Split (1)

train · 18.6k rows

file_name stringclasses 864 values	section stringclasses 208 values	text stringlengths 115 148k
10_1007_s00449-017-1757-3	chunk_1	Bioprocess Biosyst Eng CrossMark DOI 10.1007/s00449-017-1757-3 RESEARCH PAPER Interactions of phosphate solubilising microorganisms with natural rare-earth phosphate minerals: a study utilizing Western Australian monazite Melissa K. Corbett1 · Jacques J. Eksteen? · Xi-Zhi Niu3 · Jean-Philippe Croue3 . Elizabeth L. J. Watkin' Received: 11 November 2016 / Accepted: 5 March 2017 ? Springer-Verlag Berlin Heidelberg 2017 Abstract Many microbial species are capable of solubi- reaction is highly dependent on the monazite matrix struc. lising insoluble forms of phosphate and are used in agri- ture and elemental composition. culture to improve plant growth. In this study, we apply the use of known phosphate solubilising microbes (PSM) Keywords Monazite · Rare-earth elements · Phosphate to the release of rare-earth elements (REE) from the rare- solubilising microorganisms · Bioleaching · Recovery rates earth phosphate mineral, monazite. Two sources of mona- zite were used, a weathered monazite and mineral sand monazite, both from Western Australia. When incubated Introduction with PSM, the REE were preferentially released into the leachate. Penicillum sp. released a total concentration of Nested within the periodic table are 15 elements with 12.32 mg L-1 rare-earth elements (Ce, La, Nd, and Pr) atomic numbers ranging from 57 to 71 commonly referred from the weathered monazite after 192 h with little release to as the ‘Lanthanides’ [1]. Demonstrating similar physio- of thorium and iron into solution. However, cultivation chemical properties to these Lanthanides is Scandium (21) on the mineral sands monazite resulted in the preferential and Yttrium (39), and together, these 17 elements, com- release of Fe and Th. Analysis of the leachate detected monly found in the same mineral assemblages [2], are con- the production of numerous low-molecular weight organic vened together to form what are called the rare-earth ele- acids. Gluconic acid was produced by all microorganisms; ments (REE). REEs, although moderately abundant in the however, other organic acids produced differed between earth's crust, rarely occur in concentrated forms, making microbes and the monazite source provided. Abiotic leach- them economically challenging to recover with complex, ing with equivalent combinations of organic acids resulted time intensive, conventional mining approaches. Their dis- in the lower release of REE implying that other microbial tinctive electronic, optical, and magnetic properties have processes are playing a role in solubilisation of the mona- enabled them to be fundamental components of many zite ore. This study demonstrates that microbial solubilisa- imperative technologies, including mobile phones, super- tion of monazite is promising; however, the extent of the conductors, hybrid vehicles, and manufacturing industries [2]. The importance of REEs in a technologically depend- ent society cannot be understated as their unique charac- Elizabeth L. J. Watkin teristics do not permit replacement by any other metals or E.Watkin @ curtin.edu.au synthetic substitutes. As consumer demand for these prod- ucts escalates, REE requirements throughout the world are School of Biomedical Sciences, CHIRI Biosciences, Curtin University, GPO Box U1987, Perth, WA 6845, Australia expected to increase, placing stress on uncertain supply routes, and already heavily burdened markets [3]. Despite Western Australian School of Mines, Curtin University, GPO Box U1987, Perth, WA 6845, Australia more than 200 known REE-bearing ores [2], only three are considered to be primary REE mineral ores feasibly allow- Curtin Water Quality Research Centre, Department ing for the extraction of REEs: bastnasite, monazite, and of Chemistry, Curtin University, GPO Box U1987, Perth, WA 6845, Australia xenotime [4]. Acquisition of REEs obtained via the mining Springer Published online: 21 March 2017 Bioprocess Biosyst Eng of bastnasite (carbonate rich-REEs) and monazite (tho- impacts, and secure REE for future advanced technological rium-phosphate mineral) [5] is predominately performed applications. in the USA and China, contributing to resource scarcity For the application of PSMs to industrial bioprocess- and restricting commercial exploitation. Separation of the ing of rock phosphate containing rare-earth elements, ini- REEs from the ore matrix is expensive and complex and tial experiments conducted here focussed on examining involves different processes including but not restricted to: PSMs previously identified in the literature as having phos- grinding, sifting, and gravity concentration [6], low inten- phate solubilising capabilities and examining their ability sity magnetic separation [7], and flotation [8] followed by to solubilise the phosphate present in insoluble tricalcium chemical treatment with either acidic or alkaline reagents, phosphate [CasPO4, (TCP)] or a monazite source. In an under high temperatures and varying lengths of time [9]. effort to establish whether microbial mobilization of inor- As rare-earth ores may contain concentrations of thorium ganic P would allow continued microbial sustenance and and uranium up to 10% of the total ore matrix [10], poten- release REE from the ore, we report on the biogeochemi- tial radioactive waste products created during leaching lead cal changes observed, including changes in pH, organic to complicated disposal protocols or contamination of the acid production, soluble P levels, and REE concentrations REE concentrate. As phosphate precipitation is common detected in the leachate. in many natural systems [11], REEs can be found bound within igneous phosphate rock, placer deposits, and min- eral sands [12]. Methods Phosphorus (P), a vital element for all living organisms, is necessary for sustaining levels of key cellular reactions, Microbes used in this study carbon, and amino acid metabolic processes, production of The bacterial and fungal isolates used in this study are ATP, enzyme regulation, and energy transfer [13]. Micro- organisms that have the ability to obtain phosphorus from listed in Table 1. insoluble sources, known as phosphate solubilising micro- organisms (PSMs), can extricate insoluble P into a soluble DNA extraction, amplification, sequencing form by acidification, chelation, and exchange reactions and identification [14]. Commercial applications of this naturally occur- ring phenomena are currently exploited by the agricultural Bacteria industry [15] sidestepping the expensive addition of phos- phate to soils. However, this technique is currently underu- Pure isolates (Table 1) were grown overnight in 2 mL LB tilised for the recovery of minerals embedded in phospho- broth with rotation at 160 rpm at either 30 or 37°C depend- rus rock. Microbial solubilisation of phosphate from REE ing on species tested. Cells were centrifuged at 12,000xg phosphate ores has been reported [16, 17] with microbial for 20 min, DNA extracted using the MoBio DNA extrac- colonization of the mineral surface suggesting biosolubi- tion kit (MO BIO Laboratories, Inc) following manufactur- lisation, attributed to electron transfer via the secretion of er's instructions. The 16S rRNA gene region was amplified low molecular weight organic acids, such as gluconic acid, by PCR using Bioline MyTaq with 27F, 800F, 802R, and citric acid, oxalic acid, formic acid, butyric acid, and malic 1492R universal bacterial primers [24]. acid [18]. Research conducted by Qu and Lian [19], Bris- son et al. [20], and Shin [21] demonstrates that bioleaching Fungi of monazite ores by PSM belonging to the genera Pseu- domonas,Enterobacter, Serratia,Pantoea,Bacillus,and Fungal isolates (Table 1） were inoculated onto Sab- Aspergillus provided with an organic carbon source for ouraud agar (SAB) and incubated for 5 days at 30°C. growth is possible; however, recovery rates of REEs are DNA was extracted from a 5 mm diameter section of variable.
10_1007_s00449-017-1757-3	chunk_2	The application of microbial leaching is a proven fungi using the Lysing matrix A tube (MP Biomedi- technology for the recovery of copper, nickel, zinc, and cals） with 500 μL of lysis buffer [25] in a ribolyser cobalt from low-grade ores [22]. It is a well-established (MP Biomedicals) for 30 s at a speed of 4.0. RNase A industrial technique that can decrease extraction expendi- (20 μL, Sigma-Aldrich)(1 mg/ml) was added and incu- tures [23], as well as decrease the generation of damaging bated at 37°C for 5 min. NaCl (165 μL of 5 M) was environmental waste products that arise when traditionally added, mixed by inversion, and centrifuged at 21,920xg mining methodologies are engaged. The identification and for 20 min at 4°C. The supernatant was mixed with characterisation of phosphate solubilising microbes capa- 800 μL of Phenol:Chloroform:Isopropanol (25:24:1, ble of phosphate mobilization and REE release have the Sigma-Aldrich) and mixed by inversion. The sample potential to reduce recovery costs, decrease environmental was centrifuged for 5 min at 21,920xg and the top layer Springer Bioprocess Biosyst Eng Table 1 Microbial isolates Species and strain References NCBI Genbank acces- Growth used in this study sion/culture collection tempera- ture °℃ Bacteria Azospirillum brasilense DSMZ1843 Rodriguez et al. [69] NR_117478.1 37 Bacillus megaterium DSMZ 2894 Feng et al. [17] KP036928.1 37 Paenibacillus polymyxa KT783525 37 Bacillussubtilis KJ87280.1 37 Burkholderia glathei Kim et al. [67] DSMZ 50014 30 Enterobacter aerogenes ATCC 13,048 Chung et al. [60] KF516237.1 37 Klebsiella oxytoca Walpola et al. [59] AB004754 37 Klebsiella pneumoniae Chung et al. [60] AY17656 37 Pantoea agglomerans DSMZ 8570 Feng et al. [17] AM421978.1 30 Pseudomonas aeruginosa DSMZ50071 30 Pseudomonas putida DSMZ 1693 Malboobi et al. [68] FM163468.1 30 Serratia marcescens Ben Farhat et al. [71] AB909432.1 37 Fungi Aspergillus nigerDSMZ 821 Schneider et al. [70] ATCC9142 30 Aspergillus tubigensis KT033072.1 30 Penicillium sp JN7982529.1 30 removed and the process repeated. The supernatant was Monazite sample preparation and composition analysis transferred to a new tube, and an equal amount of isopro- panol added and incubated at -20°C for 2 h. Tubes were Two monazite samples from Western Australia were centrifuged at 21,920xg for 20 min at 4 °C. The superna- obtained for leaching analysis. A monazite concentrate tant was discarded, the pellet washed twice in 70% etha- sourced from the Busselton Mineral Sands deposit, Western nol, centrifuged, and then air-dried and resuspended in Australia (Cable Sands Pty Ltd) referred to as CSM, and 40 μL of TE buffer. Fungal ITS regions were amplified by a weathered lateritic monazite ore from the Mt Weld mine PCR using Bioline MyTaq with the ITS4 and ITS5 uni- (Lynas Corporation) referred to as MWM. CSM particu- versal fungal primers [26]. All PCR products were puri- late size was 90-200 μm, and the monazite concentrate was fied by Invitrogen PCR clean up kit. Sequencing was per- diluted 1:10 with Silica flour to obtain a safe Th/U working formed by the State Agricultural Biotechnology Centre, concentration. MWM was ground initially by a rod mill, Murdoch, W.A. Sequences were compared to those in the then pulverized in a ring mill and finally sieved to 1-35 μm NCBI database and isolate identity confirmed. in size. Both ores were subsequently gamma irradiated at 50 kGy for 11 h (ChemCentre, Western Australia) to ensure sterility. The mineralogical data of both ores were obtained by X-ray diffraction and elemental composition deter- Bioleaching mined by Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) (performed by CSIRO Minerals, Preparation of inocula Waterford, WA). Scanning electron microscopy-energy dis- persive X-ray spectroscopy (SEM-EDX) was performed on Bacterial species were grown in LB broth with rotation at a Zeiss Evo 40XVP SEM with an Oxford Instruments SiLi 160 rpm, at either 30 or 37°℃ depending on species tested, X-ray detector and Inca operating software (John de Laeter to exponential phase and harvested by centrifugation at Centre, Curtin University, WA). 12,000xg for 10 min. Cells were resuspended in sterile Tris-HCl buffer (pH 7), centrifuged, and washed twice Bioleaching experiments more to remove any traces of phosphate. Cells were resus- pended to a density of 1x 10* CFUs/mL in modified Piko- All leaching experiments were conducted in triplicate and vskaya (PVK) media [27] with 3% w/v glucose and pH 7. included an abiotic control in modified PVK media. The Inoculating PVK media with 5 mm plugs taken from SAB initial pH was adjusted to approximately 7 with 1 N NaOH plates grew fungal isolates. and the flasks weighed prior to and after autoclaving. Each Springer Bioprocess Biosyst Eng flask was inoculated to a concentration of 10* cells mL-1 organic acids were determined by comparing the retention and incubated for 192 h at 130 rpm at either 30 or 37°℃ times and peak areas of chromatograms of the samples with depending on species tested. standards. Organic acid identity was further confirmed by High-Resolution Mass Spectrometry. Organic standards Tricalcium phosphate included gluconic, malic, formic, butyric, citric, acetic, lac- tic, oxalic, and pyruvic. REEs (Ce, La, Nd, and Pr), and Th TCP (1% w/v) was added to modified PVK media prior to and U in solution were analysed by ICP-EOS (CSIRO Min- autoclaving. Soluble phosphate concentration was deter- erals, Waterford). mined after autoclaving by following Murphy and Riley [28] colourimetric method. pH calculations Monazite ore The concentration of protons generated by the organic acid concentrations as determined by HPLC at 192 h was cal- MWM (0.5% w/v) or 5% CSM w/v was added to flasks culated assuming a less than 10% dissociation for weak containing modified PVK media after autoclaving. Solu- acid functional groups. Malic acid (diprotic) and citric acid ble phosphate was determined via colourimetric analysis to (triprotic) were treated as individual monoprotic acids for assess the amount of P released upon ore addition to liquid. simplicity [29]. Individual pKa values for each acid were The weight of the ore was determined at 192 h (experimen- employed: tal end) for all bacterial species. Abiotic leaching with organic acids [H+]= i=1 MWM (O.5% w/v) was added to modified PVK media, Due to these simplifications, it is likely that [H+] con- centrations have been overestimated. without glucose, after autoclaving, and the following com- binations of organic acids added: gluconic acid (0.24 mM), acetic acid (2.72 mM), citric acid (0.122 mM) to mimic those produced by Penicillium sp and Aspergillus sp., glu- Results conic acid (0.007 mM), formic acid (0.007 mM), citric acid (0.03 mM) and malic acid (0.1 mM) to mimic those pro- Phosphate solubilisation from tricalcium phosphate duced by Enterobacter, and gluconic acid (0.007 mM), for- mic acid (0.0042 mM), citric acid (0.226 mM), malic acid Numerous bacterial and fungal strains from distinctly dis- (0.17 mM), and acetic acid (0.13 mM) to mimic those pro- tant genera have been reported in the literature to have duced by Pseudomonas and Pantoea. phosphate solubilising capabilities [30]. In this study, all isolates tested released phosphate into solution from insolu- Analysis ble TCP. The phosphate concentration in the media rapidly increased over the first 48 h of incubation with a contin- Samples were taken at 0, 24, 144, and 192 h, and pH deter- ued slow increase until the termination of the experiment mined (Ionode IJ series pH probe). Samples were then fil- at 192 h. The percentage phosphate leached from the TCP tered [0.20 μm (Satorius)] and assayed for soluble PO4-3 varied greatly between isolates with Aspergillus niger, A. by colourimetric method analysis. Evaporation rates were tubigensis, and Penicillium sp. leaching 75, 70, and 49% determined by weighing each flask after sampling. The of the available phosphate, respectively (Fig. 1).
10_1007_s00449-017-1757-3	chunk_3	Nota- concentrations of P were corrected for decreases in fuid bly, all three fungal species demonstrated complete clear- volume due to evaporation and the sampling process. ing of the media with large pellets formed. Of the bacte- Separation and identification of organic acids produced rial species tested, K. pneumoniae was the most effective were determined by a high-performance liquid chroma- (57% released), followed by E. aerogenes (43%), Ps. aer- tography (HPLC) instrument (Agilent 1200) coupled with uginosa (17%), B. megaterium (16%) Ps. putida (13%), a diode array detector (DAD, Agilent). Injection volume and P. agglomerans (10%) (Fig. 1). Unlike the fungal iso- was set as 50 μL for the samples. Compound separation 1 lates tested, all flasks inoculated with bacteria remained was achieved with a C18 reverse phase column (Agilent, , turbid, indicating incomplete solubilisation of the TCP in 5 μm, 4.6x250 mm). The isocratic elution flow rate was the media. There was no significant difference in the phos- 1.0 mL min-1. The mobile phase consists of 70% methanol phate leaching of remaining strains from the abiotic con- and 30% phosphate buffer (pH 2.0). A detection wavelength 1trols; these include A. brasilense (8%), B. glathei (9%), B. of 220 nm was used. Identification and concentration of subtilis (7%), P. polymyxa (8%), S. marscenes (9%). Abiotic Springer Bioprocess Biosyst Eng 100 REE release during bioleaching (REE solubility) 90 80 Concentrations of REE in the leachate after biotic incuba- 70 tion were assessed to determine release rates and solubilis- ing potential of the microorganisms. The final concentra- tion of REE leached from the MWM is shown in Fig. 2. 40 Penicillium sp was the most efficient at leaching REEs with 30 a total of 12.32 mg L-1 REE (Ce, La, Nd, and Pr) released 20 after incubation for 192 h. Concentrations of REE leached by the other fungal species, A. niger and A. tubigensis, were 24 times lower than Penicillium sp. (0.44 and 0.43 mg L-, respectively). Bacterial species, P. agglomerans, E. aero- genes, and Ps. putida released bound REEs into solution at similar concentrations (1.63, 1.93, and 1.45 mg L-1, Microbial species respectively). The remaining bacterial species released lower levels of total REEs: K. pneumonia (0.56 mg L-'), K. Fig.1 Percentage of phosphate solubilised from tricalcium phos- oxytoca (0.42 mg L-1), and B. megaterium (0.58 mg L-1). phate after 192 h incubation with microbial isolates. Data are aver- ages ± SE of two experiments each with three biological replicates When CSM was provided as a phosphate source, all species showed preferential leaching of thorium and iron (Fig. 3) over REE, with no concentration of Ce, La, or Nd recorded controls at both 30 and 37°C leached 6% of the available above 0.02 mg L-1 for any species tested. Between 4 and phosphate into solution. These isolates were excluded from 30 mg of MWM ore was lost over the duration of the further studies on monazite ores. experiment when incubated with the bacterial isolates. P solublisation from monazite bioleaching 14.0 Upon addition of the monazite source to the modified PVK 12.0 media, spontaneous release of P at time O ranged from 0.8 to 1.7 mg L-1. After 144 h incubation with a microbial iso- 10.0 late, P levels dropped to an average of 0.43 mg L-1 ± 0.11 8.0 and remained steady until the end of the experiment (data not shown). 6.0 Elemental composition of monazite ores The composition of the two monazite ores is shown in Table 2. While they had similar percentages of the REEs, Ce, La, and Nd, the Cable Sands monazite contained sig- nificantly greater levels of Th and U. The MWM contained three times the levels of Fe as the Cable Sands monazite. Microbialspecies XRD analysis of the ores determined monazite——((Ce, La, Nd, Th)PO4） and zircon (ZrSiO4) composition for CSM Fig. 2 Total concentration of REEs, Cerium (black coloured box), Lanthanum (gray coloured box), Neodymium (box with single line), and monazite ((Ce, La, Nd, Th)PO4) and Florencite (Ce, and Praseodymium (slashed box) released into leachate after 192 h La, Nd)A13+(PO4)2(OH)g for MWM. inoculation with microbial species. Data are averages of three experi- ments each with three biological replicates Table 2 ICP-EOS analysis for elemental composition of Monazite ores used in this study Sample ID Ce La Nd Th U Y Pr Fe P Ca Mg Si Ti Zr Concentration content (%) Mt weld monazite 12.6 10.1 6.25 0.162 <0.003 0.165 2.10 1.23 9.93 1.75 0.199 1.96 0.554 0.031 Cable sand monazite 14.0 9.415.46 4.98 0.194 0.851 0.184 0.458 9.45 0.693 0.079 1.72 0.325 1.80 Springer Bioprocess Biosyst Eng Thorium 50.0 lron 0.5 40.0 .7 0.4 uo 30.0 0.3 0.2 0.1 10.0 0.0 Microbialspecies Microbial species Fig.3 Final levels of thorium and iron released into solution from 192 h. Data are averages ± SE of three experiments each with three Mount Weld Monazite (black coloured box) and Cable Sands Mona- biological replicates zite (gray coloured box) after incubation with microbial species for Fig. 4 Scanning electron micrographs of undiluted Cable Sands Monazite particles demonstrating visual structural differences amongst particles of the same sample. a Rare-earth element phosphate with surface iron and pitting. b Zircon. c Clay-like particle 90 μm b 60 160μm Springer Bioprocess Biosyst Eng Fig. 5 Scanning electron microscopy with energy Spectrum 1 dispersive X-ray (SEM-EDX) spectral markers 1, 2, and 3 located on surface of Cable Sands monazite REE particle (Fig. 4a). SEM-EDX of Zircon particle (Fig. 4b, spectrum #4) and aluminium silicate clay mineral (Fig. 4c, spectrum #5). Y-axis represents number of counts, and the X-axis depicts 0.5 1.5 2.5 7.5 the energy level (keV) of those counts Spectrum2 (0 cts) e Spectrum 3 0.5 .5 ev Spectrum 4 7.5 -0.126 (0 cts) keV Spectrum 5 0.5 1.5 2.5 3.5 5.5 6.5 7.5 8 Full Scale 1552 cts Cursor: -0.126 (0 cts) Springer Bioprocess Biosyst Eng SEM examination of monazite ores isolate pH remained high after 24 h (between 5 and 6), but steadily decreased reaching 1.75-3 by 192 h (Fig. 7). Bac- The surface compositions of the two ores were examined terial isolates on MWM after 24 h had pH ranges from 3.9 by SEM-EDX. The CSM is a heterogeneous ore (Fig. 4a), to 5.7 demonstrating a greater difference in pH change than as along with phosphate bound REEs (Fig. 5, spectrum #1), observed on CSM; however, after 192 h, pH levels were numerous iron coverings (spectrum #2), and pits (spec- similar across the strains (3.1-3.8) (Fig. 7). Fungal isolate trum #3) were detected on the surface of the ore particle. pH ranged from 4.6 to 5.5 after 48 h, falling to 1.8-2.17 Additional particles containing no surface REEs were also after 192 h. recorded (Fig. 4b), composed of Zr and Si (spectrum #4) with other particles demonstrating clay, such as elemen- tal compositions (Fig. 4c) with aluminium silicates trace Organic acid production amounts of iron, calcium, titanium, and manganese (spec- trum #5). Comparatively, analysis of the MWM showed no All microorganisms used in this study produced gluconic zircon, silicates, or iron coverings (Fig. 6a). All particles acid at varying concentrations. The suite of other organic examined had a homogenous surface elemental composi- acids produced, and the concentrations they were produced tion (Fig. 6b) of REE and P. The lack of surface iron or at differed based on the monazite source provided. In addi- traces of thorium with the MWM reflects the low levels tion to gluconic acid, the bacterial species produced acetic, detected in the leachate after microbial incubation. formic, malic, and citric acids. When grown on MWM, the fungal species produced gluconic and acetic acids at higher pH changes concentrations than the bacterial species (Table 3). Growth on CSM resulted in production of fewer organic acids, with After autoclaving and addition of sterile glucose, the initial no formic acid detected for any microbial species. The con- pH of all isolates on either MWM or CSM was 6.6±0.1 at centration of gluconic acid produced by the bacterial spe- time 0.
10_1007_s00449-017-1757-3	chunk_4	After 24 h with CSM, all bacterial species demon- cies when grown on CSM was greater than when grown on strated a decrease in pH to below 4.6, finishing between 2.6 MWM; however, there was a greater variation in the range and 4.12 by 192 h (192 h data depicted in Fig. 7). Fungal of organic acids produced by the bacterial species when Fig. 6 a, b Scanning electron micrograph of Mount Weld monazite particles (a) and corresponding scanning elec- tron microscopy with energy dispersive X-ray output (b). Y-axis represents number of counts, and the X-axis depicts the energy level (keV) of those counts p ce Nd Nc N 0.5 1.5 2.5 3.5 4.3 7.5 Full Scale 1489 cts Cursor: -0.086(4 cts) ev Springer Bioprocess Biosyst Eng 6.0 a Calculated [H+] generated within the bacterial systems inoculated on MWM demonstrated production of high 5.0 enough concentrations of acid to reach the measured 4.0 pH levels. Bacterial species inoculated on CSM pro- duced less calculated acid with the pH not reaching the 3.0 observed levels at 192 h. The greatest difference between 2.0 口 calculated pH and observed pH was demonstrated by all three fungal species, each of which had a much lower actual pH than what was calculated based on organic acid productions on both MWM and CSM (Fig. 7). Microbialspecies b Abiotic leaching with organic acids 5.0 Abiotic leaching with the same ratios of organic acids 4.0 produced by the fungal species resulted in O.6 mg L-1 of 3.0 REEs in total detected in the leachate, with 0.04 mg L-1 of thorium, and undetectable levels of iron (<0.02 mg L-1) 2.0 (Fig. 8). Total REEs with abiotic organic acid leaching are greater than those seen with A. niger (0.44 mg L-1) and A. tubigensis (0.43 mg L-1), but 18 times less than what was recorded with the Penicillium species. The tho- rium levels (0.04 mg L-1) mobilized from the MWM by abiotic leaching were 3.8 times lower than when incu- bated with A. niger and 2.75 times lower than A. tubi- Fig.7 Comparison of observed pH (gray coloured box) with the calculated pH (filled diamonds) based on the concentration of the gensis. Abiotic addition of organic acids based on those organic acids detected by HLPC. Analysis of organic acids after produced by bacteria resulted in lower REE leachate lev- 192 h, incubation with either Mount Weld monazite (a) or Cable els than recorded with any bacterial species tested, with Sands monazite (b). Data are averages ±SE of three experiments (a) the highest amount (0.208 mg L-') resulting from organic or two experiments (b) each with three biological replicates acid mix number 1. Organic acid mix #2 resulted in only 0.096 mg L-1 total REEs detected in the leachate. As grown on MWM. For the fungal species, citric acid was seen with the abiotic fungal system, low Th levels were only detected in Penicillum sp. grown on MWM. detected with organic mix #1 (0.08 mg L-1) and unde- Concentrations of organic acids identified by HPLC tectable levels with organic acid mix 2 (<0.02 mg L-1). were converted to [H+] generated and assessed against Both organic acid mix #1 and #2 had undetectable levels observed pH at 192 h to determine whether enough of iron (<0.02 mg L-1) in the leachate. acid had been produced to reach pH levels recorded. Table 3 Maximum organic Microbial species Gluconic Acetic Formic Citric Malic acid concentration (mM) reached after 192 h days of MWM CSM MWM CSM MWM CSM MWM CSM MWM CSM incubation in the presence of Mount Weld monazite B. megaterium 0.0037 0.0642 0.438 0 0.0021 0 0.0934 0 0.285 0.012 (MWM) and, Cable Sands E.aerogenes 0.0071 0.114 0 0 0.0078 0 0.0336 0 0.1012 0.13 monazite (CSM). Results are K. oxytoca 0.0017 0.0123 0 0 0.0037 0 0.0307 0 0.474 0 the mean of three biological K. pneumoniae 0.005 replicates each with three 0.0045 0 0.1082 0.0037 0 0.084 0 0.081 0 technical replicates P. agglomerans 0.0073 0.054 0 0.069 0.0042 0 0.0336 0 0.1 0 Ps. aeruginosa 0.0049 0.0467 0.0825 0.0053 0 0.0352 0.059 0.16 0 Ps. putida 0.0042 0.0349 0.129 0.086 0.0042 0 0.226 0.027 0.169 0 A. niger 0.274 0.1955 1.4 0 0.0029 0 0 0 0.038 0 A. tubigensis 0.237 0.0321 2.256 0 0 0 0 0 0 Penicillium sp 0.162 0.0122 1.47 0 0.112 0 0 0 MWM Mount Weld monazite, CSM Cable Sands monazite Springer Bioprocess Biosyst Eng 0.7 were detected in the leachate and levels of all REEs were proportional to their concentrations in the ore. As Pr is 0.6 chemically similar to the other rare-earth elements, the dif- ference in leaching ratios is interesting. Levels of REEs (Ce, La, and Nd) leached from the CSM were never greater than 0.02 mg L-1. This may be attrib- 0.3 uted to the 10:1 dilution with silica flour that was necessary to work within radiation safety levels or the larger particle size of CSM. MWM surface area was greater than the CSM 0.1 due to smaller particle sizes, preferentially allowing more 0.0 colonization and microbial attachment for P mobilisation Fungal OA mix Bacterial OA mix 1 Organicacid mix and consequently greater release of REE into the leachate. Not only was the release of REE into the leachate notice- Fig. 8 Final levels of total REEs (black coloured box), thorium (gray ably different between the two ores, but also the release of coloured box), and iron (slashed box) released into solution from 'contaminants', such as thorium and iron. As the thorium Mount Weld monazite after incubation with three abiotic organic acid mixes for 192 h. Data are averages two experiments each with three concentration in the CSM matrix is 5%, it was unsurprising biological replicates to discover it at detectable levels in the leachate in com- parison with the MWM where it only constituted 0.16% of the total mass. Although Ce, La, and Nd concentrations in Discussion the CSM ore were all greater than 5% (Table 2) and should have resulted in larger amounts detected in the leachate, the Traditional REE retrieval methods enable high levels of ore matrix has had a strong effect on elemental release dur- REE recovery from various ore concentrates (>90%); how- ing solubilisation resulting in no REE concentration above ever, newer and greener methods of REE extraction and 0.02 mg L-1 for any species tested. Contrastingly, Fe makes recycling [31] are in high demand with a focus on bioleach- up 0.46% in the CSM and 1.23% in MWM, but much ing, solid-phase extraction [32], and UV light separations greater Fe levels were detected in the CSM leachate, up to [33]. Examination of the use of PSM for REE recovery is a 45 mg L-1 with A. niger (Fig. 3). As can be seen occasion- newly emerging field with a number of studies demonstrat- ally with gold particles from pyritic sources, hydrated iron ing varying degrees of success with a range of REE ores, oxides form natural coatings around the particles, which microorganisms, temperatures, and lengths of experimental are detrimental to gold recovery [34]. The presence of sur- time resulting in leaching effciencies in between 0.1-25% face iron in the CSM (Fig. 4a) and release of bound Fe dur- [19-21]. ing P mobilisation or by microbial organic acid production This study successfully demonstrated the release of REE [35] may explain why iron concentrations in the CSM lea- from a weathered monazite source when incubated with chate were so high in comparison with all other elements both bacterial and fungal isolates. Cerium was leached in measured. With continued microbial acid production, these the highest concentration by all microbial species tested on soluble Fe levels would further increase as pH decreased. MWM (as was expected due to its high percentage in the As the presence of iron during REE solvent extraction is ore). While Lanthanum constituted 10.1% of the MWM undesirable [36] with high concentrations inhibiting pre- and Nd, 6.25%, the levels of the two elements in the lea- cipitation of REEs [37], the removal of iron via bioleach- chate after incubation with E. aerogenes, P. agglomerans, ing with PSM prior to REE recovery may aid in subsequent Ps. putida, and Penicillium sp. for 192 h were often similar, REE precipitation.
10_1007_s00449-017-1757-3	chunk_5	In addition, during the solvent extrac- with Nd levels approximately 87-97% of those recorded tion process, thorium is co-precipitated along with the for La. Incubation with B. megaterium and Ps. aeruginosa REEs [38] resulting in contamination of the leach liquor. leached Nd at levels 20% greater than La, even though the By preventing the co-precipitation of Fe with the REEs total amounts of leached REEs were low. The only isolates during recovery, we propose that this preferential bioleach- that produced REEs leachate ratios according to their com- ing of thorium and iron over the REEs may provide a treat- position in the ore were the Aspergillus isolates, as levels of ment step prior to solvent extraction, potentially decreasing Nd were just under half those recorded with La. Similarly, environmental hazards associated with these radioactive a discrepancy was also recorded with the concentration of materials and decreasing rates of iron obstruction during Praseodymium in the leachate for all species bar Penicil- REE precipitation. It may also be applicable to the recovery lium sp. The isolates that leached higher levels of REEs of REE from scrap iron or for the removal of iron from neo- recorded preferential leaching of Ce, La, and Nd over Pr, dymium-iron-boron magnets during REE recycling [39] in whereas those isolates where only low quantities of REEs place of volatile, fammable solvents. Springer Bioprocess Biosyst Eng Heterotrophic metabolism by cells during growth the monazite lattice and the available surface area variable, resulted in a decreased pH of the leachate due to organic leaching kinetics could be extremely slow. acid production or in response to respiratory acidification No citric acid was detected in leachates from the Asper- and NH4+ assimilation [40]. Species incubated with CSM gillus sp. As a number of low molecular weight organic produced fewer organic acids and at lower concentrations acids have short half lives (2-6 h for citric) [50], deg- compared to those produced when incubated with MWM. radation of citric acid generated by Aspergillus prior to As bacterial metabolism of organic compounds can be sampling is a possible alternative to it being produced at altered by P availability [41], a switch to cellular respiration levels below detection. By 192 h, all fungal isolates that over metabolism would have resulted in the observed low- still had non-limiting glucose concentrations (>10%） and ering of the pH over 192 h with no corresponding organic had favourable pH levels (<pH 2) for citric acid produc- acid production. Numerous CSM particles consisted of zir- tion [51]. However, the addition of manganese sulfate con, clay minerals, and iron, resulting in reduced access to (0.002 g L-1) in the growth media may have been an inhib- surface P levels compared to MWM, affecting numerous iting factor for the production of citric acid by both Asper- biological pathways. Low-molecular weight organic acids, gillus strains. Even though concentrations were low, pellet including acetic, formic, citric, malic, and oxalic acids, clumping was visible in all fasks and previous research can increase P availability from inorganic sources [42] states that even very low concentrations of manganese with oxalic acid used to precipitate REEs from monazite (0.0004-0.002 mg L-1) [52] can decrease production of cit- ores [43]; however, no species tested either on MWM or ric acid. Unlike Aspergillus, some Penicillium species have CSM produced detectable levels of oxalic acid. It is possi- been demonstrated to produce citric and oxalic acid while ble that some oxalic acid has been lost due to oxalate-REE leaching mangniferous ores containing high levels of Mn precipitates forming; however, as industrial processes often (25.7%) [53]. In this study, it is unlikely that the levels of recover surplus oxalic acid [44] after processing, trace Mn in the growth media, or released from the MWM, had amounts remaining in the leachate should have been detect- a negative impact on the production of citric acid by Peni- able by HPLC. In this instance, it is unlikely that oxalate- cillium sp. Conversely, production of citric acid by either REE complexes have formed, thereby lowering the REE Aspergillus isolates or Penicillium in the presence of Mn content determined in the leachate. The acids detected in may have resulted in the formation of an insoluble man- this study are known to form salts with lanthanum; how- ganese citrate complex that would adhere to the ore [54] ever, lanthanum gluconate, lanthanum acetate, and lantha- reducing the overall detectable levels of citric acid in solu- num formate [45] are all water soluble, thereby not pre- tion. The high concentration of Fe2+ (10 mg L-l) released cipitating and lowering the REE leachate content. Cerium by Aspergillus from the MWM is likely to have caused the (Ce4+) gluconate forms a stable water soluble complex at inhibition of citric acid production, as media deficient in basic pH [46] precipitating with OH- as the pH increases, iron is used for commercial citric acid manufacture [55]. however, remains soluble at acidic pH as was recorded in As the Penicillium sp. released less iron during leaching of these experiments. MWM than either Aspergillus isolate, it is possible that the Numerous isolates produced formic and acetic acid after action of aconitate hydratase [56] and conversion of isocit- MWM incubation, but carboxyl 1 group organic acids have ric acid were suppressed, resulting in a lower production of poor metal complexing abilities, whereas malic and citric citric acid. In this case, itaconic or itatartaric acid may have have a high affinity for trivalent metals [47], such as the been produced [57] and two standards that were not used in REEs. As all the bacterial isolates studied here produced this study. Citric acid can release REEs at concentrations malic acid, it is important to note that malic acid is used less than 1 mM [49]. Therefore, if small amounts were as an eluent during the ion-exchange process in rare-earth continuously produced by the Aspergillus strains, a greater separation [48] and may have contributed to the matrix impact on P release would have been seen than by other solubilisation and REE release from the MWM. Aliphatic low molecular weight organic acids [58] which would have ligands of REEs with citrate are stable [49], and as other resulted in higher amounts of REEs in the leachate. Citric REEs have similar valencies, it is likely that other REE acid mediated dissolution of monazite would see REEs salts formed from complexing with these acids are also released concomitantly to P discharge [47] further accel- water soluble. We are confident that the levels of REE in erated by decreases in pH. As Penicillum sp. and all the the MWM leachate are accurate and, therefore, have not bacteria tested produced citric acid (0.0307-0.22 mM) on been unduly underestimated; however, due to the heteroge- MWM, but REE leachate levels varied greatly and it would neous nature of the CSM, precipitation of REEs with other imply that other microbial interactions and experimental minerals during solubilisation should not be excluded as a conditions are having an impact on solubility of the mon- theoretical reason as to why total REE leachate concentra- azite. HPLC analysis of the leachate resulted in a number tions were low. In addition, as REEs are strongly bound in of small unidentified peaks. A limited number of standards Springer Bioprocess Biosyst Eng were used in the analysis; however, these small peaks could Mn levels and possibly inhibited growth [65]. Both these be accounted for by either the amino acids present in the organic acids are produced by the Klebsiella sp., whereas yeast extract added to the media or other microbially pro- other species that produced these same acids remained duced organic acids.
10_1007_s00449-017-1757-3	chunk_6	It is likely that unidentified acids in unaffected by increases in soluble Mn levels. this study were important in solubilising the MWM result- Numerous strains of B. megaterium are known to solu- ing in the higher REE levels detected and lower pH in the bilise phosphate [66] and are actively applied in agriculture leachate of Penicillum sp. than for other species. for promoting plant growth. While in this study, B. mega- Abiotic leaching of REEs from MWM was tested using terium was one of the less efficient microbes at leaching organic acid combinations produced by Penicillum sp., E. REEs with a total of 0.577 mg L-1 in the leachate, and Shin aerogenes or Ps. putida and P. agglomerans. Leaching by et al. [50] did not observe any leaching of REEs from a organic acid combinations based on Penicillium sp. produc- monazite ore with B. megaterium. However, the monazite tion released the greatest concentration of REEs from the ore in that study had significantly lower levels of REEs and MWM, in proportion to their concentrations within the ore a higher level of Fe than the two monazite samples tested matrix, as was seen previously with biological leaching. here, possibly explaining the different result. Neither organic acid combination based on those produced It was also noted that the levels of soluble P levels in the by E. aerogenes, or Ps. putida and P. agglomerans resulted leachate were not correlated with the release of REEs into in REEs release from the matrix at levels higher than any solution. During glucose fermentation, microbial incor- bacterial isolate tested. In comparison with the biologically poration of inorganic phosphorus from the surrounding produced acids, abiotically added acids released undetecta- medium for conversion to organic phosphoric compounds ble levels of Fe2+ (<0.02 mg L') after 192 h incubation and is common as it would be appropriated for immediate use Th levels in the leachate were <0.08 mg L-1 for all three in various metabolic processes. This provides an explana- combinations trialed. As none of the abiotic organic acid tion as to why P levels in the leachate were very low, while mixes resulted in the release of similar levels of REE from some levels of REEs were high and as such why the meas- the MWM matrix, it appears that additional compounds urement of P should not be relied upon as an indicator of secreted by the microorganisms or possible attachment to Ore solubilisation. mineral surface for phosphate utilisation are playing addi- tional roles in solubilisation of the monazite ore. It is apparent that testing phosphate solublisation of Conclusion PSMs on an ideal P source (TCP) did not necessarily reflect the microorganism's capabilities when provided with an This study identified a number of potential PSM capable of REE ore. Klebsiella pneumoniae and K. oxytoca both solu- solubilising Western Australian monazite ores with pref- bilised TCP but were unable to release bound REEs in high erential leaching of REEs from the weathered MWM over concentrations, even though they produced similar organic iron and thorium. In contrast, leaching of CSM resulted acids to Ps. putida and E. aerogenes, both of which were in the release of iron and thorium rather than REEs when able to release REEs into solution. These two strains of inoculated with a PSM. This partisan leaching of REEs Klebsiella have been shown to solubilise inorganic phos- from the MWM compared to the CSM refects the differ- phate for plant growth promotion [59, 60]; however, in ences in ligand stability constants and REE coordination this instance, both lacked the ability to obtain P from an within ore lattice [49] and should be considered important ore source. When incubated with monazite, Klebsiella sp. when choosing both a PSM and an ore source for mineral production of malic acid, combined with the lowering of leaching. REE release appears to not only be dependent the pH, may have increased A13+ solubility [47] from the on the ore lattice but also the type and concentration of MWM, thereby increasing metal toxicity. As A13+ organic organic acid secreted during glucose fermentation, and as ligand complexes inhibit microbial metabolic activities some acids form REE salts, precipitation of released REEs of Klebsiella sp [61], inhibition of P mobilisation from may occur lowering the overall concentration of REEs in the ore surface is also likely, thereby lowering the sum of solution. Phosphate uptake by the microbial isolates for REE detected in the leachate inhibiting respiration and continued metabolic cycling occurring during monazite propagation. Pseudomonas and Pantoea are known to be solubilisation, and as such the impact that phosphate pre- more tolerant of A13+ [62, 63] and successfully released cipitation could have on the system is diminished, reducing REEs in this study. However, as the exact concentration of the loss of REEs to secondary mineral precipitates in the A13+ in the ore is currently unknown and only determined leachate. to be present by XRD, this is currently speculation. The This study successfully demonstrates the concept that release of bound Mn by malic and citric acid when pH<5 PSMs can be applied in bioleaching systems to release [64], is also known to contribute to increased extracellular bound REEs, but to be competitive with conventional Springer Bioprocess Biosyst Eng procedures, the recovery rates via this process need to be 16. Taunton AE, Welch SA, Banfield JF (2000) Geomicrobiologi- without significantly i cal controls on light rare earth element, Y and Ba distributions increased increasing operational during granite weathering and soil formation. J Alloys Compd overheads. Further examination is required to identify and 303:30-36 understand the effects of microbial produced organic acid 17. Feng MH, Ngwenya BT, Wang L, Li WC, Olive V, Ellam RM and enzymes on different REE bearing ores and to optimize (2011) Bacterial dissolution of fluorapatite as a possible source of elevated dissolved phosphate in the environment. Geochim Et leaching rates with higher slurry concentrations. Cosmochim Acta 75:5785-5796 18. Watling HR (2015) Review of biohydrometallurgical metals Acknowledgements This research was funded by the Minerals extraction from polymetallic mineral resources. Minerals 5:1-60 Research Institute of Western Australia (M434), supported by Cur- tin Health Innovation Research Institute and Minerals Engineering, Qu Y, Lian B (2013) Bioleaching of rare earth and radioactive elements from red mud using Penicillium tricolor RM-10. Biore- Curtin University. We acknowledge the use of Curtin University's sour Technol 136:16-23 Microscopy and Microanalysis Facility, whose instrumentation has Brisson VL, Zhuang W-Q, Alvarez-Cohen L (2016) Bioleaching been partially funded by the University, State and Commonwealth of rare earth elements from monazite sand. Biotechnol Bioeng Governments. We are grateful to the Lynas Corporation for funding 113:339-348 and donation of a weathered monazite ore. 21. Shin D, Kim J, Kim B, Jeong J, Lee J (2015) Use of phosphate solubilizing bacteria to leach rare earth elements from monazite bearing ore. Minerals 5:189-202 22. Domic EM (2007) A review of the development and current sta- tus of copper bioleaching operations in Chile: 25 years of suc- References cessful commercial implementation. In: Rawlings DE, Johnson DB (eds) Biomining. Springer, Berlin 1.1 Long KR, Van Gosen BS, Foley NK, Cordier D (2012) The prin- 23. Bosecker K (1997) Bioleaching: metal solubilization by microor- ciple rare earth elements deposits of the United States: a sum- ganisms. FEMS Microbiol Rev 20:591-604 mary of domestic deposits and a global perspective. In: Larsen 24. Lane DJ (1991) 16s/23s rRNA sequencing. Nucleic acid tech- RS, Wellmer F-W (eds) Non-renewable resource issues: geosci- niques in bacterial systematics, Wiley, New York entific and societal challenges, Springer, New York 25. Al-Samarrai TH, Schmid J (2000) A simple method for extrac- 2.1 EPA US (2012) Rare earth elements: a review of production, tion of fungal genomic DNA.
10_1007_s00449-017-1757-3	chunk_7	Lett Appl Microbiol 30:53-56 processing, recycling and associated environmental issues. 26. White TJ, Bruns T, Lee S, Taylor J (1990) Amplification and National Risk Management Research Laboratory direct sequencing of fungal ribosomal RNA genes for phyloge- 3. Chu S (2011) Critical materials strategy. In: Energy UDo (ed) netics. Academic Press, San Diego United States of America, p 189 27. Nautiyal CS (1999) An efficient microbiological growth medium 4. Gupta CK, Krishnamurthy N (2004) Extractive Metallurgy of for screening phosphate solubilizing microorganisms. FEMS Rare Earths, CRC Press Microbiol Lett 170:265-270 5. Neelameggham NR, Alam S, Oosterhof H, Jha AA, Wang S 28.1 Murphy J, Riley JP (1962) A modified single solution method (2014) Rare metal technology 2014. Wiley, Hoboken for the determination of phosphate in natural waters. Anal Chim 6.1 Falconer A (2003) Gravity separation: old technique/new meth- Acta 27:31-36 ods. Phys Sep Sci Eng 12:31-48 29. Lide DR, Frederikse HPR (1995) Dissociation constants of 7.2 Zhang J, Zhao B, Schreiner B (2016) Separation hydrometal- organic acids and bases. CRC handbook of chemistry and lurgy of rare earth elements. Springer, Berlin physics 8. Fuerstenau D (2013) Design and development of novel flotation Uroz S, Calvaruso C, Turpault MP, Frey-Klett P (2009) Mineral reagents for the beneficiation of mountain pass rare-earth ore. weathering by bacteria: ecology, actors and mechanisms. Trends Miner Metall Process 30:1-9 Microbiol 17:378-387 9. Jha MK, Kumari A, Panda R, Rajesh Kumar J, Yoo K, Lee JY 31. Binnemans K, Jones PT, Blanpain B, Van Gerven T, Yang Y, (2016) Review on hydrometallurgical recovery of rare earth met- Walton A, Buchert M (2013) Recycling of rare earths: a critical als. Hydrometallurgy 161:77 review. J Cleaner Prod 51:1-22 10.1 Ragheb M (2011) Thorium resources in rare earth elements. 32.1 Dave SR, Kaur H, Menon SK (2010) Selective solid-phase 11. Brady PV, House WA (1996) Surface-controlled dissolution and extraction of rare earth elements by the chemically modified growth of minerals. Brady PV (ed) Physics and chemistry of amberlite XAD-4 resin with azacrown ether. React Funct Polym mineral surfaces. CRC Press, Boca Raton 70:692-698 12.1 Emsbo P, McLaughlin PI, Breit GN, Du Bray EA, Koenig AE 33. Van den Bogaert B, Havaux D, Binnemans K, Van Gerven (2015) Rare earth elements in sedimentary phosphate deposits: T (2015) Photochemical recycling of europium from Eu/Y solution to the global REE crisis? Gondwana Res 27:776-785 mixtures in red lamp phosphor waste streams. Green Chem 13. Vyas P, Gulati A (2009) Organic acid production in vitro and 17:2180-2187 plant growth promotion in maize under controlled environment 34. Marsden JO, House CI (1960) The chemistry of gold extraction, by phosphate-solubilizing fluorescent Pseudomonas. BMC 2nd edn. Society for Mining, Metallurgy and Exploration, Inc., Microbiol 9:174 United States of America 14. Son HJ, Park GT, Cha MS, Heo MS (2006) Solubilization of 35. Johnson SE, Loeppert RH (2006) Role of organic acids in insoluble inorganic phosphates by a novel salt-and pH-tolerant phosphate mobilization from iron oxide. Soil Sci Soc Am J Pantoea agglomerans R-42 isolated from soybean rhizosphere. 70:222-234 Bioresour Technol 97:204-210 36. Jorjani E, Shahbazi M (2012) The production of rare earth ele- 15. Igual JM, Valverde A, Cervantes E, Velazquez E (2001) Phos- ments group via tributyl phosphate extraction and precipitation phate-solubilizing bacteria as inoculants for agriculture: use stripping using oxalic acid. Arab J Chem of updated molecular techniques in their study. Agronomie 37. Woyski MM, Harris RE (1963) Treatise on analytical chemistry. 21:561-568 Interscience Publishers, New York Springer Bioprocess Biosyst Eng 38. Xie F, Zhang TA, Dreisinger D, Doyle F (2014) A critical review 58． Fox TR, Comerford NB (1992) Infuence of oxalate loading on on solvent extraction of rare earths from aqueous solutions. phosphorus and aluminum solubility in spodosols. Soil Sci Soc Miner Eng 56:10-28 Am J 56:290-294 39. Vander Hoogerstraete T, Wellens S, Verachtert K, Binnemans 59. Walpola BC, Arunakumara KKIU, Yoon MH (2014) Isolation K (2013) Removal of transition metals from rare earths by sol- and characterisation of phosphate solubilizing bacteria (Klebiella vent extraction with an undiluted phosphonium ionic liquid: oxytoca) with enhanced tolerant to environmental stress. Afr J separations relevant to rare-earth magnet recycling. Green Chem Micrbiol Res 8:2970-2978 15:919-927 60. Chung H, Park M, Madhaiyan M, Seshadri S, Song J, Cho H, Sa 40. Ilmer P, Schinner F (1995) Solubilization of inorganic calcium T (2005) Isolation and characterization of phosphate solubilizing phosphates—solubilization mechanisms. Soil Biol Biochem bacteria from the rhizosphere of crop plants of Korea. Soil Biol 27:257-263 Biochem 37:1970-1974 41. Scofield V, Jacques SMSs, Guimaraes JRD, Farjalla VF (2015) 61. Brynhildsen L, Rosswall T (1989) Effects of cadmium, copper, Potential changes in bacterial metabolism associated with magnesium, and zinc on the decomposition of citrate by a Kleb- increased water temperature and nutrient inputs in tropical humic siella sp. Appl Environ Microbiol 55:1375-1379 lagoons. Front Microbiol 6:310 62.1 llmer P, Schinner F (1997） Influence of aluminum on motility 42. Bolan NS, Naidu R, Mahimairaja S, Baskaran S (1994) Influence and swarming of Pseudomonas sp. and. Arthrobacter sp. FEMS of low-molecular-weight organic acids on the solubilization of Microbiol Lett 155:121-124 phosphates. Biol Fertil Soils 18:311-319 Sulbaran M, Pérez E, Ball MM, Bahsas A, Yarzabal LA (2009) 43. Chi R, Xu Z (1999) A solution chemistry approach to the study Characterization of the mineral phosphate-solubilizing activ- of rare earth element precipitation by oxalic acid. Metall Mater ity of Pantoea agglomerans MMB051 isolated from an iron-rich Trans B 30:189-195 soil in Southeastern Venezuela (Bolivar state). Curr Microbiol 44. Alam S, Kim H, Neelameggham NR, Ouchi T, Oosterhof H 58:378-383 (2016) Rare metal technology 2016. Wiley, Hoboken 64. Jauregui MA, Reisenauer HM (1982) Dissolution of oxides of 45. Lopes CFF, Iha K, Neves EA, Suarez-Iha MEV (1996) A com- manganese and iron by root exudate components. Soil Sci Soc parison of carboxylate complexes of lanthanum (i): formate, Am J 46:314-317 acetate and propionate. J Coord Chem 40:27-34 65. Martin JE, Waters LS, Storz G, Imlay JA (2015) The Escherichia 46. Sawyer DT, Ambrose RT (1962) Cerium (IV) gluconate com- coli small protein MntS and exporter MntP optimize the intracel- plexes. Inorg Chem 1:296-303 lular concentration of manganese. PLos Genet 11:e1004977 47. Jones DL (1998) Organic acids in the rhizosphere—a critical 66. Rodriguez H, Fraga R (1999) Phosphate solubilizing bacte- review. Plant Soil 205:25-44 ria and their role in plant growth promotion. Biotechnol Adv 48. Sinha SP (1966) Complexes of the rare earths. Pergamon Press 17:319-339 Ltd, London 67.1 Kim Y-H, Bae B, Choung Y-K (2005) Optimization of biological 49. Goyne KW, Brantley SL, Chorover J (2010) Rare earth element phosphorus removal from contaminated sediments with phos- release from phosphate minerals in the presence of organic acids. phate-solubilizing microorganisms. J Biosci Bioeng 99:23-29 Chem Geol 278:1-14 68. Malboobi MA, Owlia P, Behbahani M, Sarokhani E, Moradi S, 50. Van Hees PAW, Jones DL, Godbold DL (2003) Biodegradation Yakhchali B, Deljou A, Morabbi Heravi K (2009) Solubilization of low molecular weight organic acids in a limed forest soil. of organic and inorganic phosphates by three highly efficient soil Water Air Soil Pollut Focus 3:121-144 bacterial isolates. World J Microbiol Biotechnol 25:1471-1477 51.1 Max B, Salgado JM, Rodriguez N, Cortés S, Converti A, 69. Rodriguez H, Gonzalez T, Goire I, Bashan Y (2004) Gluconic Dominguez JM (2010) Biotechnological production of citric acid production and phosphate solubilization by the plant acid. Braz J Microbiol 41:862-875 growth-promoting _bacterium Azospirillum spp. Naturwissen- 52.
10_1007_s00449-017-1757-3	chunk_8	Clark DS, Ito K, Horitsu H (1966) Effect of manganese and other schaften 91:552-555 heavy metals on submerged citric acid fermentation of molasses. 70. Schneider KD, Van Straaten P, De Orduna RM, Glasauer, S, Tre- Biotechnol Bioeng 8:465-471 vors J, Fallow D, Smith PS (2010) Comparing phosphorus mobi- 53. Acharya C, Kar RN, Sukla LB (2003) Studies on reaction mecha- lization strategies using Aspergillus niger for the mineral disso- nism of bioleaching of manganese ore. Miner Eng 16:1027-1030 lution of three phosphate rocks. J App Microbiol 108:366-374 54.1 Donati ER, Sand W (2007） Microbial processing of metal 71. Ben Farhat M, Farhat A, Bejar W, Kammoun R, Bouchaala K, sulfides. Springer, Netherlands Fourati A, Antoun H, Bejar S, Chouayekh H (2009) Characteri- 55. Adrio JL, Demain AL (2005) Microbial enzymes and biotrans- zation of the mineral phosphate solubilizing activity of Serratia formations. Humana Press Inc, Totowa marcescens CTM 50650 isolated from the phosphate mine of 56. Clark DS, Blanch HW (1997) Biochemical engineering, 2nd edn. Gafsa. Arch Microbiol 191:815-824 Taylor & Francis, London 57.1 Bonnarme P, Gillet B, Sepulchre AM, Role C, Beloeil JC, Ducrocq C (1995) Itaconate biosynthesis in Aspergillus terreus. J Bacteriol 177:3573-3578 Springer
10_1006_icar_2002_6837	chunk_1	Icarus 158, 98-105 (2002) doi:10.1006/icar.2002.6837 Hidden Mass in the Asteroid Belt G. A. Krasinsky, E. V. Pitjeva, M. V. Vasilyev, and E. I. Yagudina Institute ofApplied Astronomy,Russian Academy of Sciences,Kutuzov Quay 10,St.Petersburg 191187,Russia E-mail: kra @ quasar.ipa.nw.ru Received November 13, 2001; revised January 11, 2002 1. INTRODUCTION The total mass of the asteroid belt is estimated from an analysis of the motions of the major planets by processing high precision At present quite accurate measurements of ranging to the martian landers Vikings-1 and 2 and Pathfinder (with a typical measurements of ranging to the landers Viking-1, Viking-2, and Pathfinder (1976-1997). Modeling of the perturbing accelerations of error of about 7 m) are available. On this level of accuracy the the major planets accounts for individual contributions of 300 minor ephemeris of Mars is very sensitive to perturbations from many planets; the total contribution of all remaining small asteroids is minor planets. In the advanced ephemerides of major planets modeled as an acceleration caused by a solid ring in the ecliptic DE403 (Standish et al. 1995) and DE405 (Standish 1998) the plane. Mass Mring of the ring and its radius R are considered as perturbations from 300 asteroids have been taken into account, solve-for parameters. Masses of the 300 perturbing asteroids have which made it possible to process these measurements success- been derived from their published radii based mainly on measured fully. Nevertheless the total perturbations from other asteroids fluxes of radiation, making use of the corresponding densities. This do affect the ephemerides on a measurable level. A direct com- set of asteroids is grouped into three classes in accordance with putation of perturbations from all asteroids which have been physical properties and then corrections to the mean density for each class are estimated in the process of treating the observations. discovered up to the present epoch (in number about 130,000) In this way an improved system of masses of the perturbing asteroids cannot be reliably carried out due to the uncertainties of the has been derived. individual masses. Moreover it is natural to suppose that the to- The estimate Mring ≈ (5 ± 1) × 10-10 M。 is obtained (M。 is the tal mass of the asteroids which have not yet been discovered is solar mass) whose value is about one mass of Ceres. For the mean ra- also large enough to affect the ephemerides of the major planets dius of the ring we have R ≈ 2.80 AU with 3% uncertainty. Then the (though the individual perturbations caused by any one of them total mass Mbelt of the main asteroid belt (including the 300 aster- are negligible). Mostly these objects are too small to be observed oids mentioned above) may be derived: Mbelt (18 ± 2) × 10-10 M. from the Earth but it appears that the total hidden mass may be The value Mbelt includes masses of the asteroids which are al- detected if one studies their impact on the motion of Mars. ready discovered, and the total mass of a large number of small A prevailing part of these celestial bodies move in the asteroid asteroids—-most of which cannot be observed from the Earth. The second component Mring is the hidden mass in the asteroid belt as belt and their instantaneous positions may be considered homo- evaluated from its dynamical impact onto the motion of the major geneously distributed along the belt. Thus it seems reasonable to model the perturbations from the remaining small asteroids planets. Two parameters of a theoretical distribution of the number of (for which individual perturbations are not accounted for) by asteroids over their masses are evaluated by fitting to the impro- computing additional perturbing accelerations as being caused ved set of masses of the 300 asteroids (assuming that there is by a massive ring with a constant mass distribution in the eclip- extrapolated to the whole interval of asteroid masses and as a and radius) must be included in the set of solve-for parameters result the independent estimate Mbelt ≈ 18 × 10-10M。 is obtained while the lander ranging data are processed. In this paper results which is in excellent agreement with the dynamical finding given of such processing are described in detail. above. We have used the lunar-planetary integrator embedded in the These results make it possible to predict the total number of mi- program package ERA (Krasinsky and Vasilyev 1997). The inte- nor planets in any unit interval of absolute magnitude H. Such grator makes it possible to integrate simultaneously barycentric predictions are compared with the observed distribution; the com- parison shows that at present only about 10% of the asteroids with equations of motion of the nine major planets, Sun, and Moon, absolute magnitude H < 14 have been discovered (according to the equations of the lunar physical libration, and reduced equations derived distribution, about 130,000 such asteroids are expected of 300-350 minor planets. The mutual perturbations for the five t0 exist).@ 2002 Elsevier Science(USA) largest asteroids are accounted for; those for the other asteroids Key Words: asteroids; mass; planetary dynamics; radar. are neglected. In the present study the model has been improved by adding the perturbations from the homogeneous ecliptic ring. 98 0019-1035/02 $35.00 2002 Elsevier Science (USA) All rights reserved HIDDEN MASS IN THE ASTEROIDBELT 99 At the first step its mass has been set equal to zero, to be evaluated : IRAS was launched, which measured the infrared fuxes from a by fitting to the observations. large number of asteroids. For our analysis it is quite important to use the most accu- In the work by Krasinsky et al. (2001) we have compared the rate available values of the masses of the largest 300 perturbing masses obtained by a number of authors based on the method asteroids. In the next section the method used to derive these of the close asteroid encounters with the IRAS-based masses. It masses is outlined. appears that, except for the three largest asteroids (Ceres, Pallas, and Vesta) and seven other minor planets, the astrophysical II. MASSES OF 300 LARGEST ASTEROIDS method gives much more accurate estimates (at least by one order). In the cases where the dynamical estimates appear satis- There are two groups of methods that allow us to evaluate the factory they are in good agreement with IRAS data. Successful masses of asteroids. The first group is hereafter referred to as the dynamical determinations based on close encounters of minor astrophysical group. These methods are based on measurements planets have been produced only when very favorable conditions of the flux of radiation from the asteroid and on spectral obser- have been fulfilled (several encounters, long durations of the as- vations which give its spectral class. The important factor that trometric data, or very high accuracy of the data attained by the affects the flux is the radius of the asteroid; other factors may space telescope Hipparcos; see Bowell et al. 1994, Carpino and be modeled with sufficient accuracy (Morrison and Lebofsky Knezevic 1996, Goffin 1991, Krasinsky et al. 2001, Kuznetsov 1979). Having obtained the spectral class one can attribute a tax- 1999, Michalak 2000, 2001, Viateau and Rapaport 1998, Viateau onomic type to the asteroid; after that a corresponding density 2000, Viateau and Rapaport 2001). of the asteroid may be related to this type. With the known radius So the present study is based on the masses derived from IRAS and density, the mass of the asteroid may be easily derived. data. The astrophysical method being applied to the three largest Important additional information on asteroid radii is provided minor planets gives wrong results because these planets have by observations of occultations of stars by minor planets (Millis complicated internal structures and their mean densities cannot and Dunham 1989) and by radar observations of minor planets be restored reliably from their spectral classes.
10_1006_icar_2002_6837	chunk_2	Fortunately it ap- of the main belt (which have become a routine procedure-at pears possible to derive rather accurate masses of these asteroids present 30 asteroids of the main belt have been measured in this in the process of fitting the planetary ephemerides to the ranging way; see Magri et al. 1999). Comparison with the radar results observations from perturbations upon the orbit of Mars (Standish has confirmed the infrared astronomical satellite (IRAS) radii at and Hellings 1989, Standish 2000, Pitjeva 1997, Pitjeva 2001a). the 10% level except Asteroid 393 Lampetia for which the error Masses obtained in this way are in good agreement with a num- is about 30%. ber of dynamical estimate based on analysis of encounters with In the methods of the second group the mass of the asteroid other asteroids. has to be estimated from its perturbations upon the motion of At present 3316 radii, obtained by the astrophysical method, some other celestial body. These methods can be applied in the are published in the open NASA database (SBN) ("Small Bodies following cases: node of the NASA Planetary Data System" available at http:// pdssbn.astro.umd.edu). This set includes both the IRAS data 1. The perturbed body is another asteroid for which a close (1991 entries) and results of some ground observations. A tax- encounter with the perturbing body occurs; in this case conven- onomic code from Tholen (1989) is also given which allows tional ground-based astrometry is used. one to estimate the densities of 1098 asteroids. These data may 2. Space probes have very close encounters with some aster- be checked and appended by the radi and taxonomic codes for oids; as a result more sophisticated and precise onboard obser- about 300 asteroids from Bowell et al. (1979), Tedesco et al. vations allow very accurate and reliable estimates of the masses (1989), Howell et al. (1994), Xu et al. (1995), and Barucci of these asteroids. et al. (1997). We referred Tholen's taxonomic codes to the three 3. Several of the largest asteroids affect the motion of Mars compositional taxonomic types making use of the composi- so strongly that their masses can be estimated from an analysis tional interpretation of the asteroid taxonomy types after Bell of ranging to the martian landers. et al. (1989). These types are carbonaceous (C), silicate (S), and Until now masses of about 100 asteroids have been obtained metallic (M). The adopted correspondence is given in Table I. by dynamical methods with various levels of accuracy achi- When the ephemerides DE403/DE405 were constructed eved. Masses of the overwhelming majority of the other aster- (Standish et al. 1995, Standish 1998) mean a priori densities oids are too small to be determined by the dynamic methods s of each of the three types were used to calculate the perturbing in which ground-based astrometry is used; however, their to- accelerations from 297 asteroids, selected because they have a tal impact on the motion of Mars and Earth is not negligible. relatively large effect upon the motion of Mars. These densities Fortunately nowadays the astrophysical methods are greatly im- - have been revised in the process of fitting these ephemerides to proved and have provided a data set from which the masses may yobservational data in Standish (2000) (Table I, next line). The be estimated for about 2000 asteroids. A very important con- densities given in the last two lines of Table I have been derived tribution to this problem was made after the dedicated satellite e in a similar way from an independent analysis of practically 100 KRASINSKY ET AL. TABLE 1 10 Mo Correspondence of Tholen's Classes with Densities (in g cm-3) 15 C, D, P, T, S, K, Q, V, Tholen's classes B, G,F R, A,E M 12 DE405 Composition type C S M A priori density 1.8 2.4 5.0 Revised density, Standish (2000) 1.29 ± 0.06 2.71 ± 0.04 5.29 ± 0.53 Revised density, Pitjeva (2001a) 1.36 ± 0.03 2.67± 0.02 Revised density, this work 1.38 ± 0.02 2.71 ± 0.02 5.32± 0.07 6 Ceres 3 L the same observational data (measurements of distances to the landers and surfaces of the inner planets) while the numerical 口 1600 0 400 800 1200 2000 ephemerides EPM2000 are constructed (Pitjeva 2001a, Pitjeva 2001b) (Table I, line 4) and from the present analysis in which FIG. 1. Summing masses derived from the IRAS data. the perturbations from the solid ring are accounted for (Table I, line 5). We use 1o uncertainty for statistical errors. The errors of NEAR masses may affect only the last decimals The starting system of the masses of the minor planets was of the values given in Table II. One can see that the supposed that of DE405; the perturbing 297 asteroids and their radi and taxonomic classes were selected by J. G. Williams (1988, private error of IRAS masses for small asteroids is supported by the comparison with the NEAR results. communication to E. M. Standish, who kindly provided us with The mass distribution derived from the IRAS data, making these data). use of our estimates of the densities (the last line of Table I), At present with the available IRAS radi and Tholen's classes is presented in Fig. 1. The asteroids are ordered as their masses of the asteroids it seems possible to determine the masses for diminish (m1 > m2 > .·> mv） and the consequent sum of a larger set of the asteroids to compute the perturbations from m; in the units 10-10 Mo is depicted as a these bodies more accurately than has been computed in DE403/ DE405. So an attempt has been undertaken to account for per- function of N. The arrows on the curve mark the mass of Ceres and the total mass of the asteroids accounted for in DE405. One turbations from 351 asteroids making use of the published IRAS can see that a small but noticeable part of the asteroids which data. The results are controlled by comparison with the masses adopted in DE405, and by fitting corresponding planetary ephe- affect the motion of Mars had not yet been taken into account. merides to the lander ranging data and other measurements (see Section III) III.DYNAMICALESTIMATIONS OF THEMASS OF THE ASTEROIDBELT A priori estimates of the errors of IRAS-based radi vary in the range from 10% (for large asteroids) to 35% (for small asteroids). In our analysis of the ranging data the initial values of the It seems plausible that for any individual asteroid of known astronomical constants involved were taken from DE405. They spectral class the error of densities given in Table I may be as have been derived as a result of fitting to radiometric observations large as 20%. The corresponding total error of the individual during 1964-1994 presented on the Website of Commission mass may reach 30% for large asteroids and 100% for small 4 IAU (http://ssd.jpl.nasa.gov/iau-comm4/) without Pathfinder asteroids. and Mars Global Survey data; the cited above JPL interoffice In the following two cases the a priori estimate may be checked memoranda (Standish et al. 1995, Standish 1998) as well as by comparison with the masses of the minor planets, (433) Eros (Pitjeva 2001b) also may be found on this site. This data set was and (253) Mathilde, determined by the space mission NEAR to complemented with Russian measurements of ranging to the these asteroids (Yeomans et al. 1997, Yeomans et al. 1999). The NEAR results and those based on IRAS data are presented in inner planets (1961-1995) and the Pathfinder lander. For our aims the measurements of ranging to the martian landers Viking- Table II. 1, Viking-2, and Pathfinder are of a paramount importance. As was mentioned above the astrophysically derived masses TABLEII have allowed us to improve significantly the accuracy of the Masses of Eros and Mathilde in 10-10 M planetary ephemerides. Unfortunately, adding new perturbing asteroids to progress further appears to be not a routine work as (433) Eros (253) Mathilde our experiments have shown.
10_1006_icar_2002_6837	chunk_3	In fact the SBN database keeps two NEAR sets of radi: a set derived from IRAS observations (they are re- 0.00003362 0.00051938 ferred hereafter as the “"radiometric radi' or “system 1 of radi') IRAS 0.000058 0.000698 and a larger set from the IMPS Ground Data File ("system 2"). HIDDEN MASS IN THE ASTEROID BELT 101 So, when the radar measurements are processed several solu- tions could be obtained: 400 1. Solution 1: 300 asteroids with the masses estimated inde- pendently from the radiometric data set of system 1, corrected for scale. For 16 asteroids which have no radiometric radii in the 300- SBN database, the masses have been taken from the set used for DE405. 2. Solution 2: the same set of asteroids as in Solution 1 with simultaneous estimation of the mass and radius of the asteroid ring. 100- 3. Solution 3: 300 asteroids and the next largest 51 asteroids of the main belt for which radiometric radi are given in the SBN 400 database (system 1), plus a massive asteroid ring to represent the 100 200 300 500 smallest asteroids. FIG.2. The relation between radi: R, are the radiometric radi of System 4. Solution 4: 300 asteroids and another set of 51 asteroids of 1 and R are the radi of System 2 (in kilometers). the main belt with any SBN available radi with an estimation of the doubtful masses of five asteroids, plus the massive asteroid In the SBN there is no reference concerning these sets; informa- ring. tion about the sets is given in Tedesco (1992). Our analysis has shown that in most cases these two sets are in good agreement In producing these solutions the complete set of ranging ob- servations has been processed, corrections to the densities of the and only a slight scaling is needed to transform SBN radii R (in kilometers) to the radiometric radii R,, three classes (C, S, M) being included in the list of the solve-for parameters. Lines 1-3 of Table II present the rms o (in meters) Rr = (0.966 ± 0.001)R - (0.313 ± 0.027). (1) of the postfit residuals for the ranging observations of Viking-1, Viking-2, and Pathfinder. In the next six lines the estimated den- However, there are two families of minor planets (see Fig. 2) sities pc, ps, Pm (in g cm-3) of the three classes of asteroids for which the radi strongly disagree for the two sets. In both and the corresponding errors are given. these cases the linear dependence of R, on R again takes place The large error ±1.35 of pm for Solution 4 is due to strong but with very different values of the coefficients, correlations (up to 90%) with additional parameters of this so- lution (namely with corrections to the masses of the asteroids R, = (0.426± 0.013)R -(0.713 ±0.446) Polyhymnia, Angelina, Atala, Ismene, and Ludmila). In Solution 1 the mass of the ring was set equal to zero, in for the family B, and solutions 2-4 the mass of the ring (in 10-10 Mo) and its radius R, = (0.244 ± 0.008)R -(0.330 ± 0.138) TABLEIII for the family C. Impact of Minor Planets on Planetary Ephemerides, (o; in m, Pi In Fig. 2 the straight line A corresponds to relation (1); the in g cm-3, R in AU, M in 10-10 M) lines B and C present the linear dependence for the two other families. The families A, B, and C consist of 1555, 53, and 36 N 1 asteroids, respectively. It appears that all the asteroids from 0 of Viking-1 family B are of the same taxonomic class C, while those from 7.9 7.8 8.2 7.5 of Viking-1 family C belong to either S or M classes. It follows from Tedesco 5.6 5.4 5.5 5.7 0 of Pathfinder 3.1 3.0 8.5 3.6 (1992) that some asteroids of system 2 have default albedos, PC 1.38 1.34 1.12 1.74 and the lowest albedo 0.01, known for asteroids, has been used ±0.03 ±0.04 ±0.04 ±0.11 for them. As a result, radi of these asteroids are overstated (in ps 2.71 2.71 2.82 2.64 particular it is true for all asteroids of families B and C). Unfor- ±0.02 ±0.02 ±0.03 ±0.08 tunately several of the additional asteroids appear to belong to PM 5.32 5.30 5.35 4.21 ±0.07 ±0.13 ±0.30 ±1.35 these families and have somewhat uncertain estimates of their Rring 2.94 2.83 2.77 masses. In our analysis (Krasinsky et al. 2001) it has been shown ±0.06 ±0.06 ±0.06 that the best values of radii of the 300 asteroids (perturbations Mring 5.3 4.0 4.8 which are accounted for) correspond namely to system 2. Thus ±0.5 ±0.5 ±0.5 the radi given by system 1 must be corrected making use of N 300 300 351 351 Mtot 11.8 11.8 12.0 14.9 relation (1) obtained by comparing the two systems of radi for Msum 17.1 16.0 19.7 the asteroids of family A. 102 KRASINSKY ET AL (in AU) were included in the list of estimated parameters. Sta- 0"M tistically significant estimates have been obtained for both the 20 mass Mring (with the formal errors about 0.5 × 10-10 Mo) and mean radius Rring of the ring (with the formal error 0.06 AU). Mtot gives the sum of the masses of the asteroids from which perturbations were individually accounted for during integra- tion. From the dispersion of results for the different solutions it seems that the most plausible estimate of the mass Msum of all asteroids is Msum = (18 ± 2) × 10-10 Mo. (2) IV. DISTRIBUTION OF ASTEROID MASSES 25 50 75 100 To check this estimate we can apply a theoretical distribution 125 of the number N(r) of minor planets with radii exceeding r (the FIG. 3. Distribution of masses M(r) versus their radi r. M(r) is the total distribution is based on a collisional model of fragmentation mass of all asteroids whose radi are greater than r (in kilometers). described in Dohnanyi (1969), Hawkins (1959), and Samoilova- Yachontova (1973). For the density d N(r) of this distribution the In fact the fragmentation models predict for the power index following expression holds true taken as -3.5 in (3) values in the range from -3.53 to -3.47. We have constructed the mass distributions given by Eq. (4), dN(r) = -βr-3.5dr, 3) varying the power index for the lower and upper boundaries of this interval. The corresponding changes in the estimation of the where β > O is a constant. total mass M(0) appear to be about ±0.4 × 10-10 Mo, whose Let M(r) be the total mass of all asteroids with radi greater which is within the uncertainty of M(O). than r. Supposing that some mean density p may be used to calculate the masses of asteroids from their volumes, we obtain V. HIDDEN ASTEROID MASS after integration the expression for the distribution M(r) It seems useful to express the right part of distribution (3) M(r) =p πr3dN(r) = βr0.5 + βo, （4） in terms of the absolute magnitude H; thereafter it would be possible to compare it with all of the numbered asteroids. We apply the relation between radius r (in kilometers) of an asteroid with some constants βo and βi. The constants β and βi are and H connected by the relation logr = 3.1 - 0.2H, (7) 8π β1 =- (5) taken from Chebotarev and Shor (1976). Then instead of (3) we obtain Now we can evaluate the constants βo, βi by fitting the dis- tribution to the set of 300 asteroids studied above whose masses and radii provide the best data for such estimation. We assume that there are no significant systematic observational selection (8) effects in this region of change of r where the asteroids are large The parameter β in distribution (3) may be calculated now enough. Then we can extrapolate the derived distribution for with the help of Eq. (5), in which β1 = -0.829, as follows from r → 0 and compare M(0) with estimate (2). (6). Here the mean density p = 1.7 g cm-3 has to be expressed After fitting we have obtained for the value M(r) the following in the units 10-10 Mo km-3. With Mo = 1.99 x 1033 g we have expression (r is in kilometers; M is in 10-10 M): p = 0.0085 × 10-6. Thus we obtain M(r) = 17.65 - 0.829√r (6) β = 11.6 x 106. We see from this distribution that the total mass M(O) of the With this value for β distribution (3) shows that the expected asteroid belt is 17.6 × 10-10 Mo, whose in excellent agreement number of minor planets with radi of 1 km is 12 × 106 and with our finding based on the study of the perturbations upon the eit is about 9000 when the radi are in the range from 10 to orbit of Mars.
10_1006_icar_2002_6837	chunk_4	In Fig. 3 both the curve of this distribution and the 20 km. The typical mass of asteroids from the last subset is about experimental data (for about 2000 minor planets) are depicted. 0.0001 × 10-10 M。 with the total mass ≈1 x 10-10 M。 whose HIDDEN MASS IN THE ASTEROID BELT 103 value is not negligible if one processes the observations of the TABLE IV landers. For a unit interval of magnitudes, and in the logarithmic Expected (N,) and Observed (N。) Numbers scale, expression (8) becomes of Asteroids H No log dN ≈ 0.5H - 1.02. (9) Np 5-6 10 8 100 In Fig. 4 the line marked by the symbol “A" presents this 6-7 25 28 100 relation in the plane (log d N, H). The black circles are the values 7-8 101 89 100 obtained after the total number of minor planets in the intervals 8-9 210 280 70 of the magnitudes (H, H + 1) are calculated. (The experimental 9-10 363 900 40 data are taken from the data set at our disposal, which includes 10-11 583 3000 20 about 30,000 numbered asteroids). 11-12 1515 9000 16 12-13 4109 28000 14 One can see that for H < 8 the slope of the theoretical line 13-14 8014 90000 9 corresponds to the experimental dependence of log d N on H. 14-15 8290 280000 3 Assuming that there is no observational selection for H < 8, we can calibrate the dependence given by (9) by the experimental data in this region of magnitudes. Then instead of (9) we obtain about 130,000, and about 10% of such asteroids have already log dN ≈ 0.5H - 1.80. (10) been discovered. The line marked by the symbol “B" corresponds to this func- VI. CONCLUDING REMARKS tional dependence. It is interesting that the distribution (10) can be obtained after a small constant correction of magnitude 0.3 The main conclusions of this study may be summarized in the is applied to the starting dependence of log r on H (7). Then following way. instead of (7) we have to set There is evidence that in the asteroid belt a hidden mass exists that reveals itself by measurable effects on the motion of Mars. log r = 3.4 - 0.2H. Perturbing effects of a large number of the unobserved small Such a small correction is well within the uncertainty of rela- asteroids which contribute to this mass may be effectively de- tion (7) (for instance due to the adopted albedo or the photometric scribed by the model of a perturbing ring. The total mass of all system). asteroids of the asteroid belt (including all discovered asteroids) The data presented by Fig. 4 make it possible to estimate the is estimated as (18 ± 2) × 10-10 Mo. This estimate is in good expected number of minor planets in the asteroid belt which agreement with the theoretical distribution of asteroid masses have not yet been discovered in given intervals of absolute mag- given by fragmentation theory and based on values of parame- nitudes (see Table IV). In the last column of Table IV the ratio ters of the distribution derived from contemporary values of the of the discovered minor planets in the unit intervals of absolute masses of the 300 largest asteroids. A significant part of the mi- magnitude to the expected number of planets in this interval nor planets of the asteroid belt are too small to be observed from is given (in percent). One can see that the expected number of the Earth but their hidden masses produce measurable effects on asteroids of the main belt for which the magnitude H < 14 is the motion of Mars. In recent work (Jedicke and Metcalfe 1998) results of a new 18(N) global asteroid survey are discussed. It is assumed that the survey is complete to about absolute magnitude H ≈ 12 and argued that the power index in the distribution N(r) (Eq. (3)) is not constant and differs from the value given by the fragmentation theory of Dohnanyi (1969). If this interpretation of the survey is true then our prediction for the expected number of the minor planets (at least for smaller asteroids) seems to be overestimated. Moreover in this case our value of the summary mass of the asteroid belt (Eq. (2)) derived from the dynamical studies also may be suspected to be too large. The present work is probably a first attempt to estimate the hidden mass in the asteroid belt by direct analysis of very small perturbations in the motion of the major planets; in future new precise positional measurements provided by the ongoing NASA program of Mars exploration 13 will make it possible to essentially improve the accuracy and FIG. 4. Distribution of log(N) versus magnitudes H. reliability of the dynamical estimate of the hidden mass. 104 KRASINSKY ET AL. Indeed, after this work had been completed new measure- ACKNOWLEDGMENTS ments of distances to Mars derived from Mars Global Sur- We thank E. M. Standish and J. G. Williams (JPL) for granting necessary veyor satellite became available (see the above cited website information and helpful comments and V. A. Shor (IAA) for a useful discussion of Commission 4). New global solution with the included MGS of the results. 10-10 M。 for the total mass of the belt. These values only slightly differ from the results presented in this paper. REFERENCES Aleshkina, E. Yu., G. A. Krasinsky, and M. V. Vasilyev 1997. Analysis of LLR APPENDIX data by the program system ERA. IAU Coll. 165, 227-232. Barucci, M. A., M. T. Capria, A. Coaradini, and M. Fulchignoni 1997. Classi- Perturbing Force of the Asteroid Ring fications of asteroids using G-mode analysis.Icarus 72, 304-324. Bateman, H., and A. Eritelyi 1953. Higher Transcendental Functions, Vol. 3. Let r, v be the radius vector and orbital longitude of a perturbed body; rp, Up McGraw-Hill, New York. are those of a perturbing asteroid. The perturbing force F acting upon the body is given as Bell, J. F., D. R. Davis, W. K. Hartmann, and M. J. Gaffey 1989. Asteroids: The big picture. In Asteroids II (R. P. Binzel, T. Gehrels, and M. S. Matthews, F = -grad U, Eds.), Pp. 921-945. Univ. of Arizona Press, Tucson. Bowell, E., T. Gehrels, and B. Zellner 1979. Magnitudes, colors, types and where adopted diameters of the asteroids. In Asteroids (T. Gehrels, Ed.), pp. 1108- 1129. Univ. of Arizona Press, Tucson. 1 U=gm- Bowell, E., K. Muinonen, and L. H. Wassermann 1994. Asteroids mass determi- nation from multiple asteroids-asteroids encounters. In Abstracts for Small Bodies in the Solar System and Their Interactions with the Planets, p. 19. If the perturbing asteroids are distributed uniformly along a circular orbit in the Mariehamn, Finland. plane of the orbit of the perturbed body, then the force function may be presented Carpino, M., and Z. Knezevic 1996. Asteroid mass determination: (1) Ceres. as the integral IAU Symp. 172, 203-206. dup Chebotarev, G. A., and V. A. Shor 1976. The structure of the asteroid belt. Tr. U = gM 2π Inst. Teor Astron. 15, 60-90. [In Russian] √r2 -2rrpcos(v-vp)+r² J0 Dohnanyi, J. S. 1969. Collisional model of asteroids and their debris. J. Geophys. Res. 74, 10. where M is sum of the masses of the asteroids. We are interested in the case Goffin, E. 1991. The orbit of 203 Pompeja and the mass of Ceres. Astron. of the perturbations on the orbits of Mars and the Earth, so r < rp and the Astrophys. 249, 563-568. elliptical integral may be expressed in terms of the hypergeometric function F Hawkins, G. S. 1959. The relation between asteroids, fireballs and meteorites. (see Bateman and Erdelyi 1953) Astrophys. J. 64, 450. Howell, E. S., E. Merenyi, and L. A. Lebofsky 1994. Classification of asteroid spectra using a neural network. J. Geophys. Res. 99, 10,847-10,865. rp Jedicke, R., and T. S. Metcalfe 1998. The orbital and absolute magnitude distri- where α = r/rp and the standard notations are used: butions of main belt asteroids. Icarus 131, 245-260. Krasinsky, G. A., and M. V. Vasilyev 1997. ERA: Knowledge base for ephemeris and dynamical astronomy. IAU Coll. 165, 239-244. n=1 (c)n(1)n Krasinsky, G. A., E. V. Pitjeva, M. V. Vasilyev, and E. I. Yagudina 2001. Esti- (A.1) (q)n =q(q + 1)...(1 + n - 1). mating masses of asteroids. Comm. IAA RAN 139. Kuznetsov, V. 1999. On determination of masses of the largest asteroids. Tr. Inst. For the derivative needed to calculate the gradient the following expression may Appl. Astron. 4, 117-127. [In Russian] be applied: Magri, Ch., S. J. Ostro, K. D. Rosema, M. L. Thomas, D. L. Mitchell, D. B. Campbell, J. F. Chandler, I. I. Shapiro, J. D. Georgini, and D. K. Yeomans 1999.
10_1006_icar_2002_6837	chunk_5	Mainbelt asteroids: Results of Arecibo and Goldstone radar observa- tions of 37 objects during 1980-1995. Icarus 140, 379-407. Michalak, G. 2000. Determination of asteroid masses. I. (1) Ceres, (2) Pallas, Then for the perturbing force of the ring we obtain and (4) Vesta. Astron. Astrophys. 360, 363-374. Michalak, G. 2001. Determination of asteroid masses. I. (6) Hebe, (10) Hygiea, (15) Eunomia, (88) Thisbe, (444) Gyptis, (511) Davida, and (704) Interamnia. Astron. Astrophys. 374, 703-711. or Millis, R. L., and D. W. Dunham 1989. Precise measurement of asteroid sizes 1gM and shapes from occultations. In Asteroids II (R. P. Binzel, T. Gehrels, and rF (1.5, 1.5, 2;α²). M. S. Matthews, Eds.), pp. 148-170. Univ. of Arizona Press, Tucson. Morrison, D., and L. Lebofsky 1979. Radiometry of asteroids. In Asteroids The simplest way of calculating the hypergeometric function on the right of (T. Gehrels, Ed.), pp. 184-205. Univ. of Arizona Press, Tucson. this expression is the straightforward application of relation (A.1). In our case Pitjeva, E. V. 1997. The ephemerides of the inner planets from spacecraft range α2 ≈ 0.2 and the series quickly converges. data and radar observations 1961-1995. IAU Coll. 165, 251-256. HIDDEN MASS IN THE ASTEROIDBELT 105 Pitjeva, E. V. 2001a. Progress in the determination of some astronomical con- Tedesco, E. F., J. G. Williams, D. L. Matson, G. J. Veeder, J. C. Gradia, and stants from radiometric observations of planets and spacecraft. Astron. Astro- L. A. Lebofsky 1989. Three-parameter asteroid taxonomy classifications. In phys. 371, 760-765. Asteroids I (R. P. Binzel, T. Gehrels, and M. S. Matthews, Eds.), Pp. 1151- Pitjeva, E. V. 2001b. Modern numerical ephemerides of planets and importance 1161. Univ. of Arizona Press, Tucson. of ranging observations for their creation. Celest. Mech. Dynam. Astron. 80, Tedesco, E. F., G. J. Veeder, J. W. Fowler, and J. R. Chillemi 1992. IRAS Mi- No. 3-4, 249-271. nor Planet Survey. In Phillips Laboratory Technical Report PL-TR-92-2049, Samoilova-Yachontova, N. S., Ed. 1973. Minor Planets. Nauka, Moscow. [In p. 437. Hanscom Air-Force Base, MA. Russian] Tholen, D. J. 1989. Asteroid taxonomic classifications.In Asteroids II (R. P. Standish, M. E., and R. W. Hellings 1989. A determination of the masses of Binzel, T. Gehrels, and M. S. Matthews, Eds.), Pp. 1139-1150. Univ. of Ceres, Pallas, and Vesta from their perturbations upon the orbit of Mars. Arizona Press, Tucson. Icarus 80, 326-333. Viateau, B., and M. Rapaport 1998. The mass of (1) Ceres from its gravitational Standish, E. M., X. X. Newhall, J. G. Williams, and W. M. Folkner 1995. JPL perturbations on the orbits of 9 asteroids. Astron. Astrophys. 334, 729-735. planetary and lunar ephemerides, DE403/LE403. In Interoffice Memorandum Viateau, B. 2000. Mass and density of asteroids of (16) Psyche and (121) 314.10-127, pp. 1-22. Hermione. Astron. Astrophys. 354, 725-731. Standish, E. M. 1998. JPL planetary and lunar ephemerides, DE405/LE405. In Viateau, B., and M. Rapaport 2001. Mass and density of asteroids of (4) Vesta Interoffice Memorandum 312.F-98-048, pp. 1-18. and (11) Parthenope. Astron. Astrophys. 370, 602-609. Standish, E. M. 2000. Recommendation of DE405 for the 2001 Mars Sur- Xu, S., R. P. Binzel, T. H. Burbine, and S. J. Bus 1995. Small main-belt asteroid veyor mission and for Cassini. In Interoffice Memorandum 312. F-00-107, spectroscopic survey: Initial results. Icarus 115, 1-35. Pp. 1-7. Yeomans, D. K., and 12 colleagues 1997. Estimating the mass of Asteroid 253 Tedesco, E. F. 1989. Asteroid magnitudes, UBV colors, and IRAS albedos and Mathilde from tracking data during the NEAR flyby. Science 278, 2106-2109. diameters. In Asteroids II (R. P. Binzel, T. Gehrels, and M. S. Matthews, Eds.), Yeomans, D. K., and 14 colleagues 1999. Estimating the mass of Asteroid 433 Pp. 1090-1138. Univ. of Arizona Press, Tucson. Eros during NEAR spacecraft flyby. Science 285, 560-561.
10_1016_j_mineng_2005_10_024	chunk_1	Available online at www.sciencedirect.com SCIENCI )DIRECT MINERALS ENGINEERING ELSEVIER Minerals Engineering 19 (2006) 960-967 This article is also available online at: www.elsevier.com/locate/mineng Formation of jarosite during Fe2+ oxidation by Acidithiobacillus ferrooxidans J. Daoud, D. Karamanev * Department of Chemical and Biochemical Engineering, University of Western Ontario, London, Ont.,Canada N6G 5B9 Received 8 September 2005; accepted 26 October 2005 Available online 20 December 2005 Abstract Jarosite precipitation is a very important phenomenon that is observed in many bacterial cultures. In many applications involving Acidithiobacillus ferrooxidans, like coal desulphurization and bioleaching, it is crucial to minimize jarosite formation in order to increase efficiency. The formation of jarosite during the oxidation of ferrous iron by free suspended cells of A. ferrooxidans was studied. The pro- cess was studied as a function of time, pH and temperature. The main parameter affecting the jarosite formation was pH. Several exper- iments yielded results showing oxidation rates as high as 0.181-0.194 g/L h, with low jarosite precipitation of 0.0125-0.0209 g at conditions of pH 1.6-1.7 with an operating temperature of 35 °C. 2005 Elsevier Ltd. All rights reserved. Keywords: Bacteria; Bioleaching; Biooxidation; Reaction kinetics 1. Introduction is ferrous sulfate. The overall biochemical reaction of the oxidation of ferrous ions is Acidithioobacillus ferrooxidans, recently renamed from 4Fe2+ + O2 + 4H+ 4 4Fe3+ + 2H2O Thiobacillus ferrooxidans, is a gram-negative bacterium, (1) characterized by non-sporulating, rods, 0.5-0.6 μm wide This bioreaction is of great practical importance. by 1.0-2.0 μm long, with rounded ends, occurring singly Ore bioleaching involves the dissolution of metal sul- or in pairs, rarely in short chains. They are also known fides (MeS) by presumably an indirect mechanism, involv- to be motile by means of a single polar flagellum (Jensen ing the oxidation of Fe2+ to Fe3+ by A. ferrooxidans (1) in and Webb, 1995). A. ferrooxidans is found naturally in acid the liquid phase. Following this oxidation, the Fe3+ leaches mine drainage waters of iron and bituminous coal mines and is a dominant organism in biohydrometallurgy in the the metal through a chemical reaction (Jensen and Webb, 1995) process of ore bioleaching (Jensen and Webb, 1995). Fur- thermore, it has been utilized in the processes of desulphu- 2Fe3+ + MeS →Me2+ + S° + 2Fe2+ (2) rization of sour gases, treatment of acid mine drainage, and the desulphurization of coal. Hydrogen sulfide is usually a component of many mod- A. ferrooxidans can obtain energy from the oxidation of ern day gas streams such as biogas. It is a highly toxic com- different inorganic substances, the most common of which ponent and is usually removed by means of hydrogen sulfide scrubbers and other mechanisms like the Claus pro- cess. A. ferrooxidans can be used to eliminate hydrogen sul- fide from gas streams. This process first involves contacting Corresponding author. Tel.: +1 519 661 2111x88230; fax: +1 519 661 3498. H2S-containing gases with a solution of ferric sulfate in E-mail address: dkaramanev@eng.uwo.ca (D. Karamanev). an absorber. The solution absorbs H2S and oxidizes it to 0892-6875/$ - see front matter ? 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2005.10.024 J.Daoud, D.Karamanev / Minerals Engineering 19（2006）960-967 961 elemental sulfur, while the ferric sulfate is reduced to fer- 1.1. Jarositeformation rous sulfate (Nemati et al., 1998). This is shown in the fol- lowing reaction: A. ferrooxidans is commonly grown on 9K medium developed by Silverman and Lundgren. The ferrous iron H2S + Fe2(SO4)3 →S + 2FeSO4 + H2SO4 (3) oxidation occurs via reaction (1). This reaction goes to completion very rapidly and the Since there is consumption of hydrogen ions, the pH of hydrogen disulfde is removed effciently. Ferrous sulfate the liquid media initially increases.] However, this pH obtained from the first step is further treated by A. ferroox- increase is counteracted by the hydrolysis of ferric iron idans, according to reaction 1 in order to oxidize ferrous Fe3+ + H2O <→ FeOH²+ + H+ (4) iron to ferric iron in a bioreactor. This ferric iron obtained Fe3+ + 2H2O <→ Fe(OH) + H+ is recycled to the absorber for the first step and the cycle is (5) repeated (Nemati et al., 1998). Fe3+ + 3H2O <> Fe(OH) + H (6) The ability of A. ferrooxidans to oxidize ferrous ions is Therefore, it is quite visible that the pH of the system used in the treatment of acid mine drainage. In acid mine drainage, the main source of pollution is ferrous iron, with has effect on the extent of the oxidation and hydrolysis concentrations of up to 1 kg/m? (Jensen and Webb, 1995). reactions. Furthermore, there is a reaction in competition In order to treat this problem, the oxidation of Fe2+ with the hydrolysis reaction giving products of basic ferric to Fe3+ (by A. ferrooxidans) must precede the precipitation hydroxysulphates with the formula MFe3(SO4)2(OH)6 where M =K+, Na, NH, Ag*, or HsO+ (Jensen and of ferric iron as various insoluble oxyhydroxides (Jensen Webb, 1995). These hydroxysulphate precipitates are and Webb, 1995). In recent studies, A. ferrooxidans has known as jarosites. The following is the formula for jaro- proven to be efficient for the bacterial oxidation of ferrous site precipitation: iron in acid mine drainage. Another application of the iron oxidation by A. ferroox- 3Fe3+ + M+ + 2HSO +6H2O →MFe;(SO4)2(OH) + 8H+ idans is the desulfurization of coal (Juszczak et al., 1995). It (7) is based on the oxidation and dissolution of metal sulfides, which are naturally present in coal. Since the 9K medium contains a high concentration of There are several factors that play a role in the rate of NH ions, the jarosites produced were ammoniojarosites oxidation of ferrous ions by A. ferrooxidans. These factors with the formula NH4Fe3(SO4)2(OH)6. include ferrous/ferric iron concentration, cell and oxygen concentrations, pH, temperature, and reactor type. Past 1.1.1. Effect of jarosites studies have concluded that ferric ions competitively inhibit Jarosite formation has negative effects on many applica- ferrous ion oxidation by A. ferrooxidans; an inhibitory tions that require the use of A. ferrooxidans, especially in the effect that can be reduced by increasing cell concentration process of biological gas desulphurization. Some of the (Nyavor et al., 1996). effects include the diminishment of ferric iron used as Moreover, the pH and temperature have significant the absorbent for hydrogen disulfide, blockage of pumps, effects on the oxidation kinetics of iron by A. ferrooxidans. valves, pipes, etc., and the creation of kinetic barriers due Several studies have been conducted for the purpose of find- to the small diffusion of reactants and products through ing the optimum pH and temperature ranges for A. ferroox- the precipitation zone (Jensen and Webb, 1995). idans operation. Some findings are as shown in Table 1. The precipitation of jarosite is widely used in the zinc Therefore, the pH and temperature at which the bacte- industry to remove iron solubilized in the processing circuit rial operation and oxidation rate are at a maximum are (Dutrizac, 1999). The precipitation reaction is greatly approximately 2.0 and 30 °C, respectively. accelerated by the presence of jarosite seed, and the rate increases in a nearly linear manner with increasing seed additions (Dutrizac, 1999). Table 1 Also, in the process of coal desulfurization, the forma- Optimum pH and temperatures for A. ferrooxidans as reported by tion of jarosite on the surface of the biooxidized metal sul- different sources fide particle significantly decreases the rate of bioleaching Source Optimum pH by deactivating the surface. Jarosite formation in coal Karamanev and Nikolov (1988) 2.0 desulfurization results in residual sulfur, which cannot be Torma (1977) 2.3 removed, from coal.
10_1016_j_mineng_2005_10_024	chunk_2	Furthermore, jarosite formation in Smith et al. (1988) 2.0-2.3 immobilization matrices limits the amount of biomass Drobner et al. (1990) 2.0 retention since ferric iron deposits occupy most of the Optimum temperature (°C) available space (Jensen and Webb, 1995). Ahonen and Tuovinen (1989) 28 Okereke and Stevens (1991) 30 1.1.2. Chemistry of jarosites Smith et al. (1988) 25-30 As mentioned above, due to the abundance of NH Nemati (1996) 35 ions in the 9K medium, the jarosites produced in our 962 J.Daoud, D.Karamanev / Minerals Engineering 19（2006）960-967 experiments most closely corresponds to ammoniojaro- desired pH in each flask was obtained by adding 100% sites, NH4Fe3(SO4)2(OH)6 H2SO4 drop wise, with continuous agitation and pH mea- A recent study has been conducted in order to character- surement. Further, 20 ml of A. ferrooxidans inoculum was ize the composition of different types of jarosite produced added to each flask. The inoculum contained an average of biologically by using X-ray diffraction. It was found that 108 cells per milliliter. The flasks were then covered with ammoniojarosite has an elemental weight composition of pieces of aluminum foil. Finally, the flasks were placed in 14.6% NH, 29.1% Fe, and 11.2% S (Sasaki and Konno, the rotary shaker at ambient temperature with a rotation 2000). speed of 260 rpm. The Fe3+ and Fetotal concentrations in Several studies have been conducted in order to deter- the flasks are measured initially, at 22 h and at 46 h. The mine the importance of jarosite formation in bioreactors. ferrous and ferric iron concentrations were measured using For example, jarosite precipitation has been directly related sulfosalycilic acid as an indicator (Karamanev et al., 2002). to the number of attached cells (Pogliani and Donati, When more than 90% of Fe?+ was oxidized, the flasks 2000). However, no studies have dealt directly with deter- were removed from the shaker in order to measure the jaro- mining the amount of jarosite formation under varying cul- site precipitation. Firstly, the liquid in each fask was fil- tivation condition. The main goal of this study 1to tered using a vacuum flask and filter paper (pore size investigate jarosite precipitation under different conditions, 25 μm). The solids on the filter paper were then returned by varying pH and temperature. The results will be of sig- back to the corresponding flask by washing them with dis- nificant importance for determining the cultivation condi- tilled water. This was done in order to combine the filtered tions for minimal jarosite precipitation. jarosite and that attached to the wall of the flask. Approx- imately 10 ml of 50% H2SO4 were then added to each jaro- 2. Materials and methods site and distilled water mixture in order to dissolve the jarosite (suspended and on the walls). Finally, the total iron 2.1. Materials used concentration was measured using the sulfosalycilic method and the volume of mixture in each flask was mea- 2.1.1. Equipment sured using graduated cylinders in order to obtain the total For our procedure, the bacteria were grown in mass of jarosite present 11 × 250 ml Erlenmeyer flasks with the appropriate med- ium. The pH of each trial was adjusted using sulfuric acid 2.2.2. Temperature and monitored using a pH meter (Orion Model No. 420A). For the second part of the experiment, the appropriate The bacteria were allowed to grow in a Rotary fask shaker pH range obtained from the first part, displaying the opti- with speed and temperature adjustment (New Brunswick mum pH, was selected in order to test for the optimum Scientific Model No. G-25 R). For the ferric and total iron temperature. This allowed us to obtain the optimum pH analyses, we used a spectrophotometer (Varian Cary 50) and temperature combination in which the bacterial with the appropriate procedure (Karamanev et al., 2002). iron oxidation rate is maximal with minimum jarosite Lastly, filter paper with pore size 25 μm was used to sepa- formation. rate the jarosite produced in each fask. To start this part of the experiment, we selected the appropriate optimal pH, with minimal jarosite precipita- 2.1.2. Chemicals tion, from the first part of our experiment. We then selected All the chemicals used in this study were of analytical the appropriate number of 250 ml flasks in order to have grade, including 100% H2SO4, components of the 9K med- pH intervals of 0.1. We then followed the same procedure ium, sulfosalicylic acid and ammonium hydroxide (30%). as the first part except we set the flask shaker temperature consecutively to 25 °℃, 30 °C, 35 °℃, and 40 °℃. 2.2. Experimental method The mean ferrous iron oxidation rate (in g/L h) was determined by subtracting the Fe3+ concentration at 22 h This experiment deals with the effects of two parameters, from the initial, and dividing the result by 22 h. This gives pH and temperature, on iron oxidation and jarosite forma- more accurate results since the oxidation does not go to tion. Therefore, the experiment was divided into two parts, completion in any of the samples at 22 h, thus the mean dealing with varying pH and temperature. oxidation rate can be accurately compared for all the sam- ples at this time. 2.2.1. pH For the first part of the experiment, the effect of varying 2.3. Analytical procedures pH on jarosite formation was tested in order to find the optimal range of pH in which the bacteria operate effi- The analysis of ferrous and ferric iron concentrations at ciently with minimal jarosite precipitation. different times in our bacterial samples was done using a Eleven 250 ml Erlenmeyer flasks were used, one for each precise quantitative method, which is not affected by the pH in the range of 1.0-3.0 in intervals of 0.2 pH. 100 ml of presence of iron or A. ferrooxidans in solution. The method 9K medium was added into each of the 11 flasks. The utilizes a spectrophotometer for the colorimetric measure- J.Daoud, D.Karamanev / Minerals Engineering 19（2006）960-967 963 Table 2 Observations of Fe2+ oxidation at various pHs pH Time (h) 0 22 46 1.00 Light yellow Light yellowish lime Light yellowish lime with no visible solids on wall or suspended 1.20 Darker yellow Darker yellowish lime Darker yellowish lime with no visible solids on wall or suspended 1.39 Darker yellow Darker yellowish lime Dark yellowish lime with no visible solids on wall or suspended 1.60 Darker yellow Dark yellowish lime Light orange pekoe with no visible solids on wall or suspended 1.80 Dark yellow Light orange pekoe Darker orange pekoe with some residue on walls but no suspended visible solids 2.01 Light orange Darker orange pekoe Darker orange pekoe with some whitish muck and visible solids on wall and suspended 2.19 Darker orange Darker orange pekoe Darker orange pekoe with more whitish muck and more visible solids on wall and suspended 2.41 Darker orange Darker orange pekoe Darker orange pekoe with more whitish muck and more visible solids on wall and suspended 2.58 Darker orange Darker orange pekoe Darker orange pekoe with more whitish muck and more visible solids on wall and suspended 2.81 Darker orange Darker orange pekoe Darker orange pekoe with more whitish muck and more visible solids on wall and suspended 2.96 Dark orange Dark orange pekoe Dark orange pekoe with most whitish muck and most visible solids on wall and suspended ment of red-colored ferric-sulfosalicylate complex, which 100 90 then added, causing the 5-sulfosalicylic acid (SSA) to form 80 a yellow complex with all the iron ions, which gives the concentration of total iron in solution (Karamanev et al., 70 2002). 60- 50- 3. Results and discussions 40 % 30 22hrs This experiment consisted of two parts analyzing the 46hrs effect of pH and temperature on jarosite precipitation. 20 For each parameter, we will discuss the data and 10- observations. 0 0.5 1.5 2 2.5 3 3.5 3.1. pH pH Fig. 1. Percent completion of Fe²+ oxidation vs. pH at different times.
10_1016_j_mineng_2005_10_024	chunk_3	Several experimental runs were performed testing the effect of the various pHs, range from 1.0 to 3.0 in 0.2 inter- vals, taking samples initially, at 22 h and at 46 h. The tem- oxidation progression with time for the various pHs are perature was kept constant at 22 °C. The observations of given in Table 2. 0.14 0.14 0.12 0.12 0.1 0.1 pa 0.08 0.08 P Rate .06 三 0.04 0.04 0.02 0.02 0 0 0.5 1.5 2 2.5 3 3.5 pH Fig. 2. Comparison of jarosite produced and oxidation rate vs. pH at 46 h. 964 J.Daoud, D.Karamanev 丨 Minerals Engineering 19（2006）960-967 Therefore, the jarosite formation was only observed at temperature of 22 °C, pH of 1.6-2.0 produced the least the 46-h period with jarosites precipitating on the walls, amount of jarosite (between 0 and 0.025 g/L h) while still in suspension with the fuid and the bottom of the fasks, maintaining high Fe2+ oxidation rates of up to 0.127 g/L h. possessing a yellowish-brown color. The 2.96 pH flask was found to have the most precipitation of jarosites while 3.2.Temperature fasksof pH 1.0-1.6　displayed no　visible jarosite precipitation. The optimal pH at room temperature was determined to From Fig. 1, it can be seen that over the course of the be in the range of 1.6-2.0. Using these pH values in 0.1 experiment, conducted at room temperature of 22 °C, A. intervals and varying the temperature to 25, 30, 35, and ferrooxidans reached a degree of oxidation as high as 40 °C, the optimal pH and temperature combination was 100%. Therefore, the mass of jarosite was examined at investigated. This will give the pH and temperature opti- the end of the 46-h period in which the bacteria efficiently mums for A. ferrooxidans with minimal jarosite deposition. oxidized most of Fe2+ into Fe3+ Figs. 3-6 illustrate the relationship between oxidation Fig. 2 shows further analysis of jarosite mass produced rate of A. ferrooxidans and jarosite mass as a function of at 46 h and oxidation rate of A. ferrooxidans. At the room pH at different temperatures of 25, 30, 35, and 40 °C. At 0.186 0.06 0.184 0.05 0.182 0.18 0.176 !XO 0.174 01 0.172 0.17 1.2 1.4 1.6 1.8 2.2 pH Fig. 3. Oxidation rate and jarosite mass vs. pH at 25 °C. 0.22 0.25 0.215 0.2 7/6) 0.21 0.15 0.205 !XO 宣 0.2 0.195 1.2 1.4 1.6 1.8 2 2.2 pH Fig. 4. Oxidation rate and jarosite mass vs. pH at 30 °C. J.Daoud, D.Karamanev / Minerals Engineering 19（2006）960-967 965 0.215 0.18 0.21 0.16 0.14 0.205 0.12 7/6) 0.2 0.1 0.195 0.08 PIXO 0.19 0.06 0.185 0.04 0.18 0.02 0.175 1.2 1.4 1.6 1.8 2 2.2 pH Fig. 5. Oxidation rate and jarosite mass vs. pH at 35 °C. 0.175 0.08 0.07 0.17 0.06 0.05 osite uogepix 0.16 0.04 ass 0.03 0.155 0.02 0.15 0.01 0.145 1.2 1.4 1.6 1.8 2 2.2 pH Fig.6.Oxidation rate and jarosite mass vs. pH at 40 °C temperatures of 25 and 30 °C, the oxidation rate and the of 0.181-0.194 g/L h with low jarosite mass of 0.0125 jarosite produced curves have opposite slopes at pH below 0.0209 g. The oxidation rates attained at these conditions 1.9-2.0. However, at temperatures of 35 and 40 °C, oxida- were consistent with values obtained by Das et al. at similar tion rate and jarosite mass both increase as pH increases. conditions giving an oxidation rate of 0.15 g/L h with a The pH yielding reasonable oxidation rates with little jaro- reaction rate constant of 0.065 h-1 (Das et al., 1998). site precipitation for all temperatures, was between 1.6 and 1.7. Figs. 7 and 8 further show the effect of temperature 4. Conclusions and pH on the jarosite mass. From the analysis of Figs. 7 and 8 at pH of 1.6 and 1.7, The main purpose of this work was to investigate the respectively, it is clear that the Fe?+ oxidation rate and the Operating conditions for the Fe?+ oxidation by A. ferroox- jarosite produced are directly related, reaching a peak at idans, under which minimal amount of jarosite precipitate 30 °C. The optimum operation temperature in both cases, is produced, while the Fe2+ and rates are still high with however, was 35 °C, yielding relatively high oxidation rates respect to pH and temperature. From the experiments, 966 J.Daoud, D.Karamanev丨 Minerals Engineering 19（2006）960-967 0.25 0.02 0.018 0.016 0.014 7/6) 0.15 0.012 OS 0.01 Mass epixo 0.008 0.006 0.05 0.004 0.002 20 25 30 35 40 15 Temperature (°C) Fig. 7. Oxidation rate and jarosite mass vs. temperature at pH 1.6. 0.25 .045 D.04 D.025 osi ite 0.1 ass D.015 0.05 20 25 30 35 Temperature (°C) Fig. 8. Oxidation rate and jarosite mass vs.temperature at pH 1.7 we concluded that the optimal pH and temperature combi- References nation giving the least amount of jarosite is pH 1.6-1.7 at a temperature of 35 °C. These conditions yielded very Ahonen, L., Tuovinen, O.H., 1989. Microbiological oxidation of ferrous appealing results, giving oxidation rates of 0.181-0.194 g/ iron at low temperatures. Applied and Environmental Microbiology 55 L h, with small jarosite precipitation of 0.0125-0.0209 g. (2),312-316. Das, T., Panchanadikar, V.V., Chaudhury, G.R., 1998. Bio-oxidation of iron using Thiobacillus ferrooxidans. World Journal of Microbiology & Acknowledgements Biotechnology 14 (2), 297-298. Drobner, E., Huber, H., Stetter, K.O., 1990. Thiobacillus ferrooxidans, a facultative hydrogen oxidizer. Applied and Environmental Microbiol- We would like to thank the National Research Council ogy 56 (9), 2922-2923. of Canada (NSERC) and the University of Western Ontar- Dutrizac, J.E., 1999. The effectiveness of jarosite species for precipitating io for providing the financial support for this research. sodium jarosite. Jom 51 (12), 30-32. J.Daoud, D.Karamanev / Minerals Engineering 19（2006）960-967 967 Jensen, A.B., Webb, C., 1995. Ferrous sulfate oxidation using Thiobacillus the kinetic aspects. Biochemical Engineering Journal 1 (3), 171- ferrooxidans: a review. Process Biochemistry (Oxford) 30 (3), 225- 190. 236. Nyavor, K., Egiebor, N.O., Fedorak, P.M., 1996. The effect of ferric ion Juszczak, A.,Domka,F., Kozlowski, M., Wachowska, H., 1995. on the rate of ferrous oxidation by Thiobacillus ferrooxidans. Applied Microbial desulfurization of coal with Thiobacillus ferrooxidans Microbiology and Biotechnology 45 (5), 688-691. bacteria. Fuel 74 (5), 725-728. Okereke, A., Stevens Jr., S.E., 1991. Kinetics of iron oxidation by Karamanev, D.G., Nikolov, L.N., 1988. Influence of some physicochem- Thiobacillus ferrooxidans. Applied and Environmental Microbiology ical parameters on bacterial activity of biofilm: ferrous iron oxidation 57 (4), 1052-1056. by Thiobacillus ferrooxidans. Biotechnology and Bioengineering 31 (4), Pogliani, C., Donati, E., 2000. Immobilisation of Thiobacilus ferrooxi- 295-299. dans: importance of jarosite precipitation. Process Biochemistry Karamanev, D.G., Nikolov, L.N., Mamatarkova, V., 2002. Rapid (Oxford) 35 (9), 997-1004. simultaneous quantitative determination of ferric and ferrous ions in Sasaki, K., Konno, H., 2000. Morphology of jarosite-group compounds drainage waters and similar solutions. Minerals Engineering 15 (5), precipitated from biologically and chemically oxidized Fe ions. 341-346. Canadian Mineralogist 38 (1), 45-56. Nemati, M., 1996. An evaluation of ferrous sulphate oxidation, using Smith, J.R., Luthy, R.G., Middleton, A.C., 1988. Microbial ferrous iron immobilised Thiobacillus ferrooxidans. PhD Thesis, University of oxidation in acidic solution. JournalWater Pollution Control Fed- Manchester Institute of Science and Technology, Manchester, UK. eration 60 (4), 518-530. Nemati, M., Harrison, S.T.L., Hansford, G.S., Webb, C., 1998. Biological Torma, A.E., 1977. Advances in Biochemical Engineering, Vol. 6: New oxidation of ferrous sulfate by Thiobacillus ferrooxidans: a review on Substrates.
10_1016_S0168-1656_02_00222-5	chunk_1	ELSEVIER Journal of Biotechnology 99 (2002) 319-330 www.elsevier.com/locate/jbiotec MELISSA: a loop of interconnected bioreactors to develop life support in Space F. Godia a, J. Albiola, J.L. Montesinosa, J. Péreza, N. Creus ?, F. Cabello a, X. Menguala, A. Montras a, Ch. Lasseur b aDepartament d'Enginyeria Quimica,Escola Tecnica Superior d'Enginyeria,Universitat Autonoma de Barcelona, Bellaterra 08193Barcelona,Spain European Space Research and Technology Center, European Space Agency, PO Box 29,2200 AG Noordwijk, The Netherlands Received 9 July 2001; received in revised form 2 January 2002; accepted 16 January 2002 Abstract The development of a loop of interconnected continuous bioreactors, aimed to provide life support in space, is reported. The complete loop concept consists of four bioreactors and one higher plant compartment. For its realization the continuous and controlled operation of the bioreactors is characterized, up to the pilot scale level, first for each individual reactor, second for the interconnected reactor operation. The results obtained with the two more advanced bioreactors in the Micro Ecological Life Support System Alternative (MELISSA) loop are described more specifically. These reactors consist of a packed-bed reactor working with an immobilized co-culture of Nitrosomonas and Nitrobacter cells, and an external loop gas-lift photobioreactor for the culture of the cyanobacteria Spirulina platensis Their individual operation for long duration runs has been achieved and characterized, and their interconnected operation at pilot scale is reported. 2002 Elsevier Science B.V. All rights reserved. Keywords: Advanced life support; Closed ecological systems; Packed-bed bioreactors; Gas-lift bioreactors; Continuous operation 1. Introduction tant needs. One of them is, obviously, life support for the crew. Basically, the term life support The realization of long term manned Space comprises four main functions (Tamponnet and missions in the future, such as the establishment Savage, 1994): atmosphere regeneration for re- of planetary bases, for example on the Moon or spiration, water recycling, waste treatment and Mars, requires a number of critical technologies to generation of food. For these kind of interplane- be developed in order to supply the most impor- tary missions, involving several crew members and long distances, resupply of water, food and oxygen from Earth is not feasible, in technical and economical terms. For this reason, any life support Corresponding author. Tel.: +34-3-581-1018; fax: +34-3- 581-2013 system must be as much regenerative as possible. E-mail address: francesc.godia@uab.es (F. Godia). Currently, physical-chemical systems are well 0168-1656/02/$ - see front matter C 2002 Elsevier Science B.V. All rights reserved. PII: S0168-1656(02)00222-5 320 F.Godia et al./Journal ofBiotechnology 99(2002）319-330 developed for regeneration of oxygen from CO2 able in a reasonable residence time. The species for respiration (Gustavino et al., 1994; Eckart, responsibleof this transformation are a mixed 1994), as well as for the recycling of water culture of various strains isolated from human (Tamponet et al., 1999). However, the generation intestinal flora. The fatty acids are fed into of food is not feasible unless biological systems are compartment II where they are transformed in incorporated to life support systems. biomass, using the anaerobic photosynthetic bac- Micro Ecological Life Support System Alter- teria Rhodospirillum rubrum. Further conditioning native (MELISSA) is a model system for an of the liquid medium is done in compartment III, a advanced life support system based on different reactor with a co-culture of Nitrosomonas euro- microbial species and higher plants. It consists of a paea and Nitrobacter winogradsky cells, where the loop of interconnected bioreactors and a higher transformation of the nitrogen source from am- plant chamber, each one with a specific biotrans- monium ions to nitrate (a more easily assimilable formation task to fulfill. A general view of the nitrogen source for the following compartments) is MELISSA concept, is given schematically in Fig. achieved. Finally, compartment IVa is devoted to 1. The MELISSA system is fostered by the the culture of the cyanobacteria Spirulina platen- European Space Agency, it was originated in sis, and compartment IVb is used for higher plants 1988 (Mergeay et al., 1988), and its development growth. In them O2 is generated, and edible involvestheeffortofdifferentresearchteams biomass is produced, using the CO2 generated by (Lasseur et al., 2000). Compartment I is based the crew and other compartments, and light as on anaerobic termophilic fermentation for the energy source. Both photosynthetic microorgan- degradation of the waste material, basically or- isms can be used as edible material (Ciferri, 1983; ganic waste from the crew and non-edible parts of Vrati, 1984), and indeed Spirulina cells are already the higher plants, providing at its outlet a mixture used for human consumption. of volatile fatty acids, and CO2 and a low The development of such a complex system is percentage of the waste material hardly metaboliz- approached in a stepwise procedure. First, each Biomass CREW Wastes CO2 Wastes Compartment IVa Compartment I Compartment IVb Biomass Photoautotrophic bacteria Higher plant compartment Thermophilic anaerobic Spirulina platensis bacteria Volatile CO2 fatty acids NO CO2 Compartment III Compartment I NH4+ Photoauto/ hetrotrophic Nitrifying bacteria bacteria NitrosomonasNitrobacter Rhodospirillumrubrum NH4+ Fig. 1. General concept of the MELISSA loop for advanced life support, based on a series of four biological reactors and one higher plant compartment operating in continuous mode. F. Godia et al. I Journal of Biotechnology 99 (2002）319-330 321 compartment is studied individually. This includes (NH4)2SO4, 0.0025 g FeSO4·7H2O, 4×10-6 a number of tasks ranging from basic studies on CuSO4·5H2O, 0.71 g Na2HPO4, 0.68 g selection and characterization of strains, influence KH2PO4, 0.177 g (NH4)M07O24 · 4H2O, 4.3 x 10-6 g ZnSO4. 7H2O, 0.052 g MgSO4 · 7H2O, of different operational conditions on cell physiol- 7.4 × 10-4 g CaCl2 · 2H2O and 0.8 g NaHCO3. ogy and metabolism, kinetic studies, bioreactor design, mathematical modeling, and development The axenic strain of the cyanobacteria Arthros- of control and operational strategies for each one pira platensis (S. platensis) PCC 8005 was used in of the bioreactors. Second, the bioreactors are compartment IVa. This strain was cultured in scaled-up and operated at pilot plant level during a modified Zarrouck medium, each liter of it con- relevant period of time, since operational stability taining 2.5 g NaNO3, 1 g K2SO4, 1 g NaCl, 0.2 g is an important characteristic for this application. MgSO4·7H2O, 0.04 g CaCl2, 0.01 g FeSO4· Continuous operation of the bioreactors is studied 7H2O, 0.08 g Na2EDTA, 0.5 g K2HPO4, 10.8 g in steady-state, under command of the control NaHCO3, 7.6 g Na2CO3, 1 ml of trace solution A5 system, and the effect of different perturbations is and 1 ml of trace solution B6. One liter of trace also characterized. Finally the continuous inter- solution A5 contained 2.86 g HsBO3, 1.81 g connected operation of the different bioreactors is MnCl2·4H2O, 0.222 g ZnSO4·7H2O, 0.079 9g studied.As mentioned before these tasks are done CuSO4 ·5H2O and 0.015 g MoO3. One liter of sequentially, and this contribution focuses on the trace solution B6 was composed of 0.023 g operation of compartments IHI and IVa, since NH4VO3, 0.096 g KCr(SO4)2 ·12H2O, 0.045 g these are currently the most advanced. Compart- NiSO4 · 6H2O, 0.018 g Na2O4 · 2H2O and 0.048 g ment II, due to the very low growth rate of the Ti(SO4)2 + TiSO4. cells used, and the lack of interest in generating The compositions of the media given here are cells that cannot be used as food, has been those corresponding to the independent operation conceived as an immobilized cells packed-bed of both reactors. For the interconnection experi- column bioreactor.
10_1016_S0168-1656_02_00222-5	chunk_2	The continuous long term ments, a medium is defined as a combination of operation of the nitrification reactor is done at both, on the basis that all the compounds are different oxygen levels, since this is a critical present at the required level. parameter for this reaction. The bioreactor for compartment IVa has been conceived as a gas-lift 2.2. Bioreactor configurations and characteristics photobioreactor. The continuous operation of compartment IVa is characterized for changes in 2.2.1. Compartment III the illumination provided to the reactor, under supervision of its control system. Finally, the 2.2.1.1.Nitrosomonas and Nitrobacter cells immo- continuous interconnected operation of both re- bilization. As mentioned in the introduction, actors is demonstrated. compartment IHI has been conceived as a packed- bed reactor with immobilized cells of Nitrosomo- nas and Nitrobacter. The immobilization is done 2. Material and methods by cell adhesion on a polystyrene support (Bios- tyr), in form of 4 mm diameter beads. 2.1.Strains and culture media 2.2.1.2. Pilot scale packed-bed bioreactor for N. europaea ATCC 19178 and N. winogradskyi compartment Il. All the nitrification tests reported serotype ATCC 14123 are the strains co-cultured in this contribution were carried out at the pilot in compartment IHI. Nitrosomonas cells oxidize scale level. The total volume of this pilot reactor is ammonium to nitrite, as Nitrobacter perform a 8 1. In addition to the packed-bed central section second oxidation to nitrate. The culture medium (6.17 1 in volume) containing the immobilized was adapted from the one defined by (Wijffels and nitrifying bacteria, the reactor has a bottom Tramper, 1995), each liter of it containing 1.32 g section, magnetically stirred, for entry of the gas 322 F.Godia et al./ Journal of Biotechnology 99(2002） 319-330 and liquid inputs, as well as liquid recirculation, allow illumination from the externally mounted and a top section, for gas disengagement, as lamps, provided by six external vertical racks with depicted in Fig. 2. Liquid recirculation ratio and eight halogen lamps (Sylvania, BAB 12V 20 W, aeration rates are high, in order to improve oxygen Belgium) each.The diameter of the cylindrical transfer. The on-line instrumentation of the bior- illuminated part is 9.6 cm and has a total volume eactor is also important. Basic measurements (pH, of 3.68 1, thus, the illuminated to total volume DO, temperature) are taken in the top and bottom ratio of this reactor is 0.52. pH was measured by sections, and their values averaged by the control an in-situ probe and regulated at 9.6 by means of system. A general scheme of this pilot reactor is CO2 addition. Temperature was measured by an in presented in Fig. 2. situ probe and regulated at 36 °C by means of a hot and cold finger installed in the reactor basis. 2.2.2.Compartment IVa The gas circulation in the reactor was maintained by a membrane pump and a slight overpressure 2.2.2.1. Small scale gas-lift photobioreactor for was maintained in the reactor gas phase. compartment IVa. The basic studies for the kinetic 2.2.2.2. Pilot scale gas-lift photobioreactor j characterization of compartment IVa were con- for compartment IVa. The pilot scale reactor for ducted in a 7 1 gas-lift reactor with conventional compartment IVa is a continuous external loop design (Bioengineering AG, Wald, Switzerland), with a cylindrical main body and a gas disengage- gas-lift photobioreactor (Bioengineering AG, Wald, Switzerland), with external illumination ment top section. The walls and internal draught tube in the cylindrical body were made in glass, to that can be regulated in intensity through the control system. The scale-up procedure for this reactor has been previously described (Vernerey et al., 2001) and an scheme is presented in Fig. 3. The illuminated parts of the reactor consist of two cylindrical 15 cm diameter sections and 1.5 m height, serving as riser and downcomer for the liquid circulation in the gas-lift reactor. These columns are connected in the upper and lower parts by curved stainless-steel parts, supporting the instrumentation and external jackets for water circulation for temperature control. Mllumination is provided by a total of 350 externally mounted halogen lamps (Sylvania, BAB 12V 20 W, Bel- gium). The reactor has a total volume of 77 1, and an illuminated volume of 55 1,therefore,the illuminated to total volume ratio of this reactor is 0.71. pH was measured by an in-situ probe and 15 16 controlled at 9.6 by means of CO2 addition. Temperature was measured by an in situ probe Fig. 2. Continuous packed-bed reactor for the co-culture of N. and regulated at 36 °C by means of termostatized europaea and N. winogradsky. Main features of the bioreactor: (1) packed-bed section with immobilized culture, (2) bottom water circulation in the external reactor jackets. section for aeration, liquid distribution and instrumentation, (3) top section for gas disengagement, (4) gas sparger, (5) gas exit 2.2.2.3. Measurement of illumination intensity in condenser, (6) gas closed loop, connected to controlled oxygen/ the photobioreactors. A precise calibration between nitrogen supply to control oxygen level, (7) liquid feed, (8) the voltage applied to the lamps (that it is directly liquid recirculation, (9) liquid outlet, (10) acid addition, (11) base addition, (12) T probes, (13) oxygen probes, (14) pH regulable by the control system) and the intensity probes, (15) cold finger, (16) hot finger. of the illumination system at the external reactor F. Godia et al. I Journal of Biotechnology 99 (2002）319-330 323 2.2.2.5. Interconnection studies and experimental conditions. The experiments for the study of the interconnection between compartments III and IVa have been performed using the reactors at pilot scale previously described. The connection was done only at the liquid phase, and since the outlet of compartment IlI was basically free of cells, an on-line 0.22 μm filter was installed in the connecting stream, to retain any possible leaked cells form compartment III and avoid that they would be fed into compartment IVa. 2.2.2.6. Analytical methods. Spirulina cells con- centration measurement in dry weight was per- formed by filtration of liquid samples (Whatman glass microfiber filters GF/F), and drying at 105 °C during 24 h.Spirulina cells concentration measurement by optical density was based on absorbance measurements at 750 nm. Ammonium, nitrate and nitrite were measured using UV measurement determinations by means of different Fig. 3. Continuous external loop gas-lift reactor for the culture analysis kits (Dr Lange Nitrax, Germany) LCK of S. platensis. Main features of the bioreactor: (1) transparent 305 for ammonium, LCK 339 for nitrate and LCK cylindrical parts (illuminated section): riser (right column) and 341 nitrite. (downcomer left column), (2) stainless steel connection parts, (3) gas-liquid separator, (4) external cooling jackets, (5) liquid medium inlet, (6) liquid outlet, (7) gas inlet through sparger, (8) gas outlet, (9) condenser, (10) halogen lamps. 3. Results and discussion wall (Fr, W m-2) could be obtained on the basis 3.1.Bioreactor configuration and operation of the measurements performed with an spherical light sensor (LI-COR, Lincoln, NE, USA) placed 3.1.1. Compartment III in the center of the reactor. The calibration was The knowledge on this reactor is quite ad- done in an empty reactor, with the sensor in its vanced, as all the basic steps have been completed, center, and the measurement of the light intensity including the hydrodynamic aspects, and the at the center of the reactor was translated into the reactor has been operating continuously for two illumination intensity at the external surface, Fr, long periods (1 and 2 years, respectively). In them, taking into account the geometrical factors.
10_1016_S0168-1656_02_00222-5	chunk_3	Fi- the steady-state operation of the bioreactor has nally, the correlation of Fr with the voltage been characterized, basically in terms of ammoni- applied to the lamps allowed to obtain the final um degradation efficiency at varying loads (ob- calibration curve. tained by changing either t the ammonium concentration or the liquid flow-rate), as well as 2.2.2.4. On-line measurement of biomass in the the non steady-state operation of the reactor after photobioreactors. In all the runs carried out with a change of load. A maximum load of 1.35 kg N these two bioreactors, cell measurement was per day could be achieved, with complete performed on-line, by means of a turbidimetric conversion of ammonium to nitrate, and without sterilizable probe (turbidimeter sensor Monitek, any accumulation of nitrite. Fig. 4 shows the CT4 Dual, Dusseldorf, Germany), measuring concentration profiles at the outlet of the reactor absorbance at 746 nm. for a period of operation of 1oo days. After the 324 F.Godiaet al.1 Journal ofBiotechnology 99(2002）319-330 time(days) 45 75 90 100 0.4 100 80 0.3 60 N 0.2 9 40 0.1 20 0.0 25 50 75 100 numberofresidencetimes start-up Fig. 4. Continuous operation of the nitrifying pilot reactor during a period of 100 days. Evolution of the ammonium (O), nitrite (), nitrate (), and total nitrogen (°O', sum of the ammonium, nitrite and nitrate concentrations), at various flow-rates (residence time given by the dashed line) and ammonium load (ammonium input medium concentration given by the dotted line). start-up of the bioreactor (where the flow-rate is also be influenced by the different oxygen affinity increased stepwise as the conversion activity starts of Nitrosomonas and Nitrobacter,by diffusional progressively during the formation of the micro- limitations in the bioreactor and by the structure bial biofilm), three changes of flow-rate can be of the biofilm of cells attached to the solid support. observed, causing in all cases a non-steady period An example of this complex behavior is given in of operation right after the introduction of the Fig. 5, where the experimental profiles in the change, and the evolution towards an steady-state nitrifying column when the oxygen level is changed operation, where complete conversion is reached from 80 to 40% (top) and then further to 20% again. The information obtained from these series (bottom) are presented. As it can be observed, the of experiments is important both for the develop- first reduction brings the operation profile to a ment of the final operation strategies of this situationwhereammoniumisnotcompletely reactor and for the characterization of the effects converted, and nitrites are first accumulated after that any change in the ammonium load coming the change, but later consumed almost completely. from the rest of the loop will have. The incomplete conversion of ammonium indi- It is also very important to study in this pilot cates that Nitrosomonas cells are limited by the reactor what will be the effect of different pertur- lower oxygen supply, but this result is in principle bations in its operation. Since the ammonium contradictory with the higher affinity that Nitro- oxidation by Nitrosomonas and Nitrobacter is somonas cells have for oxygen, in respect to very dependent on oxygen up-take, the effect of Nitrobacter cells (Hendrikus et al., 1993). How- changes in the oxygen concentration in the bio- ever, the influence of oxygen diffusion limitations reactor is the most critical to study, since it will in the biofilm structure around the solid particles F. Godia et al. I Journal of Biotechnology 99 (2002）319-330 325 biofilm will drop, and probably becomes limiting 1.1 in the inner layers of the biofilm, where mainly 1.0 0.9 Nitrosomonas cells are located. In that situation of 0.8 heterogeneous special distribution of the two 60 0.7 strains in the biofilm, and low oxygen profiles, /N 0.6 Nitrosomonas cells would be more limited that 0.5 0.4 Nitrobacter, and this would explain the ammoni- 30 0.3 um accumulation in the medium. 20 0.2 On the other hand, when the oxygen concentra- 0.1 10% tion is decreased to a 20%, and after a very long 0.0- 516518520522 524 526528530532 non steady-state evolution,the situation is re- numberof residencetimes versed, and partial nitrification is observed, that is, ammonium is almost completely degraded, but 1.3 90 only part of the nitrite is further converted to 1.2 1.1 nitrate. Therefore, in this second situation the rate 1.0 70 limiting step is the activity of Nitrobacter, fact that 0.9 60 can be directly explained by the known higher 0.8 0.7 Oxygen affinity of Nitrosomonas, that it is dom- 7/ So 0.6 40 inating the co-culture interaction at this low 0.5 oxygen level. Partial nitrification has also been 0.4 0.3 observed by other authors in immobilized cells 20 0.2- nitrifying reactors (Nogueira et al., 1998; Joo et 0.1 al., 2000). 0.0 530 The observed results demonstrate the critical 535 540 545 550 555 560 number of residence times role of oxygen supply in this system, and the need Fig. 5. Step perturbation of the dissolved oxygen concentration to take into account the heterogeneous structure of in the nitrifying column from 80 to 40% (top) and from 40 to the biofilm to describe the immobilized co-culture 20% (bottom). Evolution of the concentration of ammonium behavior. The development of a mathematical (O), nitrite (), nitrate (), and total nitrogen (O', sum of model describing these series of experiments and ammonium, nitrite and nitrate concentrations). The dashed line the reactor performance both at steady-state and is the set point of dissolved oxygen concentration. The residence time employed is 10 h. The ammonium input concentration was dynamic situations is under study, as this will be a 1.1 g N 1-1 requirement to establish a control strategy. should also be considered, as well as the possible 3.1.2. Compartment IVa heterogeneous structure of the biofilm. Indeed, a Compartment IVa is the most developed of the possible interpretation of these results is that when pilot plant, since it was also the first to be studied, Oxygen supply is high enough, and due to the and all the process from the basic kinetics studies Oxygen profiles existing in the biofilm, the cells of the microorganism up to the scale-up procedure with more oxygen affinity (Nitrosomonas） will to the final size required in the pilot plant have occupy the inner layers of the biofilm, as the cells been completed successfully. The studies carried with lower affinity (Nitrobacter） will enrich the out during a period of almost 4 years allowed to outer layers, where a higher oxygen concentration test the bioreactor continuous operation at many exist. In this situation both strains would work experimental conditions of nutrients feeding and correctly, adapting to the oxygen profile in the illumination intensity. These studies were carried biofilm, as the oxygen supply would be in any case out in the 7 1 volume reactor described in Section higher that the cellular oxygen consumption. 2. A mathematical model could be then developed, However, when de external set-point of oxygen is comprising both the growth kinetics of the cells decreased to 40%, the profile of oxygen in the and the internal distribution of the light inside the 326 F.Godia et al. / Journal of Biotechnology 99(2002） 319-330 reactor, a critical aspect to define correctly the As a further step in the development of the operation of a photobioreactor, that depends on MELISSA pilot plant, the scale-up of the bior- the illumination intensity, bioreactor geometry, eactor for compartment IVa has been accom- cell concentration, and light absorption and scat- plished, based on the knowledge previously tering properties of the cells. This mathematical developed on the 7 1 bioreactor.
10_1016_S0168-1656_02_00222-5	chunk_4	A final photo- model is fully described in the literature (Cornet et bioreactor of 77 1 volume was designed according al., 1992a,b, 1998), and it is used in the control of to various criteria: maintaining cylindrical geome- the continuous operation of the bioreactor. Indeed try, maximizing the ratio between illuminated to the approach followed for this control is a total volume, moderate reactor diameter in order predictive strategy based on this non-linear math- to avoid dark areas in the center of the reactor, ematical model, including internal calculations and various constraints: size of the laboratory, using the model, reference trajectories, structura- materials availability, among others. The final design (as described in detail in Vernerey et al., tion of the manipulated variables (i.e. light in- 2001) is shown in Fig. 3, and specific details are tensity and feed flow-rate) and auto-compensation given in Section 2. This pilot reactor has been procedure (Fulget et al., 1999). The two main on- operating satisfactorily for long periods of time (1 line measurements that the control systems needs year), also under the same control system devel- are the cell concentration in the reactor and the oped previously for the 71 bioreactor. Indeed, Fig. external illumination intensity, Fr, that have been 7 shows the operation performance of this 77 1 described in Section 2. Using this system, the reactor during a step change in the S. platensis external illumination can be automatically regu- lated to obtain a given reactor performance. The productivity set point, showing very similar trends. This is of particular relevance, and makes evident overall control of the bioreactor can also manip- the advantages that in this case were obtained ulate the inlet liquid flow-rate. With these two from the development of a control system based controlled variables, the automatic control system on a mathematical model describing the interac- allows the operation of the bioreactor at a given tion of the phenomena taking place in the reactor productivity level (Fulget et al., 1999). As an (illumination, light absorption, light scattering and example of this controlled operation, the perfor- cell growth, basically). By using this approach, a mance of the 7 1 reactor bioreactor after a change successful scale-up process could be achieved, in the S. platensis cells productivity set point is based on the use of the same mathematical model provided in Fig. 6. As it can be observed, the and the physical characteristics of the new reactor control system allows to reach the new productiv- being designed. ity very smoothly and precisely, by changing appropriately the illumination intensity at the 3.1.3. Interconnected operation of compartment III reactor surface, also given in Fig. 6. A whole andIVa range of changes in operation set-points have been As mentioned in Section 1, the development of tested, and the bioreactor has been operating MELISSA is pursued at two levels: first, the set-up continuously in a robust manner for long periods and characterization of the individual bioreactors, of operation. It should also be mentioned that this second the connection between the reactors, once procedure, already completed for compartment they have been fully developed at the pilot scale IVa, is currently in progress for compartment II level. Due to the complexity of the system, both and IlI, based on the data already obtained from actions are taken stepwise. For the connection, their operation. The final goal is to build a control this has been started with the two units already system including different levels: a basic level for operating in its final pilot size, compartments III local instrumentation,a second level for each and IVa. Therefore, a series of experiments have individual bioreactor control, and a third level been carried out in which the steady-state opera- for the control of the operation of the complete tion of both compartments, working continuously loop. and interconnected has been maintained during a 3 F.Godia et al./ Journal ofBiotechnology 99(2002）319-330 327 18 500 0,24 17 0,22 16 0,20 400 0,18 (.u. Mp8 14 0,16 300 M (.4-D 13 0,14 0 [x) 0,12 s 12 0,10 200 11 0,08 0,06 100 0,04 8 0,02 0,00 0 25 50 75 100 125 150 175 Time (hours) Fig. 6. Evolution of the performance of compartment IV (7 1 reactor) operated under the control system, when a step change in cell productivity is introduced. The control action on the illumination system allows to reach and maintain the productivity set-point. (——) Productivity set-point (g d.w. 1-1 h- l); (O) on-line measured productivity (g d.w. 1-1 h- '), based on on-line cell concentration measurement, as described in Section 2; (- - -) light intensity provided at the external surface of the reactor by the illumination system; (----) liquid feed flow-rate (1 h - 1). 11 0,78 400 350 10 0,76 300 m 250 M 0,74 bo 01x) 200 0,72 150 100 0,70 50 0,68 20 40 60 80 Time (hours) Fig. 7. Evolution of the performance of compartment IV (771 pilot reactor) operated under the control system, when a step change in cell productivity is introduced. The control action on the illumination allows a similar performance as that observed before the scale-up of this compartment, (—) productivity set-point (g d.w. 1-1 h- '); (O) on-line measured productivity (g d.w. 1-1 h- 1), based on on- line cell concentration measurement, as described in Section 2; (- - -) light intensity provided at the external surface of the reactor by the illumination system; (-- --) liquid feed flow-rate (1l h - 1). 328 F.Godia et al./ Journal ofBiotechnology 99(2002）319-330 months period. It should be emphasized that this of perturbations of the normal operation of the connection is done only at the liquid phase level, loop that may occur. Therefore, a series of so the liquid at the outlet of compartment II is fed experiments have been initiated introducing a into compartment IVa. More interesting is the given perturbation in compartment IHI, and ob- study of the non steady-state operation. This has serving how this action was affecting the operation to be well characterized, since the control system of both compartments, IHI and IVa. of the complete loop will need to be designed More specifically, the results presented in Fig. 8 taking into consideration how to correct a number correspond to the introduction of a perturbation IV 1 VI VII VIII IX 600 500 400 pue 300 IN uu 200 Ammonium (N-ppm) vs. Time (h) Nitrte (N-ppm) vs. Time (h) Nitrate (N-ppm) vs. Time (h) 100 0 opobollopoobpplol 1000050L 1200 1400 1600 1800 2000 Time (h) 2,0 Ammonium (N-ppm) vs. Time (h) 600 Nitrite (N-ppm) vs. Time (h) Nitrate (N-ppm) vs. Time (h) 500 > OD(750) vs. Time (h) 1,5 400 (uu 1,0 B 300 o 200 0.5 VI VII: VIII 100IV IX 。 0,0 1200 1400 1600 1800 2000 Time (h) Fig. 8. Operation profiles of compartments Il (top) and IV (bottom) interconnected at pilot scale when different oxygen concentrations are fixed. The roman numbers indicate different periods in the reactor operation, at different percentage of oxygen saturation: IV, 40%; V, 20%; VI, 15%; VI1, 10%; VII1, 5%; IX, 40%. Ammonium concentration in the feed of compartment II was 600 mg N 1-'. 1lumination in compartment IV was 20 W m -2. Flow-rate was 14.5 1 per day (D = 0.16 h -1 for compartment II reactor and D = 0.0084 h -1 for compartment IV). F.Godia et al./ Journal of Biotechnology 99(2002） 319-330 329 in the oxygen concentration in compartment III, 4. Conclusions by means of decreasing stepwise the oxygen level in the reactor. The starting point (indicated by The completion of the MELISSA loop will serve period IV in Fig. 8) is the normal operation of the as a test for the potential of biotechnology to reactors at steady-state, at a level of 40% oxygen provide solutions for self-regenerative life support saturation in the liquid phase, and an input flow- systems in space. The complexity of the system rate of 14.5 1 per day, with an ammonium corroborates the need to proceed in a stepwise concentration of 600 mg N 1-1.
10_1016_S0168-1656_02_00222-5	chunk_5	From this point, approach to reach the final construction of the the oxygen level is consecutively lowered to 20 loop. The experiments reported in this contribu- (period V), 15 (period VI), 10 (period VII), 5% tion demonstrate that the long term continuous (period VIII), and finally raised back to 40% operation of these reactors is possible at pilot (period IX). As it can be observed, for the periods scale. Also, the successful application of control V, VI and Vll, the ammonium fed into compart- systems based on mathematical models describing ment IHI was completely transformed into nitrates the more important physical and biological aspects under steady-state operation, and those nitrates of the culture system is demonstrated, particularly were further consumed, although not completely for a complex case such a photobioreactor for S. because they were in excess, in compartment IVa platensis cells. Finally, the continuous intercon- under oxygen limitation (period VIII), partial nected operation of two bioreactors of the loop nitrification occurs in compartment IHl, nitrate (compartment II and IVa) under different condi- concentration at the outlet drops and nitrites tions of operation has also been achieved, and this accumulate. As a consequence, the nitrites con- is an important step towards the realization of the centration increases continuously in compartment complete MELISSA loop. IVa, until a steady-state is reached at the end of period VIll. It should be pointed out how the non steady-state dynamics of both reactor is very Acknowledgements different, due to the different residence times (the same flow rate is used, but the reactor volumes are The physical realization and operation of the very different). Thus, compartment IHI reaches MELISSA pilot plant at the Universitat Autono- steady-state operation much quickly than reactor ma de Barcelona has been possible through the IVa. For this period, the nitrates arriving to funding of the European Space Agency (ESTEC compartment IVa are still in excess, and the contracts 11549/95/NL/FG and 113292/98/NL/ Spirulina cell concentration in compartment IVa MV), the Plan Nacional de Biotecnologia of the is well maintained, as shown by the optical density Spanish CICYT (BIO1998-1594-E), and CIRIT profile In principle this fact also suggest that (Generalitat de Catalunya). nitrites do not affect negatively Spirulina cells. However, they seem not to be consumed at all, and this fact has an important consequence in the References operation of the loop, since due to the negative effect of nitrites on human health, their accumula- Ciferri, O., 1983. Spirulina, the edible microorganism. Micro- biol. Rev. 47, 551-578. tion should be absolutely avoided. The fact that Cornet, J.F., Dussap, C.G., Dubertet, G., 1992a. A structured nitrites are not consumed in compartment IVa is model for simulation of cultures of the cyanobacterium further corroborated in the period of operation Spirulina platensis in photobioreactors: I. Coupling between IX, in which the oxygen level is brought back to light transfer and growth kinetics. Biotechnol. Bioeng. 40, non limiting conditions (40%). In this case, com- 817-825. partment II recovers complete nitrification per- Cornet, J.F., Dussap, C.G., Cluzel, P., Dubertet, G., 1992b. A structured model for simulation of cultures of the cyano- formance, and nitrite concentration decreases at bacterium Spirulina platensis in photobioreactors: II. Iden- the outlet of compartment IVa, following precisely tification of kinetic parameters under light and mineral the wash-out curve. limitations. Biotechnol. Bioeng. 40, 826N-834N. 330 F. Godia et al. I Journal of Biotechnology 99(2002） 319-330 Cornet, J.F., Dussap, C.G., Gros, J.B., 1998. Kinetics and spectives. Proceedings of the 30th International Conference energetics of photosynthetic micro-organisms in photobio- on Environmental Systems, Toulouse, France. reactors.Adv. Biochem. Eng./Biotechnol. 59, 153-224. Mergeay, M., Verstraete, W., Dubertret, G., Lefort-tran, M., Eckart, P., 1994. Life Support and Biospheric. Herbert Utz Chipaux, C., Binot, R., 1988. MELISSA. A microorganisms Publisher, Munchen, Germany. based model for CELSS development. Proceedings of the Fulget, N., Poughon, L., Richalet, J., Lasseur, C.h., 1999. 3rd symposium on space thermal control and life support MELISSA: global control strategy of the artificial ecosys- systems. Noordwijk, The Netherlands. tem by using first principles models of the compartments. Nogueira, R., Lazarova, V., Manem, J., Melo, L.F., 1998. Adv. Space Res. 24, 397-405. Influence of dissolved oxygen on the nitrification kinetics in Gustavino, S.R., Fadden, C.D., Davenport, R.J., 1994. Con- a circulating bed biofilm reactor. Bioprocess Eng. 19, 441- cepts for advanced waste water processing systems. Pro- 449. Tamponet, C., Savage, C., Amblard, P., Laserre, J.C., Per- ceedings of the 24th International Conference on sonne, J.C., Germain, J.C., 1999. Water recovery in space. Environmental Systems, Friedrichshafen, Germany. Society of Automotive Engineering Technical paper 941500. ESA Bul1.97,56-60. Tamponnet, C., Savage, C., 1994. Closed ecological systems. J. Hendrikus, J., Laambroek, H.J., Gerards, S., 1993. Competi- Biol. Educ. 28, 167-173. tion for limiting amounts of oxygen between Nitrosomonas Vernerey, A., Albiol, J., Lasseur, C., Godia, F., 2001. Scale-up europaea and Nitrobacter winogradsky grown in mixed and design of a pilot-plant photobioreactor for the con- cultures. Microbiology 159, 453-459. tinuous culture of Spirulina platensis. Biotechnol. Progr. 17, Joo, S.-H., Kim, D.-J., Yoo, I.-K., Park, K., Cha, G.-C., 2000. 431-438. Partial nitrification in an upflow biological aerated filter by Vrati, S., 1984. Single cell protein production by photosynthetic O2 limitation. Biotechnol. Lett. 22, 937-940. bacteria grown on the clarified effluents of a biogas plant. Lasseur, C., Dixon, M., Dubertret, G., Dussap, G., Godia, F., Appl. Microbiol. Biotechnol. 19, 199-202. Gros, J.B., Mergeay, M., Richalet, J., Verstraete, W., 2000. Wijffels, R.H., Tramper, J., 1995. Nitrification by immobilized MELISSA: 10 years of research, results, status and per- cells. Enzyme Microb. Technol. 17, 482-492.
ijms-26-01905-v3	references	Citation: Fukumoto, Y.; Li, E.; Tanaka, Y.; Suzuki, N.; Ogra, Y. Evaluation of Metal Accumulation in Escherichia coli Expressing SPL2 by Single-Cell Inductively Coupled Plasma Mass Spectrometry. Int. J. Mol. Sci.2025 ,26, 1905. https://doi.org/ 10.3390/ijms26051905 Copyright: © 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/). Article
ijms-26-01905-v3	experiments	Evaluation of Metal Accumulation in Escherichia coli Expressing SPL2 by Single-Cell Inductively Coupled Plasma Mass Spectrometry Yasunori Fukumoto1,* , Enhui Li1, Yu-ki Tanaka1 , Noriyuki Suzuki1,2and Yasumitsu Ogra1 1Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 260-8675, Japan; yu-ki.tanaka@chiba-u.jp (Y.T.); noriyuki.suzuki@phar.toho-u.ac.jp (N.S.); ogra@chiba-u.jp (Y.O.) 2Faculty of Pharmaceutical Sciences, Toho University, Chiba 274-8510, Japan *Correspondence: fukumoto@faculty.chiba-u.jp
ijms-26-01905-v3	abstract	Abstract: Rare earth elements, comprising 17 elements including 15 lanthanides, are es- sential components in numerous high-tech applications. While physicochemical methods are commonly employed to remove toxic heavy metals (e.g., cadmium and mercury) from industrial wastewater, biological approaches offer increasingly attractive alternatives. Bio- mining, which utilizes microorganisms to extract valuable metals from ores and industrial wastes, and bioremediation, which leverages microorganisms to adsorb and transport metal ions into cells via active transport, provide eco-friendly solutions for resource recovery and environmental remediation. In this study, we investigated the potential of three recently identified lanthanide-binding proteins—SPL2, lanpepsy, and lanmodulin—for applications in these areas using single-cell inductively coupled plasma mass spectrometry (scICP-MS).
ijms-26-01905-v3	discussion	resulted in high expression levels and solubility. Single-cell ICP-MS analysis revealed that these recombinant bacteria accumulated lanthanum, cobalt, nickel, and cadmium, effec- tively sequestering lanthanum and cadmium from the culture media. Furthermore, SPL2
ijms-26-01905-v3	results	expression conferred enhanced bacterial tolerance to cadmium exposure. These findings establish SPL2 as a promising candidate for developing recombinant bacterial systems for heavy metal bioremediation and rare earth element biomining. Keywords: scICP-MS; single-cell inductively coupled plasma mass spectrometry; SPL2; cadmium; bioremediation; biomining
ijms-26-01905-v3	introduction	1. Introduction Rare earth elements (REEs), a group of 17 chemical elements including 15 lanthanides, yttrium, and scandium [ 1], are indispensable in numerous high-tech applications owing to their unique physicochemical properties [ 2–4]. However, conventional REE mining gener- ates substantial amounts of REE-rich liquid waste, resulting in significant resource loss and posing severe environmental and public health risks [ 5]. These wastes can contaminate ecosystems and accumulate in the food chain, potentially disrupting normal physiologi- cal processes in plants [ 6] and causing hepatotoxicity and neurotoxicity in humans and animals [ 7–9]. Therefore, developing cost-effective, adaptive, and environmentally sound methods for REE recovery from these liquid wastes is of paramount importance [10]. Biomining, which employs microorganisms to extract valuable metals from ores and mine wastes, has recently emerged as a promising alternative to conventional REE Int. J. Mol. Sci. 2025 ,26, 1905 https://doi.org/10.3390/ijms26051905 Int. J. Mol. Sci. 2025 ,26, 1905 2 of 13
ijms-26-01905-v3	methodology	offers a more sustainable approach. Advances in molecular and genetic engineering have further enhanced the efficiency of REE recovery in biomining processes. A key molecular engineering strategy involves genetically modifying microorganisms to express specific proteins, such as lanthanide-binding tags, which significantly improve REE extraction yields [10,13,14]. Heavy metals, defined as elements with high density and inherent toxicity even at low concentrations [ 15], pose a significant and growing threat to global environmental and public health [ 16]. The rapid pace of industrialization and urbanization has dramat- ically increased the release of these toxic metals into the environment through various anthropogenic activities [ 17]. While a range of physicochemical processes are currently
ijms-26-01905-v3	methodology	used for heavy metal removal, these methods are often hampered by significant limitations, including high reagent consumption and the generation of hazardous sludges [18]. Bioremediation, utilizing microorganisms to adsorb and transport metal ions into their
ijms-26-01905-v3	methodology	cells via active transport [ 19], offers an eco-friendly alternative to conventional methods for addressing heavy metal pollution. Naturally occurring bacteria, fungi, seaweeds, and algae have demonstrated significant potential for heavy metal adsorption and accumula- tion [ 20–23], offering advantages such as superior adsorption efficiency, the elimination of secondary pollution, and enhanced environmental compatibility compared with physico- chemical approaches [ 24,25]. Microorganisms employ diverse mechanisms for heavy metal detoxification, including cell surface adsorption, intracellular aggregation, mineralization precipitation, and intracellular transformation, even in cases of severe contamination such as with cadmium (Cd) [ 26,27]. Their small size, ease of culture, and rapid reproduction further contribute to their appeal. However, the practical application of naturally occurring microorganisms can be limited by their often-modest adsorption capacities and stress resistance. To overcome these limitations, recent advances in genetic engineering have focused on enhancing microbial remediation efficiency [ 28,29]. This involves introducing genes associated with enhanced heavy metal accumulation into robust and genetically tractable recipient strains [30], creating more effective bioremediation agents. Three proteins with lanthanide-binding potential—SPL2, lanmodulin, and lanpepsy— were investigated in this study. SPL2, a recently identified ubiquitin ligase localized in the outer membrane of plant chloroplasts, has an unclear physiological role. However, its cytosolic fragment exhibits structural similarity to known lanthanide-binding tags and has been shown to bind both lanthanide and calcium (Ca) ions [ 31]. Lanmodulin, identified in Methylobacterium extorquens , is characterized by four metal-binding EF-hand motifs and displays exceptionally high affinity for lanthanides, such as lanthanum (La), terbium (Tb), samarium (Sm), and neodymium (Nd), with dissociation constants in the low picomolar range—eight orders of magnitude lower than those for Ca [ 32]. Lanpepsy, a small protein induced by La exposure in Methylobacillus flagellatus , contains two PepSY domains known for their metal-binding properties. In vitro binding assays suggest that lanpepsy has four lanthanum ion-binding sites and exhibits binding activity for La ions and other lanthanides, with dissociation constants in the low micromolar range [ 33]. The subcellular localization of lanmodulin and lanpepsy has been predicted to be in the periplasm [ 32,33]. SPL2, lanpepsy, and lanmodulin were successfully expressed in Escherichia coli (E. coli ) [31–33]. Single-cell inductively coupled plasma mass spectrometry (scICP-MS) is a powerful
ijms-26-01905-v3	discussion	dividual cells. scICP-MS offers several advantages over bulk analysis using conventional ICP-MS. Namely, acid decomposition is not required, reducing the potential risks of contam- Int. J. Mol. Sci. 2025 ,26, 1905 3 of 13 ination and sample loss. In addition, inaccuracies in cell counting may affect quantification
ijms-26-01905-v3	discussion	accuracy in a bulk analysis. scICP-MS overcomes these limitations by enabling the direct analysis of individual cells. Moreover, it provides a more comprehensive characterization of samples by showing not only the average elemental content but also the variability and distribution patterns within the population [ 34–38]. In a recent study, we employed scICP-MS to directly detect histidine-tagged recombinant proteins expressed in bacteria. By labeling these proteins in situ with cobalt (Co) and nickel (Ni) ions, we were able to analyze intact bacterial cells without the need for laborious protein purification steps typically
ijms-26-01905-v3	methodology	required for in vitro studies. This approach facilitated the characterization of the metal- binding properties of recombinant proteins [ 39], demonstrating the power of scICP-MS for cellular metalloproteomics. Of the three lanthanide-binding proteins investigated in this study, SPL2 demonstrated superior characteristics, highlighting its potential for biotechnological applications in both lanthanide biomining and heavy metal bioremediation. Specifically, the metal-binding properties of SPL2 provide a strong scientific foundation for developing recombinant bacte- rial systems for the effective bioremediation of heavy metal pollution and the sustainable biomining of REEs.
ijms-26-01905-v3	results	2. Results 2.1. SPL2 and Lanpepsy Show High Expression and Solubility in Bacteria To examine the expression of potential lanthanoid-binding proteins, we expressed SPL2 (residues V291–S383) fused with flag-mCherry in E. coli . Robust expression was observed for both SPL2-flag-mCherry and lanpepsy-flag (Figure 1A). However, bacterial expression of lanmodulin proved challenging; bacteria transformed with the lanmodulin- expression plasmid exhibited significantly impaired growth, and no lanmodulin-flag expres-
ijms-26-01905-v3	discussion	binding ability [ 39], further analysis focused on soluble proteins SPL2-flag-mCherry and lanpepsy-flag. To confirm the solubility of the recombinant proteins, we analyzed the soluble fraction of bacterial lysates. Sufficient amounts of SPL2-flag-mCherry and lanpepsy- flag were detected (Figure 1B), demonstrating their good solubility and suggesting their potential to function as metal-binding proteins within bacteria. Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 3 of 14 Single-cell inductively coupled plasma mass spectrometry (scICP-MS) is a powerful
ijms-26-01905-v3	discussion	dividual cells. scICP-MS o ﬀers several advantages over bulk analysis using conventional ICP-MS. Namely, acid decomposition is not required, reducing the potential risks of con- tamination and sample loss. In addition, inaccuracies in cell counting may a ﬀect quanti ﬁ- cation accuracy in a bulk an alysis. scICP-MS overcomes these limitations by enabling the
ijms-26-01905-v3	discussion	direct analysis of individual cells. Moreover , it provides a more comprehensive character- ization of samples by showing not only the average elemental content but also the varia- bility and distribution pa tterns within the population [34–38]. In a recent study, we em- ployed scICP-MS to directly detect histidin e-tagged recombinant proteins expressed in bacteria. By labeling these protei ns in situ with cobalt (Co) and nickel (Ni) ions, we were able to analyze intact bacterial cells wi thout the need for laborious protein puri ﬁcation
ijms-26-01905-v3	methodology	steps typically required for in vitro studies. This approach facilitated the characterization of the metal-binding properties of recombinan t proteins [39], demonstrating the power of scICP-MS for cellular metalloproteomics. Of the three lanthanide-binding proteins investigated in this study, SPL2 demon- strated superior characteristics, highlighting its potential for biotechnological applications in both lanthanide biomining and heavy metal bioremediation. Speci ﬁcally, the metal- binding properties of SPL2 provide a strong scienti ﬁc foundation for developing recom- binant bacterial systems for the e ﬀective bioremediation of heavy metal pollution and the sustainable biomining of REEs.
ijms-26-01905-v3	results	2. Results 2.1. SPL2 and Lanpepsy Show High Expression and Solubility in Bacteria To examine the expression of potential la nthanoid-binding proteins, we expressed SPL2 (residues V291 ‒S383) fused with ﬂag-mCherry in E. coli . Robust expression was ob- served for both SPL2- ﬂag-mCherry and lanpepsy- ﬂag (Figure 1A). However, bacterial ex- pression of lanmodulin prov ed challenging; bacteria transformed with the lanmodulin- expression plasmid exhibited signi ﬁcantly impaired growth, and no lanmodulin- ﬂag ex-
ijms-26-01905-v3	discussion	metal-binding ability [39], further analysis focused on soluble proteins SPL2- ﬂag- mCherry and lanpepsy- ﬂag. To con ﬁrm the solubility of the recombinant proteins, we analyzed the soluble fraction of bacterial lysates. Su ﬃcient amounts of SPL2- ﬂag-mCherry and lanpepsy- ﬂag were detected (Figure 1B), demonstrating their good solubility and sug- gesting their potential to function as metal-binding proteins within bacteria. Figure 1. Visualization of recombinant prot ein expression and solubility in E. coli . (A) Whole-cell lysates of E. coli BL21(DE3) transformed with pET29b plasmids expressing mCherry, ﬂag-mCherry, mCherry-6His, SPL2- ﬂag-mCherry (SPL2 V291 ‒S383), or lanpepsy- ﬂag were analyzed by SDS- Figure 1. Visualization of recombinant protein expression and solubility in E. coli . (A) Whole-cell lysates of E. coli BL21(DE3) transformed with pET29b plasmids expressing mCherry, flag-mCherry, mCherry-6His, SPL2-flag-mCherry (SPL2 V291–S383), or lanpepsy-flag were analyzed by SDS-PAGE and stained with Coomassie Brilliant Blue (CBB). Recombinant protein expression was induced with IPTG. ( B) Soluble fractions of the same bacterial lysates, prepared under non-denaturing conditions and clarified by centrifugation, were also analyzed by SDS-PAGE and CBB staining. The calculated molecular weights of flag-mCherry-SPL2 and lanpepsy-flag are 39.7 and 19.3 kDa, respectively. Int. J. Mol. Sci. 2025 ,26, 1905 4 of 13 2.2. SPL2 Promotes La Binding in Recombinant Bacteria We employed scICP-MS to investigate the ability of recombinant bacteria to bind lanthanide. Following La exposure, SPL2 expression resulted in a rightward shift in the histogram, accompanied by an increase in the mean and distribution of the signal intensity (Figure 2). The concentration of La ions bound to the bacteria was 28.0 ±2.7 attomol/cell, whereas in bacteria lacking plasmids, it was 14.3 ±1.6 attomol/cell. This indicates that the presence of SPL2 led to an approximately twofold increase in La ion binding. In contrast, no such enhancement was observed in bacteria expressing control proteins (mCherry, mCherry-6His, and flag-mCherry). The SPL2-flag-mCherry-expressing group showed a
ijms-26-01905-v3	results	statistically significant increase in La binding compared with control groups. These results demonstrate that the SPL2 fragment facilitates enhanced La binding. While SPL2 expression tended to increase Tb binding, this difference was not statisti- cally significant (Figure 3). The flag-mCherry-expressing group also showed increased Tb binding. However, lanpepsy did not enhance the binding of La or Tb (Figure S1). Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 4 of 14 PAGE and stained with Coomassie Brilliant Blue (CBB). Recombinant protein expression was in- duced with IPTG. ( B) Soluble fractions of the same bacterial lysates, prepared under non-denaturing conditions and clari ﬁed by centrifugation, were also analyz ed by SDS-PAGE and CBB staining. The calculated molecular weights of ﬂag-mCherry-SPL2 and lanpepsy- ﬂag are 39.7 and 19.3 kDa, re- spectively. 2.2. SPL2 Promotes La Binding in Recombinant Bacteria We employed scICP-MS to investigate the ability of recombinant bacteria to bind lan- thanide. Following La exposure, SPL2 expression resulted in a rightward shift in the his- togram, accompanied by an increase in the mean and distribution of the signal intensity (Figure 2). The concentration of La ions bound to the bacteria was 28.0 ± 2.7 a ttomol/cell, whereas in bacteria lacking plasmids, it was 14.3 ± 1.6 a ttomol/cell. This indicates that the presence of SPL2 led to an approximately twof old increase in La ion binding. In contrast, no such enhancement was observed in bacter ia expressing control proteins (mCherry, mCherry-6His, and ﬂag-mCherry). The SPL2- ﬂag-mCherry-expressing group showed a
ijms-26-01905-v3	results	statistically signi ﬁcant increase in La binding compared with control groups. These results demonstrate that the SPL2 fragment facilitates enhanced La binding. While SPL2 expression tended to increase Tb binding, this di ﬀerence was not statis- tically signi ﬁcant (Figure 3). The ﬂag-mCherry-expressing group also showed increased Tb binding. However, lanpepsy did not enha nce the binding of La or Tb (Figure S1). Figure 2. Lanthanum (La) binding in SPL2-expressing bacteria. ( A‒E) Recombinant bacteria were cultured in liquid medium, and recombinant protein expression was induced. SPL2- ﬂag-mCherry and control proteins were expressed and then expo sed to 250 µM La ion for 1.5 h. Bacterial cells were collected, and La binding was assessed by scICP-MS. The histograms show the frequency dis-
ijms-26-01905-v3	experiments	tribution of La signal intensities. Representati ve results from three independent experiments are shown, with Gaussian distribution ﬁtting curves overlaid. ( F) The amount of La per bacterial cell was calculated from scICP-MS signal intensities, and the average content per cell was determined.
ijms-26-01905-v3	experiments	The graph shows data from thr ee independent experiments. , p < 0.05. **, p < 0.001. Figure 2. Lanthanum (La) binding in SPL2-expressing bacteria. ( A–E) Recombinant bacteria were cultured in liquid medium, and recombinant protein expression was induced. SPL2-flag-mCherry and control proteins were expressed and then exposed to 250 µM La ion for 1.5 h. Bacterial cells were collected, and La binding was assessed by scICP-MS. The histograms show the frequency distribution
ijms-26-01905-v3	experiments	of La signal intensities. Representative results from three independent experiments are shown, with Gaussian distribution fitting curves overlaid. ( F) The amount of La per bacterial cell was calculated from scICP-MS signal intensities, and the average content per cell was determined. The graph shows
ijms-26-01905-v3	discussion	Figure 3. Terbium (Tb) binding by recombinant bacteria. ( A‒E) Tb binding analysis by scICP-MS. The experimental procedure was identical to that de scribed in Figure 2, except that the recombinant bacteria were exposed to 250 µM Tb ion. ( F) The graph shows the averag e amount of bound Tb per
ijms-26-01905-v3	experiments	cell. Data represent means from th ree independent experiments. *, p < 0.05. 2.3. SPL2 Enhances Transition Metal Binding in Recombinant Bacteria The ability of SPL2- ﬂag-mCherry-expressing bacteria to bind transition metals was analyzed. Previously, we demonstrated that re combinant bacteria expressing 6His-tagged proteins acquire the ability to accumulate Ni and Co ions [39]. Following exposure to Ni and Co, bacteria expressing mCherry-6His sh owed a rightward shift in the histogram, whereas those expressing mCherry alone did not (Figures 4 and 5). This indicates that Ni and Co accumulation was mediated by the 6His tags, as shown previously [39]. Ni expo- sure also caused a rightward shift in the hist ogram of SPL2-expressing bacteria (Figure 4). The SPL2- ﬂag-mCherry-expressing group exhibited a signi ﬁcant increase in Ni binding compared with the mCherry and ﬂag-mCherry groups, suggesting an association be- tween the SPL2 fragment and Ni. Upon exposu re to Co, the histograms of mCherry- or ﬂag-mCherry-expressing bacteria shifted leftward, indicating decreased Co binding (Fig- ure 5). However, SPL2- ﬂag-mCherry expression enhanced Co binding compared with
ijms-26-01905-v3	experiments	cell. Data represent means from three independent experiments. *, p< 0.05. Int. J. Mol. Sci. 2025 ,26, 1905 5 of 13 2.3. SPL2 Enhances Transition Metal Binding in Recombinant Bacteria The ability of SPL2-flag-mCherry-expressing bacteria to bind transition metals was analyzed. Previously, we demonstrated that recombinant bacteria expressing 6His-tagged proteins acquire the ability to accumulate Ni and Co ions [ 39]. Following exposure to Ni and Co, bacteria expressing mCherry-6His showed a rightward shift in the histogram, whereas those expressing mCherry alone did not (Figures 4 and 5). This indicates that Ni and Co accumulation was mediated by the 6His tags, as shown previously [ 39]. Ni exposure also caused a rightward shift in the histogram of SPL2-expressing bacteria (Figure 4). The SPL2- flag-mCherry-expressing group exhibited a significant increase in Ni binding compared with the mCherry and flag-mCherry groups, suggesting an association between the SPL2 fragment and Ni. Upon exposure to Co, the histograms of mCherry- or flag-mCherry- expressing bacteria shifted leftward, indicating decreased Co binding (Figure 5). However, SPL2-flag-mCherry expression enhanced Co binding compared with mCherry and flag-
ijms-26-01905-v3	discussion	Figure 3. Terbium (Tb) binding by recombinant bacteria. ( A‒E) Tb binding analysis by scICP-MS. The experimental procedure was identical to that de scribed in Figure 2, except that the recombinant bacteria were exposed to 250 µM Tb ion. ( F) The graph shows the averag e amount of bound Tb per
ijms-26-01905-v3	experiments	cell. Data represent means from th ree independent experiments. *, p < 0.05. 2.3. SPL2 Enhances Transition Metal Binding in Recombinant Bacteria The ability of SPL2- ﬂag-mCherry-expressing bacteria to bind transition metals was analyzed. Previously, we demonstrated that re combinant bacteria expressing 6His-tagged proteins acquire the ability to accumulate Ni and Co ions [39]. Following exposure to Ni and Co, bacteria expressing mCherry-6His sh owed a rightward shift in the histogram, whereas those expressing mCherry alone did not (Figures 4 and 5). This indicates that Ni and Co accumulation was mediated by the 6His tags, as shown previously [39]. Ni expo- sure also caused a rightward shift in the hist ogram of SPL2-expressing bacteria (Figure 4). The SPL2- ﬂag-mCherry-expressing group exhibited a signi ﬁcant increase in Ni binding compared with the mCherry and ﬂag-mCherry groups, suggesting an association be- tween the SPL2 fragment and Ni. Upon exposu re to Co, the histograms of mCherry- or ﬂag-mCherry-expressing bacteria shifted leftward, indicating decreased Co binding (Fig- ure 5). However, SPL2- ﬂag-mCherry expression enhanced Co binding compared with
ijms-26-01905-v3	experiments	of bound Ni per cell. Data represent means from four independent experiments. *, p< 0.05. Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 6 of 14 Figure 5. Cobalt (Co) binding by recombinant bacteria expressing mCherry-6His and SPL2- ﬂag-
ijms-26-01905-v3	experiments	bound Co per cell. Data represent means from three independent experiments. *, p < 0.05. The Cd binding of SPL2-expressing bacter ia was also examined. SPL2 expression caused a rightward shift in the histogram (Figure 6). The SPL2- ﬂag-mCherry-expressing group showed a signi ﬁcant increase in Cd binding compared with all other groups. The concentration of Cd ions bound to the ba cteria was measured to be 0.15 ± 0.02 a ttomol/cell.
ijms-26-01905-v3	results	These results indicate that the SPL2 fragment within the SPL2- ﬂag-mCherry protein pro- moted Cd binding of the recombinant bacteria. Figure 6. Cadmium (Cd) binding by recombin ant bacteria expressing SPL2- ﬂag-mCherry. ( A‒E)
ijms-26-01905-v3	experiments	The experimental procedure was identical to that de scribed in Figure 2, with the exception that the recombinant bacteria were exposed to 250 µM Cd ion. ( F) The graph shows the average amount of bound Cd per cell. Data represent means from three independent experiments. , p < 0.05. *, p < 0.01. 2.4. La and Cd Sequestration in SPL2-Expressing Bacteria We evaluated the ability of recombinant bacteria to remove Cd and La from the cul- ture medium. When cultured in La-containing medium, La concentrations in the culture supernatant generally decreased across all bacterial strains (Figure 7A ‒C). However, at 250 and 25 µM La, the control vector-containing bacteria showed a signi ﬁcant reduction in La concentration compared with the cultur e medium not incubated with bacteria, sug- gesting that La has an a ﬃnity for bacterial cells at a baseline level. Remarkably, SPL2- ﬂag- mCherry-expressing bacteria demonstrated a signi ﬁcant reduction in La concentration at all three tested concentrations. This observat ion indicates that SPL2 expression enhances La sequestration from the culture supernatant beyond the baseline level. Figure 5. Cobalt (Co) binding by recombinant bacteria expressing mCherry-6His and SPL2-flag-
ijms-26-01905-v3	experiments	bound Co per cell. Data represent means from three independent experiments. *, p< 0.05. The Cd binding of SPL2-expressing bacteria was also examined. SPL2 expression caused a rightward shift in the histogram (Figure 6). The SPL2-flag-mCherry-expressing Int. J. Mol. Sci. 2025 ,26, 1905 6 of 13 group showed a significant increase in Cd binding compared with all other groups. The concentration of Cd ions bound to the bacteria was measured to be 0.15 ±0.02 attomol/cell.
ijms-26-01905-v3	results	These results indicate that the SPL2 fragment within the SPL2-flag-mCherry protein pro- moted Cd binding of the recombinant bacteria. Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 6 of 14 Figure 5. Cobalt (Co) binding by recombinant bacteria expressing mCherry-6His and SPL2- ﬂag-
ijms-26-01905-v3	experiments	bound Co per cell. Data represent means from three independent experiments. *, p < 0.05. The Cd binding of SPL2-expressing bacter ia was also examined. SPL2 expression caused a rightward shift in the histogram (Figure 6). The SPL2- ﬂag-mCherry-expressing group showed a signi ﬁcant increase in Cd binding compared with all other groups. The concentration of Cd ions bound to the ba cteria was measured to be 0.15 ± 0.02 a ttomol/cell.
ijms-26-01905-v3	results	These results indicate that the SPL2 fragment within the SPL2- ﬂag-mCherry protein pro- moted Cd binding of the recombinant bacteria. Figure 6. Cadmium (Cd) binding by recombin ant bacteria expressing SPL2- ﬂag-mCherry. ( A‒E)
ijms-26-01905-v3	experiments	The experimental procedure was identical to that de scribed in Figure 2, with the exception that the recombinant bacteria were exposed to 250 µM Cd ion. ( F) The graph shows the average amount of bound Cd per cell. Data represent means from three independent experiments. , p < 0.05. *, p < 0.01. 2.4. La and Cd Sequestration in SPL2-Expressing Bacteria We evaluated the ability of recombinant bacteria to remove Cd and La from the cul- ture medium. When cultured in La-containing medium, La concentrations in the culture supernatant generally decreased across all bacterial strains (Figure 7A ‒C). However, at 250 and 25 µM La, the control vector-containing bacteria showed a signi ﬁcant reduction in La concentration compared with the cultur e medium not incubated with bacteria, sug- gesting that La has an a ﬃnity for bacterial cells at a baseline level. Remarkably, SPL2- ﬂag- mCherry-expressing bacteria demonstrated a signi ﬁcant reduction in La concentration at all three tested concentrations. This observat ion indicates that SPL2 expression enhances La sequestration from the culture supernatant beyond the baseline level. Figure 6. Cadmium (Cd) binding by recombinant bacteria expressing SPL2-flag-mCherry. ( A–E) The
ijms-26-01905-v3	experiments	bound Cd per cell. Data represent means from three independent experiments. , p< 0.05. *, p< 0.01. 2.4. La and Cd Sequestration in SPL2-Expressing Bacteria We evaluated the ability of recombinant bacteria to remove Cd and La from the culture medium. When cultured in La-containing medium, La concentrations in the cul- ture supernatant generally decreased across all bacterial strains (Figure 7A–C). However, at 250 and 25 µM La, the control vector-containing bacteria showed a significant reduc- tion in La concentration compared with the culture medium not incubated with bacteria, suggesting that La has an affinity for bacterial cells at a baseline level. Remarkably, SPL2- flag-mCherry-expressing bacteria demonstrated a significant reduction in La concentration at all three tested concentrations. This observation indicates that SPL2 expression enhances La sequestration from the culture supernatant beyond the baseline level. Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 7 of 14
ijms-26-01905-v3	results	Consistent with the La removal results, Cd concentration in the supernatant showed a signi ﬁcant reduction when the SPL2-expressin g bacteria were cultured in media con- taining 250 µM and 25 µM Cd (Figure 7D ‒F). The SPL2-expressing group also exhibited a tendency for enhanced Cd removal at 50 µM Cd. These ﬁndings suggest that SPL2-ex- pressing bacteria e ﬀectively sequester Cd from the culture medium. Figure 7. Removal of La and Cd ions from cultur e media by SPL2-expressing bacteria. ( A–C) La removal. ( D–F) Cd removal. Recombinant bacteria were cultured in liquid medium, and recombi- nant protein expression was induced. Bacteria were then exposed to the indicated concentrations of La (A‒C) or Cd ions ( D‒F) for 1.5 h. Following exposure, the bacterial cells were removed by cen- trifugation, and metal ion concentrations in the resulting supernatants were measured by ICP-MS.
ijms-26-01905-v3	experiments	Graphs show means from thr ee independent experiments. p-values were calculated using Student’s t-test or Welch’s t-test, comparing each bacterial group with the no-bacteria control (culture medium without bacteria), except for pa nel F. 6His, mCherry-6His; FLAG, ﬂag-mCherry; SPL2, SPL2- ﬂag- mCherry. , p < 0.05. , p < 0.01. **, p < 0.001. 2.5. SPL2 Expression Increases Cd Tolerance of Recombinant Bacteria We examined the impact of SPL2 expression on bacterial tolerance to lanthanides and transition metals. SPL2-expressing bacteria exhibited enhanced growth on Cd-containing plates, compared with control bacteria at Cd concentrations of 0.05 mM, 0.25 mM, and 0.5 mM (Figure 8A). In contrast, SPL2 expression did not promote growth in the presence of
ijms-26-01905-v3	results	Consistent with the La removal results, Cd concentration in the supernatant showed a signi ﬁcant reduction when the SPL2-expressin g bacteria were cultured in media con- taining 250 µM and 25 µM Cd (Figure 7D ‒F). The SPL2-expressing group also exhibited a tendency for enhanced Cd removal at 50 µM Cd. These ﬁndings suggest that SPL2-ex- pressing bacteria e ﬀectively sequester Cd from the culture medium. Figure 7. Removal of La and Cd ions from cultur e media by SPL2-expressing bacteria. ( A–C) La removal. ( D–F) Cd removal. Recombinant bacteria were cultured in liquid medium, and recombi- nant protein expression was induced. Bacteria were then exposed to the indicated concentrations of La (A‒C) or Cd ions ( D‒F) for 1.5 h. Following exposure, the bacterial cells were removed by cen- trifugation, and metal ion concentrations in the resulting supernatants were measured by ICP-MS.
ijms-26-01905-v3	experiments	Graphs show means from thr ee independent experiments. p-values were calculated using Student’s t-test or Welch’s t-test, comparing each bacterial group with the no-bacteria control (culture medium without bacteria), except for pa nel F. 6His, mCherry-6His; FLAG, ﬂag-mCherry; SPL2, SPL2- ﬂag- mCherry. , p < 0.05. , p < 0.01. **, p < 0.001. 2.5. SPL2 Expression Increases Cd Tolerance of Recombinant Bacteria We examined the impact of SPL2 expression on bacterial tolerance to lanthanides and transition metals. SPL2-expressing bacteria exhibited enhanced growth on Cd-containing plates, compared with control bacteria at Cd concentrations of 0.05 mM, 0.25 mM, and 0.5 mM (Figure 8A). In contrast, SPL2 expression did not promote growth in the presence of
ijms-26-01905-v3	results	Co, La, Tb, or Ni (Figure 8B,C). These results strongly suggest that SPL2 expression Figure 7. Removal of La and Cd ions from culture media by SPL2-expressing bacteria. ( A–C) La removal. ( D–F) Cd removal. Recombinant bacteria were cultured in liquid medium, and recombinant protein expression was induced. Bacteria were then exposed to the indicated concentrations of La(A–C)or Cd ions ( D–F) for 1.5 h. Following exposure, the bacterial cells were removed by centrifugation, and metal ion concentrations in the resulting supernatants were measured by ICP-
ijms-26-01905-v3	experiments	MS. Graphs show means from three independent experiments. p-values were calculated using Student’s t-test or Welch’s t-test, comparing each bacterial group with the no-bacteria control (culture medium without bacteria), except for panel F. 6His, mCherry-6His; FLAG, flag-mCherry; SPL2, SPL2-flag-mCherry. , p< 0.05. , p< 0.01. **, p< 0.001.
ijms-26-01905-v3	results	Consistent with the La removal results, Cd concentration in the supernatant showed a significant reduction when the SPL2-expressing bacteria were cultured in media containing 250µM and 25 µM Cd (Figure 7D–F). The SPL2-expressing group also exhibited a tendency
ijms-26-01905-v3	results	for enhanced Cd removal at 50 µM Cd. These findings suggest that SPL2-expressing bacteria effectively sequester Cd from the culture medium. 2.5. SPL2 Expression Increases Cd Tolerance of Recombinant Bacteria We examined the impact of SPL2 expression on bacterial tolerance to lanthanides and transition metals. SPL2-expressing bacteria exhibited enhanced growth on Cd-containing plates, compared with control bacteria at Cd concentrations of 0.05 mM, 0.25 mM, and 0.5 mM (Figure 8A). In contrast, SPL2 expression did not promote growth in the presence of
ijms-26-01905-v3	results	Co, La, Tb, or Ni (Figure 8B,C). These results strongly suggest that SPL2 expression specifically confers tolerance to Cd, but not to the other tested metals, in the recombinant bacteria. Int. J. Mol. Sci. 2025 , 26, x FOR PEER REVIEW 8 of 14 speciﬁcally confers tolerance to Cd, but not to the other tested metals, in the recombinant bacteria. Figure 8. Tolerance of SPL2-expressing bacteria to Cd, Co, and La. Recombinant bacteria expressing SPL2- ﬂag-mCherry or containing the control vector were cultured in liquid medium, and SPL2- FLAG-mCherry expression was induced. Ba cteria were serially diluted and spo tted on LB agar plates containing the indica ted concentrations of Cd ( A), Co ( B), or La ions ( C). Plates were incu- bated for 12 h, and images were captured. The cont rol vector-containing bacteria served as a nega-
ijms-26-01905-v3	experiments	tive control. Representative results from more than two independent experiments are shown. Con- trol images for Cd and Co are identical. In panel C, pairs of the control and SPL2 images were taken from the same image.
ijms-26-01905-v3	discussion	3. Discussion Microorganisms are a ttractive candidates for bioremediation and biomining owing to their e ﬃcient and selective metal accumulation. Genetic engineering further enhances their metal-binding ability and selectivity [25–3 0]. In this regard, the expression and solu- bility of SPL2- and lanpepsy-derived recombin ant proteins suggest their potential for ge- netic engineering. In this study, we expr essed three lanthanide-bi nding proteins—SPL2, lanmodulin, and lanpepsy—in E. coli strains to develop metal-accumulating recombinant bacteria. However, lanmodulin expression in hibited bacterial growth , preventing evalua- tion of its bioremediation potential. As previously shown, recombinant protein solubility is crucial for metal binding [39]. E ﬃcient expression and solubility of SPL2 and lanpepsy in bacterial cells (Figure 1A,B) suggest their suitability for genetic engineering for biore- mediation and biomining. scICP-MS analyses identi ﬁed SPL2 as a promising cand idate for bioremediation and
ijms-26-01905-v3	methodology	biomining. This technique provides precise detection and quanti ﬁcation of metal ions at the single-cell level, revealing detailed information on their distribution and concentration within individual cells [34,35]. SPL2-expressing bacteria exhibited enhanced binding of several metals, including La, Cd, Co, and Ni (Figures 2 and 4 ‒6). This enhanced binding was directly a ttributed to the SPL2 fragment, as bacteria expressing SPL2- ﬂag-mCherry showed a signi ﬁcant increase in metal binding compared with those expressing ﬂag- mCherry alone. Importantly, the enhanced binding translated into e ﬀective sequestration of La and Cd ions from the culture supernatant by SPL2-expressing bacteria (Figure 7). These ﬁndings highlight the potential of the SPL2 cytosolic fragment (V291 ‒S383) for ac- cumulating metal ions, particularly La and Cd, in recombinant bacteria, suggesting its suitability for bioremediation and biomining applications targeting heavy metals and REE. Figure 8. Tolerance of SPL2-expressing bacteria to Cd, Co, and La. Recombinant bacteria expressing SPL2-flag-mCherry or containing the control vector were cultured in liquid medium, and SPL2-FLAG- Int. J. Mol. Sci. 2025 ,26, 1905 8 of 13 mCherry expression was induced. Bacteria were serially diluted and spotted on LB agar plates containing the indicated concentrations of Cd ( A), Co ( B), or La ions ( C). Plates were incubated for 12 h, and images were captured. The control vector-containing bacteria served as a negative control.
ijms-26-01905-v3	experiments	Representative results from more than two independent experiments are shown. Control images for Cd and Co are identical. In panel C, pairs of the control and SPL2 images were taken from the same image.
ijms-26-01905-v3	discussion	3. Discussion Microorganisms are attractive candidates for bioremediation and biomining owing to their efficient and selective metal accumulation. Genetic engineering further enhances their metal-binding ability and selectivity [ 25–30]. In this regard, the expression and solubility of SPL2- and lanpepsy-derived recombinant proteins suggest their potential for genetic engineering. In this study, we expressed three lanthanide-binding proteins—SPL2, lanmod- ulin, and lanpepsy—in E. coli strains to develop metal-accumulating recombinant bacteria.
ijms-26-01905-v3	experiments	However, lanmodulin expression inhibited bacterial growth, preventing evaluation of its bioremediation potential. As previously shown, recombinant protein solubility is crucial for metal binding [ 39]. Efficient expression and solubility of SPL2 and lanpepsy in bacterial cells (Figure 1A,B) suggest their suitability for genetic engineering for bioremediation and biomining. scICP-MS analyses identified SPL2 as a promising candidate for bioremediation and
ijms-26-01905-v3	methodology	biomining. This technique provides precise detection and quantification of metal ions at the single-cell level, revealing detailed information on their distribution and concentration within individual cells [ 34,35]. SPL2-expressing bacteria exhibited enhanced binding of several metals, including La, Cd, Co, and Ni (Figures 2 and 4–6). This enhanced binding was directly attributed to the SPL2 fragment, as bacteria expressing SPL2-flag-mCherry showed a significant increase in metal binding compared with those expressing flag- mCherry alone. Importantly, the enhanced binding translated into effective sequestration of La and Cd ions from the culture supernatant by SPL2-expressing bacteria (Figure 7).
ijms-26-01905-v3	results	These findings highlight the potential of the SPL2 cytosolic fragment (V291–S383) for accumulating metal ions, particularly La and Cd, in recombinant bacteria, suggesting its suitability for bioremediation and biomining applications targeting heavy metals and REE.
ijms-26-01905-v3	methodology	We identify SPL2 as a novel candidate for Cd bioremediation, offering an approach dis- tinct from previously reported methods. While one study has described microbial-mediated mineralization of Cd, converting it to insoluble forms, such as carbonate, phosphate salts, and cadmium sulfide [ 26], and another study has demonstrated Cd adsorption using recom- binant bacteria expressing human ferritin fused with a synthetic phytochelatin gene [ 30],
ijms-26-01905-v3	results	our findings introduce the soluble, cytosolic fragment of SPL2 as a potential tool for Cd adsorption. The presence of SPL2-flag-mCherry in the soluble fraction after cell lysis (Figure 1B) suggests intracellular sequestration and accumulation of Cd. Furthermore, SPL2-flag-mCherry expression enhanced bacterial growth upon Cd exposure (Figure 8), in- dicating that SPL2 can also improve bacterial resistance to Cd toxicity, further highlighting its potential for Cd bioremediation. The imidazole group in histidine has been extensively studied for its applications in the removal of toxic metal ions from contaminated water using polymer-based membranes, including potential industrial applications [ 40]. In our previous study, we showed that recombinant bacteria expressing 6His-tagged proteins developed the capability to accu- mulate Ni and Co ions [ 39]. Conversely, while His-tagged proteins proved effective in the bacterial system, the cytosolic fragment of SPL2 may, in turn, be applicable in bioorganic membranes and polymer-based systems. Int. J. Mol. Sci. 2025 ,26, 1905 9 of 13
ijms-26-01905-v3	methodology	4. Materials and Methods 4.1. Plasmid Vectors To investigate the properties of several proteins, synthetic open reading frames (ORFs) were designed and cloned into the pET29b expression vector. The ORFs were synthesized by Twist Biosciences (San Francisco, CA, USA). The constructs included mCherry without any tags, as well as mCherry fused to a C-terminal hexahistidine (6His) tag (mCherry-6His) or an N-terminal FLAG epitope tag (FLAG-mCherry). Additionally, a fusion protein was generated by attaching a cytoplasmic fragment of SPL2 (V291–S383) to the C-terminus of FLAG-mCherry, resulting in the SPL2-FLAG-mCherry construct. Two additional constructs were designed to express FLAG-tagged versions of lanpepsy (D24–E175) and lanmod- ulin, referred to as lanpepsy-FLAG and lanmodulin-FLAG, respectively. In the case of lanpepsy, the N-terminal intrinsically disordered region was excluded to generate lanpepsy- FLAG. These constructs were then transformed into E. coli BL21(DE3) via electroporation. Transformed bacteria were subsequently cultured on LB agar plates containing 50 µg/mL kanamycin for selection. 4.2. Expression and Solubility of Recombinant Proteins We used the IPTG induction system because it is widely utilized in the field of molecu- lar biology. Recombinant protein expression was performed in E. coli BL21(DE) as previ- ously described [ 41,42]. Briefly, a single bacterial colony was grown overnight at 37◦C in LB medium supplemented with 50 µg/mL kanamycin. The preculture was then diluted and grown at 37◦C until an OD 600of 0.4–0.6 was reached. Protein expression was induced with 0.5 mM IPTG (Takara Bio Inc., Shiga, Japan) for 3 h at 37◦C. Cells were harvested by centrifugation and washed once with buffer [50 mM Tris–HCl (pH 7.5), 10% sucrose].
ijms-26-01905-v3	methodology	Two distinct lysis methods were employed: for evaluating total protein expression, cells were directly lysed in Laemmli SDS lysis buffer by boiling at 95◦C for 5 min. For assessing protein solubility, cells were suspended in lysis buffer [20 mM NaPi (pH 6.8), 300 mM NaCl, 0.5 mM PMSF, protease inhibitor cocktail], treated with 1 mg/mL lysozyme for 15 min on ice, flash-frozen in liquid nitrogen, and subjected to three freezing-thawing cycles followed by sonication (Tomy, Tokyo, Japan, UR-20P sonicator, level 5, 10 s pulses repeated 10 times on ice). The resulting lysate was clarified by centrifugation, and the supernatant was analyzed by SDS-PAGE and Coomassie Brilliant Blue (CBB) staining.
ijms-26-01905-v3	discussion	4.3. E. coli Preparation for scICP-MS Analysis A single colony of the recombinant bacteria was cultured overnight in LB medium supplemented with kanamycin. The culture was grown at 37◦C to mid-logarithmic phase after dilution with fresh medium. Protein expression was then induced with 0.1 mM IPTG, and bacterial cells were cultured at 18◦C for 18 h. To investigate metal binding, bacterial cells were exposed to 250 µM of either lanthanum chloride (FUJIFILM Wako Pure Chemical Corporation, Inc., Osaka, Japan, Cat. No. 123-04222), terbium chloride (FUJIFILM Wako Pure Chemical Corporation, Inc., Cat. No. 206-14491), cadmium chloride (FUJIFILM Wako Pure Chemical Corporation, Inc., Cat. No. 032-00122), nickel chloride (Nacalai Tesque, Inc., Kyoto, Japan, Cat. No. 24223-92), or cobalt chloride (Nacalai Tesque, Inc., Kyoto, Japan, Cat. No. 09208-72) for 1.5 h. To remove unbound metal ions, the exposed cells were washed extensively (three times) with 0.9% sodium chloride solution (99.999%, Sigma-Aldrich, St.
ijms-26-01905-v3	discussion	Single-cell ICP-MS analysis was performed as described previously [ 36,37,39]. Metal ion-exposed bacteria were suspended and introduced into an Agilent 8900 ICP-MS/MS Int. J. Mol. Sci. 2025 ,26, 1905 10 of 13 system (Agilent Technologies, Hachioji, Japan) using a dedicated single-cell sample in- troduction system (Glass Expansion, Melbourne, Australia) consisting of a MicroMist concentric glass nebulizer, a total consumption spray chamber, and a micro syringe pump
ijms-26-01905-v3	discussion	(MSP-1D, AS ONE, Osaka, Japan). Time-resolved analysis was used to sequentially ac- quire signals of all elements of interest. The mass ( m, in grams) of La, Tb, Cd, Co, and Ni in individual bacterial cells was then calculated by comparing the measured signal intensities from the cells ( ICell) to those from an ionic standard solution ( IStd) using the following equation. m=ICell IStd−IBlk×tdwell×f×CStd×v IBlkrepresents the signal intensity of 0.9% sodium chloride blank solution; tdwell , the signal integration or dwell time, was set to 0.1 ms; f, the nebulization or transport efficiency, was determined using SiO 2nanoparticles (200 nm, Sigma-Aldrich, Cat No. 231-545-4) and an ionic silicon standard (Kanto Chemical Co., Inc., Tokyo, Japan, Cat. No. 37811- 2B), following the procedure described in our previous study [ 39];CStdrepresents the concentration of the ionic standard, set at 100 ng/mL; and vrepresents the sample flow rate, set at 0.015 mL/min. All calculations, including automatic particle baseline calibration, were performed using ICP-MS MassHunter software 5.2 (Agilent Technologies). Detailed
ijms-26-01905-v3	discussion	operational settings and instrumental conditions for scICP-MS analysis are summarized in Table 1. Table 1. ICP-MS Instrumentation and Collision/Reaction Operational Settings. Instrument Agilent 8900 ICP-MS/MS ICP incident power 1600 W Ar carrier gas 0.60 L/min Make-up gas 0.25 L/min Integration time 0.1 ms Collision/reaction cell H 2: 5.5 mL/min (La, Tb, Cd, Ni, Co, Si) O2: 0.38 mL/min (P) Signal monitoring period 40 s Sample injection rate 0.015 mL/min In the collision/reaction cell, hydrogen (H 2) gas was used to analyze La, Tb, Cd, Ni, Co, and Si, while oxygen (O 2) gas was used to analyze P . We selected the reaction gases for Si, P , Ni, and Co based on our previous manuscript [ 39]. To minimize spectral interferences arising from atmospheric and solvent-derived contaminants, H 2was used as a reaction gas
ijms-26-01905-v3	methodology	in the analysis of La, Tb, and Cd. This approach is expected to reduce interference from polyatomic species, such as nitrogen- and oxygen-containing ions. Single-cell data were analyzed using GraphPad Prism 9.0. The mean mass of metal ions in each bacterial cell was determined by fitting a Gaussian distribution to the corresponding
ijms-26-01905-v3	experiments	experiments were presented as means ±standard deviation. Statistical significance was assessed using Student’s t-test or Welch’s t-test, as appropriate. Unless otherwise stated, comparisons were made between the control vector and the mCherry-expressing groups. 4.5. Heavy Metal Tolerance of SPL2-Expressing Bacteria To assess metal tolerance, recombinant bacteria expressing SPL2-flag-mCherry were cultured overnight in LB medium containing 50 µg/mL kanamycin. Bacteria transformed with the empty pET29b vector served as a control. Following dilution of the preculture and growth at 37◦C to the mid-logarithmic phase, SPL2 expression was induced with Int. J. Mol. Sci. 2025 ,26, 1905 11 of 13 0.5 mM IPTG. After a 4 h cultivation, bacterial cultures were serially diluted (2-, 3-, 4-, 5, and 10-fold) with fresh LB medium. Three microliters of each dilution was then spotted onto LB agar plates supplemented with various concentrations of different metal ions. After 12 h at 37◦C, the plates were photographed to assess bacterial growth. A metal-free LB plate served as a negative control. 4.6. Removal of Heavy Metals by SPL2-Expressing Bacteria To analyze metal concentrations in the bacterial growth medium, SPL2-expressing bacteria were cultivated overnight at 37◦C in LB medium supplemented with kanamycin. Following dilution with fresh LB medium and growth to the mid-logarithmic phase, SPL2 expression was induced with 0.1 mM IPTG, and the bacteria were cultured at 18◦C for 18 h. Subsequently, metal ions were added, and the cultures were incubated at 37◦C for 1.5 h. Bacterial cells were then removed by centrifugation (10,000 ×g, 10 min, 4◦C), and the
ijms-26-01905-v3	discussion	resulting supernatant was collected for metal concentration analysis by ICP-MS. Data from three independent experiments were presented as means ±standard deviation. Statistical significance was determined using Student’s t-test or Welch’s t-test.
ijms-26-01905-v3	conclusion	5. Conclusions Our scICP-MS analyses demonstrated that the expression of the cytosolic fragment of SPL2 enhanced the accumulation of several metals—La, Co, and Ni—within bacterial cells. Remarkably, SPL2 also exhibited Cd-binding ability, enabling SPL2-expressing re- combinant bacteria to effectively sequester this metal from the culture medium, along with La. Furthermore, SPL2 expression conferred increased tolerance to Cd toxicity exposure
ijms-26-01905-v3	results	in the recombinant bacteria. These multifaceted findings suggest that the metal-binding ability of SPL2 offers a novel and promising strategy for bioremediation and biomining using recombinant bacteria. Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ijms26051905/s1.
ijms-26-01905-v3	methodology	Author Contributions: Conceptualization, Y.F. and Y.O.; Methodology, Y.T.; Validation, Y.F. and E.L.; Formal Analysis, Y.F. and E.L.; Investigation, E.L.; Resources, Y.T. and Y.O.; Data Curation, Y.F. and E.L.; Writing—Original Draft Preparation, Y.F. and E.L.; Writing—Review & Editing, Y.F., Y.T., N.S. and Y.O.; Visualization, Y.F. and E.L.; Supervision, N.S. and Y.O.; Project Administration, Y.F.; Funding Acquisition, Y.F., Y.T., N.S. and Y.O. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported in part by Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology [24K09793 (Y.F.), 24H00749 (Y.O.), 24K21304 (Y.O.), and 22K05345 (N.S.)]. This work was supported by an ACT-UR Grant [#4366, #4489 (Y.T.)] from Agilent Technologies. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data is contained within the article or Supplementary Material.
ijms-26-01905-v3	acknowledgment	Acknowledgments: We gratefully acknowledge Agilent Technologies for support through an ACT- UR Grant. Conflicts of Interest: The authors declare no conflicts of interest. Int. J. Mol. Sci. 2025 ,26, 1905 12 of 13
ijms-26-01905-v3	references	References 1. Kowalkowski, T.; Pastuszak, M.; Szparaga, A.; Samczynski, Z.; Polkowska-Motrenko, H.; Buszewski, B. Rare earth elements in fine fraction (<20 mum) of the Vistula River sediments. Chemosphere 2019 ,237, 124442. [PubMed] 2. Zepf, V . Rare Earth Elements: A New Approach to the Nexus of Supply, Demand and Use: Exemplified along the Use of Neodymium in Permanent Magnets ; Springer: Berlin/Heidelberg, Germany, 2013. 3. Dushyantha, N.; Batapola, N.; Ilankoon, I.M.S.K.; Rohitha, S.; Premasiri, R.; Abeysinghe, B.; Ratnayake, N.; Dissanayake, K. The story of rare earth elements (REEs): Occurrences, global distribution, genesis, geology, mineralogy and global production. Ore Geol. Rev. 2020 ,122, 103521. [CrossRef] 4. Balaram, V . Rare earth elements: A review of applications, occurrence, exploration, analysis, recycling, and environmental impact. Geosci. Front. 2019 ,10, 1285–1303. [CrossRef] 5. Gwenzi, W.; Mangori, L.; Danha, C.; Chaukura, N.; Dunjana, N.; Sanganyado, E. Sources, behaviour, and environmental and human health risks of high-technology rare earth elements as emerging contaminants. Sci. Total Environ. 2018 ,636, 299–313. [CrossRef] 6. Wang, Z.; Zhang, X.; Mu, Y. Effects of rare-earth fertilizers on the emission of nitrous oxide from agricultural soils in China. Atmos. Environ. 2008 ,42, 3882–3887. [CrossRef] 7. Basu, A.; Chakrabarty, K.; Chatterjee, G.C. Neurotoxicity of lanthanum chloride in newborn chicks. Toxicol. Lett. 1982 ,14, 21–25. [CrossRef] 8. Palasz, A.; Czekaj, P . Toxicological and cytophysiological aspects of lanthanides action. Acta Biochim. Pol. 2000 ,47, 1107–1114. [CrossRef] 9. Feng, L.; Xiao, H.; He, X.; Li, Z.; Li, F.; Liu, N.; Zhao, Y.; Huang, Y.; Zhang, Z.; Chai, Z. Neurotoxicological consequence of long-term exposure to lanthanum. Toxicol. Lett. 2006 ,165, 112–120. [CrossRef] 10. Xie, X.; Tan, X.; Yu, Y.; Li, Y.; Wang, P .; Liang, Y.; Yan, Y. Effectively auto-regulated adsorption and recovery of rare earth elements via an engineered E. coli .J. Hazard. Mater. 2022 ,424, 127642. [CrossRef] 11. Abashina, T.; Vainshtein, M. Current Trends in Metal Biomining with a Focus on Genomics Aspects and Attention to Arsenopyrite Leaching-A Review. Microorganisms 2023 ,11, 186. [CrossRef] 12. Liapun, V .; Motola, M. Current overview and future perspective in fungal biorecovery of metals from secondary sources. J. Environ. Manag. 2023 ,332, 117345. [CrossRef] [PubMed] 13. Park, D.M.; Reed, D.W.; Yung, M.C.; Eslamimanesh, A.; Lencka, M.M.; Anderko, A.; Fujita, Y.; Riman, R.E.; Navrotsky, A.; Jiao, Y. Bioadsorption of Rare Earth Elements through Cell Surface Display of Lanthanide Binding Tags. Environ. Sci. Technol. 2016 ,50, 2735–2742. [CrossRef] [PubMed] 14. Park, D.M.; Brewer, A.; Reed, D.W.; Lammers, L.N.; Jiao, Y. Recovery of Rare Earth Elements from Low-Grade Feedstock Leachates Using Engineered Bacteria. Environ. Sci. Technol. 2017 ,51, 13471–13480. [CrossRef] [PubMed] 15. Martin, K.; Hosam, M.S. Introductory Chapter: Introducing Heavy Metals. In Heavy Metals ; Hosam El-Din, M.S., Refaat, F.A., Eds.; IntechOpen: Rijeka, Croatia, 2018; pp. 3–11. 16. Vhahangwele, M.; Khathutshelo, L.M. Environmental Contamination by Heavy Metals. In Heavy Metals ; Hosam El-Din, M.S., Refaat, F.A., Eds.; IntechOpen: Rijeka, Croatia, 2018; pp. 115–133. 17. Armah, F.A.; Obiri, S.; Yawson, D.O.; Onumah, E.E.; Yengoh, G.T.; Afrifa EK, A.; Odoi, J.O. Anthropogenic sources and environmentally relevant concentrations of heavy metals in surface water of a mining district in Ghana: A multivariate statistical
ijms-26-01905-v3	methodology	approach. J. Environ. Sci. Health A 2010 ,45, 1804–1813. [CrossRef] 18. Razzak, S.A.; Faruque, M.O.; Alsheikh, Z.; Alsheikhmohamad, L.; Alkuroud, D.; Alfayez, A.; Hossain SM, Z.; Hossain, M.M. A comprehensive review on conventional and biological-driven heavy metals removal from industrial wastewater. Environ. Adv. 2022 ,7, 100168. [CrossRef] 19. Ramya, S.L.; Vinay Kumar, C.; Sudhamani, M. Application of Biosorption for Removal of Heavy Metals from Wastewater. In Biosorption ; Jan, D., Branislav, V ., Eds.; IntechOpen: Rijeka, Croatia, 2018. 20. Liu, S.H.; Zeng, G.M.; Niu, Q.Y.; Liu, Y.; Zhou, L.; Jiang, L.H.; Tan, X.F.; Xu, P .; Zhang, C.; Cheng, M. Bioremediation mechanisms of combined pollution of PAHs and heavy metals by bacteria and fungi: A mini review. Bioresour. Technol. 2017 ,224, 25–33. [CrossRef] 21. Znad, H.; Awual, M.R.; Martini, S. The Utilization of Algae and Seaweed Biomass for Bioremediation of Heavy Metal- Contaminated Wastewater. Molecules 2022 ,27, 1275. [CrossRef] 22. Salama, E.S.; Roh, H.S.; Dev, S.; Khan, M.A.; Abou-Shanab RA, I.; Chang, S.W.; Jeon, B.H. Algae as a green technology for heavy metals removal from various wastewater. World J. Microbiol. Biotechnol. 2019 ,35, 75. [CrossRef] 23. Arumugam, N.; Chelliapan, S.; Kamyab, H.; Thirugnana, S.; Othman, N.; Nasri, N.S. Treatment of Wastewater Using Seaweed: A Review. Int. J. Environ. Res. Public Health 2018 ,15, 2851. [CrossRef] 24. Alabssawy, A.N.; Hashem, A.H. Bioremediation of hazardous heavy metals by marine microorganisms: A recent review. Arch. Microbiol. 2024 ,206, 103. [CrossRef] Int. J. Mol. Sci. 2025 ,26, 1905 13 of 13 25. Roy, R.; Samanta, S.; Pandit, S.; Naaz, T.; Banerjee, S.; Rawat, J.M.; Chaubey, K.K.; Saha, R.P . An Overview of Bacteria-Mediated Heavy Metal Bioremediation Strategies. Appl. Biochem. Biotechnol. 2024 ,196, 1712–1751. [CrossRef] [PubMed] 26. Zheng, Y.; Xiao, C.; Chi, R. Remediation of soil cadmium pollution by biomineralization using microbial-induced precipitation: A review. World J. Microbiol. Biotechnol. 2021 ,37, 208. [CrossRef] [PubMed] 27. Anand, S.; Singh, A.; Kumar, V . Recent advancements in cadmium-microbe interactive relations and their application for environmental remediation: A mechanistic overview. Environ. Sci. Pollut. Res. Int. 2023 ,30, 17009–17038. [CrossRef] 28. Hansda, A.; Kumar, V .; Anshumali. A comparative review towards potential of microbial cells for heavy metal removal with emphasis on biosorption and bioaccumulation. World J. Microbiol. Biotechnol. 2016 ,32, 170. [CrossRef] 29. Saravanan, A.; Kumar, P .S.; Ramesh, B.; Srinivasan, S. Removal of toxic heavy metals using genetically engineered microbes: Molecular tools, risk assessment and management strategies. Chemosphere 2022 ,298, 134341. [CrossRef] 30. Tian, L.; Wang, D.; Liu, Y.; Wei, M.; Han, X.; Sun, X.; Yin, L.; Luo, G. Construction of Genetically Engineered Escherichia coli Cell Factory for Enhanced Cadmium Bioaccumulation in Wastewater. Water 2024 ,16, 1759. [CrossRef] 31. Tracz, M.; Gorniak, I.; Szczepaniak, A.; Bialek, W. E3 Ubiquitin Ligase SPL2 Is a Lanthanide-Binding Protein. Int. J. Mol. Sci. 2021 , 22, 5712. [CrossRef] 32. Cotruvo, J.A.; Jr Featherston, E.R.; Mattocks, J.A.; Ho, J.V .; Laremore, T.N. Lanmodulin: A Highly Selective Lanthanide-Binding Protein from a Lanthanide-Utilizing Bacterium. J. Am. Chem. Soc. 2018 ,140, 15056–15061. [CrossRef] 33. Hemmann, J.L.; Keller, P .; Hemmerle, L.; Vonderach, T.; Ochsner, A.M.; Bortfeld-Miller, M.; Gunther, D.; Vorholt, J.A. Lanpepsy is a novel lanthanide-binding protein involved in the lanthanide response of the obligate methylotroph Methylobacillus flagellatus. J. Biol. Chem. 2023 ,299, 102940. [CrossRef] 34. Meyer, S.; Lopez-Serrano, A.; Mitze, H.; Jakubowski, N.; Schwerdtle, T. Single-cell analysis by ICP-MS/MS as a fast tool for cellular bioavailability studies of arsenite. Metallomics 2018 ,10, 73–76. [CrossRef] 35. Ho, K.-S.; Chan, W.-T. Time-resolved ICP-MS measurement for single-cell analysis and on-line cytometry. J. Anal. At. Spectrom. 2010 ,25, 1114–1122. [CrossRef] 36. Tanaka, Y.; Iida, R.; Takada, S.; Kubota, T.; Yamanaka, M.; Sugiyama, N.; Abdelnour, Y.; Ogra, Y. Quantitative Elemental Analysis of a Single Cell by Using Inductively Coupled Plasma-Mass Spectrometry in Fast Time-Resolved Analysis Mode. Chembiochem 2020 ,21, 3266–3272. [CrossRef] [PubMed] 37. Tanaka, Y.; Katayama, H.; Iida, R.; Ogra, Y. Quantitative elemental analysis of human leukemia K562 single cells by inductively coupled plasma mass spectrometry in combination with a microdroplet generator. J. Anal. At. Spectrom. 2025 ,40, 216–225. [CrossRef] 38. Ogra, Y.; Tanaka, Y.; Suzuki, N. Recent advances in copper analyses by inorganic mass spectrometry. J. Clin. Biochem. Nutr. 2022 , 71, 2–6. [CrossRef] 39. Tanaka, Y.; Shimazaki, S.; Fukumoto, Y.; Ogra, Y. Detection of Histidine-Tagged Protein in Escherichia coli by Single-Cell Inductively Coupled Plasma-Mass Spectrometry. Anal. Chem. 2022 ,94, 7952–7959. [CrossRef] 40. Kaczorowska, M.A. The Use of Polymer Inclusion Membranes for the Removal of Metal Ions from Aqueous Solutions-The Latest Achievements and Potential Industrial Applications: A Review. Membranes 2022 ,12, 1135. [CrossRef] 41. Fukumoto, Y.; Kyono, R.; Shibukawa, Y.; Tanaka, Y.; Suzuki, N.; Ogra, Y. Differential molecular mechanisms of substrate recognition by selenium methyltransferases, INMT and TPMT, in selenium detoxification and excretion. J. Biol. Chem. 2024 ,300, 105599. [CrossRef] 42. Fukumoto, Y.; Nakayama, Y.; Yamaguchi, N. Human Rad17 C-terminal tail is phosphorylated by concerted action of CK1delta/epsilon and CK2 to promote interaction with the 9-1-1 complex. Biochem. Biophys. Res. Commun. 2019 ,517, 310–316. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10_1016_j_jct_2018_12_034	chunk_1	J. Chem. Thermodynamics 132 (2019) 491-496 Contents lists available at ScienceDirect TEEJOURICA HERMODYNAMIC J. Chem. Thermodynamics ELSEVIER journal homepage: www.elsevier.com/locate/jct Bio- and mineral acid leaching of rare earth elements from synthetic phosphogypsum Paul J. Antonick a.1, Zhichao Hu a.1, Yoshiko Fujita b,1, David W. Reed b,1, Gaurav Das , Lili Wu d Radha Shivaramaiah d, Paul Kim ?, Ali Eslamimanesh , Malgorzata M. Lencka , Yongqin Jiao e, Andrzej Anderko , Alexandra Navrotsky d, Richard E. Riman a,* aMaterialsScience andEngineering,RutgersUniversity,607TaylorRoad,Piscataway,NJ08854,United States bBiological and ChemicalProcessing Department,Idaho National Laboratory,Idaho Falls,ID 83415,United States OLI Systems, Inc.,240 Cedar Knolls Road,Suite 301,Cedar Knolls,NJ07927,United States PeterAockThermochemistryLaboratory andNEATRU,UniversityofCaliforniaDavis,Davis,CA95616,United Stat eBiosciences and Biotechnology Division,Physical and Life Sciences Directorate,Lawrence Livermore National Laboratory,Livermore,CA 94550,United States ＡRＴICＬE ＩNＦＯ ABSTRACT Article history: Leaching of six individual rare earth (yttrium, cerium, neodymium, samarium, europium, and yterbium) Received 31 October 2018 doped synthetic phosphogypsum samples using a suite of lixiviants was conducted. The lixiviants chosen Received in revised form 18 December 2018 for this study were phosphoric acid, sulfuric acid, gluconic acid, and a “biolixiviant" consisting of spent Accepted 19 December 2018 medium containing organic acids from the growth of the bacterium Gluconobacter oxydans on glucose. Available online 24 December 2018 The biolixiviant had a pH of 2.1 and the dominant organic acid was determined to be gluconic acid, pre- sent at a concentration of 220 mM. The leaching behaviors of the studied lixiviants were compared and Keywords: Rare earth recovery rationalized by thermodynamic simulations. The results suggest that at equivalent molar concentrations Bioleaching of 220 mM the biolixiviant was more efficient at rare earth element (REE) extraction than gluconic acid Biohydrometallurgy and phosphoric acid but less efficient than sulfuric acid. Unlike the organic acids, at pH 2.1 the mineral Mineral acid leaching acids failed to extract REE, likely due to different complexation and kinetic effects. 2019 Elsevier Ltd. 1. Introduction accomplished by the wet digestion of phosphate rock, which results in about 5 metric tons of PG waste for every ton of phos- Rare earth elements (REE) are a group of chemically similar ele- phate rock processed [8]. Several variations of the wet digestion ments that include the lanthanides, scandium, and yttrium [1]. process exist; the most common one (dihydrate process) results These elements are subject to increasing demand due to their cru- in 70-80% of the REE in the phosphate rock feed migrating to the cial roles in magnets, catalysts, phosphors, and new clean energy PG waste stream but other processes (hemihydrate, hemi- technologies [2]. In 2011, the U.s. Department of Energy released dihydrate processes) can result in PG containing more than 95% a report highlighting supply risks for REE [3]. Roughly 90% of all of the original REE [1]. It is estimated that 250 million tons of phos- REE are currently produced in China and during the last several phate rock were used by the fertilizer industry every year for phos- years interest in REE production has increased in the United States phoric acid production, and the average REE content in phosphate in order to address this supply risk [2-7]. rock was 0.046 wt%. About 100 kt of REE end up in the wet process In nature, REE are often associated with phosphate deposits [8] waste materials. In comparison, current annual world production and one potential source of REE is phosphogypsum (PG), a waste of REE oxides is about 126 kt [1]. From these numbers it is evident product of phosphoric acid production for the fertilizer industry. that REE recovery from fertilizer industry PG wastes could be a PG waste is primarily composed of gypsum (CaSO4.2H2O) but also potential solution to the global REE supply challenge. contains other minerals, unreacted phosphate rock, and some pro- Recent work indicates that most REE in PG may occur as REE cess water [9]. 90% of the world's production of phosphoric acid is phosphates adsorbed to the surface of the gypsum as a secondary phase [10]. This surface association suggests that the REE should be relatively susceptible to extraction by aqueous chemical agents, as * Corresponding author. compared to a case where REE are incorporated into the crystal lat- E-mail address: riman@rutgers.edu (R.E. Riman). tice. Among previously studied inorganic acids and organic liquids, 1 These authors contributed equally. https:/doi.org/10.1016/jct.2018.12.034 0021-9614/@ 2019 Elsevier Ltd. 492 P.J. Antonick et al./J. Chem. Thermodynamics 132 (2019) 491-496 sulfuric acid can liberate REE which are trapped in the gypsum gravimetric analysis (TGA) for purity and water content, respec- [11]. After leaching, REE recovery from the leachate can be accom- tively. XRD was performed on powders using a Bruker D8 plished by precipitation, solvent extraction, or ion exchange [12]. Discover (Bruker AXS, Inc, Madison, Wisconsin, USA) unit with a Other studies have proposed the use of bioleaching as an alter- 20 step size of 0.018° and a 0.5 s dwell time. The instrument was native to conventional hydrometallurgy in an effort to address the operated at 40 kV and 40 mA (1600 Watts) with a Vantec 1 Detec- environmental liabilities of conventional leaching processes [13]. tor, copper K-alpha source, and a horizontal goniometer. MDI Jade Leaching using sulfuric acid produced by microbially catalyzed oxi- 9 (Materials Data, Inc., Livermore, California, USA) was used for dation of sulfide ores is the most commercially exercised bioleach- pattern analysis. The Fachinformationszentrum Karlsruhe/National ing practice; 18-20% of world's mined copper is currently Institute of Standards and Technology Inorganic Crystal Structure produced this way [14]. But this type of bioleaching is generally Database (FIZ/NIST ICSD, National Institute of Standards and Tech- not practical for nonsulfidic ores such as phosphates [15,7] nology, Gaithersburg, Maryland, USA) was used to compare XRD Bioleaching can also be performed using organic acids produced patterns of samples to reference values. XRD patterns are displayed by heterotrophic microorganisms. In addition to providing acidity, in Fig. 1. Powders were also characterized using TGA on a TA organic acids can promote leaching through complexation of metal Instruments Q5000IR (New Castle, Delaware, USA) unit. A heating ions [16,17]. A biologically produced lixiviant ("biolixiviant") pro- rate of 10 °C/min to 1000 °℃ with a 50 mL/min nitrogen flow rate duced by the bacterium Gluconobacter oxydans has been shown was used. Sample loading plates were high temperature ceramic to be promising for REE recovery from other non-PG waste streams tipped platinum pans, and TA Universal Analysis (TA Instruments, [16]. This biolixiviant primarily consists of gluconic acid but likely New Castle, Delaware, USA) was used to analyze hydration states. contains other components which promote REE complexation. Biolixiviant containing a measured gluconic acid content of 2.2. Lixiviant preparation 12.5 mM leached more REE from spent fluid catalytic cracking (FCC) catalyst than a 90 mM gluconic acid solution prepared in The lixiviants included in the study were phosphoric acid, sulfu- the laboratory [16]. These catalysts are composed primarily of sil- ric acid, commercial gluconic acid, and biolixiviant. The biolixiviant ica and alumina, in contrast to the gypsum which dominates PG. used was spent medium from a culture of Gluconobacter oxydans Whether the biolixiviant can solubilize REE from waste PG is strain NRRL B58 [18], grown on glucose in a bioreactor as described unknown. previously [19].
10_1016_j_jct_2018_12_034	chunk_2	The gluconic acid content of the biolixiviant, mea- This study examined the ability of mineral acids, the gluconic sured by high performance liquid chromatography as described acid dominated biolixiviant as well as commercial gluconic acid previously [16], was 220 mM and the pH was 2.1. To determine to extract REE from synthetic PG samples that had been individu- whether the spent microbiological medium behaved differently ally doped with yttrium, cerium, neodymium, samarium, euro- from pure gluconic acid, a solution of commercial gluconic acid pium, or ytterbium. Synthetic PG was used because it enabled was prepared at 220 mM; the pH of this solution was also 2.1. the determination of the exact chemical composition of the pre- For the sulfuric acid and phosphoric acids, two sets of lixiviants leached PG (and therefore accurate determination of the percent were prepared. The first set consisted of each of the acids at titers yield) and also the direct comparison of the different lixiviants resulting in a pH of 2.1. The second set consisted of each of the because of the consistent sample compositions. Real phosphogyp- acids prepared at 220 mM in deionized water. sum wastes are variable in composition and therefore differences between the treatments could be attributable to differences in 2.3. Leaching studies starting materials, rather than leaching efficiency. The two mineral acids tested were sulfuric and phosphoric acid. Sulfuric acid is used REE-doped (REE =Y, Ce, Nd, Sm, Eu, and Yb) PG solid samples in the most common wet digestion scheme discussed above and were treated with pH 2.1 solutions of the four lixiviants at a 2% phosphoric acid is the product of that process, meaning both acids pulp density (equivalent to 20 mg solid in 1 mL lixiviant) and incu- would be readily available on site at fertilizer production plants bated at 25 °℃ for 24 h, with shaking at 150 rpm. The solutions [1]. Onsite availability of the lixiviants could improve the eco- were then filtered (0.22 μm pore size PES filters) and the metal ions nomics of REE recovery from fertilizer wastes. Each PG sample in the supernatants were measured using ICP-OES. Results were (doped with a single REE) was leached in each acid system. The tabulated to compare leaching efficiencies for each set of condi- experimental results were supplemented by thermodynamic sim- tions. A control group using deionized water as a lixiviant was also ulations which facilitated greater understanding of the systems prepared and tested. Each leaching experiment was performed in studied. triplicate. The same procedure was then performed for all six REE-doped PG with 220 mM solutions of the mineral acid lixiviants. The 2. Methods biolixiviant and gluconic acid leaching experiments were not repeated since at pH 2.1 gluconic acid was already at a concentra- 2.1. PG synthesis and solid phase characterization tion of 220 mM. All solid PG samples were prepared using the same procedure. 2.4. PG solution phase and leachate characterization Ca(NO3)2-4H2O and REE(NO3)3 at a molar ratio of 100:1 (250 mmol: 2.5 mmol) were dissolved in DI water. In a separate ICP-OES analysis for determination of aqueous ion concentra- container, Na2SO4 and NaH2PO4 at a molar ratio of 100:1 tions was performed on a Perkin Elmer Optima 7300 DV (Waltham, (250 mmol: 2.5 mmol) were also dissolved. The solutions were Massachusetts, USA) unit. Calibration curves were produced using slowly mixed under stirring and then aged for 24 h in an 80 °℃ SCP Science (Champlain, New York, USA) ICP standards and ele- water bath. The solids were collected using vacuum filtration and ment wavelengths were discarded if the R correlation value was washed with an amount of deionized water equal to three times below 99%. All measured concentration values fell within the range the reaction volume. The resulting powders were then dried for of the calibration standards (1-1000 ppm). Element wavelengths 24 h at 90 °C. (nm) used were 317.393 and 315.887 for Ca; 213.617 and The REE doped were Y, Ce, Nd, Sm, Eu, or Yb. Powders were 214.914 for P; 181.975 and 182.563 for S; 588.995 for Na; characterized using powder X-Ray Diffraction (XRD) and thermo- 371.029, 324.227, and 360.073 for Y; 418.66 for Ce; 406.109, P.J.Antonick et al./J.Chem.Thermodynamics 132(2019)491-496 493 Table 1 Empirical formulas of REE-doped PG. Sample Formula （n` YPG CaYo.010Na0.018(PO4)0.010(SO4)1.009-0.552H20 CePG CaCe0.011Na0.030(PO4)0.01(SO4)1.015-0.541H20 NdPG CaNdo.010Na0.020(PO4)0.008SO4)1.013-0.542H20 su SmPG CaSm0.010Nao.024(PO4)0.010(SO4)1.012-0.564H20 EuPG CaEu0.013Na0.021(PO4)0.010(SO4)1.015-0.560H2O YbPG CaYbo.011Na0.024(PO4)0.011(SO4)1.012-0.559H2O Table 2 REE Leaching using 220 mM Lixiviants. REE Dopant Lixiviant Leachate REE (ppm) % REE Dissolved Y H3PO4 102.9 ±12.0 85.0 ± 10.3 H2SO4 97.3 ± 6.6 80.6± 5.8 Y Bio Lix 112.2 ± 1.6 91.2 ± 1.0 10 20 30 40 50 60 Y GA Lix 102.4 ± 6.3 84.6 ± 5.2 20 (%) Ce H3PO4 10.3 ± 0.5 5.0 ±0.2 Ce H2SO4 193.0 ± 8.1 93.7 ± 3.8 Fig. 1. Normalized XRD patterns of reference (CaSO-0.8HzO, bottom) and each Ce Bio Lix 72.2 ± 1.7 36.7 ±0.8 REE-doped PG solid sample in the order of YPG, CePG, NdPG, SmPG, EuPG, and YbPG Ce GA Lix 75.0 ± 1.1 35.7 ± 0.1 (bottom to top) above the reference. Nd H3PO4 15.8 ± 4.9 7.9 ± 2.5 Nd H2SO4 174.8 ± 5.2 87.5 ± 1.9 Nd Bio Lix 84.9 ± 7.3 42.8 ± 3.2 401.225, and 430.358 for Nd; 359.26 for Sm; 381.967 and 412.970 Nd GA Lix 80.3 ± 6.6 40.4 ± 3.7 for Eu; and 289.138 for Yb. In the cases where multiple element Sm H3PO4 17.8 ± 1.4 8.5 ± 1.1 Sm H2SO4 183.4 ± 14.8 89.6 ± 7.3 wavelength calibration curves fit the criteria delineated above, Sm Bio Lix 149.3 ± 8.7 73.2 ± 2.2 results from the different element wavelengths were arithmeti- Sm GA Lix 133.9 ± 9.2 65.9 ± 2.5 cally averaged to produce the final concentration values. Empirical Eu H3PO4 44.2 ± 12.6 16.5 ± 4.7 formulas (Table 1) were determined for each powder by com- Eu H2SO4 205.4± 20.7 76.9 ± 8.1 Eu Bio Lix pletely dissolving a small amount of sample in dilute nitric acid 134.5 ± 3.0 50.0± 0.8 Eu GA Lix 84.4 ± 4.3 31.2 ± 1.5 and characterizing the solutions using inductively coupled Yb H3PO4 107.9 ±8.5 42.3 ± 3.4 plasma-optical emission spectroscopy (ICP-OES). Yb H2SO4 230.1 ± 19.5 90.6 ± 7.6 Yb Bio Lix 209.3 ± 3.3 83.7 ±2.4 Yb GA Lix 189.3 ± 8.1 75.1 ± 0.7 3. Results and discussion * Bio Lix is biolixiviant and GA Lix is commercial gluconic acid. Percent REE dis- solved is calculated based on the total REE in the initial solid sample. Error is the 3.1. Mineral acid leaching at pH 2.1 95% confidence interval based on triplicate experiments. Experiments performed at a pH of 2.1 showed no detectable REE leached for sulfuric and phosphoric acids (or the water control group); therefore, the data are not shown. For the biolixiviant between the two lixiviants with a 6.9 percent difference. Another interesting trend observed was the difference in leaching perfor- and the commercial gluconic acid, a pH of 2.1 coincided with the mance between the biolixiviant and the commercial gluconic acid; concentration of 220 mM, so the results for those lixiviants are dis- in all cases the biolixiviant was more effective than the commercial cussed in the next section. gluconic acid in solubilizing the REE. Cerium and neodymium leaching results were the most similar for the two organic acid lix- 3.2. 220 mM lixiviant leaching iviants with differences in percent leached of only 1.0 and 2.4 respectively. The greatest difference between the biolixiviant and The 220 mM lixiviant leaching results were tabulated and can the gluconic acid was observed for europium (18.8%). Fig. 3 dis- be seen in Table 2. REE concentrations measured in the leachates plays these differences calculated for each PG sample. are reported in ppm along with the percentage of REE removed from the original PG solids. Several trends were observed from this experiment. Phosphoric 3.3.Thermodynamicsimulation acid is generally the least effective leaching agent, followed by the commercial gluconic acid, biolixiviant, and finally sulfuric acid.
10_1016_j_jct_2018_12_034	chunk_3	To gain further understanding of the behavior of REE in acid Yttrium is the only element studied which does not follow this solutions, thermodynamic simulations were performed using the trend; in the case of Y-doped PG all four lixiviants performed sim- Mixed-Solvent Electrolyte (MSE) model of Wang et al [20]. The ilarly. Ytrium is also the only element for which the leaching per- MSE model is suitable for reproducing both phase and chemical formance of phosphoric acid was comparable to the other equilibria in systems containing inorganic and organic acids, salts, lixiviants. In the PG sample doped with cerium, the phosphoric organic solvents, and complexing agents. In previous studies, the acid leached a mere 5% of REE present in the solid, the lowest per- model was parameterized and verified for the key classes of com- centage observed out of all samples tested. A graphical representa- ponents that are studied here, i.e., phosphoric acid [21], sulfuric tion of the data can be seen in Fig. 2. acid [22], rare earth salts in aqueous environments [23], and The difference between the leaching efficiencies for the sulfuric organic acids and their complexes with REE [24]. acid and the biolixiviant was the greatest for the cerium doped PG; Using the MSE model, solubility of NdPO4 was calculated as a the sulfuric acid was more than twice as effective as the biolixi- function of pH in phosphoric, sulfuric, and gluconic acids (Fig. 4). viant. Ytterbium doped PG displayed a much more similar behavior The biolixiviant could not be modeled explicitly because its exact 494 P.J.Antonick et al./J.Chem.Thermodynamics 132(2019) 491-496 85.0 91.2 93.7 100 80.6 84.6 100 50 50 %) 36.7 35.7 5.0 0 H,PO4 HPO4 H2SO4 Bio Lix GA Lix H2SO4 Bio Lix GA Lix (payoea7 %) pN 100 100 89.6 87.5 73.2 65.9 50 42.8 40.4 50 Ws 7.9 8.5 0 01 H,PO4 H2SO4 Bio Lix GA Lix H,PO4 H,SO4 Bio Lix GA Lix (o 90.6 100 76.9 100 83.7 75.1 50.0 50 50 42.3 %) n3 31.2 %) 人 16.5 0 0 HsPO4 H,SO4 Bio Lix GA Lix H,PO4 H,SO4 Bio Lix GA Lix Fig. 2. Leaching effciency for REE recovery from synthetic PG with phosphoric acid (orange), sulfuric acid (red), biolixiviant (green), and gluconic acid (blue), as a percent of leached REE over initial REE in the solids. Error bars are 95% confidence interval based on triplicate experiments.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 20- 1E+1 18.8 1E+0 1E-1 15 1E-2 gluconic acid 三 1E-3 HsPO4 H2SO4 1E-4 X! PN 1E-5 10- 8.6 1E-6 7.3 1E-7 - acid conc. =220 mM 6.6 1E-8 5 pH = 2.1 !!! 1E-9- 0.5 1.5 2 2.5 2.4 3 pH 1.0 Fig. 4. Calculated solubility of NdPO4 in gluconic acid (blue line), sulfuric acid (red V Ce Nd Sm Eu Yb line), and phosphoric acid (orange line) as a function of pH. The square symbols show the solubilities in the three acids at the natural pH of 220 mM solutions. Fig.3.Average differences of REE leached from PG by biolixiviant(Bio Lix）and Dashed line is at pH 2.1. (For interpretation of the references to color in this figure gluconic acid (GA Lix). legend, the reader is referred to the web version of this article.) composition is not known; in addition to gluconic acid, the micro- plexes between the neodymium and gluconate ions.The symbols biological spent medium contains other organic components that in Fig. 4 show the solubilities at the natural pH of 220 mM acid cannot be precisely identified and quantified [16]. As expected, solutions. The pH of a 220 mM gluconic acid solution is 2.1 the solubility of rare earth phosphate is a function of pH. At a fixed whereas the natural pH of the mineral acids is substantially lower pH, the solubility in phosphoric acid is predicted to be the lowest due to the higher strength of these acids. Thus, the solubility in due to the common ion effect; however, the REE solubilities in sul- 220 mM acid solutions results from the combination of the pH furic acid and gluconic acid are relatively close. At lower pH values effect and the specific complexation/common ion effects. As shown (below ~2), the solubility in gluconic acid is predicted to be higher in Fig. 4, the predicted solubility in 220 mM acids follows the order than in sulfuric acid. This is due to the formation of aqueous com- sulfuric acid > gluconic acid > phosphoric acid. This order is consis- P.J. Antonick et al./J. Chem.Thermodynamics 132 (2019) 491-496 495 tent with the experimentally observed effectiveness of the three ducted prior to upscaling design and technoeconomic predictions non-biolixiviants (cf. Fig. 2). for potential extraction processes with these lixiviants in order to In the thermodynamic simulations, neodymium was used as a have more accurate parameters for real world processes. representative REE for modeling the PG leaching phenomena. The conclusions drawn from the simulations can be extrapolated to 4.4. Sulfuric acid leaching constraints due to double sulfate formation other REE because of similarities in the acid-base and complexa- and common ion effects tion chemistry of various REE. It should be noted that the thermo- dynamic model does not necessarily capture the full complexity of The formation of less soluble double sulfates has been reported the leaching of rare earth phosphates in 24-hour experiments in studies of sulfuric acid leaching of REE [26,27]. For this reason, because it does not include kinetic or mass-transfer effects associ- acid concentration plays an important role in limiting the final ated with dissolution from a synthetic phosphogypsum matrix. amount of REE in sulfuric acid leachates. It was found that forma- However, equilibrium solubility can be expected to be a key factor tion of double sulfates is enhanced over the solvation of REE ions in determining the dissolution and, in fact, the predicted thermo- when the concentration of sulfuric acid exceeds 12% [26]. The dynamic solubility trends are consistent with experimental leach- 220 mM sulfuric acid used in this study has a concentration around ing trends for all tested REE except yttrium. 2%, so it is unlikely that double sulfate formation inhibited the sol- vation of REE ions in the experiments performed. A recent study of 4. Conclusions industrial PG leaching using 1.5 M and 3.0 M mineral acids showed that H2SO4 was less efficient than HCl and HNO3 due to the com- 4.1. Comparison of biolixiviant and commercial gluconic acid mon ion effect [28]. Any future work dealing with higher concen- trations of sulfuric acid should take these considerations into Results show that there is an appreciable benefit to using the account. biolixiviant for PG leaching. Although both biological and gluconic acid lixiviants were primarily composed of gluconic acid, the 5. Summary biolixiviant leached more REE than the commercial gluconic acid for every element studied. This benefit was not particularly strong REE leaching from synthetic REE doped PG samples was evalu- for cerium and neodymium, but for europium the biolixiviant was ated for sulfuric acid, phosphoric acid, a biolixiviant (composed particularly effective with 18.8% greater leaching than the gluconic primarily of gluconic acid), and commercial gluconic acid both at acid. This observation warrants further research investigating why pH 2.1 and concentrations of 220 mM. Elements studied were Y, such a dramatic difference was observed for europium in the Ce, Nd, Sm, Eu, and Yb. It was found that the mineral acids leached biolixiviant as compared to the other elements studied. negligible amounts of REE at pH 2.1, but performed much better at a concentration of 220 mM. For all elements except yttrium, it was 4.2. Comparison of mineral and organic acids found that the sulfuric acid leached the most REE and phosphoric acid leached the least. The biolixiviant always outperformed the The apparent lack of any REE solubilization from the synthetic gluconic acid with the greatest difference measured for europium PGs by the mineral acids at pH 2.1 indicates the role of complexa- doped PG which leached an additional 18.8% of the europium from tion for the organic acids.
10_1016_j_jct_2018_12_034	chunk_4	However, the thermodynamic simula- the solid compared to the gluconic acid. Trends from the thermo- tions suggest that this may be primarily a kinetic effect as it was dynamic simulations using the MSE model agree with the experi- predicted that the solubility of neodymium phosphate would be mental observations but predicted much higher leaching for the comparable in gluconic acid and sulfuric acid (Fig. 4). The simula- mineral acids at pH 2.1 than were observed. This suggests that tions predict equilibrium concentrations while the leaching exper- kinetics limited REE leaching in the 24-hour experiments. iment was performed for 24 h, implying that the mineral acid systems may need more time to approach equilibrium than the Acknowledgments organic acid systems. At 220 mM acid concentrations, the trends predicted by the simulations were consistent with the empirical This research was supported by the Critical Materials Institute, results as sulfuric acid was the most effective leaching agent, and an Energy Innovation Hub funded by the U.s. Department of phosphoric acid was less effective than the organic acids despite Energy Office of Energy Efficiency and Renewable Energy, being more acidic at this concentration than the organic acids. Sul- Advanced Manufacturing Office, and conducted under Rutgers furic acid leached the most REE for every REE-doped PG except Contracts SC-13-394-436163 and SC-18-475-123975, D0E Idaho yttrium, where the studied lixiviants performed similarly. The Operations Office Contract DE-AC07-05ID14517, and Lawrence biolixiviant had a greater leaching capacity than phosphoric acid Livermore National Laboratory Contract DE-AC52-07NA27344. in every case. Studies on associated costs and environmental Accordingly, the U.s. Government retains a nonexclusive, royalty- impacts of the different lixiviants at varying concentrations should free license to publish or reproduce the published form of this con- be performed to decide if the biolixiviant is an overall better choice tribution, and allow others to do so, for U.s. Government purposes. for REE leaching than sulfuric acid. References 4.3. Deviation from real world samples [1] S. Wu, L. Wang, L. Zhao, P. Zhang, H. El-Shall, B. Moudgil, X. Huang, L. Zhang, Chem. Eng. J. 335 (2018) 774. REE content in industrial PG ranges from 0.01 to 0.4 wt% [1], a [2] K. Binnemans, P.T. Jones, B. Blanpain, T. Van Gerven, Y. Yang, A. Walton, M. much lower amount than the ~1 wt% REE of the synthetic PG pow- Buchert, J. Cleaner Prod. 51 (2013) 1. ders used for this study. It has been shown that higher REE concen- [3] D. Bauer, D. Diamond, J. Li, M. McKittrick, D. Sandalow, P. Telleen, J. Shore, J. Hackworth, C. Lieder, F. Fields, A. Campbell, D. Vashishat, B. Wanner, D. trations can lead to lower extraction efficiencies in leaching Sandalow, et al., Critical Materials Strategy, U.s. Department of Energy, 2011. experiments [25]. This means that while the numbers presented [4] G.A. Moldoveanu, V.G. Papangelakis, Hydrometallurgy 117-118 (2012) 71. from this set of experiments are useful for relative comparisons [5] D.D. Imholte, R.T. Nguyen, A. Vedantam, M. Brown, A. Iyer, BJ. Smith, J.W. Collins, C.G. Anderson, B. O'Kelley, Energy Policy 113 (2018) 294. of the different lixiviants, leaching studies of industrial PG samples [6] J. Chen, X. Zhu, G. Liu, W. Chen, D. Yang, Resour., Conserv. Recycl. 132 (2018) which have mixed REE at lower concentrations should be con- 139. 496 P.J. Antonick et al./J. Chem.Thermodynamics 132 (2019) 491-496 [7] S. Abramov, J. He, D. Wimmer, M.-L. Lemloh, E.M. Muehe, B. Gann, E. Roehm, R. [18] T. Asai, K. Shoda, J. Gen. Appl. Microbiol. 4 (1958) 289. Kirchhof, M.G. Babechuk, R. Schoenberg, H. Thorwarth, T. Helle, A. Kappler, [19] V.S. Thompson, M. Gupta, H. Jin, E. Vahidi, M. Yim, M.A. Jindra, V. Nguyen, Y. Waste Manage. 79 (2018) 735. Fujita, J.W. Sutherland, Y. Jiao, D.W. Reed, ACS Sustainable Chem. Eng. 6 (2018) [8] P. Emsbo, P.I. McLaughlin, G.N. Breit, E.A. du Bray, A.E. Koenig, Gondwana Res. 1602. 27 (2015) 776. [20] P. Wang, A. Anderko, R.D. Young, Fluid Phase Equilib. 203 (2002) 141. [9] C.R. Canovas, F. Macias, R. Perez-Lopez, M.D. Basallote, R. Millan-Becerro, J. [21] P. Wang, R.D. Springer, A. Anderko, R.D. Young, Fluid Phase Equilib. 222-223 Cleaner Prod. 174 (2018) 678. (2004)11. [10] R. Shivaramaiah, W. Lee, A. Navrotsky, D. Yu, P. Kim, H. Wu, Z. Hu, R. Riman, A. [22] P. Wang, A. Anderko, R.D. Springer, R.D. Young, J. Mol. Liq. 125 (2006) 37. Anderko, Am. Mineral. 2016 (1854) 101. [23] G. Das, M.M. Lencka, A. Eslamimanesh, A. Anderko, R.E. Riman, Fluid Phase [11] E.P. Lokshin, O.A. Tareeva, I.P. Elizarova, Russ.J. Appl. Chem. 83 (2010) 958. Equilib. 425 (2017) 16. [12] M.K. Jha, A. Kumari, R. Panda, J. Rajesh Kumar, K. Yoo, J. Lee, Y. [24] D.M. Park, D.W. Reed, M.C. Yung, A. Eslamimanesh, M.M. Lencka, A. Anderko, Y. Hydrometallurgy 161 (2016) 77. Fujita, R.E. Riman, A. Navrotsky, Y. Jiao, Environ. Sci. Technol. 50 (2016) [13] C.L. Brierley, J.A. Brierley, Appl. Microbiol. Biotechnol. 97 (2013) 7543. 2735. [14] C.L. Brierley, in: V.1. Lakshmanan, R. Roy, V. Ramachandran (Eds.), Innovative [25] E.P. Lokshin, A.V. Vershkov, Y.A. Vershkova, Metally (Moscow) (2001) 42. Process Development in Metallurgical Industry: Concept to Commission, [26] E.P. Lokshin, K.G. Ivlev, O.A. Tareeva, Russ. J. Appl. Chem. 78 (2005) 1761. Springer International Publishing, Cham, 2016, p. 109. [27] E.P. Lokshin, O.A. Tareeva, T.G. Kashulina, Russ. J. Appl. Chem. 81 (2008) 1. [15] N. Jain, D.K. Sharma, Geomicrobiol J. 21 (2004) 135. [28j M. Walawalkar, C.K. Nichol, G. Azimi, Hydrometallurgy 166 (2016) 195. [16] D.W. Reed, Y. Fujita, D.L. Daubaras, Y. Jiao, V.S. Thompson, Hydrometallurgy 166 (2016) 34. [17] V.L. Brisson, W.-Q. Zhuang, L. Alvarez-Cohen, Biotechnol. Bioeng. 113 (2016) 339. JCT 18-132
10_1007_s00792-012-0457-9	abstract	Abstract Many extremophilic microorganisms are poly- extremophiles, being confronted with more than one stress condition. For instance, some thermoacidophilic microor-ganisms are in addition capable to resist very high metal concentrations. Most likely, they have developed special adaptations to thrive in their living environments. Inorganicpolyphosphate (polyP) is a molecule considered to be primitive in its origin and ubiquitous in nature. It has many roles besides being a reservoir for inorganic phosphate andenergy. Of special interest are those functions related to survival under stressing conditions in all kinds of cells. PolyP may therefore have a fundamental part in extremo-philic microorganism’s endurance. Evidence for a role of polyP in the continued existence under acidic conditions, high concentrations of toxic heavy metals and elevated saltconcentrations are reviewed in the present work. Actual evidence suggests that polyP may provide mechanistic alternatives in tuning microbial ﬁtness for the adaptationunder stressful environmental situations and may be of crucial relevance amongst extremophiles. The enzymes involved in polyP metabolism show structure conservation amongst bacteria and archaea. However, the lack of a canonical polyP synthase in Crenarchaea, which greatlyaccumulate polyP, strongly suggests that in this phylum a different enzyme may be in charge of its synthesis.Keywords Inorganic polyphosphate /C1Acidophilic bacteria /C1Thermoacidophilic archaea /C1Metal resistance /C1 Environmental stress
10_1007_s00792-012-0457-9	introduction	Introduction Currently described as ubiquitous and versatile molecules, inorganic polyphosphates (polyP) are linear polymers consisting of dozens to hundreds of orthophosphate resi- dues (Pi) linked by high-energy phosphoanhydride bonds.The possible prebiotic origin (Yamagata et al .1991 ) together with a set of physicochemical features have placed polyP as a ‘‘key’’ molecule within early course of lifeevolution (Kulaev and Kulakovskaya 2000 ). At ﬁrst, polyPs were classically considered as energy and phos- phate reservoirs. However, their regulation and functionremained unknown for many years due to lack of speciﬁc

Subsets and Splits

No community queries yet

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