Upload folder using huggingface_hub

manual_annotation/200pdfs_in_mmd/10.1002_1097-4601(200103)33:3<182::aid-kin1012>3.0.co;2-d.mmd

@@ -0,0 +1,175 @@
+Rate Constants and Activation Energies for Ozonolysis of Isoprene Methacrolein and Methyl-Vinyl-ketone in Aqueous Solution: Significance to the In-Cloud Ozonation of Isoprene
+
+T. Pedersen
+
+1Institute of Chemistry, Laboratory 5, University of Copenhagen, 20 Universitetsparken 5, DK-2100 Kobenhaven O, Denmark 1
+
+KNU SHESTED
+
+2Risp National Laboratory, Denmark 22 February 2000; accepted 12 October 2000
+
+###### Abstract
+
+Rate constants and activation energies for the reactions of ozone with isoprene, methacrolein, and methyl-vinyl-ketone in aqueous solution have been determined at temperatures from 5 to 30\({}^{\circ}\)C, using the stopped-flow-technique and monitoring ozone decay. The rate constants at 25\({}^{\circ}\)C and the activation energies have been found to be 4.1 (\(\pm\)0.2) \(\times\) 10\({}^{\circ}\) M\({}^{-1}\) s\({}^{-1}\) and 19.9 (\(\pm\)0.5) kJ mol\({}^{-1}\) for isoprene, 2.4 (\(\pm\)0.1) \(\times\) 10\({}^{\circ}\) M\({}^{-1}\) s\({}^{-1}\) and 23.9 (\(\pm\)0.5) kJ mol\({}^{-1}\) for methacrolein, and 4.4 (\(\pm\)0.2) \(\times\) 10\({}^{\circ}\) M\({}^{-1}\) s\({}^{-1}\) and 18.0 (\(\pm\)0.5) kJ mol\({}^{-1}\) for methyl-vinyl-ketone. A UV spectrum of a transient intermediate with a lifetime of about 15 s formed during the ozonation of isoprene was obtained in the range 220 to 300 nm. It rises steadily toward 220 nm. It is suggested that the spectrum can be attributed to the two unsaturated Criege-intermediates (carbonyl oxides), which would conceivably be stabilized by resonance. Lifetime considerations indicate that the oxidation of isoprene and its first-generation reaction products, methacrolein and methyl-vinyl-ketone, by ozone and OH in the aqueous phase of a cloud environment play only a minor role compared to homogeneous gas-phase processing. (c) 2001 John Wiley & Sons, Inc. Int J Chem Kinet 33: 182-190, 2001
+
+## Introduction
+
+The present investigation originated from the down-to-earth question: "Does wet linen that has been hung out in the sun to dry smell of ozone?"-- as is a popular belief. The two isoprene daughters, methacrolein and methyl-vinyl-ketone, both have rather acrid smells that could resemble the smell of ozone when dilute. It was therefore suggested that the oxidation of the ambient isoprene molecule, followed by uptake of the two carbonyl compounds into the liquid phase of the linen,could explain the smell, as there is no easy mechanism for generating ozone under the circumstances. Be that as it may, for similar reasons, isoprene chemistry in cloud environments might be of interest.
+
+Lelieveld and Crutzen [1] studied the photochemically induced reactivity of methane and methane daughters in the atmosphere in the presence of clouds and concluded that such an environment was of importance for the overall processing of methane in the atmosphere. Isoprene (2-methyl-1,3-butadiene: ISO) is one of the major hydrocarbons of biogenic origin, its annual emission rate comparing with that of all other volatile organic compounds (excluding methane) taken together according to Guenter [2]. The reactivity of isoprene is much higher than that of methane owing to its unsaturated nature, thus increasing its ozone-forming potential (cf. Derwent [3]).
+
+In order to investigate the importance of reactions in droplets (clouds, fogs, or smoke plumes) of ISO and the first-generation reaction products (henceforth "isoprene daughters"), methacrolein (CH\({}_{2}\)=C(CH\({}_{3}\))CHO, 2-methylpropenal: MAC) and methyl-vinyl-ketone (CH\({}_{3}\)C(O)CH=CH\({}_{2}\), 3-butene-2-on: MVK), the rate constants for ozonation of ISO, MAC, and MVK were determined in aqueous solution at a number of temperatures in the range from 5 to 30degC.
+
+Although the Henry's law constant for isoprene (cf. McAucliffe [4] and Mackay and Shiu [5]) is rather small (1.30 \(\times\) 10\({}^{-2}\) M atm\({}^{-1}\)), supposedly making its reactivity in cloud water less important, the same might not be the case for the isoprene daughters, which have 2 - 3 orders of magnitude larger Henry's law constants: 4.29 M atm\({}^{-1}\) for MAC and 21.5 M atm\({}^{-1}\) for MVK (Allen et al. [6]). The same is true for the third product of ozonation of isoprene, formaldehyde, which was studied in aqueous solution in reaction with OH, by Marcovic and Sehested [7] and Bothe and Schulte-Frohlinde [8]. We shall not consider the latter
+
+Figure 1: **(a)** An oscilloscope trace of the time evolution of the ozone absorbance in the reaction between isoprene and ozone at 10°C. **(b)** Pseudo-first-order plot of the same data. The regression line has been fitted by inspection; see text.
+
+molecule in this paper; the reader is referred to the work of Lelieveld and Crutzen [1].
+
+Considerations about lifetimes are used to discuss the importance of cloud processing of the isoprene-related reaction products. A new approach assuming the validity of Henry's law throughout the processing is presented in the appendix.
+
+The gas-phase oxidation of ISO with ozone has been discussed by Paulson and Seinfeld [9] and Paulson et al. [10,11], and that of MAC and MVK by Grosjean et al. [12].
+
+## 1 Experimental
+
+The samples were used as purchased. ISO and MVK were obtained from Aldrich with stated purities of 99%; methacrolein came from ICN Biomedicals and had no stated purity. Proton NMR spectra (in CDCl3 or D2O) showed that all the samples were sufficiently pure (99% or better) to be used without further processing. NMR spectroscopy was also used to ascertain that the aqueous stock solutions were sufficiently stable if used within 6 h. There was no trace of a hydrate for MAC, and none was expected for MVK.
+
+The aqueous stock solutions were prepared by dissolving the samples in triply distilled water. For ISO, a saturated aqueous solution was first prepared in a separatory funnel. The solubility of ISO at room temperature was given as 642 g/m3 (9.42 x 10-3 M) [4]. For the other two compounds, 250 \(\mu\)L pure samples were pipetted out (with preweighed Finn-pipettes) and added to about 200 mL water in 250 mL volumetric flasks. These were subsequently filled to the mark. We found that this procedure prevented significant loss by evaporation of the rather volatile compounds MVK and MAC. From these primary stock solutions, 1.00 mL aliquots were diluted to 100 mL in volumetric flasks with water acidified with perchloric acid to pH = 2 (in order to stabilize ozone). These secondary stock solutions were replaced every 4 h. The concentrations were 47.1 \(\mu\)M for ISO, 59.5 \(\mu\)M for MAC, and 57.8 \(\mu\)M for MVK at room temperature. No correction was attempted for the lower temperatures. Ozone concentrations were kept at about 6 \(\mu\)M in order to create pseudo-first-order-reaction conditions. It would seem desirable to use a more extreme concentration ratio than has been done here (100:1, say); this, however, was not experimentally feasible (see further discussion in the next section).
+
+The stopped-flow instrument was a Hi-Tech SF51 Stopped-flow Spectrophotometer equipped with an UV lamp. This instrument has a mixing time of about 1 ms, which enables us to measure first-order rate constants up to 700 s-1. The reactions were followed by monitoring the ozone absorption maximum at 260 nm (e260 = 3292 M-1 cm-1, Hart et al. [13]). Temperatures in the thermostat were kept within \(\pm\) 0.1 K. We also tried to monitor the decay of the organic reactants, keeping ozone concentrations 10 times higher than the organic reactants; however, we found the above approach to be superior in terms of uncertainties.
+to the logarithm of the ozone concentration C\({}_{\rm Ozone}\) (absorbance divided by the absorptivity of ozone and by cell length) in Figure 1(b). In Figure 1(b), a linear fit is made to the "linear" part of the extinction vs. time curve, and it can be seen that an increasing deviation from linearity evolves after a few lifetimes.
+
+Judging from this behavior, we believe that the ratio we have been using (8-10) is satisfactory within the overall uncertainty we are aiming at (about 10% for each individual measurement). The averaging over the results of the experiments is thought to improve the accuracy by a factor of 2.5\(-\)3. We accordingly quote the final results with a 5% uncertainty.
+
+The second-order rate constants, \(k_{x}\), given in Tables 1\(-\)III are obtained by dividing the averaged first-order rate constants \(\kappa_{x}\) (\(X\) = ISO, MVK, and MAC) by the pertinent concentrations, which were given previously.
+
+## Results for Isoprene
+
+The reaction scheme for the ozonolysis of isoprene is
+
+\[\begin{array}{l}\includegraphics[width=142.26378pt]{142.26378pt}\end{array}\]
+
+The primary ozonides (5-membered rings) rapidly fall apart into stable carbonyl compounds and meta- or unstable Criegee-intermediates (carbonyl oxides). In a later section, arguments are presented for the metastability of the two carbonyl oxides in this particular case. Table 1 shows the rate constants for ozonation of isoprene determined at six different temperatures. The graph of ln(\(k\)) vs. 1/\(T\) is given in Figure 2. From the regression line, the activation energy is found to be 19.9 (\(\pm\)0.5) kJ/mole and \(k_{25}\) = 4.1 (\(\pm\)0.2) \(\times\) 10\({}^{4}\) M\({}^{-1}\) s\({}^{-1}\).
+
+## Results for Methyl-Vinyl-Ketone
+
+The activation energy is found to be 18.0 (\(\pm\)0.5) kJ/mole and \(k_{25}\) = 4.4 (\(\pm\)0.2) \(\times\) 10\({}^{4}\) M\({}^{-1}\) s\({}^{-1}\).
+
+## Results for Methacrolein
+
+The activation energy is found to be 23.9 (\(\pm\)0.5) kJ/mole and \(k_{25}\) = 2.4 (\(\pm\)0.1) \(\times\) 10\({}^{4}\) M\({}^{-1}\) s\({}^{-1}\).
+
+## An Intermediate in the Reaction of Isoprene with Ozone
+
+The oscilloscope traces of UV absorbance, used to monitor the reactions between MAC and ozone and between MVK and ozone, came to zero absorbance in the course of 50 ms at room temperature, whereas for isoprene it took about 1 min to reach the same level [part of this behavior can be seen in Fig. 1(a)]. We take this as evidence for the formation of a meta-stable intermediate with a lifetime of about 15 s in the latter case. One can think of the two possible Criegee-intermediates (more systematically called carbonyl oxides), obtainable by (1) addition of ozone and (2) the splitting off of the terminal CH2-group as formaldehyde, as candidates for this intermediate:
+
+\[\begin{array}{l} {\rm CH}_{2} = {\rm C}({\rm CH}_{3}) - {\rm CH} = {\rm O}^{ \cdot } - {\rm O}^{ \cdot } \longleftrightarrow \\ {\rm CH}_{2} = {\rm C}({\rm CH}_{3}) - {\rm CH}^{ \cdot } - {\rm O} - {\rm O}^{ \cdot } \longleftrightarrow \\ {\rm CH}_{2} ^{ \cdot } {\rm C}({\rm CH}_{3}) = {\rm CH} - {\rm O} - {\rm O}^{ \cdot } \\ \end{array}\]
+
+\[\begin{array}{l} {\rm O} - {\rm O} - {\rm O} = {\rm C}({\rm CH}_{3}) - {\rm CH} = {\rm CH}_{2} \longleftrightarrow \\ {\rm O} - {\rm O} - {\rm O} - {\rm C}({\rm CH}_{3}) = {\rm CH} - {\rm CH}_{2} ^{ \cdot } \\ \end{array}\]
+
+Although carbonyl oxides are usually too short-lived to be studied per se (see Sander [14] for a review), enhanced stability of these particular carbonyl oxides is conceivable owing to conjugation (cf. the resonance formulas shown).
+
+A low-resolution UV spectrum was obtained by adding together equal amounts of a 94 \(\mu\)M isoprene solution and a 12 \(\mu\)M ozone solution in the stopped-flow instrument (at room temperature). The monitoring wavelength was changed by 10 nm from one experiment to the next between 220 and 300 nm. Only those below 260 nm are shown in the spectrum, the absorbance being essentially zero between 250 and 300 nm. The spectrum is shown in Figure 3.
+
+A spectrum of a hypothetical reaction mixture at 300 ms obtained by modeling the reaction sequence, using the rate constants determined in this work --i.e., assuming that no intermediate is present --has been overlaid for comparison. (The spectrum has been scaled so that its absorbance matches the maximum absorbance of the intermediate at 220 nm.)
+
+Obviously, the observed spectrum cannot be attributed to the normal reactants. Also, the fact that the intermediate decays within about 15 s is taken as evidence for the candidacy of the carbonyl oxides as the intermediate. Formaldehyde is another reaction product, but it appears in all three reactions, and nothing unexpected showed up for MAC and MVK, as mentioned previously.
+
+## Significance for in-cloud chemistry
+
+The system of ISO, MVK, and MAC together with ozone and hydroxyl radicals in an air-parcel containing a cloud with liquid water content \({\rm LWC}=0.1 - 10\) g per m\({}^{3}\) (or, in terms of the dimensionless volume
+
+Figure 2: Graph of ln(_k_) vs. 1/_T_ for ozonation of isoprene.
+
+-to-volume ratio, _w_: 10-7-10-5) was analyzed in terms of the pertinent lifetimes:
+
+1. The lifetimes \(\tau_{\text{OR}(\text{kg})}\) and \(\tau_{\text{OR}(\text{kg})}\) of reactions in the gas phase
+2. The lifetimes \(\tau_{\text{OR}(\text{upo})}\) and \(\tau_{\text{OR}(\text{upo})}\) of uptake into a cloud droplet
+3. The lifetimes \(\tau_{\text{OR}(\text{aq})}\) and \(\tau_{\text{OR}(\text{aq})}\) of reaction in the aqueous phase
+4. The overall lifetime \(\tau_{\text{cloud}}\) (expression is derived in the appendix)
+
+The lifetimes (1) and (2) depend only on the distribution of ozone and OH between the two phases (Henry's law), presuming that the uptake lifetimes are short compared to the lifetimes pertaining to the homogeneous phases. The lifetime for uptake of compound \(X\) is estimated according to Schwartz [15]:
+
+\[\tau_{X}=\frac{a^{2}}{3D_{g;X}}+\frac{4a}{3\nu_{X}\alpha_{X}}\]
+
+Following Lelieveld and Crutzen [1], \(a\), the droplet radius, is taken to be 10-5 m; gas-phase diffusivities are set equal to 10-5 m2 s-1 for all reactants; \(\nu_{X}\), the mean molecular speeds, are equal to (8_kT_/_p__m__X_)1/2 \(\approx\) 500 m/s, whence the lifetimes end up around 5 \(\mu\)s. As we shall see below, this is very short compared to the homogeneous-phase chemical lifetimes, and effects from uptake are therefore neglected henceforth. This implies that Henry's law equilibrium of all gases can be safely assumed.
+
+The concentrations of ozone and OH were assumed to be 30 ppbv and 0.1 ppbv, respectively, in the gas phase (cf. Seinfeld and Pandis [16]), independent of cloud formation, while the aqueous-phase concentrations were determined using Henry's law. For OH, the value of 25 M atm-1 adopted by Lelieveld and Crutzen was used, and for ozone, 0.011 M atm-1 (cf. Kozak-Channing and Heltz [17]). Rate constants for the liquid-phase reactions between ISO, MVK, MAC, and OH were estimated as 1 x 109 M-1 s-1, the approximate diffusion limited value. The rate constants determined in this work were used for the ozone reactions in solution. Hence, the lifetimes (1) and (3) were calculated as
+
+\[\tau_{X;Y}=\frac{1}{k_{X;Y}C_{Y}},\] \[X=\text{ISO, MAC, or MVK;}\qquad Y=\text{O}_{3}\text{ or OH}\]
+
+with the pertinent rate constants and concentrations inserted. These lifetimes are given in Table 4. While the lifetimes indicate that the processes are in general speeded up in the aqueous phase, an estimate of the
+
+Figure 3: Low-resolution (1 point for each 10 nm) UV spectrum of observed intermediate. Upper spectrum is that of a hypothetical reaction mixture; see text.
+
+overall lifetime of the species is needed in order to assess the relevance of clouds. One will, of course, expect the water content of the cloud environment (expressed as the dimensionless volume-to-volume fraction \(w\) of liquid water to air) to play a crucial role in the overall lifetimes. A derivation of an overall lifetime is presented in the appendix. It presumes that the Henry's law equilibrium is universally attained-- as was argued above. Based on formula (5) in the appendix and the individual lifetimes in Table 4, we obtain the overall lifetimes of Table 5. To get insight into this table, the OH-contribution has been omitted (by artificially increasing OH-lifetimes to values that much exceed the ozone lifetimes, typically 1 mio. h). Thereby the ozone contributions can be assessed approximately. This is done in Table 6. It emerges from Tables 6 and 6 that OH dominates the picture completely. It also emerges that ISO and MVK are not much affected by the presence of a cloud. Somewhat surprisingly, MAC is special in that the cloud plays a substantial role in its oxidation. This is due to a particularly slow OH-process in the gas phase for MAC.
+
+As far as Table 6 is concerned, then, in no case does the aqueous-phase ozonation play a role that exceeds a few percent. This can be attributed to the low concentrations of O3 in the aqueous phase owing to its small Henry constant.
+
+## Discussion
+
+We have obtained rate constants and activation energies for the ozone oxidation of isoprene, methyl-vinyl-ketone, and methacrolein in aqueous solution. It is interesting to note that the order of the rate constants for ozonation of the three molecules in the gas phase is the same. For the gas phase, the rate constants (relative to ISO) ISO:MAC:MVK are 1:0.08:0.33, while in the aqueous phase the ratios are 1:0.49:0.89. The activation energies (in kJ) follow a similar trend in the two phases: g/aq: 15.9/19.9:17.6/23.9:12.6/18, being consistently lower, however, in the gas phase.
+
+During the measurements on isoprene, a transient intermediate was observed, and a low-resolution UV spectrum of it was recorded. It is a rather featureless spectrum (partly because it is of low resolution); its intensity rises steadily toward 220 nm. It is suggested that Criegee-intermediates (carbonyl oxides) may be responsible, and the possibility of resonance stabilization for the particular carbonyl oxides suggested is used as an argument for this.
+
+An expression allowing for the calculation of the lifetime of isoprene etc. in an atmosphere containing clouds is derived in the appendix. Using this expression, it can be seen that clouds are generally only of minor importance in the processing of isoprene etc. in the atmosphere; only for OH-processing is there a substantial contribution and only for methacrolein, where the lifetime drops by about 50% due to the presence of the cloud. This estimate is based on the assumption of a diffusion limited rate constant in the aqueous phase, where no data are available.
+
+## Appendix
+
+In order to evaluate the overall lifetime for isoprene, say, in a cloud environment, we commence by referring concentrations to a common volume; namely, a unit volume of the gas phase. The cloud usually contains very little water (typically 10-6 m3 per m3 of air), so that we can neglect the volume occupied by the aqueous phase when giving concentrations in the gas phase. The dimensionless ratio between the volume of water and the volume of air is called _w_--the liquid water content of a cloud. \(w\) = 10-6 in the example above. \(w\) is sometimes given a unit such as m3(aq)/m3(gas) = cm3(aq)/cm3(gas) = L(aq)/L(gas). Thus, when an aqueous-phase concentration \(C\) in moles per
+
+\begin{table}
+\begin{tabular}{c c c c} Compound & \(\tau_{\text{cloud}}\)/h at \(w\) = 10−5 & \(\tau_{\text{cloud}}\)/h at \(w\) = 10−6 & \(\tau_{\text{cloud}}\)/h at \(w\) = 10−7 \\ Isoprene & 1.08 & 1.08 & 1.08 \\ Methacrolein & 14.2 & 27.6 & 30.5 \\ Methyl-vinyl-ketone & 29.0 & 33.6 & 34.2 \\ \end{tabular}
+\end{table}
+Table 5: Overall Lifetimes for Three Cloud Types Based on Individual Lifetimes with _x_(OH) = 0.1 ppbv. \(x\)(O3) = 30 ppbv at T = 298.15 K L(aq) is referred to the gas volume, it becomes Cw and its unit is now moles per L(gas). A concentration referred to the gas volume will be given a tilde, like \(\tilde{\mathcal{C}}=Cw\). Gas-phase concentrations are already referred to the gas volume, so their symbol will not be changed. \(C_{X}\) therefore automatically implies the concentration of a gaseous species \(X\), while \(\tilde{C}_{X}\) denotes the concentration of species \(X\) in the aqueous phase, but referred to the gas volume.
+
+Henry's law constants, once made dimensionless, pertain to all units. Assuming the common atmospheric way of giving the Henry constant \(H_{X}\) for \(X\) in the unit molar \(\times\) atm\({}^{-1}\), then \(H_{X}RT\) is the dimensionless Henry constant, _irrespective of the volume units assumed_. However, when reference to a common volume is assumed, then the pertinent Henry constant becomes \(\tilde{H}_{X}=H_{X}RTw\). This constant is actually the phase distribution ratio for compound \(X\) between the gas phase and the aqueous phase. It will therefore be a crucial quantity in what follows.
+
+Now consider the total concentration (denoted by the Greek letter \(\zeta\)) of \(X\). This is given as
+
+\[\zeta_{X}=\,C_{X}+\,\tilde{C}_{X} \tag{1}\]
+
+When the individual kinetic equations for \(C_{X}\) and \(\tilde{C}_{X}\) are added together (i.e., when formulated in terms of the total concentrations \(\zeta_{X}\)), then the 1. order uptake rate expressions may be shown to cancel out, so that only the chemical rates remain (this may be seen from Eqs. (2.11) and (2.12) in [1], except that the pertinent two signs have been erroneously interconverted in these equations). Thus, assuming that \(X\) reacts with O\({}_{3}\) as well as with OH, the kinetic equation becomes
+
+\[\frac{d\zeta_{X}}{dt}=k_{X;\mathrm{OH}}C_{X}\mathrm{C}_{\mathrm{OH }}+\,k_{X;\mathrm{O}_{3}}C_{X}C_{\mathrm{O}_{3}}\\ +\,\tilde{k}_{X;\mathrm{OH}}\tilde{C}_{X}\tilde{C}_{\mathrm{OH}}+ \,\tilde{k}_{X;\mathrm{O}_{3}}\tilde{C}_{X}\tilde{C}_{\mathrm{O}_{3}} \tag{2}\]
+
+Now observe that \(k_{X;\mathrm{OH}}C_{\mathrm{OH}}\) is nothing but the reciprocal time constant \(\tau_{\mathrm{OH};\mathrm{O}_{3}}\),\(\mathrm{x}\) etc., so that this equation may be rewritten as
+
+\[\frac{d\zeta_{X}}{dt}=(\tau\tilde{\mathrm{o}}_{\mathrm{OH};\mathrm{ x}}^{\mathrm{i}}+\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{i};\mathrm{x}}^{ \mathrm{-1}})\mathrm{C}_{X}\\ +\,(\tau\tilde{\mathrm{o}}_{\mathrm{OH};\mathrm{x}}^{\mathrm{-1}} )\mathrm{+\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{i};\mathrm{x}}^{\mathrm{-1}} )\tilde{C}_{X}} \tag{3}\]
+
+We now replace the two concentrations of \(X\) by its total concentration, assuming the validity of Henry's law during the reaction --as argued for in the text. It is easily seen that this is done as follows:
+
+\[C_{X}=\frac{1}{1+\,\tilde{H}_{X}}\zeta_{X},\qquad\tilde{C}_{X}= \frac{\tilde{H}_{X}}{1+\,\tilde{H}_{X}}\zeta_{X},\\ \tilde{H}_{X}=H_{X}RTw=\frac{\tilde{C}_{X}}{C_{X}} \tag{4}\]
+
+The two fractions 1/(1 + \(\tilde{H}_{X}\)) and \(\tilde{H}_{X}\)/(1 + \(\tilde{H}_{X}\)) are the fractions of the total that are in the gas phase and in the aqueous phase, respectively. We then get
+
+\[\frac{d\zeta_{X}}{dt}=\Bigg{\{}(\tau\tilde{\mathrm{o}}_{\mathrm{OH };\mathrm{g};X}^{\mathrm{-1}}+\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{g};X }^{\mathrm{-1}})\,\frac{1}{1+\,\tilde{H}_{X}}\\ +\,(\tau\tilde{\mathrm{o}}_{\mathrm{OH};\mathrm{g};X}^{\mathrm{-1}} +\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{i};\mathrm{x}}^{\mathrm{-1}})\, \frac{\tilde{H}_{X}}{1+\,\tilde{H}_{X}}\Bigg{\}}\zeta_{X} \tag{5}\]
+
+The term in braces is an effective first-order rate constant, and its reciprocal is the time constant we seek. Hence,
+
+\[\tau_{\mathrm{cloud};X}=\Bigg{\{}(\tau\tilde{\mathrm{o}}_{\mathrm{ OH};\mathrm{g};X}+\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{g};X}^{\mathrm{-1}})\, \frac{1}{1+\,\tilde{H}_{X}}\\ +\,(\tau\tilde{\mathrm{o}}_{\mathrm{OH};\mathrm{g};X}^{\mathrm{-1}} +\,\tau\tilde{\mathrm{o}}_{\mathrm{3};\mathrm{i};\mathrm{q};X}^{\mathrm{-1}})\, \frac{\tilde{H}_{X}}{1+\,\tilde{H}_{X}}\Bigg{\}}^{-1} \tag{6}\]
+
+## Bibliography
+
+* [1] Lelieveld, J.; Crutzen, P. J. J Atmos Chem 1991, 12, 229-267.
+* [2] Guenther, A. Reactive Hydrocarbons in the Atmosphere; Hewitt, C. N., Ed.; Academic: San Diego, 1999; pp 97-118.
+* [3] Derwent, R. G. Reactive Hydrocarbons in the Atmosphere; Hewitt, C. N., Ed.; Academic: San Diego, 1999; pp 268-289.
+* [4]
+
+\begin{table}
+\begin{tabular}{l c c c} Compound & \(\tau_{\mathrm{cloud}}\)/h at \(w=10^{-5}\) & \(\tau_{\mathrm{cloud}}\)/h at \(w=10^{-6}\) & \(\tau_{\mathrm{cloud}}\)/h at \(w=10^{-7}\) \\ \hline Isoprene & 28.6 & 28.6 & 28.6 \\ Methacrolein & 365 & 368 & 369 \\ Methyl-vinyl-ketone & 78.4 & 79.6 & 79.7 \\ \hline \end{tabular}
+\end{table}
+Table 6: Overall Lifetimes Based on Individual Lifetimes for a Situation without OH and with \(x\)(O\({}_{3}\)) = 30 ppbv at T = 298.15 K * [4] McAucliffe, C. J Phys Chem 1966, 70, 1267\(-\)1275.
+* [5] Mackay, D.; Shiu, W. Y. J Phys Chem Ref Data 1981, 10, 1175\(-\)1199.
+* [6] Allen, J. M.; Balcavage, W. X.; Ramachandran, B. R.; Shrout, A. L. Environmental Toxicology and Chemistry 1998, 17, 1216\(-\)1221.
+* [7] Marcovic, V.; Sehested, K. Proc Tyhany Sympos Radiat Chem 1972, 3, 1243\(-\)1243.
+* [8] Bothe, E.; Schulte-Frohlinde, D. Z.; Naturforsch, B. Anorg Chem Org Chem 1980, 35, 1035\(-\)1039.
+* [9] Paulson, S. E.; Seinfeld, J. H. J Geophys Res 1992, 97, 20703\(-\)20715.
+* [10] Paulson, S. E.; Flagan, R. C.; Seinfeld, J. H. Int J Chem Kinet 1992, 24, 79\(-\)101.
+* [11] Paulson, S. E.; Flagan, R. C.; Seinfeld, J. H. Int J Chem Kinet 1992, 24, 103\(-\)125.
+* [12] Grosjean, D.; Williams, E. L.; Grosjean, E. Environ Sci Technol 1993, 27, 830\(-\)840.
+* [13] Hart, E.; Sehested, K.; Holcman, J. Analy Chem. 1983, 55, 46\(-\)49.
+* [14] Sander, W. A. Chem Int Ed Engl 1990, 29, 344\(-\)354.
+* [15] Schwarz, S. E. Chemistry of Multiphase Atmospheric Systems; Jaeschke, W., Ed.; Springer: Berlin, 1986; pp 415\(-\)471.
+* [16] Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics; Wiley and Sons: New York, 1998.
+* [17] Kozak-Channing, L. F.; Heltz, G. R. Environ Sci Technol 1983, 17, 145\(-\)149.

manual_annotation/200pdfs_in_mmd/10.1002_aic.690470116.mmd

@@ -0,0 +1,204 @@
+# Decolorization of Dyes Using Ozone in a Gas-Induced Reactor
+
+Yung-Chien Hsu, Jia-Tsween Chen, and Hsiang-Cheng Yang
+
+Dept. of Chemical Engineering, National Taiwan University of Science and Technology, Taipei 106, Taiwan
+
+Jyh-Herng Chen
+
+Dept. of Material and Mineral Resources Engineering, National Taipei University of Technology, Taipei 106, Taiwan
+
+###### Abstract
+
+Treatability of aqueous dye solutions using a new gas-induced ozonator was investigated, as well as feasibility of decolorizing dyes using ozone, the ozone utilization efficiency, and the chemical oxygen demand (COD) under various pH values and initial dye concentrations. Experimental results indicate that all of the dyes used can be decolorized within 15 min. For most of the dyes used, ozone decolorizing was more rapid in acidic media. The ozone utilization efficiency can be raised to 90% or higher with proper agitation speeds. The ozone utilization efficiency and COD removal rate increased with an increase in pH value; however, the COD removal remained nearly constant for various categories of dye regardless of the initial dye concentration. The kinetics of decolorization were also examined.
+
+## Introduction
+
+Dye manufacturing, textile dyeing and finishing are very common industries in Taiwan, which generate a large amount of wastewater. The composition of dye industry wastewater discharge containing dyes, mordant, sizing agents and dyeing aids are deep in color and highly polluted. Previously, a single chemical coagulation process, in combination with an activated sludge process or activated carbon, was employed to treat this type of wastewater. However, adsorption by activated carbon is an expensive and complicated process, while bio-refractory is limited owing to the low biodegradability of dyes (Grau, 1991). Recently, the Environment Protection Agency (EPA) has placed strict restrictions on the color of wastewater in Taiwan. Ozone processing is a promising alternative.
+
+Ozone is an extremely strong oxidant. It can undergo self-decomposition in an aqueous solution to form hydroxyl free radicals that have a stronger oxidizing capability (Staehelin and Hogine, 1985; Sotelo et al., 1987). Hence, ozone has many uses. In some commercialized industrial applications, it has been widely accepted as an effective disinfectant and a chemical oxidant (Gould and Weber, 1976; Beltran et al., 1993; Perkins et al., 1995). Studies have reported on the reaction between ozone and dyes (Saunders et al., 1983; Benitez et al., 1993; Gahr et al., 1994), and indicated that ozone possesses an excellent decolorizing capability. However, industrial applications have been limited by the low utilization efficiency of ozone, as well as its high production cost. The low utilization efficiency of ozone can be improved by increasing the retention time of ozone gas in an aqueous solution. In conventional processes, gas utilization efficiency can be improved by linking reactors in a series or by using a compressor to recirculate the unreacted gas back into the processing liquid. However, both of these methods are complex, and require additional accessory equipment, thus increasing the operational cost. Moreover, to improve the utilization of reactive gases, many contactors have been considered, such as the sparged loop reactor, the surface aerator, and the gas-induced reactor. Saravanan and Joshi (1994) indicated that a gas-induced reactor has all of the advantages of other reactors for gas-liquid reactions. A conventional gas-induced reactor consists mainly of a hollow shaft and impeller (Zlokarnik and Judat, 1967; Joshi and Sharma, 1997), or a standpipe with several different types of impellers (Zundelevich, 1979; Mundale and Joshi, 1995). However, their configurations are complicated and high power consumption is required for agitation. Hsu and Huang (1996, 1997) developed a new gas-induced reactor for the same purpose, in which two in-series45o pitched-blade turbines were enclosed in a draft tube. Two main characteristics of this device include gas induction and bubble recirculation around the draft tube in the liquid. In our previous work, only a heterogeneous ozone reaction with C.I. Reactive Blue 19 dye at a constant pH value and initial dye concentration was studied. At a stable gas induction, this new reactor demonstrated an ozone utilization efficiency of more than 95%. An improvement in ozone utilization efficiency without any accessory equipment is considered to be a major advantage in cost reduction. In the past, the ozone utilization efficiency on the decomposition of various dyes has rarely been investigated. Notably, each dye has its own molecular structure and differs somewhat in the degree of reaction with ozone. Therefore, this work uses eight different categories of dyes to investigate the performance of this new reactor and its probable application in wastewater treatment of textile dyeing and finishing. The capability of ozone to de-colorize and mineralize dyes was also studied. Experimental results show that, for each dye used at various pH values and initial dye concentrations, the ozone utilization efficiency remains higher than 90%. Therefore, the reaction time is effectively shortened.
+
+\begin{table}
+\begin{tabular}{c c c c} \hline C.I. & **Mordant Black 11** & **C. I. 42090** & **C.I. Acid Blue 9** & **C. I. 14645** \\ \hline \(\lambda_{\max}\) = 592 nm & & & \\ \(\lambda_{\max}\) = 587 nm & & & \\ \(\lambda_{\max}\) = 589 nm & & & \\ \(\lambda_{\max}\) = 589 nm & & & \\ \(\lambda_{\max}\) = 589 nm & & & \\ \(\lambda_{\max}\) = 589 nm & & & \\ \(\lambda_{\max}\) = 589 nm & & & \\ \(\lambda_{\max}\) = 584 nm & & & \\ \(\lambda_{\max}\) = 548 nm & & & \\ \(\lambda_{\max}\) = 548 nm & & & \\ \(\lambda_{\max}\) = 524 nm & & & \\ \(\lambda_{\max}\) = 524 nm & & & \\ \(\lambda_{\max}\) = 493 nm & & & \\ \(\lambda_{\max}\) = 493 nm & & & \\ \(\lambda_{\max}\) = 524 nm & & & \\ \(\lambda_{\max}\) = 4
+
+## Materials and Methods
+
+### Materials and procedure
+
+C.I. Mordant Black 11 (MB-11) and C.I. Acid Blue 9 (AB-9) were provided by Chint Chemicals Company (Taiwan). C.I. Reactive Blue 19 (RB-19) and C.I. Reactive Orange 16 (RB-16) were provided by Sumitomo Company (Japan). C.I. Acid Violet 42 (AV-42), C.I. Direct Blue 71 (DB-71), C.I. Basic Blue (BB-3), and C.I. Basic Red 13 (BR-13) were provided by Bayer Company (Germany). The main chemical structures and maximum absorbencies in an aqueous solution were listed in Table 1.
+
+The synthetic wastewater was prepared by dissolving the required amounts (100-500 mg/L +- 10 mg/L) of a single dye in deionized water. The pH values were adjusted to 2.0, 5.0, 7.0, 9.0, or 10.0, respectively, by using 0.1 M NaOH or H2SO4 solution (Hewes and Davison, 1971).
+
+The experiment apparatus is shown in Figure 1 (Hsu and Huang, 1996). At the beginning of each run, 7 L of dye solution at the desired concentration and pH was transferred into the reactor. The solution was maintained at 25degC. The flow rate and the concentration of the ozone gas were 210 NL/h and 20 mg/NL, respectively. The concentration of ozone was averaged and used as the feed concentration for ozone gas. The agitation speed was adjusted to 1,800 rpm and the ozone gas was simultaneously introduced into the reactor. In addition the defoamer was added in dribs to eliminate the interference for gas induction in some cases where foaming occurred. The defoamer was a commercial product (KM-73, Shinetsu, Taiwan). Since only a trace amount of defoamer was added, it was expected that it would not interfere with reactions.
+
+While the ozone concentration in the effluent gas stream was measured automatically every second by the ozone detector during the reaction, samplings of dye solution were conducted at proper intervals. The color value of each sample was measured employing the American Dye Manufacturers Institute (ADMI) method. The chemical oxygen demand (COD) values were simultaneously determined.
+
+### Methods for determination
+
+The ADMI color value of the dye solution was determined employing a UV/Vis Spectrophotometer (Hach DR-4000 colorimeter). During ozonation, the spectrophotometer scans within the visible light range (400-700 nm) and quantified the color of the dye solution.
+
+A Hach COD digester was employed to digest oxidizable organic compounds at 150degC. A UV/Vis Spectrophotometer (Hach DR-4000) was used to determine the COD value of the digested samples at 620 nm (by a built-in digestion method (Jirka and Carter, 1975), U.S. EPA approved).
+
+Equation 1 defines the ozone utilization efficiency, as suggested by Hsu and Huang (1996)
+
+\[U_{{\text{O}}_{\text{y}}} = \frac{{W_{{\text{O}}_{3},i} - W_{{\text{O}}_{3},o}}}{{W_{{\text{O}}_{3},i}}} \times 100\%\]
+
+\(U_{{\text{O}}_{\text{y}}}\) values are based on 90% decolorization of the dye solution in which a light color can be observed. \(W_{{\text{O}}_{3},i}\) and \(W_{{\text{O}}_{3},o}\) are the total input and output amount of ozone measured in the gas phase.
+
+## Results and Discussion
+
+Table 2 summarizes the results of the decolorization of the eight dyes. Among these results, Mordant Black 11 is selected to illustrate the entire scope of these experiments. In species cases, when the BB-3 and BR-13 dye solutions were adjusted to pH 11 and DB-71 was adjusted to pH 2, the appearance of the dye solutions changed, while in the spectrogram, qualitative analysis by absorption became difficult. Consequently, no investigations were conducted under these conditions.
+
+### Decolorization time of dyes
+
+#### (1) Effect of pH Values
+
+Figure 2 shows the decolorization of MB-11 with the same initial dye concentrations at various pH values. This figure also indicates that during ozonation of the ADMI color value of MB-11 disappears gradually. Furthermore, it is evident that the initial decolorization rates within approximately 2 min are almost the same at various pH values. This is because the dye molecules are the primary source of color in the initial stage. Regardless of the pH value, the dye molecules are destroyed easily by ozone and/or hydroxyl free radicals. As the decolorization process continues, after 2 min, intermediate products are generated to such an extent that the difference appears at various pH values. After approximately 4 min, the decolorization rates are much slower
+
+Figure 1: Experimental apparatus.
+
+than the initial rate for any pH value. This is because most dye molecules are destroyed and the persistent intermediate products become the primary substances. Figure 2 also indicates that a decrease in the pH value increases the dye decolorization rate. Evidently, the resonance effect within the chromophore of the dye structure greatly enhances the electron density, which in turn is affected by the ozone molecule electrophilic attack (Sotelo et al., 1989). At low pH values, a greater extent of decolorization is favored by a more direct ozone attack, since decolorization is associated with the destruction of chromophore. The increase in pH value may accelerate ozone decomposition, forming more hydroxyl free radicals (*OH) (Hogne and Bader, 1976, 1983; Gurol and Singer, 1982). This causes a decrease in available ozone molecules and a less selective ozone attack on chromophore. Consequently, a longer reaction time for decolorization is required. Similar results were also obtained for BB-3, BR-13, AV-42, AB-9, and RB-19. Figure 3 presents the decolorization behavior of RO-16. Unlike MB-11, the decolorization curves of RO-16 at various pH values are smooth, with the decolorization rate increasing as the pH increases. This reaction of RO-16 is likely due to both the resonance effects of the RO-16 dye molecules and the electrophilic attack by hydroxyl free radicals. In alkaline media (such as pH = 10) the naphthalene ring of RO-16 exhibits higher electron densities. Resonance from the dissociated hydroxyl group (that is, -\(\overset{-}{\text{O}}^{-}\)) and the -\(\overset{-}{\text{NHCOCH}}_{3}\) group, which donates electrons acting as electron releasing groups, enhances the electron density in the ring and renders the electron-rich naphthalene ring to be more susceptible to attack by 'OH radicals.
+
+_(2) Effect of Initial Dye Concentration._ Figure 4 shows the relationship between (ADMI/ADMI_o_) and time. It is evident
+
+\begin{table}
+\begin{tabular}{c c c c c c c c c c c c c c} \hline  & \multicolumn{4}{c}{C.I. Mordant Black 11} & \multicolumn{4}{c}{C.I. Direct Blue 71} & \multicolumn{4}{c}{C.I. Basic Blue 3} \\ \hline  & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) \\ pH & (mg/L) & (s) & (\%) & (s−1) & pH & (mg/L) & (s) & (\%) & (s−1) & pH & (mg/L) & (s) & (\%) & (s−1) \\ \hline
+2.0 & 200 \(\pm 10\) & 396 & 89.79 & 9.68/1.55 & — & — & — & — & — & 2.0 & 200 \(\pm 10\) & 442 & 97.16 & 3.85/6.79 \\
+5.0 & 200 \(\pm 10\) & 426 & 92.99 & — & 5.0 & 200 \(\pm 10\) & 522 & 89.19 & 5.06/- & 5.0 & 200 \(\pm 10\) & 479 & 99.10 & 3.47/5.51 \\
+7.0 & 200 \(\pm 10\) & 514 & 93.63 & 9.12/1.13 & 7.0 & 200 \(\pm 10\) & 476 & 95.25 & 5.39/- & 7.0 & 200 \(\pm 10\) & 524 & 99.17 & 3.28/4.91 \\
+10.0 & 200 \(\pm 10\) & 563 & 96.64 & 8.55/1.39 & 10.0 & 200 \(\pm 10\) & 427 & 98.98 & 5.69/- & 9.0 & 200 \(\pm 10\) & 523 & 99.07 & 2.74/5.02 \\
+7.0 & 300 \(\pm 10\) & 772 & 94.55 & 5.78/0.85 & 7.0 & 300 \(\pm 10\) & 667 & 95.93 & 3.84/- & 7.0 & 100 \(\pm 10\) & 269 & 98.33 & 6.38/11.29 \\
+7.0 & 400 \(\pm 10\) & 1037 & 95.06 & 4.16/0.76 & 7.0 & 400 \(\pm 10\) & 931 & 97.27 & 2.75/- & 7.0 & 300 \(\pm 10\) & 755 & 98.99 & 2.23/3.81 \\
+7.0 & 500 \(\pm 10\) & 1290 & 95.79 & 3.25/0.56 & 7.0 & 500 \(\pm 10\) & 1045 & 98.50 & 2.54/- & 7.0 & 400 \(\pm 10\) & 998 & 99.19 & 1.74/3.08 \\ \hline  & \multicolumn{4}{c}{C.I. Acid Blue 9} & \multicolumn{4}{c}{C.I. Reactive Blue 19} & \multicolumn{4}{c}{C.I. Acid Violet 42} \\ \hline  & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & Conc. & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) \\ pH & (mg/L) & (s) & (\%) & (s−1) & pH & (mg/L) & (s) & (\%) & (s−1) & pH & (mg/L) & (s) & (\%) & (s−1) \\ \hline
+2.0 & 200 \(\pm 10\) & 417 & 93.63 & 5.07/- & 2.0 & 200 \(\pm 10\) & 383 & 86.73 & 106.63/1.78 & 2.0 & 300 \(\pm 10\) & 182 & 95.98 & 13.21/- \\
+5.0 & 200 \(\pm 10\) & 566 & 95.63 & 3.79/- & 5.0 & 200 \(\pm 10\) & 390 & 92.12 & 9.73/1.59 & 5.0 & 300 \(\pm 10\) & 192 & 99.59 & 11.46/- \\
+7.0 & 200 \(\pm 10\) & 602 & 94.65 & 3.63/- & 7.0 & 200 \(\pm 10\) & 487 & 94.51 & 9.43/1.39 & 7.0 & 300 \(\pm 10\) & 206 & 99.25 & 10.96/- \\
+10.0 & 200 \(\pm 10\) & 631 & 96.47 & 3.44/- & 10.0 & 200 \(\pm 10\) & 483 & 97.52 & 9.37/1.11 & 10.0 & 300 \(\pm 10\) & 216 & 99.66 & 11.02/- \\
+7.0 & 300 \(\pm 10\) & 892 & 96.24 & 2.48/- & 7.0 & 300 \(\pm 10\) & 740 & 94.52 & 6.24/0.83 & 7.0 & 200 \(\pm 10\) & 143 & 99.25 & 16.01/- \\
+7.0 & 400 \(\pm 10\) & 1114 & 97.43 & 1.94/- & 7.0 & 400 \(\pm 10\) & 990 & 94.46 & 4.40/0.67 & 7.0 & 400 \(\pm 10\) & 263 & 99.14 & 8.32/- \\
+7.0 & 500 \(\pm 10\) & 1337 & 98.25 & 1.61/- & 7.0 & 500 \(\pm 10\) & 1142 & 94.34 & 3.69/5.7 & 7.0 & 500 \(\pm 10\) & 349 & 99.50 & 6.76/- \\ \hline  & \multicolumn{4}{c}{C.I. Basic Red 13} & \multicolumn{4}{c}{C.I. Reactive Orange 16} \\ \cline{2-11}  & \multicolumn{2}{c}{Conc.} & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & \multicolumn{2}{c}{Conc.} & \(t_{(X = 0.9)}\) & \(U_{Q_{1}}\) & (\(k_{1}^{\prime}\)/\(k_{2}^{\prime}\))\(\times 10^{3}\) & \multicolumn{2}{c}{} \\ pH & (mg/L) & (s) & (\%) & (s−1) & pH & (mg/L) & (s) & (\%) & (s−1) & \\ \hline  & 2.0 & 200 \(\pm 10\) & 270 & 97.20 & 6.25/10.35 & 2.0 & 200 \(\pm 10\) & 516 & 83.50 & 3.85/- \\  & 5.0 & 200 \(\pm 10\) & 285 & 97.20 & 5.76/10.88 & 5.0 & 200 \(\pm 10\) & 545 & 88.92 & 4.06/- \\
+7.0 & 200 \(\pm 10\) & 283 & 97.72 & 5.97/10.88 & 7.0 & 200 \(\pm 10\) & 470 & 95.86 & 4.65/- \\  & 9.0 & 200 \(\pm 10\) & 296 & 98.34 & 6.27/9.22 & 10.0 & 200 \(\pm 10\) & 365 & 99.72 & 6.40/- \\  & 7.0 & 300 \(\pmthat the decolorization time required is dependent upon different initial concentration of MB-11. Results also indicate that, at the concentration of \(C_{\text{do}} = 200 \pm 10\) mg/L (ADMIo = 3425 value), 90% decolorization could be reached within 10 min, whereas approximately 22 min are required for higher concentrations such as \(C_{\text{do}} = 500 \pm 10\) mg/L (ADMIo = 8,500 value). Furthermore, the decolorization times required at higher dye concentrations are greater than those at lower dye concentrations. This is due to the constant input ozone concentration, which is maintained in this study; therefore, more time is required at higher dye concentrations. All required decolorization times were extremely short. Consequently, the decolorization by ozone is remarkable and feasible for the treatment of different dye concentrations in manufacturing wastewater of textile dyeing and finishing plants. Similar results are obtained for all other dyes used.
+
+### Kinetics analysis for decolorization
+
+Saunders et al. (1983) found that the reaction between the dye and ozone is an instantaneous reaction and, in the liquid film, ozone is consumed completely in a semi-batch process. Therefore, they proposed that the kinetics model of dye with ozone was a pseudo first-order reaction. In this work, the decolorization of the eight dyes is also performed by semi-batch process. The pseudo first-order expression is assumed to describe the heterogeneous ozonation of experimental dyes with an associated instantaneous chemical reaction. Furthermore, the input ozone concentration is constant and no residual ozone remains in the bulk liquid as 90% ADMI color value is removed. In addition, ozone utilization efficiency is very high in this study (Table 2). Hence, it is possible to consider that there is an almost constant ozone influence, which can be incorporated into the overall reaction rate constant \(k^{\prime}\). However, some of the dyes have two different overall rate constants (Figure 5) and the respective \(k^{\prime}_{1}\) and \(k^{\prime}_{2}\) are calculated by
+
+\[\ln\frac{\text{ADMI}_{\text{(section 1)}}}{\text{ADMI}_{0,1}}=k^{\prime}_{1}t \tag{2}\]
+
+Figure 4: Normalized decolorization of MB-11 at different initial concentrations of dye with the same pH values.
+
+Figure 5: Variation of In(ADMIo/ADMI) of MB-11 with time at different initial dye concentration during ozone processing.
+
+Figure 3: Normalized decolorization of RO-16 at different pH values.
+
+for the first section, while
+
+\[\ln\frac{\text{ADMI}_{(\text{section}\,2)}}{\text{ADMI}_{0,2}} = k_{2}^{\prime}t\]
+
+for the second section. \(\text{ADMI}_{0,1}\) represents the initial ADMI color value at \(t\) = 0, and the \(\text{ADMI}_{0,2}\) represents the ADMI color value at the standing point of the second section. According to Figure 5, the variation of \(\ln(\text{ADMI}_{0}/\text{ADMI})\) of MB-11 with time at different initial concentrations are separated into two sections before 90% decolorization. It can be considered that ozone molecules and/or hydroxyl free radicals mainly attack the dye molecules in the first section, and attack the intermediate products in the second section. The decolorization rate in the first section is faster than that in the second section, which is because the MB-11 molecules are destroyed and form the yellow persistent decomposition intermediate products (Perkins et al., 1995). Similar results were found in RB-19 (figure not shown here). In contrast, the decolorization rates in the second section for DB-71 and AB-9 are faster than those in the first section. This may be attributed to ADMI color value of the intermediate products that are more easily decreased by ozonation. Figure 6 shows the \(\ln(\text{ADMI}_{0}/\text{ADMI})\) vs. time plot for RO-16 (others are not shown here). Due to the slight influence of intermediate products, RO-16 and other dyes (namely, DB-71, AB-9, and AV-42) are depicted by one straight line to represent the decolorization kinetics before 90% decolorization. Table 2 summarizes the overall decolorization rate constants and ozone utilization efficiency for each dye.
+
+### COD elimination
+
+#### (1) Effect of pH Value
+
+Figure 7 shows the variation of normalized COD removal of MB-11 with time at various pH values. The normalized COD removal increases with an increase in the pH value. It is considered that at higher pH values, higher fractions of ozone are decomposed to form 'OH radicals (Hoigne and Bader, 1976, 1983; Gurol and Singer, 1982). Since 'OH radicals have a stronger oxidation potential than ozone molecules, the 'OH radicals attack dye molecules and intermediates more vigorously than would be expected for a molecular ozone. As a result, the higher the pH values, the faster the mineralization will be. Singer and Gurol (1983) also obtained similar results as they treated phenol using ozone in a semibatch reactor.
+
+#### (2) Effect of Initial Concentration of Dyes
+
+Figure 8 shows the COD removal of MB-11 plotted against ozonation time to compare with literature. It is evident that during ozonation at a constant pH value, the COD removal rate appears to remain almost constant (\(k_{d} = 5.22 \pm 0.02\) mg/L * min). Tzitzi et al. (1994) reported similar results. However, in Tzitzi et al. the COD removal rate is about 1.5 mg/L * min, which is much smaller than 5.22 +- 0.02 mg/L * min, as proved herein. The removal consistency of a given amount of COD is independent of the initial dye concentration. The COD/O3 ratio is summarized in Table 3.
+
+### Ozone utilization efficiency
+
+Figure 9a shows the ozone utilization efficiency for the decolorization of MB-11 at various pH values with the same initial dye concentration. Ozone utilization efficiencies are more than 90% and increase with increasing pH value. At
+
+Figure 6: Variation of \(\ln(\text{ADMI}_{0}/\text{ADMI})\) values of RO-16 with time at different pH values during ozone processing.
+
+\(T = 298\) K, \(N = 1,800\) rpm, \(Q_{g} = 210\) NL/_h_, \(\text{C}_{0,j} = 20.0 \pm 0.1\) mg/NL, \(\text{C}_{obs} = 200 \pm 10\) mg/L.
+
+Figure 7: Effect of pH value on normalized COD removal for MB-11 during ozone processing.
+
+higher pH values, more 'OH free radicals with higher oxidation-reduction potential tend to be produced from ozone self-decomposition (Hoigne and Bader, 1976, 1983; Gurol and Vatistas, 1987). For that reason, more ozone molecules can dissolve into the aqueous solution and a higher OUE can be obtained. As showing in Figure 9b, at different initial dye concentrations with the same pH value, OUE still remains high (\(>95\%\)) and is almost independent of the initial dye concentration. Table 2 illustrates the overall results. With the exception of RO-16, DB-71, and RB-19 dye in acidic medium, an OUE of greater than 90% can be obtained. This unique characteristic is beneficial to the treatment of industrial wastewater containing various dyes.
+
+## Conclusions
+
+This work has demonstrated that the ozone utilization efficiency increased with an increase in pH value. When a newly developed gas-induced reactor was employed, ozone utiliza
+
+\begin{table}
+\begin{tabular}{c c c c c c c c} \hline  & C.I. Acid Blue 9 & \multicolumn{3}{c}{C.I. Acid Violet 42} & \multicolumn{3}{c}{C.I. Direct Blue 71} \\ \cline{2-7}  & Conc. & \multicolumn{3}{c}{} & Conc. & \multicolumn{3}{c}{} & Conc. \\ pH & (mg/L) & COD/O3 & pH & (mg/L) & COD/O3 & pH & (mg/L) & COD/O3 \\ \hline
+2.0 & \(200\pm 10\) & 0.527 & 2.0 & \(300\pm 10\) & 0.535 & — & — & — \\
+5.0 & \(200\pm 10\) & 0.515 & 5.0 & \(300\pm 10\) & 0.696 & 5.0 & \(200\pm 10\) & 0.462 \\
+7.0 & \(200\pm 10\) & 0.518 & 7.0 & \(300\pm 10\) & 0.835 & 7.0 & \(200\pm 10\) & 0.545 \\
+10.0 & \(200\pm 10\) & 0.641 & 10.0 & \(300\pm 10\) & 0.989 & 10.0 & \(200\pm 10\) & 0.569 \\
+7.0 & \(300\pm 10\) & 0.685 & 7.0 & \(200\pm 10\) & 0.678 & 7.0 & \(300\pm 10\) & 0.546 \\
+7.0 & \(400\pm 10\) & 0.652 & 7.0 & \(400\pm 10\) & 0.502 & 7.0 & \(400\pm 10\) & 0.626 \\
+7.0 & \(500\pm 10\) & 0.480 & 7.0 & \(500\pm 10\) & 0.525 & 7.0 & \(500\pm 10\) & 0.739 \\ \hline  & C.I. Mordant Black 11 & \multicolumn{3}{c}{C.I. Reactive Blue 19} & \multicolumn{3}{c}{C.I. Direct Blue 71} \\ \cline{2-7}  & Conc. & \multicolumn{3}{c}{} & Conc. & \multicolumn{3}{c}{} & Conc. \\ pH & (mg/L) & COD/O3 & pH & (mg/L) & COD/O3 & pH & (mg/L) & COD/O3 \\ \hline
+2.0 & \(200\pm 10\) & 0.460 & 2.0 & \(200\pm 10\) & 0.387 & 2.0 & \(200\pm 10\) & 0.535 \\
+5.0 & \(200\pm 10\) & 0.508 & 5.0 & \(200\pm 10\) & 0.464 & 5.0 & \(200\pm 10\) & 0.531 \\
+7.0 & \(200\pm 10\) & 0.567 & 7.0 & \(200\pm 10\) & 0.565 & 7.0 & \(200\pm 10\) & 0.623 \\
+10.0 & \(200\pm 10\) & 0.592 & 10.0 & \(200\pm 10\) & 0.599 & 10.0 & \(200\pm 10\) & 0.685 \\
+7.0 & \(300\pm 10\) & 0.538 & 7.0 & \(300\pm 10\) & 0.560 & 10.0 & \(300\pm 10\) & 0.690 \\
+7.0 & \(400\pm 10\) & 0.537 & 7.0 & \(400\pm 10\) & 0.565 & 10.0 & \(400\pm 10\) & 0.620 \\
+7.0 & \(500\pm 10\) & 0.555 & 7.0 & \(500\pm 10\) & 0.656 & 10.0 & \(500\pm 10\) & 0.601 \\ \hline \end{tabular}
+\end{table}
+Table 3: COD/O3 Ratios of Eight Species of Dyes
+
+Figure 8: Effect of initial concentration of dyes on COD removal for MB-11 during ozone processing.
+
+Figure 9: Effect of initial concentration and pH of MB-11 dye on ozone utilization efficiency during ozone processing.
+
+tion efficiency obtained can be more than 90% regardless of pH values and initial dye concentrations. Hence, ozone treatment of various dye solutions can be very rapid and efficient. Generally, only 15 to 22 min is required for a 90% dye solution decolorization at a concentration of \(500\pm 10\) mg/L. The overall decolorization rate constant depends on the various dyes and ozone utilization efficiency. In particular, MB-11, RB-19, BB-3, and BR-13 have two overall decolorization rate constants and the others used have only one. The COD elimination rates are found to be nearly constant regardless of the initial dye concentration; however, it increases with increasing pH value.
+
+## Acknowledgment
+
+The authors thank the National Science Council of R.O.C. for the financial support (Project No. NSC85-2211-E-011-027).
+
+## References
+
+- color value of dye solution during ozone processing
+* initial colour value of dye solution
+* initial colour value at \(t=0\)
+* \(\Delta\)DMI color value at the standing point of section 2
+* initial dye concentration, mg/L
+* a ozone input concentration at 273 K, 101.3 kPa, mg/NL
+* COD elimination rate constant, mg/L\({}^{\prime}\) min
+* \(k^{\prime}\) = overall decolorization rate constant, s\({}^{- 1}\)
+* \(k^{\prime}_{1}\) = overall decolorization rate constant of the first section, s\({}^{- 1}\)
+* \(k^{\prime}_{2}\) = overall decolorization rate constant of the second section, s\({}^{- 1}\)
+* \(N\) = impeller speed, rpm
+* \(Q_{g}\) = ozone input flow rate at 273 K, 101.3 kPa, NL/h
+* \(t\) = reaction time, s
+* \(T\) = temperature, K
+* \(U_{0}\) = ozone utilization efficiency, %
+* \(W_{0,\phi}\) = total weight of ozone fed to the reactor, mg
+* \(W_{0,\phi}\) = total weight of ozone leaving the reactor, mg
+* \(X\) = fractions of conversion
+* \(\lambda_{\max}\) = maximum absorption wavelength of dye in aqueous solution, nm
+
+## References
+
+- Beltran _et al._ 1993 Beltran, F. J., J. M. Encinar, and J. F. Garcia-Araya, "Oxidation by Ozone and Chlorine Dioxide of Two Distillery Wastewater Contaminants: Gallic Acid and Epicatechin," _Wat. Res._, **27**, 1023 (1993).
+* Benitez _et al._ 1993 Benitez, F. J. Beltran-Heredia, T. Gonzalez, and A. Pascual, "Ozone Treatment of Methylene Blue on Aqueous Solution," _Chem. Res. Commun._, **119**, 151 (1993).
+* Gahr _et al._ 1994 Gahr, F., F. Hermanutz, and W. Oppermann, "Ozonation--An Important Technique to Comply with New German Laws for Textile Wastewater Treatment," _Wat. Sci. Tech._, **30**, 255 (1994).
+* Gould and Weber 1976 Gould, J. P., and W. J. Weber, "Oxidation of Phenols by Ozone," _J. Wat. Pollut. Control Fed._, **48**, 47 (1976).
+* Grau 1991 Grau, P., "Textile Industry Wastewaters Treatment," _Wat. Sci. Tech._, **24**, 97 (1991).
+* Gurol and Vatistas 1987 Gurol, M. D., and R. Vatistas, "Oxidation of Phenolic Compounds by Ozone and Ozone + UV Radiation: A Comparative Study," _Wat. Res._, **21**, 895 (1987).
+* Gurol and Singer 1982 Gurol, M. O., and P. C. Singer, "Kinetics of Ozone Decomposition: A Dynamic Approach," _Environ. Sci. Technol._, **16**, 377 (1982).
+* Hewes and Davison 1971 Hewes, C. G., and R. R. Davison, "Kinetics of Ozone Decomposition and Reaction with Organics in Water," _AIChE J._, **17**, 141 (1971).
+* Hogine and Bader 1976 Hoigne, J., and H. Bader, "The Role of Hydroxyl Radical Reaction in Ozonation Process in Aqueous Solutions," _Wat. Res._, **10**, 377 (1976).
+* Hogine and Bader 1983 Hoigne, J., and H. Bader, "Rate Constant of Reaction of Ozone with Organic and Inorganic Compounds in Water-I. Non-Dissociating Organic Compounds," _Wat. Res._, **17**, 173 (1983).
+* Hsu and Huang 1996 Hsu, Y.-C., and C.-J. Huang, "Characteristics of a New Gas-Induced Reactor," _AIChE J._, **42**, 3146 (1996).
+* Hsu and Huang 1997 Hsu, Y.-C., and C.-J. Huang, "Ozone Transfer with Optimal Design of a New Gas-Induced Reactor," _AIChE J._, **43**, 2336 (1997).
+* Jirka and Carter 1975 Jirka, A. M., and M. J. Carter, "Micro Semi-Automated Analysis of Surface and Wastewaters for Chemical Oxygen Demand," _Analytical Chemistry_, **47**, 1397 (1975).
+* Joshi and Sharma 1977 Joshi, J. B., and M. M. Sharma, "Mass Transfer and Hydrodynamic Characteristics of Gas Inducing Type of Agitated Contactors," _Can. J. Chem. Eng._, **65**, 683 (1977).
+* Mundale and Joshi 2005 Mundale, V. D., and J. B. Joshi, "Optimization of Impeller Design for Gas Inducing Type of Agitated Contactors," _Can. J. Chem. Eng._, **73**, 6 (1995).
+* Perkins _et al._ 1995 Perkins, W. S., W. K. Walsh, I. E. Reed, and C. G. Namboodri, "A Demonstration of Reuse of Spent Dyebath Water following Color Removal with Ozone," _Textile Chemist and Colomist_, **28**, 31 (1995).
+* Saravanan and Joshi 1994 Saaravanan, K., and J. B. Joshi, "Gas Inducing Type Mechanically Agitated Contactors," _Ind. Eng. Chem. Res._, **33**, 2226 (1994).
+* Saunders _et al._ 1983 Saunders, F., M., J. P. Gould, and C. R. Southerland, "The Effect of Solute Competition on Ozonals of Industrial Dyes," _Wat. Res._, **17**, 1407 (1983).
+* Singer and Gurol 1983 Singer, P. C., and M. D. Gurol, "Dynamics of the Ozonation of Phenol: I. Experimental Observations," _Wat. Res._, **17**, 1163 (1983).
+* Sotelo _et al._ 1989 Sotelo, J., L. F. J. Beltran-Heredia, and J. M. Encinar, "Azo Dye Ozonation Film Theory Utilization for Kinetic Studies," _Ozone Sci. Eng._, **11**, 391 (1989).
+* Sotelo _et al._ 1987 Sotelo, J., L. F. J. Benitez, and J. Beltran-Heredia, "Ozone Decomposition in Water: Kinetics Study," _Ind. Eng. Chem. Res._, **26**, 39 (1987).
+* Stahelin and Hoigne 1985 Stahelin, J., and J. Hoigne, "Decomposition of Ozone in Water in the Presence of Organic Solutes Acting as Promoters and Inhibitors of Radical Chain Reactions," _Environ. Sci. Technol._, **19**, 1206 (1985).
+* Tzirii _et al._ 1994 Tzitzi, M., D. V. Vayenas, and G. Lyberatos, " Pretreatment of Textile Industry Wastewaters with Ozone," _Wat. Sci. Tech._, **29**, 151 (1994).
+* Zlokarnik and Judat 1967 Zlokarnik, M., and H. Judat, "Tube and Disk Stirrers--Two Efficient Stirrers for the Gassing of Liquids," _Chem. Ing. Tech._, **39**, 1163 (1967).
+* Zundelevich 1979 Zundelevich, Y., "Power Consumption and Gas Capacity of Self-Inducing Turbo Aerators," _AIChE J._, **25**, 763 (1979).
+* _Manuscript received Oct. 28, 1999, and revision received May 30, 2000._

manual_annotation/200pdfs_in_mmd/10.1002_aws2.1203.mmd

@@ -0,0 +1,135 @@
+# Implementing Ozone-BAC-GAC in potable reuse for removal of emerging contaminants
+
+Ramola Vaidya, Christopher A. Wilson, Germano Salazar-Benites, Amy Pruden, Charles Bott
+
+# Abstract
+
+Removal of contaminants of emerging concern (CECs) is an important consideration in potable reuse. The effectiveness of ozonation, biologically activated carbon (BAC) filtration, and granular activated carbon (GAC) adsorption was assessed in the removal of 96 CECs. During phase 1, O3/total organic carbon (TOC) was varied between 0 and 1.5. Ozone was successful in removing CECs, achieving a total CEC concentration of 20.5 mg/L at an O3/TOC of 1.5 compared with 70 mg/L in absence of ozone. CECs such as iohexol, meprobamate, sucalose and flame retardants, tris(2-carboxyethyl)phosphine (TCEP), tris(1-chloro-2-propyl) phosphate (TCPP), and tris(1,3-dichloro-2-propyl)phosphate (TDCPP) were detected in the ozone effluent even at the highest ozone dose. BAC filters did not contribute measurably to CEC removal, while the 20-min empty bed contact time (EBCT) GAC removed 57.6% of the influent CEC concentration, even though the adsorption capacity had been exerted through previous operation. During phase 2, the average O3/TOC ratio was 0.64, and the impact of new GAC media on CEC removal was studied. EBCT was maintained at 15 min for both GAC contactors, while the used GAC (133,600 bed volume [BV]) was replaced with new media in one GAC contactor. The GAC with new media removed CECs below the limit of quantitation, up to 10,000 BV, while removal decreased to 70% after 20,000 BV, with seven CECs--sucalose, iohexol, acesulfame-k, meprobamate, cotinine, primidone, and acetaminophen--being detected. Ozone doses for CEC removal and BV needed for adsorption of CECs on activated carbon can be determined from this study. Thus, the results from this study can help design and optimize ozone, BAC, and GAC systems used for potable reuse to target the removal of specific CECs.
+
+## Introduction
+
+The purpose of this study is to investigate the effect of removal of emerging contaminants on the generation of emerging contaminants. The purpose of this study is to investigate the effect of removal of emerging contaminants on the generation of emerging contaminants. The removal of emerging contaminants is a key factor in determining the effect of removal of emerging contaminants on the generation of emerging contaminants. The removal of emerging contaminants is a key factor in determining the effect of removal of emerging contaminants on the generation of emerging contaminants. The removal of emerging contaminants is a key factor in determining the effect of removal of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging contaminants on the generation of emerging generation of emerging contaminants on the generation of emerging generation of emerging contaminants on the generation generation generation of
+
+## 1 Introduction
+
+Advanced water treatment (AWT) systems used for potable reuse have started incorporating the use of ozone, biologically activated carbon (BAC) filtration, and granular activated carbon (GAC) adsorption in the treatment process (Hooper et al., 2020; Knopp, Prasse, Ternes, & Cornel, 2016). One of the goals of these treatment processes is the removal of contaminants of emerging concern (CECs). CECs are compounds that are currently not regulated but are frequently detected in wastewater effluent and finished drinking water, such as those included in the Contaminant Candidate List 4 (USEPA, 2020). The US Environmental Protection Agency (USEPA) has made preliminary determinations to regulate perfluorinated compounds such as perfluorooctanoic acid and perfluorooctanesulfonic acid in drinking water and is currently gathering public comments. Growing evidence and awareness of the health and ecological impacts of these CECs have led to stringent monitoring of these compounds in potable reuse applications. Removal of CECs could prove to be a limiting aspect for AWT design for potable reuse, thus emphasizing the importance of identifying factors that influence their removal. Limiting factors for ozone, BAC, and GAC treatments are of particular interest as they combine advanced oxidation, biodegradation, and sorption to optimize CEC removal.
+
+Studies have shown that ozone is a strong oxidant that can remove CECs based on their amenability to reaction with ozone (Gerrity et al., 2012; Sun et al., 2018; Sundaram et al., 2019). While the impact of increasing ozone dose on bulk organics removal in downstream BAC in a pilot-scale process treating tertiary-treated wastewater effluent has shown to be promising (Gifford, Selvy, & Gerrity, 2018), the impact of varying ozone doses on CEC removal in the downstream treatment processes, such as BAC and GAC, has not been widely studied. Furthermore, this removal can vary based on the source water and thus needs to be explored to determine how generalizable the performance is expected to be for implementation in different locations. Removal of CECs by ozonation depends on the ozone reaction second-order rate constants, as well as second-order rate constants with hydroxyl radicals (Lee et al., 2013). Lee, Howe, and Thomson (2012) showed that increasing ozone doses increased the removal of only select compounds, such as atenolol, caffeine, and sulfamethoxazole, in the downstream BAC. These compounds have moderate to high reactivity with ozone (kO3 = 3*106-650 M/s) (Broseus et al., 2009; Gerrity et al., 2012). On the other hand, Lee et al. (2012) and Sundaram et al. (2020) reported poor removal of flame retardants such as tris(2-carboxyethyl)phosphine (TCEP) and tris(1-chloro-2-propyl) phosphate (TCPP), which have low ozone reaction rate second-order constants (kO3 < 1 M/s) and moderate hydroxyl radical rate constant, probably due to known scavengers of hydroxyl radicals produced during ozonation that are typically present in wastewaters (e.g., bicarbonate, DOC, nitrite) (Gerrity et al., 2012; Pocostales, Sein, Knolle, von Sonntag, & Schmidt, 2010). Some compounds such as 1,4-dioxane or meprobamate have low reactivity with ozone (kO3 < 1 M/s) but might still be removed during ozonation due to their high reactivity with hydroxyl radicals (kOH > 109) (Gerrity et al., 2012).
+
+In addition to ozone dose, empty bed contact time (EBCT) is an important design criterion for BAC filters. The removal of certain CECs, such as N,N-Diethyl-meta-toluamide (DEET), naproxen, and ibuprofen, has been shown to be improved with a higher EBCT of 14 min compared to 5 min in biofilters containing anthracite (Halle, Huck, & Peldszus, 2015), demonstrating that these compounds are primarily removed by biodegradation. There are other studies showing the importance of BAC in the biodegradation of CECs (Abromaitis, Racys, van der Marel, & Meulepas, 2016; Greenstein, Lew, Dickenson, & Wert, 2018; Nugroho, Reungoat, & Keller, 2010). Furthermore, although GAC units are intended to sorb CECs, in fact, they will be readily colonized by microorganisms that could also contribute to CEC removal (Rattier, Reungoat, Gernjak, Joss, & Keller, 2012). Thus, mechanistically, both adsorption and biodegradation should be taken into account as being responsible for CEC removal in GAC units that have been in operation for extensive periods of time. In such situations, EBCT may have an impact on CEC removal, relative to units with fresh GAC, where EBCT should not control adsorption performance as long as the EBCT is longer than the mass transfer zone.
+
+Adsorption of CECs on GAC has been widely studied and has been primarily used for CEC removal in situations where the application of reverse osmosis is not feasible (Reungoat et al., 2010). As the adsorption capacity of GAC with respect to specific organic contaminants declines, there is poor removal of certain CECs, resulting in the breakthrough of these compounds. Although there are studies, such as Sun et al. (2018) and Sundaram et al. (2020), demonstrating the impact of bed volume (BV) processed on removal of CECs, there is little information available on the impact of the upstream treatment for CEC removal on the downstream GAC performance. Thus, there is a need to better understand the combinedimpacts of ozone dose, EBCT in BAC and GAC, and media age (measured by GAC [BV] treated) on the removal of CECs for potable reuse purposes. The objective of this study was to determine the impact of varying EBCT, ozone dose, and GAC BV treated on CEC removal in BAC and GAC. The importance of BV in removing specific CECs and the consequences on designing AWT systems are expanded upon based on the results of this study.
+
+## 2 Materials and Methods
+
+### Advanced water treatment process
+
+#### Pilot-scale process
+
+A pilot-scale advanced treatment process (13.8 L/min) consisting of coagulation flocculation and sedimentation followed by ozonation and BAC and GAC was performed at Hampton Roads Sanitation District's (HRSD's) Nansemond treatment plant, as described in a prior technical report (Pruden, Bott, Blair, Miller, & Vaidya, n.d.). Source water to the pilot consisted of fully nitrified and denitrified secondary effluent from a five-stage Bardenpho treatment process with methanol addition for enhanced denitrification. In the coagulation/flocculation/sedimentation process, the secondary effluent initially passed through two rapid mix stages, where the coagulant aluminum chlorohydrate was dosed at 25 mg/L (as product) followed by cationic floc-aid polymer (Hychem, Inc. Hyperfloc CE 824) at a dose of 0.75 mg/L (as active product). There were two rapid mix stages with a hydraulic retention time of 35 s/stage. The flocculation step consisted of three stages with tapered Gt values of 44,880; 22,440; and 11,220. The sedimentation step consisted of lamella plate settlers with a loading rate of 4.1 L/m2/min.
+
+Effluent from the settling basin was ozonated in a 0.15 m x 4.1 m (w x h) reactor with fine bubble diffusion. Four additional columns of the same dimensions were added post the ozone reactor to allow for a total contact time of 21 min. The applied ozone dose was controlled based on online measurement of ozone residual using dissolved ozone probes such that O3/total organic carbon (TOC) ratio was varied from 0 to 1.5. The dissolved ozone probes measured the ozone residual in water (mg/L). A high O3/TOC ratio of 1.5 was used to assess the impact of this dose on CEC removal in the downstream BAC and GAC and identify the CECs that are not removed by ozone even at this high dose. The O3/TOC ratio was calculated using 90% transfer efficiency and corrected for nitrite, which was measured using an online analyzer (Xylem TresCon, Rye Brook, NY). Ozone gas analyzers (IN USA, Inc., San Diego, CA) were used to determine the feed and the off gas ozone concentration and determine the gas transfer efficiency. The ozone effluent was split into two BACs operating in parallel. The two BACs were operated at different EBCTs of 5 and 10 min, denoted as BAC5 and BAC10 with loading rates of 305 and 151 L/m2/min, respectively. The effluent from BAC5 and BAC10 was then fed to GAC10 and GAC20, respectively, representing EBCTs of 10 and 20 min with loading rates of 183 and 106 L/m2/min, respectively. The activated carbon media used in the BAC filter was Calgon Filtrasorb 816 with an effective size of 1.4 mm and uniformity coefficient of 1.4 and was supported on 0.3 m of sand with an effective size of 0.45-0.55 mm and a uniformity coefficient of 1.6. The media used in GAC was Calgon Filtrasorb 400 with an effective size of 0.55-0.75 mm and a uniformity coefficient of 1.9. These BAC and GAC columns were operated for a period of 2 years prior to this study (using a different water source) (Pruden et al., n.d.; Vaidya et al., 2019). Thus, the media in the BAC were exhausted from a TOC standpoint and carried a mature biofilm, while the GAC10 and GAC20 had processed 100,000 and 50,000 BV by the time of this study, respectively. BAC was backwashed using the GAC10 effluent when the head loss in the filter increased above 3 m or when the effluent turbidity increased above 0.1 nephelometric turbidity units. This resulted in a typical run times of 2 days for BAC5 and 4 days for BAC10. The backwash frequency for GACs was once every 3-4 months, resulting in longer run times.
+
+#### Demonstration-scale process
+
+The demonstration-scale process (3.78 ML/day) was similar to the pilot-scale process and consisted of coagulation, flocculation, and sedimentation followed by ozonation. The source water was identical to the pilot-scale process. The coagulant used in the 3.78 ML/day was the same as the pilot-scale coagulation process and was fed at the same dose. Clarifiloc C-6220 was used as a polymer and added to the rapid mix stage to achieve a dose of 0.75 mg/L. Coagulation G values were identical to the pilot coagulation system. The plate loading rate for the settled basin was 8.22 L/m2/min. Ozone was supplied through a sidestream venturi injector to 50% of the process flow. The mainstream serpentine ozone contactor had an 8.5-min contact time. Ozone transfer efficiency was typically between 87% and 93%. Ozone dose was controlled to achieve >3 log virus removal and 1.5 log removal of giardia according to the USEPA Disinfection Profiling and Benchmarking Guidance Manual(USEPA, 1999). The CT needed to maintain this disinfection was calculated by measuring dissolved ozone residual with dissolved ozone probes located along the contactor. An average 0.6-0.7 O3/TOC ratio was needed to achieve >3 log virus removal with a CT of >0.4 at an average temperature of 20 degC based on the USEPA Disinfection Profiling and Benchmarking Guidance Manual. The average bromide concentration in the secondary effluent was 0.187 mg/L with a standard deviation of 0.05 mg/L. Preformed monochloramine was added at a dose of 5 mg/L Cl2 before ozonation to control bromate formation from bromide in the source water to below the drinking water primary maximum contaminant limit of 10 mg/L (Buffle, Galli, & Von Gunten, 2004). The monochloramine residual postozone was dechlorinated using sodium bisulfite prior to BAC.
+
+### Experimental phases
+
+The experimental study was divided into two phases. Phase 1 included the first five sampling events in which the pilot BAC and GAC were fed from the pilot ozone unit and the ozone/TOC ratio, corrected for nitrite, was varied from 0 to 1.5. In case of zero ozone/TOC ratio, the ozone generator was shut down, but the dissolved oxygen concentration was maintained at 35 mg/L. The EBCT for BAC and GAC was not modified, resulting in the BAC5 effluent being the influent for GAC10. Similarly, the BAC10 effluent was the influent for GAC20. Phase 2 included the last six sampling events in which the feedwater to the pilot BAC and GAC was switched to the demonstration-scale ozone system. The ozone/TOC ratio was not varied and had an average value of 0.64 with a standard deviation of 0.05. At the beginning of Phase 2, the GAC in GAC10 was replaced with fresh media to determine the impact on CEC removal, while the media in GAC20 were not changed. The EBCT for both BAC filters was maintained at 10 min. Thus, BAC5 was renamed BAC10A, and BAC10 was renamed BAC10B. Similarly, GAC EBCT was changed to 15 min, in part because of requirements for a parallel experiment, resulting in GAC10 being renamed GAC15A, with new media, and GAC20 being renamed GAC15B, with the old media.
+
+### CEC analysis
+
+A suite of 96 different CECs was analyzed in the effluents of ozone, BAC, and GAC units by Eurofins Eaton Analytical (Monrovia, CA). Samples were collected for a period of 14 months and included a total of 11 sampling events. The analysis was performed using solid-phase extraction high-performance liquid chromatography with tandem mass spectrometry (HPLC-MS-MS). Of the sample volume, 2.5 ml was extracted using Oasis HLB solid-phase extraction column. The HPLC columns used were Kinetex, 1.7 mm F5 and Waters, X-Select HSS T3, 2.5 mM, 2.1 x 75 mm Column XP. The flow rate in the HPLC column was maintained at 1.3 ml/min. The mobile phase used was a gradient organic solvent consisting of ammonium fluoride, formic acid, and acetonitrile. Analysis was carried out by using the electrospray ionization source in positive and negative ionization mode and multiple reaction monitoring. Matrix effects on electrospray ES-MS/MS signals were found to be adequately corrected by using isotopically labeled standards when available. The method-reporting limits for TOC and CECs are reported in Table S1.
+
+## 3 Results and discussion
+
+### Total CEC removal in phases 1 and 2
+
+The total concentrations and numbers of CECs detected in ozone, BAC, and GAC in phase 1 are shown in Figure 1. Ozone effluent had an average total CEC concentration of 50 ug/L while that in BAC10 was 42 ug/L and in BAC5 was 40 ug/L. Out of the 20 CECs detected in ozone effluent, an average of only 2 CECs were removed in BAC10 and 1 in BAC5. This low average removal of 15% in BAC could be attributed to varying ozone doses, thus changing the amount of CEC mass loaded on to the BAC media. Thus, the ozone effluent or BAC influent CEC concentration varied as shown in Figure 1. This resulted in sampling events when there was desorption in BAC, causing the overall CEC removal to be negative. It has been shown that an increase in EBCT is beneficial to the removal of CECs by biodegradation (Halle et al., 2015). However, there was no benefit with the longer EBCT BAC either given that the average BAC10 CEC concentration was greater than the average BAC5 CEC concentration. This shows that the CECs were recalcitrant, and increasing EBCT from 5 to 10 min did not improve the biodegradation of these compounds. There was an average of 57.6% removal in GAC20, while this removal was lower (12.5%) for GAC10 due to negative removal during one sampling event. This shows that these CECs were removed in the GAC by adsorption even after the GAC was in service for 2 years and that some of these compounds are removed by adsorption but not bio-degradation. The number of CECs detected in BAC5 and BAC10 were not different from ozone effluent, which had 20 CECs. GAC10 and GAC20 reduced these numbers to 13 and 9, respectively. This shows that GAC was better at removing total CECs compared to BAC and that the removal was a function of the BV of water processed by the GAC.
+
+For Phase 2, the advantage of replacing the media in GAC15A was apparent from the CEC removal and CECs detected in the GAC15A effluent (Figures 2 and 3). There was no desorption from the BAC indicated by any sampling events given that the ozone dose was not varied in Phase 2. The average CEC removal was 16.2% for BAC10A and 17.6% for BAC10B and was similar to Phase 1. The average removal for GAC15B dropped to 31% in Phase 2 (compared to 57.65 in GAC20 during phase 1), showing that reducing the EBCT from 20 to 15 min affected the CEC removal. This suggests that the removal in GAC after 65,000 BV is a combination of both adsorption and biodegradation and a 20-min EBCT is beneficial to CEC removal. There were 16 CECs detected in the ozone effluent with similar detections in both BACs, while the CECs detected in GAC15A and GAC15B were 4 and 12, respectively. The removal of CECs in GAC15A was close to 100%, without any CECs being detected in sampling events 1 and 2 (Figure 3). This removal declined slowly in further sampling events, showing that the activated carbon media was being exhausted.
+
+Figure 2: Total concentration (a) and number (\(\#\)CECs) (b) of contaminants of emerging concerns (CECs) detected in the effluent of ozone, biologically activated carbon (BAC), and granular activated carbon (GAC) during phase 2; 10 and 15 denote the empty bed contact time (min). GAC15A contained new media (20,000 BV), while GAC15B contained old media (80,000 BV). CECs are the 96 compounds listed inTable S1. Error bars represent _SD_ in six sampling events carried out over 7 months during Phase 2. Source: Pruden et al. forthcoming. Data subject to new analysis with permission. © The Water Research Foundation
+
+### CEC removal with varying ozone doses
+
+Ozone dose was varied between 0 and 1.5 mg/L during Phase 1. The CEC concentration in ozone effluent and BAC effluent varied with ozone dose (Figure 4). Ozone effluent had the highest total CEC concentration (sum of CECs above detection limit) of 70 mg/L during two sampling events when ozone was shut down. This concentration decreased with increasing ozone dose and was the least (20.5 mg/L) with an O3/TOC ratio of 1.47. This trend was the same for the number of CECs detected, which decreased from 35 in the absence of ozone to 6 at an O3/TOC ratio of 1.47. This shows that ozone itself plays an important role in the removal of CECs, and ozone operation will dictate the concentration of CECs in the downstream processes. As noted previously, there was no measurable removal in BAC, and the effluent CEC concentration in BAC was affected by the upstream ozone dose and ozone effluent CEC concentration. The removal in BAC was substantial as GAC20 decreased CEC concentration to 15 mg/L, with 10 or fewer CECs detected in the effluent, regardless of the ozone dose. Table S2 shows the ozone second-order rate constants of CECs that were detected when ozone was not in service and at the highest O3/TOC dose of 1.47. The CECs detected in GAC20 at the highest O3/TOC dose of 1.47 included artificial sweeteners acesulfame-k and sacralose; an x-ray contrasting agent, iohexol; a prescription drug for anxiety, meprobamate, and cotinine (a metabolite of nicotine); and flame retardant TCEP. These compounds were poorly removed by ozone, even at this highest dose, due to their low ozone second-order rate constants. Such compounds may prove to be important indicators for treatment operation. Similar results were reported by Sun et al. (2018), where removal of meprobamate and cotinine was less than 60% at O3:dissolved organic carbon ratio of 0.65, while there was no difference in the removal of flame retardants TCEP and TCPP with and without
+
+Figure 4: Total concentration (a) and number (b) of contaminants of emerging concerns (CECs) detected in the effluent of ozone, biologically activated carbon (BAC), and granular activated carbon (GAC) during Phase 1 with varying O3/total organic carbon (TOC) ratio corrected for nitrite; 5, 10, and 20 denote the empty bed contact time (min) for the BAC and GAC stages. The data cover a total of five sampling events (during a period of 6 months), with two sampling events occurring at an O3/TOC ratio of 0 when the ozone generator was shut down, but the dissolved oxygen concentration was maintained at 35 mg/L. Source: Pruden et al. forthcoming. Figure modified and reprinted with permission. © The Water Research Foundation
+
+Figure 3: Percentage of contaminants of emerging concern (CEC) removed by biologically activated carbon (BAC) and granular activated carbon (GAC) treatment steps during six sampling events in a span of 7 months during Phase 2. Removal was calculated based on the effluent concentration in the immediate upstream stage; 10 and 15 denote the empty bed contact time (min). GAC15A contained new media, while GAC15B contained old mediaozone dose. As mentioned previously, the poor removal of these compounds is probably a product of low reactivity with ozone and only moderate reactivity with hydroxyl radical that is probably offset by a relatively high scavenging rate for hydroxyl radicals. Acesulfame-k concentrations in GAC20 effluent were greater than the influent concentration by an order of magnitude, indicating desorption of previously adsorbed concentrations of acesulfame-k. Scheurer, Storck, Brauch, and Lange (2010) showed that acesulfame-k was removed by ozone at ozone doses greater than 5 mg/L. Compounds such as acetaminophen, carbamazepine, diclofenac, estrone, naproxen, triclosan, gemfibrozil, and fluoxetine were detected only when ozone was not in service. The moderate to high ozone second-order rate constants for these CECs (Table S2) make them conducive to removal by ozone and emphasizes the importance of ozone in an AWT system for the removal of such compounds.
+
+These results also show that GAC has capacity for removing some CECs through a combination of adsorption and biodegradation. The removal in GAC would then depend on upstream ozone dose and the mass of CECs removed in ozone. The mass of CECs removed in ozone could not be calculated in this case as samples were not collected in the ozone influent; nonetheless, the variation of CEC concentration in the ozone effluent can be credited to the varying ozone doses. These results from GAC show that ozone dose control and monitoring are vital during optimization of the shelf life of activated carbon media in the GAC. For AWT systems, where one of the main goals of GAC columns is CEC removal, the effluent CEC concentration can be a factor for determining the media replacement. The frequency of media replacement in GAC can affect the design of the treatment process, and one way of reducing this frequency could be to achieve the optimal CEC removal in upstream ozone. At the same time, a higher ozone dose can result in increased disinfection byproducts such as bromate and nitrosamines (Buffle et al., 2004; Marti, Pisarenko, Peller, & Dickenson, 2015). The ozone dose will then be controlled in a manner that produces the least disinfection byproducts and at the same time achieves maximum CEC removal.
+
+### CEC removal in new GAC media
+
+As noted previously, removal of CECs in GAC15A during Phase 2 decreased with sampling events and was thus a function of the GAC BV treated, as expected. Figure 5 shows this decrease in removal in CECs and the increase in CEC concentrations and number of CECs detected with increasing BV. With the CEC removal decreasing from 100% to 70% at 20,000 BV, it can be assumed that adsorption was the primary and dominant mechanism for CEC removal during this period. This removal was in the range of 30%-45% for a GAC15B, which had the same EBCT but had processed greater than 65,000 BV. Figure 6a,b compare percent CEC removal between GAC15A and GAC15B during two sampling events denoting the two different bed volumes. Figure 6a denotes the time period when the media in GAC15A were unused, while Figure 6b denotes the time after 21,000 BV of water was processed by GAC15A.
+
+The CECs that were detected in GAC15A included acesulfame-k and sucralose, the pain-relieving drug acetaminophen (a.k.a. paracetamol, was present near detection limit); cotinine (an alkaloid found in tobacco); iohexol; prescription drugs for anxiety disorders and epilepsy; merobamate; and primidone. Table S3 summarizes the octanol-water partition coefficient, log K\({}_{\text{ow}}\), for CECs that were detected in BAC or GAC effluent. It was observed that the CEC removal efficiency of GAC15A was greater than that of GAC15B for all compounds shown in Figure 6a. The removal of DEET and atrazine was comparable for both GAC columns owing to their greater log K\({}_{\text{ow}}\) values of 2.18 and 2.61, respectively. Studies have reported the consistent detection of artificial sweeteners, such as acesulfame-k and sucralose, in AWT systems, thus making them excellent treatment effectiveness indicators (Doummar & Aoun, 2018). Acesulfame-k and sucralose have a low octanol-water partition coefficient, log K\({}_{\text{ow}}\) (Table S3), thus making it difficult to remove these compounds by adsorption. Figure 6a,b
+
+Figure 5: Total contaminants of emerging concern (CEC) concentration, number of CECs detected (±CECs), and percentage of CECs removed in GAC15A (15 min of empty bed contact time) during Phase 2. BV denotes the bed volume of water processed by the media in GAC15A. The removal in GAC15A was calculated based on the effluent CEC concentration in BAC10A (10 min of empty bed contact time)
+
+show that the removal efficiency of these artificial sweeteners decreased in GAC15A after 21,000 BV and even resulted in the desorption of acesulfame-k. Desorption in acesulfame-k was observed in GAC15A after 21,000 BV and is in agreement with Scheurer et al. (2010), where acesulfame-k desorption from GAC media depended on the concentration of acesulfame-k initially adsorbed on GAC. Desorption of a compound indicates that it was previously removed by adsorption and not biodegradation. Sundaram and Pagilla (2019) showed that more than 90% of sucralose was removed by adsorption onto GAC media, while Greenstein et al. (2018) did not report removal of sucralose as the GAC media was already exhausted. Cotinine is poorly removed by ozone and biodegradation and has shown to breakthrough in an 18-min EBCT GAC at 12,500 BV (Kennedy, Reinert, Knappe, Ferrer, & Summers, 2015). Acetaminophen has been shown to be well removed by ozonation (Knopp et al., 2016; Mojiri, Vakili, Farraji, & Aziz, 2019). These detections of acetaminophen in GAC were not expected, but they were present at low average concentrations of 0.008 ug/L with a limit of quantitation of 0.005 ug/L, which could have been detected only due to sample contamination. Iohexol is poorly removed in GAC due to its high polarity and low log Kow of -3.05 and has shown to be removed by advanced oxidation using UV peroxide (Hu et al., 2019). The removal of iohexol decreased from 100% to 60% in GAC15A after 21,000 BV (Figure 6). Meprobamate is weakly reactive with ozone and has been shown to be removed by adsorption on GAC (Gerrity et al., 2012; Sundaram et al., 2019). Meprobamate was detected in the last two sampling events, showing that the removal decreases with increasing BV, again consistent with a low log Kow 0.7. Primidone has a low octanol-water partition coefficient of 0.91 (Wjekoon et al., 2013), making it difficult to be removed by adsorption. It was interesting to note that all of these compounds were detected in GAC15B at concentrations greater than those in GAC15A effluent, showing that the number of BV of water processed affects the removal of these compounds. CECs such as atrazine, bisphenol a (BPA), cariospprodol, dehydroifiedipine, and Lopressor, which have log Kow values of \(\sim\)2 or greater, were either not detected or were measured at levels near the detection limit in the effluent of both GAC units. BPA was detected in GAC15B in one sampling event but at a level near the detection limit. Flame retardants such as TCEP, TCPP, and TDCPP are weakly removed by ozone due to their low second-order rate constants (Table S2). However, these flame retardants have log Kow values of \(\sim\)2 or greater (Table S3) and hence are removed by adsorption. These compounds were removed to below the detection limit by GAC15A even after 21,000 BV but were detected consistently in the GAC15B effluent.
+
+## 4 Conclusions
+
+CEC removal in an ozone-BAC-GAC pilot was explored in an AWT system with varying ozone doses, BV, and EBCTs during two different phases. Several key conclusions from this study include the following:
+
+Figure 6: Contaminants of emerging concern (CEC) removal efficiency (%) in GAC15A and GAC15B (15 min of empty bed contact time) during one sampling event at (a) bed volume = 0 and 67,000 in GAC15A and GAC15B, respectively, and during (b) bed volume = 21,000 and 83,000 in GAC15A and GAC15B, respectively, during Phase 2. The removal in GAC15A and GAC15B was calculated based on the effluent CEC concentration in BAC10A and BAC10B, respectively. TCEP, tris(2-carboxyethyl) phosphine; TCPP, tris(1-chloro-2-propyl) phosphate; TDCPP, tris (1,3-dichloro-2-propyl)phosphate *Denotes negative removals in GAC15B for Figure 6a and negative removals in both GAC15A and GAC15B for Figure 6b
+
+* Ozone dose affected CEC removal and was critical in determining the BAC and GAC effluent CEC concentrations.
+* Removal of CECs by BAC was negligible, and changing EBCT from 5 to 10 min did not improve CEC removal.
+* Contribution of GAC toward CEC removal was important, with 20-min EBCT resulting in >55% CEC removal.
+* CECs removed by GAC with greater than 65,000 BV depended on the upstream ozone dose.
+* CEC removal in Phase 2, due to adsorption on fresh GAC media, was a function of BV. The removal decreased from 100% to 70% after 20,000 BV.
+* With increasing BV, GAC media adsorption sites are utilized by total organic carbon, causing CEC removal to decrease. The BV necessary for GAC media replacement will thus depend on the nature and concentration of CECs that initially breakthrough.
+
+## Acknowledgment
+
+The authors thank the operators and staff at HRSD's Nansemond and SWIFT treatment plants. Funding for this effort was provided in part by the Water Research Foundation Unsolicited Award U1R16 (PI Amy Pruden) and financial and in-kind support from the Hampton Roads Sanitation District and Jacobs Engineering.
+
+## Conflict of Interest
+
+There are no conflicts of interest to declare.
+
+## ORCID
+
+_Ramola Vaidya_ (c) [https://orcid.org/0000-0001-9608-6432](https://orcid.org/0000-0001-9608-6432)
+
+## References
+
+* Abromaitis et al. (2016) Abromaitis, V., Racys, V., van der Marel, P., & Meulepas, R. J. W. (2016). Biodegradation of persistent organics can overcome adsorption-desorption hysteresis in biological activated carbon systems. _Chemosphere_, _149_, 183-189. [https://doi.org/10.1016/j.chemosphere.2016.01.085](https://doi.org/10.1016/j.chemosphere.2016.01.085)
+* Broseus et al. (2009) Broseus, R., Vincent, S., Aboulfadl, K., Daneshvar, A., Sauve, S., Barbeau, B., & Prevost, M. (2009). Ozone oxidation of pharmaceuticals, endocrine disruptorsand pesticides during drinking water treatment. _Water Research_, _43_, 4707-4717. [https://doi.org/10.1016/j.watres.2009.07.031](https://doi.org/10.1016/j.watres.2009.07.031)
+* Buffle et al. (2004) Buffle, M. O., Galli, S., & Von Gunten, U. (2004). Enhanced bromate control during ozonation: The chlorine-ammonia process. _Environmental Science and Technology_, _38_(19), 5187-5195. [https://doi.org/10.1021/es0352146](https://doi.org/10.1021/es0352146)
+* Doumaru & Aoun (2018) Doumaru, J., & Aoun, M. (2018). Assessment of the origin and transport of four selected emerging micropollutants sucralose, Acesulfame-K, gemfbrozil, and iohexol in a karst spring during a multi-event spring response. _Journal of Contaminant Hydrology_, _215_(June), 11-20. [https://doi.org/10.1016/j.jconhyd.2018.06.003](https://doi.org/10.1016/j.jconhyd.2018.06.003)
+* Gerrity et al. (2012) Gerrity, D., Gamage, S., Jones, D., Korshin, G. V., Lee, Y., Pisarenko, A.,... Snyder, S. A. (2012). Development of surrogate correlation models to predict trace organic contaminant oxidation and microbial inactivation during ozonation. _Water Research_, _46_(19), 6257-6272. [https://doi.org/10.1016/j.watres.2012.08.037](https://doi.org/10.1016/j.watres.2012.08.037)
+* Gifford et al. (2018) Gifford, M., Selvy, A., & Gerrity, D. (2018). Optimizing ozone-biofililtration systems for organic carbon removal in potable reuse applications. _Ozone: Science & Engineering_, _40_(6), 427-440. [https://doi.org/10.1080/01919512.2018.1509203](https://doi.org/10.1080/01919512.2018.1509203)
+* Greenstein et al. (2018) Greenstein, K. E., Lew, J., Dickenson, E. R. V., & Wert, E. C. (2018). Investigation of biotransformation, sorption, and desorption of multiple chemical contaminants in pilot-scale drinking water biofilters. _Chemosphere_, _200_, 248-256. [https://doi.org/10.1016/j.chemosphere.2018.02.107](https://doi.org/10.1016/j.chemosphere.2018.02.107)
+* Halle et al. (2015) Halle, C., Huck, P. M., & Peldszus, S. (2015). Emerging contaminant removal by biofiltration: Temperature, concentration, and EBCT impacts. _Journal AWWA_, _107_(7), E364-E379. [https://doi.org/10.5942/jawwa.2015.107.0086](https://doi.org/10.5942/jawwa.2015.107.0086)
+* Hooper et al. (2020) Hooper, J., Funk, D., Bell, K., Noibi, M., Vickstrom, K., Schulz, C.,... Huang, C. (2020). Pilot testing of direct and indirect potable water reuse using multi- stage ozone biofiltration without reverse osmosis. _Water Research_, _169_, 115178. [https://doi.org/10.1016/j.watres.2019.115178](https://doi.org/10.1016/j.watres.2019.115178)
+* Hu et al. (2019) Hu, C. Y., Hou, Y. Z., Lin, Y. L., Deng, Y. G., Hua, S. J., Du, Y. F.,... Wu, C. H. (2019). Kinetics and model development of iohexol degradation during UV/H\({}_{2}\)O\({}_{2}\) and UV/oxidation. _Chemosphere_, _229_, 602-610. [https://doi.org/10.1016/j.chemosphere.2019.05.012](https://doi.org/10.1016/j.chemosphere.2019.05.012)
+* Kennedy et al. (2015) Kennedy, A. M., Reinert, A. M., Knappe, D. R. U., Ferrer, I., & Summers, R. S. (2015). Full- and pilot-scale GAC adsorption of organic micropollutants. _Water Research_, _68_, 238-248. [https://doi.org/10.1016/j.watres.2014.10.010](https://doi.org/10.1016/j.watres.2014.10.010)
+* Knopp et al. (2016) Knopp, G., Prasse, C., Ternes, T. A., & Comel, P. (2016). Elimination of micropollutants and transformation products from a wastewater treatment plant effluent through pilot scale ozonation followed by various activated carbon and biological filters. _Water Research_, _100_, 580-592. [https://doi.org/10.1016/j.watres.2016.04.069](https://doi.org/10.1016/j.watres.2016.04.069)
+* Lee et al. (2012) Lee, C. O., Howe, K. J., & Thomson, B. M. (2012). Ozone and biofiltration as an alternative to reverse osmosis for removing PPCPs and micropollutants from treated wastewater. _Water Research_, _46_, 1005-1014. [https://doi.org/10.1016/j.watres.2011.10.669](https://doi.org/10.1016/j.watres.2011.10.669)
+* Lee et al. (2013) Lee, Y., Gerrity, D., Lee, M., Bogeat, A. E., Salhi, E., Gamage, S.,... von Gunten, U. (2013). Prediction of micropollutant elimination during ozonation of municipal wastewater effluents: Use of kinetic and water specific information. _Environmental Science & Technology_, _47_(11), 5872-5881. [https://doi.org/10.1021/es400781r](https://doi.org/10.1021/es400781r)
+* Marti et al. (2015) Marti, E. J., Pisarenko, A. N., Peller, J. R., & Dickenson, E. R. V. (2015). N-nitrosodimethylamine (NDMA) formation from the ozonation of model compounds. _Water Research_, _72_, 262-270. [https://doi.org/10.1016/j.watres.2014.08.047](https://doi.org/10.1016/j.watres.2014.08.047)
+* Mojiri et al. (2019) Mojiri, A., Vakili, M., Farraji, H., & Aziz, S. Q. (2019). Combined ozone oxidation process and adsorption methods for the removal of acetaminophen and amoxicillin from aqueous solution; kinetic and optimization. _Environmental Technology and Innovation_, _15_, 100404. [https://doi.org/10.1016/j.eti.2019.100404](https://doi.org/10.1016/j.eti.2019.100404)* Nugroho et al. (2010) Nugroho, W. A., Reungoat, J., & Keller, J. (2010). The performance of biologically activated carbon in removing pharmaceuticals in drinking water treatment. _Journal of Applied Sciences in Environmental Sanitation_, 5(2), 131-141.
+* Pocostales et al. (2010) Pocostales, J. P., Sein, M. M., Knolle, W., von Sonntag, C., & Schmidt, T. C. (2010). Degradation of ozone-refractory organic phosphates in wastewater by ozone and ozone/hydrogen peroxide (peroxone): The role of ozone consumption by dissolved organic matter. _Environmental Science & Technology_, 44(21), 8248-8253. [https://doi.org/10.1021/es1018288](https://doi.org/10.1021/es1018288)
+* Pruden et al. (2010) Pruden, A., C. Bott, M. Blair, J. Miller, & R. Vaidya. n.d. _Characterization of organic carbon and microbial communities for the optimization of biologically-active carbon (BAC) filtration for potable reuse_. Project U1R16. Alexandria, VA: The Water Research Foundation.
+* Rattier et al. (2012) Rattier, M., Reungoat, J., Gernjak, W., Joss, A., & Keller, J. (2012). Investigating the role of adsorption and biodegradation in the removal of organic micropollutants during biological activated carbon filtration of treated wastewater. _Journal of Water Reuse and Desalination_, 2(3), 127-139. [https://doi.org/10.2166/wrd.2012.012](https://doi.org/10.2166/wrd.2012.012)
+* Reungoat et al. (2010) Reungoat, J., Macova, M., Escher, B. I., Carswell, S., Mueller, J. F., & Keller, J. (2010). Removal of micropollutants and reduction of biological activity in a full scale reclamation plant using ozonation and activated carbon filtration. _Water Research_, 44, 625-637. [https://doi.org/10.1016/j.watres.2009.048](https://doi.org/10.1016/j.watres.2009.048)
+* Scheurer et al. (2010) Scheurer, M., Storck, F. R., Brauch, H. J., & Lange, F. T. (2010). Performance of conventional multi-barrier drinking water treatment plants for the removal of four artificial sweeteners. _Water Research_, 44(12), 3573-3584. [https://doi.org/10.1016/j.watres.2010.04.005](https://doi.org/10.1016/j.watres.2010.04.005)
+* Sun et al. (2018) Sun, Y., Angelotti, B., Brooks, M., Dowbiggin, B., Evans, P. J., Devins, B., & Wang, Z. W. (2018). A pilot-scale investigation of disinfection by-product precursors and trace organic removal mechanisms in ozone-biologically activated carbon treatment for potable reuse. _Chemosphere_, 210, 539-549. [https://doi.org/10.1016/j.chemosphere.2018.06.162](https://doi.org/10.1016/j.chemosphere.2018.06.162)
+* Sundaram & Pagilla (2019) Sundaram, V., & Pagilla, K. (2019). Trace and bulk organics removal during ozone-biofiltration treatment for potable reuse applications. _Water Environment Research_, 92, 430-440. [https://doi.org/10.1002/wer.1202](https://doi.org/10.1002/wer.1202)
+* Sundaram et al. (2020) Sundaram, V., Pagilla, K., Guarin, T., Li, L., Marfil-Vega, R., & Bukhari, Z. (2020). Extended field investigations of ozone-biofiltration advanced water treatment for potable reuse. _Water Research_, 172, 115513. [https://doi.org/10.1016/j.watres.2020.115513](https://doi.org/10.1016/j.watres.2020.115513)
+* United States Environmental Protection Agency (2020) United States Environmental Protection Agency. (2020) _Contaminant candidate list (CCL) and regulatory determination_. Retrieved from [https://www.epa.gov/ccl/contaminant-candidate-list-4-ccl-4-0](https://www.epa.gov/ccl/contaminant-candidate-list-4-ccl-4-0)
+* United States Environmental Protection Agency (1999) United States Environmental Protection Agency, EPA. (1999). _Disinfection profiling and benchmarking guidance manual, office of water_ (4607). EPA 815-R-99-013.
+* Vaidya et al. (2019) Vaidya, R., Buehlmann, P. H., Salazar-Benites, G., Schimmoller, L., Nading, T., Wilson, C. A.,... Novak, J. T. (2019). Pilot plant performance comparing carbon-based and membrane-based potable reuse schemes. _Environmental Engineering Science_, 36, 1369-1378. [https://doi.org/10.1089/ees.2018.0559](https://doi.org/10.1089/ees.2018.0559)
+* Wijekoon et al. (2013) Wijekoon, K. C., Hai, F. I., Kang, J., Price, W. E., Guo, W., Ngo, H. H., & Nghiem, L. D. (2013). The fate of pharmaceuticals, steroid hormones, phytoestrogens, UV-filters and pesticides during MBR treatment. _Biovesource Technology_, 144, 247-254. [https://doi.org/10.1016/j.biortech.2013.06.097](https://doi.org/10.1016/j.biortech.2013.06.097)
+
+## Supporting Information
+
+Additional supporting information may be found online in the Supporting Information section at the end of this article.
+
+**How to cite this article:** Vaidya R, A. Wilson C, Salazar-Benites G, Pruden A, Bott C. Implementing Ozone-BAC-GAC in potable reuse for removal of emerging contaminants. _AWWA Wat Sci._ 2020; e1203. [https://doi.org/10.1002/aws2.1203](https://doi.org/10.1002/aws2.1203)

manual_annotation/200pdfs_in_mmd/10.1002_cjce.20327.mmd

@@ -0,0 +1,261 @@
+On Bubble Column Reactor Design for the Determination of Kinetic Rate Constants in Gas-Liquid Systems
+
+S. C. Cardona, F. Lopez, A. Abad, J. Navarro-Laboulais
+
+# Abstract
+
+The design of a semibatch bubble column reactor with its mathematical description is proposed for the study of ozonation reactions. The mathematical model used to describe the gas-liquid mass transfer rate in the reactor is based on the unstationary film theory and the resulting model is theoretically analysed to identify its relevant parameters. After its structural identifiability analysis, the parameters are reduced to five, that is, the gas hold-up, the ratio of diffusivities of the reacting species, the volumetric mass transfer coefficient and two time constants related with the kinetic rate constant. From the sensitivity analysis of this reduced model, we conclude that it is not sensible to the gas hold-up and the diffusivity ratio of the reacting species for optimization purposes in moderate and slow kinetic regimes. The model is tested with the reaction between the ozone and the azo-compound Acid Red 27. The experimental data match quite well the model allowing the estimation of the volumetric mass transfer coefficient together with the kinetic constant. The kinetic rate constant for the direct reaction between the ozone and the Acid Red 27 is estimated in \(k_{2} = 3723 \pm 127\) M\({}^{- 1}\)s\({}^{- 1}\) at \(21.2\pm 0.5^{-}\)C. The self-coherence of the model, the absence of hypothesis about the state of the film together with the proposed optimization procedure, allows to consider the proposed methodology as a viable alternative for the study of gas-liquid systems in semi-batch bubble columns reactors in comparison with classical approaches.
+
+## Introduction
+
+Bubble column reactors (BCR) are widely used in chemical, electrochemical, biochemical, and metallurgical industries. The absence of moving parts, their low operating and maintenance costs and the excellent mass and heat transfer rates explain the large number of applications developed with this kind of reactor against the others (Deckwer, 1992; Kantarciet al., 2005). However, the design and scale-up of bubble columns are difficult because of the complexity of the gas and liquid flow patterns coupled with mass transfer and chemical reactions. Key factors such as gas hold-up, \(e\), \(s\), volumetric mass transfer coefficient, _k_k_a, specific interfacial area, \(a\), bubble size, \(r\)32, and kinetic rate constants, _k_a, are fundamental for the proper design and the operational control of gas-liquid reactors. There is an extensive list of works, both theoretically or experimentally, oriented to the determination of the physical parameters \(e\), _k_a, \(a\), and \(r\)32 as a function of operational conditions and physico-chemical properties of the gas-liquid system (see Kantarci et al., 2005 and references therein). Although the experimental conditions for each author differ significantly to each other, there is a relatively good agreement among them about which are the best experimental and numerical procedures used to determine these magnitudes. This agreement allows the definition of dimensionless expressions useful for the estimation of these parameters.
+
+Otherwise, no such kind of agreement can be found when kinetic data are analysed in gas-liquid systems. Let us consider for instance the CO2-alkanolamine system which is of great technological importance. Recently Vaidya and Kenig (2007) have reviewed the kinetic data present in the literature concerning this system observing significant discrepancies between the kinetic rate constants. Neither an agreement between different authors is found for example in the study of the oxidation of textile azo dyes with the ozone, a system with evident applications in environmental engineering (Liakou et al., 1997; Wu and Wang, 2001; Sevimli and Kinaci, 2002; Sevimli and Sarikaya, 2002; Choi and Wiesmann, 2004; Gokcen and Ozbegle, 2005). In a significant number of papers related with this problem, the kinetic rate constants are reduced to apparent pseudo-first order rate constants which are not useful for reactor design, scale-up or control purposes. Two reasons are mainly behind these discrepancies: (i) the differences in the conception of the experimental set-up; (ii) the difficulties to relate the data with the theoretical models. The aim of this work is to contribute to the modelling of unstationary semi-batch bubble column reactors for understanding the experimental observation of gas-liquid reacting systems and for kinetic constants determination.
+
+Different kinds of devices have been proposed to do kinetic studies in gas-liquid systems. Such devices have in common that their specific interfacial area is known, the flow pattern is nearly well defined and that mostly of them operates under stationary state conditions. Let describe in the following the conventional devices used for the characterization of kinetic gas-liquid systems with a brief description of their experimental advantages and disadvantages. The Stirred-Cell Reactors consist of two CSTR, one for the gas phase and the other for the liquid one, where the gas-liquid interface has a fixed geometrical value (Kucka et al., 2003). Measuring the gas pressure depletion in the upper chamber it is possible to determine the enhancement factors when the same depletion experiment is measured in absence of chemical reaction. The principal advantage of this device is that no chemical analysis of the liquid phase is necessary for the kinetic constant determination, but only apparent kinetic rate constants could be determined. However, the inherent difficulty of this method is to relate the enhancement factor, \(E\), with the absolute kinetic rate constant. While the \(E\) definition is done in steady state conditions and thus, is a constant, the experimental enhancement factor determined in the stirred-cell reactors change with the reaction extent. An additional problem related with the experimental determination of \(E\) is how it is related with kinetic constants when the gas-liquid mass-transfer process is coupled with a complex chemical kinetic mechanism. The Laminar Jet Absorber consists in a rodlike liquid jet flowing enclosed in an absorption chamber where the absorption rate is determined in steady state conditions measuring the gas flow absorbed by the liquid jet (Rinker et al., 1995; Aboudheir et al., 2004). Assuming the penetration theory, the gas-liquid contact time could be modified changing the jet length and the liquid flow rate and, consequently, the absorption rate. In absence of chemical reaction this device allows the determination of gas-to-liquid diffusion coefficients which is a key parameter for kinetic studies (Alghawas et al., 1989). Despite of the complexity of the instrument and that only it is suitable for fast kinetic regimes, the most significant advantage of the jet absorber is the possibility to work close to true steady state conditions. Finally, the Wetted-Wall Absorber is another apparatus where a thin liquid film flows over a cylindrical rod and the absorption rate is determined measuring the differences between the inlet and the outlet gas concentrations at the chamber. Again it is assumed stationary conditions which allow the calculation of the enhancement factors from a physical absorption experiment (Cullinane and Rochelle, 2006). As it was noted for stirred-cell reactor, in the wetted-wall absorber we have the same problems about the interpretation of the experimental enhancement factors in terms of kinetic rate constants or chemical kinetic mechanisms. Finally, the Bubble Column Reactors have been widely used to study the gas-liquid reactions although the kinetic rate constants derived from these works must be considered as apparent kinetic constants because of a lack of complete description of the physical parameters of these devices (_e_, \(a\), _k_k_a, \(r\)32, etc.). Considering for instance the reactions between the ozone and the azo dyes, these reactions are almost exclusively analysed using BCR's but only a relatively small number of studies provide absolute kinetic rate constants (Benbelkacem and Debellefontaine, 2003; Benbelkacem et al., 2003, 2004a; Lopez et al., 2004). This aspect is especially important when the ozonation is compared with other wastewater treatments such as advanced oxidation processes, that is, Fenton process, UV, UV/H2O2, UV/O3, O3/H2O2, etc.
+
+The second source of discrepancies between different experiments is related with the theoretical framework linking the experiments with the physicochemical model describing the gas-liquid system. The vast majority of the references given above, together with the references therein, base the determination of the kinetic rate constants on the experimental determination of the enhancement factor which in turn is related with the absolute rate constant using some gas-liquid transfer model, for example, two-film, penetration or surface renewal models. Additionally to these models, a kinetic scheme is coupled with the transport equations leading to different expressions for the enhancement factors in accordance to the particular boundary conditions assumed for each author. An important aspect in common with all these models for the calculation of the enhancement factors is that the boundary conditions of the partial differential equations defining the problem are assumed to be in steady state regime, in other words, they are constants and bounded in a semi-infinite or in a finite space (Danckwerts, 1970; Froment and Bischoff, 1990). These stationary conditions at the boundaries ensure the analytical solution of the resulting stationary diffusion equations which allow the calculation of mass fluxes at the boundaries, and hence, the enhancement factors. However, these stationary conditions at the boundaries cannot be guaranteed in the experiments, especially when these experiments are carried out in semibatch reactors which are intrinsically unstationary in nature.
+
+The purpose of this work is to describe a BCR operating in semi-batch conditions based on the unstationary film model, allowing the direct estimation of the absolute kinetic rate constants without the intermediate calculation of the enhancement factor. Experiments of ozonation of azo dye Acid Red 27 were carried out as a reference of gas-liquid reactive system in order to check the model. The paper has been structured as follows. In the first part, the mathematical model of the unstationary interface linked to the reactor model is described in some detail. Because of the structure of the model, it is necessary to carry out a sensitivity analysis of the parameters sorting them by significance. After that, the description of the bubble column reactor specifically designed to meet the restrictions of the model follows. Then, finally, the experimental results are shown followed with their discussion.
+
+## 2.3 Unstationary gas-liquid interface model
+
+### Model Description
+
+For modelling purposes of the gas-liquid mass transfer process, we have considered the Lewis and Whitman (1924) quiescent two-film model. This theory has the advantage, against the surface renewal or eddy-diffusion theories, that the time scale is the same for the microscopic description of the gas-liquid interface and for the macroscopic reactor description. The model we propose assumes the following assumptions: (i) the hydrodynamic flow of the gas and liquid bulk in the reactor has been considered as CSTR for both phases because the height-to-diameter ratio of our reactor is very close to one; (ii) the gas film resistance has been considered negligible under the operating conditions and, consequently, the molecular diffusion in the liquid film is the only resistance to mass transport across the interface; (iii) ideal gas and Henry laws are valid under the operating conditions; (iv) there is only two reacting species, \(A\) and \(B\), where \(A\) is the gas transferred to the liquid and \(B\) is a non-volatile substance dissolved in the liquid; (v) a global second order irreversible chemical reaction is considered together with unitary stoichiometric coefficients:
+
+\[A(g\to l)+B(l)\stackrel{{ k_{2}}}{{\longrightarrow}}\text{products}(l) \tag{1}\]
+
+For describing the mass-transfer phenomenon through the gas-liquid interface, a quiescent liquid film is considered separating the gas and liquid bulk. Inside this liquid film of length \(\delta\), the substances are spatially distributed as a result of the diffusion-reaction process. Assuming an unstationary diffusion-reaction equation to model the mass transport across the film with constant diffusion coefficients, the mass balances of "\(A\)" and "\(B\)" in the gas, film and bulk phases lead to the following coupled system of ordinary and partial differential equations:
+
+\[\frac{\text{d}y(t)}{\text{d}t} =\frac{Q(1-y_{\text{in}}(t))}{V}\frac{1-\varepsilon}{ \varepsilon}\left(\frac{y_{\text{in}}(t)}{1-y_{\text{in}}(t)}-\frac{y(t)}{1-y(t)}\right)\] \[\quad+\frac{\text{RT}}{P}\frac{1}{\varepsilon}D_{A}a\left(\frac{ \partial C_{A}(z,t)}{\partial z}\right)_{z=0} \tag{2}\]
+
+\[\frac{\partial C_{A}(z,t)}{\partial t} =D_{A}\frac{\partial^{2}C_{A}(z,t)}{\partial z^{2}}-k_{2}C_{A}(z,t)C_{B}(z,t)\quad\forall z\in[0,\delta] \tag{3}\] \[\frac{\partial C_{B}(z,t)}{\partial t} =D_{B}\frac{\partial^{2}C_{B}(z,t)}{\partial z^{2}}-k_{2}C_{A}(z,t)C_{B}(z,t)\quad\forall z\in[0,\delta] \tag{4}\]
+
+\[\frac{\text{d}C_{A}^{b}(t)}{\text{d}t} =-D_{A}a\left(\frac{\partial C_{A}(z,t)}{\partial z}\right)_{z=i} -k_{2}C_{A}^{b}(t)C_{B}^{b}(t) \tag{5}\] \[\frac{\text{d}C_{B}^{b}(t)}{\text{d}t} =-D_{B}a\left(\frac{\partial C_{B}(z,t)}{\partial z}\right)_{z=i} -k_{2}C_{A}^{b}(t)C_{B}^{b}(t) \tag{6}\]
+
+where \(y(t)\) stands for the molar fraction of \(A\) in the gas phase, \(C_{A}\) and \(C_{B}\) are the concentrations at the liquid film and, \(C_{A}^{b}\) and \(C_{B}^{b}\) are the concentrations at the liquid bulk. Because the gas-liquid interface is not accessible to measurement, the only known state variables of the system of Equations (2)-(6) are \(y(t)\), \(C_{A}^{b}\) and \(C_{B}^{b}\).
+
+Practically, in all bubble columns the existence of a gas chamber on the top of the column cannot be avoided, that is, the head space, which should be considered in the model to account for the dilution effect on gas molar fraction (Figure 1). This chamber has been modelled as a CSTR with its volume as a function of the gas hold-up. Assuming that \(V_{\text{R}}\) is the geometrical volume of the reactor and \(V\) the volume of liquid, the equation accounting the head space is given by:
+
+\[\frac{\text{d}y_{\text{out}}(t)}{\text{d}t} =\frac{Q}{V}\frac{1-\varepsilon}{(1-\varepsilon)V_{\text{R}}/V-1}(1-y_{\text{in}}(t))\] \[\quad\times\left(\frac{y(t)}{1-y(t)}-\frac{y_{\text{out}}(t)}{1-y_{\text{out}}(t)}\right) \tag{7}\]
+
+where \(y_{\text{out}}(t)\) is the molar fraction of \(A\) at the exit of the head space, \(y(t)\) is the solution of the Equation (2), \(y_{\text{in}}(t)\) is the input molar fraction of \(A\) at the bottom of the BCR and \(Q\) is the gas flow rate at the same point. Notice that \(y_{\text{in}}\) has been considered as a function of time. Because of the particular design of our bubble column reactor, it is convenient to express this concentration in this way. In fact, in the experimental section, it will be detailed the structure of the reactor but we mention here that at the bottom of our column there is a gas-mixing chamber leading to the following equation for \(y_{\text{in}}(t)\):
+
+\[y_{\text{in}}(t)=y_{0}\left(1-\exp\left(-\frac{Q}{V_{B}}t\right)\right) \tag{8}\]
+
+where \(y_{0}\) is the molar fraction of \(A\) entering the mixing chamber and \(V_{B}\) is its volume. For the derivation of this equation,
+
+Figure 1.: Diagram of the bubble column reactor for ozonation studies. The diameter of the reactor is 19 cm with a total volume of 10.32 L. The volume of the gas-mixing chamber is 2.55 L. The capillaries are 14 cm long with an inner diameter of 0.4 mm. They are disposed uniformly distributed at the bottom of the reactor with 2.1 cm of separation.
+
+no mass accumulation has been considered in the mixing chamber.
+
+All the initial conditions of the model described by Equations (2)-(8) are zero except for the variables \(C_{B}\) and \(C_{B}^{b}\) which are equal to the dye initial concentration \(C_{B0}\). For the Equations (2)-(6) the initial and boundary conditions are:
+
+\[y_{\rm in}(0)=y(0)=y_{\rm out}(0)=C_{A}(z,0)=C_{A}^{b}(0)=0 \tag{9a}\] \[C_{B}(z,0)=C_{B}^{b}(0)=C_{B0}\] (9b) \[C_{A}(0,t)=\frac{P}{H}y(t)\quad C_{A}(\delta,t)=C_{A}^{b}(t)\] (10a) \[\left(\frac{\partial C_{B}(z,t)}{\partial z}\right)_{z=0}=0\quad C _{B}(\delta,t)=C_{B}^{b}(t) \tag{10b}\]
+
+The conditions (10a) ensure, for the component \(A\), the equilibrium between the gas and the liquid phase at the interface and the continuity between the liquid film and the liquid bulk. In addition, the conditions (10b) express the non-volatility of '_B_' and the continuity between the liquid film and the liquid bulk.
+
+No rigorous analytical solution of the system (2)-(8) can be envisaged by conventional mathematical techniques because of the coupling of the non-linear ordinary and partial differential equations with the mixed boundary conditions (Equation 10b). The system has been simplified using the method of lines and a finite differences scheme of Equations (3) and (4) along the spatial coordinate, \(z\), to compute an approximated numerical solution of the concentrations inside the liquid film. The simplification is based on a second-order forward, backward, and central differences scheme for the flux terms in Equations (2)-(6). Finally, considering the input gas molar fraction, \(y_{\rm o}\), the saturation concentration of the transferred gas, \(C^{\ast}=y_{\rm P}P/H\) and the initial dye concentration, \(C_{B0}\), the system (2)-(10) in dimensionless form can be written as:
+
+\[x_{\rm o}=1-\exp\left(-\frac{t}{K_{1}}\right) \tag{11}\]
+
+\[\dot{x}_{\rm I}=K_{2}\frac{1-p_{1}}{p_{1}}(K_{3}-x_{\rm o})\left(\frac{x_{\rm o}}{K_{3}-x_{\rm o}}-\frac{x_{\rm{1}}}{K_{3}-x_{\rm{1}}}\right)\]
+
+\[-NK_{4}\frac{p_{2}}{p_{1}}(3x_{\rm I}-4x_{\rm{2}}+x_{\rm{3}}) \tag{12}\]
+
+\[\dot{x}_{\rm I}=N^{2}K_{6}\frac{p_{2}^{2}}{p_{1}^{2}}(x_{\rm{I}-1}-2x_{\rm{i}}+x_{\rm{i}+1})-p_{\rm{4}}x_{\rm{i}}(1-x_{\rm{i}+N})\quad{\rm{\it i}}=2,\ldots,N \tag{13}\]
+
+\[\dot{x}_{\rm{N}+1}=\frac{-N}{2}\frac{p_{2}}{1-p_{1}}(x_{N-1}-4x_{\rm{N}}+3x_{N+1})-p_{\rm{4}}x_{\rm{N}+1}(1-x_{\rm{2N}+1}) \tag{14}\]
+
+\[\dot{x}_{\rm{N}+2}=-\frac{2}{3}N^{2}K_{6}\frac{p_{2}^{2}}{p_{1}^{2}}p_{3}(x_{N+2}-x_{N+3})+p_{\rm{5}}x_{\rm{2}}(1-x_{\rm{N}+2}) \tag{15}\]
+
+\[\dot{x}_{\rm{N}+j}=N^{2}K_{6}\frac{p_{2}^{2}}{p_{1}^{2}}p_{3}(x_{N+j-1}-2x_{N+j}+x_{N+j+1})+p_{\rm{5}}x_{\rm{j}}(1-x_{\rm{4}+j})\]
+
+\[j=3,\ldots,N \tag{16}\]
+
+\[\dot{x}_{\rm{2N}+1}=\frac{-N}{2}\frac{p_{2}}{1-p_{1}}p_{3}(x_{\rm{2N}-1}-4x_{\rm{2N}}+3x_{\rm{2N}+1})\]
+
+\[+p_{\rm{5}}x_{\rm{N}+1}(1-x_{\rm{2N}+1})\]
+
+\[\dot{x}_{\rm{2N}+2}=K_{2}\frac{1-p_{1}}{(1-p_{1})K_{5}-1}(K_{3}-x_{\rm{0}}) \left(\frac{x_{\rm{1}}}{K_{3}-x_{\rm{1}}}-\frac{x_{\rm{2N}+2}}{K_{3}-x_{\rm{2N}+2}}\right) \tag{18}\]
+
+where the dot over the symbols stands for time derivative. The Equations (11) and (12) are the dimensionless ones equivalent to Equations (8) and (2) respectively. The Equation (13) is the discrete version of Equation (3) while the Equations (15) and (16) are the discrete version of Equation (4). The dimensionless bulk concentrations are given by Equations (14) and (17) and the Equation (18) is the dimensionless concentration of \(A\) at the reactor head. With this transformation, all the initial conditions of the above system of ODEs are equal to zero. The constants of the model are:
+
+\[K_{1}=\frac{V_{B}}{Q}\quad K_{2}=\frac{Q}{V}\quad K_{3}=\frac{1}{y_{\rm{0}}}\]
+
+\[K_{4}=\frac{RT}{2H}\quad K_{5}=\frac{V_{R}}{V}\quad K_{6}=\frac{r_{\rm{22}}^{2}}{9D_{A}} \tag{19}\]
+
+and the parameters are given by:
+
+\[p_{1}=\varepsilon\quad p_{2}=k_{\rm{L}}a\quad p_{3}=\xi=\frac{D_{B}}{D_{A}} \quad p_{4}=k_{\rm{2}}C_{B0}\quad p_{5}=k_{\rm{2}}C^{\ast} \tag{20}\]
+
+### Model Analysis
+
+In a precedent work, the structural identifiability analysis was applied to identify the accessible parameters of the unsteady-state gas-liquid interface model (Navarro-Laboulais et al., 2006, 2008). This analysis is pertinent in such models where the number of state variables (\(2N+2\), in our case) is greater than the number of observable variables (\(x_{N+1}\), \(x_{2N+1}\), and \(x_{2N+2}\)). The analysis of the model described in those references, which is similar to Equations (2)-(6), had shown seven identifiable parameters:
+
+\[p_{1}^{\prime}=\varepsilon;p_{2}^{\prime}=k_{\rm{L}}a;p_{3}^{\prime}=k_{\rm{R}}a;p_{4}^{\prime}=k_{\rm{2}}C_{B0}\]
+
+\[p_{5}^{\prime}=k_{\rm{2}}C^{\ast};p_{6}^{\prime}=\frac{D_{A}}{\delta^{2}};p_{7}^{\prime}=\frac{D_{B}}{\delta^{2}} \tag{21}\]
+
+This group of parameters is the minimum number of parameters needed to characterise the reacting gas-liquid systems which is higher than the parameters considered in classical theories. We remark here that the mass transfer coefficient, \(k_{\rm{L}}\), is defined as the ratio between the diffusion coefficient of \(A\) and the size of the liquid film, that is, \(k_{\rm{L}}=D_{A}/\delta\). This definition remains unchanged even when there are chemical reactions in the film. Following this idea, it has also been defined a new mass transfer coefficient based on the diffusion of the dissolved reactant \(B\), that is, \(K_{B}=D_{B}/\delta\). Additionally, the structural analysis of the model also results in the definition of two time constants, \(p_{6}\) and \(p_{7}\), which are typical in mass transport phenomena related to bounded diffusion problems (Crank, 1999). Another significant result of this model is that the Hatta modulus is inherent to the model given by Navarro-Laboulais et al. (2006):
+
+\[Ha_{2}^{2}=\frac{k_{\rm{2}}D_{A}C_{B0}}{k_{\rm{L}}^{2}}=\frac{p_{4}^{\prime}}{p_{6}} \tag{22}\]
+
+This initial model could be reduced to the five parameters model (Equations 11-18) if (i) the diffusion coefficient of one of the species \(A\) or \(B\) is known and, (ii) the size of the bubbles in the BCRis known. Under these circumstances the parameter definition (20) holds and the Hatta modulus can be written as:
+
+\[H\text{a}_{2}^{2} = \frac{\text{9}D_{A}}{r_{32}^{2}}\left(\frac{p_{1}}{p_{2}}\right)^{2}p_{4}\]
+
+Notice that the definition of Ha considered here is stationary. It is not defined from the enhancement factor, \(E\), which classically is defined as a flux ratio (Danckwerts, 1970). If we consider an unstationary experiment, we expect that \(E\) should change with time, and thus, Ha too. However, in our model, Ha is defined implicitly and does not change with time. In our case the enhancement factor is calculated always _a posteriori_ to the kinetic rate constant determination and never is used as a derived experimental magnitude.
+
+Finally, even though the model could be reduced to the five parameters given by Equation (20), not all of them have the same effect on the observable state variables. The original set of partial and ordinary differential Equations (2)-(8) has been transformed into a set of ODE. This system is used further in the numerical algorithm for parameter determination. The same algorithm allows the sensitivity analysis of the model, that is, to analyse the effect of an infinitesimal change in the parameters on the observable state variables. This kind of mathematical analysis is local in the sense that depends on the values of the parameters around which the sensitivity of the model is analysed. Additionally, this analysis allows a better experimental design reducing or eliminating the unimportant data and determining the effect of the parameter variation on the system behaviour (Ionescu-Bujor and Cacuci, 2004). The first order relative sensitivity coefficients are defined as the partial derivative of the state variables with respect to each parameter of the model as Englezos and Kalogerakis (2001):
+
+\[S_{i,j}(t) = p_{j}\left(\frac{\partial x_{i}(t)}{\partial p_{j}}\right)_{\text{p}_{0}}\]
+
+This function is evaluated solving its differential equation simultaneously with the ODE system (11)-(18). Because the sensitivities of the model are evaluated locally near some particular point **p**0 located in the parameters space, the results are different depending on the kinetic regime under analysis. In Figure 2 the sensitivities of gas-phase and the dye concentrations against the parameters (20) are plotted for slow and fast kinetic regimes (Beltran, 2007). In Table 1 the principal figures of the sensitivities are collected. The higher the value in the table, the better the observable variable for parameter determination. From Table 1 we conclude that in the slow and fast kinetic regimes the gas hold-up, \(e\), and the diffusion coefficients ratio, \(x\), will never be accessible by dynamic measurements whichever the observable variables are selected to fit the model because their sensitivities are much lower than the other parameters. Comparing the values of the second and third columns of Table 1 for slow kinetic regimes, the differences between the relative sensitivity of bulk concentrations, \(C_{A}^{\text{b}}\) and \(C_{B}^{\text{b}}\), are not too. Also, it seems slightly more appropriate to consider \(C_{A}^{\text{b}}\) (the ozone in the solution) for the determination of kinetic rate constants instead of \(C_{B}^{\text{b}}\). However, in the fast kinetic regime the relative sensitivities point to \(C_{B}^{\text{b}}\) as the best candidate to be used for the kinetic rate constant determination. In particular, the volumetric mass transfer coefficient and the kinetic parameters \(p_{4}\) and \(p_{5}\) could be evaluated with increasing accuracy using the states variables \(x_{2N+2}\) (the ozone gas
+
+Figure 2.: Sensitivity analysis of the mathematical model under (a and b) slow kinetic conditions and (c and d) fast kinetic conditions. The simulation values are close to the experimental operating conditions. For all simulations \(P_{\text{atm}} = 960\text{ mbar}\), \(t = 21^{\circ}\text{C}\), \([\text{O}_{3}]_{9} = 57.6\text{ g m}^{-3}\), \(\Delta P_{2} = 22\text{ mbar}\), \(k_{\text{d}} = 0.005\text{ s}^{-1}\), \(\xi = 1\). \(C_{80} = 0.03\text{ mM}\), \(Q = 2\text{ N}\text{L}\text{ min}^{-1}\). (a and b) \(k_{2} = 1M^{-1}\text{ s}^{-1}\), resulting in Ha\({}_{2} = 0.0042\). (c and d) \(k_{2} = 10^{6}\text{ M}^{-1}\text{ s}^{-1}\), resulting in Ha\({}_{2} = 4.25\). (a and c) relative sensitivity \(p_{i}(\partial x_{2N+2}/\partial p_{i})\) of the dimensionless ozone gas concentration, \(x_{2N+2}\) (see Equation 18), for the five parameters (20); for \(e\) and _ξ_ξ_ the parameters are not sensitive to this observable state variable. (b and d) Relative sensitivity \(p_{i}(\partial x_{2N+1}/\partial p_{i})\) of the dimensionless dye concentration, \(x_{2N+1}\) (see Equation 17); again _ξ_ and _ξ_ are not sensitive. Notice the change of time scale for fast chemical regime.
+
+concentration, _y_A), _xN_+1 (the dissolved ozone concentration, _C_h) and _xN_+1 (the dye concentration, _C_h) in this order.
+
+In conclusion, the unstationary gas-liquid interface model (2)-(6) together with the gas chambers Equations (7) and (8) are reduced to a ODE system (11)-(18) characterised by the five parameters (20). After the sensitivity analysis of the model we conclude also that only the volumetric mass transfer coefficient, _k_L_a, and the kinetic parameters \(p\)4 and \(p\)5 would be determined by fitting the experimental data to this model directly. The gas hold-up, \(e\), and the diffusion coefficients ratio, \(e\), should be determined by independent experiments.
+
+## EXPERIMENTAL APPARATUS AND PROCEDURE
+
+The reactor designed for the study of the ozonation processes is shown on Figure 1. It consists in three parts: (i) the gas mixing chamber; (ii) the gas distributor; (iii) the reactor body comprising the upper gas chamber. The gas mixing chamber has a volume of 2.55 L and its function is to equalise the gas composition and the pressure at the entrance of each capillary. This chamber avoids the nonuniformities in flow and composition during the ozone injection. The experiments consist in a sudden injection of ozone in the reacting media without changing the total gas flow rate in the reactor. The gas distributor is formed by 61 capillaries uniformly distributed on the reactor surface (see inset in Figure 1) separated 2.1 cm between them. The capillaries are 14 cm length with an inner diameter of 0.4 mm. With this configuration and the operating conditions applied, no bubble coalescence has been observed. The study of bubble formation when some gas flows through a submerged rigid orifice has been carried out both experimental (Jamialahmadi et al., 2001) and theoretically (Gerlach et al., 2007; Das and Das, 2009). The generalised expression given in those works for the bubble volume calculation has been used here to estimate the Sauter's bubble radius needed for the mathematical unstationary model. Following the authors aforementioned:
+
+\[\frac{v_{B}}{R_{0}^{3}} = \frac{4\pi}{3}\left[ \frac{1.119}{\text{Bo}^{1.08}} + 1.406\frac{\text{Fr}^{0.36}}{\text{Ga}^{0.39}} + 0.469\text{Fr}^{0.51} \right\rbrack\]
+
+where _v_B is the bubble volume, \(R\)0 the orifice radius, and Bo, Fr, and Ga, are the Bond, Froude, and Galileo dimensionless numbers respectively. These numbers are function of water properties as density, viscosity, and surface tension (IAPWS, 1994, 2008; Tanaka et al., 2001).
+
+The total geometrical volume of the reactor, _V_B, is 10.32 L. The working solution volume, \(V\), has been fixed to 9.0 L for each experiment in order to prevent the reactor overflow by water expansion.
+
+A scheme of the instrumentation fitted to the ozonation reactor is shown in Figure 3. The ozone generator used in the experiments is an Anseros COM-AD-04 where the ozone production is changed modifying the pulse frequency in the discharge lamps. The ozonator is feed with pure oxygen (Carburos Metallicos) at constant pressure of 0.9 bar. Under these circumstances, the maximum ozone concentration measured was around 80 g Nm-3 for an oxygen flow rate of 2.0 L min-1.
+
+An analogue mass-flow meter (M+W Instrumentation GmbH, model D6210; see AMFM N2 in Figure 3) was used to measure the nitrogen flow rate used to degas the reactor. This flow was fixed by hand with a manual regulation valve. The gas mixture used in the reactions was set using three digital mass-flow controllers (Bronkhorst, Mod. EL-Flow F201CV; see DMPC O3+O2, O2, Air in Figure 3) and the mixture was measured with a digital mass-flow meter (Bronkhorst, Mod. EL-Flow F11B; see DMFM Gas Mixture in Figure 3). These digital instruments were connected through a RS485 bus, and the communication with them is done using a Dynamic Data Exchange (DDE) server. Additionally, the RS485 is connected to the computer via RS485-to-USB converter (National Instruments, NI USB-485) giving the maximum connectivity of the instruments with the minimum computer resources.
+
+The gas mixture is then sent to the reactor gas-mixing chamber or to the venting point of the system using several two-port solenoid valves (SMC, VDW Series; see V1-V6 in Figure 3). These valves are controlled with the digital port of the data acquisition card using a self-made interface which converts the state of each digital gate to a 0-24 V signal which commands the solenoid valve. With the combination of the valves shown in Figure 3 it is possible to measure the ozone gas not only at the output of the reactor, but also at the input.
+
+The ozone has been measured in the gas and in the liquid phases. For the liquid phase an electrochemical electrode sensor (ATI, Q45H/64) has been used. Instead to measure the dissolved ozone directly inside the reactor, this magnitude has been measured using a low-volume flow cell set in a closed loop where the solution is recycled with a peristaltic pump. The analyser allows the simultaneous measure of the dissolved ozone and the solution temperature. On the other hand, the ozone in the gas phase has been measured using an UV-absorption O3-meter (Anseros, Ozomat GMRTI) which is able to measure up to 200 g m-3. It is important to consider here the flow rate and the pressure at which the measure is done. The gas flow through the measuring cell in the O3 meter must be constant because if this value changes, different delay times should be applied in the kinetic curves. Then, an AMFC (Aalborg, Mod. GFC17) has been linked to the Ozomat meter, fixing the flow rate to 0.3 L min-1. The ozone-gas concentration measure depends on the pressure and the constancy of this property cannot be ensured in all the installation. In order to avoid this problem, instead of the measurement of the ozone in the units given by the instrument, that is, g m-3, it is better to change the units to molar fraction. This unit conversion has the advantage that the ozone molar fraction is insensitive to pressure changes along the reactor. Additionally, the gas pressure is measured in other two critical points in the system, the pressure at the reactor head (see P2 in Figure 3) and in the gas mixing chamber (P1 in Figure 3). All the pressure transmitters are from Druck
+
+\begin{table}
+\begin{tabular}{c c c c}  & _YA_ & _C_h & _C_h \\ Slow kinetic regime & & & \\ \(e\) & 1.388E–03 & 3.115E–03 & 1.137E–03 \\ _k_α & 0.1141 & 0.1972 & 0.2397 \\ _ξ_ & 6.747E–11 & 1.718E–09 & 7.228E–11 \\ _ρ_4 & 1.923E–02 & 4.347E–02 & 5.129E–02 \\ \(p\)5 & 1.941E–02 & 3.797E–02 & 5.197E–02 \\ Fast kinetic regime & & & \\ _ε_ & 3.777E–03 & **6.977E–03** & 3.655E–04 \\ _κ_a_ & 0.1179 & **0.5264** & 0.2361 \\ _ξ_ & 1.340E–02 & **9.138E–02** & 7.950E–03 \\ _ρ_4 & 2.564E–02 & **0.7897** & 7.192E–02 \\ _ρ_5 & 2.561E–02 & **0.7967** & 7.252E–02 \\ \end{tabular}
+\end{table}
+Table 1.: Maximum absolute values of the relative sensitivity curves plotted on Figure 2Limited (Mod PTX 1400) and give a 4-20 mA output proportional to pressure gauge.
+
+The ozonation reactor was fitted to a UV-Vis spectrophotometer to measure the abatement of the reacting substances in the reactor. An Unicam Helios-Gamma spectrophotometer fitted with a Hellma Ultra-Mini Immersion Probe (Mod 661.622) has been used in the experiments. The spectrophotometer data is acquired with the serial RS232 port of the computer.
+
+Finally, all the analogue inputs and outputs of the instruments are linked to a computer using a data acquisition card (Advantech, PCI1710-HG). The card has 8 analogue inputs configured in differential mode and 16 digital inputs/outputs which are used to control the solenoid valves states. The software controlling all the process was developed using LabView 8.20 (National Instruments) which manages properly all the devices linked to the computer.
+
+The dye used in the kinetic experiments was the Acid Red 27 (Amaranth, CAS 915-67-3, \(\text{Mr} = 604.74\,\text{g}\,\text{mol}^{- 1}\)) provided by Sigma-Aldrich (ref A1016). The absorbance measured at 520 nm has been used to follow this substance. All the solutions were set to pH 2 with HCl04 (Panreac, PA, ref. 132175) and the ozone radical reactions were blocked adding 0.01 M of tert-Butanol (2-Methyl-2-Propanol, Panreac PS, ref. 161903).
+
+## RESULTS AND DISCUSSION
+
+In the precedent sections it has been demonstrated, through the sensitivity analysis, that the gas hold-up cannot be evaluated from kinetic data although there is no theoretical limitation to do this. The model is not sensitive to this quantity and then, it must be measured with another complementary technique because \(\varepsilon\) is still needed for kinetic calculations. The gas hold-up has been estimated independently by a manometric method. The results are plotted on Figure 4 showing a good agreement between the manometric data and the equation proposed by Wilkinson et al. (1992). The gas hold-up is linear with the gas flow rate and thus, with the superficial gas velocity too. This behaviour is characteristic of the bubbly homogeneous regime (Shaikh and Al Dahhan, 2007). The usual flow rate for kinetic experiments was fixed to 2 L min-1 giving a mean bubble diameter of 3.5 mm after Equation (25). Under these conditions we ensure the no coalescence of the bubbles that can change the predicted interfacial area in the reactor given by:
+
+\[a = \frac{3\varepsilon}{r_{32}}\]
+
+where \(r\)32 is the Sauter's mean radius of the bubble calculated after Equation (25). Considering a pressure of 1100 mbar at the reactor, the interfacial area at 2 L min-1 is near 8.0 m-1.
+
+Figure 4.: Gas hold-up for the reactor determined by the manometric method. The linear relation confirms the homogeneous flow regime in the reactor. The line is the prediction of the gas hold-up in the operating conditions, according to the equation proposed by Wilkinson et al. (1992).
+
+Figure 3.: Scheme of the experimental set-up. The digital mass-flow controllers and meter (DMFC and DMFM) are connected to a computer using a R5485 protocol. The analogue mass-flow controller and meter (AMFC and AMFM) are connected to the analogue inputs or outputs of the data acquisition card. The solenoid valves (V1–V6) are controlled by the digital input/output of the acquisition card. The pressures are measured at the gas mixing chamber (P1), the reactor head (P2), and at the ozone-meter line (P3).
+
+Only the data corresponding to the dye in the solution and the ozone in the gas outlet have been fitted to the model for the determination of the second order kinetic rate constant, \(k\)2. Because an analytical solution of this system is not available, a numerical method based on the Gauss Newton algorithm has been used instead. This is a gradient based algorithm which can be implemented in optimization problems involving systems of ordinary differential equations (Englezos and Kalogerakis, 2001) and it has been previously applied for gas-liquid models (Navarro-Laboulais et al., 2008). The method consists in the minimisation of the objective function defined as the sum of the square of residuals assuming that the model could be linearised around some parameter vector. The details of this algorithm is beyond the scope of this paper and we refer the reader to the book of Englezos and Kalogerakis for a complete description of the mathematics and the programming of the method.
+
+In Unstationary Gas-Liquid Interface Model Section, it has been demonstrated that the unstationary film model is reduced to five parameters with only three of them sensitive to the observable state variables, that is, \(p\)2, \(p\)4, and \(p\)5 (see Equation 20). Because \(p\)4 and \(p\)5 are related with the kinetic rate constant, two series of experiments were designed in order to check the viability of the model. The first one consisted in maintaining the initial dye concentration constant while the initial ozone gas concentration entering the reactor was changed along the series. In this situation, \(p\)4 = \(k\)2 x C00 remains constant and \(p\)5 = \(k\)2 x C+ changes in each experiment. In the second series, the inlet ozone concentration remains constant while the initial dye concentration was changed. The results of the first and the second series of experiments are shown on Figures 5 and 6 respectively.
+
+In addition, the parameters \(p\)2, \(p\)4, and \(p\)5 could be evaluated simultaneously but the shape of the sensitivities (see Figure 2) together with the obvious proportionality between \(p\)4 and \(p\)5 makes the parameters strongly correlated. In order to uncouple these correlations an additional restriction to the minimisation of the objective function has been considered. Because the ratio \(p\)4/_p_5 only depends on the experimental conditions as a proportionality constant between the ratio of dye and ozone initial concentrations, the minimum of the objective function must be sought along the line \(p\)4/_p_5 = \(C\)20/_C+.
+
+Figure 5a shows the fit of the dimensionless dye concentration and Figure 5b the dimensionless ozone gas molar fraction in the first series of experiments. From the graphs we confirm that the gas phase measurements do not show any feature (singular or characteristics points) which could be used on kinetic rate constant determination. This experimental evidence seems to contradict previous analytical results which suggested that the measurement of the gas phase could be considered alone for kinetic rate constants determination (Navarro-Laboulais et al., 2006, 2008). The dilution and integration effect of the reactor head chamber can be in the origin of this information loss. In fact, when exclusively the gas phase data are considered for fitting the model, the algorithm does not converge to any physical plausible solution. The use of dye concentration alone or in combination with ozone gas data, leads to a convergence of the algorithm to a minimum which is faster in this second case than in the first one. Figure 6a and b shows respectively the response of the system in the second series of experiments, where the input ozone concentration remains constant and the initial dye concentration is varied. As in the previous series, the ozone gas evolution does not give relevant kinetic information but gives stability to the optimization process.
+
+All the results for the fits shown on Figures 5 and 6 are collected on Table 2. The volumetric mass transfer coefficient is almost constant for all the experiments with independence of the initial conditions even considering the different conditions of the chemical reaction. This is because the mass transfer coefficient defined in this work as _k_L = \(D\)4/_d_ does not depend on flux measurements which can introduce errors in its determination due to the lack of knowledge about the instantaneous enhancement factor. Under these circumstances, considering a confidence level of 95%, the mean volumetric mass transfer coefficient measured with the kinetic experiments is:
+
+\[k_{\text{L}}a = \left( 0.0028 \pm 0.0002 \right)\text{s}^{- 1}\]
+
+In order to check this value and because of the problems related with the dissolved ozone electrode calibration and its membrane characterization (Monzo et al., 2008), a series of oxygen absorption experiments were performed at different gas flow rates. The experiment consisted in degassing the solution with N2 and following the O2 concentration with time in an air absorption experiment. The _k_L = \(d\)1/_d_2 was calculated considering the model (11)-(18) without chemical reaction and fitting only the liquid phase. The results are shown on Figure 7. The volumetric mass transfer coefficient was evaluated at different gas flow rates showing a linear relation. For the kinetic experiments the gas flow rate is fixed to 2 L min-1 and considering an error confidence level of
+
+Figure 5.: Results for Series #1. Evolution of Acid Red 27 (a) and ozone gas concentration (b) for different ozone concentration at the input. The values are shown in the legend. \(C\)80 = 3 × 10−5 M. The lines are the best fit to the ODE system (11)–(18).
+
+95%, the oxygen mass transfer coefficient is:
+
+\[k_{\text{L}}a|_{\text{O}_{2}}=(0.0042\pm 0.0002)\,\text{s}^{-1}\]
+
+Using this value to check the previously obtained with the ozone kinetic experiments, we need to know the diffusion coefficients of the oxygen and ozone in water together with some hypothesis about the structure of the liquid film, that is, the surface renewal theory or the liquid film theory:
+
+\[\frac{k_{\text{L}}a}{k_{\text{L}}a|_{\text{O}_{2}}}=\left(\frac{D_{\text{O}_{2}}}{D_{\text{O}_{2}}}\right)^{d} \tag{27}\]
+
+where \(d\) is 1/2 for the surface renewal and 1 for the liquid film model. Considering the diffusion coefficient for the oxygen (Ferrell and Himmelblau, 1967) and for the ozone (Winkelmann, 2007), and considering a mean temperature in all the experiments of 21.2+-0.5degC, the ratio (27) is 0.7978 for the surface renewal theory and 0.6367 for the liquid film. Consequently, the resulting volumetric mass transfer coefficient for the ozone is 0.00335 s-1 for the surface renewal model and 0.00267 s-1 for the film model. Considering that all the calculations are based on the microscopic unstationary film model, this last result is more coherent with the definition of the mass transfer coefficient in the model.
+
+Once we have the mean mass transfer coefficient of the system, the data are again fitted to the model (11)-(18) but in this run considering this mean value for _k_L_a, 0.00267 s-1. Then, the problem is again simplified because the value of _k_L_a is fixed and considering that \(p\)4 and \(p\)3 are bonded by \(p\)4/_p_5 = \(C\)20/_C_+ only one parameter is fitted at one time. With the first experiment series
+
+\begin{table}
+\begin{tabular}{c c c c c c c c c} \(t\) (°C) & _P_R (mbar) & [O3]3 (g/m3) & _C_* (mg/L) & _C_* (μM) & \(C\)00 (μM) & \(C\)00/C* & _k_L_a (s−1) & \(p\)5 = \(k\)2 C* (s−1) \\ \hline Series \#1 & & & & & & & & & \\
+21 & 1094.92 & 57.6 & 13.34 & 277.9 & 30 & 0.1079 & 0.00298 ± 0.00007 & 0.76 ± 0.04 \\
+21.7 & 1087.59 & 40.1 & 8.99 & 187.2 & 30 & 0.1603 & 0.00263 ± 0.00007 & 0.59 ± 0.05 \\
+21.8 & 1096.59 & 32.2 & 7.18 & 149.6 & 30 & 0.2005 & 0.00263 ± 0.00006 & 0.65 ± 0.06 \\
+20.2 & 1094.25 & 25.6 & 6.16 & 128.3 & 30 & 0.2338 & 0.00300 ± 0.00006 & 0.21 ± 0.01 \\
+20.9 & 1093.25 & 12.8 & 2.98 & 62.1 & 30 & 0.4834 & 0.00282 ± 0.00007 & 0.16 ± 0.02 \\ \end{tabular}
+\end{table}
+Table 2: Experimental conditions for series 1 and 2
+
+Figure 6: Results for Series #2. Evolution of Acid Red 27 (a) and ozone gas concentration (b) for different initial dye concentration. The values are shown in the legend. [O3]3 ~ 14 g m−3. The lines are the best fit to the ODE system (11)–(18).
+
+Figure 7: Volumetric mass transfer coefficient \(k_{\text{L}}a|_{\text{O}_{2}}\) measured from oxygen absorption experiments at different air flow rates. The line is the best fit to data. The discontinuous lines are the confidence bands for a 95% confidence level.
+
+the new value of \(p_{5}\) is recalculated while the second series is used for recalculating \(p_{4}\). Plotting \(p_{5}\) and \(p_{4}\) against the ozone saturation concentration and the initial dye concentration respectively, by definition we must have a line which slope is the kinetic rate constant. These plots are shown on Figure 8a and b respectively. The kinetic constant for the direct reaction between the ozone and the Acid Red 27 is \(2933\pm 351\) M\({}^{-1}\) s\({}^{-1}\) from the first experimental series and \(3723\pm 127\) M\({}^{-1}\) s\({}^{-1}\) from the second series.
+
+The origin of the discrepancy between these two values can be understood if we know how the systematic errors are propagated through the experiments and the model. In the first series, the inlet ozone gas concentration is modified resulting in a change in the saturation ozone concentration in the solution. The numerical value of this quantity depends on the accuracy of the ozone gas concentration measurement or on its direct measurement in the solution. Both alternatives are not free of instrumental problems resulting in a higher inaccuracy of this quantity. In fact, considering the Figure 8b, because there is not a significant variation in the ozone for these series, the data dispersion is lower than the shown in Figure 8a.
+
+Finally, the kinetic constant determined by the methodology proposed in this paper is several orders of magnitude under the values reported in the bibliography. Most papers devoted to ozonation of organic dyes give just the apparent first order kinetic rate constants which are only valid for comparison purposes between different experiments when the geometry and operational conditions of the chemical reactor do not change very much (Sevimli and Sarikaya, 2002; Gokcen and Ozbelge, 2005; Wu et al., 2005). In a recent work, Tokumura et al. (2009) give a value of the second order kinetic rate constant for the reaction of the ozone with the Orange II (an azo compound) which is between 183 and 216 L/(g s), but the units used to express this second order constant do not make easy the comparison with the value we have obtained here because the authors express the concentration of the dye in TOC units (Tokumura et al., 2009). A special mention should be done to the work of Lopez et al. (2004) where they estimate the second order kinetic rate constant for the reaction of the ozone with the Acid Red 27 between \(2\times 10^{7}\) and \(7\times 10^{7}\) M\({}^{-1}\) s\({}^{-1}\). The mathematical model developed by the authors in a series of independent papers (Benbelkacem and Debellefontaine, 2003; Benbelkacem et al., 2003, 2004a,b) is similar to the unstationary film model shown in our work but there are significant differences in the data analysis and parameter estimation procedure. These authors base their calculation on the estimation of the enhancement factor and Hatta modulus along the reaction course. As it was stated in the introduction, the use of these magnitudes implies the existence of a steady state regime at the interfacial level which cannot be always guaranteed theoretically nor experimentally. An evidence of this last statement is that the instantaneous slope of the Hatta modulus evaluated experimentally at the different reaction stages is not null which contradicts its definition done under steady state conditions. Additionally, both models differ on the reactor definition. While in the present work the corresponding hydrodynamic equations of the reactor are coupled with the unstationary film model being able to distinguish both contributions to the observable variables, the model considered by Lopez et al. (2004) considers that the macroscopic description of the reactor does not contribute significantly to the system response.
+
+## Conclusions
+
+The experimental response of the bubble column reactor designed in this work for gas-liquid studies shows a good agreement with the unstationary film model. The model incorporates a macroscopic description of the reactor together with the microscopic description of the gas-liquid mass transfer coupled with chemical reactions. The theoretical analysis of the model gives that there are five parameters characterising the response of the system, that is, the gas hold-up, the ratio of diffusivities of the reacting species, the volumetric mass transfer coefficient and two time constants related with the kinetic rate constant. However, the sensitivity analysis of the model considering a semi-batch experiment reduces the number of accessible parameters to three because the model is not sensitive to gas hold-up and the diffusivity ratio.
+
+The model has been tested with the reaction between the ozone and the azo-compound Acid Red 27. The experimental data fitted quite well the model to calculate the mass transfer coefficient and the kinetic rate constant simultaneously. The mass transfer coefficient derived from kinetic experiments matches the one derived from oxygen absorption experiments considering the quiescent film theory for the interface description. The rate constant for the direct reaction of the ozone with the Acid Red 27 is estimated in \(k_{\mathrm{2}}=3723\pm 127\) M\({}^{-1}\) s\({}^{-1}\) at \(21.2\pm 0.5\)\({}^{\circ}\)C. Additional experiments with a deeper analysis of the model and data will be needed to explain the discrepancy between the value of this kinetic rate constant and those reported in the bibliography. The self-coherence of the model, the absence of hypothesis about the state of the film together with the applied optimization procedure, allow to consider the method proposed in this paper as a viable alternative for the study of gas-liquid systems in semi-batch bubble column reactors.
+
+Figure 8.: Kinetic constant determination from parameter \(p_{5}\) (a) calculated from data shown on Figure 5, and from parameter \(p_{4}\) (b) from data of Figure 6. The dotted and continuous lines are the confidence bands of the fit for 95% and 99% of confidence level.
+
+### Nomenclature
+
+* _e_)S/_V_ = interfacial specific area (m-1)
+* _b_B_0 _r__b_R_0 _s/_a_ = Bond number
+* \(C\) _molar concentration in the liquid phase (mol m-3)
+* _c_' \(y\)0_r_/_H_ = saturation concentration in the liquid bulk (mol m-3)
+* \(D\) diffusion coefficient in the liquid (m2 s-1)
+* \(E\) enhancement factor
+* _r_r_d/R_s_d_ = Froude number
+* _a_r__b_R_0 _s/_a_ = Galileo number
+* \(g\) gravity constant
+* \(H\) Henry constant (atm L mol-1)
+* _Ha_ Hatta modulus
+* \(K_{i}\) th constant of the model; see Equations (19) for definition _k__k__i_a_ volumetric mass transfer coefficient in the liquid side (s-1)
+* _k__L_0 _D_A/_d_ = mass transfer coefficient in the liquid side for species \(A\) (m s-1)
+* _k__B_0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 \(B\)0 _0 \(B\)0 \(B\)0 _0 \(B\)0 \(B\)0 _0 \(B\)0 \(B\)0Ferrell, R. T. and D. M. Himmelblau, "Diffusion Coefficients of Nitrogen and Oxygen in Water," J. Chem. Eng. Data **12**(1), 111-115 (1967).
+* [33]Fronent, G. F. and K. Bischoff, "Chemical Reactor Analysis and Design," 2nd ed., John Wiley & Sons, New York (1990).
+* [34]Gerlach, D., N. Alleborn, V. Buwa and E. Durst, "Numerical Simulation of Periodic Bubble Formation at a Submerged Orifice With Constant Gas Flow Rate," Chem. Eng. Sci. **62**(7), 2109-2125 (2007).
+* [35]Gokcen, F. and T. A. Ozbelge, "Enhancement of Biodegradability by Continuous Ozonation in Acid Red-151 Solutions and Kinetic Modeling," Chem. Eng. J. **114**(1-3), 99-104 (2005).
+* [36]IAPWS, "IAPWS Release on Surface Tension of Ordinary Water Substance," (1994).
+* [37]IAPWS, "Release on the IAPWS Formulation 2008 for the Viscosity of Ordinary Water Substance," (2008).
+* [38]Ionescu-Bujor, M. and D. G. Cacuci, "A Comparative Review of Sensitivity and Uncertainty Analysis of Large-Scale Systems--I: Deterministic Methods," Nuc. Sci. Eng. **147**(3), 189-203 (2004).
+* [39]Jamialahmadi, M., M. R. Zehtaban, H. Muller-Steinhagen, A. Sarrafi and J. M. Smith, "Study of Bubble Formation Under Constant Flow Conditions," Chem. Eng. Res. Des. **79**(A5), 523-532 (2001).
+* [40]Kantarci, N., F. Borak and K. O. Ulgen, "Bubble Column Reactors," Proc. Biochem. **40**(7), 2263-2283 (2005).
+* [41]Kucka, L., J. Richter, E. Y. Kenig and A. Gorak, "Determination of Gas-Liquid Reaction Kinetics With a Stirred Cell Reactor," Sep. Purif. Technol. **31**(2), 163-175 (2003).
+* [42]Lewis, W. K. and W. G. Whitman, "Principles of Gas Absorption," J. Ind. Eng. Chem. **16**, 1215-1220 (1924).
+* [43]Liakou, S., S. Pavlou and G. Lyberatos, "Ozonation of Azo Dyes," Water Sci. Technol. **35**(4), 279-286 (1997).
+* [44]Lopez, A., H. Benbelkacem, J. S. Pic and H. Debellefontaine, "Oxidation Pathways for Ozonation of Azo Dyes in a Semi-Batch Reactor: A Kinetic Parameters Approach," Environ. Technol. **25**(3), 311-321 (2004).
+* [45]Navarro-Laboulais, J., S. C. Cardona, J. I. Torregrosa, A. Abad and F. Lopez, "Structural Identifiability Analysis of the Dynamic Gas-Liquid Film Model," AIChE J. **52**(8), 2851-2863 (2006).
+* [46]Navarro-Laboulais, J., S. C. Cardona, J. I. Torregrosa, A. Abad and F. Lopez, "Practical Identifiability Analysis in Dynamic Gas-Liquid Reactors Optimal Experimental Design for Mass-Transfer Parameters Determination," Comp. Chem. Eng. **32**(10), 2382-2394 (2008).
+* [47]Monzo, P., S.C. Cardona, F. Lopez, A. Abad, J. Navarro-Laboulais, "Deconvolution of the Dynamic Response of Oxygen Electrodes. Systematic Error Correction of Mass Transfer Coefficients and Kinetic Rate Constants" in proceedings of the 18th International Congress of Chemical and Process Engineering (CHISA 2008), Prague, August 2008.
+* [48]Rinker, E. B., S. S. Ashour and O. C. Sandall, "Kinetics and Modeling of Carbon-Dioxide Absorption Into Aqueous-Solutions of N-Methyldiethanolamine," Chem. Eng. Sci. **50**(5), 755-768 (1995).
+* [49]Sevimli, M. F. and C. Kinaci, "Decolorization of Textile Wastewater by Ozonation and Fenton's Process," Water Sci. Technol. **45**(12), 279-286 (2002).
+* [50]Sevimli, M. F. and H. Z. Sarikaya, "Ozone Treatment of Textile Effluents and Dyes: Effect of Applied Ozone Dose, pH and Dye Concentration," J. Chem. Technol. Biotechnol. **77**(7), 842-850 (2002).
+* [51]Shaikh, A. and M. H. Al Dahhan, "A Review on Flow Regime Transition in Bubble Columns," Int. J. Chem. React. Eng. **5**, (2007).
+* [52]Tanaka, M., G. Girard, R. Davis, A. Peuto and N. Bignell, "Recommended Table for the Density of Water Between 0 ~C and 40 ~C Based on Recent Experimental Reports," Metrologia **38**(4), 301-309 (2001).
+* [53]Tokumura, M., T. Katoh, H. Ohata and Y. Kawase, "Dynamic Modeling and Simulation of Ozonation in a Semibatch Bubble Column Reactor: Decolorization and Mineralization of Azo Dye Orange II by Ozone," Ind. Eng. Chem. Res. **48**(17), 7965-7975 (2009).
+* [54]Vaidya, P. D. and E. Y. Kenig, "Gas-Liquid Reaction Kinetics: A Review of Determination Methods," Chem. Eng. Commun. **194**(10-12), 1543-1565 (2007).
+* [55]Wilkinson, P. M., A. P. Spek and L. L. Vandierendonck, "Design Parameters Estimation for Scale-Up of High-Pressure Bubble-Columns," AIChE J. **38**(4), 544-554 (1992).
+* [56]Winkelmann, J., "Gases in Gases, Liquids and their Mixtures," Springer, Berlin (2007).
+* [57]Wu, J. N. and T. W. Wang, "Ozonation of Aqueous Azo Dye in a Semi-Batch Reactor," Water Res. **35**(4), 1093-1099 (2001).
+* [58]Wu, C. H., C. L. Chang and C. Y. Kuo, "Decolorization of Amaranth by Advanced Oxidation Processes," React. Kinet. Catal. Lett. **86**(1), 37-43 (2005).

manual_annotation/200pdfs_in_mmd/10.1002_cjce.24109.mmd

@@ -0,0 +1,181 @@
+Ozonation of phenol and substituted phenols: Dependency of the reaction rate constant on the molecular structure
+
+[
+
+[
+
+[
+
+[
+
+[
+
+[
+
+[
+
+[
+
+###### Abstract
+
+Ozonation is a prospective method for the treatment of phenol-contaminated wastewaters. In the present work, the reaction of ozone with 10 different substituted phenols was studied at acidic pH in two model gas-liquid contactors to investigate the dependency of the ozonation rate constant on the molecular structure of the phenolic compound. While three phenols were attached to electron-donating functional groups, seven were characterized by the presence of electron-accepting groups in their structure. In a stirred cell reactor (with flat-cell interface) and mechanically agitated contactor, reaction kinetics for these two types of phenols were studied in the fast pseudo-first order and chemical reaction controlled regime, correspondingly at T = 30degC. The effect of the electron-donating or electron-accepting functional group attached to the phenolic ring on the reactivity was studied and the linear dependence of the relative oxidation rate constant (with respect to phenol) on the Hammett substituent constant was proven in a Hammett plot. The values of susceptibility factor for the Hammett plots of phenols with electron-donating and electron-accepting substituents were \(-\)7.89 and \(-\)1.25, respectively. This work is useful to establish the ozonation pathway and predict the reaction rate constant from the molecular structure of the phenolic compound.
+
+Ozonation, phenol, rate constant, stirred cell 10]Ghashyam S. Bhosale 1,2]Prakash D. Vaidya 1]Parag R. Gogate 1]Jyeshtharaj B. Joshi 1,3]Raosaheb N. Patil 2]
+
+Ozonation is a prospective method for the treatment of phenol-contaminated wastewaters. In the present work, the reaction of ozone with 10 different substituted phenols was studied at acidic pH in two model gas-liquid contactors to investigate the dependency of the ozonation rate constant on the molecular structure of the phenolic compound. While three phenols were attached to electron-donating functional groups, seven were characterized by the presence of electron-accepting groups in their structure. In a stirred cell reactor (with flat-cell interface) and mechanically agitated contactor, reaction kinetics for these two types of phenols were studied in the fast pseudo-first order and chemical reaction controlled regime, correspondingly at T = 30degC. The effect of the electron-donating or electron-accepting functional group attached to the phenolic ring on the reactivity was studied and the linear dependence of the relative oxidation rate constant (with respect to phenol) on the Hammett substituent constant was proven in a Hammett plot. The values of susceptibility factor for the Hammett plots of phenols with electron-donating and electron-accepting substituents were \(-\)7.89 and \(-\)1.25, respectively. This work is useful to establish the ozonation pathway and predict the reaction rate constant from the molecular structure of the phenolic compound.
+
+## 1 Introduction
+
+Phenolic compounds are common pollutants in the effluents from several industries, such as petroleum refineries, coal processors, and plastic manufacturing units. Phenol-contaminated wastewaters are bio-refractory, and an intermediate chemical oxidation step is often employed before biological treatment.[1, 2, 3, 4, 5, 6] The aromatic ring is broken upon oxidation, thereby resulting in the formation of easily biodegradable, short-chain aliphatic compounds.
+
+Ozone is a prospective oxidizing agent for phenolic compounds. According to Hoigne and Bader,[7] the reaction between ozone and phenolics in aqueous solutions proceeds via indirect (at values of pH above 5-6) or direct pathway (at acidic pH). If the indirect reaction pathway is prevalent, ozone is decomposed into very reactive hydroxyl radicals that rapidly oxidize the phenolic compound in a series of reactions. The solute selectivity is low under such conditions because the rate constants do not vastly differ. As evident from our recent study, the ozonation rate constant typically varies in the 105-107 mol L\({}^{-1}\) s\({}^{-1}\) range.[8]The rates of ozone consumption and the formation of hydroxyl radicals rise with rising pH. In contrast, the ozone molecule reacts at the electron-rich sites of the phenolic compound if the pathway is direct. It is known that the direct reaction is very selective in terms of the solute. It occurs via oxidation-reduction, cyclo-addition, and electro-philic substitution pathways.[9] The choice of reaction conditions and the structure of the phenolic compound dominate the reaction pathway. For several systems, both reaction pathways are predominant.[10]
+
+In the past few years, much effort has been focused on understanding several aspects of ozonation reactions, such as reaction mechanism and kinetics and the effects of reaction variables on the degradation efficiency.[11, 12, 13] However, there exists disagreement among the kinetic data published in the literature even at similar reaction conditions, possibly because different model contactors and theoretical approaches were used and the reported kinetics were possibly an overall representation of mass transfer and chemical reaction.
+
+At acidic pH values, Hoigne[14] studied ozone consumption in solutions of organic compounds using batch reactors under pseudo-first order conditions and reported rates of the direct reaction of ozone with the organic compound. Hoigne and Bader[15] reported overall second-order kinetics for the ozonation reactions of phenolic compounds by using a homogeneous stopped-flow technique. In this method, the gas is dissolved in the liquid and contacted with the liquid reactant such that there is perfect mixing. Since there is no phase transition in this apparatus, mass transfer is not considered, and the kinetics are determined from the rate of the homogeneous reaction. However, it is not practical to perform trials using high concentrations of the reacting species in this apparatus. Hoigne and Bader[15] established the linear dependence of reaction rates on the electron-donating capacity of the aromatic compound. Under acidic conditions, they reported the prevalence of electrophilic substitution over cyclo-addition and oxidation-reduction.
+
+In another work, Gurol and Nekouinaini[10] used the competition kinetic method with phenol as a reference to investigate the kinetics of nine additional aromatic compounds in a semi-batch reactor. They applied a dynamic approach and reported relative rate constants. The pH was adjusted in the 2.5-3 range. The trend in reactivity of the aromatics was similar to that reported by Hoigne and Bader.[15] A strong dependence of the rate constant on the functional group attached to the aromatic ring was established and the stoichiometric factor was found to be unity. Thus, hydroxylated phenols reacted faster than phenol, because the electron density in the aromatic ring was higher. When the relative oxidation rate constants (with respect to phenol) were plotted against the Hammett substituent constant s (based on a value of s = 0 for phenol) in a Hammett plot, a susceptibility factor (r) of -8.0 was reported. Besides, it was verified that both the methods and approaches used by Gurol and Nekouinaini[10] and Hoigne[14] provided comparable kinetic data. For example, a Hammett plot of the absolute rate constants of phenolic compounds[10] and substituted benzenes[14] was best described by a straight line of slope equal to -8.2; this value was close to the value of r (-8.0) reported by Gurol and Nekouinaini.[10] A comparison of the aforementioned works on kinetics is presented in Table 1.
+
+The aromatic compounds studied by Hoigne,[14] Hoigne and Bader,[15] and Gurol and Nekouinaini[10] were characterized by the presence of electron-donating groups and exhibited negative values of the Hammett substituent constant. However, for phenolic compounds attached to electron-accepting groups, such a comprehensive study is not available in the literature. We decided to close this gap in this work. In this study, the kinetics of 10 substituted phenols was investigated at acidic pH by using two heterogeneous techniques at T = 30degC. A stirred cell (Lewis cell) with a flat-cell interface, whose gas-liquid interfacial area is geometrically simple and familiar, was used to calculate the ozonation rate constant of highly reactive compounds in the fast pseudo-first order reaction regime. This setup involves phase transition, and the gas penetrates into the liquid phase prior to reaction. Generally, it is challenging to sort out the contributions of mass transfer and chemical kinetics in such apparatuses, due to the intricacy in finding liquid-side mass transfer coefficient and interfacial area for mass transfer. However, the enhancement factor that describes reactive absorption can be analytically obtained for a pseudo-first order irreversible reaction,[16] and the rate constant can be easily determined using a stirred cell. If the reaction pathway is known, then analysis of the liquid phase is avoidable. Recently, we reported values of the ozonation rate constant for 20 substituted phenols at neutral pH using this stirred cell.[8] The mechanically agitated contactor, wherein the gas is dispersed in the liquid and the mass transfer interfacial area is the surface area of each bubble, was used for studying the ozonation reactions of less reactive compounds in the kinetically controlled reaction regime. Similar to the stirred cell, phase change occurs and the flow pattern in this contactor is not well defined. To validate these methods, the results of this work were compared with those reported in past works. The dependency of the rate constant on the type of functional group attached to the phenolic ring was demonstrated for compounds with electron-donating as well as electron-accepting groups.
+
+## 2 Theory
+
+Ozonation reactions are heterogeneous gas-liquid reactions wherein the mass transfer of ozone is accompanied by chemical reaction in the liquid phase. In such processes, the overall reaction rate depends on the relative rates of mass transfer and chemical reaction.[17] The phenolic compounds with electron-donating groups, whose values of the Hammett constant are negative, react fast with ozone in the liquid film (fast reaction regime or regime 3). When the reaction is fast, mass transfer occurs in parallel with chemical reaction and the rate is enhanced. However, substituted phenols with electron-accepting groups, which have positive Hammett constant values, are less reactive with ozone and the reaction occurs in the bulk liquid (slow reaction regime or regime 2). The absorption process is then purely controlled by mass transfer. When the rates of chemical reaction and mass transfer are comparable, the volumetric rate is given by Equation (1):
+
+\[R_{A}a = \frac{\left[ A^{*} \right]}{\left[ \frac{1}{k_{A}a} \right] + \frac{1}{\varepsilon_{L}k_{1,1}\left[ B_{0} \right]}}\]
+
+where \(\left[ A^{*} \right]\) is the saturation concentration of ozone (denoted here as \(A\)), \(k_{L}\) is the liquid-side mass transfer coefficient, \(a\) is the gas-liquid interfacial area, \(\left[ B_{0} \right]\) is the concentration of the phenolic compound (\(B\)) in bulk liquid, \(k_{1,1}\) is the second-order rate constant and \(\varepsilon_{L}\) is the liquid holdup in the reactor.[18]
+
+If \(k_{L}a\) [\(A^{*}\)] is the rate for purely mass transfer controlled operation (regime 2), then the rate for fast reactions (regime 3) is given by Equation (2):
+
+\[R_{A}a = k_{L}a\left[ A^{*} \right]E\]
+
+where \(E\) is the enhancement factor that accounts for the improvement in the rate due to chemical reaction, and its value is equal to the Hatta number (\(Ha\)). As defined in Equation (3), \(Ha\) is the ratio of the maximum possible reaction rate to the mass transfer rate:
+
+\[Ha = \frac{\sqrt{D_{A}k_{1,1}\left[ B_{0} \right]}}{k_{L}}\]
+
+In regime 3, the inequality \(10<Ha<<(E_{i}\cdot 1)\) is satisfied. Here, \(E_{i}\) denotes the enhancement factor for instantaneous reaction and is given by Equation (4):
+
+\[E_{i} = \frac{\left[ B_{0} \right]}{z\left[ A^{*} \right]}\sqrt{\frac{D_{B}}{D_{A}}}\]
+
+where \(z\) is the stoichiometric coefficient for the reaction of ozone with the phenolic compound and \(D_{A}\) and \(D_{B}\) are the diffusivities of ozone and the phenolic compound in the liquid phase, respectively. From Equations (2) and (3), the overall rate in the fast regime can be calculated by using Equation (5):
+
+\[R_{A}a = a\left[ A^{*} \right]\sqrt{D_{A}k_{1,1}\left[ B_{0} \right]}\]
+
+For the case when the reaction occurs in the film as well as the bulk (the regime in between 2 and 3), the overall rate is given by Equation (6):
+
+## 3 Experimental
+
+### Materials
+
+The chemicals used in this study were procured from a local vendor (S.D. Fine Chemicals Pvt. Ltd., Mumbai). The phenolic compounds studied were phenol, 2-aminophenol (2-AP), 3-aminophenol (3-AP), 4-methoxy phenol (4-MP), 3-methoxy phenol (3-MP), 3-chlorophenol (3CP), 3-nitrophenol (3-NP), 2,5-dichlorophenol (2,5-DCP), 2,4-dinitrophenol (2,4-DNP), and 2,4,6-trichlorophenol. A laboratory overhead stirrer was supplied by Remi Lab World, Mumbai. The ozone generator (10 kg h\({}^{-1}\)), oxygen concentrator and UV-ozone analyzer were purchased from Eltech Engineers Ltd. (Mumbai). All the glassware used in this study was supplied by SCAM Lab Pvt. Ltd. (Mumbai).
+
+### Setup and procedure
+
+To investigate the absorption process and identify the reaction regime, various model contactors with known interfacial area are used. The details of many model contactors have been extensively described by Doraiswamy and Sharma.[17] The selection of a model contactor for any particular system is based on several factors, such as the residence time, surface renewal time, effective interfacial area, mass transfer coefficient, and the ease of operation, sampling, and fabrication. Joshi and Doraiswamy[18] assessed the merits and demerits of model contactors based on the aforesaid factors. Among all model contactors, the versatile stirred cell (with flat-cell interface) and mechanically agitated contactor (MAC) were employed in the present work, due to the relative ease in varying the reactant concentration and mass transfer coefficient.
+
+The stirred cell (or Lewis cell) has also been used in our recent work.[8] Its interfacial area (71 cm\({}^{2}\)) and k\({}_{\rm L}\) value (0.052 cm s\({}^{-1}\)) were known. The experimental procedure was described in our previous work.[8] It was verified that the ozonation reactions investigated in this contactor conformed to the fast reaction regime systems at T = 30\({}^{\circ}\)C. The MAC was a baffled glass kettle (capacity 2 L). It was equipped with five neck openings for the impellers, gas inlet and outlet, sampling point, and thermometer. The down-flow and up-flow pitched blade impellers were installed on a single shaft and connected to an overhead stirrer. A ball sparger of diameter 20 mm was used to feed ozone gas with bubble size of about 5 mm. The schematic representation of the MAC and the flowsheet are given in Figure 1. It was checked that the reactions studied in the MAC conformed to the chemical reaction controlled regime.
+
+The experiments were performed in MAC in semi-batch mode at a constant temperature (30\({}^{\circ}\)C). The pH of solution was maintained constant at a value of 3 using a phosphate buffer. The effect of impeller speed on the rate of degradation of the phenolic compounds was investigated. It was found that the rate was independent of the agitation speed above 1100 rpm. A speed of 1600 rpm was used throughout this study for the elimination of mass transfer resistance. First, the reactor was charged with 1.5 L of aqueous solution of the phenolic compound. A zero-time sample was collected by using a syringe. Ozone gas generated by ozone generator was then introduced (flow rate = 5 L min\({}^{-1}\) and concentration = 6 mg L\({}^{-1}\)) through the ball sparger located near the tip of the bottom impeller. Samples of the reaction mixture were collected at different times and analyzed by high-performance liquid chromatography (Agilent HPLC) to find the reaction rate. A reverse-phase C-18 column (5 um, 4.6 x 250 mm) with Acclaim 120 stationary phase was used. Using least-square regression, the initial rate of degradation of the substituted phenols was determined from the slope of (B)/(B\({}_{0}\)) versus time plot (see Figure 2). All experiments were repeated thrice (difference < 5%) and the mean value was considered for estimation of the intrinsic rate constant. The diffusivities of ozone and substituted phenols were estimated by using correlations suggested by Johnson and Davis[19] and Wilke and Chang.[20] Henry's coefficient (H\({}_{\rm A}=1.11\times 10^{4}\) m\({}^{3}\) kmol\({}^{-1}\) kPa) for ozone-phenol reaction in aqueous solutions was estimated from the relation of Sotelo et al.[21]
+
+## 4 Results and discussion
+
+In our earlier study,[8] we found that the rate constant for ozonation of substituted phenols at neutral pH is in the 10\({}^{5}\)-10\({}^{7}\) kmol m\({}^{-3}\) s\({}^{-1}\) range and the reactivity does not show any relation with the functional group attached to the phenol ring. It is evident that the radical reaction (indirect reaction) pathway was prevalent. This is in agreement with the past work of Hoigne and Bader.[15] However, Beltran[22] observed that the direct reaction mechanism predominates over the indirect reaction pathway at pH values less than 4. In order to understand and quantify the direct reaction pathway, the effect of a radical scavenger, t-butanol, on the degradation of 3-NP was studied here. It is known that t-butanol is a scavengercompound for hydroxyl radicals.[10] Here, it was found that there is no influence of t-butanol on the reaction rate at T = 30\({}^{\circ}\)C and pH = 3 (Figure 3). Thus, predominance of the direct reaction pathway at acidic pH was evident. A similar observation was also reported in past works.[22, 23]
+
+### Direct reaction with phenols attached to electron-donating groups
+
+Hoigne and Bader[15] determined the rate constants for aromatic compounds (e.g., substituted benzenes) by using homogenous, stopped-flow technique and reported a linear correlation with the Hammett substituent constant. The rate constant varied from 4 mol L\({}^{-1}\) s\({}^{-1}\) for benzene (based on s = 0 for benzene) to 279 mol L\({}^{-1}\) s\({}^{-1}\) for p-xylene (s = -0.34 for p-xylene). Later, Gurol and Nekouinaini[10] also investigated the kinetics of ozonation of other aromatic compounds (such as substituted phenols) and found that the reaction rate constant varies linearly with the Hammett constant. In their work, the rate constant varied from \(2.3\times 10^{3}\) mol L\({}^{-1}\) s\({}^{-1}\) for phenol (s = -0.37) to 1171 \(\times\) 10\({}^{3}\) mol L\({}^{-1}\) s\({}^{-1}\) for 4-hydroxy phenol (s = -0.74). Thus, the competition kinetic method was validated for estimating the ozonation rate constant of the fast reactions of phenolic compounds with electron-donating groups. The rate constants reported by Hoigne and Bader[15] and Gurol and Nekouinaini[10] are represented in Table 2.
+
+In this work, the rate constants for the reactions with three phenolic compounds attached to electron-donating groups (i.e., 2-AP, 3-AP, and 4-MP) were determined using the stirred cell at acidic pH. The results are represented in Table 3. It can be concluded that the rate constant for 4-MP is around 40 times higher than that for 2-AP. This results in a change in the reaction regime from in between 2 and 3 to regime 3. The shift in the reaction regime is mainly due to the higher electron-donating capacity of 4-MP. It is clear that ozone reacts with phenolic compounds at acidic pH mostly by electrophilic substitution. The rate constants presented by Hoigne,[14] Gurol and Nekouinaini,[10] and this work were collectively plotted as log (\(k_{1,\lambda}/k_{p}\)) versus Hammett substituent constant. It can be seen from the Hammett plot in Figure 4 that a linear relation prevails (susceptibility factor \(p\) = -7.89 and R2 = 0.97). This is in line with the value of \(p\) = -8.0 reported by Gurol and Nekouinaini.[10] These results validated the methodology adopted in this study. The linearity of this plot confirms the reaction mechanism for electron-donating compounds. However, data points for the validity of such linearity for Hammett constant values greater than -0.35 are unavailable. Thus, an investigation for phenols with the same stoichiometry as the reagent.
+
+\begin{table}
+\begin{tabular}{p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt}} \hline \hline \multicolumn{2}{p{14.2pt}}{**Sr. no**} & \multicolumn{1}{p{14.2pt}}{**Past works**} & \multicolumn{1}{p{14.2pt}}{**Name of compound**} & \multicolumn{1}{p{14.2pt}}{**k\({}_{1,\lambda}\) (mol L\({}^{-1}\) s\({}^{-1}\))**} & \multicolumn{1}{p{14.2pt}}{**Hammett constant**} \\ \hline
+1 & Hoigne and Bader[15] & Benzene & 4 & 0.00 \\
+2 & & Toluene & 29 & \(-\)0.17 \\
+3 & & o-xylene & 173 & \(-\)0.24 \\
+4 & & m-xylene & 179 & \(-\)0.24 \\
+5 & & p-xylene & 279 & \(-\)0.34 \\
+6 & Gurol and Nekouinaini[10] & Phenol & \(2.3\times 10^{3}\) & \(-\)0.37 \\
+7 & & 2-Methyl phenol & \(11\times 10^{3}\) & \(-\)0.44 \\
+8 & & 3-Methyl phenol & \(12\times 10^{3}\) & \(-\)0.44 \\
+9 & & 4-Methyl phenol & \(27\times 10^{3}\) & \(-\)0.54 \\ \hline
+10 & & 2,6-Dimethyl phenol & \(40\times 10^{3}\) & \(-\)0.51 \\
+11 & & 2,3-Dimethyl phenol & \(57\times 10^{3}\) & \(-\)0.51 \\ \hline
+12 & & 2,4-Dimethyl phenol & \(207\times 10^{3}\) & \(-\)0.61 \\
+13 & & 3,4-Dimethyl phenol & \(211\times 10^{3}\) & \(-\)0.61 \\
+14 & & 4-Hydroxy phenol & \(1171\times 10^{3}\) & \(-\)0.74 \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Direct reaction rate constants reported in the literature for ozone-substituted phenols
+
+\begin{table}
+\begin{tabular}{p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt} p{14.2pt}} \hline \hline \multicolumn{2}{p{14.2pt}}{**Sr. no.**} & \multicolumn{1}{p{14.2pt}}{**Name of compounds**} & \multicolumn{1}{p{14.2pt}}{**Hammett constant**} & \multicolumn{1}{p{14.2pt}}{**R\({}_{\mathbf{A}\mathbf{A}\times 10^{7}}\)**} & \multicolumn{1}{p{14.2pt}}{**k\({}_{1,\lambda}\) (mol L\({}^{-1}\) s\({}^{-1}\))**} & \multicolumn{1}{p{14.2pt}}{**Regime of reaction**} & \multicolumn{1}{p{14.2pt}}{**Ha**} & \multicolumn{1}{p{14.2pt}}{**E\({}_{\mathbf{i}}\)**} \\ \hline
+1 & 2-Aminophenol & \(-\)0.02 & 3.64 & \(3.89\times 10^{3}\) & Between 2 \& 3 & 6 & 115 \\
+2 & 3-Aminophenol & \(-\)0.02 & 4.09 & \(4.93\times 10^{3}\) & Between 2 \& 3 & 7 & 109 \\
+3 & 4-Methoxyphenol & \(-\)0.27 & 4.04 & \(148\times 10^{3}\) & 3 & 70 & 391 \\ \hline \hline \end{tabular}
+\end{table}
+Table 3: Rate constant for ozonation of phenols with electron-donating groups at T = 30\({}^{\circ}\)Celectron-accepting groups appears desirable as was the basis for the following work in this study.
+
+### Direct reaction with phenols attached to electron-accepting groups
+
+The phenols associated with electron-accepting groups are less reactive with ozone. The kinetics of the reactions of these phenols was studied in the MAC in the kinetically controlled reaction regime (or regime 1) at acidic pH. The values of rate constants are shown in Table 4.
+
+As the Hammett constant value increased, the reaction rate was lowered, and the reaction regime shifted from between 1 and 2 to regime 1 (see Equation (1)). For Hammett constant values up to 0.7, the reaction rate and mass transfer rate were comparable (see Table 4). However, for values above 0.7, the reaction rate decreased further, and the system conformed to the reaction controlled regime (\(R_{A}a=\varepsilon_{L}\)\(k_{1,1}\)[\(A^{*}\)][\(B_{0}\)]). The rate constant for the direct reaction between phenol and ozone, reported by Hoigne and Bader[115] using the stopped-flow method (\(1300\pm 300\) mol L\({}^{-1}\) s\({}^{-1}\)), is less than that reported by Gurol and Nekouainaini[101] using the competition kinetic method (\(2358\) mol L\({}^{-1}\) s\({}^{-1}\)). However, the rate constant obtained in this study (\(1140\) mol L\({}^{-1}\) s\({}^{-1}\)) is closer to that obtained by using the homogeneous stopped-flow method. A plot of log(\(k_{1,1}\)/\(k_{p}\)) versus Hammett constant is shown in Figure 5. Once again, a linear relation was observed in the Hammett plot. The value of the susceptibility factor (\(\rho\)) was \(-1.25\) (R\({}^{2}\) = 0.94).
+
+Ozone is a strong electrophilic agent. Its reactivity with aromatic compounds is inherently dependent on the electron donating and accepting capacity of the compound. The Hammett constant of a compound characterizes its electron donating and accepting capacity. This hypothesis is principally proven by linear relation of log (\(k_{1,1}\)/\(k_{p}\)) versus Hammett constant plot.
+
+## 5 Conclusions
+
+The reaction pathway of ozonation of substituted phenols depends upon pH of the reaction mixture. At neutral pH, the rate constants are independent of the structure of the phenolic compound, thereby indicating a radical reaction mechanism. However, at acidic pH, the rate constant strongly depends on the substituent group attached to the phenolic ring. The reactions with compounds bearing electron-donating groups conform to the fast reaction regime (Equation (5)) when pH is low. The rate varies linearly with the electron-donating capacity of the substituent. This indicates that the electrophilic substitution reaction mechanism governs the overall rate of reaction. However, in the case of substituted phenols with electron-accepting groups (with positive values of the Hammett constant), the reaction rate and the mass transfer rates are comparable. This work will help in predicting the rate constants of ozonation of substituted phenols with positive as well as negative values of the Hammett constant.
+
+## Nomenclature
+
+* saturation concentration of ozone in the liquids (kmol m-3)
+* concentration of the phenolic compound in the liquid phase (kmol m-3)
+* effective interfacial area (m-1)
+* diffusivity of ozone in liquid (m2 s-1)
+* diffusivity of phenol in liquid (m2 s-1)
+* enhancement factor (-)
+* enhancement factor for instantaneous reaction regime (-)
+* Hatta number (-)
+* rate constant for reaction between ozone and substituted phenol (m3 kmol-1 s-1)
+* true mass transfer coefficient (m s-1)
+* rate constant for the direct reaction of phenol with ozone (m3 kmol-1 s-1)
+* overall rate of reaction (kmol m-3 s-1)
+* stoichiometric coefficient for the reaction between ozone and phenolic compound
+
+## Greek symbols
+
+* Hammett substituent constant
+* susceptibility factor
+
+## References
+
+* [1] C. D. Adams, R. A. Cozzens, B. J. Kim, _Water Res._**1997**, _31_, 2655.
+* [2] A. B. Alvares, C. Diaper, S. A. Parsons, _Environ. Technol._**2001**, _22_, 409.
+* [3] G. Bertanza, C. Collivignarelli, R. Pedrazzani, _Water Sci. Technol._**2001**, _44_, 109.
+* [4] B. Guieysse, Z. N. Norvill, _J. Hazard. Mater._**2014**, _267_, 142.
+* [5] M. Jeworski, E. Heinzle, _Biotechnology Annual Review_**2000**, \(6\), 163.
+* [6] H. H. W. Lee, G. Chen, P. L. Yue, _Water Sci. Technol._**2001**, _44_, 75.
+* [7] J. Hoigne, H. Bader, _Water Res._**1976**, _10_, 377.
+* [8] G. S. Bhosale, P. D. Vaidya, J. B. Joshi, R. N. Patil, _Ind. Eng. Chem. Res._**2019**, _58_, 7461.
+* [9] P. Bailey, _Chem. Rev._**1958**, _58_, 925.
+* [10] M. D. Gurol, S. Nekouinaini, _Ind. Eng. Chem. Fund._**1984**, _23_, 54.
+* [11] Y. Pi, J. Wang, _Sci. China Ser. B_**2006**, _49_, 379.
+* [12] C. P. Yu, Y. H. Yu, _Ozone-Sci. Eng._**2001**, _23_, 303.
+* [13] V. Augugliaro, L. Rizzuti, _Chem. Eng. Sci._**1978**, _33_, 1441.
+* [14] J. Hoigne, in _Handbook of Ozone Technology and Applications_, Vol. 1 (Eds: R. G. Rice, A. Netzer), Ann Arbor Science, Ann Arbor, MI **1982**.
+* [15] J. Hoigne, H. Bader, _Water Res._**1983**, _17_, 173.
+* [16] W. P. M. Van Swaaij, G. F. Versteeg, _Chem. Eng. Sci._**1992**, _47_, 3181.
+* [17] L. K. Doraiswamy, M. M. Sharma, _Heterogeneous Reactions: Analysis, Examples and Reactor Design_, Vol. 2, Wiley, New York **1984**.
+* [18] J. B. Joshi, L. K. Doraiswamy, _Albright's Chemical Engineering Handbook_, CRC Press Taylor & Francis Group, Boca Raton, FL **2008**.
+* [19] P. N. Johnson, R. A. Davis, _J. Chem. Eng. Data_**1996**, _41_, 1485.
+* [20] C. R. Wilke, P. Chang, _AIChE J._**1955**, \(1\), 264.
+* [21] J. L. Sotelo, F. J. Beltran, F. J. Benitez, H. J. Beltran, _Water Res._**1989**, _23_, 1239.
+* [22] F. J. Beltran, _Reaction Kinetics for Water and Wastewater Systems_, CRC Press, Boca Raton, FL **2004**.
+* [23] M. A. Boncz, H. Brining, W. H. Rulkens, E. J. R. Sudholter, G. H. Harmsen, J. W. Bijsterbosch, _Water Sci. Technol._**1997**, _35_, 65.
+
+## How to cite this article:
+
+Bhosale GS, Vaidya PD,
+
+Gogate PR, Joshi JB, Patil RN. Ozonation of phenol and substituted phenols: Dependency of the reaction rate constant on the molecular structure. _Can J Chem Eng._ 2021;1-8. [https://doi.org/10.1002/cjce.24109](https://doi.org/10.1002/cjce.24109)

manual_annotation/200pdfs_in_mmd/10.1002_cjce.5450820616.mmd

@@ -0,0 +1,121 @@
+# Decolorization of RR-120 Dye Using Ozone and Ozone/UV in a Semi-Batch Reactor
+
+Mohammad Kazemi, Jafar S. Soltan Mohammadzadeh, Ali B. Khoshfetrat
+
+and Mohammad A.Kaynejad
+
+###### Abstract
+
+Dye manufacturing, textile dyeing and finishing processes generate a large amount of wastewater. Dyes, some mordants, sizing agents and dyeing aids are deep in colour and highly pollutant (Hsu et al., 2001). Treatment of wastewater from textile dyeing is an environmental problem that has received considerable attention. Effluent from textile processing is often discharged into municipal sewage treatment plants or waterways (Aplin and Waite, 2000). Textile dyes are not easily biodegradable, and consequently sewage treatment removes only a small amount of certain types of dyes by adsorption and settling but it fails to remove reactive dyes (Aplin and Waite, 2000). Reactive dyes are one of the most widely used classes of dyes, accounting for almost 32% by quantity and about 43% by value of world consumption of dyes by cellulosic fibres in 1993 (Holme, 1997). They also have lower fixation compared to other classes of dyes. Reactive dye that remains in the effluent has been hydrolyzed during the dyeing process and therefore cannot be reused (Cooper, 1993). The reactive dyeing process produces two different waste streams. The initial concentrated dye bath effluent with a temperature of above 50degC, has a pH of at least 11 (Grovers et al., 1998) and contains up to 0.6 g/L unfixed dye and about 40-80 g/L of salt (NaCl or Na2SO4) and small quantities of other additives.
+
+In most countries national standards place strict restrictions on the colour of wastewater. There are several possible methods for treating wastewater from textile dyeing (Hazel, 1995). Some methods such as coagulation, adsorption and nanofiltration remove dyes from wastewater and produce a secondary waste stream that requires further treatment or disposal. On the other hand, other methods like sequential anaerobic/aerobic treatment or aerobic treatment using fungi and advanced oxidation processes can degrade dyes in the effluent with a good efficiency.
+
+Advanced oxidation processes (AOPs) are a group of processes that are based on the generation of hydroxyl radicals, which are highly reactive oxidants. AOPs can oxidize a wide range of compounds that are otherwise difficult to degrade. The five AOPs that have been most widely studied are based on reaction with ozone, O3/UV, H2O2/UV, Fenton's reagent (Fe2+/H2O2) and catalytic (e.g. TiO2) UV radiation.
+
+Ozone processing is a promising alternative. Ozone is an extremely strong oxidant. It can undergo self-decomposition in an aqueous medium to form hydroxyl free radicals that have a stronger oxidation capacity (Staehelin and Hoigene, 1985; Sotelo et al., 1987). Therefore, ozone has found many applications. In some industrial applications, it has been widely accepted as an effective disinfectant and a chemical oxidant (Gould and Weber, 1976; Beltran et al., 1993; Perkins et al., 1995). Studies that have been reported on the reaction between ozone and dye (Saunders et al., 1983; Benitez et al., 1993; Cahr et al., 1994) indicate that ozone has an excellent decolorizing capability. In studies of the ozonation of dye solutions, Gahr et al. (1994) found that dye degradation was faster at lower initial concentrations, and that pH decreasedduring the reaction time. However, industrial applications have been limited due to low utilization efficiency of ozone and its high production cost (Hsu et al., 2001). The low utilization efficiency of ozone can be improved by increasing the retention time of ozone gas in an aqueous solution (Hsu et al., 2001) or by using an effective contacting scheme. To increase ozone utilization efficiency, an effective venturi injection system has been utilized in this work. As an effective ozone injection method, this system has ozone utilization efficiency (OUE) of above 95%. An improvement in ozone utilization efficiency without any additional utilities can be a major advantage in cost reduction. In this work, decolorization of a reactive red dye solution has been studied using an efficient O3 oxidation system. The effect of UV radiation on performance of ozone oxidation system in decolorization has also been investigated. Experimental results showed that, at various pH values and initial dye concentrations, the ozone utilization efficiency remains above 95%. This can reduce cost of application of ozone effectively.
+
+## Experimental
+
+C.l. Reactive Red 120 (RR-120) was provided by Chinas chemicals company. Its maximum absorbance occurs at 284.6 nm. The synthetic wastewater was prepared by dissolving the required amounts (200 to 500 +- 30 mg/L) of RR-120 in distilled water. The pH was adjusted to 3.0, 6.0, 8.0 and 11.0 (+-0.2), by using buffer solutions.
+
+Schematic diagram of the experimental apparatus is shown in Figure 1. Pure oxygen was fed to ozone generator (Ozomatic Lab 802, Wedeco, Germany) with 4 g/h capacity. Ozone concentration (measured by BMT, Press version) and volumetric flow rate were monitored. UV radiation was supplied with one 30-watt low-pressure mercury lamp with radiation wavelength of 253.7 nm.
+
+For all experiments, the ozone generator was adjusted at 2 g/h ozone generation capacity. At the beginning of each run, 5.5 l of dye solution with the desired concentration and pH was transferred into the reactor. All experiments were conducted at room temperature in a water- jacketed reactor. Ozone was injected into the dye solution through a standard ozone injector (Lab 802) using a stainless steel circulation pump (CRN2-30). The flow rate and concentration of the ozone gas were 50 NL/h and 40 g/m3, respectively.
+
+To measure excess ozone in outlet gas stream, a KI trap was used. Excess ozone reacted with KI which was then titrated with Na2SO4 solution. Each experiment was repeated twice and samples were collected at an approximate 5-min based intervals. Each sample was divided into three sub-samples that were analyzed separately. As soon as the samples were taken from the reactor, their colour values were determined using a UV/Vis spectrophotometer (Lambda 2- Perkin Elmer) at 284.6 nm. An Eco8 COD thermal reactor (VELP Scientifica, Italy) was employed to digest oxidizable organic compounds at 150degC for chemical oxygen demand (COD) analyses.
+
+Ozone utilization efficiency is defined by Equation (1), as suggested by Hsu and Hung (2001).
+
+\[OUE\% = \frac{{{\text{O}}_{3n} - {\text{O}}_{3n}}}{{{\text{O}}_{3n}}} \times 100\]
+
+OUE values are based on 90% decolorization of the dye solution in which a light colour can be observed. \({\text{O}}_{3\text{in}}\) and \({\text{O}}_{3\text{out}}\) are the total input and output ozone measured in gas phase.
+
+## Results and Discussion
+
+### Kinetic Analysis for Decolorization
+
+Reaction between dye and ozone is a relatively fast reaction, and ozone is consumed almost completely in the liquid film (Perkins et al., 1995; Hsu et al., 2001). The kinetic model of decolorization reaction can be assumed as a pseudo first order reaction (Perkins et al., 1995). In this work, decolorization reaction of RR-120 is studied in semi-batch reactor (batch with respect to the dye solution). The pseudo first order expression is assumed to describe the heterogeneous ozonation of the dye solution. Furthermore, the ozone input concentration is constant and no residual ozone remains in the bulk liquid (Perkins et al., 1995; Hsu et al., 2001). In addition, ozone utilization efficiency is high in this pilot system (above 90%). Hence, an almost constant ozone influence, which can be incorporated into the overall reaction rate constant (_k_) can be considered. The rate constant of the pseudo first order reaction can be calculated by Equation (2).
+
+\[- Ln\left\{\frac{C}{C_{0}} \right\} = kt\]
+
+in which G and \(C_{0}\) are concentration and initial concentration of dye in mg/L, \(M\) is the reaction rate constant (min-1) and t is ozonation time (min).
+
+Figure 2 shows a plot of -_Ln_ (_C_/_C_0) vs. reaction time for a typical reaction condition of initial dye concentration of 500 mg/L and different pH values in O3 oxidation system. It can be observed that the data points fall on a reasonably straight line. Based on examination of all reaction data, it was assumed that oxidation of RR-120 reaction rate follows a pseudo first order kinetics. Experimental data also revealed that for reaction of oxidation with O3/UV, assumption of pseudo first order kinetics was valid.
+
+### Comparison of Oxidation with O3 and O3/UV
+
+In Figure 3, profile of concentration of RR-120 dye with time is shown for oxidation with O3 and O3/UV at pH of 3. The figure shows that adding UV radiation to O3 does not have any detectable effect on the rate of destruction of the dye. At higher pH values, ozone decomposition and production of free hydroxyl radical is encouraged. Therefore, application of UV radiation along with ozone increases rate of oxidation reaction compared to oxidation with O3 alone at pH values of
+
+Figure 1: Schematic diagram of the experimental apparatus.
+
+7 and above. Figure 4 shows that for solution with the initial dye concentration of 200 mg/L and pH of 11, the time for 50% destruction of dye with O3 is 26 min, whereas combined O3 and UV bring about 50% destruction in 18 min. That is, addition of UV to ozone reduces the 50% destruction time by about 30% at pH of 11.
+
+### Decolorization Time of Dye
+
+Figures 5 and 6 depict the decolorization of RR-120 with the initial dye concentration of 200 mg/L at various pH values for O3 and O3/UV oxidation systems, respectively. The figures indicate that during oxidation by both O3 and O3/UV, the colour value of RR-120 is reduced exponentially. The decolorization profiles of RR-120 were approximately the same in other initial dye concentrations. At all pH values, the dye molecules are destroyed by direct ozone and/or hydroxyl free radicals mechanisms. As the decolorization process continues various intermediate products are generated and accumulated to such an extent that the difference in reaction rate becomes significant at different pH values. After that, the decolorization rates are slower than the initial rate for all pH values. This is because most
+
+Figure 4: Effect of adding UV radiation to ozone oxidation, \(C\)0 = 300 mg/L, pH = 11.
+
+Figure 5: Effect of pH on decolorization of RR-120, O3 oxidation system, \(C\)0 = 200 mg/L.
+
+Figure 3: Effect of adding UV radiation to ozone oxidation, pH = 3, \(C\)0 = 200 mg/L.
+
+Figure 6: Effect of pH on decolorization of RR-120, O3/UV oxidation system, \(C\)0 = 200 mg/L.
+
+dye molecules are destroyed and the persistent intermediate products become the primary substrates for ozone molecules and hydroxyl free radicals. The figures also indicate that decolorization at range of low pH value (e.g. pH of 3) has the lowest rate. At low pH, decolorization is favored by a more direct ozone attack (molecular mechanism) whereas at higher pH values ozone decomposition is accelerated primarily by hydroxyl free radicals, thus a less selective hydroxyl attack on dye molecules occurs. At near neutral pH (pH of 6 and 8), the combined effect of free radical and molecular reaction of oxidant agents with dye greatly enhances the rate of decolorizatioin. Figures 5 shows that reaction rate increases from acidic conditions, reaching maximum at neutral pH values and then a slight decrease at higher pH values is observed. This trend is observed for other initial concentrations. The maximum decolorization rate is observed at neutral pH values for O3 oxidation system. It is also interesting to note that the rate of decolorization for pH values of 6 and 11 have a little difference in the initial dye concentrations.
+
+The colour destruction rate at pH values of 7 and higher increases with adding UV to ozone oxidation system. Also, it was observed that there was a considerable enhancement in dye destruction at pH value of 11 owing to higher production of free hydroxyl radicals. The rate of decolorization for pH values of 8 and 11 are almost identical because of higher production of hydroxyl free radicals at pH of 11.
+
+Table 1 summarizes the results of the decolorization of the RR-120 dye with O3 and O3/UV at different initial dye concentrations and pH values. Table 1 shows that for all pH values and initial dye concentrations, OUE is higher than 95%, i.e. the amount of escaped ozone in outlet gas stream is negligible.
+
+According to Table 1, the decolorization rates are higher at lower initial dye concentrations during both O3 and O3/UV oxidation. This confirms the results of the previous reports (Gahr et al., 1994). This is due to production of less intermediates at lower initial dye concentrations. Moreover, the table shows that the rate constants for O3/UV oxidation are higher than O3 oxidation system. At high pH values the UV radiation has stronger encouraging effect on O3 oxidation system.
+
+### COD Elimination
+
+Figure 7 shows the normalized profiles of COD and dye concentration of RR-120 with time at pH values of 8 and 11 (O3 oxidation system and \(C\)0 = 300 mg/L). Comparison of variations of normalized COD and concentration with time reveals that despite an almost identical trend for reduction of both colour and COD in O3 oxidation system, the time required for mineralization is approximately 1.5 times more than that of decoloration. Figure 7 shows that mineralization process is slower than decolorization due to presence of intermediate products.
+
+### Ozone Utilization Efficiency
+
+Ozone utilization efficiency for decolorization of RR-120 at various pH values and different initial dye concentrations is presented in Table 1. As Table 1 shows ozone utilization efficiencies are above 95% and more or less constant at various pH values. From the data shown in Table 1, it can be concluded that OUE increases with increasing initial dye concentration. OUE for oxidation with O3/UV is also above 95%. The OUE in O3/UV process is slightly higher than that for oxidation with O3. This increase in OUE is most likely due to the increase in production of hydroxyl free radicals in O3/UV system compared to O3 system.
+
+## Conclusions
+
+This work showed that both decolorization and mineralization processes of RR-120 dye occur during oxidation by O3 and O3/UV systems. The pseudo first order reaction rate constant were determined for R-1209 destruction in O3 and O3/UV systems at different initial dye concentrations and pH values. Because of production of intermediates, the decolorization rate was about 1.5 times the mineralization rate. The results also showed that at low pH values (pH of 3) the decolorization rate constant had the lowest value and adding UV had no detectable effect on RR-120 decolorization rate. At near neutral pH values (pH of 8) decolorization was fastest in both oxidation systems and a considerable increase in decolorization at higher pH values (pH of 11) in O3/UV system was observed. Approximately 65 to 115 min contact
+
+Figure 7: Comparison of the rate of decolorization and mineralization in O3 oxidation system, \(C\)0 = 300 mg/L.
+
+time was required for a 90% decolorization at concentrations of between 200 to 500 +- 30 mg/L, 2 g/h ozone, at different pH values and room temperature in the experimental set up. The ozone utilization efficiency was above 95% at all pH values and initial dye concentrations (in the range of 200 to 500 mg/L). Adding UV could also accelerate decolorization and mineralization especially at pH values of 7 and above. Hence, ozone treatment of various dye solutions can be faster, more efficient and economical by adding UV and an effective ozone injection system.
+
+## Nomenclature
+
+\(C_{0}\): initial dye concentration, (mg/L) \(C\): dye concentration, (mg/L) \(COD\): chemical oxygen demand, (mg/L) \(\#\): first order reaction rate constant, (min-1) \(OUB\): ozone utilization efficiency \(\#\): time, (min)
+
+## References
+
+- Alpine, A. and T.D. Waite, "Comparison of Three Advanced Oxidation Processes For Degradation of Textile Dyes", Wat. Sci.Tech. **42**, 345-354 (2000).
+
+- Beltran, F., J.M. Encinar and J.F. Garcia-Araya, "Oxidation by Ozone and Chlorine Dioxide of Two Disttillery Wastewater Contaminants: Gallic acid and Epicatechin", Wat. Res. **27**, 1023-1032 (1993).
+
+- Benitez, F., J. Beltran-Heredia, T. Gonzalez and A. Pascual, "Ozone Treatment of Methylene Blue on Aqueous Solution", Chem. Eng. Commun. **119**, 151-166 (1993).
+
+- Cooper, P., "Removing Colour from Dyehouse Wastewaters- a Critical Review of Technology Available", J. Soc. Dyes Col. **109**, 97-100 (1993).
+
+- Gahr, F., F. Hermanutz and W. Oppermann, "Ozonation-an important Technique for Comply with New German Laws for Textile Wastewater Treatment", Wat.Sci.Tech. **30**, 255-264 (1994).
+
+- Gould, I.P. and W.J. Weber, "Oxidation of Phenols by Ozone", J. Wat. Pollut. Control Fed. **48**, 47-53 (1976).
+
+- Groves, G. R., C. A. Buckley, R. H. Turnbull and K. Treffry-Goatley, "A Guide for the Planning, Design and Implementation of Wastewater Treatment Plants in the Textile Industry", Water Research Commission Protection, Pretoria, South Africa (1988).
+
+- Hazel, B. G., "Industry Evaluation of Colour Reduction and Removal-the DEMOS Project", Society of Dyes and Colorists, Bradford, UK. (1995), pp. 59-72.
+
+- Holme, J., "Cotton Dyeing and Finishing to 2000 and Beyond", Int. Dyer **182** (3), 32-34 (1997).
+
+- Hsu, Y.C., J.T. Chen and H.C. Yang, "Decolorization of Dyes Using Ozone in a New Gas-Induced Reactor", AIChE J. **47**, 169-176 (2001).
+
+- Perkins, W.S., W.K. Walsh, I.E. Reed and C.G. Namboodi, "A Demonstration of Reuse of Spent Dyebath Water Following Colour Removal With Ozone", Textile Chemist and Colorist **28**, 31-37 (1995).
+
+- Saunders, F.M., J.P. Gould and C.R. Southerland, "The Effect of Solute Composition on Ozonalysis of Industrial Dyes", Wat.Res. **17**, 1407-1415 (1983).
+
+- Sotelo, J., F. Beltran-Heredia and J.M. Encinar, "Azo Dye Ozonation Film Theory Utilization for Kinetic Studies", Ozone Sci.Eng. **11**, 391-398 (1989).
+
+- Staehelin, J. and J. Hoigne, "Decomposition of Ozone in Water in the Presence of Organic Solutes Acting as Promoters and Inhibitors of Radical Chain Reactions", Environ. Sci. Technol. **19**, 1206-1213 (1985).

manual_annotation/200pdfs_in_mmd/10.1002_clen.201100342.mmd

@@ -0,0 +1,129 @@
+# Degradation of Disperse Blue E-4R in Aqueous Solution by Zero-Valent Iron/Ozone
+
+S. L. L., M. M., M.
+
+Guolin Ozone Equipment Co., Ltd, Qingdao, China) was used to produce O3- ZVI (iron chippings) was purchased from Wuhan Air Compressor Co., Ltd. (China). The pretreatment of ZVI was conducted according to the method described in our previous publication [15]. Two different real wastewater samples mainly containing anthraquinone dyes were assessed in this study. One was collected in a dye factory (Shanghai Hope Dyeing & Printing Co., Ltd., Shanghai, China) without being treated, named as real wastewater A (RWA). The other, named RWB, was obtained from a textile mill (Lander Textile Co., Ltd., Foshan, China), after conventional activated sludge treatment. Both samples were filtered to remove suspended solids.
+
+### Procedure
+
+A total of 500-mL E-4R aqueous solutions (1 g/L) were placed in a quartz reactor (1000 mL) and ZVI was added and O3 was bubbled for 30 min to the solution. Twenty-five milliliters of samples were then withdrawn. In this experiment, we examined the effects of O3 dosage (50, 100, 150, 200, 250, and 300 L/h), ZVI dosage (100, 200, 250, 300, 350, and 400 g), temperature (20, 30, 40, 50, 60, 70, and 80degC), pH values (1.0, 3.0, 5.0, 7.0, 9.0, 11.0, and 13.0), and ZVI particle sizes (>2, 0.9-2, 0.45-0.9, 0.125-0.45, and 0.075-0.125 mm) on the removal of E-4R. In addition, a comparative experiment was executed using ZVI alone, O3 alone, and ZVI/O3 under the following conditions: 150 L/h O3, 300 g ZVI, 40degC, pH 11, and 0.9-2 mm ZVI particle size, respectively. In order to conduct the kinetic study, samples were withdrawn at proper intervals (0, 5, 10, 15, 20, 25, 30, 35, 40, 50, and 60 min) during the processes. In the wastewater test, 500 mL real wastewater samples were treated by ZVI/O3 for 3 h with the conditions of 150 L/h O3, 300 g ZVI, 40degC, and 0.9-2 mm ZVI particle size. The water qualities of samples before and after treatment were analyzed.
+
+### Analysis
+
+The analysis of color removal was carried out on a UV1100 spectrophotometer (Beijing Rayleigh Analytical Instrument Corp., Beijing, China), described in [1]. The data were shown as color removal percentage (CR%). COD was measured on a JH-12 COD analysis apparatus (Qingdao Laoshan Electron Instrument Corp., Qingdao, China), described in [15], and results were expressed as COD removal percentage (CODR%). A Multi N/C TOC Analyzer and a Multi X 2000 AOX Analyzer (Carl Zeiss, Jena, Germany) was used to measure TOC and AOX concentrations, respectively, according to the method described in [16]. The data were shown as TOC removal percentage (TORC%) and AOX removal percentage (AOX%), respectively. The pH value was measured on a pHS-4CT pH meter (Shanghai Candy Co., Ltd., Shanghai, China). The conductivity was measured on an HQ14d digital conductivity analyzer (Hach Company, Loveland, Colorado).
+
+## Results and discussion
+
+### Degradation efficiency
+
+The CR%, CODR%, TOCR%, and AOXR% data of E-4R solution treated by ZVI alone, O3 alone, and ZVI/O3 were shown in Tab. 2. As seen in Tab. 2, the degradation efficiency of ZVI/O3 was higher than ZVI or O3 alone. The CODR% value is approximately 70 and 50% and higher than that of ZVI and O3 alone, respectively. During the internal electrolysis of ZVI process, a large amount of nascent hydrogen was produced and the species reductively degraded the organics, with the production of Fe2+. On the other hand, in the ozonation process, O3 not only oxidized the organics by itself, but also produced strong oxidative species, such as H2O2 and hydroxyl radical (*OH), to degrade organics. When the internal electrolysis and ozonation were carried out simultaneously, Fe2+, H2O2, and *OH would form Fenton's reagent, which could enhance the oxidation. Therefore, in the ZVI/O3 process, the synergistic effect of internal electrolysis and ozonation would accelerate the degradation of organics and lead to the great improvement of CODR% and TOCR%.
+
+\begin{table}
+\begin{tabular}{c c c c c c} Iron form & Applied system & Mechanism & Pollutant & Operational parameter & Reference \\ Fe(VI) & Fe(VI)H2O2 & Oxidation, coagulation, and Fenton process & Trifluraline & pH 3, 6 g/L Fe(VI), 12 g/L H2O2, 40°C & [8] \\ Fe(III) & Fe(III)-TAM/LH2O2 & Catalyzed oxidation & Textile dyes & pH 11, 10−4 M Fe(III)-TAM, 10−2 M H2O2, 25°C & [9] \\ Fe(II) & Fe(II)/H2O2 & Fenton process & Azo dye & pH 3.5, 0.27 mM Fe2+, 8.8 mM H2O2, 70°C & [10] \\ Fe(0) & Fe(0)/sand & Adsorption, and co-precipitation & Household water & [11, 12] \\  & Fe(0)/AC & Internal electrolysis & Textile dyes & pH 4, 12.0 g Fe(0), 2.3 g AC, 22°C & [11, 12] \\ \end{tabular}
+\end{table}
+Table 1: The application of iron in the water/wastewater treatments
+
+\begin{table}
+\begin{tabular}{c c c c c}  & CR\% & CODR\% & TOCR\% & AOXR\% \\ ZVI & 10.64 & 5.48 & 4.62 & 26.46 \\ O3 & 97.56 & 26.88 & 14.32 & 36.44 \\ ZVI/O3 & 99.97 & 74.51 & 43.85 & 50.60 \\ \end{tabular}
+\end{table}
+Table 2: Degradation efficiencies of different approaches
+
+Figure 1: Chemical structure of E-4R.
+
+In addition, the generation of AOX, some of which are thought to be dangerous to humans, during the wastewater treatment is a concerned problem [17]. Because there is a bromine atom connecting to the dye molecule (Fig. 1), we also evaluated the AOX concentration of E4R solution before and after the treatment. As shown in Tab. 2, the AOX6% of ZVI alone, \(\text{O}_{3}\) alone, and \(\text{ZVI}\lbrack\text{O}_{3}\) were approximately 26, 36, and 50%, respectively. In the single internal electrolysis, the low CR% and AOX6% indicated that only small amount of dye molecules were degraded and the efficiency was very low. In ozonation alone, the CR% was up to approximately 97% and the treated solution was nearly colorless, however, the AOX6% was only approximately 36%. It indicated that during the single ozonation process, the chromophoric groups of dye molecules were broken, but the cleavage of the ring and further mineralization did not fully carry through. However, during the ZVI/\(\text{O}_{3}\) process, the synergistic effect of internal electrolysis and ozonation not only broke the chromophoric groups of the dye molecules, but also carried out the ring cleavage and mineralization, leading to the increase of TOCR% and AOX6%. Therefore, ZVI/\(\text{O}_{3}\) could degrade E4-R more completely.
+
+### Degradation kinetics
+
+In order to study the degradation rates, we carried out the pseudo-first-order model, which can be presented as follows:
+
+\[\ln\frac{C_{\text{t}}}{C_{0}} = - k_{1}t\]
+
+where \(C_{\text{t}}\) is the concentration of E4-R at instant \(t\) (mg/L), \(C_{0}\) the initial E4-R concentration (mg/L), \(k_{1}\) the pseudo-first-order rate constant (min-1), and \(t\) is the time of reaction (min). Low concentration of E4-R has direct ratio relations with the absorbance according to the Lambert-Beer law. Therefore, Eq. (1) could also be calculated as:
+
+\[\ln\frac{C_{\text{t}}}{C_{0}} = \ln\frac{A_{\text{t}}}{A_{0}} = - k_{1}t\]
+
+where \(A_{0}\) (Abs) and \(A_{\text{t}}\) were the absorbance of the initial E4-R solution and the E4-R solution treated for \(t\) min at 565 nm, respectively. As shown in Fig. 2, the ZVI/\(\text{O}_{3}\) process was well suitable to pseudo-first-order model (\(R^{2} = 0.9887\)) and the pseudo-first-order rate constant was 0.0605 min-1, which was higher than that of Fe(0)/AC system [14]. However, the ZVI and \(\text{O}_{3}\) processes do not perfectly fit for pseudo-first-order and their rate constants were lower than that of the ZVI/\(\text{O}_{3}\) process. Some other work also showed that the combined iron system fit for pseudo-first-order model, such as Fe(III)-TAMI/H2O2 [9] and Fe(0)/AC [14].
+
+### Effect of \(\text{O}_{3}\) dosage
+
+In order to study the effect of \(\text{O}_{3}\) dose, different \(\text{O}_{3}\) dosages were used under the following conditions: 300 g ZVI, 40 degC, pH 11, and 0.9-2 mm ZVI particle size. As shown in Fig. 3(a), increasing the \(\text{O}_{3}\) dosage could promote increases in CR%, CODR%, and TOCR%. Increased \(\text{O}_{3}\) consumption is expected to be responsible for elevated generation of oxidative species [18]. In addition, higher \(\text{O}_{3}\) concentration in the air bubbles was favorable for the solubility of \(\text{O}_{3}\) in dye solution, which resulted in a higher rate of mass transfer [19]. However, at a certain temperature the solubility of \(\text{O}_{3}\) was definite. The CR%, CODR%, and TOCR% would approach a maximum point with the continuous increase of \(\text{O}_{3}\) doses. The results showed that the optimal \(\text{O}_{3}\) dosage was 150 l/h, and at this dosage, the CR%, CODR%, and TOCR% were up to approximately 100, 74, and 43%, respectively.
+
+### Effect of ZVI dosage
+
+To study the dose dependent effect of ZVI on CR%, CODR%, and TOCR%, the reaction ran with the condition of 150 l/h \(\text{O}_{3}\), 40 degC, pH 11, and 0.9-2 mm ZVI particle size. As shown in Fig. 3(b), the CR% was nearly unchanged with increasing ZVI dosage. However, the CODR% and TOCR% curve gradually rose with the increase of ZVI dosage from 100 to 300 g/L. However, the CODR% and TOCR% gradually decreased when the dose was >300 g/L. This was possible because the excessive ZVI limited the contact area between wastewater and ZVI and wastewater and \(\text{O}_{3}\), resulting in the inadequate utilization of ZVI and \(\text{O}_{3}\). Therefore, given the efficiency and economy, the appropriate ZVI dosage was 300 g/L.
+
+The effects of reagents in some combined iron system are more complex. Taking the Fe(VI)/H2O2 system, e.g., the contour plots of both the Fe(VI) and H2O2 concentrations were "U" shapes [8].
+
+### Effect of temperature
+
+Under different temperatures, the batch experiments of degradation of E4-R aqueous solution by ZVI/\(\text{O}_{3}\) were carried out at the conditions of 150 l/h \(\text{O}_{3}\), 300 g ZVI, pH 11, and 0.9-2 mm ZVI particle size and the result was presented in Fig. 3(c). The CR% could constantly maintain >98% at different temperatures, which clearly indicated that temperature had little effect on the degradation efficiency of E4-R treated by ZVI/\(\text{O}_{3}\).An increase of the temperature from 20 to 40 degC led to an increase of CODR% and TOCR%. Within a small temperature range, temperature has little effect on the activation energy. With a temperature increase, both mass transfer of different species and internal energy of reactive substances would increase, which would be favorable for the elevation of degradation efficiency [20]. However, when >40 degC, the CODR% and TOCR% gradually decreased, because \(\text{O}_{3}\) would decompose quickly and escape from the reaction system at a high temperature. Therefore, the optimal temperature was 40degC, which was similar with that of the Fe(III)-TAMI/H2O2 system [9], and lower than the optimal values of Fe(VI)/H2O2 and Fe(II)/H2O2 systems [8, 10].
+
+Figure 2: Pseudo-first-order degradation of E4-R. Symbols: (•) ZVI, (•) \(\text{O}_{3}\), and (•) ZVI/\(\text{O}_{3}\).
+
+### Effect of pH value
+
+The actual dye wastewaters are acid or alkaline because of the different properties of dyes. However, pH is always an important parameter affecting the treatment process. To study the relationship between degradation efficiency and pH, we treated E-4R solution of various initial pH values under the conditions of 150 l/h O3, 300 g ZVI, 40degC, and 0.9-2 mm ZVI particle size. As shown in Fig. 3(d), higher pH level was favorable for higher degradation efficiency. At a pH of approximately 11, highest values of CR%, CODR%, and TOCR% were obtained. Generally speaking, a high pH value has a promotion to zonation [21]. Moreover, under alkaline and aerobic condition, Fe2+ can form ferrous hydroxide (Fe(OH)2) and ferric hydroxide (Fe(OH)2), see Eqs. (3) and (4):
+
+\[\text{Fe}^{2 + } + 2\,\text{OH}^{ - } \rightarrow \text{Fe}(\text{OH})_{2}\]
+
+\[\text{Fe}^{2 + } + 8\,\text{OH}^{ - } + \text{O}_{2} + 2\,\text{H}_{2}\text{O} \rightarrow 4\,\text{Fe}(\text{OH})_{3}\]
+
+The formed Fe(OH)2 is a kind of coagulant with strong adsorption capability and could adsorb the dye molecules. However, at pH 13, a slight drop of degradation efficiency was observed. This possibly related to the production of large quantities of gel materials.
+
+Other combined iron systems also exhibited highest efficiency at higher pH value. The optimal pH value for Fe(III)TAML[H2O2 was 11.0 [9]. However, the effect of pH value on Fe(VI)2H2O2 system is different. At low pH value higher degradation efficiency was achieved. This is due to the fact that at a low pH the formation of Fe(III) hydroxidecomounds, such as [FeIII(OH)2+, [FeIII(OH)2+ occurs, which catalyzes the H2O2 decomposition, generating *OH [8]. Fe(II)2H2O2 and Fe(0)/AC systems also favored a lower pH value [10, 14].
+
+### Effect of ZVI particle size
+
+ZVI of different particle sizes were used to investigate the effect of the particle size at the conditions of 150 l/h O3, 300 g ZVI, 40degC, and pH 11 and 0.9-2 mm ZVI and the results were summarized in Fig. 4. The optimal particle size was 0.9-2 mm, and below or over this size would decrease the degradation efficiency. In general, the smaller particle sizes could form more microbatteries, and were beneficial to the reaction. However, in fact, too small particles easily polymerized, especially in the presence of O3, resulting to the reduction of the contact areas. Therefore, 0.9-2 mm was appropriate ZVI size in the experiment.
+
+### Real wastewater test
+
+From the preliminary results of simulated wastewater, ZVI/O3 treatment showed significant degradation efficiency. In order to evaluate
+
+Figure 4: Effect of ZVI particle size.
+
+Figure 3: Effect of operational parameters: (a) O3 dosage, (b) ZVI dosage, (c) temperature, and (d) pH value. Symbols: (•) CR%, (•) CODR%, and TOCR%.
+
+the efficiency of ZVI/O3 to treat the real wastewater, RWA and RWB were assessed. The results were presented in Tab. 3.
+
+RWA was a raw effluent without being treated. Consequently, this wastewater was strongly colored and presented a relatively high organic load. RWB was pretreated by activated sludge and the color and organics contents were relatively low. After ZVI/O3 treatment for 3 h, both the pH value of RWA and RWB decreased from weak alkali to neutrality. And the conductivity of both samples significantly increased. However, the degradation efficiencies of ZVI/O3 for the two samples were different. The efficiency for RWB was much higher than that for RWA. For example, the CR%, CODR%, TOCR%, and AOXR% for RWA were 63, 32, 28, and 31%, respectively. Comparatively, these values for RWB were 98, 66, 42, and 46%, respectively.
+
+According to the present results, ZVI/O3 was not suitable to be used as a main wastewater treatment. Conversely, this method for post-treatment of wastewater was of great interest. In the case of wastewater pretreated by activated sludge, ZVI/O3 had shown high efficiency for decolorization and mineralization.
+
+## 4 Concluding remarks
+
+An efficient and convenient procedure for the degradation of E-4R by ZVI/O3 was proposed in this study. This method showed higher degradation efficiency, compared with ZVI or O3 alone. We also discussed the effect of some operational parameters. Through the real wastewater test, ZVI/O3 showed high efficiency for the treatment of wastewater pretreated by conventional activated sludge. Therefore, ZVI/O3 is a promising method to be applied and integrated in the treatment process of dye wastewaters.
+
+## Acknowledgments
+
+The work was financially supported by Hubei Provincial Department of Education Science and Technology Research Planning projects (B class project) [No. B20111608].
+
+_The authors have declared no conflict of interest_.
+
+## References
+
+* (1) J. Fu, Y. H. Ding, G. Y. Ma, J. Yang, Q. F. Zeng, M. Y. Liu, D. S. Xia, et al., Removal of a Toxic Anthraquinone Dye by Combination of Red Mud Coagulation and Ozonation, _Ozone Sci. Eng._**2009**, 31, 294.
+* (2) M. Xu, J. Guo, G. Zeng, X. Zhong, G. Sun, Decolorization of Anthraquinone Dye by _Shenwantella decolorations_ S12, _Appl. Microbiol. Biotechnol._**2005**, 17, 246.
+* (3) C. Novotny, N. Dias, A. Kapanen, K. Malachova, M. Vandrovcova, M. Itivazara, N. Lima, Comparative Use of Bacterial, Algal and Protozoan Tests to Study Toxicity of Azo- and Anthraquinone Dyes, _Chemosphere_**2006**, 63, 1436.
+* (4) D. Sugimori, R. Banzawa, M. Kurozumi, I. Okura, Removal of Disperse Dyes by the Fungus _Cunninghamella polymorpha_, _J. Biosci. Bioeng._**1999**, 87, 252.
+* (5) X. R. Xu, H. B. Li, W. H. Wang, J. D. Gu, Degradation of Dyes in Aqueous Solutions by the Fenton Process, _Chemosphere_**2004**, 57, 595.
+* (6) C. L. Yang, J. McGrahan, Electrochemical Coagulation for Textile Effluent Decolorization, _J. Hazard. Mater._**2005**, 127, 40.
+* (7) M. M. Karim, A. K. Das, S. H. Lee, Treatment of Colored Effluent of the Textile Industry in Bangladesh Using Zinc Chloride Treated Indigenous Activated Carbons, _Anal. Chim. Acta_**2006**, 576, 37.
+* Soil Air Water_**2007**, 35, 88.
+* Soil Air Water_**2007**, 35, 459.
+* (10) C. S. D. Rodrigues, I. M. Madeira, R. A. R. Boaventura, Optimization of the Azo Dye Proccion Red H-EXT Degradation by Fenton's Reagent Using Experimental Design, _J. Hazard. Mater._**2009**, 164, 987.
+* Soil Air Water_**2009**, 37, 930.
+* Soil Air Water_**2010**, 38, 951.
+* (13) Z. M. Shen, W. H. Wang, J. P. Jia, J. C. Ye, X. Feng, A. Peng, Degradation of Dye Solution by an Activated Carbon Fiber Electrode Electrophysiology System, _J. Hazard. Mater._**2001**, 84, 107.
+* (14) H. N. Liu, G. T. Li, J. H. Qu, H. J. Liu, Degradation of Azo Dye Acid Orange 7 in Water by Fe\({}^{0}\)/Granular Activated Carbon System in the Presence of Ultrasound, _J. Hazard. Mater._**2007**, 144, 180.
+* (15) J. Fu, X. Zhen, Q. S. Li, S. Chen, S. Q. An, Q. F. Zeng, H. L. Zhu, Treatment of Simulated Wastewater Containing Reactive Red 195 by Zero-Valent Iron/Activated Carbon Combined with Microwave Discharge Electrodeless Lamp/Sodium Hypochlorite, _J. Environ. Sci._**2010**, 22, 512.
+* (16) J. Fu, G. Chen, Y. Yang, Z. M. Zhang, Q. F. Zeng, S. Q, An, H. L. Zhu, Ultraviolet Irradiation Combined with Manganese Ore Catalyzed Ozonation of 4-Chlorophenol in Aqueous Solution, _Water Sci. Technol. Water Suppl_**2010**, 10, 97.
+* (17) D. T. Sponza, Application of Toxicity Tests into Discharges of the Pulp-Paper Industry in Turkey, _Eotoxicol. Environ. Sgr._**2003**, 54, 74.
+
+\begin{table}
+\begin{tabular}{l r r r r} \hline  & \multicolumn{2}{c}{RWA} & \multicolumn{2}{c}{RWB} \\ \cline{2-5}  & Before treatment & After treatment & Before treatment & After treatment \\ \hline pH & 9.3 & 8.8 & 8.5 & 7.8 \\ Conductivity (μS/cm) & 1495 & 2287 & 2371 & 5875 \\ Color (μA\%G\_A\%S2S, and \(A_{620}\)) & 0.523, 0.348, and 0.465 & 0.211, 0.156, and 0.127 & 0.142, 0.129, and 0.061 & 0.030, 0.019, and 0.004 \\ COD (mg/L) & 565 & 385 & 127 & 43 \\ TOC (mg/L) & 275 & 198 & 78 & 45 \\ AOX (mg/L) & 1.24 & 0.85 & 0.91 & 0.49 \\ \hline \end{tabular}
+\end{table}
+Table 3: Parameters of RWA and RWB before and after treatment by ZVI/O3* [18] Z. Q, He, L. L. Lin, S. Song, M. Xia, L. J. Xu, H. P. Ying, J. M. Chen, Mineralization of C.I. Reactive Blue 19 by Ozonation Combined with Sonolysis: Performance Optimization and Degradation Mechanism. Sep. Purf. Technol.**2008**, _62_, 376.
+* [19] M. F. Sevincli, H. Z. Sarikaya, Ozone Treatment of Textile Effluents and Dyes: Effect of Applied Ozone Dose, pH and Dye Concentration. J. Chem. Technol. Biotechnol.**2002**, _77_, 842.
+* [20] M. Li, J. T. Li, H. W. Sun, Decolorizing of Azo Dye Reactive Red 24 Aqueous Solution Using Exfoliated Graphite and H\({}_{2}\)O\({}_{2}\) under Ultrasound Irradiation. Ultrason. Sonochem.**2008**, _15_, 717.
+* [21] K. Sehested, H. Corritzen, J. Holcman, C. H. Fischer, E. J. Hart, The Primary Reaction in the Decomposition of Ozone in Acidic Aqueous-Solutions. _Environ. Sci. Technol._**1991**, _25_, 1589.

manual_annotation/200pdfs_in_mmd/10.1002_ep.12306.mmd

@@ -0,0 +1,280 @@
+# Removal of Emerging Pharmaceuticals from Wastewater by Ozone-Based Advanced Oxidation Processes
+
+Fares A. Almomani, Mooryad Showaqfah, Rahul R. Bhosale, and Anand Kumar
+
+# Abstract
+
+This study investigates the use of ozone as a pretreatment process for water containing pharmaceuticals. Experiments were carried out on synthetic wastewater, surface water, and the effluent of wastewater treatment plant. The degradation efficiencies of four groups of pharmaceuticals (antibodies, estrogens, acidic, and neutral) were studied, and the effect of ozone dose and pH on the degradation efficiency was monitored. A Microx20 submission_b_ was test was used to evaluate the change in the toxicity of aqueous solutions before and after oxonation. The efficiency of oxidation of antibiotics, estrogens, and neutral pharmaceuticals increased as the ozone dose and pH increased. Ozone input dose of 108.1, 222.3, and 222.4 mg b-1-1 was found to be optimum yielding the bighest oxidation efficiency for the studied pharmaceuticals in synthetic wastewater, surface water and effluent of wastewater treatment plant, respectively. An average specific ozone dose of 2.05 for antibiotics, 1.11 for estrogens, and 1.30 mg O_/mg OOC for neutral pharmaceuticals reduced significantly the acute toxicity of the water solutions and mineralized more than 40%, 33%, and 23% of DOC in less than 1 min. The kinetics of ozone with pharmaceuticals was modeled for synthetic wastewater as an overall second-order reaction with a rate constant ranging from 103 to 106 M-13 s-1. The results indicate the effectiveness of ozone-based advanced oxidation processes in removing emerging pharmaceuticals from water and wastewater. The results showed that oxonation process is more effective than other conventional oxidation processes (CI2 and CIO2) in eliminating pharmaceuticals and reducing the toxicity of the effluent water or wastewater. 2016 American Institute of Chemical Engineers Environ Prog. 35; 982-995, 2016
+
+## Introduction
+
+The occurrence of pharmaceuticals and personal care products (PCP) in the aquatic environment threatens the purity of water [1-4], and has raised increasing concerns over the last 10 years. Pharmaceuticals are detected in surface water because they are continuously released into this ecosystem from different point sources (such as households, hospitals, and pharmaceutical industries) [5]. There are also indications that the contamination of groundwater with pharmaceuticals can be the result of human activities, such as the dumping of sludge or manure containing these pharmaceuticals in very high concentrations (50-100 mg L-1) [6,7], the migration of contaminated surface water through riverbank filtration into the groundwater [8], and artificial groundwater being recharges with contaminated water [2,3,9]. A literature review indicates that the concentration of pharmaceuticals in different water bodies can range from as low as ng/L to mg/L [10-13]. Verlichci _et al._ [14] reported that the concentration of different pharmaceuticals in raw wastewaters was in the range of 10-3 to 10-6 mg L-1. Bhandari _et al._ [15] measured average aqueous phase concentrations of ciprofloxacin, sulfinethoxazole, and azithromycin in raw wastewater at 1.44, 1.11, and 18.3 mg L-1, respectively, while the concentrations in the WWTP effluent were found to be 0.59, 1.23, and 3.25 mg/L, respectively. Raw wastewater from three hospitals in Jordan were measured for some pharmaceuticals, showing concentrations on the order of 102 mg/L to 103 mg/L for antibiotics, estrogens, acidic, and neutral pharmaceuticals.
+
+Conventional treatment processes for removing pharmaceuticals are ineffective, and several pharmaceuticals have been detected in the secondary treatment effluents of municipal wastewater treatment plants (WWTPs) [16]. As a result, there is an urgent need to develop an effective advanced treatment processes to remove emerging pharmaceuticals from conventional WWTP effluent before its final discharge into the environment.
+
+Different treatment processes, including activated carbon [1,17], nanofiltration [18,19], and reverse osmosis [20] have been reported as effective treatment options for removing different types of pharmaceuticals. However, these treatment processes are considered temporary solutions, as they only move or transfer the pharmaceuticals to the solid phase or concentrate them in a small volume of aqueous solution, which then requires further treatment. Chlorination processes are used to remove different pharmaceuticals from synthetic wastewater on a laboratory-scale setup under conditions similar to actual drinking water treatment processes. However, the removal efficiency for some pharmaceuticals did not exceed 30% over 30 min and required at least 24 h to achieve 80% removal.
+
+Ozone has been considered a promising technology for the effective treatment of different organic contaminants in water and wastewater [21-23]. The reaction of ozone with these contaminates has been shown to occur by two pathways; direct attack of ozone on the acidic sites, and indirect action through a variety of high-power oxidative radicals which can act as secondary oxidants [23,24]. The direct ozone attack is followed by an electrophilic aromatic substitution and undergoes very selective reactions [25]. The second oxidation pathway is constituted by the action of radical species generated from the ozone decomposition, mainly hydroxyl radicals, which leads to very fast reactions with organic compounds. The second oxidation pathway is less selective than the first pathway [25-27].
+
+Although ozone can be effective in degrading different types of pharmaceuticals [28,29], the mineralization of these pharmaceuticals by ozone is limited [30], and considerable amounts of oxidative byproducts may actually be more toxic than the original compounds. To minimise the threat of degradation intermediates of pharmaceuticals, it is desirable to examine the toxicity of the water before and after the ozone process, in order to determine the effect of the process on the quality of effluent wastewater. The present work aims to evaluate the efficiency of the ozonation process in removing and degrading pharmaceuticals from synthetic wastewater (SWW), surface water (SFW), and effluents of municipal wastewater treatment plant (WRTP). In the first part of this work, the efficiency of the ozonation process in degrading selected pharmaceuticals, the effect of the initial ozone dose, and the ozone reaction rate constants associated with these pharmaceuticals are investigated and determined in the case of synthetic wastewater. The change in the toxicity of the synthetic wastewater following ozonation treatment is measured and discussed, and a comparison between the reaction rate constant of ozone with other common oxidant, such as chlorine and chlorine dioxide, is provided. The second part investigates the potential performance of the ozonation reaction in the treatment of pharmaceuticals found in surface water and municipal WWTP effluents.
+
+## Materials and Methods
+
+### Chemicals
+
+The chemicals used during the experiment include potassium phosphate (Fisher Scientific), sodium phosphate monobasic (Fisher Scientific), sodium hydroxide (NaOH), phosphoric acid (Fisher Scientific), and acetonitrile (Fisher Scientific).
+
+### Experimental Setup
+
+The experimental set up used to carry out the ozonation experiments is presented in Figure 1. The setup includes an ozone system consisting of (1) an ozone generator, (2) an ozone analyzer, and (3) a gas-flow controller, as well as a reactor system consisting of (4) a magnetic stirrer, (5) a diffuser, (6) a thermometer, (7) a pH meter, and (8) an auto-sampler, finally, there is the effluent gas system, consisting of (9) a first gas-washing trap (4% KI), and (10) a second gas-washing trap (20% KI). The temperature of the aqueous solution inside the reactor is controlled at preset values (20degC) by circulating thermostatic water inside the jacket (Stream #12 in Figure 1).
+
+### Analytical Methods
+
+#### Measuring Inlet, Outlet, and Residual Ozone Concentrations
+
+Inlet and outlet gaseous ozone concentrations were determined using an ozone analyzer (Ozomat GM-6000-OEM), calibrated using the indigo method. The detection limit of ozone was found to be \(\pm\)0.05 mg/L. Residual ozone in the aqueous solution was determined by means of the indigo method (_Standard Methods_) [31].
+
+#### Measuring of Pharmaceutical Concentration
+
+The concentration of pharmaceuticals in samples was determined by high performance liquid chromatography (HPLC) using a Hewlett-Packard 1050 series HPLC equipped with an Ultra Aqueous C18 column from Restek (3.2 x 100 mm, particle size 5 mm, 80 A'). The injection volume was 100 ml. The mobile phase consisted of a binary mixture of the following solvents. (A) acidified water with 10 mM phosphoric acid and (B) pure acetonitrile. A 30 min linear gradient from 10% to 100% of B with a flow rate of 0.5 mL min-1 mobile phase gradient program was used. The separation of pharmaceuticals was monitored at the wavelength corresponding to the maximum absorbance. The detection limit of the device was determined to be 10 mg L-1 using standard solutions. Measurements were carried out in triplicate and reported as an average value at 95% confidence intervals.
+
+#### Measurement of Dissolved Organic Carbon
+
+The dissolved organic carbon (DOC) in the water samples was measured using a Shimadzu TOC analyzer with water samples that were prefiltered using a 0.45 mm filter. Measurements were carried out in triplicate and reported as an average value at 95% confidence intervals. The detection limit of the TOC analyzer was determined to be in the range of \(\pm\)0.1 mg C L-1 using potassium phthalate standard solution. Quality control experiments showed that the filtration step has no effect on the values of measured DOC.
+
+### Reaction Kinetics
+
+The change in the concentration of pharmaceuticals throughout the ozonation experiments was measured as a function of reaction time. Each kinetic experiment was repeated at least five times and averaged to fit the kinetic model at a 95% confidence level.
+
+Ozonation experiments were carried out using different molar ratios of ozone to pharmaceuticals. The selected molar ratios allow the determination of the ozonation reactions' overall kinetic constants. Kinetic experiments were carried out under both acidic and alkaline conditions.
+
+Figure 1: Experimental setup. [Color figure can be viewed in the online issue, which is available at **wileyonlinelibrary.com**.]
+
+### Studied Pharmaceuticals
+
+Table 1 presents the name, chemical formula, initial concentration of pharmaceuticals used in oxzonation experiments and the initial concentration of these pharmaceuticals in different water samples as published in literature. All the studied pharmaceuticals were obtained from Sigma-Aldrich with a purity of >98%. The studied pharmaceutical are antibiotics, estrogens, acidic pharmaceuticals, and neutral pharmaceuticals.
+
+### Water Samples
+
+Synthetic wastewater was prepared by dissolving 12 g of Na2HPO4+Y4H2O and 0.8 g of KH2PO4 in a volume of 2 L ozone-demand-free water. This recipe produces a wastewater with a pH value of around 7.8 +- 0.2. For experiments carried out at different pH values, the pH of the synthetic wastewater was regulated using 5 N NaOH or 1 M H2PO4.
+
+Water samples (synthetic (SWW), surface (SFW), and wastewater (WWTP)) were spiked with the selected pharmaceuticals to give the resulting wastewater specific concentrations in the range of 3.7 to 9.2 mg L-1 (5-40 mM) This concentration, although higher than the actual values found in raw wastewater (which often reach as much as 102 mg L-1 to 102 mg L-1), allowed accurate and fast quantitative chemical analysis with our available equipment.
+
+Local surface water (SW) was collected from Wadi El-Arab Dam, Jordan, Erfluent of municipal wastewater treatment plant (WWTP) was collected from the effluent of a secondary treatment train from a WWTP. Jordan. Water samples were filtered through a 0.45 mm Millipore filter, and stored at 4degC prior to use. Table 2 shows the characteristics of both SW and WW.
+
+### Experimental Procedure
+
+For each experiment, 2 L of aqueous solution was prepared by spiking the aqueous solution (synthetic, surface, or wastewater effluent) with one of the pharmaceuticals. The initial concentration of pharmaceuticals used in each experiment is shown in Table 1.
+
+All experiments were carried out at constant temperature (20 +- 0.5degC) and at different solution pHs of 4, 8, and 11. The prepared aqueous solution was charged in the reactor and mixed for 10 min. The ozone generator was then switched on and left for 10 min to reach steady-state ozone production. After that, the gas flow controller (Alicat Scientific), set to 0.57 L min-1, was used to continuously feed the inlet ozone gas to the reactor, being bubbled through a porous ceramic diffuser placed in the center lower half of the reactor. The ozone analyzer was used to measure the concentration of ozone in the gas stream before infecting it into the reactor. The outlet gas stream was also analyzed for residual ozone using the ozone analyzer. Ozone concentration in the water samples was analyzed using the indigo method according to the procedure in _Standard Methods_ [31]. To maintain constant pressure inside the reactor, the outlet gas line was discharged to the atmosphere after measuring the outlet concentration of ozone. The ozone in the effluent gas was trapped in two stages (with KI solutions). The autosampler was programmed to withdraw water samples at various reaction times (every 30 s) in order to measure the concentration of the pharmaceuticals.
+
+### Procedure to Determine the Reaction Requirement
+
+Reaction requirements were determine in batch experiments according to the following procedure: water solution spiked with a constant concentration of each pharmaceuticals was mixed manually with ozone at different ozone to pharmaceutical molar ration; 3:1, 5:1, and 10:1, and the solution was left to react for 10 s, 15 s, and 30 s. Then, the solution was analyzed for residual ozone and pharmaceutical concentration. The initial concentrations of all pharmaceuticals (antibiotics, estrogens, acidic, and neutral) in these experiments were fixed at 1 uM, and the initial ozone concentration was varied from 20.8 to 93.8 uM. These experiments were repeated several times (at least five times) at each experimental condition. The reaction requirements were calculated according to Eq. 1:
+
+\[\text{Reaction~requirements}\ (z) = \frac{\lbrack\text{Pharm}\rbrack_{\text{s}} - \lbrack\text{Pharm}\rbrack_{\text{f}}\rbrack_{\text{s}}}{\lbrack\text{O}_{\text{s}}\rbrack_{\text{s}}- \lbrack\text{O}_{\text{s}}\rbrack_{\text{f}}}\]
+
+### Toxicity Analysis
+
+A bioassay was used to evaluate the toxicity of the aqueous solution before and after oxzonation treatment. The test used the marine bacterium _Pbobacterium pbobacterium_ and followed the manufacturer's protocol. The procedure used for the toxicity test is outlined in our previous work [32]. MicroTM Omni software (SDI, 2002) was used to analyze the data at 5 min from the time of mixing the sample with the marine bacterium. The toxicity unit was correlated to the EC50 (concentration of tested compound that inhibits the growth of 50% of bacteria measured as light emission) according to Eq. 2:
+
+\[\text{TU}\left( 100/\text{EC}_{50} \right) = \frac{100}{\text{EC}_{50}}\]
+
+## Results and Discussion
+
+### Oxzonation of Pharmaceuticals in Synthetic Wastewater
+
+#### Oxidation of Pharmaceuticals with Ozone
+
+Preliminary experiments were carried out to determine the required ozone concentration for the oxidation of the pharmaceuticals. The required ozone concentration was judged on the basis of how much inlet ozone concentration was required to eliminate more than 90% of the pharmaceuticals in a 2-min batch experiment. The statistical analysis of 18 experiments carried out with synthetic wastewater showed that the average inlet ozone concentration of 5.5 mg L-1 is the optimum value. The ozonation experiments in synthetic wastewater thus carried out using this concentration.
+
+Figure 2A illustrates the development of antibiotic concentration as a function of ozonation reaction time carried out at 20degC and a pH of 8. The results in Figure 2A are the average of five repeated experiment at a 95% confidence level. The experiments were carried out with specific ozone dose in the range of 2.24 to 2.55 mg O/mg DOC, and with an initial ozone concentrations of 5.5 mg L-1 and 0.57 L min-1 which leads to an ozone input of 188.1 mg h-1. The structure of the studied antibiotics can be seen in Figure 2B. The oxidation reaction of the antibiotics using ozone was very fast; the concentration of antibiotics decreased progressively as a result of the reaction with ozone. A 50% reduction in the concentration of most of the antibiotics was achieved over 15 s of ozone reaction. The fast degradation of the antibiotics suggests that the reactivity of these antibiotics with the oxzonation reaction is very high. It is well known that the pH of the solution controls ozone chemistry in the water since the increase in the pH of the solution promotes ozone decomposition into hydroxyl radicals that are more reactive and less selective than the molecular ozone. At pH >7, the rate of ozone decomposition increases significantly to form *OH radicals. Thus, the experiments carried out at pH = 8 are useful to explore the contributions of *OH radicals andozone in the degradation of antibiotics. It is believed that the chemical structure of the antibiotics--which contains aromatics, long chain hydrocarbons, and high electron-density groups (e.g., phenoxide ions and tertiary amines)--determines the mechanism of the oxidation reaction (see Figure 2B). In the present case, with antibiotics, it is expected that tertiary amines react with ozone and/or \({}^{\bullet}\)OH radicals by forming either aminoxide or singlet dioxygen. In addition,
+
+\begin{table}
+\begin{tabular}{l c c c c} \hline  & & \multicolumn{2}{c}{**Used**} & & \\
+\({}^{\star}\)OH can attack and breakdown of cyclic functional groups of these antibiotics. These products give rise to a competitive reaction between O-transfer and electron transfer. The latter favors the production of amine radical cations, an ozonide radical, and hydroxyl radicals. The net result is a progressive oxidation reaction and further degradation of the antibiotic. During these experiments, antibiotics were oxidized to a very low concentration (the detection limit was 10 mg L-1) in less than 60 s using an inlet ozone concentration of 5.5 mg L-1 (corresponding to 2.24-25 mg O3/mg DOC). The amount of ozone used during the oxidation reaction was calculated by measuring the inlet and outlet ozone concentrations in the gas streams as well as the concentration of ozone in the aqueous according to Eq. 3:
+
+\[\nabla = \frac{\left\{ {Q_{\text{Gas}} \ast \left( {\left[ {\text{O}} \right\rbrack_{\text{gas,n}} - \left[ {\text{O}} \right\rbrack_{\text{gas,cat}}} \right]\Delta t} \right\} - V_{\text{water}} \ast \left[ {\text{O}} \right\rbrack_{\text{water}}}\]
+
+The amounts of ozone used to degrade the antibiotics (V) was found to range from 4.60 to 4.95 mg L-1 (azithromycin; 2.02 mg O3/mg DOC, roxinhydromycin; 1.91 mg O3/mg DOC, sulfamethoxazole; 2.02 mg V0/mg and sulfinazole; 2.27 mg O3/mg DOC), with an average value of 4.8 mg L-1 (2.05 mg V0/mg) and a standard deviation of 0.6. Although the ozonation process was very effective in degradation the pharmaceuticals, the partial oxidation of those molecules leads to the formation of byproducts which as well consume ozone. Thus, it is important to correlate the used ozone dose to the DOC mineralization as will be discussed later in this article.
+
+The contribution of ozone and hydroxyl radicals in oxidation the antibiotics was determined by repeating the previous experiment at lower pH of 4 and in the presence of tert-butanol and carbonate (well-known radical scavengers). In contrast the removals efficiencies of ozonation reaction in degradation of the antibiotics was very sensitive to the presence of tert-butanol and carbonate. Suggesting that the oxidation of antibiotics promotes a radical pathway. Martins _et al._[38] have studied the oxidation of sulfamethoxazole in single and catalyzed ozonation processes, an ozone dose input of 188.1 mg h-1 effectively lead to depletion of up to 90% of this pharmaceutical in 30 min reaction time. The study of Martins _et al._[38] and Sagi _et al._[39] confirmed that the ozonation of sulfamethoxazole promote radical pathway.
+
+Another sets of experiments were carried out at pH = 11, without _tert_-butanol and carbonate. The overall rate of oxidation (i.e., degradation) of the antibiotics at this pH was very high in comparison with the results obtained at pH 8 and 4. The fact that the ozonation of pure antibiotics was faster under alkaline conditions confirms that pharmaceuticals react more with the hydroxyl radical than with ozone.
+
+In contrast to the previous studies [40,41], the ozonation of azithromycin was slower than the other antibiotics. It is known that during ozonation reaction micropollutants are oxidized through attack by the ozone molecule itself or by the hydroxyl radical, which is derived from ozone decomposition. The ozone molecule reacts selectively with certain functional groups, but oxidation by the hydroxyl radical is not selective. The oxidation of micro-pollutant with molecular ozone suggested that ozone attacks the molecular structures at the high electron density location, such as C=C double bonds, activated aromatic systems, and no protonated amines, but it has limited reactivity to aromatic rings with ethyl, amide, or carboxyl groups. The low reactivity of azithromycin with ozone in comparison with other antibiotics can be aimed to the close proximity of pK41 and pK42 for azithromycin (specific rate constants for reactions of O3 with each individual acid-base species of an ionizable substrate) [42]. This close proximity make is difficult to determine whether the reaction of azithromycin is due primarily to reaction of O3 with the parent molecule's exocyclic tertiary amine or with its heterocyclic tertiary amine. As the experiments were carried at pH = 8 which involve both ozone pathways, it is implicit to conclude that the contribution of both path ways in oxidation of azithromycin is limited. Moreover, some of the degradation intermediates compete with the original compounds and consume ozone which also reduces the degradation efficiency. Further work to explain this trend is required.
+
+Estrogens (17-a-ethyl estradiol (EED), 178-estradiol, and estrone) showed similar trends to the antibiotics. Figure 3A presents the ozonation reaction of estrogens in synthetic wastewater with specific ozone dose in the range of 1.15 to 1.27 mg O3/mg DOC, input ozone of 188.1 mg h-1, a temperature of 20degC and a pH of 8. Ozone is a suitable oxidizing agent for estrogens, and an input ozone of 188.1 mg h-1 was sufficient to degrade all the studied estrogens to below detection limit (10 mg L-1) in less than 1 min. Chon _et al._[43] studied the oxidation of 17a-ethyl estradiol (EED) in the effluents of from the secondary sedimentation tanks of a wastewater treatment plant O3/DOC ratios from 0 to 1.24 mg O3/mg DOC. Reported results showed that EED was very reactive with ozone, and an initial concentration of 2 mM was removed to below the detection limit (EED/EED0 < 0.08) using a specific ozone dose of 0.27 mg O3/mg DOC. The result reported by Chon _et al._[43] showed that the oxidation of EED by the secondary formed hydroxyl radicals was the dominant oxidation path way. EDD (_G_ = 5 mg L-1) was also removed up to 67.5% from a prefiltered effluent sewage by ozone dose of 50 mg L-1 [44]. The reactivity of estrogens with ozone can be attributed to presence of phenolic moiety in the structure of these molecules which has a high reactivity with either ozone and/or hydroxyl radicals.
+
+The required ozone dose (V) used to oxidized the estrogens ranged from 4.9 mg L-1 to 5.0 mg/L (17a-et-thienyl estradiol; 1.02, 179-estradiol; 1.15, and extensor; 1.16 mg V0/mg DOC), with an average value of 4.9 mg L-1 (1.11 mg V0/mg DOC) and a standard deviation of 0.2. The differences in the specific ozone dose used in this study (1.02 mg V0/mg DOC) and the specific ozone dose in Chon _et al._[43] (0.27 mg V0/mg DOC) to achieve significant elimination of 17a-et-et-iently estradiol can be aimed to the differences in the initial concentration used on both studies (20 mM for this study and 2 mM for Chon _et al._'s study). Experiments carried out under acidic conditions (pH = 4) in the presence of hydroxyl radicals scavenger showed limited estrogen removal, and ozonation reactions carried out at alkaline condition (pH = 11) showed significant estrogens removals conforming that the oxidation of estrogens follow hydroxyl radicals pathway.
+
+Acidic (bezafibrate, diclofenac, and fenoprofen) and neutral pharmacokinetics (caffeine, fors formalize, propylphenzoenzo) were successfully oxidized by ozone (Figure 4A). A very rapid reaction was observed between the acidic pharmaceuticals and ozone, most being degraded in the experiments carried out at pH = 8 in less than 60 s by a specific ozone dose of 0.81, 1.1, and 1.02 mg O3/mg DOC for bezafibrate, diclofenac, and fenoprofen, respectively. The results obtained from the experiments carried out at acidic conditions (pH = 4), in the presence of radical scavenger and under the same specific ozone dose (~0.81 to 1.1 mg O3/mg DOC) showed a fast reaction between ozone and acidic pharmaceuticals suggesting that the reactivity of these pharmaceuticals follows direct ozone reaction pathway. Results from experiments carried out at alkaline pH confirm the reactivity of acidic pharmaceuticals with ozone.
+
+Neutral pharmaceuticals (caffeine, ifosfamide, propylphenylenzoene) were less affected by the ozonation reaction. Experiments carried out with an initial pharmaceuticals concentration of 30 uM and specific ozone dose in the range 0.82 to 2.18 mg O3/mg DOC caffeine; 1.43 mg O3/mg DOC, ifosfamide; 2.18 mg O3/mg DOC and propylphenylenzoene; 0.82 mg O3/mg DOC) a pH of 8, and a temperature of 20degC showed that a residuals of neutral pharmaceuticals were detected in the reaction effluent. The average % removals for five repeated experiments at 95% confidence level were 76.4%, 48.1%, and 51.0% for caffeine, ifosfamide, propylphenylenzoene, respectively. Another set of experiments was carried out with neutral pharmaceuticals using a higher inter concentration of 1.82 mg O3/mg DOC caffeine, 2.77 mg O3/mg DOC ifosfamide, and 1.04 mg O3/mg DOC propylphenylenzoene, a pH of 8, and a temperature of 20degC. This resulted in aqueous solutions with a residual neutral pharmaceutical as low as 20 uM suggesting that the ozone requirement for degrading neutral pharmaceuticals is higher than the other pharmaceuticals (antibiotic, acidic, and basic pharmaceuticals). The average used ozone dose for this set of experiments (\(\nabla\)) were 1.74 mg \(\nabla\) O3/mg DOC for caffeine, 2.70 mg \(\nabla\) O3/mg DOC for ifosfamide; and 1.03 mg \(\nabla\) O3/mg DOC for propylphenylenzoene. The reaction rate of caffeine was low even at pH 11, when the concentration of hydroxyl radicals should be higher
+
+Huerta-Fontela _et al._ [45] reported moderate caffeine removal efficiency (~76%) from drinking water containing caffeine with an initial concentration in the range of 1.0 to 1.7 mg L-1 during ozonation experiments carried out with an initial ozone concentration of 5 mg L-1 and reaction time of 20 min. Brosens _et al._ [28] also showed that drinking water with an initial caffeine concentration of 2754 to 246 ng L-1 required high ozone concentration (50 to 60 mg L-1) and long reaction time (15 min) to achieve no more than 80% removal efficiency. Rossal _et al._ [46] found that the ozonation of caffeine follows a two-stage pattern of an initial period of rapid decline followed by a second period with a lower
+
+Figure 2: (A) Evolution of antibiotics concentration as function of time during ozonation traction; \(\blacksquare\) [\(\text{Boxithromycin}_{\parallel}\) = 5 uM, \(\blacktriangledown\) [\(\text{Lazithromycin}_{\parallel}\) = 5 uM, \(\bullet\) [\(\text{Sulfathiazole}_{\parallel}\) = 20 uM, and \(\text{Xiulfamethoxazole}_{\parallel}\) = 20 uM. Temperature = 20°C, \(\text{lozonel}_{\parallel}\) = 5.5 mg/ L, pH = 8, (B) Structure of antibiotics.
+
+reaction rate. The difference in reaction between both periods was more pronounced for runs performed at a lower pH and may be associated with the rapid depletion of caffeine that took place during the first part of the runs. The mechanism of direct and indirect caffeine degradation by ozone and hydroxyl radicals was explained by initial attack of ozone to non-hindered double pond or an attack of the hydroxyl radicals to hindered double bond (C4 and C5) yielding derivatives of dimethylparabanic acid and such as 1,3/7-trimethyleuric acid [47]. Hydroxyl radicals can also attack N1, N3, and N7 positions leading to monomethylethylation and the production of theobromine, paraxanthine, and theophylline. The low reactivity between the neutral pharmaceuticals and the ozone at low ozone dose and pH of 8 can be related to the production of oxidation intermediates that are more reactive with ozone and hydroxyl radicals the case that lower their concentration in the water solution and reduce the possibilities of direct attach by ozone on N7-C4 double bond and by hydroxyl radicals on the other hindered double bond in C4-C5. Moreover, the production of organic intermediates containing (N, Cl, and P) can be considered as radical scavenger that reduce the ozone reactivity.
+
+### Organic Matter Minimization and Solution Toxicity
+
+Although the ozonation reaction is very effective in degradation the parent pharmaceutical pollutant, the partial oxidation of those molecules can lead to the formation of byproducts less prone to further oxidation. Short chain carboxylic acids such as oxamic, pyruvic, malic and oxalic acids are usually reported in literature as ozonation byproduct of pharmaceuticals [48]. For space limitation organic matter mineralization for the experiments carried out at pH of 8 will be presented in this section.
+
+The experiments carried out with synthetic wastewater and specific ozone dose of 2.41, 2.24, 2.29, and 2.50 mg O\(\text{y}\)/mg DOC reduced 46%, 45%, 44% and 40% of the dissolved organic matter (DOC) for water solution spiked with
+
+Figure 4: (A) Evolution of acidic (bezafibrate, diclofenac, and fenoprofen) and neutral pharmaceuticals concentration as function of ozonation reaction time; ○ [Bezafibrate] = 30 mM, ▚ [Diclofenac] = 30 μM, ▚ [Fenoprofen] = 30 μM, ▚ [Ifosfarálic] = 30 μM, xylPropylphenazon el] = 30 μM, ▚[Caffeine] = 30 μM, Temperature = 20°C, lozonel = 5.5 mg/L, pH = 8, (B) Structure of acidic pharmacokinetics and (C) Structure of neutral pharmacokinetics.
+
+Figure 3: (A) Evolution of estrogens concentration as function of ozonation reaction time; ▚ [17α-ethinylestradiol], = 20 μM, ▚ [17β-estradiol], = 20 μM and ▚ [Estrone], = 20 μM. Temperature = 20°C, [ozone], = 5.5mg/L, pH = 8, (B) Structure of estrogens.
+
+azithromycin (_G_ = 5 mM), roxithromycin (_G_ = 5 mM), sulfa-methoxyazole (_G_ = 20 mM), and sulfatinazole (_G_ = 20 mM), respectively.
+
+Lower levels of DOC removal were noticed for estrogens: the DOC removal rates for 17a-ethinyl estradiol (EED) (_G_ = 20 mM), 17b-estradiol (1.2 +- 20 mM), and estrone (_G_ = 20 mM) were 33%, 35%, and 34%, using specific ozone dose of 1.15, 1.27, and 1.27 mg Os/mg DOC, respectively. The high initial concentration of estrogens and the low specific ozone dose (~1.15 to 1.27 mg Os/mg DOC) can be the main reasons for the low DOC mineralization for these estrogens. Moreover, it is believed that the differences between antibiotics and estrogens in terms of organic matter mineralization are related to the differences in the oxidation byproducts that are formed: these compete with the original compounds to consume ozone and any other oxidative species. The results suggest that estrogens produce byproducts that require more ozone, consequently resulting in a decrease in degradation efficiency.
+
+Acidic pharmaceuticals showed a DOC removal rate in the range of 33% to 36% for a used specific ozone dose in the range 0.81 to 1.1 mg Os/mg DOC, while neutral pharmacists showed DOC removal only on the level of 23% to 30% with specific ozone dose of 0.82 to 1.4 mg Os/mg DOC. As the ozonation reaction did not lead to complete mineralization of the pharmaceuticals, it was essential to evaluate the effect of the ozonation process on the toxicity of the aqueous solution. The initial synthetic wastewater samples containing pharmacokinetics and water samples treated with ozone at pH = 8 were tested using Microtax(r) bioassays. Figure 5 shows the TU value for the studied pharmaceuticals before and after the ozonation process. Ozonation reaction with specific ozone dose in the range of 2.24 to 2.55 mg Os/mg DOC for antibiotic, 1.15 to 1.27 for estrogens and 0.82 to 2.18 for neutral pharmacokinetics reduced significantly the acute toxicity of the water samples compared with the initial synthetic wastewater. As an example, water containing 5 mM of Roxithromycin showed a TU value of 15, and ozonation of this aqueous solution with specific ozone dose of 2.24 mg Os/mg DOC for only 1 min reduced the TU level to 2. Similar results were observed for the other solutions, with exception of the acidic pharmaceuticals. The toxicity of the aqueous solutions containing acidic pharmaceuticals increases after the ozonation reaction. It was confirmed that, for all the studied pharmaceuticals, ozone attacks the tertiary amines, leading to the cleavage of the N-C bonds. This reaction can lead to either the destruction of the chemical structure of the pharmaceutical or the loss of the methyl group from the amino group. Li _et al._ [49] have confirmed that the loss of the methyl group from roxithromycin eliminates its apparent activity. Besides, the ozonation reaction can cause cleavage of the phenolic structure in pharmaceuticals and reduce its estrogenicity significantly. On the basis of these considerations, the oxidation of pharmaceuticals with ozone results in a significant reduction in the pharmacological effects of these species.
+
+The Global Harmonized System of Classification and Labelling of Chemicals (GHS) [50] classified the toxicity of substance into four main categories (very toxic, toxic, harmful, and not harmful) based on _L_G_, _E_Co, or _C_NV values (see Table 3). Toxicity until in _E_Q 2 and the measured _E_Co values were correlated to the classification of the GHS system as presented in Table 3. Water solution spiked with pharmaceuticals showed a toxicity units (TU) in the range of 11 to 18, with the highest values occur for antibiotics and estrogen. According to GHS classification the toxicity of these water solutions is classified as toxic. For all the studied pharmaceuticals except for acidic the ozonation reaction produced water solutions with a TU values <10, suggesting that ozone changed the toxicity of these water solutions from toxic not harmful. Similar results were reported by Zhou _et al._ [44]. The finding from the present study is in contrast with finding of Ikehata _et al._ [51] which showed that treatment of pharmaceuticals by ozonation can generate products of greater toxicity than the parent compound. The observed increase in the toxicity of acidic pharmaceuticals, for the experiments carried out with specific ozone dose of 2.24 mg Os/mg DO, suggests that ozone breaks these pharmaceuticals down to more toxic byproducts. However, ozonation of acidic pharmaceuticals with higher specific ozone dose (from 1.02 to 1.39 mg Os/mg DOC) produced water solution with acute toxicity by 33% less than the original aqueous solution, suggesting that higher specific ozone dose is required by the acidic pharmaceuticals and their oxidation byproducts to eliminate their acute toxicity.
+
+### Effects of Initial Ozone Concentration
+
+Significant degradation efficiencies (>99.9%) were reported for most of the studied pharmaceuticals with an inlet ozone concentration of 5.5 mg L-1 (specific ozone dose in the range 0.82 to 2.55 mg Os/mg DOC). The efficiency of ozone in destroying antibiotics, estrogens, and other pharmacists was recognized as being due to the high reactivity of ozone and/or hydroxyl radicals with the double bond in the compound structure. Accordingly, the cleavage of these compounds is strongly related to the ability of the oxidizing agent to attack these bonds. Higher and lower ozone concentrations were tested; there was a modest trend indicating an increase in the percentage of pharmaceuticals oxidized with increasing initial ozone dose from 1 to 5.5 mg L-1. For inlet ozone concentrations <4 mg L-1, the residual concentrations of these pharmaceuticals was detected in the aqueous solution, but no residual ozone remained in the solution, suggesting that the treatment process operates in underdose conditions. On other hand, when inlet ozone concentrations greater than 5.5 mg L-1 were used, the total elimination of the pharmaceuticals was achieved and, in most cases, residual ozone was detected in the aqueous solution. Although the presence of residual ozone can be used to remove other organic matter found in water utilities, using a high concentration of ozone will negatively affect the economics of the process. Accordingly, an ozone dose of 5.5 mg L-1 was set as the optimum value for the ozonation experiment with synthetic wastewater.
+
+### Kinetic Study
+
+The ozonation of pharmaceuticals in water is a consequence of the combination of direct and indirect, or radical, oxidation reactions. The overall kinetic of both pathways was assumed to be second order in the light of most data published on the rate of ozonation processes [24,25,27]. The rate expression of the ozonation of pharmaceuticals can be described by the following equation:
+
+\[\frac{d[\text{Pharm}]}{dt} = K_{03}[\text{O}_{3}][\text{Pharm}_{\text{i}}] + K_{04}[\text{OH}][\text{Pharm}_{\text{i}}\rbrack\]
+
+The solution of Eq. 4 requires information about the concentration of hydroxyl radical, which is generally difficult to be measured directly. Elowitz and Von Gunten [52] defined a parameter (_R_x) that relates the production of hydroxyl radical to the presence of ozone in the solution \(\left( {R_{\text{a}} = \frac{\int{{C_{\text{on}}\;\;\text{a}}}}{{\int{{C_{\text{on}}\;\;\text{a}}}}}} \right)\) which include the two oxidant species involved in the reaction, using the deferential form of _R_a Eq. 4 can be rewritten as \[\frac{d[\text{Pharm}]}{dt} = K_{\text{O3}}[\text{O}_{3}][\text{Pharm.}] + K_{\text{OH}}R_{\text{O3}}[\text{O}_{3}][\text{Pharm.}]\]
+
+which then can be reduced to Eq. 6
+
+\[\frac{d[\text{Pharm}]}{dt} = (K_{\text{O3}} + K_{\text{OH}}R_{\text{O}})[\text{O}_{3}][\text{Pharm.}]\]
+
+The reaction requirements (_z_) for the studied groups of pharmaceuticals were determined using Eq. 1 to be in the range from 0.98 to 1.05 for antibiotics, 1.03 to 1.13 for estrogens, 0.99 to 1.22 for acidic pharmaceuticals, and 1.08 to 1.15 for neutral pharmaceuticals. As mentioned previously, the half-life of most conoated pharmaceuticals was less than 15 s (except for the neutral pharmaceuticals) with an initial ozone concentration of 5.5 mg L+1 (specific ozone ratio in the range of 0.82 to 2.55 mg O3/mg DOC). For this short reaction times and high ozone concentration, the ozone to pharmaceutical molar ratio is greater than of the stoichiometric ratio, and so \([\text{O}_{3}] \approx [\text{O}_{3}]_{\text{i}} = \text{Constant}\).
+
+Accordingly, Eq. 6 can be reduced to
+
+\[\frac{d[\text{Pharm}]}{dt} = k_{\text{i},\text{i}}\lbrack\text{Pharm.}]\]
+
+where \(k_{\text{i},\text{i}} = K[\text{O}_{3}]\)
+
+Equation 8 can be written in linear form as
+
+\[\text{ln}\left( k_{\text{i},\text{i}} \right) = \text{ln}\left( K \right) + \text{ln}\left[ \text{O}_{3} \right\rbrack\]
+
+Equation 8 suggests that the plot of ln(_k_a,_i_) _versus_ ln[O3] gives a straight line with a slope equal to one and an intercept of ln(_K_). Figure 6 shows the evolution of the natural logarithm of the pseudo first-order rate constant during the ozonation of roxinromycin acid at 20degC and at the three pH values (4, 8, and 11). The plot yields a straight line (_K_2 > 0.97) with a slope very close to unity. The slope was calculated with a 95% confidence interval and was found to be \(1.021 \pm 0.052\), \(1.085 \pm 0.081\), and \(1.005 \pm 0.005\) for pH 11, 4, and 8, respectively. Different authors have confirmed that the ozonation reaction is overall second-order kinetic [53,54] and first order with respect to each reactant [24]. The results are in agreement with the published kinetics. Similar mathematical manipulation was used to calculate the rate constants for the different pharmaceuticals. For each pharmaceutical, the _k_a,\(i\) value was measured by averaging the _k_a,\(i\) value yield from five experiments for each pharmaceutical under the same condition. The maximum standard deviation of the averaged pseudo first-order constant from the individual _k_a,\(i\) values was 6%.
+
+Figure 5.: Toxicity value of initial water samples contain pharmaceuticals before and after ozonation.
+
+\begin{table}
+\begin{tabular}{c c c} \hline
+**Parameter** & **TU range** & **Class** \\ LC50/EC50/ChV & \(\leq 100\) & Very toxic \\ LC50/EC50/ChV & \(100 \times \text{EC50} \leq 10\) & Toxic \\ LC50/EC50/ChV & \(10 \times \text{EC50} \leq 1\) & Harmful \\ LC50/EC50/ChV & EC50 \textgreater{} 1 & Not harmful \\ \hline \end{tabular}
+\end{table}
+Table 3: Classification of the toxicity of water solutions following the Global Harmonized System of classification and labeling of chemicals (GHS).
+
+Table 5 shows the overall rate constants (_K_) for the pharmaceutical ozonation experiments carried out at 20degC, with initial concentration of ozone at 5.5 mg L-1 and the pH at 8. The overall rate constants obtained in this study were compared to the values reported in literature (_K_). Dodd and Huang [55] have estimated the second-order rate constant for the ozonation of azithromycin and roulithromycin at 20degC in buffered synthetic wastewater to be \(1.1\times 10^{2}\) and \(4.5\times 10^{6}\), respectively. It can be seen that the obtained results for these pharmaceuticals are in agreement with the reported values. Guten _et al._[56] used an initial ozone concentration of 5 mg L-1 to study the ozonation kinetic of sulfamethoxazole, bezahlbrate, and diclofenac in the effluent of an activated sludge process and the effluent of MBR at 25degC and a pH of 7. The differences between the rate constant values estimated in this study and the values reported by Guten _et al._[56] can be related to the differences in experimental conditions (temperature, pH, type of water used, and differences in experimental setup).
+
+#### Comparison of Ozone with Chlorine Dioxide and Chlorine
+
+As treatment alternatives, it is important to compare the rate constants of ozone with other rate constants for oxidants used commonly in water treatment, such as chlorine and chlorine dioxide. Figure 7 presents a comparison between the rate constants of ozone calculated in this study and the rate constants of chlorine and chlorine dioxide from the literature [55, 57, 58]. It can be seen in Figure 7 that ozone reacts with most of these pharmaceuticals at a high rate. The apparent rate constants for the reaction between ozone and pharmaceuticals are roughly 3 orders of magnitude greater than the rate constants of chlorine and chlorine dioxide. Accordingly, using O3 in treatment processes containing wastewater contaminated with pharmaceuticals will give better results.
+
+### Ozonation of Pharmaceuticals in Surface Water and Wastewater
+
+The water samples used in these experiments were taken from surface water and effluent from a municipal WWTP. The samples were spiked with pharmaceuticals at specific concentrations, as indicated in Table 1. Ozonation experiments were carried out at 20degC and at a pH of 8. The preliminary experiments revealed that the inlet ozone concentration required to eliminate over 90% of pharmaceuticals from surface water and wastewater is \(\sim\)6.5 mg L-1. Table 5 shows a comparative summary for the experiments carried out with surface water (SFW), effluent of municipal wastewater treatment plant (WWTP), and synthetic wastewater (SWW).
+
+### Ozonation of Pharmaceuticals in Surface Water
+
+Experiments were carried out to evaluate the potential of ozone in reducing spiked pharmaceuticals in surface water. Treating surface water spiked with antibiotics and estrogens with an initial ozone concentration of 6.5 mg L-1 at temperature of 20degC and a pH of 8 produced pharmaceutical-free
+
+\begin{table}
+\begin{tabular}{l c c c c c} \hline \hline
+**Compound** & \(K\) (M\({}^{-1}\) s\({}^{-1}\)) & _K_\({}_{c}\) (M\({}^{-1}\) s\({}^{-1}\)) & **Compound** & \(K\) (M\({}^{-1}\) s\({}^{-1}\)) & _K_\({}_{c}\) (M\({}^{-1}\) s\({}^{-1}\)) \\ \hline Azithromycin & \((0.7\pm 0.44)\times 10^{5}\) & \(1.1\times 10^{2}\)[59] & Bezafibrate & \((9.32\pm 0.15)\times 10^{2}\) & 590 [56] \\ Boxithromycin & \((9.5\pm 0.51)\times 10^{5}\) & \(4.5\times 10^{6}\)[59] & Diclofenac & \((1.06\pm 0.03)\times 10^{6}\) & \(1\times 10^{6}\)[56] \\ Sulfamethoxazole & \((5.03\pm 0.05)\times 10^{6}\) & \(2.5\times 10^{6}\)[56] & Fenoprofen & \((1.42\pm 0.03)\times 10^{5}\) & \\ Sulfathiazole & \((8.05\pm 0.03)\times 10^{5}\) & & Caffeine & \((2.30\pm 0.3)\times 10^{5}\) & \(0.25\)-\(1.05\)[45], \\ \(17\)a-ethinylestradiol & \((3.49\pm 0.5)\times 10^{6}\) & \(7\times 10^{6}\)[37] & Ifosfamide & \((1.21\pm 0.1)\times 10^{5}\) & \\ (EED) & & & & & \\ (17)b-estradiol & \((1.09\pm 0.01)\times 10^{6}\) & & & & \\ Estrone & \((9.84\pm 0.2)\times 10^{5}\) & \(6.2\times 10^{3}\)\(-9.4\) & & & \\  & \(\times\)\(10^{5}\)[59] & & & & \\ \hline \hline \end{tabular}
+\end{table}
+Table 4: The overall rate constant (_K_) for the ozonation different pharmaceuticals in synthetic wastewater.
+
+Figure 6: Relationship between initial ozone concentration (M) and pseudo first-order rate constant (1/s) for Roxithromycin at different pHs (T = 20°C and [Roxithromycin] = 5 mM. [Color figure can be viewed in the online issue, which is available at **wileyonlinelibrary.com.]**
+
+Figure 7: Comparison of apparent second-order rate constants for the reaction of selected pharmaceuticals with ozone, chlorine dioxide, and chlorine at pH 7.
+
+* [19] J. M. C. (1979). _The Theory of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of the Optical Constants of Opticalwater (to a detection limit of 10 mg L-1). Specific ozone dosage in the range of 1.4 +- 0.5 to 1.5 +- 0.4 mg O3/mg DOC for antibiotics and 0.9 +- 0.5 to 1.0 +- 0.3 mg O3/mg DOC for estrogens led to complete removal of these emerging contaminates except for azithromycin which was removed by 98.3%. Acidic and neutral pharmaceuticals added to the surface water were also treated by ozone under the same conditions. The percentage removal of the acidic pharmaceuticals using specific ozone dosage in the range of 0.9 +- 0.3 mg O3/mg DOC was in the range 98.95 to 99.2%, and neutral pharmaceuticals using specific ozone dosage in the range of 0.7 +- 0.4 to 1.1 +- 0.3 mg O3/mg DOC was in the range of 99% to 99.5%.
+
+Organic mineralization for experiments carried out with surface water was found to fall in the range 33% to 39%, 25% to 28%, 23% to 26%, and 18% to 24% for antibiotic, estrogen, acidic, and neutral pharmaceuticals, respectively. The lower organic matter mineralization for the experiments carried out with surface water can be attributed to the presence of background DOC in the surface water which completes with the pharmaceuticals in reactions with ozone. Further work is required to evaluate the toxicity of the surface water after ozonation.
+
+### Ozonation of Pharmaceuticals in Municipal Wastewater Treatment Plant effluent
+
+Similar experiments were carried out to evaluate the potential of ozone in treating pharmaceuticals added into municipal WTPT effluents (see Table 5). Experiments were carried out under the same conditions as in the case with surface water [O3] = 6.5 mg L-1, 20degC, pH 8). The specific ozone dose during this set of experiments ranged from 0.28 to 0.6 mg O3/mg DOC for antibiotics, 0.22 to 0.54 mg O3/mg DOC for estrogen, 0.19 to 0.65 mg O3/mg DOC for acidic and 0.4 to 1.4 mg O3/mg DOC for neutral pharmacokinetics. The percentage removal from municipal WTPT effluent were found to be >=98.5% >=97.5% >=99%, and >=98.6 antibiotic, estrogen, acidic, and neutral pharmaceuticals, respectively. The mineralization of organic matter were found to be in the range of 22% to 30%, 16% to 19%, 9% to 14%, and 11% to 16% in the same order, respectively. Quinomas _et al._ [60], showed that the ozonation reaction is effective in completely removing emerging contaminants (_G_i = 0.2 mg L-1) from effluents of secondary of WTPT by applied ozone of 520 mg h-1 and solution pH of 3 and 7. However, the organic matter mineralization level achieved was limited (<35% TOC removal). However, Nakada _et al._ [41] showed that ozonation removed more than 80% of phenolic antispecies, croatamtion, sulfonamide and macrolide antibiotics, and 17b-estradiol from municipal sewage treatment plant.
+
+## CONCLUSION
+
+The oxidation of different types of pharmaceuticals (antibiotic, estrogen, acidic, and neutral) with ozone was investigated in synthetic wastewater, surface water, and wastewater. In oxidation reactions, ozone significantly degrades most of the studied pharmaceuticals in aqueous solutions. The pharmaceutical degradation rate was strongly accelerated by increasing the initial ozone dose and the pH value of the solution. In synthetic wastewater, total pharmaceutical degradation was obtained within 60 s when the ozone dose was 188.1 mg h-1. A higher ozone dosage (222.3 mg h-1) was necessary to degrade the pharmaceuticals in surface and wastewater, due to the presence of natural organic matter.
+
+The ozonation reaction was found to follow an overall second-order kinetic, with first-order kinetics for both ozone and the pharmaceuticals, respectively. The overall rate constants (_K_) for the antibiotic, estrogen, acidic, and neutral pharmaceuticals was on the order of 105-105, 106, 105, and 105-106 M-1 s-1, respectively. Ozone seems to be noticeably more efficient for pharmaceutical treatment than chlorine and chlorine dioxide, because it has higher rate constants and reacts with a wide range of pharmaceuticals. The toxicity of aqueous solutions containing the investigated pharmaceuticals was found to decrease following ozone treatment. Antibodies, estrogens, and neutral pharmaceuticals showed a decrease in toxicity by approximately 66% following ozonation with an inlet ozone concentration of 5.5 mg L-1. Acidic pharmaceuticals required a higher inlet ozone dose (>5.5 mg L-1) to reduce their toxicity values.
+
+### Nomenclature
+
+\(\nabla\): Ozone used to degrade the pharmaceuticals [O3] : Concentration of ozone at any time (M) [O3] : Final concentration of ozone (M) [O3] : Ozone concentration in the outlet gas [O3] : Ozone concentration in the inlet gas [O3] : Initial concentration of ozone (M) [O4] : Concentration of ozone in the reactor [OHI] : Concentration of hydroxyl radicals at any time (M) [Pharm] : Concentration of pharmaceuticals at any time (M) [Pharm] : Final concentration of pharmaceuticals (M) [Pharm] : Initial concentration of pharmaceuticals (M) [D_M_] : Time of ozonation EC00 : Concentration of test sample at which there was a 50% reduction in metabolic activity of bacterium _K_OS : Reaction rate constant with ozone (M-1 s-1) [_K_OH : Reaction rate constant with direct hydroxyl radicals (M-1 s-1) [_Q_] : Gas-flow rate to the reactor \(t\) : Time (s) [TU] : Toxicity unit (100/EC00) [_V_] : Volume of water in the reactor
+
+## Literature Cited
+
+1. [1] : Ternes T. (2001). Pharmaceuticals and Metabolites as Contaminants of the Aquatic Environment. ESWE:Institute for Water Research and Water Technology. Schellin-strasse 158, D-65201 Westhead, Germany. Pharmaceuticals and Care Products in the Environment. Chapter 2, pp 39-54. Chapter 101. 1021/pk:2001-0071.e10.002. ACS Symposium Series, Vol. 791. ISBN13: 978084123791e ISBN: 9780841238673. Copyright (c) 2001 American Chemical Society.
+2. [2] : Huber, M., Ternes, T.A., & Von Gunten, U. (2004). Removal of estrogenic activity and formation of oxidation products during ozonation of 17alpha - ethinylestradiol, Environmental Science & Technology, 38, 5175-5186.
+3. [3] : Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., & Buxton, H.T. (2002). Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams 1999-2000: A national reconnaissance, Environmental Science & Technology, 36, 1202-1211.
+4. [4] : Li, X., Wang, Y., Yuan, S., Li, Z., Wang, B., Huang, J., Deng, S., & Yu, G. (2014). Degradation of the anti-inflammatory drug ibuprofen by electro-peroxone process, Water Research, 63, 81-93.
+5. [5] : Jones, O.A.H., Voulvoulis, N., & Lester Lee, J.N. (2005). Human pharmaceuticals in wastewater treatment processes, Critical Reviews in Environmental Science and Technology, 35, 401-427.
+6. [6] : Ternes, T.A. (1998). Occurrence of drugs in German sewage treatment plants and rivers, Water Research, 32, 3245-3260.
+7. [7] : Park, H., & Cheung, Y.K. (2007). Degradation of antibiotics (tetracycline, sulfathiazole, ampicillin) using enzymesof glutathione _S_-transferase, Human and Ecological Risk Assessment: An International Journal, 13, 1147-1155.
+* (8) S. Fram, M.S., & Belitz, K. (2011). Occurrence and concentrations of pharmaceutical compounds in groundwater used for public drinking-water supply in California, Science of the Total Environment, 409, 3409-3417.
+* (9) Sacher, F., Lange, F.T., Brauch, H., & Blankenhorn, I. (2001). Pharmaceuticals in groundwaterwaters: Analytical methods and results, of a monitoring program in Baden-Wartternberg, Germany. Journal of Chromatography A, 938, 199-210.
+* (10) Koch, D.E., Bhandari, A., Close, L., & Hunter, R.P. (2005). Azithromycin extraction from municipal wastewater and quantitation using liquid chromatography/mass spectrometry, Journal of Chromatography A, 1074, 17-22.
+* (11) Mohapatra, D.P., Barr, S.K., Tyagi, R.D., Picard, P., & Surampalli, R.Y. (2014). Analysis and advanced oxidation treatment of a persistent pharmaceutical compound in wastewater and wastewater sludge-carbamazepine, Science of the Total Environment, 470,471, 58-75.
+* (12) Espejo, A., Aguinaco, A., Amat, A.M., & Beltran, F.J. (2014). Some ozone advanced oxidation processes to improve the biological removal of selected pharmaceutical contaminants from urban wastewater, Journal of Environmental Science and Health, Part A, 49, 410-421.
+* (13) Espejo, A., Aguinaco, A., Garcia-Araya, J.F., & Beltran, F.J. (2014b). Sequential ozone advanced oxidation and biological oxidation processes to remove selected pharmaceutical contaminants from an urban wastewater, Journal of Environmental Science and Health, Part A, 49, 1015-1022.
+* (14) Verlichi, P., Al Aukidy, M., & Zambello, E. (2012). Occurrence of pharmaceutical compounds in urban wastewater: Removal, mass load and environmental risk after a secondary treatment. A review, Science of the Total Environment, 429, 123-155.
+* (15) Bhandari, A., Close, L., Kim, W., Hunter, R., Koch, D., & Surampalli, R. (2008). Occurrence of ciprofloxacin, sulfamethoxazole, and azithromycin in municipal wastewater treatment plants, Practice Periodical of Hazardous, Toxic, and Radioactive Waste Management, 12, 275-281.
+* (16) Lee, Y., Gerrity, D., Lee, M., Bogaet, A.E., Salhi, E., Gamage, S., Trenholm, R.A., Wert, E.C., Snyder, S.A., & von Gunten, U. (2013). Prediction of micropollutant elimination during ozonation of municipal wastewater effluents: Use of kinetic and water specific information, Environmental Science & Technology, 47, 5872-5881.
+* (17) Nugroho, W.A., Reungoat, J., & Keller J. (2010). The performance of biological activated carbon in removing pharmaceuticals in drinking water treatment, Journal of Applied Sciences in Environmental Sanitation, 5, 131-141.
+* (18) Radgenovic, J., Petrovic, M., Ventura, F., & Barcelo, D. (2008). Rejection of pharmaceuticals in nanoflibration and reverse osmosis membrane drinking water treatment, Water Research, 4, 3601-3610.
+* (19) Derakhshgchpoor, R., Homayoonfal, M., Akkari, A., & Mehrnia, M.R. (2013). Amoxicillin separation from pharmaceutical wastewater by high permeability polysulfone nanofiltration membrane, Journal of Environmental Health Science and Engineering, 11, 9-19.
+* (20) Kok-Kwang, N., Angela Yu-Chen, L., Tsung-Hsien, Y., & Cheng-Fang, L. (2011). Tertiary treatment of pharmaceuticals and personal care products by pretreatment and membrane processes, Sustainable Environment Research, 21, 173-180.
+* (21) Gerrity, D., Gamage, S., Jones, D., Korshin, G.V., Lee, Y., Pisarenko, A, Trenholm, R.A, Von Gunten, U, Wert, E.C, & Snyder Sh, A. (2012). Development of surrogate correlation models to predict trace organic contaminant oxidation and microbial inactivation during ozonation, Water Research, 4, 6257-6272.
+* (22) Gerrity, D., Gamage, S., Holady, J.C., Mawhinney, D.B., Quinones, O., Trenholm, R.A., & Snyder, Sh.A. (2012) Pilot-scale evaluation of ozone and biological activated carbon for trace organic contaminant mitigation and disinfection, water research 46: 6257-6272.
+* (23) Hollender, J., Zimmermann, S.G., Koepke, S., Krauss, M., McArdell, C., Ott, C., Sivon Gunten, H., & Siegrist, U. H. (2009). Elimination of organic micropollutants in a municipal wastewater treatment plant upgraded with a full-scale post-ozonation followed by sand filtration, Environmental Science & Technology, 43, 7862-7869.
+* (24) Gamal El-Din, M., Smith, D.W., Al Monnai, F., & Wang, W. (2006). Oxidation of resin and fatty acids by ozone: Kinetics and toxicity study, Water Research, 40, 392-400.
+* (25) Bailey, P.E. (1959). The reaction of ozone with organic compounds, Chemical Reviews, 59, 179-180.
+* (26) Yargeau, V., & Leclair, C. (2008). Impact of operating conditions on decomposition of antibiotics during ozonation: A review, Ozone: Science and Engineering, 30, 175-188.
+* (27) Staehelin, J., & Hoigne, J. (1985). Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions, Environmental Science & Technology, 19, 1206-1213.
+* (28) Brosus, R., Vincent, S., Aboulfall, K., Daneshvar, A., Sauv, S., Barbeau, B., & Provost, M. (2009). Ozone oxidation of pharmaceuticals, endocrine disruptors and pesticides during drinking water treatment, Water Research, 43, 4707-4717.
+* (29) Le-Minh, N., Khan, S.J., Drewes, J.E., & Stuetz, R.M. (2010). Fate of antibiotics during municipal water recycling treatment processes, Water Research, 44, 4295-4523.
+* (30) Vogna, D., Marotta, R., Napolitano, A., Andreozzi, R., & Mischia, M. (2004). Advanced oxidation of the pharmaceutical drug diclofenac with UV/H 2 O 2 and ozone, Water Research, 38, 414-422.
+* (31) APHA, WPCF. (1995). Standard methods for the examination of water and wastewater (16th Edition, Volume 16, pp. 1985), American Public Health Association.
+* (32) Al monami, F.A., & Jarrah, N. (2010). Treatment and kinetic study of cyanobacterial toxin by ozone, Journal of Environmental Science and Health Part A, 45, 719-731.
+* (33) Yasojima, M., Nakada, N., Komori, K., Suzuki, Y., & Tanaka, H. (2006). Occurrence of levofloxacin, clarithromycin and azithromycin in wastewater treatment plant in Japan, Water Science & Technology, 53, 227-233.
+* (34) Loganathan, B., Phillips, M., Movery, H., & Jones-Leep, T.L. (2009). Contamination profiles and mass loadings of macrolide antibiotics and illicit drugs from a small urban wastewater treatment plant, Chemosphere, 75, 70-77.
+* (35) Ying, L., Ranjia, J., Kyungho, C., Young, I.K., & Bullen, P.G.D. (2012). Pharmaceutical residues in wastewater treatment plants and surface waters in Bangkok, Journal of Hazardous, Toxic, and Radioactive Waste, 16, 88-91.
+* (36) Artur, P., Marzenna, R. D., & Iucjan, P. (2013) Environmental engineering IV. CRC Press, Taylor & Francis Group, The Netherlands.
+* (37) Hirsch, R., Ternes, T.A., Haberer, K., & Kratz, K.L. (1999). Occurrence of antibiotics in the aquatic environment, Science of the Total Environment, 225, 109-118.
+* (38) Martins, R.C., Cardoso, M., Dantas, R.F., Sans, C., & Esplugas, S. R.M. (2015). Quinta-Ferreira. 2015. Catalytic studies for the abatement of emerging contaminants by ozonation, Journal of Chemical Technology & Biotechnology, 90, 1611-1618.
+* (39) Sagi, G., Csay, T., Patzay, G., Sonka, E., Woinarovits, L., & Takacs, E. (2014). Oxidative and reductive degradation of sulfamethoxazole in aqueous solutions: Decomposition efficiency and toxicity assessment, Journal of Radioanalytical and Nuclear Chemistry, 301, 475-482.
+* [40] Liu, P., Zhang, H., Feng, Y., Yang, F., & Zhang, J. (2014). Removal of trace antibiotics from wastewater: A systematic study of nanofiltration combined with ozone-based advanced oxidation processes, Chemical Engineering Journal, 240, 211-220.
+* [41] Nakada, N., Shinohara, H., Murata, A., Kiri, K., Managaki, S., Sato, N., & Takada, H. (2007). Removal of selected pharmaceuticals and personal care products (PPCPs) and endocrine-disrupting chemicals (EDCs) during sand filtration and ozonation at a municipal sewage treatment plant, Water Research, 41, 4373-4382.
+* [42] Dodd, M.C., Buffle, M.O., & von Gunten, U. (2006). Oxidation of antibacterial molecules by aqueous ozone: Moiety-specific reaction kinetics and application to ozone-based wastewater treatment, Environmental Science & Technology, 40, 1969-1977.
+* [43] Chon, K., Salhi, E., & von Gunten, U. (2015). Combination of UV absorbance and electron donating capacity to assess degradation of micropollutants and formation of bromate during ozonation of wastewater effluents, Water Research, 81, 388-397.
+* [44] Zhou, H., Liu, J., Xia, H., Zhang, Q., Ying, T., & Hu, T. (2015). Removal and reduction of selected organic micropollutants in effluent sewage by the ozone-based oxidation processes, Chemical Engineering Journal, 269, 245-254.
+* [45] Huerta-Fontela, M., Galceran, M.T., & Ventura, F. (2008). Stimulatory drugs of abuse in surface waters and their removal in a conventional drinking water treatment plant, Environmental Science & Technology, 42, 6809-6806.
+* [46] Rosal, R., Rodraguez, A., Perdigian-Melian, J.A., Petre, A., Garaca-Calvo, E., Gamez, M.J., Jagera, A., & Fernandez-Alba, A.R. (2009). Degradation of caffeine and identification of the transformation products generated by ozonation, Chemosphere, 74, 825-831.
+* [47] trader, R.H., & Fay, L.B. (1995). Anticoitative reactions of caffeine: formation of 8-oxoxcaffeine (1,3,7-trimethryluric acid) in coffee subjected to oxidative stress, Journal of Agricultural and Food Chemistry, 43, 1323-1338.
+* [48] Goncalves, A., Orfao, J., & Pereira, M. (2012). Catalytic ozonation of sulfanterexoxide in the presence of carbon materials. Catalytic performance and reaction pathways, Journal of Hazardous Materials, 239240, 167-174.
+* [49] Li, X, Zhong, D., Huang, H., & Wu, S. (2001). Demethylation metabolism of roxithromycin in humans and rats, Acta Pharmacologica E Sinica, 22, 469-474.
+* [50] United Nations. (2011). Globally harmonized system of classification and labelling of chemical (GHS) (4th Edition). New York: United Nations Publications.
+* [51] Ikehata, K., Naghashkar, N.J., & El-Din, M.G. (2006). Degradation of aqueous pharmaceuticals by ozonation and advanced oxidation process: A review, Ozone: Science & Engineering, 28, 353-414.
+* [52] Elovire, M.S., & Von Gunten, U. (1999). Hydroxyl radical/ozone ratios during ozonation processes I: The R-ct concept. Ozone: Science & Engineering, 21, 239-260.
+* [53] Lovato, M.A.E., Marta, N.C.A., & Cassano, A.E. (2009). A reaction kinetic model for ozone decomposition in aqueous media valid for neutral and acidic pH, Chemical Engineering Journal, 146, 486-497.
+* [54] Zhang, P., & Liu, J. (2004). Photocatalytic degradation of trace hexane in the gas phase with and without ozone addition: kinetic study, Journal of photochemistry and Photobiology A: Chemistry, 167, 87-94.
+* [55] Dodd, M.C., & Huang, C.H. (2004). Transformation of the antibacterial agent sulfamethoxazole in reactions with chlorine: Kinetics, mechanisms, and pathways, Environmental Science & Technology, 38, 5607-5615.
+* [56] Huber M. M., Gobel A., Joss A., Hermann N., Loffler D., MacArell C.S., Ried A., Siegrist H., Ternes T. A von Gunten U. (2005) Oxidation of pharmaceuticals during ozonation of municipal wastewater effluents: A pilot study, Environmental Science and Technology, 39(11), 4290-4299.
+* [57] Pinkston, K.E., & Sedlak, D.L. (2004). Transformation of aromatic ether- and amine-containing pharmaceuticals during chlorine disinfection, Environmental Science & Technology, 38, 4019-4025.
+* [58] Huber, M.M., Korhonen, S., Ternes, T.A., & von Gunten, U. (2005). Oxidation of pharmaceuticals during water treatment with chlorine dioxide, Water Research, 39, 3607-3617.
+* [59] Belfroid, A.C., Van der Horst, A., Vethaak, A.D., Schafer, A.J., Rijs, G.B.J., Wegener, J., & Cofino, W.P. (1999). Analysis and occurrence of estrogen hormones and their glucoronideins surface water and waste water in The Netherlands, Science of the Total Environment, 225, 101-108.
+* [60] Quiaones, D.H., Alvarez, P.M., Rey, A., & Beltran, F.J. (2015). Removal of emerging contaminants from municipal WWTP secondary effluents by solar photocatalytic ozonation. A pilot-scale study, Separation and Purification Technology, 149, 132-139.

manual_annotation/200pdfs_in_mmd/10.1002_jctb.2363.mmd

@@ -0,0 +1,215 @@
+# Diclofenac removal from water by ozone and photolytic TiO2 catalysed processes
+
+Juan F. Garcia-Araya, Fernando J. Beltran
+
+Almudena Aguinaco
+
+# Abstract
+
+BACKGROUND: The aim of this work was to establish the efficiency of single oxzonation at different pH levels (5, 7 and 9) and with different TiO2 photolytic oxidizing systems (O2/UV-A/TiO2, 03/UV-A/TiO2 or UV-A/TiO2) for diclofenac removal from water, with especial emphasis on mineralization of the organic matter.
+
+RESULTS: In the case of single oxzonation processes, results show fast and practically complete elimination of diclofenac, with little differences in removal rates that depend on pH and buffering conditions. In contrast, total organic carbon (TOC) removal rates are slow and mineralization degree reaches 50% at best. As far as photocatalytic processes are concerned, diclofenac is completely removed from the aqueous solutions at high rates. However, unlike single oxzonation processes, TOC removal can reach 80%.
+
+CONCLUSION: In single oxzonation processes, direct ozone reaction is mainly responsible for diclofenac elimination. Once diclofenac has disappeared, its by-products are removed by reaction with hydroxyl radicals formed in the ozone decomposition and also from the reaction of diclofenac with ozone. In the photocatalytic processes hydroxyl radicals are responsible oxidant species of diclofenac removal as well as by-products.
+
+## INTRODUCTION
+
+Diclofenac (Fig. 1) is a popular non-steroidal anti-inflammatory pharmaceutical drug. Since it is well tolerated by the human organism, it has found widespread use around the world. Over-the-counter use is approved in some countries for minor aches and pains and fever associated with common infections.
+
+Although diclofenac is susceptible to photodegradation by complex mechanisms that depend on environmental conditions1 its presence in the water environment (triers, lakes) has been documented.1-3 Diclofenac has also been detected in many municipal sewage treatment plant effluents4-7 because classical procedures in treatment plants barely destroy it.89 Therefore, there is a need to develop new tertiary treatment processes able of removing not only diclofenac but also the degradation by-products, which could also be dangerous for humans and wildlife. Thus, the hepatotoxicity of diclofenac is at least in part due to 5- and 4'- hydroxylated by-products. Also, the diclofenac by-product 8-chlorocarbaozole-1-yl-ethanoic acid shows much higher toxicity than the parent drug.10 Until now there has been no proof that very low concentrations of diclofenac have any adverse health effects. Nevertheless, based on precautionary principles, drinking water should be free from these compounds to minimize the risk of unpredictable long-term effects.
+
+In recent years, several papers have been published dealing with diclofenac removal from water. Ozone and advanced oxidation processes such as ozone/H2O2, ozone/UV radiation or H2O2/UV radiation seem to be very effective in removal diclofenac.1-16 However these works are not strictly focused on complete elimination of diclofenac and their reaction by-products, although some of them15,16 show results regarding total organic carbon removal.
+
+The objective of the work presented here was to study the efficiency of oxzonation and different TiO2 photolytic oxidizing systems for diclofenac removal, with special emphasis on mineralization of the organic matter in water.
+
+## EXPERIMENTAL
+
+Diclofenac was obtained from Sigma-Aldrich (Spain) as the sodium salt and used as received. Ozone was generated from pure oxygen in a Sanders 301.7 model laboratory generator (Spain). As catalyst, a commercial TiO2 Degussa (Spain) P25 (70% anatase and 30% rutile) was used. The average particle size and BFT surface area, according to the manufacturer, were 30 nm and 50 m2 g-1, respectively. Diclofenac aqueous solutions (average concentration 10-4 mol L-1) was prepared in MilliQ water. In some experiments, however, the aqueous solutions were buffered at pH 5, 7 and 9 with phosphate salts.
+
+Experiments were carried out in a cylindrical borosilicate glass photo-reactor (0.45 m height, 0.08 m inside diameter), depicted in previous work.17 The reactor wall was covered with aluminium foil to avoid release of radiation. About 0.9 L of a diclofenac aqueoussolution was charged into the reactor for each experiment and the temperature was kept constant at 20 degC with a Lauda cooling thermostat system (Spain).
+
+For the processes involving ozone, an ozone-oxygen mixture was continuously bubbled into the solution through a micro-diffuser placed at the bottom of the reactor. In photolytic experiments, the aqueous solutions were irradiated with a high-pressure mercury vapour lamp (Heraeus TO718 model, Spain) immersed in a glass well placed at the centre-axis of the reactor. Although the lamp bandwidth was in the range 238-579 nm, UV-Radiation was cut off by the presence of a Duran (Spain) 50 glass immersion wall (see Table 1). The intensity of the incident radiation at 313 nm (the only wavelength involved in photolytic processes, as shown later) determined by hydrogen peroxide actinometry [18] was found to be 3 x 10-5 Einstein s-1. For experiments involving TiO2 (concentration 1.5 g L-1, the optimum concentration found in this work regarding diclofenac and total organic carbon removal), the solid was kept in suspension by magnetic stirring overnight before use.
+
+Stealthy, samples were withdrawn from the reactor and analyzed for diclofenac content, total organic carbon, hydrogen peroxide and dissolved ozone concentration. Prior to the analysis, the TiO2 present in samples was removed by centrifugation (13 000 rpm for 8 min in a 5415D Eppendorf model, Spain) and further filtration through a 0.45 mm Millex-HA filter (Millipore, Spain).
+
+Diclofenac was analyzed using a Hitachi (Spain) LaChrom Elite HPLC provided with a Phenomenex (Spain) Synergi Hydro-RP column (250 x 4.60 mm, 4 mm, 80 A), an L-2450 DA detector and EZ Chrom software for data treatment. The mobile phase was a methanol/water 30/70 v/v mixture used at a constant flow rate of 0.5 mL min-1.
+
+Dissolved ozone in aqueous solutions was analyzed by the in-digo method, [19] while the ozone in gas phase was photometrically monitored at 254 nm by means of an Anseros Ozomat analyser (Spain). Hydrogen peroxide concentration was determined through the cobalt/bicarbonate method. [20] Finally, total organic carbon (TOC) was determined by a TOC-V5Cl3 Shimadzu analyzer (Spain).
+
+## RESULTS AND DISCUSSION
+
+### Single ozonation experiments
+
+#### Diclofenac removal
+
+\(\begin{equation*}\mathrm{1}\mathrm{0}^{-4}\mathrm{mol}\,\mathrm{L}^{-1}\mathrm{ buffered}\mathrm{aqueous}\mathrm{solutions}\mathrm{solutions}\mathrm{of}\mathrm{diclofenac}\mathrm{were}\mathrm{treatedwithanoxygen-ozonemixtureatpH5,7and9.Also,anothertwoexperimentswerecarriedoutunderthefollowingthefollowingacorizationofdiclofenacbufferedsolutionsatpH7with10-3 molL-1-1-butanol(t-BuOH)andozonationofnon-buffereddiclofenacsolutions(initialpH~x5).Figure 2 shows the results obtained. Regarding the buffered experiments without t-BuOH, only for the first 2 min of ozonation diclofenac removal rates show some differences. In any case, diclofenac ozonation is a fast reaction and complete elimination is reached in 7 min without t-BuOH. On the contrary, if the solution is not buffered (initial pH~x5) diclofenac removal rate is slower and complete elimination is reached in 14 min. In the presence of t-BuOH (pH 7), diclofenac removal rate is faster and complete elimination is reached in 4 min. Some explanation can be given to these results as follows. The ozone may either react directly with organic compounds or, above a determined pH value, decompose generating radical species, such as hydroxyl radicals, that become the important oxidants. The critical values of pH above which the second type of reaction predominates, depend on the direct reactions of ozone with different substances present in the water that enhance or retard the decomposition of ozone. [21] In
+
+Figure 1.: Molecular structure of diclofenac.
+
+Figure 2.: Changes of diclofenac dimensionless concentration with time corresponding to single ozonation experiments. Conditions: temperature 20 °C gas flow rate 30 h L−1; inlet ozone gas concentration 10 mg L−1; initial diclofenac concentration 10–4 mol L−1 (average value). Symbols: pH 5 buffered; □ pH 7 buffered; □ pH 9 buffered; □ pH 5 non-buffered; □ pH 7 buffered and t-butanol 10−3 mol L−1.
+
+this work, with the exception of the experiment carried out at pH 5, ozone was not detected in the aqueous solutions of diclofenac during the first 5 min (Fig. 3). Then, it is clear that all ozone consumed in this period of time reacts directly with diclofenac and probably the first oxidation by-products. Given the activating character of amine groups for aromatic electrolytic substitution reactions, free ortho and para positions of both rings (3, 5 and 4') are strong nucleophilic points, prone to ozone attack (Fig. 1). These electrolytic substitution ozone reactions would yield the corresponding hydroxyl derivatives.13 Also, direct ozone attack may take place on the amine group.11,12,14 Partial protonation of acidic groups at pH 5 (enhanced in non-buffered reactions when the pH decreases from 5 to 3.5) deactivates the electrolytic reaction of ozone at position 3, 5 and 4' and slightly activates electrolytic ozone attack at position 6. As a consequence, removal rate of diclofenac is slower at this pH.
+
+Finally, as shown in Fig. 2, the presence of t-BuOH increases the rate of removal of diclofenac. Two explanations can be given for these results. Thus, Sein _et al._14 indicated that in the absence of t-BuOH, ozone-consuming reactions that do not contribute to a direct loss of diclofenac become much more important in the presence of hydroxyl radicals. To explain the presence of hydroxyl radicals in aqueous solutions, these authors considered the work of Munoz and von Sontnaga22 and proposed the addition of ozone to the amine group. The intermediate formed dissociates into an aminyl radical, the ozonide ion radical and a proton. The ozonide ion radical reacts with ozone and eventually generates the hydroxyl radical. However, although possible, Munoz and von Sontnaga22 consider this chain reaction as noticeable only when the rate of ozone reaction is slow, which is not the case. Another possible reason for the increase of diclofenac removal in the presence of t-butanol is the increase of the volumetric mass transfer coefficient that enhances ozone mass transfer to the water. Then the ozone available in the liquid phase increases and, as a consequence, the diclofenac removal rate also increases. Thus, Gurol and Nekouinini23 obtained a 200% increase in the oxygen mass transfer coefficient in relation to free water when a 4 x 10-4 mol L-t-BuOH was present while oxygen was bubbling in water.
+
+### TOC removal
+
+Contrary to the fast removal of diclofenac in ozonation processes (complete removal can be achieved in 7 min), TOC removal is a slower process (in 7 min less than 5% TOC removal is achieved in all cases studied; Fig. 4). Ozone breaking of hydroxylated aromatic rings leads to hydrogen peroxide21 and carboxylic acids and after 10-15 min reaction, ozone starts to accumulate in water (Fig. 3). Hydrogen peroxide can also be formed from the ozone self-decomposition reaction by elevating the pH of water. Ozone and hydrogen peroxide when simultaneously present in water, at pH values higher than 4, start to react to eventually yield hydroxyl radicals that are the main oxidizing species of carboxylic acids (for rate constant values see references 21, 24 and 25):
+
+\[\begin{matrix} {O_{3} + H\O - aromatic\ rings} \\ {HO_{2}^{-} + carboxylic\ acids} \\ \end{matrix}\]
+
+\[\begin{matrix} {O_{3} + H\O - \frac{k - 70^{-}1 - 5^{-}}{\left\{ \begin{matrix} {k - 70^{-}1 - 5^{-}} \\ {0_{2} + 0_{2} \\ \end{matrix}}} \\ \end{matrix}}H{O_{2}^{-} + O_{2}}} \\ {O_{3} + H\O_{2}^{-} \times \frac{k - 2 \times 10^{6}M^{-}1 - 5^{-}}{\left\{ \begin{matrix} {k - 2 \times 10^{6}M^{-}1 - 5^{-}} \\ {0_{2} + 0_{3}^{-} + H\O ^{-}} \\ \end{matrix}}H{O_{2}^{-} + O_{3}^{-}}}} \\ \end{matrix}\]
+
+Thus, less selective than ozone, hydroxyl radicals can react with low chain organic acids, achieving their mineralization. As a result, depending on pH, between 20 and 45% of the initial organic carbon is mineralized in about 90 min (Fig. 4). The low HO- concentration generated and the presence of phosphate salts when used as buffer (known scavengers of HO- radicals26) are responsible for the poor results achieved with regard to TOC removal. In addition, the mineralization process is more limited at pH 9 because CO2 generated in the oxidation processes remains in aqueous solution as carbonate and bicarbonate ions. These species inhibit the hydroxyl radical's action over target compounds:27
+
+\[\begin{matrix} {HCO_{3}^{-} + H\O^{\bullet}} \\ {CO_{3}^{-} + H\O^{\bullet}} \\ \end{matrix}\]
+
+\[\begin{matrix} {HCO_{3}^{-} + H\O^{\bullet}} \\ {CO_{3}^{-} + H\O^{\bullet}} \\ \end{matrix}\]
+
+### Photolytic TiO2 catalyzed experiments
+
+Since a satisfactory degree of mineralization is not achieved in single ozonation experiments, photolytic TiO2 catalyzed experiments were carried out. This system can be considered as an advanced oxidation process because the irradiation of the semiconductor titanium dioxide with an appropriate wavelength
+
+Figure 4.: Changes of total organic carbon dimensionless concentration with time corresponding to single ozonation experiments. Conditions as in Fig. 2. Symbols: **a** pH 5 buffered; **b** pH 7 buffered; **c** pH 9 buffered; **d** pH 5 non-buffered.
+
+Figure 3.: Changes of dissolved ozone concentration with time corresponding to single ozonation experiments. Conditions and symbols as in Fig. 2.
+
+in the presence of water or/and oxygen generated the hydroxyl free radical:20,29
+
+\[\begin{equation*}\mathit{TiO}_{2}+\mathit{hv}\xrightarrow{}\mathit{TiO}_{2}(e_{ab}^{-}+h_{ab}^{+})\end{equation*} \tag{6}\] \[\begin{equation*}h_{ab}^{+}+\mathit{H}_{2}\mathit{O}_{ads}\xrightarrow{}\mathit{H}_{0}^{\bullet}\mathit{H}_{ab}^{+}+H^{+}\end{equation*}\] (7) \[\begin{equation*}h_{ab}^{+}+\mathit{H}_{0ads}^{\bullet}\xrightarrow{}\mathit{H}_{0}^{\bullet}\mathit{H}_{ads}^{\bullet}\end{equation*}\] (8) \[\begin{equation*}e_{ab}^{-}+\mathit{O}_{2}\xrightarrow{}\mathit{O}_{2}^{-}\end{equation*}\] (9) \[\begin{equation*}\mathit{O}_{2}^{-}+\mathit{H}^{+}\xrightarrow{}\mathit{H}_{0}^{\bullet}\end{equation*}\] (10) \[\begin{equation*}2\mathit{H}\mathit{O}_{2}^{\bullet}\xrightarrow{}\mathit{H}_{0}2+\mathit{O}_{2}\end{equation*}\] (11) \[\begin{equation*}\mathit{H}_{2}\mathit{O}_{2}+\mathit{hv}\xrightarrow{}2\mathit{H}\mathit{O}^{\bullet}\end{equation*} \tag{12}\]
+
+The hydroxyl free radical is a powerful and non-selective oxidant and its reactions with many organic compounds, including short chain carboxylic acids, have very high rate constants.30,31 Therefore, application of TiO2 photocatalytic oxidation to diclofenac removal should lead to high mineralization levels.
+
+Because of the use of TiO2 in the experimental series, the adsorption of diclofenac was studied previously. Experimental results (not shown) indicated that, in 10-4 mol L-1 diclofenac solutions no adsorption of this organic compound takes place at pH 7. However, under non-buffered conditions (pH 5), about 20% diclofenac was adsorbed on the TiO2 surface. Therefore, non-buffered diclofenac aqueous solutions containing the appropriate amounts of TiO2 were stirred overnight to achieve adsorption equilibrium before treatment with UV radiation or/and ozone.
+
+### Diclofenac removal
+
+TiO2 photocatalytic experiments of buffered aqueous solutions (pH 7) of diclofenac were carried out by bubbling oxygen or an oxygen-ozone mixture. In addition, similar experiments were carried out under non-buffered conditions (initial pH 5). Also, a non-buffered diclofenac solution was subjected to TiO2 catalyzed photolysis without any gas being bubbled (i.e. the only oxygen source was the initially dissolved oxygen). Figure 5 shows the changes of diclofenac dimensionless concentration with time observed in these experiments. As can be seen, high diclofenac removal rates were obtained with the UV-A/TiO2 oxidation. On one hand, when oxygen is not added (that is, the only oxygen source is the initial dissolved oxygen) reaction (9) is limited after a short initial time. Thus, while there is enough dissolved oxygen, the diclofenac removal rate is high, but in a few minutes, when the oxygen is consumed, the reaction rate abruptly reduces. On the other hand, the diclofenac removal rate increases if oxygen or the oxygen-ozone mixture is fed. As a result, between 90 and 95% diclofenac is eliminated in 10 min. The 10-5% diclofenac remaining is removed at a lower rate, likely due to the mass transfer of diclofenac from the TiO2 surface to the water phase, according to the adsorption equilibrium. Under these conditions the diclofenac present in water likely competes with by-products for the hydroxyl radicals. This can be confirmed by the fact that, at pH 7, when no diclofenac is adsorbed, its removal is not only fast but also complete, between 6 and 12 min for oxygen-ozone mixture or oxygen bubbled, respectively. Then, TiO2 photocatalytic oxidation of diclofenac (buffered conditions) is the fastest process studied.
+
+### TOC removal
+
+Regarding TOC removal (Fig. 6), all the TiO2 catalyzed systems studied improve the results over those obtained if only ozone is applied (single ozonation experiments). In this sense, one can highlight the system O3/UV-A/TiO2 when applied at pH 5 (non-buffered conditions) that allows a TOC removal rate higher than 90% in 15 min. The same system under buffered conditions (pH 7) leads to only 70% TOC removal in about 60 min.
+
+As mentioned before, powerful and non-selective hydroxyl free radicals are the species responsible for mineralization. At first, according to the mechanisms proposed by Legini _et al._20 and Hoffman _et al._29 for TiO2 photocatalytic oxidation, hydroxyl free radicals are generated by two parallel paths: (a) electron transfer from adsorbed water or hydroxyl ion to h+bb hole (reactions (6) to (8)); and (b) oxygen trapping of conduction band electrons (reactions (6) and (9) to (12)). However, Baxendale and Wilson20 or Nicole _et al._18 point out the low contribution of wavelengths above 254 nm to the photolysis of the hydrogen peroxide (l = 295-299: \(\varepsilon=0.88\) L mol-1 cm-1; l = 313: \(\varepsilon=0.40\) L mol-1 cm-1; l = 365: \(\varepsilon=0.0066\) L mol-1 cm-1. Taking into account the emission spectra of the lamp (Table 1), low or negligible contribution of reaction (12) for the formation of hydroxyl radicals can be expected. Then, additional reaction(s) are necessary to explain high hydroxyl radical concentrations.
+
+Figure 5.: Changes of diclofenac dimensionless concentration with time corresponding to TiO2 photocatalytic oxidation experiments. Conditions: temperature 20 °C; gas flow rate 30 L h−1; inlet ozone gas concentration 10−1; initial diclofenac concentration 10−1; Load 10−1; Average value; TiO2 concentration: 1.5 g L−1; Symbols: \(\blacklozenge\) O3/UV-A/TiO2 (non-buffered); ▲/UV-A/TiO2 (non-buffered); ▲/UV-A/TiO2 (buffered pH 7).
+
+Figure 6.: Changes of total organic carbon dimensionless concentration with time corresponding to TiO2 photocatalytic oxidation experiments. Conditions and symbols as in Fig. 5.
+
+Hydrogen peroxide and ozone may participate in different processes that, directly or indirectly, could enhance the hydroxyl free radical concentration in the aqueous diclofenac solutions when treated with O2/UV-A/TiO2, O2/UV-A/TiO2 or UV-A/TiO2 systems. On the one hand, the superoxide ion radical can react with hydrogen peroxide as indicated by the Haber-Weiss mechanism.33
+
+\[\begin{equation*}O_{2}^{-}+H^{+}+H_{2}O_{2}\xrightarrow{}H_{2}O+O_{2}+HO^{\bullet}\end{equation*} \tag{13}\]
+
+On the other hand, ozone, when present, reacts very quickly with the superoxide ion radical to yield the ozonide ion radical that eventually leads to the hydroxyl radical.34-36
+
+\[\begin{equation*}O_{3}+O_{2}^{-}\xrightarrow{}O_{3}^{-}+O_{2}\end{equation*} \tag{14}\]
+
+\[\begin{equation*}O_{3}^{-}+H_{2}O\xrightarrow{}HO^{-}+O_{2}+HO^{\bullet}\end{equation*} \tag{15}\]
+
+\[\begin{equation*}O_{3}^{-}+H^{+}\xrightarrow{}HO_{3}^{-}\end{equation*} \tag{16}\]
+
+\[\begin{equation*}HO_{3}^{-}\xrightarrow{}O_{2}+HO^{\bullet}\end{equation*} \tag{17}\]
+
+In addition, ozone can participate in three other processes which could generate hydroxyl radicals directly or indirectly:
+
+1. By trapping conduction band electrons:37
+
+\[\begin{equation*}e_{ab}^{-}+O_{3}\xrightarrow{}O_{3}^{-}\end{equation*} \tag{18}\]
+
+and reaction (15) or (16)-(17).
+2. By direct photolysis in water which generates hydrogen peroxide:38
+
+\[\begin{equation*}O_{3}+H_{2}O+hv\xrightarrow{}H_{2}O_{2}+O_{2}\end{equation*} \tag{19}\]
+
+As the ozone extinction coefficient39 at 313 nm is 60 L mol-1 cm-1, contribution of reaction (19) is likely to be non-negligible. In any case, reaction (19) would continue with reaction (12), (13) and, especially, with reaction (20):
+3. Reacting with the hydroperoxide anion:40
+
+\[\begin{equation*}O_{3}+HO_{2}^{-}\xrightarrow{}O_{3}^{-}+HO_{2}^{\bullet}\end{equation*} \tag{20}\]
+
+The undissociated form of hydrogen peroxide reacts only very slowly with ozone.41 However, strong acceleration of the decomposition of ozone by hydrogen peroxide takes place at pH values above 5: very small concentration of hydroperoxide anion becomes kinetically effective for initiating the ozone decomposition. Staehelin and Hoigne40 calculate a critical hydrogen peroxide concentration of 10-7 mol L-1 (valid for pH < pK1022 = 11.6) above which the effect of hydroperoxide ion on the rate of ozone decomposition become kinetically effective.
+
+Hydrogen peroxide plays, therefore, an important role in the formation of hydroxyl free radicals. As a consequence, formation of hydrogen peroxide was also investigated in this work. Figure 7 shows the results obtained. In an experiment of the TiO2 photolysis of water carried out in the absence of diclofenac, very low concentrations of hydrogen peroxide were measured (symbol * in Fig. 7). This follows the mechanisms represented by reactions (6) and (9) to (12). On the contrary, when diclofenac is present in solution, significant hydrogen peroxide concentrations were determined. The curve profiles represented in Fig. 7 for the systems O2/UV-A/TiO2, O2/UV-A/TiO2 and UV-A/TiO2 are typical of an intermediate product. Thus, hydrogen peroxide concentration first increases with time and reaches a maximum value (which occurs simultaneously when diclofenac is approximately totally removed) and then it decreases to a plateau value. When only the initial dissolved oxygen is present, the hydrogen peroxide concentration measured was lower than when oxygen or oxygen-ozone mixtures are bubbled. Furthermore, under conditions of decreasing dissolved oxygen concentration, reaction (9) is limited as indicated before. Then after a few minutes of oxidation, the TOC removal also decreases and finally stops. These facts confirm, as expected, the important role of oxygen in these processes.
+
+Formation of hydrogen peroxide when ozone is present can be explained from the well established Criege mechanism of cyclo addition of ozone to unsaturated bonds,21 i.e the diclofenac rings. In a proteinic solvent such as water, the primary ozonide formed decomposes into a carbonyl compound and a zwitteriton that quickly leads to a hydroxyl-hydroperoxide compound that finally, decomposes into a carbonyl compound and hydrogen peroxide.
+
+In the absence of ozone, formation of hydrogen peroxide can be explained by the mechanism proposed by Sein _et.al._14 According to this mechanism, HO radicals, by their electrophilic nature, attack the activated sites of aromatic rings to yield a cyclohedexal-type radical. These radicals react with oxygen (which is present in excess) and the peroxyl organic radical formed decomposes into a stable hydroxylated compound and the hydroperoxide radical:
+
+\[\begin{equation*}\end{equation*}\]
+_The role of phosphates (buffer salts) and bicarbonate_
+
+As mentioned before, there are different ways to generate hydroxyl radicals. The importance of these depends on the experimental conditions applied. Thus, when a O3/UV-A/TiO2 system is applied, all possible contributions presented above could take place and TOC removal levels can rise to about 90%. The removal rate is very fast at pH 5 and slower at pH 7 because some of the substances accumulated or present in water, principally bicarbonate and phosphate ions, act as scavengers of hydroxyl radicals.26,27 Concentration of hydroxyl radicals depends on the initiation rate of radicals, _ri_, and on the scavenging factor, \(\Sigma k_{i}C_{\text{s}}\).21
+
+\[C_{i\text{PO}} = r_{i}/\sum k_{i}C_{\text{s}}\]
+
+where _C_s and _k_s are the concentration of any scavenging substance, \(S\), present in water and the rate constant of its reaction with the hydroxyl radical, respectively. In this work, the scavenging factors corresponding to bicarbonate and phosphate ions have been estimated.
+
+Taking into account the pH dependence of the H2CO3/HCO3/CO32- system (pK1 = 6.3 and pK2 = 10.3 as Lide43 reports), at pH 7, CO2 generated in oxidation processes remains as bicarbonate. Figure 8 shows the evolution of inorganic carbon (measured as bicarbonate) with time during photocatalysts experiments with dideficane. As can be seen, bicarbonate accumulates in water to reach a plateau value at the reaction time when TOC oxidation stops. The maximum bicarbonate concentration was 10.3 x 10-4 and 8.1 x 10-4 mol L-1 for dideficane O3/UV-A/TiO2 and O2/UV-A/TiO2 processes, respectively. Then, considering that the rate constant of bicarbonate ion with hydroxyl radical is 27 8.5 x 10-1 L s-1 the scavenging factor was between 8755 s-1 and 6885 s-1. Logically, this scavenging effect was not present in the oxidation processes carried out in non-buffered diclofenac solutions since at pH <5,CO2 from mineralization is mainly as carbon dioxide, which absorbs from the water due to the bubbling gas.
+
+In the same way, at pH 7, the main phosphate species are H2PO4- and HPO42- since pK2 is26.7.2. Taking into account, on the one hand, that concentrations of H2PO4- and HPO42- are 1.3 x 10-3 and 1.2 x 10-3 mol L-1, respectively, and on the other hand, rate constants of hydroxyl radicals reactions with H2PO4- and HPO42- are26 10 and 5 x 106 mol-1 L s-1, respectively, the contribution of phosphates to scavenge the hydroxyl radicals in buffered oxidation processes applied is 1.9 x 104 s-1. Compared with the non-buffered system where no phosphates were present, the scavenging factor of hydroxyl radicals is then 1.9 x 104 times higher. Thus it is not surprising that TOC removal in the buffered system is lower than in the case of the non-buffered solutions (Fig. 6).
+
+## CONCLUSIONS
+
+Major conclusions reached in this work are:
+
+* Single ozonation allows fast removal of dideficane from water. Buffering of solutions improves the removal rate.
+* However, single ozonation does not provide a satisfactory degree of elimination of TOC.
+* The use of photolytic TiO2 catalyzed systems is appropriate for dideficane and TOC removal. The O3/UV-A/TiO2 oxidation system, at pH 7, is especially recommended for this purpose.
+* The mechanism of photocatalytic ozonation involves direct ozone reaction with dideficane. Regarding TOC elimination, free radical oxidation is the main oxidation mechanism.
+
+## Acknowledgements
+
+This work has been supported by the CICYT of Spain and the ERDF of the European Commission (Project CTQ2006/04745). A. Aguinaco also thanks the Spanish Ministry of Education for a FPU grant.
+
+## References
+
+* (1) Buser HR, Fogler T and Muller MD, Occurrence and fate of the pharmaceutical drug diclofenac in surface waters: rapid photodegradation in a lake. _Environ Sci Technol_**32**:3449-3456 (1998).
+* (2) Buser HR, Muller MD and Theobald N, Occurrence of the pharmaceutical drug diclofenac acid and the herbicide mecoprop in various Swiss lakes and in the North Sea. _Environ Sci Technol_**32**:188-192 (1998).
+* (3) Ahrer W, Scheuvenik E and Buchberger W, Determination of drug residues in water by the combination of liquid chromatography or capillary electrophoresis with electro-spray mass spectrometry. _J Chromatogr_**4** 9196-978 (2001).
+* (4) Ternes TA, Occurrence of drugs in German sewage treatment plants and rivers. _Water Res_**32**:3245-3260 (1998).
+* (5) Ternes TA, Bomer M and Schmidt T, Determination of neutral pharmaceutical investiawate and rivers by liquid chromatography-electrospray tandem mass spectrometry. _J Chromatogr A_**938**:175-185 (2001).
+* (6) Hebeer T, Tracking persistent pharmaceutical residues from municipal sewage to drinking water. _J Hydrol_**266**:175-189 (2002).
+* (7) Trieer C, Singer HP, Oellers S and Muller SR, Occurrence and fate of carbamazepine, didefic acid, dideficane; ibuprofen, ketofen and naproxen in surface waters. _Environ Sci Technol_**37**:1061-1068 (2003).
+* (8) Bound P and Voluvolus N, Household disposal of pharmaceuticals as a pathway for aquatic contamination in the United Kingdom. _Environ Health_**119**:1135-1171 (2005).
+* (9) Joss A, Zabczynski S, Gobel A, Hoffmann B, Loffner D, and McArdell CS, Biological degradation of pharmaceuticals in municipal wastewater treatment: proposing a classification scheme. _Water Res_**40**:1686-1696 (2006).
+* (10) Scheuvell M, Krause S, Shah RM and Huhnerffuss H, Occurrence of diclofenac and its metabolites in surface water and effluent samples from Xarabella. _Environ Sci Technol_**37**:870-876 (2009).
+* (11) Ternes TA, Meisenheimer M, McDowell D, Sacher F, Ravach HU, and Haist-Guide B, et al., Removal of pharmaceuticals during drinking water treatment. _Environ Sci Technol_**36**:3855-3863 (2002).
+* (12) Huber MM, Canonica S, Park GF and Von Gunten U, Oxidation of pharmaceuticals during operation and advanced oxidation processes. _Environ Sci Technol_**37**:1016-1024 (2003).
+
+Figure 8.: Changes of bicarbonate concentration with time corresponding to TiO2 photocatalytic oxidation experiments. Conditions as in Fig. 5. Symbols: \(\circ\) O3/UV-A/TiO2 (buffered pH 7); \(\square\) O3/UV-A/TiO2 (buffered pH 7).
+
+- 422 (2004).
+* 6662 (2008).
+* 8306 (2005).
+* 594 (2008).
+* 1796 (2006).
+* 168 (1990).
+* 456 (1981).
+* 72 (1977).
+* 30 (2004).
+* 2033 (2000).
+* 240 (1985).
+* I: non-dissociating organic compounds. _Water Res_**17:173**
+- 183 (1983).
+* II: dissociating organic compounds. _Water Res_**17:185**
+- 194 (1983).
+* 3203 (1970).
+* 2106 (1966).
+* 698 (1993).
+* 96 (1995).
+* 386 (1988).
+* 1013 (1992).
+* 356 (1957).
+* 234 (2001).
+* and HO2/O3
+- as intermediates. _J Phys Chem_**88:2560
+- 2564 (1984).
+* as chain intermediates. _J Phys Chem_**88:5999
+- 6004 (1984).
+* 2966 (1985).
+* 1053 (1992).
+* 767 (1988).
+* 665 (1957).
+* 681 (1982).
+* 3373 (1940).
+* 1100 (1985).
+* [43] Lide DR, CRC Handbook of Chemistry and Physics, 83rd edn. CRC Press, Boca Raton (2002).

manual_annotation/200pdfs_in_mmd/10.1002_jctb.2609.mmd

@@ -0,0 +1,189 @@
+# Contaminants abatement by ozone in secondary effluents. Evaluation of second-order rate constants
+
+F. Javier Rivas, Jose Sagasti, Angel Encinas and Olga Gimeno
+
+# Abstract
+
+BACKGROUND: The ozonation of a mixture of contaminants commonly found in secondary effluents has been carried out in an artificially contaminated secondary effluent. Competitive ozonation experiments in heterogeneous and homogeneous mode have also been carried out to determine the second-order direct ozonation rate constants.
+
+RESULTS: Inlet ozone concentration, alkaline conditions and addition of 10-3 mol L-1 of H2O2 positively affected the degree of mineralization and the disappearance of chemical oxygen demand of the mixture. Reaction rates depend on pH; at pH 7 the following direct ozone rate constants were obtained: 2.7 x 105, 2.5 x 103, 2.5 x 104, 6.2 x 105, 3.2 x 105, 3.4 x 105, 8.0 x 105 and 4.6 x 105 mol-1 L1 s-1 for acetaminophen, metoprolol, caffeine, antipyrine, nonfloxacin, ketorolac, doxycycline, hydroxydiphenyl, and diclofenac, respectively.
+
+CONCLUSIONS: The positive effect on TOC and COD removal that ozone dosage exerts is not applicable to individual contaminants. An optimum ozone concentration can be found with no further improvement of the depletion rate of organics as the ozone inlet concentration is increased. Carbonates affect the oxidation of recalcitrant compounds like atrazine. Carbonate concentration must be considered when dealing with real effluents. Addition of hydrogen peroxide can increase the mineralization level obtained, however the increase in complexity and costs does not justify its addition.
+
+## INTRODUCTION
+
+Effluents from secondary waste-water treatment often contain a number of substances that have existed the biological stage. Some of these are categorized as endocrine disruptor compounds which encompass a variety of chemical categories, including natural and synthetic hormones, plant constituents, pesticides, plasticizers, personal care products, and other industrial by-products and pollutants.1
+
+Footnote 1: [https://www.ncbi.nlm.nih.gov/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub//pub/pub/pub/pub/pub//pub/pub/pub//pub/pub/pub//pub/pub/pub//pub/pub//pub/pub//pub/pub//pub/pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub///pub///pub](https://www.ncbi.nlm.nih.gov/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub/pub//pub/pub/pub/pub/pub//pub/pub/pub//pub/pub/pub//pub/pub/pub//pub/pub//pub/pub//pub/pub//pub/pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub//pub///pub///pub)(Madrid, Spain). The main properties of this effluent after filtration through 0.45 mm filters were: total organic carbon (TOC): 12.3 +- 0.8 ppm; chemical oxygen demand (COD): 26.5 +- 0.5 and 15.6 +- 3.1 ppm (before and after filtration through 0.45 mm filters); pH = 7.8 +- 0.2 ppm; inorganic carbon (IC): 17.5 +- 4.2 ppm, biological oxygen demand after 5 days (BOD): 4.1 +- 0.5 ppm.
+
+Acetaminophen, metoprolol, caffeine, antipyrine, sulfamethoxazole, flumequine, ketorolac, atrazine, hydroxydiphenylphenyl and diclofenac were purchased from Sigma Aldrich and used as received. Figure 1 shows the molecular structure of the contaminants used.
+
+Solution of pollutants in the secondary effluent was carried out by dissolving calculated amounts of the organics to obtain a final theoretical concentration of 10 ppm in each compound, in some cases, due to low solubility values, saturation conditions were used.
+
+### Experimental setup
+
+Variable influence experiments were carried out in semi-batch mode in a 1 L capacity borosilicate glass reactor. The vessel was equipped with a porous plate to bubble an oxygen-ozone gas mixture, sampling port, thermometer, outlet gas and mechanical agitation (Heidolph R2R 2020 stirrer). Details of the contact fluid-dynamic properties can be found elsewhere.4 Temperature was controlled at 20 +- 0.1 deg by immersion of the reactor into a thermostatic batch. Ozone was generated in an Erwin Sander 301.7 laboratory ozone generator capable of producing up to 12 g h-1 of ozone from pure oxygen. Samples were withdrawn from the reactor and analysed for concentration of contaminants, COD, TOC, absorbance at 254 nm, and dissolved ozone concentration.
+
+Homogeneous kinetic runs were conducted in semi-continuous mode. A high performance liquid chromatography (HPLC) pump was used to feed a constant flow-rate from an ozone-saturated aqueous solution. The solution was stabilized by the addition of a predetermined amount of tert-butyl alcohol. The aqueous solution was introduced into a perfectly mixed reactor containing an aqueous solution of the target and reference compounds.
+
+### Analytical procedure
+
+Dissolved ozone was measured by following the decolouration of 5,5,7- indigortisulphonate using a Thermo Spectronic Helios a spectrophotometer. Ozone in the gas phase was monitored with an Anseros Ozomat ozone analyser. The analysis was based on the absorbance at 254 nm. Chemical oxygen demand (COD) was determined in a Dr Lange spectrophotometer, using the standard dichromate reflux method. Biological oxygen demand (BOD) was measured following the procedure of the Warburg respirometer. For this purpose, non-acclimated microorganisms from the municipal waste-water plant of the city of Badajoz were used. In order to assess the degree of mineralization, total organic carbon (TOC) was determined by a Shimadzu TOC 5000A analyser by directly injecting the aqueous solution. Reaction media pH was measured using a Crison 507 pH meter. Details of the analytical methods applied can be found elsewhere.8
+
+Pollutants were determined by HPLC (WIR Elite La Chrom L - 2455). The mobile phase at a flow-rate of 0.2 mL min-1 was pumped through a Gemini SU C18 110R column. The initial composition was acetonitrile (10%) and acidified (25 mmol L-1 formic acid) water (90%). A gradient was applied to obtain acetonitrile 100% after 40 min. The following wavelengths (nm) were used to quantify the different contaminants: acetaminophen, 245; metoprolol, caffeine, 205; antipyrine, 243; sulfamethoxazole, 267; flumequine, 248; ketorolac, 321; atrazine, 223; hydroxybiphenyl, 243; diclofenac, 277. Error bars of 95% confidence in some figures are calculated from duplicated experiments and three analyses for each data point.
+
+## RESULTS AND DISCUSSION
+
+### Influence of operating conditions
+
+#### Effect of ozone dosage
+
+Experiments were carried out to assess the influence of ozone inlet concentration on the efficiency of the oxidation process. Ozone concentration was varied over the interval 5-40 ppm
+
+Figure 1.: Molecular structure of contaminants.
+
+keeping the other operating variables constant. Figure 2 shows the evolution of the different contaminants in solution. As seen, regardless of the operating conditions used, all of the compounds treated are easily removed from water by the simple application of ozone. Apparently, the most reactive substances are sulfamethoxazole, diclofenac and hydroxydiphenyl while caffeine, atrazine, flumequine and antipyrine show a slightly higher recalcitrance towards ozone attack. A rough way to compare the reactivity of the different contaminants was carried out by determining the simple pseudo-first-order rate constant. Thus, plotting the natural logarithm of the normalized concentration versus time gave the results shown in Table 1. From this data the following reactivity order is deduced: diclofenac \(\approx\) hydroxydiphenyl > ketorolac \(\approx\) sulfamethoxazole > acetaminophen > metoprolol > antipyrine \(\approx\) caffeine \(\approx\) flumequine > atrazine. It has to be pointed out that this order of reactivity does not correspond to the direct reaction ozone + contaminant but to the simultaneous action of molecular ozone and generated hydroxyl radicals under the conditions investigated.
+
+As a rule of thumb, ozone feed concentration exerts a positive effect on the removal rate of contaminants when increased from 5 to 30 ppm, however, further increase in CO2/plate to 40 ppm does not provide further enhancement of the process in terms of the conversion rate of organics. A different trend is, however, experienced when analysing the TOC and COD evolution profiles. Thus, Fig. 2 reveals that both parameters are reduced to a greater extent as the ozone feed concentration is increased from 5 to 40 ppm. From Fig. 2 it is observed that the maximum elimination of COD and TOC are of the order 80% and 58%, respectively. Hence, not all the organics oxidized are converted to carbon dioxide and water. The parameters average oxidation state of carbon (AOSC) and partial oxidation efficiency (\(\mu_{\text{CO}\text{parton}}\)) give information about the extent of the oxidation taking place.
+
+AOSC is defined as (units in ppm):
+
+\[\text{AOSC} = \frac{4(\text{TOC} - \frac{12}{32}\text{COD})}{\text{TOC}}\]
+
+AOSC may vary throughout the oxidation period from the minimum value -4 (for methane) to the maximum value +4 (for
+
+\begin{table}
+\begin{tabular}{c c c c c}  & CO2/plate & CO2/plate & CO2/plate & CO2/plate \\  & (ppm) & (ppm) & (ppm) & (ppm) \\  & 5.0 & 15.0 & 30.0 & 40.0 \\ acetaminophen & 0.097 (0.99) & – & 0.257 (0.98) & – \\ meteorolol & 0.104 (0.98) & 0.117 (0.98) & 0.147 (0.98) & 0.188 (0.99) \\ caffeine & 0.046 (0.98) & 0.068 (0.97) & 0.156 (0.99) & 0.138 (0.99) \\ antipyrine & 0.062 (0.99) & 0.099 (0.96) & 0.129 (0.88) & 0.153 (0.99) \\ sulfamethoxazole & 0.146 (0.99) & 0.266 (0.99) & 0.350 (0.99) & 0.351 (0.99) \\ flumequine & 0.034 (0.97) & 0.073 (0.99) & 0.159 (0.99) & 0.139 (0.99) \\ ketorolac & 0.144 (0.99) & 0.267 (0.97) & 0.404 (0.97) & 0.379 (0.98) \\ atrazine & 0.021 (0.99) & 0.031 (0.99) & 0.070 (0.99) & 0.059 (0.99) \\ hydroxydiphenyl & 0.226 (0.97) & – & 0.387 (0.96) & 0.410 (0.99) \\ diclofenac & 0.228 (0.98) & 0.353 (0.98) & 0.415 (0.99) & 0.420 (0.99) \\ \end{tabular}
+\end{table}
+Table 1: Apparent pseudo-first-order rate constants (min−1) for the ozonation of contaminants in secondary effluents at various inlet ozone concentrations (experimental conditions as in Fig. 2). **R**_a_ values in parentheses
+
+Figure 2: Ozonation of contaminants in secondary effluents. Experimental conditions: \(C_{\text{C}\text{out}} = 10\) ppm, pH = 7.7, \(T = 293\) K, \(V = 1\). Effect of ozone inlet concentration (ppm): \(\bullet\), 5.0, \(\blacksquare\), 15.0; \(\blacktriangle\), 30.0; \(\triangledown\), 40.0.
+
+carbon dioxide). The variations can be computed by (units in ppm):
+
+\[\Delta\left[ \text{A}\text{O}\text{S}\text{C}\text{C}\text{o}\text{C}\text{o}\text{-}\frac{\text{C}\text{O}\text{D}_{\text{o}}}{\text{TOC}\text{o}\text{-}\text{C}\text{O}}\text{12}\t-BuOH addition should have a negative effect on their _a_onation rate in accordance with the observed induction period. It is hypothesized that after the induction period, a kind of radical mediated process starts which is not scavenged by the tertiary alcohol. The nature of the generated radicals is unknown although the formation of organic radicals or even hydroperoxyl radicals can be considered after an initial direct _a_onation period:
+
+\[\text{RH} + \text{O}_{1}\xrightarrow{}\text{ROOH}\xrightarrow{}\text{R}^{\circ} + \text{HOO}^{\circ}\]
+
+\[\text{RH} + \text{HOO}^{\circ}\xrightarrow{}\text{R}^{\circ} + \text{H}_{\text{O}}\text{O}_{2}\]
+
+\[\text{R}^{\circ} + \text{O}_{2}\xrightarrow{}\text{ROO}^{\circ}\]
+
+\[\text{ROO}^{\circ} + \text{RH}\xrightarrow{}\text{ROOH}\]
+
+#### Effect of addition of free radical promoters
+
+To check the possibility of enhancing the elimination of COD and TOC, hydrogen peroxide was added at the beginning and after oxidation of the parent compounds in two additional series of experiments. Hydrogen peroxide concentration was tested at two levels (10-3 and 10-4 mol L-1). Regarding the concentration of the parent compounds, it has to be pointed out that hydrogen peroxide did not appreciably influence the evolution profiles of the parent organic compounds, however, some differences were experienced when the degree of mineralization was monitored.
+
+Figure 6 illustrates the results obtained in terms of normalized TOC evolution when H2O2 was added to the reactor. As observed from this figure, TOC conversion increased from roughly 60% (non-promoted run) to >80% when 10-3 mol L-1 of hydrogen peroxide was added to the reactor. Then, a further amount of H2O2 was added after 60 min to enhance, if possible, the beneficial effect of the promoter. As seen in the figure, a concentration of hydrogen peroxide of 10-4 mol L-1 was not efficient enough to improve the mineralization level when compared with single _a_onation. Also, results obtained with the addition of 10-3 mol L-1 of the promoter after 60 min were similar to those obtained when a similar amount of H2O2 was initially present. In any case, further optimization work would be valuable on this particular system.
+
+#### Kinetic considerations in ultrapure water
+
+#### Reactivity of the molecular forms of contaminants in heterogeneous experiments
+
+In a first attempt to compare the reactivity with ozone of the molecular forms of the contaminants studied, three experimental series were conducted at pH 2 by grouping substances that, _a priori_, are expected to have similar reactivity. Considering the classic mechanism of _a_onation, acidic conditions at pH 2 should be enough to prevent or minimize the indirect route of oxidation through hydroxyl radicals, although this is a matter of controversy.11
+
+Footnote 11: **R6**: 1058 – 1066
+
+Diclofenac, acetaminophen, sulfamethoxazole, ketorolac and hydroxybiphenyl were simultaneously _a_zonated under pH controlled conditions. The well-known competitive approach was used to calculate direct rate constants. However, it has to be pointed out that this competitive method is only valid when the kinetics occur in the slow or the fast pseudo-first-order regime,12 in the latter case, when the Hata (Ha) number is within the interval 3-5_i_/2, (the instantaneous reaction factor, _Ei_, is defined in Equation (10):
+
+\[\text{Ha} = \frac{\sqrt{\text{k}_{\text{O}_{3}}\,\text{D}_{\text{O}_{3}}\,\text{C}_{\text{R}}}}{\text{k}_{\text{K}}}\]
+
+\[\text{E}_{\text{i}} = \sqrt{\frac{\text{D}_{\text{O}_{3}}}{\text{D}_{\text{R}}}}\left[1+\frac{\text{D}_{\text{R}}\,\text{C}_{\text{R}}}{2_{\text{B}}\,\text{D}_{\text{O}_{3}}\,\text{C}_{\text{O}_{3}}}\right]\]
+
+Figure 4: Ozonation of contaminants in secondary effluents. Experimental conditions: _C_G_orm = 10 ppm, _C_O_ij_int = 40.0 ppm, \(T\) = 293 K, \(V\) = 1 L. Effect of initial pH: **•**, 2:0, **⃝**, 11:0, **⃝**, 7.8.
+
+In these expressions, _Di_ stands for diffusivity of species \(i\), \(k\)03 the direct rate ozonation constant referred to ozone disappearance, _k_i the individual liquid phase mass transfer coefficient and _Ci_ the concentration of species \(i\). The asterisk refers to equilibrium concentration in the gas-liquid interface.
+
+Table 3 shows the results after plotting the natural logarithm of normalized concentrations of contaminants versus the corresponding logarithm for sulfamethoxazole, the latter taken as the reference compound. The literature value13 for the direct ozonation constant of diclofenac is reported at neutral pH as \(k\)03 = 6.8 x 105 mol-1 L s-1. The rate constant for sulfamethoxazole at pH 2 has been reported14 to be 2.65 x 105 mol-1 L s-1. Based on the values reported by Dodd and co-workers3 for the mono-protonated (4.7 x 104 mol-1 L s-1) and deprotonated
+
+Figure 5.: Ozonation of contaminants in secondary effluents. Effect of addition of scavengers. Experimental conditions: _C_corr_i = 10 ppm, _C_O_ijsite = 40.0 ppm, pH = 7.7, \(T\) = 293 K, \(V\) = 1 L Scavengers (0.01 mol−1): □ carbonates; □ tert-butanol; □ none.
+
+species (\(5.7\times 10^{5}\) mol\({}^{-1}\) L s\({}^{-1}\)), at pH \(=2\) the sulfamethoxazole direct rate constant (based on the dissociation extent) can be calculated as \(3.7\times 10^{5}\) mol\({}^{-1}\) L s\({}^{-1}\). From Table 3 the slope of the Naperian logarithm for the normalized diclofenac concentration against the corresponding logarithm for sulfamethoxazole is 1.56 (R\({}^{2}\) = 0.99), consequently, assuming stoichiometric coefficients as unity, the \(k_{03}\) value for diclofenac should be in the interval \(4.1\)-\(5.8\times 10^{5}\) mol\({}^{-1}\) L s\({}^{-1}\), slightly below the reported value for neutral pH.
+
+In a similar way the rest of rate constants were obtained. A comparison of slope values in Table 3 and \(k\) ratios from Table 2 at pH 2 demonstrates an acceptable agreement in trends with some exceptions. Thus, when sulfamethoxazole is taken as the reference compound, relative reactivity (pseudo-first-order \(k\)-ratios) for diclofenac, acetaminophen, ketorolac and hydroxybiphenyl are 1.66, 1.16, 2.48, and 0.52 compared with 1.56, 0.79, 1.39 and 0.42 obtained form the simultaneous oxzonation of only these substances. When diclofenac was the reference pollutant, metoprolol and antipyrine showed \(k\)-ratios of 0.18 and 0.54, which acceptably compare with 0.15 and 0.31 obtained from the oxzonation of this second group. In any case, it seems that the simultaneous presence of contaminants influences their own reactivity. Thus, rate constants obtained in competitive experiments should be viewed with caution.
+
+### Reactivity of contaminants in homogeneous experiments
+
+Fulfilling the requirements to achieve a fast pseudo-first-order regime in heterogeneous experiments involves: (a) the manipulation of the reactant concentrations, which in some cases is not applicable (analytical limitations, solubility limitations, etc.); (b) knowledge of the fluid-dynamic properties of the reactor and the potential errors associated with their calculations; and (c) knowledge of the ozone concentration in equilibrium with the ozone outlet concentration, which in most cases cannot be calculated because of the lack of ozone at the reactor exit. To overcome all these drawbacks, rough approximations of the above parameters concentrations are normally considered together with the probable errors associated with these approximations.
+
+Therefore, in an attempt to corroborate the kinetic data, it was decided to conduct a series of oxzonation experiments in the homogeneous mode and in the presence of tert-butyl alcohol (a HO\({}^{\circ}\) radical scavenger). In an attempt to avoid the influence of the addition of various organics in their simultaneous oxidation, ozonation experiments were carried out in the presence of just two contaminants in the semi-continuous mode. The semi-continuous method applied allows for the attainment of several experimental points from a unique run. Thus, conversion values for the reference and target compounds can be ranged from almost 0 to 100%. Accordingly, errors associated with the presence of induction periods, analytical deviations, etc. are reduced/minimized when data points are analysed statistically.
+
+The mass balance describing the system is:
+
+\[\begin{array}{l}\sqrt{\frac{d(c_{\text{T}})}{dt}}+C_{\text{T}}\frac{d(\eta)}{dt}=-z_{\text{T}}V_{\text{K0}_{2}\text{C}_{\text{T}}}\text{C}_{\text{O}_{3}}\text{C}_{\text{T}}\\ \sqrt{\frac{d(c_{\text{R}})}{dt}}+C_{\text{R}}\frac{d(\eta)}{dt}=-z_{\text{R}}V_{\text{K0}_{3}\text{C}_{\text{R}}}\text{C}_{\text{O}_{3}}\text{C}_{\text{R}}\\ \frac{d(\eta)}{dt}=Q\end{array} \tag{11}\]
+
+In the above equations, \(Q\) is the flow-rate, \(V\) is the reaction volume (isolated with premetrization in a differential expression to indicate its variability), \(C\) is concentration, \(z\), the stoichiometric coefficient, and the subscripts T and R refer to target and reference compounds, respectively. To eliminate the unknown \(C_{\text{O}_{3}}\) concentration, if the reference and target compound mass balances are divided, after integration it follows that:
+
+\[\text{In}\,\frac{V_{\text{C}}}{V_{\text{C}}|_{\text{I=0}}}=\frac{z_{\text{T}}}{z_{\text{R}}}\frac{k_{03}c_{\text{T}}}{k_{03}c_{\text{R}}}\text{In}\,\frac{V_{\text{K}}}{V_{\text{K}}|_{\text{I=0}}} \tag{12}\]
+
+A plot of the left-hand side of Equation (12) versus \(\text{In}\,\frac{V_{\text{K}}}{V_{\text{K}}|_{\text{I=0}}}\) should lead to a straight line of slope \(\frac{Z_{\text{T}}}{Z_{\text{R}}}\frac{k_{03}c_{\text{T}}}{k_{03}c_{\text{R}}}\). As an example, Fig. 7 shows the corresponding plots obtained at pH 7. As observed, good linear correlation is obtained in most cases provided that target and reference compounds have a similar reactivity.
+
+When the target and reference compounds have significantly different reactivities (titalics in Table 4), rate constants should be taken with caution. Thus, in some cases the most reactive species can provoke a synergistic effect on the rate of removal of the more recalcitrant compound.
+
+\begin{table}
+\begin{tabular}{l c c} Compound & Slope (R\({}^{2}\)) & \(k_{03}\) mol\({}^{-1}\) L s\({}^{-1}\) \\ acetaminophena & 0.79 (0.99) & \(2.1\)–\(2.9\times 10^{5}\) \\ metoprololc & 0.15 (0.98) & \(6.2\)–\(8.7\times 10^{4}\) \\ caffeinec & _0.05 (0.99)_ & \(2.1\)–\(2.9\times 10^{4}\) \\ antipyrinec & 0.31 (0.98) & \(1.3\)–\(1.8\times 10^{5}\) \\ sulfamethoxazole & – & \(2.7\)–\(3.7\times 10^{5}\) \\ flumequinb & 1.06 (0.99) & \(6.4\) \\ ketorolaca & 1.39 (1.00) & \(3.7\)–\(5.1\times 10^{5}\) \\ atrazine & – & \(6.0\) \\ hydroxybiphenylb & 0.42 (0.98) & \(1.1\)–\(1.6\times 10^{5}\) \\ diclofenacc & 1.56 (0.98) & \(4.1\)–\(5.8\times 10^{5}\) \\ \end{tabular}
+\end{table}
+Table 3: Second-order rate constants (mol\({}^{-1}\) L s\({}^{-1}\)) for the oxzonation of contaminants in secondary effluents. Heterogeneous experiments at pH 2
+
+Figure 6: Oxonation of contaminants in secondary effluents, Effect of H\({}_{2}\)O\({}_{2}\) addition Experimental conditions: \(C_{\text{cm}_{1}}=10\) ppm, \(C_{\text{Q}_{\text{inter}}}=40.0\) ppm, pH \(=7.7\), \(T=293\) K, \(V=1\). H\({}_{2}\)O\({}_{2}\) addition: \(\bullet\) none; \(\bullet\) \(10^{-3}\) mol L\({}^{-1}\); \(10^{-3}\) mol L\({}^{-1}\) after 60 min; \(\bullet\) \(10^{-4}\) mol L\({}^{-1}\) after 60 min.
+
+A literature survey reported values of _k_ox for diclofenac2,13 at pH 7 of 6.8 x 105 and >1.0 x 106 mol-1 L s-1, for caffeine15 at pH 8 of 65 mol-1 L s-1 and for metropolit 16,17 239-330 (pH 2.5), 273 (pH 5.0), 1.4-2.0 x 103 (pH 7.0), 1.3 x 105 (pH 9.0) and 8.6 x 105 mol-1 L s-1 (pH 10). With the exception of caffeine, results obtained in this study are comparable with reported values.
+
+Comparing the rate constants obtained in homogeneous and heterogeneous experiments, the following discrepancies can be observed. Metoprolol, caffeine and hydroxybiphenyl show an abnormally high rate constant in heterogeneous experiments compared with values obtained from homogenous reactions. These differences can be attributed to the synergistic effect of the simultaneous ozonation of organics in heterogeneous runs, the different aqueous matrix used in the experimental series, the significantly different reactivity of the contaminant and reference compound (in the case of caffeine), the absence of tert-butyl alcohol in heterogeneous experiments or even the development of inadequate kinetic regimes in the latter runs. Additionally, for ketorolac and antipyrine, rate constants at pH 11 are slightly below the values calculated at pH 2 and pH 7, which is quite unlikely.
+
+Given the difficulties in obtaining reliable data on kinetics, more work on this particular subject is recommended.
+
+## Conclusions
+
+From this work the following conclusions can be drawn:
+
+At the operating conditions investigated, ozone dosage exerts a positive effect on TOC and COD removal. This statement is not applicable to individual contaminants. An optimum ozone concentration can be found with no further improvement of the rate of depletion of the organics as ozone inlet concentration is increased.
+
+Acidic conditions favour the elimination of highly reactive contaminants while alkaline-neutral pH is best suited to recalcitrant compounds and TOC-COD removal.
+
+The presence of tert-butyl alcohol also favours the disappearance of highly reactive compounds. The presence of carbonates mainly affects the oxidation of recalcitrant compounds like atrazine.
+
+\begin{table}
+\begin{tabular}{c c c c c c}  & pH 2 & pH 7 & pH 11 \\  & Slope (R2) & \(k_{02}\) & \(k_{02}\) & \(k_{02}\) \\  & Slope (R2) & mol−1 L s−1 & Slope (R2) & mol−1 L s−1 \\ \hline acetaminophen & 1.35 (0.99)a & 3.58 × 105 & 0.65 (0.99)a & 2.70 × 105 & 3.38 (0.98)a & 1.57 × 105 \\ meteoroprolol & 0.28 (0.97)a & 372 & _0.01 (0.99)b & 2.49 × 105 & 0.36 (0.99)a & 1.69 × 105 \\ Caffeine & 0.07 (0.98)a & 1.75 × 104 & _0.06 (0.95)a & 2.49 × 104 & 0.18 (0.97)a & 2.63 × 104 \\  & 0.31 (0.96)a & 1.16 × 103 & & & & \\  & 0.89 (0.99)f & 1.17 × 103 & & & & \\ antipyrine & 1.54 (0.99)a & 4.07 × 105 & 2.51 (0.99)b & 6.15 × 105 & 0.26 (0.88)a & 1.22 × 105 \\ sulfamethoxazole & – & 2.65 × 105 & – & 4.15 × 105 & – & 4.65 × 105 \\ nonfloxacin & 0.50 (0.99)f & 6.54 × 102 & 9.1 (0.97)c & 1.10 × 104 & 1.71 (0.98)a & 7.98 × 105 \\  & & 0.78 (0.85)a & 3.24 × 105 & & & \\ ketorolac & 1.33 (0.99)a & 3.51 × 105 & 0.82 (0.99)a & 3.40 × 105 & 0.31 (0.99)a & 1.46 × 105 \\ doxycycline & 0.97 (0.95)a & 2.57 × 105 & 1.93 (0.98)a & 8.00 × 105 & _5.19 (0.73)a & 7.56 × 105 \\  & & 2.82 (0.93)a & 7.76 × 105 & & & \\ hydroxybiphenyl & _0.04_ (0.79)a & 1.01 × 104 & 0.59 (0.99)a & 2.45 × 105 & 3.34 (0.98)a & 1.55 × 105 \\  & 1.51 (0.98)a & 5.60 × 102 & & & & \\  & 0.91 (0.99)f & 1.20 × 103 & & & & \\ diclofenac & – & 1.11 (0.99)a & 4.60 × 105 & 0.89 (0.90)a & 4.12 × 105 \\ \end{tabular}
+
+* Reference: sulfamethoxazole (taken from Ref. 14; other values in Refs 2 and 3).
+* Reference: hydroxybiphenyl (this work).
+* Reference: caffeine (this work).
+* Reference: nonfloxacin (this work).
+* Reference: metoprolol (this work).
+* Reference: phenol (from Ref. 12).
+* Reference: ketorolac (this work).
+
+\end{table}
+Table 4: Second-order rate constants (mol−1 L s−1) for the ozonation of contaminants in pure water. Homogenous experiments in the presence of 0.01 mol L−1-bBuOH
+
+Figure 7: Competitive homogeneous oxonation of contaminants in ultrapure water in the presence of 0.01 mol L−1 t-butanol at pH = 7.0. Reference-target: sulfamethoxazole-acetaminophen: hydrobiphenyl–antipyrine; a caffeine–nonfloxacin; □ sulfamethazole–ketorolac; □ sulfamethoxazole–doxycline; ○ sulfamethoxazole–ddiclofenac.
+
+Addition of hydrogen peroxide can increase the mineralization level obtained in the ozonation process.
+
+Drawbacks associated with the determination of rate constants in heterogeneous experiments can be overcome by conducting experiments in a homogeneous mode. In any case, competitive experiments must be carried out with reference and target compounds of similar reactivity.
+
+## Acknowledgements
+
+The authors thank the economic support received from the CICYT of Spain and European FEDER Funds through projects CTO 2006/04745 and CSD20060044.
+
+## References
+
+* [1] WHO, Global assessment of the state of the science of endocrine disruptors. www.who.int/psy/publications/endocrine_disruptors/endocrine_disruptors/. [Accessed in October 2010].
+* 1024 (2003).
+* 1977 (2006).
+* 656 (2009).
+* 794 (2008).
+* 5480 (2009).
+* 1637 (2010).
+* 4834 (2003).
+* 29 (2007).
+* 4382 (2009).
+* 9 (2005).
+* [12] Beltran FJ, Ozone Reaction Kinetics for Water and Wastewater Systems. Lewis Publishers/CRC Press, Boca Raton, FL (2004).
+* 6662 (2008).
+* 270 (2009).
+* 4717 (2009).
+* 59 (2009).
+* 3012 (2008).

manual_annotation/200pdfs_in_mmd/10.1002_jctb.539.mmd

@@ -0,0 +1,220 @@
+# Advanced oxidation with ozone of 1,3,6-naphthalenetrisulfonic acid in aqueous solution
+
+M Sanchez-Polo
+
+1Departamento de Quimica Inorganica, Facultad de Ciencias, Universidad de Granada, 18071, Granada, Spain
+
+J Rivera-Utrilla
+
+1Departamento de Quimica Inorganica, Facultad de Ciencias, Universidad de Granada, 18071, Granada, Spain
+
+CA Zaror
+
+2Centro Eula, Universidad de Concepcion, Concepcion, Chile2
+
+###### Abstract
+
+The kinetics of oxidation with ozone of 1,3,6-naphthalenetrisulfonic acid was analysed by studying the influence of different experimental parameters such as the concentration of _tert_-butyl alcohol (2-methyl-2-propanol), initial concentration of the acid, pH, and temperature. The rate constant of the direct reaction at 25 degC was calculated (_k_D=6.72a-1s-1). The constant of the free radical reaction was determined with the competitive kinetics method, using sodium 4-chlorobenzoate as reference compound, obtaining a value of _k_Om1=3.7x105M-1s-1. It was demonstrated that even at very acid pH values, 80% of the 1,3,6-naphthalenetrisulfonic acid was degraded by free radical reactions, so that the ozonation of this acid may be considered an advanced oxidation process.
+
+kinetics; 1,3,6-naphthalenetrisulfonic acid; ozone; oxidation; hydroxyl radical 2002 Society of Chemical Industry 77148-154 (online: 2002) doi: 10.1002/jctb.539
+
+## 1 Introduction
+
+Effluents from the textile industry present elevated concentrations of sulfonated polyphenols,1 ranging between 10 and 80 mg dm-3 and with a mean concentration of 31 mg dm-3. These effluents are composed of a great variety of chemical compounds, including naphthalenesulfonic acids.
+
+Due to their deactivating nature, sulfonic acid groups give these compounds a high solubility in water and a low reactivity against electrophilic addition reactions. This makes these compounds highly resistant to biological treatments. Reemtsma _et al_1 reported that only monosubstituted compounds can be eliminated by conventional treatments, whereas the di- and tri-sulfonates remain unaltered for a long time. Lange _et al_2 observed that the concentrations of these compounds at the point of entry into the river Rhine were similar to the concentrations measured 40 km downstream.
+
+Their low biodegradability and possible toxicity means that these contaminants are considered of environmental importance. Some researchers3 proposed the concentration of 1,3,6 naphthaleneetrisulfonic acid (NTS) as a parameter in the quality control of water for human consumption.
+
+NTS is used by the textile industry to synthesise azocolorants and has been detected in surface waters at concentrations of 20-100 ng cm-3.
+
+Conventional water treatments are not effective in eliminating this type of compound, for the above-stated reasons, so that treatments with oxidising agents are necessary. Ozonation has been successfully used in the destruction of numerous compounds with aromatic rings, with the resultant formation of molecules that are more biodegradable. Reactions of ozone with organic molecules in aqueous solution can occur by two mechanisms:4,5 through direct attack of the molecular ozone by cycloaddition or electrophilic reaction; and through indirect attack by free radicals, mainly hydroxyl radicals produced by the decomposition of the ozone in aqueous medium.6
+
+Ozone is very costly, therefore, the kinetic parameters of the reactions of ozone to oxidise pollutant organic compounds need to be determined in order to establish the feasibility of its use in water treatment and optimise the parameters required in the design of reactors. The purpose of the present study was to study the process of ozonation of NTS, selected as representative of this family of compounds. To this end, we studied the influence of different experimental parameters on the degradation process of NTS, the kinetics of this process, and the mechanism of reaction between the ozone and this contaminant. This study is part of a wider ongoing project on the decontamination of waters containing NTS through its ozonation and on the role of adsorbent solids in this process.
+
+## 2 Experimental
+
+### Materials
+
+The 1,3,6-naphthalenetrisulfonic acid was supplied by Fluka. The sodium 4-chlorobenzoate (PCBA) wassupplied by Merck. In the liquid chromatography, tetrabutylammonium bromide, TBABr (Fluka), was used as ionic exchanger and NaH2PO4 (Merck) as pH regulator in the mobile phase. _tert_-Butyl alcohol (2-methyl-2-propanol, T-BuOH) (Merck) was used to scavenge radicals.
+
+### 2.2 Experimental system
+
+The experimental system (Fig 1) consisted of a 2 dm3 agitation reactor, covered for temperature control and an OZOKAV ozone generator with a maximum capacity of 76 mg O3 min-1 operating at 25 degC and 1 atm of pressure, with sampling accessories, gas inlet and outlet, and reactive feed. The reactor operates in semi-continuous mode and is fitted with four equally spaced baffles to avoid vortexes.
+
+In each experiment, the reactor was filled with 1 dm3 of aqueous solution, which was buffered with the appropriate amount of orthophosphoric acid and sodium hydroxide to attain the desired pH values (2, 7, 9).
+
+Once the desired temperature and partial ozone pressure were adjusted, the gas stream was introduced into the reactor for 35 min, until saturation of the aqueous solution was attained. Once cm3 of an NTS solution (25 g dm-3) was then injected and the study commenced. Several samples were withdrawn from the reactor at regular time intervals to analyse concentrations of the compound under study and of the dissolved ozone. The ozonation reaction was blocked with the use of sodium nitrite.
+
+For the determination of the stoichiometric coefficient of the reaction, the reactor was used in discontinuous mode, and the experiments were conducted by mixing aqueous solutions of ozone and contaminant at known concentrations. The ozone solution was prepared by bubbling a stream of ozone-oxygen through ultrapure water until saturation was attained, while variable volumes of concentrated solution of the compound (25 g dm-3) were injected. At 120 s, a sample was taken, the amount of ozone and contaminant consumed being measured. As in the semi-continuous experiments, the ozonation reaction was blocked with sodium nitrite.
+
+### 2.3 Analytical method
+
+The ozone concentration in the gas mixture was analysed spectrophotometrically,7 using a Genesis 5 model Spectronic spectrophotometer. The concentration of ozone dissolved in the aqueous solutions was determined colorimetrically, using the Karman-Indigo method.8
+
+The concentrations of NTS and sodium 4-chloro-benzoate were followed using a Merck-Hitachi apparatus with UV detection. The mobile phase consisted of a solution of methanol-water (35/65), containing 5 mm TBABr as ionic exchanger and 10 mm NaH2PO4 as pH regulator, with a flow of 1.3 cm3 min-1.
+
+## 3 RESULTS AND DISCUSSION
+
+### 3.1 Influence of operating variables
+
+NTS ozonation experiments were conducted, modifying the concentration of radical scavengers (0-0.2 m _tert_-butyl alcohol), the initial concentration of NTS (5.75 x 10-5 -1.73 x 10-4 M), the pH (2-9), and the temperature (15-35 degC). In all these experiments, the agitation speed was maintained constant at 260 rpm and the partial pressure of ozone at 1100 Pa. The influence of each experimental parameter on the degradation of NTS is described below.
+
+#### 3.1.1 Influence of the concentration of radical scavengers
+
+A series of experiments was conducted with a pH of 2 and in the presence of _tert_-butyl alcohol, a compound that reacts very slowly with ozone9 (_k_D=0.03 m-1 s-1), although its ability to react with hydroxyl radicals10 (_k_OH=5 x 108 M-1 s-1) is widely known. This property is exploited to inhibit the decomposition of the ozone in the aqueous phase and thus prevent the development of the free radical reaction mechanism in the ozonation process. The pH was set at 2 in order to eliminate to the greatest possible degree the presence of these hydroxyl radicals in solution.
+
+The effect of the concentration of _tert_-butyl alcohol
+
+Figure 1.: Experimental system. (1) Oxygen tank, (2) ozonistar, (3) reactor, (4) spectrophotometer, (5) flowmeter, (6) cellulose trap.
+
+on the ozonation of NTS can be observed in Fig 2, which shows that a reduction in the concentration of _tert_-butyl alcohol caused an increase in the ozonation rate of the compound, thus revealing the presence of hydroxyl radicals in the solution. It can also be observed in this figure that when the _tert_-butyl alcohol concentration was over 0.01 m the ozonation rate did not decrease significantly. It can therefore be deduced that 0.01 _Mmert_-butyl alcohol is the minimum concentration that produces the inhibition of all the free radicals present in solution. As a result of these experiments, a T-BuOH concentration of 0.01 m was selected for all remaining experiments.
+
+#### 3.1.2 Effect of the initial concentration of NTS
+
+As shown in Fig 3, an increase in the concentration of NTS present in solution produced no lengthening of the time needed to accomplish its complete oxidation. At the three concentrations assayed, a minimum of 2700 s was required to achieve almost total oxidation. However, an increase in the concentration of the acid induced a marked increase in its oxidation rate.
+
+Another important fact that can be deduced from these experiments is the presence of ozone in solution regardless of the acid concentration studied. The ozone concentration remained constant and equal to its initial value (1.04 x 10-4 m) in all the experiments, suggesting that the direct ozonation of NTS is very slow.
+
+This behaviour was expected, given the molecular structure of the compound under study, depicted in Scheme 1. The three sulfonic acid groups in the molecule (highly electron-withdrawing groups) induce the deactivation of the aromatic ring. Thus, the direct ozonation of NTS is disfavoured over the free radical mechanism (indirect reaction).
+
+Bayley _et al_11 proposed that aromatic compounds can react with ozone via three distinct mechanisms: (i) 1,3 dipolar cycloaddition or ozonolysis at the bond or bonds with greatest double bond character, (ii) electrophilic attack on the most polar bonds or atoms with least localization energy, and (iii) through a conjugated addition where there is a highly reactive diene. These authors reported that 75% of the oxidation of aromatic hydrocarbons occurs by 1,3 dipolar cycloaddition in the double bond with greatest electronic density (stoichiometry of reaction: 1 mol ozone per mol degraded hydrocarbon) and 25% by electrophilic attack on the most polar bond or lowest atom-localisation energy. Pryor _et al_12 observed that the ozonation of alkenes with electron-withdrawing substituents is mainly produced via 1, 3 dipolar cycloaddition. Moreover, alkenes in which the double bond is connected to electron-donating groups react many times faster than those in which it is connected to electron-withdrawing groups. Beltran _et al_13 determined the stoichiometry of the reaction of phenanthrene with ozone to be one mol of ozone per mol of degraded hydrocarbon. This hydrocarbon reacts exclusively with the ozone via 1,3 dipolar cycloaddition.
+
+According to these findings, the ozone attack on NTS may be produced by 1,3 dipolar addition to the carbon-carbon double bonds with greatest electron density. The theoretical study of the electron density measured at the bond critical point, using the AIM theory, 14 and performed with the Gaussian 98 series of programs,15 showed that the double bonds with the greatest electron densities are those in positions 1-2, 3-4, 5-6, and 7-8, with electron densities higher than 0.34 \(\text{ea}_{0}\text{}^{- 3}\); the highest one is that in position 5-6, with a value of 0.344 \(\text{ea}_{0}\text{}^{- 3}\). Thus, the ozone attack on NTS took place as proposed in Scheme 1.
+
+#### 3.1.3 Influence of temperature
+
+The effect of temperature was investigated in the presence of _tert_-butyl alcohol at pH 2, in order to observe its influence on only the direct ozonation reaction. Figure 4 shows the results.
+
+An increase in temperature induces an increase in the rate of reaction (as discussed below, \(K_{\text{D}}\)= 5.17 m-1 s-1, 6.72 m-1 s-1, and 14.11 m-1 s-1 for 15 degC, 25 degC, and 35 degC, respectively), despite that fact
+
+Figure 3: Kinetics of NTS ozonation at different NTS concentrations. T=25 °C, pH=2, [TBuOH]=0.01 m.
+
+Figure 2: Kinetics of NTS ozonation at different T-BuOH concentrations. T=25 °C, pH=2, [NTS]=5.75 × 10−5 m.
+
+that the increase in temperature reduces the concentration of dissolved ozone. This indicates a slow reaction, because there is no consumption of the initial dissolved ozone, whose concentration in solution remains constant within the range \(6.25\times 10^{-5}\)-\(1.25\times 10^{-4}\)M, depending on the temperature used.
+
+#### 3.1.4 Influence of pH
+
+Given the catalytic action of the hydroxyl ion in the decomposition of the ozone, and the likely high contribution of free radical reactions in the oxidation of NTS, the pH should be the experimental parameter with greatest influence on the ozonation process.
+
+On the other hand, the pH of the medium determines the degree of deprotonation of the NTS. Thus, we determined the acidity constants for the different ionisable hydrogens present in this compound, using potentiometric titrations. It was not possible to determine the first acidity constant with this method because despite reaching pH 1, there was no notable change in the pH value of the solution, indicating that \(K_{1} < 1\). However, the other two constants could be detected, obtaining the values \(K_{2} = 2.3\times 10^{-4}\) and \(K_{3} = 3.75\times 10^{-10}\).
+
+We studied the range pH 2-9 in order to avoid the decomposition of the hydroxyl radicals in O-, HO2-, etc, which may also contribute strongly to the process of radical oxidation.4,16
+
+Figure 5 shows that an increase in the pH produced an increase in the oxidation rate of the compound. As discussed below, the values of the global constant, \(K_{\text{global}}\) are: \(0.0034\text{\,m}^{-1}\text{s}^{-1}\), \(0.0329\text{\,m}^{-1}\text{s}^{-1}\), and \(0.9815\text{\,m}^{-1}\text{s}^{-1}\) for pH values 2, 7, and 9, respectively. This is mainly because the increase in pH causes an increase in the concentration of hydroxyl radicals present in solution. The other reason is that, according to the p_K_a values of NTS (p_K_2  4, p_K_3  10), the species present at pH 7 and 9 is more reactive than that present at pH 2, given that it is doubly deprotonised and the electron-attractant character of the deprotonised sulfonic groups present in NTS is lesser. This produces a greater electronic density in the ring, and therefore a greater affinity for ozone. It should be noted that this increase is not linear, because the decomposition rate of ozone in aqueous medium increases exponentially with respect to the pHs, as reported by Gottschalk.17
+
+### Determination of the stoichiometry of the ozonation reaction
+
+Experiments were conducted to determine the stoichiometry of the reaction, using the reactor in discontinuous mode. In each experiment, the ozone
+
+Figure 4: Kinetics of NTS ozonation as a function of temperature. pH = 2, (TRuOH) = 0.014, (NTS) = 5.75 \(\times\) 10\({}^{-5}\)M.
+
+Figure 5: Kinetics of NTS ozonation as a function of pH. [NTS] = 5.75 × 10\({}^{-5}\)M, T = 25\({}^{\circ}\)C.
+
+concentration was maintained at a constant value (1.04 x 10-4M) while the concentration of NTS was varied (6.91 x 10-6M -8.32 x 10-4M).
+
+\[\begin{equation*}{\mathrm{M}}+z{\mathrm{O}}_{3}{\longrightarrow}{\mathrm{B}}\end{equation*}\]
+
+After 120 s, the concentrations of NTS and ozone present in solution were analysed, using eqn (1) to obtain the value of the stoichiometric coefficient (_z_).
+
+\[Z=\frac{(C_{{\mathrm{O}}_{3}})_{0}-(C_{{\mathrm{O}}_{3}})_{\mathrm{f}}}{(C_{{ \mathrm{M}}})_{0}-(C_{{\mathrm{M}}})_{\mathrm{f}}}\]
+
+Where (_C_O)0 is the initial concentration of ozone, (_C_O)f is the final concentration of ozone, (_C_M)0 is the initial concentration of NTS and (_C_M)f is the final concentration of NTS.
+
+Figure 6 shows the stoichiometric coefficient values as a function of the initial ratio (_R_) of contaminant (NTS) and ozone concentrations. It can be observed that \(z\) decreases steeply with the increase in \(R\) until a plateau is reached at a value considered to be the real value of the stoichiometric coefficient. It seems clear that at low _R_-values, the ozone is mainly consumed by ozonation reactions of the secondary products, whereas at high _R_-values, direct ozonation reaction is the only way for the ozone to disappear. It can be deduced from Fig 6 that \(z\) is one mol of ozone consumed per mol of NTS consumed.
+
+### 3.3 Study of the kinetics of NTS ozonation
+
+During ozonation, part of the dissolved ozone reacts directly with the organic compound and the other part decomposes, generating hydroxyl radicals in solution that also react with the organic compound. Thus, the kinetic equation of the ozonation of an organic compound is defined by
+
+\[\begin{equation*}{\mathrm{d}}C_{{\mathrm{M}}}/{\mathrm{d}}t=r_{{\mathrm{D}}}+r_{{\mathrm{OH}}}=(k_{{\mathrm{D}}}C_{{\mathrm{O}}_{3}}+k_{{\mathrm{OH}}}C_{{\mathrm{OH}}})C_{{\mathrm{M}}}\\ =K_{{\mathrm{global}}}C_{{\mathrm{M}}}\end{equation*}\]
+
+where the right hand side of eqn (2) represents the contribution of direct ozonation (_r_D) and radical oxidation (_r_OH) to the degradation rate of M. In the present study, the direct reaction between ozone and NTS is assumed to be second-order (first-order with respect to the ozone and first-order with respect to the NTS18). In the case of _r_OH, the second-order kinetics are evident, given that this term represents the reaction rate of an elementary step between the hydroxyl radical and the compound in the mechanism of free radical reaction.19 In eqn (2), _k_D and _k_OH are the rate constants of the direct and free radical reactions, respectively.
+
+#### 3.3.1 Determination of the direct reaction constant, _k_D
+
+We first determined the rate constant for the direct reaction of ozonation, _k_D. This constant was deduced from experiments performed under experimental conditions where the free radical process was inhibited. Thus, we proceeded to perform ozonation reactions at pH 2 in the presence of 0.01 m _ter_-butyl alcohol, the minimum concentration at which the radical process is inhibited, as mentioned above.
+
+In these experiments, the reactor was operated in semi-continuous mode, because the supply of gaseous ozone to the reactor was sustained in order to keep the concentration of dissolved ozone constant at 1.04 x 10-4M. The initial concentration of NTS was 5.75 x 10-5M, and the kinetics was determined by following the concentration of NTS as a function of the reaction time. The consumption of NTS can therefore be calculated as:
+
+\[{\mathrm{d}}C_{{\mathrm{M}}}/{\mathrm{d}}t=r_{{\mathrm{D}}}=k_{{\mathrm{D}}}C_{{\mathrm{M}}}C_{{\mathrm{O}}_{3}}\]
+
+Integrating eqn (3) gives:
+
+\[\ln[C_{{\mathrm{M}}}/C_{{\mathrm{M}}_{{\mathrm{o}}}}]=k_{{\mathrm{D}}}C_{{ \mathrm{O}}_{3}}t=k_{{\mathrm{obs}}}t\]
+
+Thus, in Fig 7, it can be seen that when ln(_C_M/_C_M) is plotted against time, the data fit a straight line, with a regression coefficient of 0.9983. It can be deduced from this figure that the value of _k_obs is 0.0007 s-1, and
+
+Figure 6.: Stoichiometric coefficient (_z_) as a function of the initial ratio of NTS and ozone concentrations (_f_) in the direct reaction between ozone and NTS. pH=2, (T-bOH)=0.01u, T=25°C. Initial concentrations: [O]=8.33 × 10 °S. [NTS]=6.91 × 10 °S. -6.67 × 10 °L.
+
+Figure 7.: Determination of direct reaction constant between ozone and NTS. pH=2, (T-bOH)=0.01u, T=25°C, [NTS]=5.75 × 10 °S. -4M.
+
+therefore the value of \(k_{\rm D}\) is 6.72 m\({}^{-1}\)s\({}^{-1}\). This value is similar to that determined by other authors for organic compounds with low electron density in the aromatic ring.[5, 9, 20]
+
+This rate constant of the direct reaction is related to the temperature by the Arrehenius equation. Thus:
+
+\[k_{\rm D}=2.8\times 10^{7}e^{(-37176/RT)} \tag{5}\]
+
+The activation energy for the direct ozonation reaction of NTS determined in this study (37176J mol\({}^{-1}\)) is very similar to that reported by other authors for compounds with a naphthalene structure[13] and for benzene derivatives with low density of charge in the ring.[21]
+
+The value of \(k_{\rm D}\) obtained is very low, implying that the oxidation of this compound was mainly by a free radical reaction. This is demonstrated by the contribution of the direct reaction to the total oxidation (\(\delta\)) at different pH values, which was calculated with eqn (6). The results are listed in Table 1.
+
+\[\delta=\frac{k_{\rm D}C_{\rm o}C_{\rm M}}{({\rm d}C_{\rm M}/{\rm d}t)}\times 100 \tag{6}\]
+
+An interesting result which can be deduced from the data in Table 1 is that at pH 2 the contribution of direct ozonation to the total oxidation of the NTS is only 20% and that this contribution decreases sharply with increases in the pH, because of the greater concentration of hydroxyl radicals in solution. The oxidation of NTS was therefore by the radical mechanism, regardless of the pH used. Thus, ozonation may be considered an advanced oxidation process for this family of compounds.
+
+#### 3.3.2 Determination of the indirect reaction constant, \(k_{\rm OH}\)
+
+Hoigne and Bader[4] proposed the use of competitive kinetics to determine very high kinetic constants (\(>\)10\({}^{4}\) m\({}^{-1}\)s\({}^{-1}\)). To employ this method, it is necessary to know the kinetic constant of one of the compounds, designated the reference compound (M1). Assuming that the reactions are taking place according to kinetics of pseudo-first-order, the constant of the compound (M2) can be calculated using eqn (7).
+
+\[k_{\rm rel}=k_{\rm M_{2}}\Big{/}k_{\rm M_{2}}=\frac{\ln[\rm M_{2}]_{\rm r}/[ \rm M_{2}]_{0}}{\ln[\rm M_{1}]_{\rm r}/[\rm M_{1}]_{0}} \tag{7}\]
+
+With this method, the kinetic constant relative to the free radical mechanism for NTS was calculated from experiments conducted in alkaline media (pH 9), in order to favour the free radical mechanism, using sodium 4-chlorobenzoate as reference compound.[10] The results are shown in Fig 8. Sodium 4-chlorobenzoate reacts very slowly with ozone and the constant of its reaction with hydroxyl radicals is known (\(k=4.4\times 10^{9}\) M\({}^{-1}\) s\({}^{-1}\)). It is also readily detected with HPLC-UV.
+
+We conducted experiments varying the [M2]0/[M1]0 ratio within the range 1:3 -3:1 (Table 2). The value of \(k_{\rm rel}=k_{\rm M_{2}}/k_{\rm M_{2}}\) was independent of the ratio of concentrations used for its determination, confirming that these are competitive reactions.
+
+Considering \(k_{\rm rel}\) to be the mean of the values displayed in Table 2, it can be deduced that the constant of indirect reaction of NTS is \(3.7\times 10^{9}\) M\({}^{-1}\)s\({}^{-1}\). This value is very high, and similar to that reported in the literature for most aromatic compounds.[22]
+
+## 4 Conclusions
+
+Our study on the oxidation of 1,3,6 naphthaleneti-sulfonic acid with ozone showed that an increase in temperature and pH enhances the degradation of this acid. The stoichiometry of the direct ozonation reaction was one mol of ozone per mol of NTS. The direct reaction constant, \(k_{\rm D}\), was 6.72 m\({}^{-1}\) s\({}^{-1}\) and the radical reaction constant, \(k_{\rm OH}\), was \(3.7\times 10^{9}\) M\({}^{-1}\) s\({}^{-1}\). The contribution of the direct reaction of ozonation was very low, even at acid pHs suggesting that the
+
+\begin{table}
+\begin{tabular}{l c c c} \hline \(pH\) & \(K_{\rm obs}\) (s\({}^{-1}\)) & \(K_{\rm globular}\) (s\({}^{-1}\)) & \(\delta\) (\%) \\ \hline
+2 & 0.0007 & 0.0034 & 20.58 \\
+7 & 0.0007 & 0.0329 & 2.13 \\
+9 & 0.0007 & 0.9815 & 0.07 \\ \hline \end{tabular}
+\end{table}
+Table 1: Contribution of the direct ozonation reaction (δ) at different pH values
+
+Figure 8: Determination of indirect reaction constant. pH=9, T=25 °C, [NTS]=1.03 × 10\({}^{-1}\)u, [PCBA]=8.43 × 10\({}^{-5}\)u.
+
+ozonation process can be considered as an advanced oxidation process for this family of contaminants.
+
+The radical reaction proved more efficacious in oxidising NTS. The removal of this acid from water will therefore be favoured under the experimental conditions that enhance the generation of free radicals, ie high pH values and absence of T-BuOH.
+
+## References
+
+- Reemtsma T, Jochimsmen J and Jekel M, Persistence of sulphonated polyphenols in the biological treatment of industrial wastewater. _Vom Wasser_**81**:353-363 (1993).
+- Lange FT, Wenz M and Brauch HJ, Trace-level determination of aromatic sulphonates in water by on-line ion-pair extraction/ion-pair chromatography and their behaviour in the aquatic environment. _J High Resol Chromatogr_**18**:243-251 (1995).
+- Fitchner S, Lange Th, Schmidt W and Brauch HJ, Determination of aromatic sulfonates in the river Elbe by on-line ion-pair extraction and ion-pair chromatography. _Penseius J Anal Chem_**353**:57-63 (1995).
+- Hoigne J and Bader H, The role of hydroxyl radical reactions in ozonation processes in aqueous solutions. _Water Res_**10**:377-386 (1976).
+- Beltran FJ, Encinar JM and Alonso MA, Nitroaromatic hydrocarbon ozonation in water: 1. Single ozonation. _Ind Eng Chem Res_**37**:25-31 (1998).
+- Stahelin J and Hoigne J, Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. _Environ Sci Technol_**19**:1206-1213 (1985).
+- Masschelein WJ, Measurement of high ozone concentrations in gases by KI titration and monitoring by UV-absorption. _Ozone Science & Engineering_**24**:489-493 (1998).
+- Bader H and Hoigne J, Determination of ozone in water by the Indigo method. _Water Res_**15**:449-456 (1981).
+- Hoigne J and Bader H, Rate constants of the reactions of ozone with organic and inorganic compounds. I. Non-dissociating organic compounds. _Water Res_**17**:173-183 (1983).
+- Hoigne J, Inter-calibration of OH radical sources and water quality parameters. _Wur. Sci Tech_**3**:1-8 (1997).
+- Bayley PS, Battenbee JE and Lane AG, Ozonation of benz[a]anthracene. _J. Am Chem Soc_**90**:1027-1033 (1968).
+- Pryor WA, Giamalva D and Church DF, Kinetics of ozonation. 1. Electron-deficient alkenes. _J. Am Chem Soc_**105**:6858-6861 (1983).
+- Beltran FJ, Orejero G, Encinar JM and Rivas J, Oxidation of polynuclear aromatic hydrocarbons in water. 1. Ozonation. _Ind Eng Chem Res_**34**:1596-1606 (1995).
+- Bader RFW, _Atoms in Molecules_, Oxford University Press, Oxford (1990).
+- Gaussian 98, Gaussian, Inc, Pittsburgh, PA (1998).
+- Von Sonnito C, Dowleit P, Fang X, Netrens R, Pan X, Nien Schuchmann M and Schuchmann HP, The fate of peroxyl radicals in aqueous solution. _Wur. Sci Tech_**35**:9-15 (1997).
+- Gottschalk C, Oxidation organischer Mikroverunreiniguen in nartischen und synthetischen Wasser mit zoon und Ozon/Wasserstoffperoxid, _PhD Thesis_, Dept Environm, Berlin (1997).
+- Hoigne J, Chemistry of aqueous ozone and transformation of pollutants by ozonation and advanced oxidation processes, in _The Handbook of Environmental Chemistry_, Ed by Hrubec J, Springer-Verlag, Berlin 85-141 (1998).
+- Beltran FJ, Orejero G and Rivas J, Oxidation of polynuclear aromatic hydrocarbons in water. 4. Ozone combined with hydrogen peroxide. _Ind Eng Chem Res_**35**:891-898 (1996).
+- Pietsch J, Schmidt W, Brauch HJ and Worch E, Kinetic and mechanistic studies of the ozonation of alicyclic amines. _Ozone Science & Engineering_**21**:23-37 (1999).
+- Rinker EB, Ashour SS, Johnson MC, Kott GJ, Rinker RG and Sandall OC, Kinetics of the aqueous-phase reaction between ozone and 2,4,6 trichlorophenol. _AIChE Journal_**45**:1802-1807 (1999).
+- Hage WR and David Yao CC, Rate constants for reaction of hydroxyl radicals with several drinking water contaminants. _Environ Sci Technol_**26**:1005-1013 (1992).

manual_annotation/200pdfs_in_mmd/10.1002_jctb.644.mmd

@@ -0,0 +1,186 @@
+Ozone treatment of textile effluents and dyes: effect of applied ozone dose, pH and dye concentration
+
+Mehmet F Sevimli, Hasan Z Sarikaya
+
+# Abstract
+
+The ozonation of wastewater supplied from a treatment plant (Samples A and B) and dye-bath effluent (Sample C) from a dyeing and finishing mill and acid dye solutions in a semi-batch reactor has been examined to explore the impact of ozone dose, pH, and initial dye concentration. Results revealed that the apparent rate constants were raised with increases in applied ozone dose and pH, and decreases in initial dye concentration. While the color removal efficiencies of both wastewater Samples A and C for 15 min ozonation at high ozone dosage were 95 and 97%, respectively, these were 81 and 87%, respectively at low ozone dosage. The chemical oxygen demand (COD) and dissolved organic carbon (OOC) removal efficiencies at several ozone dose applications for a 15 min ozonation time were in the ranges of 15-46% and 10-20%, respectively for Sample A and 15-33% and 9-19% respectively for Sample C. Ozone consumption per unit color, COD and DOC removal at any time was found to be almost the same while the applied ozone dose was different. Ozonation could improve the BOD_s_ (biological oxygen demand) COD ratio of Sample A by 1.6 times with 300 mg dm-3 ozone consumption. Ozonation of acid dyes was a pseudo-first order reaction with respect to dye. Increases in dye concentration increased specific ozone consumption. Specific ozone consumption for Acid Red 183 (AR-183) dye solution with a concentration of 50 mg dm-3 rose from 0.32 to 0.72 mg-O3 per mg dye decomposed as the dye concentration was increased to 500 mg dm-3.
+
+## Introduction
+
+Textile mill effluents are known to have extremes of pH (either alkaline or acidic, depending on the processes used) and temperature, high biological oxygen demand (BOD), high chemical oxygen demand (COD) and high concentrations of suspended solids (SS). Textile mill effluents are also characterized by high levels of color caused by residual dyes that were not fixed to fibers in the dyeing process. The dye molecules are highly structured polymers and toxic to the organism. Therefore, it is very difficult to break them down biologically.1
+
+The ozonation process has been suggested in the recent literature as a potential alternative for decolorization2-17 improvement of the biological degradation of textile effluents.9,11,17 The oxidation potential of ozone is 2.07 volts and it can oxidize organics and inorganics easily, but even high doses of ozone do not completely convert the organic matter to carbon dioxide and water, particularly for dye wastes containing surfactants and suspended matter.5,15 Much research has been undertaken with the aim of optimizing the experimental conditions of ozonation, such as pH, temperature and applied ozone dose.2,6,12,15
+
+It has been confirmed that the ozonation rate of dye-containing wastewater is controlled by the mass transfer of ozone from the gas phase to the wastewater.7,12,13 However, the driving force for ozone mass transfer is system dependent. The dependence of the driving force on the wastewater system results from the fact that the concentration of the dissolved ozone varies considerably with the rate of the self-decomposition of ozone, which depends on the wastewater's characteristics.2 Ozone always reacts with organic or inorganic matter by two different mechanisms: by direct reaction with the molecular ozone and by indirect reaction with the radical species that are formed when ozone decomposes in water.18 Usually, different reaction pathways contribute to the ozonation process simultaneously.18 At low pH, ozone only reacts with compounds with specific functional groups through selective reactions such as electrophilic, nucleophilic or dipolar addition reaction (direct path way).[19] The decomposition of ozone is faster at high pH and therefore the major oxidant in an alkaline solution is the more reactive hydroxyl radical (indirect pathway).[8] Some investigators have obtained high removal efficiencies at high pH.[3, 4, 9, 11] During the ozonation process, dyes lose their color by the oxidative cleavage of the chromophores. The cleavage of C--C double bonds and other functional groups will shift the absorption spectra of the molecule out of the visible region.[20] Researchers have observed that some dye-containing wastewater undergoes a pseudo-first order reaction with respect to dye concentration and the rate of decolorization decreases with increase in dye concentration.[12, 13] Results have showed that ozonation can improve the biodegradability of the wastewater by converting the high molecular compounds of the dye into lower molecular organics.[9, 11, 17]
+
+In this study, the impact of applied ozone dose, initial dye concentration and initial pH on the color and removal of organic matter from textile industry effluents by ozonation were investigated. Besides the evaluation of kinetic behavior, the effect of these operational parameters on ozone consumption was interpreted.
+
+## Materials and Methods
+
+### Materials
+
+Potassium iodide (KI) and sodium thiosulfate (Na2S2O3) used for the determination of ozone input and output were supplied in analytical grade by Merck. Two dyes used in the experiments, Acid Red 183 (AR-183, Color Index #: 18800) and Acid Blue 158 (AB-153, Color Index #: 14880), were of commercially available reagent grade (CIBA) and used without further purification. The chemical structures of the dyes are given in Fig 1. NaOH and H2SO4 solutions for the adjustment of the pH of wastewaters and dye solutions were of analytical grade supplied from Merck. All the solutions were prepared with DDW (distilled deionized water) (Milli-Q water, Millipore).
+
+Dye solutions of AR-183 and AB-153 with the concentrations of 50-500 mg dm-3 were used in the ozonation experiments. Dye-bath effluent prepared with 1:1 metal complex dyes and plant effluent samples taken from the equalization tank of an existing wastewater treatment plant for the present study were supplied from a wool dyeing and finishing mill. The 1:1 metal complex dye bath effluent included Neolan Yellow GR, Neolan Rose BE, Neolan Blue 2G, Neolan Blue 3R and other dye-bath additives. The characterization of the wastewaters is given in Table 1.
+
+### Methods
+
+Ozone was generated from air by a laboratory ozone generator PCI GL1. The Gas flow rate to the reactor was measured as 1.42 dm3 min-1 with a flow meter incorporated into the ozone reactor and was not changed throughout the study. Ozonation experiments were conducted at 104 kPa pressure. A schematic diagram of the bench-scale reactor system used for the experiment throughout the study is presented in Fig 2. A 1000 cm3 sample was placed in a semi-batch bubbled gas washing bottle reactor (1100 mm high and 40 mm internal diameter) having an approximate capacity of 1.4 dm3. An ozone-air mixture was supplied at the bottom of the reactor through a sintered glass plate diffuser. Two 250 cm3 gas washing bottles containing 200 cm3 2-5% KI solution were connected in series with the reactor to collect all unreacted ozone gas passing through the sample that was being treated. Ozone concentration at the inlet and outlet gas streams was determined for each ozonation sequence by titrating the ozonated KI
+
+\begin{table}
+\begin{tabular}{c c c c c c c c c}  & _COD_ & _BOD_S & _DOC_ & _Alkalinity_ & _SO_2_- & _TSS_ & _Color_ \\  & _pH_ & _(mg dm−3)_ & _(mg dm−3)_ & _(mg dm−3)_ & _(mgCaCO3_dm−3)_ & _(mg dm−3)_ & _(mg dm−3)_ & _(mg dm−3)_ & _Pl-Co_ \\ \end{tabular}
+\end{table}
+Table 1: Characterization of the samples
+
+Figure 1: Structure of dyes.
+
+Figure 2: Schematic diagram of the bench-scale reactor system.
+
+solution with 0.1 mol dm-3 sodium thiosulfate under acidic conditions according to _Standard Methods for the Examination of Water and Wastewater_.21 The applied ozone dose was maintained between 14.2 and 140 mg dm-3 min-1 to determine the impact of applied ozone dose in ozone oxidation. The pH of the samples was adjusted between 2.4 and 11.0 with 1.0-6.0 mol dm-3 of NaOH or 0.5-3.0 mol dm-3 H2SO4 solutions. All pieces contacted with ozone were stainless steel, glass or Teflon. All experiments were conducted at room temperature (20 +- 2 degC).
+
+Before color and DOC measurements, all the ozonated samples were filtered by using membrane filter paper with a pore size of 0.45 mm. True color measurements of each dye-bath effluent were carried out by a HACH DR 2000 type spectrophotometer which was calibrated according to the platinum-cobalt method.21 The concentrations of the dye solutions after each ozonation sequence were determined by a Pharmacia LKB Novaspec II type spectrophotometer at their maximum absorption wavelengths of 500 and 635 nm for AR-183 and AB-158, respectively. pH was measured with an ion analyzer (Orion SA 720). COD and BOD5 were measured according to _Standard Methods for the Examination of Water and Wastewater_.21 Dissolved organic carbon (DOC) measurements were carried out with an Ionics Carbon Analyzer 1505 type TOC analyzer.
+
+For the calculation of ozone doses, eqns (1) and (2), taken from the literature, were used.22 Equation (3) was only used in order to calculate the ozone consumption of 1 mg dye removed by ozonation.
+
+\[D_{\text{app}} = Q_{\text{gas}}\int_{0}^{t}\frac{[ O_{3_{\text{g}}} ]_{0} }{V_{\text{L}}}\text{d}t\]
+
+\[D_{\text{c}} = Q_{\text{gas}}\int_{0}^{t}\frac{[ O_{3_{\text{g}}} ]_{0} - [ O_{3_{\text{g}}} ]}{V_{\text{L}}}\text{d}t\]
+
+\[D_{\text{S}} = \frac{D_{\text{c}}}{C_{0} - C}\]
+
+where _D_app (mg dm-3), _D_c (mg dm-3) and _D_S (mg mg-1) represent applied ozone dose, consumed ozone dose and specific ozone consumption, respectively. [O3_g_]0 and [O3_g_] show the ozone concentration in the gas stream at the inlet and outlet, respectively, as mg dm-3. _Q_gas and _V_L represent gas flow rate (dm3 min-1) and liquid volume (dm3).
+
+### Kinetic Assessment
+
+The overall degradation of the dissolved organic matter contained in the wastewater by ozone is a complex process with many reactions, which cannot be individually distinguished. Therefore an approximate kinetic study of the ozonation of the textile wastewater could be performed by using any of the global parameters directly related to the organic load present in the effluent and measured by COD, DOC and color. By assuming that the reaction between ozone and organic matter follow first-order kinetics with respect to the COD, DOC and color, yielding an overall second-order kinetics, eqn (4) may be written as:
+
+\[- \frac{\text{d}C}{\text{d}t} = k[ O_{3} ]C\]
+
+where \(k\) is the second-order rate constant, [O3_] is the ozone concentration in aqueous phase and \(C\) is the pollutant concentration. The decolorization of dyes by ozone is also second-order with respect to dye.19 However, an ozonation reaction can be assumed to be a pseudo-first order reaction, implying the ozone concentration is constant and when the amount of one of the reactants is in excess.2,6,11-16,18,25,24
+
+Owing to the fact that ozone is only slightly soluble in water, the gas phase resistance is neglected. Indeed, the partition coefficient of ozone between the gas and aqueous phases is about 400 (concentration of ozone in the gas phase to concentration of ozone in the aqueous phase).25 In this study, the ozone concentrations in the gas phase were in the range 10-100 mg dm-3 Ozone concentrations in the aqueous phase changed within 0.15-0.50 mg dm-3 depending on the experimental conditions. When the ozone concentration in the aqueous phase is compared with COD and DOC concentrations (Table 1), eqn (4) may be rewritten as a pseudo-first order reaction with respect to pollutant:
+
+\[- \frac{\text{d}C}{\text{d}t} = k^{\prime}C\]
+
+where _k'_ is the overall rate constant as min-1 and is represented as follows:
+
+\[k^{\prime} = k[ O_{3} ]\]
+
+Integrating eqn (5) between \(t\) = 0 and \(t\) = \(t\), one may write:
+
+\[\ln\frac{C}{C_{0}} = - k^{\prime}t\]
+
+According to the this expression, a plot of the first term vs \(t\) should lead to a straight line for each experiment whose slope gives the overall rate constant. In this study, all the experimental data such as COD, DOC, color and dye concentration satisfied pseudo-first order kinetics.
+
+## EXPERIMENTAL RESULTS
+
+### Effect of applied ozone dose on the rate of ozonation
+
+The effect of applied ozone dose on color, COD, and DOC removal was investigated by increasing the applied ozone dose. The applied ozone dose was adjusted by changing the ozone percentage in the gas stream while maintaining the gas flow rate at 1.42 dm3 min-1. Kinetic evaluation indicated that ozonation of textile wastewater was a pseudo-first order reaction with respect to color, COD and DOC for both wastewater types, ie taken from the equalization tank (Sample A) and the dye-bath effluent (Sample C). As the applied ozone dose increased, overall rate constants of color, COD and DOC increased. This relationship has been stated by some researchers in their studies.2,15 In this study, a logarithmic relationship was found between the overall rate constants and the applied ozone doses (Fig 3). The increase in the rate constant as the ozone dose increased might be similarly explained by the two-film theory.26 According to this theory, the overall mass transfer mechanism consists of several steps: (i) diffusion of ozone through the gas phase to the interface between the gas and aqueous phases, (ii) transport across the interface to the aqueous phase boundary, and (iii) transfer into the bulk aqueous phase. Increasing the applied ozone dosage should enhance mass transfer15 and cause an increased ozone concentration in the liquid phase and an increased pseudo-first order rate constant.2 However, as the ozone concentration in the liquid phase approaches its maximum value, the process would become increasingly controlled by the rate of chemical reaction and any further improvement in ozone mass transfer would have a diminished effect on the observed reaction rate.15 This effect of shifting the process from principally mass transfer control to principally chemical reaction control should be particularly evident from the lower reactant (COD, DOC, color) concentration, as observed in Fig 4.
+
+On the other hand, when ozone dose and ozonation time were increased, the ozone utilization ratio decreased. Therefore, it was necessary to make another assessment from the viewpoint of the ozone utilization ratio for the removal efficiencies of parameters monitored during the ozonation. For instance, the ozone utilization ratios were 93 and 20% for Sample A for an ozonation time of 15 min and for applied ozone doses of 14.2 and 140 mg dm\({}^{- 3}\) min\({}^{- 1}\), respectively. Similarly, for Sample C, the ozone utilization ratios were 64 and 39% for the applied ozone doses of 24.1 and 87.2 mg dm\({}^{- 3}\) min\({}^{- 1}\), respectively.
+
+At high ozone dosage, the color of both Samples A and C decreased markedly and the efficiency of color removal increased up to 75% in a few minutes. When low ozone doses were applied, the ozonation time required to reach the same efficiency was 4-5 min. As can be seen in Fig 4, in spite of having high color removal efficiencies, limited COD and DOC removal efficiencies were obtained. Color is caused by chromosomes in the molecules (usually double bound systems) which are oxidized by ozonation to single bonds thus losing the ability to absorb visible light.20 The lower DOC and COD removal compared with the color removal can be explained by incomplete oxidation of organic materials. COD and DOC removal efficiencies for Sample A at several ozone dose applications for a 15 min ozonation time were in the ranges 15-46% and 10-20% respectively. These values were in the ranges 15-33% and 9-19% respectively for Sample C.
+
+The BODg/COD ratio, as presented in Fig 4, can be used as an indication of the biological treatability of the organic materials. Ozonation increased the BODg/COD ratio of Samples A and C continuously with increasing ozonation time and applied ozone dose except for ozone doses of 140 mg dm\({}^{- 3}\) min\({}^{- 1}\) applied for Sample A and 82.7 mg dm\({}^{- 3}\) min\({}^{- 1}\) applied for Sample C. It was an expected trend, since ozonation of organic matter causes the transformation of the compounds of larger relative molecular mass into smaller and more biodegradable ones.5,11,17 When the applied ozone dose was high as stated above, the BODg/COD ratio increased with ozonation time until it reached its maximum value and further oxidation oxidized the biodegradable by product of ozonation and as a result of this the BODg/COD ratio decreased.27,28 The ratio of BODg/COD of the non-ozonated wastewater Sample A was 0.20 and the ratio could be increased to 0.40 at the applied ozone dosage of 81 mg dm\({}^{- 3}\) min\({}^{- 1}\) for 15 min ozonation time. Ozonation could also improve the BODg/COD ratio of wastewater Sample C from 0.56 to 0.76 at nearly the same applied ozone dose for 3 min ozonation time.
+
+Figure 5 summarizes the oxidation efficiencies for the parameters as a function of the amount of consumed ozone (\(D_{\text{o}}\)). The consumed ozone per unit COD, DOC and color removal at any time was found to be almost the same even if the applied ozone doses were different. This result suggested that a desired removal efficiency of color, or other wastewater characteristics, could be achieved simply by controlling the consumed ozone.29
+
+Figure 3: Effect of applied ozone dose on apparent rate constant.
+
+### Effect of initial pH on the rate of ozonation
+
+The ozonation of wastewaters at various initial pH values (2.4-11.0) was also examined in this study. Figure 6 indicates pseudo-first order rate constants of color, COD and DOC as a function of initial pH. As can be seen in the figure, with the increase in pH, rate constants increased logarithmically for Sample B. However, the same trend could not be observed for Sample C. The rate constants were not affected by pH up to 9 however, a sharp increase in rate constants of color, COD and DOC disappearance were observed at pH 11. In general, ozone oxidation pathways include direct oxidation by ozone or radical oxidation by OH+ radicals.18,30,31 Direct oxidation is more selective and predominates under acidic conditions. Since the oxidation potential of hydroxyl radicals is much higher than that of the molecular ozone, direct oxidation is slower than radical oxidation. In this study, the amount of ozone consumed in ozonation at alkaline pH levels was greater than that consumed low pH levels. This observation may be explained by the enhancement of ozone decomposition by hydroxyl radicals at alkaline pH values.3,8,11,32
+
+Although higher removal rates of color, COD and DOC were obtained at high pH values, the differences in removal efficiencies between low and high pH conditions were slight for Sample B (Fig 7). These were 87-91% for color, 18-24 for COD and 10-14% for DOC, respectively. For Sample C, mean removal efficiencies of the color, COD and DOC were in the ranges 86-90%, 15-17% and 8-11% respectively for the pH range of 2.4-11.0. These results imply that high initial pH could enhance the removal efficiencies of color, COD and DOC.3,4,9,11,32
+
+The Ozone utilization ratios between low and high pH levels at a constant applied ozone dose were 9 and 8% for samples B and C respectively. These results show that the effect of pH on ozone utilization ratio is also slight.
+
+### Effect of initial dye concentration on the rate of ozonation
+
+The effect of initial dye concentration on the apparent rate constant with 22.6 mg dm-3 min-1 applied ozone dose at pH 7.0 as investigated. The decomposition of either AR-183 or AR-158 dye solution by ozonation
+
+Figure 4: Changes in color, COD, DOC and BOD/COD ratio with varying applied ozone dose.
+
+was in accordance with pseudo-first order kinetics. Figures 8 and 9 illustrate the results.
+
+The apparent pseudo-first order rate constant decreases with an increase in initial dye concentration. This observation has been made before.2,6,11,12,14-16 The logarithmic relationship between the apparent rate constant and the initial dye concentration derived in this study is in good agreement with results observed in previous studies (Fig 9).2,11,15 The trends of apparent reaction rate constants (_k_) for both AR-183 and AR-158 with increasing initial dye concentration were found to be similar. Reaction rate constants were obtained by solving the equations of straight lines for both dyes in Fig 9 and were obtained as follows:
+
+\[k^{\prime}_{\text{AR-183}} = 52.06\ C_{\text{dye}}^{-1.082}\]
+
+\[k^{\prime}_{\text{AB-158}} = 81.53\ C_{\text{dye}}^{-1.153}\]
+
+where the units of _k'_ and _C_dye are min-1 and mg dm-3 respectively. These equations are valid for the given initial dye concentrations in this study and these make the prediction of the rate constant from the initial dye concentration possible. Ozone is slightly soluble in water,12,15,25,26,29 the resistance of the gas phase and interface are insignificant, and the process of ozone decolorization may be assumed to be controlled by molecular diffusion of dissolved ozone and the rate of the chemical reaction. When the dye concentration is increased, the ozone concentration would no longer be increased to meet the necessity of excess dye. Therefore _k'_ values would be decreased.2,6,15 The other possible explanation is that more intermediates, which consume more ozone, are generated when the initial dye concentration is high.8,15
+
+In this study, a logarithmic relationship was also observed between initial dye concentration and specific ozone consumption. Figure 10 clearly indicates that as the initial dye concentration increased, specific ozone consumption increased according to eqns (10) and (11):
+
+\[D_{\text{S(AR-158)}} = 0.4128\ C_{\text{dye}}^{0.111}\]
+
+\[D_{\text{S(AB-183)}} = 0.0664\ C_{\text{dye}}^{0.381}\]
+
+where, _D_S indicates the specific ozone consumption as mgmg-1 dye decomposed and can be calculated from eqn (3). _C_dye is the initial dye concentration in mg dm-3. _D_S values seen in Fig 10 were obtained by dividing the final consumed ozone dose for an ozonation period by the decomposed dye. While AR
+
+Figure 5: Changes of color, COD, DOC and BOD with consumed ozone dose.
+
+Figure 6: Effect of initial pH on the apparent rate constant.
+
+183 and AB-158 dye solutions with 50 mg dm-3 of concentration consumed 0.32 and 0.63 mg ozone per decomposed mg dye respectively, the dye concentration was increased to 500 mg dm-3, specific ozone consumptions of those dye solutions rose to 0.72 and 0.82 mg ozone per decomposed mg dye, respectively. These values indicate that increasing the dye concentration decreased not only the apparent rate constant but also increased ozone consumption.
+
+## CONCLUSION
+
+In this study, wastewaters taken from the equalization tank of a treatment plant, dye-bath effluents from a dyeing and finishing mill, and two kinds of acid dye solutions were treated by ozonation in order to determine the effect of applied ozone dose, initial pH and initial dye concentration on the removal rate of color, COD, DOC and BOD. Kinetic evaluations were performed for all circumstances. According to the experimental results, the following conclusions can be drawn.
+
+When applied ozone dose was increased, color, COD and DOC removal efficiency and the apparent rate constants increased, but ozone utilization ratio decreased. The consumed ozone per unit color, COD and DOC removal at any time were found to be almost the same even if the ozone doses applied were different. High initial pH could enhance the removal efficiencies of color, COD and DOC slightly. High color removal but low COD and DOC efficiencies could be obtained by ozonation. Ozonation can increase the biodegradability of wastewater by up to 1.6 times.
+
+Decomposition of the acid dyes by ozone is a pseudo-first order reaction with respect to dye. The rate constant declines logarithmically with increased initial dye concentration. Increasing the initial dye concentration would cause more ozone to be consumed.
+
+\[\mathcal{J}\,\text{Chem}\,\text{Technol}\,\text{Biotechnol}\,\text{77} \text{842} \text{850}\,\text{(online: 2002)}\]
+
+Figure 7: Effect of initial pH on color, COOD, DOC and BODv
+
+## Acknowledgements
+
+The authors would like to thank ITU Research Fund for financial support and Ali Osman Kilictioglu and Canan Fidanci from Altinyildiz Mensucat ve Konfeksiyon AS and Izzet Apikyan from PISA Tekstil ve Boya Fabrikalari AS for supplying wastewaters and dyestuff samples during the course of the study.
+
+## References
+
+* [1] Corrcia VM, Stephenson T and Judd SJ, Characterization of textile wastewater--A review. _Environ Technol_**15**(10):917-929 (1994).
+* [2] Wu J and Wang T, Ozonation of aqueous azo dye in a semi-batch reactor. _Wat Res_**34**(4):1093-1099 (2001).
+* [3] Apkin R and Waite TD, Comparison of three advanced oxidation processes for degradation of textile dyes. _Wat Sci Tech_**42**(5-6):345-354 (2000).
+* [4] Gould JP and Groff KA, The kinetics of zoonolysis of synthetic dyes. _Osnoa-Sci Eng_**9**(2):153-164 (1987).
+* [5] Tziriti M, Vayens DV and Lyberatos G, Pretreatment of textile industry wastewater with ozone. _War Sci Tech_**29**(9):151-160 (1994).
+* [6] Chu W and Ma CW, Quantitative prediction of direct and indirect dye ozonation kinetics. _War Res_**34**(12):3153-3160 (2000).
+* [7] Saunders FM, Gould JP and Southerlend CR, The effect of solute competition on ozonolysis of industrial dyes. _War Res_**17**(10):1407-1419 (1983).
+* [8] Gahr F, Hermanus F and Oppermann W, Ozonation--An important technique to comply with new German Laws for textile wastewater treatment. _Wat Sci Tech_**30**(3):255-263 (1994).
+* [9] Sevimi MF, Color removal from textile wastewater by ozonation and the effect of ozonation on biodegradability (in Turkish), PhD dissertation, Institute of Science and Technology, Istanbul Technical University, (2000).
+* [10] Perkowski J, Kos L and Ledakowicz S, Application of ozone in textile wastewater treatment. _Osnoa-Sci Eng_**18**(1):73-85 (1996).
+* [11] Balcioglu IA and Arslan I, Partial oxidation of reactive dyestuff and synthetic textile dye-bath by the O\({}_{3}\) and O\({}_{7}\)H\({}_{2}\)O\({}_{2}\) processes. _Wat Sci Tech_**43**(2):221-228 (2001).
+* [12] Shu HY and Huang CR, Degradation of commercial azo dyes in water using ozonation and enhanced ozonation process. _Chemosphere_**31**(8):3813-3825 (1995).
+* [13] Carriere J, Jones JP and Broadcatt AD, Decolorization of textile dye solution. _Osnoa-Sci Eng_**15**:189-200 (1993).
+* [14] Arslan I and Balcioglu IA, Effect of common reactive dye auxiliaries on the ozonation of dyebouse effluents containing vinylsulphone and aminochlorotriazine dyes. _Desalination_**130**(1):61-71 (2000).
+* [15] Wu J, Eiteman MA and Law SE, Evaluation of membrane filtration and ozonation processes for treatment of reactive-dye wastewater. _J Environ Eng-ASCE_**124**(3):272-277 (1998).
+* [16] Ince NH and Tezcanli G, Reactive dyestuff degradation by combined sonolysis and ozonation. _Dyes and Pigments_**49**(3):145-153 (2001).
+
+Figure 8: Experimental results and the predicted values of the remaining fraction of the dye concentration.
+
+Figure 10: The effect of initial dye concentration on specific ozone consumption.
+
+Figure 9: Plots for the initial dye concentration versus apparent reaction rate constants.
+
+* [17] Krull R, Hemmi M, Otto P and Hempel DC, Combined biological and chemical treatment of highly concentrated residual dychouse liquors. _Wat Sci Tech_**38**(4-5):339-346 (1998).
+* [18] Hoigne J and Bader H, Rate constant of reaction of ozone with organic and inorganic compounds in water--I. Non-dissociating organic compounds. _War Res_**17**(2):173-183 (1983).
+* [19] Langlais B, Rechhow DA and Brink DR, _Ozone in Water Treatment: Application and Engineering_, Lewis Publisher, Michigan, USA (1991).
+* [20] Teramoto M, Imamura S, Yatagai N, Nishikawa Y and Teranishi H, Kinetics of the self-decomposition of ozone and the ozonation of cyanide ion and dyes in aqueous solutions. _J Chem Eng Japan_**14**(5):383-388 (1981).
+* [21] APHA/AWWA/WEF, _Standard Methods for the Examination of Water and Wastewater_, 19th edn, Washington DC, USA (1995).
+* [22] Kallas J and Munter R, Post treatment of pulp and paper industry wastewaters using oxidation and adsorption processes. _War Sci Tech_**29**(5-6):259-272 (1994).
+* [23] Benitze FJ, Beltran-Heredia J, Acero JL and Pinilla ML, Ozonation kinetics of phenolic acids present in wastewaters from olive oil mills. _Ind Eng Chem Res_**36**(3):638-644 (1997).
+* [24] Yazgan MS, Removal of endosulfan from waters by ozone oxidation (in Turkish), PhD dissertation, Institute of Science and Technology, Istanbul Technical University, (2001).
+* [25] Cogo E, Albert J, Malmary G, Coste C and Molinier J, Effect of reaction medium on ozone transfer and applications to pulp bleaching. _Chem Eng_**73**(1):23-28 (1999).
+* [26] Levespiel O, _Chemical Reaction Engineering_, John Wiley and Sons Inc, New York, NY (1972).
+* [27] Liakou S, Pavlou S and Lyberatos G, Ozonation of azo dyes. _War Sci Tech_**35**(4):279-286 (1997).
+* [28] Lopez A, Ricco G, Mascolo G, Tirawanti G, Di Pinto AC and Pasino R, Biodegradability enhancement of refractory pollutants by ozonation: a laboratory investigation on azo-dyes intermediate. _War Sci Tech_**38**(4-5):239-245 (1995).
+* [29] Zhou H and Smith DW, Process parameter development for ozonation of karst pulp mill effluents. _Wat Sci Tech_**35**(2-3):251-259 (1997).
+* [30] Hoigne J and Bader H, The role of hydroxyl radical reactions in ozonation processes in aqueous solutions. _War Res_**10**(5):377-386 (1976).
+* [31] Stahelin J and Hoigne J, Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. _Environ Sci Tech_**19**(12):120-126 (1985).
+* [32] Beltran FJ, Garcia-Araya JF and Alvarez PM, pH sequential ozonation of domestic and wine-distillery wastewaters. _War Res_**35**(4):929-936 (2001).

manual_annotation/200pdfs_in_mmd/10.1007_s10008-017-3812-z.mmd

@@ -0,0 +1,389 @@
+Degradation of paracetamol in a bubble column reactor with ozone generated in electrolyte-free water using a solid polymer electrolyte filter-press electrochemical reactor
+
+Lindomar G. De Sousa, Jose Geraldo M. C. Junior, Rodrigo M. Verly, Manoel J. M. Pires, Debora V. Franco, Leonardo M. Da Silva
+
+# Abstract
+
+A porous anode composed of \(\beta\)-PbO\({}_{2}\) was electrochemically deposited onto a carbon cloth substrate (e.g., CC/\(\beta\)-PbO\({}_{2}\)) aiming for the electrochemical ozone production (EOP) in electrolyte-free water using a solid polymer electrolyte (SPE) filter-press reactor. Scanning electron microscopy (SEM) images revealed the presence of a three-dimensional oxide structure necessary to obtain a fluid-permeable anode. X-ray analysis showed the predominance of the \(\beta\)-PbO\({}_{2}\) phase. The maximum current efficiency for the EOP was 9.5% with an ozone production rate of 1.40 g h-1. Using a constant ozone production rate of 0.5 g h-1, the oxidative degradation of paracetamol (PCT) dissolved in water was accomplished as a function of the PCT concentration (20, 30, and 50 mg L-1) and the pH (acid, natural (without adjustment), and alkaline). The UV-Vis spectrophotometric analysis showed that the degradation process is more pronounced in alkaline media with a strong reduction in the electrical energy per order (\(E_{\text{EO}}\)). A reduction of the chemical oxygen demand (COD) of up to 80% was observed. A linear correlation between data referring to COD and HPLC measurements with the UV absorbance measured at 243 nm (UV243) was verified indicating that these different techniques can be complementary to each other. The nuclear magnetic resonance (NMR) study of the ozonation by-products revealed that the oxidation of PCT occurred through the rupture of the aromatic ring. The major part of phenol's ring was oxidized to CO3-2 while no reaction occurs in the acetamide group of paracetamol during the ozonation reaction.
+
+## Introduction
+
+Ozone is a powerful oxidizing agent, capable of participating in a large number of reactions with organic and inorganic compounds [1]. Thus, in recent years, studies have been intensified in order to improve its production as well as its application in the mineralization of different organic pollutants. Among the pharmaceuticals (e.g., emerging pollutants) used in recent studies involving the ozonation, a special attention was given to paracetamol, a drug with analgesic, antipyretic, and mild anti-inflammatory effects. This drug was already found in aquatic environments since approximately 90% of the drug is excreted by the human body without being metabolized [2]. Therefore, the main route for the environmental contamination by several types of pharmaceuticals is the domestic sewage. In addition, several drugs found in polluted waters cannot be adequately treated by the conventional biological treatment methods [1, 2].
+
+Therefore, several treatment technologies based on the heterogeneous photocatalysis, ozone/Fe2+/UVA, electrochemistry, ozonation, H2O2/UV, etc. were investigated to find alternative technologies for the degradation of several drugs, as is the case of paracetamol, aiming for minimization of the deleterious impact of the drugs in the environment [3-8]. In some cases, the drug oxidation was characterized by rupture of the aromatic ring, with a partial conversion of the organic carbon content into carbon dioxide [6, 7]. In this sense, in the last years, a special attention has been given to treatment methods that are capable of generating "in situ" the hydroxyl radicals (HO--_E_o = 2.80 V), as is the case of the ozonation [1, 9, 10].
+
+Ozone applications in wastewater treatment plants contribute to the decontamination process in at least two important aspects [9, 10]: (i) increase in biodegradability of the dissolved organic matter and (ii) introduction of a considerable amount of oxygen in water, thus creating excellent conditions for the biological process implemented in a separated treatment stage. In addition, the ozone degradation in water can lead to the "in situ" formation of hydroxyl radicals [1, 10].
+
+From the above considerations, a special attention was given in the last three decades to the "electrochemical ozone production" (EOP), owing to the high concentration of the dissolved ozone obtained in this technology [1, 9, 10]. In this sense, lead dioxide is the most used anode for the EOP since this electrode material is inexpensive and highly stable at high anodic potentials [11-15], i.e., the contamination of water by Pb2+ ions can be disregarded when the anode is used in conjunction with a solid polymer electrolyte [9, 12]. Therefore, there is great interest in improving the properties of the lead dioxide anode, PbO2, as is the case of the fabrication of a mechanically stable three-dimensional anode for applications in SPE filter-press cells that operate using electrolyte-free water.
+
+The lead dioxide anodes prepared by electrodeposition commonly exhibits a compact surface morphology which is stable to wear (erosion and/or corrosion) during intense gas evolution (e.g., oxygen evolution reaction, OER) [14]. Depending on the conditions employed during the electrodeposition, the lead dioxide layer can be obtained in two crystalline forms: \(\alpha\)-PbO2 (orthorhombic structure), originated from alkaline solutions, and \(\beta\)-PbO2 (tetragonal structure), originated from acidic solutions [15]. The \(\beta\)-PbO2 phase is the preferred one in several electrochemical applications due to its higher overpotential for the OER [16] and a better electronic conductivity when compared to the \(\alpha\)-PbO2 phase [15, 17].
+
+When hydrated, PbO2 is classified as a metallic conductor (high electron density, 1020-1021 cm-3) [18]. This conductivity is attributed to the non-stoichiometry of the oxide, due to its oxygen vacancies [19] or the substitution of surface oxygen for hydroxyl radicals as a result of the film hydration [20], represented by PbO2-_x_(yH2O), where "2-_x_" is the deviation from the ideal stoichiometry and "yH2O" represents the amount of water present in the oxide structure.
+
+A survey of the literature revealed that the PbO2 layer was already electrochemically deposited onto different conducting substrates, planar or porous, such as Ti, Pt, Ti-Pt, Al, Ta, Cu, Pb, Au, carbon cloth, and stainless-steel [9, 21-30]. However, so far only a few attempts were given to the application of the porous \(\beta\)-PbO2 anode in SPE filter-press cells aiming for the environmental applications of ozone involving the water and wastewater treatments [1, 25].
+
+In the present study, the porous \(\beta\)-PbO2 layer was electrochemically deposited onto a carbon cloth substrate (serge type) in order to obtain a fluid-permeable anode for application in an SPE filter-press reactor. A systematic investigation of the EOP process was conducted using electrolyte-free water. The electrochemically generated ozone was applied on the degradation of aqueous solutions containing the drug paracetamol under semi-batch conditions using a bubble column reactor. The ozonated solutions were characterized using different techniques, including the nuclear magnetic resonance (NMR).
+
+## Experimental
+
+### Chemicals
+
+Paracetamol (CsH3NO2) (Acetaminophen--CAS: 103-90-2, purity > 99%) was purchased from Sigma-Aldrich. The pH control was made by adding some drops of a sulfuric acid solution (Merck) or by the addition of some drops of a sodium hydroxide solution (Sigma-Aldrich). Organic solvents and other chemicals employed were of high performance liquid chromatography (HPLC) grade (Fluka). Ultra-pure water used for preparing the solutions was obtained using a model MS 2000 purification system from Gehaka (Brazil) with a resistivity of 18.2 M\(\Omega\) cm at 25 degC.
+
+### Study of the OER/EOP processes using quasi-stationary polarization curves
+
+Quasi-stationary potentiostatic polarization curves were recorded in electrolyte-free water at a scan rate (\(\nu\)) of 1.0 mV s-1 from the equilibrium potential (OCP value) until a potential value corresponding to a maximum apparent current density of 150 mA cm-2. Experiments were carried out in triplicate using a model PGSTAT 128N Potentiostat from AUTOLAB (The Netherlands). The SPE cell used in the present work was previously reported by Costa and Da Silva [24]. Ultra-pure water (e.g., electrolyte-free water) was obtained using a model MS 2000 Purification System from Gehaka (Brazil) with a resistivity of 18.2 M\(\Omega\) cm at 25 degC.
+
+### Fabrication of the \(\beta\)-PbO2 layer for the SPE filter-press reactor
+
+Lead dioxide layers (\(\beta\)-PbO2) presenting a three-dimensional structure supported on fibers of a carbon cloth substrate were prepared by electrodeposition in a 0.1 dm3 single-compartment cell equipped with two graphite bars (counter electrodes) presenting each a geometric area of 16 cm2 [24]. The solution was magnetically stirred during electrodeposition. The carbon cloth CCS200 (serge type and \(e\) = 0.34 mm) furnished by Maxepoxi Co. (Brazil) used as the substrate was previously cleaned using isopropanol and then rinsed with water. A fine insulating layer (_e_ >= 0.1 mm) was applied on the edges of the substrate using silicon glue (Tekbond, Brazil) in order to avoid damages for the serge-type weaving during the cutting process using a scissor. In addition, this procedure minimized the "edge effect" that leads to a preferential electrodeposition of the lead dioxide on the edges of the substrate. A mold made of Teflon (_e_ = 0.5 mm and 4.0 cm x 5.0 cm) was used during application of the insulating layer in order to delimitate the electrodeposition area for each side of the substrate [24]. After the careful removal of the Teflon mold, the sample was left in air for 24 h at 25 degC. This procedure was repeated for the other side of the substrate. After that, the portion of the substrate devoted to electrodeposition (4.0 cm x 5.0 cm) was totally immersed in the electrodeposition solution. _b_-PbO2 layers presenting a geometric area for each side of 20 cm2 was prepared by electrodeposition using a 0.2 mol dm-3 Pb(NO3)2 + 0.01 mol dm-3 HNO3 solution at 50 degC, applying a constant apparent current density of 40 mA cm-2 for 30 min. After the electrodeposition, the uncovered part of the carbon cloth used for the electric contact was removed resulting in the final configuration of the anode used in the SPE cell. Three samples were prepared in order to check the reproducibility of the experimental findings. Electrodes were rinsed thoroughly with deionized water to remove any traces of Pb2+ ions and then air-dried and stored appropriately. The average mass obtained for the _b_-PbO2 electrodes was 100 +- 3 mg cm-2. Electrodeposition was carried out using a power source (3 A/30 V) from ICEL (Brazil). Vetec (Brazil) "purum p.a." products were used throughout.
+
+## Cathode preparation for the SPE filter-press reactor
+
+A carbon cloth (CCS200: serge type; _A_G = 20 cm2 and \(e\) = 0.34 mm) supplied by Maxepoxi Co. (Brazil) was used as the cathode, which was previously cleaned using isopropanol and then rinsed with deionized water.
+
+## Membrane electrode assembly and the SPE filter-press reactor
+
+A system comprising the electrodes (anode and cathode), solid polymer electrolyte (SPE), and perforated current collectors was assembled using a specially designed cell housing made of acrylic, in which the electrodes were pressed against the SPE (Nafion(r) 324--Teflon fabric reinforced) membrane from Dupont (Brazil) using a clamping system [9]. Fluid manifolds (water distribution channels) were machined into the intermediate acrylic plates to facilitate the supply of water to the SPE/electrode interface (active zones for electrolysis).
+
+To obtain the desired configuration for the SPE cell, a stainless-steel mesh (AISI-304: _A_G = 20 cm2, \(Q\) = 0.2 mm x 0.2 mm, and \(e\) = 0.2 mm) was placed between the electrodes (anode and cathode) and the perforated current collectors, which was also made of stainless steel (AISI-304). These auxiliary stainless-steel meshes were used to propitiate a uniform distribution of pressure on the membrane electrode assembly (MEA), since the "sandwich" (collector/mesh/anode/SPE/cathode/mesh/collector) was compressed through springs fixed at the edges of the perforated current collectors [9]. A pressure of 0.5 kgf cm-2 was applied by fastening spring-loaded screws (a clamping system) that were fixed in the perforated current collectors in order to promote an adequate mechanical/electrical contact at the electrode/SPE interface. This procedure ensured adequate compression of the SPE, providing the necessary conditions for the zero-gap configuration [24]. The general scheme of the elements composing the SPE filter-press reactor used in the present work was previously described by some of the present authors [9].
+
+Reinforced Nafion(r) 324 (Dupont, Brazil) was pre-treated by immersing it into boiling 50 _v_/_v_% HNO3 solution for 30 min and then in boiling deionized water for 2 h to provide adequate hydration of the membrane [24].
+
+The volume of the anodic and cathodic reservoirs was 2.0 dm3 (see Fig. 1). The electrolysis of electrolyte-free water was carried out by recirculation through a plug-flow batch reactor at a volumetric flow rate (_Q_) of 23.6 cm3 s-1. The hydraulic behavior of the anodic fluid inside the manifolds containing S-shaped distribution channels (dimensions: cross-sectional area (_A_SE) = 0.30 cm2, length (_l_) = 80 cm, and channel volume (_V_0) = 24 cm3) pressed against the perforated current collector (zero-gap condition) was characterized by a residence time value, _t_r = (_V_/_Q_), of 1.02 s and a fluid velocity, _u__r = (_Q_/_A_SE), of 78.7 cm s-1. In all cases, the reactor was powered using a power source (100 A/12 V) from AMZ (Brazil). All experiments were carried out using electrolyte-free water (_r_ = 18.2 Ml cm at 25 degC).
+
+## Electrochemical ozone production: characterization of the SPE filter-press reactor
+
+The gas mixture (O2/O3) that leaves the anodic compartment was separated from the electrolyte-free water that flows through a gas separator flask and directed to the spectrophotometer for the UV analysis at 254 nm (see Fig. 1).
+
+The ozone concentration in the gas phase was analyzed by measuring the UV absorbance at 254 nm, according to the methodology described by De Sousa et al. [25]. The partial current _I_EOP and the current efficiency _Ph_EOP for EOP were calculated using the following equations [25]:
+
+\[I_{\text{EOP}}(\text{A}) = \frac{AG_{2}F}{\varepsilon l}\]
+
+\[\Phi_{\text{EOP}}(\text{wt}\%) = \frac{I_{\text{EOP}}}{I_{\text{T}}},\]
+
+where \(A\) = absorbance at 254 nm; \(G\) = volumetric flow rate of the gas mixture (O2 + O3) (dm3 s-1); \(z\) = number of electrons (= 6); \(e\) = ozone molar absorptivity at 254 nm (3024 cm-1 mol-1 dm3 [12]; \(l\) = optical path length (1 cm); _I_T = total current (A); and \(F\) = 96,485 C mol-1.
+
+The specific electric energy consumption, _P_EOP was calculated using Eq. (3) [25]:
+
+\[P_{\text{EOP}}\left( {\text{Wh g}^{ - 1} } \right) = \frac{U_{2}F}{1.73 \times 10^{5}\Phi_{\text{EOP}}},\]
+
+Fig. 1: Flow diagram representing the experimental setup used for the electrochemical ozone production and ozonation of the paracetamol solutions using a bubble column reactor where \(U\) is the cell voltage.
+
+The ozone production rate, _m_EOR, was calculated using Eq. (4) [25]:
+
+\[\nu_{\text{EOP}}\left( {\text{g h}^{ - 1}} \right) = \frac{3600I_{\text{EOP}}M}{zF},\]
+
+where \(M\) is the molecular weight of O3 (48 g mol-1).
+
+The figure-of-merit denoted as "mass gain of O3 per total energy consumption," _o_EOR representing the overall reactor performance was calculated using the Eq. (5) [25]:
+
+\[\vartheta_{\text{EOP}}\left( {\text{g W}^{ - 1}\,\,{\text{h}^{ - 1}}} \right) = \frac{\nu_{\text{EOP}}}{I_{T}U}\]
+
+### Ozonation of the paracetamol solutions using the bubble column reactor
+
+For the degradation of paracetamol, the electric current applied to the reactor and the temperature of the circulating electrolyte-free water were both adjusted to provide a constant ozone production rate of 0.5 g h-1. This value was chosen based on the geometrical properties of the bubble column reactor (e.g., liquid height, volume, and the diameter of the sintered porous glass plate) in order to maximize the ozone utilization efficiency (OUE) during the ozonation of paracetamol solutions.
+
+The oxidative degradation of this drug during the ozonation process was studied under semi-batch conditions (_V_ = 1.0 dm-3) using a bubble column reactor (see Fig. 1). The gas mixture (O2/O3) was bubbled through a porous plate diffuser made of sintered glass (Schott #2 and \(A\) = 3.5 cm-2) placed at the bottom of the bubble column reactor. In addition to the evaluation of the influence of the initial concentration of the drug (20, 30, and 50 mg dm-3), the influence of the pH (2, natural (pH 6.3), and 10) on the degradation process was also studied. The pH was adjusted using a syringe containing one of the following solutions: 0.1 mol dm-3 H2SO4 or 0.1 mol dm-3 NaOH. The pH was monitored using a glass electrode connected to a model DM-22 pH meter from Digimed (Brazil).
+
+To evaluate the degradation process, samples (_V_ = 3 cm-3) were withdrawn every 5 min for the UV-Vis analysis until the total ozonation time of 60 min. In this particular study, all ozonated samples were returned to the bubble column reactor after the measurement of the absorbance. Additional samples (_V_ = 1 cm-3) were withdrawn for the chromatographic (HPLC) analysis.
+
+The ozonated samples were analyzed using a model Cary 50 spectrophotometer from Varian and a model ProStat 315 HPLC from Varian, where a model LC 018 \(C\)18 column from Supel Casil(tm) was used in the reverse mode. The UV detector was set at 243 nm to monitor the concentration of paracetamol.
+
+Also, to evaluate the degradation process, the chemical oxygen demand (COD) analysis was carried out according to the methodology reported in the standard methods of analysis [31]. In this case, COD values were determined for the treated samples and recorded as a function of the ozonation time at regular intervals of 10 min. COD is a measure of the oxygen equivalent of the organic matter content of the sample that is susceptible to oxidation by a strong chemical oxidant (e.g., potassium dichromate). In this sense, silver sulfate was used as a catalyst to promote the oxidation of certain classes of organic compounds (e.g., oxidation by-products). The sample vials (_V_ = 10 cm-3) with cap were heated at 150 degC for 2 h using the COD reactor block heating from HACH (Brazil). After cooling the samples, the COD values were determined by titration of the amount of unreacted dichromate with ferrous ammonium sulfate using ferroin as the indicator. The potassium hydrogen phthalate was used as a standard. All experiments were carried out in triplicate.
+
+The chemical structure of paracetamol is shown in Fig. 2.
+
+### Characterization of the oxidation by-products using NMR spectroscopy
+
+The NMR spectra of paracetamol and its oxidation by-products were recorded using a model FOURRIER 300 NMR spectrometer from Bruker (Switzerland) operating at 300.18 MHz for H1 and 75.48 MHz for C13, following standard protocols [32]. In this particular study, the oxidation by-products was obtained after the ozonation carried out over a period of 240 min (conditions: _v_EOP = 0.5 g h-1, \(V\) = 250 mL, [paracetamol]0 = 250 mg dm-3, and pH 10).
+
+The paracetamol and its oxidation by-products were dissolved in a mixture of H2O/D2O (90/10, _v_/_v_) containing 5 mmol dm-3 of tetramethyl-silapentane and using 10 mmol dm-3 of TMS (internal reference). In all cases, a one-dimensional (1D) NMR spectra were recorded using a 1H/13C standard 5 mm probe at 20 degC. All experiments were acquired and processed using the Topspin 3.1 software developed by Bruker BioSpin. 1H spectra were acquired with water suppression by using pre-saturation techniques with zgcppr pulse sequence. The spectral width was 6103.52 Hz for both samples, 512 numbers of scan, 2.0 number of dummy scans, and 16,384 of time domain data size. Decoupled 13C and DEPT-135 13C spectra were acquired using zgpg30 and dept135 pulse sequence, respectively. 13C spectra were
+
+Fig. 2: Chemical structure of paracetamolacquired for both samples with spectral width of 24,414.06 Hz, 4096 numbers of scan, 4 number of dummy scans, and 65,536 of time domain data size. DEPT-135 \({}^{13}\)C spectra were acquired with spectral width of 24,414.06 Hz, 2048 numbers of scan, 4 number of dummy scans, and 65,536 of time domain data size.
+
+## Results and discussion
+
+### Ex-situ characterization of the \(\beta\)-PbO\({}_{2}\) electrode
+
+#### SEM analysis
+
+Figure 3 shows the scanning electron microscopy (SEM) images obtained for the porous PbO\({}_{2}\) layer electrochemically deposited on the carbon cloth (serge type) substrate.
+
+As can be seen, the electrodeposition process resulted in a good coverage of fibers of the carbon cloth by the oxide layer. In addition, the porous structure of the carbon substrate was preserved resulting in a porous PbO\({}_{2}\) layer that exhibits a three-dimensional structure. A comparison with the literature [33; 34] revealed that the absence of a metal to support the oxide layer did not affect the morphological characteristics of the individual PbO\({}_{2}\) crystals. However, the use of carbon cloth lead to the formation of isolated grains, that is, there was the formation of a three-dimensional porous structure that can allow the water flow and transport of the gaseous products (O\({}_{2}\) and O\({}_{3}\)) through the microstructure of the MEA. As a result, the electrochemically deposited porous PbO\({}_{2}\) layer can be used as a fluid-permeable anode (FPA) in SPE filter-press reactors [9].
+
+#### XRD analysis
+
+Figure 4 shows the X-ray diffractogram obtained for the PbO\({}_{2}\) layer supported on the carbon cloth substrate.
+
+The XRD spectrum shows well defined narrow peaks, which indicate that the lead dioxide layer has good crystallinity. In addition, it was verified that the \(\beta\)-PbO\({}_{2}\) phase is predominant over the \(\alpha\)-PbO\({}_{2}\) phase. These observations are in accordance with the literature [26; 35]. An average size of the crystallites of 54 nm was obtained using the well-known Debye-Scherrer equation [36].
+
+### Kinetic study of the OER and OER/EOP processes
+
+The electrode mechanism for the OER can consists of different consecutive steps, being that each one can be the rate-determining step (RDS) [37]. The rate of each step can be influenced by the electrode composition, adsorption energy of the reaction intermediates, geometric arrangement of the atoms on the surface, etc. [18]. From the current-potential curve obtained under quasi-stationary conditions (low scan rate), the Tafel plot can be obtained permitting to identify the RDS.
+
+At least in principle, a "complete" kinetic investigation for a particular electrode reaction also requires the determination of the kinetic parameters denoted as "stoichiometric number" and "reaction order," the latter being relative to the stable species that participate in the reaction (e.g., H\({}^{+}\) or OH\({}^{-}\) ions in case of the OER) [38]. Despite these considerations, the RDS was determined in the present work based only on the Tafel slope since the use of an SPE immersed in electrolyte-free water does not allow the determination of the other aforementioned kinetic parameters. In this sense, since the values obtained for the Tafel slope in the present study were high (\(b\geq 120\) mV dec\({}^{-1}\) (25 \({}^{\circ}\)C)), it is likely to consider the RDS as being the primary water discharge step (e.g., the first electron transfer) [37] (see further discussion in this section).
+
+Following the theoretical approach proposed by Da Silva et al. [39; 40], the experimental values of "\(b\)" were used to obtain the apparent charge-transfer coefficient (\(\alpha_{\text{app}}\)) for the primary water discharge step. The ideal value of the charge-transfer coefficient (\(\alpha\)) at 25 \({}^{\circ}\)C is 0.50 for the hypothetical case of a perfectly flat electrode and in the absence of the specific adsorption [37; 38; 39; 40]. Therefore, an ideal value of 120 mV dec\({}^{-1}\) is predicted for the Tafel slope (\(b\)) in the case where the primary water discharge is the RDS [37; 39].
+According to Bockris [37], initially, the water molecule is oxidized at low current densities originating the adsorbed HO radicals (e.g., primary water discharge step, _b_ideal = 120 mV dec-1). In the next step, the HO radicals are oxidized originating the adsorbed O radicals which react between them leading to the formation of O2. As proposed by Da Silva et al. [39, 40], the O2 molecule can stand temporarily adsorbed on the electrode surface thus favoring the ozone formation reaction (e.g., O(ads) + O2(ads) - O3(ads)) and/or it can agglomerate on the electrode surface permitting the coalescence with the release of gas bubbles.
+
+In fact, the high surface concentration of the oxygenated species (HO(ads), O(ads), and O2(ads)) can propitiate the occurrence of the EOP. Different electrode mechanisms proposed to represent the OER/EOP processes, taking into account different reaction intermediates, can be found in the literature [1, 11, 16]. However, these mechanisms do not permit a direct correlation between kinetic, surface adsorbed species, and current efficiency for EOP. A comprehensive kinetic study regarding the OER/EOP processes was previously reported by Da Silva et al. [39, 40]. In these studies, an electrode mechanism was proposed in order to describe the theoretical current efficiencies for the OER and EOP processes as a function of the surface coverages of the reaction intermediates. Also, it was verified for the \(\beta\)-PbO2 electrode that the first electron transfer (primary water discharge step) present in the electrode mechanism referring to the OER/EOP processes is the RDS [39, 40]. The electrode mechanism proposed by Da Silva et al. [39, 40] is presented as follows:
+
+Electrochemical steps: _Kinetic control_
+
+\[\left( {{\text{H}}_{2}{\text{O}}} \right)_{\text{ads}} \to \left( {{\text{HO}}^{\prime}} \right)_{\text{ads}} + {{\text{H}}^{+}} + {{\text{e}}^{-}}\] RDS (6a)
+
+\[\left( {{\text{HO}}^{\prime}} \right)_{\text{ads}} \to \left( {{\text{O}}^{\prime}} \right)_{\text{ads}} + {{\text{H}}^{+}} + {{\text{e}}^{-}} \tag{6b}\]
+
+Chemical steps: _Efficiency control_
+
+\[\left( {{\text{O}}^{\prime}} \right)_{\text{ads}} \to \left[ {1 - \theta } \right]\left( {{\text{O}}^{\prime}} \right)_{\text{ads}}\] \[+ \theta \left( {{\text{O}}^{\prime}} \right)*_{\text{ads}} \left( {0 < \theta < 1} \right) \tag{6c}\]
+
+\[\left[ {1 - \theta } \right]\left( {{2\text{O}}^{\prime}} \right)_{\text{ads}} \to \left[ {1 - \theta } \right]\left( {{\text{O}}_{2}} \right)_{\text{ads}} \tag{6d}\]
+
+\[\left[ {1 - \theta } \right]\left( {{\text{O}}_{2}} \right)_{\text{ads}} \to \left[ {1 - \theta } \right]\left( {{\text{I}} - \theta } \right]\left( {{\text{O}}_{2}} \right)_{\text{ads}}\] \[+ \beta \left[ {1 - \theta } \right]\left( {{\text{O}}_{2}} \right)*_{\text{ads}} \left( {0 < \beta < 1} \right) \tag{6e}\]
+
+_Oxygen evolution_:
+
+\[\left[ {1 - \beta } \right]\cdot \left[ {1 - \theta } \right]\left( {{\text{O}}_{2}} \right)_{\text{ads}} \to {\text{O}}_{2} \tag{6f}\]
+
+_Ozone formation_:
+
+\[\theta \left( {{\text{O}}^{\prime}} \right)*_{\text{ads}} \to \beta \left[ {1 - \theta } \right]\left( {{\text{O}}_{2}} \right)*_{\text{ads}} \to \left[ {\theta + \beta \left( {1 - \theta } \right)} \right]\left( {{\text{O}}_{3}} \right)_{\text{ads}} \tag{6g}\]
+
+\[\left[ {\theta + \beta \left( {1 - \theta } \right)} \right]\left( {{\text{O}}_{3}} \right)_{\text{ads}} \to {\text{O}}_{3} \tag{6h}\]
+
+where "_th_" and "_th_" represent the surface coverages by oxygen species while "*" represents the fractional surface coverage leading to the ozone formation (e.g., EOP).
+
+As proposed by Da Silva et al. [40], the theoretical current efficiencies with respect to OER (_Ph_OER) and EOP (_Ph_EOP) processes are a function of the _th_ and \(b\) coverages as follows:
+
+\[\Phi_{\text{OER}} = \left[ {1 - \beta } \right]\cdot \left[ {1 - \theta } \right]\]
+
+\[\Phi_{\text{EOP}} = \left[ {\theta + \beta \left( {1 - \theta } \right)} \right]\]
+
+Theoretical calculations [40] showed that a maximum _Ph_EOP value is obtained for _th_ and \(b\) values tending to unity. Under these conditions, the _ozone formation_ (steps (6 g) and (6 h)) is favored over the _oxygen evolution_ (step (6f)). As a result, under these conditions the generation of oxygen is minimum, serving only as a source of adsorbed O2-species necessary for the ozone formation on the electrode surface.
+
+Figure 5 shows the Tafel plot obtained at 1.0 mV s-1 for the OER (low overpotential domain) and OER/EOP (high overpotential domain) on the CC/_b_-PbO2 electrode covering
+
+Fig. 4: X-ray diffractogram obtained for the _β_-PbO2 layer electrochemically deposited onto the carbon cloth substrate
+
+Fig. 5: Tafel plot obtained for the OER and OER/EOP processes on the CC/_β_-PbO2 electrode using electrolyte-free water (ν = 1.0 mV s−1 and \(T\) = 24 °C): (a) before and (b) after the correction for the ohmic drop the current density interval of ~ 0.1 to 150 mA cm-2. It is worth mentioning that according to the spectrophotometric analysis carried out at 254 nm (see Fig. 1) when \(j_{\rm ap}\geq 50\) mA cm-2, the EOP occurs together with the OER.
+
+As can be seen, the Tafel plot, not corrected for the ohmic drop, exhibited a straight line at low current densities and an ascending curvature at higher current densities. According to the literature [39-43], this deviation from the linearity is due to the sum of the uncompensated ohmic resistances, \(R_{\Omega}\), which may be given in the present case by \(R_{\Omega}\) = \(R_{\rm SPE}\) + \(R_{\rm film}\). However, due to the low resistivity of \(\beta\)-PbO\({}_{2}\) (\(\rho\simeq 0.95\times 10^{-4}\)\(\Omega\) cm) [15, 18] associated with the reduced film thickness (\(\varepsilon\sim 40\)\(\upmu\)m), one can argue that \(R_{\rm SPE}\)\(>\)\(>\)\(R_{\rm film}\), hence \(R_{\Omega}\cong R_{\rm SPE}\). After the correction for the ohmic drop (\(R_{\Omega}\) = 5.8 \(\Omega\) cm2, a value obtained using the classical impedance method where \(R_{\Omega}\) = \(Z_{\rm real}\)(\(\omega\rightarrow\infty\))), two Tafel slopes were observed: \(b_{1}\) (low overpotentials) and \(b_{2}\) (high overpotentials).
+
+According to the literature [11, 15, 39, 43, 44], there are many reasons for the existence of two slopes in the Tafel plots for the gas-evolving reactions (e.g., OER and OER/EOP). However, in the present case, it is more likely to consider that the change in the Tafel slope is associated with a change in the effective electrode surface area due to the competitive reactions (formation and accumulation of oxygen and ozone) occurring simultaneously on the electrode surface. While these species reduced the electrode surface area for the oxidation of water molecules, they also increased the surface resistance and thus the electron transfer from water to the electrode would become slower (reduced rate).
+
+As previously discussed by Costa and Da Silva [24], the HO radicals can remain strongly attached to the gel-layer formed on surface of the lead dioxide electrode while the O' radicals are free to move over the electrode surface. Thus, assuming that the primary water discharge step (e.g., first electron transfer) is the RDS, the current density (\(j\)) for the OER or OER/EOP processes in the high-field approximation (\(\eta\)\(>\) 0.1 V) is given as follows [45]:
+
+\[j=4Fk_{\rm ap}^{o}[H_{2}O]_{\rm MFA}{\rm exp}\left[\frac{\alpha_{\rm ap}F_{1}}{RT}\right] \tag{9}\]
+
+where \(k_{\rm ap}^{o}\) is the apparent rate constant, [H\({}_{2}\)O]\({}_{\rm MFA}\) is the water concentration inside the MEA, and \(\alpha_{\rm ap}\) is the apparent charge-transfer coefficient. The other symbols have their usual meaning. It is worth mentioning that Eq. (9) can be indeed applied considering the case when the hydroxyl radicals (HO') are present on the electrode surface at a low coverage, i.e., the HO' radical is formed in the slow step and rapidly consumed in the following rapid step [45].
+
+Since the Tafel slope is defined by \(b=(\partial\eta/\partial{\rm log}(j))_{\rm P}\), the expression for \(\alpha_{\rm ap}\) is obtained as follows [24, 45]:
+
+\[\alpha_{\rm ap}=1.985\times 10^{-4}K^{-1}{\rm V}\left(\frac{T}{b}\right), \tag{10}\]
+
+where \(b\) is the experimental Tafel slope. In this sense, the experimental \(\alpha_{\rm ap}\) value incorporates the deviations from the ideal case commonly found for the electron transfer occurring on solid electrodes [39, 45].
+
+The kinetic parameters (\(b\), \(\alpha_{\rm ap}\),\(j_{\rm o(ap)}\), and \(k_{\rm ap}^{o}\)) obtained for the OER or OER/EOP processes on the CC/\(\beta\)-PbO\({}_{2}\) electrode during electrolysis of the electrolyte-free water are gathered in Table 1.
+
+As can be seen, the OER and OER/EOP processes were characterized by Tafel slopes of 411 and 287 mV dec-1, respectively. As a result, \(\alpha_{\rm ap}\) values of 0.144 and 0.206 were obtained, respectively. The deviation from the ideal case (\(\alpha_{\rm ap}\neq 0.5\)), frequently found for solid electrodes, may be attributed to the influence of the non-uniform distribution of the electric field on the rough electrode surface and/or the adsorption of gas bubbles [39, 46]. The present kinetic findings are in agreement with those reported by Costa and Da Silva [24]. Other works [39, 40, 46] also reported high Tafel slopes for the OER on the PbO\({}_{2}\) electrode in acidic solutions. For instance, Ho and Hwang [46] found a Tafel slope of 256 mV dec-1, an \(\alpha_{\rm ap}\) of 0.30, and an apparent exchange current density of \(5.48\times 10^{-6}\) A cm-2.
+
+Electrochemical ozone production and characterization of the SPE filter-press reactor in electrolyte-free water
+
+In principle, in the case of electrochemical ozonizers using planar electrodes immersed in liquid electrolytes, all regions of the electrode surface can be electrochemically active [47]. On the contrary, in the specific case of the filter-press SPE reactors that use electrolyte-free water, the active surface sites (reaction centers) present in the three-dimensional (porous) electrodes are restricted to the "triple-contact" regions (e.g., water/SPE/electrode) [9]. The mechanisms for the electron and proton conductions in the SPE filter-press reactors using electrolyte-free water were previously discussed in the literature [1, 9, 24, 25].
+
+According to Da Silva et al. [39], the current efficiency for EOP is a function of the surface concentration of the
+
+\begin{table}
+\begin{tabular}{c c c c c} \hline \(\eta\) domain & \(b\)/mV dec−1 & \(\alpha_{\rm ap}\) & \(j_{\rm{ctop}}\)/A cm−2 & \(k^{o}{}_{\rm ap}\)/cm s−1 \\ \hline Low & 411 & 0.144 & \(7.6\times 10^{-5}\) & \(3.5\times 10^{-9}\) \\ High & 287 & 0.206 & \(3.4\times 10^{-6}\) & \(1.6\times 10^{-10}\) \\ \hline \end{tabular}
+\end{table}
+Table 1: Kinetic parameters obtained from the Tafel plot for the OER and OER/EOP processes on the CC/\(\beta\)-PbO\({}_{2}\) electrode in electrolyte-free water (\(T\) = 24 °C)active centers, which in turn depends on the surface coverage of the atomic oxygen, O(ads), and the interaction of the latter with the oxygen molecule adsorbed on the electrode surface, O2(ads). In this sense, it can be proposed in the case of SPE reactors, using porous electrodes and electrolyte-free water, that a good current efficiency for EOP can be obtained as a result of a high surface concentration of the active centers available for the surface reaction (e.g., O(ads) + O2(ads) \(\rightarrow\) O3(ads)) which formation is favored at high current densities [39, 40].
+
+It is worth mentioning that the use of the electrochemical technology for the "in situ" ozone generation has an important advantage over the well-established corona (silent electric discharge) process, since in the former case it is possible to accomplish the direct dissolution of a high concentration of ozone directly into the water stream, thus eliminating several drawbacks associated with the mass-transfer of ozone to the liquid phase [48].
+
+Figure 6 shows the dependence of the cell voltage, \(U\), the EOP current efficiency, \(\Phi_{\text{EOP}}\), the ozone production rate, \(v_{\text{EOP}}\) and the overall reactor performance represented by \(\vartheta_{\text{EOP}}\) as function of the applied current, \(I\).
+
+It can be noted that there is a great change (\(\approx 2.7\) V) in the cell voltage as a function of the applied current. In the present case, \(U\) values incorporate the uncompensated ohmic components (e.g., SPE resistance and the resistance imposed by gas bubbles confined inside the porous structure of MEA), and the cathodic (hydrogen evolution reaction (HER)) and anodic (OER + EOP) overpotentials. Thus, the specific electric energy consumption for EOP (data not shown) can be affected by different factors. Nishiki et al. [49] reported the characterization of an SPE filter-press electrochemical ozonizer that also used the Nafion(r) N324 membrane as the SPE and a boron-doped diamond (BDD) as the anode. However, it was necessary in this case a very high cell voltage of 16 V in order to obtain a low current of 0.8 A. In principle, these findings indicate a lack of good electric contact between the electrode material (e.g., BDD) and the SPE, decreasing the surface concentration of the active centers for EOP and resulting in large ohmic losses.
+
+The analysis of the EOP current efficiency revealed a maximum value of 9.5%, which is in good agreement with other studies using the lead dioxide electrode supported on different substrates [9, 12, 13, 35, 39, 40, 44, 47, 50]. In addition, it was verified that the ozone production rate increased almost linearly with the applied current. In principle, these findings indicate that the heat dissipation in the "triple-contact" regions is quite efficient, avoiding the ozone degradation induced by thermal effects [1, 12]. In fact, in some cases, the ozone production rate passes through a maximum at \(\approx 1.5\) A cm-2 [12] as a result of the ozone decomposition incurred by an inefficient heat removal in the active regions of the electrode material.
+
+The above findings reveal the importance of using highly stable anode materials since they can permit to adjust the ozone production rate in a very good manner by means of changing the applied current. Therefore, a pre-requisite for the EOP process is the choice of a stable anode since in most of the cases the electrochemical reactor must be operated covering a wide current density range (ca. 0.2-1.5 A cm-2) in order to obtain the desired mass (or concentration) of ozone necessary for a particular application, as is the case of the water treatment for different purposes, as well as the waste-water treatment process [1, 9, 10, 12].
+
+The dependence of \(\vartheta_{\text{EOP}}\) with the applied current revealed a maximum reactor performance at \(\approx I>30\) A. Therefore, great advantages can be obtained when the reactor is operated at high current values.
+
+The CC/\(\beta\)-PbO2 electrode was submitted to an endurance test under galvanostatic conditions using the electrolyte-free water (e.g., \(I\) = 23 A; \(U\) = 7.3 +- 0.2 V, \(T\) = 24 +- 2 degC, and \(t\) = 24 h) in order to verify the electrode stability for the EOP process. According to this study, the CC/\(\beta\)-PbO2 electrode remained highly stable with a constant ozone production rate of 0.50 +- 0.01 g h-1.
+
+Fig. 6: The cell voltage (_U_), the current efficiency for EOP (\(\Phi_{\text{EOP}}\)), the ozone production rate (\(v_{\text{EOP}}\)), and the mass gain of O3 per total energy consumption (\(\vartheta_{\text{EOP}}\)) as a function of the applied current (_I_). \(Q\) = 23.6 cm−3 s−1. Electrolyte-free water at 24 °C1.2 Oxonation of paracetamol under semi-batch conditions using a bubble column reactor and the spectrophotometric analysis of the treated samples
+
+The ozonation of organic compounds are dependent on the pH, the chemical reaction (intrinsic kinetics), mass-transfer of ozone, and the ozone load introduced in the aqueous phase [10, 51, 52]. Therefore, the ozonation process is affected by the type of the gas dispersion system which controls the rate of mass-transfer from the gas to the liquid phase. Also, the driving force for the mass-transfer of ozone is given by the difference in ozone concentration in the distinct phases, i.e., inside the gas phase (bubbles) and in the liquid phase (dissolved ozone) [1, 10, 47]. In fact, as previously discussed by Da Silva et al. [1, 10], the ozonation of contaminated waters (or effluents) strongly depends on the partial pressure of ozone present in the gas phase (e.g., O2 + O3) and also on the mass-transfer of ozone. In the former case, the performance of the ozonation process is governed by the ozone production rate exhibited by the ozone generator system (e.g., corona), while in the latter case it depends on the ozone solubilization in the liquid phase which is promoted by using an efficient gas-liquid contactor system.
+
+When the ozone bubbles are in contact with the water phase there is the formation of a "gas-liquid" interface, where gas absorption (solubilization) is followed by the irreversible oxidation reaction (e.g., ozonation process). As a result, two steps can control the overall ozonation process: (i) the ozone mass-transfer from the gas phase to the liquid phase and/or (ii) the chemical reaction occurring at the gas/liquid interface. According to the "film theory" [1, 10], considering that the gas is sparingly soluble in water, one has that no mass-transfer limitation is observed within the gas phase and, therefore, only the mass-transfer resistance in the liquid phase is considered in practical situations. As a result, it is very likely the formation of a "liquid film" of an average thickness, _d_L, which is established between the "liquid bulk" and the "gas-liquid" interface [10]. In addition, the mass-transfer of ozone is in the steady-state within the film in the absence of mass accumulation [1, 10].
+
+From the above considerations, when the gas mixture exhibiting a high concentration (partial pressure) of ozone is introduced into the contaminated water, the oxidative degradation of paracetamol can be mainly controlled by the elementary kinetic processes occurring in aqueous solution that results in the rupture of the aromatic ring, decreasing the absorbance at 243 nm, and originating several oxidation by-products that can remain stable in solution (e.g., persistent oxidation by-products) [52]. This degradation process comprises only the initial step of an exhaustive advanced oxidation process that can result in a good degradation of the dissolved organic matter (e.g., moderate/high reduction of COD values). However, from a practical point of view, the use of a not exhaustive (partial) degradation process can be preferred in some cases since it requires a reduced amount of ozone, i.e., the partial degradation of the target pollutant may be effective for the removal of the toxicity and, therefore, the pre-treated effluent can be directed to a final treatment process using the conventional biological treatment [10].
+
+Experimentally, the pseudo first order kinetic condition for the ozonation reaction can be ensured by applying a constant ozone concentration in the aqueous phase under semi-batch conditions, where the oxygen/ozone gas mixture is constantly supplied at the bottom of the bubble column reactor at a fixed volumetric flow rate [1, 10, 52]. Therefore, from the theoretical point of view, the ozonation of paracetamol can be described using the following equation [1, 8, 10, 52]:
+
+\[\ln\left(\frac{[C]}{[C]_{0}}\right)=-k_{ap}t, \tag{11}\]
+
+where _k_ap is the apparent rate constant (in min-1) and _k_ap = _k_[O3(a0)]a[HO(a0)]g, where \(a\) and \(g\) are empirical constants. [C] and [C]0 are the instantaneous and initial concentrations of paracetamol, respectively. Considering that the concentration can be represented by the absorbance (e.g., Lambert-Beer's law), the kinetic equation can be presented as follows:
+
+\[\ln\left(\frac{A}{A_{0}}\right)=-k_{ap}t, \tag{12}\]
+
+where the _A_/_A_o ratio represents the normalized absorbance obtained at the wavelength of maximum absorption (_l_max = 243 nm).
+
+UV-Vis spectra for the aqueous solutions containing paracetamol were recorded as a function of the ozonation time. As seen in Fig. 7 (acidic solution), the spectrum of paracetamol (_t_ = 0) was characterized by an intense band at 243 nm which was used to monitor the degradation of the dissolved organic matter (paracetamol + oxidation by-products) as a function of
+
+Fig. 7: Influence of the ozonation time on the absorption spectra of paracetamol in aqueous solution. Conditions: _pH_ 2; _μ_EOP = 0.5 g h−1, \(T\) = 24 °C, and [paracetamol]o = 50 mg dm−3the ozonation time. Different drug concentrations (20, 30, and 50 mg dm-3) and pHs (2, natural (6.3), and 10) were considered in this study. Figure 7 clearly showed that the absorbance was considerably reduced during the ozonation in acidic conditions. Similar findings were obtained for the other pH values (data not shown).
+
+In principle, the absorbance removal in the UV region indicates the occurrence of an electrophilic attack on the aromatic ring promoted by the oxidizing agents: O3 (acidic and neutral solutions) and HO' (alkaline solution) [7, 10]. Usually, the deprotonated species formed in this reaction react rapidly with the ozone in an electrophilic attack since they are strong nucleophiles [51, 52].
+
+Figure 8 shows the influence of the pH on the degradation of paracetamol.
+
+As can be seen, the removal of the absorbance was more efficient in alkaline solution, followed by the acidic and neutral solutions, respectively. In the case of the alkaline solution, after 45 min of ozonation, there was stabilization of the degradation process. Thus, the indirect (less selective) oxidation process which is mainly associated with the electrophilic attack promoted by the hydroxyl radicals on the target substance resulted in a more effective absorbance removal.
+
+Andreozzi et al. [6] reported similar findings for the degradation of paracetamol using the ozonation and the H2O2/UV system. The reaction mechanism proposed by these authors indicates that the electrophilic attack (O3 and/or HO) mainly occurs on the aromatic ring with the formation of glyoxal and oxalic acid (acidic conditions) and formic acid (alkaline conditions).
+
+As shown in Fig. 9, a good linear behavior was observed in all cases (_r_2 > 0.980) thus supporting the assumption of the pseudo first order conditions. The values of _k_ap were gathered in Table 2.
+
+As can be seen, there was a progressive reduction in _k_ap values as a function of the concentration of paracetamol. In principle, this behavior is expected since for a fixed amount of the oxidizing agent supplied to the gas/solution interface there is an increase in the amount of the target substances for the electrophilic attack propitiated by the O3 and/or HO' oxidizing agents. More precisely, in acidic or neutral solutions, the degradation of paracetamol is mainly governed by the direct ozonation since the concentration of the hydroxyl radicals can be quite low in this case [1, 7].
+
+The figure-of-merit denoted as "electrical energy per order" (_E_EO) (expressed in W h m-3 order-1) was calculated using the following relationship [53]:
+
+\[E_{\text{EO}} = \frac{38.4P}{Vk_{\text{ap}}},\]
+
+where \(P\) is the rated power of the electrochemical ozonizer (in W), \(V\) is the effluent volume (in m3), and _k_ap is the apparent rate constant (in h-1).
+
+Table 2 shows the _E_EO values consumed for treating the samples (_V_ = 1.0 dm3) as a function of the operating variables (e.g., paracetamol concentration and pH) for an ozonation time of 60 min.
+
+The analysis of data presented in Table 2 (0.020 min-1 < _k_ap < 0.159 min-1) revealed that _k_ap values are low in
+
+\begin{table}
+\begin{tabular}{c c c c} \([\text{paracetamol}]_{\text{o}}/\text{mg}\) & pH & _k_ap/min−1 & _E_EO/kW \\ dm−3 & & & h m−3 order−1 \\
+20 & 2 & 0.109 & 38 \\
+30 & 2 & 0.068 & 61 \\
+50 & 2 & 0.028 & 148 \\
+20 & nat & 0.069 & 60 \\
+30 & nat & 0.073 & 57 \\
+50 & nat & 0.020 & 207 \\
+20 & 10 & 0.159 & 26 \\
+30 & 10 & 0.111 & 37 \\
+50 & 10 & 0.088 & 47 \\ \end{tabular}
+\end{table}
+Table 2: Electrical energy per order (_E_EO) consumed during ozonation and _k_ap values obtained as a function of the operating variables
+
+Fig. 8: Dependence of the absorbance (λ = 243 nm) on the ozonation time. Conditions: pH 2; _ν_E_OP = 0.5 g h−1, \(T\) = 24 °C, and \([\text{paracetamol}]_{\text{o}} = 50\) mg dm−3
+
+Fig. 9: Pseudo first order kinetic profiles obtained as a function of the initial concentration of paracetamol. Conditions: pH 10, \(T\) = 24 °C, and \(\nu_{\text{LED}} = 0.5\) g h−1comparison with the experimental value of the volumetric mass-transfer coefficient of ozone (_k_1a = 0.41 min-1), which was determined in the present study following the procedure previously described in the literature [54]. These findings indicate that the chemical reaction comprising the oxidative degradation of paracetamol is the slow process when compared to the ozone mass-transfer in the bubble column reactor. Therefore, the ozonation process accomplished in the bubble column reactor is mainly governed by the oxidative process (e.g., direct (O3) and/or indirect (HO)) occurring in the solution phase.
+
+In addition, the analysis of the experimental findings obtained in alkaline media, where the formation of the hydroxyl radicals is very likely from the ozone degradation [10], revealed that the oxidative degradation reaction promoted by the indirect oxidation pathway is characterized by higher values of _k_ap (see Table 2). Analysis of Table 2 also revealed that the _E_EO values varied covering the wide range of 26 kW h m-3 order-1 < _E_EO < 207 kW h m-3 order-1. It was verified that the lowest energy consumption was obtained for the ozonation process carried out in alkaline solutions. The _E_EO values obtained in the present work are in good agreement with those reported in the literature. For instance, Mehrjouei et al. [55] reported _E_EO values covering the wide interval of 12 kW ;h m-3 order-1 < _E_EO < 500 kW h m-3 order-1.
+
+As will be shown in the following section of this work, all data shown above, which were based on the spectrophotometric analysis, are in good agreement with data obtained from the HPLC technique.
+
+### Reduction of COD during the ozonation process and the HPLC analysis
+
+COD measurements are necessary in order to evaluate the degree of degradation of the dissolved organic matter (e.g., paracetamol and its oxidation by-products) as a function of the ozonation time, since the spectrophotometric analysis alone might not be sufficient to accomplish this end. However, when the absorbance at 243 nm is strongly reduced this may be an indication that the organic matter is being degraded [56].
+
+COD is a parameter used to evaluate the amount of oxygen necessary to convert the organic matter in mineral substances in the presence of a powerful oxidizing agent [57]. Thus, a reduction in COD values verified as a function of the ozonation time can be associated with the degree of degradation of the dissolved organic matter.
+
+As previously discussed by Thomas et al. [58], the absorbance measured in the UV region (e.g., 254 nm) and TOC (or even COD) measurements can yield complementary information about the investigated system. In fact, the UV measurements at 254 nm are indicative of a number of total aromatic compounds present in water as a function of the ozonation time, while the reduction in COD values during the ozonation indicates an increase in the susceptibility of the dissolved organic matter to chemical oxidation with formation of mineral substances.
+
+Figure 10 shows the reduction of COD as a function of the ozonation time obtained for different paracetamol concentrations and pHs.
+
+As can be seen, there was a strong reduction in COD values mainly for the first 30 min of ozonation. In addition, the rate of COD reduction was greatly affected by the pH, especially at higher concentrations of paracetamol. This is due to the fact that when molecular ozone enters in contact with the hydroxyl anion present in aqueous solutions its catalyzed degradation rapidly initiates leading to the formation of several radical species including the hydroxyl radical [10]. Since this radical exhibits a higher oxidizing power (_E_o = 2.80 V) when compared to molecular ozone (_E_o = 2.07 V), the degradation of the
+
+Fig. 10: COD reduction as a function of the pH for different ozonation times: **a** 20 mg dm-3; **b** 30 mg dm-3, and **c** 50 mg dm-3. Conditions: \(\nu_{\text{EOP}}\) = 0.5 g h−1 and \(T\) = 24 °Cdissolved organic matter can be more pronounced in alkaline conditions.
+
+Also, Fig. 10 revealed a greater reduction in COD values obtained for the ozonation carried out without the pH adjustment (e.g., natural, pH 6.3) when compared to the case involving the acidic solution (pH 2). These findings were not verified in the previous study of the current work where the ozonated samples were analyzed based only on the spectro-photometric analysis. According to Andreozzi et al. [6], the ozone consumption in almost neutral media (pH ~ 7) is higher than in acidic solutions since in this case there is the formation of the hydrogen peroxide, which in turn can result in the formation of the hydroxyl radicals.
+
+Figure 11 shows the linear correlation of COD and HPLC with UV243 obtained for the ozonation of paracetamol.
+
+As seen, the COD values exhibited a good linear correlation with the absorbance values obtained from the spectrophotometric analysis conducted at 243 nm (e.g., UV243). Similar correlations were previously reported by Mrkva [56]. In addition, it was also possible to verify a very good linear correlation between the HPLC response (e.g., the UV detection at 243 nm) and UV243 values obtained from spectrophotometry. In principle, these findings reveal the existence of a complementarity between the different experimental techniques used to characterize the ozonated samples.
+
+## NMR characterization of ozonated samples
+
+NMR spectroscopy was used for structural analysis of paracetamol (parental substance) and its oxidation by-products formed during the ozonation process. For this particular study, a high amount of the oxidation by-products was obtained after the ozonation of 250 mL of a concentrated paracetamol solution (250 mg dm-3 and pH 10) carried out over a period of 240 min (_v_EOP = 0.5 g h-1). After that, the ozonated sample was freeze-dried and solubilized in a mixture of H2O/D2O (95:5, _v/v_) for 1H, decoupled 13C, and 13C DEPT-135 unidimensional experiments.
+
+Unequivocal 1H resonance assignments (see Table 3) of paracetamol can be performed based on the standard chemical shifts observed in the literature [59]. The intense singlet at 1.94 ppm can be assigned to the methyl protons and the two duplets about 6.69 and 7.03 ppm were related to the aromatic proton. In addition, a chemical shift of 9.42 ppm observed in the 1H spectra of paracetamol is in accordance with amidic proton resonance. 13C spectrum presented all characteristic peaks of the paracetamol structure. However, several differences could be noticed in the 1H and 13C spectra obtained for the oxidation by-products. Whereas the aromatic peaks were not observed, the amidic proton is in a lesser chemical shift when compared to the 1H paracetamol spectrum. Clearly, the different chemical environment of amidic hydrogen, due to the absence of local anisotropic effect from the aromatic ring, indicates a C-N cleavage between acetamide group and phenol followed by degradation of the aromatic ring. Moreover, no aromatic chemical shifts could be observed in both decoupled 13C and DEPT-135 spectra. At the same time, an intense and no hydrogenated peak was verified at 162.69 ppm. On the other hand, no significant changes were noticed to the carbonyl (173.09 ppm) and methyl (23.23 ppm) carbons of acetamide group. In general, these findings indicate that the major part of phenol's ring was oxidized to CO32-[60] while no reaction occurs in the acetamide group of paracetamol during the ozonation reaction.
+
+## Conclusions
+
+The EOP accomplished in electrolyte-free water using the three-dimensional lead dioxide anode electrochemically deposited onto the porous carbon cloth substrate (CC/_b_-PbO2) showed promising findings, since using a volumetric water flow rate (_Q_) of 23.6 cm3 s-1 (_T_ = 24 degC), it was possible to obtain a current efficiency for EOP of up to 9.5% with an ozone production rate of up to 1.40 g h-1 (_I_ = 50 A).
+
+\begin{table}
+\begin{tabular}{c c c c c} Nucleus/group & Paracetamol & & Oxidation by-products \\
+1H (ppm) & 13C (ppm) & 1H (ppm) & 13C (ppm) \\ Methyl & 1.94 (s) & 22.41 & 1.91 & 23.23 \\ Amidic & 9.42 (s) & – & 8.44 & – \\ Aromatic & 6.69 (d) & 115.64 & – & – \\  & 7.03 (d) & 124.65 & & \\  & & 129.25 & & \\  & & 153.30 & & \\ Carbonyl & – & 172.99 & – & 173.09 \\ CO32- & – & – & – & 162.69 \\ \end{tabular}
+\end{table}
+Table 3: 1H and 13C chemical shifts of paracetamol and its oxidation by-products obtained at 20 °C in a H2O/D2O (95:5, v/v) mixture (300 MHz)
+
+Fig. 11: Correlation of COD and HPLC with UV243 as a function of the ozonation time. Conditions: _ν_EOP = 0.5 g h−1 and \(T\) = 24 °CUsing a fixed ozone production rate of 0.5 g h-1 supplied to the bubble column reactor, the ozonation of aqueous solutions containing paracetamol was accomplished. This study clearly revealed that the degradation of the dissolved organic matter (e.g., paracetamol and its oxidation by-products) occurs more efficiently in alkaline media due to the important electrophilic attack propitiated by the hydroxyl radicals. For the best experimental conditions, a degradation degree of 80% was obtained.
+
+The ozonated samples were analyzed using different experimental techniques (UV243, COD, and HPLC). It was verified good linear correlations between the experimental findings obtained using these different experimental techniques as a function of the ozonation time. Therefore, a complementarity exists between the different techniques used to characterize the ozonated samples.
+
+NMR spectroscopy was used to characterize the oxidation by-products obtained after the ozonation of a concentrated paracetamol solution (_V_ = 250 mL; [paracetamol] = 250 mg dm-3; pH 10; \(t\) = 240 min, and _v_x_OP = 0.5 g h-1). A thorough analysis was provided in this study to elucidate the major chemical events that can occur after the oxidative degradation of paracetamol. It was verified that the major part of phenol's ring was oxidized to CO32- while no reaction occurs in the acetamide group of paracetamol during the ozonation reaction.
+
+## Funding information
+
+L.M. Da Silva wishes to thank the "Fundacao ao Amparo a Pesquisa do Estado de Minas Gerais-FAPEMIG" (Projects CEX-APO-1181-14 and CEX-112-10), "Secretaria de Estado de Ciencia, Tecnologia e Ensino Superior de Minas Gerais-SECTES/MG" (Support for the LMM LA Laboratory), and "Conselho Nacional de Desenvolvimento Cientifico e Tecnologico--CNPq" (PQ-2 grant). This work is a collaborative research project of members of the "Rede Mineira de Quimica" (RQ-MG) supported by FAPEMIG (Project: CEX - RED-00010-14).
+
+## References
+
+* [1] Du Silva LM, Franco DV, Goncalves IC, De Sousa LG (2009) In: Gersten N, Sonderby L (ed) Water purification. Nova Science, New York
+* [2] Correia B, Freitas R, Figueira E, Soares AMVM, Nunes B (2016) Oxidative effects of the pharmaceutical drug paracetamol on the edible clam _Radiulapes philipipharum_ under different salinities. Comp Biochem Physiol Part C: Toxicol Pharmacol 179:116-124
+* [3] Mameri Y, Debbachu N, Benacherine MEM, Scarghain N, Schili T (2016) Heterogeneous photodegradation of paracetamol using goethite/H2O2 and goethite/oxalic acid systems under artificial and natural light. J Photochem Photobiol A Chem 315:129-137
+* [4] Yang L, LE Y, Ray MB (2008) Degradation of paracetamol in aqueous solutions by TiO2 photocatalysts. Water Res 42:3480-3488
+* [5] Arredondo-Valdez HC, Garcia-Jimenez G, Gutierrez-Granados S, de Leon PC (2012) Degradation of paracetamol by advance oxidation processes using modified retiated vitreous carbon electrodes with TiO2 and CO4TiO2/Al2O3. Chemosphere 89:1195-1201
+* [6] Andreuzzi R, Caprio V, Marotta R, Vogna D (2003) Paracetamol oxidation from aqueous solutions by means of ozonation and H2O2/UV system. Water Res 37:993-1004
+* [7] Hamdi El Najjar N, Touffet A, Deborde M, Journel R, Leitner KV (2014) Kinetics of paracetamol oxidation by ozone and hydroxyl radicals, formation of transformation products and toxicity. Sep Purif Technol 136:137-143
+* [8] Skoumal M, Cabot PL, Centellas F, Arias C, Rodriguez RM, Garrido JA, Brillas E (2006) Mineralization of paracetamol by ozonation catalyzed with Fe2+, Cu2+ and UVA light. Appl Catal B Environ 66:228-240
+* [9] Da Silva LM, Franco DV, De Sousa LG, Goncalves IC (2010) Characterization of an electrochemical reactor for the ozone production in electrolyte-free water. J Appl Electrochem 40:855-864
+* [10] Da Silva LM, Jardim WF (2006) Trends and strategies of ozone application in environmental problems. Quim Nova 29:310-317
+* [11] Babak AA, Amadelli R, De Battisti A, Fateev VN (1994) Influence of anions on oxygen/ozone evolution on PbO3/spe and PbO3/Ti electrodes in neutral pH media. Electrochem Acta 39:1597-1602
+* [12] Stucki S, Baumann H, Christen HJ, Kotz R (1987) Performance of a pressurized electrochemical ozone generator. J Appl Electrochem 17:773-778
+* [13] Velichenko AB, Girenko DV, Kovalyov SV, Gnatenko AN, Amadelli R, Danilov FI (1998) Lead dioxide electrodeposition and its application: influence of fluoride and iron ions. J Electroanal Chem 454:203-208
+* [14] Li X, Pletcher D, Walsh FC (2011) Electrodeposited lead dioxide coatings. Chem Soc Rev 40:3879-3894
+* [15] Ho JKX, Tremiliois-Filho G, Simprga R, Conway BE (1994) Structure influence on electron catalysis and adsorption of intermediates in the anodic O2 evolution at dimorphic \(\alpha\)- and \(\beta\)-PbO2. J Electroanal Chem 366:147-162
+* [16] Sang-Do H, Kim JD, Myung KS, Rana RK, Singh KC (2006) Electrochemical production of ozone using water electrolysis cell of solid polymer electrolyte (SPE). Indian J Chem Technol 13:156-161
+* [17] Foller PC, Tobias CW (1982) The anodic evolution of ozone. J Electrochem Soc 129:506-515
+* [18] Trasatti S (1981) Electrodes of conductive metallic oxides. Elsevier, Amsterdam
+* [19] Abaci S, Pekmez K, Yildiz A (2005) The influence of nonstoichiometry on the electrocatalytic activity of PbO2 for oxygen evolution in acidic media. Electrochem Commun 7:328-332
+* [20] Devillers D, Dinh-Thi MT, Mahe E, Dauriex V, Lequeux N (2004) Electroanalytical investigations on electrodeposited lead dioxide. J Electroanal Chem 573:227-239
+* [21] Andrade LS, Ruotolo LAM, Rocha-Filho RC, Bocchi N, Biaggio SR, Iniesta J, Garcia-Garcia V, Montil V (2007) On the performance of Fe and Fe, F, doped Ti-Pt/PbO2 electrodes in the electrooxidation of the blue reactive 19 dye in simulated textile wastewater. Chemosphere 66:2035-2043
+* [22] Awad MI, Saleh MM (2010) Electrochemical generation of ozone at PbO2-loaded platinum screens. J Solid State Electrochem 14: 1877-1883
+* [23] Chen B, Guo Z, Huang H, Yang X, Cao Y (2009) Effect of the current density on electodepositing alpha-lead dioxide coating on aluminum substrate. Acta Metall Sin 22:373-382
+* [24] Costa FR, Da Silva LM (2012) Fabrication and characterization of a porous gas-evolving anode constituted of lead dioxide microfibers electroformed on a carbon cloth substrate. Electrochim Acta 70: 365-374* [25] De Sousa LG, Franco DV, Da Silva LM (2016) Electrochemical ozone production using electrolyte-free water for environmental applications. J Environ Chem Eng 4:418-427
+* [26] Mahalingam T, Velumani S, Raja M, Thankaikarasan S, Chu JP, Wang SF, Kim YD (2007) Electrosynthesis and characterization of lead oxide thin films. Mater Charact 58:817-822
+* [27] Shen PK, Wei XL (2003) Morphologic study of electrochemically formed lead dioxide. Electroph Actin Acta 48:1743-1747
+* [28] Yeo IH, Kim S, Jacobson R, Johnson DC (1989) Electrocatalysis of anodic oxygen transfer reactions: comparison of structural data with electrocatalytic phenomena for bismuth-doped lead dioxide. J Electrochem Soc 136:1395-1401
+* [29] Zhou M, Dai Q, Lei L, Ma CA, Wang D (2005) Long life modified lead dioxide anode for organic wastewater treatment: electrochemical characteristics and degradation mechanism. Environ Sci Technol 39:363-370
+* [30] Amadelli R, Velichenko AB (2001) Lead dioxide electrodes for high potential anodic processes. J Serb Chem Soc 66:835-845
+* [31] Rice EW et al. (ed), (2012) Standard methods for the examination of water and wastewater, 22 ed. American Public Health Association, New York
+* [32] Resende JM, Mornes CM, Prates MV, Cesar A, Almeida FCL, Mundim N, Valente AP, Bemquerer MP, Pilo-Veloso D, Bechinger B (2008) Solution NMR structures of antimicrobial peptides phyllosepin-1,2, and3- and biological activity: the role of charges and hydrogen bonding interactions in stabilizing helix conformations. Peptides 29:1633-1644
+* [33] Amadelli R, Maldotti A, Molinari A, Danilov FI, Velichenko AB (2002) Influence of the electrode history and effects of the electrolyte composition and temperature on O2 evolution at _b_-PbO2 anodes in acid media. J Electroanal Chem 534:1-12
+* [34] Mohd Y, Pletcher D (2005) The influence of deposition conditions and dopant ions on the structure, activity, and stability of lead dioxide anode coatings. J Electrochem Soc 152:D97-D102
+* [35] Velichenko AB, Amadelli R, Benedetti A, Gireink DV, Kovalyov SV, Danilov FI (2002) Electrophoresis and physicochemical properties of PbO2 films. J Electrochem Soc 149:C445-C449
+* [36] Mikrajuddin A, Khairurji K (2008) Derivation of Scherrer relation using an approach in basic physics course. Jurnal Nanosains & Nanotechnologie 128:32
+* [37] Bockris JOM (1956) Kinetics of activation controlled consecutive electrochemical reactions: anodic evolution of oxygen. J Chem Phys 24:817-827
+* [38] Vetter KJ (1961) Elektrochemische kinetik. Springer-Verlag, Berlin
+* [39] Da Silva LM, De Faria LA, Boods JFC (2003) Electrochemical ozone production: influence of the supporting electrolyte on kinetics and current efficiency. Electrochim Acta 48:699-709
+* [40] Da Silva LM, De Faria LA, Boods JFC (2001) Green processes for environmental application. Electrochemical ozone production. Pure Appl Chem 73:1871-1884
+* [41] da Silva LA, Alves VA, da Silva MAP, Trasatti S, Boods JFC (1997) Morphological, chemical, and electrochemical properties of Ti(TiO2 + HO2) electrodes. Can J Chem 75:1483-1493
+* [42] De Faria LA, Boods JFC, Trasatti S (1996) Electrocatalytic properties of ternary oxide mixtures of composition Ru03,Ti(0.7x)/Ce0O2: oxygen evolution from acidic solution. J Appl Electrochem 26:1195-1199
+* [43] Da Silva LM, Boods JFC, De Faria LA (2001) Oxygen evolution at Ru0O2(r)+Co0O4(1-x) electrodes from acid solution. Electrochim Acta 46:1369-1375
+* [44] Kotz ER, Stucki S (1987) Ozone and oxygen evolution on PbO2 electrodes in acid solution. J Electroanal Chem 228:407-415
+* [45] Gileadi E (2011) Physical electrochemistry: fundamentals, techniques and applications. Wiley-VCH, Weinheim.
+* [46] Ho CN, Hwang BJ (1994) Effect of hydrophobicity on the hydrophobic-modified polytetrafluoroethylene/PbO2 electrode towards oxygen evolution. J Electroanal Chem 377:177-190
+* [47] Da Silva LM, Franco DV, Forti JC, Jardim WF, Boods JFC (2006) Characterisation of a laboratory electrochemical ozonation system and its application in advanced oxidation processes. J Appl Electrochem 36:523-530
+* [48] Foller PC, Kelsall GH (1993) Ozone generation via the electrolysis of fluohere acid using glassy carbon anodes and air depolarized cathodes. J Appl Electrochem 23:996-1010
+* [49] Nishiki Y, Kitaori N, Nakamuro K (2011) Performances of small-sized generator of ozone-dissolved water using boron-doped diamond electrodes. Ozone Sci Eng 33:114-120
+* [50] Feng J, Johnson DC, Lowery SN, Carey JJ (1994) Electrocatalysis of anodic oxygen-transfer reactions: evolution of ozone. J Electrochem Soc 141:2708-2711
+* [51] Huber MM, Canonica S, Park GY, Von Gunten U (2003) Oxidation of pharmaceuticals during ozonation and advanced oxidation processes. Environ Sci Technol 37:1016-1024
+* [52] Franco DV, Jardim WF, Boods JFC, Da Silva LM (2008) Electrochemical ozone production as an environmentally friendly technology for water treatment. Clean 36:34-44
+* [53] Bolton JR, Bircher KG, Tumas W, Tolman CA (2001) Figures-of-mert for the technical development and application of advanced oxidation technologies for both electric- and solar-driven systems. Pure Appl Chem 73:627-637
+* [54] Santana MHP, Da Silva LM, Freitas AC, Boods JFC, Fernandes KC, De Faria LA (2009) Application of electrochemically generated ozone to the discoloration and degradation of solutions containing the dye Reactive Orange 122. J Hazard Mater 164:10-17
+* [55] Mehripoi M, Muller S, Moller D (2014) Catalytic and photovoltaic ozonation of tert-butyl alcohol in water by means of falling film reactor: kinetic and cost-effectiveness study. Chem Eng J 248:184-190
+* [56] Mitava M (1983) Evaluation of correlations between absorbance at 254 nm and COD of river waters. Water Res 17:231-235
+* [57] Von Sperling M (1996) Introducao a qualidade das aguas e ao tratamento de esgetos. Editora UFMG, Belo Horizonte
+* [58] Thomas O, El Khorsassani H, Touraud E, Bitar H (1999) TOC versus UV spectrophotometry for wastewater quality monitoring. Talanta 50:743-749
+* [59] Wishart DS, Knox C, Guo AC, Eisner R, Young N, Gautam B, Hau DD, Psychogios N, Dong E, Bouatta S et al (2009) HMDB: a knowledgebase for the human metabolome. Nucleic Acids Res 37:D603-D610
+* [60] Moret S, Dyson PJ, Laurenczy G (2013) Direct, in situ determination of pH and solute concentrations in formic acid dehydrogenation and CO2 hydrogenation in pressurised aqueous solutions using 1H and 13C NMR spectroscopy. Dalton Trans 42:4353-4356

manual_annotation/200pdfs_in_mmd/10.1007_s10098-018-1595-2.mmd

@@ -0,0 +1,166 @@
+Degradation of bisphenol A by combining ozone with UV and H2O2 in aqueous solutions: mechanism and optimization
+
+Ze Liu, Niels Wardenier, Seyedahmad Hosseinzadeh, Yannick Verheust, Pieter-Jan De Buyck, Michael Chys, Anton Nikiforov, Christophe Leys, Stijn Van Hulle
+
+# Abstract
+
+A laboratory-scale study on the abatement of bisphenol A (BPA) was performed by combining O3 with H2O2 and UV (O3/H2O2/UV), an ozone-based advanced oxidation processes technique (AOP). This work aimed to (1) evaluate the removal of BPA with O3/H2O2/UV, and to compare the degradation efficiency with other ozone-based AOPs (such as O3 alone, O3/H2O2, and O3/UV), (2) structurally optimize BPA abatement by using a central composite design (CCD) for experimental design purposes and/or a response surface methodology to find the optimum, and (3) identify the degradation pathways, and main intermediate products, formed during BPA abatement with O3/H2O2/UV. The degradation pathways of BPA degradation were revealed by O3/H2O2/UV on the basis of evidences of intermediate generation. The effect of initial pH, ozone, and H2O2 dose during BPA abatement was studied in detail. By increasing each of these three parameters, an enhancement of the BPA degradation efficiency is mostly observed. BPA can be degraded completely when a sufficiently high ozone dose is applied. However, excess H2O2, as a scavenger of HO-, has a negative effect on BPA abatement, resulting in a decrease in the BPA's degradation efficiency. For example, the removal decreased from 64 to 58% by enhancing the H2O2 initial dose from 0.5 to 0.75 mmol/L (at an initial pH and ozone dose of, respectively, 7 and 0.1 mg/L). The results confirmed that combining ozone with H2O2 and UV was a more efficient method than the other three ozone-based AOPs on the removal of BPA. Therefore, this method could be further applied for the treatment of real wastewaters containing BPA and other micropollutants.
+
+## Introduction
+
+Bisphenol A (BPA) is a derivative of phenol and acetone. BPA, as a plastic monomer and plasticizer, has been widely applied, such as in the production of baby bottles, safety glasses, and food packaging (Vandenberg et al. 2007). Further, it has also been used in the production of plasticizers, fire retardants, antioxidants, pesticides, and other fine chemicals (Jia et al. 2012). Concerns have been growing on the wide presence of BPA in nature, specifically due to its endocrine disrupting characteristics (Pirila et al. 2015). The United States Environmental Protection Agency (EPA) and World Wildlife Fund International have characterized BPA as one of the representative endocrine disrupting chemicals (EDCs) (Huang et al. 2012). Exposure towards embryos, neonates, and toddlers has raised concerns of the USFDA (US Food and Drug Administration) (USFDA 2010).
+
+There are a number of routes for BPA to be discharged into the aquatic environment, originating from discharges of wastewater from facilities related with BPA production, discharges of (treated) industrial and municipal wastewater, or from processing and storage facilities (Subedi et al. 2013; Lee et al. 2014; Chou et al. 2014). As such, BPA abatement is essential to obtain a cleaner and more sustainable industrial production.
+
+Recently, several methods, including biological, physical, and chemical techniques, have been applied on BPA removal from (waste)water. For example, activated carbon can adsorb BPA. However, due to BPA's low log _K_ow, adsorption on the activated carbon surface is limited (Choi et al. 2005). Additionally, activated carbon needs frequent replacement and regeneration to keep BPA removal efficiently, which increases the cost of treatment. It has been proved that activated carbon is (mostly) a more expensive method compared to ozonation (Chys et al. 2017). Biodegradation methods on the other hand usually exhibit slow BPA degradation, and its efficiency depends on many environmental factors (Bhadra et al. 2018). In particular, advanced oxidation processes (AOPs) have been demonstrated to eliminate DOC (dissolved organic compounds) from different (waste)water matrices (Matiainen and Sillanpaa 2010; Yang et al. 2014; Oturan and Aaron 2014). For example, Ohko et al. (2001) could completely mineralize 170 mM BPA using UV/TiO2, although taking up more than 5 h. Sharma et al. (2016) also reported that UV-based AOPs, such as UV-C/SPS, could remove BPA (0.22 mmol/L) more than 90%, but needed 6 h of reaction time. Moussavi et al. (2018) degraded BPA with UV/H2O2 efficiently and completely in 8 min; however, the excess amount of H2O2 used in the study (0-500 mg/L) might result in the re-scavenging of OH radicals by H2O2 and consequently result in a lower energy efficiency (and thus higher consumption). Furthermore, Muller et al. (2001) and Lester et al. (2011) have reported, respectively, O3/H2O2 showed lower energy requirement on micropollutants degradation in comparison with UV/H2O2. Accordingly, it is of great interest to develop faster and more cost-efficient AOPs. Combining ozone with UV and H2O2 (O3/H2O2/UV), one kind of ozone-based AOP methods, shows a strange ability on recalcitrant organic compounds' degradation (Arslan et al. 2017) and can be used as a further degradation method of micropollutants. Kusic et al. (2006) and Wu et al. (2008) have reported the advantage of O3/H2O2/UV for the degradation of phenol and 2-propanol. Moreover, pharmaceuticals, such as ciprofloxacin and trimethoprim, have been reported to be efficiently eliminated (Lester et al. 2011), while also Mitra et al. (2016) applied O3/H2O2/UV for the mineralization of organic compounds. It was indicated that this method showed a higher efficiency on organic compounds degradation comparing to O3 only, O3/H2O2, and UV/H2O2.
+
+When comparing with O3/H2O2/UV for example (Yu et al. 2016), it can be seen that 89.9% removal efficiency is obtained for 1 mg/L BPA by applying the photo-iron(III)-sulphite system (Fe2(SO4)3 = 0.05 mmol/L, Na2SO3 = 0.2-3 mmol/L, initial pH = 6). Although this system also has a strong capacity for degrading BPA, the release of a significant amount of iron sludge becomes an additional environment problem. Furthermore, BPA (39.9 mg/L) can be degraded completely in 1 h by the Fenton reagent (the initial concentration of Fe2+ was 0.692 mmol/L) (Young et al. 2013). BPA (11.4 mg/L) can be degraded by UV/H2O2 in less than 10 min, but only when 2 mmol/L H2O2 is applied, the efficiency of degradation can reach more than 80% (Yoon et al. 2012). However, the initial concentration of H2O2 applied for Fenton reagent and UV/H2O2 is more than 1.0 and 2 mmol/L, respectively, leading to the cost increase. This result is also consistent to the conclusion that O3/H2O2/UV is an efficient method for organic compounds degradation.
+
+However, information on the application potential of the O3/H2O2/UV process, the degradation mechanism, and the optimization of BPA removal from (waste)water is still lacking. In this study, the applicability of the O3/H2O2/UV process to degrade BPA was evaluated. A comparison between the BPA removal efficiency with the O3/H2O2/UV process and three other ozone-based AOPs, including O3 alone, O3/H2O2, and O3/UV, was made. The impact of different (initial) pH, ozone and H2O2 concentrations, and reaction time on the removal rate of BPA applying O3/H2O2/UV was investigated. In addition, the optimal settings for O3/H2O2/UV were evaluated, and the main intermediate products qualitatively detected. Furthermore, in this study, a proposition of the BPA abatement pathways is undertaken applying the O3/H2O2/UV method.
+
+## Experimental methods and material
+
+### Standards and reagents
+
+All chemicals which were used in all experiments were of analytical grade (purity of >= 98%, Sigma-Aldrich, Belgium). Hydrogen peroxide (H2O2), with a concentration of 30%, was used in the experiments. All stock solutions were prepared in demineralized water. Adjustment of pH of reaction solution was achieved by adding NaOH and sodium phosphate buffer (Pikal-Cleland et al. 2002).
+
+### BPA degradation experiments
+
+Ozone was generated by an ozone generator (COM-AD-02, Anseros) with a flow of 300 mL/min pure oxygen (Chys et al. 2015). A concentrated ozone stock solution was prepared by a subsequent bubbling process through demineralized water. According to the procedure of Bader and Hoigne (1981), the exact concentration of the ozone stock solution was determined by the indigo method.
+
+Four ozone experimental set-ups were employed. BPA solutions (200 mg/L) were dosed to 500 mL batch reactors. Experiments with different ozone doses (0.1, 0.25, 0.5, 0.75 and 1 mg/L) were performed. The applied ozone dosage was controlled through adding a known concentration of the ozone stock solution (Liu et al. 2018). For pH adjustments,a phosphate buffer (25 mmol/L) and 1 mol/L NaOH were added to each reactor (Im et al. 2012). Ozone stock solution (50 mg/L) was added to the reactors obtaining concentrations up to 0.1, 0.25, 0.5, 0,75 and 1 mg/L. Taking into account the reaction rates with ozone (1.7x104 M-1 s-1) (Deborde et al. 2005) and HO- (1.6x109 M-1 s-1) (Umar et al. 2013), samples were taken at adapted time intervals (0, 5, 10, 15, 20, 30, and 40 min) after dosing ozone. For ozone/H2O2 experiments, H2O2 was added before O3 addition with an initial concentration of 0.25, 0.5, 0.75, and 1 mmol/L.
+
+For ozone/UV, a circulating reactor system was used to evaluate its performance of BPA degradation, including a photoreceptor (Liu et al. 2018). The mixed solution was continuously recirculated by means of a peristaltic pump (800 mL min-1). For ozone/H2O2/UV experiments, H2O2 was added to the O3/UV circulating reactor system with an initial concentration of 0.25, 0.5, 0.75, and 1 mmol/L, respectively, before O3 addition (0.1, 0.25, 0.5, 0,75, and 1 mg/L).
+
+## Analytical methods
+
+The pH (Metrohm 600 pH meter) was determined of the original solution without treatment and during reaction. GC-MS analyses were done with an Agilent 6890 GC Series gas chromatograph coupled to a HP 5973 mass selective detector. A detailed description of this method is given by Liu et al. (2018).
+
+## Experimental design
+
+The optimal degradation conditions in this study were determined by applying a central composite design (CCD) with four factors and five levels (Bezerra et al. 2008). The response (_Y_) reflects the BPA removal rate. The factors (\(X_{i}\)), and their ranges and levels, are summarized in Table 1.
+
+The relationship between the response and the different factors is described by Eq. (1) below, and its coefficients were determined by Matlab:
+
+\[\begin{array}{l} {Y = \alpha_{0} + \alpha_{1} X_{1} + \alpha_{2} X_{2} + \alpha_{3} X_{3} + \alpha_{4} X_{4} + \alpha_{12} X_{1} X_{2} + \alpha_{13} X_{1} X_{3} + \alpha_{14} X_{1} X_{4} + \alpha_{23} X_{2} X_{3} + \alpha_{24} X_{2} X_{4} + \alpha_{34} X_{3} X_{4} + \alpha_{11} X_{1}^{2} + \alpha_{22} X_{2}^{2} + \alpha_{33} X_{3}^{2} + \alpha_{44} X_{4}^{2} + \end{array}\]
+
+## Results and discussion
+
+### Degradation efficiency of BPA by ozone-based AOPs
+
+An ozone dose of 0.1 mg/L and H2O2 doses of 0.25 and 0.5 mmol/L were chosen to treat 200 mg/L BPA to evaluate the performance of O3/H2O2/UV and compare it with other three different ozone-based AOPs in this part of the work.
+
+According to the results shown in Fig. 1, the degradation efficiency of BPA when combining H2O2 or UV with ozone is higher than that treated by only ozone. For similar ozone concentrations, the degradation efficiency with 0.5 mmol/L H2O2 in solution is higher than that for adding 0.25 mmol/L H2O2. These findings confirm that BPA degradation is mainly based on hydroxyl radical (HO-) reactions and degradation efficiency can be increased by increasing the HO- production efficiency (Guo et al. 2010). Figure 1 shows that for equal ozone concentrations, the degradation rate of BPA treated by O3/H2O2/UV (> 97% removal) is much higher than the degradation rate of BPA treated by the other three kinds of ozone-based AOPs methods. BPA is degraded completely (> 99% removal) after treated by O3/H2O2/UV with 0.1 mg/L O3 in only 20 min; however, the removal rate of BPA with other ozone-based AOPs is all less than 65%. It illustrates that combining O3 with H2O2 and UV can significantly improve the generation of HO-, and it is a more efficient
+
+\begin{table}
+\begin{tabular}{c c c c c c} Independent variables & Factor & Range and level \\ _Xi_ & −2 & −1 & 0 & 1 & 2 \\ Ozone concentration (mg/L) & \(X\)1 & 0.1 & 0.25 & 0.5 & 0.75 & 1 \\ H2O2 concentration (mmol/L) & \(X\)2 & 0 & 0.25 & 0.5 & 0.75 & 1 \\ Initial pH & \(X\)3 & 3.0 & 5.0 & 7.0 & 9.0 & 11.0 \\ Reaction time (min) & \(X\)4 & 0 & 5 & 10 & 15 & 20 \\ \end{tabular}
+\end{table}
+Table 1: Ranges and levels of independent variables of CCD
+
+Fig. 1: Removal efficiency of BPA (200 mg/L) for O3 only, O3/H2O2, O3/UV and O3/H2O2/UV treatments with 0.1 mg/L O3, and 0.25 and 0.5 mmol/L hydrogen peroxide at initial pH ×7 after 20 min method for BPA abatement in comparison with the three other ozone-based AOPs.
+
+### Efficiency of BPA abatement with applied O3/H2O2/ UV
+
+#### Effect of O3 concentration on BPA abatement
+
+In order to investigate the influence of different O3 concentrations on the BPA degradation with ozone alone, 500 mL of a 200 mg/L BPA solution was subjected to different ozone doses ranging between 0.1 and 1.0 mg/L (Fig. 2a). Kinetic studies were also performed in this part (shown in Fig. 2b).
+
+According to Fig. 2a, enhanced BPA removal could be noticed by increasing the O3 dosage. For instance, an O3 dose up to 0.1 mg/L resulted only 33% decrease in BPA, while almost completely eliminated after 20 min of treatment applying 1 mg/L O3. An enhanced reaction time beyond these 20 min did not result in a (noticeable) increase in BPA degradation (Fig. 2a). Most likely, all O3 was reacted after this time as dissolved O3 measurements did not show any residual after 20 min.
+
+Compared with single O3, BPA was also completely removed by UV/O3/H2O2 when using 0.1 mg/L O3 and 0.25 mmol/L H2O2 (see Fig. 1). It can be said also that the BPA removal efficiency using 0.1 mg/L O3 with 0.5 mmol/L H2O2 was even higher than during single ozonation with an O3 dose ranging from 0.1 to 0.5 mg/L (Fig. 2a). Apparently, this demonstrates that the combination of O3 with H2O2 and UV can significantly improve the BPA removal efficiency, while reducing the O3 consumption for ozone-based AOPs.
+
+The removal of BPA by ozonation generally follows a pseudo-first-order kinetic reaction. Accordingly, the reaction rate constant (_k_, min-1) was determined by plotting the logarithm of the concentration as a function of time (Fig. 2b). Following this approach, the reaction rate constant is deducible from the slope of this curve. The \(R\)2 and \(k\)
+
+Fig. 2: **a** The degradation efficiency of BPA from water treated by ozone only method with different concentration of ozone (initial pH × 7), **b** BPA degradation kinetics during ozone only experiments with different ozone doses (initial pH × 7)values, presented in Table 2, showed that the reaction constant rate increased as the initial O3 concentration increased.
+
+The results on degradation efficiency of BPA from deionized water with four kinds of ozone-based AOPs and different concentration of O3 are given in Fig. 3. BPA abatement was sharply enhanced with an increase in O3 concentration during O3 only, O3/H2O2, and O3/UV. When an initial O3 concentration of 0.1 mg/L was used with O3 alone and O3/H2O2, the BPA removal rate was 33.1 and 40.6%, respectively. However, after adding 1.0 mg/L O3 in the solution, the removal rate for both methods was more than 99%. The efficiency of BPA degrading for O3/UV was enhanced from 57.6 to 99% by increasing the initial O3 dose 0.1 mg/L up to 0.75 mg/L. However, the removal of BPA while combing O3 with UV and H2O2 did not change much (ranged from 97.4 to 99.6%), and BPA was completely removed by O3/H2O2/UV, even at low O3 doses (0.1 mg/L).
+
+### Effect of H2O2 on BPA abatement
+
+To evaluate the role of H2O2 in the process of BPA degradation with AOPs, varying amounts of H2O2 were added to the solutions with O3 concentrations between 0.1 and 1.0 mg/L and an initial pH of 7. Increasing the H2O2 dose up to 0.5 mmol/L resulted in an enhanced BPA degradation efficiency (Fig. 4), indicating an accelerated generation of OH radical. Nevertheless, an H2O2 dose above 0.5 mmol/L resulted in an opposite behaviour as the efficiency clearly decreased. This indicates that the enhancement effect of high dosages of H2O2 was counteracted by its inhibiting effect. Excessive H2O2 scavenged HO- produced in the high H2O2 concentration solutions. Thus, while the presence of H2O2 in a lower concentration during ozonation enhances the production of OH radicals, they can consume OH radicals when using higher doses. As Fig. 4 shows, the BPA abatement at a concentration of H2O2 higher than 0.5 mmol/L could be enhanced when the ozone dosage was high enough (such as
+
+\begin{table}
+\begin{tabular}{c c c} Ozone concentration (mg/L) & Correlation coefficient (_R_2) & Reaction rate constant (_k_) \\
+0.1 & 0.92772 & 0.0125 \\
+0.25 & 0.91064 & 0.0192 \\
+0.5 & 0.91937 & 0.0295 \\
+0.75 & 0.92634 & 0.0544 \\
+1.0 & 0.97405 & 0.1850 \\ \end{tabular}
+\end{table}
+Table 2: Reaction parameters in degradation process at different concentrations of ozone
+
+Fig. 3: BPA abatement (200 mg/L) applying different ozone-based AOPs at different initial ozone doses (initial pH × 7)above 0.5 mg/L); however, it was still less than that at the concentration of H2O2 ranged 0-0.5 mmol/L. This results indicated that even though the scavenging effect of excessive H2O2 was counteracted by the stimulating effect of HO-generated when sufficient ozone is present in the solutions, the scavenging effect of excessive H2O2 still influenced the removal rate of BPA.
+
+BPA abatement for 0.25 mmol/L H2O2 (see Fig. 1) enhanced less than 2% (up to 98.8% removal) by means of increasing the initial concentration of H2O2 (i.e. 0.5 mmol/L H2O2). It can be explained that when applying O3/H2O2/UV in BPA abatement experiments, the scavenging effect of H2O2 concentration on BPA abatement is reduced. Appling O3/H2O2/UV with a low H2O2 concentration (such as 0.25 mmol/L) can almost completely remove BPA from deionized water which is beneficial to reduce costs and facilitate its application in real wastewater.
+
+## Effect of initial solution pH on BPA abatement
+
+A BPA solution was treated with 0.1 mg/L ozone and 0.25 mmol/L H2O2 at various pH levels (ranging 3.0-11.0) to investigate the influence of the initial pH on the degradation efficiency of BPA.
+
+Only 22% of BPA was degraded during the treatment of O3/H2O2 with 0.1 mg/L O3 and 0.25 mmol/L H2O2 at initial pH = 3 after 20 min; however, the removal rate was enhanced to 45% when initial pH was at 11 (Fig. 5). It indicated that the initial solution pH has an impact on the degradation efficiency of BPA. The initial solution pH value affects the decomposition of O3 and H2O2, which influences the generation of HO-. At high initial pH levels, more HO- are generated which results in BPA being degraded more effectively, which is in agreement with other research (e.g. Wang et al. (2018)). However, the degradation efficiency of BPA did not show a significant enhancement when the initial solution pH level was changed from 7.0 to 11.0 (Fig. 5). This could be attributed to the production of carboxylic acids that were generated during BPA degradation, which resulted in a pH decrease of the solutions during reaction, therefore influencing the BPA degradation.
+
+## Optimization of ozone-based AOP for the abatement of BPA
+
+To reveal the optimal conditions in order to maximize the BPA abatement, a central composite design (CCD) was applied, as explained in the'materials and methods' (Sect. 2.4). A total of 30 runs are collected and shown in Table S1. These results were developed into a mathematical equation [Eq. (2)] which showed the relationship between response \(Y\) and \(X\)1, \(X\)2, \(X\)3, and \(X\)4.
+
+\[\begin{array}{l}{Y=74.33+13.76X_{1}-7.78X_{2}+14.36X_{3}}\\ {\quad+10.59X_{4}+0.086X_{1}X_{2}-0.48X_{1}X_{3}+1.35X_{1}X_{4}}\\ {\quad-0.60X_{2}X_{3}-0.38X_{2}X_{4}-0.26X_{3}X_{4}-4.67X_{1}^{2}}\\ {\quad-10.85X_{2}^{2}-2.55X_{3}^{2}-8.14X_{4}^{2}}\end{array} \tag{2}\]
+
+Regression parameters of the quadratic model were determined based on an ANOVA analysis (analyse of variance) as summarized in Table S2. In this model, the regression is significant if the model \(F\) value is 3.46. Associated \(p\) values for \(X\)1, \(X\)2, \(X\)3, \(X\)4, \(X\)2, and \(X\)2 were less than 0.05 (Table S2),
+
+Fig. 5: Effect of initial pH values on BPA (200 μg/L) degradation efficiency the AOP with 0.1 mg/L ozone, 0.25 mmol/L H2O2 and the initial pH ranged 3.0–11.0which indicated that they were significant model terms. Thus, the ANOVA showed that all linear, quadratic parameters of H2O2 concentration and reaction time were significant, whereas the quadratic parameters of ozone concentration and initial pH, and all their interactions were found to be not significant. Additionally, to assess the interactive relationship between independent factors and the response, contour, and response surface plots of the versus interactions between BPA degradation efficiency and independent variables are utilized and shown in Fig. S1.
+
+This model indicated to fit the date very well (_R_2 = 0.93, \(F\) value (lack of fit) = 0.41 indicating no lack of fit). The adequate precision index, as indication for a suitable match of the predicted responses (i.e. if higher than four), obtained a value of 6.867 (Beg et al. 2003). Therefore, this model was applied to optimize these four experimental conditions within the range of the chosen variables.
+
+Based on the results obtained under the experimental design conditions, a prediction of maximal BPA removal efficiency within the range of the different factors was aimed. As such, the optimal conditions for maximum abatement (99%) of BPA were: O3 concentration (_X_1) = 0.75 mg/L, H2O2 concentration (_X_2) = 0.38 mmol/L, initial pH (_X_3) = 9.0, and reaction time (_X_4) = 26.5 min. A test of ozone-based AOP with the predicted optimal conditions was done to validate the developed CCD model towards its performance on BPA abatement. The result showed a 99% removal after treatment which is in close agreement with the prediction of this response surface model.
+
+### Identification of intermediate products and ozone-based degradation pathways
+
+Samples were taken every 15 s in the first 2 min to make sure that the intermediate products formed in the process of BPA degradation treated by O3/H2O2/UV in the solution can be detected by GC/MS. For complete degradation, BPA solutions (1 mg/L) were treated with 1.0 mg/L ozone and 0.5 mmol/L H2O2 during a reaction time of 60 min. Various by-products were formed during BPA degradation (as listed in Table 3) and indicated a stepwise process where BPA is gradually degraded into smaller molecules such as aromatic and benzoquinone intermediates, and organic acids. According to the different identified by-products, a potential BPA degradation pathway is shown in Fig. 6.
+
+It is known that organic compounds, based on their own specific affinity, react either directly with molecular ozone or indirectly with *OH. Ozone shows a selectivity towards
+
+\begin{table}
+\begin{tabular}{c c c c} Compound & Molecular weight (m/z) & Molecular structure & References \\ Bisphenol A (BPA) & 228 & & Clara et al. (2005) \\
+4-(1-Hydroxypropan-2-yl) & 152 & & Deborde et al. (2008) \\ catechol & & HO & Deborde et al. (2008) \\ _p_-Hydroquinone & 110 & & Poerschmann et al. (2010) \\
+4-Isopropenylphenol & 121 & & Katsumata et al. (2004) \\ Phenol & 94 & & Deborde et al. (2008) \\ _p_-Benzoquinone & 108 & & Poerschmann et al. (2010) \\ \end{tabular}
+\end{table}
+Table 3: Identification of the intermediates formed in the process of BPA degradation by AOP treatment electron rich moieties, while organic compounds unselectively react with HO- (Wang and Xu 2012). Generally, due to the attack of ozone on the activated aromatic structures and the presence of phenolic groups in the solutions (Deborde et al. 2005), BPA shows a high reactivity towards ozone (Kim et al. 2007). BPA degradation is therefore attributed by reaction with both molecular ozone and HO-.
+
+Based on the intermediate products determined by GC/MS (Table 3), hydroxylated derivatives are formed at the beginning of the degradation of BPA due to the attack of phenyl groups by HO- and molecular ozone in the initial oxidation process. One-ring hydroxylated phenolic derivatives are formed, as a result of isopropylidene bridge cleavage during reaction. Following this step, the formation of 4-isopropylphenol and phenol takes place in the cleavage of BPA. 4-(1-hydroxyppan-2-yl) catechol is generated by 4-isopropylphenol combined with hydroxyl, and then dehydrates, which results in the formation of 4-isopropylphenol. This in agreement with the findings reported in a previous study (Wang et al. 2018). Furthermore, since the groups attached to the aromatic ring were on the para position, they were easily oxidized by HO-4-(1-hydroxyppropan-2-yl) and 4-isopropylphenol are converted to _p_-hydroquinone and phenol. And _p_-benzoquinone is generated by _p_-hydroquinone. According to the results, the pH of the solutions decreased from 7.0 to around 5.0 after O3/H2O2/UV treatment, which indicated that carboxylic acids were formed during the degradation of BPA. Thus, intermediate products could be further degraded into organic acids, such as malonic acid, oxalic acid, and succinic acid. In addition, this mechanism was coincided with the research of Kondrakov et al. (2014) and Yang et al. (2016)
+
+## Conclusion
+
+The degradation of BPA was efficiently evaluated by combining O3 with H2O2 and UV. The ultimate combination of these three techniques (O3/H2O2/UV) showed to be most efficient. Furthermore, by increasing the initial concentrations of O3 and H2O2 during O3/H2O2/UV treatment, it was found that the efficiency of BPA abatement can be significantly enhanced. However, H2O2 will also scavenge HO- if dosed in excess. For example, the degradation efficiency is higher when 0.5 mmol/L H2O2 (BPA removal is 93%) is applied compared to 0.75 mmol/L H2O2 (BPA removal is 84%) under the same experimental conditions. Additionally, alkaline conditions seem to be beneficial to the degradation of BPA, as the results always show a high degradation efficiency of BPA in solutions with a high initial pH (pH = 11) than that with a low initial pH (pH = 3).
+
+It is found that BPA can almost be completely removed even while working with low concentrations of O3 (0.1 mg/L) and H2O2 (0.25 mmol/L). Comparing O3/H2O2/UV with O3 only, O3/H2O2, and O3/UV, it is determined that combining O3 with H2O2 and UV can significantly
+
+Fig. 6: Proposed mechanism of BPA degradation for O3/H2O2/UVimprove the generation of HO- and is a more efficient method than the other three ozone-based AOPs.
+
+Furthermore, applying 0.75 mg/L ozone, 0.38 mmol/L H2O2, initial pH = 9, and a reaction time of 26 min resulted in maximal removal as predicted by a CCD-based RSM. Five major intermediate products were identified on the basis of qualitative analysis with GC/MS, including 4-(1-hydroxypropan-2-yl) catechol, p-hydroquinone, 4-isopropenylphenol, phenol, and p-benzoquinone. Additionally, a BPA degradation pathway with UV/O3/H2O2 treatment was proposed in this work, depending on the identified by-products with GC-MS. In a further study, the major findings and the optimal conditions found in this study can be applied to remove BPA from real wastewaters using ozone-based AOPs treatments.
+
+###### Acknowledgements.
+
+ Ze Liu is supported financially by a PhD grant of the China Scholarship Council. Furthermore, this research fits within the LED H2O project, financially supported by The Flanders Knowledge Centre Water.
+
+## References
+
+- Arslan A, Topkaya E, Ozbay B et al (2017) Application of O2/UV/H2O2 oxidation and process optimization for treatment of potato chips manufacturing wastewater. Water Environ J 31:64-71. [https://doi.org/10.1111/vej.12227-9868-8227-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-82](https://doi.org/10.1111/vej.12227-9868-8227-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-8827-82)* Moussavi et al (2018) Moussavi G, Pourakbar M, Shekohiyan S, Satari M (2018) The photochemical decomposition and detoxification of bisphenol A in the VUV/H2O2 process: degradation, mineralization, and cytotoxicity assessment. Chem Eng J 331:755-764. [https://doi.org/10.1016/j.ccj.2017.09.009](https://doi.org/10.1016/j.ccj.2017.09.009)
+* Muller et al (2001) Muller JP, Gottschalk C, Jekel M (2001) Comparison of advanced oxidation processes in flow-through pilot plants (part II). Water Sci Technol J Int Assoc Water Pollut Res 44:311-315
+* Ohko et al (2001) Ohko Y, Ando I, Niwa C et al (2001) Degradation of bisphenol A in water by TiO2 photocatalyst. Environ Sci Technol 35:2365-2368. [https://doi.org/10.1021/cse0017574](https://doi.org/10.1021/cse0017574)
+* Ouran and Aaron (2014) Ouran M A, Aaron J-J (2014) Advanced oxidation processes in water/wastewater treatment: principles and applications. A review. Crit Rev Environ Sci Technol 44:2577-2641. [https://doi.org/10.1080/1064389.2013.829765](https://doi.org/10.1080/1064389.2013.829765)
+* Pikal-Cleland et al (2002) Pikal-Cleland KA, Cleland JL, Anchordoqy TJ, Carpenter JF (2002) Effect of glycine on pH changes and protein stability during freeze-thawing in phosphate buffer systems. J Pharm Sci 91:1969-1979. [https://doi.org/10.1002/jps.10184](https://doi.org/10.1002/jps.10184)
+* Pirilla et al (2015) Pirilla M, Saouche M, Ojala S et al (2015) Photocatalytic degradation of organic pollutants in wastewater. Proc Catal 58:1085-1099. [https://doi.org/10.1007/s11244-015-0477-7](https://doi.org/10.1007/s11244-015-0477-7)
+* Poerschmann et al (2010) Poerschmann J, Trommler U, Gorecki T (2010) Aromatic intermediate formation during oxidative degradation of bisphenol A by homogeneous sub-stoichiometric Fenton reaction. Chemosphere 79:975-986. [https://doi.org/10.1016/j.chemosphere.2010.03.030](https://doi.org/10.1016/j.chemosphere.2010.03.030)
+* Sharma and Mishra (2016) Sharma J, Mishra IM, Kumar V (2016) Mechanistic study of photo-oxidation of bisphenol-A (BPA) with hydrogen peroxide (H2O2) and sodium persulfate (SPS). J Environ Manage 166:12-22. [https://doi.org/10.1016/j.jenvman.2015.09.043](https://doi.org/10.1016/j.jenvman.2015.09.043)
+* Subedi et al (2013) Subedi B, Lee S, Moon H-B, Kannan K (2013) Psychoactive pharmaceuticals in sludge and their emission from wastewater treatment facilities in Korea. Environ Sci Technol 47:13321-13329. [https://doi.org/10.1021/es404129r](https://doi.org/10.1021/es404129r)
+* Umar et al (2013) Umar M, Roddick F, Fan L, Aziz HA (2013) Application of ozone for the removal of bisphenol A from water and wastewater--a review. Chemosphere 90:2197-2207. [https://doi.org/10.1016/j.chemosphere.2012.09.090](https://doi.org/10.1016/j.chemosphere.2012.09.090)
+* USFOod et al (2010) USFOod, Drug Administration (USFDA). Update on bisphenol A for use in food contact applications. [http://www.fda.gov/NewsEvents/PublicHealthFocus/ucm197739.htm2010](http://www.fda.gov/NewsEvents/PublicHealthFocus/ucm197739.htm2010)
+* Vandenberg LN et al (2007) Vandenberg LN, Hauser R, Marcus M et al (2007) Human exposure to bisphenol A (BPA). Reprod Toxicol 24:139-177. [https://doi.org/10.1016/j.reprotox.2007.07.010](https://doi.org/10.1016/j.reprotox.2007.07.010)
+* Wang and Xu (2012) Wang JL, Xu JLJ (2012) Advanced oxidation processes for wastewater treatment: formation of hydroxyl radical and application. Crit Rev Environ Sci Technol 42:251-325. [https://doi.org/10.1080/10643839.2010.507698](https://doi.org/10.1080/10643839.2010.507698)
+* Wang et al (2018) Wang T, Huang Z-X, Miao H-F et al (2018) Insights into influencing factor, degradation mechanism and potential toxicity involved in aqueous ozonation of oxcarbazepine (CHEM46939R1). Chemosphere 201:189-196. [https://doi.org/10.1016/j.chemosphere.2018.02.062](https://doi.org/10.1016/j.chemosphere.2018.02.062)
+* Wu et al (2008) Wu JJ, Yang JS, Muruganandham M, Wu CC (2008) The oxidation study of 2-propanol using ozone-based advanced oxidation processes. Sep Purif Technol 62:39-46. [https://doi.org/10.1016/j.seppur.2007.12.018](https://doi.org/10.1016/j.seppur.2007.12.018)
+* Yang et al (2014) Yang Y, Pignatello JJ, Ma J, Mitch WA (2014) Comparison of halide impacts on the efficiency of contaminant degradation by sulfate and hydroxyl radical-based advanced oxidation processes (AOPs). Environ Sci Technol 48:2344-2351. [https://doi.org/10.1021/es4041184](https://doi.org/10.1021/es4041184)
+* Yang et al (2016) Yang Y, Guo H, Zhang Y et al (2016) Degradation of bisphenol A using ozone/persulfate process: kinetics and mechanism. Water Air Soil Pollut 227:53. [https://doi.org/10.1007/s11270-016-2746-x](https://doi.org/10.1007/s11270-016-2746-x)
+* Yoon and Jeong (2012) Yoon S-H, Jeong S, Lee S (2012) Oxidation of bisphenol A by UV/S\({}_{2}\)O: comparison with UV/H\({}_{2}\)O\({}_{2}\). Environ Technol 33:123-128. [https://doi.org/10.1080/09593330.2011.579181](https://doi.org/10.1080/09593330.2011.579181)
+* Young et al (2013) Young T, Geng M, Lin L, Thagard SM (2013) Oxidative degradation of bisphenol A: a comparison between fenton reagent, UV, UV/H\({}_{2}\)O\({}_{2}\) and ultrasound. J Adv Oxid Technol 16:89-101
+* Yu et al (2016) Yu Y, Li S, Peng X et al (2016) Efficient oxidation of bisphenol A with oxysulfur radicals generated by iron-catalyzed autoxidation of sulfite at circumneutral pH under UV irradiation. Environ Chem Lett 14:527-532. [https://doi.org/10.1007/s10311-016-0573-3](https://doi.org/10.1007/s10311-016-0573-3)

manual_annotation/200pdfs_in_mmd/10.1007_s10311-005-0103-1.mmd

@@ -0,0 +1,133 @@
+Reaction of _ortho_-methoxybenzoic acid with the water disinfecting agents ozone, chlorine and sodium hypochlorite
+
+Gulnara M. Shaydullina Natalya A. Sinkova Albert T. Lebedev
+
+G. M. Shaydullina N. A. Sinkova A. T. Lebedev ((ES)) G. M. Shaydullina N. A. Sinkova A. T. Lebedev ((ES)) Department of Organic Chemistry, Moscow State University, Leninskie gory-1/3, 119992 Moscow, Russia
+
+e-mail: lebedev@org.chem.msu.ru
+
+Tel.: +7-095-939-1407
+
+Fax: +7-095-939-1407
+
+###### Abstract
+
+Ozone, chlorine and sodium hypochlorite are commonly used as disinfecting agents for drinking water production. The reaction pathways of ozonation and chlorination of \(o\)-methoxybenzoic acid in aqueous solution were studied using gas chromatography-mass spectrometry (GC-MS) and high pressure liquid chromatography (HPLC). The results show that less than 1% of \(o\)-methoxybenzoic acid remains in reaction. The final major products using ozone oxidation are oxalic and glyoxalic acids. Phenols appear only at insufficient ozone levels. Sodium hypochlorite leads to higher levels of primary products. Molecular chlorine leads to the formation of higher amounts of polychlorinated derivatives. Model experiments allow to propose schemes of \(o\)-methoxybenzoic acid transformation under the conditions simulating water treatment processes.
+
+ortho-methoxybenzoic acid Ozone Chlorine Sodium hypochlorite Mass spectrometry GC/MS HPLC Water treatment Disinfecting agents Disinfection by-products 2005 2
+
+## Experimental
+
+### Ozonation experiments
+
+Ozonation experiments were conducted in a semi-batch reactor at ambient temperature. Ozone was generated from dried oxygen by electric discharge ozone generator. An ozonized oxygen stream with ozone concentration 20 mg/l or 80 mg/l, was bubbled at a flow rate 6 l/h to phosphate-buffered (pH 7) aqueous solution of _ortho_-methoxybenzoic acid (10\({}^{-3}\) M). The molar ratios of the studied acid to ozone were determined by taking into account ozone concentration, flow rates and time of gas mixture bubbling. The molar ratios of organic substrate (S) to ozone [S]:[O\({}_{3}\)] were 2:1, 1:1, 1:2, 1:5, 1:10 and 1:50.
+
+### Chlorination experiments
+
+Reactions of _ortho_-methoxybenzoic acid with chlorine or sodium hypochlorite were conducted in the dark at 20\({}^{\circ}\)C in 0.2 M phosphate buffer solution (pH 7) prepared from reagent grade monobasic potassium phosphate and dibasic sodium phosphate. A total of 50 ml of buffer and calculated amount of chlorinating agents was added to substrate solution and the volume was adjusted to 100 ml with distilled water. Active chlorine content was determined by iodometric titration before chlorination. In all experiments, the concentration of _ortho_-methoxybenzoic acid was 10\({}^{-3}\) M. The molar ratios of substrate to active chlorine [S]:[Cl\({}^{\rm act}\)] were 1:1, 1:5 and 1:50. After 24 h, unreacted chlorinating agents were reduced by addition of an excess of reagent grade sodium sulphite.
+
+### Analytical methods
+
+Analysis of volatile products was carried out with HP5973 GC-MS system equipped with 50 m DB-5 fused silica capillary column (50 m length, 0.25 mm i.d.). Aliquot of 5 ml of a sample was injected into a "purge and trap" concentrator and purged by helium for 10 min. Then trap sorbent was heated to 170\({}^{\circ}\)C allowing organic molecules into the gas chromatograph which operating conditions were as follows: carrier gas--helium, flow rate \(-\)1 ml/min, initial temperature \(-\)40\({}^{\circ}\)C (2 min), programming rate \(-\)10\({}^{\circ}\)C/min up to 250\({}^{\circ}\)C, then isothermal for 1 min. Electron ionization (EI) mass spectra were obtained at 70 eV electron energy with the ion source at 180\({}^{\circ}\)C. Bromofluorobenzene was used as an internal standard to quantify the products.
+
+Semi volatile products were consequently extracted at neutral and acidic pH with three sequential aliquots of freshly distilled dichloromethane. Sodium chloride was added to aqueous solutions to enhance the extraction of organic by-products. Extracts were dried over anhydrous sodium sulphate and concentrated by evaporation. Analysis was carried out with SSQ-7000 GC-MS instrument (Finnigan) using HP-5MS fused silica capillary column (30 m length, 0.25 mm i.d.). The operating conditions were as follows: carrier gas-helium, flow rate \(-\)1 ml/min, initial temperature \(-\)50\({}^{\circ}\)C (2 min), programming rate \(-\)10\({}^{\circ}\)C/min up to 280\({}^{\circ}\)C, then isothermal for 3 min. EI mass spectra were obtained at 70 eV electron energy with the ion source at 180\({}^{\circ}\)C. The background-subtracted mass spectra were matched against those in the NIST mass spectra library and interpreted on the basis of the observed fragmentation. The quantities of products in the reaction mixture were estimated using naphthalene-d8 and phenanthrene-d10 as internal standards.
+
+The concentration of non volatile products were monitored by HPLC-UV with a 1100 chromatographic system (Agilent Technologies) equipped with a Spherisorb C-18 (250x4.6 mm i.d.) column and a multiple wavelength detector, set at 196 nm. Injected samples (10 \(\mu\)l) were eluted with water acidified with TFA (pH = 1.5) at a flow rate of 1 ml/min.
+
+## Results and discussion
+
+### Background
+
+From the chemical point of view, molecular ozone in aqueous solution remains as O\({}_{3}\), or decomposes to form very reactive hydroxyl radicals. The latter is a stronger oxidizing agent (\(E^{\circ}\)=2.80 V) than the molecular ozone itself (\(E^{\circ}\)=2.07 V):
+
+\[{\rm O}_{3}+{\rm H}_{2}{\rm O}\rightarrow{\rm 2OH}^{\bullet}+{\rm O}_{2}\]
+
+Ozone and hydroxyl radicals react with organics by the following principal mechanisms: ozonolysis of unsaturated bonds, hydroxyl radical attack of aromatic rings and oxidation (Yamamoto et al. 1979).
+
+The term "aquatic chlorination" has been commonly used to name transformation of organic substrates by chlorine as well as by sodium hypochlorite in aquatic solution. However, there are some differences between reaction behavior of these chlorinating agents. Both reagents exist in aquatic solution as mixtures of different species due to equilibrium reactions with the solvent.
+
+Thus, aquatic chlorine represents a mixture of molecular chlorine, hydrochloric and hypochlorous acids with the corresponding anions:
+
+\[{\rm Cl}_{2}+{\rm H}_{2}{\rm O}\rightleftarrows{\rm HClO}+{\rm H}^{+}+{\rm Cl }^{-}\]
+
+\[{\rm HClO}+{\rm H}_{2}{\rm O}\rightleftarrows{\rm H}_{3}{\rm O}^{+}+{\rm ClO} ^{-}\]
+
+while sodium hypochlorite exists in aquatic solution mainly as hypochlorous acid and hypochlorite anion due to equilibrium reactions:
+
+\[{\rm Na}{\rm O}{\rm Cl}\rightleftarrows{\rm Na}+{\rm O}{\rm Cl}^{-}\]
+
+\[{\rm ClO}^{-}+{\rm H}_{2}{\rm O}\rightleftarrows{\rm HClO}+{\rm OH}^{-}\]
+
+Molecular chlorine is a stronger oxidizing agent (\(E^{\circ}\)=1.59) than hypochlorous acid (\(E^{\circ}\)=1.50) or hypochlorous acid (\(E^{\circ}\)=1.50).
+
+chlorite anion (_E_\({}^{\circ}\)=0.89). The mentioned species react with organic compounds by addition, substitution or oxidation (Boyce and Horning 1983). Moreover according to _ab initio_ calculations a complex of hydroxonium ion with HOCl molecule is a reactive particle in the reaction of aqueous chlorination (Lebedev et al 2004).
+
+### Ozonation by-products
+
+The results on _o_-methoxybenzoic acid ozonation are summarized in Fig. 1 and Table 1. The transformation of _o_-methoxybenzoic acid during ozonation starts with hydroxyl radical attack accompanied by decarboxylation giving methoxyphenols 2 and 3. Comparison of yields of the isomers proves that _ipso_-substitution of carboxylic group is preferable. Consequent stages result in formation of methoxybenzenediols 4, 5 and 8.
+
+\begin{table}
+\begin{tabular}{l l l l l l l l} \hline No & \multicolumn{2}{l}{[S]:[O\({}_{3}\)]} & & & & & \\  & & 2:1 & 1:1 & 1:2 & 1:5 & 1:10 & 1:50 \\ \hline
+1 & 7630 & 6510 & 5560 & 4610 & 3900 & 40.0 \\
+2 & – & – & 0.36 & 0.39 & 0.46 & – \\
+3 & – & 0.23 & 0.45 & 0.69 & 1.18 & 1.22 & – \\
+4 & – & – & 0.08 & 0.09 & 0.18 & – \\
+5 & 0.27 & 0.62 & 2.22 & 2.18 & 2.15 & – \\
+6 & – & – & 0.05 & 0.07 & 0.21 & – \\
+7 & – & – & 0.19 & – & – & – \\
+8 & – & – & 0.36 & 0.40 & 1.52 & – \\
+9 & – & 0.11 & 0.21 & 0.40 & 0.13 & 0.19 & – \\
+10 & – & – & – & – & – & – & 270 \\
+11 & 40.0 & 180 & 280 & 360 & 620 & 3150 \\
+12 & – & – & – & – & 30.0 & 800 \\
+13 & – & – & 0.28 & 0.26 & 0.23 & – \\
+14 & – & – & – & 0.04 & 0.15 & – \\ \hline \end{tabular}
+\end{table}
+Table 1: Products of _ortho_-methoxybenzoic acid ozonation in aqueous solution in a semi-batch reactor at ambient temperature and neutral pH at various molar ratios of investigated acid (S) to bubbled ozone (O\({}_{3}\)) (\(\mu\)g)
+
+Fig. 1: Degradation pathways of _ortho_-methoxybenzoic acid during ozonation in aqueous solution in a semi-batch reactor at ambient temperature and neutral pH at various molar ratios of investigated acid to bubbled ozone: (**a**): hydroxyl radical attack of aromatic ring with concurrent decarboxylation; (**b**): radical hydroxylation of aromatic ring; (**c**): fast ozonolysis of unsaturated bonds; (**d**): formation of oxalic acid as alienated off-fragment during ozonolysis of double bonds at different intermediate stages; (**e**): dehydration; (**f**): oxidation. m/z refers to mass spectrum data Ozonolysis of ring C-C bonds adjacent to electron-odonating methoxy-group led to the cleavage of the aromatic ring and formation of compounds with linear structures. Concentration of linear C6 compounds in reaction mixtures was low in all the experiments due to fast ozonation of double bonds. Concentration of unsaturated dialdehyde 6 was higher than that of ester 7. Following reactions involving double bonds of the intermediates 6 and 7 bring to several short-chain products.
+
+There were few primary products detected at the highest molar ratio of ozone to substrate. Only oxalic (11), glyoxalic (12) and maleic (10) acids were observed with rather high yields. These products originate at different intermediate stages of ozonation. A decrease of other products concentration at the highest ratios of ozone to substrate is accounted for their decay at the advanced stages of conversion with final formation of CO2 and H2O.
+
+### Chlorination by-products
+
+Reactions of _o_-methoxybenzoic acid with chlorine and sodium hypochlorite in aqueous solutions are presented in Fig. 2. Monochloro derivatives 15 and 22 are the major products (Table 2). Electrophilic substitution of two hydrogen atoms for chlorine takes place by coordinated control of both groups. Another primary reaction--chlorodecarboxylation--leads to the formation of _orthochoromethoxybenzene_ (18).
+
+Comparison of the assortment and relative amounts of reaction products demonstrates higher chlorination activity of sodium hypochlorite at the initial stages. The extent of conversion of _o_-methoxybenzoic acid with equimolar ratios of investigated agents was 5.6% in case of chlorine and 14.9% in case of sodium hypochlorite. Further deceleration of electrophilic substitution occurs due to insertion of chlorine atom into aromatic ring.
+
+It is worth mentioning the fact that only chloroform was detected in "purge-and-trap" experiments. Since CHCl3 is a dominant product of aqueous chlorination of a reactive organic molecule, transformation of _orthochoromethoxybenzoic acid stops mainly at initial stages (Fig. 2)
+
+Aromatic ring cleavage is a minor process. Brominated derivatives (16, 17) appeared due to reactions of substrate with bromine impurities in chlorinating agent solutions. An increase of a chlorinating agent concentration increased a variety of resulting organochlorines. It is to be noted that the assortment of chlorinated by-products was higher in case of chlorine in comparison with sodium hypochlorite.
+
+## Conclusions
+
+Ozonation of _o_-methoxybenzoic acid in aqueous solution led to its nearly complete disappearance, while oxalic and glyoxalic acids are the dominant by-products. Formation
+
+\begin{table}
+\begin{tabular}{l c c c c c c} \hline No & \multicolumn{2}{l}{[S]:[Cl\({}^{\rm act}\)]} & & & \\ \cline{2-7}  & Cl\({}_{2}\) & & & NaClO & & \\ \cline{2-7}  & 1:1 & 1:5 & 1:50 & 1:1 & 1:5 & 1:50 \\ \hline
+1 & 14100 & 1180 & 37.4 & 12100 & 790 & 31.5 \\
+15 & 50.4 & 712 & 184 & 171 & 946 & 1022 \\
+16 & – & – & – & – & – & 8.4 \\
+17 & – & – & 6.32 & – & – & 19.6 \\
+18 & 2.46 & 59.8 & 14.9 & 10.1 & 62.4 & 58.5 \\
+19 & – & – & 29 & – & 1,4 & – \\
+20 & – & 9.15 & 880 & – & 15.3 & 4.31 \\
+21 & – & – & 26.6 & – & – & – \\
+22 & 166 & 4060 & 1650 & 657 & 4130 & 4270 \\
+23 & – & – & 2820 & – & – & – \\ CHCl\({}_{3}\) & – & – & 21.0 & – & – & 6.70 \\ \hline \end{tabular}
+\end{table}
+Table 2: Products of _orthochor_-methoxybenzoic acid chlorination with Cl\({}_{2}\) and NaClO in aqueous solution at various molar ratios of substrate (S) to active chlorine (Cl\({}^{\rm act}\)) at ambient temperature and neutral pH (\(\mu\)g)
+
+Fig. 2: Transformation pathways of _ortho_-methoxybenzoic acid during chlorination with sodium hypochlorite and chlorine in aqueous solution at various molar ratios of substrate (S) to active chlorine (Cl\({}^{\rm act}\)) at ambient temperature and neutral pH: electrophilic substitution of hydrogen atoms for chlorine (**a**) or bromine (**b**) and electrophilic substitution in aromatic ring with concurrent decarboxylation - chlorodecarboxylation (**c**). m/z refers to mass spectrum data of reaction products in conditions of chlorination depends on molar ratios of chlorinating reagents to the organic substrate. An increase of the active chlorine concentration increased variety of resulting products. Cl\({}_{2}\) and NaCl react similarly. However, the yields of monochromiated derivatives were higher with sodium hypochlorite. On the contrary the levels of polychlorinated derivatives were higher in case of molecular chlorine. The ozonation products are definitely less toxic than the chlorination ones. Nevertheless formation of phenols at pre-ozonation stage of water treatment process could be a source of hazardous chlorophenols, haloforms, and other toxic chemicals during final chlorination.
+
+## References
+
+* Boyce and Horning (1983) Boyce SD, Horning JF (1983) Reaction pathways of trihalomethane formation from the halogenation of dihydroxyaromatic model compounds for humic acid. Environ Sci Technol 17:202-210
+* Hoigne (1982) Hoigne J (1982) Handbook of ozone technology and applications. Ann Arbor Sci Publishers, Boston, pp 378
+* Lebedev et al (1997) Lebedev AT, Moshkarina NA, Buriak AK, Petrosyan VS (1997) Water chlorination of nitrogen containing fragments of humic material. Fresenius Environ Bull 6:727-733
+* Lebedev et al (2004) Lebedev AT, Shaidullina GM, Sinkova NA, Kharchevnikova NV (2004) GC-MS comparison of the behavior of chlorine and sodium hypochlorite towards organic compounds dissolved in water. Water Res 38:3713-3718
+* Rook (1974) Rook JJ (1974) Formation of haloforms during chlorination of natural waters. Water Treat Exam 23:234-243
+* Rook (1977) Rook JJ (1977) Chlorination reactions of fulvic acids in natural waters. Environ Sci Technol 11:478-482
+* Richardson et al (2000) Richardson SD, Thruston AD, Caughran TV, Chen PH, Collette TW, Schenck KM, Lykins BW, Rav-Acha C, Glezer V (2000) Identification of new drinking water disinfection byproducts from ozone, chlorine dioxide, chloramine, and chlorine. Water Air Soil Poll 23:95-102
+* Tretyakova et al (1994) Tretyakova NY, Lebedev AT, Petrosyan VS (1994) Degradative pathways for aqueous chlorination of orchinol. Environ Sci Technol 28:606-611
+* Yamamoto et al (1979) Yamamoto Y, Niki E, Shiokawa H, Kamiya Y(1979) Ozonation of organic compounds.2.Ozonation of phenol in water. J Organic Chem 44:2137-214

manual_annotation/200pdfs_in_mmd/10.1007_s10512-012-9603-9.mmd

@@ -0,0 +1,84 @@
+# Oxidative decomposition of oxalate ions in water solutions of concentrated ozone
+
+Yu. O. Lagunova, A. F. Seliverstov, D. G. Ershov, A. G. Basiev
+
+The objective of this work is to study the oxidative decomposition of oxalate ions by concentrated ozone in water solutions under different conditions. The changes of the oxalate ion and ozone concentrations and the pH of solutions as functions of the ozonation time are determined. It is shown that oxalate ion decomposition proceeds most efficiently in an alkaline medium. The temperature dependence of the oxalate ion oxidation is determined to be extreme. The maximum effectiveness of the process depends on the pH of the medium: 50degC at pH = 10 and 70degC at pH = 2. The effect of the ozone concentration in an ozone-oxygen mixture on the effectiveness of the oxidation of oxalic acid is studied. It is shown that as the ozone concentration increases the time required for oxidation of oxalic acid decreases proportionally. The ozone consumption on oxidative decomposition of 1 g oxalic at different pH and temperatures of the solution is established. It is shown that the ozone consumption on the oxidation of oxalic acid has a minimum at 50-60degC.
+
+Localization, concentration and reprocessing of liquid rawastes simplify considerably after the complexones present in them (sodium ethylenediaminetetraacetic - EDTA, oxalic, citric, and other acids), which bind radionuclides and make it difficult to remove them by conventional physicochemical means, are removed or broken down. At present, their oxidative decomposition is the most effective method. Preliminary removal of organic impurities from wastes makes it possible to reprocess and store wastes at NPPs more efficiently and with lower capital expenditures.
+
+Oxalate ions are among the most widely used complexones. Ordinarily, potassium permanganate in an alkaline medium is used to oxidize them, which increases secondary rawastes. Other oxidizers (H2O2, ClO-, BrO-, nitric acid in the presence of catalysts) were found to be ineffective. Ozone is the most effective oxidizer; its oxidation power is greater: -2.07 V versus 1.77 and 1.69 V for H2O2 and MnO4, respectively [1]. In addition, it effectively oxidizes organic substances in solutions with a high salt background.
+
+The present work is devoted to studying the destruction of oxalate ions by means of concentrated ozone in solutions whose composition corresponds to that of decontamination solutions most often used for processing contaminated equipment as well as by activation of the oxidative process.
+
+## Experimental Part
+
+New-generation ozonators with capacity 10 g/h, making it possible to obtain high ozone concentrations (200 mg/liter and higher), developed at the Laboratory for Ozone Technologies, were studied in this work. The ozone concentration in the ozone-oxygen mixture was 180 mg/liter; the rate of feeding into the reactor was 2 ml/sec. The ozone concentration at the exit from the ozonator was varied by means of the gas flow rate and the voltage applied to the electrodes. The change of the concentration in the gas phase was recorded with an optical UV analyzer. The ozone concentration in the gas was measured in quartz cells with optical path length 1.6 mm. The ozonolysis process was conducted in a 400 cm3 reactor (height-to-diameter ratio 2:1) with a mechanical mixer (900 rpm). At this mixer rotation rate, the mixing ratehas no effect on the oxidation efficiency. The oxalate ion concentration was determined by the procedure of [2]. The specific activity of the solutions was measured by a radiometric method using the Gamma Plyus apparatus.
+
+**Results and Discussion.** Attention was focused first on studying the destruction of oxalate ions in solutions with pH characteristic for radiochemical practice (acidic and alkaline decontamination solutions, alkaline waters for special laundries, bottoms) [3, 4].
+
+In Fig. 1 it is evident that oxalate ions are slowly oxidized by ozone and most of them (approximately 95%) decompose in 150 min. The ozonogram, which records the ozone concentration at the exit from the reactor, attests that the solution
+
+Figure 1: Decomposition of oxalate ions versus time (_1_) and ozonogram of the oxidation of the oxalate ion (2) with [C\({}_{2}\)O\({}_{4}^{2-}\)] = 1 g/liter, \(T\) = 20\({}^{\circ}\)C, pH = 9.9, ozone–oxygen mixture flow rate 2 ml/sec, [O\({}_{3}\)] = 150 mg/liter.
+
+Figure 2: Rate of oxidative decomposition of the oxalate ion by ozone versus solution pH with [C\({}_{2}\)O\({}_{4}^{2-}\)] = 1 g/liter, \(T\) = 50\({}^{\circ}\)C, ozone–oxygen mixture flow rate 2 ml/sec, [O\({}_{3}\)] = 150 mg/liter (_1_) and temperature with [C\({}_{2}\)O\({}_{4}^{2-}\)] = 1 g/liter, pH = 9.9 (2).
+
+saturates in the first 3-4 min after which oxidation of the oxalate ion is observed. Once the oxidative destruction of the oxalate ion is complete, the ozone concentration in the solution no longer changes. This probably shows that oxidation proceeds in a single step via the reaction
+
+\[(\mathrm{COO}^{-})_{2}^{2-}+\mathrm{O}_{3}+\mathrm{H}_{2}\mathrm{O}\to 2\mathrm{CO}_{3}^{2-}+\mathrm{O}_{2}+2\mathrm{H}^{+}. \tag{1}\]
+
+The rate of decomposition of the oxalate ion increases to initial solution pH = 10 after which it decreases somewhat (to pH = 11.27) and drops sharply at pH = 12.57 (Fig. 2, curve _1_).
+
+The observed effect of solution pH on the oxidation of oxalate ions is related with the fact that OH- ions catalyze ozone decomposition with the formation of hydroxyl radicals in accordance with the reactions
+
+\[\mathrm{O}_{3}+\mathrm{OH}^{-}\rightarrow\mathrm{O}_{2}^{\bullet-}+\mathrm{HO}_{2}^{\bullet};\]
+
+\[\mathrm{O}_{3}+\mathrm{OH}^{-}\rightarrow\mathrm{O}_{2}+\mathrm{HO}_{2}^{-}\ \mathrm{and\ so\ on}\]
+
+\[\mathrm{HO}_{2}^{\bullet}\leftrightarrow\mathrm{H}^{+}+\mathrm{O}_{2}^{\bullet-};\quad\mathrm{H}_{2}\mathrm{O}_{2}\leftrightarrow\mathrm{HO}_{2}^{-}+\mathrm{H}^{+};\]
+
+\[\mathrm{O}_{3}+\mathrm{HO}_{2}^{-}\rightarrow\mathrm{O}_{2}+\mathrm{O}_{2}^{\bullet-}+\mathrm{{}^{\bullet}OH};\quad\mathrm{O}_{3}+\mathrm{O}_{2}^{\bullet-}\rightarrow\mathrm{O}_{3}^{-}+\mathrm{O}_{2};\]
+
+\[\mathrm{O}_{3}^{\bullet-}+\mathrm{H}_{2}\mathrm{O}\rightarrow\mathrm{{}^{ \bullet}OH}+\mathrm{OH}^{-}+\mathrm{O}_{2},\]
+
+i.e., hydroxyl ions OH- interact with ozone with formation of the radicals \({}^{\bullet}\)OH and \(\mathrm{O}_{2}^{\bullet-}\). These radicals make a considerable contribution to the subsequent decomposition of oxalate ions. The rate of interaction of the hydroxyl radicals formed with oxalate ions 1\(\cdot 10^{7}\) liters/(mole\(\cdot\)sec) is much higher than its oxidation by ozone 0.04 liters/(mole\(\cdot\)sec).
+
+As solution pH rises above 12, a stable ozonide anion-radical forms via the reaction \(\mathrm{O}_{3}+\mathrm{OH}^{-}\rightarrow\mathrm{O}_{3}^{\bullet-}+\mathrm{HO}^{\bullet}\), while the hydroxyl radical formed dissociates via the reaction \({}^{\bullet}\)OH \(\leftrightarrow\)\(\mathrm{O}^{\bullet-}+\mathrm{H}^{+}\), pH = 11.8.
+
+Both particles formed are less reactive than the hydroxyl radical, and oxidize oxalate ions much more slowly.
+
+The decomposition of the oxalate ion in an alkaline medium is temperature dependent. As solution temperature increases to 50\({}^{\circ}\)C, oxidation of oxalate accelerates significantly. An increase to 90\({}^{\circ}\)C decreases the oxidation rate. Thus,
+
+Figure 3: Ozone consumption on oxidation versus temperature with [C\({}_{2}\)O\({}_{4}^{2-}\)] = 1 g/liter, \(T\) = 20\({}^{\circ}\)C, ozone–oxygen mixture flow rate 2 ml/sec, [O\({}_{3}\)] = 150 mg/liter, pH = 9.9 (1), 2.2 (2).
+
+the temperature dependence of oxalate oxidation reaches a maximum at 50\({}^{\circ}\)C (see Fig. 2, curve 2). The extremal character of the temperature dependence is related with the fact that even though the rate of the oxidation reaction increases according to the Arrhenius law the stationary ozone concentration in solution decreases with increasing temperature as a result of decreasing ozone solubility in aqueous solutions [5]. For this reason, the increase of the reaction rate due to an increase of the temperature is compensated by an 8-10-fold decrease of the oxidizer concentration in solution.
+
+It was established that the decomposition rate of oxalate ions increases directly (by a factor of 9) as the ozone concentration increases from 20 to 180 mg/liter. Thus, the use of concentrated ozone makes it possible to accelerate the decomposition of oxalate ions significantly.
+
+The studies established that ozone is expended on oxidative destruction of oxalate ions at different solution temperature (Fig. 3, curve 1). The minimal flow rate is observed at 50\({}^{\circ}\)C and equals 0.6 \(\pm\) 0.1 g O\({}_{3}\) per 1 g oxalate, or 1.1 \(\pm\) 0.1 moles O\({}_{3}\) per 1 mole oxalate. Such consumption corresponds to the stoichiometry of reaction (1). The decomposition of oxalic acid at pH = 2.2 is similar to the behavior obtained for pH = 10 at optimal temperature of oxidative decomposition 60\({}^{\circ}\)C. The minimum consumption of ozone on oxidative decomposition of oxalic acid at 60\({}^{\circ}\)C is 0.62 \(\pm\) 0.1 g O\({}_{3}\) per 1 g oxalate, or 1.1 \(\pm\) 0.1 mole O\({}_{3}\) per mole oxalate (see Fig. 3, curve 2).
+
+One of the salts most often encountered in liquid radwastes is sodium nitrate. Studies have shown that as sodium nitrate concentration increases to 400 g/liter the oxalate-ion decomposition time triples. However, even in this case the concentration ozone and elevated temperature make it possible to conduct the process within technologically justifiable periods of time.
+
+The methods developed for oxidative decomposition of organic complexes and separation of radionuclides from aqueous media were validated on real liquid wastes from the URB-8 facility (MosNPO Radon) with specific activity 1.2\(\cdot\)10\({}^{5}\) Bq/liter and salt content 230 g/liter. Complexone decomposition was determined according to the change of the degree of cesium precipitation with nickel ferrocyanide precipitates as a function of the amount of ozone introduced into the solution (see Table 1). The \({}^{137}\)Cs removal factor for the concentrate from the URB-8 facility on nickel ferrocyanide precipitate in the absence of complexones is 65. The amount of ozone expended on the decomposition of organic complexes in 1 liter of solution of the concentrate was 1.2 g. The results presented in Table 1 show that ozone treatment of the concentrate from the URB-8 facility in the amount 1.2 g/liter makes it possible to eliminate the negative effect of complexones on cesium co-precipitation with nickel ferrocyanide precipitate.
+
+In summary, experiments confirm that oxidation of organic compounds of radwaste components can be intensified by changing the ozone concentration in the ozone-oxygen mixture, pH and temperature. They validate the regimes for optimizing the purification of liquid radwastes and make it possible to handle radwastes efficiently.
+
+\begin{table}
+\begin{tabular}{|c|c|c|} \hline \multicolumn{1}{|c|}{Ozonation time, min} & \({}^{137}\)Cs specific activity, Bq/liter & Removal factor \\ \hline
+0 & 5200 & 1 \\ \hline
+10 & 520 & 10 \\ \hline
+20 & 220 & 24 \\ \hline
+30 & 140 & 37 \\ \hline
+40 & 104 & 50 \\ \hline
+50 & 95 & 55 \\ \hline
+60 & 87 & 60 \\ \hline
+70 & 85 & 61 \\ \hline \end{tabular}
+\end{table}
+Table 1: \({}^{137}\)Cs Removal Factor for Concentrate from the URB-8 Facility Versus the Ozonation Time
+
+## References
+
+* [1] V. A. Rabinovich and Z. Ya. Khavin, _Brief Chemical Handbook_, Khimiya, Moscow (1977).
+* [2] Z. Draganich, "The spectrophotometric determination of some organic acids with copper benzidine," _Anal. Chim. Acta_, **28**, 394-397 (1963).
+* [3] N. I. Ampelogova, Yu. M. Simanovskii, and A. A. Trapeznikov, _Decontamination in Nuclear Power_, Energoatomizdat, Moscow (1982).
+* [4] B. E. Rabchikov, _Purification of Liquid Radwastes_, DeLi Print, Moscow (2008).
+* [5] B. G. Ershov, N. M. Panich, and A. F. Silverstov, "Solubility of ozone in concentrated aqueous solutions of salts," _Zh. Prikl. Khimi_, **80**, No. 11, 1787-1790 (2007).

manual_annotation/200pdfs_in_mmd/10.1007_s11172-008-0043-6.mmd

@@ -0,0 +1,116 @@
+# Kinetics and mechanism of the reactions of ozone with guaiacol, veratrol, and veratrol derivatives
+
+A. G. Khudoshin, A. N. Mitrofanova, V. V. Lunin
+
+# Abstract
+
+The kinetics of the reactions of ozone with compounds modeling structural lignin fragments, _viz._, phenol, guaiacol, veratrol, veratric aldehyde, and veratric alcohol, was studied. The reaction rate constants were calculated using the mass transfer model for the chemical reaction based on the film theory. The rate constants of the reactions of ozone with compounds of the veratryl series were calculated by the Hammett equation. The major ozonation products of the studied compounds were determined by HPLC. The ozonation mechanism was proposed.
+
+## Introduction
+
+The reactions of ozone with organic compounds were studied on a setup, whose scheme has been described previously.6 Ozone was obtained from air using a laboratory glass ozonator in an electric silent discharge (frequency 7 kHz, \(I\) = 40 mA, \(W\) = 4 kW).
+
+The ozone concentrations in a gas mixture at the outlet and inlet of the reactor were determined using a Medozon 254/3 ozonometer from the optical absorption at \(l\) = 254 nm.
+
+Ozonation was carried out at 298 K, at the initial ozone concentration (0.9 - 1.1) * 10-3 mol L-1, and at the volumetric flow rate of the gas mixture \(w\) = 25 L h-1. The volume of the reaction mixture was 0.55 L.
+
+Ozonation was carried out as follows. A solution in the reactor was saturated with ozone until the stationary ozone concentration, and then the substance to be ozonated was added. The moment, when the ozone concentrations at the inlet and outlet of the reactor became equal, was considered as the end of the reaction. Samples for the chromatographic analysis of the ozonation products were taken during ozonation.
+
+The substrates were phenol (3 * 10-4 - 4 * 10-2 mol L-1), 2-methoxyphenyl (guaiacol) (1 * 10-4 * 10-3 mol L-1),1,2-dimethoxybenzene (vertorol) (2*10-4--4*10-2 mol L-1), 3,4-dimethoxybenzaldehyde (vertatric aldehyde) (3*10-5-2*10-2 mol L-1), 3,4-dimethoxybenzyl alcohol (vertatric alcohol) (3*10-4-1*10-2 mol L-1), and 3,4-dimethoxybenzoic acid (vertatric acid) (3*10-4-1*10-2 mol L-1). The reaction with ozone was studied in aqueous solutions of sulfuric acid (pH 1.5). The reaction course was monitored by periodical sampling. The samples were analyzed by spectrophotometry (Cary 3E, Varian, USA) and HPLC (Agilent 1100 chromatograph with a spectrophotometric detector at 270 nm, Zorbax C-18 column, mixture MeCN--water--1% orthophosphoric acid as a mobile phase). Chromatographic analyses were carried out in the gradient regime according to the scheme presented below.
+
+\[\begin{array}{l}t/\min \\ 0{-}4.5\\ 4.5{-}8\\ 8{-}9\\ 9{-}10\end{array}\qquad\begin{array}{l}\mathrm{MeCN}:\mathrm{H}_{\mathrm{O}}:\mathrm{H}_{3}\mathrm{PO}_{4}\\ 30:69:1\\ 60:39:1\\ 60:39:1-30:69:1\\ \end{array}\]
+
+UV spectra were recorded for the identification of chromatographic peaks.
+
+## Results and Discussion
+
+**Kinetic model.** Various physicochemical models are used to determine the kinetic parameters of gas--liquid reactions accompanied by mass transfer. The film model is the simplest.7,8 According to this model, a film is assumed to be formed near the interface, and ozone can penetrate through this film only due to molecular diffusion.
+
+The reaction rates of ozone with various substances can be estimated by both the rate of ozone absorption and the change in the substrate concentration. For the reaction
+
+\[\begin{array}{l}(\mathrm{O}_{3})_{\mathrm{0}}\\ \mathrm{(O}_{3})_{\mathrm{l}}\\ \mathrm{(O}_{3})_{\mathrm{l}}\\ \mathrm{+}\mathrm{B}\end{array}\qquad\begin{array}{l}\mathrm{P},\\ \mathrm{where}\ (\mathrm{O}_{3})_{\mathrm{g}}\ \mathrm{and}\ (\mathrm{O}_{3})_{\mathrm{l}}\ \mathrm{is}\ \mathrm{ozone}\ \mathrm{in}\ \mathrm{the}\ \mathrm{gas}\ \mathrm{phase}\ \mathrm{and}\ \mathrm{in}\ \mathrm{the}\ \mathrm{solution}\ \mathrm{,respectively,}\ \mathrm{B}\ \mathrm{is}\ \mathrm{the}\ \mathrm{oxidizable}\ \mathrm{substance}\ \mathrm{,and}\ \mathrm{P}\ \mathrm{is}\ \mathrm{the}\ \mathrm{reaction}\ \mathrm{product}\ \mathrm{,the}\ \mathrm{adsorption}\ \mathrm{rate}\ \bar{V}\ \mathrm{can}\ \mathrm{be}\ \mathrm{written}\ \mathrm{in}\ \mathrm{the}\ \mathrm{form}\The earlier published[10] diffusion coefficients of ozone in water (\(D_{\rm O_{2}}=1.74\cdot 10^{-9}\) m\({}^{2}\) s\({}^{-1}\)) and the mass transfer coefficients for ozone (\(k_{\rm L}=2.8\cdot 10^{-4}\) m s\({}^{-1}\)) were used for the calculation of the kinetic parameters of the reactions of ozone with the substrates under study. The diffusion coefficients of phenol, veratrol, veratric alcohol, and veratric aldehyde were calculated by the equation for the diffusion of the substances in the liquid.[11]
+
+The ozone concentration on the film surface was determined according to the Henry law
+
+\[[{\rm O_{3}}]^{\ast}=K[{\rm O_{3}}]_{\rm g}, \tag{7}\]
+
+where \(K\) is the solubility coefficient for ozone. The ozone concentration in the gas phase was determined by the equation
+
+\[[{\rm O_{3}}]_{\rm g}=([{\rm O_{3}}]_{\rm in}-[{\rm O_{3}}]_{\rm out})/\ln\frac {[{\rm O_{3}}]_{\rm In}}{[{\rm O_{3}}]_{\rm out}}. \tag{8}\]
+
+The data presented in Fig. 2 show that in a wide range of concentrations of various substrates ([B] \(\sim 10^{-4}\)--\(10^{-3}\) mol L\({}^{-1}\)) the absorption rate \(\widetilde{\nu}\) is independent of the substrate concentration. This indicates that the reaction rate is controlled by mass transfer and the process occurs in the diffusion region (\(E=1\)). In this case, the interface (\(a\)) can be determined by Eq. (1).
+
+The absorption rate \(\widetilde{\nu}\) increases with [B]\({}_{0}\) ([B] \(>5\cdot 10^{-3}\) mol L\({}^{-1}\)). In the framework of the film theory, this indicates a change in the regime in which the reaction occurs.[7] In this concentration interval the reaction occurs predominantly in the film, and the reaction rate constant can be determined by the ozone absorption rate using Eqs (1)--(4).
+
+To determine the rate constants for the studied compounds by the change in the substrate concentration, the experimental data were presented in the coordinates of Eq. (5) (Fig. 3). The \(k^{\prime}\) values for various substrates were determined from these dependences.
+
+To determine the rate constant from a change in the substrate concentration, one should know the ozone consumption per one substrate molecule (\(z\)). At the initial steps of the reactions of ozone with the model lignin compounds, unsaturated substances (derivatives of muconic and maleic acids) are formed and transformed with high rates into poorly oxidizable aliphatic products. Therefore, ozone consumption in secondary processes should be taken into account. Thus, the \(z\) parameter is the amount of ozone consumed to the transformation of an initial substrate molecule and rapidly oxidizable ozonation products. It was found from the experimental data that \(z\) remains virtually unchanged during ozonation. The ozone consumption was determined by the integration of the kinetic curve.
+
+The data presented in Table 1 show that the rate constants for the studied processes determined by different methods well agree. The rate constant for the reaction of ozone with guaiacol was not determined, because this
+
+\begin{table}
+\begin{tabular}{l c c c} Substrate & [B]\({}_{0}\)/mol L\({}^{-1}\) & \(k_{\rm O_{3}}\) & \(k_{\rm B}\) \\ \cline{3-4}  & & L mol\({}^{-1}\) s\({}^{-1}\) \\ \hline Phenol & \(2.9\cdot 10^{-4}\)–\(5.3\cdot 10^{-2}\) & \(6.8\cdot 10^{3}\) & \(5.7\cdot 10^{3}\) \\ Veratrol & \(1.1\cdot 10^{-5}\)–\(3.8\cdot 10^{-2}\) & \(1.6\cdot 10^{4}\) & \(1.5\cdot 10^{4}\) \\ Veratric alcohol & \(3.1\cdot 10^{-4}\)–\(1.2\cdot 10^{-2}\) & \(3.9\cdot 10^{4}\) & \(4.4\cdot 10^{4}\) \\ Veratric aldehyde & \(3.6\cdot 10^{-5}\)–\(1.8\cdot 10^{-2}\) & \(4.1\cdot 10^{3}\) & — \\ Veratric acid & \(3.0\cdot 10^{-3}\)–\(9.0\cdot 10^{-3}\) & \(7.7\cdot 10^{3}\) & — \\ Guaiacol & \(3.0\cdot 10^{-4}\)–\(1.0\cdot 10^{-3}\) & \(>5\cdot 10^{5}\) & — \\ \end{tabular}
+\end{table}
+Table 1: Rate constants of the reactions of ozone with model compounds (\(k_{\rm O_{2}}\) and \(k_{\rm B}\) are the rate constants calculated from the ozone absorption and the change in the substrate concentration, respectively)process proceeds in the "instant" reaction regime. The rate constant was estimated from the criterion of the "instant" character of the reactions.7
+
+The commonly accepted approach to the determination of a relationship between the structure and reactivity of organic compounds is the application of the Hammett equation12 that relates the reaction rate constant \(k\) and the substituent constant \(s\). The latter takes into account the influence of the inductive (_I_) and mesomeric (_M_) effects on the electron density of the aromatic ring
+
+\[\log k = \rho \sigma + \log k_{0}.\]
+
+The plot of log_k vs_ substituent constant is shown in Fig. 4.
+
+The electron-withdrawing substituents, _viz._, aldehyde and acid groups, with negative \(I\) and \(M\) effects decrease the electron density on the aromatic ring, thus decreasing the rate constant of the reaction with such an electrophile as ozone. The electron-donor alcohol group with a small positive \(I\) effect increases the electron density and reactivity. The negative value of the \(r\) rate constant confirms the electrophilic nature of ozone in the reactions with compounds of the veratryl series.
+
+**Ozonolysis scheme.** The nature and position of substituents in the aromatic ring play a key role in the ozonolysis mechanism and strongly affect both the process rate and composition of the oxidation products.13
+
+The products of ozonation of phenol, guaiacol, veratrol, veratric acid, veratric aldehyde, and veratric alcohol were analyzed using HPLC. The chromatograms of aqueous solutions (pH 1.5) of veratrol and veratric alcohol are shown in Figs 5 and 6, respectively.
+
+The major products of veratrol ozonation are mono- and dimethyl muconates and hydroquinone derivatives; the products of guaiacol ozonation are monomethyl muconate, muconic acid, and hydroquinone derivatives; the products of phenol ozonation are muconic acid and hydroquinone.
+
+There is virtually no product peak in the case of oxidation of veratric aldehyde and veratric acid by ozone. Small amounts of veratric acid was found in the products of the ozone reaction with veratric aldehyde. Most likely, the ozonolysis rate of veratrol with electron-withdrawing substituents is much smaller than the rate of transformation of its oxidation products.
+
+The derivatives of muconic acid and its esters and minor amounts of veratric aldehyde and veratric acid were identified in the products of veratric alcohol ozonation.
+
+In all chromatograms there are nonidentified peaks with short retention times belonging, most likely, to the cleavage products of muconic acid and its derivatives. These can be maleic, oxalic, formic, and other acids and their esters.
+
+The mechanism of the reaction of ozone with the model compounds, namely, guaiacol, veratrol, and its derivatives, was proposed on the basis of the data obtained.
+
+Ozone interacts with the aromatic ring according to the electrophilic addition mechanism,5,12,13 by attacking the positions with enhanced electron density, namely, the carbon atoms linked with the hydroxy and methoxy groups. Upon the ozone attack onto the _p_-system of the aromatic ring (Scheme 1), charge-transfer complex **1** (_p_-complex) is formed and further transformed into a _s_-complex, _viz._, carbocations **2** or **3** (the latter can be formed only if there is no substituents in position 4, at X = H).
+
+Fig. 4: Rate constants of the reaction of ozone with dimethoxytoluene5 (_1_), veratric alcohol (_2_), homovaratric alcohol5 (_3_), veratric acid (_4_), and veratric aldehyde (_5_) _vs_ constant of the substituent (_σ_).
+
+Fig. 5: Chromatogram of an aqueous solution of veratrol upon ozonation: \(1\), hydroquinone or its derivatives; \(2\) and \(3\), monomethyl muconate; \(4\), veratrol; \(5\), dimethyl muconate.
+
+Fig. 6: Chromatogram of an aqueous solution of veratric alcohol upon ozonation: \(1\) and \(2\), hydroquinone or its derivatives; \(3\), veratric alcohol; \(4\), veratric acid; \(5\) and \(6\), muconic acid derivatives; \(7\), veratric aldehyde.
+
+The arbocations formed as intermediates rapidly undergo further transformations.[12] The most probable route is the transformation of s-complex **2**_via_ the mechanism proposed by Criegee[14] (Scheme 2, path _A_) with the cleavage of the C--C bond and the formation of dimethyl (for veratrol) and monomethyl (for guaiacol) muconates in agreement with the experimental data.
+
+Most likely, s-complex **2** can decompose without molozoindle formation (see Scheme 2, path _B_). This results in the formation of methyl alcohol and hydrogen peroxide, and the products are monomethyl muconate (for veratrol) and muconic acid (for guaiacol). The veratrol derivatives react with ozone by an analogous mechanism with aromatic ring opening.
+
+Hydroquinone and its derivatives that could be formed from s-complex **3** were found in the ozonolysis products of phenol, guaiacol, and veratrol. The reaction can proceed with the evolution of methanol and oxygen (Scheme 3).
+
+The proposed mechanism (Schemes 1--3) is based on the earlier[5,15] developed concepts on the mechanism of the reactions of ozone with the phenolic and non-phenolic moieties of lignin. The authors of these studies[5,15] assume, on the basis of the experimental data on determination of radicals (OH\({}^{*}\) radicals, superoxide radicals), that the reaction of ozone with guaiacol proceeds _via_ the mechanism of electrophilic addition of ozone to the substrate to form zwitterion **2**. The next step is the homolytic cleavage of the O--O bond, resulting in superoxide radicals and quinones. The feasibility of homolytic cleavage was confirmed by the quantum chemical and thermochemical methods. The heterolytic cleavage of the O--O bond or the formation of molozoindle from the zwitterion to yield muconic acid derivatives is also assumed,[15] but this route is not predominant.
+
+As shown above (see Figs 5 and 6), no quinones were found in the ozonolysis products of the studied substrates (HPLC analysis). This can be due to the fact that in an acidic medium (pH 1.5) the zwitterion is protonated and the main directions of its destruction are the heterolytic cleavage of the O--O bond (see Scheme 2, route _B_) or molozoindle formation (route _A_).
+
+The presence of veratric aldehyde and veratric acid in the ozonation products of veratric alcohol and veratric aldehyde indicates that the aromatic ring and the lateral group can be attacked by ozone.4 However, the major direction of the reactions of the model compounds with ozone is aromatic ring opening.
+
+Thus, the relationship was found between the structure of the studied compounds and their reactivity. The rate constants for the reactions of ozone with compounds of the veratryl series were determined by the Hammett equation. The structures of the guiaacyl type were shown to transform upon ozonation with higher rates than the structures of the veratryl type. The dependence of the rate constant on the nature of substituents in the aromatic ring was determined. The introduction of electron-donor substituents increases the electron density on the aromatic ring and, hence, increases the ozonation rate constant. The stability of the aromatic ring in the reactions with ozone is enhanced upon the introduction of an electron-withdrawing group. It was shown on the basis of the chromatographic data that the reactions of ozone with guiaicol, veratrol, and its derivatives could occur with aromatic ring opening or hydroxylation and with the oxidation of the lateral groups. The mechanism was proposed for the reactions of ozone with the compounds modeling the structural lignin fragments in an acidic medium.
+
+## References
+
+* [1] P. S. Bailey, _Ozonation in Organic Chemistry_, Academic Press, New York, 1982, Vol. **2**.
+* [2] M. Massatoshi, S. Tasuro, N. Yoshitish, S. Hiroshi, _J. Chem. Soc. Jpn, Chem. Ind. Chem._, 1986, **4**, 545.
+* [3] B. Ferron, J. P. Croue, M. Dore, _Ozone Sci. Eng._, 1995, **17**, 687.
+* [4] H. Kaneko, S. Hosoya, J. Nakano, _Mokuzai Gakkaishi_, 1981, **27**, 678.
+* [5] M. Ragnar, T. Eriksson, T. Reitberger, P. Brandt, _Holyforschung_, 1999, **53**, 423.
+* [6] M. M. Ksenofontova, A. N. Mitrofanova, E. A. Tveritinova, A. N. Pryakhin, V. V. Lunin, _Zh. Fiz. Khim._, 2003, **77**, 1028 [_Russ. J. Phys. Chem._, 2003, **77** (Engl. Transl.)].
+* [7] P. V. Danckwerts, _Gas-Liquid Reactions_, McGrow-Hill Book Company, New York, 1970, 256 pp.
+* [8] H. Benbelkacem, H. Debellefontaine, _Chem. Eng. Proc._, 2003, **42**, 723.
+* [9] F. J. Beltran, E. M. Rodriguez, M. T. Romeo, _J. Hazard. Mater._, 2006, **138B**, 534.
+* [10] H. Benbelkacem, H. Cano, S. Mathe, H. Debellefontaine, _Ozone Sci. Eng._, 2003, **25**, 18.
+* [11] R. C. Reid, J. M. Prausnitz, T. K. Sherwood, _The Properties of Gases and Liquids_, McGrow-Hill Book Company, New York, 1987, 741 pp.
+* [12] H. Becker, _Einfuhrung in die Elekronentheorie Organisch-Chemischer Reaktionen_, VEB Deutscher Verlag der Wissenschaften, Berlin, 1974.
+* [13] G. A. Galstyan, N. F. Tyupalo, S. D. Razumovskii, _Ozon i ego reaktsii s aromaticheskimi soedinenjyaml v zhikoi faze [Ozone and Its reactions with Aromatic Compounds in the Liquid Phase]_, Vostochnoukr. National University, Lugansk, 2003, 298 (in Russian).
+* [14] R. Criegee, _Record Chem. Progr._, 1957, **18**, 111.
+* [15] M. Ragnar, T. Eriksson, T. Reitberger, _Holyforschung_, 1999, **53**, 292.

manual_annotation/200pdfs_in_mmd/10.1007_s11270-006-9304-x.mmd

@@ -0,0 +1,170 @@
+EDTA Leaching of Cu Contaminated Soils Using Ozone/UV for Treatment and Reuse of Washing Solution in a Closed Loop
+
+Metka Udovic, Domen Lestan
+
+# Abstract
+
+Ozone and UV irradiation were used for oxidative decomposition of EDTA-Cu complexes in washing solution obtained during multi-step leaching of Cu (344,1 +- 36.5 mg kg-1) contaminated vineyard soil with EDTA as a chelant. The released Cu was absorbed from the washing solution on a commercial mixture of metal absorbing minerals, and the treated washing solution then reused for removal of soil residual Cu-EDTA complexes in a closed-loop process. Six consecutive leaching steps (6 x 2.5 mmol kg-1 of EDTA) removed 38.8 % of Cu from soils, and reduced Cu soil mobility, determined using the toxicity characteristic leaching test (TCLP), by 28.5%. The final washing solution obtained after soil remediation was colourless, with a pH close to neutral (7.5 +- 0.2) and with low concentrations of Cu and EDTA (0.51 +- 0.22 mg L-1 and 0.083 mM, respectively). The proposed remediation method has therefore potential not just to recycle and save process water, but also not to produce toxic wastewaters. Soil treatment did not substantially alter the soil properties determined by pedological analysis, and had relatively little impact on soil hydraulic conductivity and soil water sorption capacity.
+
+## Introduction
+
+Pollution of soils with toxic heavy metals has been an unfortunate by-product of industrialisation and modern agronomic practices. The major release of Cu into the land is from tailings and overburdens from copper mines, and the application of fungicides and sewage sludges. Cu is an essential element for organisms. However, at high concentrations Cu becomes toxic to plants, leading to necrosis, chlorosis, leaf discoloration and inhibition of root growth. At the cellular level, Cu inhibits photosynthesis, respiration, interferes with fatty acid, protein and pigment synthesis and with nitrogen fixation processes (Lindon & Henriques, 1992; Lou, Shen, & Li, 2004; van Assche & Clijsters, 1990). With its known antifungal and algicidal properties, elevated levels of Cu in soil could adversely affect microbially mediated soil processes.
+
+A number of stringent regulations have been established to limit the levels of toxic metals in the environment. However, the cleanup of heavy-metal contaminated sites remains highly challenging and costly. One of the permanent solutions is soil washing with solutions containing acids or chelants. Acids dissolve carbonates and other heavy metal bearing soil material and exchange heavy metals from soil colloids, while chelants desorb heavy metals from soil solid phases by forming strong and water-soluble metal-ligand coordination compounds (complexes). Since acidic solutions could cause deterioration of soil physicochemical properties, using chelants is considered to be environmentally less disturbing (Xu & Zhao, 2005).
+
+Technically, soil-washing techniques comprise soil extraction, flushing or leaching. Soil extraction refers to batch treatment of soil slurry in a reactor, followed by soil filtration. Soil flushing is an in situ soil washing technique applicable to specific soil condition in which the contaminated zone is underlain by non-permeable materials, which allows pumping and treating of washing solution. In soil leaching, the washing solution is gravitationally percolated through a soil heap or column ex situ. It is operationally simple and has potential for the economical treatment of large amounts of soil.
+
+We recently introduced a novel chelant-based soil leaching method with treatment and reuse of washing solution in a closed process loop (Fig. 1). We applied advanced oxidation processes (AOP), specifically the combination of ozone and UV, for oxidative decomposition of chelant-metal complexes in the washing solution, followed by metal absorption. AOP systems are based on the in situ generation of hydroxyl radicals (OH) under mild process conditions. *OH are powerful oxidising agents, capable of degrading a great variety of otherwise recalcitrant organic compounds (Hapeman & Torrents, 1998). Using 2.5 mmol kg-1 ethyl-enediamine tetraacetate (EDTA) as a chelant for a six-step leaching of Pb (1,243 mg kg-1) and Zn (1,190 mg kg-1) contaminated soil removed close to 50 and 20% of initial soil Pb and Zn, with almost no emission of EDTA, Pb or Zn into the environment (Finzgar & Lestan, 2006a). For leaching of Cu (412 mg kg-1) contaminated soil (the soil was from the same contaminated site as the soil used in the present study) we used [S,S] izomere of ethylene diaminedisuccinate ([S,S]-EDDS) as a chelant. Single-dose soil leaching with 5 mmol kg-1 [S,S]-EDDS removed 47.5% of Cu, with insignificant concentrations of Cu and [S,S]-EDDS in the waste washing solution (Finzgar & Lestan, 2006b). However, [S,S]-EDDS is a prohibitively expensive chelant. In soil remediation, EDTA has received much more attention because of its relatively low cost and ability to form strong soluble complexes with most soil contaminating metals. Chaney et al. (2000) reported that the price of technical grade EDTA obtained from a major US manufacturer was 4.3 US$ kg-1.
+
+The overall objective of this study was to evaluate the feasibility of using EDTA for Cu soil leaching using ozone/UV for treatment of the washing solution in a closed process loop. The study comprised:
+
+* Estimation of basic process parameters for effective soil leaching. The required EDTA reaction time and efficiency of multi-step vs. single dose soil leaching were determined in small-scale experiments.
+* Bench-scale soil column leaching study. We assessed the efficiency of the proposed remediation method by determining the percentage of Cu removal and reduction in soil Cu mobility, and by measuring residual concentrations of Cu and EDTA in the final washing solution (wastewater).
+
+Fig. 1: Flowsheet of EDTA soil leaching method with ozone/UV treatment and reuse of the washing solution in a closed process loop * The impact of the proposed remediation method on soil quality. This we assessed by determining changes in chemical and physical soil parameters.
+
+## Materials and Methods
+
+### Soil Samples and Pedological Analysis
+
+Soil samples were collected from the 0-20 cm surface layer of a vineyard in the southwestern part of Slovenia. Soil Cu contamination (344.1 +- 36.5 mg kg-1) exceeded the warning and critical limits, set at 50 and 140 mg kg-1, respectively, in EU countries (Council Directive 86/278/EEC, 1986).
+
+Soil pH was measured in a 1:2.5 (w/v) ratio of soil and 0.01 M CaCl2 water solution suspension. Soil samples were analyzed for organic matter by Walkley-Black titrations, cation exchange capacity by ammonium acetate method, soil texture by the pipette method, easily extractable P was determined colorimetrically according to the Egner-Doming method, and carbonates manometrically after soil reaction with HCl (Kalra & Maynard, 1991). The soil texture was silty-clay loam. Soil properties determined before and after remediation are summarized in Table 1.
+
+### Selected Soil Physical Properties
+
+Soil water sorption capacity was determining by inserting soil samples, sieved through a 2 mm sieve, into retaining rings placed on a ceramic plate and irrigated with deionized water for 48 h. The ceramic plate with soil sample was then placed in an extractor-pressure vessel for another 48 h. Negative pressures of 0.33 bar and 15 bar, defining soil field water capacity and wilting point, respectively, were applied. Samples were weighed and then dried for 24 h at 105degC and weighed again to determine the mass percent of water sorbed. Three replicates of the measurements were made.
+
+To determine soil hydraulic conductivity, a soil column 10 cm in length (L) and 7.4 cm in diameter was packed uniformly with dry soil (soil density 0.88 g cm-3) and slowly saturated with water. The column was oriented horizontally. Water was supplied to the inlet of the column at a constant matric potential of 70 cm H2O (H2). Water was collected at
+
+\begin{table}
+\begin{tabular}{c c c c c} Soil properties & Before remediation & After remediation \\
+0–7 cm & 7–14 cm & 14–21 cm \\ _Pedological analysis_ \\ pH (CaCl2) & 7.3 & 7.5 & 7.4 & 7.4 \\ Organic matter (\%) & 3.9 & 3.7 & 3.6 & 3.6 \\ P2O5 (mg 100g−1) & 33.4 & 35.9 & 34.8 & 35.4 \\ CO3−2 (\%) & 20.5 & 22.5 & 25.9 & 22.6 \\ CEC (mmol C+ 100g−1) & 34 & 33.8 & 35.5 & 36 \\ Sand (\%) & 11.8 & 7.5 & 6.5 & 5.0 \\ Silt (\%) & 53.3 & 56.4 & 56.8 & 60.6 \\ Clay (\%) & 34.9 & 36.1 & 36.7 & 34.4 \\ _Selected physical properties_ \\ Hydraulic conductivity (K × 103 m s−1) & a3.3 ± 0.4 & b−1.2 ± 0.8 & ab−2.2 ± 2.6 & a2.8 ± 1.8 \\ Field water capacity (–0.33 bar) & a36.5 ± 0.3 & b37 ± 0.3 & ab36.7 ± 0.2 & ab36.8 ± 0.2 \\ Wilting point (–15 bar) & a18.7 ± 0.8 & b24.8 ± 1.4 & b24.4 ± 0.1 & b24.5 ± 0.2 \\ _Selected metals_ \\ Cu (mg kg−1) & a344.1 ± 36.5 & b293 ± 6 & c235.4 ± 9.3 & e224.4 ± 9.1 \\ Cu TCLP (mg l−1) & a0.14 ± 0.014 & b0.13 ± 0.005 & d0.08 ± 0.005 & \textasciitilde{}0.09 ± 0.002 \\ Ca (mg kg−1) & a80254 ± 778 & ab3127 ± 1035 & b86360 ± 2914 & b86951 ± 4769 \\ Fe (mg kg−1) & a20038 ± 263 & a20158 ± 812 & ab19077 ± 890 & b18601 ± 573 \\ Na (mg kg−1) & a264 ± 4 & b494 ± 6 & b491 ± 12 & b508 ± 11 \\ \end{tabular}
+\end{table}
+Table 1: Selected properties of Cu contaminated soil before and after remediation the outflow end of the system, which was maintained at a matric potential of 40 cm H2O (H1). These flow conditions were imposed for 24 h to ensure that the system was at a steady-state before water flow (q) through the column was measured. Twelve replicates of the measurements were made. Saturated hydraulic conductivity (K) was calculated using a derivation of Darcy's law (Eq. 1) for isotropic soils (Dane, Jalbert, & Hopmans, 2002):
+
+\[q = {K[(H_{2} - H_{1} )/L]}\]
+
+Soil physical properties determined before and after remediation are summarized in Table 1.
+
+### The Effect of EDTA Reaction Time on Cu Leaching Efficiency
+
+Soil (150 g - small-scale experiment) was placed in perforated 250 mL polypropylene flasks with 0.5 mm plastic mesh at the bottom to retain the soil. The soil was leached in triplicates with 180 mL of 2.5 mmol kg-1 EDTA disodium salt (un-buffered). Washing solution was circulated through the soil for 72 h using a peristaltic pump (flow rate 1.2-1.5 mL min-1). Every 12 h, the sample of washing solution was collected, filtrated and the pH and Cu concentration measured.
+
+### Multi-step vs. Single-dose Soil Leaching
+
+Soil (150 g - small-scale experiment) was placed in 250 mL flasks as described above. The soil was treated in triplicates with 180 mL of 2.5, 5 and 10 mmol kg-1 EDTA disodium salt (un-buffered) as four-step, two-step and one-step (single-dose) leaching, respectively. In each leaching step, the washing solution was circulated through the soil for 24 h (reaction time) using a peristaltic pump (flow rate 1.2-1.5 mL min-1) and subsequently rinsed with 1.5 L of tap water (flow rate 1.2-1.5 mL min-1) to remove all chelant-heavy metal complexes. Washing and rinsing solutions were collected, filtrated and analysed for Cu. Cu soil removal was calculated from the Cu concentration in the solutions.
+
+### Soil Leaching Using Ozone/UV for Washing Solution Treatment
+
+EDTA leaching of Cu contaminated soils using ozone/UV for the treatment and reuse of the washing solution in a closed loop was simulated in a bench-scale study. The flow sheet of the proposed remediation method is shown in Fig. 1.
+
+Air-died soil (4.6 kg) was sieved through a 5-mm mesh sieve and placed, in triplicates, in 15 cm diameter soil columns 27 cm high. Plastic mesh (D=0.2 mm) at the bottom of the column retained the soil. The soil was treated in a six-step leaching experiment. The washing solution (2.5 mmol kg-1 EDTA in 2.4 L of un-buffered tap water) was first circulated solely through the soil column for 24-48 h (reaction time) using a peristaltic pump (flow rate 12 mL min-1). The washing solution was then circulated via a collection vessel (Fig. 1). The washing solution was pumped from the collection vessel at high flow rate into a washing solution treatment unit and back, as described below. The pH and Cu concentration in the washing solution were periodically measured. When the Cu concentration in the washing solution fell below 25 mg L-1 (after 18 h of treatment), we started a new leaching step with a new dosage of 2.5 mmol kg-1 EDTA. During the experiment, the washing solution was supplemented with tap water to compensate for the water lost during the process (approx. 10% in each leaching step).
+
+After the final, sixth leaching step, the soil from the upper (0-9 cm), middle (9-18 cm) and bottom (18-27 cm) sections of the column was collected, air-dried, homogenized, sampled and stored at 4degC before analysis.
+
+### Washing Solution Treatment Unit
+
+The washing solution treatment unit for oxidation of EDTA complexes and metal recovery consisted of a collection vessel, an ozone generator, ozonation flask (oxygen/ozone flow rate 0.15 L min-1), UV-light source in a continuous flow housing, and a metal absorption filter. A peristaltic pump was used to force the washing solution from the collection vessel (flow rate 75 mL min-1) through the unit and back, as indicated in Fig. 1.
+
+Ozone was produced in an ozone generator (V-4, Crystal Air, Surrey, British Columbia, Canada) from pure commercial oxygen (flow rate 0.45 L min-1). The ozone concentration was determined by the indigo colorimetric method (Eaton, Clesceri, & Greenberg, 1995). Ozonation with a porous oxygen/ozone sparger allowed a concentration of ozone in tap water up to 14.2 +- 1.2 mg L-1.
+
+A 320 mm long 8 W UV light bulb (MK-8, Lenttech, Delft, The Netherlands) was installed in a quartz glass and stainless steel continuous flow housing.
+
+Metals released after advanced oxidation of their EDTA complexes with ozone/UV were removed by passing washing solution through the absorption filter with 24 g of the commercial sorbent Slovakia (IPRES, Bratislava, Slovak Republic). The Slovakite (mixture of natural raw materials: dolomite, diatomite, smectite basaltic tuff, bentonite, alginite and zeolite) was replaced after each leaching step.
+
+### Cu Mobility
+
+The mobility of Cu in the soils before and after remediation was determined using TCLP analyses (US EPA, 1995), conducted in triplicate. The procedure involved shaking a 10 g soil sample in 200 ml of 0.0992 M acetic acid and 0.0643 M NaOH with a pH of 4.93 +- 0.05, for 18 h on a rotary shaker at about 300 rpm. At the end of the reaction period, the contents were filtered (Whatman No. 4 filter paper), acidified with concentrated HNO3 to pH < 2 and analysed for Cu.
+
+### Metals Determination
+
+Air-dried soils (3 g) were ground in an agate mill to pass through 0.16-mm mesh sieve. The samples were digested in aqua regia (100 mL) for 180 min. under reflux conditions and metals analyzed by AAS (Perkin-Elmer 1100-B, Norwalk, USA). Metals in the washing solution were determined by AAS directly. A standard reference material used in inter-laboratory comparisons (ALVA Boden 1) from the HBLFA Raumberg-Gumpenstein, Irdning, Austria, was used in the digestion and analysis as part of the QA/QC protocol. Reagent blank and analytical duplicates were also used where appropriate to ensure accuracy and precision in the analysis.
+
+### EDTA Determination
+
+EDTA in extractants was determined spectrophotometrically according to the procedure of Hamano, Mitsuhashi, Kojma and Aoki (1993).
+
+### Statistical Analysis
+
+The Duncan multiple range test was used to determine the statistical significance (P < 0.05) between different soil treatments, using the computer program Statgraphic 4.0 for Windows.
+
+## Results and Discussion
+
+### EDTA Reaction Time and Efficiency of Multi-Step Leaching - Small-Scale Experiments
+
+The removal of heavy metals from soil depends, among other factors, on the time allowed for the reaction to occur between the chelant and the metals. Chelant reaction time is especially important in soil leaching methods, in which the kinetics of heavy metal desorption/dissolution may be a more decisive factor than in soil extraction (Samani, Hu, & Hanson, 1998). As shown in Fig. 2, the concentration of Cu in the EDTA washing solution (2.5 mmol kg-1) increased with time. The washing solution pH also slightly increased with the reaction time. For practical reasons, we allowed a 24-48 h reaction time in further experiments.
+
+Fig. 2: Concentration of Cu and pH of washing solution after soil leaching with 2.5 mmol kg−1 EDTA using different reaction times. Error bars represent standard deviation from the mean value (_n_ = 3). Letters denote significantly different Cu concentrations according to the Duncan test (_P_ < 0.05)Four-step leaching with 2.5 mmol kg-1 EDTA removed approximately 15% more Cu than single-dose leaching with 10 mmol kg-1 EDTA (Fig. 3). A multi-step leaching approach was therefore used in the further evaluation of the proposed remediation method. Another benefit of using multi-step leaching was that lower EDTA doses generate a Cu washing solution with a less intensive blue-green colour. Intensive colouration of washing solution could block the penetration of UV irradiation, and hence prevent generation of *OH and degradation of EDTA complexes.
+
+The phenomenon, that multi-step leaching with lover EDTA doses is generally more effective than single-dose leaching was previously observed for soils contaminated with Pb and Zn (Finzgar & Lestan, 2007).
+
+### EDTA Soil Leaching Using Ozone/UV for the Washing Solution Treatment
+
+Based on the results obtained with the small-scale experiments described above, 6-step leaching with 2.5 mmol kg-1 EDTA was used in soil remediation using ozone/UV for treatment and reuse of washing solution in a closed loop. The concentration of Cu in the washing solution decreased with each leaching step, since a part of the Cu weakly bound to the soil fractions was extracted first (Fig. 4). We therefore prolonged the EDTA reaction time from the 24 h that we used in the first three leaching steps, to 48 h in the last three steps. During each leaching step, the concentration of Cu in the washing solution gradually decreased with time (Fig. 4), as Cu-EDTA complexes were washed from the soil and degraded in the washing solution treatment unit. The released Cu was permanently removed from the system by absorption. This was visually observed as complete decolouration of the washing solution, which was brightly green-blue at the beginning of each leaching step. The pH of the washing solution was slightly alkaline, within the range of 7.25-8.45. The concentration of Cu and EDTA in the final (waste) washing solution at the end of the sixth cycle was 0.51 +- 0.22 mg L-1 and 0.083 mM, respectively. The solution was colourless, with pH 7.5 +- 0.2, and presumably safe to discharge. Treating and reusing washing solution in a closed process loop enabled soil remediation in a continuous process mode. This is a significant advantage over current batch soil washing remediation methods, in which soil and washing (extraction) solutions are treated separately.
+
+The time (18 h) required to remove the Cu from the washing solution in each of the leaching steps was rather long (Fig. 4). This was presumably because of the use of a low-power UV lamp (8 W). Shu and Chang (2005), for example, reported that UV light intensity was the single most important operating parameter that affected the degradation of azo-dyes in
+
+Fig. 4: Concentration of Cu in washing solution during soil six-step column soil leaching (I–VI) with 2.5 mmol kg−1 EDTA, using ozone/UV treatment to reuse of washing solution in a closed process loop. Error bars represent standard deviation from the mean value (_n_ = 3)
+
+Fig. 3: Cu removal after multi-step soil leaching, using three different EDTA concentrations. Error bars represent standard deviation from the mean value (_n_ = 3). Letters denote significantly different treatments according to the Duncan test (_P_ < 0.05)a laboratory scale UV/H2O2 photoreceptor (H2O2 was used as a source of *OH instead of O3). They used a medium pressure mercury UV lamp, which allowed a total input electric power up to 5,000 W. Similarly, Ku, Wang and Shen (1998) studied the decomposition of EDTA in aqueous solution by UV/H2O2 and reported that the EDTA mineralization rate increased with increasing light intensity. Theoretically, the higher the UV power, the faster is the formation of *OH radicals. Another possible reason for the long treatment time is the reported recalcitrance of Cu-EDTA complexes to oxidative degradation by ozone/UV (Yang et al., 2005). Of all complexes with metals (Fe, Al, Pb, Na, Zn) Cu-EDTA had the lowest mineralization rate constant.
+
+Six-step leaching with total 15 mmol kg-1 of EDTA removed 38.8 % of Cu from the contaminated soil. We had used soil from the same contaminated site as in a previous study. Single dose leaching with 5 mmol kg-1 [S,S]-EDDS removed almost half of the initial Cu soil content (Finzgar & Lestan, 2006b). [S, SJ-EDDS was therefore, surprisingly, a much more efficient Cu chelant than EDTA. Theoretically, a higher stability constant of complex formation (Log Ks) indicates higher efficiency of chelants to extract metals. The Log Ks for Cu-EDTA (18.78 at 25degC and ionic strength u = 0.1) is slightly higher than Log Ks for Cu-[S,S]-EDDS complex formation (18.5 at 25degC m = 0.1) (Martell & Smith, 2003). It seems, therefore, that various soil factors can change chelant efficiency. Tandy et al. (2004), for example, also reported that [S,S]-EDDS efficiency in extracting Cu from the soil surpassed the efficiency of EDTA and other more common (and much cheaper) chelants. On the other hand, Lee and Kao (2004) reported that EDTA quite efficiently removed Cu from two contaminated soils and was better chelant than diethylenetriamine pentaacetate (DTPA) and citric acid.
+
+TCLP was used to assess Cu soil mobility before and after remediation (Table 1). On average, TCLP extractable Cu was reduced by only 28.5%. However, the concentration of TCLP extractable Cu was already low in the non-treated soil.
+
+Significantly less Cu was removed and Cu mobility reduced in the upper section of the soil column than in the middle and bottom sections (Table 1). A possible reason was application of the washing solution to the centre and not to the entire surface of the soil. Visually, the soil surface looked equally saturated. Nevertheless, such an application of the washing solution could create conditions for preferential flow through the upper soil section.
+
+### Effect of Proposed Remediation Method on Soil Properties
+
+Soil leaching is probably the least destructive of soil physical properties of all soil-washing techniques. Soil extraction, for example, involves stringent physical treatment that completely destroys the soil structure, rendering the soil useless as a medium for plant growth. However, in all chelant-based remediation methods, common soil cations, particularly Fe and Ca, compete with contaminating metals to form complexes with chelants (Kim, Lee, & Ong, 2003). This may lead to their removal from the soil. Ca is the dominant exchangeable cation and well known to play a crucial role in soil aggregation stability by forming cation bridges between negatively charged soil colloid particles. Removal of Ca can cause the breakdown of soil aggregates (soil structure) and thus affect the size and configuration of soil pores, which in turn reduces the soil hydraulic conductivity and water sorption capacity. Quirk (1986), for example, observed that the hydraulic conductivity of soils was highly dependent on Ca soil concentration.
+
+Using the proposed remediation method did not remove Ca from the soil (Table 1). In fact, the Ca concentration even slightly increased down the soil profile. Presumably some Ca was brought into the system with the tap water that was used to prepare the washing solution. Fe concentration slightly decreased down the soil profile (Table 1). A small part of Fe was apparently complexed and leached. These results can be presumably explained by the fact that Ca2+ forms less stable EDTA complexes (log Ks 10.65 at 25degC, u = 0.1) than competing ions: Cu2+ (18.78 at 25degC, u = 0.1), Fe3+ (log Ks 25.1 at 25degC, u = 0.1) and Fe2+ (log Ks 14.3 at 25degC, u = 0.1) (Martell & Smith, 2003). The origin of the elevated Na content in the treated soil (Table 1) was the EDTA, which was applied as a disodium salt. Na is known to replace Ca and cause dis-aggregation and swelling of soil colloids. A surplus of Na in soil could therefore result in lower hydraulic conductivity of soil.
+
+Saturated hydraulic conductivity was measured in 12 replicates due to the high sensitivity of the measurement to the uniformity (isotropy) of the soil samples, which is in turn very difficult to achieve. The difference in hydraulic conductivity between untreated soil (K=0.131 +- 0.02 cm s-1) and the upper layer of treated soil (K=0.049 +- 0.03 cm s-1) was statistically significant (Table 1). The soil hydraulic conductivity is known to vary by several magnitudes within each soil texture class (Dane et al., 2002). The observed difference might therefore not be really meaningful.
+
+Soil water sorption capacity was measured at two soil matric potentials, -0.33 and -15 bar. As shown in Table 1, the field water capacity of the treated soil did not change substantially. This indicates that soil leaching with EDTA did not significantly change the structure or share of macrophages and capillary phones, in which gravitational water is stored. The increase in wilting point water percentage after soil treatment was more significant (Table 1), but difficult to explain. At wilting point, the soil water remains in thin films around individual soil particles. The dominant effective factor is therefore soil texture, which remained relatively unchanged after soil treatment (Table 1). Other soil properties, determined by standard pedological analysis, remained relatively unchanged after soil leaching with EDTA (Table 1).
+
+## Conclusions
+
+The following conclusions can be drawn from our study:
+
+* For the Cu contaminated soil used in our study, multi-step leaching was slightly more effective than single-dose leaching with the same amount of EDTA as a chelant.
+* EDTA leaching of Cu contaminated soils using ozone/UV for treatment and reuse of washing solution in a closed loop is a technically feasible remediation method, but constrained by the efficiency of EDTA for Cu removal, which seems to be soil specific. In our laboratory set-up, the irrigation system was not properly designed to allow uniform and effective leaching throughout the column. This, however, could be less of a problem and easy solvable in pilot or full-scale operation.
+* The proposed remediation method is a continuous process, recycles and therefore uses less process water and has the potential of near-zero emissions into the environment. These are its most important strengths. The wastewater was almost Cu and EDTA free, with acceptable pH. Ozone could be captured at the exhaust of the system and quenched, for example by passing through a KJ solution.
+* Finally, EDTA leaching of Cu contaminated soil did not extensively alter the soil properties and proved to be a "soil friendly" remediation method. This too, however, could be soil-specific.
+
+###### Acknowledgements.
+
+ This work was supported by the Slovenian Ministry for Education, Science and Sport, grant J4-6134-0481-04/4.03.
+
+## References
+
+- Chaney et al. (1986) Chaney, R. L., Brown, S. L., Li, Y.-M., Angle, J. S., Stuczynski T. I., Daniels, W. L., et al. (2000). _Progress in risk assessment for soil metals, and in-situ remediation and phytoextraction of metals from hazardous contaminated soils_ (Paper presented at the US-EPA's conference Phyter-mediation, State of the Science, Boston) (May)
+* Council Directive (1986) Council Directive 86/278/EEC (1986). _On the protection of the environment, and in particular of the soil, when sewage sludge is used in agriculture_. Brussels: EC Official Journal L181.
+* Dane et al. (2002) Dane, J. H., Jalbert, M., & Hopmans, J. W. (2002). Hydraulic conductivity. In R. Lal (Ed.), _Encyclopedia of soil science_ (pp. 667-670). New York, NY: Marcel Dekker.
+* Eaton et al. (1995) Eaton, A. D., Cleseri, L. S., & Greenberg, A. E. (1995). _Standard methods for the examination of water and wastewater, 19th ed_. Washington DC: American Public Health Association.
+* Elliot & Brown (1989) Elliot, H. A., & Brown, G. A. (1989). Comparative evaluation of NTA and EDTA for extractive decontamination of Pb-polluted soils. _Water Air and Soil Pollution_, _45_, 361-369.
+* Finzgar & Lestan (2006a) Finzgar, N., & Lestan, D. (2006a). Heap leaching of Pb and Zn contaminated soil using ozone/ UV treatment of EDTA extractants. _Chemosphere_, _63_, 1736-1743.
+* Finzgar & Lestan (2006b) Finzgar, N., & Lestan, D. (2006b). Heap leaching of Cu contaminated soil with S[S]-EDDS in a closed process loop. _Journal of Hazardous Materials_, _B135_, 418-422.
+* Finzgar & Lestan (2007) Finzgar, N., & Lestan, D. (2007). Multi-step leaching of Pb and Zn contaminated soils with EDTA. _Chemosphere_, _66_, 824-832.
+* Hamano et al. (1993) Hamano, T., Mitsuhashi, Y., Kojima, N., & Aoki, N. (1993). Sensitive spectrophotometric method for the determination of ethylene-diaminetetraacetic acid in foods. _Analyst_, _118_, 909-912.
+* Hapeman & Torrents (1998) Hapeman, C. J., & Torrents, A.,(1998). Direct radical oxidation processes. In K. Kearney & T. Roberts (Eds.), _Pesticide remediation in soils and water_ (pp. 161-180). New York, NY: Wiley.
+* Kalra & Maynard (1991) Kalra, Y. P., & Maynard, D. G. (1991). _Methods manual for forest soil and plant analysis_. Edmonton: Canadian Forest Service, Northern Forestry Centre.
+
+Kim, C., Lee, Y., & Ong S-K. (2003). Factors affecting EDTA extraction of lead from lead-contaminated soils. _Chemosphere_, _51_, 845-853.
+* [Ku et al.1998] Ku, Y., Wang, L-S., & Shen, Y-S. (1998). Decomposition of EDTA in aqueous solution by UV/H2O2 process. _Journal of Hazardous Materials_, _60_, 41-51.
+* [Lee and Kao2004] Lee, C. S., & Kao, M. M. (2004). Effects of extracting reagents and metal speciation on the removal of heavy metal contaminated soils by chemical extraction. _Journal of Environmental Science and Health Part A_, _39_, 1233-1249.
+* [Lindon and Henriques1992] Lindon, F. C., & Henriques, F. S. (1992). Copper toxicity in rice: Diagnostic criteria and effect on tissue Mn and Fe. _Soil Science_, _154_, 130-135.
+* [Lou et al.2004] Lou, L. Q., Shen, Z. G., & Li, X. D. (2004). The copper tolerance mechanisms of _Elsholtizia hiaichowensis_, a plant from copper-enriched soils. _Environmental and Experimental Botany_, _51_, 111-120.
+* [Martell and Smith2003] Martell, A. E., & Smith, R. M. (2003). _NIST critically selected stability constants of metal complexes; Version 7.0_. Gaithersburg, MD: NIST.
+* [Notermann1999] Notermann, B. (1999). Biodegradation of EDTA. _Applied Microbiology and Biotechnology_, _51_, 751-759.
+* [Quirk1986] Quirk, J. P. (1986). Soil permeability in relation to solicity and salinity. _Philosophical Transactions of the Royal Society of London Series A_, _316_, 297-317.
+* [Samani et al.1998] Samani, Z., Hu, S., & Hanson, A. T. (1998). Remediation of lead contaminated soil by column extraction with EDTA: II. modeling. _Water, Air, and Soil Pollution_, _102_, 221-238.
+* [Shu and Chang2005] Shu, H-Y., & Chang, M-C. (2005). Pilot scale annular flow photoreactor by UV/H2O2 for the decolorization of azo dye wastewater. _Journal of Hazardous Materials_, _125_, 244-251.
+* [Sun et al.2001] Sun, B., Zhao, F. J., Lombi, E., & McGrath, S. P. (2001). Leaching of heavy metals from contaminated soils using EDTA. _Environmental Pollution_, _113_, 111-120.
+* [Tandy et al.2004] Tandy, S., Bossart, K., Mueller, R., Ritschel, J., Hauser, L., Schulin, R., et al. (2004). Extraction of heavy metals from soils using biodegradable chelating agents. _Environmental Science and Technology_, _38_, 937-944.
+* [US EPA1995] US EPA (1995). _Laboratory manual physical/ chemical methods, 3rd ed._ Washington DC: US Government Printing Office.
+* [Vasnsche and Clijsters1990] van Assche, F., & Clijsters, H. (1990). Effect of metals on enzyme activity in plants. _Plant, Cell and Environment_, _13_, 195-206.
+* [Xu and Zhao2005] Xu, Y., & Zhao, D. (2005). Removal of copper from contaminated soil by use of poly(amidoamine) dendrimers. _Environmental Science & Technology_, _39_, 2369-2375.
+* [Yang et al.2005] Yang, C., Xu, Y. R., Teo, K. C., Goh, N. K., Chia, L. S., & Xie, R. J. (2005). Destruction of organic pollutants in reusable wastewater using advanced oxidation technology. _Chemosphere_, _59_, 441-445.

manual_annotation/200pdfs_in_mmd/10.1016_j.apcatb.2009.08.019.mmd

@@ -0,0 +1,184 @@
+A comparison between catalytic ozonation and activated carbon adsorption/ozone-regeneration processes for wastewater treatment
+
+P.M. Alvarez, F.J. Beltran, F.J. Masa, J.P. Pocostales
+
+Departamento de Ingenieria Quimica, Universidad de Extremadura, Badajoz 06071, Spain
+
+# Abstract
+
+Two methods based on the use of granular activated carbon (GAC) and ozone to remove organic compounds from water have been investigated. Both methods have been applied to degrade an aqueous solution of gallic acid and a secondary effluent from a wastewater treatment plant (WWTP). One of the methods, namely catalytic ozonation, implies simultaneous ozonation and adsorption onto GAC. This process takes advantage of the oxidizing power of ozone and the adsorption capacity of GAC but also of the catalytic transformation of ozone into secondary oxidants on the GAC surface. The efficiency of catalytic ozonation was compared to those of single adsorption and single ozonation. It was found that the catalytic process highly improves the conversion of total organic carbon (TOC) and makes a more efficient use of ozone than the single ozonation process. To illustrate the reusability of the catalyst, the GAC was reused four times through a series of consecutive experiments. No loss of catalytic activity was observed when treating the WWTP effluent but some deactivation could be appreciated when treating the aqueous solution of gallic acid. This deactivation could be attributed to some porosity destruction and surface oxidation produced as a result of reactions of aqueous ozone on the GAC surface. The other method investigated is an adsorption-regeneration process (namely GAC/O2-regeneration) that comprises two steps: dynamic adsorption onto GAC and further regeneration of the spent GAC with gaseous ozone. The adsorption stage of the GAC/O2-regeneration experiments was carried out in a continuous flow adsorption column and breakthrough curves were obtained. It was observed that the GAC used in this work adsorbed gallic acid very efficiently but exhibited limited capacity to remove chemical oxygen demand (COD) from the WWTP effluent. The optimum ozone dose to regenerate the spent GAC after gallic acid adsorption was found to be about 0.4 g O2/g GAC, with results showing around 90% regeneration efficiency. As a result of incomplete regeneration, the GAC adsorption capacity progressively decreased with the number of adsorption-regeneration cycles. The GAC/O2-regeneration method was not successful at treating the WWTP effluent as low adsorption uptake was observed. Moreover, the GAC became damaged after regeneration because of excessive oxidation of its surface.
+
+## Introduction
+
+Since Jans and Hoigne [1] the integrated use of ozone and activated carbon has been extensively investigated as an effective advanced oxidation process (AOP) to remove toxic and/or low biodegradable organic compounds from water. The process not only takes advantage of the high adsorption capacity of activated carbon and the high oxidant power of ozone but activated carbon may also transform ozone into secondary oxidants, such as hydroxyl radicals, which can degrade adsorbed and aqueous organic compounds, eventually converting them into carbon dioxide and water (i.e., mineralization). The method is considered as a catalytic ozonation process and it has already been successfully applied to effectively mineralize a number of aqueous organic compound families [2,3]. Particularly, in a recent series of works, we used this method to degrade some polyphenols typically present in some food-processing wastewaters [4-6]. Textural and chemical properties of the activated carbon surface as well as aqueous pH were found to be key variables of the process.
+
+The removal of many organic compounds from aqueous solution can be easily accomplished by simply adsorption onto activated carbon. However, the suitability of using an activated carbon in an adsorption application depends not only on its uptake capacity but also on the possibility of reusing it several times. The most common available technology to regenerate spent activated carbons is thermal reactivation, though it has some drawbacks as large cost of regeneration facilities, high energy demand and loss of carbon due to oxidation and attrition [7]. Therefore, the research and development of other alternative regeneration methods, which are better to be realized in situ in order to avoid shipment cost, are desirable. Inthis sense, ozone-assisted regeneration of spent activated carbon might be considered. The method implies the adsorption of water pollutants onto activated carbon for a long period of time up to saturation and subsequent in situ regeneration of the spent activated carbon by reaction with gaseous ozone during a short time. This method has been already investigated to regenerate activated carbons exhausted with phenol and benzothiazole [8,9].
+
+Considering the ideas addressed above, this research focuses on the comparison between the one-step catalytic ozonation process using granular activated carbon (GAC) as a catalyst and the two-step process consisting of dynamic adsorption onto GAC followed by ozone-assisted regeneration of the spent adsorbent (i.e., GAC/O3-regeneration). Experiments were conducted on an aqueous solution of gallic acid, a phenolic compound typically present in some food-processing wastewaters, and a secondary effluent taken from a real wastewater treatment plant (WWTP) that treats a mixture of some food-processing and domestic wastewaters. The efficacies of the two processes are assessed in terms of organic compounds mineralization, ozone consumption and GAC reusability.
+
+## Materials and methods
+
+### Materials
+
+The GAC used in this study, supplied by Sigma-Aldrich, was Darco 12-20 mesh. It is produced by Chemviron Carbon (Belgium) by activation of bituminous coal with steam. The as-received GAC was boiled in distilled water for 1 h, washed repeatedly with ultrapure water (Millipore Milli-Q system), dried at 110 degC for 12 h, and stored in a desiccator at room temperature until use. Textural characterization of GAC samples was accomplished by physisorption of nitrogen at 77 K (Autosorb-1, Quantachrome) and by mercury porosimetry (Autoscan-60, Quantachrome). The BET and a-plot methods were applied to adsorption data to derive the BET surface area and micropore volume (_V_1), respectively, whereas the volume of pores of size larger than 3.5 nm (_V_2) was obtained from mercury porosimetry measurements. The surface chemistry of GAC samples was analyzed by measuring the concentrations of acidic and basic surface oxygen groups (SOG) following the Boehm's titration method [10]. The point of zero charge (PZC) was analyzed by mass titration [11]. The ash content was determined gravimetrically after combustion of GAC samples in air at 850 degC.
+
+Analytical grade gallic acid (3,4,5-trihydroxybenzoic acid, C7O3H6H6H2O) was purchased from Sigma-Aldrich. Stock solutions of this compound (0.5 g/dm3, COD ~ 540 g/m3, TOC ~ 225 g/m3) were prepared in ultrapure water (Millipore Milli-Q system) and the pH was adjusted to 6 by adding 0.1 M sodium hydroxide. A secondary effluent was collected from a full-scale WWTP located in Almedralejo (Badajoz, Spain). This WWTP receives domestic wastewater as well as wastewater from several food-processing local industry facilities including wineries, distilleries, fruit and vegetable processing industries and olive oil mills. The treatment at the WWTP consists of a screening and primary sedimentation followed by a secondary activated sludge process. To avoid samples heterogeneity linked to seasonal variations of WWTP effluent, all the samples used in this work were collected at the same time in September 2006. Immediately after collection, the samples were shipped to the laboratory, analyzed and kept frozen in PET bottles until being used in experiments. Aqueous pH and conductivity were analyzed by means of a Radiometer Copenhagen pH-meter (HPM82) and Hanna HI9033 conductivity-meter, respectively. Total and suspended solids were measured gravimetrically according to Standard Methods [12]. Chemical oxygen demand (COD) was analyzed by the colorimetric dichromate method using a Dr Lange cuvette test. Biological oxygen demand (BOD5) was followed by the respirometric method (Oxitop(r) WTW system) using an activated sludge sample from the WWTP as inoculum. A TOC-VCSH Shimadzu carbon analyzer was used to measure total organic carbon (TOC), inorganic carbon (IC) and total nitrogen (Nr). Polyphenols were analyzed by the Folin-Ciocalteau method and expressed as equivalent gallic acid concentration [13]. Phosphate and ammonium concentrations were measured using Merck Spectroquant photometric kits. UV absorbance at 254 nm was determined on diluted samples to provide an estimation of the content of olefins and aromatic compounds, which, as a rule, react fast with ozone. A Thermo Spectronic HeAlos a spectrophotometer and 1 cm quartz cells were used for the measurements.
+
+### Adsorption isotherms
+
+Equilibrium adsorption isotherms at 25 degC of gallic acid aqueous solution and WWTP effluent on GAC were generated using the bottle point method described elsewhere [14]. Different amounts of GAC were weighted and placed in bottles containing 20 cm3 of either gallic acid aqueous solution or WWTP effluent. The bottles were capped and kept in a shaking thermostatic bath for time enough to achieve adsorption equilibrium. Preliminary experiments showed that equilibrium was reached within less than 1 week. Upon equilibration, the solutions were filtered through 0.45 mm membranes and analyzed for either gallic acid concentration or COD and TOC. Gallic acid was analyzed by HPLC with a Hewlet Packard series 1110 chromatograph provided with an UV detector set at 280 nm. A Trazer Kromasil-100 (C18; 15 cm x \(A\) cm, 5 mm) column was used, the mobile phase being a mixture of water-acetonitrile-phosphoric acid (90:9:1, v/v/v). COD and TOC were analyzed by the methods cited above. The amount of solute adsorbed per gram of adsorbent at equilibrium (_q_e) was derived from the mass-balance equation (1):
+
+\[q_e = \frac{(C_{0} - C_e) \cdot V}{w}\]
+
+where \(C\)0 and _C_e are the initial and equilibrium aqueous solute concentrations, respectively (i.e., gallic acid concentration, COD or TOC), \(V\) is the volume of solution and \(w\) stands for the dry mass of GAC.
+
+### Catalytic ozonation experiments
+
+Catalytic ozonation experiments were performed at room temperature in semi-batch mode using an experimental set-up as that depicted in Fig. 1, Part A. The glass bubble column, which had a diameter of 5 cm and length 20 cm, was first loaded with 250 cm3 of either gallic acid aqueous solution or WWTP effluent. Then, a 25 N dm3/h of about 40 mg O3/dm3 ozone-oxygen mixture was produced in the ozone generator (Sander, model 301.7) and supplied to the bubble column through a porous plate situated at its bottom. The recirculation pump (Masterflex peristaltic, Cole-Parmer Instrument) produced a 2 dm3/h aqueous flow rate through the adsorption column (2 cm i.d.), which held 2 g of GAC unless otherwise specified. The concentration of ozone in the gases entering and leaving the bubble column was monitored with an Anseros Ozomat GM-6000 Pro analyzer. Aqueous samples were withdrawn from the bubble column at different times during the course of each experiment and analyzed for dissolved ozone, gallic acid, COD and TOC. Aqueous ozone concentration was determined by the indigo method [15]. For comparative purposes, blank experiments, either without GAC or without ozone, were also conducted.
+
+### GAC/O3-regeneration experiments
+
+GAC/O3-regeneration experiments were carried out at room temperature in an experimental device as that schematically shown in Fig. 1, Part B. A complete treatment cycle involved successive dynamic adsorption and GAC regeneration stages. For the dynamic adsorption stage (see Fig. 1, Part B1), 1 cm3/min flow rate of either gallic acid aqueous solution or WWTP effluent was continuously pumped from the magnetically stirred reservoir to the small-scale adsorption column (2 cm i.d.), which was packed with 2 g of fresh or regenerated GAC. Aqueous samples were taken at different times from the outlet of the column and analyzed for either gallic acid concentration or COD to obtain the breakthrough curve. Once GAC saturation was reached, the pump was stopped and the regeneration stage began. For that, the adsorption column containing the spent GAC was used as an ozonation chamber (see Fig. 1, Part B2). The spent GAC was first partly dried in flowing air and thereafter exposed to a 25 N dm3/h continuous flow of an oxygen-ozone mixture for 1 h. The concentration of ozone in the gaseous streams at the ozonation chamber inlet and outlet was continuously monitored.
+
+To investigate the nature of compounds adsorbed onto GAC, thermal analysis (TGA) of some spent and regenerated samples was carried out by heating about 200 mg of GAC samples in nitrogen flow (9 N m3/h) from room temperature up to 1173 K at a rate of 10 K/min using a Mettler TA-3000 thermobalance. Before TGA, the GAC samples were oven-dried at 110 degC for 2 h.
+
+## Results and discussion
+
+### Adsorption isotherms onto virgin GAC
+
+Fig. 2A shows the measured isotherm for gallic acid adsorption onto the virgin GAC. It can be included in the type L of the Giles classification which is characterized by an initial rapid uptake followed by the attainment of a plateau [16]. This type of isotherm has been found by a number of researchers when studying the adsorption equilibrium of different phenolic compounds onto activated carbon [17]. According to the reported in those studies, the adsorption mechanism is most likely due to dispersive forces between \(\overline{\mu}\) electrons of the aromatic ring of the gallic acid molecule and \(\overline{\mu}\) electrons of the graphene layers of the activated carbon. From Fig. 2A it is also apparent that the experimental data fit the Langmuir isotherm (Eq. (2)) better than the Freundlich isotherm (Eq. (3)). The parameters of the Langmuir model were estimated from the experimental results of Fig. 2A after applying a non-linear regression method (see the first row of Table 1 for fitting results). It is worthy to note that the maximum adsorption capacity, \(q_{m}\), and the affinity constant, \(K_{L}\), were found to be slightly higher than the values reported for the adsorption of gallic acid from aqueous solutions onto other GACs with higher surface areas than that of the used in this work [18,19]. This reveals that this particular GAC (Darco 12-40 mesh) can be a good choice for the removal of gallic acid from aqueous solutions.
+
+\[q_{e}=q_{m}\frac{K_{L}\cdot C_{e}}{1+K_{L}\cdot C_{e}} \tag{2}\]
+
+\[q_{e}=K_{F}\cdot C_{e}^{1/n} \tag{3}\]
+
+Fig. 2B shows the measured adsorption isotherms for the adsorption of the WWTP secondary effluent onto the virgin GAC.
+
+\begin{table}
+\begin{tabular}{l c c c c c c} \hline GAC & \(D_{D_{0}}\) & \(n_{D_{0}}\) & \(PR\) & \(q_{m}\) & \(K_{L}\) & \(R^{2}\) \\ sample & (gO/gGAC) & (gO/g TOC) & (\%) & (mg/g) & (dm3/g) & \\ \hline Virgin & – & – & – & \(287\pm 6\) & \(54\pm 5\) & 0.994 \\ O\_{2}-Reg. 0.1 & 0.10 & 1.44 & 44.8 & \(192\pm 9\) & \(15\pm 2\) & 0.986 \\ O\_{2}-Reg.0 & 0.21 & 3.08 & 59.9 & \(207\pm 7\) & \(23\pm 3\) & 0.987 \\ O\_{2}-Reg.0 & 0.42 & 6.16 & 87.6 & \(20\pm 10\) & \(37\pm 4\) & 0.989 \\ O\_{2}-Reg.0 & 0.58 & 8.53 & 74.6 & \(268\pm 10\) & \(20\pm 2\) & 0.991 \\ \hline \end{tabular}
+\end{table}
+Table 1: Parameters of the Langmuir isotherm determined at 25 °C for the adsorption of gallic acid onto virgin and some ozone-regenerated GACs.
+
+Figure 1: Schematic diagrams of the experimental devices for catalytic ozonation and GAC/O\({}_{3}\)-regeneration processes.
+
+The isotherms are of type S of the Giles classification [16]. Accordingly, negligible adsorption takes place at low equilibrium COD (or TOC) values but a drastic increase in the adsorption uptake occurs from a critical COD (or TOC) value up to reach the maximum adsorption capacity. This result suggests that such a fraction of the organic matter originally present in the WWTP effluent has not affinity for the GAC surface and it cannot be adsorbed. In fact, it was observed that when the WWTP effluent was brought into contact with the GAC and allowed for equilibrium, no more than 60-70% of COD and TOC could be removed from the effluent regardless of the amount of GAC used.
+
+### Degradation of gallic acid
+
+#### 3.2.1 Catalytic oxzonation experiments
+
+Fig. 3 compares some gallic acid concentration and TOC profiles obtained from single ozonation, single adsorption onto GAC and catalytic ozonation experiments. It can be observed that the degradation rate of gallic acid was quite similar when ozonation was carried out in the presence and the absence of GAC but the removal of gallic acid by single adsorption was much slower. Thus, the percentage of gallic acid removed after 2 h of treatment was about 75% by single adsorption whereas by single or catalytic ozonation gallic acid was completely degraded in less than 20 min. However, the removal of TOC by single adsorption was faster than by single ozonation likely because of the accumulation in water of intermediates of the ozone-gallic acid reaction. The most remarkable finding in Fig. 3 is that TOC removal was greatly enhanced by the simultaneous use of ozone and GAC. Thus, gallic acid was almost completely mineralized in 90 min by catalytic ozonation, whereas single ozonation barely produced a 50% TOC removal within this treatment time.
+
+The results of Fig. 3 concur with those previously reported using a slurry reactor and a GAC with surface area around 1000 m2/g [4]. They can be explained on the basis of a fast reaction between ozone and gallic acid but a relatively slow adsorption onto GAC. Accordingly, regardless of the presence of GAC, most of gallic acid reacted with ozone close to the gas-liquid interface in a fast kinetic regime of ozone absorption. The absence of aqueous ozone during the first 10 min of our experiments (see Fig. 3), when gallic acid still remained in solution, confirms the fast kinetic regime of ozone absorption. As a rule, the ozonation of aqueous gallic acid gives rise to the formation of a number of intermediates (e.g. ketomalonic and oxalic acids) which are difficult to degrade by single ozonation [4]. Because of that, TOC could not be removed to a great extent by single ozonation. It is well known that the O3/GAC catalytic system produces surface and aqueous free radicals, primarily hydroxyl radicals. These radicals arise mainly from the decomposition of ozone on basic sites of the GAC as chromene and pyrone-like structures and delocalized \(\pi\) electrons of the basal planes [20-22]. Accordingly, as ozone decomposed over GAC, the concentration of ozone in water during the catalytic ozonation experiment was lower than that during the single ozonation
+
+Figure 3: Degradation of gallic acid by single adsorption, single ozonation and catalytic ozonation experiments. Reaction conditions: \(T\) = 20 °C; pH = 6; aqueous solution volume = 250 cm3; initial gallic acid concentration = 0.5 g/d2; recirculation flow rate = 2 d/m3; GAC weight (if applied) = 2 g; gas flow rate = 25 h dm3/h; ozone concentration at the gas inlet = 40 g O3/m3. Symbols: (■), (■), adsorption experiment: (■), (■), single ozonation experiment: (■), (■), single ozonation experiment: (■), (■), catalytic ozonation experiment: (Symbols: solid acid concentration; open symbols: TOC; (<centred symbols: aqueous ozone concentration.
+
+Figure 2: Adsorption isotherms at 25 °C for gallic acid (A) and WWTP effluent (B) on the virgin GAC.
+
+experiment (see Fig. 3) but hydroxyl radicals generated from ozone decomposition over GAC, unselectively oxidized the organic compounds in solution converting them into CO2. As a result, TOC degradation was greatly enhanced by the presence of the GAC.
+
+Eq. (4) was used to estimate an ozone consumption parameter (\(\eta_{{\text{O}}_{3}}\)), that can be defined as the average amount of ozone consumed per unit mass of TOC removed in a given reaction time (_t_):
+
+\[\eta_{{\text{O}}_{3}} = \frac{{{\text{F}}_{g} \cdot \int_{0}^{{\text{e}}_{f}} \left( {{\text{C}}_{{\text{O}}_{{\text{1}}},{\text{i}}} - {\text{C}}_{{\text{O}}_{{\text{1}}},{\text{n}}} \right) \cdot {\text{d}}t}}}{{V \cdot \left( {{\text{T}}_{{\text{O}}_{{\text{0}}}} - {\text{T}}_{{\text{C}}_{{\text{1}}},{\text{f}}} } \right)}}\]
+
+where _F_g is the gas flow rate; _C_O_3,\(i\) and _C_O_3,\(n\) are the concentrations of ozone at the reactor inlet and outlet gas streams, respectively; and V is the volume of aqueous solution. Our results suggest a more efficient use of ozone when the GAC was present. Thus, \(\eta_{{\text{O}}_{3}}\) decreased from 13.1 g O3/g TOC in the single ozonation experiment to 7.4 g O3/g TOC in the catalytic ozonation experiment.
+
+To investigate catalyst deactivation behaviour, the GAC was reused in four consecutive catalytic ozonation experiments of degradation of aqueous gallic acid with a run time of 1.5 h. Fig. 4 shows the temporal profiles of normalized gallic acid concentration and TOC for these consecutive experiments. As can be seen from this figure, the removal rate of gallic acid was almost the same in all the experiments while TOC removal rate decreased to some extent after each GAC use. Thus, TOC degradation over the 1.5 h reaction period was about 95% with the fresh GAC and about 85% with the three-time reused. Moreover, \(\eta_{{\text{O}}_{3}}\) increased from 7.7 g O3/g TOC with the fresh GAC to 9.0 g O3/g TOC with the three-time reused. These results suggest slow deactivation of the GAC towards ozone transformation into hydroxyl radicals. Deactivation was likely due to changes in texture and/or surface chemistry of the GAC produced because of its progressive oxidation upon exposure to ozone [21]. To further clarify this point, a sample of virgin GAC and another of the GAC used in the four consecutive experiments (Used sample 1) were analyzed for some surface properties. From the results of Table 2, it can be seen that surface area, pore volume and ash content of the GAC did not change as significantly but noticeable changes of SOG concentrations and PZC were produced following repetitive experiments. According to data of Table 2, catalytic ozonation led to the removal of basic sites of the GAC such as chromene-like, pyrone-like and delocalized \(\pi\)-electrons of the basal planes and the fixation of acidic SOG such as carboxylic, lactone and hydroxyl groups. As a result of that, PZC decreased to some extent. Given the fact that basic sites are main responsible for ozone transformation into hydroxyl radicals, it can be concluded that the partial deactivation of the GAC towards TOC conversion was mainly due to the loss of GAC basic sites upon ozonation [22, 20, 21, 22]. In addition, the formation of acidic SOG on the entrance of micropores and the decrease of PZC might prevent the adsorption of organic compounds (i.e., gallic acid and ozonation by-products) onto the GAC to some extent, thereby contributing to decrease the overall effectiveness of the catalytic system.
+
+#### 3.2.2 GAC/O3-regeneration experiments
+
+First, a series of four GAC/O3-regeneration experiments were carried out to find the optimum dose of ozone required to regenerate the spent GAC. After each experiment, the gallic acid adsorption isotherm of a sample of the regenerated GAC was obtained. Fig. 5 shows the isotherms of the virgin GAC and those regenerated with different amounts of ozone. All isotherms of Fig. 5 could be fitted by the Langmuir equation to get the parameters listed in Table 1. The average efficiency of regeneration was calculated following a
+
+\begin{table}
+\begin{tabular}{c c c c c c c c} \hline GAC sample & Serr (m2/g) & \(V\)1 (cm3/g) & \(V\)2 (cm2/g) & Acidic SOC (quequiv./g) & Basic SOC (μequivv./g) & PZC & Ash (\%) \\ \hline Virgin sample & 710 & 0.16 & 0.51 & 792 & 196 & 6.3 & 16.2 \\ Used sample 1 & 685 & 0.14 & 0.54 & 924 & 123 & 5.5 & 15.9 \\ Used sample 2 & 643 & 0.11 & 0.59 & 1143 & 75 & 5.1 & 16.4 \\ Used sample 3 & 579 & 0.08 & 0.56 & 1478 & 44 & 4.7 & 15.1 \\ \hline \end{tabular}
+\end{table}
+Table 2: Some properties of virgin GAC and after its use in consecutive catalytic ozonation and GAC/O3-regeneration experiments to degrade either gallic acid or the WWTP effluent.
+
+method that compares the entire Langmuir isotherms of virgin and regenerated GACs. For that purpose, Eq. (5) was used:
+
+\[PR(\%)=\frac{q_{m,R}\cdot K_{LR}}{q_{m,V}\cdot K_{L,V}}\times\frac{1}{N}\sum_{i= 1}^{N}\frac{1+K_{L,V}\cdot C_{ei}}{1+K_{L,R}\cdot C_{ej}}\times 100 \tag{5}\]
+
+where the subscripts \(V\) and \(R\) refer to virgin and regenerated GACs, respectively. For the computation, 300 values of \(C_{ei}\) (i.e., \(N\) = 300) were taken regularly distributed from 1 to 301 mg/dm\({}^{3}\). Table 1 gives results for the amount of ozone consumed for GAC regeneration, expressed both per unit mass of GAC (\(D_{0_{3}}\)) and per unit mass of TOC adsorbed (\(\eta_{0_{3}}\)). Eqs. (6) and (7) were used to calculate these ozone consumptions:
+
+\[D_{0_{3}}=\frac{F_{g}\cdot\int_{0}^{t_{R}}(C_{0_{3},i}-C_{0_{3},\theta})\cdot dt }{W} \tag{6}\]
+
+\[\eta_{0_{3}}=\frac{D_{0_{3}}\times w}{2.24\cdot F_{1}\cdot\int\limits_{0}^{t_{R }}(C_{GA,i}-C_{CA,\theta})\cdot dt} \tag{7}\]
+
+where \(F_{g}\) is the gas flow rate; \(C_{0_{3},i}\) and \(C_{0_{3},\theta}\) are the concentrations of ozone at the ozonation chamber inlet and outlet, respectively, during the regeneration stage; \(w\) is the dry mass of GAC used in the experiments; _F1_ is the liquid flow rate; \(C_{CA,i}\) and \(C_{CA,\theta}\) are the concentrations of gallic acid at the adsorption column inlet and outlet, respectively, during the adsorption stage; _tA_ and _tB_ are the durations of the adsorption and regeneration stages, respectively. The coefficient 2.24 was used to convert the mass of gallic acid into the equivalent mass of carbon atoms. As shown in Fig. 5 and Table 1, the adsorption capacity of the GAC for gallic acid decreased after any GAC/O3-regeneration experiment. Nevertheless, sample O3-Reg-0.4 (i.e., \(D_{0_{3}}\) = 0.42 g O3/g GAC) exhibited larger regeneration efficiency (\(PR\)), maximum adsorption capacity (\(q_{m}\)) and adsorption affinity (\(K_{L}\)) than the other ozone-regenerated samples, suggesting that there is an optimum dose of ozone to recover most of the original GAC adsorption capacity for gallic acid. An attractive feature of the GAC/O3-regeneration experiment at optimum conditions is the efficient use of ozone. Thus, \(\eta_{0_{3}}\) was as low as 6.2 g O3/g TOC, which is even lower than the consumption measured in the catalytic ozonation experiment discussed in the previous section (i.e., 7.4 g O3/g TOC).
+
+Ozone-regeneration is intended to completely oxidize the adsorbed compounds without altering the porosity and surface chemistry of the GAC. However, experimental results in this work show that the process was not successful at the recovery of all the adsorption capacity of the virgin GAC. Regeneration efficiencies below 100% in Table 1 suggest: (i) failure of ozonation to remove all the adsorbates from the GAC surface and/or (ii) destruction of adsorptive sites and plugging of pores. In an attempt to further investigate the reasons of incomplete regeneration, some ozone-regenerated samples were analyzed for thermal desorption of adsorbed compounds and surface characteristics. Fig. 6 shows the differential thermogravimetric (DTG) profiles for the virgin GAC, a spent GAC saturated with gallic acid and three ozone-regenerated GACs. The small bands observed in the DTC curve of the virgin GAC can be due to thermal decomposition of different SOC [23]. Two peaks at about 150 and 300 deg C can be observed in the DTC curve of the spent GAC, which can be assigned to the release of physisorbed and chemisorbed gallic acid, respectively [24]. Both peaks can also be seen, though they are much weaker (note the change in _Y_-axis scale), in the DTG curve of the GAC regenerated with the lowest ozone dose (i.e., sample O3-Reg-0.1). These peaks are shifted towards lower temperatures in the DTC curves of samples O3-Reg-0.4 and O3-Reg-0.6. This result suggests that the amounts of gallic acid on these samples were lower than that in sample O3-Reg-0.1 but other adsorbed species and SOG were present there in greater extents. The peak at about 110 degC can be assigned partly to loss of trapped water while the peak centred at about 250 degC can be mainly due to the decomposition of carboxylic groups formed from the reaction of ozone with gallic acid and the GAC surface. Other peaks in ozone-regenerated samples attributable to ozonation products are the peaks at about 400-550 degC that can be associated with the thermal release of anhydrides and lactones and the broad peak centred at about 750 degC that may represent the decomposition of different groups, including phenolic groups. A peak intensity analysis in Fig. 6 leads one to think that sample O3-Reg-0.4 is cleaner from adsorbed products and SOG than the surface of the GAC regenerated with the highest dose (i.e., sample O3-Reg-0.6). Table 3 shows some parameters of the porous structure of the virgin and regenerated GACs as calculated from adsorption of nitrogen data. It is clearly seen that the regeneration with ozone, if not applied at optimum conditions, produced a significant decrease in the surface area and the volume of micropores (\(V_{1}\)). The volume of mesopores (\(V_{2}\)) of any regenerated GAC was larger than that of the virgin GAC, likely as a consequence of the widening of microporesity produced by ozone [25,26].
+
+Fig. 7 shows the gallic acid breakthrough curves generated from the results of a GAC/O3-regeneration experiment comprising four consecutive adsorption-regeneration cycles. Care was taken to keep the amount of ozone consumed at each regeneration stage
+
+\begin{table}
+\begin{tabular}{l c c c} \hline GAC sample & \(S_{\text{zerr}}\) (m\({}^{2}\)/g) & \(V_{1}\) (cm\({}^{2}\)/g) & \(V_{2}\) (cm\({}^{2}\)/g) \\ \hline Virgin sample & 710 & 0.16 & 0.51 \\ O\({}_{2}\)-Reg-0.1 & 502 & 0.05 & 0.58 \\ O\({}_{2}\)-Reg-0.4 & 697 & 0.14 & 0.62 \\ O\({}_{2}\)-Reg-0.6 & 601 & 0.10 & 0.65 \\ \hline \end{tabular}
+\end{table}
+Table 3: Some textural properties of virgin and some ozone-regenerated activated carbons.
+
+Figure 6: TGA plots for virgin, spent and some ozone-regenerated GACs.
+
+close to the optimum 0.42 g O\({}_{3}\)/g GAC. Fig. 7 reveals a progressive shift of the breakthrough curves to the left as far as the number of cycles was increased. This result is not surprising as incomplete GAC regeneration was expected to be achieved with ozone. The overall amount of gallic acid adsorbed onto the GAC during each dynamic adsorption stage decreased gradually as follows: 320 mg/g GAC (first cycle), 277 mg/g GAC (second cycle), 253 mg/g GAC (third cycle) and 240 mg/g GAC (fourth cycle). However, the initial breakthrough was almost coincident in the four dynamic adsorption stages at about 125 bed volumes. Considering the whole experiment (four cycles), the average specific ozone consumption was calculated to be 7.4 g O\({}_{3}\)/g TOC degraded. This figure is even lower than the average consumption of ozone in the four consecutive catalytic ozonation experiments (see Section 3.2.1) which was 8.2 g O\({}_{3}\)/g TOC degraded. A sample of the GAC used in the four-cycle experiment was analyzed for porous and chemical surface properties. The results are those in Table 2 corresponding to Used sample 2. From this table it can be seen that the GAC surface changed in the same way after the four consecutive O\({}_{3}\)/GAC experiments and the four-cycle GAC/O\({}_{3}\)-regeneration experiment: some destruction of porosity, fixation of acidic SOG, removal of basic SOG and decrease of PZC. However, it should be noted that these changes were more pronounced in the GAC/O\({}_{3}\)-regeneration experiment. They can explain the lowering in the GAC adsorption capacity for gallic acid. First, there is less surface available for adsorption. Second, the mentioned changes in SOG leads to removal of \(\pi\) electrons from the basal planes of carbon, thus weakening the dispersive interactions through which gallic acid adsorbs. Third, the decrease of PZC increases the polarity of the surface and favors the selectivity of adsorption towards water [26].
+
+### Degradation of a secondary effluent from a WWTP
+
+A complete characterization of the WWTP effluent used in this work is given in Table 4. It is worthy to note that it had a large content of organic matter as deduced from the high COD and TOC values, which means that the secondary treatment applied at the WWTP failed to achieve proper wastewater purification. Therefore, further treatment is needed before discharge or reuse.
+
+Fig. 8 shows some temporal profiles of normalized TOC during single ozonation, single adsorption and catalytic ozonation experiments treating the WWTP effluent. It should be stated here that the removal of COD in all these experiments (not shown) was not different from the removal of TOC. In Fig. 8, it can be clearly observed that the secondary effluent was very refractory to single ozonation as less than 25% TOC removal was achieved in 2 h of exposure to ozone. The removals of TOC and COD by single adsorption onto GAC were also limited to around 40%, which is in agreement with the equilibrium results shown in Fig. 2B. TOC and COD degradations were enhanced by the simultaneous use of ozone and GAC. Thus, for example, about 50% and 60% of TOC removals were achieved within 2 h in experiments carried out with 2 and 5 g of GAC, respectively. From Fig. 8 it is also apparent that the GAC did not suffer appreciably deactivation towards TOC degradation even after its third use, since the TOC profile did not vary with the repeated use of GAC. Regarding specific ozone consumption, in the single ozonation experiment it was found to be 18.8 g O\({}_{3}\)/g TOC degraded but it markedly decreased up to 9.6 g O\({}_{3}\)/g TOC (average value) in the catalytic ozonation experiments.
+
+A GAC/O\({}_{3}\)-regeneration experiment comprising three adsorption-regeneration cycles was carried out to treat the WWTP effluent. The amount of ozone consumed at each regeneration stage was about 0.35 g O\({}_{3}\)/g GAC. Fig. 9 shows the COD breakthrough curves.
+
+\begin{table}
+\begin{tabular}{l c} Parameter & Mean value \(\pm\) SD. \\ pH & 6.12 \(\pm\) 0.50 \\ Conductivity (m/S/cm) & 11.44 \(\pm\) 0.75 \\ Total suspended solids (g/m\({}^{3}\)) & 321 \(\pm\) 36 \\ Volatile suspended solids (g/m\({}^{3}\)) & 220 \(\pm\) 21 \\ COD (g/m\({}^{3}\)) & 522 \(\pm\) 55 \\ BODs (g/m\({}^{3}\)) & 150 \(\pm\) 100 \\ TOC (g/m\({}^{3}\)) & 171 \(\pm\) 12 \\ IC (g/m\({}^{3}\)) & 27 \(\pm\) 6 \\ Polyphenols (g/m\({}^{3}\))* & 26 \(\pm\) 5 \\ P-PO\({}^{a}\)* (g/m\({}^{3}\)) & 3.0 \(\pm\) 0.2 \\ N\({}_{\text{r}}\) (g/m\({}^{3}\)) & 16.8 \(\pm\) 1.1 \\ N-NH\({}^{a}\) (g/m\({}^{3}\)) & 14.0 \(\pm\) 0.8 \\ UV absorbance (254 nm)* & 0.40 \(\pm\) 0.03 \\ \end{tabular}
+
+* As gallic acid.
+* Measured on samples diluted five times.
+
+\end{table}
+Table 4: Quality summary of the WWTP effluent used in this work.
+
+Figure 8: Degradation of WWTP effluent by single adsorption, single ozonation and catalytic ozonation experiments. Reaction conditions: T = 20 °C; pH = 6; aqueous solution volume = 250 cm/s; initial TOC = 170 g/m\({}^{3}\); recirculation flow rate = 2 dm\({}^{3}\)/h; GAC weight (if applied) = 2 g; gas flow rate = 25 M/m\({}^{3}\); ozone concentration at the gas inlet +40 g O\({}_{3}\)/m\({}^{3}\) symbols: \(\Rightarrow\) adsorption experiment: \(\Leftrightarrow\), single ozonation experiment: \(\Leftrightarrow\), catalytic ozonation experiments: \(\blacktriangle\), GAC weight = 2 g, fresh GAC. \(\blacktriangledown\), GAC weight = 5 g, fresh GAC. \(\square\), GAC weight = 5 g, free acid.
+
+Figure 7: Gallic acid breakthrough curves from the dynamic adsorption stages of a four-cycle GAC-O\({}_{3}\)-regeneration experiment. Symbols: \(\square\), first cycle; \(\Diamond\), second cycle: \(\triangle\), third cycle: \(\triangledown\), fourth cycle.
+
+As can be seen, the column effluent had large COD values from the beginning of the adsorption stage. This result confirmed that a substantial fraction of COD of the WWTP effluent was not amenable to adsorption onto this GAC. The non-adsorbable COD fraction increased with the number of cycles, suggesting that the regeneration with ozone destroyed adsorption sites. Table 2 shows textural and chemical surface properties of a GAC sample after being used in this three-cycle experiment (Used sample 3). Destruction of porosity and modification of surface chemistry (i.e., fixation of acidic SOC, removal of basic SOC and decrease of P2C), because of ozone regeneration, are evident from results. All these findings suggest that only a minor fraction of the surface was covered by adsorbates in the spent GAC. Thus, at regeneration stages, ozone reacted mainly with the GAC itself producing loss of porosity and chemical oxidation of the surface making it more hydrophilic. As a consequence, ozone-regeneration induced changes in the GAC adsorption properties.
+
+## 4 Conclusions
+
+Catalytic ozonation using GAC as a catalyst is an AOP proved to be effective to degrade aqueous gallic acid and a secondary effluent from a full-scale WWTP. Major benefits of catalytic ozonation in comparison with single adsorption and single ozonation were faster degradation, higher degree of mineralization and better use of ozone. The GAC used in this work showed some minor deactivation towards gallic acid degradation with its repetitive use in ozonation experiments. Deactivation can be attributed to surface ozone reactions that gave rise to the fixation of acidic SOC and the removal of basic SOC. As a result, the GAC became more hydrophilic (lower P2C) thus affecting its adsorption behaviour and its ability to decompose aqueous ozone into secondary oxidants (i.e., hydroxyl radicals). No deactivation was observed after three uses of the GAC in catalytic ozonation experiments treating the WWTP effluent. Therefore, the treatment stands out as a significant candidate to be considered for full-scale operations.
+
+Dynamic adsorption onto GAC followed by ozone-regeneration of spent GAC (i.e., GAC/O\({}_{3}\)-regeneration process) can be a useful method to treat wastewater with high concentration of adsorbable compounds that react fast with gaseous ozone, as it is the case of gallic acid. However, even in this case, regeneration efficiencies lower than 90% were obtained at the conditions of this work. The GAC/O\({}_{3}\)-regeneration method failed to efficiently remove COD from the WWTP effluent as a substantial fraction of organic matter did not adsorb onto the GAC. The process also failed to regenerate the GAC as ozone destroyed its texture and reduced its surface basicity to significant extents.
+
+## Acknowledgements
+
+This work has been financed by the CICYT of Spain and the European Region Development Funds of the European Commission (Project CTQ2006/04745). Mr Pocostales also thanks the Spanish Ministry of Science and Education for a FPU grant.
+
+## References
+
+* (1) U. Jans, J. Hoigne. Ozone Sci. Eng. 20 (1998) 67-90.
+* (2) V. Fontanier, V. Farines, J. Albert, S. Baig, J. Molinier, Ozone Sci. Eng. 27 (2005) 115-128.
+* (3) A. Rodriguez, R. Rosal, J.A. Perdigon-Melo, M. Mezcu, A. Aglera, M.D. Hernando, P. Leton, A.R. Fernandez-Alba, E. Garcia-Calvo, Ozone-based technologies in water and wastewater treatment, Handbook of Environmental Chemistry, vol. 5 Part S7(2, Springer-Verlag, Berlin Heidelberg, Germany, 2008), pp. 127-175.
+* (4) F. Beltran, I. Ciriddez, J.F. Garcia-Araya, Appl. Catal. B: Environ. 63 (2006) 249-259.
+* (5) F. Beltran, I. Ciriddez, J.F. Garcia-Araya, Ind. Eng. Chem. Res. 47 (2008) 1058-1065.
+* (6) I. Ciriddez, J.F. Garcia-Araya, F. Beltran, Ind. Eng. Chem. Res. 46 (2007) 8241-8247.
+* (7) C. San Miguel, S.D. Lambert, N.D. Graham, Water Res. 35 (2001) 2740-2748.
+* (8) P.M. Alvarez, F. Beltran, V. Gomez-Serrano, J. Jaramillo, E.M. Rodriguez, Water Res. 38 (2004) 2155-2165.
+* (9) H. Valdes, C.A. Zuor, J. Hazard. Mater. 137 (2006) 1042-1048.
+* (10) H.P. Boehm, Carbon Sci. 23 (1994) 759-769.
+* (11) J.S. both, J.A. Schwarz, J. Colloid Inter. Sci. 130 (1989) 157-164.
+* (12) A.D. Eaton, L.S. Cleesert, E.W. Rice, A.E. Greenberg (Eds.), Standard Methods for the Examination of Water & Wastewater, 21st edition, APHA, AWWA, WEF, Baltimore, MD, USA, 2005.
+* (13) D.B. Now, Water Res. 17 (1983) 511-525.
+* (14) J.W. Weber (Ed.), Physicochemical Processes, Wiley, New York, 1972.
+* (15) H. Bader, J. Hoigne, Water Res. 15 (1981) 449-456.
+* (16) CH. Giles, T.H. MacFwan, S.N. Nakhwa, D. Smith, J. Chem. Soc. 786 (1960) 3973-3993.
+* (17) A. Dabrowski, P. Podkoscieiny, Z. Hubicki, M. Barczak, Chemosphere 58 (2005) 1049-1070.
+* (18) E. Utrera-Hidalgo, C. Moreno-Castillo, J. Rivera-Utrilla, M.A. Ferro-Garcia, F. Carasco-Martin, Carbon 30 (1992) 707-111.
+* (19) P.A. Garcia-Araya, F. Beltran, P. Alvarez, F.J. Mass, Adsorption 9 (2003) 107-115.
+* (20) P.M. Alvarez, J.F. Garcia-Araya, F.J. Beltran, I. Ciriddez, J. Jaramillo, V. Gomez-Serrano, Carbon 44 (2006) 3102-3112.
+* (21) M. Sanchez-Polo, U. Von Gunten, J. Rivera-Utrilla, Water Res. 39 (2005) 3189-3198.
+* (22) P.C.F. Faria, J.J.M. Orfao, M.F.R. Pereira, Ind. Eng. Chem. Res. 45 (2006) 2715-2721.
+* (23) L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.M. Orfao, Carbon 37 (1999) 1379-1389.
+* (24) J. Rivera-Utrilla, M.A. Ferro-Garcia, C. Moreno-Castillo, I. Bautista-Toledo, J.P. Poly. J. Chem. Soc. Faraday Trans. 91 (1995) 3213-3217.
+* (25) H. Valdes, M. Sanchez-Polo, J. Rivera-Utrilla, C.A. Zuor, Langmuir 18 (2002) 2111-2116.
+* (26) P.M. Alvarez, J.F. Garcia-Araya, F.J. Beltran, F.J. Mass, F. Medina, J. Colloid Interf. Sci. 283 (2005) 503-512.
+
+Figure 9: WWTP effluent breakthrough curves from the dynamic adsorption stages of a three-cycle GAC-O\({}_{3}\)-regeneration experiment. Symbols: \(\square\), first cycle; \(\bigcirc\), second cycle: \(\triangle\), third cycle.

manual_annotation/200pdfs_in_mmd/10.1016_j.apcatb.2015.11.010.mmd

@@ -0,0 +1,196 @@
+Accepted Manuscript
+
+Title: Dramatic coupling of visible light with ozone on honeycomb-like porous g-C\({}_{3}\)N\({}_{4}\) towards superior oxidation of water pollutants
+
+Author: Jiadong Xiao Yongbing Xie Faheem Nawaz Yuxian Wang Penghui Du Hongbin Cao
+
+PII: S0926-3373(15)30246-0
+
+DOI: [http://dx.doi.org/doi:10.1016/j.apcatb.2015.11.010](http://dx.doi.org/doi:10.1016/j.apcatb.2015.11.010)
+
+Reference: APCATB 14372
+
+To appear in: _Applied Catalysis B: Environmental_
+
+Received date: 3-10-2015
+
+Revised date: 3-11-2015
+
+Accepted date: 9-11-2015
+
+Please cite this article as: Jiadong Xiao, Yongbing Xie, Faheem Nawaz, Yuxian Wang, Penghui Du, Hongbin Cao, Dramatic coupling of visible light with ozone on honeycomb-like porous g-C3N4 towards superior oxidation of water pollutants, Applied Catalysis B, Environmental [http://dx.doi.org/10.1016/j.apcatb.2015.11.010](http://dx.doi.org/10.1016/j.apcatb.2015.11.010)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+Dramatic coupling of visible light with ozone on honeycomb-like porous g-C\({}_{3}\)N\({}_{4}\) towards superior oxidation of water pollutants
+
+Jiadong Xiao, Yongbing Xie, Fabeem Nawaz, Yuxian Wang, Penghui Du, Hongbin Cao, hbcao@ipe.ac.cn
+
+###### Abstract
+
+The proposed method of the proposed method is a novel method to estimate the absorption coefficient of the water column. The proposed method is able to obtain the absorption coefficient of the water column. The proposed method is able to obtain the absorption coefficient of the water column.
+
+## Abstract
+
+The proposed method is based on the concept of a "generalized stochastic process" (GPD). The proposed method is based on the concept of a "generalized stochastic process" (GPD).
+
+## 1 Introduction
+
+The \(\mathrm{O_{3}}\)/\(\mathrm{O_{3}}\
+
+## 1 Accepted Manuscript
+
+## 1 Introduction
+
+The utilization of solar energy based on semiconductor photocatalysts for the conversion of pollutants into CO\({}_{2}\) and H\({}_{2}\)O is one of the best solutions to solve environmental problems [1]. To date, various visible-light responsive materials, including TiO\({}_{2}\)[2], Ag\({}_{3}\)PO\({}_{4}\)[3], BiVO\({}_{4}\)[4], CdS [5], Bi\({}_{2}\)WO\({}_{6}\)[6], Ag@AgCl [7], C\({}_{60}\)[8] and graphitic carbon nitride (g-C\({}_{3}\)N\({}_{4}\)) [9-11] have been developed and their photocatalytic performance for water decontamination has been investigated. Among them, metal-free g-C\({}_{3}\)N\({}_{4}\) has drawn worldwide concern for its naturally narrow bandgap of 2.7 eV, permitting it to directly absorb visible light to drive chemical reactions [12]. g-C\({}_{3}\)N\({}_{4}\) can be easily obtained through direct polymerization of cheap feedstocks and possesses high thermal and chemical stability due to its tri-s-triazine ring structure [12-13]. Nevertheless, pristine bulk g-C\({}_{3}\)N\({}_{4}\) exhibits low photocatalytic efficiency owing to its low surface area and high recombination rate of photogenerated electrons and holes. Nearly 90% of charge carriers can rapidly recombine within 10 ns [14], which remarkably limits the photocatalytic activity of bulk g-C\({}_{3}\)N\({}_{4}\). Hence, it is highly necessary to increase the active surface area of g-C\({}_{3}\)N\({}_{4}\) and to promote the separation of charge carriers during photocatalytic reactions.
+
+Enormous efforts have been devoted to improving the photocatalytic activity of g-C\({}_{3}\)N\({}_{4}\), typically including doping with nonmetallic elements [15-16], deposition with metal atoms [17-18], compositing with other semiconductors [19-20] or conjugated polymers [21-22] and nanostructure engineering [10, 23-26]. Among them, porous g-C\({}_{3}\)N\({}_{4}\) (PGCN) is especially attractive due to its accessible porous framework with a large surface area. PGCN features facilitated mass transfer ability, enhanced light harvesting property and improved separation and migration efficiency of charge carriers owing to its nanoporous structure [10, 25-26]. Generally, PGCN can beobtained via a soft-templating or silica-templating method [26-28]. However, for soft template synthesis, some carbon is still residual from the templating polymers in the final product, which might reduce its catalytic activity [27]. The silica-templating synthesis involves further removal of the silica by aqueous ammonia or hydrogen fluoride [26, 28], which is hazardous and inconvenient. To develop a facile synthetic method of PGCN remains a significant issue.
+
+On the other hand, integrating photocatalysis with other advanced oxidation processes (AOPs) can be another solution to promote the separation and migration of charge carriers during photocatalytic reactions [29]. In this sense, coupling g-C\({}_{3}\)N\({}_{4}\) photocatalysis with ozone can be a promising treatment approach. Since ozone is a more powerful oxidant than oxygen, it can quickly capture photoinduced electrons during photochemical reactions. This would promote the separation of charge carriers and decomposition of ozone, and consequently generate more oxidative species to expedite the degradation and mineralization of water pollutants [29-32]. TiO\({}_{2}\) was the most widely applied in photocatalytic ozonation [29], indicating a quite limited optional scope. Till very recently, three studies announced that bulk g-C\({}_{3}\)N\({}_{4}\) could trigger a super synergy between visible-light photocatalysis and ozonation to enhance the oxidation efficiency [33-35]. However, an intensive investigation is still requisite on the mechanism for the enhanced mineralization of organics from photocatalysis to photocatalytic ozonation using PGCN as catalyst. It is also significant to study the evolution of reactive species and the mineralization pathway of water pollutants in Vis/O\({}_{3}\)/PGCN, in order to better understand the oxidation mechanism of this integrated technology.
+
+Herein, we reported a one-pot template-free method to fabricate a series of PGCN (PGCN-1, PGCN-2, PGCN-3 and PGCN-4) and applied in removal of p-hydroxybenzoic acid (PHBA). PHBA was chosen as the targeted pollutant for it is very commonly found in a great variety of agroindustrial wastewaters and especially toxic and refractory to anaerobic biological treatment [36-37]. Comparing with bulk g-C\({}_{3}\)N\({}_{4}\), PGCN-3 exhibited greatly improved photocatalytic activity for \(p\)-hydroxybenzoic acid (PHBA) degradation mainly due to its high surface area and enlarged band gap, while the final PHBA mineralization rate was nearly zero. The integration of PGCN-3 photocatalysis and ozoantion lead to almost complete mineralization of PHBA, and the mineralization could be further accelerated by increasing the ozone dosage. Trapping experiments were conducted to determine the evolution of reactive species from Vis/PGCN-3 to Vis/O\({}_{3}\)/PGCN-3. Electrospray ionization-mass spectrometry (ESI-MS) was further adopted to detect the evolution of degradation intermediates both in ozonation and Vis/O\({}_{3}\)/PGCN-3. Among the identified by-products, carboxylic acids were found to be highly recalcitrant to ozonation, while they could be rapidly eliminated by large amounts of non-selective *OH in Vis/O\({}_{3}\)/PGCN-3. Meanwhile, the complete mineralization process from the original PHBA to CO\({}_{2}\) and H\({}_{2}\)O was uncovered.
+
+## 2 Experimental
+
+### Materials and reagents
+
+Thiourea, ammonium chloride (NH\({}_{4}\)Cl), PHBA and oxalic acid (OA) were purchased from Sinopharm Chemical Reagent Co., Ltd., China. _Tert_-butanol (tBA), \(p\)-benzoquinone (pBQ) and ammonium oxalate (AO) were supplied by Xilong Chemical Co., Ltd., China. Sodium azide (NaN\({}_{3}\)) was purchased from Tianjin Fuchen Chemical Reagents factory, China. Ultrapure oxygen (purity 99.999%) and nitrogen gas (purity 99.999%) were provided by Beijing Qianxi Gas Co., Ltd., China. All chemicals used in this study were at least in analytical grade without further purification. Ultra-pure water was used for all synthesis and treatment.
+
+### Preparation of PGCN
+
+PGCN was prepared by pyrolysis of a mixture of thiourea and NH\({}_{4}\)Cl in air atmosphere. Typically, 10 g of thiourea and a certain amount of NH\({}_{4}\)Cl (2, 5, 10 or 15 g) were added into 10 mL ultrapure water in a 100 mL beaker. The beaker was then placed in a water bath with stirring at 65 \({}^{\circ}\)C for a few min. After natural dying in ambient condition, the white solid was transferred into an alumina crucible. The crucible was then put in a muffle furnace and heated to 550 \({}^{\circ}\)C with a rate of 15 \({}^{\circ}\)C/min and maintained for 4 h. The final products were collected, washed with ultrapure water and ethanol, and dried in an oven at 60 \({}^{\circ}\)C for 12 h. The samples prepared with 2, 5, 10 and 15 g of NH\({}_{4}\)Cl were labeled as PGCN-1, PGCN-2, PGCN-3 and PGCN-4, respectively.
+
+### Catalyst characterization
+
+The crystal phase was characterized by X-ray Diffraction (XRD) (X' PERT-PRO MPD) with a Cu\({}_{\text{x}a}\) irradiation (\(\lambda\)= 0.15406 nm). The morphologies and structures of the prepared samples were further investigated by field-emission transmission electron microscopy (FETEM, JEM-2100F, JEOL, Japan). The Brunauer-Emmett-Teller (BET) surface areas were measured by an automated gas sorption analyzer (Autosorb-iQ, Quantachrome, USA). Pore size distribution was calculated with the non-localized density functional theory method using adsorption data. The UV-vis diffuse reflectance spectra (DRS) of the samples were obtained using Varian Cary 5000, USA. X-ray photoelectron spectroscopy (XPS) data were obtained on an ESCALAB 250Xi instrument (Thermo Fisher Scientific, USA).
+
+### Photocatalysis, ozonation and photocatalytic ozonation experiments
+
+As shown in Scheme 1, the photocatalytic ozonation was carried out at 25 \({}^{\circ}\)C under visible light (420-800 nm) in a 500 mL cylindrical borosilicate glass reactor with a quartz cap and a porous glass plate at the bottom, containing 400 mL solution with 20 mg/L of PHBA and 0.5 g/L of catalyst. Typically, the gaseous ozone was continuously bubbled through the porous plate into the reactor with a flow rate of 100 mL/min and various concentrations (15, 40 and 65 mg/L). Meanwhile, the solution was irradiated under visible light (420-800 nm) with an average radiant flux of 360 mW/cm\({}^{2}\). Ozone was generated from ultrapure oxygen by an ozone generator (Anseros COM-AD-01, Germany). Visible light was provided by a 300 W Xenon lamp (CEL-NP2000, Aulight Co., Ltd., China) with a visible-light reflector and a 420 nm cutoff filter. Single photocatalysis and ozonation experiments were conducted using the same reactor. For photocatalytic degradation, during each run, the solution was magnetically stirring in dark for 30 min to obtain an adsorption-desorption equilibrium before reaction.
+
+### Analytical methods
+
+The concentrations of PHBA and OA were analyzed by high performance liquid chromatography (HPLC, Agilent series 1200, USA) equipped with a Zorbax SB-Aq column and a UV-Vis detector qualified at 240 nm and 210 nm, respectively. The mobile phase was a mixture of methanol and water containing 10 mM H\({}_{3}\)PO\({}_{4}\) (20/80%, v/v). Total organic carbon (TOC) was determined with a TOC-VCPH analyzer (Shimadzu, Japan). The electron spin resonance (ESR) signals of radicals spin trapped by 5,5-dimethyl-1-pyrroline (DMPO) were detected at ambient temperature on a JEOL (JES-FA200) spectrometer. The degradation intermediates were determined by ESI-MS (micOTOF-Q, Bruker, Germany). The ESI interface was operated in negative mode, and the parameters were set as follows: capillary voltage at 4500 V; nebulizer gas (N\({}_{2}\)) pressure at 0.4 bar; dry heater temperature at 180 \({}^{\circ}\)C; dry gas (N\({}_{2}\)) flow rate at 4.0 L/min. Full-scan spectra were obtained by m/z scanning from 21 to 800. All the ozone-involved samples were promptly pre-purged by ultrapure N\({}_{2}\) to remove the residual dissolved ozone before analysis.
+
+## 3 Results and discussion
+
+### Structural and optical properties of PGCN
+
+The PGCN materials were prepared by thermal condensation of mixed thiourea and NH\({}_{4}\)Cl in air atmosphere. By adjusting the mass ratio of NH\({}_{4}\)Cl to thiourea, the porous structure of g-C\({}_{3}\)N\({}_{4}\) could be effectively controlled. Fig. 1a-e shows the FETEM images of bulk g-C\({}_{3}\)N\({}_{4}\), PGCN-1, PGCN-2, PGCN-3 and PGCN-4, respectively. The bulk g-C\({}_{3}\)N\({}_{4}\) exhibited a typically bulk morphology with dense and thick nanosheets. With the addition of NH\({}_{4}\)Cl in the precursor, loose layers and abundant nanopores were gradually formed. PGCN-1, PGCN-2, PGCN-3 and PGCN-4 displayed loose and honeycomb-like nanoarchitectures. The pore size distribution curves in Fig. 1f further confirmed the nanoporous structure of PGCN. Generally, the overall number of nanopores increased with addition amount of NH\({}_{4}\)Cl in the precursor, while the pore sizes of PGCN cannot be accurately controlled by the addition amount of NH\({}_{4}\)Cl. Bulk g-C\({}_{3}\)N\({}_{4}\) had few pores, while PGCN-1 possessed a relatively larger amount of micropores (1.0 nm) and mesopores (8.5, 11.2, 16.0 and 30.1 nm). PGCN-2 had abundant micropores (1.1 nm) and less mesopores. Small mesopores (2.9 nm) and large mesopores (8.5, 11.2 and 16.0 nm) could be observed in PGCN-3. Best of all, PGCN-4 possessed both abundant micropores (1.4 nm) and small mesopores (3.0, 4.2, 5.7 and 7.5 nm). Correspondingly, PGCN-4 had the highest specific surface area (112.0 m\({}^{2}\)/g), which was 5.2 times as high as that of bulk g-C\({}_{3}\)N\({}_{4}\). The surface areas of PGCN-1, PGCN-2 and PGCN-3 were 42.5, 46.7 and 83.5 m\({}^{2}\)/g, respectively.
+
+NH\({}_{4}\)Cl as a bubbling template influenced the polymerization of thiourea from two aspects. One was the decreased polymerization degree of g-C\({}_{3}\)N\({}_{4}\). Fig. 2 presents the XRD patterns of the prepared materials, and a typical graphitic stacking C\({}_{3}\)N\({}_{4}\) structure was confirmed by two obvious peaks at 13.0\({}^{\rm o}\) and 27.4\({}^{\rm o}\), respectively [9]. The peak intensities of PGCN were lower than those of bulk g-C\({}_{3}\)N\({}_{4}\), indicating a less condensation degree and crystallinity of g-C\({}_{3}\)N\({}_{4}\). The XRD peaks at 27.4\({}^{\rm o}\) of PGCNwere broader than that of bulk g-C\({}_{3}\)N\({}_{4}\), which was attributed to the delamination of the thick bulk g-C\({}_{3}\)N\({}_{4}\) into thin nanosheets with several atomic layers [38].This was also in accordance with the thin and porous morphology of PGCN samples. The other was that the burst of bubbles from NH\({}_{4}\)Cl thermolysis would result in a large number of nanopores in the final g-C\({}_{3}\)N\({}_{4}\). Gas bubbles, which generated during the process of g-C\({}_{3}\)N\({}_{4}\) condensation, played a significant role for the generation of porous structure [10, 25]. The thermolysis of NH\({}_{4}\)Cl could release gaseous NH\({}_{3}\) and HCl during the polymerization of thiourea. Large amounts of soft bubbles were thereby generated and later burst, leading to the formation of abundant pores in the layers.
+
+The XPS measurement was conducted to determine the chemical state of PGCN-3, as shown in Fig. 3. Signals of C, N and O elements were found in the spectra survey (Fig. 3a), and no peaks assigned to S or Cl species could be detected (Fig. 3e-f). This implied that sulfur in thiourea and chlorine in NH\({}_{4}\)Cl were completely released into air during the heating process. The C1s spectra in Fig. 3b showed three main peaks, which could be assigned to adventitious carbon species (284.8 eV), C-(N)\({}_{3}\) (286.1 eV) and N-C=N (288.0 eV), respectively [39]. The high-resolution N1s spectra could be fitted into four peaks centered at about 398.4, 399.0, 400.6 and 404.2 eV, respectively (Fig. 3c) [39-40]. The main signals showed the occurrence of the sp\({}^{2}\)-bonded N involved in the triazine rings (C-N=C, 398.4 eV) and tertiary nitrogen groups (N-(C)\({}_{3}\), 399.0 eV). The weak peak at 400.6 eV indicated the presence of amino functional groups (C-N-H), originating from the defective condensation of heptazine substructures. The very weak peak at 404.2 eV was attributed to the charging effects or positive charge localization in the heterocycles.
+
+There were three main signals in O1s spectra, which could be assigned to the hydroxyl groups (-OH, 531.7 eV), O-C-N (532.3 eV) and chemisorbed H\({}_{2}\)O (533.4 eV), respectively [39]. The contained oxygen (O-C-N) in the samples presumably came from the heating treatment in the presence of air. The surface C/N atomic ratio of PGCN-3 was 1.29 in a semi-quantitative fashion. It indicated that PGCN-3 were carbon-rich and nitrogen-poor in terms of the nominal theoretical value of 0.75 for g-C\({}_{3}\)N\({}_{4}\).
+
+The optical properties of the prepared samples were examined by UV-vis DRS, as displayed in Fig. 4. The PGCN samples showed slightly reduced absorption than bulk g-C\({}_{3}\)N\({}_{4}\) especially in the region of 420-550 nm, while PGCN-4 exhibited enhanced absorption in the range of 600-800 nm. The bandgaps were determined correspondingly (Fig. 4b). The bandgaps (Eg), which were estimated from the intercept of the tangents to the plots of (ohu)\({}^{2}\) versus hu (Fig. 4b), were 2.78, 2.86, 2.86, 2.85 and 2.88 eV for bulk g-C\({}_{3}\)N\({}_{4}\), PGCN-1, PGCN-2, PGCN-3 and PGCN-4, respectively. The PGCN samples had almost equal Eg values, which were \(\sim\)0.1 eV higher than that of bulk g-C\({}_{3}\)N\({}_{4}\). This was in agreement with the reported literature [25]. The additionally released NH\({}_{3}\) (g) and HCl (g) would weaken the condensation process and reduce the degree of \(\pi\)-conjugated polymeric network, resulting in an increase of bandgap [25]. The enlarged band gap could increase the redox capability of electrons and holes and enhanced photocatalytic activity of g-C\({}_{3}\)N\({}_{4}\).
+
+### Enhanced photocatalytic activity of PGCN
+
+The photocatalytic activities of the materials were evaluated by PHBA degradation under visible light, as shown in Fig. 5. The photocatalytic degradation generally followed pseudo-zero-order kinetics according to the linear trends of C/C0 versus the irradiation time. The apparent rate constants were calculated correspondingly, as presented in Fig. 5b. It was obvious that all the PGCN samples demonstrated enhanced photocatalytic activity than bulk g-C3N4 for decomposition of PHBA under visible light. When the mass fraction of NH4Cl to thiourea reached 1.0, the resultant photocatalyst PGCN-3 exhibited the highest photocatalytic performance. However, when further increasing the proportion of NH4Cl, the photocatalytic performance of PGCN-4 significantly decreased. The specific surface area of PGCN-3 (83.5 m\({}^{2}\)/g) was 25.4% lower than that of PGCN-4 (112.0 m\({}^{2}\)/g), while the triggered PHBA removal rate constant (5.84\(\times\) 10\({}^{-2}\) mg/L \(\cdot\)min) was 39.0% higher than that of PGCN-4 (4.20\(\times\) 10\({}^{-2}\) mg/L \(\cdot\)min). Generally, high surface area can lead to enhanced adsorption and diffusion of reactants and products. However, it may also introduce more surface defects that could capture charge carriers and hinder their involvement in the photocatalytic reaction [25; 41]. The superior activity of PGCN-3 to PGCN-4 was presumably attributed to a less amount of surface defects and higher crystallinity.
+
+A series of trapping experiments were adopted using tBA [42], pBQ [42], NaN3 [43] and AO [42] as the scavenger of \(\cdot\)OH, superoxide radicals (\(\cdot\)O\({}_{2}^{-}\)), singlet oxygen (\({}^{1}\)O\({}_{2}\)) and positive holes (h\({}^{+}\)), respectively, in order to determine the involved reactive species during photocatalytic degradation of PHBA. Fig. 6 shows PHBA removal in the presence of various scavengers. The addition of tBA made a negligible difference on the degradation of PHBA, while pBQ slightly suppressed PHBA removal. It indicated that \(\cdot\)OH was not involved in the photocatalytic reactions, and \(\cdot\)O\({}_{2}^{-}\) only played a small part. NaN3 and AO could both moderately inhibit PHBA removal during photocatalysis. The photocatalytic degradation was mainly attributed to the direct reaction of PHBA with \({}^{1}\)O\({}_{2}\) and h\({}^{+}\).
+
+### Enhanced PHBA mineralization by photocatalytic ozonation
+
+Fig. 7 shows the TOC removal in mineralizing PHBA by Vis/PGCN-3, single ozonation and Vis/O\({}_{3}\)/PGCN-3, respectively. Although 17.9% of PHBA was removed in Vis/PGCN-3 at 60 min (Fig. 5a), the TOC removal rate was only 3.3%. It meant that 17.9% of PHBA reacted with \({}^{1}\)O\({}_{2}\) or h\({}^{+}\) during photocatalysis to form stable intermediates which were highly refractory to further mineralization. Compared to photocatalysis, ozonation exhibited moderately enhanced mineralization of PHBA. TOC was rapidly removed during the first 15 min in single ozonation, while the removal rate of TOC turned slower within the second 15 min. Nearly no TOC was eliminated from 30 to 60 min. This phenomenon was more obvious when improving the ozone concentration from 15 mg/L to 60 mg/L. Enhancing ozone dosage could accelerate PHBA mineralization in the first 15 min, while the final TOC removal was still quite limited. The final TOC removal was 44.2%, 46.5% and 48.6% in ozonation with an ozone concentration of 15, 40 and 65 mg/L, respectively. More importantly, PHBA mineralization was dramatically facilitated when coupling Vis/PGCN-3 with ozonation. The final TOC removal was 92.0%, 97.0% and 95.0% in Vis/O\({}_{3}\)/PGCN-3 with an ozone concentration of 15, 40 and 65 mg/L, which was 44.5%, 47.2% and 43.1% higher than the sum of that in Vis/PGCN-3 and ozonation, respectively. It indicated that PGCN-3 could trigger a vigorous synergy between photocatalysis and ozonation for efficient mineralization of water pollutants. It was also found that a higher ozone dosage would greatly expedite the mineralization of PHBA into carbon dioxide and water in Vis/O\({}_{3}\)/PGCN-3.
+
+In addition, the TOC removal in O\({}_{3}\)/PGCN-3 and Vis/O\({}_{3}\) was quite close to that in single ozonation, as shown in Fig. S1. This indicated that PGCN-3 had negligible activity in catalytic ozonation, and no synergistic effect was observed between photolysis and ozonation for PHBA mineralization. This result was in accordance with our previous work [34-35], confirming the key bridging role of PGCN-3 between visible-light irradiation and ozone molecules.
+
+### Mechanism for PHBA mineralization by Vis/O3/PGCN-3
+
+It is well known that two types of reaction mechanisms contribute to the removal of pollutants in ozone-involved AOPs. These mechanisms usually develop in consecutive stages: during the first minutes ozone reacts very fast with the unsaturated compounds to form less reactive intermediates (direct mechanism); and then these refractory compounds are eliminated by reactive radicals (indirect mechanism) [29-30]. Fig. 8a shows the PHBA removal by Vis/PGCN-3, O3 and Vis/O3/PGCN-3 with or without tBA and AO. It could be seen that PHBA reacted rapidly with ozone. The PHBA removal rate by single ozonation was as high as 69.0% within 20 min, while that was only 4.4% in Vis/PGCN-3. Among the three processes, Vis/O3/PGCN-3 revealed the highest rate of PHBA degradation. PHBA was completed eliminated by Vis/O3/PGCN-3 in 12.5 min, while 50.4% and 97.0% of PHBA were remaining in ozonation and Vis/PGCN-3, respectively. It indicated that other reactive species were generated and also responsible for PHBA degradation besides direct oxidation of ozone.
+
+As shown in Fig. 8a, the addition of tBA could significantly inhibited PHBA elimination. The PHBA removal rate of Vis/O3/PGCN-3/tBA was quite close to the sum of that of Vis/PGCN-3 and ozonation. This implied that the synergy between Vis/PGCN-3 and ozonation was dominantly attributed to the promoted generation of *OH. Large amounts of *OH were formed and mainly accounted for the enhanced mineralization of the original pollutant and refractory intermediates in Vis/O3/PGCN-3, which was in accordance with the reported UV/O3/TiO2 system [30]. The *OH generation was further confirmed by the ESR spectra in Fig. 8b. No signals assigned to *OH were found in Vis/PGCN-3 and single ozonation systems, while that can be conspicuously seen in the integrated Vis/O3/PGCN-3 process. Additionally, h\({}^{+}\) also played a very small part for PHBA degradation, according to the slight inhibition of AO. During Vis/O3/PGCN-3, PGCN-3 could absorb the visible-light photons to generate electrons upon conduction band (CB) and holes upon valance band (VB) (Eq.
+
+1). The CB minimum of g-C3N4 was reported to be \(-\) 1.3 V (versus NHE), while its VB maximum was \(+\) 1.40 V (versus NHE) [11-13]. Owing to its more negative CB potential than that of TiO2 (\(-\) 0.5 V versus NHE), ozone could quickly capture the strong reducing electrons upon CB of PGCN-3. An ozonide radical ('O3\({}^{-}\)') was thus formed (Eq. 2), and it rapidly reacted with H\({}^{+}\) in the solution to give a HO3\({}^{*}\) radical (Eq. 3), which subsequently decomposed into 'OH (Eq. 4).
+
+\[\text{PGCN-3}\xrightarrow{\text{Vis}}\text{e}^{-}+\text{h}^{+} \tag{1}\] \[\text{O}_{3}+\text{e}^{-}\rightarrow\text{'O}_{3}^{-}\] (2) \[\text{'O}_{3}^{-}+\text{H}^{+}\rightarrow\text{HO}_{3}\text{'}\] (3) \[\text{HO}_{3}\text{'}\rightarrow\text{O}_{2}+\text{'OH} \tag{4}\]
+
+Notably, h\({}^{+}\) were incapable to directly oxidize the surface hydroxyl groups or adsorbed water molecules to form 'OH, as the VB potential of g-C3N4 is less than E\({}^{0}\) ('OH/OHsuf') or E\({}^{0}\) ('OH/H2Oads) [33-35].
+
+The intermediates in PHBA mineralization by Vis/O3/PGCN-3 and ozonation were further detected through ESI-MS, in order to uncover the reaction pathway and better understand the reason for dramatically enhanced mineralization level of Vis/O3/PGCN-3. Fig. 9 shows the ESI-MS spectra of initial PHBA, intermediates in Vis/O3/PGCN-3 at 5 min, 30 min and 60 min, and final products in ozonation at 60 min. Based on this, the major by-products generated during Vis/O3/PGCN-3 were identified. Fig. 10 illustrates the identified intermediates and PHBA mineralization pathway in Vis/O3/PGCN-3. In the first stage, both ozone and 'OH could electrophilically attack the aromatic ring of PHBA to form protocatechuic acid (m/z = 153.0), which was further oxidized into 1,3-butadiene-1,2,4-tricarboxylic acid (m/z = 185.0) with ring opening. Muconic acid (m/z = 141.0) was quickly formed via decarboxylation of 1,3-butadiene-1,2,4-tricarboxylic acid. Afterwards, the non-selective 'OH dominantly accounted for further mineralization of carboxylic acids including muconic acid (m/z = 141.0), maleic/fumaric acid (m/z = 115.0),2-hydroxyl-maleic/2-hydroxyl-fumaric acid (m/z = 131.0), 2,3-dihydroxyl-maleic /2,3-dihydroxyl-fumaric acid (m/z = 147.0), and finally OA (m/z = 89.0). Further oxidation of OA would produce CO\({}_{2}\) and H\({}_{2}\)O, and the mineralization of PHBA was thus completely accomplished.
+
+It was also found that some carboxylic acids such as OA (m/z = 89.0), maleic/fumaric acid (m/z = 115.0) and 2,3-dihydroxyl-maleic/2,3-dihydroxyl-fumaric acid (m/z = 147.0) were remaining in ozonation system according to the corresponding ESI-MS peaks in Fig. 9e. It indicated that these compounds were highly recalcitrant to ozonation, which led to \(\sim\)55.8% of TOC residual in the solution (Fig. 7). OA, as one of the refractory intermediates, was then chosen as another targeted pollutant to test the oxidation efficiencies of Vis/PGCN-3, ozonation and Vis/O\({}_{3}\)/PGCN-3. As shown in Fig. 11, OA was highly stable in ozonation system as no OA was removed by single ozonation. Vis/PGCN-3 led to 13.0% of OA removal within 30 min, while that was 86.5% in Vis/O\({}_{3}\)/PGCN-3. The apparent OA mineralization rate constant in Vis/O\({}_{3}\)/PGCN-3 was 2.81\(\times\)10\({}^{-2}\) mM/L'min, which was 6.5 times as high as the sum of that in Vis/PGCN-3 and ozonation (Table S2). The integrated Vis/O\({}_{3}\)/PGCN-3 process exhibited outstanding mineralization of refractory byproducts compared to single photocatalysis and ozonation. The addition of tBA could completely block OA degradation (Fig. 11), reconfirming the dominated role of *OH in mineralizing water pollutants by Vis/O\({}_{3}\)/PGCN-3. Hence, a large amount of non-selective *OH were systematically generated triggered by ozone capturing highly reducing electrons (Eq.1-Eq.4), and could vigorously react with the pollutant and further its recalcitrant intermediates consequently to realize the complete mineralization in Vis/O\({}_{3}\)/PGCN-3.
+
+## 4 Conclusions
+
+We reported a facile one-step template-free method to fabricate honeycomb-like PGCN by mixing NH\({}_{4}\)Cl with thiourea as the precursor. The resulting PGCN-3exhibited obviously improved photocatalytic activity for PHBA degradation mainly due to its high surface area and enlarged band gap. It also strikingly triggered a vigorous synergy between photocatalysis and ozonation for PHBA mineralization. The final TOC removal in Vis/O\({}_{3}\)/PGCN-3 was 92.0% with an ozone dosage of 1.5 mg/min, which was 44.5% higher than the sum of that in Vis/PGCN-3 and ozonation. Such a remarkable enhancement was mainly attributed to the systematically promoted generation of non-selective *OH. The high CB level of PGCN-3 benefited electron capture by ozone molecules, thus significantly enhanced charge separation and the decay of ozone into abundant *OH. *OH could vigorously react with PHBA and its recalcitrant intermediates, leading to thorough mineralization in Vis/O\({}_{3}\)/PGCN-3. The present work puts forward an efficient metal-free Vis/O\({}_{3}\)/g-C\({}_{3}\)N\({}_{4}\) method for mineralization of water pollutants, and sheds light on the generation of reactive species as well as the mineralization pathway in this integrated process. It is also expected to advance the fundamental research and application of solar photocatalytic ozonation and other AOPs using sunlight as energy source.
+
+## Acknowledgements
+
+The authors greatly appreciate the financial support from the National Natural Science Foundation of China (No. 21207133) and the National Science Fund for Distinguished Young Scholars of China (No. 51425405).
+
+## References
+
+* [1] L. M. Sun, L. Xiang, X. Zhao, C. J. Jia, J. Yang, Z. Jin, X. F. Cheng, W. L. Fan, ACS Catal. 5 (2015) 3540-3551.
+* [2] A. Naldoni, M. Allieta, S. Santangelo, M. Marelli, F. Fabbri, S. Cappelli, C. L. Bianchi, R. Psaro, V. D Santo, J. Am. Chem. Soc. 134 (2012) 7600-7603.
+* [3] B. Wang, L. Wang, Z. B. Hao, Y Luo, Catal. Commun. 58 (2015) 117-121.
+* [4] M. C. Long, W. M. Cai, J. Cai, B. X. Zhou, X. Y. Chai, Y. H. Wu, J. Phys. Chem. B 110 (2006) 20211-20216.
+* [5] N. Zhang, M. Q. Yang, Z. R. Tang, Y. J. Xu, J. Catal. 303 (2013) 60-69.
+* [6] L. S. Zhang, W. Z. Wang, Z. G. Chen, L. Zhou, H. L. Xu, W. Zhu, J. Mater. Chem. 17 (2007) 2526-2532.
+* [7] P. Wang, B. B. Huang, X. Y. Qin, X. Y. Zhang, Y. Dai, J. Y. Wei, M. H. Whangbo, Angew. Chem. Int. Ed. 47 (2008) 7931-7933.
+* [8] K. J. Moor, J. H. Kim, _Environ._ Sci. Technol. 48 (2014) 2785-2791.
+* [9] X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G Xin, J. M. Carlsson, K. Domen, M. Antonietti, Nature Mater. 8 (2009) 76-80.
+* [10] M. Zhang, J. Xu, R. L. Zong, Y. F. Zhu, Appl. Catal. B: Environ. 147 (2014) 229-235.
+* [11] H. Q. Sun, G. L. Zhou, Y. X. Wang, A. Suvorova, S. B. Wang, ACS Appl. Mater. Inter. 6 (2014) 16745-16754.
+* [12] J. S. Zhang, Y. Chen, X. C. Wang, Energy Environ. Sci. 8 (2015) 3092-3108.
+* [13] Y. Zheng, L. H. Lin, B. Wang, X. C. Wang, Angew. Chem. Int. Ed. 54 (2015) 12868- 12884
+* [14] J. M. Cai, Y. M. Zhu, D. S. Liu, M. Meng, Z. P. Hu, Z. Jiang, ACS Catal. 5 (2015) 1708-1716.
+* [15] G. Liu, P. Niu, C. H. Sun, S. C. Smith, Z. G. Chen, G. Q. Lu, H. M. Cheng, J. Am. Chem. Soc. 132 (2010) 11642-11648.
+
+* [16] G. G. Zhang, M. W. Zhang, X. X. Ye, X. Q. Qiu, S. Lin, X. C. Wang, Adv.Mater. 26 (2014) 805-809.
+* [17] Y. T. Gong, P. F. Zhang, X. Xu, Y. Li, H. R. Li, Y. Wang, J. Catal. 297 (2013) 272-280.
+* [18] X. C. Wang, X. F. Chen, A. Thomas, X. Z. Fu, M. Antonietti, Adv. Mater. 21 (2009) 1609-1612.
+* [19] S. F. Chen, Y. F. Hu, S. G Meng, X. L. Fu, Appl. Catal. B: Environ. 150 (2014) 564-573.
+* [20] M. J. Munoz-Batista, M. Fernandez-Garcia, A. Kubacka, Appl. Catal. B: Environ. 164 (2015) 261-270.
+* [21] X. J. Bai, C. P. Sun, S. L. Wu, Y. F. Zhu, J. Mater. Chem. A 3 (2015) 2741-2747.
+* [22] D. S. Wang, H. T. Sun, Q. Z. Luo, X. L. Yang, R. Yin, Appl. Catal. B: Environ. 156 (2014) 323-330.
+* [23] X. J. Bai, J. Li, C. B. Cao, S. Hussain, Mater. Lett. 65 (2011) 1101-1104.
+* [24] X. H. Li, X. C. Wang, M. Antonietti, Chem. Sci. 3 (2012) 2170-2174.
+* [25] Z. Y. Wang, W. Guan, Y. J. Sun, F. Dong, Y. Zhou, W. K. Ho, Nanoscale 7 (2015) 2471-2479.
+* [26] X. C. Wang, K. Maeda, X. F. Chen, K. Takanabe, K. Domen, Y. D. Hou, X. Z. Fu, M. Antonietti, J. Am. Chem. Soc. 131 (2009) 1680-1681.
+* [27] W. Z. Shen, L. W. Ren, H. Zhou, S. C. Zhang, W. B. Fan, J. Mater. Chem. 21 (2011) 3890-3894.
+* [28] Y. J. Cui, J. S. Zhang, G. G. Zhang, J. H. Huang, P. Liu, M. Antonietti, X. C. Wang, J. Mater. Chem. 21 (2011) 13032-13039.
+* [29] J. D. Xiao, Y. B. Xie, H. B. Cao, Chemosphere 121 (2015) 1-17.
+* [30] F. J. Beltran, A. Aguinaco, A. Rey, J. F. Garcia-Araya, Ind. Eng. Chem. Res. 51 (2012) 4533-4544.
+* [31] M. Fathinia, A. Khataee, Appl. Catal. A: Gen. 491 (2015) 136-154.
+* [32] G. Marquez, E. M. Rodriguez, F. J. Beltran, P. M. Alvarez, Chemosphere 113(2014) 71-78.
+* [33] G Z. Liao, D. Y. Zhu, L. L. Li, B. Y. Lan, J. Hazard. Mater. 280 (2014) 531-535.
+* [34] J. D. Xiao, Y. B. Xie, H. B. Cao, Y. Q. Wang, Z. J. Zhao, Catal. Commun. 66 (2015) 10-14.
+* [35] J. D. Xiao, Y. B. Xie, F. Nawaz, S. Jin, F. Duan, M. J. Li, H. B. Cao, Appl. Catal. B: Environ. 181 (2016) 420-428.
+* [36] J. Beltran-Heredia, J. Torregrosa, J. R. Dominguez, J. A. Peres, Chemosphere 42 (2001) 351-359.
+* [37] J. Criquet, N. K. V. Leitner, Radiat. Phys. Chem. 106 (2015) 307-314.
+* [38] Z. Z. Lin, X. C. Wang, Angew. Chem. Int. Ed. 52 (2013) 1735-1738.
+* [39] F. Dong, Y. J. Sun, L. W. Wu, M. Fu, Z. B. Wu, Catal. Sci. Technol. 2 (2012) 1332-1335.
+* [40] B. H. Long, J. L. Lin, X. C. Wang, J. Mater. Chem. A 2 (2014) 2942-2951.
+* [41] J. S. Zhang, F. S. Guo, X. C. Wang, Adv. Funct. Mater. 23 (2013) 3008-3014.
+* [42] Z. F. Huang, J. J. Zou, L. Pan, S. B. Wang, X. W. Zhang, L. Wang, Appl. Catal. B: Environ. 147 (2014) 167-174.
+* [43] H. H. Ji, F. Chang, X. F. Hu, W. Qin, J. W. Shen, Chem. Eng. J. 218 (2013) 183-190.
+
+Figure Captions
+
+Figure 1: FETEM images of bulk g-C\({}_{3}\)N\({}_{4}\) (a), PGCN-1 (b), PGCN-2 (c), PGCN-3 (d) and PGCN-4 (e); pore size distribution of the prepared materials (f).
+
+Figure 2: XRD patterns of bulk g-C\({}_{3}\)N\({}_{4}\), PGCN-1, PGCN-2, PGCN-3 and PGCN-4.
+
+Figure 3: XPS spectra of PGCN-3 including the survey (a), C1s (b), N1s (c), O1s (d), S2p (e) and Cl2p (f).
+
+Figure 4: UV–Vis DRS spectra (a) and the bandgaps (b) of bulk g-C\({}_{3}\)N\({}_{4}\), PGCN-1, PGCN-2, PGCN-3 and PGCN-4.
+
+Figure 5: (a) Photocatalytic degradation of PHBA with various catalysts; (b) BET surface areas and PHBA photocatalytic removal rate constant of PGCN fabricated with different mass ratio of NH\({}_{4}\)Cl to thiourea as the precursor.
+
+Figure 6: Photocatalytic degradation of PHBA in the presence of various scavengers.
+
+Figure 7: TOC removal in treating PHBA by Vis/PGCN-3, O\({}_{3}\) and Vis/O\({}_{3}\)/PGCN-3 with an ozone concentration at 15, 40 and 65 mg/L (gas flow rate: 100 mL/min; light intensity: 360 mW/cm\({}^{2}\); solution volume: 400 mL; PHBA concentration: 20 mg/L; catalyst dosage: 0.5 g/L).
+
+Figure 8: (a) PHBA removal by various oxidation methods with or without tBA and AO (gas flow rate: 100 mL/min; ozone concentration: 15 mg/L; light intensity: 360 mW/cm\({}^{2}\); solution volume: 400 mL; PHBA concentration: 20 mg/L; catalyst dosage: 0.5 g/L); (b) ESR spectra of DMPO-OH adducts in Vis/PGCN-3, O\({}_{3}\) and Vis/O\({}_{3}\)/PGCN-3 systems.
+
+Figure 9: ESI-MS spectra of initial PHBA (a), intermediates in PHBA mineralization by Vis/O\({}_{3}\)/PGCN-3 at 5 min (b), 30 min (c) and 60 min (d), and final products in PHBA mineralization by ozonation at 60 min (e).
+
+Figure 10: Proposed mechanism for PHBA mineralization in Vis/O\({}_{3}\)/PGCN-3.
+
+Figure 11: Comparison of OA removal by Vis/PGCN-3, ozonation and Vis/O\({}_{3}\)/PGCN-3 (gas flow rate: 100 mL/min; ozone concentration: 40 mg/L; light intensity: 360 mW/cm\({}^{2}\); solution volume: 400 mL; OA concentration: 1 mM; catalyst dosage: 0.5 g/L).
+
+**Scheme 1.** Experimental setup of photocatalytic ozonation.

manual_annotation/200pdfs_in_mmd/10.1016_j.apcatb.2019.03.015.mmd

@@ -0,0 +1,142 @@
+A novel electro-catalytic membrane contact for improving the efficiency of ozone on wastewater treatment
+
+Kutiling Li, Lili Xu, Yong Zhang, Aixin Cao, Yujue Wang, Haiou Huang, Jun Wang
+
+# Abstract
+
+A novel electro-catalytic membrane contactor was designed to break the two constrains in O3 wastewater treatment technology simultaneously: the mass transfer of O3-water and the decomposition rate of O3 into 'OH. In the electro-catalytic membrane contactor donation (ECMCO) process, O2 and O3, in the gas chamber, diffuse through the hydrophobic membrane into the water phase; O2 is electro-reduced to H2O4 that catalyses O3 producing 'OH to degrade organic compounds rapidly. The removal rate of nitrobenzene (NB) was significantly improved by the combination of membrane contactor and electro-catalytic process. The removal rates of NB in 120 min were 85%, 55% and 23% for ECMCO, membrane contactor donation and electrolysis, respectively. O3 concentration and current density had positive effects on the removal of NB, which was related to the accumulated H2O4 concentration in aqueous phase. Gas flow had slight positive effects on NB removal when it was lower than 50 mL/min. The pH of the feed may affect the production and the form of H2O2 and hence the removal of NB. The highest removal rate in 120 min was up to 95.7% at the initial pH of 4.5. Mass transfer and electrolysis are synthetic in ECMCO process. On one hand, the mass transfer of O3 increased by 2 times due to the electro-catalytic production of H2O2. On the other hand, the existence of O3 promoted the production of H2O2. ECMCO has great potential on promoting mass transfer and decomposition of O3 into 'OH for the industrial application of O3.
+
+## Introduction
+
+Ozone (O3) can react with many organic compounds due to its high oxidation potential. It has been applied to disinfection and oxidation of drinking water and wastewater for decades [1, 2, 3].Although O3 has been used on water treatment for a long time, the cost of industrial application of O3 is still high [4]. Two main reasons are that the mass transfer efficiency between O3 and water is not high enough and the reaction rate between O3 and ozone-resistant compounds is poor [5, 6].
+
+O3 can't be stored and has to be generated on site when it is needed, because it is easy to decompose to oxygen (O2). O3 is generally produced under high voltage discharge using air or O2 as resources. Normally O3 is no more than 10% in the effluent of an ozonator and the O2 in the effluent ("90%) is scarcely used [7]. The low proportion of O3 in the gas phase determines a weak driving force for O3 diffusing into liquid phase. The dissolved O3 also easily escapes from the liquid into ambient environment due to its very low partial pressure in the air (Henry's low) [8]. Besides, the limited bubble surface provides insufficient contact interface area for mass transfer in the conventional contractors [9, 10]. These factors determine the poor utilization of O3. In order to achieve the requirement of oxidation, an excess amount of O3 has to be provided and the operation cost is usually high. The low production rate and the low partial pressure of O3 can hardly be changed, while to enlarge the contact interface in unit volume to promote mass transfer of O3 is a feasible method [11].
+
+Membrane contactor is considered to be able to provide enormous interfacial area in unit volume for O3 transfer [12, 13, 14]. In a membrane contactor, gas phase and aqueous phase are separated by a hydrophobic membrane and they flow separately. The membrane provides an interface between gas and water and acts as a distributer of O3. The effective interfacial area can be increased tremendously by improving membrane packing density. The volumetric mass transfer coefficient can be 1-2 orders magnitude higher relative to fine bubble contact [11]. In comparing with conventional contactors, membrane contact also has advantages such as the avoidance of bubble and foam formation, the reuse of oxygen in the gas phase and the lower energy consumption [15,16].
+
+Nevertheless, most reported membrane contactors for O3 based on direct ozonation. The application of membrane contact might be limited for ozone-resistant compounds in advanced wastewater treatment, herbicides removal and micropollutants mineralization. In fact an advanced oxidation process (AOP) may happen in O3 wastewater treatment process by catalyzing O3 transforming to hydroxyl free radicals ('OH) [17,18]. AOPs have attracted much attention due to their ability on mineralizing refractory organic compounds. O3 is one of the most studied reagents for producing 'OH in laboratory [19-21]. Some ozone-resistant compounds can be easily degraded by 'OH. The organic degradation can be critically improved by catalytic ozonation process in O3 wastewater treatment [22-24].
+
+Therefore, combining membrane contact ozonation (MCO) with O3 based AOPs may improve its oxidation capacity to refractory compounds. Merle et al. combined a membrane contact with peroxone (O3/H2O2) process for simultaneous micropollutant abatement and bromate minimization. The hybrid process performed better than conventional O3/H2O2 process for both ground water and surface water [25]. Li et al. coated ferrihydrite and active carbon on the surface of hollow fiber membranes in membrane contactor to catalyze O3 dissolution and transformation to 'OH, where the removal of _N,N_-diethyl-meta-toluamide in water improved from 30% to 60% [26]. These combinations of membrane contactor and catalytic ozonation improved the oxidation capacity of O3 process. Considering the high proportion of O2 in the effluent of an ozonator, it could be a good way to electro-catalytic O2 to H2O2 and implement an O3/H2O2 process by coating conductive catalyzer on the surface of membranes in a membrane contractor.
+
+In this study a novel electro-catalytic membrane contactor was designed. A nickel (Ni) foam plate and a polyvinylidene fluoride (PVDF) hydrophobic membrane were placed in the contactor (Fig. 1). The Ni foam contacted with water phase and acted as a cathode in the reactor, while the PVDF membrane affixed to the Ni foam tightly and contacted with the gas phase. In the electro-catalytic membrane contactor ozonation (ECMCO) process, O2 in the effluent of an ozonator diffuses into aqueous phase through hydrophobic membrane and is electrolytically reduced to H2O2 (Eq. (1)); O3 diffuses from gas phase into aqueous and reacts with H2O2 producing 'OH (Eqs. (2)-(8)) to degrade organic compounds rapidly.
+
+\[\begin{equation*}{\mathrm{O}}_{2}+2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-}\to{\mathrm{H}}_{2}{\mathrm{O}}_{2}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{H}}_{2}{\mathrm{O}}_{2}={\mathrm{H}}^{+}+{\mathrm{H}}_{0}{{}^{2}}^{-}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{H}}_{0}{{}^{2}}^{-}+{\mathrm{O}}_{3}\to{\mathrm{H}}_{0}{{}^{2}}^{+}+{\mathrm{O}}_{3}{{}^{-}}^{.}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{O}}_{2}{{}^{2}}^{-}+{\mathrm{H}}^{+}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{O}}-2+{\mathrm{O}}_{3}\to{\mathrm{O}}_{2}+{\mathrm{O}}_{3}^{-}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{O}}_{3}^{-}+{\mathrm{H}}^{+}={\mathrm{H}}_{0}{{}^{3}}^{.}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{H}}_{0}{{}^{3}}^{-}\to{\mathrm{O}}{\mathrm{H}}_{0}{{}^{+}}^{.}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{O}}_{3}+{\mathrm{O}}{{}^{H}}^{-}\to{\mathrm{H}}_{0}{{}^{2}}^{-}+{\mathrm{O}}_{2}\end{equation*}\]
+
+By this design, mass transfer and oxidation efficiency can be mutual promoted: H2O2 is produced on the surface of membrane, and O3 may be rapidly decomposed, which increases the driving force for O3 transfer; more O3 molecules can be catalyzed to form 'OH, and the removal efficiency of ozone-resistant compounds can be improved. Except that, some other advantages can also be achieved, such as (1) some O2 can be utilized to avoid waste relative to the MCO process; (2) H2O2 is produced on site to avoid the risky during the transportation and storage of high concentration H2O2 solution; (3) the production of H2O2 can be easily controlled by adjusting electric conditions. With these advantages, ECMCO may have the potential on reducing cost and improving oxidation efficiency in processes like advanced wastewater treatment, micropollutant removal and disinfection by-products control.
+
+The objective of this study is to construct an electro-catalytic membrane contractor to improve the organic degradation efficiency by enhancing O3 mass transfer and implementing O3/H2O2 based AOP. In order to evaluate the oxidation efficiency of ECMCO process, nitrobenzene (NB), an ozone-resistant organic, was chosen as a model contaminant. The efficiency of ECMCO was compared with that of electrolysis and conventional MCO process. The effects of operation conditions including current density, O3 concentration, gas flow rate and initial pH of the solution were investigated. Mass transfer was investigated and liner sweep voltammtry (LSV) experiments were conducted to evaluate the synergy effects on mass transfer and H2O2 production in ECMCO process.
+
+## Experimental
+
+### Chemicals and materials
+
+NB (AR, > 99.5%), Potassium titanium oxalate (AR), Sulfuric acid (AR, 95% - 98%) was purchased from Sinopharm Chemical Reagent
+
+Fig. 1: Schematic of experimental setup.
+
+Co., Ltd. (China). Na2SO4 (99.5%) and H2O2 (30 wt% solution) were from Beijing Chemical works (China). 5,5-Dimethyl-1-pyrroline-_N_-oxide (DMPO) was purchased from Sigma-Aldrich. The PVDF mem-brane (ICEQ00010) was from Merck (German) (Table 1).
+
+### Experimental setup
+
+The experimental setup is shown in Fig. 1. Flat sheet membrane was employed since it's easy to be assembled and the mechanism will not be different from hollow fiber membrane contactors. The hydrophobic membrane and Ni foam plate were installed in the self-designed module, where the hydrophobic layer contacted with the gas phase and the catalytic layer impregnated in the aqueous solution. The aqueous solution contain 0.05 M Na2SO4 and the initial concentration of NB was 30 mg/L for all experiments. The volume of the reactor was 120 mL. The anode was RuO2/Ti mesh and the effective dimension for both anode and cathode were 50 mm x 50 mm. The distance between anode and cathode was 20 mm. The current density ranged from 0.4 to 3 mA/cm2. The gas flow varied between 25 mL/min and 100 mL/min. The O3 concentration varied between 0 mg/L and 75 mg/L. When O3 concentration was 0 mg/L, the connotator was powered off and pure O2 was supplied. The initial pH of the solution varied from 2 to 11. Turn on the direct current (DC) power supply and turn off the connotator to evaluate the performance of electrolysis. Turn off the DC power supply and turn on the connotator to evaluate the performance of MCO. Turn on the DC power supply and oxonator at the same time to evaluate the performance of ECMCO. And then the removal rate of NB in each process was compared. The initial pH of NB solution was adjusted by adding 1 M HSO4 or 1 M NaOH. The replication of each test was 3, and average value was adopted.
+
+### Analysis methods
+
+The concentration of NB was measured by using High Performance Liquid Chromatography (HPLC, Agilent 1260 Infinity). The concentration of H2O2 was measured using potassium titanium oxalate method [27]. The amount of 'OH was analyzed by using electron spin resonance (ESR) method [28], and DMPO was used to trap 'OH.
+
+## Results and discussion
+
+### The performance of ECMCO on NB removal
+
+In comparing with conventional MCO and electrolysis, ECMCO process exhibited the highest removal rate for NB in 120 min (Fig. 2). The efficiency of electrolysis was the lowest and only 23% of NB was removed in 120 min. The removal efficiency of NB in MCO was much better than electrolysis, and the removal rate was 55% in 120 min. In ECMCO process the removal rate of NB in 120 min was 85%. The removal rate of ECMCO was higher than the summation removal rate of electrolysis and MCO, indicating a synthetic effects in the combination of MCO and electrolysis. The enhancement on NB removal benefited from the production of 'OH in ECMCO process that was further discussed later.
+
+### Mechanism of ECMCO process
+
+The composition of gas phase was changed to evaluate the role each component in gas phase played in ECMCO process, and the corresponding concentrations of H2O2 were shown in Fig. 3. The concentration of H2O2 was the lowest when pure N2 was charged due to the insufficient O2 in the aqueous. And the slight amount of O2 came from the dissolved O2 in the initial NB solution and oxygen evaluation reaction at the anode region. When pure O2 was charged at the gas phase, H2O2 concentration was the highest and it increased by 4 times than that when N2 was charged. It indicated that the O2 at the gas phase diffused into the aqueous through the hydrophobic layer and was utilized to produce H2O2 in the catalytic layer. When the mixture of O2 and O3 was charged at the gas phase, H2O2 concentration decreased, and it decreased with the increase of O3 concentration. It indicated that the electrolytically produced H2O2 was partially consumed by O3 which diffused into the aqueous through the hydrophobic layer. Along with the increase of O3 concentration, the driving force of O3 was increased due to the larger concentration difference across the membrane. Therefore, more O3 transferred into the aqueous phase and consumed more H2O2.
+
+The production of 'OH in ECMCO process was confirmed by ESR results in Fig. 4. The characteristic peaks in ESR spectra revealed the existence of 'OH in ECMCO process. The intensity of the peaks reflects the relative amount of 'OH to a degree [29]. The amount of 'OH increased in the initial 40 min and then leveled off. In the initial 40 min
+
+\begin{table}
+\begin{tabular}{c c c c} Membrane & Thickness (μm) & Porosity (\%) & Mean pore size (μm) \\ ICEQ00010 & 210 & 76 & 0.2 \\ \end{tabular}
+\end{table}
+Table 1: The properties of the PVDF membrane.
+
+Figure 3: H2O2 concentration with different gases charging (current density of 1 mA/cm2; gas flow rate of 50 mL/min; initial pH of 7).
+
+Figure 2: The removal of NB in different processes (current density of 1 mA/cm2; gas flow rate of 50 mL/min; O3 concentration of 50 mg/L; initial pH of 7).
+
+the accumulated H2O2 might be insufficient and limited the production of 'OH. After 40 min, the amount of 'OH kept as a constant indicating that enough H2O2 was accumulated in the system while O3 was relatively insufficient. The amount of O3 diffused into aqueous was determined by both driving force and resistance to mass transfer, which was affected by operation conditions that were discussed later.
+
+### Effects of operation conditions
+
+When O3 concentration was 0 mg/L, pure O2 was charged in the gas chamber. The removal rate of RN was just 23% (Fig. 2). H2O2 can hardly react with NB [29,30], so NB was just degraded by electrolysis. The existence of 25 mg/L O3 dramatically increased the removal rate of NB to 79%, while increasing O3 concentration from 25 to 75 mg/L only increased the removal rate from 79% to 93% (Fig. 5(a)). These results revealed that the existence of 'OH dramatically improved the removal rate of NB. NB reacts slowly with molecule O3 (0.09 +- 0.02 M-1s-1) and reacts quickly with 'OH (2.2 x 108 M-1s-1) [31]. When O3 concentration increased from 25 to 75 mg/L, the removal rates in the former 40 min were no obvious difference, since the electro-produced H2O2 did not accumulate enough and limited the removal of NB. It's in coordination with previous results in Fig. 4. After 40 min, the increased O3 concentration in the gas phase enhanced the driving force of mass transfer through the membrane, which increased the O3 consumption and the 'OH production that facilitated the removal of NB.
+
+The effects of current density on the final removal rate of NB were slight (Fig. 5(b)). The NB removal rates at 120 min were 84%, 85%, 87% and 92% when the current density was 0.4, 1, 2 and 3 mA/cm2, respectively. The removal rate of NB in former 40 min was evidently enhanced by increasing current density from 0.4 to 3 mA/cm2. The main reason was that the increase of current density enhanced the accumulation of H2O2 (Fig.S1) and promoted the production of 'OH. Besides, the electrolysis degradation of NB should have been enhanced along with the increase of current density, which also contributed to the significant difference of removal rate in the former 40 min.
+
+The effects of gas flow on NB removal were slight (Fig. 5(c)). The removal rate increased from 77% to 85% when gas flow increased from 25 to 50 mL/min. Mass transfer in membrane contact follows double-film theory. The resistances for O2 and O3 transfer were in the gas boundary layer, membrane matrix and liquid boundary in sequence. The increased gas flow intensified the turbulence of gas phase and decreased the resistance in gas boundary layer, which facilitated the mass transfer of O3. The increased gas flow might have also improved the mass transfer of O2, which might enhance the production of H2O2 and further promote the mass transfer of O3. While continuous increase of gas flow had no significant improvement on NB removal, since the resistance in water phase boundary layer and membrane matrix might dominate the mass transfer of O3, and the resistance in gas film was negligible [32].
+
+The effects of initial pH value on the removal of NB were shown in
+
+Fig. 4: The ESR spectra with elapsed time (current density of 1 mA/cm2; gas flow rate of 50 mL/min; O3 concentration of 50 mg/L; initial pH of 7).
+
+Fig. 5: Operation conditions effects on removal of NB: (a) O3 concentration; (b) current; (c) gas flow rate; (d) initial pH value (current density of 1 mA/cm2; gas flow rate of 50 mL/min; O3 concentration of 50 mg/L; initial pH of 7).
+
+Fig. 5(d). These results indicated that the ECMCO process performed very well on NB removal in a relative large range of pH. When the initial pH value was 4.5, the removal rate of NB was the highest of 95.7%. And it was the lowest when initial pH value was 7. When initial pH value was lower than 3 or higher than 9.5, the removal rate of NB decreased. The reactions in ECMCO was complicated involving several process and chain reactions (Eq. (2) - (7)). In an acid aqueous solution, the production of H2O2 can be enhanced [33], but the existence form of H2O2 was also effected by the concentration of H+. The reaction of H2O2 with O3 is slow, but that of its anion, H2O2-, is fast. Thus the concentration of H02- determines the chain reaction of producing 'OH [34,35]. When initial pH value was 4.5 the amount of H2O2- might increase due to the enhanced electro-reduction of O2 to H2O2. When the initial pH value was lower than 4.5, the increased concentration of H+ might limit the amount of H02- (Eq. (2)). Consequently, the removal rate of NB decreased when initial pH value was 2. On the contrary, in an alkalescent aqueous solution, the production of H2O2 could be inhibited to a degree, but the amount of H02- might increase due to the equilibrium shifting. Besides, OH- can react with O3 producing 'OH (Eq. (8)) [36]. When the initial pH value was higher than 9.5, the production of H2O2 might be critically inhibited inducing the decrease of removal rate. Furthermore, pH can affect the mass transfer rate of O3 and the solubility of O3 in aqueous. Solution pH can therefore have very complex effects on the 'OH production in ECMCO process, which may require further investigation.
+
+### Mass transfer of O3
+
+By measuring O3 concentrations in the effluent of electro-catalytic membrane contact and conventional membrane contactor, the mass transfer of O3 was calculated. These results were shown in Fig. 6. It can be found that the mass transfer of O3 in ECMCO process was about 2 times higher than that in MCO process. The O3 residual concentrations in bulk liquid were also tested, and there were no O3 residual at any time for both processes. It indicated that the O3 transferred through membrane were totally utilized and the O3 utilization was 2 times improved by implementing O3/H2O2 based AOP. The enhanced mass transfer rate was due to the intensified reaction of O3 with H02-. H02 had higher reaction rate constant with O3 than that of NB [34,37]. O3 in the liquid phase was consumed rapidly, which increased the concentration difference of O3 between two sides of the hydrophobic layer and hence improved the driving force for O3 transfer. Besides, the H2O2 is produced just on the surface of the hydrophobic layer, which may promote O3 mass transfer more effectively than adding H2O2 in the feed.
+
+### LSV experiments
+
+LSV experiments were conducted with charging different gases into the gas chamber (Fig. 7). When pure N2 was charged, the LSV curves revealed that the hydrogen evolution reaction happened at the potential of - 0.8 V vs. Ag/AgCl. When pure O2 was charged, the reduction current occurred when potential was lower than - 0.4 V, where O2 was electro-reduced to H2O2 through the two-electron pathway. When the mixture of O2 and O3 was charged in the gas chamber, there was a constant reduction current when voltage ranged from 0.2 V to - 0.4 V vs. Ag/AgCl. This reduction current related to the reduction of O3 [38]. With the existence of O3, O2 reduction still occurred at the potential of - 0.4 V vs. Ag/AgCl. The reduction current was much stronger in the range of - 0.5 to -0.8 V vs. Ag/AgCl than that when pure O2 was charged in the gas chamber. The enhanced reduction current indicated that the O3 promoted the production of H2O2 from the cathodic reduction of O2. The dissolved O3 consumed some H02- rapidly in the catalytic layer, which might enhance H02- releasing from the active site on cathode, so the production of H2O2 was facilitated [39]. When the cathodic potential was more negative than -1.0 V vs. Ag/AgCl, the reduction currents were the same for pure O2 and O2/O3 charging. This indicated that the cathodic O3 reduction was limited as the cathodic potential was getting more negative. This may be because of that the enhanced production of H2O2 consumed more O3, and hence the cathodic reduction of O3 was limited. This result was corresponding with what was found by Xia et. al. [38]. The LSV experiment results suggested that the existence of O3 didn't compete with the cathodic reduction of O2, but O3 actually facilitated the production of H2O2 from O2 reduction.
+
+All these results illustrated the synthetic effects between electrolysis and membrane contact mass transfer in ECMCO process: the electro-catalytic production of H2O2 facilitated the mass transfer of O3; the existence of O3 enhanced the production of H2O2 from cathodic reduction of O2. The mutual promotion of H2O2 production and O3 mass transfer determined a high production of 'OH and removal rate of NB.
+
+## 4 Conclusions
+
+A novel ECMCO process was constructed by coupling MCO with peroxone processes. The removal of NB in ECMCO was even faster than the summation of that in MCO and electrolysis due to the production of 'OH. ECMCO process can be affected by O3 concentration in the gas phase and current intensity. The effects of gas flow rate were slight. In the former 40 min, the accumulated H2O2 was usually low and
+
+Fig. 6: Mass transfer of O3 in ECMCO and MCO processes with elapsed time (current density of 25 mA/cm2; gas flow rate of 50 mL/min; O3 concentration of 50 mg/L; pH of 7).
+
+Fig. 7: LSV curves with charging different gases into the gas chamber.
+
+dominated the production of 'OH. While after 40 min, the amount of O3 transferred turned out to be the limitation for 'OH production. ECMCO can work in the range of initial pH value of 2 to 11. In ECMCO process, membrane contact mass transfer and electro-catalytic reduction of O2 were synergistic. The electro-catalytic production of H2O2 facilitated the mass transfer of O3 by 2 times, since O3 can be rapidly consumed by H2O2; and the existence of O3 enhanced the production of H2O2 from cathodic reduction of O2. Therefore, ECMCO has a great potential to improve the O3 technology in wastewater treatment by solving two critical issues on the mass transfer of O3 and the decomposition of O3 into 'OH.
+
+## Acknowledgement
+
+This research was financially supported by the special fund of State Key Joint Laboratory of Environment Simulation and Pollution Control (18L01ESPC), the National Water Pollution Control and Treatment Science and Technology Major Project (2017ZX07107-002-02), the National Key Research and Development Program of China (2016YFC0400501) and the National Natural Science Foundation of China (Grant NO. 51578533).
+
+## Appendix A Supplementary data
+
+Supplementary material related to this article can be found, in the online version, at doi:[https://doi.org/10.1016/j.apcatb.2019.03.015](https://doi.org/10.1016/j.apcatb.2019.03.015).
+
+## References
+
+* (1) T. Wadhawan, H. Simek, M. Kasi, K. Kautson, B. Fuhs, J. McEvoy, E. Khan, Dissolved organic nitrogen and its biodegradable portion in a water treatment plant with ozone oxidation, Water Res. 54 (2014) 318-326, [https://doi.org/10.1016/j.water.2014.02.009](https://doi.org/10.1016/j.water.2014.02.009).
+* (2) M. C. Dodd, M. D. Buffle, U. Von Gunten, Oxidation of antibacterial molecules by aqueous ozone: moiety-specific reaction kinetics and application to ozone-based wastewater treatment, Environ. Sci. Technol. 40 (2006) 1969-1977, [https://doi.org/10.1021/en0513606](https://doi.org/10.1021/en0513606).
+* (3) W.T.A. Andourant, M. Callaverni, J. Nopens, J. C. Wandelt, F. Vanhoucke, A. Dumoulin, P. Dejans, S.W.H. Van Hulle, Full-scale modelling of an ozone reactor for drinking water treatment, Chem. Eng. J. 157 (2010) 551-557, [https://doi.org/10.1016/j.jcp.2001.20.151](https://doi.org/10.1016/j.jcp.2001.20.151).
+* (4) C. Gortschali, A. Almya, A. Suappa, Onuation of Water and Wastewater: A Practical Guide to Understanding Ozone, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2010.
+* (5) F.R. Alves dos Santos, C.P. Borges, F.V. da Fonseca, Polymeric materials for membrane contact devices applied to water treatment by ozonation, Mater. Res. J. Mater. 18 (2015) 105-1022, [https://doi.org/10.1510/1516-349-016751](https://doi.org/10.1510/1516-349-016751).
+* (6) G.D. Onstad, S. Straub, J. Merlino, G.A. Cold, U. Von Gunten, Selective oxidation of key functional groups in organotoxins during drinking water emanation, Environ. Sci. Technol. 41 (2007) 4397-4404, [https://doi.org/10.1021/en050625327](https://doi.org/10.1021/en050625327).
+* (7) P. Janbrecht, P.A. Wilderer, C. Picard, A. Larobot, Ozone-water contacting by ecemnia membranes, Sept. Technol. 25 (2001) 341-346, [https://doi.org/10.1016/S1383-5886](https://doi.org/10.1016/S1383-5886)(01)00061-2.
+* (8) A. Rodriguez, R. Rossi, J.A. Perdigon-Saleo, M. Mercusa, A. Agliena, M. D. Hernandez, P. Izedra, A. Fernandez-Alba, E. Garcia-Calvo, Ozone-based Technologies in Water and Wastewater Treatment, (2008), [https://doi.org/10.7098/5.103](https://doi.org/10.7098/5.103).
+* (9) P.V. Shanbhag, A.K. Guha, K.K. Sarkar, Single-phase membrane ozonation of hazardous organic compounds in aqueous green, J. Hazard. Mater. 41 (1995) 95-104, [https://doi.org/10.1016/j.apcatb.2007.2](https://doi.org/10.1016/j.apcatb.2007.2).
+* (10) Z. Shen, M.J. Sammers, A.G. Collins, A novel approach to ozone-water mass transfer using hollow fiber-fiber networks, Environ. Technol. (United Kingdom). 11 (1990) 597-608, [https://doi.org/10.1080/09593339009364902](https://doi.org/10.1080/09593339009364902).
+* (11) D.S. Pines, K.N. Min, S.J. Zeng, D.A. Beckrow, Investigation of an ozone membrane contact system, Ozone Sci. Eng. 27 (2005) 209-217, [https://doi.org/10.1080/01915050045750](https://doi.org/10.1080/01915050045750).
+* (12) R.I.S. Ansen, A. Zwaighenburg, W.G.J. Van Der Meer, M. Wemling, Outside-in trimming of humic substances during ozonation in a membrane contactor, Environ. Sci. Technol. 40 (2006) 6460-6465, [https://doi.org/10.1021/en05060531](https://doi.org/10.1021/en05060531).
+* (13) T. Leikens, J. Pflattanawat, M. Butler, U. Von Gunten, W. Pronq, Oronse transfer and design concepts for NOM dec

manual_annotation/200pdfs_in_mmd/10.1016_j.apcbee.2014.03.018.mmd

@@ -0,0 +1,88 @@
+# UV/ozone Treatment of the Pyrethroid Insecticide Fenvalerate in Aqueous Solutions
+
+Nga T. T. Tran
+
+Corresponding author. Tel.: +6-011-185-93681; fax: +6-053-687-649
+
+Thanh H. Trinh
+
+W.M. Hoang
+
+Thang M. Ngo
+
+# Abstract
+
+Fenvalerate is a common Pyrethroid insecticide exits stably in water and soil. This study subjected to enhance the degradation of fenvalerate (in the form of aqueous emulsion of a commercial formulation) using UV/ozone process. Experiment results indicated that fenvalerate was decomposed rapidly under UV irradiation (99% within 10 minutes). Degradation yield also showed an increase when ozone was applied. UV/ozone degradation rates of fenvalerate followed first-order kinetics. In alkaline medium, there was a slight increase in yield. Sodium nitrate acted as a photo-sensitizer for UV irradiation process so it helped to increase reaction rate at an optimum concentration of 2.5 mM. Moreover, some degradation products were identified and tentatively assigned by GC-MS.
+
+## 1 Introduction
+
+Pyrethroids comprise a class of synthetic compounds which have structures based on those of the naturally occurring pyrethrin insecticides [1]. Synthetic pyrethroids have been widely used as pesticide in agriculture, forestry and public health due to their high insecticidal activity, long-term stability, and good compatibility [1] and [2]. Fenvalerate is one of the most potent pyrethroid insecticides, controlling a wide range of insect pests [3]. It has two chiral centers, in the acid and alcohol moieties, and thus contains four stereoisomers with\(\alpha\)S,2R-, \(\alpha\)R,2S-, \(\alpha\)R,2R-, and \(\alpha\)S,2S- configurations [3]. Fenvalerate has the potential to accumulate in soil, while its potential to pollute ground water is expected to be minimal because of its extreme lipophilic. However, fenvalerate can combine with clay, soil, sediment particulates or organic material which present in surface water in suspension form, and this raises the potential to pollute surface water [1].
+
+In natural environment, pyrethroids can be degraded through several possible processes, including photo-degradation, biodegradation, and hydrolysis [2]. Lutnicka stated that fenvalerate's half-time in an aquatic ecosystem was about 4 days [4]. Liu et al. reported their results on photo-degradation of deltamethrin and fenvalerate under simulated natural conditions. In their study, photo-degradation of both pyrethroids follow first-order kinetics under the conditions of different solvents (_n_-hexane and 50% methanol/water) and under different light irradiations [2]. Colombo studied on photo-Fenton degradation of the insecticide esfenvalerate in aqueous medium. Degradation results were significantly greater, and the rate of oxidation more rapid, using a photo-Fenton (Fe3+) process compared with its Fe2+ counterpart [1].
+
+Therefore, this research studied on UV/ozone degradation of fenvalerate in water. The effects of solvents, pH and sodium nitrate on the degradation of fenvalerate using UV/ozone were investigated. In addition, the photo-products were verified using gas chromatography - mass spectrometry (GC-MS).
+
+## Experimental
+
+### Chemicals
+
+Fenvalerate 91% was obtained from VIPESCO (Vietnam Pesticide Joint Stock Company). Acetonitrile (HPLC grade), methanol (HPLC grade) and acetone (ACS) were obtained from J. T. Baker, while acetic acid (glacial 99.8%) was from Merck.
+
+### Photodegradation Procedure
+
+Fig. 1 depicts a schematic diagram of UV/ozone system. A Lino JSC Lin 4.2x model ozone generator (Vietnam) with ozone production capacity of 2 g/h was applied as ozone source. The output flow from ozone generator was split to two streams and the active one adjusted by a rotameter (0.3+-3 L/min) flowing through the reactor. UV irradiation used a BT 014 model 14W UVC-lighting lamp (1.5 cm i.d., 20 cm length; Taiwan). The reactor was filled with 2 L of test solution and covered by an aluminum foil to avoid any UV leakage. Photo-degradations were conducted for two cases: ozone and non-ozone. In those experiments, initial concentration of fenvalerate was 50 mM containing 5% solvent which was acetone, acetonitrile or methanol. In UV/Ozone process, the effects of pH and sodium nitrate on fenvalerate degradation were also investigated.
+
+### Analytical Procedures
+
+Fenvalerate and photoproducts were analyzed by Agilent 7890A gas chromatography coupled to an Agilent 5975C mass spectrometer. Qualitative analysis was carried out on a HP-5MS (30m x 250\(\upmu\)m i.d.;
+
+Fig. 1: Schematic diagram of the experimental UV/ozone0.25\(\mu\)m). For GC method, the GC oven temperature was programmed as follows: 80\({}^{\circ}\)C, hold for 0 min, 10\({}^{\circ}\)C/min to a next temperature of 220\({}^{\circ}\)C, hold for 1 min, 5\({}^{\circ}\)C/min to a final temperature of 280\({}^{\circ}\)C, hold for 13 min. Injector and detector temperature were set at 280 and 300\({}^{\circ}\)C respectively. The 1 \(\mu\)L sample was injected into GC with split (5:1) mode. Helium was used as carrier gas at the rate of 1 mL/min.
+
+Quantitative analysis was carried out by Agilent 1200 HPLC. In this study, a Gemini C18 column (5 \(\mu\)m, 4.6 x 250 mm, Phenomenex) was applied. The mobile phase was methanol/acetic acid 0.01% (85:15). The chromatographic conditions were: oven temperature 40\({}^{\circ}\)C; flow rate 1.0 mL/min; injection volume 80 \(\mu\)L; and the Diode Array Detector (DAD) registered signals at wavelengths of 204, 212 and 254 nm.
+
+## 3 Results and Discussion
+
+### UV/ozone Process
+
+When using ozone, fenvalerate degraded slowly. The degradation yield was only 5% after 90 min of reaction time (Fig. 2a). Its degradation rate was faster (above 80%) under UV irradiation. In the 5% acetone solution, UV/ozone combination enhanced and accelerated the degradation of fenvalerate because fenvalerate was degraded not only by direct photolysis but also by hydroxyl radicals mechanism [5] and [6]. On the other hand, UV/ozone treatment was more actively than individual agent in 5% ACN solution, and photo-degradation of fenvalerate was rapidly.
+
+Fenvalerate's degradation yields of UV/ozone process in ACN or MeOH were higher compared in acetone (Fig. 2b). In 5% ACN or MeOH solution, it took only 5 minutes to degrade 99% fenvalerate, while approximately 20% removal was achieved in 5% acetone solution. The rate of photo-degradation in MeOH was lower than in ACN but significantly faster than in acetone. The Neperian Logarithm of percentage fenvalerate versus reaction time was plotted and found to be linear. Similar to Liu's result, UV/ozone degradation rate of fenvalerate could be described by the kinetic equations of pseudo-first order reactions [2].
+
+Fig. 3a demonstrates that the degradation efficiency also increased as pH of reaction solution increased. It can be explained that the hydrolytic stability of fenvalerate in alkaline medium is lower than in acidic medium [7]. On the other hand, decomposed ozone generates more free radicals in water as pH increases [5]. Higher pH lightly increases the degradation yield, but pH adjustment will increase more costs in the real treatment systems [8]. Beltra'n mentioned in his book that nitrites and nitrates, usually found in natural water, also act as indirect photosensitizers to produce secondary oxidants such as hydroxyl radicals [5]. The presence of sodium nitrate enhanced degradation yield, and reached a maximum at sodium nitrate concentration of 2.5 mM (Fig. 3b). Liu also reported a similar result within range of suitable concentration around 2 mM [2].
+
+### Photo-degradation Products
+
+Fig. 4 shows HPLC chromatograms of UV/ozone treated fenvalerate samples in different mediums. On these chromatograms, fenvalerate (peak III) was presented by the split peak due to its stereoisomers. Peaks I and II, before fenvalerate peak, were products of photo-degradation process because they did not appear at initial time of the reaction and increased due to the increase of treatment time. After 5 minutes, fenvalerate was completely degraded except 5% acetone solvent. However, peaks I and II still existed even when treatment time was extended to 90 minutes. It means that the products from photo-degradation of fenvalerate were more difficult to degrade comparing with fenvalerate. In contrast to ACN and MeOH, acetone solvent reduced the photo-degradation of fenvalerate as in Fig. 4a.
+
+To confirm structure of photo-products, 1000 mM fenvalerate sample in 100% ACN treated with UV was analyzed with GC-MS as in Fig. 5a. Some photo-products were detected using NIST 2008 library and some of them were tentatively assigned based on their mass spectrum and referred Katagi and Liu's researches on photo-degradation of fenvalerate [2] and [9]. The results are shown in Table 1 and the formation pathways of photoproducts are demonstrated in Fig. 5b.
+
+## 4 Conclusions
+
+UV/ozone process showed an enhancement and acceleration on degradation of fenvalerate compared with using single process (UV or ozone). UV/ozone degradation rate of fenvalerate could be described by the kinetic equations of pseudo-first order reactions. Degradation rate in MeOH was lower than in ACN but it was significantly faster than in acetone. The treatment process in 5% ACN solution obtained 99% removal yield after 5 minutes of reaction time. When pH was increased, the degradation efficiency was also increased. Moreover, the presence of sodium nitrate enhanced degradation yield, and the degradation efficiency reached an optimum sodium nitrate concentration of 2.5 mM. Finally, UV/ozone degradation process had more efficient than other processes so that it could be applied as a water treatment method for contaminated fenvalerate in water.
+
+## Acknowledgements
+
+This study was financially supported by the VNU-HCM, and the Key Laboratory of Chemical and Petroleum Technology, VNU-HCM.
+
+## References
+
+* [1] Colombo R., Ferreira T. C. R., Alves S. A., and Lanza M. R. V. Photo-Fenton degradation of the insecticide esfenvalerate in aqueous medium using a recirculation flow-through UV photoreactor. Journal of Hazardous Materials 2011; 198: 370-5.
+* [2] Liu P., Liu Y., Liu Q., and Liu J. Photodegradation mechanism of deltamethrin and fenvalerate. Journal of Environmental Sciences 2010; 22: 1123-8.
+* [3] Ma Y., Chen L., Lu X., Chu H., Xu C., and Liu W. Enantioselectivity in aquatic toxicity of synthetic pyrethroid insecticide fenvalerate. Ecotoxicology and Environmental Safety 2009; 72: 1913-8.
+* [4] Lutnicka H., Bogacka T., and Wolska L. Degradation of pyrethroids in an aquatic ecosystem model. Water Research 1999; 33: 3441-6.
+* Chapter 1. Marcel Dekker; 2003.
+* [6] Catalkaya E. C. and Kargi F. Color, TOC and AOX removals from pulp mill effluent by advanced oxidation processes: A comparative study. Journal of Hazardous Materials 2007; 139: 244-53.
+* [7] Lee P. W. Fate of fenvalerate (Pydrin insecticide) in the soil environment. J. Agric. Food Chem. 1985; 33: 993-8.
+* Chapter 3
+- Advanced Oxidation Processes. Center for Groundwater Restoration and Protection -National Water Research Institut; 1999.
+* [9] Katagi T. Photodegradation of the pyrethroid insecticide esfenvalerate on soil, clay minerals, and humic acid surfaces. J. Agric. Food Chem. 1991; 39: 1351-6.
+
+\begin{table}
+\begin{tabular}{l l l l l} Peak & RT (min) & Product & Fig. 4 & Fig. 5b & m/z (\({}^{\text{a}}\) Molecular ions ) \\
+1 & 6.828 & 4-chloro-benzaldehyde & & 139\({}^{\text{a}}\), 111, 75, 50 \\
+2 & 9.888 & 4-Chlorobenzoic acid, 2-methoxyethyl ester & & 139, 111, 75 \\
+4 & 13.790 & 3-phenoxy-benzaldehyde & c & 198\({}^{\text{a}}\), 181, 167, 141 \\
+6, 7 & 28.5-29.2 & 3-(4-chlorophenyl)-4-methyl-2-(3-phenoxyphenyl)pentanenitrile & I, II & b & 375\({}^{\text{a}}\), 209, 167, 125 \\
+8, 9 & 31.7-32.3 & Fenvalerate & III & a & 419\({}^{\text{a}}\), 209, 181, 167, 125 \\ \end{tabular}
+\end{table}
+Table 1: Photochemical degradation intermediates of fenvalerate

manual_annotation/200pdfs_in_mmd/10.1016_j.biortech.2012.10.086.mmd

@@ -0,0 +1,116 @@
+Ultrasonic-assisted ozone oxidation process of triphenylmethane dye degradation: Evidence for the promotion effects of ultrasonic on malachite green decolorization and degradation mechanism
+
+Xian-Jiao Zhou, Wan-Qian Guo, Shan-Shan Yang, He-Shan Zheng, Nan-Qi Ren
+
+State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, PR China
+
+###### Abstract
+
+This study aimed to prove the promotion effects of ultrasonic on malachite green (MG) decolorization in the ultrasonic-assisted ozone oxidation process (UAOOP), and propose the possible pathway of MG degradation. The decolorization of MG followed an apparent _pseudo first-order_ kinetic law (initial MG concentration 100-1000 mg/L). When ultrasonic (US) was applied with ozone simultaneously, the apparent _pseudo-first-order_ rate constant (\(\text{K}_{\text{app}}\)) increased, and the time MG decolorized to the half of initial concentration (\(T_{1/2}\)) shortened 185 s (1000 mg/L). Moreover, the stoichiometric ratio (\(\text{Z}_{\text{app}}\)) between \(\text{O}_{3}\) and MG was enhanced by US to 2.0 ml, saving 11% oxidant addition, comparing to individual ozone process. These results indicated that the application of US can reduce reaction time and dose of ozone addition. The possible pathway of MG degradation included three major approaches. And the result suggested that the reaction between MG and hydroxyl radical was substitution reaction rather than adduct reaction.
+
+## 1 Introduction
+
+As a representative substance of triphenylmethane dye, malachite green (MG) has been extensively used in textile industry for dyeing wood and silk, paper, and in leather industry (Gregory, 2009). Consequently, high concentration of MG is always present in industrial wastewaters, and it is harmful to human due to its genotoxicity, mutagenicity and carcinogenicity (Berberidou et al., 2007; Ju et al., 2008), by accumulating in fat tissue (Ju et al., 2008). Therefore, removing MG from wastewater before discharge is necessary and important (Zhou et al., 2012). MG is resistant to the conventional wastewater treatment process as it is designed to resistant the photochemical degradation (Zhang et al., 2008). Moreover, other chemical and physical treatments are not quite proper because of secondary pollution. So, it is necessary to explore novel and efficient methods to remove hazardous MG dye from wastewater (Zhou et al., 2012).
+
+Researchers have tested various methods, such as photo oxidative, H2O2 oxidation, ultrasonic process, Fenton (Berberidou et al., 2007; Meric et al., 2005; Zhang and Zheng, 2009) and ozone oxidation (Ai et al., 2010; Guinea et al., 2009). The ozone oxidation was one of the most effective oxidation processes, because of its oxidizing potential (2.07 V). Although ozone was able to break the basic structure of contaminants at the early stage of the reaction, it cost a great quantity of ozone to decolorize dyes-containing wastewater (Zhou et al., 2012).
+
+Ultrasonic-assisted techniques have been developed to promote the treat-effects in dye wastewater treatment field (Kang and Hoffmann, 1998; Zhang et al., 2006). The ultrasonic (US) assisted ozone oxidation process (UAOOP) utilized US forming cavitation micro-bubbles to degrade contamination. The UAOOP had been reported to degrade effectively dyes, phenol, chlorophenol, nitrobenzene, pesticide and pharmaceutical (He et al., 2009). The US induced micro-bubbles growing and then bubbles collapsing violently. The rapid implosion of bubbles is accompanied by adiabatic heating of the vapor phase of the bubble that yields localized transient high temperature (4200 K) and pressure (975 bars). And the conditions supply heat and energy required in the decolorization process of the dyes. As the precious studies, the UAAOOP technology can also improved the wastewater and sludge biodegradability and reduced biological toxicity (Xiong et al., 2011).
+
+Despite of advantages UAAOOP possessed, the degradation mechanism of MG during UAAOOP was unclear. Most studies believed that N-demethylation is a necessary step. Ju et al. (2008) proposed the MG degradation was the cleavage of central carbon of MG, the N-demethylation process, and then rings opening to form small molecular species, during the microwave assisted photo-degradation catalyzed by TiO2. In theory, all the above reactions are possible.
+
+This study aimed to get the evidence that the US can promote the decolorization efficiency of MG simulated wastewater, compared to individual ozone process. In addition, the possible degradation mechanism of MG was proposed and the degradation intermediates were examined.
+
+## Experimental section
+
+### Reagents and materials
+
+Malachite green (MG, CL No. 510-13-04, analytical grade), was purchased from Tianjin Chemical Reagent Co., Inc. (China) without any purification. Double distilled water was used in all the experiments, to prepare the stock solution of MG. NaOH or HCl was used to adjust the pH of the MG solution.
+
+### Apparatus and analysis methods
+
+#### Apparatus
+
+The ultrasonic-assisted ozone reactor, a sieve plate tower with 5 plates, was original designed in our laboratory, made of polymethyl methacrylate (Zhou et al., 2012). The simulated MG wastewater was 500 mL in each batch cycle. Ozone was generated using an ozone generator with gaseous flow meter (JT-Y-10B, Jingtian environmental technology Co. Ltd., Shandong, China). Ultrasonic was generated by an ultrasonic generator (FS-300, 20 kHz, Shengxi Ultrasonic Instrument Co., Shanghai, China) equipped with a titanium probe transducer, 8 mm in diameter. The MG simulated wastewater was loaded into the reactor by a metering pump (Model: YZ1515X, Baoding Longer Precision Pump Co. Ltd., Hebei, China), at the rate of 250 mL/min.
+
+#### Analysis methods
+
+The ozone concentration in gas into the reactor was determined as 7.5 x 10-4 M (35.85 mg/L). Ozone concentration in the gas was measured iodometric by absorbing the ozone containing gas in acid potassium iodide aqueous solutions and using sodium thiosulfate to reduce the liberated iodine. In water, ozone concentration was followed with the Karman indigo method (Beltran et al., 2006). HPLC-MS/MS (Finnigan, LCO DECA XP MAX) was equipped with an XBridge-C18 HPLC column (150 mm x 2.1 mm x 3.5 mm, waters) to detect and identify degradation products such as MG and other intermediates. The mobile phase was acetonitrile/water (gradient elution, V/V, the program was shown in Table 1 (S)). A 10 mL volume was injected using the auto sampler. In MS analysis, the ionization for MS was operated at APCI mode with positive ion mode. The color of the reaction medium was monitored by a spectrophotometer (SHIMADZU, UV-2550) at 617 nm to evaluate the decolorization effect of MG. Color removal rate of MG was calculated as:
+
+\[\text{Color~removal~rate}~\% = \frac{A_{0} - A_{\text{t}}}{A_{0}} \times 100\%\]
+
+where \(A\)0 and _A_t are the initial absorbance and the measured absorbance of the samples each time interval, t = 0, 2, 4, 6, 8, 10, 12 min.
+
+## Results and discussion
+
+### MG decolorization by UAAOOP
+
+#### Promotion effects of US on the stoichiometry (Zapp) between O3 consumption and MG degradation
+
+\[\text{O}_{3} + Z_{\text{app}}(\text{Malachite~Green}) \rightarrow \text{decolorized~product}\]
+
+The stoichiometry ratio (Zapp) represents the number of contaminants moles which can be oxidized by one mole O3 (Eq. (2)).
+
+The Zapp was 2.0 mol, which was received through experiments and calculating in the present study. It showed that one mole O3 decolorized 2.0 mol MG dye in UAAOOP. In the individual O3 process without ultrasonic irradiation, it was only 1.8 mol dyestuff oxidized by per mole O3 (Fig. 1), with initial concentration of MG ranged from 2.74 x 10-4 to 27.4 x 10-4 M. The Zapp value increased in UAAOOP compared to the individual O3 process. It is indicated that the oxidation efficiency was promoted by ultrasonic irradiation.
+
+#### Promotion effects of US on the pseudo-first-order degradation rate contents
+
+In present study, it was observed that the apparent _pseudo-first-order_ rate constant (_K_app) decreased with the initial MG concentration increased, in the UAAOOP system and individual O3 process (Fig. 2). The decolorization of MG in the UAAOOP could be regarded as a mass transfer process coupled with chemical reactions, i.e., gaseous ozone was absorbed into the aqueous phase and then reacted with the dissolved MG(Zhang et al., 2006). As the precious literature hypothesis (Supplement Eq. (1)-(5)), Then _K_app can be expressed as,
+
+\[K_{\text{app}} = Z_{\text{app}}k_{\text{L}}\alpha C_{\text{A}}^{\ast} \times \frac{E}{C_{\text{M}\text{G}\text{G}\text{G}}}\]
+
+The _Z_app, volumetric mass transfer coefficient _k_L and ozone equilibrium concentration _C_A_CMC(i) are all fixed. Although the enhancement factor E increases with dye concentration, it is not proportional to dye concentration. Therefore, the _K_app decreased with initial dye concentration.
+
+Fig. 1: The stoichiometric ratio of malachite green in UAAOOP and individual O3 process.
+
+As shown in Fig. 2, the MG decolorization by O3 and UAOOP followed apparent pseudo-first-order kinetics, agreeing with other works (Kusvarun et al., 2011; Zhang et al., 2006). A series of \(K_{\rm app}\) (shown in Table 1, from 0.493 to 0.066 min-1) were higher than those (from 0.468 to 0.054 min-1) achieved in individual O3 processes.
+
+From the point of dynamics theory, T1/2 represents the time MG decolorized to the half of initial concentration. For the apparent pseudo-first-order reaction, the value of T1/2 could be calculated as:
+
+\[T_{1/2}=\frac{\rm Ln2}{K}\times 60(s) \tag{4}\]
+
+Compared to the individual O3 process, T1/2 shortened 185 s at initial concentration of 1000 mg/L, and the time reduced nearly by 23% (Table 1).
+
+The \(T_{1/2}\) decreased as \(K_{\rm app}\) value increasing. The enhanced \(K_{\rm app}\) and the shorter \(T_{1/2}\) indicated that US did promote the reaction between ozone and MG. The efficiency of US alone in the MG degradation was very small, which was similar to others study (Zhang et al., 2006).
+
+### UV-vis spectrum of US assisted ozone degradation of MG
+
+The variation of absorption spectra of MG (ranged from 200 to 800 nm), during the UAOOP was shown in Fig. 1 (S). It was observed that the absorbance declined rapidly until it nearly disappeared in 10 min at the maximum absorbance peak 617 nm. The absorbance peaks at 423 and 314 nm had obviously declined, which indicated the whole conjugated chromophore structure of MG had been destroyed. It was possible that chromophoric group decomposition reaction, N-demethylation reaction or other reactions composed the main parts of the MG degradation pathway.
+
+### Degradation intermediate products of MG during the UAOOP
+
+#### Identification of the intermediates by US assisted ozone treatment
+
+Nine different intermediates were detected and identified by HPLC-MS/MS, formed in MG degradation during the UAOOP (Table 2 (S) and Fig. 2 (S)).
+
+According to the mass spectral analysis, they were assigned as following: A, (m/z 329.4) as MG with maximum absorption(617 nm); B(m/z 315.4) as (p-dimethylaminophenyl)(p-methylaminophenyl)phenylmethylmethylium(DM-PM); C, (m/z 301.3) as (p-methylaminophenyl)(p-methylaminophenyl)First, the cleavage of the central carbon reaction (decomposition of the conjugated structure) producing the main product DLBP (structure G, Criegeg intermediate (Kang and Hoffmann, 1998). It was the ozone molecule directly reacted with MG. And the decomposition of the conjugated structure directly caused the decolorization of MG. Second, the N-demethylation reaction resulted in a series of N-demethylations, which was produced through the broken bond between N-C. All N-demethylation intermediates (B, C, D, E, and F) were further degraded into small molecular species.
+
+Third, the substitution reaction of the hydroxyl radical with MG could produce H with m/z) 361, and 1 with m/z) 345. The basic structure of MG is aromatic structure, so it was usually regarded as the substitution reaction to keep the aromatic structure of secondary-products (Beltran, 2004). And the group of -NR2 was a positive group which made the MG molecule active to take part in the substitution reaction. Ortho-position and para-position were likely to be substituted by the - OH radicals. Therefore, the products H and 1 were detected by the HPLC-MS. The third approaches of MG degradation pathway was regarded as substitution reaction with.OH radicals rather than adduct reaction.
+
+## 4 Conclusions
+
+The evidence for the promotion effects of US was acquired. The enhanced _K_app indicated that UAOOP technique was able to shorten \(T\)1/2, leading to reduce the reaction time. Furthermore, the enhanced _Z_app indicated that UAOOP technique was in favor of saving O3 dose.
+
+Four N-demethylation intermediates and two substitution reaction products were detected by HPLC-MS/MS. MG molecule was directly oxidized to DLBP by ozone and DLBP structure was also identified. The MG possible degradation pathway involved three major steps: N-demethylation reactions, the substitution reaction with.OH radicals and the cleavage reaction of the central carbon.
+
+## Acknowledgements
+
+This research was supported by National Nature Science Foundation of China (Grant Nos. 51121062 and 51008105). The authors also gratefully acknowledge the financial support by State Key Laboratory of Urban Water Resource and Environment (Grant No. 2010DX11) and the Fundamental Research Funds for the Central Universities (Grant NO.HT.NSRIF.2013108).
+
+## References
+
+- Ai, Li, J.P., Zhang, L.Z., Lee, S.C., 2010. Rapid decolorization of azo dyes in aqueous solution by an ultrasound-assisted electrocatalytic oxidation process. Ultrason. Sochenol. **17** (2) 370-375.
+- Beltran, F.J., 2004. Ozone Reaction Kinetics for Water and Wastewater Systems. CRC Press, Weinheim, Germany.
+- Beltran, F.J., Rodriguez, E.M., Romero, M.T., 2006. Kinetics of the ozonation of muconic acid in water. J. Hazard. Mater. 138 (3), 534-538.
+- Berberidou, C., Poulios, I., Kekouloulakis, N.P., Mantzazinos, D., 2007. Sonolytic photocatalytic and sonophotocatalytic degradation of malachite green in aqueous solutions. Appl. Catal. B. Environ. 74 (1-2), 63-72.
+- Gregory, P., 2009. Does and dye intermediates: In: Kirk-Othmer Encyclopedia of Chemical Technology, vol. 18, pp. 1-66.
+- Guinea, E., Brillas, E., Centellas, F., Canizares, P., Rodrigo, M.A., Saez, C., 2009. Oxidation of enofloxacin with conductive-diamond electrochemical oxidation, ozonation and Fenton oxidation. A comparison. Water Res. 43 (8), 2131-2138.
+- He, Z., Zhu, R., Xu, X., Song, S., Chen, J., Xia, M., 2009. Ozonation combined with sonolysis for degradation and detoxification of m-nitrotoluene in aqueous solution. Inf. Eng. Chem. Res. 48 (12), 5578-5583.
+- Ju, Y.M., Yang, S.G., Ding, Y.C., Sun, C., Zhang, A.Q., Wang, L.H., 2008. Microwaves assisted rapid photocatalytic degradation of malachite green in 110. suspensions: mechanism and pathways. J. Phys. Chem. A 112 (44), 1172-1177.
+- Kang, J.W., Hoffmann, M.R., 1998. Kinetics and mechanism of the sonolytic destruction of methyl tert-butyl ether by ultrasonic irradiation in the presence of ozone. Environ. Sci. Technol. 32 (20), 3194-3199.
+- Kuswaran, E., Culnaz, O., Samli, A., Yildirim, O., 2011. Decolorization of malachite green, decolorization kinetics and stoichiometry of ozone-malachite green and removal of antibacterial activity with ozonation processes. J. Hazard Mater. 186 (1), 133-143.
+- Meric, S., Selcuk, H., Belgiorno, V., 2005. Acute toxicity removal in textile finishing wastewater by Fentorn's oxidation, ozone and coagulation-flocculation processes. Water Res. 39 (6), 1147-1153.
+- Xiong, Z.L., Cheng, X., Sun, D.Z., 2011. Pretreatment of heterocydic pesticide wastewater using ultrasonic/ozone combined process. J. Environ. Sci. China 23 (5), 725-730.
+- Zhang, H., Duan, L., Zhang, D., 2006. Decolorization of methyl orange by ozonation in combination with ultrasonic irradiation. J. Hazard. Mater. 138 (1), 53-59.
+- Zhang, H., Lv, Y.J., Liu, F., Zhang, D.B., 2008. Degradation of Cl acid orange 7 by ultrasound enhanced ozonation in a rectangular air-lift reactor. Chem. Eng. J. 138 (1-3), 231-238.
+- Zhang, Z., Zheng, H., 2009. Optimization for decolorization of azo dye acid green 20 by ultrasonically and H2O3 using response surface methodology. J. Hazard. Mater. 172 (2-3), 1388-1393.
+- Zhou, X.J., Guo, W.Q., Yang, S.S., Ren, N.Q., 2012. A rapid and low energy consumption method to decolorize the high concentration triphenylmethane dye wastewater: operational parameters optimization for the ultrasonic-assisted ozone oxidation process. Bioresour. Technol. 105, 40-47.

manual_annotation/200pdfs_in_mmd/10.1016_j.catcom.2009.01.021.mmd

@@ -0,0 +1,100 @@
+Catalytic efficiency and stability of cobalt hydroxide for decomposition of ozone and \(p\)-chloronitrobenzene in water
+
+ZhenZhen Xu, ZhongLin Chen, Cynthia Joll, Yue Ben, JilMin Shen, Hui Tao
+
+# Abstract
+
+Cobalt hydroxide, a stable and efficient catalyst prepared in the laboratory, has been successfully used in the decomposition of ozone and trace quantities of \(p\)-chloronitrobenzene (\(p\)CNB) in water. The cobalt hydroxide was characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM) and the Brunauer-Emmet-Teller (BET) method. The decomposition rate of aqueous ozone was increased by 1.527 times in the presence of cobalt hydroxide. Increasing the catalyst loading from 0 to 500 mg/L increased the removal efficiency of pCNB from 59% to 99%. The catalyst morphology and its composition were found to be unaltered after the catalytic reaction. After five successive recycles, the catalyst remained stable in the catalytic ozonation of \(p\)CNB.
+
+## Introduction
+
+Ozone has recently received much attention in water treatment technology, due to its high capacity for oxidation and disinfection. The ozonation process is a proven technology for removing various types of environmental pollutants, such as dyes, herbicides and other pesticides in aqueous solution [1; 2; 3]. It is well known that the reaction of ozone with organic substrates follows two reaction mechanisms [4]: (1) molecular ozone oxidation (direct reaction), and (2) free radical oxidation (indirect reaction). Recent studies on ozonation focus on the enhancement of ozone gas-li-quid mass transfer and the production of OH radicals (.OH). For lower cost and simpler operation, catalytic ozonation has been considered to be an efficient method in studies reported in the last few years [5]. Promising results have been obtained using metal oxides, supported metals or carbon catalysts as the catalyst for ozonation [6].
+
+Hydroxides are well known as the simplest compounds formed by highly charged cations, not only in neutral and basic, but also in acidic, water solutions. Hydroxides of many metals are widely used as, for example, sorbents, ion exchangers and collectors of admixtures. Despite some properties of hydroxides being highly attractive for catalysis in general, and for catalytic oxidation processes in water solutions in particular [7], hydroxides have not yet found extensive application as catalysts and have not yet been practically studied as catalysts in catalytic ozonation processes. Some hydroxides, such as FeOH, have received some attention [8; 9; 10], but, to date, studies of cobalt hydroxide as a catalyst for ozonation have not been reported.
+
+As a typical chemical raw material, halogenated nitroaromatic compounds are widely used in fields such as agricultural chemicals, medicines, dyes, lumber preservation and materials synthesis. The nitro group and the halogen atom are both electron-withdrawing groups, resulting in the density of the electron cloud of the benzene ring being decreased compared to benzene itself. Halogenated nitroaromatic compounds are therefore less susceptible to electrophilic attack, e.g. from oxidase in biological systems. As a result, halogenated nitroaromatic compounds undergo limited biodegradation and are likely to be found as contaminants in environmental systems. For example, \(p\)-chloronitrobenzene (\(p\)CNB), a typical and ordinary halogenated nitroaromatic compound, has already been detected in many lakes and rivers [11; 12]. Methods for removal of these halogenated nitroaromatic compounds from aqueous environmental systems must therefore be developed. One possible method for their removal is catalytic ozonation.
+
+The objectives of this study were (1) to identify the characteristics of cobalt hydroxide catalyst, (2) to investigate the catalytic efficiency of cobalt hydroxide for the decomposition of ozone and \(p\)CNB in water, and (3) to explore the stability of the catalyst cobalt hydroxide.
+
+## Materials and methods
+
+### Materials
+
+Milli-Q ultrapure water (specific resistance >=18 M\(\Omega\) cm) was used throughout these experiments. A stock solution of _p_CNB (99.5% purity, Chem Service, USA) was prepared with a concentration of 100 mg/L and stored in an amber flask. All other chemicals were reagent grade or HPLC grade when available, and used without further purification.
+
+The catalyst, cobalt hydroxide, was prepared in the laboratory by the alkali precipitation method. The precipitate was obtained by slowly mixing aqueous solutions of 1 mol/L NaOH and 0.5 mol/L cobalt nitrates with magnetic stirring. The suspension was aged at 60 degC and pH 12 for over 72 h. The precipitate was then collected and repeatedly rinsed with water until the pH and conductivity of the rinse water remained constant in three consecutive rinses. The precipitate was dried at 70 degC for 16 h and then ground. Particles with diameter less than 0.35 mm were used as catalyst for the decomposition of ozone and trace _p_CNB in water. The characteristics of the catalyst were investigated by X-ray diffraction (XRD), scanning electron microscopy (SEM) and the Brunauer-Emmet-Teller (BET) method, and the results are listed in Table 1.
+
+### Experimental procedure
+
+Catalytic activity experiments were carried out in a 1200 mL flat-bottomed flask as the reactor vessel. Ozone was generated from pure oxygen by an ozone generator (DHX-SS-16, Harbin Jiu Lietrochemistry Engineering Ltd, China). Ultrapure water (1 L), with pH pre-adjusted by addition of aqueous solutions of NaOH or HClO4, was transferred into the reactor. Using a silicon dispenser, ozone was bubbled into the reactor to the desired concentration, which was tested by the indigo method. Then, the catalyst (100 mg) and stock _p_CNB solution (1 mL) were immediately dosed into the reactor to achieve concentrations of 100 mg/L and 100 mg/L, respectively. A magnetic stirrer was used to achieve sufficient mixing of ozone solution with the catalyst and stock _p_CNB solution, so that a favorable mass transfer rate was expected. The reaction temperature was maintained at 20 +- 1 degC by a thermostatted water bath. Samples (50 mL) were taken at specific time intervals (0, 3, 5, 10 and 20 min), and the residual ozone was instantly quenched by 0.1 mL of aqueous Na2S2O3 solution (0.1 mol/L). The quenched samples were then analyzed by gas chromatography (GC) for _p_CNB quantification.
+
+### Analytical methods
+
+XRD powder patterns were recorded on a diffractometer (D/max-fB, Japan) using Cu Ka radiation (_l_ = 0.15418 nm, 40 kV voltage, 150 mA electric current, 0.02deg of step, and 10-90deg of scanning range). Scanning electron microscopy (SEM, Hitachi S-4700, Japan) was used in imaging. Prior to SEM measurements, the samples were mounted on a platform using PVC glue and were then gold-coated by a sputter. The plate containing the sample was placed in the electron microscope for analysis with magnification of 10,000. The textural properties such as BET specific surface area, pore volume and pore size were determined by nitrogen adsorption. Nitrogen adsorption-desorption isotherms were recorded on a commercial gas adsorption system (ASAP 2020 M, Micromeritics, USA). The density of surface hydroxyl groups, expressed in mmol per unit gram, was measured according to a method described by Tamara and Tanaka [13]. The pHosc (pH of point-of-zero charge) of the catalyst was measured with a powder addition method [14]. The isoelectric point was determined by a Zeta-Meter System (Zetasizer Nano, Malvern, British).
+
+The concentration of aqueous ozone was determined by the indigo method at 612 nm with a UV-visible spectrophotometer (T6, Beijing, China) [15], _p_CNB was extracted from water samples using hexane and the hexane extracts were analyzed using a GC equipped with an electron capture detector (68900, Agilent, USA) and a capillary column (Hewlett Packard HP-5, 15 m x 0.53 mm x 1.5 mm, USA) [12]. Metals leaching from the catalyst into the solution were determined by inductively coupled plasma atomic emission spectrometer (ICP-AES, Optima5300DV, Perkin Elmer, USA).
+
+## Results and discussion
+
+### Characterization of the catalyst
+
+The crystal structure of cobalt hydroxide (unused catalyst) was examined by XRD as shown in Fig. 1. From the peak intensity of the diffractograms, this pink powder has XRD patterns which correspond to b-Co(OH)2 in the JCPDS databank. The surface configuration with thin hexagonal platelets was clearly observed in the SEM micrograph (Fig. 2a), which showed the typical appearance of the cobalt hydroxide b. The average size of the hexagonal platelets was in the range of 50-200 nm.
+
+### Catalytic efficiency of the catalyst
+
+The decomposition of ozone in water at pH = 7.5 +- 0.1, both with and without cobalt hydroxide, was studied in order to determine the role of cobalt hydroxide on the decomposition of dissolved ozone. From Fig. 3A, the decomposition of ozone followed first-order kinetics both with and without cobalt hydroxide. The first-order decomposition of aqueous ozone was obviously enhanced in the
+
+\begin{table}
+\begin{tabular}{c c c} Parameter & Before use & After use \\ BET surface area (m2/g) & 23.9 & 29.3 \\ Pore volume (single point, cm3/g) & 0.088 & 0.132 \\ SET pore size (Å) & 147.54 & 132.82 \\ Particle size (nm) & Hexagonal & Hexagonal \\ platelets \textless{} 200 & platelets \textless{} 200 & platelets \textless{} 200 \\ pHosc (isoelectric point) & 7.34 & 7.25 \\ pHosc (isoelectric point) & 10.5 & 10.3 \\ Amount of surface hydroxyl groups & 3.20 & 3.22 \\ (mmol/g) & & \\ \end{tabular}
+\end{table}
+Table 1: Major characteristics of cobalt hydroxide.
+
+Figure 1: XRD pattern of cobalt hydroxide catalyst, (A) before use, and (B) after use, in catalytic ozonation experiments.
+
+presence of cobalt hydroxide, with its decomposition rate increasing by 1.527 times.
+
+Ozonation of _p_CNB both with and without cobalt hydroxide was then studied and the normalized concentration of _p_CNB versus time is shown in Fig. 3B. Compared to 60% removal efficiency of _p_CNB with ozonation alone, _p_CNB was more effectively degraded (98%) at the reaction time of 30 min in the catalyzed ozonation process. It is proposed that cobalt hydroxide induced and promoted the chain decomposition of ozone, which leads to the generation of hydroxyl radicals. In this way, the dominant oxidant in the process was not ozone but rather hydroxyl radicals. Hence, cobalt hydroxide showed a great effect on the catalytic decomposition of ozone and _p_CNB. Fig. 3B also shows the solution cobalt ion concentration during the course of the ozonation experiment. During the catalytic ozonation of _p_CNB, the plateau concentration (0.017 +- 0.002 mg/L) of cobalt cation in water was observed after the reaction time of 10 min. Ozonation of _p_CNB in the presence of 0.05 mg/L dissolved \(C\)02+ (added in the form of \(C\)0L2) was also investigated under the same experimental conditions, but the increase in removal of _p_CNB was less than 2%, so it is the cobalt hydroxide solid which is providing the catalytic effect on the decomposition rather than the \(C\)02+ in solution.
+
+### Catalyst dose
+
+Catalyzed ozonation of _p_CNB using various concentrations of cobalt hydroxide (50-500 mg/L) was then studied. As demonstrated in Fig. 4, increasing the cobalt hydroxide dose significantly increased the decomposition of _p_CNB when the concentration of ozone was constant, indicating that the dominant oxidant in the reaction increased when the catalyst dose was increased. Increasing the catalyst loading from 0 to 50, 100, 200 and 500 mg/L increased the _p_CNB removal efficiency from 59% to 80%, 94%, 98% and 99%, respectively, at reaction time 30 min. These results provide further evidence for the generation of some other, more effective oxidant (such as hydroxyl radical) when cobalt hydroxide is used as a catalyst in the process and illustrate the effectiveness of cobalt hydroxide as a catalyst in catalyzed ozonation. The removal efficiency of _p_CNB at the different catalyst concentrations was found to be directly related to the ozone decomposition rate, presumably because the increased amount of catalyst provided more active sites for adsorption and further decomposition of ozone and _p_CNB. However, when the catalyst loading was more than 100 mg/L, increasing the cobalt hydroxide concentrations resulted in only low increases in the removal efficiency of _p_CNB. This was possibly due to more combination of the hydroxyl radicals with each other when more hydroxyl radicals were produced at the higher catalyst concentrations. The formation of dissolved cobalt ion during the catalytic ozonation process was determined, with concentrations of cobalt ion after 30 min reaction time being found to be 0.009, 0.015, 0.039 and 0.057 mg/L when 50, 100, 200 and 500 mg/L of catalyst was used, respectively. These results suggest that increasing the
+
+Fig. 3: Cobalt hydroxide catalyzed decomposition of (A) ozone and (B) _p_CNB ([Oα] = 1 mg/L, [p-C(OH)] = 100 mg/L, pH = 7.5 ± 0.1, \(t\) = 20 ± 1 °C, _p_CNB = 100 mg/L).
+
+Fig. 2: SEM micrographs of cobalt hydroxide catalyst, α: before use, and β: after use, in catalytic ozonation experiments.
+
+catalyst loading is likely to bring proportional increases in cobalt dissolution, although the overall amount of this dissolution is minimal.
+
+### Catalyst stability
+
+Since the stability and reusability of the catalyst are important factors in catalyzed reactions, especially for practical industrial applications, the stability of the cobalt hydroxide catalyst was investigated by reuse of one sample of catalyst in five successive ozonation experiments. All of the catalyst reuse experiments were carried out under identical reaction conditions. At the end of each catalytic ozonation process, the insoluble catalyst was isolated and then rinsed gently with boiled Milli-Q ultrapure water. The washed catalyst was dried under atmospheric conditions, ready for use in the next ozonation experiment. Results of the catalyzed ozonation of _p_CNB with the recycled sample of catalyst are shown in Fig. 5. In the five catalytic ozonation experiments, 99%, 97%, 93%, 88% and 86% of decomposition of _p_CNB was observed at reaction time 30 min. The ozonation efficiencies of the first and second use of the catalyst were very similar. However, slight decreases in _p_CNB removal efficiencies were observed with further reuse of the catalyst sample, although the efficiencies still remained very high. Similar results were reported by Rivas and coworkers using cobalt-alumina catalysts in aqueous ozone decomposition reactions [16]. ICP analysis showed that, after 30 min of reaction time, 0.027, 0.034, 0.041, 0.043 and 0.062 mg/L of cobalt cation was detected in the reaction solution in the five successive reuses of the catalyst, respectively. Less than 0.1 mg/L of cobalt cation was observed in all recycles; the percentage of cobalt leached from the catalyst was less than 0.03% for all reuses of the catalyst. These results indicate that the catalyst has excellent long-term stability.
+
+The excellent consistency of the catalytic activity is likely to be attributable to the stable structure of the catalyst, which is indicated by the XRD, SEM and BET measurements. Fig. 1 shows the XRD patterns of the used and unused catalysts. There is no distinct difference between the two patterns, indicating that no obvious change in cobalt hydroxide crystal structure occurred during the catalytic reaction. From the morphology in the SEM micrograph, the used catalyst (Fig. 2b) seems to be identical to the fresh catalyst (Fig. 2a), with both existing as thin hexagonal platelets in the configuration. As can be seen in Table 1, no significant changes were observed in the BET surface area, pore volume (single point), BET pore size or particle size after use of the catalyst. The XRD, SEM and BET results confirmed that the cobalt hydroxide catalyst was stable during the decomposition of ozone and _p_CNB and that the catalyst is suitable to reuse in repeated decomposition experiments.
+
+## Conclusions
+
+Synthesised cobalt hydroxide was observed to have the typical appearance of cobalt hydroxide b, with thin hexagonal platelets. This cobalt hydroxide was found to be a useful catalyst for the decomposition of aqueous ozone and heterogeneous ozonation of trace concentrations of _p_CNB in water. The addition of cobalt hydroxide increased the decomposition efficiency of ozone and _p_CNB as compared to ozone alone. Increasing cobalt hydroxide doses resulted in significantly increased decomposition efficiencies of _p_CNB. It is proposed that cobalt hydroxide induces and promotes the chain decomposition of ozone, which leads to the generation of hydroxyl radicals. The catalyst was found to be stable and to retain its catalytic activity for up to at least five successive recycles. Studies of fresh and used catalysts by XRD, SEM and BET confirmed the stability of the catalyst. Cobalt dissolution was found to be less than 0.1 mg/L in all experiments. Overall, cobalt hydroxide catalyst was found to be an efficient and potential catalyst in the industrial application of the decomposition of ozone and _p_CNB in water.
+
+## Acknowledgements
+
+This project was supported by the National Natural Science Foundation of China (50578052) and the National High Technology Research and Development Program of China (2007A06Z339).
+
+## References
+
+* (1) S.D. Lambert, N.J.D. Graham, R.T.roll, Ozone Sci. Eng. 15 (1993) 457.
+* (2) V. Chu, C.W. Ma, Water Res. 34 (2000) 3153.
+* (3) Y.Hu, T. Morita, Y. Magara, T. Aizawa, Water Res. 34 (2000) 2215.
+* (4) J.N. Wu, K. Rudy, J. Sparks, Adv. Environ. Res. 4 (2000) 339.
+* (5) B. Legbue, N. Karpel Veil, Leitner, Metal Today 53 (1999) 61.
+* (6) F.J. Beltran, Ozone Reaction Kinetics for Water and Wastewater Systems, CRC Press, London, 2003.
+* (7) C.L. Elizavova, G.M. Zhidimov, V.N. Parmon, Catal. Today 58 (2000) 71.
+* (8) M. Muruganandham, J.J. Wu, Catal. Commun. 8 (2007) 668.
+
+* (9) J.-S. Park, H. Choi, J. Cho, Water Res. 38 (2004) 2285.
+* (10) T. Zhang, J. Ma, I. Mol. Catal. A: Chem. 279 (2008) 82.
+* (11) Z.L. Chen, J.M. Shen, X.Y. Li, F. Qi, B.B. Xu, J. Chem. Ind. Eng. 10 (2006) 2439.
+* (12) J.M. Shen, Z.L. Chen, Z.Z. Xu, X.Y. Li, B.B. Xu, F. Qi, J. Hazard. Mater. 152 (2008) 1325.
+* (13) H. Tamura, A. Tanaka, K.-Y. Mita, R. Furuichi, J. Colloid Interface Sci. 209 (1999) 225.
+* (14) G. Newcombe, R. Hayes, M. Drikas, Colloid Surf. 78 (1993) 65.
+* (15) H. Bader, J. Hogine, Water Res. 15 (1981) 449.
+* (16) J. Rivas, F. J. Beltran, E. Vera, O. Gimeno, J. Environ. Sci. Health A: Toxic. Hazard. Sub. Environ. Eng. 39 (2004) 2915.

manual_annotation/200pdfs_in_mmd/10.1016_j.cattod.2012.03.059.mmd

@@ -0,0 +1,223 @@
+Experiment and modeling of advanced ozone membrane reactor for treatment of organic endocrine disrupting pollutants in water
+
+Hung Lai Ho, Wai Kit Chan, Angelique Blondy, King Lun Yeung, Jean-Christophe Schrotter
+
+# Abstract
+
+An advanced ozone membrane reactor that uses membranes for ozone distribution, reaction contact and selective water separation was used for ozone treatment of a recalcitrant endocrine disrupting compound in water. Experiments and model calculation were employed to examine the ozonation of phthalate in the new reactor. Experimental results showed that fast ozone mass transfer rate is responsible for membrane reactor's superb performance compared with a semibatch reactor. Selective water removal further enhanced phthalate conversion and TOC removal by concentrating the pollutants in the reaction zone. Clean water was produced by membrane separation. Mathematical model was used to investigate the effect of membranes, reactor design and reaction operation on pollutant conversion and removal.
+
+## Introduction
+
+There is growing evidence that a large number of chemical compounds found in common household products ranging from medicines, cosmetics and personal care products, and household cleansers, can survive state-of-the-art wastewater treatment processes to contaminate surface and ground waters [1-10]. Many of these compounds are endocrine disruptors and studies carried out in 2001 [11] and 2003 [12] showed that a large numbers of endocrine disrupting chemicals (EDCs) survive conventional drinking water treatment processes to persist in finished, potable water. Several of the compounds were even found in samples of human blood, milk and urine [13]. Although health risk from chronic exposure to EDCs in humans has not been adequately addressed, their effects on normal hormonal processes is well documented and there is extensive evidence of their adverse effects in wildlife [14]. It is particularly disturbing that most studies in animal models [15] showed that early life stages are the most vulnerable to the actions of EDCs, putting pregnant women and children at greater risk [16].
+
+Ozone and membrane processes are technologies that showed the best promise for treatment of EDCs in water [17,18]. However, ozone treatment alone suffered from slow mineralization rates [19] and the UV/O\({}_{3}\) and O\({}_{3}\)/H\({}_{2}\)O\({}_{2}\) used by various authors [20-22] to remedy this shortcoming are often expensive and complex. Nanofiltration and reverse osmosis membrane can remove most EDCs [23,24], but the separated EDCs require further treatment. In addition, membrane fouling is a concern [25]. Ozone has been used in membrane to alleviate membrane fouling by organic matters [26,27] and improves membrane filtration processes [28]. Shanbhag et al. [29] explored in their early work the use of a silicone capillary membrane for ozone distributor for treatment of chemical micropollutants in water. More recently Karnik et al. [30] employed a catalytic membrane based on ultrafiltration and ozonation for drinking water treatment. This work examines an advanced ozone membrane reactor for treatment of recalcitrant organic EDCs in water. The membrane reactor uses membranes for ozone distribution, reaction contactor and water separation in a compact unit that synergistically combines ozone oxidation and membrane separation to achieve greater treatment efficiency. The use of multi-functional membrane reactor was shown to benefit gas-phase and liquid phase reactions [31-38]. Indeed, prior works showed the new membrane reactor increased mineralization rate, improved ozone utilization, reduced membrane fouling and enhanced clean water production [39,40]. This study investigates the design of the reactor and membranes using a mathematical model to guide the optimization and scale-up the process 
+
+## 2 Experiment and model
+
+### Potassium hydrogen phthalate (KHP) ozonation reaction kinetics
+
+The KHP ozonation reaction scheme in Fig. 1 was mainly based on the experimental results [39], with some of the pathways deduced from literature reports on dimethyl phthalate (DMP) [41]. The main intermediates and products include large intermediates with molecular structure similar to KHP (II), intermediate ketones, aldehydes and carboxylic acids (I2) and simple carboxylic acids such as formic and acetic acids (I3) that have low UV detection. The reaction rate equations are fitted to a second order reaction kinetic typical for ozonation reactions [42,43]. The reaction rates were obtained from a semibatch reactor using a glass spraper for ozone distribution. Ozone was produced from pure oxygen by corona discharge in Wedeco ozone generator. The gas pressure and ozone concentration were measured by an on-line pressure gauge and ozone analyzer (BMT 964(r)), respectively. The ozone gas volumetric flowrate was regulated by a Teflon flow meter (Cole Parmer(r)). The 250 mL semibatch reactor was filled with 150 mL of 250 ppmC, KHP solution and sparged with 200 sccm 120 ppm O3/O2. Samples were taken at fixed time intervals and purged with nitrogen gas. 10 mL samples were poured into a 50 mL conical flask and the dissolved ozone was measured by iodometric titration. 0.3 g M and six drops of 0.4 M sulfuric acid were added to the solution and titrated using 0.002 M Na2S2O3 solution and starch indicator. The pH was measured and the organics were analyzed by Water Acquity UPLC equipped with BEH C18 column and a PDA eA UV-vis detector and Shimadzu total organic analyzer (TOC-V CSH). The concentrations of the intermediates and products were obtained by calculating their sensitivity factors (SFi) from Eqs. (1) and (2).
+
+\[\text{TOC}_{\text{Cal}} = \sum\limits_{i = 1}^{N}\text{SF}_{i}\text{X}\text{P}\text{A}_{i}\]
+
+\[\text{SSE}_{\text{TOC}} = \sum\limits_{i = 1}^{N_{\text{H}}}\left( \text{TOC}_{\text{Meta}} - \text{TOC}_{\text{Cal}} \right)^{2}\]
+
+where PAi is the peak area of species \(i\) (AU); SFi is the sensitivity factor for species \(i\) (ppmC/AU) obtained at minimum SSEroc (i.e., squared errors of TOC); TOCcal is the calculated TOC (ppmC); TOCMeta is the measured TOC (ppmC); and \(N\) is number of species.
+
+### Advanced ozone membrane reactor experiments
+
+A schematic drawing of the advanced ozone membrane reactor is shown in Fig. 2. Ozone gas bubbles were fed by a 35 mm long porous stainless steel membrane (i.e., 0.2 mm) welded to the stainless steel reactor shell. The stainless steel membrane was purchased from Mott Metallurgical and had inner and outer diameters of 12.5and 15.5 mm, respectively. The membrane was heat treated in air at 773 K overnight, cleaned in 0.1 N HCl solution, before rinsing with water and ethanol. This pretreatment ensures fine gas bubbles were generated uniformly over the entire membrane. Alumina membrane from Pall-Exekia with inner and outer diameters of 6.5 and 10 mm, were cut into 75 mm lengths and end-sealed with glass enamel. A thin layer of zeolite was grown on the inner tube surface for water separation, while the coarse \(\alpha\)-Al2O3 ceramic tube served as the membrane contact. The zeolite membrane was grown by seeding and regrowth method, but unlike previous works [44-50], a template-free synthesis solution was used to grow the ZSM-5. Seeding is important for controlling the zeolite growth and membrane properties [51-55]. The membrane separator/contact was held in place by a pair of O-rings, giving the advanced ozone membrane reactor a reactor volume of 2 cm\({}^{3}\), membrane distributor area of 13.7 cm\({}^{2}\) and membrane separator/contact area of 9.2 cm\({}^{2}\). Thus, the membrane area-to-reactor volume ratio for the membrane distributor was 690 m\({}^{-1}\), and 460 m\({}^{-1}\) for the membrane separator.
+
+The advanced ozone membrane reactor operates in continuous mode. A 250 ppm phthalate solution was prepared from potassium hydrogen phthalate (KHP, 99.9%, Sigma-Aldrich). The solution was fed to the membrane reactor by a Watson-Marlow peristaltic pump. Once steady-state flow condition was reached, 130 ppm O3/O2 gas mixture was bubbled through the membrane distributor at a flow rate of 20 sccm. The ozone gas mixture was produced by Wedeco GSO 20 ozone generator and the ozone concentration was monitored by a set of internal and external (BMT) ozone analyzers. The ozone gas flow was regulated by Teflon flow meter from Cole-Palmer. The gas and liquid from the retentate outlet were separated, and the disengaged ozone gas was analyzed to measure the outlet concentration. Samples of the retentate liquid were titrated for dissolved ozone, and the remaining liquid was purged with nitrogen to remove the dissolved ozone and quench the reaction. The organics in the retentate liquid were analyzed by Water Acquity UPLC and Shimadzu TOC analyzer. Reactions carried out using the membrane distributor and contactor will be referred to as membrane zonication.
+
+Water separation and removal was accomplished by pervaporation across the zeolite membrane deposited on the inner surface of the membrane separator/contactor unit. A vacuum pressure of 60-100 Pa was maintained in the permeate-side by BOC-Edward vacuum pump. Water was selectively pervaporated across the zeolite membrane, while the organics were retained in the reaction zone for deeper oxidation. The clean permeate stream from
+
+Fig. 1: Proposed mechanism for ozonation of KHP molecules in water.
+
+the advanced ozone membrane reactor was collected in a cooled condenser and samples were withdrawn regularly. The amount was weighed to determine the average flux. The composition and organic carbon content of the reetentate and permeate samples were analyzed by UPLC and TOC to calculated the overall KHP and TOC removal efficiencies.
+
+### Advanced ozone membrane reactor modeling
+
+The advanced ozone membrane reactor consisted of three sections; the reetentate, membrane and permeate. The reactor was considered to be isothermal and isobaric. A uniform ozone distribution by the membrane distributor was assumed in the model. The fluid flows in the reetentate and membrane were considered along the axial and radial directions, and the diffusions of the organics in the membrane were calculated using a transient 1D convection and diffusion model. A pseudo steady-state, plug flow condition was assumed for the reetentate stream and well-mixed for the permeate stream. The reetentate stream and membrane separator/contactor were coupled by convective and diffusive fluxes of ozone, reactants and products across their boundaries. Water and organic flux across the ZSM-5 membrane was modeled using experimental and diffusion data, and assumed to be uniform. Thus, the composition of the permeate stream can be calculated from the component fluxes across the zeolite membrane. Ozone flux across the membrane was assumed negligible, which is consistent with experimental observation. Thus, ozone oxidation reaction can be ignored in the permeate stream. It was further assumed that the membrane was inert and do not catalyze the reaction.
+
+The material balance in the reetentate section is shown in Eqs. (1)-(5) for ozone gas (Eqs. (3) and (4)), dissolve ozone in liquid (Eqs. (5) and (7)) and organic compounds (Eqs. (6) and (7)). The concentrations of ozone and organics within the porous membrane contact are accounted for by the component material balance Eqs. (8)-(11), and their concentration profiles were calculated by finite element orthogonal collocation method. These equations along with the superficial gas velocity (Eqs. (12) and (13)) and reetentate liquid velocity (Eqs. (14) and (15)) were solved using the permeation (Eq. (19)), convective (Eq. (20)) and diffusive fluxes (Eq. (21)). Based on the concentrations at the boundary between the membrane separator and the permeate section, the permeate fluxes of water and organics were calculated and their concentrations in the permeate stream can be obtained (cf. Eqs. (16) and (20)).
+
+Material balance equations for gas phase (for \(h_{0,\text{sf}}\geq h\geq h_{0,\text{s}}\)): ozone gas
+
+\[\frac{dC_{A}^{G}}{dh}=\frac{(C_{\text{s}}^{G}-C_{A}^{G})\overset{\text{(}}{}^{ \text{s}}A_{C}-k_{\text{f}}a_{\text{s}}\beta((C_{A}^{G}/k_{H})-C_{A}^{G})}{V \text{s}}\quad\text{for}\ h\geq h_{0,\text{s}} \tag{3}\]
+
+gas phase boundary conditions:
+
+\[\text{at}\ h=h_{0,\text{s}}:\quad C_{A}^{G}=C_{\text{s},\text{d}}^{G} \tag{4}\]
+
+Material balance equations for liquid phase (for \(h_{f}\geq h\geq 0\)): dissolved ozone:
+
+\[\frac{dC_{A}^{G}}{dh}=\frac{k_{\text{f}}a_{\text{s}}\beta((C_{A}^{G}/k_{H})-C_ {A}^{G})+R_{A}^{\text{f}}+\text{Dif}_{A}^{\text{f}}}{V\text{s}} \tag{5}\]
+
+KHP and reaction products:
+
+\[\frac{dC_{A}^{\text{f}}}{dh}=\frac{R_{\text{f}}^{\text{f}}+\text{Dif}_{A}^{ \text{f}}}{V_{\text{r}}} \tag{6}\]
+
+Liquid phase boundary conditions:
+
+\[\text{at}\ h=0:\quad C_{A}^{\text{f}}=0,\quad C_{B}^{\text{f}}=C_{\text{s}, \text{d}}^{\text{f}},\quad C_{j}^{\text{f}}=0 \tag{7}\]
+
+Material balance equations in the porous membrane contact
+
+For ozone, KHP and its degradation intermediates:
+
+\[\frac{\partial C_{i}^{\text{A},j}}{\partial t}+N_{w}^{\text{w}}\frac{\partial C _{i}^{\text{A},j}}{\partial r}-D_{i}^{\text{A}}\left(\frac{\partial^{2}C_{i}^ {\text{A},j}}{\partial r^{2}}+\frac{1}{r}\frac{\partial C_{i}^{\text{A},j}}{ \partial r}\right)-R_{i}^{\text{A},j}=0 \tag{8}\]
+
+initial condition:
+
+\[\text{at}\ t=0:\quad C_{i}^{\text{A},j}=0\quad\text{for}\quad r_{\text{in}} \leq r\leq r_{\text{out}} \tag{9}\]
+
+Figure 2: Schematic drawing of the advanced ozone membrane reactor.
+
+boundary conditions:at \(r=r_{\text{out}}\):
+
+\[D_{i}^{\text{AI}}\frac{\partial C_{i}^{\text{AI},j}}{\partial r}\Bigg{|}_{r=r_{ \text{out}}}-N_{w}^{\text{f}}(C_{i}^{\text{AI},j}|_{r=r_{\text{out}}}-C_{i}^{ \text{L},j})=0 \tag{10}\]
+
+and at \(r=r_{\text{in}}\):
+
+\[D_{i}^{\text{AI}}\frac{\partial C_{i}^{\text{AI},j}}{\partial r}\Bigg{|}_{r=r_{ \text{in}}}-N_{w}^{\text{f}}(C_{i}^{\text{AI},j}|_{r=r_{\text{in}}}+n_{i}^{ \text{f}})=0 \tag{11}\]
+
+The superficial gas flow velocity (\(\dot{q}\)) is calculated from:
+
+\[\frac{d\dot{q}}{dh}=0\ \ \ \text{for}\ \ h<h_{0,\phi}\ \ \text{and}\ \ h>h_{0,\text{f}} \tag{12}\]
+
+\[\frac{d\dot{q}}{dh}=\frac{\dot{q}\dot{q}s}{L_{0,i}A_{c}}=\bar{\nu}s\ \ \ \text{for}\ h_{0,\phi}\leq h\leq h_{0,\text{f}} \tag{13}\]
+
+While the retentate liquid flow velocity (\(\dot{v}^{\prime}\)) is obtained from:
+
+\[\frac{d\dot{v}^{\prime}}{dh}=0\ \ \ \text{for}\ h>h_{\text{mf}} \tag{14}\]
+
+\[\frac{d\dot{v}^{\prime}}{dh}=\frac{N_{w}^{\text{f}}\text{a}_{m}}{L_{m}A_{c}}\ \ \ \text{for}\ \ 0\leq h\leq h_{\text{mf}} \tag{15}\]
+
+Also,
+
+Permeate concentration (\(C_{i}^{\text{P}}\)):
+
+\[C_{i}^{\text{P}}=\frac{\sum_{j=1}^{N_{\text{oA}}n_{i}^{\text{f}}}r_{i}^{j}}{ \sum_{j=1}^{N_{\text{oA}}n_{i}^{\text{f}}}N_{w}^{\text{f}}} \tag{16}\]
+
+Gas volume holdup (\(\beta\)):
+
+\[\beta=8.42(\nu^{8}) \tag{17}\]
+
+Lumped ozone mass transfer coefficient (\(k_{\text{L}}a_{\text{B}}\)):
+
+\[k_{\text{L}}a_{\text{B}}=0.1995(\dot{q})^{-0.246} \tag{18}\]
+
+Permeation flux (\(N_{w}^{\text{f}}\)):
+
+\[-N_{w}^{\text{f}}=P_{w}^{\text{f}}(P_{w}^{\text{f}}x_{w}^{\text{f}}|_{r=r_{ \text{in}}}-P_{w}^{\text{f}}) \tag{19}\]
+
+Solute permeation flux (\(N_{i}^{\text{f}}\)):
+
+\[-N_{i}^{\text{f}}=P_{i}^{\text{r}}(C_{i}^{\text{AI},j}|_{r=r_{\text{in}}}-C_{i}^{\text{P}}) \tag{20}\]
+
+Convective flux (\(\text{C}_{i}^{\text{f}}\)):
+
+\[\text{C}_{i}^{\text{f}}=N_{w}^{\text{f}}C_{i}^{\text{L},j} \tag{21}\]
+
+Diffusive flux (\(\text{D}\text{f}_{i}^{\text{f}}\)):
+
+\[\text{D}\text{f}_{i}^{\text{f}}=-D_{i}^{\text{AI}}\left(\frac{\partial C_{k}^{\text{AI},j}}{\partial r}\right)\Bigg{|}_{r=r_{\text{out}}} \tag{22}\]
+
+## 3 Results and discussion
+
+### KHP ozonation in semibatch reactor
+
+The simplified reaction mechanism between phthalate and ozone in Fig. 1, organizes the reaction products into three main groups (i.e., I1, I2 and I3). It is believed that ozone reacts with KHP according to Criege mechanism where molecular ozone forms primary ozonide with the unsaturated bonds in the aromatic ring before decomposition into carbonyl and carboxylic compounds shown in Fig. 1. I1 consists of large intermediates deduced from the previous study on DMP ozonation. I2 are products of ozone reaction with I1 intermediates and includes citric, adipic and malonic acids, while I3 are mainly simple (C1-C4) carboxylic acids. This allows the reaction to be modeled from five reaction rate equations (cf. Table 1), instead of eighteen if all reaction intermediates and products are considered. The organic molecules can react directly with the dissolve ozone, or with the hydroxyl radicals generated when ozone decomposes. The latter reaction was ignored as it is negligible at room temperature and solution pH less than 5 [56]. According to Hoigne et al. [42,43] ozone reactions with most organic compounds can be adequately described by an irreversible, second order reaction rate equation. Thus, the ozonation of KHP and organic intermediates and products (I1, I2 and I3) were described by second order rate equations shown in Table 1.
+
+Fig. 3a plots the semibatch ozonation of KHP at room temperature. The concentrations of KHP, I1, I2 and I3 are plotted in Fig. 3a along with the total organic carbon (TOC) content. The results show that in a semibatch reactor, KHP was completely reacted in 15 min of ozonation to produce the reaction intermediates I1. The I1 intermediates are more reactive than KHP (cf. Table 2) and react rapidly
+
+\begin{table}
+\begin{tabular}{l l} \hline Species & Rate equations \\ \hline Ozone & \(R_{\text{a}}=\frac{C_{i}^{\text{L}}}{d\dot{q}}=-z_{\text{a}}k_{\text{L}}C_{i}^{\text{L}}c_{\text{A}}^{\text{C}}-z_{\text{a}}k_{\text{C}}C_{i}^{\text{C}}-z_{\text{a}}k_{\text{C}}C_{i}^{\text{C}}C_{\text{A}}^{\text{C}}-z_{\text{a}}k_{\text{C}}C_{i}^{\text{C}}-z_{\text{a}}k_{\text{C}}C_{i}^{\text{C}}C_{\text{A}}^{\text{C}}-\) \\  & \(k_{\text{a}}c_{\text{a}}^{\text{L}}-k_{\text{a}}c_{\text{a}}^{\text{L}/2}\) \\ KHP & \(R_{\text{a}}=\frac{C_{i}^{\text{L}}}{c_{\text{a}}^{\text{L}}}=-k_{\text{b}}C_{i}^{\text{L}}c_{\text{B}}^{\text{L}}\) \\ I1 & \(R_{\text{c}}=\frac{C_{i}^{\text{L}}}{c_{\text{a}}^{\text{L}}}=-k_{\text{c}}C_{i}^{\text{L}}c_{\text{A}}^{\text{C}}-k_{\text{c}}C_{i}^{\text{L}}c_{\text{C}}^{\text{L}}\) \\ I2 & \(R_{\text{b}}=\frac{C_{i}^{\text{L}}}{c_{\text{a}}^{\text{L}}}=z_{\text{a}}k_{\text{C}}C_{i}^{\text{L}}c_{\text{B}}^{\text{L}}+z_{\text{a}}k_{\text{C}}C_{i}^{\text{L}}c_{\text{B}}^{\text{L}}-k_{\text{c}}C_{i}^{\text{L}}c_{\text{B}}^{\text{L}}\) \\ I3 & \(R_{\text{b}}=\frac{C_{i}^{\text{L}}}{d\dot{q}}=-z_{\text{a}}k_{\text{C}}C_{i}^{\text{L}}c_{\text{C}}^{\text{L}}+z_{\text{a}}k_{\text{C}}C_{i}^{\text{L}}c_{\text{B}}^{\text{L}}-k_{\text{c}}C_{i}^{\text{L}}c_{\text{C}}^{\text{L}}\) \\ \hline \end{tabular}
+\end{table}
+Table 1: Reaction rate equations.
+
+Figure 3: (a) Plots of KHP, intermediates, and TOC during ozonation of 250 ppmc KHP solution at 25 °C in a semibatch reactor. (b) A plot of calculated and experimental concentrations of KHP and organic carbons from semibatch reactions at 25, 40 and 60 °C. Please note that symbols – experimental data and lines – model calculation ([O\({}_{3}\)] = 120 ppm, \(Q_{3}\) = 200 \(\text{sccm}\), and \(V_{\text{max}}\) = 150 mL)
+
+with ozone to produce less toxic I2 and I3 products. The I3 products consisting of simple carboxylic acids are refractory to ozone (cf. Table 2) and are responsible for the high TOC content of the reaction mixture (Fig. 3a). They also have larger stoichiometric coefficient and consume more ozone per molecule than KHP,11 and I2. Fig. 3a shows that there is excellent agreement between experimental data and model calculations. A semibatch reactor model based on the work of Benbelkacem and De beleflontaine [57] was used in the calculation along with the experimental rate kinetics in Tables 1 and 2. The calculated and experimental concentrations of KHP and organic carbon from experiments carried out at 25, 40 and 60 degC were plotted in Fig. 3b. The plot shows that there is good agreement between experiment and model over the different temperatures and reaction conditions used in the study.
+
+### KHP ozonation in advanced ozone membrane reactor
+
+KHP ozonation was performed in the advanced ozone membrane reactor shown in Fig. 2. The reaction was carried out with and without water separation across the zeolite pervaporation membrane. The reaction at 25 and 40 degC are plotted in Fig. 4a as a function of hydraulic retention time. The data show that a retention time of 6 min in the membrane reactor is sufficient to reacts all KHP in the water (Fig. 4a) and temperature affects TOC removal (Fig. 4b) more than KHP conversion during membrane ozonation. Selective water separation by zeolite membrane pervaporation during advanced ozone membrane reactor operation resulted in a significant enhancement in KHP and TOC removals. KHP was completely reacted within 3 min of ozonation (Fig. 4a) and TOC removal was doubled compared to membrane ozonation (Fig. 4b). Model calculations were conducted to investigate the membrane and advanced ozone membrane reactor performance for the KHP ozonation reaction. Rate equations from semibatch reaction experiments (Tables 1 and 2) were used along with the experimental ozone mass transfer rate (Eq. (16)), membrane permeation rates (Eq. (17)) and ozone solubility and diffusion data [58,59]. Fig. 4 shows that there is good agreement between experimental data and model calculations. The model successfully captured the behavior of the reaction in both membrane reactor configurations.
+
+when ozone is fed at high concentration but low flow rate as high ozone gas concentration and large interfacial area from small ozone gas bubbles favor fast ozone mass transfer rate and reaction. Fig. 5 shows that high TOC removal is also obtained at this ozone feed condition. The TOC removal has an asymptote of 60% at room temperature due to the refractory nature of the I3 products. However as most of the I3 products are carboxylic acids, it is possible to use biological method for their removal.
+
+The selective removal of water in the advanced ozone membrane reactor concentrates the organics in the reaction zone resulting in faster reaction and more complete degradation of organics as shown by the twofold increase in TOC removal (Fig. 4b). Model calculation was used to examine the effects of membrane selectivity and flux on KHP and TOC removal in the advanced ozone membrane reactor (Fig. 6a). The calculation shows that membrane flux is more important than membrane selectivity, and that a H2O/organic separation of 10 is sufficient to enhance ozone degradation of the organic pollutant. Increasing the membrane selectivity from 10 to 100,000 resulted in less than 10% increase in TOC removal, while a fourfold increase in membrane flux resulted in over 20% improvement. The relative area of the membrane separator and distributor was examined in Fig. 6b. It is important to note that the calculations were based on the tube-and-shell reactor arrangement shown in Fig. 2 and therefore the membrane area ratio is constrained to values less than 0.9. The plots show that for all residence time, KHP and TOC removal increase as the area of the separation membrane increases. This suggests that it would be beneficial to consider capillary or spiral wound membranes to attain a larger separation membrane area.
+
+## 4 Concluding remarks
+
+This work shows that an advanced ozone membrane reactor that uses membranes for ozone distribution, reaction contact and water separation could attain both higher EDC and TOC removals for KHP ozonation at a shorter residence time compared to a traditional semibatch reactor. A mathematical model based on experimental transport, permeation and reaction data gave an accurate description of the advanced ozone membrane reactor operation. The model was used to examine the reactor and reaction to explore membrane properties, reactor design as well as operating conditions to achieve the best pollutant removal and TOC degradation for the minimum amount of ozone use. Calculations showed that the best result could be obtained from redesigning the reactor to accommodate a larger separation membrane area. The study also showed that for ozonation of KHP, it is possible to achieve complete KHP conversion, but TOC removal approaches an asymptote at 60% at room temperature due to the refractory nature of the I3 products.
+
+## Acknowledgments
+
+The authors gratefully acknowledge funding from the Hong Kong Innovation Technology Fund (ITS/108/09FP) and financial supports of Chiaphua Industries Ltd., Anjou Recherches and Veolia Environment. We thank the Consulate General of France in Hong Kong and the Foreign Ministry of France for financial support for Ms. Angelique Blondy.
+
+## References
+
+* (1) G. Loraine, M. Pettiggrove, Environmental Science and Technology 40 (2006) 687-695.
+* (2) B.L. Tan, D.W. Harker, J.F. Muller, F.D. Leusch, LA. Tremblay, H.F. Chapman, Chemosphere 69 (2007) 644-654.
+* (3) A.J. Watkinson, E.J. Murphy, S.O. Scatano, Water Research 41 (2007) 4164-4174.
+* (4) T. Moluda, Y. Kobayashi, R. Nagao, N. Yamashita, H. Tanaka, S. Tanaka, S. Fujii, C. Konishi, I. Houw, Water Science and Technology 57 (2008) 65-71.
+* (5) Y.J. Zhang, S.U. Geissen, C. Cal, Chemosphere 73 (2008) 1151-1161.
+* (6) D. Simzak, J. Fujiwara, S. Manabe, M. Matsuda, M. Asami, S. Kunikane, Water Science and Technology 58 (2008) 1129.
+* (7) D. Koplan, E. Furlong, M. Meyer, E. Thurman, S. Zaugg, L. Barber, Environmental Science and Technology 36 (2002) 1202-1211.
+* (8) M.Y. Kozato, D.W. Kolpin, K.K. Barnes, E.T. Furlong, M.T. Meyer, S.D. Zaugg, L.B. Barber, M.E. Thurman, Science of the Total Environment 402 (2008) 201-216.
+* Sources, Fate, Effects and Risks, 2nd ed., Springer, 2004.
+* (10) S.S. Sunder, P. Westerhoff, Y. Yoon, D.L. Sedlak, Environmental Engineering Science 20 (2003) 449-469.
+* (11) P.E. Stackelberg, E. Furlong, M. Meyer, S. Zaugg, A. Henderson, D. Reissman, Science of the Total Environment 329 (2004) 98-113.
+* (12) P.E. Stackelberg, J. Gils, E.T. Furlong, M.T. Meyer, S.D. Zaugg, R.L. Lippincott, Science of the Total Environment 377 (2007) 255-272.
+* (13) Center for Disease Control and Prevention, Third National Report on Human Exposure to Environmental Chemicals, National Center for Environmental Health Pub. No. 05-0570,2005.
+* (14) World Health Organization, Global assessment of the state-of-the-science of Endocrine Disruptors, The International Programme on Chemical Safety, WHO/FCS/EDC/03/3, 2003.
+* (15) T.E. Stoker, L.G. Parks, L.E. Gray, R.L. Cooper, Critical Reviews in Toxicology 30 (2000) 197-252.
+* (16) A.C. Collier, Ecohealth 4 (2007) 164-171.
+* (17) T.A. Ternes, M. Meisenheimer, D. McDowwell, F. Sacher, H.J. Brauch, B.H. Gulde, (2002) 3855-3863.
+* (18) P. Westerhoff, Y. Yoon, S. Snyder, E. Wert, Environmental Science and Technology 39 (2005) 6849-6863.
+* (19) V. Camel, A. Bermond, Water Research 32 (1998) 3208-3222.
+* (20) M.C. Huber, S. Canonic, G.Y. Park, U.V. Gunten, Environmental Science and Technology 37 (2003) 5665-5670.
+* (21) R. Ermawati, S. Morimura, Y.Q. Tang, K. Liu, K. Kida, Journal of Bioscience and Bioengineering 103 (2007) 277-31.
+* (22) V.P. Zhang, J.L. Zhou, Water Research 39 (2005) 3991-4003.
+* (23) A.M. Comfort, R.C. Andrews, D.M. Bagley, C.Y. Hao, Journal of Membrane Science 313 (2008) 323-335.
+
+Fig. 6: Calculated percent removals of KHP (solid lines) and TOC (dash lines) as a function of (a) membrane selectivity at different membrane fluxes (\(\tau\) = 3.5 min), and (b) membrane areas ratio at different residence time ([KHP] = 250 ppmC, [O\({}_{3}\)] = 130 ppm, \(\Omega_{3}\) = 20 sccm, and \(\tau\) = 25°C).
+
+* (24) K. Kimura, S. Toshima, G. Amy, Y. Watanabe, Journal of Membrane Science 245 (2004) 71-78.
+* (25) S.K. Hong, M. Elimelech, Journal of Membrane Science 132 (1997) 159-181.
+* (26) S. van Geluwe, L. Braeken, B. van der Bruggen, Water Research 45 (2011) 3551-3570.
+* (27) B. Karnik, S.H.R. Davies, K.C. Chen, D.R. Jaglowski, M.J. Baumann, S.J. Masten, Water Research 39 (2004) 728-734.
+* (28) B. Schlichter, V. Mavrov, H. Chmiel, Desalination 156 (2003) 257-265.
+* (29) P.V. Shanbhag, A.K. Ciuha, K.S. Sarkar, Industrial and Engineering Chemistry Research 37 (1998) 4388-4398.
+* (30) B.S. Karnik, S.H. Davies, M.J. Baumann, S.J. Masten, Environmental Science and Technology 39 (2005) 7656-7661.
+* (31) K.L. Yeung, R. Aravind, J. Szegner, A. Varma, Studies in Surface Science and Catalysis 101 (1996) 1349-1358.
+* (32) M.A. Pena, D.M. Carr, K.L. Yeung, A. Varma, Chemical Engineering Science 53 (1998) 3821-3834.
+* (33) K.L. Yeung, R. Aravind, R.J.X. Zawada, J. Szegner, G. Cao, A. Varma, Chemical Engineering Science 49 (1994) 4823-4838.
+* (34) Y.S. Cheng, M.A. Pena, M.L. Yeung, Journal of the Taiwan Institute of Chemical Engineers 40 (2009) 281-288.
+* (35) Y. Guo, X.F. Zhang, H.Y. Zou, H.O. Liu, J.Q. Wang, KL Yeung, Chemical Communications (2009) 5858-5900.
+* (36) J.M. Mata, W.N. Lau, C.Y. Lee, P.L. Yue, C.K. Chan, KL. Yeung, Chemical Engineering Science 58 (2003) 959-962.
+* (37) Y.S.S. Nan, K.L. Yeung, A. Gvarillidis, Applied Catalysis A 281 (2005) 285-293.
+* (38) K.L. Yeung, X.F. Zhang, W.N. Lau, R. Martin-Aranda, Catalysis Today 110 (2005) 26-37.
+* (39) S. Heng, KL. Yeung, M. Djafer, J.C. Schroterter, Journal of Membrane Science 289 (2007) 67-75.
+* (40) S. Heng, KL. Yeung, A. Julbe, A. Ayral, J.C. Schroterter, Microporous and Mesoporous Materials 115 (2008) 137-146.
+* (41) Y.H. Chen, N.C. Shang, D.C. Hsieh, Journal of Hazardous Materials 157 (2008) 260-268.
+* (42) J. Hogine, H. Bader, Water Research 17 (1983) 173-183.
+* (43) J. Hogine, H. Bader, Water Research 17 (1983) 185-194.
+* (44) L.Y. Au, J.L.H. Chau, C.T. Ariso, K.L. Yeung, Journal of Membrane Science 183 (2001) 269-291.
+* (45) L.Y. Au, W.Y. Muir, S.Lu, C.T. Ariso, K.L. Yeung, Microporous and Mesoporous Materials 47 (2001) 203-216.
+* (46) W.C. Wong, I.Y.Y. Au, P.S. Lau, C.T. Ariso, K.L. Yeung, Journal of Membrane Science 193 (2001) 141-161.
+* (47) Y.S. Cheng, M.A. Pena, L.J. Ferro, D.C.W. Hui, K.L. Yeung, Journal of Membrane Science 204 (2002) 329-340.
+* (48) J.H. Chau, A.Y. Leung, K.L. Yeung, Lab-on-a-Chip 3 (2003) 53-55.
+* (49) J.H. Chau, K.L. Yeung, Chemical Communications (2002) 960-961.
+* (50) J. Motuzas, S. Heng, P.S.Z. Lau, K. Yeung, J.L. Beresnevicius, A. Julbe, Microporous and Mesoporous Materials 99 (2007) 197-205.
+* (51) G.H. Yang, X.F. Zhang, S.Qi, K.L. Yeung, J.Q. Wang, Journal of Physics and Chemistry of Solids 68 (2007) 26-31.
+* (52) X.F. Zhang, H.O. Liu, KL. Yeung, Materials Chemistry and Physics 96 (2006) 42-50.
+* (53) S.M. Lal, L.Y.Y. Au, K.L. Yeung, Microporous and Mesoporous Materials 54 (2002) 63-77.
+* (54) S.M. Kwan, K.L Yeung, Chemical Communications (2008) 3631-3633.
+* (55) S. Feng, P.S. Lau, K.L. Yeung, M. Djafer, J.C. Schroterter, Journal of Membrane Science 243 (2004) 69-78.
+* (56) M.S. Egoritz, U. von Gunten, H.-P. Kaiser, Ozone Science and Engineering 22 (2000) 123-150.
+* (57) H. Benbelkemem, H. De beliefontaine, Chemical Engineering and Processing 42 (2003) 723-732.
+* (58) J.A. Roth, D.E. Sullivan, Industrial and Engineering Chemistry Fundamentals 20 (1981) 137-140.
+* (59) P.N. Johnson, R.A. Davis, Journal of Chemical and Engineering Data 41 (1996) 1485-1487.

manual_annotation/200pdfs_in_mmd/10.1016_j.cattod.2018.01.013.mmd

@@ -0,0 +1,178 @@
+Accepted Manuscript
+
+Title: Promising application of SiC without co-catalyst in photocatalysis and ozone integrated process for aqueous organics degradation
+
+Authors: Yongbing Xie, Jin Yang, Yue Chen, Xuelian Liu, He Zhao, Yujie Yao, Hongbin Cao
+
+PII: S0920-5861(18)30013-0
+
+DOI: [https://doi.org/10.1016/j.cattod.2018.01.013](https://doi.org/10.1016/j.cattod.2018.01.013)
+
+Reference: CATTOD 11203
+
+To appear in: _Catalysis Today_
+
+Received date: 30-10-2017
+
+Revised date: 22-12-2017
+
+Accepted date: 8-1-2018
+
+Please cite this article as: Yongbing Xie, Jin Yang, Yue Chen, Xuelian Liu, He Zhao, Yujie Yao, Hongbin Cao, Promising application of SiC without co-catalyst in photocatalysis and ozone integrated process for aqueous organics degradation, Catalysis Today [https://doi.org/10.1016/j.cattod.2018.01.013](https://doi.org/10.1016/j.cattod.2018.01.013)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+**Promising application of SiC without co-catalyst in photocatalysis and ozone integrated process for aqueous organics degradation**
+
+Yongbing Xiea, Jin Yanga,b, Yue Chena, Xuelian Liua,b, He Zhaoa, Yujie Yaoa and Hongbin Caoa,b*
+
+a Beijing Engineering Research Center of Process Pollution Control, Division of Environment Technology and Engineering, Institute of Process Engineering, Chinese Academy of Science, Beijing 100190, China
+
+b School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China
+
+*Corresponding author:
+
+Tel: +86-01082544845, E-mail: hbcao@ipe.ac.cn
+
+## Highlights
+
+* A commercial SiC is very active in photocatalytic ozonation of aqueous organics
+* UV is more powerful than visible light in the SiC catalyzed combining process
+* Photo generated electron reduction of oxygen and ozone are the key steps
+* The high conducting band position of SiC benefit to the electron reduction
+* The activity of the commercial SiC is comparable to P25 TiO\({}_{2}\)
+
+## Abstract
+
+SiC is a newly developed photocatalyst, but it is often used together with a co-catalyst rather than solely for its relatively low activity. Here we reported the high activity of a commercial SiC in a photocatalysis and ozone combined process(photocatalytic ozonation). The commercial SiC showed very weak activity in photocatalysis of oxalic acid (OA) and para-hydroxybenzoic acid (PHBA) degradation, and its performance in catalytic ozonation was also not satisfied. However, the organics degradation and mineralization rates dramatically increased in the SiC photocatalytic ozonation. The different operation parameters were optimized, and the main reactive radicals in this process and its generation pathways were studied. It was found that photo-generated electron reduction of ozone and oxygen are the main pathways to produce hydroxyl radical, which was responsible to the high oxidizing ability of this process.
+
+Keywords: SiC; photocatalytic ozonation; Penicillin G; band structure; reaction mechanism.
+
+## 1 introduction
+
+Silicon carbide (SiC) has a very good chemical stability, a high thermal stability and thermal transfer ability, these make it a good support for catalyst used at high temperatures [1, 2]. It is also known as the third generation semiconductor, and has been used in the photocatalytic processes for water purification or water splitting to generate hydrogen [3, 4, 5].
+
+However, in most of these cases, it was not solely used but composited with other semiconductors [6, 7], metals/metal oxides [8, 9] or both of them [10, 11]. TiO\({}_{2}\) is the most widely used co-catalyst with SiC to prepare SiC-TiO\({}_{2}\) nanoparticles [3, 10] or TiO\({}_{2}\)/SiC nanocomposite film [12], for the application in water splitting or volatile organic compounds degradation. Recently, CdS [13, 14], BiVO\({}_{4}\)[6], C\({}_{3}\)N\({}_{4}\)[15], SnO\({}_{2}\)[16], graphene [17] and Ag\({}_{3}\)PO\({}_{4}\)[18] were also used to composite with SiC to synthesize a more active material. These co-catalysts played key roles in enhancing the light adsorption and separation of photo generated hole-electron couples, and thus increased the activity of pristine SiC. Two SiC materials with different phases can also combine together to perform a high activity [19]. Comparing the vast majority of literatures on TiO\({}_{2}\), WO\({}_{3}\), BiVO\({}_{4}\) and C\({}_{3}\)N\({}_{4}\) photocatalysis, very few papers have been reported on SiC photocatalysis, which is due to its low intrinsic photocatalytic activity. In some cases using SiC/TiO2 materials, SiC was even considered as the catalyst support but not the active component [20, 21].
+
+Ozone is a very strong oxidant and has been adopted in wastewater purification for many years [22]. When ozone is combined with photocatalysis, which is called photocatalytic ozonation, an obvious synergy have been observed in this systems using different kinds of photocatalysts, such as WO3, C3N4, BiFeO3 and TiO2, under visible light or UV light irradiation [23-26]. Recently, we discovered that the band structure of C3N4 is very crucial to its activity in the photocatalytic ozonation process [27], and the same trends were also observed in the photocatalytic ozonation process with WO3 or BiVO4 catalysts (not published yet). Under light irradiation, a semiconductor with high conducting band (CB) position can produce electrons with strong reducing ability. This benefited to the electron reduction of ozone molecule to produce 'O5' radical and it further evolved into hydroxyl radical [25], and thus a high oxidizing efficiency in photocatalytic ozonation was achieved. Based on our recent findings, SiC is expected to have an excellent performance in the photocatalytic ozonation for its high CB position, but no papers have been published on this topic till now.
+
+In this paper, a commercial SiC was applied in the photocatalytic ozonation under visible light or UV light irradiation for the first time, and its high catalytic activity was confirmed in several organics degradation. The physical structure, optical and electrochemical properties of the SiC were characterized, and its activities under different reaction conditions were intensively studied. The reactive radical in this process was recognized and its generation pathways were thoroughly discussed.
+
+## 2 Experimental
+
+### Chemicals and reagents
+
+Oxalic acid (OA) and para-hydroxybenzoic acid (PHBA) were purchased from Sinopharm Chemical Reagent Co., Ltd., China. Penicillin G was purchased from J&K Scientific. The black silicon carbide (SiC) powder was obtained from Aldrich, with a particle size of about 400 mesh. All chemicals were at least in chemical grade and were used without further purification. Ultrapure oxygen (purity 99.999%) was used to produce ozone. Ultrapure water was used for preparing all solution in this work.
+
+2.2 Materials characterization
+
+X-ray diffraction (XRD) was used to characterize the crystal phase of SiC (CuKa irradiation, X' PERT-PRO MPD). The morphology of SiC was observed by scanning electron microscopy (SEM, JSM-7610F, JEOL) and high resolution transmission electron microscopy (HRTEM, JEM-2100F, JEOL), respectively. The surface area and pore structure were measured with N2 physical adsorption at 77K (Autosorb-iQ, Quantachrome). The UV-Vis diffuse reflectance spectra (UV-Vis DRS) were recorded on a spectrometer equipped with a 60 mm diameter integrating sphere (Varian Cary 5000). The valence band (VB) of the material was analyzed by X-ray photoelectron spectroscopy (XPS) on an ESCALAB 250Xi instrument (Thermo Fisher Scientific, USA).
+
+The photoluminescence spectra (PL) of the photocatalysts were obtained at room temperature on a spectrofluorometer (NanoLOG-TCSPC, Horiba Jobin Yvon, USA) with an excitation wavelength of 400 nm. Electrochemical impedance spectroscopy (EIS) was performed by an electrochemical analyzer (Autolab PGSTAT302 N, Metrohm, Switzerland) in a three-electrode system with 0.2 M Na2SO4 as the electrolyte. A Pt electrode and a standard calomel electrode (SCE) were used as the counter electrode and the reference electrode, respectively. The working electrodes were prepared as follows: 5 mg of photocatalyst was dispersed in 1 mL of ethanol under sonication for 30 min to produce a slurry, then 10 mL of the slurry was dip-coated onto a glassy carbon electrode and exposed in air for 30 min to eliminate ethanol for further test. The measurement was conducted under open-circuited potential with signal amplitude of 5 mV and a frequency from 10\({}^{6}\) to 10-1 Hz.
+
+2.3 Ozonation, photocatalysis and photocatalytic ozonation
+
+All experiments were conducted in a 500 ml glass reactor under irradiation of a 300W Xenon lamp (CEL-NP2000, Aulight, China) at 25\({}^{\circ}\)C. The wavelength of the simulated visible light and UV light were 420-800 nm and 200-400 nm, respectively. Ozone was generated from pure oxygen by an ozone generator (Anseros COM-AD-01,Germany), and was introduced into the solution through a porous glass plate. The detailed ozone flowrate and concentration were listed in the session of results and discussion.
+
+Three organics, oxalic acid (OA), para-hydroxybenzoic acid (PHBA) and Penicillin G with different structural complexity were studied as the target pollutants. The concentrations of OA, PHBA and Penicillin G in solution were detected by high performance liquid chromatograph (HPLC, Agilent series 1260, USA) equipped with a UV-Vis spectrometer qualified at 210 nm. The mineralization extent of the organics was analyzed by a total of organic (TOC) analyzer (Shimadzu TOC-VCPH, Japan). The generation of H2O2 during the reactions was monitored with a photometric method using peroxidase (POD) and N, N-diethyl-p-phenylenediamine (DPD).
+
+## 3 Results and discussion
+Fig 1 showed the structure characterization of the commercial SiC. In the XRD pattern (Fig. 1A), three intensive peaks appeared at 35.6\({}^{\text{o}}\), 60.0\({}^{\text{o}}\) and 71.7\({}^{\text{o}}\), which represented the (111), (220) and (311) facets of cubic structured \(\beta\) phase of SiC [9]. This was completely consistent with the JCPDS 29-1128. A closed loop presented in the nitrogen physical adsorption curve, but the very small value indicated negligible pore volume. The surface area of the SiC was calculated to be 0.33 m\({}^{2}\)/g (Fig. 1B), according to the BET method. In the SEM (Fig 1C), the SiC particles showed irregular shapes and the particle size was generally in the range of 10-30 mm. In the HRTEM figure, the lattice with an interplanar space about 0.248 nm was observed, which was characteristic of (111) facet.
+
+### Optical property and band structure of SiC
+
+The UV-Vis DRS result showed that the commercial \(\beta\)-SiC has an adsorption in the visible light range, as a broad peak at 450-800 nm presented (Fig 2A). The bandgap of SiC varied from 2.3 to 3.3 eV in many published literatures [5-6, 19], due
+
+Fig. 1: XRD (A), physical adsorption (B), SEM (C) and HRTEM (D) of SiC.
+
+Fig. 2: UV-Vis DRS spectra (A), bandgap (B) and VB-XPS (C) of SiC.
+
+to their different structure and properties. In this paper, the bandgap of the SiC material was calculated to be 2.82 eV from the Tauc plot (Fig. 2B). The VB position was detected from XPS and it was 1.28 eV. Thus, the CB position was calculated to be -1.54 eV for this commercial SiC.
+
+### Catalytic activities of SiC in organics removal
+
+Being a simple aliphatic organics, OA is resistant to ozone attack but can be degraded by hydroxyl radical (OH). Here we firstly evaluated the activity of SiC in the catalytic ozonation, photocatalysis and photocatalytic ozonation of OA under UV or visible light (Fig. 2A). It was obvious that OA was hardly degraded in the three single processes. However, it was completely removed in a UV-O\({}_{3}\) process within 60 min, in which the light intensity was 157 mW/cm\({}^{2}\) and no SiC was involved. The addition of SiC further greatly accelerated OA degradation and its removal rate at 30 min achieved 93%. It was also noticed that the reaction rate in the Vis-O\({}_{3}\)-SiC process was much lower (light intensity: 180 mW/cm\({}^{2}\)), and the OA removal rate at 60 min was only 20.3%. When the intensity of visible light increased, quite different results was obtained (Fig. 2B). The removal rate of OA came to 59.4% at 30 min with a light intensity of 276 mW/cm\({}^{2}\). This is lower than the UV-O\({}_{3}\)-SiC process, but much higher than that in the UV-O\({}_{3}\) process without SiC (37.6%). Further increasing the light intensity can slightly accelerate OA removal, but this is not recommended for the
+
+Fig. 2: OA degradation in different processes (A) and visible light photocatalytic ozonation with different light intensity (B). Reaction conditions: SiC: 0.2 g/L, OA: 180 mg/L, solution: 300 mL, ozone: 20 mg/L and 100 mL/min.
+
+economic consideration. The adsorption of OA on SiC was also analyzed under the same condition. Only 1.8% of OA was absorbed on the surface of SiC after 60 min under continuous stirring. This meant adsorption took a very little part in the catalytic degradation of OA.
+
+The kinetic of oxalic acid degradation in UV-O\({}_{3}\)-SiC and Vis-O\({}_{3}\)-SiC processes was analyzed and compared. Using the data in the first 30 min, the rate constants were calculated to be 61.96\(\times\)10\({}^{-3}\) and 6.12\(\times\)10\({}^{-3}\) mmol-L\({}^{-1}\)-min\({}^{-1}\) in UV-O\({}_{3}\)-SiC and Vis-O\({}_{3}\)-SiC processes, 1.12\(\times\)10\({}^{-3}\) and 0.68\(\times\)10\({}^{-3}\) mmol-L\({}^{-1}\)-min\({}^{-1}\) in UV-SiC and Vis-SiC processes and 6.89\(\times\)10\({}^{-6}\) mmol-L\({}^{-1}\)-min\({}^{-1}\) in O\({}_{3}\)-SiC process, respectively. This showed that SiC is very inactive in catalytic ozonation, and the performance in UV or Visible light photocatalysis was much closer. But the rate constant in UV-O\({}_{3}\)-SiC was about 10 times of that in Vis-O\({}_{3}\)-SiC process, which meant SiC is more promising in the UV-O\({}_{3}\) combined process. Accordingly, the rate constant in UV-O\({}_{3}\)-SiC was 54.7 times of the sum of those in O\({}_{3}\)-SiC and UV-SiC processes, indicating a great synergetic effect in the UV-O\({}_{3}\)-SiC process. A weaker synergy was also found in the Vis-O\({}_{3}\)-SiC process, and the rate constant in Vis-O\({}_{3}\)-SiC was 8.92 times of the sum of those in O\({}_{3}\)-SiC and Vis-SiC.
+
+PHBA is a representative aromatic compound which is easily oxidized by ozone molecule. It was further adopted as the simulated pollutant, and its degradation and total mineralization was evaluated in single or combined processes (Fig. 3). The adsorption ability of SiC to PHBA was firstly detected, and it was found that about 1.0% of PHBA was removed in 60 min under the same reaction condition. In the UV or UV-SiC processes, PHBA degradation rates were lower than 10% at 60 min (Fig. 3A). When ozone was involved, it was quickly completely degraded within 15 or 30 min. Though the removal rates of PHBA were very close to each other in the ozone involved processes, the curves in TOC removal were quite different. It was noticed that SiC exhibited a moderate activity in catalytic ozonation, as the TOC removal rate increased from 38.4% in ozonation to 53.4% in SiC catalytic ozonation. The same trend was revealed in Fig 3A, as the complete degradation time of PHBA was reduced from 30 to 15 min after the addition of SiC in ozonation. SiC was also very active in photocatalytic ozonation of PHBA, and the TOC removal rate reached 95.8% at 60 min. It is 1.91 times of the sum of ozonation (38.4%) and SiC photocatalysis (1.72%), and 1.78 times of the sum of photolysis (0.3%) and SiC catalytic ozonation (53.4%). This indicated that SiC triggered a synergy between UV light and ozone for the organics mineralization.
+
+Furthermore, reaction parameters were optimized in the UV photocatalytic ozonation. We used a more complex organics penicillin G as the simulated pollutant of pharmaceutical wastewater. It was also easily oxidized by ozone molecule, but many intermediates will be generated and total mineralization became difficult. The adsorption of penicillin G on SiC was evaluated firstly and it was found that 4.8% was removed. As penicillin G was completely degraded in a few minutes, the TOC removal rates were only compared in the parameters optimization. Fig. 4A showed the influence of SiC dosage to the penicillin G mineralization, and the UV light intensity was 157 mW/cm\({}^{2}\) in this series of reactions. In all cases, SiC showed a positive effect in the organics degradation after its addition into the UV-O\({}_{3}\) system. The TOC removal rate increased from 38.7% to 66.4% when the dosage of SiC varied from 0.2 to 0.6 g/L, which was 2.45 times of that in the UV-O\({}_{3}\) process. But a higher dosage of SiC was detrimental to the light adsorption, as a result, the TOC removal rate enormously reduced to about 51%. The intensity of visible light was very important to OA degradation (Fig. 2B), it can also influence the process.
+
+Fig. 2B showed the importance of visible light intensity to OA degradation, and this was also evaluated in the SiC-UV-O\({}_{3}\) process for penicillin G mineralization (Fig. 4B). The oxidizing ability of the process was also positively related with the UV light
+
+Fig. 4: The mineralization of Penicillium G with different dosage of SiC (A), light intensity (B) and H\({}_{2}\)O\({}_{2}\) generation (C) in photocatalytic ozonation. Reaction conditions: Penicillin G: 35.6 mg/L (0.1 mM), solution: 400 mL, ozone: 20 mg/L and 100 mL/min.
+
+intensity, but the difference was not so much when the light intensity varied from 157 to 392 mW/cm\({}^{2}\). We also compared the visible light photocatalytic ozonation with a light intensity of 493 mW/cm\({}^{2}\), but the performance was much lower than the UV photocatalytic ozonation.
+
+The concentration of H\({}_{2}\)O\({}_{2}\) in the solution was monitored during photocatalytic ozonation of penicillin G, and compared in the systems with different light sources. The intensities of UV and visible light in the system were 392 and 493 mW/cm\({}^{2}\), respectively. In these two cases, the concentration of H\({}_{2}\)O\({}_{2}\) both increased and reached the maximum value in the first stage, and then the concentration decreased in different rates. There are two reaction pathways for H\({}_{2}\)O\({}_{2}\) generation in this system: (a) ozone evolution under light irradiation [28] and (b) direct ozonation of organics [29]. The accumulation of H\({}_{2}\)O\({}_{2}\) in the solution in the first stage was due to the combination of pathways (a) and (b). Much more H\({}_{2}\)O\({}_{2}\) was produced in the UV photocatalytic ozonation than that in visible light photocatalytic ozonation, because O\({}_{3}\) was more easily transformed to H\({}_{2}\)O\({}_{2}\) under UV irradiation. After 30 minutes, the concentration of H\({}_{2}\)O\({}_{2}\) decreased to different extents because persistent organic intermediates were produced and thus the pathway (b) was weakened. H\({}_{2}\)O\({}_{2}\) was also consumed in the peroxone process and it reacted with O\({}_{3}\) to produce \({}^{\bullet}\)OH [30]. This is a possible origin of \({}^{\bullet}\)OH generation and will be discussed in the next session.
+
+\({}^{\star}\)OH is widely recognized as the main reactive species in the photocatalytic ozonation. _t_-BA quenching is widely regarded as an effective and simple method to detect \({}^{\star}\)OH for their high reaction rate [31]. The addition of t-BA will consume \({}^{\star}\)OH and decrease its concentration, and thus the organic degradation will be suppressed to different contents. Here we added 7400 mg/L of _t_-BA into the OA solution, and found that OA degradation was almost completely inhibited in both cases with UV light or visible light (Fig. 5). OA was totally degraded in the UV/O\({}_{3}\)SiC system within 30 min, while this did not work after the addition of _t_-BA. This meant that \({}^{\star}\)OH is the dominant reactive species in photocatalytic ozonation of OA with SiC catalyst, which is the same to the systems with C\({}_{3}\)N\({}_{4}\), WO\({}_{3}\) and TiO\({}_{2}\) catalysts [23, 25-27].
+
+There are several possible pathways in which \({}^{\star}\)OH will be produced in photocatalytic ozonation [28]. The first (i) is electron reduction of O\({}_{3}\) to produce \({}^{\star}\)O\({}_{3}\), and it is evolved into HO\({}^{\star}\) and then \({}^{\star}\)OH; the second (ii) is electron reduction of O\({}_{2}\) to produce \({}^{\star}\)O\({}_{2}\) and it reacts with O\({}_{3}\) to produce \({}^{\star}\)OH; the third (iii) is the ozone photolysis to produce H\({}_{2}\)O\({}_{2}\) and it reacts with O\({}_{3}\) to produce \({}^{\star}\)OH. If the semiconductor is active in catalytic ozonation, \({}^{\star}\)OH can also be produced from catalytic decomposition of ozone (iv). The key steps for \({}^{\star}\)OH generation in the SiC photocatalytic ozonation will be discussed later.
+
+Fig. 5: Tert-butanol (_t_-BA) quenching in UV and visible light photocatalytic ozonation of OA. Reaction conditions: SiC: 0.6 g/L, OA: 180 mg/L, t-BA: 100 mM (7400 mg/L), solution: 300 mL, ozone: 20 mg/L and 100 mL/min, UV light intensity: 157 mW/cm\({}^{2}\) and visible light intensity: 493 mW/cm\({}^{2}\).
+
+The solution pH is very important to many processes for organics degradation, especially to the photocatalytic ozoantion. Here we studied the effect of pH in the UV photocatalytic ozonation. The initial pH of the OA solution was 2.5, and it was adjusted to near neutral (6.8) or alkaline (11.0) with NaOH solution, respectively. Fig. 6 indicated that the acid solution greatly benefited this process. When the pH increased from 2.5 to 6.8 and 11.0, the OA removal rates at 30 min decreased from 98.7% to 20.7% and 8.3%, respectively. The pH change in these solutions without bubbling was also monitored. In the alkaline solution, the pH gradually decreased from 11.0 to 10.5 possibly due to the very low removal rate of oxalate. In the acid solution and near neutral solution, the pH gradually increased from 6.8 and 2.5 to 7.6 and 4.8, respectively.
+
+combine with \(\bullet\)O\({}_{3}\) to generate HO\({}^{*}\) and it quickly decomposed into \(\bullet\)OH.
+
+P25 is the most easily accessible TiO\({}_{2}\) with high activity, which has been regarded as the benchmark in many photocatalytic reactions. To get an intuitive understand of the activity of SiC, it was further compared with P25 TiO\({}_{2}\) under UV and solar light. In both UV-O\({}_{3}\) processes with SiC or TiO\({}_{2}\), OA was completely removed in 30 min, and TiO\({}_{2}\) was slightly more active than SiC (Fig. 7A). Considering the great difference between the surface area of TiO\({}_{2}\) and SiC, the optimized dosage of TiO\({}_{2}\) may be lower than 0.6/L. So we further compared the activity of TiO\({}_{2}\) and SiC with a dosage of 0.2 g/L in photocatalytic ozonation. OA was almost totally removed in 30 min with TiO\({}_{2}\), while the removal rate was 93.0% with SiC. Furthermore, their
+
+Fig. 7: Photocatalytic ozonation of OA with SiC and TiO\({}_{2}\) catalysts (A) and stability test of SiC (B). Reaction conditions: OA: 180 mg/L, solution: 300 mL, ozone: 100 mL/min, 20 mg/L (UV) and 30 mg/L (solar), light intensity: 157 mW/cm\({}^{2}\) (UV) and 330mW/cm\({}^{2}\) (solar).
+
+activities of in simulated integration of solar light and ozone were compared. The ozone concentration was increased to 30 g/L, the light intensity was 330mW/cm\({}^{2}\)and the catalyst dosage was 0.3 g/L. OA was totally removed in 30 min again, and still it was less active than TiO\({}_{2}\). But the difference between their catalytic activities was not so much, considering that the surface area of TiO\({}_{2}\) was about 150 times higher than SiC. This indicated the great potential of SiC in photocatalytic ozonation, because its surface area can be greatly enhanced with other synthesis methods, and a much better performance in photocatalytic ozonation can be expected.
+
+The catalytic stability of SiC in OA degradation was also tested under the same UV-O\({}_{3}\) reaction condition. In each case, the solid catalyst was filtered from the solution and washed with ultrapure water and used without further treatment. Fig 7 (B) showed the activity of SiC in five cycles of UV photocatalytic ozonation of OA. It was noticed that SiC was very stable in this process. Though the OA removal rates in the third and fourth cycles slightly decreased, almost total degradation of OA was achieved again in the fifth cycle. Meanwhile, the quality of SiC catalyst in each cycle may slightly vary because of the possible mass loss during the catalyst re-collection from solution. we regarded that this commercial SiC was very stable in the photocatalytic ozonation.
+
+To understand the similar activity in photocatalytic ozonation, the intrinsic properties of SiC and TiO\({}_{2}\) in photo-generated electron-hole separation and electron transfer were evaluated. In Fig. 8(A), a strong peak of SiC appeared in in PL spectra,and this meant a low electron-hole separation efficiency. In the contrast, a weaker peak over TiO\({}_{2}\) showed its higher efficiency in charge separation. Meanwhile, the electron transfer ability of the two materials was also compared via EIS analysis. From the curves in Fig. 8(B), the resistances were calculated to be 96.0 \(\Omega\) for SiC and 100.7 \(\Omega\) for TiO\({}_{2}\). Band structure is another important parameter determining the activity of a semiconductor in photocatalysis and photocatalytic ozonation. It is widely known that the bandgap of TiO\({}_{2}\) is around 2.7 eV and the CB position is at around -0.91 V. The bandgap of SiC in this work was slightly wider (2.82 eV), but the CB position was much higher (-1.54 eV). This meant a much stronger reducing ability of the photo generated electrons, which is very important in photocatalytic ozonation because this will benefit ozone reduction to produce *OH [25].
+
+To conclude, this commercial SiC had a similar bandgap and electron trasnfer ability to TiO\({}_{2}\). Though it has a much weaker charge seperation capacity than TiO\({}_{2}\), this shortage is coverd in the presence of ozone molecule, because the electron reduction of ozone greatly promote the separation of electron-hole pairs. Especially the SiC has a much lower CB position, and this is definitely good for ozone molecule capturing electron. As a result, the SiC showed comparable activity to TiO\({}_{2}\) in photocatalytic ozonation of OA.
+
+## 4 Conclusions
+
+In this paper, we firstly attempted the application of a commercial SiC material in photocatalytic ozonation for wastewater treatment. Though it was very inert both in photocatalysis and catalytic ozonation, the SiC photocatalytic ozonation showed high efficiency in oxalic acid degradation and para-hydroxybenzoic acid and Penicillin G mineralization. Several reaction parameters in this process were optimized, and it was found UV irradiation and acid solution were both more conducive to this reaction system. The low efficiency of SiC in electron-hole separation was compensated by the presence of ozone molecule. The reduction of ozone and oxygen by photo generated electron are the key steps to produce *OH in SiC photocatalytic ozonation. This can exactly take the advantage of the very high CB position of SiC. The contribution of \(\bullet\)OH as the main reactive species was confirmed by \(t\)-BA quenching in the degradation of OA. This work inspires the new application of SiC in photocatalytic ozonation for wastewater treatment, and further exploration is expected to reveal how to synthesize a more active SiC material.
+
+## Acknowledgements
+
+The authors greatly appreciate the financial support from the Beijing Natural Science Foundation (8172043), National Science Fund for Distinguished Young Scholars of China (No. 51425405), National Natural Science Foundation of China (51378487) and Major Science and Technology Program for Water Pollution Control and Treatment (2015ZX07202013).
+
+## References
+
+* [1] M.J. Ledoux, S. Hantzer, C.P. Huu, J. Guille, M.P. Desaneaux, J. Catal. 114 (1988) 176-185.
+* [2] J. Zou, X. Mu, W. Zhao, P. Rukundo, Z.J. Wang, Catal. Commun. 84 (2016) 116-119.
+* [3] H. Yamashita, Y. Nishida, S. Yuan, K. Mori, M. Narisawa, Y. Matsumura, T. Ohmichi, I. Katayama, Catal. Today 120 (2007) 163-167.
+* [4] H. Liu, G. She, L. Mu, W. Shi, Mater. Res. Bull. 47 (2012) 917-920.
+* [5] D.H. van Dorp, N. Hijnen, M. Di Vece, J.J. Kelly, Angew. Chem. Int. Edit. 48 (2009) 6085-6088.
+* [6] D. Wang, Z. Guo, Y. Peng, W. Yuan, Chem. Eng. J. 281 (2015) 102-108.
+* [7] C. Gomez-Solis, I. Juarez Ramirez, E. Moctezuma, L.M. Torres Martinez, J. Hazard. Mater. 217 (2012) 194-199.
+* [8] D. Wang, W. Wang, Q. Wang, Z. Guo, W. Yuan, Mater. Lett. 201 (2017) 114-117
+* [9] Z. Wang, F.A. Armstrong, Catal. Sci. Technol. 6 (2016) 7038-7041.
+* [10] Y. Li, Z. Yu, J. Meng, Y. Li, Int. J. Hydrogen Energ. 38 (2013) 3898-3904.
+* [11] Y. Zhang, Y. Xu, T. Li, Y. Wang, Particuology 10 (2012) 46-50.
+* [12] T. Zou, C. Xie, Y. Liu, S. Zhang, Z. Zou, S. Zhang, J. Alloy. Compd. 552 (2013) 504-510.
+* [13] Y. Peng, Z. Guo, J. Yang, D. Wang, W. Yuan, J. Mater. Chem. A 2 (2014) 6296-6300.
+* [14] Y. Peng, G. Han, D. Wang, K. Wang, Z. Guo, J. Yang, W. Yuan, Int. J. Hydrogen Energ. 42 (2017) 14409-14417.
+* [15] B. Wang, J. Zhang, F. Huang, Appl. Surf. Sci. 391 (2017) 449-456.
+* [16] X. Liao, J. Chen, M. Wang, Z. Liu, L. Ding, Y. Li, J. Alloy. Compd. 658 (2016)642-648.
+* [17] Y. Wang, X. Guo, L. Dong, G. Jin, Y. Wang, X.-Y. Guo, Int. J. Hydrogen Energ. 38 (2013) 12733-12738.
+* [18] Z. Chen, F. Bing, Q. Liu, Z. Zhang, X. Fang, J. Mater. Chem. A 3 (2015) 4652-4658.
+* [19] Y. Zhang, Y. Hu, H. Zeng, L. Zhong, K. Liu, H. Cao, W. Li, H. Yan, J. Hazard. Mater. 329 (2017) 22-29.
+* [20] N.A. Kouame, D. Robert, V. Keller, N. Keller, C. Pham, P. Nguyen, Catal. Today 161 (2011) 3-7.
+* [21] A.N. Kouame, R. Masson, D. Robert, N. Keller, V. Keller, Catal. Today 209 (2013) 13-20.
+* [22] U. von Gunten, Water Res. 37 (2003) 1443-1467.
+* [23] E. Mena, A. Rey, S. Contreras, F.J. Beltran, Catal. Today 252 (2015) 100-106.
+* [24] J. Yin, G. Liao, J. Zhou, C. Huang, Y. Ling, P. Lu, L. Li, Sep. Purif. Technol. 168 (2016) 134-140.
+* [25] J. Xiao, Y. Xie, F. Nawaz, Y. Wang, P. Du, H. Cao, Appl. Catal. B.: Environ. 183 (2016) 417-425.
+* [26] Y. Jing, L. Li, Q. Zhang, P. Lu, P. Liu, X. Lu, J. Hazard. Mater. 189 (2011) 40-47.
+* [27] J. Xiao, Y. Xie, F. Nawaz, S. Jin, F. Duan, M. Li, H, Cao, Appl. Catal. B: Environ. 181 (2016) 420-428.
+* [28] A.C. Mecha, M.S. Onyango, A. Ochieng, C.J.S. Fourie, M.N.B. Momba, J. Catal. 341 (2016) 116-125.
+* [29] L. Gu, X.W. Zhang, L.C. Lei, Ind. Eng. Chem. Res. 47 (2008) 6809-6815.
+* [30] J.P. Pocostales, M.M. Sein, W. Knolle, C.V. Sonntag, T.C. Schmidt, Environ. Sci. Technol. 44 (2010) 8248-8253.
+* [31] J. Xiao, Y. Xie, F. Nawaz, S. Jin, F. Duan, M. Li, H. Cao, Appl. Catal. B: Environ. 181 (2016) 420-428.
+* [32] J. Nawrocki, B. Kasprzyk-Hordern, Appl. Catal. B: Environ. 99 (2010) 27-42.
+* [33] M.M. Sein, A. Golloch, T.C. Schmidt, C. von Sonntag, Chemphyschem 8 (2007) 2065-2067.

manual_annotation/200pdfs_in_mmd/10.1016_j.cattod.2018.06.005.mmd

@@ -0,0 +1,188 @@
+## Accepted Manuscript
+
+Title: Polymeric Catalytic Membrane for Ozone Treatment of DEET in Water
+
+Authors: Ying Li, King Lun Yeung
+
+PII: S0920-5861(18)30723-5
+
+DOI: [https://doi.org/10.1016/j.cattod.2018.06.005](https://doi.org/10.1016/j.cattod.2018.06.005)
+
+Reference: CATTOD 11494
+
+To appear in: _Catalysis Today_
+
+Received date: 1-7-2017
+
+Revised date: 18-4-2018
+
+Accepted date: 4-6-2018
+
+Please cite this article as: Li Y, Yeung KL, Polymeric Catalytic Membrane for Ozone Treatment of DEET in Water, _Catalysis Today_ (2018), [https://doi.org/10.1016/j.cattod.2018.06.005](https://doi.org/10.1016/j.cattod.2018.06.005)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+# Polymeric Catalytic Membrane for Ozone Treatment of DEET in Water
+
+Ying LI
+
+King Lun YEUNG
+
+###### Abstract
+
+The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR Department of Chemical and Biological Engineering and 4Division of Environment and Sustainability; The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong SAR
+
+Tel:+852-23587123; Fax:+852-23580054; Email: kekyeung@ust.hk
+
+###### Abstract
+
+The ozone treatment of N,N-diethyl-meta-toluamide (DEET) in water was carried out in a membrane reactor that uses ozone-resistant polymer membranes for ozone distribution, catalytic contactor and water separation. Iron oxide (ferrihydrite) nanoparticles supported on powdered activated carbon (CAT1) were used as catalyst, and were coated on the surface of the membranecontactor/distributor. Ferrihydrites promote ozone dissolution and transformation to reactive hydroxyl radicals, and accelerate the conversion and mineralization of DEET. The PAC substrate helped retain the organic pollutants in the reaction zone leading to greater conversion and deeper mineralization. Consistently, the catalytic membrane contactor/distributor displayed the highest DEET conversion and TOC reduction over the entire studied range of ozone dosage and reaction residence time. The selective removal water by the membrane separator has the desired effect of concentrating the organic pollutants resulting in higher conversion and mineralization rate. The compact membrane reactor unit equipped with catalytic membrane contactor/distributor and membrane water separator outperformed a semi-batch ozone reactor with 60% DEET conversion compared to 20%, and 30% TOC reduction versus 5%.
+
+Catalytic ozone membrane reactor; Membrane contactor; Ferrihydrite/PAC; DEET
+
+## 1. Introduction
+
+Numerous compounds from pharmaceutical and personal care products (PPCPs) are becoming ubiquitous in aquatic environment [1-4]. The environmental concentration of these pollutants ranges from tens of ng L-1 to thousands of \(\mu\)g L-1, but even at trace quantities, they have potent impact on health and environment [5,6]. They are known to survive conventional wastewater and water treatment processes to contaminate finished drinking water [7,8]. Many of these pollutants are endocrine disrupting compounds (EDCs) that can interfere with the normal hormonal functions of the body, while others are suspected carcinogens. Studies have shown that they are associated with prostate and breast cancers, metabolic diseases and disorders in reproduction, cardiovascular and neuroendocrinology systems, and are believed to play a role in the growing prevalence of these diseases [1-4].
+
+Physical, chemical and biological treatment processes including advanced oxidation, adsorption, membrane separation and biodegradation showed limited success due to the broad range of compounds being addressed and their very different physicochemical properties, reactivity, and fluctuating concentrations, as well as to their complex transformation pathways [6,9-11]. A combination of one or more processes was reported to be more effective in treating these potent pollutants [12-16]. Prior works [17-19] showed an advanced ozone membrane reactor that combines adsorption, ozone oxidation and membrane separation in a single unit operation to be particularly effective for treatment of recalcitrant EDCs in water. Adsorbents captured and trapped the pollutants in the reactor for more complete conversion and mineralization by ozone, while membrane separation produced clean water with less than 2 ppm total organic carbon (TOC). Ceramic membranes were used for ozone distribution and water separation in the advanced ozonemembrane reactor to achieve improved ozone utilization and enhanced water production [17,20].
+
+The catalytic membrane reactor in this work employed polymeric membranes instead of ceramic membranes. Ozone-resistant polyvinylidene fluoride (PVDF) microfiltration and ultrafiltration membranes were respectively used for ozone distributor/contact and membrane separator [21-23]. These hollow fiber membranes possess large surface area-to-volume ratio \(>\) 7000 m\({}^{-1}\) allowing a large membrane area to be packed into a small volume resulting in compact reactor design. Also, their competitive cost makes them increasingly attractive for industrial applications [24-29]. The PVDF microfiltration membrane serves a gas distributor producing fine ozone bubbles uniformly over the entire reactor volume, while simultaneously providing the contact surface for the adsorption and ozonation of the pollutants. Coating the catalyst on the membrane distributor/contact locates the catalysts where the ozone concentration is highest. This also avoids loose catalyst particles that could fuel and damage the ultrafiltration membrane. Selective water separation by hollow fiber, PVDF ultrafiltration membranes produces clean water and has the desired effects of concentrating the pollutants in the reactor to enhance their conversion and mineralization.
+
+N,N-diethyl-meta-toluamide (DEET) was chosen as the model pollutant in this study due to their prevalence in both surface and ground waters [30,31]. DEET is the most common active ingredient in topical insect repellant and studies indicate that DEET causes brain cell death in the region of the brain responsible for cognition, behavior and motor functions [32]. It therefore poses a significant risk to children's developmental health [33]. Iron oxide catalysts supported on high surface area activated carbon captured and transformed the pollutants during ozone reaction [34-37]. The catalyst was deposited in a thin layer on the membrane distributor/contact surface in such a way as to minimize gas flow resistance across the membrane.
+
+## 2 Experimental Methods
+
+### 2.1 Materials
+
+Polyvinylidene fluoride (PVDF) hollow fiber microfiltration (1.3 mm O.D., 1.0 mm I.D.) and ultrafiltration (1.3 mm O.D., 1.0 mm I.D.) membranes were supplied by Pall Corporation and Dow Water and Process Solutions, respectively. The microfiltration membrane has a nominal pore size of 0.1 \(\upmu\)m, while the ultrafiltration membrane has a pore size of 0.03 \(\upmu\)m. The catalysts were prepared from FeSO\({}_{4}\)\(\bullet\) 7H\({}_{2}\)O (99.8%) purchased from Sigma-Aldrich and powdered activated carbon (PAC) from Calgon Carbon WPA. The DEET (97%) was supplied by Sigma-Aldrich and was used as purchased without further purification.
+
+### 2.2 Catalyst preparation and characterization
+
+Iron oxide (ferrihydrite) on activated carbon (CAT1) was prepared from 1.0 M FeSO\({}_{4}\) solution using a modified procedure adopted from Leone and Gennari [38] and Park et al. [36]. Briefly, two grams of PAC was placed in a flask before adding 30 mL 0.1 M FeSO\({}_{4}\) solution. The suspension was stirred for an hour at room temperature, followed by dropwise addition of 2.0 M NaOH solution until the solution pH reached 7.0-8.0. The resulting catalyst (CAT1) was collected after a series of filtration and washing procedure followed by drying in an oven at 105 \({}^{\circ}\)C and 12 h.
+
+A second catalyst (CAT2) was prepared by physical mixing of iron oxide (ferrihydrite) nanoparticles and activated carbon. The ferrihydrite was prepared by slow addition of 1 M NaOH solution to 500 mL 0.2 M FeSO\({}_{4}\) solution to bring the pH to 7.0-8.0. The preparation was carried out at room temperature under well-mixed conditions, and the resulting product was dialyzed toremove the anions for five days until the water conductivity was lower than 3 \(\upmu\)Scm-1. The iron oxide nanoparticles were obtained after freeze-drying [38,39]. The catalysts were characterized by Powder X-ray diffraction (XRD, Philips PW1830), X-ray photoelectron spectroscopy (XPS, Physical Electronics 5600 multi-technique system) and transmission electron microscopy (TEM, JEM 2010F (JEOL)). The surface area and pore size distribution of the catalyst were determined by nitrogen physisorption (Coulter SA 3100). Both CAT1 and CAT2 had 80 mg Fe/g catalyst loading.
+
+Batch adsorption measurements were made at 25 \({}^{\circ}\)C from 200 ppm DEET (150 ppmC) solution and 1 g/L of the catalyst. The concentration was monitored over time by ultra-performance liquid chromatography (UPLC, Water Acquity(tm)) and analyzed for total organic content by a Shimadzu TOC-V. Adsorption studies were also carried out on an ozonated 200 ppm DEET solution. The solution was prepared by reacting a 150 mL DEET solution with 150 sccm of 120 ppm O\({}_{3}\) gas/O\({}_{2}\) produced by WEDECO GSO 20 ozone generator. The reaction was quenched after 60 minutes by purging nitrogen gas to give a solution with 110 to 120 ppmC organics. A 1 g/L of the catalyst was added and the adsorption was allowed to reach equilibrium at 25 \({}^{\circ}\)C.
+
+### Catalytic membrane preparation and characterization
+
+The PAC, CAT1 and CAT2 were first dispersed in ethanol under ultrasonic agitation for 30 min. Nafion ionomer (Sigma-Aldrich) was added as a binder and the suspension was allowed to thicken into a paste. The catalyst paste containing 18 % solids and 40% Nafion ionomer was brushed uniformly on the surface of the hollow fiber PVDF microfiltration membranes, and allowed to dry at 60 \({}^{\circ}\)C for 12 h in an oven to give a membrane with 30 wt.% of solids. The prepared catalytic membranes were examined under an optical microscope for defects before use. Scanning electron microscope (SEM JEOL-6700F) was also used to observe the detailed morphology and microstructure of the membrane and coating.
+
+### Catalytic membrane reactor and reaction
+
+A schematic drawing of the catalytic membrane reactor is shown in Fig. 1. Four fifty-five millimeter long coated PVDF microfiltration membranes served as ozone gas distributor as well as the catalytic contactor. Four fifty-five millimeter long PVDF ultrafiltration membranes were used for water separation. The hollow fiber membranes were sealed and held in place by a pair of O-rings. The reactor volume was 6 cm\({}^{3}\), while the membrane distributor/contactor and membrane separator have an area of 9 cm\({}^{2}\) each. This gave a membrane area-to-volume ratio of 150 m\({}^{2}\).m\({}^{3}\).
+
+Figure 1: A schematic drawing of catalytic membrane reactor with membrane distributor,
+
+contactor and separator.
+
+The catalytic membrane reactor was fed with 200 ppm DEET solution by Warson-Marlow peristaltic pump at 1.2 - 6.0 mL per min. Once a steady-state flow was established, 60 sccm of 120 ppm O\({}_{3}\)/O\({}_{2}\) gas mixture was fed through the membrane distributor/contactor into the flowing liquid solution. Gas and liquid were disengaged in a gas-liquid separator at the reactor exit. The unreacted ozone gas was decomposed in an ozone destructor. Samples from the retenate were titrated for ozone, and quenched by purging nitrogen gas. Water separation by PVDF ultrafiltration membrane was carried out by applying a vacuum on the permeate side of the reactor (ca. 60 - 100 Pa) using a BOC-Edward vacuum pump. Samples taken from the retenate and permeate streams were analyzed by UPLC and TOC to determine the DEET and TOC conversions.
+
+## 3. Results and Discussion
+
+### 3.1 Characteristics of catalytic adsorbents
+
+The X-ray diffraction of PAC detected the presence of SiO\({}_{2}\), Al\({}_{2}\)O\({}_{3}\) and Fe\({}_{2}\)O\({}_{3}\) in the commercial powdered activated carbon as shown in Fig. 2a. X-ray photoelectron spectroscopy indicated that there was roughly 0.2, 0.4 and 0.6 atomic percent of Si, Al and Fe on the surface of PAC (Fig. 2b). Supporting 80 mg/g ferrihydrites on PAC did not significant change the diffraction pattern. XPS measured 11.4 at.% Fe on CAT1. The ferrihydrites were high dispersed and has an estimated size of about 2 nm according to transmission electron microscopy in Fig. 3a. Energy dispersive X-ray spectroscopy in Fig. 3b confirmed that these are nanoparticles consist of iron element. The textural properties of PAC and CAT1 are summarized in Table 1. It can be seen that the addition of 80 mg/g Fe did not significantly change the surface area, pore size and pore volume from the original PAC.
+
+Figure 3: (a) Transmission electron image and (b) energy dispersive X-ray spectrum of the CAT1 catalyst.
+
+Figure 2: (a) X-ray diffraction graphs and (b) X-ray photoelectron spectra of PAC support and CAT1 catalyst.
+
+Figure 4 plots the adsorption of DEET and DEET ozonation mixture after 60 min reaction by PAC and CAT1. The DEET adsorptions on PAC and CAT1 (cf. Fig. 4a) were rapid and within ten minutes the concentration reached ninety percent of the steady-state concentration (_0.9C\({}_{s}\)_). The steady-state concentrations of 34 and 40 ppm DEET were obtained from PAC and CAT1, respectively. Figure 4b shows the adsorption from the reaction mixture was equally rapid, but CAT1 was able to lower the TOC by 20% more compared to PAC, to less than 60 ppmC. Ferrihydrites have greater capacity for adsorption of carboxylic and oxygenated compounds that are primary products of ozone reactions of DEET. Indeed prior studies showed that the affinity of the hydroxyl groups on the iron oxide particles for the carboxylic groups of natural organic matters (NOMs) was mainly responsible for their large adsorption capacity for NOM [40, 41].
+
+\begin{table}
+\begin{tabular}{c c c c} \hline  & Surface Area & Pore Diameter & Pore Volume \\  & (m\({}^{2}\)g\({}^{-1}\)) & (nm) & (cm\({}^{3}\)g\({}^{-1}\)) \\ \hline PAC & 760 & 2.4 & 0.46 \\ CAT1 & 730 & 2.7 & 0.49 \\ \hline \end{tabular}
+\end{table}
+Table 1: Textural properties of PAC support and CAT1 catalyst.
+
+### Properties of catalytic PVDF membrane
+
+The commercial PVDF microfiltration membrane was coated with uniform layer of PAC, CAT1 and CAT2. Figure 5a is a cross-section of the original microfiltration membrane. The micrograph clearly showed the graded structure of the membrane. Following the coating of the CAT1 layer in Fig. 5b, a 20 \(\upmu\)m thick, dense layer of catalyst was deposited on the surface of the membrane. Figures 5c and 5b show that the surface of the original membrane and CAT1 catalytic membranes are distinctly different. The catalyst particles were clearly trapped and embedded in the membrane pores as can see from the high magnification inset pictures. In a separate study, high pressure O\({}_{3}\)/O\({}_{2}\) mixture was sparged through the CAT1 catalytic membrane distributor/contactor in DDI water to observe whether the coated particles could be dislodged. No particles was observed in the water following the test indicating that Nafion ionomer was suitable binder for the catalyst
+
+Figure 4: Adsorption kinetics of (a) DEET and (b) DEET ozonation mixture after 60 min reaction by PAC and CAT1. C\({}_{0}\) and C\({}_{t}\) indicated the concentrations at times 0 and t, respectively.
+
+coating. Similar results were observed for PAC and CAT2 coated PVDF microfiltration membranes.
+
+### Catalytic ozonation of DEET in membrane reactor
+
+Both the original PVDF microfiltration membrane and the coated PVDF catalytic membranes produced fine bubbles of ozone in the reactor retentate. A substantial increase in DEET conversion was observed compared to reaction in semi-batch reactor even when an inert, uncoated PVDF microfiltration membrane was used (cf. Fig. S-1). The DEET conversion in the semi-batch reactor was low due to the refractory nature of DEET. The reaction carried out for 150 mL 200 ppm DEET
+
+Figure 5: Scanning electron micrographs of (a) & (b) cross-sections and (c) & (d) surfaces of the (a) & (c) original PVDF membrane and (b) & (d) CAT1 coated PVDF catalytic membrane. The insets were taken at higher magnification.
+
+and 150 sccm 120 ppm O\({}_{3}\) gave a conversion of 20 % with a TOC reduction of 5 % for the semibatch reactor using a gas sparger. A much higher DEET and TOC conversions of 45 % and 12 % when a PVDF membrane was used for ozone distribution. The result showed more than two-fold increase in DEET degradation at very short residence time of 4 min in the membrane reactor.
+
+The performance of the membrane reactors are reported in Figs. 6 and 7. Coating the PVDF membrane distributor/contactor with PAC captures has the desired effects of retaining the organic pollutants in the reaction zone for a longer time resulting in higher DEET conversion of about two-to-threefold from 5% to 15% at a residence time (\(\tau\)) of 1.5 min, and from 23% to 45% at a residence time (\(\tau\)) of 7.0 min as shown in Fig. 6a. However, the degree of mineralization remains identical for the PAC-coated and uncoated PVDF membrane (Fig. 6b). This is due to the refractory nature of the intermediate byproducts of DEET ozonation. Mixing ferrihydrite nanoparticles with PAC (CAT2) before coating onto PVDF membrane distributor/contactor did not increase DEET conversion, but improved the mineralization of DEET particularly at the longer residence time. It can be seen from the plots in Fig. 6b, that TOC reduction increased from 5% to 18% (\(\tau=6\) min) with the addition of ferrihydrites. Studies have suggested that ferrihydrite can enhance ozone dissolution and initiate ozone transformation to reactive hydroxyl radicals [42-43]. Figures 6a and 6b show that CAT1 coated PVDF catalytic membrane distributor/contactor showed the highest DEET conversions and TOC reductions over the entire range of residence time investigated. On average, a threefold increase in DEET conversion and a fivefold increase in TOC reduction were obtained from CAT1 catalytic membrane compared to the original, uncoated PVDF membrane.
+
+Figures 6c and 6d show the effects of doubling the ozone feed rate from 10 to 20 sccm. Increasing the ozone feed flow rate not only increases the ozone dosage, but more importantly, it enhances the mass transfer rate. It has the immediate effect of enhancing the mineralization of
+
+DEET as shown in Fig. 6d as the ozone dosage increased. The enhanced mass transfer rate is believed to be responsible for the higher DEET conversion rate of CAT2 in Fig. 6c. Increasing the ozone feed flow rate to the maximum allowable operational limit for the reactor of 60 sccm led to a slight improvement in DEET conversions for the coated membrane distributor/contactor (Fig. 6e). Highest DEET conversion and mineralization was obtained using the CAT1 coated PVDF membrane contactor/distributor at the higher ozone feed rate (i.e., 60 sccm) and reactor residence time (\(\tau\) = 6 min).
+
+Figure 6: Plots of DEET conversion and TOC reduction in a membrane reactor using coated and uncoated PVDF membrane contactor/distributor as function of reactor residence time for ozone
+
+feed flow rates of (a) & (b) 10 sccm, (c) & (d) 20 sccm, and (e) & (f) 60 sccm. Please note symbols represent experimental data, and the lines were drawn to guide the eyes (200 ppm DEET, 120 ppm O3/O2).
+
+### Catalytic ozonation of DEET in membrane reactor with a membrane water separator
+
+Experiments had shown that catalytic membrane contactor/distributor increases both DEET conversion and mineralization compared to a membrane reactor using an inert membrane contactor/distributor. Figure 7a shows that selective water separation using a PVDF ultrafiltration membrane can further increase DEET conversion from 30 to 38% (uncoated), 30 to 39% (PAC), 45 to 55% (CAT2) and 48 to 60% (CAT1). Selective water removal has the effect of concentrating DEET in the retentate resulting in higher conversion rate. It has similar effects on the reaction byproducts and consequently resulting in deeper mineralization as shown in Fig. 7b. The coated PVDF membranes were more stable (Figure S-2) under ozone as coating layer protects the membrane from wetting. The original PVDF microfiltration membrane was hydrophilic resulting in greater water penetration into the membrane pores [44] and generation of more reactive hydroxyl radicals that attacks the membrane resulting in the unstable performance. The CAT1 catalyst coated PVDF membrane contactor/distributor gave the highest performance reaching 60% DEET conversion at a short residence time of 4 min and a sustained TOC reduction of 30%.
+
+## 4 Conclusion
+
+This work successfully demonstrated the use of polymeric membranes as contactor, distributor and separator in a membrane reactor for ozone degradation of refractory organic pollutant, DEET in water. The use of PAC to capture and retain the organic pollutants in the reaction zone of the membrane reactor has the desired effect of increasing conversion and mineralization of DEET. Adding ferrihydrites known for its ability to promote the dissolution and transformation of ozone into active hydroxyl radicals led to deeper mineralization. Preparing ferrihydrite supported on activated carbon catalyst required care in order to obtain highly dispersed catalyst. A catalytic membrane contactor/distributor using CAT1 as catalytic layer gave the best DEET conversion and mineralization. Longer residence time and higher ozone dosage has the desired effects of deeper
+
+Figure 7: (a) Overall DEET conversion and (b) overall TOC reduction for the coated and uncoated PVDF membrane contactor/distributor with and without simultaneous water removal across a PVDF ultrafiltration membrane.
+
+oxidation of DEET. Finally, the use of a PVDF ultrafiltration membrane for water separation can further enhance the DEET conversion and mineralization rates by further concentrating the organic pollutants in the retentate. This showed that it is possible to implement a compact membrane reactor for efficient treatment of recalcitrant and refractory organic pollutants in water.
+
+## Acknowledgement
+
+The authors gratefully acknowledge the financial support from the Hong Kong Research Grant Council (605312), National Natural Science Foundation of China General Program (21376196) and Shenzhen Basic Research Program (JCYL 20150630094001158).
+
+## References
+
+* (1) M.J. Benotti, R.A. Trenholm, B.J. Vanderford, J.C. Holady, B.D. Stanford, S.A. Snyder, Pharmaceuticals and Endocrine Disrupting Compounds in U. S. Drinking Water, Environ. Sci. Technol. 43 (2009) 597-603.
+* A review from global views, Microchem. J. 110 (2013) 292-300.
+* (3) K.C. Makris, S.A. Snyder, Screening of pharmaceuticals and endocrine disrupting compounds in water supplies of Cyprus, Water Sci. Technol. 62.(2010) 2720-2728.
+* (4) N. Salgueiro-gonzalez, I. Turnes-carou, V. Besada, S. Muniategui-lorenzo, Science of the Total Environment Occurrence, distribution and bioaccumulation of endocrine disrupting compounds in water, sediment and biota samples from a European river basin, Sci. Total Environ. 529 (2015) 121-130.
+* (5) E.G.B. Xu, S. Liu, G. Ying, G.J.S. Zheng, J.H.W. Lee, K.M.Y. Leung, The occurrence and ecological risks of endocrine disrupting chemicals in sewage effluents from three different sewage treatment plants, and in natural seawater from a marine reserve of Hong Kong, Mar. Pollut. Bull. 85 (2014) 352-362.
+* (6) N. Bolong, A. F. Ismail, M.R. Salim, T. Matsuura, A review of the effects of emerging contaminants in wastewater and options for their removal, Desalination. 238 (2009) 229-246.
+* (7) A. Kumar, I. Xagoraraki, Science of the Total Environment Pharmaceuticals, personal care products and endocrine-disrupting chemicals in U. S. surface and finished drinking waters: A proposed ranking system, Sci. Total Environ. 408 (2010) 5972-5989.
+
+* [8] P. Westerhoff, Y. Yoon, S. Snyder, E. Wert, Fate of Endocrine-Disruptor, Pharmaceutical, and Personal Care Product Chemicals during Simulated Drinking Water Treatment Processes, Environ. Sci. Technol. 39 (2005) 6649-6663.
+* [9] C. Jung, A. Son, N. Her, K. Zoh, J. Cho, Y. Yoon, Journal of Industrial and Engineering Chemistry Removal of endocrine disrupting compounds, pharmaceuticals, and personal care products in water using carbon nanotubes: A review, J. Ind. Eng. Chem. 27 (2015) 1-11.
+* [10] T. Basile, A. Petrella, M. Petrella, G. Boghetich, V. Petruzzelli, S. Colasuonno, D. Petruzzelli, Review of Endocrine-Disrupting-Compound Removal Technologies in Water and Wastewater Treatment Plants: An EU Perspective, Ind. Eng. Chem. Res. 50 (2011) 8389-8401.
+* [11] A. Garcia-rodriguez, V. Matamoros, The ability of biologically based wastewater treatment systems to remove emerging organic contaminants -- a review, Environ. Sci. Pollut. Res. 21 (2014) 11708-11728.
+* [12] M. Clara, B. Strenn, O. Gans, E. Martinez, N. Kreuzinger, H. Kroiss, Removal of selected pharmaceuticals, fragrances and endocrine disrupting compounds in a membrane bioreactor and conventional wastewater treatment plants, Water Res. 39 (2005) 4797-4807.
+* [13] K. Choo, R. Tao, M. Kim, Use of a photocatalytic membrane reactor for the removal of natural organic matter in water: Effect of photoinduced desorption and ferrihydrite adsorption, J. Memb. Sci. 322 (2008) 368-374.
+* [14] R. Molinari, L. Palmisano, E. Drioli, M. Schiavello, Studies on various reactor configurations for coupling photocatalysis and membrane processes in water purification, J. Memb. Sci. 206 (2002) 399-415.
+
+* [15] S.O. Ganiyu, E.D. van Hullebusch, M. Cretin, G. Esposito, M.A. Oturan, Coupling of membrane filtration and advanced oxidation processes for removal of pharmaceutical residues: A critical review, Sep. Purif. Technol. 156 (2015) 891-914.
+* [16] B.S. Karnik, S.H. Davies, M.J. Baumann, S.J. Masten, Fabrication of catalytic membranes for the treatment of drinking water using combined ozonation and ultrafiltration, Environ. Sci. Technol. 39 (2005) 7656-7661.
+* [17] W.K. Chan, J. Jouet, S. Heng, K. L. Yeung, J.C. Schrotter, Membrane contactor/separator for an advanced ozone membrane reactor for treatment of recalcitrant organic pollutants in water, J. Solid State Chem. 189 (2012) 96-100.
+* [18] H.L. Ho, W.K. Chan, A. Blondy, K.L. Yeung, J.C. Schroter, Experiment and modeling of advanced ozone membrane reactor for treatment of organic endocrine disrupting pollutants in water, Catal. Today. 193 (2012) 120-127.
+* [19] S. Heng, K.L. Yeung, M. Djafer, J.C. Schroter, A novel membrane reactor for ozone water treatment, J. Memb. Sci. 289 (2007) 67-75.
+* [20] S. Heng, K.L. Yeung, A. Julbe, A. Ayral, J.C. Schroter, Preparation of composite zeolite membrane separator / contactor for ozone water treatment, Micropor. Mesopor. Mat. 115 (2008) 137-146.
+* [21] Z. Yao, Y. Li, Y. Cui, K. Zheng, B. Zhu, H. Xu, L. Zhu, Tertiary amine block copolymer containing ultra filtration membrane with pH-dependent macromolecule sieving and Cr (VI) removal properties, Desalination. 355 (2015) 91-98.
+* [22] S. Rajabzadeh, S. Yoshimoto, M. Teramoto, M. Al-marzouqi, H. Matsuyama, CO 2 absorption by using PVDF hollow fiber membrane contactors with various membrane structures, Sep. Purif. Technol. 69 (2009) 210-220.
+
+* [23] J. Zuo, T. Chung, PVDF hollow fibers with novel sandwich structure and superior wetting resistance for vacuum membrane distillation, Desalination. 417 (2017) 94-101.
+* [24] S. Bampering, T. Suwannachart, S. Atchariyawut, R. Jiraratananon, Ozonation of dye wastewater by membrane contactor using PVDF and PTFE membranes, Sep. Purif. Technol. 72 (2010) 186-193.
+* [25] P. Wang, T. Chung, A New-Generation Asymmetric Multi-Bore Hollow Fiber Membrane for Sustainable Water Production via Vacuum Membrane Distillation; Environ. Sci. Technol. 47 (2013) 6272-6278.
+* [26] I.F.J. Vankelecom, Polymeric Membranes in Catalytic Reactors, Chem. Rev. 102 (2002) 3779-3810.
+* [27] L. Francis, N. Ghaffour, A.S. Alsaadi, S.P. Nunes, G.L. Amy, PVDF hollow fiber and nanofiber membranes for fresh water reclamation using membrane distillation, J. Mater. Sci. 49 (2014) 2045-2053.
+* [28] D.S. Pines, K. Min, S.J. Ergas, D.A. Reckhow, Investigation of an Ozone Membrane Contactor System, Ozone Sci. Eng. 27 (2005) 209-217.
+* [29] S. Rajabzadeh, S. Yoshimoto, M. Teramoto, M. Al-marzouqi, Y. Ohmukai, T. Maruyama, H. Matsuyama, Effect of membrane structure on gas absorption performance and long-term stability of membrane contactors, Sep. Purif. Technol. 108 (2013) 65-73.
+* [30] K. Quednow, W. Puttmann, Temporal concentration changes of DEET, TCEP, terbutryn, and nonylphenols in freshwater streams of Hesse, Germany: possible influence of mandatory regulations and voluntary environmental agreements, Environ. Sci. Pollut. Res. 16 (2009) 630-640.
+* [31] L. Diskowitzky, E. Heruwati, F. Ariyani, H.E. Irianto, J. Schwarzbauer, Exceptionally high concentrations of the insect repellent N, N -diethyl- m -toluamide ( DEET ) in surface waters from Jakarta, Indonesia, Environ. Chem. Lett. 12 (2014) 407-411.
+* [32] M. Kitis, S.S. Kaplan, Advanced oxidation of natural organic matter using hydrogen peroxide and iron-coated pumice particles, Chemosphere. 68 (2007) 1846-1853.
+* [33] D.L. Sudakin, W.R. Trevathan, DEET: A Review and Update of Safety and Risk in the General Population, J. Toxicol. 41 (2003) 831-839.
+* [34] H.C. Wang, H.S. Liang, M.B. Chang, Chlorobenzene oxidation using ozone over iron oxide and manganese oxide catalysts, J. Hazard. Mater. 186 (2011) 1781-1787.
+* [35] M. Goncalves, M.C. Guerreiro, L.C.A. de Oliveira, C.S. De Castro, A friendly environmental material: Iron oxide dispersed over activated carbon from coffee husk for organic pollutants removal, J. Environ. Manage. 127.(2013) 206-211.
+* [36] H.S. Park, J.R. Koduru, K.H. Choo, B. Lee, Activated carbons impregnated with iron oxide nanoparticles for enhanced removal of bisphenol A and natural organic matter., J. Hazard. Mater. 286 (2015) 315-24.
+* [37] M. Gui, V. Smuleac, L.E. Ormsbee, D.L. Sedlak, D. Bhattacharyya, Iron oxide nanoparticle synthesis in aqueous and membrane systems for oxidative degradation of trichloroethylene from water, J. Nanoparticle Res. 14 (2012) 861-876.
+* [38] P. Leone, M. Gennari, Role of Ferrihydrite in Adsorption of Three Imidazolinone Herbicides, J. Agric. Food Chem. 49 (2001) 1315-1320.
+* [39] H. Park, Y. Kim, B. An, H. Choi, Characterization of natural organic matter treated by iron oxide nanoparticle incorporated ceramic membrane-ozonation process, Water Res. 46 (2012) 5861-5870.
+* [40] M. Kitis, S.S. Kaplan, E. Karakaya, N.O. Yigit, G. Civelekoglu, Adsorption of natural organic matter from waters by iron coated pumice, Chemosphere. 66 (2007) 130-138.
+* [41] K.S. Tay, N.A. Rahman, M. Radzi, B. Abas, Degradation of DEET by ozonation in aqueous solution, Chemosphere. 76 (2009) 1296-1302.
+* [42] F.J. Rivas, F.J. Beltra, Iron type catalysts for the ozonation of oxalic acid in water, Water Res. 39 (2005) 3553-3564.
+* [43] W. Ling, Z. Qiang, Y. Shi, T. Zhang, B. Dong, Journal of Molecular Catalysis A : Chemical Fe ( III ) -loaded activated carbon as catalyst to improve omethoate degradation by ozone in water, J. Mol. Catal. A. Chem. 342-343 (2011) 23-29.
+* [44] W. Rongwong, C. Fan, Z. Liang, Z. Rui, R.O. Idem, Investigation of the effects of operating parameters on the local mass transfer coefficient and membrane wetting in a membrane gas absorption process, J. Memb. Sci. 490 (2015) 236-246.

manual_annotation/200pdfs_in_mmd/10.1016_j.cattod.2020.01.042.mmd

@@ -0,0 +1,153 @@
+Catalytic ozonation of naphthalene acids in the presence of carbon-based metal-free catalysts: Performance and kinetic study
+
+[
+
+[
+
+[
+
+[
+
+[
+
+[
+
+[
+
+###### Abstract
+
+This study examined the use of carbon-based metal-free materials as catalysts for the catalytic ozonation of a model napthemic acid (NA) compound (1,3-Adamantanedicarboxylic acid; ADA). Adsorption, single ozonation and catalytic ozonation experiments were performed using ADA concentration of 50 mg/L, 0.5 g/L of carbon materials, and 30 mg/L of applied ozone (O3) dose, at pH 8. The results showed that, after 15 min of reaction, less than 5 % of ADA was removed by adsorption, while 33 % of ADA was removed by single ozonation. For the catalytic ozonation study, among the materials tested, the best performance was obtained using carbon xerogels prepared at pH 5.5 (C5.5), reaching 65 % of ADA removal after 15 min of reaction. The first-order rate constant of O3 + C5.5 (0.0631 min-1) was almost two times higher than that of single ozonation (0.0258 min-1). Hydroxyl radicals were the main responsible species for the ADA degradation. In general, the results confirmed that catalytic ozonation process is an efficient and promising approach for environmental applications.
+
+[O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozonation, [O3]C. Alberts, Ozon,promoter of the radical reaction is still not well understood. Similarly, Orge et al. [31] studied the ozonation of selected organic pollutants (oxalic acid and the textile dye CI Reactive Blue 5) in the presence of carbon xerogels with different properties. The results of this study showed that the pore sizes of CXs played a key role in the degradation of the selected pollutants.
+
+To the best of our knowledge, there is no information available in the published literature on CX as a catalyst to enhance the oxidation of NAs by O2. Therefore, the objectives of this study were: (i) to prepare and characterize CX materials with different textural properties; (ii) to investigate the influence of textural properties of the CX and granular activated carbon (GAC) on the catalytic decomposition of O3 and degradation of a model NA compound; (iii) to study the reaction mechanism by using tert-butyl alcohol (TBA) as a "OH scavenger; and (iv) to assess the reusability of the carbon materials.
+
+## Materials and methods
+
+### Materials and reagents
+
+Dicyclohexylacetic acid (DCHA), trans-4- pentylcyclohexane carboxylic acid (TPCA), and 1,3-Adamantanedicarboxylic acid (ADA), optima water, and acetonitrile (HPLC grade) were purchased from Sigma-Aldrich (Canada). Resorcinol (99 %), formaldehyde (37 wt.% in water, stabilized by 10 - 15 wt.% methanol), hydrochloric acid (> 37 %), sodium hydroxide (97 %), sulfuric acid (98 %), sodium tetraborate, and acetic acid were purchased from Fisher Scientific Co. (Canada). Ammonium acetate was purchased from Fluka Analytical. GAC was purchased from Calgon Carbon* (Pennsylvania, USA). TBA and sodium thiosulfate were obtained from Fisher Scientific Co. (Canada). Ultra-dry oxygen for O3 generation was obtained from Praxair (Canada). All chemicals were used as received without further purification. Milli-Q ultrapure water (Millipore Corporation) was used throughout the work. Borate buffer solution (pH 8) was prepared by mixing 0.05 M sodium tetraborate and sulfuric acid at a ratio of 1.25:1 in order to mimic the pH of OSPW.
+
+### Synthesis of carbon xerogels
+
+Carbon xerogels were prepared at pH 5.5 and 6.9 by the conventional sol-gel approach, using formaldehyde and resorcinol in a molar ratio of 2:1 [24,32,33]. Two pHs were selected because the polymerization occurs only in a small pH window (1-2 pH units). If the pH is too low, precipitation occurs, while if the pH is too high, condensation is hindered. This dramatic effect of the initial pH on the pore structure is discussed elsewhere [34]. Detailed procedure is provided in the Support Information (SI).
+
+### Textural and surface characterization
+
+The textural characterization of the CX and GAC was based on the nitrogen adsorption isotherms, determined at -196 degC with an Autosorb Quantumchrome 1 M pH instrument, USA. Surface characterization of CX and GAC was performed using scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDX), X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared (FT-IR). The detailed methods of textural and surface characterization have been provided in the SI.
+
+### Adsorption experiments
+
+Three model NA compounds with different structures, including DCHA, TPCA, and ADA were used for the preliminary adsorption study. A volume of 500 mL of solution was used, consisting of a 50 mg/L solution of model NA at pH 8 borate buffer (50 mM). The carbon material adsorbents (0.25 g, particle size between 0.6 and 1.4 mm) were introduced into the solution by stirring at 200 rpm to start the adsorption experiments. Each experiment was conducted in replicates. Blank tests (in the absence of carbon materials) were also carried out.
+
+### Single and catalytic ozonation experiments
+
+Single and catalytic ozonation experiments with a model compound (ADA) were performed in a 1 L headspace-free batch reactor. First, stock O3 solution was prepared by placing borate buffer solution (pH 8) in a reactor and saturating with O3 then to begin the reaction, ADA was spiked to have a final concentration of 50 mg/L and the catalyst (0.5 g/L) was added; this was considered time zero for the reaction. In addition, single ozonation (no catalyst) and adsorption experiments (no ozone) were carried out under the same conditions. At predetermined times, 2 mL of the reaction mixture was withdrawn by syringe, quenched with Na2S2O3 and filtered through nylon syringe 0.2 mm filters for analysis.
+
+Catalyst reutilization tests were carried out with the three carbon samples, in order to evaluate the stability of the catalyst in catalytic ozonation processes. For mechanism study, both single and catalytic ozonation experiments were performed in the presence of the radical scavenger TBA with concentration ranging from 2 to 100 mM.
+
+### Analytical methods
+
+A direct UV spectrophotometry method was used to determine the O3 concentration from the absorbance of the solution at 260 nm [35]. Model NA samples were analyzed using a liquid chromatogram mass spectrometer (LC-MS). The detailed methods have been provided in the SI.
+
+## Results and discussion
+
+### Characterization of carbon samples
+
+The carbon materials used in this work are described in Table 1, together with their relevant textural properties. The characterization results showed that commercial GAC was a microporous material, with the highest BET surface area (S\({}_{\text{BET}}\) = 976 m2/g), almost 1.5 times higher than CX5.5 (S\({}_{\text{BET}}\) = 573 m2/g) and 2.5 times higher than CX6.9 (S\({}_{\text{BET}}\) = 391 m2/g). GAC also showed a large volume of microporous (V\({}_{\text{micro}}\) = 0.386 cm2/g) and microporous surface area (S\({}_{\text{micro}}\) = 839 m2/g). However, the total pore volume of CX5.5 (V\({}_{\text{total}}\) = 1.545 cm2/g) was 2.5 times higher than that of GAC (V\({}_{\text{total}}\) = 0.599 cm2/g), from which 87 % was mesoporous volume (V\({}_{\text{mean}}\) = 1.340 cm2/g) and the remaining 13 % was microporve volume (V\({}_{\text{micro}}\) = 0.205 cm2/g). A large average pore diameter (D\({}_{\text{p}}\) = 11 nm) was found for CX5.5 and 3 nm for
+
+\begin{table}
+\begin{tabular}{l c c c c c c c} Sample & S\({}_{\text{BET}}\) & S\({}_{\text{micro}}\) & V\({}_{\text{micro}}\) & V\({}_{\text{micro}}\) & V\({}_{\text{macro}}\) & V\({}_{\text{macro}}\) & V\({}_{\text{macro}}\) & D\({}_{\text{sp}}\) & Particle size \\  & (m2/g) & (m2/g) & (cm2/g) & (cm2/g) & (cm2/g) & (cm2/g) & (mm) & (mm) \\ \hline CX5.5 & 573 & 438 & 0.205 & 1.340 (87 & 1.545 & 11 & 0.6-1.4 \\  & & & & & & & \\  & & & & & & & \\ C6.9 & 391 & 256 & 0.116 & 0.157 (58 & 0.273 & 3 & 0.6-1.4 \\  & & & & & & & \\  & & & & & & & \\  & & & & & & & \\  & & & & & & & \\  & & & & & & & \\ \end{tabular}
+\end{table}
+Table 1: Textural properties of the carbon samples tested in the adsorption and catalytic donation of model NAs.
+
+CX6.9, concluding that carbon xerogels were mesoporous materials. The carbon xerogel result revealed that the pH of the initial solution clearly played a definitive role in the development of porosity [36]: a slight pH decrease led to an almost six-fold higher pore volume and about 32 % higher surface area. The main difference between CX5.5 and CX6.9 was the total pore volume and the average pore diameter. Mesooper volume values reflect that these are the pores preferentially formed when using a lower solution pH.
+
+In order to obtain the surface morphology and elemental composition of carbon samples, SEM and SEM-EDX analyses were performed. The results (Figure S1 (a-c) and Table S2) show that CXs were mainly composed of carbon (93 wt.%) and oxygen (7 wt.%) with no trace elements. On the other hand, GAC was mainly composed of carbon (90 wt.%) and oxygen (7 wt.%) with trace elements, including aluminium (0.85 wt.%), silicon (0.70 wt.%), sulphur (0.56 wt.%), and iron (0.75 wt.%). Similar results were obtained by Alvarez et al. [37], demonstrating that CX is mainly carbonaceous with a negligible amount of trace elements. Fig. 1(a-c) shows the XPS deconvolution peaks for the C1s for GAC, CX5.5 and CX6.9 samples, respectively. The dominant peak in the C1s region was found at 284.5 eV for all carbon materials as it represents graphite carbon (C-C). A second peak was observed at 285.6 eV for all materials, which indicates C-O bonds [38]. The FT-IR spectroscopy analysis of the materials is shown in Fig. 1d. The surface properties within the range 1300 - 1000 cm\({}^{-1}\) can be assigned to various C-O bonds, such as ethers, phenols and hydroxyl groups. The FT-IR spectrum of the materials with the range 2000 to 1500 cm\({}^{-1}\)was related to aldehyde, ketones, acids, aromatic ring carbon, and ester. The stretching vibration of the C=C bond usually gives rise to a moderate band in the region 2300-2400 cm\({}^{-1}\)[39]. From XPS and FT-IR analyses, we observed that even though the precursors and synthesis conditions were different, the surface properties (mainly surface functional groups) were similar among the GAC, CX5.5 and CX6.9.
+ozonation experiments with the model compound ADA. Fig. 3a depicted the O\({}_{3}\) decomposition as a function of time in the presence and absence of carbon materials. The result revealed that the presence of carbon materials in general accelerated the decomposition of O\({}_{3}\) (Fig. 3a). For instance, after 10 min of reaction, O\({}_{3}\) decomposition was 48 %, 82 %, 79 % and 68 % for O\({}_{3}\) only, O\({}_{3}\) + GAC, O\({}_{3}\) + CX5.5, and O\({}_{3}\) + CX6.9, respectively. Huge increments in the O\({}_{3}\) decomposition were observed for GAC and CX5.5; this could be related to their surface area and pore volume, respectively as observed elsewhere [41, 42].
+
+First-order kinetic expression was used to determine the O\({}_{3}\) decomposition rate. The first-order rate constants of O\({}_{3}\)/CX5.5 (0.1367 min\({}^{-1}\)) and O\({}_{3}\)/GAC (0.1325 min\({}^{-1}\)) were almost 2.4 times higher than that of single ozonation (0.0567 min\({}^{-1}\)) (Fig. 3b). Therefore, these results confirmed that the presence of carbon materials enhanced the transformation of O\({}_{3}\) into secondary radicals such as 'OH. Similar studies [26, 27, 43] have reported that activated carbon can accelerate O\({}_{3}\) decomposition, resulting in the formation of 'OH. Possible reaction mechanism explaining the results was proposed by Alvarez et al. [44] as follows: (i) a fraction of dissolved O\({}_{3}\) could be transferred from the liquid bulk to the surface (external and internal) of carbon particles; (ii) adsorption of O\({}_{3}\) on the carbon pore sites; and (iii) reaction of adsorbed and/or non-adsorbed O\({}_{3}\) with some surface functionalities of carbon materials, mainly direct transformation of O\({}_{3}\) into active radicals.
+
+For comparison purpose, ADA adsorption kinetic experiment was performed (Fig. 4a). The result indicated that ADA was difficult to be adsorbed on the carbon materials and could, therefore, be neglected. However, according to Fig. 4b, the addition of O\({}_{3}\) alone or O\({}_{3}\) with small amount of carbon catalyst resulted in a major increase in the degradation of ADA. To prove that catalytic ozonation occurred, a control adsorption experiment of CX5.5 treated with O\({}_{3}\) and then added to ADA, compared to the untreated CX5.5 adsorption experiment was conducted. The adsorption results and the surface chemistry characterization results (not shown), showed insignificant change in terms of efficiency and surface chemistry. Thus, it can be considered that the removal of ADA in the presence of O\({}_{3}\) and carbon materials mainly occurred by catalytic ozonation.
+
+In general, the results showed that, after 15 min of reaction, less than 5 % of ADA was removed by adsorption for all carbon materials, while 33 % of ADA was removed by single ozonation. For the catalytic ozonation study, after 15 min of reaction, 42 %, 60 % and 65 % of ADA were removed using CX6.9, GAC, and CX5.5 as catalyst, respectively. All materials acted effectively as ozonation catalysts, but among the catalyst materials tested, the best performance was obtained by using CX5.5, reaching 97 % of ADA removal after 60 min of reaction. This observation could be explained by the large mesopore or total pore volume and pore diameter present in the CX5.5 that could control the kinetics of the process by facilitating the diffusion of O\({}_{3}\) or ADA through the pore network. Even though the higher surface area also played an important role for the GAC, its microporosity nature minimized the efficiency in terms of ADA removal. Similarly, Sanchez-Polo et al. [42] concluded that the O\({}_{3}\) decomposition on GAC was greatly affected by the volume of mesopores as the process might be controlled by pore diffusion.
+
+The experimental data were analyzed considering first-order kinetic model and fitted significantly well (Fig. 4c). According to the first-order kinetic model, the first-order rate constant of O\({}_{3}\) + CX5.5 (0.0631 min\({}^{-1}\)) was almost 2.5 times higher and O\({}_{3}\) + GAC (0.0499) was 2 times higher than that of single ozonation (0.0258 min\({}^{-1}\)). The results led to conclude that the presence of carbon accelerated the
+
+Figure 3: (a) Decomposition of ozone in water and (b)first-order kinetic in the presence and absence of carbon materials differing in their textural characteristics (0.5 g/L carbon materials, 30 mg/L of applied O\({}_{3}\) dose, and pH 8 in borate buffer).
+
+Figure 2: Removal of NAS during adsorption. Initial NA concentration of 50 mg/l, 0.5 g/L carbon materials, pH 8 in borate buffer, adsorption time of 3 and 24 h.
+
+transformation of O3 into surface radical species and/or oxygen-containing radicals in solution, such as 'OH [45].
+
+### Reaction mechanism study
+
+The most likely reaction mechanism from Eqs. (1) through (9) are demonstrated as below. Generally, O3 decomposition in aqueous solution is initiated by the presence of hydroxide (HO-) ions (Eq. 1); in this process pH plays a major role [46]. Moreover, in the presence of carbon materials, O3 decomposition could be accelerated (Eq. (2)). According to literature, the two possible pathways of generating free radicals are: (i) O3 molecules first adsorb and react on the surface of carbon, yielding surface free radicals (Eq. (3)); and (ii) carbon material acts as an initiator of the decomposition of O3, yielding free radicals; such as 'OH in solution (Eq. 2) [46]. In our study, ADA degradation possibly occurred on the surface of the carbon, between the surface radicals and adsorbed reagent (ADA), according to Eqs. (4) and (5). Moreover, adsorbed species might also interact with O3 (Eq. (6) or 'OH from the bulk (Eq. (7)). Similar assumption have been presented by Beltran et al. [47] and Faria et al. [46], regarding the ozonation of purvice acid and oxalic acid in the presence of activated carbon, respectively. In addition, the degradation of ADA in the bulk occurred indirectly via 'OH (Eq. (8)) and directly via molecular O3 (Eq. 9). However, the rate constant of the reaction between NAs and molecular O3 was very low compared to 'OH, therefore, the 'OH was considered to be the main reaction mechanism.
+
+O3 (in the presence of OH-)- OH (1) O3 + C ~OH (2) O3 + C ~C ~C (3) R + C ~C ~R (4) C-R + C ~O ~P (5) C-R + O3 ~P C-R + O3 ~P C-R + O3 ~P Where C = Carbon; R = Reagent (ADA); C ~O = Any oxygen-containing active species on the surface (including hydroxyl radical); C-R = Adsorbed reagent; P = Products
+
+To confirm the proposed mechanism, the oxidative species (O3, surface free radicals or 'OH radicals) responsible for the ADA degradation process were identified by using a powerful 'OH scavenger (TBA). TBA was used to inhibit the degradation of ADA by scavenging 'OH in bulk solution and on the surface of the catalyst because of the minimal adsorption of TBA on the carbon material [26]. Fig. 5 shows that the ADA removal was inhibited by the presence of TBA in the entire non-catalytic (Fig. 5a) and catalytic ozonation process (Fig. 5b-d). This result indicates that 'OH oxidation in bulk solution was involved in all the processes. ADA removal generally decreased with increasing TBA concentration from zero to 100 mM. In general, the result shows that the TBA quenched all the hydroxyl in the bulk at any concentration. Due to the lower affinity of TBA to be adsorbed on the carbon surface, there was no chance to be close to the surface to quench those radicals on the surface, attacking the C-R. However, at higher concentration of TBA, some of the 'OH might be quenched. Therefore, according to the experimental results obtained in this work, the degradation of ADA most likely occurred through Eq. 7 and Eq. 8. This mechanism is very complex and still not understood very well, however, similar result was obtained by Faria et al. [46]. On the other hand, Guzman-Perez et al. [48] investigated the oxidation of atrazine by ozone mixed with AC and found that the reaction mainly occurred in the bulk liquid phase due to
+
+Fig. 4: (a) ADA adsorption experiments, (b) ozonation experiments, and (c) first-order kinetic (50 mg/L of ADA, pH 8 in borate buffer, 30 mg/L of applied O3 dose, and catalyst concentration = 0.5 g/L).
+
+the inhibition effect of TBA.
+
+### Relation of the textural properties of carbon materials with the rate constant
+
+Figure S4 shows the rate constant of catalytic ozonation of ADA normalized by the BET surface area of the materials. It can be seen that CX5.5 had a better performance, since it reacted fast per unit of area. Even though GAC showed a higher surface area compared with carbon xerogel, the rate constant per surface area revealed that GAC performance was two times lower than that of CX. These results suggested that the mesoporous material, having high total pore volume, may positively affect the reaction by providing wider and available pores for better diffusion of O3 and ADA. Similarly, with porous material, O3 can easily access or interact with the available sites in order to be decomposed and generate 'OH radical. Moreover, the analysis of the results showed that there was a correlation between the rate constant of ADA by catalytic ozonation and the total pore volume of the tested carbon materials.
+
+### Reutliination of the catalyst
+
+Reusability tests on O3/CX and O3/GAC system were conducted to assess the possible deactivation of the catalyst during catalytic ozonation. The CX5.5, CX6.9 and GAC were reused for three times consecutively in the presence of O3 and under the same experimental conditions. Figure S5 shows that 93 %, 92 %, and 89 % removal of ADA was obtained for GAC after 60 min of reaction in three successive runs, respectively. Similarly, 97 %, 96 % and 94 % of ADA removal for CX5.5 and 89 %, 85 % and 80 % of ADA removal for CX6.9 were obtained after 60 min of reaction in three successive runs, respectively. Slight differences were observed in terms of removal efficiency for the three consecutive runs of all the materials. We also determined the recovery percentage of the solid material in each utilization test and it was consistently above 98 %. The 2 % might be loss of material during the recovery steps (washing and weighing the sample). This demonstrated that no huge deactivation occurred on the carbon catalyst by the oxidative conditions in the reactor.
+
+## 4 Conclusions
+
+This study investigated the preparation, characterization and the use of carbon materials (CX and GAC) as catalysts for the catalytic ozonation. In this work, we also compared the degradation of a model NA compound (ADA) using adsorption, ozonation and catalytic ozonation process. The following conclusions could be drawn:
+
+1. CX was synthesized by sol-gel method from resorcinol and formaldehyde at two pHs (i.e., pH 5.5 and 6.9). The characterization results showed that CX prepared at pH 5.5 (CX5.5) was a mesoporous material with large surface area (573 m\({}^{2}\)/g) and high pore volume (1.55 cm\({}^{3}\)/g), which was mainly composed of carbon (93.20 %) and oxygen (6.71 %). The pH of the initial solution clearly played a great role in the development of porosity: a slight
+
+Figure 5: ADA degradation experiments in the presence of TBA for (a) ozone, (b) ozone + GAC (c) ozone + CX5.5, and (d) ozone + CX6.9 (50 mg/l. of ADA, pH 8 in buffer, 30 mg/l. of applied O3 dose, 0 −100 mM of TBA, and catalyst concentration = 0.5 g/l).
+
+PH decrease led to an almost five-fold higher pore volume and about 30 % higher surface area.
+2. The first-order rate constants of O3/CX5.5 (0.1367 min\({}^{-1}\)) and O3/GAC (0.1325 min\({}^{-1}\)) were almost 2.4 times higher than that of single oxonation (0.0567 min\({}^{-1}\)). The main responsible species for the NA degradation was 'OH present in the bulk solution. In general, the presence of carbon materials with mild O3 dose accelerated the O3 decomposition rate and model NA degradation.
+3. First-order rate constant of ADA could be directly related to some textural properties of CX (i.e., surface area and pore volume). Although GAC showed a higher surface area compared with CX, the rate constant of per surface area revealed that GAC performance was two times lower than that of CX. The mesoporous material with high total pore volume, positively affected the reaction by providing wider and available pores for better diffusion of O3 and ADA.
+
+## CRediT authorship contribution statement
+
+**Selamawit Ashagre Messele:** Conceptualization, Methodology, Writing - original draft. **Pamela Chelme-Ayala:** Writing - review & editing. **Mohamed Gamal El-Din:** Supervision.
+
+###### Acknowledgements.
+
+ This work was funded by Natural Sciences and Engineering Research Council of Canada (NSERC) Senior Industrial Research Chair (IRC) in Oil Sands Traillies Water Treatment through the support by Syncrude Canada Ltd., Suncor Energy Inc., Canadian Natural Resources Ltd., Imperial Oil Resources, Tech Resources Limited, EPCOR Water Services, Alberta Innovates, and Alberta Environment and Parks. As a part of the University of Alberta's Future Energy Systems research initiative, this research was made possible in part thanks to funding from the Canada First Research Excellence Fund.
+
+## Appendix A Supplementary data
+
+Supplementary material related to this article can be found, in the online version, at [https://doi.org/10.1016/j.cattod.2020.01.042](https://doi.org/10.1016/j.cattod.2020.01.042).
+
+## References
+
+* (1) D. Lyder, J. Feng, R. Rivard, A. Gallie, E. Clouis, Fuel 89 (2010) 760-767.
+* (2) C. Hsu, J.V. Hendley, K.M. Pew, R. Frank, P. Yang, K.R. Solomon, J. Chromatogr. A 1067 (2005) 277-284.
+* (3) V. W. Rogers, M. Wickstrom, K. Liber, M.D. MacKinnon, Toxicol. Sci. 66 (2002) 347-355.
+* (4) J.M. Masiriyah, Z. Zhou, Z. Xu, J. Czarnecki, H. Hamza, Can. J. Chem. Eng. 82 (2004) 628-654.
+* (5) E. W. Allen, J. Environ. Eng. Sci. 7 (2008) 123-138.
+* (6) Government of Alberta, Alberta Energy: Oil Sands, Available at: [http://www.energy.thesis.com/S0719/aS](http://www.energy.thesis.com/S0719/aS) (Accessed December 2013 (2011).
+* (7) C. Li, L. Fu, J. Stafford, M. Boelovic, M. Gamal El-Din, Sci. Total Environment. 601-602 (2017) 1755-1802.
+* (8) A. Mahaffey, M. Dahe, Environ. Rev. 25 (2017) 97-114.
+* (9) R. Huang, Y. Chen, M. Gamal El-Din, Environ. Sci. Technol. 50 (2016) 6433-6441.
+* (10) R. Huang, Y. Chen, M.N.A. Mehner, P. Chelme-Ayala, S. Dong, M.D. Ibrahim, C. Wang, N. Klameth, S.A. Hughes, J.V. Hoslingky, K.M. Pem, C. Brown, A. Mahaffey, M. Gamal El-Din, Water Res. 128 (2018) 129-137.
+* (11) C. Wang, N. Klameth, S.A. Messele, A. Singh, M. Boelovic, M. Gamal El-Din, Water Res. 100 (2016) 476-485.
+* (12) A. Afral, P. Chelme-Ayala, J.W. Martin, M. Gamal El-Din, Ozone Sci. Eng. 37 (2014) 45-54.
+* (13) M.N.A. Mehner, N. Klameth, M.S. Islam, K.N. McPhedran, M. Gamal El-Din, Chemosphere 180 (2017) 149-159.
+* (14) A. Afral, P. Dremezic, A.A. Peres-Estrada, Y. Chen, J.W. Martin, M. Gamal El-Din, Environ. Sci. Technol. 46 (2012) 10727-10734.
+* (15) Z. Fang, R. Huang, P. Chelme-Ayala, Q. Shi, C. Xu, M. Gamal El-Din, Sci. Total Environ. 604 (2019) 133686.
+* (16) M.A. Mehner, P. Chelme-Ayala, M. Gamal El-Din, Water Res. 125 (2017) 62-71.
+* (17) F. Vilasenska, M.R. Pereira, J.M. Grifo, J.L. Figueiredo, J. Colloid Inter. Sci. 293 (2006) 128-136.
+* (18) S.A. Carlabinger-Ammosmisr, M.F.R. Pereira, J.L. Figueiredo, Water Res. 45 (2011) 4583-4591.
+* (19) J.I. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.M. Orfalo, Carbon 37 (1999) 1379-1389.
+* (20) J.I. Figueiredo, M.F.R. Pereira, M.A.A. Freitas, J.M. Orfalo, Ind. Eng. Chem. Panda. 60 (2007) 4110.
+* (21) P. Serp, J.I. Figueiredo (Eds.), Carbon Materials for Catalysis, Wiley, Hoboken, New Jersey, 2009.
+* (22) J. Athar, N.A. Amin, J. Wang, J. Water Supply Res. Technol 61 (2012) 413-426.
+* (23) P.V. Smant, P. Goncalves, M.A. Freitas, M.R. Pereira, J.L. Figueiredo, Carbon 42 (2004) 132-1325.
+* (24) N. Mahata, M.J.R. Pereira, F. Suarez-Garcia, A. Martinez-Alonso, J.M.D. Tascon, J.L. Figueiredo, J. Colloid Interfect. Sci. 324 (2008) 1805-1525.
+* (25) L. Li, W. Zhu, Z. Zhang, Z. Hao, Y. Hu, W. Hu, W.

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2007.11.019.mmd

@@ -0,0 +1,156 @@
+# Decolorization of aqueous textile reactive dye by ozone
+
+Jiangning Wu
+
+j3wu@ryerson.ca
+
+Huu Doan
+
+Simant Upreti
+
+Department of Chemical Engineering, Ryerson University, 350 Victoria Street, Toronto, Ontario, Canada M5B 2K3
+
+Received 23 September 2006; received in revised form 11 November 2007; accepted 20 November 2007
+
+###### Abstract
+
+The aqueous solution of a model textile reactive dye, C.I. Reactive Blue 15, was ozonated in a semi-batch reactor. Results of kinetic study showed that ozonation of the aqueous reactive dye was a pseudo-first-order reaction with respect to the dye. The apparent rate constant increased with both the applied ozone dose and temperature, but declined logarithmically with initial dye concentration. It was also found that the volumetric mass transfer coefficient of ozone increased linearly with initial dye concentration, applied ozone dose and temperature, respectively. The experimental data further indicated that ozonation effectively removed chemical oxygen demand (COD) and enhanced the biodegradability of the aqueous dye solutions.
+
+keywords: Textile reactive dye; Ozone; Color removal; Wastewater treatment; Ozone mass transfer +
+Footnote †: journal: Chemical Engineering Journal
+
+## 1 Introduction
+
+Textile effluents are characterized by heavy color resulting from dyes remaining in the water. Colored wastewater not only affects the aesthetic merit and water transparency of receiving waterbodies [1,2], but there are also environmental concerns about the possible toxicity and carcinogenicity of some organic dyes. In the case of reactive dyes commonly used for cotton dyeing, the quality of the wastewater is further degraded because around 30% of the dyes applied remain in the effluents [3, 4, 5, 6]. Since dyes were intentionally designed to resist degradation, conventional biological wastewater treatment methods are ineffective in removing the color [1,2,5,7, 8, 9].
+
+Ozone is very effective in decolorizing textile effluents [4,6]. Ozone can also convert biorefractory dyes in wastewater into biodegradable species so that effective biological treatment can follow [3,10,11]. It has been reported that the rate-limiting step in the ozonation of dye-containing wastewater is the mass transfer of ozone from the gas-phase to the liquid-phase [12, 13, 14, 15]. However, the driving force for ozone mass transfer, i.e., the difference between the concentration of the dissolved ozone and the equilibrium ozone concentration at the gas-liquid interface, varies with water quality parameters and operating conditions [10,16]. Consequently, the efficacy of ozone treatment of textile wastewater depends on water quality parameters such as the characteristics of the dye, concentration of the dye, etc. It also depends on operating parameters such as the applied ozone dose and temperature.
+
+In this study, the aqueous solution of a model textile reactive dye, C.I. Reactive Blue 15, was ozonated in a semi-batch reactor. Experiments were designed to quantify the effects of several water-quality and operating parameters on the decolorization of this representative system. The extent of biodegradability increased after ozonation for this system was also investigated.
+
+## 2 Experimental
+
+### Materials and methods
+
+C.I. Reactive Blue 15 was purchased from Sigma-Aldrich Canada (Oakville, Ontario). All other chemicals used were reagent grade and obtained from Sigma-Aldrich or BDH Inc. (Toronto, Ontario). Chemical oxygen demand (COD) vials and biological oxygen demand (BOD) bottles were all supplied by VWR Canlab (Mississauga, Ontario).
+
+The color of the aqueous dye solutions were measured by an integration method developed previously [6,10]. This method involved scanning the absorbance of a sample from 400 to 700 nm (Beckman DU 650 spectrophotometer, Beckman Instruments Canada Inc., Mississauga, Ontario) and integrating the area under the absorbance curve. This integrated area is expressed as the integrated absorbance units (IAU), which is directly proportional to the sample color. The integration method is simpler than the American Dye Manufactures Institute (ADMI) tristimulus filter method but the two methods have been shown to yield similar results [6,17].
+
+The dissolved ozone concentration was measured by the indigo method [17]. The 5-day BOD (BOD5) and COD of the samples were all measured by standard methods [17].
+
+### Apparatus and procedure of ozonation
+
+Ozone was generated from pure oxygen by a Model GL-1 ozone generator (PCI-WEDECO Environmental Technologies, Charlotte, NC). The concentration of ozone in the output gas of the ozone generator was measured by an ozone monitor (Model HC-400) from the same company. The applied ozone dose could be readily adjustable by varying ozone weight percentage and/or gas flowrate. In this study it was adjusted by varying ozone weight percentage while keeping the gas flowrate and pressure constant. The ozone weight percentages examined were 1.0, 1.5, 2.0, 2.5 and 4.0%, respectively. The reactor was a 500 mL gas washing bottle equipped with a porous gas diffuser (VWR Canlab, Mississauga, Ontario). Excess ozone leaving the reactor was destroyed by a catalytic ozone-destruct unit filled with Carllite catalyst (Carus Chemical Company, Peru, IL). More details were described elsewhere [6,10].
+
+The aqueous dye solution was prepared by dissolving C.I. Reactive Blue 15 in distilled, deionized water. The dye concentration of the dye solution varied to simulate those dye concentrations found in residual dyebaths and diluted effluents [6].
+
+The ozonation process started when the ozone-oxygen mixture from the ozone generator was sparged into the reactor, which was operated in semi-batch mode by feeding the ozone-containing gas continuously. The operating pressure was 82.73 kPa (12 psi). Samples were taken at appropriate time intervals to analyze color and concentration of dissolved ozone, respectively. The BOD5 and COD of the samples were measured before and after 30 min ozonation.
+
+## 3 Results and discussion
+
+### Decolorization kinetics
+
+For the decolorization of Reactive Blue 15, the first-order behavior with respect to the dye concentration was observed in all experimental runs:
+
+\[-\frac{\text{d}C_{\text{dye}}}{\text{d}t}=kC_{\text{dye}} \tag{1}\]
+
+where \(C_{\text{dye}}\) is the concentration of the dye and \(k\) is the apparent rate constant. This observation is in agreement with previous reports involving ozonation of aqueous organic chemicals. The apparent rate constant \(k\) is the product of dissolved ozone concentration, \(C_{\text{ozone}}\), and the intrinsic rate constant [10,12,13,18,19,20].
+
+Fig. 1 shows the variation of \(k\) with initial dye concentration, applied ozone dose (\(D_{\text{ozone}}\)) and temperature. It is observed from Fig. 1 that \(k\) increases with the applied ozone dose and temperature, but declines logarithmically with initial dye concentration.
+
+According to Fig. 1, \(k\) declines logarithmically with the initial dye concentration as:
+
+\[k=0.045C_{\text{dye}}^{-0.8258} \tag{2}\]
+
+where the units of \(k\) and \(C_{\text{dye}}\) are min\({}^{-1}\) and g/L, respectively. The fact that \(k\) varied with the initial dye concentration at a constant temperature further confirmed that it is an apparent rate constant instead of a real one, because the latter is the function of temperature only [19]. The decrease in \(k\) could be attributed to the generation of more intermediates with the increase in \(C_{\text{dye}}\). These intermediates consumed more ozone, and \(C_{\text{ozone}}\) was thus lower with the higher initial dye concentration. Consequently, the apparent rate constant declined with \(C_{\text{dye}}\). The logarithmic relationship between the apparent rate constant and the initial dye concentration derived in this study is in agree
+
+Figure 1: Dependence of the apparent rate constant of decolorization \(k\) on initial dye concentration, applied ozone dose and temperature. (a) \(D_{\text{ozone}}\) = 26.1 mg/L min, \(T\) = 20 \({}^{\circ}\)C. (b) \(C_{\text{dye}}\) = 1.0 g/L, \(T\) = 20 \({}^{\circ}\)C. (c) \(C_{\text{dye}}\) = 1.0 g/L, \(D_{\text{ozone}}\) = 26.1 mg/L min.
+
+ment qualitatively with that observed in our previous study on a textile wastewater system containing more than one reactive dye and a number of proprietary auxiliary ingredients [6]. The linear log(_k_)-log(_C_dye) relationships observed from the two systems of different compositions indicate that the linearity between log(_k_) and log(_C_dye) seems valid for the ozonation of reactive dyes regardless of the co-existence of other compounds. Revelation of such linearity makes it possible to predict the rate constant from the initial dye concentration, provided that this relationship is further verified by testing more textile wastewater systems.
+
+For the increase of \(k\) with the applied ozone dose and temperature observed in Fig. 1, the explanations are that higher applied ozone dose increased the dissolved ozone concentration, _C_ozone, which makes the decolorization faster; while higher temperature resulted in larger rate constant according to the Arrhenius equation [19].
+
+### Ozone mass transfer
+
+In a completely mixed semi-batch reactor with no chemical reaction, the mass balance of ozone is expressed as [21]:
+
+\[\frac{\text{d}C_{\text{ozone}}}{\text{d}t}=k_{\text{L}}^{\text{o}}a(C^{*}-C_{ \text{ozone}})-r \tag{3}\]
+
+where _C_* is the equilibrium ozone concentration at the gas-liquid interface, _k_L__a_ is the physical volumetric mass transfer coefficient of ozone [21], and \(r\) is the rate of ozone self-decomposition. In a solution of low pH, the rate of ozone decomposition is negligible [22]. By neglecting ozone self-decomposition and integrating Eq. (3) from _C_ozone = 0 at \(t\) = 0 to _C_ozone = _C_ozone at \(t\) = \(t\) yields
+
+\[\ln(C^{*}-C_{\text{ozone}})=-k_{\text{L}}^{\text{o}}at+\ln C^{*} \tag{4}\]
+
+Values of _k_L__a_ can be obtained by plotting the left-hand side of Eq. (4) versus the ozonation time, \(t\). For the values of _C_*, they were determined when the concentrations of dissolved ozone stop changing with the gas-liquid contact time.
+
+The dye solutions were ozonated to examine the dependence of the rate of ozone mass transfer on initial dye concentration, applied ozone dose, and temperature. The applied ozone dose was adjusted by changing the ozone weight percentage in the gas mixture while maintaining the gas flowrate at 0.52 L/min and pressure at 82.73 kPa (12 psi). During ozonation, the color of the dye solutions decreased with the time whereas the concentration of the dissolved ozone initially increased with the time, but it would soon reach an equilibrium. The representative profiles can be found in Fig. 2.
+
+When dye solutions instead of organic-free water were ozonated, the adsorption of ozone into water was enhanced. The enhancement factor, \(E\), is the ratio of the amount absorbed into a quiescent liquid in a given time when reaction occurs to the amount absorbed in the same time in the absence of reaction [23]. By measuring the ozone absorption rates in the presence and absence of chemical reactions, respectively, the values of \(E\) can be obtained. Applying the following
+
+Fig. 3: Dependence of the volumetric mass transfer coefficient of ozone on initial dye concentration, applied ozone dose and temperature. (a) _D_ozone = 26.1 mg/L min, \(T\) = 20 °C. (b) _C_dye = 1.0 g/L, \(T\) = 20 °C. (c) _C_dye = 1.0 g/L, _D_ozone = 26.1 mg/L min.
+
+Fig. 2: Color and dissolved ozone concentration of the aqueous dye solution as functions of ozonation time (initial dye concentration = 1.0 g/L, applied ozone dose = 26.1 mg/L min, \(T\) = 20 °C).
+
+equation [24]:
+
+\[E = \frac{k_{\rm L}a}{k_{\rm L}^{\rm o}a} \tag{5}\]
+
+values of \(k_{\rm L}a\) can be obtained.
+
+The variations of \(k_{\rm L}a\) with initial dye concentration, applied ozone dose and temperature are plotted in Fig. 3. It is observed that \(k_{\rm L}a\) increases linearly with \(C_{\rm dye}\), \(D_{\rm ozone}\) and \(T\), respectively. The best fit of the data in Fig. 3 yields the following model:
+
+\[k_{\rm L}a = -0.170+2.491\times 10^{-2}C_{\rm dye}+5.310\times 10^{-3}D_{\rm ozone} \tag{6}\] \[+5.866\times 10^{-3}T\]
+
+where the units of \(C_{\rm dye}\), \(D_{\rm ozone}\) and \(T\) are g/L, mg/L min and \({}^{\circ}\)C, respectively. Results of statistical analysis indicate that \(k_{\rm L}a\) has a strong correlation with the three parameters (\(R^{2}=0.9533\)).
+
+The reason why \(k_{\rm L}a\) increases with \(C_{\rm dye}\) can be drawn from the reaction kinetics and absorption theory [23]. When the initial dye concentration increased, more dissolved ozone was consumed which resulted in a lower \(C_{\rm ozone}\) as described earlier in decolorization kinetics. Therefore, the driving force of ozone mass transfer from the gas-phase to the liquid-phase, \(C^{*}-C_{\rm ozone}\), increased, so did the rate of ozone mass transfer.
+
+Fig. 3 also indicates the increase of \(k_{\rm L}a\) with \(D_{\rm ozone}\). With increased ozone input, the rate of decolorization increased as shown in Fig. 1b, the increased rate of reaction reduced \(C_{\rm ozone}\) and consequently, enhanced the mass transfer of ozone. This is a typical example of gas absorption enhanced by chemical reaction [25].
+
+As for the increase of \(k_{\rm L}a\) with temperature, two effects of temperature should be noted. When temperature rises, the concentration of the dissolved ozone decreases. However, the reaction proceeds faster at higher temperature. According to the results shown in Fig. 1, the overall effect of the temperature rise in the tested temperature range was to increase the rate of dye ozonation, which in turn enhanced the ozone mass transfer.
+
+### The effect of ozonation on biodegradability
+
+The ratio of BOD\({}_{5}\)-to-COD is usually used to measure the biodegradability of the wastewater [26]. A larger BOD\({}_{5}\)-to-COD ratio indicates a higher biodegradability of the wastewater. In this study, the aqueous reactive dye was ozonated to examine the effect of ozonation on biodegradability of the dye solutions. The ozonation was conducted with the gas flowrate of 0.52 L/min and pressure of 82.73 kPa (12 psi) for 30 min. The length of ozonation was selected following a series of preliminary tests, which revealed that changes in BOD\({}_{5}\) and COD were negligible after 30 min. Results are listed in Table 1 where after ozonation, the increase in biodegradability is obvious. The ratio of BOD\({}_{5}\)-to-COD increased 18.7-68.5 times. In the meantime, ozonation removed 51.7-84.6% of COD. Therefore, ozonation of the aqueous reactive dye enhanced the biodegradability and removed COD effectively.
+
+## 4 Conclusions
+
+The decolorization of aqueous C.I. Reactive Blue 15 was a pseudo-first-order reaction with respect to the dye. The apparent rate constant increased with the applied ozone dose and temperature. However, it decreased with initial dye concentration. A logarithmic relationship was derived between the apparent rate constant and initial dye concentration.
+
+\begin{table}
+\begin{tabular}{c c c c c c c}  & (COD)\({}_{b}\)a (mg/L) & (COD)\({}_{30}\)a (mg/L) & COD removal (\%) & (BOD\({}_{5}\))b (mg/L) & (BOD\({}_{5}\))b (mg/L) & Increase in BOD\({}_{5}\)-COD ratioa  \\ \hline \(D_{\rm ozone}\) = 26.1 mg/L min, \(T\) = 20 \({}^{\circ}\)C & & & & & \\ \(C_{\rm dye}\) (g/L) & & & & & & \\ \(0.5\) & 469 & 84.9 & 81.9 & 3.85 & 33.1 & 47.5 \\ \(1.0\) & 956 & 253.3 & 73.5 & 4.06 & 39.8 & 37.0 \\ \(1.5\) & 1426 & 480.6 & 66.3 & 4.34 & 42.2 & 28.9 \\ \(3.0\) & 2804 & 1354.3 & 51.7 & 5.25 & 47.3 & 18.7 \\ \(C_{\rm dye}\) = 1.0 g/L, \(T\) = 20 \({}^{\circ}\)C & & & & & \\ \(D_{\rm ozone}\) (mg/L min) & & & & & & \\ \(16.7\) & 956 & 263.9 & 72.4 & 4.06 & 34.8 & 31.1 \\ \(26.1\) & 956 & 253.3 & 73.5 & 4.06 & 39.8 & 37.0 \\ \(34.8\) & 956 & 173.0 & 81.9 & 4.06 & 40.9 & 55.7 \\ \(43.6\) & 956 & 159.7 & 83.3 & 4.06 & 41.3 & 60.9 \\ \(69.7\) & 956 & 147.2 & 84.6 & 4.06 & 42.8 & 68.5 \\ \(C_{\rm dye}\) = 1.0 g/L, \(D_{\rm ozone}\) = 26.1 mg/L min & & & & & \\ \(T\)(C) & & & & & & \\ \(10\) & 956 & 271.5 & 71.6 & 4.06 & 37.1 & 32.2 \\ \(20\) & 956 & 253.3 & 73.5 & 4.06 & 39.8 & 37.0 \\ \(30\) & 956 & 234.2 & 75.5 & 4.06 & 41.0 & 41.2 \\ \(40\) & 956 & 209.4 & 78.1 & 4.06 & 41.7 & 46.9 \\ \end{tabular}
+\end{table}
+Table 1: COD removal and biodegradability increase of Reactive Blue 15 after 30 min ozonation (gas flowrate = 0.52 L/min, gas pressure = 82.73 kPa (12 psi))Mass transfer of ozone during the decolorization of aqueous reactive dye in the semi-batch reactor was increased linearly with initial dye concentration, applied ozone dose and temperature, respectively. A model was developed to predict the volumetric mass transfer coefficient in the experimental range.
+
+Ozonation improved the biodegradability of the aqueous reactive dye solution. The increase in biodegradability ranged from 18.7 to 68.5 times. Concurrently, ozonation removed wastewater COD effectively. The COD removal ranged from 51.7 to 84.6%
+
+## Acknowledgement
+
+The authors wish to express their gratitude to the Natural Sciences and Engineering Research Council of Canada (NSERC) for providing financial support for this work.
+
+## References
+
+* [1] W.C. Tincher, Mills will face new effluent challenges, Textile World (May) (1993) 60-62.
+* [2] I.M. Banat, P. Nigam, D. Singh, R. Marchant, Microbial decolorization of textile-dye-containing effluents: a review, Bioresour. Technol. 58 (1996) 217-227.
+* an important technique to comply with new German laws for textile wastewater treatment, Water Sci. Technol. 30 (1994) 255-263.
+* [4] S.E. Law, J. Wu, M. Eiteman, Ozone decolorization of cotton dyebouse wastewater, in: Proceedings of the ASAE Annual International Meeting, Phoenix, Arizona, July 14-18, 1996.
+* [5] P.C. Vandevivere, R. Biznchi, W. Verstraete, Treatment and reuse of wastewater from the textile wet-processing industry: review of emerging technologies, J. Chem. Technol. Biotechnol. 72 (1998) 289-302.
+* [6] J. Wu, M. Eiteman, S.E. Law, Evaluation of membrane filtration and ozonation processes for treatment of reactive-dye wastewater, J. Environ. Eng. 124 (1998) 272-277.
+* [7] S. Liakou, S. Pavlou, G. Lyberatos, Ozonation of azo dyes, Water Sci. Technol. 35 (1997) 279-286.
+* [8] G. Mishra, M. Tripathy, A critical review of the treatments for decolorization of textile effluent, Colourage 10 (1993) 35-38.
+* [9] E. Razo-Flores, M. Luijten, B. Donlon, G. Lettinga, J. Field, Biodegradation of selected azo dyes under methanogenic conditions, Water Sci. Technol. 36 (1997) 65-72.
+* [10] J. Wu, T. Wang, Ozonation of aqueous azo dye in a semi-batch reactor, Water Res. 35 (2001) 1093-1099.
+* [11] A. Lopez, G. Ricco, G. Mascolo, G. Tiravanti, A.C.D. Pinto, R. Passino, Biodegradability enhancement of refractory pollutants by ozonation: a laboratory investigation on an azo-dyes intermediate, Water Sci. Technol. 38 (1998) 239-245.
+* [12] J. Carriere, P. Jones, A.D. Broadbent, Decolorization of textile dye solutions, Ozone Sci. Eng. 15 (1993) 189-200.
+* [13] F.M. Saunders, J.P. Gould, C.R. Southerland, The effect of solute competition on ozonolysis of industrial dyes, Water Res. 17 (1983) 1407-1419.
+* [14] H. Shu, C. Huang, Degradation of commercial azo dyes in water using ozonation and UV enhanced ozonation process, Chemosphere 31 (1995) 3813-3825.
+* [15] M. Tzitzi, D.V. Vayenas, G. Lyberatos, Pretreatment of textile industry wastewaters with ozone, Water Sci. Technol. 29 (1994) 151-160.
+* [16] J. Hoigne, Mechanisms, rates and selectivities of oxidations of organic compounds initiated by ozonation of water, in: R.G. Rice, A. Netzer (Eds.), Handbook of Ozone Technology and Applications, Ann Arbor, Silicon, Ann Arbor, MI, 1982, pp. 341-379.
+* [17] APHA, in: A.E. Greenberg, L.S. Clesceri, A.D. Eaton (Eds.), Standard Methods for the Examination of Water and Wastewater, 21st ed., American Public Health Association, Washington, D.C., 2005.
+* [18] B. Langlais, D.A. Reckhow, D.R. Brink, Ozone in Water Treatment: Application and Engineering, Lewis Publishers, Boca Raton, FL, 1991, pp. 31-54.
+* [19] H.S. Fogler, Elements of Chemical Reaction Engineering, Prentice Hall PTR, Upper Saddle River, New Jersey, 1999.
+* [20] W. Chu, C. Ma, Quantitative prediction of direct and indirect dye ozonation kinetics, Water Res. 34 (2000) 3153-3160.
+* [21] J. Roth, D.E. Sullivan, Solubility of ozone in water, Ind. Eng. Chem. Fundam. 20 (1981) 137-140.
+* [22] M. Gurol, P. Singer, kinetics of ozone decomposition: a dynamic approach, ES&T 16 (1982) 377-383.
+* [23] P.V. Danckwerts, Gas-Liquid Reactions, McGraw-Hill Book Company, New York, 1970, pp. 34 and 105.
+* [24] V. Tufano, R. Andreozzi, V. Caprio, M. D'Amore, A. Insola, Optimal operating conditions for lab-scale ozonation reactors, Ozone Sci. Eng. 16 (1994) 181-195.
+* [25] R.B. Bird, W.E. Stewart, E.N. Lightfoot, Transport Phenomena, second ed., John Wiley & Sons, Inc., Toronto, 2002.
+* [26] N.J. Karrer, G. Ryniner, E. Heinzle, Applicability test for combined biological-chemical treatment of wastewater containing biorefractory compounds, Water Res. 31 (1997) 1013-1020.

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2010.03.009.mmd

@@ -0,0 +1,150 @@
+Oxidation of hydrochlorothiazide by UV radiation, hydroxyl radicals and ozone: Kinetics and elimination from water systems
+
+Francisco J. Real, Juan L. Acero, F. Javier Benitez, Gloria Roldan, Luz C. Fernandez
+
+Departamento de Ingenieria Quimica y Quimica Fisica, Universidad de Extremadura, Avda. de Elvas s/n, 06071 Badoloz, Spain
+
+A. B. S. T. R. A. A. B. S. T. R. A. B. S.
+
+## 2 Experimental
+
+### Standards, reagents and water systems
+
+Hydrochlorothiazide was obtained from Sigma-Aldrich (Buchs SG, Schweiz) with the highest purity available. Other chemicals were of analytical grade or higher. Solutions of HCTZ, analytical reagents, ozone, and phosphate buffers were prepared with ultrapure water produced from a Milli-Q(Millipore, Bedford, USA) water purification system.
+
+Two water systems were used to investigate the oxidation of the selected pharmaceutical under realistic water treatment conditions. The first one was a surface water (SW) from the public reservoir "Pena del Aguila" located near Badajoz (Extremadura region, South-West of Spain); and the second one was a secondary effluent (SE) generated in the municipal wastewater treatment plant of Badajoz. Samples were immediately processed or stored in a refrigerator (<4 degC) inside glass bottles. The values measured for pH, TOC content, absorbance at 254 nm, and alkalinity for these waters are listed in Table 1. Specifically, UV absorbance and TOC content constitute a significant indication of the total dissolved organic matter (DOM) present in these waters.
+
+### Experimental procedures
+
+The reactor used in the photochemical experiments of hydrochlorothiazide consisted of a 500 mL cylindrical glass vessel with an external jacket surrounding the reactor, and a water stream was pumped from a thermostatic bath in order to maintain the temperature at the designed value within +-0.5 degC. For these photodegradation experiments, a low pressure vapor mercury lamp (TNN 15/32, nominal electrical power 15 W; Heraeus, Madrid, Spain) which emitted monochromatic radiation at 254 nm was used. This lamp was located inside the reactor in axial position and was protected by a quartz sleeve which housed the lamp. Several experiments were performed at different temperatures (10-40 degC) and pH values (3-9), and in the presence and absence of hydrogen peroxide. For every experiment conducted, the reactor was filled with 350 mL of HCTZ (10 mM) solutions (plus hydrogen peroxide in the combined UV/H2O2 experiments) at the selected pH (10 mM phosphate buffer). Samples were periodically withdrawn from the reactor to measure the residual HCTZ concentrations.
+
+The oxidation experiments by Fenton's reagent were carried out in 250 mL Erlenmeyer flasks submerged in a thermostatic bath at 20 degC, and the solutions were homogenized by means of a magnetic stirrer. For every experiment conducted, each flask was filled with the aqueous solution containing the selected pharmaceutical, previously buffered at the selected pH (in the range 2-4) by adding a perchloric acid/perchlorate solution (10 mM). The required amounts of ferrous ion and hydrogen peroxide were also added to the reactor, thus starting the reaction. Sodium sulfite was used to quench the reaction in the samples withdrawn from the reactor at regular time intervals. Additional Fenton's reagent experiments were also performed for the determination of the rate constants of the reaction between HCTZ and hydroxyl radicals by competition kinetics. The selected reference compound was p-chlorobenzoic acid (p-CBA).
+
+The ozonation experiments were carried out in heterogeneous conditions with respect to ozone (which was fed in a gas stream) in the same reactor used for the photochemical experiments, which was also provided with an additional outlet for the exit of the effluent gas. Ozone was produced from borthetled synthetic air in a laboratory ozone generator (Sander, mod. 300.5; Ingetesca, Barcelona, Spain). The ozone-air gas flow rate in every experiment was set at 40 L h-1, and the ozone partial pressure was 0.049 kPa. This stream was introduced through a porous plate into the reactor, which contained a solution of 350 mL of HCTZ (10 mM) and tert-butylalcohol (t-BuOH, 0.01 M) as OH radical scavenger, and buffered at the selected pH (3-9) by adding phosphoric acid/phosphate solution (0.01 M). Competition kinetics was also applied to determine the second-order rate constant at 20 degC for the direct reaction between ozone and HCTZ, using metoprolol (10 mM) as the reference compound. At regular reaction times, samples were taken from the reactor for analysis.
+
+Additional experiments were carried out with HCTZ dissolved in two water systems (SW and SE) at 20 degC and the natural pH of each water. Thus, photodegradation experiments were performed in the same reactor above described to investigate the elimination of HCTZ (1 mM) in the selected waters. Another set of homogeneous ozonation experiments was carried out in a 500 mL flask reactor. Ozone stock solutions were prepared by dissolving an ozone stream in ice-cold ultrapure water until the saturation was reached. Each run was initiated by injecting the volume of the ozone stock solution required to achieve the desired initial O3 dose into the flask, which contained a solution of HCTZ and p-CBA (1 mM) in the selected water. The degradation of p-CBA throughout the ozonation process provided a measurement of OH radical exposure. At regular reaction times, two samples were withdrawn: one was directly introduced into an indigo solution to determine the remaining ozone concentration; the second one was introduced into a vial containing potassium thiosulfate (0.1 M) to quench the residual ozone, and the pharmaceutical concentration was then measured. Finally, some ozonation experiments were performed in static dose mode by varying the initial ozone concentration (0-10 mg L-1). These experiments were started by adding different amounts of the ozone stock solution to aliquots of 17 mL of HCTZ solutions (1 mM). The residual HCTZ concentration was analyzed after 2 h, time enough for complete ozone consumption.
+
+### Analytical methods
+
+The pharmaceutical HCTZ, as well as p-CBA and metoprolol (used as reference compounds), were analyzed by HPLC in a Waters Chromatograph equipped with a 2487 Dual l Detector and a Waters Nova-Pak C18 Column (5 mm 150 mm x 3.9 mm). The detection was performed at 226 nm for HCIZ, 238 nm for p-CBA and 222 nm for metoprolol. The mobile phase was a mixture of methanol and 0.01 M aqueous phosphoric acid solution (10:90 in volume), the elution flow rate was 1 mL min-1 and the injection volume was 50 mL in all samples. The ozone concen
+
+\begin{table}
+\begin{tabular}{l l l} \hline Parameter & Reservoir water (SW) & Secondary effluent (SE) \\ \hline pH & 7.3 & 8.0 \\ TOC (mg L−1) & 4.3 & 23.3 \\
+254 nm absorbance (cm−1) & 0.118 & 0.191 \\ Allatility (mg L−1 CaCO3) & 30 & 250 \\ \hline \end{tabular}
+\end{table}
+Table 1: Water quality characteristics of the two real waters.
+
+Figure 1: Chemical structure of HCTZ.
+
+tration in the inlet gas ozone-oxygen stream was determined iodometrically, while the ozone concentration in the stock solutions was determined directly by measuring their UV absorbance at 258 nm (\(\varepsilon\) = 3150 L mol\({}^{-1}\) cm\({}^{-1}\)). Remaining ozone concentration in the homogeneous ozonation experiments was analyzed by the Indigo method [11].
+
+## 3 Results and discussion
+
+### Oxidation of hydrochlorothiazide in ultrapure water
+
+In a first stage, the oxidation of the selected pharmaceutical HCTZ dissolved in ultrapure water was conducted by using different oxidizing systems: UV radiation, Fenton's reagent and ozone, with the aim of establishing the influence of some operating variables on the degradation process as well as of determining the main kinetic parameters. The initial concentration of HCTZ in these experiments was 10 mM.
+
+Firstly, HCTZ photodegradation was studied out by means of the monochromatic radiation described in Section 2. These experiments were performed at different temperatures (10-40 degC) and pH values (3-9). Additionally, two experiments were also conducted in the presence of hydrogen peroxide at initial concentrations of \(5\times 10^{-5}\) and \(1\times 10^{-4}\) M. Table 2 compiles the runs performed and their experimental conditions. In order to examine the effect of these variables on the photodecomposition reaction, Table 2 also summarizes the HCTZ removal (\(X_{\text{UV}}\)) obtained after 5 min of reaction.
+
+From these \(X_{\text{UV}}\) values, it is deduced that no significant effect was exerted by the pH, because similar removals were obtained at any pH, which suggests that only the direct photodecomposition contributes to the global elimination of HCTZ. On the other hand, the temperature of the reaction had a positive effect on the degradation rate as would be expected, with increasing removals when the temperature was increased.
+
+In a first approach the kinetic study is conducted by considering that the photochemical process follows first-order kinetics. According to this, a plot of \(\ln[\text{HCTZ}]_{0}/[\text{HCTZ}]\) vs. reaction time should result in a straight line, whose slope provides the first-order rate constant \(k_{\text{UV}}\). Following this procedure and after regression analysis (\(r^{2}\) > 0.99), the rate constants that are compiled in Table 2 were deduced. As observed for the variation of pH, the values are quite similar, with an average value of 0.152 min\({}^{-1}\). On the contrary, the rate constants increased with temperature, from 0.110 to 0.170 min\({}^{-1}\). Finally, the addition of hydrogen peroxide affected positively as was commented, and \(k_{\text{UV}}\) increased from 0.146 min\({}^{-1}\) (Expt. UV-2) to 0.202 and 0.278 min\({}^{-1}\) (Expts. UVH-1 and UVH-2).
+
+A more rigorous kinetic study was based on the evaluation of the quantum yield (\(\phi\)) for the photodegradation of the selected pharmaceutical compound. The reaction model described in detail in a previous work [12] was used, which provides a general equation for the rate of disappearance of a general micropollutant as a function of the absorbed radiation flux \(W_{\text{abs}}\):
+
+\[[\text{HCTZ}]=[\text{HCTZ}]_{0}-\frac{\phi}{\sqrt{}}\int_{0}^{t}W_{\text{abs}}\,\,\text{dt} \tag{1}\]
+
+The procedure for determining \(W_{\text{abs}}\), was also described elsewhere [12], and the integral term \(\left\{W_{\text{abs}}\,\text{df}\,\text{was calculated numerically, by fitting the experimental data ($W_{\text{abs}}$, $t$) to a polynomial expression by least squares regression, and integrating the resulting function. Previously, the radiation intensity emitted by the lamp into the reactor was evaluated by chemical actinometry experiments, using hydrogen peroxide as actinometer, being the obtained value of \(2.0\times 10^{-6}\) E s\({}^{-1}\). At the same time, the molar extinction coefficient at 254 nm of HCTZ was determined at different pH values, with a value of 6650 mol\({}^{-1}\) L cm\({}^{-1}\) valid in the pH range 3-9. A plot of the pharmaceutical concentration [HCTZ] vs. the corresponding integral \(\left\{W_{\text{abs}}\,\text{df}\,\text{led to a straight line, whose slope provided the overall quantum yield $\phi$ of the photodegradation, values that are also listed in Table 2. It can be seen that the trend previously observed for \(k_{\text{UV}}\) is exactly reproduced in the quantum yields: almost no influence with the variation of the pH (an average value of \(0.041\pm 0.004\) molEins\({}^{-1}\) is proposed at any pH at 20 degC); while increasing \(\phi\) values when the temperature was increased. Then, an Arrhenius-type expression can be proposed, and after linear regression analysis, 54 mol Eins\({}^{-1}\) and 2.14 kcal/mol were deduced for the \(\phi_{0}\) and \(E_{\text{a}}\) parameters. It must be noted that this kinetic procedure cannot be applied to the oxidation experiments by the combination UV/H\({}_{2}\)O\({}_{2}\) because its reaction mechanism is different than the single photoreaction mechanism, and additional reactions with different rate constants must be taken into account.
+
+In a later stage, individual oxidation experiments of HCTZ with Fenton's reagent were performed at 20 degC as described in Section 2, by varying the initial concentrations of ferrous ions and hydrogen peroxide (2.5-10 \(\times\) 10\({}^{-5}\) for both reactants), as well as the pH (2-4).
+
+As it is known, in the Fenton's reagent ferrous ion reacts with hydrogen peroxide to generate OH radicals. These radicals generated oxidize most organic compounds. From the experimental results obtained it can be concluded that both species, Fe\({}^{2+}\) and H\({}_{2}\)O\({}_{2}\), promote a direct effect on the pharmaceutical removal, which is consequent with the most important reaction of the Fenton's reagent, responsible of the generation of the oxidant OH radical [13].
+
+In addition, decomposition experiments of HCTZ were performed by varying the pH from 2 to 4, which is proposed as the optimum range of pH by other authors [14, 15]. The results confirm that the pH 3 is the optimum, being the obtained HCTZ removal highest; while at pH 2 and 4, the removals were lower. Thus, the decrease in the degradation at pH 4 can be attributed to a decrease of the free iron species in the solution, due to the precipitation of ferric oxyhydroxides. On the other hand, the decrease in the degradation at pH 2 is due to the inhibition of the formation Fe (III)-peroxy complexes, which are the precursors of iron (II) regeneration [16].
+
+\begin{table}
+\begin{tabular}{l c c c c c c} \hline \hline Expt. & \(T\) (\({}^{\circ}\)C) & pH & \([\text{H}_{2}\text{O}_{2}]_{0}\times 10^{2}\) (M) & \(X_{\text{UV}}\) (x) (5 min) & \(k_{\text{UV}}\) (min\({}^{-1}\)) & \(\phi\) (mol Eins\({}^{-1}\)) \\ \hline UV-1 & 20 & 3 & & 45.2 & 0.164 & 0.046 \\ UV-2 & 20 & 5 & & 43.2 & 0.149 & 0.040 \\ UV-3 & 20 & 7 & & 41.2 & 0.146 & 0.038 \\ UV-4 & 20 & 9 & & 43.8 & 0.151 & 0.038 \\ UV-5 & 10 & 7 & & 30.5 & 0.110 & 0.028 \\ UV-6 & 30 & 7 & & 53.1 & 0.164 & 0.046 \\ UV-7 & 40 & 7 & & 58.9 & 0.170 & 0.048 \\ UVH-1 & 20 & 7 & 5 & 55.8 & 0.202 & - \\ UVH-2 & 20 & 7 & 10 & 65.8 & 0.278 & - \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Experimental conditions and kinetics parameters obtained in the photodegradation of HCTZ in UPt water. \([\text{HCTZ}]_{0}\) = 10 mM.
+
+In general terms, the rate equation for the removal of an organic compound (here HCTZ) by means of hydroxyl radicals can be written in the form:
+
+\[-\frac{d[{\rm HCTZ}]}{d{\rm r}} = k_{{\rm OH-HCTZ}}\left[ {\rm{}^{*}OH} \right]\left[ {\rm{}^{*}HCTZ} \right]\]
+
+where _k_OH-HCTZ is the second-order rate constant between OH radicals and the pharmaceutical compound. Generally, the radical rate constants present high values (in the range of 107-1010 M-1 s-1) [17,18]. The determination of the unknown rate constant _k_OH-HCTZ cannot be made directly, and competition kinetics is frequently used, which is based in the simultaneous oxidation of the target compound (HCTZ in the present case) and a reference compound R whose rate constant for the reaction with OH radicals _k_OH-R is previously known. The application of this model provides this relationship for the evaluation of the rate constant _k_OH-HCTZ:
+
+\[\ln\frac{\left[ {\rm{}^{*}{\rm{}}{\rm{tion process were the same as for the UV degradation in ultrapure water. In the ozonation experiments, p-CBA was used as a probe compound to determine OH radical exposure. Several conditions were maintained constant in all these experiments: the temperature was 20 degC and the initial concentrations of HCTZ and p-CBA were 1 uM. These experiments were performed at the pH of each water, 7.3 and 8.0 for SW and SE, respectively.
+
+Regarding to the photodegradation process, it is again considered that the process follows first-order kinetics, and therefore, the experimental terms \(\ln[(\text{HCTZ}_{0}\rbrack_{0}/[\text{HCTZ}\rbrack)\) were plotted vs. reaction time. After regression analysis, the slopes were evaluated and provided the first-order rate constant \(k_{\text{UV}}\). The values obtained are 0.117 and 0.094 min-1 for the SW and SE waters, respectively, below the value of 0.146 min-1 obtained in the experiment in UP water at the same operating conditions. From these rate constants, it can be observed a lower removal rate in the SE, intermediate in the SW, and higher rate in UP water. These results can be explained by the fact that the dissolved organic matter (DOM) present in the SW and the SE could absorb some amount of UV radiation, in a higher extent in the SE due to a higher content in DOM (see Table 1). On the contrary, UP water does not contain any amount of organic matter, and therefore, the radiation absorbed is totally consumed in the degradation of the pharmaceutical, providing a higher elimination rate (\(k_{\text{UV}}\) = 0.146 min-1). Nevertheless, HCTZ was almost completely eliminated after 30 min, even in the SE.
+
+In the ozonation experiments performed in static dose mode, different ozone doses were added to both water matrices containing the pharmaceutical compound. Fig. 3 shows the remaining concentration of HCTZ obtained after complete ozone consumption. As it is seen, an increase in the ozone dose leads to a logical increase in the removal of HCTZ. It is also observed that the degradation extent was higher in the in the SW than in the SE, as a consequence of the DOM present which consumes part of the oxidant, being this DOM lower for SW and higher for the SE. Thus, while 3 mg L-1 of ozone was enough to completely remove HCTZ from the SW, an ozone dose higher than 10 mg L-1 was needed in the SE. In conclusion, in water systems with higher amount of dissolved organic matter, the amount of oxidant available to react with micropollutants is lower, requiring higher doses of oxidant to reach the desired degree of pollutant elimination.
+
+Time-resolved experiments were run with an initial ozone dose of 4 mg L-1 for SW water and 10 mg L-1 for SE, as typically found in full-scale drinking water and secondary effluent treatments. Initial ozone decay (data not shown) was very fast due to fast reactions of ozone with organic and inorganic matter present in these waters (instantaneous ozone demand) in which hydroxyl radicals are generated. Then, ozone consumption became slower until complete depletion which was reached after 10 and 1 min in SW and SE, respectively. HCTZ elimination was very fast, being completely removed after 30 s in both water matrices. Therefore, HCTZ can be completely removed in water matrices of diverse quality when the ozone exposure is measurable (ozone residual after 30 s of reaction). These results confirm the high efficiency of ozone to remove the selected pharmaceutical in different waters.
+
+In order to predict the ozonation process of HCTZ in different water matrices, a kinetic model is proposed, which is based on the previously calculated rate constants \(k_{\text{O}_{\text{2-HCTZ}}}\) and \(k_{\text{OH-HCTZ}}\) as well as the ozone and OH radicals concentration decays. The ozone depletion in natural waters can be followed by measuring its concentration at regular reaction times. However the knowledge of the OH radical evolution presents more problems, since there is no method for the direct measurement of its concentration. Then, in order to measure the transient OH radical concentration during ozonation processes, Elovitz and von Gunten [24] introduced the \(R_{\text{ct}}\) parameter, defined as the ratio between the OH radicals and \(\text{O}_{3}\) exposures:
+
+\[R_{\text{ct}}=\frac{\int[^{\bullet}\text{OH}]\,\text{d}t}{\int[\text{O}_{3}]\, \text{d}t} \tag{5}\]
+
+In this Eq. (5) the ozone exposure can be evaluated from the integration of the ozone concentration (directly measured) vs. time data; while the \(R_{\text{ct}}\) parameter must be determined by the measurement of the decay of an ozone-resistant probe compound, which reacts rapidly with OH radicals and presents a very low reactivity towards ozone. In this study, the probe compound selected was again p-CBA, whose rate constants with ozone and OH radicals are \(k_{\text{O}_{\text{2-p-CBA}}}=0.15\,\text{M}^{-1}\,\text{s}^{-1}\) and \(k_{\text{OH-p-CBA}}=5\times 10^{9}\,\text{M}^{-1}\,\text{s}^{-1}\), respectively [25,17]. As the reaction between OH radicals and p-CBA follows second-order kinetics, the integration of its rate equation and the introduction of the defined \(R_{\text{ct}}\) parameter (Eq. (5)) leads to the following expression for the probe compound concentration profile:
+
+\[\ln\left(\frac{[\text{p-CBA}]_{0}}{[\text{p-CBA}]_{\text{f}}}\right)=k_{\text{OH -p-CBA}}\int_{0}^{t}[\text{OH}]\,\text{d}t=k_{\text{OH-p-CBA}}\cdot R_{\text{ct} }\cdot\int_{0}^{t}[\text{O}_{3}]\,\text{d}t \tag{6}\]
+
+Fig. 3: Influence of the type of water matrix on the final removals of hydrochlorothiazide using different ozone doses. Experimental conditions: \(T\) = 20 °C; pH = 7.3 for SW and pH = 8.0 for SE; [HCTZ]0 = 1.0 μM.
+
+Fig. 2: Apparent rate constants for the oxidation of hydrochlorothiazide with ozone. Comparison of experimental (symbols) and calculated values from Eq. (4) (line). Experimental conditions: [HCTZ]0 = 10 μM; \(T\) = 20 °C; \(\text{pO}_{3}\) in the gas inlet: 0.049 kPa.
+
+According to Eq. (6), the \(R_{\rm{ct}}\) value and the ozone exposure were obtained, the evaluation of the OH radicals exposure at any reaction time, and consequently, the knowledge of the OH radicals concentration profile in a ozonation process is obtained by using Eq. (5).
+
+Following the described procedure, Fig. 4 presents the plots mentioned for both water matrices (Expts. OZ-SW and OZ-SE), and the \(R_{\rm{ct}}\) values listed in Table 4 were determined. As can be observed, two \(R_{\rm{ct}}\) values were deduced for each experiment: higher \(R_{\rm{ct}}\) values in the initial period of the reaction indicate a higher amount of OH radicals formed from ozone decomposition, being this decomposition lower in the second period.
+
+As was initially proposed, the final goal is the prediction and modeling of the oxidation of HCTZ in water systems by ozone and OH radicals as oxidant species. For this purpose, the \(R_{\rm{ct}}\) values determined are useful. In effect, assuming second-order kinetics for the reaction between many micropollutant (HCTZ in this case) and both oxidants, the reaction rate for this compound when present in any type of water can be written in the form:
+
+\[\left[{\rm{HCTZ}}\right]_{\rm{f}} = \left[{\rm{HCTZ}}\right]_{\rm{0}} \tag{7}\] \[\exp\left\{-\left(\int_{\rm{0}}^{t}\left[{\rm{O}}_{\rm{3}}\right] \rm{d}t\right)\cdot\left(k_{\rm{OH-HCTZ}}\cdot R_{\rm{ct}}+k_{\rm{O}_{\rm{3}- HCTZ}}\right)\right\}\]
+
+As \(k_{\rm{OH-HCTZ}}\), \(k_{\rm{O}_{\rm{3}-HCTZ}}\) and the \(R_{\rm{ct}}\) parameter have been previously determined, Eq. (7) can be applied for the determination of the theoretical concentrations of HCTZ in the experiments carried out, being these values compared with the experimental results obtained. The quite satisfactory agreement between values predicted by Eq. (7) and experimental values confirms the goodness of this kinetic approach.
+
+Finally, the \(R_{\rm{ct}}\) parameter is also useful for determining the relative importance of OH radicals and \({\rm{O}}_{\rm{3}}\) reaction pathways in the oxidation of a micropollutant present in water systems. Thus, the fraction of HCTZ degraded in the present study by OH radicals can be expressed in the form:
+
+\[f_{\rm{OH}} = \frac{k_{\rm{OH-HCTZ}}\left[{}^{\bullet}{\rm{OH}}\right]\left[{ \rm{HCTZ}}\right]}{k_{\rm{OH-HCTZ}}\left[{}^{\bullet}{\rm{OH}}\right]\left[{ \rm{HCTZ}}\right]+k_{\rm{O}_{\rm{3}-HCTZ}}\left[{\rm{O}}_{\rm{3}}\right]\left[{ \rm{HCTZ}}\right]} \tag{8}\] \[= \frac{k_{\rm{OH-HCTZ}}R_{\rm{ct}}}{k_{\rm{OH-HCTZ}}R_{\rm{ct}}+k_ {\rm{O}_{\rm{3}-HCTZ}}}\]
+
+Eq. (8) was applied to the results obtained in the experiments carried out in the waters tested, being the percentages obtained for each compound also compiled in Table 4. From these values it is clearly deduced that the direct ozonation pathway predominates over the radical pathway, with almost no influence of the type of water; and this effect more especially in the second period, where the global degradation due to the radical pathway was 8.2 and 1.8%. These results coincide with those of previous authors, which showed that significant OH radical exposure can be experienced during the first seconds of reaction [26,27].
+
+## Acknowledgements
+
+The authors wish to gratefully acknowledge financial support from the _Ministerio de Educacion y Ciencia of Spain_ through the Project CTQ2007-60255 and the Project CONSOLIDER-INGENIO CSD2006-00044.
+
+## References
+
+* [1] T. Heberer, U. Dunniber, C. Reilich, H.J. Stan, Detection of drugs and drug metabolites in ground water samples of a drinking water treatment plant, Fres. Environ. Bull. 6 (1997) 438-443.
+* [2] S. Mompelat, B. Lebot, O. Thomas, Occurrence and fate of pharmaceutical products and by-products, from resource to drinking water, Environ. Int. 35 (2009) 803-814.
+* [3] S.A. Snyder, E. Wert, H.D. Lei, P. Westerhoff, Y. Yoon, Removal of EDCs and Pharmaceuticals in Driding and Reuse Treatment Processes, American Water Works Association Research Foundation (AWWARF), Denver, Colorado, 2007.
+* [4] MJ. Benotti, A. R. Trenholm, B.J. Vanderford, J.C. Holady, B.D. Stanford, G.S. Snyder, Pharmaceuticals and endocrine disrupting compounds in U.S. drinking water, Environ. Sci. Technol. 43 (2009) 597-603.
+* [5] G.C. Daughton, T.A. Termes, Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ. Health Persp. 107 (1999) 907-938.
+* a suitable design parameter to evaluate the capacity of wastewater treatment plants to remove micropollutants, Water Res. 39 (2005) 97-106.
+* [7] S. Suarez, M. Carphali, E. Omid, L. Jem, How are pharmaceutical and personal care products (PPCs) removed from urban wastewaters? Rev. Environ. Sci. Biotechnol. 7 (2008) 125-138.
+* [8] S.A. Snyder, D. Li, Villeenave, E.M. Snyder, J.P. Giesy, Identification and quantification of estrogen receptor agonists in wastewater effluents, Environ. Sci. Technol. 35 (2001) 3620-3625.
+* [9] M.J. Martinez-Bueno, A. Aguera, M.J. Gomez, M.D. Hernando, J.F. Garcia-Reyes, A.R. Fernandez-Alba, Application of IC/quality-linafter ion trap mass spectrometry and time of flight mass spectrometry to the determination of pharmaceuticals and related contaminants in wastewater, Anal. Chem. 79 (2007) 9372-9384.
+* [10] M.Gomez, M.J. Martinez-Bueno, A. Aguera, M.D. Hernando, A.R. Fernandez-Alba, M. Mezeuca, Evaluation of ozone-based treatment processes for wastewater containing microdeterminants using LC-QTRAN-MS and LC-TOF/MS, Water Sci. Technol. 57 (2008) 41-48.
+* [11] H. Bader, J. Hogue, Determination of ozone in water by the Indigo method, Water Res. 15 (1981) 449-456.
+* [12] F.J. Benitez, J. Beltran-Heredia, J.L. Acero, F.J. Ruhio, Oxidation of several chlorophenolic derivatives by UV irradiation and hydroxyl radicals, J. Chem. Technol. Biotechnol. 76 (2001) 3213-320.
+* [13] F.J. Benitez, F.J. Real, J.L. Acero, C. Garcia, E.M. Llanos, Kinetics of phenylurea herbicides oxidation by Fenton and photo-Fenton processes, J. Chem. Technol. Biotechnol. 82 (2007) 65-73.
+
+\begin{table}
+\begin{tabular}{c c c c c} Expt. & \(R_{\rm{ct}}\times 10^{4}\) & \(R_{\rm{ct}}\times 10^{4}\) & \(f_{\rm{OH}}\) (\%) & \(f_{\rm{OH2}}\) (\%) \\ OZ-SW & 61.7 & 8.4 & 39.7 & 8.2 \\ OZ-SE & 82.7 & 4.8 & 24.2 & 1.8 \\ \end{tabular}
+\end{table}
+Table 4: Oxidation of HCTZ in real waters by ozone: \(R_{\rm{ct}}\) values and fraction of HCTZ degraded by OH radicals.
+
+* [14] B.G. Kwon, D.S. Lee, N. Kang, J von, Characteristics of fp-chlorophenyl oxidation by Fenton's reagent, Water Res. 33 (1999) 2110-2118.
+* [15] W.Z. Tang, C.P. Huang, 2.4-Dichlorophenol oxidation kinetics by Fenton's reagent, Environ. Technol. 17 (1996) 1371-1378.
+* [16] J. De Laat, H. Gallard, Catalytic decomposition of hydrogen peroxide by Fe(III) in homogeneous aqueous solution: mechanism and kinetic modelling, Environ. Sci. Technol. 33 (1999) 2726-2732.
+* [17] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical review of rate constants for oxidation of hydrated electrons, hydrogen atoms and hydroxyl radicals (AOH/aO-) in aqueous solutions, J. Phys. Chem. Ref. Data 17 (1988) 513-886.
+* [18] W.K. Haag, D.C. Yao, Rate constants for reaction of hydroxyl radicals with several drinking water contaminants, Environ. Sci. Technol. 26 (1992) 1005-1013.
+* [19] J. Hoigne, Chemistry of aqueous ozone and transformation of pollutants by zoranotation and advanced oxidation processes The Handbook of Environmental Chemistry, vol. 5, Springer-Verlag, J. Hrubec, Berlin, Germany, 1998 (Part C).
+* [20] F.J. Benitez, J.A. Acero, F.J. Real, G. Roldan, Ozonation of pharmaceutical compounds: rate constants and elimination in various water matrices, Chemosphere 77 (2009) 53-59.
+* [21] J. Sangster, US EPI Estimation Program Interface (EPI) Suite (2000) KOWWIN V.16, EPA and Syracuse Research Corporation, 1994.
+* [22] J. Hoigne, H. Bader, Rate constants of reactions of ozone with organic and inorganic compounds in water. I. Non-dissociating organic compounds, Water Res. 17 (1983) 173-183.
+* [23] J. Benner, E. Salhi, T. Ternes, U. von, Cunzen, Ozonation of reverse osmosis concentrate: kinetics and efficiency of beta blocker oxidation, Water Res. 42 (2008) 3003-3012.
+* [24] M.S. Ebourne, U. von Cuntzen, Hydroxyl radical/ozone ratios during zoranotation processes. I. The \({R}_{n}\) concept, Ozone Sci. Eng. 21 (1999) 239-260.
+* [25] D.C. Yao, W.R. Haag, Rate constants for direct reactions of ozone with several drinking water contaminants, Water Res. 25 (1991) 761-773.
+* [26] M.O. Buffie, J. Schumacher, E. Salhi, M. Jekel, U. von Gutten, Measurement of the initial phase of ozone decomposition in water and wastewater by means of a continuous quench-flow system: application to disinfection and pharmaceutical oxidation, Water Res. 40 (2006) 1884-1894.
+* [27] E.C. Werf, F.L. Rosario-Ortiz, S.A. Snyder, Effect of ozone exposure on the oxidation of trace organic contaminants in wastewater, Water Res. 43 (2009) 1005-1014.

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2011.08.014.mmd

@@ -0,0 +1,222 @@
+# Kinetics of the ozone oxidation of Reactive Orange 16 azo-dye in aqueous solution
+
+Chedly Tizaoui, Naser Grima
+
+# Abstract
+
+Reactive azo-dyes present environmental concerns since they are persistent organics which are difficult to remove with conventional treatment processes. In this study, the ozonation of Reactive Orange 16 dye, C.I. 177757 (synonym Remazol Brilliant Orange 3R) was investigated at different experimental conditions in a semi-batch ozone reactor (25-100 mg/l, dye; 20-80 g/m3* NTP O3 gas (NTP: 0*C and 1 atm); pH 2, 7, 11). Ozone was very effective to remove completely the colour within a short period of time (few minutes). High gas-phase ozone concentrations and low dye concentrations resulted in high decolourisation rates. The decolourisation of the dye has also improved as pH increased from 2 to 11. Direct ozone reactions are assumed to predominate and hydroxyl radicals formed as a result of ozone decomposition at high pH reacted mostly with the products of the main decolourisation reaction. The reaction was assumed second order and Danckwerts model was used to determine the values of its rate constant, \(k\)2, at different pHs. Values of \(k\)2 were 2.5 x 105, 3.2 x 105, and 1.4 x 106 L/mols for pH values of 2, 7, and 11, respectively. The stoichiometric ratio of the reaction was found equal to 3 mol O3/mol dye and a degradation pathway was proposed.
+
+## Introduction
+
+The textile industry consumes large quantity of water and chemicals for various operations such as washing, dyeing, rinsing and finishing. Accordingly, it generates large flows of wastewater which are characterised by high COD, large amount of suspended solids, largely fluctuating pH, high temperature, unbound colourants, and dye impurities, auxiliaries and surfactants [1-3]. This raises difficulties in recycling or reusing the wastewater in the dyeing process because of changes in the character of the residual dye in the bath (i.e. hydrolysis of the dye) and the presence of dyeing auxiliaries and organic substances that affect the dyeing process [4,5]. As a result large amounts of wastewater with poor quality are disposed of. The textile industry consumes on average about 150 L of water per kilogram of cloth processed [1]. The discharge of coloured textile wastewaters into the environment is a major environmental issue due to the noticeable colour and some dyes may have carcinogenic and/or teratogenic health effects. Although biological treatment seems the most cost effective treatment of wastewaters, it generally fails to achieve a treated effluent with a quality that complies with the ever tightening environmental legislations. This is due to the toxic and refractory nature of most textile wastewaters [6]. In addition, conventional physical and chemical treatment methods are inefficient for colour removal [7]. For this reason combination of biological treatment with oxidation such as ozonation, wet air oxidation, hydrogen peroxide treatment and other advanced oxidation processes is very attractive [8-10]. It is common that the main purpose of oxidation, in particular ozonation, is to remove first the residual non-biodegradable colour from the wastewater and to achieve as a second objective the removal of other constituents such as surfactants or partial oxidation of dissolved organic carbon (DOC) to improve the biodegradation of the wastewater [11].
+
+Due to the fact that ozone is very effective in removing colour from textile wastewaters, numerous published studies have emerged over the last decades and continue to grow significantly [2,12-16]. In particular the ozonation of azo-dyes has gained considerable interest due to the facts that:
+
+* azo-dyes form the largest group of dyes used in the textile industry (over 50% of all dyes) [7,17,18];
+* significant input of the azo-dye (more than 50%) is lost in the dyedhouse effluent due to poor fixation efficiency to the fibre [7,19];
+* azo-dyes are difficult to biodegrade in sewage treatment works due to their inherent nature to withstand microbial biodegradation;
+* azo-dyes are highly soluble in water so they do not transfer to sludge at significant extent;
+* ozone is highly reactive with the azo (-N=N-) double bond, so colour can easily be removed from the wastewater with ozone.
+
+Ozone reacts with organic molecules through a combination of molecular and radical reactions. The radical reactions generally involve hydroxyl radicals (*OH), which are produced as a result of ozone decomposition in water, are generally fast but non-selective. Ozone radical reactions may be promoted at high pH (i.e. pH> approx. 7), by addition of H2O2, UV irradiation, and catalysts. The hydroxyl radicals are very strong oxidants having a redox potential of 2.80 V higher than molecular ozone (E0* = 2.07 V). Treatment processes involving hydroxyl radicals are called Advanced Oxidation Processes (AOPs), which are widely used to enhance the degradation rates of the organic molecules [20]. On the other hand, ozone molecular reactions are selective and in most cases they proceed through electrophilic attack; though ozone can react through nucleophilic or oxygen transfer pathways. The electrophilic attack occurs on sites possessing high negative charge density including multiply bonded species such as -C=C- or -N=N-, and atoms such as N, P, O, S. Direct reactions of ozone with compounds having ortho- and para-directing substituent (i.e. electron donors) such as -OH, -CH3, -OCH3, -NH2 may also occur [21]. Several studies have shown that it is the molecular ozone reaction mechanism that is responsible for the partial oxidation of azo dyes (i.e. colour removal) [22, 23, 24] and it was suggested that the first stage of decolourisation occurs through ozone reaction with the double bond azo chromophoric group-N=N- or with the double bond of the -C=C- connecting aromatic rings. Generally the decolourisation with ozone occurs rapidly due to the rapid destruction of the conjugated chains of the dye molecules that are responsible for colour. Rate constants of ozone reactions with dyes in the range of 103 and higher than 107 L/mol s have been reported in the literature [12, 25, 26].
+
+Due to the gaseous nature of ozone, most ozonation studies have been carried out in semi-batch gas/liquid reactors (e.g. bubble column reactor) [27]. Although the experimental system is simple, analysis of the results obtained in semi-batch gas/liquid reactors is difficult. This is mainly due to the complex equations that describe the rate of gas absorption accompanied by a chemical reaction. As a result, a large number of studies paid little attention to the effect of gas-liquid mass transfer on the observed rates and they reported erroneous rate constants using for example simple first-order kinetics, which does not apply for most cases [28, 29, 30]. Hence several ozonation data presented in the literature should be used carefully since such data is system-dependent and are not intrinsic to the compound studied.
+
+In the present study, Reactive Orange 16 (RO16) (Remazol Brilliant Orange 3R, C.1 77757), an anionic sulphonated reactive azo dye, was selected as a model pollutant because it is difficult to biodegrade [31]. Hence it presents environmental concerns since it has potential to persist in the environment and is hard to remove by conventional treatment processes as is the case for most azo dyes [32, 33]. Spagni et al. [33] have shown that RO16 dye was hard to remove which required them to use alternating anaerobic and aerobic bioreactors and extensive biomass acclimation. Moreover, they showed that the biodegradation of RO16 dye resulted in significant concentrations of recalcitrant aromatic amines, which are expected to be highly carcinogenic [34]. RO16 dye belongs to the group of vinyltsulfone dyes and it is used for dyeing cellulose fibres. Fig. 1 shows the molecular structure of RO16 dye and its UV/Vis spectrum. Besides, ozonation data of RO16 dye are not available. Hence this study was carried out and its objectives were to investigate the efficiency of ozone in removing RO16 dye in solution and determine reaction kinetic data, which are currently not available for the semi-batch reactor. Results on the effect of the operating parameters such as ozone concentration, initial dye concentration, pH of solution and reaction time together with the kinetic data are useful to optimise the decolourisation process of RO16 dye. The effect of mass-transfer on the observed kinetics was also studied and a tentative mechanism for RO16 dye degradation with ozone was also proposed.
+
+## Kinetic model
+
+In gas/liquid reactors where mass transfer is accompanied by a chemical reaction, eight different kinetic regimes ranging from extremely slow reaction to extremely fast reaction, each with its particular rate equation, may take place depending on the operating conditions and the physicochemical properties of the system including gas and liquid reactant concentrations, reaction rate constant and order, mass transfer coefficients, Henry's constant, and solutes diffusivities [35, 36]. Hatta number, _Ha_, and the instantaneous enhancement factor (or instantaneous reaction factor), _Ei_, are two important parameters used to characterise the regime under which the system is operating. Two theories are widely used to describe gas/liquid reactions, which are the double-film theory [37] and Danckwerts surface renewal theory [38]. Other models including the penetration model (i.e. Higbie model) and the filamentation model have also been used to describe gas absorption [39]. Depending on the pair (_Ha_, _Ei_) and considering the theory used, different equations may be used to describe the gas absorption rate. The fundamentals of this topic have been discussed extensively in the literature [35, 36, 37, 38, 40].
+
+In this study, it was assumed that ozone reaction with the dye (O3 + Bb - Products), where B means the dye and \(b\) the stoichiometric ratio, is irreversible and of second order, which is a common case for ozone reactions with organics [41, 42, 43, 44]. In addition it was assumed that the reaction develops in the fast kinetic regime, which is also a common case for ozone reactions with dyes; this has also been confirmed by the absence of dissolved ozone during the experiments in this study. For a second-order reaction, _Ha_ and _Ei_ are given by Eqs. (1) and (2), respectively (Danckwerts theory). When a chemical reaction takes place in the liquid phase, the gas absorption rate increases as compared to that without reaction. To account for this increase due to the chemical reaction, the rate of ozone absorption, _N_O2, is hence written as in Eq. (3); where \(E\) is the enhancement factor due to the chemical reaction. Using the Danckwerts model, De Coursey [40] developed an explicit equation to determine \(E\) directly from _Ha_ and _Ei_ (Eq. (4)). Moreover, the rate of decolourisation of the dye is considered stoichiometrically proportional to the ozone absorption rate hence Eq. (5). Integration of Eq. (5) leads to Eq. (6) which was used to calculate the change of dye concentration as function of time \(t\). The second-order rate constant \(k\) was determined by fitting the experimental data to Eq. (6) by minimising the objective function (Eq. (7)) as being the sum of
+
+Fig. 1: Molecular structure and UV/Vis spectrum of RO16 dye.
+
+the squared differences between experimental and calculated ratio dye concentrations relative to the initial concentration at different times. Integration of \(N\)03 relative to time was made numerically and Microsoft Excel Solver was used to minimise the objective function using \(k\)2 as a parameter.
+
+\[H\alpha = \frac{\sqrt{D_{\text{O}_{3}}k_{2}C_{\text{B}}}}{k_{\text{L}}}\]
+
+\[E_{i} = \sqrt{\frac{D_{\text{O}_{3}}}{D_{\text{B}}}}\left( {1 + \frac{C_{\text{B}}D_{\text{B}}}{bC_{\text{O}_{3}}^{\ast}D_{\text{O}_{3}}}} \right)\]
+
+\[N_{\text{O}_{3}} = E_{k}aC_{\text{O}_{3}}^{\ast}\]
+
+\[E = - \frac{H\alpha^{2}}{2(E_{i} - 1)} + \sqrt{\frac{H\alpha^{4}}{4(E_{i} - 1)^{2}} + \frac{E_{i}H\alpha^{2}}{(E_{i} - 1)} + 1}\]
+
+\[\frac{\text{d}C_{\text{B}}}{\text{dt}} = - bN_{\text{O}_{3}}\]
+
+\[C_{\text{B}} = C_{\text{B0}} - b\int_{0}^{x}N_{\text{O}_{3}}\text{ dt}\]
+
+\[obj(k_{2}) = \sum_{i = 1}^{n}\left( \frac{C_{\text{B,exp}}(t_{i})}{C_{\text{B0}}} - \frac{C_{\text{B,calc}}(k_{2},t_{i})}{C_{\text{B0}}} \right)^{2}\]
+
+where \(k\)2 is the second-order reaction rate constant; _C_B and _C_B0 are the concentrations of the dye at a time t, and 0, respectively; _k_L is the liquid-side mass transfer coefficient; \(t\) is the reaction stoichiometric coefficient for the dye; _D_B and _D_O are the diffusivities of the dye and ozone, respectively; _C_O_3 is the liquid equilibrium ozone concentration.
+
+The diffusivity of the dye (_D_B) was estimated from Wilke-Chang equation [45] taking into account of the Le bas method [46] for estimating the molar volume of the dye molecule. A value of _D_B = 3.24 x 10-10 m2/s was calculated, which falls within the range of diffusion coefficients obtained for azo-dyes by other researchers [39]. Ozone diffusivity (_D_O3 = 1.74 x 10-9 m2/s) was obtained from Beltran [47]. _C_O3 was calculated by applying Henry's law: _C_O3 (mol/L) = 55.56_PA_/_H_, where _PA_ is the ozone partial pressure in the gas phase in atm; \(H\) is the Henry's law constant given by Eq. (8) as function of pH and temperature [48]. It was assumed that _t_-butanol, which was added at relatively low concentrations in some experiments, did not affect significantly the equilibrium concentration [49].
+
+\[H\,\left( {\text{atm}/\text{mol}/\text{mol}} \right) = 3.84 \times 10^{7}(10^{\text{pH} - 14})^{0.035}\,\exp\left( {- \frac{2428}{T}} \right).\]
+
+## Materials and methods
+
+### Materials
+
+Reactive Orange RO16 dye (C.I. 177757) was purchased from Sigma-Aldrich, UK, and was used as received. The dye has a molecular mass of 617.53 g/mol and its UV/Vis spectrum is shown in Fig. 1 alongside its molecular structure. The dye shows a maximum absorbance at _l_max = 496 nm. Dye solutions at given concentrations were prepared by dissolving corresponding masses of the dye powder in 1 L volumetric flask of deionised water at room temperature to achieve concentrations in the range 25-500 mg/L. The dye was easy to dissolve in water and only a gentle mixing was sufficient to achieve its complete dissolution in water. In some experiments, _t_-butanol (Sigma-Aldrich, UK) was used as a radical scavenger to evaluate the significance of radical reactions to the overall reaction. Sodium hydroxide, phosphoric acid and sulphuric acid (Fisher, UK) were used to set the pH at the desired value (2, 7, 11) and all solutions were prepared with deionised water (Milli-Q system, resistivity >18 MO cm).
+
+### Experimental setup and analytical methods
+
+The experimental setup is shown in Fig. 2. A 500 mL semi-batch gas/liquid reactor was used to study the decolourisation of R016 dye. Ozone was produced from pure, dry oxygen (BOC, UK) by a laboratory air-cooled corona discharge ozone generator (model LAB2B, Ozonia Triogen, UK). The ozone gas concentration was easily set using a variable frequency control knob fitted to the generator at one of the desired values of 20, 40, 60 or 80 g/m3 normal temperature and pressure (NTP). The gas flow rate was fixed at 400 mL/min and the experiments were carried out at room temperature (20 +- 2 degC). A UV ozone analyser (BMT 963 VNT, BMT Mestschnik, Germany) was used to measure the ozone gas concentration. The analyser was interfaced with a P-C and the data were collected at 2 s intervals. The gas mixture (i.e. O2 and O3) was fed to the reactor through a three-ways valve after the generator has reached stability. The off-gas was diverted to an ozone destructor containing an alumina catalyst to remove any un-reacted ozone in the gas stream leaving the reactor. The dye concentration was measured using a diode array UV/Visible spectrophotometer (HP8453, Agilent, UK) from a preliminary calibration curve that was made at _l_max = 496 nm. The relationship between the absorbance at 496 nm and dye concentration was linear in the range of concentrations used, so the Beer-Lambert law was valid, and the determination factor was \(R\)2 = 0.9999. Samples were withdrawn continuously in a closed loop from the reactor by a peristaltic pump to the UV/Vis flow-through cell and returned back to the reactor (Fig. 2). Spectra were collected at 10 s intervals. The concentration of _t_-butanol, when used, was 0.2 M.
+
+Intermediates of the degradation reaction were detected by gas chromatography/mass spectrometry (GC/MS) (Thermofisher DSQ-II). The gas chromatograph was equipped with a capillary column (HP-5, cross linked 5% phenyl methyl siloxane, 30 m x 320 mm x 0.25 mm). The chromatographic conditions were those used in the literature [50]. The GC/MS samples were prepared by ozonating a solution of R016 dye at initial concentration of 500 mg/L and pH of 2.26 adjusted by sulphuric acid. The volume of each sample was 200 mL and extracted three times with 20 mL dichloromethane. The extracted phase was purged to dryness using pure air at room temperature and then adjusted to 2 mL with dichloromethane. GC analysis was carried out on samples collected at 0, 5, 15, 30, 45 and 60 min of reaction time.
+
+## Results and discussion
+
+### Ozone mass transfer parameters
+
+Mass transfer parameters are very important information in the study of gas/liquid reactors. Knowledge of the values of the mass transfer coefficient, _k_L, the specific interfacial area, \(a\), and the equilibrium concentration of ozone in water, _C_O3, is required to couple the mass transfer with chemical reaction so the observed reaction kinetics can be used to determine the intrinsic kinetic data. The mass transfer parameters were determined as discussed by Tizaoui et al. [51]. Values of _k_L, \(k_{a}\) and \(a\) developed in this study are shown in Table 1, which are in good agreement with literature [52]. Equilibrium liquid ozone concentration was calculated by Henry's law (Eq. (8)). Since the reaction between ozone and the dye is fast, it was assumed that the rate of ozone auto-decomposition was negligible.
+
+### Stoichiometric ratio
+
+The reaction stoichiometry was determined by injecting a solution of RO16 dye at a given concentration (1.15 x 10-3 mol/L) into an aqueous ozone solution at various known initial concentrations in the range (0.47-2.36) x 10-4 mol/L. With these molar concentrations, the number of moles of RO16 dye was sufficient to consume all ozone initially present and leave a measurable concentration of the dye. Fig. 3 shows the change of the number of moles of ozone reacted as function of the number of moles of dye removed. The slope of the line gives the stoichiometric ratio between ozone and RO16 dye reaction, which was taken as three moles of ozone per mole of dye. This value is in agreement with other studies on ozonation of dyes [29,53].
+
+### Effect of operating parameters on dye decolourisation
+
+#### Effect of pH and radical scavenger (_t_-butanol)
+
+It is well established that pH plays a significant role in ozone reactions [47]. High pH values lead to accelerated ozone decomposition accompanied by increased generation of hydroxyl radicals. Although hydroxyl radicals are highly reactive, results for the ozonation of dyes showed that the removal rates have not always been improved at high pH. For example data from Kusvuran et al. [29] show that an increase of pH from 3 to 10 results in reduction of the removal percentages from 90% to 50% after 4 min ozonation of Basic Yellow 28 dye. Hence, careful interpretation of results related to ozone oxidation of dyes at various pHs should be made. In this study, the effect of pH on the decolourisation of RO16 dye is shown in Fig. 4. In the absence of _t_-butanol, the curves show that a change of pH from 2 to 7 did not result in significant changes of the decolourisation rates of RO16 dye; indeed a removal percentage of about 86% was achieved after 3 min ozonation at both pHs. However, when the pH increased to 11, the decolourisation of the
+
+\begin{table}
+\begin{tabular}{c c c c} \hline
+[_t_-butanol] (mol/L) & \(k_{k}\) × 104 (ms−1) & _k__k_α × 102 (s−1) & _α_ (m−1) \\ \hline
+0 & 0.67 & 2.48 & 37 \\
+0.2 & 1.02 & 5.85 & 57 \\ \hline \end{tabular}
+\end{table}
+Table 1: Mass transfer parameters at pH 7.
+
+Figure 3: Determination of the stoichiometric ratio for ozone-RO16 dye solution reaction.
+
+Figure 2: Experimental setup.
+
+dye was faster than for the other two pHs and resulted in a removal percentage of 95% after 3 min ozonation. The increased rate at pH 11 may due to two effects (i) at such high pH, the production of hydroxyl radicals is significant which leads to faster removal of the dye, and (ii) at high pH, the H-N-R group of the dye molecule deprotonates, which makes the sites more reactive for the electrophilic attack by ozone. It is likely that at pH 7, the production of hydroxyl radicals as a result of ozone decomposition is significant. However, since no change in decolourisation rate occurred at pH 7 in comparison to pH 2, it is likely that the increase in the rate at pH 11 was due to increased ionisation of the dye. In addition, at high pH, the naphthalene ring of R016 dye molecule should exhibit high electron density [54], which renders it more susceptible to electrophilic attacks by molecular ozone and possibly hydroxyl radicals.
+
+Fig. 4 also shows the effect of the radical scavenger _t_-butanol (0.2 mol/L) at the different pHs. At pH 2, the addition of _t_-butanol did not affect the decolourisation rate of the dye as compared to the experiment without _t_-butanol. Effectively, at pH 2, the production of hydroxyl radicals from ozone decomposition is negligible because of the very low concentration of hydroxide ions (the initiator of ozone decomposition). As a result the removal of colour at pH 2 proceeds solely through molecular ozone reaction (i.e. direct mechanism). Maciejewska et al. [22] have also found that in acidic solutions the removal of Reactive Blue 81 proceeds solely through ozone direct mechanism. It is worth noting that the addition of _t_-butanol at pH 2 did not affect significantly the mass transfer since only less than 5% increase in the volumetric mass transfer coefficient _k__t_a was observed in this study at pH 2 with _t_-butanol. On the other hand, when pH values have increased to 7 and 11, it is clear from Fig. 4 that the addition of _t_-butanol increased the rate of R016 dye decolourisation, though the increase is more significant at pH 7. It was suggested that in the presence of _t_-butanol, less ozone
+
+Fig. 5: Effect of pH and _t_-butanol (th) on spectra changes as function of time C_in_ =25 mg/L and _C_O_2 = 20 g/m3 NTP (a) pH 2, no tb; (b) pH 7, no tb; (c) pH 11, no tb; (d) pH 2, with tb; (e) pH 7, with tb; (f) pH 11, with tb.
+
+is wasted on scavenging reactions and the utilisation of ozone for colour removal is therefore enhanced. This means that in the absence of _t_-butanol at high pH, ozone decomposition to hydroxyl radicals is significant and indirect hydroxyl-radical mediated reactions with non-coloured saturated bonds as well as organic and inorganic compounds in solution (i.e. products of the main reaction) is expected to be significant. Moreover the addition of _t_-butanol to the aqueous solution at pH 7 and 11 enhances the mass transfer rate as a result of increased volumetric mass transfer coefficient, _k__k__a_[51]. This results in enhanced decolourisation rates.
+
+Fig. 5 shows the changes of the UV/Vis spectra as function of time at different pH values and in the absence and presence of _t_-butanol. The peak observed in the visible region (i.e. 496 nm) is due to the orange colour of the chromophore. All spectra show that although the absorbance at 496 nm, which is attributable to the parent dye molecule (i.e. colour), disappeared totally after about 10 min oxonation, absorbances in the UV region, which are essentially due to the products of the main reaction, remained relatively significant. This clearly indicates that ozone was very effective in removing colour, possibly due to a rapid reaction with the azo group. However, the results indicate that the removal of colour did not imply complete degradation of the parent dye molecule. It is also clear from the results that reaction products follow the same trend as colour removal though at different degrees depending on the pH and the presence or absence of the radical scavenger _t_-butanol. Without _t_-butanol, the spectra profiles (Fig. 5 a-c) show significant decrease of the absorbance in the UV region as the pH increased from 2 to 11; for example the absorbance at 253 nm decreased after 7 min oxonation by 31, 47 and 82% at pHs 2, 7 and 11, respectively. This clearly proves that the products of the main reaction are degraded by hydroxyl radicals, which are produced in increasing quantities as the pH increased. On the other hand, in the presence of _t_-butanol (Fig. 5 d-f), the spectra profiles for all pHs are almost similar and the reduction of absorbance in the UV region is less pronounced when pH increased as that for experiments without _t_-butanol. This again confirms that hydroxyl radicals play an important role in oxidising the reaction products rather than the parent molecule.
+
+#### 4.3.2 Effect of inlet ozone gas concentration and initial dye concentration
+
+The inlet ozone gas concentration has a significant effect on the removal rate. It was found in the literature that ozone gas concentration can have positive, negative, or no effect on colour removal of dyes [55]. Similarly several studies have also shown that initial dye concentration plays a significant role on removal rates with ozone [30,56]. Hence careful consideration should be given to the effect of gas ozone and initial dye concentrations. In this study, inlet ozone concentrations (C03) of 20, 40, 60 and 80 g/m3 NTP were used to remove the dye at initial concentrations (C80) of 25, 50, 70 and 90 mg/L at pH 7 and no _t_-butanol. The time at which 90% of the initial dye concentration was removed (f90) was used to illustrate the effects of ozone gas concentration and initial dye concentration on colour removal rates. Lower f90 values indicate higher decolourisation rates, while higher f90 values indicate lower rates. Fig. 6a illustrates the changes of C8/C80 as function of time at various inlet ozone gas concentrations for an initial dye concentration of 25 mg/L; similar trends were also obtained for the other initial dye concentrations. The figure clearly shows that as the gas ozone concentration increased from 20 to 80 g/m3 NTP, the decolourisation rate has almost doubled. Table 2 presents f90 values for the various ozone and dye concentrations used in this study. It is evident that the decolourisation time f90 decreases with increasing ozone gas concentration and increases with increasing the initial dye concentration. Hence, the decolourisation times required at higher inlet ozone gas concentrations are smaller than those at lower inlet ozone concentrations. For
+
+\begin{table}
+\begin{tabular}{c c c c c}  & \multicolumn{3}{c}{initial dye concentration (mg/l)} \\  & **25** & **50** & **70** & **90** \\ \hline \(C_{\text{O}_{1}}\) (g/m3 NTP) & & & & \\
+**20** & 3.4 & 8.6 & 12.8 & 17.2 \\
+**40** & 2.5 & 4.6 & 7.3 & 9.4 \\
+**60** & 1.9 & 3.5 & 4.9 & 6.2 \\
+**80** & 1.5 & 2.3 & 3.6 & 5.1 \\ \hline \end{tabular}
+\end{table}
+Table 2: Time observed for 90% colour removal in minutes (pH 7, no _t_-butanol).
+
+Figure 6: Effect of inlet ozone gas and initial dye concentrations on colour removal ((a) Ca0 = 25 mg/L, pH 7, no _t_-butanol).
+
+example at an initial dye concentration of 25 mg/L, \(t\)90 decreased from 3.4 min to 1.5 min when ozone gas concentration increased from 20 to 80 g/m3 NTP. While for example at the inlet ozone gas concentration of 20 g/m3 NTP, \(t\)90 increased from 3.4 min to 17.2 min when the initial dye concentration increased from 25 to 90 mg/L. Higher ozone gas concentration implies higher equilibrium liquid concentration, which results in higher ozone absorption rate (Eq. (3)) that leads to higher decolourisation rates (Eq. (5)). Whereas for a given inlet ozone gas concentration, the decolourisation times required at higher initial dye concentrations are greater than those at lower initial dye concentrations. Clearly, the data show that as the inlet ozone gas concentration (_C_O3) increased, the removal rates became faster since \(t\)90 decreased as function of _C_O3 by over 70% as shown in Fig. 6b. However, such increase in rates was not proportional to the increase in the inlet ozone gas concentration and the rates tend to increase little when the inlet ozone concentration increased further (Fig. 6a and b). On the other hand, the colour removal rates tend to decrease linearly with the initial dye concentration (Fig. 6c). Based on the previous observations, an attempt was made to propose an empirical equation giving the relationship between \(t\)90 and (_C_O3, _C_B0). A power law with respect to the inlet ozone gas concentration was most suitable while a linear relationship with respect to the initial dye concentration was appropriate. Hence a two-parameter equation was proposed (Eq. 9). The two parameters \(a\) and \(n\) in Eq. (9) were determined by least-square regression analysis for all experimental results shown in Table 2. Eq. (9) forms a simple way to predict the performance of the ozonation system when different combinations of inlet ozone gas and initial dye concentrations are used, which is useful for design and control purposes. Similar equations, but with different parameter values, can also be established from experimental results if higher removals are required (e.g. for 95% removal) or different reactor systems are used.
+
+\[t_{90} = \alpha C_{{\text{O}}_{{\text{3}}}}^{- n}C_{{\text{B}}0}\]
+
+where \(a\) = 2.69 units and \(n\) = 0.89 with \(t\)90 in minutes, _C_O3 in g/m3 NTP, and _C_B0 in mg/L.
+
+Based on the results obtained in this study, ozone is feasible to decolourise RO16 dye and is practical to treat wastewaters from dyeing processes and finishing plants. Eventually, for a given initial dye concentration, the inlet ozone gas concentration can be chosen to satisfy a required process criteria (e.g. a given decolourisation time).
+
+### Profile of reaction products versus time
+
+UV/Vis spectra were used to illustrate the profile of products change over time. This was done assuming Beer-Lambert law is valid (dilute systems) for any wavelength, in particular for a wavelength of 388 nm (a maximum absorbance in the UV region as observed in Fig. 1). According to Beer-Lambert law, the measured absorbance at 388 nm (Absmeas,388) at any time \(t\) is the sum of the absorbance exhibited by the parent dye molecule (Absd4,388) and that by the reaction products (Absp3,388) (Eq. (10)). It should be noted that ozone, if there was any in solution, does not exhibit any absorbance at 388 nm.
+
+\[\text{Abs}_{\text{meas,388}} = \text{Abs}_{\text{d4,388}} + \text{Abs}_{\text{p},388}\]
+
+The absorbance resulting from the parent dye molecule at 388 nm (Absd4,388) was determined from the measured absorbance in the visible region at 496 nm. A preliminary study based on measuring the absorbance values at 388 nm and 496 nm for different dye concentrations gave a linear relationship of the form Absd4,388 = \(\alpha_{388/496} \times \text{Abs}_{\text{d4,496}}\); where \(\alpha_{388/496} = 0.6179\) with a determination coefficient \(R\)2 = 0.9995. With the assumption that absorbance at 496 nm (Absd4,496) during oxidation is only due to the colour of the parent dye molecule, the absorbance at 388 nm resulting from the products is then given by Eq. (11).
+
+\[\text{Abs}_{\text{p},388} = \text{Abs}_{\text{meas,388}} - \alpha_{388/496} \times \text{Abs}_{\text{d4,496}}\]
+
+The change of Absp3,388 as function of time for different pH values in the absence and presence of _t_-butanol is shown in Fig. 7. The figure shows that the absorbance at 388 nm due to reaction products increased rapidly (within approximately the first 2.5 min at pH 2 and 1 min at pH 11) up to a maximum after which it declined slowly. For experiments without the radical scavenger _t_-butanol, the maxima were observed at 2.2, 2.8 and 0.9 min for pHs of 2, 7 and 11, respectively, while in the presence of _t_-butanol, the maxima were obtained at almost similar times (except for pH 7) of 2.3, 1.7 and
+
+Fig. 7: Profiles of ozonation products at different pH values in absence and presence of _t_-butanol (_C_B0 = 25 mg/L and _C_O3 = 20 g/m3 NTP).
+
+0.7 min for the pHs 2, 7 and 11, respectively. These times correspond to times at which about 80% of the initial dye has been removed. The maxima were highest at pH 2 and lowest at pH 11 indicating that reaction products are substantially degraded at high pHs to simple molecules (e.g. acids, aldehydes), which do not absorb significantly in the UV region. It is clear from this result that hydroxyl radicals, which are produced at significant concentrations at high pH, are the major contributing factor in the degradation of the reaction products. Further ozonation after the maxima times leads to the degradation of the products, which explains the decrease of AbsP3,388 after these maxima. The profiles at pH 2 and 7 (Fig. 7) present tails at the end of the experiments indicating that the degradation of the products identified at 388 nm becomes very slow possibly due to other competitive reactions.
+
+### Kinetic study
+
+The determination of the second-order reaction rate constant, \(k\)2, was made at different pH values by fitting the experimental results to Eq. (7) using \(k\)2 as a parameter to change in Excel Solver. In order to fulfil the assumptions, the model was applied only for data
+
+\begin{table}
+\begin{tabular}{c c c c c c c} \hline Symbol & Name & Structure & Sample time (min) & & & \\ \cline{3-7}  & & 5 & 15 & 30 & 45 & 60 \\ \hline D1 & Nitrosobenzene & & & & & \\ D2 & Nitrobenzene & & & & & \\ D3 & Benzene-1,4-diol & HO & & & & \\ D4 & 1,4-Benzoquinone & & & & & \\ D5 & Acetamide & & & & & \\ D6 & Phthalic acid & & & & & \\ D7 & Maleic acid & & & & & \\ D8 & Oxalic acid & & & & & \\ D9 & Acetic acid & & & & & \\ D10 & Formic acid & & & & & \\ \end{tabular}
+\end{table}
+Table 3: Identified intermediates (_c_th0 = 500 mg/L \(C\)0 = 60 g/m3 NTP).
+
+Figure 8: Agreement between model and experimental results (dashed lines represent ±5% error).
+
+Figure 9: Variation of the decimal logarithm of \(k\)2 as function of pH.
+
+of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 8 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. The values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 8 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 8 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 8 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 8 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 9 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 9 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 9 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.4. Fig. 10 shows good agreement between the calculated and experimental data within an error of \(\pm\) 5%. Experiments which presented low Hatta numbers (i.e. \(Ha\)\(<\) 1) were not used for calculating \(k_{2}\). The effect of pH on the rate constant \(k_{2}\) is shown in Fig. 9. The figure shows that as the pH increased from 2 to 7, \(k_{2}\) increased only slightly from \(2.5\times 10^{5}\) to \(3.2\times 10^{5}\) L/mol s. However a further increase to pH 11 resulted in significant increase in \(k_{2}\) by over 5 times as compared to that at pH 2 (at pH 11, \(k_{2}\) = 1.4 \(\times\) 10\({}^{6}\) L/mol s). Regarding the fact that there is no available data about the rate constant for the reaction between ozone and R016 dye, the rate constants obtained in this study cannot be compared directly to other studies. However, rate constants for the reaction between ozone and other azo dyes have been reported to vary between \(3.7\times 10^{3}\) and \(7.1\times 10^{5}\) L/mol s [12, 57]. As can be seen, the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.4 are consistent with the values of \(C_{\rm B0}\)/\(C_{\rm B0}\)\(>\)0.
+
+obtained in this study are in the same range as those obtained in the literature. Clearly the values of \(k\)2 obtained as function of pH agree well with the observations made in Section 4.3.1.
+
+### Possible degradation mechanism
+
+It is well known that the electrophilic reactions of ozone take place on sites with strong electronic density [41,43,58]. In particular, aromatics substituted with electron-donor groups are highly reactive with ozone on carbons located in the ortho and para positions due to high electronic densities on these carbons. Whereas, aromatics substituted with electron-withdrawing groups (-COOH, -NO2) are weakly ozone reactive. The molecular structure of RO16 dye contains both electron-donating groups (-OH, -NHR) and electron-withdrawing groups (-SO3 and -SO2R') hence ozone reactions should occur initially on certain preferential sites. In order to illustrate this, molecular modelling was made with HyperChemTM 8.0.9 package. The molecular geometry was first optimised using the BIO Molecular Mechanics Force Field available in HyperChemTM 8.0.9 and the charge density at each atom was simulated using the semi-empirical CNDO method. Sites with strong electronic density were identified and the analysis showed that it is likely that ozone attack would take place at the azo group. Besides the electrophilic attack, ozone undergoes cycloaddition reactions with unsaturated bonds and leads to the formation of compounds having the carbonyl group (-C=O) or the nitroso group (-N=O). With this in mind and in order to illustrate a possible reaction mechanism, further analysis was made with GC/MS to identify the intermediate products. The GC/MS resulted in a number of compounds, which are shown in Table 3. Based on the molecular modelling and the GC/MS results, a possible degradation mechanism for RO16 dye with ozone was postulated and is shown in Fig. 10. In the very first step of degradation, ozone attack might take place on the azo (-N=N-) group leading to discolouration of the solution. The breakage of the azo-bond leads to the formation of nitro groups **S1** (2-(4-nitrosphenyl)sulfonylethyl hydrogen sulphate) and **S2** (6-acetamido-4-hydroxy-3-nitroso-naphthalene-2-sulfonic acid). Gutowska et al. [59] have also postulated that the oxidation of Reactive Orange 113 dye, which has similar structure to RO16 dye, proceeds first through attack of the azo group and leads to the formation of a nitro group as well. Through cleavage of the S-C bond in compound **S1** and with further ozone attack, **D1** (nitrosobenzene) and **D2** (nitrobenzene) are formed. Through further ozone attack, **D1** and **D2** are converted to **D3** (benzene-1.4-diol) and further to **D4** (1.4-benzouquinone). On the other hand, it was postulated that the bonds C-S and C-N in **S2** are cleaved leading to the formation of intermediates **S3** (N-(8-hydroxy-7-nitroso-2-naphthyl) and **S4** (2-nitrosaphthalene-1-ol) and the release of **D5** (acetamide). Further attacks of ozone leads to the conversion of **S3** and **S4** to **S5** (naphthalene-1,2,4-triol) then **S6** (2-hydrooxynaphthalene-1.4-dione) and eventually conversion to phthalic acid **D6**. It is believed that under electrophilic attack of ozone, a 1,3-dipolar cyclo-addition of ozone opens the aromatic rings of **D4** and **D6** leading to the formation of maleic (**D7**), oxalic (**D8**), acetic (**D9**) and formic (**D10**) acids [60,61]. Other studies on the ozonation of azo-dyes have also identified by GC/MS and/or LC/MS phenolic compounds, quinones, formic, oxalic and acetic acids [50,62,63]. Besides, oxidation and cleavage of the sulphonic groups from the aromatic rings lead to the release of sulphate during ozonation. Nitrogen from the dye's molecule may also be released either as a gas or as nitrate [61]. Extended oxidation of the intermediates leads to complete mineralisation to produce carbon dioxide and water [64]. It was observed in this study that pH values have dropped significantly to values as low as about 3.5 form initial pHs as high as 11, which also supports that acids were formed as a result of the dye oxidation. This significant drop of pH can be attributed to carboxylic acids but primarily to the formation of sulphuric acid that resulted from the rupture of HSO3 [65]. Moreover, conductivity measurements in this study have also showed significant increase by over 2.5 times of the initial value, which supports the formation of ionic species such as acids, sulphates and possibly nitrates during RO16 dye ozonation [66].
+
+## 5 Conclusions
+
+The decolourisation of the azo-dye Reactive Orange 16 (CL 177757) with ozone was performed at different conditions (pH of 2, 7, 11; initial dye concentration of 25, 50, 70 and 90 mg/L; ozone gas concentration of 20, 40, 60 and 80 g/m3 NTP; with and without radical scavenger _t_-butanol). At all conditions, ozone was found effective to remove the dye within very short periods of time (from 1.5 min to less than 20 min). The reaction was assumed of second order (1,1) and the rate constants at different pH values were determined using Danckwerts model. The second-order rate constant increased with pH and its values were \(2.5 \times 10^{5}\), \(3.2 \times 10^{5}\), and \(1.4 \times 10^{6}\) L/mol s for pH values of 2, 7, and 11, respectively. These values agree well with second-order rate constants reported for other azo dyes in the literature. It was suggested that the main mechanism for dye decolourisation is through molecular ozone and the high rate constants obtained at high pHs are possibly due to increased electronic density on the naphthalene ring and deprotonation of the H-N-R group of the dye molecule as the pH increases. On the other hand, although hydroxyl radicals may contributed to the decolourisation of the dye molecule at high pH, certainly they played a significant role in reacting with the products of the main reaction. A degradation pathway was proposed for the oxidation of RO16 dye with ozone and the reaction intermediates were identified. Results of this work emphasise the effectiveness of ozone technology for treating wastewaters from the textile industry.
+
+## References
+
+* (1) X. Lu, L. Liu, R. Liu, J. Chen, Textile wastewater reuse as an alternative water source for dyeing and finishing processes: a case study. Desalination 258 (2010) 229-232.
+* (2) Y. Aajnapuulu, N. Sreedhara Chary, D. Samuel Suman Raj, Decolourization of industrial effluents--available methods and emerging technologies--a review, Rev. Environ. Sci. Biotechnol. 4 (2005) 245-273.
+* (3) A.Y. Zahrin, C. Tzanou, N. Hillat, Coagulation with polymers for nanofiltration pre-treatment of highly concentrated dyes: a review, Desalination 266 (2011) 1-16.
+* (4) M. Muthukumar, D. Sarganamani, M. Senthilkumar, N. Selvakumar, Studies on decolouration, toxicity and the possibility for recycling of acid dye effluents using ozone treatment, Desy Fig. 64 (2005) 39-44.
+* (5) C. Allegere, P. Moulin, M. Maissen, F. Charbitt, Treatment and reuse of reactive dye effluents, J. Membr. Sci. 269 (2006) 15-34.
+* (6) M. Jeworski, E. Heinz, Combined chemical-biological treatment of wastewater containing refractory pollutants, Biotechnol. Annu. Rev. 6 (2000) 163-196.
+* (7) A. Lopez-Lopez, I.S. Pic, H. Deblelefontaine, Ozonation of azo dye in a semi-batch reactor: a determination of the molecular and radical contributions, Chemosphere 66 (2007) 2120-2126.
+* (8) O. Leitzke, Wastewater treatment with the combination of biology and ozonization, in: A. Vogelpohl (Ed.), Chausellar Umwett-Akademie, Oxidation of Water and Wastewater Coslar, Germany, 1996.
+* (9) X.J. Wang, S.L. Chen, X.Y. Gu, K.Y. Wang, Y.Z. Qian, Biological aerated filter treated textile washing wastewater for reuse after ozonation pre-treatment, Water Sci. Technol. 58 (2008) 919-923.
+* (10) S. Ledakowicz, M. Solecka, Influence of ozone and advanced oxidation processes on biological treatment of textile wastewater, Ozone: Sci. Eng. 23 (2001) 327-332.
+* (11) C. Gottschalk, J.A. Lipa, A. Saupe, Ozonation of Water and Waste Water, Wiley-VCH, Weinheim, 2000.
+* (12) A.C. Gomes, J.C. Nunes, R.M.S. Simoes, Determination of fast ozone oxidation rate for textile dyes by using a continuous quench-flow system,J. Hazard. Mater. 178 (2010) 57-65.
+* (13) M. Khadhaoui, H. Trabelsi, M. Ksibi, S. Bouguerra, E. Bleuch, Discoloration and detoxification of a ozone red dye solution by means of ozone treatment for a possible water reuse, J. Hazard. Mater. 161 (2009) 974-981.
+* (14) Y. Awaar, A. Batibay, Ozone application as an alternative method to the chemical treatment technique for textile wastewater, Fresenius Environ. Bull. 19 (2011) 2788-2794.
+
+* (15) C.A. Somensi, E.L. Simionato, S.L. Bertoli, A. Wisniewski Jr., C.M. Radetski, Use of ozone in a pilot-scale plant for textile wastewater pre-treatment: physicochemical efficiency, degradation by-products identification and environmental toxicity of treated wastewater, J. Hazard. Mater. 175 (2011) 235-240.
+* (16) F. Gahr, F. Hermentarz, W. Oppermann, Ozonation-an important technique to comply with new German laws for textile waste-water treatment, Water Sci. Technol. 30 (1994) 255-263.
+* (17) W. Jing, L. Liua, Z. Jiti, L. Hong, L. Guangfei, J. Ruofei, Y. Fenglin, Enhanced biodegradation of azo dyes by electropolymerization-immobilized redox mediator, J. Hazard. Mater. 168 (2009) 1098-1104.
+* (18) C. Tizouzi, N. Karoida, M. Abbruavska, Kinetic study of the manganese-based catalytic hydrogen peroxide oxidation of a persistent azo-dye, J. Chem. Technol. Biotechnol. 85 (2010) 234-242.
+* (19) C. McMullan, C. Meehan, A. Conneely, N. Kirby, T. Robinson, P. Nigam,I.M. Banat, R. Marchant, W.F. Simyth, Microbial decolololoration and degradation of textile dyes, Appl. Microbiol. Biotechnol. 56 (2001) 81-87.
+* (20) W.H. Gaze, J.W. Kang, D.H. Chapin, The chemistry of water-treatment processes involving ozone, hydrogen-peroxide and ultraviolet-radiation, Ozone: Sci Eng. 9 (1987) 335-352.
+* (21) B. Langlais, D.A. Reckhlow, D.R. Brink, Ozone in Water Treatment: Application and Engineering, Lewis Publishers, Boca Raton, 1991.
+* (22) R. Maciejewski, S. Lederowicz, L. Gebickea, I. Perkowski, Ogzonation kinetics of Reactive Blue 81 in aqueous solution, in: Proceedings of the International Specialized Symposium IOA, Toulouse, France, 2000, p. 75.
+* (23) P. Ball, C.M. Nicholls, Azo-hydranzone tautomersil of hydroxyazo compounds-a review, Dyes Bigm. 3 (1982) 5-26.
+* (24) A. Lopez, H. Benbelkacem, J.S. Pic, H. Debelkentzen, Oxidation pathways for ozonation of azo dyes in a semi-blatt reactor: a kinetic parameters approach, Environ. Technol. 25 (2004) 311-321.
+* (25) M.R. Detry, A. Friedman, Oxidation of tellunoprylium dyes with ozone, chlorine, and bromine-differing replacement and kinetic-behavior with respect to oxidations of oxygen-containing, sulfur-containing, and selenium-containing dyes, Oragonmetallics 13 (1994) 533-540.
+* (26) F. Munoz, C. von Sonntag, Determination of fast ozone reactions in aqueous solution by competition kinetics, J. Chem. Soc.-Perkin Trans. 2 (4) (2000) 661-664.
+* (27) A.K. Bin, M. Roustan, Mass transfer in ozone reactors, in: Proceedings of the International Specialised Symposium on IOA 2000, Fundamental and Engineering Concepts for Ozone Reactor Design, Toulouse, 2000, pp. 99-131.
+* (28) U. Filipkova, J. Rodziewicz, M. Krzemienieniewski, E. Duska, The removal of C.I. Reactive Black 8 and Reactive Orange 16 by ozonation, Pol. J. Nat. Sci. 23 (2008) 659-666.
+* (29) E. Kusuraru, O. Gulanza, A. Samil, M. Erbil, Detection of double bond-zone stoichiometry by an iodimetric method during ozonation processes, J. Hazard. Mater. 175 (2010) 410-416.
+* (30) Z. He, S. Song, M. Xia, J. Qiu, H. Ying, B.L. Vialang, J. Chen, Mineralization of C.I. Reactive Yellow 84 in aqueous solution by snoolytic ozonation, Chemosphere 69 (2007) 191-199.
+* (31) C.-Y. Chen, Photocatalytic degradation of azo dye reactive orange 16 by TIO2, Water, Air Soil Pollut. 202 (2009) 335-342.
+* (32) U. Pagga, D. Brown, The degradation of dyestuffs: Part II behaviour of dyestuffs in aerobic biodegradation tests, Chemosphere 15 (1986) 479-491.
+* (33) A. Spagni, S. Grilli, S. Casu, D. Mattioli, Treatment of a simulated textile wastewater containing the azo-dye reactive orange 16 in an anaerobic-biofilm mono-aerobic membrane bioreactor, Int. Biodetator. Biodegrad. 64 (2010) 676-681.
+* (34) M.A. Brown, S.C. Dewito, Predicting azo-dye toxicity, Crit. Rev. Environ. Sci. Technol. 23 (1993) 249-324.
+* (35) O. Levenspiel, Chemical Reaction Engineering, 3rd ed. John Wiley & Sons, New York, 1999.
+* (36) J.-C. Charpentier, Mass transfer rates in gas-liquid absorbers and reactors, in: Advances in Chemical Engineering, Academic Press, 1981, pp. 1-133.
+* (37) D.W. VanKweelen, P.J. Hofrijzer, Kinetics of gas-liquid reactions, I: General theory, Recl. Trav. Chim. Pay. B 67 (1948) 563-568.
+* (38) Danckwerts, Gas-Liquid Reactions, McGraw-Hill, New York, 1970.
+* (39) J. Navarro-Laboluts, L. Capahblanca, A. Abad, S.C. Cardona, F. Lopez, J.I. Torregrosa, Mathematical model for monitoring gas-liquid reactors by means of continuous flow analysis, Ozone: Sci. Eng. 28 (2006) 17-27.
+* (40) W.J. Decoursey, Absorption with chemical reaction: development of a new relation for the Danckwerts model, Chem. Eng. Sci. 29 (1974) 1867-1872.
+* (41) M.D. Gurol, S. Nekouininail, Kinetic-behavior of ozone in aqueous-solutions of substituted phenols, Ind. Eng. Chem., Fundam. 23 (1984) 54-60.
+* (42) F. Xiong, N.J. DeGraham, Rate constants for herbicide degradation by ozone, Ozone: Sci. Eng. 14 (1992) 283-301.
+* (43) J. Hoigne, H. Bader, Rate constants of reactions of ozone with organic and inorganic-compounds in water. I. Non-dissociating organic-compounds, Water Res. 17 (1983) 173-183.
+* (44) J. Hoigne, H. Bader, Rate constants of reactions of ozone with organic and inorganic compounds in water.II. Dissociating organic compounds, Water Res. 17 (1983) 185-194.
+* (45) C.R. Wilke, P.C. Chang, Correlation of diffusion coefficients in dilute solutions, J. Atch. 11 (1955) 264.
+* (46) C. Lekas, The Molecular Volumes of Liquid Chemical Compounds, Longmans, New York, 1915.
+* (47) F. Beltran, Ozone Reaction Kinetics for Water and Wastewater Systems, Lewis Publishers, Boca Raton Florida, 2004.
+* (48) L. Roth, D.E. Sullivan, Solubility of ozone in water, Ind. Eng. Chem., Fundam. 20 (1981) 137-140.
+* (49) Y. Ku, L.K. Wang, Decomposition of 2-chlorophenol in aqueous solutions by ozone and W/

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2013.11.057.mmd

@@ -0,0 +1,248 @@
+Removal of trace antibiotics from wastewater: A systematic study of nanofiltration combined with ozone-based advanced oxidation processes
+
+Pengxiao Liu, Hanmin Zhang, Yujie Feng, Fenglin Yang, Jianpeng Zhang
+
+# Abstract
+
+This work did a systematic investigation on removal of trace antibiotics from wastewater treatment plant (WWTP) effluent through nanofiltration (NF), and disposal of the NF concentrate by advanced oxidation processes (AOPs). Four antibiotics, namely, norfloxacin (NOR),floxacin (OPL), roithromycin (ROX) and azithromycin (AZI), which had high detection frequencies in effluents from WWTPs in Dalian (China), were selected as the target microplutants. High rejections of antibiotics (>98%) were obtained in all sets of NF experiments. UV254 photolysis, ozonation and UV/O3 process were employed to treat NF concentrate. Results demonstrated that UV254 photolysis was not effective in degrading the four antibiotics, while ozone-based processes exhibited high removal efficiencies in 30 min. A sy energetic effect between O3 and UV was observed in degradation of the selected antibiotics during UV/O3 treatment. Generation of hydroxyl radicals in the process was testified using electron paramagnetic resonance (EPR) spin trapping technology. In treatment of NF concentrate from real secondary effluent, UV/O3 process achieved excellent removal efficiencies of antibiotics (>87%), a partial removal of dissolved organic carbon (DOC) (40%), an increase of BOD3/COD ratio (4.6 times), and a reduction of acute toxicity (58%). The study revealed that nanofiltration could efficiently remove antibiotics from WWTP effluent, and meanwhile, UV/O3 process was able to further eliminate the antibiotics in the NF concentrate effectively.
+
+## 1 Introduction
+
+In recent years, pharmaceuticals and personal care products (PPCPs) have been identified as emerging contaminants threatening natural environment and human health [1,2]. Antibiotic is one important category of these trace organic contaminants [3]. Occurrence of antibiotics in the environment has been reported worldwide, such as in rivers of Europe [4,5], surface water of America [6,7], rivers of Australia [8] and seas and rivers of Asia [9,10]. Although detected environmental levels were generally low in waters (from ng L-1 to mg L-1), antibiotics were considered as "pseudo-persistent" contaminants due to their continual introduction into environment and permanent presence [8]. The presence of antibiotics in environmentally relevant levels has been associated with chronic toxicity to some non-target organisms and evolution of antibiotic-resistant bacteria [11]. The potential impact of antibiotic residues on ecosystem has received a growing international concern.
+
+Wastewater treatment plants (WWITPs) have been generally suggested to be a dominant source of antibiotics present in the environment [5,8]. A review article revealed that most of antibiotics in WWTP influent were only partially removed after the whole treatment processes in spite of varied biological treatment technologies [12]. Our previous research has also demonstrated that six WWITPs in Dalian (China) exhibited poor elimination for seven antibiotics (belonging to three groups) present in influents [13]. Meanwhile, the target antibiotics have been detected in various locations of coastal waters nearby the discharge outfalls of WWTP effluents. Therefore, diminishing the release of those trace contaminants from WWITS could be an effective approach to limit their occurrence in the aquatic environment.
+
+Additional treatment steps, downstream of conventional biological process, such as membrane processes [14,15], adsorptive treatment processes [16], advanced oxidation processes (AOPs) [17,18], and the combined ones [19,20], have been suggested and investigated to eliminate micropollutants in WWITPs. Among these technologies, membrane filtration processes, particularly reverse osmosis (RO) and nanofiltration (NF), showed promising application potential with merits of high product quality, low capital investment, easily being scaled-up, etc. [14]. Besides, effluent from RO or NF is an important water resource which can be reused in agriculture, urban areas, industry, aquifer recharge, etc. [21]. The implementation of wastewater reuse cannot only reduce the pollution to the effluent-receiving waters, but also mitigate the demand for natural freshwater sources.
+
+However, one main issue of RO or NF application in WWITPs is the disposal of concentrate which contains a wide range of organic pollutants including natural organic matter, refractory chemicals from public use like PPCPs, and soluble microbial products (SMPs) [19]. Although direct or indirect discharge of concentrate into the environment is not regulated currently in some areas, such behaviours would inevitably lead to pollution to ecosystem. Thus, increasing attention has been paid to this issue in recent years [22]. AOPs, such as, ozonation, Fenton process, and electrochemical oxidation, have been applied to treat RO concentrate [22]. Nevertheless, there is a lack of information about a systematic investigation of high-pressure membrane process as a tertiary step in WWITPs removing micropollutants with the membrane concentrate being disposed at the same time.
+
+In our study, four antibiotics, norfloxacin (NOR), ofloxacin (OFL), roxithromycin (ROX) and azithromycin (AZI), which have been detected with high frequencies in effluents from six WWITPs in Dalian in our previous work [13], were selected as the target micropollutants. NF process, which requires lower pressure and has a higher permeate flux than RO process, was applied to remove the target micropollutants with both model solutions and real secondary effluent as feed. The NF concentrate was treated using UV254 photolysis, ozonation and UV/O3 process, respectively. Dissolved organic carbon (DOC), UV254 absorbance, biodegradability, and ecotoxicity were evaluated for the raw water and treated samples. The objective of the study is to make an initial attempt to achieve zero discharge of micropollutants from WWITPs by employing NF process as a tertiary polishing step with the NF concentrate being disposed by an applicable AOP at the same time.
+
+## 2 Materials and methods
+
+### Reagents and chemicals
+
+Antibiotic standards of OFL and ROX were purchased from Dr. Ehrenstorfer (Augsburg, Germany), and NOR and AZI were obtained from National Institutes for Food and Drug Control (NIFDC, Beijing, China). The purity of the standards is equal to or higher than 95%. The physico-chemical properties of the analytes are summarized in Table S-1 in the Supplementary materials, Chemical structures are shown in Fig. 1. Individual stock solutions of each analyte were prepared at a concentration of 200 mg L-1 in water with 0.2% (in volume) 1.0 mol L-1 HCl to increase their solubility. All stock solutions were stored in the dark at -15 degC and new stock solutions were prepared monthly. Feed solutions for NF and AOPs experiments were prepared by spiking a certain amount of stock solutions into ultrapure water. 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) (purity higher than 97%) was purchased from Dojindo (Kumamoto, Japan). All other chemicals used in the study were of analytical grade. Ultrapure water was produced by a Milli-Q water purification system (Millipore, USA).
+
+### Bench-scale nanofiltration experiments
+
+#### 2.2.1 Nanofiltration system
+
+One commercial NF membrane, NFX (Synder Filtration, Vacaville, CA, USA), was used in our study. The nominal molecular weight cut-off (MWCO) of NFX is in the range of 150-300 Da. More information about the properties of NFX membrane is detailed in Table S-2 in the Supplementary materials. A laboratory-scale
+
+Fig. 1: Chemical structures of the selected antibiotics and the attacked sites by ozone or hydroxyl radical.
+
+cross-flow filtration unit was used for all the NF experiments. The layout of the unit is shown in Fig. S-1 and described in the Supplementary materials. The effective membrane area of the flat membrane cell is 75 cm2. The temperature of the feed solution was kept constant at 20 +- 0.5 degC for all the experiments.
+
+#### 2.2.2 Experimental protocols
+
+Prior to use, membranes were left to soak in pure water for 24 h. Soaked membranes were then compacted in NF system using pure water at the pressure of 1 MPa for 2 h. A new membrane was used for each experiment. Each experiment was proceeded for 8 h with an initial feed volume of 4 L and a cross-flow velocity of 0.35 m s-1. The rejection of antibiotics during the filtration is defined as \(R = (1 - C_{p}|C_{t})\) 100%, where \(C_{p}\) and \(C_{r}\) are concentrations of antibiotics in permeate and concentrate, respectively.
+
+Two sets of experiments were conducted with model solutions containing different components (with or without background organics) (shown in Table 1). Each antibiotic was spiked to make an initial concentration of 200 mg L-1 in the feed solutions. Sodium alginate was selected as background organics to represent extracellular polymeric substances (EPS) in secondary effluent. At regular time intervals, 2 mL samples were withdrawn from both concentrate and permeate at the same time, and then kept at -15 degC until analysis.
+
+One set of experiments was performed with real secondary effluent from one WWTP, in which a cyclic activated sludge treatment technology was employed. The WWTP served 0.2 million inhabitants with the capacity of 60,000 m3 d-1. Prior to experiments, the effluent was filtered by Whatman filter paper No. 1 to remove suspended matter. The characteristic of the effluent was listed in Table 2. The effluent was directly used as the NF feed solution without any antibiotics spiked, and the applied pressure was 0.4 MPa.
+
+### NF concentrate treatment experiments using AOPs
+
+#### 2.3.1 Experimental set-up
+
+UV254 photolysis, ozonation and UV/O3 experiments were all carried out in a 5-L cylindrical acrylic reactor (450 mm height, 120 mm diameter). A 15 W low-pressure mercury lamp which emits UV light mainly at 254 nm was introduced into the centre of the reactor and kept separated from the feed water using a quartz cooling jacket. The temperature of the feed solution was maintained at 20 degC by the circulating water. Ozone gas was fed into the reactor through a microporous titanic plate situated at the bottom of the reactor. The flow rate of 1 L min-1 and ozone concentration of 4 mg L-1 were applied in our study, which was based on previous literatures [23,24]. The ozone generator (CF-G-3-10G, Qingdao Guolin Industry Co., China) was used to produce ozone from pure oxygen (99.5%).
+
+#### 2.3.2 Experiments with model solution
+
+Model solution was prepared according to the characteristics of NF concentrate produced at the water recovery of 75% during NF experiment. It contained 20 mmol L-1 NaCl, 2.5 mmol L-1 NaHCO3, 1.5 mmol L-1 CaCl2, 40 mg L-1 alginate and the four antibiotics spiked at 600 mg L-1 for each analyte with pH of 7.9. Feed volume was 4 L for each experiment. UV254 photolysis, ozonation and UV/O3 experiments were conducted individually with the UV lamp or ozone generator turning on or off. All the experiments proceeded for 30 min. At regular time intervals, 20 mL samples were withdrawn from the reactor, then 200 mL sodium nitrite solution (1 mol L-1) was immediately added into the samples to quench the residual ozone. Samples were then kept at -15 degC until analysis.
+
+#### 2.3.3 Experiments with NF concentrate from real secondary effluent
+
+Total 8 L of NF concentrate was obtained from nearly 32 L real secondary effluent. Each 4 L concentrate was used for ozonation or UV/O3 experiment without any antibiotics spiked. Both experiments proceeded for 30 min. At regular time intervals, 20 mL samples were withdrawn from the reactor. Samples were taken and shaken in centrifuge tubes, then placed at room temperature for 2 h to remove residual ozone, which was according to the method reported in previous study [25]. Sodium nitrite was not added as it could interfere the measurement of UV254 absorbance or UV absorption spectra.
+
+### Analytical methods
+
+High performance liquid chromatography coupled with tandem mass spectrometry (HPLC-MS/MS) was applied to analyze the target antibiotics. The detailed procedure was described in our previous study [13]. The method of quantification of antibiotic concentrations is described in the Supplementary materials.
+
+UV/Vis spectrophotometers, model 752 (Shanghai jinghua Instrument Ltd., China) and UV-1700 (Shimadzu, Japan), were used to analyze the absorbance at 254 nm (UV254) and absorption spectra, respectively. The value of specific UV absorbance (SIVA) was calculated using the equation: SUVA = 100 (UV254/DOC). DOC was measured using a total organic carbon analyzer (TOC-VCPH, Shimadzu, Japan). COD and BODs were measured according to the Chinese SEPA standard methods [26]. The ozone concentration in the process gas from the generator was determined by KI method [27]. Scanning electron microscopy (SEM, Quanta 450, FEI, USA) was applied to analyze the surface of virgin and fouling mem
+
+\begin{table}
+\begin{tabular}{c c c c c c} \hline \hline  & First set of experiments & Second set of experiments \\ \hline Feed components & 10 mmol L−1 NaCl, 1 mmol L−1 NaHCO3, 0.5 mmol L−1 & 10 mmol L−1 NaCl, 1 mmol L−1 NaHCO3, 0.5 mmol L−1 CaCl2, 15 mg L−1 \\ pH & 7.0 & alginate, 200 mg L−1 for each antibiotic spiked & 7.0 \\ Applied pressure & 0.2 MPa and 0.4 MPa & 0.2 MPa, 0.4 MPa and 0.6 MPa \\ \hline \hline \end{tabular}
+\end{table}
+Table 1: NF experiments of model feed solutions.
+
+\begin{table}
+\begin{tabular}{c c c c c c c} \hline \hline  & pH & Conductivity (μ5 cm−1) & DOC (mg L−1) & NO3−N (mg L−1) & NH4+-N (mg L−1) & UV254 (cm−1) \\ \hline Secondary effluent & 7.2 & 970 & 5.6 & 10 & 5.6 & 0.109 \\ NF concentrate & 8.2 & 1810 & 19.2 & 18.4 & 9.8 & 0.347 \\ NF permeate & 7.1 & 610 & 0 & 5.3 & 2.1 & 0 \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Characteristics of the real secondary effluent, NF concentrate and NF permeate.
+
+branes. Membrane coupon extraction procedure is described in the Supplementary materials.
+
+The electron paramagnetic resonance (EPR) spin trapping experiments were carried out using a Electron-Spin Resonance Spectrometer (Bruker A200, Bruker Instrument, Germany). The instrumental settings for EPR spin trapping experiments were as follows: microwave frequency, 9.445 GHz; microwave power, 20.40 mW; modulation amplitude 0.1 mT; center field, 336.00 mT; sweep width, 10.00 mT; sweep time, 5.2 s. DMPO was used as spin-trapping agent.
+
+The luminescent marine bacteria _Photobacterium phosphoreum_ (T3 mutation) were used to evaluate the acute ecotoxicity of water samples. The bioluminescence intensities of _P. phosphoreum_ were measured by a LuminMax-C Luminometer (Maxwell Sensors Inc., USA). The tests were carried out according to the national standard method of China [28].
+
+## Results and discussion
+
+### Removal of the selected antibiotics by NF
+
+Removal efficiencies of the four antibiotics through NF were investigated with different feed solutions at varied operation pressures. Fig. 2 shows the removal behaviour at the pressure of 0.2 M\(\mathrm{\SIUnitSymbolOhm}\) in the first set of experiments, in which background organics were not added. Rejections of the four antibiotics were above 98%. Higher rejections were obtained at 0.4 MPa (>99%). The results indicated that NFX membrane was efficient in removing the selected antibiotics. As molecular weights of the four antibiotics are higher than the MWCO of NFX membrane, steric exclusion is bound to play a predominant role in the rejection.
+
+It was notable that antibiotic concentrations in feed increased overall due to the continuously decreasing feed volume over time. However, a decreasing trend was observed at early stage. It was believed that this phenomenon was not derived from the penetration of antibiotics through the membrane since antibiotics in permeate were detected at very low concentrations. Adsorption of antibiotics onto membrane was speculated to be the reason as fluoroquinolones and macrolides exhibit high sorption property due to their high solid-water partition coefficients (_K_d) [13].
+
+Results of membrane extraction experiments confirmed the adsorption of antibiotics onto membrane. Adsorption amounts of AZI or ROX were higher than those of NOR or OFL (see Table 3). This could be attributed to different properties of the antibiotics and the specific characteristic of NF membrane. Under the experimental conditions, NOR and OFL were present in neutral forms, while AZI and ROX were positively charged, according to their pKa values. Polyamine membrane was usually negatively charged in solution at pH 7 [14; 29]. Therefore, electrostatic interaction between the positively charged species and the negatively charged membrane could enhance the adsorption of AZI and ROX onto membrane. An overall mass balance calculation was carried out to estimate the adsorption amounts of antibiotics onto membrane as well (see Table 3). Nevertheless, the calculated results did not agree well with the results of direct extraction measurement. The reason could be the inadequate desorption of the antibiotics from membrane in extraction experiments, or their adsorption onto other components of
+
+Fig. 2: Concentrations of the four antibiotics in NF concentrate and permeate during NF experiment. Experimental conditions: model solutions containing background inorganic solute and antibiotics spiked at 200 μg L−1 for each one; operation pressure of 0.2 MPa; cross-flow velocity of 0.35 m s−1; temperature of 20 °C.
+
+the NF system. The same situation was reported in previous study [30].
+
+In the second set of experiments, alginate was added in feed to evaluate the influence of background organics on antibiotic removal and membrane flux. Rejections of antibiotics were over 99%. Lower adsorption amounts of them onto membrane were obtained compared to those in the first set of experiments, which was probably because the adsorption sites on the membrane were pre-occupied by alginate.
+
+### Permeate flux decline and membrane fouling
+
+Permeate flux decline was investigated in each experiment. In the first set of experiments, the flux did not show obvious decline. A flux reduction of 3% and a water recovery of 30% were obtained at 0.2 MPa after 8 h, while they were 14% and 72% at 0.4 MPa, respectively.
+
+In the second set of experiments, permeate flux decline was more severe than that in the first set. Fig. 3 shows the flux decline at varied pressures. Filtration at 0.2 MPa showed the lowest flux reduction (12%) after 8 h. At 0.4 MPa, the permeate flux dropped by 31% after 8 h, and a drop of 37% was obtained at 0.6 MPa after 7 h. The severe flux decline was supposed to arise from the membrane fouling caused by background organics in feed. Organic fouling exhibits complex interactions between chemical functional groups of organic foulants and those of the polymeric membrane skin layer [29]. It could lead to serious flux decline due to the cake layer formed on the surface of the membrane [29,31]. As shown in Fig. 4(c and d), an alginate fouling layer on the membrane could be observed through the SEM images. Water recoveries were 28% and 58% at 0.2 and 0.4 MPa within 8 h respectively, and 77% was obtained at 0.6 MPa within 7 h.
+
+### Nanofiltration of real secondary effluent
+
+Concentrations of the target antibiotics in secondary effluent were detected in the range of 221-372 ng L-1 (see Table 4). After
+
+\begin{table}
+\begin{tabular}{c c c c c} \hline \hline Antibiotic & Concentration in different samples & (ng L−1) & \\ \cline{2-5}  & Secondary effluent & NF concentrate & Ozonation treated samples & UV/O\({}_{\text{t}}\) treated samples \\ \hline NOR & 221 & 715 & 87 & 41 \\ ORL & 253 & 863 & 32 & – \\ AZI & 296 & 1060 & – & – \\ ROX & 372 & 1402 & – & – \\ \hline \hline \end{tabular} _Note:_ − − − represents not available, not been detected or concentrations below LOQs.
+
+\end{table}
+Table 4: Antibiotic concentrations detected in secondary effluent, NF concentrate and the treated samples.
+
+\begin{table}
+\begin{tabular}{c c c c c} \hline \hline Different approaches & \multicolumn{4}{c}{Overall adsorption amounts (μβ)} \\ \cline{2-5}  & NOR & OFL & AZI & ROX \\ \hline Direct extraction measurement & 36 & 18 & 63 & 42 \\ Mass balance calculation & 102 & 83 & 192 & 107 \\ \hline \hline \end{tabular}
+\end{table}
+Table 3: Adsorption amounts of antibiotics onto NF membrane.
+
+Figure 4: SEM images of virgin and fouling NF membranes, (a) and (b) are images of virgin membrane; (c) and (d) are images of fouling membrane after NF with model solution containing alginate as feed at 0.4 MPa; (e) and (f) are images of fouling membrane after NF with real secondary effluent as feed at 0.4 MPa.
+
+Figure 3: Permeate flux decline during NF at varied pressure. Experimental conditions: model solutions containing background inorganic solute, alginate, and antibiotics spiked at 200 μg L−1 for each one; real secondary effluent; cross-flow velocity of 0.35 m s−1; temperature of 20 °C.
+
+filtration for 8 h, NOR and OFL were detected with concentrations below the limit of quantification (LOQ) in the permeate, while AZI and ROX were not found. Characteristics of the NF permeate and concentrate are shown in Table 2. The flux decline over time was observed, but milder than that using model solution (see Fig. 3). This might be due to the fast formation of an alginate fouling layer on the membrane surface in model solution experiments. Besides, it was believed that the presence of Ca2+ could cause an increase in organic fouling as Ca2+ could bridge carboxylate functional groups on neighbouring alginate molecules, and make the fouling layer denser [29]. A flux reduction of 27% was obtained at the end of the experiment. SEM analysis confirmed the formation of fouling layer during the filtration (see Fig. 4(e and f)).
+
+### Treatments of model NF concentrate using AOPs
+
+UV254 photolysis, ozonation and UV/O3 process were used to treat model NF concentrate. Results are shown in Fig. 5. In addition, generation of reactive oxygen species (ROS) during the three treatment processes was investigated using EPR spin trapping technique with DMPO as spin-trapping agent. The spectra are shown in Fig. 6.
+
+#### 3.4.1 _Uv254 photolysis_
+
+In UV254 photolysis experiments, removal of NOR was comparatively effective after irradiation for 30 min (removal efficiency of 49%). In contrast, UV254 irradiation seemed not efficient in degrading the other three antibiotics.
+
+Degradation of organic pollutants by direct UV photolysis has been reported in previous literatures [17,32]. The mechanism is considered to be that organic substrates absorb UV light to form radicals which then react with oxygen, or form electronically excited states, and then an electron from the excited state of the substrates transfers to ground state molecular oxygen [33]. Besides, UV light with shorter wavelengths (\(\lambda < 190\) nm) could photolyze
+
+Fig. 5: Removal profile of antibiotics in model NF concentrate by three AOPs. Experimental conditions: model solutions containing background inorganic solute, alginate, and antibiotics spiked at 600 μg L−1 for each one; ozone concentration and flow rate were 4 mg L−1 and 1 L min−1, respectively.
+
+Fig. 6: Spectra of electron paramagnetic resonance (EPR) spin trapping tests for the three treatment processes. (a) UV254 photolysis; (b) ozonation; (c) UV/O3 process.
+
+H2O into *OH and *H [34], which was not the case in our study employing a UV lamp emitting light at 254 nm. This was supported by the result of EPR test of UV254 process. No signal of ROS was observed in the spectrum (see Fig. 6). Under a certain wavelength and intensity of UV irradiation, the degradation of organic substrate by UV photolysis largely depends on the chemical structure of the target compound [32]. Kim et al. found that sulphonamides such as sulfamethoxazole and sulfamonomethoxine, and quinolones such as norfloxacin and malidixic acid were effectively removed by UV photolysis (removal efficiencies of 86-100%) [35]. In contrast, macrolides such as clarithromycin, erythromycin and azithromycin were removed by 4-7% only. These results were well in accordance with our findings. Among the four antibiotics in our study, only NOR contains a N-H bond, which is believed to be easily broken by UV radiation [32]. This could be the reason that it exhibited higher sensitivity to UV254 photolysis than the other three.
+
+#### 3.4.2 Ozonation
+
+Compared to UV254 photolysis, ozonation was much more efficient in degrading the four antibiotics (see Fig. 5). OFL, AZI, and ROS were degraded with removal efficiencies above 99% after 10 min, while the same removal efficiency for NOR was obtained after 20 min. The ozonation rate of NOR was slower than those of the other three antibiotics, especially in the first 5 min. The reason should be that the diversity of structures of the four antibiotics resulted in different ozone affinity.
+
+During ozonation, oxidation of pollutants could occur through ozone or *OH, which was dependant on the ratio of ozone and OH concentration, the corresponding kinetics, the background of the treated water, etc. [36,37]. However, there was no significant characteristic signal of *OH could be observed in the EPR spectrum of ozonation process. Thus, ozone was believed to be the main reactive species during ozonation of antibiotics in our experiments. Ozone is a selective oxidant and attacks certain functional groups such as C=C double bonds, activated aromatic systems and non-protonated amines, which are electron-rich functional groups [36]. It has been suggested that N(4) (see Fig. 1) is the main site that O3 attacks during ozonation of fluoroquinolones [37,38]. However, compared to secondary amine, tertiary amine, in which a methyl group results in a higher electron density at N(4) site, shows a higher ozone affinity and a faster ozonation rate [36,39]. Among the four antibiotics, NOR contains a secondary amine, while the other three have tertiary amines, which is the reason for a slower ozonation rate of NOR than those of the other three.
+
+In addition, fluoroquinolones' reactivity toward O3 is strongly dependent on pH, which is intrinsically governed by deprotonation of N(4) amine [37]. The difference in pKa2 values makes NOR and OFL exhibit different ionization behaviours at a certain pH, and consequently leads to different protonated states of the amine groups. The pKa2 values of NOR and OFL are 8.85 and 7.65, respectively [40]. Therefore, during the ozonation of NF concentrate at pH around 7.9, NOR exhibited neutral or zwitterionic forms, whereas OFL exhibited anionic forms that resulted in a deprotonated amine at N(4) site. Thus, it led to a faster reaction rate for OFL than that for NOR. In fact, it was demonstrated by Marquez et al. that the increasing pH of the solution led to an increasing reactivity of molecular ozone towards OFL [41]. The second-order rate constants (_k_O3-OFL) of OFL-ozone reaction were \(2.0\times 10^{6}\) and \(1.2\times 10^{7}\) M\({}^{-1}\) s\({}^{-1}\) at pH 7 and 8, respectively.
+
+Both AZI and ROS were effectively degraded during ozonation. It was in accordance with the previous report that AZI and ROS were fast-reacting substrates towards O3 with apparent second-order rate constants (_k_O3, app) of \(5.2\times 10^{6}\) and \(3.1\times 10^{5}\) M\({}^{-1}\) s\({}^{-1}\) at pH 7.7 [37].
+
+#### 3.4.3 UV/O3 treatment
+
+As show in Fig. 5, UV/O3 treatment showed the highest degradation efficiencies for all the four antibiotics. Removal efficiencies were in the range of 85-99% after 5 min, while they achieved over 99% after 10 min. Reaction rates during UV/O3 process were faster than those during ozonation, especially in the first 10 min. The trend was extremely obvious for the degradation of NOR.
+
+UV/O3 treatment may involve three processes: UV photolysis, direct ozone oxidation, and indirect ozone oxidation. In our study, UV photolysis and ozonation were investigated individually. Results demonstrated that UV254 photolysis was only effective in degrading NOR among the four antibiotics, while ozonation was highly effective in degrading OFL, AZI and ROS, but less effective in degrading NOR. During UV/O3 treatment, a synergetic effect between O3 and UV irradiation, not merely an addictive effect, was observed in degradation of the selected antibiotics, which implied that indirect ozone oxidation was involved in the reaction. This was testified by the results of the EPR spin trapping experiments. In the EPR spectrum of UV/O3 process, a remarkable four-line signal was monitored with hyperfine splitting constants \(a^{N}\) = \(a^{N}\) = 1.488 mT and \(g\) = 2.0032, which are representatives of DMPO-OH adducts [42]. The result indicated that hydroxyl radicals were produced as reactive species during the treatment of antibiotics by UV/O3 process.
+
+Known as a strong oxidizing agent, hydroxyl radicals could react very fast with a large number of moieties. Some ozone-refractory pharmaceuticals could be degraded by *OH during ozonation of municipal wastewater effluents [43]. It was reported that the quinolone ring of fluoroquinolones could only be oxidized by *OH during ozonation process [38]. In the research of ozonation mechanism of ROX, Radjenovic et al. found that several abundant ozonation products were generated due to the *OH attack on some moieties, such as the cleavage of the dadinoase moiety and the attack on C-atom of the =N-O-CH2-O (see Fig. 1) [44]. In addition, the second-order rate constants (_k_OH, app) of the reaction of OH with AZI and ROS were \(2.9\times 10^{9}\) and \(5.4\times 10^{9}\) M\({}^{-1}\) s\({}^{-1}\) at pH 7 [37], which were much higher than those of ozone with them (shown previously).
+
+To sum up, UV/O3 process showed the best effectiveness in removing antibiotics in NF concentrate. UV irradiation combined with ozone enhanced the effective utilization of ozone. To further explore the feasibility and efficiency of ozone-based processes, NF concentrate of real secondary effluent was used as feed in AOPs experiments.
+
+### Treatments of NF concentrate from real secondary effluent using AOPs
+
+#### 3.5.1 Removals of antibiotics and DOC
+
+Ozonation and UV/O3 process were selected to treat NF concentrate from secondary effluent. First of all, the removals of the four target antibiotics were investigated. Concentrations of the antibiotics in NF concentrate and treated samples are shown in Table 4. After ozonation for 30 min, AZI and ROS were not detected, while removal efficiencies of NOR and OFL were 87% and 96%, respectively. For UV/O3 treatment, NOR was the only one found in the treated sample with removal efficiency of 94%.
+
+DOC reduction of 28% and 40% were obtained after ozonation and UV/O3 treatment, respectively. UV254 of the samples were also measured, and removal efficiencies of 36% and 51% were achieved for ozonation and UV/O3 processes, respectively. The relatively high UV254 removals suggested that O3-based processes were effective in degrading substrates with C=C, C=O bonds and aromatic ring present in NF concentrate.
+
+#### UV absorbance spectra
+
+The spectrophotometric spectra (200-400 nm) of the treated samples at different treatment time are shown in Fig. 7. For all samples, an absorbance peak appeared around 215 nm, and the absorbance decreased monotonically in the range of 215-400 nm. The peak was supposed to be the absorbance of NO3; which was detected with high concentration in the NF concentrate (see Table 2). This was testified by measuring the spectra of NaNO3 solution. With the increase of treatment time, both treatment processes decreased the absorbance of samples in the 230-300 nm range, indicating partial oxidation of chromophoric moieties of organics. Liu et al. suggested that the decrease of UV absorbance around 250 nm was due to the transformation of organics into products with low aromaticity and low molecular weight after oxidation [45]. In addition, it could be observed that the absorbance fall in 230-300 nm was more significant during UV/O3 treatment than that during ozonation after 30 min, which indicated that the former process was highly efficient in cleaving the moieties affording strong UV absorbance.
+
+#### Biodegradation and ecotoxicity assessment
+
+Fig. 8 shows the SUVA and biodegradability index (BODs/COD ratios) of the raw NF concentrate and samples treated by ozonation or UV/O3 process. SUVA was known as a useful parameter for estimating the dissolved aromatic carbon content in aquatic systems [46]. Both ozonation and UV/O3 treatment decreased the SUVA value of the raw NF concentrate with reduction of 12% and 19% respectively, showing that both processes were capable of destroying the aromatic structure of organics in NF concentrate. This could be attributed to electrophilic substitution reactions of aromatic compounds with electrophilic species, such as ozone and OH [36].
+
+The raw NF concentrate showed a low BODs/COD ratio of 0.05, which implied that the organic contents were refractory and difficult to be further biodegraded. After the treatment by ozonation and UV/O3 process, BODs/COD ratios increased 3.2 and 4.6 times, respectively. The results indicated that biodegradability of NF concentrate was improved after treatment, nevertheless, not satisfactory compared to that reported in previous research [19], which could be due to the difference of water quality, ozone dose, treatment time, etc.
+
+Acute ecotoxicity of raw water and treated samples was tested through evaluating the inhibition of bioluminescence from _P. phosphoreum_. Results are shown in Fig. 9. For the raw solution spiked with antibiotics, an inhibition ratio of 21% was detected. Both ozonation and UV/O3 process treated samples showed a reduction of toxicity. The remaining toxicity of the treated samples must derive from the degradation products since high removal efficiencies
+
+Figure 8: SUVA and BODs/COD ratio of raw NF concentrate, ozonation treated sample and UV/O3 treated sample.
+
+Figure 7: UV absorbance spectra change of NF concentrate from secondary effluent during ozonation (a) and UV/O3 process (b). (The inset in (b); comparison of spectra for the two processes.)
+
+Figure 9: Biduminescence inhibition ratio of model solution, NF concentrate and treated samples by ozonation and UV/O3 process. (A: model solution and treated samples, four antibiotics were spiked initially at 1 mg L−1 for each one; B: NF concentrate from secondary effluent and treated samples.)of antibiotics were achieved by both ozonation (above 99%) and UV/O3 treatment (above 99%) after 30 min. The ecotoxicity of NF concentrate from secondary effluent was comparatively prominent (inhibition ratio of 41%). UV/O3 treatment achieved the highest reduction of toxicity of the raw water with inhibition ratio dropped by 58%.
+
+## 4 Conclusions
+
+In summary, this work demonstrated that nanofiltration was an efficient approach to remove antibiotics in WWTP effluent, and the residual antibiotics in NF concentrate could be effectively eliminated by ozone-based AOPs. A synergetic effect between O3 and UV irradiation was observed in degrading antibiotics during UV/O3 treatment. The treated NF concentrate by UV/O3 process showed excellent removals of antibiotics (>87%), a partial reduction of DOC (40%), an increase of biodegradability (4.6 times), and a reduction of ecotoxicity (58%). Therefore, the treated NF concentrate is prone to be easily purified by activated sludge in case it is recirculated back to the biological treatment unit, which is easy-implemented in WWTPs. Thus, zero discharge of micropollutants from WWTPs is possible through the proposed scheme in the study. However, further research, such as optimization of experimental conditions and tests for more micropollutants, is still needed.
+
+## Acknowledgments
+
+The research is financially supported by the Fundamental Research Funds for the Central Universities and the National Natural Science Foundation of China (NSFC 51278079). And we also gratefully acknowledge the support from the Open Project of State Key Laboratory of Urban Water Resource and Environment (Harbin Institute of Technology, NO. ESK201301).
+
+## Appendix A Supplementary material
+
+Supplementary data associated with this article can be found, in the online version, at [http://dx.doi.org/10.1016/j.cej.2013.11.057](http://dx.doi.org/10.1016/j.cej.2013.11.057).
+
+## References
+
+* (1) K. Fent, A.A. Weston, D. Caminada, Ecotoxicology of human pharmaceuticals, August. Toxicol. 76 (2006) 122-159.
+* (2) S.K. Nefetan, T.J. Collins, Human pharmaceuticals in the aquatic environment: a challenge to green chemistry, Chem. Rev. 107 (2007) 2319-2364.
+* (3) A.J. Watkinson, E.J. Murthy, S.D. Costanzo, Removal of antibiotics in conventional and advanced wastewater treatment: implications for environmental discharge and wastewater recycling. Water Res. 41 (2007) 4164-4176.
+* (4) Q. Tao Dinh, F. Allot, E. Moreau-Guignon, J. Burin, M. Chevreuil, P. Labadie, Measurement of trace levels of antibiotics in river water using on-line enrichment and tripic-quadrupole L-MS/MS, Talanta 85 (2011) 1283-1245.
+* (5) M. Al Aukidy, P. Verlichi, A. Jelic, M. Drorovic, D. Barcelo, Monitoring release of pharmaceutical compounds: occurrence and environmental risk assessment of two WWTP effluents and their receiving bodies in the Po Valley, Italy, Sci. Total Environ. 438 (2015) 15-25.
+* (6) J.M. Cha, S. Yang, K.H. Carlson, Trace determination of p-lactam antibiotics in surface water and urban wastewater using liquid chromatography combined with electrospray tandem mass spectrometry. J. Chromatogr. 115 (2006) 46-57.
+* (7) C. Wang, H. Shi, C.D. Adams, S. Gamagedara, L. Stayton, T. Timmons, Y. Ma, Investigation of pharmaceuticals in Missouri natural and drinking water using high performance liquid chromatography-tandem mass spectrometry, Water Res. 45 (2011) 1818-1828.
+* (8) A.J. Watkinson, E.J. Murthy, D.W. Kolpin, S.D. Costanzo, The occurrence of antibiotics in an urban watershed: from wastewater to drinking water, Sci. Total Environ. 407 (2009) 2711-2723.
+* (9) S. Zou, W. Xu, R. Zhang, J. Tang, Y. Chen, G. Zhang, Occurrence and distribution of antibiotics in coastal water of the Bohai Bay. China: impacts of river discharge and aquaculture activities, Environ. Pollut. 159 (2011) 2913-2920.
+* (10) A. Shimizu, H. Takada, T. Koike, A. Takeshita, M. Saha, Rinawati, N. Nakada, A. Murata, T. Suzuki, S. Suzuki, N.H. Chiem, B.C. Tuyen, P.H. Viet, M.A. Siringan, C. Kwon, M.P. Zakaria, A. Reugnagus, Ubiquitous occurrence of sulfanamides in tropical Asian waters, Sci. Total Environ. 452 (2013) 108-115.
+* a review
+- Part II, Chemosphere 75 (2009) 435-441.
+* (12) I. Michael, I. Rizzo, C.S. Mordell, C.M. Manaia, C. Merlin, T. Schwartz, C. Dagot, D. Fatta-Kassinos, Urban wastewater treatment plants as hotspots for the release of antibiotics in the environment: a review, Water Res. 47 (2013) 957-995.
+* (13) H. Zhang, P. Liu, V. Feng, F. Yang, Fate of antibiotics during wastewater treatment and antibiotic distribution in the effluent-receiving waters of the Yellow Sea, northern China, Mar. Pollut. 73 (2013) 282-290.
+* (14) J.L. Acero, F. J. Benitez, A. Lal, F.J. Real, F. Teva, Membrane filtration technologies applied to municipal secondary effluents for potential reuse, J. Hazard. Mater. 177 (2010) 390-398.
+* (15) J.L. Acero, F.J. Benitez, F. Kra, A.L. Leal, Retention of emerging micropollutants from UP water and a municipal secondary effluent by ultrafiltration and nanofiltration, Chem. Eng. J. 163 (2010) 264-272.
+* (16) C. Djilani, R. Zaghdoud, A. Modarressi, M. Rogalski, F. Djazi, A. Lallam, Elimination of organic micropollutants by adsorption on activated carbon prepared from agricultural waste, Chem. Eng. J. 189-190 (2012) 203-212.
+* (17) N. De la Cruz, J. Gimenez, S. Szulgas, D. Grandjean, L.F. de Alencato, C. Pulgarin, Degradation of 32 emergent contaminants by UV and neutral photo-fenton in domestic wastewater effluent previously treated by activated sludge, Water Res. 46 (2012) 1947-1957.
+* (18) W.T.M. Audenaert, D. Vanderendonck, S.W.H. Van Hulle, L. Nopens, Comparison of ozone and HO center not induced conversion of effluent organic matter (EOM) using random and UV/H2O2 treatment, Water Res. 47 (2013) 2387-2398.
+* (19) A. Justo, O. Gonzalez, J. Accha, S. Perez, D. Barcelo, C. Sans, S. Esplugas, Pharmaceuticals and organic pollution mitigation in reclamation osmosis phones by UV/H2O2 and ozone, J. Hazard. Mater. (2013) <[http://dx.doi.org/10.1016/j.ha/maa.2013.1005.1036](http://dx.doi.org/10.1016/j.ha/maa.2013.1005.1036).
+* (20) S. Miralles-Cuevas, A. Arquds, M.I. Maldonado, J.A. Sanchez-Perez, S. Malato Rodriguez, Combined nanofiltration and photo-fenton treatment of water containing micropollutants, Chem. Eng. J. 224 (2013) 89-95.
+* (21) D. Bixo, C. Thoeye, J. De Koning, D. Joskinsovic, D. Savic, T. Wittgens, T. Melin, Wastewater reuse in Europe, Desailtion 187 (2008) 89-101.
+* (22) A. Pore-Gonzalez, A. Murtagi, R. Ibanez, I. Ortiz, State of the art and review on the treatment technologies of water reverse osmosis concentrates, Water Res. 46 (2012) 267-283.
+* (23) C.O. Lee, K.J. Howe, B.M. Thomson, Ozone and biofiliration as an alternative to reverse osmosis for removing PCPCs and micropollutants from treated wastewater, Water Res. 46 (2012) 1005-1014.
+* (24) W.T.M. Audenaert, D. Vanderendonck, S.W.H. Van Hulle, L. Nopens, Comparison of ozone and HO induced conversion of effluent organic matter (EOM) using ozonation and UV/H2O2 treatment, Water Res. 47 (2013) 2387-2398.
+* (25) T. Zhou, T.-T. Lim, S.-S. Chin, A.G. Fane, Treatment of organics in reverse osmosis concentrate from a municipal wastewater regulation plant: feasibility test of advanced oxidation processes with/without pretreatment, Chem. Eng. J. 166 (2011) 932-939.
+* (26) Chinese-SERA, Water and Wastewater Monitoring Methods, third ed., Chinese Environmental Science Publishing House, 1997.
+* (27) QIF-3028-29. (28) QIF-3028-29. (29). The measures of ozone concentration, output and specific energy consumption for ozone generator, 1994 (in Chinese).
+* determination of the acute toxicity
+- luminescence bacteria test, 1995 (in Chinese).
+* (30) K.L. Tu, A.R. Chivas, L.D. Nghiem, Effects of membrane fouling and scaling on boron rejection by nanofiltration and reverse osmosis membranes, Desailtion 279 (2011) 269-277.
+* (31) E. Steinle-Darling, E. Litwiller, M. Reinhard, Effects of sorption on the rejection of trace organic contaminants during nanofiltration, Environ. Sci. Technol. 44 (2010) 2592-2598.
+* (32) T. Fujioka, S.J. Khan, J. McDonald, R.K. Henderson, Y. Pousasde, J.E. Dwess, L.D. Nghiem, Effects of membrane fouling on N-nitrosamine rejection by nanofiltration and reverse osmosis membranes, J. Membr. Sci. 427 (2013) 3311-3119.
+* (33) I. Kim, H. Tanaka, Photodegradation characteristics of PCPs in water with UV treatment, Environ. Int. 35 (2009) 793-802.
+* (34) O. Legrini, E. Oliveros, A.M. Braun, Photochemical processes for water treatment, Chem. Rev. 93 (1993) 671-698.
+* (35) G. Heit, A. Neuner, P.-Y. Satay, A.M. Braun, Vacuum-UV (172 nm) actinometry. The quantum yield of the photolysis of water, J. Phys. Chem. A 102 (1998) 5551-5561.
+* (36) I. Kim, N. Yamashita, H. Tanaka, Performance of UV and UV/H2O2 processes for the removal of pharmaceuticals detected in secondary effluent of a sewage treatment plant in Japan, J. Hazard. Mater. 166 (2009) 1134-1140.
+* (37) U. Von Gutten, Ozonation of drinking water: Part I, oxidation kinetics and product formation, Water Res. 373 (2003) 143-1467.
+* (38) M.C. Dodd, M.O. Buffle, U. Von Gutten, Oxidation of antibacterial molecules by aqueous ozone: moiety-specific reaction kinetics and application to ozone-based wastewater treatment, Environ. Sci. Technol. 40 (2006) 1969-1977.
+
+* [38] C. Liu, V. Nanabobina, G.V. Korshin, W.J. Jiang, Spectroscopic study of degradation products of ciprofloxacin, northloxacin and lomerloxacin formed in comanored wastewater, Water Res. 46 (2012) 5235-5246.
+* [39] B.D. Witte, H.V. Langehove, K. Hemelseyt, K. Demesterre, P.D. Wispelaere, V. Van Speybroeck, J. Dewulf, Levofloxacin ozonation in water: rate determining process parameters and reaction pathway elucidation, Chemosphere 76 (2005) 683-689.
+* [40] N.M. Vieno, T. Tulkanen, L. Kronberg, Analysis of neutral and basic pharmaceuticals in sewage treatment plants and in recipient rivers using solid phase extraction and liquid chromatography-tandem mass spectrometry detection, J. Chromatogr. 1134 (2006) 101-111.
+* [41] C. Marquez, E.M. Rodriguez, J.J. Beltran, P.M. Alvarez, Determination of rate constants for ozonation of olfloxacin in aqueous solution, Ozone Sci. Eng. 35 (2013) 186-195.
+* [42] A. Keszler, R. Kalyanaraman, N. Hogg. Comparative investigation of superoxide trapping by cyclic nitrogen spin traps: the use of singular value decomposition and multiple linear regression analysis, Free Radical Biol. Med. 35 (2003) 1149-1157.
+* [43] M.M. Huber, A. Gobel, A. Joss, N. Hermann, D. Loffler, C.S. Mcardell, A. Ried, H. Siegrist, T.A. Ternes, U. von Gunten, Oxidation of pharmaceuticals during ozonation of municipal wastewater effluents: a pilot study, Environ. Sci. Technol. 39 (2005) 4290-4299.
+* [44] J. Reidenovic, M. Godehardt, M. Petrovic, A. Hein, M. Farre, M. Jekel, D. Barcelo, Evidencing generation of persistent ozonation products of antibiotics rouktronym and interruption, Environ. Sci. Technol. 43 (2009) 6808-6815.
+* [45] S. Liu, M. Lim, R. Fabris, C.W.K. Chow, M. Drikas, C., Korshin, R. Annal, Multi-wavelength spectroscopic and chromatography study on the photocatalytic oxidation of natural organic matter, Water Res. 44 (2010) 2525-2532.
+* [46] L.I. Weishaar, G.R. Alken, B.A. Bergamaschi, M.S. Fran, R. Fujii, K. Mopper, Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon, Environ. Sci. Technol. 37 (2003) 4702-4708.

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2014.06.071.mmd

@@ -0,0 +1,328 @@
+Ozone oxidation kinetics of Reactive Blue 19 anthraquinone dye in a tubular _in situ_ ozone generator and reactor: Modeling and sensitivity analyses
+
+Kishora K. Panda, Alexander P. Mathews
+
+# Abstract
+
+Modeling of ozone absorption and reaction in an _in situ_ ozone generator and reactor.
+
+Reaction kinetics and stoichiometry of primary and secondary product formation represented well.
+
+Second order reaction rate constants are invariant with initial dye concentration.
+
+Absorption and reaction model predict extents of reaction in the liquid film and the bulk liquid.
+
+## 1 Introduction
+
+There has been a rapid growth in industrial production worldwide in recent years and concomitant increase in generation of waste streams that is impacting the environment and human health. The appearance of recalcitrant compounds that are toxic, mutagenic, carcinogenic or endocrine disrupting in water supplies has created a need for improvements in existing technologies, and the development of new technologies to contain this problem. Many dyes, pharmaceutical products, drugs, hormones and synthetic organic chemicals are not captured by current waste treatment processes and are discharged to receiving waters. Compounds such as 17b-estradiol, estrone, bisphenol A, and metabolites of alkyl phenol polyethoxyles are present in ng/L to g/L concentrations in receiving waters and disrupt endocrine functions of aquatic organisms [1,2]. As a result, there is increased impetus to improve existing technologies, and to develop new technologies that provide effective treatment of wastewaters at affordable costs. New oxidation techniques using ozone, supercritical water, UV photons, ultrasound irradiation, electron beam irradiation, and non-thermal plasmas are being developed or improved upon to address this problem [3]. Among these oxidation techniques, ozone has been gaining popularity as an oxidant by itself or in combination with hydrogen peroxide or UV to produce hydroxyl radicals for oxidation of recalcitrant organics and the disinfection of water supplies. The oxidation potential of ozone is 1.5 times that of chlorine, and it does not generate chlorinated disinfection byproducts that are carcinogenic. Hence, the potential for use of ozone oxidation technology is high due to its high technical feasibility. However, the widespread use of ozone in water and wastewater treatment applications is limited at present due to the relatively high cost of ozonation. This is in part due to mass transfer limitations stemming from the low solubility of ozone, and improper choice of contacts without consideration of the operative reaction regime.
+
+Ozone is typically produced in bulk through high voltage electrical discharge across a thin gap of dry air or oxygen. Conventional ozonators are typically of bubble column design, and the generated ozone is transported through ozone resistant pipes and bubbled into the bulk liquid. If deep tanks are not used, there will be substantial loss of ozone to the gas phase, and this ozone must be destroyed before the exhaust gas can be vented. The electrical corona discharge process can generate in addition to ozone, other short-lived species such as OH radical, OH+ ion, atomic hydrogen, electrons, and oxygen atomic radicals [3-5]. The generation of these species depends upon the reactor configuration and the voltage used. These reactive species can enhance oxidation rates if they can react with contaminants in the fluid before decay. Several different designs have been reported in the literature attempting to generate ozone _in situ_ to utilize the reactive species to enhance oxidation rates [4-6]. These are studies mainly aimed at evaluating the effectiveness of the active species formed near the corona in the oxidation of organic compounds. It may be noted that the ozonator design and gas flow pattern affects the transfer and availability of these species for reaction with organics. So an efficient reactor design must provide a high mass transfer rate of ozone from gas to liquid phase, immediate reaction of any short-lived species on generation, and feasibility for practical application in water and wastewater treatment with reduced cost. These aspects were considered in this research to develop an _in situ_ ozone reactor system.
+
+The _in situ_ reactor design developed in this research utilizes a porous ceramic tubular electrode around the periphery of which ozone is generated by corona discharge. The generated ozone and reactive species migrate through the pores of the electrode and react with the contaminants in the fluid flowing inside the tubular electrode. Panda and Mathews [7] have reported on the enhanced mass transfer obtained using this system. The voltage used in this study is not sufficiently high to generate significant quantities of reactive species. The goal of this paper is to examine the kinetics of oxidation of a high molecular weight organic compound (Reactive Blue 19 dye) using the porous electrode ozonator, and to present modeling and sensitivity analysis data pertaining to absorption and reaction in the _in situ_ porous electrode ozonator.
+
+## Materials and methods
+
+### Materials
+
+Ultra high purity oxygen (minimum purity of 99.994%, OX UHP 300+) was obtained from Airgas Inc., (Radner, PA). for use in the production of ozone. Reactive Blue 19 (RB-19) dye (Remazol Brilliant Blue R) was obtained from Sigma-Aldrich Company (St. Louis, MO), and used without purification. The dye solution was prepared in distilled water and the pH was adjusted to 2 using sulfuric acid. A schematic of the _in situ_ ozone generator-reactor system is shown in Fig. 1. Ozone is generated around the periphery of the electrode by high voltage discharge, and the ozone and reactive species diffuse immediately and react with the contaminants flowing inside the porous electrode. Fabrication details of the laboratory unit have been reported elsewhere [7]. The electrical circuitry was designed to supply less hazardous high frequency (15-20 kHz) pulsed AC HT (5-10 kV) from a regular 110 V wall terminal [8].
+
+### In situ reactor dye azonation studies
+
+The experimental set-up is shown in Fig. 2. Due to the small operating volume of the reactor (0.09 L), the porous electrode reactor was operated in recycle mode. Recirculation unit 1 of volume 1.08 L was used in dye oxidation kinetics studies for recirculation of the solution through the reactor. Since the generated ozone is consumed in the reactor, ozone generation must be assessed independently. The ozone generation rate was determined using mass balance approach and verified using the potassium iodide method [9]. Prior to starting the dye ozonation experiment, ozone generation rate in the _in situ_ ozonator was estimated by running distilled and deionized water at pH 2 from a second recirculation reservoir (unit 2) of volume 0.4 L placed in parallel with the recirculation reservoir holding the dye solution. Ozone generation was estimated using a mass balance equation:
+
+\[G_{\text{in}} = G_{\text{OSS}} + \frac{k_{d}C_{A,\text{LOS}}^{\text{m}}V}{Q_{g}}\]
+
+here _G_OSS and _C_A,LOS are the steady state off-gas and dissolved ozone concentrations respectively, _k_d the ozone decomposition rate constant, \(V\) the total volume of water in the recirculation unit and the reactor, _Q_g the gas flow rate, and \(m\) the reaction order for the decomposition of ozone in water. After finishing this step, the ozone generation was stopped. All the ozonated water from the reservoir and the reactor was then drained, and the reservoir was filled with distilled water. The reactor and the connecting tubes were then rinsed with distilled water several times by circulating water from the second reservoir filled with DDI water. During all these cleaning procedures, gas was continuously flowing through the ozonator and passing through the ozone gas analyzer. After the ozone analyzer showed zero reading and the reactor was cleaned, two three-way valves were operated to bring the reservoir with dye solution in line with the reactor. The dye solution was then pumped through the _in situ_ ozonator and the _in situ_ ozonator was turned on. Water samples of about 20 mL were drawn at set intervals from the bottom of the recirculation unit by opening the outlet valve and analyses were made immediately for dissolved ozone, dye and COD concentrations. Two ml samples were used for measuring the COD concentrations. After the absorbance measurement, the sample solution along with make-up dye solution with concentration equal to the concentration of sample dye solution, with a total volume of 20 mL was returned to the recirculation unit. This ensured constant dye solution level in the reservoir. The off-gas ozone concentration readings were noted from the online analyzer BMT 964, BMT Messertechnik (Berlin, Germany) at the same time intervals. The experiment was carried out until dissolved ozone concentration in the recirculation unit and off-gas readings reached constant values.
+
+### Analytical methods
+
+RB-19 dye concentrations were measured by determining absorbance at 590 nm with a UV-Visible (Cary 50, Varian, Inc.) spectrophotometer. Ozonation of RB-19 dye produces several primary and secondary ozonation products. Degradation of RB-19 involves the destruction of chromophore components followed by the formation of anthraquinone derivatives as primary products. Further oxidation results in the generation of several products including 1,3-inandone, furans, and organic acids [10]. It is clear that dye decolorization reaction can be followed by determining absorbance values at 590 nm. In the case of ozonation products, chemical oxygen demand (COD) was used as a surrogate parameter to follow the reaction to steady-state conditions. A linear
+
+Figure 1: In situ ozone generator and reactor.
+
+Figure 2: Experimental set-up for ozone absorption and reaction studies in the tubular in situ reactor.
+
+relationship was obtained in the calibration of dye concentration with corresponding COD values. During the dye ozonation studies, dissolved ozone concentrations were determined using the Indigo method.
+
+### Gas holdup measurement
+
+Gas holdup in the _in situ_ reactor was measured by filling the reactor with water and collecting excess water that was forced out of the reactor upon the introduction of gas at a specified flow rate. Gas holdup can be calculated from this measurement for each gas flow rate [11]. To estimate the corresponding gas holdup while water is circulated in the reactor, the correlation (Eq. (2)) proposed by Hughmark [12], and Danckwerts [13] was used:
+
+\[\frac{u_{g}}{\varepsilon_{g0}} = \frac{u_{g}}{\varepsilon_{g}} - \frac{u_{i}}{1 - \varepsilon_{g}}\]
+
+here, _ug_ is the superficial gas velocity, _ui_ the superficial liquid velocity (cocurrent flow), _e_gp the gas holdup at no water flow, and _e_g the gas holdup during water flow.
+
+### Stoichiometric factor for RB-19 dye decolorization reaction
+
+The stoichiometric factor \(z\)1 for the reaction between dissolved ozone and the dye was determined experimentally by reacting ozone with excess dye so as to minimize any secondary reaction of ozone with the primary products of ozonation. The initial mole ratio of dye concentration to ozone concentration was chosen to be between 5 and 10. A volume of 30 mL of dye solution of concentration 80, 100, or 160 mg/L was contacted with 10 mL of stock ozone solution with dissolved ozone concentrations ranging from 1.8 to 7 mg/L in a 50 mL glass reaction vessel. The dissolved ozone concentration was measured immediately prior to the addition of dye to the reaction vessel. After 10 min of reaction time, the residual dye and ozone concentrations were measured. In all experiments, no dissolved ozone was detected at the end of the reaction. The stoichiometric factor was then calculated as \(z\)1 = D[qpe]/D[O3]. All experiments were conducted in triplicate.
+
+## Modeling ozonation process in _in situ_ reactor
+
+A heterogeneous gas-liquid reaction such as ozonation involves two processes, ozone absorption and simultaneous chemical reaction. Depending on the relative rate of individual processes, ozonation reaction rate can be limited by one of the processes. Based on the reaction rate of ozone with the reactant in water, the kinetic regimes could range from slow to instantaneous [14]. In the slow reaction regime, ozonation efficiency is limited by the reaction rate, while in fast to instantaneous reaction regime it is limited by mass transfer rate. Also the mass transfer rate is enhanced in the presence of chemical reaction affecting the overall reaction rate. Therefore, it is necessary to take mass transfer into consideration while studying any gas-liquid reaction. In the _in situ_ reactor, ozonation experiments were conducted in a closed recirculation mode and there could be a change in kinetic regime during the course of reaction due to decrease in concentration of the dye. A mathematical model has been developed to simulate the dye ozonation process in the _in situ_ reactor by taking ozone mass transfer, variable kinetic regime, secondary reaction of ozone with the byproducts of ozonation, and ozone self-decomposition reaction into consideration. The model consists of two parts. The first part describes the mass transfer and reaction inside the hydrodynamic film, and the second part considers the reaction of transferred ozone from the film into the bulk liquid. The following assumptions were made in developing the model:
+
+* The _in situ_ reactor is divided into a series of \(n\) number of equal sized reactors (tanks) where both liquid and gas phases are completely mixed.
+* Both liquid and gas phases are completely mixed inside the recirculation unit, and headspace respectively.
+* The two-film theory is used to describe ozone mass transfer from gas phase to liquid phase, and mass transfer resistance is negligible inside the gas film as ozone is sparingly soluble in water.
+* Self-decomposition reaction of ozone is _m_th order in the liquid phase and negligible in the gas phase.
+* Mass transfer coefficient is constant along the height of the column.
+* Temperature, pressure, and liquid volume are maintained constant.
+
+Considering the liquid film and the bulk liquid separately in the analysis, mass balance equations can be written for individual components in each zone.
+
+### Mass balance for the film
+
+The mass transfer of ozone is assumed to take place from the gas phase to the liquid phase through an intermediate hypothetical liquid film of depth _d_L. The mass transport within the film is assumed to be at steady state and hence there is no accumulation of mass inside the film. Also, decolorization reaction between molecular ozone and the organic compound is fast enough to make the secondary reaction of ozone with byproducts minimum. The reaction between ozone and organic compound is often described as a second order reaction with a first order rate with respect to individual reactants. Thus for a gas A (ozone) diffusing into the liquid film and undergoing a second order irreversible reaction (A + \(z\)1B - P; \(r_{A}\) = \(k\)1CaCa and \(r_{B}\) = \(z\)1Fa) with compound B (dye) within the film, the mass balance on both the compound A and B at steady state can be written in dimensionless form as given below:
+
+\[\frac{d^{2}A}{d\text{X}^{2}} - Ha^{2}A\text{B} = 0\]
+
+\[\frac{d^{2}B}{d\text{X}^{2}} - \frac{Ha^{2}}{E_{\text{inst}} - 1}\text{A}B = 0\]
+
+\[A(0) = 1;\quad B(1) = 1\]
+
+\[\left( \frac{dA}{d\text{X}} \right)_{\text{X} - 1} = (Ha^{2} - R)A;\quad\left( \frac{dB}{d\text{X}} \right)_{\text{X} - 0} = 0\]
+
+The dimensionless variables are defined by the following equations:
+
+\[X = \frac{x}{\delta_{\text{L}}}\quad A = \frac{C_{A}}{C_{A}}\quad B = \frac{C_{B}}{C_{B,b}}\]
+
+\[Ha = \frac{\sqrt{k_{1}C_{B,b}D_{A}}}{k_{l}}\]
+
+\[E_{\text{inst}} = 1 + \frac{D_{B}C_{B,b}}{z_{1}D_{A}C_{A}}\]
+
+\[R = \frac{k_{1}C_{B,b}\varepsilon_{l}}{k_{l}a}\]
+
+\[Ha\]
+
+ is the Hatta number, _E_inst the instantaneous enhancement factor, and \(R\) the reaction number, \(k_{l}\) the mass transfer coefficient, and _Ka_a the volumetric mass transfer coefficient.
+
+### Enhancement factor (E) and depletion factor (D)
+
+The Hatta number is a diffusion-reaction parameter that is of significance in gas-liquid reactions. It is a measure of the reaction time relative to that of diffusion. There are several diffusion-reaction regimes that can be defined based on the value of \(Ra\)[14]. When the chemical reaction rate is slow, the rate of ozone diffusion from the interface into the liquid film is equal to the rate of diffusion of ozone out of the film into the bulk, and the chemical reaction takes place completely in the bulk liquid. This corresponds to \(0.02<Ha<0.3\). With an increase in reaction rate, the reaction takes place partly in the film and partly in the bulk liquid (\(0.3<Ha<3\)). When the rate becomes very large, the ozone reacts completely within the film, and no reaction takes place in bulk. To account for the proportion of ozone reacting in the film and the bulk liquid, the concept of depletion factor, \(D\), analogous to the enhancement factor has been introduced [15, 16]. \(D\) is defined as the ratio between the flow rate of ozone entering the liquid bulk in the presence of a chemical reaction and the flow rate entering at the same surface during physical absorption alone [15]. The film theory can be used to explain this concept and to determine the values of the enhancement and depletion factors in a gas-liquid reaction.
+
+During a slow reaction regime, the ozone concentration profile (profile 1) in the liquid film is a straight line (PR) and equal concentration gradients at both ends of the line indicate complete depletion of ozone from the film into the bulk liquid (Fig. 3). With increase in reaction rate, the profile (e.g., profile 2) becomes curved and the gradient at the interface (PP\({}^{\prime}\)) becomes higher than that at the liquid bulk end (RR\({}^{\prime}\)). In this case, reaction is shared between the film and the bulk liquid. In the presence of a sufficiently fast reaction (profile 3), the concentration gradient at the bulk liquid boundary end becomes zero making depletion of ozone to bulk liquid zero and 100% of reaction occurs in the film.
+
+The enhancement factor is the ratio of mass transfer rate in the presence of chemical reaction to that for physical absorption alone, and can be correlated to the ratio of the concentration gradients in presence of chemical reaction (slope of line PP\({}^{\prime}\)) and physical absorption of ozone (slope of line PR). In the same analogy, the depletion factor is the ratio of depletion of ozone into bulk liquid in presence of chemical reaction to that for physical absorption alone, can be correlated to the ratio of the concentration gradients in the presence of chemical reaction (slope of line RR\({}^{\prime}\)) and physical absorption of ozone (slope of line RP). Mathematically \(E\) and \(D\) can be expressed as:
+
+\[E=\frac{\text{Slope of line PP}^{\prime}}{\text{Slope of line PR}}=\frac{-\left(\frac{A_{C}}{A_{A}}\right)_{n=0}}{\left(C_{A}-C_{A,A}\right)_{n=0}}=\frac{-D_{A}\left(\frac{A_{C}}{A_{A}}\right)_{n=0}}{k_{L}\left(C_{A}-C_{A,A}\right)} \tag{11}\]
+
+\[D=\frac{\text{Slope of line RR}^{\prime}}{\text{Slope of line RP}}=\frac{-\left(\frac{A_{C}}{A_{A}}\right)_{n=k_{h}}}{\left(C_{A}-C_{A,B}\right)_{n=k_{h}}}=\frac{-D_{A}\left(\frac{A_{C}}{A_{A}}\right)_{n=k_{h}}}{k_{L}\left(C_{A}-C_{A,B}\right)} \tag{12}\]
+
+So the ozone absorption rate from the gas phase to the liquid film (\(r_{A,\text{abs}}\)) and from the liquid film into the bulk liquid (\(r_{A,\text{abs}}\)) can be written respectively as Eqs. (13) and (14):
+
+\[r_{A,\text{abs}}=EK_{t}a(C_{A}^{*}-C_{A,B}) \tag{13}\]
+
+\[r_{A,b}=DK_{t}a(C_{A}^{*}-C_{A,B}) \tag{14}\]
+
+### Mass balance for the bulk liquid
+
+The dye solution is circulated from the recirculation unit through the porous tubular reactor. The gas enters normal to the direction of liquid flow in the reactor. Owing to the hydrostatic pressure from the liquid flowing vertically up in the reactor, the gas flow rate varies linearly along the height of the reactor. Modeling the reactor as a number of well-mixed reactors in series is a rational approach for this type of gas liquid contacting system, and the value of the number of equivalent reactors (\(n\)) must be determined through tracer tests. It is assumed that the reaction in the film is at steady state, and that there is no accumulation of products inside the film. Also, it is assumed that negligible reaction occurs between ozone and byproducts formed from the degradation of the dye in the film, whereas ozone reacts simultaneously with the dye and the byproducts in the bulk liquid. This secondary reaction of ozone with the byproducts is taken into account in the model by quantifying the byproducts in terms of an easily measured surrogate parameter, COD.
+
+The reaction rate equations of the absorbing gas A (ozone) with a non-volatile compound B (dye), and with primary products (P1) formed from dye oxidation are given by Eqs. (15) and (16). The reactions are assumed to be first order with respect to each of the reactants and the unknown stoichiometric coefficients are
+
+Figure 3: Effect of chemical reaction on concentration profile of ozone and dye inside the hydrodynamic film, and representation of \(E\) and \(D\).
+
+represented by \(z\)1, \(z\)2, \(z\)3, and \(z\)4. As noted in Section 2.2, the experimental studies were conducted using distilled water at a pH of 2. Only molecular ozone is present at this pH, and there are no hydroxyl radicals or scavengers present in this system:
+
+\[A + z_{1}B{\rightarrow}z_{2}P1\]
+
+\[A + z_{3}P1{\rightarrow}z_{2}z_{4}P2\]
+
+Mass balance equations can now be written for the reactor, the recirculation unit and the headspace. Dividing the tubular reactor into \(n\) compartments as determined from RTD studies [7], the mass balance equations for gas phase ozone, liquid phase ozone, dye, primary products, and secondary products can be written for each compartment \(i\) (_i_ = 1, _n_) as:
+
+\[\frac{dG_{\text{in}}}{dt} = \left( \frac{n}{e_{\text{g}}V_{\text{g}}} \right)\left( {\text{Q}_{\text{g}i - 1}} \right)\text{G}_{\text{g}i - 1}\text{(t)} + q_{i}G_{\text{in}} - \text{Q}_{\text{g}}G_{\text{in}}\text{(t)} \right)\]
+
+\[- \frac{(1 - e_{\text{g}})}{e_{\text{g}}}\text{E}_{\text{g}}K_{\text{i}}a\left( \alpha G_{i} - C_{\text{AB}}\text{(t)} \right)\]
+
+\[\frac{dC_{\text{AB}}}{dt} = \left( \frac{nQ_{i}}{(1 - e_{\text{g}})V_{\text{g}}} \right)\left( {\text{C}_{\text{AB}i - 1}} \right)\left( {\text{C}_{\text{AB}}\text{(t)}} \right) + D_{i}K_{\text{i}}a\left( \alpha G_{i} - C_{\text{AB}}\text{(t)} \right)\]
+
+\[- k_{1}C_{\text{AB}}C_{\text{AB}} - k_{2}C_{\text{AB}}C_{\text{P1B}} - k_{4}C_{\text{AB}}^{\text{m}}\text{(t)};\]
+
+\[\frac{dC_{\text{BB}}}{dt} = \left( \frac{nQ_{i}}{(1 - e_{\text{g}})V_{\text{g}}} \right)\left( {\text{C}_{\text{BB}i - 1}} \right)\left( {\text{C}_{\text{BB}}\text{(t)}} \right)\]
+
+\[- z_{1}\left[ \left( E_{i} - D_{i} \right)K_{\text{i}}a\left( \alpha G_{i} - C_{\text{AB}}\text{(t)} \right) + k_{1}C_{\text{AB}}\text{(t)}C_{\text{BB}}\text{(t)} \right]\]
+
+\[\frac{dC_{\text{PB}}}{dt} = \left( \frac{nQ_{i}}{e_{\text{g}}V_{\text{g}}} \right)\left( {\text{C}_{\text{P1B}i - 1}} \right)\left( {\text{C}_{\text{P1B}i - 1}} \right)\left( {\text{C}_{\text{P1B}}\text{(t)}} \right)\]
+
+\[+ z_{2}\left[ \left( E_{i} - D_{i} \right)K_{\text{i}}a\left( \alpha G_{i} - C_{\text{AB}}\text{(t)} \right) + k_{1}C_{\text{AB}}\text{(t)}C_{\text{BB}}\text{(t)} \right]\]
+
+\[- z_{3}k_{2}C_{\text{AB}}C_{\text{PB1}}\]
+
+\[\frac{dC_{\text{PB2}}}{dt} = \left( \frac{nQ_{i}}{e_{\text{g}}V_{\text{g}}} \right)\left( {\text{C}_{\text{P2B}i - 1}} \right)\left( {\text{C}_{\text{P2B}}\text{(t)}} \right) + z_{4}k_{2}C_{\text{AB}}C_{\text{PB1}}\]
+
+\[Q_{\text{g0}} = 0;\text{ }C_{\text{AB0}}\text{(t)} = C_{\text{AB}}\text{(t)};\text{ }C_{\text{BR0}} = C_{\text{BU}};\text{ }C_{\text{P1B0}} = C_{\text{PU}};\text{ }C_{\text{P2B0}} = C_{\text{P20}};\]
+
+Mass balances for gas and liquid phase ozone, dye, primary products, and secondary products in the recirculation unit are given in Eqs. (23-27):
+
+\[\frac{dG_{\text{0}}}{dt} = \frac{Q_{\text{g}}}{V_{\text{hs}}}\left( {G_{\text{in}}\text{(t)} - G_{\text{0}}\text{(t)}} \right)\]
+
+\[\frac{dC_{\text{AU}}}{dt} = \frac{Q_{i}}{V_{\text{U}}}\left( {C_{\text{AR}}\text{(t)} - C_{\text{AU}}\text{(t)} } \right) - k_{1}C_{\text{AU}}C_{\text{BB}} - k_{2}C_{\text{AU}}C_{\text{P1,U}} - k_{4}C_{\text{AU}}^{\text{m}}\]
+
+\[\frac{dC_{\text{BU}}}{dt} = \frac{Q_{i}}{V_{\text{U}}}\left( {C_{\text{BR}}\text{(t)} - C_{\text{BU}}\text{(t)} } \right) - z_{1}k_{1}C_{\text{AU}}\text{(t)}C_{\text{BU}}\text{(t)}\]
+
+\[\frac{dC_{\text{PU}}}{dt} = \frac{Q_{i}}{V_{\text{U}}}\left( {C_{\text{P1B}\text{(t)}} - C_{\text{PU}}\text{(t)} } \right) + z_{2}k_{1}C_{\text{AU}}\text{(t)}C_{\text{BU}}\text{(t)}\]
+
+\[- z_{3}k_{2}C_{\text{AU}}\text{(t)}C_{\text{PU}}\text{(t)}\text{(t)}\]
+
+\[\frac{dC_{\text{P2U}}}{dt} = \frac{Q_{i}}{V_{\text{U}}}\left( {C_{\text{P2B}\text{(t)}} - C_{\text{P2U}}\text{(t)} } \right) + z_{4}k_{2}C_{\text{AU}}\text{(t)}C_{\text{PU}}\text{(t)}\]
+
+The initial conditions for each compartment in the reactor (_i_ = 1, _n_) and the recirculation unit are:
+
+\[\text{Reactor}:G_{\text{R}}\text{(0)} = 0;\text{ }C_{\text{AB}}\text{(0)} = 0;\text{ }C_{\text{BB}}\text{(0)} = C_{\text{BB}}\text{; }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{ }C_{\text{P1B}}\text{(0)} = 0;\text{ }C_{\text{P2B}}\text{(0)} = 0;\text{generation rate (\(G_{\text{in}}\)), initial concentrations of dye, ozone, and COD, and the diffusivity values for ozone and dye in water (\(D_{\text{A}}\) and \(D_{\text{B}}\)).
+
+Based on the input data and the initial concentrations of dye (\(C_{\text{in}}\)), values of three dimensionless numbers, \(Ha\), \(E_{\text{in}}\) and \(R\) corresponding to time \(t=0\) are calculated. Then for these values of \(Ha\), \(E_{\text{in}}\) and \(R\), the set of mass balance equations within the film (Eqs. (3-6)) calculate the profile of ozone and dye concentration inside the diffusion film. From the profile of ozone, \(E\) and \(D\) are calculated according to the Eqs. (11) and (12) respectively. The slopes of the ozone profile used in the numerator of these two equations were calculated from the solution of the Eqs. (3-6) using the central difference formula for first derivative. The values of \(E\) and \(D\) become the input to the set of mass balance equations (Eqs. (17-29)) for the bulk liquid whose solution calculates the value of concentrations of dye, dissolved ozone, primary products (\(P1\)) and secondary products (\(P2\)) at the next time step (\(t=0+\Delta t\)). Then, the program goes back to calculate the new values of \(Ha\), \(E_{\text{in}}\), and \(R\) based on this new value of dye concentration in bulk. This iteration continues until the end of the time value in the experimental data is reached.
+
+#### Solution of mass balance equations in the film
+
+Due to the nonlinear nature of the set of differential equations (Eqs. (3-6)) in the film, numerical solution using finite difference techniques has been adopted to determine the concentration profiles of A and B in the diffusion film as exact analytical solutions are not possible. The dimensionless distance in liquid film from \(X=0\)-\(1\) is divided into \(2n\) number of nodes with an equal distance \(h\) between two adjacent nodes. The central difference formula is used to transform the nonlinear differential equations to a set of nonlinear algebraic equations with \(2n\) unknowns of \(y\)[16]. The system of equations is solved using "fsolve" optimization tool available in MATLAB software program.
+
+#### Solution of bulk liquid mass balance equations
+
+When complex chemistry is involved in a reaction simulation involving the evolution in time of a system defined by nonlinear differential equations, the governing equations are usually highly nonlinear and stiff. The differential equations are characterized as stiff when the time or length scales are vastly disparate, and the system has some solutions that change very rapidly compared with other solutions, or some solutions which change very rapidly at some times and slowly at other times. Owing to severe stability restrictions, explicit numerical algorithms are inappropriate for solving stiff problems and implicit methods must be used. An implicit solution for stiff problems available in MATLAB software (ode23th) was used to solve the set of equations (Eqs. (17-29)) in the bulk liquid. The solutions of the equations for the film calculates the values of \(E\) and \(D\) at time \(t\) and these values are used in the mass balance equation for bulk liquid and the reactor system (Eqs. (17-29)) to predict the concentration of dye, ozone, \(P1\), \(P2\), off-gas at time \(t+\Delta t\). A MATLAB program was written to solve these two sets of equations. The sensitivity of the time step was evaluated in the solution of the model.
+
+## Results and discussion
+
+### General characteristics of dye ozonation in tubular porous electrode reactor
+
+Reactive dyes such as anthraquinone and azo dyes constitute a major share of the worldwide market for different classes of dyes [17]. Anthraquinone based dyes are characterized by a very stable structure and are more resistant to chemical and biological oxidation than azo dyes due to their fused ring structure [18]. Dyes which generate highly colored wastewater from industries such as textile, and pulp and paper, are also classified as acidic, direct, disperse, and reactive dyes. Among these dye classes, the reactive dyes have lower rates of fixation in the fabric and contribute significantly to the color in textile wastewater [19]. In addition, the discharge of untreated dye effluents may cause toxic and mutagenic effects in aquatic organisms in the receiving waters. In this study a reactive anthraquinone dye Reactive Blue 19 is selected as the model organic compound.
+
+The dye oxidation experiments were conducted in a closed loop system by circulating dye solution from the recirculation unit through the _in situ_ reactor as described in Section 3.2. Figs. 5 and 6 show the experimental data for the dye, dissolved ozone, and off-gas ozone concentrations, for initial dye concentrations of 40 and 80 mg/L respectively. Dye decolorization and oxidation data were obtained until all the ozone demanding substances were reacted, and the water became saturated with dissolved ozone.
+
+Figure 5: Determination of reaction rate constant (\(k_{\text{z}}\)) from fitting model to the experimental COD data for ozonation of BB-19 dye in in situ reactor (\(C_{\text{in}}=40\) mg/L, pH = 2, \(Q_{\text{z}}=0.05\) l/min, \(Q_{\text{z}}=0.1\) l/min, and \(G_{\text{in}}=46.6\pm 0.5\) g/nm\({}^{3}\)).
+
+Some general conclusions can be made on dye oxidation kinetics from the profiles in Figs. 5 and 6. The entire experimental period can be divided into two major segments representing two different processes. First, the dye decolorization reaction occurs rapidly by reaction of ozone with the dye chromophores, and is characterized by the disappearance of the blue color. This has also been observed by other researchers (Fanchiang, 2009). Absence of dissolved ozone during this period implies that the reaction rate is very high and the rate of reaction is possibly limited by ozone mass transfer rate. Also, COD values decreased during this period but at a slower rate than dye as observed from the gradients of the concentration profiles for COD and dye during this period. Second, after complete disappearance of blue color and up to the end of the experimental period, the process is governed by the simultaneous process of ozone dissolution and the oxidation of primary products of ozone- tion. The reaction between the primary products and ozone is slow, and hence, dissolved ozone begins to appear in the bulk liquid. Once the ozone demand from the primary products is reduced to zero, the dissolved ozone concentration steadily increases to reach equilibrium value. The off-gas ozone concentration increases to a steady state value close to the feed gas ozone concentration.
+
+As noted earlier, there are numerous products of dye ozonation, and it would be impractical to track and characterize individual compounds under field conditions (Fanchiang, 2008). Chemical oxygen demand (COD) was used as a surrogate parameter to represent primary and secondary products of ozonation as indicated in Section 4.2. In order to observe formation of any byproducts that had absorbance in the visible region, dye samples withdrawn during the experiment were scanned in the spectrum of wavelengths ranging from 400 to 650 nm in a spectrophotometer. It was observed that ozonation of dye solution progressively decreased the absorbance peak for the dye at 590 nm and did not develop any new peaks in the visible region of spectra that would have interfered with the dye measurement.
+
+### Determination of stoichiometric factors for dye decolorization reaction
+
+The stoichiometric factor for the reaction between RB-19 dye and ozone was determined separately as given in Section 2.2. From four runs, a value of \(8.5\pm 0.3\) mg of dye/mg of ozone was obtained for \(z_{1}\). The unit for \(z_{1}\) is expressed in mg of dye/mg of ozone instead of mole for consistency with the units of COD (mg/L). In molar units, the value for \(z_{1}\) is \(0.654\pm 0.026\) mol dye/mole ozone.
+
+The stoichiometric factor, \(z_{2}\), is the amount of byproduct formed from unit mass of dye reacting with ozone. The amount of byproducts formed during the ozonation of dye was expressed in COD values. The COD measurement value of dye wastewater at any point during the ozonation is the sum of the contributions of COD from the dye (B), the byproducts (\(P1\)) and the products (\(P2\)). So the COD of dye solution can be expressed as:
+
+\[\text{COD},\text{mg}/\text{L}=\text{COD}_{\text{B}}+\text{COD}_{\text{Pt}}+ \text{COD}_{P2} \tag{30}\]
+
+where \(\text{COD}_{\text{B}}\) is the COD contributed by the dye, \(\text{COD}_{\text{Pt}}\) is the COD contributed by the byproducts (\(P1\)), and \(\text{COD}_{\text{P2}}\) is the COD contributed by the products (\(P2\)). COD contribution from the dye was determined from a standard curve developed between the dye solution and the exerted COD. The slope of the standard regression line represented by the factor \(c\) was used to calculate the value of \(\text{COD}_{\text{B}}\) for a particular dye concentration. So Eq. (30) can be rearranged to obtain the value of \(\text{COD}_{\text{Pt}}\) (Eq. (31)):
+
+\[\text{COD}_{\text{Pt}}=\text{COD}-c\text{ C}_{\text{B}}-\text{COD}_{P2} \tag{31}\]
+
+For the case when the contribution of COD from \(P2\) is minimum, i.e., at the initial periods of dye decolorization when formation of final product is minimum, one can obtain the value of \(\text{COD}_{\text{Pt}}\) from the experimental data on COD and dye concentration. The stoichiometric factor was then computed as \(z_{2}=z_{1}\) (\(\Delta[\text{COD}_{\text{Pt}}]/\Delta[\text{Dye}]\)) from the experimental data on dye ozonation at three initial dye concentrations of 40, 80, and 160 mg/L in the _in situ_ reactor. The values of \(z_{2}\) obtained in three experimental runs are given in Table 1.
+
+### Determination of stoichiometric factors for reaction between ozone and primary products
+
+The reaction between the dye and ozone predominates as long as the dye is present in the water due to the relatively high reaction rate of ozone with the dye than with the byproducts. So the stoichiometric factor for the reaction between byproducts and ozone was estimated from the experimental data by calculating the amount of ozone consumed per mg of COD reduction after the blue color completely disappeared. The results calculated from three experimental data with different initial dye concentrations of 40,
+
+Figure 6: Determination of reaction rate constant (\(k_{2}\)) from fitting model to the experimental COD data for ozonation of RB-19 dye in _in situ_ reactor (\(C_{\text{Bn}}=80\) mg/L, pH = 2, \(Q_{\text{e}}=0.05\) L/min, \(Q_{\text{e}}=0.1\) L/min, and \(G_{\text{m}}=46.6\pm 0.5\) g/Nm\({}^{3}\)).
+
+80, and 160 mg/L are given in Table 1. Hsu et al. studied the ozonation of Reactive Blue 19 dye and concluded that about 1-2 mg of ozone is required per mg of COD oxidation. Thus an average value of 0.65 for z\({}_{3}\) i.e., about 1.54 mg of ozone required per mg of COD reduction obtained in this study is in agreement with the literature data.
+
+After continuously ozonating the dye solution until the dissolved ozone concentration reached a value close to equilibrium, complete COD removal was not observed implying that dye ozonation does not attain complete mineralization of the organics. The stoichiometric factor z\({}_{4}\) was determined from the experimental data on COD value at the end of complete color removal and the amount of COD left at the end of experiment. This factor was calculated from the experimental data for the COD left at the end of ozonation and COD as byproducts produced at the end of 100% color removal. The values obtained for the stoichiometric factor z\({}_{4}\) are summarized in Table 1.
+
+### Model input parameters
+
+Parameters in the model besides reaction rate constants were either estimated from literature correlations or determined through experiments. The diffusivity of ozone and RB-19 dye in water at 21 degC were calculated from the Wilke and Chang equation and found to be 1.36 x 10-9 m\({}^{2}\)/s and 3.8 x 10-10 m\({}^{2}\)/s respectively [20]. The liquid mass transfer coefficient was estimated using the correlation (Eq. (32)) developed by Calderbank and Moo-Young [21] for bubble sizes below 2.5 mm and found to be 5.7 x 10-5 m/s:
+
+\[k_{k}=0.31(g\nu)^{1/3}\left(\frac{D_{A}}{\nu}\right)^{2/3} \tag{32}\]
+
+where v is the kinematic viscosity of the liquid and \(D_{A}\) the diffusivity of ozone in liquid. The overall mass transfer coefficient, gas holdup, and ozone decomposition reaction rate constant values were obtained from the experimental results on ozone mass transfer studies in the _in situ_ reactor. The values of \(n\) for the number of compartments is series for particular gas and water flow rates in the tubular reactor, were obtained from tracer studies [7].
+
+### Determination of reaction rate constants
+
+The model (Eqs. (3-6) and (17-29)) is used to extract the two rate constants for reaction between ozone and dye (\(k_{1}\)), and ozone and primary products (\(k_{2}\)) in two steps. In the first step, the reaction rate constant value, \(k_{1}\) was determined by fitting the model output for dye concentration to the experimental data. The value of \(k_{1}\) was varied during the model execution until the model output fitted the experimental data satisfying the minimum value of the objective function given by Eq. (33):
+
+\[R=\sqrt{\sum\left(\frac{C_{B,U\text{ exp}}-C_{B,U\text{med}}}{C_{B,U\text{ exp}}}\right)^{2}} \tag{33}\]
+
+here C\({}_{B,U\text{ med}}\), C\({}_{B,U\text{ exp}}\) are the concentrations of dye in the recirculation unit obtained through model and observed in experiment respectively. In order to minimize the interference of ozone reaction with byproducts, the model was fitted to the initial 0-30 min of the ozonation time depending on the dye concentration, when the secondary reaction of ozone with byproduct is negligible.
+
+In the second step, the rate constant for reaction between ozone and primary products, \(k_{2}\) was determined by fitting the model output for concentration of COD to the experimental data. With the value of \(k_{1}\) being known, the value of \(k_{2}\) was determined by fitting the model (Eqs. (3-6) and (17-29)) output for COD concentration to the experimental data. The value of \(k_{2}\) was varied during the model execution until the model output fitted the experimental data satisfying the minimum value of the objective function given by Eq. (34):
+
+\[R=\sqrt{\sum\left(\frac{C_{COD,U\text{ exp}}-C_{COD,U\text{ med}}}{C_{COD,U\text{ exp}}}\right)^{2}} \tag{34}\]
+
+Here \(C_{COD,U\text{ med}}\), \(C_{COD,U\text{ exp}}\) are the concentrations of COD in recirculation unit obtained through model and observed in experiment respectively.
+
+The values obtained for \(k_{1}\) at the three experimental runs are given in Table 2. An average value of 1.51 +- 0.13 x 10-2 L/mg. sec was estimated for the decolorization reaction rate constant between dye and ozone in the _in situ_ reactor. The literature on the kinetics of reaction of ozone with RB-19 dye was reviewed [22,23]. All these studies expressed the decolorization reaction through pseudo-first order reaction kinetics. Chu and Ma [22] studied the ozonation of RB-19 in a semi batch reactor at pH values of 4, 7 and 10, and obtained pseudo-first order rate constants ranging from 3.6 x 10-4 s-1 at pH 4 to 3.0 x 10-4 s-1 at pH 7. As this particular dye shows a long tail in the decolorization profile after about 90% color removal, Hsu et al., [24] expressed the decolorization profile in two different segments by incorporating two (\(k_{1}\) and \(k_{2}\)) rate constants. They reported a value of 1.06 x 10-2 s-1 and 1.78 x 10-3 s-1 for \(k_{1}\) and \(k_{2}^{\prime}\) respectively. Loll et al., [23] compared ultrasound enhanced ozonation of RB-19 with ozonation alone for dye decolorization. The apparent first order rate constant values were obtained to be 0.0019, 0.0029, and 0.0037s-1 for gas phase ozone concentrations of 4.6, 5.4, and 6.2 mg/L respectively. There was no study that reported the second order reaction rate constant between ozone and RB-19 dye in the available literature. Hence the second order kinetics rate constant obtained in this research cannot be compared to the reported results of the literature for RB-19 dye ozonation.
+
+There have been some attempts recently to fit a second order rate equation to the primary decolorization reaction for other anthraquinone and azo dyes [25, 26, 27]. Gomes et al., [25] attempted to fit rate data from a 15-min time period to a second-order rate model without considering diffusion effects. According to the authors, the fitting obtained was "unacceptable", and they fit the data with an exponential decay function. They obtained the rate
+
+\begin{table}
+\begin{tabular}{c c c c} \hline Initial dye conc. & Stoichiometric coefficients & \\ (mg/L) & \(z_{2}\) (mg COD/mg & \(z_{3}\) (mg COD/mg & \(z_{4}\) (mg COD/mg \\ Dye) & ozone) & COD) \\ \hline
+40 & 8.20 & 0.63 & 0.17 \\
+80 & 8.16 & 0.65 & 0.23 \\
+160 & 8.20 & 0.68 & 0.23 \\ Average value & 8.16 ± 0.04 & 0.65 ± 0.02 & 0.21 ± 0.04 \\ \hline \end{tabular}
+\end{table}
+Table 1: Stoichiometric coefficient values determined from experimental data.
+
+\begin{table}
+\begin{tabular}{c c c c c c c} \hline C\({}_{\text{60}}\) (mg/g) & \(k_{1}\) (1×10\({}^{\text{10}}\)) & \(R^{2}\) & \(k_{2}\) (×10\({}^{\text{10}}\)) & \(R^{2}\) & \(E\) & \(D\) \\ L\({}_{1}\) & (Img 1s 1−s−1) & & (Img 1s−1) & & (Img 1s−1) & & \\ \hline
+40 & 1.60 & 0.1089 & 1.58 & 0.3903 & 1.08 & 0.97 \\
+80 & 1.38 & 0.0860 & 1.34 & 0.3998 & 1.16 & 0.92 \\
+160 & 1.53 & 0.1141 & 1.50 & 0.3588 & 1.34 & 0.83 \\ Average value & 1.51 ± 0.13 & & 1.46 ± 0.13 & & \\ \hline \end{tabular}
+\end{table}
+Table 2: Reaction rate constants (\(k_{1}\) and \(k_{2}\)), enhancement factor (\(E\)), and depletion factor (\(D\)) for _in situ_ reactor (\(G_{m}\) = 4.6 ± 0.5 g/Nm3, pH = 2, Q = 0.1 L/min, and Q = 0.05 L/min).
+
+constant by fitting the initial data points over a 400 ms time period, and reported a value of \(7.7\times 10^{-2}\) L mg\({}^{-1}\) s\({}^{-1}\) for Acid Orange 7 (A07) dye at a pH of 3.0. Gomes et al., [26], in a previous study using a quench-flow system, obtained a value of \(3.4\times 10^{-2}\) L mg\({}^{-1}\) s\({}^{-1}\) for A07 dye at a pH of 2.5. The rate constant value of \(1.6\times 10^{-2}\) L mg\({}^{-1}\) s\({}^{-1}\) obtained in our work for RB-19 dye at a pH of 2 is consistent with the above data. The reported results are also consistent with that of Tizaoui and Grima [27], who included diffusion effects in their analysis, and reported a value of \(4.04\times 10^{-1}\) L mg\({}^{-1}\) s\({}^{-1}\) for the ozonation of Reactive Orange 16 (R016) dye. Decolorization reaction is relatively fast for this dye compared to RB-19, and was complete in about 4.5 min.
+
+The results obtained for the secondary reaction rate constant (\(k_{2}\)), enhancement factor (\(E\)), and depletion factor (\(D\)) from the model fit to the experimental data for three different initial dye concentrations are presented in Table 2. As shown in these Figures, an average value of \(1.46\pm 0.13\times 10^{-4}\) L/mg. sec. for \(k_{2}\) fitted the model output for COD to the experimental COD data obtained in experiments at three different initial dye concentrations. The model output for off-gas and dissolved ozone concentrations also fit well to the experimental data indicating the model's ability to represent the physical system well.
+
+There have been very few studies reported in the literature that has incorporated the secondary reaction of ozone with the byproducts formed from the ozonation of primary compound using a gross parameter for characterizing the byproducts. Wu and Masten [28] used a first order reaction with respect to ozone to take into account the competitive reaction involving ozone and the byproducts formed during the ozonation of phenolic and indolic compounds. They obtained values of \(350\)-\(400\) s\({}^{-1}\) for phenols, \(400\)-\(450\) s\({}^{-1}\) for indoles, and \(40{,}000\) s\({}^{-1}\) for stored swine manure slurry water. Separate rate expressions are often used to take into account the secondary reactions of ozone with primary products of ozonation when the type of byproducts and their reaction kinetics with ozone are known. For example, Singer and Gural [29] identified catechol, hydroquinone, and _cis_-muconic acid to be the significant products formed from ozonation of phenol in an acidic medium, and incorporated separate rate expressions for phenol and intermediate compounds after determining reaction rate constants for individual intermediates. For any real wastewater or intermediates formed during ozonation of synthetic water that contains many complex compounds, a gross parameter such as COD may be the effective way to take into consideration secondary reaction of ozone with the degradation products of primary compounds [28]. In a study conducted on ozonation of an azo dye, Orange II, in batch experiments, Liakou et al. [30], adopted a holistic approach by developing the kinetics in terms of measurable quantities such COD and BOD. In this approach, it was not necessary to have detailed knowledge of the byproducts and their reaction kinetics with ozone. They estimated a value of \(1.38\times 10^{-2}\) L/g. sec. for reaction rate constant for reaction between COD and ozone that included the stoichiometric factor. Though the corresponding value obtained in this study for RB-19 ozonation is not directly comparable to their results, it may be noted that the values are in the same order of magnitude.
+
+### Enhancement (\(E\)) and depletion (\(D\)) factors
+
+The values for \(E\) and \(D\) at the beginning reaction are listed in Table 2. \(E\) and \(D\) values vary as a function of the initial dye concentration as the reaction regime is determined by the Ha number. For the three concentrations studied, the values of \(Ha\) are 0.52, 0.74, and 1.05 for 40 mg/L, 80 mg/L, and 160 mg/L respectively. These values of \(0.3<Ha\times 3\) fall in the intermediate regime, and the reaction can be expected to occur partly in the film and partly in the bulk liquid. The extent of reaction in the film can be obtained as \((E-D)|E\). Hence, the extent of reaction occurring in the film ranges from 10%, 21%, and 38% for initial dye concentrations of 40 mg/L, 80 mg/L, and 160 mg/L respectively. The value of \(E\) will decrease with time to 1.0, and the value of \(D\) will increase to 1.0 due to the decrease in concentration of the dye in the recycle reactor. That is, as reaction time progresses, a higher and higher proportion of the reaction will occur in the bulk liquid. In a continuous flow tubular _in situ_ reactor with no recycle, \(E\) will decrease along the length of the reactor, whereas \(D\) will increase along the length of the reactor. In cases where the reaction rate constants are very high, as in the case of reactions involving hydroxyl radicals, the \(Ha\) number can be sufficiently high (\(Ha>3\)) for the reaction to occur completely in the film. In addition, as the thickness of the hydrodynamic film increases Ha will increase, and more of the reaction can be expected to occur in the film. More detailed analysis of the variation of \(E\) and \(D\) as the reaction progresses, and the effect of \(Ha\) are given elsewhere [31].
+
+### Model simulation and sensitivity analysis
+
+The ozone absorption and reaction model was used to predict the concentration of dissolved ozone, off-gas, dye, and COD concentrations, and the results along with experimental data are shown in Fig. 7 for an initial dye concentration of 40 mg/L. As evident from Fig. 7, the model predictions for the dye, COD, and dissolved ozone concentration profiles are quite close to the experimental results. The model prediction for off-gas ozone showed a maximum deviation of 14% from the experimental data. Considering the variation in generation rates of ozone in any electrical discharge based ozone generator, the deviation between the model and the experimental result is quite reasonable. However, it was noticed from the fluctuations in the off-gas concentration profile that the low gas flow rate of 0.02 L/min might have affected the ozone generation rate in the _in situ_ generator. The simultaneous accurate prediction of the concentration profiles indicate that the model represents the physical system well and can be used for predicting dye decolorization and COD reduction with time in the tubular _in situ_ reactor.
+
+Several absorption and reaction parameters used in the model were varied over a range (\(\pm 20\)% to \(\pm 50\)%) of its base value while maintaining the remaining parameters constant, to determine the the effect of uncertainty in parameter values on the simulation results for dye, COD, dissolved and off-gas ozone concentrations. As seen in Figs. 8 and 9, uncertainty (\(\pm 20\)%) in the absorption parameters \(\alpha\) and \(K_{\mathrm{d}}\), can significantly affect the predicted COD and dissolved ozone profiles. The predicted concentration profiles are also quite sensitive the reaction stoichiometric coefficients \(z_{1}\), \(z_{2}\), and \(z_{3}\) (Figs. 10, 11, and 12). In the case of \(z_{4}\) the stoichiometric coefficient for the secondary products, uncertainty ((\(\pm 20\)%) in parameter values mainly affects the COD concentration profile during the secondary product formation time period (Fig. 13). The predicted profiles are not highly sensitive to the decolorization rate constant \(k_{1}\) (Fig. 14) and the second ozonation reaction rate constant \(k_{2}\) (Fig. 15). The sensitivity analyses indicate that the absorption parameters and the reaction stoichiometry parameters must be determined reasonably accurately to obtain good model predictions. A well calibrated model will be useful in optimizing system design for ozonation systems and in effective operation of such systems.
+
+### Effect of initial dye concentration on dye ozonation kinetics
+
+Ozonation of RB-19 dye was conducted in the _in situ_ reactor at three different initial dye concentrations of 40, 80, and 160 mg/L with the ozone generation rate of \(46.6\pm 0.5\) g/Nm\({}^{3}\), pH of 2, and feed gas flow rate of 0.05 L/min. The reaction rate constant valuesFigure 8: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to partition coefficient (a) (pH 2, \(C\)00 = 40 mg/L, _G_m = 46.6 g/Nm3; _Q_m = 0.05 L/min).
+
+Figure 7: Comparison of experimental and predicted dissolved ozone, off-gas, dye, and COD concentrations in the _in situ_ reactor (_C_00 = 40 mg/L, \(Q\)0 = 0.1 L/min, _Q_e = 0.02 L/min (10/2 kPa), pH = 2, _G_m = 47.5 g/Nm3, and \(T\) = 21 °C).
+
+Figure 9: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to mass transfer coefficient (_K_i_,_d_) (pH 2, \(C\)00 = 40 mg/L, _G_m = 46.6 g/Nm3; _Q_m = 0.05 L/min).
+
+Figure 11: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to stoichiometric factor, \(z\)2 (pH 2, \(C\)00 = 40 mg/L, _C_n = 46.6 g/Nm3: \(Q_{d} = 0.05\) L/min).
+
+Figure 12: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to stoichiometric factor, \(z\)3 (pH 2, \(C\)00 = 40 mg/L, _C_n = 46.6 g/Nm3: \(Q_{d} = 0.05\) L/min).
+
+Figure 10: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to stoichiometric factor, \(z\)1 (pH 2, \(C\)00 = 40 mg/L, _C_n = 46.6 g/Nm3: \(Q_{d} = 0.05\) L/min).
+
+Figure 14: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to reaction rate constant, \(k\)1 (pH 2, _C_no * 40 mg/L, _G_m * 46.6 g/Nm3: \(Q_{g}\) = 0.05 L/min).
+
+Figure 13: Sensitivity of dye, COD, dissolved ozone and off gas concentration profiles to stoichiometric factor, _z_a (pH 2, _C_no * 40 mg/L, _G_m * 46.6 g/Nm2: \(Q_{g}\) = 0.05 L/min).
+
+for \(k_{1}\) and \(k_{2}\) obtained from the data are consistent (Table 2) and invariant with the initial dye concentration. Several researches have used a pseudo-first order rate constant for reaction of ozone with the dye, and in these cases the rate constant was reported to decrease with increase in dye concentrations [23,24,32,33]. Chu and Ma [21] studied the ozonation of two azo dyes and one anthraquinone dye at three initial dye concentrations of 14.4, 7.2 and \(3.6\times 10^{-2}\) mmol/L. The pseudo-first order rate constant decreased in all the cases but did not follow any specific pattern. They conducted experiments in excess ozone and expressed the decolorization kinetics as a pseudo-first order reaction with respect to dye. Higher secondary reaction of ozone with increased availability of byproducts at increased dye concentrations makes ozone less available for reaction with parent dye compounds. This resulted in a decrease in the pseudo-first order rate constant as it is dependent on the concentration of ozone. The work presented in this paper provides a better framework for modeling reactions of ozone with high-molecular weight organic compounds where secondary reactions occur at a slower rate than the primary reaction. Moreover, the proper inclusion of the steps of mass transfer and reaction in the hydrodynamic film is important in determining the true intrinsic rate constants.
+
+There are not many studies available in literature dealing with the effect of initial dye concentration on the second order reaction rate constant. Beltran and Alvarez [34], studied the ozonation of phenol and three azo dyes (Direct Yellow 27, Direct Blue 1, and Black Acid 52) in an agitated cell and determined the second order rate constant using the kinetic equations derived from Danckwerts' model for gas-liquid absorption. As shown in their article, the rate constant did not follow a consistent pattern to make conclusions on the dependency of the rate constant on dye concentration. For the dye Direct Yellow 27, the rate constant decreased from \(5.97\times 10^{2}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) to \(4.85\times 10^{2}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) when the dye concentration was doubled from a value of 0.25 mol m\({}^{-3}\). A direct relationship between rate constant and dye concentration was observed in case of Direct Blue 1 (\(9.5\times 10^{4}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) for \(C=3.5\) mol m\({}^{-3}\) and \(6.38\times 10^{4}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) for \(C=0.7\) mol m\({}^{-3}\)). No significant effect was found for Black Acid 52 (\(5.24\times 10^{4}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) for \(C=1.05\) mol m\({}^{-3}\) and \(5.66\times 10^{4}\) m\({}^{3}\) mol\({}^{-1}\) s\({}^{-1}\) for \(C=0.87\) mol m\({}^{-3}\)). The intrinsic second order reaction rate constant should be a constant and invariant with changes in dye concentration. For fast reactions, as is the case with many dye decolorization reactions, the overall rate of reaction is controlled by the gas absorption rate. If the dye diffusion rate is sufficiently high, the dye concentration in the hydrodynamic film will be the same as in the bulk liquid, and the enhancement factor will vary as the square root of initial dye concentration [14]. If the intrinsic reaction rate is quite low (\(Ha<0.02\)), all of the reaction will occur in the bulk liquid, and both dissolved ozone and off-gas ozone concentrations will steadily increase. This is what is observed in the case of the secondary reaction of ozone with the primary products of the decolorization reaction. In a mass transfer limited reaction regime, the chemical reaction rate depends on the amount of ozone transferred from the gas phase to the liquid phase. Increasing the dye concentration increases the enhancement factor, and thus the amount of ozone transferred to water. This increases the initial rate of dye decolorization and the net removal of dye within a certain period of ozonation. Ignoring these effects can result in conclusions that are conflicting and cloudy. The results obtained in this study for the intrinsic rate constant values do not show any significant dependency of the rate constant on the initial dye concentration.
+
+## 5 Conclusions
+
+The absorption and reaction kinetics of ozonation of an anthraquinone dye, Reactive Blue-19, was studied experimentally in a novel tubular _in situ_ ozone generator. In this reactor, ozone is generated _in situ_ around the periphery of a tubular porous electrode, and the generated ozone diffuses immediately and reacts with the contaminant in the fluid flowing through tubular reactor. A comprehensive mathematical model was developed to represent absorption of ozone and reaction with the dye. The model developed considers ozone absorption and reaction in the hydrodynamic film and bulk liquid. The reaction kinetics was represented by two parallel reactions, one of ozone with the dye, and a second reaction of ozone with the primary products from the first reaction. The primary and secondary products of ozonation were represented by chemical oxygen demand as the surrogate parameter. Stoichiometric constants for the reactions, and reaction kinetic constants were obtained from experimental data for several initial dye concentrations. The stoichiometric coefficient value obtained for \(z_{1}\) of 8.5 mg dye/mg O\({}_{3}\), is consistent with the value of 4.3 mg dye/mg O\({}_{3}\) obtained for R016 dye by Tizaoui and Grima [27]. Stoichiometric coefficients were also estimated for the secondary reaction of ozone with the primary products. The predicted COD profiles were found to be quite sensitive to the stoichiometric coefficient values. The rate constants obtained were found to be invariant with changes in initial dye concentrations. At low initial dye concentrations very little of the reaction occurred in the hydrodynamic film, and most of the reaction occurred in the bulk liquid. However, as the initial dye concentration increased, a higher proportion of the primary decolorization reaction was found to occur in the film. The profiles for dye, dissolved ozone, off-gas ozone, and COD were not found to be sensitive to changes in the rate constant \(k_{1}\). The latter three profiles were somewhat sensitive to the secondary reaction rate constant, \(k_{2}\). The model developed can be used for optimization of the _in situ_ tubular electrode system, and in choosing appropriate contacts for ozonation depending whether the reaction occurs in the film or the bulk liquid. The methodology developed can be used to better characterize reaction kinetics in dye ozonation systems.
+
+## Acknowledgment
+
+The authors acknowledge and thank the U.S. National Science Foundation for partial financial support for this research (Grant No. BES-0209343).
+
+## References
+
+* (1) C. Baronti, R. Curini, G. D'Ascenzo, A. Di Corcia, A. Gentili, R. Samperi, Monitoring natural and synthetic estrogens at activated sludge sewage treatment plants and in a receiving water, Environ. Sci. Technol. 34 (2000) 5059-5066.
+* (2) M.D. Cargouet, A. Perdiz, A. Moutassim-Souali, S. Tamiser-Karolx, V. Levi, Assessment of river contamination by estrogenic compounds in Paris area (France), Sci. Total Environ. 324 (2004) 55-66.
+* (3) W.FLM. Hoebe, E.M. van Veldhuizen, C.M.W. Kroesen, Gas phase corona discharge for oxidation of phenol in an aqueous solution, J. Phys. D 32 (1999) L133-L137.
+* (4) W.T. Stin, S. Yaicoumi, C. Tsouris, S. Dai, A pulseless corona-discharge process for the oxidation of organic compounds in water, Ind. Eng. Chem. Res. 39 (2000) 4408-4414.
+* (5) N. Sano, T. Kawashima, J. Fujikawa, T. Fujimoto, T. Kitai, T. Kanki, Decomposition of organic compounds in water by direct contact of corona discharge: influence of discharge conditions, Ind. Eng. Chem. Res. 41 (2002) 5906-5911.
+* (6) AK Sharma, D.M. Carnaioli, G.B. Josephson, S.C. Goheen, Destruction of pentachlorophenol using glow discharge plasma process, Environ. Sci. Technol. 34 (2000) 2267-2272.
+* (7) KK Fanda, A.P. Mathews, Mass transfer of ozone in a novel in situ ozone generator and reactor, J. Environ. Eng. 134 (2008) 860-869 (ASCE).
+* (8) K. Padmanabhan, S. Ananthi, K. Patel, Portable ozone generator, Electron. You 32 (2000) 645-51.
+* (9) KL Klabenes, R.C. Renner, B.A. Hegg, A.C. Hill, Practical design model for calculating bubble diffuser cofactor ozone transfer efficiency, Ozone: Sci. Eng. 10 (1988) 173-214.
+
+* (10) J.M. Fanching, D.H. Tseng, Degradation of anthraquinone dye C1. Reactive Blue tip in aqueous solution by ozonation, Chemosphere 77 (2009) 214-221.
+* (11) J. Dudley, Mass transfer in bubble columns: a comparison of correlations, Water Res. 29 (1995) 1129-1138.
+* (12) G. Huglimark, Holoph and mass transfer in bubble columns, Ind. Eng. Chem. Process Des. Rev. 6 (1967) 218-220.
+* (13) P.V. Dandekwers, Gas-Liquid Reactions, McGraw-Hill, New York, USA, 1970.
+* (14) J.C. Charpentier, Mass transfer rates in gas liquid absorbers and reactors, in: Advances in Chemical Engineering, Academic press, USA, 1981, pp. 1-133 (Vol. 11).
+* (15) S. Schluter, T. Schulzke, Modeling mass transfer accompanied by fast chemical reactions in gas-liquid reactors, Chem. Eng. Technol. 22 (1999) 742-746.
+* (16) H. Benbelaekem, H. Debeliefendiene, Modeling of a gas-liquid reactor in batch conditions, Study of the intermediate regime when part of the reaction occurs within the film and part within the bulk, Chem. Eng. Process. 42 (2003) 713-732.
+* (17) H. Zollinger, Color chemistry: Synthesis, Properties, and Applications of Organic Dyes and Pigments, third ed., Wiley-VCH, Weinheim, Germany, 2003.
+* (18) D. Deng, J. Guo, G. Zeng, G. Sun, Dekolfoization of anthraquinone, triphenylmethmethane and azo dyes by a new isolated Bacillus cereus strain DC11, Int. Biodoter. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod. Biod Biod

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2015.04.070.mmd

@@ -0,0 +1,207 @@
+## Accepted Manuscript
+
+Oxidation of microcystin-LR in water by ozone combined with UV radiation: The removal and degradation pathway
+
+Jing Chang, Zhong-lin Chen, Zhe Wang, Jing Kang, Qian Chen, Lei Yuan, Jimin Shen
+
+PII: S1385-8947(15)00552-5 DOI: [http://dx.doi.org/10.1016/j.cej.2015.04.070](http://dx.doi.org/10.1016/j.cej.2015.04.070)
+
+Reference: CEJ 13559
+
+To appear in: _Chemical Engineering Journal_
+
+Received Date: 22 December 2014
+
+Revised Date: 3 April 2015
+
+Accepted Date: 11 April 2015
+
+Please cite this article as: J. Chang, Z-l. Chen, Z. Wang, J. Kang, Q. Chen, L. Yuan, J-m. Shen, Oxidation of microcystin-LR in water by ozone combined with UV radiation: The removal and degradation pathway, _Chemical Engineering Journal_ (2015), doi: [http://dx.doi.org/10.1016/j.cej.2015.04.070](http://dx.doi.org/10.1016/j.cej.2015.04.070)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+Oxidation of microcystis-LR in water by ozone combined with UV radiation: The removal and degradation pathway
+
+Jing Chang, Zhong-lin Chen, Zhe Wang, Jing Kang, Qian Chen, Lei Yuan, Ji-min Shen
+
+State Key Laboratory of Urban Water Resource and Environment, School of Municipal & Environmental Engineering, Harbin Institute of Technology, Harbin 150090, China
+
+###### Abstract
+
+In this study, the performance of the combined UV/O\({}_{3}\) process for the degradation of microcystis-LR (MC-LR) in water was investigated, and the possible degradation pathway of the toxin was elucidated according to reaction intermediates. Results demonstrated that the UV/O\({}_{3}\) process was a more effective method for the removal and mineralization of MC-LR in water, compared with UV and O\({}_{3}\)-alone processes. Increasing the solution pH (\(5-10\)) and dissolved organic carbon concentration (\(0-5.3\) mg/L) could inhibit the degradation of the toxin. Nine reaction intermediates of MC-LR were detected by LC/MS, and four initial reaction sites of the toxin were observed: the conjugated diene of Adda, the double bonds of Mdha and two free acid groups of MeAsp and Glu. The degradation pathways of the toxin during UV/O\({}_{3}\) process involved isomerization, hydroxylation and oxidative cleavage of the Adda side chain, oxidation of Mdha and decarboxylation of MeAsp and Glu. The pathway of the oxidation and cleavage of Adda and Mdha could result from both HO\({}^{\star}\) and O\({}_{3}\) oxidation, but the isomerization of Adda moiety and the 
+
+###### Contents
+
+* 1 Introduction
+* 2 Microcystins (MCs) are a group of monocyclic heptapeptide hepatotoxins produced by the toxic species of cyanobacteria [1]. With occurrences of cyanobacteria (blue-green algae) blooms, MC contamination in water becomes a serious aqueous environmental problem [2-5]. In many freshwaters, the toxin production can persist throughout the year [2, 4, 5] and can occur at levels [33 (microgram per liter to milligram per liter) that cause human health problems [2, 5, 6]. Thus far, more than 90 variants of MCs have been identified [7]. The general structure of MCs contains [35] D-alanine, D-_erythro-b_-methylaspartic acid, Adda, \(\gamma\)-linked-D-glutamic acid, _N_-methyl dihydroalanine and two variable L-amino acids [8]. Adda is an unusual C20 amino acid [37] (3-amino-9-methoxy- 2,6,8-trimethyl-10-phenyl-4(E),6(E)-dienoic acid) and is nontoxic [8]. However, the Adda amino acid is essential for the expression of toxicity of MCs [1].
+* 3omerization/oxidation of the Adda side chain could significantly decrease the toxicity associated with MCs or generate nontoxic products [9-12].
+* 41 Among the MCs, microcystin-LR (MC-LR) has been given more attention because it is common in eutrophic water and has an acute toxicity (LD\({}_{50}\): 50 \(\mu\)g/kg, mice) [11]. The World Health Organization (WHO) has recommended a guideline value of 1 \(\mu\)g/L for MC-LR in drinking water[13]. However, traditional water treatment processes, such as coagulation/flocculation, clarification or filtration, have limited ability to remove extracellular MC-LR from water [14, 15]. Therefore, more alternatives have been studied to remove the toxin from water [7, 16]. Advanced oxidation processes (AOPs) have received significant attention because they can generate a strong oxidant hydroxyl radical (HO') that can non-selectively oxidize a wide variety of organic matter with a high reaction rate constant [17-20]. The performances of AOPs on the degradation and detoxification of MCs have been widely investigated: examples include UV/TiO2, ultrasonic irradiation, UV/H2O2, O3/Fe(II) and Fenton [12, 21-25].
+* Except for the above AOPs, the combined UV254nm/O3 process is another attractive option because of its effective oxidation and destruction of toxic and refractory organics, bacteria and viruses in water [18]. UV/O3 technology involves multiple chemical reactions for the degradation of pollutants [26]. The main reactions involve radical (HO') oxidation, direct reaction with ozone and direct UV photolysis [26-28]. The HO is the principal active species in photolytic ozonation and its generation involves several chemical processes [26, 28]. The photolysis of aqueous ozone directly produces hydrogen peroxide as an initiation step (reaction (1)) [28]. Then, HO' can be formed through the following chemical processes: UV photolysis of hydrogen peroxide (reaction (2)), and the decomposition of O3 initiated by peroxide and hydroxide ions (reactions (3)-(9)) [20, 28, 29]. Thus, the UV/O3 process also involves the behaviors of UV/H2O2 and O3/H2O2 processes, but of the three; the UV/O3 process provides the maximum yield of HO' per oxidant [26].
+* O3 + H2O + _hv_ → O2 + H2O2
+* H2O2 + _hv_ → 2 HO'
+* O3 + H2O2 → HO2~+ HO' + O2
+* H2O2 + HO2~+ H+
+* \(k\) = 1.0 x 10-2 M-1S-1
+* H2O2 + HO2~+ H+
+* _pK_a_ = 11.6\[\begin{array}{ll}67&{\rm O_{3}+OH}\rightarrow{\rm HO_{2}}^{-}+{\rm O_{2}}&k=70 \;{\rm M^{-1}S^{-1}}\end{array} \tag{5}\]
+* \[\begin{array}{ll}68&{\rm O_{3}+HO_{2}}^{-}\rightarrow{\rm O_{2}}^{-}+{\rm HO^{ \prime}}+{\rm O_{2}}&k=5.5\times 10^{6}\;{\rm M^{-1}S^{-1}}\end{array}\] (6)
+* \[\begin{array}{ll}69&{\rm O_{3}+O_{2}}^{-}\rightarrow{\rm O_{3}}^{-}+{\rm O_{2 }}&k=1.6\times 10^{9}\;{\rm M^{-1}S^{-1}}\end{array}\] (7)
+* \[\begin{array}{ll}70&{\rm O_{3}}^{-}+{\rm H^{+}}\rightarrow{\rm HO_{3}}^{*} &k=5\times 10^{10}\;{\rm M^{-1}S^{-1}}\end{array}\] (8)
+* \[\begin{array}{ll}71&{\rm HO_{3}}^{*}\rightarrow{\rm HO^{\prime}}+{\rm O_{2 }}&k=1.4\times 10^{5}\;{\rm M^{-1}S^{-1}}\end{array}\] (9)
+
+Many researchers have investigated the performances of the individual UV and ozone processes for the degradation of MC-LR [30]. The study by Tsuji et al.[31] showed that 10 mg/L of MC-LR in distilled water could be decomposed completely using a UV-C lamp with 2.55 mW/cm\({}^{2}\) for 10 min. The decomposition was accompanied by isomerizations of the Adda moiety of MC-LR, and the isomers of MC-LR could be further photolyzed [31, 32]. Ozone oxidation is an effective method to remove MC-LR from water within seconds to minutes, which are dependent on the water quality parameters [33]. Rositano et al. [34] found that at pH = 7, 0.02 mg/L and 0.22 mg/L of ozone could remove 27% and 100% of MC-LR (1 mg/L) in pure water, and both of the reactions took only 15 s. However, in natural water, the ozone of MC-LR could be significantly affected by different water quality, and it seems that the presence of an ozone residual is necessary for complete removal of the toxin [35, 37]. Differing from UV radiation, the oxidation of MC-LR by ozone was initiated by oxidation and cleavage of the double bonds in Adda and Mdha amino acids [38, 39]. For the combined UV/O\({}_{3}\) process, Liu et al. [40] investigated its effectiveness in removal of MC-LR in litrated water (pH: 7.0 - 7.6; total organic carbon: 3.2 - 4.7 mg/L; alkalinity as CaCO\({}_{3}\): 32 - 45 mg/L) from a waterworks in Harbin, China. Results showed that approximate 80% of MC-LR (100 mg/L) could be removed after 1.0 min with an initial ozone concentration of 1.0 mg/L and a UV (254 nm) intensity of 2.6 mW/cm\({}^{2}\), while 60% and 25% of MC-LR were degraded in UV- and O\({}_{3}\)-alone processes.
+
+Based on above, the study on oxidation of MC-LR by the combined UV/O\({}_{3}\) process is limited.
+* 91 As an attractive water treatment technology, it is necessary to evaluate its potential to remove MCs from water. In addition, the degradation pathway of the toxin during treatment is also necessary to be investigated. Because of the multiple chemical reactions in the UV/O\({}_{3}\) process, the products of photolysis can be further oxidized by HO' and O\({}_{3}\) and at the same time, the oxidation products of HO' and O\({}_{3}\) can be further photolyzed [27]. Therefore, the degradation pathways of the toxin during the UV/O\({}_{3}\) process may not be necessarily a combination of the pathways in the two individual processes. Based on above, in this study, the performance of the combined UV/O\({}_{3}\) process for the degradation of MC-LR was investigated, and the results were compared with that obtained from treatment by ozone and UV alone. The comparison involved the removal efficiency of MC-LR, the influence of the initial solution pH and dissolved organic carbon (DOC) concentration, the mineralization of MC-LR and the reaction intermediates. In addition, the possible degradation pathway of MC-LR under the combined UV/O\({}_{3}\) process was proposed.
+* 92 Materials and methods
+* 93 Chemicals and materials
+* 94 The stock solution of MC-LR was prepared by dissolving 1mg of solid toxin (\(\geq\) 95%, Enzo, USA) in 20 mL of methanol (HPLC grade, J&K, China). The stock solution was stored at -20 \({}^{\circ}\)C and was stable for up to 6 months. Before the experiments, a specific volume of the stock solution was dried by a gentle N\({}_{2}\) stream at 50 \({}^{\circ}\)C and then redissolved in the same volume of ultrapure water (Millipore Corp, USA). The concentration of MC-LR in ultrapure water was measured and an appropriate volume of the stock solution was added to the reaction solution. A commercial humic acid purchased from Tianjin Guangfu Fine Chemical Research Institute, China, was purified by repeated pH adjustment, precipitation, and centrifugation [41]. Then, the stock solution of humic acid was prepared as described by Shawwa and Smith [33]. All organic solvents used for analysis were HPLC grade (J&K, China). All other chemical reagents, such as hydrochloric acid, sodium hydroxide and sodium thiosulfate, were of analytical grade without further purification (Tianjin Benchmark, China). All solutions were prepared with ultrapure water.
+
+### 2.2. UV/O3 experiments
+
+All experiments were conducted in a locally fabricated photo-reaction apparatus [22]. A low-pressure mercury vapor UV lamp (8W, CREATOR, China) which emitted light primarily at \(\lambda\) = 253.7 nm was used. The UV lamp was turned on to warm up for 30 min to ensure stable output before the experiments. The reactor was a 13-mL glass dish(internal diameter 3.4 cm) with a quartz cover. The distance from the UV lamp to the surface of reaction solution was 5 cm, and the UV light intensity (at 5 cm) through the quartz cover was measured by a calibrated UV-254 radiometer (Photoelectric Instrument Factory of Being Normal University, China). The average UV intensity in the water could be obtained after several corrections, according to the study by Bolton and Linden [42]. The average UV dose in the water was calculated as average UV dose (mJ/cm\({}^{2}\)) = average UV intensity (mW/cm\({}^{2}\)) x reaction time (s) [42]. The temperature in the apparatus was \(30\pm 2^{\circ}\)C.
+
+Before the experiments, stock solutions of aqueous ozone were obtained by bubbling ozone-containing oxygen through ultrapure water (1 L, 4 \({}^{\circ}\)C) on a magnetic stirrer. The ozone concentration of the stock solution was monitored at 258 nm with a UV spectrophotometer [35]. The stock solutions containing steady-state ozone concentrations of 10 mg/L and of 44 mg/L were prepared for different experiments in this study. The range of the volumes of the stock solutions added to the reaction solution was \(0.025-9\) mL. The residual ozone of the sample was determined by the indigo method [43] and the method detection limit was 5.8 ug/L. The experiments were conducted as follows: (1) for the combined UV/O3 process, the reaction solution containing MC-LR was added to the reactor and placed under the UV lamp, and a calculated volume of the ozone stock solution was simultaneously added beneath the surface of the solution and then covered with the quartz cover; (2) for the UV-alone process, the procedure was the same as described above, but no ozone stock solution was added; and (3) for the O3-alone process, the UV lamp was turned off, and the experiment was conducted as described in the combined UV/O3 process. The [O3]0 / [MC-LR]0 (mol/mol) ratios used ranged from 1 to 321 in this study. The range of initial concentration of MC-LR was 1 - 10 mg/L, and of O3 was 0 - 39 mg/L. The initial pH value of the reaction solution without pH adjusted was 5.9 +- 0.1. Hydrochloric acid and sodium hydroxide were used to adjust pH when the influence of initial pH (5 - 10) was investigated. Different initial DOC concentrations (0.5 - 5.3 mg/L) were prepared by adding appropriate volumes of the humic acid stock solution to the reaction solutions. For all above experiments, the total volume of the reaction solution was 10 mL. The reactions involving ozone were quenched by the addition of 100 mL of sodium thiosulfate (0.1 mol/L). Each experiment was repeated three times.
+150 2.3 Analysis
+151 MC-LR was analyzed with high performance liquid chromatography (HPLC) (Agilent 1200LC). HPLC was equipped with a quaternary pump and a UV detector, and MC-LR was measured at 238 nm. The HPLC column was Zorbax Eclipse XDB-C18 (4.6 x 150 mm, 5 mm) (Agilent, USA). The mobile phase was water containing 0.05% trifluoroacetic acid (v/v) and acetonitrile, and the ratio was 67:33. The injection volume of the sample was 100 mL and the flow rate was 1 mL/min. The method detection limit for MC-LR was 5.93 ug/L.
+
+Reaction intermediates were analyzed with liquid chromatography / mass spectrometry (LC/MS). The LC column was Eclipse Plus-C18 (2.1 x 150 mm, 3.5 mm) (Agilent, USA) with a guard column Eclipse Plus-C18 (2.1 x 12.5 mm, 5 mm) (Agilent, USA). The mobile phase was methanol and water, both containing 0.05% (v/v) acetic acid. A gradient elution was used as follows: 0 - 2 min, 161 30% methanol; 2 -12 min, 30% - 80% methanol; 12 - 20 min, 80% methanol; 20 - 22 min, 80% - 30% methanol; 22 - 31 min, 30% methanol. Injection volume of the sample was 25 mL A Thermo 163 Finnigan LCQ Deca XP Plus (PDA / ESI) system was used for MS and MS analysis. Mass spectra data of MC-LR (molecular weight: 994, C_a_H_4N_10O_12, Fig.1) and its reaction intermediates were obtained in the positive ion mode by full scan from _m/z_ 200 to 1200. The other details on HPLC and LC/MS analysis can be found in an earlier study [39].
+
+The total organic carbon (TOC) concentration of the reaction solution was measured by a total organic carbon analyzer (Multi N/C 2100s, Analytikjena, Germany). The details can be found in the supplementary materials. The pH value of the solution was measured by pH 720 (WTW, Germany).
+
+## 3. Results and discussion
+
+### 3.1. Degradation of MC-LR by the UV/O3 process
+
+The oxidation of MC-LR (1 mg/L) in pure water by the UV/O3 process is shown in Fig. 2, where the average UV light intensity in the water was 1.88 mW/cm2 and the initial molar ratios of O3 to MC-LR ([O3]o/ [MC-LR]o) were 1.0 and 1.58 (O3: 48 mg/L and 76 mg/L). Results showed that under the same condition, the combined UV/O3 process was more effective for MC-LR degradation compared with the UV- and O3-alone processes. In the UV-alone process, MC-LR was gradually degraded from 39% at 0.5 min to 66% at 5 min. However, in the O3-alone process, the oxidation of MC-LR could complete within a very short time, as observed by other studies [33, 34]. Approximate 54% and 72% of MC-LR were removed within 0.5 min when the used initial ozone concentrations were 48 mg/L and 76 mg/L. For the combined process, with an ozone dose of 76 mg/L, the UV/O3 process could rapidly degrade 81% of MC-LR within 0.5 min; then the degradation of MC-LR was insignificant. This was because that the low ozone concentration was used and ozone was completely consumed at the beginning of the reaction. Therefore, the immediate drop of the concentration of MC-LR resulted from the degradation by UV, O3 and HO'. Then, the degradation of MC-LR was insignificant which mainly attributed to UV radiation only. Results in Fig.S1 showed that with an initial ozone concentration of 9.125 mg/L ([O3]6/[MC-LR]6=2.5), more than 99.5% of MC-LR (1 mg/L) could be degraded by the UV/O3 process within 1.5 min. Compared to other AOPs (UV/H2O2 or UV/TiO2) that involve photochemical processes, the capital and operational costs of the UV/O3 process are higher due to the production of ozone. However, the required reaction times of this process are very short and the required oxidant doses are much lower (compared to H2O2) [22, 44, 45, 46], which may be the advantages of the UV/O3 process used to remove the toxin from water.
+* 3.1.2 The effect of solution pH
+* 10) on the degradation of MC-LR by the UV/O3 process is investigated, as shown in Fig. 3. It was observed that as the pH value increased from 5 to 10, the degradation efficiency of MC-LR was decreased by 10%. This phenomenon was more obvious in O3-alone process, which was consistent with previous studies [23, 33, 34, 49]. But in UV-alone process, the degradation of MC-LR differed insignificantly. Therefore, as the pH value increased, the decrease in UV/O3 effectiveness mainly resulted from the effect of pH on the ozone
+process. For ozone based processes, the solution pH determines the stability of dissolved ozone molecules in water and also indicates the manner in which the ozone process occurs [50]. In both UV/O3 and O3 processes, the higher pH values promote the decomposition of O3 and the generation of HO*, according to reactions (4) - (9) [28, 29]. The HO* is a stronger oxidant and the second order rate constant of HO* with MC-LR is 5 orders of magnitude higher than O3[35]. However, in Fig. 3, the removal efficiency of MC-LR decreased as pH increased in these two processes, especially in O3-alone process. Previous studies also observed the phenomenon in O3-alone process and the authors concluded that MC-LR probably reacts with ozone as opposed to hydroxyl radicals [23, 34, 49]. One possible reason is that molecular ozone is a selective oxidant and it has a high selectivity towards the double bonds in Adda and Mdha amino acids of MC-LR [35]. In contrast, the HO* oxidation is non-selective, which indicated that the HO* is more likely to be consumed by other substances, e.g., the reaction intermediates [51]. Therefore, with increasing the pH value, the reduced efficiency of UV/O3 possibly resulted from the fact that the direct ozone reaction was inhibited.
+* 3.1.3 The effect of DOC concentration
+* 3.1.3 The effect of DOC concentration
+* 3.1.3 Results in Fig. 2 showed that MC-LR was efficiently removed by UV/O3 in pure water. However, MC-LR exists in real water mixed with other natural organic matter (NOM), which has a significant influence on the reactions of HO*, UV and O3 processes [23, 33, 35, 52, 53, 54, 55]. Thus, the effects of DOC concentration on MC-LR degradation by the UV/O3 process were investigated, as shown in Fig. 4. Humic acid is a principal component of humic substances, which are important organic constituents of NOM in natural water [56]. Humic acid generally contains rich unsaturated structures that could be preferentially oxidized by HO* and O3 and have ultraviolet absorbance at 254 nm [53]. Thus, humic acid can compete with MC-LR for the oxidants (HO* and O3) and UVabsorbance in the UV/O3 process. Results in Fig. 4 showed that with an initial ozone concentration of 1.5 mg/L, the efficiency of three processes was not significantly affected at the DOC concentrations <= 1.4 mg/L. The UV/O3 and O3-alone processes could oxidize MC-LR (1 mg/L) to below the detection limit (5.93 mg/L) and UV-alone process could remove more than 50% of MC-LR. However, the efficiency of the UV/O3 process decreased by 16% when the DOC concentration increased from 1.4 to 5.3 mg/L. This decrease was 4% in the UV-alone process and 231 42% in the O3-alone process. Moreover, data in Fig. S2 showed that for samples with DOC value of 232 5.3 mg/L, the ozone concentration that was required to oxidize MC-LR (1 mg/L) to below the detection limit increased to 4.5 mg/L in the O3-alone process and no residual ozone was detected. Results indicated that as the DOC concentration increased, significant competing reactions occurred between humic acid and MC-LR for reaction with ozone. In the UV/O3 process, the fast consumption of ozone by the humic acid decreased the generation of HO' (reactions (6) - (9)) and inhibited the reaction of HO' and MC-LR. Moreover, the direct ozonation of MC-LR could also be inhibited. Therefore, in Fig. 4, the decrease of efficiency of the UV/O3 process is mainly because much of the ozone was consumed by humic acid. Based on above, with the presence of DOC (0.5 - 5.3 mg/L), the UV/O3 process had a higher potential to remove MC-LR compared with two individual processes under the experiment conditions used (Fig. 4). However, ozone based processes were more affected by DOC, and depending on water quality, UV process may become a preferred option.
+* 3.2 The mineralization of MC-LR by the UV/O3 process
+* 3.3 Results above showed that with sufficient doses of UV and ozone, the UV/O3 process was an effective method to completely remove MC-LR from water within a very short time. However, this conclusion was based on the removal efficiency of MC-LR only. For the mineralization of MC-LR,it was relatively difficult and there were significant differences between the combined UV/O3 process and the individual processes, as shown in Fig. 5. Under the same ozone and UV doses, the UV/O3 process performed better for the mineralization of the toxin (2.5 mg/L) after 10 min of reaction compared with the two individual processes. With a UV dosage of 1128 mJ/cm2, TOC removal rates of the toxin were 8% - 38% when 90 - 167 of [O3]0 / [MC-LR]0 (initial O3: 10.8 - 20 mg/L) were used, but the UV- and O3-alone processes were ineffective. As the values of [O3]0 / [MC-LR]0 increased, the efficiency of UV/O3 gradually increased and the efficiency of O3-alone process increased significantly. As the value of [O3]0 / [MC-LR]0 increased to 321 (initial O3: 38.8 mg/L), the efficiency differences of the UV/O3 and O3 processes decreased significantly. The better performance of UV/O3 for mineralization of the toxin is mainly attributed to its multiple chemical reactions to degrade the organic matter as previously discussed, especially the oxidation by the principal active species, HO*. HO* has a high oxidation potential (2.8 V) and is less discriminating in the types of functional groups they will attack, which has advantage for the mineralization of organic substances [18, 51, 51]. However, complete mineralization of the toxin was unrealistic. One possible reason was that as the mineralization increased, the continuously formed carbon dioxide existed in water with the form of HCO3-/ CO32- and thus, the scavenging effect of them on hydroxyl radicals became more significant [58].
+
+### 3.3 The degradation pathway of MC-LR by the UV/O3 process
+
+To study the degradation pathways of MC-LR under the UV/O3 system, the reaction intermediates of MC-LR were detected by LC/MS. The intermediates from the UV- and O3-alone processes under the same conditions were also detected and compared. In total, nine intermediates(_m/z_ = 791.4, 795.4, 815.4, 827.3, 835.5, 855.3, 995.5 (two intermediates), 1029.5) were observed in the UV/O3 process, as shown in Table S1. Among these, the two intermediates with _m/z_ 995.5 were also observed in UV-alone process. Moreover, six intermediates (_m/z_ = 795.4, 815.4, 827.3, 835.5, 855.3, 1029.5) were also found in O3-alone process, and the structures (Table S1) were identified in an earlier study [39].
+* Therefore, three intermediates, with _m/z_ 995.5 (two) and 791.4, were required to be identified in this study. For the intermediates with _m/z_ 995.5, the analysis of LC/MS and photo-diode array (PDA) was conducted to assist in identification of the _m/z_ 995.5, as shown in Fig. S3 and Table S2. According to previous studies [31, 32], the two intermediates with _m/z_ 995.5 were identified as [tricyclo-Adda5]-MC-LR and [4(E), 6(Z)-Adda5]-MC-LR. They were the isomers of MC-LR resulting from the isomerization of the Adda moiety. The related details can be found in the supplementary materials. The intermediates with _m/z_ 791.4 had 44 Da of difference in molecular weight to the _m/z_ 835.5. It was possible that _m/z_ 791.4 was a decarboxylation product of _m/z_ 835.5. This finding could be supported by the analysis of MS2 spectrums of _m/z_ 791.4 (Fig. S4, Table S3 and S4). Results showed that the decarboxylation occurred at either the Glu or MeAsp amino acid. It was difficult to definitively distinguish the reaction site by the analysis of the product ions observed in MS2 spectrums because the Glu and MeAsp amino acids have the same molecular weights, and the intramolecular rearrangements of the peptides can also occur in MS2 detection [59, 60].
+* According to the above intermediates, the proposed degradation pathways of MC-LR in the presence of the UV/O3 process are shown in Fig. 6. Under UV radiation, the isomers of MC-LR formed through the isomerization of the Adda side chain, as observed by previous studies [31, 32].
+
+Both MC-LR and its isomers could be oxidized by the dihydroxylation of either C\({}_{4}\)-C\({}_{5}\) or C\({}_{6}\)-C\({}_{7}\) double bonds of the Adda side chain, and the intermediates with _m/z_ 1029.5 were formed. These dihydroxylation products could be further oxidized and cleaved at the diene structure, resulting in the formation of _m/z_ 835.5 (a type of ketone) and _m/z_ 795.4 (a type of aldehyde). As observed in previous studies, the dihydroxylation of the diene bonds followed by oxidative cleavage was a significant degradation pathway of the Adda moiety of MC-LR during AOPs [12, 24, 61, 62].
+* [299] Except for the oxidation of the Adda side chain, the oxidation of peptide ring could be initiated at three amino acids: Mdha, Glu and MeAsp, as shown in Fig. 6. Among these, the double bonds of Mdha were the main initial reaction sites in the peptide ring, as observed by several AOP studies [12, 24, 61-63]. Antoniou et al. [24] isolated abundant intermediates resulting from hydroxyl radical addition, oxidation and cleavage of the Mdha amino acid. Similar to the previous study, the oxidation of Mdha was observed in the UV/O3 process, and the three corresponding intermediates were _m/z_ 855.3, 815.4 and 827.3 (Fig. 6). As shown in Fig. 6, _m/z_ 835.5 could be oxidized to _m/z_ 855.3 by the oxidative cleavage of Mdha; then, _m/z_ 855.3 could transform to _m/z_ 815.4 by further oxidation of the residual Adda moiety. Meanwhile, _m/z_ 795.4 could also transform to _m/z_ 815.4 through Mdha oxidation. Moreover, _m/z_ 795.4 could be oxidized to another product with _m/z_ 827.3, forming a carbonyl in Mdha and a carboxylic at C\({}_{4}\) in the residual Adda moiety. Except for the oxidation of the Mdha amino acid, the decarboxylation of MeAsp and Glu were observed (Fig. 6).
+* [311] Fang et al. [63] has found the decarboxylation pathway of MeAsp and Glu during BiOBr photocatalysis of MC-LR. However, the decarboxylation pathway has not been observed in MC-LR reactions with HO' and O\({}_{3}\)[7, 64] because the free carboxyl group in the polypeptide molecule is always chemically insensitive to the oxygen-centered free radicals [63, 65] and molecular ozone [66, 67]. In this study, the decarboxylation pathway was observed and the corresponding products with * [316]_m/z_ 791.4 were formed. One possible reason was the UV radiation during the UV/O\({}_{3}\) process. According to the study by Takano et al. [68], UV radiation (wavelength over 160 nm) can cause the \(\alpha\)-decarboxylation reaction of aspartic acid and glutamic acid. Thus, the formation of _m/z_ 791.4 in this study may result from the decarboxylation of _m/z_ 835.5 under UV radiation. To support this supposition, the reaction solution after 10 min of ozonation alone was subjected to UV radiation for 321 10 min to investigate the change of intermediates. As shown in Fig. S5, UV radiation had a positive effect on the degradation of the intermediates in the UV/O\({}_{3}\) process, and _m/z_ 791.4 was gradually formed during UV radiation.
+* [324] Based on the above results, the oxidation of MC-LR by the combined UV/O\({}_{3}\) process could initiate at four sites of the toxin: the conjugated double bonds of the Adda chain, the double bonds of Mdha and the free acid groups of the McAsp and Glu amino acids (Fig. 6). The degradation pathways of MC-LR involved the isomerization, hydroxylation and oxidative cleavage of the Adda side chain, the oxidation and cleavage of Mdha, and the decarboxylation of Glu and McAsp. In comparison to the UV- and O\({}_{3}\)-alone processes, the combined UV/O\({}_{3}\) process had more pathways to degrade MC-LR, which contained pathways observed during the two individual processes [32, 38, 39]. Results indicated that the oxidative cleavage of Adda and Mdha could result from both HO' and O\({}_{3}\) reactions, but the isomerization of Adda and the decarboxylation of Glu and McAsp mainly resulted from UV radiation.
+* [334] Through the above degradation pathways, the MC-LR molecule lost the Adda side chain, opened the heptapeptide ring and transformed to linear polypeptides. Moreover, the decarboxylation of amino acids may increase the oxidation possibility of the planar peptide ring, because the loss of the free acid group could decrease the hindering effects caused by the functional groups in the peptide ring [24] and increase the chance of oxidants attacking the peptide rings. In other AOPs(UV/H2O2, UV/TiO2, and ultrasonic irradiation) [12, 24, 62], abundant hydroxyl substitution/addition products of MC-LR (e.g., _m_/z 1011.5, 1027.5, 1045.5, 1063.5) could be found and were even the dominant intermediates. They resulted from substitution reactions on the aromatic ring and addition reactions on the Adda conjugated diene and Mdha double bonds. However, these substitution/addition products of MC-LR were difficult to detect in this study, although HO' is the principal active species in the UV/O3 process. This difference could be largely attributed to the direct ozone reaction, which had a high selectivity towards the oxidation of the double bonds [35, 39]. Moreover, the HO' was more prone to oxidize C-C than C-C and C-H based on the second-order rate constants [17]. Therefore, the oxidative cleavage of the double bonds in Adda and Mdha occurred in most intermediates in this study. This can simplify or destroy the stable ring structure, which is conducive to the further degradation of these intermediates.
+* 3.3.3 The yields and degradation of the intermediates
+* 3.3.4 The yields and accumulation trends of the intermediates during 10 min of reaction are shown in Fig. 7. In the UV/O3 process, the yields of _m_/z 795.4 and 835.5 were much higher than that of the other intermediates (Fig. 7 (a) and (b)). This result indicated that among the above degradation pathways, the oxidation of the Adda side chain was the dominant degradation pathway. Compared with the Adda moiety, the oxidation of the peptide ring was relative difficult, which was consistent with results reported by other AOPs [24, 61, 62].
+* 3 min and then decreased gradually, except for the secondary product of _m_/z 791.4 which was formed through UV radiation of the primary product _m_/z 835.5. The trends of [tricyclo-Adda5]-MC-LR and [4(E), 6(Z)-Adda5]-MC-LR are not shown because of their weak
+signals in the mass spectrums. However, their trends in the UV-alone process are shown in Fig. 7 (c). [4(E), 6(Z)-Adda5]-MC-LR gave a maximum yield faster than [tricyclo-Adda5]-MC-LR. After their yields reached maximum values, the two isomers of MC-LR were further degraded with increasing duration of UV radiation. In comparison, the intermediates produced from O3-alone process had different trends (Fig. 7 (d)). The intermediates could accumulate within 1 +- 3 min, and then, the yields changed insignificantly after 3 min. This was because after 2 min of the reaction, most of the ozone was consumed and the residual ozone concentration was below the method detection limit (5.8 mg/L). Because of the low reactivity of the functional groups in the polypeptide structure (e.g., amides and guanidine) during ozonation [35, 66], the degradation of these intermediates required much higher ozone doses than that used in Fig. 7, as shown in a previous study [39]. Therefore, under the same ozone and UV doses, the UV/O3 process had a higher potential to degrade not only MC-LR but also the intermediates with high molecular weights compared with the individual processes. This was because the UV/O3 process had three main chemical reaction mechanisms (HO, O3 and UV) that could result in the degradation of the intermediates.
+
+## 4 Conclusions
+
+This study investigated the combined UV/O3 process for the removal of MC-LR and proposed the possible degradation pathway of the toxin during the treatment. Results showed that the UV/O3 process was an effective method to remove MC-LR from water, and it performed better than UV- and O3-alone processes under the same conditions. The initial solution pH and DOC values could influence the efficiency of the UV/O3 process. The degradation efficiency of MC-LR decreased with increasing pH values (from 5 to 10) and DOC concentrations (from 1.4 to 5.3 mg/L). Compared with the individual processes, the combined UV/O3 process has a significant advantage for the mineralization of MC-LR. Approximately 60% of the TOC concentration was removed by UV/O3 process under the maximum UV dosage (1128 mJ/cm2) and ozone concentration ([O3]0 / [MC-LR]0 (mol/mol) = 321) used.
+* 388 The structures of the reaction intermediates showed that the degradation of MC-LR by the UV/O3 process could be initiated at four sites of the toxin: the conjugated double bonds of the Adda chain, the double bonds of Mdha and two free acid groups of MeAsp and Glu. The possible degradation pathways of the toxin involved isomerization, hydroxylation and oxidative cleavage of the Adda side chain, the oxidative cleavage of Mdha and the decarboxylation of Glu and MeAsp.
+* 393 The oxidation of Adda and Mdha could result from both radical (HO*) reaction and direct O3 reaction, but the isomerization of Adda and the decarboxylation of the two amino acids mainly resulted from UV radiation. Through the above degradation pathways, the Adda moiety, which was essential for the expression of MC toxicity, was modified or destroyed in all of the intermediates.
+* 397 The oxidation of the Adda side chain was the dominant degradation pathway of MC-LR in this study. Moreover, the UV/O3 process had a higher potential for the simultaneous degradation of MC-LR and its intermediates than the individual processes, under the same UV and ozone doses.
+* 400 **Acknowledgement**
+* 401 This work was supported by the funds for the State Key Laboratory of Urban Water Resource and Environment (Harbin Institute of Technology) (Nos. 2014DX02, 2014TS03), the National Important Items of Science and Technology for the Control and Treatment of Water Pollution (Grant No.2014ZX07405002), National Natural Science Foundation of China (Grant No. 51208186) and National Key Technology Support Program (Grant No. 2013BAD21B01).
+* 406 **Appendix A. Supplementary data*** [408] **References**
+* [409] **R.M. Dawson, The toxicology of microcystins, Toxicon. 36 (1998) 953-962.**
+* [410] **C.D. Williams, M.T. Aubel, A.D. Chapman, P.E. D'Aluto, Identification of cyanobacterial toxins in Floridas freshwater systems, Lake Reserv. Manag. 23 (2007) 144-152.**
+* [412] **S. Bogialli, F.N. di Gregorio, L. Lucentini, E. Ferretti, M. Ottaviani, N. Ungaro, P.P. Abs, M.C. de Grazia,**
+* [413] **Management of a toxic cyanobacterium bloom (planktontrix rubescens) affecting an Italian drinking water basin: A case study, Environ. Sci. Technol. 47 (2013) 574-583.**
+* [414] **M. Ruiz, L. Galanti, A.L. Ruibal, M.I. Rodriguez, D.A. Wunderlin, M.V. Ame, First report of microcystins and anatoxin-a co-occurrence in San Roque Reservoir (Cordoba, Argentina), Water Air Soil Pollut. 224 (2013) 1593.**
+* [415] **C.M. Cook, E. Vardaka, T. Lanaras, Toxic cyanobacteria in Greek freshwaters, 1987-2000: Occurrence, toxicity, and impacts in the Mediterranean region, Acta Hydrochim. Et Hydrobiol. 32 (2004) 102-124.**
+* [416] **A. Zamyadi, S.L. MacLeod, Y. Fan, N. McQuaid, S. Dorner, S. Sauve, M. Prevost, Toxic cyanobacterial breakthrough and accumulation in a drinking water plant: A monitoring and treatment challenge, Water Res. 46 (2012) 1511-1523.**
+* Implications on human health, Anti-Cancer Agents Med. Chem. 11 (2011) 19-37.**
+* [418] **D.P. Botes, P.L. Wessels, H. Kruger, Structural studies on cyanogeniosins-LR, -YR, -YA, and -YM, peptide toxins from microcystis aeruginosa, J. Chem. Soc., Perkin Trans. L(1985) 2747-2748.**
+* [419] **K. Harada, K. Ogawa, K. Matsuura, H. Murata, M. Suzuki, M.F. Watanabe, Y. Itezono, N. Nakayama, Structural determination of geometrical-isomers of microcystis LR and RR from cyanobacterial by 2-dimensional NMR spectroscopic techniques, Chem. Res. Toxic. 3 (1990) 473-481.**
+* [420] **J.S. An, W.W. Carmichael, Use of a colorimetric protein phosphatase inhibition assay and enzyme linked immunosorbent assay for the study of microcystis and nodulararis, Toxicon. 32 (1994) 1495-1507.**
+* [421] **K.L. Rinehart, M. Namikoshi, B.W. Choi, Structure and biosynthesis of toxins from blue-green algae (cyanobacteria), J. Appl. Phycol. 6 (1994) 159-176.**
+* [422] **W.S. Zong, F. Sun, X.J. Sun, Oxidation by-products formation of microcystin-LR exposed to UV/H**2O**2: Toward the generative mechanism and biological toxicity, Water Res. 47 (2013) 3211-3219.**
+* [423] **WHO, Guidelines for drinking-water quality, 4th ed., World Health Organization, Geneva. (2011).**
+* treatment processes
+- laboratory and pilot
+- scale experiments, Toxic. Assess. 3 (1988) 643-656.**
+* treatment processes on the removal of hepatotoxins from microcystis and oscillatoria cyanobacteria
+- a laboratory study, Water Res. 23 (1989) 440 979-984.**
+* [426] **I.A. Westrick, D.C. Szlag, B.J. Southwell, J. Sinclair, A review of cyanobacteria and cyanotoxins removal/inactivation in drinking water treatment, Anal. Bioanal. Chem. 397 (2010) 1705-1714.**
+* [427] **G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals (OH/O') in aqueous solution, J. Phys. Chem. Ref. Data. 17 (1988) 445 513-851.**
+* [428] **O. Legrini, E. Oliveros, A.M. Braun, Photochemical processes for water-treatment, Chem. Rev. 93 (1993) 447 671-698.**
+* [429] **L.A. Lawton, P.K.J. Robertson, Physico-chemical treatment methods for the removal of microcystins (cyanobacterial hepatotoxins) from potable waters, Chem. Soc. Rev. 28 (1999) 217-224.*** [20] T. G aroma, M.D. Gurol, Degradation of tert-butyl alcohol in dilute aqueous solution by an Oy/UV process, Environ. Sci. Technol. 38 (2004) 5246-5252.
+* [21] W.H. Song, T. Teshiba, K. Rein, K.E. O Shea, Ultrasonically induced degradation and detoxification of microcystin-LR (cyanobacterial toxin), Environ. Sci. Technol. 39 (2005) 6300-6305.
+* [22] R.P. Qiao, N. Li, X.H. Qi, Q.S. Wang, Y.Y. Zhuang, Degradation of microcystin-RR by UV radiation in the presence of hydrogen peroxide, Toxicon. 45 (2005) 745-752.
+* [23] F. Al Momani, D.W. Smith, M.G. El-Din, Degradation of cyanobacteria toxin by advanced oxidation processes, J. Hazard. Mater. 150 (2008) 238-249.
+* [24] M.G. Antoniou, J.A. Shoemaker, A.A. De La Cruz, D.D. Dionysiou, Unveiling new degradation intermediates/pathways from the photocatalytic degradation of microcystin-LR, Environ. Sci. Technol. 42 (2008) 8877-8883.
+* [25] E.R. Bandala, D. Martinez, E. Martinez, D.D. Dionysiou, Degradation of microcystin-LR toxin by Fenton and photo-Fenton processes, Toxicon. 43 (2004) 829-832.
+* [26] W.H. Glaze, J.W. Kang, D.H. Chapin, The chemistry of water-treatment processes involving ozone, hydrogen peroxide and ultraviolet-radiation, Ozone-Sci. Eng. 9 (1987) 335-352.
+* [27] H.W. Prengie, Experimental rate constants and reactor considerations for the destruction of micropollutants and trihalomethane precursors by ozone with ultraviolet radiation, Environ. Sci. Technol. 17 (1983) 743-747.
+* [28] G.R. Peyton, W.H. Glaze, Destruction of pollutants in water with ozone in combination with ultraviolet-radiation. 3. photolysis of aqueous ozone, Environ. Sci. Technol. 22 (1988) 761-767.
+* [29] J. Staehelin, J. Hoigne, Decomposition of ozone in water-rate of initiation by hydroxide ions and hydrogen peroxide, Environ. Sci. Technol. 16 (1982) 676-681.
+* [30] V.K. Sharma, T.M. Triantis, M.G. Antoniou, X.X. He, M. Pelaez, C.S. Han, W.H. Song, K.E. O'Shea, A.A. de la Cruz, T. Kaloudis, A. Hiskia, D.D. Dionysiou, Destruction of microcystins by conventional and advanced oxidation processes: A review, Sep. Purif. Technol. 91 (2012) 3-17.
+* [31] K. Tsuji, T. Watanuki, F. Kondo, M.F. Watanabe, S. Suzuki, H. Nakazawa, M. Suzuki, H. Uchida, K. Harada, Stability of microcystins from cyanobacteria 2. Effect of UV light on decomposition and isomerization, Toxicon. 33 (1995) 1619-1631.
+* [32] K. Kaya, T. Sano, A photoactivation mechanism of the cyanobacterial hepatotoxin microcystin-LR by ultraviolet irradiation, Chem. Res. Toxic. 11 (1998) 159-163.
+* [33] A.R. Shawwa, D.W. Smith, Kinetics of microcystin-LR oxidation by ozone, Ozone-Sci. Eng. 23 (2001) 161-170.
+* [34] J. Rositano, B.C. Nicholson, P. Pieronne, Destruction of cyanobacterial toxins by ozone, Ozone-Sci. Eng. 20 (1998) 481-223-238.
+* [35] G.D. Onstad, S. Strauch, J. Meriluoto, G.A. Codd, U. von Gunten, Selective oxidation of key functional groups in cyanobacteria during a drinking water ozonation, Environ. Sci. Technol. 41 (2007) 4397-4404.
+* [36] J. Rositano, G. Newcombe, B. Nicholson, P. Sztajnbok, Ozonation of NOM and algal toxins in four treated waters, Water Res. 35 (2001) 23-32.
+* [37] S. Brooke, G. Newcombe, B. Nicholson, G. Klass, Decrease in toxicity of microcystins LA and LR in drinking water by ozonation, Toxicon. 48 (2006) 1054-1059.
+* [38] H.F. Miao, F. Qin, G.J. Tao, W.Y. Tao, W.Q. Ruan, Detoxification and degradation of microcystin-LR and -RR by ozonation, Chemosphere. 79 (2010) 355-361.
+* [39] J. Chang, Z.L. Chen, Z. Wang, J.M. Shen, Q. Chen, J. Kang, L. Yang, X.W. Liu, C.X. Nie, Ozonation degradation of microcystin-LR in aqueous solution: Intermediates, byproducts and pathways, Water Res. 63 (2014) 52-61.
+* [40] X.W. Liu, Z.L. Chen, N. Zhou, J.M. Shen, M.M. Ye, Degradation and detoxification of microcystin-LR in drinking water by sequential use of UV and ozone, J. Environ. Sci.-China. 22 (2010) 1897-1902.
+* [41] J. Kilduff, W.J. Weber, Transport and separation of organic macromolecules in ultrafiltration processes, Environ. Sci. Technol. 26 (1992) 569-577.
+
+* [42] J.R. Bolton, K.G. Linden, Standardization of methods for fluence (UV dose) determination in bench-scale UV experiments, J. Environ. Eng. Asce. 129 (2003) 209-215.
+* [43] H. Bader, J. Hoigne, Determination of ozone in water by the indigo method, Water Res. 15 (1981) 449-456.
+* [44] X.X. He, M. Pelaez, J.A. Westrick, K.E. O'Shea, A. Hiskia, T. Triantis, T. Kaloudis, M.I. Stefan, A.A. de la Cruz, D.D. Dionysiou, Efficient removal of microcystin-LR by UV-C/H\({}_{2}\)O\({}_{2}\) in synthetic and natural water samples, Water Res. 501 46 (2012) 1501-1510.
+* [45] G.S. Shephard, S. Stockenstrom, D. De Villiers, W.J. Engelbrecht, E.W. Sydenham, G.F.S. Wessels, Photocatalytic degradation of cyanobacterial microcystin toxins in water, Toxicon. 36 (1998) 1895-1901.
+* [46] G.S. Shephard, S. Stockenstrom, D. de Villiers, W.J. Engelbrecht, G.F.S. Wessels, Degradation of microcystin toxins in a falling film photocatalytic reactor with immobilized titanium dioxide catalyst, Water Res. 36 (2002) 506 140-146.
+* [47] X.C. Jin, S.R. Wang, Y. Pang, F.C. Wu, Phosphorus fractions and the effect of pH on the phosphorus release of the sediments from different trophic areas in Taihu Lake, China, Environ. pollut. 139 (2006) 288-295.
+* [48] J.R. Schmidt, S.W. Wilhelm, G.L. Boyer, The fate of microcystins in the environment and challenges for monitoring, Toxins. 6 (2014) 3354-3387.
+* [49] F.A. Al Momani, N. Jarrah, Treatment and kinetic study of cyanobacterial toxin by ozone, J. Environ. Sci. Health 512 Part a-Toxic/Hazard. Subst. Environ. Eng. 45 (2010) 719-731.
+* [50] M. Mehrjouei, S. Mueller, D. Moeller, A review on photocatalytic formation used for the treatment of water and wastewater, Chem. Eng. J. 263 (2015) 209-219.
+* a review., Environ. Health Perspect. 69 (1986) 151-157.
+* [52] K.P. Wang, J.S. Guo, M. Yang, H. Junji, R.S. Deng, Decomposition of two haloacetic acids in water using UV radiation, ozone and advanced oxidation processes, J. Hazard Mater. 162 (2009) 1243-1248.
+* [53] P. Westerhoff, G. Aiken, G. Amy, J. Debroux, Relationships between the structure of natural organic matter and its reactivity towards molecular ozone and hydroxyl radicals, Water Res. 33 (1999) 2265-2276.
+* [54] M. Welker, C. Steinberg, Rates of humic substance photosensitized degradation of microcystin-LR in natural waters, Environ. Sci. Technol. 34 (2000) 341-3419.
+* [55] H. Mazur-Marzec, J. Meriluoto, M. Plinski, The degradation of the cyanobacterial hepatotoxin nodulararin (NOD) by UV radiation, Chemosphere. 65 (2006) 1388-1395.
+* [56] A. Latifoglu, M.D. Gurol, The effect of humic acids on nitrobenzene oxidation by ozonation and Oy/UV processes, Water Res. 37 (2003) 1879-1889.
+* [57] H. Kusic, N. Koprivanac, A.L. Bozic, Minimization of organic pollutant content in aqueous solution by means of AOPs: UV- and ozone-based technologies, Chem. Eng. J. 123 (2006) 127-137.
+* [58] J.M. Shen, Z.L. Chen, Z.Z. Xu, X.Y. Li, B.B. Xu, F. Qi, Kinetics and mechanism of degradation of p-chloronitrilebenzine in water by ozonation, J. Hazard. Mater. 152 (2008) 1325-1331.
+* [59] R.W. Vachet, B.M. Bishop, B.W. Erickson, G.L. Glish, Novel peptide dissociation: Gas-phase intramolecular rearrangement of internal amino acid residues, J. Am. Chem. Soc. 119 (1997) 5481-5488.
+* [60] M.G. Antoniou, J.A. Shoemaker, A.A. de la Cruz, D.D. Dionysiou, LC/MS/MS structure elucidation of reaction intermediates formed during the TiO\({}_{2}\) photocatalysis of microcystin-LR., Toxicon 51 (2008) 1103-1118.
+* [61] I. Liu, L.A. Lawton, P.K.J. Robertson, Mechanistic studies of the photocatalytic oxidation of microcystin-LR: An investigation of byproducts of the decomposition process, Environ. Sci. Technol. 37 (2003) 3214-3219.
+* [62] W.H. Song, A.A. De La Cruz, K. Rein, K.E. O'Shea, Ultrasonically induced degradation of microcystin-LR and -RR: Identification of products, effect of pH, formation and destruction of peroxides, Environ. Sci. Technol. 40 (2006) 538 3941-3946.
+* [63] Y.F. Fang, Y.P. Huang, J. Yang, P. Wang, G.W. Cheng, Unique ability of BiOBr to decarboxylate D-Glu and -DeAsp in the photocatalytic degradation of microcystin-LR in water, Environ. Sci. Technol. 45 (2011) 1593-1600.
+
+* [64] J. Andersen, C. Han, K. O'Shea, D.D. Dionysiou, Revealing the degradation intermediates and pathways of visible light-induced NF-TiO\({}_{2}\) photocatalysis of microcystin-LR, Appl. Catal. B Environ. 154 (2014) 259-266.
+* [65] W.M. Garrison, Reaction-mechanisms in the radiolysis of peptides, polypeptides, and proteins, Chem. Rev. 87 (1987) 381-398.
+* [66] J. Hoigne, H. Bader, Rate constants of reactions of ozone with organic and inorganic compounds in water-II. dissociating organic compounds, Water Res. 17 (1983) 185-194.
+* [67] U.v. Gunten, Ozonation of drinking water: Part I. Oxidation kinetics and product formation, Water Res. 37 (2003) 548 1443-1467.
+
+[MISSING_PAGE_POST]
+
+Figure 1:
+
+Figure 2:
+
+Figure 3:
+
+Figure 4:
+
+Figure 5:
+
+Figure 6:
+
+Figure 7:
+
+Figure 1: The \(\Delta\)-dependence of the \(\Delta\)-
+
+* [583] Degradation of MC-LR was investigated in the combined UV/O\({}_{3}\) process.
+* [584] UV/O\({}_{3}\) was more efficient to remove/mineralize MC-LR than UV and O\({}_{3}\) alone.
+* [585] High pH and DOC inhibited the degradation of the toxin.
+* [586] Degradation of MC-LR was initiated at four sites of the toxin.
+* [587] A possible degradation pathway of MC-LR is proposed.
+* [588]

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2016.06.136.mmd

@@ -0,0 +1,282 @@
+Accepted Manuscript
+
+Photocatalytic-assisted ozone degradation of metabollor aqueous solution
+
+C.A. Orge, M.F.R. Pereira, J.L. Faria
+
+PII: S1385-8947(16)30943-3
+
+DOI: [http://dx.doi.org/10.1016/j.cej.2016.06.136](http://dx.doi.org/10.1016/j.cej.2016.06.136)
+
+Reference: CEJ 15446
+
+To appear in: _Chemical Engineering Journal_
+
+Please cite this article as: C.A. Orge, M.F.R. Pereira, J.L. Faria, Photocatalytic-assisted ozone degradation of metabollor aqueous solution, _Chemical Engineering Journal_ (2016), doi: [http://dx.doi.org/10.1016/j.cej.2016.06.136](http://dx.doi.org/10.1016/j.cej.2016.06.136)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+**Photocatalytic-assisted ozone degradation of metabollor aqueous solution**
+
+**C.A. Orge\({}^{\star}\), M.F.R. Pereira, J.L. Faria**
+
+Laboratory of Separation and Reaction Engineering - Laboratory of Catalysis and Materials (LSRE-LCM)
+
+Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, s/n, 4200-465 Porto, Portugal.
+
+carlaorge@fe.up.pt, fpereira@fe.up.pt, jlfaria@fe.up.pt
+
+\({}^{\star}\) Corresponding author: Carla A. Orge
+
+e-mail: carlaorge@fe.up.pt
+
+tel: +351225081400
+
+fax: +351225081440Abstract
+
+Photocatalytic-assisted ozone degradation of metolachlor (MTLC) aqueous solutions was investigated using neat TiO\({}_{2}\) (prepared by sol-gel method) and TiO\({}_{2}\)/carbon composite (prepared from commercial available metal oxide and carbon phase) as catalysts.
+
+In terms of MTLC degradation, O\({}_{3}\) on its own is enough to achieve 100% removal, but the introduction of light increased the rate of removal. On the other hand, the combination of O\({}_{3}\) with light and the tested catalysts is mandatory to reach high mineralization in short reaction times. After 60 min of reaction, the TOC removal was 87% and 75% in the presence of the prepared composite and TiO\({}_{2}\), respectively.
+
+The concentration of two short chain carboxylic acids, oxalic and oxamic acids, was followed during MTLC degradation. The amount of these acids decreased when O\({}_{3}\) and light were combined.
+
+In general, nitrogen ions, such as nitrate and ammonium, were detected in the studied processes. All treatments released ammonium and light based processes also produced nitrate.
+
+Micro\({}^{\circledR}\) analysis showed that the combined process in the presence of the prepared catalysts led to a remarkable reduction in the toxicity of the treated solution, decreasing the inhibition of luminescent activity of _Vibrio Fisheri_ from 74% to 12%.
+
+Introduction
+
+The pollution caused by pesticides has significantly increased due to their extensive use in agriculture. Their residues are known to be persistent in surface water and ground water supplies posing a major problem in many countries [1]. Pesticides are highly noxious, sometimes non-biodegradable and very mobile throughout the environmental [2]. Chloroacetamides are the most widely used herbicides and their operation mode consists in inhibiting the early development of susceptible weeds by preventing biosynthesis of very long fatty acid chains, thus affecting cell integrity [3-5]. Among the most commonly used chloroacetamides are acetochlor, alachlor and metolachlor (MTLC) [3, 6] and they have been detected in surface water and ground water from 0.1 to 10 ug/L [7-10]. MTLC is listed in the Drinking Water Contaminant Candidate List of the US Environmental Protection Agency [11]. MTLC and its metabolites are alleged or confirmed carcinogens.
+
+Many conventional water treatment processes, such as coagulation and chlorination, have been found to be ineffective to remove alachlor and MTLC [12]. Therefore, successfully elimination of pesticides in aqueous environment requires application of unconventional processes [1, 2].
+
+Sunlight is an efficient degradation pathway of MTLC in soil, with the drawback of leaving behind some of its hazardous metabolites. It is estimated that about 50% of applied MTLC degrades in eight days on sunlit soil [13]. The degree of photodegradation diminishes rapidly with the deepening of soil incorporation. In water, it undergoes mesolytic degradation to give primarily 4-(2-ethyl-6-methylphenyl)-5-methyl-3-morpholine [14].
+
+The photocatalytic degradation of MTLC using pure TiO\({}_{2}\) and Ag modified TiO\({}_{2}\) was investigated by Sakkas et al. [15]. Working with low to moderate amounts of TiO\({}_{2}\), they concluded that in the early degradation steps the toxicity of the solution increases due to the formation of more toxic by-products.
+
+In the presence of organic matter the photodegradability of MTLC is hindered, while in the presence of nitrate the opposite is observed, because the production of HO\({}^{*}\) radicals is increased [16].
+
+Simulated drinking water samples containing a mixture of chloroacetamide herbicides and chloroacetamide derivatives were subjected to different treatments by Hladik et al. [12]. Coagulation provided little removal of the parent and derivative compounds, but chlorination was able to remove completely the by-products. However, ozonation proved to be even more efficient than chlorination [12].
+
+According to Munoz et al., homogeneous Fenton-like oxidation (H\({}_{2}\)O\({}_{2}\)/Fe\({}^{3+}\)) of monochlorophenols is strongly dependent of the operating conditions [17]. The stoichiometric amount of H\({}_{2}\)O\({}_{2}\) and Fe\({}^{3+}\) is the key to achieve suitable results.
+
+The use of organic matter from compost to promote the photocatalytic degradation of two herbicides and a fungicide by solar light was studied by Coelho et al. [18]. Following their findings the main conclusion was that hydroxyl radicals are the principal species involved in the reactions, mainly due to their high reactivity.
+
+On the other hand, Avetta et al. [19] proposed a different mechanism suggesting the participation of singlet oxygen species in the photodegradation of monochlorophenols in the presence of soluble bio-based substances (SBO). For different organic substrates the toxicity assays showed a progressive, up to a complete, detoxification of the system, mediated by the singlet oxygen species with no significant contribution of the present SBO.
+
+Even though a direct comparison of the efficiency of TiO\({}_{2}\) and ZnO may not be precise due to differences in their surface properties, Fennoll et al. verified that ZnO was more efficient during photocatalytic degradation of a mixture of triazina and chloroacetanilide herbicides [20].
+
+Other studies reported the ozonation of MTLC, individually or in a mixture of emerging pollutants, in semi-batch and continuous operation [21; 22]. In terms of toxicity, the by-products released with O\({}_{3}\) alone are more toxic than the parent compound, but the addition of the carbon material reduces this impact. Multi-walled carbon nanotubes and carbon nanofibers grown on a honeycomb corderite were used as catalysts.
+
+A study on the toxicity of photoproducts formed during UV-treatments of three chloroactamide herbicides showed that 90% of the original pesticide was converted in compounds with more or equal toxicity than the parent compound [3].
+
+In summary, this study focuses on using photocatalytic-assisted ozone process for MTLC degradation. To the best of our knowledge, the present work describes the MTLC degradation using photocatalytic ozonation by the first time. For that purpose two catalysts were prepared, one composite made of commercial TiO2 and multi-walled carbon nanotubes and another consisting in TiO2 synthetized by the sol-gel procedure. Experiments without catalysts and with only ozone and radiation were also performed in order to better understand the results. The performance of the prepared materials was compared with the commercial TiO2 (P25).
+
+## 2 Experimental
+
+### Reagents and materials
+
+MTLC (C15H2CINO2, PESTANAL Analytical Standard), nitric acid (HNO3, >=65%), oxalic acid (C2H2O4, >= 99%), oxalic acid (C2H3O3N, >=98%), _tert_-butanol ((CH3)3OH, >=99.5%), titanium (IV) isopropoxide (Ti[OCH(CH3)2]4, 97%), 2-ethyl-6-methylaniline (C2H5Ca13NHz, >=99.5%) and 2,6-pyridine dicarboxylic acid (C7H8NO4, 99%) were purchased from Sigma-Aldrich. Sodium carbonate (Na2CO3, >= 99%) was obtained from Fluka. Acetonitrile (C2H3N, HPLC gradient grade) and sulfuric acid (H2SO4, 95-98%) was supplied by Fisher Scientific. Methanol (CH3OH, MS grade) was acquired from VWR International. Ultrapure water was supplied by a Milli-Q water system. Commercial TiO2, sample P25, was supplied by Evonik Degussa Corporation. The commercial multi-walled carbon nanotubes, MWCNT, were supplied by Nanocyl (ref. 3100).
+Photocatalytic ozonation of MTLC was performed in a glass immersion photochemical reactor according to the experimental conditions described in [23]. The initial concentration of MTLC was 20 ppm and the reactor was loaded with 0.5 g L-1 of catalyst. The reaction system was the same for all tested processes; however, in the case of photolytic reactions the ozone was replaced by oxygen and for ozonation experiments the radiation source was turned off.
+
+In the experiments carried out with _tert_-butanol, the radical scavenger is presented in excess with a concentration 10 times higher than the initial concentration of the parent compound (C_tert_-butanol = 0.7 mmol L-1).
+
+For this study, different catalysts were tested in the kinetic reactions. Recent researches of our group have confirmed that composites based on TiO\({}_{2}\) and carbon nanotubes present high catalytic activity in the photocatalytic oxidative degradation of several pollutants [23; 24; 25; 26; 27]. The sol-gel method has been widely applied to prepare TiO\({}_{2}\) photocatalysts to be used in the development of prototypes at laboratorial scale [28; 29; 30]. Composite of 90:10 (w/w) P25 and MWCNT was synthetized by the hydration-dehydration technique, sample P2590 MWCNT\({}_{10}\)[25; 27]. In the composite preparation, a selected amount of MWCNT was dispersed in water under ultrasonication. P25 was added to the suspension 30 min later and the mixture was heated up to 80 oC and magnetically stirred until the water was completely evaporated. The resulting composite was dried at 110 oC overnight. TiO\({}_{2}\) sample was obtained by the sol-gel technique [30]. The preparation consisted in the slowly addition of Ti[OCH(CH\({}_{3}\))\({}_{2}\)]\({}_{4}\) to ethanol. After 30 min under continuous stirring, nitric acid was added. The solution was loosely covered and kept stirring until the homogeneous gel formed. After grinding the xerogel, a fine powder was obtained and afterwards it was calcined at 400 oC in a nitrogen flow for 2 h.
+
+The textural characterization of the materials was obtained from the N\({}_{2}\) equilibrium adsorption/desorption isotherms, determined at -196 oC with a Quantachrome Instruments NOVA 4200e apparatus. The relative amount of TiO\({}_{2}\) in the composite was determined by thermogravimetric analysis (TG) under air in a STA 409 PC/4/H Luxo 2 Netzsch thermal analyser. Detailed results of characterization were reported in a previous work [23].
+
+### Analytical techniques
+
+The MTLC concentration was monitored by HPLC, using a Hitachi Elite LaChrom device fitted with a diode array detector (DAD). The separation of the pollutant was attained using a Lichrocart C18-RP Puroshper Star (250 mm x 4.6 mm, 5 mm) column with an isocratic mobile phase containing 60% of acetonitrile and 40% of water. The concentration of short-chain carboxylic acids resulted from MTLC degradation, oxalic acid (OXA) and oxamic acid (OMA), was followed by a Hitachi Elite LaChrom HPLC provided with an UV-Vis detector and an Alltech QA-1000 chromatography column operating with an isocratic mobile phase of 5 mmol L-1 H2SO4.
+
+Analyses with an Ultra High Performance Liquid Chromatography with tandem Mass Spectrometry (UHPLC-MS/MS) were carried out to verify the presence of some MTLC intermediates reported in the literature. For UHPLC-MS/MS analyses, a Shimadzu Corporation apparatus (Tokyo, Japan) was used, consisting of a Nexera UHPLC equipment, coupled to a LCMS-8040 triple quadrupole mass spectrometer detector with an electrospray ionization source operating in both positive and negative ionization modes. The column Kinetex(tm) 1.7 mm XB-C18 100 A (100 x 2.1 mm i.d.)
+
+(Phenomenex, Inc., California, USA) was operated under reversed mode with a mobile phase consisting of a mixture of methanol and water (70/30, v/v) with a flow rate of 0.22 mL min-1, temperature of 20 degC and a volume of injection of 20 mL. The details of operation mode of UHPLC-MS/MS are reported in [31].
+
+In order to evaluate the mineralization degree of the processes, the total organic carbon (TOC) was obtained with a Shimadzu TOC-5000A apparatus.
+
+The ions formed during MTLC degradation were accomplished by the technique of ion chromatography by a MetroOHM 881 Compact IC Pro with 863 Compact Autosampler as described in a previous work [32].
+
+Microta(r) acute toxicity analyses, the procedure of which is described by standard ISO/DIS 11348-3 [33], were performed with the purpose of evaluating the toxicity caused by compounds released during the MTLC degradation. The experimental details are reported in [34]. Summarising, this procedure measures the inhibition of the light emission of bioluminescent bacteria (_V. fischeri_) promoted by the toxic effect of the tested chemicals, during 30 min of incubation at 15 degC.
+
+## 3 Results
+
+### MTLD degradation
+
+The experimental results corresponding to MTLC degradation during single ozonation (O\({}_{3}\)), photolysis (Light), photolytic ozonation (PO\({}_{3}\)) and photocatalytic ozonation (PCO\({}_{3}\)) corresponding to MTLC degradation are depicted in Figure 1.
+
+In order to better evaluate the performance of the different processes, the data corresponding to MTLC decay were fitted during the first 15 min of reaction.
+
+The PCO\({}_{3}\) follows a pseudo-first-order kinetic rate model. The PCO\({}_{3}\) of MTLC can be described by a Langmuir-Hinshelwood kinetic model [35] and assuming that due to combined effects of light and O\({}_{3}\) the concentration of HO\({}^{\bullet}\) with respect to MTLC is quasi constant.
+
+According to this model, in the absence of any solid catalyst, the evolution of MTLC concentration during the oxidation process was found to be well described by the following equation:
+
+\[-\frac{\mathrm{d}C_{\mathrm{MTLC}}}{\mathrm{dt}}=k_{\mathrm{hom}}C_{\mathrm{MTLC}} \tag{1}\]where \(k_{\rm hom}\)(min-1) represents the first-order apparent rate constant and \(C_{MTLC}\)(mmol L-1) is the concentration of the MTLC in each instant. Integration of Eq. (1), considering \(C_{MTLC}=C_{MTLC,0}\), when \(t=0\), leads to:
+
+\[\ln\frac{C_{MTLC,0}}{C_{MTLC}}=k_{\rm hom}t \tag{2}\]
+
+When the prepared materials are introduced, both homogeneous and heterogeneous degradation occur. Therefore, the MTLC removal rate is the sum of the two contributions:
+
+\[-\frac{dC_{MTLC}}{dt}=(k_{\rm hom}+k_{\rm het})C_{MTLC} \tag{3}\]
+
+where \(k_{\rm het}\) (min-1) represents the first-order apparent rate constant for the heterogeneous degradation. Integration of Eq. (3), considering \(k_{\rm app}=k_{\rm hom}+k_{\rm het}\) and \(C_{MTLC}=C_{MTLC,0}\), when t = 0, leads to:
+
+\[\ln\frac{C_{MTLC,0}}{C_{MTLC}}\in k_{\rm app}t \tag{4}\]
+
+The corresponding rate constants are given in the Table 1.
+
+\begin{table}
+\begin{tabular}{c c c}
+1 & **System (Catalyst)** & \(k_{\rm app}\)**×10\({}^{2}\) ± 25\% (min-1)** \\
+0 & \(O_{3}\) (-) & 16\({}^{*}\) \\ Light (-) & 4\({}^{*}\) \\ PO\({}_{3}\) (-) & 27\({}^{*}\) \\ PCO\({}_{3}\) (P25\({}_{0.9}\)MWCNT\({}_{0.1}\)) & 28 \\ PCO\({}_{3}\) (TiO\({}_{2}\)) & 12 \\ PCO\({}_{3}\) (P25) & 36 \\ \end{tabular}
+\end{table}
+Table 1: First-order apparent rate constants of non-catalytic and catalytic runs of MTLC decay.
+
+MTLC is an aromatic compound, it presents a high delocalization of electrons and exhibits enhanced reactivity towards O\({}_{3}\). This behavior is in accordance with the literature, which suggests that the degradation of MTLC is mainly due to the direct reaction with ozone [21].
+
+On the other hand, photolysis did not easily remove MTLC. Sakkas et al. also verified that MTLC concentration in aqueous solutions did not decrease under direct irradiation [15]. The presence of light was not enough to completely remove MTLC from the solution in a short period of reaction.
+
+The combination of O\({}_{3}\) with near UV/Vis light led to a slightly faster decay of MTLC than single ozonation (1.7 times higher).
+
+In the case of PCO\({}_{3}\), MTLC degradation depends on the type of catalyst. The commercial sample (P25) presents the highest removal rate (k\({}_{\text{app}}\) = 36\(\times\)10\({}^{2}\) min-1), leading to a total conversion after 5 min of reaction. When the prepared composite was combined with O\({}_{3}\) and light (k\({}_{\text{app}}\) = 28\(\times\)10\({}^{2}\) min-1) no significant improvement was verified when compared to the non-catalytic-combined run (k\({}_{\text{hom}}\) = 27\(\times\)10\({}^{2}\) min-1), but with the synthesized TiO\({}_{2}\) a decrease in the degradation rate was observed. Although the different catalysts did not present the same performances, total MTLC removal was observed until 30 min of reaction with all O\({}_{3}\) based processes, as expected [21].
+
+In order to better understand the influence of HO\({}^{\ast}\) radicals in the reaction mechanism of MTLC degradation, experiments were carried out in the presence _tert_-butanol, a well known hydroxyl radical scavenger. _Tert_-butanol reacts very rapidly with hydroxyl radicals (k = 5 x 10\({}^{8}\) M-1 s-1) [36] and very slowly with ozone ( k = 0.03 M-1 s-1) [37, 38].
+
+These results are depicted in Figure 2.
+
+The MTLC degradation was practically not affected by the addition of the radical scavenger _tert_-butanol during O\({}_{3}\) based processes, both in the presence and absence of catalyst or light. A very similar trend was observed by Restivo et al. [21]. This result suggested that MTLC reacts mainly with O\({}_{3}\) under the current experimental conditions.
+
+In the case of photolysis, a significant influence was verified by the addition of _tert_-butanol, which means that HO* radicals play a key role in the presence of light alone. Benitez et al. also reported that MTLC photooxidation rate was negatively affected by the presence of _tert_-butanol, independently of type of water [39].
+
+### 3.2 TOC removal
+
+In order to evaluate the efficiency of removing the by-products released during MTLC degradation, analyses of TOC were carried out at 60, 120 and 180 min of reaction. The results of normalized TOC content are depicted in Figure 3.
+
+The non-catalytic individual methods, single ozonation and photolysis, present a low TOC removal, less than 20% after 180 min of reaction. This occurs because the by-products of the oxidation of MTLC are less reactive than MTLC to O3. This behavior is in accordance with what was previous reported [27, 40], only a slow decrease of organic matter and no variation was observed after 8 h during photolysis under Xe irradiation. On the other, when these two processes were combined the TOC content significantly decreased. O3 and light together removed 40% of TOC after 60 min of reaction and the amount of TOC removed doubled after 120 min of reaction; however, its value remains in the last hour of the reaction. Independently of the catalyst introduced, when the prepared samples were combined with near UV/Vis light, a significant amount of organic matter was quickly removed. The best performance in terms of mineralization degree was verified with the prepared composite, leading to 87% of TOC removal after 60 min of reaction. This value remained practically unchanged until the end of reaction (91% of TOC removal after 3 h). In the case of TiO2 sample, a TOC depletion of 75% and 85% was verified in the first and third hour of reaction, respectively. The commercial catalyst presented a similar profile.
+With the aim of evaluating the presence of synergetic effects during PCO\({}_{3}\) with the prepared materials, catalytic ozonation (CO\({}_{3}\)) and photocatalysis (PC) were individually carried out. Adsorption experiments (ADS) were also performed. The results of normalized MTLC concentration and TOC content are presented in Figure 4.
+
+Both CO\({}_{3}\) and PC in the presence of prepared samples led to a fast decay of MTLC in terms of TOC removal, PC presents better results than CO\({}_{3}\). In the case of the composite during PC, a high content of TOC was removed after 120 min of reaction (84%). This value is very similar to that obtained by PCO\({}_{3}\) (90%), although the combined method achieved a higher removal after only 1 h of reaction. PC with TiO\({}_{2}\) removed approximately 50% of organic matter. The prepared composite had a higher adsorption capacity than the neat TiO\({}_{2}\). This improvement can be attributed to the presence of the carbon phase in the composite.
+
+Summarizing, CO\({}_{3}\) and PC in the presence of prepared samples are effective in MTLC degradation; however, high mineralization rates are only accomplished in a short period of reaction by PCO\({}_{3}\).
+
+### 3.4 By-products analysis
+
+Short chain carboxylic acids are final intermediates of a wide range of organic compounds. Thus, the formation of OXA and OMA was followed during MTLC degradation (Figure 5).
+
+In the case of photolysis none of the acids was detected.
+
+During single ozonation the concentration of OXA increases all the time, as expected since OXA is refractory to O\({}_{3}\) alone [41, 42]. P\({}_{3}\) produced the highest amount of OXA, however at the end of the reaction all acid was degraded. This result was expected since O\({}_{3}\) and light together are able to remove OXA from the solutions [32]. In the remaining processes only a small content was verified until 120 min of reaction.
+
+As happens with OXA, the profile concentration of OMA depends on the applied process. Analysing the results of the individual methods, photolysis did not produce OMA and single ozonation released a slight amount after 120 min of MTLC degradation. In the case of PO3, its concentration increases until approximately 0.030 mmol L-1 at 120 min of reaction and after that it starts to decrease. PCO3 in the presence of TiO2 released the highest amount of OMA (approximately 0.100 mmol L-1), but it disappeared in 1 h. In PCO3 with P250.9MWCNT0.1 only a small amount of OMA was observed in the first hour of reaction. This decrease in OMA concentration is expected in view of previous results because PCO3 with an appropriate catalyst can eliminate OMA [23].
+
+In order to evaluate the concentration of ions released during MTLC degradation, ion chromatography analyses were carried out at specific times of reaction. Table 2 presents the concentration of ions and the nitrogen-balance of identified species at the end of the reaction for all tested processes.
+
+\begin{table}
+\begin{tabular}{c c c c} \hline \hline System (Catalyst) & NOx3 (mmol L-1) & OMA (mmol L-1) & Nr (mmol L-1) & Nr (mmol L-1) \\ \hline \hline O3 (-) & - & 0.015 & 0.008 & 0.023 \\ Light (-) & 0.030 & 0.012 & - & 0.042 \\ PO3 (-1) & 0.014 & 0.015 & 0.022 & 0.052 \\ PCO3 (P250.9MWCNT0.1) & - & 0.035 & 0.001 & 0.037 \\ PCO3 (TiO2) & - & - & - & - \\ PCO3 (P25) & 0.020 & 0.035 & - & 0.055 \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Concentration of ions and N-balance of identified species (Nr) at 180 min of reaction.
+
+PO\({}_{3}\) and PCO\({}_{3}\) with commercial sample, P25, presented the highest amount of identified species (N\({}_{\text{max}}\) = 0.07 mmol L\({}^{-1}\)).
+* The presence of non-identified nitrogen containing species, especially in the cases of O\({}_{3}\) alone and PCO\({}_{3}\) with TiO\({}_{2}\) must be taken into account. The adsorption of the parent compound and the possible adsorption of nitrogenated oxidation by-products may also be taken in account for the TiO\({}_{2}\) material. In addition, the degradation of MTLC could lead to the formation of nitrogen compounds in gas phase, such as N\({}_{\text{2}}\) or nitrogen oxides.
+* With the aim of evaluating the presence of other intermediates during MTLC degradation, UHPLC-MS/MS analyses were carried out. Only the presence of 2-ethyl-6-methylaniline was confirmed during photolysis at 120 and 180 min of reaction under the experimental conditions used. This compound is a known intermediated from the oxidation of MTLC and it is characterized by high toxicity [15, 43]. The chromatographic analysis suggests that the concentration of 2-ethyl-6-methylaniline decreased after 120 min till 180 min. The presence of this intermediate during photolysis at the end of reaction is in accordance with the remaining results. Since MTLC was slowly degraded during photolysis, it is expected that the primary intermediates, as 2-ethyl-6-methylaniline, are detected at this time of reaction, instead of final by-products as short chain carboxylic acids, which are verified in the remaining processes.
+* Non observation of this compound during the remaining processes suggested that the reaction mechanism is different from that of the O\({}_{3}\) based processes. When the reaction was carried only with light, 2-ethyl-6-methylaniline was detected; when O\({}_{3}\) was used alone or combined, other primary by-products were formed (their identification was not possible at this stage with the existing analytical methods, due to their trace concentrations).
+
+### Bio-toxicity of MTLC degradation products
+
+In order to assess the toxicity caused by compounds produced during MTLC oxidation, the acute toxicity of the untreated MTLC solution and solutions submitted to different treatments during 180 min was evaluated by Microtax(r) bioassays. The marine bacterium used to study the toxic substances was _Vibrio fischeri_. The results presented in Figure 6 were obtained after 30 min of exposure by determination of inhibition percentage in the bacteria luminescence caused by each sample.
+
+On all tested processes, the final solution has less toxicity than the untreated solution, even when the non-catalytic individual methods are applied. The toxicity observed during MTLC photolysis can be attributed to the formation of 2-ethyl-6-methylaniline that was identified during UHPLC-MS/MS analyses [15, 16].
+
+According to the results, both non-catalytic and catalytic combined methods in the presence of the prepared samples presented the most pronounced decrease. Concerning the PCO\({}_{3}\) systems that use the prepared catalysts (excluding the neat P25) the measured percent of initial inhibition is 12%, representing an abatement of 84% with relation to the value of the non-treated sample. PCO\({}_{3}\) with the neat commercial sample (P25) was less efficient than with the prepared catalyst. In addition, PO\({}_{3}\) also led to a treated solution with less toxicity than the catalytic reaction with P25. This difference in the toxicity of the final solutions suggested that the introduction of a carbon phase in the composite, as well as the presence of different crystalline phases on TiO\({}_{2}\) changed the paths of the reaction mechanism, and consequently the final by-products formed.
+
+## 4 Conclusions
+
+The present work focuses on the photocatalytic-assisted ozone degradation of MTLC aqueous solutions and reports this combined method for the first time. Two different catalysts were used, neat TiO\({}_{2}\) synthetized by the sol-gel method and TiO\({}_{2}\)/carbon-nanotubes composite made of commercial TiO\({}_{2}\) and multi-walled carbon nanotubes. The degradation of MTLC molecule can be easily achieved by O\({}_{3}\) alone. However, this is not enough to achieve a high mineralization and a toxicity abatement. Indeed,considerable amounts of organic matter are still present in solution after 60 min of reaction in the absence of a (photo)catalyst. Additionally, the application of the combined process using the prepared materials leads to a significant decrease in the inhibition of the luminescent activity of _Vibrio Fisheri_.
+
+The introduction of a carbon phase in the composite increased the TOC removal in comparison with the parent TiO\({}_{2}\), removing 90% of organics after 60 min of reaction.
+
+The best toxicity reduction was achieved by PCO\({}_{3}\) system that used the prepared catalysts (not the commercial one), reaching 84% of abatement on the percent of initial inhibition effect, after 180 min of reaction.
+
+The PCO\({}_{3}\) system was demonstrated to have great potential for MTLC degradation, since the pollutant was removed very fast, a high mineralization degree was easily reached and the final solution had low toxicity.
+
+## Acknowledgements
+
+This work was financed by FCT and FEDER through COMPETE 2020 (Project UID/EQU/50020/2013 - POCI-01-0145-FEDER-006984). C. A. Orge acknowledges the research fellowship BPD/90309/2012 received from FCT. The authors are grateful to Ana R. Ribeiro for UHPLC-MS/MS analyses, as well as the appropriate interpretation of the results.
+
+## References
+
+* (1) H. Yang, L. Yang, Z. Hu, H. Liu, H. Pang, B. Yi, General transformation law of chlorinated acetanilide herbicides by reactive oxidative species (ROSs) during photocatalytic process, Journal of Water Process Engineering 8 (2015) e66-e74.
+* (2) R. Otero, D. Esquivel, M.A. Ulibarri, F.J. Romero-Salguero, P. Van Der Voort, J.M. Fernandez, Mesoporous phenolic resin and mesoporous carbon for the removal of S-Metolachlor and Bentazon herbicides, Chemical Engineering Journal 251 (2014) 92-101.
+
+[3] Y. Souissi, S. Bouchonnet, S. Bourcier, K.O. Kusk, M. Sablier, H.R. Andersen, Identification and ecotoxicity of degradation products of chloroacetamide herbicides from UV-treatment of water, Science of The Total Environment 458-460 (2013) 527-534.
+* [4] P. Boger, B. Matthes, J. Schmalfuss, Towards the primary target of chloroacetamides -new findings pave the way, Pest Management Science 56 (2000) 497-508.
+* Section C Journal of Biosciences 57 (2002) 843-852.
+* [6] L. Xin-Yu, S. Zhen-Cheng, L. Xu, Z. Cheng-Gang, Z. Hui-Wen, Assessing the effects of acetochlor on soil fungal communities by DGE and clone library analysis, Ecotoxicology 19 (2010) 1111-1116.
+* [7] T.C. Cavalier, T.L. Lavy, J.D. Mattice, Persistence of Selected Pesticides in Ground-Water Samples, Ground Water 29 (1991) 225-231.
+* [8] J.E. Barbash, G.P. Thelin, D.M. Kolpin, R.J. Gilliom, Major Herbicides in Ground Water, Journal of Environmental Quality 30 (2001) 831-845.
+* [9] I.K. Konstantinou, D.G. Hela, T.A. Albanis, The status of pesticide pollution in surface waters (rivers and lakes) of Greece. Part I. Review on occurrence and levels, Environmental Pollution 141 (2006) 555-570.
+* [10] C. Rocha, E.A. Pappas, C.h. Huang, Determination of trace triazine and chloroacetamide herbicides in tile-fed drainage ditch water using solid-phase microextraction coupled with GC-MS, Environmental Pollution 152 (2008) 239-244.
+* CCL 4, United States Environmental Protection Agency [http://www2.epa.gov/ccl/chemical-contaminants-ccl-4](http://www2.epa.gov/ccl/chemical-contaminants-ccl-4) (2015).
+* [12] M.L. Hladik, A.L. Roberts, E.J. Bouwer, Removal of neutral chloroacetamide herbicide degradates during simulated unit processes for drinking water treatment, Water Research 39 (2005) 5033-5044.
+
+[13] Herbicide Handbook, Weed Science Society of America, 7th Edition ed., Champaign, 1994.
+* [14] R. Mathew, S.U. Khan, Photodegradation of Metolachlor in Water in the Presence of Soil Mineral and Organic Constituents, Journal of Agricultural and Food Chemistry 44 (1996) 3996-4000.
+* [15] V.A. Sakkas, I.M. Arabatzis, I.K. Konstantinou, A.D. Dimou, T.A. Albanis, P. Falaras, Metolachlor photocatalytic degradation using TiO2 photocatalysts, Applied Catalysis B: Environmental 49 (2004) 195-205.
+* [16] A.D. Dimou, V.A. Sakkas, T.A. Albanis, Metolachlor Photodegradation Study in Aqueous Media under Natural and Simulated Solar Irradiation, Journal of Agricultural and Food Chemistry 53 (2005) 694-701.
+* [17] M. Munoz, Z.M. de Pedro, J.A. Casas, J.J. Rodriguez, Assessment of the generation of chlorinated byproducts upon Fenton-like oxidation of chlorophenols at different conditions, Journal of Hazardous Materials 190 (2011) 993-1000.
+* [18] C. Coelho, L. Cavani, A.t. Haile, G. Guyot, C. Ciavatta, C. Richard, Rates of production of hydroxyl radicals and singlet oxygen from irradiated compost, Chemosphere 85 (2011) 630-636.
+* [19] P. Avetta, F. Bella, A. Bianco Prevot, E. Laurenti, E. Montoneri, A. Arques, L. Carlos, Waste Cleaning Waste: Photodegradation of Monochlorophenols in the Presence of Waste-Derived Photosensitizer, ACS Sustainable Chemistry & Engineering 1 (2013) 1545-1550.
+* [20] J. Fenoll, P. Hellin, C.M. Martinez, P. Flores, S. Navarro, Semiconductor-sensitized photodegradation of s-triazine and chloroacetanilide herbicides in leaching water using TiO2 and ZnO as catalyst under natural sunlight, Journal of Photochemistry and Photobiology A: Chemistry 238 (2012) 81-87.
+* [21] J. Restivo, J.J.M. Ortao, S. Armenise, E. Garcia-Bordeje, M.F.R. Pereira, Catalytic ozonation of metolachlor under continuous operation using nanocarbon materialsgrown on a ceramic monolith, Journal of Hazardous Materials 239-240 (2012) 249-256.
+* [22] S. Derrouiche, D. Bourdin, P. Roche, B. Houssais, C. Machinal, M. Coste, J. Restivo, J.J.M. Orfao, M.F.R. Pereira, Y. Marco, E. Garcia-Bordeje, Process design for wastewater treatment: catalytic ozonation of organic pollutants, Water Science and Technology 68 (2013) 1377-1383.
+* [23] C.A. Orge, M.F.R. Pereira, J.L. Faria, Photocatalytic ozonation of model aqueous solutions of oxalic and oxamic acids, Applied Catalysis B: Environmental 174-175 (2015) 113-119.
+* [24] C.G. Silva, J.L. Faria, Photocatalytic Oxidation of Phenolic Compounds by Using a Carbon Nanotube-Titanium Dioxide Composite Catalyst, ChemSusChem 3 (2010) 609-618.
+* [25] M.J. Sampaio, C.G. Silva, R.R.N. Marques, A.M.T. Silva, J.L. Faria, Carbon nanotube-TiO2 thin films for photocatalytic applications, Catalysis Today 161 (2011) 91-96.
+* [26] R.R.N. Marques, M.J. Sampaio, P.M. Carrapico, C.G. Silva, S. Morales-Torres, G. Drazic, J.L. Faria, A.M.T. Silva, Photocatalytic degradation of caffeine: Developing solutions for emerging pollutants, Catalysis Today 209 (2013) 108-115.
+* [27] M.J. Sampaio, R.R.N. Marques, P.B. Tavares, J.L. Faria, A.M.T. Silva, C.G. Silva, Tailoring the properties of immobilized titanium dioxide/carbon nanotube composites for photocatalytic water treatment, Journal of Environmental Chemical Engineering 1 (2013) 945-953.
+* [28] W. Wang, C.G. Silva, J.L. Faria, Photocatalytic degradation of Chromotrope 2R using nanocrystalline TiO2/activated-carbon composite catalysts, Applied Catalysis B: Environmental 70 (2007) 470-478.
+* [29] C.G. Silva, J.L. Faria, Effect of key operational parameters on the photocatalytic oxidation of phenol by nanocrystalline sol-gel TiO2 under UV irradiation, Journal of Molecular Catalysis A: Chemical 305 (2009) 147-154.
+
+* [30] C.G. Silva, J.L. Faria, Anatase vs. rutile efficiency on the photocatalytic degradation of clofibric acid under near UV to visible irradiation, Photochemical & Photobiological Sciences 8 (2009) 705-711.
+* [31] A.R. Ribeiro, M. Pedrosa, N.F.F. Moreira, M.F.R. Pereira, A.M.T. Silva, Environmental friendly method for urban wastewater monitoring of micropollutants defined in the Directive 2013/39/EU and Decision 2015/495/EU, Journal of Chromatography A 1418 (2015) 140-149.
+* [32] C.A. Orge, J.L. Faria, M.F.R. Pereira, Removal of oxalic acid, oxalic acid and aniline by a combined photolysis and ozonation process, Environmental Technology 36 (2014) 1075-1083.
+* determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (luminescent bacteria test), Part 3: Method Using Freeze-Dried Bacteria (2005).
+* [34] A.G. Goncalves, J.J.M. Orfao, M.F.R. Pereira, Catalytic ozonation of sulphamethoxazole in the presence of carbon materials: Catalytic performance and reaction pathways, Journal of Hazardous Materials 239-240 (2012) 167-174.
+* [35] D. Ollis, C.G. Silva, J. Faria, Simultaneous photochemical and photocatalyzed liquid phase reactions: Dye decolorization kinetics, Catalysis Today 240, Part A (2015) 80-85.
+* [36] J. Hoigne, Inter-calibration of OH radical sources and water quality parameters, Water Science and Technology 35 (1997) 1-8.
+* [37] J. Hoigne, H. Bader, Rate constants of reactions of ozone with organic and inorganic compounds in water--II: Dissociating organic compounds, Water Research 17 (1983) 185-194.
+* [38] P.C.C. Faria, J.J.M. Orfao, M.F.R. Pereira, Activated carbon catalytic ozonation of oxalic and oxalic acids, Applied Catalysis B: Environmental 79 (2008) 237-243.
+
+* [39] F. Javier Benitez, J.L. Acero, F.J. Real, C. Maya, Modeling of photooxidation of acetamide herbicides in natural waters by UV radiation and the combinations UV/H2O2 and UV/O3, Journal of Chemical Technology & Biotechnology 79 (2004) 987-997.
+* [40] L. Gutowski, O. Olsson, C. Leder, K. Kummerer, A comparative assessment of the transformation products of S-metolachlor and its commercial product Mercantor Gold(r) and their fate in the aquatic environment by employing a combination of experimental and in silico methods, Science of The Total Environment 506-507 (2015) 369-379.
+* [41] C.A. Orge, J.J.M. Ortao, M.F.R. Pereira, Catalytic ozonation of organic pollutants in the presence of cerium oxide-carbon composites, Applied Catalysis B: Environmental 102 (2011) 539-546.
+* [42] C.A. Orge, J.J.M. Ortao, M.F.R. Pereira, A.M. Duarte de Farias, M.A. Fraga, Ceria and cerium-based mixed oxides as ozonation catalysts, Chemical Engineering Journal 200-202 (2012) 499-505.
+* [43] O. Osano, W. Admiraal, H.J.C. Kamer, D. Pastor, E.A.J. Bleeker, Comparative toxic and genotoxic effects of chloroacetanilides, formamidines and their degradation products on Vibrio fischeri and Chironomus riparius, Environmental Pollution 119 (2002) 195-202.
+
+Figure captions
+* 1 Figure 1: Normalized concentration (C/C0) of MTLC in aqueous solution as a function of time for O3, Light, PO3 and PCO3.
+* 2 Influence of _tert_-butanol on the dimensionless MTLC concentration during O3, Light, PO3 and PCO3 with prepared composite.
+* 3 Figure 3: Normalized TOC content (TOC/TOC0) for MTLC in aqueous solution as a function of time for O3, Light, PO3 and PCO3.
+* 4 Figure 2: Normalized MTLC concentration (C/C0) and TOC content (TOC/TOC0) as a function of time for CO3, PC and ADS experiments in the presence of P2590MWCNT10 (a) and (b) and TiO2 (c) and (d).
+* 5 Figure 3: OXA (a) and OMA (b) concentration for single ozonation (O3), photolysis (Light), PO3 and PCO3.
+* 6 Figure 6: Results of Microtox(r) tests at 180 min of reaction, with exposure time at bacteria _Vibrio fischeri_ of 30 min.
+* 7 Figure 7:Figure 4:
+
+[MISSING_PAGE_EMPTY:25]
+
+[MISSING_PAGE_EMPTY:26]
+
+Figure 4:
+
+Figure 5:
+
+Figure 6:
+
+**Research Highlights**
+
+* O\({}_{3}\) assisted photocatalytic degradation of metolachlor was investigated
+* O\({}_{3}\) by itself is enough to achieve total removal
+* High mineralization was only easily attained with O\({}_{3}\), light and tested samples
+* The combined method led a pronounced decrease in the toxicity of the solutions
+* Prepared catalysts presented remarkable performance during metolachlor degradation
+Figure 1: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**: **A**A

manual_annotation/200pdfs_in_mmd/10.1016_j.cej.2018.10.093.mmd

@@ -0,0 +1,254 @@
+Accepted Manuscript
+
+Coupling of electrocoagulation and ozone treatment for textile wastewater reuse Lucyna Bilinska, Kazimierz Blus, Marta Gmurek, Stanislaw Ledakowicz
+
+PII: S1385-8947(18)32044-8 DOI: [https://doi.org/10.1016/j.cej.2018.10.093](https://doi.org/10.1016/j.cej.2018.10.093)
+
+Reference: CEJ 20156
+
+To appear in: _Chemical Engineering Journal_
+
+Received Date: 5 July 2018
+
+Revised Date: 13 September 2018
+
+Accepted Date: 10 October 2018
+
+Please cite this article as: L. Bilinska, K. Blus, M. Gmurek, S. Ledakowicz, Coupling of electrocoagulation and ozone treatment for textile wastewater reuse, _Chemical Engineering Journal_ (2018), doi: [https://doi.org/10.1016/j.cej.2018.10.093](https://doi.org/10.1016/j.cej.2018.10.093)
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+# Coupling of electrocoagulation and ozone treatment for textile wastewater reuse
+
+Lucyna Bilinska\({}^{\text{a}\text{*}}\), Kazimierz Blus\({}^{\text{a}}\), Marta Gmurek\({}^{\text{b}}\), Stanislaw Ledakowicz\({}^{\text{b}}\)
+
+###### Abstract
+
+Industrial textile wastewater is usually characterized by high pH, intense color and extremely high salinity, especially in case of wastewater generated after dyeing. Due to the complex matrix of textile wastewater, the selection of an appropriate treatment method can be a challenge. This paper presents the advantages of coupling electrocoagulation (EC) and ozonation (O\({}_{3}\)) as one-step (EC+O\({}_{3}\)) and two-step (EC\(\rightarrow\)O\({}_{3}\)) operations. The combination of EC and O\({}_{3}\), which both worked well in salty environment, gave very good results in the caseof aqueous solutions and real industrial wastewater containing Reactive Black 5 (RB5). Very high color removal, more than 95%, was achieved in a very short treatment time (less than 18 min., while ca. 60 min. was required in the case of individual O3). At the same time the applied ozone dose was reduced from 1.69 g/L, when using O3 only, to less than 0.3 g/L for EC\(\rightarrow\)O3. The results of color removal and mineralization analysis indicated the high efficiency of both the EC+O3 and EC\(\rightarrow\)O3 processes; however, the cost evaluation revealed that the EC\(\rightarrow\)O3 treatment was more advantageous. EC preformed as a single treatment resulted in an increase in toxicity, but coupling EC with O3 reduced this undesirable effect. Very promising results were obtained in recycling trials. The color difference (DE\({}_{\text{cmc}}\), according to ISO 105-J03) of textiles dyed with purified wastewater (brine) were between 0.28 and 0.98 (below the limiting value of 1.5).
+
+**Keywords:** textile wastewater; electrocoagulation; ozonation; brine recycling; toxicity
+
+**Highlights:**
+
+* **Ozonation (O3) and electrocoagulation (EC) work well in a salty wastewater matrix**
+* **EC\(\rightarrow\)O3 treatment gives more than 95% color removal in a very short time (18 min.)
+* **Ozone dose can be reduced more than 5 times when EC is coupled with O3 treatment**
+* **Two-step EC\(\rightarrow\)O3 treatment was more economically justified**
+* **Recycling of purified brine in a subsequent dyeing operation gave very promising results**
+
+## 1 Introduction
+
+The manufacturing of textiles is a complex issue and consists of many operations, such as desizing, washing, bleaching, dyeing and printing [1]. Therefore, real industrial textile wastewater has an extremely complex matrix and contains many other chemical ingredients in addition to dyes. A very high pH and salinity are the most apparent characteristics of the textile wastewater matrix, especially in the case of reactive dyeing effluent, and these parameters should be taken into consideration when the selection of an appropriate treatment method is concerned.
+
+Ozonation (O3) is an efficient method of textile wastewater treatment. Ozone is a strong oxidant that can decompose many substances, including dyes. In our previous studies, it was found that among advanced oxidation processes (AOPs) ozonation is the least affected by textile auxiliaries, which compose the wastewater matrix; moreover, ozonation works well in a saline environment [2, 3, 4]. A publication by Bilinska et al. [2] indicates that chromophores of azo dyes, which are responsible for color, are the main targets of "the ozone attack" and are oxidized first. Due to this fact, the decolorization of textile wastewater by ozone is not disturbed by auxiliary agents present in the wastewater.
+
+The second method of treatment, that can be considered for the purification of salty wastewater is electrocoagulation (EC). The basis of EC is the electrochemical dissolution of a sacrificial anode material (M), which is usually iron (steel) or aluminum, and elution of the metal ion (M\({}^{\mathrm{n+}}\)) into the aqueous phase (reaction (1)) takes place:
+
+\[\mathrm{M-ne^{\cdot}\to M^{\mathrm{n+}}} \tag{1}\]
+
+At the same time, hydrogen and hydroxyl ions are generated due to the cathodic decomposition of water (2):
+
+\[\mathrm{nH_{2}O+ne^{\cdot}\to nOH^{\cdot}+\frac{n}{2}H_{2}\gamma} \tag{2}\]The electrode-processes described above (eqs. 1, 2) result in the elution of metal cations and hydroxyl anions into the reaction bulk; in this way, insoluble metal hydroxides, such as Fe(OH)\({}_{3}\) or Al(OH)\({}_{3}\), which are coagulants, can be generated in situ [5]. Numerous literature reports have proved the effectiveness of EC in dye removal from aqueous solution [16, 17, 18, 19, 20, 21, 22]. Based on these reports, a certain salt content in the reaction bulk is required to keep a low ohmic resistance during the process. Therefore, the natural presence of salts provided an advantage when industrial textile wastewater treatment by EC was investigated [5, 7, 23, 24, 8, 16, 17, 18, 19, 20, 21, 22]. Even though the salt content in the reaction bulk is considered a key parameter during EC, none of these papers presented the results for high salinity, which is characteristic of dyeing effluents (NaCl conc. 30 - 100 g/L). In most reports, the salt concentration was form 0 to 5 g/L. The highest salt concentration investigated during EC was 37.5 g/L [25]; however, the study was not focused on brine recycling after treatment.
+
+In a few papers, EC was combined with ozone, but these papers only considered synthetic wastewater, not industrial wastewater [26, 27, 28]. Based on this short literature overview, no EC study to date has focused on a real textile wastewater matrix. In fact, only equalized effluents from textile industries have been examined, and the wastewater produced after dyeing has not yet been investigated, especially wastewaters with extremely high salinity and alkalinity. Moreover, no paper was found in which ozone-assisted EC was carried out for industrial wastewater treatment. Although, a few studies investigated the recycling of ozone-treated textile wastewater for subsequent dyeing operations [29, 30, 31], such an investigation was not found for EC or ozone-assisted EC.
+
+In this study ozonation and EC, which both work well in a salty environment, were selected for the dedicated treatment of industrial dyeing effluent, that was highly polluted by residual dye, NaCl and alkaline compounds. The investigated processes were as follows:* O\({}_{3}\) (single treatment),
+* EC (single treatment),
+* EC combined with O\({}_{3}\) carried out simultaneously in one reactor as one-step process,
+* EC combined with O\({}_{3}\) run separately, one by one as two-step process.
+
+For the combined treatment, EC was selected as the first step in the sequence. The reason for this choice was to remove the pollutants of higher molecular weight (not oxidized dye molecules) by coagulation first and then, as the second step, to use ozone only for the residual dye. This way, the ozone dose could be minimized.
+
+The objects of the study were aqueous solutions and industrial wastewater containing the most popular textile dye, Reactive Black 5, in purified and industrial form, respectively.
+
+The main objective of this study was to develop a system for the purification and recycling of brine from textile wastewater. The presented treatment methods were evaluated in terms of their feasibility for industrial implementation. Color removal, mineralization, toxicity and cost of treatment were indicators for choosing the most appropriate process. However, the most important factors from an industrial point of view were short treatment time and the possibility of purified wastewater reuse. Therefore, the purified wastewater was a source of brine in subsequent dyeing operations, which were carried out on cotton fabric in four various shades. The research was conducted in close cooperation with industrial and scientific partners, and the findings of the study are planned to be implemented into industry.
+
+## 2 Experiment
+
+### Materials
+
+_Wastewater treatment_. Reactive Black 5 (RB5), with a molecular mass of 991 g/mol and \(\lambda_{\text{max}}\) at 596 nm, was obtained from Boruta-Zachem (Poland) as a purified reagent. The chemical structure of the hydrolyzed form of this dye is presented in Table 1. NaOH (AR) and HCl (AR), which were used for pH adjustment, were purchased from Stanlab (Poland) and Chempur (Poland), respectively. Na\({}_{2}\)SO\({}_{3}\) (AR) was purchased from Chempur (Poland). The substances present in the industrial textile wastewater were Bezactiv Black SNN (industrial product based on RB5, CHT Switzerland AG (Switzerland)), an industrial dyeing assistant - Perigen LDR (SAA - a mixture of naphthaleneesulfonic acid and carboxylates, Textilchemie Dr. Petry Co. (Germany)) and NaCl, NaOH, and Na\({}_{2}\)CO\({}_{3}\) (technical products). The industrial wastewater indicators are given in Table 1. The wastewater was taken directly from the dyeing machine (Thies (Germany) TRD type with an operating volume 8000 L) after the indicated dyeing process.
+
+### 2.2.Methods
+
+\begin{table}
+\begin{tabular}{l l l} \hline
+**Indicator** & **RB5 aqueous solution** & **Industrial wastewater** \\ \hline pH & 12 & 11.82 \\ Conductivity, mS/cm & 36.91 & 57.56 \\ NaCl conc., g/L & 35 & 53.66 \\ COD, mgO\({}_{2}\)/L & 915\(\pm\)4 & 1315\(\pm\)5 \\ TOC, mgO\({}_{2}\)/L & 157\(\pm\)13 & 264\(\pm\)20 \\ conc. of the dye, mg/L & 500 & 790 \\ \hline
+**Chemical structure of RB5** & & \\ \hline \end{tabular}
+\end{table}
+Table 1: Industrial wastewater characteristics_Wastewater treatment._ EC was carried out in continuous 2 L reactor with 11 exchangeable electrodes, each with an area of 100 cm\({}^{2}\). The experimental setup is presented in Figure 1. (components 1 to 7). Ferrous or aluminous electrodes were used during experiments. Before each use of the ferrous electrodes, oxide removal from the electrodes was carried out. The electrodes were placed in a 6% HCl solution, then polished and rinsed with distilled water. To clean the aluminous electrodes, polishing and rinsing was sufficient. To maintain a consistent EC treatment time, two current values were used I = 2 A for RB5 aqueous solutions and I = 10 A for wastewater (current densities, j, equal to 20 mA/cm\({}^{2}\) and 100 mA/cm\({}^{2}\), respectively). A DC laboratory power supply with adjustable voltage in the range between 0 - 150 V with the maximal current output of 25 A was used during experiment. Mixing in the reactor was ensured by recirculation flow of the reaction mixture. Due to the high flow rate, the temperature during the process was kept at an approx. constant level (23\(\pm\)5\({}^{\circ}\)C).
+
+Ozonation was carried out in the 1 L glass semi batch stirred cell. Stirring was provided by Wigo ES 21 (Poland) magnetic stirrer set to 200 r.p.m. Ozone was produced by an Ozone Generator (Poland) fed with oxygen from a compressed gas cylinder (O\({}_{2}\) purity 99.5%). The ozone concentrations at the inlet and outlet of the reactor were measured with a BMT 963 Vent analyzer. Two combinations of ozone feeding parameters were used (C\({}_{03}\) 5 mg/L, Q\({}_{in}\) 20 L/h for RB5 solution and C\({}_{03}\) 42 mg/L, Q\({}_{in}\) 40 L/h for the industrial wastewater). Na\({}_{2}\)SO\({}_{3}\) was used to quench the ozone in the samples taken from the reactor. The scheme of the ozone experimental setup is presented elsewhere [32].
+
+During the EC+O\({}_{3}\) process an ozone diffuser was placed inside the EC reactor, shown in Figure 1 (C\({}_{03}\) was 5 mg/L, Q\({}_{in}\) was 20 L/h and the current was 2 A). In the case of the EC\(\rightarrow\)O\({}_{3}\) process, EC (the current was 2 A for the RB5 aqueous solutions and 10 A for wastewater) was performed first, and ozonation (C\({}_{03}\) 5 mg/L, Q\({}_{in}\) 20 L/h for RB5 solution and C\({}_{03}\) 42 mg/L, Q\({}_{in}\) 40 L/h for the wastewater) was performed second.
+
+Color was determined by a spectrophotometer (Helios Thermo, US) and calibration plot based on the Lambert Beer law were used to determine the concentrations of the samples. COD (chemical oxygen demand) was measured with the HACH LCK 514 and 314 test kits. TOC (total organic carbon) was measured in HACH IL 550TOC-TN apparatus. The samples were diluted before COD and TOC testing to avoid the influence of salt.
+
+Toxicity tests against the marine luminescent bacteria _Vibrio fischeri_ were performed according to ISO 11348-3 [33] by using a Microtox Model 500 Analyzer (Modern Water Inc., Newark, Delaware, USA). Due to the very high toxicity of the effluent, before toxicity testing effluents were diluted 1:10 with MQ water for use as reagents. The tests were performed by running the 81.9% basic test protocol (Microtox Omni 4.2, Modern Water Inc.), which consists of nine dilutions. According to the procedure, freeze-dried-bacteria were
+
+Figure 1: Experimental setup for the combined EC+O\({}_{3}\) process. Ozone supply system: I. gas (oxygen) cylinder, II. gas dryer, III. ozone generator, IV. rotameter, V. ozone meter, VI. diffuser; EC system: 1. mono-polar anode, 2. mono-polar cathode, 3. set of bi-polar electrodes, 4. electrochemical cell, 5. power supply, 6. magnetic stirrer, 7. circulation pump.
+
+reconstituted with water to provide a stock suspension of test organisms, which was kept at 5\({}^{\circ}\)C and used to perform the test. A correction factor was applied due to the loss of luminescence of the control sample (reduction in light emitted without exposure to the toxicant). Microtox osmotic adjusting solution was added to all samples at a concentration of 10% to provide osmotic protection to the test organisms. Toxicity was expressed as EC50, the pollutant concentration reducing 50% of the initial luminescence. In this work, measurements were made at 5 and 15 min of exposures.
+
+Experimental data were calculated and mean squared error (MSE) values were determined using the software Origin(r) 9.1 Pro.
+
+_Recycling trials._ Cotton dyeing with five types of reactive dyes in various shades (listed in section 2.1) was carried out in a LABOMAT BEA-12 (laboratory dyeing machine made by Mathis AG) using undiluted purified wastewater. A typical gradient dyeing method at temperature of 60\({}^{\circ}\)C was employed. The volume of the dyeing liquor was 120 mL, and the liquor ratio was 1:12. All dyesuffs and auxiliary solutions, as well as dyeing liquors, were prepared automatically by the DOSORAMATM system.
+
+The quality of the dyed textiles was determined by the DECMC color matching parameter, measured spectrophotometrically with the DataColor(tm) system in accordance with ISO 105-J03. Moreover, the color fastnesses against washing and rubbing were investigated in accordance with ISO 105-C06 and ISO 105-X12, respectively.
+
+## 3 Results and discussion
+
+### EC study
+
+The first part of the study concerned the color removal efficiency of the EC process when iron or aluminous electrodes were used. The experiment was carried out for aqueous solution and industrial wastewater containing RB5 (characteristics in Table 1). The results obtained during this experiment indicated that EC was a highly effective treatment. In Figure 2, a very low color value, between 10 and 20%, remained after a short treatment time (10 min) in both cases, RB5 aqueous solution and industrial wastewater. However, prolonging the treatment time did not result in a noticeable color decrease. On the other hand, a comparison of the experiments performed with the use of iron and aluminous electrodes showed a higher efficiency in color removal when using the former electrodes (in an alkaline reaction medium).
+
+Even though the experimental results were in agreement with those presented in some previous papers [5], the pH values of the reaction environment during EC treatment were extremely high, higher than those in the referred papers. When the pH value of the examined medium is close to 12, as in textile wastewater, the mechanism of EC becomes more complex.
+
+Figure 2: Color removal in RB5 aqueous solution and industrial wastewater by EC using aluminum (Al) and iron (Fe) electrodes
+
+The aluminum and iron ions produced by the electrodissolution of sacrificial anodes can be converted to many forms of oxides and hydroxides. Depending on pH, various forms of iron, Fe\({}^{2+}\), Fe\({}^{3+}\) Fe(OH)\({}_{2}^{+}\), Fe(OH)\({}^{2+}\), Fe(OH)\({}^{+}\), Fe(OH)\({}_{2}\), Fe(OH)\({}_{3}\), Fe(OH)\({}_{4}^{-}\), and aluminum, Al\({}^{3+}\) Al(OH)\({}_{2}^{+}\), Al(OH)\({}^{2+}\), Al(OH)\({}_{3}\), Al(OH)\({}_{4}^{-}\) coexist in the reaction bulk at different equilibrium ratios [5]. When the pH of the reaction bulk is higher than 11, the content of insoluble hydroxides decreases, and more soluble hydroxides appear, such as Fe(OH)\({}_{4}^{-}\) or Al(OH)\({}_{4}^{-}\)[5]. Moreover, in a high pH environment, some hydroxides, oxides, hydrates and polymeric species like Fe(H\({}_{2}\)O)\({}_{6}^{3+}\), Fe(H\({}_{2}\)O)\({}_{5}\)(OH)\({}^{2+}\), Fe(H\({}_{2}\)O)\({}_{4}^{+}\), Fe(H\({}_{2}\)O)\({}_{8}\)(OH)\({}_{2}^{4+}\) and Fe(H\({}_{2}\)O)\({}_{6}\)(OH)\({}_{4}^{2+}\) as well as Al(H\({}_{2}\)O)\({}_{6}^{3+}\), Al(H\({}_{2}\)O)\({}_{5}\)(OH)\({}^{2+}\), Al\({}_{2}\)(OH)\({}_{2}^{4+}\), Al\({}_{6}\)(OH)\({}_{15}^{3+}\) and Al\({}_{8}\)(OH)\({}_{20}^{4+}\), can be more readily produced than in a moderate pH environment [5, 34]. In an alkaline environment (pH \(>\) 10) amphoteric hydroxides, such as Fe(OH)\({}_{3}\) and Al(OH)\({}_{3}\), can be transformed into hydroxy ferrates, Na[Fe(OH)\({}_{4}\)] or Na\({}_{5}\)[Fe(OH)\({}_{8}\)] \(\times\) 5 - 6 H\({}_{2}\)O, or hydroxy aluminate, Na[Al(OH)\({}_{4}\)], respectively. These compounds are able to adsorb the dye molecules and the side products of EC while they are in liquid phase in dissolved form.
+
+The consequence of the described issue can be a lack of progressive color removal when increasing the time of EC treatment. A kind of equilibrium state between the dye in the liquid and solid phases was observed with the use of both iron and aluminum electrodes, which may be caused by the presence of metal (Fe and Al) compounds in soluble form. Moreover, the equilibrium ratio can slightly change over time, which is the reason why the color value is higher after 20 min than after 10 min of treatment (Figure 2).
+
+The lower efficiency of the aluminum electrodes can be explained by the high pH of the reaction mixture as well. Extremely high sludge production via EC with these electrodes was observed. The actual dissolution of aluminum electrodes was much higher than the theoretical value, obtained from Faraday's law. Due to the alkaline environment, the aluminum electrodes self-dissolved into sodium aluminates, according to reaction (3):\[2\mathrm{Al}+6\mathrm{OH}+3\mathrm{H}_{2}\mathrm{O}\rightarrow\mathrm{Al}(\mathrm{ OH})_{6}^{3-}+3\mathrm{H}_{2} \tag{3}\]
+
+The aluminates were much weaker coagulants than the aluminum hydroxides. The production of aluminates may be motivated by a decrease in pH value, which was measured within the EC system with aluminum electrodes (pH\({}_{\mathrm{initial}}\) 11.82, pH\({}_{\mathrm{final}}\) 10.58). When iron electrodes were used, the pH of the reaction mixture was stable, and the weight of the dissolved electrodes was close to the theoretical value (theoretical m\({}_{\mathrm{Fe}}\) 2.604 g, actual m\({}_{\mathrm{Fe}}\) 2.40 g; theoretical m\({}_{\mathrm{Al}}\) 0.84 g, actual m\({}_{\mathrm{Al}}\) 6.04 g, when EC time was 30 min and was 50 mA/cm\({}^{2}\)). Because the use of iron electrodes resulted in a higher and faster decolorization rate, iron electrodes were used in further stages of the study.
+
+### Coupling of ozone and EC
+
+The main purpose of the study was to compare EC, O\({}_{3}\), EC+O\({}_{3}\) (one-step process) and EC\(\rightarrow\)O\({}_{3}\) (two-step process) treatment methods. The color removal of RB5 aqueous solutions was the indicator of treatment efficiency, and the results are presented in Figure 3 A.
+
+When EC was applied as a single treatment, 10 minutes of the process resulted in ca. 85% color removal from RB5 aqueous solution. Therefore, EC can be considered as an effective treatment. However, after the reported time, no further color was removed, as previously explained in section 3.1; the cause of this phenomenon was an extremely high pH value during EC (pH 12). A greater than 85% color reduction could not be achieved using EC only, even after the prolongation of treatment time.
+
+When O\({}_{3}\) was investigated as a single treatment, complete decolorization was achieved, and a relatively long time (50 minutes) was required to remove more than 90% of the color. The mechanism of decolorization by O\({}_{3}\) in an alkaline medium and the presence of NaCl was determined to be oxidation via ozone and hydroxyl radical; this mechanism was described in our previous papers [2, 3, 4].
+
+In contrast to O\({}_{3}\), the EC+O\({}_{3}\) and EC\(\rightarrow\)O\({}_{3}\) processes were very efficient in color removal more than 90% removal in less than 20 min was achieved. Coupling EC with O\({}_{3}\) during EC+O\({}_{3}\) treatment brings benefits from the simultaneous oxidation and coagulation of dye molecules and by-products. However, considerations such as the occurrence of oxidative/reductive environments, floc formation, electrode dissolution and ferrates residual need to be discussed in further depth. There are only a few literature reports considering the simultaneous use of EC and O\({}_{3}\) for dye removal [26, 27, 28, 35], and all indicated a synergetic effect from the coupled treatment. All authors in these papers attributed the enhanced efficiency of EC+O\({}_{3}\) to Fe(OH)\({}_{n}\) coagulation combined with extra oxidation caused by hydroxyl radical produced in the catalytic O\({}_{2}\)/Fe\({}^{2+}\) system. This phenomenon could partially take place in this study as well. However, during both EC and EC+O\({}_{3}\) the reduced form of RB5, the hydrazine form, could be identified. The reduction of RB5 azo bonds to hydrazine bonds occurred, and a characteristic violet color was observed during the transitional stage of the process. By using thin layer chromatography (TLC), the presence of two amines as side-products of reductive conditions in the postreaction mixture was indicated (results not shown). Therefore, during the EC process, reducing conditions, which are caused by hydrogen evolution at the cathode, predominate.
+
+This observation is even more surprising when taking into account the extremely high NaCl content in the reaction mixture. In the presence of NaCl, the elution of chlorine gas on the anode surface is highly probable (reaction (4)):
+
+\[\text{2nCl}^{\text{-}}\text{ - }\text{ 2ne}^{\text{-}}\rightarrow\text{nCl}_{2} \tag{4}\]When NaOH is present at the same time in the reaction bulk, the production of sodium hypochlorite can take place (5), followed by a disproportionation reaction (6) [36]:
+
+\[\text{Cl}_{2}+2\text{NaOH}\rightarrow\text{NaOCI}+\text{NaCl}\]
+
+\[\text{NaOCI}\rightarrow\text{Na}^{+}+\text{ClO}^{\cdot}\]
+
+Both chlorine gas and sodium hypochlorite are strong oxidants, and global EC efficiency may be enhanced due to their occurrence [36]; however, the oxidative potential of the ClO ion (0.88 V), which dominates in an alkaline reaction medium (reaction (6)), is far lower than that of hypochlorite acid (1.49 V), which occurs in an acidic environment [37]. Moreover, it is highly probable that the anodic production of chlorine gas is a limiting factor for the elution of oxygen, which has a lower standard electrode potential and therefore could have not appeared on the anode surface. In this way oxidative conditions could be limited during EC.
+
+In summary, reductive conditions characterize the EC process more strongly than oxidative conditions, despite the fact that the formation of oxidative chlorine species cannot be excluded. The production of a reduced transitional form of dye could be observed even during EC+O\({}_{3}\) process. Therefore, when selection of treatment method is considered the reductive conditions during the EC can be an advantage and then oxidative conditions during O\({}_{3}\) treatment can be used (when the treatments performed as sequential operations in the case of EC+O\({}_{3}\)). At the same time, for EC\(\rightarrow\)O\({}_{3}\), the residual ferrous compounds can be removed by filtration before the O\({}_{3}\) step; furthermore, corrosion from ozone treatment can be limited by performing ozonation as a subsequent operation for shorter time.
+
+Figure 3 B shows the advantage of EC coupled with ozonation over ozonation performed as a single process, when considering ozone dose and treatment time. The ozone dose was calculated following Chung and Kim [38] using equation (7):
+
+EC\(\rightarrow\)O\({}_{3}\) is more efficient than EC+O\({}_{3}\) when treatment time and ozone dose are considered. Dosing of ozone directly into the EC reactor can lead to the higher consumption of ozone. It is highly probable that the majority of ozone in EC+O\({}_{3}\) is used in unproductive way, which results in a longer treatment time in comparison with the EC\(\rightarrow\)O\({}_{3}\) process. In contrast to EC+O\({}_{3}\), in the EC\(\rightarrow\)O\({}_{3}\) system, the sludge was separated before the O\({}_{3}\) step by filtration. Therefore, this unproductive use of ozone in EC+O\({}_{3}\) could be caused by additional oxidation of the sludge, as well as decomposition of the electrode surface. Due to this fact, the corrosivity of ozone should also be taken into consideration in the case of the EC+O\({}_{3}\) process. It is highly probable that electrodes can be consumed faster during such a process, and the advantage of extra hydroxyl radical production reported by Song et al., 2007, 2008 [27, 28], is not significant in comparison with the results of the EC\(\rightarrow\)O\({}_{3}\) process.
+
+Using a bubble diffuser to apply ozone gas inside the EC reactor can negatively affect floc formation; this effect could be a reason why Behin and coworkers [26] applied ozone-assisted EC in a reactor equipped with a separated cell (for separation of the O\({}_{3}\) and EC processes).
+
+The disadvantages of EC\(\rightarrow\)O\({}_{3}\) are the need to install an additional reactor and a pump and the additional energy and time needed for wastewater transfer.
+
+The EC\(\rightarrow\)O\({}_{3}\) process was observed to be the most advantageous in the processing of RB5 aqueous solution; therefore, this process was investigated for industrial wastewater treatment.
+
+Figure 3: EC, ozonation and EC-ozonation (one-step and two-step) of RB5 aqueous solutions: A) relative color removal vs time, B) ozone dose (mg/L) and treatment time required to achieve more than 90% color removal
+
+The results of EC\(\rightarrow\)O\({}_{3}\) were compared to those obtained using O\({}_{3}\). In Figure 4 A, a long treatment time was required to remove color from industrial wastewater by O\({}_{3}\). In Figure 4 B, the coupling of EC and O\({}_{3}\) allowed the treatment efficiency to be significantly increased. A time of 8 min. of EC treatment was enough to reach 90% color removal in wastewater, and prolonging the EC process was pointless because of the equilibrium state between the dye in the liquid and solid phases as described earlier in section 3.1. The use of ozone as the second step of this treatment resulted in the very high removal of residual color in a very short time (10 min of ozonation was enough because most of dye had been previously removed by EC). In this way, the overall treatment time was decreased from 60 min (in the case of ozonation only) to 18 min for two-step EC-ozonation process. At the same time, the applied total ozone dose was reduced from 1.68 to 0.28 gO\({}_{3}\)/L.
+
+### Mineralization
+
+A preliminary assessment of mineralization was conducted for all tested treatment methods. Figure 5 presents the COD/COD\({}_{0}\) and TOC/TOC\({}_{0}\) values calculated for the treatments of RB5 aqueous solutions and industrial wastewater, in which more than 90% color removal was achieved (and ca. 85% in the case of the EC process). All treatments resulted in partial COD removal; however, a decrease in TOC, which is considered equivalent to mineralization, was
+
+Figure 4: Color removal in industrial wastewater by A) O\({}_{3}\), B) EC\(\rightarrow\) O\({}_{3}\)not observed for EC and O\({}_{3}\) performed as single processes in RB5 solutions. The coupling of EC with O\({}_{3}\) resulted in significantly higher COD and TOC removal, and synergy was observed in this case.
+
+Based on the presented results, the degree of color removal was greater than the degree of mineralization. This observation is in agreement with our previous studies concerning O\({}_{3}\) treatment of textile dyes [2, 3]. Although it can be assumed that decolorization is the first stage the oxidative decomposition of dye during the O\({}_{3}\) process, in the case of EC, the issue seems to be more complex. EC is a process in which the coagulation, flotation and decomposition of pollutants can take place simultaneously [34]. Even though color removal was observed, the cause of this removal is not only the transfer of dye into the solid phase by coagulation. As mentioned in section 3.2, some dye transformation was observed during EC, as was the formation of reductive by-products. The lack of TOC removal may be caused by the occurrence of some by-products, which could have reminded in the liquid phase (the reaction mixture). At the same time, an increase in the BOD/COD ratio was observed for the EC treatment of RB5 (results not shown), which also confirms this observation.
+
+More than 40% of COD was removed from the RB5 solutions and wastewater when EC was coupled with O\({}_{3}\), and the TOC removal was significant as well. The highest TOC removal was achieved for EC+O\({}_{3}\), which may be the cumulative outcome of oxidative-reductive conditions as well as an extremely high chlorine concentration and alkalinity. These mechanisms need more detailed investigation; despite, the existence of a few literatures reports of dyes treatment by ozone-assisted EC [26, 27, 28, 35], the oxidative degradation mechanism proposed in these reports seems to be inadequately comprehensive and there is a need to investigate the process in further depth.
+
+### 3.4.Cost evaluation
+
+Both EC and O3, which were investigated in this study, require electrical power for operation and can be considered high energy-consuming methods. As far as industrial implementation is concerned, the efficiency and the cost of treatment are also determinants of method selection. When considering the results of laboratory-scale experiments, the electrical energy per order (EE/O) is an indicator that can give preliminary information about the consumption of energy during treatment when a specified purification level is required. Based on previous work by Arislan Alaton et al. [39] and Azbar et al. [40], equation (8) was used to calculate EE/O values.
+
+\[EE/O=\frac{P\times t\times 1000}{V\times 60\times\log\frac{C_{init}}{C_{fin}}}{,(\text{kWh~{}m}^{-3}~{})} \tag{8}\]
+
+Figure 5: Mineralization expressed as values of COD/COD0 and TOC/TOC0 calculated for RB5 aqueous solution and industrial wastewaterwhere P is electric power (kW), t is treatment time (h), V is the volume of treated wastewater (L), and Cinit and Cfin are the initial and final concentration of contaminant, respectively (g/m-3).
+
+In this study, the purification level for method comparison was 90% color reduction (as previously described in section 3.2). Therefore, the value of log \(\frac{C_{init}}{C_{fin}}\) used to calculate the EE/O indicator was equal to 1, and the treatment time (t) and electric power (P) corresponded to 90% color reduction (and 85% for EC).
+
+\begin{table}
+\begin{tabular}{c c c c c c c} \hline \hline A) & RB5 & & & & & \\ \hline Treatment & U, V & I, A & t, min. & V, L & log (Cinit/Cfin) & EE/O, kWh m-3 \\ \hline EC & 12.5 & 2 & 10 & 1 & 0.82 & 5.08 \\ O3 & 154 & 0.057 & 45 & 1 & 1 & 6.58 \\ one-step: EC & 12.5 & 2 & 18 & 1 & 1 & 7.5 \\ O3 & 154 & 0.057 & 18 & 1 & 1 & 2.63 \\ EC+ O3 & & & & & & 10.13 \\ two-step EC & 12.5 & 2 & 8 & 1 & 1 & 3.33 \\ O3 & 154 & 0.057 & 3 & 1 & 1 & 0.44 \\ EC\(\rightarrow\)O3 & & & & & & 3.77 \\ \hline B) & wastewater & & & & & \\ \hline Treatment & U, V & I, A & t, min & V, L & log (Cinit/Cfin) & EE/O, kWh m-3 \\ \hline EC & 37.5 & 10 & 8 & 1 & 0.82 & 60.98 \\ O3 & 230 & 0.164 & 45 & 1 & 1 & 28.29 \\ one-step EC & 37.5 & 10 & 18 & 1 & 1 & 112.5 \\ O3 & 154 & 0.057 & 18 & 1 & 1 & 2.63 \\ EC+ O3 & & & & & & 115.13 \\ two-step: EC & 37.5 & 10 & 8 & 1 & 1 & 50 \\ O3 & 230 & 0.164 & 10 & 1 & 1 & 6.29 \\ EC\(\rightarrow\)O3 & & & & & & 56.29 \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Operating parameters and EE/O values for the EC, O3, EC+O3 and EC\(\rightarrow\)O3 processesBased on the EE/O values presented in Table 2, EC is highly energy-consuming treatment method, especially when a higher current is used (10 A). Unfortunately, the current intensity I is the main operating factor in EC treatment, and the consequence is high energy consumption. Therefore, the treatment time of EC should be as short as possible to make this process economically justified. More advanced study of the use of a lower current supply or reverse power supply, which prevents electrodes passivation, could be an option to decrease energy consumption within EC. In both RB5 aqueous solution and industrial wastewater, the EE/O values were more than twice as high for the EC+O\({}_{3}\) process than for EC\(\rightarrow\)O\({}_{3}\), and when industrial application is considered, the costs would be 11,86 and 5.80 USD m\({}^{-3}\), respectively. However, cost evaluations based on laboratory results can be only roughly estimated.
+
+### 3.5 Toxicity
+
+NaCl is often used as a supporting electrolyte. As presented in Table 1, both the RB5 solution and the industrial wastewater contained very high concentrations of chloride ions (21.24 g CI- and 32.56 g CI-, respectively). The electrochemical process of anodic oxidation converts chloride ions into to active chlorine, which can decolorize dyes [14, 41, 42, 43, 44]. Moreover, free chlorine oxidizes the anodes to generate metal oxide coagulants: ferrous iron is converted to insoluble ferric iron [14]. In addition, Cl- could be oxidized to Cl\({}_{2}\)(g) in an acidic medium or to ClO in alkaline water [14, 44]. Based on these processes, high NaCl concentrations not only increase conductivity but also contribute strong oxidizing agents. Decolorization of dye wastewater is obtained via EC, and after treatment, residual chlorine/hypochlorite and toxic may occur [14, 41, 42, 44]. However, in the literature, much lower concentrations of chloride electrolytes are usually studied (approximately 1 - 5 gNaC/L [14, 41, 42, 43, 44]). To evaluate the toxicities of the decolorized solutions after all processes, the _V. fischeri_ light inhibition test was employed. A toxicity assessment was carried out for samples containing RB5 after 90% decolorization (after 45 min of O\({}_{3}\), 10 min of EC, 16.7 min of EC+O\({}_{3}\), 18 min of EC\(\rightarrow\)O\({}_{3}\)) and for wastewater samples after at least 80% of wastewater decolorization (after 30 min of O3, 8 min of EC, 18 min of EC+O3, and 18 min of EC\(\rightarrow\)O3).
+
+As seen in Figure 6, the RB5 solution and industrial wastewater indicated very similar EC50 values (16.06% and 17.71%, respectively). Although the decolorization of the RB5 aqueous solution was equal to 90%, the toxicity surprisingly increased for all processes, even for O3. After 45 min of O3 treatment of RB5 solution, the EC50 was 7.86%, while after 10 min of EC treatment, the EC50 was 0.419%. The low EC50 value after EC is related to toxic products such as chlorine/hypochlorite and dye by-products [14, 42]. The combined EC and ozonation process also indicated high toxicity; however, after treatment, the toxicity was lower than that observed in the case of single EC. Moreover, there seems to be no difference between EC+O3 and EC\(\rightarrow\)O3 (EC50 values were equal to 1.583% and 1.435%, respectively). When industrial wastewater is considered, O3 decreased the toxicity (EC50 increased to 38.76%), but after EC, a slight improvement was observed (EC50 19.255 %). This differences could indicate that after 8 minutes of EC, the process should be stopped, not because the highest decolorization is reached but because the lowest toxicity is reached. However, in the case of combined EC and ozonation, the toxicity decreased, but there was still no difference between EC+O3 and
+
+Figure 6: Toxicity assessment of RB5 solution and industrial wastewater toward _V. fischeri_ after combined EC and ozonation
+
+EC\(\rightarrow\)O\({}_{3}\) from a toxicity perspective (EC\({}_{50}\) values were equal to 9.002% and 10.19%, respectively). The obtained results show that lower toxicity was obtained when industrial wastewater was treated. Furthermore, EC did not to lead to higher toxicity. These processes should be further investigated in order to explain this strange behaviour in toxicity changes.
+
+### Recycling trials
+
+The possibility of recycling wastewater treated by the EC\(\rightarrow\)O\({}_{3}\) process was investigated, as an extremely important point of this research. Dyeing was performed using purified wastewater without any dilution. The quality of the color was assessed using a DE\({}_{\text{cmc}}\) indicator, which demonstrates the difference between the color of the calibration sample and the color of the test sample. The standards were textile materials dyed with fresh water.
+
+The values of the DE\({}_{\text{cmc}}\) color matching parameters measured according to ISO 105-J03 are presented in Table 3. The obtained DE\({}_{\text{cmc}}\) values for the samples of cotton fabrics dyed in various shades with recycled brine were between 0. 28 and 0.98 and were below the limiting value equal to 1.0. Based on these results, the differences in color between standards, dyed with fresh water, and samples, dyed with recycled brine, were very low, and only an experienced observer would notice a slight difference in shades. Moreover, Table 3 presents the results of color fastness against washing and rubbing in accordance with ISO 105-J03, ISO 105-C06 and ISO 105-X12. The obtained color fastness parameters are typical for textiles dyed with reactive dyes, and the fastness of recycled samples was not worse than that of standard samples. Therefore, the use of recycled brine as a dyeing liquor had no adverse effect on shades and color fastnesses. When wastewater reuse for next dyeing is considered, color removal during treatment is the most important factor, and mineralization (presented in section 3.3) seems to be a secondary factor.
+
+These promising results may be a motivation for further, more advanced human-ecological research, e.g., analysis according to the Oeko-tex(r) Standard and a more detailed toxicity assessment based on the preliminary results discussed in section 3.4.
+
+## 4 Conclusions
+
+Based on this study, the combination of EC and ozonation, both as one-step (EC+O\({}_{3}\)) and two-step (EC\(\rightarrow\)O\({}_{3}\)) treatments, gave very good results in color removal and moderate results in the mineralization of RB5 aqueous solution and wastewater containing this dye. The use of EC as a single process did not give satisfactory color removal even after an extended treatment time. Ozonation carried out as a single operation required a long treatment time to achieve high color removal. At the same time, cost analysis showed an advantage of EC\(\rightarrow\)O\({}_{3}\) in the cost of the two-step process was less than half that of EC+O\({}_{3}\). Therefore, as a general remark, the application of EC a short time as the first step and then O\({}_{3}\) as the second step of treatment can be recommended for industrial wastewater treatment. In accordance with this approach, the ozonation time could be reduced from 60 minutes (in the case of ozonation only) to 10 minutes (when EC is coupled with ozonation). Thus, the total ozone dose was reduced by more than 5 times, while the final color removal was 98%.
+
+\begin{table}
+\begin{tabular}{c c c c c c c c} \hline \hline  & & & & & Fastness against & & & \\ \cline{4-8} Type of sample & & & & & & & & \\ \cline{4-8} Synozol Yellow KHL (C.I. Reactive Yellow 145) & 2 & 0.44 & 5 & 5 & 4/5 & 4/5 & 4 & 4 \\ Synozol Red K3-BS (C.I. Reactive Red 195) & 2 & 0.28 & 5 & 5 & 4 & 4 & 3/4 & 4 \\ Synozol Blue KBR (C.I. Reactive Blue 221) & 2 & 0.50 & 5 & 5 & 4/5 & 4/5 & 4/5 \\ Bezaktiv Black SNN (C.I. Reactive Black 5) & 6 & 0.70 & 5 & 4/5 & 4/5 & 3 & 3/4 \\ \cline{4-8} Bezaktiv Black SNN (C.I. Reactive Black 5) & 8 & 0.98 & 5 & 5 & 3 & 3 & 2/3 & 2/3 \\ \hline \hline \end{tabular}
+\end{table}
+Table 3: DE and color fastnesses against washing and rubbing in accordance with ISO105-J03 ISO 105-C06 and ISO 105-X12The experiment where EC\(\rightarrow\)O\({}_{3}\) was used led to the production of ready-to-use brine from wastewater, which was successfully recycled for use in a next textile dyeing processes. The values of the DE\({}_{\text{cmc}}\) color matching parameter and color fastness were very promising and proved that the use of recycled wastewater do not have an adverse influence on the shade and durability of color in textiles. The moderate mineralization achieved in the experiment was recognized as a secondary factor because it had no influence on dyeing quality, however, the occurrence of by-products and the toxicity of the purified brine need further investigation.
+
+## References
+
+* [1] E. Kalliala, P. Talvenmaa, Environmental profile of textile wet processing in Finland, J. Clean. Prod. 8 (2000) 143-154. doi:10.1016/S0959-6526(99)00313-3.
+* [2] L. Bilinska, M. Gmurek, S. Ledakowicz, L. Bilinska, Textile wastewater treatment by AOPs for brine reuse, 9 (2017) 420-428. doi:10.1016/j.psep.2017.04.019.
+* Biodegradability, toxicity and cost assessment, Chem. Eng. J. 306 (2016), doi:10.1016/j.cej.2016.07.100.
+* [4] L. Bilinska, M. Gmurek, S. Ledakowicz, Application of advanced oxidation technologies for decolorization and mineralization of textile wastewaters, J. Adv. Oxid. Technol. 18 (2015).
+* [5] M. Kobya, O.T. Can, M. Bayramoglu, Treatment of textile wastewaters by electrocoagulation using iron and aluminum electrodes, J. Hazard. Mater. 100 (2003) 163-178. doi:10.1016/S0304-3894(03)00102-X.
+* [6] I.A. Senji, M. Ozacar, The decolorization of C.I. Reactive Black 5 in aqueous solution by electrocoagulation using sacrificial iron electrodes, J. Hazard. Mater. 161 (2009) 1369-1376. doi:10.1016/j.jhazmat.2008.04.100.
+* [7] A.S. Fajardo, R.C. Martins, D.R. Silva, C.A. Martinez-Huitle, R.M. Quinta-Ferreira, Dye wastewaters treatment using batch and recirculation flow electrocoagulation systems, J. Electroanal. Chem. 801 (2017) 30-37. doi:10.1016/j.jelchem.2017.07.015.
+* [8] B. Merzouk, K. Madani, A. Sekki, Using electrocoagulation-electroflotation technology to treat synthetic solution and textile wastewater, two case studies, Desalination. 250 (2010) 573-577. doi:10.1016/j.desal.2009.09.026.
+* [9] D.G. Bassyouni, H.A. Hamad, E.S.Z. El-Ashtoukhy, N.K. Amin, M.M.A. El-Latif, Comparative performance of anodic oxidation and electrocoagulation as clean processes for electrocatalytic degradation of diazo dye Acid Brown 14 in aqueous medium, J. Hazard. Mater. 335 (2017) 178-187. doi:10.1016/j.jhazmat.2017.04.045.
+* [10] B.K. Nandi, S. Patel, Effects of operational parameters on the removal of brilliant green dye from aqueous solutions by electrocoagulation, Arab. J. Chem. 10 (2017) S2961-S2968. doi:10.1016/j.arabjc.2013.11.032.
+* [11] M.K. Mbacke, C. Kane, N.O. Diallo, C.M. Diop, F. Chauvet, M. Comtat, T. Tzedakis, Electrocoagulation process applied on pollutants treatment- experimental optimization and fundamental investigation of the crystal violet dye removal, J. Environ. Chem. Eng. 4 (2016) 4001-4011. doi:10.1016/j.jece.2016.09.002.
+* [12] S. Aoudj, A. Khelifa, N. Drouiche, M. Hecini, H. Hamitouche, Electrocoagulation process applied to wastewater containing dyes from textile industry, Chem. Eng. Process. Process Intensif. 49 (2010) 1176-1182. doi:10.1016/j.cep.2010.08.019.
+* [13] C. Phalakornkule, S. Polgumhang, W. Tongdaung, B. Karakat, T. Nayut, Electrocoagulation of blue reactive, red disperse and mixed dyes, and application in treating textile effluent, J. Environ. Manage. 91 (2010) 918-926. doi:10.1016/j.jenvman.2009.11.008.
+* [14] M.C. Wei, K.S. Wang, C.L. Huang, C.W. Chiang, T.J. Chang, S.S. Lee, S.H. Chang, Improvement of textile dye removal by electrocoagulation with low-cost steel wool cathode reactor, Chem. Eng. J. 192 (2012) 37-44. doi:10.1016/j.cej.2012.03.086.
+* [15] M. Kobya, E. Demirbas, O.T. Can, M. Bayramoglu, Treatment of levafix orange textile dye solution by electrocoagulation, J. Hazard. Mater. 132 (2006) 183-188. doi:10.1016/j.jhazmat.2005.07.084.
+* [16] M.M. Hossain, M.I. Mahmud, M.S. Parvez, H.M. Cho, Impact of current density, operating time and pH of textile wastewater treatment by electrocoagulation process, Environ. Eng. Res. 18 (2013) 157-161. doi:10.4491/eer.2013.18.3.157.
+* [17] A.S. Naje, S. Chelliapan, Z. Zakaria, S.A. Abbas, Electrocoagulation using a rotated anode: A novel reactor design for textile wastewater treatment, J. Environ. Manage. 176 (2016) 34-44. doi:10.1016/j.jenvman.2016.03.034.
+* [18] F. Ghanbari, M. Moradi, A comparative study of electrocoagulation, electrochemical Fenton, electro-Fenton and peroxi-coagulation for decolorization of real textile wastewater electrical energy consumption and biodegradability improvement, J. Environ. Chem. Eng. 3 (2015) 499-506. doi:10.1016/j.jece.2014.12.018.
+* [19] D.R. Manenti, A.N. Modenes, P.A. Soares, F.R. Espinoza-Quinones, R.A.R. Boaventura, R. Bergamasco, V.J.P. Vilar, Assessment of a multistage system based on electrocoagulation, solar photo-Fenton and biological oxidation processes for real textile wastewater treatment, Chem. Eng. J. 252 (2014) 120-130. doi:10.1016/j.cej.2014.04.096.
+* [20] M. Kobya, E. Gengec, E. Demirbas, Operating parameters and costs assessments of a real dyehouse wastewater effluent treated by a continuous electrocoagulation process, Chem. Eng. Process. Process Intensif. 101 (2015) 87-100. doi:10.1016/j.cep.2015.11.012.
+* [21] I. Zongo, A.H. Maiga, J. Wethe, G. Valentin, J.P. Leclerc, G. Paternotte, F. Lapicque, Electrocoagulation for the treatment of textile wastewaters with Al or Fe electrodes: Compared variations of COD levels, turbidity and absorbance, J. Hazard. Mater. 169 (2009) 70-76. doi:10.1016/j.jhazmat.2009.03.072.
+* [22] A. Aouni, C. Fersi, M. Ben Sik Ali, M. Dhahbi, Treatment of textile wastewater by a hybrid electrocoagulation/nanofiltration process, J. Hazard. Mater. 168 (2009) 868-874. doi:10.1016/j.jhazmat.2009.02.112.
+* [23] M. Bayramoglu, M. Eyvaz, M. Kobya, Treatment of the textile wastewater by electrocoagulation. Economical evaluation, Chem. Eng. J. 128 (2007) 155-161. doi:10.1016/j.cej.2006.10.008.
+* [24] S. Zodi, O. Potier, F. Lapicque, J.P. Leclerc, Treatment of the textile wastewaters by electrocoagulation: Effect of operating parameters on the sludge settling characteristics, Sep. Purif. Technol. 69 (2009) 29-36. doi:10.1016/j.seppur.2009.06.028.
+* [25] B.K. Korbahti, A. Tanyolac, Electrochemical treatment of simulated textile wastewater with industrial components and Levafix Blue CA reactive dye: Optimization through response surface methodology, J. Hazard. Mater. 151 (2008) 422-431. doi:10.1016/j.jhazmat.2007.06.010.
+* [26] J. Behin, N. Farhadian, M. Ahmadi, M. Parvizi, Ozone assisted electrocoagulation in a rectangular internal-loop airlift reactor: Application to decolorization of acid dye, J. Water Process Eng. 8 (2015) 171-178. doi:10.1016/j.jwe.2015.10.003.
+* [27] S. Song, Z. He, J. Qiu, L. Xu, J. Chen, Ozone assisted electrocoagulation for decolorization of C.I. Reactive Black 5 in aqueous solution: An investigation of the effect of operational parameters, Sep. Purif. Technol. 55 (2007) 238-245. doi:10.1016/j.seppur.2006.12.013.
+* [28] S. Song, J. Yao, Z. He, J. Qiu, J. Chen, Effect of operational parameters on the decolorization of C.I. Reactive Blue 19 in aqueous solution by ozone-enhanced electrocoagulation, J. Hazard. Mater. 152 (2008) 204-210. doi:10.1016/j.jhazmat.2007.06.104.
+* [29] E. Hu, S. Shang, X. Tao, S. Jiang, K. Chiu, Regeneration and reuse of highly polluting textile dyeing effluents through catalytic ozonation with carbon aerogel catalysts, J. Clean. Prod. 137 (2016) 1055-1065. doi:10.1016/j.jclepro.2016.07.194.
+* [30] M. Senthilkumar, M. Muthukumar, Studies on the possibility of recycling reactive dye bath effluent after decolouration using ozone, Dye. Pigment. 72 (2007) 251-255. doi:10.1016/j.dyepig.2005.08.019.
+* [31] S. Ledakowicz, R. Zylla, K. Pazdzior, J. Wrebiak, J. Sojka-Ledakowicz, Integration of Ozonation and Biological Treatment of Industrial Wastewater From Dyehouse, Ozone Sci. Eng. 39 (2017) 357-365. doi:10.1080/01919512.2017.1321980.
+* Doctoral thesis, 2017.
+* Water quality -
+- Determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (Luminescent bacteria test) -
+- Part 3: Method using freeze-dried bacteria, 2007. [https://www.iso.org/standard/40518.html](https://www.iso.org/standard/40518.html) (accessed June 8, 2018).
+* [34] S. Garcia-Segura, M.M.S.G. Eiband, J.V. de Melo, C.A. Martinez-Huitle, Electrocoagulation and advanced electrocoagulation processes: A general review about the fundamentals, emerging applications and its association with other technologies, J. Electroanal. Chem. 801 (2017) 267-299. doi:10.1016/j.jelechem.2017.07.047.
+* [35] M.A. Garcia-Morales, G. Roa-Morales, C. Barrera-Diaz, V.M. Miranda, P.B.
+
+Hernandez, T.B.P. Silva, Integrated advanced oxidation process (Ozonation) and electrocoagulation treatments for dye removal in denim effluents, Int. J. Electrochem. Sci. 8 (2013) 8752-8763.
+* [36] P. Sakthisharmila, P.N. Palanisamy, P. Manikandan, Removal of benzidine based textile dye using different metal hydroxides generated in situ electrochemical treatment-A comparative study, J. Clean. Prod. 172 (2018) 2206-2215. doi:10.1016/j.jclepro.2017.11.192.
+* [37] S. Garcia-Segura, J.D. Ocon, M.N. Chong, Electrochemical oxidation of real wastewater effluents -- A review, Process Saf. Environ. Prot. 113 (2018) 48-67. doi:10.1016/j.psep.2017.09.014.
+* [38] J. Chung, J.-O. Kim, Application of advanced oxidation processes to remove refractory compounds from dye wastewater, Desalin. Water Treat. 25 (2012) 235-240. doi:10.5004/dwt.2011.1935.
+* [39] I.A. Alaton, I.A. Balcioglu, D.W. Bahnemann, Advanced oxidation of a reactive dyebath effluent: Comparison of O3, H2O2/UV-C and TiO2/UV-A processes, Water Res. 36 (2002) 1143-1154. doi:10.1016/S0043-1354(01)00335-9.
+* [40] N. Azbar, T. Yonar, K. Kestioglu, Comparison of various advanced oxidation processes and chemical treatment methods for CCD and color removal from a polyester and acetate fiber dyeing effluent, Chemosphere. 55 (2004) 35-43. doi:10.1016/j.chemosphere.2003.10.046.
+* [41] A.A. Bukhari, Investigation of the electro-coagulation treatment process for the removal of total suspended solids and turbidity from municipal wastewater., Bioresour. Technol. 99 (2008) 914-21. doi:10.1016/j.biortech.2007.03.015.
+* [42] S.-H. Chang, K.-S. Wang, H.-H. Liang, H.-Y. Chen, H.-C. Li, T.-H. Peng, Y.-C. Su, C.-Y. Chang, Treatment of Reactive Black 5 by combined electrocoagulation-granular activated carbon adsorption-microwave regeneration process, J. Hazard. Mater. 175 (2010) 850-857. doi:10.1016/J.JHAZMAT.2009.10.088.
+* [43] B. Merzouk, B. Gourich, A. Sekki, K. Madani, C. Vial, M. Barkaoui, Studies on the decolorization of textile dye wastewater by continuous electrocoagulation process, Chem. Eng. J. 149 (2009) 207-214. doi:10.1016/J.CEJ.2008.10.018.
+* [44] M. Chafi, B. Gourich, A.H. Essadki, C. Vial, A. Fabregat, Comparison of electrocoagulation using iron and aluminium electrodes with chemical coagulation for the removal of a highly soluble acid dye, Desalination. 281 (2011) 285-292. doi:10.1016/J.DESAL.2011.08.004.

manual_annotation/200pdfs_in_mmd/10.1016_j.cep.2005.06.007.mmd

@@ -0,0 +1,363 @@
+Wastewater dye destruction using ozone-loaded Volasil(tm)245 in a continuous flow liquid-liquid/ozone system
+
+D.B. Ward, C. Tizaoui
+
+c.tizaoui@bradford.ac.uk
+
+M.J. Slater
+
+School of Engineering Design and Technology, University of Bradford, Bradford BD7 1DP, UK
+
+Received 13 July 2004; received in revised form 24 May 2005; accepted 2 June 2005
+
+###### Abstract
+
+Volasil(tm)245 (decamethylcyclopentasiloxane), originally developed as a heat exchange fluid, has been found to dissolve 10 times more ozone than does water. Extracting wastewater contaminants to ozone-loaded Volasil(tm)245 is proposed as a means of enhancing reaction kinetics and thus, providing more rapid wastewater decontamination. Using a continuous flow pilot plant, ozone-loaded Volasil(tm)245 contact was applied to dye-contaminated waters. Process kinetics and efficacy have been assessed in terms of colour, COD and TOC reductions, and also in relation to direct ozone gas contact. Design and operating conditions for the prototype Volasil(tm)245 liquid-liquid/ozone wastewater treatment plant are proposed. It was found that counter-current ozone-loading of the solvent is required instead of co-current loading. The dyes used were not extracted into the solvent and expected kinetic enhancement was not realised compared to direct gas/water treatment. Other contaminant chemicals are being studied to assess the process further.
+
+Ozone; Wastewater; Decontamination; Dyes; Volasil(tm)245; Liquid-liquid extraction; SMV static mixers +
+Footnote †: journal: Acoustics and Geoscience
+
+## 1 Introduction
+
+The industrial discharge of polluted wastewaters is the subject of an increasingly stringent and expanding array of regulatory controls, driven by public opinion, health concerns and environmental forecasts. Not least of these has been the European Union Water Framework Directive 2000/60/EC [1], which was adopted by all member states in December 2003. Such legislative pressures have motivated the development of evermore effective and innovative wastewater treatment technologies.
+
+Ozone-based technologies have proved themselves very useful in meeting the challenge of improved industrial wastewater quality. Ozone is a powerful oxidising agent with a wide range of applications. These include the treatment of waters containing organic contaminants [2], predominantly for:
+
+1. the transformation of toxic or biocidal micro-pollutants (e.g. aromatic, chloro-aromatics and specifically, pesticides),
+2. the partial oxidation of biologically refractory compounds (mostly applied as a pre-treatment to biodegradation), and
+3. the removal of colour.
+
+In order for ozone to oxidise a wastewater contaminant, the respective substances must be mutually exposed. Conventional treatment systems dissolve ozone directly into wastewater using gas-liquid contacting devises (e.g. bubbling columns, gas ejectors, etc.). However, this approach incorporates two potentially serious limitations that may severely retard the oxidation process. Firstly, current ozone generating technology is incapable of producing gas concentrations of more than several percent (even when supplied with pure oxygen) and secondly, the solubility of ozone in water is relatively poor (\(\sim\)0.2 mg/L per mg/L in contacting gas phase at 20\({}^{\circ}\)C [3]). These factors combine to ensure wastewater contaminants encounter relatively low ozone con centrations in solution and thus, may undergo low rates of breakdown as a consequence.
+
+It is proposed that oxidation kinetics may be improved by exposing contaminants to higher concentrations of ozone. The feasibility of using ozone-loaded solvents as a means of achieving this has been investigated at bench scale [4; 5; 6; 7; 8; 9; 10]. The general process involves contacting contaminated water with an immiscible solvent containing a high concentration of dissolved ozone. Prior saturation, or loading, of the solvent is achieved by contacting with ozone gas (in oxygen or air mixture). The loaded solvent is then contacted with the water phase in order to promote mass transfer of contaminants and/or ozone. The site of the reaction can be expected to depend upon the solvent/water distribution coefficient of the contaminants involved. Contaminants extracted into the solvent phase encounter an oxidant-enriched environment and hence, the potential to undergo more rapid degradation. Compounds resisting extraction may also be subject to accelerated oxidation, as ozone in the water is rapidly replenished from an elevated concentration within the solvent (i.e. by enhanced ozone mass transfer). Following adequate contact time, the two phases are separated by gravity and the solvent reloaded for re-use. The technique can be termed liquid-liquid/ozone water treatment. Fig. 1 illustrates the basic stages involved.
+
+Another potential advantage of liquid-liquid/ozone treatment lies in its ability to deliver ozone without direct gas bubbling. The bubbling of gases through a wastewater may be problematic where contaminants possess high volatility (e.g. dichloromethane). Such contact can result in sparging and thus, the transferral (i.e. escape) of compounds to the exiting gas phase. This can then necessitate an additional gas-cleaning requirement. It may also be desirable to avoid the liberation of compounds to the gas phase in cases where such substances are particularly toxic or pungent.
+
+The process is also suggested to improve the efficiency of ozonation by selectively extracting and oxidising target pollutants [7]. In conventional gas contact systems, the presence of free radical scavengers, such as carbonate (CO32-) and bicarbonate (HCO3-), can further retard the oxidation of target compounds by the wasteful consumption of oxidant in preferential reactions. The selective extraction of contaminants to an ozone-plenful, scavenger-free, phase may avoid this problem.
+
+In selecting an appropriate solvent, ozone solubility is amongst the most important criteria. This must be significantly greater within the solvent than in water and thus, allow the achievement of a much higher dissolved concentration when exposed to ozone gas of a given partial pressure. Very low solubility of the solvent in water and non-toxicity are also needed. Previous work in this field has involved the fluoro-carbon solvents FC40(tm) and FC77(tm) (3M Co.). However, more recent research has identified the heat exchange fluid Volasil(tm)245 (decamethylcyclopentasiloxane) (VWR International, UK) as a potential liquid-liquid/ozone solvent with distinct advantages over fluorocarbons [10].
+
+Ward et al. [10] examined ozone solubility within a number of fluorocarbon and polydimethylsiloxane solvents and concluded that FC77(tm) and Volasil(tm)245 to be superior in this respect. For both these solvents, ozone solubility was determined to be 10 times greater than that found in water. Volasil(tm)245 was, however, preferred due to its significantly lower vapour pressure (<5.3 mm Hg for Volasil(tm)245 but 42 mm Hg for FC77(tm) and hence, its lesser tendency to undergo evaporation during loading.
+
+Volasil(tm)245 is considered additionally suitable as it possesses a low toxicity (LD50 oral rats = 2 g/kg), low water solubility (17 mg/L) and can hold in solution a wide variety of organic compounds. Furthermore, by exposing the substance to ozone for 100 h without noticeable detriment, Ward et al. [10] demonstrated Volasil(tm)245 to be resistant to ozone attack. Some general physical properties are given in Table 1.
+
+A continuous flow Volasil(tm)245 liquid-liquid/ozone pilot rig has been developed and its performance demonstrated
+
+\begin{table}
+\begin{tabular}{l c c c} Property & Unit & Quantity & Source \\ Density & kg/m3 & 956 at 20 °C & a \\ Viscosity & g/ms & 4.0 at 20 °C & a \\ Molar mass & g/mol & 370 & a \\ Flash point & °C & 72 & a \\ Water solubility & μ\(\mu\)g/L & 17 at 20 °C & a \\ Vapor pressure & mm Hg & \textless{}5.3 at 20 °C & a \\ Interfacial tension with water & mN/m & 24 at 20 °C & b \\ Henry’s constant (Ozone) & atm./mole-fraction & 34 at 20 °C & b \\ \end{tabular}
+\end{table}
+Table 1: General physical properties of Volasil(tm)245
+
+Figure 1: Basic stages of liquid–liquid/ozone wastewater treatment.
+
+with regard to the treatment of dye-contaminated wastewaters. It is the purpose of this paper to report upon:
+
+1. the design and construction of the continuous flow liquid-liquid/ozone water treatment rig,
+2. the demonstration of rig operation with regard to various dye solutions,
+3. the determination of suitable operating conditions,
+4. the effects of liquid-liquid/ozone treatment on the various dye solutions in terms of reductions in colour, TOC and COD, and
+5. the comparison of reaction kinetics with regard to conventional ozone-gas contact.
+
+As to this, the paper consists of three closely related but discrete parts: preliminary tests, ozone-loading tests and dye destruction tests.
+
+## 2 The dyes
+
+The dye compounds used in this investigation included Drimarene Brilliant Red (Clariant UK Ltd.) and Remazol Brilliant Orange 16 (Sigma-Aldrich Co. Ltd., UK). Both these dyes contain polycyclic ring structures similar to many toxic organic compounds (e.g. dioxins, PAHs, PCBs, pesticides, etc.) and thus, their behaviour within a liquid-liquid/ozone system may reflect that of more crucial contaminants. Details as to the characteristics of both these dyes are as follows.
+
+### Drimarene Brilliant Red K-4BL, colour index no. 147
+
+It is not possible to ascertain the exact molecular formula for this dye, as the manufacturers could not release this information. However, Rozsa [11] speculated its molecular structure to be as in Fig. 2.
+
+As Drimarene Brilliant Red (DBR) is an azo dye, its structure can be surmised as having an azo-link (-N=N-), joining two aromatic rings, each containing a sulphonic group and possibly one or more aliphatic chains. Through examining the mass spectrum of the dye, Rozsa [11] also suggested its molar mass to be approximately 408 g/mol (the first ionisation peak being at \(\sim\)408 \(m/z\)).
+
+### Reactive Orange 16 (Remazol Brilliant Orange), colour index no. 177757
+
+The chemical formula of Reactive Orange 16 (RO16) is shown in Fig. 3.
+
+Archibald and Roy-Arcand [12] have given a simplified reaction mechanism for the treatment of azo-dyes with ozone (see Eq. (1)). This describes the oxidation and splitting of the azo-link and the production of a nitro compound. However, our paper does not speculate as to the specific chemical reactions taking place but comments upon the actual effects of treatment with regard to dye solution characteristics (i.e. pH, colour, TOC and COD):
+
+\[\mathrm{R}_{1}\mathrm{-N}\mathrm{=N}\mathrm{-}\mathrm{R}_{2}\rightarrow\ \mathrm{R}\mathrm{-NHOH}\ \rightarrow\ \mathrm{RNO}\ \rightarrow\ \mathrm{RNO}_{2} \tag{1}\]
+
+## 3 Preliminary dye tests
+
+Preliminary tests were carried out in order to gain a basic understanding of dye behaviour in a liquid-liquid/ozone system and thus, allow informed design of the proposed rig. Specifically, tests were devised to determine: (i) decolouration kinetics, (ii) distribution of dyes and reaction products between phases, and (iii) rapidity of solvent/water phase separation.
+
+Tests were performed at bench scale using aqueous dye solutions of DBR (200 mg/L) and RO16 (310 mg/L), respectively (made up using distilled water). Each test involved contacting 100 mL of ozone-loaded Volasil(tm)245 (\(\sim\)170 mg/L ozone concentration) with an equal volume of dye solution. Contact was carried out within a 250 mL separating funnel. Agitation of the two phases was achieved by means of a motorised impeller. In both cases, an ozone:dye molar ratio of approximately 8:1 was used. Tizaoui [13] determined that a minimum of 4 mol of ozone was required to decolour 1 mol of DBR dye. Hence, in the case of DBR, an ozone excess of 100% was allowed.
+
+Ozone-loading of the solvent was carried out in a 100 mL dreschel bottle supplied with ozone-enriched gas (0.4 L/min, 1.2 bar absolute pressure, \(\sim\)25\({}^{\circ}\)C). Ozone was produced using a BMT 803 generator (BMK Messtechnik, Germany) fed by pure oxygen. Gas exiting the dreschel bottle was passed through an ultraviolet ozone analyzer (model BMT963, BMK Messtechnik, Germany). Dissolved ozone
+
+Figure 3: Chemical structure of Reactive Orange 16.
+
+Figure 2: Assumed chemical structure of Drimarene Brilliant Red [11].
+
+concentration in the solvent was determined by applying Henry's Law (\(H\) = 34 bar/mole-fraction [10]).
+
+Following initial contact with the ozone-loaded solvent, \(\sim\)10 mL samples of the mixture were drained from the reactor after intervals of 0.5, 1, 2, 5, 10, and 30 min. Each sample was collected in a 10 mL measuring cylinder, prior charged with \(\sim\)10 mg sodium sulphite crystals (Na\({}_{2}\)SO\({}_{3}\)) (Griffin & George Ltd., UK). Sodium sulphite was added to destroy any residual ozone remaining in the sample and thus, halt the reaction at the moment of collection. Samples were then left to stand for 30 min to allow separation of the phases. After this period, 2 mL sub-samples of both solvent and aqueous phases were respectively pipetted into a spectrophotometer cell (path length, 5 mm). Using a Hewlett-Packard 8451A Diode Array Spectrophotometer, each sub-sample was analysed for light absorbance over the UV-vis range (190-700 nm).
+
+With respect to the aqueous phase of both dye solutions, Figs. 4 and 5 show the absorbance profiles obtained before and after contact with ozone-loaded solvent. As can be seen in both cases, dramatic declines in absorbance were detected
+
+Fig. 4: Absorbance profile obtained before and after contact with ozone-loaded solvent for DBR dye solution.
+
+Fig. 5: Absorbance profile obtained before and after contact with ozone-loaded solvent for RO16 dye solution.
+
+over almost the entire UV-vis range. This suggests that significant dye degradation occurred. This was corroborated by observations made during each test (by naked eye), which noted a rapid transition from vibrant colour to almost colourless solution. However, as can also be seen from Figs. 4 and 5, post-contact absorbance profiles reveal the persistence of peaks in the UV range--thus, suggesting the presence of residual reaction products.
+
+The distinctive colour of a dye is denoted by the characteristic absorbance peak/s in the visible range (i.e. 400-700 nm). In the case of DBR and RO16, these occur at approximately 540 and 494 nm, respectively (see Figs. 4 and 5, \(t\) = 0). Examining the decline of such peaks with time allows the kinetics of decolouration to be quantitatively assessed. Using Eq. (2) and absorbance data acquired from each aqueous sub-sample drained from the reactor at time \(t\), Fig. 6 shows the extent and rapidity of decolouration for both dye solutions.
+
+As can be seen from Fig. 6, approximately 90% colour loss was achieved within the first 30 s of contact in both cases. Completion of the reactions was within 1-2 min, after which time decoloration of DBR and RO16 dye solutions had reached \(\sim\)98%. Hence, an effective liquid-liquid/ozone rig must be able to afford contact times of at least this approximate duration:
+
+\[\text{decolouration (\%)}=\left(1-\frac{A_{t}}{A_{0}}\right)\times 100 \tag{2}\]
+
+where \(A\)0 is the absorbance peak height of sub-sample taken at \(t\) = 0, and \(A_{t}\) is the absorbance peak height of sub-sample taken at \(t\) = \(t\).
+
+Spectrophotometer analysis of Volasil(tm)245 sub-samples showed no absorbance peaks within the solvent phase and thus, an absence of either dye or reaction products. This would suggest both dyes and products to resist extraction to the solvent phase whilst undergoing ozone reactions in the aqueous phase. To confirm this assertion with respect to the un-reacted dye compounds, both dye solutions were contacted with Volasil(tm)245 as before but in the absence of ozone. Inspection of the solvent before and after contact could detect no colour change (solvent remained colourless and clear). Further, spectrophotometer analysis registered no presence of UV-vis absorbance peaks in the post-contact solvent, further suggesting that both dyes resist extraction.
+
+In the aftermath of all of the above tests, complete separation of both phases occurred within 2-3 min of impeller deactivation, demonstrating separation to take place within a reasonable time frame.
+
+## 4 Liquid-liquid/ozone rig design
+
+Fig. 7 shows a schematic diagram of the continuous flow liquid-liquid/ozone rig. The system comprised of four sections: (i) ozone generation, (ii) solvent-loading, (iii) solvent-water contact, and (iv) off-gas cleaning. Each section is described as follows.
+
+### Ozone generation section
+
+Ozone gas was produced using a LAB2B Ozone Generator (Ozonia Triogen Ltd., UK) fed with dry oxygen from a pressurized cylinder. A flow meter, pressure gauge and valve was used to regulate/measure gas stream conditions before entry into the solvent-loading section. To determine the ozone gas concentration, a flow sample was diverted through an ultra
+
+Figure 6: Decolouration kinetics of DBR and RO16 dyes experienced during contact with ozone-loaded solvent in preliminary tests.
+
+violet ozone analyzer (model BMT963, BMK Messtechnik, Germany).
+
+### Solvent-loading section
+
+The purpose of this section was to achieve a dissolved concentration of ozone within the solvent. The system was comprised of \(4\,\mathrm{m}\times 1\,\mathrm{m}\) vertically mounted glass tubes connected by U-bends (QVF Process Systems Ltd., UK), the tube diameter being approximately \(0.015\,\mathrm{m}\) throughout (i.e. DN15). Gas and solvent were passed through the system concurrently. For the promotion of contact, SMV static mixing elements (Sulzer (Chemtech) Ltd., UK) were mounted at the base of each tube carrying upward flow (to generate bubbles); tubes carrying downward flow were packed with \(6\,\mathrm{mm}\) glass beads. At the section exit, gas and ozone-loaded solvent were disengaged by gravity in a gas/liquid separator. Spent gas was drawn-off at the top of the separator (via a glass wool demister for the removal of solvent droplets), from where it passed into the off-gas cleaning section. Loaded-solvent was allowed to flow under head pressure from the base of the separator and directly into the solvent-water contact section.
+
+### Solvent-water contact section
+
+This section contacted ozone-loaded solvent with contaminated water and allowed exchange and/or oxidation reactions to occur. The section consisted of \(15\,\mathrm{m}\times 1\,\mathrm{m}\) vertically mounted glass tubes (\(0.015\,\mathrm{m}\) diameter, DN15) connected with U-bends (QVF Process Systems Ltd., UK). Water and solvent flowed co-currently through the system. In order to encourage mixing of the two liquid phases, each alternate tube was fitted at the base with a SMV static mixing element (Sulzer (Chemtech) Ltd., UK). Passage of the liquids through the static mixers created and maintained a dispersion of solvent droplets. Lower U-bends were fitted with drain valves to allow the extraction of point samples. Valves fitted to the upper U-bends allowed the build-up of any gas to be vented if required. Solvent/water separation was achieved by means of a cylindrical gravity separation tank (\(L=0.9\,\mathrm{m}\), \(D=0.1\,\mathrm{m}\)). The solvent, being the lighter phase, separated as
+
+Figure 7: Schematic flow diagram showing the liquid–liquid/ozone pilot rig.
+
+a supernatant layer and was selectively drained off and recycled through the system. The water phase (the subnatant) was allowed to drain to a treated water holding tank for sampling and/or disposal.
+
+### Off-gas cleaning section
+
+To facilitate destruction of any residual ozone in the exiting oxygen of the solvent-loading section (and ozone analyser), off-gas was channelled through a series of cleaning stages before being safely delivered to the laboratory air extraction system. Ozone destruction was achieved by passage through firstly, a bed of zeolite pellets (aluminium oxide, 3 mm sieve) and then a solution of potassium iodide (KI) (Avocado Research Chemicals Ltd., UK). A demister of glass wool subsequently removed any entrained droplets of KI solution before the gas was dried by means of silica gel (to prevent moisture build-up in pump and flow meter). A diaphragm pump was used to draw the gas through the system under negative pressure and so minimise the possibility of ozone escape to the laboratory atmosphere. A TX2000 personal ozone monitor (Oldham s.a., France) was used to check that safe limits were not exceeded in the working area.
+
+## Solvent-loading tests
+
+To more fully understand and therefore optimise operation of the designed co-current solvent-loading system, a series of tests was performed with the following objectives:
+
+1. determine if dissolved ozone equilibrium concentrations are achieved, and
+2. determine suitable operating conditions for solvent-loading.
+
+### Solvent-loading section
+
+In order to determine the effectiveness of solvent-loading carried out by the rig, it was necessary to know the maximum concentration of dissolved ozone acquirable by the solvent under given inlet conditions. Assuming ozone free solvent at the inlet and steady-state conditions, a mass balance over the solvent-loading section gives:
+
+\[\dot{V}_{\mathrm{g}}C_{\mathrm{g,in}}=(\dot{V}_{\mathrm{g}}C_{\mathrm{g,out}}) +(\dot{V}_{\mathrm{s}}C_{\mathrm{s,out}}) \tag{3}\]
+
+where \(V_{\mathrm{g}}\) is the gas volumetric flow rate (L/min), \(V_{\mathrm{s}}\) the solvent volumetric flow rate (L/min), \(C_{\mathrm{g,in}}\) the gas ozone concentration at inlet (mg/L), \(C_{\mathrm{g,out}}\) the gas ozone concentration at outlet (mg/L), and \(C_{\mathrm{s,out}}\) is the solvent ozone concentration at outlet (mg/L).
+
+Previous study [10] has shown that at equilibrium, the concentration of ozone within the solvent was approximately two times that of the contacting gas phase. This factor was determined using an absolute pressure of \(\sim\)1.2 bar and a solvent temperature of \(\sim\)25 \({}^{\circ}\)C [10]. Hence, assuming the rig to be operated under similar conditions, outlet gas concentration (\(C_{\mathrm{g,out}}\)) can be described by Eq. (4) when equilibrium is achieved:
+
+\[C_{\mathrm{g,out}}=\frac{C_{\mathrm{s,out}}}{2} \tag{4}\]
+
+By substituting Eq. (4) into (3) and rearranging for \(C_{\mathrm{s,out}}\), the equilibrium concentration of ozone-in-solvent can be expressed in terms of the inlet conditions (Eq. (5)):
+
+\[C_{\mathrm{s,out}}=\frac{C_{\mathrm{g,in}}}{0.5+\left(\frac{V_{\mathrm{g}}}{V_ {\mathrm{s}}}\right)^{-1}} \tag{5}\]
+
+Three sets of inlet flow conditions (A-C) were investigated for a range of gas/solvent flow ratios (i.e. 2.5, 5 and 10). For each set of flow conditions, equilibrium ozone-loading concentrations were calculated based on gas inlet concentrations of 20, 40 and 60 mg/L (NTP). The results of these calculations are given in Table 2.
+
+### Practical solvent-loading test
+
+To determine whether predicted ozone-loading concentrations could be achieved in practice, the rig was operated under each of the conditions given in Table 2. In addition to these, inlet gas pressure (absolute) and solvent temperature were regulated to 1.2 \(\pm\) 0.2 bar and 25 \(\pm\) 2 \({}^{\circ}\)C, respectively (i.e. conditions similar to those applied by Ward et al. [10] in assessing dissolved ozone equilibrium in Volasil(tm)245). To prevent dissolved ozone from being recycled along with the solvent (and thus, falsely elevate loading concentration on subsequent passages), solvent exiting the loading system was depleted of ozone by contact/reaction with an excess of DBR dye in aqueous solution in the solvent-water contact section (see Fig. 7).
+
+Under each combination of operating conditions, 10 mL solvent samples were drained from the sampling points
+
+\begin{table}
+\begin{tabular}{l l l l l l} \hline Flow condition & Gas inlet flow & Solvent inlet flow (L/min) & Predicted [ozone-in-solvent], mg/L (experimental [ozone-in-solvent] as \% of predicted) \\ \cline{3-6}  & (L/min) & flow (L/min) & 20 mg/L gas inlet & 40 mg/L gas inlet & 60 mg/L gas inlet \\ \hline A & 1.5 & 0.6 & 22 (94) & 44 (100) & 67 (92) \\ B & 1.5 & 0.3 & 29 (104) & 57 (92) & 86 (93) \\ C & 3.0 & 0.3 & 33 (101) & 67 (91) & 100 (91) \\ \hline Av. = 95\%, S.D. = 5\%. & & & & & \\ \end{tabular}
+\end{table}
+Table 2: Predicted equilibrium solvent-loading concentrations for various rig inlet conditionsinstalled in the two lower U-bends shown in Fig. 7 (sample points 1 and 2--i.e. approximately 1 and 3 m along the solvent-loading section tube length). Immediately after collection, a 2 mL sub-sample was transferred via pipette to a quartz spectrophotometer cell (path length, 5 mm). The cell was then sealed with a PTFE stopper, placed in a Hewlett-Packard 8451A Diode Array Spectrophotometer and analysed for UV absorbance at 254 nm (i.e. the wavelength most susceptible to absorption by ozone [2]).
+
+In order to interpret the acquired UV absorption data and thus, quantify the concentration of ozone in solvent samples, a calibration study was performed. Using the solvent-loading equipment described in the preliminary tests (i.e. dreschel bottle supplied with flow of ozone-in-oxygen, etc.), a quantity of Volasil(tm)245 was successively enriched with dissolved ozone to concentrations of 0, 40, 60, 80 and 120 mg/L. Samples taken from the dreschel bottle at the various concentrations were subjected to spectrophotometer analysis at 254 nm. Plotting UV absorbance data against dissolved concentration generated the required calibration curve, which was found to obey the Beer-Lambert law (i.e. ozone concentration was directly proportional to absorbance).
+
+The results of each solvent-loading test are given in Fig. 8. The figure compares experimentally obtained ozone-loading concentrations (taken at samples points 1 and 3 m along the section tube length) with those predicted at equilibrium by Eq. (5).
+
+In an attempt to quantify the degree to which experimentally obtained values match those predicted by Eq. (5) (and thus, demonstrate the achievement of dissolved equilibrium with the contacting gas phase), Table 2 also gives experimentally determined concentrations (value taken at 3 m) as a percentage of their predicted counterparts. As can be seen, a reasonably accurate and precise agreement is suggested (Av. = 95%, S.D. = 5%). Allowing for some experimental error, this would imply the dissolution of ozone within the rig to have reached the predicted equilibrium concentration. Furthermore, with reference to Fig. 8, samples taken after 1 m of gas/solvent contact yield similar results to those taken after 3 m, thus indicating equilibrium to be achieved within the first meter of reactor tubing.
+
+### Determining suitable conditions for solvent-loading
+
+Based on Eq. (3) and with respect to ozone quantities within the inlet gas, Fig. 9 shows how the'mass% dissolved' (i.e. the absorption extent) and 'dissolved concentration% achieved' can be expected to vary with gas/solvent volume flow ratio. As can be seen, dissolved ozone concentration is predicted to increase with volume flow ratio in a semi-logarithmic fashion. However, coupled with this increase is a similar decline in absorption extent and hence, an increase in the fraction of ozone lost to the off-gas (i.e. wasted ozone). This is characteristic of co-current flow operation and an unavoidable limitation of such a system. (A counter-current system may therefore be needed.)
+
+With regard to operating the liquid-liquid/ozone rig during dye tests, suitable gas/solvent flow ratios were considered to lie in the 2-5 range. This allowed significant solvent-loading concentration to be achieved, whilst avoid
+
+Fig. 8: Comparison of predicted equilibrium and experimental ozone-in-solvent concentrations obtained under flow conditions A–C.
+
+ing excessive ozone/oxygen losses to the off-gas and thus, over-burdening of the gas cleaning system.
+
+## 6 Liquid-liquid/ozone rig dye tests
+
+The principle objectives of liquid-liquid/ozone dye tests were as noted above. All tests were conducted using the same 7 L of Volasil(tm)245, which was continuously recycled through the system.
+
+### Sampling and analysis
+
+Liquid-liquid samples taken from the rig were collected from the seven lower U-bend sampling points shown in Fig. 7 (i.e. sample points 3-9). Assuming inform plug flow, the residence time endured by each sample prior to collection was calculated based on reactor tube dimensions and flow conditions. Samples were also taken from the aqueous outlet of the liquid-liquid separator.
+
+#### 6.1.1 Decolouration
+
+As in preliminary tests, samples were collected in 20 mL glass tubes charged with \(\sim\)50 mg sodium sulphite for the purpose of quenching the reaction at the moment of collection. Following phase separation, 2 mL aqueous sub-samples were subjected to UV-vis spectrophotometer analysis using a 5 mm path length quartz cell (Hewlett-Packard 8451A Diode Array Spectrophotometer). Based on visible light absorbance at 540 and 494 nm (for DBR and RO16 dye solutions, respectively), the extent of decolouration was quantified by means of Eq. (2).
+
+#### 6.1.2 Chemical oxygen demand (COD)
+
+Samples taken for COD analysis were collected in 100 mL separating funnels. Unfortunately, due to the nature of the analytical technique, a reaction-quenching reagent (e.g. sodium sulphite) could not be used, as any such substance would present an additional COD (i.e. cause interference). Instead, samples were allowed 2-3 min separation time, after which the solvent phase (containing residual ozone) was removed. Samples were then left to stand overnight, in order to ensure separation of any solvent micro-droplets still present in the aqueous phase (and thus, avoid their contributing to the final COD result). Following this period, 2 mL aqueous sub-samples were taken and analysed in the 0-150 mg/L detection range. COD measurement was carried out in accordance with US EPA 8000 using a HACH reactor digestion equipment and colorimeter (model DR/890).
+
+#### 6.1.3 Total organic carbon (TOC)
+
+TOC measurement was carried out using an _Isco TOC Analyser_ (Isco Inc., USA). Unfortunately this device requires at least \(\sim\)300 mL of aqueous sample in order to reliably determine the TOC content. This was considered to be too great a volume to drain from individual sampling points without significant disruption to the system. Hence, samples for TOC analysis were restricted to collection from the aqueous outlet of the liquid-liquid separator (see Fig. 7). In order to ensure separation of any solvent micro-droplets, the separator contents were left to stand overnight before sampling.
+
+### Process stability
+
+In order to show that results acquired in subsequent tests were the product of pertinent operating conditions and not
+
+Figure 9: Predicted variation in ‘mass % dissolved’ and ‘dissolved concentration % achieved’ with gas/solvent volume flow ratio.
+
+unsteady-state behaviour (or sampling/analysis inconsistencies), the stability of liquid-liquid/ozone rig operation was demonstrated.
+
+Operating the rig under conditions given in Table 3, three sets of samples were drained from the seven lower U-bend sampling points at \(t\) = 5, 15 and 30 min after initial activation of the rig. The aqueous phase of each sample was then subjected to decolouration analysis using the spectrophotometer.
+
+Decolouration results are given in Fig. 10. Over the period of rig operation, similar degrees of decolouration were achieved after similar residence times throughout the solvent-water contact system, thus, suggesting stable operation. Furthermore, this indicates steady-state operation to have been achieved within the first 5 min of operation. Using a total solvent volume of 7 L, a solvent flow rate of 0.5 L/min and test duration of 30 min, complete recycling of the Volasil(tm)245 phase would have occurred approximately twice.
+
+Observations made during the test showed that Sulzer SMV mixing elements performed well. As desired, passage of the two phases through the elements produced and maintained a reasonably uniform dispersion of solvent droplets within water (\(\sim\)1 mm diameter droplets).
+
+### Liquid-liquid/ozone treatment of dye solution
+
+With respect to DBR and RO16 dye solutions, the effect of liquid-liquid/ozone water treatment was examined over a range of ozone:dye molar ratios (i.e. 2, 4, 6 and 10). Effects were measured in terms of changes in colour, COD and TOC. Table 4 gives the respective characteristics of raw 200 mg/L DBR and RO16 dye solutions (made using distilled water).
+
+\begin{table}
+\begin{tabular}{l c c c c c c} O\({}_{3}\);DBR mole ratio & Volume flows (L/min) & Concentrations (mg/L) \\ \cline{2-7}  & Gas & Solvent & Water & O\({}_{3}\)-in-gas & O\({}_{3}\)-in-solvent & Dye-in-water \\ \hline Process stability test & & & & & & \\
+4:1 & 2.0 & 0.5 & 0.4 & 65 & 80 & 200 \\ DBR dye tests & & & & & & \\
+2:1 & 1.2 & 0.4 & 0.6 & 60 & 72 & 200 \\
+4:1 & 1.2 & 0.6 & 0.375 & 60 & 60 & 200 \\
+6:1 & 1.2 & 0.6 & 0.25 & 60 & 60 & 200 \\
+10:1 & 1.2 & 0.6 & 0.3 & 60 & 60 & 100 \\ RO16 dye tests & & & & & & \\
+2:1 & 1 & 0.4 & 0.6 & 43 & 60 & 200 \\
+4:1 & 1 & 0.5 & 0.5 & 64 & 60 & 200 \\
+6:1 & 1.2 & 0.6 & 0.4 & 63 & 60 & 200 \\
+10:1 & 1.2 & 0.6 & 0.5 & 65 & 60 & 100 \\ \hline \end{tabular}
+\end{table}
+Table 3: Operating conditions applied during various tests
+
+Figure 10: Comparison of decolouration kinetics measured after 5, 15 and 30 min of continuous rig operation.
+
+#### 6.3.1 Treatment of DBR dye solution
+
+Table 3 gives the operating conditions employed during tests involving DBR dye solution.
+
+With respect to treated samples (collected from the liquid-liquid separator aqueous outlet; see Fig. 7), Fig. 11 compares the effects of increased ozone: dye molar ratios in terms of colour, COD and TOC. Decolouration appears to be the most significant effect of liquid-liquid/ozone treatment, with modest reductions in COD and TOC being achieved. Optimum treatment benefit was reached using a molar ratio of 4:1. Despite the use of higher ratios, little advantage could be discerned. This would suggest the dye to be broken down into reaction products of an ozone-refractory nature. Treated DBR solution was found to possess a pH of \(\sim\)3.5, perhaps suggesting the formation of organic acids.
+
+Assuming DBR decolouration to superficially behave as first order, reaction kinetics are shown in Fig. 12. In all cases, completion of the reaction was achieved within approximately 1.5 min. Optimum molar ratio conditions were reached at 4:1. Increasing the amount of available ozone beyond this relative amount was not shown to significantly increase the rate of reaction. This would suggest DBR decolouration to be reaction rate controlled at molar ratio conditions of 4:1 and above.
+
+The kinetics of COD reduction can be seen in Fig. 13. Elevating the amount of available ozone beyond the molar ratio of 4:1 was not shown to yield any significant increase in the rate of COD reduction.
+
+The potential for further COD reduction is implied by the persistence of a slight upward trend in the results. This would suggest some slower COD-reducing reactions to be still ongoing. However, the shape of the reaction curves obtained also suggests the bulk of the reaction to be complete and therefore, a prolonged residence time would yield little further reduction.
+
+#### 6.3.2 Treatment of RO16 dye solution
+
+Operating conditions employed during tests involving RO16 dye solutions are given in Table 3.
+
+Fig. 14 compares the effects of liquid-liquid/ozone treatment on RO16 solution at various ozone: dye molar ratios. With respect to colour, complete reduction was achieved by applying a molar ratio of 6:1. Impacts on COD and TOC were much less dramatic (\(\sim\)30% at 10:1). However, unlike the
+
+\begin{table}
+\begin{tabular}{l c c c c c} Dye & Colour & Absorbance (path length = 5 mm) & COD (mg/L) & TOC (mg/L) & pH \\  & & @510 mm & @494 mm & & \\ DBR & Red & 1.47 & – & 150 & 52 & 6 \\ RO16 & Orange & – & 2.01 & 155 & 43 & 6 \\ \end{tabular}
+\end{table}
+Table 4: Characteristics of raw 200 mg/L DBR and RO16 dye solutions
+
+Fig. 11: Comparison of liquid–liquid/ozone treatment effects with respect to DBR solution processed using various ozone:dye molar ratios.
+
+behaviour of DBR dye, increased ozone availability beyond 6:1 was shown to yield further COD and TOC reductions.
+
+Assuming RO16 decolouration to superficially behave as first order, Fig. 15 shows the reaction kinetics at the various molar ratios. Increasing the amount of available ozone was demonstrated to yield faster rates of decolouration.
+
+Reaction kinetics for the reduction of COD are shown in Fig. 16. Relatively rapid COD decline is suggested to occur within the first \(\sim\)30 s of contact, during which the bulk of the reduction is achieved. Similar to DBR behaviour, potential continuation of the reaction (beyond the residence time allowed by the rig) is indicated by the persistence of a slight upward trend in the results. However, due to the slowness of the reaction at this point, little additional reduction would be gained by allowing a significantly greater residence time.
+
+Similar to DBR, treated RO16 solutions were found to have experienced a decline in pH to \(\sim\)3.5. Again, this may indicate organic acids to be amongst the products of dye ozonation.
+
+Fig. 12: Comparison of pseudo first order decolouration kinetics of DBR dye solution treated at various ozone: dye molar ratios in the liquid–liquid/ozone rig.
+
+Fig. 13: Comparison of COD reduction kinetics of DBR solution treated at various ozone: dye molar ratios in the liquid–liquid/ozone rig.
+
+### Comparison with conventional ozone gas contact
+
+In order to demonstrate the performance of a liquid-liquid/ozone system in relation to a conventional ozone gas contact system, the rig was modified to operate under liquid-gas/ozone conditions.
+
+In respective liquid-gas/ozone tests, DBR and RO16 dye solutions were exposed to an ozone:dye molar ratio of 6:1. Ozone-in-oxygen (62 mg/L for DBR, 42 mg/L for RO16) and dye solution (200 mg/L) were vigorously mixed by passage through the former liquid-liquid contact section of the rig. In each test, gas and liquid flows were adjusted to 0.8 and 0.35 L/min.
+
+With respect to treated DBR and RO16 samples (collected from the former liquid-liquid separator outlet), the extent and kinetics of colour reduction, COD and TOC were almost iden
+
+Fig. 14: Comparison of liquid–liquid/ozone treatment effects with respect to RO16 solution processed using various ozone:dye molar ratios.
+
+Fig. 15: Comparison of psudo first order decolouration kinetics of RO16 dye solution treated at various ozone:dye molar ratios in the liquid–liquid/ozone rig.
+
+tical to those obtained during liquid-liquid/ozone operation (within \(\pm\)2%).
+
+### Predicted cost of liquid-liquid/ozone dye wastewater treatment
+
+Assuming ozone generation to demand \(\sim\)8 kW h/kg [14] and the cost of that energy to be \(\sim\)0.07 EUR/kW h, the energy cost of ozone generation can therefore, be estimated at \(\sim\)0.56 EUR/kg (by corona discharge). Based on this assumption and with respect to decolouration, Table 5 estimates the ozone generating energy costs for the decolouration of wastewaters contaminated with DBR and RO16.
+
+As discussed above, the co-current flow arrangement of the solvent-loading section ensures that only a fraction of the ozone generated is dissolved into the solvent (the balance breaking through to off-gas) (see Fig. 9). A counter-current arrangement to achieve 100% utilisation is strongly indicated.
+
+Volasil(tm)245 is available at the cost of \(\sim\)7.5 EUR/L. However, as solvent is recycled through the liquid-liquid/ozone system, replacement costs may be minimised by careful handling of the material (minimising entrainment loss in the separator).
+
+### Suggested process improvements
+
+During operation of the liquid-liquid/ozone rig, potential design alterations were noted that may potentially improve process operation. These are as follows.
+
+#### Counter-current solvent-loading
+
+In order to avoid inefficient use of ozone and for the achievement of higher solvent-loading concentrations, it is recommended that practical application of liquid-liquid/ozone technology should incorporate a counter-current flow solvent-loading section (rather than co-current). As discussed, at equilibrium Volasil(tm)245 will achieve twice the ozone concentration of the contacting gas phase. However, due to the present co-current design of the rig's solvent-loading section, dissolved ozone equilibrium is reached with the depleted outlet gas concentration (not the fresh inlet gas concentration). Hence, less than optimal loadings are achieved and a significant proportion of the total ozone generated is wasted. Despite these negative aspects, a co-current solvent-loading system was developed for this investigation, as it was considered to offer ease of design, construction and operation.
+
+#### Relocation of liquid-liquid separator
+
+As can be seen in Fig. 7, the present rig design locates the liquid-liquid separator approximately level with the base of liquid-liquid contact section. As such, solvent-water flow is allowed to cascade \(\sim\)1 m before plunging directly into the separator. Despite the achievement of careful mixing within the liquid-liquid contact section itself (via the use of static mixers), this final violent handling of the mixture enables the creation of solvent micro-droplets (visible to the naked
+
+Figure 16: Comparison of COD reduction kinetics of RO16 solution treated at various ozone:dye molar ratios in the liquid–liquid/ozone rig.
+
+\begin{table}
+\begin{tabular}{l l l l l} Dye & Dye concentration & O\({}_{2}\):dye molar requirement for decolouration & Cost (€/m\({}^{2}\)) \\ Dye & Dye concentration & O\({}_{2}\):dye molar requirement for decolouration & 100\% O\({}_{3}\) utilisation & 30\% O\({}_{3}\) utilisation \\ DBR & 200 & 4:1 & 0.05 & 0.18 \\ RO16 & 200 & 6:1 & 0.05 & 0.17 \\ \end{tabular}
+\end{table}
+Table 5: Estimated energy cost of ozone generation for the decolouration of waste waters contaminated with DBR and RO16eye), which were observed to exit the separator's aqueous outlet. No such droplets were observed to occur within the reactor tubes themselves. After deactivation of the pumps (and thus, the halting of tube flow), tube contents were seen to quickly separate into two discrete phases--each free of visible droplets.
+
+It is recommended that the separator be relocated to a position approximately level with the upper U-bends of the liquid-liquid contact section--thus, avoiding the 1 m drop and the resultant excessive agitation of the phases. Practical application of the technology should perhaps include the added precaution of a coalescer, fitted downstream of the separator.
+
+## 7 Conclusions
+
+Within this study, a continuous flow liquid-liquid/ozone water treatment system has being designed, constructed and operated at laboratory scale. Successful operation of the system was demonstrated with respect to various dye solutions using ozone-loaded Volasil(tm)245. Loading of the solvent was achieved to equilibrium with the gas phase at the outlet. Contact, dispersion and mixing of the solvent in dye-contaminated water was achieved within a tubular flow system using in-line Sulzer SMV static mixing elements. Following water phase contact, the solvent was disengaged and repeatedly recycled through the system.
+
+Dye treatment was investigated with respect to the DBR and RO16 solutions. The impact of treatment was measured in terms of colour, COD and TOC reductions. Almost complete decolouration of DBR and RO16 dye solutions was achieved by applying minimum ozone:dye molar ratios of 4:1 and 6:1, respectively (in the case of DBR this assumes Rosa's 408 g/mol molar mass assumption to be correct). In the case of RO16, increasing the availability of ozone beyond 6:1 was shown to yield continued improvements in reaction kinetics and final solution quality. Conversely, DBR reaction kinetics and treated solution quality were not significantly improved by ozone availabilities in excess of 4:1. This would suggest DBR breakdown to: (a) become reaction rate controlled (rather than mass transfer controlled) and, (b) form ozone-refractory products, thus, implying a practical limit of effective DBR treatment by ozone.
+
+Reductions in COD and TOC were less dramatic than reductions in colour. Even when relatively higher ozone availabilities were applied (i.e. 10:1 moles ozone:dye), both COD and TOC were found to decline by only \(\sim\)30%. This is typical of ozone treatment in general, and suggests that the process is only applicable to situations where wastewater decolouration is the priority.
+
+In tests comparing liquid-liquid/ozone treatment to that of conventional gas contact (i.e. gas-liquid/ozone), no significant differences could be discerned in terms of final solution quality or reaction kinetics. However, the possible benefits of liquid-liquid/ozone treatment were not fully explored. Neither dye was found to be soluble in Volasil(tm)245 and therefore, extraction to the ozone-enriched solvent was not realised. Furthermore, due to the co-current design of the rig's solvent-loading section, less than optimal solvent-loading concentrations were achieved and therefore, the effects of contact with a more heavily loaded solvent (i.e. twice the ozone concentration of the gas inlet) were not demonstrated.
+
+Dye reactions were shown to occur within time frames of \(\sim\)1.5-2 min--even under gas contact conditions. Such relatively brief reaction times present little practical need for improvement and hence, with respect to the dyes investigated, the added complexity and cost of liquid-liquid/ozone treatment is unlikely to be justified on the grounds of enhanced kinetics alone. Moreover, any decrease in reaction time would be offset by the time required to separate treated wastewater and solvent (2-3 min).
+
+Investigations continue in order to assess the advantages of liquid-liquid/ozone treatment with respect to more toxic wastewater pollutants, such as chlorinated organics (e.g. dichlorophenol, dichlorobenzene and dichloromethane). As the results on solvent insoluble dye degradation have clearly shown, the key to successful operation is likely to depend on the solubility of a pollutant in Volasil(tm)245.
+
+## Acknowledgements
+
+The authors would like to thank EPSRC (UK Government) for funding this project (contract no. GR/N 323890), VWR International (Merck, Poole, UK) for supplying the Volasil(tm)245 solvent and Sulzer Chemtech (UK) Ltd. (Farnborough, UK) for providing SMV static mixing elements.
+
+## References
+
+* (1) Anon, The Water Framework Directive: Guiding Principles on the Technical Requirements, Environment Agency, UK, Bristol, 2002, ISBN 1857058674.
+* (2) C. Gottschalk, J.A. Libra, A. Saupe, Ozonation of Water and Waste Water, Wiley/Vch, Weinheim, Germany, 2000, p. 30 and 69.
+* (3) B. Langlais, D.A. Recklow, D.R. Brink, Ozone in Water Treatment Application and Engineering, Lewis Publishers, USA, 1991, p. 113.
+* (4) D. Bhattacharya, C.E. Hamrin jr., R.P. Northey, Oxidation of hazardous organics in a two-phase fluorocarbon-water system, J. Haz. Waste Haz. Mater. 3 (4) (1986) 405-427.
+* (5) F.A. Stich, D. Bhattacharyya, Ozonolysis of organic compounds in a two-phase fluorocarbon-water system, Environ. Prog. 6 (4) (1987) 224-229.
+* (6) C.Y. Changand, J.N. Chen, Ozonolysis of 2,4-dichlorophenol in a two-phase solvent/water system, Water Sci. Technol. 29 (9) (1994) 343-346.
+* (7) D. Bhattacharya, T.F. Van Dierdonck, S.D. West, A.R. Freshour, Two-phase ozonation of chlorinated organics, J. Haz. Mater. 41 (1994) 73-93.
+* (8) A.K. Guha, P.V. Shanbhag, K.K. Sarkar, D.A. Vaccari, D.H. Trivedi, Multiphase ozonolysis of organics in wastewater by a novel membrane reactor, Am. Inst. Chem. Eng. J. 41 (1995) 1998-2012.
+* (9) A.R. Freshour, S. Mawhinney, D. Bhattacharyya, Two-phase ozonation of hazardous organics in single and multicomponent systems, Water Res. 30 (9) (1995) 1949-1958* [10] D.B. Ward, C. Tizaoui, M.J. Slater, Ozone-loaded solvents for use in water treatment, Ozone Sci. Eng. 25 (6) (2003) 485-495.
+* [11] M.A. Rozsa, Studies on Water Treatment using Ozone and adsorbents, PhD Thesis, University of Bradford, Bradford, UK, 2000.
+* [12] F. Archibald, L. Roy-Arcand, The use of ozone to decolour residual direct paper dyes in kraft paper machine whitewater, Ozone Sci. Eng. 19 (6) (1997) 549-565.
+* [13] C. Tizaoui, Investigation of a New Technique for Water Treatment Using Adsorbed Ozone, PhD Thesis, University of Bradford, Bradford, UK, 2001.
+* [14] Water Quality Association Ozone Task Force, Ozone for point-of-entry and small water systems, in: Water Treatment Application: A Reference Manual, Water Quality Association, Lisle, IL, USA, 1997, p. 12.

manual_annotation/200pdfs_in_mmd/10.1016_j.cep.2021.108337.mmd

@@ -0,0 +1,255 @@
+# Developing multiplexed plasma micro-reactor for ozone intensification and wastewater treatment
+
+Ainy Hafeez, Nasir Shezad, Fahed Javed, Tahir Fazal, Muhammad Saif ur Rehman, Fahad Rehman
+
+# Abstract
+
+The development of non-thermal plasma-based wastewater treatment technologies is a challenging topic of research these days. The conventional domestic ozonators based on corona, are susceptible to arcing, produce less ozone due to low input power, and yield larger bubbles limiting mass transfer rate. The article proposes a new design-multiplexed system containing six Corona-DBD hybrid packed-bed plasma micro-reactors connected in parallel to significantly improve the said design problems. The reactor gives enhanced ozone generation due to characteristics of corona (high concentration of radicals), DBD (homogeneous discharge), and packing material (stronger electric field). The diffuser configuration yields high number density of microbubbles enhancing mass transfer. The optimized multiplexed reactor improves ozone generation by 4.8 folds at 5.8 kHz, 20 kHz, 70 mA, 1 LPM, and 5 cm electrode length producing 480.8 ppm ozone in 20 s. The reactor efficiency was evaluated using Methyl orange degradation. The energy yield of the plasma reactor was also evaluated for MO solution. At optimized reactor conditions, 86 % Methyl orange was removed at pH 3, 10 ppm, and 900 mL in 8 min, and COD was reduced to 74 %. The multiplexed design offers an effective replacement for conventional ozone generators for about 1 L capacity in less than 10 min.
+
+## 1 Introduction
+
+Ozone is frequently used in household appliances such as water filters and pollutants degradation. These ozone generators are usually based on corona discharge [1,2] as shown in Fig. S1 in the supplementary information (SI). Operation at atmospheric pressure and low power requirement can be considered as inherent advantages of corona discharge [3,4]. However, counter-intuitively, the application of corona discharge is limited by the low input power requirement. Electrodes in this discharge are exposed to each other, therefore, high input power provides the current a direct path to flow which results in arcing [5,33]. The concentration of produced ozone is also reduced due to an unnecessary long tube connecting the corona reactor and the diffuser-used to dose/bubble the ozone. While passing through the tubes the ozone colloids with itself and with the walls of the tube and converts back to oxygen. Usually, a ceramic-based spherical sparger is used as a diffuser. The diffuser is microporous and requires a high-pressure drop for the gas to pass through it. The hydrophobic nature of the diffuser produces larger bubbles [6]. The flow pattern around the diffuser also induces coalescence making bubble size even bigger [7, 8, 9]. The mechanism of bubble formation around a spherical diffuser is shown in Fig. S1. The larger bubbles rise in a turbulent regime reducing the overall mass transfer from the ozone bubble to liquid/substrate.
+
+To overcome the limitations of low input power requirement, loss of ozone during transfer, large bubble, and coalescence, a DBD-Corona hybrid reactor integrated with sintered borosilicate diffuser has been proposed to simultaneously produce and dose ozone. The presence of a dielectric barrier not only prevents arcing but also allows to use of larger input power and produces relatively uniform discharge as compared to corona discharge [10,8]. This allows the reactor to work under high throughput conditions. The use of dielectric in plasmas also produces high-density energetic electrons, with an energy range of \(1-10\) eV [11]. The corona and DBD characteristics could be integrated to make the DBD-Corona hybrid discharge reactor to take advantage of both discharges simultaneously for various applications [11]. The insight of DBD, Corona, and Corona-DBD hybrid discharge is given in our previousstudies [11, 12].
+
+The hybrid reactor is AC driven and has pointed to plane electrode geometry with glass as a dielectric barrier between them [31, 35]. The pointed electrode (wire) provides streamers that produce a high concentration of radicals and glass dielectric makes electric discharge homogeneous as reported in our previous studies [11, 13]. In the current manuscript, the hybrid plasma micro-reactor is filled with inert glass beads to make a packed-bed reactor. The glass beads increase dielectric constant and create submicron channels assisting the plasma ignition and sustainability. The larger dielectric strength of the reactor allows high power deposition in the plasma which increases electric field strength. At the close contact points of beads and beads/electrodes, the electric fields become even stronger [14]. The number density of electrons reduces in packed-bed plasma reactors as it is inversely proportional to electric field strength. Therefore, packed-bed plasma reactors are characterized by low electron density and stronger electric field [15]. The packed-bed plasma reactors have greater energy efficiency for ozone generation as reported in other studies [16, 17].
+
+On contrary to the commercial ozonators, the plasma micro-reactor in the current study has optimized the minimal length required for the ozone to reach the wastewater to minimize the loss of ozone by colliding with itself and the tube walls. To assist the formation of smaller bubbles, a hydrophilic nature sintered borosilicate glass diffuser has been used in the current study. The thin and flat-surfaced sintered borosilicate diffuser produces microbubbles in the laminar regime at a lower gas flow rate [7]. The comparison of the flat sintered borosilicate diffuser and spherical ceramic diffuser could be seen in Fig. S2 in SI.
+
+To increase the throughput, six hybrid plasma reactors are connected in parallel to make a multiplexed plasma micro-reactor system. The efficiency of the multiplexed Corona-DBD hybrid plasma micro-reactor was evaluated for ozone generation using the Indigo method [11, 18]. The parameters affecting ozone generation - voltage, frequency, electrode length, air flow rate was investigated and optimized. The efficiency of the multiplexed hybrid reactors was also evaluated for its application in complex organic compound degradation-Methyl orange (MO). MO, a reactive azo dye, was used as a model compound [19, 20]. The hybrid reactor was operated at optimized conditions and parameters related to MO- pH, initial concentration, and volume of solution was studied.
+
+## Materials and methods
+
+### Multiplexed Corona-DBD hybrid plasma micro reactor
+
+The schematic diagram of the multiplexed Corona-DBD hybrid plasma micro-reactor is shown in Fig. 1. The hybrid reactor was designed to be operated at atmospheric pressure to treat different solutions. The multiplexed system consisted of six hybrid plasma micro-reactor connected in parallel to treat larger volumes of synthetic wastewater. Each micro reactor consisted of a borosilicate glass tube with an inner diameter of 7 mm and an outer diameter of 10 mm. The live electrode was a 1.3 mm stainless steel wire, passing through the center of the reactor and the ground electrode was 0.1 mm thick Aluminum tape with 1 cm length, wound around the glass tube. The upper part of each reactor upper part was a 300 mL glass funnel with a sintered borosilicate glass diffuser fixed at its bottom. The diameter of the grade-3 (15-40 microns) sintered borosilicate glass diffuser was 3 in. Glass beads were added in the plasma zone of each reactor as in our previous study [11]. Air was supplied using an air compressor (Malic MY5001). Indigo solution was added to the reactor after a steady flow of air to avoid leakage of the solution. A variable voltage (0-40 kV) and frequency (20-65 kHz) power supply (Amazing1, USA PVMS00/DI-DRIVE10) was used to provide AC voltage. When power was supplied to the reactor electrodes, electrical discharge occurred in between two electrodes extending to a length equal to the length of the ground electrode i.e., 1 cm. In a non-thermal DBD plasma reactor, the electron temperature (T\({}_{o}\)) is 3-5 eV while gas temperature (T\({}_{o}\)) varies from 0.05 to 0.1 eV, the electron density is 10\({}^{18}\)-10\({}^{21}\) m\({}^{-3}\)[8], the vibrational temperature is 4500 K and the rotational temperature is 700 K. The relationship between various particles' temperature in non-thermal
+
+Figure 1: The schematic diagram of multiplexed plasma micro-reactor.
+
+plasma can be conventionally represented by \(\text{Te}>\text{T}\text{v}>\text{T}\text{r}\approx\text{Ti}\approx\text{To}\)[21]. The voltage in the multiplexed hybrid micro-reactor was measured using a passive high voltage probe purchased from UNI-T with model P21. The current was measured using the CT2 current probe purchased from Tektronix [22,23]. A digital oscilloscope (UNI-T UTD2000 M) was used to display the voltage and current waveforms. The schematic experimental setup is shown in Fig. 2.
+
+The parameters affecting ozone formation- peak voltage (4-6 kV), peak current (20-83 mA), frequency (20-26 kHz), airflow rate (0.5-5 LPM), and electrode length (1-5 cm) were investigated in multiplexed hybrid plasma micro-reactor. The operational parameters were optimized for maximum ozone formation. The effect of dye parameters - pH (3-11), concentration (10-50 ppm), and the volume of dye solution (150-300 ml) was varied to investigate optimum conditions for the complete degradation of dye. The energy yield of MO degradation was also investigated. Further, the effect of oxonation on COD was studied for each parameter of dye.
+
+### Ozone quantification
+
+Ozone was quantified using Bader and Hoign's indigo method in a multiplexed hybrid plasma micro-reactor. The method follows fast kinetics and is sensitive to small changes in ozone concentration [8,11]. Potassium indigo trisulphonate, Phosphoric acid (85% v/v), and sodium dihydrogen phosphate were purchased from Sigma Aldrich. Indigo stock solution was prepared by dissolving 0.77 gm of potassium indigo trisulphonate in phosphoric acid (1 mL), diluting up to 1 L. The standard solution of Indigo was prepared by mixing 10 g sodium dihydrogen phosphate, 20 mL indigo stock solution, and 7 mL of phosphoric acid and diluting the mixture with distilled water to make 1 L solution. The pH of the solution was maintained around 2 using Sodium dihydrogen phosphate and phosphoric acid as a buffer. At pH 2, amines remain protonated and all the ozone produced reacts with \(\text{C}=\text{C}\)[24]. The calibration curve of indigo solution is given in Fig. S3 in section-S1 of SL.
+
+In each reactor, 150 mL of indigo solution was used with a total of 900 mL in each experimental run of the multiplexed reactor, and plasma was ignited for 20 s. Experiments were repeated thrice. The samples were analyzed using UV/Vis Spectrophotometer (Biobase Biodustry BK-UV1900PC) at \(\text{a}_{\text{max}}\) 600 nm. The moles of indigo decomposed were calculated during the reaction and moles of ozone formed were calculated since one mole of indigo decomposed is equal to one mole of ozone produced.
+
+### Preparation of dye solutions
+
+Methyl Orange with purity 98 % was purchased from Sigma Aldrich. All the solutions were prepared in ultra-pure deionized water with conductivity 1-1.5 \(\text{s}\text{p}\text{cm}^{-1}\). The MO stock solution of 1000 ppm was prepared and stored in 4 degC. The standard solutions of MO with various concentrations (10-50 ppm) were prepared by diluting the stock solution. The pH of the wastewater was measured using Hanna Instruments and the pH was adjusted using 0.1 M HCl and 0.1 M NaOH.
+
+## Analytical method
+
+The stock solution of MO was diluted with ultra-pure water to prepare a standard solution of concentration 200 ppm. The pH of the solution was initially maintained at pH 7. In each reactor of the multiplexed system, 150 mL of MO standard solution was added. Experiments were run for till complete MO degradation. The absorbance of the untreated and treated samples was measured using UV-vis Spectrophotometer at 464 nm. Each experiment was repeated three times for standard error estimation. The removal efficiency of MO was calculated using Eq. (1).
+
+\[\eta = \frac{\text{C}_{\text{o}} - \text{C}_{\text{i}}}{\text{C}_{\text{o}}}*100\]
+
+Fig. 2: Experimental setup for multiplexed Corona-DBD plasma micro-reactor ozone generation system.
+
+Where \(C_{\text{o}}\) and \(C_{\text{t}}\) is the MO concentration in ppm before and after ozone treatment respectively and \(\eta\) is the MO degradation efficiency (%).
+
+The degradation of MO in a multiplexed plasma micro-reactor has also been studied in terms of mineralization using chemical oxygen demand (COD). The experiments for COD determination were carried out using the Photometer System (Aqualytic, AL2000, Germany). For COD measurements, a zero cuvette was prepared by adding 2 mL of TOC-free water in the reagent cuvettes, gently swirled, and heated for 150 degC for 2 h in the heating chamber. The zero cuvettes were then allowed to cool to room temperature. The dye solutions (2 mL), before and after the plasma treatment was added in the reagent cuvettes with range 0-1500 mg/1 and placed in the heating chamber for 2 h at the same temperature. The sample cuvettes were then allowed to cool to room temperature. COD of the samples were measured by placing zero cuvettes in the COD meter and performing zero tests. After zero adjustment, sample cuvettes were placed in the COD meter one by one. The initial and final COD of MO solutions were determined before and light irradiation and after light irradiation (4 h), respectively [25]. The percentage of COD removal was calculated using Eq. (2).
+
+\[\eta=\frac{COD_{1}-COD_{2}}{COD_{1}}*100 \tag{2}\]
+
+Where CODi and CODi are the chemical oxygen demand of MO before and after ozone treatment, respectively.
+
+The plasma micro-reactors degradation efficiency has been calculated using energy yield (G), which is the amount of MO degraded per Kilo Watt-hour of energy to achieve a particular degradation efficiency/conversion. The energy yield is given by the following expression Eq. (3) [26].
+
+\[G\left(\frac{g}{KWh}\right)=\frac{C_{\text{t}}V_{\text{t}}\eta(\%)}{100\ P*t} \tag{3}\]
+
+Where Vt is the total volume of the treated solution, P is the power input, and t is the discharge time to achieve a particular degradation efficiency.
+
+## Results and discussion
+
+### Operational parameters of the multiplexed hybrid micro-reactor for Ozone quantification
+
+#### 4.1.1 Effect of voltage and frequency
+
+In the multiplexed plasma reactor, six reactors have been connected in parallel, and reactor system efficiency for ozone generation has been investigated. The effect of voltage and frequency on ozone formation is observed by varying peak voltage between 4.0 kV-6.1 kV and frequency between 20 kHz-26 kHz, as shown in Fig. 3 keeping air flow rate at 1 LPM with 1 cm length of plasma zone operating under atmospheric pressure. The lower and upper limits of voltage and frequency were set according to the breakdown voltage of air and arcing point in the reactor. The current could adjust itself according to multiplexed reactor impedance. The standard error in absorbance was 2.5 %.
+
+The effect of voltage on ozone formation could be observed by fixing frequency at a point. At 4-4.9 kV, the plasma started igniting at 26 kHz and was able to sustain at lower frequency 20 kHz with a little change in ozone concentration. At 5.2-5.8 kV, the plasma started ignited at a lower frequency-24 kHz and became intense at 20 kHz with maximum ozone formation at 5.8 kV at 20 kHz. At 6.1 kHz, the ozone concentration was again reduced. It can be seen in Fig. 3 that there is no regular trend of ozone formation with voltage variations. There is a rise and fall of ozone generation with an increase in voltage. However, there is an increase in ozone formation at higher voltage i.e., an increase from 4.0 kV to 5.8 kV, which increases ozone concentration from 50 to 350 ppm. The increase in voltage is related to high electrical energy density and hence more energetic electrons. The more energetic electrons give more frequent collisions with the air molecules and dissociate oxygen into excited O atoms as depicted in (4) and (5) [27]. The O atoms carry out third body recombination reaction with O2, O, or N2 as a third body in case of the air feed gas, given in reaction (6) [28; 29]. In air plasma, N2 molecules also get dissociated by electron impact, which generally starts from the electronic vibrations of molecules by electrons excitation (the first stage) and transfer of particles from vibrational levels to electronic decay (the second stage). The N2 dissociation rate is limited by the second stage. The electronic excited N produced during the reactions influence the discharge species accumulation. The O atoms produced in plasma discharge leading to ozone, increased three times due to O2 dissociation reactions involving excited states of N2 molecules. The dissociation reactions of N2 and O2 produce number of active species- NO, N2O, NO2, N2O5. However, the concentration and rate of O atoms and O3 is much higher than N species. The detailed mechanism of air plasma discharge reactions is given in [30]. When voltage is further increased to 3.4 kV which is the arcing point at 20 kHz, ozone concentration is reduced to 300 ppm. It is due to the increase in power density at higher voltage, which leads to side reactions such as the recombination of O atoms with each other, or with already produced ozone which results in the formation of O2 molecules, given in reactions (7) and (8) [31; 32].
+
+\[e + O_{2}\to e+O+O \tag{4}\] \[O + O_{2} + M\to O_{3}^{+}+M\to O_{3}+M\] (5) \[O + O\ + M\to O_{2}+M\] (6) \[O + O_{3}^{-}\to 2O_{2}\] (7) \[O + O_{3}\to 2O_{2} \tag{8}\]
+
+The effect of frequency variations (20-26 kHz) on ozone formation was observed by keeping voltage fixed. It could be seen in Fig. 3 that ozone concentration is decreased when frequency increases. Since the multiplexed hybrid plasma is AC driven, there is a polarity shift in the system. At lower frequency, charges find more retention time on the electrode which increases the current deposition [33; 34; 35]. This results in increased electric field strength in the micro-reactor due to which ozone concentration increases [29]. Therefore, ozone concentration is found to be maximum at 5.8 kV peak voltage and 20 kHz. Also, since the reactors are connected in parallel, the voltage in all reactors remains the same, however, the current is distributed. Due to unequal current distribution in the reactors, the maximum current could be generated through the reactor with the least resistance [36]. The reactor performance for ozone formation is shown at the optimum conditions of voltage and frequency keeping other factors constant. The ozone formation was calculated by taking average indigo decomposition in six reactors. The six reactors are denoted by R1 to R6.
+
+Fig. 4 shows the variations in current drawn by multiplexed hybrid plasma micro-reactor at different points of voltage and frequency. At a
+
+Figure 3: The effect of voltage and frequency on ozone formation at 1 LPM, 1 cm electrode length and, 1 atm.
+
+particular voltage, when the frequency is varied, the current drawn by the hybrid plasma microreactor varies in an irregular pattern. But it could be seen that the overall current has an inverse relationship with frequency. When the voltage is varied at a particular frequency, the current in the plasma reactor also varies. And it could be seen that when voltage is increased, current also increases at a frequency [37]. The variations in voltage and frequency affect ozone generated in the plasma micro-reactors.
+
+#### 4.1.2 Effect of airflow rate
+
+The influence of flow rate variations (0.5-1 LPM) was observed on ozone formation at 5.8 kV\({}_{\text{pk}}\) and 20 kHz frequency with a 1 cm electrode, given in Fig. S. The standard error in absorbance was 3%. It could be seen in the figure that when flowrate was increased to 1 LPM the graph for ozone concentration rises to a maximum (350 ppm) and then falls sharply after 1 LPM and reaches a steady-state up to 5 LPM. In hybrid plasma micro-reactors, various reactions take place, a few of which are given in Eqs. (4)-(8) due to which an optimum micro discharge strength ([O]/[O2]) = 10\({}^{-3}\) exists for ozone formation [38,39]. At a lower flow rate (1 LPM), the residence time of air is low and the reactions in the plasma discharge make the concentration of O2 higher enough to reach optimum micro discharge strength. It is because the greater number of O2 are dissociated in O atoms which follow third body recombination reaction to form higher O3 concentration. Whereas, at higher flow rates (\(>\) 1 LPM), O atoms concentration becomes greater which leads to side reactions are given in Eqs. (7) and (8). The side reactions quench O3 and produce O2 molecules and ozone concentration reduces to 295 ppm at 5 LPM. By variations in air flow rate the concentration of the active species produced in the plasma zone varies. Therefore, flow rate affects micro discharge strength and hence ozone formation. Furthermore, at a lower flowrate, the resistance of the air in the active plasma zone increase, and the greater number of molecules get collisions with energetic electrons to produce O3 through a series of reactions. Some other studies report the same trends [11,36].
+
+In the multiplexed reactor system, sintered borosilicate glass diffusers are used which make ozone and air to pass through them in the form of bubbles. Therefore, variations in flow rate affect the bubble formation pattern. At a lower flow rate, the bubbly regime is there, and bubbles formed are of small size. Smaller bubbles increase the interfacial area between ozone and indigo which increases mass transfer rate and hence higher ozone concentration [40]. With further increase in flow rate, the bubbly regime is converted to a heterogeneous regime, and bubbles of larger size are formed, which reduces interfacial area and hence mass transfer between two phases [41,42]. Thus, ozone concentration is reduced. Liu et al. [43] used a DBD plasma reactor for wastewater treatment with a small piece of sintered glass at the bottom to bubble O3 and air in the solution at 1LPM air flowrate. The reactor was able to produce 40 ppm of ozone which might be limited due to design and majorly the mass transfer limitation due to the small diffuser. In another study, Tichonavas et al. [44] investigated dyes degradation using DBD reactor made of glass tube with copper ground electrode immersed in water, and live electrode was passing through centre of the walls. The V\({}_{\text{pk}}\) was varied between 30-50 kV. The flow rate of air was maintained at 14.5 LPM and a ceramic diffuser was attached at the bottom of the DBD reactor. The concentration of ozone produced in the reactor was around 1-2 ppm. The lower efficiency of the reactor could be due to the usage of a ceramic diffuser which produces larger bubbles and requires a higher flow rate. The larger bubbles in the solution give less gas-liquid mass transfer and a higher flow rate leads to less ozone formation. The multiplexed plasma micro-reactor design in the present study produces 400 ppm of ozone in 20 s at 1 LPM. The higher efficiency of the reactor is attributed to the smaller bubbles produced due to sintered borosilicate glass diffuser, lower air flow rate, a reduced distance between plasma discharge and wastewater as discussed in the introduction section. In the multiplexed plasma, the reaction conditions were fixed, and ozone formation and MO decomposition were taken an average of the six reactors attached in parallel.
+
+#### 4.1.3 Effect of electrode length
+
+The effect of the plasma zone area on the plasma zone was investigated by increasing the ground electrode length of each hybrid reactor from 1 cm to 6 cm as shown in Fig. 6. The multiplexed hybrid reactor was operated at 5.8 kV\({}_{\text{pk}}\), 20 kHz, with 76 mA I\({}_{\text{pk}}\), 1 LPM airflow rate at 1 atm. The standard error in absorbance was 5%. It could be seen in the graph that with the increase in the length of the plasma zone, ozone concentration was gradually increased to 480.8 ppm when the length of the plasma zone was increased to 5 cm. The increase in ozone formation with electrode length is due to the increase in the active plasma zone and the discharge power [45]. The longer the plasma zone, the greater is time air gets to stay in the plasma zone and the higher is the molecule's interaction rate with each other. This causes more reactions in the plasma zone leading to the formation of higher ozone concentration in the presence of increased discharge power [46,47]. In the multiplexed
+
+Figure 4: Variations in current with frequency and voltage at 1 LPM, 1 cm electrode length and, 1 atm.
+
+Figure 5: The effect of air flow rate on ozone formation at 5.8 kVpk, 20 kHz 1 cm electrode length and, 1 atm.
+
+Figure 6: The effect of electrode length on ozone formation at 5.8 kVpk, 20 kHz, 1 LPM air flow rate and, 1 atm.
+
+plasma microreactor, the ozone formation is calculate by taking average of indigo decomposition in six reactors placed in parallel. The concentration of ozone produced in the multiplexed plasma micro-reactors is much higher than the conventional ozone generators. For instance, one of the ozone generators made by ozone tech produces 240 ppm of ozone per minute [48] and another one made by primocre produces 5 ppm of ozone in one minute [49].
+
+### Operational parameters for dye removal
+
+To see the effect of parameters- pH, concentration, and volume of methyl orange solution, the solution was treated in a single reactor at the optimized conditions (5 kV peak voltage, 36 mA peak current, 26.02 kHz frequency, 1 LPM air flow rate, and 4 cm electrode length in the presence of glass beads) in Fig. 9 of our previous study [12] and the parameters were optimized. The single plasma reactor produced 76 ppm of ozone empty channel plasma micro-reactor and 100 ppm in the packed-bed plasma micro-reactor. So, there was a 24 % increase in ozone formation in packed-bed plasma micro-reactor as reported in our previous study. Initially, 150 mL of 20 ppm MO at pH 7 was treated until the complete degradation of MO. Parameters were investigated one by one. The effect of ozone on and variations in MO parameters was also investigated on COD removal. After getting the optimum results of each parameter, pH, concentration, and volume of MO solution, the experiments were run at the optimized conditions of the multiplexed hybrid micro-reactor to scale up the wastewater treatment application.
+
+#### 4.2.1 Effect of pH on MO degradation
+
+The influence of pH was investigated on MO degradation by varying it from 3 to 11, as shown in Fig. 7. The reactor was operated at the optimized conditions of a single reactor. Experiments were run with 150 mL of 20 ppm solution till complete degradation of MO. It could be seen in Fig. 7 that when pH is increased from 3 to 7, MO removal decreases. However, with a further increase in pH to 11, degradation efficiency of MO increases but it is still lower than degradation at pH 3. Therefore, the maximum degradation of MO takes place at pH 3 and 97 % MO is removed in 8 min. MO exhibits different structures at different pH of the solution. In an acidic medium, MO has a quinoid structure which is easy to be degraded by oxidation, whereas, in the alkaline medium the structure is benzenoid which is difficult to be degraded. Further, Ozone solubility is higher at lower pH. Therefore, the degradation of MO is higher at pH 3. Similar results are shown in other studies [50].
+
+In air plasma, various reactive species of oxygen and nitrogen (RONs) are produced such as O-, OH-, N2O, NOx, etc. The discharge in the non-thermal plasma produced in this study is below the sintered borosilicate glass diffuser. Most of the active species formed in plasma are short-lived, therefore, only O3 being stable is the main species that gets diffused in wastewater in the form of microbubbles, along with air [51]. The NOx produced in the reactor also transfers to the wastewater depending upon the position of the plasma discharge [52,53]. The air plasma produced in the present study is ex-situ. Therefore, a very low concentration of NOx is transferred into the solution. Liu et al. [43] made a comparison between nitrogen species produced in-situ and ex-situ air plasma was performed. The study showed that in ex-situ or distant plasma, the concentration of nitrogen species (NOx) was negligible (below 5 ppm), whereas the formation of NOx such as nitrite (NO2) and nitrate (NO3) in-situ plasma was significantly higher [43]. It is because, for in-situ air plasma discharge, more energy or power is required to produce the same ozone concentration as in ex-situ. And higher energy leads to the NOx formation. The (NO2) and (NO3) react with water to produce HNO2 and HNO3 respectively as shown in Eqs. (9) and (10). HNO2 being highly unstable converts into NO2 or NO depending upon the solution pH [54]. In the solution, NO2 also reacts with O3 (Eq. (11)) to produce NO3 ions and O2. Moreover, the amines present in MO also produce nitrogen-containing functional groups by the attack of O3 and OH-, which get completely oxidized and form nitrates [44].
+
+\[2NO_{2} + H_{2}O \rightarrow HNO_{2} + H^{+} + NO_{3}^{-}\]
+
+\[3HNO_{2} \rightarrow H_{2}O + 2NO + H^{+} + NO_{3}^{-}\]
+
+\[O_{1} + NO_{2}^{-} \rightarrow O_{2} + NO_{3}^{-}\]
+
+The COD experiments depict the oxygen required for mineralization or oxidation of complex organic compounds into CO2 and H2O molecules [55]. Therefore, the mineralization of MO was studied using COD analysis to confirm the degradation of MO into smaller and less toxic compounds [56]. COD was also investigated at the optimized condition of pH. The COD of untreated MO at pH 3, 78 % reduction in 8 min of plasma treatment.
+
+#### 4.2.2 Degradation pathway of MO
+
+MO is characterized by the azo group (N=N), which is sensitive to ozone because of its high affinity towards unsaturated bonds. The reactivity of ozone with MO depends upon the pH of the solution. The degradation pathway of MO with O3 is given in the Eqs. (12-18) [57]. Moreover, during the MO degradation process, the pH of the solution further reduces which shows MO conversion into acidic products with the reaction time [58]. When the pH of the solution is varied, the active species in the solution changes since ozone has two reaction pathways in solutions [59,60]. In acidic to neutral medium, ozone decomposes with MO directly [44]. Whereas, in an alkaline medium, some of the ozone reacts with water to form OH- radicals and both react with MO [61]. During the ozonation of MO, a negligible amount of H2O2 is formed, therefore, it is not detected [58,62]. Therefore, O3 and OH- are two main species playing important role in MO degradation. Moreover, since the plasma micro-reactor used in the study is non-thermal plasma, the temperature rise does not usually exceed room temperature [30,63]. The ozone decomposes at a higher temperature but since the temperature is not so high, therefore, a significant change in ozone decomposition does not usually happen [8,64]. The decomposition of MO begins with ozone attack on N=N bond through electrophilic substitution and it generates less complex intermediate products. In the initiation reaction, ozone decomposes in water and produces OH- ions which act as an initiator to produce Hydroporey radical (HO2) and HO2- being unstable radical converts into O2- ion. In the chain reaction, O2- reacts with O3 and produce O3- ions which are very reactive and combine with H+ to produce HO3 Radicals. OH- radicals also react with O3 and produce some reactive species like HO4. HO2- and some other radicals act as promotors and maintain the chain reaction. In termination reactions OH- reacts with MO and converts it into smaller and simpler molecules which may include CO2. H2O, CO2.3, NO23, SO24, etc. [65] as indicated by COD of the solution after treatment.
+
+**Initiation reaction:**
+
+\[OH^{-} + O_{1} \rightarrow O_{2}^{-} + HO_{2}\]
+
+Fig. 7: The effect of pH on 150 mL of 20 ppm MO degradation at 5 kVpk, 26 kHz, 1 LPM air flow rate, 4 cm, and 1 atm in single-channel plasma micro-reactor.
+
+\[HO_{2} \to H^{+} + O_{2}^{-}\]
+
+**Radical chain-reaction:**
+
+\[O_{2}^{-} + O_{3} \to O_{3}^{-} + O_{2}\]
+
+\[O_{3}^{-} + H^{+} \to HO_{3}\]
+
+\[OH + O_{3} \to O_{2}^{-} + HO_{4}\]
+
+\[HO_{4} \to O_{1} + HO_{2}\]
+
+**Termination reaction:**
+
+\[OH + MO \to degradation\]
+
+#### 4.2.3 Effect of concentration on MO degradation
+
+The influence of initial MO concentration (10-50 ppm) was investigated on degradation efficiency at pH 3, and 150 mL solution in a single reactor operated under optimized conditions, shown in Fig. 8. The Figure shows that the removal of MO decreases from 97.5 % to 89 % with the increase in initial dye concentration in 8 min of treatment. It is because the experiments were carried out at fixed conditions of ozone generation and MO, therefore, the amount of ozone and OH- radicals formed in the solution was the same. Increasing MO concentration increases dye molecules to be degraded by a fixed number of O\({}_{3}\) gas molecules and OH- radicals. Due to which the decoloration efficiency of MO decreases with an increase in initial dye concentration [66,50]. The G of MO was also investigated as a function of degradation efficiency by varying concentration of the solution (see Fig. 9). It could be seen in the graph that the MO concentration with higher degradation efficiency, the G is low. And when the concentration is high, G is higher due to the presence of a greater number of MO molecules affected by the plasma discharge [26]. Therefore, G is directly dependent on the concentration of the solution. When the MO solution is treated with plasma for a longer time, most of the molecules in the solution are degraded, due to which fewer molecules are available in the water for decomposition. This reduces degradation of MO in the solution per KWh of energy, and hence G [67]. Similar results have been reported by several researchers [26,62, 68]. The change in COD was also studied at the optimized conditions of initial MO concentration at pH 3. The COD of the solution was reduced to 83 % after 8 min.
+
+#### 4.2.4 Effect of solution volume on MO degradation
+
+The effect of the volume of MO solution was investigated in the range 150-300 mL with pH 3 and 10 ppm concentration as shown in Fig. 10. The single reactor was operated under optimized conditions. The standard error in absorbance was 4 %. The Figure shows that with the increase in volume from 150 to 300 mL, Mo degradation efficiency is reduced. The decrease in degradation efficiency by increasing volume could be attributed to the fact that at given reactor conditions and dye parameters, a fixed number of ozone and OH radicals are formed. By increasing the volume of MO solution, a greater number of MO molecules are to be degraded by ozone and OH radicals which reduce the decoloration efficiency of MO.
+
+The G of MO was also studied with respect to degradation efficiency by varying volume of the solution as shown in Fig. 11. The graph shows that when G is dependent upon the volume of the MO solution. When the volume of the solution is less, there is low G and vice versa. It is because, at fixed reaction conditions, the number of MO molecules is greater in larger volumes. Therefore, G is higher for larger volumes [26]. Moreover, G is indirectly related to the degradation efficiency of MO because, at higher degradation efficiency, the amount of MO molecules removal per kWh of energy is lower [62]. The effect of COD was also studied at the optimized conditions of the volume of the MO solution at pH 3 and 10 ppm MO concentration. The COD of the MO solution was reduced to
+
+Figure 8: The effect of concentration on 150 mL of MO degradation at pH 3 and 5 KWh, 26 kHz, 1LPM air flow rate, 4 cm, and 1 atm in single-channel plasma micro-reactor.
+
+Figure 10: The effect of solution volume on 10 ppm MO degradation at pH 3 and 5.0 kVpk, 26 kHz, 1 LPM air flow rate, 4 cm, and 1 atm in single-channel plasma micro-reactor.
+
+Figure 9: Energy yield Vs Degradation efficiency with the change of initial concentration of MO solution at 180 W input power, 150 mL solution.
+
+83 % in 8 min of plasma treatment.
+
+### Degradation of MO in multiplexed plasma micro-reactor
+
+The optimization of MO parameters-pH, initial concentration, and volume of solution was carried out in a single plasma micro-reactor operated at its optimized conditions. The dye was then degraded in a multiplexed plasma micro-reactor and the reactor was operated at the optimized conditions. 5.8 kVp\({}_{\text{bd}}\) 20 kHz, 1 LPM, and 5 cm electrode length. Similarly, MO parameters were also maintained at the optimized conditions-pH 3 and 150 ml of 10 ppm solution. Therefore, a total of 900 mL of solution was treated. The degradation of MO in multiplexed plasma micro-reactor with time could be seen in Fig. 12. 48 % of MO was degraded in 2 min and around 86 % of degradation took place in 8 min of reaction time. However, the efficiency of all the reactors was different due to uneven plasma formation owing to the parallel connections of the reactors. The COD of the solution could be observed to reduce by 74 % in the multiplexed system. The multiplexed plasma micro-rector gives the advantage of producing more ozone as compared to the single reactor. The multiplexed system produces 480 ppm of ozone per 20 s while single reactor has a limited capacity of ozone formation-100 ppm in 20 s. Thus, there is a 4.8-fold increase in ozone formation in the multiplexed system. Moreover, the increase in reactor throughput is possible by the single power source. The multiplexed system enables to treat of larger volumes at once while maximizing the contact area between ozone and wastewater.
+
+### Conclusion
+
+The multiplexed reactors containing six packed-bed Corona-DBD hybrid plasma micro-reactors connected in parallel have been employed to scale up ozone generation and water treatment applications. The shortcomings of conventional ozone generators have been significantly improved in the current design. The Corona-DBD hybrid design of the plasma micro-reactor takes advantage of the concentration of the high radical of the corona discharge and homogenous discharge of DBD. Moreover, packing the plasma micro-reactor with glass beads increases the dielectric constant and electric field strength which enhances ozone generation. The point of ozone generation is kept close to water to be treated so that ozone may not destroy before reaching the targeted solution due to collisions with itself and walls of the tubes. Furthermore, the configuration of the diffuser used in the study yields a higher number density of microbubbles, which increases the mass transfer rate between air and targeted solution. The multiplexed reactor was optimized for maximum ozone generation, which produced 480 ppm of ozone at 5.8 Kp\({}_{\text{bd}}\), 76 mA I\({}_{\text{bd}}\), 20 kHz, 1 LPM, and 5 cm electrode length. The ozone generation in the multiplexed system is much higher than the conventional ozone generator which produces 240 ppm/min ozone. The efficiency of MO degradation was then investigated by varying dye parameters. MO was removed to 86 % at pH 3, 10 ppm, and 900 mL solution in 8 min in multiplexed reactors. At these conditions, COD was reduced to 74 % which shows MO degradation.
+
+### Author's contributions
+
+Airy Hafeez: Conceptualization, Methodology, Experimentation, Formal data analysis, Writing - original draft
+
+Nasir Shezad: Data Analysis and Writing
+
+Fahed Javed: Visualization, Data curation
+
+ Tahir Fazal: Experimentation
+
+Muhammad Saif ur Rehman: Supervision, Providing Resources
+
+Fahed Rehman: Supervision and ensuring that the descriptions are accurate and agreed by all authors.
+
+### Funding statement
+
+The research work was supported by Indigenous scholarship phase II Batch V, Higher Education Commission Pakistan, and NRPU (2017)-7924, Higher Education Commission Pakistan.
+
+### Declaration of Competing Interest
+
+The authors report no declarations of interest.
+
+### Acknowledgments
+
+We thank Dr. Fahad Rehman for his supervision in the completion of the current research article. We are grateful to Dr. Muhammad Saif ur Rehman for his technical guidance and support. We thank Mr. Nasir Shezad, Mr. Fahed Javed, and Mr. Tahir Fazal for helping in experimentation, data analysis, and proofreading of the article.
+
+## Appendix A Supplementary data
+
+Supplementary material related to this article can be found, in the online version, at doi:[https://doi.org/10.1016/j.cep.2021.108337](https://doi.org/10.1016/j.cep.2021.108337).
+
+## References
+
+* (1) H. Bao, L. Bao, H. Haridas, D. Masju, S. Swerda, H. Chamskaya, Design and Characterization of Cold Plasma Ozonator for Wastewater Treatment, Recent Trends in Waste Water Treatment and Water Resource Management, Springer, 2009, pp. 159-168.
+* (2) N. Sain, M.R. Ghezzar, C. Guyon, F. Abdelmalek, M. Tatoulian, A. Addou, New prototype for the treatment of falling film liquid effluents by gliding arc discharge part II: planaracyclic activity of TiO2 thin film deposited by magnetron sputtering, Chem. Eng. Process. Imengt. 2015 (2015) 32-40.
+* (3) K. Mao, Y. Hayashi, M. Yamada, S. Takahashi, N. Takada, H. Kanda, M. Goto, Atmospheric pressure peaked discharge plasma in capillary slug flow system for dye decomposition, Chem. Eng. Proc.-Process. Intent. 135 (2010) 133-140.
+* (4) M. Chezaz, N. Sain, S. Bachrach, F. Abdelmalek, A. Addou, New prototype for the treatment of filling film liquid effluents by gliding arc discharge part I: application to the dissociation and degradation of antonaguiomatic Acid Green 25, Chem. Eng. Process. Imengt. 2013 (2013) 42-50.
+* (5) F. Rehman, W.S. Abdul Majeed, W.B. Zimmerman, Hydrogen production from water vapor plasmalysis using DBD Corona hybrid reactor, Energy fuels 273 (2013) 2748-2761.
+* (6) W.B. Zimmerman, M.K. Al-Mahbadani, H.H. Bandulsasem, Evaporation dynamics of microtubules, Chem. Eng. Sci. 101 (2013) 865-877.
+* (7) F. Rehman, G.J. Medley, H. Bandulsasem, W.B. Zimmerman, Flatidic oscillator-mediated microbubble generation to provide cost effective mass transfer and mixing efficiency to the wastewater treatment plants, Electron. Res. 137 (2015) 32-39.
+* (8) F. Rehman, Y. Liu, W.B. Zimmerman, The role of chemical kinetics in using O3 generation as proxy for hydrogen production from water vapour plasmolysis, Int. J. Hydrogen Energy 41 (2016) 6180-6192.
+* (9) W.B. Zimmerman, M. Zandi, H.H. Bandulsasem, V. Tesar, D.J. Gilmour, K. Ying, Design of an air-lit loop bioreactor and pilot scales studies with fluidic oscillator induced microbubbles for growth of a microbubble Daniella salima, Appl. Energy 88 (2011) 3357-3369.
+* (10) B. Jiang, J. Zheng, S. Qin, M. Wu, Q. Zhang, Z. Yan, Q. Xue, Review on electrical discharge plasma technology for wastewater remediation, Chem. Eng. 236 (2014) 348-368.
+
+Figure 12: Removal of 900 mL of MO in multiplexed plasma micro-reactor at 5.8 kVp, 20 kHz, 1 LPM air flow rate, 5 cm electrode length, and 1 atm.
+
+* [11] A. Hafeez, S.A.A. Taqvi, T. Fazal, F. Aved, Z. Khan, U.S. Amjad, A. Bokhari, N. Shehab, M. Rashidi, S. Behnam, Optimization on cleaner intensification of ozone production using Artificial Neural Network and Response Surface Methodology: parameter and comparative study, J. Clean. Prod. (2019), 119833.
+* [12] H. Alny, Javed F. Rahul, Faisal Tahir, Sherad Nasir, Amjad Um-Saima, Muhammad Safir for Behnam, Behnam Shah, Internationalization of ozone generation and degradation of azo dye in non-thermal hybrid corona-DBD plasma micro-reactor, Chem. Eng. Process. 159 (2021) 108205.
+* [13] Y. Liu, P. Rehman, W.B. Zimmermann, Reaction engineering of carbon monoxide generation by treatment with atmospheric pressure, low power CO22 DBD plasma, Fuel 209 (2017) 117-126.
+* [14] Z. Cui, X. Zhang, Y. Tian, Y. Li, J. Tang, Effects of glass beads packing on SF to abatement by packed blend plasma, Plasma Chem. Plasma Process. 1995 (2019) 1-17.
+* [15] Z. Zhou, Z. Liang, Y. Liu, Y. Ma, X. Zhu, X. Zhang, Q. Li, Z. Ren, Identification of degradation of Sunset Yellow using scaled bed in a pulsed high-voltage hybrid gas-liquid discharge system: optimization of operating parameters, degradation mechanism and pathways, Chem. Eng. Process. Process. Intensti. 115 (2017) 23-33.
+* [16] M. Southakaran, J.O. Jo, Q.H. Trinh, V.S. Mok, Characteristics of packed-bed plasma reactor with dielectric barrier discharge for treating, Appl. Chem. Eng. 26 (2015) 495-504.
+* [17] H.-X. Ding, A.-M. Zhu, X.-F. Yang, C.-H. Li, Y. Xu, Removal of formaldehyde from gas streams via packed-bed dielectric barrier discharge plasmas, J. Phys. D Appl. Phys. 38 (2005) 4160.
+* [18] H. Bader, Determination of Ozone in Water by the Indigo Method: A Submitted Standard Method, 1982.
+* [19] A. Perez, T. Pomjarski, L. Chairez, Effect of additives on ozone-based decomposition of Reactive Black 5 and Direct Red 28 dyes, Water Environ. Res. 85 (2013) 291-300.
+* [20] Q. Zheng, Y. Dai, X. Han, Decoloration of azo dye G Reactive Black 5 by protonation in aqueous solution: influence factors, degradation products, reaction pathway and toxicity assessment, Water Sci. Technol. 73 (2016) 1050-1510.
+* [21] A. Fridman, A. Chierkov, A. Guisch, Non-thermal atmospheric pressure discharges, J. Phys. D Appl. Phys. D Appl. Phys. 38 (2005) R1.
+* [22] Z. Fan, H. Qi, Y. Liu, H. Yan, C. Ren, Effects of airflow on the distribution of filaments in atmospheric AC dielectric barrier discharge, Phys. Plasmas 23 (2016), 123520.
+* [23] K. Gazidi, L. Doanh, B. Held, P. Clement, Electrical, thermal and optical parametric study of optical ionization waves produced with a compact sp-s pulsed DBD-based reactor, Plasma 1 (2018) 23-44.
+* [24] D. Kushinizu, A. Skawamoto, J. Lozano-Pardo, Y. Shimomae, Efficient compact micro BSD plasma reactor for ozone generation for industrial application in liquid and gas phase systems, World Acad. Sci. Eng. Technol. 8 (2014) 80-83.
+* [25] T. Fazal, A. Razza, F. Aved, A. Hafeez, N. Rashidi, U.S. Amjad, M.S.U. Behnam, A. Faisal, F. Behnam, Integrating adsorption and photocatalysis: a soft effective strategy for textile wastewater treatment using hybrid biochar-TiO2 composite, J. Hazard. Mater. 390 (2020), 21623.
+* [26] B. Jiang, J. Zheng, Q. Liu, M. Wu, Degradation of azo dye using non-thermal plasma advanced oxidation process in a circulatory airtight reactor system, Chem. Eng. J. 204 (2012) 32-39.
+* [27] C. Ding, D. Yuan, Z. Wang, Y. He, S. Kumar, Y. Zhu, X. Ren, Ozone production influenced by increasing gas pressure in multichannel dielectric barrier discharge for positive and negative pulse modes, Ozone Sci. Eng. 40 (2018) 228-236.
+* [28] L. Kim, M. Jonnis, L. Demaire, G. Vidali, Formation of ozone by solid state reactions, J. Chem. Soc. Faraday Trans. 20 (2013) 19750-19755.
+* [29] B. Eliason, U. Kogelstratz, Modeling and applications of silent discharge plasmas, IEEE Trans. Plasma Sci. 19 (1991) 309-323.
+* [30] J. Kovalyi, A. Kovalnyi, A. Matveev, V. Siskovic, Kinetic scheme of the non-equilibrium discharge in nitrogen-oxygen mixtures, Plasma Sources Sci. Technol. 1 (1992) 207.
+* [31] A. Kiamandana, E. Yulinanto, A. Prakoso, A. Faruq, A. Qusundin, M. Hendra, E. Saminta, M. Restrijournals, P. Fratiun, F. Araintu, The comparison of ozone production with dielectric barrier discharge plasma reactors series and parallel at atmospheric pressure, J. Phys. Conf. Ser. (2019) 02100. IOP Publishing.
+* [32] X. Zhang, J. Lee, H.G. Im, M.S. Chu, Ozone production with dielectric barrier discharge: effects of power source and humidity, IEEE Trans. Plasma Sci. 44 (2016) 2288-2296.
+* [33] T.-L. Sung, S. Teti, C.-M. Liu, R.-C. Hsiao, P.-C. Chen, Y.-H. Wu, C.-K. Yang, K. Teti, S. Ono, K. Elsham, Effect of pulse power characteristics and gas flow rate on ozone production in a cylindrical dielectric barrier discharge ozonnier, Vacuum 90 (2013) 65-69.
+* [34] A. Ogata, K. Miyane, K. Mizuno, S. Kubishima, M. Terada, Decomposition of benzene in air in a plasma reactor effect of reactor type and operating conditions, Plasma Chem. Plasma Process. 22 (2002) 537-552.
+* [35] V. Neira, A. Kumar, H. Dwivedi, Atmospheric non-thermal plasma sources, Int. J. Eng. 2 (2005) 583-68.
+* [36] M. Nur, A. Sisan, Z. Muthiliu, F. Avitanto, A.W. Kiamandana, I. Nurhasah, S. Sumarishi, P.J. Wilawa, G. Gunawan, A. Usman, Evaluation of novel integrated dielectric barrier discharge plasma as ozone generator, Bull. Chem. Res. Lett. 21 (2017) 24-31.
+* [37] O. League, L. Palla, M. Taroulian, D. Mantovani, S. Ognier, Treatment of 4-chloromagnetic acid by plasma-based advanced oxidation processes, Chem. Eng. Process. Process. Intensti. 72 (2013) 82-89.
+* [38] Y. Sun, X. Liu, K. Dong, C. Xie, Corona discharge effect on charge and energy transfer in dielectric barrier discharge, Phys. Scr. 94 (2019), 085602.
+* [39] S. Jodzis, T. Racczynski, Ozone synthesis and decomposition in oxygen-fed pulsed system: effect of ozone concentration, power density, and residence time, Ozone Sci. Eng. 41 (2015) 69-79.
+* [40] A. Kenova, Decolorization of wastewater containing direct dye by oxonation in a batch buildin column reactor, Destination 158 (2003) 233-240.
+* [41] C. L. Savary, E. Schlieber, M. Lischau, U. Hamid, M. Schubert, Hydrodynamics, mixing and mass transfer in a pilot-scale bubble column with dense intervals, Chem. Eng. Sci. 02 (2019) 64019.
+* [42] S. Patel, R. Agarwal, S.K. Majumder, P. Das, P. Ghosh, Kinetics of ozonation and mass transfer of Pharmaceutical-degraded by ozone fine bubbles in a plant prototype, Heat Mass Transf. 2019) 1-13.
+* [43] Y. Liu, S. Mei, D. Yu-Sou, S. Cavadias, S. Ogiler, Carbamazepine removal from waterley dielectric barrier discharge: comparison of ex situ and in situ discharge on water, Chem. Eng. Process. Intensti. 56 (2012) 10-18.
+* [44] M. Tichonova, E. Krugly, V. Raeys, R. Rippler, V. Kauzlien, I. Sisalistititititi, Deformation of various textile dyes as wastewater pollutants under dielectric barrier discharge: plasma treatment, Chem. Eng. J. 229 (2013) 9-19.
+* [45] M. Bahri, F. Haghighat, S. Rohani, H. Kazemian, Impact of design parameters on the performance of non-thermal plasma air purification system, Chem. Eng. J. 302 (2016) 204-212.
+* [46] S. Jodzis, W. Padlowski, Kinetic and energetic analysis of the ozone synthesis process in oxygen-fed DBD reactor: Effect of power density, gap volume and residence time, Ozone Sci. Eng. 38 (2016) 86-99.
+* [47] Y.H. Lee, S.G. Pavlostatsis, Decoloration and toxicity of reactive anti-azutioquinone textile dyes under methanogenic conditions, Water Res. 38 (2004) 1838-1852.
+* [48]**Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonetech**, **Omonet**, **O

manual_annotation/200pdfs_in_mmd/10.1016_j.chemosphere.2009.03.048.mmd

@@ -0,0 +1,184 @@
+Levofloxacin ozonation in water: Rate determining process parameters and reaction pathway elucidation
+
+Bavo De Witte, Herman Van Langenhove, Karen Hemelsoet, Kristof Demeestere, Patrick De Wispelaere, Veronique Van Speybroeck, Jo Dewulf
+
+# Abstract
+
+Ozonation of the quinolone antibiotic levofloxacin was investigated with focus on both the levofloxacin degradation rate and degradation product formation. Degradation was about 2 times faster at pH 10 compared to pH 3 and 7 explained by direct ozonation at the unprotonated _N_a, one of the tertiary amines of the piperazinyl substituent. H2O2 concentration (2-100 mM) had only limited effect. Liquid chromatography - high resolution mass spectrometry revealed degradation at the piperazinyl substituent and the quinone moiety, with the relative importance of both pathways being strongly affected by changes in pH. Levofloxacin N-oxide concentrations reached up to 40% of the initial levofloxacin concentration during ozonation at pH 10. Degradation at the quinone moiety resulted in isatin and anthranilic acid type metabolites, probably formed through reaction with hydroxyl radicals. Ab initio molecular orbital calculations predicted radical attack mainly at C2 of the quinolone moiety. This is the carbon atom with the largest Fukui function. Reaction with ozone is expected to mainly occur at _N_a, characterized by the largest negative charge.
+
+## Introduction
+
+Fluorquinolones are synthetic antibiotics inhibiting bacterial DNA synthesis through binding with DNA gyrase and topoisomerase IV enzyme (Hooper, 1999). Nalidixic acid, the first fluoroquinolone, was introduced in 1962. First and second generation quinolones are active against Gram-negative bacteria and atypical pathogens. The latter are pathogens that can cause community-acquired pneumonia (Lee et al., 2002). The activity of third and fourth generation quinolones is extended to Gram-positive and anaerobic bacteria, respectively (Oliphant and Green, 2002). Ciprofloxacin, belonging to the second generation and introduced in 1987, was the mostly prescribed quinolone in Europe in 2003. However, a shift towards levofloxacin and maxilloxacin, introduced in 1996 and 1999, respectively, is noticed (Ferech et al., 2006).
+
+The increased use of quinolones has led to increased bacterial resistance (Jacoby, 2005). This can be partially due to the release of antibiotics into the environment. After administration, quinolones are only partially metabolized and their biotic transformation in the environment is slow (Huang et al., 2001) leading to wastewater treatment plant effluent concentrations up to 5.6 mg L-1 for ciprofloxacin (Batt et al., 2006). By consequence, physical-chemical removal technologies, such as advanced oxidation processes (AOPs), are a suitable alternative method for their removal from wastewater. AOPs are characterized by the generation of hydroxyl radicals at ambient conditions. Advanced oxidation of ciprofloxacin has been extensively studied (Dodd et al., 2006; Siminiceanu and Bobu, 2006; Paul et al., 2007; De Witte et al., 2008, 2009). In contrast, literature data on advanced oxidation of more recently introduced quinolones is scarce. In this paper, the ozonation of levofloxacin is discussed for the first time. The effect of process parameters pH and H2O2 is tested and degradation products are identified based on UV and high resolution mass spectrometry (HRMS) detection. Reactive sites are predicted based on ab initio molecular orbital calculations. This approach is widely applicable and has proven to be successful for a broad variety of molecules (Geerlings et al., 2003; Hemelsoet et al., 2005; Van Speybroeck et al., 2006).
+
+## Materials and methods
+
+### Chemicals
+
+Levofloxacin (>=98%) was delivered by Fluka (Germany). Other chemicals used were of reagents grade and were previously reported (De Witte et al., 2008).
+
+### Experimental setup and analytical procedures
+
+Levofloxacin ozonation was performed in a bubble column containing 1.75 L water buffered with 10.12 mM phosphate buffer (pH 3 and 7) or 2.53 mM borax buffer (pH 10). Initial levofloxacin concentration mounted 45.3 mM (16.4 mg L-1). The experimental set-up was identical as recently reported for ciprofloxacin ozonation (De Witte et al., 2008). Based on research on ciprofloxacin ozonation (De Witte et al., 2009), 2-100 mM H2O2 was added to the reactor during peroxone experiments. For experiments with radical scavengers, 30.45 mM _t_-butanol was added.
+
+Ozone in the gas flow was measured by an ozone analyzer (Anseros O2omat GM) by UV-light absorption at 253.7 nm. For levofloxacin determination, 5 mL liquid samples were taken and analyzed by liquid chromatography (LC)-UV spectroscopy identical to a previously described procedure (De Witte et al., 2009). Quantification of levofloxacin (295 nm) and its degradation products took place at the UV-absorbance maximum +4.5 nm. For identification of degradation products, 25 mL samples were concentrated by a factor of 125 by solid phase extraction. Compounds were separated by gradient LC and detected by UV and HRMS (De Witte et al., 2008). Comparisons with UV- and MS-spectra of analogous products (De Witte et al., 2008) and the parent compound allowed level 2 or full identification (De Witte et al., in press). Polyethylene glycols (PEG) were used as HRMS internal standard for determination of the accurate mass and chemical formula of the degradation products. An additional energy of 100 V was applied to the electrospray ionization needle (collision induced dissociation, CID) for enhancement of degradation product fragmentation. With CID, PEG ions were not stable as internal standard. They were used as external standard for determination of accurate mass of MS-fragmentation products. If the measured _m_/_z_ of the protonated compounds deviated less than 5 ppm from the theoretical values in the case of internal standards and less than 15 ppm in the case of external standards, the chemical formula was restrained.
+
+### Ab initio molecular orbital calculations
+
+All ab initio calculations were carried out using the Gaussian 03 software package. Density functional theory (DFT) (Parr and Yang, 1989) was applied due to its excellent cost-to-reliability performance compared to post-Hartree-Fock methods. Geometries were optimized using the B3LYP functional (Lee et al., 1988; Becke, 1993) and 6-31+G(d,p) Pople basis set. Subsequent single-point energy computations were performed using the meta-hybrid BMK functional (Boese and Martin, 2004) in combination with the large 6-311++G(3df,2p) basis set. DFT-based reactivity indicators, and in particular frontier orbital-related properties (i.e., Fukui functions (Fukui, 1973)) were computed at the BMK/B3LYP level of theory. Compared to the frontier orbitals HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molecular orbital) the Fukui functions contain more detailed information, taking also orbital relaxation effects into account. For more explanation on their definition we refer to the handbook of Parr and Yang (1989) or a review of Geerlings et al. (2003). Atomic charges and condensed-to-atoms values of the Fukui function were examined using the Mulliken population scheme (Mulliken, 1955).
+
+## Results and discussion
+
+### Parameter study
+
+The levofloxacin degradation curve for ozonation at pH 7 is presented as Supplementary material (Fig. S1, \(n\) = 3) and resulted in a half-life time of \(12.8\pm 0.2\) min (_n_ = 3) and more than 95% and 99% of the initial levofloxacin concentration was removed at 40 and 50 min of ozonation, respectively. The ozone consumption during 60 min of ozonation, calculated from the difference between the inlet and outlet gaseous ozone concentration, mounted \(0.61\pm 0.01\) mmol compared to 0.44 mmol for the blank experiment without levofloxacin. Half-life time (t1/2) as well as levofloxacin degradation rate constants at 10% degradation (_k_10%) was considered to compare experimental results. For \(k\)10% determination, a quadratic equation was fitted to the levofloxacin data points up to 95% degradation and the first derivative at 10% degradation was calculated.
+
+As can be seen from Table 1, degradation rate constants (_k_10%) are approximately two times faster at pH 10 compared to pH 7 and 3. Next, levofloxacin degradation is also faster compared to previously reported ciprofloxacin degradation at similar conditions (De Witte et al., 2009). Differences with ciprofloxacin are larger at pH 10 (t1/2 = 7.8 versus 13.8 min) compared to pH 7 and 3 (t1/2 = 12.8 and 16.0 min versus 15.9 and 17.6 min, respectively). Levofloxacin has a pKa-value of 6.20 for the carboxylic group, 5.20 for the N1-atom and 8.20 for the N4-atom of the piperazinyl substituent (Fig. 1) (Lin et al., 2004). Protonated amines are practically unreactive towards ozone whereas the lone electron pair of the unprotonated amine can react fast with ozone, leading to higher er degradation rates at higher pH (Munoz and von Sontag, 2000). Moreover, the N4-atom belongs to a tertiary amine group whereas ciprofloxacin has a secondary amine group at its piperazinyl substituent. Methyl groups are better electron donors than hydrogen
+
+\begin{table}
+\begin{tabular}{l l l l l} pH & H2O2 (μM) & t1/2 (min) & \(k\)10\({}^{\mathrm{m}}\)(mM min\({}^{-1}\)) & Ozone consumption during 60 min (mmol) \\
+3 & – & 16.0 & 1.77 ± 0.05 & 0.55 \\
+7 & – & 12.8 ± 0.2\({}^{\mathrm{a}}\) & 1.96 ± 0.10 & 0.61 ± 0.01\({}^{\mathrm{a}}\) \\
+10 & – & 7.8 & 3.80 ± 0.05 & 0.65 \\
+7 & 2 & 10.9 & 2.62 ± 0.17 & 0.62 \\
+7 & 10 & 11.9 & 2.17 ± 0.04 & 0.61 \\
+7 & 25 & 10.6 & 2.62 ± 0.10 & 0.62 \\
+7 & 50 & 11.6 & 2.31 ± 0.15 & 0.62 \\
+7 & 100 & 11.8 & 2.47 ± 0.12 & 0.59 \\
+3 & 10 & 15.6 & 1.77 ± 0.23 & 0.55 \\
+3 & 100 & 16.2 & 1.81 ± 0.18 & 0.59 \\
+10 & 10 & 8.1 & 3.75 ± 0.40 & 0.63 \\
+10 & 100 & 9.9 & 2.99 ± 0.54 & 0.65 \\ \end{tabular}
+\end{table}
+Table 1: Levofloxacin half-life time, degradation rate constants at 10 wt% degradation and ozone consumption during 60 min of ozonation for experiments at 45.3 μM initial levofloxacin concentration and varying pH and H2O2 concentration (O2 Lüntz = 2500 ppm, = 4.87 mg L−1, = 7.27 ± 0.1 °C).
+
+atoms. This will result in a higher electron density at the N\({}_{4}^{\prime}\)-atom for tertiary amines and, by consequence, faster ozonation rates and larger pH effects for tertiary amines compared to secondary amines. Differences between tertiary and secondary amines are in agreement with Munoz and von Sonntag (2000), who found rate constants of \(4.1\times 10^{6}\) M\({}^{-1}\) s\({}^{-1}\) for triethylamine versus \(9.1\times 10^{5}\) M\({}^{-1}\) s\({}^{-1}\) for diethylamine.
+
+At pH 7, addition of 2-100 mM H\({}_{2}\)O\({}_{2}\) increased the levofloxacin degradation rate constants (\(k_{10\mu 3}\)) by 11-34% (Table 1). The effect of H\({}_{2}\)O\({}_{2}\) addition on the levofloxacin degradation rate is, however, limited compared to the pH effect. Moreover, no trend in levofloxacin degradation rate as well as ozone consumption is derived as function of the H\({}_{2}\)O\({}_{2}\) amount added whereas an optimum concentration could be expected because H\({}_{2}\)O\({}_{2}\) can both promote and scavenge hydroxyl radicals (Gogate and Pandit, 2004). The limited effect of H\({}_{2}\)O\({}_{2}\) addition may indicate that the radical chain mechanism, described by Staehelin and Hogine (1985), is of minor importance for levofloxacin degradation compared to direct ozonation because levofloxacin contains several amine groups and an aromatic moiety, both reacting fast with ozone (\(10^{3}-10^{11}\) M\({}^{-1}\) s\({}^{-1}\), Munoz and von Sonntag (2000), Vandersmissen et al. (2008)). A second possibility is that ozonation of amines promotes OH-radical formation (Buffle, 2005), which initiates the radical chain mechanism and reduces the effect of H\({}_{2}\)O\({}_{2}\) addition.
+
+At pH 3, addition of 10 and 100 mM H\({}_{2}\)O\({}_{2}\) did not affect levofloxacin degradation (Table 1), probably because H\({}_{2}^{-}\) instead of H\({}_{2}\)O\({}_{2}\) plays the major role in OH-radical formation (pK\({}_{\rm{a}}\) of H\({}_{2}\)O\({}_{2}\) = 11.8, Beltran (2004)). At pH 10, however, addition of 100 mM H\({}_{2}\)O\({}_{2}\) reduced levofloxacin degradation efficiency due to scavenging of hydroxyl radicals at higher H\({}_{2}^{-}\) concentrations. Buxton et al. (1988) reported reaction constants of \(7.5\times 10^{9}\) and \(2.7\times 10^{7}\) M\({}^{-1}\) s\({}^{-1}\) between OH-radicals and H\({}_{2}\)O\({}_{2}\), respectively.
+N-oxide formation was not detected during ciprofloxacin oxidation (De Witte et al., 2008). Probably, the electron donating capacity of a methyl group compared to a hydrogen atom substituted at N\({}_{4}\)of the piperazinyl affects not only reaction rate but also degradation product formation. According to Munoz and von Sontnag (2000), ozonation of tertiary amines leads to ozonide ammonium zwitterions mainly followed by loss of dioxygen resulting in compound 2. The secondary amine (compound 3) can be formed together with formaldehyde through dissociation of the ozonide ammonium zwitteriton into the ozonide radical ion and the amine radical cation or through reaction of a tertiary amine with hydroxyl radicals (Munoz and von Sontnag, 2000). Compounds 4 and 5 are formed after multiple reactions, leading to enhanced degradation at the piperazinyl substituent.
+
+Compounds 6-10 have UV spectra different from levofloxacin which indicates degradation at the chromophore, i.e. the quinolone moiety. Based on UV-spectrum, molecular formula and MS fragmentation, these degradation products were found to be isatin (compounds 6, 8 and 10) or anthranilic acid analogues (compounds 7 and 9). Similar degradation products were identified during ciprofloxacin ozonation (De Witte et al., 2008). Formation of isatin and anthranilic acid analogues by means of hydroxyl radicals is possible through formation of intermediates A and B (Fig. 1) whereas direct ozonation can also lead to anthranilic acid analogues (Karl et al., 2006; De Witte et al., 2008).
+
+In conclusion, nine levofloxacin ozonation products were identified indicating degradation at the piperazinyl substituent and the quinolone moiety. No degradation products were found corresponding to degradation at the oxazinyl group (Fig. 1). However, the structure could not be identified for every degradation product, detected by HPLC-MS (Table 2).
+
+### Effect of pH, H\({}_{2}\)O\({}_{2}\) and t-butanol on degradation product formation
+
+Without sample concentration, LC-UV peak areas could be determined for seven degradation products at all investigated process conditions and different ozonation times. The chromatographic separation for samples at 20 min ozonation at pH 7 is given in Supplementary material, Fig. S3.
+
+For 2 out of the 7 compounds, no molecular formula could be obtained with HRMS using the threshold error value of 5 ppm. The maximum concentration for the two unidentified compounds were observed at pH 3 and 7 (data not shown). For the other compounds, maximum peak areas are plotted in Fig. 2. Addition of 10 mM H\({}_{2}\)O\({}_{2}\) at pH 3, 7 and 10 does not affect degradation product concentrations (Fig. 2). pH, on contrary, is an important parameter with respect to degradation product formation. Compounds 8 and 11 are mainly formed at pH 7 while addition of the radical scavenger t-butanol in excess at pH 7 excludes their formation. Compound 8 shows degradation at the quinolone moiety. Compound 11 was not identified but reveals a strong change in UV-spectrum compared to levofloxacin, also suggesting degradation at the quinolone moiety. The results suggest that degradation at the quinolone moiety is mediated by OH-radicals in agreement with previously reported ciprofloxacin degradation (De Witte et al., 2008). At pH 3, no OH-radicals are expected (Rivas et al., 2005). At pH 10, the N\({}_{4}^{-}\)atom of the piperazinyl is unprotonated and becomes an important reactive centre for ozone probably hampering high concentrations of compounds 8 and 11 at this pH.
+
+\begin{table}
+\begin{tabular}{l l l l l l l l} \hline \(\mathrm{f}_{\mathrm{R}}^{\mathrm{o}}\)(min) & [M+H]\({}^{\mathrm{o}}\) measured & Nominal mass (Da) & Molecular formula & Error\({}^{\mathrm{o}}\)(ppm) & Difference with levofloxacin & Degradation site moiety\({}^{\mathrm{a}}\) & No. \\ \hline
+3.82 & 365.159 & 364 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & \(-1.21\) & -\(\mathrm{c}_{4}\) & +H\({}_{2}\)O\({}_{3}\) & \\
+**9.45** & **338.152** & **337** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **1.83** & \(\mathrm{-c}_{2}\) & & **Qui** & **7** \\
+11.49 & 312.135 & 311 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & \(-1.06\) & -\(\mathrm{c}_{2}\)H\({}_{4}\) & & \\
+12.05 & 338.152 & 337 & \(\mathrm{c}_{15}\)H\({}_{20}\),N\({}_{4}\)F & 2.63 & -\(\mathrm{c}_{2}\) & & \\
+**15.23** & **320.141** & **319** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **0.04** & -\(\mathrm{c}_{3}\)H\({}_{20}\) & & **Qui** & **6** \\
+**15.88** & **294.126** & **293** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **2.70** & **-\(\mathrm{c}_{4}\)H\({}_{20}\)** & & **Pip. Qui** & **10** \\
+**17.69** & **354.145** & **353** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & \(-3.77\) & \(\mathrm{-c}_{2}\) & **+O** & **Pip. Qui** & **9** \\
+18.59 & 376.129 & 375 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & 0.18 & \(-\mathrm{H}_{2}\) & +O & \\
+**20.34** & **336.135** & **335** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **0.32** & -\(\mathrm{c}_{4}\)H\({}_{2}\) & & **Pip** & **4** \\
+20.68 & 354.145 & 353 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & \(-3.40\) & -\(\mathrm{c}_{2}\) & +O & & 11 \\
+20.88 & 362.151 & 361 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & \(-1.44\) & _Lewofloxacin_ & & & \\
+21.68 & 314.068 & 313 & \(\mathrm{c}_{13}\)H\({}_{20}\),N\({}_{4}\)F & 0.74 & -\(\mathrm{c}_{3}\)H\({}_{2}\) & +O\({}_{1}\) & & \\
+**21.77** & **348.137** & **347** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **3.67** & **-\(\mathrm{c}_{3}\)H\({}_{2}\)** & & **Pip** & **3** \\
+**22.99** & **336.136** & **335** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **0.44** & **–\(\mathrm{c}_{3}\)H\({}_{2}\)** & & **Pip**, Qui** & **8** \\
+**27.73** & **378.146** & **377** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **0.25** & & **+O** & **Pip** & **2** \\
+29.63 & 326.151 & 325 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & \(-0.31\) & \(-\mathrm{c}_{3}\) & & \\
+40.48 & 298.073 & 297 & \(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F & 0.93 & \(-\mathrm{c}_{3}\)H\({}_{2}\) & +O\({}_{2}\) & & \\
+**41.47** & **270.078** & **278** & **\(\mathrm{c}_{14}\)H\({}_{20}\),N\({}_{4}\)F** & **0.21** & \(-\mathrm{c}_{4}\)H\({}_{3}\) & & **Pip** & **5** \\
+**42.32** & 364.130 & 363 & \(\mathrm{c}_{15}\)H\({}_{20}\),N\({}_{4}\)F** & \(-0.10\) & \(-\mathrm{c}_{2}\) & +O & & \\ \hline \end{tabular}
+
+* The numbering follows the reaction pathway (Fig. 1), except for compound 11 for which no structure was identified.
+* HPLC retention time based on MS detection.
+* Difference between measured and theoretical mass.
+* pip = Piperazinyl and qui| quinolonic moiety.
+
+\end{table}
+Table 2: Levofloxacin degradation products\({}^{\mathrm{o}}\).
+
+Fig. 2: Maximum peak area of compounds 2, 3, 4, 8 and 11 during levofloxacin ozonation as a function of pH, H\({}_{2}\)O\({}_{2}\) and t-butanol concentration. The peak area of compound 2 (N-oxide) was divided with a factor 5 for reasons of visibility.
+
+Compounds 2 and 3 contain the same chromophore as levofloxacin while the UV-spectrum of compound 4 is only slightly shifted (data not shown). Therefore, concentration estimations can be based on the levofloxacin UV response factor. Concentrations of the N-oxide compound 2 reach up to 0.3, 11.4 and 18.1 mM at pH 3, 7 and 10, respectively. Addition of t-butanol at pH 7 increased concentrations up to 25.4 mM. These results indicate that compound 2 formation is favored by direct ozonation at the unprotonated tertiary amine while its degradation is partially affected by the presence of OH-radicals.
+
+Compounds 3 and 4 reveal equal or lower concentrations at pH 10 (up to 0.9 and 0.4 mM, respectively) compared to pH 7 (up to 1.0 and 2.3 mM, respectively) while t-butanol addition at pH 7 reduces their concentration to 0.5 and 0.3 mM (Fig. 2). This indicates that direct ozonation at the partially unprotonated N-atom can not be the main reaction pathway at pH 7: hydroxyl radicals are suggested to be more important for formation of these products at this pH.
+
+High concentration of the N-oxide compound 2 proves that direct ozonation is important for levofloxacin ozonation. Moreover, the levofloxacin degradation rate as a function of pH can be linked to N-oxide formation. However, the influence of t-butanol addition on the formation of compounds 3, 4, 8 and 11 reveals that the radical chain mechanism also contributes to levofloxacin degradation, affecting degradation at the quinone moiety. Since the carbonyl and carboxyl at the quinolone moiety are essential for binding at the DNA gyrase or topoisomerase IV target (Chu and Fernandes, 1989), the radical chain mechanism is likely to inactivate the drug instantly.
+
+### Rationalization of reactivity by ab initio molecular orbital calculations
+
+It was investigated whether the experimentally observed oxidation products can be further rationalized in terms of ab initio computations. A conformational analysis was performed, rotating the carboxyl group on one hand and varying the planarity of the piperazinyl substituent on the other. The latter leads to a twist boat configuration for the ring substituent (dihedral angle of 59.5deg, Fig. 3), whereas the quinolone moiety and carboxyl functional group accords with an almost planar substructure (maximal dihedral angle of 3.8deg, Fig. 3). The most stable conformer was ultimately obtained taking into account an explicit hydrogen molecule, forming an external hydrogen bridge with the hydrogen-atom of the carboxyl group. The optimized geometry of the gas-phase levofloxacin molecule with an explicit water solvent molecule is depicted in Fig. 3.
+
+The reactivity and site selectivity of the optimized geometry was examined using DFT-based reactivity indicators. According
+
+\begin{table}
+\begin{tabular}{l c c c} \hline  & Charges & \(f\) & \(f\) \\ \hline N1 & −0.843 & 0.051 & 0.102 \\ C2 & 0.155 & −0.027 & **0.128** \\
+015 & −0.759 & 0.057 & **0.082** \\
+016 & −0.791 & 0.034 & 0.054 \\ C17 & 0.698 & 0.007 & −0.028 \\
+018 & −0.638 & 0.007 & 0.017 \\ N1 & −0.802 & **0.175** & 0.102 \\ N4 & **–0.941** & 0.048 & 0.031 \\ C2 & 0.407 & −0.004 & −0.004 \\ \hline \end{tabular}
+\end{table}
+Table 3: Mulliken atomic charges and condensed-to-atoms Fukui functions \(f\)- and \(f\)-corresponding with an electrophilic and radical attack, respectively.
+
+Figure 4: Three-dimensional iso-surfaces of the frontier orbitals (a, b) and the Fukui function for radical attack (c) for the levofloxacin molecule.
+
+Figure 3: Optimized geometrical structure of the levofloxacin molecule. Important bond lengths (in Å) and dihedrals (in degrees) are indicated.
+
+to the early work of Klopman (1968), reactions can be classified as predominantly frontier-orbital- or charge-controlled.
+
+The radical Fukui function \(f\)0 is used to examine the interaction with hydroxyl radicals. The preferred site is identified by the maximal value of the Fukui function. The condensed-to-atoms values (Table 3) indicate enhanced reactivity at the levofloxacin C2 and (to a lesser extent) the N1 and N'-atom. Reactions occurring at the nitrogen atoms can, however, be affected by steric hindrance effects which are not encapsulated in the definition of the reactivity indicators. The three-dimensional iso-surfaces of the HOMO and LUMO frontier orbitals and the radical Fukui function are shown in Fig. 4. It can be seen that the combination of the HOMO and LUMO (taking the mean average) can be regarded as an initial approximation of the Fukui function. Compared to the condensed-to-atoms results, the \(f\)0 iso-surface has the advantage to disregard the effect of the population analysis scheme. The C2-atom is clearly found to be a suitable site for radical attack. This is in agreement with the formation of isatin and anthranilic acid analogues through degradation at the quinolone moiety. These products were excluded when the radical scavenger t-butanol was added. Moreover, Karl et al. (2006) proved that degradation of the quinolone enro-fluxoxin by hydroxyl radicals with formation of isatin and anthranilic acid analogues proceeds through breaking of the C2-C3 bond. By consequence, radical interactions of levofloxacin with hydroxyl radicals can be modeled as frontier-orbital controlled.
+
+Interaction with ozone corresponds to an electrophilic attack. The electrophilic Fukui function _f_- does not support the experimental observations as the _f_- values indicate N1 to be the most reactive centre (Table 3). However, due to the less favorable overlap between the frontier orbitals of levofloxacin and ozone, the interaction will be dominated by charge/charge effects and is thus expected to occur at the sites with the highest electron density. The Mulliken scheme shows the largest negative charge at the N4-atom of the piperazinyl substituent, which is indeed the most reactive centre towards ozone. Ozone nor OH-radicals reactions are predicted at the oxazinyl group, confirming the lack of products with degradation at this part of the molecule.
+
+## Conclusions
+
+Levofloxacin ozonation at different pH and different H2O2 amounts revealed strong influence of pH on levofloxacin degradation rate as well as reaction pathways whereas H2O2 addition had only limited effect. At pH 10, the tertiary amine at the piperazinyl substituent is unprotonated leading to fast ozonation at this site of the molecule and high concentrations of the N-oxide degradation product. At pH 7, degradation at the quinolone moiety is also observed, probably mediated by reaction with hydroxyl radicals. This was confirmed by ab initio molecular orbital calculations which predicted the carbon atom of the quinolone moiety with the largest value of the Fukui function as the most reactive centre for radical attack.
+
+## Acknowledgements
+
+We acknowledge financial support from the Flemish Government for the MAT 95XP-Trap in the framework of the Flemish investment support for heavy research equipment. This work is supported by the Fund for Scientific Research-Flanders (FWO) and the Research Council of Ghent University.
+
+## Appendix A Supplementary material
+
+Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.chemosphere.2009.03.048.
+
+## References
+
+* Batt _et al._ (2006) Batt, A.I., Bruce, I.B., Aga, D.S., 2006. Evaluating the vulnerability of surface waters to antibiotic contamination from varying wastewater treatment plant discharges. Environ. Pollut. 142, 295-302.
+* Becke (1993) Becke, A.D., 1993. Density-functional thermochemistry. 3. The role of exact exchange. J. Chem. Phys. 98, 5648-5652.
+* Beltran (2004) Beltran, F.J., 2004. Ozone Reaction Kinetics for Water and Wastewater Systems. CRC Press, Boca Raton, FL.
+* Boese and Martin (2004) Boese, A.D., Martin, J.M.L., 2004. Development of density functionals for thermochemical kinetics. J. Chem. Phys. 121, 3405-3416.
+* Buffie (2005) Buffie, M.O., 2005. Mechanistic investigation of the initial phase of ozone decomposition in drinking water and wastewater. Ph.D. Dissertation, Swiss Federal Technology, Zurich.
+* Burton _et al._ (1988) Burton, G.V., Greenstock, C.L., Phillip Helman, W., Ross, A.B., 1988. Critical review of rate constants for reaction of hydrated electrons, hydrogen atoms, and hydroxyl radicals (OH/O ) in aqueous solution. J. Phys. Chem. Ref. Data 17, 513-586.
+* Chu and Fernandes (1989) Chu, D.T., Fernandes, P.R., 1989. Structure-activity relationships of the fluorquinolones. Antimicrob. Agents Chemother. 33, 131-135.
+* De Witte _et al._ (2008) De Witte, B., Dewulfi, J., Demesetere, K., Van De Veyreux, V., De Vingelenez, P., Van Langenhoe, H., 2008. Ozonation of ciprofloxacin in water: HRMS identification of reaction products and pathways. Environ. Sci. Technol. 42, 4889-4895.
+* De Witte _et al._ (2009) De Witte, B., Dewulfi, J., Demesetere, K., Van Langenhoe, H., 2009. Ozonation and advanced oxidation by the provzone process of ciprofloxacin in water. J. Hazard. Mater. 161, 701-708.
+* De Witte _et al._ (2006) De Witte, B., Van Langenhoe, H., Demesetere, K., Dewulfi, J., in press. Advanced oxidation of pharmaceuticals: chemical analysis and biological assessment of degradation products. Crit. Rev. Environ. Sci. Technol.
+* Dodd _et al._ (2006) Dodd, M.C., Buffie, M.O., Von Guttenauer, U., 2006. Oxidation of antibacterial molecules by aqueous ozone: mepley-specific reaction kinetics and application to ozone-based wastewater treatment. Environ. Sci. Technol. 40, 1969-1977.
+* Ferech _et al._ (2006) Ferech, M., Coenen, S., Malhotra-Kumar, S., Dworakova, K., Hendrickx, E., Suetens, C., Gossens, H., 2006. European surveillance of antimicrobial consumption (ESAC): outpatient quinolone use in Europe. J. Antimicrob. Chemother. 58, 423-427.
+* Fukui (1973) Fukui, F., 1973. Theory of Orientation and Stereoselection. Springer-Verlag, Berlin,.
+* Geerlings _et al._ (2003) Geerlings, P., De Proft, F., Langenheuer, W., 2003. Conceptual density functional theory. Chem. Rev. 103, 1793-1800.
+* Gogate and Pandit (2004) Gogate, P.R., Pandit, A.R., 2004. A review of imperative technologies for wastewater treatment II: hybrid methods. Adv. Environ. Res. 8, 553-597.
+* Hemelseed _et al._ (2005) Hemelseed, K., Lestrasheghe, D., Van Speybroeck, V., Waroquier, M., 2005. Bifunctional acid-base catalyzed reactions in zeolites from the HSAb viewpoint. Chem. Phys. Lett. 419, 10-15.
+* Hooper (1999) Hooper, D.C., 1999. Mode of action of fluoroquinolones. Drugs 58 (Suppl. 2), 6-10.
+* Huang _et al._ (2001) Huang, C., Renew, J.E., Smely, K.L., Pinkston, K., Sedlak, D.L., 2001. Assessment of potential antibiotic contaminants in water and preliminary occurrence analysis. Water Resour. Update 120, 30-40.
+* Jacoby (2005) Jacoby, G.A., 2005. Mechanisms of resistance to quinolones. Clin. Infect. Dis. 41, S120-S126.
+* Karl _et al._ (2006) Karl, W., Schneider, J., Wetzstein, H.-G., 2006. Outlines of an "exploding" network of metabolites generated from the fluoroquinolone enrofloxacin by the brown rot fungus _Glooseophyll stimatum_. Appl. Microbiol. Biotechnol. 71, 101-113.
+* Klopman (1968) Klopman, G.J., 1968. Chemical reactivity and concept of charge- and frontier-controlled reactions. J. Am. Chem. Soc. 203, 223-234.
+* Kultanhulved _et al._ (2004) Kultanhulved, P., Barbuch, R.S., Vij, P., Rener, G.A., Martiriz, E.L., Hadden, C.E., Goodwin, LA, Ehlhardt, W.J., 2004. Selective reduction of N-oxides to amines: application to drug metabolism. Drug Metab. Disps. 32, 966-972.
+* Lee _et al._ (1988) Lee, C.T., Yang, W.T., Park, R.C., 1988. Development of the Calle-Salvert correlation-energy formula into a function of the electron-density. Phys. Rev. 87, 785-789.
+* Lee _et al._ (2002) Lee, S.J., Lee, M.G., Jeon, M.J., Jung, K.S., Lee, H.K., Kishimoto, T., 2002. Atypical pathogens in adult patients admitted with community-acquired pneumonia in Korea. Jpn. J. Infect. Dis. 55, 157-159.
+* Lin _et al._ (2004) Lin, C.-E., Deng, Y.-J., Liao, W.-S., Sun, S.-W., Lin, W.-Y., Chen, C.-C., 2004. Electrophoretic behavior and R\({}_{\text{x}}\) determination of quinolones with a piperazinyl substituent by capillary zone electrophoresis. J. Chromatogr. A 1051, 283-290.
+* Mulliken (1955) Mulliken, R.S., 1955. Electronic population analysis on LCAO-MO molecular wave functions. J. Chem. Phys. 23, 1833-1840.
+* Munoz (2000) Munoz, F., von Sontate, C., 2000. The reactions of ozone with tertiary amines including the complexing agents nitrilotriacetic acid (NTA) and ethylenediaminetracacetic acid (EDT) in aqueous solution. J. Chem. Soc. Perkin. T. 2, 2029-2033.
+* Olightam and Green (2002) Olightam, C.M., Green, C.M., 2002. Quinolones: a comprehensive review. Am. Fam. Physician 65, 455-464.
+* Parr _et al._ (1984) Parr, R.G., Yang, W., 1984. Density-Functional Theory of Atoms and Molecules. Oxford University Press, New York.
+* Paul _et al._ (2007) Paul, T., Miller, F.I., Strathmann, T.J., 2007. Visible-light-mediated TiO, photocatalysts of fluoroquinolone antibacterial agents. Environ. Sci. Technol. 41, 4720-4727.
+
+* [2005] Rivas, F.J., Beltran, F.J., Acedo, B., Garcia Araya, J.F., Carbajo, M., 2005. Kinetics of the ozone-ph-chiorobenzene acid reaction. Ozone-Sci. Eng. 27, 3-9.
+* [2006] Siminicanu, I., Bobu, M.M., 2006. A comparative study of the ciprofloxacin degradation in water by several advanced oxidation processes. Rev. Chim.-Bucharest 57, 1082-1085.
+* [2005] Staehelin, J., Hoigen, J., 1995. Decomposition of ozone in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. Environ. Sci. Technol. 19, 1206-1213.
+* [2009] Van Speybroeck, V., Moonen, K., Hemelsoet, K., Stevens, C.V., Waroquier, M., 2006. Unexpected four-membered over six-membered ring formation during the synthesis of azatherocyclic phosphonates: experimental and theoretical evaluation. J. Am. Chem. Soc. 128, 8468-8478.
+* [2008] Vandersmissen, K., De Smedt, F., Vinckier, C., 2008. The impact of traces of hydrogen peroxide and phosphate on the ozone decomposition rate in "pure water". Ozone-Sci. Eng. 30, 300-309.

manual_annotation/200pdfs_in_mmd/10.1016_j.chemosphere.2017.01.007.mmd

@@ -0,0 +1,209 @@
+# Effects of oil dispersant on ozone oxidation of phenanthrene and pyrene in marine water
+
+Yanyan Gong, Dongye Zhao
+
+[MISSING_PAGE_POST]
+
+# Highlights Highlights
+
+# Highlights
+
+# Highlights Highlights
+
+# Highlights
+
+# Highlights Highlights
+
+# Highlights
+
+[MISSING_PAGE_POST]
+
+# Highlights Highlights Highlights
+
+# Highlights Highlights
+
+# Highlights Highlights Highlights
+
+# Highlights Highlights Highlights
+
+# Highlights Highlights Highlights Highlights
+
+Scelfo and Tjeerdema, 1991).
+
+Polycyclic aromatic hydrocarbons (PAHs) are a class of important oil hydrocarbons that are of great environmental concern due to their potential toxicity and environmental persistency (Nam et al., 2008). The BPs Macondo well oil contained approximated 3.9% of PAHs by weight. The DWH oil spill released approximately 2.1 x 107 kg of PAHs into the Gulf of Mexico (Reddy et al., 2012).
+
+Once released into the environment, PAHs undergo a number of physical and chemical processes, such as dissolution and volatilization (Liu et al., 2012), adsorption (Yang et al., 2005), bio-accumulation (Baumard et al., 1998), biodegradation (Baumard et al., 1998), and photodegradation (D'Auria et al., 2009). Another potentially important, yet overlooked, abiotic process affecting the fate of PAHs in the Gulf coast is oil degradation by tropospheric ozone, which is produced by reaction of sunlight with volatile organic compounds and nitrogen oxides in air. High levels of ozone have been widely reported at the ground level along the Gulf coast. For example, based on the 2010 monitoring data, the 8-h ozone level in Alabama air ranged from 60 to 92 ppb (EPA, 2015). Ozone levels over an oil slick may be much higher than the normal value due to the heavy evaporation of hydrocarbons from leaked oil reaching the surface (Ryerson et al., 2011).
+
+Ozone is one of the most effective oxidants (E0 = +2.07 V) and has been widely applied to degrade various organic chemicals including PAHs in engineered processes (Broseus et al., 2009; Chelme-Ayala et al., 2011; Liu et al., 2014; Marquez et al., 2014). Two primary mechanisms have been proposed for ozone oxidation of PAHs: (1) direct attack by O3 via cycloaddition or electrophilic reaction; and (2) indirect attack by free radicals (primarily hydroxyl radical, OH*) resulting from decomposition of ozone (Masten and Davies, 1994; Zhao et al., 2011). Beltran et al. (1995) examined the role of hydroxyl radical scavengers on ozone oxidation of fluorene, phenanthrene, and acenaphnthene in aqueous solutions, and concluded that the ozonation of fluorene was due to both direct and hydroxyl radical reactions while phenanthrene and acenaphnthene was only due to direct reactions with ozone.
+
+The ozonation efficiency of PAHs in water depends on several factors including ozone concentration, pH, and temperature (Beltran et al., 1995). Beltran et al. (1995) observed that the oxidation rate of fluorene increased with increasing ozone partial pressure from 116 to 1015 Pa, with increasing pH from 2 to 12, and with increasing reaction temperature from 4 to 20 degC. However, little is known on the effects of oil dispersants on the ozone oxidation kinetics of PAHs. Moreover, the influences of other factors such as aqueous ozone concentration, pH, ionic strength (IS), and temperature on PAHs degradation in the presence of oil dispersant have not yet been explored.
+
+The overall goal of this study was to determine effects of a stereotype oil dispersant (Corexit EC9500A) on the ozone degradation rates of PAHs in seawater. Phenanthrene and pyrene were selected to represent typical oil-related PAHs. The specific objectives were to: (1) investigate effects of various concentrations of the dispersant on the ozone degradation rate of phenanthrene and pyrene in seawater; and (2) examine effects of aqueous ozone concentration, pH, IS, and temperature on ozone degradation of pyrene in dispersant solutions.
+
+## Materials and methods
+
+### Materials
+
+Sewater was collected from the top 30 cm of the water column from Grand Bay, Al, USA in December, 2010. The latitudes/longitudes of the sampling site were 30.37926/88.30684. The seawater sample was stored in sealed containers at 4 degC in the refrigerator. Before use, the seawater was first passed through 0.45 mm membrane filters of cellulose acetate to remove suspended solids, and then sterilized at 121 degC for 35 min via autoclaving. Separate tests confirmed that the membrane filters did not retain phenanthrene or pyrene in the solutions. Detailed properties of the seawater sample have been described elsewhere (Gong et al., 2015; Gong et al., 2014). In brief, pH of the seawater was 8.8, dissolved organic matter (DOM) was 0.43 mg/L as total organic carbon (TOC), and IS was 0.7 M. Phenanthrene and pyrene in the seawater were 0.0029 and 0.0028 mg/L, respectively.
+
+All chemicals used in this study were analytical grade or higher. Phenanthrene, pyrene, and methanol were purchased from Alfa Aesar (Ward Hill, MA, USA). NaOH and NaCl were obtained from Fisher Scientific (Fair lawn, NJ, USA). Acetonitrile (HPLC grade) was purchased from EMD Millipore Corporation (Billerica, MA, USA). HCl was acquired from BDH Aristar (West Chester, PA, USA). Corexit EC9500A was acquired through the courtesy of Nalco Company (Naperville, IL, USA). The critical micelle concentration (CMC) of Corexit EC9500A was determined to be 22.5 mg/L from our prior work (Gong et al., 2014).
+
+### Experimental apparatus
+
+A schematic of the experimental set-up for ozonation is depicted in Fig. 1. Ozone was generated from dry and pure air using an AZZ ozone generator (Model HB5735B, AZZ Ozone Inc., Louisville, Kentucky, USA), which is able to generate a maximum of 1 g ozone h-1. Gaseous ozone was passed through the surface of the reaction solution which was continuously mixed using a magnetic stirrer and a stir bar. The flow of ozone into the reactor was regulated at 500 mL/min using an Aalborg mass flow controller (Model GFC17, Orangeburg, New York, USA). Ozone concentration in the gas phase was analyzed by an ozone monitor M106-L (2B Technologies, Inc., Boulder, CO, USA) through measuring the ultra violet absorbance at 254 nm. Excess ozone was passed into two gas absorption bottles containing 2% KI solution. All tubes from the ozone generator to the reactor and the gas absorption bottles were made of Teflon to avoid adsorption of the gas.
+
+### Effects of dispersant on ozone oxidation of phenanthrene and pyrene
+
+Separate stock solutions of phenanthrene (1.4 g/L) and pyrene (0.3 g/L) were prepared in methanol, which were shaken overnight to assure complete dissolution. Then, the solutions were diluted with seawater to obtain a phenanthrene solution of 400 mg/L and a pyrene solution of 60 mg/L, respectively, to simulate PAHs-contaminated seawater during and after the DWH oil spill. The concentrations chosen here are based on: (1) the values reported in previous studies which investigated the ozone oxidation of PAHs. For instance, Beltran et al. (1995) used a phenanthrene concentration of 516 mg/L, and Corless et al. (1990) employed pyrene concentrations from 10 to 200 mg/L; and (2) the solubility of these two compounds in seawater. The solubility of phenanthrene and pyrene was measured to be 766 and 135 mg/L in the seawater.
+
+Batch ozone degradation kinetic tests were carried out in a well-controlled glass reactor with a surface area of 78 cm\({}^{2}\) and a volume of 650 mL. In each batch, the reactor was filled with 300 mL of a seawater solution (phenanthrene = 400 mg/L or pyrene = 60 mg/L), and stirred gently with a magnetic stirrer to simulate the ocean wave actions and maintain uniform PAHs distribution. Control tests (carried out without turning on the ozone generator) indicated that phenanthrene/pyrene loss due to volatilization and sorption to the reactor wall was negligible. During the tests, 1 mL each of the solution was sampled from the reactor at predetermined times and analyzed for phenanthrene/pyrene remaining. To investigate the effects of the dispersant, the tests were conducted in the presence of 0, 18, and 180 mg/L of Corexit EC9500A. The dispersant concentrations were calculated based on the initially added amounts of the dispersant. All experiments were conducted in duplicate at \(22\pm 1\) degC.
+
+Experiments were also carried out in the glass reactor without phenanthrene/pyrene to determine the aqueous ozone concentrations. A phosphate buffer with a pH of 8.5 was prepared in DI water with 0.005 M NaH2PO4 and 0.005 M Na2HPO4. As the gaseous ozone passed through the surface of the reaction solution, samples were withdrawn at predetermined times. The ozone concentration was measured colorimetrically following the Indigo method (Bader and Hoigne, 1981). Ozone concentrations in the presence of 0, 18, and 180 mg/L of Corexit EC9500A in seawater were measured under otherwise identical conditions.
+
+### Effects of environmental factors on ozone degradation of pyrene in dispersant solutions
+
+Effects of aqueous ozone concentration, pH, IS, and temperature were investigated through similar pyrene kinetic experiments in the presence of 18 mg/L of the dispersant. To study the ozone concentration effect, the aqueous ozone concentration was increased from 0.07 to 0.87 mg/L while the PAH and dispersant concentrations were kept fixed. To examine the pH effect, the initial solution pH was adjusted to 5.0 and 8.5 using 0.1 M HCl. After 2 h, the solution pH slightly changed to 5.3 and 8.2, respectively, probably due to the relatively low concentrations of pyrene (40 mg/L) and dispersant (18 mg/L) during the experiments. To test the IS effect, IS of the reaction solution was varied from 0.01 to 0.7 M. DI water was added to the original seawater IS (0.7 M) to dilute it down to 0.01 M. To investigate the temperature effect, the experiments were carried out at 10 and 22 degC, which represent the lowest and highest seawater temperatures in the Grand Bay area.
+
+### Analytical methods
+
+Phenanthrene and pyrene concentrations were determined using an HPLC system (HP series 1100, Hewlett Packard, CA, USA) equipped with a UV detector and a Zorbax SB-C18 column (150 x 468 mm). The mobile phase consists of 70% acetonitrile, 30% water, and 0.1% phosphoric acid. For phenanthrene, the injection volume was 80 mL, the mobile phase flow rate was 12 mL/min, and the optimal UV detection wavelength was found to be 250 nm, which afforded a phenanthrene detection limit of 4 mg/L. For pyrene, the injection volume was 100 mL, the mobile phase flow rate was 1.0 mL/min, and the optimal UV detection wavelength was determined to be 240 nm, which afforded a pyrene detection limit of 2.5 mg/L.
+
+## Results and discussion
+
+### Effects of oil dispersant on ozone oxidation of PAHs in seawater
+
+The ozone oxidation of a given compound in water can be due to direct attack by molecular ozone and/or the indirect reaction with free radicals (Staehelin and Hoigne, 1985). The radicals are from the decomposition of ozone, which is initiated by the hydroxide ion or other substances present at trace concentrations. The pseudo-first order kinetic law was used to investigate the degradation kinetics of the PAHs (Ning et al., 2015; Lin et al., 2014):
+
+\[\text{In}(\text{C}_{\text{f}}/\text{C}_{0})=-kt \tag{1}\]
+
+where \(C\)0 and _C_f are the reactant concentrations at irradiation time 0 and \(t\), respectively, and \(k\) is an apparent rate constant. For phenanthrene, \(k\) is the rate constant due to direct O3 oxidation, while for pyrene, \(k\) equals to the sum of the rate constants due to direct O3 oxidation and indirect degradation by radicals such as hydroxyl radicals. The \(k\) value was obtained by fitting Eqn. (1) to the corresponding experimental kinetic data (Fig. 2).
+
+Fig. 2a and b display an unusual two-stage kinetic profile (stage 1 and stage 2). For phenanthrene, a slow kinetic rate in the first 60 min with or without the dispersant was observed, which was followed by a much faster rate thereafter. Similarly, for pyrene, a slow initial reaction rate was observed in the first 45 min without dispersant and in the first 90 min with the dispersant, followed by a much faster rate. Accordingly, the pseudo-first order kinetic model was applied to the two stages separately to interpret the ozone degradation data. Table 1 lists the resultant kinetic parameters. For phenanthrene, the degradation rate constant in stage 1 (_k_) was determined to be 0.041 min-1, which increased to 0.0214 min-1 (a 52% increase) in stage 2. Likewise, the pyrene degradation rate constant in stage 2 (0.0643 min-1) was 1.3 times greater than that in stage 1 (0.0285 min-1). It is noteworthy that pyrene was more easily degraded than phenanthrene with much higher reaction rate constants in both stages. The difference can be attributed to the different degradation mechanisms. It has been reported that the oxidation of phenanthrene was primarily through the direct attack of ozone (Beltran et al., 1995), while the oxidation of pyrene involved both the direct reaction and the indirect reaction with radicals (Yao et al., 1998). Yao et al. (1998) reported that at the early
+
+Figure 1: Schematic of the experimental set-up including online ozone generation, ozonation reactor, and exhaust gas treatment apparatuses.
+
+stage of ozonation, the ring cleavage of pyrene first occurred at the 4, 5 positions, forming phenanthrene-type products; upon further ozonation, pyrene and the phenanthrene-type products were further degraded, resulting in an increase in the concentrations of biphenyl-type products.
+
+The two-stage degradation kinetics were in consistent with the dissolved ozone concentration changes in the seawater (without phenanthrene/pyrene) over the reaction time (Fig. 3). The overall ozonation rates of both phenanthrene and pyrene increased with the increment in the residual ozone concentration. During the first 60 min, the aqueous ozone concentration remained constant at -0.008 mg/L probably due to the initial ozone demand, i.e., reactions with other chemical compositions (e.g., DOM and sulfide compounds) in seawater (Liu et al., 2001), which can be revealed by comparing with the much higher ozone level when DI water was used as the control. After the initial ozone consumption, the ozone concentration sharply increased to 0.04 mg/L at 90 min, and further to 0.11 mg/L at 120 min. In dilute aqueous solutions, the self-decomposition of ozone is very important compared to the consumption of ozone by organic compounds, as such, a smaller percentage of applied ozone would go to the oxidation of the target chemicals (Yao et al., 1998). Since the ozonation of phenanthrene was mainly direct ozonation, the degradation was slower in the first 60 min at low ozone concentrations and faster thereafter at high ozone concentrations. While for pyrene, the degradation was slower in the first 45 min and faster thereafter probably due to the contributions of both direct and indirect ozonation.
+
+Fig. 2 shows that the ozonation rates of phenanthrene and pyrene decreased with increasing dispersal concentration. At 60 min, 64% of phenanthrene was degraded when no dispersant was present, and the degradation was lowered to 48% and 19%, respectively, in the presence of 18 and 180 mg/L of the dispersant. The rate constant \(k\)1 of phenanthrene was suppressed from 0.0141
+
+\begin{table}
+\begin{tabular}{l l l l l l l l} \hline \multicolumn{2}{l}{Corecir ECG500A} & \multicolumn{2}{l}{Phenanthrene} & \multicolumn{2}{l}{Pyrene} \\ \cline{2-7} \multicolumn{2}{l}{(mg/L)} & \multicolumn{1}{l}{0-60 min} & \multicolumn{2}{l}{60-120 min} & \multicolumn{2}{l}{0-45 min without dispersant or 0–90 min} & \multicolumn{2}{l}{45-120 min without dispersant or 90–120 min with dispersant} \\ \multicolumn{2}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} & \multicolumn{1}{l}{} \\  & & & & & & & & \\  & & & & & & & & \\  & & & & & & & & \\ \hline
+0 & (1.41 \(\pm\) 0.01) × 10−2 & 0.990 & (2.14 \(\pm\) 0.03) × 10−2 & 1 & (2.85 \(\pm\) 0.06) × 10−2 & 0.999 & (6.43 \(\pm\) 1.04) × 10−2 & 1 \\
+18 & (0.96 \(\pm\) 0.02) × 10−2 & 0.997 & (1.32 \(\pm\) 0.05) × 10−2 & 0.998 & (0.90 \(\pm\) 0.01) × 10−2 & 0.978 & (3.15 \(\pm\) 0.07) × 10−2 & 1 \\
+180 & (0.35 \(\pm\) 0.01) × 10−2 & 0.997 & (0.43 \(\pm\) 0.01) × 10−2 & 0.999 & (0.42 \(\pm\) 0.01) × 10−2 & 0.958 & (1.34 \(\pm\) 0.04) × 10−2 & 1 \\ \hline \end{tabular}
+**Note:**\(k_{1}\) and \(k_{2}\) (min−1): ozone oxidation rate constant in stage 1 and stage 2, respectively; \(R^{2}\): coefficient of determination, \(R^{2}=1-\frac{\sum(n-y_{\text{proj}})^{2}}{\sum(n-y_{\text{proj}})^{2}}\), where \(y_{\text{i}}\) and \(y_{\text{proj}}\) are observed data and model values, respectively, and \(\overline{y}\) is the mean of the observed data.
+
+\end{table}
+Table 1: Pseudo-first order ozone oxidation rate constants for phenanthrene and pyrene in the absence or presence of Corecir EC9500A. Errors refer to the standard error.
+
+Fig. 3: Concentration histories of aqueous ozone in the reactor systems with different media: DI water, seawater, a monomeric dispersant solution (Corecir EC9500A – 18 mg/L), and micellar dispersant solution (Corecir EC9500A – 180 mg/L). Solution volume – 300 mL, solution pH – 8.5, temperature – 22 °C. Data plotted as mean of duplicates and the error bars (calculated as standard error) indicate data reproducibility.
+
+Fig. 2: First-order kinetic plots of **(a)** phenanthrene and **(b)** pyrene ozonation in seawater and in a monomeric dispersant solution (Coreit EC9500A – 18 mg/L) and micellar dispersant solution (Corecit EC9500A – 180 mg/L), Initial phenanthrene – 400 μg/L, initial pyrene – 60 μg/L, solution volume – 300 mL, solution pH – 8.0–8.5, temperature – 22 °C. Data plotted as mean of duplicates and the error bars (calculated as standard error) indicate data reproducibility.
+
+min\({}^{-1}\) (without dispersant) to 0.0096 min\({}^{-1}\) (by 32%) (with 18 mg/L dispersant) and to 0.0035 min\({}^{-1}\) (by 75%) (with 180 mg/L dispersant), and \(k_{2}\) from 0.0214 min\({}^{-1}\) (without dispersant) to 0.0132 min\({}^{-1}\) (by 38%) (with 18 mg/L dispersant) and 0.0043 min\({}^{-1}\) (by 80%) (with 180 mg/L dispersant). Similar retardation phenomena were also observed for pyrene. Within 60 min, the degradation of pyrene was decreased from 82% without dispersant to 38% and 21% with 18 and 180 mg/L of the dispersant, respectively. \(k_{1}\) of pyrene was suppressed from 0.0285 min\({}^{-1}\) (without dispersant) to 0.0090 min\({}^{-1}\) (by 68%) (with 18 mg/L dispersant) and to 0.0042 min\({}^{-1}\) (by 85%) (with 180 mg/L dispersant), and \(k_{2}\) from 0.0643 min\({}^{-1}\) (without dispersant) to 0.0315 min\({}^{-1}\) (by 51%) (with 18 mg/L dispersant) and to 0.0134 min\({}^{-1}\) (by 79%) (with 180 mg/L dispersant).
+
+The dispersant Corexit EC9500A can affect the oxidation of PAHs in two contrasting ways. On the one hand, the dispersant may accelerate the degradation by lowering the surface tension, which may reduce the gas transfer barrier between the ozone gas phase and the liquid phase, and thereby leading to an increase in the soluble ozone concentration in the solution (Chu et al., 2006); on the other hand, the dispersant itself can compete for the reactive ozone and free radicals resulting in a decrease in the aqueous ozone concentration and thereby reduced ozonation rate (Amat et al., 2007; Ikehata and El-Din, 2004). The overall effect depends on the extent of these two contrasting effects. Fig. 3 compares the changes of the aqueous ozone concentration during the reaction period in seawater with or without the dispersant. Within 120 min, the ozone concentration was decreased from 0.11 mg/L in dispersant-free seawater to 0.07 mg/L in the 18 mg/L dispersant solution and further to 0.03 mg/L in the 180 mg/L dispersant solution. In this study, the inhibitive effects outweighed the promoting effects.
+
+In addition, at the dispersant concentration of 180 mg/L, which is much higher than the CMC value (22.5 mg/L), the dispersant can further inhibit the ozone degradation through the "cage effect", i.e., the accumulation of pyrene in the micelles, which inhibit access of ozone and free radicals to pyrene due to elevated mass transfer resistance (Chu and Jia, 2009; Zhang et al., 2011).
+
+### Influence of reaction conditions on ozonation of pyrene in dispersant solution
+
+The ozone degradation kinetics of pyrene in the presence of 18 mg/L of Corexit EC9500A was tested as a function of dissolved ozone concentration, pH, IS, and temperature. The first-order rate constants of \(k_{1}\) and \(k_{2}\) are listed in Table 2.
+
+#### 3.2.1 Effects of ozone concentration
+
+Fig. 4 shows the pyrene degradation kinetic data at various concentrations of ozone. As the aqueous ozone concentration increased from 0.07 to 0.87 mg/L, the rate constant \(k_{1}\) increased from 0.0090 to 0.0637 min\({}^{-1}\) (a factor of 6), and \(k_{2}\) boosted from 0.0315 to 0.290 min\({}^{-1}\) (by 8 times), respectively. Apparently, the pyrene oxidation rate is proportional to the aqueous ozone concentration even though 18 mg/L of the dispersant was competing for the reactive species.
+
+#### 3.2.2 Effects of pH
+
+pH has a significant influence on the ozonation of pyrene in the dispersant solutions. Fig. 5 presents effects of pH on ozone degradation of pyrene. The rate constant \(k_{1}\) increased from 0.0637 to 0.117 min\({}^{-1}\) (an 83% increase), and \(k_{2}\) increased from 0.290 to 0.779 min\({}^{-1}\) (by 1.68 times) as the pH decreased from 8.5 to 5.0.
+
+Solution pH significantly influences ozone decomposition in water. As pH increased, ozone decomposition occurs via the following five-step chain reactions as shown in Eqns. (2)-(6) (Kasprzyk-Hordern et al., 2003):
+
+\[O_{3}+H_{2}O\to 2H\Odegradation of pyrene. However, the gain in the indirect ozonation is at the expense of a loss in direct ozonation. Based on our observations in this case, it was estimated that the loss in direct ozonation outweighed the gain in the indirect reactions, resulting in a net decrease in the overall pyrene degradation rate at the higher pH. Further work is needed to more accurately quantify the relative contributions of direct and indirect ozonation to pyrene degradation. Different from our findings, Beltran et al. (1995) investigated effects of pH on oxidation rate of fluorene by ozone and found that the increase of pH led to an increase in the oxidation rate. This inconsistency may be attributed to the dispersant which might compete for the radicals more fiercely than for the ozone.
+
+#### Effects of IS
+
+Fig. 6 shows that increasing IS from 0.01 to 0.70 M did not significantly affect the degradation rate of pyrene in the dispersant solution (for \(k_{1}\): \(p=0.414\), and for \(k_{2}\): \(p=0.748\) at the 0.05 level of significance).
+
+At elevated IS, NaCl can compete for dissolved ozone and radicals with pyrene, resulting in a slower degradation rate. In the experimental pH of 8.5, the following reactions can take place (Muthukumar and Selvakumar, 2004):
+
+\[\begin{equation*}{\mathrm{{\it NaCl}}}{\mathrm{{\it\leftrightarrow}}}{\mathrm{{ \it Na}}}^{+}+{\mathrm{{\it Cl}}}^{-}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{{\it Cl}}}^{-}+{\mathrm{{\it OH}}}^{\bullet}{\mathrm{{\it \leftrightarrow}}}{\mathrm{{\it Cl}}}^{\bullet}+{\mathrm{{\it OH}}}^{-}\end{equation*}\]
+
+\[\begin{equation*}{\mathrm{{\it Cl}}}^{-}+{\mathrm{{\it Cl}}}^{\bullet}{\mathrm{{\it \leftrightarrow}}}{\mathrm{{\it Cl}}}_{2}+{\mathrm{{\it e}}}^{-}\end{equation*}\]
+
+In addition, elevated IS can increase the interfacial concentration of pyrene at the ozone-solution interface, resulting in a faster degradation rate.
+
+In the presence of the dispersant, some of the lighter and less-soluble components in the dispersant (e.g., Span 80) may accumulate more PAH to the top layer of the water column, which is conducive to the contact and reactions with ozone in the headspace air. However, the dispersant also tends to offset the salting out effect due to lowered surface tension and enhanced solubilization of the PAH. Moreover, elevated IS may push more pyrene to the dispersant/solvent cages. In this study, all these contrasting effects of IS in the presence of the dispersant appeared to have mutually offset. As a result, the overall effect on the reaction rate was insignificant.
+
+#### Effects of temperature
+
+The effect of temperature on the oxidation rate of pyrene by ozone was investigated at pH 8.5. Fig. 7 shows that decreasing temperature from 22 to 10 degC enhanced the rate constant \(k_{1}\) from
+
+Fig. 5: Effects of pH on ozone degradation of pyrene in 18 mg/l. dispersant solution. Data plotted as mean of duplicates and the error bars (calculated as standard error) indicate data reproducibility.
+
+Fig. 6: Effects of IS on ozone degradation of pyrene in 18 mg/l. dispersant solution. Data plotted as mean of duplicates and the error bars (calculated as standard error) indicate data reproducibility.
+
+0.0637 to 0.0893 min-1 (_p_ = 0.003), and \(k\)2 from 0.290 to 0.643 min-1 (_p_ = 0.001), respectively.
+
+Higher temperature accelerates the molecular collision rate, increases the reaction rate constant and the volumetric mass transfer coefficient, accelerating the degradation rate (Zhao et al., 2004), however, it decreased the solubility of ozone, and thereby, inhibited the production of radicals (Beltran et al., 1995). Zhao et al. (2004) reported a modest decrease in the conversion of cationic red X-GRL dye from 88, 86, to 85% as the temperature increased from 15, 20 to 25 degC. The much enlarged effect observed in our study can be attributed to the dispersant effects: (1) competition of the dispersant with pyrene for free radicals and ozone, and (2) interactions between pyrene and the dispersant.
+
+## Conclusions
+
+This study investigated effects of a stereotype oil dispersant Corexit EC9500A on ozone oxidation of phenanthrene/pyrene in the seawater. The primary findings are summarized as follows:
+
+1. Ozone can effectively degrade phenanthrene and pyrene with or without the dispersant. In the dispersant solutions, the ozone degradation followed a two-stage first order kinetics, i.e., a slower initial degradation rate followed by a much faster rate thereafter. Pyrene was more easily degraded than phenanthrene based on the reaction rate constants.
+2. The presence of 18 and 180 mg/L of the dispersant inhibited the phenanthrene and pyrene degradation. The rate constant \(k\)1 of phenanthrene was suppressed by 32% and 75%, and \(k\)2 by 38% and 80%, respectively, at 18 and 180 mg/L of the dispersant. Similarly, \(k\)1 of pyrene was suppressed by 68% and 85%, and \(k\)2 was reduced by 51% and 79%, respectively.
+3. The pyrene degradation rate constants (_k_1 and \(k\)2) in the 18 mg/L dispersant solution increased by 6 and 8 times when the aqueous ozone concentration was raised from 0.07 to 0.87 mg/L. Decreasing pH from 8.5 to 5.0 increased \(k\)1 value by 83% and \(k\)2 by 1.68 times. Decreasing temperature from 22 to 10 degC enhanced degradation rate constants of \(k\)1 and \(k\)2 by 40% and 1.2 times, respectively. The effect of IS in the range of 0.01-0.70 M was insignificant.
+
+The results provide useful information for understanding the roles of oil dispersants on environmental weathering/degradation and fate of persistent oil components in natural and engineered systems.
+
+## Acknowledgements
+
+The research was partially supported by MESC BP-GRI and the U.S. Department of the Interior Bureau of Ocean Energy Management.
+
+## References
+
+- Amat et al. (2007) Amat, A.M., Arques, A., Miranda, M.A., Vincente, R., Segui, S., 2007. Degradation of two commercial anionic surfactants by means of ozone and/or UV irradiation. Environ. Eng. Sci. 24, 790 -794.
+- Bader and Hoigne (1981) Bader, H., Hoigne, J., 1981. Determination of ozone in water by the indigo method. Water Res. 15, 449 -456.
+- Baumard et al. (1998) Baumard, P., Budinski, H., Carrigues, P., Sorbe, J.C., Burgeot, T., Bellocq, J., 1998. Concentrations of PAHs (polycyclic aromatic hydrocarbons) in various marine organisms in relation to those in sediments and to trophic level. Mar. Pollut. Bull. 36, 951 -960.
+- Beltran et al. (1995) Beltran, F.J., Oveiero, G., Encinar, JM, Rivas, J., 1995. Oxidation of polynuclear aromatic hydrocarbons in water. I. 2.0. Isolation, Inf. Eng. Chem. Res. 34, 1596 -1606.
+- Brosius et al. (2009) Brosius, R., Vincente, S., Abouffald, K., Daneshvar, A., Sauve, S., Barbeau, B., Prevost, M., 2009. Ozone oxidation of pharmaceuticals, endocrine disruptors and pesticides during drinking water treatment. Water Res. 43, 470 -471.
+- Cheline-Ayala et al. (2011) Cheline-Ayala, P., El-Min, M.G., Smith, D.W., Adams, C.D., 2011. Oxidation kinetics of two pesticides in natural waters by agroantigen and ozone combined with hydrogen peroxide. Water Res. 45, 2517 -2526.
+- Chu et al. (2006) Chu, W., Chan, K.H., Graham, N.J.D., 2006. Enhancement of ozone oxidation and its associated processes in the presence of surfactant: degradation of atrazine. Chemosphere 64, 931 -936.
+- Chu et al. (2009) Chu, W., Jia, J., 2009. The photodegradation and modeling of a typical NAPI, trifluorolecthene, by monochromatic UV irradiations. Environ. Sci. Technol. 43, 1455 -1459. -
+
+Fig. 7: Effects of temperature on ozone degradation of pyrene in 18 mg/L dispersant solution. Data plotted as mean of duplicates and the error bars (calculated as standard error) indicate data reproducibility.
+
+* Corless et al. (1990) Corless, C.E., Reynolds, G.L., Graham, N.J.D., Perry, R., 1990. Cozonation of pyrene in aqueous solution. Nat. Res. 24, 1119-1123.
+* Dururia et al. (2009) Dururia, M., Emanuele, L., Racioppi, R., Veluzzi, V., 2009. Photochemical degradation of crude oil: comparison between direct irradiation, photocatalysis, and photocatalysis on zeolite. J. Hazard. Mater. 164, 32-38.
+* EPA (2015) EPA (2015) Monitor values report. Retrieval of Juaya 2315, from [http://www.epa.gov/jartata/j.erp.mon.html](http://www.epa.gov/jartata/j.erp.mon.html).
+* Kong et al. (2015) Kong, Y., Fu, J., O'Reilly, S.E., Zhao, D., 2015. Effects of oil dispersants on photodegradation of pyrene in marine water. J. Hazard. Mater. 287, 142-150.
+* Cong et al. (2014) Cong, Y., Zhao, X., Cai, Z., O'Reilly, S.E., Hao, X., Zhao, D., 2014. A review of oil, dispersed oil and sediment interactions in the aquatic environment: influence on the fate, transport and remediation of oil spills. Nat. Pollut. 79, 16-33.
+* Cong et al. (2014) Cong, Y., Zhao, X., O'Reilly, S.E., Qian, J., Zhao, D., 2014. Effects of oil dispersant and oil on sorption and desorption of phenanthrene with Gulf Coast marine sediments. Environ. Pollut. 185, 240-249.
+* Hekata and El-Din (2004) Hekata, K., El-Din, M.G., 2004. Degradation of recalcitrant surfactants in wastewater by ozonation and advances oxidation processes: a review. Ozone-Sci. Eng. 26, 327-343.
+* Kasprzyk-Hordern et al. (2003) Kasprzyk-Hordern, B., Zlolek, M., Nawrocki, J., 2003. Catalytic ozonation and methods of enhancing molecular ozone reactions in water treatment. Appl. Catal. E-Environment 46, 639-669.
+* Kajivninski et al. (2014) Kajivninski, E.B., Kido Soule, M.C., Valentine, D.L., Boysen, A.K., Longnecker, K., Redmond, M.C., 2014. Fate of dispersants associated with the Deepwater Horizon oil spill. Environ. Sci. Technol. 45, 1298-1306.
+* Lin et al. (2014) Lin, C., Zhang, W., Yuan, M., Peng, C., Ren, Y., Wei, C., 2014. Degradation of polycyclic aromatic hydrocarbons in a coding wastewater treatment plant residual by an O3/ultraviolet fluidized bead reactor. Environ. Sci. Pollut. Res. 21, 20320-10338.
+* Liu et al. (2001) Liu, Q., Schurter, L.M., Muller, C.A., Moisio, S., Francisco, J.S., Margerun, D.W., 2001. Kinetics and mechanisms of aqueous ozone reactions with bromide, sulfite, hydrogen sulfide, iodide, and nitrite loss. Inorg. Chem. 04, 4436-4442.
+* Liu et al. (2014) Liu, Y., Li, X.L., Liu, J., Shi, C., Zhu, A., 2014. Ozone catalytic oxidation of benzene over AgMn/H2SM-5 catalysts at room temperature: effects of Mn loading and water content. Chin. J. Catal. 33, 1465-1474.
+* Liu et al. (2012) Liu, Z., Liu, J., Zhu, Q., Wu, W.W., 2012. The weathering of oil after the Deepwater Horizon oil spill: insights from the chemical composition of the oil from the sea surface, salt analyses and sediments. Environ. Res. Lett. 7, 035302.
+* Marquez et al. (2014) Marquez, G., Rodriguez, E.M., Maldonado, M.I., Alvarez, P.M., 2014. Integration of ozone and solar TiO2-photocatalytic oxidation for the degradation of selected pharmaceutical compounds in water and wastewater. Separ. Purif. Technol. 136, 18-26.
+* Masten and Davies (1994) Masten, S.J., Davies, S.H., 1994. The use of ozonation to degrade organic contaminants in wastewaters. Environ. Sci. Technol. 28, 180A-185A.
+* Muthukumar and Selvakumar (2004) Muthukumar, M., Selvakumar, N., 2004. Studies on the effect of inorganic salts on decoloration of acid type effluents by ozonation. Drugs presents 621, 221-228.
+* Nam et al. (2008) Nam, J.I., Thomas, O.G., Jaward, F.M., Steinnes, E., Gustafsson, O., Jones, K.C., 2008. PAHs in background soils from Western Europe: influence of atmospheric deposition and soil organic matter. Chemosphere 70, 1596-1602.
+* Ning et al. (2015) Ning, X., Shen, I., Sun, J., Lin, C., Zhang, Y., Yang, Z., Chen, S., 2015. Degradation of polycyclic aromatic hydrocarbons (PMs) in textile dye sludge by o3H4O2 treatment. RSC Adv. 5, 38021-38029.
+* Reddy et al. (2012) Reddy, C.M., Arey, J.S., Sewald, J.S., Syba, S.P., Lemkau, K.L., Nelson, R.K., Carmichael, C.A., McIntyre, C.P., Fenwick, J., Ventura, G.T., Van Moo, B.A., Camildi, R., 2012. Science Applications in the Deepwater Horizon Oil spill Special Feature: Composition and fate of Gas and Oil Release to the Water Column during the Deepwater Horizon Oil spill. PNS. 101242108.
+* Ryerson et al. (2011) Ryerson, T.B., Alkin, K.C., Aseguine, W.M., Atlas, E., Blake, D.R., Brock, C.A., Fehsenfeld, F.C., Gao, R.S., de Gouy, J.A., Fahey, D.W., Holloway, J.S., Luck, D.A., Iueb, R.A., Meinardi, S., Middledewick, A.M., Murphy, D.M., Neumann, J.A., Nowak, J.B., Parrish, D.D., Pestisch, J., Perring, A.E., Pollack, I.B., Ravisharma, A.R., Roberts, J.M., Schwarz, J.P., Spackman, J.R., Stark, H., Warneke, C., Watts, L.A., 2011. Atmospheric emissions from the Deepwater Horizon spill constrain air-water partitioning. Hydrocarbon fate, and leak rate. Geophys. Res. Lett. 38, L07803.
+* Sammatro et al. (2013) Sammatro, P.W., Kolian, S.R., Warby, R.A.F., Boudlin, J.L., Suhra, W.A., Porter, S.A., 2013. Distribution and concentrations of petroleum hydrocarbons associated with the BP/Deepwater Horizon oil spill, Gulf of Mexico. Mar. Pollut. Bull. 73, 129-143.
+* Scello and Tejerdema (1991) Scello, G.M., Tejerdema, R.S., 1991. A simple method for determination of Corexit 9527 in natural waters. Mar. Environ. Res. 31, 69-78.
+* Staehleith (1985) Staehleith, J., Hogue, I., 1985. Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. Environ. Sci. Technol. 110, 1206-1213.
+* Yang et al. (2005) Yang, G., Zhao, Y., Lu, X., Gao, X., 2005. Adsorption of methanol on marine sediments. Colloid. Surf. A 264, 179-186.
+* Yao et al. (2018) Yao, J., Huang, Z., Masten, S.J., 1998. The ozonation of pyrene: pathway and product identification. Water Res. 32, 3001-3012.
+* Zhang et al. (2011) Zhang, Y., Wong, J.W.C., Liu, P., Yuan, M., 2011. Heterogeneous photocatalytic degradation of phenanthrene in surfactant solution containing TiO2 particles. J. Hazard. Mater. 191, 136-143.
+* Zhao et al. (2011) Zhao, W., Liao, Q., Zhang, J., Yang, Y., Dai, J., Zhao, D., 2011. Oxidation of cationic Red X-GRI by ozonation combined with UV radiation in aqueous solution: degradation, kinetics, and modeling. Chem. Eng. J. 171, 628-639.
+* Zhao et al. (2004) Zhao, W., Shi, H., Wang, D., 2004. Ozonation of Cationic Red X-GRI in aqueous solution: degradation and mechanism. Chemosphere 57, 1189-1199.

manual_annotation/200pdfs_in_mmd/10.1016_j.chemosphere.2017.07.040.mmd

@@ -0,0 +1,374 @@
+## Accepted Manuscript
+
+Ozone pretreatment of process waste water generated in course of fluoroquinolone production
+
+Fares Daoud, David Pelzer, Sebastian Zuehlke, Michael Spiteller, Oliver Kayser
+
+PII: S0045-6535(17)31083-4
+
+DOI: 10.1016/j.chemosphere.2017.07.040
+
+Reference: CHEM 19573
+
+To appear in: _ECSN_
+
+Received Date: 6 December 2016
+
+Revised Date: 23 June 2017
+
+Accepted Date: 9 July 2017
+
+Please cite this article as: Daoud, F., Pelzer, D., Zuehlke, S., Spiteller, M., Kayser, O., Ozone pretreatment of process waste water generated in course of fluoroquinolone production, _Chemosphere_ (2017), doi: 10.1016/j.chemosphere.2017.07.040.
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+**Graphical abstract**
+
+## 1 Introduction
+
+The _n_-body problem is the problem of the _n_-body problem* [41] LC, liquid chromatography; HR, high resolution; MS, mass spectrometry; TOC, total organic carbon; MW, molecular weight; WWTP, waste water treatment plant; HESI, heated electrospray ionization
+* [42]**Highlights**
+* [43] Ozonation of original waste water from production of ciprofloxacin and moxifloxacin
+* [44] Waste water at fluoroquinolones saturation
+* [45] Identification of new fluoroquinolone transformation products
+* [46]
+* [1] **1. Introduction**
+* [2] **Antibiotics, such as fluoroquinolones, are essential drugs to cure human beings and animals that suffer from bacterial infections with severe consequences for their health. The first synthetic quinolone antibiotic ever developed was nalidixic acid (Lesher et al., 1962). While it was found to possess antibacterial property, its activity against Gram-negative microorganisms has been proved but was limited (Blondeau, 2004). Structurally, quinolones are characterized by the presence of an exocyclic oxygen and a carboxyl group at the naphthydridine nucleus (Appelbaum et al., 2000). Improved activity against Gram-negative organisms was achieved through addition of piperazine, while the development of third generation quinolones involved also a fluorination. The newly established group of fluoroquinolones attained a broader spectrum of activity and showed enhanced effectiveness against Gram-positive bacteria (Appelbaum et al., 2000). Fluoroquinolones belong to the family of gyrase inhibitors. They irreversibly inhibit microbial topoisomerase II (gyrase A subunit), an enzyme indispensable for the replication of DNA (Blondeau, 2004).**
+* [3] **Bayer launched their first fluoroquinolone ciprofloxacin in 1981, followed by the second generation drug, moxifloxacin (Appelbaum et al., 2000). Both antibiotics are composed of the quinolone-core carrying diverse moieties at defined locations. The common structural elements are the cyclopropyl substituent, the essential carboxylic group, the exocyclic oxygen and the fluorine atom, while specific substituents, the piperazine moiety for ciprofloxacin and the pyrrolo pyridine ring system in the case of moxifloxacin constitute their differentiating characteristics. The latter antibiotic additionally carries a methoxy group (Figure 1). All fluoroquinolones are chemically synthesized ingredients.**
+* [4] **In the course of production of fluoroquinolones, waste water is generated at several steps of the synthesis. This waste water has to be disposed of, but common WWTP's are not suitable because even**
+* [5] **low concentrations of fluoroquinolones can impair the effectiveness of biological waste water treatment (Dodd et al., 2006). While it is effective against bacterial infections in humans and animals, the biological activity of fluoroquinolones also causes a considerable challenge in biological waste water treatment. Therefore, in industrial countries, fluoroquinolone-containing waste water from production sites is not entering municipal or industrial WWTP's but is subjected to incineration or comparable treatment methods. As incineration of liquids characterized by a low heat of combustion is very expensive and energy-consuming (Zhao et al., 2009), ozonation-based treatment was investigated in this study as an alternative. In the last twenty years, several attempts to remove the concentration of fluoroquinolones via ozonation have been made (Dodd et al., 2006), but all cited reports relied on analysis of either WWTP effluents or trace-level contaminated samples, both with high and low concentrations of fluoroquinolones. The investigated original process waste water discharged directly after the final separation step of the fluoroquinolone production, showed high concentrations close to the saturation point.Incineration has become an effective method for the elimination of process waste water containing significant levels of highly bioactive compound. The objective of this work was to evaluate the potential of ozonation for the effective removal of contaminants from process waste water,* [88] generated in course of industrial production of fluoroquinolones. The ozonation process shall assure an adequate reduction of the main contaminants and all newly formed transformation products with minor burden for the receiving waters. Therefore knowledge about the transformation products and their achievable concentrations as well as the impact on the aquatic biota is required.
+
+[MISSING_PAGE_POST]
+
+123inlet concentration and flow rate of ozone were held constant during ozonation (100-120 g m-3 and 1 L 124min-1, respectively). Ozone concentration in the off-gas from the reactor was measured and its excess collected in a liquid trap before releasing to ambient atmosphere. During ozonation, the inlet, off-gas and dissolved ozone concentrations were continually measured, as were the temperature and pH. Prior to ozonation, the pH of the process water was adjusted to either pH 3, pH 7 and pH 10 through an 128addition of sulphuric acid (H2SO4) or sodium hydroxide (NaOH). pH had been maintained during ozonation by continuous addition of acid or caustic, but no buffer has been used. The maximum ozonation time was 300 min. TOC concentration was measured at defined time intervals. Therefore, liquid samples of approx. 30 ml were manually withdrawn through the sampling port into small 40 ml 132flasks. Thereof, 1.5 ml samples were retrieved for LC-HRMS analysis. Immediately after collection, the samples were flushed for approx. 30 s with nitrogen, in order to remove residual ozone.
+134
+135
+
+### 2.3 Analytical measurements
+
+pH and temperature were monitored by means of the PT100 compensated pH-electrode K100PR (Dr. 137A. Kuntze GmbH, Meerbusch, Germany). The mass flow was determined with the digital mass flow meter D-6300 (M+W Instruments GmbH, Leonhardsbuch, Germany). In gaseous streams, ozone concentration was measured by the ozone analyzers BMT 964 (BMT Messtechnik GmbH,
+140Berlin,Germany), consisting of a double beam UV-photometer at a wavelength of 254 nm. The first ozone analyzer was located in the inlet gas stream, in line with the ozone generator; the second - in the reactor off-gas. Additionally, the concentration of the dissolved ozone was monitored inside the 143reactor by means of the potentiometric double electrode sensor Krypton K System (Dr. A. Kuntze GmbH, Meerbusch, Germany) with AuAu 600-OO-2-1-PG electrodes. All aforementioned data were collected with the Multi-Channel Recorder RSG30 (Endress+Hauser Messtechnik GmbH+Co. KG,
+146Weil am Rhein, Germany) and saved on SD-card, followed by data export to Excel (Microsoft
+147Windows). TOC was determined using the TOC-V\({}_{\text{CPN}}\) analyzer (Shimadzu Deutschland GmbH,
+148Duisburg, Germany) with the measurement range of 0.004-25.000 mg L-1. ASI-V autosampler
+149(Shimadzu Deutschland GmbH, Duisburg, Germany) is for feeding a water sample to TOC.
+150
+
+### 2.4 LC-HRMS measurements
+
+Compounds were identified and quantified by LC-HRMS (LTQ-Orbitrap spectrometer, Thermo
+153Scientific, Waltham, USA). Separation was achieved with a Surveyor-LC HPLC system (Thermo
+154Fisher Scientific, Bremen, Germany) encompassing a quaternary, low-pressure mixing pump with
+155vacuum degassing, an autosampler with a temperature-controlled tray (T = 8 degC), and a column oven
+156(25 degC). Injection volume was 10 ml. For mass spectrometric detection, nitrogen was used as the sheath gas (6 arbitrary units) and helium served as the collision gas. Compound separation was
+158performed on the Nucleodur Gravity C18 column (\(3\times 50\) mm, 1.8 mm, Macherey-Nagel, Dueren, Germany) with the following solvent system: water (+ 0.5% formic acid) (A)/methanol (+ 0.1%
+* [160] formic acid) (B) (flow rate, 0.3 mL min-1). Samples were analyzed using the following gradient
+* [161] program: 85% A isocratic for 3 min, linear gradient to 50% B within 9 min, to 100% B within 0.5 min
+* [162] and held for 11 min; the system returned to its initial conditions (85% A) within 0.5 min and was
+* [163] equilibrated for 7 min. The spectrometer was operated in HESI positive mode (1 spectrum/s; mass
+* [164] range, 100-650) with the nominal mass resolving power of 60000 at m/z 400. Automatic gain control
+* [165] was used to provide high-accuracy mass measurements within 2 ppm deviation using one internal lock
+* [166] mass, m/z 391.284290; bis-(2-ethylhexyl)-phthalate. External calibration was performed at the
+* [167] concentration levels of 0.5, 2, 10, 50, 200, 1000, and 5000 ng mL-1 in water:methanol (v,v, 80:20).
+* [168] Calibration curves were generated using the reference standards available at Bayer Corp. Sum
+* [169] formulae were derived by comparison of measured and theoretical mass-to-charge ratios (m/z) within
+* the arrangement of atoms and chemical
+* [171] bonding, tandem and triple-stage mass spectrometry (MS-MS and MSa) experiments were performed
+* [172] by Pelzer (2015). Collision induced dissociation (CID) was used for fragmentation, helium served as
+* [173] collision gas. Unknown components could be identified by comparison of fragmentation patterns with
+* [174] fragmentation of the reference standards and pure substances (see Supporting Information E and F).
+* [175]
+* [176] **2.5 Ecotoxicological parameters**
+* [177] Ozonated process water cannot be released to receiving water bodies without additional biological
+* [178] treatment. In order to evaluate its remaining toxicity, ecotoxicological parameters as well as the
+* [179] decrease in TOC content (Zahn-Wellens testing, DIN EN 9888) were measured. Therefore, ozonated
+* [180] waste water prior and after biological treatment has been tested. In order to simulate realistic
+* [181] conditions for biological treatment, tests have been conducted in an aerated reactor with originally
+* [182] activated sludge from the connected WWTP. The experiments were performed at a laboratory certified
+* [183] for ecotoxicological evaluation (Gobio GmbH, Aarbergen, Germany). In order to analyze the toxicity
+* [184] for aquatic biota, tests with three trophic levels have been conducted. For safe disposal without impact
+* [185] on the aquatic biota, the minimum dilution factor (G) for the drainage of process water into receiving
+* [186] waters had to be determined. Because bacteria, algae and invertebrates react differently on
+* [187] contaminants all trophic levels have to be taken into account, thereof the highest dilution factor has to
+* [188] be considered for safe disposal. The evaluation of dilution factors (G) for bacteria (GL), algae (GA)
+* [189] and invertebrates (GD) has been performed according to DIN EN 38412 L30, L33 and L34. Due to the
+* [190] fact that the Zahn-Wellens test has to be performed with a maximum TOC of 400 mg L-1, the
+* [191] investigated process water was appropriately diluted.
+* [192]
+* [193] **3. Results and discussion**
+* [194] **3.1 Ozonation**
+* [195] As the degree of fluoroquinolone degradation and the generation of their transformation products
+* [196] significantly depend on the pH, the ozonation experiments were performed under acidic, neutral and * [197] basic conditions (pH 3, 7 and 10). The ozone consumption was calculated according to Gottschalk et al. (2010). For ciprofloxacin, the highest ozone consumption rates were measured in neutral and basic reaction environments, while under acidic conditions the obtained values were 3-4 times lower (see Supporting Information B.1). The curves display over the 300 min ozonation time a straight line, which proves a constant reaction of ozone with available reactants. For moxifloxacin, the highest ozone consumption was registered under basic conditions (see Supporting Information B.2). The curves demonstrate under neutral and acidic conditions straight lines, but under basic conditions, the curve converges to a limiting curve which is an indication of a decelerating ozone reaction. In both cases, the obtained results can be attributed to the increased generation of hydroxyl radicals under basic conditions (Hoigne and Staehelin, 1982). These results prove that best ozonation results could be achieved under basic conditions. Based on these results further MS-experiments have been limited to the transformation products generated under basic conditions.
+* [209]
+* [210] **3.2 Mass spectrometric identification and quantitation**
+* [211] To check the variety of transformation products, samples after 300 min ozonation at all tested pH
+* [212] values were analyzed. Full degradation kinetics was studied in detail for conditions with the best degradation of fluoroquinolones. For ciprofloxacin (initial concentration 3244 mg L-1), the degradation
+* [214] was most pronounced under basic conditions (pH 10), reaching \(>\) 99.8%, with diminishing values
+* [215] recorded in alternative reaction environments (98.8% at pH 7 and 73.9% at pH 3). Degradation of
+* [216] moxifloxacin (initial concentration 5637 mg L-1) was so fast, that detection limit (1ng mL-1) was
+* [217] reached within defined ozonation time for all reaction conditions. Subsequent, in-depth analyzes were
+* [218] focused on ozonation under basic conditions.
+* [219] Based on available reference standards, calibration curves were delineated for quantification of all
+* [220] detected components. Although reference samples were accessible for only a few relevant compounds,
+* [221] concentration values of other transformation products were calculated assuming comparable ionization
+* [222] to that of parent antibiotic in full scan mass spectrometric detection. Relevant target compounds were
+* [223] quantified after distinct time of ozonation (0, 15, 30, 60, 120, 180, 300 min) and the obtained values
+* [224] were used for determination of degradation kinetics and calculation of rate constants. Verification of
+* [225] the structure of the transformation product was done by means of MS-MS and MS3.
+* [226] Summing up the content of Ciprofloxacin together with all other by-products and transformation
+* [227] products resulted in approx. 7200 mg L-1 at time 0 min. The total concentration kept constant for the
+* [228] first 120 min and then dropped down to 2600 mg L-1 at 300 min ozonation time. Transformation
+* [229] products with lower molecular weight were expected, but a summation of all transformation products
+* [230] should stay stable. Nevertheless reduction of all compounds was science-based on the fact, that small
+* [231] transformation products below 100 g/mol have not been identified by the used analytics. Additionally
+* [232] partly mineralization could have occurred which resulted in a carryover of components with low vapor * [233] pressure into the gaseous phase. Same was observed with the TOC which dropped from 28200 mg L-1 down to 18100 mg L-1 (see Supporting Information D).
+* [234] Summing up the content of Mooxifloxacin together with all other by-products and transformation products resulted in approx. 5675 mg L-1 at time 0 min. The total concentration of mooxifloxacin and all adjacent transformation products decreased rapidly. After 180 min all transformation products were below detection limit. Most probably all transformation products were degraded to small molecules (< 100 g mol-1) and partly exited the reactor via offgas. During ozonation, the corresponding TOC showed a decrease to almost 50% (see Supporting Information D). Figures 2 and 5 show mass spectra extracted from the total ion chromatograms of ciprofloxacin and moxifloxacin after 0 and 300 min of ozonation, as well as the sum formulae of compounds corresponding to the most prominent signals.
+* [244] _Ciprofloxacin_
+* Prior to ozone treatment, the process waste water samples contained, in addition to ciprofloxacin (m/z 246 332.1405), its decarboxylated form (m/z 288.1506) and a chloroquineonline (vs. fluoroquinolone) derivative (m/z 348.1108) (Figure 2). Post-ozonation, all main constituents of the original process water sample were almost completely degraded, while a great number of minor components could be determined (some are highlighted).
+
+Figure 2: Mass spectrum (extracted from total ion chromatogram) of ciprofloxacin process water before ozonation (**A**) and mass spectrum (extracted from total ion chromatogram) of ciprofloxacin process water after 300 min of ozonation (**B**) (*impurity from the LC-MS-system)
+
+## 6 Conclusion
+
+The exact masses, sum formulae and retention times of 22 transformation products were fully elucidated by LC-HRMS (see Supporting Information G.1).
+
+Various authors had worked on potential transformation products in the last couple of years. To our knowledge, no data on direct ozone pre-treatment of process waste water have been published to date.
+
+However, results of several reports provide valuable insights concerning potential fluoroquinolone transformation products. Other investigations focused on degradation via photolysis (Burhenne et al.,1997; van Doorslaer et al., 2011; Calza et al., 2008; Tuerk et al., 2012; Burhenne et al., 1997; Kabasci, 2007), ozonation (de Witte et al., 2008; Heynderickx et al., 2011; Kabasci, 2007; Kovalova et al., 2013; Rodriguez et al., 2008), or other advanced oxidation processes for the generation of OH radicals (Tuerk, 2006; Wang et al., 2010; de Witte et al., 2009; Rodriguez et al., 2008). Degradation of fluoroquinolones has been recently reviewed by Sukul and Spiteller (2007).
+
+Thus, most of the postulated structures (and equivalent molecular masses) have been identified in previous research reports, using various analytical methods. Three of the identified transformation products were available as reference substances within this study. These molecules were, therefore, unambiguously identified by their exact mass, fragmentation pattern and retention time, as compared to the standards. Their calibration curves were further established. Most of the other detected transformation products were identified based on their mass spectrometric fragmentation. Four compounds were reported for the first time, others compared with previously postulated structures.
+
+Figure 3 shows the decrease in concentration of ciprofloxacin within 300 min, from 3244 to 61 mg L-1 (98.1%). Concurrent to the antibiotic depletion, levels of several intermediates increased during ozonation. Initial and final concentrations of occurred reaction by-products as well as transformation products have been quantified based on available reference standards. Within 300 min of ozonation time the chloroquine compound started at 1981 mg L-1 and degraded down to 7.3 mg mg L-1 (99.6%), the ethylene diamine compound started at 88 mg L-1, increased up to 853 mg L-1 (120 min) and declined to 153 mg L-1. The defluorinated compound started at 188 mg L-1 and decreased below detection limit. Most of the 22 transformation products were quantified at concentrations below 1 % of initial ciprofloxacin content and are, therefore, not shown in Figure 3.
+
+Ciprofloxacin degradation occurred at various functional moieties of the molecule (Figure 4). The piperazinyl substituent and the carboxyl group were most prone to disintegration. Additionally, defluorination at C-6 was observed, C17H19N3O3 (MW313), which is one of the known side-products of the synthesis of the antibiotic. Further, loss of the propyl moiety at N-1 leads to C14H14FN3O3 (MW291), a transformation product generated also by a combination of ozone and UV light (Tuerk et al., 2012). Three different products were generated from the parent molecule via oxidation of the piperazinyl substituent, either through its N-formylation (C18H18FN3O4, MW359), hydroxylation, (C17H18FN3O4, MW347) or addition of an exocyclic oxygen to the ring system, C17H16FN3O4 (MW345). In the two latter cases, localization of oxygen binding could not be determined by means of the applied analytical methods. MW359 was also induced by ozone/UV (Tuerk et al., 2012) or 303 photolytic degradation (Cardoza et al., 2005). MW 347 was previously generated by photolytic reaction (Calza et al., 2008), ozonation (de Witte et al., 2008) or a combination of ozone and UV light (Tuerk et al., 2012). MW345 has been identified by photolytic degradation (Burhenne et al., 1997).
+* While cleavage of the piperazine ring leads to the aldehyde metabolite, C16H16FN3O4 (MW333) its further oxidation generates a compound with two aldehyde groups in the ring system, C17H16FN3O5 (MW361). MW333 has also been detected by photolytic reaction (Calza et al., 2008), ozonation (de Witte et al., 2008), photo-catalytic degradation (Wang et al., 2010) as well as ClO2 induced oxidation (Paul et al., 2010). MW361 was identified by Wang et al. (2010) and Paul et al. (2010). In the consecutive step, the piperazine component is further degraded to an ethylene diamine moiety, C15H16FN3O3 (MW305), a reference standard from Bayer Corp., but also a transformation product that has been detected in the course of several different degradation processes (Burhenne et al., 1997; Calza et al., 2008; de Witte et al., 2008; An et al., 2010; de Witte et al., 2009; Thabaj et al., 2007;
+
+Figure 3: Degradation of ciprofloxacin and formation of its most abundant transformation products (c/c0 > 5 %) in course of waste water ozonation (relative to ciprofloxacin concentration at time 0 min)
+
+* [315] Cardoza et al., 2005; Paul et al., 2010). An et al. (2010) identified comparable transformation products of nonfloxacin, levofloxacin and lemofloxacin.
+* [317] Instead of the aforementioned generation of a double aldehyde within the piperazine moiety, further oxidation of the primary aldehyde to a carboxyl group is possible, C16H16FN3O5 (MW349). De Witte (2008) identified also by ozonation a transformation product with the same chemical formula but proposed its alternative structure, an aldehyde group at a different location of the piperazine moiety.
+* [318] Exact location cannot be verified with currently available fragmentation pattern. Based on the structural formula of MW305, full cleavage of the piperazine moiety leads to a compound with an amino group at C-7, C13H11FN2O3 (MW 262). Also here a broad variety of degradation processes lead to this transformation product (Burhenne et al., 1997; de Witte et al., 2008; Wang et al., 2010; An et al., 2010; Tuerk et al., 2012; de Witte et al., 2009; Cardoza et al., 2005; Paul et al., 2010). MW305 can be further oxidized to C14H11FN2O4 (MW 290), a transformation product previously identified by Paul et al. (2010) by photo-catalysis.
+
+Figure 4: Ozonation pathway of ciprofloxacin. Newly identified degradation products in grey boxes highlighted; highlighted compounds within dotted-line boxes have been published previously (de Witte et al., 2008; de Witte et al., 2009) as molecules defined by the same sum formulae but of different chemical structure
+
+* [333] As indicated before, a decarboxylated compound, C\({}_{16}\)H\({}_{18}\)FN\({}_{3}\)O (MW287) was also identified. Its further degradation entails either a cleavage of the piperazine moiety or its oxidation. The latter leads to a molecule comparable to MW359, containing an N-formylated piperazine substituent but no carboxyl group, C\({}_{17}\)H\({}_{18}\)FN\({}_{3}\)O\({}_{2}\) (MW315). Also, in this case, de Witte (2009) identified by ozonation a transformation product with the equivalent chemical formula but of disparate structural formula.
+* [334] Again, fragmentation pattern does not offer valuable clues to the location of the aldehyde group.
+* a molecule with a pyrrol moiety and two exocyclic oxygen atoms. In this case the fragmentation pattern (see
+* [336] Supporting Information E.4) supports the assumed chemical structure, because the C\({}_{2}\)H\({}_{5}\)N-fragment
+* [337] has only been discovered in conjunction with a carbon monoxide fragment. Full degradation of the piperazine moiety, on the other hand, resulted in a molecule with an amino group at C-7, C\({}_{12}\)H\({}_{11}\)FN\({}_{2}\)O
+* a new transformation product that has not been published before. MW218, as well as MW289, can be further oxidized to two previously unknown molecules, C\({}_{13}\)H\({}_{11}\)FN\({}_{2}\)O\({}_{2}\) (MW246) and C\({}_{16}\)H\({}_{16}\)FN\({}_{3}\)O\({}_{3}\) (MW317). MW246 contains an aldehyde moiety at the C-7 amino group and MW317
+* [339] carries two aldehyde groups at both nitrogen atoms of the ethylene diamine moiety.
+* [340] Furthermore, the concentration of the chlorinated ciprofloxacin metabolite (MW347) declined in course of the ozonation from 1981.0 (in the original waste water sample) to 7.3 mg L\({}^{-1}\) (99.6%), while
+* [341] the decarboxylated metabolite (MW287) content was reduced from 1565.0 to 33.0 mg L\({}^{-1}\) (97.9%).
+* [342] With regard to the fact that decarboxylated ciprofloxacin compounds are products of its ozonation and
+* [343] were already present in the process water at time 0 min, the degree of degradation might have been
+* [344] even higher. Based on the exponential shape of the ciprofloxacin degradation curve pseudo-first-order
+* [345] rate constants can be determined from a plot of ln(c/c\({}_{0}\)) versus time with rate constant k being the
+* [346] slope of the straight line (Prasse et al., 2012). For the degradation of Ciprofloxacin pseudo-first order
+* [347] reaction seemed to be the appropriate assumption indicated by a high linearity factor (R\({}^{2}\)=0.934).
+* [348] Based on the concentration measured at different time intervals, the apparent rate constant for a
+* [349] pseudo-first order reaction (see Supporting Information C.1) was 0.0115 min-1.
+* [350] Nevertheless 4 transformation products have only been identified with respect to their mass formula
+* [351] but MS\({}^{2}\) and MS\({}^{3}\) data were not appropriate to identify the chemical structure.
+* [352] _Moxifloxacin_
+* [353] In case of waste water samples derived from moxifloxacin (m/z 402.1825) production, the
+* [354] decarboxylated form of the antibiotic (m/z 288.1506) could also be detected (Figure 5). It also shows
+* [355] the mass spectrum of the compound of interest after 300 min of ozonation. Similar to ciprofloxacin, ozone treatment of moxifloxacin resulted in detection of a high number of unknown degradation products. The exact masses, sum formulae and retention times of 13 transformation products were fully elucidated by LC-HRMS (see Supporting Information G.2).
+
+Figure 5: Mass spectrum (extracted from total ion chromatogram) of moxifloxacin before ozonation (**A**) and mass spectrum (extracted from total ion chromatogram) of moxifloxacin after 300 min of ozonation (**B**)
+
+* [377] None of the identified transformation products were available as reference substances. Thus, their identity was established through exact mass determination and comparative analyzes of their fragmentation patterns (as obtained in course of MS-MS and MS\({}^{3}\) experiments) and retention time values, relative to known related compounds. All transformation products detected have been identified for the first time.
+* [382] As depicted in Figure 6, the concentration of moxifloxacin decreased within 300 min of ozonation from 5637 to \(<\) 1 mg L-1 (\(>\) 99.9%). Levels of several transformation products rose and declined in course of the waste water ozone treatment. Thereof highest concentrations occurred for MW 292 with a peak of 178 mg L-1 at 15 min and for MW 307 with a maximum of 296 mg L-1 at 30 min ozonation time. After determination of degradation kinetics, based on measurements of individual product levels over time (rate constant for a pseudo-first order reaction established at 0.167 min-1, see Supporting Information C.2), the ozonation pathway of moxifloxacin, as illustrated in Figure 7, was elucidated.
+* [389] Moxifloxacin degradation occurs at different functional moieties of the molecule, as in case of ciprofloxacin.
+* [390] Loss of the carboxyl group at C-3 leads to C20H24FN3O2 (MW357). Another transformation product is generated by cleavage of the pyrrolo piperazine moiety at C-7 (C19H24FN3O4, MW377), while oxidative addition of two hydroxyl groups at the quinolone ring leads to C21H24FN3O6 (MW433).
+
+Figure 6: Degradation of moxifloxacin and formation of its most abundant transformation products (c/c0 > 5%) in course of waste water ozonation (relative to moxifloxacin concentration at time 0 min)
+
+* [402] Further degradation of the pyrrolo piperazine moiety of MW377 can generate C15H13FN2O4 (MW304) and, in a consecutive step, oxidation leads to C14H12FNO7 (MW325) with three additional hydroxyl groups at the quinolone ring.
+* [406] Before mentioned transformation products MW292 and MW307 have been identified with respect to their molecular formula, but MS2 and MS3 data did not give appropriate evidence about the molecular structure. For MW292 the ozonation pathway shows at least the most probable structure.
+* [409]
+* [410] **3.3 Ecotoxicological investigations**
+* [411] _Ciprofloxacin_
+* [412] Ozonated process water generated in course of ciprofloxacin production was characterized by the following EC50 dilution factors: G1L = 40 for bacteria, GA = 24 for algae and GD = 48 for daphnia. The Zahn-Wellens- test indicated TOC degradation of 34.6% after 7 days. Finally, the waste water was analyzed regarding its toxic impact on bacteria, algae and daphnia populations, after being submitted to biological degradation. The respective EC50 dilution factors were: G1L = 18, GA = 9 and GD = 18.
+* [417] The measured values indicated that, for safe disposal into receiving waters, the discharged ozonated water required dilution by a factor of 48, while the further biologically treated waste water had to be
+
+Figure 7: Proposed ozonation pathway of moxifloxacin. All depicted compounds are newly identified degradation products and, to our knowledge, have not been published before. Position of one additional hydroxyl group (framed by dotted lines) may vary for MW433 and MW447
+
+* [419] diluted at least 18 times. That means that toxicity to aquatic biota can be improved by treatment in industrial waste water treatment plants by the factor of approx. 2.5.
+* [420]_Moxifloxacin_
+* [421] The EC50 dilution factors for bacteria, algae and daphnia, characteristic of the investigated
+* [422] moxifloxacin production-derived ozonated process water, were: G\({}_{\text{L}}\) = 192, G\({}_{\text{A}}\) = 128 and G\({}_{\text{D}}\) = 32,
+* [423] respectively. The Zahn-Wellens- test indicated TOC degradation of 86.7% after 7 days. The ozonated
+* [424] waste water was then subjected to biological degradation treatment and, once again, analyzed
+* [425] regarding its toxic impact on bacteria, algae and daphnia. The respective EC50 dilution factors were:
+* [426] G\({}_{\text{L}}\) = 12, G\({}_{\text{A}}\) = 6 and G\({}_{\text{D}}\) = 6. Thus, for safe disposal into receiving waters, the discharged ozonated
+* [427] water had to be diluted 192 times, whereas the biologically treated waste water required dilution with
+* [428] the receiving water by a factor of at least 12, for safe discharge. Toxicity to aquatic biota can be
+* [429] improved by treatment in industrial waste water treatment plants by the factor of 16.
+* [430] Due to residual biological activity of the investigated samples, appropriate dilution of the ozonated
+* [431] waste water was required before transferring to receiving water bodies. Assuming an average
+* [432] industrial WWTP effluent of approx. 3000 m\({}^{\text{3}}\) d\({}^{\text{-1}}\), the established dilution factors were high enough to
+* [433] ascertain safe disposal of the ozonated moxifloxacin-containing process water. While ciprofloxacin
+* [434] was not sufficiently degraded under the process parameters described above, safe disposal of the waste
+* [435] water generated in course of its production could be achieved through prolonged ozonation. This
+* [436] hypothesis, however, requires further verification.
+* [437] Moreover, further ecotoxicological data are needed to ultimately determine the toxicity of
+* [438] ciprofloxacin-containing process water. The residual toxic effects may be attributed to the fact that
+* [439] many metabolites of ciprofloxacin degradation (Figure 4) still encompass the keto-acid moiety of the
+* [430] core structure which is an essential prerequisite for biological activity (Appelbaum et al., 2000). In
+* [431] case of moxifloxacin, the concentration of transformation products is much lower.
+* [432]
+* [433]
+* [434] **4 Conclusions**
+* [435] The current study provides conclusive evidence that ozonation of ciprofloxacin- and moxifloxacin-
+* [436] containing process water can be effectively implemented as a treatment prior to biological degradation
+* [437] procedure. Both fluoroquinolones degraded to a high number of transformation products. The
+* [438] undertaken ozonation experiments resulted in degradation of 98.5% of ciprofloxacin and nearly 100%
+* [439] of moxifloxacin. In case of the former antibiotic, prolonged ozonation (beyond the applied 300 min) is
+* [430] assumed to lead to still higher degree of decomposition, while moxifloxacin content in the analyzed
+* [431] samples dropped below the limit of detection already after 60 min. In both cases, best results were
+* [432] achieved under basic conditions. Numerous degradation products were identified via LC-HRMS, MS-
+* [433] MS and MS\({}^{\text{3}}\). Four transformation products of ciprofloxacin (highlighted compounds in Figure 4) and * [456] all five moxifloxacin transformation products (Figure 7) are reported for the first time. Additionally, the structure of three already published products of ciprofloxacin could be enlightened.
+* [458] All transformation products identified herein show an intact quinolone ring; most, still encompass functional moieties. Therefore, the ozonated process waste will exhibits antibacterial properties [460] (Sukul and Spiteller, 2007), but much less as indicated by the ecotoxicological data. These can be reduced with prolonged ozonation times. Nevertheless, the proposed ozone pre-treatment technology seems to be a valuable alternative for the removal of fluoroquinolones from their production waste water compared to the predominantly implemented incineration.
+* [464]
+* [465] **Appendices**
+* [466] A. Experimental setup, B. Ozone consumption, C. Calculation of rate constants, D. TOC, E. MSn
+* [467] spectra of new identified ciprofloxacin degradation products, F. MSn spectra of new identified
+* [468] moxifloxacin degradation products, G. List of all transformation products
+* [469]
+* [470] **Acknowledgements**
+* [471] F.D. is grateful to the Bayer Healthcare AG for providing a fully equipped laboratory reactor for ozonation trails. M.S. and S.Z. acknowledge the financial assistance of the Ministry of Innovation,
+* [473] Sciences, Research and Technology of the State of North Rhine-Westphalia, Germany and the German
+* [474] Research Foundation (DFG) in purchasing a high-resolution mass spectrometer.
+
+* [477]**References**
+* [478] Akmehmet Balcioglu, I., Oetker, M., 2003. Treatment of pharmaceutical waste water containing antibiotics by O3 and O3/H2O2 processes. Chemosphere 50 (1), 85-95.
+* [480] An, T., Yang, H., Song, W., Li, G., Luo, H., Cooper, W.J., 2010. Mechanistic considerations for the advanced oxidation treatment of fluoroquinolone pharmaceutical compounds using TiO2 heterogeneous catalysis. The Journal of Physical Chemistry A 114, 2569-2575.
+* [483] Andreozzi, R., Campanella, L., Fraysse, B., Garric, J., Gonella, A., Lo Giudice, R., Marotta, R., Pinto, G., Pollio, A., 2004. Effects of advanced oxidation processes (AOPs) on the toxicity of a mixture of pharmaceuticals. Water Science and Technology 50 (5), 23-28.
+* [486] Appelbaum, P.C., Hunter, P.A., 2000. The fluoroquinolone antibacterials: past, present and future perspectives. International Journal of Antimicrobial Agents 16, 5-15.
+* [488] Blondeau, J.M., 2004. Fluoroquinolones: mechanism of action, classification, and development of resistance. Survey of Ophthalmology 49 (2), S73-S78.
+* [490] Burhenne, J., Ludwig, M., Spiteller, M., 1997. Photolytic degradation of fluoroquinolone carboxylic acids in aqueous solution: Part 1. Environmental Science and Pollution Research International 4 (2), 10-15.
+* [491] Burhenne, J., Ludwig, M. Spiteller, M., 1997. Photolytic degradation of fluoroquinolone carboxylic acids in aqueous solution: Part 2. Environmental Science and Pollution Research International 4 (2), 61-67.
+* [492] Calza, P., Medana, C., Carbone, F., Giancotti, V., Baiocchi, C., 2008. Characterization of intermediate compounds formed upon photoinduced degradation of quinolones by high-performance liquid chromatography/high-resolution multiple-stage mass spectrometry. Rapid Communications in Mass Spectrometry 22 (10), 1533-1552.
+* [493] Cardoza, L.A., Knapp, C.W., Larive, C.K., Belden, J.B., Lyday, M., Graham, D.W., 2005. Factors affecting the fate of ciprofloxacin in aquatic field systems. Water, Air and Soil Pollution161 (1-4), 502 383-398.
+* [494] De Witte, B., Dewulf, J., Demeestere, K., van de Vyvere, V., de Wispelaere, P. van Langenhove, H., 2008. Ozonation of ciprofloxacin in water: HRMS identification of reaction products and pathways. Environmental Science and Technology 42 (13), 4889-4895.
+* [495] De Witte, B., Dewulf, J., Demeestere, K., van Langenhove, H., 2009. Ozonation and advanced oxidation by the peroxone process of ciprofloxacin in water. Journal of Hazardous Materials 161 (2-3), 701-708.
+* [506] Dodd, M.C., Buffle, M.O., von Gunten, U.,2006. Oxidation of antibacterial molecules by aqueous ozone: moiety-specific reaction kinetics and application to ozone-based wastewater treatment. Environmental Science and Technology 40 (6), 1969-1977.
+* [512] Gottschalk, C., Libra, J.A., Saupe, A., 2010. Ozonation of water and waste water, page 86-87, 2\({}^{\text{nd}}\) Ed. Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim.
+
+* [514] Heynderickx, P.M., Demeestere, K., Dewulf, J., de Witte, B., van Langenhove, H., 2011. Formal bimolecular kinetic model for the ozonation of ciprofloxacin in the liquid phase. Journal of Advanced Oxidation Technologies 14 (1) 71-80.
+* [517] Kabasci, S., 2007. Oxidative treatment of partial streams of hospital effluents to eliminate persistent, highly active pharmaceuticals, Part 2: scale-up of the process, installation and optimization of a pilot plant (Oxidative Behandlung von Krankenhausabwasser-Teilstroemen zur Beseitigung von persistenten, hochwirksamen Pharmazeutika, Teil 2: Scale-up des Verfahrens, Aufbau und Optimierung einer Demonstrationsanlage). Fraunhofer Institut fuer Umwelt-, Sicherheits- und Energietechnik UMSICHT.
+* [518] Kovalova, L., Siegrist, H., von Gunten, U., Eugster, J., Hagenbuch, M., Wittmer, A., Moser, R., McArdell, C.S., 2013. Elimination of micropollutants during post-treatment of hospital wastewater with powdered activated carbon, ozone, and UV. Environmental Science and Technology 47 (14),7899-7908.
+* [519] Lesher, G.Y., Froelich, E.J., Gruett, M.D., Bailey, J.H., Brundage, R.P., 1962. 1-8-Naphthyridine derivates. A new class of chemotherapeutic agents. Sterling-Winthorp Research Institute, Division of Sterling drug Inc., Rensselaer, New York
+* [520] Paul, T., Dodd, M.C., Strathmann, T.J., 2010. Photolytic and photocatalytic decomposition of aqueous ciprofloxacin: transformation products and residual antibacterial activity. Water Research 44 (10), 532 3121-3132.
+* [521] Pelzer, D., 2015. Identification and purification of ozonation products of 6-fluoroquinolones (Identifizierung und Aufreinigung der Ozonisierungsprodukte von 6-Fluorchinolonen). Master thesis, Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology, Technical University of Dortmund, Germany.
+* [522] Prasse, C., Wagner, M., Schulz, R., Ternes, T.A.,2012. Oxidation of the antiviral drug acyclovir and its biodegradation product carboxy-acyclovir with ozone: kinetics and identification of oxidation products. Environ. Sci. Technol. 46, 2169-2178
+* [523] Rodriguez, A., Rosal, R., Perdigon-Melon, J. A., Mezcua, M., Agueera, A., Hernando, M. D., Leton, P., Fernandez-Alba, A.R., Garcia-Calvo, E., 2008. Ozone-based technologies in water and wastewater treatment. The Handbook of Environmental Chemistry 5 (S/2), 127-175.
+* [524] Staehlin, J., Hoigne, J., 1982. Decomposition of ozone in water: rate of initiation by hydroxide ions and hydroxide peroxide. Environmental Science and Technology 16, 676-682.
+* [525] Sukul, P., Spiteller, M., 2007. Fluorquinolone antibiotics in the environment. Reviews of Environmental Contamination and Toxicology 191, 131-162.
+* [526] Thabaj, K., Kulkarni S., Chimatadar S., Nandibewoor S., 2007. Oxidative transformation of ciprofloxacin by alkaline permanganate : a kinetic and mechanistic study. Polyhedron 26 (17), 549 4877-4885.
+
+* [550] Tuerk, J., 2006. Presentation: Degradation of pharmaceuticals in partial streams of hospital effluents
+* [551] using AOP (Praesentation: Abbau von Pharmazeutika in Krankenhausabwasser-Teilstroemen
+* [552] mittels AOP). Bereich Umweltmedizin-Analytik, Institut fuer Energie-und Umwelttechnik e.V.
+* [553] (IUTA).
+* [554] Tuerk, J., Boergers, A., vom Eyser, C., Richard, J., Dopp, E., Bester, K., 2012. Evaluation and
+* [555] prevention against toxic oxidation by-products in oxidative water treatment of fluoroquinolones
+* [556] (Untersuchungen zur Bewertung und Vermeidung von toxischen Oxidationsnebenprodukten bei
+* [557] der oxidativen Abwasserbehandlung von Fluorquinolonen). Mitt Umweltchem Oekotox18. Jahrg.
+* [558] 2012/Nr.2.
+* [559] Van Doorslaer, X., Demeestere, K., Heynderickx, P.M., van Langenhove, H., Dewulf, J., 2011. UV-A
+* [560] and UV-C induced photolytic and photocatalytic degradation of aqueous ciprofloxacin and
+* [561] moxifloxacin: reaction kinetics and role of adsorption. Applied Catalysis B: Environmental 101,
+* [562] 540-547.
+* [563] Wang, P., He, Y.-L., Huang, C.-H., 2010. Oxidation of fluoroquinolone antibiotics and structurally
+* [564] related amines by chlorine dioxide: reaction kinetics, product and pathway evaluation. Water
+* [565] Research 44 (20), 5989-5998.
+* [566] Zhao, W., van der Voet, E., Huppes, G., Zhang, Y., 2009. Comparative life cycle assessments of
+* [567] incineration and non-incineration treatments for medical waste. Int J Life Cycle Assess, 14:114-121.
+
+* [569]**List of figures**
+* [570]**Figure 1.** Structure of ciprofloxacin (A) and moxifloxacin (B)
+* [571]**Figure 2.** Mass spectrum (extracted from total ion chromatogram) of ciprofloxacin process water
+* [572] before ozonation (A) and mass spectrum (extracted from total ion chromatogram) of ciprofloxacin
+* [573] process water after 300 min of ozonation (B)
+* [574]**Figure 3.** Degradation of ciprofloxacin and formation of its most abundant transformation products
+* [575] (c/c0 > 5 %) in course of waste water ozonation (relative to ciprofloxacin concentration at time 0 min)
+* [576]**Figure 4.** Ozonation pathway of ciprofloxacin. Newly identified degradation products in grey boxes
+* [577] highlighted; highlighted compounds within dotted-line boxes have been published previously (de
+* [578] Witte et al., 2008; de Witte et al., 2009) as molecules defined by the same sum formulae but of
+* [579] different chemical structure
+* [580]**Figure 5.** Mass spectrum (extracted from total ion chromatogram) of moxifloxacin before ozonation
+* [581] (A) and mass spectrum (extracted from total ion chromatogram) of moxifloxacin after 300 min of
+* [582] ozonation (B)
+* [583]**Figure 6.** Degradation of moxifloxacin and formation of its most abundant transformation products
+* [584] (c/c0 > 5%) in course of waste water ozonation (relative to moxifloxacin concentration at time 0 min)
+* [585]**Figure 7.** Proposed ozonation pathway of moxifloxacin. All depicted compounds are newly identified
+* [586] degradation products and, to our knowledge, have not been published before. Position of additional
+* [587] hydroxyl groups (framed by dotted lines) may vary for MW433 and MW447* [588]**Appendices**
+* [589]**Ozone pretreatment of process waste water generated in course of fluoroquinolone production**
+* [590]
+* [591] Fares Daoud\({}^{\text{a}}\), David Pelzer\({}^{\text{b}}\), Sebastian Zuehlke\({}^{\text{b}}\), Michael Spiteller\({}^{\text{b}}\), Oliver Kayser\({}^{\text{e}^{\ast}}\)
+* [592]
+* [593]\({}^{\text{a}}\) Bayer Technology Services GmbH, Friedrich-Ebert-Str. 217-333, 42096 Wuppertal, Germany
+* [594]\({}^{\text{b}}\) Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology,
+* [595] Technical University of Dortmund, Otto-Hahn Str. 6,
+* [596] Germany
+* [597]\({}^{\text{c}}\) Department of Technical Biochemistry, Technical University of Dortmund, Email-Figge-Strasse 66,
+* [598] Germany
+* [600]\({}^{\ast}\)Corresponding author: Oliver Kayser, Department of Technical Biochemistry, Technical University
+* [601] of Dortmund, Emil-Figge-Str. 66, 44227 Dortmund, Germany; tel: +49 231 755-7487, fax: +49 231 602 755-7489, e-mail: Oliver.Kayser@bci.tu-dortmund.de; www.tb.bci.tu-dortmund.de
+* [603]**A. Experimental setup**
+* [604]**A. Experimental setup*** [608]**B. Ozone consumption**
+* [609] **(B.1)**
+* [611]**A(O3): absorbed ozone dose (mg L-1)
+* [612]**Q**G: gaseous mass flow (l/s)
+* [613]**V**L: reaction volume (l)
+* [614]**t: time (s)
+* [615]**c**O**non,0: inlet ozone concentration in the gaseous phase (mg L-1)
+* [616]**c**O**non,e: outlet ozone concentration in the gaseous phase (mg L-1)Figure B.2: Absorbed ozone dose in course of ozonation of moxifloxacin-containing waste water at pH 10, pH 7 and pH 3
+
+* [626]**C. Calculation of rate constants**
+* [627]**Rate constants were calculated according to the following equations:**
+* [628]****\(c\left(M_{{}_{t}}\right)=c\left(M_{{}_{t=0}}\right)\exp^{-k_{{}_{\mathit{app}}}\cdot c\left(O_{{}_{1}}\right)\cdot t}\)****
+* [629]**Simplification:**
+* [630]****\(c\left(M_{{}_{t}}\right)=c\left(M_{{}_{t=0}}\right)\exp^{-k_{{}_{\mathit{app}}}\cdot t}\)****
+* [631]****\(\ln\frac{c\left(M_{{}_{t}}\right)}{c\left(M_{{}_{t=0}}\right)}=k_{{}_{\mathit{app}}}\cdot t\)****
+* [632]****\(c\left(M_{{}_{t=0}}\right)=c\left(M_{{}_{t=0}}\right)\exp^{-k_{{}_{\mathit{app}}}\cdot t}\)****
+* [633]****\(c\left(M_{{}_{t=0}}\right)=k_{{}_{\mathit{app}}}\cdot t\)****
+
+[MISSING_PAGE_POST]
+
+* [645] TOC decreased to about 60% under basic conditions and to 80-90 % under neutral/ acidic conditions for both investigated process waste waters. The incipiently increase in TOC during the first 60 min is founded in the initial high concentration of the target compound and the associated by-products, resulting in partial precipitation. These precipitated compounds cannot be detected with the current TOC measurement.
+
+Figure E.1: MS\({}^{\text{a}}\) spectrum of MW218
+
+Figure E.2: MS\({}^{\text{a}}\) spectrum of MW246
+
+Figure E.3: MS\({}^{3}\) spectrum of MW261
+
+Figure E.5: MS\({}^{2}\) spectrum of MW315
+
+Figure E.6: MS\({}^{2}\) spectrum of MW317
+
+Figure E.7: MS\({}^{2}\) spectrum of MW317
+
+Figure F.1. MS\({}^{2}\) spectrum of MW357
+
+Figure F.2. MS\({}^{2}\) spectrum of MW377
+
+Figure F.3: MS\({}^{2}\) spectrum of MW433
+
+Figure F.4: MS\({}^{2}\) spectrum of MW477
+
+### G. Transformation products
+
+#### G1. Transformation products of ciprofloxacin
+
+\begin{tabular}{||c|c|c||c|c|c||c|c||c|c||} \hline
+**Molecular** & **Molecular** & **f**a** & **[M*H]** & **Errora** & **[M*Na]a** & **Errora** & **Highest content** & **Remarks** \\
+**forumar** & **weight** & & & & & & **A/Aschex** & \\ \hline
+**C1H10**_M_N_F_ & **331** & **5.05** & **332.14051** & **-0.0453** & **354.122500** & **-0.1812** & **100\% at 0 min** & **reference standard** \\ \hline _C12H10 N2F_ & _218_ & _4.20_ & _219.0927_ & _0.5319_ & _241.074500_ & _1.1968_ & _9.7\% at 180 min_ & _New_ \\ \hline _C14H10_ & _20 N2F_ & _246_ & _7.5-8.1_ & _247.087750_ & _-0.0772_ & _269.069650_ & _0.1057_ & _3.0\% at 180 min_ & _New_ \\ \hline _C14H10O N2F_ & _261_ & _0.66-1.3_ & _262.13495_ & _0.2489_ & _284.116950_ & _0.0383_ & _3.3\% at 60 min_ & _New_ \\ \hline C12H10O N2F & 262 & 10.1-10.8 & 263.082650 & -0.0153 & 285.064550 & 0.1564 & 10.8\% at 300 min & _Previously published_ \\  & & & & & & & & \\  & & & & & & & & \\  & & & & & & & & \\  & & & & & & & & \\  & & & & & & & & \\  & & & & & & & & \\ \hline
+**C18H19O N2F** & **287** & **1.0-1.9** & **288.15064** & **0.0871** & **310.132350** & **0.9055** & **48.3\% at 0 min** & **reference standard** \\ \hline C18H10O2 N2F & 289 & 5.5-6.9 & 290.12988 & 0.1729 & 312.111600 & 0.9512 & 6.8\% at 180 min & prev. publ. with different structure \\  & & & & & & & & (de Witte et al., 2008) \\ \hline C18H10O4 N2F & 290 & 10.4-10.8 & 291.077550 & 0.0379 & 313.059450 & 0.1931 & 2.0\% at 180 min & previously published \\  & & & 10.8 & & & & & (Paul et al., 2010) \\ \hline C18H10O3 N2F & 291 & 2.8-3.3 & 292.10913 & 0.2233 & **–** & & & 0.1\% at 30 min & previously published \\  & & & 12.9 & & & & (Tuerb et al., 2012) \\ \hline C18H10O2 N2F & _303_ & _4.2-4.9_ & _304.145550_ & _0.0990_ & _326.127450_ & _0.2474_ & _2.7\% at 120 min_ & _No structure identified_ \\ \hline
+**C18H10O3N2F** & **305** & **2.6-3.9** & **306.12479** & **0.1803** & **328.106650** & **0.4588** & **12.9\% at 120 min** & **reference standard** \\ \hline
+* [692] **G2. Transformation products of moxifloxacin**

manual_annotation/200pdfs_in_mmd/10.1016_j.chemosphere.2018.06.067.mmd

@@ -0,0 +1,294 @@
+## Accepted Manuscript
+
+1. Degradation characteristics of two typical N-heterocycles in ozone process: Efficacy, kinetics, pathways, toxicity and its application to real biologically pretreated coal gasification wastewater
+
+Hao Zhu, Wencheng Ma, Hongjun Han, Chunyan Xu, Yuxing Han, Weiwei Ma
+
+PII: S0045-6535(18)31142-1
+
+DOI: 10.1016/j.chemosphere.2018.06.067
+
+Reference: CHEM 21599
+
+To appear in: _ECSN_
+
+Received Date: 20 December 2017
+
+Revised Date: 9 June 2018
+
+Accepted Date: 9 June 2018
+
+Please cite this article as: Zhu, H., Ma, W., Han, H., Xu, C., Han, Y., Ma, W., Degradation characteristics of two typical N-heterocycles in ozone process: Efficacy, kinetics, pathways, toxicity and its application to real biologically pretreated coal gasification wastewater, _Chemosphere_ (2018), doi: 10.1016/j.chemosphere.2018.06.067.
+
+This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
+
+**Graphical abstract**Degradation characteristics of two typical N-heterocycles in ozone process: Efficacy, kinetics, pathways, toxicity and its application to real biologically pretreated coal gasification wastewater
+
+Hao Zhu\({}^{\text{a}}\), Wencheng Ma\({}^{\text{a}}\), Hongjun Han\({}^{\text{a}}\), Chunyan Xu\({}^{\text{a},\bullet}\), Yuxing Han\({}^{\text{b},\bullet\bullet}\), Weiwei Ma\({}^{\text{a}}\)
+
+\({}^{\text{a}}\) State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, China
+
+\({}^{\text{b}}\) School of Engineering, South China Agriculture University, Guangzhou 510642, China
+
+(E-mail address: 2534495668@qq.com (H. Zhu); damahit@163.com (W. Ma); han13946003379@163.com (H. Han); 15946091166@139.com (C. Xu); yuxinghan@scau.edu.cn (Y. Han); 501234355@qq.com(W. Ma)
+
+*Corresponding author
+
+Tel: +86 159-4609-1166; Fax: +86 451 86283082.
+
+E-mail address: 15946091166@139.com (Chunyan Xu).
+
+**Corresponding author
+
+Tel: +86 180-4505-3995; Fax: +86 451 86283082.
+
+E-mail address: yuxinghan@scau.edu.cn (Yuxing Han).
+
+## Abstract
+
+Ozonation of pyridine and indole was investigated both in aqueous solution and biologically pretreated coal gasification wastewater (BPCGW). Experimental results showed that the removal of indole was hardly affected by pH value. Direct reaction rate constant of ozone with pyridine increased from 0.18 M-1 s-1 (protonated pyridine) to 3.03 M-1 s-1 (molecular pyridine), and that with molecular indole was 8.6\(\times\)105 M-1 s-1. Seven and five transformation intermediates were observed for pyridine and indole, respectively. Ozonation pathways were proposed as hydroxylation, opening and cleavage of the aromatic ring. It was found that ammonia nitrogen (NH\({}_{3}\)-N) increased by 3.3 mg L-1 in ozone process, suggesting the broken of the C-N bonds of pyridine, indole and other N-heterocyclic compounds. In terms of biochemical oxygen demand to chemical oxygen demand (BODs/COD), toxicity and resazurin dehydrogenase activity (DHA), the biodegradability was improved after ozone treatment, indicating the possibility of ozone combined with biosystem for the treatment of BPCGW. The results of gas chromatograph and mass spectrometry (GC-MS) indicated that primary products during first 10 min might lead to the obstinate toxicity, which was further proved by US Environmental Protection Agency (US-EPA) test. This study would assist in obtaining a better understanding of the application of ozonation pretreatment in BPCGW.
+
+## 1 Introduction
+
+Pyridine and indole, as an important class of aromatic N-heterocycles, are often used as raw materials and solvents in the manufacture of dyes, pesticides, pharmaceuticals and other fine chemicals (Chen et al., 2013; Fetzner, 1998). In addition, they occur widely in coking wastewater, pharmaceutical wastewater and coal gasification wastewater (Lai et al., 2009; Wang et al., 2012; Zhang et al., 2009). Among them, coal gasification wastewater is one kind of toxic and refractory organic wastewater which mainly contains phenolic compounds, polynuclear aromatic hydrocarbons, long-chain alkanes and nitrogenous heterocyclic compounds (NHCs) (Wang et al., 2011). NHCs including pyridine and indole accounted for 60% of the recalcitrant compounds in coal gasification wastewater (Li et al., 2013). Even after biological treatments, there are still residual recalcitrant compounds due to the presence of biologically inhibitory organic substances (Uma and Sandhya, 1997). Heterocyclic structures of pyridine and indole improved their solubility compared with that of homocyclic analogues, making them easier access to the soil and ground water (Kuhn and Suffita, 1989; Stuermer et al., 1982). Once released into the environment, they exhibited high toxicity, mutagenicity and carcinogenicity among various life forms (Kamath and Vaidyanathan, 1990; Li et al., 2008; Meyer et al., 1999; Zhu et al., 2016). Therefore, new treatment strategies for elimination of pyridine and indole are made more acute before they enter the biosphere.
+
+Various technologies have been used for the treatment of wastewater containing pyridine and indole, such as biological methods (Chen et al., 2013; Liu et al., 2015; Luo et al., 2010; Sun et al., 2011; Zhang et al., 2009), Fenton-like oxidation (Li et al., 2014), catalytic oxidation (Singh and Shang, 2016) and adsorption (Zhu et al., 2016). For biodegradable organic pollutants, biological processes are economical andenvironmentally friendly technologies (Garcia-Pena et al., 2012; Moussavi and Heidarizad, 2011), while NHCs have toxic and inhibitory impact on the biological processes at high concentrations (Chen et al., 2013; Jing et al., 2013; Padoley et al., 2011). Thus, chemical oxidation as a pretreatment process could be used to reduce toxicity and to increase the biodegradability of wastewater, which would favor the subsequent biological process and improve overall effectiveness (Moussavi et al., 2014). Recently, ozonation as a physicochemical pretreatment has attracted much attention due to its capability of transforming refractory compounds into smaller and more biodegradable components to improve the overall biodegradability (Carballa et al., 2007; Lee and Gunten, 2010; Wert et al., 2009; Zhuang et al., 2014). As a strong oxidizing agent (E=2.08 V) (Rajeswari and Kannani, 2009), ozonation has been proved to be a promising method for the oxidation and destruction of a large number of organic pollutants in wastewater (Ikehata et al., 2006), while complete mineralization is often not achieved (Christophoridis et al., 2016). As a consequence, in some cases, highly toxic intermediates may be formed inevitably (Feng et al., 2016). These intermediates always differ in toxicity and may be more difficult to be removed. Therefore, study on the biodegradability and toxicity of ozonated intermediates is a field of growing interest (Aken et al., 2015; Coelho et al., 2009; De Torres-Socias et al., 2013). For a comprehensive assessment on the environment, it is necessary to elucidate the reaction kinetics, since it provides a time-dependent view of remediation of pyridine and indole. From the practical point of view, full investigations on the influencing factors and stoichiometry in ozone process as well as application of ozone pretreatment to real wastewater will be meaningful.
+
+So far, studies on the utilization of ozone process for degradation of pyridine and indole are limited, especially in the treatment of BPCGW. The primary objects of this study were as follows: (1) to investigate the efficacy of pyridine and indole under various pH values and ozone dosages; (2) to quantify the reaction rate constants of ozone with pyridine (protonated and molecular) and indole (molecular) as well as the rate constant of hydroxyl radicals (\({}^{\bullet}\)OH) with molecular pyridine; (3) to propose degradation pathways of pyridine and indole in ozone process; (4) to study the degradation characteristics of pyridine and indole in real BPCGW and the relationship between intermediates and toxicity. The major contribution of this work is the improved understanding of fundamental processes governing removal of pyridine and indole by ozone.
+
+## Materials and methods
+
+### Chemical and reagents
+
+Pyridine, indole, tert-butyl alcohol (TBA), disodium hydrogen phosphate, monometallic sodium orthophosphate and sodium thiosulfate of analytical grade were purchased from Tianjin Kermel Chemical Reagents Company (Tianjin, China) and used for ozonation experiments without further purification. Intermediates were obtained from Aladdin Industrial Corporation (Shanghai, China). All solutions were prepared with deionized water.
+
+The real BPCGW was collected from the effluent of secondary settling tank in the full-scale wastewater treatment facility, which was described in detail in Supplementary Material. Details on wastewater characterization were as follows: 150-180 mg L-1 of COD, 0.05-0.08 of BOD5/COD ratio, 60-70 mg L-1 of total organic carbon (TOC), 50-60 mg L-1 of total nitrogen, 20-30 mg L-1 of NH3-N and pH value of 7.5-8.5.
+
+### Experimental set-up and ozone procedure
+
+The experiments were performed in a cylindrical water-jacketed reactor (internaldiameter of 9 cm and working height of 25 cm) with a total working volume of 1 L. The system was operated in a semi-batch mode including O3 supply and emission equipment. Ozone was produced in dry air by an ozone generator (ANSEROS, COM-AD-01) and was introduced continuously into the solution at the bottom of the reactor by means of a porous glass diffuser with a desired gas flow rate of 40 L h-1 in the experiment. The effluent O3 was further treated with a 20% KI solution. The solution was perfectly mixed by a magnetic stirrer agitated at 400 rpm. All experiments were carried out at room temperature.
+
+For ozone experiments in aqueous solution, the solution of pyridine and indole was first prepared and each run was initiated when the produced ozone was stable. During the run, aliquots were taken from the reactor at predefined time points and quenched by adding a concentrated solution of sodium thiosulfate or purged with high-purity N2 (applied in section 3.5.2) for 5 min to remove residual ozone. Following above procedures, various pH values of 5, 7 and 9 (adjusted by phosphate buffer) as well as ozone dosages of 1, 2; 4 and 8 mg L-1 were performed. Meanwhile, the effect of TBA on the degradation of pyridine and indole was investigated. In addition, a control test of pure oxygen purging was conducted in order to discard the effect of volatilization during the experiments. Kinetic study and identification of intermediates were presented in section 2.3 and 2.6, respectively.
+
+For ozone experiments in real BPCGW, degradation characteristics of pyridine and indole were first studied. Then, the relationship between intermediates and toxicity were evaluated in terms of changes of BOD5/COD, toxicity and DHA as well as analysis of GC-MS and US-EPA test.
+
+Experiments were conducted in triplicate and average values were presented.
+
+#### 2.3.1 Determining the reaction rate constant of ozone with pyridine and indole
+
+The reaction rate constant of ozone with pyridine was studied using direct oxidation method. Pyridine presents one dissociation constant (pKa=5.23) (Tekle-Rottering et al., 2016), so it is interesting to study the rate constant of ozone with the molecular and protonated forms of pyridine. k (Eq. (1)) was determined under the condition of acid pH buffered with phosphate. In addition, TBA (0.05 M) was added to aqueous solutions to scavenge radical reactions in the experiments. Aliquots were taken from the reactor every 5 min for 30 min to determine the residual concentration of pyridine and dissolved ozone. The rate constant for the reaction was calculated using:
+
+\[k = a_{1}k_{1} + a_{2}k_{2}\]
+
+where k is the apparent rate constant; k1 and k2 is the specific rate constant for the molecular and the protonated form, respectively; and a1 and a2 mean respective fractions of pyridine at the acid pH value. The values of a1 and a2 can be calculated from the concentration of H+ and the dissociation constant. In the test, specific ozone dosages were measured according to previous study (Shen et al., 2008).
+
+The reaction rate constant of ozone with indole was determined by the competition method using HSO3-1 as the reference compound similar to the previous study (Yao and Haag, 1991). In brief, 25 mL indole (600 mg L-1) and 10 mL NaHCO3 (750 mg L-1) as well as phosphate were added into six conical flasks (500 mL) in order to reach the settled concentration and pH value of 4.5. Then, different concentrations of ozonated water (15 mg L-1) of 0, 100, 200, 300, 400 and 500 mL were slowly introduced into each conical flask. After that, deionized water was used to adjust equal volume of solution. Finally after the reaction, the concentration of indole and HSO3- of each conical flask was measured.
+
+#### Determining the reaction rate constant of *OH with pyridine
+
+In this experiment, competitive kinetics method with para-chlorobenzoic acid (pCBA) as a reference compound was used to determine the reaction rate constant of *OH with pyridine. pCBA was chosen since it hardly reacted with ozone and the reaction rate of *OH with pCBA was as high as 5\(\times\)109 M-1 s-1 (Haag and Yao, 1992). The experiment was carried out at pH value of 9 buffered with phosphate. In alkaline conditions, the reaction with ozone can be ignored due to its short half-lives. Pyridine (20 mg L-1) and pCBA (5 mg L-1) were spiked into the solution. Five milliliters of the aliquot was withdrawn every 5 min for 30 min and the remaining concentration of pyridine and pCBA was analyzed by High Performance Liquid Chromatography (HPLC).
+
+#### Stoichiometry
+
+The experiments were carried out in a series of volumetric flasks (20 mL). Each flask consisted of ozone solution (1 mL 15 mg L-1) and TBA (1 mL 15 mg L-1) as well as different volumes of pollutant solution (1-9 mL) to reach the settled molar C0/ozone0 ratio (1:1-9:1). Then pure water was put into the flask to make the solution up to 15 mL. The system was controlled at pH value of 5 (adding a certain amount of sodium hydroxide into 0.2 M sodium dihydrogen phosphate solution) for pyridine (considering its pKa of 5.23) and 3 for indole. When ozone was completely consumed, the concentration of the pollutants was analyzed by HPLC.
+
+### Biodegradability index, toxicity assessment and DHA
+
+The activated sludge used in the bioassays for biodegradability and toxicity was taken from the aerobic tank in the full-scale wastewater treatment plant (Jia et al., 2015).
+
+BOD5/COD was adopted as biodegradability index. Toxicity assessment wasdescribed in a previous study (De Torres-Socias et al., 2013). Briefly, the test was conducted in a glass reactor with 1 L of endogenous activated sludge at a constant temperature of 25 \(\square\). The dissolved oxygen concentration was measured by dissolved oxygen meter. This test compared the oxygen uptake of the target sample with that of the reference sample. 30 mL of sample was added into the glass reactor to obtain the OUR-sample and 30 ml of distilled water with 0.5 g of sodium acetate g volatile suspended solid was injected to gain the OUR-reference. The inhibition percentages were calculated by the Eq. (2):
+
+\[\mathit{Inhibition}(\%)=100\times(1-\frac{\mathit{OUR}-\mathit{sample}}{ \mathit{OUR}-\mathit{reference}}) \tag{2}\]
+
+DHA referred to the previous method (Liu, 1983). Resazurin, as an electron receptor that could substitute for the electron transport of the respiratory chain, was applied to measure the dehydrogenase activity of microorganisms to characterize the biodegradability due to its easy operation and strong electrophilic ability. The apparatus used for the experiments contained seven centrifugal tubes (20 mL) at temperature of 25 \(\square\). First, endogenous activated sludge (2.5 mL) and the sample (2.5 mL) were put into the tube and contacted for 30 min. Then resazurin (0.5 mL, 100 mg L\({}^{-1}\)) was added to the tube, 30 min later, sodium carbonate (5 mL, 1 M) (Strotmann et al., 1993) was added to terminate the reaction. After that, the tube was centrifuged at a rate of 3000 rpm for 5 min and the absorbance of supernatant was measured at 610 nm. The DHA can be calculated by the Eq. (3):
+
+\[\mathit{DHA}=(\frac{\Delta E_{\mathit{5}\mathit{10}}\times 10^{6}\times V}{ \mathit{\varepsilon t}}) \tag{3}\]
+
+where \(\Delta\)E is the change of absorbance; \(\mathit{\varepsilon}\) is molar absorption coefficient; V is the total volume; t is the reaction time of resazurin.
+The concentration of pyridine and indole was analyzed using a liquid chromatograph (HPLC, LC-1200, Agilent, USA) with an Agilent C\({}_{18}\) (150 mm\(\upalpha\)4.6 mm). The mobile phase was methanol to water (60:40) and the wavelength of pyridine and indole was 253 nm and 287 nm, respectively. pCBA was determined according to the previous study (Wen et al., 2011). COD, BOD\({}_{5}\) and NH\({}_{3}\)-N were measured in accordance with standard methods for examination of water and wastewater (APHA, 1998). Ozone concentration in the gas was monitored by ozone gas detection instrument and dissolved ozone concentration was determined via the indigo method (Bader and Hoigne, 1981). TOC was monitored via a TOC analyzer (TOC-VCPH, Shimadzu, Japan). pH value was determined with a pH meter (pHS-3C, Leici, China).
+
+The identification of intermediates was performed by an Agilent 1290 LC equipped with an Agilent 6460 Triple Quadrupole mass spectrometer system with multiple-reaction monitoring (LC-MS/MS-MRM). The mass data were acquired under positive ion mode. Intermediates were confirmed in terms of the match with parent/daughter ion. The parameters were as follows: Gas temperature: 300 \(\square\); Source voltage: 3.5 KV; Cone voltage: 32 V; Collision energy: 16 V.
+
+The identification of formic acid, acetic acid and oxalic acid was conducted by ion chromatography (IC) according to the previous study (Qiang et al., 2013).
+
+## 3 Results and discussion
+
+### Efficacy
+
+To investigate the degradation of pyridine and indole by ozone, the experiments were performed at different pH values and with different ozone dosages. Ozone was introduced into 20 mg L\({}^{\text{-1}}\) pyridine (reaction time 30 min) and 10 mg L\({}^{\text{-1}}\) indole (reaction time 60 s), respectively with an influent ozone dosage of 1, 2, 4 and 8 mgL-1.
+
+As shown in Fig. 1a, it was demonstrated that the degradation of pyridine increased with the increase of ozone dosage at different pH values. The best removal efficiency was observed at pH value of 9 with ozone dosage of 8 mg L-1, at which, complete degradation of pyridine was achieved. Especially, the difference of removal efficiency at pH value of 5 and 7 might also be related with its different structures (molecular and protonated forms of pyridine),since dissociation constant of pyridine was 5.23, indicating that protonated form of pyridine might have a lower reactivity with ozone than that of the molecular form, which would be proved in section 3.2.1.
+
+The degradation of indole showed a different trend, as revealed in Fig. 1b. It was found that the increase of ozone dosage was beneficial for degradation process, while the influence of pH value on the removal efficiency of indole was not significant. At different ozone dosages, the maximum change of indole degradation caused by pH value was only 2.8%. In addition, the introduction of TBA had little effect on the removal efficiency of indole, indicating the dominating role of ozone in influencing the degradation of indole rather than *OH. Control test of pure oxygen purging indicated that there was no volatilization of pyridine and indole during the experiments (data not shown).
+
+* [265] **Fig. 1.** Influence of ozone dosage on the degradation of pyridine and indole at different pH values (Conditions for a: 20 mg L-1 of pyridine, gas flow of 40 L h-1 and reaction time of 30 min; for b: 10 mg L-1 of indole, gas flow of 40 L h-1 and reaction time of 60 s).
+* [266] **Fig. 2.** Kinetics
+* [267] **Fig. 3.**2.**1. Determining the reaction rate constant of ozone with pyridine and indole
+* [268] **Fig. 4.** The reaction of pyridine with aqueous ozone and *OH in ozone process can be described by second-order kinetics (Bourgin et al., 2013): \[\frac{\text{d}[\mathit{pyridine}]}{\text{d}t}=-k\bullet\text{on}[\mathit{pyridine}][\bullet\mathit{OH}]-k\infty[\mathit{pyridine}][\text{O}_{3}]\] (4)
+
+In this study, when influent ozone gas was continuously introduced into the solution, the concentration of aqueous ozone could be assumed to be constant in ozone process. At pH value of 3 and 5, when the concentration of TBA was higher than that of pyridine (20 mg L-1), the reaction of ozone with pyridine played a dominant role and the influence of OH could be ignored. Thus, the Eq. (4) can be rewritten as: \[\frac{d[\mathit{pyridine}]}{\text{d}t}=-ko_{3}[\mathit{pyridine}][\mathit{O}_{3}]\] (5) \[=\ln\frac{[\mathit{pyridine}]}{[\mathit{pyridine}]_{0}}=k_{o_{3}}\int[ \text{O}_{3}]\text{d}t\] (6)
+
+According to the Eq. (6), k can be concluded from the plots of pyridine versus ozone exposure. As shown in Fig. 2a, the solution pH exerted a strong impact on k value. At pH 5, pyridine was mainly presented as molecular form and its k value was 1.391 M-1 s-1, while the k value decreased to 0.194 M-1 s-1 at pH 3, which might be ascribed to the increasing fractions (99.98%) of the protonated species of target pyridine with decreasing pH to 3. Moreover, according to Eq. (1), the reaction rate constant of ozone with pyridine was calculated and the value was 3.03 M-1 s-1 for molecular form and 0.18 M-1 s-1 for protonated form, respectively, which further explained the observed increase of the apparent reaction rate constant with decreasing pH value.
+
+The reaction rate constant of ozone with indole was determined by competitive kinetic method. In this study, HSO3' was adopted as the reference compound and the reaction rate constant could be obtained using:
+
+\[k_{\mathit{indole}\bullet O_{3}} = \frac{- \ln([\mathit{Indole}]/[\mathit{Indole}]_{0})}{- \ln([M\,]/[M\,]_{0})}k_{\text{M}\bullet O_{3}}\]
+
+where kindoleO3 and kM+O3 are the reaction rate constant of ozone with indole and HSO3'. As the kM-O3 is known from the literature, the kindoleO3 can be easily calculated.
+
+As could be seen from Fig. 2b, the rate (calculated using Eq. (7)) was about 2.69 times higher for indole than that of HSO3'. KM was 3.2x105 M-1 s-1 obtained from the literature (Hoigne et al., 1985), thus, the reaction rate constant between indole and O3 was 8.6x105 M-1 s-1.
+
+It could be seen that indole was more reactive toward ozone than pyridine, which might be due to their different structures. For indole, it has one aromatic ring constituted by five atoms and some of the bond angles may not be the relatively stable sp2 hybrid orbital, which is associated with the stretch energy. The higher reactivity of ozone with indole may be the result of its higher stretch energy than that of pyridine, while it requires further study.
+
+Fig. 2: Plots used to determine the reaction rate constant of ozone with pyridine (a) and indole (b) (Conditions for a: 20 mg L-1 of pyridine, ozone dosage of 4 mg L-1, gas flow of 40 L h-1 and reaction time of 30 min; for b: 30 mg L-1 of indole, 15 mg L-1 of NaHCO3, different initial ozone concentrations of 0, 3, 6, 9, 12, 15 mg L-1, and reaction time of 3 min).
+
+Fig. 3: Plots used to determine the reaction rate constant of ozone with pyridine (a) and indole (b) (Conditions for a: 20 mg L-1 of pyridine, ozone dosage of 4 mg L-1, gas flow of 40 L h-1 and reaction time of 30 min; for b: 30 mg L-1 of indole, 15 mg L-1 of NaHCO3, different initial ozone concentrations of 0, 3, 6, 9, 12, 15 mg L-1, and reaction time of 3 min).
+
+* [328] **Fig. 3. Plots used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [334] **Fig. 4. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [335] **Fig. 5. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [336] **Fig. 6. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [337] **Fig. 7. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [338] **Fig. 8. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [339] **Fig. 9. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [340] **Fig. 10. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [341] **Fig. 11. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [342] **Fig. 12. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [343] **Fig. 13. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [344] **Fig. 14. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [345] **Fig. 15. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [346] **Fig. 16. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [347] **Fig. 17. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [348] **Fig. 18. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [349] **Fig. 19. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [350] **Fig. 20. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [351] **Fig. 21. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [352] **Fig. 22. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [353] **Fig. 23. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [354] **Fig. 24. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [355] **Fig. 25. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [356] **Fig. 26. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [357] **Fig. 27. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [358] **Fig. 28. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [359] **Fig. 29. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [360] **Fig. 30. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [361] **Fig. 31. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [362] **Fig. 32. Plot used to determine the rate constant of *OH with pyridine using the competition method (pCBA as the competitor). (a) The residual concentration of pyridine and pCBA with the time. (b) The relationship between pyridine and pCBA concentrations (Conditions for a and b: 20 mg L-1 of pyridine, 5 mg L-1 of pCBA, ozone dosage of 4 mg L-1, pH value of 9, gas flow of 40 L h-1 and reaction time of 30 min).**
+* [363] **Fig. 33. Plot used to determine the rate low, some intermediates produced by other reactions might consume ozone, resulting in a higher z value (Beltran-Heredia et al., 2001). As the concentration of pyridine and indole increased, ozone was considered to be consumed only by the target pollutants, thus, z tended to be a stable value of 4.0 and 0.25, which was the stoichiometry of ozone with pyridine and indole, respectively.
+* 3.4. Degradation pathways
+* 3.5. The intermediates of ozonation of pyridine and indole determined by LC-MS/MS and IC were shown in Table 1. The mass spectrum of P1 to P6 was provided in the supplementary material.
+* 3.5. The intermediates in ozone process and its US-EPA test analysis.
+* 3.6. The intermediates in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and the water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and the water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in the water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in the water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water were calculated by calculating the mean of the total number of water molecules in water and water. The total number of water molecules in water and water are calculated by calculating the mean of the total number of water molecules in water and water.
+
+C(2)-C(3) and C(6)-C(7). Then, low-molecular-weight carboxylic acids such as formic acid, acetic acid, oxalic acid and iminodiacetic acid were produced with the cleavage of aromatic ring and CO2 and H2O were the final oxidation products (Zhang et al., 2008). In the previous study, P3 was also observed in the ozone process (Tekle-Rottering et al., 2016). Their research paid more attention to the discussion on mechanism of generation of primary products in ozone process. While a new insight into reaction scheme with other two hydroxylation products and iminodiacetic acid was found in this study.
+
+### 3.5 Ozone process applied to real BPCGW
+
+#### 3.5.1 Degradation and mineralization
+
+As shown in Fig. 5, when the removal efficiency of pyridine and indole was 81.2% and 100%, the residual TOC after 60 min (58 mg L-1, decreased from the initial 67.9 mg L-1) was observed with a mineralization rate of 14.6%, indicating that the residual pollutants after reactions were difficult to be oxidized by ozone. Degradation of indole in real wastewater was much faster than that of pyridine due to its high reactivity with ozone, as studied in kinetic study. Furthermore, formation of NH3-N
+
+Figure 4: Proposed degradation pathways of pyridine and indole.
+
+was found and its concentration increased by 3.3 mg L-1 in ozone process, suggesting possible cleavage of C-N bonds of pyridine and indole as well as other N-containing compounds. NH3-N was also observed in the oxidation of indigo (Qu et al., 2015) and diclofenac (Coelho et al., 2009) and its formation might be due to the electron-transfer reactions involving C-centered radicals and imine intermediates (Vera et al., 2017).
+
+removal to ozone dosage was 0.31. In addition, Fig. 6 revealed the inhibition percentage of the treated solutions decreased from 53% to 36%. This was comparable with the increase of the biodegradability index. In order to assess the biodegradability of the ozonated wastewater better, the test of DHA was also performed. As shown in Fig. 6, DHA increased from 177.4 to 288.1 nano-mole/h, indicating improved biodegradability. In first 10 min, the increased toxicity was in agreement with the decreased DHA, indicating the generation of more bio-inhibitory compounds, which would be discussed in section 3.5.3. Similarly for the treatment of BPCGW, physicochemical pretreatment combined with biological processes were proved to be very effective methods in previous studies (Hou et al., 2015; Jia et al., 2015; Xu et al., 2015; Hou et al., 2015). All these results indicated that BPCGW was expected to be treated by a biosystem after partial ozone process in the future.
+
+5.3 The relationship between intermediates and toxicity was evaluated by GC-MS and US-EPA test. Ecological Structure Activity Relationships program was employed in the US-EPA test, which had been widely used by environmental assessors, chemical
+
+Fig. 6: Evolution of biodegradability index, toxicity and resazurin dehydrogenase activity (DHA) in ozone process (Conditions: ozone dosage of 4 mg L−1, pH value of 7.6 and gas flow of 40 L h−1).
+
+suppliers and other regulatory agencies for toxicity monitoring (Tay and Madehi, 2015).
+* Reports on some other pharmaceutical and personal care products have observed an increased toxicity after ozonation (Kuang et al., 2013; Rosal et al., 2009). Sometimes, the toxicity increased at the first, and then decreased with continuous ozone (Dantas et al., 2007; Gomez-Ramos et al., 2011). Based on above results, it was found that intermediates during the first 10 min might lead to the obstinate toxicity, thus, GC-MS of raw wastewater and ozonated wastewater (10 min) as well as US-EPA test were performed.
+* Before ozonation, as shown in Table S1, it could be seen that the major toxic and refractory compounds of raw wastewater were phenols and their derivatives, polycyclic aromatic hydrocarbons, long-chain hydrocarbons and NHCs. After 10 min ozonation, the toxicity increased accompanied with decreased DHA, as revealed in Fig. 6. According to Table S1, most of phenolic compounds were reduced because of their high reactivity with O3 (k03=1.8x106 M-1 S-1) and *OH (k.OH=.6.1x109 M-1 S-1) (Jin et al., 2012), while the relative percentage of 1,2-benzenediol and resorcinol improved after ozonation, which might be due to the hydroxylation of phenols in ozone process (Amado-Pina et al., 2016). Combined with the result of US-EPA test, it could be inferred that 1,2-benzenediol and resorcinol might contribute to the rise in toxicity. In addition, some hydrocarbons including decane, 1-decene, 4-methyl-, tridecane, tetradecane and heptadecane with higher toxicity increased after ozonation, which perhaps resulted from ring opening reactions of other aromatic compounds. This would be another reason for the increased toxicity. Similar conclusions were also drawn by other studies that the increased toxicity after ozonation was originated from ring-opened structure (Kuang et al., 2013; Larcher et al., 2012). The ozonated sample * [448] exhibited lower toxicity than the raw wastewater until 35 min later, which might be attributed to the presence of low toxic substances such as hydroxypyridines and some low-weight molecular acids, as revealed in Table 1. There could be more non-identified oxidation intermediates that were also responsible for the increased or decreased toxicity. Overall, the degradation of real BPCGW can pose potential toxicity at the initial period, while there is no need to worry about this problem after 60 min.
+
+## 4 Conclusion
+
+The study presented the characteristics of ozone with pyridine and indole in aqueous solution and real BPCGW. pH value did not to seem to affect indole degradation. In addition, reaction rate constant ozone with indole was higher than that with pyridine. What's more, when ozone pretreatment was applied to treat real wastewater, formation of NH3-N was observed, implying the breakage of C-N bonds. The results also demonstrated that ozonation could improve the biodegradability of real BPCGW in terms of BOD5/COD, toxicity and DHA, indicating the potential of ozone integrated with biosystem for this type of wastewater. Finally, intermediates such as 1,2-benzenediol, resorcinol and some hydrocarbons produced during the first 10 min might result in the increase of toxicity. This study would assist in obtaining a better understanding of the application of ozone pretreatment in BPCGW.
+
+## Acknowledgements
+
+This work was supported by National key research and development program-China (2016YFB0600502).
+
+## Appendix A Supplementary information
+
+Supplementary data about this article can be found in the online version.
+
+* [472]**References**
+* [473] APHA, 1998. Standard Methods for the Examination of Water and Wastewater, twentieth ed. A.P.H. Association, Washington, DC, USA.
+* [475] Aken, P.V., Broeck, R.V.D., Degreve, J., Dewil, R., 2015. The effect of ozonation on the toxicity and biodegradability of 2,4-dichlorophenol-containing wastewater. Chem. Eng. J. 280, 728-736.
+* [477] Amado-Pina, D., Roa-Morales, G., Barrera-Diaz, C., Balderas-Hernandez, P., Romero, R., Campo, E.M.D., Natividad, R., 2016. Synergic effect of ozonation and electrochemical methods on oxidation and toxicity reduction: Phenol degradation. Fuel. 198, 82-90.
+* [480] Bader, H., Hoigne, J., 1981. Determination of ozone in water by the indigo method. Water Res. 15, 449-456.
+* [482] Beltran-Heredia, J., Torregrosa, J., Dominguez, J.R., Peres, J.A., 2001. Kinetics of the reaction between ozone and phenolic acids present in agro-industrial wastewaters. Water Res. 35, 1077_1085.
+* [484] Bourgin, M., Violleau, F., Debrauwer, L., Albet, J., 2013. Kinetic aspects and identification of by-products during the ozonation of bitertanol in agricultural wastewaters. Chemosphere. 90, 486 1387-1395.
+* [487] Carballa, M., Manterola, G., Larrea, L., Ternes, T., Omil, F., Lema, J.M., 2007. Influence of ozone pre-treatment on sludge anaerobic digestion: removal of pharmaceutical and personal care products. Chemosphere. 67, 1444-1452.
+* [490] Chen, Y., Xie, X., Ren, C., Dai, C., 2013. Degradation of N-heterocyclic indole by a novel endophytic fungus Phonolysis liquidambari. Bioresource Technol. 129, 568-574.
+* [491] Christophoridis, C., Nika, M.C., Aalizadeh, R., Thomaidis, N.S., 2016. Ozonation of ranitidine: Effect of experimental parameters and identification of transformation products. Sci. Total Environ. 170, 557-558.
+* [492] Coelho, A.D., Sans, C., Aguera, A., Gomez, M.J., Esplugas, S., Dezotti, M., 2009. Effects of ozone pre-treatment on diclofenac: intermediates, biodegradability and toxicity assessment. Sci. Total Environ. 407, 3572-3578.
+* [493] Dantas, R.F., Canternino, M., Marotta, R., Sans, C., Esplugas, S., Andreozzi, R., 2007. Bezafibrate removal by means of ozonation: Primary intermediates, kinetics, and toxicity assessment. Water Res. 41, 2525-2532.
+* [504] Dantas, R.F., Contreras, S., Sans, C., Esplugas, S., 2008. Sulfamethoxazole abatement by means of ozonation. J. Hazard. Mater. 150, 790-794.
+* [505] De Torres-Socias, E., Fernandez-Calderero, L., Oller, I., Trinidad-Lozano, M.J., Yuste, F.J., Malato, S., 2013. Cork boiling wastewater treatment at pilot plant scale: Comparison of solar photo-Fenton and ozone (O\({}_{3}\), O\({}_{9}\)/H\({}_{2}\)O\({}_{2}\)). Toxicity and biodegradability assessment. Chem. Eng. J. 234, 232-239.
+* [506] Feng, M., Yan, L., Zhang, X., Sun, P., Yang, S., Wang, L., Wang, Z., 2016. Fast removal of the antibiotic fluumequin for aqueous solution by ozonation: Influencing factors, reaction pathways, and toxicity evaluation. Sci. Total Environ. 541, 167-175.
+* [507] Fetzner, S., 1998. Bacterial degradation of pyridine, indole, quinoline, and their derivatives under different redox conditions. Appl. Microbiol. Biot. 49, 237-250.
+* [511] Garcia-Pena, E.I., Zarate-Segura, P., Guerra-Blanco, P., Poznyak, T., Chairez, I., 2012. Enhanced Phenol and Chlorinated Phenols Removal by Combining Ozonation and Biodegradation. Water, Air, & Soil Pollution. 223, 4047-4064.
+* [512] Gomez-Ramos, M.M.; Mezcua, M., Aguera, A., Fernandez-Alba, A.R., Gonzalo, S., Rodriguez, A., Rosal, R., 2011. Chemical and toxicological evolution of the antibiotic sulfamethoxazole under ozone treatment in.water solution. J. Hazard. Mater. 192, 18-25.
+* [513] Haag, W.R., Yao, C.C.D., 1992. Rate constants for reaction of hydroxyl radicals with several drinking water contaminants. Environ. Sci. Technol. 26, 1005-1013.
+* [514] Hoigne, J., Bader, H., Haag, W.R., Staehelin, J., 1985. Rate constants of reactions of ozone with organic and inorganic compounds in water--III. Inorganic compounds and radicals. Water Res. 19, 993-1004.
+* [515] Hou, B., Han, H., Zhuang, H., Xu, P., Jia, S., Li, K., 2015. A novel integration of three-dimensional electro-Fenton and biological activated carbon and its application in the advanced treatment of biologically pretreated Lurgi cool gasification wastewater. Bioresource Technol. 196, 721-725.
+* [516] Ikehata, K., Naghashkar, N.J., El-Din, M.G., 2006. Degradation of Aqueous Pharmaceuticals by Ozonation and Advanced Oxidation Processes: A Review. Ozone Science & Engineering. 28, 353-414.
+* [517] Jia, S., Han, H., Zhuang, H., Hou, B., 2015. The pollutants removal and bacterial community dynamics relationship within a full-scale British Gas/Lurgi cool gasification wastewater treatment using a novel system. Bioresource Technol. 200, 103-110.
+* [518] Jia, S., Han, H., Zhuang, H., Xu, P., Hou, B., 2015. Advanced treatment of biologically pretreated coal * [532] gasification wastewater by a novel integration of catalytic ultrasound oxidation and membrane bioreactor. Bioresource Technol. 189, 426-429.
+* [534] Jin, X., Peldszus, S., Huck, P.M., 2012. Reaction kinetics of selected micropollutants in ozonation and advanced oxidation processes. Water Res. 46, 6519-6530.
+* [535] Jing, J., Li, J., Feng, J., Li, W., Yu, W., 2013. Photodegradation of quinoline in water over magnetically separable Fe\({}_{2}\)O\({}_{3}\)/TiO\({}_{2}\) composite photocatalysts. Chem. Eng. J. 219, 355-360.
+* [536] Kamath, A.V., Vaidyanathan, C.S., 1990. New pathway for the biodegradation of indole in Aspergillus niger. Applied & Environmental Microbiology. 56, 275-280.
+* [537] Kuang, J., Huang, J., Wang, B., Cao, Q., Deng, S., Yu, G., 2013. Ozonation of trimethoprim in aqueous solution: Identification of reaction products and their toxicity. Water Res. 47, 2863-2872,
+* [538] Kuhn, E.P., Suffita, J.M., 1989. Microbial degradation of nitrogen, oxygen and sulfur heterocyclic compounds under anaerobic conditions: Studies with aquifer samples. Environmental Toxicology & Chemistry, 8, 1149-1158.
+* [539] Lai, P., Zhao, H., Zeng, M., Ni, J., 2009. Study on treatment of coking wastewater by biofilm reactors combined with zero-valent iron process. J. Hazard. Mater. 162, 1423-1429.
+* [540] Larcher, S., Delbes, G., Robaire, B., Yargeau, V., 2012. Degradation of 17\(\overline{a}\)-ethinylestradiol by ozonation-identification of the by-products and assessment of their estrogenicity and toxicity. Environ. Int. 39, 66-72.
+* [541] Lee, Y., Gunten, U.V., 2010. Oxidative transformation of micropollutants during municipal wastewater treatment: Comparison of kinetic aspects of selective (chlorine, chlorine dioxide, ferrate VI, and ozone) and non-selective oxidants (hydroxyl radical). Water Res. 44,555-566.
+* [542] Li, H., Han, H., Du, M., 2013. Effect of nitrate concentration on performance of pre-denitrification moving bed biofilm reactor system in treating coal gasification wastewater. Desalination & Water Treatment. 51, 5996-6002.
+* [543] Li, J., Cai, W.J., Cai, J.J., 2008. The characteristics and mechanisms of pyridine biodegradation by Streptomyces sp. J. Hazard. Mater. 165, 950-954.
+* [544] Li, N., Lu, X., Zhang, S., 2014. A novel reuse method for waste printed circuit boards as catalyst for wastewater bearing pyridine degradation. Chem. Eng. J. 257, 253-261.
+* [545] Liu, D., 1983. Resazurin reduction method for activated sludge process control. Environ. Sci. Technol. 17, 407-411.
+* [546] Liu, X., Chen, Y., Zhang, X., Jiang, X., Wu, S., Shen, J., Sun, X., Li, J., Lu, L., Wang, L., 2015. Aerobic granulation strategy for bioaugmentation of a sequencing batch reactor (SBR) treating high strength pyridine wastewater. J. Hazard. Mater. 295, 153-160.
+* [547] Luo, Y., Zhang, R., Liu, G., Li, J., Li, M., Zhang, C., 2010. Electricity generation from indole and microbial community analysis in the microbial fuel cell. J. Hazard. Mater. 176, 759-764.
+* [548] Meyer, S., Simone Cartellieri, A., Steinhart, H., 1999. Simultaneous Determination of PAHs, Hetero-PAHs (N, S, O), and Their Degradation Products in Creosote-Contaminated Soils. Method
+* [549] Development, Validation, and Application to Hazardous Waste Sites. Anal. Chem. 71, 169-176.
+* [550] Moussavi, G., Aghapour, A.A., Yaghmaeian, K., 2014. The degradation and mineralization of catechol using ozonation catalyzed with MgO/GAC composite in a fluidized bed reactor. Chem. Eng. J. 249, 302-310.
+* [551] Moussavi, G., Heidarizad, M., 2011. The performance of SBR, SCR, and MSCR for simultaneous biodegradation of high concentrations of formaldehyde and ammonia. Separation & Purification Technology. 77, 187-195.
+* [552] Padoley, K.V., Mudflar, S.N., Banerjee, S.K., Deshmukh, S.C., Pandey, R.A., 2011. Fenton oxidation: A pretreatment option for improved biological treatment of pyridine and 3-cyanopyridine plant wastewater. Chem. Eng. J. 166, 1-9.
+* [553] Qiang, Z., Ling, W., Tian, F., 2013. Kinetics and mechanism for omethoate degradation by catalytic ozonation with Fe(III)-loaded activated carbon in water. Chemosphere. 90, 1966-1972.
+* [554] Qu, R., Xu, B., Meng, L., Wang, L., Wang, Z., 2015. Ozonation of indigo enhanced by carboxylated carbon nanotubes: Performance optimization, degradation products, reaction mechanism and toxicity evaluation. Water Res. 68, 316-327.
+* [555] Rajeswari, R., Kanmani, S., 2009. A study on synergistic effect of photocatalytic ozonation for carbaryl degradation. Desalination, 242, 277-285.
+* [556] Rosal, R., Gonzalo, M.S., Boltes, K., Leton, P., Vaquero, J.J., Garcia-Calvo, E., 2009. Identification of intermediates and assessment of ecotoxicity in the oxidation products generated during the ozonation of clofibric acid. J. Hazard. Mater. 172, 1061-1068.
+* [557] Shang, N.C., Yu, Y.H., Ma, H.W., Chang, C.H., Liou, M.L., 2006. Toxicity measurements in aqueous solution during ozonation of mono-chlorophenols. J. Environ. Manage. 78, 216-222.
+* [558] Shen, J.M., Chen, Z.L., Xu, Z.Z., Li, X.Y., Xu, B.B., Qi, F., 2008. Kinetics and mechanism of * [592] degradation of p-chlornitrobenzene in water by ozonation. J. Hazard. Mater., 152, 1325-1331.
+* [593] Singh, S., Shang, L.L., 2016. Catalytic performance of hierarchical metal oxides for per-oxidative degradation of pyridine in aqueous solution. Chem. Eng. J. 309, 753-765.
+* [595] Strotmann, U.J., Butz, B., Bias, W.R., 1993. The dehydrogenase assay with resazurin: practical performance as a monitoring system and Ph-dependent toxicity of phenolic compounds. Ecotoxicology & Environmental Safety. 25, 79-89.
+* [598] Stuermer, D.H., Ng, D.J., Morris, C.J., 1982. Organic contaminants in groundwater near an underground coal gasification site in northeastern Wyoming. Environ. Sci. Technol. 16, 582-587.
+* [600] Sun, J., Xu, L., Tang, Y.Q., Chen, F., Liu, W., Wu, X., 2011. Degradation of pyridine by one Rhodococcus strain in the presence of chromium (VI) or phenol. J. Hazard. Mater. 191, 62-68.
+* [602] Tay, K.S., Madehi, N., 2015. Ozonation ofloxacin in water: By-products, degradation pathway and ecotoxicity assessment. Sci. Total Environ. 520, 23-31.
+* [604] Tekle-Rottering, A., Reisz, E., Jewell, K.S., Lutze, H.V., Ternes, T.A., Schmidt, W., Schmidt, T.C., 2016. Ozonation of pyridine and other N-heterocyclic aromatic compounds. Kinetics, stoichiometry, identification of products and elucidation of pathways. Water Res. 102, 582-593.
+* [607] Uma, B., Sandhya, S., 1997. Pyridine degradation and heterocyclic nitrification by Bacillus coagulans. Can. J. Microbiol. 43, 595-598.
+* [609] Vera, G.A.D.D., Gernjak, W., Weinberg, H., Farre, M.J., Keller, J., Gunten, U.V., 2017. Kinetics and mechanisms of nitrate and ammonium formation during ozonation of dissolved organic nitrogen. Water Res. 108, 451-461.
+* [612] Wang, W., Han, H., Yuan, M., Li, H., Fang, F., Wang, K., 2011. Treatment of coal gasification wastewater by a two-continuous UASB system with step-feed for COD and phenols removal. Bioresource Technol. 102, 5454-5460.
+* [613] Wang, Z., Xu, X., Gong, Z., Yang, F., 2012. Removal of COD, phenols and ammonium from Lurgi coal gasification wastewater using A\({}_{\text{O}}\)-MBR system. J. Hazard. Mater. 78, 235-236.
+* [617] Wen, G., Ma, J., Liu, Z., Zhao, L., 2011. Ozonation kinetics for the degradation of phthalate esters in water and the reduction of toxicity in the process of \({\text{O}}_{\text{H}}{\text{Z}}_{\text{O}}{\text{Z}}_{\text{O}}{\text{Z}}_{ \text{J}}\). Hazard. Mater. 195, 371-377.
+* [618] Wert, E.C., Rosarioortiz, F.L., Snyder, S.A., 2009. Effect of ozone exposure on the oxidation of trace organic contaminants in wastewater. Water Res. 43, 1005-1014.
+* [621] Xu, P., Han, H., Zhuang, H., Hou, B., Jia, S., Xu, C., Wang, D., 2015. Advanced treatment of biologically pretreated coal gasification wastewater by a novel integration of heterogeneous Fenton oxidation and biological process, Bioresource Technol. 182, 389-392.
+* [624] Xu, P., Ma, W., Hou, B., Shi, Z., 2016. A novel integration of microwave catalytic oxidation and MBR process and its application in advanced treatment of biologically pretreated Lurgi coal gasification wastewater Separation & Purification,Technology. 177, 233-238.
+* [625] Yao, C.C.D., Haag, W.R., 1991. Rate constants for direct reactions of ozone with several drinking water contaminants. Water Res. 25, 761-773.
+* [626] Zhang, C., Li, M., Liu, G., Luo, H., Zhang, R., 2009. Pyridine degradation in the microbial fuel cells. J. Hazard. Mater. 172, 465-471.
+* [631] Zhang, T., Lu, J., Ma, J., Qiang, Z., 2008. Comparative study of ozonation and synthetic goethite-catalyzed ozonation of individual NOM fractions isolated and fractionated from a filtered river water. Water Res., 42, 1563-1570.
+* [632] Zhu, Q., Mogridge, G.D., Aime, M., Mantle, M.D., Gladden, L.F., D Agostino, C., 2016. Adsorption of pyridine from aqueous solutions by polymeric adsorbents MN 200 and MN 500. Part 1: Adsorption performance and PFG-NMR studies. Chem. Eng. J., 306, 67-76.
+* [633] Zhuang, H., Han, H., Hou, B., Jia, S., Zhao, Q., 2014. Heterogeneous catalytic ozonation of biologically pretreated Lurgi coal gasification wastewater using sewage sludge based activated carbon supported manganese and ferric oxides as catalysts. Bioresource Technol., 166, 178-186.
+
+**Highlights**
+
+1. The removal of indole was hardly affected by pH value.
+
+2. The reaction rate constants of ozone with pyridine and indole were calculated.
+
+3. Seven and five intermediates were found for pyridine and indole in ozone process.
+
+4. Primary products during first 10 min might lead to the obstinate toxicity.

manual_annotation/200pdfs_in_mmd/10.1016_j.desal.2010.11.013.mmd

@@ -0,0 +1,196 @@
+# Degradation kinetics of DEP in water by ozone/activated carbon process: Influence of pH
+
+Tatianne Ferreira de Oliveira, Olivier Chedeville, Benoit Cagnon, Henri Fauduet
+
+# Abstract
+
+Efficiency of the ozone (O3)/Activated Carbon (AC) process to remove phthalates was studied by investigating the degradation kinetics of diethylphthalate (DEP) chosen as model pollutant. The influence of pH was determined by performing experiments at five pH values (2.5 +- 7.2). The kinetic contribution of radical mechanisms (_d_(_t_)) was estimated by using a radical scavenger (terr-butyl alcohol). Single oxzonation of DEP was first performed. Pseudo-first order modelling shows that the reaction depends on pH, with kinetic constants varying from 0.0036 min-1 (pH = 2.5) to 0.6129 min-1 (pH = 7.2). Ozonation was then performed in the presence of a commercial AC (Pica L27), whose chemical and textural properties had been previously determined. Pseudo-first-order modelling shows a significant increase in DEP degradation kinetics: kinetic constants rose from 0.239 min-1 (pH = 2.5) to 0.862 min-1 (pH = 7.2). Estimation of the kinetic contribution of heterogeneous reactions to DEP removal (_d_(_t_)) shows that in acidic conditions, reactions are essentially located on the AC surface (_d_(_d_(_t_) = 98.7% at pH = 2.5) and occur in the bulk liquid when pH increases (_d_(_t_)) = 29.2% at pH = 7.5). Lastly, estimation of the kinetic contribution of radical mechanisms (_d_(_t_)) shows that the increase in the DEP degradation kinetics is mainly due to radical reactions, with _d_(_t_) increasing from 87.9% (pH = 2.5) to 92.0% (pH = 7.5). These radical reactions may be promoted by deprotonated acid groups present on the AC surface.
+
+## Introduction
+
+In recent years, the presence of micropollutants such as phthalates has increased in water and sediments, causing a major environmental concern. Phthalates are intensively used as plasticizers in industry or as additives in cosmetics or in printing inks, uses which require a world production of several million tonnes per year [1]. Recent studies show that many surface waters and sediments contain a high concentration of phthalates coming from industrial wastewaters [2,3]. These pollutants are not easily detectable and are suspected of being endocrine disruptors and/or carcinogenic compounds [4,5]. Given this potential health hazard, a tolerance limit of phthalates in industrial wastewater has therefore been fixed in Europe at 1.3 kg L-1 (diethylhexyl phthalate (DEHP) is the standard). A further stipulation is that a 30% reduction in phthalate concentrations in industrial wastewaters has to be achieved by 2015 (WFD, 2000/60/EC). With wastewater regulation becoming more stringent, it is necessary to develop new wastewater treatment methods and/or to optimize existing processes [6]. Ozone (O3)/activated carbon (AC) coupling is an advanced oxidation process (AOP), which appears to be a cost efficient process [7], that combines several actions: ozonation of compounds in the bulk liquid by direct and indirect pathways (homogeneous reactions), adsorption on AC, and direct and indirect ozonation of compounds on the AC surface (heterogeneous reactions). Moreover, this coupling enhances the generation of hydroxyl radicals (HO') by interaction between O3 and AC surface groups [8-10]. This method presents a great potential for the removal of micropollutants [11], but depending on the systems, on experimental conditions and on the molecules, the process cannot yet be easily controlled. It is therefore very important to understand the reaction mechanisms and to optimize the treatment, by promoting the generation of radicals resulting from O3/AC interaction.
+
+The aim of this work was to study the efficiency of the O3/AC process to remove diethylphthalate (DEP), chosen as the model pollutant, by investigating its degradation kinetics. The study was performed with a commercial activated carbon (Pica L27) whose textural and chemical properties had been previously determined. DEP removal kinetics were successively studied in the presence ofO\({}_{3}\) (single ozonation), and O\({}_{3}\)/AC. Moreover, the influence of pH was determined by performing experiments at five pH values (2.5\(<\)pH\(<\)7.5). To determine the nature (molecular or radical) of reactions, the same experiments were carried out with tert-butyl alcohol (tBuOH) as radical scavenger. For each experiment, the results were modelled with a first-order model in order to determine the kinetic contribution of the actions of O\({}_{3}\)/AC coupling.
+
+## Material and method
+
+### Experimental
+
+#### 2.1.1 Ozonation pilot
+
+DEP degradation kinetics were performed in a 1 L batch reactor (Fig. 1), surmounted by a dropping funnel (DF). The reactor was thermostated by a cryothermostat (C) and was mechanically stirred (M). The stirring mobile (impeller) and its dimensions (7 cm in diameter) were chosen to favour mass transfer between the three phases and to limit AC attrition. Ozone, produced from pure oxygen in a BMT S03 N ozone generator (O\({}_{3}\) C) supplied by the BMT company, was introduced through a porous diffuser (D). Ozone concentrations in the inlet and the outlet gas were measured with a BMT 964 ozone analyser (O\({}_{3}\) A). The outlet gas was sent to an ozone catalytic destructor (CD) to eliminate unconverted ozone. The gas circuit was deviced with teflon pipes. The valves, flowmeters and non-return valves, provided by the EM-Technik Company, were made of PFA. All these materials are inert with respect to ozone.
+
+#### 2.1.2 Kinetic experiments
+
+The reactor was filled with 500 mL of phosphate buffered solution, prepared with a mixture of H\({}_{3}\)PO\({}_{4}\) (purity up to 85%, obtained from Sigma Aldrich) and NaH\({}_{2}\)PO\({}_{4}\) (purity up to 99%, obtained from Sigma Aldrich) at pH 2.5, 3.5, 5.6, 6.2 or 7.2 and thermostated at 20 degC. For experiments performed with AC, 2 g of AC were introduced into the reactor. The dropping funnel was filled with a solution containing 0.05 g of DEP (purity up to 99.5%, obtained from Sigma Aldrich) dissolved in 250 mL of the same buffered solution. The initial DEP concentration was 0.2 g L\({}^{-1}\). For experiments performed with tBuOH (purity up to 99.5%, purchased from Across Organics), 0.5 g of this radical inhibitor was introduced into the dropping funnel with the solution containing DEP in buffered solution ([tBuOH] = 30[DEP]). O\({}_{3}\) was fed into the reactor during about 10 min to achieve saturation of the solution, monitored by the carmin indigo method proposed by Bader and Hoigne [12]. The gas flow rate was 40 NL h\({}^{-1}\) and O\({}_{3}\) concentration in the inlet gas was fixed at 50 g Nm\({}^{-3}\). This concentration was chosen according to ozone generator properties (O\({}_{3}\) concentration remains constant if this one is up to 50 g Nm\({}^{-3}\)) and experiments duration. When ozone saturation was obtained, the DEP solution was introduced into the reactor. Samples were withdrawn with a syringe at suitable time intervals and were filtered through a 0.45 mm membrane filter. A volume of 2 mL of filtered sample was introduced into 100 mL of sulphite solution (0.1 M) to remove dissolved ozone. The concentration of dissolved ozone in the reactor was monitored by the carmin indigo method to ensure that saturation was maintained throughout the experiment. The DEP concentration in samples was monitored by HPLC (Kontron 325 system) using a hypersi C18 column (250 mm long x 4.6 mm i.d, Thermo Scientific) with a UV detector (Spectra Physics 200) at 228 nm. Mobile phase was an acetonitrile/water mixture (70:30) introduced at a flowrate of 1 mL min\({}^{-1}\).
+
+### Determination of AC properties
+
+The AC used in this study (Pica L27) was a commercial activated carbon provided by Pica. It was washed before each experiment to eliminate any residual acidity due to its activation treatment, which could perturb the study of its intrinsical chemical properties effects on the process. The porosity of the adsorbent material was characterized through a conventional nitrogen adsorption isotherm at \(-\)196 degC (77 K) using a Micromeritics ASAP 2020. The sample was previously degassed at 250 degC for 48 h under a residual vacuum of less than 10\({}^{-4}\) Pa. The nitrogen adsorption isotherm was analyzed according to Dubinin's theory [13,14]. Both the specific microporous volume \(W_{0}\) (cm\({}^{3}\) g\({}^{-1}\)) and the mean pore size \(L_{0}\) (nm) were estimated from the linear part of the Dubinin-Radushkevich (D-R) plots [13]. The Sing \(\alpha\)s plots were also used to determine the specific external surface \(S_{\text{ext}}\) (m\({}^{2}\) g\({}^{-1}\)) [14], assuming that for slit-shaped micropores the specific microporous surface \(S_{\text{micro}}\) (m\({}^{2}\) g\({}^{-1}\)) could be estimated using the specific microporous volume and the mean pore size [13]. Boehm titration was used to determine the oxygen surface groups [15] and determination of the pHP\({}_{\text{ZC}}\) was obtained by the method proposed by Rivera-Utrilla [16].
+
+### Kinetic model
+
+DEP degradation by O\({}_{3}\)/AC coupling results from different effects. Reactions can occur in the bulk liquid (direct or indirect oxidation by ozone) or on the AC surface (adsorption, oxidation by O\({}_{3}\) or HO\({}^{\prime}\)). A simplified kinetic model taking all these effects into account was used to describe DEP adsorption [8, 9, 10].
+
+It is generally admitted that ozonation kinetics is correctly described by a second-order model. Therefore, direct (Eq. (1)) and indirect (Eq. (2)) ozonation kinetics were modelled as:
+
+\[-\frac{d[DEP]}{dt}=k_{1}\cdot[DEP]\cdot[O_{3}] \tag{1}\]
+
+\[-\frac{d[DEP]}{dt}=k_{2}\cdot[DEP]\cdot[HO^{\prime}] \tag{2}\]
+
+where [_DEP_] is the concentration of DEP (mol L\({}^{-1}\)), \(t\) is the time (s), [O\({}_{3}\)] is the concentration of dissolved ozone (mol L\({}^{-1}\)), [_HO\({}^{\prime}\)_] is the concentration of hydroxyl radicals (mol L\({}^{-1}\)), and \(k_{1}\) and \(k_{2}\) are the kinetic constants of respectively direct and indirect ozonation of DEP in the bulk liquid (L mol\({}^{-1}\) s\({}^{-1}\)).
+
+It was assumed that the O\({}_{3}\) concentration remained constant throughout the kinetic experiments (this assumption was verified by measuring the dissolved O\({}_{3}\) concentration during the experiments). This assumption means that no O\({}_{3}\) diffusion limitation occurs (chemical regime, _Ha_\(<\)3) [17]. According to Von Gunten [18], it can be assumed that hydroxyl radical concentration was also maintained constant during the experiments. For experiments performed only
+
+Figure 1: Ozzonation pilot. (O\({}_{3}\) A) ozone analyser, (DF) dropping funnel, (C) cyberthermal, (O\({}_{3}\)) ozone generator, (M) stirring motor, (D) porous diffuser, (CD) catalytic destructor.
+
+with O3 (homogeneous phase), the DEP degradation kinetics can be described by a pseudo-first-order model:
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{\mathit{homo}}[\mathit{DEP}] \tag{3}\]
+
+with:
+
+\[k_{\mathit{homo}}=k_{1}[\mathit{O}_{3}]+k_{2}[\mathit{HO}^{\prime}] \tag{4}\]
+
+From the experiments performed with O3 and tBuOH the kinetics of direct ozonation were determined, as radical reactions were scavenged. In accordance with Eq. (1), the degradation kinetics of DEP can be written as follows:
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{1}^{\prime}[\mathit{DEP}] \tag{5}\]
+
+with
+
+\[k_{1}^{\prime}=k_{1}[\mathit{O}_{3}] \tag{6}\]
+
+The kinetic contribution of the indirect pathway to DEP degradation by ozonation can be estimated by Eq. (7):
+
+\[\delta^{\mathit{02ind}}(\overline{x})=\frac{k_{\mathit{homo}}-k_{1}^{\prime}}{ k_{\mathit{homo}}}\cdot 100 \tag{7}\]
+
+In this study, in order to obtain a simplified model taking into account all the effects of O3/AC coupling, it was assumed that the adsorption kinetics could be modelled by a first-order model:
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{3}\cdot[\mathit{DEP}] \tag{8}\]
+
+where \(k\)3 is the kinetic constant of DEP adsorption (s-1).
+
+The DEP oxidation kinetics by O3 (Eq. (9)) or HO7 (Eq. (10)) on the AC surface (heterogeneous reactions) were modelled as follows:
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{4}\cdot[\mathit{DEP}]\cdot[\mathit{O}_{3}] \tag{9}\]
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{5}\cdot[\mathit{DEP}]\cdot[\mathit{HO}^{\prime}] \tag{10}\]
+
+where \(k\)4 and \(k\)5 are respectively the kinetic constants of direct and indirect oxidation of DEP on the AC surface (L mol-1 s-1). For experiments performed with O3 and AC, all the presented effects occur. In accordance with previous equations, the kinetics of DEP degradation can be written as in Eq. (11):
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{1}[\mathit{DEP}][\mathit{O}_{3}]+k_{2}[\mathit{ DEP}][\mathit{HO}^{\prime}]+k_{3}[\mathit{DEP}]+k_{4}[\mathit{DEP}][\mathit{O}_{3}]\] \[+k_{5}[\mathit{DEP}][\mathit{HO}^{\prime}] \tag{11}\]
+
+O3 concentration being maintained constant, Eq. (11) can be simplified to:
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{\mathit{global}}[\mathit{DEP}] \tag{12}\]
+
+with:
+
+\[k_{\mathit{global}}=k_{\mathit{homo}}+k_{\mathit{hetero}} \tag{13}\]
+
+\[k_{\mathit{hetero}}=k_{3}+k_{4}[\mathit{O}_{3}]+k_{5}[\mathit{HO}^{\prime}] \tag{14}\]
+
+The kinetic contribution of heterogeneous reactions to DEP degradation (_d_\(\delta^{\mathit{hetero}}\)) can be estimated by Eq. (15):
+
+\[\delta^{\mathit{hetero}}=\frac{k_{\mathit{global}}-k_{\mathit{homo}}}{k_{ \mathit{global}}}\cdot 100 \tag{15}\]
+
+For experiments performed with both O3, AC and tBuOH, radical reactions are scavenged and, in accordance with Eq. (11), DEP degradation kinetics can be modelled as in Eq. (16):
+
+\[-\frac{d[\mathit{DEP}]}{dt}=k_{\mathit{obs}}[\mathit{DEP}] \tag{16}\]
+
+with
+
+\[k_{\mathit{obs}}=k_{1}[\mathit{O}_{3}]+k_{3}+k_{4}[\mathit{O}_{3}] \tag{17}\]
+
+The kinetic contribution of radical reactions to DEP degradation (_d_\(\delta^{\mathit{HO}}\)) is estimated by Eq. (18):
+
+\[\delta^{\mathit{HO}}(\overline{x})=\frac{k_{\mathit{global}}-k_{\mathit{obs}}} {k_{\mathit{global}}}\cdot 100 \tag{18}\]
+
+## Results and discussion
+
+### Chemical and textural properties of AC
+
+The properties of L27 AC are presented in Table 1. These results show that this AC has a high content of acid functions (1.57 meg g-1) and few basic functions (0.18 meg g-1). Moreover, using Boehm's titration it was determined that L27 AC presents a high amount of carboxylic functions (0.81 meg g-1) and smaller amount of phenolic (0.30 meg g-1) and lactonic (0.46 meg g-1) functions. The acidic behaviour of L27 AC was confirmed by the determination of the pHrace value (pHrace = 3.0). Results obtained with N2 adsorption at 77 K show that this AC has a high specific microporous volume and external surface, which could favour intraparticle diffusion.
+
+### Ozonation of DEP
+
+Ozonation of DEP was performed at different pH values between 2.5 and 7.2. The evolution of DEP concentration during this treatment is presented on Fig. 2. Results show that DEP can be completely removed by ozonation but that the degradation kinetics depends strongly on the pH value. For example, after 10 min of treatment, DEP removal was about 4.3% at pH 2.5 and 100% at pH 6.2 and 7.2. Moreover, DEP degradation at pH 2.5 was 36.4% after 150 min of treatment (not shown here). The results of fig. 2 were used in Eqs. (3), (5) and (7) to determine the values shown in Table 2. The determination of _k_homo shows that the first-order model used correctly describes the ozonation kinetics of DEP (r2>0.987). The higher the pH value, the faster the degradation, with the kinetic constant _k_homo increasing from 0.0036 min-1 (pH 2.5) to 0.6129 min-1 (pH 7.2). The same phenomenon has also been observed with other non-dissociating organic compounds [19]. It could be explained by the formation of hydroxyl radicals resulting from the increase in interactions between O3 and HO- when pH increases.
+
+\begin{table}
+\begin{tabular}{c c c c c c c} pHrace & Total acid & Total basic & _S_{\mathit{mut}} & _S_{\mathit{mut}} & _W_{\mathit{in}} & _L_{\mathit{in}} \\  & (mg g−1) & (mg g−1) & (m2 g−1) & (m2 g−1) & (m2 g−1) & (m2 g−1) & (A) \\
+3 & 1.57+ & 0.18 & 4.44 & 6.16 & 0.57 & 18.5 \\ \end{tabular}
+\end{table}
+Table 1: Chemical and textural properties of L27 activated carbon.
+
+To determine the kinetic contribution of the radical mechanism in the DEP degradation by ozone, the same experiments were performed with the introduction of BtuOH as a radical scavenger. This compound was chosen because it is known to react slowly with O3 (_k_OX, m_OH_=3.10-2 L mol-1 s-1) and very rapidly with HO' (_k_OH, r_H_=0.5108 L mol-1 s-1) [19]. The values of \(k\)1 (Table 2) obtained from Eq. (5) (r2~0.985) show that the degradation kinetics of DEP is much slower in the presence of BtuOH, with \(k\)1 increasing from 0.0005 min-1 to 0.0098 min-1 when the pH rises from 2.5 to 7.2. The values of \(d\)03/nd (Eq. (7)), which increase from 86.1% (pH 2.5) to 98.4% (pH 7.2), confirm that during ozonation, DEP is mainly removed by the indirect pathway, even at acidic pH. The propagation of radical mechanisms by DEP itself could explain these results. According to Pi [20], the reaction between hydroxyl radicals (initiated by interaction between O3 and HO-) and the aromatic ring lead to the formation of H2O2 which has a great ability to generate radicals by interaction with O3.
+
+### 03/AC coupling
+
+Experiments were performed in the same experimental conditions and with the addition of 2.0 g of L27 AC. The results presented on Fig. 3 show that DEP degradation in the presence of AC is significantly faster than when only O3 was used, especially at acidic pH. In comparison with the results obtained with single ozonation, after 10 min of treatment, DEP removal by O3/AC coupling rose to 93% at pH 2.5. For all other pH, DEP was totally removed after 10 min of treatment. The results obtained from Eqs. (12) and (15) and presented on Table 3 show that the global first-order model used correctly describes DEP degradation kinetics (r2~0.987). These results confirm that the higher the pH value is, the faster the DEP degradation, with _k_global rising from 0.290 min-1 (pH 2.5) to 0.866 min-1 (pH 7.2). The degradation kinetics of DEP in the presence of O3 and AC, _k_global (Table 3), is significantly higher than _k_homo at each pH condition (Table 2), showing that the presence of AC promotes the degradation kinetics of DEP. Nevertheless, comparison with the results obtained by single ozonation shows that the effect of AC on DEP degradation is greater in the most acidic conditions and becomes weaker when the pH value increases. The determination of \(d\)0_hetero_ (Table 3) shows that DEP degradation is mainly located on the surface of AC at acidic pH (_d_hetero = 98.7% at pH 2.5) and is located in the bulk liquid when the pH rises (_d_hetero = 29.2% at pH 7.2). These results can be explained by the evolution of the ozonation kinetic constant with pH: when the pH decreases, ozonation becomes very slow (Fig. 2), whereas the presence of AC significantly accelerates the reaction rate. This phenomenon may be due to three effects: DEP adsorption and DEP ozonation on the AC surface by direct or indirect reactions.
+
+(pH 7.2). Thus, DEP is mainly degraded by radical mechanisms during treatment by the O3/AC process. This is confirmed by the values of _d_H0*, presented in Table 4, which range from 78.5% (pH 3.5) to 92% (pH 6.2). These results show that O3/AC coupling enhances the degradation kinetics of DEP by radical mechanisms. It appears that L27 AC, which presents an acidic surface, favours these mechanisms. According to Valdes [8], this may be due to the presence of deprotonated acid surface groups which can initiate radical reactions by interaction with O3. The high amount of these acid surface groups on L27 AC therefore leads to an efficient removal of DEP. Moreover, the great efficiency of O3/AC coupling could also be explained by the high microporous volume and external surface of L27 AC which favour intraparticle diffusion.
+
+## Conclusion
+
+The aim of this work was to study the efficiency of O3/AC coupling to remove DEP, chosen as a model pollutant in the degradation of phthalates. A kinetic study of DEP degradation was conducted at different pH conditions (2.5< pH<7.2). Single ozonation was first performed. It appeared that total degradation of DEP can be obtained, but that the degradation kinetics is strongly linked to pH. For example, after 10 min of treatment, DEP degradation is complete when the pH value is greater than 6.2, whereas it is less than 5% at pH = 2.5. It was shown that this phenomenon can be explained by HO7 generation resulting from interaction between O3 and HO-. Then, DEP removal by O3/AC coupling was performed. Results show that the presence of AC significantly enhances DEP degradation, especially in the most acidic conditions. Results were correctly modeled by a global pseudo-first-order model, which made it possible to estimate the nature and the location of the reaction. It was shown that in all experimental conditions, DEP was mainly removed by radical reactions (78.5<_d_H0* (%)<89.5). Given the oxidative potential of hydroxyl radicals, O3/AC coupling, which has been demonstrated to achieve fast and complete DEP removal, could also completely mineralize this compound. Moreover, it has also been shown here that the more acidic the pH, the more reactions on AC occur. To obtain an efficient wastewater treatment process, it would be necessary to study and optimize it by studying the influence of some parameters (temperature, ratio of AC dose and O3 amount, pollutant concentration). Moreover, it would be interesting to determine the influence of textural and chemical properties on radical generation.
+
+## Acknowledgments
+
+The authors wish to thank Xavier Bourrain and the Agence de l'Eau Loire Bretagne for their technical and financial support and Pica S.A for gratuitously supporting AC.
+
+## References
+
+- A. Gomez-Hens, M.P. Aguilar-Caballos, Social and economic interest in the control of phthalic acid esters, Trends Anal. Chem. 22 (2003) 847-857.
+
+- M. Munzer, J. Thewirch, D. Bahmenan, Titanium dioxide mediated photocatalytic degradation of 1, 2-dimethyl phthalate, J. Photochem. Photobiol. A: Chem. 143 (2003) 213-219.
+
+- I. Hai-Yan, J. Qu, H. Liu, Removal of a type of endocrine disruptors
+- di-n-butyl phthalate from water by ozonation, J. Environ. Sci. 18 (2006) 845-851.
+
+- A.P. Wexel, P. Van Vaardinger, R. Posthunus, GH. Cromentuijt, D.T.H.M. Sjinn, Environmental risk limits for two phthalates, with special emphasis on endocrine disruptive properties, Ecotvino. Exviron. Sci. 46 (2000) 305-325.
+
+- M. Witrasek, J. Angerov, M. Koloss-Gehring, S.D. Schafer, W. Klockenbach, L. Dobler, AK. Gunsel, A. Muller, G.A. Wiessmuller, Fetal exposure to phthalates
+- a pilot study, International J. Phys. Environ. Health 212 (2009) 492-498.
+
+- H. Deboulil, O. Chedeville, B. Cagnon, V. Cajqueret, C. Dorte, Influences of pH, temperature and activated carbon properties on the interaction ozone/activated carbon for a wastewater treatment process, Des. 254 (2010) 12-16.
+
+- T.A. Kurianwan, W. Lo, G.Y.S. Chan, Degradation of reactant compounds from stabilized landfill leachate using a combination of ozone-GAC adsorption treatment, J. Haz. Mat. 137 (2006) 443-455.
+
+- H. Valdes, A.C. Zorst, Heterogeneous and homogeneous catalytic ozonation of benzothiazole promoted by activated carbon: kinetic approach, Chem. 65 (2006) 1131-1136.
+
+- P.C. Faria, M.F.R. Pereira, J.J.M. Orlso, Ozone decomposition in water catalyzed by activated carbon: influence of chemical and textural properties, Ind. Eng. Chem. Res. 45 (2006) 2715-2721.
+
+- M. Sanchez-Polo, R. Leyva-Ramos, J. Rivera-Utrilla, Kinetics of 1, 3, 6-naphthalet-nertsulphonic acid donation in presence of activated carbon, Original Carbon 43 (2005) 962-969.
+
+- M. Sanchez-Polo, J. Rivera-Utrilla, Ozonation of naphthalenetterisulphonic acid in the presence of activated carbons prepared from petroleum coke, App. Catal. B 67 (2006) 113-120.
+
+- Baderft,. Hoingel, Determination of ozone in water by the indigo method, Wat. Res. 15 (1981) 449-456.
+
+- H. Stocelli, Porosity in carbon, in: J. Patrick (Ed.), Arnold, London, 1995, pp. 67-92.
+
+- H.F. Stocelli, M.V. Lopez-Ramon, D. Hugi-Cleary, A. Guillot, Microgore sizes in activated carbons determined from the Dubinin-Radushkevich equation, Carbon 39 (2001) 1115-1116.
+
+- K.S.W. Sing, D.H. Everett, R.A.W. Haul, I. Moscou, RA. Pierotti, J. Rouquerol, T. Siemienewska, Reporting physisorption data for gas/solid systems, IUPAC, Pure Appl. Chem. 57 (1985) 603.
+
+- H.P. Boehm, Surface oxides on carbon and their analysis: a critical assessment, Carbon 40 (2002) 145-149.
+
+- J. Rivera-Utrilla, M. Sanchez-Polo, Ozonation of 1, 3, 6-naphthaleternisulphonic acid catalysed by activated carbon in aqueous phase, Appl. Catal. B: Environ. 39 (2002) 319-329.
+
+- O. Chedeville, M. Deboac, M. Ferrante Almanza, C. Porte, Use of an ejector for phenol containing water treatment by ozonation, Sep. Purit. Technol. 57 (2007) 201-208.
+
+- U. Von Gutten, Ozonation of drinking water: Part I. Oxidation kinetics and product formation, Wat. Res. 37 (2003) 1443-1467.
+
+- J. Hoingel, H. Bader, Rate constants of reactions of ozone with organic and inorganic compounds in water
+- 1: non-dissociating organic compounds, Wat. Res. 17 (1983) 173-183.
+
+- Y. Pi, J. Schumacher, M. Jekel, Decomposition of aqueous ozone in the presence of aromatic organic solutes, Wat. Res. 39 (2005) 83-88.
+
+\begin{table}
+\begin{tabular}{c c c c} pH & \(k_{\text{ads}}^{-1}\) (min) & r2 & \(\delta^{\text{in}}\)O (\%) \\
+2.5 & 0.035 & 0.978 & 87.9 \\
+3.5 & 0.035 & 0.992 & 78.5 \\
+5.6 & 0.074 & 0.993 & 85.8 \\
+6.2 & 0.073 & 0.986 & 92.0 \\
+7.2 & 0.091 & 0.995 & 89.5 \\ \end{tabular}
+\end{table}
+Table 4: Experimental determination of rate constant in the degradation kinetics of DEP by the O3/AC process in the presence of radical scavenger ((Ba0H)).

manual_annotation/200pdfs_in_mmd/10.1016_j.dyepig.2006.01.005.mmd

@@ -0,0 +1,153 @@
+Oxidative decomposition of Acid Brown 159 dye in aqueous solution by H2O2/Fe2+ and ozone with GC/MS analysis
+
+Wojciech K. Jozwiak
+
+Magdalena Mitros
+
+Jounna Kaluzna-Czaplinska
+
+Ryszard Tosik
+
+Institute of General and Ecological Chemistry, Technical University of Lodz, 90-924, Lodz, 116 Zeremskiego, Poland
+
+Received 11 September 2005; received in revised form 20 December 2005; accepted 13 January 2006
+
+###### Abstract
+
+The oxidative degradation of metal-complex Co(II) Acid Brown 159 in aqueous solution by H2O2/Fe2+ and ozone was investigated. Optimal conditions for decolorization and total decomposition processes were found for dye oxidation during H2O2/Fe2+ and ozone treatments. The full 100% decolorization is attained during first minutes of H2O2/Fe2+ and ozone oxidation processes. The complete mineralization of aqueous solutions of Acid Brown 159 dye cannot be achieved even under the optimal reaction conditions. Fenton's process appeared to be more effective in comparison with ozonation. A tentative mechanistic pathway for the oxidative degradation of Acid Brown 159 dye in aqueous solution was postulated.
+
+c 2006 Elsevier Ltd. All rights reserved.
+
+Fenton process; Ozonation; Metal-complex azo dye; Oxidation; Decolorization; Decomposition +
+Footnote †: journal: Accepted for publication in _The Journal of Chemical Biology_
+
+## 1 Introduction
+
+Metal complexes of monoazo compounds are principally useful as trivalent chromium and cobalt complexes for dyeing of protein and polyamide fibers [1].
+
+Metal-complex dyes are very versatile in terms of applications. Virtually all substrates, apart from a few synthetic fibers, can be dyed and printed with this class of dyes. Countless shades from greenish yellow to deep black can be generated, depending upon the metal, the dye ligands, and the combination of dye ligands in mixed complex dyes [2]. In commercial terms the most important chelated metals are chromium, cobalt, copper, iron, and nickel. The resulting dyeing of chromium and cobalt complex dyes is in general dull but exhibits a high standard of fastness, particularly light fastness. Because of the dullness, these metal complexes are chiefly used for deep colors procedure for which a large amount of dye must be applied. Their use is restricted to nitrogen-containing substrates, such as wool, nylon, and leather, since they have only a little affinity to cellulosed fibers. However, with the advent of reactive dyes, chromium and cobalt complex dyes containing a fiber-reactive group also find application in cellulose dyeing.
+
+Acid Brown 159 is one of the metal-complex derivatives belonging to the overwhelming majority of synthetic dyes currently used in the industry. The impact and toxicity of dyes that are released in the environment have been very important and extensively studied [3]. Our knowledge concerning dyes behavior in the environment and health hazards involved in their use is still incomplete. Standard wastewater treatments appeared ineffective because of the chemical stability of most dye pollutants that makes them non-biodegradable. A wide range of methods have been developed for the removal of synthetic dyes from waters and wastewaters to decrease their impact on the environment. The applied methods involve adsorption, decolorization by photocatalysis, and/or by oxidation process, biological decomposition, etc. [3]. The efficiency of various methods of dye removal, such as chemical precipitation, chemical oxidation, adsorption and their effects on biological treatment were reported in earlier papers [4,5]. Strongly dependent on oxidant type, chemical oxidation pretreatment is a prerequisite condition for the subsequent activated sludge process [6].
+
+Hydrogen peroxide can effectively decolorize dye from wastewaters in the presence of Fe(II) sulfate [7,8]. Optimal conditions for decolorization were found to be different for each type of dye, indicating that the development of a general oxidation method for a mixture of dyes would be very difficult. Thus, compromise must be made that is suitable for the decomposition of each dye at a reasonable oxidation rate [9]. UV/H\({}_{2}\)O\({}_{2}\) treatment can be successfully used for the decolorization of acid, basic and reactive dyes [10]. The influence of operating parameters on the decolorization of reactive dye by ozone has been studied in detail [11]. The results indicated that the decomposition rate increased with increasing pH and temperature [12].
+
+This paper is devoted to Acid Brown 159 in aqueous solution neutralized by advanced oxidation process (AOP) involving ozonation (O\({}_{3}\)) and Fenton (H\({}_{2}\)O\({}_{2}\)/Fe\({}^{2+}\)) processes. The operating parameters such as oxidant dosages, initial dye concentration, pH of solution and reaction time were determined to find optimum conditions for complete decolorization and total oxidation of dye solution. The toxicity of initial and final solutions was determined using Texoallert tests. Also, on the basis of GC/MS analysis a tentative mechanistic approach of Acid Brown 159 dye decomposition in aqueous solution is postulated.
+
+## Experimental
+
+### Materials
+
+Acid Brown 159 is a synthetic mixture of two Co-complex dyes with a molar ratio equal to 1:2; complex A about 70%, molecule of 2-aminophenyl-4-sulphonoamid coupling with 1-phenyl-3-methylpirazol-5 and complex B about 30%, molecule of 2-aminophenyl-4-sulphonoamid coupling with 2-naphthol. This dye was obtained from Dyestuff Industry Works "Polfa-Pabianice" in Poland and it is applied in tanning and textile industries. Molecular formulas of these two complexes of Acid Brown 159 dye are as follows:
+
+### Apparatus and procedure
+
+The H\({}_{2}\)O\({}_{2}\)/Fe\({}^{2+}\) Fenton oxidation experiments were carried out in a reactor with constant volume. Concentration of dye in demineralized water, 100 mg/dm\({}^{3}\), was applied. Varied doses of FeSO\({}_{4}\) compound were added to the reactor, which contained dye solution and then pH was adjusted by means of H\({}_{2}\)SO\({}_{4}\) and NaOH solutions. After stirring the reactor content, the initial UV absorbance (\(A_{0}\)) of the solution was determined at the specified wavelength 458 nm. Then the specified amount of oxidant H\({}_{2}\)O\({}_{2}\) was added to the reactor and the influence of reaction time on decolorization (UV absorbance \(A\)) and chemical oxygen demand (COD) were measured. From the measured absorbance values percentage of decolorization was calculated according to the formula: \(\alpha\) = (1 - \(A/A_{0}\))100%. Experiments of Acid Brown 159 oxidation by hydrogen peroxide were carried out at 40 \({}^{\circ}\)C.
+
+The ozonation experiments were carried out in a reactor containing aqueous solution of dye, 100 mg/dm\({}^{3}\), saturated in oxygen containing about 17 ppm of ozone with volume velocity 15 dm\({}^{3}\)/h. In the course of the experimental run the following parameters were measured: concentration of ozone, redox potential, pH, temperature, time and UV absorbance.
+
+Total organic content (TOC) was determined in TOC 505 Shimadzu analyzer. The FT-IR spectra of Acid Brown 159 dye compound were recorded on a Shimadzu 8150 spectro-photometer over the range 4000\(-\)400 cm\({}^{-1}\) using pellets of solid dye and KBr mixture. The concentrations of Co and Fe were analysed by ICP method.
+
+### GC/MS analysis
+
+Samples were prepared by the extraction of dye decomposition products with hexane and tetrachloromethane solvents and finally solutions were concentrated in SPE (NEXUS, Varian) apparatus with octadecyl columns. An Agilent 6890N gas chromatograph (GC) with 5973 mass-selective (MS) detector and Chem Station data system was employed to identify the intermediates of dye Acid Brown 159 solution degradation. The capillary column HP-5MS (cross-linked 5% phenyl-methylsilicone) was 30 m \(\times\) 0.25 mm \(\times\) 0.25 mm. The temperature ramp was programmed from 40 \({}^{\circ}\)C to 80 \({}^{\circ}\)C with linear rate 10 \({}^{\circ}\)C/min, then to 280 \({}^{\circ}\)C at 4 \({}^{\circ}\)C/min. Helium was used as the carrier gas with a volume flow 0.9 cm\({}^{3}\)/min. The MS quadrupole of mass analyser, ion source temperature was 150 \({}^{\circ}\)C, ionization at 70 eV. Comparison of the experimental mass spectra with those stored in NIST 98 Library identified GC of fragmentation products of dye degradation.
+
+### The test for the inhibition of oxygen consumption by activated sludge
+
+The test for the inhibition of oxygen consumption by activated sludge was performed on the basis of oxygen consumption rate changes at the constant temperature 20 \({}^{\circ}\)C. The oxygen consumption rate was measured in an OxiTop OC 100 device. Total volume of the solution was equal to 432 cm\({}^{3}\) in all the experiments. The oxygen consumption in dye solution was measured as a function of dye amount. The special program (EN ISO 8192) was used according to the formula: \(I\) = (\(R_{\rm B}-R_{\rm T}\))/\(R_{\rm B}\), where \(R_{\rm B}\) and \(R_{\rm T}\) are the rates of oxygen consumption of blank and controlled samples, respectively. On base of the obtained value \(I\) an influence of the examined dye solution on the oxygen consumption was estimated as inhibition when \(I\)  0.1 or stimulation when \(I\)  0.1.
+
+## 3 Results and discussion
+
+### 3.1 IR spectra of dye Acid Brown 159 identification
+
+The metal-complex dye Acid Brown 159 compound can be identified by its characteristic IR spectrum, presented in Fig. 1. The assignment of registered IR bands is as follows. In the range of 2800\(-\)3620 cm\({}^{-1}\) the broad band and small features can be assigned to hydrogen bonded OH and NH stretching vibrations. The most characteristic bands are ring stretching C==C and C==N vibrations appearing in the range of 1600\(-\)1520 cm\({}^{-1}\), doublet of C==C skeletal vibrations of the aromatic ring in the range 1500\(-\)1400 cm\({}^{-1}\), singlet of C\(-\)CH\({}_{3}\) vibrations at 1380 cm\({}^{-1}\). In the range of 1360\(-\)1260 cm\({}^{-1}\) stretching "out of the plane" vibrations C\(-\)C, and in 1160\(-\)1120 cm\({}^{-1}\) C\(-\)H stretching vibrations appear. In the range of 1080\(-\)1040 cm\({}^{-1}\) coupling conjugation between S, O and N is observed. The characteristic "out of the plane" hydrogen deformation modes in naphthalene ring located in the range of 720\(-\)480 cm\({}^{-1}\) are overlapped with those characteristic of Co\(-\)O interactions in 550\(-\)400 cm\({}^{-1}\), and also "out of the aromatic ring plane" vibrations are overlapped with C\(-\)S and C\(-\)N vibrations.
+
+### 3.2. Oxidation
+
+The influence of the experimental parameters such as hydrogen peroxide concentration (mg of H\({}_{2}\)O\({}_{2}\) per 1 mg of a dye), Fe\({}^{2+}\) ion concentration and pH on the effectiveness of solution decolorization is presented in Fig. 2. The degree of decolorization higher than 80% can be easily achieved after 2 h of dye solution oxidation. The effect of pH of the reaction habitat ranging from 1 to 5 was investigated. The degree of dye decolorization higher than 70% (curve a) was achieved in pH range 2\(-\)3. The effect of ferrous ions was investigated in the concentration range from 0 to 50 mg/dm\({}^{3}\) (curve b).
+
+Figure 1: FT-IR spectra of Co-complex dye Acid Brown 159 compound.
+
+Figure 3: COD values as a function of oxidation time of Acid Brown 159 dye solution by H\({}_{2}\)O\({}_{2}\)/Fe\({}^{2+}\).
+
+Figure 2: Dependence of the degree of Acid Brown 159 dye solution decolorization on H\({}_{2}\)O\({}_{2}\) dose, Fe\({}^{2+}\) concentration and pH value of aqueous solution after 2 h of oxidation.
+
+The process of decolorization does not take place in the absence of Fe\({}^{2+}\) ions. The efficiency was the highest when the concentration of Fe\({}^{2+}\) ions was in the range 5\(-\)50 mg/dm\({}^{3}\). The increase in H\({}_{2}\)O\({}_{2}\) concentration up to 2 mg per 1 mg of dye leads to 100% complete decolorization (curve c) at 40 \({}^{\circ}\)C. Thus, the following parameters were chosen as optimal conditions for further experiments of Acid Brown 159 oxidation: hydrogen peroxide concentration with the ratio of 2 mg of H\({}_{2}\)O\({}_{2}\) per 1 mg of a dye, Fe\({}^{2+}\) ion concentration 10 mg/dm\({}^{3}\) and pH = 2.5.
+
+The application of the above optimal conditions of dye solution oxidation leads to the results of the influence of reaction time on chemical oxygen demand (COD) values which are presented in Fig. 3. After 5 h of oxidation reaction the reduction of initial COD value was about 50% and after 24 h more than a 70% decrease in COD value was observed.
+
+The results of ozonation for Acid Brown 159 in aqueous solutions with concentration of 100 mg/dm\({}^{3}\) are presented in Fig. 4. The time necessary to achieve nearly full decolorization \(\alpha\) = 100% was 7 min. Also, the ozone consumption and redox potential are included in Fig. 4.
+
+The comparative effects of the oxidation by hydrogen peroxide and ozone on the changes of chemical oxygen demand (COD) and the total organic carbon (TOC) in aqueous Acid Brown 159 dye solution are presented in Table 1. Results of TOC analysis indicate that organic compounds are not completely oxidized to final products of oxidation, H\({}_{2}\)O and CO\({}_{2}\). The decreases in TOC values were 90.4% and 54.1% after 2 h of oxidation by H\({}_{2}\)O\({}_{2}\) and ozone, respectively. The reduction of COD values was about 73.7% for hydrogen peroxide and 56.1% for ozone treatment.
+
+### UV\(-\)vis spectra
+
+The comparison of UV\(-\)vis spectra of Acid Brown 159 in aqueous solutions before and after 2 h of oxidation by hydrogen peroxide and ozone is presented in Fig. 5. Both processes were performed in optimal conditions, which were chosen after preliminary measurements. The spectra of dye initial solution represent UV bands characteristic of \(-\)N\(=\)N\(-\) groups (514 cm\({}^{-1}\)) and related it to the benzene and naphthalene rings bonded to the \(-\)N\(=\)N\(-\) groups (220 and 322 cm\({}^{-1}\)) [13\(-\)15]. Absorption bands in visible region (458, 557 nm) are relevant to the whole conjugated structure. Thus, the \(\pi\)\(-\)\(\pi\)* transition of electrons in the azo group connecting phenyl and naphthyl rings is responsible for the band. Within near ultraviolet region
+
+\begin{table}
+\begin{tabular}{l l l l l l l l l l} Dye (Acid Brown 159) & Before oxidation & After oxidation by H\({}_{2}\)O\({}_{2}\) & After oxidation by O\({}_{3}\) & H\({}_{2}\)O\({}_{2}\) & & O\({}_{3}\) & \\ \hline COD & TOC & COD & TOC & COD & TOC & \(\eta_{\mathrm{COD}}\) & \(\eta_{\mathrm{TOC}}\) & \(\eta_{\mathrm{COD}}\) & \(\eta_{\mathrm{TOC}}\) & \(\eta_{\mathrm{TOC}}\) \\ (mg of O\({}_{2}\)/dm\({}^{3}\)) & (ppm) & (mg of O\({}_{2}\)/dm\({}^{3}\)) & (ppm) & (mg of O\({}_{2}\)/dm\({}^{3}\)) & (ppm) & (\%) & (\%) & (\%) & (\%) \\ \hline
+114 & 36 & 30 & 3.5 & 50 & 16.6 & 73.7 & 90.4 & 56.1 & 54.1 \\ \hline \end{tabular}
+\end{table}
+Table 1: An effect of the oxidation by hydrogen peroxide and ozone on changes of chemical oxygen demand (COD) and the total organic carbon (TOC) in aqueous Acid Brown 159 dye solution
+
+Figure 4: Dependence of the ozonation parameters of dye solutions on ozonation time. (a) \(\alpha\)\(-\) Degree of decolorization (%), (b) \(E\)\(-\) redox potential (mV), and (c) Z\({}_{\mathrm{O_{2}}}\)\(-\) ozone consumption (mg/dm\({}^{3}\)).
+
+(260 nm), the absorption band results from the un-saturated character of benzene and naphthalene rings. The dramatic changes of UV spectra represent a disappearance of both azo and aromatic groups in the course of Acid Brown 159 dye degradation. Ozone treatment appears much more efficient than Fenton's process but this statement refers to the decolorization rather than to the process of the entire oxidation (compare COD and TOC values in Table 1).
+
+### The inhibition test
+
+The relationship \(I\) versus log(\(\Delta_{\text{COD}}\)) for the Acid Brown 159 dye in solutions before and after hydrogen peroxide or ozone oxidation is illustrated in Fig. 6. It is rather obvious that in the case of hydrogen peroxide oxidation the measured values \(I\) were about 0.1. After ozone oxidation a different range of value \(I\) from 0.1 to 0.5 is observed. Acid Brown 159 solution before oxidation gave value _I_\(\ll\) 0.1. On the basis of the above behavior one can estimate a dye concentration which is toxic for microorganisms of activated sludge and an adequate value I \(\gg\) 0.1 is observed.
+
+### GC/MS of dye Acid Brown 159
+
+The main intermediate compounds of dye Acid Brown 159 degradation were identified by GC/MS method and they are presented as GC chromatogram in Fig. 7. The major peaks were assigned to the most important compounds and they are listed in Table 2 and also their most significant ions are included according to the decreasing order of ion intensities in Fig. 7a. The long chains of carbons and alcohols obtained after hydrogen peroxide oxidation are shown in Fig. 7b, whereas aldehyde and conjugated acid compounds are presented in Fig. 7c after ozone oxidation.
+
+One can anticipate that at the very beginning of dye molecule degradation process the rupture of azo (\(-\)N\(=\)N\(-\)) and metal oxygen (Co\(-\)O) groups takes place. Further, oxidation probably occurs via a variety of amine and/or nitro type of aromatic compounds formation, which are gradually transformed into long linear compounds as a result of aromatic ring opening by hydroxyl radical, originating from initial oxidation agents, H\({}_{2}\)O\({}_{2}\) or O\({}_{3}\). The results of GC/MS analysis show linear acids and aldehydes after Fenton's oxidation and alcohols and hydrocarbons in the case of ozonation process. The advanced oxidation can lead to deep mineralization of dye solution. The removal of cobalt and/or iron ions from dye solutions can be effectively accomplished in the form of precipitated metal hydroxides during alkalization process. A general oxidative decomposition of Acid Brown 159 dye pathway mechanism is schematically presented in Fig. 8.
+
+## 4 Conclusion
+
+Optimal conditions for decolorization and total decomposition processes were found for dye oxidation during H\({}_{2}\)O\({}_{2}\)/Fe\({}^{2+}\) and ozone treatments. An entire decolorization of Acid Brown 159 dye in aqueous solution during the first minutes of oxidation process was obtained.
+
+The complete mineralization of Acid Brown 159 dye solution cannot be achieved even under the optimal reaction conditions. Fenton's process appeared more effective than ozonation in deep oxidation of dye solution. A tentative mechanistic pathway for the oxidative degradation of Acid Brown 159 dye in aqueous solution was postulated.
+
+Figure 5: UV\(-\)vis spectral changes of Acid Brown 159 solutions before and after 2 h of oxidation by H\({}_{2}\)O\({}_{2}\) and O\({}_{3}\).
+
+Figure 6: The inhibition \(I\) of log(\(\Delta_{\text{COD}}\)) for Acid Brown 159 dye solutions before and after hydrogen peroxide and ozone oxidation.
+
+\begin{table}
+\begin{tabular}{l l l l l l} No. & Retention time & Compound identified & Formula & \(M_{\rm W}\) & The major \\  & on GC & by MS & & ions (_mL_) \\ \hline
+1 & 4.87 & Acetophenone & C\({}_{3}\)H\({}_{4}\)O & 105 & 105, 77, 120, 51, 50 \\
+2 & 5.01 & 1-methyl-1H-benzimidazol-2-amine & C\({}_{3}\)H\({}_{8}\)N\({}_{3}\) & 147 & 147, 119, 105, 90, 77 \\
+3 & 6.66 & methyl N-hydroxybenzene carbvimidoate & C\({}_{3}\)H\({}_{8}\)NO\({}_{2}\) & 133 & 133, 151, 73, 55, 68 \\
+4 & 8.37 & N-(3,4-dimethyl-2,6-dinitrobenzoyl) pentan-3-amine & C\({}_{13}\)H\({}_{19}\)N\({}_{3}\)O\({}_{4}\) & 252 & 281, 162, 192, 208 \\
+5 & 53.16 & N-methylloaniline & C\({}_{2}\)H\({}_{8}\)N & 106 & 106, 77, 79, 51, 65 \\ \end{tabular}
+\end{table}
+Table 2: GC/MS analysis of fragmentation products
+
+Fig. 7: Total ion GC/MS chromatogram of Acid Brown 159 oxidative decomposition products. (a) Before oxidation, (b) after H\({}_{2}\)O-oxidation, and (c) after O\({}_{3}\)oxification. The major peaks are identified in Table 2.
+
+The oxidation by hydrogen peroxide had no serious impact on changes of toxicity features of activated sludge. More toxic compounds were formed in ozone oxidation process.
+
+## Acknowledgements
+
+This study was supported by the Polish State Committee for Scientific Research under Grant No. 3 T09 B 126 27.
+
+## References
+
+* (1) Schwarzenbach G. Helv Chim Acta 1952;35:2344.
+* (2) Zollinger H. Diazo chemistry I: aromatic and heteroaromatic compounds. Weinheim: VCH; 1994.
+* (3) Forges E, Cserha'i T, Oros G. Removal of synthetic dyes from wastewaters: a review. Environ Int 2004;30:953-71.
+* (4) Guaratini CCI, Zanoni MVB. Textile dyes. Quinn Nova 2000;23:71-8.
+* (5) Oros G, Cserha'i T, Forgacs E. Strength and selectivity of the fungicidal effect of diazobenzene dyes. Fresenius Environ Bull 2001;10: 319-22.
+* (6) Shaul GM, Holdsworth TJ, Dempsey CR, Dostal KA. Fate of water soluble azo dyes in the activated sludge process. Chemosphere 1999;22:107-19.
+* (7) Kuo WG. Decolorizing dye wastewater with Fenton's reagent. Water Res 1992;26:881-6.
+* (8) Shu Hung-Yee, Chang Ming-Chin, Fan Huau-Jung. Decolorization of azo dye acid black 1 by the UV/H\({}_{2}\)O\({}_{2}\) process and optimization of operating parameters. J Hazard Mater 2004;B113:201-8.
+* (9) Tang WZ, Chen RA. Decolorization kinetics and mechanism of commercial dyes by H\({}_{2}\)O\({}_{2}\)/iron powder system. Chemosphere 1996;32: 1947-58.
+* (10) Yang YQ, Wyatt DT, Baharshy M. Decolorization of dyes using UV/H\({}_{2}\)O\({}_{2}\) photochemical oxidation. Text Chem Color 1998;30:27-35.
+* (11) Hitchcock DR, Law SE, Wu JN, Williams PL. Determining toxicity trends in the ozonation of synthetic dye wastewaters using the nematode _Caenorhabditis elegans_. Arch Environ Contam Toxicol 1998;34: 259-64.
+
+Figure 8: Tentative pathway of Acid Brown 159 oxidative degradation.
+
+* [12] Wu JN, Wang TW. Effects of some water-quality and operating parameters on the decolorization of reactive dye solutions by ozone. J Environ Sci Health 2001;36(Part A):1335-47.
+* [13] Wang A, Qu J, Liu H, Ge J. Degradation of azo dye Acid Red 14 in aqueous solution by electrokinetic and electrooxidation process. Chemosphere 2004;55:1189-96.
+* [14] Solozhenko EG, Soboleva NM, Goncharuk VV. Decolourization of azodye solutions by Fenton's oxidation. Water Res 1995;29:2206-10.
+* [15] Wu F, Deng NS, Hua HL. Degradation mechanism of azo dye C.I. Reactive Red 2 by iron powder reduction and photooxidation in aqueous solutions. Chemosphere 2000;41:1223-38.

manual_annotation/200pdfs_in_mmd/10.1016_j.envadv.2020.100017.mmd

@@ -0,0 +1,165 @@
+Elimination of volatile organic compounds in paint drying by absorption reaction in water combined with the ozone oxidation technique
+
+Juan Martin Alvarez, Carlos J. Seijas, Gustavo L. Bianchi
+
+# Abstract
+
+The aim of this work was to evaluate the treatment of Volatile Organic Compounds (VOCs) released during the painting and drying process of an industrial paint by absorbing them in a water curtain and then oxidizing them with ozone. For this purpose, laboratory equipment was built consisting of a paint drying cabin and a bubble column where the VOCs were absorbed and treated. Tests were carried out with simulated effluents using distilled water and service water, to which 3.5 g/L of formaldehyde, 11.6 mg/L of butyl acetate, 5.3 mg/L of ethanol and 5.6 mg/L of methyl ethyl ketone were added, according to data provided by the paint manufacturer; these tests were then repeated using the industrial paint currently used in the processes. The water saturation with the VOCs and the simultaneous treatment by means of ozone dosage was studied, taking the Chemical Oxygen Demand (COD) value as a treatment effectiveness parameter. VOC saturation is achieved for exposure times between 5 and 15 min in the aqueous effluent. The system of absorption and ozonation is effective for the elimination of the released VOCs, achieving a 94% reduction in the COD of the wastewater. Finally, it was observed that when O3 is injected together with the paint, the reduction of COD in the effluent is favored due to a first instance of direct ozonation of the VOCs, and a second, slower indirect stage, generated by the presence of free radicals from the reaction of O3 with water [hydroxy] radical (HO**), superoxide anion radical (O2***) and hydroperoxyl radical (HO**2***)] favored by the pH of the solution. The reaction kinetics of VOC oxidation is of first order.
+
+## Introduction
+
+Volatile organic compounds (VOCs) are those organic compounds that have a Reid vapour pressure of more than 10.3 Pa under normal temperature and pressure conditions, which encompasses a group of carbon-based chemicals that evaporate easily at room temperature (Kamal et al., 2016; Ojala et al., 2011). VOCs emissions can come from a wide range of sources including chemical industries, paper production, food processing, paint drying, transportation, oil refineries, petrochemicals, etc. (Morin et al., 2019; Zhang et al., 2016; Drobek et al., 2015). Typically, aldehydes, aromatics and halo-hydrocarbons are the most common contaminants in the large family of industrial VOCs.
+
+VOCs emission can be controlled using methods based on recovery and destruction. Recovery-based techniques include absorption, adsorption, membrane separation and condensation. According to Kujawa (2015) the high cost of adsorbents and the need for their frequent regeneration are the main limitations of the adsorption processes. In methods based on destruction through oxidative processes, VOCs are converted to carbon dioxide and water. Within the oxidative processes, there are several methods such as catalytic oxidation (Wang et al., 2017; Chen et al., 2020, 2020; Shu et al., 2018; Li et al., 2018, 2018), photocatalytic oxidation (Deng et al., 2017; Krichevskaya et al., 2017) and plasma oxidation (Zhu et al., 2015; Chang et al., 2019), which have given positive results in the treatment of VOCs.
+
+According to Zhu et al. (2017)Zhu (2017), the catalytic ozone oxidation method is a promising alternative technique for low temperature VOC oxidation (Huang et al., 2015; Zhao et al., 2011). Zhu et al. found that relative humidity plays an important role in the process of catalytic oxidation of ozone, as the presence of water significantly improved the oxidative removal of HCHO. They concluded that the positive effects of relative humidity on the process of catalytic oxidation of ozone gives a more favorable oxidation route for HCHO in a wet gas stream.
+
+Ozone oxidation is another potential technique that has been applied to decompose VOCs. Because of its strong oxidizing ability, ozone was commonly used as a cleaning agent in some air purifiers for the removal of VOCs. Yu-Hua Li and co-workers (2018) developed a continuous flow system that combines two reactors, one with the ozone oxidation technique and the other with the photocatalytic oxidation technique (UV / TiO2 + O3) to treat the cooking fumes. They found that VOCs concentrations decreased abruptly when ozone was injected into the ozone oxidation reactor, showing that ozone can react effectively with VOCs in oil fumes.
+
+Ozone, as mentioned above, is a powerful oxidant with a redox potential of 2.07 V in an alkaline solution, so it can oxidize many inorganic and organic substances (Yang et al., 2020; De Araujo et al., 2020; Tabla-Hernandez et al., 2020; Swetha et al., 2017) in'media. In general, the ozonation of pollutants can be done in two different ways a) direct reaction by the ozone molecule and b) indirect oxidation by the OH* produced by the decomposition of ozone (Sonntag, 2012). If the ozone molecule reacts directly with the pollutants, it can happen within four possible reaction mechanisms (Wang and Chen, 2020) among which oxidation-reduction reactions, cycloaddition reactions, electrophilic substitution reactions, and nucleophilic reactions may be involved; if the reaction is indirect, the pH value of the solution where the reaction occurs has a significant influence on both the efficiency of direct ozonation and on the generation of OH*. Filho et al. (2019) found that an increase in the relative humidity improved the degradation of the contaminants, which was mainly associated to the production of OH* from the reaction between O* radicals and water molecules.
+
+Water curtain cabins are used for painting processes (Fig. 1). The cabins consist of several elements that allow collecting the paint sprays in the water tank, as well as capturing the solvents incorporated into the paints.
+
+A water film traps the VOCs and micro-drops of paint remaining from the application process. This method avoids the use of filters, which reduces their maintenance. The disadvantage of this closed system is the accumulation of VOCs in the water tank. Consequently, this study aims to develop an innovative technique for the decomposition of VOCs by ozone oxidation, simulating the operation process of the water curtain cabin. This strategy of VOC absorption in aqueous solution prior to treatment has already been reported by Xie et al. (2019) through the use of a wet scrubber.
+
+### Experimental method
+
+Based on the information provided by the company, the operating plant has an effluent that presents the following contaminants: 0.0035 mg/L of formaldehyde, 0.0116 g/L of butyl acetate, 0.0053 g/L of ethanol, 0.0056 g/L of methyl ethyl ketone with a COD value of 7910 mg/L. The solubility of the contaminants in water at 20 \({}^{\circ}\) C is 7 g/L for butyl acetate, 2.9 g/L for methyl ethyl ketone and 40% v/v for formaldehyde.
+
+In a first stage, the wastewater was simulated in laboratory using the chemical reagents mentioned above, in analytical grade. An effluent of similar concentration to that which would be generated in the water tank of the paint cabin was prepared with distilled water. 250 mL of this effluent were treated by bubbling O3, and chemical oxygen demand (COD) measurements before and after treatment were then compared. The ozone was generated by an Ozonizer equipment that generates an ozone flow of 650 mg/h.
+
+In a second stage, the effluent was simulated using the service water that the company has in its process plant. As in the previous stage, 250 mL of this wastewater was treated by bubbling O3. To determine the effectiveness of the technique, different physicochemical parameters were evaluated for both the service water, the simulated wastewater, and the supernatant of the treated effluent. At this stage, a pollutant concentration tendency was made with respect to the COD values, in order to obtain the concentration ratio that should be reached in the plant up to the maximum effluent discharge value established at 250 mg/L DOC.
+
+A third stage was designed with a closed-circuit device, to simulate water curtain cabins used for painting processes. This device is presented in Fig. 2.
+
+The paint sample was injected through the lower injection nozzles and distributed on the base of the drying booth. An impeller was used to circulate the flow and release the paint solvents (VOC), and the air circuit was forced to close through a bubbling column in water. This water is incorporated into the system through the fill/drain tap. The washed air is returned to the drying cabin through the return duct, generating a circulation circuit so that all the VOCs emitted by the paint are absorbed into the water column. To achieve the necessary pressure difference and generate the bubbling column, a secondary impeller was included on the return pipe. In this system, 600 mL of plant service water were loaded, and 60 mL of paint were injected through the 9 lower injection points, to then generate the evaporation of the solvents and bubble them into the water reservoir. Successive samples were taken every 5 min for 30 min, and solvent contamination was checked by measuring the COD. Subsequently, 200 mL of that sample were taken and O3 was bubbled up to a maximum time of 8 h and the effectiveness of the method was verified by also evaluating by the final COD measurement.
+
+In a fourth stage, the ozonation equipment was incorporated in the bubble column (Fig. 2b), so that the unconsumed O3 is absorbed by the return pipe and generates a secondary gas phase oxidation. Two tests were carried out in this fourth stage; the first with the ozonation equipment off, similar to stage 3, but with gradual injection of paint, 10 mL, so that a saturation curve could be built every 5 min. In the second test, the ozone generator was turned on from the beginning, and the delay in the saturation was confirmed with VOCs in the absorption water. Once the saturation was reached, the paint injection was suspended, and the O3 oxidation test continued in the closed circuit. During the treatment phase, once the maximum of absorbed VOCs in the service water had been reached, in addition to taking samples for the determination of COD, UV spectrophotometry curves were made. The equipment used for these tests was a Hach DR 5000 spectrophotometer.
+
+### Results and discussion
+
+VOC decomposition efficiency using the ozone oxidation technique in the simulated laboratory effluent
+
+#### Using distilled water
+
+The initial COD value of the effluent simulated with distilled water was 9950 mg/L. 270 mg of O3 were added by bubbling to the simulated effluent, and a slight turbidity was observed in the sample. The final COD was filtered and measured, and its value was 1050 mg/L. The VOCs concentrations decreased sharply when ozone was injected into the ozone oxidation reactor, which shows that ozone can react effectively with the VOCs from the simulated paint drying in aqueous media.
+
+Figure 1: Water curtain paint cabin.
+
+A subsequent bubbling addition of 30 mg of O3 to the sample reduced the COD value to 1000 mg/L (Fig. 3a).
+
+#### Using service water
+
+The service water has a negligible initial COD value of 10 mg/L, which increases considerably to 6880 mg/L when the analytical reagents are added. Subsequently, 250 ml of the simulated effluent was taken and 120 mg of O3 was bubbled into it, reducing the COD value to 1022 mg/L (Fig. 3b). The physicochemical parameters evaluated in the service water and in the simulated wastewater before and after the O3 oxidation test are presented in Table 1.
+
+In both cases, using distilled and service water, the VOC concentration decreased rapidly when ozone was injected into the oxidation reactor. This result corroborates the evidence for the beneficial role of water in the ozone oxidation process as documented in previous reports (Yu and Lee, 2007; Zhao et al., 2011; Zhu et al., 2017; Huang et al., 2015). On the other hand, the pH value of the reaction solution has a significant influence on both the efficiency of direct ozonation and the generation of OH (indirect ozonation). At high pH (especially above 8),
+
+\begin{table}
+\begin{tabular}{l l l l} \hline \hline Parameters & Service water & Simulated wastewater & Treated wastewater \\ \hline COD (mg/L) & 10 & 6880 & 1022 \\ pH & 8,14 & 8,19 & 8,94 \\ Conductivity (μS/cm) & 1428 & 1435 & 1460 \\ Hardness (mg/L) & 360 & 378 & 351 \\ Alkalinity (mg CaCO3/L) & 550 & 500 & 450 \\ Silica (mg/L) & 30,2 & 28,5 & 27,4 \\ Aluminium (mg/L) & \(\leq\) 0.05 & \(\leq\) 0.05 & \(\leq\) 0.05 \\ \hline \hline \end{tabular}
+\end{table}
+Table 1: Physicochemical parameters of the samples.
+
+Figure 3: Ozone oxidation test. Simulated wastewater with a) distilled water and b) service water.
+
+Figure 2: a) Water curtain cabin simulator equipment. b) Cabin with attached ozone generator.
+
+the abundance of OH- can improve the generation of OH, being this a usually optimal pH for ozonation treatments (Wang and Chen, 2020; Chen et al., 2020a, 2020b; De Witte et al., 2010).
+
+Two types of tests, referred to as E1 and E2, were carried out with the service water and effluent simulated in the laboratory, using the closed-circuit device in Fig. 2. Table 2 refers in its first two pairs of columns to E1, and the remaining pair to E2. 600 mL of service water were saturated after 5 minutes and then this value remained constant until it reached 30 min. Afterwards, 200 mL of the sample saturated with VOC were taken and O3 was injected during 6.5 h in the ozonation reactor, in the same way the simulated sample was ozonated.
+
+Fig. 4 illustrates the variation in VOC decomposition efficiencies with the injected ozone concentration. Direct chemical reactions between ozone and VOCs would be limited when ozone concentrations were low, resulting in lower VOC decomposition efficiencies.
+
+Finally, the equipment shown in Fig. 2 was used to carry out three tests with 600 ml of service water and the paint VOCs. The first test (E1) was performed to construct the saturation curve of the VOCs in the water, without adding ozone; the second test (E2) was performed to evaluate the effect of ozonation; the third test (E3) of long duration was performed to record the evolution of the reaction kinetics. The results obtained are shown in Table 3. It can be seen in the second column that the 600 ml sample is saturated with VOC after 20 min, with 40 mL of paint injected; then, in the last two columns, the contamination and simultaneous treatment tests are shown. In the first stage, ozone is injected continuously, and 10 mL of paint are added every 5 min until the first 20 min (E2). Once the 20 min of the test have been reached, the paint is no longer injected into the system, but the ozone bubbling continues until the end of the test at 90 min (E3).
+
+Fig. 5 shows the UV absorbance curve from the saturation of the service water to the end of the treatment. There is an absorbance maximum at about 268 nm, which decreases as ozone is injected. Fig. 6 clearly shows the difference between E1 (only paint) and the first section of E2 (ozone + paint), as the VOC concentrations decreased when ozone was injected. When stopping the injection of paint and continuing with the dosage of ozone in line, E3, the change of slope in the reaction kinetics of the ozonation mechanism is clearly observed. The first stage is a direct ozonation of the VOCs in the presence of ozone and the second is a slower reaction, generated by the presence of *OH, favored by the pH of the solution. At the end of the test, after dosing 2.38 g of ozone, a COD removal of 94% is achieved.
+
+\begin{table}
+\begin{tabular}{l l l l} \hline \hline \multicolumn{4}{c}{Contamination} \\  & \multicolumn{2}{c}{E1} & \multicolumn{2}{c}{Contamination and treatment} \\  & \multicolumn{2}{c}{E2 and F3} \\ \hline Time (min) & COD (mg/L) & Ozone (g) & COD (mg/L) \\ \hline
+
+[MISSING_PAGE_POST]
+
+ \hline \hline \end{tabular}
+\end{table}
+Table 3: Contamination and simultaneous treatment of service water.
+
+Figure 4: E2 – Ozone treatment. Ozonation of VOCs as a function of dosed ozone.
+
+\begin{table}
+\begin{tabular}{l l l l l l} \hline \hline E1 \({}_{\text{s}}\) - Service water contamination & E2 – Ozone treatment & & & \\ \hline Saturation time & COD (mg/L) & Contaminated service water & & & Simulated wastewater \\
+0 & 10 & Dosed ozone (g)/ g initial COD & COD (mg/L) & Dosed ozone (g)/ g initial COD & COD (mg/L) \\ \hline
+5 & 11650 & 0 & 11870 & 0 & 11610 \\
+10 & 11710 & 0.27 & 10220 & 0.56 & 11200 \\
+15 & 11770 & 0.68 & 6240 & 1.12 & 4788 \\
+20 & 11820 & 0.96 & 3890 & 1,68 & 2000 \\
+25 & 11820 & 1.23 & 2460 & 2,24 & 665 \\
+30 & 11870 & 1.78 & 500 & & \\ \hline \hline \end{tabular}
+\end{table}
+Table 2: Service water absorption of VOCs and ozonation treatment.
+
+According to Merenyi (2009), in the direct reaction of O\({}_{3}\) with water the following reactions occur:
+
+\[\mathrm{O}_{3}+\mathrm{OH}^{-}\rightarrow\mathrm{HO}_{4}^{-}\]
+
+\[\mathrm{OH}^{-}\leftrightarrow\mathrm{HO}_{2}\cdot+\mathrm{O}_{2}^{-}\]
+
+In the presence of ozone, OH\(\bullet\) would be generated through the following equations:
+
+\[\mathrm{O}_{2}^{-}\leftrightarrow\mathrm{O}_{3}\rightarrow\mathrm{O}_{2}+ \mathrm{O}_{3}^{-}\bullet\]
+
+\[\mathrm{O}_{3}^{-}\rightarrow\mathrm{O}_{2}+\mathrm{O}^{-}\bullet\]
+
+\[\mathrm{O}^{-}\star+\mathrm{H}_{2}\mathrm{O}\rightarrow\bullet\mathrm{OH}+ \mathrm{OH}^{-}\]
+
+Usually the concentrations of radicals are low (less than \(10^{-12}\) M), but their reaction rates with the organic compounds are very high, which makes the contribution in the total oxidation of organic compounds significant (Kasprzyk-Hordem et al., 2003; Boncz, 2002). The radicals formed as secondary oxidants, the hydroxyl radical (HO\(\bullet\)) and also the superoxide anion radical (O\({}_{2}\)\(\bullet\)) (Tian et al., 2020, 2020) are very reactive, but not very specific (Schmitt et al., 2020). As mentioned, the free radicals formed in the decomposition of ozone almost always intervene in the oxidation processes. The most important radicals found in a solution containing ozone are the hydroxyl radical (HO\(\bullet\)), the superoxide radical (O\({}_{2}\)\(\bullet\)) and the hydroperoxyl radical (HO\({}_{2}\)\(\bullet\)) (Tian et al., 2020, 2020). Ozone is the most selective oxidant (three orders of magnitude between the reaction rates for the slowest and the fastest
+
+Figure 5: UV spectra during oxonation. The decrease in absorbance evidences the conversion of the contaminants into CO\({}_{2}\), O\({}_{2}\) and OH\({}^{-}\) (pH raised from 8.1 to 9)
+
+Figure 6: Contamination and simultaneous oxonation of absorbed VOCs.
+
+reaction) and O*- and HO* are the least selective, as can be seen from the results tests in Fig. 7, where a rapid reaction is observed up to the addition of 1 g of ozone (approximately 60 min of reaction) and then transitions to a moderate reaction rate ( Deng, 2020). This rapid reduction of effluent COD followed by a slower regime was also reported by Suryawan et al. (2020) in the treatment of textile effluents and by Yang et al. (2019) in the ozonation of effluents from the chlor-alkali industry.
+
+If the data from the E?3 test is considered, and the natural logarithm of COD versus dosed ozone is plotted, the points fit on a straight line (Fig. 8). This observation is repeated when analyzing the variation of the logarithm of the absorbance at 268 nm, also versus the grams of ozone injected, which indicates that the ozonation of the absorbed pollutants is due to a first order kinetic:
+
+\[\frac{dC}{dt}=-k.C\]
+
+\[\int_{C_{v}}^{C_{f}}\ln C=-k.t\]
+
+Figure 8: Representation of ln COD and ln of absorbance at 268 nm versus ozone dosed. The oxidation reaction follows a first order kinetic.
+
+Figure 7: Regime transition in ozonation treatment. In the rapid stage the ratio of g/L of COD removed per gram of ozone is 9.02; in the moderate stage it decreases to 1.08.
+
+Since the added ozone is a function of the treatment time, it can be expressed according to the regression of Fig. 8 where:
+
+\[\frac{d\;COD}{d\;gr\;O_{3}}=-1,2175\;\cdot COD\]
+
+\[\ln COD=-1,2175\;\cdot gr\;O_{3}+9,0657\]
+
+According to this model, an effluent with less than 200 mg/L will be achieved by adding 3.1 g of ozone, that is, in about 4 h 45 min of treatment. As for the products obtained by the treatment, Kharel et al. (2020) found that, in the ozonation of pharmaceutical effluents, a proportion of 0.7 to 1.0 mg O3 for each mg of dissolved organic carbon (DOC) reduces the formation of harmful by-products to insignificant levels. According to their recorded values, this corresponds to about 0.28 to 0.40 mg O3 each mg of COD; the effluent treated in this work yields a proportion of 0.35 mg O3 each mg of initial COD, remaining within the range mentioned by Kharel. In turn, Gunten records that the reaction of ozone with formaldehyde, through the formation of the formate ion, is (Gunten, 2003):
+
+\[\text{CHO}_{2}^{-}+\text{O}_{3}\rightarrow\text{CO}_{2}+\text{O}_{2}+\text{OH}^{-}\]
+
+This reaction is consistent with the fact that the pH of the treated effluent increased from 8.1 to about 9 in the treatments performed. In our case study, in addition to the oxidation of formaldehyde, the ozonation of the rest of the contaminants (butyl acetate, ethanol, methyl ethyl ketone) gives the same reaction products, based on the UV curves where a strong reduction in effluent absorbency can be seen.
+
+## Conclusions
+
+This study successfully developed a continuous flow reaction system to effectively remove VOCs from paint drying. The continuous flow reaction system combines two techniques: the water VOC absorption technique, and the ozone oxidation technique.
+
+Ozone in water undergoes a decomposition process and is converted into hydroxyl (HO*) radicals, the superoxide radical (O2-*) and the hydroperoxyl (HO2*) radical, for subsequent VOC reactions, and the oxidation reaction follows a first order kinetic.
+
+It was shown that the decomposition efficiency of VOCs in water increased with the concentration of ozone, reaching a COD removal of 94%. The closed circuit allowed the treatment efficiency to be increased by recapturing any ozone that might escape from the water in the bubble column.
+
+## Declaration of Competing Interest
+
+None.
+
+## References
+
+* De Araujo et al. (2020) De Araujo, L.G., Prado, E.S.A.P., Miranda, F.D., Vicente, R., Sobrinho, A.S.D., Filho, G.P., Marumo, J.T., 2020. Physicochemical modifications of radioactive oil sludge by ozone treatment. J. Environ. Chem. Eng. 8 (5). doi:10.1016/j.jecmose.2020.104128.
+* Bonc (2002) Bonc, M., 2002. Selective oxidation of organic compounds in waste water by ozone-based oxidation processes. Wageninger University Promot. Prof.Dr.H.F.W.R. Rultken, Prof.Dr.H.F.R. Rultken, Prof.Dr.H.F.R. Rultken, Prof.Dr.H.F.R. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.H.F. Rultken, Prof.Dr.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. Rultken, Prof.H.F. F Xie, R., Liu, G., Liu, D., Liang, S., Lei, D., Dong, H., Huang, H., Leung, D.Y.C., 2019. Wet scrubber coupled with heterogeneous UV/femton for enhanced VOCs oxidation over Fe/2SiSe catalyst. Chemosphere 227, 401-408. doi:10.1016/j.chemosphere.2019.03.160.
+* Yang et al. (2019) Yang, Z., Wang, P., Zhang, Y., Zan, X., Zhu, W., Jiang, Y., Zhang, L., Yasin, A., 2019. Improved donation efficiency for polymerization mother liquid from polyvinyl chloride production using tandem reactors. Molecules 24 (24), 4436. doi:10.3390/molecules24/44436.
+* Yang et al. (2020) Yang, Z., Zhang, Y., Zhu, W., Zan, X., Zhang, L., Liu, Y., 2020. Effective oxidative degradation of coal gasification wastewater by ozonation: a process study. Chemosphere 255, 126963. doi:10.1016/j.chemosphere.2020.126963.
+* Yu and Lee (2007) Yu, R., Lee, G.W.M., 2007. Decomposition of gas-phase toluene by the combination of ozone and photocatalytic oxidation process (TiO2/UV, TiO2/UV/O3, and UV/O3). Appl. Catal. B Environ. 75 (1-2), 29-38. doi:10.1016/j.apcatb.2007.03.006.
+* Zhang et al. (2016) Zhang, Z., Jiang, Z., Shangguan, W., 2016. Low-temperature catalysis for VOCs removal in technology and application: a state-of-the-art review. Catal. Today 264, 270-278. doi:10.1016/j.catatol.2015.10.04.
+* Zhao et al. (2011) Zhao, D.Z., Li, X.S., Shi, C., Fan, H.Y., Zhu, A.M., 2011. Low-concentration formaldehyde removal from air using a cycled storage-discharge (CSD) plasma catalytic process. Chem. Eng. Sci. 66 (17), 3922-3929. doi:10.1016/j.cesd.2011.05.019.
+* Zhu et al. (2017) Zhu, B., Li, X.S., Sun, P., Liu, J.L., Ma, X.Y., Zhu, X., Zhu, A.M., 2017. A novel process of ozone catalytic oxidation for low concentration formaldehyde removal. J. Catal. 38 (10), 1759-1769. doi:10.1016/j.jcssd.2006.17(26)2890.0.
+* Zhu et al. (2015) Zhu, X., Gao, X., Qin, R., Zeng, Y., Qu, R., Zheng, C., Tu, X., 2015. Plasma-catalytic removal of formaldehyde over Cu-Ce catalysts in a dielectric barrier discharge reactor. Appl. Catal. B Environ. 170-171, 293-300. doi:10.1016/j.apcatb.2015.01.032.

manual_annotation/200pdfs_in_mmd/10.1016_j.eti.2019.100404.mmd

@@ -0,0 +1,323 @@
+Combined ozone oxidation process and adsorption methods for the removal of acetaminophen and amoxicillin from aqueous solution; kinetic and optimisation
+
+Amin Mojiri, Mohammadtaghi Vakili, Hossein Farraji, Shuok Qarani Aziz
+
+# Abstract
+
+The growing use of pharmaceuticals raises questions on their potential risk to human health and water quality. This research aimed to introduce a combined treatment technique with high performance in removing pharmaceutical micropollutants (MPs) from aqueous solution. The research included two steps, ozone treatment (first step) and adsorption technique (second step). The elimination of acetaminophen (ACT) and amoxicillin (AMX) with ozone reactor, first step, was optimised by artificial neural network (ANN). The optimisation process included two independent variables, namely, initial concentration of MPs and ozone dosage. On the basis of ANN, the linear regression coefficient denoted by R2 between predicted and experimental MP removals was close to 1. Result displayed that the prediction by the trained ANN is acceptable. Approximately 0.17 mg/L (84.8%) of ACT and 0.16 mg/L (82.7%) of AMX were removed at the initial concentration of 0.2 mg/L and ozone dosage of 15 mg/L. Beside it, ozonation experiments showed that the rate of constant (m-1s-1) for ACT and AMX were 2.63x106 and 5.98\(\times 10^{6}\) respectively. After treating by ozone reactor, water was subjected to pass through the cross-linked chitosan/bentonite as a fixed-bed column, for second step. ACT and AMX were not detected after step 2.
+
+## 1 Introduction
+
+Water pollution has worsened in several areas around the world, particularly in areas with threats of high contamination, such as across Africa, Asia, America and Europe (Evans et al., 2019). The exponential increment in world population, recent industrialisation, civilisation, agricultural and household activities leads to greater levels of organic and inorganic contaminants in water (Yadav et al., 2019). One of the vital pollutants is categorised as emerging micropollutants (MP). MPs are primarily synthetic organic chemicals that have been recently found in natural environments. MPs can affect aquatic and human life (Ahmed et al., 2017). Examples of MPs are steroid sex hormones, pharmaceuticals, personal-care products, illicit drugs, flame retardants and perfluorinated compounds, which can enter wastewater networks after use in households and industries. Moreover, MPs have been detected in sewage treatment plant influents and effluents in many countries (Prieto-Rodriguez et al., 2013). Pharmaceutical MPs are commonly found in various environmental compartments. The growing use of pharmaceuticals has raised questions regarding their potential risk to human health, environment and water quality (Mahmoud et al., 2017). Therefore, this research investigates the removal of pharmaceutical MPs coming from amoxicillin (AMX) and acetaminophen (ACT), which are widely applied in many countries.
+
+AMX is an antibiotic for the treatment of several bacterial infections. The formula for amoxicillin is C16H19N3O5S (IU-PAC ID of (25,5R,6R)-6-[([2R)-2-amino-2-(4-hydroxyphenyl)-acetyl]amino)-3,3-dimethyl-7-oxo-4-thia-1-azabicyclo[3.2.0] heptane-24-carboxylic acid). As a water soluble solid compound, AMX has weak acid properties (Kidak and Dogan, 2018). Amoxicillin is the member of beta-lactams group. Beta-lactams have 65% of the total world market for antibiotics (Demirezen et al., 2019). Acetaminophen (C8H9NO2, N-acetyl-p-aminophenol), known as paracetamol, is frequently applied over-the-counter non-steroidal anti-inflammatory drug used for the treatment of pain and fever (Katarzyn and Tyszczuk-Rotko, 2018). Therefore, ACT was selected in the current study because of its high usage of about 1.45 x 105 tons per year and high solubility of 12.78 g/L in aqueous solution (Wang et al., 2019). Daas and Hamdaoui (2014) stated that conventional treatment procedures at urban wastewater treatment plants fail to remove completely pharmaceutical substances. Thus, employing combined treatment techniques may have high performance in eliminating these pollutants completely. One of the attractive physical/chemical treatment ways that can be combined with other methods is advanced oxidation process (AOP).
+
+AOPs are based on the generation of hydroxyl radicals (OH) in solutions and considered for the degradation of biorefractory or hazardous organic compounds in water systems (Benitez et al., 2011). AOPs, such as ozonation, photocatalysis and photo-Fenton techniques, have attracted attention for the elimination or transformation of emerging contaminants (Valcarcel et al., 2012). Ozone is an oxidant for drinking water treatment because of its oxidation and disinfection potential. Under water treatment conditions, ozone can produce hydroxyl radicals that can react fast with a wide range of molecules (Gomez et al., 2017), especially organic compounds. Ternes et al. (2003) removed approximately 85% of an antibiotic (trimethoprim) by using a modified ozonation method. The ozonation technique can be combined with other physical/chemical methods, such as adsorption methods. Kurniawan et al. (2006) and Mojiri et al. (2017) reported a combined treatment method, including ozonation and adsorption methods, in treating wastewater.
+
+Adsorption is a mass transfer process by which a substance is transferred from the liquid phase to the surface of a solid and then bound by physical or chemical interactions or both (Kurniawan et al., 2006). The use of low-cost and environment friendly materials, such as chitosan and bentonite, to eliminate MPs has attracted research attention (Xu et al., 2012; Zha et al., 2013). Chitosan has high hydrophilicity due to a large number of hydroxyl groups in its glucose units, which makes chitosan theoretically suitable for organic and inorganic adsorption (Huang et al., 2018). Bentonite is primarily an expandable montmorillonite clay (Alexander et al., 2018). Low-cost adsorption has been widely used as a parameter for removing pollutants from water. Approximately 81% of AMX was removed by modified organobentonite (Zha et al., 2013). In the study, the removal of ACT and AMX were optimised with the artificial neural network (ANN). Artificial neural network modelling methods have gained significant attention in modelling wastewater treatment techniques during recent years. Beneficially, a neural network model has unique ability of learning non-linear functional relations. They do not need any prior structural knowledge of the relations that exist between vital variables and procedures to be modelled (Guclu and Dursun, 2010). Due to its accuracy and simplicity of prediction, the applying of machine learning, remarkably artificial neural network (ANN), to model wastewater treatment procedure becomes a talented alternative realisable with the development of computing skills. ANN models applies a set of non-linear equations to establish complex patterns and relations between input and output, and hence, it might be applied in estimation, prediction, simulation and classification (Bekkari and Zeddouri, 2018). Silva Riberio et al. (2019) investigated the using of ANN in eliminating boron from wastewater by electrocoagulation. In the basis of the reasonable R\({}^{2}\) and squared error which confirmed the well fit of the ANN model. Therefore, using just some input parameters may be employed during water or wastewater treatment based on the goals, and the model can be updated during the treatment process in the pilot-scale with further inputs.
+
+Hence, contemporary research has chiefly designed removal of acetaminophen and amoxicillin by combining ozonation and cross-linked chitosan/bentonite fixed-bed column. Some researchers have attempted to design a combined treatment technique with optimum performance. However, no existing designed system is stated in the literature. Furthermore, this study applies artificial neural network (ANN) to optimise the removal of ACT and AMX.
+
+## Materials and methods
+
+### Materials
+
+AMX (Table 1), ACT (Table 1), bentonite and chitosan were supplied by Sigma-Aldrich Co. Distilled water and methanol (>=99.9%) were provided by Merck Co. with high-pressure liquid chromatography (HPLC) grade, and other chemicals were purchased from Merck.
+
+### Ozone reactor
+
+For first step of this study, a reactor with a working volume of 2.1 L (height = 45 cm, radius = 4 cm) was used as the ozonation reactor. A column ozone chamber was used for ozone gas diffusion. The cooling system and water bath were applied to maintain the internal reflex temperature at <15oC and obtain the optimal half-life of the dissolved ozone (30 min) in water (Mojiri et al., 2017b). Ozone generator (BMT Messtechnik, Germany) feeding with pure dry oxygen was used to produce ozone. Ozone dosage was monitored by an ultraviolet gas ozone detector (BMT 964). Reaction time was set to 25-30 min based on preliminary experiments (Rodayan et al., 2014). The ozone dose (mg/L) varied from 3 (Odabasi and Buyukgungor, 2016) to 15 (Ternes et al., 2003). After monitoring the off-O3 by detector, it was gone to a bubble column for ozone destruction.
+
+### Producing cross-linked chitosan/bentonite as fixed-bed column and designed fixed-bed column
+
+In preparing the cross-liked chitosan/bentonite (CCB), 4 g of chitosan was firstly dissolved in 200 mL of acetic acid solution (2 wt.%) with continuous stirring in accordance with Dotto et al. (2016). Subsequently, 4 g of bentonite was augmented to the chitosan solution and stirred for 2 h at 45oC. The pH of the solution was adjusted to nearly 6.0 with NaOH. Then, to prepare chitosan/bentonite bead, the gel was dripped into a precipitation bath containing 250 mL of 20% ethanol -NaOH (2 M) solution by a disposable syringe. The CCB gel thereby coagulated to uniform beads. Finally, cross-linked chitosan/bentonite was washed with distilled water and dried in an oven at 60oC for 24 h and then sieved. On the basis of the autosorb (Quantachrome IQ, Germany) tests for cross-linked chitosan/bentonite, the BET surface area (m2/g), the Langmuir surface area (m2/g), the micropore area (m2/g) and the micropore volume (cc/g) were 237, 453, 116 and 28.4, respectively. The zeta potential of CCB) was monitored (Mojiri et al., 2019).
+
+A glass column with radius 8 mm (16 mm diameter) and 15 cm length was filled by CCB as a fixed-bed column adsorption for second step of this research. On the basis of preliminary studies, the contact time was set around 25 min (Rua-Gomez et al., 2012).
+
+\begin{table}
+\begin{tabular}{l c c c c} Name & Chemical formula & Chemical structure & Molecular weight (g/mol) & pKa & Log \(K_{\mathrm{OW}}\) \\ Acetaminophena (ACT) & G\({}_{4}\)H\({}_{8}\)NO\({}_{2}\) & & 151.1 & 9.38 & 0.46 \\  & & & NH & & \\ Amoxicillinb (AMX) & C\({}_{46}\)H\({}_{8}\)N\({}_{4}\)O\({}_{5}\)S & & & & \\ \end{tabular}
+\end{table}
+Table 1: Physical/chemical properties of ACT (A) and AMX (B).
+
+### Synthetic aqueous solution and experimental procedure
+
+A stock solution of each MP (ACT and AMX) was prepared by dissolving 10 mg of MP in 10 mL of solvent mixture methanol/water (50:50) (Stavbar et al., 2017). This study applied two steps (Fig. 1), namely, MP removal by ozone reactor and fixed-bed column adsorption. The initial concentrations of ACT and AMX ranged from 0.2 mg/L (Moreira et al., 2016) to 2.2 mg/L (Roza et al., 2017) in the ozone reactor. After treating water in the ozone reactor, water was moved through the fixed-bed column (cross-linked chitosan/bentonite). The contact time was set to 30 min (Rua-Gomez et al., 2012) in the fixed-bed column.
+
+The experiment was carried out at the neutral pH of 7 to reduce treatment cost and follow up actual conditions (Jin et al., 2012).
+
+### Analytical methods
+
+The analytical methods for MPs (ACT and AMX) were obtained from the literature (Weng et al., 2018). MP concentrations were tested by HPLC (LC-20AT, Shimadzu International Trading (Shanghai) Co., Ltd.). The UV detector had an analytical column of 2.1 mm \(\times\) 100 mm with particle size of 1.7 \(\mu\)m and working wavelengths of 208 nm and 230 nm for AMX and ACT, respectively. The mobile phase was a mixture of NaH\({}_{2}\)PO\({}_{4}\) and acetonitrile with a volumetric ratio of 60/40 with an injection flow rate of 1 mL/min. Limit of detection (LOD) and limit of quantification (LQQ) were estimated in according to the 3 \(\sigma\)/s and 10 \(\sigma\)/s criteria, respectively; where \(\sigma\) defines the standard deviation of the peak area and s specifies the slope of the corresponding calibration curve (Attimarad, 2011). LOD (\(\mu\)g/L) and LOQ (\(\mu\)g/L) for AMX and ACT were 0.03 and 0.09, and 0.10 and 0.30 respectively. Lempart et al. (2018) expressed that LOD defines the lowest quantity of a substance that may be identified from the absence of that substance within a certain confidence interval. In this study, when concentrations were below the LOD or LQQ, we mentioned to "not detected".
+
+The zeta potential of cross-linked chitosan/bentonite (CCB) was monitored by the zeta potential meter (Zetasizer nan-ZS90 Malvern) at 25\({}^{\circ}\)C in different pH (3 to 9) (Mojiri et al., 2019). Based on Soros et al. (2019), zeta potential was analysed via two different techniques. First, water samples were analysed for their background electrical charges. Second, by using a titration process, zeta potential was analysed during the titration of water samples with chitosan. For this measurement, water samples comprising a certain amount of bentonite at neutral pH were titrated with chitosan stock solution at different doses. Water pH was repeatedly measured during the titration. Titration graphs presenting changes in zeta potential while chitosan was being added into the water sample were generated displaying chitosan dose and pH at which the point of zero charge occurred. Fourier transform infrared (FTIR) spectra was monitored via the Thermoscientific Nicolet 380 with KBR, ranged 400 to 4000 cm\({}^{-1}\).
+
+### Statistical analysis
+
+The effectiveness of MP removal was estimated on the basis of initial and final aqueous phase concentrations (Eq. (1)).
+
+\[\text{Removal}\ \%=\frac{C_{i}-C_{f}}{C_{i}}\times 100, \tag{1}\]
+
+where initial and final concentrations are denoted by \(C_{i}\) and \(C_{f}\), respectively.
+
+Figure 1: Schematic of experiments.
+
+ANN with Levenberg-Marquardt (LM) training algorithm was established for correlating the effectiveness of ACT and AMX elimination from water by using the ozone reactor. This algorithm was evaluated by using Matlab (version R2015a). In the feed-forward neural network, the data are only moved forward in one direction (Tanzifi et al., 2017). In this study, 60% of 70 datasets were applied for the training and development of the neural network, and 20% and 20% were applied for validation and testing, respectively. The initial concentrations of each MP and ozone dosage were considered as input data for the model, and the removal efficacy of each MP was defined as output. The ANN topology included two neurons in the input layer, four neurons in the hidden layer and one neuron in the output layer. R\({}^{2}\) and mean square error (MSE) values were analysed for the selection of the best model (Eqs. (2) and (3)).
+
+\[MSE=\frac{1}{N}\sum_{i=1}^{N}(|y_{\text{prd},i}-y_{\text{exp},i}|)^{2}, \tag{2}\]
+
+\[R^{2}=1-\frac{\sum_{i=1}^{N}(y_{\text{prd},i}y_{\text{exp},i})}{\sum_{i=1}^{N} y_{\text{prd},i}-y_{m}}, \tag{3}\]
+
+where \(y_{\text{prd},i}\) is the predicted value by using the ANN model; \(y_{\text{exp},i}\) is the experimental value; N is the number of datasets; and \(y_{\text{m}}\) is the average of the experimental values.
+
+### Kinetic study of ozonation
+
+The changing concentrations of ACT and AMX during the ozonation experiments was considered as a function of contact time. For this purpose, in this part of the current research, initial concentrations of ACT and AMX were set to 1 uM, and the preliminary ozone concentration varied from 20.8 to 93.8 uM. The experiments were repeated thrice. This method was described by Almomani et al. (2016) in detail. The reaction requirements were assessed by Eq. (4).
+
+\[\text{Reaction Requirements }(z)=\frac{MP_{i}-MP_{f}}{Ozone_{i}-Ozone_{f}}, \tag{4}\]
+
+where MPi and MPf denote the initial concentration and the final concentration of MPs, respectively, while Ozonei and Ozonef denote the initial concentration and the final concentration of ozone, respectively.
+
+On the basis of the described method by El Najjar et al. (2014), the rate constants of the reaction of ozone with ACT and AMX were defined by using the competitive kinetic technique. Phenol (PHN) was selected as competitor (\(k_{\text{03/PHN}}=2.79\times 10^{6}\) M\({}^{-1}\)s\({}^{-1}\), with pH of approximately 7.1). Batch experiments were done in beakers, and the pH was set to 7-7.1. For each experiment, tert-butanol (3.16 mM) was introduced as the HO radical scavenger. After complete ozone consumption (i.e. 24 h), the residual concentrations of MPs (ACT and AMX) and PHN were monitored. Under these conditions, as the MPs and the reference compound PHN were concurrently existing in the reaction system, and based on second-order rate laws, the kinetic expressions for the ozone reaction can be written as follows (Eq. (5)):
+
+\[\ln\frac{[\text{MP}]_{T,n}}{[\text{MP}]_{T,n}}=\frac{k_{\text{03/MP}}}{k_{\text {03/PHN}}}\ln\frac{[\text{PHN}]_{T,n}}{[\text{PHN}]_{T,n}}, \tag{5}\]
+
+where \(k_{\text{03/MP}}\) and \(k_{\text{03/PHN}}\) denote the second-order rate constants for the response of ozone with MP and PHN, respectively, and \(n\) is the ozone dosage. The ratio defined by kO3/MP/kO3/PHN was derived from the slope of \(\ln[(\text{MP}]_{T,n}/[\text{MP}]_{T,0})\) versus \(\ln[(\text{PHN}]_{T,n}/[\text{PHN}]_{T,0})\) for each experiment.
+
+### Adsorption study
+
+Batch adsorption studies were done by using different dosages (up to 10 g/L) of CCB in fixed ACT and AMX concentration (2.0 mg/L), pH (7) and adsorption time (30 min). Beakers with working volumes of 100 mL were shaken at 300 rpm for 30 min.
+
+The capacity of adsorption (mg/g\({}^{-1}\)) was estimated with the following Eq. (6):
+
+\[q_{e}=\frac{(C_{0}-C_{eq})V}{m_{s}}, \tag{6}\]
+
+where the initial MP concentration is denoted by \(q_{e}\); \(C_{eq}\) is the MP concentration (mgL\({}^{-1}\)) at equilibrium; \(V\) is the solution volume (L); and \(m_{s}\) is the mass of the adsorbent (g).
+
+### Studies of empty bed contact time (EBCT), mass transfer zone and breakthrough curves
+
+The contact time between the water phase and the adsorbent is named as empty bed contact time (EBCT). EBCT (min) mainly measures the critical depth and the contact time for the adsorbent (Beji et al., 2018). EBCT was assessed with the following relation between packed bed volume V (mL) and liquid stream flow rate Q (mL/min) (Eq. (7)):
+
+\[\text{EBCT}=\frac{V}{Q} \tag{7}\]On the basis of the experimental data from the parametric studies, bed height, flow rate, and concentration, the breakthrough curves were plotted by the normalised concentration of ACT and AMX in terms of time. Normalised concentration (Ct/C0) is explaining the concentration (Ct) at time t, and C0 (initial concentration) (Muthamileslu et al., 2018). For estimating effects of flow rate on breakthrough curves, the Q varied from 4 mL/min to 12 mL/min (Beji et al., 2018) and concentration was set at 2 mg/L. And for estimating effects of initial concentration of MPs on breakthrough curves, the concentration varied from 1 mg/L to 3 mg/L and Q was set at 8 mL/min.
+
+The zone of the bed where maximum of the adsorption occurs and moves up the bed is described by the mass transfer zone (MTZ) (Beji et al., 2018) (Eq. (8)).
+
+\[\text{MTZ}=L\frac{t_{e}-t_{b}}{t_{e}} \tag{8}\]
+
+where L(Z) = bed height or length (cm), _t_e = the required time to touch the breakthrough (min), _t_b = the required time to touch the exhaust point (min)
+
+### Desorption and reusability studies
+
+The economic usability of the cross-linked chitosan/bentonite (CCB) was monitored by referring to regeneration studies. CCB was regenerated by soaking in 100 mL methanol for 2-3 h in the batch experiments and then washed by using distilled water for the desorption and regeneration of CCB. Seven adsorption/desorption cycles were done. After each cycle, the residual concentration of MPs was monitored (He et al., 2017).
+
+## 3 Results and discussions
+
+In this research, ACT and AMX were removed by a combined treatment method, including ozonation and cross-linked chitosan/bentonite (a fixed-bed column). Water was firstly treated in the ozone reactor, and water was moved through the fixed-bed column to completely remove ACT and AMX. The removal efficiency (Table 2 and Fig. 2) was optimised by ANN during the water treatment in the ozone reactor.
+
+### Antibiotic removal by ozone reactor and ozone utilisation study
+
+As shown in Table 2 and Fig. 2, the maximum ACT removal (84.8% or 0.170 mg/L) occurred at the ozone dosage 15 mg/L and ACT initial concentration of 0.2 mg/L. The minimum ACT removal (57.2% or 1.258 mg/L) occurred at the ozone dosage 3 mg/L and ACT initial concentration of 2.2 mg/L. Snyder et al. (2006) reported an approximate 80% removal of ACT by ozonation in 24 min. The high removal of ACT during ozonation with 10 to 40 mg/L of ozone dosage was reported by Kuzmanovic et al. (2013).
+
+As shown in Table 2 and Fig. 2, the maximum AMX elimination (82.7% or 0.165 mg/L) occurred at the ozone dosage of 15 mg/L and AMX initial concentration of 0.2 mg/L. The minimum AMX elimination (56.0% or 1.231 mg/L) occurred at the ozone dosage of 3 mg/ and AMX initial concentration of 2.2 mg/L. Liu et al. (2014) reported approximately 87% of antibiotic removal by UV/O3 in 30 min. Here, the initial antibiotic concentration was 0.2 mg/L. Alajmi (2014) removed 84% of an antibiotic (trimethoprim) at the ozone dosage of 15 mg/L. These studies support the current research.
+
+Rame et al. (2018) stated that ozone may be an excellent choice for eliminating antibiotics from water and wastewater supplies, and antibiotics may be rapidly destroyed by ozone. In ozonation systems, MPs may be oxidised by the direct reaction with ozone (Eq. (9)). Hydroxyl radicals can be produced through the decomposition of ozone (Eq. (10)). An indirect reaction by the non-selective and highly reactive hydroxyl radicals (Eq. (11)). (Hansen et al., 2016).
+
+\[R+\text{O}_{3}\longrightarrow R_{\text{OX}} \tag{9}\]
+
+\[\text{H}_{2}\text{O}+\text{O}_{3}\longrightarrow\text{O}_{2}+2\text{HO} \tag{10}\]
+
+\[R+\text{HO}\longrightarrow R_{\text{OX}} \tag{11}\]
+
+As shown in Table 2, elimination efficacies can be increased by increasing ozone dosage. Yaghmaeian et al. (2017) reported that enhancing pharmaceutical degradation with increasing ozone dosage may be linked to the increase in the number of ozone molecules. Therefore, the chances of interaction of pharmaceutical molecules can increase with a rise in number of ozone molecules. Yanyan et al. (2018) stated that increasing ozone dosage leads to the increased removal of ACT in neutral pH. Ling et al. (2019) stated that the increase of pH value is useful for the transformation of O3 to OH in ozonation. In these previous studies, the consumption of O3 in neutral conditions can be achieved by effective decomposition.
+
+### Optimisation with ANN
+
+The values of R\({}^{2}\) and MSE are shown in Table 3. The best model with low MSE and reasonable R values (i.e. close to 1 for all training, validation and testing phases) was considered (Gadekar and Ahammed, 2019). Linear regression analysis between experimental and predicted removal data of each MP was completed to assess network efficacy. The predicted and experimental MP removal data yielded linear regression coefficient (R\({}^{2}\)) values close to 1 (i.e. their slopes approach 1). This result indicates that the prediction by the trained ANN model is acceptable. Fig. 3 shows the changes in MSE values by the LM training algorithm by selecting several functions, such as pure linear, transig and log sigmoid. The training was completed after 56 and 13 epochs for ACT (a) and AMX (b), respectively. These findings further prove that the ANN model is sufficiently trained at the end of the training phase (Uddin et al., 2018). Fig. 4 shows the model prediction versus experimental values. Nagendra and Khare (2006) expressed that ANN models need known input data set without any assumptions It displays quick information processing and is able to develop a mapping of the input and output variables. Therefore, the current ANNs models and predictions can be applied in a water treatment plant to control treatment process in removing AMX and ACT. Based on the input concentration of MPs, the input concentration of ozone may be changed to reach optimum removal efficiency.
+
+### Kinetic study of ozonation
+
+The ozonation of MPs in water is a consequence of the integration of direct and indirect (or radical) oxidation reactions. The overall kinetic of both reactions is supposed to be in the second order (Almomani et al., 2016). As this experiment was done in neutral pH, the elementary reactions of ozone with neutral and ionic species of ACT and AMX can lead to the
+
+Figure 2: Elimination efficacies for Acetaminophen (A) and Amoxicillin (B).
+
+pH dependence of the rate constant (El Najjar et al., 2014). Then, the rate of the ozonation of organic MPs can be defined by Eq. (12).
+
+\[\frac{d[\text{MP}]_{\text{T}}}{dt}=K_{0_{3}}\left[\text{O}_{3}\right]\left[\text{MP}\right] \tag{12}\]
+
+A graph of ln[PRCI]T,n/[PRCI]T,o versus ln[PHN]T,n/[PHN]T,o is shown in Fig. 5. The ozonation was set with pH of 7 \(\pm\) 0.1 and temperature of 22 degC. The rates of constant (m-1s-1) for ACT and AMX were 2.63\(\times\)10\({}^{6}\) and 5.98 \(\times\)10\({}^{6}\), respectively. El Najjar et al. (2014) reported 2.57 m-1s-1 as the constant rate at pH 7.2. Ikehata et al. (2006) reported 6\(\times\)10\({}^{6}\) m-1s-1 at the constant rate at pH 7. The constant rate (m-1s-1) for antibiotics removal by ozonation was reported to be from 3.8\(\times\)10\({}^{5}\) to 1.7\(\times\)10\({}^{7}\) by Ben et al. (2012).
+
+### Antibiotics removal by the fixed-bed column
+
+Fixed-bed column, including cross-linked chitosan/bentonite, was prepared for the second step of this study. Fig. 6 shows the FTIR results of chitosan, bentonite and chitosan/bentonite. For chitosan (Fig. 6a), peaks 3462 and 2938 correspond to O-H and C-H, respectively, which are in line with the findings of Yasmeen et al. (2016) and Abukhadra et al. (2019). C=O and N-N group were justified by peaks 1655 and 1462, respectively (Abukhadra et al., 2019). For bentonite (Fig. 6b), peak 3422 corresponds to hydroxyl groups (El-Dib et al., 2016). Peaks 793 and 1044 correspond to Si-O and
+
+\begin{table}
+\begin{tabular}{l l l l l l l l l} \hline Run & Ozone dosage (mg/L) & Initial concentration of MP (mg/L) & Acetaminophen removal & Amoxicillin removal & & \\  & & & (\%) & (mg/L) & SD & (\%) & (mg/L) & SD \\ \hline
+
+[MISSING_PAGE_POST]
+
+ \hline \end{tabular}
+\end{table}
+Table 2: Removal of acetaminophen and amoxicillin by ozone reactor.
+
+\begin{table}
+\begin{tabular}{l l l l l l l} \hline Micropolitanant & R\({}^{2}\) & MSE & & \\  & Training & Validation & Test & Training & Validation & Test \\ \hline For Acetaminophen removal & 0.994 & 0.994 & 0.924 & 0.574 & 0.874 & 6.425 \\ For Amoxicillin removal & 0.986 & 0.986 & 0.971 & 0.941 & 0.825 & 2.419 \\ \hline \end{tabular}
+\end{table}
+Table 3: R\({}^{2}\) and MSE values for the removal of each pollutant in the selection of the best model.
+
+Al-O, respectively (Huang et al., 2017). For cross-linked chitosan/bentonite (Fig. 6c), peak 3422 represents an interfacial interaction between chitosan and bentonite surfaces, as peak 3431 linked to OH groups on the clay surface has changed to 3422 (El-Dib et al., 2016). Based on the Fig. 7, zeta potential of CCB was positive in pH (3) to (5.9) it is in line with finding of Rijith et al. (2016). Zeta potentials (mV) were 41 to 12 in pHs 3 to 5 respectively. The zero point during zeta potential testing was reached at 6 of pH. After that zeta potential converted to negative which could be supported by findings of Cabuk et al. (2016). Zeta potentials (mV) were \(-\)5 to \(-\)16 in pHs 7 to 9, respectively.
+
+After treating water in the ozone reactor in each run, water was moved through a fixed-bed column, including cross-linked chitosan/bentonite. The remaining ACT and AMX from each run of ozone reactor were lower than the LOD and LOQ, not detected, after step 2. Roza et al. (2017) removed more than 90% of pharmaceuticals by using O3/PAC (ozone/powdered activated carbon). Liu et al. (2016) reported 85% of AMX removal through advanced treatment (ozonation/biological activated carbon). Xiao et al. (2018) stated that bentonite is a widely available and inexpensive material, with a large surface area, and its positive ionic exchange capacity and adsorptive properties can effectively transport and eliminate pollutants. Lozano-Morales et al. (2018) reported that the use of clay cationic adsorbents, such as bentonite, is a reasonable technique in removing pharmaceuticals.
+
+was reported by Mashayekh-Salehi and Moussavi (2016) during ACT removal with adsorption technique. For the AMX elimination, R2, b and Q (mg/g) were 0.838, 0.370 and 0.008, respectively. The value of b = 0.01 was reported by de Franco et al. (2017).
+
+Fig. 4: Model prediction versus experimental values for the optimum topology for acetaminophen removal (A), and amoxicillin removal (B). * The figures for training, validation data, test and all were obtained by training dataset (60% of all data), validation dataset (20% of all data), test dataset (20% of all data), and all data, respectively based on the maximum R2.
+
+The Freundlich adsorption isotherm (Eq. (14)) is a curve representing the concentration of a solute on the surface of an adsorbent to the concentration of the solute in the liquid with which it is in contact.
+
+\[q_{m}=K_{\text{f}}C_{\text{e}}^{1/n}, \tag{14}\]
+
+\begin{table}
+\begin{tabular}{l l l l l l l} \hline Parameters & Langmuir isotherm & & & Freundlich isotherm & & \\ \cline{2-7}  & \(\overline{Q_{m}}\) (mg/g) & b & R\({}^{2}\) & \(K_{\text{f}}\) (mg/g(L/mg)\({}^{1/n}\)) & 1/n & R\({}^{2}\) \\ \hline ACT & 0.008 & 0.318 & 0.841 & 0.175 & \(-\)3.160 & 0.966 \\ AOX & 0.008 & 0.370 & 0.838 & 0.268 & \(-\)7.034 & 0.959 \\ \hline \end{tabular}
+\end{table}
+Table 4: Langmuir and Freundlich isotherms study for Removal ACET and AMX by CB.
+
+Figure 5: Determination of the second-order rate constant \(k_{03/\text{ACT}}\) (A) and \(k_{03/\text{AMX}}\) (B).
+
+Figure 6: FTIR results of chitosan/bentonite. * chitosan (a), bentonite (b) and chitosan/bentonite (c),
+
+Figure 7: Zeta Potential of CCR.
+
+where \(K_{\rm f}\) is the adsorption capability of the adsorbent (mg\({}^{1-(1/n)}\)/L\({}^{1/n/g-1}\)), and n is a fixed variable representing adsorption intensity.
+
+Table 4 and Fig. 8 show the Freundlich isotherm regression, constants and correlation coefficients for removing ACT and AMX. For the ACT removal, R\({}^{2}\), 1/n and \(K_{\rm f}\) (mg\({}^{1-(1/n)}\)/L\({}^{1/n}\)/g\({}^{-1}\)) were 0.966, \(-\)3.160 and 0.175, respectively. The values of R\({}^{2}\) = 0.99 and \(K_{\rm f}\) = 0.17 were reported by Jung et al. (2015). For the AMX elimination, R\({}^{2}\), 1/n and \(K_{\rm f}\) (mg\({}^{1-(1/n)}\)/L\({}^{1/n}\)/g\({}^{-1}\)) were 0.959, \(-\)7.034 and 0.268, respectively. The values of \(K_{\rm f}\) = 0.17 and R\({}^{2}\) = 0.97 were reported by de Franco et al. (2017) during AMX removal with the adsorption method. Mojiri et al. (2017a) stated that relatively high 1/n values yield weak adsorption bonds. Increasing log (\(C_{\rm e}\)) can lead to decreasing log (x/m), and consequently a negative 1/n (slope of the line). In terms of R\({}^{2}\), Freundlich is more favourable than Langmuir. The result in the current work agrees with those of other research (Ahile et al., 2015).
+
+### Empty bed contact time (EBCT) and breakthrough curves
+
+One of the vital parameters is flow rate (Q) during investigating the performance of continuous fixed-bed adsorption column. The influence of the Q on the adsorption of ACT and AMX by using CCB was estimated by different the flow rate (4, 8 and 12 mL/min) while the bed length and influent MPs concentration were set at 15 cm and at 2 mg/L, respectively. With increasing the fellow rate (mL/min) from 4 to 8, MTZ and EETC were decreased. It is in line with findings of Chowdhury et al. (2015). In the high Q, the column was gained the exhausted point earlier. It means the fixed-bed column was saturated quickly in high Q. In the basis of Table 5, the \(t_{\rm b}\) (min) and \(t_{\rm e}\) were 20, 30 and 40, and 90, 100 and 110 for ACT removal respectively in Q (4 mL/min), Q (8 mL/min) and Q (12 mL/min). And based on Table 6, the \(t_{\rm b}\) (min) and \(t_{\rm e}\) were 20, 30 and 40, and 90, 100 and 110 for ACT removal respectively in Q (4 mL/min), Q (8 mL/min) and Q (12 mL/min).
+
+Another vital parameters during assessing the performance of fixed-bed column is initial concentration of MPs, MTZ and EETC decreased with increasing initial concentration (mg/L) from 1 to 3. It may be proved with finding of Beji et al. (2018). Beji et al. (2018) expressed that at higher influent concentration, the breakthrough curves might be sharper and breakthrough occurred quicker. This may be clarified by the statement that higher adsorbent sites were being occupied with the rise of MPs concentration.
+
+### Desorption and reusability studies
+
+One of the ways to reduce cost is by reusing adsorbents in water and wastewater treatment. In regeneration-adsorbent situations, choosing the appropriate desorbents (i.e. inorganic desorbents (NaOH, H\({}_{2}\)SO\({}_{4}\) and HCl) or organic desorbents (ethanol, methanol and acetic acid)) is essential (Zhou et al., 2015). Pharmaceutical MPs are highly soluble in alcohols due to the presence of hydroxyl groups. Hence, methanol was selected as the desorbent in the current study. Mojiri et al. (2019) stated that NaOH and HCl cannot efficiently desorb MPs. Beside this claim, they stated that the restoration
+
+Figure 8: Langmuir and Freundlich isotherms Regression; Langmuir Regression for ACT (A) and AMX (B), Freundlich Regression for ACT (C) and AMX (D).
+
+capacity of methanol is higher than that of ethanol. In this study, after six cycles with initial concentrations of 2 mg/L the elimination effectiveness of the cross-linked magnetic chitosan/bentonite remained unaffected.
+
+## 4 Conclusion
+
+Pharmaceutical MPs affect human health and aquatic environments. However, conventional treatment techniques for domestic sewage treatment plants cannot effectively and completely remove pharmaceutical MPs. Hence, the performance of combined ozonation and cross-linked chitosan/bentonite in removing pharmaceuticals was investigated. The main achievements of this study are as follows:
+
+1. At the initial concentration of 0.2 mg/L and ozone dosage of 15 mg/L (i.e. first step), the ozone reactor removed 0.17 mg/L (84.8%) of ACT and 0.16 mg/L (82.7%) of AMX.
+
+2. The removal of ACT and AMX by using the ozone reactor was optimised by ANN. High R\({}^{2}\) and low MSE values confirmed that ANN can optimise ACT and AMX elimination.
+
+3. On the basis of the ozonation experiments, the rate of constants (m\({}^{-1}\)s\({}^{-1}\)) for ACT and AMX were 2.63\(\times\)10\({}^{6}\) and 5.98 \(\times\)10\({}^{6}\), respectively.
+
+4. After treating water in the ozone reactor, water was moved through a fixed-bed column (i.e. cross-linked chitosan/bentonite). The fixed-bed column also removed the remaining ACT and AMX.
+
+5. On the basis of the R\({}^{2}\) values, Freundlich and Langmuir isotherms can justify the removal of ACT and AMX from water by cross-linked chitosan/bentonite.
+
+6. Desorption studies were done by soaking cross-linked chitosan/bentonite in methanol. After six cycles, the elimination effectiveness of the cross-linked magnetic chitosan/bentonite remained unaffected.
+
+## Acknowledgments
+
+Authors would like to express their gratitude to the Institute of Scientific Research Amin-Azma Iran for providing fellowship with the reference number "A201901".
+
+## Appendix A Supplementary data
+
+Supplementary material related to this article can be found online at [https://doi.org/10.1016/j.eti.2019.100404](https://doi.org/10.1016/j.eti.2019.100404).
+
+## References
+
+* Abukhadra et al. (2019) Abukhadra, M.R., Adili, A., Bakry, B.M., 2019. Green fabrication of bentonite/chitosan@cobalt oxide composite (BE/CH@Co) of enhanced adsorption and advanced oxidation removal of Congo red dye and Cr (VI) from water. Int. J. Biol. Macromol. 126, 302-413. [http://dx.doi.org/10.1016/j.ijbiomac.2018.12.225](http://dx.doi.org/10.1016/j.ijbiomac.2018.12.225).
+* Ahile et al. (2015) Ahile, U.J., Adejo, S.O., Tughgba, M.S., Tyohemba, R.L., Ama, S.O., 2015. Kinetic and equilibrium studies for the adsorption of amoxicillin from aqueous solution on carbonized ground-nurit shells. In: 2nd International Conference on Chemical, Biological, and Environmental Sciences, ICGBES'15, May 20-21, Dubai (UAE). [http://dx.doi.org/10.17758/IAAST40515011](http://dx.doi.org/10.17758/IAAST40515011).
+* Ahmed et al. (2017) Ahmed, M.B., Zhou, J.L., Ngo, H.M., Guo, W., Thornidakis, N.S., Xu, J., 2017. Progress in the biological and chemical treatment technologies for emerging contaminant removal from wastewater: A critical review. J. Hazard. Mater. 323, 274-298. [http://dx.doi.org/10.1016/j.jhazmat.2016.04.045](http://dx.doi.org/10.1016/j.jhazmat.2016.04.045).
+* Alajmi (2014) Alajmi, H.M., 2014. Effect of physical, chemical and biological treatment on the removal of five pharmaceuticals from domestic wastewater in laboratories reactors and a full-scale plant (Msc. thesis). Newcastle University.
+* Alexander et al. (2018) Alexander, J.A., Ahmad Zaini, M.A., Surajuoden, A., Alju, E.N.L., Omezia, A.L., 2018. Surface modification of low-cost bentonite adsorbents--A review. Part. Sci. Technol. 37 (5), 8-13. [http://dx.doi.org/10.1080/0272635.12018.1438548](http://dx.doi.org/10.1080/0272635.12018.1438548).
+
+\begin{table}
+\begin{tabular}{l l l l l l l} \hline G\({}_{0}\) (mg/L) & Z (cm) & Q (ml/min) & t\({}_{0}\) (min) & t\({}_{\mathrm{e}}\) (min) & MTZ (cm) & EBCT (min) \\ \hline
+2.0 & 15 & 4 & 20 & 90 & 11.6 & 7.5 \\
+2.0 & 15 & 8 & 30 & 100 & 10.5 & 3.7 \\
+2.0 & 15 & 12 & 40 & 110 & 9.5 & 2.5 \\
+1.0 & 15 & 8 & 40 & 120 & 10.0 & 3.7 \\
+3.0 & 15 & 8 & 20 & 90 & 11.6 & 3.7 \\ \hline \end{tabular}
+\end{table}
+Table 6: Parameters in fixed-bed column for AMX adsorption by chitosan/bentonite.
+
+\begin{table}
+\begin{tabular}{l l l l l l l} \hline G\({}_{0}\) (mg/L) & Z (cm) & Q (ml/min) & t\({}_{0}\) (min) & t\({}_{\mathrm{e}}\) (min) & MTZ (cm) & EBCT (min) \\ \hline
+2.0 & 15 & 4 & 20 & 110 & 12.2 & 7.5 \\
+2.0 & 15 & 8 & 30 & 100 & 10.5 & 3.7 \\
+2.0 & 15 & 12 & 40 & 90 & 8.3 & 2.5 \\
+1.0 & 15 & 8 & 30 & 120 & 11.2 & 3.7 \\
+3.0 & 15 & 8 & 20 & 90 & 11.6 & 3.7 \\ \hline \end{tabular}
+\end{table}
+Table 5: Parameters in fixed-bed column for ACT adsorption by chitosan/bentonite.
+
+* Almonani et al. (2016) Almonani, F.A, Shawagdah, M., Bhosale, R.R., Kumar, A., 2016. Removal of emerging pharmaceuticals from wastewater by ozone-based advanced oxidation processes. Environ. Prog. Sustain. Energy 35 (4), 982-995. [http://dx.doi.org/10.1002/ep](http://dx.doi.org/10.1002/ep).
+* Attimazad (2011) Attimazad, M., 2011. Simultaneous determination of paracetamol and Iornoxicam by RF-HPLC in bulk and tablet formulation. Pharm. Methods 2 (1), 61-66. [http://dx.doi.org/10.4103/2229-4708.81084](http://dx.doi.org/10.4103/2229-4708.81084).
+* Beij et al. (2018) Beij, R., Hamidl, W., Sersavioui, A., Seffen, 2018. Adsorption of phosphorus by alkaline Tunisian soil in a fixed bed column. Water Sci. Technol. 78 (4), 751-763. [http://dx.doi.org/10.2166/wst.2018.341](http://dx.doi.org/10.2166/wst.2018.341).
+* Bekkari and Zedouri (2018) Bekkari, N., Zeddouri, A., 2018. Using artificial neural network for predicting and controlling the effluent chemical oxygen demand in wastewater treatment plant. Manag. Env. Quality 30 (3), 593-608. [http://dx.doi.org/10.1108/MEQ-04-2018-0084](http://dx.doi.org/10.1108/MEQ-04-2018-0084).
+* Ben et al. (2012) Ben, W., Qiang, Z., Pan, X., Nie, Y., 2012. Degradation of veterinary antibiotics by ozone in swine wastewater pretreated with sequencing batch reactor. J. Environ. Eng. 138 (3), 272-277.
+* Benitez et al. (2011) Benitez, F.J., Acero, J.L., Real, F.J., Roldan, G., Casas, F., 2011. Comparison of different chemical oxidation treatments for the removal of selected pharmaceuticals in water matrices. Chem. Eng. 1, 168 (3), 1149-2256. [http://dx.doi.org/10.1016/j.cej.2011.02.001](http://dx.doi.org/10.1016/j.cej.2011.02.001).
+* Cabut et al. (2016) Cabut, M., Alan, Y., Unal, H.I., 2016. Enhanced electrokinetic properties and antimicrobial activities of biodegradable chitosan/organo-bentonite composites. Carbohyd. Polym. 161, 71-81. [http://dx.doi.org/10.1016/j.carbonl.2016.12.067](http://dx.doi.org/10.1016/j.carbonl.2016.12.067).
+* Chowdhury et al. (2015) Chowdhury, Z.Z., Abd Hamidl, S.B., Zain, S.M., 2015. Evaluating design parameters for breakthrough curve analysis and kinetics of fixed bed columns for Cu(1) cations using lignocellulosic insecticides using ibcrosculosis. BioResource 10 (1), 732-749.
+* Das et al. (2014) Das, A., Hamdoui, O., 2014. Removal of non-steroidal anti-inflammatory drugs ibuprofen and ketoprofen from water by emulsion liquid membrane. Environ. Sci. Pollut. Res. 21, 2154-2164. [http://dx.doi.org/10.1007/s11356-013-2140-9](http://dx.doi.org/10.1007/s11356-013-2140-9).
+* Demirezen et al. (2019) Demirezen, D.A., Yildiz, I.S., Yilmaz, D.D., 2019. Amoxicillin degradation using green synthesized iron oxide nanoparticles: Kinetics and mechanism analysis. Env. Nanotech. Monitor. Manag. 11, 100219. [http://dx.doi.org/10.1016/j.exom.2019.100219](http://dx.doi.org/10.1016/j.exom.2019.100219).
+* Dotto et al. (2016) Dotto, G.L., Rodrigues, F.K., Tanabe, E.H., Frohlich, R., Berutoh, D.A., Martins, T.R., Foletto, E.L., 2016. Development of chitosan/bentonite hybrid composite to remove hazardous anionic and cationic dyes from colored effluents. J. Environ. Chem. Eng. 4, 3230-3239. [http://dx.doi.org/10.1016/j.jece.2016.07.004](http://dx.doi.org/10.1016/j.jece.2016.07.004).
+* Di-Dib et al. (2016) Di-Dib, F.I., Tawfik, F.M., Eshao, G., Hefni, H.H.H., ElMetwally, A.E., 2016. Remediation of distilleries wastewater using chitosan immobilized Bentonite and Bentonite based organocyles. Int. J. Biol. Macromol. 86, 750-755. [http://dx.doi.org/10.1016/j.ijbiomac.2016.01.108](http://dx.doi.org/10.1016/j.ijbiomac.2016.01.108).
+* El Najjar et al. (2014) El Najjar, N.H., Touffet, A., Deborde, M., Journel, R., Vel Leitner, N.K., 2014. Kinetics of paracetamol oxidation by ozone and hydroxyl radicals, formation of transformation products and toxicity. Sep. Puril. Technol. 136, 137-143. [http://dx.doi.org/10.1016/j.seppur.2014.09.004](http://dx.doi.org/10.1016/j.seppur.2014.09.004).
+* Evans et al. (2019) Evans, A.E., Mateo-Sageas, J., Qadir, M., Boelee, E., Ippolito, A., 2019. Agricultural water pollution: key knowledge gaps and research needs. Curr. Opin. Environ. Sustain. 36, 20-27. [http://dx.doi.org/10.1016/j.cosus12.018.10.003](http://dx.doi.org/10.1016/j.cosus12.018.10.003).
+* de Francu et al. (2017) de Francu, M.A., de Carvalho, C.B., Bonetto, M.M., Soares, R.D.P., Feris, LA, 2017. Removal of amoxicillin from water by adsorption onto activated carbon in batch process and fixed bed column: Kinetics, isotherms, experimental design and breakthrough curves modelling. J. Clean. Prod. 161, 947-956. [http://dx.doi.org/10.1016/j.jclepp.2017.05.197](http://dx.doi.org/10.1016/j.jclepp.2017.05.197).
+* Gadekar et al. (2019) Gadekar, M.R., Ahammed, M.M., 2019. Modelling dye removal by adsorption onto water treatment residuals using combined response surface methodology-artificial neural network approach. J. Environ. Manag. 231, 241-248. [http://dx.doi.org/10.1016/j.jenvman.2018.10.017](http://dx.doi.org/10.1016/j.jenvman.2018.10.017).
+* Gomez et al. (2017) Gomez, J., Costa, R., Quinta-Ferreira, R.M., Martins, R.C., 2017. Application of ozonation for pharmaceuticals and personal care products removal from water. Sci. Total Environ. 586, 265-283. [http://dx.doi.org/10.1016/j.scitometry.2017.01.216](http://dx.doi.org/10.1016/j.scitometry.2017.01.216).
+* Cuclu and Dursun (2010) Cuclu, D., Dursun, S., 2010. Artificial neural network modelling of a large-scale wastewater treatment plant operation. Bioproc. Biosyst. Eng. 33 (9), 1051-1058. [http://dx.doi.org/10.1007/s00449-010-0430-x](http://dx.doi.org/10.1007/s00449-010-0430-x).
+* Ozone lifetime and required ozone dose. Chem. Eng. J. 290, 507-514. [http://dx.doi.org/10.1016/j.ccp.2016.01.027](http://dx.doi.org/10.1016/j.ccp.2016.01.027).
+* He et al. (2017) He, J., Li, Y., Cai, X., Chen, K., Zheng, H., Wang, C., Zhang, K., Lin, D., Kong, L., Liu, J., 2017. Study on the removal of organic micropollutants from aqueous and ethanol solutions by HAP membranes with tunable hydrophilicity and hydrophobicity. Chemosphere 174, 380-389. [http://dx.doi.org/10.1016/j.chemosphere.2017.02.008](http://dx.doi.org/10.1016/j.chemosphere.2017.02.008).
+* Huang et al. (2018) Huang, Y., Lee, X., Macazo, F.C., Grattieri, M., Cai, R., Mintner, S.D., 2018. Fast and efficient removal of chromium (VI) anionic species by a reusable chitosan-modified multi-walled carbon nanotube composite. Chem. Eng. J. 339, 259-267. [http://dx.doi.org/10.1016/j.ccp.2018.01.133](http://dx.doi.org/10.1016/j.ccp.2018.01.133).
+* Huang et al. (2017) Huang, Z., Li, Y., Chen, W., Shi, J., Zhang, N., Wang, X., Li, Z., Gao, L., Zhang, Y., 2017. Modified bentonite adsorption of organic pollutants of dye wastewater. Mater. Chem. Phys. 202, 266-276. [http://dx.doi.org/10.1016/j.matchemphys.2017.09.028](http://dx.doi.org/10.1016/j.matchemphys.2017.09.028).
+* Ikehata et al. (2006) Ikehata, K., Naghashkar, N.J., El-Din, M.G., 2006. Degradation of aqueous pharmaceuticals by ozonation and advanced oxidation processes: A review. Ozone: Sci. Eng. 28 (6), 353-414. [http://dx.doi.org/10.1080/01919510600985937](http://dx.doi.org/10.1080/01919510600985937).
+* Jin et al. (2012) Jin, X., Pedkaya, S., Huck, P.M., 2012. Reaction kinetics of selected micropollutants in ozonation and advanced oxidation processes. Water Res. 46, 6519-6530. [http://dx.doi.org/10.1016/j.wattres.2012.09.026](http://dx.doi.org/10.1016/j.wattres.2012.09.026).
+* Jung et al. (2015) Jung, C., Oh, J., Yoon, Y., 2015. Removal of acetaminophen and naproxen by combined coagulation and adsorption using biochar: influence of combined sewer overflow components. Environ. Sci. Pollut. Res. 22 (13), 10058-10069. [http://dx.doi.org/10.1007/s11356-015-4191-6](http://dx.doi.org/10.1007/s11356-015-4191-6).
+* Katarzyn and Tsyzczuk-Rotko (2018) Katarzyn, C., Tsyzczuk-Rotko, K., 2018. Visible-light-driven photocatalytic removal of acetaminophen from water using a novel MWCNT-TiO2-SiO2 photocatalysts. Sep. Puril. Technol. 206, 343-355. [http://dx.doi.org/10.1016/j.seppur.2018.06.025](http://dx.doi.org/10.1016/j.seppur.2018.06.025).
+* Kdak and Dogan (2018) Kdak, R., Dogan, S., 2018. Medium-high frequency ultrasound and ozone based advanced oxidation for amoxicillin removal in water. Ultrasonic Sonochem. 40, 131-139. [http://dx.doi.org/10.1016/j.altnson.2017.01.033](http://dx.doi.org/10.1016/j.altnson.2017.01.033).
+* Kurniawan et al. (2017) Kurniawan, T.A., Io, W.H., Chan, G.Y.S., 2017. A concentration of recalcitant compounds from stabilized landfill leachate using a combination of ozone-GAC adsorption treatment. J. Hazard. Mater. 137 (1), 443-455. [http://dx.doi.org/10.1016/j.jhazmat.2006.02.020](http://dx.doi.org/10.1016/j.jhazmat.2006.02.020).
+* Kuzmanovic et al. (2013) Kuzmanovic, M., Banjac, Z., Ginebreda, A., Petrovic, M., Barcelo, D., 2013. Prioritization: Selection of environmentally occurring pharmaceuticals to be monitored. In: Petrovic, M., Perez, S., Barcelo, D. (Eds.), Analysis, Removal, Effects and Risk of Pharmaceuticals in the Water Cycle. Elsevier, ISBN: 978-0-4446-2657-8.
+* Lempart et al. (2018) Lempart, A, Kudlek, E., Dudziak, M., 2018. Determination of micropollutants in water samples from swimming pool systems. Water 10 (8), 1083. [http://dx.doi.org/10.3390/w10081083](http://dx.doi.org/10.3390/w10081083).
+* Ling et al. (2016) Ling, Y., Liao, C., Xu, P., Li, 2016. Fast mineralization of acetaminophen by highly dispersed Ag-g-GN4 hybrid assisted photocatalytic ozonation. Sep. Puril. Technol. 216, 1-8. [http://dx.doi.org/10.1016/j.seppur.2019.01.057](http://dx.doi.org/10.1016/j.seppur.2019.01.057).
+* Liu et al. (2016) Liu, J., Sun, Q., Zhang, C., Li, H., Song, W., Zhang, N., Jia, X., 2016. Removal of typical antibiotics in the advanced treatment process of productive drinking water. Desalin. Water Treat 57 (24), 11386-11391. [http://dx.doi.org/10.1080/19443994.2015.1040848](http://dx.doi.org/10.1080/19443994.2015.1040848).
+* Liu et al. (2016) Liu, P., Zhang, H., Feng, Y., Yang, F., Zhang, J., 2016. Removal of trace antibiotics from wastewater: A systematic study of nanofiltration combined with ozone-based advanced oxidation processes. Chem. Eng. J. 240, 211-220. [http://dx.doi.org/10.1016/j.cep.2013.11.057](http://dx.doi.org/10.1016/j.cep.2013.11.057).
+* Lozano-Morales et al. (2018) Lozano-Morales, V., Gardi, I., Nir, S., Undabeytia, T., 2018. Removal of pharmaceuticals from water by clay-cation starch sorbents. J. Clean. Prod. 190, 703-711. [http://dx.doi.org/10.1016/j](http://dx.doi.org/10.1016/j).
+
+* Mashayekh-Salehi and Moussavci (2016) Mashayekh-Salehi, A., Moussavci, G., 2016. Removal of acetaminophen from the contaminated water using adsorption onto carbon activated with NH4Cl Desalin. Water Treat 57 (27), 12861-12873. [http://dx.doi.org/10.1080/19443994.2015.1051588](http://dx.doi.org/10.1080/19443994.2015.1051588).
+* Mojiri et al. (2017a) Mojiri, A., Ahmad, Z., Tajuddin, M.R., Arshad, M.F., Barrera, V., 2017a. Molybdenum(VI) removal from aqueous solutions using bentonite and powdered cockle shell; optimization by response surface methodology. Global NEST J. 19 (2), 232-240. [http://dx.doi.org/10.3095/jsnj.001971](http://dx.doi.org/10.3095/jsnj.001971).
+* Mojiri et al. (2019) Mojiri, A., Kazeroon, R.A., Gholami, A., 2019. Cross-linked magnetic chitosan activated biochar for removal of emerging micropollutants from water: Optimization by the artificial neural network. Water 11 (3), 551. [http://dx.doi.org/10.3390/w1130551](http://dx.doi.org/10.3390/w1130551).
+* Mojiri et al. (2017b) Mojiri, A., Ziyang, L., Hui, W., Ahmad, Z., Tajuddin, R.M., Abu Amr, S.S., Kindachi, T., Aziz, H.A., Farraji, H., 2017b. Concentrated landfill leachate treatment with a combined system including eicro-ozonabin and composite adsorbent augmented sequencing batch reactor process. Process. Saf. Environ. Protect 111, 253-262. [http://dx.doi.org/10.1016/j.spec.2017.07.013](http://dx.doi.org/10.1016/j.spec.2017.07.013).
+* Moreira et al. (2016) Moreira, N.F.F., Sousa, J.M., Macedo, G., Ribeiro, A.R., Barreiros, L., Pedrosa, M., Faria, J.L., Pereira, M.F.P., Castro-Silva, S., Segundo, M.A., Manaia, C.M., Nunes, O.C., Silva, A.M., 2016. Photocatalytic ozonation of urban wastewater and surface water using immobilized TiO2 with LEDs: Micropollutants, antibiotic resistance genes and estrogenicity. Water Res. 94, 10-22. [http://dx.doi.org/10.1016/j.watres.2016.02.003](http://dx.doi.org/10.1016/j.watres.2016.02.003).
+* Muthanilsehi et al. (2018) Muthanilsehi, P., Karthikeyan, R., Kapoor, A., Prabhakar, S., 2018. Continuous fixed-bed studies for adsorptive remediation of phenol by garlic peel powder. Int. J. Indus. Chem. 9 (4), 379-390. [http://dx.doi.org/10.1007/s40090-018-0166-z](http://dx.doi.org/10.1007/s40090-018-0166-z).
+* Nagendra and Khare (2006) Nagendra, S.M.s, Khare, M., 2006. Artificial neural network approach for modelling nitrogen dioxide dispersion from vehicular exhaust emissions. Ecol. Model 190, 99-115. [http://dx.doi.org/10.1016/j.enrod.2005.01.062](http://dx.doi.org/10.1016/j.enrod.2005.01.062).
+* Noumoradi et al. (2018) Noumoradi, H., Moghadam, K.F., Jafari, A., Kamarehe, B., 2018. Removal of acetaminophen and ibuprofen from aqueous solutions by activated carbon derived from Quercus Brantii (Oak) acorn as a low-cost biosorbent. J. Environ. Chem. Eng. 6 (6), 6807-6815. [http://dx.doi.org/10.1016/j.jece.2018.10.047](http://dx.doi.org/10.1016/j.jece.2018.10.047).
+* Odabasi and Buyukgungor (2016) Odabasi, S.U., Buyukgungor, H., 2016. Removal of micropollutants in water with advanced treatment processes. In: 1st International black Sea Congress on Environmental Sciences (IBCESS). August 31-September 3, Giresun Turkey. p. 7.
+* Prieto-Rodriguez et al. (2013) Prieto-Rodriguez, L., U., Ilamet, I., Ilamet, N., Aguera, A., Rodriguez, E.M., Malato, S., 2013. Application of solar AOPs and ozonation for elimination of micropollutants in municipal wastewater treatment plant effluents. Water Res. 47, 1521-1528. [http://dx.doi.org/10.1016/j.watres.2012.11.002](http://dx.doi.org/10.1016/j.watres.2012.11.002).
+* Rame et al. (2018) Rame, R., Pranoto, H., Winahyu, R.K.K., Sofie, M., Raharjo, B.H., Utomo, A.S., 2018. Catalytic Ozonation Based Advanced Oxidation Process for Effective Treating Wastewater from Hospital and Community Health Centre Facility by FLASH WWT Catalyst System in Indonesia. In: IOP Conf. Series: J. Physics: Conf. Series, vol. 1095, p. 012030. [http://dx.doi.org/10.1088/1742-6596/1095/1/012030](http://dx.doi.org/10.1088/1742-6596/1095/1/012030).
+* Rijith et al. (2016) Rijith, S., Anirudhan, T.S., Sumi, V.S., Anitha, P.K., Abhilsash, S., Shibii, S.M.A., 2016. Carbohydrate functionalized chitosan/pentonite composite matrix as a cation exchanger for the removal of Pb(II) from aqueous media: Kinetic and equilibrium studies. Orient. J. Chem. 31 (2), 1113-1120. [http://dx.doi.org/10.1305/joc/ijoc/310263](http://dx.doi.org/10.1305/joc/ijoc/310263).
+* Rodayan et al. (2014) Rodayan, A., Segura, P.A., Vargaeva, V., 2014. Ozonation of wastewater: Removal and transformation products of drugs of abuse. Sci. Total Environ. 487, 763-770. [http://dx.doi.org/10.1016/j.scitotenv.2013.11.023](http://dx.doi.org/10.1016/j.scitotenv.2013.11.023).
+* Roza et al. (2017) Roza, O., Baeza, C., Nunez, K., Rossner, A., Urutria, R., Mansilla, H.D., 2017. Organic micropollutants (OMPs) oxidation by ozone: Effect of activated carbon on toxicity abatement. Sct. Total Environ. 590/591, 430-439. [http://dx.doi.org/10.1016/j.scitotenv.2016.12.120](http://dx.doi.org/10.1016/j.scitotenv.2016.12.120).
+* Rua-Gomez et al. (2012) Rua-Gomez, P.C., Guedez, A.A., Ania, C.O., Puttmann, W., 2012. Upgrading of wastewater treatment plants through the use of unconventional treatment technologies: Removal of lidocaine, tramadol, venlafaxine and their metabolites. Water 4, 650-669. [http://dx.doi.org/10.3390/w4030650](http://dx.doi.org/10.3390/w4030650).
+* Silva Ribeiro et al. (2018) Silva Ribeiro, D.T., Grosel, C.D., Merma, A.G., dos Santos, B.F., Torem, M.L., 2018. Removal of boron from mining wastewaters by electrocoagulation method: modelling experimental data using artificial neural networks. Miner. Eng. 131, 8-13. [http://dx.doi.org/10.1016/j.mineng.2018.10.016](http://dx.doi.org/10.1016/j.mineng.2018.10.016).
+* Snyder et al. (2006) Snyder, S., Wert, E.C., Rexing, D.J., Zegers, R.E., Drury, D.D., 2006. Ozone oxidation of endocrine disruptors and pharmaceuticals in surface water and wastewater. Ozone: Sc. Eng. 28, 445-460. [http://dx.doi.org/10.1080/019150601039726](http://dx.doi.org/10.1080/019150601039726).
+* Soros et al. (2019) Soros, A., Ambuyere, J.E., Stauber, C.C., Sobsey, M.D., Casanova, L.M., 2019. Torbidity reduction in drinking water by coagulation-flocculation with chitosan polymers. J. Water Health 17 (2), 204-218. [http://dx.doi.org/10.1266/w1b.2019.114](http://dx.doi.org/10.1266/w1b.2019.114).
+* Stavbar et al. (2017) Stavbar, S., Hrncic, M.K., Premzl, K., Kolar, M., Turk, S.S., 2017. Sub- and super-critical water oxidation of wastewater containing amoxicillin and ciprofloxacin. J. Supercrit. Fluid 128, 73-78. [http://dx.doi.org/10.1016/j.surbin.2017.05.013](http://dx.doi.org/10.1016/j.surbin.2017.05.013).
+* Tanzifi et al. (2017) Tanzifi, M., Hosseini, S.H., Kiedati, A.D., Olazar, M., Karimipour, K., Rezizenhem, R., Ali, L., 2017. Artificial neural network optimization for methyl orange adsorption onto polyaniline nano-adsorbent: Kinetic, isotherm and thermodynamic studies. J. Mol. Liq. 244, 189-200. [http://dx.doi.org/10.1016/j.jmolio.2017.08.12](http://dx.doi.org/10.1016/j.jmolio.2017.08.12).
+* Ternes et al. (2003) Ternes, T.A., Stuber, J., Nadine Herrmann, N., McDowell, D., Ried, A., Kampmann, M., Teiser, B., 2003. Ozonation: a tool for removal of pharmaceuticals, contrast media and mask fragrances from wastewater? Water Res. 37 (8), 1976-1982. [http://dx.doi.org/10.1016/S0043-1354/02/00570-5](http://dx.doi.org/10.1016/S0043-1354/02/00570-5).
+* Uddin et al. (2018) Uddin, M.K., Khan Rao, R.A., Chandra Mouli, K.V.V., 2018. The artificial neural network and Box-Behnken design for Cu2+ removal by the posterior sludge from water samples: Equilibrium, kinetic and thermodynamic studies. J. Mol. Liq. 266, 617-627. [http://dx.doi.org/10.1016/j.mollio.2018.06.08](http://dx.doi.org/10.1016/j.mollio.2018.06.08).
+* Valcarci et al. (2012) Valcarci, V., Martinez, F., Gonzalez-Alonso, S., Segura, Y., Catala, M., Molina, R., Montero-Rubio, J.C., Mastrolanni, N., de Adda, M., Postigo, C., Barcelo, D., 2012. Drugs of abuse in surface and tap waters of the taps river basin: heterogeneous photo-fenton process is effective in their degradation. Environ Int 41, 35-43. [http://dx.doi.org/10.1016/j.envm.2011.12.006](http://dx.doi.org/10.1016/j.envm.2011.12.006).
+* Wang et al. (2019) Wang, S., Wu, J., Lu, X., Gong, Q., Ding, J., Dan, B., Rie, P., 2019. Removal of acetaminophen in the Fe\({}^{2+}\)/persulfate system: Kinetic model and degradation pathways. Chem. Eng. J. 358, 1091-1100. [http://dx.doi.org/10.1016/j.cej.2018.09.145](http://dx.doi.org/10.1016/j.cej.2018.09.145).
+* Weng et al. (2018) Weng, X., Cai, W., Lan, R., Sun, Q., Chen, Z., 2018. Simultaneous removal of amoxicillin, ampicillin and penicillin by clay supported Fe/Ni bimetallic nanoparticles. Environ. Poll. 236, 562-569. [http://dx.doi.org/10.1016/j.envol.2018.01.100](http://dx.doi.org/10.1016/j.envol.2018.01.100).
+* Xiao et al. (2018) Xiao, Y., Li, Y., Ning, Z., Li, P., Yang, P., Liu, C., Liu, Z., Xu, F., Hynds, P.D., 2018. Organic contaminant removal efficiency of sodium bentonite/clay (BC) mixtures in high permeability regions utilizing reclaimed wastewater: A meso-scale study. J. Contam. Hydrol. 210, 1-14. [http://dx.doi.org/10.1016/j.jenvol.2018.01.008](http://dx.doi.org/10.1016/j.jenvol.2018.01.008).
+* Xu et al. (2012) Xu, J., Wang, L., Zhu, Y., 2012. Decontamination of bisphenol a from aqueous solution by graphene adsorption. Langmuir 28 (22), 8118-8425. [http://dx.doi.org/10.1012/la301476p](http://dx.doi.org/10.1012/la301476p).
+* Yadav et al. (2015) Yadav, V.B., Gadi, R., Kalra, S., 2015. Clay based nanocomposites for removal of heavy metals from water: A review. J. Environ. Manag. 232, 803-817. [http://dx.doi.org/10.1016/j.jenvm.2018.11.20](http://dx.doi.org/10.1016/j.jenvm.2018.11.20).
+* Yaghmare et al. (2017) Yaghmare, K., Moussavci, G., Mashayekh-Salehi, A., Mohseni-Bandpei, A., Satari, M., 2017. Oxidation of acetaminophen in the ozonation process catalyzed with modified mgo nanoparticles: Effect of operational variables and cytotoxicity assessment. Process. Saf. Environ. Prot. 109, 520-528. [http://dx.doi.org/10.1016/j.jspec.2017.04.020](http://dx.doi.org/10.1016/j.jspec.2017.04.020).
+* Yangan et al. (2018) Yangan, L., Kurimawan, T.A., Albadarim, A.B., Walker, G., 2018. Enhanced removal of acetaminophen from synthetic wastewater using multi-walled carbon nanotubes (MWCNTs) chemically modified with NaOH, HNO3/\({}_{H}\)2SO4, ozone, and/or chitosan. J. Mol. Liq 251, 369-377. [http://dx.doi.org/10.1016/j.mollio.2017.12](http://dx.doi.org/10.1016/j.mollio.2017.12).

manual_annotation/200pdfs_in_mmd/10.1016_j.fct.2005.08.015.mmd

@@ -0,0 +1,148 @@
+# Degradation of trichothecene mycotoxins by aqueous ozone
+
+J. Christopher Young
+
+Corresponding author. Tel.: +519 780 8033; fax: +519 829 2600.
+
+Honghui Zhu
+
+Food Research Program, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, Ont., Canada N1G 5C9
+
+Ting Zhou
+
+Food Research Program, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, Ont., Canada N1G 5C9
+
+24 January 200524due to a matrix effect. The corn was ground and thus porous whereas ozone may not have been able to penetrate the whole wheat kernels as readily.
+
+More recently, ozone gas was demonstrated as being effective in chemically modifying (McKenzie et al., 1997) a variety of non-trichothecene mycotoxins (aflatoxins B1, B2, G1, and G2, cyclopiazonic acid, fumanisin B1, ochratoxin A, patulin, secalonic acid, and zearalenone) and reducing their biological activity in the bioassays used (McKenzie et al., 1998; Lemke et al., 1999). More studies are required to establish the extent of hazard reduction. Ozonation of triacetoxyscirpenol was conducted as part of a structural determination (Sigg et al., 1965).
+
+Ozone is an unstable and a strong oxidant, capable of reacting with a wide variety of contaminants in water, which makes it suitable for the treatment of waste and drinking water to remove undesirable contaminants (Hoigne, 1998). Two primary oxidation pathways in water have been proposed (Hoigne, 1982; Staehelin and Hoigne, 1982; Staehelin and Hoigne, 1985): direct oxidation by molecular ozone and indirect oxidation by free radicals, which are formed during the ozone decomposition in water. The relative importance between these two pathways is affected by pH, UV light, ozone concentration, and the presence of radical scavengers. The later pathway can be further divided into three steps: initiation, propagation, and termination to form various free radicals. Among them is hydroxyl free radical, which is much more powerful an oxidant than ozone molecule itself. Thus, the reactions between these free radicals and organic contaminants are often considered to be much faster but less selective.
+
+Since the only previous study of the reactivity of ozone with trichothecenes was with DON and ozone in air, the purpose of this study was to investigate the reaction between ozone and DON in an aqueous medium and to determine if this is a general reaction with other trichothecenes. Furthermore, since the reactive species of aqueous ozone is dependant upon pH, we also studied the effect of pH on the course of trichothecene ozonation.
+
+## Materials and methods
+
+### Mycotoxin samples and reagents
+
+The mycotoxins 3-acetyldeoxynivalenol (3ADON), 15-acetyldeoxynivalenol (15ADON), diacetoxyscirpenol (DAS), 4-deoxynivalenol (DON), fusaremon X (FUS), HT-2 toxin (HT2), 15-monoacetoxyscirpenol (MAS), neosolanio1 (NEO), T-2 triol (T2T), and verrucarol (VER) were obtained from Sigma-Aldrich (Oakville, ON). The structures of the various mycotoxins are shown in Fig. 1. Methanol was obtained from Caledon (Georgetown, ON) and oxygen was obtained from Praxair (Kitchener, ON).
+
+### Ozone in water
+
+Water saturated with ozone was prepared by bubbling the gas, generated by passage of purified extra dry oxygen through a Triogen Model LAB2B generator (Ozonia North America, Elmwood Park, NJ), for 30 min through a 500 mL gas wash bottle containing NANOpure (Barnstead) water and chilled in an ice bath. For safety reasons, excess ozone was neutralized by passage through saturated aqueous sodium thiosulfate. The concentration of ozone in water was determined by the indigo colorimetric method of Bard and Hoigne (1981), which involves spectro-photometric (Milton Spectronic 1000 plus) measurement at 600 nm.
+
+### Ozonation of pure mycotoxins
+
+In a typical experiment with each of the mycotoxins, about 2.5 mg of mycotoxin was placed into a 1.8 mL glass vial and dissolved in water. Ozone saturated water of known concentration was then added in a ratio such that the total aqueous volume was 1.0 mL and gave the desired final concentration of ozone. The reaction was repeated with DON, MAS and T2T in 0.1 M ammonium acetate buffered solutions at pH 4-9.
+
+### Analysis of mycotoxins and reaction products
+
+The reaction mixtures were then analyzed directly without any work up by taking a 20 mL aliquot and injecting onto a 4.6 x 150 mm liquid chromatographic (LC) column packed with Agilent Zorbax Eclipse XDB-C18, 3.5 mm particle size. The column was eluted at 1 mL/min with a linear gradient of methanol-water changing from 1:3 to 3:1 over 15 min. Starting
+
+Fig. 1: Structures of the trichothecenes studied.
+
+materials and products were detected with a Finnigan SpectraSystem UV6000LP ultraviolet (UV) detector and a Finnigan LCQ Deca ion trap mass spectrometer (MS) operated in the atmospheric pressure chemical ionization (APCI) positive ion mode. The mass spectrometer was tuned for maximum response for DON. Machine operating conditions were as follows: shear gas and auxiliary flow rates were set at 80 and 0 (arbitrary units); voltages on the capillary, tube lens offset, multipole 1 offset, multipole 2 offset, lens, and entrance lens were set at 15.00, 30.00, \(-\)5.00, \(-\)7.00, \(-\)16.00, and \(-\)60.00 V, respectively; capillary and vaporizer temperatures were set at 200 and 450 degC, respectively; and the discharge needle current was set at 10 mA. Compounds were quantified on the basis of integrated peak areas using absorbance units (UV) or ion counts (MS). All reactions were run in triplicate and the results averaged.
+
+## 3 Results and discussion
+
+### Unbuffered ozonation
+
+At the outset, the mycotoxins were treated with saturated aqueous ozone (concentration typically about 25-30 ppm). The LC-UV-MS analyses showed that all of the starting material had been destroyed and no products could be detected. McKenzie et al. (1997) also remarked that no products were observed when their series of non-trichothecene mycotoxins were treated with ozone.
+
+The treatments were repeated except that molar ratios of ozone to trichothecenes ranging from \(\sim\)0.3 to 5.0 (equivalent to \(\sim\)0.14-1.8 ppm ozone in water) were chosen. The solutions were unbuffered and had a pH of 6.5. Under these conditions, the disappearance of the trichothecenes and appearance of products could be followed. Several phenomena were immediately apparent. Reaction with ozone reached its endpoint quickly; sequential reanalyses of several reaction mixtures revealed that there were no further changes over time when the product mixtures were compared with the initial analysis made shortly after mixing of reagents. In all instances, the appearance of products was transitory; they too disappeared as the ozone concentration increased. A final observation was that some mycotoxins disappeared more readily than others.
+
+Fig. 2 shows the disappearance profiles between the various trichothecenes for different amounts of ozone. The median relative standard deviations (RSD) of these determinations was 20.6%. The reaction between ozone and a trichothecene required less ozone for those compounds (DAS, MAS, and VER) lacking an oxygen at the carbon 8 position (Fig. 2a). Those trichothecenes with an allylic oxygen at C8 required more ozone (Fig. 2b and c).
+
+Fig. 2: Reductions, relative to starting material, in amounts of trichothecenes upon reaction with varying amounts of aqueous ozone. Data points represent averages of reactions run in triplicate and quantified by liquid chromatography–mass spectrometry. (a) Trichothecenes lacking oxygenation at C. • = recurcarcol, • = 15-monoacetylospermol; • = diacetylospermol; • = diacetylospermol; • = diacetylospermol; • = diacetylospermol; • = 15-monoacetylospermol; • = diacetylospermol; •Table 1 shows the relative amounts of ozone required to effect a 50% reduction (R50) in each trichothecene and confirms this effect. The mean R50 values depended upon the oxidation state of C8 and increased from methylene (no oxygen) \(<\) hydroxyl (free or esterified) \(<\) keto. Hoigne and Bader (1983) observed a 1.8-fold increase in rate of reaction with ozone when the allylic hydroxyl in hex-1-ene-3-ol was moved further away from the double bond (hex-1-ene-4-ol). These results are consistent with the well established electrophilic nature of attack of ozone on
+
+\begin{table}
+\begin{tabular}{l l l l} \hline \hline Groups & Trichothecenes & Mean\({}^{\pm}\pm\) std err & Group means\({}^{\mathrm{h,c}}\) \\ \hline
+8-methylene & VER & 0.65 \(\pm\) 0.09 & 0.77 \(\pm\) 0.06\({}^{\mathrm{a}}\) \\  & MAS & 0.75 \(\pm\) 0.05 & \\  & DAS & 0.93 \(\pm\) 0.16 & \\
+8-hydroxy & NEO & 1.14 \(\pm\) 0.12 & 1.40 \(\pm\) 0.11\({}^{\mathrm{b}}\) \\  & HT2 & 1.36 \(\pm\) 0.23 & \\  & T2T & 1.67 \(\pm\) 0.28 & \\
+8-keto & DON & 1.44 \(\pm\) 0.05 & 1.75 \(\pm\) 0.13\({}^{\mathrm{e}}\) \\  & FUS & 1.50 \(\pm\) 0.10 & \\  & 3ADON & 1.62 \(\pm\) 0.16 & \\  & 15ADON & 2.44 \(\pm\) 0.26 & \\ \hline \end{tabular}
+
+* Means of three replicates of each compound and their standard errors.
+* Averages of means of each group and their standard errors.
+* Means with different letter in the column are significantly different according to the Fisher’s protected LSD test at \(P=0.05\) level (UNISTAT 5.5, London, UK).
+
+\end{table}
+Table 1: Molar ratios of ozone/trichothecene required to effect a 50% reduction in trichothecene levels
+
+\begin{table}
+\begin{tabular}{l l} \hline \hline Trichothecene & Ozonation products \\ \hline DON & 2.79\({}^{\mathrm{a}}\) (100)\({}^{\mathrm{b}}\); 2.92 (44); 3.21 (8); 3.59 (6); 4.16 (15) \\
+15ADON & 3.10 (64); 4.05 (100); 4.52 (34); 5.39 (23); 6.44 (43); 7.08 (82) \\
+3ADON & 6.38 (8); 6.77 (26); 7.03 (100) \\ FUS & 3.20 (7); 3.48 (100); 3.90 (3); 4.29 (4); 4.96 (5) \\ NEO & 4.25 (15); 5.14 (7); 5.45 (57); 6.10 (100) \\ T2T & 8.18 (35); 8.34 (53); 9.54 (83); 9.98 (32); 10.80 (47); 11.47 (100) \\ HT2 & 10.80 (48); 11.11(100); 11.90 (31); 12.54 (25); 12.90 (27); 13.47 (50) \\ DAS & 6.17 (95); 6.32 (100); 6.71 (21); 8.58 (26); 9.02 (24); 10.74 (34) \\ MAS & 2.63 (79); 2.88 (39); 3.09 (100), 3.38 (10); 3.89 (7) \\ VER & 2.11 (14); 2.50 (83); 2.76 (13); 3.09 (100); 3.35 (40); 3.46 (40); 3.69 (32); 4.13 (7) \\ \hline \end{tabular}
+
+* Retention times in minutes.
+* Relative amounts based on peak heights.
+
+\end{table}
+Table 2: Products from reaction of trichothecenes with unbuffered aqueous ozone
+
+Fig. 3: Variation in total amounts of products from reaction of trichothecenes with varying amounts of aqueous ozone. Values are relative to responses for the starting trichothecene taken as 100. Data points represent averages of reactions run in triplicate and quantified by liquid chromatography–mass spectrometry. (a) trichothecenes lacking oxygenation at C8. \(\bullet\) – semiconacid; \(\blacksquare\) – 15-monoacetylseripenol; \(\blacktriangle=\) diacetoxysicripenol; (b) trichothecenes with free or esterified hydroxide at C8: \(\bullet\) – neosaloniol; \(\blacksquare\) – T-2 trio; \(\blacktriangle=\) HT-2 toxin; (c) trichothecenes with a keto group at C8: \(\blacklozenge=\) deoxynivalenol; \(\blacksquare\) – fusarenon X; \(\blacksquare\) – 3-acetyldeoxynivalenol; \(\blacktriangle=\) 15-acetyldeoxynivalenol.
+
+olefins and the rate reducing effects of nearby electron-withdrawing groups (Bailey, 1978).
+
+At the lower ozone concentrations, the appearance of transitory products could be followed. Fig. 3 shows the relative amounts of the sum of all intermediate products for a given trichothecene as they are formed and subsequently degraded in unbuffered solution. The median RSD of these determinations was 21.2%. The oxidation state at C8 had an effect on the ratio of ozone required to form the intermediate products as well as result in their degradation. Trichothecenes lacking oxygen at C8 required less ozone to afford the intermediate products, which appeared in proportionally higher yield and then disappeared. The requirement for more ozone to reach maximum yield and then effect the disappearance of intermediates increased with allylic hydroxide and was even greater for a ketone at C8. The amounts of ozone needed for total destruction of identifiable products were also in the order methylene \(<\) hydroxyl \(<\) keto at C8 (data not shown). Table 2 shows that the mix of intermediate products at maximum yield for each of the trichothecenes studied was quite complex with three to eight compounds observed.
+
+The reactions with ozone did not terminate at the stage shown in Fig. 3. Additional ozone reacted with the intermediate products and presumably degraded them to much simpler molecules (such as acids, aldehydes, ketones, CO2) (Hoigne, 1998) that were not detected by UV or MS.
+
+### Site of ozonation
+
+The olefinic position is one of the most reactive sites for reaction of ozone with organic compounds (Bailey, 1978; Razumovskii and Zaikov, 1984 Hoigne, 1998; Rakovsky and Zaikov, 1998). There is an olefinic double bond at the C9-10 position of trichothecenes. Due to extended conjugation, the 8-keto trichothecenes (3ADON, 15ADON, DON, and FUS) are the only ones studied that exhibited significant UV absorption (at 217 nm). When monitored by either LC-MS or LC-UV, the disappearance profiles of these compounds in the presence of ozone were virtually superimposable, which suggests that reaction occurs at the C9-10 double bond. Furthermore, none of the products from these compounds exhibited any significant UV absorption.
+
+This site of reaction was also supported by MS data. Under the APCI \(+\)ve ion conditions employed, all the trichothecenes showed MS fragmentation patterns consistent with sequential loss of hydroxyl (as water) and/or acyl groups. Table 3 summarizes the observed fragmentation patterns. Reaction of ozone with olefins has been well studied (Bailey, 1978; Razumovskii and Zaikov, 1984; Rakovsky and Zaikov, 1998) and leads to the proposed mechanism for reaction with trichothecenes illustrated in Fig. 4. Two of the expected keto aldehyde products can be explained by isomerization at C11 during or immediately after formation of the aldehyde at C10. The final postulated product(s) arises from the net addition of two atoms of oxygen to the molecule. When the MS of the major product from each of the trichothecenes was offset by 32 mass units and compared with the MS of the starting compounds, there was virtually complete matching of the fragmentation patterns. See Table 3 for complete MS patterns of the major products. The combined UV and MS data suggest that aside for the change in oxidation state at C9-10, the remainder of the molecules was left intact at this stage of the reaction.
+
+This data is consistent with that reported for the reaction of pure ozone with triacetoxyscirpenol at \(-\)70 degC (Sieg et al., 1965), which resulted in the cleavage of the C9-10 double bond and enabled isolation of the intermediary ozonide. Catalytic hydrogenation over Pd subsequently afforded the proposed keto aldehyde characterized by infra red and proton nuclear magnetic resonance spectroscopy.
+
+### Buffered ozonation
+
+The above discussion is predicated on the mechanism of reaction being direct oxidation by molecular ozone. To test the possible indirect oxidation by free radicals formed during the ozone decomposition in water at elevated pH, the ozonation reaction of representative C8 methylene, hydroxyl, and keto trichothecenes (MAS, T2T, and DON, respectively) was studied over the range pH 4-9. Fig. 5 shows the disappearance of the trichothecenes at the different pH values. The median RSD of these determinations was 6.6%. Each of these mycotoxins was rapidly degraded by ozone at or below pH 6. The R50 values at pH 6 were each about 30% higher than the corresponding values given in Table 1. The oxidation state at C8 influenced the outcome of the reaction. MAS, with a methylene at this position, was the most prone to degradation at pH 7 and 8. Surprisingly, at pH 9, where OH' radicals might be expected to be present (Hoigne and Bader, 1978; Hoigne, 1982), there was little or no reaction with any of the trichothecenes. Because the reactions were analyzed by LC-MS, the reaction solution could only be buffered with relatively volatile ammonium acetate. OH' radicals are known to react with ammonia (Hoigne and Bader, 1978), so perhaps the buffer reagent scavenged the radicals before they could react with DON.
+
+Fig. 6 shows the relative amounts of the sum of all intermediate products. The median RSD of these determinations was 6.1%. The highest yields of total products were observed at pH 4 and 5, and essentially no products observed above pH 7. pH also influenced the mix of products: MAS afforded more polar products (earlier eluting) at pH 6
+
+Fig. 5: Reductions, relative to starting material, in amounts of trichothecenes upon reaction with varying amounts of aqueous ozone at different pH. Data points represent averages of reactions run in triplicate and quantified by liquid chromatography–mass spectrometry. \(\blacklozenge=\) pH 4; \(\Diamond=\) pH 5; \(\blacklozenge=\) pH 6; \(\Diamond=\) pH 7; \(\blacksquare=\) pH 8; \(\square=\) pH 9 (a) monoacetylospermol; (b) T-2 triol; (c) deoxyynivelon.
+
+Fig. 6: Reductions, relative to starting material, in amounts of trichothecenes upon reaction with varying amounts of aqueous ozone at different pH. Data points represent averages of reactions run in triplicate and quantified by liquid chromatography–mass spectrometry. \(\blacklozenge=\) pH 4; \(\Diamond=\) pH 5; \(\blacklozenge=\) pH 6; \(\Diamond=\) pH 7; \(\blacksquare=\) pH 8; \(\square=\) pH 9 (a) monoacetylospermol; (b) T-2 triol; (c) deoxyynivelon.
+
+and fewer polar products at pH 8; DON yielded different major products at pH 4 and 7.
+
+Several studies (see Sundstol Eriksen et al., 2004 and references cited therein) have shown that de-epoxy trichothenes are markedly less toxic than their parents. Since the epoxy moiety appears to have been retained in the intermediary degradation products, it is possible that there may be some residual toxicity. However, it is likely that under the more drastic conditions, the epoxide ring is also degraded.
+
+In summary, aqueous ozone has been shown to degrade a wide variety of trichothenes to presumably simple products. The identity of the breakdown products under the different conditions used in the study remain to be confirmed and their toxicity evaluated. Since aqueous ozone reacts quickly and leaves no residue, it shows promise as a reagent for decontamination of trichothenee-contaminated grains. Studies to determine the efficacy of this technology to such matrices are underway.
+
+## Safety note
+
+Ozone is a toxic gas, so all preparations were conducted in a fume hood. The trichothecenes are also toxic and appropriate personal protective equipment was worn when these materials were handled.
+
+## Acknowledgments
+
+The authors thank Kuanji Wei for technical assistance and Dr. Hongde Zhou (University of Guelph) for helpful discussions.
+
+## References
+
+* [1] Anon, 2003. Mycotoxins: Risks in plant, animal, and human systems. Task Force Report No. 139, Council for Agricultural Science and Technology, Ames, Iowa.
+* [2] Bailey, P.S., 1978. Ozonation in Organic Chemistry Volume 1 Olefinic Compounds. Academic Press, New York.
+* [3] Bard, H., Hoigne, J., 1981. Determination of ozone in water by the indigo method. Water Res. 15, 449-456.
+* [4] Buck, W.B., Cote, L.M., 1991. Handbook of Natural Toxins. In: Keeler, R.F., Tu, A.T. (Eds.), Toxicology of Plant and Fungal Compounds, Vol. 6. Marcel Dekker, Inc., New York, pp. 523-555.
+* [5] Charmley, L.L., Prelusky, D.B., 1994. Decontamination of _Fusarium_ mycotoxins. In: Miller, J.D., Trenholm, H.L. (Eds.), Mycotoxins in Grain: Compounds Other Than Aflatoxin. Eagan, St. Paul, MN.
+* [6] DeVries, J.W., Trucksess, M.W., Jackson, L.S., 2002. Mycotoxins and Food Safety: Proceedings of an American Chemical Society Symposium held in Washington, DC, USA, on 21-23 August 2000. Kluwer Academic Publishers, Dordrecht, Netherlands, p. 295.
+* [7] Hoigne, J., 1982. Mechanisms, rates and selectivities of oxidations of organic compounds initiated by ozonation in water. In: Rice, R.G., Netzer, A. (Eds.), Handbook of Ozone Technology and Applications. Ann Arbor Science Publishers, Ann Arbor, Mich.
+* [8] Hoigne, J., 1998. Chemistry of aqueous ozone and transformation of pollutants by ozonation and advanced oxidation processes. In: Hrubec, J. (Ed.), Quality and Treatment of Drinking Water II. Springer-Verlag, Berlin, pp. 83-141.
+* [9] Hoigne, J., Bader, H., 1978. Ozonation of water: kinetics of oxidation of ammonia by ozone and hydroxyl radicals. Environ. Sci. Technol. 12, 79-84.
+* [10] Hoigne, J., Bader, H., 1983. Rate constants of reactions of ozone and inorganic compounds in water--I Non-dissociating organic compounds. Water Res. 17, 173-183.
+* [11] Karlovsky, P., 1999. Biological detoxification of fungal toxins and its use in plant breeding, feed and food production. Nat. Toxins 7, 1-23.
+* [12] Lemke, S.L., Mayaru, K., Ottinger, S.E., McKenzie, K.S., Wang, N., Fickey, C., Kubena, L.F., Phillips, T.D., 1999. Assessment of the estrogenic effects of zearaleno after treatment with ozone utilizing the mouse uterine weight bioassay. J. Toxicol. Environ. Health Part A 56, 283-295.
+* [13] McKenzie, K.S., Kubena, L.F., Denvir, A.J., Rogers, T.D., Hitchens, G.D., Bailey, R.H., Harvey, R.B., Buckley, S.A., Phillips, T.D., 1998. Aflatoxicosis in turkey points is prevented by treatment of naturally contaminated corn with ozone generated by electrolysis. Poultry Sci. 77, 1094-1102.
+* [14] McKenzie, K.S., Sarr, A.B., Mayura, K., Bailey, R.H., Miller, D.R., Rogers, T.D., Norred, W.P., Voss, K.A., Plattner, R.D., Kubena, L.F., Phillips, T.D., 1997. Oxidative degradation and detoxification of mycotoxins using a novel source of ozone. Food Chem. Toxicol. 35, 807-820.
+* [15] Miller, J.D., Trenholm, H.L. (Eds.), 1994. Mycotoxins in Grain: Compounds Other Than Aflatoxin. Eagan, St. Paul, MN.
+* [16] Pittet, A., 1998. Natural occurrence on mycotoxins in foods and feeds--an updated review. Revue de Medecine Veterinaire 149, 479-492.
+* [17] Placinta, C.M., D'Mello, J.P.F., Macdonald, A.M.C., 1999. A review of worldwide contamination of cereal grains and animal feed with _Fusarium_ mycotoxins. Anim. Feed Sci. Technol. 78, 21-37.
+* [18] Rakovsky, S., Zaikov, G., 1998. Kinetics and Mechanism of Ozone Reactions with Organic and Polymeric Compounds in Liquid Phase. Nova Science Publishers Inc., Commack, NY.
+* [19] Razumovski, S.D., Zaikov, G.E., 1984. Ozone and its Reactions with Organic Compounds. Elsevier, New York.
+* [20] Sigg, H.P., Mauli, R., Flury, E., Hauser, D., 1965. Die Konstitution von Diacetoxyscirpenol. Helv. Chim. Acta 48, 962-988.
+* [21] Staehelin, J., Hoigne, J., 1982. Decomposition of ozone in water: rate of initiation by hydroxide ions and hydrogen peroxide. Environ. Sci. Technol. 16, 676-681.
+* [22] Staehelin, J., Hoigne, J., 1985. Decomposition of ozone in water in presence of organic solutes acting as promoters and inhibitors of radical chain reactions. Environ. Sci. Technol. 19, 120-126.
+* [23] Sundstol Eriksen, G., Pettersson, H., Lundh, T., 2004. Comparative cytotoxicity of deoxynivalent, nivalenol, their acetylated derivatives and de-epoxy metabolites. Food Chem. Toxicol. 42, 619-624.
+* [24] Trigo-Stockli, D.M., 2002. Effect of processing deoxynivalenol and other trichothences. In: DeVries, J.W., Trucksess, M.W., Jackson, L.S. (Eds.), Mycotoxins and Food Safety: Proceedings of an American Chemical Society Symposium held in Washington, DC, USA, on 21-23 August 2000. Kluwer Academic Publishers, Dordrecht, Netherlands, pp. 181-188.
+* [25] Young, J.C., 1986. Reduction in levels of deoxynivalenol in contaminated corn by chemical and physical treatment. J. Agric. Food Chem. 34, 465-467.
+* [26] Young, J.C., Subryan, L.M., Potts, D., McLaren, M.E., Gobran, F.H., 1986. Reduction in levels of deoxynivalenol in contaminated wheat by chemical and physical treatment. J. Agric. Food Chem. 34, 461-465.

manual_annotation/200pdfs_in_mmd/10.1016_j.foodcont.2015.09.013.mmd

@@ -0,0 +1,173 @@
+# Kinetic study of aflatoxins' degradation in the presence of ozone
+
+S. Agriopoulou
+
+A. Koliadima
+
+G. Karaiskakis
+
+J. Kapolos
+
+# Abstract
+
+The degradation of aflatoxins AFB1, AFB2, AFG1 and AFG2 after treatment of their solutions in triple distilled water with ozone was studied and the ability of ozone, even at low concentrations, to degrade aflatoxins was proved. Ozone concentrations of 8.5, 13.5, 20, 25 and 40 ppm were applied at different temperatures on aflatoxin solutions in triple distilled water of 10 ppb and 2 ppb and the complete and rapid elimination of AFB1 and AFG1 was observed while AFB2 and AFG2 remain more or less stable. The kinetic equations for the degradation procedure were calculated by applying ozone on neutral buffer solutions of aflatoxin at 298.15, 308.15, 318.15 and 328.15 K at ozone concentrations of 8.5, 13.5 and 20 ppm and the rate constants, were determined. The degradation of aflatoxins was described by a first order kinetic equation. Finally, the activation energies during degradation of aflatoxins were calculated from the Arrhenius equation.
+
+## 1 Introduction
+
+Aflatoxins (AFs) are widely distributed mycotoxins produced by strains of _Aspergillus flavus_, _Aspergillus parasiticus_ and _Aspergillus nominus_ (Kurtzman, Horn, & Hesseline, 1987). Additionally other species which produce aflatoxins are _Aspergillus pseudotamari_, _Aspergillus ochracoreseus_, _Aspergillus rambelli_, _Aspergillus toxicarius_, _Emericella austrella_, _Emericella olivicola_ and _Emericella venzeulensis_ (Reiter, Zentek, & Razzazi, 2009). Moreover other fungi of the genera _Aspergillus_ (e.g., _A. ochracuseus_ and _A. ochracunins_) produce another important mycotoxin, ochratoxin A (OTA) (Sarigiannis, Kapolos, Koliadima, Tsegenidis, & Karaiskakis, 2014). Aflatoxins are extremely harmful to the health of humans and animals, as showing carcinogenic, mutagenic, teratogenic and immunosuppressive actions (Zinedine & Maines, 2009). The known aflatoxins are about twenty, while the four main aflatoxins which are mainly found in tropical and subtropical climates (Inan, Pala, & Doymaz, 2007), are referred as aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin G1 (AFG1) and aflatoxin G2 (AFG2). The aflatoxins M1 (AFM1) and M2 (AFM2), which are hydroxylated metabolites of AFB1 and AFB2, can be found in all kinds of milk and dairy products from animals that have consumed contaminated feed with aflatoxins (Giray, Girgin, Engin, Aydm, & Sahin, 2007; Hussain & Anwar, 2008). Aflatoxins are frequently found in many human foodstuffs or animal feeds (Ali et al., 2005; Chiavaro et al., 2001; Nasir & Jolley, 2002; Park, 2002). From all the food-contaminating aflatoxins, AFB1 is usually the predominant mycotoxin (Jolly et al., 2006) and the International Agency for Research on Cancer (IARC) has classified aflatoxin AFB1 in the most toxic group (1 carcinogen), which primarily affects the liver (IARC, 1993). As food and health bodies have not established a tolerable daily intake (Tolerable Daily Intake, TDI) for humans, contamination in food, should be reduced to the lowest possible level (Pietri, Bertuzzi, Agosti, & Donadini, 2010).
+
+Several factors have been reported as means of reducing the levels of aflatoxins. These are natural, such as mechanical separation of the clean product from the contaminated product (Karaca, 2010), heating at high temperatures, effects of radiation (Karaca, 2010; Reddy et al., 2009) and light, grinding, washing, and use of adsorbents (Karaca, 2010) or chemicals, such as ammoniation (Reddy et al., 2009), influence of acids and bases, influence of oxidizing agents or with various inorganic and organic chemicals (Karaca, 2010). Between the oxidizing agents used for the reduction or elimination of aflatoxins, a powerful oxidant is ozone. Ozone is a highly unstable trioatomic oxygen molecule (O\({}_{3}\)) formed by the addition of an oxygen atom to a molecular diatomic oxygen (O\({}_{2}\)) that can be generated readily and economically for application to several commodities (USA, 1997). It is one of the most potential oxidants that has several advantages, most prominently the absence of detectable residues on treated products. Ozone can be generated on-site, eliminating the need to store or dispose ofchemical containers. Ozone has been used to reduce various types of mycotoxins, including the aflatoxins (Chen et al., 2014; Inan et al., 2007; Luo et al., 2014; Zorlugenc, KiroGlu Zorlugenc, Oztekin, & Evliya, 2008). Ozone reacts across the 8, 9 double bond of the terminal furan ring of aflatoxin through electrophilic attack, causing the formation of primary ozonides followed by rearrangement into monozonide derivatives such as aldehydes, ketones and organic acids (Inan et al., 2007).
+
+Although there have been many studies on ozone reactions with aflatoxins at many important agricultural products, very little kinetic data have been reported. In order to effectively treat aflatoxins with ozone and design appropriate treatment processes, reaction kinetics must be determined.
+
+In the present study the influence of ozone on the degradation of aflatoxins AFB1, AFB2, AFG1 and AFG2 in triple distilled water was investigated, as well as the appropriate kinetic equations for the above procedure in neutral buffer solutions were determined. Determination of activation energy was also performed from the Arrhenius equation, for the degradation of aflatoxins AFB1 and AFG1, in the presence of ozone, in neutral buffer solutions. As far as we know rate constants and activation energies for the degradation of aflatoxins in the presence of ozone are calculated for the first time.
+
+## Materials and methods
+
+### Chemicals and reagents
+
+Acetonitrile (ACN) and methanol (MeOH) (all of HPLC grade) were purchased from Merck (Darmstadt, Germany). Standards solutions of AFB1, AFB2, AFG1, AFG2, were obtained from the company R-Biopharm (Rhone Diagnostic Technologies Ltd., Glasgow, UK). All solutions were prepared with triple distilled water. Triple distilled water was prepared in a special ultra-clean water system (Purelab Ultra, Elga Labwater, Marlow, UK) for producing ASTM Type I. The freshly prepared mobile phase was degassed for 20 min before use with a Branson 2510 (Branson Ultrasonic Corp., CT, USA) ultrasonic bath. Phosphate buffer solution was prepared with potassium dihydrogen phosphate (34.02 g) and disodium hydrogen phosphate (35.40 g) (all purchased form Merck) in 1 L triple distilled water. The final pH of the solution was adjusting to 7 by adding a few drops of 0.5 M NaOH. The solution was then stored at 4 degC in a sealed flask until use.
+
+### Alatoxin standards in water
+
+In order to study the degradation of aflatoxins in triple distilled water in the presence of different concentrations of ozone, aflatoxin standards in water were prepared. The stock solution of the four aflatoxins AFB1, AFB2, AFG1 and AFG2, was reconstituted by dissolving the content of standard in 10 mL of pure acetonitrile, obtaining 5 mg/mL of total aflatoxins (2 mg/mL of each aflatoxin AFB1 and AFG1 and 0.5 mg/mL of each aflatoxin AFB2 and AFG2) and stored at 4 degC. The preparation of standard solutions of aflatoxins included in a first stage the preparation of a stock solution, concentration of 200 ng/mL in acetonitrile. From this solution, were prepared by appropriate dilutions in water, working solutions with concentration of 1, 1.5, 2.5, 5, 7.5 and 10 ng/mL and stored at 4 degC until use in order to construct the four calibration curves.
+
+Six-point calibration curves, with triplicate injections, covering the range of 1-10 ng/mL for each aflatoxin, were constructed. Calibration curves were drawn using the average response for each standard concentration employed.
+
+### Ozone production
+
+Ozone was produced by using an ozone generator (TG-10 Ozone Generator, Ozone Solutions, IA, USA) for the transformation of O2 molecules to O atoms and then forming O3 by using high voltage current. Ozone was canalized to a constant conditions environmental chamber (KBF 115, Binder, Tuttlingen, Germany) in order to control temperature and relative humidity and after that, through a two port valve, was lead to an ozone analyzer (Ozone analyzer Model UV-100, ECO SENSORS, INC.). When a constant concentration of ozone was measured, the gas stream was guided throw the two port valve to a jacked beaker which was maintained at a constant temperature by mean of a heating circulator (F12-ED, Julabo Seelbach, Germany), (cf. Fig. 1).
+
+For the ozonolysis experiments an appropriate volume of a stock solution, containing all the aflatoxins in CH3CN, was added in 50 mL volumetric flask. Two different aflatoxin solutions in triply distilled water were prepared with final concentrations of 10 ppb for AFB1 and AFG1 and 2.5 ppb for AFB2 and AFG2 (first solution) and 2 ppb for AFB1 and AFG1 and 0.5 ppb for AFB2 and AFG2 (second solution). After that, each solution was added in the jacked beaker which was maintained at a constant temperature (298.15 K, 308.15 K, 318.15 K, 328.15 K and 338.15 K). Each aflatoxin solution in triple distilled water was exposed for 20 min on gaseous stream containing ozone at five different concentrations of 8.5, 13.5, 20, 25 and 40 ppm.
+
+In order to obtain kinetic results for the degradation procedure, instead of triply distilled water, buffer solution at pH 7 was used for preparing the aflatoxin solutions of the concentration of 10 and 2 ppb. In contrast to the use of five ozone concentrations that were used for the ozonolysis, for the kinetic study the first three concentrations were used, because the two largest gave no results due to the rapid rate of the phenomenon of degradation. In order to study the influence of ozone on aflatoxin concentration as well as to determine the kinetic equations, every two minutes, a sample of 500 mL was collected and analyzed. For each experiment fresh aflatoxin solutions were prepared. The quantification of aflatoxins was calculated by the peak area according to the calibration curve.
+
+### HPLC analysis of aflatoxins
+
+The aflatoxin concentrations were determined by using an LC-20A Prominence HPLC system (Shimadzu, Kyoto, Japan) equipped with a spectrofluorometric detector (RF-20A/20Axs), a system controller (CBM-20A/20Alite), a column oven (CTO-20A/20AC), an autosampler (SIL-20AHT/20ACHT) and an on-line degasser (DCU-20A3/20A5). The spectrofluorometric detector was operated at an excitation wavelength of _l_ex = 365 nm and an emission wavelength of _l_em = 430 nm. Water-methanol-acentonitrile (56:22:22 v/v/v) at a flow rate of 1 mL/min under isocratic conditions was used as mobile phase, the pressure was 113-115 bar and the total run time was 17 min. Separation was completed by Mycotox(tm) reversed-phase column (C18, 4.6 x 250 mm, 5 mm) (Pickering Laboratories Inc., California, USA). The column temperature was set to 42 degC and sample volumes of 10 mL were injected in triplicate. The samples were analyzed after post-column derivatization with a Pinnacle PCX instrument (Pickering Laboratories Inc., California, USA). The reagent used was iodine (100 mg/L of l2 in water), the volume of the reactor was 1.4 mL, the flow rate was 0.3 mL/min and the reactor temperature was 90 degC. The solution of iodine was degassed using a nitrogen gas stream for 15 min. aflatoxin AFG2 was eluted first followed by AFG1, AFB2 and AFB1 with retention times of approximately 7.5, 8.6, 9.3 and 10.8 min, respectively. The chromatograms were analyzed by LC solution LC-Assist software (Shimadzu). Alfatoxin concentration was expressed in ug of aflatoxinper kg of each standard.
+
+## 3 Results and discussion
+
+In order to investigate the behavior of the aqueous solutions of the four aflatoxins as well as the response of the detector of the HPLC, calibration curves of different aflatoxins' concentrations, described in the experimental section, against their peak area were produced. The correlation coefficients, R2, for all calibration curves were better than 0.994 which indicated a good linear fitting of the experimental data. From the ozonolysis experiments the time for the complete degradation of AFB1 and AFG1 as well as the % percentage of the degradation of AFB2 and AFG2 were found. From the results the less resistant to ozonolysis aflatoxins proved to be the AFG1 and AFB1, while the more resistant proved to be the AFG2 and AFB2. Also, aflatoxin AFG1 seems to be the less resistant aflatoxin, whilst aflatoxin AFB2 is the most resistant. The sensitivity of aflatoxins in ozone was AFG1 > AFB1 > AFG2 > AFB2. AFB1 has the same behavior as AFG1 and AFB2 as AFG2. These differences in the degradation, agree with the literature (McKenzie et al., 1997) and can be attributed to the different chemical structure of AFB1 and AFG1 and especially to the double bond between the 8 and 9 carbon atom of the furan ring in the molecule of aflatoxins AFB1 and AFG1, contrary to aflatoxins AFB2 and AFG2.
+
+Specifically, aflatoxins AFB1 and AFG1 with initial concentration of 2 ppb and 10 ppb after treatment with 13.5 ppm ozone rapidly degraded at any temperature, after 3-7 min and 3-4 min, respectively (cf. Figs. 2 and 3).
+
+From the above two aflatoxins AFG1 was less resistant than the AFB1. The same behavior was observed when the concentrations of aflatoxins AFB1 and AFG1 increased 5 times from 2 to 10 ppb. In a study carried out by Maeba et al. (Maeba, Takamoto, Kamimura, & Miura, 1988) although different solvent was used (4% dimethyl sulfoxide), AFB1 and AFG1 were sensitive to ozone and easily degraded with 1.1 mg/L of ozone within 5 min at room temperature.
+
+Regarding the ozonolysis of aflatoxins in all treatment temperatures, at all concentrations of ozone and to all initial concentrations of aflatoxins, AFB2 and AFG2, proved to be more resistant than aflatoxins AFB1 and AFG1. The more sensitive to the ozonolysis between aflatoxin AFB2 and AFG2 seemed to be the AFG2, while the more stable appeared to be the AFB2. Regarding the range of the degradation, for aflatoxin AFG2 was from 53.9% to 42.1%, whereas the AFB2 remained quite stable, after 20 min treatment and only a degradation from 29.6% to 17.4% was detected. The temperature at which the largest degradation appeared for both aflatoxins AFG2 and AFB2, was the temperature of 308.15 K, the concentration of ozone caused the greatest degradation in both aflatoxins was the concentration of 40 ppm, while the initial concentration caused the largest degradation was that of 0.5 ppb. In fact, during the degradation of aflatoxins AFB2 and AFG2, none of the examined conditions, were able to fully degrade both aflatoxins. In a study carried out by Maeba et al. (Maeba et al., 1988) aflatoxins AFB2 and AFG2 were rather resistant to ozone, requiring 50-60 min to degrade them completely with 34.3 mg/L of ozone.
+
+We consider that the reaction of the decomposition of aflatoxins by ozone in neutral buffer solutions is a first order reaction depending on the concentration of aflatoxins, as the concentration of ozone is in excess and remains stable during the experiments.
+
+By using the kinetic equation for the first order reactions
+
+\[\text{lnc} = \text{lnc}_{0} - k_{\text{app}} t\]
+
+the apparent rate constants (_k_app) for the degradation of aflatoxins in neutral buffer solutions and at all the working temperatures were calculated from the slopes of the plots of Inc versus time t, where \(c\) is the aflatoxin concentration. Such plots are appeared in Figs. 4 and 5. The validity of the Eq. (1) was verified from the experimental data. The R2 values were in the range 0.944-0.999.
+
+The apparent rate constants, (_k_app), for the degradation of aflatoxins AFB1 and AFG1, in the whole working temperature range are quoted in Table 1.
+
+The true rate constants for the degradation of aflatoxins AFB1 and AFG1, were obtained by dividing the apparent rate constants, with the concentration of dissolved ozone in water. In order to estimate the concentrations of dissolved ozone in water, the Henry's constants for pH 7 for all the working temperatures were initially calculated with the aid of Eq. (2) (Rinker et al., 1999):
+
+\[H^{ - 1}\sigma_{3} = \frac{6.91 \times 10^{8}}{\rho_{\text{water}}}\left[ \text{OH}^{ - }\right]^{0.0035}\exp\left( \frac{-2428}{T} \right)\]
+
+Figure 1: Schematic representation of ozone treatment system.
+
+where, \(H^{-1}\)O\({}_{3}\) is the constant of Henry's Law for ozone in water, \(\rho_{\rm water}\) is the density of water at the temperature of the experiment and \(T\) is the temperature. The values of the Henry constants are given in Table 2.
+
+Subsequently, we calculated the concentrations of dissolved ozone in water by using the Eq. (3) (Santer, 1999):
+
+\[C_{\rm aq}=H^{-1}\mathrm{o}_{3}\cdot P\mathrm{o}_{3} \tag{3}\]
+
+where \(P\mathrm{o}_{3}\) is the pressure of ozone. The values of dissolved ozone in water by the application of three ozone concentrations 8.5, 13.5 and 20 ppm in neutral buffer solution of aflatoxins AFB1 and AFG1, for
+
+Fig. 4: Variation of In (ARC1/ppb) with the time in buffer of neutral pH at a temperature of 328.15 K, after treatment with 13.5 ppm O\({}_{2}\).
+
+Fig. 5: Variation of In (ARB1/ppb) with the time in buffer of neutral pH at a temperature of 328.15 K, after treatment with 13.5 ppm O\({}_{2}\).
+
+Fig. 3: Degradation of aflatoxins AFB1 (c) and AFG1 (d) with concentration of 10 ppb after treatment with 13.5 ppm O\({}_{3}\) versus time, at different temperatures.
+
+Fig. 2: Degradation of aflatoxins AFB1 (a) and AFG1 (b) with concentration of 2 ppb after treatment with 13.5 ppm O\({}_{3}\) versus time, at different temperatures.
+
+all the working temperatures are given in Table 3.
+
+Finally, was calculated the true rate constants for the degradation of aflatoxins AFB1 and AFG1, by dividing the apparent rate constants, with the concentration of dissolved ozone in water according to the Eq. (4) (cf. Table 4):
+
+\[k_{\rm true}=\frac{k_{\rm zero}}{C_{\rm ad}} \tag{4}\]
+
+The true rate constants that were observed during decomposition of aflatoxins AFB1 and AFG1, in the presence of ozone, at all temperatures were greater than those for the decomposition of aflatoxins AFB2 and AFG2. As one can see from Table 4, the true rate constants depend on temperature and increase with increasing temperature. Higher temperatures with ozone treatment substantially increased the degradation of aflatoxins AFB1 and AFG1. Although the true rate constants should be increase with the concentration of ozone this actually not happening. This can be attributed to the fact that as the concentration of ozone increases, the concentration of dissolved ozone in water increases, too. Therefore because the true values of the rate constants for the degradation of aflatoxins AFG1 and AFB1, were obtained by dividing the apparent rate constants with the concentration of dissolved ozone in water, these rate constants are normalized and do not increase as ozone concentration increases.
+
+The activation energies for the degradation of aflatoxins were calculated from the Arrhenius equation:
+
+\[\ln k_{\rm true}=\ln A-\frac{E_{\rm a}}{R}\frac{1}{T} \tag{5}\]
+
+where \(A\) is the pre-exponential factor, \(E_{\rm a}\) is the activation energy, \(R\)
+
+\begin{table}
+\begin{tabular}{l l l l l l} \hline Temperature (K) & \(k_{\rm true}\) AFB1 (s−1) & & \(k_{\rm true}\) AFG1 (s−1) \\ \cline{2-6}  & 8.5 ppm O3 & 13.5 ppm O3 & 20 ppm O3 & 8.5 ppm O3 & 13.5 ppm O3 & 20 ppm O3 \\ \hline
+298.15 & 12.10 & 6.22 & 5.61 & 11.00 & 6.56 & 12.62 \\
+308.15 & 13.49 & 8.77 & 10.64 & 16.45 & 9.64 & 20.1 \\
+318.15 & 15.08 & 11.80 & 17.98 & 20.81 & 15.25 & 31.69 \\
+328.15 & 17.05 & 14.78 & 35.86 & 29.83 & 21.5 & 49.81 \\ \hline \end{tabular}
+\end{table}
+Table 4: True rate constants (\(k_{\rm true}\)) for the degradation of aflatoxins AFB1 and AFG1, in neutral buffer solutions, obtained by HPLC analysis, at various temperatures.
+
+\begin{table}
+\begin{tabular}{l l l} \hline Temperature (K) & \(10^{3}\times k_{\rm zero}\) AFB1 (s−1) & \(10^{3}\times k_{\rm zero}\) AFG1 (s−1) \\ \cline{2-4}  & 8.5 ppm O3 & 13.5 ppm O3 & 20 ppm O3 \\ \hline
+298.15 & 1.81 & 1.51 & 2.01 & 1.62 & 1.52 & 4.50 \\
+308.15 & 1.60 & 1.62 & 3.03 & 1.91 & 1.81 & 5.60 \\
+318.15 & 1.42 & 1.83 & 4.02 & 2.01 & 2.31 & 7.11 \\
+328.15 & 1.33 & 1.32 & 6.60 & 2.31 & 2.63 & 9.12 \\ \hline \end{tabular}
+\end{table}
+Table 1: Apparent rate constants (\(k_{\rm zero}\)) for the first order degradation of aflatoxins AFB1 and AFG1, in neutral buffer solutions, obtained by HPLC analysis, at various temperatures.
+
+\begin{table}
+\begin{tabular}{l l l l l l} \hline Temperature (K) & \(10^{4}\times C_{\rm ad}\) (ppm) & & \(k_{\rm true}\) AFG1 (s−1) \\ \cline{2-6}  & 8.5 ppm O3 & 13.5 ppm O3 & 20 ppm O3 & 8.5 ppm O3 & 13.5 ppm O3 & 20 ppm O3 \\ \hline
+298.15 & 12.10 & 6.22 & 5.61 & 11.00 & 6.56 & 12.62 \\
+308.15 & 13.49 & 8.77 & 10.64 & 16.45 & 9.64 & 20.1 \\
+318.15 & 15.08 & 11.80 & 17.98 & 20.81 & 15.25 & 31.69 \\
+328.15 & 17.05 & 14.78 & 35.86 & 29.83 & 21.5 & 49.81 \\ \hline \end{tabular}
+\end{table}
+Table 2: Henry’s constants, H−10S for the decomposition of ozone in neutral buffer solutions at various temperatures.
+
+Figure 6: Arrhenius plots (\(\ln k_{\rm true}\) versus 1/T) for the ozonolysis of aflatoxins AFG1 (a) and AFB1 (b) after treatment with 13.5 ppm O3 on neutral buffer solution.
+
+is the gas constant and \(T\) is the temperature value. Using the calculated values of the true rate constants for each ozone concentration and plotting ln \(k_{\rm true}\) vs 1/\(T\) (cf. Fig. 6), the activation energies at each ozone concentration of the degradation of aflatoxins were determined from the slopes of the graphical representations. The activation energies for aflatoxin AFB1 and AFG1 are given in Table 5.
+
+From Table 5 it can be concluded as the concentration of ozone increases, the activation energies increase too, for both aflatoxins.
+
+## 4 Conclusion
+
+As a general conclusion one could say that the gaseous ozone was highly effective in the degradation mainly of aflatoxins AFB1 and AFG1, despite the degradation of aflatoxins AFB2 and AFG2. Especially, between the aflatoxins AFB1 and AFG1 more sensitive to treatment with ozone proved to be the AFG1, at the higher ozone concentration and the AFB1 at the lower concentrations. The found activation energies show that the degradation of aflatoxins is an activated process depending not only on the nature of the aflatoxins but also on the ozone concentration.
+
+## References
+
+* Ali et al. (2005) Ali, N., Hashim, N. H., Saad, B., Safan, K., Nakajima, M., & Yoshizawa, T. (2005). Evaluation of a method to determine the natural occurrence of aflatoxins in commercial traditional herbal medicines from Malaysia and Indonesia. _Food and Chemical Toxicology_, 43, 1763-1772.
+* Chen et al. (2014) Chen, R., Ma, F., Li, P.-W., Zhang, W., Ding, X.-X., Zhang, Q., et al. (2014). Effect of ozone on aflatoxins detoxification and nutritional quality of peanuts. _Food Chemistry_, 146, 284-288.
+* Chiavaro et al. (2001) Chiavaro, E., Dall'Asta, C., Galaverna, G., Blancardi, A., Gambarelli, E., Dossena, A., et al. (2001). New reversed-phase liquid chromatographic method to detect aflatoxins in food and feed with polydectrins as fluorescence enhancers added to the eluent. _Journal of Chromatography A_, 397, 31-40.
+* Gray et al. (2007) Gray, B., Girgin, G., Engin, A. R., Aydin, S., & Sahin, G. (2007). Aflatoxin levels in wheat samples consumed in some regions of Turkey. _Food Control_, 18, 23-29.
+* Hussain & Anwar (2008) Hussain, L., & Anwar, J. (2008). A study on contamination of aflatoxin M\({}_{1}\) in raw milk in the Punjab province of Pakistan. _Food Control_, 19, 393-395.
+* IARC (1993) IARC (International Agency for Research on Cancer). (1993). IARC monographs on the evaluation of carcinogenic risks to humans. In _Some naturally occurring substances: Food items and constituents_, _heterocyclic aromatic amines and myoclonaus_ (vol. 56, pp. 1-609). 1990. France.
+* Inan et al. (2007) Inan, F., Pala, M., & Doymaz, I. (2007). Use of ozone in detoxification of aflatoxin B\({}_{1}\) in red pepper. _Journal of Standard Products Research_, 43, 425-429.
+* Jolly et al. (2006) Jolly, P., Jiang, Y., Ellis, W., Awuah, R., Nueda, O., Phillips, T., et al. (2006). Determinants of aflatoxin levels in chanalians: sociodemographic factors, knowledge of aflatoxin and food handling and consumption practices. _International Journal of Hygiene Environmental Health_, 2006, 345-358.
+* Karaca (2010) Karaca, H. (2010). Use of ozone in the citrus industry. _Ozone: Science and Engineering_, 232, 122-129.
+* Kurtzman et al. (1987) Kurtzman, C. P., Horn, B. W., & Hesseltime, C. W. (1987). _Aspergillus nomius_, a new aflatoxin-producing species related to _Apergillus flavus and Aspergillus paracitis_. _Anton wur Leuveninechnol_, 53, 147-158.
+* Luo et al. (2014) Luo, X., Wang, R., Wang, L., Li, Y., Bian, Y., & Chen, Z. (2014). Effect of ozone treatment on aflatoxin B\({}_{1}\) and safety evaluation of ozonized corn. _Food Control_, 37, 71-176.
+* Maeha et al. (1988) Maeha, H., Takamoto, Y., Kamimura, M., & Miura, T. (1988). Destruction and detoxification of aflatoxins with ozone. _Journal of Food Science_, 53, 667-668.
+* McKenzie et al. (1997) McKenzie, K. S., Sarr, A. B., Mayara, K., Bailey, R. H., Miller, D. R., Rogers, T. D., et al. (1997). Oxidative degradation and detoxification of mycotoxins using a novel source of ozone. _Food and Chemical Toxicology_, 35, 807-820.
+* Nair & Jolley (2002) Nair, M. S., & Jolley, M. E. (2002). Development of a fluorescence polarization assay for the determination of aflatoxins in grains. _Journal of Agriculture and Food Chemistry_, 50, 3116-3121.
+* Park (2002) Park, D. L. (2002). Effect of processing on aflatoxin. _Advances in Experimental Medicine and Biology_, 504, 173-179.
+* Pierti et al. (2010) Pierti, A., Bertazzi, T., Agosti, R., & Donadini, C. (2010). Transfer of aflatoxin B\({}_{1}\) and fumomish B from naturally contaminated raw materials to beer during an industrial brewing process. _Food Additives & Contaminants_. Part A. _27_, 1431-1439.
+* Reddy et al. (2009) Reddy, K. R. N., Abbas, H. K., Abel, C. A., Shier, W. T., Oliveira, C. A. F., & Raghavender, C. R. (2009). Myoclon contamination of commercially important agricultural commodities. _Toxic Reviews_, 28, 154-168.
+* Reiter et al. (2009) Reiter, E., Zentek, J., & Razzazi, E. (2009). Review on sample preparation strategies and methods used for the analysis of aflatoxins in food and feed. _Molecular Nutrition A Food Research_, 53, 508-524.
+* Rinker et al. (1999) Rinker, E. B., Ashour, S. S., Johnson, M. C., Kott, G. J., Rinker, R. G., & Sandall, O. C. (1999). Kinetics of the aqueous-phase reaction between ozone and 2,4,6-tri-chiorophenol. _AIChE Journal_, 15, 1802-1807.
+* Santer (1999) Santer, R. (1999). Compliant of Henry's law constants for inorganic and organic species of potential importance in environmental chemistry (Version 3). Sander, R. _1999_[http://www.helmworks.com](http://www.helmworks.com).
+* Sarigiannis et al. (2014) Sarigiannis, Y., Kapolos, J., Kolaidima, A., Tsegenitidis, T., & Karaiskakis, G. (2014). Oktracutan A levels in Greek retail wines. _Food Control_, 42, 139-143.
+* USA (1997) USA. (1997). U.S. Food and drug Administration. Substances generally recognized as safe, proposed rule. _Federal Register_, 62, 18937-18964.
+* Zinedine & Minaes (2009) Zinedine, A., & Minaes, J. (2009). Occurrence and legislation of mycotoxins in food and feed from Morocco. _Food Control_. 20(4), 334-344.
+* Zorlugene et al. (2008) Zorlugene, B., Zorlugene, F. K., Oztekin, S., & Evlra, I. B. (2008). The influence of gaseous ozone and ozonated water on microbial flora and degradation of aflatoxin B\({}_{1}\) in dried figs. _Food and Chemical Toxicology_, 46, 3593-3597.
+
+\begin{table}
+\begin{tabular}{l c c c} \hline \hline \multicolumn{1}{c}{
+\begin{tabular}{c} Aflatoxin \\ \end{tabular} } & \(E_{\rm a}\) (k|/mol) \\  & 8.5 ppm O\({}_{3}\) & 13.5 ppm O\({}_{3}\) & 20 ppm O\({}_{3}\) \\ \hline AFB1 & 9.3 \(\pm\) 0.3 & 23.6 \(\pm\) 1.1 & 49.4 \(\pm\) 2.3 \\ AFG1 & 26.2 \(\pm\) 1.6 & 32.6 \(\pm\) 1.1 & 37.2 \(\pm\) 0.5 \\ \hline \hline \end{tabular}
+\end{table}
+Table 5: Values of activation energy, \(E_{\rm a}\), for the degradation of aflatoxins AFB1 and AFG1, after treatment with various ozone concentrations, in neutral buffer solutions.

manual_annotation/200pdfs_in_mmd/10.1016_j.jenvman.2018.12.068.mmd

@@ -0,0 +1,233 @@
+Simulated solar driven photolytic ozonation for the oxidation of aqueous recalcitrant-to-ozone tritosulfuron. Transformation products and toxicity
+
+Rafael R. Solis
+
+Departamento de Ingenieria Química y Química Fisica, Universidad de Extremadura, Avda. Elvas s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain.
+
+Olga Gimeno
+
+Departamento de Ingenieria Química y Química Fisica, Universidad de Extremadura, Avda. Elvas s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain.
+
+F. Javier Rivas
+
+Departamento de Ingenieria Química y Química Fisica, Universidad de Extremadura, Avda. Elvas s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain.
+
+Fernando J. Beltran
+
+Departamento de Ingenieria Química y Química Fisica, Universidad de Extremadura, Avda. Elvas s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain. Instituto Universitario del Agua, Cambio Climítico y Santamibilidad (IACYS), Universidad de Extremadura, Avda. de la Investigacion s/n, 06006, Badajoz, Spain.
+
+###### Abstract
+
+This work reports the combination of ozone and solar radiation as an advanced oxidation process to remove the herbicide tritosulfuron (TSF) in water. Firstly, the recalcitrance of TSF has been assessed, obtaining an ozonation second order rate constant of 5-154 M\({}^{-1}\) min\({}^{-1}\) in the range of pH from 5 to 8; while the rate constant with HO- was found to be (1.8-3.1)10\({}^{\circ}\)M\({}^{-1}\) s\({}^{-1}\). Secondly, the simultaneous application of simulated solar radiation in between 300 and 800 nm and ozone resulted positive in the oxidation rate of TSF. Mineralization extent was also higher. Less effective oxidation was achieved after limiting the radiation to the range 360-800 nm or 390-800 nm; also completely inappropriate for mineralization. Thirdly, the detected transformation products (TPs) demonstrated the vulnerability of TSF molecule to be attacked by HO- in the sulfurphone bridge. The combination of ozone and radiation of 300-800 nm led to the most effective removal of the TPs. Finally, after the photolytic ozonation treatment toxicity was also evaluated in terms of phototoxicity towards the germination and root elongation of _Lactuca Sativa_ seeds, and toxicity by immobilization tests of _Daphnia Magna_.
+
+## 1 Introduction
+
+The development and contribution of pesticides (named as 'Green Revolution') have improved the products quantity and quality of the food industry. Pesticides with artificial origin have been used systematically after the 1950s decade due to a more demanding and skilled agriculture market. The banned used of the first developed pesticides, highly toxic and persistent and, hence bio-accumulating in the environment, has promoted stricter legislation and the application of other modern pesticides formulations less aggressive to the environment. Nevertheless, pesticides and their introduction into the environment due to agriculture is considered an important source of organic contaminants of emerging concern (Diamond et al., 2011).
+
+To date, sulfonylure herbicides have been developed and commercialized worldwide in over 89 countries. Their use is common in all major agronomic crops, land/pasture, forestry and vegetation management (Kramer et al., 2008). In particular, tritosulfuron (TSF) is a broad-spectrum post-emergence herbicide developed and commercialized by BASF Company in 2004 under the trade name 'Biathlon" as a water-soluble formulation. TSF acts mainly through the treated leaves, and not via the soil.
+
+Advanced Oxidation Processes (AOPs) have gained attention in the research community due to their powerful ability to remove organic aqueous pollutants. Intensive scientific exploration of AOPs for water treatment has been conducted. Efficient hydroxyl radical production has been researched, combining UV radiation, hydrogen peroxide, homogeneous catalysis (like Fenton's reagent), ozone, or photocatalysis, among others (Serpone et al., 2017). Regarding ozone-based AOPs, much study has been focused on improving pollutant or/and mineralization rates by using UVC ozonation (Kuo, 1999), ozone combined with hydrogen peroxide (peroxone) (Katsoyiannis et al., 2011), homogeneous or heterogeneous catalytic ozonation (Kasprzyk-Hordern et al., 2003), or the recently explored photocatalytic ozonation (Mehrijouei et al., 2015; Xiao et al., 2015; Beltran and Rey, 2017).
+
+Photocatalytic ozonation emerges as a better performance technology if the results in terms of organic removal, and overall mineralization extent, are analyzed (Xiao et al., 2015). Nevertheless, implementation of heterogeneous catalysts involves the recovering or immobilization of the catalytic solid, and the replace due to inactivation. Additionally, the oxidation by the combination of ozone and UV radiation also covers the oxidation of a wide range of pollutants of diverse reactivity (Otturan and Aaron, 2014). However, the use of UV radiation requires large amounts of electrical energy, raising the cost of deprivation with regard to a real application (Miklos et al., 2010). For that reason, this technology has been poorly implemented in real applications, different for drinking water disinfection (Parson, 2004;Menuer et al., 2006). Nevertheless, UV-based technologies represent a powerful tool to remove organic pollutants, microorganisms and antibiotic resistance genes (Sousa et al., 2017).
+
+So far, few works have focused on the combination of ozone and alternative to UVC (254 nm) radiation. Recently, Somathilake et al. (2018) found that UVA photo-assisted ozonation was appropriated for the mineralization of aqueous carbamazepine. In fact, the synergism between ozone and UVA radiation has also been explored with other organics, successfully oxidized and partially mineralized, such as aniline and 4-chlorophenol (Sauelda and Brillas, 2001).
+
+Some works have considered photolytic ozonation with solar or visible radiation for comparison purposes with other more complex technologies, without exploring in detail the photolytic ozonation system. For example, Rey et al. (2016) achieved comparable effectiveness of photolytic and photocatalytic ozonation of metoprolol. Similar conclusions can be extracted from the photolytic ozonation of DET with solar or visible radiation (Mena et al., 2017). Other works also support the ability to combine ozone and solar radiation, which might avoid the use of extra hydroxyl radical promotion via the addition of photocatalysts (Marquez et al., 2014; Quijnones et al., 2015). Only few recent research have tested the ozone decomposition rate in the presence of solar simulated radiation, and its application to a mixture of pollutants in simulated effluent of wastewater treatment plant (Chavez et al., 2016) or river water (Solis et al., 2019).
+
+The aims of this work have been: a) the study of photolytic ozonation with simulated solar radiation for the oxidation of the ozone recalcitrant TSF at different wavelength ranges; b) the TSF ozone and hydroxyl radical kinetics; c) the tentative identification of transformation products; and, d) the toxicity evaluation of raw and treated TSF samples with phytotoxicity tests of germination-root length elongation using _Lacuca Sativa_ seeds, and immobilization assays of _Daphnia Magna_.
+
+## Experimental section
+
+### Chemicals
+
+Tritosulfuron (TSF, C13H2F8N3O4S, CAS: 142469-14-5) was analytical standard grade (> 99%) and acquired from Sigma-Aldrich" (Germany). Chemicals used for analytical purposes were analytical grade and purchased from Panacea" (Spain). All test and store solutions were prepared with Milli-Q" ultrapure water coming from an Integral 5 system (18.2 MO cm). HPLC-grade acetonitrile (Panreac", Spain) was used for TSF analytical HPLC analysis and MS-grade acetonitrile for qualitative LC-MS-QTOF identification of transformation products.
+
+### Experimental installation and procedure
+
+Solar photolytic ozonation assays were developed in a Suntest CPS + simulator (1500W, air-cooled Xe arc lamp) in which a 500 mL borosilicate glass spherical reactor was placed, homogeneously maintained under magnetic stirring. The emitted simulated solar radiation was restricted to different ranges by using filters named as Daylight (300-800 nm), Storelight (360-800 nm) and Visiblelight (390-800 nm). Ozone was generated in an Anseros COM-AD-01 device and gaseous ozone concentration was monitored in an Anseros-GM apparatus. Fig. 1 shows a scheme of the experimental setup and Fig. S1 depicts the absorption spectra of the herbicide tritosulfuron and the emission spectrum of the simulated solar radiation with the different filters used.
+
+Semi-continuous experiments were carried by feeding ozone and radiation at the same time. The experiments that required darkness were conducted by covering the reactor with aluminum foil in order to maintain similar temperature profiles for comparison purposes. Before starting, the reactor was filled in with 500 mL of TSF solution, dissolved in ultrapure water. At different times samples were extracted from the aqueous solution for analysis. In ozone processes, an inert gas, e.g. nitrogen, was bubbled through samples to remove residual dissolved ozone. The removal of accumulated dissolved ozone was carried out in order to quench the reaction with molecular ozone in all samples, with the exception of those taken for dissolved ozone quantification.
+
+### Aqueous analyses
+
+Aqueous concentration of tritosulfuron was determined by Liquid Chromatography in an HPLC with Diode-Array detection. The apparatus used was a UPLC Shimadzu Prominence LC-AD. A mixture of acetonitrile (A) and water acidified with 0.1% of H2PO4 (B) was pumped at a flow rate of 0.5 mL min-1. Core-shell C18 Kinetex" (150 x 4.6 mm, 5 mm) was used as stationary phase (thermally maintained at 30 degC) and a 40:60 (v/v A:B) as the mobile phase. Quantification was conducted at 227 nm.
+
+Total Organic Carbon (TOC) and Inorganic Carbon (IC) were determined in a Shimadzu TOC-VCSH analyzer equipped with automatic sample injection.
+
+Imorganic and short-chain organic acids were determined by Ion Chromatography (IC) coupled to a conductivity detector. A Methrom" 881 Compact IC pro equipped with chemical suppression, 863 Compact autosampler and anionic-exchange column (MetroSep A sup 5, 250 x 4.0 mm, particles of 5 mm) thermally maintained at 45 degC was used. The used mobile phase program consisted of a 0.7 mL min-1 gradient of Na2CO3 aqueous solution from 0.6 mM to 14.6 mM in 50 min.
+
+Dissolved ozone concentration in aqueous solution was analyzed by the indigo method (Bader and Hoigne, 1981).
+
+The generated hydrogen peroxide was quantified by the colorimetric method based on the cobalt oxidation and complexation with bicarbonate (Masschelein et al., 1977), valid for H2O2 concentrations lower than 50 mM. In UVC photolytic decomposition of H2O2 used for the hydroxyl radicals' rate constant (k100*, TSF) determination, the H2O2 concentration was spectrophotometrically determined with the titanium (IV) oxysulfate reagent (Eisenberg, 1943). A basic 20 Crison" pH-meter equipped with a 50 11T was used for pH measurement.
+
+### Transformation products identification
+
+Transformation products of TSF oxidation via ozonation and photolytic ozonation were analyzed and monitored by HPLC coupled to a Quadrupole Time of Flight (HPLC-QTOF). Experiments with 10 mg L-1 of tritosulfuron were carried out for the identification of the main intermediates. In each analysis, 5 mL of aqueous sample were injected in an Agilent 1260 HPLC coupled to an Agilent 6520 Accurate Mass QTOF LC/MS. A Zorbax Eclipse Plus C18 column (3.5 mm, 4.6 x 100 mm) was used for the chromatographic separation at 30 degC. A mixture of pure MilliQ" water (phase A) and acetonitrile (phase B) was pumped at a flow rate of 0.4 mL min-1 with the following gradient: A:B with a 90:10 ratio was kept during 2 min and changed to 10:90 in 23 min, keeping thereafter 2 min for equilibration. The QTOF conditions were as follows: ESI(-) mode, gas temperature 325 degC, drying gas 10 mL min-1, nebulization 45 psig, Vcap 3500 V, fragmentation 100 V, acquisition m/z range 100-1000. MS spectra were processed with Agilent Mass Hunter Qualitative Analysis B.04.00 software assistance. Suspected and potential candidates list based on computational (_in silico_) prediction tools, such as University of Minnesota Pathway Prediction System and PathPred (Bletsou et al., 2015) were also used.
+
+### Toxicity to Daphnia Magna & phytotoxicity to Lacuca Sativa assays
+
+Immobilization to _Daphnia Magna_ assays was considered for the toxicity analysis before and after the oxidation treatments by using the commercial test kit DAPHTOXKIT F" (MicroBio Tests Inc., Belgium). The procedure followed for eggs hatch and feeding protocol was in accordance to the OECD guidelines for acute immobilization tests (OECD, 2004). The pH of the extracted samples was adjusted to 7 +- 0.1 before analysis. The immobility of the _D. Magna_ neonates at 24 h was registered.
+
+Phytotoxicity assays based on the seed germination-root elongation of _La Lactate Sattu_ were also used to assess the acute toxicity through the reaction time of ozonation and photolytic ozonation treatments. Briefly, fifteen seeds of _L. Sattu_ (_Battuia blonde of Paris_ lettuce from Vilmorin', France) were placed in a Petri dish equipped with paper disc and 4 mL of aqueous sample were transferred, moistening the paper disc. Then, Petri dishes were incubated in a germination chamber isolated from light at 22 degC. After 5 days, the root length of each germinated seed (L) was measured. Additionally, a blank control with ultrapure water was done (Lo) in order to calculate the percentage root growth of the tested samples (L/Lo).
+
+## Results and discussion
+
+### Reactivity and kinetics of triosulfuron in aqueous ozone systems
+
+The reactivity of TSF with ozone was evaluated by determining the second order rate constant of this reaction, k03.TSF, considering that the slow-kinetic regime develops (Beltran, 2004). Fig. 2 depicts the changes of the normalized concentration of triosulfuron and the monitored dissolved ozone concentration profile with time in presence of 5 mM of tert-butanol (t-BuOH), as HO- quencher. For the determination of k03.TSF values, the media of dissolved ozone values after 90 min of ozonation was considered for the calculation, as explained in supplementary material.
+
+Following the procedure previously described (see also Supplementary information) k03.TSF values were obtained and the corresponding Hatta number for slow kinetic regime checked (see Table 1). As it is observed, the k03.TSF calculated values vary from 5 to 150 M-1 min-1 as pH increases from 5 to 8. The dissociation of the molecule makes the anion species 30 times more reactive than the neutral one.
+
+The calculated second-order rate constant was used to extrapolate and calculate the values of the rate constant with the protonated and deprotonated species according to the following equation (Bemner et al., 2008):
+
+\[k_{{\text{O}}_{3},{\text{TSF}}} = \alpha k_{{\text{O}}_{4},{\text{d}}{\text{p}}{\text{p}}{\text{p}}} + (1 - \alpha)k_{{\text{O}}_{5},{\text{p}}{\text{p}}{\text{p}}}\]
+
+where a, the dissociation grade, is defined as:
+
+\begin{table}
+\begin{tabular}{c c c c} pH & k03.TSF \(\pm\) error (M−1 min−1) & R2 & Ha 10−1 \\
+5.01 & 5.5 ± 0.4 & 0.991 & 2.08 \\
+5.50 & 4.7 ± 0.3 & 0.993 & 1.94 \\
+6.01 & 11 ± 1 & 0.971 & 2.96 \\
+6.52 & 35 ± 2 & 0.993 & 5.28 \\
+6.80 & 54 ± 5 & 0.996 & 7.07 \\
+7.00 & 118 ± 12 & 0.992 & 10.8 \\
+8.03 & 154 ± 11 & 0.990 & 11.1 \\ \end{tabular}
+\end{table}
+Table 1: Second-order rate constant of TSF-ozone reaction at different pHs and corresponding Hatta number (Ha).
+
+Fig. 1: Experimental set-up scheme. 1: Oxygen tank; 2: Ozone Generator; 3: gas-phase ozone analyzer; 4: flowmeter; 5: simulated solar radiation apparatus; 6: borosilicate glass reactor; 7: magnetic stirrer; 8: sampling.
+
+\[\alpha=\frac{1}{1+\frac{C_{H^{\prime}}}{K_{a}}}=\frac{1}{1-10^{H_{\text{K}}-pH}} \tag{2}\]
+
+By using a non-linear least squares regression analysis, experimental data of \(\text{k}_{\text{OX,TSF}}\) were fitted to equation [1] to obtain the rate constants of the reactions of ozone with the protonated and deprotonated forms of TSF and the pKa of TSF equilibrium in water. The pKa of the sulfonamide group by itself is 10.1 in solution; however, it is shifted to between 5 and 6.5 depending on the substituents (Kamp et al., 2003). It has to be noted that the pKa reported in the bibliography, 4.69 (PPDB, 2018) does not match the one determined here, 6.93. This sort of discrepancy has already been observed in previous works (Von Sonnig and von Gunten, 2012) and it was reported as reactive pKa. According to these authors, such difference is highly related to the reactivity of the functional groups. Thus, the sigmoidal line in Fig. 3 depicts the modelled \(\text{k}_{\text{OX,TSF}}\) with pH. The modelled equation [1] gave negligible value for the second-order rate constant of the ozone-TSF protonated form, and 171 M\({}^{-1}\) min\({}^{-1}\) for the ozone-TSF dissociated form. The low reactivity towards ozone may be due to the presence of the s-triazine aromatic ring as reported in the literature for similar compounds (Acero et al., 2000; Alvarez et al., 2016) and the deprotonation of the molecule plays an important role during ozone attack.
+
+The second order rate constant of TSF-hydroxyl radical reaction was determined with 254 nm photolysis in the presence H2O2 in excess, (see Supplementary information). TSF abatement in this system for the pHs studied are shown in Fig. S3. Least squares analysis of experimental data to fit equation [88] gave the values of the second-order rate constant in the range of pH = 4-9 presented in Table 2. As can be appreciated, typical rate constant values (average: 2.5\(\cdot\)10\({}^{9}\) M\({}^{-1}\) s\({}^{-1}\)) were obtained. At higher pH, lower values were obtained probably due to the dissociation equilibrium of hydroxyl radical, pKa = 11.54 (Poskrebyshev et al., 2002).
+
+### Solar assisted photolytic aconation
+
+#### 3.2.1 TSF oxidation
+
+Experiments in the presence of solar radiation were carried out in order to elucidate the effect produced with the simultaneous ozone application. The normalized ozone abatement of TSF and the corresponding mineralization (TOC evolution) with time are depicted in Fig. 4. For comparison purposes, photolysis, including all the emitting radiation, and single aconation were experienced. As can be observed, TSF is not photolyzed with Daylight (300-800 nm) radiation. The lack of overlap between TSF radiation absorption spectrum and that of Daylight emission spectrum explains this result (see Fig. S1). On the other hand, although this herbicide is highly recalcitrant to ozone, the 90% TSF removal achieved after 2 h of single oxonation can be explained by the action of hydroxyl radicals. Taking equation [S2] in mind, the experimental conditions followed by ozonation experiment, and the calculated value of \(\text{k}_{\text{OX,TSF}}\), it is possible to simulate the contribution of direct ozonation, by numerically solving the differential equation [S2], i.e. Euler's method (grey line of Fig. 4). From the simulated direct ozonation, only a 3% of TSF oxidation could be achieved in 2 h.
+
+Alternatively, the importance of the hydroxyl radical role can also be examined by calculating the percentage of contribution of the radical pathway (\(\eta_{\text{hyd}}\)) in the process:
+
+\[\eta_{\text{hyd}}=-\frac{1}{\text{k}_{\text{OX,TSF}}\text{C}_{\text{obs}}} \tag{3}\]
+
+where k\({}^{\prime}\)Obs is the apparent pseudo-first order rate constant of TSF oxidation observed in the ozonation process, and \(\text{C}_{\text{0,s}}\) the dissolved ozone concentration. Thus, taking into account the time dissolved ozone concentration profile during the reaction, available in Fig. 5, and given the fact that \(\text{k}_{\text{OX}}\) = 1.15 +- 0.04 h\({}^{-1}\), more than 99% contribution of hydroxyl radical oxidation is deduced during the ozonation
+
+\begin{table}
+\begin{tabular}{c c c} pH & (\(\text{k}_{\text{OX,TSF}}\) \(\leq\) error) 10\({}^{-6}\) (M\({}^{-1}\)s\({}^{-1}\)) & R\({}^{2}\) \\
+4.01 & 2.82 ± 0.14 & 0.998 \\
+5.09 & 3.14 ± 0.09 & 0.999 \\
+6.05 & 2.43 ± 0.11 & 0.999 \\
+7.03 & 2.45 ± 0.10 & 0.999 \\
+8.13 & 1.76 ± 0.09 & 0.998 \\
+9.10 & 1.27 ± 0.12 & 0.996 \\ \end{tabular}
+\end{table}
+Table 2: Second-order rate constant of TSF-hydroxyl radical reaction at different pHs.
+
+Figure 3: pH dependence of the second-order rate constant of TSF-ozone reaction. Comparison of experimental and calculated values from equation [1] (line).
+
+process. Therefore, the assistance of ozone decomposition into hydroxyl radicals via the application of solar radiation seems to be an effective strategy to accelerate the oxidation kinetics of TSF.
+
+The simultaneous application of solar radiation and ozone led to a complete abatement of TSF in less than 60 min of treatment. If the pseudo-first order rate constant of the process is calculated (_k_'obs), just like a mere tool for comparison purpose, the following values are attained: 4.0 +- 0.1 (R2 = 0.99), 3.7 +- 0.1 (R2 = 0.98), 3.8 +- 0.4 h-1 (R2 = 0.99) for photolytic ozonation using Daylorlight (300-800 nm), Storelight (360-800 nm) and Visibellight (390-800 nm), respectively. This means that the presence of radiation with all the UV portion spectrum leads to k' values almost 3.5 higher than in the absence of radiation. Differences between Storelight and Visibellight were minimal, with approximately 2.8 folded values to single ozonation.
+
+#### 3.2.2 Disolved ozone and hydrogen peroxide evolution
+
+The analysis of dissolved ozone concentration profiles with time is a useful tool to elucidate the enhanced formation of radicals in the simultaneous presence of radiation and ozone. Fig. 5 (left) shows the evolution of dissolved ozone concentration with time in the experiments carried out in the presence and absence of radiation. As can be inferred from Fig. 5 (left), when only ozone is applied a stable plateau of 2.1 mg L-1 is reached in less than 10 min, under the experimental conditions applied. The addition of Daylorlight solar radiation has a decomposition effect, decreasing the value of dissolved ozone concentration to around 0.6 mg L-1. If the less energetic radiation Storelight or Visiblelight is applied, the ozone decomposition is not as remarkable as what was observed with Daylorlight, reducing the concentration to 1.5 mg L-1. The fact that ozone presents a little absorption of light in the range 300-320 nm (Chavez et al., 2016; Oh et al., 2016) explains the differences appreciated in the lowest dissolved ozone value reached during the photolytic ozonation with the Daylorlight filter. The other two filters showed poorer ozone decomposition into radicals due to the almost negligible absorption of ozone.
+
+Hydrogen peroxide is a well-known intermediate of aqueous ozonation processes (Stineheim et al., 1984). Fig. 5 (right) depicts the evolution of the generated hydrogen peroxide concentration through the course of the reaction time. Some interesting results can be appreciated when analyzing this figure. First, the H2O2 concentration-time profile observed during the ozone application reaches a stable value in the range of 14-28 uM. If radiation is simultaneously added, higher and faster production of H2O2 is appreciated, with the exception of visible-range radiation. In particular, the combination of ozone and solar radiation in the range 300-800 nm leads to the highest and faster H2O2 formation, reaching a maximum concentration of 52 uM in 30 min, followed by almost total H2O2 decomposition in 120 min. In the presence of Storelight, H2O2 concentration reached a slightly lower maximum (50 uM) followed by a lesser decrease if compared to Daylorlight. Regarding Visibellight ozonation, a similar profile to that observed during ozonation is registered. In this latter case, a slight decrease in H2O2 concentration is also observed after 30 min of treatment. These results can be explained by analyzing the interaction of radiation and ozone or H2O2.
+
+Firstly, the higher UV range leads to an enhanced ozone decomposition to form more H2O2. Secondly, with higher UV range (especially from l \(\sim 360\) nm), direct photolysis of H2O2 also increases (Chu and Anastanio, 2005; Jacobi et al., 2004) and hence H2O2 decomposition.
+
+#### 3.2.3 Mineralization & final oxidation products
+
+In terms of mineralization, differences between Daylight and Storelight or Visiblelight were much higher (see Fig. 4 bottom). The combination of ozone and solar radiation with all the spectra lead to 80% mineralization in 2 h. Single ozonation or visible assisted photozonation achieved a poor 20%. The application of Storelight and ozone slightly improved the results of ozonation, achieving 30% of TOC removal.
+
+An analysis of the final oxidation products, i.e. inorganic and organic anions, gives information about the nature of the TOC remaining in these processes. In this sense, Fig. 6 shows the evolution of the released inorganic and organic anions during the application of ozone and photolytic ozonation with the different filters. Due to the presence of F, N and S in the TSF molecule, fluoride, nitrate, and sulfate were detected during the process. The most efficient technology was the application of Daylight radiation and ozone. In this system, a fast fluoride released is appreciated according to the oxidation of the parent TSF molecule in the first fifteen minutes; reaching a final 61% in 120 min. In the case of nitrate, all ozone involving systems lead to similar inorganic nitrogen release (~20%), with the exception of photolytic ozonation under Daylight radiation, in which nitrate concentration increased in the last final period (36%). No relevant differences were registered in sulfate profiles, with a maximum 75-79% release.
+
+More interesting conclusions can be inferred when comparing the evolution of the formic and oxalic acid concentration profiles. Thus, the application of ozone leads to a low formation of formic and oxalic acids, as expected, due to inefficient mineralization rate of single ozonation. Oxalic acid gradually increased until ~2 mg L-1 whereas formic acid presented a maximum of ~1 mg L-1 at 45 min with a small decrease. This behavior can be explained by taking in mind the reactivity of these
+
+Fig. 5: Dissolved ozone (left) and generated hydrogen peroxide (right) evolution during ozonation and photolytic ozonation of TSF. Experimental conditions as shown in Fig. 4.
+
+acids towards ozone. It is well known that formic acid is several magnitude orders more reactive to direct ozonation than oxalic acid (Hogine and Bader, 1983). When radiation is simultaneously applied with ozone, a higher maximum of formic acid is registered with the exception of visible radiation. Daylight + ozone is the most efficient in terms of formic acid production and its oxidation, being completely removed after 90 min. Finally, ozone and ozone + Visibelight led to similar gradual oxalic acid release whereas Storelight accelerated the gradual generation of this organic acid. However, Daylight + ozone produced the highest oxalic acid production which a maximum in the proximity of ~6 mg L-1 at 45 min, and then, it was almost completely oxidized after 2 h of reaction. In conclusion, the powerful mineralization achieved in photolytic ozonation when applying the UV to the highest extent, that is Daylight (300-800 nm), is intimately related to oxalic and formic acids removal with this system.
+
+#### 3.2.4 Transformation products, proposed mechanism and evolution
+
+Degradation pathway was proposed by analyzing the species involved during TSF oxidation by ozone or ozone combined with radiation (Daylight or Visibelight) by means of LC-QTOF technique. Eight transformation products (TPs) were successfully recognized. Table 3 shows the retention times at which TPs were registered, the formula and structure proposed, the associated error to the experimental mass (in ppm) and the oxidation process in which they were detected. In general, no differences were noticed in the nature of the TPs if ozonation is compared to photolytic ozonation processes. Hydroxyl radical plays, by far, the major role in the TSF oxidation even during single ozonation, the mechanism of oxidation being similar in both technologies. Fig. 7 depicts a proposed mechanism of oxidation routes based on the detected TPs.
+
+Figure 7: Proposed mechanism of TSF oxidation during oxonation and photolytic oxonation processes based on the detected TPs.
+
+2005; Bosch et al., 2001; Ellis and Mabury, 2000). The formation of trifluoracetic acid has also been reported under UVA or solar radiation. The pH and the degree of electron donating-withdrawing of the _ortho_ substituent to -CF3 groups define the yield production of fluoride or trifluoroacetic acid (Ellis and Mabury, 2000).
+
+The evolution of the tentatively detected TPs through the oxidation time was studied by registering the peak area of the Extracted Ion Chromatograms (EC). Fig. S4 depicts the EICs peak area versus time for ozonation, Visiblelight-photolytic _ozonation_, and Daylight-photolytic _ozonation. Total Ion Chromatograms (TICs) at 30 and 60 min are also presented in Fig. S4. Peak areas for all TPs are not proportional to the concentration in the same extent, i.e. intensity of each detected mass strongly depends on operating conditions and easiness to be ionized. Nevertheless, EICs peak areas can be assumed as a tool for comparison of TPs evolution during the oxidation technology. TP1, TPS, and TP8 appear in higher intensities; and they could be considered as the major intermediates. The profile of the rest appears in one less magnitude of intensity.
+
+TP1 reached maximum formation after 5-10 min of oxidation, its intensity is higher according to the following order: Daylight photolytic _ozonation_ > Visiblelight _o_zonation > _o_zonation. The release of TPS and TPS was also faster when applied \(\text{O}_{3} + \text{Daylight}\) radiation, reaching their maxima at 30 min to significantly decrease at 60 min. Ozonation and Visible-photolytic _o_zonation accumulated TPS and TPS with reaction time. The higher accumulation of intermediates at 30 min for Daylight + ozone is also observed by comparing the TICs for the three systems (Fig. S4). In 30 min, Daylight + ozone leads to complete removal of the parent compound with more production of TPs that appear as peaks of higher intensity in the TTC.
+
+TP4, TP6, and TP8 are accumulated in the aqueous media for ozonation and Visiblelight-photolytic _o_zonation, whereas Daylight-photolytic _o_zonation is capable of removing them completely after 60 min.
+
+TP2 appears after further oxidation of TP6. Only Daylight combined with ozone is able to degrade TP6 and launch the formation of TP2. Softer oxidation via \(\text{O}_{3}\) or \(\text{O}_{3} + \text{Visiblelight}\) radiation seems to be inefficient for the transformation of TP6 which accumulates and impedes the formation of TP2. Similar trends were observed for TP3, generated after oxidation of TPS. As TPS is oxidized, the formation of TP3 is triggered; this behavior only being observed in Daylight photolytic _o_zonation.
+
+Fig. 8.: Peak areas evolution of the Extracted Ion Chromatograms (EICs) for the identified Transformation Products (TPs) in the ozonation (top-left) and photolytic _o_zonation with Daylight (300–800 nm) (down) or Visiblelight (390–800 nm) (top-right) radiation. Experimental conditions as shown in Fig. 4.
+
+#### 3.2.5 Phytotoxicity to Lacuca Sativa and immobilization to Daphnia Magna
+
+Toxicity bioassays have positively been used as a reliable tool to evaluate whether effluent detoxification takes place (Rizzo, 2011). This especially applies to those cases where partial oxidation of organic compounds is appreciated. The study of immobilization of Daphnia Magna is postulated as the most popular test due to its reproducibility and easy procedure. Moreover, the seed germination of Lacuca Sativa assay as target species has been successfully tested to study the changes in toxicity terms during AOPs processes (Andreozzi et al., 2008). In this work, both tests have been considered for the evaluation of the detoxification of TSF. Fig. 9 depicts the results obtained.
+
+The inhibition growth on other target vegetables has been reported for similar sulfonylurea herbicides (Kotoua-Sykka et al., 1993). TSF shows a low-moderate inhibition on the germination-root elongation of _L. Sativa_. As can be appreciated in Fig. 9 (A & B), the untreated samples containing 10 mg L-1 TSF led to ~50% of inhibition on the lettuce growth. Fig. 9A shows how photolytic (Daylight radiation) oxzonation process is able to completely remove the phytotoxic character. Thus, seeds reach after the treatment (120 min) a 100% growth length. In contrast, the application of only ozone led to 85% growth length after the same time of treatment. Therefore, after oxzonation, a 15% of phytotoxic content remains, likely due to the influence of accumulated intermediates. By comparing oxzonation and photolytic oxzonation, the addition of Daylight radiation does not only remove phytotoxicity to higher extent but also leads to a faster evolution of the percentage growth of _L. Sativa_. When radiation is filtered to have a less UV-containing range, ozone + photolysis is less effective to remove phytotoxicity (Fig. 9B). The order of effectiveness is Daylight > Storelight > Visiblelight.
+
+Ecotoxicity test using the commercial DAPHTOXKIT F magna" was used to evaluate the toxicity of the untreated and treated samples after 120 min (Fig. 9C). The untreated sample presented 80% of immobilization of the _daphuis_ after 24 h of exposition. Other sulfonylurea have been reported to be considerably toxic to _D. Magna_(Zaltauskaite and Brazaityte, 2013) with over 80% mortality of crustaceans. When applying 120 min of oxzonation, less than 50% inhibition is appreciated. If solar radiation is applied in the oxzonation process, the toxicity is reduced to only 25% in the best of the cases. This, once again, proves the effectiveness in terms of detoxification of ozone application with the all UV available range in the solar spectrum (ozone + Daylight).
+
+## 4 Conclusions
+
+From the obtained results the following conclusions were reached:
+
+Application of simulated solar radiation in ozone processes postulates as a promising alternative to enhance the removal rate of oxidation recalcitrant organics, such as TSF. The application of solar radiation (Daylight) and ozone is the most effective process in terms of TSF oxidation. Storelight and visible-photolytic ozonation also improve TSF removal rates if compared to single ozonation. However, in terms of mineralization the role the UV radiation plays results of high importance. Only Daylight + ozone improved the mineralization rate, reaching 80% of TOC oxidation. This behavior is intimately related to the ability of this system to remove refractory organics to ozone final organic acids, like oxalic acid.
+
+From the TPs tentatively identified from LC-QTOF technique, the cleavage of TSF in the sulfonyl urea bridge is proposed. The s-triazine aromatic and the-CF3 groups showed reactance to oxidation, the application of ozone + Daylight being the most effective for degradation as fluoride release confirms.
+
+Finally, from the analysis of toxicity of the treated samples through _L. Sativa_ phytotoxicity analyses and _D. Magna_ immobilization, higher detoxification of Daylight-photolytic ozonation process is observed.
+
+## Acknowledgements
+
+The authors are grateful to Junta de Extremadura (Project IB16022), co-financed by the European Funds for Regional Development, for economically supporting this work. Moreover, it is also acknowledged the '_Servicio de Andilisi Elemental y Molecular (SAEM)_' of '_Servicios de Apoyo a la Investigacion de la Universidad de Extremadura (SAIUex)_' for the helping with the intermediate products analyses.
+
+## Appendix A Supplementary data
+
+Supplementary data to this article can be found online at [https://doi.org/10.1016/j.jenvman.2018.12.068](https://doi.org/10.1016/j.jenvman.2018.12.068).
+
+[MISSING_PAGE_POST]
+
+Bocel, F., Conquerella, M.C., Marin, M.L., Miranda, M.A., 2001. Photochemistry of 2-hydroxy-4-trifluoromethylebenzoic acid, major metabolite of the photosensitizing plastic anti-aggregant drug trithillul. Photochem. Photobiol. 75 (45), 463-468.
+* [10] Chivera, A.M., Rey, A., Beltran, F.J., Alvarez, P.M., 2016. Short photo-oxidation: a novel treatment method for the degradation of water pollutants. J. Hazard. Mater. 317 (46-43).
+* [11] Chu, L., Anastanio, C., 2005. Formation of hydroxyl radical from the photolysis of frozen hydrogen peroxide. J. Phys. Chem. A, 109 (28), 6264-6271.
+* [12] Diamond, J.M., Lalime, H.A., Malmiviruk, K.B., Thornton, K.W., Bartell, S.M., Kidd, K.A., 2011. Prioritizing contaminants of emerging concern for ecological screening assessments. Environ. Toxicol. Chem. 30 (11), 2385-2394.
+* [13] Eisenberg, G.M., 1943. Colorimetric discrimination of hydrogen peroxide. Ind. Eng. Chem. Anal. 81 (53), 237-238.
+* [14] Ellis, D., Malmyruk, S.A., 2000. The aqueous photolysis of TPM and related tri-fluoromethylphenols: An alternate source of trifluoroacetic acid in the environment. Environ. Sci. Technol. 4 (3), 622-637.
+* [15] European Food Safety Authority (HFSA), 2007. Opinion on the toxicological relevance of the soil and groundwater metabolite TRSA of trithulururur in the context of the human risk assessment. ETSA J. 621 (1-33).
+* [16] Hoijne, J., Jaker, H., 1983. Rate content of reactions of ozone with organic and inorganic compounds in water: II: dissociating organic compounds. Water Res. 17 (2), 188-194.
+* [17] Jacobi, H.W., Kwakye-Avuah, B., Schrems, O., 2004. Photochemical decomposition of hydrogen peroxide (H\({}_{2}\)O\({}_{3}\)) and formaldehyde (HCHO) in artificial snow. Ann. Glacid. 39 (29-33).
+* [18] Kamp, K., Kijlullah, N., Corkey, B.E., Berggren, P.O., Hamilton, J.A., 2003. Sulfonylures rapidly cross phospholipid bilayer membranes by a free-diffusion mechanism. Diabetes 52, 2526-2531.
+* [19] Kasprisdorf, H., Zioledi, M., Naweed, J., 2003. Catalytic formation and methods of enhancing molecular ozone reactions in water treatment. Appl. Catal. B. Environ. 46 (4), 639-669.
+* [20] Katsuyama, I.A., Cannonic, S., von Gunten, U., 2011. Efficiency and energy requirements for the transformation of organic micropollutants by ozone, O\({}_{3}\)/H\({}_{2}\)O, and UV HQ,2O Water Res. 45 (13), 3811-3822.
+* [21] Kototakis-Vika, E., Elefthebrationration, I.G., Gaglamas, A.A., Sficas, A.G., 1993. Phytotoxicity and persistence of chlorosulfan, trithulurururally, trithulurururally, trithulururally in three-slit cells. Wear. Res. 33 (5), 355-357.
+* [22] Kowal, S., Halas, P., Waters, F., Schmidt, T.C., 2013. Fully automated standard addition method for the quantification of 29 polar peptide metabolites in different water bodies using ICMS/MS, Anal. Bioanal. Chem. 405 (10), 6337-6351.
+* [23] Kramer, W., Schirmer, U., Jeschke, P., Witschel, M., 2008. Modern Crop Protection Compounds. Methods, vol. 11987-CV Verlag GmbH & Co., KGa., Nov. 1999. Synergistic effects of combination of photolysis and donation on destruction of chlorophenol in water. Chemosphere 39 (11), 1853-1860.
+* [24] Lam, M.W., Young, C.J., Malmyruk, S.A., 2005. Aqueous photochemical reaction kinetics and transformations of fluorosine. Environ. Sci. Technol. 39 (2), 513-522.
+* [25] Miyatose, G., Rodriguez, E.M., Beltran, F.J., Alvarez, P.M., 2014. Solar Photocatalytic formation of a mixture of pharmaceutical compounds in water. Chemosphere 113, 71-78.
+* [26] Muschelcin, W., Denis, M., Ledent, R., 1977. Spectrophotometric determination of residual hydrogen peroxide. Water Sew Works 124 (8), 69-72.
+* [27] Mehloipou, M., Muller, M., Muller, M., 2015. A review on photocatalytic generation used for the treatment of water and wastewater. Chem. Eng. 2, 263 (1), 209-219.
+* [28] Mens, E., Rey, A., Rodriguez, E.M., Beltran, F.J., 2017. Nanostructured CeO2 as catalysts for different X0b based in the application of ozone and simulated solar radiation. Catal. Today 280 (1), 74-79.
+* [29] Mendez-Arriaga, P., Osu, T., Oyama, T., Gimenez, J., Espinosa, S., Hidaka, H., Serpone, N., 2011. Photooxidation of the antidepressant drug fluoroserine (Prozae) in aqueous media by hybrid catalytic/ozonium processes. Water Res. 45 (9), 2782-2794.
+* [30] Munzer, L., Cannonic, S., von Gunten, U., 2006. Implications of sequential use of UV and ozone for drinking water quality. Water Res. 40 (9), 1864-1876.
+* [31] Miklin, J.B., Henry, C., Jelekt, M., Linden, K.G., Drewes, J.E., Hubner, U., 2010. Evaluation of advanced oxidation processes for water and wastewater treatment. A critical review. Water Res. 139, 118-131.
+* [32] OECD, 2004. Daphnia is. Acute immobilisation Test, OECD Guideline for the Testing of Chemicals, Guideline. pp. 202.
+* [33] Oh, J.S., Yajima, H., Hashida, K., Ono, T., Ishihima, T., Serizawa, I., Puruta, H., Hatta, A., 2016. Insitu UV absorption spectroscopy for observing dissolved ozone in water. J. Photopovil. Sci. Technol. 29 (3), 427-432.
+* [34] Ouran, M.A., Aaron, J., 2014. Advanced oxidation processes in water/susteur treatment principles and applications. A review. Grit. Rev. Environ. Sci. Technol. 44 (23), 2577-2641.
+* [35] Parson, S., 2004. Advanced Oxidation Processes for Water and Wastewater Treatment. IVR Publishing, London.
+* [36] Poufersbyche, G.A., Neta, P., Huie, R.E., 2002. Temperature dependence of the acid dissociation constant of the hydroxyl radical. J. Phys. Chem. A 106, 11488-11491.
+* [37]**Pesticide Properties Database (PPDB)**. [http://sitem.herts.ac.uk/aerv/pdb/en/](http://sitem.herts.ac.uk/aerv/pdb/en/), Accessed date: 12 June 2018.
+* [38] Quinones, D.J., Alvarez, P.M., Rey, A., Beltran, F.J., 2015. Removal of emerging contaminants from municipal WWTF secondary effluents by barbotocatalytic on-nation. A pilot-scale study. Sparr. Pcentral. Lett. 149, 132-139.
+* [39] Remstman, T., Alder, L., Bansiak, U., 2013. Emerging pesticide metabolites in groundwater and aurbor water as determined by the application of a multimethod for 150 pesticide metabolites. Water Res. 47 (15), 5355-5365.
+* [40] Rey, A., Quinones, D.J., Alvarez, P.M., Beltran, F.J., Plucinski, P.K., 2016. Simulated solar-light assisted photocatalytic on atomic and molecular organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic neutral organic organic neutral organic neutral organic neutral organic neutral organic organic neutral organic neutral organic neutral organic organic neutral organic neutral organic organic neutral organic neutral organic organic neutral organic neutral organic organic neutral organic neutral organic organic neutral organic neutral organic organic neutral organic organic neutral organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic neutral organic organic organic neutral organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic neutral organic organic organic organic neutral

manual_annotation/200pdfs_in_mmd/10.1016_j.jhazmat.2003.07.001.mmd

@@ -0,0 +1,194 @@
+Ozonation and H\({}_{2}\)O\({}_{2}\)/UV treatment of clofibric acid in water: a kinetic investigation
+
+Roberto Andreozzi
+
+Corresponding author. Tel.: +39-081-7682251; fax: +39-081-5936936. roberto.andreozzi@unina.it
+
+Vincenzo Caprio
+
+Raffaele Marotta
+
+Anita Radovnikovic
+
+Dipartimento di Ingegneria Chimica, Facolta di Ingegneria, Universita degli Studi di Napoli "Federico II", Piazzale V. Tecchio 80, 80125 Naples, Italy
+
+10 March 200310 March 2003233-246
+
+###### Abstract
+
+The presence of pharmaceuticals or their active metabolites in surface and ground waters has been recently reported as mainly due to an incomplete removal of these pollutants in sewage treatment plants (STP). Advanced oxidation processes may represent a suitable tool to reduce environmental release of these species by enhancing the global efficiency of reduction of pharmaceuticals in the municipal sewage plant effluents. The present work aims at assessing the kinetics of abatement from aqueous solutions of clofibric acid (a metabolite of the blood lipid regulator clofibrate) which has been found in surface, ground and drinking waters. Ozonation and hydrogen peroxide photolysis are capable of fast removal of this species in aqueous solution, with an almost complete conversion of the organic chlorine content into chloride ions for the investigated reaction conditions. A validation of assessed kinetics at clofibric acid concentrations as low as those found in STP effluents is presented for both systems.
+
+Clofibric acid; Drugs; Ozonation; Hydrogen peroxide photolysis; AOP processes; Kinetics 10 March 2003233-246
+
+## 1 Introduction
+
+Following the first studies in the 1980s [1; 2; 3], a recent increasing interest in the presence of pharmaceuticals or their active metabolites in the aquatic environment is documented [4; 5; 6; 7; 8; 9; 10]. A fairly high number of drugs belonging to different pharmaceutical classes has been detected in surface and ground waters at concentration ranging from nanograms to micrograms per liter. Previous investigations generally conclude that the main source of these environmental pollutants are the effluents of the sewage treatment plants (STP). In fact,after their use such substances are often excreted unmetabolised, directly into the sewage system as parent compounds. Many of these pharmaceuticals have been reported to be only partially removed in STP and are thus discharged in surface waters. Additional sources are manure, through which veterinary drugs are introduced in the environment [4,11], and incorrect disposal of personal care products and unused pharmaceuticals in domestic refuse. Most scientific papers dealing with this topic outline the need to assess potential risks of pharmaceuticals for human and environmental health since the concentrations at which they are found are orders of magnitude lower than those causing acute toxic effects on aquatic organisms [12]. The impact on sexual differentiation in fish of contraceptive steroid 17a-ethylnoestradiol, which has been demonstrated at concentration of nanograms per liter is reported to support this view [13].
+
+Recently, adverse effects on rainbow trout (_Oncorhynchus mykiss_) exposed to diclofenac, an anti-inflammatory drug, at a concentration of 1 mg l-1 have been documented [14]. Moreover data on invertebrates gave LOEC for carbamazepine at concentrations of about 20 mg l-1 (J. Garric, unpublished data).
+
+Clearly, even in cases in which adverse effects are proved for some pharmaceuticals, their use cannot be abandoned, and some measures devoted to reduce the environmental risk have to be adopted. Therefore, wastewater treatment by means of advanced technologies capable of ensuring an enhanced removal of these species with respect to that achieved in conventional biological processes have thus to be considered. Among existing treatment strategies, ozonation and hydrogen peroxide photolysis have already achieved a high level of development, which makes their adoption at an industrial scale quite flexible. Following this point of view, a recent work studied the removal of carbamazepine through the use of ozonation process [15]. The oxidative degradation in aqueous solution of paracetamol by means of ozonation and H2O2/UV photolysis was also investigated by the authors [16].
+
+The present work evaluates the removal of clofibric acid (Fig. 1), a human metabolite of clofibrate (its ethyl ester form), used as a blood lipid regulator, from aqueous solutions by means of both ozonation and hydrogen peroxide photolysis. The presence of this compound, which has been reported to be highly persistent once introduced in the surface waters [20], has been documented in STP effluents [10,17], rivers [18,19], lakes [20], ground and drinking waters [21]. Previous studies [22,34] report that ozonation effectively removes clofibric acid from drinking water but only scant indications on reaction kinetics were given.
+
+## Materials and methods
+
+Ozonation was tested in a semicontinuous stirred tank Pyrex glass reactor (1.090 l), with constant temperature \(T=298\) K, operated in the batch mode with respect to liquid
+
+Fig. 1: Chemical structure of clofibric acid. CAS number: 882-09-7.
+
+phase. The apparatus used for the studies was similar to that previously described [23]. The initial clofibric acid concentrations were in the range 1.0 x 10-3 to 1.5 x 10-3 M. An ozonized oxygen stream of 2 vol.%, generated by an ozone-generator (Fischer 502) was fed at a flow rate of 361h-1 to the reactor containing the aqueous solution.
+
+The ozone concentration in the outlet gaseous stream (O3, freeboard) was evaluated by continuous UV monitoring at 253 nm (\(\varepsilon_{\text{O}_{3}}=3200\,\text{M}^{-1}\,\text{cm}^{-1}\)) using of a Varian UV spectrophotometer equipped with a quartz cell (optical length = 2.0 x 10-2 dm).
+
+Clofibric acid solutions were buffered at desired pHs by adding of H3PO4, KH2PO4 and Na2HPO4 salts. The ionic strength was adjusted at a constant value of 0.1 M with addition of NaCl salt.
+
+Batch ozonation experiments were performed at low clofibric acid concentrations (5.0 x 10-8 M). For these low concentration experiments a 0.81 aqueous solution at pH = 5.0 was previously saturated with ozone by bubbling an ozonized gaseous stream (ozone concentration in the liquid bulk was 1.0 x 10-5 M).
+
+Once saturation was attained, gaseous feeding was stopped and the substrate charged in the reactor by rapidly injecting 0.8 cm3 of a concentrated clofibric acid aqueous solution (5.0 x 10-5 M). The reaction was quenched at the desired time by sparging the aqueous solution with stream of nitrogen. After quenching, the solution was recovered and 0.31 were concentrated to a final volume of 2.0 ml for analysis.
+
+The UV/H2O2 experiments were carried out at 298 K in a batch cylindrical glass jacketed reactor with an outer diameter of 9.5 cm and a height of 28 cm wrapped with an aluminium foil. At the top, the reactor had inlets for feeding reactants and an outlet for withdrawing samples. The reactor was equipped with a 17 W (power input) low-pressure lamp (by Helios Italquartz) with a monochromatic wavelength emission at 254 nm [24] enclosed in a quartz sleeve, which was immersed in the solution in the centre of the reactor. The radiation power (2.7 x 10-6 E s-1) was measured by means of H2O2 actinometric measurements [25]. The reactor was open to air, and mixed with a magnetic stirrer placed at the bottom.
+
+Clofibric acid solutions were adjusted to the desired pH value with dilute HClO4 and NaOH mixtures. Samples were taken at fixed reaction times and analysed. For photolytic experiments at low concentrations of clofibric acid (5.0 x 10-8 M), the reaction was stopped by switching off the lamp, and the solutions were recovered and concentrated by evaporation for the analyses similar to the ozonation experiments.
+
+The substrate was analysed by Hewlett-Packard HPLC (HP 1100 L) equipped with a diode array detector and a Synergi C12 4u MAX-RP column using a 40:60 buffered aqueous solution: acetonitrile as mobile phase flowing at 1.0 ml min-1. The buffered aqueous solution was prepared with 4 ml H3PO4 (85 wt.%), 50 ml methanol in 11 HPLC water.
+
+An Orion 96-17B combination electrode was used to detect the free chloride produced during the ozonation and UV/H2O2 processes. The total organic carbon (TOC) was monitored by means TOC analyzer (Shimadzu 5000 A).
+
+The pH of the aqueous solutions was determined using an Orion 960 pH meter with a glass pH electrode. All of the reagents except hydrogen peroxide (Fluka, 30 wt.% not stabilized) were purchased from Sigma-Aldrich.
+
+## 3 Results and discussion
+
+Preliminary experiments were carried out to assess the capability of two chosen systems to remove clofibric acid. Fig. 2 reports the results obtained during 1 h of an ozonation experiment with an aqueous solution of clofibric acid at initial concentration of \(1.5\times 10^{-3}\) M. Complete clofibric acid disappearance was observed after 20 min of ozonation with a mineralization degree equal to 34.0%. It is noteworthy to observe that at the same reaction time, the initial chlorine content in the substrate was released as chloride ions, thus indicating that no hazardous chlorinated intermediates were formed. Moreover, prolonged ozonation treatment up to 60 min allowed the achievement of a mineralization degree of 49.1%. The results obtained in a photolytic runs with [H2O2]0 = 1.0 M and [S]0 = 1.0 \(\times\) 10\({}^{-3}\) M are shown in Fig. 3. An almost complete removal of clofibric acid was achieved in 60 min of treatment with a satisfactory efficiency of chlorine release as chloride although the degree of mineralization recorded at this reaction time was poor.
+
+## 4 Reaction kinetics
+
+### Ozonation
+
+In a gas-liquid reactor the oxidation process develops according to different regimes of absorption with reaction. The reactor used in the present investigation was previously characterized by determining the mass transfer coefficient \(k_{\rm L}^{0}\) (\(4.26\times 10^{-3}\) cm s\({}^{-1}\)) and the volumetric mass transfer coefficient (in the absence of chemical reaction) \(k_{\rm L}^{0}a\) (\(0.045\) s\({}^{-1}\)) at the adopted stirrer speed (\(380\) rpm) and ionic strength (\(0.1\) M). It has been demonstrated in previous papers [27, 36] that in this reactor ozonation processes of organic species develop under (slow/fast) kinetic regimes for Hatta number [\([D_{\rm O_{3}}zk_{\rm O_{3}}[{\rm S}]_{0}]^{0.5}(k_{\rm L}^{0})^{-1}\)] \(<2.0\) and under a quasi-diffusional regime for \(Ha\) number values up to \(25\). For higher values a diffusional regime establishes.
+
+A careful choice of clofibric acid starting concentration was thus necessary to perform kinetic experiments. In fact, for low values a complete oxidation of the substrate is achieved in a too short time scales thus hindering the collection of a significant number of experimental samples during a single run. For high values of the starting concentration, the diffusionalregime of absorption with reaction could establish and render all of the collected data useless for the determination of reaction kinetics. The concentrations adopted were around 1.0 x 10-3 M and resulted, for most cases, in the achievement of a kinetic regime of absorption with reaction (slow or fast) [26], sometimes in a quasi-diffusional one [27]. In these conditions, a fluidynamic submodel published elsewhere [27] was coupled with an overall ozonation reaction:
+
+\[{\rm Clofibric\,acid}+z{\rm O}_{3}\stackrel{{ k_{{\rm O}_{3}}}}{{\to}}{\rm products}\]
+
+and used for analysis of the collected data.
+
+The values of the parameters _k_O3 and \(z\) shown in Table 1 were estimated as those which minimize for each run the sum of the squares of the differences between experimental and calculated data. Fig. 4a and b shows various examples of the agreement between experimental data and those calculated by the model (solid lines) when the best values of the parameters _k_O3 and \(z\) are used. It is noteworthy to observe that the percentage standard deviations (Table 1) for the substrate concentration in the liquid bulk and ozone in the freeboard, which give a measure of the model adequacy, are as low as those found in the analytical determination of these species. At pH = 3.0, both of the models obtained by assuming a process development under a "slow" and "fast" kinetic regime of absorption with reaction gave satisfactorily results with very similar values of _s_'s. In this case, the most appropriate regime was not singled out. At pH = 6.5 poor results were found by using a stoichiometric coefficient \(z\) = \(a\) = 2.00 (\(\sigma_{\rm s}\) = 16.6%, \(\sigma_{\rm O_{3}}\) = 14.8%). On the other hand, lower percentage standard deviations were calculated by adopting an overall stoichiometric coefficient as a linear function of the reaction time (_z_ = \(a\) + _bt_) with \(a\) = 2.00. In this case, the best value of the parameter \(b\) was estimated along with _k_O3 by means of the optimization procedure described earlier.
+
+It can be easily verified that for the values of _k_O3 and \(z\) estimated with this model and using an ozone diffusivity (_D_O3) of 1.77 x 10-5 cm2 s-1 [37], Hatta numbers lower than 2.1 are found.
+
+\begin{table}
+\begin{tabular}{c c c c c c c} pH & [S]0 (mM) & _k_O3 (M−1 s−1) & \(z\) & _σ_s (\%) & _s_O3 (\%) & Regime \\
+2.0 & 0.85 & 29.8 ± 1.52 & 2.00 & 3.27 & 3.66 & Kinetic regime “slow” \\
+2.5 & 0.92 & 39.7 ± 2.30 & 2.00 & 2.63 & 3.03 & Kinetic regime “slow” \\
+3.0 & 0.86 & 103.5 ± 5.6 & 2.00 & 4.32 & 3.41 & Kinetic regime “slow” \\
+3.0 & 0.86 & 91.3 ± 4.82 & 2.00 & 4.20 & 3.89 & Kinetic regime “fast” \\
+3.5 & 0.88 & 164.5 ± 22 & 2.00 & 4.66 & 6.75 & Kinetic regime “fast” \\
+4.0 & 0.90 & 155.8 ± 12.2 & 2.00 & 3.26 & 4.04 & Kinetic regime “fast” \\
+4.5 & 1.03 & 311.5 ± 29.7 & 2.00 & 5.13 & 2.44 & Kinetic regime “fast” \\
+5.0 & 1.16 & 842.1 ± 79.2 & 2.00 ± 0.04 & 3.73 & 3.93 & “Quasi-diffusive” regime \\
+5.5 & 0.90 & 1580 ± 105.7 & 2.11 ± 0.03 & 2.34 & 3.10 & “Quasi-diffusive” regime \\
+6.0 & 0.98 & 1041.1 ± 109.3 & 2.02 ± 0.04 & 3.84 & 2.64 & “Quasi-diffusive” regime \\
+6.5 & 0.88 & 2550 ± 251 & \(z\) = \(a\) + _bt_; \(a\) = 2.00; & 5.22 & 6.83 & “Quasi-diffusive” regime \\  & & & \(b\) = 0.28 ± 0.03 & & & \\ \end{tabular}
+\end{table}
+Table 1: Kinetic parameters for ozonation of clofibric acid, calculated for different pH valuesFigure 4: Comparison between experimental (full symbols) and calculated data (solid lines) for ozonation of clofibric acid at different pH: \(T=298\) K; (\(\blacksquare\)) substrate; (\(\blacksquare\)) ozone in the freeboard phase.
+
+Moreover the data in Table 1 indicate that the system reactivity increases with increasing the pH of the solution. This result can be explained by taking into account that the substrate dissociates in aqueous solution:
+
+\[\begin{equation*}{\mathrm{c}}{\mathrm{C}}{\mathrm{H}}_{3}{\mathrm{C}}{\mathrm{O}}{\mathrm{H}}_{5}{\mathrm{C}}{\mathrm{O}}{\mathrm{H}}_{5}{\mathrm{C}}{\mathrm{O}}{\mathrm{H}}_{3}{\mathrm{C}}{\mathrm{O}}{\mathrm{H}}_{5}{\mathrm{C}}{\mathrm{O}}{\mathrm{O}}{\mathrm{H}}_{3}{\mathrm{C}}{\mathrm{O}}{\mathrm{O}}{\mathrm{C}}{\mathrm{O}}{\mathrm\[\frac{{\rm d}[{\rm H}_{2}{\rm O}_{2}]}{{\rm d}t} = -\frac{\phi{\rm H}_{2}{\rm O}_{2}}{V_{\rm sol}}\,I_{0}[1-\exp(2.3l( \varepsilon_{\rm s}[{\rm S}]+\varepsilon_{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{ \rm O}_{2}]))]f_{{\rm H}_{2}{\rm O}_{2}} \tag{2}\] \[+\ -k_{\rm h}[{\rm HO}^{\bullet}][{\rm H}_{2}{\rm O}_{2}]+k_{\rm t }[{\rm HO}_{2}{}^{\bullet}]^{2}\]
+
+where \(\phi_{\rm s}\) and \(\phi_{{\rm H}_{2}{\rm O}_{2}}\) are the primary quantum yields of the direct photolysis at 254 nm of clofibric acid and hydrogen peroxide (\(\phi_{{\rm H}_{2}{\rm O}_{2}}=0.5\) mol E\({}^{-1}\)[30; 31]), \(V_{\rm sol}\) the volume of the aqueous solution (0.421), \(I_{0}\) the measured lamp UV-light intensity at 254 nm (2.7 x 10\({}^{-6}\) E s\({}^{-1}\)), \(l\) the optical pathlength (0.201 dm) of the reactor, \(\varepsilon_{\rm s}\), \(\varepsilon_{{\rm H}_{2}{\rm O}_{2}}\) and \(k_{\rm s}\) are, respectively, the molar extinction coefficients at 254 nm for the substrate (380 M\({}^{-1}\) cm\({}^{-1}\)), hydrogen peroxide (18.6 M\({}^{-1}\) cm\({}^{-1}\)) and reaction products, \(f_{\rm s}\) and \(f_{{\rm H}_{2}{\rm O}_{2}}\) represent the UV fraction absorbed by the substrate and hydrogen peroxide.
+
+The mass balances on HO and HO\({}_{2}\) radical species are
+
+\[\frac{{\rm d}[{\rm HO}^{\bullet}]}{{\rm d}t}=2\phi_{{\rm H}_{2}{\rm O}_{2}}\frac {W_{\rm abs}}{V_{\rm sol}}-k_{\rm h}[{\rm HO}^{\bullet}][{\rm H}_{2}{\rm O}_{2} ]-k_{\rm s}[{\rm HO}^{\bullet}][{\rm S}] \tag{3}\]
+
+\[\frac{{\rm d}[{\rm HO}_{2}{}^{\bullet}]}{{\rm d}t}=k_{\rm h}[{\rm HO}^{\bullet} ][{\rm H}_{2}{\rm O}_{2}]-2k_{\rm t}[{\rm HO}_{2}{}^{\bullet}]^{2} \tag{4}\]
+
+where \(W_{\rm abs}\) is the radiation power absorbed by the solution. \(W_{\rm abs}\) can be expressed as:
+
+\[W_{\rm abs}=I_{0}\left[\,1-\exp(-2.3l(\varepsilon_{\rm s}[{\rm S}]+\varepsilon _{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]))\,\right]f_{{\rm H}_{2}{\rm O }_{2}} \tag{5}\]
+
+By assuming the "steady-state" hypothesis for radical species [35], the stationary HO\({}^{\bullet}\) and HO\({}_{2}{}^{\bullet}\) concentrations can be expressed as
+
+\[[{\rm HO}]_{\rm SS}=\frac{2\phi_{{\rm H}_{2}{\rm O}_{2}}}{V_{\rm sol}}\frac{I_ {0}[1-\exp(-2.3l(\varepsilon_{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]+ \varepsilon_{\rm s}[{\rm S}]))]}{k_{\rm h}[{\rm H}_{2}{\rm O}_{2}]+k_{\rm s}[{ \rm S}]}f_{{\rm H}_{2}{\rm O}_{2}} \tag{6}\]
+
+\[[{\rm HO}_{2}]_{\rm SS}^{2}=\frac{k_{\rm h}}{k_{\rm t}}\frac{\phi_{{\rm H}_{2}{ \rm O}_{2}}}{V_{\rm sol}}\frac{I_{0}[1-\exp(-2.3l(\varepsilon_{{\rm H}_{2}{\rm O }_{2}}[{\rm H}_{2}{\rm O}_{2}]+\varepsilon_{\rm s}[{\rm S}]))][{\rm H}_{2}{\rm O }_{2}]}{k_{\rm h}[{\rm H}_{2}{\rm O}_{2}]+k_{\rm s}[{\rm S}]}f_{{\rm H}_{2}{\rm O }_{2}} \tag{7}\]
+
+and substituting in Eqs. (1) and (2) gives
+
+\[\frac{{\rm d}[{\rm S}]}{{\rm d}t} = -\frac{\phi_{\rm s}}{V_{\rm sol}}\,I_{0}[1-\exp(-2.3l(\varepsilon_ {{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]+\varepsilon_{\rm s}[{\rm S}]))]f_ {\rm s} \tag{8}\] \[+-k_{\rm s}\frac{2\phi_{{\rm H}_{2}{\rm O}_{2}}}{V_{\rm sol}}\frac{I_ {0}[1-\exp(-2.3l(\varepsilon_{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]+ \varepsilon_{\rm s}[{\rm S}]))][{\rm S}]}{k_{\rm h}[{\rm H}_{2}{\rm O}_{2}]+k_{ \rm s}[{\rm S}]}f_{{\rm H}_{2}{\rm O}_{2}}\]
+
+\[\frac{{\rm d}[{\rm H}_{2}{\rm O}_{2}]}{{\rm d}t} = -\frac{\phi_{{\rm H}_{2}{\rm O}_{2}}}{V_{\rm sol}}\,I_{0}[1-\exp(-2.3l( \varepsilon_{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]+\varepsilon_{\rm s}[{\rm S}]))]f_{{\rm H}_{2}{\rm O }_{2}} \tag{9}\] \[+\ -k_{\rm h}\frac{\phi_{{\rm H}_{2}{\rm O}_{2}}}{V_{\rm sol}}\frac{I_ {0}[1-\exp(-2.3l(\varepsilon_{{\rm H}_{2}{\rm O}_{2}}[{\rm H}_{2}{\rm O}_{2}]+ \varepsilon_{\rm s}[{\rm S}]))][{\rm H}_{2}{\rm O}_{2}]}{k_{\rm h}[{\rm H}_{2}{\rm O}_{2}]+k_{ \rm s}[{\rm S}]}f_{{\rm H}_{2}{\rm O}_{2}}\]
+
+Preliminary photolytic runs without hydrogen peroxide addition allowed the determination of quantum yield of the direct photolysis of clofibric acid at 254 nm at pH = 5.5 (\(\phi_{\rm s}=1.08\times 10^{-2}\pm 2.37\times 10^{-4}\) mol E\({}^{-1}\)). Once the value of the parameter \(k_{\rm s}\) is known, Eqs. (8) and (9) can be integrated with the initial conditions: \(t=0\), \([{\rm S}]=[{\rm S}]_{0}\) and \([{\rm H}_{2}{\rm O}_{2}]=[{\rm H}_{2}{\rm O}_{2}]\)[H2O2]0 and the concentrations of the substrate and hydrogen peroxide calculated at varying reaction time. Unfortunately _k_s was not known "a priori" and its value was estimated through the adoption of an optimization procedure [32] by using the experimental data collected in the runs at different hydrogen peroxide starting concentrations (1, 10, 20 and 30 mM) for the same initial concentration of substrate (2.0 x 10-5 M). A mean value for _k_s equal to (2.38 +- 0.18) x 109 M-1 s-1 was thus estimated. A comparison between experimental and calculated data for the oxidation of clofibric acid by UV/H2O2 at pH = 5.5 is shown for different initial concentrations of hydrogen peroxide in Fig. 5.
+
+It is noteworthy to stress that a failure of the model could be expected both when H2O2 levels decrease with respect to those adopted in this investigation (keeping constant the substrate concentrations) or when the substrate concentration increases (working at the same H2O2 levels as in the present experiments). In these conditions, the consumption of OH radicals by the oxidation by-products and their light absorption cannot be neglected
+
+Figure 5: Comparison between experimental (full symbols) and calculated data (solid lines) for UV/H2O2 oxidation of clofibric acid at pH = 5.5 with different H2O2 concentrations: \(T=298\) K.
+
+as assumed for the development of the above-reported model. No effect of the pH on the kinetic constant _k_s was observed in the range 4.0-7.0.
+
+## 5 Kinetic model validation at low concentration
+
+Relevant concentrations of pharmaceuticals in the environment are on the order of few micrograms per liter [10]. It is time consuming to use diluted aqueous solutions for kinetic investigations since each sample needs to be concentrated for the analysis by means of common equipments such as HPLC and GC-MS.
+
+For this reasons a clofibric acid starting concentration in the range 1.0 \(\times\) 10\({}^{-3}\) to 1.5 \(\times\) 10\({}^{-3}\) M was employed for both of the systems aqueous solutions in the first part of the present work. However, to demonstrate that kinetic constants estimated in these runs can be conveniently used in process design for the treatment of real effluents, an attempt was done to model the behaviour of both systems by reducing the starting concentration equal to the maximum value at which clofibric acid has been found in real STP effluents (5.0 \(\times\) 10\({}^{-8}\) M) [33].
+
+Figure 6: Comparison between experimental (full symbols) and calculated data (solid line) for oxidation of clofibric acid with UV/H\({}_{2}\)O\({}_{2}\) at pH = 5.5 with [H\({}_{2}\)O\({}_{2}\)]\({}_{0}\) = 1.0 \(\times\) 10\({}^{-2}\) M: \(T\) = 298 K.
+
+Fig. 6 shows the results obtained by submitting to a photolytic run an aqueous solution containing 10 \(\mu\)g 1\({}^{-1}\) of clofibric acid with an initial hydrogen peroxide of 1.0 \(\times\) 10\({}^{-2}\) M in the experimental apparatus previously described (full circles) along with those predicted by the model (continuous line) by using a value for \(k_{\rm s}\) equal to 2.38 \(\times\) 10\({}^{9}\) M\({}^{-1}\) s\({}^{-1}\).
+
+Similar results have been found for the ozonation experiments. Fig. 7 shows experimental data (full circles) compared with concentrations predicted by using the mean value of the kinetic constants at pH = 5.0 (\(k_{\rm O_{3}}\) = 842.1 M\({}^{-1}\) s\({}^{-1}\)) found during the ozonation runs at higher starting concentrations (continuous line). Although standard deviations up to 18.0% were associated to clofibric acid determination, the comparison between experimental data and those predicted by the models developed in the first part of the paper are encouraging. However, an improvement in the analytical determination is required for a definitive validation of the assessed kinetics.
+
+## 6 Conclusions
+
+The removal of acid clofibric from aqueous solutions has been studied using ozonation and H\({}_{2}\)O\({}_{2}\)/UV systems. Both these systems are able to quickly remove this pharmaceutical
+
+Figure 7: Comparison between experimental (full symbols) and calculated data (solid lines) for ozonation of clofibric acid at pH = 5.0: [dissolved ozone]\({}_{0}\) = 1.0 \(\times\) 10\({}^{-5}\) M; \(T\) = 298 K.
+
+compound with an almost complete conversion of the initial chlorine content into chloride ions. Reaction kinetics have been evaluated in experimental runs with the initial substrate concentration in the range \(1.5\times 10^{-3}\) to \(2.0\times 10^{-5}\) M. A dependence of the ozonation kinetic constants upon the pH has been recorded, in agreement with the capability of the studied species to dissociate in aqueous solution into the more reactive clofibrate ion (\(29.8\,\mathrm{M}^{-1}\,\mathrm{s}^{-1}\) at \(\mathrm{pH}=2.0\) and \(2550\,\mathrm{M}^{-1}\,\mathrm{s}^{-1}\) at \(\mathrm{pH}=6.5\)). No influence of the pH of the solution on the kinetic constant of OH radical attack on the substrate has been observed during H\({}_{2}\)O\({}_{2}\) photolytic experiments (\(2.38\times 10^{9}\,\mathrm{M}^{-1}\,\mathrm{s}^{-1}\)). An attempt to validate the assessed reaction kinetics at low environmentally relevant clofibric acid concentrations has been successfully performed.
+
+## Acknowledgements
+
+The authors wish to thank the Commission of the European Communities for the financial support of this work under Grant No. EVK1-CT-2000-00048.
+
+## References
+
+* [1] C.D. Watts, B. Crathorne, M. Fielding, C.P. Steel, in: Book of Abstracts of the 3rd European Symposium on Organic Micropolutants in Water, Oslo, Norway, 1983.
+* [2] M.L. Richardson, J.M. Bowron, J. Pharm. Pharmacol. 30 (1985) 1.
+* [3] G. W Aherne, J. English, V. Marks, Ecotoxicol. Environ. Safety 9 (1985) 735.
+* [4] T. Heberer, Toxicol. Lett. 131 (2002) 5.
+* [5] D.W. Kolpin, E.T. Furlong, M.T. Meyer, E.M. Thurman, S.D. Zaugg, L.B. Barber, H.T. Buxton, Environ. Sci. Technol. 36 (2002) 1202.
+* [6] K. Kummerer, Chemosphere 45 (2001a) 957.
+* [7] O.A.H. Jones, N. Voulvoulis, J.N. Lester, Environ. Technol. 22 (2001) 1383.
+* [8] S.E. Jorgensen, B. Halling-Sorensen, Chemosphere 40 (2000) 691.
+* [9] C.G. Daughton, T.A. Ternes, Environ. Health Perspect. 107 (1999) 907.
+* [10] T.A. Ternes, Water Res. 32 (1998) 3245.
+* [11] G. Hamscher, S. Sczesny, H. Hoper, H. Nau, Anal. Chem. 35 (17) (2002) 1509.
+* [12] K. Kummerer, in: K. Kummerer (Ed.), Pharmaceuticals in the Environment: Sources, Fate, Effects and Risks, first ed., Springer-Verlag, Berlin, 2001b.
+* [13] N.J. Ayscough, J. Fawell, G. Franklin, W. Young, R&D Technical Report P390, Environment Agency Dissemination Centre, Swindon, Wilts SN5 8YF, 2000.
+* [14] R. Triebskorn, A. Heyd, H. Casper, R. Kohler Heinz, H. Ferling, R. Negele, J. Schwaiger, in: Book of Abstracts of SETAC Europe 12th Annual Meeting: Challenges in Environmental Risk Assessment and Modelling, Linking Basic and Applied Research, Vienna, Austria, 2002.
+* [15] R. Andreozzi, R. Marotta, G. Pinto, A. Pollio, Water Res. 36 (11) (2002) 2869.
+* [16] R. Andreozzi, V. Caprio, R. Marotta, D. Vogna, Water Res. 37 (2003) 993.
+* [17] M. Stumpf, T. Ternes, R.D. Wilken, S.V. Rodrigues, W. Baumann, Sci. Total Environ. 225 (1999) 135.
+* [18] T. Heberer, K. Schmidt-Baumler, H.J. Stan, Acta Hydrochin. Hydrobiol. 26 (5) (1998) 272.
+* [19] E. Zuccato, D. Calamari, M. Natangelo, R. Fanelli, Lancet 355 (2000) 1789.
+* [20] H.R. Buser, M.D. Muller, N. Theobald, Environ. Sci. Technol. 32 (1998) 188.
+* [21] T. Heberer, H.J. Stan, Int. J. Environ. Anal. Chem. 67 (1997) 113.
+* [22] C. Zwiener, F.H. Frimmel, Water Res. 34 (6) (2000) 1881.
+* [23] R. Andreozzi, A. Insola, V. Caprio, M.G. D'Amore, Water Res. 26 (5) (1992) 639.
+* [24] R. Andreozzi, V. Caprio, A. Insola, R. Marotta, Water Res. 34 (2) (1999) 463.
+
+* [25] I. Nicole, J. De Laat, M. Dore, J.P. Duguet, C. Bonnel, Water Res. 24 (2) (1990) 157.
+* [26] R. Andreozzi, A. Insola, V. Caprio, M.G. D'Amore, V. Tufano, Ind. Eng. Chem. Res. 30 (1991) 2098.
+* [27] R. Andreozzi, V. Caprio, A. Insola, V. Tufano, Chem. Eng. Commun. 143 (1996) 195.
+* [28] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data 17 (2) (1988) 513.
+* [29] B.H.J. Bielski, D.E. Cabelli, R.L. Arudi, A.B. Ross, J. Phys. Chem. Ref. Data 14 (4) (1985) 1041.
+* [30] J.H. Baxendale, J.A. Wilson, Trans. Faraday Soc. 53 (1957) 344.
+* [31] D.H. Volman, J.C. Chen, J. Am. Chem. Soc. 81 (1959) 4141.
+* [32] G.V. Reklaitis, A. Ravindran, K.M. Regsdell, in: Engineering Optimization, Wiley, New York, 1983.
+* [33] C. Hignite, D.L. Azarnoff, Life Sci. 20 (1977) 337.
+* [34] T.A. Ternes, M. Meisenheimer, D. McDowell, F. Sacher, H.J. Brauch, B. Haist-Gulde, G. Preuss, U. Wilme, N. Zulei-Seibert, Environ. Sci. Technol. 36 (2002) 3855.
+* [35] J. De Laat, P. Berger, T. Poinot, N.K. Vel Leitner, M. Dore, in: Proceedings of the 12th Ozone World Congress of the International Ozone Association, Lille, May 1995, p. 373.
+* [36] R. Andreozzi, R. Marotta, J. Hazardous Mater. B69 (1999) 303.
+* [37] R.C. Reid, J.M. Prausnitz, B.E. Poling, The Properties of Gases and Liquids, Wiley, New York, 1983.