- Email: [email protected]
Ozonation of indomethacin: Kinetics, mechanisms and toxicity
Accepted Manuscript Title: Ozonation of indomethacin: Kinetics, mechanisms and toxicity Author: Yue Zhao Jiangmeng Kuang Siyu Zhang Xiang Li Bin Wang Jun Huang Shubo Deng Yujue Wang Gang Yu PII: DOI: Reference:
S0304-3894(16)30454-X http://dx.doi.org/doi:10.1016/j.jhazmat.2016.05.023 HAZMAT 17712
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
12-2-2016 6-5-2016 8-5-2016
Please cite this article as: Yue Zhao, Jiangmeng Kuang, Siyu Zhang, Xiang Li, Bin Wang, Jun Huang, Shubo Deng, Yujue Wang, Gang Yu, Ozonation of indomethacin: Kinetics, mechanisms and toxicity, Journal of Hazardous Materials http://dx.doi.org/10.1016/j.jhazmat.2016.05.023 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.
Ozonation of indomethacin: Kinetics, mechanisms and toxicity Yue Zhaoa, Jiangmeng Kuanga, Siyu Zhangb, Xiang Lia, Bin Wanga, Jun Huanga, Shubo Denga, Yujue Wanga, Gang Yu*a
a
School of Environment, Beijing Key Laboratory for Emerging Organic Contaminants Control, State Key
Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), Tsinghua University, Beijing 100084, China b
Key Laboratory of Pollution Ecology and Environmental Engineering, and Institute of Applied Ecology,
Chinese Academy of Science, Shenyang 110016, China * Corresponding author. Tel.: +86 10 62787137; fax: +86 10 62794006. E-mail: [email protected]
Highlights
Kinetic constants of neutral and anionic IM with O3 and ·OH were calculated.
Ozone rather than hydroxyl radical played a dominant role in the ozonation of IM.
Experiments combined with DFT calculation were used to determine intermediates.
Six intermediates and three organic acids were found.
Toxicity of IM against P. phosphoreum was completely eliminated.
Abstract Ozonation of a commonly used non-steroidal anti-inflammatory drug indomethacin (IM) was studied. Kinetic constants of IM with ozone and hydroxyl radicals were measured at an order of magnitude of
105 M-1 s-1 and 109 M-1 s-1 , respectively. IM was degraded within 7 min under the lowest ozone dose, but TOC removal was only 50% even under the highest ozone dose used in the experiments. Ozone rather than hydroxyl radicals was found to be the main oxidant during reaction, with a contribution rate of 80% under pH 7. Six intermediates were identified by high resolution mass spectrometer. Nitrogen atom, C-C double bond and benzene ring were found to be the main reaction sites. Electrophilic attack or Criegee cyclo-addition were proved to be the most probable pathways at the first step. The formation mechanism of one of the ozone products was first proposed during the experiment, then confirmed by the density functional theory (DFT) calculation. Acetic acid, formic acid and oxalic acid were detected as small molecule organic products. The toxicity change during ozonation was measured by luminescent bacterium with results showing that the toxicity can be reduced to zero when ozone dose was high enough.
Key words Ozonation, Pharmaceuticals and Personal Care Products, Kinetics, Mechanism, Toxicity
1. Introduction Pharmaceuticals and personal care products (PPCPs) are a group of emerging contaminants that are largely used throughout the world [1]. Because of their incomplete removal during traditional secondary treatment process [2-4], PPCPs are released into the aquatic environment consecutively [5]. As a result, they become ubiquitous in nearly all environmental compartments with a concentration level up to μg L-1 (i.e. surface water, ground water, sewage sludge and sediments) [6-10]. To date, ecological toxicities of several PPCPs have been reported on both lab scale and field scale [11-14]. In addition, as a sub-category of PPCPs, antibiotics may lead to drug resistance of bacteria when widely present in the environment [15-16]. Non-steroidal anti-inflammatory drugs (NSAIDs), a group of PPCPs, are widely used for their antipyretic, analgesic and anti-inflammatory properties. However, they are always associated with the adverse effects of the gastrointestinal (GI) mucosa injury. Indomethacin (IM) is one of the most damaging NSAIDs for the GI mucosa [17]. In Germany, IM was detected in the effluents of all 49 sewage treatment plants, where the maximum IM concentration was 0.60 μg L-1. As for the surface water, IM was detected in about 80% (34/43) samples with the concentration being 0.20 μg L-1 at most [18]. In fact, other studies show the wide detection of IM [19-21]. In conventional sewage treatment plants, IM cannot be easily removed. The removal efficiency is less than 20% in the primary treatment and less than 10% in the secondary treatment [22]. As a consequence of its occurrence and toxicity, it is important to improve the removal efficiency of IM. To improve the removal efficiency and thus reduce the environmental emission of PPCPs and other
micro-contaminants, the application of tertiary treatment process, or advanced treatment process is increasing in wastewater treatment plants (WWTPs) [2]. As one of commonly used advanced treatment processes, ozonation shows a high performance on removing PPCPs and consequently has been applied in many waste water and drinking water treatment plants [23-26]. However, poor mineralization, as a main shortcoming of ozonation [27-29], may lead to higher ecology and health risk because of the possible higher toxicity of ozonation products compared with parent compounds [27,30]. To identify ozonation products, elucidate reaction mechanism and evaluate toxicity change after ozonation, studies on ozonation of some widely used PPCPs have been carried out [31-34]. A portion of ozonation products were identified and several mechanisms such as cyclo-addition and electrophilic attack were proposed. Most studies employed liquid chromatography tandem mass spectrometer (LC-MS/MS) to detect and identify ozonation products, and further, inferred the reaction pathways and mechanisms based on time profiles and speculative structure of products. Although high resolution mass spectrometer is a powerful instrument to identify chemical structures, the "bottom-up" approach (i.e. infer reaction pathways and mechanisms on the basis of experimental results) may lose some important information due to analytical method, stability of products and sampling time intervals. Thus, the picture drawn is yet incomplete. Quantum chemical calculation can predict reaction products and corresponding mechanisms from a "top-down" approach. However, in most cases the calculation load is too large due to the complexity of most PPCPs molecules. A combination of these two approaches may solve the problems mentioned above. In previous studies, only the degradation curve of the ozonation of IM had been studied [35]. According to previous considerations, the present work was designed to assess the ozonation of IM thoroughly. For this purpose, the kinetic constants of IM with ozone and hydroxyl radicals were calculated respectively. The contribution of each oxidant was further evaluated. Additionally, several ozonation products were identified by liquid chromatography tandem mass spectrometry and ionic chromatography. The reaction mechanisms of several key steps were proposed on the basis of quantum chemical calculation. Also, some unidentified products were predicted based on reaction mechanisms inferred. Last but not least, the toxicity of products mixture was evaluated by luminescent bacteria assay.
2.Materials and methods 2.1 Chemicals All chemical reagents and organic solvents used were of analytical grade (A.R. grade) or higher. Indomethacin, 4-methoxycinnamic acid, acetophenone and para-chlorobenzoic acid were purchased from Sigma-Aldrich (Germany). Lincomycin was purchased from Ehrenstorfer (Germany). Acetonitrile and methanol were purchased from Fisher Scientific (USA). Ammonium acetate was provided by Alfa Aesar (UK). Tert-Butanol (t-BuOH) and phosphate were purchased from Modern Eastern Fine Chemical (China). Ultrapure water was produced by a filtration system (Milli-pore, USA).
2.2 Ozonation experiments setup The ozonation experiments were carried out in a bubble reaction system (shown in Figure S1). A stream of ozone/oxygen mixture was bubbled into a 300 mL reactor filled with 25 μM IM solution at a flow rate of 250 mL min-1. Samples were withdrawn at certain time intervals and residual ozone was purged out with nitrogen. All experiments were carried out in triplicate.
2.3 Determination of kinetic constants Competition kinetic experiments based on the liquid-liquid reaction of saturated ozone solution (750 μM). IM (25 μM) and competition compound C (25 μM, 4-methoxycinnamic acid and lincomycin) mixture were carried out to measure the kinetic constant of IM with ozone and hydroxyl radicals. A stream of ozone/oxygen mixture was bubbled into a 100 mL conical flask filled with 50 mL ultrapure water in an ice-bath for a few hours to produce the saturated ozone solution with a concentration of 700 ~ 800 μM. The phosphate buffer (approximately 8 mM) was used to adjust solution pH and temperature was kept at 20 °C. For measurement of kinetic constant with ozone, different volumes (0 ~ 2 mL) of saturated ozone solution were added into a 50 mL reaction solution of IM and competition compound C mixture. A magnetic stirrer was used to ensure adequate mixing of ozone and reaction solution. The whole system is shown in Figure S2. To scavenge radical reaction, 30 mM t-BuOH was added into the reaction solution before the addition of ozone. The remaining concentration of IM and C were analyzed after the reaction was completed. All experiments were carried out in triplicate. Details on the equations and the experimental design can be found in the Supplementary Material.
2.4 Measurement of oxidation contribution The experimental process was similar with the measurement of the kinetic constant with ozone. Ozone solution was added into reaction solution containing 25 μM IM and 1 μM AP. After the reaction was completed, the concentration of remaining IM and AP were analyzed. All experiments were carried out in triplicate. Details on the equations and the experimental design can be found in the Supplementary Material.
2.5 Analysis methods The concentration of IM, MC, AP and LCM were all measured via an HPLC-UV composed with a 515 HPLC pump, a 717 Auto sampler, and a 2487 dual λ Absorbance Detector (Waters, USA). A reverse phase TC-C18 column (150 mm × 4.6 mm, 5-micron, Agilent, USA) was used with a mobile phase consisting of water and methanol (2 mM ammonium acetate and 0.01% formic acid). The flow rate, injection volume and oven temperature were 1 mL min-1, 50 μL and 30 °C, respectively. Three different isocratic programs of elution were used to measure the four tested analytes. Detailed composition of mobile phase and detection wavelength for each compound please refer to Table S2. Total organic carbon (TOC) was measured by a TOC analyzer TOC-VCPH (Shimadzu, Japan). The concentration of the ozone solution was measured by a
UV spectrometer at 258 nm, using O3 3000 M-1 cm-1 [36]. A high-performance liquid chromatography (HPLC, Ultimate3000, Dionex, USA) followed by electrospray ionization and tandem mass spectrometry (ESI-MS/MS, API3200, AB Sciex, USA) was conducted to measure the relative amount of intermediates. A reverse phase column SB-Aq (150 mm × 3 mm, 3.5 - micron, Agilent, USA) was used as the stationary phase. The mobile phase was a mixture of water (2mM ammonium acetate, phase A) and acetonitrile (phase B). The elution started at 20% of B for 2 min, increased linearly to 40% in 10 min, held isocratically for 5 min, returned to initial condition in 0.1 min and maintained for column regeneration for another 5 min, resulting in a total run time of 22.1 min. The flow rate, injection volume and oven temperature were 0.3 mL min-1, 10 μL and 35 °C, respectively. A hybrid quadrupoleorbitrap mass spectrometer (Q-exactive, Thermo Fisher Scientific, USA) was used to obtain accurate mass and identify chemical structure of intermediates. The mass spectra were obtained in the ESI (-) mode, while the data were collected in the Q1 full scan mode, with a scan range of 120 - 450 Da. A Dionex ICS-2000 ionic chromatography (IC) was used to measure the small carboxylic acids and chloride formed during the ozonation. The flow rate was set at 1.00 mL min-1. The suppressor current was set at 13 mA and 62 mA for carboxylic acids and chloride respectively. 5 mM KOH and 25 mM KOH were selected as the eluent solution for carboxylic acids and chloride, respectively.
2.6 Density functional theory (DFT) calculation DFT calculations were performed with Gaussian 09 software package [37]. Geometry optimizations were performed at the B3LYP/6-31+G(d,p) level. Geometries of transition states (TS) were optimized to saddle points and confirmed by frequency analysis. Single point energies of the optimized geometries were calculated at the B3LYP/6-31++G(d,p) level. Thermal corrections for Gibbs energies were calculated at 298 K, 1 atm. Energies of singlet biradicals like O3 were corrected by a broken symmetry (BS) approach incorporated with approximate spin-projection (ASP) [38]. Details on correction processes were described in another work [39]. Intrinsic reaction coordinate (IRC) calculations starting with optimized geometries of TSs were performed to confirm that TSs connected designated reactants. The integral equation formalism polarized continuum model (IEFPCM) based on the self-consistent-reaction-field (SCRF) [40] was employed for all calculations including geometry optimizations for considering solvent effects. Gibbs free energy of reactions (ΔΔGr) was calculated by ΔGproducts - ΔGreactants. Energy barriers (ΔΔGa‡) were calculated by ΔGTS - ΔGreactants.
2.7 Toxicity assay Luminescent bacterium Photobacterium phosphoreum, one of the most sensitive species [41], was used to evaluate the acute toxicity during the ozonation of IM. Luminescent intensity was measured by a photo detector (BHP9511, Hamamatsu Photonics, China). The acute toxicity was evaluated by equation 9.
Inhibition
L0 L L0
(1)
Here, L0 is the luminescent intensity of blank solution, and L is the luminescent intensity of a sample.
3. Results and discussions 3.1 Kinetics From the degradation curve of IM we can concluded that even in the lowest ozone dose, IM was completely removed within 7 min (Figure 1a). However, the mineralization was poor as shown in Figure 1b.The lowest ozone dose could only remove 20% TOC after a 30 min ozonation. Elevating ozone dose could improve mineralization to some extent, but not to the full extent. When ozone influent concentration increased to 35 mg L-1, there was still more than 50% TOC remaining in the solution after a 30 min ozonation (Figure 1b). This indicates that some products were generated during the ozonation.
To evaluate the ozonation kinetics quantitatively, we measured the kinetic constants of IM reacting with ozone and hydroxyl radicals respectively (Figure 2). Both neutral and anionic IM showed high kinetic constants with ozone on a magnitude of 105 M-1 s-1 (Figure 2a). Furthermore, because of the electrophilic property of ozone, the kinetic constant of IM showed an increase at a higher pH condition when the percentage of anionic form increased. However, the influence of pH on kinetic constant of IM was not as significant as other compounds such as phenols and amines ever showed [42-44]. The kinetic constants of neutral form ( 5.65 105 M-1 s-1 ) and anionic form ( 8.54 105 M-1 s-1 ) of IM were still on the same order of magnitude. This small increase can be explained by the fact that the dissociation moiety of IM is the carboxyl group which cannot be oxidized by ozone but may active ozone preferred sites by dissociation. The kinetic constant of IM with hydroxyl radicals was on the magnitude of 109 M-1 s-1 and did not show much difference between the neutral ( 6.71109 M-1 s-1 ) and the anionic ( 5.51109 M-1 s-1 ) form (Figure 2b). The hydroxyl radicals are non-selective oxidant and reacts very fast with most organic compounds, so the dissociation of IM did not affect the kinetic constant with hydroxyl radicals very much.
3.2 Mechanism Six intermediates with four distinctive molecular weights were observed (Figure 3). Molecular structures of the six OPs were proposed according to results of MS2 spectra (Figure S5, S7, S8, S9, S11 and S13). Each intermediate was named as OP plus molecular weight. Isomers detected at different retention time were differentiated by name-1, name-2, etc. The chemical structures of the MS2 fragments can be seen in Figure S6, S10, S12 and S14. MS2 fragment ions of m/z 261 in Figure S10, m/z 273 and m/z 245 in Figure S12, m/z 245 in Figure S14 were interpreted as radicals. Generation of radical fragments in ESI-MS has been reported by several studies [45-49]. The interpretation of MS2 spectra can be seen in the Supplementary Material.
To elucidate the mechanism clearly, we measured the relative contribution of ozone and hydroxyl
radicals under different pH. The contribution of ozone (when n (O3) : n (IM) = 1:1) under pH 3, pH 7 and pH 10 was (96 ± 0.2)%, (80 ± 0.6)% and (74 ± 4.5)%, respectively. The contribution of ozone when n (O3) : n (IM) = 0.5 : 1 and 1.5 : 1 is shown in Figure S3. It is well known that the hydroxide ion is a radical promoter [50], so the increase of pH will accelerate the self-decomposition of ozone, thus leading to a decrease in the concentration of ozone and an increase in the concentration of hydroxyl radicals. It is rational that the contribution of ozone decreased when pH increased. However, due to the high kinetic constant of IM with ozone, ozone still played a major role in the oxidation of IM even at a high pH of 10. Thus, in the discussion afterwards, we will focus on the reaction mechanism of IM with ozone. In OP333, the methoxy substituted benzene ring was cleaved. As methoxy moiety is an electron donating substituent, its existence will enhance the reaction activity of benzene ring. Ozone can cleave benzene ring via Criegee mechanism [44]. Thus the reaction pathway to OP333 was proposed as Figure 4. Theoretically, an electrophilic attack on the chlorinated benzene ring is disfavored by an electron withdrawn effect of the chlorine atom. But a hydroxylation product of the chlorinated benzene ring was still observed, i.e. OP391. In OP391, besides the hydroxylation of the chlorinated benzene ring, the C-C double bond of the pyrrole ring was also hydroxylated by the ozone attack. In order to confirm that the OP391 was generated by the ozone attack, hydroxyl radicals scavenging experiments were performed. Though with a less amount, OP391 was still detected after scavenging hydroxyl radicals by t-BuOH, indicating the contribution of both ozone and hydroxyl radicals to the formation of OP391. The hydroxylation of the chlorinated benzene ring was proposed to proceed through an oxygen transfer reaction (Figure 5).
Three products with the same molecular weight of 277 but different retention time were observed, named as OP277-1, OP277-2 and OP277-3. Because of the existence of the isomers, there are two structures for OPP277-1, denoted as OP277-1 and OP277-1’. The three products were proposed to be generated through the electron abstraction of ozone. A long-range electron transfer from IM to O3 could occur or an IM-O3 complex was formed before the electron transfer. For OP277, two isomers (OP277-1, OP277-2) with different substituted positions of the hydroxyl group were proposed. The OP277-1, 1’ (Figure 6a, 6b) and OP277-2 (Figure 6d) were proposed to be generated through the formation of a C- and N-centered radical precursor, respectively. During the formation of OP277-3 (Figure 6c), one oxygen atom was transferred from ozone to IM accompanied with the electron abstraction of ozone, therefore, a quinoid precursor was formed. This reaction followed the well accepted electron and oxygen transfer mechanism [44]. The formation pathway of OP156 (i.e. pCBA) was not as obvious as that of other OPs. According to the structure of OP156 and the widely accepted ozonation mechanisms for attacking N atoms [44], OP156 was assumed to be generated from C-N bond cleavage initiated by O3 attacking the N atom (reaction 1). However, the process was found to be moderately endothermic (ΔΔGr = 67 kJ mol-1), meaning that a reverse reaction to form IM is thermodynamically favorable. This result raised doubt on the assumed formation pathway of OP156. In order to resolve the question and unveil the formation process of OP156, further transformation of the intermediate IM-O1 generated through the reaction 1 was investigated.
Two probable transformation pathways leading to the formation of pCBA were proposed for IM-O1, i.e. hydrolysis (reaction 2), or intramolecular rearrangement (reaction 3) followed by hydrolysis (reaction 4). Energies calculated for the reactions are shown in Table S4.The hydrolysis reaction of IM-O1 releases heat of 183 kJ mol-1, meaning that a total reaction of reactions 1 and 2 is highly exothermic and can occur spontaneously leading to the formation of pCBA. The second transformation pathway for IM-O1 is thermodynamically feasible too. The intramolecular rearrangement reaction of IM-O1 is highly exothermic (ΔΔGr = -170 kJ mol-1), though the following hydrolysis reaction requires heat of 84 kJ mol-1. Moreover, the intermediate IM-O3 of reaction 4 can transform to a more thermodynamically stable form (reaction 5) by releasing heat of 97 kJ mol-1. Therefore, it can be inferred that IM-O1 is highly unstable and tends to decompose to form pCBA once formed, though the formation of IM-O1 is moderately endothermic. In order to evaluate the importance of the two transformation pathways for producing pCBA, energy barriers of the processes were calculated. Transition states of the elementary reactions 2 – 4 are shown in Figure S15. As shown in Table S4, the hydrolysis reaction 2 proceeds through a moderate energy barrier of 82 kJ mol-1. Nevertheless, the intramolecular rearrangement reaction 3 has a barrier of 62 kJ mol-1, which is lower than that of hydrolysis (reaction 2), implying that the intramolecular rearrangement is energetically more favorable than hydrolysis for IM-O1. Hydrolysis of IM-O2 (reaction 4) meets a high energy barrier of 226 kJ mol-1, indicating this process occurs slowly. This high energy barrier may result in the accumulation of IM-O2, which could impede intramolecular rearrangement (reaction 3) and facilitate hydrolysis of IM-O1 (reaction 2). Reaction 1:
Cl
Cl
O N O
O N
+ O3
O
O
O IM
+ 1 O2 O
O
O
IM-O1
(2)
Reaction 2:
Cl OH N
O N O O
+ H2O
O
O Reaction 3:
+ O
O
O IM-O1
Cl
O O IM-OH
pCBA
(3)
Cl
Cl
O
O N O O
O N O
O
O
IM-O1
O O
IM-O2
(4)
Reaction 4:
Cl O N H
O O N O
+ H2O
O
+
O IM-O3
O
O
O
O IM-O2
Cl
O pCBA
(5)
Reaction 5:
OH N
O N H O
O
O
O O
O IM-O3
IM-OH
(6)
3.3 Ozonation Products To get more insights on degradation pathways of IM, formation profiles of intermediates were determined at pH 3, 4.5, 7, 10 by ESI-MS/MS (Figure 7). Generation yields of intermediates were determined according to peak areas. For OP333, the highest yield was observed at pH 3. Its generation decreased with an increase of pH. Since ozone molecules could hydrolyze to form hydroxyl radicals at alkaline pH, a higher yield at pH 3 implied OP333 was generated from direct reaction of ozone with IM. A Criegee ozonation mechanism of ozone reactions was proposed for the generation of OP333. The yield of either OP391 or OP156 was the highest at pH 10, and decreased at low pH. This is consistent with an increase in generation of hydroxyl radicals at alkaline pH. Therefore, OP391 was proposed to be formed through a hydroxyl radical induced reaction of IM [51-52]. OP156 (pCBA) was detected at both acidic and basic conditions, implying this product could be formed either by ozone or by hydroxyl radical reactions. The yield of OP277-1, OP277-2 and OP277-3 increased from pH 3 to 7, indicating hydroxyl radicals may accelerate the formation of the three OP277s. But the yield of OP277-1, OP277-2 and OP277-3 at pH 10 was low. This may due to the high concentration of hydroxyl radicals which degraded OP277s. In general, all OPs at pH 3, 4.5, 7 and 10 reached their highest concentrations within the first 5 min. The
intermediates disappeared after 30 min of ozonation, indicating that the intermediates could be completely degraded. In contrast, carboxylic acids behaved differently during the reaction. Three carboxylic acids (i.e. formic acid, acetic acid and oxalic acid) were detected as small molecule organic products. From the time profile (Figure 8) it can be found that these acids were formed from the beginning and could not be further degraded. Studies on the ozonation of other PPCPs also showed that oxidation terminated in the stage of small carboxylic acids [27,53-54], as well as other organic compounds such as cresol [55]. After 30 min of the ozonation, these three acids account for 40% TOC of the products mixture. Although dechlorinated intermediates were not detected by LC-MS/MS, chloride was detected using IC during reaction (Figure 8). After 30 min of the ozonation, 15 μM i.e. 60% of chlorine was emitted into the solution in the form of chloride.
3.4 Toxicity assessment Studies on some other PPCPs have found an increase of toxicity after ozonation [41,56-57]. Sometimes, the toxicity first increases, and then decreases [28,58]. In our study, the toxicity of IM ozonation decreased during ozonation. Firstly, a comparison between the degradation of IM and the toxicity change during ozonation (Figure 1a and Figure 9) was made. It showed that although IM disappeared within 7 min at the lowest ozone dose, the toxicity did not behave the same, indicating that the toxicity during the reaction was mainly contributed by the ozonation products. The comparison between the result of TOC removal and toxicity change during the ozonation was then made (Figure 1b and Figure 9), from which an interesting accordance can be observed that both TOC and toxicity started to decrease at the beginning and then reached a platform at almost the same reaction time. Under the highest ozone dose, although TOC was not completely removed, the toxicity was eliminated. The percentage of inhibition did not change a lot from 10 minutes’ to 60 minutes’ ozonation. This indicated that although the sort of the small molecule organic products might change, the toxicity of these products after 10 minutes' ozonation were stable and not that toxic to Photobacterium phosphoreum. The initially detected toxicity was mainly contributed by IM and intermediates. As section 3.2 shows, at 30 min into the reaction, approximately 40% of the remaining TOC were formic acid, acetic acid and oxalic acid, which are common metabolites of micro-organisms and not so toxic to luminescent bacteria. It is reasonable to infer that the other small molecule organic products are also not toxic and probably are other small carboxylic acids. Moreover, high dechlorination rate may also reduce the toxicity of the products mixture. All those above imply that complete mineralization may not be necessary for eliminating toxicity during ozonation of some PPCPs. Despite the incomplete TOC removal, ozone was found out to be a promising technology to this emergency contaminant. One inevitable problem is that the test organism is another crucial factor influencing risk assessment. Five organisms were used for toxicity assay in the study of Magdeburg [59], according to which different sensitivity and even trend were shown. Thus, on a lab scale, further studies regarding different assay organism are still needed.
4. Conclusions The kinetic constants of neutral and anionic IM with ozone are 5.65 105 M-1 s-1 and
8.54 105 M-1 s-1 , respectively. The kinetic constants of neutral and anionic IM with hydroxyl radicals are 6.71109 M-1 s-1 and 5.51109 M-1 s-1 , respectively. The ozonation attained a total abatement of IM in a short time, but poor mineralization was achieved. Although the kinetic constants of neutral and anionic IM with hydroxyl radicals are far higher than the kinetic constants of neutral and anionic IM with ozone, ozone rather than hydroxyl radicals plays a major role in the oxidation of IM. The limited mineralization infers the existence of ozonation products. Six major intermediates are
found. The mechanism of IM ozonation was detected by both experimental method and DFT calculation method. Nitrogen atom, C-C double bond and benzene ring are the main reaction sites. The first step is usually electrophilic attack or Criegee cyclo-addition, followed by either oxygen transfer, electron transfer, intramolecular rearrangement or hydrolysis. The degradation of IM finally gave rise to the formation and accumulation of small molecule products such as organic compounds (carboxylic acids) and inorganic ions (chloridion), which have negligible toxicity compared to IM and intermediates.
Novelty Statement A thorough study was conducted on the ozonation of a non-steroidal anti-inflammatory drug indomethacin (IM), on which few researches had been done before. After the degradation of IM ozonation and TOC removal were detected, the kinetic constants of IM with ozone and hydroxyl radicals were then measured. The contribution of ozone and hydroxyl radicals during reaction was detected. Both experiments and DFT calculation were used to determine the primary ozonation products. Final ozonation products and the toxicity were studied. Despite the poor mineralization, the toxicity can be fully degraded, indicating ozonation to be a promising technology to remove IM.
Acknowledgments This study was supported by the National High Technology Research and Development Program of China (2013AA06A305), the Program for Changjiang Scholars and Innovative Research Team in University (IRT1261), and the Collaborative Innovation Center for Regional Environmental Quality.
Reference [1] C.G. Daughton, T.A.Ternes, Pharmaceuticals and personal care products in the environment: Agents of subtle change? Environ. Health Persp. 107 (1999) 907-938. [2] Y. Luo, W. Guo, H.H. Ngo, N. Long Duc, F.I. Hai, J. Zhang, S. Liang and X.C. Wang, A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment, Sci. Total Environ. 473 (2014) 619-641. [3] W.C. Li, Occurrence, sources, and fate of pharmaceuticals in aquatic environment and soil, Environ. Pollut. 187 (2014) 193-201. [4] D. Fatta-Kassinos, S. Meric, A. Nikolaou, Pharmaceutical residues in environmental waters and wastewater: current state of knowledge and future research, Anal. Bioanal. Chem. 339 (2011) 251-275. [5] J.B. Carbajo, A.L. Petre, R. Rosal, Continuous ozonation treatment of ofloxacin: Transformation products, water matrix effect and aquatic toxicity, J. of Hazard. Mater. 292 (2015) 34-43. [6] Y.W. Bai, W. Meng, J. Xu,Y. Zhang, C.S. Guo, Occurrence, distribution and bioaccumulation of antibiotics in the Liao River Basin in China, Environ, Sci-Proc. Imp. 16 (2014) 586-593. [7] G.H. Dai, J. Huang, W.W. Chen, B. Wang, G. Yu, S.B. Deng, Major Pharmaceuticals and Personal Care Products (PPCPs) in Wastewater Treatment Plant and Receiving Water in Beijing, China, and Associated Ecological Risks, B. Environ. Contam. Tox. 92 (2014) 655-661. [8] A.Y.C. Lin, W.N. Lee, X.H. Wang, Ketamine and the metabolite norketamine: Persistence and phototransformation toxicity in hospital wastewater and surface water, Water Res. 53 (2014) 351-360. [9] R. Meffe, I. de Bustamante, Emerging organic contaminants in surface water and groundwater: A first overview of the situation in Italy, Sci. Total Environ. 481 (2014) 280-295. [10] M.M.P. Tsui, H.W. Leung, P.K.S. Lam, M.B. Murphy, Seasonal occurrence, removal efficiencies and preliminary risk assessment of multiple classes of organic UV filters in wastewater treatment plants, Water Res. 53 (2014) 58-67. [11] P.P. Fong, A.T. Ford, The biological effects of antidepressants on the molluscs and crustaceans: a review, Aquat. Toxicol. (Amsterdam, Netherlands) 151 (2014) 4-13. [12] M.L. Hedgespeth, P.A. Nilsson, O. Berglund, Ecological implications of altered fish foraging after exposure to an antidepressant pharmaceutical, Aquat. Toxicol. (Amsterdam, Netherlands) 151 (2014) 84-87. [13] J. Liu, G. Lu, D. Wu, Z. Yan, A multi-biomarker assessment of single and combined effects of norfloxacin and sulfamethoxazole on male goldfish (Carassius auratus), Ecotox. Environ. Safe. 102 (2014) 12-17. [14] S.D. Melvin, M.C. Cameron, C.M. Lanctot, Individual and Mixture Toxicity of Pharmaceuticals Naproxen, Carbamazepine, and Sulfamethoxazole to Australian Striped Marsh Frog Tadpoles (Limnodynastes peronii), J. Toxicol. Env. Heal. A 77 (2014) 337-345.
[15] F. Baquero, Low-level antibacterial resistance: a gateway to clinical resistance, Drug Resist. Updat. 4 (2001) 93-105. [16] K. Drlica, The mutant selection window and antimicrobial resistance, J. Antimicrob. Chemoth. 52 (2003) 11-17. [17] C. Carrasco-Pozo, R.L. Castillo, C. Beltrán, A. Miranda, J. Fuentes, M. Gotteland, Molecular mechanisms of gastrointestinal protection by quercetin against indomethacin-induced damage: role of NF-κB and Nrf2, J. Nutr. Biochem. 27 (2015) 289-298. [18] T.A. Ternes, Occurrence of drugs in German sewage plants and rivers, Water Res. 32 (1998) 3245-3260. [19] J.W. Kim, H.S. Jang, J.G. Kim, H. Ishibashi, M. Hirano, K .Nasu, N. Ichikawa, Y. Takao, R. Shinohara, K. Arizono, Occurrence of Pharmaceutical and Personal Care Products (PPCPs) in Surface Water from Mankyung River, South Korea, J. Health Sci. 55 (2009) 249-258. [20] L. Lishman, S.A. Smyth, K. Sarafin, S. Kleywegt, J. Toito, T. Peart, B. Lee, M. Servos, M. Beland, P. Seto, Occurrence and reductions of pharmaceuticals and personal care products and estrogens by municipal wastewater treatment plants in Ontario, Canada, Sci. Total Environ. 367 (2006) 544-558. [21] R. Rosal, A. Rodriguez, J. Antonio Perdigon-Melon, A. Petre, E. Garcia-Calvo, M. Jose Gomez, A. Aguera, A.R. Fernandez-Alba, Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation, Water Res. 44 (2010) 578-588. [22] Q. Sui, J. Huang, S.B. Deng, G. Yu, Q. Fan, Occurrence and removal of pharmaceuticals, caffeine and DEET in wastewater treatment plants of Beijing, China, Water Res. 44 (2010) 417–426. [23] V. Homem, L. Santos, Degradation and removal methods of antibiotics from aqueous matrices - A review, J. Environ. Manage. 92 (2011) 2304-2347. [24] E.A. Murphy, G.B. Post, B.T. Buckley, R.L. Lippincott, M.G. Robson, Future Challenges to Protecting Public Health from Drinking-Water Contaminants, Annu. Rev. Publ. Health 33 (2012) 209-224. [25] R.L. Oulton, T. Kohn, D.M. Cwiertny, Pharmaceuticals and personal care products in effluent matrices: A survey of transformation and removal during wastewater treatment and implications for wastewater management, J. Environ. Monitor. 12 (2010) 1956-1978. [26] M.F. Rahman, E.K. Yanful, S.Y. Jasim, Endocrine disrupting compounds (EDCs) and pharmaceuticals and personal care products (PPCPs) in the aquatic environment: implications for the drinking water industry and global environmental health, J. Water Health 7 (2009) 224-243. [27] F.J. Beltran, A. Aguinaco, J.F. Garcia-Araya, A. Oropesa, Ozone and photocatalytic processes to remove the antibiotic sulfamethoxazole from water, Water Res. 42 (2008) 3799-3808. [28] R.F. Dantas, M. Canterino, R. Marotta, C. Sansa, S. Esplugas, R. Andreozzi, Bezafibrate removal by means of ozonation: Primary intermediates, kinetics, and toxicity assessment, Water Res. 41 (2007) 2525-2532. [29] S. Esplugas, D.M. Bila, L.G.T. Krause, M. Dezotti, Ozonation and advanced oxidation technologies to remove endocrine disrupting chemicals (EDCs) and pharmaceuticals and personal care products
(PPCPs) in water effluents, J. Hazard Mater. 149 (2007) 631-642. [30] R.F. Dantas, C. Sans, S. Esplugas, Ozonation of Propranolol: Transformation, Biodegradability, and Toxicity Assessment, J. Environ. Eng.-Asce 137 (2011) 754-759. [31] J. Benner, T.A. Ternes, Ozonation of Metoprolol: Elucidation of oxidation pathways and major oxidation products, Environ. Sci. Technol. 43 (2009a) 5472-5480. [32] A. Lajeunesse, M. Blais, B. Barbeau, S. Sauve, C. Gagnon, Ozone oxidation of antidepressants in wastewater -Treatment evaluation and characterization of new by-products by LC-QToFMS, Chem. Cent. J. 7 (2013) 1-11. [33] C. Liu, V. Nanaboina, G. Korshin, Spectroscopic study of the degradation of antibiotics and the generation of representative EfOM oxidation products in ozonated wastewater, Chemosphere 86 (2012) 774-782. [34] C. Prasse, M. Wagner, R. Schulz, T.A. Ternes, Oxidation of the Antiviral Drug Acyclovir and Its Biodegradation Product Carboxy-acyclovir with Ozone: Kinetics and Identification of Oxidation Products, Environ. Sci. Technol. 46 (2012) 2169-2178. [35] R.R. Giri, H. Ozaki, S. Ota, Degradation of common pharmaceuticals and personal care products in mixed solutions by advanced oxidation techniques, Int. J. Environ. Sci. Te. 7 (2010) 251-260. [36] J. Benner, T.A. Ternes, , Ozonation of Propranolol: Formation of Oxidation Products, Environ. Sci. Technol. 43 (2009b) 5086-5093. [37] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta Jr., F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian Inc., Wallingford CT, 2009. [38] T. Saito, S. Nishihara, Y. Kataoka, Y. Nakanishi, Matsui, T., Y. Kitagawa, T. Kawakami, M. Okumura, K. Yamaguchi, Transition state optimization based on approximate spin-projection (AP) method, Chem. Phys. Lett. 483 (2009) 168–171. [39] S. Zhang, G. Yu, J. Chen, B. Wang, J. Huang, S. Deng, 2014, Unveiling formation mechanism of carcinogenic N-nitrosodimethylamine in ozonation of dimethylamine: a density functional theoretical investigation, J. Hazard Mater. 279 (2014) 330–335. [40] J. Tomasi, B. Mennucci, R. Cammi, Quantum mechanical continuum solvation models, Chem. Rev. 105 (2005) 2999–3093. [41] J. Kuang, J. Huang, B. Wang, Q. Cao, S. Deng, G. Yu, Ozonation of trimethoprim in aqueous solution:
Identification of reaction products and their toxicity, Water Res. 47 (2013) 2863-2872. [42] M.C. Dodd, M.O. 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. [43] J. Hoigné, 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. [44] C. von Sonntag, U. von Gunten, Chemistry of Ozone in Water and Wastewater Treatment: From Basic Principles to Applications, International Water Assn, 2012. [45] E.M. Thurman, I. Ferrer, O.J. Pozo, J.V. Sancho, F. Hernandez, The even-electron rule in electrospray mass spectra of pesticides, Rapid Commun. Mass Spectrom. 21 (2007) 3855-3868. [46] K. Levsen, H.M. Schiebel, J.K. Terlouw, K.J. Jobst, M. Elend, A. Preiβ, H. Thiele, A. Ingendoh, Even-electron ions: a systematic study of the neutral species lost in the dissociation of quasi-molecular ions, J. Mass Spectrom. 42 (2007) 1024-1044. [47] A.B. Attygalle, J. Ruzicka, D. Varughese, J.B. Bialecki, S. Jafri, Loss of benzene to generate an enolate anion by a site-specific double-hydrogen transfer during CID fragmentation of o-alkyl ethers of ortho-hydroxybenzoic acids, J. Mass Spectrom. 42 (2007) 1224-1234. [48] G. Xu, T. Huang, J. Zhang, J.K. Huang, T. Carlson, S. Miao, Investigation of collision-induced dissociations involving odd-electron ion formation under positive electrospray ionization conditions using accurate mass, Rapid Commun. Mass Spectrom. 24 (2010) 321-327. [49] F.M. Wendel, L.E. Christian, E.J. Machek, S.E. Duirk, M.J. Plewa, S.D. Richardson, T.A. Ternes, Transformation of iopamidol during chlorination. Environ. Sci. Technol. 48 (2014) 12689-12697. [50] J. Staehelin, 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 (1985) 1206-1213. [51] T.H. Lay, J.W. Bozzelli, J.H. Seinfeld, Aatmospheric photochemical oxidation of benzene: benzene plus OH and the benzene-OH adduct (hydroxyl-2,4-cyclohexadienyl) plus O2, J. Phys. Chem. 100 (1996), 6543-6554. [52] K.S. Tay, N.A. Rahman, M.R. Bin Abas, Characterization of atenolol trasformation products in ozonation by using rapid resolution high-performance liquid chromatography/quadrupole-time-of-flight mass spectrometry, Microchem. J. 99 (2011), 312-326. [53] C. Quispe, F.M. Nachtigall, M.F.R. Fonseca, R.M. Alberici, L. Astudillo, J. Villasenor, M.N. Eberlin, L.S. Santos, Monitoring of beta-Blockers Ozone Degradation via Electrospray Ionization Mass Spectrometry, J. Brazil. Chem. Soc. 22 (2011), 919-928. [54] M. Skoumal, P.L. Cabot, F. Centellas, C. Arias, R.M. Rodriguez , J.A. Garrido, E. Brillas, 2006, Mineralization of paracetamol by ozonation catalyzed with Fe2+, Cu2+ and UVA light, Appl. Catal. B-Environ. 66 (2006) 228-240. [55] M.C. Valsania, F. Fasano, S.D. Richardson, M. Vincenti, Investigation of the degradation of cresols in the treatments with ozone, Water Res. 46 (2012) 2795-2804. [56] M.N. Abellan, W. Gebhardt, H.F. Schroder, Detection and identification of degradation products of sulfamethoxazole by means of LC/MS and -MSn after ozone treatment, Water Sci. Technol. 58 (2008)
1803-1812. [57] R. Rosal, M.S. Gonzalo, K. Boltes, P. Leton, J.J. Vaquero, E. Garcia-Calvo, Identification of intermediates and assessment of ecotoxicity in the oxidation products generated during the ozonation of clofibric acid, J. Hazard Mater. 172 (2009) 1061-1068. [58] M. D. M. Gomez-Ramos, M. Mezcua, A. Agueera, A.R. Fernandez-Alba, S. Gonzalo, A. Rodriguez, R. Rosal, Chemical and toxicological evolution of the antibiotic sulfamethoxazole under ozone treatment in water solution, J. Hazard Mater. 192 (2011) 18-25. [59] A. Magdeburg, D. Stalter, J. Oehlmann, Whole effluent toxicity assessment at a wastewater treatment plant upgraded with a full-scale post-ozonation using aquatic key species, Chemosphere 88 (2012) 1008-1014.
(a)
(b) Figure 1. Degreadation curve of IM (a) and TOC (b) under different ozone doses
(a)
(b) Figure 2. Kinetic constants of IM with ozone (a) and hydroxyl radicals (b) under different dissociation degree of IM
Figure 3. Structures of IM and ozonation products
Figure 4. Formation mechanism of OP333
Figure 5. Formation mechanism of OP391
(a)
(c) Figure 6. Formation mechanism of OP277-1 (a), OP277-1’ (b), OP277-2 (c), OP277-3 (d)
(b)
(d)
(a)
(b)
(c)
(d)
(e)
(f)
Figure 7. Time profile of OP156 (a), OP277-1 (b), OP277-2 (c), OP277-3 (d), OP333 (e) and OP 391 (f) under the influent gaseous ozone concentration of 2 mg L-1
23
Figure 8. Time profile of carboxylic acid and chloride under the influent gaseous ozone concentration of 2 mg L-1
24
Figure 9. Toxicity changes during ozonation under different ozone doses
25
S0304-3894(16)30454-X http://dx.doi.org/doi:10.1016/j.jhazmat.2016.05.023 HAZMAT 17712
To appear in:
Journal of Hazardous Materials
Received date: Revised date: Accepted date:
12-2-2016 6-5-2016 8-5-2016
Please cite this article as: Yue Zhao, Jiangmeng Kuang, Siyu Zhang, Xiang Li, Bin Wang, Jun Huang, Shubo Deng, Yujue Wang, Gang Yu, Ozonation of indomethacin: Kinetics, mechanisms and toxicity, Journal of Hazardous Materials http://dx.doi.org/10.1016/j.jhazmat.2016.05.023 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.
Ozonation of indomethacin: Kinetics, mechanisms and toxicity Yue Zhaoa, Jiangmeng Kuanga, Siyu Zhangb, Xiang Lia, Bin Wanga, Jun Huanga, Shubo Denga, Yujue Wanga, Gang Yu*a
a
School of Environment, Beijing Key Laboratory for Emerging Organic Contaminants Control, State Key
Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), Tsinghua University, Beijing 100084, China b
Key Laboratory of Pollution Ecology and Environmental Engineering, and Institute of Applied Ecology,
Chinese Academy of Science, Shenyang 110016, China * Corresponding author. Tel.: +86 10 62787137; fax: +86 10 62794006. E-mail: [email protected]
Highlights
Kinetic constants of neutral and anionic IM with O3 and ·OH were calculated.
Ozone rather than hydroxyl radical played a dominant role in the ozonation of IM.
Experiments combined with DFT calculation were used to determine intermediates.
Six intermediates and three organic acids were found.
Toxicity of IM against P. phosphoreum was completely eliminated.
Abstract Ozonation of a commonly used non-steroidal anti-inflammatory drug indomethacin (IM) was studied. Kinetic constants of IM with ozone and hydroxyl radicals were measured at an order of magnitude of
105 M-1 s-1 and 109 M-1 s-1 , respectively. IM was degraded within 7 min under the lowest ozone dose, but TOC removal was only 50% even under the highest ozone dose used in the experiments. Ozone rather than hydroxyl radicals was found to be the main oxidant during reaction, with a contribution rate of 80% under pH 7. Six intermediates were identified by high resolution mass spectrometer. Nitrogen atom, C-C double bond and benzene ring were found to be the main reaction sites. Electrophilic attack or Criegee cyclo-addition were proved to be the most probable pathways at the first step. The formation mechanism of one of the ozone products was first proposed during the experiment, then confirmed by the density functional theory (DFT) calculation. Acetic acid, formic acid and oxalic acid were detected as small molecule organic products. The toxicity change during ozonation was measured by luminescent bacterium with results showing that the toxicity can be reduced to zero when ozone dose was high enough.
Key words Ozonation, Pharmaceuticals and Personal Care Products, Kinetics, Mechanism, Toxicity
1. Introduction Pharmaceuticals and personal care products (PPCPs) are a group of emerging contaminants that are largely used throughout the world [1]. Because of their incomplete removal during traditional secondary treatment process [2-4], PPCPs are released into the aquatic environment consecutively [5]. As a result, they become ubiquitous in nearly all environmental compartments with a concentration level up to μg L-1 (i.e. surface water, ground water, sewage sludge and sediments) [6-10]. To date, ecological toxicities of several PPCPs have been reported on both lab scale and field scale [11-14]. In addition, as a sub-category of PPCPs, antibiotics may lead to drug resistance of bacteria when widely present in the environment [15-16]. Non-steroidal anti-inflammatory drugs (NSAIDs), a group of PPCPs, are widely used for their antipyretic, analgesic and anti-inflammatory properties. However, they are always associated with the adverse effects of the gastrointestinal (GI) mucosa injury. Indomethacin (IM) is one of the most damaging NSAIDs for the GI mucosa [17]. In Germany, IM was detected in the effluents of all 49 sewage treatment plants, where the maximum IM concentration was 0.60 μg L-1. As for the surface water, IM was detected in about 80% (34/43) samples with the concentration being 0.20 μg L-1 at most [18]. In fact, other studies show the wide detection of IM [19-21]. In conventional sewage treatment plants, IM cannot be easily removed. The removal efficiency is less than 20% in the primary treatment and less than 10% in the secondary treatment [22]. As a consequence of its occurrence and toxicity, it is important to improve the removal efficiency of IM. To improve the removal efficiency and thus reduce the environmental emission of PPCPs and other
micro-contaminants, the application of tertiary treatment process, or advanced treatment process is increasing in wastewater treatment plants (WWTPs) [2]. As one of commonly used advanced treatment processes, ozonation shows a high performance on removing PPCPs and consequently has been applied in many waste water and drinking water treatment plants [23-26]. However, poor mineralization, as a main shortcoming of ozonation [27-29], may lead to higher ecology and health risk because of the possible higher toxicity of ozonation products compared with parent compounds [27,30]. To identify ozonation products, elucidate reaction mechanism and evaluate toxicity change after ozonation, studies on ozonation of some widely used PPCPs have been carried out [31-34]. A portion of ozonation products were identified and several mechanisms such as cyclo-addition and electrophilic attack were proposed. Most studies employed liquid chromatography tandem mass spectrometer (LC-MS/MS) to detect and identify ozonation products, and further, inferred the reaction pathways and mechanisms based on time profiles and speculative structure of products. Although high resolution mass spectrometer is a powerful instrument to identify chemical structures, the "bottom-up" approach (i.e. infer reaction pathways and mechanisms on the basis of experimental results) may lose some important information due to analytical method, stability of products and sampling time intervals. Thus, the picture drawn is yet incomplete. Quantum chemical calculation can predict reaction products and corresponding mechanisms from a "top-down" approach. However, in most cases the calculation load is too large due to the complexity of most PPCPs molecules. A combination of these two approaches may solve the problems mentioned above. In previous studies, only the degradation curve of the ozonation of IM had been studied [35]. According to previous considerations, the present work was designed to assess the ozonation of IM thoroughly. For this purpose, the kinetic constants of IM with ozone and hydroxyl radicals were calculated respectively. The contribution of each oxidant was further evaluated. Additionally, several ozonation products were identified by liquid chromatography tandem mass spectrometry and ionic chromatography. The reaction mechanisms of several key steps were proposed on the basis of quantum chemical calculation. Also, some unidentified products were predicted based on reaction mechanisms inferred. Last but not least, the toxicity of products mixture was evaluated by luminescent bacteria assay.
2.Materials and methods 2.1 Chemicals All chemical reagents and organic solvents used were of analytical grade (A.R. grade) or higher. Indomethacin, 4-methoxycinnamic acid, acetophenone and para-chlorobenzoic acid were purchased from Sigma-Aldrich (Germany). Lincomycin was purchased from Ehrenstorfer (Germany). Acetonitrile and methanol were purchased from Fisher Scientific (USA). Ammonium acetate was provided by Alfa Aesar (UK). Tert-Butanol (t-BuOH) and phosphate were purchased from Modern Eastern Fine Chemical (China). Ultrapure water was produced by a filtration system (Milli-pore, USA).
2.2 Ozonation experiments setup The ozonation experiments were carried out in a bubble reaction system (shown in Figure S1). A stream of ozone/oxygen mixture was bubbled into a 300 mL reactor filled with 25 μM IM solution at a flow rate of 250 mL min-1. Samples were withdrawn at certain time intervals and residual ozone was purged out with nitrogen. All experiments were carried out in triplicate.
2.3 Determination of kinetic constants Competition kinetic experiments based on the liquid-liquid reaction of saturated ozone solution (750 μM). IM (25 μM) and competition compound C (25 μM, 4-methoxycinnamic acid and lincomycin) mixture were carried out to measure the kinetic constant of IM with ozone and hydroxyl radicals. A stream of ozone/oxygen mixture was bubbled into a 100 mL conical flask filled with 50 mL ultrapure water in an ice-bath for a few hours to produce the saturated ozone solution with a concentration of 700 ~ 800 μM. The phosphate buffer (approximately 8 mM) was used to adjust solution pH and temperature was kept at 20 °C. For measurement of kinetic constant with ozone, different volumes (0 ~ 2 mL) of saturated ozone solution were added into a 50 mL reaction solution of IM and competition compound C mixture. A magnetic stirrer was used to ensure adequate mixing of ozone and reaction solution. The whole system is shown in Figure S2. To scavenge radical reaction, 30 mM t-BuOH was added into the reaction solution before the addition of ozone. The remaining concentration of IM and C were analyzed after the reaction was completed. All experiments were carried out in triplicate. Details on the equations and the experimental design can be found in the Supplementary Material.
2.4 Measurement of oxidation contribution The experimental process was similar with the measurement of the kinetic constant with ozone. Ozone solution was added into reaction solution containing 25 μM IM and 1 μM AP. After the reaction was completed, the concentration of remaining IM and AP were analyzed. All experiments were carried out in triplicate. Details on the equations and the experimental design can be found in the Supplementary Material.
2.5 Analysis methods The concentration of IM, MC, AP and LCM were all measured via an HPLC-UV composed with a 515 HPLC pump, a 717 Auto sampler, and a 2487 dual λ Absorbance Detector (Waters, USA). A reverse phase TC-C18 column (150 mm × 4.6 mm, 5-micron, Agilent, USA) was used with a mobile phase consisting of water and methanol (2 mM ammonium acetate and 0.01% formic acid). The flow rate, injection volume and oven temperature were 1 mL min-1, 50 μL and 30 °C, respectively. Three different isocratic programs of elution were used to measure the four tested analytes. Detailed composition of mobile phase and detection wavelength for each compound please refer to Table S2. Total organic carbon (TOC) was measured by a TOC analyzer TOC-VCPH (Shimadzu, Japan). The concentration of the ozone solution was measured by a
UV spectrometer at 258 nm, using O3 3000 M-1 cm-1 [36]. A high-performance liquid chromatography (HPLC, Ultimate3000, Dionex, USA) followed by electrospray ionization and tandem mass spectrometry (ESI-MS/MS, API3200, AB Sciex, USA) was conducted to measure the relative amount of intermediates. A reverse phase column SB-Aq (150 mm × 3 mm, 3.5 - micron, Agilent, USA) was used as the stationary phase. The mobile phase was a mixture of water (2mM ammonium acetate, phase A) and acetonitrile (phase B). The elution started at 20% of B for 2 min, increased linearly to 40% in 10 min, held isocratically for 5 min, returned to initial condition in 0.1 min and maintained for column regeneration for another 5 min, resulting in a total run time of 22.1 min. The flow rate, injection volume and oven temperature were 0.3 mL min-1, 10 μL and 35 °C, respectively. A hybrid quadrupoleorbitrap mass spectrometer (Q-exactive, Thermo Fisher Scientific, USA) was used to obtain accurate mass and identify chemical structure of intermediates. The mass spectra were obtained in the ESI (-) mode, while the data were collected in the Q1 full scan mode, with a scan range of 120 - 450 Da. A Dionex ICS-2000 ionic chromatography (IC) was used to measure the small carboxylic acids and chloride formed during the ozonation. The flow rate was set at 1.00 mL min-1. The suppressor current was set at 13 mA and 62 mA for carboxylic acids and chloride respectively. 5 mM KOH and 25 mM KOH were selected as the eluent solution for carboxylic acids and chloride, respectively.
2.6 Density functional theory (DFT) calculation DFT calculations were performed with Gaussian 09 software package [37]. Geometry optimizations were performed at the B3LYP/6-31+G(d,p) level. Geometries of transition states (TS) were optimized to saddle points and confirmed by frequency analysis. Single point energies of the optimized geometries were calculated at the B3LYP/6-31++G(d,p) level. Thermal corrections for Gibbs energies were calculated at 298 K, 1 atm. Energies of singlet biradicals like O3 were corrected by a broken symmetry (BS) approach incorporated with approximate spin-projection (ASP) [38]. Details on correction processes were described in another work [39]. Intrinsic reaction coordinate (IRC) calculations starting with optimized geometries of TSs were performed to confirm that TSs connected designated reactants. The integral equation formalism polarized continuum model (IEFPCM) based on the self-consistent-reaction-field (SCRF) [40] was employed for all calculations including geometry optimizations for considering solvent effects. Gibbs free energy of reactions (ΔΔGr) was calculated by ΔGproducts - ΔGreactants. Energy barriers (ΔΔGa‡) were calculated by ΔGTS - ΔGreactants.
2.7 Toxicity assay Luminescent bacterium Photobacterium phosphoreum, one of the most sensitive species [41], was used to evaluate the acute toxicity during the ozonation of IM. Luminescent intensity was measured by a photo detector (BHP9511, Hamamatsu Photonics, China). The acute toxicity was evaluated by equation 9.
Inhibition
L0 L L0
(1)
Here, L0 is the luminescent intensity of blank solution, and L is the luminescent intensity of a sample.
3. Results and discussions 3.1 Kinetics From the degradation curve of IM we can concluded that even in the lowest ozone dose, IM was completely removed within 7 min (Figure 1a). However, the mineralization was poor as shown in Figure 1b.The lowest ozone dose could only remove 20% TOC after a 30 min ozonation. Elevating ozone dose could improve mineralization to some extent, but not to the full extent. When ozone influent concentration increased to 35 mg L-1, there was still more than 50% TOC remaining in the solution after a 30 min ozonation (Figure 1b). This indicates that some products were generated during the ozonation.
To evaluate the ozonation kinetics quantitatively, we measured the kinetic constants of IM reacting with ozone and hydroxyl radicals respectively (Figure 2). Both neutral and anionic IM showed high kinetic constants with ozone on a magnitude of 105 M-1 s-1 (Figure 2a). Furthermore, because of the electrophilic property of ozone, the kinetic constant of IM showed an increase at a higher pH condition when the percentage of anionic form increased. However, the influence of pH on kinetic constant of IM was not as significant as other compounds such as phenols and amines ever showed [42-44]. The kinetic constants of neutral form ( 5.65 105 M-1 s-1 ) and anionic form ( 8.54 105 M-1 s-1 ) of IM were still on the same order of magnitude. This small increase can be explained by the fact that the dissociation moiety of IM is the carboxyl group which cannot be oxidized by ozone but may active ozone preferred sites by dissociation. The kinetic constant of IM with hydroxyl radicals was on the magnitude of 109 M-1 s-1 and did not show much difference between the neutral ( 6.71109 M-1 s-1 ) and the anionic ( 5.51109 M-1 s-1 ) form (Figure 2b). The hydroxyl radicals are non-selective oxidant and reacts very fast with most organic compounds, so the dissociation of IM did not affect the kinetic constant with hydroxyl radicals very much.
3.2 Mechanism Six intermediates with four distinctive molecular weights were observed (Figure 3). Molecular structures of the six OPs were proposed according to results of MS2 spectra (Figure S5, S7, S8, S9, S11 and S13). Each intermediate was named as OP plus molecular weight. Isomers detected at different retention time were differentiated by name-1, name-2, etc. The chemical structures of the MS2 fragments can be seen in Figure S6, S10, S12 and S14. MS2 fragment ions of m/z 261 in Figure S10, m/z 273 and m/z 245 in Figure S12, m/z 245 in Figure S14 were interpreted as radicals. Generation of radical fragments in ESI-MS has been reported by several studies [45-49]. The interpretation of MS2 spectra can be seen in the Supplementary Material.
To elucidate the mechanism clearly, we measured the relative contribution of ozone and hydroxyl
radicals under different pH. The contribution of ozone (when n (O3) : n (IM) = 1:1) under pH 3, pH 7 and pH 10 was (96 ± 0.2)%, (80 ± 0.6)% and (74 ± 4.5)%, respectively. The contribution of ozone when n (O3) : n (IM) = 0.5 : 1 and 1.5 : 1 is shown in Figure S3. It is well known that the hydroxide ion is a radical promoter [50], so the increase of pH will accelerate the self-decomposition of ozone, thus leading to a decrease in the concentration of ozone and an increase in the concentration of hydroxyl radicals. It is rational that the contribution of ozone decreased when pH increased. However, due to the high kinetic constant of IM with ozone, ozone still played a major role in the oxidation of IM even at a high pH of 10. Thus, in the discussion afterwards, we will focus on the reaction mechanism of IM with ozone. In OP333, the methoxy substituted benzene ring was cleaved. As methoxy moiety is an electron donating substituent, its existence will enhance the reaction activity of benzene ring. Ozone can cleave benzene ring via Criegee mechanism [44]. Thus the reaction pathway to OP333 was proposed as Figure 4. Theoretically, an electrophilic attack on the chlorinated benzene ring is disfavored by an electron withdrawn effect of the chlorine atom. But a hydroxylation product of the chlorinated benzene ring was still observed, i.e. OP391. In OP391, besides the hydroxylation of the chlorinated benzene ring, the C-C double bond of the pyrrole ring was also hydroxylated by the ozone attack. In order to confirm that the OP391 was generated by the ozone attack, hydroxyl radicals scavenging experiments were performed. Though with a less amount, OP391 was still detected after scavenging hydroxyl radicals by t-BuOH, indicating the contribution of both ozone and hydroxyl radicals to the formation of OP391. The hydroxylation of the chlorinated benzene ring was proposed to proceed through an oxygen transfer reaction (Figure 5).
Three products with the same molecular weight of 277 but different retention time were observed, named as OP277-1, OP277-2 and OP277-3. Because of the existence of the isomers, there are two structures for OPP277-1, denoted as OP277-1 and OP277-1’. The three products were proposed to be generated through the electron abstraction of ozone. A long-range electron transfer from IM to O3 could occur or an IM-O3 complex was formed before the electron transfer. For OP277, two isomers (OP277-1, OP277-2) with different substituted positions of the hydroxyl group were proposed. The OP277-1, 1’ (Figure 6a, 6b) and OP277-2 (Figure 6d) were proposed to be generated through the formation of a C- and N-centered radical precursor, respectively. During the formation of OP277-3 (Figure 6c), one oxygen atom was transferred from ozone to IM accompanied with the electron abstraction of ozone, therefore, a quinoid precursor was formed. This reaction followed the well accepted electron and oxygen transfer mechanism [44]. The formation pathway of OP156 (i.e. pCBA) was not as obvious as that of other OPs. According to the structure of OP156 and the widely accepted ozonation mechanisms for attacking N atoms [44], OP156 was assumed to be generated from C-N bond cleavage initiated by O3 attacking the N atom (reaction 1). However, the process was found to be moderately endothermic (ΔΔGr = 67 kJ mol-1), meaning that a reverse reaction to form IM is thermodynamically favorable. This result raised doubt on the assumed formation pathway of OP156. In order to resolve the question and unveil the formation process of OP156, further transformation of the intermediate IM-O1 generated through the reaction 1 was investigated.
Two probable transformation pathways leading to the formation of pCBA were proposed for IM-O1, i.e. hydrolysis (reaction 2), or intramolecular rearrangement (reaction 3) followed by hydrolysis (reaction 4). Energies calculated for the reactions are shown in Table S4.The hydrolysis reaction of IM-O1 releases heat of 183 kJ mol-1, meaning that a total reaction of reactions 1 and 2 is highly exothermic and can occur spontaneously leading to the formation of pCBA. The second transformation pathway for IM-O1 is thermodynamically feasible too. The intramolecular rearrangement reaction of IM-O1 is highly exothermic (ΔΔGr = -170 kJ mol-1), though the following hydrolysis reaction requires heat of 84 kJ mol-1. Moreover, the intermediate IM-O3 of reaction 4 can transform to a more thermodynamically stable form (reaction 5) by releasing heat of 97 kJ mol-1. Therefore, it can be inferred that IM-O1 is highly unstable and tends to decompose to form pCBA once formed, though the formation of IM-O1 is moderately endothermic. In order to evaluate the importance of the two transformation pathways for producing pCBA, energy barriers of the processes were calculated. Transition states of the elementary reactions 2 – 4 are shown in Figure S15. As shown in Table S4, the hydrolysis reaction 2 proceeds through a moderate energy barrier of 82 kJ mol-1. Nevertheless, the intramolecular rearrangement reaction 3 has a barrier of 62 kJ mol-1, which is lower than that of hydrolysis (reaction 2), implying that the intramolecular rearrangement is energetically more favorable than hydrolysis for IM-O1. Hydrolysis of IM-O2 (reaction 4) meets a high energy barrier of 226 kJ mol-1, indicating this process occurs slowly. This high energy barrier may result in the accumulation of IM-O2, which could impede intramolecular rearrangement (reaction 3) and facilitate hydrolysis of IM-O1 (reaction 2). Reaction 1:
Cl
Cl
O N O
O N
+ O3
O
O
O IM
+ 1 O2 O
O
O
IM-O1
(2)
Reaction 2:
Cl OH N
O N O O
+ H2O
O
O Reaction 3:
+ O
O
O IM-O1
Cl
O O IM-OH
pCBA
(3)
Cl
Cl
O
O N O O
O N O
O
O
IM-O1
O O
IM-O2
(4)
Reaction 4:
Cl O N H
O O N O
+ H2O
O
+
O IM-O3
O
O
O
O IM-O2
Cl
O pCBA
(5)
Reaction 5:
OH N
O N H O
O
O
O O
O IM-O3
IM-OH
(6)
3.3 Ozonation Products To get more insights on degradation pathways of IM, formation profiles of intermediates were determined at pH 3, 4.5, 7, 10 by ESI-MS/MS (Figure 7). Generation yields of intermediates were determined according to peak areas. For OP333, the highest yield was observed at pH 3. Its generation decreased with an increase of pH. Since ozone molecules could hydrolyze to form hydroxyl radicals at alkaline pH, a higher yield at pH 3 implied OP333 was generated from direct reaction of ozone with IM. A Criegee ozonation mechanism of ozone reactions was proposed for the generation of OP333. The yield of either OP391 or OP156 was the highest at pH 10, and decreased at low pH. This is consistent with an increase in generation of hydroxyl radicals at alkaline pH. Therefore, OP391 was proposed to be formed through a hydroxyl radical induced reaction of IM [51-52]. OP156 (pCBA) was detected at both acidic and basic conditions, implying this product could be formed either by ozone or by hydroxyl radical reactions. The yield of OP277-1, OP277-2 and OP277-3 increased from pH 3 to 7, indicating hydroxyl radicals may accelerate the formation of the three OP277s. But the yield of OP277-1, OP277-2 and OP277-3 at pH 10 was low. This may due to the high concentration of hydroxyl radicals which degraded OP277s. In general, all OPs at pH 3, 4.5, 7 and 10 reached their highest concentrations within the first 5 min. The
intermediates disappeared after 30 min of ozonation, indicating that the intermediates could be completely degraded. In contrast, carboxylic acids behaved differently during the reaction. Three carboxylic acids (i.e. formic acid, acetic acid and oxalic acid) were detected as small molecule organic products. From the time profile (Figure 8) it can be found that these acids were formed from the beginning and could not be further degraded. Studies on the ozonation of other PPCPs also showed that oxidation terminated in the stage of small carboxylic acids [27,53-54], as well as other organic compounds such as cresol [55]. After 30 min of the ozonation, these three acids account for 40% TOC of the products mixture. Although dechlorinated intermediates were not detected by LC-MS/MS, chloride was detected using IC during reaction (Figure 8). After 30 min of the ozonation, 15 μM i.e. 60% of chlorine was emitted into the solution in the form of chloride.
3.4 Toxicity assessment Studies on some other PPCPs have found an increase of toxicity after ozonation [41,56-57]. Sometimes, the toxicity first increases, and then decreases [28,58]. In our study, the toxicity of IM ozonation decreased during ozonation. Firstly, a comparison between the degradation of IM and the toxicity change during ozonation (Figure 1a and Figure 9) was made. It showed that although IM disappeared within 7 min at the lowest ozone dose, the toxicity did not behave the same, indicating that the toxicity during the reaction was mainly contributed by the ozonation products. The comparison between the result of TOC removal and toxicity change during the ozonation was then made (Figure 1b and Figure 9), from which an interesting accordance can be observed that both TOC and toxicity started to decrease at the beginning and then reached a platform at almost the same reaction time. Under the highest ozone dose, although TOC was not completely removed, the toxicity was eliminated. The percentage of inhibition did not change a lot from 10 minutes’ to 60 minutes’ ozonation. This indicated that although the sort of the small molecule organic products might change, the toxicity of these products after 10 minutes' ozonation were stable and not that toxic to Photobacterium phosphoreum. The initially detected toxicity was mainly contributed by IM and intermediates. As section 3.2 shows, at 30 min into the reaction, approximately 40% of the remaining TOC were formic acid, acetic acid and oxalic acid, which are common metabolites of micro-organisms and not so toxic to luminescent bacteria. It is reasonable to infer that the other small molecule organic products are also not toxic and probably are other small carboxylic acids. Moreover, high dechlorination rate may also reduce the toxicity of the products mixture. All those above imply that complete mineralization may not be necessary for eliminating toxicity during ozonation of some PPCPs. Despite the incomplete TOC removal, ozone was found out to be a promising technology to this emergency contaminant. One inevitable problem is that the test organism is another crucial factor influencing risk assessment. Five organisms were used for toxicity assay in the study of Magdeburg [59], according to which different sensitivity and even trend were shown. Thus, on a lab scale, further studies regarding different assay organism are still needed.
4. Conclusions The kinetic constants of neutral and anionic IM with ozone are 5.65 105 M-1 s-1 and
8.54 105 M-1 s-1 , respectively. The kinetic constants of neutral and anionic IM with hydroxyl radicals are 6.71109 M-1 s-1 and 5.51109 M-1 s-1 , respectively. The ozonation attained a total abatement of IM in a short time, but poor mineralization was achieved. Although the kinetic constants of neutral and anionic IM with hydroxyl radicals are far higher than the kinetic constants of neutral and anionic IM with ozone, ozone rather than hydroxyl radicals plays a major role in the oxidation of IM. The limited mineralization infers the existence of ozonation products. Six major intermediates are
found. The mechanism of IM ozonation was detected by both experimental method and DFT calculation method. Nitrogen atom, C-C double bond and benzene ring are the main reaction sites. The first step is usually electrophilic attack or Criegee cyclo-addition, followed by either oxygen transfer, electron transfer, intramolecular rearrangement or hydrolysis. The degradation of IM finally gave rise to the formation and accumulation of small molecule products such as organic compounds (carboxylic acids) and inorganic ions (chloridion), which have negligible toxicity compared to IM and intermediates.
Novelty Statement A thorough study was conducted on the ozonation of a non-steroidal anti-inflammatory drug indomethacin (IM), on which few researches had been done before. After the degradation of IM ozonation and TOC removal were detected, the kinetic constants of IM with ozone and hydroxyl radicals were then measured. The contribution of ozone and hydroxyl radicals during reaction was detected. Both experiments and DFT calculation were used to determine the primary ozonation products. Final ozonation products and the toxicity were studied. Despite the poor mineralization, the toxicity can be fully degraded, indicating ozonation to be a promising technology to remove IM.
Acknowledgments This study was supported by the National High Technology Research and Development Program of China (2013AA06A305), the Program for Changjiang Scholars and Innovative Research Team in University (IRT1261), and the Collaborative Innovation Center for Regional Environmental Quality.
Reference [1] C.G. Daughton, T.A.Ternes, Pharmaceuticals and personal care products in the environment: Agents of subtle change? Environ. Health Persp. 107 (1999) 907-938. [2] Y. Luo, W. Guo, H.H. Ngo, N. Long Duc, F.I. Hai, J. Zhang, S. Liang and X.C. Wang, A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment, Sci. Total Environ. 473 (2014) 619-641. [3] W.C. Li, Occurrence, sources, and fate of pharmaceuticals in aquatic environment and soil, Environ. Pollut. 187 (2014) 193-201. [4] D. Fatta-Kassinos, S. Meric, A. Nikolaou, Pharmaceutical residues in environmental waters and wastewater: current state of knowledge and future research, Anal. Bioanal. Chem. 339 (2011) 251-275. [5] J.B. Carbajo, A.L. Petre, R. Rosal, Continuous ozonation treatment of ofloxacin: Transformation products, water matrix effect and aquatic toxicity, J. of Hazard. Mater. 292 (2015) 34-43. [6] Y.W. Bai, W. Meng, J. Xu,Y. Zhang, C.S. Guo, Occurrence, distribution and bioaccumulation of antibiotics in the Liao River Basin in China, Environ, Sci-Proc. Imp. 16 (2014) 586-593. [7] G.H. Dai, J. Huang, W.W. Chen, B. Wang, G. Yu, S.B. Deng, Major Pharmaceuticals and Personal Care Products (PPCPs) in Wastewater Treatment Plant and Receiving Water in Beijing, China, and Associated Ecological Risks, B. Environ. Contam. Tox. 92 (2014) 655-661. [8] A.Y.C. Lin, W.N. Lee, X.H. Wang, Ketamine and the metabolite norketamine: Persistence and phototransformation toxicity in hospital wastewater and surface water, Water Res. 53 (2014) 351-360. [9] R. Meffe, I. de Bustamante, Emerging organic contaminants in surface water and groundwater: A first overview of the situation in Italy, Sci. Total Environ. 481 (2014) 280-295. [10] M.M.P. Tsui, H.W. Leung, P.K.S. Lam, M.B. Murphy, Seasonal occurrence, removal efficiencies and preliminary risk assessment of multiple classes of organic UV filters in wastewater treatment plants, Water Res. 53 (2014) 58-67. [11] P.P. Fong, A.T. Ford, The biological effects of antidepressants on the molluscs and crustaceans: a review, Aquat. Toxicol. (Amsterdam, Netherlands) 151 (2014) 4-13. [12] M.L. Hedgespeth, P.A. Nilsson, O. Berglund, Ecological implications of altered fish foraging after exposure to an antidepressant pharmaceutical, Aquat. Toxicol. (Amsterdam, Netherlands) 151 (2014) 84-87. [13] J. Liu, G. Lu, D. Wu, Z. Yan, A multi-biomarker assessment of single and combined effects of norfloxacin and sulfamethoxazole on male goldfish (Carassius auratus), Ecotox. Environ. Safe. 102 (2014) 12-17. [14] S.D. Melvin, M.C. Cameron, C.M. Lanctot, Individual and Mixture Toxicity of Pharmaceuticals Naproxen, Carbamazepine, and Sulfamethoxazole to Australian Striped Marsh Frog Tadpoles (Limnodynastes peronii), J. Toxicol. Env. Heal. A 77 (2014) 337-345.
[15] F. Baquero, Low-level antibacterial resistance: a gateway to clinical resistance, Drug Resist. Updat. 4 (2001) 93-105. [16] K. Drlica, The mutant selection window and antimicrobial resistance, J. Antimicrob. Chemoth. 52 (2003) 11-17. [17] C. Carrasco-Pozo, R.L. Castillo, C. Beltrán, A. Miranda, J. Fuentes, M. Gotteland, Molecular mechanisms of gastrointestinal protection by quercetin against indomethacin-induced damage: role of NF-κB and Nrf2, J. Nutr. Biochem. 27 (2015) 289-298. [18] T.A. Ternes, Occurrence of drugs in German sewage plants and rivers, Water Res. 32 (1998) 3245-3260. [19] J.W. Kim, H.S. Jang, J.G. Kim, H. Ishibashi, M. Hirano, K .Nasu, N. Ichikawa, Y. Takao, R. Shinohara, K. Arizono, Occurrence of Pharmaceutical and Personal Care Products (PPCPs) in Surface Water from Mankyung River, South Korea, J. Health Sci. 55 (2009) 249-258. [20] L. Lishman, S.A. Smyth, K. Sarafin, S. Kleywegt, J. Toito, T. Peart, B. Lee, M. Servos, M. Beland, P. Seto, Occurrence and reductions of pharmaceuticals and personal care products and estrogens by municipal wastewater treatment plants in Ontario, Canada, Sci. Total Environ. 367 (2006) 544-558. [21] R. Rosal, A. Rodriguez, J. Antonio Perdigon-Melon, A. Petre, E. Garcia-Calvo, M. Jose Gomez, A. Aguera, A.R. Fernandez-Alba, Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation, Water Res. 44 (2010) 578-588. [22] Q. Sui, J. Huang, S.B. Deng, G. Yu, Q. Fan, Occurrence and removal of pharmaceuticals, caffeine and DEET in wastewater treatment plants of Beijing, China, Water Res. 44 (2010) 417–426. [23] V. Homem, L. Santos, Degradation and removal methods of antibiotics from aqueous matrices - A review, J. Environ. Manage. 92 (2011) 2304-2347. [24] E.A. Murphy, G.B. Post, B.T. Buckley, R.L. Lippincott, M.G. Robson, Future Challenges to Protecting Public Health from Drinking-Water Contaminants, Annu. Rev. Publ. Health 33 (2012) 209-224. [25] R.L. Oulton, T. Kohn, D.M. Cwiertny, Pharmaceuticals and personal care products in effluent matrices: A survey of transformation and removal during wastewater treatment and implications for wastewater management, J. Environ. Monitor. 12 (2010) 1956-1978. [26] M.F. Rahman, E.K. Yanful, S.Y. Jasim, Endocrine disrupting compounds (EDCs) and pharmaceuticals and personal care products (PPCPs) in the aquatic environment: implications for the drinking water industry and global environmental health, J. Water Health 7 (2009) 224-243. [27] F.J. Beltran, A. Aguinaco, J.F. Garcia-Araya, A. Oropesa, Ozone and photocatalytic processes to remove the antibiotic sulfamethoxazole from water, Water Res. 42 (2008) 3799-3808. [28] R.F. Dantas, M. Canterino, R. Marotta, C. Sansa, S. Esplugas, R. Andreozzi, Bezafibrate removal by means of ozonation: Primary intermediates, kinetics, and toxicity assessment, Water Res. 41 (2007) 2525-2532. [29] S. Esplugas, D.M. Bila, L.G.T. Krause, M. Dezotti, Ozonation and advanced oxidation technologies to remove endocrine disrupting chemicals (EDCs) and pharmaceuticals and personal care products
(PPCPs) in water effluents, J. Hazard Mater. 149 (2007) 631-642. [30] R.F. Dantas, C. Sans, S. Esplugas, Ozonation of Propranolol: Transformation, Biodegradability, and Toxicity Assessment, J. Environ. Eng.-Asce 137 (2011) 754-759. [31] J. Benner, T.A. Ternes, Ozonation of Metoprolol: Elucidation of oxidation pathways and major oxidation products, Environ. Sci. Technol. 43 (2009a) 5472-5480. [32] A. Lajeunesse, M. Blais, B. Barbeau, S. Sauve, C. Gagnon, Ozone oxidation of antidepressants in wastewater -Treatment evaluation and characterization of new by-products by LC-QToFMS, Chem. Cent. J. 7 (2013) 1-11. [33] C. Liu, V. Nanaboina, G. Korshin, Spectroscopic study of the degradation of antibiotics and the generation of representative EfOM oxidation products in ozonated wastewater, Chemosphere 86 (2012) 774-782. [34] C. Prasse, M. Wagner, R. Schulz, T.A. Ternes, Oxidation of the Antiviral Drug Acyclovir and Its Biodegradation Product Carboxy-acyclovir with Ozone: Kinetics and Identification of Oxidation Products, Environ. Sci. Technol. 46 (2012) 2169-2178. [35] R.R. Giri, H. Ozaki, S. Ota, Degradation of common pharmaceuticals and personal care products in mixed solutions by advanced oxidation techniques, Int. J. Environ. Sci. Te. 7 (2010) 251-260. [36] J. Benner, T.A. Ternes, , Ozonation of Propranolol: Formation of Oxidation Products, Environ. Sci. Technol. 43 (2009b) 5086-5093. [37] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta Jr., F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian Inc., Wallingford CT, 2009. [38] T. Saito, S. Nishihara, Y. Kataoka, Y. Nakanishi, Matsui, T., Y. Kitagawa, T. Kawakami, M. Okumura, K. Yamaguchi, Transition state optimization based on approximate spin-projection (AP) method, Chem. Phys. Lett. 483 (2009) 168–171. [39] S. Zhang, G. Yu, J. Chen, B. Wang, J. Huang, S. Deng, 2014, Unveiling formation mechanism of carcinogenic N-nitrosodimethylamine in ozonation of dimethylamine: a density functional theoretical investigation, J. Hazard Mater. 279 (2014) 330–335. [40] J. Tomasi, B. Mennucci, R. Cammi, Quantum mechanical continuum solvation models, Chem. Rev. 105 (2005) 2999–3093. [41] J. Kuang, J. Huang, B. Wang, Q. Cao, S. Deng, G. Yu, Ozonation of trimethoprim in aqueous solution:
Identification of reaction products and their toxicity, Water Res. 47 (2013) 2863-2872. [42] M.C. Dodd, M.O. 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. [43] J. Hoigné, 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. [44] C. von Sonntag, U. von Gunten, Chemistry of Ozone in Water and Wastewater Treatment: From Basic Principles to Applications, International Water Assn, 2012. [45] E.M. Thurman, I. Ferrer, O.J. Pozo, J.V. Sancho, F. Hernandez, The even-electron rule in electrospray mass spectra of pesticides, Rapid Commun. Mass Spectrom. 21 (2007) 3855-3868. [46] K. Levsen, H.M. Schiebel, J.K. Terlouw, K.J. Jobst, M. Elend, A. Preiβ, H. Thiele, A. Ingendoh, Even-electron ions: a systematic study of the neutral species lost in the dissociation of quasi-molecular ions, J. Mass Spectrom. 42 (2007) 1024-1044. [47] A.B. Attygalle, J. Ruzicka, D. Varughese, J.B. Bialecki, S. Jafri, Loss of benzene to generate an enolate anion by a site-specific double-hydrogen transfer during CID fragmentation of o-alkyl ethers of ortho-hydroxybenzoic acids, J. Mass Spectrom. 42 (2007) 1224-1234. [48] G. Xu, T. Huang, J. Zhang, J.K. Huang, T. Carlson, S. Miao, Investigation of collision-induced dissociations involving odd-electron ion formation under positive electrospray ionization conditions using accurate mass, Rapid Commun. Mass Spectrom. 24 (2010) 321-327. [49] F.M. Wendel, L.E. Christian, E.J. Machek, S.E. Duirk, M.J. Plewa, S.D. Richardson, T.A. Ternes, Transformation of iopamidol during chlorination. Environ. Sci. Technol. 48 (2014) 12689-12697. [50] J. Staehelin, 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 (1985) 1206-1213. [51] T.H. Lay, J.W. Bozzelli, J.H. Seinfeld, Aatmospheric photochemical oxidation of benzene: benzene plus OH and the benzene-OH adduct (hydroxyl-2,4-cyclohexadienyl) plus O2, J. Phys. Chem. 100 (1996), 6543-6554. [52] K.S. Tay, N.A. Rahman, M.R. Bin Abas, Characterization of atenolol trasformation products in ozonation by using rapid resolution high-performance liquid chromatography/quadrupole-time-of-flight mass spectrometry, Microchem. J. 99 (2011), 312-326. [53] C. Quispe, F.M. Nachtigall, M.F.R. Fonseca, R.M. Alberici, L. Astudillo, J. Villasenor, M.N. Eberlin, L.S. Santos, Monitoring of beta-Blockers Ozone Degradation via Electrospray Ionization Mass Spectrometry, J. Brazil. Chem. Soc. 22 (2011), 919-928. [54] M. Skoumal, P.L. Cabot, F. Centellas, C. Arias, R.M. Rodriguez , J.A. Garrido, E. Brillas, 2006, Mineralization of paracetamol by ozonation catalyzed with Fe2+, Cu2+ and UVA light, Appl. Catal. B-Environ. 66 (2006) 228-240. [55] M.C. Valsania, F. Fasano, S.D. Richardson, M. Vincenti, Investigation of the degradation of cresols in the treatments with ozone, Water Res. 46 (2012) 2795-2804. [56] M.N. Abellan, W. Gebhardt, H.F. Schroder, Detection and identification of degradation products of sulfamethoxazole by means of LC/MS and -MSn after ozone treatment, Water Sci. Technol. 58 (2008)
1803-1812. [57] R. Rosal, M.S. Gonzalo, K. Boltes, P. Leton, J.J. Vaquero, E. Garcia-Calvo, Identification of intermediates and assessment of ecotoxicity in the oxidation products generated during the ozonation of clofibric acid, J. Hazard Mater. 172 (2009) 1061-1068. [58] M. D. M. Gomez-Ramos, M. Mezcua, A. Agueera, A.R. Fernandez-Alba, S. Gonzalo, A. Rodriguez, R. Rosal, Chemical and toxicological evolution of the antibiotic sulfamethoxazole under ozone treatment in water solution, J. Hazard Mater. 192 (2011) 18-25. [59] A. Magdeburg, D. Stalter, J. Oehlmann, Whole effluent toxicity assessment at a wastewater treatment plant upgraded with a full-scale post-ozonation using aquatic key species, Chemosphere 88 (2012) 1008-1014.
(a)
(b) Figure 1. Degreadation curve of IM (a) and TOC (b) under different ozone doses
(a)
(b) Figure 2. Kinetic constants of IM with ozone (a) and hydroxyl radicals (b) under different dissociation degree of IM
Figure 3. Structures of IM and ozonation products
Figure 4. Formation mechanism of OP333
Figure 5. Formation mechanism of OP391
(a)
(c) Figure 6. Formation mechanism of OP277-1 (a), OP277-1’ (b), OP277-2 (c), OP277-3 (d)
(b)
(d)
(a)
(b)
(c)
(d)
(e)
(f)
Figure 7. Time profile of OP156 (a), OP277-1 (b), OP277-2 (c), OP277-3 (d), OP333 (e) and OP 391 (f) under the influent gaseous ozone concentration of 2 mg L-1
23
Figure 8. Time profile of carboxylic acid and chloride under the influent gaseous ozone concentration of 2 mg L-1
24
Figure 9. Toxicity changes during ozonation under different ozone doses
25