Degradation kinetics and mechanism of oxytetracycline by hydroxyl radical-based advanced oxidation processes

Degradation kinetics and mechanism of oxytetracycline by hydroxyl radical-based advanced oxidation processes

Accepted Manuscript Degradation Kinetics and Mechanism of Oxytetracycline by Hydroxyl Radicalbased Advanced Oxidation Processes Yiqing Liu, Xuexiang H...

606KB Sizes 7 Downloads 78 Views

Accepted Manuscript Degradation Kinetics and Mechanism of Oxytetracycline by Hydroxyl Radicalbased Advanced Oxidation Processes Yiqing Liu, Xuexiang He, Yongsheng Fu, Dionysios D. Dionysiou PII: DOI: Reference:

S1385-8947(15)01291-7 http://dx.doi.org/10.1016/j.cej.2015.09.034 CEJ 14190

To appear in:

Chemical Engineering Journal

Received Date: Revised Date: Accepted Date:

30 June 2015 9 September 2015 10 September 2015

Please cite this article as: Y. Liu, X. He, Y. Fu, D.D. Dionysiou, Degradation Kinetics and Mechanism of Oxytetracycline by Hydroxyl Radical-based Advanced Oxidation Processes, Chemical Engineering Journal (2015), doi: http://dx.doi.org/10.1016/j.cej.2015.09.034

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.

Degradation Kinetics and Mechanism of Oxytetracycline by Hydroxyl Radical-based Advanced Oxidation Processes Yiqing Liua,b, Xuexiang Heb, Yongsheng Fua, Dionysios D. Dionysioub,*

a

Faculty of Geosciences and Environmental Engineering, Southwest Jiaotong University, Chengdu 611756, China

b

Environmental Engineering and Science Program, University of Cincinnati, Cincinnati, Ohio 45221-0012, United States *

Corresponding author’s email: [email protected] Tel: +1-513-556-0724; Fax: +1-513-556-4162

1

Abstract In this study, the degradation kinetics and transformation mechanism of oxytetracycline (OTC) by UV-254 nm and UV-254 nm/H2O2 were investigated. The removal of OTC increased with increasing initial H2O2 dosage while excess H2O2 acted as an inhibitor of HO•. The observed UV fluence based pseudo first-order rate constant of OTC (kobs) decreased while the degradation rate at the initial 13 min increased with increasing initial OTC concentrations. Presence of different water constituents in the reaction system had a different degree of influence on the degradation efficiency of OTC. Though after 10 h irradiation, there was only a limited elimination of total organic carbon (TOC), successful transformation of OTC was demonstrated by the detection of thirty-one degradation byproducts in the UV/H2O2 system. Potential degradation mechanisms for OTC were proposed exhibiting five main degradation pathways, including hydroxylation (+16 Da), secondary alcohol oxidation (−2 Da), demethylation (−14 Da), decarbonylation (−28 Da) and dehydration (−18 Da). This study indicates that UV-254 nm/H2O2 is an effective technology for the removal of OTC from an aquatic environment.

Keywords Oxytetracycline; Advanced oxidation processes (AOPs); Hydroxyl radical; UV-254 nm/H2O2; Mechanism.

2

1. Introduction Antibiotics, one of the largest groups of pharmaceuticals and personal care products (PPCPs), have received increasing attention and scientific interest in recent years [1-6]. Oxytetracycline (OTC), for example, is an important broad spectrum antibiotic and has been used in livestock for the disease prevention and growth promotion worldwide [5,6]. In the United States, it is one of the ten approved antibiotic growth promoters [7]. Most OTC is excreted without undergoing metabolism leading to its frequent detection in surface water, sewage water, groundwater, drinking water, seawater and sediment throughout the world [1,4,5,8,9]. Up to 0.34 µg L−1 of OTC has been detected in surface water in the United States [10], while 20-800 mg L−1 of OTC has been reported in the effluent from the wastewater treatment plant (WWTP) of one of the biggest OTC producers in China [11]. OTC occurrence in natural environment can affect the selection of genetic variants of microorganisms resulting in the development of drug resistant bacteria or pathogens, which may subsequently induce a potential risk to the ecosystem and human health [2,4,5,12,13]. It is thus important to remove such a compound from contaminated water. Conventional water treatment methods, such as biodegradation and chlorination, cannot remove OTC efficiently due to its bio-resistant property and chemical stability (Fig. 1) [14,15]. Therefore, it is necessary to develop more efficient and environmentally friendly technologies for the removal of this microcontaminant. Advanced oxidation processes (AOPs) via the generation of highly reactive free radicals, such as hydroxyl radical (HO•) and sulfate radical (SO4•−), have been considered as a promising alternative for the destruction of recalcitrant organic pollutants [16-19]. Hydroxyl radical-based 3

AOPs (HR-AOPs) are a traditional and highly efficient process, because HO• is a non-selective oxidant ( E 0 = 2.72 V , depending on the experimental conditions [18]) having high second-order rate constants with organic compounds in the range of 108-1010 M−1 s−1 [20]. A number of technologies have been applied to generate HO• including Fenton process, photo-Fenton process, UV-254 nm/H2O2 and photocatalysis. Due to the high HO• quantum yield from the direct dissociation of H2O2 ( Φ = 1.0 , i.e., 1.0 mole of HO• being generated with the absorption of 1.0 mole UV photons at 254 nm), as shown in Eq. (1) [21], and a potential complete mineralization of contaminants, UV-254 nm/H2O2 has been widely studied and used for the degradation of various organic contaminants [22-25]. The feasibility of UV/H2O2 in removing OTC has been investigated by several researchers [24-27]. However, the systematic degradation of OTC by UV/H2O2 has not been evaluated so far, including the effects of oxidant dosage, contaminant concentration, common water quality parameters and mineralization. H 2 O 2 /HO −2 + hv → 2 HO •

(1)

Φ = 1. 0

In addition to kinetics, information on the transformation mechanism of OTC by UV/H2O2 is also very limited. In the study by Yuan et al. [27], UV direct photolysis resulted in the detection of three main byproducts as identified by LC-MS/MS while UV/H2O2 led to six main byproducts as identified by GC-MS. Four main byproducts were reported by Jeong et al. [28] from γ–irradiation of OTC in water where HO• and e −aq acted as the reactive species. Reaction byproducts and pathways were also studied by photocatalysis using TiO2 under simulated solar irradiation [29]. Nevertheless, the number of reported byproducts was very limited and the reaction pathways were still unclear, especially for the degradation of OTC by UV/H2O2. 4

The objective of this work was to systematically investigate the photochemical degradation of OTC by UV/H2O2. In our previous study [30], the influence of pH and bicarbonate/carbonate on the removal of OTC was extensively evaluated. In this work, the role of HO• and the effect of H2O2 dosage, OTC concentration and background water matrix such as inorganic anions, metal cations, and natural organic matter (NOM) were studied. The consumption of H2O2 and mineralization efficiency in terms of total organic carbon (TOC) elimination were also investigated. Finally, the degradation mechanism was proposed based on the identified reaction byproducts.

2. Materials and methods 2.1. Materials Oxytetracycline hydrochloride (> 95%) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Hydrogen peroxide (50%, w/w) was purchased from Fisher Scientific (Pittsburgh, PA, USA). Suwannee River humic acid standard Ⅰ (HA) and fulvic acid standard Ⅱ (FA) were obtained from International Humic Substances Society (IHSS, University of Minnesota, St. Paul, MN, USA) and used as representatives of NOM. Other chemicals, such as sodium chloride (NaCl), sodium sulfate (Na2SO4), sodium nitrate (NaNO3), calcium nitrate tetrahydrate (Ca(NO3)2 · 4H2O), magnesium nitrate hexahydrate (Mg(NO3)2 · 6H2O), Ferric nitrate nonahydrate (Fe(NO3)3 · 9H2O), and cupric sulfate pentahydrate (CuSO4 · 5H2O) were all ACS grade and used as received. Milli-Q water (Milli-pore Corp., Billerica, MA) was used to prepare all aqueous solutions. 5

2.2. Analysis The quantification of OTC was carried out by a high performance liquid chromatography (HPLC, Agilent 1100 Series). The detailed measurement information can be found elsewhere [30]. An Agilent 6540 ultra-high definition accurate-mass quadrupole time-of-flight tandem mass spectrometer (QTOF/MS) coupled with an Agilent 1290 infinity HPLC system (LC-QTOF/MS) was used for the detection and identification of degradation byproducts. An Agilent ZORBAX Eclipse XDB-C18 narrow-bore rapid resolution column (2.1 × 50 mm, 3.5 µm) was utilized as a stationary phase. The mobile phase consisted of A: 0.1% formic acid in H2O and B: 0.1% formic acid in acetonitrile, with a gradient elution of 5% B linearly increased to 90% in the initial 5 min, kept for 1 min, and back to 5% B in the next 0.1 min. The flow rate was 0.2 mL min−1, with the sample injection volume of 5 µL and column temperature of 30 oC. The mass spectrum (m/z 30 to 500) was analyzed in positive ion mode by electrospray ionization (ESI) with the drying gas temperature of 350 oC, drying gas flow of 7 L min−1 and collision energy of 25 eV. Data were analyzed through Agilent MassHunter B.04.00 workstation software. TOC was measured by a Shimadzu VCSH-ASI TOC Analyzer. The concentration of H2O2 was analyzed by a spectrophotometric method using a Hewlett Packard UV-vis spectrophotometer [31].

2.3. Photochemical experiments 6

A bench scale photochemical apparatus equipped with two 15 W low-pressure Hg UV lamps with monochromatic emission at λ max = 254 nm was utilized for photochemical experiments. The average UV fluence rate through the reaction volume was determined to be 0.1 mW cm−2 by three different methods [32]. For OTC degradation experiments, the initial concentrations of OTC, H2O2 and phosphate buffer were 10 µM, 0.5 mM and 5 mM, respectively, unless stated otherwise. At each UV fluence interval, 0.15 mL of reaction solution was sampled and mixed with 0.1 mL 0.025 N Na2S2O3 prior to HPLC analysis. For the mineralization and degradation mechanism study, a higher initial OTC concentration of 40 µM was used. At given time, the irradiated solution was sampled and analyzed immediately without adding any quenching agent. All the above experiments were conducted in triplicate except for mechanism study. The error bars represent the standard error of the mean in the figures.

3. Results and discussion 3.1. Role of HO• on the degradation of OTC in UV/H2O2 system In our previous study, we found that the degradation rate of OTC was much higher by UV/H2O2 than by direct UV photolysis which could probably be attributed to the generation of highly reactive HO• [30]. In order to confirm the role of HO•, three different alcohols, i.e., tert-butanol

(t-BuOH,

k HO • /t − BuOH = 6.0 × 10 8 M −1 s −1

[20]),

methanol

(MeOH,

k HO • /MeOH = 9.7 × 10 8 M −1 s −1 [20]), and isopropanol (i-PrOH, k HO • /i -PrOH = 1.9 × 10 9 M −1 s −1

[20]), were added into the reaction solutions. These alcohols are well known HO• scavengers. Although their second-order rate constants with HO• are lower than that of OTC 7

( k HO • /OTC = 7.18 × 10 9 M −1 s −1 at pH 7 [30]), their concentrations were 1000 times higher (10 mM vs 10 µM), which could result in a complete scavenging of HO•. As shown in Fig. 2, the degradation of OTC by UV/H2O2/alcohols was almost the same as that under UV only condition. The observed UV fluence based pseudo first-order rate constants of OTC (kobs) in UV and UV/H2O2 processes were found to be (0.54 ± 0.02) × 10−3 and (4.85 ± 0.12) × 10−3 cm2 mJ−1, respectively. The contribution of UV direct photolysis and HO• oxidation in UV/H2O2 process was calculated as 11.1% and 88.9%, respectively [25], indicating the significant contribution of HO• in the degradation of OTC in UV/H2O2 system.

3.2. Effect of initial H2O2 concentration Oxidant dosage is an important parameter in evaluating the applicability of UV/H2O2 process. Fig. 3 describes the effect of initial H2O2 concentration on kobs. For the concentrations below 0.5 mM, there was a linear increase of kobs with the increase in H2O2 concentration, which was consistent with the direct dissociation of H2O2 in generating HO• as shown in Eq. (1). Though not proportional, further increase in kobs was also observed at higher initial H2O2 concentrations, which could probably result from the scavenging effect of the excess H2O2, specifically the competitive radical reactions (Eqs. (2)-(4) [20]). Similar results have been observed by Lopez-Penalver et al. [26]. In their study, the tetracycline degradation rate constant increased proportionally with the H2O2 concentration in the range of 0.02-0.2 mM and this behavior disappeared at the H2O2 concentration of 2 mM.

8

HO • + H 2 O 2 → HO 2 + H 2 O

k = 2.7 × 10 7 M −1 s −1

(2)

HO • + HO 2 → O 2 + H 2 O

k = 6.6 × 10 9 M −1 s −1

(3)

HO • + HO • → H 2 O 2

k = 5.5 × 10 9 M −1 s −1

(4)





3.3. Effect of initial OTC concentration

Fig. 4 exhibits the changes in kobs and degradation rate of OTC under different initial OTC concentrations while keeping the initial H2O2 concentration at 0.5 mM. The kobs decreased from (1.31 ± 0.06) × 10−2 cm2 mJ−1 to (1.86 ± 0.04) × 10−3 cm2 mJ−1 when the OTC concentration increased from 2 µM to 30 µM. Possible explanations for this result include: (1) the much higher molar absorptivity of OTC (11550 M−1 cm−1 at pH 7.0 [30]) than H2O2 (19.6 M−1 cm−1) would lead to the significant increase in optical density of the solution with the increase in initial OTC concentration, which could not only affect the direct photolysis efficiency of OTC but also decrease the light absorption by H2O2 and generate subsequently fewer HO• to attack OTC [32,33]; (2) a variety of byproducts, as shown in Table S1 in supporting information (SI), were formed during the degradation of OTC and their concentrations were expected to increase with increasing initial OTC concentration, which could result in the competition for HO• between OTC and the byproducts [34]. On the contrary, the degradation rate of OTC, calculated by the change of OTC concentration in the initial 13 min, increased with the increasing OTC concentration, as shown in Fig. 4. This could be elucidated by the elevated number of OTC molecules exposing to the reactive radicals at higher initial concentration, leading to a higher degradation rate [18]. 9

3.4. Effect of background water matrix

Inorganic anions (e.g., Cl−, SO42−, NO3−, CO32−, and HCO3−), metal cations (e.g., Ca2+, Mg2+, Fe3+, and Cu2+) and NOM are commonly present in natural water. For the potential practical application of UV/H2O2 AOP in the treatment of OTC in natural water or wastewater, it is critical to understand their effect on the destruction of OTC. 3.4.1. Effect of inorganic anion

The effect of CO32− and HCO3− on the degradation of OTC by UV/H2O2 has been extensively investigated previously [30]. This work focused specifically on the influence of other common anions, i.e., Cl−, SO42− and NO3−. The obtained results are shown in Fig 5a. For Cl−, which can react with HO• to form a transient intermediate (ClOH•−) with a second-order rate constant of (4.3 ± 0.4) × 109 M−1 s−1, as shown in Eq. (5) [35], no inhibition effect was observed. One influencing factor was the rapid regeneration of HO• through the dissociation of the formed ClOH•−, as presented in Eq. (6). Through the protonation reaction, ClOH•− can also be converted to Cl•, which can be subsequently transformed into Cl2•−, as shown in Eqs. (7) and (8). Neta et al. [36] pointed out that the formation of Cl2•− through the oxidation of Cl− with HO• was difficult at pH 7 but much more efficient in acidic solutions. Therefore, Eq. (6) might be the main reaction followed by the formed ClOH•− in current conditions [37]. The presence of SO42− at 1 mM level did not affect the degradation of OTC either, because it is nonreactive with HO• [38]. Although NO3− is a well-known photosensitizer, it showed no apparent influence on the transformation of OTC by UV/H2O2. Partial reason for this observation was the inefficient 10

HO• scavenging effect of NO3− ( k < 5 × 10 5 M −1 s −1 [39]). In addition, the amount of HO• formed by the photolysis of NO3− was very limited due to its low molar absorption coefficient at 254 nm ( ε = 4 M −1 cm −1 ) and low HO• quantum yield ( Φ = 0.09 ) [40], which could barely affect the steady-state concentration of HO• generated by the photolytic decomposition of H2O2 in the system. Cl − + HO • → ClOH •−

k = (4.3 ± 0.4) × 10 9 M −1 s −1

(5)

ClOH •− → Cl − + HO •

k = (6.1 ± 0.8) × 10 9 M −1 s −1

(6)

ClOH •− + H + → Cl • + H 2 O

k = (2.1 ± 0.7) × 1010 M −1 s −1

(7)

Cl • + Cl − → Cl 2

k = 2.1 × 1010 M −1 s −1

(8)

•−

3.4.2. Effect of metal cation

OTC can strongly bind with metal ions, especially Ca2+ and Mg2+, at different sites including the C1-C3 tricarbonyl system, C10-C12 phenolic diketone moiety and C4 dimethylammonium group (Fig. 1) [41]. The formed metal complexes may have different destruction performance in different reaction systems. Under simulated sunlight irradiation, Chen et al. [42] found that at pH 7.3, the photodegradation of OTC was slightly inhibited with the addition of Mg2+ but was enhanced in the presence of Ca2+; while at pH 9.0, both cations inhibited the degradation efficiency. To the best of our knowledge, no previous research has investigated the influence of Ca2+ and Mg2+ on the removal of OTC in the UV/H2O2 system. It was found in this study that the degradation of OTC was hardly affected by these two cations, as shown in Fig 5b. This observation suggested the unchanged sensitivity of OTC-metal 11

complex toward HO• reaction. The effect could probably be different for Fe3+ and Cu2+ which could not only bind to OTC but also act as catalysts for Fenton-like reaction [43,44]. It is well-known that HO• can be generated in UV/Fe3+/H2O2 system through two pathways including: (1) classical thermal Fenton-type chemistry for Fe3+/H2O2 system (Eqs. (9)-(13)) [44-46]; (2) photosensitizing effect of FeOH2+ and/or Fenton reaction caused by Fe2+ formed in situ (Eqs. (14) and (12) ) [45,46]. However, pathway (1) is sensitive to pH and is important in a relatively narrow pH range of 2.5-4.0 [45]; as for pathway (2), FeOH2+ is mainly present between pH 2.5 and 5 [47]. As a result, HO• could barely be generated by photo-Fenton process in current reaction conditions where the pH of solution was 7 (5 mM phosphate buffer) and Fe3+ might exist mainly as Fe(OH)2+, or suspended Fe(OH)3 and FePO4 [47]. Therefore, it was not surprising that the presence of Fe3+ showed little effect on the degradation of OTC (Fig. 5b). In contrast, the degradation of OTC increased slightly in the presence of Cu2+. This observation could be attributed to the formation of HO• by Cu2+–catalyzed Fenton-like system (Eqs. (9)-(13)) [48,49]. Several studies have indicated that Cu2+ can effectively activate H2O2 in a broader pH range, especially at near neutral pH values [49,50]. M n + + H 2 O 2 ↔ MOOH (n −1) + + H + (M : Fe or Cu; for Fe, n = 3; for Cu, n = 2)

(9)

MOOH (n −1) + → M (n −1) + + HO 2

(10)



M n + + HO 2 → M (n −1) + + O 2 + H +

(11)

M (n−1) + + H 2 O 2 → M n + + HO • + HO −

(12)

HO • + M (n −1) + → M n + + HO −

(13)

FeOH 2+ + hv → Fe 2+ + HO •

(14)



12

3.4.3. Effect of NOM

NOM can compete with target contaminant and H2O2 for the incident UV light, which may influence the direct UV photolysis of OTC as well as the formation of HO• generated through the homolysis of H2O2 with the activation under UV-254 nm irradiation [51,52]. In addition, NOM, which has a high second-order rate constant with HO• ( k = 2.23 × 10 8 L (mol C) −1 s −1 ) [53], can also act as a major HO• scavenger in the reaction system, as shown in Eq. (15). In order to investigate the effect of NOM (quantified by dissolved organic carbon, DOC) on the degradation of OTC, 3 mg DOC L−1 of HA or FA was added in the reaction solutions. As expected in Fig. 5c, the degradation of OTC was inhibited significantly in the presence of NOM. HO • + NOM → Products

k = 2.23 × 10 8 L (mol C) −1 s −1

(15)

3.5. Mineralization of OTC and consumption of H2O2

As shown in Fig 6, although the parent compound was strongly degraded, the removal of TOC was very slow. At 10 h UV irradiation, no residual OTC was detected while only around 9.5% TOC was removed. This result was different from those obtained by Yuan et al. [27] or by Lopez-Penalver et al. [26]. Yuan et al. found that only 2.91% and 11.61% TOC was reduced for OTC at UV fluence of 7632 and 11448 mJ cm−2 in UV/H2O2 process, respectively; while Lopez-Penalver et al. reported that 27.8% TOC was removed at 60 min UV-254 nm irradiation. These differences were probably caused by the difference in reaction conditions. Their studies 13

also indicated that a good TOC removal could be achieved by increasing the UV fluence and/or the oxidant dosage [19]. The consumption of H2O2 during the destruction of OTC was not significant (Fig. 6). There was only around 20% H2O2 decomposition in 10 h. Sufficient oxidant could induce further degradation of byproducts, which was important for a complete mineralization.

3.6. Identification of reaction byproducts and degradation pathways

A relatively high initial OTC concentration of 40 µM was used for the identification of reaction byproducts. Results from the degradation kinetics of OTC by UV only and UV/H2O2 are presented in Fig. S1 in SI, indicating again the importance of HO• for the effective removal of OTC. Thirty-one degradation byproducts in UV/H2O2 and five byproducts in UV only were detected. The mass to charge ratio (m/z), retention time (RT) and formula of these byproducts are shown in Table S1 in SI. The highlighted byproducts were further identified by mass spectrometry and their evolutions are displayed in Figs. 7a and b. In this case, due to the lack of standards, the volume obtained from the MassHunter data analysis software was used directly as a reference. As shown in Table S1 and Figs. 7a and b, more byproducts were detected in UV/H2O2 than in UV only system, which could be attributed to the reaction with HO•. Generally, as an electrophilic agent, HO• is considered to react mainly through hydrogen abstraction, hydroxyl addition to double bond and electron transfer [17,54]. In case of OTC, aromatic ring (ring D) could be attacked by HO• through hydroxyl addition. Secondary alcohol and tertiary amine moieties could also react with HO• through hydrogen abstraction due to the 14

increase in the electron density of the α-carbon to the oxygen of the hydroxyl group and the nitrogen of the amino group, respectively. The keto/enol sites at C1-C3 and C11-C12 were sensitive toward HO• reaction through hydroxyl addition to double bond [55], though C1-C3, due to the electron withdrawing effect of the adjacent amide, was probably relatively less reactive. Therefore, in addition to the mass spectrum analysis of parent compound OTC and certain byproducts shown in Figs. S2-S7 in SI, the discussed theoretical principles of photochemistry and radical chemistry were applied in the structural assignment of the reaction byproducts. A potential OTC degradation mechanism was thus proposed showing five different degradation pathways including hydroxylation (+16 Da), secondary alcohol oxidation (−2 Da), demethylation (−14 Da), decarbonylation (−28 Da) and dehydration (−18 Da), as shown in Scheme 1. (1) Hydroxylation, represented by pathway (1) in Scheme 1, was an important reaction process. As stated above, OTC might be hydroxylated by HO• through addition to unsaturated carbon or hydrogen abstraction on the saturated α-carbon of hydroxyl group or amino group. The second-order rate constants of HO• with phenol, enolic acetylacetone, i-PrOH and protonated trimethylamine were reported to be 1.4 × 1010, 9.9 × 109, 1.9 × 109 and 4 × 108 M−1 s−1, respectively [20,55-57]. According to the group contribution theory [58], aromatic ring and keto/enol at C11-C12 were thus the most likely target sites to be attacked by HO•. Hence, the detected monohydroxylation byproducts m/z 477 (isomer-1 and isomer-2) in this study might be formed by hydroxylation at these two sites, which could be supported by their mass spectra (Figs. S3 and S4 in SI). Kamel et al. [59] and Vartanian et al. [41] both reported that fragment 15

ions m/z 58, 98, 126 and 154 for OTC all originated from intact ring A. These fragment ions, presented in the mass spectra of OTC and these two m/z 477 isomers, indicated again the hydroxylation not likely to occur in the C1-C3 keto/enol. The detailed mechanisms for hydroxyl addition to aromatic ring and keto/enol are shown in Scheme S1 in SI. In the studies by Jeong et al. [28] and Pereira et al. [29], only one monohydroxylation byproduct, i.e., m/z 477 (isomer-1), was reported. The dihydroxylation byproducts, i.e., m/z 493 (isomer-1 and isomer-2), and multiple hydroxylated byproducts m/z 509 and 525 could also be formed through the same mechanism in UV/H2O2, as presented in Scheme 1. In UV only reaction system, the same two monohydroxylation byproducts were also detected (Fig. 7b and Table S1). Under UV irradiation of OTC, molecular oxygen could get an electron from the excited OTC forming a superoxide radical anion which could break down to yield H2O2, as shown in Eqs. (16)-(18) [60]. Therefore, hydroxylation byproducts could also be generated under UV activation of OTC. OTC + O 2 + hv → OTC •+ + O 2 O2

•−

+ H + → HO 2

(16)

•−

(17)



(18)

2 HO 2 → H 2 O 2 + O 2 •

(2) Secondary alcohol oxidation, illustrated by pathway (2) in Scheme 1, resulted from hydrogen abstraction by HO• at C5 leading to the formation of a carbon center radical intermediate, as shown in Scheme S2 in SI [61]. The formed m/z 459 could be further hydroxylated to m/z 475 and 491 (Scheme 1). (3) Demethylation (pathway (3) in Scheme 1) involved the removal of one methyl group of dimethylammonium group at C4 which was confirmed by the mass spectrum of demethylation 16

byproduct m/z 447 (Fig. S5 in SI) where the fragmentation from fragment ions m/z 412 to 381 by losing a methylamine was different from that of parent compound OTC (Fig. S2) from m/z 426 to 381 through a loss of dimethylamine. Yuan et al. [27] also detected the demethylation byproduct in the degradation of OTC by UV only. The relative abundance of m/z 447 in UV/H2O2 was higher than that in UV only (Figs. 7a and b) indicating that HO• could be more effective in removing the methyl group. Khan et al. [17] also reported that the presence of HO•/SO4•− could increase the removal of lateral alkyl groups of atrazine. The detailed mechanistic steps for the demethylation of OTC by HO• are presented in Scheme S3 in SI [17]. The m/z 463 and 479 could then be formed by pathway (1) from the m/z 447 in UV/H2O2 only, as shown in Scheme 1. (4) Decarbonylation, shown by pathway (4) in Scheme 1, referred to a loss of CO from the ring structure of OTC. Choudhury et al. [62] found that CO could be generated through the rapid dissociation of •COCH3 formed from the C−C bond cleavage of enolic acetylacetone in the thermal decomposition of acetylacetone, as shown in Eqs. (19) and (20). The activation energy for this reaction was reported to be 74.6 kcal mol−1 [62]. The high energy UV irradiation could probably also result in such a bond cleavage, as has been demonstrated by the photodissociation of enolic acetylacetone at 266, 248 and 193 nm [63]. Similarly, UV-254 nm irradiation with energy of 112 kcal mol−1 had a potential to break the keto/enolic bond. There are two enolic acetylacetone moieties in the OTC structure (Fig. 1), one in ring A (C12a-C4), and the other in ring B-C (C10a-C12a). The latter could undergo the photoenolization reaction by γ-hydrogen abstraction due to the presence of phenolic group at C11 [64-66]. The formed 17

enol tautomer was unstable and could rapidly reketonize to the initial compound at room temperature [64,65]. In addition, Wagner et al. [67] reported that the rate constants for the Norrish typeⅠcleavage (α-cleavage) of tert-alkyl phenyl ketones are only 1/100 as large as for the aliphatic ketones. Decarbonylation regularly accompanies α-cleavage [67]. This reaction in this study might thus occur more likely in the first enolic acetylacetone moiety in ring A. Due to the lower energy of C1-C12a (sp3 hybridization) than C1-C2 (sp2 hybridization), the excited OTC could be more likely to undergo α-cleavage at C12a−C1 bond generating a diradical intermediate. Through the loss of CO, the formed diradical could produce the other diradical compound which could generate decarbonylation byproduct m/z 433 by ring closure, as shown in Scheme S4 in SI [68,69]. The absence of fragment ions m/z 98, 126 and 154 for m/z 433 further supported this proposed pathway (Fig. S6 in SI). The produced m/z 433 could further form byproducts m/z 449, 465, 481 and 497 by hydroxylation and m/z 431 (isomer-1) by secondary alcohol oxidation (Scheme 1). Only m/z 433 was found in UV only system demonstrating again the extremely limited HO• reaction in this system (Fig. 7b and Table S1). CH 3COCH = C(OH)CH 3 → • COCH 3 + • CH = C(OH)CH 3 •

(19) (20)

COCH 3 → • CH 3 + CO

Though less probable, bond cleavage at the C1-C2 (sp2 hybridization) could also occur due to the high excitation energy from UV photons. This was supported by the detection of m/z 495 (+34 Da, +2OH) and its hydroxylation byproducts. In the process of decarbonylation, the formed diradical intermediate might also be attacked by oxygen giving rise to the byproduct m/z 451 that could also generate its hydroxylation byproducts through pathway (1) (Schemes 1 and 18

S4). (5) Dehydration (pathway (5) in Scheme 1) engaged in the removal of H2O from the OTC ring structure. In the study by Choudhury et al. [62], the formation of H2O through a four-center molecular channel was found in the thermal pyrolysis of enolic acetylacetone, as shown in Eq. (21). The activation energy of this reaction was 60.8 kcal mol−1 [62]. Such dehydration would not likely occur due to the instability of cyclic cumulene. Alternatively, with the tautomerization of the C11-C12 keto/enol, the dehydration at C6-C5a could lead to a stable aromatic ring at C. Hasan et al. [70] also found that 5a,6-anhydrotetracycline was a major product in the photolysis of tetracycline using 3500–Å lamp in the presence of β– mercaptoethanol. Though dehydration byproduct m/z 443 was not detected in this study, m/z 415, formed probably from m/z 443 by decarbonylation and/or m/z 433 by dehydration (Scheme 1), was found to be the most significant byproduct in both UV only and UV/H2O2 systems in terms of obtained signal volume (Figs. 7a and b). The mass spectrum of m/z 415 is presented in Fig. S7, suggesting again the occurrence of decarbonylation in ring A according to the absence of fragment ions m/z 58, 98, 126 and 154. The formation of m/z 431 (isomer-2) and m/z 413 could then be achieved from the m/z 415 through pathways (1) and (2), respectively (Scheme 1). The formed m/z 413 could be further oxidized to m/z 429 and 445 by HO• (Scheme 1). CH 3 COCH = C(OH)CH 3 → H 2 O + CH 3 COCH = C = CH 2 (50%) + CH 3 COC ≡ CCH 3 (50%) (21)

4. Conclusions 19

This study investigated systematically the degradation kinetics and transformation mechanism of OTC by UV and UV/H2O2. It was found that although OTC could be degraded under direct UV irradiation, HO• contributed much more significantly in UV/H2O2. The increased initial H2O2 concentration improved the observed UV fluence based pseudo first-order rate constant (kobs) while the excess H2O2 could act as a HO• scavenger. The kobs decreased while the initial degradation rate increased with increasing initial OTC concentration. Presence of common inorganic anions (i.e., Cl−, SO42− and NO3−) and metal cations (i.e., Ca2+, Mg2+ and Fe3+) did not affect the OTC degradation under current reaction conditions; while Cu2+ improved slightly the destruction of OTC. On the contrary, the presence of NOM significantly inhibited the removal of OTC. Mineralization of OTC could be obtained with a limited elimination of TOC. Consumption of H2O2 was not significant during the degradation of OTC. Degradation mechanism was further evaluated revealing five different reaction pathways including hydroxylation (+16 Da), secondary alcohol oxidation (−2 Da), demethylation (−14 Da), decarbonylation (−28 Da) and dehydration (−18 Da) based on the detected thirty-one degradation byproducts by LC-QTOF/MS in the UV/H2O2 system. This study suggests that UV-254 nm/H2O2 AOP is capable of removing antibiotics such as OTC from the contaminated water and provides valuable information on the potential application of this process for the removal of OTC in various aquatic compartments.

Acknowledgments

20

This work was funded by the Cyprus Research Promotion Foundation through Desmi 2009-2010 which is co-funded by the Republic of Cyprus and the European Regional Development Fund of the EU under contract number NEA IPODOMI/STRATH/0308/09. Yiqing Liu is also thankful to China Scholarship Council (CSC) Scholarship (201307000035) for providing financial support.

Appendix A. Supporting information

Supporting information related to this article can be found at.

21

References

[1] K. Kummerer, Antibiotics in the aquatic environment–a review–PartⅠ, Chemosphere 75 (2009) 417-434. [2] K. Kummerer, Antibiotics in the aquatic environment–a review–PartⅡ, Chemosphere 75 (2009) 435-441. [3] S.K. Khetan, T.J. Collins, Human pharmaceuticals in the aquatic environment: a challenge to green chemistry, Chem. Rev. 107 (2007) 2319-2364. [4] R. Hirsch, T. Ternes, K. Haberer, K.-L. Kratz, Occurrence of antibiotics in the aquatic environment, Sci. Total Environ. 225 (1999) 109-118. [5] A.K. Sarmah, M.T. Meyer, A.B.A. Boxall, A global perspective on the use, sales, exposure pathways, occurrence, fate and effects of veterinary antibiotics (VAs) in the environment, Chemosphere 65 (2006) 725-759. [6] N. Kemper, Veterinary antibiotics in the aquatic and terrestrial environment, Ecol. Indic. 8 (2008) 1-13. [7] S. Yang, K. Calson, Evolution of antibiotic occurrence in a river through pristine, urban and agricultural landscapes, Water Res. 37 (2003) 4645-4656. [8] S. Mompelat, B. Le Bot, O. Thomas, Occurrence and fate of pharmaceutical products and by-products, from resource to drinking water, Environ. Int. 35 (2009) 803-814. [9] I. Michael, L. Rizzo, C.S. McArdell, 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. [10] D.W. Kolpin, E.T. Furlong, M.T. Meyer, E.M. Thurman, S.D. Zaugg, L.B. Barber, H.T. Buxton, Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: a national reconnaissance, Environ. Sci. Technol. 36 (2002) 1202-1211. [11] K. Li, A. Yediler, M. Yang, S. Schulte-Hostede, M.H. Wong, Ozonation of oxytetracycline and toxicological assessment of its oxidation by-products, Chemosphere 72 (2008) 473-478. [12] J.L. Martinez, Environmental pollution by antibiotics and by antibiotic resistance determinants, Environ. Pollut. 157 (2009) 2893-2902. [13] L. Rizzo, C.M. Manaia, C. Merlin, T. Schwartz, C. Dagot, M.C. Ploy, I. Michael, D. Fatta-Kassinos, Urban wastewater treatment plants as hotspots for antibiotic resistant bacteria and genes spread into the environment: a review, Sci. Total Environ. 447 (2013) 345-360. [14] A.J. Watkinson, E.J. Murby, 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. [15] T.A. Ternes, M. Meisenheimer, D. McDowell, F. Sacher, H.-J. Brauch, B. Haist-Gulde, G. Preuss, U. Wilme, N. Zulei-Seibert, Removal of pharmaceuticals during drinking water treatment, Environ. Sci. Technol. 36 (2002) 3855-3863. [16] M. Klavarioti, D. Mantzavinos, D. Kassinos, Removal of residual pharmaceuticals from aqueous systems by advanced oxidation processes, Environ. Int. 35 (2009) 402-417. [17] J.A. Khan, X. He, N.S. Shah, H.M. Khan, E. Hapeshi, D. Fatta-Kassinos, D.D. Dionysiou, 22

Kinetic and mechanism investigation on the photochemical degradation of atrazine with activated H2O2, S2O82− and HSO5−, Chem. Eng. J. 252 (2014) 393-403. [18] N.S. Shah, X. He, H.M. Khan, J.A. Khan, K.E. O’Shea, D.L. Boccelli, D.D. Dionysiou, Efficient removal of endosulfan from aqueous solution by UV-C/peroxides: a comparative study, J. Hazard. Mater. 263 (2013) 584-592. [19] X. He, S.P. Mezyk, I. Michael, D. Fatta-Kassinos, D.D. Dionysiou, Degradation Kinetics and mechanism of β-lactam antibiotics by the activation of H2O2 and Na2S2O8 under UV-254 nm irradiation, J. Hazard. Mater. 279 (2014) 375-383. [20] 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) 513-886. [21] J.H. Baxendale, J.A. Wilson, The photolysis of hydrogen peroxide at high light intensities, Trans. Faraday Soc. 53 (1957) 344-356. [22] F. Yuan, C. Hu, X. Hu, J. Qu, M. Yang, Degradation of selected pharmaceuticals in aqueous solution with UV and UV/H2O2, Water Res. 43 (2009) 1766-1774. [23] B.A. Wols, C.H.M. Hofman-Caris, Review of photochemical reaction constants of organic micropollutants required for UV advanced oxidation processes in water, Water Res. 46 (2012) 2815-2827. [24] 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. [25] I. Kim, N. Yamashita, H. Tanaka, Photodegradation of pharmaceuticals and personal care products during UV and UV/H2O2 treatments, Chemosphere 77 (2009) 518-525. [26] J.J. Lopez-Penalver, M. Sanchez-Polo, C.V. Gomez-Pacheco, J. Rivera-Utrilla, Photodegradation of tetracyclines in aqueous solution by using UV and UV/H2O2 oxidation processes, J. Chem. Technol. Biotechnol. 85 (2010) 1325-1333. [27] F. Yuan, C. Hu, X. Hu, D. Wei, Y. Chen, J. Qu, Photodegradation and toxicity changes of antibiotics in UV and UV/H2O2 process, J. Hazard. Mater. 185 (2011) 1256-1263. [28] J. Jeong, W. Song, W.J. Cooper, J. Jung, J. Greaves, Degradation of tetracycline antibiotics: mechanisms and kinetic studies for advanced oxidation/reduction processes, Chemosphere 78 (2010) 533-540. [29] J.H.O.S. Pereira, V.J.P. Vilar, M.T. Borges, O. Gonzalez, S. Esplugas, R.A.R. Boaventura, Photocatalytic degradation of oxytetracycline using TiO2 under natural and simulated solar radiation, Sol. Energy 85 (2011) 2732-2740. [30] Y. Liu, X. He, X. Duan, Y. Fu, D.D. Dionysiou, Photochemical degradation of oxytetracycline: influence of pH and role of carbonate radical, Chem. Eng. J. 276 (2015) 113-121. [31] A.O. Allen, C.J. Hochanadel, J.A. Ghormley, T.W. Davis, Decomposition of water and aqueous solutions under mixed fast neutron and gamma radiation, J. Phys. Chem. 56 (1952) 575-586. [32] X. He, M. Pelaez, J.A. Westrick, K.E. O’Shea, A. Hiskia, T. Triantis, T. Kaloudis, M.I. 23

Stefan, A.A. de la Cruz, D.D. Dionysiou, Efficient removal of microcystin-LR by UV-C/H2O2 in synthetic and natural water samples, Water Res. 46 (2012) 1501-1510. [33] A. Aleboyeh, Y. Moussa, H. Aleboyeh, The effect of operational parameters on UV/H2O2 decolourisation of acid blue 74, Dyes Pigments 66 (2005) 129-134. [34] H. Ghodbane, O. Hamdaoui, Decolorization of antraquinonic dye, C.I. Acid blue 25, in aqueous solution by direct UV irradiation, UV/H2O2 and UV/Fe (Ⅱ) processes, Chem. Eng. J. 160 (2010) 226-231. [35] G.G. Jayson, B.J. Parsons, Some simple, highly reactive, inorganic chlorine derivatives in aqueous solution, J. Chem. Soc. Faraday Trans. 1. 69 (1973) 1597-1607. [36] P. Neta, R.E. Huie, A.B. Ross, Rate constants for reactions of inorganic radicals in aqueous solution, J. Phys. Chem. Ref. Data 17 (1988) 1027-1284. [37] C.-H. Liao, S.-F. Kang, F.-A. Wu, Hydroxyl radical scavenging role of chloride and bicarbonate ions in the H2O2/UV process, Chemosphere 44 (2001) 1193-1200. [38] N. Gao, Y. Deng, D. Zhao, Ametryn degradation in the ultraviolet (UV) irradiation/hydrogen peroxide (H2O2) treatment, J. Hazard. Mater. 164 (2009) 640-645. [39] P. Neta, R.E. Huie, Rate constants for reactions of nitrogen oxide (NO3) radicals in aqueous solutions, J. Phys. Chem. 90 (1986) 4644-4648. [40] G. Mark, H.-G. Korth, H.-P. Schuchmann, C. von Sonntag, The photochemistry of aqueous nitrate ion revisited, J. Photochem. Photobiol. A. 101 (1996) 89-103. [41] V.H. Vartanian, B. Goolsby, J.S. Brodbelt, Identification of tetracycline antibiotics by electrospray ionization in a quadrupole ion trap, J. Am. Soc. Mass Spectrom. 9 (1998) 1089-1098. [42] Y. Chen, H. Li, Z. Wang, T. Tao, C. Hu, Photoproducts of tetracycline and oxytetracycline involving self-sensitized oxidation in aqueous solutions: effects of Ca2+ and Mg2+, J. Environ. Sci. 23 (2011) 1634-1639. [43] J.W. Moffett, R.G. Zika, Reaction kinetics of hydrogen peroxide with copper and iron in seawater, Environ. Sci. Technol. 21 (1987) 804-810. [44] C. Walling, Fenton’s reagent revisited, Acc. Chem. Res. 8 (1975) 125-131. [45] J.J. Plgnatello, Dark and photoassisted Fe3+–catalyzed degradation of chlorophenoxy herbicides by hydrogen peroxide, Environ. Sci. Technol. 26 (1992) 944-951. [46] Y. Sun, J.J. Plgnatello, Photochemical reactions involved in the total mineralization of 2,4-D by Fe3+/H2O2/UV, Environ. Sci. Technol. 27 (1993) 304-310. [47] B.C. Faust, J. Hoigne, Photolysis of Fe (Ⅲ)-hydroxy complexes as sources of OH radicals in clouds, fog and rain, Atmos. Environ. 24A (1990) 79-89. [48] J. Soler, A. Garcia-Ripoll, N. Hayek, P. Miro, R. Vicente, A. Arques, A.M. Amat, Effect of inorganic ions on the solar detoxification of water polluted with pesticides, Water Res. 43 (2009) 4441-4450. [49] D.A. Nichela, A.M. Berkovic, M.R. Costante, M.P. Juliarena, F.S. Garcia Einschlag, Nitrobenzene degradation in Fenton-like systems using Cu ( Ⅱ ) as catalyst. Comparison between Cu (Ⅱ)- and Fe (Ⅲ)-based systems, Chem. Eng. J. 228 (2013) 1148-1157. [50] F.J. Millero, V.K. Sharma, B. Kam, The rate of reduction of copper (Ⅱ) with hydrogen 24

peroxide in seawater, Mar. Chem. 36 (1991) 71-83. [51] C.-H. Liao, M.D. Gurol, Chemical oxidation by photolytic decomposition of hydrogen peroxide, Environ. Sci. Technol. 29 (1995) 3007-3014. [52] J.A. Khan, X. He, H.M. Khan, N.S. Shah, D.D. Dionysiou, Oxidative degradation of atrazine in aqueous solution by UV/H2O2/Fe2+, UV/S2O82−/Fe2+ and UV/HSO5−/Fe2+ processes: a comparative study, Chem. Eng. J. 218 (2013) 376-383. [53] P. Westerhoff, S.P. Mezyk, W.J. Cooper, D. Minakata, Electron pulse radiolysis determination of hydroxyl radical rate constants with Suwannee River fulvic acid and other dissolved organic matter isolates, Environ. Sci. Technol. 41 (2007) 4640-4646. [54] 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. [55] R.K. Broszkiewicz, T. Soeylemez, D. Schulte-Frohlinde, Reactions of OH radicals with acetylacetone in aqueous solution, Z. Naturforsch, 37b (1982) 368-375. [56] E.J. Land, M. Ebert, Pulse radiolysis studies of aqueous phenol. Water elimination from dihydroxycyclohexadienyl radicals to form phenoxyl, Trans. Faraday Soc. 63 (1967) 1181-1190. [57] M. Simic, P. Neta, E. Hayon, Pulse radiolytic investigation of aliphatic amines in aqueous solution, Int. J. Radiat. Phys. Chem. 3 (1971) 309-320. [58] D. Minakata, K. Li, P. Westerhoff, J. Crittenden, Development of a group contribution method to predict aqueous phase hydroxyl radical (HO•) reaction rate constants, Environ. Sci. Technol. 43 (2009) 6220-6227. [59] A.M. Kamel, H.G. Fouda, P.R. Brown, B. Munson, Mass spectral characterization of tetracyclines by electrospray ionization, H/D exchange, and multiple stage mass spectrometry, J. Am. Soc. Mass Spectrom. 13 (2002) 543-557. [60] X. He, A.A. de la Cruz, A. Hiskia, T. Kaloudis, K.E. O’Shea, D.D. Dionysiou, Destruction of microcystins (cyanotoxins) by UV-254 nm-based direct photolysis and advanced oxidation processes (AOPs): influence of variable amino acids on the degradation kinetics and reaction mechanisms, Water Res. 74 (2015) 227-238. [61] X. He, G. Zhang, A.A. de la Cruz, K.E. O’Shea, D.D. Dionysiou, Degradation mechanism of cyanobacterial toxin cylindrospermopsin by hydroxyl radicals in homogeneous UV/H2O2 process, Environ. Sci. Technol. 48 (2014) 4495-4504. [62] T.K. Choudhury, M.C. Lin, Homogeneous pyrolysis of acetylacetone at high temperatures in shock waves, Int. J. Chem. Kinet. 20 (1990) 491-504. [63] H.P. Upadhyaya, A. Kumar, P.D. Naik, Photodissociation dynamics of enolic-acetylacetone at 266, 248, and 193 nm, J. Chem. Phys. 118 (2003) 2590-2598. [64] L.A. Baker, M.D. Horbury, S.E. Greenough, P.M. Coulter, T.N.V. Karsili, G.M. Roberts, A.J. Orr-Ewing, M.N.R. Ashfold, V.G. Stavros, Probing the ultrafast energy dissipation mechanism of the sunscreen oxybenzone after UVA irradiation, J. Phys. Chem. Lett. 6 (2015) 1363-1368. [65] R. Haag, J. Wirz, P.J. Wagner, The photoenolization of 2-methylacetophenone and related 25

compounds, Helv. Chim. Acta 60 (1977) 2595-2607. [66] P.J. Wagner, C.-P. Chen, A rotation-controlled excited-state reaction. The photoenolization of ortho alkyl phenyl ketones, J. Am. Chem. Soc. 98 (1976) 239-241. [67] P.J. Wagner, Chemistry of excited triplet organic carbonyl compounds, Top. Curr. Chem. 66 (1976) 1-52. [68] S.W. Benson, G.B. Kistiakowsky, The photochemical decomposition of cyclic ketones, J. Am. Chem. Soc. 64 (1942) 80-86. [69] F.E. Blacet, A. Miller, The photochemical decomposition of cyclohexanone, cyclopentanone and cyclobutanone, J. Am. Chem. Soc. 79 (1957) 4327-4329. [70] T. Hasan, M. Allen, B.S. Cooperman, Anhydrotetracycline is a major product of tetracycline photolysis, J. Org. Chem. 50 (1985) 1755-1757.

26

Captions Figures: Figure 1: Structure of OTC. Figure 2: Degradation of OTC at different conditions. [OTC]0 = 10 µM, [H2O2]0 = 0.5 mM,

[t-BuOH]0 = [MeOH]0 = [i-PrOH]0 = 10 mM, no phosphate buffer added. Figure 3: Effect of initial H2O2 concentration on kobs in UV/H2O2 system. [OTC]0 = 10 µM, 5

mM phosphate buffer at pH 7.0. Note: The equation only applies for the first four data points. Figure 4: Effect of initial OTC concentration on kobs and degradation rate (calculated by the

change of OTC concentration in the initial 13 min) in UV/H2O2 system. [H2O2]0 = 0.5 mM, 5 mM phosphate buffer at pH 7.0. Figure 5: Effect of inorganic anion (a), metal cation (b), and NOM (c) on the degradation of

OTC in UV/H2O2 system. [OTC]0 = 10 µM, [H2O2]0 = 0.5 mM, [Cl−]0 = 3 mM, [SO42−]0 = [NO3−]0 = [Ca2+]0 = [Mg2+]0 = 1 mM, [Fe3+]0 = [Cu2+]0 = 1 µM, [HA]0 = [FA]0 = 3 mg DOC L−1, 5 mM phosphate buffer at pH 7.0. Figure 6: Mineralization of OTC in terms of TOC elimination and consumption of H2O2 in

UV/H2O2 system. [OTC]0 = 40 µM, [H2O2]0 = 0.5 mM, no phosphate buffer added. Figure 7: Evolution of major reaction byproducts by UV/H2O2 (a) and UV only (b). [OTC]0 =

40 µM, [H2O2]0 = 0.5 mM, no phosphate buffer added.

Schemes: Scheme 1: Proposed degradation mechanisms of OTC in UV/H2O2 system: (1) hydroxylation, 27

(2) secondary alcohol oxidation, (3) demethylation, (4) decarbonylation and (5) dehydration. The byproduct inside the brackets was not detected in this study. [OTC]0 = 40 µM, [H2O2]0 = 0.5 mM, no phosphate buffer added.

28

Figure 1 H3C HO

8 9

7

D 10

6a 10a

CH3 6

C 11

5a 11a

CH3

OH

N

5

4

B 12

4a

A

12a

OH 3 2

NH 2

1

OH

OH

O

OH

29

O

O

Figure 2

1.0

0.8

C/C 0

0.6

0.4 UV only UV/H2O2 + 10 mM t-BuOH 0.2

UV/H2O2 + 10 mM MeOH UV/H2O2 + 10 mM i-PrOH UV/H2O2

0.0

0

40

80

120

160

200

UV Fluence (mJ cm-2)

30

240

280

320

Figure 3

kobs ( x 10-3 cm2 mJ-1)

10

8

6

4 y = 7.550 E-3 x + 1.038 E-3 2 R = 0.999

2

0

0.0

0.5

1.0

Concentration of H2O2 (mM)

31

1.5

Figure 4

14

kobs ( x 10-3 cm2 mJ-1)

0.30

10 0.25 8 0.20 6 0.15 4 0.10

2

0.05

0

0

5

10

15

20

Concentration of OTC (µM)

32

25

30

-1

12

Degradation rate (µM min )

0.35

Figure 5 (a)

1.0

0.8

C/C 0

0.6

0.4 -

UV/H2O2 + 3 mM Cl

0.2

UV/H2O2 + 1 mM SO42UV/H2O2 + 1 mM NO3

0.0

-

control

0

40

80

120

160

200

UV Fluence (mJ cm-2)

33

240

280

320

Figure 5 (b)

1.0

0.8

C/C 0

0.6

0.4 UV/H2O2 + 1 mM Ca2+ 2+

0.2

UV/H2O2 + 1 mM Mg

3+

UV/H2O2 + 1 µM Fe control

0.0

UV/H2O2 + 1 µM Cu2+ 0

40

80

120

160

200

UV Fluence (mJ cm-2)

34

240

280

320

Figure 5 (c)

1.0

0.8

C/C 0

0.6

0.4

0.2

-1

UV/H2O2 + 3 mg DOC L HA UV/H2O2 + 3 mg DOC L-1 FA

0.0

control

0

40

80

120

160

200

UV Fluence (mJ cm-2)

35

240

280

320

Figure 6

1.0

0.8

C/C0

0.6 TOC change H2O2 change

0.4

OTC change

0.2

0.0

0

2

4

6

Time (h)

36

8

10

Figure 7 (a)

m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z

8e+6

Volume

6e+6

4e+6

2e+6

511 493-1: 493-2: 477-1: 477-2: 459 451 449 447 433 431-1: 431-2: 415

0

0

1

2

3

4

5

Time (h)

37

6

7

8

9

0.97 1.50 1.75 2.23

1.98 2.44

Figure 7 (b)

1.2e+6

m/z m/z m/z m/z m/z

1.0e+6

Volume

8.0e+5

477-1: 1.75 477-2: 2.23 447 433 415

6.0e+5 4.0e+5 2.0e+5 0.0

0

1

2

3

Time (h)

38

4

5

6

Scheme 1 H3C OH

CH3

HO

CH3

N

OH N H2 OH OH

H3C N

CH3 O

HO

O

OH

CH3

HO

HO

CH3

N

OH

O

O

OH

O

OH

O

HO

NH2

CH3

HO

HO

OH

OH

OH

O

OH

OH

O

(1)

O

HO

OH

O

HO

O

HO

OH

CH 3

HO

CH3

HO

HO

CH 3

OH

HN OH NH2 OH

OH O

O

OH

O

O

OH

O

O

m/z 46 3: C21H22N 2 O10

(1) CH3 CH3

N

HO

CH 3

HO

OH

HN OH

OH

NH2

NH2 OH

O

O

HO

O

OH

m/z 49 1: C22H22N 2O11

O

CH3 N

(1)

OH O

O

H3C

NH2 OH

OH

(1)

m/z 49 3-2: C22H24N2O 11

CH3 OH

OH

O

m /z 447: C21H22N2O 9

OH

OH

O

m/z 493-1: C22H 24N 2O11

N

OH

O

NH2

(1)

O

O

OH

OH

H3C CH 3

CH 3

HO

(1)

HO

O

N H2 O

N H2

H3C CH3

N

HO

OH

HN

OH

OH

(1)

OH

m/z 47 5: C 22H22N 2O10

O

OH

OH

OH

OH

OH

m /z 477-2: C22H 24N 2O10

H3C

CH3

CH3

HO

N H2

O

(1)

H3C N

CH3 OH

m/z 477-1: C22 H 24N 2O10

(2)

CH 3 O

N

N H2

m/z 459: C22H22N2O 9

HO

C H3

OH

OH

OH

(1)

CH3

HO

N H2 O

(3) demethylation -14 Da (-CH 2)

(1) H3C

H3C OH

OH OH

O

(1) hydroxylation +16 Da (+O)

CH3 OH

O

m/z 461: C 22H24 N2O 9

(2) secondary alcohol oxidation -2 Da (-2H)

O

O

OH

OH O

OH

O

(1) H3C HO

CH 3

OH

N

CH3 O

HO

NH2 OH OH

O

OH O

O

m/z 525: C 22H24N 2O13

39

O

O

m/z 47 9: C21H22N 2O11

m/z 50 9: C22H24N 2 O12

HO

O

OH

OH

O

O

Scheme 1 (continued) H3C OH

CH3

HO

CH3

N

OH N H2 OH OH

H3C CH3

OH

O

HO

OH

CH3

N

O

O

OH

O

(5)

CH3

N

(2)

CH3

HO

HO

OH

CH3

N

CH3

HO

HO

OH

OH

O

H3C

OH

CH3

HO

HO

OH

OH

m/z 431-1: C21H 22N 2O8

O

O

CH3

N

OH

CH3 HO

OH N H2

CH3

HO

OH

OH

O

OH

HO

O

O

H3C N

OH

O

HO

O

OH

OH

OH

O

CH3

HO

OH

OH

N

CH 3

NH2

H3C CH3

O

N

HO

CH3 OH N H2

OH

OH OH

O

O

O

O

m/z 445: C 21H20N2O 9

40

OH

CH3

O

O

OH O

OH O

NH2 OH

O

m/z 54 3: C22H26N 2O14

m/ z 497: C21H 24N 2O12

(1)

N

OH OH

O

OH

O

OH

HO

O

O

OH

m/z 429: C 21H20N2O 8

CH 3

HO

OH

N H2

O

O

OH

HO OH

O

H3C

CH3

N

OH

N H2 OH

OH

N H2

HO

HO

O

OH

CH3

N

OH OH

O

NH2 OH

(1)

H 3C

H3C

(1)

m /z 413: C21H20N2O 7

OH

m/z 499: C21 H26N2O 12

O

CH3

m/z 52 7: C22H26N 2O13

(1)

CH3 OH

O

OH

O

CH3

N

OH

OH

HO

(2) O

CH3

HO

HO

m/ z 481: C 21H 24N 2O11

(2)

OH

OH

OH

H3C

NH2 OH

OH O

OH

OH

O

m/z 431-2: C21H 22N 2O8

CH3

O

CH 3

N

O

OH

(1)

OH

OH OH

CH 3

HO

N H2

(1)

HO

HO

m/z 483: C 21H26N2O 11

H 3C

O

H3C

CH3

N

OH

O

m/z 465: C21 H24N2O 10

(1)

OH

OH

OH

OH OH

O

(1)

N H2

O

H3C

OH

NH2 OH

m/z 51 1: C22H26N 2O12

OH

CH3

HO

HO

OH

OH O

O

H3C

CH3

N

N H2 OH

OH

(1)

H3C

CH3

N

CH3

N

OH

OH

OH

(1) CH3 O

OH

OH

m/z 467: C 21H26 N2O 10

m/z 449: C 21H24 N2O 9

m/z 415: C21H22N2O 7

CH 3

HO

HO

N H2

O

O

H3C

CH3

N

OH

OH

O

OH

OH O

OH

(1)

N H2 OH

O

NH2 OH

m/z 495: C 22H26N 2O11

OH

O

HO

OH

O

H3C

OH

OH O

OH

OH

(1)

N H2 OH

O

(1)

OH

OH

OH OH

m/z 451: C 21H26N2O 9

H3C

H3C OH

OH

O

m /z 433: C21H24N2O 8

(4)

CH3

N

N H2 OH

OH

OH

CH 3

HO

N H2 OH

m/z 443: C22 H22N2O 8

CH3

CH3

N

OH

N H2 O

H3C

OH

CH3

HO

OH

OH OH

ring cleavage H3C

H3C CH3

OH

OH

O

(4) decarbonyl ation -28 Da (-CO)

CH3

N

OH

O

m /z 461: C22H24N2O 9

(5) dehydration -18 Da (-H2O)

O

UV-254 nm

UV photolysis

H2O2

Reaction pathways: 1. Hydroxylation: +16 Da

HO•

H3 C

HO

CH3

OH

CH3

2. Secondary alcohol oxidation: −2 Da

N

(Only in UV/H2O2)

OH

3. Demethylation: −14 Da NH 2

4. Decarbonylation: −28 Da

OH OH

O

OH

O

5. Dehydration: −18 Da

O

Oxytetracycline (MW 460, C22H24N2O9)

41

Highlights  HO• played a much more significant role than UV in degrading OTC by UV/H2O2.  Ca2+ and Mg2+ had no effects on OTC degradation though they can bind with OTC.  The presence of Cu2+ slightly increased the OTC removal in UV/H2O2 process.  Limited TOC elimination was obtained during the mineralization studies of OTC.  Five reaction pathways were proposed based on identified byproducts by LC-QTOF/MS.

42