English

• Antibiotics, as a class of emerging contaminants, have become an environmental topic of acute concern due to their pseudo-persistence and ecological risk in aquatic systems [1-3]. As necessities, antibiotics are widely used in human treatments, animal husbandry, and aquaculture [4-6]. Then, they are mainly released from hospitals and households into wastewater treatment plants (WWTPs) [7,8]. China has attached great importance to eliminate the emerging contaminants. However, most antibiotics cannot be incompletely removed by conventional wastewater treatment processes and are therefore frequently discharged into the aquatic environment [9,10]. Their aqueous ubiquity in surface waters has been reported in many countries, including European countries, the United States, and China [11-13]. Even trace amounts of antibiotics persisting in the aquatic environment can have long-term adverse effects [14]. Antibiotic pollutants are resistant to biodegradation in the environment, but their presence in water systems induces the generation of antibiotic resistant bacteria and poses a threat to ecosystem and human health [15]. Therefore, it is necessary to control such pollutants in wastewater to prevent them from entering the aquatic environment.

  Conventional wastewater treatment processes including adsorption, sedimentation, and biodegradation cannot remove antibiotics effectively [16-18]. Ultraviolet (UV) based advanced oxidation processes (AOPs) have been reported to effectively degrade antibiotics [6,19,20]. UV/persulfate (UV/PS) and UV/chlorine have received increasing attention in the degradation of antibiotics in wastewater because of their strong oxidation ability, high efficiency, and limited secondary pollution [21-23]. Many previous studies have investigated influencing factors of the UV/PS and UV/chlorine processes [24-27], and the pH values of the reaction system were found to significantly affect the degradation efficiency of certain pollutants [28,29]. For instance, although the degradation of sulfachloropyridazine in the UV/PS process at different pH conditions followed pseudo-first-order reaction kinetics, the fastest reaction rate appeared at pH 5 [30]. Yang et al. [31] observed the fastest degradation of ciprofloxacin (CIP) at pH 7 and pH 5, respectively, in the UV/PS and UV/chlorine processes, and deduced that the phenomena might be attributed to the diverse reactivities of CIP toward reactive species (RS) at different pHs. As the molecular structures contain ionizable groups (e.g., −COOH and −NH_n), most antibiotics are ionizable and will undergo acid-base dissociation, exhibiting different dissociated species depending on the pH of the water [32-34]. These dissociated species have been proven to have unique physicochemical properties [34]. As for the UV/PS and UV/chlorine processes, previous studies have found that pH has a significant effect on the degradation of ionizable antibiotics by SO₄^•− and Cl^•, but the specific mechanisms involving the degradation kinetics and transformation products are not well elucidated [23,27,31]. Therefore, it is necessary to clarify the degradation behavior and related mechanisms of antibiotics in different dissociated forms in the two processes.

  This study provides a detailed examination of SO₄^•−/Cl^•-mediated degradation of sulfathiazole (ST), sarafloxacin (SAR), and lomefloxacin (LOM) in UV/PS and UV/chlorine systems. ST, SAR, and LOM (purity > 98%) were purchased from J & K Scientific Ltd. (Beijing, China). Their chemical structures and molecular weights are shown in Table S1 (Supporting information). The experiments were carried out using a merry-go-round photochemical reactor with a 500 W mercury lamp (Fig. S1 in Supporting information). In order to avoid direct photolysis of the target compounds, 400 nm cut-off filters were used to adjust the emission spectra of the light source (λ > 400 nm). The light intensity at the solutions was 1.83 mW/cm² at 420 nm. In the competition kinetic experiments, Na₂S₂O₈ (40 µmol/L) and NaClO (40 µmol/L) were used to generate SO₄^•− and Cl^•, respectively. The initial concentrations of the antibiotics were set at 5 µmol/L based on the better ratio of the model compounds to the RS [35,36]. The bimolecular reaction rate constants k_RS,S (k_SO4^•−,S and k_Cl^•,S) for the three model substances (S) with SO₄^•− and Cl^• were calculated by Eqs. 1−4 [33,37],

  (1)

  (2)

  (3)

  (4)

  where k_RS,R denotes the bimolecular reaction rate constants of the references (R), k_{SO4^•−,BA} = 1.2 × 10⁹ L mol⁻¹ s⁻¹, k_Cl^•,BA = 1.8 × 10¹⁰ L mol⁻¹ s⁻¹, k_•OH,BA = 5.9 × 10⁹ L mol⁻¹ s⁻¹, k_•OH,NB = 3.9 × 10⁹ L mol⁻¹ s⁻¹; [^•OH]_SS and [Cl^•]_SS represent the steady-state concentrations of ^•OH and Cl^•, respectively. Vibrio fischeri was selected to examine the 15-min acute toxicities of the samples during degradation processes according to the international standard method (ISO11348–3–2007). The experimental information is detailed in the supporting information.

  In all control experiments without light irradiation or RS photosensitizers, no significant degradation of ST, SAR, and LOM was observed (P > 0.1), suggesting that their pyrolysis and hydrolysis were negligible. As shown in Fig. S2 (Supporting information), these model compounds had no light absorption at λ > 400 nm, and thus did not undergo direct photolysis ( < 2%). When exposed to light irradiation (λ > 400 nm), all three compounds disappeared rapidly in the competition kinetic experiments, indicating that they were effectively degraded by SO₄^•−/Cl^•. It can be seen from the degradation curves in Fig. 1 that the apparent reactions of the three individual substances with the RS were conformed to follow pseudo-first-order kinetics (R² > 0.95), depending on the pH. However, the degradation of the substances by the RS (SO₄^•−/Cl^•) was essentially second-order bimolecular reactions. Thus, the calculated values for the bimolecular reaction rate constants (k_SO4^•−,S and k_Cl^•,S) are shown in Table S2 (Supporting information). The k_SO4^•−,S ranged from (1.80 ± 0.05) × 10⁹ L mol⁻¹ s⁻¹ for ST (pH 8) to (8.73 ± 0.33) × 10¹⁰ L mol⁻¹ s⁻¹ for LOM (pH 10), while the k_Cl^•,S ranged from (3.44 ± 0.23) × 10⁹ L mol⁻¹ s⁻¹ for SAR (pH 10) to (3.31 ± 0.002) × 10¹¹ L mol⁻¹ s⁻¹ for LOM (pH 8).

  Figure 1

  

  Figure 1.  Kinetic profiles for the reactions of sulfathiazole (ST), sarafloxacin (SAR) and lomefloxacin (LOM) with SO₄^•−/Cl^• under different pH conditions.

  DownLoad: Full-Size Img PowerPoint

  In comparison with the k_^•OH,S values of ST [33], SAR [38], and LOM [39], which were all in the same order of magnitude (10⁹), the k_SO4^•−,S values of ST, SAR, and LOM differed by orders of magnitude (10⁹−10¹⁰). Similarly, their k_Cl^•,S values were not in the same order of magnitude. For example, Lei et al. [40] determined the k_Cl^•,S values of 88 trace organic contaminants ranged from 3.10 × 10⁹ L mol⁻¹ s⁻¹ to 4.08 × 10¹⁰ L mol⁻¹ s⁻¹. The significant differences in k_SO4^•−,S and k_Cl^•,S values among different classes of reactants could be attributed to the higher selectivity of SO₄^•− and Cl^• [28]. SO₄^•− could oxidatively degrade alkanes, alcohols, organic acids, ethers, esters, and most organic compounds containing unsaturated bonds, whereas Cl^• was more sensitive to substances containing electron-rich groups such as olefins, phenols, and anilines [40,41].

  As shown in Fig. 2a, both k_SO4^•−,S and k_Cl^•,S of each compound were dependent on pH (P < 0.05). As for ST, k_SO4^•−,S and k_Cl^•,S were slightly higher at pH 2 than at pH 5 and 8. However, k_SO4^•−,S was maximum at pH 5 for SAR reacting with SO₄^•−, while LOM was the most reactive toward SO₄^•− at pH 10. In addition, both the k_Cl^•,S values of SAR and LOM were the greatest when reacting with Cl^• at pH 8. The pH dependence of the reactivities in UV/PS and UV/chlorine reactions was caused by Coulomb’s force and chemical structures. The coulombic repulsion between reactants under alkaline conditions leads to lower reactivities [42]. Meanwhile, the molecular structures and deprotonation degrees might have impacts on the pH dependence [33,43], indicating that the RS oxidative reactivities of various dissociated forms of these ionizable antibiotics need to be further differentiated.

  Figure 2

  

  Figure 2.  The bimolecular reaction rate constants of ST, SAR, and LOM with SO₄^•−/Cl^•. (a) is for different pH conditions, and (b) is for different dissociated forms. S represents the model substances.

  DownLoad: Full-Size Img PowerPoint

  To quantify the reactivities of each protonated form towards SO₄^•− and Cl^•, the bimolecular reaction rate constants (k_SO4^•−,i, k_Cl^•,i) of different dissociated forms (i) were obtained by matrix calculations, and the results are shown in Fig. 2b (detailed data listed in Table S3 in Supporting information). The various dissociated forms of the individual antibiotics exhibited different reactivities (from H₂ST⁺ to ST⁻; H₂FQs⁺ to FQs⁻). For ST, the cationic forms (H₂ST⁺) were more reactive toward SO₄^•− and Cl^•. The cationic SAR (H₂SAR⁺) had the highest reactivity towards SO₄^•−, while the anionic and cationic LOM (LOM⁻ and H₂LOM⁺) were more reactive with SO₄^•−. Interestingly, both SAR and LOM in the neutral forms (HFQs⁰) showed the fastest reactions when reacting with Cl^•. Compared with the SO₄^•− mediated reaction, the reactivities of FQs toward Cl^• varied by 3 orders of magnitude based on the profile (pH 8 or 10).

  As the pH increased, the dissociated forms gradually changed (Fig. S3 in Supporting information), which can explain why the oxidation reactivities were significantly related to pH (Fig. 2). Overall, the cationic forms (H₂ST⁺, H₂SAR⁺, and H₂LOM⁺) were more highly reactive towards SO₄^•− in most cases, while the neutral forms (e.g., HSAR⁰ and HLOM⁰) reacted the fastest with Cl^• for most of the antibiotics tested.

  Furthermore, we identified 31 significant intermediates generated from SO₄^•−/Cl^•-mediated degradation of the three antibiotics in the UV/PS and UV/chlorine processes. The proposed chemical structures and tentative transformation pathways are presented in Figs. 3−5. Table S4 (Supporting information) shows detailed information about these degradation products, including retention times (t_R), molecular weights (M_w), and MS fragment m/z. The total ion chromatograms and MS spectra in positive ionization mode are shown in Figs. S4 and S5 (Supporting information). There are different intermediates and transformation pathways corresponding to the two reactions of the individual antibiotics (Figs. 3−5). As for ST (Fig. 3), the transformation products of the reaction with SO₄^•− were simple, mainly because SO₄^•− was easy to selectively attack electron-rich groups like anilines [41], resulting in the S−N breaking and the formation of P155 and P100. The products were also detected for the photolysis of ST [44]. In contrast, the primary products in the reaction process of ST with Cl^• were abundant, with more reaction pathways involving five-membered heterocyclic ring opening, hydroxylation, and dealkylation. This could be attributed to the diversity of RS (Cl^• and ^•OH) in the UV/chlorine process [45]. The ^•OH preferred to experience multi-site oxidation with phenyl hydroxylation and heterocyclic cleavage [33], while Cl^• was easy to attack single bonds and amino groups [21,40]. Therefore, the five-membered heterocycles in the ST structure were easily cleaved.

  Figure 3

  

  Figure 3.  Primary transformation products and pathways of ST during UV/PS and UV/chlorine processes. The products are labeled "Pn", with n standing for the molecular weights.

  DownLoad: Full-Size Img PowerPoint

  Figure 4

  

  Figure 4.  Products and primary pathways of SAR reacting with SO₄^•−/Cl^•.

  DownLoad: Full-Size Img PowerPoint

  Figure 5

  

  Figure 5.  Transformation products and primary pathways of LOM during UV/PS and UV/chlorine processes.

  DownLoad: Full-Size Img PowerPoint

  For SAR (Fig. 4), the same transformation products, P359 and P383, were observed in both UV/PS and UV/chlorine processes. This is mainly due to the susceptibility of SAR’s piperazine ring and −F group to be attacked by the RS, resulting in ring opening and defluorination reactions [46]. Meanwhile, there were different degradation products (P232, P351, and P381, etc.) generated during the two processes because of the different oxidation capacities and attack sites of SO₄^•− and Cl^•. Based on the transformation products of SAR, then primary reaction pathways were proposed. These included: defluorination; hydroxylation; and piperazinyl cleavage. Importantly, SAR underwent chlorination during the UV/chlorine process, indicating the participation of Cl^• in the reaction. The chlorinated product (P351) is of concern due to its complex biological effects [47-49]. Compared with SAR, the more degradation products (P273, P331, and P349) of LOM (Fig. 5) were similar for UV/PS and UV/chlorine processes, mainly because LOM has more alkane structures and is susceptible to attack by SO₄^•− and Cl^• [41]. Based on these transformation products, the proposed primary pathways of LOM are defluorination, piperazinyl ring opening, hydroxylation, etc.

  The toxicity of the 3 target antibiotics during the SO₄^•−/Cl^• mediated degradation processes was assayed, for which the results are shown in Fig. 6. There were certain similarities and differences in the toxicity changes for the individual antibiotics towards the two RS. For ST, the toxicities of both reaction systems first increased, and then decreased with time, implying the generation of some more toxic intermediates than the parent compound. However, toxicities for the Cl^• mediated reaction solution of ST were more enhanced at the initial period (0−t_1/2) of degradation, which can be attributed to the abundant intermediates that contain the basic parent structure (Fig. 3).

  Figure 6

  

  Figure 6.  Concentration changes and solution luminescence inhibition rates of ST, SAR, and LOM during UV/PS and UV/chlorine processes.

  DownLoad: Full-Size Img PowerPoint

  As for SAR and LOM, their toxicity profiles were diverse between the two reaction systems (Fig. 6). The toxicities observed during the reaction of SAR with SO₄^•− showed an overall decreasing trend with time, while the toxicities were initially reduced and then enhanced during the reaction between SAR and Cl^•. Unlike SAR, the toxicity trends of LOM during the reaction with SO₄^•− and Cl^• were similar, which may be due to the formation of similar transformation products (Fig. 5).

  The observed toxicity profile was related to the formation and accumulation of transformation products, of which some would be more toxic and others less toxic than the parent compounds. The toxicities of these individual intermediates were assessed using ECOSAR v2.0 software, with their 96 h LC₅₀ (fish), 48 h LC₅₀ (daphnia), and 96 h EC₅₀ values (green algae) shown in Table S5 (Supporting information). It can be found that ST, SAR, and LOM all generated intermediates with higher toxicities than the parents during the degradation processes, such as the transformation products P100, P218, and P246 for ST, P301, P351, P357, and P383 for SAR, as well as P275 and P289 for LOM. This suggested that the toxicities of the reaction systems would persist in these initial intermediates, showing good consistence with the results of the vibrio fischeri toxicity experiments. Therefore, the ecological risks of the treated antibiotic wastewaters by the UV/PS and UV/chlorine processes need to be considered before treated-wastewater release into the aquatic environment.

  In summary, this study provides new insights into the SO₄^•−/Cl^•-mediated degradation kinetics, products, toxicities, and related mechanisms of the three representative antibiotics in UV/PS and UV/chlorine processes. The transformation kinetics were found to be dependent on pH (P < 0.05), which was attributable to the disparate reactivities of their individual dissociated forms when reacting with RS (e.g., SO₄^•− and Cl^•). These reactions generated different products, and demonstrated diverse transformation pathways, involving S–N cleavage, hydroxylation, defluorination, and chlorination reactions. This has implications when assessing the degradative fate of these chemicals during the AOPs that are increasingly used in tertiary wastewater treatment processes.

  Furthermore, we found that many primary by-products exhibited similar structures to the parent compounds and revealed the toxicity profiles of ST, SAR, and LOM reaction systems before and after degradation. The toxicity responses were dependent on the major intermediates and primary pathways. Many intermediates were proven to be more toxic than their individual parents, raising the wider issue of extended potency for these compounds with regards to their ecotoxicity. These results provide novel knowledge for understanding the UV-based AOPs to treat wastewater containing antibiotics.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Jinshuai Zheng: Writing – review & editing, Writing – original draft, Methodology, Data curation, Conceptualization. Junfeng Niu: Writing – review & editing. Crispin Halsall: Writing – review & editing. Yadi Guo: Writing – review & editing. Peng Zhang: Writing – review & editing, Supervision. Linke Ge: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

  Acknowledgments

  This work was supported by the Key Research and Development Program of Shaanxi Province (No. 2024SF-YBXM-567), the National Natural Science Foundation of China (Nos. 21976045, 22076112), and the China Scholarship Council (CSC) Scholarship (No. 202308610123).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110202.

  
  
• 1. [1]

     E. Brillas, Sci. Total Environ. 819 (2022) 153102. doi: 10.1016/j.scitotenv.2022.153102

  2. [2]

     K. Brooks, J. Eze, O. Onalenna, T.O. Rahube, J. Hazard. Mater. Adv. 9 (2023) 100232.

  3. [3]

     H. Malik, R. Singh, S. Kaur, et al., J. Infect. Public Health 16 (2023) 172–182. doi: 10.1016/j.jiph.2023.11.002

  4. [4]

     J. Ren, H. Shi, J. Liu, et al., J. Environ. Manage. 335 (2023) 117546. doi: 10.1016/j.jenvman.2023.117546

  5. [5]

     X. Wang, J. Jing, M. Zhou, R. Dewil, Chin. Chem. Lett. 34 (2023) 107621. doi: 10.1016/j.cclet.2022.06.044

  6. [6]

     R. Guo, Y. Chen, B. Liu, et al., Chin. Chem. Lett. 33 (2022) 3809–3817. doi: 10.1016/j.cclet.2021.11.061

  7. [7]

     F. Yan, L. An, X. Xu, W. Du, R. Dai, Sci. Total Environ. 906 (2024) 167737. doi: 10.1016/j.scitotenv.2023.167737

  8. [8]

     C. Zhang, Y. Chen, S. Chen, et al., Ecotoxicol. Environ. Saf. 255 (2023) 114817. doi: 10.1016/j.ecoenv.2023.114817

  9. [9]

     J. Wang, R. Zhuan, L. Chu, Sci. Total Environ. 646 (2019) 1385–1397. doi: 10.1016/j.scitotenv.2018.07.415

  10. [10]

      Y. Pan, Y. Zhang, M. Zhou, J. Cai, Y. Tian, Water Res. 153 (2019) 144–159. doi: 10.1016/j.watres.2018.12.063

  11. [11]

      C.E. Givens, D.W. Kolpin, L.E. Hubbard, et al., Sci. Total Environ. 904 (2023) 166753. doi: 10.1016/j.scitotenv.2023.166753

  12. [12]

      D. Han, Q. Hou, J. Song, et al., Environ. Res. 235 (2023) 116653. doi: 10.1016/j.envres.2023.116653

  13. [13]

      U. Szymańska, M. Wiergowski, I. Sołtyszewski, et al., Microchem. J. 147 (2019) 729–740. doi: 10.1016/j.microc.2019.04.003

  14. [14]

      G. Liu, H. Zheng, J. Lu, Trends Environ, Anal. Chem. 16 (2017) 16–23.

  15. [15]

      N. Guo, Y. Wang, L. Yan, et al., Water Res. 117 (2017) 95–101. doi: 10.1016/j.watres.2017.03.058

  16. [16]

      J. Dutta, A.A. Mala, Water Sci. Technol. 82 (2020) 401–426.

  17. [17]

      A.C. Reis, B.A. Kolvenbach, O.C. Nunes, P.F. Corvini, New Biotechnol 54 (2020) 34–51. doi: 10.1016/j.nbt.2019.08.002

  18. [18]

      D. Zhi, D. Yang, Y. Zheng, et al., J. Environ. Manage. 251 (2019) 109598. doi: 10.1016/j.jenvman.2019.109598

  19. [19]

      Z. Li, J. Wang, J. Chang, B. Fu, H. Wang, Sci. Total Environ. 857 (2022) 159172.

  20. [20]

      J. Liu, H. He, Z. Shen, H.H. Wang, W. Li, J. Hazard. Mater. 429 (2022) 128398. doi: 10.1016/j.jhazmat.2022.128398

  21. [21]

      N. Li, J. Ye, H. Dai, et al., Water Res. 235 (2023) 119926. doi: 10.1016/j.watres.2023.119926

  22. [22]

      Z. Li, L. Wang, Y. Liu, et al., Water Res. 195 (2021) 116973. doi: 10.1016/j.watres.2021.116973

  23. [23]

      Y. Zhang, L. Li, Z. Pan, et al., Chem. Eng. J. 379 (2020) 122354. doi: 10.1016/j.cej.2019.122354

  24. [24]

      E. Bazrafshan, P. Tavassoli, A.A. Zarei, Water Sci. Technol. 2017 (2018) 126–133. doi: 10.2166/wst.2018.096

  25. [25]

      N. Huang, W. Wang, Z. Xu, Q. Wu, H. Hu, J. Environ. Manage. 237 (2019) 180–186. doi: 10.1016/j.jenvman.2019.02.055

  26. [26]

      M.Y. Kilic, W.H. Abdelraheem, X. He, K. Kestioglu, D.D. Dionysiou, J. Hazard. Mater. 367 (2019) 734–742. doi: 10.1016/j.jhazmat.2018.06.062

  27. [27]

      Y. Zhang, J. Zhang, Y. Xiao, V.W.C. Chang, T.T. Lim, Chem. Eng. J. 302 (2016) 526–534. doi: 10.1016/j.cej.2016.05.085

  28. [28]

      M. Xu, J. Deng, A. Cai, et al., Chem. Eng. J. 413 (2021) 127533. doi: 10.1016/j.cej.2020.127533

  29. [29]

      Z. Xu, C. Shan, B. Xie, Y. Liu, B. Pan, Appl. Catal. B 200 (2017) 439–447. doi: 10.1016/j.apcatb.2016.07.023

  30. [30]

      J. Dan, Q. Wang, K. Mu, et al., Environ. Sci. Water Res. Technol. 6 (2020) 2510–2520. doi: 10.1039/D0EW00239A

  31. [31]

      H. Yang, Y. Li, Y. Chen, G. Ye, X. Sun, Water Environ. Res. 91 (2019) 1576–1588. doi: 10.1002/wer.1144

  32. [32]

      L. Ge, Q. Dong, C. Halsall, et al., Environ. Sci. Pollut. Res. 25 (2018) 15726–15732. doi: 10.1007/s11356-018-1765-0

  33. [33]

      L. Ge, P. Zhang, C. Halsall, et al., Water Res. 149 (2019) 243–250. doi: 10.1016/j.watres.2018.11.009

  34. [34]

      X. Wei, J. Chen, Q. Xie, et al., Environ. Sci. Technol. 47 (2013) 4284–4290. doi: 10.1021/es400425b

  35. [35]

      X. Liu, Y. Liu, S. Lu, et al., Chem. Eng. J. 385 (2020) 123987. doi: 10.1016/j.cej.2019.123987

  36. [36]

      Z. Wu, S. Gong, J. Liu, J. Shi, H. Deng, J. Water Process Eng. 58 (2024) 104870. doi: 10.1016/j.jwpe.2024.104870

  37. [37]

      J. Fang, Y. Fu, C. Shang, Environ. Sci. Technol. 48 (2014) 1859–1868. doi: 10.1021/es4036094

  38. [38]

      L. Ge, G. Na, S. Zhang, et al., Sci. Total Environ. 527 (2015) 12–17.

  39. [39]

      X. Luo, X. Wei, J. Chen, et al., Water Res. 166 (2019) 115083. doi: 10.1016/j.watres.2019.115083

  40. [40]

      Y. Lei, S. Cheng, N. Luo, X. Yang, T. An, Environ. Sci. Technol. 53 (2019) 11170–11182. doi: 10.1021/acs.est.9b02462

  41. [41]

      X. Li, J. Shen, H. Cao, et al., Chem. Eng. J. 474 (2023) 146256. doi: 10.1016/j.cej.2023.146256

  42. [42]

      X. Tong, S. Wang, L. Wang, Chemosphere 256 (2020) 126997.

  43. [43]

      T. Mill, Chemosphere 38 (1999) 1379–1390. doi: 10.1016/S0045-6535(98)00540-2

  44. [44]

      A.L. Boreen, W.A. Arnold, K. McNeill, Environ. Sci. Technol. 38 (2004) 3933–3940. doi: 10.1021/es0353053

  45. [45]

      T.K. Kim, T. Kim, Y. Cha, K.D. Zoh, Water Res. 185 (2020) 116159. doi: 10.1016/j.watres.2020.116159

  46. [46]

      N. Li, R. Li, X. Duan, et al., Environ. Sci. Technol. 55 (2021) 16163–16174. doi: 10.1021/acs.est.1c06244

  47. [47]

      S. Kali, M. Khan, M.S. Ghaffar, et al., Environ. Pollut. 281 (2021) 116950. doi: 10.1016/j.envpol.2021.116950

  48. [48]

      B.T. Oba, X. Zheng, M.A. Aborisade, et al., Environ. Pollut. 291 (2021) 118239. doi: 10.1016/j.envpol.2021.118239

  49. [49]

      A. Riu, A. le Maire, M. Grimaldi, et al., Toxicol. Sci. 122 (2011) 372–382. doi: 10.1093/toxsci/kfr132

• 

  Figure 1  Kinetic profiles for the reactions of sulfathiazole (ST), sarafloxacin (SAR) and lomefloxacin (LOM) with SO₄^•−/Cl^• under different pH conditions.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  The bimolecular reaction rate constants of ST, SAR, and LOM with SO₄^•−/Cl^•. (a) is for different pH conditions, and (b) is for different dissociated forms. S represents the model substances.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Primary transformation products and pathways of ST during UV/PS and UV/chlorine processes. The products are labeled "Pn", with n standing for the molecular weights.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Products and primary pathways of SAR reacting with SO₄^•−/Cl^•.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Transformation products and primary pathways of LOM during UV/PS and UV/chlorine processes.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Concentration changes and solution luminescence inhibition rates of ST, SAR, and LOM during UV/PS and UV/chlorine processes.

  下载: 全尺寸图片 幻灯片
•