English

• Fenton-like processes utilizing peroxydisulfate (PDS) are attractive for eliminating antibiotics and disinfecting due to the generation of sulfate (SO₄^•−) and hydroxyl (HO^•) radicals [1-5]. Antibiotics are typically targeted in the free state during oxidation processes. However, recent studies have shown that Fe(Ⅲ) in the Fenton-like system tends to form complex with antibiotics containing ionizable functional groups [6-9]. It was observed that the degradation preferences and the ecotoxic effects of degradation products are completely different for free and bound-state antibiotics in the same system [10-12]. Therefore, it is necessary to carefully consider the complexes in the oxidation processes.

  Natural organic matter (NOM) are commonly found in water bodies. Previous studies have extensively investigated NOM as a reducing ligand in the Fenton system [13-15]. For example, it has been indicated that the formation of complexes between humic acid (HA) as a natural ligand and Fe(Ⅲ) is essential for the activation of H₂O₂ [16,17]. Furthermore, NOM was known as a natural photosensitizer. When exposed to light, NOM enters an excited state and reacts with water and oxygen to generate reactive oxygen species (ROS), including singlet oxygen (¹O₂), hydroxyl radicals (HO^•), and superoxide radicals (^•O₂⁻) [18,19]. However, the role of NOM considered as a photosensitizer in Fenton-like systems has been less studied.

  For this research, levofloxacin (LVF) was chosen as the primary pollutant. In a previous study, we confirmed the production of Fe(Ⅲ)-LVF complexes and calculated the proportion of these complexes using a mathematical model [7]. Nevertheless, the specifics of where Fe(Ⅲ) and LVF bind remain uncertain. Furthermore, the impact of HA on the complexes during the process should be taken into account. In a previous study, we employed nanosecond transient absorption spectroscopy to substantiate the generation of LVF^* in the visible light spectrum [7]. Interestingly, we have observed that the electrons from excited state LVF (LVF^*) can be transferred to PDS to produce more HO^•. Additionally, earlier research has shown that creating binding chemical bonds can efficiently promote electron transfer to Fe(Ⅲ) through ligand-to-metal charge transfer (LMCT) [19-22]. For example, swift electron movement takes place inside the compound created by the reaction between HA-diclofenac (DCF) and KMnO₄, known as HA-DCF-KMnO₄ [23]. It accelerated the elimination of diclofenac through permanganate treatment. Therefore, this study investigates whether the LVF/HA-Fe(Ⅲ)-PDS formed through the interaction of Fe(Ⅲ)-LVF/HA and PDS could have a similar effect and discuss the role of photosensitizers in this effect.

  Based on it, this study constructed the Vis/Fe(Ⅲ)/PDS photo-Fenton-like system and selecting HA as the typical NOM to investigate its effects. The identification of complexes and main functional groups of the ligands was carried out using UV-vis spectroscopy, X-ray photoelectron spectroscopy (XPS), and Fourier transform infrared spectroscopy (FTIR). The impact of HA on the ROS was assessed. The effects of complexation and photosensitization were initially determined through complexation masking and turning off the light, respectively. To further substantiate the interaction between complexation and photosensitization, electrochemical techniques and density functional theory (DFT) simulations were utilized.

  The details of chemicals, analytic methods, and DFT calculation were shown in Texts S1-S5 (Supporting information). The degradation tests were conducted using a photocatalyst equipped with LED lamps (CEL-LB70, Beijing China Education AU-Light Technology Co., Ltd.) that produced white light spanning wavelengths from 400 nm to 800 nm. A magnetic stirrer was positioned at the center of the reaction chamber to facilitate uniform mixing of the solution. In a standard test, LVF (2 mg/L), Fe(Ⅲ) (40 µmol/L), HA (2 mg/L), and PDS (2.5 mmol/L) were successively introduced into the reaction mixture at an initial pH of 7. The reaction was initiated by switching on the LED lamps (17.50 W/m²) pointed towards. Samples of 1.50 mL were extracted at regular intervals and promptly combined with 20 µL of 1 mol/L NaNO₂ to halt the ongoing oxidation reaction.

  Fig. 1a illustrates the degradation behavior of LVF across various systems. In contrast to the Vis/Fe(Ⅲ)/PDS, incorporating HA greatly improved LVF elimination, boosting the degradation rate from 82.63% to almost complete removal. Particularly, the promotion mechanism of HA was closely related to light, since HA showed a poor promotion effect under dark conditions (Fig. 1a). Fe(Ⅲ) and irradiation were also important influencing factors, and the degradation rates in Vis/HA/PDS and Fe(Ⅲ)/HA/PDS were only 27.61% and 81.74%, respectively. It was noticed that the contaminants exhibited a fluctuating decrease in the first 10 min (Fig. 1b). The UV-vis spectra of the reaction solutions, both filtered and unfiltered using a 0.22 µmol/L membrane, were examined. A significant decrease in ΔAbs (ΔAbs = Abs(unfiltered) − Abs(filtered)) was detected with the addition of acid (Fig. S2 in Supporting information), indicating the occurrence of flocculation and adsorption due to the formation of iron hydroxide. Furthermore, stable ΔAbs were observed in the systems where both Fe(Ⅲ) and HA were present, suggesting that new substances might have been produced by Fe(Ⅲ) and HA [24,25]. Furthermore, the influence of experimental parameters, including PDS dosage, HA dosage, Fe(Ⅲ) dosage, light intensity, and initial pH were also investigated. Specific details are shown in Text S6 and Fig. S4 (Supporting information).

  Figure 1

  

  Figure 1.  The degradation of LVF in different system (a) within 30 min and (b) within 10 min. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = 2 mg/L, [PDS] = 2.50 mmol/L, [LVF] = 2 mg/L, pH 7, and light intensity 17.50 W/m².

  DownLoad: Full-Size Img PowerPoint

  Previous studies indicated that complexes might be form between Fe and HA, and also HA and Fe are likely to bind to LVF [7,26]. UV-vis spectroscopy is used to confirm the presence of complexes in the system, and a red-shift or hyperchromatic/hypochromatic effect is observed when complexes are formed [27]. Variations in absorption coefficients and spectral bands between the mixture spectra and the mathematically additive spectra suggest the creation of Fe(Ⅲ)-LVF, Fe(Ⅲ)-HA, and HA-LVF complexes, as illustrated in the Figs. S5-S7 (Supporting information). Additionally, the minor red shift detected in the Fe(Ⅲ)-HA and Fe(Ⅲ)-LVF complexes suggests that electron delocalization is directed towards forming complexes, with Fe(Ⅲ) serving as an electron acceptor [28,29].

  To understand the characteristics of the functional groups participating in the complexation, the FTIR spectra of the complexes were recorded at various Fe(Ⅲ)/LVF, Fe(Ⅲ)/HA, HA/LVF, and Fe(Ⅲ)/HA/LVF ratios. The FTIR absorption wavelengths of different functional groups are listed in Table S2 (Supporting information). As shown in Fig. S8 (Supporting information), When the Fe/LVF mass ratio was increased from 1:4 to 1:24, the transmission intensities at 1711, 1626, and 1057 cm⁻¹ were notably lower compared to the spectra of LVF alone. This suggests that the carboxyl and piperazine rings were involved in complexation [30,31]. To further verify the order of complexation among the functional groups, a 2D-COS analysis of FTIR spectra was performed [10]. Six positive autopeaks were located on the diagonal line of the synchrotron spectrum, indicating that the carboxyl group, piperazine ring, and phenolic hydroxyl group in the LVF are sensitive to Fe mass alterations (Figs. 2a and b) [30,32]. The symbols of the cross-correlation peaks in the spectra are shown in Table S3 (Supporting information). Based on Noda's rule, the relative sensitivities of the LVF functional groups involved in complexation with Fe are summarized as follows: piperazine ring > carboxyl group > phenolic hydroxyl group [33].

  Figure 2

  

  Figure 2.  The 2D-COS spectra from FTIR. Synchronous (a) and asynchronous (b) of Fe(Ⅲ)-LVF. Synchronous (c) and asynchronous (d) of Fe(Ⅲ)-HA. Synchronous (e) and asynchronous (f) of HA-LVF. Synchronous (g) and asynchronous (h) of Fe(Ⅲ)-LVF-HA. A positive correlation is indicated by red, a negative correlation by blue, and the strength of the correlation is indicated by the darkness of the color.

  DownLoad: Full-Size Img PowerPoint

  For Fe(Ⅲ)-HA complexes, the intensities of the carboxylate vibration peaks (1595 and 1398 cm⁻¹) significantly diminished with the progressive addition of Fe(Ⅲ) (Fig. S9 in Supporting information) [30]. Additionally, two positive autopeaks appeared on the diagonal of the synchronization spectrum, further confirming the sensitivity of the carboxylate group to the change in Fe mass (Fig. 2c). The asynchronous spectrum has four negative cross peaks (1595/1278, 1595/1010, 1398/1278, and 1398/1010 cm⁻¹) in Fig. 2d. Based on Noda's method: Fe will react with the hydroxyl group in HA first, followed by the carboxyl group.

  For the HA-LVF complex, the intensity of the carboxylate characteristic peaks (1581 and 1359 cm⁻¹) diminished as the LVF concentration increased. Furthermore, a heightened intensity of the hydroxyl C-O absorption peak at 1047 cm⁻¹ was detected (Fig. S10 in Supporting information) [10]. Due to the fact that LVF does not contain hydroxyl groups, it suggests that hydrolytic hydroxylation occurs during the complexation reaction. The generated hydroxyl groups further undergo dehydration with HA, which is supported by the increased transmission intensity of the keto group C=O (1620 cm⁻¹). Two prominent autopeaks at 1359 and 1581 cm⁻¹ were detected along the diagonal in the synchronous spectra (Figs. 2e and f), suggesting that the carboxyl group is the major binding site.

  The FTIR spectra of the complexes with different Fe(Ⅲ)/HA/LVF ratios are shown in Fig. S11 (Supporting information). Significant changes in transmission intensities was observed at 1626, 1469, 1263, 1049, and 1010 cm⁻¹ as Fe(Ⅲ) addition gradually increased. This indicates that the carbonyl group, piperazine moiety, carboxylic acid group, and hydroxyl group are crucial binding sites for the creation of the HA-Fe(Ⅲ)-LVF ternary complex. Six distinct autopeaks were identified along the diagonal line in the synchrotron spectrum (Fig. 2g), indicating that the keto group, piperazine ring, carboxyl group, and hydroxyl group are sensitive to Fe(Ⅲ) mass alterations. These functional groups serve as crucial binding sites for the creation of binary complexes such as Fe(Ⅲ)-LVF, Fe(Ⅲ)-HA, and HA-LVF. According to the cross-correlation peak symbols in Table S6 (Supporting information), the relative sensitivities of functional groups involved in complexation in HA and LVF to Fe are as follows: keto group > carboxyl group > piperazine ring C-N > phenolic hydroxyl group (Figs. 2g and h).

  XPS was employed to examine the chemical states of elements within the complexes [33-36]. Upon introducing Fe(Ⅲ), the C-N bond in LVF dropped from 71.91% to 29.84%, while the C-O bond fell from 73.73% to 15.63% (Fig. 3a), indicating the involvement of the piperazine ring and carboxyl group in the reaction with Fe(Ⅲ) [7]. The C=O bonding decreased from 79.09% to 25.42% (Fig. 3b), indicating that the carboxyl group was the primary functional group involved. Additionally, Fe(Ⅲ)-LVF contained solely Fe(Ⅲ), whereas the Fe(Ⅲ)-HA complex exhibited both Fe(Ⅲ) and Fe(Ⅱ) (Fig. 3c). The outcome mirrors the variation in Fe(Ⅱ) levels detected with o-phenanthroline, suggesting that HA plays a vital role in facilitating the Fe(Ⅲ)/Fe(Ⅱ) cycle [37,38].

  Figure 3

  

  Figure 3.  XPS spectra in different systems of (a) N 1s, (b) O 1s and (c) Fe 2p.

  DownLoad: Full-Size Img PowerPoint

  The distinct quartet pattern for the DMPO-HO^• adduct, detected in the electron paramagnetic resonance (EPR) spectra of the Vis/HA/PDS system (Fig. 4a) indicates the generation of hydroxyl radicals (HO^•). However, the observed intensity of the DMPO-HO^• adduct in the Vis/PDS/Fe(Ⅲ) system appears to be markedly complex. This phenomenon can be attributed to the high oxidation potential exhibited by this system, whereby the DMPO-HO^• adduct is subject to oxidation, resulting in the formation of the DMPO_X product [39]. The signal detection of the DMPO-^•O₂⁻ adduct confirmed the formation of ^•O₂⁻ (Fig. 4b). Same adduct species were detected in the systems with and without HA, suggesting that the addition of HA has a minimal effect on ROS species. To further assess the effects of ROS on LVF degradation, tert-butanol (TBA), ethanol (EtOH), and superoxide dismutase (SOD) were used as quenchers (Fig. 4c) [39-42]. The inhibitory effects of TBA, EtOH and SOD were nearly similar, indicating that HO^• and ^•O₂⁻ were the major ROS. As shown in Fig. 4f, the conversion of PMSO is lower than 50%, and the inhibition of degradation is lower than 10%, indicating that less Fe(Ⅵ) is present in the system [43].

  Figure 4

  

  Figure 4.  EPR spectra for various systems: (a) DMPO-HO^•/DMPO_X and (b) DMPO-^•O₂⁻. (c) Effect of scavengers on LVF degradation in the Vis/Fe(Ⅲ)/HA/PDS system; (d, e) Levels of 7-hydroxycoumarin in different systems; (f) concentration of PMSO and PMSO₂ in the Vis/Fe(Ⅲ)/HA/PDS system. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = 2 mg/L, [PDS] = 2.50 mmol/L, [LVF] = 2 mg/L, [TBA] = [EtOH] = 1 mol/L, [SOD] = 150 mg/L, [PMSO] = 200 µmol/L, [EDTA] = 40 µmol/L, [coumarin] = 2 mmol/L, pH 7, light intensity 17.50 W/m².

  DownLoad: Full-Size Img PowerPoint

  Due to HO^• being an important ROS for LVF degradation, 7-hydroxycoumarin was used as a probe to further demonstrate and semi-quantify the formation of HO^• in different systems [44]. When photo, Fe(Ⅲ) and PDS were simultaneously present in the system, a significant 7-hydroxycoumarin signal was detected (Figs. 4d and e). The addition of HA further increased the 7-hydroxycoumarin concentration. Notably, the levels of 7-hydroxycoumarin were found to be considerably elevated when LVF was present compared to its absence. The LVF^* and HA^* produced by light can interact with water and oxygen to form HO^• [45,46]. Previous study has proposed that LVF^* can also directly activate PDS to generate HO^•, and this mechanism may also apply to HA, which has similar photosensitizing properties [47]. The Excitation-Emission-Matrix (EEM) spectra were utilized to additionally track changes in the levels of photosensitizing compounds. Results from Figs. S14 and S15 (Supporting information), HA alone and LVF alone solutions exhibited a significant fluorescence signal. The signal gradually weakened in the first 20 min and disappeared at 30 min (Figs. S16 and S17 in Supporting information), which related to the proliferation rate of 7-hydroxycoumarin in the Vis/Fe(Ⅲ)/HA/PDS/LVF setup. The findings suggest a strong correlation between the rise in HO^• production and the presence of photosensitizers like HA^* and LVF^*. Additionally, it was noted that EDTA greatly reduced HO^• production in the Vis/Fe(Ⅲ)/HA/PDS/LVF setup (Fig. 4d). The binding chemical bond has the potential to accelerate the electron transfer between the photosensitizing substances (HA^* and LVF^*) and PDS through complex molecules formed by the interaction of complexed-iron and PDS (HA/LVF-Fe(Ⅲ)-PDS) [24,25,48].

  To analyze the promotion of binding chemical bonds, ethylenediaminetetraacetic acid (EDTA) was utilized as a complexation masking agent. The complexation stabilization constant of EDTA with Fe(Ⅲ) is usually reported to be 7.08 [28]. In this study, the complexation stabilization constants of Fe(Ⅲ)-HA and Fe(Ⅲ)-LVF were calculated to be 5.06 and 4.48, respectively, and the details are shown in Text S7 (Supporting information). Compared to HA and LVF, EDTA is more competitive in complexing with iron.

  The capacity of HA and LVF to donate electrons was verified by measuring the Fe(Ⅱ) levels in various setups. As shown in Fig. 5a, the Fe(Ⅱ) concentration increased from 0 µmol/L to 6.30, 1.37, and 7.70 µmol/L by adding HA, LVF, and HA+LVF, respectively. The outcome mirrors the variation in Fe(Ⅱ) levels as determined by XPS analysis, suggesting that HA plays a vital role in facilitating the Fe(Ⅲ)/Fe(Ⅱ) cycle [37,38]. Particularly, the Fe(Ⅱ) concentration increased further after exposure to light radiation. An EDTA complexation masking agent was added to the HA/LVF/Fe(Ⅲ) system to assess the synergistic effect of photosensitization and complexation. The lower concentration of Fe(Ⅱ) should be consistent for the HA/LVF/Fe(Ⅲ) system with and without EDTA after the light is turned off, assuming that the generation of Fe(Ⅱ) occurs solely through the LMCT process. However, the generation of Fe(Ⅱ) decreased by 2.11 µmol/L after the addition of EDTA under dark conditions, while EDTA inhibited Fe(Ⅱ) generation by 4.22 µmol/L under light conditions. This suggests that, in addition to the electron supply from the electron-donating functional groups of the ligand, the photoelectrons generated by the photosensitization of the ligand also contribute to the increase in Fenton-like efficiency. There is a photosensitization-LMCT process. The increase in Fe(Ⅲ) redox potential under light also confirms this (Fig. 5b). A higher redox potential is favorable for Fe(Ⅲ) to gain electrons.

  Figure 5

  

  Figure 5.  (a) Fe(Ⅱ) concentration and (b) redox potentials in different systems. (c, d) Energy barriers of radical pathways in Fe/PDS, Fe-LVF/PDS and Fe-HA/PDS systems. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = [LVF] = 2 mg/L, pH 7, and light intensity 17.50 W/m².

  DownLoad: Full-Size Img PowerPoint

  Furthermore, the complexed iron has a lower redox potential than free iron, decreasing from 0.63 V (free Fe(Ⅲ)) to 0.43 V (Fe(Ⅲ)-LVF), 0.59 V (Fe(Ⅲ)-HA) and 0.57 V (HA-Fe(Ⅲ)-LVF) (Fig. 5b). The decrease in redox potential can increase the potential difference for the reaction with the PDS, thereby enhancing the thermodynamic feasibility of the Fenton reaction [16,17]. Furthermore, the energy barrier for radical generation through Fe-activated PDS was investigated by transition state calculations. Based on the characterization results and a previous study [7], the molecular formula of the Fe-LVF was postulated as shown in Fig. S18 (Supporting information). HA is an intricate substance, primarily characterized by carboxyl and phenolic hydroxyl groups, which are the key functional groups driving its effects. It was noted that tartaric acid (TA), citric acid (CA), and glycolic acetate (GA), all of which share similar structures, facilitated LVF breakdown (Fig. S19 in Supporting information). TA was selected as a representative HA model, and its complex with Fe(Ⅲ) was hypothesized, as illustrated in Fig. S20 (Supporting information). The DFT results showed (Figs. 5c and d) that the activation energies for the interactions of Fe(Ⅱ), Fe-HA, and Fe-LVF with PDS were 1.38, 0.19, and 0.23 eV, respectively. Compared to free iron, the energy barriers of complexed iron were reduced by 1.19 and 1.15 eV, respectively, indicating that complexed iron can activate PDS more effectively. The formation of complexed iron can enhance the thermodynamic feasibility of the Fenton-like reaction.

  Results show that both HA and LVF can effectively promote the Fenton reaction, and the mechanism is closely associated with complexation and photosensitization. The details are as follows. Firstly, the photosensitizing substances (HA^* and LVF^*) can directly transfer electrons to water, oxygen, and PDS, which promote the formation of HO^•. Compared to free iron, complexed iron improves the potential difference with PDS and reduces the reaction energy barrier for Fe-activated PDS. This can enhance the thermodynamic feasibility of Fenton-like reactions. Furthermore, the photoexcitation of HA and LVF can be transferred to Fe(Ⅲ) via LMCT. Then, the electrons are rapidly transferred to PDS through complex molecules formed by the interaction of iron-complexed and PDS (HA/LVF-Fe(Ⅲ)-PDS), ultimately enhancing the activation efficiency of PDS.

  Furthermore, the degradation products of LVF were examined using UPLC-Q-TOF MS/MS, and four possible degradation pathways were proposed with existing works. Specific details are shown in Text S8 (Supporting information). The generalization of the HA-promoted mechanism to the degradation of other pollutants is also observed and extended to other small molecule carboxylic acids with similar structures. Details are presented in Text S9 (Supporting information).

  This research discovered that enhancement of HA on LVF breakdown in the Fenton-like system was notably more effective under illuminated conditions compared to darkness. The proposed promotion mechanism included complexation and photosensitization. FTIR and XPS analyses revealed that the primary locations for Fe(Ⅲ)-HA and Fe(Ⅲ)-LVF formation are the phenolic hydroxyl group, carboxyl group, and piperazine ring. The formation of complexed iron significantly reduces the redox potential of iron and increases the potential difference with PDS, which is responsible for accelerating electron transfer. Chemical calculations demonstrate that complexed iron reduces the energy barriers for PDS activation by 1.19 eV (Fe-HA) and 1.15 eV (Fe-LVF), respectively, compared to free iron. Furthermore, the complex molecules formed by the interaction of iron-complexed and PDS (HA/LVF-Fe(Ⅲ)-PDS) accelerated electron transfer between the photosensitizers (HA^* and LVF^*) with Fe(Ⅲ) and PDS. These findings could offer fresh perspectives on the combined elimination of contaminants through natural organic compounds and light exposure.

  Declaration of competing interests

  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

  Xuejia Li: Writing – original draft, Investigation, Data curation. Yang Liu: Writing – review & editing, Supervision, Funding acquisition. Jian Wei: Supervision, Resources, Funding acquisition. Yujia Xiang: Methodology, Data curation. Xinruo Wang: Investigation, Data curation. Hanchang Wang: Validation, Methodology. Heng Zhang: Supervision, Software, Investigation. Bo Lai: Supervision, Resources.

  Acknowledgments

  This work was supported by the Natural Science Foundation of China (No. 42107073), Central Guidance for Local Science and Technology Development Fund Projects (No. 2024ZYD0030), Natural Science Foundation of Sichuan Province (No. 2024NSFSC0130), and the Sichuan Science and Technology Program (No. 2024NSFTD0014).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110811.

  
  
• 1. [1]

     C. Yu, Z. Wu, H. Shi, et al., Chin. Chem. Lett. 35 (2024) 109334. doi: 10.1016/j.cclet.2023.109334

  2. [2]

     L. Lai, H. Zhou, Y. Hong, et al., Chin. Chem. Lett. 35 (2024) 108580. doi: 10.1016/j.cclet.2023.108580

  3. [3]

     M. Yang, Z. Hou, X. Zhang, et al., Environ. Sci. Technol. 56 (2022) 11635–11645. doi: 10.1021/acs.est.2c01261

  4. [4]

     W. Xiang, M. Huang, X. Wu, et al., Chin. Chem. Lett. 33 (2022) 1275–1278. doi: 10.1016/j.cclet.2021.07.072

  5. [5]

     Y. Shang, Y. Kan, X. Xu, Chin. Chem. Lett. 34 (2023) 108278. doi: 10.1016/j.cclet.2023.108278

  6. [6]

     R. Yin, Y. Chen, J. Hu, et al., Appl. Catal. B 283 (2021) 119663. doi: 10.1016/j.apcatb.2020.119663

  7. [7]

     A. Wang, Y. Shi, Y. Liu, et al., J. Hazard. Mater. 462 (2024) 132755. doi: 10.1016/j.jhazmat.2023.132755

  8. [8]

     Y. Yuan, Z. Zhou, X. Zhang, et al., Chin. Chem. Lett. 34 (2023) 107932. doi: 10.1016/j.cclet.2022.107932

  9. [9]

     M. Huang, S. Peng, W. Xiang, et al., Chem. Eng. J. 429 (2022) 132372. doi: 10.1016/j.cej.2021.132372

  10. [10]

      B. Yang, C. Wang, X. Cheng, et al., Water Res. 202 (2021) 117379. doi: 10.1016/j.watres.2021.117379

  11. [11]

      M. Huang, Y.S. Li, C.Q. Zhang, et al., Proc. Natl. Acad. Sci. U. S. A. 119 (2022) e2202682119. doi: 10.1073/pnas.2202682119

  12. [12]

      M. Huang, Y. Han, W. Xiang, et al., Environ. Sci. Technol. 55 (2021) 15361–15370. doi: 10.1021/acs.est.1c03758

  13. [13]

      Z. Wang, Y. Du, P. Zhou, et al., Chem. Eng. J. 454 (2023) 140096. doi: 10.1016/j.cej.2022.140096

  14. [14]

      B. Yang, X. Cheng, Y. Zhang, et al., J. Hazard. Mater. 407 (2021) 124853. doi: 10.1016/j.jhazmat.2020.124853

  15. [15]

      X. Zhou, X. Li, Y. Xiang, et al., Chin. Chem. Lett. 36 (2025) 110664. doi: 10.1016/j.cclet.2024.110664

  16. [16]

      W. Huang, M. Brigante, F. Wu, et al., Environ. Sci. Technol. 47 (2013) 1952–1959. doi: 10.1021/es304502y

  17. [17]

      F. Jiang, Y. Li, W. Zhou, et al., Chem. Eng. J. 387 (2020) 124048. doi: 10.1016/j.cej.2020.124048

  18. [18]

      Z. Wang, J. Lv, S. Zhang, et al., Environ. Sci. Technol. 55 (2021) 1769–1778. doi: 10.1021/acs.est.0c07132

  19. [19]

      K.P. Mangalgiri, L. Blaney, Environ. Sci. Technol. 51 (2017) 12310–12320. doi: 10.1021/acs.est.7b03482

  20. [20]

      H. Dong, C. Sans, W. Li, Z. Qiang, Sep. Purif. Technol. 171 (2016) 144–150. doi: 10.1016/j.seppur.2016.07.033

  21. [21]

      X. Shi, Y. Li, Z. Zhang, et al., Chem. Eng. J. 372 (2019) 1113–1121. doi: 10.1016/j.cej.2019.04.195

  22. [22]

      Y. Tong, P. Zhou, Y. Liu, et al., J. Hazard. Mater. 428 (2022) 128202. doi: 10.1016/j.jhazmat.2021.128202

  23. [23]

      Y. Zhou, Z. Zeng, J. Fu, et al., Environ. Sci. Technol. 58 (2024) 4019–4028. doi: 10.1021/acs.est.3c10703

  24. [24]

      C. Zhang, D. Zhang, Z. Li, et al., Environ. Sci. Technol. 48 (2014) 6318–6325. doi: 10.1021/es501056n

  25. [25]

      C. Peng, A. Sundman, C. Bryce, et al., Environ. Sci. Technol. 52 (2018) 5753–5763. doi: 10.1021/acs.est.8b00953

  26. [26]

      X. Huang, Z.L. Ye, J. Cai, L. Lin, Water Res. 206 (2021) 117756. doi: 10.1016/j.watres.2021.117756

  27. [27]

      L.J. Yang, Q.L. Liu, M.X. Wang, et al., Spectrochim. Acta. A: Mol. Biomol. Spectrosc. 166 (2016) 1–7. doi: 10.1016/j.saa.2016.04.048

  28. [28]

      Y. Zhang, S. Wu, P. Sun, Sci. Total Environ. 899 (2023) 165702. doi: 10.1016/j.scitotenv.2023.165702

  29. [29]

      D. Tian, H. Zhou, H. Zhang, et al., Chem. Eng. J. 428 (2022) 131166. doi: 10.1016/j.cej.2021.131166

  30. [30]

      W. Yan, J. Zhang, C. Jing, J. Colloid Interface Sci. 390 (2013) 196–203. doi: 10.1016/j.jcis.2012.09.039

  31. [31]

      I.M. Le-Deygen, A.S. Safronova, I.M. Kolmogorov, et al., Russ. J. Bioor. Chem. 48 (2022) 710–719. doi: 10.1134/s1068162022040148

  32. [32]

      Y. Fan, C. Zheng, A. Huo, et al., Ecotoxicol. Environ. Saf. 181 (2019) 34–42. doi: 10.1016/j.ecoenv.2019.05.076

  33. [33]

      W. Yan, H. Wang, C. Jing, Environ. Sci. Technol. 50 (2016) 4343–4349. doi: 10.1021/acs.est.6b00066

  34. [34]

      A. Shchukarev, D. Korolkov, Open Chem. 2 (2004) 347–362. doi: 10.2478/BF02475578

  35. [35]

      M. Sun, P. Zhou, J. Peng, et al., Chem. Eng. J. 435 (2022) 134836. doi: 10.1016/j.cej.2022.134836

  36. [36]

      Y. Liu, H. Guo, Y. Zhang, et al., Chem. Phys. Lett. 653 (2016) 101–107. doi: 10.1016/j.cplett.2016.04.069

  37. [37]

      Y. Wang, T. Pan, Y. Yu, et al., Water Res. 185 (2020) 116136. doi: 10.1016/j.watres.2020.116136

  38. [38]

      Y. An, X. Li, Z. Liu, et al., Chemosphere. 317 (2023) 137883. doi: 10.1016/j.chemosphere.2023.137883

  39. [39]

      S.D. Stan, M.A. Daeschel, J. Agric. Food Chem. 53 (2005) 4906–4910. doi: 10.1021/jf047918k

  40. [40]

      G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data 17 (1988) 513–886. doi: 10.1063/1.555805

  41. [41]

      Y. Yao, H. Chen, C. Lian, et al., J. Hazard. Mater. 314 (2016) 129–139. doi: 10.1016/j.jhazmat.2016.03.089

  42. [42]

      H. Wang, H. Yao, P. Sun, et al., Environ. Sci. Technol. 50 (2016) 145–153. doi: 10.1021/acs.est.5b03696

  43. [43]

      B. Shao, H. Dong, L. Feng, et al., J. Hazard. Mater. 384 (2020) 121303. doi: 10.1016/j.jhazmat.2019.121303

  44. [44]

      H. Guan, L. Zhu, H. Zhou, H. Tang, Anal. Chim. Acta. 608 (2008) 73–78. doi: 10.1016/j.aca.2007.12.009

  45. [45]

      M. Chen, J. Xu, R. Tang, et al., J. Hazard. Mater. 440 (2022) 129784. doi: 10.1016/j.jhazmat.2022.129784

  46. [46]

      Z. Guo, J. Wang, X. Chen, et al., Water Res. 188 (2021) 116568. doi: 10.1016/j.watres.2020.116568

  47. [47]

      A. Wang, P. Zhou, D. Tian, et al., Appl. Catal. B: Environ. 316 (2022) 121631. doi: 10.1016/j.apcatb.2022.121631

  48. [48]

      Y. Zhou, Z. Zeng, J. Fu, et al., Environ. Sci. Technol. 48 (2024). 6318-6325. doi: 10.1109/tvt.2023.3345156

• 

  Figure 1  The degradation of LVF in different system (a) within 30 min and (b) within 10 min. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = 2 mg/L, [PDS] = 2.50 mmol/L, [LVF] = 2 mg/L, pH 7, and light intensity 17.50 W/m².

  下载: 全尺寸图片 幻灯片
  

  Figure 2  The 2D-COS spectra from FTIR. Synchronous (a) and asynchronous (b) of Fe(Ⅲ)-LVF. Synchronous (c) and asynchronous (d) of Fe(Ⅲ)-HA. Synchronous (e) and asynchronous (f) of HA-LVF. Synchronous (g) and asynchronous (h) of Fe(Ⅲ)-LVF-HA. A positive correlation is indicated by red, a negative correlation by blue, and the strength of the correlation is indicated by the darkness of the color.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  XPS spectra in different systems of (a) N 1s, (b) O 1s and (c) Fe 2p.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  EPR spectra for various systems: (a) DMPO-HO^•/DMPO_X and (b) DMPO-^•O₂⁻. (c) Effect of scavengers on LVF degradation in the Vis/Fe(Ⅲ)/HA/PDS system; (d, e) Levels of 7-hydroxycoumarin in different systems; (f) concentration of PMSO and PMSO₂ in the Vis/Fe(Ⅲ)/HA/PDS system. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = 2 mg/L, [PDS] = 2.50 mmol/L, [LVF] = 2 mg/L, [TBA] = [EtOH] = 1 mol/L, [SOD] = 150 mg/L, [PMSO] = 200 µmol/L, [EDTA] = 40 µmol/L, [coumarin] = 2 mmol/L, pH 7, light intensity 17.50 W/m².

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Fe(Ⅱ) concentration and (b) redox potentials in different systems. (c, d) Energy barriers of radical pathways in Fe/PDS, Fe-LVF/PDS and Fe-HA/PDS systems. Experimental parameters: [Fe(Ⅲ)] = 40 µmol/L, [HA] = [LVF] = 2 mg/L, pH 7, and light intensity 17.50 W/m².

  下载: 全尺寸图片 幻灯片
•