English

• Advanced oxidation processes (AOPs) are widely applied in wastewater treatment especially for the efficient removal of refractory organic pollutants [1-3]. Amongst peroxymonosulfate (PMS) based homogeneous AOPs has been considered as a promising technique for wastewater remediation due to its strong ability to oxidize contaminants and produce long half-life periods of SO₄^•− [4-6]. However, the homogeneous system generally suffers from harsh pH environment requirement, large consumption of oxidant, extensive use of ionic catalysts at concentration dozens or even hundreds of times higher than their emission standard, and the excess slug production [7,8]. Thus, considerable efforts have been devoted to developing heterogeneous catalysts for PMS activation. Along this line, advanced heterogeneous catalysts such as Co-based catalysts [7,9], Fe-based catalysts [2,10-12] and metal–N–C [13-15] have been reported. Amongst, Co-based heterogeneous catalysts show excellent performance of activating PMS, but most of them are limited by insufficient utilization and regeneration rate of Co²⁺ (the active sites), as well as the leaching of the Co [16].

  In previous studies, two effective approaches have been proposed to solve the above issues. The first one is to disperse the Co-based catalysts on a host with large surface area and strong interaction with the catalysts, thus inhibiting the aggregation of catalysts and diminishing the possible Co leaching [17]. Another promising approach is to design Co-based composite metal oxide (e.g., perovskite catalysts [1], spinel catalysts [18] and double hydroxide catalysts [19]), which obviously improve the structural stability of catalysts from metal leaching. LaCoO₃, a typical example of perovskite catalysts, was widely used due to its reasonable high stability and catalytic activity, however, intrinsic high state of Co (i.e., Co³⁺) in the farmwork hinders its continuous activation of PMS. As a matter of fact, perovskite A_xB_yO_z with low-state A-site metals as the active site would be more prone to donate electrons for catalytic activation. In addition, B-site metals are essential for the regeneration of the A-site active site and the structural stability of the catalyst [20,21].

  Based on above design principles, we herein proposed perovskite cobalt titanate (CoTiO₃) as a promising PMS activator and suspected that the presence of Co–O–Ti bond would cause a strong interaction between Co and Ti, thus promoting the rapid transfer of electrons in the catalytic process. To verify the above hypothesis, CoTiO₃ nanocatalyst was synthesized by a simple one-step hydrothermal method and employed as PMS activator. As a result, CoTiO₃/PMS system removes 98.2% of hydroxychloroquine (HCQ, drugs for effective treatment of COVID-19) within 20 min at low dose of PMS (0.5 mmol/L), shows high tolerance to the environmental pH range (3.5–10.6) and significant versatility for various refractory organics. Combined with the material characterization and DFT calculations, the mechanism for the excellent catalytic performance of heterogeneous CoTiO₃ was revealed. Furthermore, the possible degradation pathway of HCQ and toxicity studies of intermediate products in CoTiO₃/PMS systems were proposed based on GCMS– and T.E.S.T. analysis. In addition, CoTiO₃–CNT–PVDF membrane was designed to further demonstrate the practical application of the catalysts using continuous flow-through mode approach.

  X-ray diffraction (XRD) pattern of the obtained catalysts can be assigned to perovskite CoTiO₃ (JCPDS No.15–0866) as shown in Fig. 1a, confirming its successful synthesis using a simple one-step hydrothermal method. X-Ray photoelectron spectroscopy (XPS) were collected to analyze the chemical states of elements in CoTiO₃. As shown in Fig. 1b, both peaks with binding energies of 786.8 eV and 780.6 eV originated from Co²⁺ [18,22,23], which was usually reported as a highly active center for PMS activation. Two major peaks are observed at 458.6 eV and 463.9 eV for Ti 2p spectrum (Fig. 1c) [24], which can be assigned to Ti³⁺ and Ti⁴⁺ with percentage of 51.5% and 48.5%, respectively. In addition, the splitting energy of 5.7 eV between two Ti 2p peaks is well in agreement with the TiO₆ octahedra [25,26]. The O 1s spectra of CoTiO₃ consists of three peaks at 529.9, 531.4 and 532.7 eV (Fig. 1d), which can be assigned to the Co–O–Ti bond, oxygen vacancies and surface-bonded hydroxyl groups [27,28]. The Fourier transform infrared spectroscopy (FTIR) (Fig. 1e) reveal the vibration of Ti-O-Ti (800–500 cm⁻¹) and Co–O–Ti (1091 cm⁻¹) [29-32]. The structure was further investigated by Raman spectroscopy (Fig. 1f), where the vibration peaks at 521 cm⁻¹ and 689 cm⁻¹ can be assigned to symmetric bending motion of TiO₆ and CoO₆ octahedra, respectively [33,34]. It is worth noting that the formation of Co–O–Ti linkage and oxygen vacancy are beneficial to the electron transfer in the catalytic activation process, improving the redox cycle efficiency of Co/Ti, whose mechanism will be discussed in following parts. The scanning electron microscope (SEM) images in Fig. 1g demonstrates the nano-block morphology of CoTiO₃ with lateral length less than 500 nm. The characteristic d-spacing of 0.2537 nm shown by HRTEM (Fig. 1h) can be assigned to (110) plane of CoTiO₃. Moreover, the uniform elemental distribution with composition of CoTi_0.99O_3.03 was confirmed by energy dispersive X-ray spectroscopy (EDS) analysis, which is consistent with CoTiO₃ (Fig. S1 in Supporting information).

  Figure 1

  

  Figure 1.  Material characterizations of CoTiO₃. (a) XRD pattern (inset shows the crystal structure of CoTiO₃). High-resolution XPS spectra of (b) Co 2p, (c) Ti 2p and (d) O 1s. (e) FTIR spectra. (f) Raman spectra. (g) SEM images. (h) HRTEM image (insets show TEM image and SAED pattern).

  DownLoad: Full-Size Img PowerPoint

  The catalytic performance was evaluated by degradation of HCQ in CoTiO₃/PMS system. As exhibited in Fig. 2a and Fig. S2 (Supporting information), CoTiO₃/PMS system with low CoTiO₃ and PMS dosages (i.e., 0.5 g/L and 0.5 mmol/L, respectively) can efficiently remove 98.2% of HCQ (10 mg/L) within 20 min, which is much faster than that of PMS (11.2%) or CoTiO₃ (19.1%) alone. Besides, the degradation rate only reaches 62.3% in Co₃O₄/PMS systems under identical conditions, reflecting that monometallic cobalt oxide catalyst showed a significantly lower activity than CoTiO₃. To further demonstrate the excellent degradation performance of CoTiO₃/PMS system, it was compared with the well-known efficient homogeneous systems with same molar of ionic catalysts (i.e., Fe²⁺/PMS, Cu²⁺/PMS and Co²⁺-Ti⁴⁺/PMS). It is found that the degradation of HCQ followed the first-order reaction model (Fig. S3 in Supporting information). As shown in Fig. 2b and Fig. S4 (Supporting information), the degradation rate constant for CoTiO₃/PMS system (0.5683 min⁻¹) was higher than that of any reported homogeneous systems herein, including its homogenous counterpart with the same molar concentration of metal ions (i.e., Co²⁺-Ti⁴⁺/PMS system). The above results demonstrate that there is no synergistic effect between Co²⁺ and Ti⁴⁺ in the homogeneous system, while in the heterogeneous CoTiO₃, the presence of Co–O–Ti accelerates the electron transfer within the system, which is more conducive to the catalytic process. Besides, we also found that the pH of CoTiO₃/PMS system is maintained in the neutral environment (pH mainly lies between 6 and 7), while the Co²⁺-Ti⁴⁺/PMS system lies in a strong acidic environment with pH range between 3 and 3.5 as shown in Fig. S5 (Supporting information). As is well-known, strong acid environment would inhibit the oxidation ability of the generated ^•OH [35], so the Co²⁺-Ti⁴⁺/PMS system exhibits lower degradation efficiency. Moreover, the mineralization efficiency of HCQ within 20 min is as high as 71.6% (Fig. 2c), demonstrating the strong oxidation capacity of CoTiO₃/PMS system. More importantly, the CoTiO₃/PMS system can degrade more than 90% of HCQ within 20 min after 5 cycling experiments and the leaching amount of Co ions is always less than 0.25 mg/L, which is lower than environmental standard stipulated by European Union (1 mg/L) [36], clearly demonstrating its good reusability and stability (Fig. 2d and Fig. S6 in Supporting information). As is well-known, the practical application of homogenous PMS system is significantly hindered by the harsh requirement on environmental pH and massive production of heavy metal-containing sludge [37]. In stark contrast, the CoTiO₃/PMS system possesses high tolerance to the environmental pH and the removal efficiency of 10 mg/L HCQ at 20 min is higher than 80% in the pH range of 3.5–10.6 (Fig. 2e and Fig. S7 in Supporting information). Furthermore, the versatility of CoTiO₃/PMS system is also confirmed by the efficient degradation of several typical organic pollutants like endocrine disruptor (i.e., BPA), pharmaceutical (i.e., CBZ), antibiotic (i.e., SMX), and dye (i.e., MB) as shown in Fig. 2f and Fig. S8 (Supporting information). The degradation rate of above pollutants by CoTiO₃/PMS system can achieve higher than 85% in a short period of time (< 20 min). In addition, it is worth mentioning that the CoTiO₃/PMS system can maintain the high catalytic performance in the presence of common ionic and organic species (i.e., CO₃²⁻ (0–10 mmol/L), Ca²⁺ (0–5 mmol/L) and humic acid (HA, 0–20 mg/L)) (Fig. S9 in Supporting information). All these results clearly demonstrate that the CoTiO₃/PMS system holds great potential regarding to the practical application in organic wastewater treatment.

  Figure 2

  

  Figure 2.  (a) Degradation efficiencies of HCQ and (b) the first order kinetic reaction rate constant in first 5 min in different systems. (c) Total organic carbon (TOC) removal of HCQ in CoTiO₃/PMS system. (d) Cycle experiments of CoTiO₃/PMS conducted with recycled CoTiO₃. (e) The influence of initial pH on HCQ degradation in CoTiO₃/PMS system. (f) Degradation of different pollutants in CoTiO₃/PMS system.

  DownLoad: Full-Size Img PowerPoint

  The linear sweep voltammetry (LSV) result shows distinctly increased current after adding PMS in pure CoTiO₃ suspension, indicating efficient electron transfer between CoTiO₃ and PMS for generating reactive species (Fig. 3a). EPR measurement was carried out with DMPO trapping agent to verify the existing radicals in CoTiO₃/PMS system. As shown in Figs. 3b and c, the characteristic signals of DMPO–SO₄^•–, DMPO–^•OH and DMPO–O₂^•− can be clearly seen during the reaction process in CoTiO₃/PMS system, confirming the presence of SO₄^•−, ^•OH and O₂^•−. Meanwhile, the typical three-line spectra (1:1:1) of stable TEMP-¹O₂ was not observed in EPR test, indicating little ¹O₂ is produced in the system (Fig. S10 in Supporting information). To further reveal the major reactive species in HCQ degradation over CoTiO₃/PMS system, a series of quenching experiments were conducted. As shown in Figs. 3d and e, with the addition of MeOH (^•OH and SO₄^•− quencher, 100 mmol/L) [38] and TBA (^•OH quencher, 100 mmol/L) [39], the degradation efficiency of HCQ in CoTiO₃/PMS system was decreased by 76.2% and 33.1%, respectively. Moreover, upon the addition of BQ (O₂^•− quencher, 4 mmol/L) [40] in CoTiO₃/PMS system, the HCQ degradation efficiency is decreased by 23.3%. With the addition of FFA (¹O₂ quencher, 100 mmol/L), the degradation efficiency of HCQ was just decreased by 2%, suggesting a negligible degradation contribution from non-free radical pathways enabled by ¹O₂. The above results clearly indicate SO₄^•− and ^•OH are the major reactive species for HCQ degradation in CoTiO₃/PMS system, while O₂^•− showed slight contribution. The overall contribution is as follows: SO_{4contribution}^•− > ^•OH_contribution > O_{2contribution}^•−. Furthermore, the concentration evolution of major reactive oxidation species (i.e., SO₄^•− and ^•OH) in CoTiO₃/PMS and Co²⁺-Ti⁴⁺/PMS during the activation process were monitored by EPR. As shown in Fig. S11 (Supporting information), the amount of DMPO-SO₄^•− and DMPO-^•OH in the homogeneous system decreased significantly in 10 min. In contrast, the free radicals in CoTiO₃/PMS system were found to increase gradually over the same time period as shown in Fig. 3f and Fig. S10, verifying the continuous catalysis of CoTiO₃ due to the fast regeneration of its active site.

  Figure 3

  

  Figure 3.  (a) LSV curves of CoTiO₃ with/without PMS and HCQ. EPR spectra of (b) SO₄^•−, ^•OH and (c) O₂^•− in CoTiO₃/PMS system using DMPO trapping agent at different time intervals. (d) Removal efficiency and (e) inhibition ratio of HCQ degradation in the presence of different quenchers and (f) the corresponding quantitative concentration of SO₄^•− and ^•OH in CoTiO₃/PMS system using DMPO trapping agent at different time intervals.

  DownLoad: Full-Size Img PowerPoint

  To uncover the mechanism of the excellent catalytic performance of CoTiO₃ in activating PMS, the adsorption/activation of PMS on CoTiO₃ were explored by performing density functional theory (DFT) calculations. Computational details can be found in Supporting information. As shown in Fig. 4a, dynamic catalytic process for PMS activation via CoTiO₃ with high exposure of (104) facets was revealed. The HSO₅⁻ was randomly placed on the CoTiO₃ at the beginning, but it adsorbed on Co sites and Ti sites, thus forming a cyclic electron chain by HSO₅^- and Co–O–Ti together. Afterwards, the HSO₅^- was activated to form SO₄^•− and ^•OH via breaking the O–O bond. We also have found that the energy of product was much lower than that of the reactant (Fig. 4b), revealing that the PMS activation is favorable in thermodynamics, which is consistent with the superior catalytic performance of CoTiO₃. Furthermore, the Bader charge results indicate that the valence electrons of Co sites decreased significantly after PMS activation, while only a slight number of valence electrons were transferred at Ti site. Due to the higher density of electron clouds around Co, it is more inclined to act as an electron "supplier" to PMS, while Ti acts as a "feeder" and transmits its electrons through Co–O–Ti bond (Fig. 4c and Fig. S12 in Supporting information). Besides, the valence electrons on HSO₅^- before and after activation is −0.524 and −1.755, respectively. Based on the above theoretical calculations, it is clear that the PMS is efficiently activated to generate SO₄^•− and ^•OH by accepting the electrons from CoTiO₃ (~1.231 eV).

  Figure 4

  

  Figure 4.  (a) Adsorption and activation of HSO₅⁻. (b) Gibbs free energy (△G) of different states of HSO₅⁻ on CoTiO₃ crystal. (c) Analyse of Bader charge before and after activation.

  DownLoad: Full-Size Img PowerPoint

  To confirm the understanding based on DFT calculations, the high resolution XPS spectra before and after activation PMS were explored. As can be seen in Fig. 5a, the percentage of Co²⁺ in the catalyst after use for 5 degradation cycles decreased from 100% to 53.1%, respectively, as they donate electrons to activate PMS (Eq. 1). This analysis is supported by the fact that the oxidation potential of CoTiO₃ (~0.28 V, Fig. 5d) is even much lower than that oxidation potential of the common efficient ionic catalysts (e.g., Co²⁺, Fe²⁺ and Cu²⁺), confirming its stronger electron-donating capacity [41-43]. On the other hand, Co²⁺ still maintained a high proportion (i.e., 53.1%) after 5 degradation cycles, indicating its efficient regeneration, which is also supported by the continuous decomposition of PMS in the CoTiO₃/PMS system (Fig. S13 in Supporting information) and excellent cycle stability for HCQ degradation. Meanwhile, Figs. 5b and c showed the slight decrease of Ti³⁺ (from 52.7% to 43.9%) and surface oxygen (from 24.2% to 18.7%) on the used CoTiO₃, which is mainly due to the regeneration of Co²⁺ through the bridge of Co–O–Ti bond (Fig. S14 in Supporting information).

  Figure 5

  

  Figure 5.  XPS spectra of the fresh CoTiO₃ and used CoTiO₃ after 5 degradation cycles: (a) Co 2p, (b) Ti 2p and (c) O 1s. (d) The CV curve of CoTiO₃. (e) Schematic illustration of PMS activation for HCQ removal.

  DownLoad: Full-Size Img PowerPoint

  Combined with the material characterization and DFT calculations, CoTiO₃ catalyst is expected to work as follows during the PMS activation. First and foremost, the formation of Co–O–Ti bond (as verified by FTIR, Raman and XPS analysis) induces the higher electron density of Co site, leading to the formation of Co²⁺, which in return serves as active sites to contribute electrons for PMS activation [41]. During this process, SO₄^•− and ^•OH could be generated for HCQ removal (Eqs. 1 and 2). Then, according to the standard reduction potentials in Langes' Chemistry Handbook (Eqs. 3-5), the reduction potential of Co³⁺ is 1.82 V, while Ti³⁺ is involved, the E⁰ decreases to 1.63 V. Therefore, the reduction of Co³⁺ by Ti³⁺ is more thermodynamically feasible, verifying catalysts containing Co–O–Ti bond could provide fast regeneration of Co²⁺ [34,44]. Finally, Ti⁴⁺ further efficiently reacts with HSO₅⁻ to achieve recovery (Eq. 6) [45]. In addition, the regeneration of Co²⁺ can be also achieved through the reduction by PMS and electron-rich intermediates (i.e., Os, Eq. 7) [7,46], which reduces the amount of surface oxygen on the used CoTiO₃. Based on the above investigations, it can be concluded that the formation of Co–O–Ti linkage facilitates the rapid recycling of Co²⁺/Co³⁺ on CoTiO₃, enabling continuously efficient activation of PMS. Schematic illustration of CoTiO₃ activated PMS for HCQ removal is shown in Fig. 5e. In addition, fast regeneration of Co²⁺ ensures the high concentration of SO₄^•− and ^•OH produce, thus making HCQ more inclined to be converted into low or non-toxic products as confirmed by the degradation pathway and toxicity analysis (Figs. S15 and S16 in Supporting information).

  $ \mathrm{Co}^{2+}+\mathrm{HSO}_5^{-} \rightarrow \mathrm{Co}^{3+}+\mathrm{SO}_4^{\bullet-}+\mathrm{OH}^{-} $

  (1)

  $ \mathrm{SO}_4^{\bullet-}+\mathrm{OH}^{-} \rightarrow \mathrm{SO}_4^{2-}+\;^{\bullet}\mathrm{OH} $

  (2)

  $ \mathrm{Co}^{3+}+\mathrm{e}^{-} \rightarrow \mathrm{Co}^{2+}, E^0=1.82 \mathrm{~V} $

  (3)

  $ \mathrm{Ti}^{4+}+\mathrm{e}^{-} \rightarrow \mathrm{Ti}^{3+}, E^0=0.19 \mathrm{~V} $

  (4)

  $ \mathrm{Co}^{3+}+\mathrm{Ti}^{3+} \rightarrow \mathrm{Co}^{2+}+\mathrm{Ti}^{4+}, E^0=1.63 \mathrm{~V} $

  (5)

  $ \mathrm{Ti}^{4+}+\mathrm{HSO}_5^{-}+\mathrm{H}^{+}+\mathrm{V}_{\mathrm{O}}^{\bullet \bullet} \rightarrow \mathrm{Ti}^{3+}+\mathrm{SO}_5^{\bullet-}+\mathrm{H}^{+}+\mathrm{O}_{\mathrm{O}}^{\times} $

  (6)

  $ \mathrm{Co}^{3+}+\mathrm{HSO}_5^{-}+\mathrm{H}^{+}+\mathrm{V}_{\mathrm{O}}^{\bullet \bullet} \rightarrow \mathrm{Co}^{2+}+\mathrm{SO}_5^{\bullet-}+\mathrm{H}^{+}+\mathrm{O}_{\mathrm{O}}^{\times} $

  (7)

  where $\mathrm{V}_0^{\bullet \bullet}$ is the double-charged oxygen vacancy, and $\mathrm{O}_{\mathrm{O}}^{\times}$ is the oxygen ion in CoTiO₃.

  The separation and recovery of powder catalysts are key challenges limiting the practical applications of heterogenous AOPs. To this end, CoTiO₃–CNT–PVDF catalytic membrane reactor was designed to demonstrate its ability to remove microcontaminants in wastewater via continuous flow-through mode (Fig. 6a). As shown in Fig. 6b and Fig. S17 (Supporting information), the HCQ could be efficiently degraded (> 90%) through continuous flow within 540 min via CoTiO₃–CNT–PVDF membrane with little cobalt ion leaching. The excellent catalytic performance is consistent with the catalyst agglomeration-free property of the membrane, which enables the efficient use of the catalyst (Fig. S18 in Supporting information). Furthermore, compared with previously developed membrane-based AOPs (Fig. 6c and Table S3 in Supporting information), the CoTiO₃–CNT–PVDF membrane/PMS system delivers a highest removal efficiency and longest operation duration for degrading various organic pollutants. To understand the excellent stability, the XRD patterns and SEM images of the composite membrane were collected before and after the reaction. As shown in Fig. 6d, the crystal structure of the membrane is well maintained after long time operation, although the peak intensity slightly decreases. Fig. 6e shows the membrane morphology and skeleton are well maintained after the reaction, which is consistent with the superior stability of the membrane reactor. It can be concluded that when the catalysts, PMS and pollutant are confined together in the channel of membrane, the generated active species would quickly interact with the pollutant in situ, so as to accurately and effectively remove the pollutant.

  Figure 6

  

  Figure 6.  (a) Schematic diagram of the preparation of CoTiO₃–CNT–PVDF membrane. (b) HCQ degradation efficiency of CoTiO₃–CNT–PVDF membrane. (c) Comparison of CoTiO₃–PVDF–CNT membrane/PMS system with previously developed catalytic membranes in Table S2 (Supporting information). (d) XRD patterns and (e) SEM images of CoTiO₃–CNT–PVDF membrane before and after reaction.

  DownLoad: Full-Size Img PowerPoint

  To sum up, in CoTiO₃/PMS system, Co is more inclined to act as an electron "supplier" to PMS due to its higher electron density, while Ti acts as a "feeder" and transmits its electrons through Co–O–Ti bond to ensure cobalt regeneration, then facilitating the redox cycle of Co²⁺/Co³⁺ during PMS activation and thus accelerating the superior degradation of various pollutants. In practical application demonstration, a designed CoTiO₃–CNT–PVDF membrane reactor can effectively remove HCQ pollutant via practically feasible filter-through mode, whose performance is superior to previously developed membrane-based AOPs. This work reports a catalytic model of CoTiO₃–PVDF membrane/PMS, which provides new strategies for the continuous, rapid, economical and stable removal of stubborn contaminants from wastewater.

  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

  Aiping Liang: Writing – review & editing, Writing – original draft, Resources, Methodology, Investigation. Chaolin Li: Writing – review & editing, Supervision. Chen Ling: Methodology, Investigation. Hengpan Duan: Methodology, Investigation. Wenhui Wang: Conceptualization, Writing – review & editing, Supervision, Funding acquisition.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 52100084 and 52170155) and Shenzhen Natural Science Fund (the stable support plan program, No. GXWD20231129152058003).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110788.

  
  
• 1. [1]

     C.Y. Wang, C. Hu, F. Chen, et al., Adv. Funct. Mater. 33 (2023) 2301144. doi: 10.1002/adfm.202301144

  2. [2]

     F.B. Yu, C. Jia, X. Wu, et al., Nat. Commun. 14 (2023) 4975. doi: 10.1038/s41467-023-40691-2

  3. [3]

     A.W. Wang, M. Du, J.X. Ni, et al., Nat. Commun. 14 (2023) 6733. doi: 10.1038/s41467-023-42542-6

  4. [4]

     R. Ji, J. Chen, T. Liu, et al., Chin. Chem. Lett. 33 (2021) 643–652.

  5. [5]

     W. Liu, Y. Chen, L. Wang, et al., Chem. Eng. J. 481 (2024) 148632. doi: 10.1016/j.cej.2024.148632

  6. [6]

     C. Li, W. Liu, X. Chen, et al., Chin. Chem. Lett. 35 (2024) 109652. doi: 10.1016/j.cclet.2024.109652

  7. [7]

     L. Peng, Y. Shang, B. Gao, et al., Appl. Catal. B: Environ. 282 (2021) 119484. doi: 10.1016/j.apcatb.2020.119484

  8. [8]

     Y. Wang, Y. Pan, H. Zhu, et al., Acta Phys.: Chim. Sin. 40 (2024) 2304050. doi: 10.3866/PKU.WHXB202304050

  9. [9]

     Q. Zhang, D. He, X. Li, et al., J. Hazard. Mater. 384 (2020) 121350. doi: 10.1016/j.jhazmat.2019.121350

  10. [10]

      C.J. Su, C.L. Li, W.H. Wang, Rare Metal. 42 (2023) 4005–4014. doi: 10.1007/s12598-023-02335-8

  11. [11]

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

  12. [12]

      M.X. Ran, H. Xu, Y. Bao, et al., Angew. Chem. Int. Ed. 135 (2023) e202303728. doi: 10.1002/ange.202303728

  13. [13]

      Y. Gao, T.W. Wu, C.D. Yang, et al., Angew. Chem. Int. Ed. 60 (2021) 22513–22521. doi: 10.1002/anie.202109530

  14. [14]

      Y. Qi, J. Li, Y. Zhang, et al., Appl. Catal. B: Environ. 286 (2021) 119910. doi: 10.1016/j.apcatb.2021.119910

  15. [15]

      D. Zhou, Z. Li, X. Hu, et al., Small 20 (2024) 2311691. doi: 10.1002/smll.202311691

  16. [16]

      J. Hou, X. He, S. Zhang, et al., Sci. Total Environ. 770 (2021) 145311. doi: 10.1016/j.scitotenv.2021.145311

  17. [17]

      W. Ding, W. Xiao, W. Huang, et al., J. Clean Prod. 257 (2020) 120457. doi: 10.1016/j.jclepro.2020.120457

  18. [18]

      M. Xu, J. Li, Y. Yan, et al., Chem. Eng. J. 369 (2019) 403–413. doi: 10.1016/j.cej.2019.03.075

  19. [19]

      Z.H. Xie, C.S. He, Y.L. He, et al., Water Res. 232 (2023) 119666. doi: 10.1016/j.watres.2023.119666

  20. [20]

      D. Manos, K. Miserli, I. Konstantinou, Catalysts 10 (2020) 1299. doi: 10.3390/catal10111299

  21. [21]

      S. Zhang, H. Zheng, P.G. Tratnyek, Nat. Water. 1 (2023) 666–681. doi: 10.1038/s44221-023-00098-1

  22. [22]

      Y. Pang, L. Kong, D. Chen, et al., J. Hazard. Mater. 384 (2020) 121447. doi: 10.1016/j.jhazmat.2019.121447

  23. [23]

      X. Zhou, M. Luo, C. Xie, et al., Appl. Catal. B: Environ. 282 (2021) 119605. doi: 10.1016/j.apcatb.2020.119605

  24. [24]

      H. Wang, Q. Gao, H. Li, et al., Chem. Eng. J. 368 (2019) 377–389. doi: 10.1016/j.cej.2019.02.124

  25. [25]

      M.C. Yao, X.J. Wu, L.L. Xu, et al., Rare Metal. 41 (2022) 972–981. doi: 10.1007/s12598-021-01832-y

  26. [26]

      D. Wang, F.X. Yin, B. Cheng, et al., Rare Metal. 40 (2021) 2369–2380. doi: 10.1007/s12598-021-01731-2

  27. [27]

      B. Barrocas, A.J. Silvestre, A.G. Rolo, et al., Phys. Chem. Chem. Phys. 18 (2016) 18081–18093. doi: 10.1039/C6CP01889K

  28. [28]

      Y. Deng, M. Xing, J. Zhang, Appl. Catal. B: Environ. 211 (2017) 157–166. doi: 10.1016/j.apcatb.2017.04.037

  29. [29]

      T. Acharya, R.N.P. Choudhary, Mater. Chem. Phys. 177 (2016) 131–139. doi: 10.1016/j.matchemphys.2016.04.005

  30. [30]

      W.C. Hung, Y.C. Chen, H. Chu, et al., Appl. Surf. Sci. 255 (2008) 2205–2213. doi: 10.1016/j.apsusc.2008.07.079

  31. [31]

      Q. Yang, H. Choi, D.D. Dionysiou, Appl. Catal. B: Environ. 74 (2007) 170–178. doi: 10.1016/j.apcatb.2007.02.001

  32. [32]

      Y. Zhang, Z. He, J. Zhou, et al., Chemosphere 307 (2022) 135681. doi: 10.1016/j.chemosphere.2022.135681

  33. [33]

      W. Xu, T. Zhang, R. Bai, et al., J. Mater. Chem. A 8 (2020) 9677–9683. doi: 10.1039/c9ta13851j

  34. [34]

      H. Li, H. Wang, Q. Gao, et al., J. Mater. Chem. A 8 (2020) 20953–20962. doi: 10.1039/d0ta06469f

  35. [35]

      J. Lee, U. von Gunten, J.H. Kim, Environ. Sci. Technol. 54 (2020) 3064–3081. doi: 10.1021/acs.est.9b07082

  36. [36]

      J. Tang, J. Wang, Environ. Sci. Technol. 52 (2018) 5367–5377. doi: 10.1021/acs.est.8b00092

  37. [37]

      K. Liu, J. Lu, Y. Ji, Water Res. 84 (2015) 1–7. doi: 10.1109/IAIN.2015.7352235

  38. [38]

      S. Qu, W. Wang, X. Pan, et al., J. Hazard. Mater. 384 (2020) 121494. doi: 10.1016/j.jhazmat.2019.121494

  39. [39]

      C. Su, R. Li, C. Li, et al., Appl. Catal. B: Environ. 310 (2022) 121330. doi: 10.1016/j.apcatb.2022.121330

  40. [40]

      Y. Nosaka, A.Y. Nosaka, Chem. Rev. 117 (2017) 11302–11336. doi: 10.1021/acs.chemrev.7b00161

  41. [41]

      H. Ma, Y. Zhang, L. Zhu, et al., Chem. Eng. J. 413 (2021) 127480. doi: 10.1016/j.cej.2020.127480

  42. [42]

      J. Li, Y. Wan, Y. Li, et al., Appl. Catal. B: Environ. 256 (2019) 117–126.

  43. [43]

      S.K. Rani, D. Easwaramoorthy, I.M. Bilal, et al., Appl. Catal. A: Gen. 369 (2009) 1–7. doi: 10.1016/j.apcata.2009.07.048

  44. [44]

      F. Pan, H. Ji, P. Du, et al., J. Hazard. Mater. 402 (2021) 123779. doi: 10.1016/j.jhazmat.2020.123779

  45. [45]

      J. Ma, L. Chen, Y. Liu, et al., J. Hazard. Mater. 418 (2021) 126180. doi: 10.1016/j.jhazmat.2021.126180

  46. [46]

      H. Xu, N. Jiang, D. Wang, et al., Appl. Catal. B: Environ. 263 (2020) 118350. doi: 10.1016/j.apcatb.2019.118350

• 

  Figure 1  Material characterizations of CoTiO₃. (a) XRD pattern (inset shows the crystal structure of CoTiO₃). High-resolution XPS spectra of (b) Co 2p, (c) Ti 2p and (d) O 1s. (e) FTIR spectra. (f) Raman spectra. (g) SEM images. (h) HRTEM image (insets show TEM image and SAED pattern).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) Degradation efficiencies of HCQ and (b) the first order kinetic reaction rate constant in first 5 min in different systems. (c) Total organic carbon (TOC) removal of HCQ in CoTiO₃/PMS system. (d) Cycle experiments of CoTiO₃/PMS conducted with recycled CoTiO₃. (e) The influence of initial pH on HCQ degradation in CoTiO₃/PMS system. (f) Degradation of different pollutants in CoTiO₃/PMS system.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) LSV curves of CoTiO₃ with/without PMS and HCQ. EPR spectra of (b) SO₄^•−, ^•OH and (c) O₂^•− in CoTiO₃/PMS system using DMPO trapping agent at different time intervals. (d) Removal efficiency and (e) inhibition ratio of HCQ degradation in the presence of different quenchers and (f) the corresponding quantitative concentration of SO₄^•− and ^•OH in CoTiO₃/PMS system using DMPO trapping agent at different time intervals.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Adsorption and activation of HSO₅⁻. (b) Gibbs free energy (△G) of different states of HSO₅⁻ on CoTiO₃ crystal. (c) Analyse of Bader charge before and after activation.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  XPS spectra of the fresh CoTiO₃ and used CoTiO₃ after 5 degradation cycles: (a) Co 2p, (b) Ti 2p and (c) O 1s. (d) The CV curve of CoTiO₃. (e) Schematic illustration of PMS activation for HCQ removal.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Schematic diagram of the preparation of CoTiO₃–CNT–PVDF membrane. (b) HCQ degradation efficiency of CoTiO₃–CNT–PVDF membrane. (c) Comparison of CoTiO₃–PVDF–CNT membrane/PMS system with previously developed catalytic membranes in Table S2 (Supporting information). (d) XRD patterns and (e) SEM images of CoTiO₃–CNT–PVDF membrane before and after reaction.

  下载: 全尺寸图片 幻灯片
•