Efficient photodegradation of 2-chloro-4-nitrophenol over Fe-doped BiOCl nanosheets with oxygen vacancy†

Haoyuan Wu ^a, Xiangming Liu ^b, Hua Xu *^b, Xinmin Yang ^a and Jinhua Ye *^ac
aTJU-NIMS International Collaboration Laboratory, School of Materials Science and Engineering, Tianjin University, Tianjin 300072, P.R. China
^bSchool of Chemistry and Environmental Engineering, Wuhan Institute of Technology, No. 206, Guanggu 1st Road, Donghu New & High Technology Development Zone, Wuhan 430205, P.R. China. E-mail: XU.Hua@wit.edu.cn
^cInternational Center for Materials Nanoarchitectonics (WPI-MANA), National Institute for Materials Science (NIMS), 1-1 Namiki, Tsukuba 305-0044, Japan. E-mail: Jinhua.YE@nims.go.jp

First published on 11th June 2021

---

---

1. Introduction

Photodegradation of organic pollutants has emerged as one of the promising routes to convert persistent organic pollutants into more biologically degradable and less toxic substances.^1–3 It is well-known that the photocatalytic reaction is a free radical chain reaction,^4,5 where the holes and electrons generated from the excited semiconductor could activate either OH/H₂O or O₂ into reactive oxygen species (˙OH/H₂O₂ or ˙O₂⁻) to degrade organic contaminants.^6–8 Thus, promoting the transformation of reactive oxygen species (ROS) is essential for the participation of large number of active free radicals in heterogeneous photocatalytic reactions.^9,10

Remarkably, being a type of 2D material, more attention was paid on the surface oxygen vacancies (OVs) existing on the BiOCl plates covered with the active (001) facet.^11–15 Previous studies demonstrated that OVs on BiOX (X = F, Cl, Br, and I) could benefit CO₂ adsorption and activation.^16,17 It has been reported that the vacancy associates existing on the surface of ultrathin BiOCl nanosheets promote the adsorption of organic pollutants and charge separation of electron–hole pairs.¹⁸ The BiOCl/BiOBr heterojunction with rich OVs possessed a more negative defect energy level than E⁰ (O₂/˙O₂⁻) = −0.33 V (vs. NHE); thus, more photogenerated electrons trapped by defect states could reduce O₂ into ˙O₂⁻ for effective carbamazepine photodegradation.¹⁹ Moreover, OVs serve as active sites not only for the direct activation of adsorbed O₂ to reactive oxygen species (such as ˙O₂⁻)²⁰ but also may induce the surface Fenton reaction based on its electron-donor nature for the H₂O₂ dissociation into ˙OH,²¹ which endows favourable ambience for the photodegradation of phenolic compounds.^22,23 To achieve more ˙OH with high oxidation potential (E⁰ = 1.8–2.7 V vs. NHE),²⁴ our previous study built a solid–gas interfacial Fenton reaction over Fe–C₃N₄, with the Fe³⁺ decorated surface acting as a trigger for H₂O₂ conversion into ˙OH, thus resulting in high apparent quantum yield upon the photooxidation of isopropanol.²⁵ The Fe-doped Bi/BiOBr heterojunction with H₂O₂ in the reaction system also proved that a similar Fenton-like reaction would occur to produce more ˙OH for the rhodamine B photodegradation.²⁶

Herein, we introduce the Fe dopant in the OV-associated BiOCl nanosheets (Fe-BOC) to build Fenton-like catalysts without the extra addition of H₂O₂. 2-Chloro-4-nitrophenol, which is affiliated to nitrophenol types with photostability and high toxicity, was selected as a model target organic pollutant. In contrast to the undoped BiOCl nanosheets (BOC), the photo-degradation tempo of 2-C-4-NP by Fe-BOC was elevated by 3.1 times upon the production of more ˙OH radicals via a superficial photo-induced Fe²⁺ reaction with H₂O₂. Moreover, the better charge separation of Fe-BOC also resulted in more surface e⁻ reacting with O₂ to produce ˙O₂⁻ involved in the 2-C-4-NP photodegradation. This study demonstrates that creating more oxygen vacancies over BiOCl together with functionalized Fe-doping could largely improves the photodegradation efficiency of organic contaminants.

2. Experimental section

2.1 Synthesis of Fe-doped BiOCl nanosheets

Fe-Doped BiOCl nanosheets were fabricated via a facile solvothermal treatment with slight modification of the BiOCl nanosheet, as reported in previous literature.¹⁸ Typically, 1 mmol of Bi(NO₃)₃·5H₂O and 400 mg (average mol wt 360000) of polyvinylpyrrolidone (PVP) were first dissolved in 25 ml of a mannitol solution (0.1 M) under magnetic stirring, and then, 5 ml of a saturated NaCl solution was added dropwise, followed by addition of 0.04 mmol of Fe(NO₃)₃·9H₂O. The resulting solution was subsequently transferred to a 50 ml Teflon autoclave. After the reaction at 160 °C for 3 h, the precipitate was collected and washed with ethanol and distilled water, and then dried at 70 °C for 12 h in the oven. The obtained sample was denoted as Fe-BOC. The dopant concentration of Fe was expected to be 4 mol%. The BiOCl nanosheets with no Fe doping (denoted as BOC) were prepared using a similar procedure but without the addition of Fe(NO₃)₃·9H₂O. Fe-Doped BiOCl nanosheets with various molar ratios of Fe and Bi of 0.02, 0.04, 0.06, and 0.10 were denoted as Fe_0.02-BOC, Fe_0.04-BOC (Fe-BOC), Fe_0.06-BOC and Fe_0.10-BOC, respectively.

2.2 Photodegradation of 2-C-4-NP

The photodegradation test was carried out using 2-C-4-NP as the target organic pollutant. 50 ml of the aqueous suspension containing 10 mg L⁻¹ of 2-C-4-NP and 30 mg of the catalyst was placed in a quartz glass beaker. A 300 W Xe arc lamp (the light intensity was measured to be 300 mW cm⁻²) was utilized as a light source, and a 45 × 45 mm cooling water circulation shutter window was placed between the light and the reaction cell to avoid the heat effects. Prior to irradiation, the suspension was magnetically stirred in dark for 30 min to get adsorption equilibrium. After light irradiation, 2 ml of the suspension was sampled every 10 min. The scavenger tests were performed as the above procedure with certain amounts of different scavengers: 3 mM of triethanolamine (TEOA); 3 mM of tert-butanol (TBA); 100 μL of 2500 units per mL superoxide dismutase (SOD); 100 μL of 2500 units per mL catalase (CAT). The concentration of 2-C-4-NP was detected by a 1220 Infinity HPLC (Agilent, America) fitted with an Agilent ZORBAX SB-C18 Column (150 mm × 4.6 mm × 5 μm) maintained at 30 °C and the VWD detector was set at 280 nm; the mobile phase was adopted with a methanol:water (v/v, 60:40) mixture and the flow rate was kept at 0.7 ml min⁻¹. The total organic carbon (TOC) of the reaction solution was determined by a 1030 W TOC analyzer (OI Analytical, America). The intermediate products were detected by a 8860 gas chromatograph (Agilent, America) equipped with a HP-5MS UI capillary column (30 m × 0.25 mm × 0.25 μm film thickness), and a 5977B MSD mass spectrometer was used as detector. The temperature program was set to 50 °C for 4 min and then ramped to 160 °C at 5 °C min⁻¹ finally going up to 260 °C at 10 °C min⁻¹. To identify the intermediate compounds, the solution obtained after the photocatalytic reaction was freeze-dried and redissolved in 0.4 ml of anhydrous pyridine with 0.2 ml hexamethyldisilazane and 0.1 ml chlorotrimethylsilane, and then shaken vigorously for complete silylation. After centrifugation, the obtained sample was ready for detection.

2.3 Photoelectrochemical measurement

BOC or Fe-BOC anodes were prepared via the spin-coating method. 10 mg of sample was dispersed into 1 ml of ethanol with 50 μL of naphthol as an agglomerant, and then the above suspension was uniformly dispersed on an FTO substrate and dried at 70 °C in the oven. Photoelectrochemical measurements were performed on a CHI electrochemical station (CHI 660e) using the three-electrode mode with 0.5 M Na₂SO₄ as the electrolyte. The BOC/Fe-BOC photoanodes, Pt wire, and Ag/AgCl electrode were used as the working electrode, counter electrode, and reference electrode, respectively. The illumination source was a 300 W Xe arc lamp. The photocurrent curves were recorded under a bias potential of −0.2 V. Electrochemical impedance spectra were recorded at an applied potential of 0.2 V over the frequency range of 1 MHz–0.01 Hz. Current–potential (I–V) curves were recorded via the linear-sweep-voltammetry (LSV) mode, and the voltage-scan speed was 10 mV s⁻¹.

2.4 Characterization

X-ray diffraction (XRD) patterns were recorded on a D8 Advanced X-ray diffractometer (Bruker, Germany). The morphology was observed on a field-emission transmission electron microscope (JEM-F200, JEOL, Japan). UV-visible diffuse reflectance spectra were obtained on a UV-visible spectrophotometer (UV-2700, Shimadzu, Japan), and then converted into absorption spectra via Kubelka-Munk transformation. X-ray photoelectron spectroscopy (XPS) was performed on an Escalab 250Xi (Thermo SCIENTIFIC, America). All binding energies were corrected with C 1s peaks of contaminant carbon at 284.8 eV. Electron spin resonance spectra were recorded on a JES-FA 200 (JEOL, Japan) spectrometer with a microwave power of 1.0 mW and modulation frequency of 100 KHz. Photoluminescence (PL) spectra were recorded on a fluorescence spectrophotometer (Fluorolog-3, HORIBA Scientific, America).

2.5 Analytical method

The amount of H₂O₂ was detected via the N,N-diethyl-p-phenylenediamine sulfate (DPD)/peroxidase (POD) method.²⁷ Briefly, 50 mg of DPD was dissolved in 5 ml of 0.05 M H₂SO₄, denoted as solution A. 10 mg of POD was dissolved in 10 ml of H₂O, denoted as solution B. 15 ml of 0.5 M K₂HSO₄ solution and 85 ml of 0.5 M KH₂SO₄ solution were mixed together to obtain solution C. 2 ml of the filtered solution to be assayed was added into a mixture, which was composed of 2.2 ml of H₂O, 0.1 ml of solution A, 0.1 ml of solution B, and 0.8 ml of solution C. The concentration of H₂O₂ was monitored on a UV-vis spectrometer at 551 nm. The concentration of ˙OH was detected by the PL method using coumarin (COU) (0.1 mM) as the probe, and the excitation wavelength was set at 332 nm.²⁸ The after-reaction solution was filtered to record the PL intensity at 456 nm for every 10 min-light irradiation.

3. Results and discussion

As shown in Fig. 1, the XRD patterns of the as-prepared samples could be indexed to BiOCl with a tetragonal structure (JCPDS No. 6-249). The diffraction peaks at 11.9, 25.8, 32.5, 33.4, 40.9, 46.6, 54.1, 58.6 and 68.1° corresponded to the (001), (101), (110), (102), (112), (200), (211), (212) and (220) planes of BiOCl. It is noteworthy to mention that the XRD patterns did not show any clear change after Fe doping. The morphology of Fe-BOC was observed by TEM images, as shown in Fig. 2a. Fe-BOC consisted of squarish nanosheets with sizes in the 20–100 nm range. The high-resolution TEM (HRTEM) image (Fig. 2b) shows the distance of the lattice fringe to be 0.275 nm, corresponding to the lattice spacing of the (110) atomic plane of BiOCl. The related fast Fourier transformation (FFT) revealed the angle between (110) and (200) planes to be 45°, indicating the [001] zone axis of tetragonal BiOCl. Moreover, the energy-dispersive X-ray spectroscopy (EDS) elemental mapping analysis (Fig. 2c–f) shows that the distribution of Fe was homogeneous over the BiOCl nanosheets.

The optical properties of the as-prepared samples are shown in Fig. 3. UV-visible absorption spectra reveal that the absorption edges of BOC and Fe-BOC were similar at approximately 370 nm. For BiOCl as an indirect-gap semiconductor, based on the Tauc equation⁸ of αhυ = A(hυ − E_g),² in which α, hυ, A and E_g are the absorption coefficient, incident photon energy, proportionality constant, and band gap, respectively. Plotting a tangent line from the absorption spectra, the bandgaps of BOC and Fe-BOC were deduced to be 3.2 and 3.15 eV, respectively (inset of Fig. 3).

XPS was further applied to study the valence states of elements. As shown in Fig. 4a, the peaks of Bi 4f_7/2 and 4f_5/2 for BOC are located at 158.9 and 164.2 eV, respectively, with a spin-orbital doublet splitting of 5.3 eV, indicating the oxidation state of Bi³⁺ in BOC.²⁹ These peaks shifted to a lower binding energy by 0.2 eV for Fe-BOC, which might have resulted from the electron transfer occurring from Fe to Bi; thus, the density of electron clouds around Bi increased.³⁰ There was no obvious change for either O 1s or Cl 2p XPS spectra after Fe doping (Fig. 4b and c), the O 1s peak at 529.6 eV was assigned to surface lattice oxygen (O–Bi), the other peaks at around 531.6 and 533.0 eV represented the oxygen vacancy related ˙O₂⁻ and surface OH group, respectively.³¹ As shown in Fig. 4d, the peaks at 724 and 711 eV (with a binding energy difference between the two peaks of 13 eV) for Fe-BOC were indexed to Fe 2p_3/2 and 2p_1/2 in its ferric valence state, respectively, while other peaks at 723.0 and 708.5 eV were assigned to Fe²⁺.³² EPR spectra (Fig. S2†) reveal that Fe-BOC exhibit a similar g value = 2.1 to that of the standard Fe³⁺ sample, further confirming the existence of Fe³⁺ in Fe-BOC.³³

Next, the photodegradation test was carried out over both BOC and Fe-BOC adopting 2-C-4-NP as the target organic pollutant. As shown in Fig. 5a and S6,† the photodegradation of 2-C-4-NP can approximately be deemed to be a first-order reaction. The photodegradation kinetic constants were calculated to be 0.011, 0.025, 0.035, 0.031, and 0.029 for BOC, Fe_0.02-BOC, Fe_0.04-BOC (Fe-BOC), Fe_0.06-BOC, and Fe_0.10-BOC, respectively (Fig. S9†). Among which, Fe_0.04-BOC (Fe-BOC) with the optimized Fe amount exhibited 3.1 times enhanced photocatalytic activity in contrast to that of BOC.

Reactive oxygen species (ROS) such as ˙OH, ˙O₂⁻ and H₂O₂, play important roles in the degradation of organic pollutants. Herein, triethanolamine (TEOA), tert-butanol (TBA), superoxide dismutase (SOD), and catalase (CAT) were adopted as the scavengers of h⁺, ˙OH, ˙O₂⁻, and H₂O₂, respectively. As shown in Fig. 5b, for BOC, neither ˙O₂⁻ nor H₂O₂ contributed greatly on the 2-C-4-NP photodegradation. After the addition of TBA for ˙OH catching, the photocatalytic activity of BOC dropped by 17%. A further h⁺ scavenger experiment showed that the 2-C-4-NP degradation would be inhibited by nearly 70% without h⁺ for BOC. For Fe-BOC, the photocatalytic activity dropped to 42%, 47%, 66% and 90% with the addition of the scavenger of h⁺, ˙OH, H₂O₂ and ˙O₂⁻, respectively. In contrast to BOC, the contribution of ˙OH and H₂O₂ on the photoactivity of Fe-BOC was much larger, which might have resulted from the Fenton-like reaction aroused by the Fe dopant. In the Fe-related Fenton-like reaction, the existence of H₂O₂ generated during the photocatalytic reaction over Fe-BOC, dissociates and produces the highly reactive oxygen species ˙OH for the 2-C-4-NP oxidation.

The Fenton-like reaction will occur when the dopant Fe coexists with H₂O₂. In the photocatalytic reaction, one e⁻ will reduce O₂ into ˙O₂⁻ (eqn (2)), while ˙O₂⁻ will further be reduced to H₂O₂ by another e⁻ (eqn (3)). For Fe-BOC, Fe³⁺ could be reduced by e⁻ into Fe²⁺ (eqn (4)). Thus, the coexistence of Fe²⁺ and H₂O₂ would generate ˙OH (eqn (5)). We then quantitatively investigated the amount of H₂O₂ and ˙OH in the photoreaction system. As shown in Fig. 6a, the concentration of H₂O₂ in the Fe-BOC-based system was lower than that in BOC, while 3 times more ˙OH was detected for Fe-BOC compared to that of BOC (Fig. 6b), which might have resulted from the quick dissociation of H₂O₂ into ˙OH over Fe-BOC via the Fenton-like reaction as shown in Scheme 1. In addition, as shown in Fig. 6c and d, the typical EPR signals assigned to ˙OH and ˙O₂⁻ were also detected. Clearly, with the increase in the photoreaction time, both the concentration of ˙OH and ˙O₂⁻ increased, and the reactive oxygen species (˙OH and ˙O₂⁻) in Fe-BOC were always much more than that of BOC.

The intermediate products during the photocatalytic reaction were investigated to study the reaction pathway of the 2-C-4-NP photo-degradation in either BOC or Fe-BOC. The degradation and mineralization of 2-C-4-NP, which is a typical example of highly toxic nitrophenols (NPs), include two vital segments for nitro oxidation and dechlorination to harmless and environmentally friendly products.^34,35 The detected compounds are listed in Fig. S13;† thus, the proposed reaction pathway is shown in Fig. 7. Initially, the nitro group (–NO₂) of the 2-C-4-NP molecule is attacked by the ˙OH radical to generate chlorohydroquinone (CHQ). Then, the chloro group (–Cl) of CHQ is released by the ˙O₂⁻ radical, resulting in the formation of hydroquinone (HQ). Thereafter, the conjugated structure of the benzene ring is broken into small molecule acids and alcohols, such as 2-butenedioic acid, lactic acid, acetic acid and ethylene glycol.

	(1)

	(2)

˙O2− + H+ + e− → H2O2	(3)

	(4)

Fe2+ + H2O2 → Fe3+ + ˙OH + OH−	(5)

To further investigate the charge separation properties of the samples, photoluminescence spectra were recorded. The PL intensity of Fe-BOC was lower than that of BOC (Fig. 8a), indicating a better charge separation of the former. In addition, the dynamic PL spectra (Fig. 8b) recorded under a laser light excitation of 405 nm revealed that the decay times of the carriers in Fe-BOC were shorter than that in BOC. As shown in Fig. 8c, the EIS arc of BOC was larger than that of Fe-BOC under light irradiation, suggesting Fe-BOC exhibited a faster interfacial charge transfer, further leading to the enhanced photocurrent (Fig. 8d). All the above-mentioned results suggest that Fe-BOC exhibited better charge separation ability; thus, more e⁻ could migrate to the surface and react with the O₂ adsorbed in the OV sites to produce more ˙O₂⁻, which is well in accordance with our EPR test (Fig. 6d).

Accordingly, we can conclude that the enhanced photocatalytic activity of Fe-BOC arose from the much more generated reactive oxygen species (˙OH and ˙O₂⁻). Upon light excitation, h⁺ and e⁻ are generated over Fe-BOC, and then h⁺ could either oxidize the organic pollutant 2-C-4-NP directly or oxidize H₂O into reactive species such as the ˙OH radical, while e⁻ reduced O₂ into ˙O₂⁻ or H₂O₂. With the existence of dopant Fe, a Fenton-like reaction would occur to produce more ˙OH radicals, which contributes greatly on the 2-C-4-NP degradation. Moreover, the oxygen vacancy over BiOCl would be beneficial for O₂ adsorption; thus, Fe-BOC has much better charge separation, which motivates more e⁻ to migrate to the surface and activate O₂ into˙O₂⁻ for the dechlorogenation of 2-C-4-NP.

4. Conclusion

In this study, Fe-doped BiOCl nanosheets were fabricated for the efficient photodegradation of 2-C-4-NP. With the optimization of the amount of Fe, Fe-BOC exhibited 3.1 times higher activity compared to BOC. The enhanced activity was attributed to the dopant Fe induced Fenton-like reaction and better charge separation, which aroused more efficient O₂ activation, producing much more reactive species (˙OH and ˙O₂⁻) involved in the photodegradation process. This study provides a facile way to promote numerous reactive oxygen species for more effective photodegradation of organic pollutants.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by Science Foundation of Wuhan Institute of Technology (Grant Number K201962) and National Natural Science Foundation of China (Grant Number 21633004).

Notes and references

1. C. Wang, J. Li, X. Lv, Y. Zhang and G. Guo, Energy Environ. Sci., 2014, 7, 2831–2867 RSC .
2. L. Ye, Y. Su, X. Jin, H. Xie and C. Zhang, Environ. Sci.: Nano, 2014, 1, 90 RSC .
3. J. Levec and A. Pintar, Catal. Today, 2007, 124, 172–184 CrossRef CAS .
4. J. Zhao, C. Chen and W. Ma, Top. Catal., 2005, 35, 269–278 CrossRef CAS .
5. S. Ahmed, M. G. Rasul, W. N. Martens, R. Brown and M. A. Hashib, Desalination, 2010, 261, 3–18 CrossRef CAS .
6. Y. Nosaka and A. Y. Nosaka, Chem. Rev., 2017, 117, 11302–11336 CrossRef CAS PubMed .
7. L. Yu, X. Zhang, G. Li, Y. Cao, Y. Shao and D. Li, Appl. Catal., B, 2016, 187, 301–309 CrossRef CAS .
8. C. Wang, X. Zhang, H. Qiu, G. Huang and H. Yu, Appl. Catal., B, 2017, 205, 615–623 CrossRef CAS .
9. S. G. Schrank, H. J. José, R. F. P. M. Moreira and H. F. Schröder, Chemosphere, 2005, 60, 644–655 CrossRef CAS PubMed .
10. M. Klavarioti, D. Mantzavinos and D. Kassinos, Environ. Int., 2009, 35, 402–417 CrossRef CAS PubMed .
11. C. Mao, H. Cheng, H. Tian, H. Li, W. Xiao, H. Xu, J. Zhao and L. Zhang, Appl. Catal., B, 2018, 228, 87–96 CrossRef CAS .
12. S. Wu, J. Xiong, J. Sun, Z. D. Hood, W. Zeng, Z. Yang, L. Gu, X. Zhang and S. Z. Yang, ACS Appl. Mater. Interfaces, 2017, 9, 16620–16626 CrossRef CAS PubMed .
13. Q. Wang, Z. Liu, D. Liu, W. Wang, Z. Zhao, F. Cui and G. Li, Chem. Eng. J., 2019, 360, 838–847 CrossRef CAS .
14. L. Ye, X. Jin, Y. Leng, Y. Su, H. Xie and C. Liu, J. Power Sources, 2015, 293, 409–415 CrossRef CAS .
15. Q. Wang, W. Wang, L. Zhong, D. Liu, X. Cao and F. Cui, Appl. Catal., B, 2018, 220, 290–302 CrossRef CAS .
16. X. Ren, M. Gao, Y. Zhang, Z. Zhang, X. Cao, B. Wang and X. Wang, Appl. Catal., B, 2020, 274, 119063 CrossRef CAS .
17. J. Zhao, Z. Miao, Y. Zhang, G. Wen, L. Liu, X. Wang, X. Cao and B. Wang, J. Colloid Interface Sci., 2021, 593, 231–243 CrossRef CAS PubMed .
18. M. Guan, C. Xiao, J. Zhang, S. Fan, R. An, Q. Cheng, J. Xie, M. Zhou, B. Ye and Y. Xie, J. Am. Chem. Soc., 2013, 135, 10411–10417 CrossRef CAS PubMed .
19. G. Liu, H. Xu, D. Li, Z. Zou, Q. Li and D. Xia, Eur. J. Inorg. Chem., 2019, 2019, 4887–4893 CrossRef CAS .
20. U. Aschauer, J. Chen and A. Selloni, Phys. Chem. Chem. Phys., 2010, 12, 12956–12960 RSC .
21. H. Li, J. Shang, Z. Yang, W. Shen, Z. Ai and L. Zhang, Environ. Sci. Technol., 2017, 51, 5685–5694 CrossRef CAS PubMed .
22. S. Sun and A. T. Lemley, J. Mol. Catal. A: Chem., 2011, 349, 71–79 CrossRef CAS .
23. D. Sánchez-Rodríguez, M. G. Méndez Medrano, H. Remita and V. Escobar-Barrios, J. Environ. Chem. Eng., 2018, 6, 1601–1612 CrossRef .
24. A. Y. Zhang, T. Lin, Y. Y. He and Y. X. Mou, J. Hazard. Mater., 2016, 311, 81–90 CrossRef CAS .
25. Y. Li, S. Ouyang, H. Xu, X. Wang, Y. Bi, Y. Zhang and J. Ye, J. Am. Chem. Soc., 2016, 138, 13289–13297 CrossRef CAS .
26. X. Geng, W. Li, F. Xiao, D. Wang and L. Yang, Catal. Sci. Technol., 2017, 7, 658–667 RSC .
27. H. Bader, V. Sturzenegger and J. Hoigné, Water Res., 1988, 22, 1109–1115 CrossRef CAS .
28. Q. Xiang, J. Yu and P. K. Wong, J. Colloid Interface Sci., 2011, 357, 163–167 CrossRef CAS PubMed .
29. H. Xing, H. Ma, Y. Fu, X. Zhang, X. Dong and X. Zhang, J. Renewable Sustainable Energy, 2015, 7, 063120 CrossRef .
30. N. Zhang, L. Li, Q. Shao, T. Zhu, X. Huang and X. Xiao, ACS Appl. Energy Mater., 2019, 2, 8394–8398 CrossRef CAS .
31. H. Li, J. Shang, H. Zhu, Z. Yang, Z. Ai and L. Zhang, ACS Catal., 2016, 6, 8276–8285 CrossRef CAS .
32. G. Bhargava, I. Gouzman, C. M. Chun, T. A. Ramanarayanan and S. L. Bernasek, Appl. Surf. Sci., 2007, 253, 4322–4329 CrossRef CAS .
33. S. K. Misra, S. I. Andronenko, A. Thurber, A. Punnoose and A. Nalepa, J. Magn. Magn. Mater., 2014, 363, 82–87 CrossRef CAS .
34. Z. Xiong, H. Zhang, W. Zhang, B. Lai and G. Yao, Chem. Eng. J., 2019, 359, 13–31 CrossRef CAS .
35. G. Anuradha, K. Meenu, C. Archana, T. Masahiro, K. C. Asit and K. J. Rakesh, Environ. Sci. Technol., 2010, 44, 1067–1077 Search PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1cy00807b

This journal is © The Royal Society of Chemistry 2021