Ferrocenylselenoether and its cuprous cluster modified TiO₂ as visible-light photocatalyst for the synergistic transformation of N-cyclic organics and Cr(VI)

Abstract

In this study, fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ nanosystems based on ferrocenylselenoether and its cuprous cluster were developed and characterized by X-ray photoelectron spectroscopy (XPS), high-resolution transmission electron microscopy (HR-TEM), energy dispersive X-ray spectroscopy (EDX), and electron paramagnetic resonance (EPR). Under optimized conditions, 0.2 g L⁻¹ catalyst, 20 mM H₂O₂, and initial pH 7, good synergistic visible light photocatalytic tetracycline degradation and Cr(VI) reduction were achieved, with 92.1% of tetracycline and 64.5% of Cr(VI) removal efficiency within 30 minutes. Mechanistic studies revealed that the reactive species ˙OH, ˙O₂⁻, and h⁺ were produced in both systems through the mutual promotion of Fenton reactions and photogenerated charge separation. The [Cu₂I₂(fcSe)₂]_n@TiO₂ system additionally produced ¹O₂ from Cu⁺ and ˙O₂⁻. The advantages of the developed nanosystems include an acidic surface microenvironment provided by Se⋯H⁺, resourceful product formation, tolerance of complex environments, and excellent adaptability in refractory N-cyclic organics.

---

Introduction

Nitrogen-containing cyclic organics, N-cyclic organics in brief, are among the most common functional compounds in biological, pharmaceutical, and material fields.^1,2 With the rapid development of industry and global consumption, the N-cyclic organics are continuously discharged into the environment.^3,4 Due to their high stability and high chemical oxygen demand, it is difficult to rely solely on degradation by microorganisms under natural conditions.⁵ These N-cyclic organics have become major harmful pollutants in soil and aquatic ecosystems, which have toxic effects on living organisms, including teratogenesis, carcinogenesis, and mutagenesis.^6–9 Simultaneously, the highly toxic and non-biodegradable Cr(VI) is one of the most coexisting toxic metals and can cause severe threats to the natural environment and public health.^10–12 It is very significant to develop environmentally friendly technologies to simultaneously transform N-cyclic organics and the coexisting toxic Cr(VI) from complex ecosystems.

The common methods available to treat N-cyclic organics in the environment include chemical oxidation,^13–16 biodegradation¹⁷ and physical adsorption.¹⁸ As an approach of advanced oxidation processes (AOPs), Fenton chemistry has gradually shown the prominent prospect of industrial application in wastewater treatment.^5,19,20 In Fenton chemistry methodology, organic pollutants can be degraded by highly oxidative hydroxyl radicals (˙OH) produced by Fe(II) and H₂O₂ reactions. To solve the slow Fe²⁺ recovery that inhibits the efficiency of Fenton chemistry, the efficient redistribution of local electrons in the catalyst system and/or the excess electrons generated under an external field play an important role.²⁰ Recently, the design of heterogeneous photo-Fenton system based on semiconductors has witnessed rapid development.²¹ Ding and Wang demonstrated a nanocomposite catalyst, Fe₃O₄@β-CD/g-C₃N₄, that enriched the photoinduced electrons to promote Fe(II) regeneration on the catalyst surface.²² Zhu et al. designed the electron self-sufficient core–shell BiOCl@Fe(III)–BiOCl nanosheet photocatalyst to promote the Fe(III)/Fe(II) recycling in the photo-Fenton reaction.²³ In this regard, the most reported photo-Fenton systems used inorganic iron salts as Fenton reagents, which suffer from low stability, iron sludge formation, and strict acidic pH limitation. Also, the charge-separation efficiency is another key point for enhancing the photocatalytic performance. It is imminent to develop novel photocatalytic nanoplatforms with features, such as suitable catalysts with stable active metal sites, environmental benignity, high adaptability, and resource regeneration.

Ferrocene is a sandwich-shaped organometallic compound containing ferrous ion, owning the advantages of kinetic stability, easy modification, biocompatibility, and redox capability. Ferrocenyl compounds have been confirmed as efficient Fenton reactors,^24–26 and the visible absorption bands of ferrocene allow the photo-promoted Fe(II)–Fe(III) redox pair recycle through the Fenton reaction to improve ˙OH release.^27,28 Alternatively, stable copper(I) clusters based on selenoethers can effectively enhance photocatalytic performance.^29–31 These systematic studies have prompted us to further investigate the applications of ferrocenyl derivatives in waste water treatment. In the current study, we sensitized TiO₂ with ferrocenylselenoether fcSe and its cuprous iodide cluster [Cu₂I₂(fcSe)₂]_n as heterogeneous photocatalysts (Scheme 1). Choosing Tetracycline (TC) and Cr(VI) as template reactions, the obtained fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ were applied in the visible light photo-Fenton simultaneous efficient transformation at neutral aqueous solution. The two nanosystems were further proved to possess excellent adaptability in the degradation of seven representative N-cyclic organic contaminants, that are, antibiotics Ciprofloxacin and imidazole, textile dyes Toluidine Blue, Methyl Orange, Methylene Blue, Malachite Green, and Basic Violet. The innovations of two photocatalytic nanosystems, fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ in this study are: (1) Fenton reaction inhibited the recombination of the photogenerated charges; (2) the effective recycling between ferrocene ([Fe²⁺(η⁵-C₅H₅)₂]) and ferrocenium ([Fe³⁺(η⁵-C₅H₅)₂]) pumped by photoexcited TiO₂; (3) the incorporation of selenium atoms in nanosystems lead to an acidic microenvironment for efficient Fenton reaction in neutral aqueous solution.

Scheme 1 (a) In situ preparation and (b) photocatalytic transformation study of N-cyclic organics and Cr(VI) of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n @TiO₂.

Experimental section

Materials

All starting materials were analytical-grade reagents and purchased from Aladdin or Source Leaf and used without further purification unless otherwise specified. TiO₂ was commercial P25 (75% anatase, 25% rutile). 1,2,3-Triselena[3]ferrocenophane fcSe₃ (fc = [Fe(η⁵-C₅H₄)(η⁵-C₅H₄)]) was prepared according to the literature methods.³²

Characterization

¹H NMR spectra were recorded on a Bruker DRX spectrometer (400 MHz) at room temperature. The ESI-mass spectra were determined by a Micromass Quattro II triple-quadrupole mass spectrometer and analyzed using the MassLynx software. The single-crystal X-ray diffraction measurement was performed using a Bruker SMART APEX II CCD diffractometer. Transmission electron microscopy (TEM, Hitachi HT7700), high-resolution transmission electron microscopy (HRTEM, JEOL 2100F), and energy-dispersive X-ray (EDX) spectroscopic analysis were performed to gain the structure information. The elemental analysis and oxidation state study of the samples were carried out by X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe), and the binding energies were corrected for specimen charging effects using the C 1s level at 284.8 eV as the reference. UV-vis diffuse reflectance spectroscopy (UV-VIS DRS) was performed on a PerkinElmer Lambda 950 UV/vis/nir spectrophotometer with BaSO₄ as a reference. FT-IR spectroscopy was performed using a Fourier transform infrared spectrometer with KBr as a reference. Inductively coupled plasma emission spectrometer (ICP) experiments were performed on an Agilent 7900 ICP-MS. Brunauer–Emmett–Teller (BET) results were obtained on an ASAP 2460. Electron paramagnetic resonance (EPR) analysis was performed using Bruker A300. Malvern Zetasizer Nano ZS90 was used to determine the zeta potentials and the hydrodynamic diameters of liposomes.

The visible light source is a Xenon Lamp MC-PF300C coupled with a power meter (MC-PM100C) for photo-Fenton reaction, purchased from Beijing Merry Change Technology Co., Ltd, and equipped with a 420 nm filter to filter out UV light. Gaseous products in photocatalytic degradation were measured by gas chromatography (GC-9790plus, FULI) with a flame ionization detector (FID) and a thermal conductivity detector (TCD). The liquid products were measured by ion chromatography (Dionex Aquion Rfic), LC-MS (MSQ PLUS/U3000), and HPLC (Agilent 1260).

Preparation of L0, fcSe, fcSe@TiO₂, and [Cu₂I₂(fcSe)₂]_n@TiO₂

As shown in Scheme 2a, 1,1′-bis((4-carboxybenzyl)seleno)ferrocene fcSe was synthesized via a similar route described by us previously.²⁷ The detailed synthetic procedures, ¹H NMR and MS spectra of the compounds, crystal structure and crystal data and structure refinement details of L0 and fcSe are provided in ESI (Fig. S1–S5 and Tables S1, S2†). The data for L0 (CCDC 2216502) is available from the Cambridge Crystallographic Data Center via https://www.ccdc.cam.ac.uk/.

Scheme 2 Synthetic routes of (a) fcSe, (b) fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂.

The stepwise in situ self-assembly of fcSe and [Cu₂I_2(fcSe)2]_n on the TiO₂ P25 surface as shown in Scheme 2b.

fcSe@TiO₂. The TiO₂ P25 (0.07 g) was dispersed in 50 mL dichloromethane solution of fcSe (0.2 g, 0.34 mmol). After sonication for 30 min, the mixture was refluxed for 24 h. The precipitate was collected by centrifugation, then washed with dichloromethane (3 × 10 mL) and dried under vacuum to obtain fcSe@TiO₂.

[Cu₂I₂(fcSe)₂]_n@TiO₂. CuI (0.07 g, 0.37 mmol) was dissolved in 50 mL acetonitrile solution, and fcSe@TiO₂ (0.10 g) was then dispersed into the solution. The mixture was let to react at room temperature for 24 h in the dark. The precipitate was collected by centrifugation, then washed with acetonitrile (3 × 10 mL) and dried under vacuum to obtain [Cu₂I₂(fcSe)₂]_n@TiO₂.

Photocatalytic activity evaluation

In the photocatalytic degradation experiment, TC (20 mg L⁻¹), Ciprofloxacin (20 mg L⁻¹), Toluidine Blue (20 mg L⁻¹), Methyl Orange (20 mg L⁻¹), Methylene Blue (20 mg L⁻¹), Malachite Green (20 mg L⁻¹), Basic Violet (20 mg L⁻¹), and imidazole (1 g L⁻¹) were selected as model N-cyclic organic pollutants to represent the antibiotics and textile dyes in the industrial wastewater, respectively. The initial pH of the reaction system was adjusted by 1 mol L⁻¹ NaOH or HCl solution.

In the photocatalytic reaction, 0.010 g catalyst was added into 50 mL aqueous solution containing the target N-cyclic organic. After stirring for 40 min in the dark at room temperature, 100 μL of 30 wt% H₂O₂ was added, and at the same time, the visible light photo-Fenton reaction was started. During the photo-Fenton reaction, 3.0 mL sample was collected from the suspension at an interval of 5 min and immediately filtered using 0.45 μm membrane filters.

The concentration of most of the target N-cyclic organics was detected by the UV-vis absorption intensity, and the detection wavelengths for TC, CIP, TB, MO, MB, MG and BV were 357 nm, 279 nm, 640 nm, 463 nm, 664 nm, 617 nm, and 585 nm, respectively. The concentration of imidazole was detected by HPLC with ZORBAX Eclipse Plus C18 column and acetonitrile as eluent (0.8 ml min⁻¹ flow rate).

The degradation efficiency was calculated using eqn (1):

C₀ was the initial concentration, and C_t was the N-cyclic organic concentration during the reaction.

The apparent rate constant k of catalysts was estimated by a pseudo-first-order kinetics equation (eqn (2)):

In the reactive specie scavenger experiments, the scavengers used were 1.0 mM 4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPOL) for ˙O₂⁻, carotene for ¹O₂, and isopropanol (IPA) for ˙OH and ammonium oxalate (AO) for h⁺, respectively. k_S represents the apparent rate constant k of catalysts in the above scavenger experiments, that is, k_T when using TEMPOL, k_C when using carotene, k_I when using IPA, and k_A when using AO.³³ The fractional contribution of different reactive species R is calculated by equation (eqn (3)):

The photoreduction of Cr(VI) to Cr(III) was analyzed using a pretreatment method of solvent extraction separation and enrichment combined with ICP analyses.

The conversion rate of CO and HCOOH was calculated by equation (eqn (4)):

n_CO is the molar amount of CO and n_HCOOH is the molar amount of HCOOH in the degradation products, and n_∑C is the total carbon molar amount in N-cyclic organic.

Results and discussion

Characterizations of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂

The X-ray photoelectron spectra (XPS) confirmed the successful assembly of fcSe and [Cu₂I₂(fcSe)₂]_n on the surface of TiO₂ P25 (Fig. 1a). In the high-resolution XPS Spectra of fcSe@TiO₂, stable Fe 2p state (Fig. 1b) was obtained with the binding energy values at 708.2 and 721.2 eV, revealing Fe²⁺ species of the ferrocene unit; the Se 3d was fitted to the peak with a binding energy value of 55.9 eV (Fig. 1c); the energy separation 5.7 eV of the Ti 2p doublet peaks, Ti 2p_3/2 (458.5 eV) and Ti 2p_1/2 (464.2 eV), perfectly agrees with the characteristic signals for TiO₂ Ti⁴⁺ state described in literature works (Fig. S6†).^29,34 In the high-resolution XPS spectra of [Cu₂I₂(fcSe)₂]_n@TiO_2, as well as the similar Fe 2p, Se 3d, and Ti 2p spectra, the confirmed Cu 2p spectrum (Cu 2p_3/2 933.1 eV and Cu 2p_1/2 953.8 eV) (Fig. 1d) and I 3d spectrum (I 3d_5/2 620.1 eV and I 3d_3/2 631.0 eV) (Fig. 1e) indicated the successful incorporation of CuI.^29,34,35 Transmission electron microscopy (TEM) images suggest that the construction of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ maintained the overall dimensions or morphology of TiO₂ nanoparticles with a diameter of 20 nm (Fig. S7, S8† and 2a–c). The high-resolution transmission electron microscopy (HR-TEM) of [Cu₂I₂(fcSe)₂]_n@TiO₂ in Fig. 2a–c shows a coated layer with approximately 5.00 nm thickness on the TiO₂ surface; well-defined lattice fringes in substrate TiO₂ with the lattice spacing was estimated to be ca. 0.325 nm, these are equal to the lattice parameter in the (110) facet of anatase TiO₂.³⁶ [Cu₂I₂(fcSe)₂]_n@TiO₂ was also analyzed using Scanning TEM (STEM) coupled with energy dispersive X-ray spectroscopy (EDX) to confirm the presence of Ti, Cu, Fe, I, and Se, which suggests the successful preparation of core–shell nanoparticles [Cu₂I₂(fcSe)₂]_n@TiO₂ (Fig. 2d and 2e–i). The quantitative mapping analysis manifested that the molar ratio of Cu/Fe = 2.0 far exceeded the bulk molar ratio of Cu/Fe = 1.0 measured by ICP, suggesting that the content of Cu decreased from the exterior to the interior. In the FTIR spectra of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ (Fig. S9†), the C=O stretching vibration is at 1677 cm⁻¹, as well as asymmetric and symmetric stretching bands of the carboxylate anion (COO⁻) due to the splitting of carboxylate groups when complexed with surface Ti centers, υ_as at 1600 cm⁻¹ and υ_s at 1424 cm⁻¹.³⁷ The value of Δυ_a–s = 176 cm⁻¹ in the current two nanosystems suggests fcSe is chemically adsorbed on the TiO₂ surface through the bidentate bridging form,³⁸ which further facilitates electron transfer between ferrocene and TiO₂ surface.^39–41

Fig. 1 (a) XPS survey spectrum of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂, high-resolution XPS spectra of (b) Fe 2p and (c) Se 3d in fcSe@TiO₂, (d) Cu 2p and (e) I 3d in [Cu₂I₂(fcSe)₂]_n@TiO₂.

Fig. 2 (a–c) TEM and HRTEM images, (d) EDX-mapping spectrum and (e–i) elemental mapping image of [Cu₂I₂(fcSe)₂]_n@TiO₂.

The relative BET specific surface areas of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ were found to be 36.1 m² g⁻¹ and 48.3 m² g⁻¹ (Fig. S10†), slightly decreased from that of pure TiO₂ (54.2 m² g⁻¹), indicating that the surface modification did not produce holes and cavities.

The UV-vis diffuse reflectance spectra (DRS) of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ in Fig. 3a revealed the broad and strong visible absorption, primarily attributed to the presence of the ferrocene moiety.^29,30 Tauc plot was applied to calculate the band gap energy E_g via the Kubelka–Munk method (Fig. 3b and S11†), giving 2.3 eV for fcSe@TiO₂ and 2.2 eV for [Cu₂I₂(fcSe)₂]_n@TiO₂. The introduction of fcSe and [Cu₂I₂(fcSe)₂]_n significantly decreases the E_g of free TiO₂ (3.2 eV) and increases the visible light absorption, owing to the potential to accelerate the photocatalytic reaction.

Fig. 3 (a) UV-vis-DRS spectra and (b) the band gap estimation of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂.

Visible light photocatalytic degradation of TC

The photocatalytic activities of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ on TC degradation were initially evaluated by the degradation of TC (20 mg L⁻¹) under the role of each constituent, including visible light (λ > 420 nm), H₂O₂ (20 mM) and catalysts (0.2 g L⁻¹). As shown in Fig. 4a and calculated by eqn (1), according to the decreased adsorption of TC at 357 nm, the TC degradation efficiency (93.1%) was obtained by fcSe@TiO₂ + H₂O₂ + hν within 30 min, exhibiting a significant enhancement as compared with TiO₂ + H₂O₂ + hν (67.8%) and fcSe + H₂O₂ + hν (23.8%). Calculated by eqn (2), the apparent rate constant k value of fcSe@TiO₂ + H₂O₂ + hν was 0.064 min⁻¹, 2 times of TiO₂ + H₂O₂ + hν (0.033 min⁻¹) and 8 times of fcSe + H₂O₂ + hν (0.009 min⁻¹) (Fig. 4b).

Fig. 4 (a) Degradation rate and (b) the corresponding pseudo-first-order kinetic curves of TC under various experimental conditions in the fcSe@TiO₂ system (blue line). (c) Photo-Fenton degradation efficiency and (d) the corresponding pseudo-first-order kinetic curves of TC on various experimental conditions in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red line).

Without visible light irradiation, fcSe@TiO₂ +H₂O₂, TC removal efficiency decreased to 46.8% with k = 0.011 min⁻¹, suggesting that visible light could significantly promote Fenton chemistry by providing photogenerated electrons in the recovery of Fe(II) from Fe(III). As for the photocatalytic process without H₂O₂ addition (fcSe@TiO₂ + hν), 45.7% TC degradation was observed, indicating that H₂O₂ is not the only source of reactive species. The system of hν + H₂O₂ could hardly degrade TC, illustrating the vital role of fcSe@TiO₂. Therefore, three constituents (visible light, H₂O₂, and fcSe@TiO₂) are indispensable for this photo-catalytic process and their interaction is responsible for the efficient degradation of TC. Similar results were obtained for [Cu₂I₂(fcSe)₂]_n@TiO₂ (Fig. 4c and d). The TC degradation efficiency (91.3%) and k = 0.062 min⁻¹ were obtained by [Cu₂I₂(fcSe)₂]_n@TiO₂ + H₂O₂ + hν within 30 min.

We next moved on to screen different factors influencing the TC degradation efficiency by the two studied nanosystems, including catalyst dosage, H₂O₂ concentration, and initial pH on TC degradation (Fig. 5). In the three studied catalyst dosages (0.1 g L⁻¹, 0.2 g L⁻¹, and 0.3 g L⁻¹), the best TC degradation efficiency was found to be 93.1% for fcSe@TiO₂ and 91.3% for [Cu₂I₂(fcSe)₂]_n@TiO₂ at 0.2 g L⁻¹ dosage (Fig. 5a and d). As shown in Fig. 5b and e, 10 mM/20 mM/30 mM H₂O₂ concentrations were studied, the TC degradation efficiency showed no significant enhancement for fcSe@TiO₂ (from 92.2% to 93.1%) and medium enhancement for [Cu₂I₂(fcSe)₂]_n@TiO₂ (from 83.4% to 91.3%). These agree with literature works, excessive H₂O₂ can cause oversaturation of the reaction sites and result in the scavenging of ˙OH.^42–44

Fig. 5 TC Degradation rate under different initial conditions. (a) Catalyst dosage, (b) H₂O₂ concentration and (C) initial pH in fcSe@TiO₂ system (blue line). (d) Catalyst dosage, (e) H₂O₂ concentration, and (f) initial pH in [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red line).

It is known that most reported Fenton systems are limited by strict acidic pH requirements.²⁰ However, our two catalysts have much wider pH tolerance. The Z-average hydrodynamic diameters and zeta potentials of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ at pH = 3, 5, 7, and 9 were obtained from dynamic light scattering (DLS) measurements (Table S3†). The larger Z-average hydrodynamic diameters and more positive zeta potentials were obtained in an acidic solution. The results of the study indicated that fcSe has the ability to form Se⋯ H⁺ bond, which may effectively create an acidic microenvironment that could potentially enhance the Fenton reaction. As shown in Fig. 5c and f, the degradation efficacy of TC was maintained around 90% under acidic or neutral conditions, pH = 3, 5, 7. Under alkaline conditions, pH = 9, TC rates decreased to 70.5% for fcSe@TiO₂ and 73.1% for [Cu₂I₂(fcSe)₂]_n@TiO₂, which is due to electrostatic repulsion and the additional decomposition of H₂O₂.⁴⁵ So, the following studies are all based on the optimal conditions, which are 0.2 g L⁻¹ catalyst and 20 mM H₂O₂ and initial pH = 7.

In actual wastewater systems, it is inevitable that various ionic species exist and pose a negative effect on the photocatalytic performance of the catalyst.^46,47 Herein, five different anions Cl⁻, Cr₂O₇²⁻, HCO₃⁻, H₂PO₄⁻, SO₄²⁻ and three different cations K⁺, Na⁺, NH₄⁺ at a concentration of 10 mM, were selected to study their influence on TC degradation. It was found that in the fcSe@TiO₂ system, Cl⁻, Cr₂O₇²⁻, SO₄²⁻ and K⁺, Na⁺, NH₄⁺ showed negligible effect on TC degradation (Fig. 6a); the same phenomena was observed for Cr₂O₇²⁻, HCO₃⁻, Na⁺, NH₄⁺ in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system (Fig. 6b). A significant inhibiting effect was observed in the presence of H₂PO₄⁻, the TC degradation efficiency decreasing to 71.9% for fcSe@TiO₂ and 75.1% for [Cu₂I₂(fcSe)₂]_n@TiO₂, which was attributed to the H₂PO₄⁻ scavenger of ˙OH and/or the attachment to the surface of TiO₂.^48,49 The introduction of Cl⁻ slightly inhibited the degradation of TC (79.8%) in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system, which can be due to the less oxidizable ˙Cl and ˙ClOH⁻ formed from Cl⁻ and ˙OH.⁵⁰ Therefore, it is delightful to find that fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ have good tolerance of most coexisting inorganic anions and cations, so no pre-treatment is needed to remove them before the photodegradation of TC wastewater.

Fig. 6 Effect of coexisting ions on the degradation of TC in (a) fcSe@TiO₂ (blue) and (b) [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red). Synergetic photodegrade of TC and photoreduction of Cr(VI) in (c) fcSe@TiO₂ (blue) and (d) [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red). Experimental conditions: TC concentration = 20 mg L⁻¹, H₂O₂ concentration = 20 mM, catalyst dosages = 0.2 g L⁻¹, initial pH = 7.

We then further explored simultaneous photodegradation of TC and reduction of Cr(VI). Under neutral conditions, 92.1% of TC and 64.5% of Cr(VI) removal efficiency were achieved with fcSe@TiO₂, and 89.2% of TC removal and 41.2% of Cr(VI) reduction efficiency with [Cu₂I₂(fcSe)₂]_n@TiO₂, along with unchangeable efficient degradation efficiency of TC (Fig. 6c and d). Moreover, the presence of TC degradation increased the Cr(VI) reduction rate 3.76-fold in fcSe@TiO₂/H₂O₂/hν system and 3.72-fold in [Cu₂I₂(fcSe)₂]_n@TiO₂/H₂O₂/hν system. According to the reported studies,^10,51 three aspects could interpret the good synergetic activity: (1) the reduction of Cr(VI) benefits from the acidic microenvironment obtained from Se in fcSe; (2) the photogenerated electrons effectively reduce Cr(VI); (3) TC degradation consumes reactive species and suppresses the re-oxidation of Cr(III).

The reusability and stability of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ catalysts were investigated by successive TC degradation and Cr(VI) reduction processes (Fig. S12†). The catalytic performance for TC degradation remained invariant during the initial four successive cycles, then was slightly reduced to around 88% in the fifth cyclic run. The same phenomena were observed for Cr(VI) reduction. From the relevant characterizations of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ catalysts after the fifth degradation process, no obvious change in the size or shape of nanoparticles was observed in TEM images (Fig. S13 and S14†), while FTIR spectra indicated physical adsorption on the surface of the photocatalysts which may slightly decrease the catalytic performance (Fig. S15†). The leaching of Cu and Fe concentration was a key issue related to the stability of nanocatalysts. Our ICP analyses revealed low concentrations of leachable iron in the solutions in comparison to literature reports;^52,53 after five cycles, only 0.048 mg L⁻¹ of fcSe@TiO₂ (1.46% of 3.28 mg L⁻¹) and 0.009 mg L⁻¹ of [Cu₂I₂(fcSe)₂]_n@TiO₂ (0.30% of 3.10 mg L⁻¹). Leachable copper in the solution after five cycles is 0.141 mg L⁻¹ of [Cu₂I₂(fcSe)₂]_n@TiO₂ (3.98% of 3.55 mg L⁻¹). Based on the ICP analysis, it is not surprising to find the increased stability of [Cu₂I₂(fcSe)₂]_n@TiO₂ compared to fcSe@TiO₂.

The photocatalytic removal efficiency of TC of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ were compared with the reported catalysts in the literature works (Table S4†). Considering the catalyst dosage, the irradiation time and interfering ions, our two catalysts exhibited excellent degradation ability of TC (93.1% and 91.3%) under the conditions of low catalyst dosage (0.2 g L⁻¹), short reaction time (30 min), and good tolerance of most coexisting inorganic ions. Moreover, it was worth noting that good synergetic redox photocatalytic activity of the photodegrade of TC and photoreduction of Cr(VI) to Cr(III). In summary, all these results demonstrated that fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ are efficient and stable photocatalysts, with potential industrial applications. Although various copper compounds have been studied as efficient H₂O₂ activators in Fenton-like reactions,^54–59 [Cu₂I₂(fcSe)₂]_n@TiO₂ catalyst system shows no significant improvement as compared with fcSe@TiO₂.

Mechanism study

To study the reactive species generated in the two heterogeneous photo-catalytic systems, radical trapping experiments were firstly performed by employing different scavengers, isopropanol (IPA) for ˙OH, ammonium oxalate (AO) for h⁺, TEMPOL for ˙O₂⁻ and carotene for ¹O₂ respectively.⁶⁰ In the fcSe@TiO₂ system (Fig. 7a), the TC degradation efficiency decreased from 93.1% to 84.6% by IPA, 81.8% by AO, 78.7% by TEMPOL, and 89.0% by carotene. In the [Cu₂I₂(fcSe)₂]_n@TiO₂ system (Fig. 7b), the TC degradation efficiency decreased from 91.3% to 73.4% by IPA, 69.4% by AO, 76.2% by TEMPOL, and 76.9% by carotene. From the collective results of the above degradation rates under various scavengers (Fig. 7c), combining the fractional contribution R of each active specie calculated from first-order rate constant (k_s), eqn (3) (Fig. 7d–f), it can be indicated that ˙OH, ˙O₂⁻ and h⁺ were responsible for TC degradation in the fcSe@TiO₂ system, while additional ¹O₂ existed in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system.

Fig. 7 Effects of different scavengers on the photo-catalytic degradation rate of TC in (a) fcSe@TiO₂ (blue), (b) [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red) and (c) collective results in bar graph. The according pseudo-first-order kinetic curves (d) in fcSe@TiO₂ system (blue), (e) [Cu₂I₂(fcSe)₂]_n@TiO₂ system (red) and (f) the fractional contribution of different reactive species R in the bar graph. Experimental conditions are 25–30 °C, pH = 7, catalyst dosage = 0.2 g L⁻¹, H₂O₂ concentration = 20 mM.

The spin-trapping EPR experiment with 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was performed to attest the generation of ˙OH and ˙O₂⁻ radicals in the two systems, singlet oxygen sensor green (SOSG) was selected as ¹O₂ probe through the change of the intense fluorescence emission centered at 525 nm due to the cycloaddition reaction with ¹O₂.⁶¹ As depicted in Fig. 8a–f, no obvious signals were detected in the dark. After 5 min of visible light irradiation, the characteristic peaks of DMPO-˙OH and DMPO-˙O₂⁻ adducts were observed, respectively. fcSe@TiO₂ system displayed 2.0 times and 1.2 times higher DMPO-˙OH and DMPO-˙O₂⁻ intensities signals than those in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system. Thus, it could be concluded that fcSe@TiO₂ is able to generate a higher concentration of ˙OH and ˙O₂⁻ radicals than [Cu₂I₂(fcSe)₂]_n@TiO₂, which is in congruence with their different removal efficiencies in the first five minutes as shown in Fig. 4a. In fcSe@TiO₂ system there is no significant change of SOSG fluorescence emission near 525 nm before and after illumination, implying the absence of ¹O₂; while in [Cu₂I₂(fcSe)₂]_n@TiO₂ system the significantly enhanced emission indicated ¹O₂ produced. When the ˙O₂⁻ scavenger TEMPOL^39,61 was added into the [Cu₂I₂(fcSe)₂]_n@TiO₂ system, the fluorescence intensity of SOSG in response to ¹O₂ decreased sharply to about 8.0%, indicating the generation of ¹O₂ from ˙O₂⁻ (Fig. S16†).

Fig. 8 EPR spectra of (a) DMPO-˙OH, (b) DMPO-˙O₂⁻ and (c) fluorescence change of SOSG in response to ¹O₂ generated using fcSe@TiO₂ (blue line). EPR spectra of (d) DMPO-˙OH, (e) DMPO-˙O₂⁻ and (f) fluorescence change of SOSG in response to ¹O₂ generated using [Cu₂I₂(fcSe)₂]_n@TiO₂ (red line). Experimental conditions are 25–30 °C, pH = 7, catalyst dosage = 0.2 g L⁻¹, H₂O₂ concentration = 20 mM.

According to the results of reactive species capture, EPR, and ¹O₂ detection, we infer that fcSe and [Cu₂I₂(fcSe)₂]_n on TiO₂ tune the generation of reactive species through photogenerated charge and the photo-Fenton process further affect the pathways in TC photocatalytic degradation, and finally determine the degradation rate.

To further reveal the photocatalytic degradation pathways of TC, the intermediates and final products formed during the 5 min, 10 min, and 30 min process were analyzed through GC and LC-MS, and chemical structures of the corresponding degradation products are summarized in Table S5.† Based on the observed m/z peaks and related literature, probable degradation pathways are presented in Scheme 3.^62–64

Scheme 3 Proposed photo-Fenton degradation pathways of TC in fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ system.

In the fcSe@TiO₂ system, three possible initial degradation pathways were generated by the independent attack of ˙O₂⁻, ˙OH, and h⁺. For pathway I, TC molecule underwent 1,3-dipolar addition of double bond to give I1 (m/z = 461) (blue part in Scheme 3), due to the attacks of ˙O₂⁻. After that, compound I1 was further demethylated to form compound I2 (m/z = 433) with loss of both methyl groups in the amino moiety. Through the dehydration-oxidation induced by h⁺, pathway II was dealcoholization on methylcyclohexan-1-one to give I3 (m/z = 427) (yellow part in Scheme 3); I3 was oxidatively degraded, and removed the methyl groups on the dimethylamine. Then, a series of ring opening and the side chain degradation to obtain I4 (m/z = 353); I3 could also be converted into I5 (m/z = 337) by deamidation along with the ring opening. In pathway III, I6 (m/z = 417) intermediate was generated through demethylation of the dimethylamino group when TC reacted with ˙OH (red part in Scheme 3). Subsequently, I7 (m/z = 447) was formed by ring opening and deamination of the compound I6.

The significant difference in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system was that there only existed pathway II and III. As electron-rich functional groups, double bond, amide group and dimethylamino group in TC molecule were more vulnerable to the attack of electrophilic ¹O₂. Therefore, compound I7 was much easier to generate when I6 was attacked by ¹O₂ in pathway III.

The TC stepwise degradation goes through ring opening, decarboxylation, dealcoholization and dealkylation processes, and finally, to form HCOOH, CO, CO₂, H₂O, NO₃⁻. Conversion rate η_CO+HCOOH was then calculated by eqn (4), to give 7.2% in fcSe@TiO₂ and 6.1% in [Cu₂I₂(fcSe)₂]_n@TiO₂ (Table 1). The yield of CO and HCOOH was low under visible light irradiation, which meant the selectivity of TC to these products was poor. When white light was applied, much-enhanced transformation of TC to CO and HCOOH (22.8%) was found in fcSe@TiO₂.

Table 1 Evaluation of fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ in the degradation of TC under different irradiation light^a

	Visible light (λ > 420 nm)		White light
	fcSe@TiO2	[Cu2I2(fcSe)2]n@TiO2	fcSe@TiO2	[Cu2I2(fcSe)2]n@TiO2
a Experimental conditions are 25–30 °C, pH = 7, catalyst dosage = 0.2 g L^−1, H2O2 concentration = 20 mM.b Conversion rate of CO and HCOOH ηCO+HCOOH was calculated by eqn (4).
Removal efficiency (%)	93.1	91.3	97.4	96.2
ηCO+HCOOH (%)^b	7.2	6.1	22.8	7.4

---

Based on the above analyses, the proposed degradation mechanism of TC over fcSe@TiO₂ or [Cu₂I₂(fcSe)₂]_n@TiO₂ catalysts is presented in Fig. 9. Firstly, visible-light irradiated fcSe@TiO₂ or [Cu₂I₂(fcSe)₂]_n@TiO₂ to produce photoexcited e⁻ and h⁺. The h⁺ in the valence band (VB) is directly involved in the degradation process and H⁺ evolution from H₂O. At the same time, a Fenton-like reaction between the ferrocene group and H₂O₂ released ˙OH and ferrocenium group; in this step, Se could form hydrogen bonds with H⁺ to form an acid microenvironment, which further improves the efficiency of the photo-Fenton reaction. Then, the O–Ti bonds between fcSe and TiO₂ surface allow part e⁻ migration to promote the reduction conversion of ferrocenium ([Fe³⁺(η⁵-C₅H₅)₂])/ferrocene ([Fe²⁺(η⁵-C₅H₅)₂]). Meanwhile, the remaining electrons accumulated in the CB could effectively reduce Cr(VI) and react with O₂ to form reactive ˙O₂⁻ radicals.^64,65 The redox cycle between different valence states Cu^I/Cu^II could also contribute to the generation of ¹O₂ from ˙O₂⁻.⁶⁶ These processes benefit the inhibition of the recombination of e⁻/h⁺, and produce reactive species (h⁺, ˙OH, ˙O₂⁻ and ¹O₂) for synergistical transformation of TC and Cr(VI).

Fig. 9 Proposed TC degradation mechanism in fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ system.

Screen the application scope of N-cyclic organics

A distinguished system in water treatment is expected to have widespread applicability in the degradation of different pollutants. We then evaluated fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ in the degradation of seven representative N-cyclic organic contaminants. As shown in Table S6,† the degradation efficiency in the fcSe@TiO₂ system within 30 min was 86.2% for Ciprofloxacin, 93.9% for Methylene Blue, 94.2% for Toluidine Blue, 92.0% for Pigment Green, 92.3% for Basic Violet, 17.1% for Methyl Orange, and 16.6% for imidazole; while in the [Cu₂I₂(fcSe)₂]_n@TiO₂ system, 64.9% for Ciprofloxacin, 41.4% for Methylene Blue, 88.9% for Toluidine Blue, 93.7% for Pigment Green, 86.2% for Basic Violet, 16.0% for Methyl Orange and 25.9% for imidazole, respectively. Within these studied organics, only Methyl Orange and imidazole showed very low degradation efficiency, possibly due to the stable bonds in small molecular structures. Resourceful HCOOH and CO were found in final products with the conversion rate η_CO+HCOOH from 3.0% to 20.8%. These demonstrated that as-prepared fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂ possessed excellent adaptability to various N-cyclic organic pollutants.

Conclusions

Two novel visible light photocatalysts, fcSe@TiO₂ and [Cu₂I₂(fcSe)₂]_n@TiO₂, were successfully constructed to achieve efficient simultaneous transformation of refractory N-cyclic organic contaminants and Cr(VI) reduction under neutral conditions. Within just 30 minutes, 92.1% of TC and 64.5% of Cr(VI) removal efficiency were achieved with fcSe@TiO₂, 89.2% of TC removal and 41.2% of Cr(VI) reduction efficiency with [Cu₂I₂(fcSe)₂]_n@TiO₂. Mechanistic studies inferred that fcSe and [Cu₂I₂(fcSe)₂]_n on TiO₂ tune the generation of reactive species through photogenerated charge and photo-Fenton process. The Fenton reaction inhibited the recombination of the photogenerated charges; meanwhile, the efficient photogenerated electrons could accelerate ferrocenium/ferrocene reduction conversion in the Fenton cycle. Se atom can also form interactions with H⁺, thereby providing an acidic microenvironment, which would be the key factor for the acidic Fenton reaction, acidic Cr(VI) reduction, and H⁺ evolution from H₂O over a hole in the valence band. The two nanosystems exhibited good tolerance towards most coexisting inorganic anions and cations, as well as stability after five cycling runs. This study provides an efficient and practical approach for visible light photocatalytic transformation of N-cyclic organics and Cr(VI) under mild reaction conditions, with resourceful products, complex environment tolerance, and wide adaptability in wastewater treatment.

Author contributions

Zhuo Yang, Jinshan Wang, Chao Wang: photocatalysts preparation and photocatalytic study. Wei Ji: experiments design and original draft writing. Su Jing: conceptualization, methodology, supervision, writing-reviewing and editing. Aimin Li, Elísabet Pires, Wenzhong Yang: results and discussion and manuscript comment.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work is funded by the Primary Research & Development Plan of Jiangsu Province (BE2019708 and BE2021712), and the National Natural Science Foundation of China (Grant No. 22175090).

References

1. E. Vitaku, D. T. Smith and J. T. Njardarson, J. Med. Chem., 2014, 57, 10257–10274.
2. D. Sun, D. N. Confair and J. A. Ellman, Acc. Chem. Res., 2021, 54, 1766–1778.
3. F.-K. Zhao, L. Yang, H. Yen, Q.-Y. Feng, M. Li and L.-D. Chen, Nat. Commun., 2023, 14, 6094.
4. S. M. Zainab, M. Junaid, N. Xu and R. N. Malik, Water Res., 2020, 187, 116455.
5. P. Chaturvedi, B. S. Giri, P. Shukla and P. Gupta, Bioresour. Technol., 2021, 319, 124161.
6. P. Kovalakova, L. Cizmas, T. J. McDonald, B. Marsalek, M.-B. Feng and V. K. Sharma, Chemosphere, 2020, 251, 126351.
7. Q.-L. Yang, Y. Gao, J. Ke, P. L. Show, Y.-H. Ge, Y.-H. Liu, R.-X. Guo and J.-Q. Chen, Bioengineered, 2021, 12, 7376–7416.
8. J. Sharma, S. Sharma and V. Soni, Reg. Stud. Mar. Sci., 2019, 7, 24.
9. T. Yu, H. Chen, T. Hu, J. Feng, W.-L. Xing, L. Tang and W.-W. Tang, Appl. Catal., B, 2024, 342, 123401.
10. S. Guo, W. Yang, L. You, J. Li, J. Y. Chen and K. Zhou, Chem. Eng. J., 2020, 393, 124758.
11. X. Yin, W. Liu and J. Ni, Chem. Eng. J., 2014, 248, 89–97.
12. Y. Wang, L. Rao, P. Wang, Z. Shi and L. Zhang, Appl. Catal., B, 2020, 262, 118308.
13. B. Zhu, S. Chen, C. Li, G. Jiang, F. Liu, R. Zhao and C. Liu, Appl. Catal., B, 2022, 318, 121881.
14. C. Shi, S. Yu and C. Li, Chem. Eng. J., 2022, 441, 136052.
15. S. Wu, H. Hu, Y. Lin, J. Zhang and Y. H. Hu, Chem. Eng. J., 2020, 382, 122842.
16. W. Shi, C. Hao, Y. Fu, F. Guo, Y. Tang and X. Yan, Chem. Eng. J., 2022, 433, 133741.
17. H. Tan, D. Kong, Q. Li, Y. Zhou, X. Jiang, Z. Wang, R. E. Parales and Z. Ruan, Environ. Pollut., 2022, 305, 119299.
18. C. Xia, H. Huang, D. Liang, Y. Xie, F. Kong, Q. Yang, J. Fu, Z. Dou, Q. Zhang and Z. Meng, Chem. Eng. J., 2022, 443, 136398.
19. D. Meyerstein, Nat. Rev. Chem, 2021, 5, 595–597.
20. Z. Tang, P. Zhao, H. Wang, Y. Liu and W. Bu, Chem. Rev., 2021, 121, 1981–2019.
21. Y. Zhu, R. Zhu, Y. Xi, J. Zhu, G. Zhu and H. He, Appl. Catal., B, 2019, 255, 117739.
22. H. Wang, C.-Y. Zhang, X.-L. Zhang, S.-J. Wang, Z.-X. Xia, G.-F. Zeng, J. Ding and N.-Q. Ren, Chem. Eng. J., 2022, 429, 132445.
23. Z.-H. Wu, J. Shen, W.-L. Li, J.-S. Li, D.-H. Xia, D.-F. Xu, S.-Y. Zhang and Y.-F. Zhu, Appl. Catal., B, 2023, 330, 122642.
24. C.-A. Li, W. Ji, J. Qu, S. Jing, F. Gao and D.-R. Zhu, Dalton Trans., 2018, 47, 7463–7470.
25. H.-Y. Zhou, M. Li, J. Qu, S. Jing, H. Xu, J.-Z. Zhao, J. Zhang and M.-F. He, Organometallics, 2016, 35, 1866–1875.
26. J. Qu, Q. Xia, W. Ji, S. Jing, D. Zhu, L. Li, L. Huang, Z. An, C. Xin, Y. Ni, M. Li, J. Jia, Y. Song and W. Huang, Dalton Trans., 2018, 47, 1479–1487.
27. J. Zhou, X. Zhu, Q. Cheng, Y. Wang, R. Wang, X. Cheng, J. Xu, K. Liu, L. Li, X. Li, M. He, J. Wang, H. Xu, S. Jing and L. Huang, Inorg. Chem., 2020, 59, 9177–9187.
28. T. Zhou, Q. Cheng, L. Zhang, D. Zhang, L. Li, T. Jiang, L. Huang, H. Xu, M. Hu and S. Jing, Chem. Eng. J., 2022, 438, 135637.
29. W. Ji, J. Qu, C. A. Li, J. W. Wu, S. Jing, F. Gao, Y. L. Lv, C. Liu, D. R. Zhu, X. M. Ren and W. Huang, Appl. Catal., B, 2017, 205, 368–375.
30. G.-J. Zha, W. Ji, Z.-H. Qi, W.-J. Qiu, A.-M. Li, D.-R. Zhu and S. Jing, J. Catal., 2022, 405, 313–321.
31. W. Ji, J. Qu, S. Jing, D. R. Zhu and W. Huang, Dalton Trans., 2016, 45, 1016–1024.
32. M. R. Burgess, S. Jing and C. P. Morley, J. Organomet. Chem., 2006, 691, 3484–3489.
33. F. Guo, H. Zhang, H. Li and Z. Shen, Appl. Catal., B, 2022, 306, 121092.
34. V. Kumaravel, S. Rhatigan, S. Mathew, J. Bartlett, M. Nolan, S. J. Hinder, P. K. Sharma, A. Singh, J. A. Byrne, J. Harrison and S. C. Pillai, J. Phys. Chem. C, 2019, 123, 21083–21096.
35. Y. Xu, J. Wan, L. Huang, J. Xu, M. Ou, Y. Liu, X. Sun, S. Li, C. Fang, Q. Li, J. Han, Y. Huang and Y. Zhao, Energy Storage Mater., 2020, 33, 432–441.
36. D. Dambournet, Acc. Chem. Res., 2020, 55, 696–706.
37. A. B. Grafton and C. M. Cheatum, Chem. Phys., 2021, 512, 3–12.
38. Y. Ma, Y. Lu, G. Hai, W. Dong, R. Li, J. Liu and G. Wang, Sci. Bull., 2020, 65, 658–669.
39. A. Vittadini, A. Selloni, F. P. Rotzinger and M. Grätzel, J. Phys. Chem. B, 2000, 104, 1300–1306.
40. M. Nilsing, P. Persson and L. Ojamäe, Chem. Phys. Lett., 2005, 415, 375–380.
41. M. W. Mara, D. N. Bowman, O. Buyukcakir, M. L. Shelby, K. Haldrup, J. Huang, M. R. Harpham, A. B. Stickrath, X. Zhang, J. F. Stoddart, A. Coskun, E. Jakubikova and L. X. Chen, J. Am. Chem. Soc., 2015, 137, 9670–9684.
42. L. Yang, L. Li, Z. Liu, C. Lai, X. Yang, X. Shi, S. Liu, M. Zhang, Y. Fu, X. Zhou, H. Yan, F. Xu, D. Ma and C. Tang, Chemosphere, 2022, 294, 133736.
43. S. Huang, Q. Zhang, P. Liu, S. Ma, B. Xie, K. Yang and Y. Zhao, Appl. Catal., B, 2020, 263, 118336.
44. Y. Yao, G. Wu, F. Lu, S. Wang, Y. Hu, J. Zhang, W. Huang and F. Wei, Environ. Sci. Pollut. Res., 2016, 23, 21833–21845.
45. H. Qi, X. Sun and Z. Sun, Chem. Eng. J., 2021, 403, 126270.
46. F. Chen, Q. Yang, X. Li, G. Zeng, D. Wang, C. Niu, J. Zhao, H. An, T. Xie and Y. Deng, Appl. Catal., B, 2017, 200, 330–342.
47. Q. Li, G. Wei, Y. Yang, L. Gao, L. Zhang, Z. Li, X. Huang and J. Gan, Chem. Eng. J., 2021, 424, 130537.
48. N. Li, Y. Wang, X. Cheng, H. Dai, B. Yan, G. Chen, L. Hou and S. Wang, Water Res., 2022, 222, 118896.
49. Y. Ren, L. Lin, J. Ma, J. Yang, J. Feng and Z. Fan, Appl. Catal., B, 2015, 165, 572–578.
50. Y. Liu, X. He, Y. Fu and D. D. Dionysiou, Chem. Eng. J., 2016, 284, 1317–1327.
51. J. Gong, W. Zhang, T. Sen, Y. Yu, Y. Liu, J. Zhang and L. Wang, ACS Appl. Nano Mater., 2021, 4, 4513–4521.
52. R. Liu, Y. Xu and B. Chen, Environ. Sci. Technol., 2018, 52, 7043–7053.
53. J. Sun, J. Wan, Y. Wang, Z. Yan, Y. Ma, S. Ding, M. Tang and Y. Xie, J. Hazard. Mater., 2022, 429, 128299.
54. E. Ma, K. Wang, Z. Hu and H. Wang, J. Colloid Interface Sci., 2021, 603, 685–694.
55. G. Pan and Z. Sun, Chemosphere, 2021, 283, 131257.
56. H. Hu, K. Miao, X. Luo, S. Guo, X. Yuan, F. Pei, H. Qian and G. Feng, J. Cleaner Prod., 2021, 318, 128632.
57. J. Xu, X. Zheng, Z. Feng, Z. Lu, Z. Zhang, W. Huang, Y. Li, D. Vuckovic, Y. Li, S. Dai, G. Chen, K. Wang, H. Wang, J. K. Chen, W. Mitch and Y. Cui, Nat. Sustain., 2021, 4, 233–241.
58. N. Schmachtenberg, S. Silvestri, J. da S. Salla, G. L. Dotto, D. Hotza, S. L. Jahn and E. L. Foletto, J. Environ. Chem. Eng., 2019, 7, 102954.
59. Y. Tian, N. Jia, H. Ma, G. Liu, Z. Xiao, Y. Wu, L. Zhou, J. Lei, L. Wang, Y. Liu and J. Zhang, J. Hazard. Mater., 2021, 419, 126359.
60. Y. Yang, G. Zeng, D. Huang, C. Zhang, D. He, C. Zhou, W. Wang, W. Xiong, B. Song, H. Yi, S. Ye and X. Ren, Small, 2020, 16, 2001634.
61. B. Chen, Y. Yang, Y. Wang, Y. Yan, Z. Wang, Q. Yin, Q. Zhang and Y. Wang, ACS Appl. Mater. Interfaces, 2021, 13, 18533–18544.
62. C. Liu, H. Dai, C. Tan, Q. Pan, F. Hu and X. Peng, Appl. Catal., B, 2022, 310, 121326.
63. D. Xing, Z. Cui, Y. Liu, Z. Wang, P. Wang, Z. Zheng, H. Cheng, Y. Dai and B. Huang, Appl. Catal., B, 2021, 290, 120029.
64. K. Dou, C. Y. Peng, R. C. Wang, H. P. Cao, C. Yao, J. F. Qiu, J. L. Liu, N. Tsidaeva and W. Wang, Chem. Eng. J., 2023, 455, 140813.
65. C. Y. Yang, J. Wang, R. Wang, W. X. Zhu, L. Zhang, T. Du, J. Sun, M. Q. Zhu, Y. Z. Shen and J. L. Wang, Appl. Catal., B, 2023, 322, 122084.
66. Q. Wang, D. Zhou, K. Lin and X. Chen, Chem. Eng. J., 2021, 419, 129667.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2216502. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3ra07390d

---