Raw materials like anhydrous ethanol, Butyl titanate, Ferric nitrate, Hydrochloric acid, Sodium hydroxide, Methylene blue, Sodium chloride, Potassium chloride, Sodium sulfate, Sodium carbonate, Isopropanol, n-Butanol, Cerium nitrate, Ruthenium trichloride, Tin tetrachloride, Chloroiridic acid were purchased from Tianjin Beilian Chemical Reagents Company in analytical grade, and used as received without further purification. The coal gasification slag (CGS) was obtained from the Yanzhou Coal Industry Yulin Energy and Chemical Company. Titanium sheets (dimensions 50 mm × 30 mm × 2 mm) were purchased from Shenzhen Quanfu Metal Material Co., Ltd., China. These sheets were utilized as cathodes for the preparation of Ti/RuO₂-IrO₂-SnO₂-CeO₂ anode.

The Zeiss Sigma 300 scanning electron microscope (SEM, Zeiss, Germany) and transmission electron microscope (JEM, 2100F TEM (JEOL, Japan) were used for morphology and elemental composition analyses of the Ti/RuO₂-IrO₂-SnO₂-CeO₂ electrode and Fe³⁺-TiO₂@CGS particles. The concentrations of MB in the solution samples were analyzed by spectrometry (UV-2450, Shimadzu, Japan). The absorption/desorption curves and pore size distributions of Fe³⁺-TiO₂@CGS three-dimensional electrodes were analyzed by BET specific surface area (CN61M/SZB-9, Replete, China). The functional groups on the samples were tested by fourier transform infrared spectrometer (FT-IR, IS50, Nicolet, America). X-ray diffraction (XRD measurements were recorded by a Rikagu diffractometer (D₈ Advance, BRUKE, Germany) with Cu-radiation (λ = 1.54178 Å) at 40 kV voltage and 30 mA current. The X-ray photoelectron spectroscopy (XPS) profiles were obtained on a Scientific ESCALAB 250Xi photoelectron spectrometer (Thermo Fisher, America).

The microstructures of the Ti/RuO₂-IrO₂-SnO₂-CeO₂ electrodes were viewed by SEM and the results are shown in Figure 2A. A fragmented structure containing cracks and a dense structure without cracks were observed on the electrode surface. These features were caused by the difference in thermal expansion coefficients of the metal coating and metallic titanium substrate. The formation of the cracked fragmented structure may be beneficial in terms of increasing the specific surface area of the electrode, facilitating contact with contaminants, and improving electrocatalytic performance. More importantly, the presence of the dense structure could enhance the gas erosion resistance of the electrode surface, which would effectively alleviate the penetration of O₂ and electrolyte into the coating and titanium substrate, thereby inhibiting anode failure and extending its lifetime. The SEM micrographs of the electrode after use revealed no obvious changes, indicating the good stability of the prepared DSA (Figure 2B).

The elemental composition of the DSA coating surface was determined by EDS (Figures 2C,D). The surface of the coated electrode was mainly composed of five elements (Sn, Ru, O, C, and Ir), while the element Ce was absent in the spectra. This can be attributed to the incomplete dissolution of Ce in isopropanol, resulting in lower Ce content in the coating. According to literature reports (Lassali et al., 2000), the addition of iridium oxide into the coating may effectively reduce the chlorine evolution potential of the electrode, as well as enhance the corrosion resistance of the coating toward the oxygen evolution reaction, thereby enhancing the lifetime of the electrode. The addition of Sn-containing compounds could improve the formation of multi-crystals grain on the coating surface, thereby raising the specific surface area. Moreover, Ru and Ce, as active elements in the coating, can enhance the electrocatalytic activity of the anode.

The SEM micrographs of Fe³⁺-TiO₂@CGS particles prepared by the sol-gel method revealed carbon in CGS existing in a flocculent amorphous morphology (Figures 3A–D ), with a relatively well-developed porous structure on the surface. In Figure 3C, white crystals looked uniformly distributed on CGS particles without agglomeration. The average particle size was estimated to 10 nm, and the structure of the particles appeared denser, indicating Fe³⁺-doped TiO₂ widely distributed on the CGS carrier surface. Some TiO₂ particles were embedded in the micropores of CGS (Figure 3D ), showing the pore size of CGS compatible with the size of TiO₂ crystals. However, the adsorption channels inside CGS were not completely blocked. Also, the dimensions of some TiO₂ grains looked larger than the surface micropores of CGS, hindering them from entering the micropores. Therefore, the as-prepared Fe³⁺-TiO₂@CGS particles may exhibit a certain adsorption capacity during “absorption + photocatalysis + electrocatalysis” process of Fe³⁺-TiO₂@CGS particles in the photoelectrical system designed for wastewater degradation.

As shown in Figure 3E, the gasification slag contained C, O, Ti, Si, Al, and Fe elements, in addition, the weight percentage of C and O elements were estimated to 58.38% and 18.08%, respectively. Thus, the prepared particles were successfully loaded on the Fe³⁺-doped TiO₂. Moreover, the TiO₂ particles were uniformly distributed on the CGS surface (TEM micrographs of Fe³⁺-TiO₂@CGS particles, Figures 3F,G ). The lattice spacings of 0.35 nm and 0.25 nm were also kept well with the facet of TiO₂ (101) and Fe₂O₃ (110), respectively (Figure 3G). Therefore, TiO₂ and Fe₂O₃ nanoparticles were successfully loaded at the pores of CGS particles, confirming the feasibility of the preparation of Fe³⁺ -TiO₂ @ CGS particles.

The absorption-desorption curves and pore size distribution profiles of Fe³⁺-TiO₂@CGS particles are presented in Figure 4A. The specific surface area, the average pore diameter, and the pore volume of Fe³⁺-TiO₂@CGS particles were determined as 264.3259 m²·g⁻¹, 0.247755gnn, and 0.247138 cm³·g⁻¹, respectively. Hence, the pores remained incompletely occupied after doping the gasification slag with Fe³⁺-loaded TiO₂. Table 2 revealed a decline of 59.01% in specific surface area of Fe³⁺-TiO₂@CGS particles after loading. Also, all gasification slag before and after loading displayed flourishing pore structures. Therefore, the high microporous structure of gasification slag was not destroyed during the doping process, conducive to the full contact between particle surface and pollutants.

To further confirm the structures of CGS and Fe³⁺-TiO₂@CGS, the FT-IR spectra were obtained. In the FT-IR sepctra (Figure 4B), it is found that the main skeleton of CGS is carbon-based material with corresponding vibration peaks. After the formation of Fe³⁺-TiO₂@ CGS composite material, the structure changes slightly with some vibration peaks disappear or appear. Combined with TEM and XRD experimental results, it reveals that the composite material is prepared successfully.

As shown in Figure 4C, it can be seen that there are two obvious characteristic peaks at 25.8° and 44.7°, assigned to the diffraction peaks of graphite crystallites of the gasification slag. When TiO₂ is loaded, the diffraction peaks are revealed in the composite material with the diffraction peak of Fe₂O₃, the reason is that the main skeleton of Fe³⁺-TiO₂@CGS composite material is tends to be amorphous structure compared with the raw materials. This phenomenon indicates that TiO₂ and Fe³⁺ particles are successfully composited on the CGS along with good stability and modifiability. Furthermore, the XRD patterns of the as-prepared Fe³⁺-TiO₂@CGS particles revealed peaks at 2θ of 25.24°, 37.94°, 48.03°, 54.97°, 62.58°, 69.96°, and 74.98° (Figure 4C). These peaks can be attributed to anatase crystal planes (101), (004), (200), (105), (204), (220), and (215), consistent with the characteristic peaks of TiO₂ standard card (JCPDS 021–1272). The peaks at 2θ of 26.51° and 35.32° were assigned to the rutile crystal plane (110) and (101), respectively. Note that characteristic peaks corresponding to the TiO₂ standard card (JCPDS 021–1276) were consistent with experimental results. The peaks at 2θ of 32.48° and 42.75° were linked to the crystal planes (104) and (113), respectively. The characteristic peaks corresponding to the Fe₂O₃ standard card (JCPDS 001–1053) were also consistent with experimental data, indicating gasification slag successfully loaded on nano-TiO₂ and mainly composed of anatase crystals. Fe₂O₃ crystals may exist, and Fe³⁺ was not fully infiltrated into TiO₂ as no obvious characteristic peak of amorphous carbon was observed. The latter may be due to the weak peak intensity of amorphous carbon, which was overlayed by TiO₂ characteristic peaks. The XPS diagram of Fe³⁺-TiO₂@CGS showed four characteristic peaks of C, O, Fe and Ti (Figure 4D). The peaks of C1s, O1s, Fe2p, Ti2p located at 284.8 eV, 530.3 eV, 711.7 eV and 459.0 eV. In Figure 4E, the peak at 284.8 eV is corresponded to the C1s, and the others at 286.5 eV, 288.9 eV are assigned to the C-O and O-C═O bonds of CGS. The peak at 530.3 eV of O1s, which indicated the forming of Ti-O-Fe bond in the TiO₂ lattices. The presence of -OH group was ascribed to the peak at 531.9 eV. Furthermore, the peaks at 711.7 eV and 724.8 eV in Fe2p corresponded to Fe2p_3/2 and Fe2p_1/2 of the metal-ion Fe³⁺. There are two peaks at 459.0 eV and 464.7 eV in the spectrum of Ti2p, which corresponding to Ti2p_3/2 and Ti2p_1/2 (Hassan et al., 2016; Singh et al., 2019). In Figure 4D, Fe³⁺ also entered the lattice of TiO₂, and the introduction of Fe³⁺ played an irreversible role in the electron trap. In other words, defects were introduced into the lattice after replacement of Ti⁴⁺ by Fe³⁺. The cross between Fe2p state and TiO₂ crystal valence band broadened the valence band and narrowed the energy gap, which can reduce electron-hole recombination and effectively improve photocatalytic efficiency.

To assess the photoelectric coupling effect of the three-dimensional electrode, the degradation of MB at a concentration of 600 mg·L⁻¹ and volume of 500 mL was carried out in the photoelectrical reactor (Figure 1). The process was performed either with photocatalysis under UV light, electrocatalysis with the three-dimensional electrode, or a combination of both in synergistic photoelectric catalysis experiments. The conditions consisted of a Fe³⁺-TiO₂@CGS particle of 5 g, NaCl electrolyte concentration of 0.15 mg·L⁻¹, treatment time of 1 h, and current density of 15.63 mA·cm⁻².

From Figure 5A, the decomposed MB fractions were estimated to 16.8% under UV irradiation alone and 38.5% when employing electrocatalysis alone. The decomposition of MB using the photoelectric synergistic catalysis system via a three-dimensional electrode was recorded as 61.3%, a value higher than the sum of both separately conducted catalytic methods. The synergistic factor (C = η₃/(η₁+η₂)×100%) was calculated as 1.11, indicating an obvious photoelectric synergistic effect in this system (Xu et al., 2008; Hou et al., 2017a).

The conditions of the (Gan et al., 2008) photoelectrocatalytic would influence the treatment effect of wastewater, including electrolyte type, current density, and initial pH of the solution. Based on previous works (Zeng et al., 2022; Zhang et al., 2022), an in-depth analysis of the supporting electrolyte, current density, and initial pH of the solution was carried out. The influencing mechanism of the degradation effect of methylene blue, as well as the optimal operating conditions, were obtained.

In order to study the effects on the degradation of MB, sodium salts (NaCl, Na₂CO₃, and Na₂SO₄) were employed as supporting electrolytes in separate experiments. For each experimental condition, the initial MB concentration was kept at 600 mg·L⁻¹ along with 5 g of Fe³⁺-TiO₂@CGS particles and a high-pressure Mercury lamp at a power of 600 W. The system was also purged with at the flow of 200 L·h⁻¹. At the initial pH 3, the electrolytes were added to adjust the conductivity to 6,000 μS·cm⁻¹, and the reaction was carried out for 60 min. Figures 5B–D gathers the results of the effects of process conditions, including electrolyte type, current density, and initial pH of the solution on the degradation rates of MB. The plots of MB concentration as a function of time in the photoelectric catalytic system containing different supporting electrolytes are provided in Figure 5B. After 10 min reaction, the concentration of MB decreased significantly in the presence of all three supporting electrolytes. After the addition of Na₂SO₄ to the system, the decomposition rate was accelerated. However, the degradation in NaCl and Na₂CO₃ as supporting electrolytes looked more effective than in Na₂SO₄ for the reaction proceeding for 20 min. The calculated degradation yields at 30 min reaction time were 99.78%, 99.19%, and 61.22%, with insignificant performance differences between NaCl and Na₂CO₃. The introduction of Cl⁻ in alkaline Na₂CO₃ solution as an electrolyte might be explained by the increase in the pH value to 10. When 4 mol·L⁻¹ hydrochloric acid was added to the solution, the pH value returned to 3. Therefore, the introduced Cl⁻ anions might determine the decomposition performance of the system. In the following 30 min reaction, the degradation in the three supporting electrolytes ceased, with final degradation yields determined as 99.95%, 99.32%, and 71.08%, respectively.

The conductivity of the solution may significantly influence the degradation process of organic matter (Solarska et al., 2006). The addition of the supporting electrolyte in the coupled photoelectric process could enhance the current efficiency, resulting in the effective degradation of wastewater pollutants. In the designed photoelectrical treatment system for MB, the reactions of Cl⁻ can be promoted under photoelectric coupling by introducing Cl⁻ to the supporting electrolyte. Certain concentrations of active chlorine may induce strong oxidizing properties (Wang et al., 2007). Also, large numbers of active substances, such as HO·, O⁻, and Cl⁻ may be generated under photocatalysis, conducive to increasing the degradation rates of organic matter (reactions 1–4). 2 C l − − 2 e − → C l 2 (1) C l 2 + H 2 O → H C l O + H C l (2) C l 2 + H R → R C l + H C l (3) H C l O + H R → R C l + H 2 O (4)

Using Na₂SO₄ as the supporting electrolyte, MB was decomposed at a certain degradation rate during the first 20 min of the reaction, leading to a decrease in the subsequent period. The reason for this might have to do with the generation of a certain amount of H₂O₂ from SO₄ ^2- in the Na₂SO₄ system via reactions (5) and (6) under photoelectric excitation. In turn, the formed H₂O₂ may promote the degradation of MB solution. At prolonged reaction times, H₂O₂ would gradually decompose in the presence of light, and SO₄ ^2- can capture H⁺ or HO· produced in the system to generate relatively weak oxidizing agent SO₄ ^2- via reactions (7) and (8) (Li et al., 2020). These mechanisms would explain the degradation rate in solution through this system, indicating the inorganic salt of NaCl was a better supporting electrolyte in the photoelectrical treatment of MB than Na₂CO₃ and Na₂SO₄. S O 4   2 − − 2 e − + h v → S 2 O 8   2 − (5) S 2 O 8   2 − + 2 H 2 O → 2 H S O 4 − + H 2 O 2 (6) H O ⋅ + S O 4   2 − → S O 4 − + O H − (7) h + + S O 4   2 − → S O 4 − (8)

To clarify the effect of the current density on the degradation of MB, the same above reaction conditions were employed to study the influence of the supporting electrolyte by selecting NaCl as the electrolyte. The reactions were carried out for 60 min at current densities of 15.63, 18.76, and 21.89 mA·cm⁻².

As shown in Figure 5C, the current density significantly affected the photoelectrical degradation of MB in solution. A good performance was achieved for both degradation efficiency and reaction rate. After 30 min reaction time, the degradation yield at the current density of 18.76 mA·cm⁻² exceeded 99%, while the same yield was achieved after 50 min at a current density of 15.63 mA·cm⁻². Therefore, within a certain concentration range, MB can be effectively degraded at increased current densities. However, the degradation effect varied slightly when the current density further rose from 18.76 mA·cm⁻²–21.89 mA·cm⁻².

The observed variation can be explained by the increase in the amount of electricity passing through the photoelectrical system in the same period with the current density, promoting the generation of active chlorine from Cl⁻. This considerably contributed to the degradation of wastewater, leading to better degradation of organic matter. Furthermore, the increase in current density provided sufficient energy for the reactions (9) and (10) to enhance the amount of HO· in the system, thereby promoting the degradation of organic matter. However, the excess H₂O₂ consumed strongly-oxidizing reagents, such as photo-generated holes or HO· in the system, inhibiting the degradation of wastewater. Consequently, the reaction rate gradually decreased or even ceased to occur. However, higher current densities in the coupled photoelectric process under laboratory conditions shortened the reaction time, enhancing the difficulty of reaction occurrence and appearance of intermediate products. Besides, the voltage increase with current density would more likely damage the electrode surface and greatly shortens its lifetime. By considering the above factors, the coupled photoelectric degradation of MB solution would be optimal to carry out at a current density of 18.76 mA·cm⁻², which was used for subsequent experiments. O 2 + 2 H + + 2 e − → H 2 O 2 (9) H 2 O 2 + h v → 2 H O ⋅ (10)

To investigate the effects of different initial pH values on the degradation of MB solution, the reaction was carried out at initial pH values of 3, 5, 7, 9, and 11 for 60 min under the same above conditions using NaCl as supporting electrolyte at the current density of 18.76 mA·cm⁻².

In the photoelectric catalysis process, Figure 5D shows the treatment effects of different initial pH values on MB decomposition. The reaction process of MB decreased significantly under all five conditions. However, the degradation rate first declined and then rose with pH value. In the neutral solution, the degradation displayed relatively low rates with a yield of only 61.38% after a reaction time of 20 min, while 90% decomposition was achieved under other conditions in the same time frame.

For the photoelectrical degradation of MB solution, the initial pH of the solution would affect the degradation of organic matter, as well as the surface charge of the catalyst to a certain extent. At pH < 7, H⁺ concentration existed at a relatively high levels, thereby reaction (9) would naturally proceed in the forward direction, leading to increased H₂O₂ concentration in the solution. The generated H₂O₂ underwent reaction (10) to form large amounts of HO· under the photoelectric process, thereby enhancing the ability of the degradation of MB in solution. As pH rose, OH⁻ under alkaline conditions promoted reaction (11) with dissolved oxygen generated under aeration conditions, thereby generating HO₂· that can directly oxidize organic matter. HO₂· also underwent reactions (12) and (10) to generate HO·, which can degrade organic matter. Meanwhile, the protonated TiO₂ surface was positively charged at low pH values, conducive to the transfer of photo-generated electrons to the TiO₂ surface. At much higher pH values, the surface became more negatively charged, thereby promoting the transfer of holes from the inside of particles to the surface. Therefore, the photoelectrical degradation of MB exhibited higher reaction rates under both lower and higher pH values than under neutral conditions. O 2 + 2 O H − → 2 H O 2 ⋅ (11) 2 H O 2 ⋅ → H 2 O 2 + O 2 (12)

Overall, MB can effectively be degraded in solution by the designed photoelectrical reactor. The trend of the MB concentration varied under different conditions owing to the effects of supporting electrolyte, current density, initial pH of the solution, and other factors. At pH 11, current density of 18.76 mA·cm⁻², and NaCl as the supporting electrolyte, the degradation yield of wastewater reached 99.98% after 60 min of photoelectrical treatment, demonstrating an outstanding degradation effect. At 60 min under these conditions, the instantaneous concentration of CO₂ detected by the miniature detector was just only 3,386 ppm, indicating that the amount of CO₂ released was very small, and a large amount of MB was converted into organic intermediates in solution.

During the initial 30 min of photoelectrochemical coupling treatment of MB solution in NaCl supporting electrolyte at pH 11 under a current density of 18.76 mA·cm⁻², the color of the solution significantly decreased from light blue to colorless. To gain a better understanding of the changes in MB with reaction time, the solutions at different stages were diluted five times and then subjected to spectroscopy. As shown in Figure 6A, the initial solution of MB exhibited two obvious characteristic peaks at 664 nm and 292 nm, related to the chromophore structure (thiol) and aromatic ring structure absorption peak of MB, respectively (Xiang et al., 2017; Aritonang et al., 2018)^. The intensities of these peaks decreased with reaction time. In addition, MB was almost fully degraded after 30 min, consistent with the previous results. To further quantify the MB degradation capacity in a three-dimensional photoelectric system (Figure 6B), the MB degradation amounts were expressed by: R=(A₀-A_t/A₀)×100%, where R is the MB degradation rate (Li et al., 2019; Wang et al., 2019), A₀ represents the initial absorbance, and A_t is the absorbance at a certain time. Under optimal conditions, the degradation rate of MB reached 99.56% at 30 min. The three-dimensional photoelectric system showed large degradation capacities of MB dyes by absorption, electrochemical action, generated HO·, and molecular interactions. Such excellent performances testify to the usefulness of Fe³⁺-TiO₂@CGS in the treatment of water pollution for target dye.

The degradation of MB was obvious in the constructed three-dimensional electrode photoelectric system. Based on relevant literature (Li et al., 2015; Qin et al., 2015; Hou et al., 2016; Hou et al., 2017b), as shown in Figure 7, the degradation effect may be related to several factors. The first had to do with direct anodic oxidation, in which the active metal component M on DSA anode surface can react with water to form free radicals, which when adsorbed on the electrode would form adsorbed free radicals M(HO·). The active metal component can also interact with free radicals to produce stronger oxides MO with higher valence. Finally, the organic functional group R of MB can directly be degraded on the anode surface through the oxidation of M(HO·) and MO attached to the anode surface. Second, in the three-dimensional electrochemical reaction system, oxygen would be generated by electrolysis and the dissolved oxygen would reduce at the anode to form the strong oxidizing substances H₂O₂, which, in turn, further decompose to form strong oxidizing free radicals HO· and HO₂· to degrade MB. Third, TiO₂ is an n-type semiconductor with band gap of 3.2 eV (anatase). When it receives ultraviolet light with a wavelength less than or equal to 387.5 nm, it will generate electrons and hole. The electron-hole h⁺ generated by the excitation of TiO₂ in Fe³⁺-TiO₂@CGS particles would react with the pollutants adsorbed on the catalyst surface to degrade MB. The H₂O and OH⁻ adsorbed on the catalyst surface in solution would be in contact with the electron holes to form the active group HO·, conducive to degradation of MB. Meanwhile, O₂ ⁻· and HO₂· produced by the reaction of dissolved oxygen with photogenerated electrons could degrade MB. Fourth, Fe³⁺-TiO₂@CGS particles in the three-dimensional electrode photoelectric system would be repolarized due to voltage excitation, similar to the anode of the system. The surface of the Fe³⁺-TiO₂@CGS particles can adsorb MB and directly capture electrons of some unstable organic groups during the reaction, leading to the degradation of MB. In addition, HO· and HO₂· produced in the reaction would degrade MB effectively. Therefore, the pore structure and adsorption capacity of Fe³⁺-TiO₂@CGS particles were conducive to improving the degradation ability of MB.

The gasification slag with high carbon residue contents not only has an adjustable pores and functional groups that can enrich MB on its surface, but also can be used as a carrier for Fe³⁺ and TiO₂ for the further preparation of Fe³⁺-TiO₂@CGS particles. As is known to all, TiO₂ is a semiconductor material in the system of photoelectric, electron transition from valence band to conduction band, it will produce strong reactive oxidation groups and holes, but also can further promote the ability of electrochemical reaction. In addition, the doping of Fe³⁺ contributes to the formation of Fe³⁺/Fe²⁺ in solution, which leads to the formation of Fenton reaction (Fe²⁺+H₂O₂→Fe³⁺+HO⁻ + HO·)at the cathode. And then the MB is oxidized to organic intermediates, and further degraded into CO₂ and H₂O under the action of photoelectric fields. Undoubtedly, the different components of Fe³⁺-TiO₂@CGS particles are the fundamental factor for synergistic interaction of the absorption, photocatalysis, three-dimensional electrode electrocatalysis, and Fenton technology for the treatment of the industrial waste water. All in all, it will propose a basic idea for the treatment of the organic pollutants.