Enhancement of Z-scheme behavior by coupling local WO_3−x–WS₂ with Au nanoparticles for efficient photoelectrochemical degradation of methylene blue

Abstract

The removal of water pollutants has become the foremost priority in environmental remediation. To overcome this issue, photoelectrochemical (PEC) organic-dye degradation is preferred as it meets the demands of sustainability and eco-friendliness. For efficiently harvesting solar energy and facilitating carrier migration, semiconductors with appropriate band gaps, high light absorption capacity, and favorable reaction kinetics have been explored to drive PEC organic-dye conversion. In this work, we prepared a WO_3−x–WS₂ heterostructure embedded with Au nanoparticles by hydrothermal and photodeposition strategies. The optimal WO_3−x@Au–WS₂ ternary catalyst achieved the removal of 98.52% methylene blue in a neutral aqueous solution within 60 min under visible light, which was 1.7 times higher than that of bare WO_3−x. In the WO_3−x@Au–WS₂ ternary heterostructure, a Z-scheme of electron transfer paths was formed to promote the migration of photogenerated carriers. Our study offers new insights into the catalytic mechanism of ternary heterostructures, providing a promising strategy for improving the efficiency of wastewater treatment.

---

1. Introduction

Photoelectrocatalytic (PEC) degradation of methylene blue (MB) has attracted widespread attention in recent years as an efficient method for water pollution control.^1–4 Benefiting from the reconciliation of photocatalysis^5–7 and electrocatalysis,^8–10 photoelectrocatalysis with an internal electric field that inhibits the recombination of photogenerated electron and hole pairs can reduce energy consumption and improve the catalytic degradation performance. The superior efficiency of PEC degradation fundamentally arises from three key characteristics: (i) the generation of abundant light-excited charge carriers,¹¹ (ii) a lowered energy barrier for dye oxidation,¹² and (iii) the efficient production of highly active free radicals, notably hydroxyl radicals (·OH) and superoxide radicals (·O₂⁻), on the catalyst surface. Consequently, great efforts in recent decades have focused on developing catalysts with optimized bandgap structures, high charge carrier mobility, and robust chemical stability to maximize PEC degradation performance.^13,14

With a wide band gap (2.6–2.8 eV), WO₃, as a semiconductor¹⁵ that is capable of generating photogenerated electron–hole pairs under visible-light irradiation, suffers from a weak visible-light absorbance (<460 nm) and a gradual loss of photoactivity due to the accumulation of peroxo-species on the surface.^16–19 Researchers are devoted to designing WO₃-based heterostructures²⁰ for the purpose of improving the separation of photogenerated charge carriers.^21–24 Previous research studies have indicated that the combination of WO₃ with suitable band-gap semiconductors to form a Z-scheme heterostructure is a feasible approach to promote the separation of charge carriers.^25,26 Cheng et al. have prepared a ternary film of CuO–Cu₂O/WO₃, achieving a novel Z-scheme for the efficient PEC removal of deoxynivalenol. The degradation efficiency reached 87.6% after 180 minutes of PEC treatment. The experimental findings demonstrated that constructing a Z-scheme heterostructure enables efficient charge carrier separation and inhibits carrier complexation.²⁷ Wang et al. prepared a PCN/WO₃ Z-scheme heterostructure through a solid-state annealing method for the efficient photoreforming of polylactic acid plastics. They found that the formation of a Z-scheme heterostructure can effectively promote charge separation and enhance the redox driving force of the catalyst, which could improve the catalytic activity.²⁸ Therefore, the construction of a Z-scheme heterostructure is indeed a feasible way to improve the degradation performance of catalysts.^29–32

In situ sulfurization is an effective means of achieving heterogeneous structures.^33–36 WO₃ can form the Z-scheme heterogeneous structure with WS₂ to improve light absorption for highly efficient PEC degradation.^37,38 WS₂ has a 2D layer structure made of building blocks of WS₂ stacked together by van der Waals and exhibits a low band gap (1.4–2.4 eV).^39,40In situ sulphation growth of WS₂ is based on WO₃ structures with better interfaces for efficient PEC degradation.⁴¹ Meanwhile, Au acting as bridges within the catalyst structure can further enhance charge carrier transfer and inhibit electron–hole recombination.^42–45 The formation of a Z-scheme heterostructure confers distinct advantages such as enhanced charge separation efficiency and spatially separated redox sites, which enable more stable and enhanced redox potentials than conventional heterostructures.^46–49

Current WO₃-based heterostructures still suffer from a persistent issue of weak interfacial bonding,^50,51 resulting in high charge-transfer resistance.^52,53 Furthermore, metallic nanoparticles such as Au are often mono-functional and lack an integrated synergistic design.^54–56 Margaux Desseigne et al. systematically investigated the effects of the dispersion and loading amounts of Au nanoparticles on photocatalytic performance in their study on the Au/WO₃ composite system. It was pointed out that although Au enhances the visible-light response, the system still faces challenges such as nanoparticle aggregation and insufficient optimization of the metal–semiconductor interface, which limit the charge transfer efficiency across the interface.⁵⁷ Ali Can Güler et al. incorporated Au nanoparticles into BiVO₄/WO₃ photoanodes to enhance charge carrier separation through the LSPR of Au and the formation of a Schottky junction. Despite the notable improvements achieved, Au was primarily regarded as a metallic enhancer rather than an integral component of a synergistically designed system.⁵⁸ In this work, we demonstrate the rational design and construction of a Z-scheme heterostructure catalyst, WO_3−x@Au–WS₂, for PEC degradation of MB as a target pollutant under visible light. Through controlling the hydrothermal synthesis time and temperature, the synthesized WO_3−x@Au–WS₂ catalyst has achieved the best performance degradation of 98.52% in 60 minutes, which is better than most WO₃-based materials. We detected ·OH and ·O₂⁻ free radicals using ESR, which can better improve the catalytic degradation. Furthermore, the electronic structure of WO_3−x@Au–WS₂ was analyzed by density functional theory (DFT). It was found that the formation of the ternary structure leads to local metallization of the material owing to the presence of Au, which further increases the electron transfer rate and thus the degradation performance. The successful preparation of the WO_3−x@Au–WS₂ Z-scheme heterostructure paves the way for exploring the Z-scheme heterostructure system for efficient degradation.

2. Experimental section

2.1 Materials

Propanol (C₃H₆O, AR, 99.7%), ethanol (C₂H₅OH, AR, 99.7%), thiourea (CH₄N₂S, AR), methylene blue (C₁₆H₁₈CIN₃S·3H₂O), sodium oxalate (Na₂C₂O₄, AR, 99.8%) and anhydrous sodium sulphate (Na₂SO₄, AR, 99.0%) were purchased from Sinopharm Chemical Reagent Co. Hydrochloric acid (HCl, 36%) was purchased from Henan Kaifeng Dongda Chemical Co. Sodium citrate (C₆H₅Na₃O₇·2H₂O, 99%), sodium tungstate (Na₂WO₄·2H₂O, AR, 99.5%), and chloroauric acid trihydrate (HAuCl₄·3H₂O, 48–50% Au-based) were purchased from Shanghai Macklin Biochemistry Technology Co., Ltd. Ammonium chloride (NH₄Cl, 99.5%) was purchased from Tianjin Damao Chemical Reagent Factory. FTO conductive glass (7 Ω, transmittance ≥80%) was purchased from Wuhan Jingge Solar Technology Co. A 30% aqueous solution (H₂O₂) of hydrogen peroxide was purchased from Tianjin Tianli Chemical Reagent Co. Tungstic acid (H₂WO₄, AR) was purchased from Chemical Reagent No. 2 Factory. The deionized water used in this experiment was homemade in the laboratory with a resistivity test value of 18.2 MΩ cm⁻¹.

2.2 Preparation of catalyst material samples

The WO₃ seed layer was deposited onto cleaned FTO substrates. The pH of the precursor solution, prepared using sodium tungstate as the tungsten source with sodium oxalate and ammonium chloride additives, was adjusted with concentrated hydrochloric acid. WO_3−x films supported on FTO were then hydrothermally synthesized under systematically varied temperatures and durations. Subsequently, Au nanoparticles were deposited onto WO_3−x films via photodeposition, yielding WO_3−x@Au composites. These WO_3−x and WO_3−x@Au substrates were then subjected to in situ sulfurization to produce WO_3−x–WS₂ and WO_3−x@Au–WS₂ heterostructures. The fabricated materials underwent comprehensive structural and photoelectrocatalytic characterization. Full experimental details are provided in the SI.

2.3 Material characterization

The phase composition and chemical state of the samples were characterized by X-ray diffraction (XRD, X'PERT MPD Pro), X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha) and Raman spectroscopy (Raman, Horiba LabRAM HR Evolution). The surface morphology of the samples was obtained by field emission scanning electron microscopy at 10 kV (FE-SEM, FEI Nova NANOSEM 400), TEM, HR-TEM, EDS and elemental mapping (JEM-F200). Free radical and retrograde studies of the samples during the reaction were carried out using a steady-state/transient fluorescence spectrometer (Edinburgh FLS1000). The detection of absorbance was carried out using a visible photometer (V-T3). Ultraviolet-visible (UV-vis) spectra were recorded using a Pgeneral TU-1901 Dual-beam UV-vis spectrophotometer equipped with an integrating sphere, and BaSO₄ powders were used as a reflectance standard.

2.4 Experiments for PEC degradation

The photoelectrocatalytic activity of the prepared samples was evaluated by simulating the degradation of MB solution under visible light. In the photoelectrocatalysis experiment, a magnetic stirrer was placed at the bottom of a quartz glass reactor, the bias voltage was set to 0.7 V vs. Ag/AgCl, and the xenon lamp intensity was adjusted to 100 mW cm⁻². The reactor was placed approximately 5 cm away (20 mW cm⁻²) from the xenon lamp light source. Then, 30 mL of 10 ppm MB solution was added to the reactor and the prepared sample was held submerged in the solution using an electrode holder. Before the start of the photoreaction, it was left for 30 minutes in dark to reach adsorption equilibrium. After the start of the photoreaction, 3 mL of the mixed solution was taken at intervals and centrifuged, and the supernatant was extracted. The absorbance at 664 nm (absorption peak of MB) was determined using a UV-visible photometer.

3. Results

Different samples of WO_3−x were synthesized on the seed layer (Fig. 1a) by a hydrothermal method. The microstructures of the synthesized samples were observed by scanning electron microscopy (SEM). We first synthesized WO_3−x at 180 °C with different hydrothermal times (Fig. S1). At hydrothermal times of 4 h and 8 h, the morphology of WO_3−x still has agglomerated clusters in addition to lamellar structures. After 12 h, the morphology of WO_3−x changes almost to a uniform lamellar structure with a thickness of about 0.1 μm (Fig. 1b). When the reaction time was further extended to 36 h, WO_3−x appeared as denser and thicker. Afterwards, we synthesized WO_3−x by hydrothermal treatment for 12 h at different temperatures (Fig. S2). The WO_3−x synthesized at 120 °C demonstrates uniform distribution. As the temperature increases, dissolution–recrystallization processes progressively transformed the material into lamellar architectures. At 180 °C, a lamellar structure thinner than the sample prepared at 140 °C and 160 °C was formed. In contrast, WO_3−x synthesized at 200 °C showed overgrowth.

Fig. 1 SEM images of (a) WO₃ seed layers, (b) WO_3−x synthesized at 180 °C for 12 h, and (c) WO_3−x–WS₂ composite catalyst. (d) Schematic of morphological changes after hydrothermal treatment and sulphurisation. Comparison of the efficiency of WO_3−x catalysts synthesized at (e) 4 h, 8 h, 12 h, and 36 h for the degradation of MB and (f) at 160 °C, 180 °C, and 200 °C for the degradation of MB. (g) Comparison of the MB degradation efficiency of WO_3−x and WO_3−x–WS₂.

The optimal reaction conditions were judged by our methylene blue degradation experiments on all samples. From Fig. 1d and e and S3, it can be seen that the performance of WO_3−x synthesized at a hydrothermal temperature of 180 °C and a hydrothermal time of 12 h is optimal. The degradation rate reached 56.72% within 60 min. In addition, the samples synthesized at 180 °C for 12 h showed good stability in the repeatability test (Fig. S4 and S5). From these points of view, we sulfurized the optimal WO_3−x to form a WO_3−x–WS₂ binary catalyst (Fig. S6) with an average thickness of about 2 μm (Fig. S11). Fig. 1c shows that the morphology of the synthesized sample has not changed significantly after sulphurisation. The degradation rate of the binary WO_3−x–WS₂ was 80.2% in 60 min, which was 1.41 times higher than that of single WO_3−x (Fig. 1f).

In order to further improve the degradation efficiency, WO_3−x@Au–WS₂ was prepared by photodeposition. In the HR-TEM image of the ternary junction of WO_3−x@–Au–WS₂ (Fig. 2c), a lamellar structure, WO_3−x bulk material, with a lateral dimension of about 1 μm and a width of about 0.2 μm can be observed. Meanwhile, the as-prepared WS₂ appears to be layered and in close contact with the WO_3−x bulk. According to EDS (Fig. 2c), W, S and O elements are evenly distributed. In addition, gold nanoparticles of about 14 nm are bound to WO_3−x and WS₂ due to photodeposition. The lattice spacings are 3.64 Å for the (200) plane of WO_3−x and 6.18 Å for the (002) plane of WS₂ (Fig. 2b), whereas the lattice spacing of the (111) and (200) faces of Au is 2.34 Å and 2.02 Å (Fig. 2a).

Fig. 2 High-resolution transmission electron microscopic (HR-TEM) images of WO_3−x@Au–WS₂ junction particles. (a) Lattice fringes of Au and (b) WO_3−x and WS₂. (c) TEM image and EDS mapping of WO_3−x@Au–WS₂ (O, W, S and Au). (d) XRD patterns of WO_3−x, WO_3−x–WS₂ and WO_3−x@Au–WS₂. (e) Raman spectra of WO_3−x, WO_3−x–WS₂ and WO_3−x@Au–WS₂. (f) Magnified view of Raman spectral regions.

The XRD analysis was performed to determine the crystal phase composition of the synthesized materials. The XRD patterns of WO_3−x, WO_3−x–WS₂ and WO_3−x@Au–WS₂ are displayed in Fig. 2d. The diffraction peaks of WO_3−x at 22.54°, 38.35°, and 44.70° correspond to the (001), (210), and (211) facets (JCPDS No. 75-2187). The diffraction peaks at 14.07°, 24.39°, 28.34°, and 36.73° correspond to the (002), (001), (004) and (102) planes of WS₂ (JCPDS No. 08-0237). The (200) and (111) crystal faces appearing at 45.97° and 40.92° are distinct phases of Au (JCPDS No. 04-0784). It follows that WO_3−x is predominantly monoclinic. After sulphurisation, the WO_3−x diffraction peaks shift to lower angles, which confirms the doping of S atoms into the WO_3−x lattice, resulting in a shortening of W–O bond lengths. In ternary WO_3−x@Au–WS₂, the (001) surface diffraction angle decreases. At the same time, the peak position of WS₂ decreased. Upon Au loading, electron injection and stress release expand the WO_3−x lattice and cause an increase in the WS₂ layer spacing, exacerbating the changes in the XRD peaks.

The Raman spectra were recorded to analyze the atomic vibrations in WO_3−x, WO_3−x–WS₂, and WO_3−x@Au–WS₂ (Fig. 2e). WO_3−x at 264.65 cm⁻¹ and 336.23 cm⁻¹ corresponds to the O–W–O bending vibration mode. The Raman peaks at 687.62 cm⁻¹ and 802.45 cm⁻¹ are assigned to W–O–W telescopic vibrations. The E¹_2g (352.77 cm⁻¹) and A¹_g (417.25 cm⁻¹) peaks of WS₂ appear in WO_3−x–WS₂ and WO_3–x@Au–WS₂. The peak of WO_3−x is blue-shifted, indicating that the formation of the WO_3−x–WS₂ heterostructure leads to lattice distortion. The peak spacing between E¹_2g and A_1g of WS₂ increases to 65.52 cm⁻¹, which may be due to Au-induced charge transfer or localized stresses, leading to interlayer interactions in WS₂. The WO_3−x peak, however, undergoes a minor blue-shift compared to that in the binary system, indicating that the effect of Au on WO_3−x is weaker than that of WO_3−x@Au–WS₂. Furthermore, the blue-shift observed in WO_3−x suggests that the localized surface plasmon resonance (LSPR) of Au strengthens W–O bonds, likely shortening their equilibrium length.⁵⁹

The photoluminescence (PL) spectra were recorded to explore the transfer and recombination of photogenerated charge carries inside the bare WO_3−x, WO_3−x–WS₂, and WO_3−x@Au–WS₂. As shown in Fig. 3b, the PL intensity of WO_3−x–WS₂ and WO_3−x@Au–WS₂ dramatically decreases in comparison with WO_3−x, indicating the reduced charge carrier recombination. Fig. 3c shows that the decreased average lifetime of photoelectrons in WO_3−x (1.19 μs) WO_3−x–WS₂ (0.91 μs) and WO_3−x@Au–WS₂ (0.74 μs) fitted from the time-resolved PL spectra proves that the constructed heterostructures can facilitate rapid carrier transfer, which is conducive to enhancing the PEC performance.⁶⁰ UV-visible absorption spectroscopy (Fig. 3a) reveals a progressive red-shift in the absorption onset from ∼392 nm for WO_3−x to 410 nm for the WO_3−x–WS₂ binary composite, and further to 436 nm for ternary WO_3−x@Au–WS₂. This shift signifies enhanced visible-light harvesting across the series. The extended absorption profile of WO_3−x@Au–WS₂ is attributed to the LSPR of Au nanoparticles, which intensifies visible-light absorption and promotes the generation of photogenerated charge carriers for catalytic reactions.

Fig. 3 (a) UV-visible DRS curves, (b) PL spectra, and (c) time-resolved PL spectra of WO_3−x, WO_3−x–WS₂ and WO_3−x@Au–WS₂.

Characteristic peaks of W 4f, O 1s, S 2p and Au 4f have been revealed by XPS analysis (Fig. 4). It is noteworthy that the presence of oxygen vacancies and adsorbed oxygen may affect the chemical shifts of the characteristic peaks by altering the surface electronic states. The synthesized WO_3−x exhibits energies of 37.68 eV and 35.54 eV, corresponding to W 4f_5/2 and W 4f_7/2 states of W⁶⁺, respectively. In WO_3−x–WS₂, XPS reveals the characteristic W⁴⁺ states at binding energies of to 34.38 (W 4f_5/2) and 32.28 eV (W 4f_7/2). The WO_3−x@Au–WS₂ composites exhibit analogous W⁴⁺ signatures shifted to higher energies at approximately 34.52 eV and 32.36 eV, respectively. In WO_3−x–WS₂ and WO_3−x@Au–WS₂, the partial conversion of W⁶⁺ to W⁴⁺ leads to a decrease in the relative content of W⁶⁺. The XPS feature signal of W⁶⁺ shows a significant attenuation. The deconvolution of O 1s spectra reveals three components across all materials: chemisorbed or dissociated oxygen (O_c), oxygen vacancies (O_v), and lattice oxygen in W–O bonds. Notably, the W–O binding energy decreases significantly in both WO_3−x–WS₂ (530.48 eV) and WO_3−x@Au–WS₂ (530.37 eV) relative to pristine WO_3−x. This destabilization is particularly pronounced in WO_3−x@Au–WS₂, where Au promotes interfacial charge transfer to adjacent W atoms and induces bond polarization. The S 2p spectra of WO_3−x–WS₂ and WO_3−x@Au–WS₂ show characteristic doublets (S 2p_3/2 and S 2p_1/2). A small but positive shift (∼0.2 eV) in WO_3−x@Au–WS₂ indicates reduced electron density at sulfur sites, which may be attributed to Au-mediated charge transfer. Concurrently, a significant negative shift (1.6–1.7 eV) of Au 4f peaks indicates electron enrichment from oxygen vacancy donors. The complementary binding energy shifts in S 2p (positive) and Au 4f (negative) could evidence interfacial electron transfer, while Au acts as a bridge to enhance the electron transfer between WO_3−x and WS₂.⁶¹

Fig. 4 X-ray photoelectron spectroscopy (XPS) results of W 4f, S 2p, O 1s, and Au 4f orbitals.

The photoelectrocatalytic activity of the prepared catalysts was determined by measuring the degradation of MB solution under UV irradiation.⁶² According to the standard curve of methylene blue (Fig. S7), we can get that the absorbance shows a linear relationship with the concentration of the solution. Fig. 5a shows the degradation efficiency of the synthesized materials for photoelectrocatalytic degradation of methylene blue. As can be seen from Fig. 5a, the ternary WO_3−x@Au–WS₂ degradation is the best. The degradation rate of WO_3−x alone in degrading MB was only 55.1% at 60 min, the degradation rate of WO_3−x@Au on MB reached 73.9% at 60 min, and the degradation rate of WO_3−x–WS₂ on MB reached 80.2% at 60 min. In contrast, the degradation rate of WO_3−x@Au–WS₂ had reached 98.3% at 50 min, and MB almost completely was degraded to 98.52% at 60 min. Moreover, all synthesized materials exhibit excellent stability while retaining their degradation efficacy over multiple cycles (Fig. 5d–f). In comparison with other studies of WO₃-based materials, our prepared WO_3−x@Au–WS₂ shows excellent performance (Fig. 5g and h, Table S3). In order to evaluate the role of photocatalysis and electrocatalysis during photoelectrocatalysis, we performed separate photocatalytic and electrocatalytic tests using ternary WO_3−x@Au–WS₂. Transient photocurrent response under chopped light illumination (Fig. 5c) demonstrates rapid and repeatable response of WO_3−x, WO_3−x–WS₂ and WO_3−x@Au–WS₂, in line with the LSV curves (Fig. S9a). In addition, the stable photocurrents after multiple cycles demonstrated great reproducibility of the WO_3−x@Au–WS₂ photoelectrode (Fig. S13). The universal applicability of the material for degrading organic dyes was also investigated. Degradation tests conducted on RhB and MO (Fig. S14) demonstrate that WO_3−x@Au–WS₂ retains high photoelectrocatalytic activity, achieving stable and efficient degradation of these diverse pollutants within 60 minutes. According to the LSV data (Fig. S9a), it can be seen that the catalyst photogenerated electrons increase after the addition of light, while the photocurrent decreases without light, which prove photocatalysis in the PEC process, whereas electrocatalysis plays the role of applying an electric field for the carrier mobility to play a promotional role only. Cyclic voltammetry (CV) (Fig. S8) revealed that the WO_3−x@Au–WS₂ structure exhibits a higher double-layer capacitance (Fig. S9b), indicating an increased number of electrochemically active sites and consequently enhanced catalytic activity. WO_3−x@Au samples were prepared (Fig. S10), and the results indicate that incorporating Au enhances the degradation efficiency.

Fig. 5 (a) Comparative degradation performance diagram of WO_3−x, WO_3−x–WS₂, WO_3−x@Au and WO_3−x@Au–WS₂. (b) Comparison of WO_3−x@Au–WS₂ degradation performance under dark, photocatalytic, electrocatalytic and photoelectrocatalytic conditions. (c) I–t curves for different materials at 0.7 V vs. Ag/AgCl. Stability of (d) WO_3−x, (e) WO_3−x–WS₂, and (f) WO_3−x@Au–WS₂. (g and h) Degradation efficiency of pollutants by different WO₃-based catalysts.

In order to verify the photoelectrocatalytic generation of hydroxyl and superoxide radicals in the reaction, ESR tests by utilizing DMPO for the capture of hydroxyl radicals and superoxide radicals were carried out. In an aqueous H₂O solvent, DMPO is more likely to capture hydroxyl radicals and less likely to capture superoxide radicals. In methanol solvents, the capture of superoxide radicals with DMPO is easier. As illustrated in Fig. 7a and b, the formation of hydroxyl radicals (·OH) was confirmed through trapping with DMPO, as evidenced by the characteristic quartet signal with an intensity ratio of 1/2/2/1 corresponding to DMPO–OH. In addition, a distinct signal for DMPO–˙O₂ with a 1/1/1/1 peak intensity ratio was observed, indicating the concurrent generation of superoxide radicals. These results verify that the reactions (H₂O + h⁺ → H⁺ + ·OH, O₂ + e⁻ → ·O₂⁻) occur during the photoelectrochemical process, thereby leading to the degradation of MB.

To elucidate the PEC enhancement mechanism, the electronic structure and absorption properties have been investigated based on first-principles calculations. First, an atomic structure model of WO_3−x–WS₂ was constructed (Fig. S12a). The electronic band structure and partial density of states (PDOS) diagrams reveal that WO_3−x–WS₂ retains its semiconductor characteristics (Fig. 6a). Furthermore, the band gap is reduced due to orbital hybridization. For WO_3−x@Au (Fig. S12c), electron injection from Au into WO_3−x results in Fig. 6b: (i) an increased PDOS near the conduction band of WO_3−x, (ii) a reduced band gap, and (iii) an emerging metallization tendency. To determine the optimal Au position, we analyzed WO_3−x–WS₂@Au (Fig. S14a) and WO_3−x@Au–WS₂ (Fig. S14c), both of which exhibit metallization in the presence of Au (Fig. 6c and e). Fig. 6c reveals partial overlap between the WO_3−x valence band peak and Au localized states, while the WS₂ peak maintains spatial separation. As for E_F, the PDOS of Au exhibits spillover into WO_3−x, yet shows negligible overlap with WS₂. This electronic configuration hinders electron transfer from WO_3−x to WS₂, which correlates with the charge density distribution observed in Fig. 6d. Fig. 6e exhibits near-complete spectral overlap of the valence band peaks from WO_3−x, Au, and WS₂. At E_F, the Au-bridged interface significantly enhances the PDOS of WS₂. This configuration establishes an electron transfer channel through Au, facilitating efficient carrier injection from WO_3−x to WS₂ with improved charge mobility. As shown in Fig. 6d and S14b, electrons transfer from Au to WO_3−x and become localized at the WO_3−x–WS₂ interface in the WO_3−x–WS₂@Au heterostructure. In contrast, for the WO_3−x@Au–WS₂ structure (Fig. 6f and S14d), the electron cloud delocalizes across the Au nanoparticle. These spatial distribution profiles, corroborated by XPS analysis, collectively demonstrate that Au serves as an electronic bridge mediating the interaction between WO_3−x and WS₂.

Fig. 6 Band structure diagrams and PDOS of (a) WO_3−x–WS₂, (b) WO_3−x@Au, (c) WO_3−x–WS₂@Au and (e) WO_3−x@Au–WS₂. Side views of the charge density difference of (d) WO_3−x–WS₂@Au and (f) WO_3−x@Au–WS₂. The blue and yellow regions indicate the electron density loss and gain, respectively. Red: S atoms; grey: W atoms; yellow: O atoms; gold: Au atoms.

Based on the collective experimental evidence, a direct Z-scheme charge transfer mechanism within the WO_3−x@Au–WS₂ heterostructure is proposed, which effectively suppresses charge recombination and enhances photoelectrocatalytic (PEC) degradation performance. Under visible-light irradiation, both WO_3−x and WS₂ generate electron–hole pairs. In WO_3−x, the photogenerated holes in the valence band (VB) exhibit strong oxidation potential, while the electrons in the conduction band (CB) possess weak reduction capability.⁶³ In contrast, WS₂ yields highly reductive electrons in its CB and weakly oxidative holes in its VB. A Z-scheme pathway is established wherein the less reactive electrons from the CB of WO_3−x migrate to Au nanoparticles and subsequently recombine with the VB holes of WS₂.^64,65 This process preserves the highly oxidative holes in WO_3−x and the strongly reductive electrons in WS₂, thereby maintaining their respective redox capacities.⁶⁶ The accumulated holes in WO_3−x oxidize H₂O to generate hydroxyl radicals (·OH), while the accumulated electrons in the CB of WS₂ reduce dissolved oxygen to superoxide radicals (·O₂⁻). These reactive oxygen species effectively drive the degradation of MB. XPS valence band spectra (Fig. 4) provide direct evidence that the negative shift of 1.61 eV in the Au 4f_7/2 peak indicates electron accumulation on Au, while positive shifts in W 4f_7/2 (+0.18 eV) and S 2p_3/2 (+0.20 eV) suggest electron depletion in WO_3−x and WS₂, confirming electron transfer from WO_3−x to Au and then to WS₂, which is consistent with a Z-scheme configuration.^67,68 DFT calculations (Fig. 6e) further reveal significant orbital hybridization between the d-orbitals of Au and the O 2p orbitals of WO_3−x and S 3p orbitals of WS₂ near the Fermi level, forming a continuous electron transport channel that lowers the energy barrier for interfacial charge transfer. Under an applied bias of 0.7 V vs. Ag/AgCl, the electric field synergistically promotes charge separation and further restrains carrier recombination,^54,61 reinforcing the Z-scheme pathway. Additionally, the localized surface plasmon resonance (LSPR) effect of Au enhances the local electromagnetic field and increases the carrier generation rate. As illustrated in Fig. 7c and d, the preserved holes in WO_3−x and electrons in WS₂ facilitate dual-site redox catalysis. MB adsorbed on the catalyst surface is either directly oxidized by holes or attacked by ·OH, forming intermediates that are further oxidized by ·O₂⁻ and ultimately mineralized to CO₂ and H₂O. This concerted mechanism significantly enhances the overall PEC degradation efficiency.

Fig. 7 ESR result: (a) ·O₂⁻ in H₂O; (b) ·OH in MeOH. (c) Structure of the ternary heterostructure energy band. (d) PEC degradation mechanism diagram.

4. Conclusions

In summary, the WO_3−x@Au–WS₂ Z-scheme heterostructure was successfully prepared and used for the PEC degradation of MB. The ternary WO_3−x@Au–WS₂ composite demonstrates enhanced PEC degradation performance (98.52% of MB degradation at 60 minutes). Comprehensive characterization and computational analysis reveal that Au nanoparticles enhance the catalytic activity of WO_3−x and WS₂ through two synergistic core mechanisms: (i) acting as an electronic bridge that facilitates Z-scheme heterojunction formation between WO_3−x and WS₂, which simultaneously induces localized metallization to accelerate interfacial electron transfer; (ii) the localized surface plasmon resonance (LSPR) of Au enhancing light absorption. Additionally, oxygen vacancies in WO_3−x optimize interfacial contact, synergizing with the above mechanisms to contribute to the light degradation performance. These factors work synergistically, thereby collectively contributing to the high degradation performance observed. This study provides a way to design an efficient photocatalyst for the degradation of organic pollutants.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI).

Supplementary information: additional experimental details, characterization data (XTD, XPS, Raman, TEM, SEM, UV-vis, EDS, ESR), Experiments for PEC degradation; the preparation of catalysts; DFT calculations, and comparative performance tables; electrocatalytic behavior of catalysts, and the tables of degradation efficiency of pollutants by different WO₃-based catalysts. See DOI: https://doi.org/10.1039/d5cy01071c.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. U21A20317, U23A20606, 22109121, and 22309141), the fund from Key Laboratory of Hubei Province for Coal Conversion and New Carbon Materials, Wuhan University of Science and Technology (No. WKDM202204) and the National Key Research and Development Program of China (No. 2022YFB3207200). The numerical calculation was supported by the High-Performance Computing Center of Wuhan University of Science and Technology.

References

1. D. Liu, R. Tian, J. Wang, E. Nie, X. Piao, X. Li and Z. Sun, Chemosphere, 2017, 185, 574–581.
2. J. Lin, R. Zong, M. Zhou and Y. Zhu, Appl. Catal., B, 2009, 89, 425–431.
3. F. Nandjou and S. Haussener, J. Phys. D: Appl. Phys., 2017, 50, 124002.
4. N. Kaur, N. Kodan, D. Sharma, A. Ghosh, P. Bisht, R. Singh and B. R. Mehta, Renewable Energy, 2025, 240, 122229.
5. R. Shi, H. F. Ye, F. Liang, Z. Wang, K. Li, Y. Weng, Z. Lin, W. F. Fu, C. M. Che and Y. Chen, Adv. Mater., 2018, 30, 1705941.
6. Z.-G. Zhao and M. Miyauchi, Angew. Chem., Int. Ed., 2008, 47, 7051–7055.
7. F. Xu, C. Lai, M. Zhang, B. Li, S. Liu, M. Chen, L. Li, Y. Xu, L. Qin and Y. Fu, J. Colloid Interface Sci., 2021, 601, 196–208.
8. Y. Zhou, X. Ren, Q. Guo, H. Zhang, J. Wang, Z. Huang, F. Ma, R. Chen and H. Ni, ACS Appl. Energy Mater., 2025, 8, 4418–4427.
9. X. Jiang, W. Chu, X. Ren, F. Ma, R. Chen, S. Ning, Y. Zhang, C. Zeng, L. Shi and L. Ren, ACS Nano, 2025, 19(18), 17336–17346.
10. X. Jiang, X. Ren, R. Chen, F. Ma, W. He, T. Zhang, Y. Wen, L. Shi, L. Ren and H. Liu, J. Electroanal. Chem., 2023, 949, 117860.
11. Z. Zheng, C. Sun, M. Ma, T. Wang, L. Wang, J. Fu, Q. Zhou, Y. Liang and W. Wang, J. Colloid Interface Sci., 2025, 683, 585–599.
12. H. Gao, L. Zhu, G. Zhang, X. Xu and F. Yang, J. Colloid Interface Sci., 2025, 688, 677–687.
13. M. I. Din, R. Khalid, J. Najeeb and Z. Hussain, J. Cleaner Prod., 2021, 298, 126567.
14. T. H. Jeon, D. Monllor-Satoca, G. H. Moon, W. Kim, H. I. Kim, D. W. Bahnemann, H. Park and W. Choi, Nat. Commun., 2020, 11, 967.
15. V. B. Shinde, Y. Maeda, T. Katayama, A. Furube, T. Yano and P. Koinkar, Mod. Phys. Lett. B, 2023, 37, 2340005.
16. T. Paik, M. Cargnello, T. R. Gordon, S. Zhang, H. Yun, J. D. Lee, H. Y. Woo, S. J. Oh, C. R. Kagan and P. Fornasiero, ACS Energy Lett., 2018, 3, 1904–1910.
17. M. Ling, C. S. Blackman, R. G. Palgrave, C. Sotelo-Vazquez, A. Kafizas and I. P. Parkin, Adv. Mater. Interfaces, 2017, 4, 1700064.
18. X. Ren, Z. Liu, T. Zhang, X. Jiang, Q. Fang, Y. Li, F. Ma, R. Chen, H. Zhang and H. Ni, J. Mater. Sci., 2024, 59, 10724–10743.
19. K. Yan, S. Li, J. Liu, X. Liu and J. Zhang, J. Colloid Interface Sci., 2025, 681, 356–364.
20. K. G. Motora, C.-M. Wu, T. F. Chala, M.-H. Chou, C.-F. J. Kuo and P. Koinkar, Appl. Surf. Sci., 2020, 512, 145618.
21. L. Kong, X. Guo, J. Xu, Z. Mo and L. Li, J. Photochem. Photobiol., A, 2020, 401, 112760.
22. P. Wu, Z. Liu, D. Chen, M. Zhou and J. Wei, Appl. Surf. Sci., 2018, 440, 1101–1106.
23. H. Sun, W. Hua, Y. Li and J.-G. Wang, ACS Sustainable Chem. Eng., 2020, 8, 12637–12645.
24. S. Li, J. Liu, X. Liu, K. Yan and J. Zhang, Electrochim. Acta, 2024, 484, 144058.
25. Z. Li, S. Xu, Y. Shi, X. Zou, H. Wu and S. Lin, Chem. Eng. J., 2021, 414, 128814.
26. J. Jin, J. Yu, D. Guo, C. Cui and W. Ho, Small, 2015, 11, 5262–5271.
27. L. Cheng, T. Jiang and J. Zhang, Sci. Total Environ., 2021, 776, 145840.
28. R.-Z. Wang, Z. Lin, Y.-Q. Wang, K.-N. Zhang, G.-H. Zhang, J. Zhang, S. S. Mao and S.-H. Shen, Rare Met., 2024, 43, 3771–3783.
29. X. Li, X. Song, C. Ma, Y. Cheng, D. Shen, S. Zhang, W. Liu, P. Huo and H. Wang, ACS Appl. Nano Mater., 2020, 3, 1298–1306.
30. Q. Xu, L. Zhang, B. Cheng, J. Fan and J. Yu, Chem, 2020, 6, 1543–1559.
31. D. Xu, B. Cheng, S. Cao and J. Yu, Appl. Catal., B, 2015, 164, 380–388.
32. H. Zhong, X. Ji, C. Yang, C. Tang, X. Hu, J. Cai, Q. Yin, J. Wang, X. Hu and X. Tan, ACS Catal., 2024, 14, 6272–6291.
33. B. Yang, S. Qin, D. J. Xue, C. Chen, Y. S. He, D. Niu, H. Huang and J. Tang, Progr. Photovolt.: Res. Appl., 2017, 25, 113–122.
34. Y. Huo, X. Yang, J. Zhu and H. Li, Appl. Catal., B, 2011, 106, 69–75.
35. B. Zheng, W. Zheng, Y. Jiang, S. Chen, D. Li, C. Ma, X. Wang, W. Huang, X. Zhang and H. Liu, J. Am. Chem. Soc., 2019, 141, 11754–11758.
36. T. Zhang, X. Ren, W. Cao, H. Zou, X. Jiang, F. Ma, R. Chen, H. Qiao, Y. Zhang and H. Liu, J. Colloid Interface Sci., 2025, 679, 569–577.
37. X. Wang, S. Li, Z. Yuan, Y. Sun, Z. Tang, X. Gao, H. Zhang, J. Li, S. Wang and D. Yang, Angew. Chem., 2023, 135, e202303794.
38. K.-S. Lin, A. Furube, T. Katayama, P. Koinkar and C. M. Wu, Mod. Phys. Lett. B, 2025, 39, 2441007.
39. M. Barakat, R. Kumar, T. Almeelbi, B. A. Al-Mur and J. O. Eniola, J. Cleaner Prod., 2022, 330, 129942.
40. Z. Liang, X. Zhang, J. Yang, Y. Cheng, H. Hou, S. Hussain, J. Liu, G. Qiao and G. Liu, J. Hazard. Mater., 2023, 443, 130316.
41. D.-B. Seo, S. Yoo, V. Dongquoc, T. N. Trung and E.-T. Kim, J. Alloys Compd., 2021, 888, 161587.
42. H. Qiao, H. Liu, Z. Huang, Q. Ma, S. Luo, J. Li, Y. Liu, J. Zhong and X. Qi, Adv. Energy Mater., 2020, 10, 2002424.
43. O. Hurtado-Aular, R. Añez and A. Sierraalta, Surf. Sci., 2021, 714, 121907.
44. G. Liu, P. R. Narangari, Q. T. Trinh, W. Tu, M. Kraft, H. H. Tan, C. Jagadish, T. S. Choksi, J. W. Ager and S. Karuturi, ACS Catal., 2021, 11, 11416–11428.
45. J. Mai, Y. Fang, J. Liu, J. Zhang, X. Cai and Y. Zheng, Mol. Catal., 2019, 473, 110389.
46. J. Low, C. Jiang, B. Cheng, S. Wageh, A. A. Al-Ghamdi and J. Yu, Small Methods, 2017, 1, 1700080.
47. J. Li, H. Yuan, W. Zhang, B. Jin, Q. Feng, J. Huang and Z. Jiao, Carbon Energy, 2022, 4, 294–331.
48. L. Wang, C. Bie and J. Yu, Trends Chem., 2022, 4, 973–983.
49. Q. Xu, L. Zhang, J. Yu, S. Wageh, A. A. Al-Ghamdi and M. Jaroniec, Mater. Today, 2018, 21, 1042–1063.
50. S. Dong, J. Dai, Y. Yang, A. Zada and K. Qi, Molecules, 2024, 29, 4561.
51. J. Zuo, H. Guo, S. Chen, Y. Pei and C. Liu, Catal. Sci. Technol., 2023, 13, 3963–3973.
52. A. O. Onjwaya, M. L. Malati, J. C. Ngila and L. N. Dlamini, Dalton Trans., 2024, 53, 7694–7710.
53. N. Zhang, H. Li, B. Yao, S. Liu, J. Ren, Y. Wang, Z. Fang, R. Wu and S. Wei, Dalton Trans., 2023, 52, 6284–6289.
54. M. Humayun, H. Ullah, J. Cao, W. Pi, Y. Yuan, S. Ali, A. A. Tahir, P. Yue, A. Khan and Z. Zheng, Nano-Micro Lett., 2020, 12, 7.
55. N. Paul, T. Katayama, A. Furube and P. Koinkar, J. Mater. Sci.: Mater. Electron., 2025, 36, 982.
56. N. Paul, A. Sawate, S. Sugano, T. Katayama, A. Furube and P. Koinkar, Surf. Rev. Lett., 2025, 32, 1–11.
57. M. Desseigne, V. Chevallier, V. Madigou, M.-V. Coulet, O. Heintz, H. Ait Ahsaine and M. Arab, Catalysts, 2023, 13, 1333.
58. A. C. Güler, M. Masar, M. Urbánek, M. Machovský, M. M. Elnagar, R. Beranek and I. Kuritka, ACS Appl. Energy Mater., 2025, 8, 4090–4102.
59. M. Tayebi, Z. Masoumi and B.-K. Lee, Ultrason. Sonochem., 2021, 70, 105339.
60. Q. Ma, R. Song, F. Ren, H. Wang, W. Gao, Z. Li and C. Li, Appl. Catal., B, 2022, 309, 121292.
61. T. Kim, Y. Im, S. Kim, M. S. Goh, H. Moon, S.-M. Park and M. Kang, J. Alloys Compd., 2023, 932, 167629.
62. Q. Ma, Z. Ning, M. Fang, G. Zhang, H. Guo, J. Zhou and T. Wang, Chem. Eng. J., 2024, 498, 155551.
63. S. Zhang, S. Chen, D. Liu, J. Zhang and T. Peng, Appl. Surf. Sci., 2020, 529, 147013.
64. X. Wang, G. Hai, B. Li, Q. Luan, W. Dong and G. Wang, Chem. Eng. J., 2021, 426, 130822.
65. D. Zhou, D. Li and Z. Chen, Front. Chem., 2024, 12, 1359895.
66. D. Liu, S. Chen, R. Li and T. Peng, Acta Phys.-Chim. Sin., 2021, 37(6), 2010017–2010048.
67. Z. Jiang, W. Zhang, L. Jin, X. Yang, F. Xu, J. Zhu and W. Huang, J. Phys. Chem. C, 2007, 111, 12434–12439.
68. M. Karbalaei Akbari, N. S. Lopa and S. Zhuiykov, Coatings, 2024, 14, 880.

---