Enhanced visible-light-driven photocatalytic performances using Bi₂WO₆/MS (M = Cd, Zn) heterostructures: facile synthesis and photocatalytic mechanisms

Abstract

The formation of type II heterostructures is proved to be an effective method to improve the photocatalytic performance of semiconductors. In this report, a surface functionalization method using 3-mercaptopropionic acid (MPA) as the surface functionalizing agent was adopted for the fabrication of Bi₂WO₆/MS (M = Cd and Zn) heterostructures. The composition and microstructures of the as-prepared heterostructures were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-resolution TEM (HRTEM) and X-ray photoelectron spectroscopy (XPS). The as-prepared Bi₂WO₆/CdS and Bi₂WO₆/ZnS heterostructures exhibit enhanced photocatalytic activities for the degradation of Rhodamine B (Rh B) as compared to pure Bi₂WO₆, CdS, ZnS nanoparticles and Degussa P25. The enhanced photocatalytic activities of the as-prepared heterostructures may relate to the effective electron–hole separation during the photocatalytic process. A series of scavengers (benzoquinone for O₂˙⁻, ammonium oxalate for h⁺, AgNO₃ for e⁻, and t-BuOH for ˙OH) were employed to investigate the possible mechanism of the photocatalytic process. The results clearly indicate that the photogenerated holes are the main active species during the photocatalytic process.

---

1. Introduction

Photocatalysts for the degradation of organic pollutants using solar light energy have attracted increasing interest in the past few years because of the growing environmental concerns and energy demand.^1–5 Compared with the traditional methods, semiconductor-based photocatalysis offers a greener route for the elimination of a variety of contaminants, especially for azo dyes.⁶ For this purpose, a great number of visible-light-driven photocatalysts have been successfully synthesized, which have been proved to be effective for the elimination of organic pollutants.^7–12 Among these photocatalysts, Bi₂WO₆ has attracted increasing attention because of its excellent visible-light-driven photocatalytic activity. The unique layered structure of Bi₂WO₆ favors the charge transfer process and retards the recombination of holes and electrons during the photocatalytic process,¹³ making it an ideal candidate for the photo-degradation of organic pollutants. As a result, a variety of Bi₂WO₆ nano-structures with different morphologies including nest,¹⁴ flake-ball,¹⁵ disk,¹⁶ cage,¹⁷ flowers^18–20 have been successfully prepared. Although these Bi₂WO₆ nanostructures have been found to display strong visible light response, the photocatalytic activity is expected to be further improved to satisfy the large-scale application under visible light irradiation. According to the previous reports, two formidable challenges still remain.^21,22 One centers on the difficult migration and high recombination of photo-generated electron–hole pairs, which will greatly reduce the photocatalytic activity of Bi₂WO₆. A second entails its wide band gap, which severely precludes the employment of visible light with wavelength longer than ∼450 nm.

It is widely believed that the coupling of two semiconductors is an effective way to enhance the photocatalytic performance, which would lead to a more efficient photoelectron–hole separation or expansion of the spectral absorption range.^23,24 The electrons excited by the visible light would transfer from conduction band of semiconductor with low band gap to the conduction band of Bi₂WO₆, which will facilitate the charge separation and improve the visible-light photocatalytic activities of the heterostructures dramatically. As a result, a variety of Bi₂WO₆ based heterostructures have been successfully prepared, including Bi₂WO₆/WO₃,²⁵ Bi₂WO₆/Bi₂S₃,²⁶ Bi₂WO₆/BiOBr,²⁷ Bi₂WO₆/TiO₂,²⁸ Bi₂WO₆/Bi₂O₃ ²⁹ and so on. All of these works clearly indicate that the coupling of another semiconductor is indeed an effective method to improve the photocatalytic activities of Bi₂WO₆. For example, Huang and co-workers have succeeded in the preparation of layered Bi₂WO₆/BiIO₄ heterojunctions by a one-step hydrothermal preparation strategy. The degradation rate for Rh B was almost 34.4 and 4.1 times higher than the corresponding values of BiIO₄ and Bi₂WO₆, respectively.³⁰ Fu and co-workers have also reported the preparation of Bi₂WO₆/Ag₃PO₄ hierarchical heterostructures, whose degradation rate for phenol under visible light irradiation was about 1.7 and 1.1 times higher than pure Bi₂WO₆ and Ag₃PO₄, respectively.³¹ However, the synthesis of the Bi₂WO₆ based heterostructure is still a big challenge for the chemists, especially by a simple and economic method.

Recently, the coupling of metal sulfides with other semiconductors has been found to be an effective route for the preparation of photocatalysts with high photocatalytic activities.^32–37 Thus, it is speculated that the sensitization property of metal sulfides will be helpful in overcoming the drawbacks of Bi₂WO₆ and lead to the formation of highly efficient photocatalysts. In this study, two Bi₂WO₆ based heterostructure (Bi₂WO₆/CdS and Bi₂WO₆/ZnS) have been successfully prepared using a simple surface functionalization route. Evaluated by the degradation of Rhodamine B (Rh B), the as-formed Bi₂WO₆/MS (M = Cd and Zn) heterostructures exhibit enhanced photocatalytic activities for the degradation of Rh B under visible light irradiation. Based on experimental results and the band gap positions, a possible photodegradation mechanisms of these heterostructures were also systematically investigated.

2. Experimental section

2.1 Preparations of flower-like Bi₂WO₆

All the reagents were of analytical grades and used without further purification. Flower-like Bi₂WO₆ structures was prepared by a hydrothermal process as we have reported previously.³⁸ In a typical process, 2 mmol Bi(NO₃)₃·5H₂O and 1 mmol Na₂WO₄·2H₂O were dissolved separately in 22.5 mL de-ionized water. Then, the two solutions were mixed under vigorous stirring. The resulting slurries were stirred for 30 min and then transferred into a 50 mL Teflon-lined autoclave. The autoclaves were maintained at 200 °C for 12 h and then cooled to room temperature naturally. The precipitates were collected by centrifugation, washed four times with deionized water and ethanol, and then dried at 60 °C for 6 h in vacuum.

2.2 Preparations of Bi₂WO₆/CdS and Bi₂WO₆/ZnS heterostructures

For the growth of CdS and ZnS nanoparticles onto flower-like Bi₂WO₆ structures, a surface functionalization route using 3-mercaptopropionic acid (MPA) as the functionalizing agent was adopted.³⁹ The synthetic process of Bi₂WO₆/CdS heterostructures is illustrated in Scheme 1. In a typical process, 20 μL of MPA was added to the suspension containing 1.0 g Bi₂WO₆. The suspension was stirred for 4 h at room temperature to complete the surface functionalization of the Bi₂WO₆ surfaces. Afterward, 0.1 g Cd(CH₃COO)₂·4H₂O was added to the above suspension under constant magnetic stirring. After stirring for 2 h, 0.09 g of Na₂S·9H₂O was added slowly to the system and stirred constantly for 1 h at room temperature. The solution gradually turned to be faded yellow, which suggest the formation of Bi₂WO₆/CdS heterostructures, and the nominal content of CdS with respect to Bi₂WO₆ is about 23.4% (molar ratio). Subsequently, the resulting product was separated by centrifugation and washed repeatedly with water and absolute alcohol and then dried in vacuum at 60 °C for 6 h. As for the preparation of Bi₂WO₆/ZnS heterostructure, Cd(CH₃COO)₂·4H₂O added in the second step was replaced by Zn(NO₃)₃·6H₂O, keeping the other reactions conditions unchanged. By calculation, the nominal content of ZnS with respect to Bi₂WO₆ is about 23.5% (molar ratio).

Scheme 1 Schematic illustration for the synthetic process of Bi₂WO₆/CdS heterostructures.

2.3 Preparations of CdS and ZnS nanoparticles

For comparison purpose, CdS and ZnS nanoparticles were prepared by a simple precipitation method. Take the synthesis of CdS nanoparticles for example, 1.0 g of Cd(CH₃COO)₂·4H₂O was firstly dissolved in 40 mL of deionized water to form a transparent solution. In the next step, 0.9 g Na₂S·9H₂O was added into the above solution and stirred for 1 h at room temperature. Finally, the product was separated by centrifugation, washed repeatedly with deionizer water and ethanol and then dried at 60 °C for 6 h in vacuum. ZnS nanoparticles were synthesized by a similar process during which Zn(NO₃)₃·6H₂O was used as the starting material.

2.4 Sample characterizations

The phase and composition of the samples were characterized by X-ray powder diffraction (XRD) on a Philips X'Pert Pro Super diffractometer with Cu Kα radiation (λ = 1.5416 Å). The XRD pattern was recorded in the 2θ range of 10–70° with scan rate of 0.05° s⁻¹. The S-4800 (Hitachi) field emission scanning electron microscope (FESEM) equipped with a GENESIS4000 energy dispersive X-ray spectroscope was applied to observe the size and morphology of the product. The transmission electron microscope (TEM) image was taken on a Hitachi H-7650 transmission microscope, using an accelerating voltage of 120 kV. High-resolution transmission electron microscopy (HRTEM) images were obtained on a JEOL-2010 TEM at an acceleration voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) was recorded on Thermo ESCALAB 250 system with a monochromatic Al Kα source with 1486.6 eV of energy, 15 kV of voltage, and 150 W of power. The Brunauer–Emmett–Teller (BET) tests were carried out via a Quantachrome autosorb IQ-C nitrogen adsorption apparatus. All the as-prepared samples were degassed at 150 °C for 4 h prior to nitrogen adsorption measurements. UV-vis diffuse reflectance spectra (DRS) were recorded on a UV-vis spectrometer (Shimadzu UV-2550) by using BaSO₄ as a reference and were converted from reflection to absorbance by the Kubelka–Munk method. The photoluminescence (PL) spectra of the photocatalysts were detected using an F-7000 FL Spectrophotometer.

2.5 Photocatalytic experiments

In a general process, the photocatalytic activities of the photocatalysts were evaluated by the degradation of rhodamine B (Rh B). In a typical process, 80 mg of the as-synthesized photocatalysts were dispersed in 80 mL 1.0 × 10⁻⁵ mol L⁻¹ rhodamine B (Rh B) solution and then magnetically stirred in the dark for 30 min to reach the adsorption–desorption equilibrium. The solution was then exposed to a 500 W Xe lamp with a cutoff filter (λ > 400 nm). And the working distance from the Xe lamp to the beaker is kept to be 20 cm. The optical power was measured to be 25.8 mW cm⁻² during the whole photocatalytic process. To avoid the increase of the temperature during the irradiation process, the beaker containing the dispersion of Rh B and photocatalyst was cooled by circulating water during the whole photocatalysis process. The temperature of the solution was maintained to be 20 °C during the whole process to prevent the thermal catalytic effect. At given irradiation time, adequate aliquots (6 mL) of the suspension were extracted and centrifuged at 9000 rpm to remove the residual photocatalyst for analysis. The concentration of Rh B in the solution was calculated by the intensity of UV-Vis absorption (Agilent Cary 5000E UV-visible spectrophotometer).

2.6 Active species trapping experiments

To explore the possible photocatalytic mechanism, active species trapping experiments was carried out using various kinds of scavengers such as benzoquinone (BQ) (a quencher of O₂˙⁻), ammonium oxalate (AQ) (a quencher of h⁺), AgNO₃ (a quencher of e⁻), and t-BuOH (a quencher of ˙OH). And the amount of scavenger was 10 mM except for BQ, which was 1 mM. The whole process is carried out in the similar way as the photocatalytic experiments.

2.7 Photocurrent measurements

The ITO glass was cut into 3 cm × 0.7 cm slices and washed with water for two times before being bathed in 1 M NaOH solution for 10 min, and then washed with water several times and dried before use. For the preparation of the photoelectrode, 10 mg of photocatalyst was dispersed in 1 mL of absolute ethanol by ultrasonication firstly. In the next step, 10 μL of the resulting dispersion (10 mg mL⁻¹) was drop-casted onto a piece of ITO slice with a fixed area of 0.7 cm² and dried in air at room temperature to form Bi₂WO₆/MS modified ITO electrode. The photocurrent were measured by using an electrochemical analyzer (CHI660B, Chen Hua Instruments, Shanghai, China) with a standard three-electrode configuration, in which a Pt wire was employed as the counter electrode, a saturated Ag/AgCl electrode as the reference electrode and indium tin oxide (ITO) as the working electrode. The 500 W Xe lamp was used as the photo source, and the measured optical power on the ITO electrode was determined to be 70 mW cm⁻². The electrolyte solution of photocurrent was phosphate buffered saline (0.1 mol L⁻¹) and the pH value was adjusted to 7.4.

3. Results and discussion

3.1 Sample characterizations

The addition of MPA in the synthetic process is of vital importance for the formation of Bi₂WO₆ based heterostructures. Take the formation of Bi₂WO₆/CdS as an example, no CdS will be observed on the nanoplates of Bi₂WO₆ without MPA (ESI, Fig. S1a†). To further investigate the effect of MPA, FTIR was employed. Pure MPA was reported to possess several typical peaks centering at 2667, 2573, 1711 and 1411 cm⁻¹, which can be assigned to the O–H(ν_O–H), S–H(ν_S–H), C=O(ν_C=O) and C–O(ν_C–O) stretching bands, respectively.^40,41 As for the sample obtained by adding MPA into the dispersion of Bi₂WO₆, the peaks corresponding to ν_S–H disappears (ESI, Fig. S1b,† black line). Because the pK_a of the thiol moiety is quite high (10.3), the disappearance of S–H peak can't be assigned to the deprotonation of sulfhydryl group in MPA.⁴² Meanwhile, the solution pH during the synthetic process is about 6, which also suggest that the MPA added during the synthetic process will not be the sulfur source for the formation of the heterostructures. The disappearance of this peak indicate the coordination interaction between Bi³⁺ and MPA, in which Bi³⁺ is bounded to the sulfhydryl group. When Cd²⁺ is added in to the solution during the synthetic process, the intermediate product was also separated and investigated with FTIR (ESI, Fig. S1b,† red line). The peaks corresponding to ν_C=O shift from 1637 to 1560 cm⁻¹, which clearly indicate the coordination interaction between carboxyl group of MPA and Cd²⁺. This kind of coordination interaction favors for the adsorption of Cd²⁺ onto the surfaces of Bi₂WO₆ nanoplates, which is of vital importance for the fabrication of Bi₂WO₆/CdS heterostructures. In the next step, when Na₂S was added, CdS nanoparticles finally formed on the surfaces of Bi₂WO₆ nanoplates. Based on the above experimental results and analysis, the role of MPA molecule can be considered to be a “linker”, which connect the host material and modifiers together, leading to the formation of the Bi₂WO₆/MS (M = Cd and Zn) heterostructures.

Bi₂WO₆ with flower-like structure were prepared by a hydrothermal process at 200 °C for 12 hours. The phase purities of the sample is examined by X-ray diffraction (XRD), and the corresponding XRD pattern is displayed as Fig. 1a. The diffraction peaks on Fig. 1a can be indexed to be the orthorhombic phased Bi₂WO₆ (JCPDS Card, no. 73-1126),⁴³ indicating the formation of Bi₂WO₆. The XRD patterns of the samples modified with CdS (BW-Cd) and ZnS (BW-Zn) are displayed as Fig. 1b and c. Compared to Fig. 1a, no obvious difference can be observed on these two XRD patterns. Although no peak corresponding to CdS or ZnS is detected on the XRD pattern, the existence of CdS or ZnS can't be precluded. This phenomenon may be related to the small amount of CdS or ZnS and their high dispersion in the samples.⁴⁴ Meanwhile, the synthetic processes were carried out under room temperature, which may lead to the poor crystallinities of MS (M = Cd or Zn). The poor crystallinities of MS would make them difficult to be detected by the XRD examination. To verify the existence of CdS and ZnS in samples BW-Cd and BW-Zn, the chemical compositions of the as-prepared composite photocatalysts were further investigated using X-ray fluorescence (XRF). The results of XRF clearly indicate the existence of CdS and ZnS in sample BW-Cd and BW-Zn. For sample BW-Cd, the atomic ratio between Bi, W, Cd and S is determined to be 2 : 1 : 0.18 : 0.18. While for sample BW-Zn, the corresponding value is determined to be 2 : 1 : 0.14 : 0.14.

Fig. 1 XRD patterns of (a) Bi₂WO₆, (b) Bi₂WO₆/CdS and (c) Bi₂WO₆/ZnS photocatalysts.

The chemical state of the constituent elements for the as-obtained two samples (BW-Cd and BW-Zn) were further investigated by XPS analysis, which is based on the limited escape depth of electrons with relatively low kinetic energy and reveals information on the topmost few angstroms of the surface.⁴⁵ The peak positions in all the XPS spectra were calibrated with C 1s at 284.60 eV. Fig. 2a is the overall XPS spectrum of sample BW-Cd, which clearly indicates the formation of Bi₂WO₆/CdS heterostructure. Fig. 2b and c are the high resolution XPS spectra of S 2s and Cd 3d, respectively. The binding energy centering at 226.4 eV can be assigned to S 2s. Meanwhile, Cd 3d exhibits characteristic peaks at 411.95 and 405.2 eV, corresponding to Cd 3d_3/2 and Cd 3d_5/2. Two peaks centering at 158.85 eV and 164.15 eV can be ascribed to Bi 4f_7/2 and Bi 4f_5/2 ⁴⁶ (ESI, Fig. S2a†). And the peaks centering at 37.2 and 35.05 eV correspond to W 4f_5/2 and W 4f_7/2,^47,13a respectively (Fig. S2b†). The formation of Bi₂WO₆/ZnS heterostructure can be also verified by the XPS investigation of sample BW-Zn, which clearly indicate the existence of Bi, W, Zn and S in sample BW-Zn (Fig. 2d). The peaks locating at 225.8 eV (shown in Fig. 2e) and 1021.75 eV (shown in Fig. 2f) belong to S 2s and Zn 2p peaks,⁴⁸ respectively. Peaks centering at 37.45 and 35.3 eV, as shown in Fig. S2c,† correspond to W 4f_5/2 and W 4f_7/2. The binding energies for Bi 4f_7/2 and Bi 4f_5/2 are 159 and 164.3 eV, which can be observed in Fig. S2d.† For sample BW-Cd, the atomic ratio of Bi : W : Cd : S is determined to be 2 : 1 : 0.35 : 0.35. While for sample BW-Zn, the corresponding atom ratios of Bi : W : Zn : S is determined to be 2 : 1 : 0.22 : 0.22. It should be noted that the content of CdS or ZnS determined by the XPS are much larger than the corresponding values determined by the XRF investigation, which suggest that MS (M = Cd, Zn) nanoparticles is mainly dispersed on the surfaces of flower-like Bi₂WO₆ structures. Meanwhile, the content of CdS determined by XPS is much larger than the corresponding value of ZnS. This phenomenon can be attributed to the different measure modes between XRF and XPS. Because XPS is a micro-area analysis technique, the relatively high value for CdS content may relate to the non-uniform distribution of CdS nanoparticles on the naoplates of Bi₂WO₆.

Fig. 2 XPS analysis of (a) Bi₂WO₆/CdS sample, (b) S 2s, (c) Cd 3d, (d) Bi₂WO₆/ZnS sample, (e) S 2s, (f) Zn 2p.

The morphologies and microstructures of samples BW-Cd and BW-Zn were further investigated by SEM, TEM and HRTEM, and the corresponding results are shown in Fig. 3. Fig. 3a is the SEM image of sample BW-Cd, which clearly indicate that the sample is composed of flower-like microstructures with diameter of about 2.0 μm, which agrees well with the microstructure of bare Bi₂WO₆ in our previous report.³⁸ Fig. 3b is the high-magnification TEM image of the Bi₂WO₆ nanoplates, suggesting the existence of small nanoparticle on the surfaces of these nanoplates. The sizes of these nanoparticles is about 5–10 nm. These small nanoparticles can be designated to be CdS according to the EDS analysis (ESI, Fig. S3a†). To gain further insight into the detailed structure of sample BW-Cd, HRTEM was employed. Fig. 3c is the HRTEM image of sample BW-Cd, which clearly indicate the formation of the Bi₂WO₆/CdS heterostructure. The typical lattice spacing, being determined to be 0.315 nm, is consistent with the (1 1 3) planes of Bi₂WO₆. The lattice spacing taken on the tiny nanocrystals is about 0.242 nm, which is consistent with the (1 0 2) planes of CdS (JCPDS no. 89-2944). Fig. 3d–f shows the SEM, TEM, and HRTEM images of Bi₂WO₆/ZnS heterostructures, which also suggest that the existence of ZnS nanoparticles on the surfaces of Bi₂WO₆ nanoplates. Fig. 3f is the HRTEM image of sample BW-Zn, which clearly suggest the formation of Bi₂WO₆/ZnS heterostructures. The typical lattice spacing, being determined to be 0.415 nm, is consistent with the (0 0 4) planes of Bi₂WO₆. The lattice spacing taken on the tiny nanocrystals is about 0.546 nm, which is consistent with the (0 0 1) planes of ZnS (JCPDS no. 80-0020).

Fig. 3 (a) SEM, (b) TEM and (c) HRTEM images of sample BW-Cd; (d) SEM, (e) TEM and (f) HRTEM images of sample BW-Zn.

The BET specific surface areas of Bi₂WO₆, BW-Cd and BW-Zn were determined by nitrogen adsorption BET method. And the measurement indicate that the specific surface area of pure Bi₂WO₆, BW-Cd and BW-Zn were about 13.6, 13.9 and 13.7 m² g⁻¹, respectively. No obvious change in BET surface area (as displayed in ESI Fig. S4†) has been observed among these three samples, which uncovered that the surface area may not be the influencing factor for the different photocatalytic performances of BW-Cd and BW-Zn.

The optical properties of the as-prepared photocatalysts are investigated using the UV-vis diffuse reflectance spectrum (DRS) (Fig. 4a and c). Bare Bi₂WO₆ exhibit fundamental absorbance edge at 450 nm because of the intrinsic band-gap transition,⁴⁹ and the absorbance edge of bare CdS and ZnS is at ∼540 nm and ∼338 nm, respectively. The adsorption edge of the sample BW-Cd is at 477 nm, which lies between the absorbance edges between Bi₂WO₆ and CdS. Compared with bare Bi₂WO₆, the absorbance edge of sample BW-Cd has an obvious red shift because of the introduction of CdS onto the Bi₂WO₆ nanoplates. As a result, the colors of samples BW-Cd changes from white to yellow (inset of Fig. 4a), resulting from the interfacial interactions between Bi₂WO₆ and CdS. As for sample BW-Zn, the absorbance edge is determined to be 575 nm, which show obvious red shift as compared to both bare Bi₂WO₆ and ZnS. Liu and co-workers have investigate the optical property of ZnO/ZnAl₂O₄ heterostrucures,⁵⁰ whose absorbance edge also exhibit obvious red shift as compared to both ZnO and ZnAl₂O₄. This kind of change is attributed to the middle band gap between the CB bottom of ZnO and the VB top of ZnAl₂O₄. In our synthesis, the red shift of absorbance edge for sample BW-Zn could also be attributed to the middle band gap formed between the CB bottom of Bi₂WO₆ and the VB top of ZnS, which is mainly caused by the interfacial interaction between ZnS and Bi₂WO₆. This phenomena have also been reported by several groups for a variety of semiconductor heterostructures such as Bi₂MoO₆/TiO₂,⁵¹ and TiO₂/MMT.⁵²

Fig. 4 (a) UV-vis absorption spectra of Bi₂WO₆, CdS and BW-Cd, and (b) Kubelka–Munk plot for band gap calculation of Bi₂WO₆, CdS, and BW-Cd. The inset in (a) shows the snapshots of Bi₂WO₆ and BW-Cd samples; (c) UV-vis absorption spectra of Bi₂WO₆, ZnS and BW-Zn, and (d) Kubelka–Munk plot for band gap calculation of Bi₂WO₆, ZnS, and BW-Zn. The inset in (c) shows the snapshots of Bi₂WO₆ and BW-Zn samples.

The optical band gaps of the as-form heterostructures can be calculated using the Tauc plot, which is described as follows:⁵³

where h is the Planck's constant, ν is the frequency of vibration, α is the absorption coefficient, E_g is the band gap of the semiconductor. Among them, n is determined from the type of optical transition of a semiconductor, and the value of n for the direct semiconductor is 1/2, while for indirect transition, n = 2.³⁹ By calculation, n is determined to be 1/2 for the as-formed heterostructures. And the band gaps of Bi₂WO₆, CdS, ZnS, BW-Cd and BW-Zn were estimated to be 3.08, 2.33, 3.78, 2.74, and 2.86 eV, respectively (Fig. 4b and d).

Based on the result of DRS, the band edge positions of the as-prepared semiconductors can be theoretically predicted using the electronegativity concept.^54,55 The CB and VB potentials of the semiconductor at the point of zero charge can be calculated by the following equation:

E_CB = X + E₀ − 0.5E_g

and

E_VB = E_CB + E_g

where E_g represents the band gap energy of the semiconductor, E_CB is the conduction band potential, E_VB is the valence band potential, E₀ is scale factor relating to the reference electrode redox level to the absolute vacuum scale (E₀ = −4.5 eV for normal hydrogen electrode), and X is the electronegativity of the semiconductor, which is defined as the geometric average of the absolute electronegativity of the constituent atoms. The calculated CB and VB potential of Bi₂WO₆, CdS, ZnS, BW-Cd, BW-Zn are shown in Table 1. The band gap position of bare Bi₂WO₆ is measured to be 3.08 eV, and the E_CB and E_VB are calculated to be 0.36 eV and 3.44 eV, respectively. For CdS and ZnS nanoparticles, both the values of E_CB and E_VB are lower than the corresponding values of Bi₂WO₆, corresponding to the type-II band gap alignment. This kind of band gap alignment will favor for the separation of electrons and holes during the photocatalytic process and lead to the enhanced photocatalytic activities for the heterostructures.

Table 1 Calculated values of band gap, conduction band (CB) and valence band (VB) of sample Bi₂WO₆, CdS, ZnS, BW-Cd and BW-Zn

Sample	Band gap (eV)	Conduction band (eV)	Valence band (eV)
Bi2WO6	3.08	0.36	3.44
CdS	2.33	−0.56	1.76
ZnS	3.78	−1.12	2.66
Bi2WO6/CdS	2.74	0.36 (Bi2WO6)	3.44 (Bi2WO6)
		−0.56 (CdS)	1.76 (CdS)
Bi2WO6/ZnS	2.86	0.36 (Bi2WO6)	3.44 (Bi2WO6)
		−1.12 (ZnS)	2.66 (ZnS)

---

3.2 Visible-light-driven photocatalytic activities

Dyes are pervasive commercial chemicals which have arisen unique environmental problems.^56,57 So it is imperative to remove them before draining them to the environment. Compared to technologies involving physical or biological treatments,^58,59 semiconductor-based photocatalysis can achieve significant organic dye pollutant degradation, which is considered as an alternative treatment strategy to solve the environmental problems successfully. To investigate the photocatalytic activities of the as-prepared heterostructures, rhodamine B (Rh B) is chosen as the target pollutant. For comparison purpose, the blank experiment (without photocatalysts), the photocatalytic activities of bare Bi₂WO₆, CdS nanoparticles, ZnS nanoparticles and Degussa P25 were also investigated. The photodegradation efficiencies of Rh B over BW-Cd, BW-Zn, bare Bi₂WO₆, CdS, ZnS, Degussa P25 and the blank experiment under visible light are shown in Fig. 5a. The blank experiment shows that Rh B molecules are very stable and the photolysis is negligible. In addition, both of the as-formed heterostructures (BW-Cd and BW-Zn) display enhanced photocatalytic activities as compared to bare Bi₂WO₆, CdS nanoparticles and ZnS nanoparticles. The photocatalytic activities among these samples follow the sequence: BW-Cd > BW-Zn > Bi₂WO₆ > P25 > CdS > ZnS. Among these samples, sample BW-Cd exhibits the highest photocatalytic activity toward the degradation of Rh B under visible light, and almost 98.7% of the Rh B can be degraded in 90 min (Fig. 5b). The photocatalytic activity of sample BW-Zn is a little poorer than sample BW-Cd and 87.1% of the Rh B can be degraded after irradiating for 90 minutes (Fig. 5b). Considering the fact that the two samples possess the similar BET surface areas, the difference in photocatalytic activities may relate to the different band gaps of the two heterostructures. ZnS nanoparticles exhibit the lowest photocatalytic activity, which may result from its large band gap. According to the band gap calculation above, the band gap of ZnS is 3.78 eV, indicating the poor absorbance of light in the visible region. Although the band gap of P25 is also large (3.2 eV), P25 shows a notably higher photoactivity than ZnS. Besides the band gap, the photocatalytic activity of semiconductor is also greatly influenced by a series of factors, such as BET surface areas and crystallinity.^60,61 Fig. S5† are the XRD patterns of the as-obtained ZnS nanoparticles and P25, which clearly indicate that the ZnS nanoparticles are poorly crystallized than P25. The difference in the crystallinity could also be reflected by the corresponding SAED pattern taken on these two samples. The poor crystallinity of ZnS nanoparticles could be verified by the ring-like ED pattern. Meanwhile, the BET surface areas of P25 and ZnS nanoparticles are determined to be 50.4 m² g⁻¹ and 21 m² g⁻¹, which shows that P25 possess a much larger surface area than ZnS nanoparticles. So the difference in photocatalytic activities between P25 and ZnS nanoparticles could be attributed to their different crystallinities and surface areas.

Fig. 5 (a) Photocatalytic performance at different time intervals; (b) degradation efficiency of blank, Bi₂WO₆, CdS and ZnS nanoparticles, Degussa P25, Bi₂WO₆/CdS and Bi₂WO₆/ZnS heterostructures for the degradation of Rh B solution under visible light irradiation.

As the band gap of P25 is large, it can be anticipated that the photocatalytic activity under visible light will be handicapped. To compare it with BW-MS (M = Cd or Zn), the Xe lamp was replaced by mercury lamp to offer useful photons for both P25 and BW-MS systems. The photocatalytic activities sequence among the three samples follow the sequence: P25 > BW-Zn > BW-Cd, which is in the opposite sequence as compared to the photocatalytic activities under visible light (Fig. S6†). Therefore, it can be concluded that the as-synthesized BW-MS (M = Cd or Zn) is a good visible-light responsive but not a good UV-light responsive photocatalyst.

To quantitatively understand the reaction kinetics of Rh B degradation in our experiments, Langmuir–Hinshelwood model was employed. This model is well-established for photocatalytic experiments when the concentration of the organic pollutant is in the millimolar range,⁶² which is expressed as follows:

where C₀ is the initial concentration of Rh B (mol L⁻¹), C is the concentrations of pollutant in solution at desired time, and k is the apparent first-order rate constant. The apparent rate constant k is calculated to be 0.04633 and 0.0230 min⁻¹ for samples BW-Cd and BW-Zn, respectively (Fig. 6). The photodegradation efficiency and apparent rate constant k of all the samples are summarized in Table 2. The photodegradation rate for sample BW-Cd and BW-Zn is 3.72 and 1.85 times than bare Bi₂WO₆, which can be attributed to the effective electron–hole separations at the interfaces of the two semiconductors.

Fig. 6 Plot of ln(C₀/C) as a function of visible irradiation time for photocatalysis of Rh B solution containing: blank, Bi₂WO₆, CdS and ZnS nanoparticles, Degussa P25, Bi₂WO₆/CdS and Bi₂WO₆/ZnS heterostructures under visible light irradiation.

Table 2 Photocatalytic efficiency and rate constant of Bi₂WO₆, CdS, ZnS, P25, BW-Cd and BW-Zn

Sample	Photocatalytic efficiency	Rate constant (min^−1)
Bi2WO6	69.5%	1.244 × 10^−2
CdS	39.1%	5.420 × 10^−3
ZnS	2.3%	2.472 × 10^−4
P25	41.9%	5.940 × 10^−3
Bi2WO6/ZnS	87.1%	2.301 × 10^−2
Bi2WO6/CdS	98.7%	4.633 × 10^−2
Blank (no photocatalyst)	1.7%	1.862 × 10^−4

---

The stability of the Bi₂WO₆/MS (M = Cd or Zn) heterostructures were also investigated by the degradation of Rh B under visible-light irradiation (shown in Fig. S7†). It should be noted that the photocatalytic activity of BW-Cd and BW-Zn catalyst exhibit obvious loss after five cycles. And the photodegradation efficiency decreased to 86.2% and 90% of their initial photocatalytic efficiency, respectively. According to the previous report, both CdS and ZnS are easily suffered from photocorrosion in aqueous media containing oxygen when exposed to visible light.^63–66 Thus, the decrease in the photocatalytic activities of the as-obtained heterostructures may result from the photocorrosion of metal sulfides during the photocatalytic process.

The improved photocatalytic activities of BW-Cd and BW-Zn is closely to the enhanced separation efficiency of photogenerated electrons and holes. To clarify this view, the photocurrent and photoluminescence spectra of bare Bi₂WO₆, BW-Cd and BW-Zn are investigated. Electrons in the conduction band and holes in the valence band were generated in the semiconductor under irradiation. As well known, the separation efficiency of electrons and holes play a crucial role during the photocatalytic reaction.⁶⁷ A higher photocurrent indicate the presence of photogenerated electrons and holes longer lives and hence the higher the photocatalytic activity can be achieved.⁶⁸ From the I–t curves (shown in Fig. 7), it can be seen that the BW-Cd and BW-Zn exhibit higher transient photocurrent density than bare Bi₂WO₆, which indicate an enhanced photo-induced electrons and holes separation efficiency. And the enhancement in the separation efficiency of electrons and holes for these two samples is higher than bare Bi₂WO₆, resulting from the interactions between Bi₂WO₆ and MS (M = Cd and Zn).

Fig. 7 Transient photocurrent response for the sample Bi₂WO₆, BW-Cd, and BW-Zn. The inset shows the amplifying photocurrent of bare Bi₂WO₆.

Photoluminescence (PL) spectrum is also a useful technique to survey the separation efficiency of the photogenerated charge carriers in a semiconductor, because the PL emission mainly results from the recombination of free carriers.⁶⁹ In general, lower PL intensity means lower recombination rate of the electron–hole pairs under light irradiation. The PL spectra (excited at 300 nm) of bare Bi₂WO₆, BW-Cd and BW-Zn at room temperature is shown in Fig. 8. It can be found that the PL emission intensities of samples BW-Cd and BW-Zn composite are dramatically weakened as compared to bare Bi₂WO₆. Sample BW-Cd exhibits the lowest emission peaks, which indicates that the formation of the heterostructures between Bi₂WO₆ and CdS would greatly facilitate the separation of the photo-generated charge carriers. As a result, the photocatalytic activity of the BW-Cd was greatly enhanced, and that is the reason why sample BW-Cd exhibits the highest photocatalytic activity.

Fig. 8 PL spectra of sample Bi₂WO₆, BW-Cd, and BW-Zn.

3.3 Mechanism of photocatalysis

As for the photodecomposition process of Rh B, two photocatalytic mechanisms are generally proposed. One is achieved by the degradation of the conjugated structure via a photocatalytic process,⁷⁰ during which a decrease of the absorption band without shifting of the maximum absorbance wavelength at 555 nm could be observed. The second pathway is really a photosensitized process, during which the degradation of Rh B occurs via the N-deethylation of Rh B in a stepwise process. To be specific, Rh B initially tetra-ethylated (λ_max = 555 nm) becomes tri, di-, mono-, and no-ethylated (λ_max is, respectively, 539, 522, 510, and 498 nm).^71–73 And this process is usually featured by the blue shifting of the absorption band. According to the absorption spectra variation of Rh B upon irradiation with the visible light in the presence of BW-Cd and BW-Zn (ESI, Fig. S8†), the occurrence of the photosensitization process could be confirmed. To figure out the main driving force in the photocatalytic process, a series of comparative experiments were carried out. In the first step, a 510 nm cutoff filter was employed during the whole photocatalytic process, which could only allow the transmission of light with wavelength longer than 510 nm. In this region, both BW-Cd and BW-Zn can't be effectively excited, and the photocatalytic process mainly relate to the photosensitized process. The photodegradation efficiency of BW-Cd sharply decreased when the 510 nm cutoff filter was employed, and only 5.4% of Rh B was degraded in 90 minutes. As for sample BW-Zn, the photodegradation efficiency can reach to 77% in 90 minutes, which is a little smaller than the corresponding value when the 400 cutoff filter was employed. This result clearly demonstrate that the photodecomposition of Rh B over BW-Cd is mainly driven by the first pathway (photocatalytic process). While for sample BW-Zn, the photodecomposition of Rh B is a photosensitization process indeed. So, although both photocatalytic process and photosensitization process exist during the photodecomposition of Rh B, the influence of these two pathways are totally different for this two samples. To further clarify our view, 4-chlorophenol (4-CP) was chosen as the target pollutant, because it has no light absorption in the visible-light region and no photosensitization in the visible light region. The initial concentration of 4-CP was 7.5 × 10⁻⁵ mol L⁻¹, and the dosage of photocatalyst was same to the typical photocatalytic experiments. The photocatalytic of the three sample bare Bi₂WO₆, BW-Zn and BW-Cd over 4-CP were shown in Fig. 9b, which clearly indicate that all of the three samples exhibit photocatalytic activities under visible light. Among the three sample, BW-Cd exhibits the highest photocatalytic activity. It should be noted that the photocatalytic activity of BW-Zn exhibits much lower photocatalytic activity as compared to BW-Cd, which also suggest that the photosensitization process play a key role in the photodecomposition process.

Fig. 9 (a) Photodegradation efficiency of BW-Cd and BW-Zn under different visible light (λ > 510 nm and λ > 400 nm) irradiation; (b) visible-light photodegradation of 4-CP in the presence of Bi₂WO₆, BW-Cd and BW-Zn.

In order to reveal the active species during the photocatalytic process, the trapping experiments were conducted. A series of quenchers such as benzoquinone (BQ, scavenger for O₂˙⁻), ammonium oxalate (AO, scavenger for h⁺), AgNO₃ (scavenger for e⁻), and t-BuOH (scavenger for ˙OH) were introduced to the Rh B solution before the addition of photocatalyst to ascertain the ruling active species,⁷⁴ and the corresponding results are shown in Fig. 10. The amount of these quenchers is 10 mM except for the BQ, which is 1 mM to avoid the reaction with Rh B. For both the two heterostructures, the addition of t-butanol (hydroxyl radicals scavenger) only cause a small decrease on the photodegradation rate of Rh B, indicating that hydroxyl radicals are not the driving force for the degradation of Rh B. On the contrary, the introduction of ammonium oxalate (the holes scavenger) cause an inhibition phenomenon on the photodegradation of Rh B, suggesting that the photo-generated holes play vital roles in the degradation of Rh B. Meanwhile, the photodegradation of Rh B was also substantially reduced by the addition of AgNO₃ (electrons scavenger) and BQ (superoxide radical anions scavenger), indicating that both electron and superoxide radical anions also exist in the solution.

Fig. 10 Effects of a series of scavengers on the photodegradation efficiency of Rh B using sample (a) BW-Cd, and (b) BW-Zn.

For sample BW-Cd, the enhanced photocatalytic performance can be well ascribed to the interfacial interaction between CdS and Bi₂WO₆. And the process of photodegradation can be described as follows:

CdS + hv → CdS(e_CB⁻…h_VB⁺)

CdS(e_CB⁻) + Bi₂WO₆ → CdS + Bi₂WO₆(e_CB⁻)

Bi₂WO₆(e_CB⁻) + O₂ → Bi₂WO₆ + O₂˙⁻

CdS(h_VB⁺) + Rh B → CO₂ + H₂O

While for BW-Zn, the photosensitization process play an important role in the photocatalytic process. Besides the band gap coupling of the two semiconductors, Rh B can also absorb incident photon flux owing to the intermolecular π–π* transition. So the photodegradation process of BW-Zn is different from sample BW-Cd, which could be described as follows:

Rh B + hv → Rh B(e⁻…h⁺)

Rh B(e⁻) + ZnS → ZnS(e_CB⁻) + Rh B

ZnS(e_CB⁻) + O₂ → ZnS + O₂˙⁻

Bi₂WO₆ + hν → Bi₂WO₆(e_CB⁻) + Bi₂WO₆(h_VB⁺)

Bi₂WO₆(e_CB⁻) + O₂ → Bi₂WO₆ + O₂˙⁻

Bi₂WO₆(h_VB⁺) + ZnS → Bi₂WO₆ + ZnS(h_VB⁺)

ZnS(h_VB⁺) + Rh B → CO₂ + H₂O

where e⁻_CB and h⁺_VB stand for the electron in the conduction band and hole in the valence band, respectively.

Based on the above experimental results, the schematic diagrams for the photocatalytic processes in the presence of BW-Cd and BW-Zn are presented in Scheme 2. When BW-Cd is irradiated, electrons are excited from the valence band (VB) of CdS to its conduction band (CB). Then, the CB-electrons of CdS could easily transfer to the CB of Bi₂WO₆, driven by the decrease in potential energy. This process could facilitate their participation in an electron reduction reaction of oxygen to activate molecular oxygen (e.g., the formation of superoxide radical anion). As a result, the recombination of electron–hole pairs were efficiently inhibited and the lifetime of charge carriers were prolonged, which would lead to the enhanced photocatalytic performances of the as-formed heterostructures.

Scheme 2 Schematic diagram showing separation of photo-generated charge carriers in sample BW-Cd and BW-Zn.

As for BW-Zn, the photocatalytic mechanism is different to BW-Cd. Because of the large band gap of ZnS, ZnS can't be excited under visible light. So, when BW-Zn was irradiated with visible light, electrons are easily excited from the VB of Bi₂WO₆ to its CB. Then the photo-generated holes will transfer to the VB of ZnS because of the inter-built electric field. Meanwhile, Rh B dye could absorb the incident photon flux upon visible light irradiation. Then, the photo-generated electrons were transferred to the excited state of the dye owing to the intramolecular π–π* transition and the dyes were oxidized. The photoelectrons of the excited state were immediately injected into the conducting band of Bi₂WO₆. The photoelectrons in the CB were then captured by O₂, and the succeeding reactions can lead to the decomposition of the dyes. Because of the formation of heterostructures, the separation efficiency of photo-generated electrons and holes are greatly enhanced, which will lead to the fast decomposition of Rh B in our experiment.

4. Conclusions

In this report, Bi₂WO₆/CdS and Bi₂WO₆/ZnS heterostructures was successfully prepared by a surface functionalization method using 3-mercaptopropionic (MPA) as the surface functionalizing agent. The formation of these type II heterostructure is proved to be an effective method to improve the photocatalytic performance of Bi₂WO₆. And the photodegradation efficiency of sample BW-Cd and BW-Zn for Rh B was approximately 3.72 and 1.85 times higher than bare Bi₂WO₆, respectively. The enhanced photocatalytic activities of the as-obtained heterostructures could be attributed to the facilitated interfacial charge transfer and inhibited electron–hole recombination, which has been confirmed by the photocurrent and PL measurement. The trapping experiments indicate the dominant active species in the photocatalytic process are photo-generated holes. Based on the experimental results, the photocatalytic mechanisms for samples BW-Cd and BW-Zn are proposed. Thus, these novel heterostructured materials are supposed to have potential applications in waste water treatment.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant no. 21373106) and Program for Scientific Research Innovation Team in Colleges and Universities of Shandong Province.

References

1. A. Hagfeldt and M. Gratzel, Chem. Rev., 1995, 95, 49–68.
2. Z. G. Zou, J. H. Ye and K. Sayama, Nature, 2001, 414, 625–627.
3. N. Z. Bao, L. M. Shen, T. Takata and K. Domen, Chem. Mater., 2008, 20, 110–117.
4. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
5. W. Morales, M. Cason, O. Aina, N. R. Tacconi and K. Rajeshwar, J. Am. Chem. Soc., 2008, 130, 6318–6319.
6. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahneman, Chem. Rev., 1995, 95, 6996.
7. Z. G. Zou, J. H. Ye, S. Kazuhiro and A. Hironori, Nature, 2001, 414, 625–627.
8. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
9. K. Maeda, K. Teramura, D. L. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Nature, 2006, 440, 295.
10. X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76–80.
11. (a) Z. G. Yi, J. H. Ye, N. Kikugawa, T. Kako, S. X. Ouyang, H. Stuart-Williams, H. Yang, J. Y. Cao, W. J. Luo, Z. S. Li, Y. Liu and R. L. Withers, Nat. Mater., 2010, 9, 559–564; (b) Y. P. Bi, S. X. Ouyang, N. Umezawa, J. Y. Cao and J. H. Ye, J. Am. Chem. Soc., 2011, 133, 6490–6492.
12. X. X. Xu, C. Randorn, P. Efstathiou and J. T. S. Irvine, Nat. Mater., 2012, 11, 595–598.
13. (a) H. B. Fu, C. S. Pan, W. Q. Yao and Y. F. Zhu, J. Phys. Chem. B, 2005, 109, 22432–22439; (b) Y. Tsunoda, W. Sugimoto and Y. Sugahara, Chem. Mater., 2003, 15, 632–635.
14. J. Wu, F. Duan, Y. Zheng and Y. Xie, J. Phys. Chem. C, 2007, 111, 12866–12871.
15. F. Amano, K. Nogami, R. Abe and B. Ohtani, J. Phys. Chem. C, 2008, 112, 9320–9326.
16. C. X. Xu, X. Wei, Y. M. Guo, H. Q. Wu, Z. H. Ren, G. Xu, G. Shen and G. R. Han, Mater. Res. Bull., 2009, 44, 1635–1641.
17. M. Shang, W. Z. Wang and H. L. Xu, Cryst. Growth Des., 2009, 9, 991–996.
18. L. S. Zhang, W. Z. Wang, Z. G. Chen, L. Zhou, H. L. Xu and W. Zhu, J. Mater. Chem., 2007, 17, 2526–2532.
19. Z. Chen, L. W. Qian, J. Zhu, Y. P. Yuan and X. F. Qian, CrystEngComm, 2010, 12, 2100–2106.
20. X. F. Cao, L. Zhang, X. T. Chen and Z. L. Xue, CrystEngComm, 2011, 13, 306–311.
21. L. Wu, J. H. Bi, Z. H. Li, X. X. Wang and X. Z. Fu, Catal. Today, 2008, 131, 15–20.
22. Z. J. Zhang, W. Z. Wang, M. Shang and W. Z. Yin, J. Hazard. Mater., 2010, 177, 1013–1018.
23. S. B. Zhu, T. G. Xu, H. B. Fu, J. C. Zhao and Y. F. Zhu, Environ. Sci. Technol., 2007, 41, 6234–6239.
24. S. M. Lopez, M. C. Hidalgo, J. A. Navio and G. Colon, J. Hazard. Mater., 2011, 185, 1425–1434.
25. M. S. Gui, W. D. Zhang, Y. Q. Chang and Y. X. Yu, Chem. Eng. J., 2012, 197, 283–288.
26. Z. J. Zhang, W. Z. Wang, L. Wang and S. M. Sun, ACS Appl. Mater. Interfaces, 2012, 4, 593–597.
27. J. X. Xia, J. Di, S. Yi, H. Xu, J. Zhang, Y. G. Xu, L. Xu, H. M. Li and M. X. Ji, RSC Adv., 2014, 4, 82–90.
28. Q. C. Xu, D. V. Wellia, Y. H. Ng, R. Amal and T. T. Y. Tan, J. Phys. Chem. C, 2011, 115, 7419–7428.
29. X. N. Li, R. K. Huang, Y. H. Hu, Y. J. Chen, W. J. Liu, R. S. Yuan and Z. H. Li, Inorg. Chem., 2012, 51, 6245–6250.
30. H. W. Huang, S. B. Wang, N. Tian and Y. H. Zhang, RSC Adv., 2014, 4, 5561–5567.
31. G. K. Fu, G. N. Xu, S. P. Chen, L. Lei and M. L. Zhang, Catal. Commun., 2013, 40, 120–124.
32. Y. Hu, Y. Liu, H. Qian, Z. Li and J. Chen, Langmuir, 2010, 26, 18570–18575.
33. D. Jing and L. Guo, J. Phys. Chem. C, 2007, 111, 13437–13441.
34. D. Barpuzary and M. Qureshi, ACS Appl. Mater. Interfaces, 2013, 5, 11673–11682.
35. G. S. Li, D. Q. Zhang and J. C. Yu, Environ. Sci. Technol., 2009, 43, 7079–7085.
36. Y. X. Yu, W. X. Ouyang, Z. T. Liao, B. B. Du and W. D. Zhang, ACS Appl. Mater. Interfaces, 2014, 6, 8467–8474.
37. Z. Fang, Y. F. Liu, Y. T. Fan, Y. H. Ni, X. W. Wei, K. B. Tang, J. M. Shen and Y. Chen, J. Phys. Chem. C, 2011, 115, 13968–13976.
38. Y. W. Mi, S. Y. Zeng, L. Li, Q. F. Zhang, S. N. Wang, C. H. Liu and D. Z. Sun, Mater. Res. Bull., 2012, 47, 2623–2630.
39. M. R. Kim, Y. Kang and D. Jang, J. Phys. Chem. C, 2007, 111, 18507–18511.
40. A. Ihs and B. Liedberg, J. Colloid Interface Sci., 1991, 144, 282–292.
41. N. Mahapatra, S. Panja, A. Mandal and M. Halder, J. Mater. Chem. C, 2014, 2, 7373.
42. U. Srinivasan, P. A. Mieyal and J. J. Mieyal, Biochemistry, 1997, 36, 3199–3206.
43. S. Y. Zeng, R. F. Tang, S. X. Duan, L. Li, C. H. Liu, X. L. Gu, S. S. Wang and D. Z. Sun, J. Colloid Interface Sci., 2014, 432, 236–245.
44. L. Ge and J. Liu, Mater. Lett., 2011, 65, 1828–1831.
45. P. Tierno and W. A. Goedel, J. Phys. Chem. B, 2006, 110, 3043–3050.
46. K. L. Zhang, C. M. Liu, F. Q. Huang, C. Zheng and W. D. Wang, Appl. Catal., B, 2006, 68, 125–129.
47. M. S. Gui, W. D. Zhang, Q. X. Su and C. H. Chen, J. Solid State Chem., 2011, 184, 1977–1982.
48. G. Marci, V. Augugliaro, M. J. López-Muñoz, C. Martin, L. Palmisano, V. Rives, M. Schiavello, R. J. D. Tilley and A. M. Venezia, J. Phys. Chem. B, 2001, 105, 1026–1032.
49. A. Kudo, I. Tsuji and H. Kato, Chem. Commun., 2002, 17, 1958–1959.
50. L. Zhang, J. H. Yan, M. J. Zhou, Y. P. Yu, Y. Liu and Y. N. Liu, Trans. Nonferrous Met. Soc. China, 2014, 24, 743–749.
51. N. Li, L. Zhu, W. D. Zhang, Y. X. Yu, W. H. Zhang and M. F. Hou, J. Alloys Compd., 2011, 509, 9770–9775.
52. G. K. Zhang, X. M. Ding, F. S. He, X. Y. Yu, J. Zhou, Y. J. Hu and J. W. Xie, Langmuir, 2008, 24, 1026–1030.
53. P. Sippel, D. Denysenko, A. Loidl, P. Lunkenheimer, G. Sastre and D. Volkmer, Adv. Funct. Mater., 2014, 24, 3885–3896.
54. D. N. Ke, T. Y. Peng, L. Ma, P. Cai and P. Jiang, Appl. Catal., A, 2008, 350, 111–117.
55. L. Ren, J. Lei, J. B. Wang, M. Qiu and Y. Yu, Nanotechnology, 2009, 20, 405602.
56. K. Vinodgopal, I. Bedja and P. V. Kamat, Chem. Mater., 1996, 8, 2180–2187.
57. S. Chatterjee, A. K. Tyagi and P. Ayyub, J. Nanomater., 2014, 2014, 1–7.
58. E. Forgacs, T. Cserháti and G. Oros, Environ. Int., 2004, 30, 953–971.
59. S. Senthilkumaar, K. Porkodi, R. Gomathi, A. Geetha Maheswari and N. Manonmani, Dyes Pigm., 2006, 69, 22–30.
60. S. Ikeda, N. Sugiyama, S. Murakami, H. Kominami, Y. Kera, H. Noguchi, K. Uosaki, T. Torimoto and B. Ohtani, Phys. Chem. Chem. Phys., 2003, 5, 778–783.
61. B. Ohtani, R. M. Bowman, D. P. Colombo Jr, H. Kominami, H. Noguchi and K. Uosaki, Chem. Lett., 1998, 27, 579–580.
62. V. A. Sakkas, I. M. Arabatzis, I. K. Konstantinou, A. D. Dimou, T. A. Albanis and P. Falaras, Appl. Catal., B, 2004, 49, 195–205.
63. H. Zhang and Y. F. Zhu, J. Phys. Chem. C, 2010, 114, 5822–5826.
64. D. E. Dunstan, A. Hagfeldt, M. Almgren, H. O. G. Siegbahn and E. Mukhtart, J. Phys. Chem., 1990, 94, 6197–6804.
65. W. R. Becquerel, J. Chem. Phys., 1960, 32, 1505–1514.
66. H. Gerischer, Pure Appl. Chem., 1980, 52, 2649–2667.
67. H. Xu, J. Yan, Y. G. Xu, Y. H. Song, H. M. Li, J. X. Xia, C. J. Huang and H. L. Wan, Appl. Catal., B, 2013, 129, 182–193.
68. Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 115, 7355.
69. H. Q. Li, Y. M. Cui and W. S. Hong, Appl. Surf. Sci., 2013, 264, 581–588.
70. (a) X. K. Li and J. H. Ye, J. Phys. Chem. C, 2007, 111, 13109–13116; (b) T. Watanabe, T. Takizawa and K. Honda, J. Phys. Chem., 1977, 81, 1845–1851.
71. (a) P. Qu, J. C. Zhao, T. Shen and H. Hidaka, J. Mol. Catal. A: Chem., 1998, 129, 257–268; (b) T. X. Wu, G. M. Liu, J. C. Zhao, H. Hidaka and N. Serpone, J. Phys. Chem. B, 1998, 102, 5845–5851.
72. P. Lei, C. Chen, J. Yang, W. Ma, J. Zhao and L. Zang, Environ. Sci. Technol., 2005, 39, 8466–8474.
73. C. Chen, W. Zhao, J. Li and J. Zhao, Environ. Sci. Technol., 2002, 36, 3604–3611.
74. Y. Wang, K. Deng and L. Zhang, J. Phys. Chem. C, 2011, 115, 14300–14308.

---

Footnote

† Electronic supplementary information (ESI) available: EDS spectra and BET spectra for the two heterostructures, XPS patterns of Bi 4f and W 4f of BW-MS samples are provided in supporting information. See DOI: 10.1039/c5ra04655f

---