Facile Synthesis of Heterostructured WS2/Bi2MoO6 as High-Performance Visible-Light-Driven Photocatalysts

In this paper, novel WS2/Bi2MoO6 heterostructured photocatalysts were successfully fabricated via a facile solvothermal growth method using pre-exfoliated layered WS2 nanoslices as a substrate. The structure, morphology, and optical properties of the as-prepared WS2/Bi2MoO6 samples were characterized by XRD, XPS, SEM, TEM (HRTEM), and UV-vis diffuse reflectance spectra (DRS). Results confirmed the existence of an excellent nanojunction interface between layered WS2 nanoslices and Bi2MoO6 nanoflakes. Under visible light (>420 nm), the WS2/Bi2MoO6 composites exhibit significantly enhanced photocatalytic activity compared with pure Bi2MoO6 toward the decomposition of rhodamine B (RhB). Meanwhile, the active species trapping experiments indicated that holes (h+) were the main active species during the photocatalytic reaction. The enhanced photocatalytic performance can be ascribed to the effective light harvesting, fast photogenerated electron–hole pairs separation, and excellent charge carrier transport of the WS2/Bi2MoO6 heterostructures. Moreover, the prepared WS2/Bi2MoO6 composites also show good structural and activity stability in repeatability experiments.


Background
The photocatalysis is widely regarded as one of the most promising environmental remediation technique due to the clean energy utilization method [1,2]. Generally, some accepted that high-efficient photocatalysts with wide forbidden gap, such as TiO 2 and ZnO, can only utilize ultraviolet light irradiation [3]. As to practical application, photocatalysis strategy will be a huge boost once a photocatalyst can favorably absorb the abundant solar energy in visible region. For this purpose, many attempts to probe visible-light photocatalyst for sufficient solar energy utilization by using the narrow band semiconductor [4][5][6]. Despite the single-phase photocatalyst can be excited smoothly by visible light, it still manifests low energy conversion efficiency due to poor charge separation efficiency resulting from rapid recombination of photo-induced electrons and holes [7]. It is widely accepted that the heterostructure can improve the separation probability of light-induced charge because the contact interfacial region of heterojunction will provide an internal electric field to restrain the recombination probability, thus resulting in an efficient photocatalytic performance. In general, the designed heterostructure will adopt at least one narrow band semiconductor to harvest more visual-light energy and then to generate more photo-induced charges [8,9].
As a novel photocatalyst, Bi 2 MoO 6 has received attention in the field of visual-light-driven photocatalysis because it possesses distinct sandwiched layered structure [10,11]. As previously mentioned, the pure Bi 2 MoO 6 is not suitable for the utilization as an efficient visible-light photocatalyst due to the high recombination probability of photogenerated charge carrier. Therefore, some effective strategies to meet this challenge by using the architecture of proper hybrid nanostructure and especially the introduction of two dimensional (2D) nanosheets have been proved as an effectual approach to strengthen interfacial charge transfer between two components in the process of photocatalytic reaction. Obviously, it is anticipated that the heterostructure between Bi 2 MoO 6 and 2D layered material will increase photocatalytic efficiency by visuallight irradiation [8].
Layered transition metal dichalcogenides (TMDs) are widely regarded as a kind of promising loading material because of their analogous graphene reticular structure [12,13]. Especially, monolayer and few layers of TMDs have important application for catalysis and energy storage due to their distinct electronic properties and high specific surface areas [14,15]. For example, monolayered and few-layer MoS 2 have recently paid the attention of the scientific community in photocatalysis research, which ascribes the lack of interlayer coupling and the absence of inversion symmetry resulting in the photoelectric property that differ markedly from those of the bulk [14,16,17]. From the material design perspective for an efficient visible-light-driven sensitized heterojunctional photocatalyst, the primary concern is that the hybrid narrow band gaps (1.1-1.7 eV) can closely match the solar spectrum [18]. In fact, the typical 2D layered semiconductors, such as MoS 2 or g-C 3 N 4 , have received significant attention to explore potential photocatalysis applications, which lead to TMD nanosheet which is often utilized as a supporter to establish the heterostructured composite photocatalysts via different energy band hybrid strategies [19,20]. For instance, the hierarchical MoS 2 /Bi 2 MoO 6 composites exhibited an efficient performance for photocatalytic oxidation of rhodamine B under visible-light irradiation [21]. However, the mono-or few-layer heterostructured architecture of WS 2 /Bi 2 MoO 6 as a visual-light photocatalyst has not been reported.
Herein, we demonstrated a facile strategy to fabricate heterostructured WS 2 /Bi 2 MoO 6 composite via a facile solvothermal growth method using pre-exfoliated layered WS 2 nanoslices as a supporter. The WS 2 /Bi 2 MoO 6 exhibits excellent photocatalytic activity towards the degradation of rhodamine B (RhB) under visible-light (λ > 420 nm) irradiation. According to the microstructure characterization analysis of XRD, XPS, SEM, and TEM, the possible photocatalytic mechanism of the few-layer WS 2 /Bi 2 MoO 6 composite was also elucidated. It is believed that the formation of junctions between Bi 2 MoO 6 and WS 2 can allow the prompt migration of photogenerated charge and reduce the self-agglomeration. It is postulated that the excellent photocatalytic activity of WS 2 /Bi 2 MoO 6 should be ascribed to its high migration efficiency of photo-induced carriers and the interfacial electronic interaction. These results also probably provide a valuable perspective to insight into the design of other heterostructured photocatalysts.

Preparation of the Few-Layer WS 2 Nanoslices
The liquid exfoliation of layered commercial WS 2 was accomplished following the modified report method [22].
Briefly, 50 mg commercial WS 2 powder (purchased from Aladdin Industrial Corporation) was added to 20 mL of ethanol/water with EtOH volume fractions of 40% added as dispersion solvent. The sealed flask was sonicated for 10 h, and then the dispersion was centrifuged at 3000 rpm for 20 min to remove aggregations. Finally, the supernatant was collected to obtain few-layer WS 2 nanoslices. To determine the concentrations of 2D nanosheets in the supernatant, we estimated the mass remaining in the supernatant by measuring the UV-vis absorption spectrum at fixed wavelength of 630 nm. The calculation result by virtue of Lambert-Beer Law indicated that the exfoliated WS 2 dispersion concentration was about 0.265 ± 0.02 mg/ml.

Synthesis of Hierarchical WS 2 /Bi 2 MoO 6 Composites
The WS 2 /Bi 2 MoO 6 samples were synthesized using a facile solvothermal method. Typically, 2 mmol of Bi(NO 3 ) 3 ·5H 2 O was added to 10 mL of ethylene glycol solution containing dissolved Na 2 MoO 4 ·2H 2 O with the Bi/Mo molar ratio of 2:1 under magnetic stirring. An appropriate amount of exfoliated WS 2 nanoslices was dispersed into 20 mL ethanol and ultrasonicated at room temperature for 45 min. Then, it was slowly added into the above solution, followed by stirring for 10 min to form a homogeneous phase. The resulting solution was transferred into a 50-mL Teflonlined stainless steel autoclave and kept at 160°C for 10 h. Subsequently, the autoclave was cooled to room temperature gradually. Finally, the precipitate was centrifuged and washed with ethanol and deionized water several times and dried in a vacuum oven at 80°C for 6 h. According to this method, WS 2 /Bi 2 MoO 6 composites with different WS 2 mass ratios (1, 3, 5, and 7 wt%) were synthesized. For comparison, the blank Bi 2 MoO 6 was prepared in the absence of WS 2 using the same experimental conditions.

Characterization of Photocatalysts
Structure and morphology of the sample was investigated by scanning electron microscopy (SEM; JEOL JSM-6701F, Japan), transmission electron microscopy (TEM; JEOL 2100, Japan), high-resolution transmission electron microscopy (HRTEM; JEOL 2100, Japan), and powder X-ray diffraction (XRD; Bruker D8 Advance using Cu-Kα radiation source, λ = 1.5406 Å, USA). The ultraviolet-visible diffuse reflectance spectra (DRS) of samples were performed at room temperature in the range of 200-800 nm on a UV-vis spectrophotometer (Cary 500 Scan Spectrophotometers,Varian, USA) equipped with an integrating sphere attachment. The electronic states of surface elements of the catalysts were identified using X-ray photoelectron spectroscopy (XPS; Shimadzu Corporation, Japan, Al-Kα X-ray source).

Measurement of Photocatalytic Activity
In all catalytic activity of experiments, 50 mg of the samples were added to aqueous RhB solution (50 mL, 10 mg/L) magnetically stirred in a Pyrex glass vessel and then radial irradiated with a 300 W Xe arc lamp (PLS-SXE 300, Beijing Perfect Company, Labsolar-III AG) to provide visible light with λ ≥ 420 nm by an ultraviolet UVCUT-420 nm cut-off filter (CE Aulight. Inc). The distance between the ultraviolet filter and aqueous RhB solution was about 6.5 mm. And the power density of visible light was 150 mW/cm 2 , which was estimated by the optical power meter (PD130, Thorlabs, USA). Prior to irradiation, the suspension was kept in the dark under magnetically stirred for 30 min to ensure the establishing of an adsorption/desorption equilibrium. At given time intervals, 2 mL were collected from the suspension and immediately centrifuged; the concentration of RhB after illumination was monitored at 553 nm by using UV-vis spectrophotometer (Shimadzu UV-2550, Shimadzu Corporation, Japan). The relative concentrations (C/C 0 ) of the RhB were determined by the absorbance (A/A 0 ) at 553 nm. All experiments were carried out at least in duplicate. The reported values were within the experimental error range of ±2%. Combining with Lambert-Beer law, the photocatalytic degradation rate constant (k) of RhB was obtained using the following formula: where C is RhB concentration at reaction time t, C 0 is the adsorption/desorption equilibrium concentration of RhB at the starting reaction time, and A and A 0 are the corresponding absorbance values.
In addition, to identify the active species generated during photocatalytic reactivity, various scavengers were added into the solution of RhB, including 2 mM isopropanol (IPA, a quencher of ·OH), 2 mM disodium ethylenediamine tetraacetic acid (EDTA; a quencher of h + ), and 2 mM p-benzoquinone (BQ; a ·O 2 − scavenger), and 40 mL/min N 2 (an electron quencher). The comparative trials of photocatalytic degradation were performed under the same reaction conditions as those mentioned above.

Micostructure and Morphology Analysis
In order to confirm the composition and crystal structure of the as-prepared samples, an XRD study was carried out. As shown in Fig. 1, it can be found that the pure WS 2 , five peaks located at 14 In the case of the few-layer WS 2 /Bi 2 MoO 6 composite materials, the XRD pattern only displays the characteristic diffraction peaks of hexagonal phase WS 2 and orthorhombic phase Bi 2 MoO 6 . Furthermore, compared with the standard data for Bi 2 MoO 6 (no. 76-2388), the existence of few-layer WS 2 did not change the diffraction peak positions of Bi 2 MoO 6 in the composite sample, indicating Bi 2 MoO 6 nanoflakes grown on few-layer WS 2 nanoslices rather than incorporated into the WS 2 lattice. There is no trace of any impurity phase under the present resolution, which suggests the high purity of the as-prepared samples.
The morphologies of the as-synthesized samples were investigated using SEM. For comparison, SEM images of the bulk raw WS 2 without sonicated treatment and exfoliated nanoslices are shown in Fig. 2a, b. The former displays a distinct multi-layer laminated morphology with about 20 μm in thickness, while the latter exhibits 2D sheet-like morphology with thickness varying from dozens of nanometers to 1-2 μm. The results demonstrate that the layered commercial WS 2 have been stripped to fewlayer WS 2 nanoslices. Figure 2c shows the SEM image of pure Bi 2 MoO 6 . It can be seen that the Bi 2 MoO 6 exhibited microsphere morphology with rough surfaces. Closer examination reveals that the microspheres consist of numerous secondary Bi 2 MoO 6 nanoplates. Furthermore, when Bi 2 MoO 6 was deposited onto the 2D few-layer WS 2 via a facile solvothermal process (Fig. 2d), it can be clearly seen that the surfaces of WS 2 nanoslices were uniformly covered by numerous two-dimensional Bi 2 MoO 6 nanoplates (Fig. 2d) and that formed a WS 2 /Bi 2 MoO 6 hierarchical structure. Further information about the nano-structure of the few-layer WS 2 /Bi 2 MoO 6 composites was obtained from TEM (HRTEM) images. It is easy to observe in Fig. 3a that WS 2 (purple arrows) shows a clear nanosheet structure which is similar to that of graphene, proving that graphene-like tungsten disulfide is obtained. Meanwhile, Bi 2 MoO 6 nanoplates with diameters of about 50-100 nm were observed to grow on the WS 2 nanosheets. HRTEM images (Fig. 3b, c) taken from Fig. 3a clearly display the resolved lattice fringes of 0.274 and 0.227 nm, which corresponds to the (200) planes of orthorhombic phase of Bi 2 MoO 6 and the (103) planes of WS 2 , respectively. Therefore, the experimental results indicated that a coherent and tight heterojunction interface between fewlayer WS 2 and Bi 2 MoO 6 was formed, which can benefit better charge separation and efficient electron transfer within the hybrid structure in comparison with pure Bi 2 MoO 6 .

Electronic Structure and Spectrum Analysis
The elemental composition and oxidation states of the fewlayer WS 2 /Bi 2 MoO 6 composites were further determined by XPS spectra. Figure 4a shows the survey XPS spectra of the few-layer WS 2 /Bi 2 MoO 6 (5 wt%) sample, which exhibits W, S, O, Bi, Mo, and C peaks. No peaks corresponding to other elements are observed. The peak for Bi 4f in the Bi 2 MoO 6 ( Fig. 4b) which appeared at 164.4 and 159.2 eV belonged to Bi 4f 5/2 and Bi 4f 7/2 of Bi 3+ ions [23]. The Mo 3d binding energy (Fig. 4c) of 235.6 and 232.5 eV is consistent with the Mo 3d 3/2 and Mo 3d 5/2 of Mo 4+ ions [23]. The asymmetric peaks of O 1 s (Fig. 4d) are located at 530.0 eV, which are characteristic of the Mo-O [24]. However, the binding energies of Bi 4f, Mo 3d, and O 1 s in the XPS spectra (Fig. 4b-d) of the hierarchical WS 2 /Bi 2 MoO 6 slightly shift (about 0.2 eV) toward lower binding energies as compared with the pure Bi 2 MoO 6 . Meanwhile, in the hierarchical WS 2 / Bi 2 MoO 6 composite, the values of W 4f 5/2 (34.2 eV) and W 4f 7/2 (32.0 eV) peaks (Fig. 4e) corresponding to WS 2 are slightly lower (about 0.2 eV) than the pure WS 2 (34.4 and 32.2 eV). Similarly, the high-resolution S 2p spectrum (Fig. 4f ) also slightly shifts toward lower binding energies of 0.3 eV. These results could be ascribed to the strong interaction between WS 2 and Bi 2 MoO 6 resulting in an inner shift of Bi 4f, Mo 3d, O 1 s W 4f, and S 2p orbits [21,25]. Therefore, by combining the XRD, SEM, TEM, and XPS investigations, it revealed that there are both WS 2 and Bi 2 MoO 6 species in the hierarchical WS 2 /Bi 2 MoO 6 composite and that the heterojunctions are formed in their contact interface.   Figure 5a shows a comparison of the UV-vis diffuse reflectance spectra (UV-Vis-DRS) of the WS 2 , Bi 2 MoO 6 and hierarchical WS 2 /Bi 2 MoO 6 composite with different WS 2 contents. It can be clearly seen that the absorption spectrum of pure Bi 2 MoO 6 extends from the UV region to visible light at about 450 nm. When WS 2 combined with Bi 2 MoO 6 , the absorption spectrum of the hierarchical composite exhibits an obviously redshift and more intensive absorption within the visible-light range in comparison with pure Bi 2 MoO 6 . Meanwhile, when the content of WS 2 increased to a relatively high (3 to 7 wt%), the hierarchical composite display surprisingly strong absorption around 450-800 nm. These results clearly indicate that the composite photocatalyst could absorb more photons during photocatalytic reaction. Therefore, it can be revealed that the addition of WS 2 nanoslices is beneficial for the visible-light absorbance of the WS 2 /Bi 2 MoO 6 composite.
In addition, the optical band gap energies (E g ) of samples were calculated by the following equation [26]: where α, hv, A, and E g are absorption coefficient, photon energy, proportionality constant, and bandgap, respectively. The value of n is determined by the type of transition (direct (n = 1) or indirect (n = 4)) [27,28]. A plot of (ahv) 2 versus (hv) is converted according the UV-Vis-DRS. As shown in Fig. 5b, the E g values of pure WS 2 and Bi 2 MoO 6 have been estimated to be 1.47 and 2.72 eV, respectively.

Photocatalytic Activity
The photocatalytic activities of the as-prepared samples were measured by degrading rhodamine B (RhB) under visible-light irradiation. For comparison, photocatalytic activities of pure Bi 2 MoO 6 and mechanically mixed samples (5% WS 2 and 95% Bi 2 MoO 6 ) have also been investigated. As shown in Fig. 6a, the self-degradation effect of RhB under visible-light irradiation could be ignored. It can be clearly seen that the photodegradation rate of RhB by the pure Bi 2 MoO 6 was only~39% after 100 min of visible-light irradiation. Obviously, all the hierarchical WS 2 /Bi 2 MoO 6 composites show better photocatalytic performance than the pure Bi 2 MoO 6 .~48,~74,~95, and~88% of RhB were degraded using 1% WS 2 / Bi 2 MoO 6 , 3% WS 2 /Bi 2 MoO 6 , 5% WS 2 /Bi 2 MoO 6 , and 7% WS 2 /Bi 2 MoO 6 , respectively. The results indicate that the optimal WS 2 content in WS 2 /Bi 2 MoO 6 composite exists when the mass ratio is 5%. Meanwhile, it was noted that the WS 2 /Bi 2 MoO 6 (5 wt%) composite exhibits remarkably superior photocatalytic activity than the mechanically mixed 5% WS 2 and 95% Bi 2 MoO 6 . This strongly suggests that an effective nanojunction interface contact and strong interactions between WS 2 and Bi 2 MoO 6 are extremely useful to enhance the migration, transport, and separation processes of photogenerated carriers. Furthermore, such superior photocatalytic performances could be attributed to the good crystallization and high specific surface area of composites and the small sheet thickness of the WS 2 substrate. In addition, the pseudo-first-order kinetics model was used to fit the experimental data of the photocatalytic degradation of the RhB solution, and the results are given in Fig. 6b. The rate constant k is 0.0280 min −1 for the hierarchical WS 2 /Bi 2 MoO 6 (5 wt%) composites, which is 3.8 and 7.1 times greater than those of mechanically mixed WS 2 and Bi 2 MoO 6 and pure Bi 2 MoO 6 , respectively. These results indicated that RhB could be degraded more efficiently by the hierarchical WS 2 /Bi 2 MoO 6 composite photocatalyst. Figure 7 shows the UV-vis adsorption spectra changes of RhB solution degradation over the WS 2 /Bi 2 MoO 6 (5 wt%) composite photocatalyst, which was performed to further study the photocatalytic degradation process of RhB. It can be seen that the main absorption peak of  RhB gradually shifted from 552 to 537 nm, corresponding to the stepwise formation of a series of N-de-ethylated intermediates. As the visible-light irradiation process continues, the peak located at 537 nm is continued to shift and decrease, which indicates that the RhB molecules were further decomposed into smaller molecular fragments and the structure of RhB was also destroyed in the end. The two-step transition processes for photodegradation of RhB were also reported in several previous studies [29,30]. Meanwhile, the suspension loses color gradually in the experiment, which further indicates that the structure of RhB has been destroyed in the end.

Catalyst Stability
The photocatalytic stability of the hierarchical WS 2 / Bi 2 MoO 6 composites was investigated by repeatability experiments for degradation of RhB, as shown in Fig. 8a. It can be found that the photocatalytic activity of WS 2 / Bi 2 MoO 6 remains stable in the first two-cycle experiments. After four recycles, the catalysts did not show an obvious decrease in photocatalytic activity, demonstrating that WS 2 /Bi 2 MoO 6 composite retained a relatively high degradation activity during the photodegradation process. Furthermore, the catalyst samples collected after four cycles was characterized by XRD measurement (Fig. 8b). It can be seen that the crystal structure and phase composition of WS 2 /Bi 2 MoO 6 composite do not change after four photocatalytic reactions. Thus, the good structural stability ensures the WS 2 /Bi 2 MoO 6 composite efficient photocatalysts working under visible-light irradiation.
Possible Photocatalytic Mechanism Figure 9 shows the trapping experiment of main active species in the photocatalytic process of the WS 2 /Bi 2 MoO 6 composite. Isopropanol (IPA), 1,4-benzoquinone (BQ), and disodium ethylenediamine tetraacetic acid (EDTA) acted as the scavengers for ·OH, ·O 2 − , and h + , respectively. It can be observed that the addition of 2 mM IPA or BQ in the RhB solution had a little effect on the rate constant k app , suggesting that ·OH and ·O 2 − are the secondary active species during the photocatalytic reaction, not the main active species during the photocatalytic reaction. On the contrary, the k app for degradation of RhB obviously decreased after the addition of 2 mM EDTA. Therefore, it can be confirmed that h + play a key role for degradation of RhB. Furthermore, N 2 was bubbled into the RhB solution at the rate of 40 mL/ min to ensure that the reaction was operated without O 2 as an electron quencher. The degradation of RhB showed a slight decrease in comparison with the case of the airequilibrated solution and further indicated that ·O 2 − played a minor role.
To explain the enhanced photocatalytic performance, conduction band (CB) and valence band (VB) of WS 2 and Bi 2 MoO 6 potentials should be calculated. For a semiconductor, the bottom CB and top VB can be estimated by  [31]: E CB = X − E 0 − 0.5E g and E VB = E CB + E g , where E CB (E VB ) is the CB (VB) edge potential; X is the electronegativity of the semiconductor; E 0 is the energy of free electrons of the hydrogenscale (~4.5 eV vs NHE); and E g is the band gap energy of the semiconductor obtained from the UV-visible diffuse reflectance absorption. The X values for WS 2 and Bi 2 MoO 6 are calculated to be 5.66 and 5.55 eV, respectively [28,32,33]. Thus, E CB and E VB values of WS 2 are determined to be +0.43 and +1.9 eV and Bi 2 MoO 6 are −0.31 and +2.41 eV, respectively.
On the basis of the above results, a possible photocatalytic mechanism scheme of the WS 2 /Bi 2 MoO 6 composite photocatalyst is shown in Fig. 10. It can be found that WS 2 and Bi 2 MoO 6 are excited under visible-light irradiation and generate electrons and holes in their CB and VB, respectively. The electrons on CB of Bi 2 MoO 6 will easily transfer WS 2 due to the CB potential of Bi 2 MoO 6 (−0.31 eV) is more negative than the CB potential of WS 2 (0.43 eV) [29,30]. The few-layer WS 2 nanoslices could act as effective electron collectors, which was favorable to the separation of electron-hole pairs in Bi 2 MoO 6 . Therefore, this fast electron and hole transfer process can decrease the recombination of charges and prolong the lifetime of holes on VB of Bi 2 MoO 6 [34]. The CB potential of WS 2 (+0.43 eV) is more positive than E 0 (O 2 /·O 2 − ) (−0.046 eV) which suggests that the ·O 2 − radicals were not formed through electrons reducing the dissolved O 2 [35]. However, a few electrons on the CB of Bi 2 MoO 6 can react with dissolved O 2 to yield ·O 2 − radicals because its potential (−0.31 eV) is more negative than E 0 (O 2 /·O 2 − ). Thus, the ·O 2 − active species played a minor role. Meanwhile, the photoinduced holes on VB of Bi 2 MoO 6 could not also directly oxidize the adsorbed H 2 O molecules to ·OH radicals because its potential (+2.41 eV) was lower than E 0 (·OH/H 2 O) (+2.68 V) [36]. Finally, the main active species holes and minor active species ·O 2 − act as a strong oxidizing agent to oxidize the organic pollutants (RhB) to CO 2 and H 2 O. Therefore, the hierarchical WS 2 / Bi 2 MoO 6 composites exhibit improved photocatalytic activity.

Conclusions
In summary, a novel WS 2 /Bi 2 MoO 6 heterostructured photocatalysts were successfully fabricated via a facile solvothermal growth method using pre-exfoliated layered WS 2 nanoslices as a substrate. The hierarchical WS 2 /Bi 2 MoO 6 exhibits excellent photocatalytic activity towards the degradation of rhodamine B (RhB) under visible-light irradiation. Based on the results of a series of structure and performance tests, it is believed that there formed a tight nanojunction interface between layered WS 2 nanoslices and Bi 2 MoO 6 nanoflakes, which make the photo-induced electrons be easily transferred to the WS 2 substrate. As a result, the recombination of charges was decreased and the lifetime of holes was prolonged. Therefore, the hierarchical WS 2 /Bi 2 MoO 6 composites exhibit much higher visible-light-driven photocatalytic activity than the pure Bi 2 MoO 6 . Furthermore, the WS 2 / Bi 2 MoO 6 composites are very stable under visible-light irradiation and cycling photocatalytic tests. Thus, the as-prepared WS 2 /Bi 2 MoO 6 photocatalyst has potential application for pollutant abatement.