Synthesis of novel and stable g-C3N4-Bi2WO6 hybrid nanocomposites and their enhanced photocatalytic activity under visible light irradiation

Graphitic carbon nitride (g-C3N4) nanosheets with a thickness of only a few nanometres were obtained by a facile deammoniation treatment of bulk g-C3N4 and were further hybridized with Bi2WO6 nanoparticles on the surface via a solvothermal method. The composite photocatalysts were characterized by powder X-ray diffraction, scanning electron microscopy, transmission electron microscopy, UV–vis diffuse reflection spectroscopy and X-ray photoelectron spectroscopy (XPS). The HR-TEM results show that the nano-sized Bi2WO6 particles were finely distributed on g-C3N4 sheet surface, which forms heterojunction structure. The UV–vis diffuse reflectance spectra (DRS) show that the absorption edge of composite photocatalysts shifts towards lower energy region in comparison with those of pure g-C3N4 and Bi2WO6. The degradation of methyl orange (MO) tests reveals that the optimum activity of 8 : 2 g-C3N4-Bi2WO6 photocatalyst is almost 2.7 and 8.5 times higher than those of individual g-C3N4 and Bi2WO6. Moreover, the recycle experiments depict high stability of the composite photocatalysts. Through the study of the influencing factors, a possible photocatalytic mechanism is proposed. The enhancement in both photocatalytic performance and stability was caused by the synergistic effect, including the effective separation of the photogenerated electron-hole pairs at the interface of g-C3N4 and Bi2WO6, the smaller the particle size and the relatively larger specific surface area of the composite photocatalyst.

HL, 0000-0001-8358-3480 Graphitic carbon nitride (g-C 3 N 4 ) nanosheets with a thickness of only a few nanometres were obtained by a facile deammoniation treatment of bulk g-C 3 N 4 and were further hybridized with Bi 2 WO 6 nanoparticles on the surface via a solvothermal method. The composite photocatalysts were characterized by powder X-ray diffraction, scanning electron microscopy, transmission electron microscopy, UVvis diffuse reflection spectroscopy and X-ray photoelectron spectroscopy (XPS). The HR-TEM results show that the nano-sized Bi 2 WO 6 particles were finely distributed on g-C 3 N 4 sheet surface, which forms heterojunction structure. The UV-vis diffuse reflectance spectra (DRS) show that the absorption edge of composite photocatalysts shifts towards lower energy region in comparison with those of pure g-C 3 N 4 and Bi 2 WO 6 . The degradation of methyl orange (MO) tests reveals that the optimum activity of 8 : 2 g-C 3 N 4 -Bi 2 WO 6 photocatalyst is almost 2.7 and 8.5 times higher than those of individual g-C 3 N 4 and Bi 2 WO 6 . Moreover, the recycle experiments depict high stability of the composite photocatalysts. Through the study of the influencing factors, a possible photocatalytic mechanism is proposed.
2018 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited. 1

. Introduction
Semiconductor photocatalysts have drawn much attention in the past decades because they represent a promising technology to use natural sunlight energy to promote chemical reactions, such as watersplitting, pollutant degradation and organic transformation [1,2]. Up to now, a mass of oxide and sulfide semiconductor photocatalysts have been developed for photocatalytic reactions, for example, the extensively investigated titanium dioxide and zinc sulfide [2,3]. However, the traditional photocatalysts are active only in the UV region and have high electron-hole recombination rates, which led to its inability to make full use of solar energy and reduce the photocatalytic performance [4]. Thus, it is still a challenging task looking for the sustainable, high-efficiency and visible-light-responsive photocatalytic materials.
Recently, non-metallic polymer photocatalyst g-C 3 N 4 , which shows superior photocatalytic activity for hydrogen production through water-splitting under visible light irradiation, was reported by Wang et al. [5]. Unlike traditional wide band-gap photocatalysts, this novel g-C 3 N 4 photocatalyst possesses a narrow band gap (2.71 eV) and favourable thermal and chemical stability, which make it a unique electronic structure and highly condensed [5]. In addition, carbon and nitrogen elements, the composition of g-C 3 N 4 , are abundant in natural source. All these merits of g-C 3 N 4 make it a valuable photocatalyst for solar energy-driven applications. However, the photocatalytic activity of g-C 3 N 4 is still restricted due to the high electron-hole recombination rate and low specific surface area (generally below 10 m 2 g -1 for the bulk g-C 3 N 4 ) [6]. To further improve the photocatalytic activity of g-C 3 N 4 , many strategies have been proposed, including element doping, designing a textural porosity, as well as coupling g-C 3 N 4 with heterogeneous semiconductor composites [7][8][9].
To reduce the recombination rates of photogenerated electron-hole pairs, construction of heterojunctions by coupling of two semiconductors was proved to be an effective strategy. Up to now, several g-C 3 N 4 -based heterojunction composites have been reported, for instance, g-C 3 N 4 -TiO 2 , g-C 3 N 4 -ZnO, g-C 3 N 4 -CdS, g-C 3 N 4 -BiOBr and g-C 3 N 4 -TaON [10][11][12][13][14]. All these composites exhibit enhanced photocatalytic performance with respect to the sole component, which was ascribed to the effective separation of charged carriers. Considering the band structures of two coupled semiconductors, three types of heterojunction structures are usually formed. Type II heterojunction was regarded as the most effective band structure to separate charged carriers, in which the electrons transfer from one semiconductor to the other, while the holes migrate reversely [15]. According to the conduction band (CB) and valence band (VB) potential value of g-C 3 N 4 (−1.13 and 1.58 eV, respectively), it was considered that the Bi 2 WO 6 (0.46 and 3.26 eV, respectively) is one of the most suitable components to form type II heterojunction with g-C 3 N 4 based on energy levels [16][17][18]. Furthermore, the crystal structure of Bi 2 WO 6 comprises accumulated layers of alternating bismuth oxide (Bi 2 O 2 ) 2+ and tungsten oxide (WO 4 ) 2− sheets, which is favourable for charge transfer in plane.
This paper presents a facile strategy to produce hybrid g-C 3 N 4 nanosheets with Bi 2 WO 6 nanoparticles by two steps. First, bulk g-C 3 N 4 was synthesized by thermal polycondensation of melamine. To make the specific surface area of g-C 3 N 4 larger, the bulk g-C 3 N 4 was treated by a simple deammoniation process at 500°C for 3 h, and then the g-C 3 N 4 nanosheets with several nanometres thickness were obtained. Second, using g-C 3 N 4 nanosheets as bases, Bi 2 WO 6 nanoparticles were deposited onto g-C 3 N 4 surface by a solvothermal process and a series of composite photocatalysts with different weight ratio were prepared. It was found that the two components coexisted and closely constructed a heterojunction structure. Furthermore, owing to the larger specific surface area and well-matched band structures, the activity of the obtained composed catalysts was significantly high than those of the pure Bi 2 WO 6 and g-C 3 N 4 respectively. In addition, these composed catalysts were very stable and could be used multiple times, retaining a relatively high photocatalytic activity. Finally, through the analysis of the capture experiment, the possible photocatalytic mechanism was put forward.

Synthesis
All chemicals used were commercially available and used without further purification. Bulk g-C 3 N 4 was prepared by the thermal treatment of 4 g melamine at 550°C for 4 h in a muffle furnace under an ambient pressure. Further deammoniation treatment was performed at 500°C for 3 h [19,20]. After reaction, the alumina crucible was cooled to room temperature. The resultant yellow product was collected for further use.
The g-C 3 N 4 -Bi 2 WO 6 composite photocatalysts were obtained by hybridization of g-C 3 N 4 nanosheet with Bi 2 WO 6 nanoparticles through a solvothermal method. In a typical procedure, a certain amount of g-C 3 N 4 sheets were added into 30 ml triethylene glycol (TEG) and sonicated for 30 min, and then a certain percentage of Bi(NO 3 ) 3 ·5H 2 O dissolved in 20 ml TEG by stirring and added dropwise to the above suspension. The mixture was stirred for 1.5 h at room temperature to form a homogeneous solution. Meanwhile, a certain amount of Na 2 WO 4 ·2H 2 O solid was added to 20 ml ethylene glycol (EG) to obtain a uniform solution. After stirring and dissolving, the solution was added rapidly to the above mixture suspension and then stirred for another 3 h at room temperature. After that, the solution was transferred into a Teflon-lined steel autoclave and kept at 200°C for 12 h. Subsequently, the precipitate was collected by centrifugal separation, washed with distilled water and ethanol several times, and then dried at 60°C for 12 h. Finally, the obtained g-C 3 N 4 -Bi 2 WO 6 photocatalysts were ground for further use. In the same procedure, different mass ratios of g-C 3 N 4 -Bi 2 WO 6 at 2 : 8, 5 : 5 and 8 : 2 were prepared and denoted as 2 : 8 g-C 3 N 4 -Bi 2 WO 6 , 5 : 5 g-C 3 N 4 -Bi 2 WO 6 and 8 : 2 g-C 3 N 4 -Bi 2 WO 6 , respectively. The pure Bi 2 WO 6 sample was synthesized according to the same method without g-C 3 N 4 .

Characterization
XRD patterns were recorded on a PANalytical X'pert PRO powder diffractometer equipped with a Cu Kα radiation source (λ = 0.15405 nm). FT-IR spectra were measured on a Thermo Electron Nicolet-Nexus 470 FT-IR spectrometer (KBr disc). X-ray photoelectron spectroscopy (XPS) of the photocatalysts was performed on an ESCALAB 250Xi (Thermo Electron Corporation, USA). SEM was employed to observe the morphology of samples (Hitachi S-4800 microscope). TEM and HR-TEM were applied to characterize the microstructure of the samples (JEOL, JEM-2100). UV−vis diffuse reflectance spectra (DRS) were measured using a UV−vis spectrometer (Shimadzu, UV-3600, Japan). The photoluminescence (PL) spectra of the photocatalysts were obtained by a VARIAN fluorescence spectrophotometer.

Photocatalytic tests
The photocatalytic activity of the catalysts was tested by degrading the methyl orange (MO) solution under a visible light using a 500 W xenon lamp with a 420 nm cut-off filter as the light source. The specific test procedure was as follows: 50 mg of the sample was dispersed in 50 ml of methyl orange solution (5 mg l −1 ) and stirred for 120 min under dark conditions to achieve adsorption equilibrium. After the start of the experiment, for a certain time period, about 3 ml suspensions were collected and centrifuged to wipe off the photocatalytics. The concentrations of residued MO were then monitored by a UV-vis spectrometer at a wavelength of 463 nm. 3. Results and discussions 3. 1 Figure 1 shows the XRD patterns of the as-prepared samples. The pure g-C 3 N 4 sample shows two diffraction peaks at 12.97°and 27.93°, corresponding to the (100) and (002) planes, which are attributed to the periodic arrangement of the basic structural units in the g-C 3 N 4 and the interlayer build-up of the cyclic aromatic substance [6]. For pure Bi 2 WO 6 , there is a series of narrow and pointed diffraction peaks, which are in good agreement with the orthorhombic phase of Bi 2 WO 6 (JCPDS 39-0256) without any impurity. After the two materials are compounded, the g-C 3 N 4 -Bi 2 WO 6 composite was similar to the pure Bi 2 WO 6 diffraction pattern. This phenomenon can be ascribed to two aspects: one is that the (002) of g-C 3 N 4 and (131) of Bi 2 WO 6 diffraction peaks were located in a similar position (approx. 28°), Soc. open    FT-IR spectra of the as-synthesized g-C 3 N 4 , Bi 2 WO 6 and g-C 3 N 4 -Bi 2 WO 6 . and the other is that the g-C 3 N 4 layer was too thin, thus the diffraction intensity is relatively weak with respect to those of Bi 2 WO 6, which make g-C 3 N 4 diffractions invisible in the composite [21].

. Characterization
The FT-IR spectra of g-C 3 N 4 , Bi 2 WO 6 and g-C 3 N 4 -Bi 2 WO 6 photocatalysts are shown in figure 2a. The absorption peaks at 3428 cm −1 and 1628 cm −1 in the spectrum of pure Bi 2 WO 6 are attributed to the stretching vibration and bending vibration of the O-H, respectively. For g-C 3 N 4 sample, the 3155 cm −1 band can be attributed to the N-H stretching mode. The enlarged spectrum from 400 to 1600 cm −1 is shown in figure 2b. The main absorption peaks at 400-800 cm −1 in pure Bi 2 WO 6 sample correspond to Bi-O, W-O stretching and W-O-W bridging stretching modes [22]. In the case of pure g-C 3 N 4 , the intense band at 808 cm −1 belongs to the characteristic breathing mode of s-triazine, and the strong bands in the 1200-1600 cm −1 region with peaks at 1243, 1324, 1401 and 1569 cm −1 correspond to typical stretching vibration modes of C=N and C-N heterocycles [23,24]. All these peaks can be found in g-C 3 N 4 -Bi 2 WO 6 photocatalyst, indicating the presence of g-C 3 N 4 and Bi 2 WO 6 components in composite photocatalysts.
The elemental composition of the samples was measured by XPS. We can see that the g-C 3 N 4 mainly consisted of C, N and a little amount of O elements (figure 3a). The O element that appeared here may be ascribed to O 2 adsorbed on the surface during polymerization process, which usually occurs in synthetic g-C 3 N 4 materials [25]. As shown in figure 3b, two peaks at 157.58 and 162.88 eV can be attributed to Bi 4f 7/2 and 4f 5/2 of Bi 3+ ions. The XPS of W 4f electrons is shown in figure 3c, indicating the + 6 valence of    literature [30]. After combining the two semiconductors, a gradual red shift appeared as the amount of g-C 3 N 4 increased; this should be the result of the interaction between g-C 3 N 4 and Bi 2 WO 6 in the heterojunction [31,32]. The band-gap values were also estimated from the intercept of tangents to plots of (Ahv) 1/2 versus photon energy [31], as shown in figure 4b. It is worth noting that the g-C 3 N 4 -Bi 2 WO 6 composite shows narrower band gap in comparison with pure g-C 3 N 4 and Bi 2 WO 6 . The results indicate that the g-C 3 N 4 -Bi 2 WO 6 heterojunction complex photocatalyst has a good visible light response, which can enhance its photocatalytic activity.
As seen from the SEM image of figure 5a, g-C 3 N 4 has a hierarchical flower-like morphology, which was assembled by many nanoflakes with a thickness of only a few nanometres. The nanoflake structure was further characterized by TEM observation. As shown in figure 5c, a typically two-dimensional lamellar structure with a thickness of several nanometres of g-C 3 N 4 nanoflakes was observed, which is similar to the reported reference [33]. The SEM image of 8 : 2 g-C 3 N 4 -Bi 2 WO 6 composite is exhibited in figure 5b. Clearly, with the coupling of Bi 2 WO 6 nanoparticles, the hierarchical structures of g-C 3 N 4 were disassembled into discrete nanoflakes. With the increase of Bi 2 WO 6 content, the g-C 3 N 4 nanoflake was  completely covered by Bi 2 WO 6 nanoparticles, which would have a negative effect on the photocatalysis because of the decrease in photocatalytic reaction site on the g-C 3 N 4 nanoflake surface. Figure 5d shows the TEM image of pure Bi 2 WO 6 sample. It can be observed that Bi 2 WO 6 formed irregular nanoparticles with a mean size of approximately 30 nm. From the TEM image of the 8 : 2 g-C 3 N 4 -Bi 2 WO 6 sample (figure 5e), we know that Bi 2 WO 6 nanoparticles were closely covered on the g-C 3 N 4 nanosheet surface. The HR-TEM image of figure 5f shows lattice fringe spacing of 0.316 and 0.273 nm, which belongs to (113) and (060) lattice planes of cubic Bi 2 WO 6 . Because of low crystallinity of the g-C 3 N 4 , it is hard to find the lattice fringe of g-C 3 N 4 [23]. The result obviously shows that g-C 3 N 4 nanosheets form a close heterogeneous contact with Bi 2 WO 6 . This close contact interface accelerates the migration of photogenerated electrons and holes between g-C 3 N 4 and Bi 2 WO 6 , inhibiting the recombination of photogenerated electron-hole pairs and improving the photocatalytic activity.

Photocatalytic performance
Based on the above conclusions, we evaluated the photocatalytic activity of the synthesized samples by degrading methyl orange under visible light irradiation. As shown in figure 6a, the highest degradation rate of 95.88% after 120 min irradiation was obtained over 8 : 2 g-C 3 N 4 -Bi 2 WO 6 composite sample. Meanwhile, no other peaks in the UV region indicate the full decomposition of aromatic component. To compare the photocatalytic performance of different photocatalysts, the degradation rate of MO was determined by the characteristic absorption peak of methyl orange at 463 nm. The C/C 0 versus irradiation time is plotted in figure 6b. The concentration of MO solution without photocatalyst did not change after 120 min degradation, indicating that methyl orange is quite stable, and ruled out the possibility of the occurrence of its self-degradation. As the Bi 2 WO 6 sample has a relatively high photogenerated electron-hole pairs recombination rate, the pure Bi 2 WO 6 sample exhibits its low photocatalytic activity. Similarly, only 68.8% of MO was degraded by the g-C 3 N 4 sample for the same irradiation time. It was expected that the construction of proper heterojunction can enhance the photocatalytic activity through effective separation of photogenerated electron-hole pairs. As proved in this paper, all the g-C 3 N 4 -Bi 2 WO 6 composite photocatalysts present enhanced photocatalytic ability for the degradation of MO compared to pure Bi 2 WO 6 and g-C 3 N 4 . Besides, the highest activity was obtained over 8 : 2 g-C 3 N 4 -Bi 2 WO 6 composite, and its degradation rate was 1.4 and 3.2 times higher than that of pure g-C 3 N 4 and Bi 2 WO 6 , respectively. It should be noted that too much Bi 2 WO 6 content in composite photocatalyst (such as 2 : 8 g-C 3 N 4 -Bi 2 WO 6 ) would distinctly reduce the photocatalytic activity of the composite. Previous research also pointed out that the dispersion and size of deposited nanoparticles could influence their photocatalytic efficiency [34,35]. On g-C 3 N 4 nanosheets, Bi 2 WO 6 nanoparticles with smaller size and favourable dispersibility meant higher photocatalytic activity. The smaller size and higher dispersibility of Bi 2 WO 6 on the g-C 3 N 4 sheets meant higher photocatalytic activity. But if the loading density of Bi 2 WO 6 nanoparticle was too high, the photocatalytic site on the g-C 3 N 4 sheet surface will be covered, which would damage the heterojunction structure and reduce synergistic effect between the two components [36]. As a result, only Bi 2 WO 6 nanoparticles coated on g-C 3 N 4 with proper size and  dispersion would enhance the photocatalytic activity of composite photocatalyst. In this work, it was found the 8 : 2 g-C 3 N 4 -Bi 2 WO 6 composite sample displayed the highest catalytic performance due to the optimal structure. The kinetics of photocatalytic degradation of methyl orange reflects the reaction rate of the photocatalyst. The change of methyl orange concentration can be obtained by the first-order kinetic equation ln (C 0 /C t ) = k t to obtain the corresponding apparent degradation rate constant (k) T, where C 0 and C are the concentrations of pollutant in solution at time t 0 and t, respectively, and k is the apparent first-order rate constant [37]. As shown in figure 7a, all fitting curves of t for ln (C/C 0 ) are approximated by a straight line, and thus the corresponding kinetic constants (k) are calculated. Observably, the rate constant k is calculated to be 0.00999, 0.01104, 0.0166, 0.02659 and 0.00313 min −1 for pure g-C 3 N 4 , 2 : 8 g-C 3 N 4 -Bi 2 WO 6 , 5 : 5 g-C 3 N 4 -Bi 2 WO 6 , 8 : 2 g-C 3 N 4 -Bi 2 WO 6 and pure Bi 2 WO 6 samples, respectively (figure 7b). That is, 8 : 2 g-C 3 N 4 -Bi 2 WO 6 catalyst has the best photocatalytic activity, the degradation rate of methyl orange is almost 2.7 and 8.5 times higher than that of either individual g-C 3 N 4 or Bi 2 WO 6 . The above results show that the introduction of Bi 2 WO 6 could effectively enhance the visible light photocatalytic activity of g-C 3 N 4 Apart from the photocatalytic performance, the stability of photocatalysts in the practical application also has a very important significance. The g-C 3 N 4 -Bi 2 WO 6 composite photocatalyst was circulated four times under the same conditions. After every cycle, the sample was centrifuged, washed and dried at 60°C. After four cycles, the degradation rate of methyl orange was still 84.26% (figure 8), which shows a high photocatalytic stability of the composite after four recycling runs. Thus, the stability and recyclability of g-C 3 N 4 -Bi 2 WO 6 composites are excellent.  Figure 9. Band structure schematic of g-C 3 N 4 -Bi 2 WO 6 heterojunction and possibly occurring reaction mechanism of MO on the surface.

Possible photocatalytic mechanism
It can be seen from the photocatalytic experiments that the novel g-C 3 N 4 -Bi 2 WO 6 composite has shown a good photocatalytic effect for degrading the MO under visible light. On the basis of the above experimental results, a possible mechanism for the g-C 3 N 4 -Bi 2 WO 6 composite photocatalyst is proposed. The process of electron-hole separation and transport in interface is shown in figure 9.
It is known from the literature that the band-gap positions of g-C 3 N 4 were determined at −1. 13 and +1.58 eV [4], while those of Bi 2 WO 6 were estimated at 0.46 and +3.26 eV. Once the composite photocatalyst is irradiated by visible light, both g-C 3 N 4 and Bi 2 WO 6 can be stimulated and generate photogenerated electron-hole pairs. Because it has a well-matched staggered band-gap structure and a close interface, the electrons on the conduction band of g-C 3 N 4 will be transferred to the conduction band of Bi 2 WO 6 . At the same time, holes on the VB of Bi 2 WO 6 reversely transfer to the VB of g-C 3 N 4 . This process effectively facilitates the separation of photogenerated electron-holes, thereby inhibiting the recombination of photogenerated electron-hole pairs, which could improve the photogenerated electronhole pair separation effectively and decrease the possibility of photogenerated charge recombination greatly. Therefore, Bi 2 WO 6 nanoparticles and g-C 3 N 4 nanosheet in the g-C 3 N 4 -Bi 2 WO 6 complex material could form the optimal heterojunction structures, which result in the improvement of the photocatalytic activity.
As we know, photogenerated holes, ·OH radicals and ·O 2 − are the three main major active species in the degradation process [13,38]. Therefore, to study the role of these reactive groups in the photocatalytic process, a series of free radicals capture experiments were conducted by using ethylenediaminetetraacetate (EDTA, 6 mmol l −1 ), benzoquinone (BQ, 0.5 mmol l −1 ) and tert-butanol (TBA, 6 mmol l −1 ) as effective scavengers for holes, ·O 2 − and ·OH radicals respectively [39]. In the free radicals capture experiments (figure 10a), when TBA was added to the system, the degradation efficiency of MO was reduced from 95.89% to 80.53% compared with the photocatalytic experiment without the capture agent, indicating that the photogenerated holes are not the primary active group for the degradation of methyl orange, and they are also not the sources of ·OH radicals. It is worth noting that the degradation of MO is slightly enhanced when the hole trapping agent EDTA was added, which means that the hole is not the active group for degrading the MO, and the reason for the increase in catalytic activity is that EDTA is a capture agent, so that the photogenerated electron-hole could be effectively separated. However, when BO was added, the degradation rate of MO was only 45.02%, which demonstrates that the main active species should be ·O 2 − in the degradation of MO [40,41].
For further insight into photogenerated electron-hole pair behaviour of the g-C 3 N 4 -Bi 2 WO 6 composite and to verify the above-mentioned mechanism, PL spectra of the as-prepared photocatalysts were obtained. Generally, in semiconductor materials, the migration and separation of photogenerated carriers lead to the generation of fluorescence spectra [42,43]. Figure 10b presents the PL spectra of pure g-C 3 N 4 , pure Bi 2 WO 6 and 8 : 2 g-C 3 N 4 -Bi 2 WO 6 composite samples excited with a 365 nm light. At room temperature, the luminescent characteristic peaks of pure g-C 3 N 4 are centred at 440 nm, which is attributed to the radiation recombination process of self-trapped excitations [44]. Compared with those of pure g-C 3 N 4, the position of the emission peaks of the 8 : 2 g-C 3 N 4 -Bi 2 WO 6 sample was almost unchanged, but the intensity was greatly reduced, which indicates that the photogenerated charges recombination rate was controlled in the heterojunction semiconductors. The PL results support the above discussion on the photocatalytic experiments and proposed mechanism strongly. Over all, the PL spectra of the as-synthesized g-C 3 N 4 , Bi 2 WO 6 and 8 : 2 g-C 3 N 4 -Bi 2 WO 6 photocatalysts at room temperature. g-C 3 N 4 -Bi 2 WO 6 composite photocatalysts can effectively separate photogenerated electron-hole pairs and inhibit recombination of the charges, which has a very promising application in environmental purification.

Conclusion
In conclusion, the g-C 3 N 4 -Bi 2 WO 6 composite photocatalyst with a heterojunction structure was prepared by a simple solvothermal method. This novel g-C 3 N 4 -Bi 2 WO 6 composite catalyst exhibits outstanding visible light response photocatalysis, attributing to the effective separation of electron-hole pair at heterojunction interfaces. Among the as-prepared various weight ratios of g-C 3 N 4 -Bi 2 WO 6 samples, the 8 : 2 g-C 3 N 4 -Bi 2 WO 6 sample displayed best photocatalytic activity because of its optimum structure. The investigation of the photocatalytic mechanism showed that the degradation of methyl orange by the g-C 3 N 4 -Bi 2 WO 6 sample is mainly through ·O 2 − radicals, and ·OH was verified to be inappreciable.
We believe this work provides some new insights for fabrication of highly efficient heterojunction photocatalysts for environmental remediation.