Extensive researches have been done on improving photocatalytic properties of materials in recent years [1-2]. Photocatalytic degradation of organic compounds can be done by some of the semiconductors in order to purify wastewater from the industries and households [3-9]. Most researches were focused on the metal oxides such as Ti, Nb, Ta, In and etc. [10-12]. It was recently reported that metal tungstates possess special physicochemical properties due to their interesting self-trapped excitations . Metal tungstates were found to have a good photocatalytic properties under ultraviolet (UV) and visible light. It has been reported that ZnWO4 is an efficient catalyst for dye degradation [14-15]. Therefore, significant attention has been paid to metal tungstates for their unique luminescence and structural properties . One of the most practical tungstates is cadmium tungstate (CdWO4). CdWO4 with a monoclinic wolframite structure has high average refractive index, thermal stability, high density (7.9 gr/cm3), low afterglow, low radiation damage and high X-ray absorption coefficient . At room temperature, CdWO4 with photoluminescence (PL) peak about 460 nm has been used as an X-ray scintillator . The high efficiency, short decay time, high chemical stability, and high stopping power are its advantages as a scintillator material. It has also a promising application as an advanced medical X-ray detector in computerized tomography . The preparation of CdWO4 nanorods and nanofibers for improving the properties of CdWO4 through hydrothermal synthesis has been reported by many groups [20-22]. Wang et al. had synthesized CdWO4 nanorods by hydrothermal method. They reported a good luminescence with a PL peak at 435 nm . Ye et al. have found that the CdWO4 photocatalytic activities are comparable to the known ZnWO4 and TiO2 semiconductors under UV and visible light irradiation . CdWO4 has synthesized by various methods such as co-precipitation , sol–gel , reverse-micelle , solvothermal , solid-state metatheticreaction , spray pyrolysis , molten salt , and hydrothermal [20-22,30].
Hydrothermal method offers many advantages, such as mild synthesis conditions, high degree of crystallinity, high purity, and narrow particle size distribution of product. These advantages made this method as a very popular synthesis method. A lot of works have been done to enhance the photocatalytic properties of CdWO4. Ye et al. reported that CdWO4 doped with Eu+3 have better photocatalytic degradation of methyl orange (MO) than pure CdWO4 . Recently, Xu et al. used reduced graphene oxide (RGO) to enhance the photocatalytic properties of CdWO4 . They used different weight percentages of RGO and found that the CdWO4-RGO with 2% RGO has the best results.
In this work, we have synthesized CdWO4 with 2% RGO by hydrothermal route at different reaction temperatures as 140, 160 and 180oC. The photocatalytic activities of as-prepared samples were investigated by degradation of methylene blue (MB) under the visible light irradiation.
MATERIALS AND METHOD
The synthesis of CdWO4 and CdWO4-RGO have been done by hydrothermal method. For preparing CdWO4, 2.198 g of Na2WO4.2H2O and 2.056 g of Cd(NO3)2.4H2O were dissolved in 30 and 20 ml of distilled water individually. Afterward, the Na2WO4.2H2O was added drop wise into cadmium nitrate solution. For preparing of CdWO4-RGO, 0.03 ml of RGO solution (1 mg/ml)  added to the CdWO4 solution and stringed for 20 min. Finally, suspensions were sealed into a Teflon-lined autoclave and maintained for 12 h at 180 °C for pure CdWO4 and at 140, 160 and 180°C for CdWO4-RGO samples named as CdWO4-RGO140, CdWO4-RGO160 and CdWO4-RGO180, respectively. Subsequently, the autoclave was cooled to room temperature naturally. The resulting samples were washed several times with distilled water and finally with ethanol in order to remove the impurities and dried at 60°C in a vacuum oven for 24 h.
The crystalline structure of the prepared CdWO4 and CdWO4-RGO samples were analyzed by using an X-ray diffraction (XRD) (Philips-PW 1800 X-ray powder diffractometer) method with monochromatized Cu-Kα radiation (λ = 1.541874 Å) in the range of 2θ )20–40° (. The morphology of the nanostructures was depicted by a scanning electronic microscope (SEM). The SEM samples had been coated by a gold thin film using a desktop sputtering system (Nanostructured Coating Co.-Iran). Fourier transform infrared (FT-IR) spectra were recorded in the range of 400–4000 cm−1 using Nicolet spectrometer. Room temperature photoluminescence (PL) spectra were taken on a Perkin-Elmer LS55 equipped with a 450 W Xe lamp as an excitation source. Raman spectra were analyzed on a microscopic Raman spectrometer (SENTERRA 2009, Bruker). The concentration of MB in the irradiation process was analyzed by using a UV-visible spectrophotometer (Shimadzu, MPC-2200).
In order to investigate the photocatalytic activity of the as prepared CdWO4 and CdWO4–RGO samples, degradation experiments of MB were performed under visible light. For this analyze, 0.01 g of samples were dispersed into a 10 mL MB solution (MB concentration: 10 mgl−1) and then irradiated by visible light (400 W tungsten lamp) under continuous stirring. Before the irradiation, the suspension was maintained in the dark for 3 h to reach the complete adsorption–desorption equilibrium. All photocatalytic experiments were accomplished at the same conditions. The distance between the surface of the MB solution and the light sources were about 20 cm. The photocatalytic performance was indirectly monitored by relating the optical absorbance to the MB degradation amount using a double beam UV–visible spectrophotometer at a wavelength of 664 nm. The blank experiment without catalyst was also investigated, and its value with less than 1% degradation after 180 min illumination can be neglected.
RESULTS AND DISCUSSION
Fig. 1 shows the XRD patterns of the CdWO4 and CdWO4-RGO nanocomposites. All XRD peaks for the samples were indexed by monoclinic wolframite phase of CdWO4 which is well consistent with the reported data (JCPDS, card No. 14-0676). The distinctive peaks for all samples centered at 2θ: 23.5, 29.1, 29.7, 30.7, 35.5 and 35.7o which are matched well with (110), (−111), (111), (020), (002), and (200) crystal planes of CdWO4, respectively, which is consistent with the results of Ye et al. . Lattice constants were obtained as a =5.15, b = 5.8, c = 5.13 Å and β=94.47o for all samples. The peak broadening in the XRD patterns clearly indicates that small nanocrystals are present in the samples. The results also show that the addition of RGO has no distinct effect on the crystal structure of CdWO4 as reported in the Ref.  but XRD pattern of the pure CdWO4 has sharper peaks in comparison with the CdWO4-RGO nanocomposites. In order to calculate micro strains and grain size according to the Williamson-Hall equation , the diffraction line broadening is due to crystallite size (D) and lattice strain (ε):
where λ is the X-ray wavelength (1.54 Å), k is the Scherrer constant (0.9), and θ is the Bragg angle. The Williamson-Hall plot is presented for CdWO4-RGO140 in Fig. 2. The slope of the fitted line is attributed to lattice microstrains. The positive slope of the fitted line shows that a tensile microstrain is induced in the crystal lattice of the sample. Table 1 shows grain sizes and lattice microstrains for all samples. CdWO4-RGO160 has the highest lattice tensile strain. The induced lattice strain is commonly originated from defects and lattice distortions in the crystals . The small value for the pure CdWO4 is due to absence of graphene.
The FT-IR spectra of CdWO4 and CdWO4- RGO samples were measured in the wavenumber region of 400-4000 cm-1 and are shown in Fig. 3. A symmetrical stretching vibrations of W-O-W bond in [WO4]2- is represented by the band at 725 and 890 cm-1 . The bending and stretching vibrations of Cd-O (580 cm-1) and W-O (820 cm-1) were also identified in the pure CdWO4 and CdWO4-RGO nanocomposites. The strong band centered at 1384 cm-1 is assigned to the symmetrical and asymmetrical vibrations of C-O bond. This peak is considerable in the nanocomposites that indicates presence of graphene oxide in the samples. By increasing the reaction temperature from 140 to 180°C, this peak is decreased which can be related to reduction of graphene oxides. The broad absorption band centered at 3430 cm-1 is attributed to the vibration of H-O bonds for surface hydration layers that is higher in the nanocomposite samples than the pure CdWO4. The weak absorption located at 1635 cm-1 is associated with the deformation vibration of H-O-H bonds of water molecules, and centered peak at 2360 cm-1 related to C-H stretching vibration [36,37].
Fig. 4 shows morphology of the different samples by SEM. It can be seen in the Fig. 4 (a) that the pure CdWO4 sample has nanorods structures and these nanorods are distributed homogenously. From the images of nanocomposite samples in the Figs. 4(b)-(d), we can see CdWO4 nanorods which have grown among RGO sheets. As shown in Fig. 4 (a), CdWO4 nanorods in the pure sample have 50-100 nm diameter and ~1 µm length. By adding graphene to the nanocomposite samples, these dimensions are considerably decreased.
Fig. 5 shows TEM image of the pure CdWO4 sample. The obtained CdWO4 nanostructures appear to have uniform rod-like morphologies with 30–100 nm width. This result is corresponded with SEM image.
Raman spectroscopy is widely employed to study the ordered/disordered crystal structures of carbonaceous materials. According to the Fig. 6(a), Raman spectrum of the pure CdWO4 has only one strong vibration at 896 cm-1 and several weak vibrations at 776, 705, 686 and 548 cm-1. The vibration modes located at 896 cm-1 are correspond to the normal W–O vibrations of the WO6 octahedral, while the modes located at 776 and 686 cm-1 involve motions of WO6 octahedral against the Cd2+ [38,39]. Bands in the range of 500–600 cm-1 are characteristic of symmetric W–O–W stretching modes . Therefore, the Raman analysis confirmed that monoclinic CdWO4 were successfully prepared. In Fig. 6(b), we can also see the Raman spectrum of CdWO4-RGO160. The result is similar to the pure CdWO4 but the intensity of peaks are declined. In addition, CdWO4-RGO160 exhibited D line at 1318 cm−1 and G line at 1591 cm−1. ID/IG (intensity ratio of D and G lines) is 1.46.
Fig. 7 shows the PL spectra of the pure CdWO4 and CdWO4-RGO nanocomposites. The obtained nanorods exhibit a very strong emission band in the range of 400–500 nm. The emission peak is at 450 nm for the pure CdWO4. Moreover, it is at 451, 454 and 453 nm for CdWO4-RGO140, CdWO4-RGO160 and CdWO4-RGO180, respectively. This PL emission is caused by the 1A1→ 3T1 transitions within the WO66- complex . When RGO was added to CdWO4, the photoemission intensity of the samples fell down due to quenching of photoemission. It can be seen from the figure that the highest intensity is related to the pure CdWO4 and the lowest intensity is related to CdWO4-RGO160. This indicates that photoelectron transfer from CdWO4 to RGO in CdWO4-RGO160 is more effective than others which impedes the recombination of photo induced electrons and holes.
As a main characterization of prepared samples, the photocatalytic properties of the pure CdWO4 and CdWO4-RGO nanocomposites have been investigated and compared with each other. MB was considered as a model dye to evaluate the photocatalytic degradation performance of the samples. The photocatalytic activities of the as-prepared samples were investigated under visible light irradiation for 180 min to degradation of MB. The results are shown in the Fig. 8. Based on this figure, the activity of CdWO4-RGO nanocomposites are better than the pure CdWO4 and especially CdWO4-RGO160 has the highest activity under the visible light irradiation. This activity can be related to the retardation of the recombination time in this sample which confirmed by PL results. By monitoring the MB absorption peak at 665 nm, the plots of the degradation percentage vs. reaction time were obtained for the pure CdWO4 and CdWO4-RGO nanocomposites under visible light irradiation. The MB degradation percentage was calculated by the following equation:
Degradation (%) = (2)
where c0 and c are the initial and final concentrations of MB, respectively. This result is shown for CdWO4-RGO160 in Fig. 9(a).
The kinetics of these degradation reactions were investigated, too. In order to calculate kinetics of degradation reactions,  relation was used, where c0 is the initial concentration of a pollutant, c is MB concentration at time t, and k is the rate constant of the first-order reaction. The amount of k for CdWO4-RGO160 can be obtained by determining the slope of ln c/c0 vs. time (Fig. 9(b)). Table 2 shows the rate constants of photodegradation for all samples of current work as compared to other works.
The charge separation and transfer in the reaction system of the nanocomposites are shown in Fig. 10. Under visible light irradiation, the excited electrons jump from the valence band (VB) of CdWO4 nanorods to the conduction band (CB), and diffuse to the interface of CdWO4 and RGO. With a good conductivity, RGO could be a good acceptor of photo-induced electrons. Hence, the recombination would be effectively suppressed and it has been confirmed by the PL results. As a result of this effect, much more holes (h+) could diffuse to the surface of CdWO4 and cause the formation of hydroxyl radicals (•OH), which subsequently degrade MB. Meanwhile, the transferred electrons might react with the dissolved oxygen to form the superoxide radicals (•O2), which would also react with the dye molecules.
The hydrothermal method as an easy and feasible method was used for synthesizing the pure CdWO4 and CdWO4-RGO nanocomposites. Reaction hydrothermal temperature as 140, 160 and 1800C have been investigated for CdWO4-RGO nanocomposites. Synthesized samples contained CdWO4 nanorods with lower than 100 nm diameter and 1 µm length. Raman spectra proved the presence of RGO in the nanocomposites. The photocatalysis experiment results showed that the CdWO4-RGO nanocomposites have higher photocatalytic activity than the pure CdWO4 sample for MB degradation. Among the nanocomposites, CdWO4-RGO with reaction temperature of 160 oC had the best photocatalytic degradation effect and the lowest PL intensity due to the longest recombination time of photo-induced electrons and holes.
Authors would like to thank from Dr. S. Safa for his assistance and guidance in this research.
CONFLICT OF INTEREST
The authors declare that there are no conflicts of interest regarding the publication of this manuscript.