Developing the Ternary ZnO Doped MoS2 Nanostructures Grafted on CNT and Reduced Graphene Oxide (RGO) for Photocatalytic Degradation of Aniline

Transition metal sulfide semiconductors have achieved significant attention in the field of photocatalysis and degradation of pollutants. MoS2 with a two dimensional (2D) layered structure, a narrow bandgap and the ability of getting excited while being exposed to visible light, has demonstrated great potential in visible-light-driven photocatalysts. However, it possesses fast-paced recombination of charges. In this study, the coupled MoS2 nanosheets were synthesized with ZnO nanorods to develop the heterojunctions photocatalyst in order to obtain superior photoactivity. The charge transfer in this composite is not adequate to achieve desirable activity. Therefore, heterojunction was modified by reduced graphene oxide (RGO) nanosheets and carbon nanotubes (CNTs) to develop the RGO/ZnO/MoS2 and CNTs/ZnO/MoS2 ternary nanocomposites. The structure, morphology, composition, optical and photocatalytic properties of the as-fabricated samples were characterized through X-ray diffraction (XRD), Fourier Transform Infrared (FTIR), Field Emission Scanning Electron Microscopy (FESEM), Transmission Electron Microscopy (TEM), Energy-Dispersive X-ray (EDX), elemental mapping, Photoluminescence (PL), Ultraviolet–Visible spectroscopy (UV-VIS), and Brunauer-Emmett-Teller (BET) techniques. The photo-catalytic performance of all samples was evaluated through photodegradation of aniline in aqueous solution. The combination of RGO or CNTs into the ZnO/MoS2 greatly promoted the catalytic activity. However, the resulting RGO/ZnO/MoS2 ternary nanocomposites showed appreciably increased catalytic performance, faster than that of CNTs/ZnO/MoS2. Charge carrier transfer studies, the BET surface area analysis, and the optical studies confirmed this superiority. The role of operational variables namely, solution pH, catalyst dosage amount, and initial concentration of aniline was then investigated for obtaining maximum degradation. Complete degradation was observed, in the case of pH = 4, catalyst dosage of 0.7 g/L and aniline concentration of 80 ppm, and light intensity of 100 W. According to the results of trapping experiments, hydroxyl radical was found to be the main active species in the photocatalytic reaction. Meanwhile, a plausible mechanism was proposed for describing the degradation of aniline upon ternary composite. Moreover, the catalyst showed excellent reusability and stability after five consecutive cycles due to the synergistic effect between its components. Total-Organic-Carbon concentration (TOC) results suggested that complete mineralization of aniline occurred after 210 min of irradiation. Finally, a real petrochemical wastewater sample was evaluated for testing the catalytic ability of the as-fabricated composites in real case studies and it was observed that the process successfully quenched 100% and 93% of Chemical Oxygen Demand (COD) and TOC in the wastewater, respectively.

Wastewater sampling method. A real petrochemical wastewater sample was prepared from the effluent of a petrochemical company, in Iran. The sample was kept in a 10 L container at 4 °C 53 .
Synthesis of MoS 2 nanosheets. For preparation of MoS 2 nanosheets, 1.9 g of thiourea, as a source of sulfur content was dissolved in 80 mL of DI water. Simultaneously, 1.03 g of sodium molybdate was added to this solution accompanied by constant stirring. Next, the mixture was poured into a 100 mL Teflon-sealed autoclave and was heated for 15 hour at 200 °C. The black product was filtered and was washed several times using ethanol and DI water. Finally, the sample was annealed for 2 hour at 700 °C to improve its crystallinity 34 . Synthesis of Zno nanorods. In order to develop ZnO nanorods, 2.195 g of Zn(CH 3 COOH) 2 was dissolved in 80 mL of ethanol. Then, 0.8 g of NaOH was added to the solution and was stirred for 20 min. Next, the solution was poured into a 100 mL Teflon-sealed autoclave and was heated at 150 °C for 12 hour. After the reaction, ZnO nanorods were collected through centrifugation and were washed using ethanol and water. In order to improve the crystallinity and to obtain the Wurtzite phase, the sample was annealed at 700 °C for 2 hour 53,54 . Synthesis of binary and ternary samples. CNT/ZnO/MoS 2 nanocomposite was synthesized by a typical hydrothermal method. First, a certain amount of CNTs (0.07, 0.1 and 0.13 gr CNTs) was dispersed into a mixture of ethanol (40 mL) and water (40 mL) through ultrasonic treatment at 60 watts for 3 hour. Then, 0.8 g of MoS 2 and 0.2 g of ZnO were added to the above-mentioned solution and were sonicated again for 1 hour. The resulting solution was put into a 100 mL hydrothermal autoclave and was kept at 250 °C for 15 h. Then, the prepared composite was centrifuged and was washed several times using distilled water and was dried at 60 °C overnight. Finally, the nanocomposite was annealed for 2 hour at 300 °C for obtaining intimate contact between the components. Similar procedure was carried out to prepare samples of RGO/ZnO/MoS 2 nanocomposite. Also, the binary nanocomposite (ZnO/MoS 2 ) was prepared without using carbonaceous materials 55 . characterization. X-ray diffraction (XRD) experiment was conducted using PHILIPS PW1730 (Netherland) Diffractometer with Cu K α radiation in scanning range of 2θ = 5°-70°. A THERMO AVATAR (USA) Fourier Transform Infrared (FTIR) spectrometer was utilized to obtain the spectra with the wavelength range of 400-4000 cm −1 . The surface morphology of the samples was evaluated using Field Emission Scanning Electron Microscopy (FESEM-FEI MOVA NANOSEM 450, USA). The Transmission Electron Microscopy (TEM) was obtained by Zeiss EM 900 apparatus. Energy-Dispersive X-ray (EDX) spectroscope and elemental mapping were applied in order to determine the composition of the samples. The Barrett-Joyner-Halenda (BJH) method was applied to determine the pore size and Brunauer-Emmett-Teller (BET) method was employed to assess the specific surface area based on the nitrogen adsorption/desorption isotherms using BELSORP MINI II (Japan) instrument. All photocatalysts samples were degassed at 120 °C for 2 hour before the analysis. Photoluminescence (PL) spectra were obtained using AvaSpec-2048 apparatus. UV-vis spectra of the samples were analyzed by an Avantes (AvaSpec-2048-TEC) spectrophotometer in the range of 200-1200 nm. The mineralization was studied through determination of Total-Organic-Carbon-concentration (TOC) using TOC-VCSH Shimadzu analyzer, Japan. The components of real wastewater sample prepared from the effluent of the petrochemical company were identified using Agilent 7890 GC-MS, USA. COD determination was done using open-reflux method, based on the available standard methods for water and wastewater examination. experimental procedure. Aniline was applied as a study model to measure the photocatalytic performance of samples at ambient temperature. First, all the as-fabricated samples were evaluated under the same condition in order to select the best photocatalyst in terms of photocatalytic performance. Initially, 1.2 g of the nanocomposite was added into 100 mL of 100 ppm aniline solution at pH = 7. Prior to the reaction, the solution was vigorously stirred in dark to establish the adsorption-desorption equilibrium time between the photocatalyst and aniline solution. Then, the experiment was initiated under the exposure to light for 120 min using two 50 W LED lamps, as a source of visible light and they were placed 10 cm away from the reaction media. During the reaction, 2 mL of aniline solution was taken at a regular interval of 15 min using a syringe and after centrifugation, the variation in the aniline concentration was monitored using UV-vis spectrophotometer (Unico, USA) at maximum absorbance of 230 nm (λ max = 230 nm). The results showed that RGO10%/ZnO/MoS 2 had the highest photocatalytic performance among all samples, therefore; it was chosen for performing the optimization of the operational parameters. Finally, the effect of operational parameters such as pH, the catalyst dosage amount, and aniline concentration was evaluated during the course of study. The value for removal and total mineralization of aniline in the photocatalytic degradation process was calculated by the following equations 56 : where C 0 represents initial concentration of aniline, and C t represents the concentration of aniline at time t. TOC 0 represents the initial TOC value and TOC t represents the value of TOC solution at time t.
The pH ZPC of the as-fabricated sample was calculated by adding 0.1 g of catalyst sample to 25 mL of NaCl solution. Then, the pH of the solution was adjusted by adding specific amounts of HCl or NaOH to reach to the pH in the range of 2-12. After that, the solutions with different pH values were shaken constantly for 48 hour. Finally, the pH ZPC of the catalyst was obtained similar to the initial pH value.

Results and discussion
characterization of nanocomposites. In order to characterize the crystallinity and phase purity of MoS 2 , ZnO20%/MoS 2 , RGO10%/ZnO/MoS 2 and CNTs10%/ZnO/MoS 2 nanocomposite samples, their XRD patterns were evaluated. Figure 1 shows the XRD patterns related to the as-fabricated samples. Figure 1 32 . The abovementioned characteristic peaks of MoS 2 were also observed in the XRD pattern for binary and ternary nanocomposites ( Fig. 1(b-d)). Figure 1 37 . In the XRD pattern for RGO/ZnO/MoS 2 nanocomposite, the sharp diffraction peak of GO at 2θ = 10° did not exist, however; a very weak diffraction peak emerged at 2θ = 26° which is attributed to (002) reflections of RGO. This implies that GO sheets were transformed into RGO after hydrothermal reduction ( Fig. 1(c)). Figure 1(d) shows that there is not any obvious peak related to CNT. This could be due to the low content and amorphous property of CNT. A broad noticeable diffraction peak was observed at 2θ = 14.49° with interlayer space of about 0.61 nm indicating that the samples might compose of a few layers of MoS 2 55 . In addition, the crystallite sizes of the samples were measured by the Scherrer's equation. The mean crystallite size of the MoS 2 , ZnO20%/MoS 2 , CNT10%/ZnO/MoS 2 , and RGO10%/ZnO/MoS 2 samples were found to be equal to 66 nm, 64 nm, 54 nm, and 53 nm, respectively at the highest peak for (002) plane. It can be concluded that, carbonaceous materials caused prevention in the growth of the crystallinity for the MoS 2 nanosheets.
FTIR was conducted to determine the functional groups in the samples (Fig. 2). The results revealed that MoS 2 showed the following absorption peaks at 470 cm −1 , 1100 cm −1 , 1640 cm −1 , and 3436 cm −1 . The absorption peaks at 470 cm −1 and 1640 cm −1 are assigned to the presence of Mo-S and Mo-O bands in the sample, respectively. The peak at 1100 cm −1 and 3436 cm −1 were formed as a result of the hydroxyl stretching vibration due to the absorbed water molecules. All these characteristic peaks also existed in the binary and ternary samples. Moreover, in the case of ZnO20%/MoS 2 , the peak of Mo-S bond at 470 cm −1 had overlap with the Zn-O peak (at 500 cm −1 ) resulted in formation of a large peak at 470 cm −1 . In the spectra of CNT10%/ZnO20%/MoS 2 , the peaks at 2700 cm −1 and www.nature.com/scientificreports www.nature.com/scientificreports/ 1720 cm −1 were formed due to the vibration of -COOH groups and C=O stretching, respectively. Also, in the spectra of RGO10%/ZnO20%/MoS 2 , the peaks appearing at 2854 cm −1 and 2922 cm −1 were formed due to the C-H stretching vibration, and the peak centered at 1726 cm −1 is corresponded to the C=O stretching frequency from the residual carboxyl groups of RGO 34,37,[57][58][59] .
The FESEM images of the as-fabricated samples along with their corresponding EDX results are presented in Fig. 3(a-d). Figure 3(a) clearly shows the formation of large MoS 2 nanosheets. As mentioned before, the strong peak at 2θ = 14.4 suggests that the fabricated MoS 2 is few-layered according to the FESEM image of MoS 2. The average thickness of each sheet is about 11 nm. Figure 3(b) displays the FESEM image of ZnO20%/MoS 2 nanocomposite. ZnO particles mainly with a rod-like form and hexagonal shape were compactly deposited on the surface of MoS 2 nanosheets while random dispersion was also observed. For better evaluation of the size and morphology of ZnO particles, at the Fig. 3(e,f) the TEM image is provided which confirmed the FESEM result and indicate that rod-like ZnO nanostructures are distributed on the surface of MoS 2 nanosheets. It can be seen that the length and diameter of ZnO nanorods are about 100-200 nm and 50-75 nm, respectively. For the sample of CNTs10%/ZnO20%/MoS 2 , CNTs with an average diameter of 20 nm obviously adhered to the ZnO20%/MoS 2 material which is considered to be beneficial for efficient charge transfer. Finally, in the RGO10%/ZnO20%/MoS 2 structure, after incorporation of RGO nanosheets, the morphology of the MoS 2 nanosheets and ZnO nanorods was not changed. Instead, they were embedded into the MoS 2 nanosheets, leading to provision of enlarged surface area and improvement of electrical performance for the ternary sample. The results of EDX analysis only showed the peaks for Mo, S, Zn, O, and carbon elements, which confirmed the successful formation of all nanocomposite samples and high elemental purity of them. Additionally, elemental mapping was used for further identification of elemental content and distribution of CNTs/ZnO/MoS 2 and RGO/ZnO/MoS 2 nanocomposite samples (Fig. 4). Homogeneous distribution of Mo, Zn, C, and O was observed in the ternary composites, indicating well-organized structure with desirable interaction.
The adsorption-desorption isotherm of samples ( Fig. 5A), surface area, BJH results ( Fig. 5B) and pore sizes ( Table 1) were calculated to investigate the effect of ZnO and carbonaceous materials on the enhancement of surface area for the MoS 2 . Figure 5A shows that pure MoS 2 and ZnO20%/MoS 2 both showed similar type V isotherms with H 3 type hysteresis loop, while the ZnO/MoS 2 -carbon materials showed type IV isotherms. These findings indicated that the pore volumes are supplied by mesopores in all samples providing efficient channels for mass transport and enhancing the interface contacts between catalyst surface and pollutant. Pure MoS 2 had a surface area of 59.94 m 2 g −1 and the addition of ZnO nanorods resulted in a slight rise in the surface area, yielding a value of 60.95 m 2 g −1 for ZnO20%/MoS 2 . After modification using carbonaceous materials, the BET surface area reached 344.97 m 2 g −1 and 420.75 m 2 g −1 for CNT10%/ZnO20%/MoS 2 and RGO10%/ZNO20%/MoS 2 , respectively, indicating that carbonaceous materials especially RGO are considered to be excellent support material for ZnO/MoS 2 . The pore size studies revealed that the average pore size of ternary composites was lower than that of pure MoS 2 and binary composite. However, the extended surface area of the ternary photocatalysts is still capable to supply more active sites for the adsorption of aniline and enhancing the photocatalytic performance 52 .
The UV-vis analysis of as-fabricated photocatalysts was performed to investigate the optical properties. Figure 6(A) shows the UV-vis adsorption spectra for MoS 2 , ZnO20%/MoS 2 , CNT10%/ZnO20%/MoS 2 and www.nature.com/scientificreports www.nature.com/scientificreports/ RGO10%/ZNO20%/MoS 2 in the range of 190-800 nm. These results are attributed to the color of composite influencing the adsorption capability of the composite powders. Incorporation of ZnO into MoS 2 catalyst caused a decrease in the cut-off wavelength which has led to higher bandgap value due to color conversion from black to gray for MoS 2 and ZnO20%/MoS 2 catalysts, respectively. However, adding graphene and CNT to the final composite caused an increase in the cut-off wavelength. Moreover, the adsorption edge of RGO10%/ZNO20%/ MoS 2 and CNT10%/ZnO20%/MoS 2 composites showed a red-shift near the wavenumbers of bare MoS 2 , which is attributed to successful interaction between graphene nanosheets and CNT layers. Hence, the presence of RGO or  www.nature.com/scientificreports www.nature.com/scientificreports/ in Table 2. The results showed that the bandgap value for photocatalyst containing graphene was higher than that of photocatalyst containing CNT. The obtained results can be attributed to the high charge migration rate between graphene layers and semiconductors and the potential charactristic of graphene in light adsorption 60 .
The efficient separation of photo-induced electron-hole pairs is considered as a prominent factor for an ideal photocatalyst. Therefore, the photoluminescence (PL) emission spectra formed as a result of the recombination  www.nature.com/scientificreports www.nature.com/scientificreports/ of photo-excited charge carriers on the surface of the semiconductor was investigated for further study of the migration, separation, and recombination of carriers. The PL spectrum of the bare MoS 2 , binary, and ternary composites are illustrated in Fig. 6(B). Bare MoS 2 showed an intense PL emission centered at around 388 nm with an excitation wavelength of 340 nm. The PL spectra for other samples were similar to that of bare MoS 2 , but the presence of ZnO, CNT, and RGO caused to the reduction in their PL intensity. This reduction was found to be significant when the carbonaceous materials were incorporated. This could happen since, carbonaceous materials possess excellent electro-conductivity and high electron storage capacity, therefore; the migration of photo-excited electrons could be accelerated from the conduction band of MoS 2 to the surface of these materials, and therefore, the recombination of charges would be prevented. RGO10%/ZnO20%/MoS 2 had lower recombination rate compared to CNT10%/ZnO20%/MoS 2 which is accounted to be better for photocatalytic reaction 60,61 .   In order to investigate the effect of ZnO substrate, different amounts of ZnO loading (0, 10, 20, and 40 wt.%) were used (Fig. 7A). Furthermore, to investigate the structural stability and photosensitization of aniline under the visible light irradiation, a blank test was carried out in a case without catalyst involvement. Figure 7(A) shows that no obvious self-degradation was observed in the absence of a catalyst, suggesting that aniline is persistent and cannot be degraded under visible light. As expected, in the presence of the catalyst, the single-component had an inadequate degradation rate by 35% which is due to the insufficient absorption sites and fast recombination of charges. The degradation was further improved by loading ZnO content. Increasing the amount of ZnO from 10% to 20% caused an increase in the degradation efficiency from 44% to 58%. However, when the content of ZnO increased to 40%, the catalytic performance decreased to 52%. The reduction in the degradation efficiency of the sample containing the excessive amount of ZnO content might be attributed to the shading effect and increasing opacity of sample blocking the adsorption light of MoS 2 , and high amount of ZnO content could cause severe agglomeration which results in the reduction of charge transfer capability. It can be concluded that, the photo-activity of the samples depends on the mass ratio of MoS 2 and ZnO, indicating that the proper amount of ZnO is of great importance for the synergistic effect between MoS 2 and ZnO. Therefore, the selected amount of ZnO content in the ZnO/MoS 2 is equal to 20 wt.%, and it is considered to be proper for further investigation [62][63][64][65] .
Study on the role of carbonaceous materials substrate. In order to investigate the effect of carbonaceous materials as support material for catalytic behavior of ZnO20%/MoS 2 , the degradation of aniline was evaluated. Figure 7(B) clearly shows that the presence of RGO had a substantial effect on the photocatalytic ability of ZnO20%/MoS 2 . This result is attributed to the excellent properties of RGO nanosheets incorporated into ZnO20%/MoS 2 which are explained as follows: (1) incorporation of RGO caused an increase in the surface area and enhancement in the adsorption capacity through providing a 2D support material for ZnO20%/MoS 2 . As a result, more pollutant molecules can be absorbed through absorption sites. (2) More importantly, incorporation of RGO would result in a delay in the recombination of charges by separating them effectively. This delay could be attributed to the characteristics of carbonaceous materials including having remarkable electron storage capacity and the capability to act as the electron sink, such that photo-induced electrons can be accumulated on their structures. Subsequently, these electrons can be trapped by the dissolved oxygen to form superoxide and hydroxide radicals. Concurrently, the holes in the VB could be reacted with the pollutants to degrade them or react with water molecules and produce more.OH radical. With the increase in RGO content from 7% to 10%, the degradation increased gradually from 68.5% to 84%, while by increasing RGO substrate to 13%, the degradation rate was observed to have a decreasing trend (Fig. 7B).
The major reasons for a decreased catalytic ability caused by excessive incorporation of RGO included the role of RGO as a center of recombination of charges and also the shielding effect of the overloaded RGO causing less light to reach to the surface of the catalyst. Same performance was observed after modification of ZnO20%/MoS 2 using CNTs (Fig. 7C). The comparison made between photo-activity performances of ternary nanocomposites showed that RGO is a better alternative for promoting the efficiency of ZnO20%/MoS 2 nanocomposite. Although graphene and CNT have similar superior properties in common, such as great adsorption capacity, high electron transition rate, and large specific surface area, they were not similar in enhancing the photocatalytic performance, when used as a composite with ZnO/MoS 2 . After being exposed to 120 min of irradiation, degradation rate of aniline over the optimized RGO10%/ZnO20%/MoS 2 reached to 84% while for CNT10%/ZnO20%/MoS 2 , this value was equal to 76%. Adsorption capacity of the composite sample is one of the major factors which influence its behavior. Obviously, the adsorption capacity of RGO modified ZnO/MoS 2 is higher than that of CNTs modified ZnO/MoS 2 catalyst, which should be due to its larger S BET . Based on the optical studies, slight narrowed band gap of RGO10%/ZnO20%/MoS 2 compared with CNT10%/ZnO20%/MoS 2 suggests that interaction between RGO and ZnO/MoS 2 is stronger than that of CNT modified ZnO/MoS 2 51,66 .
In addition, the PL studies confirmed the improved light-harvesting properties in RGO10%/ZnO20%/ MoS 2 . Charge carrier transfer studies provided more evidence regarding the effective migration and separation of charges as well as a reduction in the recombination of charges in RGO10%/ZnO20%/MoS 2 compared to CNT10%/ZnO20%/MoS 2 . All these findings strongly suggest that RGO has more effect on the photo-degradation of aniline under visible light than CNTs.
In order to make a quantitative comparison between the photo-activity of all nanocomposite samples, the kinetics study was carried out. The photocatalytic degradation of aniline was found to follow the pseudo-first order equation (Fig. S1(A)).
where C 0 represents the initial concentration of aniline, C represents the concentration of aniline at time t, and k represents the apparent reaction rate constant. The rate constants values was obtained from Fig. S1(B). According to the k value, RGO10%/ZnO20%/MoS 2 sample showed a rapid increase in photo-degradation rate suggesting that RGO is beneficial for promoting the photo-activity. As a result, this nanocomposite was chosen as the desired ternary sample for the following investigation.
Study on the role of pH. Solution pH is considered as one of the main parameters that can dramatically influence the photocatalytic reaction since it influences the formation of active radicals and the surface charge of the adsorbate. Due to the formation of intermediate species which may alter the pH of the solution, the pH was monitored during the process. Figure 7(D) depicts the degradation profile of aniline as a function of time at different pH values. Clearly, the degradation rate reached to its maximum level in acidic medium, and it decreased along with increasing pH value. This can be explained according to the surface charge of photocatalyst and the state of aniline at different pH values. Typically, the zero point of charge (pH zpc ) for the RGO10%/ZnO20%/MoS 2 influenced its surface charge at different pH values. For the RGO10%/ZnO20%/MoS 2 , the pH zpc was reported to be equal to 2.8. Thus, at pH value < 2.8, the surface of the composite is positively charged, while at pH value > 2.8, it is negatively charged. Besides, the acid dissociation constant (pKa) of aniline was reported to be equal to 4.6 meaning that aniline at values below than this value is positively charged and at values above than this value, it is negatively charged. As a consequence, at high value of pH, both aniline and catalyst are negatively charged. Thus, the interaction between them becomes repulsive, resulting in a decrease in the degradation rate. Conversely, at pH = 4, the electrostatic attraction forces between the positively charged aniline and negatively charged RGO10%/ZnO20%/MoS 2 leads to an increase in the degradation rate. Moreover, the recombination of e-h pairs is less likely to occur in the acidic medium which is considered to be favorable for photocatalytic reaction. It should be noted that due to the excess amount of H + ions at the low value of pH, more electrons migrate to the surface of the photocatalyst to react with O 2 to generate active species (superoxide and hydroxyl radicals). Prior studies reported a similar effect of pH value on photo-degradation of aniline [67][68][69] .
Study regarding the role of nanocomposite dosage. Catalyst dosage is another key parameter influencing the efficiency of the photocatalytic reaction. To determine the role of catalyst dosage, different amounts of RGO10%/ ZnO20%/MoS 2 were used varying from 0.4 to 1 g/L. The results are shown in Fig. 7(E). It was found that by Scientific RepoRtS | (2020) 10:4414 | https://doi.org/10.1038/s41598-020-61367-7 www.nature.com/scientificreports www.nature.com/scientificreports/ increasing the amount of catalyst dosage from 0.4 g/L to 0.7 g/L, the degradation rate increased from 72% to 95%. However, beyond this value, a slight decrease was observed in the rate of reaction. The increase in the degradation rate caused by increasing the catalyst dosage was accompanied by the abundance of active sites, and the generation of active radicals. In higher amounts of catalyst dosage, the decrease in the activity was observed, which is due to the fact that the number of aniline molecules are not enough in the reaction medium to be absorbed on the semiconductor sites. In other words, the additional amount of catalyst substrates does not have any effect in the reaction. Moreover, when high amounts of catalyst is used, the reaction mixture becomes opaque, and the penetration of light to the surface of substrates would be hindered 70,71 .
Study on the role of aniline concentration. After adjusting the catalyst dosage and pH of the solution, in order to determine the proper amount of aniline concentration, the photo-degradation process was carried out using three different concentration of aniline and the results are presented in Fig. 7(F). It was observed that the increase the aniline concentration resulted in the decrease in the removal rate. The trends for degradation of aniline were similar to the trends for degradation of other organic pollutants. A possible explanation for this observation is that, as the concentration of aniline increases, a significant amount of light might be absorbed by aniline molecules rather than by catalyst, which in turn less active species would be produced. Besides, the catalyst sample has a restricted number of active sites which are saturated with aniline molecules. Therefore, the adsorption of OH − on the catalyst surface would be decreased which leads to the reduction in the formation of .OH and .O 2 radicals. Furthermore, the formation of intermediates during the photocatalytic process influences the overall aniline degradation rate. Since by increasing the initial concentration, the amount of formation of intermediates increases which competitively attach to the surface of the catalyst and also competitively react with oxidant species 70,72 .
Herein, for a better evaluation, the catalytic activity of the as-fabricated nanocomposite in the degradation of aniline was compared with those reported in previous studies (see Table 3). The RGO/ZnO/MoS 2 showed performance than did the nanomaterials used in previous works, since it decompose higher amount of aniline in less time under visible light. To sum up, RGO/ZnO/MoS 2 is an efficient photocatalyst for decontamination of aqueous medium containing organic pollutants.
Mineralization. The TOC analysis was performed to identify the remained intermediates in the reaction medium 73 . Therefore, an experiment was carried out in the optimal aniline degradation condition which is as follows: pH value of 4, RGO10%/ZnO20%/MoS 2 dosage of 0.7 g/L, and initial aniline concentration of 80 ppm. The complete degradation of aniline was achieved over the time period of 120 min, however, the TOC removal was found to be by 40%, and the complete removal of TOC was obtained during the time period of 210 min because aniline molecules were degraded into other organic intermediates. Figure S2 shows the results obtained during mineralization process of aniline.
Study on the effect of radical scavenger. Trapping experiments were carried out with the purpose of identifying the active species in degradation of aniline and determining the mechanism of the photocatalytic reaction. Holes (h + ) and radicals such as hydroxyl radicals and superoxide radicals are the common oxidative species in the photocatalytic process. To identify the predominant species for aniline degradation over the RGO10%/ZnO20%/ MoS 2 nanocomposite in the reaction, three active species of scavengers, namely, disodium ethylene diamine tetra acetic acid (Na-EDTA, 10 mL), benzoquinone (BQ, 10 mL), and 10 mL of isopropanol (IPA, 10 mL) were added to the reaction media for evaluating the effect of radical scavengers h + , . O 2 , and . OH, respectively. Figure 8(a) shows that after addition of radical scavengers, the degradation percentage achieved in the order of EDTA > BQ > IPA. EDTA had less effect on the degradation percentage which suggests that h + contributed to the minor extent in the removal of aniline and only a few number of h + species were produced in the degradation process. On the other hand, the reaction efficiency was greatly impeded following the addition of IPA and BQ, suggesting that . OH and . O 2 are the main active species in the photocatalytic reaction 74 .
According to the abovementioned information, a probable mechanism for degradation of aniline over the RGO/ZnO/MoS 2 photocatalyst was proposed using the information presented in   www.nature.com/scientificreports www.nature.com/scientificreports/ and then translate into hydroxyl radicals. In this case, the charges are successfully separated to avoid being recombined again. Besides, given that RGO has superior electron mobility and storage capacity, it can act as an electron reservoir. It can advance the suppression of electron, hole recombination and extend the lifetime of charge carriers, as confirmed by the PL spectra discussed earlier. The accumulated electrons on the conduction band of ZnO can move to RGO sheets. These electrons can reduce the oxygen into superoxide radicals. These radicals can directly degrade aniline or can be translated to hydroxyl radicals for further degradation of aniline. In conclusion, the synergistic effects of MoS 2 , ZnO, and RGO facilitates the separation of charges and consequently, the degradation of aniline is improved by the use of RGO10%/ZnO20%/MoS 2 .
Reusability and stability. In addition to catalytic efficiency, the reproducibility and stability is another essential feature of the photocatalyst, which is referred to its cost-effective practical usage. For investigation of the reusability of the RGO10%/ZnO20%/MoS 2 sample, the experiment of photocatalytic degradation of aniline at optimal condition (pH value of 4, photocatalytic dosage of 0.7 g/L and aniline concentration of 80 ppm) was repeated for five times. After each run, RGO10%/ZnO20%/MoS 2 sample was collected using centrifuge, and was washed using distilled water. Then, it was dried, and reused for the next run. Figure 8(B) shows that there is no significant loss of catalytic activity after five cycles of experiment and the negligible loss is maybe due to the loss in the amount of RGO10%/ZnO20%/MoS 2 during its preparation for each cycle, or reduction of surface area caused by absorbing the untreated by-products to the surface of the photocatalyst. Furthermore, after the five cycles, the sample was subjected to XRD, FTIR, BET, and FESEM analysis to investigate the structural, textural, and morphological stability of RGO10%/ZnO20%/MoS 2 (Fig. S4(A-D)). The XRD and FTIR patterns of the recovered sample ( Fig. S4(A,B)) shows that, no additional peak was appeared and the crystalline nature of photocatalyst has been preserved after five consecutive cycles. Figure S4(C) shows the FESEM image of the nanocomposite after five cycles, and it reveals that the morphological features remained unchanged. Figure S4(D) exhibits the texture properties of the samples after five runs, and it indicates that there was not any obvious change in the type of nitrogen adsorption-desorption isotherms. However, BET measurement data shows slight decrease in the S BET which might be attributed to the fact that during the cycles, a number of untreated intermediates attached on the nanocomposite surface and block the pores, resulting in the reduction of surface area of the nanocomposite.
Consequently, RGO10%/ZnO20%/MoS 2 showed excellent reusability and stability without any photo-corrosion in the photocatalytic reaction, which makes it to be considered for long-run utilization in the tests for the elimination of organic pollutants.
Real wastewater treatment. In order to determine whether the catalyst sample fabricated here was proper for practical purposes, a real petrochemical wastewater sample was prepared to be treated using the RGO10%/ ZnO20%/MoS 2 . The properties of petrochemical wastewater sample are presented in Table S1. In order to find the untreated wastewater components, GC-MASS analysis was carried out. The main components of untreated  (Fig. S5).
The efficiency of the as-fabricated photocatalyst in terms of TOC and COD removal for the petrochemical wastewater sample was found to be equal to 93% and 100%, respectively, in the presence of RGO10%/ZnO20%/ MoS 2 photocatalyst under the optimum degradation condition (pH value of 4, catalyst dosage of 0.7 g/L) over the time period of 440 min (Figs. S6 and S7). Photocatalyst is identified as a reliable material for efficient long term mineralization and partial oxidation of organic compounds 75 . conclusions In summary, ternary RGO/ZnO/MoS 2 and CNTs/ZnO/MoS 2 nanocomposites were synthesized with different amount of RGO and CNTS contents by a typical hydrothermal method and their catalytic ability was evaluated in terms of photo-degradation of aniline. The RGO/ZnO/MoS 2 showed the best efficiency compared to that of CNTs/ZnO/MoS 2 , ZnO/MoS 2 , and MoS 2 . The outcoming of the present study suggested that RGO has a substantial effect on the catalytic behavior of ZnO/MoS 2 . The results of characterization also confirmed the significant effect regarding the incorporation of RGO. The UV-vis analysis and PL spectra showed that RGO/ZnO/MoS 2 possess extended visible absorption, and as a result the separation of photo-induced charges was promoted. The BET analysis showed that the surface area also increased greatly in the presence of RGO. In addition, the degradation process was conducted under various operational parameters, and complete degradation of aniline was achieved at pH value of 4, catalyst dosage of 0.7 g/L, and aniline concentration of 80 ppm after being exposed to light irradiation for 120 min. Trapping experiments were performed with the purpose of recognizing active radicals in the reaction and the results suggested that . OH played a paramount role in degradation progress. In the real wastewater sample, the COD and TOC ratios decreased to zero and 7%, respectively after 440 min under the same operational conditions. The obtained results revealed that RGO/ZnO/MoS 2 has great potential for the remediation of wastewater containing different kinds of organic pollutants due to its tremendous catalytic ability, stability and reusability.