A novel MoS2/ tourmaline /graphene ternary composite with enhanced visible-light photocatalytic property

In this study, MoS2-graphene-tourmaline (MoS2-GR-T) composite photocatalyst was successful synthesized via one-step hydrothermal method. Raman spectra revealed that graphene oxide was reduced to graphene. Scanning electron microscopy images showed that MoS2 was dispersed well on graphene. Transmission electron microscope images showed that MoS2 and tourmaline contacted well with graphene. Analysis of UVvisible diffuse reflectance spectra implied that the bandgap energies of MoS2, MoS2-T and MoS2-GR-T samples were 2.01 eV, 1.91 eV and 1.79 eV, respectively. The photocatalytic performances are evidenced under Xenon lamp irradiation utilizing 2 Rhodamine B dye as the model compound. Compared with MoS2 and MoS2-GR, the MoS2-graphene-tourmaline (MoS2-GR-T) composites exhibited the excellent photocatalytic activity for the degradation about irradiating for 60 min were 93.9% under visible-light. The enhanced photocatalytic activity of MoS2-GR-T composite could be attributed to the exposed adsorption-photocatalytic active sties, the improved light adsorption ability and the promoted charge separation efficiency. The introduction of tourmaline reduced the band gap explored by analysis of UV-visible diffuse reflectance spectra. This work demonstrated that the charge efficiency of photocatalysts could be promoted by coupling both metal-free co-catalyst and polar mineral.

However, fast recombination of the photo-generated e -/h + pairs restricted its photocatalytic property [25]. There has been significant progress in solving the aforementioned problems. Co-catalysts [15,[26][27][28] and supporting materials [20,[29][30] were regarded as effective strategies to improve the photocatalytic property of MoS2. The co-catalysts were studied based on the heterojunction exist in the interface of catalysts, which could transfer photo-induced e -/h + pairs to promote the photocatalytic property [28,31]. While, supporting materials were used to disperse catalysts to expose more active sites due to its large surface area [20,29]. GR is a kind of two-dimensional carbon species and is available to use as supporting and electron-transfer materials in photocatalytic field because of its higher specific surface area and outstanding electronic properties [32][33][34].
It was widely demonstrated that ternary photocatalysts could result in higher photocatalytic property due to synergistic effect of the components [51][52]. For example, BiPO4-MoS2-graphene composite showed enhanced photocatalytic activity than that of BiPO4, MoS2, and MoS2-BiPO4 composite for photocatalytic degradation of dyes [53].
The photocatalytic activity of the samples was evaluated by degradation of Rhodamine B (RhB) under visible-light irradiation.

Results and Discussion
The crystal structures of photocatalysts were further characterized by XRD displayed in Fig.1. There was a strong diffraction peak located at 2θ =26.56° (Fig. 1a) for graphite, indicating good crystallization nature. The (001) plane of GO showed a broad diffraction peak at 2θ = 10.69° (Fig. 1b). A broad peak appeared at 2θ=25.22° after GO went through hydrothermal reaction (Fig. 1b), suggesting effective deoxygenation of GO [54]. The XRD pattern of tourmaline powders (Fig. 1c) was well indexed to that of Fe-rich tourmaline (JCPDS 85-1811) and no other peak of impurity was observed. The peaks located at Raman spectra were employed to further determine the co-existence of the samples (Fig.   2). Two Raman peaks appeared at 1350 cm -1 and 1601 cm -1 (Fig. 2a) could be assigned to the D band (signal from the disordered carbon) and G band (signal from the sp2 hybridized carbon). The ID/IG (I meant the band intensity) ratio of GO, GR and MoS2-GR-T samples were 0.96, 1.19 and 1.33, respectively. The increased ID/IG ratio indicated that small graphene domains have been re-established [55,62]. For MoS2-GR-T sample, the D band of GR shifted to 1344 cm -1 and 1586 cm -1 , respectively, indicating successful synthesis of ternary MoS2-GR-T composite. The peaks appeared at 377 cm -1 and 403 cm -1 (Fig. 2b) were originated from the 2 1 vibration model (Mo and S atoms vibrate along the in-plane direction and oppositely to each other) and A1g vibration model (S atom vibrate along the perpendicular-to-plane direction) of Mo-S bond, respectively [63][64], and a red shift for the two peaks were observed for MoS2-T-GR sample.
The SEM images of morphology of the samples were observed in Fig. 3. As it can be seen in Fig. 3a, there were wrinkles on the surface of GO. The tourmaline was granular appearance particles ( Fig. 3b) with the average size of ～600 nm. The pristine MoS2 was flower-like sphere with a diameter of ～800 nm (Fig. 3c). The introduction of tourmaline resulted in the formation of smaller MoS2 particles (～400 nm) (Fig. 3d), synergized with GR to make MoS2 grow along and dispersed well on the substrates (Fig. 3e). The  It is of importance to explore the effect of the initial pH of RhB aqueous solution on the photocatalytic degradation rate as the photocatalytic oxidization reaction of dyes usually occurred in aqueous solution. The results showed that the degradation rate decreased with increase of pH (Fig. 7c). RhB is an aromatic amino acid with amphoteric characteristics due to the presence of both amino group and carboxyl group. Therefore, the charge state of RhB depends on pH values of the solution [67][68]. When pH was below 3.10, RhB was positively charged, and when pH was above 3.10, RhB was negatively charged [69]. Results from zeta potential analysis revealed that MoS2-GR-T composite was negatively charged with pH varied at the range of 2.02-8.03 (Fig. 7d).
Photocatalytic property was affected mainly by the adsorption ability of photocatalysts [70]. When pH was above 2.02 and below 3.10, the adsorption between RhB and MoS2-GR-T composite was enhanced mainly by the electrostatic attraction, and when pH was visible-light irradiation while they were not observed in the dark (Fig. 8b).
Potential mechanism for the enhanced photocatalytic property with the composite was schematically illustrated in Fig. 9. (1) MoS2 was dispersed well on the surface of graphene and exposed more adsorption-photocatalytic active sites.
MoS2 (eVB -) + graphene Graphene (e -) + MoS2 (2) Graphene (e -) + Tourmaline Graphene + Tourmaline (e`) The separated e-reacted with the oxygen dissolved in the water to produce superoxide radical anion (•O 2-). The generated •O 2was a kind of strong oxidants towards the decomposition of organic dyes [74]. Accumulated h + on the surface of photocatalyst could oxidize dyes directly [70,[75][76]. Dyes could be degraded in the presence of MoS2-T-GR composite.    Co. Ltd, China. All the reagents were analytical grade and were used as received without further purification. Doubly distilled water was used throughout this study.

Synthesis of graphene oxide
The graphene oxide (GO) used in this work was prepared via an ultrasound-assisted modified Hummer's method. In detail: 1 g of natural flake graphite powder, 1 g of NaNO3, 6 g of KMnO4 and 46 ml of concentrated H2SO4 were added into a beaker and the obtained mixture was immersed in ice water, magnetically stirred for 2 h. The temperature of the mixture was maintained at 35 °C by water bath for 30 min. 92 ml of distilled water was then added into above mixture and the temperature was kept at 98 °C for 15 min. The obtained brown paste was diluted with 280 ml of distilled water and was treated with 20 ml of H2O2 until color of the suspension turn into bright yellow. The suspension was then rinsed with 5% HCl aqueous solution and a large amount of distilled water until the pH=7±0.3. Finally, the obtained bright yellow slurry was dispersed in distilled water and sonicated for 4 h to obtain GO aqueous colloid. The GO aqueous colloid was dried using a freezing dryer and the resultant product was ground into powders in the agate mortar.

Photocatalytic and trapping experiments
Photocatalytic property of the as-synthesized samples was evaluated as follows: 50 mg of the as-synthesized samples were mixed with 100 ml of 5 mg•L -1 RhB aqueous solution and the obtained suspension was stirred in the dark. 4 ml of the suspension were extracted every 20 min and was centrifuged to acquire the clarified solution.
Absorbance of the solution was measured using a unico 2600 UV-vis spectrophotometer at 554 nm wavelength. Adsorption experiments were proceeded until the adsorption-desorption equilibrium had been established. Then the suspension was exposed to visible-light resource (A 300 W Xe lamp with an UV cut-off filter).
Trapping experiments were carried out the same as photocatalytic experiments except

Supporting Information
Supporting Information File 1 File Name: Additional characterization data