Effective charge separation and enhanced photocatalytic activity by the heterointerface in MoS₂/reduced graphene oxide composites

Abstract

MoS₂/reduced graphene oxide (rGO) composites were fabricated via a facile one-step hydrothermal reaction. In the composites, MoS₂ clusters composed of several nanosheets are intertwined tightly with rGO sheets. As the amount of rGO in the composites is increased, the crystallinity of MoS₂ changes little, but the diameter of MoS₂ clusters decreases considerably and they become more dispersive, resulting in a large specific surface area. As compared to the bare MoS₂, the as-prepared MoS₂/rGO composites exhibit remarkably enhanced light adsorption and photocatalytic activity for the degradation of Rhodamine B (RhB) under visible light irradiation. Photoluminescence (PL) spectra measurements suggest that the enhanced photocatalytic properties can be ascribed to the effective charge separation across the heterointerface in MoS₂/rGO composites. It is also demonstrated that the optical and photoelectric properties of two-dimensional (2D) materials could be substantially enhanced by the heterojunction with an appropriate band edge alignment, since they possess the largest specific surface area.

---

1. Introduction

Environmental pollution and energy crisis are two key issues restricting the rapid development of human society at the current stage. Semiconductor-based photocatalysts have received enormous interest as an effective method to alleviate the environmental deterioration by toxic organic pollutants.^1,2 The fundamental requirements for high-efficiency photocatalysts are as follows: (1) large specific surface area so that the organic pollutants can contact with catalysts sufficiently; (2) strong light absorption in the visible range; (3) the photon-generated electron–hole pairs can be separated rapidly.^3,4 To this end, nanocomposites with an appropriate band gap and band edge alignment are commonly favored. A great number of research works have been done in this field from both the experimental and theoretical aspects.^5,6

As well known, the materials with a band gap of about 1.6 eV are desirable for high-efficient absorption of solar energy. However, it is difficult to find out the suitable materials in nature. In recent years, it has been proved that the band gap of two-dimensional (2D) materials can be considerably tuned through changing the sheet thickness and strain state. For example, the band gap of MoS₂ sheets changes from 1.3 eV in bulks to 1.97 eV in single-layer counterpart,⁷ and few-layer MoS₂ exhibits a very high light absorption coefficient up to 7 × 10⁵ cm⁻¹ in the visible wavelength region.^8,9 Moreover, 2D materials possess the largest specific surface area and might provide more sites for photocatalytic reaction. Hence, a great number of research works focused on coupling 2D nanosheets of MoS₂ or graphene with TiO₂,^10,11 Bi₂MoO₆,^12,13 Bi₂WO₆,¹⁴ g-C₃N₄,^15,16 ZnO¹⁷ and so on to fabricate co-catalysts for photocatalytic degradation of dyes or photocatalytic hydrogen evolution. The results demonstrated that the photo-generated electrons could be easily transferred onto MoS₂ or graphene sheets, leading to effective separation of photo-generated electrons and holes, and thus improved photocatalytic performance. Supposing that two kinds of 2D materials are combined into heterojunction with an appropriate band edge alignment, the strong interface effect should enhance the separation of photon-generated electron–hole pairs and the photocatalytic activity more effectively. Meng et al. demonstrated good photocatalytic activity of MoS₂ nano-platelets on rGO sheets for hydrogen evolution reaction (HER) which is dominated by the photo-generated electrons.¹⁸ It was reported that MoS₂/rGO composites are usually p type.¹⁸ That is, holes are the majority carriers, and electrons are the minority ones. In the photocatalytic hydrogen evolution reaction, the electrons will be consumed, and light illumination should be continued in order to induce more electrons and holes. Since a great number of holes occupy the states at the top of valence band in p type semiconductor, the optical excitation becomes more and more difficult. However, in the photocatalytic process for degradation of organic pollutants, the holes are consumed, and thus the optical excitation is easier than that in HER process. Hence, the kinetics should be distinct from that in the HER process. Li et al.¹⁹ synthesized MoS₂/rGO composites via microwave-assisted method and studied photocatalytic degradation of MB under visible light. It was found that MoS₂ sheets in the composites exhibited irregularly aggregated morphology and lower crystallinity with a wide full width at half maximum (FWHM) of XRD peaks. As well known, lower crystallinity means that more defects are involved, and they usually act as the centers for the recombination of photo-generated electrons and holes. In such a case, the substantial recombination of photogenerated electrons and holes might take place at the defect centers, and affect the photocatalytic activity to some degree.^19,20 So it is still a challenge to fabricate dispersive MoS₂/rGO composites with higher crystallinity.

In this work, hydrothermal process is adopted to prepare MoS₂/rGO composites at a considerably lowered growth speed. X-ray diffraction (XRD), Raman spectra, X-ray photoelectron spectroscopy (XPS) and high-resolution transmission electron microscopy (HRTEM) are used to characterize the structures of the MoS₂/rGO composites. It is found that MoS₂ in the as-fabricated samples commonly exhibit dispersive flower-like morphology with enhanced crystallinity and edge-terminated feature, which are tightly intertwined with rGO sheets. The adsorption ability to organic pollutants will be substantially enhanced for the ultra-thin MoS₂ nanosheets due to large specific surface area, and the edge-terminated nanosheets could provide more active sites for photo-catalytic degradation. Photoluminescence (PL) spectra, absorption spectra and photocatalytic test are conducted to study the effects of heterogeneous interface on the separation of carriers as well as on the photocatalytic properties, and the physical mechanism is discussed.

2. Experimental details

2.1. Samples preparation

Firstly, 1 mmol Na₂MoO₄·2H₂O and 5 mmol thiourea were dissolved in 60 mL distilled water with magnetic stirring for 30 minutes. Then different amount of GO was dispersed into the above clear solution with sonication for 1 h, resulting in a gray dispersion. Next, the homogeneous solution was transferred into a 100 mL Teflon-lined autoclave and kept at 210 °C for 24 h. The precipitates were collected, washed three times with distilled water followed by ethanol two times, and then dried at 80 °C in a vacuum oven overnight to obtain the MoS₂/rGO hybrids. In order to study the effect of the amount of rGO sheets on the growth and photocatalytic activity of MoS₂/rGO hybrids, we prepared a series of MoS₂/rGO composites using different weight percentages of GO relative to Na₂MoO₄·2H₂O. The samples prepared using 5% and 10% GO are presented in this paper and they are labeled as MoS₂/rGO-5 and MoS₂/rGO-10. The bare MoS₂ was obtained according to the above route for preparation of MoS₂/rGO but in the absence of GO.

2.2. Microstructure characterization

XRD (SHIMADZU XRD-7000S diffractometer) was adopted to analyze the phase structure in the as-prepared samples. Scanning electron microscopy (SEM, FEI Quanta 600S) and HRTEM (JEOL JEM 2100F) were adopted to characterize the morphology and heterogeneous interface of the samples. Raman spectra were examined by a Horiba HR800 spectrometer with a 633 nm laser as the excitation light source. N₂ adsorption–desorption isotherms were measured on Micromeritics ASAP2010. The specific surface areas were determined from nitrogen adsorption using the Brunauer–Emmett–Teller (BET) method. XPS measurements were performed on Thermo Scientific K-Alpha XPS spectrometer. PL spectra were detected with a PTI QM40 spectrometer using a 532 nm line from a xenon lamp. The UV-vis absorption spectra of samples were measured on a Hitachi U-4100 spectrophotometer.

2.3. Photocatalytic tests

RhB was used as the organic pollutant to evaluate the photocatalytic activity of the samples under visible light illumination by a 1 kW xenon lamp. Firstly, 10 mg photocatalyst was added into 40 mL of 20 mg L⁻¹ RhB solution with ultrasonication for 30 min, and the homogeneous solution was stirred in the dark for 3 h to reach adsorption–desorption equilibrium. Secondly, the photocatalyst was collected from the solution by using the centrifugation method, and then added to 40 mL of 10 mg L⁻¹ RhB solution which is close to the concentration in adsorption–desorption equilibrium. In this way, the real initial concentration of RhB solution at the beginning of degradation could be maintained as a constant value of 10 mg L⁻¹. During the process of degradation reaction, 4 mL mixed solution was taken out for each 30 min, and then the photocatalyst was separated by centrifugation at 8000 rpm for 5 min. The reduction in the concentration of RhB solution was measured by the Hitachi U-4100 UV-vis spectrophotometer. In order to evaluate the stability of photocatalyst, it was collected by centrifugation at 8000 rpm for 5 min after degradation reaction, then washed with deionized water for 3 times to eliminate contaminants on surface, and dried at 80 °C for 6 h for another cycle.

2.4. Photoelectrochemical measurements

The photoelectrodes were prepared by dip-coating method. 4 mg of the samples was suspended in 0.1 mL isopropanol to make a slurry with ultrasonication for 20 min, and the slurry was dip-coated on a 2 cm × 2 cm F-doped SnO₂-coated glass (FTO glasses). To attach the samples on FTO glasses, the photoelectrodes were calcined at 500 °C under a constant flow of Ar gas for 1 h. The photoelectrochemical experiments were done in an Autolab PGSTAT 128N electrochemical station with a xenon lamp. The photoelectrodes acted as the working electrode, and a Pt sheet and a saturated calomel electrode (SCE) were used as counter and reference electrode, respectively. 0.5 mol L⁻¹ Na₂SO₄ solution was used as the electrolyte. Before measurement, the electrolyte was purged with nitrogen for 30 min.

3. Results and discussion

Fig. 1 shows the XRD patterns of the as-synthesized MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 composites. The diffraction peaks could be perfectly indexed to 2H–MoS₂ (JCPDS 37-1492). The strongest peak at 2θ = 14.1° corresponds to (002) crystal plane, indicating a typical lamellar structure along the c axis.²⁰ The FWHM values of (002) peak are 1.14°, 1.17°, and 1.20° for the three samples. The close FWHM values indicate similar crystallinity of MoS₂ in all the samples, and the content of rGO hardly affects the crystallinity of MoS₂. This is in agreement with the previously reported results.²¹ As compared to the XRD patterns of MoS₂/rGO in ref. 18, the FWHM value in our work is substantially reduced. This indicates improved crystallinity in our samples, and should result in enhanced photocatalytic activity. Fig. S1† shows the XRD patterns of MoS₂/rGO-5 composites prepared for a reaction time of 12 h, 18 h, 24 h, and 30 h [ESI†]. It also indicates that the reaction time influences the crystallinity of samples little.

Fig. 1 XRD patterns of as-prepared MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 samples.

Fig. 2(a) displays the Raman spectra of the samples. The two dominant peaks at 378 cm⁻¹ and 405 cm⁻¹ correspond to the in-plane E¹_2g and out-of-plane A_1g vibration modes in the hexagonal MoS₂ lattice, respectively.²² Two peaks at 1334 cm⁻¹ and 1584 cm⁻¹ matching very well to the D- and G-bands of graphene are identified from the Raman spectra of MoS₂/rGO-5 and MoS₂/rGO-10 samples. The results suggest that the hybrid structure of MoS₂ and rGO is successfully fabricated through the one-step method. Furthermore, the relative change of the integrated intensities of E¹_2g and A_1g can provide some interesting information about the terminated structure of MoS₂.^22,23 Fig. 2(b) shows the intensity ratio of E¹_2g and A_1g peaks of the as-prepared samples. Obviously, the intensity ratio of the MoS₂/rGO composites is lower than that of bare MoS₂. The higher intensity ratio of E¹_2g and A_1g peaks stands for the formation of surface-terminated terrace structure, the lowered intensity ratio is related to the edge-terminated structure. Hence, the edge-terminated MoS₂ nanosheets are preferred in MoS₂/rGO-5 and MoS₂/rGO-10 composites. In essential, the free energy of the edge sites is higher than that of the terrace sites by 2 orders of magnitude. So the edge-terminated configuration is unstable. But it could be produced in non-equilibrium process.²² In the growth process of MoS₂/rGO composites by hydrothermal method, the functional groups on GO nanosheets could promote the nucleation and growth of MoS₂. So the growth rate of MoS₂ in the solution with GO available is much faster. This facilitates the formation of unstable edge-terminated MoS₂.²² The rim and edge sites on MoS₂ are considered as the active sites in photo-catalytic degradation of dyes, because they have a strong interaction with dyes molecules via unstable dangling bonds.^24,25 So the MoS₂ nanosheets with exposed edge sites are critical for photo-catalytic degradation of dyes, and they have also attracted great interests in other fields.^26–28

Fig. 2 (a) Raman spectra, (b) the intensity ratio of E¹_2g and A_1g lattice vibration mode in MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 samples.

XPS spectra of MoS₂/rGO-5 composite as well as those of GO are measured to study the reduction degree of rGO. As shown in Fig. 3(a), the high-resolution XPS spectrum of C 1s in GO can be decomposed into three components: (1) the C–C, C=C, or C–H at the binding energy of 284.8 eV; (2) C–OH at 286.7 eV; (3) C–O–C or C=O at 288.0 eV characteristic of oxidized state of graphene. The result is consistent with that reported previously.²⁹ The XPS peaks of C 1s, Mo 3d, S 2p in the MoS₂/rGO-5 sample can be clearly identified from Fig. 3(b–d), respectively. The high-resolution spectrum of C 1s in Fig. 3(b) can be decomposed into two peaks at 284.8 eV and 286.7 eV, and the peak of oxygen-containing functional groups is substantially reduced as compared to that in Fig. 3(a). So GO has been effectively reduced to rGO during hydrothermal process.³⁰ The Mo 3d spectrum in Fig. 3(c) can be fitted into two peaks at 229.4 eV and 232.6 eV, corresponding to Mo 3d_5/2 and Mo 3d_3/2, respectively, while the S 2p spectrum in Fig. 3(d) can be resolved into two peaks at 162.3 eV and 163.4 eV. It suggests that Mo⁴⁺ and S²⁻ ions are involved in the samples.²¹ Namely, MoS₂/rGO composites are successfully fabricated.

Fig. 3 High-resolution XPS spectra: (a) C 1s in the GO sample; (b) C 1s, (c) Mo 3d, and (d) S 2p in the MoS₂/rGO-5 composite.

Fig. 4(a–c) show the SEM images of the surface morphologies of MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 composites, respectively, and Fig. 4(d–f) present the corresponding high-resolution images. As shown in Fig. 4(a) and (d), the bare MoS₂ is composed of sphere-like clusters with an average diameter of about 2.34 μm, and the cluster consists of many nanosheets with the thickness of about 20 nm. It is reasonable to deduce that the similar clusters in Fig. 4(b) and (c) are also MoS₂, and the thinner chiffon-like sheets must be rGO. As exhibited in Fig. 4(b) and (c), the MoS₂ clusters are tightly intertwined with rGO sheets, and the intimate contact makes the electronic interaction between MoS₂ and rGO possible.³¹ The mean diameter of clusters in MoS₂/rGO-5 and MoS₂/rGO-10 is about 0.93 μm and 0.70 μm, respectively. Notably, as the content of rGO is increased, the diameter of MoS₂ clusters gradually decreases and they become more dispersive. The detailed measurements are presented in Fig. S2 and S3.† It means that the GO sheets can promote the nucleation of MoS₂ but suppress the aggregation of MoS₂ clusters. N₂ adsorption–desorption isotherm curves are measured and employed to evaluate the specific surface area by BET method. The results are shown in Fig. S4.† The specific surface area of MoS₂, MoS₂/rGO-5, and MoS₂/rGO-10 is 2.6 m² g⁻¹, 23.5 m² g⁻¹, and 55.8 m² g⁻¹, respectively. Obviously, the composites have much larger specific surface area as compared to the pure MoS₂. The dye molecules can be adsorbed on the dispersive MoS₂ clusters easily. Fig. 4(g–i) shows the HRTEM images of the samples. Lamellar structures with an interlayer separation of 0.63 nm can be identified in all the samples, corresponding to the (002) planes of MoS₂. The results are in good agreement with the XRD patterns. Furthermore, (002) planes are the exposed facets of MoS₂ in the as-synthesized samples. This confirms that the interface between MoS₂ and rGO may be formed by (002) planes.

Fig. 4 SEM (a–f) and TEM (g–i) images of the samples (MoS₂: (a), (d), and (g); MoS₂/rGO-5: (b), (e), and (h); MoS₂/rGO-10: (c), (f), and (i)).

In order to investigate the growth process of MoS₂ on GO sheets, a series of MoS₂/rGO-5 samples at different reaction time were prepared, and Fig. S5† shows the SEM images of the morphologies. No particles can be observed on rGO sheets after 1.5 h reaction [Fig. S5(a)†]. When the reaction time is elongated to 3 h, a great number of particles with an average diameter of 0.45 μm appear [Fig. S5(b)†]. EDS (Energy Dispersive Spectrometer) analysis indicates that the particles are MoS₂, as depicted in the inset of Fig. S5(h).† It can be seen from Fig. S5† that the MoS₂ clusters distinctly grow up with the elongating reaction time, and the separation between nanosheets becomes larger and larger. As shown in Fig. S5(f),† the nanosheets of MoS₂ clusters prepared at a reaction time of 24 h are more dispersive and bigger than those obtained in other conditions [Fig. S5(b)–(e) and (g)†]. So, we choose the samples prepared at a reaction time of 24 h to study the effect of rGO sheets on the photocatalytic activity. The effect of the functional groups of GO on the growth of MoS₂ is studied by using graphene sheets as the matrix rather than GO. As shown in Fig. S6,† no similar particles are available on graphene sheets after 3 h reaction. It indicates that the functional groups of GO facilitate the nucleation of MoS₂, as reported in the other graphene-based composites.^32,33 Fig. 5 schematically displays the mechanism for the formation of MoS₂/rGO hybrid. When Na₂MoO₄ is introduced into the suspension of GO sheets, MoO₄²⁻ will interact with the functional groups of GO, resulting in the selective nucleation of MoS₂ on GO sheets.²⁶ This interaction may differ from that in the presence of hexadecyltrimethyl ammonium bromide (CTAB) in which the CTAB usually acts as a bridge to overcome the charge incompatibility between GO and anions.³⁴ During the hydrothermal reaction, H₂S is generated through decomposition of CN₂H₄S at 210 °C and the GO sheets are reduced to rGO by H₂S, meanwhile, the MoO₄²⁻ is transformed into MoS₂ clusters. Finally, the flower-like clusters consisting of nanosheets are formed.

Fig. 5 Schematic illustration for the synthesis of MoS₂/rGO composite.

Fig. 6(a) and (b) present the UV-visible absorption and PL spectra of the samples. According to the previous reports,³⁵ four characteristic excitonic peaks A, B, C, and D can be resolved at 408, 465, 617, and 678 nm from the absorption spectra of MoS₂ and are marked by black arrows. Peaks A and B are direct excitonic transitions at the point K in the Brillouin zone due to the spin–orbit splitting of the top of valence band, while peaks C and D are from the deep level in the valence band to conduction band at the point M.^35,36 As compared to the bare MoS₂, the absorption intensity of MoS₂/rGO-5 and MoS₂/rGO-10 is significantly enhanced over the entire range of wavelength investigated owing to the intrinsic light absorption of rGO.³⁷ Therefore, the hybrid structure, in particular, the MoS₂/rGO-5, is a good harvesting carrier for visible light. The lower absorption of MoS₂/rGO-10 as compared with MoS₂/rGO-5 may be related to the stronger scattering of light by the excess rGO sheets. PL spectra have been widely employed to probe the transfer behavior of the photo-generated carriers across the interface in the composites.²⁹ As shown in Fig. 6(b), broad and intense emission peaks exist in the PL spectra of MoS₂, which can be assigned to a combination of deep level emissions and indirect band-edge recombination in MoS₂. However, substantial quenching of PL is observed in MoS₂/rGO-5 and MoS₂/rGO-10 samples, indicating that the recombination of photo-generated electrons and holes is effectively suppressed by the heterojunction between MoS₂ and rGO sheets. This should improve the photocatalytic activity. The remarkable quenching of PL indirectly proves that MoS₂ clusters are tightly intertwined with rGO sheets in the composites.

Fig. 6 (a) UV-visible absorption and (b) photoluminescence spectra of the as-prepared MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 samples.

Photocatalytic degradation of RhB in aqueous solution was conducted under visible light irradiation to evaluate the photocatalytic performances of MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10. Since the adsorption of organic molecules on photocatalyst is a key step during photodegradation process, we firstly investigate the adsorption characteristics of RhB on the MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10. Fig. 7(a) displays the adsorption percentage of RhB after reaching the adsorption–desorption equilibrium in the dark. The absorption efficiencies of bare MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 composites are 24.20%, 55.56% and 71.74%, respectively. Obviously, the MoS₂/rGO-5 and MoS₂/rGO-10 composites exhibit improved adsorptivity of RhB as compared with the bare MoS₂. The enhancement was also observed in TiO₂–rGO composites,³⁸ and is mainly due to the strong interaction between RhB molecules and the decorated rGO sheets via a conjugational π–π stacking with a face-to-face orientation.³⁹ As the amount of rGO in the composites is increased, the contribution from the more dispersive MoS₂ clusters and nanosheets to the improved adsorptivity is not negligible.

Fig. 7 (a) Absorption of RhB after reaching the adsorption equilibrium in the dark, (b) photocatalytic degradation profiles of RhB solution under visible light irradiation, (c) photocatalytic degradation reaction kinetics, and (d) recycling test on the MoS₂–rGO-5 composite for degradation of RhB under visible light irradiation.

Commonly, the normalized temporal concentration change (C/C₀) of dye solution is proportional to the normalized maximum absorbance (A/A₀),⁴⁰ so the change in concentration of RhB solution can be easily characterized by the absorption spectra of RhB at 553.5 nm. Fig. 7(b) shows the photocatalytic degradation profiles of RhB solution with MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 under visible light irradiation. It can be found that, after photodegradation for 180 min, the concentration of RhB solution decreases by about 28% in the presence of MoS₂, however, the degradation is remarkably enhanced by MoS₂/rGO-5 and MoS₂/rGO-10 composites. The photocatalytic efficiency is quantitatively evaluated by exploring the kinetic process. In pseudo-first-order kinetics, the photocatalytic degradation process can be described by

where k_app is the apparent reaction rate constant (in unit of min⁻¹), C₀ is the concentration of RhB at the adsorption equilibrium, C is the residual concentration at time t. k_app is calculated by the gradient of ln(C₀/C) with respect to time (t).⁴¹ The results are shown in Fig. 7(c). The linear fit between ln(C₀/C) and reaction time (t) with a slope very close to unity is characteristic of pseudo-first-order reaction kinetics. The k_app of MoS₂/rGO-5 and MoS₂/rGO-10 is 0.00378 min⁻¹ and 0.00240 min⁻¹, respectively, higher than that of bare MoS₂ (0.00184 min⁻¹). Hence, rGO plays a vital role in improving the photocatalytic degradation of RhB. However, the excess rGO in MoS₂/rGO-10 samples results in strong light scattering and weak irradiation on MoS₂. In such a case, the photocatalytic activity is lowered as compared to MoS₂/rGO-5 samples on the contrary.³ It is suggested that an appropriate amount of rGO is desirable in the composites for the best photocatalytic performance. Similar observations have also been reported in TiO₂/rGO composites.³⁸ Fig. 7(d) shows four recycling runs tests on the MoS₂/rGO-5 composite for degradation of RhB under visible light irradiation. The C/C₀ decreases with time in almost the same tendency during the four cycling tests, that is, the photocatalytic activity of the MoS₂/rGO-5 composite degrades little. Hence, the MoS₂/rGO-5 composite should exhibit excellent stable recycling performance for photocatalytic degradation of organic pollutions.

The photocurrents of MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 under visible light irradiation were measured through dip-coating the as-prepared samples on FTO electrodes. The potential of the working electrodes against Pt counter electrode was set at 0.8 V. In the photocurrent measurement, when the MoS₂/rGO composites are irradiated by visible light, the electrons on the top of valence band of MoS₂ will be excited to the bottom of conduction band, and holes are left on valence band, then the electrons and holes will flow towards the opposite directions resulting in electronic current, and the photocurrent response is detectable when the light irradiation is switched on and off. The photocurrent magnitude is related to the separation efficiency of photo-generated electrons and holes. The results are displayed in Fig. 8. An obvious photocurrent response is observed for each switch-on/off event. In contrast, the photocurrent of MoS₂/rGO-5 and MoS₂/rGO-10 is higher than that of bare MoS₂. The improved photoelectric property can be ascribed to the effectively suppressed recombination of photo-generated electrons and holes by the heterojunction between MoS₂ and rGO sheets. The MoS₂/rGO-5 composites exhibit the fastest and the highest photocurrent response, indicating the superior separation efficiency, which is in good agreement with the results of photo-degradation tests.

Fig. 8 Photocurrent responses of MoS₂, MoS₂/rGO-5 and MoS₂/rGO-10 samples.

The energy band alignment and trapping experiments are required to understand the physical mechanism for the enhanced photocatalytic properties. The work function of graphene which is prepared by reduction of GO through a hydrothermal route is 4.40 eV.¹⁰ Hence, the conduction band bottom of rGO is at −4.40 eV with respect to the absolute vacuum (AV) level. The electron affinity of MoS₂ is 4.3 eV, and thus the conduction band bottom is at −4.3 eV vs. AV level and the valence band top is at −5.5 eV vs. AV level.⁴² As reported previously,⁴³ the redox potentials vs. normal hydrogen electrode (NHE) of a semiconductor in electrochemical scale can be converted from the values with respect to AV level by:

E_(NHE) = −E_(AV) − 4.50 (2)

So the redox potentials of the conduction band bottom and valence band top of MoS₂ is at −0.2 V and 1.0 V vs. NHE, respectively. The top of valence band of MoS₂ is more negative than the redox potentials of H₂O/˙OH (2.72 V vs. NHE) and OH⁻/˙OH (1.89 V vs. NHE),⁴⁴ so the ˙OH radicals is not available in the photocatalytic process. As the bottom of conduction band of MoS₂ is more positive than the standard redox potential of O₂/O₂˙⁻ (−0.28 V vs. NHE),¹² the photo-generated electrons in MoS₂ cannot be trapped by adsorbed O₂ to produce O₂˙⁻ radical. But the redox potentials of the excited RhB (RhB*) (−1.42 eV) is negative than that of O₂/O₂˙⁻ (−0.28 V), the energy of photogenerated electrons due to the dye-sensitization effect is enough to trap O₂ to produce O₂˙⁻ radicals. Therefore, the trapping experiments of O₂˙⁻ radicals and holes have been done to detect the active species during photodegradation process and to judge whether the dye-sensitization effect dominates the degradation process. Fig. 9 presents the results. As well known, nitrogen and methanol are commonly used as the scavenger of O₂˙⁻ radicals and holes in photocatalytic reaction, respectively. It can be found that the degradation efficiency decreases from 49.56% to 33.33% by 33% if nitrogen gas is bubbled into RhB solution, but to 22.20% by 55% if methanol is introduced into the solution. It means that both O₂˙⁻ radicals and photogenerated holes from MoS₂ are responsible for the photocatalytic activity, but the holes dominate the degradation process. Accordingly, Fig. 10 schematically illustrates the physical mechanism for enhanced photocatalytic activity of MoS₂/rGO composites. Because the energy level of rGO is slightly lower than the conduction band of MoS₂, the photo-generated electrons produced in MoS₂ will flow onto rGO sheets, and the rGO works as an electron sink to rapidly capture the photo-generated electrons,⁴⁵ thus the photo-generated holes are left on the MoS₂. The heterointerface between MoS₂ and rGO effectively retard the recombination of photo-induced electrons–holes pairs in MoS₂. Because the photogenerated electrons from MoS₂ could not trap O₂˙⁻ to produce O₂˙⁻ radicals, so the O₂˙⁻ radicals should come from the dye-sensitization effect in which the excited dye will supply electrons to graphene and MoS₂ sheets without any more reaction.^5,10,39 Moreover, the conduction band of MoS₂ is higher than that of rGO, the electron transfer from RhB* to rGO is more favorable. But the injected electrons on rGO are easily recombined with RhB˙⁺ radicals adsorbed on surface.^5,39 In spite of this, RhB* adsorbed on MoS₂ can provide electrons to MoS₂, some electrons will be separated onto graphene by the heterointerface, and some electrons unrecombined with RhB˙⁺ radicals will trap O₂ to produce O₂˙⁻ radicals. The heterointerface promotes the separation between the injected electrons and RhB˙⁺ radicals on MoS₂, and enhances the photocatalytic degradation. So the photogenerated holes in MoS₂ are mainly responsible for the photocatalytic degradation of RhB, and the sensitization effect of RhB molecules under visible light accelerates the photodegradation process further.

Fig. 9 The trapping experiments of O₂˙⁻ radicals and holes.

Fig. 10 Suggested mechanism for the photocatalytic degradation of RhB by MoS₂/rGO composites under visible light irradiation.

4. Conclusions

In summary, we have fabricated MoS₂/rGO heterostructures through a facile hydrothermal method. In the composites, MoS₂ clusters composed of several nanosheets are intertwined with rGO sheets tightly. As the amount of rGO in the composites is increased, the crystallinity of MoS₂ changes little, but the diameter of MoS₂ clusters decreases considerably and they become more dispersive, resulting in a large specific surface area. It is found that, as compared to bare MoS₂, the as-prepared MoS₂/rGO photocatalysts exhibit higher adsorptivity of RhB. The photo-generated holes are mainly responsible for the photocatalytic degradation of RhB and, the electrons from the sensitization effect of RhB molecules under visible light can trap O₂ to produce O₂˙⁻ radicals and accelerate the photodegradation process. The heterointerface between MoS₂ and rGO results in an efficient separation of photo-generated electron–hole pairs in MoS₂, and retards the recombination of between the injected electrons and RhB˙⁺ radicals. Consequently, the photocatalytic activity for degradation of RhB under visible light irradiation is substantially improved. It opens a new avenue for fabrication of heterostructure photocatalyst.

Acknowledgements

This work was jointly supported by National Natural Science Foundation of China (Grant No. 51271139, 51471130, 51302162), Fundamental Research Funds for the Central Universities.

References

1. L. Liu, W. Yang, Q. Li, S. Gao and J. K. Shang, ACS Appl. Mater. Interfaces, 2014, 6, 5629–5639.
2. M. Lu, C. Shao, K. Wang, N. Lu, X. Zhang, P. Zhang, M. Zhang, X. Li and Y. Liu, ACS Appl. Mater. Interfaces, 2014, 6, 9004–9012.
3. T. Xu, L. Zhang, H. Cheng and Y. Zhu, Appl. Catal., B, 2011, 101, 382–387.
4. X. Zhang, B. Zhang, D. Huang, H. Yuan, M. Wang and Y. Shen, Carbon, 2014, 80, 591–598.
5. Z. Xiong, L. L. Zhang, J. Ma and X. S. Zhao, Chem. Commun., 2010, 46, 6099–6101.
6. T. Hua, O. Shuxin, B. Yingpu, N. Umezawa, M. Oshikiri and Y. Jinhua, Adv. Mater., 2012, 24, 229–251.
7. A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C.-Y. Chim, G. Galli and F. Wang, Nano Lett., 2010, 10, 1271–1275.
8. K.-K. L. Dung-Sheng Tsai and D.-H. Lien, ACS Nano, 2013, 7, 3905–3911.
9. S. Wi, M. Chen, H. Nam, A. C. Liu, E. Meyhofer and X. Liang, Appl. Phys. Lett., 2014, 104, 232103.
10. J. S. Lee, K. H. You and C. B. Park, Adv. Mater., 2012, 24, 1084–1088.
11. M. Shen, Z. Yan, L. Yang, P. Du, J. Zhang and B. Xiang, Chem. Commun., 2014, 50, 15447–15449.
12. E. Gao, W. Wang, M. Shang and J. Xu, Phys. Chem. Chem. Phys., 2011, 13, 2887–2893.
13. Y. Chen, G. Tian, Y. Shi, Y. Xiao and H. Fu, Appl. Catal., B, 2015, 164, 40–47.
14. Z. Sun, J. Guo, S. Zhu, J. Ma, Y. Liao and D. Zhang, RSC Adv., 2014, 4, 27963.
15. Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 115, 7355–7363.
16. S. W. Hu, L. W. Yang, Y. Tian, X. L. Wei, J. W. Ding, J. X. Zhong and P. K. Chu, Appl. Catal., B, 2015, 163, 611–622.
17. T. N. Reddy, J. Manna and R. K. Rana, ACS Appl. Mater. Interfaces, 2015, 7, 19684–19690.
18. F. Meng, J. Li, S. K. Cushing, M. Zhi and N. Wu, J. Am. Chem. Soc., 2013, 135, 10286–10289.
19. J. Li, X. Liu, L. Pan, W. Qin, T. Chen and Z. Sun, RSC Adv., 2014, 4, 9647–9651.
20. J. Guo, X. Chen, S. H. Jin, M. M. Zhang and C. H. Liang, Catal. Today, 2015, 246, 165–171.
21. F. Xiong, Z. Cai, L. Qu, P. Zhang, Z. Yuan, O. K. Asare, W. Xu, C. Lin and L. Mai, ACS Appl. Mater. Interfaces, 2015, 7, 12625–12630.
22. D. S. Kong, H. T. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
23. F. Li, L. Zhang, J. Li, X. Lin, X. Li, Y. Fang, J. Huang, W. Li, M. Tian, J. Jin and R. Li, J. Power Sources, 2015, 292, 15–22.
24. M. Daage and R. R. Chianelli, J. Catal., 1994, 149, 414–427.
25. H. Li, K. Yu, X. Lei, B. Guo, C. Li, H. Fu and Z. Zhu, Dalton Trans., 2015, 44, 10438–10447.
26. Y. Li, H. Wang, L. Xie, Y. Liang, G. Hong and H. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.
27. T. Lin, J. Wang, L. Guo and F. Fu, J. Phys. Chem. C, 2015, 119, 13658–13664.
28. T. Y. Wang, D. L. Gao, J. Q. Zhuo, Z. W. Zhu, P. Papakonstantinou, Y. Li and M. X. Li, Chem.–Eur. J., 2013, 19, 11939–11948.
29. T. Lavanya, K. Satheesh, M. Dutta, N. V. Jaya and N. Fukata, J. Alloys Compd., 2014, 615, 643–650.
30. M.-Q. Yang, C. Han and Y.-J. Xu, J. Phys. Chem. C, 2015, 119, 27234–27246.
31. J. Qin, X. Zhang, Y. Xue, N. Kittiwattanothai, P. Kongsittikul, N. Rodthongkum, S. Limpanart, M. Ma and R. Liu, Appl. Surf. Sci., 2014, 321, 226–232.
32. H. Wang, L.-F. Cui, Y. Yang, H. Sanchez Casalongue, J. T. Robinson, Y. Liang, Y. Cui and H. Dai, J. Am. Chem. Soc., 2010, 132, 13978–13980.
33. Y. Liang, H. Wang, H. S. Casalongue, Z. Chen and H. Dai, Nano Res., 2010, 3, 701–705.
34. Z. Wang, T. Chen, W. Chen, K. Chang, L. Ma, G. Huang, D. Chen and J. Y. Lee, J. Mater. Chem. A, 2013, 1, 2202–2210.
35. Z. X. Gan, L. Z. Liu, H. Y. Wu, Y. L. Hao, Y. Shan, X. L. Wu and P. K. Chu, Appl. Phys. Lett., 2015, 106, 233113.
36. J. P. Wilcoxon, P. P. Newcomer and G. A. Samara, J. Appl. Phys., 1997, 81, 7934.
37. Y. Hou, Z. Wen, S. Cui, X. Guo and J. Chen, Adv. Mater., 2013, 25, 6291–6297.
38. Y. Zhang, Z.-R. Tang, X. Fu and Y.-J. Xu, ACS Nano, 2010, 4, 7303–7314.
39. J. Zhang, Z. Xiong and X. S. Zhao, J. Mater. Chem., 2011, 21, 3634–3640.
40. J. Du, X. Lai, N. Yang, J. Zhai, D. Kisailus, F. Su, D. Wang and L. Jiang, ACS Nano, 2011, 5, 590–596.
41. D. Sarkar, C. K. Ghosh, S. Mukherjee and K. K. Chattopadhyay, ACS Appl. Mater. Interfaces, 2013, 5, 331–337.
42. S. McDonnell, A. Azcatl, R. Addou, C. Gong, C. Battaglia, S. Chuang, K. Cho, A. Javey and R. M. Wallace, ACS Nano, 2014, 8, 6265–6272.
43. Y. Xu and M. A. A. Schoonen, Am. Mineral., 2000, 85, 543–556.
44. T. Tachikawa, M. Fujitsuka and T. Majima, J. Phys. Chem. C, 2007, 111, 5259–5275.
45. H. Yu, J. Tian, F. Chen, P. Wang and X. Wang, Sci. Rep., 2015, 5, 13083.

---

Footnotes

† Electronic supplementary information (ESI) available: XRD patterns of MoS₂/rGO-5 composites prepared in different time, the measurements of the average diameter of MoS₂ clusters, the aggregation of MoS₂ clusters in MoS₂/rGO-5 and MoS₂/rGO-10, N₂ adsorption–desorption isotherm curves of MoS₂, MoS₂/rGO-5, and MoS₂/rGO-10, SEM images of surface morphologies of MoS₂/rGO-5 composites at different reaction time, and the surface morphologies of the composites synthesized by using graphene as the matrix. See DOI: 10.1039/c6ra10923c

‡ Long Zhang and Lan Sun contributed equally to this work.

---