1. Introduction
Increasing environmental pollution and energy consumption are the two most serious problems facing humanity today, due to which, the development of techniques for the efficient utilization of solar energy has attracted much attention.1,2) As a renewable clean energy source, solar energy can be used to decompose water into hydrogen and oxygen over a photocatalyst, which can be further stored in the form of chemical energy or converted into a fuel.3-6) In view of their low cost, unique structure and excellent physicochemical properties, both TiO2 and MoS2 are of great interest as photocatalyst materials.7-9) However, there are inherent disadvantages when using TiO2 or MoS2 alone that severely limit their practical application, but it has been observed that TiO2 and MoS2 are highly complementary.10,11) This finding has led to much research effort in recent years to form MoS2/TiO2 composites to improve photocatalytic activity.12-16) However, although both TiO2 and MoS2 can be easily prepared using low-cost methods, the reported synthesis procedures for MoS2/TiO2-based composites are complex, expensive, and are not satisfactory in terms of performance.17) The most advantageous among these processes is that involving ex-situ synthesis in being low cost and is scalable. However, TiO2 and MoS2 in the obtained MoS2/TiO2-based composite material have weak interfacial interaction and their dispersion in the composite is highly inhomogeneous. Hence, a considerable fraction of MoS2 does not make sufficient contact with the TiO2 skeleton, which affects the stability and electrical conductivity of the composite.18) In in-situ synthesis, the most common methods used to make a strong interfacial contact between MoS2 and TiO2 in the prepared MoS2/TiO2 composite are through hydrothermal and solvothermal reactions. However, it has been observed that the MoS2/TiO2-based composite prepared by this method exhibits structural instability due to a lattice mismatch between TiO2 and MoS2. This results in a considerable amount of MoS2 that is not adequately contacted to TiO2, leading to an unstable interface during photocatalysis.19-20) The interface between the constituent components is an important criterion that affects the performance of composite materials; hence, adhesion at the interface must be strictly controlled in MoS2/TiO2-based composites. At present, the photocatalytic activity of MoS2/TiO2 composites is still low under visible light because of the lack of active catalytic reaction sites to enable the effective separation of electron-hole pairs. Therefore, several strategies have been employed to increase the photocatalytic hydrogen production efficiency of these composites, for example, by extending the light utilization range to the visible region, improving bonding at the interface or through additives such as graphene that can be supported on MoS2/TiO2 to form a ternary composite. Graphene (RGO) has a high specific surface area and excellent electrical conductivity and has therefore been widely used as a co-catalyst to prevent agglomeration and enhance the electron transfer ability of the catalyst. Therefore, the introduction of RGO into the MoS2/TiO2 composite is expected to effectively improve the photocatalytic activity.21-24) Xiang synthesized a TiO2/MoS2/RGO (T/MG) composite photocatalyst by a two-step hydrothermal method, in which, TiO2 nanoparticles were grown on the surface of layered MoS2/RGO (MG). Under the UV light irradiation, the photogenerated electrons on TiO2 could be transferred to both MoS2 and G promoters, which effectively improved the separation and migration efficiency of the charge carriers. As a result, the ternary T95M5G sample showed a high photocatalytic activity of 2066 μmol−1h−1, which was 4 times that of the binary T/100 M0G sample.15) Li and his team used glucose to improve contact between TiO2 and MoS2 to promote the adhesion of MoS2 to the TiO2 surface in MoS2@TiO2 composites.16) The strong interfacial interaction between MoS2 and TiO2 and the large-area contact significantly improved stability and resulted in effective charge transfer. In this study, we have developed a method to increase the transmission efficiency of photogenerated electron holes by adsorbing MoS2 on the surface of graphene. Moreover, the morphology and interfacial properties of TiO2 were modified by hydrofluoric acid using glucose and a surfactant as chelating agents. These two modifications enhanced interaction at the MoS2/G-TiO2 interfaces and greatly improved both H2 production efficiency and stability of the MoS2-G/TiO2 composite under visible light.
2. Experimental Procedure
2.1. Materials
Sodium molybdate dihydrate (Na2MoO4·2H2O), thiourea (H2NCSNH2), hydrochloric acid (HCl), polyvinylpyrrolidone ((C6H9NO)n), sodium dodecylbenzenesulfonate (C18H29NaO3S), tetrabutyl titanate (C16H36O4Ti), diethanolamine (C4H11NO2), ethanol (C2H6O) and glacial acetic acid (C2H4O2) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Graphene oxide slurry was obtained from Shan dong Yuhuang New Energy Technology Co., Ltd. All reagents were of analytical grade and no further purification was required. Distilled water was further purified before use.
2.2. Synthesis of MoS2/G Photocatalyst
Sodium molybdate (0.363 g) and thiourea (0.324 g) used as molybdenum and sulfur sources, respectively, were mixed with a known quantity of graphene oxide and dissolved in 20 mL of 1.48 mmol/L sodium dodecylbenzene sulfonate solution in water. Hydrochloric acid (3 mmol) of was added dropwise to this solution After stirring for 30 min, the resulting solution was transferred to a Teflon-lined autoclave (30 mL) and heated at 220°C for 20 h. The obtained precipitate was collected by centrifugation, washed three times with distilled water and ethanol and dried in an oven at 60°C for 12 h.
2.3. Synthesis of MoS2/G-TiO2 Photocatalyst
MoS2/G-TiO2 photocatalyst was synthesized by a hydrothermal method. In the first step, MoS2/G (24 mg), glucose (50 mg) and 0.9 g of polyvinylpyrrolidone (PVP) were added to a mixture of acetic acid (4 mL) and ethanol (l6 mL). Next, 1 mL of hydrofluoric acid was added dropwise and stirred for 20 min using a magnetic stirrer. The solution was then sonicated for 45 min using an ultrasonic cell grinder (1000 W). Tetrabutyl titanate (1.7 mL) and diethanolamine (0.4 mL) were added to ethanol (7.5 mL) and stirred for 30 min, and this solution was added dropwise to the MoS2/G mixed solution under magnetic stirring and left stirring for 45 min. Finally, the mixed solution was transferred to a polytetrafluoroethylene-lined autoclave (30 mL) and heated to 180°C for 10 h. The obtained precipitate was collected by centrifugation, washed three times with distilled water and ethanol and dried in an oven at 60°C for 12 h. The samples were labeled as 1GT-2M, 1GT-4M, 1GT-5M, 1GT-6M, 1GT-8M, and 1GT-10M, depending on the MoS2 content. Samples with different G contents were labeled 5MT-0.25G, 5MT-0.5G, 5MT-0.75G, 5MT-1G, 5MT-1.25G, and 5MT-1.5G.
2.4. Characterization
Crystal structures were determined by X-ray powder diffraction (XRD) using a Bruker D8 Advance diffraction with Cu Kα radiation (λ = 1.5406 Å, 40 keV, 40 mA). Raman measurements were performed using a Via Reflex micro-Raman spectrometer with excitation at a wavelength of 532 nm. The size and morphology of the samples were investigated by scanning electron microscopy (SEM) (JOEL, JSM-7500F). Ultraviolet (UV)-visible absorption spectra were taken with the help of a UV-vis spectrophotometer (SolidSpec-3700, Japan). The laser beam was focused by a 50× objective lens to a ~ 1 μm spot on the surface of the sample. Specific surface areas were determined from N2 adsorption isotherms measured at 77 K (JW-BK132F) using the Brunauer-Emmett-Teller (BET) equation
2.5. Photocatalytic hydrogen production performance
An On-line photocatalytic hydrogen production system (AuLight, Beijing, and CELSPH2N) was used to measure the photocatalytic hydrogen production at the ambient temperature of 20°C. Photocatalytic H2 production experiments were carried out in a closed gas circulation and evacuation system fitted with a top Pyrex window. The photocatalyst (25 mg) was dispersed in 50 mL of methanol-water mixture (containing 10 mL of methanol and 40 mL of deionized water). Photocatalytic experiments were conducted in a single compartment Pyrex reactor of volume ~ 196 cm3 having a flat window ~ 19.6 cm2 area for illumination. A 300-W Xe lamp (545 mW/cm2) equipped with an optical cutoff filter of 420 nm was employed for visible-light excitation; the intensity of the light source was estimated to be 180 mW/cm2. Gas evolution was observed only under photo-irradiation, and the evolved gases were analyzed using an online gas chromatograph (SP7800, TCD, 5 Å molecular sieve, N2 carrier gas). To evaluate the stability of the photocatalyst, the photocatalyst was separated from the suspension after the first 8 h of hydrogen production run, washed with water, and dried at 60°C. The recovered photocatalyst was then used in the next hydrogen production run under the same conditions.
3. Results and Discussion
3.1. Characterization of catalysts supports
3.1.1. XRD analysis
The results of XRD analysis of MoS2, TiO2, 5MT and 5MT-1G are shown in Fig. 1. In the XRD pattern of 5MT and 5MT-1G, there are several strong and narrow peaks corresponding to (002), (101), (103) and (105) crystal planes of MoS2 (standard PDF card, No. JCPDS37-14920) indicating the high crystallinity of the nanosheets.25-28) The diffraction peaks at 26.5°, 37.5° and 47.8° can be attributed to reflections from the (101), (004) and (200) crystal planes of TiO2 (standard PDF cards JCPDS75-1621). Diffraction peaks due to carbon species in the photocatalyst could not be observed due to its low mass ratio and the low diffraction intensity of G.29-31) Therefore, Raman spectroscopy was used to analyze the composition of the material.32)
3.1.2. Raman spectral analysis
The Raman spectrum of pure MoS2, which is taken to be the reference sample, is shown in Fig. 2. Raman spectra of both 5MT and 5MT-1G show the two characteristic peaks at 387 and 406 cm−1 corresponding to MoS233) originating from E12g and A1g vibration modes, which represent, respectively, the inter-layer displacement of S atoms and the outward symmetric displacement of S atoms along the c-axis. In addition, the frequency difference between the two characteristic Raman peaks is - 19 cm−1, which confirms that the MoS2 sheet is very thin.34-36) Peaks at 151, 514 and 635 cm−1 are characteristic of TiO2.15) As compared to 5MT, two more peaks of 1350 cm−1 and 1600 cm−1 can be seen in the 5MT-1G sample, which are characteristic peaks of G. Raman spectra thus demonstrate that the composite MoS2/G-TiO2 photocatalyst was successfully prepared.33)
3.1.3. SEM characterization
SEM images (Fig. 3(a)) show that graphene oxide is a large area nanosheet where self-assembled MoS2 spheres are seen adsorbed on the graphene surface, the edges of the graphene sheet are slightly folded (Fig. 3(b)). The MoS2 spheres are self-assembled from MoS2 monolayers and have uniform diameters in the range 50-100 nm. An ultra-thin nano-MoS2 layer was clearly observed in Fig. 3(c), indicating that the MoS2 crystal form in the MoS2/G material also grows along the sheet. In Fig. 3(d), the morphology of the 5MT-1G catalyst can be seen, which consists of a sphere composed of MoS2 sheets and graphene that are uniformly dispersed in the TiO2 matrix during ultrasonic dispersion to form a perfectly mixed MoS2 /G-TiO2 photocatalyst.
3.1.4. UV-vis spectroscopic analysis
Figure 4 shows the UV-vis absorption spectra of the catalysts. All samples absorbed both UV and visible light as expected, confirming the photocatalytic activity of the composite catalyst in visible light. The pure TiO2 sample shows an absorption edge at ~ 387 nm corresponding to a band gap of 3.2 eV15) and the absorption onset of MoS2 is at 461 nm. The exciton peak at 688 nm corresponding to the band gap of 1.8 eV emitted by brillouin region K in MoS2 can be clearly observed in the composite materials and in pure MoS2, proving the successful surface modification of TiO2 by MoS2.25) As compared to pure TiO2, the absorption of 5MT, 1GT, and 5MT-1G is stronger in the visible range. Upon the addition of MoS2 and graphene, a red shift in the absorption edge occurs accompanied by increase in absorption strength.
3.1.5. Surface area of the catalysts
Table 1 lists the specific surface area values of the catalysts. As-prepared 1GT has a surface area of 69 m2/g, while the surface area of 5MT is found to be 101.7 m2/g, thus showing that the SSA values are increased, respectively, by 5.7 and 8.4 times, as compared to that of pure TiO2 (12.2 m2/g). A dramatic increase in surface area is observed for layered MoS2 and RGO-embedded TiO2 systems. Thus, the specific surface area of the 5MT-1G photocatalyst is as high as 205.8 m2/g. Combining MoS2 and RGO improves the structure of TiO2, leads to a favorable synergistic effect, creates more reactive sites, increases visible light absorption, and results in better charge separation and electron transfer efficiency.
3.2. Analysis of the photocatalytic hydrogen evolution activity and its mechanism
As shown in Fig. 5(a) and 5 (b), we have studied the effect addition of MoS2 and G of different mass loading on the photocatalytic hydrogen production activity of TiO2. The photocatalytic efficiency of pure TiO2 under visible light is only 176 μmolg−1h−1 and the photocatalysts 1GT and 5MT showed slightly higher photocatalytic activity, with H2 generation rates of 401 μmolg−1h−1 and 557 μmolg−1h−1, respectively. Photocatalytic hydrogen production is increased with increase in both MoS2 and G content. It can be speculated that the quantum confinement effect of MoS2 promotes charge separation and the excellent charge transfer capacity of G enhances the photocatalytic activity. When the MoS2 content is 5 wt.% with 1 wt.% G, the H2 generation rate reaches a maximum of 1989 μmolg−1h−1, which is 12.9 times higher than that of TiO2. Further increasing the content of MoS2 in the catalyst results in a decrease in photocatalytic activity, this could be due to the higher MoS2 and G concentration in TiO2, which is not conducive to charge separation and photo-generated electron migration.
A tentative mechanism is proposed to explain the high H2 production rate of the MoS2/G-TiO2 photocatalyst, as illustrated in Fig. 6. Under visible light, photo-generated electrons in TiO2 are excited and transferred onto the surface. Being a good two-dimensional layered conductor, the redox potential of graphene is slightly lower than the CB of anatase TiO2. Graphene can combine with the cluster formed from nano-layered MoS2 to enhance its electrical conductivity. The cluster formed by self-assembly of nano MoS2 layer has a large number of exposed edges and unsaturated active S atoms and therefore acts as a good co-catalyst. The photogenerated electrons in the CB of TiO2 can be transferred to the clusters in nano MoS2 layers through the graphene sheets (which act as a conductive electron transport “highway”) enabling fast charge transfer (Fig. 6). Some of the electrons approaching the edge of MoS2 react directly with adsorbed H+ in H2O to produce H2 due to the presence of unsaturated active S atoms in the co-catalyst, which can accept electrons and act as active sites for H2 generation. Furthermore, the stability of 5MT-1G was tested by repeating the photocatalytic H2 production four times (Fig. 5(c)) and the change in hydrogen production was less than 10%. This result demonstrates the high stability and excellent catalytic activity of our composite catalyst in photocatalytic hydrogen production.
4. Conclusions
The MoS2/G-TiO2 photocatalyst prepared in this work has a high visible light catalytic hydrogen production activity and excellent stability. Results show that the special heterojunction structure in the composite enhances the separation efficiency of photogenerated carriers and improves charge transfer efficiency. Moreover, the high specific surface area effectively increases the number of active sites at the reaction sites and also tailors the forbidden band width to increase visible light utilization.