Abstract

SnS2 and Sn1−xTixS2 ( = 0, 0.1, 0.3, 0.5, and 0.7 mol) materials were designed using solvothermal method with the aim to enhance hydrogen production from water/methanol water photosplitting. Scanning electron microscopy revealed hexagonal plates with one side, 3.0 μm in length, in the SnS2 materials. Pure SnS2 showed absorption band edges of above 660 nm, and the absorption was shifted to low wavelengths with the insertion of Ti ions. The evolution of H2 from MeOH/H2O (1 : 1) photosplitting over SnS2 hexagonal plates in the photocatalytic liquid system was 0.016 mL h−1 g−1, and the evolutions were enhanced in Sn1−xTixS2. In particular, 0.049 mL h−1 g−1 of H2 gas was produced in Sn0.7Ti0.3S2 without electrolytes and it increased significantly to more than 90.6% (0.47 mL h−1 g−1 evolutions) at higher pH using 0.1 M of KOH. Based on the UV-visible absorption spectra, the high photocatalytic activity of Sn1−xTixS2 was attributed to the existence of an appropriate band-gap state that retarded recombination between the electrons and holes.

1. Introduction

Many attempts have been shown to produce hydrogen as a renewable energy carrier to satisfy future demands because of its versatility and friendly properties. With this in mind, the photocatalytic formation of hydrogen and oxygen on semiconductors, such as pure TiO2 (anatase) [1, 2] and metal loaded-TiO2 [36] or non-metal loaded-TiO2 [79], has been studied extensively because of their relatively low band gap and high corrosion resistance. On the other hand, these materials are activated only by UV because of their large energy band gap (e.g., 3.2 eV for anatase TiO2). To improve the utilization of solar energy, considerable research effort has focused on shifting the photocatalytic hydrogen producing activity of TiO2 into the visible wavelength above 450 nm, which accounts for ~42% of solar energy. Recently, studies of metal sulfide photocatalysts, particularly ZnS, Bi2S3, and CdS-loaded TiO2, have covered topics ranging from synthesis to applications in new photocatalytic reaction mechanisms [1015]. A narrow band gap makes it possible to absorb longer wavelengths compared to the wide band gaps of conventional metal oxide semiconductor systems.

Some researchers have reported SnS2 materials. In particular, nanostructures of SnS2 often demonstrated shape and size-dependent physical and chemical properties that are of technological and scientific importance [16, 17]. Consequently, a great deal of effort has been focused on designing methods for the synthesis of SnS2 with different characteristics [18, 19] to exploit their potential. SnS2 has abundant optical, electrical, and photoelectric properties [20, 21] and has a band gap of 2.25 eV [22]. Owing to its photoconductivity, it is also considered a prospective candidate for solar cells and optoelectronic devices [20, 21]. Crystals of SnS2 consist of planar triple layers (sandwiches, S–Sn–S with strong ion-covalent bonding) that are coupled weakly to one another by van der Waals forces. One of the most salient properties of layered SnS2 is their ability to act as a host for atomic and molecular guest species, which are accommodated at the empty sites bounded by van der Waals forces between the adjacent close packed chalcogen layers [23]. Owing to this property, lithium can be inserted into SnS2, meaning that SnS2 can become a promising candidate as a cathode material in the preparation of lithium batteries [2426]. In addition, the absence of surface states associated with dangling bonds, which can act as recombination centers for photoinjected electrons, makes SnS2 an ideal model system for dye sensitization studies and contributes to the high quantum yields [27]. Recently, the application of SnS2 is shifting to its photocatalytic activity. Yang et al. [28] attributed the greatly enhanced photocatalytic activity of the /TiO2 composites to the matching band potentials and efficient charge transfer and separation at the tight-bonding interface between and TiO2. On the other hand, there are few reports of SnS2 for photocatalytic hydrogen production, particularly the replacement of the SnS2 framework with Ti ions (S2).

In this study, the SnS2 submicron hexagonal plate shaped particle was synthesized by a solvothermal method. Ti ions were incorporated into SnS2 at various molar ratios (Sn1−xTixS2) to regulate the potential energies of the valence and conduction bands for efficient hydrogen production from MeOH/H2O photosplitting. The relationship between their spectroscopic properties and the photocatalytic performance on the production of H2 is discussed and the characteristics of SnS2 and Sn1−xTixS2 were determined by X-ray diffraction (XRD), transmission electron microscopy (TEM), UV-visible absorption spectroscopy, and photoluminescence (PL) spectroscopy.

2. Experimental

SnS2 and Sn1−xTixS2 were prepared using a solvothermal method, as shown in the experimental flowchart in Scheme 1. To prepare the sol mixture, SnCl4, sulfur powder, and TiCl4 were used as the Sn, S, and Ti precursors, respectively. First, moles ( is Ti concentration, , 0.1, 0.3, 0.5, and 0.7 mol) of SnCl4 were dissolved in N, N-dimethylformamide as a solvent. 2 moles of S powders and 0, 0.1, 0.3, and 0.5 moles of TiCl4 were added slowly to the solution with constant stirring and stirred to homogeneity for 2 h. Subsequently, the final solution was transferred to an autoclave for the thermal treatment. Sn and Ti ions were sulfurized during thermal treatment at 200°C for 24 h in a nitrogen environment. The resulting precipitates were obtained, washed with acetone, and dried at 50°C for 24 h. Finally, four different materials, SnS2, Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, and Sn0.5Ti0.5S2 were prepared.

479508.sch.001

The synthesized SnS2 and Sn1−xTixS2 particles were examined by XRD (X’Pert Pro MPD PANalytical 2-circle diffractometer at the Yeungnam University Instrumental Analysis Center) using nickel-filtered CuKα radiation (30 kV and 30 mA). The sizes and shapes of the SnS2 and Sn1−xTixS2 particles were examined by field emission scanning electron microscopy (FE-SEM, S-4100, Hitachi). The UV-visible absorption spectra of SnS2 and Sn1−xTixS2 particles were obtained using a Cary 500 spectrometer with a reflectance sphere. PL spectroscopy was also performed to determine the number of photoexcited electron hole pairs using a photoluminescence mapping system (LabRamHR, Sci-Tech Instruments).

Photosplitting in a MeOH/H2O solution was carried out using a liquid photoreactor designed in our laboratory, as shown in Figure 1. For water photosplitting, 0.5 g of SnS2 or Sn1−xTixS2 particles were added to 1.0 L distilled water in a 2.0 L Pyrex reactor. UV-lamps (6 × 3 W cm−2 = 18 W cm−2, 30 cm length  ×  2.0 cm diameter; Shinan Com., Korea) emitting radiation of 365 nm were used. Methanol/water (MeOH/H2O) photosplitting was carried out for 10 h with constant stirring, and hydrogen evolution was measured at an interval of 1 h. The concentration of each added electrolyte was 0.1 moles. The hydrogen gas (H2) produced during water photosplitting was analyzed by TCD-type gas chromatography (GC, model DS 6200; Donam Instruments Inc., Korea). To determine the products and intermediates, the GC was connected directly to the water decomposition reactor. The following GC conditions were used: TCD detector; Carbosphere column (Alltech, Deerfield, IL, USA); 140°C injection temp: 30°C initial temp.; 150°C final temp.; 300°C detector temp.

3. Results and Discussions

Figure 2 shows XRD patterns of the SnS2 and Sn1−xTixS2 particles. The XRD peaks in the SnS2 materials corresponded to pure hexagonal phase SnS2 (Berndtite-2T, P-3m1, JCPDS card number 23-677), whereas those of Sn1−xTixS2 revealed the formation of SnS2 and anatase TiO2 (tetragonal phase, JCPDS card number 65-5714) mixtures. The XRD patterns in hexagonal phase SnS2 showed the main peaks at 15.0, 28.3, 32.2, 42.2, 50.1, 52.6, 55.3, 60.9, 67.3, 70.5, 82.4, and 88.4° 2θ, which were assigned to the (001), (100), (101), (102), (110), (111), (103), (201), (202), (113), (211), and (212) planes, respectively. In the case of Sn0.9Ti0.1S2, although there was very little reduction in intensity, the peak positions for the SnS2 structure were not changed and no peaks assigned to TiO2 were observed despite the addition of Ti ions to the framework because of the very small amount. The peaks for TiO2 at 25.38, 38.08, 48.28, 54.8, and 63.8° were assigned to the (101), (004), (200), (105), and (204) crystal planes in Sn0.7Ti0.3S2 [29]. The peaks for Sn0.5Ti0.5S2 were very weak, which might be caused by structural damage to the SnS2 hexagonal phases depending on the level of substitution of Ti ions. This result can be predicted as follows more specifically. The sample nonadded Ti ingredient (SnS2) exhibits a complete hexagonal crystal structure. However, the SnS2 framework turns into a new crystal structure (orthorhombic SnS) with an increase of Ti concentration. Sn0.9Ti0.1S2 including Ti 0.1 moL presented to the same framework of SnS2; however two crystal structures (SnS2 and SnS) coexisted in Sn0.7Ti0.3S2 with Ti concentration increasing to be 0.3 moL. Here the crystal growth directions in two structures are different, and thus the two crystals are competitively grown, and eventually the peak intensities in XRD patterns are decreased. At this time, the anatase TiO2 crystals rapidly formed between Ti components which not fully inserted into SnS2 framework are shown in Sn0.5Ti0.5S2 and Sn0.7Ti0.3S2 samples. It is maybe attributed to too slow crystal growth of SnS2 and SnS. The result was also similar in Sn0.5Ti0.5S2. Otherwise, the Sn0.3Ti0.7S2 expressed an orthorhombic SnS crystal structure; thus the crystal growth rate increased and finally the strong peak intensities in XRD patterns were shown in Sn0.3Ti0.7S2. On the other hand, the full width at half maximum (FWHM) of the 25.38 2θ peak for Sn0.7Ti0.3S2 and Sn0.5Ti0.5S2 was estimated. The Scherrer’s equation was used to estimate the crystallite size [30]. The calculated crystallite sizes of the Sn0.7Ti0.3S2 and Sn0.5Ti0.5S2 were 13.7 and 10.0 nm, respectively. Surprisingly, the diffraction peaks of the Sn0.3Ti0.7S2 were rather similar to the orthorhombic SnS (JCPDS card, number 01-0984, Pmcn, = 3.99 A, = 4.34 A, = 11.2 A) structure. This suggests that the hexagonal SnS2 structure can be transferred to an orthorhombic SnS structure when many Ti ions are inserted.

Figure 3 shows low-magnification SEM images of SnS2 and Sn1−xTixS2 particles. SEM revealed hexagonal plates of one side 3.0 μm in the SnS2 materials. In contrast, the morphologies of the Sn1−xTixS2 materials varied according to the level of Ti ions insertion: broken sheets, coexisting forms with TiO2 nanoparticles, and peanut-shaped have been appeared in Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, and Sn0.5Ti0.5S2, respectively. Otherwise, microrods with a similar morphology to SnS were observed in Sn0.3Ti0.7S2. On the other hand, the specific surface areas of SnS2, Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, Sn0.5Ti0.5S2, and Sn0.3Ti0.7S2 exhibited 38.72, 44.24, 54.20, 68.52, and 48.30 cm2/g, respectively. Typically, it is well known that the surface area differs on the particle shape and size. However, the specific surface area values in this study seem to be related to the bulk pores formed by the contacts between the particles rather than the particle size and shape. Especially the larger surface area in the case of Sn0.5Ti0.5S2 sample can be considered by bulk pores much formed between the smaller and round-shaped particles. Generally, the larger surface area has, the catalytic performance increases. However, the surface area in our study was not directly related to the catalytic activity. The surface area in Sn0.7Ti0.3S2 was greater than Sn0.5Ti0.5S2; however, the photocatalytic activity (hydrogen production) was more excellent in Sn0.5Ti0.5S2. It can be concluded in this study that the proper band gap and light absorption ability of samples were more advantageously affected to the catalytic activity.

Figure 4 shows the UV-visible diffuse reflectance spectra of the as-synthesized SnS2 and Sn1−xTixS2 powders. All products displayed optical absorption capabilities over the entire visible light spectrum (400–700 nm). The broad spectrum response suggests that the as-synthesized SnS2 hexagonal plates should be an excellent visible light responsive photocatalyst for MeOH/H2O splitting. With Ti ion insertion, the curves were shifted to shorter wavelengths. The optical band-gaps were determined based on the theory of optical absorption for direct band gap semiconductors [31]: . The curves of versus for the as-synthesized SnS2 hexagonal plates were plotted. By extrapolating the straight line portion of the plots of versus to , the Eg values of the as-synthesized SnS2, Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, and Sn0.5Ti0.5S2 were estimated to be 1.87 (670 nm), 1.88 (660 nm), 1.90 (650 nm), and 2.25 (550 nm) eV, respectively. Generally the band gaps of SnS2 and SnS are 2.2 and 1.08 eV [32]. However, the absorption in the visible region was attributed to the transition from the ground state to a few defect related deep states. This excitation character of the absorption spectra indicated the excellent crystal quality of the semiconductor. A red-shift is observed in the optical absorption spectra of the SnS2 hexagonal columns, which indicates that the SnS2 particles were too large to show quantum confinement related effects. This absorption in the visible region was attributed to the transition from the ground state to a few defect related deep states. Consequently, the band gap can be varied according to the particle size or crystal defects.

Photoluminescence (PL) spectroscopy measures the spectrum emitted by the recombination of photogene-rated minority carriers and is a direct way of measuring the band gap. On the other hand, the large quantity of impurities induces a large free carrier density in the bands. Consequently, different carrier interactions cause remarkable modifications of the line shape and spectral energy of the PL features. Figure 5 shows the PL spectra of SnS2 and Sn1−xTixS2 taken at room temperature. The spectra exhibited a strong emission peak at 549.9 nm corresponding to green emission. The strong PL peaks might be related to crystalline defects induced during growth. SnS2 exhibits a luminescence peak for near band emission at 549.78 nm (2.25 eV). The intensities of PL curves of samples were smaller with an increase of Ti insertion, and those were decreased in the following order: SnS2 < Sn0.9Ti0.1S2 < Sn0.5Ti0.5S2 < Sn0.7Ti0.3S2. Otherwise, the emission bands of TiO2 (at 420 nm) were not observed because SnS2 is a real light sensitizer. Typically, a smaller PL intensity indicates better photoactivity [33].

Figure 6 summarizes the evolution of H2 from MeOH/H2O photosplitting over the SnS2, Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, and Sn0.5Ti0.5S2 photocatalysts in a batch-type liquid photosystem. The catalytic activity of Sn1−xTixS2 was enhanced considerably compared to that of pure SnS2. 0.08 mL of H2 gas was collected over 0.5 g SnS2 photocatalyst after MeOH/H2O photosplitting for 10 h. SnS2 easily absorbed longer wavelengths in UV-visible absorption but the recombination between the excited electrons and holes also rapidly generated rapid catalytic deactivation. In contrast a significant amount of H2 gas was collected over the Sn1−xTixS2, and the amount of H2 produced reached 0.245 mL (0.049 mL h−1 g−1) over 0.5 g Sn0.7Ti0.3S2. Sn1−xTixS2 will show stronger oxidation-reduction ability than pure SnS2 with decreased electron-hole recombination due to the wider band gap, which increases the photocatalytic performance. This can be related to the PL result. The PL intensity of Sn0.9Ti0.1S2 sample was larger, and this fact means that the recombination between holes and electrons in Sn0.9Ti0.1S2 sample is larger than those in Sn0.5Ti0.5S2. The holes and electrons generated from Sn0.9Ti0.1S2 eventually did not play well in catalytic reaction. Consequently, the recombination time between electrons and holes for sample with mixed structure is expected to be longer rather than a sample having a perfect crystal structure like as SnS or SnS2. In particular, the catalytic performance for MeOH/H2O decomposition over Sn0.7Ti0.3S2 was enhanced further in the alkali solution electrolytes (B). Hydrogen production was increased dramatically in the alkali solution (NaOH and KOH) due to the generation of more OH radicals from alkali compounds, and it is reaching up to 0.47 mL h−1 g−1 evolution in a KOH solution. In acidic solutions as like H2SO4 and CH3COOH, the hydrogen production decreased because the unstable S ions in the Sn0.7Ti0.3S2 catalyst which dissolved in acidic solution oxidized with O ions in water and resulted in forming ions. The ions combined with the hydrogen ions generated during the MeOH/H2O photosplitting process to form H2SO4, and thus the hydrogen production was decreased. Otherwise, when neutral Na2SO4 compounds are added into the reaction solution, there is already the excess amount of ions, and thus ions formations are suppressed by the principle of Le Chatelier. Consequently, the catalytic activity in Na2SO4 solution relatively increased compared to it in the solution with nonadditives.

The UV-visible absorption and PL spectra indicated two photocatalysis models, as shown in Scheme 2. In model (a), combination between TiO2 and SnS2 was perfectly formed as in the Sn0.9Ti0.1S2 composite, despite the initial photoreaction being slower because of the larger band gap but the recombination of electrons and holes was slower, resulting in an increase in catalytic performance. In model (b), if TiO2 particles are physically loaded on the surface of SnS2 particles, the first electronic transition would occur in the SnS2 semiconductors from the 365 nm radiation source, and the electrons at a higher conduction band in SnS2 move to the TiO2 conduction band, whereas the holes in TiO2 move to the valence band of SnS2, which is formed by the current recycle. Therefore, the recombination of excited electrons and holes will be suppressed during photocatalysis. These phenomena increase the evolution of OH radicals formed from the electrons and holes that will eventually enhance the photocatalytic performance of MeOH and H2O degradation. Therefore, the Sn1−xTixS2 composites exhibited better catalytic performance than the TiO2 and SnS2 monocatalysts.

479508.sch.002

4. Conclusions

This reported the development of new photocatalysts using a metal sulfide framework for hydrogen production from MeOH/H2O splitting. SnS2 and Sn1−xTixS2 exhibiting activity in the visible radiation were synthesized using a solvothermal method. UV-visible absorption spectroscopy revealed estimated band gaps of 1.87, 1.88, 1.90, and 2.25 eV for the as-synthesized SnS2, Sn0.9Ti0.1S2, Sn0.7Ti0.3S2, and Sn0.5Ti0.5S2, respectively. The PL intensity of SnS2 and Sn1−xTixS2 decreased with the addition of Ti and was smallest in Sn0.7Ti0.3S2. A significant amount of H2 gas was collected over the Sn0.7Ti0.3S2 photocatalyst, and the amount reached 0.049 mL h−1 g−1 without an electrolyte with a maximum yield of 1.88 mL after 8 h in a KOH solution. This is due most likely to the slower recombination of electrons and holes, which enables higher catalytic performance for Sn0.7Ti0.3S2.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgment

This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Ministry of Education, Science and Technology (no. 2012R1A1A3005043), for which the authors are very grateful.