Tandem Photocatalysis of Graphene-Stacked SnS2 Nanodiscs and Nanosheets with Efficient Carrier Separation.

Being an optical semiconductor, tin disulfide (SnS2) attracts increasing interest in the fields of heterogeneous photocatalysis and photovoltaics. However, support from a graphene sheet in the form of a nanocomposite is expected to increase the stability and effectiveness of a SnS2 material in potential applications. We report here novel nanocomposites of graphene-oxide-stacked hexagonal-shaped pristine SnS2 nanodiscs (NDs of two different sizes) and nanosheets synthesized using an in situ one-pot hydrothermal synthesis process and the application of the nanocomposite as an efficient heterogeneous photocatalyst. The as-synthesized morphology-oriented nanoparticles and nanocomposites were comprehensively characterized, and finally, excellent photocatalytic activity of reduced graphene oxide/SnS2 nanocomposites under visible-light irradiation was analyzed using UV-vis spectroscopy, high-performance liquid chromatography, and gas chromatography. While precisely manipulating the nanocomposite formation, we observed efficient visible-light-driven photocatalytic application of graphene-stacked SnS2 NDs in the quantitative synthesis of aniline (99.9% yield, absolute selectivity) from nitrobenzene (>99.9% conversion), in the reduction of toxic Cr(VI) to nontoxic Cr(III), and in the degradation of mutagenic organic dyes. A possible synergetic electrical and chemical coupling leads to effective carrier separation in the semiconductor and charge transport in the nanocomposite, which finally gives rise to efficient tandem photocatalysis reactions.

Introduction

Graphene analogues of various inorganic layered materials with two-dimensional (2D) morphology, mainly the transition metal chalcogenides, have attracted considerable interest in recent years.[1,2] Chemically modified graphene has many uses, such as in sensors, energy materials, paper-like materials, field-effect transistors, and biomedical applications, because of its excellent electrical, thermal, and mechanical properties.[3−5] Among various transition metal chalcogenides, MoS2,[6] MoSe2,[7] WS2,[8] WSe2,[9] VS2,[10] and HfS2[11] have layered structures preferably used in the formation of binary nanocomposites with graphene or reduced graphene oxide (RGO) for various applications. Tin disulfide (SnS2) is another interesting semiconductor material with a layered structure analogous to graphene, showing acceptability in potential applications such as sensing, anodic material, photovoltaics, and energy storage.[12−14] SnS2 exists in the CdI2 structure and can be synthesized in a wide range of morphologies;[15] moreover, because of the numerous vacant sites present in the sandwiched structure of SnS2, it can be an important host material for various intercalation reactions. This metal disulfide with a moderate optical band gap usually has light-absorbing capabilities in the visible-light region, thus serving as a promising class of sensitizer for solar-light-driven photocatalysts. In terms of its lower toxicity and wider spectral response, SnS2 is considered as a potential material for photocatalytic activity as well.[16−18]

However, it would be interesting if nanoparticles (NPs) of layered SnS2 could be embedded in layered graphene oxide sheets to make nanocomposites and to extend the scope of applications that use nanocomposites.[19,20] In this scenario, exciton generation and charge separation in the SnS2 semiconductor and charge transport in graphene should result in a profitable synergetic effect.[21−23] This could be a new concept for accessing most of the active sites of the semiconductor’s NPs and the functional groups of partially reduced GO for solar light harvesting and utilizing the excitons generated thereof for multiple photocatalytic reactions. A high surface area is expected for this system, which helps with the adsorption of substrate molecules or reactants for photocatalytic reactions.

Herein, we report a low-cost and environmentally friendly one-pot hydrothermal approach for the synthesis of 2D SnS2 nanodiscs (NDs) and nanosheets (NSs) and also developed an in situ synthesis method to make their binary nanocomposites with RGO (see Scheme ). The as-synthesized nanocomposites (RGO/SnS2) were found to give an excellent efficiency in the photocatalytic synthesis of aniline from nitrobenzene (NB) (complete conversion of the substrate and absolute selectivity of a product, quantitative yield), degradation of dyes [methylene blue (MB) and rhodamine blue (Rh B)], and reduction of carcinogenic Cr(VI) to nontoxic Cr(III) in visible-light exposures under ambient conditions. Hence, these visible-light-driven multiple photocatalytic reactions make our RGO/SnS2 (GSnS2) ND and NS samples promising heterogeneous photocatalysts for multiple important environmental remediation reactions, including the removal of carcinogenic and mutagenic dyes and toxic Cr(VI), the removal of volatile NB from polluted water, or the production of aniline under green conditions. Nevertheless, ND morphology is found to be more effective than NS morphology in the present study.

Scheme 1

Schematic View of the Synthesis of an RGO/SnS2 Nanocomposite by a One-Pot Hydrothermal Approach in the Presence of Either SDS or AOT Surfactant

Results and Discussion

A hydrothermal approach to synthesize SnS2 particles in the presence of an anionic surfactant, sodium dodecyl sulfate (SDS), gives a hexagonal ND-like morphology but with a wide particle size distribution. Thermal dissociation of thioacetamide (TAA) in an aqueous solution gives sulfur ions at high temperature (180 °C),[24] which would react with Sn4+ for the nucleation of SnS2. SDS plays an important role in the formation of the 2D irregular hexagonal ND-like morphology of SnS2 particles by considerably accelerating the growth of the {100} and {101} facets specifically rather than other facets. However, during in situ synthesis of the binary nanocomposite GSnS2, it is predicted that Sn4+ is tightly anchored on the surface of exfoliated graphene oxide (GO) sheets by electrostatic interactions[25] owing to the presence of many functional groups, such as hydroxyl, carbonyl, and epoxy groups, over the surface of a GO sheet (shown in Scheme ). After that, dissociated S ions from TAA would result in nucleation, growth, and aggregation, and finally, SDS-oriented growth of SnS2 NDs over the GO sheet is obtained. The high temperature used and the presence of sulfur ions reduced the initial GO sheets to RGO by excreting residual oxide functional groups,[26] giving the RGO nanocomposite of SnS2 as GSnS2. Because of the presence of RGO sheets as substrates, the SnS2 NPs are uniformly distributed on the RGO surface. RGO could bind to SnS2 nuclei and restrain their self-assembly into NDs. Moreover, the presence of SnS2 particles between RGO sheets prevents the restacking of graphene layers and hence preserves their intriguing properties relative to a few-layer structure. In another hydrothermal approach, we synthesized NSs of SnS2 by changing the metal precursor, S source, and type of surfactant. It followed a similar kind of formation and growth mechanism as that mentioned for NDs; however, the morphology transformation is facilitated by the amount of dioctyl sulfosuccinate sodium salt (AOT) surfactant.[12] The reaction time and temperature were the same in both cases, which were 10 h and 180 °C, respectively. As predicted, AOT strongly binds to sulfur-terminated {001} facets, hence considerably accelerating the growth of the (100) and (101) planes of the layered structure beyond the disc to sheet morphology.

Structure, Surface, and Morphology

The powder X-ray diffraction (XRD) patterns of as-synthesized ND samples are compared in Figure a with that of bulk pristine SnS2. All reflections in the diffraction patterns of as-synthesized SnS2 and GSnS2 were well-indexed with bulk hexagonal SnS2, with calculated lattice parameters a = b = 3.648 Å and c = 5.894 Å (space group P3̅m1; JCPDS file no. 22-0951).[12−14] Interestingly, it was found that the intensity of the (001) peak is lower and much more broad than the corresponding bulk XRD peak of hexagonal SnS2, which indicates restricted growth of SnS2 particles along this crystal facet and hence results in a 2D hexagonal flake-like morphology. In addition, a small peak in GSnS2 that belongs to RGO appeared at 26° corresponding to the (100) plane, which is usually indexed at 9° in pristine GO (shown in the inset of Figure a), hence confirming the formation of an RGO-stacked SnS2 nanocomposite (GSnS2).[27] Raman spectroscopy determined that hexagonal SnS2 has a pure phase by locating the characteristic A1g mode at 314 cm–1.[28] In the case of the GSnS2 sample, the intensity ratio (ID:IG) is about 2.206, which is higher than that of graphene sheets, with the D band located at 1361 cm–1 and the G band located at 1604 cm–1, as shown in Figure b. On the growth of SnS2 NPs over graphene sheets, a more disordered carbon structure is expected because of defects generated on the formed nanocomposite. The formation of a nanocomposite is further confirmed from FTIR measurements (see Figure c), where an intense peak for the surface hydroxyl groups of GO completely vanished in the GSnS2 sample and the peak intensities of the carbonyl and epoxy groups decreased.

Figure 1

(a) XRD patterns of as-synthesized SnS2 NDs, GSnS2 nanocomposite, GO, and bulk SnS2. (b) Raman spectra of as-synthesized SnS2 NDs, GSnS2, and GO. (c) FTIR and (d) BET characterization of as-synthesized SnS2 NDs and GSnS2 nanocomposite.

Brunauer–Emmett–Teller (BET) plots using the multipoint BET equation indicate a type IV isotherm with H3/H4 type hysteresis, as shown in Figure d. Specific surface areas of 29.4 and 85.5 m2/g were determined from N2 adsorption–desorption BET isotherm plots of SnS2 and GSnS2, respectively. Thus, the high surface area of the SnS2/RGO nanocomposite compared with that of SnS2 NPs is due to the presence of graphene sheets, which additionally enhance the catalytic property of GSnS2 by offering more surface area for substrate adsorption. Furthermore, optical absorption measurements were carried out followed by band gap determination for the as-synthesized SnS2 and GSnS2 NDs (Figure S1). A direct band gap value of 2.32 eV was determined by the Tauc and Davis–Mott model (see the Supporting Information for calculations) for both SnS2 and GSnS2, which is consistent with earlier reports for SnS2 samples,[29] confirming a pure phase for our as-synthesized SnS2 and GSnS2. The important aspect of the obtained band gap and the semiconductor is that 2.32 eV corresponds to a visible-light wavelength of 534 nm, which is the most intense region in the solar spectrum. Thus, we expect that photons will be maximally absorbed by SnS2 and GSnS2 and that charge separation for photocatalytic reactions will be maximal.

Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images of as-synthesized SnS2 ND particles and their nanocomposites detailing their morphologies, topographies, and crystal structures are summarized in Figure . As-synthesized SnS2 has a ND-like morphology and was found to be aligned in two different orientations on the TEM grid, one that is flat-lying and another that is face-to-face stacked on its edges, as seen in Figure a. The distribution of an ND particle diameter (edge to edge) was found to be broad; however, the average size was calculated to be 35 × 9 nm (diameter, thickness). Further crystallographic orientations of hexagonal NDs were confirmed from HRTEM analyses by studying the facets that were flat-lying or standing on their edges in a stacked pillar. Distinctive sets of lattice fringes, which correspond to the reticular planes of hexagonal pristine SnS2 lying flat on a hexagonal facet , were identified. The 2D fast Fourier transform (2D FFT) calculated from Figure c, shown in Figure d, also further confirmed the presence of (100) planes, whereas HRTEM from a side view indexed the presence of another characteristic plane (001) with a lattice spacing of 0.58 nm for hexagonal SnS2. The FFT shown in Figure f with a bright spot at d001 = 0.58 nm indicates that hexagonal NDs are preferably stacked along their hexagonal facet in the ⟨001⟩ direction. Herein, the morphology of the 2D SnS2 NDs is mainly attributed to the accelerated growth of six degenerate crystalline planes {100} and prohibited growth of the {001} plane. Scanning electron microscopy (SEM) analyses demonstrated the topographies and elemental ratios (electron diffraction for X-ray analysis, EDAX) for SnS2 and GSnS2 NDs (Figure S2). A membrane-like morphology was more easily visualized for graphene in GSnS2 over SnS2 NDs relative to their respective TEM images. EDAX analysis gave a 1:2 stoichiometric ratio of Sn to S. Moreover, carbon (41%) from RGO and oxygen (29%) were also observed in the GSnS2 system. Thus, from XRD and low-magnification TEM and HRTEM data, it is evident that SnS2 formed a layered morphology analogous to graphene and a stacked topology, which is helpful for adsorbing more substrate and for the efficient movement of the charge carriers (electrons and holes) generated from photoexcitation.

Figure 2

TEM and HRTEM analyses of as-synthesized SnS2 NDs and their RGO nanocomposites (GSnS2). (a) Low-magnification TEM image of as-synthesized SnS2 NDs. (b) Particle size distribution (histogram) showing an average particle diameter (inset: particle thickness distribution). (c) HRTEM image of a flat-lying particle showing clear lattice fringes (inset: (100) plane of SnS2). (d, e) 2D FFT images calculated from panels c and f, respectively. (f) HRTEM image of face-to-face edge-stacked particles showing clear reticular (001) lattice panes. (g) Low-resolution TEM image of as-synthesized GSnS2 NDs. (h) HRTEM image of face-to-face edge-stacked particles. (i) 2D FFT image calculated from panel h. (j) Depiction of the layered crystal structure of SnS2 seen from the [0,1,0] zone axis.

Two other sets of ND samples (SnS2 ND15 and GSnS2 ND15) were prepared by increasing the reaction time to 15 h, as mentioned in the Experimental Section (see Figure S3 for TEM image and XRD). For the SnS2 material, the powder XRD pattern suggested a pure phase and the TEM image depicted a larger (∼90 nm) hexagonal ND-like morphology. A control reaction was also carried out for 5 h; however, this reaction yielded phase impurity (Figure S3c).

Structural, surface, and morphology characterizations of SnS2 NSs and the corresponding composites with RGO are elaborated in Figure . All characteristic XRD peaks in Figure a in as-synthesized SnS2 and GSnS2 NSs are well-indexed to the hexagonal structure of SnS2 (JCPDS no. 23-0677; space group P3m1).[12,13] On the other hand, the intensity of the (100) plane was found to be higher than its intensity in the bulk pattern compared to that of the (001) and (101) planes, revealing the growth of SnS2 NPs along the ⟨100⟩ direction and hence the possible formation of a 2D-type morphology in this case as well. The specific surface area was calculated from BET plots (Figure b) and found to be ∼20 m2 g–1 for the SnS2 NSs, which increased for the graphene nanocomposite up to 100 m2 g–1. The N2 adsorption–desorption curve indicates that the structure is mesoporous, which, according to the IUPAC classification, is a type IV isotherm with an H3-type hysteresis loop attributed to non-rigid aggregates of plate-like particles giving rise to slit-type pores.[30] The mesoporosity of the GSnS2 NS sample is observed from the pore size distribution curve, with a pore size of 3–4 nm and pore volume of ∼0.132 cm3 g–1, shown in the inset of Figure b. This high surface area and porosity are major advantages for the efficient adsorption of reactant molecules on our proposed photocatalyst’s surface. The low-magnification TEM images in Figure c,d depict the NS morphology of the as-synthesized SnS2 particles with an average diameter of 140–160 nm. Stacking of RGO sheets is seen in the case of the GSnS2 NS sample. A top-view HRTEM image of the GSnS2 NS sample shows clear lattice fringes of the (100) plane with d = 0.31 nm (Figure e). Interestingly, we were able to see reticular planes of the edges of SnS2 NSs in only a few particles, and one such edge is shown in Figure f, where the sheet thickness was found to be 7.5 nm. Furthermore, RGO-stacked SnS2 NSs are seen in SEM imaging (Figure f). Optical absorption measurements have been carried out followed by band gap determination of as-synthesized SnS2 NSs (Figure S4), and the calculated direct band gap value is 2.45 eV.

Figure 3

(a) XRD patterns of as-synthesized SnS2 NS, GSnS2 NS composite, and bulk SnS2. (b) BET characterization of the as-synthesized GSnS2 NS composite measured at 77 K (inset: pore volume distribution vs pore radius). (c, d) Low-magnification TEM images of as-synthesized SnS2 and GSnS2 NSs. The inset of panel d is the NS diameter distribution. (e, f) HRTEM images of SnS2 NSs in a GSnS2 NS sample viewed from the top and from the edge side, respectively. (g) SEM image of RGO/SnS2 NSs.

Spectroscopy Studies of SnS2 and GSnS2 NDs

To study the exciton separation behavior of the semiconductor and its nanocomposite, we carried out photoluminescence (PL) and time-resolved photoluminescence (TRPL) measurements. A strong PL peak at 425 nm was observed for SnS2 NDs, which is weak in the case of the GSnS2 ND sample (Figure a), suggesting that the recombination of photoinduced charge carriers is inhibited greatly by interfacial charge transfer between RGO and SnS2 NDs. To understand this process, TRPL spectroscopy was performed and data are shown in Figure b. The emission decay data were fitted triexponentially. On calculating average lifetime (⟨τ⟩), it was found that τ for GSnS2 NDs (0.8 ns) is less than that for SnS2 NDs (1.05 ns). This shortening of the lifetime in GSnS2 NDs indicates the emergence of a nonradiative pathway, that is, the delocalization of electrons from SnS2 to RGO and hence effective carrier separation.[31] This happens because of the transfer of electrons from the conduction band (CB) of SnS2 to the Fermi level of RGO, and this assumption is corroborated with the quenching of emission.

Figure 4

(a) Normalized PL and optical absorption spectra of SnS2 and GSnS2 ND samples. (b) TRPL spectra of SnS2 and GSnS2 NDs showing the corresponding average lifetime (⟨τ⟩).

Photocatalytic Degradation of Dyes

To determine the optimal amount of photocatalyst for all reactions, we carried out a few controlled reactions by varying the amount of the catalyst with time, keeping the substrate concentration the same. In the photochemical reactor vessel, 25 mL of 0.01 mM MB was taken and a dye degradation reaction was carried out by using 5.0, 7.5, 10.0, 12.5, or 15.0 mg of GSnS2 NDs. The results are summarized in Table , which shows that 12.5 mg of catalyst is the optimum amount for maximum efficiency at a given time. Another controlled reaction was carried out to determine the SnS2:GO ratio that achieves the best photocatalytic performance. For this purpose, GSnS2 ND composites were prepared by varying the amount of GO. Figure S5 shows a comparison of the degradation of 0.1 mM Rh B in 15 min using bare SnS2 NDs, GSnS2 NDs with 5 mg of GO, GSnS2 NDs with 10 mg of GO, and GSnS2 NDs with 15 mg of GO, resulting in 75.2, 91.8, 99, and 95.2% degradation, respectively. Hence, we found that the GSnS2 ND with 10 mg of GO is the best photocatalyst for the proposed photocatalytic reactions, and we used the same ratio for further studies.

Table 1

Variation of the Amount of the GSnS2 ND Photocatalyst in the Degradation of 25 mL of 0.01 mM MB Dye at 25 °C To Determine the Maximum Photocatalytic Efficiency in a 5 min Reaction

entry no.	amount of photocatalyst (mg)	% conversion
1	0	0
2	5	55
3	7.5	64
4	10	79
5	12.5	99.9
6	15	99

Therefore, aniline synthesis, dye degradation, and Cr(VI) reduction reactions were carried out by taking 12.5 mg of photocatalyst. Similar controlled reactions were carried out using GSnS2 NS as the photocatalyst (data not shown), where 12.5 mg of the catalyst sample was also found to be the most effective. Furthermore, no reaction occurred in the absence of GSnS2 samples, implying the necessity of the photocatalyst for the conversion.

For the investigation of the SnS2 and GSnS2 nanocomposites as visible-light photocatalysts for the degradation of MB and Rh B dyes, the obtained results are summarized in Figure (see the Supporting Information for calculations).[32] Complete degradation (99.9%) of 0.01 mM MB dye takes place in 5 min in the presence of the GSnS2 ND photocatalyst (Figure a), where the intensity of the peak at λmax = 655 nm decreased gradually without shifting the peak position to the baseline, indicating complete degradation. However, the amount of time required for the complete degradation of MB dye using GSnS2 ND15 and GSnS2 NS was found to be 8 and 14 min, respectively (Figure b,c). A similar trend appeared when all three nanostructured GSnS2 samples were studied for the photocatalytic degradation of Rh B dye. The amount of time required for the complete degradation of Rh B dye using GSnS2 NDs, a corresponding 15 h sample, and GSnS2 NSs was found to be 15, 18, and 35 min, respectively. In all of these cases, the activities of simple SnS2 samples (without RGO) were also compared and found to be less effective than the corresponding RGO nanocomposites.

Figure 5

Comparison of visible-light-driven degradation of MB and Rh B dyes in the presence of 12.5 mg of GSnS2 nanocomposite photocatalysts. (a–c) Photocatalytic degradation of MB (0.01 mM, 25 mL). Optical absorption spectra showing a gradual decrease in the characteristic peak for MB at λmax = 655 nm with 99.9% degradation in 5, 8, and 14 min using GSnS2 NDs, GSnS2 ND15, and GSnS2 NSs, respectively. (d–f) Photocatalytic degradation of Rh B (0.01 mM, 25 mL) with a gradual decrease in the characteristic peak at λmax = 554 nm with 99% degradation in 15, 18, and 35 min using GSnS2 NDs, GSnS2 ND15, and GSnS2 NSs, respectively.

Apparent rate constants (kap, pseudo-first-order kinetics) for each of the photocatalytic reactions were calculated, and the reaction kinetics graphs are shown in Figure . kap was found to be maximal for the degradation reactions in which GSnS2 ND was used as the photocatalyst (Figure a,d). kap gradually decreased to a minimum for the degradation of both dyes when GSnS2 NS was used as the photocatalyst (Figure c,e). Thus, Figures and 6 together imply that GSnS2 ND is the best RGO/SnS2 nanocomposite-based photocatalyst among the three RGO/SnS2 samples with different morphologies studied in the present work. To compare the catalytic activity of our photocatalyst with those reported in the literature, we calculated the activity parameter K = kap/m, where kap is the apparent rate constant and m is the mass of the catalyst loaded. The activity parameters of some recently reported highly active catalysts were compared with our measured activity parameter. The high activity of our catalyst was noted compared with that of a few of the best reported activities for the same photocatalytically induced reaction[33−37] using different types of NPs and their composites.

Figure 6

[A/A0] and ln[A0/A] (inset) vs time plot to calculate the apparent rate constants of the photocatalytic MB dye degradation reactions depicted in Figure using samples (a) GSnS2 NDs, (b) GSnS2 ND15, and (c) GSnS2 NSs. [A/A0] and ln[A0/A] (inset) vs time plot to calculate the apparent rate constants of the photocatalytic Rh B dye degradation reactions depicted in Figure using samples (d) GSnS2 NDs, (e) GSnS2 ND15, and (f) GSnS2 NSs.

The reusability of the photocatalysts was studied over 10 continuous cycles, and the photocatalytic activities versus the number of cycles are shown in Figure a–d. GSnS2 NDs were found to be excellent in this case as well, which showed only a 5% decrease in photocatalytic efficiency even after 10 cycles (Figure a,b), which is better than that of the simple SnS2 ND, GSnS2 ND15, and GSnS2 NS samples. The enhanced photocatalytic activity of the as-synthesized GSnS2 nanocomposite is due to the presence of more active sites as compared with that of bare SnS2 NDs, which facilitates the adsorption of dye molecules, efficient charge separation in SnS2, and transportation of photoelectrons during photocatalysis. On the other hand, the enhanced activity of GSnS2 NDs over GSnS2 ND15 and GSnS2 NSs is attributed to the former having smaller particles and possibly to its smaller band gap. Notably, in any case, no degradation of the dye is observed under dark for a prolonged period in the presence of the GSnS2 ND catalyst with the highest activity (Figure e), corroborating the photocatalytic nature of our as-synthesized nanocomposite products.

Figure 7

Recyclability test of GSnS2 photocatalysts up to 10 cycles at 25 °C. (a, b) Recyclability of GSnS2 ND catalysts for the degradation of MB and Rh B dyes, respectively. (c, d) Recyclability of GSnS2 NS catalysts for the degradation of MB and Rh B dyes, respectively. (e) Test under dark conditions for all photocatalytic reactions (degradation of dyes, hydrogenation of NB, and reduction of Cr(VI) in the presence of GSnS2 NDs).

At this point, the question arises as to whether the dyes are truly decomposed or just decolorized to a colorless solution. To study this phenomenon, we chose the crucial dye MB for the decomposition study and carried out high-performance liquid chromatography (HPLC) and optical absorption measurements (in the UV region) to observe the nature of both the mixture after the photocatalytic reaction and the original MB solution. As per the HPLC data, there is only one peak observed in a standard (0.01 mM) MB solution at a retention time of 30 min (Figure S6a), whereas no such peak was observed at the same retention time in the MB-degraded extract obtained after a 5 min photocatalytic reaction using GSnS2 NDs, which indicates the complete degradation of MB instead of decolorization. Furthermore, from the optical absorption spectra, it was found that two peaks appeared for MB at positions of 291 and 655 nm because of its aromatic ring and heteroaromatic skeleton, respectively (Figure S6b). With increasing photocatalytic reaction time, the sharp peak at 655 nm disappeared because of the cleavage of a C=N bond. At the same time, a significant reduction in the 291 nm peak was observed, which is due to the multistep breaking of an aromatic fragment to CO2 molecules.[38]

To check the long-term durability of the GSnS2 ND sample, characterizations of the photocatalysts were done after 50 cycles of Rh B degradation. The photocatalysts were separated after 50 cycles (only 20% efficiency loss) and washed with distilled water, and then XRD and TEM analyses were performed (see Figure S7). The XRD pattern after catalysis shows a pure phase of SnS2 without the observation of any other impurity or phase, and the TEM image shows the retainment of the ND-like morphology, which proves the long-term stability of the photocatalyst beyond degradation.

Photocatalytic Synthesis of Aniline from Nitrobenzene

The GSnS2 ND sample was found to be the best among the three photocatalysts used in the present study, as demonstrated earlier. Hence, we used a GSnS2 ND sample for further photocatalytic activity exploration. The as-synthesized nanocomposite GSnS2 ND was used for the synthesis of aminoarenes from nitroarenes. Interestingly, the results are very impressive, and finally, we obtained >99.9% conversion of 2.0 mM NB to aniline in the presence of 12.5 mg of GSnS2 NDs. The reaction was carried out under green conditions with the addition of ammonium formate as a quencher for photogenerated holes under a nitrogen atmosphere and visible-light illumination.[39,40]Figure a shows the optical absorption spectra of disappearing NB and newly forming aniline in the presence of the GSnS2 and SnS2 samples. The characteristic peak of NB at 267 nm decreases with an increase in time, with a 99.9% yield and selectivity (calculated from GC data; see Figure c). GC was performed after extracting the reaction mixture and dissolving it in a chloroform solvent. Aniline was observed within 90 min, and the peak position was shifted toward λ = 300 nm. The present visible-light (≥400 nm)-driven reaction was found to be very efficient at room temperature and atmospheric pressure compared to that in recent reports with other photocatalysts under similar reaction conditions,[39] hence making the as-synthesized GSnS2 NDs applicable as a future photocatalyst for such organic conversions over a wide range. The apparent rate constant was found to be 0.027 min–1 for this reaction when the GSnS2 nanocomposite was used as the photocatalyst (Figure b). After each photocatalytic reaction, the catalyst was recycled by air-drying and reused. Only a 7% decrease in the efficiency of the catalyst up to 5 cycles was observed (Figure d), indicating the stability and reusability of GSnS2. The minor efficiency loss may be attributed to a loss of the catalyst during the centrifugation and washing processes.

Figure 8

Visible-light-driven photocatalytic conversion of 2.0 mM NB to aniline using GSnS2 as the photocatalyst at 25 °C under a N2 atmosphere. (a) Optical absorption spectra showing a gradual decrease in the characteristic peak of NB at λ = 267 nm with an increase in time. (b) [A/A0] vs time plot to calculate the apparent rate constant of the NB-to-aniline photoconversion reaction (inset: ln[A0/A] vs t plot). No photocatalytic conversion was observed under dark reaction conditions (shown by black triangles). (c) Gas chromatograph (GC) of the photocatalytic reaction product (* indicates an unresolved product). (d) Reusability test of the GSnS2 photocatalyst up to 5 cycles for the hydrogenation of a NB reaction.

Photoreduction of Cr(VI) to Cr(III)

Figure shows the results of the photocatalytic reduction of Cr(VI) in an acidic solution (at pH 3) using SnS2 and GSnS2 as catalysts under visible-light irradiation. Results of dark condition reactions are depicted in Figure e. The reduction of Cr(VI) occurs quite rapidly, with an apparent reaction rate 0.040 min–1, and approaches 94% reduction to reach equilibrium in 90 min in the presence of the GSnS2 catalyst, in contrast to SnS2, which shows only 71% reduction, hence making as-synthesized GSnS2 as an efficient photocatalyst for Cr(VI) reduction compared to photocatalysts in earlier reports.[41] On performing the recyclability test for 5 cycles, only a 3% efficiency loss was observed for GSnS2, whereas the efficiency loss was 20–25% in the case of SnS2 NDs (shown in Figure S8). To check whether Cr(VI) is really reduced to Cr(III) rather than adsorbed on the catalyst surface to produce the decrease in the intensity of the optical absorption peak at λmax = 372 nm, a qualitative analysis was carried out on the supernatant solution after the photocatalysis reaction. To 1 mL of the supernatant solution of the reaction mixture was added a 0.1 M NaOH solution (2 mL), which produced a gray-green gelatinous precipitate of Cr(OH)3, which was further dissolved in the presence of additional acid (H2SO4); hence, a light green solution of Cr(III) was obtained,[42] confirming that the photocatalytic reduction reaction took place. TEM analysis was performed on the separated catalyst after the completion of the 5th reaction cycle, and no adsorbed chromium ions were observed as an adsorbate over the catalyst surface, as shown in Figure S9.

Figure 9

Photocatalytic reduction of toxic Cr(VI) to nontoxic Cr(III) using GSnS2 as the photocatalyst. (a) Optical absorption spectra showing a gradual decrease in the characteristic peak intensity at λmax = 372 nm, corroborating 94% reduction of Cr(VI) in 90 min. (b) [A/A0] and ln[A0/A] (inset) vs time plot to calculate the apparent rate constant of the Cr(VI) photoreduction reaction.

Mechanism of Photocatalytic Reactions

A schematic diagram to understand the mechanism of all three photocatalytic reactions using GSnS2 nanomaterials is shown in Figure and is based on a few of our new findings and a few earlier reports.[39,43,44] The reactions occurred by electron–hole pair generation from the SnS2 semiconductor under visible-light-driven conditions followed by efficient exciton separation. It is well-known that the Fermi level of graphene lies lower than the CB of SnS2. Hence, excitation of the semiconductor with suitable wavelengths leads to the drifting of electrons from the SnS2 CB to the Fermi level of graphene. Notably, reverse movement is not expected; otherwise, we would not observe any photocatalytic reaction. Graphene-supported GSnS2 nanocomposites have more active sites because of the presence of functional groups, and the electron transfer process is fast. In NB reduction, as shown in Figure a, the NB molecule accepts an electron and a hydrogen ion from water under a nitrogen environment and is converted to the aniline molecule. Ammonium formate acts as a quencher for photogenerated holes, confirming the occurrence of a free radical reaction. In addition, the presence of 2D monolayers of graphene also increases the concentration of the reactants accumulating over the surface of the photocatalysts because of the high surface area. Hence, the graphene support provided an increased opportunity for spatial contact between photogenerated electrons and reactant NB molecules.

Figure 10

Schematic representation showing all three photocatalytic reactions using the graphene-stacked SnS2 ND composite system (GSnS2) as the photocatalyst under visible-light irradiation. (a) Mechanism of photocatalytic conversion of NB to aniline using ammonium formate as a hole quencher, (b) mechanism of photocatalytic degradation of dyes (MB and Rh B), and (c) mechanism of photoreduction of Cr(VI) to Cr(III) in acidic medium.

In the photodegradation of dyes and the photoreduction of Cr(VI), as shown in Figure b,c, mainly oxyradicals, such as hydroxyl, superoxide, and HO2• radicals, are formed in the aqueous solution under visible-light irradiation when sufficient e– and h+ separation occurs from graphene-supported SnS2 semiconductor NDs. The cationic dyes are adsorbed over the surface of the photocatalyst and react with OH• to generate a cationic dye radical (dye+•),[36] which leads to nontoxic byproducts. In the case of chromium reduction, at pH 3, aqueous Cr(VI) is reduced to Cr(III) by more hydrogen ions (8H+), and 4 water molecules are generated in the whole process, as shown in Figure c.

From the above analyses, it is clear that the efficiency of the present photocatalytic reactions is based on (a) the suitable band gap of SnS2, (b) the position of the band gap in visible-light irradiation (solar spectrum) for maximum photon absorption that results in (c) maximum exciton generation, (d) effective electron mobility by graphene sheets to the reactants, and (e) better adsorption of reactants/substrates due to the high surface area with recycling. The overall photocatalytic activities of the RGO/SnS2 ND composite, which are better than those in recent reports[33−37] on similar materials, in the synthesis of aniline for the degradation of dyes (MB and Rh B) and Cr(VI) reduction reactions are summarized in Table .

Table 2

Photocatalytic Activities of the GSnS2 ND Nanocatalyst in the Synthesis of Aniline from NB, Degradation of MB and Rh B Dyes, and Reduction of Cr(VI) to Cr(III)

entry	substrate/reactant	reaction time (min)	degradation/conversion (%)	kap (min–1)	activity parameter (min–1 g–1)
1	NB	90	99.9	0.027	2.16
2	MB	5	99.9	0.647	51.76
3	Rh B	15	99	0.471	37.68
4	Cr(VI)	90	94	0.040	3.2

Conclusions

In summary, 2D SnS2 hexagonal NDs (of two different sizes) and NSs and their graphene-loaded nanocomposites (GSnS2) were successfully synthesized by a simple, economic, one-pot hydrothermal approach. Superior visible-light-driven photocatalytic conversions were demonstrated by all three materials and their corresponding RGO-based nanocomposites. However, it is emphasized that as-synthesized GSnS2 NDs exhibited visible-light (≥400 nm)-driven powerful photocatalytic activity in the synthesis of aniline from NB by a hydrogenation reaction. The conversion of substrate was >99.9%, and the yield and selectivity of the product, aniline, were 99.9% at 25 °C and atmospheric pressure. The same nanocomposite catalyst very efficiently degraded mutagenic dyes (MB and Rh B) to nontoxic products in a shorter time period. In addition, as a multifunctional photocatalyst, GSnS2 NDs effectively reduced carcinogenic Cr(VI) to nontoxic Cr(III) under ambient conditions. In all cases, the reusability of the GSnS2 photocatalyst was excellent over many cycles. The layered structure, specific 2D morphology, and enhanced electron mobility in the stacked graphene sheets due to the small size, large specific surface area, and efficient charge carrier separation in the material led to an excellent photocatalytic behavior of GSnS2. All of these attractive features of low-cost GSnS2 obtained from a simple synthesis make it ideal as an efficient visible-light-driven heterogeneous photocatalyst for industrial waste/polluted water treatment and in the synthesis of commodity organic chemicals.

Materials and Methods

Materials

Graphite powder (99%), tin(IV) acetate (Sn(CH3COO)4, 99%), tin(IV) bis(acetylacetonate) dichloride (99%), AOT (96%), hexamethyldisilazane (HMDS), and TAA (99%) were purchased from Sigma Aldrich, USA. Ortho-phosphoric acid (H3PO4, 98%) and hydrochloric acid (HCl, 80%) were obtained from Thomas Baker, India. Sulfuric acid (98%), potassium permanganate (KMnO4, 98.5%), ammonium formate (98%), MB (99%), Rh B (99%), sodium sulfide (Na2S, 98.5%), and potassium dichromate (K2Cr2O7, 99%) were obtained from Merck, India. Hydrogen peroxide (H2O2, 59%) was obtained from Fischer Scientific, India. SDS (99%) was obtained from SRL, India. NB (99%) was obtained from Spectrochem, India. All chemicals were used without any further purification.

Synthesis of SnS2 NDs

In a typical synthesis process, 1 mmol of tin(IV) acetate (in 2 mL of water) was added to 3 mmol of SDS (dissolved in 5 mL of water), and the mixture was stirred for 5 min. To this solution was added 4 mmol of TAA (dissolved in 1 mL of water), and it was further stirred for 30 min. Thereafter, 2 mL of HMDS was added to the mixture while stirring. The total volume of a 50 mL reaction mixture was made by adding 40 mL of water. The autoclave was sealed and maintained at 180 °C for 10 h. After the reaction was completed, the product was rinsed with absolute ethanol and distilled water two times and finally dispersed in 1 mL of ethanol for further characterization. This sample is SnS2 ND. In another reaction, all of the above reaction parameters were kept the same except for the reaction time, which was increased to 15 h to produce larger particles. This sample is SnS2 ND15.

One-Pot Synthesis of RGO/SnS2 ND Nanocomposites (GSnS2 ND)

RGO nanocomposites of SnS2 NDs were synthesized in a similar manner as that mentioned above (Scheme ), except that a solution of 10 mg of as-synthesized exfoliated GO sheets (synthesized by a modified version of Hummer’s method)[45] in 10 mL of water was added first to the Sn(IV) precursor solution before adding the S precursor (TAA) solution. The rest of the process is the same as that described earlier. This sample is GSnS2 ND. Another set of RGO/SnS2 nanocomposites was prepared following a similar procedure, except that the reaction time was increased to 15 h. This sample is GSnS2 ND15.

Synthesis of SnS2 NSs

Tin(IV) bis(acetylacetonate) dichloride (1 mmol) dissolved in a minimum amount of distilled water (2 mL) was added to 5 mmol of AOT, which was dissolved in 10 mL of water and stirred for 30 min. Thereafter, 3 mmol of Na2S (dissolved in 1 mL of distilled water) was added slowly to the above mixture and stirred for another 30 min. This reaction mixture was then transferred to an autoclave that was sealed and maintained at 180 °C for 10 h. After completing the reaction, the product was rinsed with absolute ethanol and distilled water several times and dispersed in 1 mL of ethanol for further characterization or dried for photocatalytic studies. This sample is SnS2 NS.

One-Pot Synthesis of RGO/SnS2 NS Nanocomposites (GSnS2 NS)

RGO nanocomposites of SnS2 NSs were synthesized in a similar manner as that mentioned earlier (Scheme ), except that a solution of 10 mg of as-synthesized exfoliated GO sheets (synthesized by modified version of Hummer’s method)[45] in 10 mL of water was added first to the Sn(IV) precursor solution before adding an S precursor (Na2S) solution. The rest of the process is the same as that described earlier. This sample is GSnS2 NS.

Characterization

XRD patterns of the as-synthesized products were collected at room temperature using a Bruker D8 Advance diffractometer system using a monochromatized Cu Kα radiation (λ = 1.54056 Å) source. Optical absorption measurements were carried out using a Perkin Elmer Lambda 35 UV–visible spectrophotometer. A diluted, well-dispersed solution of NDs in absolute ethanol was used for the absorption study. PL spectra were recorded on a Varian Cary Eclipse fluorescence spectrophotometer with an intense Xenon flash lamp. A TRPL study was carried out at room temperature using a Horiba Jobin Yvon Fluoro Hub and a single photon counting controller having a Nano LED of 370 nm and pulse duration of 1.2 ns. The band edge PL lifetime was examined by monitoring time-resolved fluorescence spectroscopy. Time-decay spectra were obtained for nanocomposites excited at λex = 370 nm. TEM and phase-contrast HRTEM measurements were performed using an FEI Technai G2-20 transmission electron microscope operating at an accelerating voltage of 200 kV. SEM and EDAX measurements were performed using a JEOL JSM 6610 at 20 kV, with a width distance of 10 mm and spot size of 30. EDAX was performed at a resolution of 135.2 eV. BET surface area and porosity measurements of the as-synthesized material were evaluated using a surface area and pore size analyzer (Gemini-V, Micromeritics, USA) at 77 K. Before analysis, samples were degassed in situ at 100 °C for 8 h. FTIR spectra were recorded by the KBr method using a PerkinElmer FTIR 2000 spectrophotometer. GC was performed using a Shimadzu-14A instrument connected to a HP-5 capillary column (30 m, HP-5). An HPLC technique was performed using a Waters 2996 multiwavelength (λ) fluorescence detector instrument.

Photocatalytic Activity Test

All of the photocatalytic tests were performed under standard conditions, as reported recently,[46] and under visible-light irradiation. A conventional mercury vapor lamp of 250 W, 12 750 lumens (Philips, HPL-N, Color Rendering Index 45 Ra8, color temperature 4100 K) was used as the light source, and the wavelength distribution of the light source was 400–700 nm (white light). The distance between the light source and the reaction mixture was 12 cm, and the optical irradiance at the sample position was calculated using a solar meter to be 525 × 100 lux at room temperature. A photocatalyst (SnS2 or GSnS2; 12.5 mg) was added to 25 mL of an NB (2.0 mM), MB (0.01 mM), Rh B (0.01 mM), or Cr(VI) (0.5 mM of pH 3) aqueous solution, and then the reaction mixture was ultrasonicated for 15 s to completely disperse the catalyst in the reaction solution. After this, all reaction mixtures were stirred under dark for 1 h to ensure the establishment of the adsorption–desorption equilibrium between the catalyst and reactants under the N2 environment at room temperature. Then, the suspension was positioned inside a cylindrical vessel surrounded by circulating water to control the temperature and irradiated with the 250 W light source. The circulation of water over the reaction vessel ensured the cancellation of IR heating of the substrates. After regular intervals of time, 1 mL of a sample was taken and centrifuged at 4000 rpm to remove the catalyst, and concentrations of the reaction mixture were measured in the supernatant solution by recording absorption spectra and noting the gradual decrease in the characteristic absorption peak at 655 (MB), 554 (Rh B), 372 (Cr(VI)), or 267 nm (NB).

Qualitative Inorganic Analysis of Cr(III)

To 1 mL of the supernatant solution of the reaction mixture after the completion of the photocatalytic reduction of Cr(VI) was added a 0.1 M NaOH solution (2 mL), which produced the gray–green gelatinous precipitate of Cr(OH)3; this was dissolved in the presence of additional acid (H2SO4), and a light green solution of Cr(III) was confirmed.[42]