An investigation into the solar light-driven enhanced photocatalytic properties of a graphene oxide–SnO₂–TiO₂ ternary nanocomposite

Abstract

A novel graphene oxide (GO)–SnO₂–TiO₂-based ternary nanocomposite was prepared via a one-step solvothermal process. The structure, morphology, and optical properties were characterized by a series of techniques, including X-Ray Diffraction (XRD), Field-Emission Scanning Electron Microscopy (FE-SEM), Energy Dispersive Spectroscopy (EDS), UV-vis Diffuse Reflectance Spectroscopy (DRS), Photoluminescence, Raman spectroscopy, Nitrogen Adsorption–Desorption, and X-Ray Photoelectron Spectroscopy (XPS). Various microscopic images of the ternary nanocomposite showed that the SnO₂ and TiO₂ nanoparticles are firmly covered over GO, thereby increasing the surface area of the resultant nanocomposite. The photocatalytic activity of ternary GO–SnO₂–TiO₂ and binary GO–SnO₂ and GO–TiO₂ materials were studied through the photodegradation of congo red and methylene blue under solar radiation. The degradation efficiency of GO–SnO₂–TiO₂ was found to be 96% for methylene blue dyes within 60 min and 98% for congo red within 70 min, which is much higher than the binary composites. Furthermore, a photoelectrochemical study was performed to provide further insight into the photocatalytic activity, which further confirmed the superiority of the novel ternary nanocomposite in photocurrent generation. The enhanced photocatalytic properties of the ternary nanocomposite can be attributed to enhanced light absorption, efficient charge transfer process, high surface area, as well as superior durability of the composite. In addition, a possible reaction mechanism has been postulated. Our results have demonstrated that by carefully introducing GO with suitable metal oxides, highly efficient photocatalysts can be designed that would absorb a wider range of the solar spectrum.

---

1. Introduction

Organic dyes cause serious environmental problems due to their high toxicity to aquatic creatures and carcinogenic and mutagenic effects on humans.^1–3 Such dyes have high resistivity towards aerobic degradation, which increases their lifetime in water bodies, thereby requiring streamlined research for their removal.⁴ Methylene blue (MB) and congo red (CR) are such dyes which behave as suspected carcinogen and is toxic to many organisms.^5–7 At present, photocatalytic degradation is widely preferred as a green technique to offer great potential for the degradation of dye molecules into harmless substances.⁸ So far, several semiconductor-based photocatalysts, such as oxides,^9,10 sulfides,^11,12 and oxynitrides^13,14 have been investigated for the degradation of toxic organic pollutants in waste-water. Among all the studied semiconductors, metal oxides have shown the most promising activities for degradation because of their suitable band energies. Nanosized semiconductor metal oxides, such as TiO₂, In₂O₃, Bi₂O₃, SnO₂, ZnO, and Fe₂O₃ are widely used in photocatalysis.^15–18 In particular, nanocrystalline TiO₂ and SnO₂ have been used extensively in the debasement of natural contaminations and water splitting.^19–23 They have been emerged as much preferred photocatalysts because of the high negative reduction potentials of the excited electrons and rapid generation of electron–hole pairs by photoexcitation, as well as their lack of toxicity, eco-friendliness, high stability, and low cost. However, the major drawback in using these semiconductor materials is their large bandgaps due to which they utilize only UV light (4–5% of the solar spectrum) and which also significantly limits their use of the visible light portion (∼50%) of the solar spectrum. Therefore, current research interest focuses on the design of novel photocatalyst which could absorb a wider portion of the solar spectrum to generate enhanced photocatalytic activity and photostability.

In this regard, formation of a heterojunction between two semiconductor materials of different band gaps has been found to be a novel strategy to improve the separation efficiency of photogenerated charge carriers. Towards this objective, mainly binary metal oxide nanocomposites have been synthesized for variety of photocatalytic applications.^24–36 However, even then, a complete and faster degradation rate is not yet achieved and problem for charge recombination still exist.

Therefore, exploring a highly efficient TiO₂ and SnO₂-based photocatalyst through a considerable design is needed. Recently, the coupling of graphene oxide with semiconductor materials is found to be a promising strategy for improving the photocatalytic activities of semiconductors.³⁷ Graphene oxide, with a two-dimensional honeycomb structure, has attracted extensive attention in recent years. Major emphasis has been paid to the coupling of graphene oxide with some semiconductors,³⁸ which has shown a significant improvement of the photocatalytic ability by virtue of its superior charge transport properties,³⁹ intense light absorption property and the unique flexible sheet-like structure of the graphene component.^40–42 For example, in the case of GO–TiO₂ nanocomposites, the GO is considered to serve as an efficient acceptor for the photogenerated electrons, thus significantly suppressing charge recombination and enhancing the photocatalytic rate of the nanocomposite as compared to that of pure TiO₂ nanoparticles.^43,44 Unfortunately, the electron transfer and the recombination of photo-generated charge carriers still exist in graphene oxide-based binary composite, decreasing its photocatalytic performance. To solve this problem, more recently, design and development of ternary composite has been found to be an ideal solution in which the band-edge levels of each component in the composite would have a step wise structure which help in achieving a efficient charge transfer process, thereby increasing its photocatalytic performance. Till date, few ternary nanocomposites such as ZnO–rGO–Ag,⁴⁵ ZnS–Ag₂S–RGO,⁴⁶ and RGO–TiO₂–ZnO⁴⁷etc. have been demonstrated as an efficient photocatalysts. In this direction, a ternary composite of GO, TiO₂ and SnO₂ should be a promising photocatalyst as the energy levels of these materials follow the order: TiO₂ (CB) > GO > SnO₂ (CB). Such a step-wise structure would help in transferring photo-induced electrons from TiO₂ CB via GO to SnO₂, which can efficiently separate the photo-induced electrons and hinder the charge recombination. Moreover, design of such composite would help in harvesting a wider portion of solar radiation, including ultraviolet light and visible light for a successful photodegradation process. Additionally, the use of solar radiation as an energy source reduces costs and also beneficial to environment. Even though literature have reported on the use of GO–SnO₂ and GO–TiO₂ nanocomposite, to the best of our knowledge, the design of ternary nanocomposite based on (GO–SnO₂–TiO₂) system has not been studied so far.

In this study, we report a ternary graphene oxide-based nanocomposite, specifically referring to graphene oxide–SnO₂–TiO₂ (GST) system. By employing the degradation of Methylene Blue (MB) and Congo Red (CR) under solar light as a model reactions, we systematically looked into the photocatalytic behavior and properties of GST. We found that the ternary GST nanocomposite exhibited an improved activity for successive photocatalytic reactions as compared to the other binary nanocomposite such as GO–SnO₂, and GO–TiO₂. Photoelectrochemical studies were also carried out and the results corroborate with the enhanced photocatalytic activities of the ternary composite. The mechanism behind the observed synergistic effect under sunlight are discussed in detail. The high performance photocatalyst anticipated to provide new trend of thought in proposing other GO-based multicomponent hybrid photocatalyst toward specific application.

2. Experimental

2.1. Reagents and materials

Graphite powder, tetrabutyl titanate (TBT), tin(IV) chloride penta-hydrate (SnCl₄·5H₂O), isopropanol, and ethanol were purchased from Sigma Aldrich. All the chemicals were of analytical grade and used as received. 18 Milli-Ω water was used in all experiments.

2.2. Preparation of graphene oxide–TiO₂ nanocomposite

Graphene oxide (GO) was prepared from natural graphite flakes according to the enhanced Hummers method.⁴⁸ Graphene oxide–TiO₂ nanocomposites were prepared via a simple solvothermal method according to our previous report.⁴⁹ Typically, 100 mg of GO was first dispersed in 60 ml of isopropanol by sonication for 1 h to obtain a clear brown dispersion of graphene oxide. 0.8 ml of tetra butyl titanate (TBT) was added into the dispersion and the mixture was stirred constantly for 30 min at room temperature. Afterward, 2 ml of deionized water was added drop-wise, followed by stirring for another 30 min. The resulting pale yellow sol was then transferred into a 100 ml Teflon-lined stainless steel autoclave and solvothermaly treated at 180 °C for 18 h. The product was washed first with millipore water three times by centrifugation, followed by using ethanol three times.

2.3. Preparation of graphene oxide–SnO₂ nanocomposite

Graphene oxide–SnO₂ nanocomposites were prepared via a solvothermal method. In a typical synthesis, a required amount of 150 mg of SnCl₄·5H₂O and 100 mg of GO were added in 50 ml of isopropanol, followed by sonication for 1 h to obtain a completely dispersed solution. Afterward, 2 ml of deionized water was added dropwise, followed by stirring for another 30 min. After that, the dispersion was transferred into a 100 ml Teflon-lined stainless-steel autoclave and heated at 180 °C for 16 h. The products were then collected by centrifugation and washed several times with millipore water and ethanol in order to remove the chloride ion.

2.4. Preparation of graphene oxide–SnO₂–TiO₂ (GST) nanocomposite

The synthesis of GST nanocomposite is schematically illustrated in Scheme 1. To synthesize this composite, GO, tin(IV) chloride pentahydrate (SnCl₄·5H₂O) and tetra butyl titanate (TBT) were used as precursor. Typically, 100 mg of GO was first dispersed in 60 ml of isopropanol by sonication for 1 h to obtain a clear brown dispersion of graphene oxide. 150 mg tin(IV) chloride pentahydrate (SnCl₄·5H₂O) as the source of tin was added to the above dispersion under continuous stirring for 30 min. 0.8 ml of tetra butyl titanate (TBT) was added into the dispersion once the solution was clear. Afterward, 2 ml of deionized water was added dropwise, followed by stirring for another 30 min. The solution was then transferred into a Teflon-lined stainless steel autoclave and heated at 180 °C for 18 h. After completion of the reaction, the solution was centrifuged at 3000 rpm for 2 h, and the resulting material was washed with millipore water and ethanol to remove impurities. Finally, the obtained product was dried at 75 °C over-night prior to being characterized.

Scheme 1 Schematic diagram for the synthesis of the GST nanocomposite.

2.5. The evaluation of photocatalytic activities

The photocatalytic activity was evaluated by the degradation of congo red (CR) and methylene blue (MB) dyes under solar light. In a typical process, 40 mg of the photocatalyst was suspended in 100 ml aqueous solution of congo red (20 mg l⁻¹). Before being exposed to any form of illumination, the suspensions were stirred in the dark for 1 hour to ensure that adsorption/desorption equilibrium of CR on the sample surface was established and UV-vis absorption spectra of the suspension was then taken in order to compared with that of the original CR solution to note any changes in the absorbance value if any. Subsequently, the suspension was irradiated with direct sunlight. All experiments were conducted under similar climatic condition on the sunny days of May–June 2014 in Rourkela city (geographical location 24.25 North Latitude and 84.88 East Longitude), between 1000 hours and 1700 hours (surrounding temperature, 35–40 °C). At a given time interval of irradiation (20 minutes), 3 ml of the suspension was withdrawn, followed by centrifugation at a rate of 9000–10 000 rpm for 7 minutes. UV-vis absorption spectra of the supernatant was then measured using the UV-visible spectrophotometer. Similar conditions were employed for the photocatalytic degradation of methylene blue dye.

2.6. Photoelectrochemical measurements

Photoelectrochemical measurements were performed in a three-electrode in-house developed Electrochemical Workstation where Pt wire and Ag/AgCl were used as counter and reference electrodes, respectively. Thin films of the samples coated onto Indium-doped Tin Oxide (ITO) substrates were used as working electrodes. Typically, 15 mg of the sample was dissolved in 20 ml of methanol and ultrasonicated for half an hour to allow the uniform dispersion of the sample. Then, the suspension were spin-coated onto ITO substrates which were placed onto a hot plate at 80 °C. A 0.5 M Na₂SO₄ aqueous solution was used as an electrolyte. The source of photon was generated by a 150 W high pressure mercury vapor lamp (λ = 400–450 nm), passed through a flat circular quartz window, equipped on the side of the three-electrode cell. The photoinduced current–voltage (I–V) curves were measured at an input voltage of 2.0 V. The gap between the switching on and turning off of the light was 50 s.

3. Characterization

The UV-vis absorbance spectra were obtained on UV-2450, Agilent technologies. FTIR spectra of the product were recorded using a Perkin-Elmer FTIR spectrophotometer using NaCl support. The catalyst was analyzed by X-ray diffraction study using PHILIPS PW 1830 X-ray diffractometer with CuKα source. Raman spectra were recorded using a BRUKER RFS 27 spectrometer with 1064 nm wavelength incident laser light. Field emission scanning electron microscopy (FESEM) of the sample was recorded by Nova NanoSEM/FEI. The compositional information of the products was performed using EDX (JEOL JSM-6480 LV). Transmission electron micrographs (TEM) of the sample were recorded using PHILIPS CM 200 equipment using carbon coated copper grids. Nitrogen adsorption/desorption isotherm was obtained at 77 K on a Quantachrome Autosorb 3-B apparatus. The specific surface area and pore size distribution were acquired by emulating BET equation and BJH method, respectively. The dielectric parameters were measured using a computer-controlled impedance analyzer (HIOKI IMPEDANCE ANALYZER 1352) as a function of frequency (100 Hz to 1 MHz) at room temperature. The thermogravimetric (TG) measurement was carried out using a NETZSCH, STA409C, Germany thermal analyser under a flow of N₂ gas with a temperature ramp of 10 °C min⁻¹ from room temperature to 800 °C. A commercial electron energy analyser (PHOIBOS 150 from Specs GmbH, Germany) and a non-monochromatic Mg Kα X-ray source (hv = 1253.6 eV) have been used to perform XPS measurements with the base pressure of <1 × 10⁻⁹ mbar.

4. Results and discussion

4.1. Structure and morphology characterization

In this work, graphene oxide was synthesized using hummer's method as discussed in the experimental section. Later on, the desired ternary compound GO–SnO₂–TiO₂ nanocomposite containing three components, i.e., GO, SnO₂ and TiO₂ were prepared by a solvothermal method. The presence of various chemical groups on each GO, GO–TiO₂, GO–SnO₂ and GO–SnO₂–TiO₂ was characterized by XRD and FTIR spectroscopy. X-ray diffraction analysis was employed to determine the crystal phases of the synthesized samples which are shown in Fig. 1. The X-ray diffraction peak of GO show a sharp peak at 2θ = 10.5° which relates to the (002) reflection of stacked GO sheets. In the diffraction pattern of the GO–TiO₂ composite (JCPDS no. 21-1272), characteristic signals of the TiO₂ nanoparticles are coming at 25.6°, 38.1°, 48.5°, 55.6° and 62.9° as shown in Fig. 1b. The GO–SnO₂ composite was also characterized by XRD analysis, and its diffraction pattern is shown in Fig. 1c. The diffraction peaks, (100), (110), (101), (200), (211), (310) and (301) as seen in the XRD profile can be indexed to the rutile phase of SnO₂ (JCPDS no. 41-1445).^50–52 Finally, the XRD spectra of the GO–SnO₂–TiO₂ nanocomposite in Fig. 1d showed the same peaks related to the TiO₂ along with certain peaks related to the SnO₂ nanoparticles on graphene oxide surface. The presence of all of these peaks in this diffractogram confirms the effective formation of the ternary nanocomposite. In all the binary and ternary nanocomposites, the absence of GO peak implies the reduction of GO during the solvothermal process, in which supercritical (SC) solvent play the role of reducing agent.^53,54 This reduction process is believed to be analogous to the H⁺ catalyzed dehydration of alcohol, where water acts as a source of H⁺ for the protonation of hydroxyl groups.⁵⁵ Further, the broad diffraction peaks of the nanocomposites imply the small crystallite size of the TiO₂ and SnO₂ nanoparticles.

Fig. 1 X-ray diffraction patterns of GO, GS, GT and GST nanocomposite.

To evaluate the chemistry of all the functional groups in GO and the resulting binary and ternary composite, FTIR analysis was performed. Fig. 2 shows the Fourier transform infrared (FTIR) spectrum of GO, GST, as well as the GT and GS nanocomposites. In the case of GO, the peaks at 1060 and 1706 cm⁻¹ can be attributed to stretching of C–O and C=O groups, while the peaks at 1570–1210 cm⁻¹ is due to the stretching vibrations of phenolic group C–OH and C=C group, respectively. Additionally, the broad peak at 3410 cm⁻¹ is due to the stretching vibrations of O–H stretching mode. In the GT composite, most of the functional groups such as –OH and –COOH are lowered suggesting the effective reduction of GO during the synthesis. Moreover, the bands at 1400–1600 cm⁻¹ develop due to the Ti–O–C vibration suggesting the effective interaction between Ti and C. In addition, the absorption signals in the range 450–670 cm⁻¹ can be attributed to the Ti–O–Ti bond resulting on the surface of the GT composite.⁵⁶ Similar to GT, in the case of GS composite, most of the oxidized functional groups disappeared. New peaks at 661 and 539 cm⁻¹ are due to O–Sn–O and Sn–O stretching vibrations, respectively.⁵⁷ All these characteristic peaks of both SnO₂ and TiO₂ can be found in the spectrum of GST nanocomposite which suggests the successful incorporation of both the nanoparticles into the GO sheets. Both the XRD pattern and FTIR spectrum confirmed the coexistence of TiO₂ and SnO₂ in the GST ternary nanocomposites.

Fig. 2 Fourier transform infrared (FTIR) spectra of (a) GO (b) GS (c) GT and (d) GST nanocomposite.

The X-ray photoelectron spectroscopy (XPS) was further used to investigate the detailed elemental compositions of the nanocomposites. Fig. 3a shows the survey XPS spectrum of GO–SnO₂–TiO₂ nanocomposite sample. The peaks observed at 284.8, 458.8, 486.8, and 532.2 eV corresponds to C 1s, Ti 2p, Sn 3d, and O 1s core-levels at their respective standard values of the binding energy, which indicate the presence of these elements in this ternary nanocomposite sample. Fig. 3b presents the narrow scan XPS spectrum of Sn 3d core-level where two clear doublet peaks at 488 eV and 496.5 eV correspond to Sn 3d_5/2 and Sn 3d_3/2 electronic states, respectively are visible.⁵⁸ The binding energy difference between the two lines is 8.5 eV, which is close to the standard reference value of that in SnO₂. Clearly, the XPS spectrum of Sn 3d confirms that Sn ions are in the +4 oxidation state. The narrow scan XPS spectrum of the Ti 2p core-level can be seen in Fig. 3c with two peaks at around 460 eV and 465.5 eV, which suggest the presence of Ti 2p_3/2 and Ti 2p_1/2, respectively, in the ternary nanocomposite sample. The position of two peaks originating for Ti suggests the oxidation state of Ti⁴⁺ in anatase TiO₂.⁵⁹ However, the Ti 2p_3/2 (459 eV) and Ti 2p_1/2 (464.5 eV) peaks in Fig. 3c show a small shift towards somewhat higher binding energy with respect to the reported values,⁵⁹ which might be due to the formation of Ti–O–Sn bonds on the surface of TiO₂. Moreover, the smaller spin–orbit splitting of about 5.5 eV in Ti 2p core-level indicates the interaction between the lattice Ti and oxide functional groups of GO.^60–65 The XPS spectrum of C 1s of GO (Fig. 3d) represents four components that correspond to the carbon atoms in different functional groups, arising from the non-oxygenated ring C (C=C/C–C, 284.5 eV), the C in C–O bonds (286.6 eV) of epoxy and alkoxy group, carbonyl C (C=O, 287.8 eV), and the carboxylate carbon (O–C=O, 288.8 eV) groups. On comparing to GO, the respective peak intensities of C 1s spectra of GO–SnO₂–TiO₂ nanocomposite (Fig. 3e) reduces, which implies that during synthesis of ternary nanocomposite some of the hydroxyl and epoxide functional groups have been solvothermally reduced by supercritical (SC) solvent. Similar observations are reported in other literatures.^53,54,66 The removal of these groups indicates the good electronic conductivity among its constituents, which might induce the GO sheet to serves as a conductive channel between SnO₂ and TiO₂ nanoparticles groups.^67–69 The conductivity of GO mainly relies on carrier transport within the carbon plane, as a result, functional groups attached to the plane, e.g. epoxy and hydroxyl groups, are the main influencing factor on its conductivity, while groups attached to the edge, e.g. carboxyl and carbonyl group, have less influence. Therefore, the removal of these groups can restore pathways for carrier transport within GO and increase its conductivity.¹⁹Fig. 3f shows the XPS spectrum of the O 1s core level, which can be deconvulated into two peaks: 530.2 and 532.3 eV, corresponding to lattice oxygen from the Ti–O bond and Sn–O, respectively.

Fig. 3 (a) XPS survey spectrum of GST nanocomposite, (b) and (c) Sn 3d, and Ti 2p core level spectra of GST nanocomposite, (d) and (e) C 1s spectra for GO and GST nanocomposite, (f) O 1s spectra of GST nanocomposite.

In order to understand the morphologies of our nanocomposites, FESEM analysis was performed. Fig. 4 shows the FESEM images of GO and binary/ternary nanocomposites. The FE-SEM image of GO (Fig. 4a) shows the three-dimensional inter-linked structure where typical wrinkles and folds of the graphene oxide sheet can be clearly seen which possibly formed due to the exfoliation and restacking process of GO. The GT nanocomposite (Fig. 4b) depicts a crinkled texture, where TiO₂ nanoparticles spread over the GO sheet. The surface of GO sheet is uniformly anchored with TiO₂ nanoparticles with homogeneous distribution of nanoparticles. In the case of GS nanocomposite (Fig. 4c), it is clearly observed that GO sheets are densely decorated by SnO₂ nanosheets. In both of the cases, the SnO₂ and TiO₂ nanoparticles are well-dispersed on the GO sheets and they possibly acted as a spacer.^70,71 The ternary GST nanocomposite shows a rippled type of structure with uniform distribution of SnO₂ and TiO₂ nanoparticles on the graphene oxide surface (Fig. 4d). Later on, elemental mapping was carried to determine the elemental distribution of the individual components on the surface of GO and the results are displayed in Fig. 4e. The elements C, Ti, Sn, and O are present in the testing area with uniform distribution, indicating the uniform distribution of TiO₂ and SnO₂ nanoparticles over the entire matrix. No traces of other impurities were seen tin the spectra, suggesting that the synthesized ternary composite are pure. The results of XRD and FESEM demonstrated that the structural property of the metal oxides are preserved during the solvothermal synthesis.

Fig. 4 FESEM image of (a) GO, (b) GS, (c) GT (d) GST nanocomposite (e) EDX spectra and EDS elemental mapping of GST nanocomposite (inset shows the atomic percentage of various elements).

The microstructure of the final GST nanocomposite was further investigated through transmission electron microscopy (TEM) as depicted in Fig. 5. The figure shows that the SnO₂ and TiO₂ nanoparticles were distributed over the entire GO sheets, accounting for the successful synthesis of the desired nanocomposite. The sizes of metal oxide nanoparticles on GO were in several nanometres range (6–8 nm) as shown in Fig. 5a. Because of interfacial interactions and preferential heterogeneous nucleation, numerous SnO₂ and TiO₂ nanocrystals were densely deposited onto the graphene oxide sheets, as shown in Fig. 5a. The selected area electron diffracted pattern show rings for (110) and (101) lattice plane for TiO₂ and (110), (101) and (211) lattice planes of SnO₂, respectively. The corresponding HRTEM image in Fig. 5c and d showed clear lattice fringes, which allowed for the identification of crystallographic spacing. The coexistence and connection of SnO₂ and TiO₂ can be confirmed by the high-resolution TEM (HRTEM, Fig. 5d). There are two main distinct lattice fringes. The spacing of 0.33 nm and 0.26 nm were characteristic of (110) and (101) lattice planes of the rutile SnO₂ and anatase TiO₂ structure, respectively.^60,72 The crystal lattice fringes of GO was not clearly seen due ultrafine structure.⁷³ The results suggested the formation of anatase TiO₂ with rutile SnO₂ on GO, which was consistent with the XRD results.

Fig. 5 TEM, SAED and HRTEM image of (a–d) GST nanocomposite.

To obtain further understanding of the structural changes occurs during the chemical processing from GO to GST nanocomposite and to study the effect of attachment of nanoparticles on the GO layers, Raman spectra were obtained between 800 and 2000 cm⁻¹ and are shown in Fig. 6. The Raman spectrum of pure graphene oxide displays a D-band at 1310 cm⁻¹, typically assigned to defects and disarranged structures of GO lattice.⁷⁴ Another peak at 1603 cm⁻¹ was also seen which is assigned to the G-band obtained due to the vibration of sp²-bonded carbon atoms in the hexagonal lattice.⁷⁵ In comparison GST nanocomposite had slight shift in peak positions. It was found that the D and G-Raman band of the ternary nanocomposites are positioned at a slightly lower value position like 1292 cm⁻¹ and 1594 cm⁻¹, respectively. Thorough search of available literatures reveal that when metal nanoparticles are deposited on the surface of GO, the intensity ratio of D and G-band (I_D/I_G) usually increase, suggesting the presence of electronic interaction between the nanoparticles and GO.⁷⁶ In our case, the intensity ratio of composite (I_D/I_G = 1.42) is found to be larger than that of the pure graphene oxide (I_D/I_G = 1.08). An increase in I_D/I_G value indicates the generation of defects in sp² hybridized conjugated carbon domains during ternary nanocomposite synthesis. Additionally, the peak position of G-band was shifted from 1603 cm⁻¹ in GO to 1594 cm⁻¹ in GST composite. These Raman results suggest the presence of electronic interaction between SnO₂, TiO₂ and GO in the composite, thus confirming the successful incorporation of the SnO₂ and TiO₂ particles onto the GO surface. Furthermore, these results are consistent with the FESEM, TEM and XPS results. It is important to note that similar geometrical confinement of metal oxide nanoparticles decorated on graphene oxide nanosheets had been reported in various literatures, which help to suppress the agglomeration of nanoparticles.^77–79

Fig. 6 Raman spectra of (a) GO and (b) GST nanocomposite.

Additionally, the effect due to the incorporation of third material in GO in terms of thermal stability was studied in GST ternary nanocomposite by TGA analysis under nitrogen atmosphere (see ESI Fig. S1†). The initial look at TGA plot suggests that GST ternary nanocomposite has highest thermal stability than GT nanocomposite and pure GO. The TGA results of the pure GO, GT and GST are summarised in Table 1. Pure GO showed the thermal degradation minima at (i) 100 °C, (ii) 200–220 °C and (iii) 600 °C. The first degradation temperature is the same for all the materials, and it is due to the loss of moisture and impurities captured in the material. The second degradation of 31% mass loss around 200–220 °C is ascribed due to the pyrolysis of the labile oxygen-containing groups in the forms of CO, CO₂ and steam. The third degradation of 60% mass loss is found above 600 °C due to decomposition of carbon chain. The total weight loss of GS binary nanocomposite is around 50% at 800 °C, whereas it is 32% for the GST ternary nanocomposite. Due to the incorporation of third material, the thermal motion of the metal oxide nanoparticle becomes restricted inside the pores of graphene oxide sheets, enhancing the thermal stability of the ternary composite.

Table 1 BET surface areas and the pore volume of GST, GT and GS nanocomposite

Sample	BET surface area (m^2 g^−1)	Pore volume (cm^3 g^−1)
GST	160	0.11
GT	110.8	0.28
GS	60.8	0.42

---

The absorption edge of light plays an important role in photocatalysis. In order to have a photocatalytic reaction with high yield, the designated photocatalyst should have high-intensity optical absorbance both in UV and visible region to cover a larger portion of the solar spectrum.⁸⁰ The absorbance spectra of the SnO₂, TiO₂, GO, GST, GS and GT samples were recorded using a UV-vis spectrophotometer over the range of 180–800 nm and are depicted in Fig. 7. In the case of SnO₂ and TiO₂, intense absorbance in the ultraviolet region (350 and 360 nm, respectively) was found. Pure GO shows the characteristic absorption edge at 260 nm, indicating GO only absorb UV light. The wavelength threshold of GS composite photocatalyst is estimated to be 380 nm. Compared with pure GO, introduction of SnO₂ nanoparticles increased the absorbance. This may be due to an increase in surface electric charge of the oxides in the GS composite. In comparison, GT shows a broad absorption at 380–400 nm due to the introduction of TiO₂ nanoparticles. In comparison to bare SnO₂ and TiO₂ nanoparticle, GST nanocomposite showed a broad background absorption in the visible region. As shown in figure, there was an obvious red shift of ca. 50–60 nm in the case of GST was observed (inset of Fig. 7). As a result of this extended photo responding range of ca. 420–425 nm (corresponding to the violet-blue region in electromagnetic spectrum), a more efficient utilization of the solar spectrum could be achieved and sunlight becomes energetic enough to induce the electron transitions. Similar observations were also reported in other graphene-based and other nanocomposite systems.^81–84 The bandgap energies of the photocatalysts were determined from the Tauc plot, a plot between (F(R)hv)^1/2versus hv. F(R) is the Kubelka–Munk function derived from reflectance spectra F(R) = (1 − R²)/2R equation and hv is the photon energy.⁸⁵ The estimated band gap energy from the intercepts of the tangents to the plots are 3.6, 3.3, 3.9, 3.2, 3.13 and 3.0 eV for the samples SnO₂, TiO₂, GO, GS, GT and GST, respectively. It can be found that with the introduction of SnO₂ and TiO₂ into GO, narrowing of band gap in GST composite was observed. This narrowing should be attributed to the chemical bonding between SnO₂, TiO₂, and GO, that is the formation of Ti–O–C and Sn–O–C, similar to the case reported in GO–TiO₂ and others.^86–88 This suggests that the minimum energy required to excite an electron from the valence band to the conduction band will be considerably lower in GST composite and it has a larger redox potential for the photocatalytic decomposition of organic contaminants under sunlight irradiation.^84,89

Fig. 7 UV-vis spectra and Tauc's plot of the synthesized nanocomposites.

5. Sunlight driven photocatalysis

The photocatalytic activities of as-synthesized GO-based binary and ternary composites were carried out under sunlight for the degradation of MB and CR dyes. To have a better evaluation of the photocatalytic efficiency, the degradation behaviour of bare SnO₂ and TiO₂ nanoparticles was also carried out. Blank experiments (without light or catalyst) demonstrated negligible changes, suggesting the degradation behaviour was driven primarily by a photocatalytic process. Fig. 8a shows the comparison of the photodegradation rate of MB with a concentration of 20 mg l⁻¹ in the presence of photocatalyst under exposure to sunlight for different durations. With the increase of the irradiation time, the absorption peak corresponding to MB at 663 nm diminishes gradually, testifying the degradation of MB. The removal rate of MB for SnO₂ and TiO₂ was 12% and 21%, respectively (Fig. S2†). By contrast, when graphene oxide was introduced into SnO₂ and TiO₂, the removal rate was increased to 36% and 44%, respectively. The ternary nanocomposite shows maximum removal efficiency of 96% for MB dye. As shown in Fig. 8b, GST ternary nanocomposite possesses a faster reaction rate than other nano-photocatalyst system. Furthermore, it can be seen that the decomposition rate of photocatalyst follow: GST > GT > GS > TiO₂> SnO₂. The enhanced degradation rate of GST lies on the synergic effect between catalytic activity and adsorption capability of nanostructured SnO₂ and TiO₂ with GO. Here, the degradation of MB can proceed via direct reduction by photo-induced electron or direct oxidation in the presence of holes. It is important to note that GO sheets show a tendency to get agglomerated back in graphitic structure due to strong van der Waal interaction. The incorporation of SnO₂ and TiO₂ on the GO surface helps to reduce the agglomeration of the graphene oxide sheets. Moreover, during the photocatalysis process the inter-link connection between SnO₂, TiO₂ and GO permits the easy stepwise electrons transfer, which can significantly enhances the charge carrier separation and significantly supress the charge recombination to large extent. Hence, this process helps to improve the photocatalytic activity.

Fig. 8 (a)Absorbance spectra of MB under sunlight in the presence of GST nanocomposite, (b) comparison of photocatalytic degradation curve of MB using GS, GT, SnO₂, TiO₂ and GST nanocomposite (c) absorbance spectra of CR under sunlight in the presence of GST nanocomposite (d) comparison of photocatalytic degradation curve of CR using GS, GT, SnO₂, TiO₂ and GST nanocomposite.

Considering the good performance of GST on dye treatment, photocatalytic degradation activities of congo red (CR) was also investigated. Fig. 8c and d displays the UV-vis absorption spectra of dye in the presence of GST and Fig. S3† shows the absorption spectra of dye in the presence of GT, GS, SnO₂ and TiO₂ under sunlight at different time intervals. It was observed that CR was hardly degrading even after exposure for 70 min to sunlight in the absence of photocatalyst. When the catalyst was used, the main peaks at 498 nm decreased continuously with the increase of irradiation time, indicating that the dyes were decomposed. It was also noted that GST composite exhibited best photocatalytic performance than binary nanocomposite and metal oxide nanoparticles with degradation efficiency of 98% in 70 min. In comparison, the removal rate of CR for SnO₂ and TiO₂ was 20% and 26%, respectively whereas for GO–TiO₂ and GO–SnO₂ it was found to be 42% and 58%, respectively.

The recyclability of a photocatalyst is one of the important factors for practical applications. Recycling reactions were carried out to explore the photo-stability of the ternary nanocomposite by adding used ternary nanocomposite to fresh CR and MB dyes. As shown in Fig. 9, GST ternary composite demonstrates significant photo stability even after five successive cycles. The photo catalytic activity of ternary nanocomposite photo catalyst retains over 88% and 83% for MB and CR dyes, respectively after five successive experimental runs in sunlight-light irradiation. The very slight decrease in activity might be partly caused by loss of the photocatalyst particles during each round of collection and rinsing. Moreover, the XRD patterns of the ternary composite before and after the photocatalytic reactions revealed an intact crystalline nature of the metal oxides, therefore verifying the photostability of the ternary nanocomposite as shown in Fig. S4.†

Fig. 9 Recyclability of GST nanocomposite in five successive experiments for photocatalytic degradation of CR and MB dyes in aqueous media under sunlight irradiation.

To have a better understanding of the reaction kinetics of the dye MB and CR dye degradation, the experimental data were fitted by pseudo-first-order and pseudo-second-order kinetic model. A linear fitting between a pseudo-first-order kinetic model and experimental data indicated that the photocatalytic degradation of CR and MB dye follows a pseudo-first order model. Its kinetics can be expressed using ln(C_t/C₀) = −k_appt, where k_app is the apparent reaction rate constant, Co the initial concentration of organic dye, t the reaction time and C_t is the concentration of dye at the reaction time t. The apparent constant values can be deduced from the linear fitting of ln(C_t/C₀) vs. irradiation time and the calculated values are presented in Fig. S5.† The initial degradation rate constant for 20 mg l⁻¹ of the model dye with different catalysts was calculated and the results are presented in Fig. S5a and b.† The order of degradation rate of CR and MB is as follows; GST nanocomposite > GT nanocomposite > GS nanocomposite > TiO₂ > SnO₂. The degradation rate constant for GST nanocomposite (k = 0.046 min⁻¹) was higher than that of binary nanocomposite such as GT (k = 0.010 min⁻¹) and GS (k = 0.006 min⁻¹). Similarly, the pure TiO₂ (k = 0.003 min⁻¹) and SnO₂ (0.002 min⁻¹) have lower value of rate constant than ternary nanocomposite. The order of rate constant follow the same trend for MB dye, but their degradation rate is lower than CR. The GST nanocomposite shows the highest photocatalytic rate constant value of 0.046 min⁻¹ and 0.026 min⁻¹ for CR and MB dyes, respectively.

In order to understand the reasons behind this enhanced photocatalytic performance of the ternary composite, some other characterizations, including PL, electrochemical analysis, and nitrogen adsorption–desorption were carried out. Photoluminescence (PL) measurement is an important tool to measure the fluorescence decay of the sample which in turn can help to understand the transfer mechanism of photon-induced electrons among the nanocomposite. Fig. 10 shows the photoluminescence (PL) spectra of nanocomposites which are used to analyze the efficiency of charge carrier trapping and the lifetime of electron/hole pair re-combination. The samples were scanned by incident light of wavelengths ranging from 400 nm to 500 nm at room temperature. Distinct peaks were obtained at 436 nm, 448 nm and 466 nm for pure TiO₂ as a result of electron–hole pair recombination.^90,91 All these emission bands can be assigned to the free exciton emission of TiO₂. For GT sample, the visible and UV emission were suppressed which can be due to the charge transfer from the trapped states and conduction band of TiO₂ to GO.⁹² Specifically, the GST ternary composite shows the weakest PL intensity among the three samples, indicating the longest lifetime and the most efficient separation efficiency of electron–hole-pairs. This phenomenon suggests the incorporation of GO with SnO₂ and TiO₂ can improve the separation of photo induced electrons and holes. It can be inferred that GST should possess the best photocatalytic properties which is consistent with the results of photocatalytic activity.

Fig. 10 Photoluminescence spectra of synthesized (a) TiO₂, (b) GT and (c) GST nanocomposite.

In a typical photocatalytic reaction by nanocomposite, the porous structure and high surface area play an very important role for the diffusion of charge carriers which in turn enhance the photocatalytic activity due to the availability of more active sites at the surface for the adsorption of reactant molecules. Further, from XRD analysis, it was found that metal oxide nanoparticles possibly act as a spacer during the nanocomposite synthesis as they help in avoiding restacking of GO sheets and irreversible agglomeration due to van der Waals interactions. In reality, this should lead to an increase in the surface area of the composite.^70,71 To gain insights about this possible impact, N₂ adsorption–desorption isotherm of binary and ternary nanocomposite were carried out and are shown in Fig. 11a–c. All the samples exhibit type IV isotherm with an associated H1-type hysteresis loop based on the IUPAC classification.⁹³ The surface areas of GS and GT binary nanocomposite are found to be 60.8 m² g⁻¹ and 110.8 m² g⁻¹ with the pore volume of 0.11 cm³ g⁻¹ and 0.28 cm³ g⁻¹, respectively as shown in Table 1. After the formation of the GST ternary nanocomposite, the surface area was significantly increased to 160 m² g⁻¹, and the corresponding pore volume was also increased to 0.42 cm³ g⁻¹. It is anticipated that, this observed increase in the surface area could be one of the factors responsible for the enhanced photocatalytic activity of the GST nanocomposite towards dye degradation by providing efficient adsorption and transport of charge carriers. Moreover, the high surface area justify the idea that the metal oxide nanoparticles possibly acted as spacer during the nanocomposite synthesis as they help in avoiding restacking of GO sheets and irreversible agglomeration due to van der Waals' interaction. After the formation of the GST ternary nanocomposite, the surface area was significantly increased to 160 m² g⁻¹, and the corresponding pore volume was also increased to 0.42 cm³ g⁻¹. It is believed that introduction of GO in the composite reduces the aggregation of SnO₂ and TiO₂ particles that lead to a higher surface area in GST, similar to that observed in other systems.^94–96 In addition, synergistic effect between the individual component play an important role for larger surface area.^97,98

Fig. 11 N₂ adsorption–desorption isotherm of GT, GS and GST nanocomposite.

To further confirm the separation capability of the photogenerated electron–hole pairs and prolongation of the lifetime of the carriers, photo-electrochemical measurements were carried out. The dark currents and photo-excited currents under vertical light beam illumination are measured using a standard two-probe technique. Fig. 12a illustrates the photo response switching behaviour of the different nanocomposite such as GST, GT, GS, and GO. It can be observed that the photocurrent can be reproducibly switched from the “ON” state to the “OFF” state by periodically turning the light on and off at a low bias voltage of 0.5 V bias. Upon illumination, the photocurrent rapidly increased to a stable value of approximately 3.4 μA for GST nanocomposite, and then drastically decreased to its initial level when the light was turned off, indicating the excellent stability and reproducible characteristics of the photodetector. Fig. 12b shows the response time and recovery time of our device and were found to be around 26 s and 30 s, respectively. In comparison with other GO-based nanocomposite photodetector devices, our results reveal that the GST photodetector exhibited faster photo response characteristic. It was clearly observed that photocurrent of GST composite is enhanced significantly, demonstrating the longer lifespan and more efficient charge separation of the ternary composite in comparison to others. The I–V characteristics of the flexible GST photodetector in the dark and under 420 nm visible light illumination are shown in Fig. S6c and d.† The room temperature I–V curves are measured at a bias ranging from −3 to +3 V in air. It was observed that the I_vis/I_dark of GST nanocomposite is higher than the other GO-based composites (Fig. S6d†).

Fig. 12 (a) The photocurrent-response of GST nanocomposite, GT nanocomposite, GS nanocomposite and GO at applied potential 0.5 V (b) the rise time (t_r) and the decay time (t_d) of different nanocomposite.

To determine further the advantage of the ternary GST nanocomposite over binary GS, GT nanocomposite and pure GO, we have also performed electrochemical impedance spectroscopy (EIS) Nyquist plot, a powerful tool to probe the charge transfer processes. Fig. 13 shows the Nyquist plots of the GST, GS, GT and pure GO electrode materials under visible-light irradiation obtained in 0.2 M Na₂SO₄ electrolyte solution. As can be seen from the figure, the GST ternary nanocomposite exhibits the smallest impedance arc radius among all the samples. Since high-frequency arc in the Nyquist plots is related to the charge-transfer limiting process and ascribed to the double-layer capacitance (C_dl) in parallel with the charge transfer resistance (R_ct) at the contact interface between electrode and electrolyte solution, smallest impedance arc of the ternary GST heterostructure suggest the presence of least charge transfer resistance than that in binary and GO.^99–102 This in turn should contribute to the best separation efficiency of photogenerated electron–hole pairs and enhanced photo activity of the ternary GST nanocomposite. This data is consistent with the BET data where surface area of the ternary hybrid (160 m² g⁻¹) is larger than its binary GS and GT nanocomposite (60.8 m² g⁻¹ and 110.8 m² g⁻¹), suggesting that the ternary system could provide more active sites for photocatalytic reactions to proceed.⁶⁰ It is interesting to find that in all the other samples one arc or semicircle was present, suggesting the involvement of only surface charge transfer step in the photocatalytic reaction.¹⁰³

Fig. 13 Impedance spectra of synthesized GO, GT, GS and GST nanocomposite.

For further understanding the photochemical behaviour of the photocatalyst, the photocatalytic degradation of MB and CR mixed aqueous solutions were also investigated under similar experimental conditions. The characteristic absorption peaks of MB and CR at 498 nm and 665 nm, respectively decreases subsequently with time. After 75 min solar light irradiation, all the absorption peaks of the MB and CR dyes solution is subsequently decreased with no observation of any new peak, indicating the extraordinary photocatalytic performance of GST nanocomposite as shown in Fig. S7.†¹⁰⁴

The order of the photocatalytic efficiency of the prepared samples follow: GST > GT > GS > TiO₂> SnO₂. This behaviour can be attributed to the following reasons. Sunlight contains 4% of UV light, 40–50% of visible light and remaining constitutes the IR.^80,105 It is well known that pure SnO₂ and pure TiO₂ nanoparticles are a wide band gap semiconducting materials and it has also been revealed from the results of UV-vis studies. Therefore, pure SnO₂ and TiO₂ nanoparticles absorb only UV light and do not show efficient photo catalytic degradation of dye under solar irradiation. On the other hand, when graphene oxide is incorporated with both SnO₂ and TiO₂ nanoparticles to synthesize GST nanocomposite, absorption in the UV and visible region could be well noticed. Owing to this, GST nanocomposite probably generates more free charge carriers to induce surface chemical reactions under sunlight irradiation. This can be supported by the fact that the same GST composite showed only 72% and 62% for MB and CR degradation under visible light (with UV filter) and 30% and 39% for MB and CR under UV light (see ESI Fig. S8†) irradiation, respectively. Based on the above analyses, a possible photocatalytic mechanism over the ternary GST nanocomposite was proposed and shown in Fig. 14. To have a thorough understanding, a schematic of the band alignment has been drawn between the three components with regard to their energy band values. The calculated work function of graphene oxide sheets and the conduction band (CB) edge of TiO₂ are found to be −4.4 eV and −4.2 eV versus vaccum, respectively, which are less negative than SnO₂ CB edge of (−4.6 eV vs. vaccum). Owing to this favorable energy level, electrons will transfer from TiO₂ to GO or SnO₂ when they are in contact. Under the irradiation of sunlight, the electrons (e⁻) are excited from the valence band (VB) of TiO₂ to its conduction band (CB), leaving holes (h⁺) in the VB. Immediately, the photogenerated electrons transfer from the CB of TiO₂ to GO, in which GO serves as a electron collector and transporter. Subsequently, the electrons can be readily trapped by SnO₂ owing to their intimate interfacial contact which in turn retard the recombination of photogenerated electron–hole pairs. The decrease in intensity of the PL signal in the composite suggest that enhanced charge separation of the photogenerated electron–hole pairs. The excited electrons were trapped by the adsorbed molecular oxygen on the SnO₂ surface to produce superoxide anion (O₂˙⁻) radicals. Electrons left on TiO₂ also could be trapped to form O₂˙⁻ radicals. The photogenerated holes on VB of TiO₂ can directly oxidate water to generate OH˙ radicals. These superoxide radical and hydroxyl radicals generated from GST nanocomposite can cause the oxidative decomposition of organic dyes (MB and CR) to CO₂, H₂O, and other mineralization products.^106,107 The intimate interfacial contact among individual components may also play an important role in the significantly enhanced photoactivity over the ternary nanocomposite. Moreover, due to the adsorption driven by the π–π conjugation between dye and aromatic graphene oxide, more organic molecules are adsorbed on the catalyst surface resulting a greater probability of photocatalytic reaction.¹⁰⁸

Fig. 14 Photocatalytic mechanism for the degradation of GST nanocomposite under sunlight.

6. Comparison with other reported systems

Later on, to check the efficiency of our photocatalyst, the degradation efficiency of the present system are compared with the earlier reports and are presented in Table 2. It should be noted that the earlier studies have used high power xenon and mercury lamps for the degradation of very low concentration of dye with more amount of photocatalyst. But from Table 2, it can be seen that our catalyst exhibited higher degradation efficiency for high concentration dye with less amount of photocatalyst in lesser time in mild condition as compared to the other reported system. These results confirmed that our ternary GST photocatalyst is more effective and is superior than that of those observed in the previous reports.^47,109–114

Table 2 Degradation percentages of dye obtained by different photocatalysts

Photocatalyst	Dyes	Concentration (mg L^−1)	Irradiated source	Volume/catalyst ml : mg	% degradation : time	References
RGO–TiO2–ZnO	MB	0.3	300 W xenon lamp	100 : 100	92 : 120 min	47
Cu–P25–G	MB	10	Xenon lamp	100 : 100	98 : 100 min	109
TiO2/Cu2O/rGO	MB	25	Sunlight	100 : 80	100 : 300 min	110
GO–CdS/CoFe2O4	MB	10	300 W xenon lamp	250 : 100	98 : 120 min	111
ZnS/CdS/Ag2S	CR	12	Sunlight	50 : 50	98 : 120 min	112
CuInSe2–ZnO	CR	30	400 W + 500 W Hg vapour lamp	20 : 25	70 : 120 min	113
RGO/PANI/Cu2O	CR	10	450 W xenon	50 : 25	98 : 20 min	114
GO–SnO2–TiO2	MB	20	Sunlight	50 : 10	96 : 60 min	Present work
GO–SnO2–TiO2	CR	20	Sunlight	50 : 10	98 : 70 min	Present work

---

7. Conclusions

In summary, ternary GO–SnO₂–TiO₂ composite was successfully fabricated through an one step solvothermal method and its properties were thoroughly evaluated. The XRD diffraction patterns and EDS techniques confirmed that the final nanocomposite is a combination of the GO, SnO₂, and TiO₂ and the FESEM and TEM images demonstrated both SnO₂ and TiO₂ nanoparticles are uniformly distributed across the surface of the GO sheets. It was found that the ternary nanocomposite exhibited promising photocatalytic performances towards the degradation of MB and CR dyes under solar light. The favourable energy level between GO, SnO₂, and TiO₂ in the ternary nanocomposite results in faster photogenerated electron relay, longer lifetime of photogenerated electron–hole pairs, lower photoinduced hole–electron recombination rates, and increased surface area. Additionally, photoelectron chemical studies confirmed the separation of photogenerated electron–hole pairs in the ternary composite. It is expected that our work will shed new insight for the judicious fabrication of graphene-based nanocomposite photocatalysts, which would have the potential to improve their capacity for potential photocatalytic reactions significantly.

Acknowledgements

The authors are thankful to Department of Science & Technology (DST), Govt. of India, for funding and the Sophisticated Analytical Instrument Facility, IIT Madras for providing FT-RAMAN facility. XPS facility at IIT Delhi is partially funded by FIST grant of DST, India.

References

1. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
2. S. Erdemoglu, S. K. Aksu, F. Sayılkan, B. Izgi, M. Asilturk, H. Sayılkan, F. Frimmel and S. Guçer, J. Hazard. Mater., 2008, 155, 469–476.
3. J. Wang, Y. Jiang, Z. Zhang, G. Zhao, G. Zhang, T. Ma and W. Sun, Desalination, 2007, 216, 196–208.
4. R. Ramakrishnan, S. Kalaivani, J. A. I. Joice and T. Sivakumar, Appl. Surf. Sci., 2012, 258, 2515–2521.
5. R. K. Wahi, W. W. Yu, Y. Liu, M. L. Mejia, J. C. Falkner, W. Nolte and V. L. Colvin, J. Mol. Catal. A: Chem., 2005, 242, 48–56.
6. L. Lin, R. Y. Zheng, J. L. Xie, Y. X. Zhu and Y. C. Xiel, Appl. Catal., B, 2007, 76, 196–202.
7. A. Purwanto, H. Widiyandari, T. Ogi and K. Okuyama, Catal. Commun., 2011, 12, 525–529.
8. S. G. Kumar and L. G. Devi, J. Phys. Chem. A, 2011, 115, 13211–13241.
9. G. Madhusudanaa, P. S. Kumar, D. P. Kumar, V. V. S. S. Srikanth and M. V. Shankar, Mater. Lett., 2014, 128, 183–186.
10. D. P. Kumar, M. V. Shankar, M. M. Kumari, G. Sadanandam, B. Srinivas and V. Durgakumar, Chem. Commun., 2013, 49, 9443–9445.
11. M. Muruganandham, R. Amutha, E. Repo, M. Sillanpaa, Y. Kusumotoc and M. A.-A. Mamun, J. Photochem. Photobiol., A, 2010, 216, 133–141.
12. S. Peng, L. Li, Y. Wu, L. Jia, L. L. Tian, M. Srinivasan, S. Ramakrishn, Q. Yan and S. G. Mhaisalkar, CrystEngComm, 2013, 15, 1922–1930.
13. M. V. Shankar, S. Nelieu, L. Kerhoas and J. Einhorn, Chemosphere, 2007, 66, 767–774.
14. S. Shi, M. A. Gondal, A. A. Al-Saadi, R. Fajgar, J. Kupcik, X. Chang, K. Shen, Q. Xu and Z. S. Seddig, J. Colloid Interface Sci., 2014, 416, 212–219.
15. Y. Min, K. Zhang, Y. Chen and Y. Zhang, Chem. Eng. J., 2011, 175, 76–83.
16. R. Nakano, R. Chand, E. Obuchi, K. Katoh and K. Nakano, Chem. Eng. J., 2011, 176–177, 260–264.
17. Y. Shi, H. Li, L. Wang, W. Shen and H. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 4800–4806.
18. J. S. Jang, K. Y. Yoon, X. Xiao, F.-R. F. Fan and A. J. Bard, Chem. Mater., 2009, 21, 4803–4810.
19. S. Pei and H.-M. Cheng, Carbon, 2012, 50, 3210–3228.
20. C. X.-L. Zhang Xiao-Yan, Acta Phys.-Chim. Sin., 2009, 25, 1829–1834.
21. N. Zhang, Y. Zhang and Y.-J. Xu, Nanoscale, 2012, 4, 5792–5813.
22. Y. Yu, J. C. Yu, J.-G. Yu, Y.-C. Kwok, Y.-K. Che, J.-C. Zhao, L. Ding, W.-K. Ge and P.-K. Wong, Appl. Catal., A, 2005, 289, 186–196.
23. G. Demazeau, J. Mater. Chem., 1999, 9, 15–18.
24. V. Štengl, S. Bakardjieva, N. Murafa, V. Houšková and K. Lang, Microporous Mesoporous Mater., 2008, 110, 370–378.
25. S. K. Panda, A. Dev and S. Chaudhuri, J. Phys. Chem. C, 2007, 111, 5039–5043.
26. X. Wang, Y. Li, M. Wang, W. Li, M. Chena and Y. Zhaoab, New J. Chem., 2014, 38, 4182–4189.
27. M. R. Vaezi, J. Mater. Process. Technol., 2008, 205, 332–337.
28. C. F. Lin, C. H. Wu and Z. Nanonn, J. Hazard. Mater., 2008, 154, 1033–1039.
29. J. Nayak, S. N. Sahu, J. Kasuya and S. Nozaki, Appl. Surf. Sci., 2008, 254, 7215–7218.
30. M. Ristic, S. Popovic and S. Music, Mater. Lett., 2004, 58, 2494–2499.
31. K. K. Senapati, C. Borgohain and P. Phukan, Catal. Sci. Technol., 2012, 2, 2361–2366.
32. C. Wang, X. Wang, B.-Q. Xu, J. Zhao, B. Mai, P. A. Peng, G. Sheng and J. Fu, J. Photochem. Photobiol., A, 2004, 168, 47–52.
33. A. Dodd, A. McKinley, M. Saunders and T. Tsuzuki, Nanotechnology, 2006, 17, 692–698.
34. L. Zheng, Y. Zheng, C. Chen, Y. Zhan, X. Lin, Q. Zheng, K. Wei and J. Zhu, Inorg. Chem., 2009, 48, 1819–1825.
35. X. Zhou, T. Zhou, J. Hu and J. Li, CrystEngComm, 2012, 14, 5627–5633.
36. T. Gao and T. Wang, Chem. Commun., 2004, 2558–2559.
37. W. Han, L. Ren, L. Gong, X. Qi, Y. Liu, L. Yang, X. Wei and J. Zhong, ACS Sustainable Chem. Eng., 2014, 2, 741–748.
38. Q. Xiang and J. Yu, J. Phys. Chem. Lett., 2013, 4, 753–759.
39. Q. Xiang, J. Yu and M. Jaroniec, Chem. Soc. Rev., 2012, 41, 782–796.
40. J. Low, J. Yu, Q. Li and B. Cheng, Phys. Chem. Chem. Phys., 2014, 16, 1111–1120.
41. Z. Gao, N. Liu, D. Wu, W. Tao, F. Xu and K. Jiang, Appl. Surf. Sci., 2012, 258, 2473–2478.
42. N. R. Khalid, Z. Hong, E. Ahmed, Y. Zhang, H. Chan and M. Ahmad, Appl. Surf. Sci., 2012, 258, 5827–5834.
43. Y. Ma, F. L. Formal, A. Kafizas, S. R. Pendlebury and J. R. Durrant, J. Mater. Chem. A, 2015, 3, 20649–20657.
44. T. Xian, H. Yang, L. Di, J. Ma, H. Zhang and J. Dai, Nanoscale Res. Lett., 2014, 9, 327–336.
45. A. Meng, J. Shao, X. Fan, J. Wang and Z. Li, RSC Adv., 2014, 4, 60300–60305.
46. D. A. Reddy, R. Ma, M. Y. Choi and T. K. Kim, Appl. Surf. Sci., 2015, 324, 725–735.
47. F. T. Johra and W.-G. Jung, Appl. Catal., A, 2015, 49, 52–57.
48. Y. Xu, K. Sheng, C. Li and G. Shi, ACS Nano, 2010, 4, 4324–4330.
49. K. F. Zhou, Y. H. Zhu, X. L. Yang, X. Jiang and C. Z. Li, New J. Chem., 2011, 35, 353–359.
50. S. J. Ding, D. Y. Luan, F. Y. C. Boey, J. S. Chen and X. W. Lou, Chem. Commun., 2011, 47, 7155–7157.
51. J. S. Chen, Y. L. Tan, C. M. Li, Y. L. Cheah, D. Y. Luan, S. Madhavi, F. Y. C. Boey, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2010, 132, 6124–6130.
52. S. J. Ding, J. S. Chen, G. G. Qi, X. N. Duan, Z. Y. Wang, E. P. Giannelis, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2011, 133, 18522–18525.
53. H. Wang, J. T. Robinson, X. Li and H. Dai, J. Am. Chem. Soc., 2009, 29, 9910–9911.
54. S. Dubin, S. Gilje, K. Wang, V. C. Tung, K. Cha, A. S. Hall, J. Farrar, R. Varshneya, Y. Yang and R. B. Kaner, ACS Nano, 2010, 4, 3845–3852.
55. Y. Zhou, Q. Bao, L. A. L. Tang, Y. Zhong and K. P. Loh, Chem. Mater., 2009, 13, 2950–2956.
56. J. Fan, S. Liu and J. Yu, J. Mater. Chem., 2012, 22, 17027–17036.
57. S.-K. Park, S.-H. Yu, N. Pinna, S. Woo, B. Jang, Y.-H. Chung, Y.-H. Cho, Y.-E. Sung and Y. Piao, J. Mater. Chem., 2012, 22, 2520–2525.
58. A. Kumar, L. Rout, R. S. Dhaka, S. L. Samala and P. Dash, RSC Adv., 2015, 5, 39193–39204.
59. P. Benjwal and K. K. Kar, Mater. Chem. Phys., 2015, 160, 279–288.
60. N. Zhang, Y. Zhang, X. Pan, M.-Q. Yang and Y.-J. Xu, J. Phys. Chem. C, 2012, 116, 18023–18031.
61. J. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
62. H. L. Guo, X. F. Wang, Q. Y. Qian, F. B. Wang and X. H. Xia, ACS Nano, 2009, 3, 2653–2659.
63. H. K. Jeong, Y. P. Lee, R. J. W. E. Lahaye, M. H. Park, K. H. An, I. J. Kim, C. W. Yang, C. Y. Park, R. S. Ruoff and Y. H. Lee, J. Am. Chem. Soc., 2008, 130, 1362–1366.
64. H. P. Cong, X. C. Ren, P. Wang and S. H. Yu, ACS Nano, 2012, 6, 2693–2703.
65. C. Z. Zhu, S. J. Guo, Y. X. Fang and S. J. Dong, ACS Nano, 2010, 4, 2429–2437.
66. D. Briggs and G. Beamson, High resolution XPS of organic polymers-Tbe Scienta ESCA 300 data base, John Wiley and Sons, New York, 1992.
67. S. Dutta, R. Sahoo, C. Ray, S. Sarkar, J. Jana, Y. Negishib and T. Pal, Dalton Trans., 2015, 44, 193–201.
68. J. Ma, X. Wang, Y. Liu, T. Wu, Y. Liu, Y. Guo, R. Li, X. Sun, F. Wu, C. Li and J. Gao, J. Mater. Chem. A, 2013, 1, 2192–2201.
69. M. Zong, Y. Huang, Y. Zhao, X. Sun, C. Qu, D. Luo and J. Zheng, RSC Adv., 2013, 3, 23638–23648.
70. H. Song, L. Zhang, C. He, Y. Qu, Y. Tian and T. Lv, J. Mater. Chem., 2011, 21, 5972–5977.
71. X. S. Zhou, L. J. Wan and Y. G. Guo, Adv. Mater., 2013, 25, 2152–2157.
72. W. Han, L. Ren, X. Qi, Y. Liu, X. Wei, Z. Huang and J. Zhong, Appl. Surf. Sci., 2014, 299, 12–18.
73. G. D. Park, J. S. Choa and Y. C. Kang, Nanoscale, 2015, 7, 16781–16788.
74. Q. Guo and X. Qin, ECS Solid State Lett., 2013, M41–M43.
75. J. Cherusseri and K. K. Kar, RSC Adv., 2015, 5, 34335–34341.
76. S. Li, W. Xie, S. Wang, X. Jiang, S. Peng and D. He, J. Mater. Chem. A, 2014, 2, 17139–17145.
77. S. Yang, W. Yue, J. Zhu, Y. Ren and X. Yang, Adv. Funct. Mater., 2013, 23, 3570–3576.
78. X. Liu, J. Cui, J. Sun and X. Zhang, RSC Adv., 2014, 4, 22601–22605.
79. Y.-X. Wang, Y.-G. Lim, M.-S. Park, S.-L. Chou, J. H. Kim, H.-K. Liu, S.-X. Dou and Y.-J. Kim, J. Mater. Chem. A, 2014, 2, 529–534.
80. Y. Fan, D. Han, B. Cai, W. Ma, M. Javed, S. Gan, T. Wu, M. Siddiq, X. Dong and L. Niu, J. Mater. Chem. A, 2014, 2, 13565–13570.
81. H. Zhang, X. Lv, Y. Li, Y. Wang and J. Li, ACS Nano, 2010, 4, 380–386.
82. D. A. Reddy, R. Ma and T. K. Kim, Ceram. Int., 2015, 41, 6999–7009.
83. S. Xu, L. Fu, T. S. H. Pham, A. Yu, F. Han and L. Chen, Ceram. Int., 2015, 41, 4007–4013.
84. S. Gayathri, P. Jayabal, M. Kottaisamy and V. Ramakrishnan, J. Phys. D: Appl. Phys., 2015, 48, 415305–415316.
85. L. Rout, P. Rengasamy, B. Ekka, A. Kumar and P. Dash, Nano, 2015, 4, 1550059.
86. S. Sakthive and H. Kisch, Angew. Chem., Int. Ed., 2003, 42, 4908–4911.
87. W. Ren, Z. H. Ai, F. L. Jia, L. Z. Zhang, X. X. Fan and Z. G. Zou, Appl. Catal., B, 2007, 69, 138–144.
88. H. Zhang, X. Lv, Y. Li, Y. Wang and J. Li, ACS Nano, 2010, 4, 380–386.
89. T. Surendar, S. Kumara and V. Shanker, Phys. Chem. Chem. Phys., 2014, 16, 728–735.
90. J. Yang, J. Zheng, H. Zhai, X. Yang, L. Yang, Y. Liu, J. Lang and M. Gao, J. Alloys Compd., 2010, 489, 51–55.
91. F. Vietmeyer, B. Seger and P. V. Kamat, Adv. Mater., 2007, 19, 2935–2940.
92. G. Williams and P. V. Kamat, Langmuir, 2009, 25, 13869–13873.
93. X. Liu, J. Cui, J. Sun and X. Zhang, RSC Adv., 2014, 4, 22601–22605.
94. R. Saravanan, M. M. Khan, V. K. Gupta, E. Mosquera, F. Gracia, V. Narayanan and A. Stephen, J. Colloid Interface Sci., 2015, 452, 126–133.
95. R. Saravanan, M. M. Khan, V. K. Gupta, E. Mosquera, F. Gracia, V. Narayanan and A. Stephen, RSC Adv., 2015, 5, 34645–34651.
96. R. Rajendran, L. K. Shrestha, K. M. M. Subramanian, R. Jayavel and K. Ariga, J. Mater. Chem. A, 2014, 2, 18480–18487.
97. T. He, Z. Zhou, W. Xu, Y. Cao, Z. Shi and W. P. Pan, Compos. Sci. Technol., 2010, 70, 1469–1475.
98. K. Jayanthi, S. Chawla, K. N. Sood, M. Chhibara and S. Singh, Appl. Surf. Sci., 2009, 255, 5869–5875.
99. N. Jiang, Z. Xiu, Z. Xie, H. Li, G. Zhao, W. Wang, Y. Wu and X. Haoa, New J. Chem., 2014, 38, 4312–4320.
100. E. Gao and W. Wang, Nanoscale, 2013, 5, 11248–11256.
101. N. Zhang, Y. Zhang, X. Pan, X. Fu, S. Liu and Y.-J. Xu, J. Phys. Chem. C, 2011, 115, 23501–23511.
102. A. Iwase, Y. H. Ng, Y. Ishigur, A. Kudo and R. Amal, J. Am. Chem. Soc., 2011, 133, 11054–11057.
103. J. Fan, S. Liu and J. Yu, J. Mater. Chem., 2012, 22, 17027–17036.
104. A. K. Gupta, A. Pal and C. Sahoo, Dyes Pigm., 2006, 69, 224–232.
105. W.-S. Wang, H. Du, R.-X. Wang, T. Wen and A.-W. Xu, Nanoscale, 2013, 5, 3315–3321.
106. A. Fujishima, T. N. Rao and D. A. Tryk, J. Photochem. Photobiol., C, 2000, 1, 1–21.
107. L. Ye, Y. Su, X. Jin, H. Xie and C. Zhang, Environ. Sci.: Nano, 2014, 1, 90–112.
108. Z. Chen, N. Zhang and Y.-J. Xu, CrystEngComm, 2013, 15, 3022–3030.
109. Z. Jin, W. Duan, B. Liu, X. Chen, F. Yang and J. Guo, Appl. Surf. Sci., 2015, 356, 707–718.
110. B. M. Almeida, M. A. Melo, J. Bettini, J. E. Benedetti and A. F. Nogueira, Appl. Surf. Sci., 2015, 324, 419–431.
111. S. Singh and N. Khare, RSC Adv., 2015, 5, 96562–96572.
112. K. Kalpana and V. Selvaraj, RSC Adv., 2016, 6, 4227–4236.
113. M. Bagheri, A. R. Mahjoub and B. Mehri, RSC Adv., 2014, 4, 21757–21764.
114. J. Miao, A. Xie, S. Li, F. Huang, J. Cao and Y. Shen, Appl. Surf. Sci., 2016, 360, 594–600.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra02067d

---