Introduction

Acquisition of clean hydrogen energy by splitting of water using plentiful solar energy is considered as an ideal way to resolve the global renewable energy demand and environment problems [1,2,3,4]. In particular, photocatalytic or photoelectrochemical splitting water is one of the most ideal ways considering resource sustainability, environmental, and cost issues [5, 6]. The urgent work for water splitting by photocatalysis is to design and develop semiconductor photocatalysts with appropriate band gap to make best use of solar energy and band edges to meet oxidization and reduction water requirement as well as high quantum yield and high stability [7]. Up to now, the development of photocatalysts experienced from binary oxides (TiO2, ZnO, Fe2O3) [8], ternary oxides (SrTiO3, K4Nb6O17, NaTaO3) [9], to multi elements compounds (K4Ce2M10O30(M = Ta, Nb) [10], especially solid solution compounds (GaN: ZnO, ZnGeN2-ZnO) [11], and series of (oxy) nitrides (Ta3N5, TaON, LaTiO2N) [12, 13], (oxy) sulfides (Sm2Ti2S2O5, Cu2ZnSnS4) [14] based on band engineering methods, as well as from p block photovoltaic cell semiconductors candidates such as GaInP/GaAs, GaPN, GaAsPN, p-InGaN, etc. [15]. Additionally, the morphologies of film or powder with nanowire, nanorod/nanotube, and nanobelt etc. are extensively controlled [16]. Unfortunately, most of them failed to satisfy the mentioned above requirements simultaneously.

SnO2 is a well-known semiconductor with band gap about 3.6–3.8 eV. However, wide band gap and low conduction band edge (more positive than that of H+/H2) of SnO2 restrict its utilization as a photocatalyst for water splitting [17]. In the most cases, SnO2 was used as a part of composite or coupled photocatalysts, such as in SnO2-TiO2 [18], SnO2-ZnO [19] for its lower conduction band edges to facilitate the transferring photo-generated electrons from the host photocatalyst.

In this communication, visible-light-driven SnO2-x was synthesized by facile one-pot solvothermal method from the precursors of SnCl2·2H2O. The as-prepared powder was identified through X-ray diffraction (XRD) as pure SnO2 phase, and ultraviolet–visible spectroscopy (UV-vis) spectrum indicated its absorption edge about 570 nm, corresponding to band gap of 2.17 eV, presenting good potential to respond to visible light. The photoelectrochemical and photocatalytic water splitting activities under visible light were presented.

Methods

Reagents

All chemicals of analytical grade were purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China, and used as received without further purification. The super pure water (18.25 MΩ cm) was used as solvent for photoelectrode preparation and photocatalytic measurement.

Preparation of Powder SnO2-x

SnO2-x was prepared by conventional solvothermal method with 0.02 mol SnCl2·2H2O (SnCl4·5H2O) dissolved into 100 mL methanol solvent and stirring for 30 min. Then, adjusting the pH value from initial 1.0 to 3.0 by dipping 0.02 mol/L NH3·H2O slowly with stirring, getting white floccule. After reacting for 2 h, the mixture was transferred into a 200 mL Teflon-lined autoclave and heated at 423 K for 20 h. The yellow slurry was obtained from washing with deionized water and ethanol several times, and dried at 343 K for 12 h, got the targeted sample.

Preparation of SnO2-x Electrode

Porous thin film electrodes were prepared by electrophoretic deposition method on conductive fluorine-doped tin oxide glass (FTO, Ahahi Glass Co.). The electrophoretic deposition was carried out in an acetone solution (40 mL) containing as-prepared powder (40 mg) and iodine (15 mg), which was dispersed by sonication for 3 min. The coated area was controlled to be ca. 1.5 × 4 cm. This procedure resulted in the formation of SnO2-x layer with uniform thickness of ca. 2 μm, with good reproducibility.

Photocatalytic Evaluation

The photoelectrochemical measurement was performed by three-electrodes configuration mode consisted of a working electrode (prepared electrode), a counter electrode (Pt mesh), and a reference electrode (Ag/AgCl) as well as electrolyte (0.1 M aqueous Na2SO4 solution) on electrochemical workstation (Autolab PGSTAT 204, Switzerland), and the pH value of the electrolyte solution was adjusted to 4.05 by 0.1 M H2SO4. The solution was purged with Ar for over 10 min before the measurements. The electrodes were irradiated through silicon glass window by a Xe lamp (300 W, Cermax) fitted with a cut-off filter (Hoya L-42) to block light of wave length less than 420 nm.

The photocatalytic activities were carried out in a Pyrex side-irradiation-type reaction vessel connected to a glass closed gas circulation system. A flow of cooling water was used to maintain the reaction system at room temperatures. Then, 0.2 g powder was dispersed into 200 mL solution, irradiated by 300 W Xe-lamp fitted with a cut-off filter (Hoya L-42) to block light of wave length less than 420 nm. The evolved gas was analyzed by gas chromatography with thermal conductivity detector (TCD) detector and Ar as carrier.

Characterizations

The sample was identified by X-ray powder diffraction on Geiger-flex RAD-B, Rigaku; Cu Kα). Scanning electron microscopy (SEM) images were obtained on field-emission scanning electron microscopy (FE-SEM; S-4700, Hitachi). UV-vis diffuse reflectance spectrum was recorded by spectrophotometer (JASCO, V-670). The Brunauer-Emmett-Teller (BET) surface area was measured using a BELSORP-mini instrument (BEL Japan) at 77 K. The elements and valence states of the samples were analyzed by X-ray photoelectron spectroscopy (XPS) (Thermo Fisher K-Alpha, America). Transmission electron microscope (TEM) and high-resolution transmission electron microscopy (HRTEM) images of samples were performed on Tecnai G2 F20 transmission electron microscopy at 200 kV accelerating voltage.

Results and Discussion

The as-prepared powder was identified by XRD patterns. The compositions, absorption properties, and crystallite of as-prepared samples closely depended on the preparation conditions, such as tin precursors (SnCl2·2H2O, SnCl4·5H2O), pH values, and consequently further heat treatment. As an example, this sample prepared by SnCl2·2H2O with methanol as solvent and adjusted pH value to 3.0 by NH3·H2O, XRD pattern identified its pure SnO2 phase with poor crystalline (Fig. 1a), and UV-vis spectrum (Fig. 1c) revealed its absorption edge is about 570 nm, corresponding to band gap of 2.17 eV, showing great potential to response to visible light. While for these SnO2 from precursors of SnCl4·5H2O and SnCl4·5H2O with SnCl2·2H2O (molar ration 1:1) under the same procedures above, their absorption edges are almost the same at about 370 nm. Moreover, with the precursor SnCl4·5H2O, we cannot get visible-light-driven SnO2 by co-precipitation method in air and by hydrothermal method in water. Furthermore, for precursor SnCl2·2H2O in methanol solvent, with the increase of pH value, the obtained powder became the mixture of SnO2 and SnO (Fig. 1b). The XPS of the as-prepared powder was measured to characterize the elemental compositions and chemical states, as shown in Fig. 2. The survey scan spectra (Fig. 2a) of the SnO2 and SnO2-x (SnCl2·5H2O as precursor) sample clearly indicate the obvious peaks of Sn, C, and O. Figure 2b showed that the binding energy of Sn 3d in SnO2-x decreased by 0.2 eV as compared to pure SnO2 (from 486.9 to 486.7 eV for Sn 3d5/2, and from 495.4 to 495.2 eV for Sn 3d3/2). As shown in Fig. 2c, the Sn 3d5/2 signal of SnO2-x sample centered at 486.7 eV can be deconvoluted by the multi-Gaussian function into two parts centered at 486.8 and 485.8 eV assigned to Sn4+ and Sn2+, which confirmed the presence of Sn2+ dopants in the prepared SnO2-x because of the formation of oxygen vacancies which (cut down) the binding energy of Sn 3d to preserve charge neutrality [20]. Figure 2d showed that O 1s transition peak shifted 0.2 eV (from 530.6 to 530.4 eV) upon self-doping of Sn2+, and the formation of oxygen vacancies was also considered to enhance the absorption in the visible-light region [21, 22]. The optical absorption properties of prepared SnO2-x as different precursors were studied by UV-vis DRS spectroscopy (Fig. 1c). The visible light response ability of prepared SnO2-x by SnCl2·2H2O as precursor was attributed to the incorporation of Sn2+ into the lattice of SnO2-x [20]. These obvious differences in control preparation conditions indicated that the visible-light-driven mechanism for as-prepared SnO2-x had great relation with Sn2+ species in oxygen inefficient situation.

Fig. 1
figure 1

XRD patterns of prepared SnO2 a with different precursors, b different pH values of SnCl2·2H2O as precursor (JCPDS#72-1147 and 85-0712 of SnO2 and SnO), and c UV-vis DRS spectra with different precursors

Full size image
Fig. 2
figure 2

Survey XPS spectra (a), the Sn 3d XPS spectra (b), the O 1s XPS spectra (d) of SnO2 and prepared SnO2-x, and c the Sn 3d5/2 XPS spectra of prepared SnO2-x

Full size image

The microstructure of prepared SnO2-x was obtained by SEM, TEM, and HRTEM. The SEM images illustrated regular spherical particle in diameter of about 1–2 μm (Fig. 3a, b), while their BET surface areas are about 100 m2/g, and the crystal size is about 2.5 nm from BET measurement, which is consistent with that from calculation by Scherrer equation. As shown in Fig. 3c, we can see that prepared SnO2-x showed regular spherical particle consisted with SEM image. The HRTEM image (Fig. 3d) indicated that the lattice fringes measured with a spacing of 0.33 nm were clearly visible, corresponding to the (110) atomic plane of SnO2 with a tetragonal cassiterite phase.

Fig. 3
figure 3

SEM (a and b), TEM (c), and HR-TEM (d) images of prepared SnO2-x

Full size image

The photocurrent effect on as-prepared SnO2-x electrode under visible light (λ > 420 nm) was shown in Fig. 4a. For this photoelectrode without any treatment, although there showed obvious photoanodic current, a N-type semiconductor responsive characters, the photocurrent properties are not so normal in slow increase and decrease responsive to light on and off, which may be ascribed to the surface capacity effect. For this with further heat treatment at 150 °C in air, there showed not only the increase of current density, but also the improvement of its photocurrent properties. From Fig. 4a, the as-prepared SnO2-x posed with onset potential less than 0 V Vs reversible hydrogen electrode (RHE), that is to say, the as-prepared SnO2-x with conduction band located negative than that of H+/H2, indicating that the as-prepared SnO2-x can split water without bias potential. To make certain the potential of the band edges for as-prepared SnO2-x, the photocatalytic water decomposition in powder for half reaction under visible light was carried out in a gas circular system. As shown in Fig. 4b, c, the as-prepared SnO2-x demonstrated obvious H2 and O2 evolution activities under visible light irradiation(λ > 420 nm) with the presence of electron donor (methanol) and acceptor (AgNO3) respectively even without any co-catalyst loading and modification. And with the loading of Pt (1 wt.%) by in-situ photo-deposition method from H2PtCl6, the activities were prompted greatly. The hydrogen and oxygen evolution activities under visible light further confirmed that the as-prepared SnO2-x poses appropriate band edges to meet the requirement for water redox reaction. The wavelength dependence on photocurrent (Fig. 4d) showed good agreement with absorption edge, indicating the band transition properties. The photocurrent density of prepared SnO2-x time dependance was measured under visible-light irradiation at a bias potential of 0.6 V Vs RHE (Fig. 4e). After 10,000 s of irradiation, the photocurrent density is slowly reduced to zero. It can be found that the stability of prepared SnO2-x is poor which is due to the oxidation of Sn2+.

Fig. 4
figure 4

Photocurrent effect of prepared SnO2-x. a The photocatalytical activities of H2 evolution (b) and O2 evolution (c). The wavelength dependence on photocurrent effect for prepared SnO2-x (d). e The I-T curve for this parpared SnO2-x

Full size image

SnO2, a known wide band-gap semiconductor, phases with different oxygen composition. Non-stoichiometry of SnO2, in particular oxygen deficiency or impurity dopants, can donate electrons into the conduction band, and the conduction band is a single band of s-type character that is strongly dispersed with a minimum at the T-point of the Brillouin zone, which make it a good electron conduction [23]. Additionally, for these visible-light-driven Sn2+ including compounds Sn2Nb2O7 (SnNb2O6), and Sn2+ ion-exchange Sn2+/K4Nb6O17, Sn2+/KTiNbO5, it was ascribed that the Sn 5 s2 contributes to the top of the valence band, and locates in about 0.7~1.4 eV negative than that of O 2 p [24]. So here, for as-prepared SnO2-x, the visible-light-driven mechanism maybe ascribed to the energy levels that are formed between Sn2+ 5 s orbital and O 2p orbital. On the other hand, the valence state of Sn2+ is more negative than that of Sn4+ (illustrated in Scheme 1) resulting in doping in the lattice that will cause charge imbalance to form oxygen vacancies, which has an effect on the surface properties and charge transfer of the catalyst.

Scheme 1
scheme 1

Schematic diagram for the band structure of pure SnO2 and prepared SnO2-x photocatalyst

Full size image

Conclusion

Cost-effective stannic oxide photocatalyst has been successfully synthesized by facile one-pot solvothermal method from SnCl2·2H2O and methanol. It is significant to show visible light responsive ability and photoelectrolysis water decomposition activities. The visible-light-driven mechanism for this SnO2-x maybe ascribed to self-doping by Sn2+ generating oxygen vacancies to preserve charge neutrality which can enhance the performance of photocatalyst. Further work focusing on the improvement of activities and stability are under investigations.