Facile preparation of a ZnS/ZnO nanocomposite for robust sunlight photocatalytic H₂ evolution from water

Abstract

A ZnS/ZnO nanocomposite was facilely and cost-effectively prepared for visible light photocatalytic H₂ evolution. ZnS/ZnO showed good H₂ evolution activity (187 μmol g⁻¹ h⁻¹) and stability (nearly linear H₂ production rate even after 16 h) without any co-catalyst. Furthermore, ZnS/ZnO revealed superior catalytic performance in sunlight-driven H₂ evolution (1807 μmol g⁻¹ h⁻¹). Due to the electronic hybridization of ZnO band structures with ZnS-surface-states, the band bending and surface dipole moment of ZnO could occur. Under visible-light irradiation, the electrons at the ZnO/ZnS interfaces can be excited from the bent valence band level to the conduction bend of ZnO for water reduction, while the holes trapped by ZnS-surface-states can be quenched by the sacrificial reagent (S²⁻/SO₃²⁻). The created ZnS/ZnO interfaces with ZnS-surface-states led to a lower band gap energy and enabled the visible light response.

---

Introduction

Increasing awareness of the importance of the energy crisis and environmental pollution has stimulated great progress in the development of green renewable energy. Hydrogen is regarded as a potential fuel for renewable energy. Photocatalytic water decomposition into hydrogen is a valuable approach to utilize solar energy, which is recognized as a green and promising way to produce clean energy and solve the fossil fuel shortage problem and environmental issues.^1,2 In recent years, visible light photocatalytic hydrogen generation from water by using semiconductor materials has attracted increasing attention.^3–6 In spite of these intensive achievements, developing robust, stable and inexpensive photocatalysts for water splitting remains a serious challenge within the field.

As important II–VI semiconductors, ZnO and ZnS contain only earth-abundant elements and have been intensively studied in a wide range of UV sensors, lasers, field emitters, nanogenerators, solar cells, photocatalysis, etc.⁷ Unfortunately, both ZnO and ZnS had a large band gap (E_g = 3.37 and 3.67 eV for ZnO and ZnS, respectively), and thus there is almost no adsorption in the visible light region, which largely limits their application in visible-light-driven water splitting. Recently, the heterostructures of ZnO and ZnS have attracted theoretical and experimental interest because the combination of these two wide bandgap semiconductors can yield a novel material with the photoexcitation threshold energy lower than the individual components, leading to improved physical and chemical properties.^8,9 According to our knowledge, apart from ZnS/ZnO hybrid nanowires,¹⁰ few literature was reported on ZnS/ZnO nanocomposites used as effective catalysts for visible light photocatalytic hydrogen production. The preparation of reported ZnS/ZnO hybrid nanowires was conducted by a high-temperature hydrothermal method involving the use of expensive, toxic and smelly 16-mercaptohexadecanoic acid, which makes it difficult to be scaled up and does serious harm to environmental protection and people's health. Therefore, new preparation strategy of ZnS/ZnO heterostructure is greatly desirable to make the best use of ZnO and ZnS for solar energy application.

Herein, a ZnS/ZnO nanocomposite is easily prepared by the photodeposition of ZnS on the surface of ZnO nanoparticles in photocatalytic process, and ZnO nanoparticles can be abundantly and conveniently obtained by a facile precipitation-anneal method using inexpensive and nontoxic inorganic salt and ammonium hydroxide as raw materials. The ZnS/ZnO revealed good catalytic activity and stability in visible-light-driven water splitting and natural sunlight-driven hydrogen evolution.

Experimental section

Preparation

The ZnS/ZnO nanocomposite is formed in photocatalytic hydrogen production process using ZnO nanoparticles as raw materials. To obtain ZnO nanoparticles, 20 mL of 6.0 mol L⁻¹ aqueous ammonia was slowly added into 150 mL of 0.01 mol L⁻¹ Zn(CH₃COO)₂ aqueous solution. After stirring for 3 days, the resulting precipitates were separated, washed, dried and annealed at 500 °C (ESI Fig. S1†).

Characterization

X-ray diffraction (XRD) patterns were recorded on a D8 X-ray diffractometer (Bruker AXS, German) to identify the composition and phase of sample. Transmission electron microscopy (TEM) image was collected on a JEM-2100 transmission electron microscope (JEOL, Japan) to examine the morphology and size of sample. Energy-dispersive X-ray spectroscopy (EDX) was taken on the TEM. Raman spectra were recorded on a confocal microscopic Raman spectrometer (Renishow In-Via) with a 785 nm laser light irradiation from 100 to 700 cm⁻¹ at a duration time of 10 s. To detect surface species of sample, X-ray photoelectron spectroscopy (XPS) analysis was conducted using an ESCALAB 250 Xi (Thermo, USA) X-ray photoelectron spectrometer with Al Kα line as the excitation source (hν = 1484.6 eV) and adventitious carbon (284.6 eV for binding energy) was used as reference to correct the binding energy of sample. The photoluminescence (PL) spectra of the photocatalyst were measured on a Cary Eclipse spectrometer (Varian, USA) with an excitation wavelength of 325 nm. Inductively coupled plasma atomic emission spectrometer (ICP-AES, Leeman Prodigy, USA) was used to detect the element content in the sample.

Photocatalytic hydrogen production

The photocatalytic experiments were performed in a 37 mL flask at ambient temperature using a 300 W Xe lamp equipped with UV cut off filter (λ > 400 nm). 10 mg of ZnO was added in an aqueous solution (total volume = 10 mL) containing 0.45 M Na₂S and 0.45 M Na₂SO₃, and then the suspension was stirred for 15 min in the dark. Before each experiment, the suspension was purged with gas mixture of nitrogen and methane (V_nitrogen : V_methane = 4 : 1) for 40 min to remove air. Methane served as the internal standard. Hydrogen gas evolution was measured by gas chromatography (SP-6890, nitrogen as a carrier gas) with a thermal conductivity detector (TCD).

Photoelectrochemical measurement

A working electrode was prepared as follows: 20 mg of sample was dispersed in 4 mL water. 150 μL of the resultant suspension was uniformly droped onto a 1 × 1 cm² indium-tin oxide glass (ITO glass) and then dried at 70 °C for 1 h to obtain the working electrodes. Photocurrents were measured on a CHI600D electrochemical analyzer (Chenhua Instruments Co. Shanghai) in a standard three-electrode system by using the prepared electrode as the working electrode, a Pt net as the counter electrode, and Ag/AgCl as the reference electrode. A 300 W Xe-lamp equipped with a band-glass light filter (>400 nm) served as a light source. The electrolyte was a buffer solution (pH = 6) containing 0.30 mol L⁻¹ hexamethylenetetramine and 0.10 mol L⁻¹ hydrochloric acid.

Results and discussion

In this work, ZnS/ZnO nanocomposite was formed by the photodeposition of ZnS on the surface of ZnO in photocatalytic H₂ production process. When light was bombarded on the surface of ZnO, the holes and electrons generated (eqn (1)). The surface of ZnO underwent dissolution in alkaline sulfide solution for the formation of ZnS on the surface of ZnO particles.¹¹ Na₂S + Na₂SO₃ solution acted as a hole scavenger by three possible routes (eqn (2), (3) and (4)).^12–14 The production of S₂²⁻ ions (eqn (3)) efficiently reacted with SO₃²⁻ ions (eqn (5)) with the generation of S₂O₃²⁻ and S²⁻.¹⁵ In the presence of H⁺, S₂O₃²⁻ ions (eqn (4) and (5)) transformed to HSO₃²⁻ ions and S (eqn (6)). Finally, it became a fact that the formation of ZnS on the surface of ZnO appeared to occur (eqn (7) and (8)) in the aqueous Na₂S + Na₂SO₃ solution with photoirradiation.

ZnO + hν → ZnO (e_CB⁻ + h_VB⁺) (1)

SO₃²⁻ + H₂O + 2h_VB⁺ → SO₄²⁻ + 2H⁺ (2)

2S²⁻ + 2h_VB⁺ → S₂²⁻ (3)

SO₃²⁻ + S²⁻ + 2h_VB⁺ → S₂O₃²⁻ (4)

S₂²⁻ + SO₃²⁻ → S₂O₃²⁻ + S²⁻ (5)

S₂O₃²⁻ + H⁺ → HSO₃⁻ + S (6)

Zn²⁺ + S²⁻ → ZnS (7)

Zn²⁺ + S + 2e⁻ → ZnS (8)

Fig. 1a showed XRD patterns of ZnO and ZnS/ZnO. ZnO displayed the characteristic XRD peaks corresponding to (100), (002), (101), (102), (110), (103), (202), (112) and (201) reflections of wurtzite ZnO (JCPDS no. 36-1451). Diffraction peaks related to impurity was not observed. For ZnS/ZnO, new XRD peaks indexed to (111), (220) and (311) planes of wurtzite ZnS (JCPDS no. 05-0566) appeared, which indicated that ZnS formed on the surface of ZnO. The decreased peak intensity of ZnO in ZnS/ZnO illustrated that the formation of ZnS tended to low the crystallization of ZnO. Fig. 1b presented Raman spectra of ZnO and ZnS/ZnO. A typical Raman band was located at 437 cm⁻¹ for ZnO, which corresponded to the E2 (high) vibrational mode implying the existence of the wurtzite structure.¹⁶ At the same time, there are three new Raman bands at 153, 260 and 342 cm⁻¹ for ZnS/ZnO. The two Raman bands at 342 cm⁻¹ and 260 cm⁻¹ matched the corresponding longitudinal (LO) and transverse (TO) phonon modes of wurtzite ZnS while the Raman band at 153 cm⁻¹ was caused by the small size of the as-synthesized ZnS.^17,18 Similarly, in comparison with ZnO, the decreased Raman band intensity of ZnO was observed for ZnS/ZnO.

Fig. 1 XRD patterns (a) and Raman spectra (b) of ZnO and ZnS/ZnO.

Fig. 2a presented the TEM micrograph of ZnO. It was clearly observed that the size of ZnO nanoparticles was about 50–80 nm. For ZnS/ZnO nanocomposite, the generated ZnS particles with small size of ca. 5–10 nm accumulated on ZnO surface, as shown in Fig. 2b. Fig. 2c demonstrated a representative high-resolution TEM micrograph of ZnS/ZnO. The lattice fringes with the spacing of 0.25 nm and 0.31 nm were in good agreement with the interplanar spacing of the (101) plane of wurtzite ZnO and the interplanar spacing of the (111) plane of wurtzite ZnS, respectively. Fig. 2d gave the EDX spectrum of ZnS/ZnO. The Zn, O and S were obviously observed and the total atomic percents of O and S were similar to that of Zn, further indicating the ZnS/ZnO composite structure.

Fig. 2 TEM images of ZnO (a) and in situ ZnS/ZnO (b), HRTEM (c) image and EDX spectrum (d) of ZnS/ZnO.

Fig. 3 gave XPS analysis results of ZnS/ZnO. Fig. 3a showed the scan survey spectra for the representative ZnS/ZnO. All of the peaks on the curve can be ascribed to Zn, O, S and C elements. The presence of C element mainly came from the hydrocarbon contaminants that commonly existed for XPS. Therefore, it can be concluded that the sample was composed of Zn, O and S only, which was consistent with the XRD results. The position of Zn2p3/2 and Zn2p5/2 peaks for the sample (Fig. 3b) were at about 1021.1 eV and 1044.2 eV, which confirmed that Zn element existed mainly in the form of Zn²⁺ chemical state on the sample surface. Fig. 3c showed the O1s peak for ZnS/ZnO, which can be fitted into two peaks. The lower energy peak located at 531.0 eV corresponded to the oxygen atoms coordinated with Zn atoms, while the higher energy peak centered at 532.5 eV can be ascribed to the oxygen adsorbed on ZnS/ZnO surface. The position of the S2p peak (Fig. 3d) was at about 161.0 eV, which confirmed that S element existed mainly in the form of S²⁻ chemical state on the sample surface.

Fig. 3 (a) Survey scan XPS spectra in the binding energy range 0–1100 eV and high-resolution spectra, (b) Zn2p, (c) O1s, (d) S2p of ZnS/ZnO.

Fig. 4a showed the photocatalytic activity of ZnS/ZnO for H₂ generation under visible light irradiation (λ > 400 nm). H₂ production enhanced with the increase of irradiation time. During 2 h, the corresponding H₂ evolution was 374 μmol g⁻¹, which was higher than that of the reported materials.^10,19 The results illustrated that the prepared ZnS/ZnO was an efficient catalyst for photocatalytic H₂ evolution from water. In addition, the effects of different Na₂S/Na₂SO₃ concentrations were investigated and the results were given in Fig. 4b. The photocatalytic H₂ production first increased with an increase in the Na₂S and Na₂SO₃ concentration. Generally, a certain increase of Na₂S/Na₂SO₃ concentration could be expected for the better diffusion of the reacting species to the surface of photocatalysts. However, a decrease in the H₂ evolution was found with further increase of Na₂S and Na₂SO₃ concentration. There existed an optimal concentration of 0.45 mol L⁻¹ Na₂S/0.45 mol L⁻¹ Na₂SO₃ for H₂ evolution. Obviously, the H₂ evolution rate was more sensitive to S²⁻ anion in comparison with SO₃²⁻. Since Na₂S and Na₂SO₃ played an important role in formation of ZnS/ZnO, it is very necessary to investigate the effects of different Na₂S/Na₂SO₃ concentrations on the ratio of ZnS to ZnO, especially for the effect of Na₂S due to its high responsibility to photocatalytic H₂ evolution. The contents of ZnS in ZnS/ZnO composites under different Na₂S concentrations were summarized in Table 1. As the Na₂S concentration increased from 0.15 to 0.45 M, the weight percent of ZnS increased from 16.58% to 20.70%. Obviously, an increase amount of ZnS-surface-states formed, and thereby led to the increasing photocatalytic H₂ evolution activity. Meanwhile, it was well known that a certain increase of Na₂S concentration as a hole scavenger could be expected for the better diffusion of the reacting species to the surface of photocatalysts subsequently causing the enhance of photocatalytic H₂ evolution activity. However, with the further increase of Na₂S concentration to 0.55 M, the weight percent of ZnS reached to 21.83%. In this way, although more rich ZnS-surface-states existed, the photocatalytic H₂ evolution slightly decreased. According to previously reported literature,^15,20 the largest H₂ evolution can not be obtained in the presence of concentrated Na₂S as a hole scavenger. This possibly explained the decreasing photocatalytic H₂ activity for ZnS/ZnO with more rich ZnS-surface-states at the condition of 0.55 M Na₂S used.

Fig. 4 (a) Photocatalytic hydrogen production during 2 h in a 10 mL aqueous solution containing 0.45 M Na₂S and 0.45 M Na₂SO₃ suspended with 10 mg of ZnO under visible light irradiation (λ > 400 nm); (b) effect of Na₂S/Na₂SO₃ concentration on the photocatalytic hydrogen production during 2 h in a 10 mL aqueous solution containing 0.45 M Na₂SO₃/0.25 M Na₂S suspended with 10 mg of ZnO under visible light irradiation (λ > 400 nm).

Table 1 Content of S for ZnS/ZnO under different Na₂S concentrations determined by ICP-AES^a

Samples	Na2S concentration (M)	S (wt%)	ZnS (wt%)
a The weight percent of ZnS was calculated by the content of S.
ZnS/ZnO	0.15	5.47	16.58
ZnS/ZnO	0.25	5.95	18.04
ZnS/ZnO	0.35	6.48	19.64
ZnS/ZnO	0.45	6.83	20.70
ZnS/ZnO	0.55	7.20	21.83

---

In view of practical applications, besides catalytic activity, the stability and durability are also indispensable to photocatalysts. In order to evaluate the stability of ZnS/ZnO, we performed the long time curse of H₂ evolution experiment under visible light. Fig. 5 presented the H₂ evolution as the function of irradiation time. The H₂ production was nearly linear even after 16 h of irradiation without renewing the sacrificial agents, and more than 2112 μmol g⁻¹ H₂ was produced during the course of the 16 h experiment, indicating the good stability of the ZnS/ZnO sample for photocatalytic H₂ production.

Fig. 5 Photocatalytic hydrogen production in a 10 mL aqueous solution containing 0.45 M Na₂S and 0.45 M Na₂SO₃ suspended with 10 mg of ZnO under visible light irradiation (λ > 400 nm).

The ultimate goal of photocatalyst is high-efficiency application of sunlight and solar energy. Therefore, H₂ evolution was also conducted under direct sunlight outdoors (ESI Fig. S2†). As shown in Fig. 6, H₂ production increased with the prolongation of sunlight irradiation time. During 2 h, H₂ evolution reached 3580 μmol g⁻¹, which was nearly 10 times higher than that under visible light irradiation from Xe lamp. When the sunlight irradiation time prolonged to 4 h, the corresponding H₂ production increased to 7230 μmol g⁻¹. These results indicated that ZnS/ZnO was a robust catalyst for sunlight-driven H₂ evolution from water. Such heterostructure may be expected to be promising candidates as effective sunlight photocatalyst and potential technological applications.

Fig. 6 Photocatalytic hydrogen production during 4 h in a 10 mL aqueous solution containing 0.45 M Na₂S and 0.45 M Na₂SO₃ suspended with 0.01 g of ZnO under sunlight irradiation in Wuxi city 2014. Experimental condition: outdoor temperature: 30–32 °C, time: 10:00–14:00.

According to our knowledge, the reported studies on ZnS/ZnO nanocomposites visible light response have revealed that the band alignment can be shifted at the ZnS/ZnO interface for their visible light activities.^21–23 Especially, Lahiri et al. have observed an increase of the work function of ZnO upon the formation of ZnS monolayer on the ZnO surface, which was attributed to band bending and surface dipole moment due to the electronic hybridization of ZnO band structures with the surface states induced by the ZnS monolayer, resulting in an effective band gap of about 2.8 eV for the visible-light excitation.²⁴

In our ZnS/ZnO nanocomposite, ZnS particles with size of several nanometers formed on ZnO surface. As shown, the Raman analysis (Fig. 1b) and TEM, HRTEM images (Fig. 2b and c) had proved that almost each ZnO nanoparticle was surrounded by many ZnS nanocrystals with directly contacted interfaces. This would create numerous ZnS-surface-states on ZnO nanoparticle surfaces.²⁴ Due to the electronic hybridization of ZnO band structure with the ZnS-surface-states, the band bending of ZnO occurred.²⁴ Thus, as illustrated in Fig. 7, it became possible that the electrons at the ZnO/ZnS interfaces could be excited from the bent valence band level to the conduction bend of ZnO which can be applied for water reduction, while the photogenerated holes trapped by ZnS-surface-states would be subjected to quenching by the sacrificial reagent (S²⁻/SO₃²⁻). In this way, the effective band gap at the ZnO/ZnS interface would be narrowed, which enabled excitation under visible-light irradiation. The similar results have been reported in photocatalytic H₂ evolution with ZnS/ZnO hybrid nanowires photocatalysts.¹⁰

Fig. 7 Graphical representation for the photocatalytic H₂ production.

In order to prove the visible light response for the as-prepared ZnS/ZnO, the absorption spectra of ZnO and ZnS/ZnO were recorded and shown in Fig. 8. The excitonic absorption peak of ZnO was around 355 nm. However, the ZnS/ZnO nanocomposite showed the absorption band at about 380 nm with an absorption edge up to 460 nm, which was red-shifted in comparison with the absorption feature of ZnO. The observation verified that ZnS-surface states on ZnO surface were responsible for the absorption in the visible region. In order to verify the charge transfer mechanism, photoluminescence (PL) emission spectra and photoelectrochemical (PEC) I–t curves were performed to study the charge recombination and transfer behavior of ZnO and ZnS/ZnO as presented in Fig. 9. Fig. 9a gave the PL spectra of ZnO and ZnS/ZnO with an excitation wavelength of 325 nm. The emission peak at about 495 nm for ZnO corresponded to the band gap for the recombination of photogenerated electrons and holes.¹² Compared ZnO with ZnS/ZnO, the peak intensity at 495 nm of ZnO obviously decreased due to the presence of ZnS-surface-states enhancing the charge transfer. Fig. 9b presented the PEC I–t curves of ZnO and ZnS/ZnO electrodes under intermittent visible light irradiation at −0.22 V vs. Ag/AgCl. Compared with ZnO, ZnS/ZnO hybrid photocatalyst showed the higher photocurrent density, which indicated the efficient photogenerated charge transfer between ZnO and ZnS-surface-states for ZnS/ZnO.

Fig. 8 UV-vis-diffuse reflection spectra of ZnO and ZnS/ZnO.

Fig. 9 Comparison of photoluminescence spectra (a) and transient photocurrent responses (I–t) at −0.22 V vs. Ag/AgCl (b) of ZnO and ZnS/ZnO.

Conclusions

In conclusion, a novel preparation strategy of ZnS/ZnO heterostructure was designed and realized by the photodeposition of ZnS on the surface of ZnO nanoparticles, while ZnO nanoparticles can be abundantly and conveniently obtained by a facile precipitation-anneal method using inexpensive and nontoxic inorganic salt and ammonium hydroxide as raw materials. ZnS/ZnO showed the good H₂ evolution activity and stability without any co-catalyst under visible light irradiation. Additionally, ZnS/ZnO was confirmed as an efficient catalyst for sunlight-driven H₂ evolution. The electronic hybridization of ZnO band structures with ZnS-surface-states could cause the band bending and surface dipole moment of ZnO. Under visible light irradiation, the electrons at the ZnO/ZnS interfaces could be excited from the bent valence bend level induced to the conduction bend of ZnO for water reduction, while the holes trapped by ZnS-surface-states can be subjected to quenching by the sacrificial reagent (S²⁻/SO₃²⁻). The created ZnS/ZnO interfaces with ZnS-surface states led to a lower band gap energy of the composite allowing for the excitation under visible-light irradiation. This work could provide new insights into the facial fabrication of photocatalysts with high performance and facilitate their practical application in solar energy field.

Acknowledgements

The authors gratefully acknowledge the support from the National Natural Science Foundation of China (no. 20903048, 21275065), the Fundamental Research Funds for the Central Universities (JUSRP51314B and JUDCF13015), the Postgraduate Innovation Project of Jiangsu Province (CXZZ13-0743) and MOE & SAFEA for the 111 Project (B13025).

Notes and references

1. Q. Wang, J. Li, Y. Bai, J. Lian, H. Huang, Z. Li, Z. Lei and W. Shangguan, Green Chem., 2014, 16, 2728–2735.
2. Z. Yan, X. Yu, Y. Zhang, H. Jia, Z. Sun and P. Du, Appl. Catal., B, 2014, 160–161, 173–178.
3. S. R. Lingampalli, U. K. Gautam and C. N. R. Rao, Energy Environ. Sci., 2013, 6, 3589–3594.
4. P. Gao, Z. Liu and D. D. Sun, J. Mater. Chem. A, 2013, 1, 14262–14269.
5. Y. P. Xie, Z. B. Yu, G. Liu, X. L. Ma and H. M. Cheng, Energy Environ. Sci., 2014, 7, 1895–1901.
6. X. Wang, L. Yin and G. Liu, Chem. Commun., 2014, 50, 3460–3463.
7. X. Huang, M. Wang, L. Shao, M. G. Willinger, C. S. Lee and X. M. Meng, J. Phys. Chem. Lett., 2013, 4, 740–744.
8. X. Huang, M. Wang, M. G. Willinger, L. Shao, D. S. Su and X. M. Meng, ACS Nano, 2012, 6, 7333–7339.
9. W. Chen, H. Ruan, Y. Hu, D. Li, Z. Chen, J. Xian, J. Chen, X. Fu, Y. Shao and Y. Zheng, CrystEngComm, 2012, 14, 6295–6305.
10. Z. Wang, S. W. Cao, S. C. J. Loo and C. Xue, CrystEngComm, 2013, 15, 5688–5693.
11. T. Arai, S. Senda, Y. Sato, H. Takahashi, K. Shinoda, B. Jeyadevan and K. Tohji, Chem. Mater., 2008, 20, 1997–2000.
12. P. Gomathisankar, K. Hachisuka, H. Katsumata, T. Suzuki, K. Funasaka and S. Kaneco, ACS Sustainable Chem. Eng., 2013, 1, 982–988.
13. P. Gomathisankar, K. Hachisuka, H. Katsumata, T. Suzuki, K. Funasaka and S. Kaneco, Int. J. Hydrogen Energy, 2013, 38, 8625–8630.
14. P. Gomathisankar, K. Hachisuka, H. Katsumata, T. Suzuki, K. Funasaka and S. Kaneco, RSC Adv., 2013, 3, 20429–20436.
15. N. Bao, L. Shen, T. Takata and K. Domen, Chem. Mater., 2008, 20, 110–117.
16. A. P. Bhirud, S. D. Sathaye, R. P. Waichal, L. K. Nikama and B. B. Kale, Green Chem., 2012, 14, 2790–2798.
17. C. Lan, K. Hong, W. Wang and G. Wang, Solid State Commun., 2003, 125, 455–458.
18. Q. Wu, H. Cao, S. Zhang, X. Zhang and D. Rabinovich, Inorg. Chem., 2006, 45, 7316–7322.
19. H. Zhou, L. Ding, T. Fana, J. Ding, D. Zhang and Q. Guo, Appl. Catal., B, 2014, 147, 221–228.
20. Z. Yan, H. A. Wu, A. Han, X. Yu and P. Du, Int. J. Hydrogen Energy, 2014, 39, 13353–13360.
21. Y. Hu, H. Qian, Y. Liu, G. Du, F. Zhang, L. Wang and X. Hu, CrystEngComm, 2011, 13, 3438–3443.
22. J. Schrier, D. O. Demchenko and L. W. Wang, Nano Lett., 2007, 7, 2377–2382.
23. Z. Wang, J. Wang, T.-K. Sham and S. Yang, J. Phys. Chem. C, 2012, 116, 10375–10381.
24. J. Lahiri and M. Batzill, J. Phys. Chem. C, 2008, 112, 4304–4307.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra13203c

---