Preparation of DyVO₄/WO₃ heterojunction plate array films with enhanced photoelectrochemical activity

Abstract

In this work, DyVO₄/WO₃ heterojunction plate arrays were first fabricated on FTO using a hydrothermal method for WO₃ vertical plate arrays and a dipping–annealing process for the deposition of DyVO₄ nanoparticles. The samples were characterized by various techniques, including X-ray diffraction (XRD), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Photoelectrochemical activities were investigated by linear sweep voltammetry (LSV) and incident photon to current conversion efficiency (IPCE). The DyVO₄/WO₃ heterojunction electrode exhibited a maximum photocurrent density of 0.78 mA cm⁻² at +1.2 V (vs. Ag/AgCl), while the photocurrent density of pure WO₃ was just 0.49 mA cm⁻² under illumination. And the highest IPCE value increased from 27.4% to 54.1% after the DyVO₄ nanoparticle deposition. The enhanced PEC performance was attributed to the longer electron lifetime, increased carrier density and high charge separation efficiency at the interface of heterojunction, which were confirmed by electrochemical impedance spectroscopy (EIS) and Mott–Schottky analysis. The study demonstrates that metal orthovanadates may be good heterojunction candidates to couple with WO₃ to provide promising photoanodes for water splitting.

---

1. Introduction

Tungsten oxide (WO₃) has been most widely used as a promising photocatalytic material, because of its small band gap (∼2.6 eV) for effective absorption of the solar spectrum,¹ resistance to photocorrosion,² good stability in acidic solution and a deep valence band to promote the oxygen evolution reaction.³ One practical problem associated with photocatalysts is the undesired electron/hole recombination process which represents the major energy-wasting step, thus limiting the quantum yield.⁴ Many scientific strategies have been put forward to prevent the electron–hole pair recombination and enhance the photoelectrochemical (PEC) performance, such as doping with transition metal ions,⁵ deposition of noble metals,⁶ surface modification⁷ and formation of hybrid heterojunctions.⁸ In fact, many investigations have verified that the coupling with a narrow band gap semiconductor could not only offer the driving forces to effectively separate and transfer photogenerated electron–hole pairs, but also extend the light response, such as CdS/WO₃,⁹ Cu₂O/WO₃ (ref. 10) and Fe₂O₃/WO₃.¹¹

Recently, orthovanadate materials have caused extensive scientific research owing to their interesting optical, catalytic and electrical properties.¹² They are considered as a possible candidate for photocatalytic material. For example, Zou et al. reported that wolframite-type and monoclinic indium vanadate (InVO₄) has a prospective photocatalytic production of H₂ from water under visible light irradiation.¹³ Xu et al. found that yttrium vanadate (YVO₄) nanopowders possess excellent photocatalytic properties for degradation of methyl orange.¹⁴ Notably, only a portion of vanadate salts, which cation has a d¹⁰s⁰ electron configuration or has partially filled 4f orbitals, have a small band gap and could be excited by visible light. The most representative is bismuth vanadate (BiVO₄). The valence band (VB) of BiVO₄ is composed of hybridized Bi 6s and O 2p orbitals. It largely disperses the VB, which narrows the band gap and favors the mobility of photogenerated holes in the VB.¹² For the same reason, lanthanide orthovanadates also have a small band gap and could be promising visible light driven photocatalysts. Up to date, several lanthanide orthovanadates (such as lanthanum vanadate – LaVO₄,¹⁵ cerium vanadate – CeVO₄,¹⁶ and yttrium vanadate – YVO₄ (ref. 14)) have been studied for their PEC activity. Dysprosium vanadate (DyVO₄) is an important lanthanide vanadate with interesting properties, such as the optical properties with a small band gap of 2.3 eV for visible light absorption,¹⁷ and the surface catalytic properties for efficient photodegradation of 2-propanol and benzene.¹² He et al. prepared a novel g-C₃N₄/DyVO₄ heterostructured photocatalyst for Rhodamine B photodegradation, which promoted both the light response and the electron–hole pair separation, thus the higher photocatalytic activity was achieved.¹⁸ However, to the best of our knowledge, few reports focused on the PEC water splitting of DyVO₄ and its composites have been published.

In this paper, we choose WO₃ plate-like arrays as the base material, which offer large surface areas with increased density of active redox sites¹⁹ and provide direct pathways for electron transport.²⁰ Then, a heterojunction film of DyVO₄/WO₃ was formed on the WO₃ surface through a simple dipping-annealing process. The as-prepared DyVO₄/WO₃ heterojunction plates arrays films were characterized by X-ray diffraction (XRD), scanning electron microscope (SEM) and transmission electron microscope (TEM) and X-ray photoelectron spectroscopy (XPS). The enhanced PEC performance of heterojunction film was achieved.

2. Experimental section

2.1 Synthesis of WO₃ plate-like arrays films

The WO₃ plate-like arrays films were prepared according to our previous report.²¹ In the typical hydrothermal synthesis process, 0.231 g of sodium tungstate (Na₂WO₄) was used as the tungsten source to react with 3 M hydrochloric acid (HCl) to form tungsten acid (H₂WO₄), with 0.2 g of ammonium oxalate ((NH₄)₂C₂O₄) as the structure-directing agent. The hydrothermal synthesis process was carried out at 140 °C for 6 h. Then the films were taken out and dried in the oven.

2.2 Synthesis of DyVO₄/WO₃ heterojunction films

A dipping-annealing process was used to synthesize the DyVO₄ nanoparticles on the surface of as-prepared WO₃ plates arrays films. The precursor solution was prepared according to our previous report:²² Dy(NO₃)₃·5H₂O in glacial acid (0.4 M, 3 mL) and vanadyl acetylacetonate in acetylacetone (0.06 M, 20 mL) were mixed with a 1 : 1 mol ratio of Dy : V. Next, the WO₃ films were vertically dipped into above solution for 1 h. Finally, the as-prepared films were annealed at 500 °C in air for 1 h.

2.3 Characterization

The crystalline phase of the electrodes was characterized by X-ray diffraction (XRD, D/Max2250, Rigaku Corporation, Japan) with Cu Kα (λ = 0.15406 nm) radiation. The microscopic morphologies were investigated by a field emission scanning electron microscope (FESEM, Nova NanoSEM 230) and high resolution transmission electron microscope (HRTEM, G2 F20). The surface electronic states were analyzed by X-ray photoelectron spectroscopy (XPS) and the UV-vis spectra were obtained using a diffused reflectance ultraviolet and visible spectrophotometer (DR-UVS, Shimadzu 2450 spectrophotometer).

2.4 Photoelectrochemical measurements

The photoelectrochemical measurements were carried out in a typical three-electrode electrochemical cell under AM 1.5 G illuminations with an electrochemical analyzer (Zennium, Zahner, Germany). The as-prepared films, Ag/AgCl electrode and Pt sheet were used as the working electrode, the reference electrode and the counter electrode, respectively. Visible light irradiation was performed under a 500 W Xe lamp (CHF-XM35, Beijing Trusttech Go. Ltd.) with a 400 nm cutoff filter to remove UV irradiation and the light intensity was adjusted to 100 mW cm⁻². The linear sweep voltammetry (LSV) was measured in 0.2 M Na₂SO₄ solution (pH = 7) with a scanning speed of 20 mV s⁻¹. The Mott–Schottky plots were obtained at the AC frequency of 1 kHz. The electrochemical impedance spectra (EIS) were measured at 0.8 V (vs. Ag/AgCl) with the AC frequence of 10 kHz to 100 mHz.

3. Results and discussion

3.1 Characterization of the synthesized photoanodes

The XRD patterns of WO₃ and DyVO₄/WO₃ are shown in Fig. 1. The sharp diffraction peaks at 23.1°, 23.6° and 24.5° assign to monoclinic WO₃ phase according to JCPDS data (JCPDS 43-1035). There is not any noticeable change in the phase by loading DyVO₄, which suggested that the crystallinity of WO₃ was not influenced by an introduction of DyVO₄. Furthermore, the concentration of DyVO₄ was too low to be detected by XRD.

Fig. 1 XRD pattern of the DyVO₄/WO₃ film.

The microstructures and surface morphology of the films were characterized by SEM. It can be seen that the pure WO₃ film reveals numbers of platelike arrays growing vertically on the FTO substrate with smooth surface (Fig. 2a), while the surface morphology of DyVO₄/WO₃ heterojunction film is rough (Fig. 2b). It is evident that a number of tiny nanoparticles are aggregated to formed a thin shell on the surface of the WO₃ plates. The local composition of DyVO₄/WO₃ film was analyzed by energy dispersive X-ray spectrometer (EDS). The result of EDS analysis shows that the loading molar content of DyVO₄ was about 1.4%. XPS was also employed to analyze the surface composition of the DyVO₄/WO₃ film. The result of XPS analysis shows that the loading molar content of DyVO₄ was 3.6%, which value is larger than examined by EDS. This is due to the different penetration of X-rays between EDS and XPS. The X-rays of XPS only has a penetration depth of tens of nanometers, while the analysis depth of EDS can reach micrometer level.²³ The difference between EDS and XPS results indicates the DyVO₄ nanoparticles are distributed on the surface of WO₃ plates.

Fig. 2 SEM images of WO₃ film (a), DyVO₄/WO₃ heterojunction film (b).

To observe the microstructures and binding mode of DyVO₄ nanoparticles and WO₃ plates, TEM images of DyVO₄/WO₃ heterojunction film is presented in Fig. 3. It can be seen that the base material WO₃ has platelike structures from Fig. 3a. We can also easily recognize the DyVO₄ nanoparticles with size of 10–20 nm which were well attached to the WO₃ plates from the HRTEM image (Fig. 3b). The lattice d-spacing of 0.262 nm examined by Fourier transform corresponds to the (202) plane for the WO₃ plates, and lattice d-spacing of 0.302 nm (112) for the DyVO₄ nanoparticles. The TEM-mapping analysis was conducted to demonstrate the spatial distribution of DyVO₄. The result show that the DyVO₄ particles covered onto the surface of WO₃ plates (Fig. S2†). Combined elemental mapping of W, V with HRTEM image (Fig. 3b), a conclusion can be drawn that WO₃ and DyVO₄ might not form any solid solution but remain segregated phases.

Fig. 3 TEM image of DyVO₄/WO₃ film (a), HRTEM image of DyVO₄/WO₃ film (b).

The composition of DyVO₄/WO₃ heterojunction film was detected by X-ray photoelectron spectroscopy (XPS). The analysis of XPS data was performed using the XPS Peak Fitting Program version 4.1. Fig. 4a shows the overall XPS survey spectrum, which confirms the existence of all the elements of DyVO₄ and WO₃. It is observed that W 4f_7/2 and W 4f_5/2 peaks are located at 35.4 eV and 37.5 eV, respectively (Fig. 4b), which is consistent with the W⁶⁺.²⁴ The O 1s main peak in Fig. 4c was dissociated into two peaks: W–O (530.3 eV) and surface hydroxide –OH (531.6 eV).²⁵ In addition, the peaks at 152.3 eV and 153.8 eV correspond to Dy 4d (Fig. 4e), and the V 2p peak was observed at 517.4 eV (Fig. 4d), which are in good agreement with the reported studies.²⁶ Simultaneously, the potentials for top of the valence band (VB) for the samples were estimated (Fig. 4f). This result indicates that the introduction of DyVO₄ nanoparticles results a 0.3 eV negative shift for the VB top edge of WO₃.

Fig. 4 XPS spectra of the DyVO₄/WO₃ heterojunction film: (a) survey scan; (b) W 4f; (c) O 1s; (d) V 2p; (e) Dy 4d; (f) VB-XPS spectra.

The UV-vis diffuse reflectance spectra of the WO₃ and DyVO₄/WO₃ films are depicted in Fig. 5a. For pure WO₃ film, the optical absorption edge is around 450 nm. The introduction of DyVO₄ onto WO₃ results in better absorption and a little red shift in the visible light range was observed. The band gap energy (E_g) of the samples was calculated by the Tauc formula:²⁷

(αhν)^1/2 = A(hν − E_g) (1)

where α, h, ν and A are the absorption coefficient, the Plank's constant, the frequency of the radiation and a constant, respectively. The pure WO₃ film has a band gap of about 2.70 eV, which is consistent with other reports.^28–30 With the contribution of DyVO₄ nanoparticles (E_g = 2.30 eV (ref. 12)), the DyVO₄/WO₃ heterojunction film has a narrower band gap of 2.68 eV. The loading amount of DyVO₄ is too low to affect the optical absorption properties of WO₃ film.

Fig. 5 UV-vis absorption spectra (a) and the band gap determination (b) of WO₃ and DyVO₄/WO₃ heterojunction films.

3.2 Photoelectrochemical measurements of as-prepared photoanodes

Linear sweep voltammetry (LSV) measurement was employed to investigate the PEC activities of the DyVO₄/WO₃ and WO₃ films under visible irradiation (Fig. 6). The LSV was carried out with a potential range from 0 V to 1.5 V (vs. Ag/AgCl) with a scan rate of 20 mV s⁻¹. The photocurrent densities of both films increased with applied potential under illumination. At +1.2 V (vs. Ag/AgCl), the photocurrent density of DyVO₄/WO₃ film was 0.78 mA cm⁻², while the bare WO₃ film exhibited a photocurrent of 0.49 mA cm⁻². The DyVO₄/WO₃ heterojunction film exhibited an almost 1.59 times higher photocurrent density than the pure WO₃ film under illumination. The enhanced PEC performance is probably due to the efficient photo-generated carriers' separation at the DyVO₄/WO₃ interface.

Fig. 6 Photocurrent densities (a) and IPCE (b) values of WO₃ and DyVO₄/WO₃ films.

To study the relationship between the PEC activity and the different wavelength of incident light, the incident-photon-to-current-efficiencies (IPCE) measurements were performed for WO₃ and DyVO₄/WO₃ films at a bias voltage of 1.2 V (vs. Ag/AgCl) in 0.2 M Na₂SO₄ solution (Fig. 6b). IPCE can be expressed as IPCE = (1240I)/(λJ_light),³¹ where I (mA cm⁻²), λ (nm) and J_light (mW cm⁻²) are the photocurrent density, the wavelength and power density of incident light, separately. Both two films presented strong photo responses at 300–460 nm. However, the DyVO₄/WO₃ heterojunction film exhibited larger IPCE values than that of the WO₃ film over the whole photo response region. The highest IPCE value of 54.11% was observed for DyVO₄/WO₃ film at 350 nm, which is almost 2 times higher than that of WO₃ film (∼27.38%). Combining the UV-vis absorption spectra with the IPCE data, it demonstrates that the low content of deposited DyVO₄ nanoparticles are not acted as a light photosensitizer.

Considering the importance of the stability of a photoelectrode for its practical application, the photocurrent–time plots of both films were obtained at +1.2 V vs. Ag/AgCl under discontinuous illumination. The results are shown in Fig. 7. After 5000 s, the pristine WO₃ film shows about 47% decrease in photocurrent from 0.49 to 0.26 mA cm⁻², while the DyVO₄/WO₃ film shows just about 17% decrease from 0.78 to 0.65 mA cm⁻². This indicates that the introduction of DyVO₄ improves the stability of WO₃ film.

Fig. 7 Photocurrent–time plots of the photoelectrodes.

To elucidate the strong correlation between formation of heterojunction and the enhanced photocurrent and IPCE values, Mott–Schottky analysis was conducted under the visible light irradiation with a frequency of 1 kHz in a 0.2 M Na₂SO₄ aqueous electrolyte. As presented in Fig. 8, both two films exhibited an n-type characteristic with the positive slop, which reflected that electrons served as the major carriers in DyVO₄/WO₃ heterojunction film. The electron density can be calculated based on the Mott–Schottky equation:³²

N_d = (2/εε₀q)[d(1/C²)/dV]⁻¹ (2)

where ε and ε₀ are the dielectric constant of the semiconductor (50 for WO₃ (ref. 33)) and the vacuum permittivity (8.85 × 10⁻¹⁴ F cm⁻²), q the electron charge (1.60 × 10⁻¹⁹ C), N_d/cm⁻³ the carrier density, and V the potential applied at the electrode, respectively. The electron density of WO₃ nano-plate arrays film was calculated to be 9.1 × 10¹⁹ cm⁻³, which agrees with previous studies,^34,35 while the N_d value of DyVO₄/WO₃ film was 4.0 × 10²⁰ cm⁻³. The formation of heterojunction thus led to an increase of 1 order of magnitude in the carrier density. The bigger electron density in the DyVO₄/WO₃ film is believed to be a major factor contributing to the pronounced photocurrent density enhancement. The higher N_d value of DyVO₄/WO₃ film can be explained by following reason: the valence band (VB) of DyVO₄ is composed of hybridized Dy 4f and O 2p orbitals, which narrows the band gap and favors the mobility of photogenerated holes in the VB.¹²

Fig. 8 Mott–Schottky plots of WO₃ and DyVO₄/WO₃ films.

The flat band potential (V_fb) at semiconductor/electrolyte interface can also be estimated by the Mott–Schottky equation:³⁶

1/C² = (2/εε₀qN_d)[(V − V_fb) − kT/q] (3)

By extrapolating the X-intercepts of the liner region in Mott–Schottky plots, V_fb of pure WO₃ film was ∼0.08 V vs. Ag/AgCl, while that of DyVO₄/WO₃ heterojunction film shifted to 0.01 V. Although the negative shift of the flat band was not prominent, it indicated fewer barrier for oxygen evolution, larger accumulation of electrons and better separation of photo-generated carriers in the heterojunction film.³⁷

To investigate the electron transport properties in the electrode, the capacitance and resistance of the electrode materials which indicated the separation efficiency between the photo-generated electrons and holes,³⁸ electrochemical impedance spectroscopy (EIS) of the as-prepared films were measured under the visible light illumination with a 0.8 V bias. Fig. 9a shows the typical Nyquist curve of the DyVO₄/WO₃ heterojunction film and pure WO₃ film at an AC frequency varying from 10 kHz to 100 mHz. One semicircle is observed for each sample, suggesting that the faradaic charge transfer is the limiting step for the oxidation process in the electrode surface.³² The apparent difference in the impedance spectra was observed that the EIS Nyquist curve of DyVO₄/WO₃ film has a smaller circular radius than that of the pure WO₃ film. The equivalent circuit consists of solution resistance (R₁), charge transfer resistance (R₂) in parallel to the constant phase element (CPE₁).³⁹ The resistance R₁, R₂ were calculated to be 25.3 Ω, 904.7 Ω for the WO₃ film, and 35.4 Ω, 488.9 Ω for the DyVO₄/WO₃ film, respectively. The R₁ values of both films are comparable while the R₂ values are very different. This suggests that DyVO₄/WO₃ film owns lower charge transfer resistance than WO₃ film, corresponding to the better PEC performance. Similarly, Bode plots in Fig. 9b reflect the efficient electron lifetime in WO₃ and DyVO₄/WO₃ films. The lifetime (τ_e) can be calculated by τ_e = 1/(2πf_max), where the f_max is the frequency at which the frequency peak appears in the Bode plot.⁴⁰ Larger τ_e values indicate that electrons have longer lifetime and faster diffusion rate in the photoelectrodes.⁴¹ As seen in Fig. 8b, the f_max value for WO₃ is 67.7 Hz, while for DyVO₄/WO₃ is 18.6 Hz. The corresponding τ_e is 2.4 ms and 8.6 ms. This suggests that DyVO₄/WO₃ possesses an almost 4-fold improved lifetime of electrons compared with pure WO₃ film.

Fig. 9 EIS Nyquist plots (a) and Bode plots (b) of WO₃ and DyVO₄/WO₃ films.

Actually, it is well known that the PEC activity of photocatalyst mainly depends on whether the electron–hole pairs can be separated effectively. The schematic for the DyVO₄/WO₃ heterojunction film illustrates the photo-generated carriers transport process, which can explain the possible mechanism for the enhanced PEC properties as shown in Fig. 10. Under the visible light irradiation, both DyVO₄ and WO₃ would be excited and produce the photo-generated electrons and holes, simultaneously. The DyVO₄ has a more negative conduction band (CB) edge potential (−0.32 eV vs. NHE) and a valence band (VB) edge potential (1.98 eV vs. NHE) compared with WO₃ (CB: 0.74 eV vs. NHE, VB: 3.44 eV vs. NHE),^18,42 indicating that the photo-generated electrons of DyVO₄ can easily transport into WO₃, then transfer to the back-contacted FTO substrate through a direct pathway provided by the plates. Similarly, the photo-generated holes of WO₃ will move to DyVO₄ and take part in the oxygen evolution reaction. Thus, the photo-generated electron–hole pairs will be separated at the interface of the heterojunction and reduce the probability of recombination, which promote the PEC water splitting.

Fig. 10 Depiction of the energy level diagram of the DyVO₄/WO₃ film and the charge separation process.

4. Conclusion

In summary, the DyVO₄/WO₃ heterojunction electrode consisting of two-dimensional WO₃ plates arrays and DyVO₄ nanoparticles were fabricated through a simple dipping-annealing method. Compared with the pure WO₃ film, the DyVO₄/WO₃ heterojunction film possessed a higher photocurrent and a better PEC performance under visible light irradiation. On the basis of EIS, Mott–Schottky and IPCE, the DyVO₄/WO₃ heterojunction film owned longer electron lifetime, larger carriers density and more negative flat band potential. Besides that, the enhanced PEC performance of the heterojunction film was mostly attributed to the high separation and collection efficiency of photo-generated electron–hole pairs at the interface. The study demonstrates that metal orthovanadates may be good heterojunction candidates to couple with WO₃ to provide promising photoanodes for water splitting.

Acknowledgements

This study was supported by the National High Technology Research and Development Program of China (2011AA050528), the National Nature Science Foundation of China (51304253).

References

1. M. Yagi, S. Maruyama, K. Sone, K. Nagai and T. Norimatsu, J. Solid State Chem., 2008, 181, 175–182.
2. Z. Jiao, J. Wang, L. Ke, X. W. Sun and H. V. Demir, ACS Appl. Mater. Interfaces, 2011, 3, 229–236.
3. F. Amano, D. Li and B. Ohtani, Chem. Commun., 2010, 46, 2769–2771.
4. M. Qamar, Q. Drmosh, M. I. Ahmed, M. Qamaruddin and Z. H. Yamani, Nanoscale Res. Lett., 2015, 10, 54–59.
5. W. Li, F. Zhan, J. Li, C. Liu, Y. Yang, Y. Li and Q. Chen, Electrochim. Acta, 2015, 160, 57–63.
6. M. Qamar, Z. H. Yamani, M. A. Gondal and K. Alhooshani, Solid State Sci., 2011, 13, 1748–1754.
7. Y. Choi, S. Kim, M. Seong, H. Yoo and J. Choi, Appl. Surf. Sci., 2014, 324, 414–418.
8. J. Zhu, W. Li, J. Li, Y. Li, H. Hu and Y. Yang, Electrochim. Acta, 2013, 112, 191–198.
9. L. J. Zhang, S. Li, B. K. Liu, D. J. Wang and T. F. Xie, ACS Catal., 2014, 4, 3724–3729.
10. S. Wei, Y. Ma, Y. Chen, L. Liu, Y. Liu and Z. Shao, J. Hazard. Mater., 2011, 194, 243–249.
11. P. Zhao, C. X. Kronawitter, X. Yang, J. Fu and B. E. Koel, Phys. Chem. Chem. Phys., 2014, 16, 1327–1332.
12. Y. He, L. Zhao, Y. Wang, H. Lin, T. Li, X. Wu and Y. Wu, Chem. Eng. J., 2011, 169, 50–57.
13. J. Ye, Z. Zou, M. Oshikiri, A. Matsushita, M. Shimoda, M. Imai and T. Shishido, Chem. Phys. Lett., 2002, 356, 221–226.
14. H. Xu, H. Wang and H. Yan, J. Hazard. Mater., 2007, 144, 82–85.
15. Y. He, J. Cai, L. Zhang, X. Wang, H. Lin, B. Teng, L. Zhao, W. Weng, H. Wan and M. Fan, Ind. Eng. Chem. Res., 2014, 53, 5905–5915.
16. W. Congsen, C. Xiaodong and L. Xue-Hong, Adv. Mater. Res., 2014, 1081, 146–150.
17. S. J. Duclos, A. Jayaraman, G. Espinosa, A. Cooper and R. Maines, J. Phys. Chem. Solids, 1989, 50, 769–775.
18. Y. He, J. Cai, T. Li, Y. Wu, Y. Yi, M. Luo and L. Zhao, Ind. Eng. Chem. Res., 2012, 51, 14729–14737.
19. J. Zhang, P. Zhang, T. Wang and J. Gong, Nano Energy, 2015, 11, 189–195.
20. F. Amano, D. Li and B. Ohtani, J. Electrochem. Soc., 2011, 158, K42–K46.
21. J. Yang, W. Z. Li, J. Li, D. B. Sun and Q. Y. Chen, J. Mater. Chem., 2012, 22, 17744–17752.
22. C. Liu, J. Li, Y. Li, W. Li, Y. Yang and Q. Chen, RSC Adv., 2015, 5, 71692–71698.
23. X. Ma, J. Gong, S. Wang, F. He, H. Guo, X. Yang and G. Xu, J. Mol. Catal. A: Chem., 2005, 237, 1–8.
24. S. Han, J. Li, X. Chen, Y. Huang, C. Liu, Y. Yang and W. Li, Int. J. Hydrogen Energy, 2012, 37, 16810–16816.
25. M. M. Mohamed, S. A. Ahmed and K. S. Khairou, Appl. Catal., B, 2014, 150, 63–73.
26. F. Guo, W. Shi, X. Lin, X. Yan, Y. Guo and G. Che, Sep. Purif. Technol., 2015, 141, 246–255.
27. A. Subrahmanyam and A. Karuppasamy, Sol. Energy Mater. Sol. Cells, 2007, 91, 266–274.
28. C. Santato, M. Odziemkowski, M. Ulmann and J. Augustynski, J. Am. Chem. Soc., 2001, 123, 10639–10649.
29. W. Z. Li, J. Li, X. Wang, J. Ma and Q. Y. Chen, Int. J. Hydrogen Energy, 2010, 35, 13137–13145.
30. W. Z. Li, J. Li, X. Wang, J. Ma and Q. Y. Chen, Appl. Surf. Sci., 2010, 256, 7077–7082.
31. Y. Ling, G. Wang, D. A. Wheeler, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 2119–2125.
32. G. Wang, Y. Ling, D. A. Wheeler, K. E. George, K. Horsley, C. Heske, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 3503–3509.
33. E. Salje and K. Viswanathan, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1975, 31, 356–359.
34. R. Sivakumar, A. M. E. Raj, B. Subramanian, M. Jayachandran, D. Trivedi and C. Sanjeeviraja, Mater. Res. Bull., 2004, 39, 1479–1489.
35. L. Su, L. Zhang, J. Fang, M. Xu and Z. Lu, Sol. Energy Mater. Sol. Cells, 1999, 58, 133–140.
36. N. T. Hahn and C. B. Mullins, Chem. Mater., 2010, 22, 6474–6482.
37. J. Zhang, J. H. Bang, C. Tang and P. V. Kamat, ACS Nano, 2009, 4, 387–395.
38. W. D. Zhang, L. C. Jiang and J. S. Ye, J. Phys. Chem. C, 2009, 113, 16247–16253.
39. S. K. Pilli, T. G. Deutsch, T. E. Furtak, L. D. Brown, J. A. Turner and A. M. Herring, Phys. Chem. Chem. Phys., 2013, 15, 3273–3278.
40. F. Wang, Y. Wang, X. Zhan, M. Safdar, J. Gong and J. He, CrystEngComm, 2014, 16, 1389–1394.
41. C. Zha, L. Shen, X. Zhang, Y. Wang, B. A. Korgel, A. Gupta and N. Bao, ACS Appl. Mater. Interfaces, 2013, 6, 122–129.
42. H. Katsumata, Y. Tachi, T. Suzuki and S. Kaneco, RSC Adv., 2014, 4, 21405–21409.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra22272a

---