Facile fabrication and photoelectrochemical properties of a CuO nanorod photocathode with a ZnO nanobranch protective layer

Abstract

In this study, the photoelectrochemical properties of CuO/ZnO photoelectrodes fabricated with nanorod and film structures were investigated and compared, and the effect of surface morphology on their photoelectrochemical performance was discussed in detail. The experimental results demonstrated that the CuO/ZnO photoelectrode with nanorod structures showed superior photoelectrochemical properties compared to that of the photoelectrode with the film structure. The electrochemical impedance analysis and UV-vis spectroscopy results confirmed that the hierarchical nanorod-like structure of the CuO/ZnO photoelectrode was advantageous for effective light absorption, and reduced charge transfer resistance at the electrode/electrolyte interface. At the same time, the ZnO layer effectively contributed to the suppression of photocorrosion in the CuO, and the photoelectrodes with a ZnO layer demonstrated 82.13% better stability in photoelectrochemical conditions.

---

1. Introduction

The global increase in energy demand and the limited availability of fossil fuels has spurred the development of alternative and renewable energy resources.¹ The effective utilization of solar energy and its conversion into useful and clean H₂ fuel has become more promising based on the continuing development of the photoelectrochemical (PEC) cell.^2,3 Over the last decade, several wide bandgap metal oxide semiconductors (such as TiO₂, ZnO, SrTiO₃, WO₃) have been extensively studied as potential photoelectrode materials for PEC cells due to their high chemical stability, low cost and non-toxicity.^4–11 However, the solar-to-hydrogen production efficiency of PEC cells with these materials is quite low, due to their large bandgap.¹² Thus, much present research is focused on developing and examining novel narrow bandgap semiconducting materials for PEC cells, which are active in visible light wavelengths. Among the visible light active materials, copper oxide (CuO) is very promising for PEC hydrogen generation because it has a narrow bandgap and a suitably positioned conduction band edge. Moreover, CuO is non-toxic, earth abundant, and can be practically realized as cost-effective photoelectrodes for PEC cells.^13–17

Nevertheless, CuO has a few drawbacks, such as low stability in aqueous media and short diffusion length of charge carriers, which limits its wide application. The instability of CuO is associated with its decomposition potential position, located within the bandgap, which results in CuO being reduced by photoinduced electrons rather than water molecules.^18–22 Recent research on CuO photoelectrodes have shown that the instability of the CuO can be effectively suppressed via a suitable protecting layer. Some works have reported protection of CuO or Cu₂O through atomic layer deposition (ALD) of multiple thin layers, by surface modification, and incorporation with a graphene based materials.^20,22–24 Although the reported works have demonstrated the enhanced stability of the electrodes and good PEC performance, the high-cost and complex deposition techniques, as well as the achieved relatively low stability of these methods, limits their wide application.

Alternatively, photoelectrodes composed of one-dimensional (1D) nanostructures such as nanowires or nanotubes are more advantageous for PEC application since they provide efficient transport pathways for the photogenerated charges, and enhanced light absorption, with minimum recombination.^{9,16,20,25,26} Thus, the use of one dimensional CuO photoelectrodes would make it possible to overcome the charge carrier diffusion length limitation of CuO, and can be expected to have superior PEC properties compared to the bulk. There are reports available in the literature demonstrating the fabrication and evaluation of CuO nanostructured electrodes with some protection layers to improve their efficacy and stability for PEC cells. Weina Shi et al. have reported the fabrication of Cu₂O nanowires by electrochemical anodization method followed by heat treatment.²³ The stability of the electrode was improved up to 61.3% by introducing a thin carbon protective layer on the Cu₂O nanowires. Another relatively simple method of fabricating CuO/ZnO core/shell nanowire photoelectrodes was reported by X. Zhao.²⁷ The photo conversion efficiency and stability of the electrodes were enhanced due to the formation of p–n junctions along the p-CuO core and n-ZnO protective shell, respectively.

Although there have been published many research works on the fabrication and evaluation of PEC performance using CuO nanostructured electrodes, few of them have provided a comparison of the PEC properties of CuO nanostructures versus film/bulk based electrodes. In this study, we examined the comparative photoelectrochemical performances of hierarchical CuO/ZnO nanorod structures and flat CuO/ZnO film based photoelectrodes. The effect of the morphological structures of the fabricated photocathode and the role of a ZnO passivation layer are discussed in detail.

2. Experiment details

2.1 Preparation of photoelectrodes

Metallic Cu NRs were grown on a Cu substrate by electrochemical deposition through a porous membrane as a template, in an electrolyte composed of 125 g L⁻¹ of CuSO₄·xH₂O and 100 g L⁻¹ of H₂SO₄ (Fig. 1(a)). Then, the Cu NR substrate was thermally oxidized at 300 °C for 2 h to form CuO nanorods on the Cu substrate (Fig. 1(b)). Another set of photoelectrodes comprised of CuO thin films were fabricated by the thermal oxidation of pre-cleaned Cu foil (thickness of 1 mm). Then, a thin ZnO seed layer was deposited on both the CuO NR and CuO film substrates by immersing them into 0.1 mM zinc acetate (Zn(O₂CCH₃))₂ solution and drying at 100 °C on a hot plate (Fig. 1(c)). This process was repeated 8 times in order to obtain the proper density of ZnO seed layer. Afterwards, both CuO NR and CuO film substrates were transferred into a hydrothermal reactor containing a hydrothermal solution composed of an equimolar ratio of 0.25 mM Zn(NO₃)₂ and (CH₂)₆N₄ and kept at 90 °C for 5 h (Fig. 1(d)). After the hydrothermal reaction, the samples were thoroughly washed with DI water and dried under N₂ stream.

Fig. 1 Schematic illustration of the CuO/ZnO NR and CuO/ZnO film photoelectrode preparation. (a) Metallic Cu NR growth, (b) thermal oxidation of Cu NRs and plate to transform into CuO, (c) ZnO seed layer coating, (d) hydrothermal growth of ZnO NR layer.

As a result, two types of photoelectrodes were obtained, namely, CuO NR with ZnO nanobranches (CuO/ZnO NR), and CuO film with ZnO NRs grown on top (CuO/ZnO film).

2.2 Instrumentation

The surface morphology and micro-structural characterizations of the nanostructured samples were performed by using a field emission scanning electron microscope (FE-SEM, JSM-6500F, JEOL, Japan) and X-ray diffraction (XRD, RIGAKU/SWXD X-MAX/2000-PC, Japan). The optical properties of the electrodes were evaluated by using UV-visible reflectance spectroscopy (Shimadzu 3600 UV-visible NIR spectrophotometer, Japan). The photocurrent measurements were performed using an Autolab PGSTAT302N potentiostat in a three electrode configuration consisting of a Pt counter electrode and a saturated Ag/AgCl reference electrode in 0.1 M KOH electrolyte. Working electrode with 1 cm² area was illuminated using a 1 kW xenon lamp from which infrared wavelengths were filtered by water. Illuminating light intensity measured by a thermopile detector was 100 mW cm⁻². Electrochemical impedance spectroscopy was measured in wide frequency range of 0.01–10⁵ Hz in three electrode cell with Pt counter electrode and a saturated Ag/AgCl reference electrode. Fabricated photoelectrodes were used as a working electrode in 0.1 M KOH electrolyte.

3. Results and discussion

In this study, we fabricated CuO/ZnO hierarchical structures with a ZnO NR protective layer, as schematically illustrated in Fig. 1. The field-emission scanning electron microscope images of as prepared CuO/ZnO photoelectrodes are shown in Fig. 2. The surface and cross-sectional FE-SEM micrographs of the CuO/ZnO NR electrode (shown in Fig. 2(a) and (b)) clearly reveal the formation of CuO NRs covered with regular ZnO NR branches, forming a CuO/ZnO hierarchical structure. The inset of Fig. 2(a) and (b) shows an enlarged view of individual CuO/ZnO NR structures, which indicates that ZnO NR branches have been continuously grown along the entire length of the CuO NRs due to the homogenous ZnO seed layer coating. The surface and cross-sectional FE-SEM images of the CuO/ZnO film electrodes are shown in Fig. 2(c) and (d), respectively. The average length of the regular ZnO NRs on the CuO film is 420–750 nm.

Fig. 2 FE-SEM images of the CuO/ZnO photoelectrodes. (a) and (b) Surface and cross-sectional view of CuO/ZnO-NR photoelectrodes, respectively. (c) and (d) Surface and cross-sectional view of CuO/ZnO-film photoelectrodes, respectively.

Fig. 3(a) shows bright-field transmission electron microscopy (TEM) image of CuO/ZnO nanorods bundle. It is clearly seen that CuO trunk (indicated with yellow dash lines) is covered with ZnO NR branches (indicated with white arrowheads). Enlarged view of the ZnO NR branch/CuO trunk interface (white boxed region) is shown in the Fig. 3(b). Inset of the Fig. 3(b) shows high-magnification TEM image of the ZnO branches, where lattice fringes of the ZnO (002) plane is shown. Scanning transmission electron microscope (STEM)-energy dispersive X-ray spectroscopy (EDX) elemental mapping (Fig. 3(c) and (d)) shows that trunk part of the nanostructures is composed of Cu and O elements whereas branches are composed of ZnO and O.

Fig. 3 TEM and TEM-EDS analysis of the CuO/ZnO-NRs. (a) Bright-field TEM images of the CuO/ZnO NR bundle. (b) High magnification TEM image of the CuO/ZnO interface taken from white boxed region in (a). Inset shows high-resolution image of ZnO NR branch. (c) STEM and (d) EDS elemental mapping of the CuO/ZnO-NR bundle.

The formation of CuO and ZnO on the CuO/ZnO NR electrode was confirmed by XRD analysis and is presented in Fig. 4(a). The strong peaks corresponding to the Cu substrate are well matched with a pristine Cu substrate, which was used as a reference, and can be assigned to the Cu (111), (200) and (220) planes (PDF# 01-070-3038). The XRD pattern of the CuO/ZnO hierarchical structures demonstrate the presence of well-defined CuO peaks corresponding to the (−111), (111) and (103) planes at 35.5°, 38.6° and 60.94°, respectively.¹⁹ The reflection peaks corresponding to (100), (002) and (102) at 31.66°, 34.32°, 47.46° were assigned to the hexagonal ZnO (PDF# 01-075-1533), which confirms the successful loading of ZnO on the CuO NR electrode. In addition, XRD patterns of a pure CuO and ZnO NRs grown on Cu substrate were provided as a control samples for clear comparison. The appearance of a few Cu₂O peaks can be ascribed to the existence of a thin Cu₂O layer beneath the CuO which could have formed due to the relatively low oxidation temperature (300 °C).²⁸

Fig. 4 Microstructural and optical properties of CuO/ZnO photoelectrodes. (a) XRD pattern of the CuO/ZnO photoelectrode. (b) UV-vis spectrum of the CuO/ZnO-NR and CuO/ZnO – film electrodes.

Fig. 4(b) shows the UV-vis absorbance spectra of the CuO/ZnO-film and CuO/ZnO-NR electrodes. Both electrodes demonstrate strong absorption below 380 nm and moderate absorption up to 800 nm in the wide range of visible light, corresponding to the absorption band edges of ZnO and CuO, respectively. Apparently, the CuO/ZnO-NR electrode has superior light absorption properties over both the short and long wavelength regions of the spectrum, which can be attributed to the unique hierarchical structure of the electrode, where a more pronounced light scattering and trapping effect can be observed compared to the CuO film based electrode.²⁹ The effect of enhanced absorption properties of the nanostructured electrode was confirmed by comparing the absorbance spectra of ZnO thin film and NRs as given in the Fig. S1.† Sputtered ZnO thin film and hydrothermally grown ZnO NRs control samples were prepared on an amorphous SiO₂/Si substrate in order to exclude the substrate effect on optical properties. As can be seen, both ZnO thin film and NR samples demonstrate identical absorption peaks centered at 380 nm. However, intensity of the absorption peak for the ZnO NR is relatively higher than that of thin film. The Kubelka–Munk function was used to transform the UV-vis absorbance spectra to determine the optical bandgap energies of the electrodes:

F(R) = α = (1 − R)²/2R (1)

where “R” is the reflectance and α is the optical absorption coefficient.³⁰ The relation of the incident photon energy and bandgap “E_g” to the transformed Kubelka–Munk function is given as:

[F(R)hν]ⁿ = A(hn − E_g) (2)

where, A is a transition probability constant, and n is an optical abortion index.

The optical bandgap energies of the electrodes can be determined by extrapolating the linear part of the curve to the energy axis from the [F(R)hν]ⁿ vs. (hν) plot, as presented in the inset of Fig. 4(b). Both electrodes demonstrate two overlapping shoulders at 1.48 eV and 3.2 eV corresponding to the bandgap energies of CuO and ZnO, respectively, and these results are in good agreement with other reported data.³¹ The obtained bandgap values of the electrodes suggest that no significant bandgap modification or doping occurred with the CuO–ZnO incorporation.

The photoelectrochemical properties of the bare CuO film and CuO NR photoelectrodes were first evaluated by measuring their photocurrent–potential characteristics under light illumination. Fig. 5(a) shows linear sweep voltammograms of the CuO NR and CuO film electrodes measured in the three electrode configuration in 0.1 M KOH electrolyte. The light illumination was continuously chopped during the linear sweep voltammetry measurements in order to reveal dark current and photocurrent in a single sweep. Both the CuO film and CuO NR photoelectrodes produced a negative photocurrent, of −0.85 mA cm⁻² and −1.13 mA cm⁻² at −0.5 V vs. Ag/AgCl, respectively. The cathodic photocurrent generation indicates the p-type conductivity of the CuO based electrodes. Evaluation of the efficiencies of the photoelectrodes was calculated in terms of applied bias photon-to-current efficiency (ABPE) using the following equation:

where, J_ph (mA cm⁻²) is a measured photocurrent density at a given applied bias V_b (V vs. Ag/AgCl) and 1.23 (V) is the thermodynamic minimum potential needed to split water. P_total is the incident light intensity (mW cm⁻²). Based on our results, the ABPE for CuO NR, and CuO film were 0.83%, and 0.6%, respectively. The obtained ABPE was higher for the CuO NR electrode compared to that of the CuO film electrode. This can be attributed to the nanostructured geometry of the CuO NR, which can provide a reduced recombination rate and superior charge transfer ability, and is thereby more beneficial for effective charge transport compared to the film based electrode.³²

Fig. 5 Photoelectrochemical properties of bare CuO NR and film electrodes. (a) Photocurrent–voltage characteristics for CuO NR and CuO film photoelectrodes. (b) and (c) Transient photocurrent response for CuO film and CuO NR photoelectrodes, respectively. (d) Transient time constant characteristics for CuO NR and CuO film.

Further, the photocurrent transients of the CuO NR and film-based electrodes were measured in order to quantitatively investigate the morphology and geometry dependent charge transport/recombination properties of the electrodes.³³ Fig. 5(b) and (c) shows the photocurrent transients for the CuO film and NR electrodes, respectively. The transient curves of the CuO film electrode show an increase in the cathodic photocurrent at the “light on” position followed by an exponential decrease with time. The maximum cathodic photocurrent spike (initial photocurrent – I_i) is induced by electron–hole pair separation, and the decay in the photocurrent indicates that recombination processes are taking place. Another spike is observed at the “light off” position due to the back reaction of conduction band electrons with holes trapped at the electrode surface.³⁴

On the other hand, no spikes at the light on and off conditions were observed for the CuO NR electrode (Fig. 5(c)), and the photocurrent decay is not as dramatic as it was for the CuO film electrode. The recombination process occurring in the photoelectrodes can be expressed by the following equation:^35,36

where t is time, τ is transient time constant, I_t is a current at t time, and subscripts i and f indicate the initial and final states. The transient time constant expresses the recombination rate and can be estimated by the inverse slope on the ln(D) vs. time plot, as shown in Fig. 5(d). The transient time constant for the CuO film and NR electrodes was found to be 1.44 s and 3.67 s, respectively. Therefore, it can be concluded that the recombination process in the CuO NR electrode is almost two times slower than that of the film electrode. Hence, an electrode with nanorod morphology can result in enhanced carrier concentration and longer electron lifetime, which explains the higher photocurrent generation as evidenced in the Fig. 5(a).

The photoelectrochemical properties of the CuO electrodes covered with the ZnO NRs protective layers are further illustrated in Fig. 6(a). The generated photocurrent and calculated ABPE efficiency values for the CuO/ZnO-film and CuO/ZnO-NR photoelectrodes were −0.5 mA cm⁻², −0.81 mA cm⁻² and 0.37%, 0.61% at −0.5 V vs. Ag/AgCl, respectively. In this case, the superior PEC performance of the CuO/ZnO-NR electrode is ascribed to not only the nanorod shape, but also to the unique hierarchical structure. The nano trunk-branch feature of the CuO/Zn-NR electrode provides higher surface area, more reactive sites and better charge carrier diffusion at the electrode/electrolyte interface compared to the planar CuO/ZnO-film.

Fig. 6 Photoelectrochemical properties of CuO/ZnO-NR and CuO/ZnO film photoelectrodes. (a) Photocurrent–voltage characteristics for CuO/ZnO-NR and CuO/ZnO film photoelectrodes. (b) Nyquist and (c) Bode phase plots obtained for CuO/ZnO-NR and CuO/ZnO film electrodes. (d) Schematic illustration of light absorption and charge transfer properties in the CuO/ZnO-NR and CuO/ZnO film electrodes.

The interfacial charge transfer properties of the electrodes were studied by electrochemical impedance spectroscopy (EIS) analysis over the frequency range of 0.01–10⁵ Hz.^37,38 Fig. 6(b) shows the Nyquist plots of the CuO/ZnO-film and CuO/ZnO-NR electrodes. Both electrodes demonstrate semicircle characteristics in the overall frequency region of the plot, which represents the charge transfer process occurring at the electrode/electrolyte interface, and the diameter of the semicircle indicates the charge transfer resistance R_ct. The CuO/ZnO-NR electrode shows a lower R_ct compared to the CuO/ZnO-film electrode, which confirms the effective charge transfer property of the electrode.

According to the Bode phase plot shown in Fig. 6(c), both electrodes exhibit a presence of time constant in the middle frequency region, which can be assigned to the charge transfer process occurring at the electrode/electrolyte interface. The observed time constant is inversely associated with the electron lifetime, as follows;^39–41

τ = 1/2πω_max (6)

where τ is electron lifetime, ω_max is maximum frequency. Therefore, a decrease in frequency would imply a longer electron lifetime. Indeed, the frequency peak for the CuO/ZnO-NR electrode is shifted to the lower frequency region compared to the CuO film based electrodes. The values of electron lifetime were calculated to be 30.74 ms and 82.48 ms for the CuO/ZnO-film and CuO/ZnO-NR electrodes, respectively. The longer electron lifetime is an indication of charge recombination suppression and effective interfacial charge transfer at the CuO/ZnO-NR/electrolyte interface.

Consequently, based on the collected PEC and EIS data, it can be concluded that the CuO/ZnO-NR electrode has superior performance compared to the CuO/ZnO-film electrode, which can be ascribed to the collective effect of several factors that emerge from the nanotrunk-branched structure of the electrode, as schematically shown in Fig. 6(d). These are (i) enhanced light absorption owing to the multiple light scattering and trapping, (ii) reduced charge recombination in the CuO NR due to the decreased charge diffusion length, (iii) reduced charge transfer resistance at the electrode/electrolyte interface (longer electron lifetime) which is due to the favorable electron transfer path along both the lateral and longitudinal axes of the nanorods, as well as their large surface area and abundant reaction sites.

The effect of the ZnO NR layer on the stability of the CuO electrode was studied by measuring the photocurrent at a fixed −0.5 V vs. the Ag/AgCl potential for an extended period, as demonstrated for the CuO NR electrodes in Fig. 7. Fig. 7(a) shows the chronoamperometry (photocurrent–time curves) results for the bare CuO NR and CuO/ZnO-NR electrodes. As can be seen, the photocurrent for the bare CuO electrode immediately starts to decline and continuously decreases during measurement, which is an indication of photoelectrode decomposition.

Fig. 7 Stability measurement results of bare CuO and CuO/ZnO-NR photoelectrodes. (a) Short-period chronoamperometry results of bare CuO and CuO/ZnO-NR photoelectrodes with light on/off cycles. (b) Long-period chronoamperometry results of bare CuO and CuO/ZnO-NR photoelectrodes plotted with photocurrent density values corresponding to the light on state from (a).

On the other hand, the CuO NR electrode with the ZnO NR layer retains its photocurrent value. The photocurrent values at the “light on” states were collected from the chronoamperometry measurements, recorded for 20 min, and plotted against corresponding time; they are presented in Fig. 7(b). Obviously, the photocurrent decay is much slower for the CuO/ZnO-NR electrode compared to the bare CuO NR.

The photo stability of the electrodes was quantified as the ratio of the photocurrent at the end of the measurement to that at the beginning, based on Fig. 7(b). The calculated stability of the bare CuO NR was only 19.8%, whereas stability of 82.13% was obtained for the CuO/ZnO-NR electrode. Surface morphology and structural composition of the CuO/ZnO-NR electrode after stability test were examined by FE-SEM and XRD. From low and high magnification FE-SEM images, shown in Fig. S2(a) and (b),† we can conclude that hierarchical morphology of the CuO/ZnO-NR electrode is still preserved after stability test. However, as shown in Fig. S2(c) and (d),† there are a few mechanical damages and contaminations that could be introduced at sample preparation stage (sample mounting for photocurrent measurement setup and cleaning after measurement). Fig. S3† shows XRD pattern of the CuO/ZnO-NR after stability test, which demonstrate similar microstructural composition as before measurement (see Fig. 4(a)). The enhanced photo stability was assigned to the barrier effect of the ZnO NRs, which protected the surface of the CuO electrode from direct contact with the electrolyte, while at the same time not suppressing its PEC properties.

4. Conclusion

The key findings of this work suggest how the structural morphology of CuO photoelectrodes (nanorod and planar) affect their photoelectrochemical performance, in terms of photo-conversion efficiency, and the dynamics of photoinduced charge carriers in the electrode and at electrode/electrolyte interfaces. The experimental findings revealed that the charge recombination rate in the photoelectrode with the CuO nanorod backbone was two times lower than for the planar CuO electrode, due to the favorable electron diffusion path and effective charge collection in the electrode. Furthermore, the densely grown ZnO NRs acted as an effective surface passivation layer, protecting CuO from photocorrosion and contributing to the enhancement of the photoelectrode stability, which was increased by 82.13%.

Acknowledgements

The authors would like to acknowledge the support from the National Research Foundation of Korea through the Ministry of Science, ICT & Future Planning (Grant Number: 2014M1A7A1A03045383, 2010-0020077 and 2014R1A2A1A01006421).

References

1. S. R. C. A. Grimes and O. K. Varghese, Light, Water, Hydrogen, Springer, US, 2008.
2. S. Dunn, Int. J. Hydrogen Energy, 2002, 27, 235–264.
3. K. Villa, X. Domènech, S. Malato, M. I. Maldonado and J. Peral, Int. J. Hydrogen Energy, 2013, 38, 12718–12724.
4. S. K. Biswas and J.-O. Baeg, Int. J. Hydrogen Energy, 2013, 38, 3177–3188.
5. C. H. Hsu and D. H. Chen, Int. J. Hydrogen Energy, 2011, 36, 15538–15547.
6. A. Ikram, S. Sahai, S. Rai, S. Dass, R. Shrivastav and V. R. Satsangi, J. Power Sources, 2014, 267, 664–672.
7. R. Liu, X. Wang, H. Zhou, T. Wang, J. Zhang, Y. Xu, C. He, B. Wang, J. Zhang and H. Wang, Int. J. Hydrogen Energy, 2013, 38, 16755–16760.
8. S. Rai, A. Ikram, S. Sahai, S. Dass, R. Shrivastav and V. R. Satsangi, Renewable Energy, 2015, 83, 447–454.
9. U. Shaislamov and B. L. Yang, Int. J. Hydrogen Energy, 2013, 38, 14180–14188.
10. J. Zhang, P. Zhang, T. Wang and J. Gong, Nano Energy, 2015, 11, 189–195.
11. R. Mohan, K. Krishnamoorthy and S. J. Kim, Chem. Phys. Lett., 2012, 539–540, 83–88.
12. R. Abe, J. Photochem. Photobiol., C, 2010, 11, 179–209.
13. Y. K. Hsu, C. H. Yu, Y. C. Chen and Y. G. Lin, Electrochim. Acta, 2013, 105, 62–68.
14. U. Shaislamov, K. Krishnamoorthy, S. J. Kim, A. Abidov, B. Allabergenov, S. Kim, S. Choi, R. Suresh, W. M. Ahmed and H.-J. Lee, Int. J. Hydrogen Energy, 2016, 41, 2253–2262.
15. P. Wang, H. Wu, Y. Tang, R. Amal and Y. H. Ng, J. Phys. Chem. C, 2015, 119, 26275–26282.
16. Q. Zhang, K. Zhang, D. Xu, G. Yang, H. Huang, F. Nie, C. Liu and S. Yang, Prog. Mater. Sci., 2014, 60, 208–237.
17. K. Krishnamoorthy and S. J. Kim, Mater. Res. Bull., 2013, 48, 3136–3139.
18. Z. Zhang and P. Wang, J. Mater. Chem., 2012, 22, 2456.
19. X. Guo, P. Diao, D. Xu, S. Huang, Y. Yang, T. Jin, Q. Wu, M. Xiang and M. Zhang, Int. J. Hydrogen Energy, 2014, 39, 7686–7696.
20. A. A. Dubale, W.-N. Su, A. G. Tamirat, C.-J. Pan, B. A. Aragaw, H.-M. Chen, C.-H. Chen and B.-J. Hwang, J. Mater. Chem. A, 2014, 2, 18383–18397.
21. F. Amano, T. Ebina and B. Ohtani, Thin Solid Films, 2014, 550, 340–346.
22. A. Paracchino, V. Laporte, K. Sivula, M. Grätzel and E. Thimsen, Nat. Mater., 2011, 10, 456–461.
23. W. Shi, X. Zhang, S. Li, B. Zhang, M. Wang and Y. Shen, Appl. Surf. Sci., 2015, 358, 404–411.
24. Q. Huang, F. Kang, H. Liu, Q. Li and X. Xiao, J. Mater. Chem. A, 2013, 1, 2418.
25. U. Shaislamov and B. L. Yang, J. Mater. Res., 2012, 28, 418–423.
26. U. Shaislamov, H. Kim and B. L. Yang, J. Mater. Res., 2013, 28, 497–501.
27. X. Zhao, P. Wang and B. Li, Chem. Commun., 2010, 46, 6768–6770.
28. L. De Los Santos Valladares, D. H. Salinas, A. B. Dominguez, D. A. Najarro, S. I. Khondaker, T. Mitrelias, C. H. W. Barnes, J. A. Aguiar and Y. Majima, Thin Solid Films, 2012, 520, 6368–6374.
29. C. Cheng, W. Ren and H. Zhang, Nano Energy, 2014, 5, 132–138.
30. S. Emin, F. F. Abdi, M. Fanetti, W. Peng, W. Smith, K. Sivula, B. Dam and M. Valant, J. Electroanal. Chem., 2014, 717–718, 243–249.
31. C. Y. Chiang, Y. Shin, K. Aroh and S. Ehrman, Int. J. Hydrogen Energy, 2012, 37, 8232–8239.
32. K. Zhu, N. R. Neale, A. Miedaner and A. J. Frank, Nano Lett., 2007, 7, 69–74.
33. K. Zakrzewska, M. Radecka, J. Przewoźnik, K. Kowalski and P. Czuba, Thin Solid Films, 2005, 490, 101–107.
34. M. Radecka, A. Wnuk, A. Trenczek-Zajac, K. Schneider and K. Zakrzewska, Int. J. Hydrogen Energy, 2015, 40, 841–851.
35. A. Kusior, A. Wnuk, A. Trenczek-Zajac, K. Zakrzewska and M. Radecka, Int. J. Hydrogen Energy, 2015, 40, 4936–4944.
36. X. Shi, K. Zhang and J. H. Park, Int. J. Hydrogen Energy, 2013, 38, 12725–12732.
37. T. Lopes, L. Andrade, H. A. Ribeiro and A. Mendes, Int. J. Hydrogen Energy, 2010, 35, 11601–11608.
38. C. Zhang, M. Shao, F. Ning, S. Xu, Z. Li, M. Wei, D. G. Evans and X. Duan, Nano Energy, 2015, 12, 231–239.
39. S. Herná, D. Hidalgo, A. Sacco, A. Chiodoni, A. Lamberti, V. Cauda, E. Tresso and G. Saracco, Phys. Chem. Chem. Phys., 2015, 17, 7775–7786.
40. N. Yusoff, S. V. Kumar, A. Pandikumar, N. M. Huang, A. R. Marlinda and M. N. An'amt, Ceram. Int., 2014, 41, 5117–5128.
41. S. P. Lim, A. Pandikumar, H. N. Lim, R. Ramaraj and N. M. Huang, Sci. Rep., 2015, 5, 11922.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra18832j

---