Electrochemical energy-storage performances of nickel oxide films prepared by a sparking method

Abstract

In this work, nickel oxide (NiO) films have been prepared by a sparking method on flexible chromium/gold coated polyethylene terephthalate substrates and investigated for electrochemical energy-storage applications. Structural characterizations by scanning/transmission electron microscopies, X-ray diffraction, X-ray photoelectron spectroscopy and a UV-vis spectrophotometer reveal that the film comprises polycrstalline NiO nanoparticles with diameters in the range of 3.0–6.0 nm loosely agglomerated into a porous foam-like network. The nanoporous sparked NiO films, exhibit remarkable energy-storage behavior with a high average specific charge capacity of 402.75 C g⁻¹ at a discharge current of 1 A g⁻¹ and a good capacity retention of 88% after 1000 cycles at a high discharge current of 40 A g⁻¹. Thus, the sparking method is a promising alternative route for the preparation of high-performance electrochemical energy-storage devices.

---

Introduction

Recently, transition metal oxide (TMO) nanoparticles have been widely studied for energy-storage applications including fuel cells, electrochromic capacitors and batteries because of their advantageous properties including high specific surface area, high electrocatalytic activity and good structural stability.^1–4 In electrochemical energy-storage applications, TMOs such as RuO₂, V₂O₅, Fe₃O₄, MnO₂ and NiO have been shown to exhibit promising performances with high specific capacitances and high energy densities owing to their excellent charge storage characteristics.^5–9 Among various TMO, NiO is a particularly attractive choice^10–16 because of its exceptionally high theoretical specific capacitance (3750 F g⁻¹), excellent redox activity, low toxicity, low cost, and environmental friendly nature.^11–13 The performances of NiO-based electrochemical energy-storage devices are related to its structural, chemical and electrical properties, which depend considerably on the preparation method. NiO has been synthesized by a variety of chemical and physical approaches.^{14,15,17–22} Chemical methods including chemical bath deposition, electrodeposition, chemical precipitation, spray pyrolysis, hydrothermal process and sol gel process usually generate harmful chemical wastes or excess chemical impurities, which come from the chemical precursors or other chemical agents used in the processes. Physical methods such as thermal decomposition/oxidation, thermal evaporation and pulsed laser deposition can avoid these shortcomings. However, most of them require either high vacuum system or high-temperature operation, which are expensive and impractical for low-cost energy-storage applications. In addition, the high temperature requirement complicates its use and application with other low-temperature components such as plastic substrates.

Therefore, an alternative physical synthesis method should be developed to achieve NiO nanostructures with high pseudo-capacitive characteristics at low cost and low temperature. Sparking technique is a fairly new fabrication process that can create porous nanostructured films of metal and metal oxide with uniform composition and well-controlled thickness in one step at low cost.^23–25 It utilizes arcing between two metal tips to form nanoparticulate films under atmospheric condition so that an expensive vacuum system is not required and it can be easily scaled up for commercial use. Recently, various metal oxide nanoparticle films including TiO₂, ZnO and In₂O₃ have been prepared by sparking method and applied for photocatalysts, double-layered photoelectrodes in dye-sensitized solar cells and gas sensors.^23–25 However, there is no report of NiO nanostructured films prepared by the sparking method for energy-storage applications. In this work, NiO films are deposited by the sparking method and its electrochemical energy-storage performances are investigated by cyclic voltammetry (CV) and galvanostatic charge–discharge (GCD) measurements.

Materials and methods

The schematic illustration of the sparking apparatus for the fabrication of NiO nanoparticles (NiO-NPs) and NiO films is shown in Fig. 1. Nickel is sparked off between nickel anode and cathode tips under a high applied voltage onto a substrate placed directly below the wires. The sparking machine was equipped with an array of nickel electrodes that would be scanned in one dimension while the substrate would be sliding automatically in another dimension with controllable rates by step motors. The design of sparking system allowed rapid deposition of nanoparticle films over a large area and improved the uniformity and quality of these films. Nickel wires (99.98%, Advent Research Material Ltd, Oxford, UK) with 0.25 mm in diameter were used as the anode and cathode electrodes. Before use, nickel wires were washed with acetone, absolute ethanol, and de-ionized water in an ultrasound bath for 15 min and then dried by nitrogen gas at room temperature. The wires were cut to form sparking tips and then aligned with a gap between anode and cathode of ∼1 mm. The NiO films as working electrodes were deposited by direct sparking from an array of nickel wires at 3 kV and 10 mA onto chromium/gold (Cr/Au) current collector layers on flexible polyethylene terephthalate (PET) substrates. The high applied voltage induced high-temperature arcing plasma in the air gap via field ionization process. Electrons and ions in the plasma bombarded at the two tip surfaces, producing vaporized nickel nanoparticles that were oxidized in air at the high plasma temperature into NiO-NPs and deposited onto underlying substrates. The 50 nm thick Cr and 200 nm thick Au layers were prepared by dc sputtering at an argon pressure of 3 × 10⁻³ mbar and a dc current of 0.2 A. The Cr layer was used as an adhesive layer between Au and PET.²⁶ The substrate was repeatedly scanned under the arcing electrode to obtain a uniformly thick NiO film. It should be noted that the sparking process parameters were chosen based on the previous studies reported by our group,²⁷ which showed that higher sparking voltage could produce smaller average particle size. In particular, the maximum sparking voltage (3.0 kV) and sample sliding speed (1 cm min⁻¹) of the sparking machine were selected to obtain uniform sparked films with very fine nanometer-sized particles and high porosity. The mass of sparked NiO films was measured by high-precision microbalance with 6 digit resolution (M5P, Sartorius). The film mass was calculated by subtracting the substrate mass measured before NiO film deposition from the final mass measured afterwards. The average mass density of the film was determined from 8 samples (1 × 1 cm²) to be 0.0375 mg cm⁻². The NiO electrodes would be characterized for electrochemical energy-storage performances without post-deposition annealing.

Fig. 1 Schematic illustration of the sparking system for fabrication of NiO films.

Material characterizations

The morphologies and nanostructures of NiO-NPs and NiO films were characterized using a field emission scanning electron microscope (FESEM, JEOL JSM-6335F) and transmission electron microscopy (TEM, JEOL JEM-2010). TEM samples were prepared by drop coating of NiO-NPs dispersion in acetone on a holey carbon/copper grid and TEM was operated at 200 kV and 108 mA. The selected area electron diffraction pattern (SAEDP) was also recorded to verify the crystal structure of NiO-NPs. The surface topography of NiO films was examined using an atomic force microscope (AFM, Nano Scope IIIa, Digital Instruments) equipped with a standard Si tip scanned over an area of 1 × 1 μm² in air at room temperature. The average surface roughness of the film was then calculated from AFM images of 4 different areas on a sample. In addition, the specific surface area and pore size distribution of the film were determined by Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) nitrogen adsorption measurements (Autosorb 1 MP, Quantachrome). The samples were outgassed at 60 °C for 4 hour before the measurements. The band gap energy of sparked NiO film was determined by optical absorption spectroscopy using UV-vis spectrophotometer (Varian Cary 50, USA). The UV-vis sample was prepared by sparking of NiO-NPs on a glass substrate and UV-vis spectrum was measured in the wavelength range of ∼300–800 nm. Besides, the oxidation state and chemical composition of the sparked NiO film were evaluated by X-ray photoelectron spectroscopy (XPS) using an AXIS Ultra DLD-X-ray photoelectron spectrometer and a monochromatic Al Kα X-ray excitation source.

Electrochemical measurement

The electrochemical energy-storage performances of the NiO films were evaluated at room temperature by CV and GCD measurements using an μ-Autolab Type III electrochemical work station (ECO-Chemie, Metrohm, Switzerland). The electrochemical measurements were conducted in a three-electrode configuration comprising a silver/silver chloride (Ag/AgCl) reference electrode, a platinum (Pt) wire counter electrode, the NiO working electrode and 1.0 M KOH electrolyte.

Results and discussion

Fig. 2 shows typical top-view and cross-sectional-view SEM images of a sparked NiO film. From the top-view image (Fig. 2(a) and (b)), it can be seen that the film comprises very fine nanometer-sized particles loosely agglomerated into porous foam-like network. In addition, it contains a large number of pores with various sizes ranging from a few to hundreds nanometers. The cross-sectional view image (Fig. 2(c)) demonstrates that the NiO layer prepared by sparking method is a homogenous and regular film with high surface roughness. The average layer thickness is estimated to be ∼2.6 ± 0.27 μm. The result confirms that a uniform film of loosely packed nanoparticles with microns in thickness can be effectively produced by the specially designed sparking system with electrode array and two-dimensional scanning.

Fig. 2 The SEM images of the sparked NiO film: top-view at (a) low and (b) high magnifications and (c) cross-sectional-view.

The surface topography of the sparked NiO film was examined and compared with that of Cr–Au on PET substrate as illustrated in Fig. 3. It is seen that the NiO sparked film displays a bumpy surface comprising agglomerated secondary nanoparticles with diameters of less than 30 nm (Fig. 3(c) and (d)) while the substrate is much smoother with very shallow nanoprotrusions (Fig. 3(a) and (b)). The result is in agreement with the SEM observations in the top area of the porous foam network (Fig. 2(b)). The root-mean-square (RMS) surface roughness of the substrate and NiO sparked film are determined from the AFM data to be 4.4 ± 0.4 and 40.2 ± 3.6 nm, respectively. Thus, the RMS roughness of NiO sparked film is almost 10-fold as high as that of substrate. The high RMS roughness value of the film implies a large effective surface area and high rate of the electrochemical reaction.²⁸

Fig. 3 (a) 2D, (b) 3D AFM topographic images of Cr–Au on PET substrate, (c) 2D and (d) 3D AFM topographic images of NiO sparked surfaces.

Fig. 4(a) displays a typical TEM image of NiO-NPs. It shows fine nanoparticles with uniform diameters randomly scattered on the surface. The corresponding HR-TEM image in Fig. 4(b) reveals that the nanoparticles are mostly spheroidal with varying diameters in the range of 3.0–6.0 nm. The results indicate that the sparking technique produces homogeneous nanoparticles with narrow particle size distribution. In addition, lattice fringes can be observed on some nanoparticles with the inter planar distance of ∼0.21 nm, which is corresponding to the d-spacing on (200) of cubic NiO.^29,30 The corresponding SAED pattern of NiO-NPs is shown in Fig. 4(c). It demonstrates ring diffraction patterns, identifying that the NiO particles have a polycrystalline structure. From indexing the diffraction pattern, the four main diffraction rings are found to closely match with the (111), (200), (220), and (311) planes of the cubic NiO phase (JCPDS 78-0429).¹³ Thus, as-produced sparked powders are confirmed to be NiO nanoparticles with polycrystalline structure. Fig. 4(d) illustrates a typical EDS spectrum of the NiO-NPs on a carbon/copper TEM grid. It confirms all expected elements, including nickel (Ni), oxygen (O), copper (Cu) and carbon (C) of the nanoparticles and TEM grid. By excluding the elements from the grid, the atomic percentages of O and Ni are determined to be 47.6% and 52.4%, respectively. The composition is close to the ideal stoichiometric NiO.

Fig. 4 (a) TEM image, (b) HR-TEM image, (c) SAED pattern and (d) EDX spectrum of sparked NiO-NPs.

Fig. 5 displays the N₂ adsorption–desorption isotherm of sparked NiO film. It can be seen that the film shows type VI pattern according to IUPAC classification,^31,32 corresponding to a stepwise multilayer adsorption process, which occurs due to a multimodal pore distribution. The estimated BET surface area from the adsorption data is ∼16 m² g⁻¹, which is considered a high value for a thin film structure. The calculated BJH pore size distribution of the NiO film (inset of Fig. 5) demonstrates two maximum pore sizes at ∼2.8 and ∼10.4 nm, indicating the mesoporous structure. The high film porosity will provide an efficient transport pathway for ions to the interior of the sample, which is beneficial for electron transfer and adsorption, leading to a high charge capacity.

Fig. 5 Nitrogen adsorption and desorption isotherms of NiO film (inset: corresponding BJH pore size distribution).

The GIXRD pattern of as-deposited NiO films on Au/Cr coated PET substrates is shown in Fig. 6(a). It is seen that the diffraction peaks can match well with the polycrystalline cubic NiO (JCPDS: 78-0429) and gold (JCPDS: 04-0784). For NiO, the diffraction on (111) plane is only clearly visible white diffractions on (200) and (220) are very weak and almost the same as background noise. The result is due to the fact that the GIXRD technique is not very sensitive to plane perpendicular to the film surface. Nevertheless, this method allows a significant reduction of diffraction from the Au film on PET substrate. It can also be noticed that there is a significant background slope at low diffraction angles due to the effect of amorphous PET substrate. In addition, it is observed that the NiO peaks appear to be quite narrow, implying quite large crystallite size of NiO nanoparticles in the film, which is in contrast to the observed nanoparticle size determined by TEM of 3–6 nm. The paradox could be due to the fact that diffraction signal of NiO is not much larger compared with that background noise so that the peak shape and related crystallite size are highly uncertain. Thus, GIXRD only allows the confirmation of NiO phase in the film while the actual crystallite size of NiO nanoparticles in the film cannot yet be determined but it can be implied that the value should be smaller than the size of TEM primary particles.

Fig. 6 (a) GIXRD pattern of sparked NiO film on Au/Cr coated PET substrate and (b) UV-vis spectrum (inset: plot of (αhν)² vs. hν).

Fig. 6(b) illustrates a typical UV-vis spectrum of sparked NiO-NPs on glass substrate. It is seen that NiO-NPs exhibit strong absorption in the near-UV region, which can mainly be attributed to the band gap absorption.³³ The value of the band gap (E_g) can be calculated based on the fundamental absorption equation, which is given by:³⁴

αhνⁿ = (hν − E_g) (1)

where α is the absorption coefficient, E_g is the energy band gap, hν is photon energy, A is a constant relative to the materials and n is a characteristic number, 1/2 and 2 for indirect and direct band gap transitions, respectively. For NiO, n = 2 since it is a direct band gap semiconductor. The inset graph presents the plots of (αhν)² as a function of hν. It is seen that the plot (black curve) is linear only in the high energy region (>3.6 eV) where fundamental band gap absorption occurs. The absorption at lower energy comes from other process such as absorption via defect states and thus does not conform to eqn (1). E_g of NiO-NPs is estimated to be around 3.47 eV by extrapolation of the linear region to the energy intercept (blue line). The value is in accordance with the value of NiO nanostructures in literature^35,36 and thus confirms the formation of NiO nanoparticles.

The surface compositions and chemical states of sparks NiO are further confirmed by using XPS. Fig. 7(a) shows the Ni 2p core levels of sparked NiO-NPs. The core levels, Ni 2p_3/2 and Ni 2p_1/2, can be deconvoluted in to four doublet pairs at ∼853.8 : 871.4, ∼854.9 : 872.5, ∼855.9 : 873.4 and ∼857.2 : 874.7 eV, respectively. These doublet peaks may be assigned to Ni²⁺ of NiO, Ni²⁺ of Ni(OH)₂, Ni³⁺ of Ni₂O₃ and Ni³⁺ of NiOOH, respectively.^37–39 For the oxygen element (Fig. 7(b)), the O 1s core level spectrum can be similarly decomposed into four components located at ∼529.3, ∼530.9, ∼532.0 and ∼533.4 eV, respectively. The peak at 529.3 corresponds to the lattice oxygen of NiO while the ones at ∼530.9, 532 and 533.4 eV can be assigned to adsorbed oxygen species, adsorbed OH– group due to humidity^12,40,41 and adsorbed –C–O– groups due to hydrocarbon impurities.⁴² The results indicate that NiO-NPs formed by sparking method are complex nickel oxide.

Fig. 7 (a) Ni 2p and (b) O 1s XPS spectra of sparked NiO film.

The electrochemical characteristics of the sparked NiO film studied using CV and GCD are illustrated in Fig. 8. Fig. 8(a) shows a comparison of the CV curves in the potential window of 0 to 0.45 V (vs. Ag/AgCl) of the Cr–Au electrode on PET substrate with and without the sparked NiO films. It is evident that only NiO films exhibits high and sharp redox peaks, indicating its highly effective faradaic nature.⁴³ In contrast, the Cr–Au electrode on the PET substrate gives very low signal current and negligible peak. Thus, the NiO film will contribute almost all of charge storage capacity from the structure. The anodic and cathodic peaks at ∼0.41 and ∼0.32 V can be attributed to the redox reactions between NiO film and KOH electrolyte, in which NiOOH is oxidized to NiO and NiO is reduced to NiOOH, respectively. The reversible reaction is given by⁴⁰

NiO + OH⁻ ↔ NiOOH + e⁻ (2)

Fig. 8 (a) CV curve of the Cr–Au on PET substrate vs. sparked NiO film at a scan rate of 10 mV s⁻¹, (b) CV graph of NiO film at different scan rates and (c) oxidation peak current vs. square root of scan rate of sparked NiO film.

According to the XPS data, Ni(OH)₂ is also formed at the electrode surface. Thus, Ni(OH)₂ may also be reversibly converted to NiOOH according the redox reaction:^44,45

Ni(OH)₂ + OH⁻ ↔ NiOOH + H₂O + e⁻ (3)

However, the contribution of the reaction (3) to the redox peaks in CV should be relatively small since the fraction of Ni(OH)₂ found from the XPS data is considerably smaller than that of NiO.

Fig. 8(b) illustrates the CV profiles of the NiO film at different scan rates. It is seen that the CV peak current density increases monotonically with increasing scan rate, suggesting a good rate capability of the NiO films. In addition, the redox peak potential also increases accordingly with increasing scan rate. Fig. 8(c) shows the plot of the oxidation peak current vs. square root of scan rate. It is evident that the oxidation peak current is linearly proportional to the square root of scan rate, indicating that the redox reaction at NiO electrode is diffusion limited.⁴⁶ The characteristic is different from a typical capacitive behavior, in which peak current is proportional to the scan rate.

The energy-storage characteristic of NiO electrode is then evaluated by GCD technique. Fig. 9(a) and (b) displays GCD curves of NiO films at low and high ranges of discharge current density, respectively. It can be seen that all discharge curves consists of three regions with initial short fast discharge (0.45–0.4 V) followed by a long and slow discharge region (0.4–0.3 V) and final rapid discharge section (0.3–0 V), which is consistent with a typical characteristic of battery-like energy-storage device and in accordance with the faradaic CV behavior. The long discharge region corresponds to the potential at which redox peaks occur in the CV curve when the accumulated charges from redox reactions are released. Due to its battery-like characteristics, the performance of the sparked NiO electrode should be expressed in term of specific charge capacity (or in short specific capacity), which can be estimated from the GCD profile according to the defining equation:

where I = discharge current, t = discharge time and m = mass of electrode material. From calculation, the sparked NiO electrode has the specific capacity values of 402.75, 368.55, 336.6, 308.7, 284.85, 275.85, 263.25 and 260.1 C g⁻¹ at 1, 2, 4, 8, 16, 20, 28 and 40 A g⁻¹, respectively. The specific charge capacity is plotted as a function of current density as displayed in Fig. 9(c). It is evident that the specific capacity rapidly decreases as the current density increases from 1 to 8 A g⁻¹ but then becomes very slow decreasing as the current density increases further from 8 to 40 A g⁻¹.

Fig. 9 (a–b) GCD data of NiO films at low and high ranges of discharge current density, (c) specific capacity curves at a variety of current densities (inset: photograph of bent electrode before testing), (d) GCD profiles of repeated 10 cycles at 40 A g⁻¹ and (e) cycling performance of sparked NiO films at 40 A g⁻¹.

Moreover, the cycling performances of sparked NiO films at 40 A g⁻¹ is demonstrated in Fig. 9(d) and (e). It can be seen that NiO electrode shows highly repeatable charging/discharging profile in the first 10 cycles (Fig. 9(d)). Furthermore, it has a good cycling ability with a good specific capacity retention of 88% for 1000 cycles of operation at the highest current density of 40 A g⁻¹ (Fig. 9(e)). It is seen that the loss of capacity occurs mostly during first 300 cycles and the loss becomes much less as the number of cycles increases further up to 1000. The initial loss of capacity may be due to the agglomeration of the NiO particles^13,47 or structural change of NiO nanoparticles. This problem may be alleviated by appropriate post-deposition annealing of NiO film on the Au/PET substrate. Nevertheless, the sparked NiO film can maintain 88% of initial capacity up to 1000 cycles, which is still comparable with some other reports from porous NiO on Ni foam.¹³

In most reports on NiO-based electrochemical energy-storage devices, the performances were reported in term of specific capacitance instead of specific capacity due to the misconception of psuedocapacitor for nickel or cobalt oxide and hydroxide.⁴⁸ To compare the results with other reports, the average specific capacitance is converted from the specific capacity by dividing with the potential window used in this study of 0.45 V. The average specific capacitance values of sparked NiO electrode are calculated to be 895, 819, 748, 686, 633, 613, 585 and 578 F g⁻¹ (or 33.6, 30.7, 29.1, 28.0, 27.1, 25.7, 25.0, 24.4, 23.7, 23.4, 23.0, 22.0 and 21.7 mF cm⁻²) at 1, 2, 4, 8, 16, 20, 28 and 40 A g⁻¹, respectively. The values are comparable with some NiO-based devices, NiO nanoflowers, nanoslices and nanoparticles,¹² NiO flake-like and hierarchical porous ball-like,⁴⁹ porous thin film NiO nanowires,⁵⁰ mesoporous NiO nanoflake arrays⁵¹ and NiO flower-like microspheres,⁴⁷ which report the specific capacitance in the range of 500–2000 F g⁻¹ or 1–50 mF cm⁻² at similar current densities. It should be noted that some work reported very high specific capacitance value but with narrower potential window or lower current density. Moreover, a preliminary mechanical durability test was performed. The electrochemical characteristics and specific capacity of sparked NiO film were not changed considerably after a typical rolling process. In addition, no physical delamination or damage of sparked NiO film was observed. Thus, the sparked NiO film could have sufficient mechanical stability for capacitor fabrication process.

The excellent electrochemical energy-storage performances of sparks NiO film may be attributed to large specific surface area of NiO NPs for redox reactions and charge storage. In addition, the porous structure of NiO films produced by sparking process assists the transportation of ions/electrons between electrode and electrolyte and reduce ion transfer resistance to the electrodes, which could result in a low effective series resistance.²² The sparking technique can produce more porous structure compared with several other chemical methods since no binder is included, leading to the fully accessible surface area of NiO-NPs for electrochemical reaction. Thus, the sparked NiO film is considered to be an attractive choice for electrochemical energy-storage applications due to low cost, simplicity and high charge capacity. Although the results achieved with the present sparking process parameters are considered decent, some parameters including scanning speed and the number of scanning cycles may be further optimized for optimal structure and electrochemical energy storage performances. The further optimization is under way and the results may be reported elsewhere.

Conclusions

In conclusion, NiO films were successfully fabricated by a sparking method and characterized for electrochemical energy-storage applications. The sparked NiO films exhibited excellent electrochemical energy storage performances with a high specific charge capacity of 402.75 C g⁻¹ at a discharge current of 1 A g⁻¹ and 88% capacity retention after 1000 cycles at a high discharge current of 40 A g⁻¹. The remarkable specific charge capacity could be attributed to the high electroactive surface area of highly porous films containing NiO nanoparticles. The key advantages of this technique include its convenient, low-temperature operation, and the low-cost system. Thus, the sparking method holds a promise as a practical and effective preparation technique of highly porous metal oxide films suitable for energy-storage applications.

Acknowledgements

The authors gratefully acknowledge the financial support by Thailand Graduate Institute of Science and Technology (TGIST), the Materials Science Research Center, Department of Physics and Materials Science, Faculty of Science, Chiang Mai University and the Graduate School, Chiang Mai University. Also, we would like to thank National Electronics and Computer Technology Center (NECTEC), Pathumthani, Thailand and Research Laboratory for Analytical Instrument and Electrochemistry Innovation, Faculty of Science, Chiang Mai University for electrochemical test. Moreover, the authors would like to acknowledge Mrs B. Kuntalue, Mr M. Kongtungmon T. Sakorn and E. Kuntarak for helping with TEM, SEM operation and AFM. In addition, the authors would like to thank Dr G. S. Roberts for language improvement.

Notes and references

1. B. T. Raut, S. G. Pawar, M. A. Chougule, S. Sen and V. B. Patil, J. Alloys Compd., 2011, 509, 9065–9070.
2. S. Pereira, A. Gonçalves, N. Correia, J. Pinto, L. Pereira, R. Martins and E. Fortunato, Sol. Energy Mater. Sol. Cells, 2014, 120, 109–115.
3. H. Dalvand, G. Reza Khayati, E. Darezereshki and A. Irannejad, Mater. Lett., 2014, 130, 54–56.
4. G. Özkan and E. Özçelik, J. Power Sources, 2005, 140, 28–33.
5. C. J. Hung, P. Lin and T. Y. Tseng, J. Power Sources, 2014, 259, 145–153.
6. K. M. Kim, J. H. Nam, Y.-G. Lee, W. I. Cho and J. M. Ko, Curr. Appl. Phys., 2013, 13, 1702–1706.
7. J. Zhu, L. Cao, Y. Wu, Y. Gong, Z. Liu, H. E. Hoster, Y. Zhang, S. Zhang, S. Yang, Q. Yan, P. M. Ajayan and R. Vajtai, Nano Lett., 2013, 13, 5408–5413.
8. Q. Wang, L. Jiao, H. Du, Y. Wang and H. Yuan, J. Power Sources, 2014, 245, 101–106.
9. R. S. Kalubarme, M.-S. Cho, K.-S. Yun, T.-S. Kim and C.-J. Park, Nanotechnology, 2011, 22, 395402.
10. A. I. Inamdar, Y. Kim, S. M. Pawar, J. H. Kim, H. Im and H. Kim, J. Power Sources, 2011, 196, 2393–2397.
11. A. Jena, N. Munichandraiah and S. A. Shivashankar, J. Power Sources, 2013, 237, 156–166.
12. S.-I. Kim, J.-S. Lee, H.-J. Ahn, H.-K. Song and J.-H. Jang, ACS Appl. Mater. Interfaces, 2013, 5, 1596–1603.
13. H. Wang, H. Yi, X. Chen and X. Wang, Electrochim. Acta, 2013, 105, 353–361.
14. Y. Jiang, D. Chen, J. Song, Z. Jiao, Q. Ma, H. Zhang, L. Cheng, B. Zhao and Y. Chu, Electrochim. Acta, 2013, 91, 173–178.
15. X.-h. Xia, J.-p. Tu, X.-l. Wang, C.-d. Gu and X.-b. Zhao, J. Mater. Chem., 2011, 21, 671–679.
16. M.-S. Wu, C.-Y. Huang and K.-H. Lin, J. Power Sources, 2009, 186, 557–564.
17. M.-S. Wu, Y.-A. Huang, C.-H. Yang and J.-J. Jow, Int. J. Hydrogen Energy, 2007, 32, 4153–4159.
18. M.-S. Wu, C.-H. Yang and M.-J. Wang, Electrochim. Acta, 2008, 54, 155–161.
19. M. M. Rahman, S.-L. Chou, C. Zhong, J.-Z. Wang, D. Wexler and H.-K. Liu, Solid State Ionics, 2010, 180, 1646–1651.
20. X.-M. Liu, X.-G. Zhang and S.-Y. Fu, Mater. Res. Bull., 2006, 41, 620–627.
21. I. Porqueras and E. Bertran, Thin Solid Films, 2001, 398–399, 41–44.
22. H. Wang, Y. Wang and X. Wang, Electrochem. Commun., 2012, 18, 92–95.
23. W. Thongsuwan, T. Kumpika and P. Singjai, Curr. Appl. Phys., 2008, 8, 563–568.
24. K. Hongsith, N. Hongsith, D. Wongratanaphisan, A. Gardchareon, S. Phadungdhitidhada, P. Singjai and S. Choopun, Thin Solid Films, 2013, 539, 260–266.
25. K. Inyawilert, A. Wisitsora-at, A. Tuantranont, P. Singjai, S. Phanichphant and C. Liewhiran, Sens. Actuators, B, 2014, 192, 745–754.
26. I. Nam, S. Park, G.-P. Kim, J. Park and J. Yi, Chem. Sci., 2013, 4, 1663–1667.
27. T. Kumpika, W. Thongsuwan and P. Singjai, Surf. Interface Anal., 2007, 39, 58–63.
28. S. Cho, K.-H. Shin and J. Jang, ACS Appl. Mater. Interfaces, 2013, 5, 9186–9193.
29. H. Long, T. Shi, H. Hu, S. Jiang, S. Xi and Z. Tang, Sci. Rep., 2014, 4 DOI:10.1038/srep07413.
30. A. Kumar, A. Saxena, A. De, R. Shankar and S. Mozumdar, Adv. Nat. Sci.: Nanosci. Nanotechnol., 2013, 4, 025009.
31. V. Kruefu, A. Wisitsoraat and S. Phanichphant, J. Nanomater., 2015, 2015, 8.
32. J. H. de Boer, B. C. Lippens, B. G. Linsen, J. C. P. Broekhoff, A. van den Heuvel and T. J. Osinga, J. Colloid Interface Sci., 1966, 21, 405–414.
33. S. Nandy, S. Goswami and K. K. Chattopadhyay, Appl. Surf. Sci., 2010, 256, 3142–3147.
34. M. Salavati-Niasari, N. Mir and F. Davar, J. Alloys Compd., 2010, 493, 163–168.
35. V. Patil, S. Pawar, M. Choug, P. Godse, R. Sakhare, S. i. Sen and P. Joshi, J. Surf. Eng. Mater. Adv. Technol., 2011, 1, 35–41.
36. L.-P. Zhu, G.-H. Liao, Y. Yang, H.-M. Xiao, J.-F. Wang and S.-Y. Fu, Nanoscale Res. Lett., 2009, 4, 550–557.
37. J.-Z. Huang, Z. Xu, H.-L. Li, G.-H. Kang and W.-J. Wang, Trans. Nonferrous Met. Soc. China, 2006, 16, 1301–1306.
38. Y. Li and Y. He, RSC Adv., 2014, 4, 16879–16884.
39. M. V. Bykova, D. Y. Ermakov, V. V. Kaichev, O. A. Bulavchenko, A. A. Saraev, M. Y. Lebedev and V. A. Yakovlev, Appl. Catal., B, 2012, 113–114, 296–307.
40. J. B. Wu, R. Q. Guo, X. H. Huang and Y. Lin, J. Power Sources, 2013, 243, 317–322.
41. A. Fihri, R. Sougrat, R. B. Rakhi, R. Rahal, D. Cha, M. N. Hedhili, M. Bouhrara, H. N. Alshareef and V. Polshettiwar, ChemSusChem, 2012, 5, 1241–1248.
42. X. Feng, N. Dementev, W. Feng, R. Vidic and E. Borguet, Carbon, 2006, 44, 1203–1209.
43. G. X. Pan, X. H. Xia, F. Cao, P. S. Tang and H. F. Chen, Electrochem. Commun., 2013, 34, 146–149.
44. R. A. Huggins, Advanced Batteries: Materials Science Aspects, Springer Science & Business Media, 2008.
45. R. Mohd Syafiq Hamdan and M. R. Othman, Int. J. Electrochem. Sci., 2013, 8, 4747–4760.
46. B. Vidhyadharan, N. K. M. Zain, I. I. Misnon, R. A. Aziz, J. Ismail, M. M. Yusoff and R. Jose, J. Alloys Compd., 2014, 610, 143–150.
47. D. Du, Z. Hu, Y. Liu, Y. Deng and J. Liu, J. Alloys Compd., 2014, 589, 82–87.
48. T. Brousse, D. Bélanger and J. W. Long, J. Electrochem. Soc., 2015, 162, A5185–A5189.
49. S. K. Meher, P. Justin and G. Ranga Rao, ACS Appl. Mater. Interfaces, 2011, 3, 2063–2073.
50. A. Paravannoor, R. Ranjusha, A. M. Asha, R. Vani, S. Kalluri, K. R. V. Subramanian, N. Sivakumar, T. N. Kim, S. V. Nair and A. Balakrishnan, Chem. Eng. J., 2013, 220, 360–366.
51. X. Yan, X. Tong, J. Wang, C. Gong, M. Zhang and L. Liang, J. Alloys Compd., 2014, 593, 184–189.

---