Plasma-modified SnO₂:F substrate for efficient cobalt selenide counter in dye sensitized solar cell

Abstract

Cobalt selenide (Co_0.85Se) counter electrodes (CEs) were synthesized in situ on plasma-treated fluorine-doped tin oxide (FTO) substrates using a hydrothermal approach. FTO glass substrates were treated using O₂/Ar direct current (DC) plasma for 5 min prior to the cobalt selenide growth. It was found that Co_0.85Se developed horizontally and vertically oriented, submicron or micron sized, and tremelliform-like structures on the plasma-modified FTO surface. This unique Co_0.85Se nanomaterial had a much larger accessible surface area, more active catalytic sites, and better catalytic properties compared to the case without plasma treatment. The electronic and ionic processes in dye sensitized solar cells (DSSCs) based on cobalt selenide CEs with or without plasma treatment as well as the Pt CE were analyzed and compared. The device with the Co_0.85Se on the plasma-treated FTO produced an energy conversion efficiency of 8.04%, which is significantly superior to that for the DSSC with the Pt CE (7.66%) and also higher than that (7.88%) for the device with the Co_0.85Se CE on the pristine FTO without plasma treatment. Plasma treatment of transparent conducting oxides has been proposed as an effective method for in situ deposition of high-quality inorganic compound CE nanomaterials and improving the electrocatalytic activities of inorganic compound CEs.

---

1. Introduction

Dye sensitized solar cells (DSSCs) have been intensively studied because their conversion efficiencies exceed 11%, and the accompanying economic, scientific, and technical advantages of these cells have the potential to be widely applied.^1–3 The classical DSSC has a sandwich-type structure consisting of a counter electrode (CE), an electrolyte, and a dye-sensitized photoanode. The function of a CE is to reduce the redox species used as a mediator in regenerating the sensitizer after electron injection in a liquid-state/quasi-solid state DSSC, or collect the holes from the hole-conducting material in a solid state DSSC.^1,4 The standard CE used in DSSCs is fluorine-doped tin oxide (FTO) glass supported platinum (Pt), which is normally prepared by a sputtering or thermal decomposition method. A FTO-supported Pt CE has high conductivity, excellent electrocatalytic activity, and good chemical stability. However, the high cost of the noble metal Pt and the possible decomposition of the Pt in the I₃⁻/I⁻ redox couple electrolyte restrain the large-scale fabrication and long-term stability of the DSSCs. Therefore, considerable efforts have been made to explore low-cost and stable alternatives to replace Pt CEs. Many low-cost materials, such as conducting polymers,⁵ graphene,^6,7 carbon,^8–10 nitrides,^11,12 sulfides,^13–15 and selenides^16–18 have been successfully used for CEs. Despite numerous investigations, only a few materials have shown better performance than platinum.¹⁹ Recently, Gong et al. synthesized Co_0.85Se and Ni_0.85Se non-Pt CEs with a low-temperature hydrothermal approach.¹⁶ The power conversion efficiency (PCE) of the DSSCs based on the grown Co_0.85Se CE is 0.76% higher than that of the device based on the Pt CE. The Co_0.85Se CE exhibited obviously higher electrocatalytic activity than Pt for the reduction of triiodide, but Ni_0.85Se showed inferior catalytic activity than Pt. In another study, NiSe₂ was used as the CE, and the DSSCs produced a PCE of 8.69%, which is higher than that obtained with the Pt CE-based DSSCs (8.04%).²⁰ CoSe prepared by an electrodeposition method was employed as the CE for DSSCs. After optimization, the DSSCs gave a PCE of 7.30%, higher than the DSSCs using a Pt CE (6.91%).²¹ Guo et al. prepared NbSe₂ nanosheets (NSs) and nanorods (NRs) via a solvothermal approach. The prepared NSs and NRs were used as CEs for DSSCs, which produced PCE values of 7.34 and 6.78%, close to Pt CE-based DSCs (7.90%).²² A ternary selenide CuInGaSe₂ CE prepared by a magnetron sputtering technology was reported to give a PCE of 7.13% in its DSSC device, comparable to the efficiency of the DSSC using a Pt CE (6.89%).¹⁸ We also explored a simple, eco-friendly screen-printing process for quaternary selenide Cu₂ZnSn(S,Se)₄ (CZTSSe) CEs and achieved a PCE of 5.75%.¹⁷ The catalytic activity of selenides as CEs are significantly affected by the morphological, physical (i.e., particle size, porosity, crystal structure), and chemical properties of nanoselenides; however, insufficient research has been performed in these areas. Moreover, the mechanism of the catalytic activity of selenides for regeneration of the electrolyte is poorly understood.

Low-temperature plasma is an effective tool for the fabrication and manipulation of nanomaterials and thin films.²³ It has been previously applied for surface modification/optimization of various nanomaterials and thin films, including different transparent conducting oxides (TCOs) such as indium tin oxide (ITO),²⁴ zinc oxide (ZnO),²⁵ FTO,²⁶ and TiO₂ films^27–29 with different gas combinations such as argon, hydrogen, oxygen, carbon tetrafluoride, or sulfur hexafluoride. Enhancement of the DSSC performance has been demonstrated by plasma treatments of TiO₂ films due to the increase in surface hydrophilicity and nanoparticle packing density and improvement of oxide's stoichiometry, leading to increased dye loading as well as improved electron transport and reduced recombination.^28,29

In this study, we propose the use of plasma-treated FTO glass as a substrate for the growth of cobalt selenide CEs for DSSCs via an in situ deposition method. We aim to advance the understanding of the structural features of cobalt selenide CEs assisted by plasma treatment and their effect on the photovoltaic performance of the device. The FTO glass substrates were treated using 20 sccm O₂/30 sccm Ar plasma for 5 min prior to the nano-selenide growth. Cobalt selenide (Co_0.85Se) was synthesized in situ on plasma-treated FTO substrates using a modified hydrothermal method.¹⁶ For comparison, cobalt selenide on a pristine FTO glass substrate was prepared using the same hydrothermal approach, and Pt on a pristine FTO glass substrate was fabricated using the thermal decomposition method. The grown cobalt selenide was applied directly as CEs to assemble DSSCs without any post-treatments. The catalytic properties of cobalt selenide nanomaterials for reduction of I₃⁻ were examined by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The cell with the Co_0.85Se on the plasma-treated FTO produces an energy conversion efficiency of 8.04%, which is higher than that for the DSSC with the Pt CE (7.66%), and also higher than that (7.88%) for the device with the CoSe CE on pristine FTO without plasma treatment.

2. Experimental

2.1 Preparation of cobalt selenide and Pt CEs

Se powders, CoCl₂·6H₂O, and N₂H₄·H₂O were purchased from the Sinopharm Chemical Reagent Co., Ltd, China. The reagents were used as received without further purification. Transparent conductive glass (F-doped SnO₂, FTO, 15 Ω sq⁻¹, Nippon Sheet Glass Co., Ltd, Japan) was used as the substrate material. FTO substrates were cleaned by sonication in glass detergent (Hui Jie Washing Ltd, Shenzhen), acetone, and isopropyl alcohol, and subsequently dried in a vacuum oven. The cleaned FTO glass was treated using 20 sccm O₂/30 sccm Ar plasma for 5 min prior to the selenide nanomaterial growth. The direct current (DC) plasma power was set at 30 W. The working pressure was approximately 40 Pa. Next, Se powders (0.048 mmol, 99.0% in purity) and CoCl₂·6H₂O (0.04 mmol) were dissolved in 11 mL of deionized water. The resulting mixture was transferred to a 100 mL Teflon-lined autoclave. Then, 3.0 mL of N₂H₄·H₂O (85 wt%) was added in the autoclave with vigorous stirring for 10 min. The plasma-treated FTO substrates were put into the autoclave and placed at an angle against its Teflon liner wall with the substrate-conducting layer facing down. The autoclave was sealed and maintained at 120°C for 8 h, which is a shorter amount of time than that reported in.¹⁶ For comparison, cobalt selenide and Pt were prepared on pristine FTO glass substrates without plasma treatment. The former was prepared using the same hydrothermal method as described above. The latter was printed using a paste based on H₂PtCl₆ dispersed in a mixture of terpineol and ethylcellulose. The printed layers were heated at 385°C for 20 min.³⁰

2.2 Fabrication of DSSCs

The TiO₂ working electrodes with a total thickness of approximately 16 μm on the FTO glass plates were prepared according to the procedures described in our previous work.⁶ The as-prepared photoanodes were dipped in a dye solution (0.5 mM N719 (Solaronix) in acetonitrile and tert-butyl alcohol (volume ratio of 1 : 1)) at room temperature for 20 h. The cells were sealed with a Surlyn 1702 (Dupont) gasket with a thickness of 60 μm. A drop of electrolyte solution (0.05 M I₂, 1 M MPII, 0.5 M guanidine thiocyanate, and 0.5 M tert-butylpyridine in acetonitrile) was injected into the inter-space between the photoanode and CE. Finally, the holes on the back of the CE were sealed with a Surlyn film and covered with a thin glass slide under heat.

2.3 Characterization

The morphologies and microstructures of the formed inorganic compound counter-electrode layers were characterized by field emission scanning electron microscopy (FESEM, S4800, Hitachi and JSM-7001F, JEOL) and X-ray diffractometry (XRD, Rigaku ULTIMA IV, D/tex detector, Cu-Kα: λ = 0.15406 nm). The thickness of the layers was measured with a profilometer (Dektak 6M). EIS and CV measurements of DSSCs were recorded with a galvanostat (PG30.FRA2, Autolab, Eco Chemie B. V Utrecht, Netherlands) under illumination of 100 mW cm². The electrolyte was the same as that used with the DSSCs. The frequency range was from 10 to 100 KHz, and the applied bias voltage and ac amplitude were set at open-circuit voltage and 10 mV, respectively, between the counter electrode and the working electrode. The impedance spectra were analyzed by an equivalent circuit model that interprets the characteristics of the DSSCs.^31,32 Photocurrent–voltage (I–V) measurements were performed using an AM 1.5 solar simulator equipped with a 1000 W xenon lamp (Model no. 91 192, Oriel, USA). The solar simulator was calibrated using a standard silicon cell (Newport, USA). The light intensity was 100 mW cm⁻² on the surface of the test cell. I–V curves were measured using a computer-controlled digital source meter (Keithley 2440). The area of the solar cells is 0.196 cm².

3. Results and discussion

The XRD pattern of the precipitate from the autoclave reactions is shown in Fig. 1. All of the observed diffraction peaks could be perfectly indexed to the hexagonal system with the lattice constants a = 3.615 Å, c = 5.283 Å, which matched well with the standard data file of Co_0.85Se (JCPDS file no. 52-1008). The three strongest peaks of the pattern, at 2θ = 33.3°, 44.9° and 51.1°, are assigned to the (101), (102) and (110) planes of the hexagonal close-packed (hcp) Co_0.85Se.

Fig. 1 XRD pattern of the obtained Co_0.85Se powders.

SEM was used to study the surface microstructure and morphology of the grown cobalt selenide nanomaterials. Fig. 2 shows SEM images of Co_0.85Se nanomaterials grown in situ on the pristine (a–c) and the plasma-treated (d–f) FTO substrates, and Pt (g and h) on the pristine FTO substrate. Co_0.85Se has significantly different surface features on the pristine and the plasma-treated FTO substrates. However, they are composed of widely divided cobalt selenide nanosheets or aggregates. Without plasma treatment, Co_0.85Se tended to grow in two dimensions (2D), extending parallel to the FTO surface, and formed flat, mono dispersive nanosheets. The typical lateral dimensions of the formed Co_0.85Se nanosheets is between 100 nm and 300 nm, as shown in Fig. 2(a–c). In contrast, with the assistance of the plasma treatment, Co_0.85Se grew in three dimensions (3D), considerably developed both parallel and perpendicular to the FTO surface, and formed tremelliform nanosheets with much larger lateral dimensions of 300–1500 nm, as shown in Fig. 2(d–f). Many Co_0.85Se tremelliform nanocrystals clustered together on the FTO, clearly shown in Fig. 2(d). Fig. 2(g) and (h) show typical SEM images of the platinum film of the electrode prepared by thermal decomposition. The formed Pt particles (1–5 nm in size) were unevenly distributed. Similar to the situation of Co_0.85Se nanomaterials grown in situ on the plasma-treated FTO substrates (Fig. 2(d)), some platinum particles aggregated to form large clusters of approximately 1–2 μm in size on the SnO₂ particles of the FTO substrate, as shown as in Fig. 2(g) and (h). Nevertheless, Pt has a much smaller grain size and much higher grain density than Co_0.85Se grown on either the pristine or the plasma-treated FTO substrates as shown in Fig. 2(c), (f) and (h).

Fig. 2 SEM images of Co_0.85Se grown in situ on the pristine (a–c) and the plasma-treated (d–f) FTO substrates, and Pt (g and h) on the pristine FTO substrate.

From the previously described SEM measurement results, the formed flat Co_0.85Se on the pristine FTO shown in Fig. 2(a–c) is analogous to the curled leaf or graphene-like Co_0.85Se that was grown by the same chemical system as ours but with a longer time (12 h).¹⁶ Due to reduction or elimination of surface contamination, the plasma treatment improved the wettability of the FTO substrate and enhanced its surface energy,³³ increasing the growth rate, the grain size, and specific area of the Co_0.85Se nanomaterial. Tremelliform Co_0.85Se nanosheets with larger accessible surface areas were successfully synthesized in situ on the plasma-treated FTO substrate as shown in Fig. 2(d–f). This result is consistent with that reported by Liu et al.³⁴ In their work, tremelliform Co_0.85Se sheet powders were fabricated through a chemical reaction of Co(NO₃)₂·6H₂O, Na₂SeO₃, and N₂H₄·H₂O in an autoclave at 140°C for 24 h. Our tremelliform-like Co_0.85Se nanosheets, by contrast, were obtained at lower temperature and shorter time. Compared to the case without plasma treatment, the size and loading of Co_0.85Se obviously increases on the FTO substrate, which can enhance the surface area and active catalytic sites in the CE material. Liu et al. previously characterized the specific surface areas of as-deposited Co_0.85Se tremelliform nanosheet powders using the Brunauer–Emmer–Teller (BET) method.³⁴ They found that the tremelliform Co_0.85Se had a specific surface area of 55.1 m² g⁻¹, and this value was much higher than those of graphene-like Co_0.85Se nanocrystallines (11.8 m² g⁻¹) and Pt nanoparticles (5.8 m² g⁻¹) reported by Gong et al.¹⁶ From the SEM images in Fig. 2(a)–(c) of this study and Fig. 3(a) and (b) of ref. 16, it can be seen that the surface area of Co_0.85Se flat nanosheets grown without plasma treatment in our work is smaller than that (11.8 m² g⁻¹) in the latter. The formed Co_0.85Se tremelliform nanosheets on the plasma-treated FTO possess a much larger specific area, which can be expected to form a considerably larger contact surface area between the CE and electrolyte and speed up the diffusion of the electrolyte, thus improving the electrocatalytic activity in the DSSCs. This effect was proven by the subsequent electrochemical and photovoltaic measurements. Therefore, the plasma treatment of FTO glass prior to the deposition of cobalt selenide improved the morphological property and the quality of the deposited cobalt selenide CE.

Fig. 3 CV curves of the Pt electrode and Co_0.85Se CEs with and without DC plasma treatment. The scan rate and scan voltage range are 50 mV s⁻¹ and −0.4–1.0 V, respectively.

CV experiments were carried out to characterize the electrocatalytic activity of Pt and in situ-grown Co_0.85Se CEs toward triiodide reduction in an I⁻/I₃⁻ redox solution. The Co_0.85Se CEs with or without plasma treatment behaved quite similarly to the Pt/FTO electrode and exhibited two distinct pairs of oxidation and reduction peaks.³⁵ The relative negative or left pair is assigned to redox reaction (1), and the positive one is assigned to redox reaction (2).³⁶

I₃⁻ + 2e⁻¹ ↔ 3I⁻ (1)

3I₂ + 2e⁻¹ ↔ 2I₃⁻ (2)

Cobalt selenides show similar cathodic and anodic peaks compared to Pt, indicating that these selenides are effective in catalyzing the reduction of triiodide to iodide. The characteristics of the left pair peaks are at the focus of our analysis because the CE is responsible for catalyzing the reduction of I₃⁻ to I⁻ in a DSSC. Thus, the magnitude of the current density at the left reduction peak (I_p) is directly proportional to the ability of the electrode to reduce the I₃⁻ species. From the bottom left corner of the CV images shown in Fig. 3, it can be seen that the absolute I_p value of the Co_0.85Se CE with DC plasma treatment was 2.03 mA cm⁻², which was significantly higher than that of the Co_0.85Se sample without plasma treatment (1.04 mA cm⁻²), demonstrating higher catalytic activity towards the I⁻/I₃⁻ redox reaction and a faster reaction rate. These could be attributed to the unique physical structure of the former sample shown in Fig. 2. Compared to the case without plasma treatment, the size and loading of Co_0.85Se obviously increases on the FTO substrate, enhancing the surface area and active catalytic sites in the CE material. Moreover, the formed 3D cobalt selenide tremelliform network can allow the rapid diffusion of I₃⁻ to access the active sites and allow catalytic reduction toward I⁻, thereby improving the electrocatalytic activity in the solar cell. The absolute I_p value of the Pt CE is 0.98 mA cm⁻². The first and second highest peak current densities of Co_0.85Se CE with or without DC plasma pre-treatment, respectively, reveal that Co_0.85Se is a remarkable electrochemical catalyst for the reduction of I₃⁻.

To further evaluate the electrocatalytic activity of the as-prepared CEs for the reduction of triiodide, EIS tests were performed, as shown in Fig. 4. The corresponding resistances were fitted with Nova software (v.1.9, Metrohm Autolab) in terms of the equivalent circuit shown in the inset of Fig. 4. Generally, the typical Nyquist plots of the DSSCs show three semicircles in the measured frequency from 0.1 Hz to 100 kHz. The Ohmic serial resistance (R_s) is associated with the series resistance of the electrolytes and electrical contacts in the DSSCs. R_ct1, R_ct2, and R_ct3 correspond to the charge transfer processes occurring at the counter electrode (corresponding to the first arc), the TiO₂–dye–electrolyte interface (corresponding to the second arc), and the Warburg diffusion process of I⁻/I₃⁻ in the electrolyte (corresponding to the third arc), respectively.^31,32 The R_s values of the Co_0.85Se-based DSSCs on the pristine and the plasma-treated FTO slides are 18.79 Ω cm² and 17.61 Ω cm², respectively. Both values are higher than the R_s (11.9 Ω cm²) of the DSSCs fabricated using the Pt CE. Considering that Co_0.85Se and Pt-based DSSCs were made by the same types of electrolyte and FTO substrate, the higher R_s can be attributed to the lower conductivity of Co_0.85Se compared to the metal Pt. However, compared to the case of the device based on the Pt CE, the charge transfer resistance (R_ct1) at the counter electrode was dramatically decreased when Co_0.85Se CEs made with plasma pre-treatment were used. The R_ct1 of the DSSCs with Co_0.85Se CEs on the plasma-treated FTO was only 7.05 Ω cm². This value is much smaller than that (14.88 Ω cm²) of Pt-based DSSCs, and also smaller than that (12.49 Ω cm²) of the cell based on the Co_0.85Se CE made without plasma treatment, suggesting that the Co_0.85Se CE with plasma treatment had the best electrocatalytic activity for I₃⁻ reduction.

Fig. 4 Nyquist plots of DSSCs assembled on the Pt electrode and Co_0.85Se CEs with and without DC plasma treatment.

Fig. 5 shows photocurrent density–voltage (J–V) curves of DSSCs fabricated with two kinds of Co_0.85Se CEs and a Pt electrode under a light intensity of 100 mW cm⁻². The average measured diode parameters for these cells based on different electrodes along with statistics are summarized in Table 1. The photovoltaic parameters including short circuit current density (J_sc), open circuit voltage (V_oc), power conversion efficiency (η), and fill factor (FF) are listed in the table. The DSSCs based on the Pt CE showed a J_sc of 16.79 mA cm⁻², a V_oc of 0.72 V, a FF of 63.36%, and therefore, an overall η of 7.66%. The DSSCs fabricated with Co_0.85Se CEs with or without plasma treatment showed enhanced FF values of 67.78% and 67.03%, respectively. The DSSCs with Co_0.85Se CEs on the plasma-treated FTO showed a J_sc value of 16.25 mA cm⁻² and an FF value of 67.78%, and thus, the highest η value of 8.04%, which is higher than that (7.66%) for the DSSCs with the Pt CE, and also superior to that of the device with Co_0.85Se CEs on the pristine FTO.

Fig. 5 J–V curves of DSSCs assembled on the Pt electrode and Co_0.85Se CEs with and without DC plasma treatment.

Table 1 Photovoltaic performance parameters for different CEs

CE	Voc (V)	Jsc (mA cm^−2)	FF (%)	η (%)
Pt	0.72 ± 0.01	16.79 ± 0.28	63.36 ± 0.01	7.66 ± 0.23
Co0.85Se (no plasma)	0.73 ± 0.01	16.10 ± 0.31	67.03 ± 0.02	7.88 ± 0.26
Co0.85Se (DC plasma)	0.73 ± 0.01	16.25 ± 0.34	67.78 ± 0.02	8.04 ± 0.28

---

Previous studies have demonstrated that a counter electrode usually affects three aspects of cell performance: the first is the electrical property or sheet resistance; the second is the electrochemical property or catalysis efficiency, which is determined typically by the inverse of charge transfer resistance; and the last is the optical property or reflection of illumination.^6,37–40 The redox reaction resistance R_ct1 at the counter electrode, the resistance R_ct3 of carrier transport by ions in the electrolyte, and resistance R_s due to the sheet resistance of TCO contribute to the internal series resistance of the cell.³⁹ These resistances bring negative effects on the fill factor, and thus, the energy conversion efficiency. The DSSCs based on the Co_0.85Se CE with DC plasma treatment showed the lowest resistance R_ct1. In view of the fact that Co_0.85Se and Pt-based DSSCs were made by the same types of electrolyte, TiO₂ photoanode, and FTO substrate, the lowest value of R_ct1 led to the lowest internal series resistance and resulted in the highest fill factor and the highest energy conversion efficiency of the device, as shown in Fig. 5 and Table 1. The first and second lowest value of R_ct1 for the DSSCs based on the two Co_0.85Se CEs also indicated that the electrochemical property or catalysis efficiency of the Co_0.85Se CE is better than that of Pt. The conclusions regarding the catalytic activity and the photovoltaic performance derived from the CV, EIS, and J–V data for the cobalt selenide and Pt CEs are consistent.

4. Conclusions

We have investigated plasma treatment of FTO glass for improving the performance of cobalt selenide CEs for DSSCs. The FTO glass substrates were treated using O₂/Ar DC plasma prior to the cobalt selenide growth. The surface morphology of cobalt selenide synthesized in situ on the plasma-modified FTO substrate changed dramatically compared to the case without plasma modification. EIS, CV, and J–V measurements show consistent results, which can be associated with the unique physical structures of the Co_0.85Se CEs. With plasma treatment, the size and loading of Co_0.85Se was obviously increased, and tremelliform Co_0.85Se nanosheets were formed on the FTO substrate. This Co_0.85Se CE has enhanced active catalytic sites, a larger accessible surface area, and improved catalytic properties. The charge transfer resistance (R_ct1) at the counter electrode is considerably decreased when Co_0.85Se CEs with plasma treatment are used, compared to the case of the Pt CE or the Co_0.85Se CEs without plasma treatment. The cell with Co_0.85Se on the plasma-treated FTO produces an energy conversion efficiency of 8.04%. It provides the best photovoltaic performance when compared to the DSSCs with the Pt CE (7.66%) and the device with the Co_0.85Se CE on the pristine FTO without plasma treatment (7.88%).

Acknowledgements

This work was supported by National Natural Science Foundation of China (no. 11274119, 61275038) and Large Instruments Open Foundation of East China Normal University.

References

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–724.
2. A. Kay and M. Grätzel, Sol. Energy Mater. Sol. Cells, 1996, 44, 99–117.
3. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. Curchod and N. Ashari-Astani, Nat. Chem., 2014, 6, 242–247.
4. M. Grätzel, J. Photochem. Photobiol., A, 2004, 164, 3–14.
5. S. Ahmad, J. H. Yum, Z. Xianxi, M. Grätzel, H. J. Butt and M. K. Nazeeruddin, J. Mater. Chem., 2010, 20, 1654–1658.
6. D. Zhang, X. Li, H. Li, S. Chen, Z. Sun, X. J. Yin and S. M. Huang, Carbon, 2011, 49, 5382–5388.
7. X. H. Miao, K. Pan, Q. J. Pan, W. Zhou, L. Wang, Y. P. Liao, G. H. Tian and G. F. Wang, Electrochim. Acta, 2013, 96, 155–163.
8. K. Suzuki, M. Yamamoto, M. Kumagai and S. Yanagida, Chem. Lett., 2003, 32, 28–29.
9. D. W. Zhang, X. D. Li, S. Chen, F. Tao, Z. Sun, X. J. Yin and S. M. Huang, J. Solid State Electrochem., 2010, 14, 1541–1546.
10. Y. P. Liao, K. Pan, L. Wang, Q. J. Pan, W. Zhou, X. H. Miao, B. J. Jiang, C. G. Tian, G. H. Tian, G. F. Wang and H. G. Fu, ACS Appl. Mater. Interfaces, 2013, 5, 3663–3670.
11. M. X. Wu, X. Lin, Y. D. Wang, L. Wang, W. Guo, D. Qi, X. J. Peng, A. Hagfeldt, M. Gratzel and T. L. Ma, J. Am. Chem. Soc., 2012, 134, 3419–3428.
12. Q. W. Jiang, G. R. Li and X. P. Gao, Chem. Commun., 2009, 6720–6722.
13. X. Xin, M. He, W. Han, J. Jung and Z. Lin, Angew. Chem., Int. Ed., 2011, 50, 11739–11742.
14. H. K. Mulmudi, S. K. Batabyal, M. Rao, R. R. Prabhakar, N. Mathews and Y. M. Lam, Phys. Chem. Chem. Phys., 2011, 13, 19307–19309.
15. Z. Li, F. Gong, G. Zhou and Z. S. Wang, J. Phys. Chem. C, 2013, 117, 6561–6566.
16. F. Gong, H. Wang, X. Xu, G. Zhou and Z. S. Wang, J. Am. Chem. Soc., 2012, 134, 10953–10958.
17. J. Shen, D. Zhang, J. Li, X. Li, Z. Sun and S. Huang, Nano-Micro Lett., 2013, 5, 281–288.
18. X. Y. Cheng, Z. J. Zhou, Z. L. Hou, W. H. Zhou and S. X. Wu, Sci. Adv. Mater., 2013, 5, 1193–1198.
19. S. Thomas, T. G. Deepak, G. S. Anjusree, T. A. Arun, V. S. Nair and A. Sreekumaran Nair, J. Mater. Chem. A, 2014, 2, 4474–4490.
20. F. Gong, X. Xu, Z. Li, G. Zhou and Z. S. Wang, Chem. Commun., 2013, 49, 1437–1439.
21. Z. Zhang, S. Pang, H. Xu, Z. Yang, X. Zhang, Z. Liu, X. Wang, X. Zhou, S. Dong and X. Chen, RSC Adv., 2013, 3, 16528–16533.
22. J. Guo, Y. Shi, C. Zhu, L. Wang, N. Wang and T. Ma, J. Mater. Chem. A, 2013, 1, 11874–11879.
23. K. Ostrikov, in Plasma Nanoscience: Basic Concepts and Applications of Deterministic Nanofabrication, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2008.
24. C. C. Wu, C. I. Wu, J. C. Sturm and A. Kahn, Appl. Phys. Lett., 1997, 70, 1348–1350.
25. A. Y. Polyakov, N. B. Smirnov, A. V. Govorkov, K. Ip, M. E. Overberg, Y. W. Heo, D. P. Norton, S. J. Pearton, B. Luo, F. Ren and J. M. Zavada, J. Appl. Phys., 2003, 94, 400–406.
26. M. S. Choi, Y. J. Lee, J. D. Kwon, Y. Jeong, J. Y. Park, Y. C. Ang, P. K. Song and D. H. Kim, Curr. Appl. Phys., 2013, 13, 1589–1593.
27. X. Zhang, M. Jin, Z. Liu, D. A. Tryk, S. Nishimoto, T. Murakami and A. Fujishima, J. Phys. Chem. C, 2007, 111, 14521–14529.
28. Y. S. Kim, B. J. Yoo, R. Vittal, Y. Lee, N. G. Park and K. J. Kim, J. Power Sources, 2008, 175, 914–919.
29. M. Konstantakou, T. Stergiopoulos, V. Likodimos, G. C. Vougioukalakis, L. Sygellou, A. G. Kontos, A. Tserepi and P. Falaras, J. Phys. Chem. C, 2014, 118, 16760–16775.
30. N. Pappas, M. Grätzel, Process for manufacturing an electrode for an electrochemical device, Polytechnique Federale Lausanne DE EP0852804, 1999.
31. Y. H. Lai, C. Y. Lin, J. G. Chen, C. C. Wang, K. C. Huang, K. Y. Liu, K. F. Lin, J. J. Lin and K. C. Ho, Sol. Energy Mater. Sol. Cells, 2010, 94, 668–674.
32. G. Zhu, X. J. Wang, H. L. Li, L. K. Pan, H. C. Sun, X. J. Liu, T. Lv and Z. Sun, Chem. Commun., 2012, 48, 958–960.
33. F. Lisco, A. Abbas, B. Maniscalco, P. M. Kaminski, M. Losurdo, K. Bass, G. Claudio and J. M. Walls, J. Renewable Sustainable Energy, 2014, 6, 011202.
34. C. C. Liu, J. M. Song, J. F. Zhao, H. J. Li, H. S. Qian, H. L. Niu, C. J. Mao, S. Y. Zhang and Y. H. Shen, Appl. Catal., B, 2012, 119–120, 139–145.
35. Z. Huang, X. Liu, K. Li, D. Li, Y. Luo, H. Li, W. Song, L. Chen and Q. Meng, Electrochem. Commun., 2007, 9, 596–598.
36. M. Wu, X. Lin, A. Hagfeldt and T. Ma, Chem. Commun., 2011, 47, 4535–4537.
37. A. Hauch and A. Georg, Electrochim. Acta, 2001, 46, 3457–3466.
38. Q. Wang, J. E. Moser and M. Grätzel, J. Phys. Chem. B, 2005, 109, 14945–14953.
39. L. Han, N. Koide, Y. Chiba and T. Mitate, Appl. Phys. Lett., 2004, 84, 2433–2435.
40. L. Han, N. Koide, Y. Chiba, A. Islam, R. Komiya, N. Fuke, A. Fukui and R. Yamanaka, Appl. Phys. Lett., 2005, 86, 213501.

---