Sputtering and sulfurization-combined synthesis of a transparent WS₂ counter electrode and its application to dye-sensitized solar cells

Abstract

In this work, continuous and large-area tungsten sulfide (WS₂) films, deposited by radio frequency sputtering followed by a sulfurization process, were applied as a low-cost platinum (Pt)-free counter electrode (CE) for dye-sensitized solar cells (DSSCs). The composition and structure of WS₂ films were confirmed using X-ray diffraction, field-emission scanning electron microscopy, Raman spectroscopy and X-ray photoemission spectroscopy techniques. The WS₂ CE was phase pure and considerably transparent. The cyclic voltammetry, electrochemical impedance spectroscopy and Tafel curve showed that the WS₂ CE possesses high electrocatalytic activity and fast reaction kinetics for the reduction of tri-iodide to iodide, which can be attributed to its inherent catalytic property. Finally, TiO₂-based DSSC with an optimized WS₂ CE (sputtered for 10 min) showed as high as 6.3% power conversion efficiency, which was comparable to the performance of DSSC with a Pt-based CE (6.64%). Our study demonstrated the feasibility to develop low-cost, transparent, catalytically active, stable and abundant metal chalcogenide catalysts by an RF sputtering method to replace Pt CE for photovoltaic application.

---

Introduction

The dye-sensitized solar cells (DSSCs) is a promising candidate for replacing traditional silicon-based solar cells due to its simple fabrication process, environmental friendliness, high power conversion efficiency (PCE), availability of easy and low-cost sensitizers, simple production, high stability and wider range of design possibilities.^1,2 To date, a myriad of efforts have been made to improve the PCE of DSSCs by using platinum (Pt) counter electrode (CE), cobalt(II/III) electrolytes and zinc porphyrin.³ Among them, conventional Pt as a CE is an expensive noble metal, which greatly restricts the commercial production of DSSCs. Pt can also be decomposed to PtI₄ by redox couple in electrolyte.^4,5 Many attempts have been made to replace the conventional Pt CE with low-cost alternatives such as graphite,⁶ carbon black,⁷ carbon nanotubes,⁸ conducting polymers,⁹ metal chalcogenides,^10,11 nitrides,¹² carbides¹³ and phosphides.¹⁴ On the other hand, environmentally benign, abundant and non-expensive transition metal di-chalcogenides (MX₂, M = Mo, W and X = S, Se) have been the subject of extensive investigation due to their strong catalytic activities and potential electrochemical applications.^15–17 Among the varieties of layered transition metal sulfides, tungsten sulfide (WS₂) is composed of three atomic layers: a tungsten layer is sandwiched between two sulfur layers, and the triple layers are stacked and held together by weak van der Waals interactions. It is an indirect band gap semiconductor with an energy gap of ∼1.2 eV in bulk form.^16,18 Due to its analogous structure to graphene and potential electrocatalytic activity, WS₂ is considered as versatile for electrochemical applications. For example, WS₂ is extensively used for variety of applications such as solid state lubricants, catalysts for hydride sulfurization and hydrogen evolution reaction and anode materials for lithium ion batteries.^19–21 However, only a few reports have so far focused on the application of WS₂ as CE in DSSCs. Wu et al.¹¹ synthesized molybdenum sulphide (MoS₂) and WS₂ by a hydrothermal method. They employed them as CE materials for DSSCs and achieved PCEs of 7.59% and 7.73%, respectively. Recently, Yue et al.²² reported DSCCs with a PCE of 6.41% under a simulated solar illumination of 100 mW cm⁻² (AM 1.5) using multi-wall carbon nanotubes decorated with WS₂ (MWCNTs-WS₂) CE, which were obtained using a hydrothermal method. However, to avoid the complex solution-based preparation processes which are not eco-friendly, development of a pure WS₂ thin film directly grown onto a conductive substrate deserves further exploration. In this study, we present a scalable and reproducible RF sputtering method for the fabrication of a large area, continuous WS₂ film onto fluorine-tin-oxide (FTO)/glass substrate. The TiO₂-based DSSCs with an optimized WS₂ CE exhibited an impressive 6.3% PCE, which is comparable to DSSCs result with a Pt CE (6.64%).

Experimental details and device fabrication

Pieces of FTO/glass substrates (∼2 × 2 cm²) were ultrasonically degreased in acetone, methanol, isopropyl alcohol and deionized water and then baked at 120 °C for 5 min. After loading the pieces in a sputter chamber, the chamber was evacuated by a rotary pump and a turbomolecular pump combination to achieve the pressure at ∼1 × 10⁻⁷ torr. Next, WO₃ thin films were deposited onto FTO/glass substrates using a W target by RF magnetron sputtering. During the film deposition, the Ar and O₂ flow ratio was maintained at 5 and 15 sccm, respectively and the RF power was fixed at 180 W. The substrate temperature was fixed at room temperature. The sputtering time was varied (5, 10, 15 min) to control the thickness of WS₂ film. After taking out the samples from the sputter chamber, the as-sputtered films were placed downstream of a chemical vapor deposition (CVD) chamber and heated. The as-sputtered films were subjected to sulfurization process at 650 °C for 30 min to form WS₂ film and also to improve the crystalline quality of the films. A pure sulfur powder (99.99%) was placed upstream of the CVD chamber, and a heating filament for the sulfur boat was fixed at 120 °C. The sulfur powder was evaporated at 120 °C using a carrier gas, Ar (100 sccm), and the pressure of the CVD chamber was kept at 2 × 10⁻² mTorr. The distance between the sulfur boat and the substrate and the flow of the carrier gas were fixed for all the experiments. The amount of the sulfur powder was optimized at 0.5–0.8 g for the controlled growth of WS₂ film.

2.1 Fabrication of DSSCs using Pt CEs

A Pt CE was prepared by using a drop-cast method, as reported previously.^23,24 In brief, a thin TiO₂ blocking layer with 100 nm thickness was deposited onto a FTO/glass substrate. A 0.2 M TiCl₄ (20–30 wt% in HCl) was prepared using H₂O₂ at 50–55 °C and crystallized at 350 °C for 60 min. After preparation of the solution, few drops were placed onto the FTO/glass substrate and put on a hot plate at 70 °C for 45 min. A TiO₂ layer (thickness ∼4 μm) with an average particle size of ∼20 nm was coated by the doctor blade method. A light scattering TiO₂ layer (thickness ∼3 μm) was applied over the previously deposited TiO₂ and sintered again at 450 °C for 30 min. The prepared TiO₂ photoanode was dipped in 0.3 mM N719 (ethanol + acetonitrile) dye for 24 h so as to form working electrode for DSSCs. The DSSCs were fabricated by injecting a liquid electrolyte (0.005 M I₂, 0.1 M Lil, 0.6 M tetrabutylammonium iodide, and 0.5 M 4-tert-butylpyridine in acetonitrile) through an aperture between the dye-sensitized TiO₂ electrode and the CEs (MoS₂ & Pt in the present case).

2.2 Characterization techniques

The phase purity analysis and structure confirmation of WS₂ CEs were studied using a Raman microscope at 514 nm wavelength. The structural formation and crystal orientation were verified by using X-ray diffraction (XRD, Rigaku) with Cu-Kα radiation operated at 50 kV and 300 mA. The morphological evolution of WS₂ films as a function of RF sputtering time was confirmed from the field-emission scanning electron microscopy (FE-SEM, Nova nano SEM200-100 FEI) images. Chemical configurations were determined by X-ray photoelectron spectroscopy (XPS), model PHI 5000 Versa Probe (Ulvac-PHI) using a monochromatic Al K_α X-ray source (1486.6 eV). The data were collected from a spot-size of 100 × 100 μm². The carbon 1s peak (284.6 eV) was used as a reference for internal calibration.

2.3 Photo-electrochemical measurements

A solar simulator (150 W Xe lamp, Sun 2000 solar simulator, ABET 5 Technologies, USA) equipped with an AM 1.5G filter was used to generate simulated sunlight with a corrected intensity of 1 sun (100 mW cm⁻²). The photocurrent density–applied voltage (J–V) spectra were obtained using a Keithley 2400 source meter. Incident photon to current conversion efficiency (IPCE) spectra were obtained without bias potential under illumination with respect to a calibrated Melles-Griot silicon diode, and by changing the excitation wavelength (McScience K3100 spectral IPCE measurement system and Polaronix® K102 signal amplifier). Electrochemical impedance spectroscopy (EIS) analysis was performed by IviumStat Technologies, Netherlands. The frequency of applied sinusoidal AC voltage signal varied from 150 kHz to 0.1 MHz.

Results and discussion

The schematic diagram of the WS₂ RF sputtering preparation process and DSSCs module is illustrated in Fig. 1.

Fig. 1 Schematic diagram for the preparation of WS₂ CE and its role in DSSCs.

Fig. 2a–c shows optical microscopy images of WS₂ films obtained at different sputtering times. The observed films seem to be uniform and continuous. Additionally, the films become darker (thicker) with increasing the sputtering time. Also, 3D images of prepared WS₂ films for different sputtering time are presented in Fig. (S1).†

Fig. 2 (a–c) Optical images of WS₂ films obtained at different sputtering times; (a) 5, (b) 10 and (c) 15 min and (d) the Raman spectra of the corresponding WS₂ films (sputtered for 5, 10 and 15 min).

Raman analysis was performed to perceive the crystal quality of WS₂. Raman spectra of the annealed WS₂ thin films are shown in Fig. 2d. From Fig. 2d, two characteristic WS₂ Raman peaks related to the in-plane vibration of W and sulfur atoms, E_2g¹ mode at ∼354.01 cm⁻¹ and out-of-plane vibration of sulfur atoms, A_1g mode at ∼418.4 cm⁻¹, were obtained. The frequency difference (Δk) between two Raman peaks is closely related with the layer number and can be used to determine the thickness of atomically thin WS₂.^25,26 The Δk increased with increasing sputtering time. The obtained Δk were ∼64.3 cm⁻¹, ∼65 cm⁻¹ and ∼65.5 cm⁻¹ for 5 min, 10 min and 15 min sputter time, respectively, corresponding to few-layer WS₂ films [Fig. 2d].

The crystallinity and purity of the WS₂ thin films were analysed by XRD analysis. From the Fig. 3, all the reflections were corresponding solely to WS₂ and substrate. The diffraction pattern revealed intense peaks at the positions of 37.2°, 42.9°, 51.1°, 54.6°, 61.6°, 65.7°, 70.9°, 78.3°, 80.5°, 83.5° and 87.5°, which were assigned to (103), (006), (105), (106), (107), (114), (202), (109), (205), (206) and (118) lattice planes of WS₂ crystal structure, respectively, evidencing that there was no trace of WO₃ after sulfurization process. The substrate-related FTO peaks were clearly noticed in the XRD patterns. Presence of intense XRD peaks was an indication of the polycrystalline nature and the resultant diffraction peaks corroborated well with the standard patterns for the hexagonal crystal structure of WS₂ (JCPDS # 87-2417 file). The XRD results demonstrated no other phase transformation or other impurities in the crystalline film. The observed XRD reflections were highly sharp, indicating the high crystallinity of the WS₂ film surface. No other significant variations were observed in the XRD patterns for WS₂ films prepared at different sputtering time periods.

Fig. 3 XRD patterns of WS₂ films sputtered at different sputtering times.

Surface morphology acts as a key role in the development of thin film solar cells.²⁷ FE-SEM was used to see the surface morphology of the WS₂ material over FTO/glass substrate [Fig. 4a–c]. Some crystallites larger than the normal due to the agglomeration process were detected. The valleys and hillocks were noticed over the sample prepared at 5 min-sputtering time [Fig. 4a]. The various sizes and shapes of crystallites were noticed for WS₂ sputtered for 10 min [Fig. 4b]. The highest crystallite sizes were found for WS₂ film sputtered for 15 min (Fig. 4c). The elemental mapping graph of WS₂ (Fig. 4d) confirmed the presence of Sn (from substrate), tungsten, and sulfur; consistent to XPS results presented below.

Fig. 4 FE-SEM micrographs of WS₂ CE sputtered for; (a) 5, (b) 10, and (c) 15 min. (d) Elemental mapping of WS₂ (sputter time 10 min).

To further study the chemical composition and surface electronic states of WS₂, XPS analysis was carried out (sample sputtered for 10 min). The survey XPS spectra (Fig. 5a) indicate the presence of sulfur, carbon, and oxygen etc., elements in the WS₂ film. The C_1s peak, related to sp³ hybridized carbon, was observed at 284.8 eV [Fig. 5b]. Fig. (5c and d) shows the XPS spectrum of WS₂ film. The W 4f core-level spectrum showed three peaks corresponding to the W 4f_5/2 and W 4f_7/2 levels for W⁴⁺ and W 5p_3/2 electronic states at 34.8, 32.7, and 38.0 eV, respectively. The S 2p core-level spectrum exhibited two peaks at 163.5 and 162.3 eV, corresponding to the S 2p_1/2 and S 2p_3/2 states, respectively. These results are well-consistent with the reported values for WS₂ single crystals.

Fig. 5 XPS spectra of WS₂ film sputtered at 10 min; (a) survey spectra, (b) carbon related C_1s peak, (c) tungsten doublet peak (W⁴⁺), and (d) sulphur (S²⁻).

In this study, the fabricated transparent WS₂ CEs were applied for DSSCs (Fig. 1). Fig. S2† shows J–V characteristic curves of DSSCs with WS₂ CEs with different thicknesses. The obtained photovoltaic performance parameters are summarized in Table S1.† The observed PCEs (η) of TiO₂-based DSSCs with WS₂ CEs sputtered for 5, 10 and 15 min were 5.4%, 6.3% and 5.8%, respectively. The highest efficiency was obtained from WS₂ CE sputtered at 10 min. J. Wu et al.²⁸ demonstrated DSSCs with WS₂ CE, but the PCE of DSSCs based on pristine WS₂ CE was limited to 5.32%. The PCE could be enhanced by inclusion of high-conductive MWCNT along with WS₂ in the presence of glucose so called as G-A WS₂/MWCNT hybrid CE. The PCEs of our WS₂-based electrode were also superior to that of the previously reported plastic and carbon-coated WS₂ CEs.²⁹

Fig. 6a presents J–V characteristic curves of DSSCs with Pt and WS₂ CEs under illumination of 100 mW cm⁻². J–V curves of two DSSCs are nearly rectangular and identical, demonstrating an excellent catalytic behaviour of WS₂ as like Pt. DSSCs employing WS₂ CE exhibits PCE of 6.3%, short-circuit current (J_sc) of 13.43 mA cm⁻², open-circuit voltage (V_oc) of 0.71 V, and fill factor (FF) of 0.66 [Table 1]. The IPCE curves of TiO₂-based DSSCs with Pt and WS₂ CEs (Fig. 6b) also present a similar behavior, suggesting that almost similar (to Pt) catalytic activity is obtained by the transparent WS₂ CE.

Fig. 6 (a) J–V characteristic curves, (b) IPCE measurements of TiO₂-based DSSCs with WS₂ and Pt CEs.

Table 1 DSSCs parameters of WS₂ (sputtered at 10 min) and Pt-CEs

	DSSCs type	Jsc (mA cm^−2)	Voc (V)	FF (%)	η (%)
1	TiO2–Pt CE	16.50	0.66	0.61	6.64
2	TiO2–WS2 CE	13.43	0.71	0.66	6.3

---

The PCE value for WS₂ CE is close to that obtained for Pt CE (6.64%). The slight difference in the V_oc and J_sc values for the DSSCs with Pt and WS₂ CE can be attributed to several factors i.e., (a) surface roughness difference, (b) conductivity, (c) catalytic activity, and (d) chemical stability etc. Electrocatalytic activity is highly dependent on catalyst morphology, grain size, surface texture, crystalline structure and electrical conductivity of CE.^30,31 The PCE dependence in DSSCs with different WS₂ film thicknesses (sputter time periods) could be attributed to the different morphologies of WS₂ films obtained for different time periods (Fig. 4). The catalytic activity of WS₂ originates from the sulfur edges as reported previously. It is believed that with increasing the sputtering time more active sulfur edges are formed which ultimately enhances the electro-catalytic activity.^30,31

In this work, transparent (few-layers) WS₂ electrode was used for the first time as the CE material in DSSCs. To estimate the electrocatalytic performance of the WS₂ towards oxidation and reduction, cyclic voltammetry (CV) studies were performed. Fig. 7a shows the CVs of the system for Pt and WS₂ in the potential interval between −0.8 and 0.8 V. In this process, electrons are injected into photo-oxidized dye from I⁻ ions in the electrolyte [eqn (1)], and the produced I₃⁻ ions are reduced on the CE [eqn (2)]. In this interval, the anodic peak is contributed by the reaction (1) in an anodic sweep and the cathodic peak is contributed by the reaction (2) in a cathodic sweep, respectively.

I₃⁻ + 2e⁻ ⇔ 3I⁻ (1)

3I₂ + 2e⁻ ⇔ 2I₃⁻ (2)

Fig. 7 Cyclic voltammetry curves of; (a) I⁻/I₃⁻ redox system using Pt and WS₂ CEs, (b and c) WS₂ CE at different scan rates and durability test for 100 consecutive cycles at a 10 mV s⁻¹ scan rate. (d) Tafel polarization curves of symmetrical cells obtained with Pt and WS₂ CEs.

More CV tests of WS₂ CE for the redox reaction were carried out by changing scan rates (25–150 mV s⁻¹) as shown in Fig. 7b. It was observed that the peak current densities were gradually increased with the increasing the scan rate. Simultaneously, the diffusion coefficient D of Pt and WS₂ CEs can be estimated from the Randles–Sevcik equation:

I_p = Kn^3/2AC(D)^1/2V^1/2 (3)

where I_p represents the peak current, K is the constant of 2.69 × 10⁵, n is the number of electrons transferred in the redox event, A is the electrode area, C represents the bulk concentration of I₃⁻ species, D is the diffusion coefficient and V means the scan rate. The diffusion coefficient of I₃⁻ for the Pt (∼7.56 × 10⁻⁵) is larger than that for the WS₂ CE (∼4.98 × 10⁻⁵), presumably arising from the high active surface area and surface roughness difference of both electrodes.

To investigate the electrochemical stability of the WS₂ CE, 100 consecutive CV cycles were performed in the potential region from −0.8 to 0.8 V vs. Ag/AgCl [Fig. 7c]. After 100 consecutive scans, the CV shape of WS₂ was changed slightly and the current densities of its redox peaks remained stable, suggesting that the WS₂ CE possessed the characteristics of reversible activity, excellent electrochemical, chemical stability and strong adhesiveness on the FTO glass substrate. TiO₂ was used as a dye material and plays an important role due to many advantages such as having the most efficient photoactivity, the highest stability and the lowest cost.^32,33 Two types of photochemical reactions occur on the TiO₂ surface when irradiated with ultraviolet light. One includes the photo-induced redox reactions of adsorbed substances, and the other is the photo-induced hydrophilic conversion of TiO₂ itself.^32,34

Tafel polarization measurements were performed on the symmetrical cells using WS₂ (10 min sputter time) and Pt CEs, and the corresponding curves are shown in Fig. 7d. Also, Tafel polarization measurements of WS₂ obtained for different deposition time periods are given in Fig. (S4).† The tangent slope of Tafel curves provides the information about the exchange current density (J_o). The commercial Pt electrode has larger J_o compared with sputtered WS₂ electrode, suggesting its higher electrocatalytic activity and lower charge-transfer resistance at the electrolyte–electrode interface.

Furthermore, EIS spectrums for symmetrical cells were carried out to examine the overall cell resistance. The Nyquist plots (Fig. 8a) for the symmetric cells with the Pt and WS₂ CEs illustrate impedance characteristics. The semicircle in the high frequency region (left) is due the charge-transfer process of the electrolyte–counter electrode interface, which changes inversely with the catalytic activity of the CE on the reduction of tri-iodide, and the corresponding constant phase element (CPE). The one in the low frequency region (right) is assigned to the Nernst diffusion process of tri-iodine ions. The impedance values (calculated using Z-View curve fitting software) are presented in Table 2. The R_ct value of the WS₂ and Pt CEs is 7.2 and 5.0 Ω cm², respectively. In addition, R_s is another important factor affecting the fill factor and performance of DSSCs.³⁵ Due to semiconducting behaviour, the WS₂ CE showed relatively higher R_s value than Pt CE.^36,37 Higher J_sc value in Pt CE DSSCs could be due to an availability of its larger active surface area for redox reaction.³⁸

Fig. 8 Nyquist plots for (a) Pt and (b) WS₂ CEs with inset of (b) as an equivalent circuit.

Table 2 EIS fitting parameters used for WS₂ and Pt CEs

CE	Pt	WS2
Rs (Ω cm^2)	3.1	19.5
R1 (Ω cm^2)	8.11	13.39
CPE1−t × 10^−3	2.45	7.29
CPE1−p	0.88	0.90
R2 (Ω cm^2)	1.96	2.11
CPE2−t × 10^−3	0.98	0.50
CPE2−p	0.60	0.096

---

Conclusion

In summary, continuous and large-area transparent WS₂ films deposited onto FTO substrate by RF magnetron sputtering were used as a low-cost and platinum-free counter electrode materials for dye-sensitized solar cells. X-ray diffraction and XPS results have verified the phase purity and surface composition of WS₂ deposited onto FTO substrate. CV, EIS and Tafel curve measurements indicated that the WS₂-based CE hold good electrocatalytic activity and fast reaction kinetics for the redox reaction. Finally, the DSSC with an optimized WS₂ CE (sputtered for 10 min) showed as high as 6.3% PCE, which is comparable to the performance of DSCC with Pt CE (6.64%). The present work demonstrated the feasibility to develop low-cost, transparent, efficient and abundant metal chalcogenide electrocatalysts (WS₂) with high catalytic activity and stability by RF sputtering method to replace Pt electrocatalyst for photovoltaic applications.

Acknowledgements

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (2010-0020207, 2012R1A1A2007211), by the Human Resources Development of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Trade, Industry & Energy (No. 20154030200630), and by the MOTIE (Ministry of Trade, Industry & Energy) (10052928) and KSRC (Korea Semiconductor Research Consortium) support program for the development of the future semiconductor device. Authors, RSM and MN extend their gratitude to the Visiting Professor Program (VPP) Unit of King Saud University (KSU), Saudi Arabia for the financial support.

References

1. B. O'regan and M. Grätzel, Nature, 1991, 353, 737–740.
2. E. Stathatos, P. Lianos, U. Lavrencic-Stangar and B. Orel, Adv. Mater., 2002, 14, 354.
3. A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–634.
4. E. Olsen, G. Hagen and S. E. Lindquist, Sol. Energy Mater. Sol. Cells, 2000, 63, 267–273.
5. A. Kay and M. Grätzel, Sol. Energy Mater. Sol. Cells, 1996, 44, 99–117.
6. M. J. Ju, J. C. Kim, H.-J. Choi, I. T. Choi, S. G. Kim, K. Lim, J. Ko, J.-J. Lee, I.-Y. Jeon and J.-B. Baek, ACS Nano, 2013, 7, 5243–5250.
7. H. Choi, H. Kim, S. Hwang, W. Choi and M. Jeon, Sol. Energy Mater. Sol. Cells, 2011, 95, 323–325.
8. T. Chen, L. Qiu, Z. Cai, F. Gong, Z. Yang, Z. Wang and H. Peng, Nano Lett., 2012, 12, 2568–2572.
9. L. Kavan, J.-H. Yum and M. Grätzel, Nano Lett., 2011, 11, 5501–5506.
10. C.-T. Li, C.-P. Lee, Y.-Y. Li, M.-H. Yeh and K.-C. Ho, J. Mater. Chem. A, 2013, 1, 14888–14896.
11. M. Wu, Y. Wang, X. Lin, N. Yu, L. Wang, L. Wang, A. Hagfeldt and T. Ma, Phys. Chem. Chem. Phys., 2011, 13, 19298–19301.
12. G. R. Li, F. Wang, Q. W. Jiang, X. P. Gao and P. W. Shen, Angew. Chem., Int. Ed., 2010, 49, 3653–3656.
13. A. Hauch and A. Georg, Electrochim. Acta, 2001, 46, 3457–3466.
14. M. Wu, J. Bai, Y. Wang, A. Wang, X. Lin, L. Wang, Y. Shen, Z. Wang, A. Hagfeldt and T. Ma, J. Mater. Chem., 2012, 22, 11121–11127.
15. J. N. Coleman, M. Lotya, A. O'Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. de and R. J. Smith, Science, 2011, 331, 568–571.
16. Y. Zhang, Y. Zhang, Q. Ji, J. Ju, H. Yuan, J. Shi, T. Gao, D. Ma, M. Liu and Y. Chen, ACS Nano, 2013, 7, 8963–8971.
17. D. Voiry, H. Yamaguchi, J. Li, R. Silva, D. C. Alves, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy and G. Eda, Nat. Mater., 2013, 12, 850–855.
18. A. L. Elías, N. Perea-López, A. Castro-Beltrán, A. Berkdemir, R. Lv, S. Feng, A. D. Long, T. Hayashi, Y. A. Kim and M. Endo, ACS Nano, 2013, 7, 5235–5242.
19. R. Bhandavat, L. David and G. Singh, J. Phys. Chem., 2012, 3, 1523–1530.
20. A. Sobczynski, A. Yildiz, A. J. Bard, A. Campion, M. A. Fox, T. Mallouk, S. E. Webber and J. M. White, J. Phys. Chem., 1988, 92, 2311–2315.
21. L. Rapoport, Y. Bilik, Y. Feldman, M. Homyonfer, S. Cohen and R. Tenne, Nature, 1997, 387, 791–793.
22. G. Yue, J. Wu, J.-Y. Lin, Y. Xiao, S.-Y. Tai, J. Lin, M. Huang and Z. Lan, Carbon, 2013, 55, 1–9.
23. S. B. Ambade, R. B. Ambade, R. S. Mane, G.-W. Lee, S. F. Shaikh, S. A. Patil, O.-S. Joo, S.-H. Han and S.-H. Lee, Chem. Commun., 2013, 49, 2921–2923.
24. X. Xin, M. He, W. Han, J. Jung and Z. Lin, Angew. Chem., Int. Ed., 2011, 50, 11739–11742.
25. H. Li, Q. Zhang, C. C. R. Yap, B. K. Tay, T. H. T. Edwin, A. Olivier and D. Baillargeat, Adv. Funct. Mater., 2012, 22, 1385–1390.
26. C. Lee, H. Yan, L. E. Brus, T. F. Heinz, J. Hone and S. Ryu, ACS Nano, 2010, 4, 2695–2700.
27. G. Yue, J.-Y. Lin, S.-Y. Tai, Y. Xiao and J. Wu, Electrochim. Acta, 2012, 85, 162–168.
28. J. Wu, G. Yue, Y. Xiao, M. Huang, J. Lin, L. Fan, Z. Lan and J.-Y. Lin, ACS Appl. Mater. Interfaces, 2012, 4, 6530–6536.
29. Y. Wang, S. Li, Y. Bai, Z. Chen, Q. Jiang, T. Li and W. Zhang, Electrochim. Acta, 2013, 114, 30–34.
30. B. Hinnemann, P. G. Moses, J. Bonde, K. P. Jørgensen, J. H. Nielsen, S. Horch, I. Chorkendorff and J. K. Nørskov, J. Am. Chem. Soc., 2005, 127, 5308–5309.
31. T. F. Jaramillo, K. P. Jørgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
32. M. Thomalla and H. Tributsch, J. Phys. Chem. B, 2006, 110, 12167–12171.
33. M. Shanmugam, T. Bansal, C. A. Durcan and B. Yu, Appl. Phys. Lett., 2012, 100, 153901.
34. N. G. Park and K. Kim, Phys. Status Solidi A, 2008, 205, 1895–1904.
35. P. Joshi, L. Zhang, Q. Chen, D. Galipeau, H. Fong and Q. Qiao, ACS Appl. Mater. Interfaces, 2010, 2, 3572–3577.
36. L. Han, N. Koide, Y. Chiba, A. Islam, R. Komiya, N. Fuke, A. Fukui and R. Yamanaka, Appl. Phys. Lett., 2005, 86, 213501.
37. N. Koide, A. Islam, Y. Chiba and L. Han, J. Photochem. Photobiol., A, 2006, 182, 296–305.
38. C. H. Yoon, R. Vittal, J. Lee, W.-S. Chae and K.-J. Kim, Electrochim. Acta, 2008, 53, 2890–2896.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra18613g

---