Tungsten carbide/porous carbon core–shell nanocomposites as a catalyst support for methanol oxidation

Abstract

Carbon-encapsulated tungsten carbide (WC@C) was prepared by a microwave-assisted synthesis method with resorcinol-formaldehyde resin (RF) as carbon source. WC was encapsulated by porous carbon layer to form core–shell structure which could protect tungsten oxide from occupying the active sites on the surfaces of the WC@C. The characteristics of WC@C composites were determined by X-ray diffraction, scanning electron microscopy, transmission electron microscopy, energy dispersive X-ray spectroscopy and Brunauer–Emmet–Teller gas adsorption. Platinum nanoparticles were uniformly distributed on WC@C to synthesize a new electrocatalyst Pt–WC@C. The electro-catalytic performances of prepared Pt–WC@C, commercial Pt/C and PtRu/C toward methanol oxidation were compared by cyclic voltammetry, chronoamperometry and CO stripping test. It was found that Pt–WC@C exhibited higher catalytic activity for methanol oxidation than that of commercial Pt/C and PtRu/C catalysts. Especially, the Pt–WC@C achieved the long-term stability which was attributed to the effective protection by the carbon porous shell structure.

---

1. Introduction

Direct methanol fuel cells (DMFC) are considered as an attractive future electro-catalytic energy conversion technology for power applications due to their high theoretical specific energy of methanol and low operating temperature.^1,2 Currently, the most effective electro-catalyst for DMFC is the Pt metal or its alloy.^3–5 Although these catalysts exhibit desirable electrochemical activities, both Pt and its alloy are expensive due to the scarce world reserves of Pt. In addition, strong absorption of CO on the surface of Pt blocks the active sites for methanol oxidation and results in loss of activity.⁶ Consequently, the search for a low-cost, high-activity electro-catalyst in the anode compartment of DMFC is of growing interest as demand continues to rise. Although the Pt–Ru alloy catalysts are expected to be the best candidate catalyst for DMFCs. However, their high cost still hinders fuel cells commercialization and Ru dissolution from Pt–Ru catalysts is always a significant problem.^7,8 Several approaches have been attempted including loading small amount of Pt on non-noble metals.^9–13 Among the non-noble metals, tungsten carbide (WC) has attracted attention from electrocatalysis researchers because it exhibits the better synergistic effects with Pt.^14–17 WC surface is active toward the dissociation of H₂O to produce surface hydroxyl groups, which are propitious to the oxidation of the absorbed CO intermediate.¹⁸ It is also found that WC could be used as a support to increase the dispersion of the precious metal, and this WC based catalysts showed improved activity for methanol oxidation reaction (MOR).^19,20 However, WC materials prepared by conventional methods usually have high crystallinity and low surface area because of the high temperature in reduction and carburization process, which weakens the combination with platinum.²¹ The previous study suggests that another negative point about WC was WC particles were thermodynamically unstable and easy to be oxidized in aqueous system at room temperature. The resulted oxidized surface layer mainly consisting of WO₃ covered the surface of WC catalysts and became the barrier for the active sites.²² The study about voltammetric behavior of a mixture of WC and Pt also proved that soluble tungstate species, produced by chemical oxidation of WC, deposited on Pt in a form of hydrous tungsten oxide and partially covered Pt surface.²³ In this case, the WC based catalysts can't maintain their contents and structures over the long time. Consequently this change in the surface layer of WC catalysts will result in a loss of electrochemical activity. Herein, we intend to design a core–shell structure to physically isolate WC catalysts from the acid environment to avoid leaching, which however should not impede the catalytic properties of WC. This design is inspired from the researches including Pd/C core–shell nanocomposites for oxidation of alcohols, porous Fe₃O₄/C core–shell nanorods with good electromagnetic properties and other catalysts with similar core–shell features.^24–26 In these core–shell systems direct contact of metal particles with harsh environments including acid medium, oxygen, and sulfur contaminations is avoided. However, this protection does not impede the activation of catalysts. On the contrary, the catalysts have a rather high activity and long-term stability. This enhancement is discussed by DFT calculations, indicating that the catalytic activity could arise from the electron transfer from metal particles to the carbon shells leading to a decreased local work function on the carbon surface.²⁷

In this work, we fabricated a novel support WC@C by a microwave-assisted synthesis method with using resorcinol-formaldehyde resin (RF) as the carbon source. Typically, carbon-based nanomaterials have been used as promising supports for low temperature fuel cells because carbon-based material has the low electrical resistance to facilitate electron transport during the electrochemical reactions.²⁸ We design a carbon shell to inhabit or at least deplete agglomeration of the particles, which can improve the dispersity of the WC particles. Moreover, the carbon shell can protect WC from being oxidized in the electrolyte and thus increase their structure stability during cycling. We also use a porous structure in the carbon shell for obtaining maximum contact with fuel and strong interaction between catalyst and support.

Different from our previous work, in this study we used RF as in situ carbon source instead of the usual external carbon source such as CO gas. And the prepared porous WC@C material by microwave-assisted heating has a uniform size distribution and stable microsphere structure. By loading a relatively small amount of Pt on the WC@C microsphere we produced a new electrocatalyst for methanol oxidation which is far more active and stable than the commercial Pt/C and PtRu/C catalysts.

2. Experimental

2.1. Catalyst preparation

A solution containing 2 ml of formaldehyde (37%) and 1 g resorcinol was synthesized under vigorous stirring at room temperature to make RF. Ammonium metatungstate (AMT, (NH₄)6H₂W₁₂O₄₀·xH₂O) aqueous solution, prepared by dissolving 2.5 g of AMT in 20 ml distilled water, was then added into the resorcinol-formaldehyde solution with stirring for 30 min. Afterwards the mixed solution was heated by microwave system (Biotage, Initiator EXP) at 85 °C for 10 min, then to 150 °C and kept for 20 min. The as-made products were dried at 80 °C. All chemicals used were of analytical grade.

The resulting precursor was carburized in a tube furnace system where was swept by nitrogen gas to remove air. The precursor was heated to 400 °C at rate of 5 °C min⁻¹ and kept for 1 h, then the temperature was raised to 900 °C, held for 4 h, and then cooled down to room temperature. The whole in situ reduction carbonization was performed under argon and hydrogen without adding other carbon source. The obtained product was a black powder denoted as WC@C. The formation process of the WC@C composite is schematized in Fig. 1. For comparison, WC was synthesized by the same method, but the reduction carburization was completed under CO atmosphere. The noble metal Pt (10 wt%) particles were loaded on the WC@C by microwave heating according to an established method.²⁹ The product was marked as Pt–WC@C, and the 20% Pt–WC@C (Pt 20 wt%) material was prepared by the same method for comparison.

Fig. 1 Schematic illustrating the method to use microwave-assisted synthesis to form WC@C core–shell nanocomposites.

2.2. Catalyst characterization

The morphology and structure of catalysts were characterized by transmission electron microscopy (TEM) and scanning transmission electron microscope (STEM) using a Tecnai G2 F30 S-Twin microscope (FEI, Netherlands) at a voltage of 200 kV and a current of 103 mA, coupled with energy dispersive X-ray spectrometer (EDX, Thermo NORAN VANSTAGE ESI). The morphologies of the materials were also observed by scanning electron microscopy (SEM, Hitachi S-4700II) using an accelerating voltage of 3.0 kV. The X-ray diffraction patterns (XRD, Thermo ARL SCINAG X TRA) were used to characterize the crystalline structure of the products with Cu K_α source operated at 45 kV. N₂ adsorption–desorption was examined by ASAP 2020 (Micromeritics, America) and the specific surface area of WC@C was measured by Brunauer–Emmett–Teller (BET) method. Raman spectra were obtained at room temperature by Raman spectrometer (JOBIN YVON Lab RAM HR UV800) at an excitation wavelength of 514.5 nm.

2.3. Electrochemical measurements

The electrochemical measurements were carried out by CHI 660C electrochemical workstation in a standard three-electrode cell. A 3 mm diameter glass carbon electrode (GC) deposited with catalyst was used as the working electrode, Pt foil and saturated calomel electrode (SCE) were used as counter electrode and reference electrode respectively. 5 mg Pt–WC@C sample was dispersed in 0.1 ml isopropyl alcohol and 20 μl Nafion (5 wt%, Du Pont Corp. USA) under ultrasonic treatment for 10 min. Then the ink was deposited on the surface of GC-electrode and dried at 60 °C to remove the solvent. The Pt loading of the all electrodes is 8.3 μg.

The cyclic voltammetry (CV) curves were measured in 0.5 M H₂SO₄ with 0.5 M CH₃OH at a scan rate 50 mV s⁻¹. The chronoamperometry test (CA) was performed in the same electrolyte 0.5 M H₂SO₄ containing 0.5 M CH₃OH at potential of 0.2 V. For the CO stripping voltammetry, CO was bubbled into the 0.5 M H₂SO₄ for 30 min while keeping the potential at −0.14 V after the electrolyte was purged with nitrogen for 30 min. Then the nitrogen was purged again for 30 min to remove trace of dissolved CO. Finally the CO stripping voltammetry was measured from −0.2 to 1.0 V at a scan rate 50 mV s⁻¹ at room temperature. The other electrochemistry experiments were carried out at 50 ± 0.5 °C. Both the CV and CA were obtained after the electrolyte was purged with high purity nitrogen for 20 min. The commercial Pt/C (20% Pt on Vulcan XC-72R, Johnson Matthey Corp.) and PtRu/C catalyst (10% Pt : 10% Ru on Vulcan XC-72R, Johnson Matthey Corp.) were also tested for comparison.

3. Results and discussion

Cyclic voltammograms of WC@C and pure WC recorded in 0.5 M H₂SO₄ at a scan rate 50 mV s⁻¹ from −0.2 to 1.0 V is given in Fig. 2. The voltammogram of the WC@C powder is stable over time while an anodic peak is observed at 0.8 V on pure WC. This anodic peak represents the voltammetric behavior of WO₃, which indicates that the WC surface is stable in its oxidized state. The oxidation of WC induced by anodic polarization has been reported by Lee et al.³⁰ They found the similar result that a sharp anodic peak at 0.8 V could be observed for pure WC in aqueous solution. In the presence of water and/or oxygen, WC is oxidized to surface oxide and/or soluble W(VI) species via the following reactions:^31,32

WC + 5H₂O → WO₃ + CO₂ + 10H⁺ + 10e⁻ (1)

WC + 6H₂O → WO₄²⁻ + CO₂ + 12H⁺ + 10e⁻ (2)

Fig. 2 Cyclic voltammograms of WC and WC@C in 0.5 M H₂SO₄ at 50 mV s⁻¹ from −0.2 to 1.0 V.

Different from the pure WC, no anodic peak can be observed on WC@C in the whole scan range suggesting the carbon shell encapsulated on WC can prevent the WC from further oxidation.

The degree of graphitization of the carbon in WC@C composite was investigated by Raman spectra, shown in Fig. 3. The spectra show two Raman characteristic peaks at 1310 and 1590 cm⁻¹. The peak at around 1310 cm⁻¹, be named the D-band, can be attributed to the presence of sp³ defects within the carbon. The D-band is associated with the A_1g vibration of disordered carbon atoms. The peak at 1590 cm⁻¹, named the G-band, corresponds to vibration of graphitic carbon atoms. The G-band is associated with the Raman-active E_2g mode. The intensity ratio of the D-band to G-band (I_D-band/I_G-band) is indicative of the graphite site density.³³ The ratio value of I_D-band/I_G-band for the prepared WC@C is 1.182, implying a micro-ordered graphitic structure for carbon shells.

Fig. 3 Raman spectra of WC@C material.

Fig. 4 displays the XRD patterns of the WC@C and Pt–WC@C samples. It can be found that the sample WC@C is mainly composed of hexagonal WC phase whose main peaks could be clearly identified according to the JCPDS card no. 00-002-1055, suggesting that most of tungsten source AMT has been in situ carbonized to WC in the presence of the resorcinol formaldehyde resin. The diffraction peaks at 31.507°, 35.550° and 48.255° can be assigned to the crystal planes of WC (001), (100) and (101), respectively. Some diffraction peaks of W₂C are also detected the WC@C sample. The diffraction peaks at 39.939°, 46.176° and 67.660° correspond to the crystal planes of face-centered-cubic (fcc) platinum cell (111), (200) and (220). In addition, the average particle size of Pt calculated using the Debye–Scherrer equation³⁴ from the plane (200) is 5.18 nm. No characteristic peak of graphite-like carbon is observed at 2θ = 25.5°. This absence of the graphitic structure implies that only amorphous carbon aerogel can be formed from resorcinol formaldehyde resin.

Fig. 4 XRD patterns of the WC@C and Pt–WC@C particles.

The structural and morphological characteristics of WC@C samples were observed by SEM and TEM, as shown in Fig. 5. It can be found from Fig. 5a and b, the microsphere appears to be an aggregate of small WC particles that are wrapped with an amorphous carbon shell. The same results are also proved by TEM images. In Fig. 5c and d, WC@C particles consist of a core of WC (black area) and a shell of carbon (gray area) which confirms the metal WC particles are encapsulated by the carbon shell. Fig. 6 shows the high-resolution transmission electron microscope (HRTEM) images of WC particles. The d-spacings of the core structure are 0.125 and 0.145 nm, corresponding to WC (102) and (110) planes. No obvious lattice pattern of the carbon shell is observed which explains the absence of the graphitic structure in the XRD patterns. The EDS analysis (Fig. 6) for the shell indicates the single element of carbon while the elements of the core consist of W and C which also testifies the formation of core–shell structure.

Fig. 5 SEM (a and b) and TEM (c and d) images of WC@C particles.

Fig. 6 HRTEM image of WC@C particle and EDS patterns of (a) part a and (b) part b in the HRTEM image.

The possible mechanism for the synthesis process is schematized in Fig. 1. It is well known that resorcinol and formaldehyde can form a Bakelite type polymer.³⁵ In our study, initially a mixture was formed when adding resorcinol, formaldehyde and AMT in the aqueous solution under stirring. After the condensation polymerization a cross-linking polymer network was fabricated within microwave heating system where offered a hermetic and high-pressure environment. During this polymerization process, AMT immersed in the cross-linking network and was entrapped in the RF. Then the polymer composite was heated at high temperature under argon and hydrogen flow. The entrapped AMT was reduced and carburized to WC, and part of the polymer took the role as the carburized source to form WC, and the other contributed to the carbon-encapsulated shell. In this way, the uniformly dispersed WC@C particles were synthesized after carbonization.

The platinum particles were loaded on WC@C by the microwave heating polylol reduction method in alkaline media. Both TEM and HRTEM images in Fig. 7 show that Pt nanoparticles are well supported on WC@C. This designed structure is supposed to affect the stability of catalytic reaction due to the improved interaction between active component and support. The HRTEM image exhibits the lattice fringe spacing of Pt is about 0.139 nm, which corresponds to (220) plane. The histograms show that the particle size of Pt is about 5 nm in diameter, which agree with the calculation result of Debye–Scherrer equation in XRD.

Fig. 7 TEM and HRTEM images of Pt–WC@C catalyst, and histograms of the particle size distributions of Pt particles about the Pt–WC@C catalysts.

Fig. 8A gives the N₂ adsorption/desorption isotherm of the WC@C sample. With pressure increasing, the isotherm shows a sharp increase in the amount of adsorption followed by saturation, and there is a hysteresis between adsorption and desorption branches. This typical Type IV represents characteristics of mesoporous materials. The BET surface area of WC@C microspheres calculated from data at the lower N₂ pressures is about 270 m² g⁻¹. The mesoporous structure is also confirmed from the small-angle XRD pattern (Fig. 8B). The sample WC@C shows reflection clearly at 2θ = 1° which suggests the existence of ordered mesopores on surface of the sample. This porous structure was formed when RF networks were filled with hydrogen gas and then fully removed during the carbonization.

Fig. 8 (A) The nitrogen adsorption/desorption isotherm of WC@C microspheres. (B) Small-angle X-ray scattering pattern of WC@C particles.

Fig. 9 presents the cyclic voltammograms of commercial Pt/C, PtRu/C and Pt–WC@C in 0.5 M H₂SO₄ solution. All the catalysts clearly show the hydrogen adsorption and desorption peaks on platinum surface at the potentials from −0.2 V to 0.1 V and the Pt–O formation at the potential of 0.4–0.5 V in the backward sweep. These typical regions demonstrate the presence of polycrystalline Pt. It can be noticed that Pt–WC@C has higher current peak for the hydrogen adsorption/desorption than that of commercial Pt/C and PtRu/C. This is the evidence for explaining the spillover effects between Pt and WC. Spillover of hydrogen from a metal to an oxide or carbon surface is well studied for many catalytic reactions involve hydrogen because most metal catalysts consist of small metal particles supported on either high surface area oxides or carbon.^36,37 For example, the synergistic effect between WO₃ and Pt relies on formation of tungsten bronzes by the faradaic process of intercalation/de-intercalation of H atoms into WO₃, which is spillover of hydrogen, showing as following:

xH⁺ + xe⁻ + WO₃ → H_xWO₃ (3)

Fig. 9 Cyclic voltammograms of commercial Pt/C, PtRu/C and Pt–WC@C catalysts in 0.5 M H₂SO₄ electrolyte.

However, in our case, spillover of hydrogen happens from Pt to WC surfaces as well as carbon shell. The similar phenomenon was discussed in the research work about Pt/WC.^38–40 Not like oxide and carbon, the process and mechanism of H⁺ spillover from Pt to WC was not fully understood. An accepted explanation discussed that the much higher ECSA value for the Pt/WC catalyst could be accounted for by active participation of WC in electrochemical hydrogen oxidation together with Pt. Another synergistic effect between WC and Pt was also discussed. Since the WC can also absorb OH_ads, thereby finally relieve more Pt active sites by facilitating the oxidation of CO_ads and other intermediates which cover the Pt–WC@C interface. Thus, we can find a fine correlation between the activity of methanol oxidation and ECSA values of the catalysts.

The electrocatalytic properties for methanol electrooxidation of commercial Pt/C, PtRu/C and Pt–WC@C catalysts were compared by CVs measurement in 0.5 M H₂SO₄ + 0.5 M CH₃OH in Fig. 10. As shown in Fig. 10, Pt–WC@C shows higher peak current density compared with commercial Pt/C and PtRu/C for methanol oxidation. In the methanol electrooxidation, the first anodic peak is usually assigned to the oxidation of methanol molecules on the electrode surface, and the backward anodic peak is generally attributed to the continuous oxidation of incompletely oxidized carbonaceous intermediates accumulated on the catalyst surface during the forward scan, such as CO_ads, COOH_ads, and COH_ads.⁴¹ The ratio of the forward anodic peak current density (I_f) to the reverse anodic peak current density (I_b), i.e., I_f/I_b, can be used to describe the catalyst tolerance to carbonaceous species accumulation. The high I_f/I_b ratio indicates excellent oxidation of methanol during the reverse anodic scan and less accumulation of residues on the catalyst.⁴² The ratios of the forward anodic peak current density (I_f) to the reverse anodic peak current density (I_b), i.e., I_f/I_b on Pt–WC@C is much higher than that on commercial Pt/C and PtRu/C (see Table 1). It is meaningful for us to notice that the Pt–WC@C catalyst gives larger I_f/I_b value (1.48) than that on commercial Pt/C (0.88) and PtRu/C (1.11), which means that the Pt is better dispersed on the WC@C support can be much more tolerant to the intermediate carbon species and consequently facilitate the oxidization of methanol to carbon dioxide much more efficient. The above higher catalytic activity demonstrated by the Pt–WC@C catalyst can be ascribed to the larger ECSA and uniform small size Pt, especially the synergistic effect between WC and Pt.

Fig. 10 Cyclic voltammograms of commercial Pt/C, PtRu/C and Pt–WC@C catalysts in 0.5 M H₂SO₄ and 0.5 M CH₃OH electrolyte.

Table 1 Comparison of different parameters between Pt/C, PtRu/C and Pt–WC@C catalysts

Samples	If (A g^−1 Pt)	Ib (A g^−1 Pt)	If/Ib
Pt–WC@C	509.94	343.71	1.48
PtRu/C	389.17	351.85	1.11
Commercial Pt/C	263.76	299.76	0.88

---

In order to demonstrate the Pt particles can be better dispersed on the WC@C support, we compared the catalytic activities and TEM images of the commercial Pt/C and 20% Pt–WC@C catalysts (see Fig. 11). The higher catalytic activities displayed by the 20% Pt–WC@C catalyst can be ascribed to the better dispersed Pt on the novel support WC (see Fig. 11a and (b)). It is show that 20% Pt can be also better dispersed on the WC@C support.

Fig. 11 TEM images of commercial Pt/C (a), 20% Pt–WC@C (b), and (c) cyclic voltammograms of commercial Pt/C and 20% Pt–WC@C catalysts in 0.5 M H₂SO₄ and 0.5 M CH₃OH electrolyte.

The CO tolerance of the prepared electrocatalyst can be evaluated by the CO stripping experiment as shown in Fig. 12. Before the stripping experiment, electrodes were immersed in 0.5 M H₂SO₄ solution where was bubbled with CO for 30 min and then purged with high pure N₂ for 30 min. Compared with the other samples, no obvious CO absorption-stripping peak can be observed on the support WC@C (inset in Fig. 12). One possible reason is that CO is relatively weakly bonded on WC@C.⁴³ The experimental CO uptake of WC samples was around 7 mmol g⁻¹, which been several orders of magnitude less than that of Pt metal.⁴⁴ CO adsorption experiments of water and CO on C/W (111) show that the presence of surface hydroxyls hinders the adsorption of CO. Thus, even after coadsorbing water and CO on the WC@C surface, only trace amounts of CO can be oxidized, which is not detectable. Compared with commercial Pt/C, PtRu/C and Pt–WC@C catalysts for CO oxidation, the Pt–WC@C could have the higher activity for CO oxidation is the evidence of promoting effect of WC in terms of H₂O dissociation and CO oxidation,^43,45 implying that the poisoning-tolerance and performance of the DMFC could be improved in the involvement of WC.

Fig. 12 The CO stripping data on WC@C (inset), commercial Pt/C, PtRu/C and Pt–WC@C catalysts in 0.5 M H₂SO₄ aqueous solution.

To evaluate the stability of Pt–WC@C catalyst for methanol oxidation, the chronoamperometry test was recorded and compared with commercial Pt/C and PtRu/C catalyst as shown in Fig. 13. In the initial period, the potentiostatic current decreases rapidly for both two catalysts, perhaps due to charge with the electrical double layer. After 1000 s, the current densities of the three catalysts are stable, but the Pt–WC@C maintains at a higher steady state than the commercial Pt/C, PtRu/C because WC can activate water and form surface hydroxyls easily, and hence enhance the CO tolerance of the catalysts. Moreover, the carbon shell could extend the function of WC even in the long cycles of reaction as it can keep the activity of WC from being oxidized.

Fig. 13 Chronoamperometry curves for commercial Pt/C, PtRu/C and Pt–WC@C catalysts in 0.5 M CH₃OH and 0.5 M H₂SO₄ aqueous solution at potential of 0.6 V.

Concerning the further work, the DMFC performances of prepared electrocatalysts will be evaluated by confirming the half-cell results.

4. Conclusion

In this study, the WC@C support was prepared by microwave-assisted synthesis method with using RF as in situ carbon source. The electro-oxidation of methanol and CO stripping test show that Pt–WC@C catalyst has the best performance for methanol oxidation and CO tolerance compared with those of commercial Pt/C and PtRu/C catalysts. It is that WC is encapsulated by porous carbon shell structure. This porous carbon shell structure can suppress the agglomeration of catalysts and provide more interfaces between different components. More important, this carbon shell structure retards the oxidation of WC surface and prohibits tungsten oxide from occupying the active sites on the surfaces of the Pt–WC@C catalysts. Therefore, more active sites remain and the synergistic effect of Pt and WC is improved, finally results in the enhanced activity and stability.

Acknowledgements

This research was funded by International Science & Technology Cooperation Program of China (2010DFB63680) and National Natural Science Foundation of China (21376220), partially supported by Zhejiang Provincial Natural Science Foundation of China (LY16B060009) and Science and Technology Key Project of Fujian Province (2014H0038).

References

1. F. Maillard, E. Peyrelade and Y. Soldo-Olivier, Electrochim. Acta, 2007, 52, 1958–1967.
2. M. A. J. Cropper, S. Geiger and D. M. Jollie, J. Power Sources, 2004, 131, 57–61.
3. J. H. Choi, K. J. Jeong, Y. J. Dong, J. H. Han, T. H. Lim and J. S. Lee, J. Power Sources, 2006, 163, 71–75.
4. J. Prabhuram, T. S. Zhao, Z. X. Liang and R. A. Chen, Electrochim. Acta, 2007, 52, 2649–2656.
5. R. Ganesan, D. J. Ham and J. S. Lee, Electrochem. Commun., 2007, 9, 2576–2579.
6. K. Chandrasekaran, J. C. Wass and J. O. Boclcris, J. Electrochem. Soc., 1990, 137, 518–524.
7. E. Antolini, J. Solid State Electrochem., 2011, 15, 455–472.
8. D. C. Papageorgopoulos, M. P. de Heer, M. Keijzer, J. A. Z. Pieterse and F. A. de Bruijn, J. Electrochem. Soc., 2004, 151, A763–A768.
9. J. Hosseini and A. Bodaghi, J. Solid State Electrochem., 2011, 15, 795–800.
10. K. Drew, G. Girishkumar, K. Vinodgopal and P. V. Kamat, J. Phys. Chem. Lett., 2005, 109, 11851–11857.
11. J. M. Skowronski and A. Wazny, J. Solid State Electrochem., 2005, 9, 890–899.
12. J. Prabhuram, T. S. Zhao, Z. X. Liang and R. Chen, Electrochim. Acta, 2007, 52, 2649–2656.
13. S. J. Liu, C. H. Huang, C. K. Huang and W. S. Hwang, Electrochem. Commun., 2009, 11, 1792–1795.
14. C. A. Ma, N. Brandon and G. H. Li, J. Phys. Chem. C, 2007, 111, 9504–9508.
15. R. B. Levy and M. Boudart, Science, 1973, 181, 547–549.
16. C. A. Ma, J. F. Sheng, N. Brandon, C. Zhang and G. H. Li, Int. J. Hydrogen Energy, 2007, 32, 2824–2829.
17. S. Shanmugam, D. S. Jacob and A. Gedanken, J. Phys. Chem. B, 2005, 109, 19056–19059.
18. M. B. Zellner and J. G. Chen, Catal. Today, 2005, 99, 299–307.
19. G. S. Raman, J. H. Dong and S. L. Jae, Electrochem. Commun., 2007, 9, 2576–2579.
20. M. K. Jeon, K. R. Lee, W. S. Lee, H. Daimonb, A. Nakaharab and S. IhlWoo, J. Power Sources, 2008, 185, 927–931.
21. D. J. Ham, Y. K. Kim, S. H. Han and J. S. Lee, Catal. Today, 2008, 132, 117–122.
22. K. M. Andersson and L. Bergströom, Int. J. Refract. Met. Hard Mater., 2000, 18, 121–129.
23. M. D. Obradovic, B. M. Babic, A. Kowal, V. V. Panic and S. L. Gojkovic, J. Serb. Chem. Soc., 2008, 73, 1197–1209.
24. W. J. Kang, H. B. Li, Y. Yan, P. P. Xiao, L. L. Zhu, K. L. Tang, Y. C. Zhu and Y. T. Qian, J. Phys. Chem. C, 2011, 115, 6250–6256.
25. Y. J. Chen, G. Xiao, T. S. Wang, Q. Y. Ouyang, L. H. Qi, Y. Ma, P. Gao, C. L Zhu, M. S. Cao and H. B. Jin, J. Phys. Chem. C, 2011, 115, 13603–13608.
26. M. H. Wang, X. Q. Wang, Q. Yue, Y. Zhang, C. Wang, J. Chen, H. Q. Cai, H. L. Lu, A. A. Elzatahry, D. Y. Zhao and Y. H. Deng, Chem. Mater., 2014, 26, 3316–3321.
27. D. H. Deng, L. Yu, X. Q. Chen, G. X. Wang, L. Jin, X. L. Pan, J. Deng, G. Q. Sun and X. H. Bao, Angew. Chem., 2013, 125, 389–393.
28. J. M. Lee, S. B. Han and J. Y. Kim, Carbon, 2010, 48, 2290–2296.
29. C. B. Xu, M. Q. Shi, L. Z. Kang and C. A. Ma, Mater. Lett., 2013, 91, 183–186.
30. K. Lee, K. Ishihara, S. Mitsushima, N. Kamiya and K. Ota, Electrochim. Acta, 2004, 49, 3479–3485.
31. J. D. Voorhies, J. Electrochem. Soc., 1972, 119, 219–222.
32. M. H. Ghandehari, J. Electrochem. Soc., 1980, 127, 2144–2147.
33. D. G. McCulloch, S. Prawer and A. Hoffman, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 5905–5917.
34. A. Pozio, M. D. Francesco, A. Cemmi, F. Cardelllini and L. Giorgi, J. Power Sources, 2002, 105, 13–19.
35. S. A. Al-Muhtaseb and J. A. Ritter, Adv. Mater., 2003, 15, 101–114.
36. W. C. Conner and J. L. Falconer, Chem. Rev., 1995, 95, 759–788.
37. J. G. Santiesteban, D. C. Calabro, W. S. Borghard, C. D. Chang, J. C. Vartuli, Y. P. Tsao, M. A. Natal-Santiago and R. D. Bastian, J. Catal., 1999, 183, 314–322.
38. R. Ganesan, D. J. Ham and J. S. Lee, Electrochem. Commun., 2007, 9, 2576–2579.
39. M. K. Jeon, H. Daimon, K. R. Lee, A. Nakahara and S. I. Woo, Electrochem. Commun., 2007, 9, 2692–2695.
40. A. Hassan, V. A. Paganin and E. A. Ticianelli, Appl. Catal., B, 2015, 165, 611–619.
41. Y. X. Chen, A. Miki, S. Ye, H. Sakai and M. F. Osawa, J. Am. Chem. Soc., 2003, 125, 3680–3681.
42. J. M. Lee, S. B. Kim and Y. W. Lee, Appl. Catal., B, 2012, 111–112, 200–207.
43. H. H. Hwu, B. D. Polizzotti and J. G. Chen, J. Phys. Chem. B, 2001, 105, 10045–10053.
44. Y. Hara, N. Minami, H. Matsumoto and H. Itagaki, Appl. Catal., A, 2007, 332, 289–296.
45. Z. J. Mellinger, E. C. Weigert, A. L. Stottlemyer and J. G. Chen, Electrochem. Solid-State Lett., 2008, 11, 63–67.

---