PtRu/Ti anodes with varying Pt : Ru ratio prepared by electrodeposition for the direct methanol fuel cell

PAPER www.rsc.org/pccp Physical Chemistry Chemical Physics PtRu/Ti anodes with varying Pt : Ru ratio prepared by electrodeposition for the direct methanol fuel cell Zhi-Gang Shao,* ab Fuyun Zhu,w a Wen-Feng Lin,* a Paul A. Christensen a and Huamin Zhang b Received 6th April 2006, Accepted 10th May 2006 First published as an Advance Article on the web 23rd May 2006 DOI: 10.1039/b604939g PtRu/Ti anodes with varying Pt : Ru ratio were prepared by electrodeposition of a thin PtRu catalyst layer onto Ti mesh for a direct methanol fuel cell (DMFC). The morphology and structure of the catalyst layers were analyzed by SEM, EDX and XRD. The catalyst coating layer shows an alloy character. The relative activities of the PtRu/Ti electrodes were assessed and compared in half cell and single DMFC experiments. The results show that these electrodes are very active for the methanol oxidation and that the optimum Ru surface coverage was ca. 9 at.% for DMFC operating at 20 1C and 11 at.% at 60 1C. The PtRu/Ti anode shows a performance comparable to that of the conventional carbon-based anode in a DMFC operating with 0.25 M or 0.5 M methanol solution and atmosphere oxygen gas at 90 1C. 1. Introduction The liquid feed direct methanol fuel cell (DMFC) is considered as a potential power source for stationary and transportation applications because of characteristics such as simple construction, easy operation, liquid fuel and high efficiency. 1,2 However, obstacles still prevent their widespread commercial applications, 3 5 e.g. low activity of methanol electrooxidation catalysts, methanol crossover from the anode to the cathode, carbon dioxide gas management and water management. 6 Unlike other fuel cells, the liquid feed DMFC suffers from mass transport limitations predominantly at the anode due to the low diffusion coefficient of methanol in water and the release of carbon dioxide gas bubble. 5 A methanol concentration gradient exists within the thickness of the catalyst layer, this results in poor utilization of the catalyst. 7 In addition, a methanol diffusion limiting current is found in DMFC with the low concentration methanol solution, 8 which inhibits getting high power density. Therefore, the conventional anode structure based on the gas diffusion electrodes employed in proton exchange membrane fuel cells is not ideal for the transport of the methanol and carbon dioxide. Lately, a novel anode structure prepared by thermal decomposition of the corresponding metal chloride solutions on titanium substrate has been developed. 9,10 In this structure, a fine Ti mesh which benefits the methanol transfer and carbon dioxide removal is used as a substitute for the conventional gas diffusion layer in the anode. This paper further develops the Ti mesh anode a School of Chemical Engineering and Advanced Materials, Bedson Building, University of Newcastle upon Tyne, Newcastle upon Tyne, UK NE1 7RU. E-mail: wenfeng.lin@ncl.ac.uk b Dalian Institute of Chemical Physics, Chinese Academy of Sciences, China. E-mail: zhgshao@dicp.ac.cn w Present address: Transformer Examining and Repairing Department of Nantong Power Supply Company, No. 90, South Yuelong Road, Nantong, Jiangsu, P.R. China 226006, Tel: 0086-513-5163619. concept in terms of the activity of anodes prepared by the electrochemical deposition of PtRu on Ti mesh. It is well known that the ratio of Pt to Ru has a dramatic effect on the performance of the catalyst for methanol oxidation. However, there is one controversial point of the optimum Pt Ru atomic relation. Iwasita et al. 11 showed that the activity of smooth Pt Ru electrodes with Ru contents between 10 and 40 at.% was similar. Gasteiger et al. 12 proposed an optimum surface composition which was near to 10 at.% of Ru at room temperature based on the need of having three sites on Pt for methanol adsorption. Chu and Gilman 13 reported a maximum in activity for 50 at.% Ru in the bulk between 25 and 65 1C. Some results indicate that the optimum surface composition was ca. 20at.%Ru, 14 40 at.% Ru 15 and 50 at.% 16,17 for unsupported Pt Ru alloy. However, the Pt Ru catalyst is usually evaluated by the three-electrode cell in the literature work and few works have been done to evaluate the Pt Ru catalyst in practical DMFC. In this paper, Pt Ru catalysts are studied not only by a three-electrode cell, but also by single DMFC experiments. In addition, although much work has been performed on Pt Ru catalysts for methanol oxidation, including unsupported Pt Ru alloys, 16,18,19 Pt Ru supported on carbon, 20 22 very few have investigated the electrodes with the catalyst layer coated on Ti mesh by electrodeposition for methanol oxidation in DMFC. In this work, PtRu/Ti anodes with varying Pt : Ru ratio are prepared by electrodeposition PtRu catalyst layer onto Ti mesh. The morphology and structure of the catalyst layers were analyzed by SEM, EDX and XRD. The relative activities of the PtRu/Ti electrodes were assessed and compared in half-cell and single DMFC experiments and compared with a conventional carbon-based anode in a DMFC. 2. Experimental 2.1 Anode preparation The anode was prepared by direct electrodeposition of PtRu catalyst on titanium mesh. The Ti mesh (2.5 2.5 cm 2, Delker 2720 Phys.Chem.Chem.Phys., 2006, 8, 2720 2726 This journal is c the Owner Societies 2006

corporation, 99.9% Ti) was immersed in 10% oxalic acid at 80 1C for 1 h and then rinsed with Millipore water (18 MO cm). Electro-deposition of Pt and Ru onto the Ti mesh was performed in a conventional three-electrode cell at room temperature at a constant potential of 100 mv vs. Ag/AgCl. The reference electrode was a silver/silver chloride (Ag/AgCl) electrode in saturated KCl, its potential vs. RHE is 0.199 V at room temperature. The counter electrode was a 2.5 2.5 cm 2 Pt foil (Good fellow, 99.9%). A solution containing 2 mm H 2 PtCl 6,10mMH 2 SO 4 and appropriate quantity RuCl 3 was used as the electrolyte, the quantity of RuCl 3 was added according to the required precursor Pt Ru atomic ratio. The desired catalyst loading was achieved by controlling the electrodeposition time. The loading of PtRu catalyst in the electrode is about 2 mg cm 2 for half-cell test. For single DMFC testing, just one side of the electrode that faces the membrane is active, so the loading of PtRu catalyst in the electrode was calculated as about 4 mg cm 2. The electrodes so prepared are denoted as PtRu/Ti. For comparison, conventional porous carbon anodes were prepared according to the procedure reported by Wilson and co-workers. 23 Briefly: a thin, microporous layer of uncatalysed (Ketjen black 600) carbon, bound with 10 wt% Nafion s from a solution of 5 wt% Nafion s dissolved in a mixture of water and lower aliphatic alcohols (Aldrich) was spread over a 0.35 mm thick, teflonised (20%) carbon cloth support (E-Tek, type A ). The catalysed ink, consisting of unsupported Pt Ru black (atomic ratio Pt : Ru ¼ 1 : 1, Johnson Matthey), 25 wt% Nafion s and iso-propanol was painted onto the Teflon blanks. The anode catalyzed layer was then transferred onto the membrane by the decal method. 2.2 Half-cell tests Half-cell performance testing was performed in a conventional three-electrode cell. The reference electrode was a Ag/AgCl electrode in saturated KCl, which was connected to the cell by a glass capillary. In this paper, unless otherwise specified, all electrode potentials are quoted versus the Ag/AgCl reference electrode. The counter electrode was a 2.5 cm 2.5 cm Pt foil. The working electrode was a 1 cm 0.5 cm Ti mesh coated with PtRu catalyst. A Voltlab PGZ301 Dynamic-EIS was used to control potential/current. Firstly, cyclic voltammetry was carried out on all of the anodes in N 2 -deaired 0.5 M H 2 SO 4 with and without added 0.5 M CH 3 OH at room temperature (20 1C) and 60 1C. The electrodes were cycled between 0.2 and 0.7 V at a scan rate of 50 mv s 1 until the stable state of the electrode was reached. And then, linear sweep voltammetry (LSV) was performed for the anodes in N 2 -deaired 0.5 M H 2 SO 4 in the presence of 0.5 M CH 3 OH between 0.2 V and 0.5 V at a scan rate of 1mVs 1. 2.3 Preparation of membrane electrode assembly (MEA) The cathodes were also prepared according to the procedure used in references. 24 A Teflonised (20%) Toray TGPH-090 carbon paper with a thin microporous layer of uncatalysed (XC-72R) carbon, bound with 50 wt% PTFE was used as the diffusion layer and A mixture including catalysts (50% Pt/C, Johnson Matthey), 30 wt% Nafion in iso-propanol was spread on the diffusion layer as the catalyst layer. The PtRu/Ti electrodes were used as the anodes after being sprayed on 5 wt% Nafion solution (EW 1100), equivalent to a dry Nafion loading of about 0.6 mg cm 2. The cathodes and anodes were placed on either side of a pre-treated Nafion s 117 membrane (Aldrich). This pre-treatment involved heating the membrane at 80 1C for 2 h in 3 vol% H 2 O 2 and 2 h in 0.5 M H 2 SO 4 before washing in boiling Millipore water for 2 h. The assembly was hot-pressed at 70 kg cm 2 for 3 min at 135 1C. 2.4 Single cell tests For single cell testing, the geometrical area of all the anodes and cathodes was ca. 6.25 cm 2 and the catalyst loadings of all the anodes and cathodes were 4 and 3.5 mg cm 2, respectively. Details of the testing system of the single fuel cell are described fully elsewhere. 25 A dilute methanol solution was fed to the anode inlet at a flow rate of 10 ml min 1 by a peristaltic pump without pre-heating and back-pressure. Atmospheric oxygen gas was fed to the cathode inlet at room temperature and a flow rate of 200 ml min 1 without pre-heating and humidification. 2.5 Physico-chemical characterization A Jeol JSM-5300LV scanning electron microscope (SEM) was employed to investigate the structures and morphology of the mesh electrodes. Energy dispersive X-ray (EDX) spectroscopy was also employed for the characterisation of the electrodes by a RONTEC spectrometer (made in Germany). X-Ray diffraction (XRD) patterns were obtained on a Philips Xpert Pro diffractometer operating at 40 kv accelerating voltage and 40 ma of beam current and using Cu Ka 1 radiation. Diffraction peaks were attributed following the Joint Committee of Powder Diffraction Standards (JCPDS) cards. 3. Results and discussion 3.1 Electrode preparation and characterization The relative content of surface Ru sites in the PtRu/Ti as a function of Ru/(Pt þ Ru) atomic ratio in precursor solution is reported in Fig. 1. The Ru content of PtRu/Ti surface is measured by EDX. From Fig. 1, a straight line is observed. It means that, in our experiment, the Ru content of PtRu/Ti surface is smaller than that in precursor solution and they have a linear relation. This may be a result of the fact that the electrodeposition rate of Ru on the surface of PtRu/Ti electrode is directly proportional to the concentration of Ru in the precursor solution. This result implies that the electrodeposition method is convenient to control the catalyst loading and composition of the Pt : Ru atomic ratio. Some alloy surfaces show strong segregation effects, which also depend on the ratio between the mixed metals. The phase diagram of the PtRu system 26 shows that an fcc phase is formed at Pt concentrations above 40%, an hcp phase is formed at Pt concentrations below 20% and a two-phase region in-between. The Pt surface enrichment of the Pt Ru alloy catalyst prepared by electrodeposition is reported by Richarz et al. 27 It may result from This journal is c the Owner Societies 2006 Phys. Chem. Chem. Phys., 2006, 8, 2720 2726 2721

Fig. 1 Surface concentration of Ru on PtRu/Ti vs. concentration in precursor solution. the sufficient surface mobility of the metal ions during deposition. Arico et al. 15 also found a Pt surface enrichment for the Pt Ru alloy with low Ru content in the bulk. It is consistent with our results. Fig. 2 shows SEM images of the PtRu/Ti electrodes with different Pt : Ru atomic ratio. It is seen that the morphology of catalyst surface is different with the change of Ru content in the catalyst. The coating layer with low Ru content (4.3 at.%) is porous and there are many catalyst clusters in the surface. The coating becomes smooth when the Ru content in the coating layer increases. This effect may be attributed to the forming of PtRu alloy with the increase of Ru content, which weakens the self-agglomeration of Pt particles. A typical EDX spectrum is shown in Fig. 3a and the atomic ratio of Ru/(Pt þ Ru) is about 30%. The characteristic peak of Ti can not be found in Fig. 3a. It indicates that Ti mesh is almost fully covered by the PtRu catalyst layer. A crosssectional view of the anode is shown in Fig. 3b and it can be seen that the PtRu layer is ca. 3mm thick. The XRD results of plain Ti mesh, commercial Pt Ru black and the PtRu/Ti is shown in Fig. 4. From the XRD pattern of commercial Pt Ru black in Fig. 4, it is clearly observed that typical peaks of the bimetallic Pt Ru alloy phase are at 40.7, 46.6, 69.1 and 82.71 in 2y and typical peaks of the Ti substrate is also clearly visible. For the PtRu/Ti electrodes, the peaks of the Pt Ru alloy phase are at 40.2, 46.4, 68.3 and 82.41 in 2y and they are at lower 2y angles than that of the commercial Pt Ru black. It implies that the former has a larger lattice parameter and less Ru content than the latter according to Vegard s law. 28 The mean particle size of the agglomerates was evaluated from the line broadening of the (2 2 0) peaks by using the Scherrer formula. It is estimated to be about 2.6 and 3.3 nm for the commercial Pt Ru black and the electrodeposited catalyst, respectively. The particle size of the electrodeposited catalyst is slightly bigger than that of the commercial Pt Ru black. It may be explained by the fact that the conditions of catalyst preparation are different. Fig. 2 SEM micrographs of the PtRu/Ti anodes with: (a) 4.3 at.% Ru; (b) 11 at.% Ru; (c) 18.7 at.% Ru; (d) 30 at.% Ru on the catalyst layer surface. 2722 Phys.Chem.Chem.Phys., 2006, 8, 2720 2726 This journal is c the Owner Societies 2006

Fig. 5 Steady state cyclic voltammograms of Ti mesh and PtRu/Ti (30 at.% Ru) in 0.5 M H 2 SO 4 þ 0.5 M CH 3 OH, scan rate 50 mv s 1, room temperature. Fig. 3 (a) EDX spectra analysis of the PtRu/Ti (30 at.% Ru); (b) scanning electron micrographs of the cross-section of PtRu/Ti anode (30 at.% Ru): I: PtRu catalyst layer; II: Ti mesh. 3.2 Half-cell tests Fig. 5 compares the CVs of the PtRu/Ti and Ti mesh in the presence of 0.5 M H 2 SO 4 þ 0.5 M CH 3 OH and it is clearly seen that titanium has little activity of methanol oxidation and the effect of the plain Ti mesh can be neglected. The onset potential of methanol oxidation is observed at around 200 mv for PtRu/Ti anode. Linear sweep voltammetries for the PtRu/Ti electrode with 30 at.% Ru in 0.5 M H 2 SO 4 þ 0.5 MCH 3 OH at 20 and 60 1C are shown in Fig. 6. From Fig. 6, it is seen that the activity of the methanol oxidation for PtRu/Ti electrode markedly increases with the operating temperature. This result is to be expected from the increase of methanol oxidation kinetics with temperature. Fig. 7 shows mass activities of methanol oxidation for the PtRu/Ti electrodes at constant potentials of 0.2, 0.3 and 0.4 versus Ag/AgCl during linear sweep voltammetry measurements as a function of Ru/(Ru þ Pt) atomic ratio at 20 1C. The atomic ratio of Ru and Ru þ Pt is obtained by EDX measurements. The potentials 0.2, 0.3 and 0.4 V were chosen because these covered the practical DMFC operating range. As may be seen from Fig. 7, the optimum Ru/(Pt þ Ru) atomic ratio at room temperature is 9 at.% with separately highest output current densities of 0.9, 6.8 and 18.0 ma mg 1 cm 2 at 0.2, 0.3 and 0.4 V, respectively. This feature is different with that reported by Arico et al., 15 they found that the optimum surface composition was ca. 40 at.% Ru for unsupported PtRu alloy with a 1 : 1 bulk composition as anode catalyst in DMFC. Gasteiger et al. 12 reported that an optimum surface composition was near 10 at.%. Chu and Gilman 13 reported a maximum in the activity for 50 at.% Ru in the bulk between 25 and 65 1C. This can be explained by the Fig. 4 XRD pattern of Ti mesh, the PtRu/Ti electrode and unsupported Pt Ru black (atomic ratio Pt : Ru ¼ 1 : 1) from Johnson Matthey. Fig. 6 Linear sweep voltammetries for the PtRu/Ti with 11 at.% Ru in 0.5 M H 2 SO 4 þ 0.5 M CH 3 OH, scan rate 1 mv s 1. This journal is c the Owner Societies 2006 Phys. Chem. Chem. Phys., 2006, 8, 2720 2726 2723

Fig. 7 Activity of the PtRu/Ti as a function of Ru/(Pt þ Ru) atomic ratio, taken from LSVs (scan rate ¼ 1mVs 1, PtRu loading around 2mgcm 2, 0.5 M H 2 SO 4 þ 0.5 M CH 3 OH) at 20 1C. The activities were calculated on the basis of the PtRu total loading. fact that the conditions of catalyst preparation and the electrolyte are different. According to bifunctional the mechanism of methanol oxidation, 13,17,29 31 Pt is the species responsible for dissociative dehydrogenation of methanol, while the Ru sites provide a source of oxygenated species for removal of carbon containing fragments. The optimum alloy surface composition would both maximize the adsorption of methanol through a large number of Pt adsorption sites as well as allow the oxidative removal of CO via hydroxide shuttling by Ru at low potentials. Gasteiger et al. 32 assume that the adsorption of methanol requires an ensemble of three adjacent Pt atoms, the number of 3-fold Pt sites available for methanol adsorption is larger on the 10 at.% than on the 50 at.% Ru surface and on the 10 at.% Ru surface most of the Pt ensembles are adjacent to a Ru atom to allow the supply of oxygen-like species. Therefore, 10 at.% Ru surface of Pt Ru alloy displays the highest activity for methanol oxidation. The mass activities for the PtRu/Ti electrodes at 60 1C observed at constant potentials of 0.2, 0.3 and 0.4 V have been plotted versus the Ru at.% content in the catalysts surface (Fig. 8). As shown in Fig. 8, it is observed that the change in methanol oxidation behavior is remarkable and the electrode Fig. 8 Activity of the PtRu/Ti as a function of Ru/(PtþRu) atomic ratio, taken from LSVs (scan rate ¼ 1mVs 1, PtRu loading around 2 mg cm 2, 0.5 M H 2 SO 4 þ 0.5 M CH 3 OH) at 60 1C. The activities were calculated on the basis of the PtRu total loading. Fig. 9 Polarization curves for DMFCs with different PtRu/Ti anodes at 20 1C, with atmospheric oxygen gas feed at the cathode and 1 M CH 3 OH feed at the anode. with 11 at.% Ru content displays a best performance and the electrode with 30 at.% Ru content shows a comparable performance to that with 11 at.% Ru content. It is also interesting to note that the optimum range of Ru content with good activity shifts from 4 11 at.% at 20 1C to 11 30 at.% at 60 1C. This may be interpreted on the basis of bifunctional mechanism of methanol oxidation. The chemisorption and subsequent dehydrogenation of methanol on the Ru sites is significantly less favored than on Pt sites, but is strongly activated by temperature. 33,34 Therefore a Ru rich sample might be expected to perform better at higher operating temperature. 3.3 Single cell performance Fig. 9 shows a comparison of the performance of DMFCs at 20 1C with the different anodes using aqueous 1 M methanol as the anode feed and atmospheric oxygen gas as the cathode feed. As shown in Fig. 9, the better performance is obtained using the anode with 11 at.% Ru content than that with 30 at.% Ru content. This result is consistent with that obtained in the above-mentioned half cell tests. The effect on the performance of DMFCs of raising the temperature to 60 1C is demonstrated in Fig. 10. It is seen that the cell with 30 at.% Ru anode displays a comparable performance to that with 11 at.% Ru anode. It indicates that there are significant temperature effects on the activity of the 30 at.% Ru anode. This result suggests that the bifunctional mechanism appears to govern the methanol oxidation on PtRu/Ti anodes in DMFC. Fig. 11 displays a comparison of the performance of the DMFCs at 90 1C with different anode structures using aqueous 0.25 M or 0.5 M methanol as the anode feed and atmospheric oxygen gas as the cathode feed. As shown in Fig. 11, at lower current densities, the cell with the PtRu/Ti anode (30 at.% Ru) gave a competitive performance to that employing the conventional anode (PtRu black, Pt : Ru ¼ 1 : 1), i.e. 510 mv at 100 ma cm 2 compared to 525 mv, respectively, in 0.5 M methanol solution. However, at the higher current densities, the cell with the novel PtRu/Ti anode gave a better performance than the cell with a 2724 Phys.Chem.Chem.Phys., 2006, 8, 2720 2726 This journal is c the Owner Societies 2006

4. Conclusions Fig. 10 Polarization curves for DMFCs with different PtRu/Ti anodes at 60 1C, with atmospheric oxygen gas feed at the cathode and 1 M CH 3 OH feed at the anode. conventional carbon-based anode, 300 mv at 200 ma cm 2 compared to 0 mv, in 0.25 M methanol solution. The diffusion limiting current densities observed with the DMFC employing the conventional anode were ca. 130 ma cm 2 and 250 ma cm 2 in 0.25 M and 0.5 M methanol solution, respectively. However, under the same conditions, the diffusion limiting current densities can not be found with the DMFC using the novel PtRu/Ti anode. In addition to the improvements in performance detailed above, the novel PtRu/Ti mesh electrodes have many advantages over conventional anodes, including: (i) simplicity; (ii) low cost; (iii) flexibility in terms of shape; (iv) collecting current easily. Although there is much less surface area contacting with the membrane in PtRu/Ti anode structure, it still shows a comparable and even better performance than the conventional anode, which indicates that a high catalyst utilisation is achieved in the PtRu/Ti anode. Fig. 11 Comparison of the performance of the DMFC with different anode structures: (a) 0.25 M methanol, conventional anode; (b) 0.25 M methanol, PtRu/Ti anode; (c) 0.5 M methanol, conventional anode; (d) 0.5 M methanol, PtRu/Ti anode. The data were obtained at 90 1C with atmospheric oxygen gas feed at the cathode (200 ml min 1 ) and with dilute methanol solution feed at the anode (10 ml min 1 ) without back-pressure. PtRu/Ti anodes with varying Pt : Ru ratio have been prepared by electrodeposition of the catalyst direct on Ti mesh for the liquid feed DMFC. The electrodeposition method is found to be convenient for controlling the catalyst loading and composition of Pt : Ru atomic ratio. The catalysts show an alloy character as determined by XRD. It is found that the surface composition of PtRu/Ti anodes has a significant influence on the performance of DMFCs. The optimum Ru surface coverage was ca. 9 at.% for DMFC operating at 20 1C and 11% at 60 1C. However, the anode with 30 at.% Ru displays a comparable performance to that with 11 at.% Ru at 60 1C. Compared to a conventional porous carbon-based anode, the PtRu/Ti anode shows an enhanced performance in a DMFC operating with a low concentration of methanol. Acknowledgements The authors would like to acknowledge the funding from EU Framework 5 craft and the Knowledge Innovation Program of the Chinese Academy of Sciences (Grant No.DICPR200502). References 1 X. Ren, M. S. Wilson and S. Gottesfeld, J. Electrochem. Soc., 1996, 143, L12 L15. 2 D. Kim, E. A. Cho, S.-A. Hong, I.-H. Oh and H. Y. Ha, J. Power Sources, 2004, 130, 172 177. 3 A. S. Arico, V. Baglio, E. Modica, A. Di Blasi and V. Antonucci, Electrochem. Commun., 2004, 6, 164 169. 4 Z.-G. Shao, X. Wang and I.-M. Hsing, J. Membr. Sci., 2002, 210, 147 153. 5 H. Yang, T. S. Zhao and Q. Ye, J. Power Sources, 2005, 139, 79 90. 6 G. Q. Lu, F. Q. Liu and C.-Y. Wang, Electrochem. Solid-State Lett., 2005, 8, A1 A4. 7 M. Wohr, S. R. Narayanan and G. Halpert, Abstract 797, The Electrochemical Society Meeting Abstracts, Vol. 96-2, San Antonio, TX, Oct 6 11, 1996. 8 K. Scott, W. M. Taama, S. Kramer, P. Argyropoulos and K. Sundmacher, Electrochim. Acta, 1999, 45, 945 957. 9 C. Lim, K. Scott, R. G. Allen and S. Roy, J. Appl. Electrochem., 2004, 34, 929 933. 10 Z.-G. Shao, W.-F. Lin, F. Zhu, P. A. Christensen, M. Li and H. Zhang, Electrochem. Commun., 2006, 8, 5 8. 11 T. Iwasita, H. Hoster, A. John-Anacker, W. F. Lin and W. Vielstich, Langmuir, 2000, 16, 522 529. 12 H. A. Gasteiger, N. Markovic, P. N. Ross, Jr and E. J. Cairns, J. Electrochem. Soc., 1994, 141, 1795 803. 13 D. Chu and S. Gilman, J. Electrochem. Soc., 1996, 143, 1685 1690. 14 B. A. L. de Mishima, H. T. Mishima and G. Castro, Electrochim. Acta, 1995, 40, 2491 2500. 15 A. S. Arico, P. L. Antonucci, E. Modica, V. Baglio, H. Kim and V. Antonucci, Electrochim. Acta, 2002, 47, 3723 3732. 16 M.-S. Loffler, H. Natter, R. Hempelmann and K. Wippermann, Electrochim. Acta, 2003, 48, 3047 3051. 17 M. Watanabe and S. Motoo, J. Electroanal. Chem., 1975, 60, 275 283. 18 Z. Jusys, J. Kaiser and R. J. Behm, Electrochim. Acta, 2002, 47, 3693 3706. 19 J. Solla-Gullon, F. J. Vidal-Iglesias, V. Montiel and A. Aldaz, Electrochim. Acta, 2004, 49, 5079 5088. 20 J. W. Guo, T. S. Zhao, J. Prabhuram, R. Chen and C. W. Wong, Electrochim. Acta, 2005, 51, 754 763. 21 L. Jiang, G. Sun, X. Zhao, Z. Zhou, S. Yan, S. Tang, G. Wang, B. Zhou and Q. Xin, Electrochim. Acta, 2005, 50, 2371 2376. 22 Y. Takasu, T. Kawaguchi, W. Sugimoto and Y. Murakami, Electrochim. Acta, 2003, 48, 3861 3868. 23 M. S. Wilson and S. Gottesfeld, J. Appl. Electrochem., 1992, 22, 1 7. This journal is c the Owner Societies 2006 Phys. Chem. Chem. Phys., 2006, 8, 2720 2726 2725

24 C. Lim and C. Y. Wang, J. Power Sources, 2003, 113, 145 150. 25 R. W. Reeve, P. A. Christensen, A. J. Dickinson, A. Hamnett and K. Scott, Electrochim. Acta, 2000, 45, 4237 4250. 26 J. M. Hutchinson, Platinum Met. Rev., 1972, 16, 88. 27 F. Richarz, B. Wohlmann, U. Vogel, H. Hoffschulz and K. Wandelt, Surf. Sci., 1995, 335, 361 371. 28 E. Antolini, Mater. Chem. Phys., 2003, 78, 563 573. 29 V. S. Bagotzky and Y. B. Vassilyev, Electrochim. Acta, 1967, 12, 1323 1343. 30 K. A. Friedrich, K.-P. Geyzers, U. Linke, U. Stimming and J. Stumper, J. Electroanal. Chem., 1996, 402, 123 128. 31 T. Frelink, W. Visscher and J. A. R. v. Veen, Langmuir, 1996, 12, 3702 3708. 32 H. A. Gasteiger, N. Markovic, P. N. Ross and E. J. Chairns, J. Phys. Chem., 1993, 97, 12020 12029. 33 S. Wasmus and A. Kuver, J. Electroanal. Chem., 1999, 461, 14 31. 34 K. A. Friedrich, K. P. Geyzers, A. J. Dickinson and U. Stimming, J. Electroanal. Chem., 2002, 524 525, 261 272. 2726 Phys.Chem.Chem.Phys., 2006, 8, 2720 2726 This journal is c the Owner Societies 2006