Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

B876

0013-4651/2007/154共8兲/B876/7/$20.00 © The Electrochemical Society

Electro-oxidation of Ethanol at Sputter-Deposited Platinum–Tin Catalysts Pascale Bommersbach,* Mohamed Mohamedi,**,z and Daniel Guay** Institut National de la Recherche Scientifique, Centre Energie, Matériaux et Télécommunications, University of Quebec, Varennes, Quebec J3X 1S2, Canada The electro-oxidation of ethanol was studied in Pt/Sn catalysts, prepared by sputter-deposition. The performances of four different Pt/Sn catalysts in the proportions of 95/5, 80/20, 50/50, and 25/75 were investigated using cyclic voltammetry, chronoamperometry, and electrochemical impedance spectroscopy. Pt50 /Sn50 appears to be the best electrocatalyst among the four studied alloys; the current density of the ethanol oxidation peak is the highest for Pt50 /Sn50, and this catalyst also reveals the lowest chargetransfer resistance, which tends to decrease in spite of the application of higher potentials. © 2007 The Electrochemical Society. 关DOI: 10.1149/1.2747534兴 All rights reserved. Manuscript submitted October 25, 2006; revised manuscript received April 16, 2007. Available electronically June 22, 2007.

In the last two decades, the development of alternative power sources has become a crucial issue of research and development. One of the main goals in this quest is the reduction of emission of greenhouse gases like carbon dioxide 共CO2兲. With this in mind, fuel cells appear to be very promising as power supplies for automotive, portable, or stationary applications.1 While hydrogen as the fuel in a polymer electrolyte membrane fuel cell 共PEMFC兲 allows higher electric efficiency to be achieved, its production and storage are still problematic. The use of hydrogen carriers like alcohols in direct alcohol fuel cells 共DAFCs兲 appears advantageous for three main reasons: 共i兲 the use of liquid alcohols simplifies the problem of storage, 共ii兲 in a DAFC alcohol is not reformed into hydrogen gas but is oxidized directly, and 共iii兲 the theoretical mass-energy density of most alcohols is high. Methanol has been considered a strong contender in the realm of alternative fuels. However, methanol is a toxic compound to human and animal life. It is also a pollutant, so its extensive use is inconceivable because of environmental hazards 共methanol is miscible with water, which could also involve contamination of ground water in the case of a spillage兲. A good substitute is ethanol, which has a very positive environmental, health, and safety footprint with no major uncertainties or hazards. Ethanol is a hydrogen-rich liquid and it has more energy density 共8.0 kWh/kg兲 compared to methanol 共6.1 kWh/kg兲. Ethanol can be obtained in great quantity from biomass through a fermentation process from renewable resources like sugar cane, wheat, corn, or even straw. Biogenerated ethanol 共or bioethanol兲 is thus attractive because it will not change the natural balance of carbon dioxide in the atmosphere. This is in sharp contrast to the use of fossil fuels. The use of ethanol would also overcome both the storage and infrastructure challenge of hydrogen for fuel cell applications. In a fuel cell, the oxidation of any fuel requires the use of a catalyst in order to achieve the current densities required for commercially viable fuel cells, and platinum-based catalysts are some of the most efficient materials for the oxidation of small organic molecules. Methanol is a relatively inexpensive and reactive fuel for DAFCs, and its electrooxidation has been extensively studied for more than three decades.1-8 The complete electro-oxidation of ethanol in CO2 is not easy to achieve; it involves the release of 12 electrons per ethanol molecule, the cleavage of the C–C bond without forgetting the formation of CO intermediates that could poison the anode catalysts. A number of catalysts has thus been tested and Pt-based catalysts are some of the most efficient materials.9-34 Lamy and co-workers were the pioneers in ethanol electrooxidation studies.9-12,14 Their first investigations began before the

* Electrochemical Society Student Member. ** Electrochemical Society Active Member. z

E-mail: [email protected]

1990s. Thanks to their in situ spectroscopic methods 共infrared and UV/visible reflectance spectroscopy兲 and electrochemical studies, our understanding of ethanol oxidation mechanism has greatly improved. Many authors looked at Pt/Sn catalysts10,14-20,23-25 whose, superior performances in ethanol oxidation compared to pure Pt was attributed to the so-called bifunctional mechanism; Sn or its oxides could supply a surface with species containing oxygen to remove strongly adsorbed species like CO. These studies showed that PtSn catalysts somehow lower the onset potential of oxidation of ethanol compared to Pt electrodes. Lamy et al. tested performances of different Pt/Sn catalysts. Pt90:Sn10 obtained by electrodeposition9-12 revealed the best electrocatalyst properties, whereas the optimum composition in tin was in the range 10–20 atom % for catalysts prepared by coimpregnation-calcination-reduction.14 Spinacé et al. studied catalysts prepared by the alcohol reduction process. They found that Pt/Sn electrocatalysts with Pt/Sn molar ratios of 50:50 are more active than electrocatalysts with 75:25 or 25:75 molar ratios,17 in agreement with Xin and co-workers studies.18 Despite these few works, the optimum atomic composition of PtSn has not yet been established mainly because of the preparation method that has an effect on the results. To further examine this issue, the present investigation has been undertaken to examine in detail the effect of Sn content in the PtSn alloy catalysts synthesized by the sputtering deposition method. PtSn catalysts with different Pt/Sn atomic ratios and their electrochemical activity toward ethanol oxidation have been evaluated using cyclic voltammetry 共CV兲, chronoamperometry 共CA兲, and electrochemical impedance spectroscopy 共EIS兲. Experimental Electrocatalysts preparation.— The deposition was performed with a multisputter-target machine 共Anelva-350S-C兲. The chamber was evacuated to a base pressure of 10−4 Pa. The deposition was carried out at 300°C in the chamber filled with 99.999% pure argon gas to 10 Pa under a substrate rotation speed of 20 rpm. Sputterdeposited Pt and PtSn films were fabricated on a gold flag substrate 共geometric surface area was 0.39 cm2兲. Bulk chemical compositions of samples were determined by energy dispersive X-ray 共EDX兲 analysis; thereafter, the nomination of Ptx /Sny means a catalyst with an atomic composition of x% of Pt and y% of Sn, determined by EDX. Electrochemical measurement.— The electrochemical studies were carried out in a deaerated solution of 0.5 M H2SO4 + 0.5 M C2H5OH. A Pt coil and an Ag/AgCl were used as counter and reference electrodes, respectively. All potentials reported in this article are with respect to the relative hydrogen electrode 共RHE兲. Current-potential 共I-E兲 and chronoamperometric curves were measured by a potentiostat 共Autolab兲. Prior to electrochemical measure-

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

Figure 1. CVs recorded in 0.5 M H2SO4, scan rate = 50 mV s−1 at Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, and Pt25 /Sn75 catalysts.

B877

Figure 2. CV curves for ethanol oxidation at Pt catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH, scan rate = 50 mV s−1.

at which the ethanol starts to be oxidized兲 starts at about 0.5 V/RHE and the current density of the first oxidation peak is 0.44 mA cm−2. ments in ethanol-containing solution, the surface of the working electrode was cleaned electrochemically by potential cycling in 0.5 M H2SO4. All the electrochemical measurements were carried out at 23 ± 0.5°C. As the real surface area is impossible to determine for the alloys, the current densities are thus referred to the geometric surface area of the electrodes. Electrochemical impedance spectra were obtained with the Autolab; the impedance data were collected over a frequency range between 100 kHz and 10 mHz, and a sine wave with 5 mV amplitude was applied. The EIS spectra were numerically modeled to an appropriate electrical equivalent circuit using the Equivcrt software program, which uses a nonlinear least-squares 共NLLS兲 fitting routine to optimize the impedance characteristics of a proposed equivalent circuit to fit the experimental data.35 Results and Discussion Ptx /Sny catalysts behavior in sulfuric acid solution.— CVs of Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, and Pt25 /Sn75 electrocatalysts in 0.5 M H2SO4 solution are shown in Fig. 1. Depending on the alloy composition, the voltammograms of Fig. 1 were recorded at different upper anodic limits to prevent tin leaching 共or dissolution兲 from the alloy. The scan rate was 50 mV s−1. The Pt–Sn catalysts do not exhibit a clearly distinguished hydrogen oxidation region 共0 to + 0.2 V/RHE兲; however, the currents in the double layer are larger on Pt80 /Sn20 and Pt50 /Sn50 alloys. Furthermore, in terms of current density values, the order of activities is Pt50 /Sn50 ⬎ Pt80 /Sn20 ⬎ Pt95 /Sn5 ⬎ Pt25 /Sn75 共with 0.8, 0.19, 0.07, and 0.05 mA cm−2, respectively, at 0.1 V/RHE兲. Ethanol oxidation in Pt catalyst.— Figure 2 presents Pt catalyst behavior toward ethanol oxidation. The CV was recorded with 50 mV s−1 in 0.5 M H2SO4 + 0.5 M C2H5OH solution. As it was already seen in the literature,19,20 the voltammogram of Fig. 2 presents the following shape. During the positive potential scanning it exhibits two oxidation peaks; the first one 共at 0.870 V/RHE兲 is due to the ethanol oxidation and the second one 共appearing in our case at ca. 1.2 V/RHE兲, is due to side products such as acetaldehyde and acetic acid.10,27,28 During the backward sweep process 共above 0.6 V/RHE兲 the oxidation peak is often attributed to the further oxidation of the adsorbed intermediate species of ethanol.13,21,23 The onset potential for the formation of a product is a determining element to evaluate the activity of an electrode regarding the electrochemical reaction. In the case of platinum catalyst, the onset oxidation potential 共the onset potential value is defined as the value

Ethanol oxidation at Ptx /Sny catalysts .— CVs obtained at Ptx /Sny catalysts 共Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, and Pt25 /Sn75兲 recorded at 50 mV s−1 in 0.5 M H2SO4 + 0.5 M C2H5OH solution are shown in Fig. 3. Compared to platinum catalyst, the presence of tin in the catalyst involves different behavior according to tin content. In Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50 catalysts, the onset potential is shifted toward more negative values 共about 0.2 V/RHE兲 in comparison with the one recorded at the Pt catalyst 共0.5 V/RHE兲. Concerning the first oxidation peak, current density values of these three catalysts are higher than the current recorded with Pt catalyst. The value of the onset oxidation potential for the Sn-rich alloy Pt25 /Sn75 is about 0.41 V/RHE, and the oxidation current peak density is weak, almost comparable to that of pure Pt. We have to emphasize that a sputtered pure Sn electrode did not exhibit any noticeable activity toward ethanol electro-oxidation. Figure 4 summarizes the onset oxidation potentials and forward current peak density values for each catalyst. While looking at the three best catalysts, Pt50 /Sn50 demonstrates a very high catalytic

Figure 3. CV curves for ethanol oxidation on Ptx /Sny catalysts in 0.5 M H2SO4 + 0.5 M C2H5OH, scan rate = 50 mV s−1.

B878

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

Figure 4. Onset oxidation potentials 共--䊊--兲 and oxidation forward-peak current densities 共兩兲 recorded for ethanol oxidation on each catalyst 共Pt, Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, and Pt25 /Sn75兲 in 0.5 M H2SO4 + 0.5 M C2H5OH, scan rate = 50 mV s−1.

activity for ethanol oxidation 共almost 10 mA cm−2兲 compared to Pt95 /Sn5 共1.4 mA cm−2兲, while Pt80 /Sn20 also supplies a relatively satisfactory current density value near 9.0 mA cm−2. Thus, Pt50 /Sn50 and Pt80 /Sn20 seem to be the best performing electrocatalysts.

Figure 5. Chronoamperometric curves recorded at 0.54 V/RHE in 0.5 M H2SO4 + 0.5 M C2H5OH on Ptx /Sny catalysts. 共Inset兲 Chronoamperometric responses for Pt and P25 /Sn75 catalysts.

Stationary potential electrolysis.— The better performance of Pt50 /Sn50 and Pt80 /Sn20 compared to Pt95 /Sn5 or Pt25 /Sn75 was also observed by short-time steady-state current densities at a constant potential. In Fig. 5, all chronoamperometric curves were obtained in 0.5 M H2SO4 + 0.5 M C2H5OH solution by applying 0.54 V/RHE. The initial high current corresponds mainly to the double-layer charging process. The Pt50 /Sn50 electrode also demonstrates the highest current density 共⬃2 mA cm−2兲, while only 0.02 mA cm−2 is recorded for the Pt25 /Sn75 catalyst.

EIS.— EIS is widely used for investigating the performance of electrochemical systems such as fuel cells.22,29-34 Our interest in using this technique is to assess the electrocatalytic activity in terms of charge-transfer resistance of the catalysts toward the oxidation of ethanol. Impedance measurements of Fig. 6-10, respectively, at Pt, Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, and Pt25 /Sn75 electrodes were performed in 0.5 M H2SO4 + 0.5 M C2H5OH solution at various electrode potentials.

Figure 6. Electrochemical impedance spectra at different applied potentials on Pt catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH.

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

B879

Figure 7. Electrochemical impedance spectra at different applied potentials on Pt95 /Sn5 catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH.

In the absence of an effect from diffusion, which is not expected or observed with the concentrations of ethanol used in this study, the presence of two semicircles in the complex plane indicates the reaction proceeds via one adsorbed intermediate with a lifetime long enough to be detected.36-39 Taking into account the complexity of the complete electro-oxidation reaction of ethanol, more than one adsorbed intermediate is probably to be implicated in the reaction mechanism. Nevertheless, our impedance spectra indicate that the rate of turnover of all but one intermediate is very fast. The total impedance of such a system can be expressed in terms of the equivalent circuit shown in Fig. 11. The physical significance of parameters for systems presenting adsorbed intermediates was discussed by Harrington,37 Armstrong,38 and Cao.39 To obtain a satisfactory fit of the data to the circuit in Fig. 11, it was necessary to replace the double-layer capacitance with a constant phase element 共CPE兲. Seland et al. have used a similar equivalent circuit for impedance study of methanol oxidation on platinum electrodes.40 The charge-transfer resistance, Rct, is determined mostly by the semicircle at high frequency during the fitting process and therefore is the best parameter with which to follow the kinetics of ethanol oxidation. Figure 12 shows the change in Rct deduced from the fitting process as a function of potential. This resistance pattern change is very interesting and has never been reported. With the exception of Pt25Sn75, at least five different potential regions are observed between 0.2 and 1.2 V vs RHE for the Pt–Sn alloys, likely indicating a change in the reaction mechanism. It is beyond the

scope of this work to discuss in detail the complex reaction mechanism; however, in terms of charge-transfer resistance, the following observations can be made. 1. In the whole potential range investigated, the chargetransfer resistance increases following Pt25 /Sn75 ⬎ Pt ⬎ Pt95 /Sn5 ⬎ Pt50 /Sn50 ⬵ Pt80 /Sn20, which means that the kinetics of oxidation of ethanol are slower following the same order. This is fairly in agreement with the results of CV and chronamperometry. 2. As to Pt25 /Sn75, the activity of this alloy was detected only within 0.5 and 0.8 V/RHE, where a semicircle could be recorded. Below and above these potentials the catalyst behaved as pure capacitance. 3. Pt95 /Sn5, Pt50 /Sn50, and Pt80 /Sn20 displayed similar profiles of the evolution of the charge-transfer resistance with the potential. These three catalysts showed a decrease in Rct from 0.2 to 0.6 V/RHE potential range, which indicates an increase in the catalytic activity for ethanol oxidation. On the contrary, Pt25 /Sn75 displayed an increase in Rct within the same potential range, which suggests a loss in the catalytic activity. As to pure Pt, Rct decreased up to 0.42 V and then started to increase for higher potentials. In summary, within the potential range of 0.2–0.42 V/RHE, the rate at which ethanol is electro-oxidized at Pt electrode starts to be enhanced, which is indicated by the detection of a charge-transfer semicircle. Nevertheless, Rct increases with the increase in potential within 0.42 and 0.6 V/RHE. This reveals a slow reaction rate of

B880

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

Figure 8. Electrochemical impedance spectra at different applied potentials on Pt80 /Sn20 catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH.

ethanol oxidation caused by the poisoning of reaction intermediate COads, which is strongly adsorbed on Pt and blocks incessant adsorption and dehydrogenation of ethanol. Therefore, the increase in Rct is due to increasing coverage of intermediates. The scenario is different for the Pt95 /Sn5, Pt50 /Sn50, and Pt80 /Sn20 alloys, where Rct continues to decrease up to 0.6 V/RHE, which indicates that CO coverage starts to decrease. This is confirmed by the fact that semiinductive patterns 共clockwise capacitive-inductive loop兲 appeared within this potential region. The semiinductive patterns are characteristic for systems with adsorbed intermediates or for showing transition between passive and active states. Bai and Conway36 have

demonstrated that a semiinductive behavior in systems with adsorbed species is associated with a change in the sign of the coverage dependence on potential. The appearance of a semi-inductive pattern is proof that the CO coverage starts decreasing at this potential. The process responsible for this behavior is likely the oxidation of CO by oxygenated species on the interface. This change in the rate at which ethanol is electro-oxidized can be interpreted by an additional activation in the removal of the adsorbed reaction intermediate, which would cause the coverage of the reaction intermediate to cease increasing with potential. The chemical nature of the

Figure 9. Electrochemical impedance spectra at different applied potentials on Pt50 /Sn50 catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH.

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

Figure 10. Electrochemical impedance spectra at different applied potentials on Pt25 /Sn75 catalyst in 0.5 M H2SO4 + 0.5 M C2H5OH.

oxygen species formed within the 0.42–0.6 V/RHE range is unknown and needs further study using in situ spectroelectrochemical methods. From 0.6 to about 0.84 V, the diameter of the primary semicircle begins to increase with increasing potential at the Pt95 /Sn5, Pt50 /Sn50, and Pt80 /Sn20 alloys and the inductive response is no longer seen within this potential range. This impedance pattern change has also been observed for methanol oxidation and has been very recently reported by Hsing et al.41 on Pt/C electrode and by Melnick and Palmore on polished polycrystalline Pt.42 The fact that the primary semicircle at high frequency increases in diameter with potential indicates that adsorption sites for ethanol are inhibited. This is further supported by the observation of negative values for impedance for higher potentials. One can see that the semicircle flips over to the second quadrant of the complex plane 共Fig. 7-9兲. The corresponding phase-shift plots 共Bode plot兲 show an abrupt jump between 90 and −90° 共data not shown兲, explicitly, and this drastic phase change results in a counterclockwise impedance profile in the Nyquist plot, showing that the two characteristics frequencies are equal. A negative resistance appears when the current decreases with increasing voltage 共see Fig. 2兲, which usually results from pas-

B881

Figure 12. Charge-transfer resistance evolution vs the applied potential on Pt95 /Sn5, Pt80 /Sn20, Pt50 /Sn50, Pt25 /Sn75, and Pt catalysts in 0.5 M H2SO4 + 0.5 M C2H5OH.

sivation of the electrode surface. The formation of an inactive species might have caused the inhibition of the ethanol adsorption and its subsequent oxidation. Conclusion Performances of several Ptx /Sny catalysts, prepared by sputterdeposition, were compared with respect to the electro-oxidation of ethanol in H2SO4 solution. Electrochemical investigations showed that adding tin into platinum improved the electrocatalytic activity of the catalyst, i.e., the onset potential value shifted more than 200 mV toward negative values in the presence of tin. Pt50 /Sn50 and Pt80 /Sn20 catalysts, for which higher current-density values are recorded during ethanol electro-oxidation, are more outstanding than Pt95 /Sn5 or Pt25 /Sn75 electrode. The kinetics of ethanol oxidation on the catalysts were also studied by EIS. Upto an applied potential of 0.6 V/RHE, charge transfer is carried out faster on Pt50 /Sn50 and Pt80 /Sn20 catalysts than on the Pt95 /Sn5 catalyst 共Rct values are ten times lower for Pt50 /Sn50 or Pt80 /Sn20 catalysts than for Pt95 /Sn5兲. Moreover, from 0.5 to 0.6 V/RHE, a pseudo-inductive behavior describing the depoisoning of the catalysts surface by electro-oxidation of the COads was observed for Pt95 /Sn5, Pt50/Sn50, and Pt80 /Sn20 catalysts. This study has shown the beneficial effect of adding a certain amount of tin to platinum for the enhancement of ethanol electrooxidation that is relevant to fuel cell applications. We can conclude that within the Ptx /Sny system, both Pt50 /Sn50 and Pt80 /Sn20, with similar performances are the most promising catalysts for ethanol oxidation. Further studies by means of differential electrochemical mass spectroscopy are in progress in our laboratory to further shed light on the nature of reactions products and intermediates. Acknowledgments This work was supported by the Natural Sciences Engineering Research Council of Canada Strategic Grant “Development of Novel Nanosized Electrocatalysts for Direct Bioethanol Fuel Cells,” The Canadian Foundation for Innovation 共CFI兲, and INRS-EMT.

Figure 11. Equivalent circuit used for the fitting of impedance spectra of Fig. 6-10. Rs is the solution resistance, CPE is the constant phase element representing the double-layer capacitance, Rct is the charge-transfer resistance, and RL and L are the resistance and inductance of the adsorbed layer, respectively.

Institut National de la Recherche Scientifique assisted in meeting the publication costs of this article.

References 1. C. Lamy, J.-M. Léger, and S. Srinivasan, in Modern Aspects of Electrochemistry, Vol. 34, J. O’M. Bockris and B. E. Conway, Editors, Plenum Press, New York

B882

Journal of The Electrochemical Society, 154 共8兲 B876-B882 共2007兲

共2000兲. 2. C. Lamy and J.-M. Léger, in Proceedings of the 2nd International Symposium on New Materials for Fuel Cell and Modern Battery Systems, Ecole Polytechnique, pp. 477–488 共1997兲. 3. C. Coutanceau, A. F. Rakatondrainibe, A. Lima, E. Garnier, S. Pronier, J.-M. Léger, and C. Lamy, J. Appl. Electrochem., 34, 61 共2004兲. 4. E. A. Batista, G. R. P. Malpass, A. J. Motheo, and T. Iwasita, J. Electroanal. Chem., 571, 273 共2004兲. 5. J. Huang, H. Yang, Q. Huang, Y. Tang, T. Lu, and D. L. Akins, J. Electrochem. Soc., 151, 11 共2004兲. 6. J.-H. Choi, K.-W. Park, B.-K. Kwon, and Y.-E. Sung, J. Electrochem. Soc., 150, 973 共2003兲. 7. M. Umeda, H. Ojima, M. Mohamedi, and I. Uchida, J. Power Sources, 136, 10 共2004兲. 8. B. Beden, J.-M. Léger, and C. Lamy, in Modern Aspects of Electrochemistry, Vol. 22, J. O’M. Bockris and B. E. Conway, Editors, Chap. 2, p. 97, Plenum Press, New York 共1992兲. 9. J.-M. Léger, S. Rousseau, C. Coutanceau, F. Hahn, and C. Lamy, Electrochim. Acta, 50, 5118 共2005兲. 10. C. Lamy, S. Rousseau, E. M. Belgsir, C. Coutanceau, and J.-M. Léger, Electrochim. Acta, 49, 3901 共2004兲. 11. S. Rousseau, C. Coutanceau, C. Lamy, and J.-M. Léger, J. Power Sources, 158, 18 共2006兲. 12. C. Lamy, A. Lima, V. Lerhun, F. Delime, C. Coutanceau, and J.-M. Léger, J. Power Sources, 105, 283 共2002兲. 13. S. Song and P. Tsiakaras, Appl. Catal., B, 63, 187 共2006兲. 14. F. Vigier, C. Coutanceau, F. Hahn, E. M. Belgsir, and C. Lamy, J. Electroanal. Chem., 563, 81 共2004兲. 15. S. Tanaka, M. Umeda, H. Ojima, Y. Usui, O. Kimura, and I. Uchida, J. Power Sources, 152, 34 共2005兲. 16. W. J. Zhou, S. Q. Song, W. Z. Li, Z. H. Zhou, G. Q. Sun, Q. Xin, S. Douvartzides, and P. Tsiakaras, J. Power Sources, 140, 50 共2005兲. 17. E. V. Spinacé, M. Linardi, and A. Oliveira Neto, Electrochem. Commun., 7, 365 共2005兲. 18. W. J. Zhou, S. Q. Song, W. Z. Li, G. Q. Sun, Q. Xin, S. Kontou, K. Poulianitis, and P. Tsiakaras, Solid State Ionics, 175, 797 共2004兲.

19. H. Wang, Z. Jusys, and R. J. Behm, J. Power Sources, 154, 351 共2006兲. 20. T. Iwasita and E. Pastor, Electrochim. Acta, 39, 531 共1994兲. 21. C.-G. Lee, T. Itoh, M. Mohamedi, M. Umeda, I. Uchida, and H.-C. Lim, Electrochemistry (Tokyo, Jpn.), 71, 7 共2003兲. 22. Y. Bai, J. Wu, J. Xi, J. Wang, W. Zhu, L. Chen, and X. Qiu, Electrochem. Commun., 7, 1087 共2005兲. 23. S. Q. Song, W. J. Zhou, Z. H. Zhou, L. H. Jiang, G. Q. Sun, Q. Xin, V. Leontidis, S. Kontou, and P. Tsiakaras, Int. J. Hydrogen Energy, 30, 995 共2005兲. 24. F. Colmati, E. Antolini, and E. R. Gonzalez, J. Power Sources, 157, 98 共2006兲. 25. L. Jiang, G. Sun, Z. Zhou, W. Zhou, and Q. Xin, Catal. Today, 93–95, 665 共2004兲. 26. L. Jiang, G. Sun, S. Sun, J. Liu, S. Tang, H. Li, B. Zhou, and Q. Xin, Electrochim. Acta, 50, 5484 共2005兲. 27. A. O. Neto, M. J. Giz, J. Perez, E. A. Ticianelli, and E. R. Gonzalez, J. Electrochem. Soc., 149, 272 共2002兲. 28. N. Fujiwara, K. A. Friedrich, and U. Stimming, J. Electroanal. Chem., 472, 120 共1999兲. 29. W. Sugimoto, K. Aoyama, T. Kawaguchi, Y. Murakami, and Y. Takasu, J. Electroanal. Chem., 576, 215 共2005兲. 30. D. Chakraborty, H. Bischoff, I. Chorkendorff, and T. Johannessen, J. Electrochem. Soc., 152, 2357 共2005兲. 31. E. Frackowiak, G. Lota, T. Cacciaguerra, and F. Béguin, Electrochem. Commun., 8, 129 共2006兲. 32. S. S. Gupta and J. Datta, J. Power Sources, 145, 124 共2005兲. 33. J. T. Mueller and P. M. Urban, J. Power Sources, 75, 139 共1998兲. 34. M. Ciureanu, S. D. Mikhailenko, and S. Kaliaguine, Catal. Today, 82, 195 共2003兲. 35. B. A. Boukamp, Equivalent Circuit Users Manual, 2nd ed., University of Twente, The Netherlands 共1989兲. 36. L. Bai and B. E. Conway, Electrochim. Acta, 38, 1803 共1993兲. 37. D. A. Harrington and B. E. Conway, Electrochim. Acta, 32, 1703 共1987兲. 38. R. D. Armstrong, R. E. Firmann, and H. R. Thirsk, Faraday Discuss. Chem. Soc., 56, 244 共1973兲. 39. C. N. Cao, Electrochim. Acta, 35, 831 共1990兲. 40. F. Seland, R. Tunold, and D. A. Harrington, Electrochim. Acta, 51, 3827 共2006兲. 41. I.-M. Hsing, X. Wang, and Y.-J. Leng, J. Electrochem. Soc., 149, A615 共2002兲. 42. R. E. Melnick and G. T. R. Palmore, J. Phys. Chem. B, 105, 1012 共2001兲.