Silver-coated ion exchange membrane electrode

0 downloads 0 Views 196KB Size Report
The SPE electrodes were used for carbon dioxide (CO2) reduction with 0.2 M K2SO4 as the ... Keywords: Electrochemical reduction; CO2; Ag; SPE; Ion exchange membrane ..... of Electrochemistry and Photoelectrochemistry, The Electroche-.
Electrochimica Acta 48 (2003) 2651 /2657 www.elsevier.com/locate/electacta

Silver-coated ion exchange membrane electrode applied to electrochemical reduction of carbon dioxide Y. Hori *,1, H. Ito, K. Okano, K. Nagasu, S. Sato Faculty of Engineering, Department of Applied Chemistry, Chiba University, Yayoi-cho, Inage-ku, Chiba 2638522, Japan Received 20 January 2003; received in revised form 26 April 2003; accepted 27 April 2003

Abstract Silver-coated ion exchange membrane electrodes (solid polymer electrolyte, SPE) were prepared by electroless deposition of silver onto ion exchange membranes. The SPE electrodes were used for carbon dioxide (CO2) reduction with 0.2 M K2SO4 as the electrolyte with a platinum plate (Pt) for the counterelectrode. In an SPE electrode system prepared from a cation exchange membrane (CEM), the surface of the SPE was partly ruptured during CO2 reduction, and the reaction was rapidly suppressed. SPE electrodes made of an anion exchange membrane (SPE/AEM) sustained reduction of CO2 to CO for more than 2 h, whereas, the electrode potential shifted negatively during the electrolysis. The reaction is controlled by the diffusion of CO2 through the metal layer of the SPE electrode at high current density. Ultrasonic radiation, applied to the preparation of SPE/AEM, was effective to improve the electrode properties, enhancing the electrolysis current of CO2 reduction. Observation by a scanning electron microscope (SEM) showed that the electrode metal layer became more porous by the ultrasonic radiation treatment. The partial current density of CO2 reduction by SPE/AEM amounted to 60 mA cm 2, i.e. three times the upper limit of the conventional electrolysis by a plate electrode. Application of SPE device may contribute to an advancement of CO2 fixation at ambient temperature and pressure. # 2003 Elsevier Science Ltd. All rights reserved. Keywords: Electrochemical reduction; CO2; Ag; SPE; Ion exchange membrane

1. Introduction The greenhouse effect owing to minor components in the atmospheric air may cause irreversible changes to the global environment in the coming several decades. Mitigation of this effect is required, and various research works are promoted with regard to fixation of carbon dioxide (CO2) and conversion to valuable substances. Electrochemical reduction of CO2 is an attractive alternative, since this reaction can convert CO2 directly to useful substances by one step. This process may be applied to a novel energy storage method for utilization

* Corresponding author. Tel./fax: /81-3-3630-4086. E-mail address: [email protected] (Y. Hori). 1 ISE member.

of intermittent renewable energies such as solar and wind energies [1]. Among numerous electrochemical processes, reduction of CO2 at solid metal electrodes is simple and practically advantageous, since this process provides relatively high current density. High-purity copper electrode is the only electrode at present that can reproducibly yield methane, ethylene, and alcohols in aqueous electrolytes at high current densities [2,3] as confirmed by many workers. However, low solubility of CO2 in aqueous solution limits the current density to 20 mA cm2 or so. This problem with regard to transport process may be solved by some modified CO2 reductions, such as electrolysis under high pressure [4,5], gas diffusion electrode [6 /14], and metal-coated ion exchange membrane electrode (occasionally called ‘‘solid polymer electrolyte’’ or SPE) [15 /20]. Among these devices the

0013-4686/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0013-4686(03)00311-6

2652

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657

SPE process is a safe technology, employed under atmospheric pressure. A review of these attempts has recently been published [21]. SPE electrodes were applied to CO2 reduction as previously reported by some workers. Ito and co-workers [15] used a cation exchange membrane (CEM) (Nafion 315) coated with Au as the electrode metal. The reaction product was CO, but the current density of CO2 reduction was much lower than that obtained by a conventional electrolysis using an Au plate electrode. The low current density is probably due to poor electric contact of Au particles on the SPE electrode. DeWulf and Bard [16] prepared a Cu SPE electrode using a CEM (Nafion 115). Their measurements of CO2 reduction with 1 mM H2SO4 as the electrolyte solution (counter solution) gave a total current efficiency of 19% as a maximum with much lower partial current than the value obtained by a bulk Cu metal electrode. Cook et al. [17] reported CO2 reduction with a Cu-deposited CEM (Nafion 117). The maximum current density was 30 mA cm 2, and the current efficiency of C2H4 formation was less than 6.5%. They also investigated various metals containing Ag as the electrode metal of SPE [18]. They reported that none of the electrode metals they tested showed high electrocatalytic activity for CO2 reduction. Kunugi and co-workers [19] employed an anion exchange membrane (AEM) (Selemion AMV) as well as a CEM (Nafion 117) for preparation of SPE electrodes. They showed that the maximum partial current density of CO2 reduction was less than ca. 1.3 mA cm 2 for both SPE electrodes with the electrolyte solution of 0.5 M K2SO4. The major reaction product was C2H4 for the SPE with the AEM, and HCOOH for the SPE with the CEM. The electrocatalytic activity was stably sustained for 5 h. Kunugi and Yumiyama [20] also reported CO2 reduction at Bi-coated SPE electrodes. The product from CO2 was mainly HCOOH with slight amount of CO. The maximum partial current density of CO2 reduction was 18 mA cm 2 at /1.9 V vs. SCE. As reviewed above, SPE electrodes except Bi-coated one did not successfully accomplish to enhance CO2 reduction rate in comparison with those by bulk metal electrodes in aqueous media. The gaseous products are easily separated from the electrolyte solution. Thus, we studied SPE electrodes coated with Cu and Ag as the electrocatalysis for the electrochemical reduction of CO2. Both metal electrodes give gaseous products such as CH4 and C2H4 or CO. The SPE electrodes with Cu prepared by usual electroless deposition failed to give high partial current density of CO2 reduction. The present paper reports electrochemical reduction of CO2 at silver-coated SPE electrodes prepared from AEM and CEM. The rate of CO2 reduction was highly enhanced at SPE electrodes made of AEM.

2. Experimental 2.1. Preparation of SPE electrodes The SPE electrodes were prepared from a CEM (Nafion 117) and an AEM (Selemion AMV manufactured by Asahi Glass Co. Ltd.). Metallic silver was deposited on the membranes from 0.01 M AgNO3 by electroless deposition at 40 8C using 0.015 M NaBH4 as the reductant solution. SPE electrodes with the AEM (SPE/AEM) was prepared as follows. AgNO3 solution and NaBH4 solution were placed at both sides of the AEM which was vertically mounted in a Pyrex cell, and porous silver layer was deposited on the membrane at the AgNO3 solution side. The deposition was carried out for 90 min. SPE electrodes with the CEM (SPE/CEM) was not successfully prepared in a similar way, since Ag  penetrates into the CEM and silver metal was deposited inside the membrane. In order to obtain silver deposited only at the membrane and solution interface, we initially placed pure water and AgNO3 solution at both sides of the membrane for 30 min. After penetration of Ag ion in the membrane, the AgNO3 solution was removed from the cell. The cell was washed and then NaBH4 solution was poured into the compartment where the AgNO3 solution was previously placed. Ag was reduced to metallic Ag on the membrane for 30 min. This procedure was repeated three times, and we obtained Ag-coated CEMs. We also studied the effect of ultrasonic radiation during the preparation of SPE/AEM electrodes. After an electroless deposition for 90 min, an ultrasonic radiation was applied to SPE/AEM electrodes for 3 min. A part of deposited Ag metal loosely bound to the electrode was removed during this treatment. Then, the procedure of the electroless deposition was repeated for 30 min. 2.2. Evaluation of SPE electrodes The SPE electrodes appeared smooth, and greyish white in color without metallic lustre. The surfaces of the electrodes were observed by a scanning electron microscope (SEM) Model ABT-32T manufactured by Topcon Co. Ltd. The amounts of deposited Ag on SPE/AEM electrodes were analyzed by chronopotentiometry. The deposited Ag on the SPE electrodes was dissolved in concentrated HNO3. The solution was evaporated to dryness, and the residue was dissolved in 0.1 M HNO3. Chronopotentiometric measurements were conducted with various constant currents with a Pt wire electrode. The transient times were analyzed in accordance with Sand’s equation [22]:

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657

2it1=2 p1=2 nFAD1=2 C where i denotes the current applied to the electrode, t the transient time, n (/1) the number of electrons involved with the reaction, F the Faraday constant, A the surface area of the electrode, D the diffusion constant of Ag  ion, and C the concentration of Ag  ion in the test solution. A good working straight line was obtained between t1/2 and C /i for the standard solution, passing the origin. 2.3. Electrolytic cell for the reduction of CO2 The SPE electrodes prepared by the procedures described above were mounted vertically in an electrolytic cell using Neoprene O rings. A gold collector ring was also mounted together in the electrolytic cell for the electric contact with the SPE electrode. The size of the SPE electrodes was 25 mm diameter as the working part. The ion exchange membrane side of the SPE electrode is in contact with the electrolytic solution, and the metaldeposited side faced the compartment where CO2 gas was continuously supplied at a constant flow rate ca. 100 ml min 1 at the atmospheric pressure. A platinum plate (Pt) (20 /20 mm2) worked as the counterelectrode in the electrolytic solution. A Luggin capillary was placed close to the SPE electrode from the solution side with an Ag/AgCl reference electrode connected. Constant current electrolyses were conducted using a potentiogalvanostat (Model 2020 manufactured by Toho Giken Co. Ltd.). The electric charge consumed in the electrolysis was measured by a coulombmeter (Model MF 201 manufactured by Hokuto Denko Co. Ltd.). The electrode potential is given with respect to SHE. We did not compensate the ir drop across the membrane and the electrolyte solution between the Luggin capillary tip and the electrode metal. 2.4. Analysis of the products The effluent gas from the gas compartment of the electrolytic cell was introduced to gas chromatographs, and analyzed every 5 min. The electrolytic solution was analyzed for formic acid after the electrolysis by a liquid chromatograph. Other experimental details may be referred to our previous publication [3].

3. Results and discussion 3.1. Electrochemical reduction of CO2 at an Ag-coated SPE/CEM electrode An Ag-coated SPE/CEM electrode was tested for CO2 reduction at a constant current density of 20 mA cm 2 with the electrolyte solution 0.2 M K2SO4. CO2 was

2653

reduced to CO similarly to a solid Ag plate electrode. H2 also appeared as a product. However, the electrode shortly lost electrocatalytic activity; CO formation dropped and hydrogen evolution prevailed in 10/15 min. The SPE/CEM electrode surface was found partly ruptured soon after the beginning of the electrolysis. A small amount of liquid came out slowly from the ruptured part of the electrode, covering the electrode and preventing the supply of CO2 to the electrode. The liquid was KOH and K2CO3 as analyzed qualitatively. The electrolysis cell was then modified so that the liquid from the ruptured part of the electrode could easily run out without staying at the electrode. The modified electrolysis cell sustained the electrode activity more than 30 min, giving nearly constant current efficiencies of 65 and 29% for CO and H2, respectively, with the electrode potential of /1.5 to /1.6 V. Thus, we judged that SPE/CEM electrode is not suitable for CO2 reduction, and did not investigate any more. 3.2. Ag-coated SPE/AEM electrode Electrolysis measurements using SPE/AEM were carried out with 0.2 M K2SO4 solution as the electrolyte. No visual change of the SPE electrodes was observed during the electrolyses. The reduction products are CO and HCOOH in agreement with the results obtained by the present authors using an Ag plate electrode in 0.1 or 0.5 M KHCO3 aqueous solutions at the current density of 5 mA cm 2 [2,23]. The current efficiency for CO and H2 remained constant. Table 1 shows the results obtained with an Ag-coated SPE electrode prepared without ultrasonic radiation. The activity of the electrodes stood more than 2 h with regard to CO2 reduction, whereas the potential shifted to the negative direction. We attempted to modify the preparation of the SPE electrodes in order to improve the electrocatalytic property. Among these attempts, ultrasonic radiation as described above was effective. Table 2 shows the average current efficiencies of CO, HCOOH, and H2 in the electrolysis measurements for 2 h with two Agcoated SPE/AEM electrodes prepared by the present method. The current efficiency of CO formation is much higher at 20 and 50 mA cm 2 than those given in Table 1. The electrocatalytic activity of the electrode is maintained up to 100 mA cm2, whereas, the electrode potential shifted to more negative values during the electrolysis. Table 2 tabulates the electrode potentials at the beginning and the end of the electrolysis runs as well as the values averaged during each electrolysis period. The variation of the electrode potential during the electrolysis is greater at higher current density. The

2654

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657

Table 1 Average current efficiencies of products from electrochemical reduction of CO2 at an Ag-coated AEM electrode with 0.2 M K2SO4 as the electrolyte solution Current density (mA cm2)a

Electrode potential (V vs. SHE) b

20 50

c

Current efficiency (%) d

Average

Initial

Final

CO

HCOOH

H2

Total

/1.49 /1.80

/1.47 /1.69

/1.50 /1.90

74.4 62.7

6.3 10.8

20.6 26.2

101.3 99.7

The electrode was prepared without ultrasonic radiation. a Controlled current electrolyses. b Electrode potential averaged for the electrolysis period. c Electrode potential at the beginning of the electrolyses. d Electrode potential at the end of the electrolyses.

amounting to or exceeding 0.15 at 100 mA cm 2. Higher fraction of HCOOH may result from more negative potential at higher current density and higher pH at the metal electrolyte interface. The total current efficiency was less than 100%. In addition to the analytical error of 2 /3% for each product from the gas chromatographic and the liquid chromatographic techniques, HCOOH formed at the metal membrane interface of the SPE electrodes will partly remain in the AEM. The experimental error due to HCOOH remaining in the membrane will be the higher for experimental runs in which the faradaic yield of HCOOH is higher.

heat generated due to the ir drop and the overpotentials at the electrode may cause slight local expansion or deformation of the electrode membrane, deteriorating the electric contact between the metal particles of the SPE electrodes. Thus, the current distribution may not be totally homogeneous within the SPE electrode under high current densities. We indicate the experimental data from two electrodes in Table 2. The discrepancy between the two electrodes may be derived from such reason. Stabler structure of metal layer needs to be developed for longer term electrolysis at high current density. Nevertheless, it is remarkable that the partial current density for CO2 reduction amounts to 60 mA cm 2, three times the upper limit of the electrochemical reduction of CO2 in a conventional electrolysis in an aqueous electrolyte solution using a metal plate electrode. The major fraction of the products from an Ag plate electrode is CO in 0.1 or 0.5 M KHCO3 aqueous solutions at the current density of 5 mA cm 2, and the HCOOH to CO ratio was less than 0.05 [2,23]. Tables 1 and 2 show that the HCOOH to CO ratio increases with the increase of the current density,

3.3. Diffusion limited transport of CO2 to the electrode The partial current of CO2 reduction, obtained from the sum of the current efficiencies of CO and HCOOH multiplied by the current density, is plotted against the electrode potential as presented in Fig. 1. The partial current shows a saturation against the potential, suggesting that the CO2 transport is limited by diffusion process.

Table 2 Average current efficiencies of products from the electrochemical reduction of CO2 at two Ag-coated AEM electrodes with 0.2 M K2SO4 as the electrolyte solution Current density (mA cm2)a

20 20 50 50 100 100

Electrode potential (V vs. SHE)

Current efficiency (%)

Averageb

Initialc

Finald

CO

HCOOH

H2

Total

/1.31 /1.30 /1.46 /1.80 /2.87 /2.96

/1.30 /1.27 /1.45 /1.69 /2.82 /2.72

/1.33 /1.32 /1.47 /1.89 /2.95 /3.05

92.1 92.3 64.6 69.0 47.9 52.7

2.7 3.9 7.3 11.3 12.1 8.0

2.3 1.1 20.4 11.2 27.0 35.9

97.1 97.3 92.3 91.5 87.0 96.6

The electrodes were prepared with ultrasonic radiation treatment. a Controlled current electrolyses. b Electrode potential averaged for the electrolysis period. c Electrode potential at the beginning of the electrolyses. d Electrode potential at the end of the electrolyses.

Electrode No.

1 2 1 2 1 2

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657

2655

Fig. 1. Partial current density of CO2 reduction at silver-coated AEM electrodes prepared with ultrasonic radiation. The open circles and triangles correspond, respectively, to two different electrodes.

The electric resistance of the AEM (Selemion) is 2.0 / 3.5 V cm2 in accordance with the manufacturer. Assuming the resistance as 2.8 V cm2, we can estimate the ir drop across the membrane as 0.28 V at the current density 100 mA cm2, and 0.14 V at 50 mA cm 2. The specific resistance of the electrolyte solution 0.2 M K2SO4 is estimated as 2.5 V cm from the molar conductivity [24]. Since the gap between the Luggin capillary tip and the SPE membrane is less than 1 mm in our experimental setup, the ir drop is less than 0.25 V at 100 mA cm 2, and 0.13 V at 50 mA cm 2. Thus, the total contribution of the ir drop is not more than 0.53 V at 100 mA cm2, and 0.27 V at 50 mA cm 2. Taking into account these values and the shape of the correlation curve in Fig. 1, we conclude that the CO2 transport is limited by diffusion process at higher current density. 3.4. Features of SPE/AEM electrodes prepared with and without ultrasonic radiation The amounts of Ag metal deposited on the SPE/AEM electrodes were measured by chronopotentiometry. The analyses were carried out with electrodes prepared by the identical process as described above. The amount was 6.1 mg cm2 for an electrode prepared without ultrasonic radiation, and 2.1 mg cm 2 with ultrasonic radiation. The surfaces of both the SPE/AEM electrodes with and without ultrasonic radiation appeared smooth. SEM images with 1000 times magnification are given in Figs. 2 and 3. Fig. 2 shows that the surface is composed of very fine particles ca. 1 mm or less together with small number of larger particles, whereas, Fig. 3 indicates that the surface is covered mainly with relatively coarse particles of 5 mm or so. The electrode prepared with ultrasonic radiation is apparently composed of porous metal layer with coarse particles, providing higher gas permeability than the one prepared without ultrasonic radiation. Thus, CO2 diffuses more easily to the electrode electrolyte interface through the porous electrode metal layer. The electrode

Fig. 2. SEM image of a silver-coated AEM electrode prepared without ultrasonic radiation. Magnification: /1000. Horizontal lines are the fibers which reinforce the membrane.

Fig. 3. SEM image of a silver-coated AEM electrode prepared with ultrasonic radiation. Magnification: /1000.

without ultrasonic radiation is composed of aggregates of fine particles, in which CO2 gas is not supplied rapidly enough to the electrode/electrolyte interface. These features may rationalize the difference between the two electrodes. The active surface area of the SPE electrodes may be roughly estimated from these values. If all the particles are assumed to be spheres with homogeneous size, the total surface area of spheres of the diameter 1 mm is 34.8 cm2 cm 2 for the silver loaded on the electrode 6.1 mg cm 2; the value for the particles of 5 mm for 2.1 mg cm 2 is 2.4 cm2 cm2. The active domain of the SPE electrodes will be composed of three phase regions, i.e. gas /solid/liquid; the actual active surface area of the SPE electrodes will be considerably less than these values.

2656

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657

3.5. Difference of the SPE electrodes between AEM and CEM The difference of SPE electrodes between AEM and CEM may be rationalized as follows. OH  is generated at the electrode in the cathodic reduction of CO2 in aqueous media. CO2 H2 O2e 0 CO2OH The resultant OH  will react with gaseous CO2, forming HCO3 or CO2 3 . Since the present reaction system employs K2SO4 as the electrolyte solution, K  ions are abundantly supplied to the electrode in SPE/ CEM system. Thus, KHCO3 or K2CO3 is formed at the metal membrane interface. However, neither OH , HCO3 nor CO2 is eliminated from the metal mem3 brane interface in CEM. Thus, K2CO3 must be accumulated and flow out from the electrode, consequently peeling the metal from the membrane. According to Kunugi and co-workers [19], both Cucoated SPE electrodes from CEM and AEM showed stable electrode performance for CO2 reduction for 5 h. We presume that liquid containing KOH or K2CO3 may have run out from their SPE/CEM electrode as described above, since they employed 0.5 M K2SO4 as the electrolyte solution. The current density of their measurements was 25 mA cm 2 or less. Thus, the liquid coming out from the SPE electrode would have easily flowed down from the electrode part due to the structure of their electrolytic cell. Cook et al. [18] reported an Ag-coated SPE electrode prepared from a CEM (Nafion 117) with Pt deposited as the counterelectrode. Hydrogen stream was supplied to the counterelectrode as the reductant. Their Ag SPE electrode did not promote CO2 reduction. In their Ag SPE electrode system, H  is continuously formed in the anodic oxidation of H2, and supplied through the membrane to the cathode where CO2 is reduced. Thus, the Ag membrane interface must be strongly acidic, and the cathodic reaction is prevalently hydrogen evolution reaction. CO2 reduction is practically suppressed. AEM provides high anionic conductivity. OH and CO2 will be easily eliminated from the metal mem3 brane interface, permeating through the membrane to the electrolyte solution. Thus, CO2 reduction will not be prevented with an AEM electrode, and can stand for long time. Hence, AEM is more suitable for preparation of SPE for CO2 reduction.

4. Conclusions Ag/SPE electrodes prepared from an AEM was applied to CO2 reduction, giving the partial current density as high as 60 mA cm 2. Ag/SPE prepared from a CEM was not suitable for CO2 reduction, since OH ,

formed in the CO2 reduction cannot HCO3, and CO2 3 be eliminated from the metal membrane interface. Ultrasonic radiation during the preparation of SPE electrode is effective to promote CO2 reduction. SEM observation showed that the Ag layer treated by ultrasonic radiation was more porous than that without ultrasonic radiation. The porous Ag layer probably enhances the transport rate of CO2 to the electrode. The electrolysis current was still controlled by the CO2 transport through the electrode metal layer at higher current density. Well-designed porous metal layer for SPE will highly promote the rate of CO2 reduction. The present results show that application of SPE device may contribute to an advancement of CO2 fixation technology at ambient temperature and pressure.

Acknowledgements This work was supported by Research Institute of Innovative Technology for the Earth of Japan (RITE) and New Energy and Industrial Technology Development Organization (NEDO).

References [1] Y. Hori, in: M. Tomkiewicz, R. Haynes, H. Yoneyama, Y. Hori (Eds.), Proceedings of the Symposium on Environmental Aspects of Electrochemistry and Photoelectrochemistry, The Electrochemical Society, 1993, p. 1. [2] Y. Hori, K. Kikuchi, S. Suzuki, Chem. Lett. (1985) 1695. [3] Y. Hori, A. Murata, R. Takahashi, J. Chem. Soc., Faraday Trans. 1 85 (1989) 2309. [4] M. Azuma, K. Hashimoto, M. Hiramoto, M. Watanabe, T. Sakata, J. Electrochem. Soc. 137 (1990) 1772. [5] T. Saeki, K. Hashimoto, A. Fujishima, N. Kimura, K. Omata, J. Phys. Chem. 99 (1995) 8440. [6] M.N. Mahmood, D. Masheder, C.J. Harty, J. Appl. Electrochem. 17 (1987) 1159. [7] N. Furuya, T. Yamazaki, M. Shibata, J. Electroanal. Chem. 431 (1997) 39. [8] M. Shibata, K. Yoshida, N. Furuya, J. Electrochem. Soc. 145 (1998) 595. [9] N. Furuya, K. Matsui, S. Motoo, Denki Kagaku 56 (1988) 288. [10] R.L. Cook, R.C. MacDuff, A.F. Sammells, J. Electrochem. Soc. 137 (1990) 607. [11] S. Ikeda, T. Ito, K. Azuma, K. Ito, H. Noda, Denki Kagaku 63 (1995) 303. [12] M. Schwartz, R.L. Cook, V.M. Kehoe, R.C. MacDuff, J. Patel, A.F. Sammells, J. Electrochem. Soc. 140 (1993) 614. [13] K. Hara, A. Kudo, T. Sakata, M. Watanebe, J. Electrochem. Soc. 142 (1995) L57. [14] F. Philipps, E. Feil, A. Jirka, E. Mennenkamp, A. Bandi, in: T.N. Veziroglu (Ed.), Hydrogen Energy Prog. XI, Proceedings of the 11th World Hydrogen Energy Conference, Vol. 2, 1996, p. 1317. [15] M. Maeda, Y. Kitaguchi, S. Ikeda, K. Ito, J. Electroanal. Chem. 238 (1987) 247. [16] D.W. DeWulf, A.J. Bard, Catal. Lett. 1 (1988) 73. [17] R.L. Cook, R.C. MacDuff, A.F. Sammells, J. Electrochem. Soc. 135 (1988) 1470.

Y. Hori et al. / Electrochimica Acta 48 (2003) 2651 /2657 [18] R.L. Cook, R.C. MacDuff, A.F. Sammells, J. Electrochem. Soc. 137 (1990) 187. [19] S. Komatsu, M. Tanaka, A. Okumura, A. Kunugi, Electrochim. Acta 40 (1995) 745. [20] A. Kunugi, W. Yumiyama, Denki Kagaku 66 (1998) 326. [21] Y. Hori, in: W. Vielstich, A. Lamm, H. Gasteiger (Eds.), Handbook of Fuel Cell Technology, vol. 2 (Chapter 52), Wiley, Chichester, 2003.

2657

[22] D.T. Sawyer, J.L. Roberts, Experimental Electrochemistry for Chemists (Chapter 8), Wiley, New York, 1974. [23] Y. Hori, H. Wakebe, T. Tsukamoto, O. Koga, Electrochim. Acta 39 (1994) 1833. [24] The Electrochemical Society of Japan (Ed.), Handbook of Electrochemistry, 5th ed., Maruzen, Tokyo, 2000, p. 110.