Three electrodes-E1 (0.18 Mg Pt cm(-2)), E2 (0.38 Mg pt cm(-2)), E3 (0.4 Mg pt cm(-2) without a gas-diffusion layer)-are fabricated and compared with a commercial product (E-TEK). The performance of the electrodes increases with increase in Pt loading in the catalyst layer. The performance of the E2 electrode is superior to that of E1, E-TEK or E3. Elimination of the diffusion layer between the carbon-cloth substrate and the catalyst layer affects the performance of electrode E3 in particular. The power density shows a similar pattern to current density. The difference in performance between E2 and E-TEK electrodes may be due to the difference in the method of fabrication. Increase in exchange current density results in an increase in efficiency. The curves for E1, E2 and E-TEK electrodes appear to stabilize at constant efficiency, which indicates maximum efficiency at a lower exchange current density, compared with the E2 electrode, which does not approach a steady efficiency even at an exchange current density of 1 mA cm(-2). This means that the E2 electrode has greater efficiency than E1, E3 or E-TEK electrodes. Voltage and irreversibility curves for the four electrodes meet at different voltage operational limits; namely, 0.48, 0.55, 0.46 and 0.42 V at 1.2, 0.85, 0.4 and 0.3 mA cm(-2), for E-TEK, E2, E1 and E3 electrodes, respectively. Hence, while these electrodes can be operated conveniently, the E2 electrode with a 0.38 mg Pt cm-2 loading can be operated at optimum conditions of 0.55 V and 0.85 mA cm(-2). (C) 2003 Elsevier Science B.V. All rights reserved.