NUMEROUS studies have been made of the electrochemical reduction of oxygen-oxide layers on noble metal electrodes, and of the electrochemical reduction of oxygen on these and on bare metal surfaces in acid solutions."'s The effect of various concentrations of peroxide on these systems has also been investigated.1~9~13*1* There is, however, a lack of data concerning the electrochemical behaviour of platinumplatinum-metal alloys, possibly because of the high electrocatalytic activity of platinum itself, and partly because there is little theory that might predict marked increases in activity on alloying.
Recently Bond and Webster16 have shown that co-reduced platinum and ruthenium oxides give a more active catalyst than either oxide alone for certain hydrogenation reactions. "Enhancement Factors" (rate at alloy/rate at Pt) of up to 4 were obtained for ruthenium contents of 3-30 at-% depending on the system. The catalysts were probably in alloyed form rather than mechanical mixtures. Nishimura and Taguchils have also shown that co-reduced mixed platinum and rhodium oxides give catalytic properties not produced by either oxide alone, and other similar examples have been reported. l7 We have now found that platinum-ruthenium and platinumrhodium alloys are more active electrochemically' than bright platinum for the reduction of oxygen in acid solution.
Galvanostatic measurements were carried out at 25 f O*l"C in sulphuric acid (pH 1.28) prepared from triply distilled water and Hopkin and Williams "low in lead" reagent and cathodically pre-electrolysed. Oxygen was bubbled through the flowing solution during the experiments, a fresh solution being used for each run. A saturated calomel electrode was used as a reference electrode in the same solution but was separated from the main compartment by a sintered glass disk and a Haber-Luggin capillary. The counter electrode was a spiral of bright platinum wire, and the working electrodes (kindly supplied by Johnson, Matthey) were in the form of flags welded to platinum wire and sealed into glass. Before testing they were ground on 6/O's emery paper, flamed and quenched in concentrated hydrochloric and nitric acids, then cathodically polarized to hydrogen evolution in test solution. Finally they were inserted in the cell and again cathodically polarized to hydrogen evolution to give reproducible "bare" surfaces. The current was then rapidly reduced to about 1O-6 A/cm2 ready for the first measurements. Potentials were recorded on a Dynacap pH Metee. The values plotted for any given current density were taken when the * Ma&script