Recent publications 1 3 have critically examined the correlations between hydrogen overvoltage and E(M-H), the strength of the bond between adsorbed hydrogen and the electrode surface. In the absence of experimental data on the latter, Krishtalik 1 extracted estimates of E(M-H) from kinetic and mechanistic information on the h.e.r. (hydrogen evolution reaction) while Trasatti 3 adopted the more conventional approach of assuming that for transition metals, the bond strength of"electrolytic" adsorbed hydrogen could be equated with that of hydrogen adsorbed from the gas-phase, and for non-transition metals he used the bond strength of binary hydrides. If log i 0 (where i 0 is the exchange current density of the h.e.r.) is plotted against E(M-H) obtained by either method, the "volcano" plot predicted by Parsons 4 results. There is, however, a major anomaly in the case of the transition metals Ni, Fe and Co. In Krishtalik's plot, these metals fall on the branch of the curve where the exchange current is increasing with increasing M-H bond Strength, whereas Trassatti places them on the decreasing branch.
The source of this anomaly lies in the experimental evidence, cited by Krishtalik, that the absence of a high electrode capacitance during hydrogen evolution on Ni, Fe and Co is indicative of weak adsorption of the reactive hydrogen-atom intermediate and E(M-H) for this species is therefore substantially lower than for hydrogen adsorbed from the gas phase. Trasatti has addressed this point and concluded that"the electrochemical properties of metals depend on electronic structure in exactly the same way as do their catalytic (or general surface) properties", the inference being that the strength of the M-H bond is not influenced by the solvent. However, in gas-phase adsorption a variety of types of M-H bonds can be formed