Trends in the Exchange Current for Hydrogen Evolution
J. K. Nørskov, T. Bligaard, A. Logadottir, J. R. Kitchin, J. G. Chen, S. Pandelov, and U. Stimming, “Trends in the Exchange Current for Hydrogen Evolution”, Journal of The Electrochemical Society 152(3), J23-J26 (2005). https://doi.org/10.1149/1.1856988
This summary was written by a language model from the paper and has not been checked line by line. Rely on the published paper for details.
Question
Metals differ by many orders of magnitude in the exchange current density they deliver for the hydrogen evolution reaction (HER), the rate of H2 evolution per unit area at the equilibrium potential. The paper asks where these variations come from, whether they can be explained by the strength of the hydrogen-metal bond computed from first principles, and why the Pt group metals are the best HER electrocatalysts (Nørskov et al. 2005).
Approach
The authors build a consistent database of hydrogen chemisorption energies with density functional theory (DFT), using a plane wave pseudopotential code with ultrasoft pseudopotentials and the RPBE functional. Most surfaces are modeled as three-layer fcc(111) slabs in a 2x2 cell (bcc(110) for W, Mo and Nb), with H in the fcc site at coverages of 1/4 and 1 monolayer. They compile measured exchange currents from the literature, on both polycrystalline and single crystal electrodes, for Au, Ag, Pd, Pt, Rh, Ir, Ni, W, Co, Cu, Mo, Re and Nb, and plot them against the calculated binding energies (Nørskov et al. 2005).
To turn binding energies into free energies, they add a zero point energy correction of 0.04 eV (computed for H on Cu(111) and assumed to hold for all metals) and an entropy term equal to half the gas-phase entropy of H2, which gives the simple relation that the adsorption free energy equals the chemisorption energy plus 0.24 eV. They then construct free energy diagrams at the equilibrium potential (pH 0, 1 bar H2, 300 K) and develop a minimal kinetic model for the proton reduction step that includes only effects arising from the thermochemistry. The model assumes a metal-independent prefactor for proton transfer, a Langmuir coverage of adsorbed H, and an extra activation equal to the adsorption free energy when that step is endothermic. Finally, they compute the energies of dissociative water adsorption to OH and O on several metals to judge whether the surfaces would be oxidized under reaction conditions (Nørskov et al. 2005).
Key findings
Plotting the measured exchange currents against the calculated hydrogen chemisorption energies gives a clear volcano. The reactive metals sit on the strong-binding leg, while Cu, Ag and Au lie on the weak-binding leg. In the free energy diagrams, H adsorption on Pt is close to thermoneutral at equilibrium, whereas Ni and Mo bind H too strongly and Au too weakly. With a single fitted parameter (a prefactor of 200 per second per site), the kinetic model reproduces the trends and the position of the maximum, with a hydrogen adsorption free energy of zero separating the two legs. On the strong-binding side, the rate falls because few free sites remain for H recombination; on the weak-binding side, proton transfer to the surface becomes increasingly uphill (Nørskov et al. 2005).
The model underestimates the currents for metals far from the peak. For Ag and Au the authors attribute this to steps and defects that bind H more strongly than the close-packed surface. For Mo, W and Nb they argue the measured values reflect surface oxides rather than metals, which the water dissociation calculations support: among the metals examined there, only Mo and W are predicted to be covered by oxygen. Adding a water overlayer changes the H adsorption energy on Pt(111) by less than 0.02 eV, so solvation effects on H binding are neglected. High-coverage energies follow the same trends as low-coverage ones. For Pt(111), a two-state picture comparing 1/4 and 1 monolayer suggests the weakly bound high-coverage hydrogen has the lower overall barrier and is the likely active species, which the authors offer as a possible explanation of the two types of hydrogen invoked in the electrochemistry literature (Nørskov et al. 2005).
Significance
The paper distinguishes itself from earlier HER volcanoes by using a systematic, computed descriptor of the intrinsic hydrogen-metal interaction as the x-axis, which gives a simple thermochemical explanation of the volcano and of the superior activity of Pt. The authors suggest the approach could help predict bimetallic electrocatalysts for hydrogen evolution and for hydrogen oxidation in fuel cells. The hydrogen adsorption free energy defined here became a standard descriptor for computational screening of HER catalysts. The work prompted a published comment by Schmickler and Trasatti, to which the authors replied in (No̸rskov et al. 2006).