Direct Borohydride Fuel Cells (DBFCs) have the potential to generate high power densities for use in portable power applications. The volumetric energy density of an aqueous sodium borohydride solution, the DBFC anode fuel, at its solubility limit exceeds that of pure methanol and is about half that of gasoline. Despite the potential of DBFC's, current applications are limited in part, by the lack of an effective anode electrocatalyst. Though a number of pure metals and alloys have been tested as anodes, gold and silver electrodes are uniquely capable of completing direct oxidation at high coulombic efficiency. Overpotentials on Au anodes, however, limit the overall cell efficiency to low values and little is known of the elementary electrocatalytic mechanisms, reaction intermediates, or limiting elementary steps . Difficulties associated with experimental characterization of elementary kinetics for the complex 8 electron reaction motivate our application of density functional theory (DFT) methods. In this study, DFT calculations are used to examine borohydride electro-oxidation over the Au(111), and compare with Pt(111) surfaces for which non-selective hydrolysis paths compete with direct oxidation. Possible stable surface intermediates and limiting elementary steps are identified and results are discussed in comparison with experimental data from the literature.

The thermodynamic driving force to adsorption of the BH₄^- ion differs substantially between the Au(111) and Pt(111) surfaces. Over the Au(111) surface, adsorption with electron transfer becomes favorable only at large anode overpotentials. The BH₄ species adsorbs molecularly over Au, followed by an activated B-H dissociation step. Initial adsorption of the reactant ion therefore limits the performance of Au anodes. Dissociative adsorption is favorable at all potentials of interest over Pt(111). However, subsequent hydrogen evolution competes with the oxidation reaction, leading to non-selective hydrolysis which limits the DBFC Coulombic efficiency. On both Pt and Au, initial dehydrogenation steps occur over small reaction barriers, and direct oxidation of adsorbed BH₄ species is limited by O-H dissociation steps occurring late in 8 electron pathway.

These initial results provide insight into rational design approaches for transition metal alloy catalysts for the borohydride oxidation reaction. In addition to the direct relevance of the reported results to DBFCs, the accuracy in the various methods applied to model the electrochemical interface will be discussed. Specifically, comparison between “vacuum slab” and solvated, variable potential methods in calculating ion adsorption equilibrium constants and surface reaction activation barriers will be presented.