Identification of Adsorption Ensembles on Bimetallic Alloys

A strategy for counting ensembles for various adsorbates on bimetallic alloys is developed. A Monte Carlo method is used to simulate the structure of a model Au/Pd(111) alloy using a gold−palladium interaction parameter ϵ/kT = 0.64 for the alloy and for a random distribution, ϵ/kT = 0.0. The method is illustrated by counting the ensembles for the adsorption of hydrogen and oxygen and for the conversion of ethylene to ethylidyne, and the results are compared with experiment. The adsorbate coverages are measured as a function of alloy composition primarily using temperature-programmed desorption and the surface composition of the alloy measured using low-energy ion scattering. It is found that hydrogen adsorbs on 3-fold sites for gold coverages less than ∼0.4 monolayers (ML) but on bridge sites at higher coverages. In the case of oxygen adsorption, this saturates at a coverage of 0.25 ML on the clean surface in a (2 × 2) structure. Thus, adjacent sites are excluded on the clean surface and this effect must be taken into account to successfully calculate the oxygen coverage on the alloy. The gold-coverage-dependence of ethylidyne formation on the alloy can be modeled by including an adsorption site to accommodate the resulting hydrogen atom. While better fits to the experimental data are obtained using an interaction parameter ϵ/kT = 0.64, satisfactory fits are also found for a random distribution.