A kinetic Monte Carlo algorithm was developed to simulate the temperature-programmed desorption of oxygen on Rh(100). The interactions between surface oxygen atoms were derived by a bond order conservation (BOC) model that was parameterized from a database of ab-initio density functional quantum chemical results. The intrinsic activation barriers and heats of adsorption were taken from the DFT-calculated database. The effect of coverage on these intrinsic parameters was calculated internal to the simulation at each site throughout the simulation using the modified BOC model. The Monte Carlo simulation was subsequently used to follow the coverage-dependent desorption kinetics for O/Rh(100). The simulated temperature-programmed desorption peaks of 820, 925, and 1250 K compare favorably with experimental peak temperatures at 810, 920, and 1200 K. Below 0.3 ML, oxygen prefers the fourfold hollow sites and orders in a (2×2) arrangement. Between 0.3 and 0.5 ML, oxygen begins to reorder due to repulsive lateral interactions. The results shown here indicate that oxygen begins to occupy bridge sites. Although reconstruction is not explicitly considered, it would likely demonstrate similar effects of reducing repulsive interactions through a change in the surface structure.
Bibliographical noteFunding Information:
The authors wish to acknowledge the support from the NSF CAREER award (CTS-9702762) and the Ford Motor Company for a Young Investigator Award.