Monte Carlo methods in surface science and catalysis.

Abstract

We have studied the behavior of a model bimolecular Langmuir-Hinshelwood heterogeneous catalytic reaction and a model adsorption system using Monte-Carlo methods. In the model reaction system, we have studied the fractal properties of the surface structures arising from Monte-Carlo simulation of the reaction in various kinetic regimes in the limit of irreversible adsorption of reactant. We have classified three fractal attributes of these surfaces and shown that these structures are transient, evolving to a poisoned state. We also studied a reversible version of this model and found that a noise-induced transition to bistability can be induced in the system through manipulation of the rate of reactant desorption. The noise-induced transition arises through the interaction of noise inherent in statistical fluctuations in the adsorption of reactant with spatial degrees of freedom in our system. We have analyzed the probability density function, the correlation integral, and the power spectrum of time series of the rate of reaction and fractional surface coverage of reactant in this model and we have confirmed the existence of the noise-induced transition and quantified the role of noise in the transition. We have shown that this measure can be efficaciously utilized to study noise-induced phenomena. Our studies with the model adsorption system have focussed on the underst and ing of equilibrium in adsorption systems for which the potential energy surface is inherently heterogeneous, in general, and in modelling the adsorption of CO on Pt(111), in particular. Our contribution to the underst and ing of adsorption in these systems has been our development of a novel "two-site" lattice-gas model in which the equilibrium of adsorbed species is governed by both adsorbate-adsorbate and adsorbate-potential energy surface interactions. We have utilized Monte-Carlo simulations to generate equilibrium surfaces with two adsorbate-adsorbate interaction potentials, a dipole-dipole and a Lennard-Jones potential. We have developed criteria for determining the ground states of our model systems and we have studied the pattern-formation and order-disorder phase transitions of adsorbed species in these systems. We have also measured the coverage-dependence of the heat of adsorption and the infrared shift of the linear species due to dipole-dipole vibrational coupling. The dipole-dipole model accurately represents the behaviour of the CO-Pt(111) systems at coverages below 1/3.