Density profile evolution and nonequilibrium effects in partial and full spreading measurements of surface diffusion

We study the nature of nonequilibrium effects in the collective diffusion coefficient D-C(theta) vs the coverage theta as extracted from Boltzmann-Matano analysis of spreading coverage profiles. We focus on the temporal behavior of the profiles and study how the corresponding nonequilibrium effects in D-C(theta) depend on the initial density gradient and the initial state from which the spreading starts.

To this end, we carry out extensive Monte Carlo simulations for a lattice-gas model of the O/W(110) system. Studies of submonolayer spreading from an initially ordered p(2x1) phase at theta = 1/2 reveal that the spreading and diffusion rates in directions parallel and perpendicular to rows of oxygen atoms are significantly different within the ordered phase.

Aside from this effect, we find that the degree of ordering in the initial phase has a relatively small impact on the overall behavior of D-C(theta). Also, although we find that nonequilibrium effects are clearly present in submonolayer spreading profiles, D-C(theta) determined from such data approaches its asymptotic equilibrium behavior much more rapidly than in the case of full spreading.

Nevertheless, in both cases there are noticeable deviations from equilibrium results that persist even at very long times and are strongest in ordered phases and in the vicinity of phase boundaries. These conclusions are confirmed by complementary studies of the temporal behavior of the order parameter phi(theta).

Finally, we use D-C(theta) and phi(theta) to determine the locations of phase boundaries and find such data to be clearly time dependent during full spreading. We conclude that nonequilibrium effects seem to be an inherent feature in profile evolution studies of surface diffusion in all cases where ordering plays a prominent role.

This warrants particular care to be taken with profile spreading experiments. (C) 2001 American Institute of Physics.