Statistical Properties of Stepped Surfaces: Melding Quantitative Experiments, Simple Analytical Models, and Computer Simulations

Theodore L. Einstein (einstein@physics.umd.edu)

Dep't of Physics

University of Maryland

College Park, MD 20742-4111

The talk begins with a discussion about general features of stepped surfaces and what makes them so interesting from both a technological and a scientific perspective. Experiments can now monitor quantitatively various statistical properties of stepped surfaces. In order to make sense of them, we use simple models which depend on a few macroscopic parameters: step stiffness, step-step interaction strength, and a kinetic parameter associated with the dominant mechanism for atomic transport. Computer simulations are used to test the consequences of these models, to ascertain whether the viewpoint accounts for different phenomena in a consistent way. One can also start from estimates of microscopic energies, but then results depend sensitively on one's a priori ability to recognize the important, rate-limiting processes. We first consider the equilibrium terrace-width distribution of vicinal surfaces, in particular as a way to deduce step-step interactions. Then we consider the fluctuations of isolated steps and vicinal surfaces in the continuum limit. We also show how the same approach can be used to describe the Brownian motion single-layer clusters of atoms or vacancies on surfaces.

Work supported by NSF grant MRSEC-DMR-96-32521 and done in collaboration with S.V. Khare and O. Pierre-Louis.