Scale decomposition of molecular beam epitaxy

Download

PDF Download (260Kb)

Description/Abstract

In this work, a study of epitaxial growth was carried out by means of the wavelets formalism. We showed the existence of a dynamic scaling form in a wavelet discriminated linear molecular beam epitaxy (MBE) equation where diffusion and noise are the dominant effects. We determined simple and exact scaling functions involving the scale of the wavelets when the system size is set to infinity. Exponents were determined for both correlated and uncorrelated noise. The wavelet methodology was applied to a computer model simulating linear epitaxial growth; the results showed very good agreement with analytical formulation. We also considered epitaxial growth with the additional Ehrlich–Schwoebel effect. We characterized the coarsening of mounds formed on the surface during the nonlinear phase using the wavelet power spectrum. The latter has an advantage over other methods, in the sense that one can track the coarsening in both frequency (or scale) space and real space simultaneously. Wavelets analysis also provides a quantitative tool for the characterization of the mounded surfaces through its concise scale discrimination. We showed that the averaged wavelet power spectrum (also called scalegram) over all the positions on the surface profile identified the existence of a dominant scale a*, which increases with time following a power law relation of the form a*~t^n, where n=1/3 .