Transient diffusion of beryllium and silicon in gallium arsenide

Transient diffusion is an increasingly important phenomenon as thermal budgets for real processes decrease and diffusion during sample growth becomes more important. To fully characterize dopant diffusion in gallium arsenide, an understanding must be developed of the dominant atomistic processes for a given dopant, as well as the sources of transient effects under a given set of experimental conditions. Theoretical, experimental, and simulation results were obtained to understand transient diffusivities of beryllium and silicon in grown-in and implanted samples. In implanted samples, by understanding implant damage and modeling the evolution of point defect populations, the observed transient effects can be explained. Such phenomena cannot account for the time-dependent diffusivity observed when the dopant is introduced during molecular beam epitaxial growth. Transient diffusivities for grown-in beryllium were investigated and explained by modeling the evolution of point defect populations as they increase beyond their equilibrium levels at the growth temperature to achieve equilibrium at the anneal temperature.