Kinetic Monte Carlo simulations of Cu/Cu(001) homoepitaxial growth using anisotropic diffusion local surface

A quantitative understanding of kinetic effects in metal homoepitaxy is essential for controlling thin-film morphology. Here, we investigate the influence of substrate temperature T and deposition rate F on the morphological evolution of homoepitaxial Cu/Cu(001) layers using an atomistic kinetic Monte Carlo (kMC) model that uniquely resolves diffusion events according to the local atomic environment. This is achieved by introducing the concept of diffusion local surface. The model explicitly incorporates crystallographic facet–dependent diffusion barriers, step-edge barrier for downward diffusion and restricted funneling within a unified framework. It reproduces a rich variety of morphologies depending on deposited thickness and kinetic limitations, including growth instabilities characterized by square-based mound formation aligned along ⟨110⟩ directions, mound steepening, and facet development. The early stages of growth are analyzed in the sub-monolayer regime up to the onset of film continuity, defined by the effective thickness θ cont . This latter exhibits a nonmonotonic temperature dependence, reflecting the competition between random adsorption, anisotropic surface diffusion, upward and downhill mass transport. At intermediate temperatures (200–250 K), θ cont follows a power-law dependence on deposition rate with negative scaling exponents, characteristic of three-dimensional growth, while a deposition-rate-independent regime emerges near 300 K. At higher temperatures (400–450 K), enhanced step-descent diffusion promotes smoother, quasi–layer-by-layer growth. The model reproduces the experimentally observed bell-shaped temperature dependence of surface roughness, with a maximum at a characteristic temperature T w = 300 K for F = 0.1 ML/s, followed by reentrant smooth growth at lower temperatures. We further show that T w depends on both deposited thickness and deposition rate. More generally, this local-environment–resolved kMC framework provides a predictive tool for linking microscopic diffusion anisotropy to mesoscale morphology in fcc metal homoepitaxy and related growth systems.