Multiscale diffusion Monte Carlo simulation of epitaxial growth

We present an accelerated kinetic Monte Carlo (KMC) simulation algorithm for molecular beam epitaxial growth during step flow. The acceleration is achieved by allowing adatoms far from the step edges to execute larger jumps with correspondingly reduced rates. The computational complexities of a number of different algorithmic implementations are analyzed and compared. The scaling of the accelerated algorithms are verified by test runs performed on a one-dimensional model. Both performance and accuracy are evaluated. One to two orders of magnitude increases of efficiency are achieved while preserving physical accuracy.