Self-learning kinetic Monte Carlo Simulations of Self-diffusion of small Ag clusters on Ag (111) surface

The self-diffusion of two-dimensional small Ag islands (containing up to $10$ atoms) on Ag(111) surface has been studied using and self-learning kinetic Monte Carlo [J. Phys.: Condens. Matter 24, 354004 (2012)] simulations. A variety of concerted, multi-atom and single-atom processes were automatically revealed in these simulations. The size dependence of the diffusion coefficients, effective energy barriers as well as key diffusion processes responsible for island diffusion are reported. In addition, we have compared activation barriers for concerted diffusion processes with those obtained from Density Functional Theory (DFT) calculations.