Extended pattern recognition scheme for self-learning kinetic Monte Carlo simulations

Abstract

We report the development of a pattern recognition scheme that takes into account both fcc and hcp adsorption sites in performing self-learning kinetic Monte Carlo (SLKMC-II) simulations on the fcc(111) surface. In this scheme, the local environment of every under-coordinated atom in an island is uniquely identified by grouping fcc sites, hcp sites and top-layer substrate atoms around it into hexagonal rings. As the simulation progresses, all possible processes, including those such as shearing, reptation and concerted gliding, which may involve fcc-fcc, hcp-hcp and fcc-hcp moves are automatically found, and their energetics calculated on the fly. In this article we present the results of applying this new pattern recognition scheme to the self-diffusion of 9-atom islands (M(9)) on M(111), where M = Cu, Ag or Ni.

Self-diffusion of small Ni clusters on the Ni(111) surface: A self-learning kinetic Monte Carlo study

“…Moreover, whereas previous studies have used an on-lattice SLKMC method, [33][34][35][36] in which adatoms were restricted to fcc occupancy, in the present study both fcc and hcp occupancies are allowed and are detectible by our recently developed improved pattern-recognition scheme. 37 The remainder of the paper is organized as follows. In Sec.…”

Self-learning kinetic Monte Carlo simulations of self-diffusion of small Ag islands on the Ag(111) surface

“…Although not shown here, we have also observed reptation processes 37 as well and activation barriers for the intermediate processes were ever lower than single-atom diffusion processes, but former happens only when the shape of the island is non-compact.…”

“…Further details on simulation cell size, calculation of energy barriers for various diffusion processes and other details can be found elsewhere 37 .…”

“…Furthermore, activation barriers of single-atom processes depend on various factors including the type of step-edge, namely A-or B-type, (see Fig.1(b) in Ref. 37) along which atom diffuses. Depending on the type of material either the fcc-island or the hcp-island or mixed-island can be energetically favorable, but in case of Ag(111) surface it is always the fcc-island with few exceptions as.…”

“…On the other hand, as diffusion processes are rare events, molecular dynamics (MD) simulation cannot capture every microscopic processes possible. Therefore, to do a systematic study of small Ag island diffusion on Ag (111) surface we have used on-lattice self-learning kinetic Monte Carlo (SLKMC-II) method 37 , which automatically finds all the relevant atomistic processes and their activation barriers on-the-fly for system under study. This paper is organized as follows.…”