Article ID Journal Published Year Pages File Type
1562638 Computational Materials Science 2010 12 Pages PDF
Abstract

Large-scale atomistic simulations of silicon offer the possibility to support new developments in semiconductor technology. However, it is not an easy task to describe silicon by a reliable empirical potential. A good candidate for such a potential is the Modified Embedded Atom Method scheme [B-J. Lee, M.I. Baskes, Phys. Rev. B 62 (2002) 8564], but one needs an optimal parameter set for a dependable outcome. A recent new parametrization [M. Timonova, B-J. Lee, B.J. Thijsse, Nucl. Instrum. Meth. B 255 (2007) 195] (called MEAM-L) has shown promising results in simulations of surface sputtering, and it was anticipated that this would also be a step forward towards a better overall description of silicon. In this paper we present an extensive test of the MEAM-L potential by calculating crystal, liquid and amorphous phase properties, and phase transitions. We also make a comparison with results for existing silicon potentials: Stillinger–Weber, Tersoff, Modified Tersoff (MOD), and the Environment-Dependent Interatomic Potential (EDIP). This comparison shows that, except for the high value of the melting temperature, the MEAM-L potential is superior over other potentials by most of the criteria.

Related Topics
Physical Sciences and Engineering Engineering Computational Mechanics
Authors
, ,