Toward a predictive atomistic model of ion implantation and dopant diffusion in silicon

We review the development and application of kinetic Monte Carlo simulation s to investigate defect and dopant diffusion in ion implanted silicon. In t hese types of Monte Carlo models, defects and dopants are treated at the at omic scale, and move according to reaction rates given as input parameters. These input parameters can be obtained from first principles calculations and/or empirical molecular dynamics (MD) simulations or can be extracted fr om fits to experimental data. Time and length scales differing several orde rs of magnitude can be followed with this method, allowing for direct compa rison with experiments. The different approaches are explained and some res ults presented. (C) 1998 Elsevier Science B.V. All rights reserved.