Structure of Si(114) determined by global optimization methods

TL;DR

This paper uses advanced global optimization techniques to determine the atomic structure of the Si(114) surface, confirming a model consistent with experimental observations and density functional calculations.

Contribution

It introduces two independent global optimization methods for surface structure determination and validates the most stable model with ab-initio calculations.

Findings

Identified a stable Si(114) surface structure matching experimental data

Validated the structure with density functional theory calculations

Demonstrated effectiveness of parallel-tempering and genetic algorithms in surface modeling

Abstract

In this article we report the results of global structural optimization of the Si(114) surface, which is a stable high-index orientation of silicon. We use two independent procedures recently developed for the determination of surface reconstructions, the parallel-tempering Monte Carlo method and the genetic algorithm. These procedures, coupled with the use of a highly-optimized interatomic potential for silicon, lead to finding a set of possible models for Si(114), whose energies are recalculated with ab-initio density functional methods. The most stable structure obtained here without experimental input coincides with the structure determined from scanning tunneling microscopy experiments and density functional calculations by Erwin, Baski and Whitman [Phys. Rev. Lett. 77, 687 (1996)].