Temperature induced metallicity of the Si(001) surface: Insights from molecular dynamics simulations with machine learned interatomic potentials

Abstract

The temperature-dependent structural dynamics of Si dimers on the reconstructed Si(001) surface are investigated using molecular dynamics simulations with a newly developed machine learned interatomic potential trained on a database derived from density functional theory-based calculations. We find a finite probability of dimers occupying the higher energy symmetric configurations (associated with metallic behavior) even at temperatures as low as 300 K and that this probability continues to grow with increasing temperature as the lower energy asymmetric dimers flip rapidly with rates ranging from 106 to 109 s−1. Furthermore, above 700 K, some dimers are found to dissociate, leading to the presence of Si adatoms on the surface. These results are in accord with the experimental observations of metallicity on Si(001) with an onset around 400 K, which increases with increasing temperature followed by an abrupt rise around 700 K. By capturing the dynamics of dimer flipping, which on average reflect a domination of the symmetric configuration over the ground state asymmetric geometry, these simulations provide a rationale for the origin of the observed metallicity.