A general potential for molecular dynamics of ion-sputtered surfaces

TL;DR

This paper develops a new molecular dynamics potential that accounts for many-particle collisions and contaminants, improving the simulation of ion-sputtered surfaces and their self-organization phenomena.

Contribution

A novel general potential for MD simulations that incorporates many-particle perturbations and contaminants, enabling more accurate modeling of surface erosion and self-organization.

Findings

Phase shift in MD equilibrium points varies with contaminants and materials.

Existing potentials are insensitive to environment and contaminant effects.

Surface perturbations are specific to the material's surface environment.

Abstract

Erosion of surface atoms of solid materials by ion bombardment (surface-sputtering) causes nano-ripples and quantum dots to self-organise on the surfaces. The self-organisation had been shown, in some sputtering experiments, to be influenced by unexpected contaminants (ions) from vacuum walls. Existing inter-atomic-interaction potentials of Molecular Dynamics (MD) simulations for studying this are unsuitable because they assume two-particle collisions at a time instead of many (including contaminants)-particle collisions (Wider-area Perturbations, (WP)). We designed this study to develop a suitable potential that incorporates WP of the MD. We developed the general potential to account for the possibility of WP due to contaminants (both foreign and local to the material) consequently shifting the eqiulibrium points of the MD the material. For instance, dynamics of Au and Fe were studied with O bombardments/contamination (oxygenated environments), and those of CSiGe were studied with W, Ti, and O. It was found that the phase-shift, of the eqiulibrium points of the MD simulations, for Au stabilised for just five atoms of oxygen while that of Fe did not, when interacting in separate oxygenated environments. A different phase shift was recorded for CSiGe which reflected the usage of different contaminants, under subsequent bombardments with W, Ti and O, in transformations to CSiGeW, CSiGeWTi, and CSiGeWTiO. These results and those for the sampled superconducting materials, such as SrNdCuO and LiFeAs, showed a sensitivity of the general potential, and insensitivity of the existing potentials, to the type and environment of the atoms of these materials. Furthermore, results obtained clearly show that the MD perturbations in a material, via its surface, is peculiar to the surface.