From the role of Ar ions in the sputter deposition of Al films to a comprehensive surface surrogate model

The interaction of heavy particles with surfaces in thin film sputter depositions is often studied by means of computer simulations. While forward sputtered particles in comparison to entrapped working gas atoms are assumed to have a predominant effect on the surface state, their individual contribution as well as correlation is scarcely clarified due to their inherent overlay. This issue is addressed in this work by comparing two case studies of sputter deposition of Al thin films in Ar working gas (i) with and (ii) without implanted Ar atoms. The surface interactions are described by a hybrid reactive molecular dynamics/force-bias Monte Carlo approach. The entrapment of Ar atoms is found to facilitate persistent Al interstitials and surface reconstructions. Finally, the results are used to train an artificial neural network (i.e., conditional variational autoencoder) to generalize the findings and propose a surface surrogate model that may be readily coupled to plasma transport simulations.