Thin film processing by means of sputter deposition inherently depends on the interaction of energetic particles with a target surface and the subsequent transport of film forming species through the plasma. The length and time scales of the underlying physical phenomena span orders of magnitudes. A theoretical description which bridges all time and length scales is not practically possible.A unified model approach which describes the dynamics of both the solid and the gas-phase, however, remains desired. In fact, advantage can be taken particularly from the well-separated time scales of the fundamental surface and plasma processes by evaluating both independently. Initially, surface properties may be a priori calculated from a surface model and stored for a number of representative incident particle energy distribution functions. Subsequently, the surface data may be provided to gas-phase transport simulations via appropriate model interfaces (e.g., analytic expressions or look-up tables) and utilized to define insertion boundary conditions. During run-time evaluation, however, the maintained surface data may prove to be not sufficient (e.g., too narrow input data range). In this case, missing data may be obtained by interpolation (common), extrapolation (inaccurate), or be supplied on-demand by the surface model (computationally inefficient).In this work, a potential alternative is established based on machine learning techniques using artificial neural networks. As a proof of concept, a multilayer perceptron network is trained and verified with sputtered particle distributions obtained from transport of ions in matter (TRIM) based simulations for Ar projectiles bombarding a Ti-Al composite. It is demonstrated that the trained network is able to predict the sputtered particle distributions for unknown, arbitrarily shaped incident ion energy distributions. It is consequently argued that the trained network may be readily used as a machine learning based model interface (e.g., by quasi-continuously sampling the desired sputtered particle distributions from the network), which is sufficiently accurate also in scenarios which have not been previously trained.

Compressive stresses in sputter deposited thin films are generally assumed to be caused by forward sputtered (peened) built-in particles and entrapped working gas atoms. While the former are assumed to be predominant, the effect of the latter on interaction dynamics and thin film properties is scarcely clarified (concurrent or causative). The overlay of the ion bombardment induced processes renders an isolation of their contribution impracticable. This issue is addressed by two molecular dynamics case studies considering the sputter deposition of Al thin films in Ar working gas. First, Ar atoms are fully retained. Second, they are artificially neglected, as implanted Ar atoms are assumed to outgas anyhow and not alter the ongoing dynamics significantly. Both case studies share common particle dose impinging Al(001) surfaces. Ion energies from 3 to 300 eV and [Formula: see text] flux ratios from 0 to 1 are considered. The surface interactions are simulated by hybrid reactive molecular dynamics/force-biased Monte Carlo simulations and characterized in terms of mass density, Ar concentration, biaxial stress, shear stress, ring statistical connectivity profile, Ar gas porosity, Al vacancy density, and root-mean-squared roughness. Implanted Ar atoms are found to form subnanometer sized eventually outgassing clusters for ion energies exceeding 100 eV. They fundamentally govern a variety of surface processes (e.g., forward sputtering/peening) and surface properties (e.g., compressive stresses) in the considered operating regime.

Simulations of thin film sputter deposition require the separation of the plasma and material transport in the gas phase from the growth/sputtering processes at the bounding surfaces (e.g., substrate and target). Interface models based on analytic expressions or look-up tables inherently restrict this complex interaction to a bare minimum. A machine learning model has recently been shown to overcome this remedy for Ar ions bombarding a Ti-Al composite target. However, the chosen network structure (i.e., a multilayer perceptron, MLP) provides approximately 4×106 degrees of freedom, which bears the risk of overfitting the relevant dynamics and complicating the model to an unreliable extent. This work proposes a conceptually more sophisticated but parameterwise simplified regression artificial neural network for an extended scenario, considering a variable instead of a single fixed Ti-Al stoichiometry. A convolutional β-variational autoencoder is trained to reduce the high-dimensional energy-angular distribution of sputtered particles to a low-dimensional latent representation with only two components. In addition to a primary decoder that is trained to reconstruct the input energy-angular distribution, a secondary decoder is employed to reconstruct the mean energy of incident Ar ions as well as the present Ti-Al composition. The mutual latent space is hence conditioned on these quantities. The trained primary decoder of the variational autoencoder network is subsequently transferred to a regression network, for which only the mapping to the particular low-dimensional space has to be learned. While obtaining a competitive performance, the number of degrees of freedom is drastically reduced to 15 111 (0.378% of the MLP) and 486 (0.012% of the MLP) parameters for the primary decoder and the remaining regression network, respectively. The underlying methodology is very general and can easily be extended to more complex physical descriptions (e.g., taking into account dynamical surface properties) with a minimal amount of data required.