Atomistic simulation methodologies for modelling the nucleation, growth and structure of interfaces

Abstract

There have been many studies applying atomistic simulation techniques to investigate the structure and energetics of surfaces and interfaces. Almost all start by defining the basic structure of the interface, which is then simulated by static or dynamical methods. A different approach is adopted here, where we allow interfacial structures to evolve during the course of the simulation. In particular, three atomistic simulation methodologies for constructing models for thin film interfaces have been developed, including 'atom deposition', where the thin film is 'grown' by sequentially depositing atoms onto a support material to obtain information on nucleation and growth mechanisms; 'layer-by-layer' growth, where monatomic layers of a material are successively deposited on top of a substrate surface; and finally, 'cube-on- cube' whereby the whole of the thin film is placed directly on top of the substrate, before dynamical simulation and energy minimisation. The methodologies developed in this study provide a basis for simulating the nucleation, growth and structure of interface systems ranging from small supported clusters to monolayer and multilayer thin film interfaces. In addition, the layer-by-layer methodology is ideally suited to explore the critical thickness of thin films. We illustrate these techniques with studies on systems with large negative misfits. The calculations suggest that the thin films (initially constrained under tension due to the misfit) relax back to their natural lattice parameter resulting in the formation of surface cracks and island formation. The cube-on-cube methodology was then applied to the SrO/MgO system, which has a large (+ 20%) positive misfit. For this system, the SrO thin film underwent an amorphous transition which, under prolonged dynamical simulation, recrystallised revealing misfit-induced structural modifications, including screw-edge dislocations and low angle lattice rotations.