Study of Strain and Temperature Dependence of Metal Epitaxy

TL;DR

This paper investigates how strain and temperature influence metal epitaxy by combining density functional theory and kinetic Monte Carlo simulations to predict the evolution of metallic crystal structures.

Contribution

It introduces a methodology integrating DFT and KMC to accurately simulate metal growth processes considering strain and temperature effects.

Findings

DFT provides detailed diffusion properties on metallic surfaces.

Strain significantly affects adatom diffusion energy barriers.

Ab initio KMC simulations can predict mesoscopic growth structures.

Abstract

Metallic films are important in catalysis, magneto-optic storage media, and interconnects in microelectronics, and it is crucial to predict and control their morphologies. The evolution of a growing crystal is determined by the behavior of each individual atom, but technologically relevant structures have to be described on a time scale of the order of (at least) tenths of a second and on a length scale of nanometers. An adequate theory of growth should describe the atomistic level on very short time scales (femtoseconds), the formation of small islands (microseconds), as well as the evolution of mesoscopic and macroscopic structures (tenths of seconds). The development of efficient algorithms combined with the availability of cheaper and faster computers has turned density functional theory (DFT) into a reliable and feasible tool to study the microscopic aspects of growth phenomena (and many other complex processes in materials science, condensed matter physics, and chemistry). In this paper some DFT results for diffusion properties on metallic surfaces are presented. Particularly, we will discuss the current understanding of the influences of strain on the diffusion (energy barrier and prefactor) of a single adatom on a substrate. A DFT total energy calculation by its nature is primarily a static calculation. An accurate way to describe the spatial and temporal development of a growing crystal is given by kinetic Monte Carlo (KMC). We will describe the method and its combination with microscopic parameters obtained from ab initio calculations. It is shown that realistic ab initio kinetic Monte Carlo simulations are able to predict an evolving mesoscopic structure on the basis of microscopic details.