Kinetic Monte Carlo Simulation of Nanoalloy Crystal Growth

University Of Tennessee Knoxville, Knoxville TN

Investigators

Abstract

The goal of this research project is to develop realistic, accurate, and efficient computer-simulation techniques for modeling alloy materials. The project focuses on simulation of the growth of nanoclusters, especially alloys, ranging in size from a few hundred to a few million atoms. Interest in nanoclusters is broad, arising in large part due to the tunable nature of their properties. The morphology, structure, and stability, along with the thermal, electronic, and chemical properties of clusters can vary dramatically with size, composition, and atomic ordering. Often, there are "magic sizes" that exhibit particularly unusual structures and properties. In some cases, the addition of as little as a single dopant atom can create a dramatically different material. Applications are diverse, including optics, catalysis, nanoelectronics, and biomedicine. While there is a large body of work on nanoclusters, the techniques under development in this project are new and uniquely suited to simulating the growth and evolution of atomistic-scale structures. The successful development of these techniques will fill a significant gap in existing simulation methods. This project focuses on kinetic Monte Carlo (KMC) simulation of the growth of nanoclusters. Essentially all of the work on nanocluster simulation to date is based on either global optimization (the search for ground states) or equilibrium Monte Carlo simulations. In contrast, KMC simulation is designed for following the dynamics of nonequilibrium processes and is therefore uniquely suited to simulating the growth and evolution of atomistic-scale structures. KMC methods fall into two broad categories: lattice and off-lattice, both of which will be used in this work. Lattice-based models are much simpler and orders of magnitude faster, allowing them to work with much larger systems for longer simulation times. This type of simulation will be used to study the growth of dendritic clusters containing several million atoms. This portion of the project also contains a significant modeling component, the main thrust of which is to derive general results for the way surface energy and equilibrium crystal shapes are determined by the formulation of various growth models. These results will then be used to identify parameter values and model configurations that are likely to yield interesting dendrite morphologies during the subsequent simulation of the growth. The off-lattice portion of the project will be focused on the growth of smaller clusters, containing hundreds or perhaps thousands of atoms. The main focus here will be on developing faster methods. The testing ground for these new methods will be simulating the growth of alloy nanoparticles with core-shell structures. This can be viewed as the nanoparticle analog of heteroepitaxial thin films.