Fast Kinetic Monte Carlo Simulation of Crystal Growth and Evolution

点击翻译按钮获取中文摘要

Schulze0707443This project concerns both the development of fast KineticMonte-Carlo (KMC) algorithms and their application to simulatingnano- and micron-scale crystal growth. Viewed broadly, KMC isthe numerical task of simulating stochastic models. While thefocus of the project is primarily on crystal growth, KMCsimulations arise in many contexts, including micro-magnetics,ecology, and chemical processes. The algorithms that aredeveloped are applicable to all such models. In this work,however, two new algorithms with single event execution timesindependent of the system size are used to examine dendriticgrowth during solidification. Dendritic growth refers to acommon interfacial instability that occurs during the formationof solid materials -- snow flakes being a familiar example. These instabilities have a number of practical implications formaterials processing and are also of fundamental scientificinterest. Dendrites that form in both pure and compositematerials are studied. This project concerns the development of computer programs thatsimulate the behavior of solid materials on length scales thatare so small that individual atoms are resolved. It thereforehas direct impact on the area of nanotechnology, which is rapidlyemerging as an area of national strategic interest. The programsthat arew developed are the most efficient of all known ways forperforming this particular type of simulation. These simulationssimulate the growth and evolution of solid phase materials byimitating the random motion of individual atoms using millions ofrandom numbers, which themselves are generated by a separatecomputer program. A particular type of solid phase growth from aliquid material that gives rise to a snow-flake like crystalstructure is a principal area of study, but the computer programsthemselves are designed in a way that also makes them useful in amuch broader class of problems. Students are involved in theproject.