Multiscale simulations of plasticity and fracture: the atomic-scale mechanisms of hydrogen embrittlement in engineering alloys.

点击翻译按钮获取中文摘要

Molecular dynamics (MD) and atomistic simulations have the potential to impact areas ranging from nanotechnology, to drug design, to the next generation of advanced materials. However, the predictive capabilities of MD are limited in several ways. The two most significant limitations are the prohibitively small numbers of atoms that can be studied and the limited availability (or sometimes complete lack of availability) of accurate interatomic potentials for many important combinations of atomic species. This research pursues the development of methodologies, algorithms and tools that help to overcome these obstacles. At the same time, we will use these tools to study important scientific questions related to the mechanical failure of materials at the nanoscale. These questions have consequences for the design of materials with enhanced properties, for our understanding of nanostructures, and for the fundamentals of fracture and failure in structural materials.**Understanding and predicting fracture is, of course, of fundamental importance to engineering. The more accurate and reliable our understanding of the process, the better equipped we are to confidently design lightweight, long-serving structures that will not fail suddenly and catastrophically. This research focusses on applying atomic scale models to advance our understanding of fundamental aspects of plasticity and fracture in structural engineering metals, specifically addressing the important role that hydrogen plays in making certain important metals more brittle. For example, we will address hydrogen embrittlement effects in aluminum alloys, and the role of hydrides (precipitated hydrogen compounds) in the fracture behaviour of zirconium alloys. The research methodology will use computer simulation techniques that have been developed by the PI's research group and other researchers. For example, we will use multiscale methods that extend the size of the simulation that can be run with atomistic accuracy. This is accomplished through judicious choice of which regions of the problem are treated using a fully atomistic description, allowing us to study hydrogen-aluminum interactions within realistic configurations of atoms around a stressed crack tip. As well, we will employ computational methods to explore atomic configurations and accurately determine the activation energy associated with key processes like dislocation nucleation. Accurate knowledge of such data is of fundamental importance to any larger scale models aiming to predict fracture toughness. Finally, the project will both utilize and build upon the recent openKIM.org project to develop new accurate models to describe the zirconium-hydrogen system. This will allow us to study the complex interactions between cracks and embedded hydrides.