Place: Seminar of Institute of High-Performance Computing, Singapore
Due to the availability of powerful parallel computers, efficient algorithms and realistic material descriptions, the numerical modeling of material behavior under various conditions plays an increasingly important role in research, development and processing of advanced materials. In particular nanostructured materials, like nanocrystals or nanocomposites, have drawn a lot of attention, because of their unique mechanical and functional properties. One common feature of all nanostructured materials is that internal interfaces, like grain boundaries or phase boundaries, dominate their observable, i.e. macroscopic behavior. Two examples for multiscale simulations will be given that demonstrate two different approaches: In the first one, information about the strength of interfaces is first calculated by quantum mechanical density functional theory (DFT) methods and then transferred to a continuum mechanical simulation scheme. In this scheme interfaces are described by the cohesive zone model and plasticity is modeled on the level of individual dislocations in an elastic continuum. With this hierarchical scale bridging scheme results are gained in form of macroscopic observables, like the fracture toughness of layered structures or the fracture toughness of polycrystals. The second approach uses large-scale atomistic simulations with several million atoms to calculate quantities that can immediately be compared with experiments. This is demonstrated for the example of nanoindentation in the presence of grain boundaries, where results of atomistic simulation can immediately be compared with experimental data. This approach is also shown to be useful when investigating the properties of metallic nanocomposites.