Regensburg 2010 – scientific programme

Parts | Days | Selection | Search | Downloads | Help

MM: Fachverband Metall- und Materialphysik

MM 26: Poster Session

MM 26.25: Poster

Tuesday, March 23, 2010, 14:45–16:30, Poster C

Automated dislocation detection in atomistic simulations — •Alexander Stukowski, Jonathan Schäfer, and Karsten Albe — FB Materialwissenschaft, TU Darmstadt, Germany

We present a novel method that allows us to directly extract dislocation lines and their associated Burgers vectors from atomistic simulations. Our dislocation detection algorithm is based on an automated Burgers circuit analysis to trace dislocation cores followed by a vectorization step, thereby achieving the transition from the atomistic system to a discrete dislocation representation.

The dislocation analysis can be efficiently performed on-the-fly during molecular dynamics (MD) simulations with a high time resolution. This enables the investigation of rapid dislocation processes, whose observation was otherwise impeded by other crystal defects or simply by the huge amount of data produced by large-scale MD simulations.

Our analysis method can reliably detect perfect, partial, and twinning dislocations. For the first time, the dislocation density in MD simulations can now directly be quantified. In principle, even more comprehensive data not accessible before can be extracted, including dislocation mobility and dislocation reactions.

The new analysis method is applied to large-scale deformation simulations of nanocrystalline metals to study the density and character of dislocations as a function of strain. Being able to directly visualize and measure dislocation activity gives new insights into the interplay of grain-boundary mediated processes and dislocation plasticity in such materials.