Regensburg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 6: Computational Materials Modelling: Defects / Alloys
MM 6.5: Vortrag
Montag, 5. September 2022, 16:45–17:00, H44
An efficient method to access the grain boundary parameter space with atomistic simulations — •Timo Schmalofski1, Martin Kroll2, Rebecca Janisch1, and Holger Dette2 — 1ICAMS, Ruhr-University Bochum, 44780 Bochum, Germany — 2Department of Mathematics, Ruhr-University Bochum, 44780 Bochum, Germany
A grain boundary (GB) is a two dimensional defect in solids with significant influence on different material properties. It describes the interface between two grains with different orientations and is thus defined by five macroscopic degrees of freedom (DOF), 2 from the rotation axis, 1 from the misorientation angle and 2 from the grain boundary normal vector. The GB energy as a function of the DOF can be obtained e.g. by atomistic simulations. However, a systematic sampling of the 5D grain boundary parameter space, or even lower-dimensional subspaces of it, comes with several challenges. To overcome them, a sampling method is needed, which only needs a small number of data points and can automatically find the cusps (deep minima) in the energy while sampling. Recently we introduced a sequential sampling technique which fulfills both [1] in the 1D subspace of symmetrical tilt grain boundaries. Now this sequential sampling technique will be evaluated for a 2D analysis of the energy as a function of GB plane inclination for fixed misorientations. [1] Kroll, M., Schmalofski, T., Dette, H. and Janisch, R. (2022), Efficient Prediction of Grain Boundary Energies from Atomistic Simulations via Sequential Design. Adv. Theory Simul. 2100615.