Parts | Days | Selection | Search | Updates | Downloads | Help

MM: Fachverband Metall- und Materialphysik

MM 10: Poster Session 1

MM 10.23: Poster

Monday, September 5, 2022, 18:00–20:00, P2

Al-Ge solid solubility prediction using machine-learned forcefield potentials and phonon calculations — •Ondřej Fikar and Martin Zelený — Institute of Materials Science and Engineering, Faculty of Mechanical Engineering, Brno University of Technology, Brno, Czech Republic

This work is focused on a theoretical study of the phase stability of Al rich solid solution in Al-Ge alloy. The solubilities of the solid solution were first determined using temperature-dependent free energies of pure elements and solid solutions of various chemical compositions obtained from ab initio calculations based on density functional theory. All total energy calculations were performed by Vienna Ab initio Simulation Package (VASP) with the help of Projector Augmented-wave potentials. Contributions of vibrational free energy and electron free energy were obtained from Phonopy package. Subsequently, a forcefield potential for Al-Ge alloy using machine learning routines as implemented in the VASP package was created. The trained forcefield potential was then used to again carry out phonon calculations. The results were compared to the previous phonon calculations carried out without machine learning.