1. Development of interatomic potential for Al-Tb alloys using a deep neural network learning method
- Author
-
Matthew J. Kramer, Ze-Jin Yang, Kai-Ming Ho, L. Tang, Tongqi Wen, and Cai-Zhuang Wang
- Subjects
Diffraction ,Materials science ,Alloy ,Computer Science::Neural and Evolutionary Computation ,General Physics and Astronomy ,FOS: Physical sciences ,Interatomic potential ,02 engineering and technology ,engineering.material ,01 natural sciences ,Molecular physics ,Molecular dynamics ,Condensed Matter::Materials Science ,Ab initio quantum chemistry methods ,0103 physical sciences ,Atom ,Physics::Atomic and Molecular Clusters ,Physical and Theoretical Chemistry ,010306 general physics ,Condensed Matter - Materials Science ,Artificial neural network ,Materials Science (cond-mat.mtrl-sci) ,Disordered Systems and Neural Networks (cond-mat.dis-nn) ,Condensed Matter - Disordered Systems and Neural Networks ,Computational Physics (physics.comp-ph) ,021001 nanoscience & nanotechnology ,Molecular geometry ,engineering ,0210 nano-technology ,Physics - Computational Physics - Abstract
An interatomic potential for the Al–Tb alloy around the composition of Al90Tb10 is developed using the deep neural network (DNN) learning method. The atomic configurations and the corresponding total potential energies and forces on each atom obtained from ab initio molecular dynamics (AIMD) simulations are collected to train a DNN model to construct the interatomic potential for the Al–Tb alloy. We show that the obtained DNN model can well reproduce the energies and forces calculated by AIMD simulations. Molecular dynamics (MD) simulations using the DNN interatomic potential also accurately describe the structural properties of the Al90Tb10 liquid, such as partial pair correlation functions (PPCFs) and bond angle distributions, in comparison with the results from AIMD simulations. Furthermore, the developed DNN interatomic potential predicts the formation energies of the crystalline phases of the Al–Tb system with an accuracy comparable to ab initio calculations. The structure factors of the Al90Tb10 metallic liquid and glass obtained by MD simulations using the developed DNN interatomic potential are also in good agreement with the experimental X-ray diffraction data. The development of short-range order (SRO) in the Al90Tb10 liquid and the undercooled liquid is also analyzed and three dominant SROs, i.e., Al-centered distorted icosahedron (DISICO) and Tb-centered ‘3661’ and ‘15551’ clusters, respectively, are identified.
- Published
- 2020