Development of Interatomic Potential for Al-Tb Alloy by Deep Learning
Method
release_okoywvn2o5defexnavtucl5x44
by
L. Tang, Z. J. Yang, T. Q. Wen, K. M. Ho, M. J. Kramer, C. Z. Wang
2020
Abstract
An interatomic potential for Al-Tb alloy around the composition of Al90Tb10
was developed using the deep learning method with DeePMD-kit package. The
atomic configurations and the corresponding total potential energies and forces
on each atom obtained in the ab initio molecular dynamics (AIMD) simulations
are collected to train a deep neural network model to construct the interatomic
potential for Al-Tb alloy. We show the obtained deep neural network model can
well reproduce the energies and forces calculated by AIMD. MD simulations using
the neural network interatomic potential also describe the structural
properties of Al90Tb10 liquid well, such as the partial pair correlation
functions and the bond angle distribution, in comparison with the results from
AIMD. Furthermore, the developed neural network interatomic potential predicts
the formation energies of crystalline phases of Al-Tb system with the same
accuracy as ab initio calculations. The structure factor of Al90Tb10 metallic
glass obtained by MD simulation using the developed neural network interatomic
potential is also in good agreement with the experimental X-ray diffraction
data.
In text/plain
format
Archived Files and Locations
application/pdf 2.3 MB
file_pz3wwvnfxff2vjhoucw7ac5hzu
|
arxiv.org (repository) web.archive.org (webarchive) |
2001.06762v1
access all versions, variants, and formats of this works (eg, pre-prints)