Deep-learning Potential Molecular Dynamics Simulations of the Mechanical Properties of Al-Sc Alloys
Year: 2025
Type: article
Abstract: The deep potential (DP) model based on high-throughput first-principles calculation is expected to solve the problem of potential function scarcity in the molecular dynamics simulation of rare earth aluminium alloys. Firstly, the Al-Sc binary potential function was trained by deep-learning method, and the accuracy of the DP potential function was tested by comparing the interatomic force, energy, energy-volume curve and elastic modulus of DP and DFT. The results show that the interatomic force, energy, energy-volume curve and elastic modulus of precipitated phase predicted by DP are very close to the first-principles calculation results, which proves that the DP potential function developed in this paper has high accuracy. On this basis, the DP potential function was used to study the uniaxial tensile deformation of aluminium single-crystal alloys with different Sc contents. It is found that Al-0.1%Sc alloy has the highest tensile strength, which will provide a reference for the composition optimization of high-performance rare earth Al-Sc alloys.
Source: Chinese Rare Earths
Authors: Juan LI, Hong-tao XUE, Zhi-jun ZHANG, Zhen CHANG, Fu-ling TANG, Jun-qiang REN, Xue-feng LU, Jun-chen LI
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4417019429
FWCI:
Cites:
Cited by:
Related to: