Exploring the mechanism of plastic deformation in BCC Mg-Li-Al alloys via Machine learning Molecular dynamics simulations

Gaoshang Zhang,Wei Mei,Wentao Zhang,Kuang Yu
DOI: https://doi.org/10.1016/j.commatsci.2024.113396
IF: 3.572
2024-09-27
Computational Materials Science
Abstract:Alloys stand at the forefront of materials science, featuring superior mechanical properties vital for diverse industrial applications. However, understanding the intricate plastic behavior of alloys still remains a challenge. This study delves into the plastic deformation mechanisms of body-centered cubic (BCC) Mg-Li-Al alloys, which is an important ultra-light structural material. An accurate machine learning potential energy surface is constructed based on first-principle calculations, enabling us to observe the plastic deformation of both single crystal and polycrystalline Mg-Li-Al at atomistic scale. Our simulation unveils a complex and unique BCC-FCC-BCC-HCP martensitic phase transition in single crystal Mg-14Li-3Al alloy, giving rise to a strong twin strengthening effect under large pulling strain. Such effect is absent in polycrystalline Mg-Li-Al, and can potentially be exploited to develop new ultra-strong light-weight alloys. For polycrystalline Mg-Li-Al system, we systematically predict its plastic strengths in different composition ratios, which provides a useful guide for its further optimization. Through this study, we show that facilitated by machine learning potential energy surface, it is now much more convenient to understand the plastic behaviors at microscopic level, thus opening new gates to the design of alloy materials.
materials science, multidisciplinary
What problem does this paper attempt to address?