EXTENDED ABSTRACT: Multicomponent materials obtained through alloying, doping, entropy increasing and other strategies provide huge design space for material research and development on the one hand, and on the other hand these materials often have superior performance. Atomic scale computational simulation of multicomponent materials requires exploring a huge space of components, atomic configurations, and crystal structures, which consumes a lot of resources. It is necessary to develop and apply efficient computational simulation methods. This report will first introduce the latest progress in efficient computing methods, and provide an overview of this field. Next, we will introduce our research in efficient calculation and multiscale simulation in two types of materials: transition metal oxygen-family compounds and high-entropy alloys. The compounds formed by transition metals and oxygen family elements have rich crystal structures and properties. Based on the integration of first-principles calculations, monte carlo simulations, cluster expansion, and machine learning, we systematically studied the composition-phase-configuration-property map of transition metal oxygen family compounds. We found that the van der Waals interlayer composition gradient and the local atomic fine configuration (short-range order, etc.) have significant effects on electronic properties. For TiZr-based high entropy alloy system, the integration of first-principles calculations and monte carlo simulations was used to explore the short-range order in these systems and its influence on mechanical properties such as modulus and stacking fault energy. At the same time, the influence and mechanism of adding a small amount of oxygen to the high entropy metal matrix on the atomic structure and mechanical properties of the material were studied. Combined with high-throughput calculations and statistical analysis, the stable local atomic configuration formed by the interaction between oxygen and short order was obtained.
Keywords: Multicomponent materials, Efficient computation, Multiscale simulation, Machine learning
REFERENCES:
[1] T. Zhang, L. Zhu, H. Liu, J. Zhou, Z. Sun, MGE Advances, 1(2), (2023) e7
[2] L. Zhu, J. Zhou, Z. Sun, J. Phys. Chem. Lett., 13 (2022) 3965.
[3] Q. Chen, M. Chen, L. Zhu, N. Miao, J. Zhou, G. Ackland, Z. Sun,, ACS Appl. Mat. Interfaces, 12 (2020) 45184.