Research progress in high-entropy alloys driven by high throughput computation and machine learning

Abstract

High-entropy alloys have attracted great attention in various fields due to their high-entropy effect, severe lattice distortion, slow diffusion and special and excellent material performance due to the combination of various alloying elements in equal or near-equal molar proportions. Its high strength and hardness, fatigue resistance, excellent corrosion resistance, radiation resistance, near-zero thermal expansion coefficient, catalytic response, thermoelectric response and photoelectric conversion make high-entropy alloys have potential applications in many aspects. High-throughput computation and machine learning technology have rapidly become powerful tools to explore the huge composition space of high-entropy alloys and comprehensively predict material properties. The basic concepts of high-throughput computing and machine learning were introduced in this paper as well as the advantages of first-principles calculation, thermodynamic/kinetic calculation and machine learning in the research of high-entropy alloys. The application research status of high-entropy alloy composition screening, phase and microstructure calculations and performance prediction were summarized. In the final part, the existing problems, and the solutions and future prospects of this field were summarized, including developing tools for first-principles calculations and machine learning of high-entropy alloys, building high-quality databases for high-entropy alloys and integrating high-throughput computing with machine learning to globally optimize the mechanical property and service performance of high-entropy alloys.

Keywords

• high-entropy alloy
• thermodynamic
• first principles
• machine learning
• performance optimization