The TiZrHfNb-based Bcc refractory high entropy alloy shows excellent strength and toughness matching, low density (compared with Ni-Co based superalloy) and high temperature softening resistance, which is expected to be applied in aerospace, high temperature mold and ultra-high temperature coating fields. Combining high-throughput computations and machine learning, rapid screening design and performance prediction and optimization are realized in the composition space of TiZrHfNb-based four-element, six-element and seven-element refractory high-entropy alloys, speeding up the research and development of high temperature structural materials with excellent properties, shortening the development time and reducing the calculation cost.

High-throughput computations and machine learning driven design of refractory high-entropy alloys

The high-efficiency preliminary screening of refractory high-entropy alloys is carried out based on high-throughput thermodynamic calculations. The key material parameters of refractory high-entropy alloys are designed based on first principal calculation. Composition design and performance optimization of TiZrHfNb-based four-element, six-element and seven-element refractory high entropy alloys driven by machine learning. The research results form a complete set of high-throughput computations and composition screening techniques for refractory high-entropy alloys, and provide theoretical guidance for obtaining alternative components with matching strength and toughness properties. Based on high-throughput computations technology and machine learning methods, it can provide a theoretical design basis for the development and progress of high-temperature materials, and accelerate the development of new high-temperature materials for aerospace.