GMHEA: A New Software for the Multi-Variable Optimization of High-Entropy Alloy by the Improved Genetic Algorithm and...

Area of Science:

• Materials Science
• Computational Materials Science
• Alloy Design

Background:

• High-entropy alloys (HEAs) offer exceptional properties due to complex compositions.
• Identifying stable HEA structures is challenging due to vast configurational space and high computational costs.
• Existing methods suffer from low search efficiency and inadequate sampling.

Purpose of the Study:

• To introduce GMHEA, a novel software package for multi-variable optimization of HEAs.
• To address limitations of current methods in HEA structure identification.
• To accelerate the design and development of HEAs.

Main Methods:

• Integration of an improved genetic algorithm (GA) with effective medium theory (EMT) for rapid energy calculations.
• Utilized adaptive genetic operators (cut-and-splice pairing, soft mutation, strain mutation) and fingerprint-based structure comparison.
• Incorporated variable-cell relaxation for robust convergence to stable structures.

Main Results:

• GMHEA demonstrates a balance between accuracy and efficiency, with average computation times per atom ranging from 23.13 to 52.19 s.
• Identified HEA configurations exhibit short-range order (SRO) and optimized interatomic distances, correlating with thermodynamic stability.
• Successfully applied to multi-component HEAs like NiₓCuₓPdₓAgₓPtₓAuₓ (x=1-10).

Conclusions:

• GMHEA is a powerful tool for accelerating HEA design and development.
• Enables rapid identification of stable atomic structures, crucial for material performance.
• Has broad implications for applications in catalysis, energy storage, and advanced manufacturing.