Machine learning-assisted discovery of Cr, Al-containing high-entropy alloys for high oxidation resistance

A Machine Learning (ML) integrated workflow was utilized to guide the design of Cr, Al-containing five-element high-entropy alloys (HEAs) for achieving an enhanced high-temperature oxidation resistance. ML directs the design of HEAs to a chemical composition consisting of Fe, Cr, Al, Ni, and Cu for enhanced oxidation resistance. The oxidation behavior of AlxCrCuFeNi (x = 0, 0.25, 0.5, 1) HEAs at 1100 degrees C in air was systematically inves-tigated and the oxidation mechanism was elucidated. The experimental validation agrees well with the ML prediction, demonstrating that ML could be used as a powerful tool for designing alloys with optimized oxidation resistance.

Automated summary

A machine learning integrated workflow was used to guide the design of Cr, Al-containing five-element high-entropy alloys (HEAs) with enhanced high-temperature oxidation resistance. ML directed the design of HEAs with a chemical composition of Fe, Cr, Al, Ni, and Cu for improved oxidation resistance. The oxidation behavior of AlxCrCuFeNi (x = 0, 0.25, 0.5, 1) HEAs at 1100 degrees C in air was systematically studied and the oxidation mechanism was elucidated. Experimental validation agreed well with the ML prediction, showing that ML can be a powerful tool for designing alloys with optimized oxidation resistance.