Machine-learning-guided descriptor selection for predicting corrosion resistance in multi-principal element alloys
More than $270 billion is spent on combatting corrosion annually in the USA alone. This work presents a machine learning-based approach to select corrosion resistant alloys. The focus is on a new class of alloys called high-entropy alloys (HEAs). Given the vast search space due to the variety of compositions and features to be considered, we demonstrate feature optimization to predict the corrosion resistance of any given HEA, based upon the existing corrosion data available on HEAs. Our model reveals that the most important features that determine the corrosion resistance of a given alloy are the pH of the medium, the halide concentration, mixing entropy, atomic % of element with the maximum reduction potential, and a parameter based on atomic radii. Our approach offers a swift tool for screening a vast number of HEAs and selecting the best corrosion resistant HEAs.
Published: January 13, 2023
Roy A., M. Taufique, H. Khakurel, R. Devanathan, D.D. Johnson, and G. Balasubramanian. 2022.Machine-learning-guided descriptor selection for predicting corrosion resistance in multi-principal element alloys.npj Materials Degradation 6, no. 1:Art. No. 9.PNNL-SA-159832.doi:10.1038/s41529-021-00208-y