Elemental Diffusion Coefficient Prediction in Conventional Alloys Using Machine Learning

Arjun S. Kulathuvayal,Yi Rao,Yanqing Su
DOI: https://doi.org/10.1063/5.0222001
2024-01-01
Chemical Physics Reviews
Abstract:This paper presents the Machine Learned Diffusion Coefficient Estimator, a comprehensive machine learning framework designed to predict diffusion coefficients in impure metallic (IM) and multi-component alloy (MCA) media. The framework incorporates five machine learning models, each tailored to specific diffusion modes: (1) impurity and (2) self-diffusion in IM media, and (3) self, (4) impurity, and (5) chemical diffusion in MCA media. These models use statistical aggregations of atomic descriptors for both the diffusing elements and the diffusion media, along with the temperature of the diffusion process, as features. Models are trained using the random forest and deep neural network algorithms, with performance evaluated through the coefficient of determination (R-2), mean squared error (MSE), and uncertainty estimates. The models within this framework achieve an impressive R-2 score above 0.90 with MSE less than 10(-16) m(2)/s, demonstrating high predictive accuracy and reliability for diffusion coefficient.
What problem does this paper attempt to address?