Machine learning aided materials design platform for predicting the mechanical properties of Na-ion solid-state electrolytes

TL;DR

This study develops a machine learning model to predict the mechanical properties of Na-ion solid-state electrolytes, aiding rapid materials discovery and optimization for energy storage applications.

Contribution

A novel machine learning regression model trained on extensive data to accurately predict mechanical properties of Na-SSEs, facilitating accelerated materials screening.

Findings

Model achieved R2 scores of 0.72 and 0.87 for shear and bulk modulus.

Predicted properties validated with first principles calculations.

Optimized screening platform reduces prediction uncertainty.

Abstract

Na-ion solid-state electrolytes (Na-SSE) exhibit high potential for electrical energy storage owing to their high energy densities and low manufacturing cost. However, their mechanical properties critical to maintain structural stability at the interface are still insufficiently understood. In this study, a machine learning based regression model was developed for predicting the mechanical properties of Na-SSEs. As a training set, 12,361 materials were obtained from a well-known materials database (Materials Project) and were represented with their respective chemical and structural descriptors. The developed surrogate model exhibited a remarkable accuracy (R2 score) of 0.72 and 0.87, with a mean absolute error of 11.8 GPa and 15.3 GPa for the shear and bulk modulus, respectively. This model was then applied to predict the mechanical properties of 2,432 Na-SSEs, the properties of which have been validated with first principles calculations. Finally, the optimization process was performed to develop an ideal materials screening platform by adding the new minimized dataset, wherein the prediction uncertainty is reduced. We believe that the platform proposed in this study can accelerate the search for Na-SSEs with ideal mechanical properties at minimum cost.