[PO-D2] Poster Session 2

[P2-30] Optimizing elastic moduli of the silicate glasses through high-throughput atomistic modeling and machine learning techniques

Yong-Jie Hu¹, Ge Zhao², Tyler Del Rose¹, ^○Liang Qi¹ (1.Dept. of Materials Science and Engineering, University of Michigan, United States of America, 2.Department of Statistics, Pennsylvania State University, United States of America)

Chemical design of the silicate glass with high elastic moduli is of great interest. However, it is difficult to find a universal expression to predict the elastic moduli according to the glass composition before synthesis since the elastic moduli are a complex function of several material properties at different length scales. This work presents a computational framework to efficiently predict the elastic moduli of the silicate glass across a multicomponent compositional design space, including 11 types of additive oxides, by integrations of high-throughput molecular dynamic (MD) calculations and machine learning (ML) techniques. Our newly developed ML model can predict the elastic moduli for k-nary silicate glass systems, using the learning datasets generated from MD calculations for only binary and ternary systems. The usefulness of our model is illustrated by identifying the most relevant materials descriptors that determine the elastic moduli and screening for the silicate glasses with high stiffness.