Predicting densities and elastic moduli of SiO2-based glasses by machine learning

• Published in: npj computational materials Vol. 6; no. 1
• Main Authors: Hu, Yong-Jie, Zhao, Ge, Zhang, Mingfei, Bin, Bin, Del Rose, Tyler, Zhao, Qian, Zu, Qun, Chen, Yang, Sun, Xuekun, de Jong, Maarten, Qi, Liang
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 20.03.2020; Nature Publishing Group
• Subjects: 639/301/1023/218; 639/301/1034; Characterization and Evaluation of Materials; Chemical bonds; Chemistry and Materials Science; Composition; Computational Intelligence; Computer simulation; Density; Elastic properties; Glass; Learning algorithms; Machine learning; Materials Science; Mathematical and Computational Engineering; Mathematical and Computational Physics; Mathematical Modeling and Industrial Mathematics; Mathematical models; Modulus of elasticity; Molecular dynamics; Silicon dioxide; Statistical methods; Statistical models; Theoretical; Training
• ISSN: 2057-3960; 2057-3960
• DOI: 10.1038/s41524-020-0291-z

Chemical design of SiO 2 -based glasses with high elastic moduli and low weight 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 interatomic bonds and their ordering at different length scales. Here we show that the densities and elastic moduli of SiO 2 -based glasses can be efficiently predicted by machine learning (ML) techniques across a complex compositional space with multiple (>10) types of additive oxides besides SiO 2 . Our machine learning approach relies on a training set generated by high-throughput molecular dynamic (MD) simulations, a set of elaborately constructed descriptors that bridges the empirical statistical modeling with the fundamental physics of interatomic bonding, and a statistical learning/predicting model developed by implementing least absolute shrinkage and selection operator with a gradient boost machine (GBM-LASSO). The predictions of the ML model are comprehensively compared and validated with a large amount of both simulation and experimental data. By just training with a dataset only composed of binary and ternary glass samples, our model shows very promising capabilities to predict the density and elastic moduli for k-nary SiO 2 -based glasses beyond the training set. As an example of its potential applications, our GBM-LASSO model was used to perform a rapid and low-cost screening of many (~10 5 ) compositions of a multicomponent glass system to construct a compositional-property database that allows for a fruitful overview on the glass density and elastic properties.