Machine learning glass transition temperature of polymethacrylates

Abstract

The glass transition temperature, Tg, is an important thermophysical property for polymethacrylates, which can be difficult to determine experimentally. Data-driven modeling approaches provide alternative methods to predict Tg in a rapid and robust way. Here, we develop the Gaussian process regression model to shed light on the relationship between quantum chemical descriptors and the glass transition temperature for the polymethacrylate. A total of 37 samples with the glass transition temperature ranging from 203 K to 428 K are examined. The model is highly stable and accurate that contributes to fast and low-cost estimations of the glass transition temperature.

Topics & Concepts