Glass transition temperature (Tg) is important for understanding the physical and mechanical properties of a polymer material because it relates to the thermal energy required to transition between a hard glassy state and a soft rubbery one. Over the years, various models have been developed for predicting this thermal property from molecular structure to aid in designing novel polymers in selected classes. This work builds on those efforts by utilizing both machine learning (ML) and quantum chemistry (QC) techniques to develop models that can predict Tg values from the molecular structure under different data availability scenarios and for a wide variety of polymer types. For the ML model, a graph convolutional network (GCN) was used to map topological polymer features; this model was trained against a dataset of more than 7500 Tg values and resulted in a root mean square error (RMSE) of 38.1 °C. The QC-based regression model was trained on 83 Tg values and produced an RMSE of 34.5 °C. In conclusion, this work demonstrated that while both model techniques produce accurate predictions and are suitable for different data availability scenarios, the QC-based regression model offered a more interpretable model framework with significantly less training data.
Hickey, Kevin, et al. "Applying machine learning and quantum chemistry to predict the glass transition temperatures of polymers." Computational Materials Science, vol. 238, Mar. 2024. https://doi.org/10.1016/j.commatsci.2024.112933
@article{osti_2328569,
author = {Hickey, Kevin and Feinstein, Jeremy and Sivaraman, Ganesh and MacDonell, Margaret and Yan, Eugene and Matherson, Carlos and Coia, Scott and Xu, Jason and Picel, Kurt},
title = {Applying machine learning and quantum chemistry to predict the glass transition temperatures of polymers},
annote = {Glass transition temperature (Tg) is important for understanding the physical and mechanical properties of a polymer material because it relates to the thermal energy required to transition between a hard glassy state and a soft rubbery one. Over the years, various models have been developed for predicting this thermal property from molecular structure to aid in designing novel polymers in selected classes. This work builds on those efforts by utilizing both machine learning (ML) and quantum chemistry (QC) techniques to develop models that can predict Tg values from the molecular structure under different data availability scenarios and for a wide variety of polymer types. For the ML model, a graph convolutional network (GCN) was used to map topological polymer features; this model was trained against a dataset of more than 7500 Tg values and resulted in a root mean square error (RMSE) of 38.1 °C. The QC-based regression model was trained on 83 Tg values and produced an RMSE of 34.5 °C. In conclusion, this work demonstrated that while both model techniques produce accurate predictions and are suitable for different data availability scenarios, the QC-based regression model offered a more interpretable model framework with significantly less training data.},
doi = {10.1016/j.commatsci.2024.112933},
url = {https://www.osti.gov/biblio/2328569},
journal = {Computational Materials Science},
issn = {ISSN 0927-0256},
volume = {238},
place = {United States},
publisher = {Elsevier},
year = {2024},
month = {03}}
Argonne National Laboratory (ANL); IL (United States)
Sponsoring Organization:
USDOE; USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE Office of Energy Efficiency and Renewable Energy (EERE), Office of Sustainable Transportation. Bioenergy Technologies Office (BETO); USDOE Office of Science (SC)