Machine Learning-assisted Prediction of Polymer Glass Transition Temperature: A Structural Feature Approach

Bardia Afsordeh; Hadi Shirali

doi:10.1007/s10118-025-3361-3

Your Location：

Home >

Browse articles >

Machine Learning-assisted Prediction of Polymer Glass Transition Temperature: A Structural Feature Approach

RESEARCH ARTICLE | Updated：2025-08-25

- Machine Learning-assisted Prediction of Polymer Glass Transition Temperature: A Structural Feature Approach
- Chinese Journal of Polymer Science Vol. 43, Issue 9, Pages: 1661-1670(2025)
- Affiliations：
  
  a.Department of Polymer Engineering and Color Technology, Amirkabir University of Technology, Tehran, Iran
  b.Department of Polymer Engineering and Color Technology, Amirkabir University of Technology, PO Box 15875-441, Tehran, Iran
- Author bio：
  
  hadishirali@aut.ac.ir
- Funds：
- DOI：10.1007/s10118-025-3361-3
  CLC：
- Received：11 March 2025，
  
  Revised：17 April 2025，
  
  Accepted：17 April 2025，
  
  Published Online：01 July 2025，
  
  Published：05 September 2025
- Accepted：
Scan QR Code
Afsordeh, B.; Shirali, H. Machine learning-assisted prediction of polymer glass transition temperature: a structural feature approach. Chinese J. Polym. Sci. 2025, 43, 1661–1670

Bardia Afsordeh, Hadi Shirali. Machine Learning-assisted Prediction of Polymer Glass Transition Temperature: A Structural Feature Approach[J]. Chinese journal of polymer science, 2025, 43(9): 1661-1670.
Afsordeh, B.; Shirali, H. Machine learning-assisted prediction of polymer glass transition temperature: a structural feature approach. Chinese J. Polym. Sci. 2025, 43, 1661–1670 DOI： 10.1007/s10118-025-3361-3.

Bardia Afsordeh, Hadi Shirali. Machine Learning-assisted Prediction of Polymer Glass Transition Temperature: A Structural Feature Approach[J]. Chinese journal of polymer science, 2025, 43(9): 1661-1670. DOI： 10.1007/s10118-025-3361-3.

摘要

This study utilizes machine learning to predict the glass transition temperature (

) of polymers using a dataset of polymer structures. Key descriptors like flexibility and side chain occupancy were analyzed

with Extra Trees and Gaussian Process Regression achieving the best accuracy

highlighting the potential of data-driven appro

aches in polymer science.

Abstract

Machine learning (ML) has emerged as a powerful tool for predicting polymer properties

including glass transition temperature (

)

which is a critical factor influencing polymer applications. In this study

a dataset of polymer structures and their

values were created and represented as adjacency matrices based on molecular graph theory. Four key structural descriptors

flexibility

side chain occupancy length

polarity

and hydrogen bonding capacity

were extracted and used as inputs for ML models: Extra Trees (ET)

Random Forest (RF)

Gaussian Process Regression (GPR)

and Gradient Boosting (GB). Among these

ET and GPR achieved the highest predictive performance

with

values of 0.97

and mean absolute errors (MAE) of approximately 7–7.5 K. The use of these extracted features significantly improved the prediction accuracy compared to previous studies. Feature importance analysis revealed that flexibility had the strongest influence on

followed by side-chain occupancy length

hydrogen bonding

and polarity. This work demonstrates the potential of data-driven approaches in polymer science

providing a fast and reliable method for

prediction that does not require experimental inputs.

关键词

Keywords

references

Calori, I. R.; Braga, G.; de Jesus, P. D. C. C.; Bi, H.; Tedesco, A. C. Polymer scaffolds as drug delivery systems. Eur. Polym. J. 2020 , 129 , 109621..

Fan, J.; Weerheijm, J.; Sluys, L. Dynamic compressive mechanical response of a soft polymer material. Mater. Des. 2015 , 79 , 73−85..

Fan, J.; Weerheijm, J.; Sluys, L. High-strain-rate tensile mechanical response of a polyurethane elastomeric material. Polymer 2015 , 65 , 72−80..

Fan, J.; Wang, C. Dynamic compressive response of a developed polymer composite at different strain rates. Compos. B Eng. 2018 , 152 , 96−101..

Gibbs, J. H.; DiMarzio, E. A. Nature of the glass transition and the glassy state. J. Chem. Phys. 1958 , 28 , 373−383..

Flory, P. J. Statistical thermodynamics of semi-flexible chain molecules. Proc. R. Soc. Lond. Ser. A 1956 , 234 , 60−73..

Kauzmann, W. The nature of the glassy state and the behavior of liquids at low temperatures. Chem Rev. 1948 , 43 , 219−256..

Bikiaris, D.; Prinos, J.; Botev, M.; Betchev, C.; Panayiotou, C. Blends of polymers with similar glass transition temperatures: a DMTA and DSC study. J. Appl. Polym. Sci. 2004 , 93 , 726−735..

Alves, N.; Mano, J. F.; Balaguer, E.; Dueñas, J. M.; Ribelles, J. G. Glass transition and structural relaxation in semi-crystalline poly(ethylene terephthalate): a DSC study. Polymer 2002 , 43 , 4111−4122..

Gibout, S.; Franquet, E.; Haillot, D.; Bédécarrats, J.-P.; Dumas, J.-P. Challenges of the usual graphical methods used to characterize phase change materials by differential scanning calorimetry. Appl Sci. 2018 , 8 , 66..

Gusenkov, G.; Krestov, G. Problems of non-linear theory in differential scanning calorimetry. J. Therm. Anal. 1978 , 13 , 369−386..

Van Der Sman, R. Predictions of glass transition temperature for hydrogen bonding biomaterials. J. Phys. Chem. B 2013 , 117 , 16303−16313..

Eom, Y.; Kim, S.-M.; Lee, M.; Jeon, H.; Park, J.; Lee, E. S.; Hwang, S. Y.; Park, J.; Oh, D. X. Mechano-responsive hydrogen-bonding array of thermoplastic polyurethane elastomer captures both strength and self-healing. Nat. Commun. 2021 , 12 , 621..

Lugger, S. J.; Houben, S. J.; Foelen, Y.; Debije, M. G.; Schenning, A. P.; Mulder, D. J. Hydrogen-bonded supramolecular liquid crystal polymers: smart materials with stimuli-responsive, self-healing, and recyclable properties. Chem. Rev. 2021 , 122 , 4946−4975..

Bocharova, V.; Wojnarowska, Z.; Cao, P. F.; Fu, Y.; Kumar, R.; Li, B.; Novikov, V.; Zhao, S.; Kisliuk, A.; Saito, T. Influence of chain rigidity and dielectric constant on the glass transition temperature in polymerized ionic liquids. J. Phys. Chem. B 2017 , 121 , 11511−11519..

Yin, H.; Yang, B.; Chua, Y. Z.; Szymoniak, P.; Carta, M.; Malpass-Evans, R.; McKeown, N. B.; Harrison, W. J.; Budd, P. M.; Schick, C. Effect of backbone rigidity on the glass transition of polymers of intrinsic microporosity probed by fast scanning calorimetry. ACS Macro Lett. 2019 , 8 , 1022−1028..

Micoulaut, M. Linking rigidity transitions with enthalpic changes at the glass transition and fragility: insight fr om a simple oscillator model. J. Phys.: Condens. Matter 2010 , 22 , 285101..

Aharoni, S. M. Increased glass transition temperature in motionally constrained semicrystalline polymers. Polym. Adv. Technol. 1998 , 9 , 169−201..

Gatsiou, C. A.; Bick, A.; Krokidis, X.; Economou, I. G. Atomistic and coarse-grained simulations of bulk amorphous amylose above and below the glass transition. Macromolecules 2022 , 55 , 2999−3010..

Hadipeykani, M.; Aghadavoudi, F.; Toghraie, D. A molecular dynamics simulation of the glass transition temperature and volumetric thermal expansion coefficient of thermoset polymer based epoxy nanocomposite reinforced by CNT: a statistical study. Phys. A Stat. Mech. Appl. 2020 , 546 , 123995..

Khan, P.; Khan, M. N.; Khan, N. A. Shear and cooling induced regulation of mechanical properties and glass transition temperature of isotactic polypropylene. Mater. Today Proc. 2023 , 72 , 2705−2712..

Alcobaça, E.; Mastelini, S. M.; Botari, T.; Pimentel, B. A.; Cassar, D. R.; de Leon Ferreira, A. C. P.; Zanotto, E. D. Explainable machine learning algorithms for predicting glass transition temperatures. Acta Mater. 2020 , 188 , 92−100..

Tao, L.; Varshney, V.; Li, Y. Benchmarking machine learning models for polymer informatics: an example of glass transition temperature. J. Chem. Inf. Model. 2021 , 61 , 5395−5413..

Zhang, Y.; Xu, X. J. Machine learning glass transition temperature of polymers. Heliyon 2020 , 6 , e05055..

[Zhang, Y.; Xu, X. Machine learning glass transition temperature of styrenic random copolymers. J Zhang, Y.; Xu, X. J. Machine learning glass transition temperature of styrenic random copolymers. J . Mol . Graph . Model . 2021 , 103 , 107796..

[Wilamowski, B. M. Neural networks and fuzzy systems for nonlinear applications. 2007 11th International Conference on Intelligent Engineering Systems , Budapest, Hungary, 2007 , 13-19..

Bose, B. K. Neural network applications in power electronics and motor drives: an introduction and perspective. IEEE Trans. Ind. Electron. 2007 , 54 , 14−33..

[Bell, J. What is machine learning? Machine Learning and the City: Applications in Architecture and Urban Design . 2022 , 207-216..

Xie, C. H.; Qiu, H. K.; Liu, L.; You, Y.; Li, H. F.; Li, Y. Q.; Sun, Z. Y.; Lin, J. P.; An, L. J. Machine learning approaches in polymer science: Progress and fundamental for a new paradigm. SmartMat 2025 , 6 , e1320..

Zhang, M. Y.; Li, J.; Kang, L.; Zhang, N.; Huang, C.; He, Y. Q.; Hu, M. H.; Zhou, X. F.; Zhang, J. Machine learning-guided design and development of multifunctional flexible Ag/poly(amic acid) composites using the differential evolution algorithm. Nanoscale 2020 , 12 , 3988−3996..

Wu, F. Y.; Yin, J.; Chen, S. C.; Gao, X. Q.; Zhou, L.; Lu, Y.; Lei, J.; Zhong, G. J.; Li, Z. M. Application of machine learning to reveal relationship between processing-structure-property for polypropylene injection molding. Polymer 2023 , 269 , 125736..

Higuchi, C.; Horvath, D.; Marcou, G.; Yoshizawa, K.; Varnek, A. Prediction of the glass-transition temperatures of linear homo/heteropolymers and cross-linked epoxy resins. ACS Appl. Polym. Mater. 2019 , 1 , 1430−1442..

Huang, Q.; Chen, Z. W.; Lin, Z. W.; Li, W. M.; Yu, W. J.; Zhu, L. Enhancing copolymer property prediction through the weighted-chained-SMILES machine learning framework. ACS Appl. Polym. Mater. 2024 , 6 , 3666−3675..

Pugar, J. A.; Childs, C. M.; Huang, C.; Haider, K. W.; Washburn, N. R. Elucidating the physicochemical basis of the glass transition temperature in linear polyurethane elastomers with machine learning. J. Phys. Chem. B 2020 , 124 , 9722−9733..

Wen, C. Y.; Liu, B. H.; Wolfgang, J.; Long, T. E.; Odle, R.; Cheng, S. F. Determination of glass transition temperature of polyimides from atomistic molecular dynamics simulations and machine-learning algorithms. J. Polym. Sci. 2020 , 58 , 1521−1534..

Chen, G.; Tao, L.; Li, Y. Predicting polymers' glass transition temperature by a chemical language processing model. Polymers 2021 , 13 , 1898..

Alesadi, A.; Cao, Z. Q.; Li, Z. F.; Zhang, S.; Zhao, H. Y.; Gu, X. D.; Xia, W. J. Machine learning prediction of glass transition temperature of conjugated polymers from chemical structure. Cell Rep. Phys. Sci. 2022 , 3 , 100911..

Barbosa-da-Silva, R.; Stefani, R. QSPR based on support vector machines to predict the glass transition temperature of compounds used in manufacturing OLEDs. Mol. Simul. 2013 , 39 , 234−244..

Uddin, M. J.; Fan, J. T. Interpretable machine learning framework to predict the glass transition temperature of polymers. Polymers 2024 , 16 , 1049..

Goswami, S.; Ghosh, R.; Neog, A.; Das, B. Deep learning based approach for prediction of glass transition temperature in polymers. Mater. Today Proc. 2021 , 46 , 5838−5843..

Armeli, G.; Peters, J. H.; Koop, T. Machine-learning-based prediction of the glass transition temperature of organic compounds using experimental data. ACS Omega 2023 , 8 , 12298−12309..

Okui, N. Relationships between melting temperature, maximum crystallization temperature and glass transition temperature. Polymer 1990 , 31 , 92−94..

Boyer, R. F. Mechanical motions in amorphous and semi-crystalline polymers. Polymer 1976 , 17 , 996−1008..

Biau, G.; Scornet, E. A random forest guided tour. Test. 2016 , 25 , 197−227..

Geurts, P.; Ernst, D.; Wehenkel, L. Extremely randomized trees. Mach. Learn. 2006 , 63 , 3−42..

Schulz, E.; Speekenbrink, M.; Krause, A. A tutorial on Gaussian process regression: modelling, exploring, and exploiting functions. J. Math. Psychol. 2018 , 85 , 1−16..

Deringer, V. L.; Bartók, A. P.; Bernstein, N.; Wilkins, D. M.; Ceriotti, M.; Csányi, G. Gaussian process regression for materials and molecules. Chem. Rev. 2021 , 121 , 10073−10141..

Friedman, J. H. Greedy function approximation: a gradient boosting machine. Ann. Statist. 2001 , 29 , 1189−1232..

[MARK, J. E. James E. Mark - Polymer Data Handbook-Oxford University Press, USA . Oxford University Press, 2009 ..

Tudoran, M.; Putz, M. Molecular graph theory: From adjacency information to colored topology by chemical reactivity. Curr. Org. Chem. 2015 , 19 , 359−386..

Estrada, E. Spectral moments of the edge adjacency matrix in molecular graphs. 1. Definition and applications to the prediction of physical properties of alkanes. J. Chem. Inf. Comput. Sci. 1996 , 36 , 844−849..

Panyukov, S. Theory of flexible polymer networks: Elasticity and heterogeneities. Polymers 2020 , 12 , 767..

Privalko, V. P. Conformation and flexibility parameter of polymeric chains. Macromolecules 1973 , 6 , 111−113..

Mavrantzas, V. G. Using Monte Carlo to simulate complex polymer systems: recent progress and outlook. Front. Phys. 2021 , 9 , 661367..

Tzounis, P. N.; Anogiannakis, S. D.; Theodorou, D. N. General methodology for estimating the stiffness of polymer chains from their chemical constitution: a single unperturbed chain Monte Carlo algorithm. Macromolecules 2017 , 50 , 4575−4587..

Barrat, J. L.; Baschnagel, J.; Lyulin, A. Molecular dynamics simulations of glassy polymers. Soft Matter 2010 , 6 , 3430−3446..

von der Lieth, C. W.; Stumpf-Nothof, K.; Prior, U. A bond flexibility index derived from the constitution of molecules. J. Chem. Inf. Comput. Sci. 1996 , 36 , 711−716..

Horn, N. R. A critical review of free volume and occupied volume calculation methods. J. Membr. Sci. 2016 , 518 , 289−294..

White, R. P.; Lipson, J. E. G. Polymer free volume and its connection to the glass transition. Macromolecules 2016 , 49 , 3987−4007..

Askadskii, A. A. Methods for calculating the physical properties of polymers. Rev. J. Chem. 2015 , 5 , 83−142..

[Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A. G.; Taylor, R. Tables of bond lengths determined by X-ray and neutron diffraction. Part 1. Bond lengths in organic compounds. J. Chem. Soc. Perkin Trans. 2 , 1987 , S1..

Slater, J. C. Atomic radii in crystals. J. Chem. Phys. 1964 , 41 , 3199−3204..

Lewis, D. F. V. Hydrogen bonding in human P450-substrate interactions: a major contribution to binding affinity. Sci. World J. 2004 , 4 , 1074−1082..

Fajans, K. II. Die eigenschaften salzartiger verbindungen und atombau. Z. Für Kristallogr. Cryst. Mater. 1924 , 61 , 18−48..

Allred, A. L. Electronegativity values from thermochemical data. J. Inorg. Nucl. Chem. 1961 , 17 , 215−221..

Adnan, M. S.; Safaei-Farouji, M.; Amiri-Ramsheh, B.; Hemmati-Sarapardeh, A. Modeling apparent viscosity of waxy crude oils doped with polymers using tree-based models, radial basis function neural networks, and Gaussian process regression. Geoenergy Sci. Eng. 2024 , 235 , 212689..

Chen, Z. D.; Li, D. Z.; Liu, J.; Gao, K. Application of Gaussian processes and transfer learning to prediction and analysis of polymer properties. Comput. Mater. Sci. 2023 , 216 , 111859..

Boscato, G.; Civera, M.; Fragonara, L. Z. Recursive partitioning and Gaussian process regression for the detection and localization of damages in pultruded glass fiber reinforced polymer material. Struct. Contr. Health Monit. 2021 , 28 , e2805..

Haruna, S. I.; Zhu, H.; Umar, I. K.; Shao, J. W.; Adamu, M.; Ibrahim, Y. E. Gaussian process regression model for the prediction of the compressive strength of polyurethane-based polymer concrete for runway repair: a comparative approach. IOP Conf. Ser. Earth Environ. Sci. 2022 , 1026 , 012007..

Schneider, H. A. Flexibility and phase transitions of polymers. J. Appl. Polym. Sci. 2003 , 88 , 1590−1599..

Li, T. Y.; Li, H.; Wang, H. Q.; Lu, W.; Osa, M.; Wang, Y. Y.; Mays, J.; Hong, K. L. Chain flexibility and glass transition temperatures of poly( n -alkyl (meth)acrylate)s: Implications of tacticity and chain dynamics. Polymer 2021 , 213 , 123207..

[Rong, W. R.; Fan, Z. Y.; Yu, Y.; Bu, H. S.; Wang, M. Influence of entanglements on glass transition of atactic polystyrene. J. Polym. Sci. Part B Polym. Phys . 2005 , 43 , 2243−2251..

Singh, M. K.; Hu, M.; Cang, Y.; Hsu, H.-P.; Therien-Aubin, H.; Koynov, K.; Fytas, G.; Landfester, K.; Kremer, K. Glass transition of disentangled and entangled polymer melts: single-chain-nanoparticles approach. Macromolecules 2020 , 53 , 7312−7321..

Bender, C. J. Theoretical models of charge-transfer complexes. Chem. Soc. Rev. 1986 , 15 , 475−502..

Goetz, K. P.; Vermeulen, D.; Payne, M. E.; Kloc, C.; McNeil, L. E.; Jurchescu, O. D. Charge-transfer complexes: new perspectives on an old class of compounds. J. Mater. Chem. C 2014 , 2 , 3065−3076..

[Fryd, M. Structure— T g relationship in polyimides. In Polyimides: Synthesis, Characterization, and Applications Volume 1 , Springer: 1984 ; pp 377-383..

Yen, H. J.; Lin, K. Y.; Liou, G. S. High Tg, ambipolar, and near-infrared electrochromic anthraquinone-based aramids with intervalence charge-transfer behavior. J. Polym. Sci. Part A Polym. Chem. 2012 , 50 , 61−69..

Khan, I. M.; Alam, K.; Alam, M. J. Exploring charge transfer dynamics and photocatalytic behavior of designed donor-acceptor complex: characterization, spectrophotometric and theoretical studies (DFT/TD-DFT). J. Mol. Liq. 2020 , 310 , 113213..

Mitsopoulou, C. A. Identifying of charge-transfer transitions and reactive centers in M(diimine)(dithiolate) complex es by DFT techniques. Coord. Chem. Rev. 2010 , 254 , 1448−1456..

Gracia-Fernández, C. A.; Gómez-Barreiro, S.; López-Beceiro, J.; Tarrío Saavedra, J.; Naya, S.; Artiaga, R. Comparative study of the dynamic glass transition temperature by DMA and TMDSC. Polym. Test. 2010 , 29 , 1002−1006..

Liu, P.; Yu, L.; Liu, H. S.; Chen, L.; Li, L. Glass transition temperature of starch studied by a high-speed DSC. Carbohydr. Polym. 2009 , 77 , 250−253..

Rahman, M. S.; Al-Marhubi, I. M.; Al-Mahrouqi, A. Measurement of glass transition temperature by mechanical (DMTA), thermal (DSC and MDSC), water diffusion and density methods: a comparison study. Chem. Phys. Lett. 2007 , 440 , 372−377..

Baldanza, A.; Loianno, V.; Mensitieri, G.; Scherillo, G. Modelling changes in glass transition temperature in polymer matrices exposed to low molecular weight penetrants. Philos. Trans. A Math. Phys. Eng. Sci. 2023 , 381 , 20210216..

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

Pre-trained Mol2Vec Embeddings as a Tool for Predicting Polymer Properties

Data-Driven Exploration of Polymer Processing Effects on the Mechanical Properties in Carbon Black-Reinforced Rubber Composites

Inferring the Physics of Structural Evolution of Multicomponent Polymers via Machine-Learning-Accelerated Method

Related Author

Ivan Zlobin

Nikita Toroptsev

Gleb Averochkin

Alexander Pavlov

Zi-Long Wan

Wan-Chen Zhao

Hao-Ke Qiu

Shu-Shuai Zhou

Related Institution

A. N. Nesmeyanov Institute of Organoelement Compounds of Russian Academy of Sciences, Vavilova St., 28, Bld. 1

Center NTI "Digital Materials Science: New Materials and Substances", Bauman Moscow State Technical University, 2S1 Baumanskaya St., 5/1

Xinjiang Laboratory of Phase Transitions and Microstructures in Condensed Matters, College of Physical Science and Technology, Yili Normal University, Yining

College of Chemistry, Jilin University

School of Applied Chemistry and Engineering, University of Science and Technology of China

⁰