Predicting Glass Transition Temperature of Conjugated Polymers via Cheminformatics from Monomer Structures

Abstract

We report a cheminformatics-driven model for predicting glass transition temperature (T_g) of conjugated polymers (CPs) from monomer structure. Our model, based on four key molecular descriptors that capture backbone flexibility, electronic structure, charge distribution, and heteroatom interactions, achieves high predictive accuracy (R^2 = 0.85 for the test set) and outperforms previous approaches. Molecular dynamics (MD) simulations further validate the descriptor-Tg relationships by linking segmental motion and chain rigidity to thermal behavior. This integrated framework accelerates the design of CPs with tailored thermal properties and provides mechanistic insights into the interplay between molecular topology and electronic effects.