Rapid, Accurate and Reproducible Prediction of the Glass Transition Temperature Using Ensemble-Based Molecular Dynamics Simulation

• DOI: 10.1021/acs.jctc.4c01364
• Publication Date: 2025-01-29T07:21:21Z
• Authors: James L. Suter; Werner A. Müller; Maxime Vassaux; Alexandros Anasta...; Martin Simmons; David Tilbrook; Peter V. Coveney
• Keywords: Uncertainty Quantification; Multiscale Modeling; Ensemble forecasting; Ensemble average

For the computational design of new polymeric materials, accurate methods for determining glass transition temperature (Tg) are required. We apply an ensemble approach in molecular dynamics (MD) and examine its predictions Tg their associated uncertainty. separate uncertainty into aleatoric contributions arising from dynamical chaos that due to scenarios chosen compute Tg. propose a scenario computing Tg, where density–temperature behavior is computed by running all temperatures concurrently, rather than invoking sequential approach, thereby significantly reducing wall-clock time days several hours without increasing On comparing concurrent on six highly cross-linked epoxy resins cured with aromatic amines, we find excellent agreement our experimentally determined using mechanical analysis both scenarios. The confidence intervals found scale as N–0.5, N number members ensemble, implying ensembles comprised at least ten replicas required predict MD 95% less 20 K. optimal simulation protocol 4 ns burn-in followed 2 production run time.