Molecular scale simulations on thermoset polymers: A review

Abstract

This article reviews the field of molecular simulations of thermoset polymers. This class of polymers is of interest in applications ranging from structural components for aerospace to electronics packaging and predictive simulations of their response is playing an increasing role in understanding the molecular origin of their properties and complementing experiments in the search for tailored materials for specific applications. It focuses on modeling and simulation of the process of curing to predict the molecular structure of these polymers and their thermomechanical response by all-atom molecular dynamics simulations. Results from Monte Carlo and coarse-grained simulations are briefly summarized.

Evolution of network topology of bifunctional epoxy thermosets during cure and its relationship to thermo-mechanical properties: A molecular dynamics study

“…In recent years, with the help of supercomputers, MD simulations have become a powerful tool to simulate the curing process of epoxy thermosets [ 14 ] starting from the uncured resin and making relatively straightforward assumptions about chemical reactions. Hamerton et al [ 15 ] carried out the first fully atomistic MD simulation on trifunctional polycyanurate.…”

Integration of Atomistic Simulation with Experiment Using Time−Temperature Superposition for a Cross‐Linked Epoxy Network

“…There is a sustained interest in the use of molecular dynamics (MD) simulations of cross‐linked epoxy and other network polymers for materials research 1. However, the timescales accessible by MD simulations and thermomechanical experiments are vastly mismatched.…”

“…Further, the computational and experimental a T ‐ T trends show excellent agreement. However, Young's modulus and other quantities that rely on the virial stress tensor are subject to large statistical fluctuations, and hence, a relatively high degree of uncertainty 1…”

Bayesian calibration of coarse-grained forces: Efficiently addressing transferability

“…1–3 In such settings, these tools are becoming an integral component of the decision-making process, given their ability to rapidly search design spaces for novel systems that warrant further experimental study. While it is generally agreed that this procedure can significantly reduce material development costs, a key theme has repeatedly emerged from practical implementations thereof: in order to be informative for commercial applications, computational models require techniques that can rapidly calibrate inputs and assess uncertainties in simulated predictions.…”