For glass-forming polymers, direct quantitative comparison of atomistically detailed molecular dynamics simulations with thermomechanical experiments is hindered by the vast mismatch between the accessible timescales. Recently, we demonstrated the successful application of the time- temperature superposition (TTS) principle to perform such a comparison for the volumetric properties of an epoxy network. Here, we follow-up and study the local translational dynamics of the same network computationally. The mean squared displacement (MSD) and time-scaling exponent trends of select atoms of the network are calculated over a temperature range that spans the glass transition. Using TTS, both trends collapse onto master curves that relate the reduced MSD and time-scaling exponent to the reduced time at a reference temperature. Because the reduced time of these computational master curves extends to 109 s, they can be directly compared with experimental creep compliance for the same material from the literature. A quantitative comparison of the three master curves is performed to provide an integrated view that relates atomic-level dynamics with macroscopic thermomechanics. The time-shift factors needed for TTS in simulation show excellent agreement with experiment in the literature, further establishing the veracity of our approach.
Macromolecular Theory and Simulations
epoxy, glass transition, mechanical properties, molecu-lar dynamics, networks, simulations, thermodynamics, thermosets