Modelling the dynamic physical properties of vulcanised polymer models by molecular dynamics simulations and machine learning

K Yoshida and Y Kanematsu and DSR Rocabado and T Ishimoto, COMPUTATIONAL MATERIALS SCIENCE, 221, 112081 (2023).

DOI: 10.1016/j.commatsci.2023.112081

The performance trade-off of rubber materials is a major barrier to their development. Thus, to overcome such limitations, there is a strong need to understand the mechanisms related to the influence of the polymer structure on the physical properties of rubber materials. Although a number of studies have analysed the relationship between the structures of monomers and the properties of polymers, few studies have focused on the contribution of polymer dynamics to property analysis. Thus, we performed molecular dynamics simulations on vulcanised natural rubber to calculate the dynamic structural features and physical properties. Machine learning was used to analyse the effect of changes in the polymer structure (i.e., sulfur crosslinking) on the physical properties. Furthermore, the structural parameters involved in the physical properties were extracted. These findings pro-vide clues to understanding the mechanism of simultaneous control of multiple properties in a trade-off rela-tionship, which has previously been considered difficult.

Return to Publications page