Thermo-mechanical properties of cross-linked epoxy based systems: A molecular simulation study

Molecular simulations have the ability to provide a direct insight into the role of specific chemical interactions in the behavior of polymer nanocomposites, which is often difficult to obtain in experiments. In this work, molecular simulations are used to investigate the effect of the specific interfacial interactions between the matrix and the filler in cross-linked epoxy-carbon nanotube (CNT) nanocomposites. The poor interfacial interactions between cross-linked epoxy and pristine well-dispersed single-walled CNTs cause a depression in the glass transition temperature (Tg) of nearly ~ 66 K in the nanocomposite compared to the neat polymer, and an enhanced compressibility in the interphase region. As a repercussion of this higher compressibility, the filler-induced reinforcement is negatively affected; the value of the Young’s modulus (E) for this nanocomposite is found to be unchanged compared to that of the neat polymer. On the other hand, the substitution of the pristine CNTs by amido-amine functionalized CNTs eliminates the Tg depression, significantly reduces the compressibility of the interphase region, and therefore causes an increase in the value of E of the nanocomposite by about 50% compared to the neat epoxy.

Furthermore, molecular simulations are useful for investigating the mechanical behavior of materials at ballistic strain-rates (i.e. strain-rates in excess of 10E+4 1/s), where the design and the analysis of experiments can be particularly challenging. In this work, molecular simulations have been used to calculate the Young’s modulus of cross-linked epoxy as a function of high strain-rates and temperature. It is shown that similar to experiments, the time-temperature superposition principle can be applied to simulation data that can then be collapsed onto a master curve, which, in turn, can supplement the experimental master curve with data on the material behavior at ballistic strain rates. It is thus demonstrated that molecular simulations can be a powerful tool in the arsenal of material scientists to establish causality and to predict structure-property relationships at length- and time-scales that are difficult to access in experiments.