The effect of cell-size, cross-linking ratio, and force fields on the determination of properties of epoxy crosslinked by heuristic protocol

Abstract

This research explores the effect of the cell size, cross-linking ratio, and the force fields used in the molecular dynamic simulation for determining the mechanical and thermal properties of cross-linked epoxy formed with a heuristic cross-linking procedure. The effects of the abovementioned variables on density, Young’s modulus, shear modulus, bulk modulus, and glass transition temperature values by molecular dynamics (MD) simulation were evaluated. Epoxy resin diglycidyl ether of bisphenol A and hardener diethyl toluene diamine were used in modeling the epoxy. A Heuristic method for reactive molecular dynamics (REACTER) protocol was used as the cross-linking procedure. Firstly, six structure cells were prepared in different cell sizes with a crosslinking ratio of 75%, and a mechanical analysis of all cells was performed. Then, the largest cell was prepared for three different crosslink ratios and its mechanical and thermal properties were calculated. Finally, the mechanical properties of the largest cell were calculated using the three different force fields namely the COMPASS, DREIDING, and UNIVERSAL. The results were also compared with the molecular dynamic simulation results performed using the other crosslinking procedures, and experimental results available in the literature. In comparison, it was observed that the results obtained with MD simulations coincided with the experimental data. It has been concluded that using the largest cell gives closer results to the experimental data but the processing time is also increasing rapidly. Moreover, it was also observed that the increase in the crosslinking ratio caused an increase in the mechanical properties of the epoxy and a significant increase in the glass transition temperature. Finally, compared to other force fields, it is seen that the mechanical analysis results obtained with the COMPASS force field comply more with the experimental data.