A Molecular Dynamics Study of Epoxy-Based Networks: Cross-Linking Procedure and Prediction of Molecular and Material Properties

Abstract

Molecular modeling of thermosetting polymers has been presented with special emphasis on building atomistic models. Different approaches to build highly cross-linked polymer networks are discussed. A multistep relaxation procedure for relaxing the molecular topology during cross-linking is proposed. This methodology is then applied to an epoxy-based thermoset (EPON-862/DETDA). Several materials properties such as density, glass transition temperature, thermal expansion coefficient, and volume shrinkage during curing are calculated and found to be in good agreement with experimental results. Along with the material's properties, the simulations also highlight the distribution of molecular weight buildup and inception of gel point during the network formation.

A Molecular Dynamics Study of Crosslinked Phthalonitrile Polymers: The Effect of Crosslink Density on Thermomechanical and Dielectric Properties

“…Consistent Valence Force Field (CVFF) is used in all simulations [15,16]. The non-bonded van der The initial uncrosslinked molecular model structure was established using a procedure similar to that used by Varshney et al [12]. In accordance with experimental measurements conducted by Keller et al [13] a molar ratio of 28:1 (BPh:m-APB) with a density of 1.2 g/cm 3 is selected in this study.…”

Multiscale Modeling of Multiphase Fluid Flow

“…This effort aims at addressing some of these limitations of continuum based CFD methods by introducing a multiscale framework for a hybrid MD and CFD approach to modeling heat transfer in fluid flow. MD simulations have long shown great promise as tools in statistical mechanics and materials science [3][4][5][6] and are recently becoming more practical for heat transfer problems [7][8][9][10]. Although atomistic molecular dynamics can provide insight into the molecular nature of heat transfer, the spatial and timescales accessible by MD are of the order of tens of nanometers and hundreds of nanoseconds and therefore not practical neither sufficient to model macro-scale processes.…”

Multi-scale modelling of carbon nanotube reinforced crosslinked interfaces

“…There are two approaches to obtain crosslinked networks: (i) to use representative crosslinked units with a certain degree of crosslinking, then performing classical MD runs for equilibration and data collection 36,53 (ii) to create a simulation box including primary molecules of resin and curing agent without any crosslink, then performing a cyclic set of minimization, equilibration and dynamics runs to reach a crosslinked structure by forming a chemical bond with a probability depending 10 on the reactivity of groups whenever two atoms participating in a crosslink are close in space 2,9,12 .Our previous studies 36,53 indicated that the first approach is ineffective in the construction of 3D complex structures such as epoxy networks due to the equilibration problems and the associated computational cost. Moreover, the effect of nanofillers (nanofibers or nanotubes) on crosslinking is disregarded through this method.…”

“…Computational modelling offers a tool to investigate how strength and stiffness of the composite depend on the degree of cross-linking in the matrix and reinforcement provided by nanofillers. Due to their excellent mechanical and thermal stability neat thermoset epoxy resins have been investigated thoroughly in the literature, and as a matrix, they have been simulated extensively by molecular dynamics (MD) [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18] . For a proper understanding of their equilibrated structures, there are several attempts which perform coarse-graining to simulate these crosslinked networks on 35 larger time and length scales 2,19,20 .…”