Simulations of Semi-Crystalline Polymers and Polymer Composites in order to predict Electrical, Thermal, Mechanical and Diffusion Properties

Abstract: Several novel computer simulation models were developed for predicting electrical, mechanical, thermal and diffusion properties of materials with complex microstructures, such as composites, semi-crystalline polymers and foams. A Monte Carlo model for simulating solvent diffusion through spherulitic semicrystalline polyethylene was developed. The spherulite model, based on findings by electron microscopy, could mimic polyethylenes with crystallinities up to 64 wt%. Due to the dendritic structure of the spherulites, the diffusion was surprisingly independent of the aspect ratio of the individual crystals. A correlation was found between the geometrical impedance factor (τ) and the average free path length of the penetrant molecules in the amorphous phase. A new relationship was found between volume crystallinity and τ. The equation was confirmed with experimental diffusivity data for Ar, CH4, N2 and n-hexane in polyethylene. For electrostatics, a novel analytical mixing model was formulated to predict the effective dielectric permittivity of 2- and 3-component composites. Results obtained with the model showed a clearly better agreement with corresponding finite element data than previous models. The analytical 3-component equation was in accordance with experimental data for nanocomposites based on mica/polyimide and epoxy/ hollow glass sphere composites. Two finite element models for composite electrostatics were developed. It is generally recognized that the fracture toughness and the slow crack growth of semicrystalline polymers depend on the concentrations of tie chains and trapped entanglements bridging adjacent crystal layers in the polymer. A Monte Carlo simulation method for calculating these properties was developed. The simulations revealed that the concentration of trapped entanglements is substantial and probably has a major impact on the stress transfer between crystals. The simulations were in accordance with experimental rubber modulus data. A finite element model (FEM) including diffusion and heat transfer was developed for determining the concentration of gases/solutes in polymers. As part of the FEM model, two accurate pressure-volume-temperature (PVT) relations were developed. To predict solubility, the current "state of the art" model NELF was improved by including the PVT models and by including chemical interactions using the Hansen solubility parameters. To predict diffusivity, a novel free-volume diffusion model was derived based on group contribution methods. All the models were used without adjustable parameters and gave results in agreement with experimental data, including recent data obtained for polycarbonate and poly(ether-etherketone) pressurized with nitrogen at 67 MPa.