Prediction of the effective transverse thermal conductivity of carbon based textile composites with varying constituent properties and reinforcement geometry

摘要

The National Research Council Canada and the University of Ottawa are developing a predictive capability for thermal properties in textile-reinforced composites based on unit cell models. This predictive capability may be used to reduce test time and cost required for thermal characterisation and will provide input required to develop process maps which may be used for material selection of temperature critical structures. This paper covers the modelling method, presents simulation results and compares these to experimental data. Simulation results were produced as a parametric study including tow section, tow width, tow thickness and tow spacing, fibre volume fraction, and both specific heat and thermal conductivity for the fibres and resin. The material modelled during this initial study consists of unidirectional isotropic carbon fibre tows in an epoxy resin matrix. Simulated effective transverse thermal conductivity of the composite material ranged from 0 43.3 to 0 664 W/m°C with varying geometric parameters and material properties. These results can be compared to 0.57 W/m°C (experimentally derived) for a 2/2 twill carbon fibre composite. The parametric study shows that in addition to modifying the overall fibre volume fraction (V_f), geometric parameters also affect transverse thermal conductivity for unit cells with constant Vf. The paper concludes with perspectives for future development.