Thermal Properties of Orthotropic Materials: Experimental Study and Correlation with Microstructure

Predictive simulation tools are needed for aircraft structures sizing, to enhance safety, reliability and performance. But the main difficulty arises from the lack of accurate knowledge regarding their input parameters, particularly thermal properties of composite materials. A lot of techniques [1] already exist to characterise orthotropic composite materials but they have to be associated or repeated for an extensive and full thermal characterisation. To overcome this problem, a new experimental device based on a non-stationary method was developed to identify simultaneously the specific heat CP and the components of the effective thermal conductivity tensor [λ] of such composite materials. In the same way, numerical techniques are utilized to estimate effective thermal properties. We calculate thermal properties from microstructure images since the microstructure and the properties of each phase have a strong impact on the calculation of the effective properties of these materials [2]. The fibres distribution in composites deviates generally hard-core random structure and often exhibits fibre-rich and -poor regions. The representative volume element (RVE) is then a crucial input to predict reliable results. In this contribution, we first develop the experimental set-up designed to estimate composite thermal properties. In a second part, we investigate a characterization of the microstructure of unidirectional (UD) composite using image processing techniques and stereological spatial descriptors. We finally define the RVEs of this microstructure from spatial descriptors as well as effective thermal conductivities (λx,y).