A microstructure-based model for the thermal conductivity of carbon fibers

In this paper, a model is developed to predict the thermal conductivity of carbon fibers (CFs) based on the microstructure information and phase composition obtained experimentally. The microstructures of four different CFs are investigated by SEM and XRD. The contents of graphite crystallites, amorphous carbon, and microvoids are estimated by the internal standard method. Crystallinity, crystallite size, and the orientation of crystallite are considered in the model. The predicted thermal conductivities are in good agreement with the reported values. The contribution of the microstructural factors on the thermal conductivity is evaluated. It is revealed that crystallinity is the major factor that affects the thermal conductivity of CFs. Furthermore, the predictions of the presented model are also compared with the prediction of empirical models based on the measured electrical resistivity. The results show that the prediction of our model has a higher generality. The fundamental understanding on the thermal conductivity of CFs may benefit the development of highperformance CFs.