Modeling thermal conductivity of polymer/carbon fiber composites prepared by SCFNA method

Abstract The spatial confining forced network assembly (SCFNA) has proven to be a promising method for greatly improving the thermal conductivity (TC) of polymer composites via transforming self‐assembled networks into force‐assembled networks in polymer matrices. Most of the models currently in use were derived to predict the TC of polymer composites with self‐assembled networks; however, these are not suitable for predicting the TC of polymer composites with force‐assembled networks. To address this issue, a TC prediction model for polymer/carbon fiber (CF) composites obtained using the SCFNA method was developed. First, a literature TC prediction model, built by abstracting a representative volume element (RVE) from a CF self‐assembled network, was utilized to determine the key parameters related to the CF distance. Thereafter, a modified RVE model was constructed based on the morphological characteristics of the force‐assembled network of CFs. In conjunction with the thermal resistance method, the TC values of the polymer composites were calculated and successfully validated using experimental data obtained by the SCFNA method. In this model, the compression ratio (ε) and a fitting coefficient f were applied to describe the relationship between self‐assembled and force‐assembled networks, and furthermore, the actual filler content in the force‐assembled network was calculated. It was found that as the ε value and filler content increased, f accordingly decreased; however, the calculated actual filler volume content in the force‐assembled network increased. These results possibly reflect the evolution of the CF microstructure obtained by the SCFNA method. Highlights Filler gaps in self‐assembled networks were studied using a literature model. The relationship between self‐assembled and force‐assembled networks was established. A thermal conductivity model of the force‐assembled network was established. The actual filler volume content in the force‐assembled network was calculated.