Overcoming the singularity of 1D embedment enhances computational efficiency of CNT nanocomposite thermal analysis multifold

Abstract

In the digital era, novel smart materials require digital design with the more increasing demand on computational performance, the smaller scale we approach. Nanocomposites present an ultimate challenge, where the morphology of filler particles and their interactions with polymer have to be addressed. For carbon nanotube (CNT)-like particles, computational efficiency would increase multifold if we were able to replace these complex interactions with an equivalent 1D geometry. Unfortunately, for thermal analysis, it results in a singularity of infinite temperature. In this study, relying on undocumented yet possibilities in Abaqus software, we develop a technique to overcome the singularity and apply it to an aligned-CNT nanocomposite. Digital twin is populated with 3D particle morphology obtained by electron tomography, and numerical simulations demonstrate close reproducibility of experimentally measured values for homogenized thermal conductivity.