Can machine learning help us discover better materials? This study employs a combination of high-throughput molecular dynamics (MD) simulation and active learning (AL) to identify polymer blends with enhanced thermal conductivity (TC) compared to their single-component counterparts. Researchers initially determined the TC of approximately 600 amorphous single-component polymers and 200 blends via MD simulations.
They identified the optimal representation method for polymer blends and utilized an AL framework to explore the TC of roughly 550,000 unlabeled blends. This framework proved highly effective in accelerating the discovery of high-performance polymer blends for thermal transport. The
Published in npj Computational Materials, this study is highly relevant to the journal's focus. It combines computational materials science with machine learning to discover polymer blends with enhanced thermal conductivity, a key area of interest for researchers in this field.