Predicting Polymer Grafted Nanoparticles Assembly using Deep Learning

Grafting polymer chains on nanoparticles' surfaces is a well-known route to control their self-assembly and distribution in a polymer matrix. A wide variety of self-assembled structures are achieved by changing the grafting patterns on an individual nanoparticle's surface. However, accurate estimation of the effective potential of mean force between a pair of grafted NPs that determines their assembly and distribution in a polymer matrix is an outstanding challenge in nanoscience. Here, we propose a new Deep-Learning (DL) method that learns the interaction between a pair of grafted nanoparticles from the MD trajectory of a cluster of polymer-grafted nanoparticles. Subsequently, we carry out the DL potential of mean force-based molecular simulation that predicts the self-assembly of a large number of polymer-grafted NPs into various anisotropic superstructures, including percolating networks and bilayers depending on the nanoparticles' concentration in three dimensions. The deep learning potential of mean force-predicted self-assembled superstructures is consistent with the actual superstructures of polymer-grafted NPs. This DL framework is very generic and can accelerate the characterization and prediction of the self-assembly and phase behavior of polymer-grafted and un-functionalized NPs in free space or a polymer matrix.