Linking polysulfamide design to morphology using molecular simulation and machine learning

Polysulfamides are a new family of polymer that are analogous to polyureas with a sulfamide group in place of polyurea’s carbonyl group. Polysulfamides exhibit increased thermal stability, adjustable glass transition temperatures with changing polymer backbone structure, and degradability in green conditions (Chemical Science, 2020, 11, 7807-7812). To consider polysulfamide as a sustainable alternative for polyurea, there is a need for a fundamental understanding of how polysulfamide’s morphology and physical properties change with varying polysulfamide designs. In this poster, we will share our work linking polysulfamide morphology to changing design using molecular simulations and machine learning (ML). Our new coarse-grained polysulfamide model enables molecular simulations that complement experiments in our collaborators’ labs. The positional and orientational order of polymers in the simulation is compared to extent of crystallinity observed in experiments using X-ray diffraction, atomic force microscopy, wide-angle X-ray scattering, infrared spectroscopy, and differential scanning calorimetry. Using ML models that automate the quantification of crystallinity from experimental data will be useful to directly compare experimentally observed trends with simulation trends.