(35b) Factors Influencing the Self-Assembly of Glycolipids into Three-Dimensional Network Phases

Glycolipids are amphiphiles that self-assemble into a variety of spatially periodic thermotropic liquid crystalline phases, including technologically useful 3D network morphologies such as hexagonally-perforated lamellae and double gyroid. Their assembly into and applications as functional 3D network materials require comprehensive knowledge of how amphiphile structural motifs stabilize these mesophases. In this work, using molecular dynamics simulations, a framework is developed to systematically discover glycolipid structural motifs that drive 3D network formation. It features external guiding fields to find appropriate system sizes for the simulations and to overcome the free energy barriers for forming network phases. This strategy is implemented for monosaccharide glycolipids to evaluate the influence chain length, chain branching, and unsaturation in the hydrophobic alkyl tail group on the phase behavior. Radial and angular distributions of molecules along with a deep neural network, the PointNet, are utilized to detect short-range assembly patterns and to classify network phases. This simulation study advances the current understanding of the supramolecular self-assembly of glycolipids.