Chemical separations play a crucial role in today’s technological advancements by providing high purity materials for component synthesis. Purification methodologies used to refine these reagents are often energy intensive, accounting for 10-15% of the world’s energy consumption. his data-science approach seeks to uncover atomic, molecular and aggregate correlations across constituent phases and to link these correlations through multiscale simulations. Insights will be gained into the underlying energy drivers within a chemical framework that will allow new approaches to low-energy chemical separations.