What data are most valuable to screen ionic liquid entrainers for hydrofluorocarbon refrigerant reuse and recycling?

Separating azeotropic mixtures of hydrofluorocarbons (HFCs) for reuse and recycle is environmentally and economically imperative. While ionic liquid (IL)-enabled HFC separations show promise, Edisonian trial-and-error screening for the optimal IL entrainer is intractable and expensive. Here we propose an open-source, equation-oriented modeling framework to rapidly translate HFC/IL solubility data into regressed thermodynamic models which can be used for process design under uncertainty and rapid IL screening. Moreover, we use data science and process systems engineering tools to contemplate which data are the most valuable for IL screening. We find that binary solubility data collected at multiple temperatures is adequate for separation process design and newly available ternary solubility measurements should be reserved for validation. Additionally, we use uncertainty quantification analyses to show up to 10% experimental error is acceptable for IL screening decisions. Informed by these results, we recommend a multi-step workflow for IL screening.