Predicting solvent effects on relative volatility behavior in extractive distillation using isothermal titration calorimetry (ITC) and molecular modeling (MM)

A solvent screening approach for extractive distillation, combining isothermal titration calorimetry (ITC) and molecular modeling (MM), was investigated based on three close-boiling mixture cases. The investigated mixtures are octanoic acid (HOct) and levulinic acid (HLev); diethylmethylamine (DEMA) and diisopropylether (DIPE); 2-butanol (2-BuOH) and 2-butanone (2-BuO). Solvent effects were predicted based on ITC and MM. The correlation between the interaction energy of the solvent and mixture constituents and the effect on the relative volatility (validated with VLE experiments) was shown. Successful solvents showed moderate interactions (between binary mixture interaction energy and −60 kJ/mol, with MM calculated). Too low interaction energy lacks effect and too high interaction energy compromised regeneration. For mixtures without strong intermolecular interactions (e.g. 2-BuOH and 2-BuO), a preferential exothermic interaction with the high-boiling component is needed. The developed method with solvent selection guidelines offers an effective approach for systems with strong non-idealities.