Application of MOSCED To Predict Limiting Activity Coefficients, Hydration Free Energies, Henry’s Constants, Octanol/Water Partition Coefficients, and Isobaric Azeotropic Vapor–Liquid Equilibrium

Modified Separation of Cohesive Energy
Density (MOSCED) is a solubility
parameter-based method to predict limiting activity coefficients.
In addition to making quantitative predictions, MOSCED may additionally
be used to understand the underlying molecular-level driving forces
for intuitive solvent selection and formulation. A major improvement
of MOSCED over similar solubility parameter methods is that it splits
the association term. We show by example how this change allows MOSCED
to better model the molecular interactions of associating fluids.
While parametrized to predict limiting activity coefficients, we demonstrate
the ability to predict hydration-free energies, Henry’s constants
in water, and octanol/water partition coefficients. We focus on water
as MOSCED was previously found to perform substantially worse when
water was the solvent. Comparison is made to molecular simulation
and mod-UNIFAC, and we find for the studied reference set that predictions
with MOSCED were in better agreement with experimental data. Comparison
to mod-UNIFAC was additionally made to model binary isobaric azeotropic
vapor–liquid equilibrium. Overall, for these nonideal systems,
the quantitative agreement with experiment is better with mod-UNIFAC,
although the difference with MOSCED is relatively small. An interactive
software suite (MOSCED-CAD) has also been developed to calculate the
properties described in the present study, in addition to binary interaction
parameters for the Wilson and the nonrandom two-liquid (NRTL) equations
for use in a process simulator.