Developing a Predictive Form of MOSCED for Nonelectrolyte Solids Using Molecular Simulation: Application to Acetanilide, Acetaminophen, and Phenacetin

The Modified Separation of Cohesive Energy Density Model (MOSCED) is an efficient, analytic method to predict infinite dilution activity coefficients over a range of temperatures. Its predictability makes MOSCED an attractive engineering design tool. However, its use is limited. When trying to model a novel compound, reference data must first be available to regress the necessary MOSCED parameters. Here, we propose the use of molecular simulation to generate the reference dataset. In this fashion, MOSCED can be made a truly predictive engineering design tool. This combines the predictive strength of molecular simulation with the efficiency of MOSCED to create a powerful new tool. Here, we use molecular simulation to generate MOSCED parameters for the nonelectrolyte solid solutes acetanilide, acetaminophen, and phenacetin. Adopting the melting point temperature and enthalpy of fusion of these compounds from available experimental data, we are able to predict equilibrium solubilities. Predictions using the new predictive MOSCED are in good agreement with available experimental solubility data for acetaminophen in nonaqueous solvents.