Preferential solvation in pharmaceutical processing: Rigorous results, critical observations, and the unraveling of some significant modeling pitfalls

Abstract

We probe the solvent effects on the preferential solvation of pharmaceutical species in mixed-solvent environments, identify its universal molecular-based signature, and characterize explicitly the macroscopic-to-microscopic formal connections between the thermodynamic non-idealities and the precisely defined fundamental structure making/breaking functions $S_{\alpha \beta }\left(T,P,x_{\alpha }\right)$. For that purpose, we link the thermodynamic response of the solute triggered by changes in the mixed-solvent environment to either a linear combination of $S_{\alpha \beta }\left(T,P,x_{\alpha }\right)$ or the universal preferential solvation function $PS\left(T,P,x_{\alpha }\right)$. Then, we illustrate the proposed approach by analyzing the solvation behavior of a series of pharmaceutical solutes in both aqueous-organic and mixed-organic environments at ambient state conditions. Moreover, we briefly discuss the tenets of a popular local composition-based model of preferential solvation, present a forensic analysis of the Kirkwood-Buff inversion expressions invoked in its implementation, and identify some pervasive modeling pitfalls as well as their associated common causes and concomitant consequences. Finally, we highlight the significance behind the analysis of preferential solvation phenomena, provide some pertinent observations on the findings, and offer a consistent outlook.