Alchemical free energy methods for solubility of O₂, N₂ and CO₂ in perfluorocarbons

Solubility of gases in liquids is governed by a complex interplay of electrostatic interactions, entropy, and molecular polarizability. For poorly soluble gases like O₂, N₂, and CO₂, small changes in solvent structure can lead to large variations in solubility, with direct consequences for chemical reactivity and transport. Fluorinated organic solvents dissolve gases at much higher levels than water, enabling applications across medicine, energy, and materials processing. However, accurately predicting gas solubility across diverse fluorinated solvents remains a major challenge, limiting efforts to modify, enhance, or replace existing fluorinated solvents. Here, we present a new computational method to predict the solubility of O₂, N₂, and CO₂ in 17 chemically diverse fluorinated solvents and their aqueous mixtures. We identify key limitations in existing alchemical free energy approaches which we overcome by incorporating quantum mechanical and empirical corrections to achieve accuracies <1 kcal mol^-1 in free energy of solvation. To demonstrate broader relevance, we apply our method to study the influence of dissolved oxygen on the environmental transport behavior of four common per- and polyfluoroalkyl substances (PFAS): 4 : 2, 6 : 2, 8 : 2, and 10 : 2 fluorotelomer alcohols. Our results reveal that dissolved oxygen affects PFAS partitioning between air and water and impacts both salting-in and salting-out effects. Notably, oxygen and PFAS appear to co-localize at concentrations far below the critical micelle concentration thresholds, suggesting that gas/solute interactions play an underappreciated role in PFAS transport and distribution in aquatic environments. This work advances predictive solubility modeling and provides new insights into gas-mediated transport mechanisms for persistent environmental pollutants.