Henry's Constant Analysis for Water and Nonpolar Solvents from Experimental Data, Macroscopic Models, and Molecular Simulation

Abstract

Experimental data, equations of state (EoS), and Monte Carlo simulations are used to analyze the Henry's law constant of solutes in water and in organic solvents at different temperatures. EoS are incapable of correlating the experimental data for light hydrocarbons dissolved in water. Novel simulation methodologies are used for methane in water and in ethane. Results are analyzed with respect to the free energy of cavity formation for hosting the solute molecule in the solvent and the free energy of interactions between the solute molecule and the solvent. It is shown that the hydrophobic phenomenon is driven, to a large extent, by the weak intermolecular interactions between water molecules and nonpolar solute molecules. IntroductionThe Henry's law constant is used widely to describe the low solubility of light solutes in a variety of solvents at relatively low pressure over a wide temperature range. 1-3 Experimental data and accurate correlations have been reported for many solutes in different solvents such as water, 1 light and heavy hydrocarbons, 2,3 and others. 2 Typically Henry's constant increases with temperature for relatively low-temperature values, goes through a maximum, and then decreases for higher temperatures. A relatively high Henry's constant value corresponds to low solubility and vice versa. Traditionally, the low solubility of a solute in a solvent has been associated with differences in energetic interactions between solvent molecules and solute molecules or differences in molecular size or both. A typical example of the first case is the dissolution of n-alkanes in water. N-alkane molecules interact through weak London forces, whereas water molecules exhibit strong hydrogenbonding interactions resulting in three-dimensional structures. A representative example for the second case is that of light gases (hydrogen, nitrogen, etc.) dissolved in a heavy n-alkane.In this work, a generalized thermodynamic framework is presented for the analysis of the Henry's constant and its variation with temperature. Henry's constant is expressed as a product of two terms: The first term involves pure solvent thermodynamic properties only. The second term is a function of the excess chemical potential of the solute at the state conditions examined. The latter is subdivided into a term associated with the formation of a cavity in the solvent to host the solute molecule and a term that accounts for the free energy of inserting the solute into the cavity by turning on the energetic

Henry’s Law Constant from Molecular Simulation: A Systematic Study of 95 Systems

“…For example, Boulougouris et al [12] calculated the solubility of CH 4 in liquid C 2 H 6 and of the same solute in liquid water. Due to their technical importance, the solubility of larger hydrocarbons such as n-butane, n-hexane, cyclohexane, or benzene in liquid water was also studied [13,14]. Other systems, including CO 2 in liquid water [15] or O 2 in liquid benzene [16], were investigated as well.…”

A generalized direct‐particle‐deletion scheme for the calculation of chemical potential and solubilities of small‐ and medium‐sized molecules in amorphous polymers

“…It was shown in ref 17. that the quality of both predictive methods is comparable (i.e., below the statistical uncertainty of the results).…”

“…The new particle deletion approach of Boulougouris et al15–18 is designed to overcome the “insertion problem” that plagues classical simulation approaches to the prediction of solubility and phase equilibria based on the infinite dilution excess chemical potential μ 2 ex,∞ .…”

“…A few years ago a new and more straightforward particle deletion algorithm was introduced by Boulougouris et al to overcome the insertion problem of Widom‐related prediction methods in principle 15–17. There, after construction and equilibration of an atomistic packing model for a polymer or a liquid mixture, containing a reasonably small number of (dissolved) penetrant molecules, a Monte Carlo (MC) or molecular dynamics (MD) data production run is performed for about 100–1000 ps.…”

Simulation of the (vapor+liquid) equilibria of binary mixtures of benzene, cyclohexane, and hydrogen

“…represents the excess contribution [19,41]. In equations (15) and (16) q is the molar density of solvent, and l ex;1 i is the excess chemical potential of solute i at infinite dilution (l ex i ¼ l i À l id i ).…”