James B. Falabella, Chemical and Biomolecular Engineering, Georgia Institute of Technology, 311 Ferst Dr. NW, Atlanta, GA 30332-0100, Xin-Sheng Chai, Institute of Paper Science and Technology, Georgia Institute of Technology, 311 Ferst Drive NW, Atlanta, GA 30332-0100, and Amyn S. Teja, Georgia Institute of Technology, 311 Ferst Drive NW, Atlanta, GA 30332-0100.
The determination of accurate air-water partition coefficients for volatile organic compounds (VOC) and greenhouse gases (GHG) is essential in pollution and global climate modeling. In the atmosphere, oceans, as well as in groundwater, the concentration of dissolved salts is sufficiently high to significantly affect the air-water partitioning behavior of the VOC or GHG. There is therefore a great need for partitioning data for common VOCs and GHGs in aqueous environments containing salts. Unfortunately, such data are not readily available, and even when available, there are large disagreements between different sets of data. Henry's constants of several VOCs in aqueous salt solutions were therefore measured using a new headspace gas chromatography (HTHSGC) method to create a set of internally consistent air-water partition coefficients for modeling. The VOCs studied included the homologous series of 1-alkanols, 2-ketones, and organic sulfides, as well as components of gasoline including toluene, ethylbenzene, o-xylene, methyl tertbutyl ether, and ethyl tertbutyl ether. An investigation of solubility parameters for the hydrogen bond, polar, and dispersion forces gives reason for the range of salt effects observed with VOCs and GHGs in solutions of the same ionic strength. A model with temperature-independent parameters based on dilute solution theory was developed using the library of data to discover useful trends in the salt effect parameters for classes of VOCs and GHGs. The model correlated Henry's constants of methane over temperature ranges as wide as 300 K, salt concentrations up to 4 molal, and pressures up to 1000 bar. Extrapolations of up to 50 K, 1 molal salt and 100 bar pressure can also be performed with the model to eliminate the need for additional experiments. The temperature-independent salt effect parameter was found to be directly proportional to the critical volume of the VOC. Also, all homologous VOCs followed the same linear trend with size, allowing prediction of the salt effect.