No abstract

No abstract

In the present study we report the infinite dilution activity coefficients (γ ∞ ) of 11 alkanes in 67 organic liquids at 25 °C measured by fully automated headspace gas chromatography. This study is part of a more comprehensive one whose main objective is to compare the predictive ability of several semitheoretical models to estimate alkane γ ∞ values. To do this it is obviously necessary to collect reliable experimental data spanning a wide range of chemical properties. The solutes include five linear (pentane to nonane), four branched (2-methylpentane, 2,4-dimethylpentane, 2,5-dimethylhexane, 2,3,4-trimethylpentane), and two cyclic (cyclohexane and ethylcyclohexane) alkanes. The solvents include several homologous series: n-alkanes, 1-alkenes, n-alcohols, 2-ketones, nitriles, acetates, and 34 other common volatile and nonvolatile solvents. Several modifications introduced to the previously reported instrumentation and to the quantitation method are discussed. The data presented are the average of at least two independent measurements with a precision better than 5%.

An inert gas stripping gas chromatography method was described for the determination of infinite dilution activity coefficients ( ^) of organic compounds in water. The infinite dilution activity coefficients for 31 compounds including many halogenated and aromatic compounds were measured. In general, the y% values measured by gas stripping are in very good agreement with y% values estimated from the inverse solubilities and other experimental measurements. This suggests that when there is no directly measured data available, y^v alues estimated from the inverse of solubility can be used for the development and test of the models predicting limiting activity coefficients. A major advantage of this method relative to headspace gas chromatography that improves its precision and accuracy is the fact that the absolute detector response is unimportant. The detector only needs to be linear.