Fundamental questions about entropy IV. The hydrophobicity of alkanes: A nearly pure entropic effect

Abstract

The low solubility of low-molecular weight alkanes in water is not due to the necessity of breaking H-bonds in water upon introducing the solute. It is also not due to the breaking of a special kind of cohesion forces "hydrophobic interactions" in the organic liquid. Experimental evidences are given demonstrating that the hydrophobic effect of water and alcohols with respect to liquid alkanes is nearly purely entropic. The introduction of the alkane has practically no influence on the fraction but extends the standardized domain < DOmA> of one molecule of the hydrogen bonded liquid. The hydrophobic effect is essentially due to this extension of the territory of the mobile order. TR a good approximation the increase of the chemical potential of the alkane caused by this effect is given by 6 = -nAT(dASAmobiieorder/anB). On the other hand, contrary to the assumptions of multicomponent theories, the combinatorial entropy of the mixture is practically not affected by the presence of the ti-bonds. From the value of the combinatorial entropy derived in part two, and neglecting all enthalpic effect2 one derives an-equatiotpredicting the solubilities of alkanes in water (in volume fractions) QB = exp (0.5(VB/18-1) + 0.51nV6/18 -2Vd18). This equation, that does not contain any adptableQarameter, predicts the solubility of all the alkanes within 50% error in a domain which varies from 10 to 10-volume fraction.It is widely known that hydrocarbons are very scarcely soluble in water. Several theories have been presented in the past to explain this poor solubility but, in general, they do not take the risk of making quantitative predictions. For instance, the theory of "ideal associated solutions", the description of which can be found for instance in Marcus [l], a multicomponent theory which neglects all effects not directly related to H-bonding, would predict a complete miscibility of pentane and water! In general, people are convinced that the hydrophobic effect is due to the necessity of breaking hydrogen bonds of water or of the alcohols in order to create the cavity where the solute molecule can be placed. This is erroneous. Other people think that there exist in the pure alkanes a special kind of cohesion forces, "hydrophobic interactions", which should bind the alkane molecules together [2]. For lower alkanes this is also erroneous [3].In this article we will give experimental evidence for our statement that, to a good approximation, we may completely neglect the H-bonds which are broken in water (or in alcohols) upon dissolving pentane in these solvents and also that this dissolution does not imply that some special "hydrophobic interactions" have to be broken for the pentane molecules before leaving their own phase.We will show that the hydrophobic effect is due nearly only to thejncrease of the mobile order, described in the previous part of this article, in alcohols or in water when an inert substance is dissolved in these solvents (Inert in the sense that it cannot form itself H-bonds with the solvent).More...