The solubility of anthracene was measured in binary and ternary co-solvent-water systems. The binary systems consisted of water and a completely miscible organic solvent (CMOS); while the ternary system incorporated a partially miscible organic solvent (PMOS) into the binary systems. The data were used to test the following model:(UNFORMATTED TABLE OR EQUATION FOLLOWS) log Sᵃ(c,p,w) = log Sᵃ(w) + f(c) σᵃ(c) + [(Sᴾ(w) 10 (f(c) σᴾ(c))/D(p)] σᵃ(p). (TABLE/EQUATION ENDS) The terms on the right of the equality sign are the aqueous solubility of anthracene, the solubility of anthracene in CMOS-water, and the solubility of anthracene due to the incorporation of the PMOS, respectively. This model predicts that the incorporation of a PMOS, as a solubilized solute, in CMOS-water mixtures can lead to an increase in the solubility of anthracene due to the cosolvency effect of the PMOS. The results indicate a good correlation between the observed vs. predicted increase in solubility. The deviations observed may be explained by the interactions between the solvent components.

The solubilities of benzene, naphthalene and anthracene were measured in five binary solvent systems. These systems consised of water and one of the following water miscible organic solvents: acetone, acetonitrile, methanol, ethanol, and isopropanol. The measurements were made at intervals of 0.1 volume fractions of the organic cosolvent. Solubility data were also collected for the above solutes in mixed cosolvents. solvent systems containing three In addition, the solubilities of and six five other aromatic solutes were measured in the binary solvent systems of methanol/water and acetone/water. The data was used to test the log-linear solubility model of Yalkowsky (1981). The model predicts a linear relationship between the solubility of a solute in a binary solvent system (S(m)) and the volume fraction of cosolvent present (f(c)) log S(m) = σf(c) + log S(w) Where S(w) is the solubility of the solute in water and σ is the proportionality constant and slope of the curve. The model is easily extended to multiple mixed solvents by combining the σ values from the binary solvent systems. log (S(m)/S(w)) = Σ₁ (σ₁£₁) A method was developed to estimate σ in a given binary solvent system from the octanol-water partition coefficient of the solute. Combining this method with the generalized solubility equation of Yalkowsky to estimate S(w), allows a priori estimates of solubility in mixed solvents. Maximum deviations in the binary solvent systems studied were related to maxima in excess density. In the alcoholic binary solvent systems the minima were related to minima in the heats of mixing of the two cosolvents. The herbicide atrazine deviated dramatically from the model. The system was examined for possible changes in the crystal structure of atrazine. It was found that some crystal modification occured in the presence of mixed solvents. The rate of the change appears to be dependant on the concentration of the cosolvent. A change or modification in the crystal violates one of the basic assumptions of the log-linear model. The assumption is that the crystal contributes equally to the solubility behavior irrespective of the solvent system. It was determined that atrazine undergoes a polymorphic transition in the systems studied. It is postulated that this polymorphism is responsible for the anomolous solubility behavior observed for atrazine.

No description available.