Thermodynamic modeling of several aqueous alkanol solutions containing amino acids with the perturbed-chain statistical associated fluid theory equation of state
The perturbed-chain statistical associated fluid theory EoS was applied to model the solubilities of glycine,
DL-alanine, L-serine, L-threonine, and L-isoleucine in pure water, pure alcohols (ethanol, 1-propanol, and
2-propanol) and in mixed solvent systems. Three pure component nonassociating parameters for the amino
acids were fitted to the densities, activity and osmotic coefficients, vapor pressures, and water activity of
their aqueous solutions. The solubilities of amino acids in pure and mixed solvent systems were calculated
on the basis of the phase equilibrium conditions for a pure solid and a fluid phase. The hypothetical melting
properties of each amino acid were fitted, to accurately correlate the solubilities in pure water. Only one
temperature independent binary parameter is required for each amino acid/solvent pair. The model can accurately
describe the solubility of the amino acids in water, but the correlation for the solubility in pure alcohols was
not so satisfactory. The solubility in mixed solvents (ternary systems) was predicted on the basis of the modeling
of the solubility in pure solvents, without any additional fitting of the parameters, and the results achieved
were reasonable. Fitting the binary parameter for the pair amino acid/alcohol not to the solubility in pure
alcohol, but to the solubility in the mixed solvent system, the description of the solubility in the mixed solvent
systems was clearly improved and the results were in fair agreement with the experimental data for all mixture
compositions. The results showed a global root-mean-square deviation in mole fraction of 0.0032 for correlation
and 0.0070 for prediction.
