Effect of solution nonideality on cholesterol supersaturation for liquid antisolvent crystallization

Abstract

The present work aims to analyze the effect of solution non-ideality on the magnitude of supersaturation of cholesterol (3) for crystallization from its solution in organic solvent (2) using water (1) as antisolvent. Real supersaturation (S) has been computed as the ratio of activities (a 3/a 3*) at the supersaturated (x 3) and saturated (x 3*) mole fractions respectively, by multiplying the ideal solution supersaturation (Sx = x 3/x 3*) with the activity coefficient ratio [γ3/γ3*]. The values of [γ3/γ3*] are estimated using the Pharma Modified UNIFAC model. The x 3* data have been measured experimentally in aqueous organic-solvent mixtures at different values of x 1 in ethanol at three temperatures (T) in the range of 290.5 to 299.5 K, and in acetone and acetonitrile at 299.5 K. These data have been correlated in terms of x 1 and T separately for computation of Sx , [γ3/γ3*], and S. The quantitative impact of solution non-ideality on the sensitivity of S to the variations of x 1, x 3, [x 3/x 3*], T and nature of solvent are investigated computationally. It is observed that [γ3/γ3*] is less than 1.0 and is relatively more significant at lower x 1 and higher Sx , though it is almost invariant with T and solvent. The value of Sx at 299.5 K is found higher than S by 15% at Sx = 2 and by 162% at Sx = 10. The results clearly establish that it is necessary to include the solution non-ideality in the estimation of solute supersaturation, as otherwise significant errors may result in computing nucleation and growth rates.