Solubility of lovastatin in a family of six alcohols: Ethanol, 1-propanol, 1-butanol, 1-pentanol, 1-hexanol, and 1-octanol

Accurate experimental determination of solubility of active pharmaceutical ingredients (APIs) in solvents and its correlation, for solubility prediction, is essential for rapid design and optimization of isolation, purification, and formulation processes in the pharmaceutical industry. An efficient material-conserving analytical method, with in-line reversed HPLC separation protocol, has been developed to measure equilibrium solubility of lovastatin in ethanol, 1-propanol, 1-butanol, 1-pentanol, 1-hexanol, and 1-octanol between 279 and 313 K. Fusion enthalpy ΔH^fus, melting point temperature, T_m, and the differential molar heat capacity, ΔC_P, were determined by differential scanning calorimetry (DSC) to be 43,136 J/mol, 445.5 K, and 255 J/(mol K), respectively. In order to use the regular solution equation, simplified assumptions have been made concerning ΔC_P, specifically, ΔC_P = 0, or ΔC_P = ΔS. In this study, we examined the extent to which these assumptions influence the magnitude of the ideal solubility of lovastatin, and determined that both assumptions underestimate the ideal solubility of lovastatin. The solubility data was used with the calculated ideal solubility to obtain activity coefficients, which were then fitted to the van't Hoff-like regular solution equation. Examination of the plots indicated that both assumptions give erroneous excess enthalpy of solution, H^∞, and hence thermodynamically inconsistent activity coefficients. The order of increasing ideality, or solubility of lovastatin was butanol > 1-propanol > 1-pentanol > 1-hexanol > 1-octanol.