A comparison between semiempirical equations to predict the solubility of pharmaceutical compounds in supercritical carbon dioxide

The aim of this work is to determine, depending on the operation conditions, which semiempirical equation provides the best fit to solubility data of pharmaceutical compounds in supercritical CO2. Solubility data from 27 different pharmaceutical solutes were collected from literature and the different density-based models (Chrastil, Adachi-Lu, del Valle-Aguilera, Sparks, Kumar-Johnston, Bartle, Méndez Santiago-Teja) together with the Yuʹs model and Gordilloʹs model were employed. The results showed that, in general, Sparks’ equation provides the best fit to the solubility data for this kind of solids in supercritical CO2. However, at certain specific conditions, the best correlation is obtained using Gordilloʹs equation. By means of a brief comparison with Peng-Robinson equation of state, semiempirical equations present a more accuracy prediction compared to cubic equations of state, and present no drawbacks such as properties estimation and computational difficulties.