Semi-empirical correlation of solid solute solubility in supercritical carbon dioxide: Comparative study and proposition of a novel density-based model

Abstract

A comprehensive data set on experimental solubility of 210 solid solutes in supercritical CO 2 counting 5550 data points has been used for comparison of the correlation performance of 21 empirical models. On the basis of the comparison results a new eightparameter density-based model has been proposed. The comparison shows that the three-parameter models are the least accurate. The results also show that models that relate the logarithm of the solubility to the logarithm of solvent density and temperature are more accurate than models that include the pressure. When comparing the overall correlating performance in terms of average absolute relative deviation the proposed model is by far the best with an average absolute relative deviation lying in the range 0.17 e81.99% and an average value of 8.88%.

Dimensionless Empirical Model to Correlate Pharmaceutical Compound Solubility in Supercritical Carbon Dioxide

“…From the category-II models, the Méndez-Santiago and Teja model [20] has an inferior correlating ability with highest global mean AARD (22.312 %) and the model from Keshmiri et al [27] correlated much better with minimum global mean AARD (12.978 %). However, there is an insignificant change in Hozhabr et al [28], Jouyban et al [22], Keshmiri et al [27], Khansary et al [29], Kumar and Johnston [15], and Belghait et al [34], are found to be statistically significant (paired t test, p < 0.05). The proposed model global mean AARD is found to be significantly different from the category I-III models global mean AARD (paired t test, p < 0.05).…”

“…Concerning category‐I models, the Belghait et al model is able to predict much better with the least global mean AARD (12.979 %) and the Chrastil model exhibits a poor correlating ability with the corresponding global mean AARD being 19.032 %. From the category‐II models, the Méndez‐Santiago and Teja model has an inferior correlating ability with highest global mean AARD (22.312 %) and the model from Keshmiri et al correlated much better with minimum global mean AARD (12.978 %).…”

“…R 2 differences between the new model and category I–III models other than Bian et al (2016) , Hozhabr et al , Jouyban et al , Keshmiri et al , Khansary et al , Kumar and Johnston , and Belghait et al. , are found to be statistically significant (paired t test, p < 0.05).…”

“…Among the six parameter models, the model from Bian et al (2011) is the best one for estimating the solubility data after the new model, and the model from Gordillo et al was found to have poor accuracy in estimating solid solubility in ScCO 2 . Furthermore, the models from Keshmiri et al with five and from Belghait et al with eight parameters are highly efficient in estimating the pharma solubility after the proposed model. Notably, the model from Hozhabr et al is the only four‐parameter model which is able to estimate drug solubility very much closer to other models with five to eight parameters.…”

Solubility of cefuroxime axetil in supercritical CO2: Measurement and modeling

“…Such properties are not available for many drug solutes and have to be estimated through group contribution methods. Therefore, the use of semi-empirical density-based models is a good alternative for modeling drug solute solubilities because of their simplicity [15].…”