Prediction of the Solubility in Supercritical Carbon Dioxide: a Hybrid Thermodynamic/QSPR Approach

First published in
Computer aided chemical engineering  
Number in series
38
Start Page
1587
End Page
1592
Citation
Computer Aided Chemical Engineering (38): 1587-1592 (2016)
Contribution to Conference
26th European Symposium on Computer Aided Process Engineering 2016  
Publisher DOI
10.1016/B978-0-444-63428-3.50269-1
Scopus ID
2-s2.0-84994232511
Publisher
Elsevier Science
Numerous studies are devoted to the experimental investigation of equilibria in supercritical fluids (SCFs), in particular to the problem of solubility. However, prediction of the solubility entirely based on molecular structure still remains challenging. The aim of this contribution is two-fold. First, we present a compendium of solubility data for high-boiling point compounds in supercritical carbon dioxide and its mixtures with modifiers. Literature data from late 70s till present were collected from more than 300 original publications (ca. 600 compounds, more than 20,000 experimental points). The data are organized in a database that offers diverse functionality. Second, we exemplify the usage of the database by prediction of the solubility employing two models, semi-empirical Chrastil model and more physically sound regular solution model. Both models ultimately require empirical parameters which were related to the molecular structure of the solute (quantitative structure – property relationship, QSPR). Presented hybrid thermodynamic/QSPR approach allows to exceed the limits of the experimental data and to predict the solubility at given pressure and temperature based entirely on molecular structure.
Subjects
chrastil equation
database
regular solution
solubility
supercritical fluid
DDC Class
540: Chemie