Prediction of CO2 solubility in ionic liquids using the PSRK model

• The PSRK model is successfully used to predict CO2 solubility in ILs.
• The UNIFAC PSRK matrix is extended to include new IL groups.
• Experimental VLE data were subject to consistency test.
• The model shows good predictability for ILs having longer alkyl chain.

The predictive Soave–Redlich–Kwong (PSRK) model was applied to predict CO2 solubility in imidazolium-based ionic liquids containing bis(trifluoromethylsulfonyl)imide [Tf2N], tetrafluoroborate [BF4], and hexafluorophosphate [PF6] anions. The UNIFAC PSRK matrix was extended to include these new groups. Binary interaction parameters were obtained by regressing experimental vapor–liquid equilibrium data available in the literature. Experimental data were subjected to a consistency test, and only consistent data were considered during the regression. The model predictions showed satisfactory agreement with the experimental data at pressures up to 5 MPa, with an overall root mean square deviation (RMSD) of 0.31. The prediction ability of the model was tested for ionic liquids containing long alkyl chains, exhibiting an overall RMSD of 0.40. The results indicated that the PSRK model provides a reasonable prediction of CO2 solubility in ionic liquids at pressures not in the vicinity of the critical pressures of CO2 and ionic liquids.

Figure optionsDownload as PowerPoint slide