Prediction of carbon-dioxide activity coefficient for solubility in ionic liquids using multi-non-linear regression...

Area of Science:

• Chemical Engineering
• Thermodynamics
• Materials Science

Background:

• Activity coefficients quantify deviations from ideal behavior in solutions, vital for understanding solute-solvent interactions.
• Carbon dioxide (CO2) capture is critical for industrial processes and environmental sustainability.
• Ionic liquids are promising solvents for CO2 separation, but their behavior requires accurate modeling.

Purpose of the Study:

• To develop a predictive correlation for CO2 activity coefficients in ionic liquids.
• To cover a wider range of temperatures and pressures than previously available in literature.
• To provide a tool for assessing the efficiency of ionic liquids in CO2 capture.

Main Methods:

• Multi-nonlinear regression analysis was used to develop the correlation.
• COSMO-RS (Conductor-like Polarizable Continuum Model for Solvation) was employed for initial data generation.
• Boxplot analysis was performed for outlier detection and parameter range validation.
• Predicted values were validated against experimental data.

Main Results:

• A novel correlation was established for CO2 activity coefficients in ionic liquids.
• The correlation covers pressures from 1-50 bar and temperatures from 298.15-333.15 K.
• The developed correlation achieved an Average Absolute Error (AAE%) of 19.53% and a Root Mean Square Error (RMSE) of 0.465 when compared to experimental data.

Conclusions:

• The developed correlation accurately predicts CO2 activity coefficients in ionic liquids.
• This tool can facilitate the selection and design of ionic liquids for efficient CO2 capture.
• The findings contribute to advancing CO2 separation technologies.