(383f) Computer Aided Design of Ionic Liquids through a New Group Contribution Approach Based on Surface Charge Density and COSMO-SAC Predictions | AIChE

(383f) Computer Aided Design of Ionic Liquids through a New Group Contribution Approach Based on Surface Charge Density and COSMO-SAC Predictions

Authors 

Karunanithi, A. T. - Presenter, University of Colorado Denver
Farahipour, R., University of Colorado Denver
The practical use of computer-aided ionic liquid design (CAILD) approach towards the design of solvents for separation processes, such as CO2 absorption and liquid-liquid extraction, is greatly hindered by the very limited availability of UNIFAC interaction parameters that cover diverse ionic liquid building blocks (i.e. different cation cores, anions, and side chain groups) and other molecular groups (e.g. CO2). The non-availability of interaction parameters can be linked to the limited thermodynamic data available for systems involving ionic liquids. COSMO-RS and COSMO-SAC methods provide a fully predictive approach for activity coefficients that rely on surface charge density obtained from quantum chemical calculations. However, these methods are not directly compatible with CAMD or CAILD approach as electronic structure calculations are possible only through ab initio quantum chemical programs such as Gaussian or Turbomole and they are specific for each ionic liquid as a whole. Further, these calculations are extremely time consuming and sometimes can take several hours for a single compound. We have recently developed a novel methodology that involves group contribution prediction of ionic liquid surface charge density (sigma profiles) and cavity volumes. This is further integrated with COSMO-SAC methodology to enable prediction of solution thermodynamics (activity coefficient) of systems involving ionic liquids. This approach is fully predictive in nature and does not require fitting group interaction parameters using experimental data or quantum chemical calculations. Further, this method is fully compatible with CAMD approach. This predictive thermodynamic model is integrated with optimization methods to reverse engineer novel ionic liquids. We demonstrate the utility of the model through case studies.