Thermodynamic Properties Of Organic Molecules In Ionic Liquids By COSMO-RS Method

Ionic liquids (ILs) are novel "green solvents", which are expected to replace the extensively used volatile organic solvents, so that a new type of environmental friendly chemical process can be designed. However, the high cost of synthesis and measurement seriously limited their practical applications due to the vast number of ILs. Therefore, it is very necessary to find an effective method to predict the thermodynamic properties of ionic liquid system.At present, some models can be used for correlating experimental data and phase behavior for ionic liquids systems, such as classical UNIFAC. However, the parameters of these models must be optimized by a large amount of existing experimental data. In the case of absence or lacking of experimental data, a predictive model is highly appreciated. COSMO-RS (conductor-like screening model for real system) is a first-principle based model which can effectively predict the thermodynamic properties without adjustable parameters from experiments. In this research, we use COSMO-RS method to predict thermodynamic properties of organic molecules in ionic liquids. The main contents are as follows: 1. Using COSMO-RS method to predict the infinite dilution activity coefficient (γ’) of organic solutes in ionic liquids. The conformational influence of cation, anion and solute on γ∞is discussed. The result shows that the different conformers of cation, anion, and solute will influence the predictive results of COSMO-RS. In the case of intramolecular interaction is strong (e.g., intramolecular hydrogen bond), the difference between lnγ∞is more than±1, which must be considered in the solvents screening.2. Using COSMO-RS method to calculate the lnγ∞(infinite dilution activity coefficient) of alkanes, alkenes, aromatic hydrocarbons, alcohols and other polar solutes in ionic liquids and comparing the predictive data with the experimental data. The results show that the accuracy of COSMO-RS method can be controlled within+1. COSMO-RS method can predict lnγ∞with the change of carbon number (both the solutes and ionic liquids) correctly. In summary, it is a qualitative and semi-quantitative predictive model, which is an ideal tool for solvent screening.3. Based on above results, we used COSMO-RS method to screen the ionic liquids for the separation of benzene and n-hexane. Using the distribution coefficient (D’∞i=1/lg(ri∞V)) and selectivity (S1/2∞/r1∞) as the screening criterion, where the volume of ionic liquid is also considered. The screening result for the513ionic liquids which are formed by27cations and19anions showed that the ionic liquids containing cyano group are suitable for the separation process. 4. Given that the accuracy is not enough in the quantitative prediction of COSMO-RS method, we have tried some different methods to improve the accuracy. Using dimer to deal with the ionic liquids, we used the improved method to calculate lnyγ∞LLE and VLE containing ionic liquids. The results show that the calculation method of dimer improved the prediction accuracy.