Application of a Group Contribution Equation of State for the Thermodynamic Modeling of Gas + Ionic Liquid Mixtures Marı ´a Dolores Bermejo, David Me ´ndez, and A ´ ngel Martı ´n* High Pressure Processes Group, Department of Chemical Engineering and EnVironmental Technology, UniVersity of Valladolid, Prado de la Magdalena s/n 47011 Valladolid, Spain The group contribution equation of state (GC-EoS) is used to describe the phase behavior of gas + ionic liquid mixtures. With this equation, by application of the group contribution concept, if the parameters of the characteristic functional group of a family of ionic liquids are calculated using experimental data of a reduced number of ionic liquids of the family, then the phase behavior of all ionic liquids of the same family can be predicted. With this work, the parameter table of the GC-EoS is extended to systems that are comprised of an ionic liquid of the methylimidazolium bis[(trifluoromethyl)sulfonyl]imide family [-mim][Tf 2 N] and a gas (H 2 , CO, C 2 H 4 ,O 2 , SO 2 , CH 3 OH, N 2 O, or Xe). Furthermore, a compilation of all GC-EoS parameters for gas + ionic liquid systems currently available in the literature is presented. 1. Introduction Ionic liquids (ILs) are substances that are composed entirely of ions that are liquid at ambient or close-to-ambient temper- atures. Their low vapor pressure and excellent solvating properties have generated enormous interest in using these substances to replace conventional organic solvents. Most ionic liquids are nonflammable, which reduces the fire and explosion hazards. A wide liquid range and good thermal stability are other favorable properties of ILs. 1 By appropriate selection of the anion and cation, the properties of the resulting IL can be modified. The determination of the physical properties of an ionic liquid often is a lengthy and costly process. Considering the large number of ILs that can be prepared, it would be advantageous to have methods to predict the physical properties of potential ionic liquids, which could be used for a preliminary selection of suitable ILs for a certain application. 2 In particular, predictive methods of phase equi- librium parameters of systems with ILs would be very useful, because these parameters must be known in many separation and reaction applications. Several authors have developed models to calculate the phase behavior of systems with ILs. Here, some of the most representative works in this field are mentioned. Methods using molecular simulation 3-6 were developed to explain and predict phase equilibrium and properties of ILs. Excess Gibbs energy models such as NRTL and UNIQUAC 7 were developed for describing both liquid-liquid (LL) and liquid-vapor (LV) equilibrium. Cubic equations of state (EoSs) have also been tested, 8,9 and although they certainly are not the most suitable way to describe these systems, they are able to model the solubility of gases and supercritical fluids in ILs satisfactorily and to predict the solubility of ILs in supercritical CHF 3 qualitatively. To describe the behavior of the CO 2 + IL system at high pressure, the tPC SAFT 10,11 and the soft-SAFT 12 were used. Other authors used different EoSs that incorporate the “group contribution concept”: if the parameters of the charac- teristic functional group of a family of ILs are fitted using data of a reduced number of ILs of the family, then the phase behavior of all the ILs of the same family can be predicted. 13-16 Breure et al. 17 described the capability of the group contribu- tion equation of state (GC-EoS) originally developed by Skjold- Jørgensen 18 to represent the phase behavior of systems con- sisting of carbon dioxide (CO 2 ) and the previously commonly used ionic liquids 3-methylimidazolium tetrafluoroborate ([-mim][BF 4 ]) and 3-methylimidazolium hexafluorophosphate 1-substituted ([-mim][PF 6 ]). One of the main advantages of this method is that it relies on the group contribution concept. Bermejo et al. 19 used the same methodology to describe the phase behavior of ILs based on the 3-methylimidazolium nitrate 1-substituted [-mim][NO 3 ] functional group. Ku ¨hne et al. 20 applied this model to describe some ternary (CO 2 + organic + IL) mixtures. Schilderman et al. 21 used this model to calculate the solubility of CO 2 in 3-methylimidazolium bis[(trifluometh- yl)sulfonyl]imide 1-substituted ([-mim][Tf 2 N]) ILs. Martı ´n et al. 22 extended the parameter table of the GC-EoS to systems that are comprised of ILs with a Tf 2 N anion and 1-alkyl-2,3- dimethyl imidazolium ([-dmim]), 1-alkyl-3-methyl-pyridinium ([-mpy]), and 1-alkyl-1-methyl-pyrrolidinium ([-mpyrr]) cat- ions as well as different gases (CO 2 , SO 2 , and O 2 ). In this case, the groups admit substituents in position 1 of the respective cycles of the cation. The objective of this work is to extend the GC EoS to systems that contain ILs of the family [-mim][Tf 2 N] with several gases as H 2 , CO, C 2 H 4 ,O 2 SO 2 , CH 3 OH, N 2 O, and Xe. To do so, pure group parameters for the Xe and N 2 O not previously available in the literature were calculated. In addition, and to apply the equation to ILs with fluorinated substituents, pure and binary interaction parameters of the groups CF 3 and CF 2 were calculated. To do so, new pure group parameters and binary interaction parameters for the new functional groups were estimated based on experimental information about the ILs of this family that was found in the literature. Moreover, a compilation of all available parameters for application of the GC-EoS to the IL + gas systems is presented. 2. GC-EoS Model The GC-EoS was proposed by Skjold-Jørgensen 18 to calculate vapor-liquid (VL) equilibria of nonideal mixtures at pressures up to 30 MPa. It is based on the generalized van der Waals function, combined with the local composition principle. It is expressed in terms of the residual Helmholtz energy (A R ) as * To whom correspondence should be addressed. E-mail: mamaan@iq.uva.es. Ind. Eng. Chem. Res. 2010, 49, 4966–4973 4966 10.1021/ie901989f 2010 American Chemical Society Published on Web 04/23/2010