Explaining the Differential Solubility of Flue Gas Components in Ionic

Mar 12, 2009 - Meijin Jin , Yucui Hou , Weize Wu , Shuhang Ren , Shidong Tian , Li Xiao , and Zhigang Lei. The Journal of Physical Chemistry B 2011 11...
0 downloads 0 Views 564KB Size
J. Phys. Chem. B 2009, 113, 4739–4743

4739

Explaining the Differential Solubility of Flue Gas Components in Ionic Liquids from First-Principle Calculations B. Ram Prasad and Sanjib Senapati* Department of Biotechnology, Indian Institute of Technology Madras, Chennai 600036, India ReceiVed: June 14, 2008; ReVised Manuscript ReceiVed: January 14, 2009

Flue gas is greatly responsible for acid rain formation and global warming. New generation ionic liquids (ILs) have potential in controlling the flue gas emissions, as they acquire high absorptivity for the component gases SO2, CO2, etc. The association of the IL-gas interactions to the absorptivity of gas molecules in ILs is, however, poorly understood. In this paper, we present a molecular level description of the interactions of ILs with SO2, CO2, and N2 and show its implications to the differential gas solubility. Our results indicate that the IL anion-gas interactions play a key role in deciding the gas solubility in ILs, particularly for polar gases such as SO2. On the other hand, regular solution assumption applies to N2 solubility. In accordance with the previous theoretical and experimental findings, our results also imply that the IL anions dominate the interactions with gas molecules while the cations play a secondary role and the underlying fluid structures of the ILs remain unperturbed by the addition of gas molecules. Introduction Ionic liquids (ILs) are organic salts that are liquid at room temperature and are composed of organic cations and organic or inorganic anions. Owing to their unique properties such as low volatility, nonflammability, and chemical stability, ILs are regarded as green solvents.1-3 This in conjunction with their ease of preparation at different cation-anion combinations provides a great opportunity to obtain task-specific ILs for a range of applications in the chemical industries.4,5 The areas of application include gas solubilities and separations, cellulose processing, catalysis, extraction, high-temperature pyrochemical processing, etc. The gas solubility in ILs has gained tremendous interest in recent years. This is because of some major problems that the traditional amine-based solvents experience. For example, a trace of sulfur-bearing compound in gas mixtures can degrade the solvent capacity of amine-based solvents substantially.6 ILs can potentially be used as liquid absorbents for permanent gases and as solvents for gas separations without such problems. Thus, Anthony et al.7 reported the solubility and the associated enthalpy and entropy of absorption of nine different gases in 1-n-butyl-3-methylimidazolium hexafluorophosphate, [BMIM][PF6]. Anderson et al.4 measured the solubilities of CO2, CH4, C2H6, C2H4, O2, and N2 in 1-hexyl-3-methylpyridinium bis[(trifluoromethyl)sulfonyl]imide, [HMPY][TF2N]. They found that CO2 has the highest solubility followed by ethylene and ethane. On the other hand, oxygen and nitrogen had immeasurably low solubility. In an attempt to improve CO2 solubility in ILs, Muldoon et al.8 examined the CO2-philicity in a number of ILs with a systematic increase in fluorination. Their results indicate that ILs containing increased fluoroalkyl chains on either the cation or anion have improved CO2 solubility compared to the less fluorinated ILs. Recently, the solubility data of hydrogen in [BMIM][PF6] were reported for a temperature range of 313-373 K and pressure up to 9 MPa. The results show that * To whom correspondence should be addressed. Phone: +91-44-22574122. Fax: +91-44-2257-4102. E-mail: [email protected].

the solubility of hydrogen in an IL is extremely low, which increases slightly with temperature.9 Flue gas, the combustion exhaust gas from power plants, is greatly responsible for acid rain formation and global warming.10,11 New generation ILs could play a key role in controlling the flue gas emissions, since they show high solubility for component gases CO2, SO2, etc. After the great success with CO2 solubility,4,7,8 the researchers are now focusing on designing ILs that absorb SO2 gas at ambient conditions. Thus, Huang et al.12 prepared 1,1,3,3-tetramethylguanidine (TMG)-based ILs [TMG][BF4] and[TMG][TF2N], which absorbed a large amount of SO2 gas without any chemical transformation. They have also found that [BMIM][BF4] and [BMIM][TF2N] can successfully be used for SO2 gas absorption. However, their separation test for mixed SO2-N2 gases was unsuccessful with [BMIM] ILs. Anderson et al.10 used [HMIM][TF2N] and [HMPY][TF2N] to dissolve SO2 in ILs. The SO2 solubility is found to be extremely high and is much greater than that of CO2 at a particular temperature and pressure. Although success is attained in absorbing CO2 and SO2, nitrogen gas, which contributes more than two-thirds to flue gas, is immiscible in any of the commercially available ILs.4,7,10,12 Thus, a better understanding of the solubility of these gases in ILs is necessary. In this paper, we present a molecular level description of the interactions of flue gas components SO2, CO2, and N2 with ILs and try to relate that to the differential solubility of these gases in the ionic liquids. We note that the theories of gas solubility in ILs suffer disagreement and are debated in the literature. While a number of studies have speculated the solute-anion (gas-IL anion) interactions as the major determinant of gas solubility,13-18 studies based upon regular solution theory assume this contribution to be negligibly small compared to that of the solvent-solvent and solute-solute interactions.19,20 Following the former prescription, we paid special attention to the gas-IL anion interactions but also address its relevance to the regular solution theory assumption of gas solubility.

10.1021/jp805249h CCC: $40.75  2009 American Chemical Society Published on Web 03/12/2009

4740 J. Phys. Chem. B, Vol. 113, No. 14, 2009

Prasad and Senapati

TABLE 1: BSSE-Corrected Interaction Energies (kJ/mol) of the Ionic Liquid Anions with SO2, CO2, and N2 Gasa anion -

F ClBrNO3BF4PF6N(CN)2CH3COON3-

SO2

CO2

N2

-152.05 -62.51 -54.73 -56.67 -40.01 -32.78

-82.78 -18.67 -16.21 -25.89 -18.79 -11.08

-10.30 -5.10 -4.31 -6.56 -6.24 -4.67 -5.24 -8.86 -8.76

a The corresponding geometries were optimized at the MP2 level with the 6-311+G* basis set.

Methods of Calculation Ab initio calculations were performed using the Gaussian03 program.21 Preliminary geometry optimizations were performed at the level of density functional theory (DFT) using the 6-311G basis set. More exact calculations of the geometries, energies, and vibrational frequencies were carried out at the second-order Moller-Plesset (MP2)22 level to include the effects of electron correlation. For MP2 level calculations, we employed the 6-311+G* basis set. Various different initial configurations are considered, in which the gas molecules were placed at different locations around the IL anions. Optimizations of these configurations were performed without any constraints in the geometry. The lowest energy structure among all the explored geometries was designated as the optimized structure. All the complexes involving IL cations, i.e., cation-anion ion pairs, cation-gas complexes, and ion pair-gas complexes were optimized at the Hartree-Fock (HF) level using the 6-311+G* basis set. Again various different initial configurations optimized to the same final structure with nearly identical energies at the HF/6-311+G* level. The basis set superposition errors (BSSEs) were corrected using the counterpoise method of Boys and Bernadi.23 The atom-centered point charges were determined via fits to the electrostatic potentials obtained from the calculated wave functions using the CHelpG subroutine of Gaussian03. Results and Discussion Thus far, the association of the gas-IL interaction to the solubility of gas in ILs is poorly understood and debated in the literature. An investigation of specific chemical and molecular interactions between the gas and the ILs, such as dipole-dipole, dipole-induced dipole, and dispersion forces, can be used to extract information about the gas solubilities in ILs.13,24 With this conjecture, we have calculated the interaction energies of some of the room temperature ILs with SO2, CO2, and N2. We start with the calculation of IL anion-gas interactions, following the well-accepted concept that the anion plays a key role in determining the gas solubility in ionic liquids.13-18 The calculated energy values of the IL anion-gas complexes are included in Table 1. The results in Table 1 show that anion-SO2 complexes are stronger than anion-CO2 complexes and much stronger than the complexes of anion-N2. All anion-CO2 complexes are also found to be stronger than N2-based complexes. Interestingly, the solubility pattern of these gases in ILs follows the same trend, i.e., solubility of SO2 > solubility of CO2 . solubility of N2.10 This implies that the high solubility of SO2 or CO2 in ILs can be explained, at least in part, due to the intermolecular complex formation.

Figure 1. Minimum energy structures of the complexes of SO2 with (a) fluoride, (b) nitrate, (c) tetrafluoroborate, and (d) hexafluorophosphate anion. The dotted lines represent the possible modes of interaction, with interatomic distances in angstroms.

TABLE 2: Atomic Charges and Dipole Moment in Isolated and IL Anion-Bound SO2 IL anion

δS (e)

δO1 (e)

δO2 (e)

dipole moment (D)

isolated SO2 FClBrNO3BF4PF6-

0.778 0.814 0.806 0.804 0.800 0.794 0.792

-0.389 -0.407 -0.403 -0.402 -0.400 -0.397 -0.396

-0.389 -0.407 -0.403 -0.402 -0.400 -0.397 -0.396

2.51 2.93 2.83 2.81 2.77 2.70 2.68

Figure 1 shows the minimum energy structures of the IL anion-SO2 gas complexes. In all the halide-SO2 complexes, the halide ion is seen to be located above the plane of SO2 and to lie equidistant from the two oxygen atoms with an X-S-O angle of about 100°. The negatively charged ion thus maximizes the attractive interactions with oppositely charged S while minimizing the repulsive forces with electronegative oxygen atoms. In the case of the nitrate-SO2 complex, the plane of SO2 aligns quasi-perpendicular to the plane of nitrate ion, with the S atom sitting slightly out of plane of the anion. The S atom resides close to one of the nitrate oxygens, and the SO2 oxygens point outward with respect to the nitrate ion to optimize the interactions. In the BF4-SO2 complex, SO2 again leans quasiperpendicular to the plane assumed by two B-F bond vectors, with the S atom lying on the plane. The SO2 ogygens rest above and below this plane, as do the other two tetrafluoroborate fluorines. The oxygens direct opposite to fluorine atoms to better stabilize the complex. The SO2 molecule, in the PF6-SO2 complex, stabilizes near one face of the octahedral anion with S atom distances of 2.82, 3.22, and 3.22 Å from three vertices. The oxygen atoms flank out symmetrically to generate a stable complex of intermolecular energy -32.78 kJ/mol. Notably in all these complexes the anions do not disturb the molecular structure of SO2, in spite of forming strong complexes with it (Table 1). This is quantified by calculating the dipole moment of isolated SO2 and of SO2 in each complex. The values are presented in Table 2. The polar SO2 molecule is found to have a permanent dipole moment of 2.51 D. The dipole moment as well as the charge distribution in SO2 does not change much in the complexes. The dipole moment value of SO2 increases slightly by 0.17-0.42 D upon complex formation. The change in partial charges upon complexation is even minute. The charge on the S atom increases to 0.814 from 0.778 and to -0.407 from -0.389 on O atoms. This means that polar-polar or

Solubility of Flue Gas Components in ILs

J. Phys. Chem. B, Vol. 113, No. 14, 2009 4741

Figure 2. Minimum energy structures of the complexes of CO2 with the same IL anions as described in Figure 1.

TABLE 3: Influence of IL Anions on the Structural Parameters of CO2a IL anion

δC (e)

δO1 (e)

δO2 (e)

isolated CO2 FClBrNO3BF4PF6-

0.960 0.852 0.954 0.956 0.954 0.956 0.958

-0.480 -0.426 -0.477 -0.478 -0.477 -0.478 -0.479

-0.480 -0.426 -0.477 -0.478 -0.477 -0.478 -0.479

a

dipole ∠OCO ∆OCO moment (deg) (deg) (D) 0.00 1.47 0.32 0.24 0.28 0.18 0.13

180.00 137.90 171.48 173.43 172.56 175.05 176.31

0.0 42.10 8.52 6.57 7.44 4.95 3.69

∆OCO denotes the change in the OCO angle (∠OCO).

dipolar-dipolar interactions play a key role in stabilizing these complexes and can be responsible for high SO2 solubility in polar ILs. Carbon dioxide has two polar covalent C-O bonds. However, the bond dipole moments cancel each other since they point in exactly opposite directions. Hence, CO2 has no molecular dipole moment, but the molecule is easily polarizable and can achieve a temporary dipole moment if the linear geometry is distorted. This is exactly what happens when CO2 contacts many of the ILs. The strong electric field due to the polar ILs perturbs the linear structure of CO2 and gives rise to an electric dipole. Figure 2 presents the optimized geometries of anion-CO2 complexes as obtained from our calculations. All the IL anions that we have considered in this study are found to disturb the linear geometry of CO2, with the F- ion having the largest influence. This and other halide anions lie exactly on the plane of CO2 and optimize the interactions by bending the O-C-O angle. However, the effect decreases with increasing halogen size. With respect to the plane of the nitrate ion, carbon dioxide aligns perpendicular and locates symmetrically between two nitrate oxygens to minimize repulsions. Even then the linearity in CO2 was lost. The influence of BF4- and PF6- on the CO2 structure was weak due to the charge delocalization on multiple sites. Nevertheless, the CO2 geometry was distorted to some extent in both cases. This extent of distortion in CO2 can be quantified by calculating the O-C-O angle. The calculated values are included in Table 3. As expected, the largest field strength due to F- (highest charge density) distorts the angle as high as by 42°. The respective distortion due to nitrate, tetrafluoroborate, and hexafluorophosphate is 7.44°, 4.95°, and 3.69°. This nonlinearity induces a temporary dipole moment in CO2 (Table 3), which is now capable of interacting with polar ILs. Thus, a dipole-induced dipole interaction exists between IL and CO2, which may be responsible for the CO2 solubility in ILs.

Figure 3. Minimum energy structures of the complexes of N2 with (a) nitrate, (b) dicyanamide, (c) acetate, and (d) azide anion. The dotted lines represent the possible interactions via π electron clouds. The distances of the centers of two planes are shown in angstroms.

TABLE 4: NtN Stretching Frequency in N2 upon Complexation with IL Anions IL anion

frequency (cm-1)

isolated N2 FClBrNO3BF4N(CN)2CH3COON3-

2175.65 2173.78 2174.93 2174.56 2166.57 2173.23 2169.77 2170.62 2137.50

The nitrogen molecule has only a few electrons which are tightly controlled by the nuclear charges. The molecule thus has a low polarizability, and the field due to the ionic liquid cannot induce an electric dipole. The energy values in Table 1 indicate that the interaction of N2 is governed by weak and nonspecific dispersion forces between the gas and IL anions.15 The minimum energy configurations of the N2-anion complexes, however, show a distinct feature particularly for anions with π electrons, such as nitrate. The N2 molecule occupies a position above the plane of the NO3- anion, leading to a parallel orientation at the perfect interplanar distance of 3.3 Å for extensive π-stacking interactions. This is shown in Figure 3. To recheck this finding, a few more anions with π electrons, e.g., N(CN)2-, CH3COO-, and N3-, were examined. In all cases, the N2 molecule was held parallel at an optimal 3.2-3.4 Å distance for efficient π-π interactions with the IL anions. The minimum energy structures of these complexes are also included in Figure 3. Interestingly, with the exception of F-, the anions which best stabilize the complexes are found to be the ones with π orbitals and capable of offering π-stacking stabilization to N2 (Table 1). The best energy in the acetate-N2 complex can be explained due to interplanar π-π interactions plus the interactions between the acidic hydrogens in the anion with the π electrons of N2. The calculated stretching frequencies of the N2 molecule in complexes at the MP2 level are listed in Table 4. Note that the optimized structures are true minima since no imaginary frequency was observed in the harmonic frequency analysis. The frequency values are consistent with the interaction energies where the anions which are found to damp the natural

4742 J. Phys. Chem. B, Vol. 113, No. 14, 2009

Prasad and Senapati

Figure 5. Optimized structure of the ionic liquid cation-anion pair, 1,3-dimethylimidazolium nitrate. The dotted line represents the hydrogen bond between the cation and the anion.

Figure 4. Optimized structures of the complexes of 1,3-dimethylimidazolium cation with (a) SO2, (b) CO2, and (c) N2. The dotted lines represent the hydrogen bond(s) between the cation and the gas molecule.

TABLE 5: BSSE-Corrected Interaction Energies (kJ/mol) of the IL Cation and Ion Pair with SO2, CO2, and N2 Gasa species +

[MMIM] [MMIM][NO3]a ∆h

SO2

CO2

N2

-32.29 -396.71 -34.94

-8.40 -376.23 -14.46

-3.60 -364.22 -2.45

a The energy between the IL ion pair and gas complexes is denoted as Ea. The corresponding geometries were optimized at the HF level with the 6-311+G* basis set. The interaction energy (Eb) among the IL ion pairs, [MMIM][NO3], at the HF level was computed to be -361.77 kJ/mol. The enthalpy of gas absorption is calculated as26 ∆h ) Ea - Eb.

frequency of N2 molecule the most are NO3-, N(CN)2-, N3-, and CH3COO-. Thus, the poor solubility of N2 gas in previously tested ILs is due to weak dispersion interactions, which can be improved by selecting IL anions capable of providing π-stacking stability to N2. It is to be noted that the solubilization of nonpolar benzene into polar ILs was reported to occur via π-π interactions.25 In flue gas separations, however, one should opt for decreased N2 solubility to improve the selectivity of ILs for other gases, and hence, ILs with anions of π-stacking capability should be avoided. After investigating the IL anion-gas interactions, we wanted to get an estimate of the IL cation-gas interaction strength. Noting the great performances of imidazolium cation-based ILs, we chose 1,3-dimethylimidazolium, [MMIM]+, as the model cation in this study. Figure 4 shows the minimum energy structures of [MMIM]+-SO2, [MMIM]+-CO2, and [MMMIM]+-N2, and the corresponding energies are tabulated in Table 5. Interestingly, the gas molecules stabilize on the same plane of the imidazolium cation ring, though the geometry optimization was initiated by (i) aligning the solutes perpendicular to the plane of the cation ring and (ii) placing the solutes parallel to but above the plane of the cation ring such that a potential π-π stacking stabilization could take place (see the Supporting Information). The gas molecules are also found to engage in an interaction where their electronegative

oxygen (nitrogen in the case of N2) atom aligns with the most positively charged hydrogen on the imidazole ring. In the case of [MMIM]+-SO2, the bent structure of the gas molecule is seen to take the advantage of a second H-bond with the side chain methyl of the cation, which leads to an enhanced stability of the complex. This means H-bonding does play a dominant role in cation-gas interactions. Nevertheless, the interaction strength of a gas molecule with a cation is always weaker than with any IL anion that we have considered in this study (refer to Table 1). Next we check the mode of interactions among the IL cation and anion. We chose 1,3-dimethylimidazolium nitrate, [MMIM][NO3], as the model ionic liquid and proceed to optimize its structure by exploring two different initial geometries. In one of these interaction geometries, the anion was placed above the plane of the cation in parallel, and in the other, it was kept in a perpendicular orientation (see the Supporting Information). Upon energy minimization, both the geometries were found to optimize to the same final structure, similar to IL cation-gas complexes. The anion was seen to be stabilized on the plane of the imidazolium cation whose ring proton H (C2) forms a hydrogen bond with the anion. The optimized geometry of this complex is shown in Figure 5, and the corresponding interaction energy can be found in Table 5. Finally, to check how gas molecules complex with IL ion pairs, we have energy-minimized the interaction geometries of [MMIM][NO3]-CO2, and [MMIM][NO3]-SO2, [MMIM][NO3]-N2. A series of independent calculations involving different initial structures of the complexes were carried out for each system. The several possible geometries that we optimized include solute (gas molecule) above the plane of the nitrate ion in a parallel orientation, solute above the plane of the cation but in a perpendicular orientation, solute sandwiched between parallel cation and anion, and solute lying parallel above the parallel planes of anion and cation (see the Supporting Information). The obtained best energy structures from these calculations are displayed in Figure 6. The respective energy values are included in Table 5. In all cases, the solute leans more toward the anion with an orientation almost similar to that of the anion-gas complexes. In other words, the anion is found to dominate the interactions with the gas molecule, while the cation plays a secondary role. A similar observation was reported from the past experimental and theoretical studies.16-18 Our computed enthalpy of -14.46 kJ/mol for the absorption of CO2 in [MMIM][NO3] also matches quite well with the experimentally measured range of -12.5 to -14.3 kJ/mol for various imidazolium ionic liquids.15 Figure 6 also implies that the cation-anion molecular structural distribution changes very

Solubility of Flue Gas Components in ILs

J. Phys. Chem. B, Vol. 113, No. 14, 2009 4743 hence, any correlation ignoring it will introduce a small error. The large anion-gas intermolecular forces for SO2 may be explained by dipole-dipole interactions, while the weaker interactions between anion and CO2 and anion and N2 may be interpreted through dipole-induced dipole and nonspecific dispersion forces, respectively. In gas-IL intermolecular forces, the IL anion is found to dominate the interactions with the gas molecule while the cation plays a secondary role. This observation is in good agreement with the previous experimental and theoretical studies. In agreement with previous studies, the results also show that the underlying fluid structure of the IL remains relatively unperturbed by the addition of gas. Finally, it is worth mentioning that though the intermolecular interactions play a major role in gas solubility, additional factors such as free volume in ILs could determine the ultimate solubility. Acknowledgment. The financial support of the Department of Science and Technology (DST), Government of India, is gratefully acknowledged. B.R.P. thanks Prof. S. Mangala Sunder Krishnan for support. Supporting Information Available: Complete ref 21 and the various different initial configurations of IL cation-gas, IL ion pair, and ion pair-CO2 complexes. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes

Figure 6. Minimum energy structures of the complexes of 1,3dimethylimidazolium nitrate with (a) SO2, (b) CO2, and (c) N2. The stabilization due to the ionic liquid anion is found to dominate the interaction.

little upon addition of gas into the system. That is, the underlying fluid structure of the IL is relatively unperturbed by the addition of SO2, CO2, or N2, presumably due to the strong Coulombic interactions responsible for the organization of the liquid. This result also agrees with the previous theoretical and experimental findings that CO2 does not affect the fluid structures of imidazolium-based ionic liquids.16 At this stage, it is very tempting to comment on the contradiction in the literature (the gas-anion interactions versus regular solution theory) from our calculated interaction energy values. Table 1 gives the gas-anion interaction energies, and Table 5 contains the IL ion pair interactions. If we consider the IL ion pair energy -361.77 kJ/mol as the solvent-solvent interaction, then the ratio of the gas-anion interaction energies (Table 1) to the solvent-solvent interaction energy would have the following range (excluding the F- anion): (a) SO2 interaction ratio 9.1-17.3%, (b) CO2 interaction ratio 3.1-7.2%, and (c) N2 interaction ratio 1.2-2.5%. Given these numbers, we could submit that the regular solution assumption applied to N2 solubility; however, it does not apply to SO2 solubility. In fact, it also shows why for CO2 the literature disagrees. The CO2-anion interaction is small, and its range is narrow; therefore, any correlation that ignores the CO2-anion interaction would be introducing only a 2% error. Summary and Conclusions Our results shed some light on the existing controversy of gas solubility in ionic liquids. We found that while the high solubility of SO2 in ILs can be explained due to gas-anion interactions, the regular solution assumption applies to N2 solubility. The CO2-anion interaction range is narrow, and

(1) Gorman, J. Sci. News 2001, 160, 156–158. (2) Welton, T. Chem. ReV. 1999, 99, 2071–2084. (3) Zhao, H.; Xia, S.; Ma, P. J. Chem. Technol. Biotechnol. 2005, 80, 1089–1096. (4) Anderson, J. L.; Dixon, J. K.; Brennecke, J. F. Acc. Chem. Res. 2007, 40, 1208–1216. (5) Keskin, S.; Kayrak-Talay, D.; Akman, U.; Hortacsu, O. J. Supercrit. Fluids 2007, 43, 150–180. (6) Rao, A. B.; Rubin, E. S. EnViron. Sci. Technol. 2002, 36, 4467–4475. (7) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2002, 106, 7315–7320. (8) Muldoon, M. J.; Aki, S. N; V, K.; Anderson, J. L.; Dixon, J. K.; Brennecke, J. F. J. Phys. Chem. B 2007, 111, 9001–9009. (9) Kumelan, J.; Kamps, A. P. S.; Tuma, D.; Maurer., G. J. Chem. Eng. Data 2006, 51, 11–14. (10) Anderson, J. L.; Dixon, J. K.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2006, 110, 15059–15062. (11) Parson, E. A.; Keith, D. W. Science 1998, 282, 1053–1054. (12) Huang, J.; Riisager, A.; Wasserscheid, P.; Fehrmann, R. Chem. Commun. 2006, 38, 4027–4029. (13) Blanchard, L. A.; Gu, Z.; Brennecke, J. F. J. Phys. Chem. B 2001, 105, 2437–2444. (14) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2001, 105, 10942–10949. (15) Anthony, J. L.; Anderson, J. L.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2005, 109, 6366–6374. (16) Cadena, C.; Anthony, J. L.; Shah, J. K.; Morrow, T. I.; Brennecke, J. F.; Maginn, E. J. J. Am. Chem. Soc. 2004, 126, 5300–5308. (17) Kazarian, S. G.; Briscoe, B. J.; Welton, T. Chem. Commun. 2000, 20, 2047–2048. (18) Cammarata, L.; Kazarian, S. G.; Salter, P. A.; Welton, T. Phys. Chem. Chem. Phys. 2001, 3, 5192–5200. (19) Camper, D.; Scovazzo, P.; Koval, C.; Noble, R. Ind. Eng. Chem. Res. 2004, 43, 3049–3054. (20) Kilaru, P. K.; Scovazzo, P. Ind. Eng. Chem. Res. 2008, 47, 910–919. (21) Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (22) Moller, C.; Plesset, M. S. Phys. ReV. 1934, 46, 618–624. (23) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553–556. (24) Bhargava, B. L.; Balasubramanian, S. Chem. Phys. Lett. 2007, 444, 242–246. (25) Lachwa, J.; Bento, I.; Duarte, M. T.; Lopes, J. N. C.; Rebelo, L. P. N. Chem. Commun. 2006, 23, 2445–2447. (26) Wang, Y.; Li, H.; Han, S. J. Phys. Chem. B 2006, 110, 24646–24651.

JP805249H