On the High-Pressure Solubilities of Carbon Dioxide in Several IonicLiquidsIsabel Mejía, Kathleen Stanley, Roberto Canales, and Joan F. Brennecke*

Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, United States

ABSTRACT: The solubility of carbon dioxide in nine ionicliquids (ILs), at 298 K, 313 K, and 333 K and pressures up to 9MPa, is presented. The solubility of CO2 in the selected ILsincreases with increasing pressure and decreases withincreasing temperature, as expected. The effect of severaldifferent anions on the solubility of CO2 with a common 1-ethyl-3-methylimidazolium cation is studied in this work. Theanions compared were hydrogen sulfate ([emim][HSO4]),methylsulfate ([emim][MeSO4]), methane sulfonate ([emim]-[MeSO3]), thiocyanate ([emim][SCN]), and diethylphos-phate ([emim][DEP]). The results for 1-ethyl-3-methyl-imidazolium hydrogen sulfate are particularly interesting andcan be explained based on the stronger attractive interactions between the IL and CO2, which is supported by quantumcalculations. Other ILs investigated were ethyl(tributyl)phosphonium diethylphosphate ([P2444][DEP]), 1-hexyl-3-methyl-imidazolium trifluoromethane sulfonate ([hmim][OTf]), 1-(2-hydroxyethyl)-3-methylimidazolium trifluoroacetate ([OHemim]-[TFA]), and trihexyl tetradecylphosphonium bis(trifluoromethylsulfonyl)imide ([P66614][Tf2N]). Using the new data for the[emim]+ ILs, in concert with data from the literature, these ILs were selected primarily to study the effect on the solubility ofchanging the cation. Of particular note is the high solubility of CO2 in the trihexyltetradecylphosphonium based IL comparedwith its imidazolium equivalent. A brief discussion is presented to explain the observed solubility results.

■ INTRODUCTIONIonic liquids (ILs) are arguably among the most interesting andimportant solvents to be developed in recent years. They arehighly versatile due to their remarkably low vapor pressures,generally high thermal and chemical stability, nonflammability,and the ability to tune the chemical and physical properties bythe incorporation of various functional groups into the anionsand cations. As a result, they are effective media for reactions,1

have been considered as solvents for CO2 separation,2−8 can be

used as absorbents in absorption refrigeration systems,9,10 servein reactive catalysis,11−13 and have been evaluated for a widevariety of other separation processes.14−18 CO2 solubilities inthe ILs are important for a number of these applications.In addition, supercritical or near-critical CO2 has been

studied by several authors as a way to recover valuable productsfrom IL mixtures. For instance, CO2 can be used to separate theproduct from an IL + catalyst reaction mixture, leaving the ILand the catalyst ready for reuse.15,17 Scurto et al.19

demonstrated that CO2 could be used to separate ILs fromaqueous solutions. Subcritical and supercritical CO2 has alsobeen used successfully to induce separations of IL and organiccompounds.16,17,19,20 It is well-known that high-pressure CO2can affect the solvent strength of mixtures, especially when thesolvent swells significantly with the addition of CO2. Thus, thisis another situation where high pressure IL + CO2 phasebehavior is important.The solubilities and thermophysical properties of a number

of IL + CO2 mixtures have been studied in recent years.21−41

Despite the growing body of data, additional fundamentalunderstanding of CO2 solubility in ILs is needed in order toincrease and improve the process applications of IL + CO2systems. Experimental measurements of thermophysical andthermodynamics properties of ILs21−23,34,41−45 and measure-ments of solubilities of CO2 in ILs at low and highpressures24−39 have been complemented by spectroscopymeasurements,46−50 charge models,10,24,51 and molecularsimulations41,52,53 to understand the mechanism of solvationand the interactions controlling the solubility of IL + CO2systems.It is known from the literature that functionality on both the

anion and the cation,2,54,55 as well as the free volume, canimprove the solubility of CO2 in the IL. Increasing thefluorination of the anion (e.g., using the Tf2N

− anion) andobtaining a cation with longer alkyl chains (higher free volume)enhances the “CO2-philic” nature of the IL, inducing anincrease in the solubility of CO2.

36,38,56,57 However, Raveen-dran and Wallen46 concluded that the incorporation ofacetylation is better than fluorination for the design ofrenewable CO2-philic compounds and even the incorporationof SO bonds could be an effective approach to the design ofCO2-philic materials. These are some of the reasons for therapid development of new ILs with a wide variety of

Received: June 6, 2013Accepted: August 12, 2013

Article

pubs.acs.org/jced

© XXXX American Chemical Society A dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXX

combinations of anion and cation. This opens up opportunitiesnot only for the design of new ILs with higher CO2 solubility,but also for the design of ILs which are less toxic and morereadily biodegradable (i.e., “greener” ILs).In this study, solubilities of carbon dioxide are shown in nine

different ILs, 1-ethyl-3-methylimidazolium hydrogen sulfate([emim][HSO4]), 1-ethyl-3-methylimidazolium methylsulfate([emim][MeSO4]), 1-ethyl-3-methylimidazolium methane sul-fonate ([emim][MeSO3]), 1-ethyl-3-methylimidazolium thio-cyanate ([emim][SCN]), 1-ethyl-3-methylimidazolium diethyl-phosphate ([emim][DEP]), ethyl(tributyl)phosphonium dieth-ylphosphate ([P2444][DEP]), 1-hexyl-3-methylimidazoliumtrifluoromethane sulfonate ([hmim][OTf]), 1-(2-hydroxyeth-yl)-3-methylimidazolium trifluoroacetate ([OHemim][TFA]),and trihexyltetradecylphosphonium bis(trifluoromethyl-sulfonyl)imide ([P66614][Tf2N]). The structures, names, andabbreviations for these compounds are found in Table 1. Thesolubilities are reported at moderate pressures at 298 K, 313 K,and 333 K. A brief discussion is presented to explain the effectof the anion and the cation on the solubilities of CO2.

■ MATERIALS AND METHODSChemicals. The ILs were purchased commercially from

EMD Chemicals, Inc. ([emim][HSO4] and [emim][DEP]),Sigma Aldrich ([emim][SCN]), Iolitec ([emim][MeSO4],

[emim][MeSO3], and [hmim][OTf]), and Cytec Industries,Inc. ([P2444][DEP]). All of the ILs had a purity of ≥ 0.99 massfraction except for [emim][HSO4], which was reported as ≥0.98 pure in mass fraction.9 The [P66614][Tf2N] was synthesizedin our laboratory by mixing a mole ratio of 1:1.2 oftrihexyltetradecyl bromide ([P66614][Br], Cytec Industries,Inc., purity ≥ 0.95 mass fraction) and lithium bis-(trifluoromethylsulfonyl)imide (LiTf2N, 3M-Fluorad, purity ≥0.995 mass fraction) at ambient conditions to form [P66614]-[Tf2N], which is immiscible with water. The LiBr byproductwas removed by repeated water washes. [P66614][Tf2N] isbelieved to be greater than 0.99 mass fraction purity, asdetermined by NMR analysis and halide analysis (Cole-Parmerbromide selective ion conductivity probe, model EW-27504-02). [OHemim][TFA] was also synthesized in our laboratoryfollowing the procedure described elsewhere.10 The watercontent of the ILs was measured with a Karl Fisher Coulometer(Metrohm 831). Each IL had a water content less than 0.0005weight fraction except for [emim][DEP], which had a watercontent around 0.0009 weight fraction.

Solubility Measurements. The experimental apparatusused for determining the solubility of carbon dioxide in the ILhas been described in previous publications.36,39 The apparatusconsists of two zones: a high-pressure CO2 delivery system andan equilibrium zone (the cell side). Both of them are provided

Table 1. Structure, Names, and Abbreviations of the ILs Used in This Study

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXB

with constant temperature air and water baths, respectively, tomaintain accurate temperature control. The principal compo-nents of the CO2 delivery side are a positive displacementRuska pump (model 2200) used to pump mercury andcompress the carbon dioxide, a Heise Bourdon gauge (0 to 20.7MPa with rated accuracies of ± 0.02 MPa) and T typethermocouples with an accuracy of ± 1 K. The cell side iscomposed of a special cell holder, built in our laboratory. Thecell is a sapphire tube (Saphikon Inc.) of 4” length, with amaximum working pressure of 20.7 MPa, a magnetic bar to mixthe IL and CO2, a cathetometer (Titan Tool Supply Co. Inc.)to measure the height of the liquid inside the cell (± 0.005mm), and a Heise pressure transducer [(0 to 20.7) ± 0.01 MPa,model 901A] to measure the pressure in the cell side.In a typical solubility experiment, a dry IL sample (between 1

g and 1.5 g) is loaded in the sapphire cell inside a gloveboxunder a nitrogen atmosphere. The cell is attached to the cellholder and immersed in the water bath at the desiredtemperature. The cell side is then purged with carbon dioxideat about 0.15 MPa at least three times to displace any traces ofair, and the initial liquid level is measured with thecathetometer. Once the temperature is in equilibrium, carbondioxide is pumped from the CO2 delivery side to the cell side,filling the lines and the headspace of the cell at a definedpressure. Sufficient CO2 is present so that vapor−liquidequilibrium exists. After stirring the vapor−liquid meniscus, acertain amount of CO2 is absorbed by the liquid phase, anexpansion in the liquid volume is observed, and a pressure dropof the gas phase in the cell side has occurred. The sample ismixed until equilibrium is reached, that is, no pressure changesare detectable. At this point, the pressure, temperature, and thenew liquid level are recorded, and the procedure is repeatedagain up to the maximum desired pressure. On the deliveryside, there is a pressure drop caused by the release of CO2 tothe cell side. The pressure is increased back to the initialpressure on the delivery side with a Ruska pump, and thevolume change required is registered for each solubility point.The Span−Wagner58 equation of state was used to calculate

the density of the CO2 and determine the amount of CO2transferred from the delivery system to the cell side. Knowingthe volume of the lines on the cell side and the headspace of thecell, it is possible to determine the solubility of CO2 in the ILby difference, once again using the Span−Wagner equation ofstate for the vapor phase. Assuming that the vapor is pure CO2is a very good assumption for the liquids investigated here sincethe ILs have very low volatility. Using a cathetometer todetermine the height changes of the liquid level, one can alsoget the molar volume of the liquid mixture. The combinedexpanded uncertainty of the molar volume is calculated fromthe standard uncertainty of the cell volume which is determinedfrom the height measurement with the cathetometer, the ILmass (± 0.0002 g), and the moles of CO2 dissolved in theliquid. This last uncertainty is determined from propagation ofthe uncertainties of the temperatures, Ruska pump volume, andvolumes of the cell and lines.Propagation of errors due to the uncertainties in temper-

ature, pressure, and volume measurements results in very smalluncertainties in the CO2 mole fractions, as describedelsewhere.59 However, based on repeatability of the measure-ments, the actual experimental uncertainty is estimated asapproximately ± 0.01 mole fraction for pressures greater than 3MPa and ± 0.02 mole fraction for pressures less than 3 MPa.Important factors not included in the error propagation that

contribute to this uncertainty include impurities in the ILsamples and potential small leaks in the system. To furthervalidate the system operation, several experiments wereconducted to measure the solubility of carbon dioxide int o l u ene and 1 -h e x y l - 3 -me thy l im id a zo l i um b i s -(trifluoromethylsulfonyl)imide ([hmim][Tf2N]). Those resultswere compared with literature data,36,60,61 obtaining goodagreement within the expected experimental uncertainty (±0.01 mole fraction). The uncertainty in the molar volumes wasestimated from the cathetometer reading error and the CO2solubility uncertainty.

■ RESULTS AND DISCUSSIONEffect of the Anion on CO2 Solubilities. The solubility of

CO2 in [emim][SCN], [emim][HSO4], [emim][MeSO4],[emim][MeSO3], [emim][DEP], [emim][TFA],30 and[OHemim][TFA] at 298 K and 313 K are shown in Figures1 and 2, and data are tabulated in Tables 2 to 7, respectively. As

Figure 1. CO2 solubility in ILs at 298 K. ◆, [emim][DEP]; ■,[emim][MeSO4]; ×, [emim][MeSO3]; ○, [hmim][Tf2N];

36 △,[OHemim][TFA]; ∗, [emim][TFA].30

Figure 2. CO2 solubility in ILs at 313 K. ◆, [emim][DEP]; ■,[emim][MeSO4]; +, [emim][HSO4]; ×, [emim][MeSO3]; ◇,[emim][SCN]; ○, [hmim][Tf2N];

36 △, [OHemim][TFA]; ∗,[emim][TFA].30

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXC

typical for many ILs, the solubility of CO2 increases whenincreasing pressure. The CO2 solubility is as high as 0.69 molefraction at around 8.3 MPa and 313 K for [emim][HSO4]. Thedata shown are clearly consistent with the expected temper-ature dependence of the solubility of IL + CO2 systems. Forinstance, at 3.8 MPa the solubility of CO2 in [emim][MeSO4]at 298 K is 0.41 mole fraction, while at 313 K it is 0.24 molefraction at the same pressure. The same trend is followed by allthe ILs shown in both figures.The interest in our study is the effect of different anions on

the solubility of CO2 when testing ILs with a common cation,

1-ethyl-3-methylimidazolium ([emim]+). The anions comparedare [MeSO3]

−, [HSO4]−, [TFA]−, and [MeSO4]

−. Figure 2shows the solubility of CO2 in [emim]+ based ILs with thepreviously mentioned anions at 313 K. The order of CO2solubility at this temperature in the [emim]+ ILs with differentanions is: [MeSO3]

− ≈ [MeSO4]− < [TFA]− < [HSO4]

−. Asimilar tendency is shown at 298 K for [MeSO3]

−, [MeSO4]−,

Table 2. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[emim][SCN] (2)a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

299.4 0.0980 155299.4 1.09 0.08 147 2299.5 1.34 0.14 137 4298.8 2.20 0.21 128 4299.0 3.34 0.27 119 6298.6 4.27 0.33 112 8313.0 0.0995 159313.3 1.69 0.14 140 4313.3 3.03 0.23 128 6313.3 4.28 0.30 118 8313.3 5.67 0.34 82 8

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Table 3. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[emim][HSO4] (2)

a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

313.3 0.0999 159313.3 1.65 0.24 121 6313.3 2.92 0.36 103 8313.3 4.18 0.47 84 8313.3 5.82 0.56 71 8313.3 7.01 0.62 61 8313.3 8.29 0.69 51 8333.6 0.0985 161333.6 1.44 0.20 128 6333.6 2.87 0.31 113 8333.6 4.20 0.38 101 8333.6 5.59 0.43 93 10332.8 7.61 0.51 83 12332.8 8.29 0.54 76 12332.8 9.70 0.60 66 12

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Table 4. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[emim][MeSO4] (2)

a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

299.5 0.0986 177299.5 1.13 0.15 155 6299.6 2.47 0.29 133 8299.6 3.86 0.41 113 10299.6 5.21 0.51 96 12313.3 0.0991 179313.2 1.33 0.03 176 8313.2 2.59 0.16 155 12313.2 3.77 0.24 143 14313.3 4.85 0.29 135 18313.3 6.14 0.34 129 22313.3 7.48 0.38 122 30332.9 0.0996 179333.0 1.31 0.01 182 8333.0 2.65 0.08 170 14333.0 4.07 0.14 161 18333.0 5.20 0.23 145 20333.0 6.73 0.29 137 24333.0 7.93 0.34 129 28

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Table 5. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[emim][MeSO3] (2)

a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

298.7 0.0990 166298.7 1.24 0.12 152 4298.7 2.06 0.22 137 6298.6 3.18 0.27 131 10298.5 4.18 0.30 128 12298.4 5.17 0.33 125 16313.5 0.0989 168314.1 1.42 0.05 162 4313.4 2.84 0.15 149 6313.3 4.43 0.26 133 8313.3 5.56 0.32 126 10

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXD

and [TFA]−, but the solubility of [MeSO3]− is lower compared

with [MeSO4]−. It was expected that CO2 solubility in the ILs

with fluorinated anions would be higher than in those withoutfluorinated anions since fluoroalkyl groups are known to beCO2-philic. By contrast, the particularly high solubility of CO2in the nonfluorinated [emim][HSO4] was very interesting andsomewhat surprising. If the attention is focused on [HSO4]

−

and [MeSO4]−, a lower CO2 solubility in [HSO4]

− is expecteddue to its shorter (i.e., nonexistent) alkyl chain, which generallymeans a smaller free volume. As suggested by Blanchard et al.,34

there is generally greater solubility in ILs with larger molarvolumes. This idea has subsequently been expanded upon byother authors, including the premise that free volume is themost and, perhaps, only important factor in determining CO2solubility.28,57,62,63 However, it is shown in Figure 2 that thesolubility of CO2 in [HSO4]

− is significantly higher than thesolubility of CO2 in [MeSO4]

−, and this cannot be explained bythe free volume concept. Thus, it is evident that solubility ofCO2 in these two molecules has to be analyzed in terms of theinteraction energies between the IL and the CO2.It is very interesting to analyze the effect of the charge on the

oxygen atoms of the [HSO4]− and [MeSO4]

− anions.According to numerous molecular simulations53,64−70 thecarbon of CO2, which has a partially positive charge, interactspreferentially with the anion, undoubtedly due to the negativecharges on the atoms of the anion. With this point of view, tounderstand the higher solubility of CO2 in [emim][HSO4]compared with [emim][MeSO4], one needs to examine thecharge on the oxygen atoms in the two anions. Ficke andBrennecke10 studied some properties of ILs with these anionswith water and reported CHELPG charges for some of theanions and cations studied in this article. They observed thatoxygen atoms are much more negative in [HSO4]

− (−0.72)than in [MeSO4]

− (−0.50) or ethylsulfate ([EtSO4]−) (−0.56).

In fact, the total negative charge on the oxygen atoms in[HSO4]

− is greater than in [MeSO4]−. These results point out

two important things: (1) increasing the length of the alkylchain attached to the sulfate anion affects the partial charges onthe oxygen atoms of the sulfate functional group, and (2) freevolume is not the only variable that affects the solubility of CO2in ILs. Higher free volume means more “void space” available inthe IL liquid phase for CO2 to fill. However, the molar volume(and thus the free volume) of [emim][HSO4] is less than[emim][MeSO4] (see Tables 3 and 4), even though the CO2solubility is much higher. Therefore, enthalpic interactionsbetween CO2 and the anion, in this case likely due to Lewisacid−Lewis basic (LA−LB) interactions, are also important indetermining physical CO2 solubility in ILs. The more negativecharge on the anion oxygens of [emim][HSO4] likely leads tostronger interactions between the partially positive charge ofthe carbon of CO2 and compensate for the smaller free volume,explaining the higher solubility of CO2 in [emim][HSO4].Raveendran and Wallen46 also established that during theinteraction of a sulfonyl group with CO2 the SO bond ishighly polarized, resulting in a stronger LA−LB interactions.The explanation for the solubility of CO2 in [emim][MeSO3]

compared with [emim][MeSO4] can also be made byexamining the partial charges. The partial charge of the oxygenatoms in [MeSO3]

− are large (−0.79), but there are only threeoxygen atoms compared to the four negatively charge oxygenatoms in [emim][MeSO4] (each with a charge of about −0.50).As a result, there is almost no difference in CO2 solubility inthese two ILs at 313 K. At 298 K there is a higher CO2

Table 6. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[emim][DEP] (2)a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

298.3 0.0995 238298.4 1.20 0.22 189 12298.5 2.46 0.41 146 12298.6 3.31 0.49 131 14298.6 4.89 0.61 104 14298.7 5.47 0.65 97 14313.8 0.0981 242313.8 1.30 0.19 198 12313.7 2.54 0.30 175 16313.7 3.59 0.39 157 18313.7 4.60 0.46 141 20313.7 5.79 0.54 125 20313.7 7.12 0.60 111 22333.3 0.0987 241333.3 1.22 0.11 218 12333.3 2.59 0.19 202 20333.3 4.01 0.30 179 24333.3 4.81 0.37 161 24333.2 5.57 0.41 155 26

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Table 7. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[OHemim][TFA] (2)a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

299.2 0.0995 181299.2 1.19 0.26 136 6299.3 2.40 0.38 118 8299.4 3.84 0.50 98 8299.4 5.31 0.58 87 10313.3 0.0989 180313.8 1.31 0.13 162 8313.0 2.46 0.24 145 10313.1 3.90 0.34 129 12313.4 5.29 0.46 109 12313.4 6.67 0.54 96 14314.2 8.12 0.63 80 14333.6 0.0984 187333.6 1.14 0.09 172 8333.5 2.49 0.22 151 10333.6 3.97 0.32 134 12333.5 5.37 0.41 119 14333.4 6.86 0.49 106 14333.6 8.23 0.61 82 12

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXE

solubility in the IL with the [MeSO4]− anion at higher

pressures.Figure 2 also shows the solubility of CO2 in 1-ethyl-3-

methylimidazolium trifluoroacetate ([emim][TFA]) at pres-sures up to 2 MPa reported by Shiflett and Yokozeki.30 Thesedata are at a slightly higher temperature than the other datashown in the figure (323 K instead of 313 K). Nonetheless, it isclear that CO2 solubility in [emim][TFA] is similar to[emim][MeSO3], [emim][MeSO4], and [emim][SCN] andnot nearly as high as in other fluorinated ILs (e.g.,[hmim][Tf2N]). Shiflett and Yokozeki30 explained thesolubility of CO2 in [emim][TFA] by considering that theCO2 is acting as a Lewis acid (rather than a Lewis base), whichcould reduce its solubility in the IL, despite the favorable effectof the fluorinated group.In another study on the effect of the acetate ([AC]−) and

[TFA]− anions on CO2 solubility at higher pressures, Carvalhoet al.24 identified the acetate group as the most favored “CO2-philic” functional group based on the carbonyl functionality.They explained that the higher solubility of CO2 in [AC]− ILscomparing with [TFA]− ILs is due to chemisorption effects. Inour laboratory, acetate anions (but not trifluoroacetate anions)were found to react with dialkylimidazolium cations bydeprotonation at the C2 position. The resulting carbene canreadily react with CO2. Thus, it is not necessarily the acetateanion that is CO2-philic, but the acetate anion is sufficientlybasic to produce the imidazolium carbene, which is “CO2-philic”. Unfortunately, this mechanism is not completelyreversible since the formed acetic acid from the deprotonationis volatile and can evaporate. In fact, the smell of acetic acid waspervasive when experiments were conducted with acetate ILsand CO2 in our laboratory. However, the chemistry of CO2with acetate ILs remains an active area of investigation fornumerous research groups.30,71−74

Another interesting comparison shown in Figures 1 and 2 isbetween the solubility of CO2 in [OHemim][TFA] (measuredin this study) and the solubility of CO2 in [emim][TFA](reported by Shiflett and Yokozeki).30 The addition of the OHgroup to the cation alkyl chain does not increase the solubilityvery much; that is, at 298 K and 1.3 MPa the solubility of CO2in [emim][TFA] is 0.20 mole fraction, while at the samepressure and temperature the solubility in [OHemim][TFA] isabout 0.26 mole fraction. This is consistent with many reportswhere the primary interactions of CO2 in ILs are with theanion. Once again, referring the partial charges reported byFicke and Brennecke,10 this can be understood in more detail.The addition of the OH group to the cation does not affect thepartial charge of the acidic hydrogen of the cation at position 2.The acidic hydrogen at the C2 position is the primary locationfor cation−anion interactions in imidazolium ILs. Therefore,the association of the anion with the cation near the C2hydrogen should be similar for [OHemim][TFA] and [emim]-[TFA]. However, in [OHemim][TFA] the oxygen atom of theOH group has a highly negative partial charge (−0.68) whichcould interact with the carbon of CO2 and increase itssolubility. Also, as shown in Figures 1 and 2, the solubility doesnot increase substantially with the addition of the hydroxylgroup. The most probable reason for that is the presence of ahighly positive charge (0.44) on the hydrogen of the OHgroup, which provides a new location on the cation forenhanced cation/anion interactions. Therefore, the anionassociation at the OH group precludes any significant OH +CO2 interactions, resulting in only a modest increase in CO2

solubility. Clearly, there is no chemical reaction of the CO2with the hydroxyl group on the cation.The effect of the anion can also be examined by comparing

several ILs with the common 1-hexyl-3-methylimidazolium([hmim]+) cation. Figure 3 shows the solubility of CO2 in

different [hmim]+ ILs at 313 K with the following anions:[Tf2N]

−,36 [OTf]− and [eFAP]−.39 The solubility behavior is[OTf]− < [Tf2N]

− < [eFAP]−. Increasing the fluorination ofthe anion increases the solubility dramatically, as reported byMuldoon et al.39 We attribute this to both stronger interactionsof the CO2 with the electronegative fluorine atoms and thehigher free volume of anions with fluoroalkyl chains.

Effect of Cation Alkyl Chain Length. Figure 4 comparesthe solubility of CO2 in [emim][Tf2N]

26 and [emim][OTf]27

with [hmim][Tf2N]36 and [hmim][OTf] at 313 K. Solubility

data for [hmim][OTf] is shown in Table 8. All systems showthe same expected tendency; the solubility of CO2 increaseswith higher pressures. The measurements indicate a highersolubility of CO2 in both ILs with the [Tf2N]

− anion. The

Figure 3. Solubility of CO2 in ILs with several different anions and acommon cation ([hmim]+) at 313 K. □, [hmim][OTf]; ○,[hmim][Tf2N];

36 ▲, [hmim][eFAP].39

Figure 4. Comparison of the solubility of CO2 in [emim]+ and[hmim]+ based ILs at 313 K. ○, [hmim][Tf2N];

36 ●, [emim]-[Tf2N];

26 ▼, [emim][OTf];27 □, [hmim][OTf].

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXF

solubility in both [hmim]+ ILs appears to be slightly higherthan the corresponding [emim]+ ILs, but the difference is verysmall. For example, the solubility of CO2 in [emim][Tf2N] at313 K and 5.7 MPa is 0.59 mole fraction, while in[hmim][Tf2N] it is 0.63 mole fraction at a similar pressure,so the solubility just increases 7 %. Several authors havediscussed this increase in terms of a larger free volume, whichappears to be a reasonable explanation.28,34,62,63

The effect of cation alkyl chain length is further explored inFigure 5, where the new data for [hmim][OTf] are comparedat 313 K with measurements for [emim][OTf] and [bmim]-[OTf] from the literature.27,31,36 At higher pressures, thesolubility of CO2 in [hmim][OTf] is slightly higher than in theother two ILs. When the pressure is 8 MPa the solubility ofCO2 in [hmim][OTf] is 0.60 mole fraction, while for

[bmim][OTf] it is around 0.54 mole fraction. Perhaps thedata for [bmim][OTf] from Soriano et al.31 would show betterthe expected trend of increasing solubility with a longer alkylchain length if pressures were extended to values higher than 6MPa. Nonetheless, the increase in solubility with increasingalkyl chain length is small, frequently not much greater than theuncertainty in the data. Although not shown in Figure 5, onecould compare the new data for [hmim][OTf] from this workwith the obtained by Shin and Lee33 for the same IL, wheretheir results show a 20 % higher solubility values than the onesobtained in this study. These authors report solubility data forthe [hmim][Tf2N] + CO2 system which are also higher thanthose reported by Aki et al.36 Therefore, some systematic errorin the data reported by Shin and Lee is suspected.Comparing the solubility of CO2 in [emim][OTf]27 and

[emim][Tf2N]26 (Figure 4) reveals a much larger difference

than one might expect. This difference could be attributed tostrong Lewis acid−base interactions between SO and CO2,as proposed by Raveendran and Wallen46 or to large differencesin the free volume. The reported partial molar enthalpy fordissolution of CO2 in [hmim][Tf2N] is around −12.1 kJ·mol−1.39 A rough estimation of those enthalpy effects (asexplained below) yields a value around −11 kJ·mol−1 for[hmim][OTf]. Thus, these two ILs have very similar CO2 + ILinteraction energies. Therefore, the difference in free volume isa good explanation for the higher solubility of CO2 in [Tf2N]

−

than [OTf]− ILs.Phosphonium versus Imidazolium Based ILs. Figure 6

shows the solubility of CO2 in [P66614][Tf2N], [hmim]-

[Tf2N],36 [emim][MeSO3], [P66614][MeSO3],

37 [emim][DEP],and [P2444][DEP]. Our new data for the solubility of CO2 in[P66614][Tf2N] and [P2444][DEP] are shown in Tables 9 and 10,respectively. For all of the cases shown, the solubility in thephosphonium based ILs is higher than in the equivalentimidazolium based ILs. The figure also includes the datareported by Song et al.32 and Carvalho et al.75 for [P66614]-[Tf2N]. Solubility data from this work is comparable withCarvalho et al., but Song et al. show a higher solubility atpressures up to 5 MPa compared with the other two, while at

Table 8. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[hmim][OTf] (2)a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

297.8 0.0988 270298.1 1.20 0.33 189 14298.2 2.31 0.47 155 16298.3 3.55 0.60 124 16298.4 4.99 0.68 105 16298.2 5.96 0.76 86 14313.1 0.0980 268313.1 1.04 0.05 262 20313.1 2.37 0.24 218 26313.1 3.29 0.30 205 30313.1 4.35 0.43 173 30313.1 5.93 0.49 163 38313.1 8.02 0.60 134 46

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Figure 5. Solubility of CO2 imidazolium-based ILs with a commonanion ([OTf]−) at 313 K. □, [hmim][OTf]; ▷, [bmim][OTf];36 ▶,[bmim][OTf];31 ▼, [emim][OTf].27

Figure 6. Comparison of the solubility of CO2 in phosphonium andimidazolium based ILs at 313 K. ×, [emim][MeSO3]; ★, [P66614]-[MeSO3];

37 ○, [hmim][Tf2N];36 solid ⬠, [P2444][DEP]; −, [emim]-

[DEP];◀, [P66614][Tf2N];◁, [P66614][Tf2N];32 ⬢, [P66614][Tf2N].

75

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXG

higher pressures their results are consistent with the data fromCarvalho et al.Figure 7 shows the molar volumes of the IL + CO2 mixtures

for the ILs investigated in this work and for [hmim][Tf2N].36

As a first approximation, one can presume that molar volumerelates to the “free volume” available in the IL mixtures.[P66614][Tf2N] clearly has the highest molar volume and thehighest CO2 solubility. This increased free volume madeavailable by the “floppy” long alkyl chains is the likelyexplanation for the higher solubility of CO2 in [P66614][Tf2N]than in [hmim][Tf2N]. Likewise, [P2444][DEP] + CO2 mixtureshave much larger molar volumes and somewhat higher CO2solubilities than [emim][DEP] + CO2 mixtures. Althoughmolar volumes are not available for the [P66614][MeSO3]investigated by Zhang et al.,37 larger values than those reportedhere for [emim][MeSO3] can be anticipated. Thus, we believe

the primary explanation for higher CO2 solubilities inphosphonium ILs than imidazolium ILs with the same anionis the larger free volume.The case of the alkyl phosphate anion is also interesting.

Park76 studied the interactions between phosphorus com-pounds and CO2; their results shows strong interactions

Table 9. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[P66614][Tf2N] (2)

a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

298.1 0.0999 754298.1 1.57 0.52 372 56298.1 2.92 0.68 259 48298.1 4.27 0.79 180 34298.1 5.77 0.85 139 32298.2 0.0991 741298.2 1.60 0.53 359 54298.2 2.93 0.70 241 42298.2 4.28 0.79 173 32298.2 5.79 0.85 132 30313.2 0.0994 747313.2 1.67 0.46 421 76313.2 2.99 0.60 321 74313.2 4.37 0.69 258 70313.2 5.75 0.77 202 60313.2 7.29 0.83 155 52313.2 0.0991 763313.2 1.68 0.47 426 82313.2 3.07 0.59 334 86313.2 4.50 0.68 274 86313.2 5.90 0.76 214 74313.2 7.31 0.82 164 62333.1 0.0992 762333.1 1.65 0.39 479 76333.1 3.09 0.52 384 82333.1 4.50 0.60 329 88333.1 5.92 0.68 267 80333.1 7.28 0.75 216 68333.1 0.0993 754333.1 1.72 0.37 485 100333.1 3.17 0.51 394 110333.1 4.59 0.61 320 106333.2 6.03 0.70 259 94333.1 7.29 0.76 208 78

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Table 10. Experimental VLE Data for Temperature T,Pressure p, Mole Fraction of CO2 x1, Molar Volume Vm, andthe Combined Expanded Uncertainty of the Molar VolumeΔVm, for the Liquid Phase of the Binary System CO2 (1) +[P2444][DEP] (2)

a

T p Vm ΔVm

K MPa x1 cm3·mol−1 cm3·mol−1

298.6 0.0985 390298.8 1.13 0.20 316 16299.0 1.96 0.37 256 18299.1 2.76 0.46 224 20299.2 3.53 0.52 202 22299.3 4.32 0.59 179 22299.4 5.11 0.65 160 22313.3 0.0998 394313.3 1.23 0.24 304 26313.3 2.55 0.38 256 32313.3 3.81 0.48 224 36313.3 4.74 0.54 203 38313.3 6.01 0.62 173 38313.3 7.37 0.71 139 34333.7 0.0995 396333.7 1.31 0.10 363 32333.7 2.71 0.28 295 38333.6 3.64 0.37 264 40333.7 4.89 0.42 247 48333.7 6.08 0.47 232 54333.7 7.50 0.56 194 52333.7 8.41 0.61 173 50

aStandard uncertainties u are u(T) = 1 K, u(p) = 0.01 MPa, and theestimated combined expanded uncertainty in CO2 solubility, u(x) =0.01 mole fraction for p > 3 MPa and u(x) = 0.02 mole fraction for p <3 MPa, as explained in the text.

Figure 7. Molar volumes of IL + CO2 mixtures at 313 K. △,[OHemim][TFA]; □, [hmim][OTf]; ◆, [emim][DEP]; ■, [emim]-[MeSO4]; solid ⬠, [P2444][DEP]; +, [emim][HSO4]; ◀, [P66614]-[Tf2N]; ×, [emim][MeSO3]; ◇, [emim][SCN]; ○, [hmim][Tf2N].

36

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXH

between the partial charges on the oxygen atom of the

phosphate and the carbon of CO2. This could explain why the

difference in solubility of CO2 in [emim][DEP] and

[P2444][DEP] is smaller than in the other cases. Despite the

larger free volume of [P2444][DEP], the difference in solubility

is not great. Therefore, [P2444][DEP] is a good example of how

both free volume and strong anion/CO2 interactions determine

the CO2 solubility.

Expansivity. Figure 8 quantifies the volume expansion (orlack thereof) of the ILs investigated upon the addition of CO2.The molar expansivity was calculated according to eq 1,36

Δ =−

·VV

V T P x V T P

V T P%

( , , ) ( , )

( , )100m,L 1 m,2 0

m,2 0 (1)

where Vm,L represents the molar volume of the liquid at a givenpressure, temperature, and mole fraction of CO2 and Vm,2represents the molar volume of the pure IL at the sametemperature and atmospheric pressure. The values are negativebecause the molar volumes of the mixtures are much smallerthan the pure component molar volumes. The molarexpansivities do not vary significantly with temperature so alltemperatures investigated are shown with a single symbol foreach IL in Figure 8.Results indicate that the volume of the ILs does not change

dramatically upon the addition of CO2, even at highconcentrations of CO2. This is in sharp contrast to organicsolvents (e.g., acetonitrile) which expand significantly uponaddition of CO2. The other striking feature is that the molarexpansivity is essentially the same for all of the ILs investigated.Once again, this is simply a reflection of the fact that none ofthe ILs increase very much in volume when CO2 is added, dueto the strong electrostatic forces between the ions.The volume expansivity is directly related to the overall

solvent strength of the IL + CO2 mixtures and is the principalconcept for the CO2 antisolvent processes. Lower solvationstrength (which corresponds to large volume expansion) meansa high probability of precipitation of a dissolved solid

compound from the liquid phase. Since the solvation strengthfor these IL does not change substantially upon addition ofCO2, precipitation of solutes from ILs by addition of CO2would likely be quite difficult. However, it has been shown thataddition of small amounts of cosolvent can compensate for thiseffect.17,20

Enthalpy of Gas Dissolution. Enthalpies of dissolution ofCO2 + IL systems can be calculated from the temperaturedependence of Henry’s law constants. The value of the partialmolar enthalpy indicates the strength of the physicalinteractions between the IL and the CO2.

38 Of course, theapplication of Henry’s law is only valid at low pressures and lowsolubilities. However, an approximation at higher solubilitiescould be made by using eq 2,

Δ = − = ∂∂

⎛⎝⎜

⎞⎠⎟H H H R

PT

ln(ln )

x1 1 1

id

1 (2)

where H1 is the partial molar enthalpy of the gas solute insolution, Hid is the enthalpy of the pure ideal gas, T is thetemperature of the system, P is the partial pressure of the gas,and x1 is the mole fraction of gas dissolved in the IL.36 Ourestimates for the partial molar enthalpies of the systemsinvestigated here show a significant degree of uncertainty sinceonly three temperatures were investigated. Nonetheless, thevalues seem quite reasonable and ranged between −12 kJ·mol−1and −27 kJ·mol−1. However, the uncertainty in the values is toolarge to compare the relative strength of the physicalinteractions of CO2 with the different ILs.

CO2 Solubility as a Function of Molality. Carvalho andCoutinho57 argued that the solubility of CO2 in ILs onlydepends on free volume for pressures up to 5 MPa and the bestway to show this is to plot the solubility as a function ofmolality, where all the data should fall on a common curve.Figure 9 shows the solubilities of CO2 in the ILs for all of the

ILs investigated here, plotted as a function of molality (mol·kg−1). The data are reported in this way to minimize thedifferences in the solubilities due to differences in the molecularweights of the ILs. Clearly, when data for a wide variety ofdifferent ILs are included on the graph, there is no clear trend.The data do not fall on a single curve, as argued by Carvalho

Figure 8. Volume expansivity of the ILs upon the addition of CO2. △,[OHemim][TFA]; □, [hmim][OTf]; ◆, [emim][DEP]; ■, [emim]-[MeSO4]; solid ⬠, [P2444][DEP]; +, [emim][HSO4]; ◀, [P66614]-[Tf2N]; ○, [hmim][Tf2N].

36

Figure 9. Solubilities of CO2 as a function of molality at 313 K. △,[OHemim][TFA]; □, [hmim][OTf]; ◆, [emim][DEP]; ■, [emim]-[MeSO4]; solid ⬠, [P2444][DEP].

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXI

and Coutinho.57 This has also been point out recently by deLoos and co-workers.77 Thus, it is clear that CO2 solubility inILs depends both on free volume and the strength ofintermolecular and intramolecular interactions. CO2 is non-polar but has a significant quadrupole moment so thecontribution of enthalpic interactions is expected.

■ CONCLUSIONSThis article presents new data for the CO2 solubility and molarvolumes for nine ILs ([emim][HSO4], [emim][MeSO4],[emim][MeSO3], [emim][SCN], [emim][DEP], [P2444]-[DEP], [hmim][OTf], [OHemim][TFA], and [P66614][Tf2N])at 298 K, 313 K, and 333 K and pressures to 9 MPa. CO2solubility increases with increasing pressure and decreases withincreasing temperature, as expected. The CO2 solubility in[emim][HSO4] is significantly higher than in [emim][MeSO4],even though [emim][MeSO4] has a much larger molar (andfree) volume. This can be explained by the larger negativepartial charges on the oxygen atoms in [HSO4]

− than in[MeSO4]

−. The addition of a hydroxyl group ([OHemim]-[TFA]) does not increase the CO2 solubility significantlycompared to [emim][TFA]. We interpret this as the anionassociation with the OH group precluding any significant OH +CO2 interactions, resulting in only a modest increase in CO2solubility. Clearly, there is no chemical reaction of the CO2with the hydroxyl group on the cation. Increasing the alkylchain length on the cation only slightly increases the CO2solubility, as has been observed previously. Increasing theconcentration of fluoroalkyl chains (e.g., [emim][Tf2N]compared with [emim][OTf]) has a much larger effect. CO2solubility in tetra-alkylphosphonium ILs is higher than in theimidazolium counterparts. This is attributed to the larger freevolume afforded by the multiple alkyl chains on thephosphonium cations. None of the ILs expand to anysignificant extent, even when very large amounts of CO2dissolve, as has been observed previously. CO2 solubilities inILs do not collapse on a single curve when plotted as a functionof molality, as suggested by Carvalho and Coutinho.57 Both freevolume and enthalpic interactions play a role in determiningCO2 solubility in ILs.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: jfb@nd.edu. Telephone: +1 (574) 631-5847. Fax: +1(574) 631-8366.Present AddressI.M.: School of Chemical Engineering, Universidad del Valle,P.O. Box 25360, Cali, Colombia.NotesThe authors declare no competing financial interest.

■ REFERENCES(1) Earle, M.; Forestier, A.; Olivier-Bourbigou, H.; Wasserscheid, P.Organic Synthesis. In Ionic Liquids in Synthesis; Wiley-VCH VerlagGmbH & Co. KGaA: Weinheim, Germany, 2003; pp 174−288.(2) Bates, E. D.; Mayton, R. D.; Ntai, I.; Davis, J. H. CO2 Capture bya Task-Specific Ionic Liquid. J. Am. Chem. Soc. 2002, 124, 926−927.(3) Anderson, J. L.; Dixon, J. K.; Maginn, E. J.; Brennecke, J. F.Measurement of SO2 Solubility in Ionic Liquids. J. Phys. Chem. B 2006,110, 15059−15062.(4) Shiflett, M. B.; Yokozeki, A. Separation of Carbon Dioxide andSulfur Dioxide Using Room-Temperature Ionic Liquid [bmim]-[MeSO4]. Energy Fuels 2010, 24, 1001−1008.

(5) Anthony, J. L.; Brennecke, J. F.; Holbrey, J. D.; Maginn, E. J.;Mantz, R. A.; Rogers, R. D.; Trulove, P. C.; Visser, A. E.; Welton, T.Physicochemical Properties of Ionic Liquids. In Ionic Liquids inSynthesis; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim,Germany, 2003; pp 81−93.(6) Maginn, E. Design and Evaluation of Ionic Liquids as Novel CO2

Absorbents. Quarterly Technical Report (10/01/06−12/31/06);DOE Award Number: DE-FG26-04NT42122; National EnergyTechnology Laboratory: Morgantown, WV, 2007.(7) Anthony, J. L.; Aki, S. N. V. K.; Maginn, E. J.; Brennecke, J. F.Feasibility of using ionic liquids for carbon dioxide capture. Int. J.Environ. Technol. Manage. 2004, 4, 105−115.(8) Bara, J. E.; Carlisle, T. K.; Gabriel, C. J.; Camper, D.; Finotello,A.; Gin, D. L.; Noble, R. D. Guide to CO2 Separations in Imidazolium-Based Room-Temperature Ionic Liquids. Ind. Eng. Chem. Res. 2009,48, 2739−2751.(9) Ficke, L. E.; Novak, R. R.; Brennecke, J. F. Thermodynamic andThermophysical Properties of Ionic Liquid + Water Systems. J. Chem.Eng. Data 2010, 55, 4946−4950.(10) Ficke, L. E.; Brennecke, J. F. Interactions of Ionic Liquids andWater. J. Phys. Chem. B 2010, 114, 10496−10501.(11) Reetz, M. T.; Wiesenhofer, W.; Francio, G.; Leitner, W.Continuous Flow Enzymatic Kinetic Resolution and EnantiomerSeparation using Ionic Liquid/Supercritical Carbon Dioxide Media.Adv. Synth. Catal. 2003, 345, 1221−1228.(12) Wong, H.; Han, S.; Livingston, A. G. The effect of ionic liquidson product yield and catalyst stability. Chem. Eng. Sci. 2006, 61, 1338−1341.(13) Sheldon, R. Catalytic reactions in ionic liquids. Chem. Commun.2001, 2399−2407.(14) Blanchard, L. A.; Brennecke, J. F. Recovery of Organic Productsfrom Ionic Liquids Using Supercritical Carbon Dioxide. Ind. Eng.Chem. Res. 2001, 40, 287−292.(15) Blanchard, L. A.; Hancu, D.; Beckman, E. J.; Brennecke, J. F.Green processing using ionic liquids and CO2. Nature 1999, 399, 28−29.(16) Scurto, A. M.; Aki, S. N. V. K.; Brennecke, J. F. CO2 as aSeparation Switch for Ionic Liquid/Organic Mixtures. J. Am. Chem.Soc. 2002, 124, 10276−10277.(17) Saurer, E. M.; Aki, S. N. V. K.; Brennecke, J. F. Removal ofammonium bromide, ammonium chloride, and zinc acetate from ionicliquid/organic mixtures using carbon dioxide. Green Chem. 2006, 8,141−143.(18) Plechkova, N. V; Seddon, K. R. Applications of ionic liquids inthe chemical industry. Chem. Soc. Rev. 2008, 37, 123−150.(19) Scurto, A. M.; Aki, S. N. V. K.; Brennecke, J. F. Carbon dioxideinduced separation of ionic liquids and water. Chem. Commun. 2003,572−573.(20) Wu, W.; Li, W.; Han, B.; Jiang, T.; Shen, D.; Zhang, Z.; Sun, D.;Wang, B. Effect of Organic Cosolvents on the Solubility of IonicLiquids in Supercritical CO2. J. Chem. Eng. Data 2004, 49, 1597−1601.(21) Kumełan, J.; Tuma, D.; Maurer, G. Partial molar volumes ofselected gases in some ionic liquids. Fluid Phase Equilib. 2009, 275,132−144.(22) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. Solubilities andThermodynamic Properties of Gases in the Ionic Liquid 1-n-Butyl-3-methylimidazolium Hexafluorophosphate. J. Phys. Chem. B 2002, 106,7315−7320.(23) Ficke, L. E.; Rodríguez, H.; Brennecke, J. F. Heat Capacities andExcess Enthalpies of 1-Ethyl-3-methylimidazolium-Based Ionic Liquidsand Water. J. Chem. Eng. Data 2008, 53, 2112−2119.(24) Carvalho, P. J.; Alvarez, V. H.; Schroder, B.; Gil, A. M.;Marrucho, I. M.; Aznar, M.; Santos, L. M. N. B. F.; Coutinho, J. A. P.Specific Solvation Interactions of CO2 on Acetate and TrifluoroacetateImidazolium Based Ionic Liquids at High Pressures. J. Phys. Chem. B2009, 113, 6803−6812.(25) Revelli, A.-L.; Mutelet, F.; Jaubert, J.-N. High Carbon DioxideSolubilities in Imidazolium-Based Ionic Liquids and in Poly(ethyleneglycol) Dimethyl Ether. J. Phys. Chem. B 2010, 114, 12908−12913.

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXJ

(26) Schilderman, A. M.; Raeissi, S.; Peters, C. J. Solubility of carbondioxide in the ionic liquid 1-ethyl-3-methylimidazolium bis-(trifluoromethylsulfonyl)imide. Fluid Phase Equilib. 2007, 260, 19−22.(27) Soriano, A. N.; Doma, B. T.; Li, M.-H. Carbon dioxide solubilityin 1-ethyl-3-methylimidazolium trifluoromethanesulfonate. J. Chem.Thermodyn. 2009, 41, 525−529.(28) Carvalho, P. J.; Alvarez, V. H.; Machado, J. J. B.; Pauly, J.;Daridon, J.-L.; Marrucho, I. M.; Aznar, M.; Coutinho, J. A. P. Highpressure phase behavior of carbon dioxide in 1-alkyl-3-methylimida-zolium bis(trifluoromethylsulfonyl)imide ionic liquids. J. Supercrit.Fluids 2009, 48, 99−107.(29) Kumełan, J.; Tuma, D.; Maurer, G. Solubility of CO2 in theIonic Liquids [bmim][CH3SO4] and [bmim][PF6]. J. Chem. Eng. Data2006, 51, 1802−1807.(30) Shiflett, M. B.; Yokozeki, A. Phase Behavior of Carbon Dioxidein Ionic Liquids: [emim][Acetate], [emim][Trifluoroacetate], and[emim][Acetate] + [emim][Trifluoroacetate] Mixtures. J. Chem. Eng.Data 2009, 54, 108−114.(31) Soriano, A. N.; Doma, B. T.; Li, M.-H. Carbon dioxide solubilityin some ionic liquids at moderate pressures. J. Taiwan Inst. Chem. Eng.2009, 40, 387−393.(32) Song, H. N.; Lee, B.-C.; Lim, J. S. Measurement of CO2

Solubility in Ionic Liquids: [BMP][TfO] and [P14,6,6,6][Tf2N] byMeasuring Bubble-Point Pressure. J. Chem. Eng. Data 2010, 55, 891−896.(33) Shin, E.-K.; Lee, B.-C. High-Pressure Phase Behavior of CarbonDioxide with Ionic Liquids: 1-Alkyl-3-methylimidazolium Trifluor-omethanesulfonate. J. Chem. Eng. Data 2008, 53, 2728−2734.(34) Blanchard, L. A.; Gu, Z.; Brennecke, J. F. High-Pressure PhaseBehavior of Ionic Liquid/CO2 Systems. J. Phys. Chem. B 2001, 105,2437−2444.(35) Ren, W.; Sensenich, B.; Scurto, A. M. High-pressure phaseequilibria of {carbon dioxide (CO2) + n-alkyl-imidazolium bis-(trifluoromethylsulfonyl)amide} ionic liquids. J. Chem. Thermodyn.2010, 42, 305−311.(36) Aki, S. N. V. K.; Mellein, B. R.; Saurer, E. M.; Brennecke, J. F.High-Pressure Phase Behavior of Carbon Dioxide with Imidazolium-Based Ionic Liquids. J. Phys. Chem. B 2004, 108, 20355−20365.(37) Zhang, S.; Chen, Y.; Ren, R. X.-F.; Zhang, Y.; Zhang, J.; Zhang,X. Solubility of CO2 in Sulfonate Ionic Liquids at High Pressure. J.Chem. Eng. Data 2005, 50, 230−233.(38) Anthony, J. L.; Anderson, J. L.; Maginn, E. J.; Brennecke, J. F.Anion Effects on Gas Solubility in Ionic Liquids. J. Phys. Chem. B 2005,109, 6366−6374.(39) Muldoon, M. J.; Aki, S. N. V. K.; Anderson, J. L.; Dixon, J. K.;Brennecke, J. F. Improving Carbon Dioxide Solubility in Ionic Liquids.J. Phys. Chem. B 2007, 111, 9001−9009.(40) Kilaru, P. K.; Condemarin, R. A.; Scovazzo, P. Correlations ofLow-Pressure Carbon Dioxide and Hydrocarbon Solubilities inImidazolium-, Phosphonium-, and Ammonium-Based Room-Temper-ature Ionic Liquids. Part 1. Using Surface Tension. Ind. Eng. Chem. Res.2008, 47, 900−909.(41) Yokozeki, A.; Shiflett, M. B.; Junk, C. P.; Grieco, L. M.; Foo, T.Physical and Chemical Absorptions of Carbon Dioxide in Room-Temperature Ionic Liquids. J. Phys. Chem. B 2008, 112, 16654−16663.(42) Jacquemin, J.; Husson, P.; Mayer, V.; Cibulka, I. High-PressureVolumetric Properties of Imidazolium-Based Ionic Liquids: Effect ofthe Anion. J. Chem. Eng. Data 2007, 52, 2204−2211.(43) Gardas, R. L.; Costa, H. F.; Freire, M. G.; Carvalho, P. J.;Marrucho, I. M.; Fonseca, I. M. A.; Ferreira, A. G. M.; Coutinho, J. A.P. Densities and Derived Thermodynamic Properties of Imidazolium-,Pyridinium-, Pyrrolidinium-, and Piperidinium-Based Ionic Liquids. J.Chem. Eng. Data 2008, 53, 805−811.(44) Crosthwaite, J. M.; Muldoon, M. J.; Dixon, J. K.; Anderson, J. L.;Brennecke, J. F. Phase transition and decomposition temperatures,heat capacities and viscosities of pyridinium ionic liquids. J. Chem.Thermodyn. 2005, 37, 559−568.

(45) Rodríguez, H.; Brennecke, J. F. Temperature and CompositionDependence of the Density and Viscosity of Binary Mixtures of Water+ Ionic Liquid. J. Chem. Eng. Data 2006, 51, 2145−2155.(46) Raveendran, P.; Wallen, S. L. Cooperative C−H···O HydrogenBonding in CO2−Lewis Base Complexes: Implications for Solvation inSupercritical CO2. J. Am. Chem. Soc. 2002, 124, 12590−12599.(47) Yokozeki, A.; Kasprzak, D. J.; Shiflett, M. B. Thermal effect onC-H stretching vibrations of the imidazolium ring in ionic liquids. Phys.Chem. Chem. Phys. 2007, 9, 5018−5026.(48) Kazarian, S. G.; Briscoe, B. J.; Welton, T. Combining ionicliquids and supercritical fluids: ATR-IR study of CO dissolved in twoionic liquids at high pressures. Chem. Commun. 2000, 2047−2048.(49) Cammarata, L.; Kazarian, S. G.; Salter, P. A.; Welton, T.Molecular states of water in room temperature ionic liquids. Phys.Chem. Chem. Phys. 2001, 3, 5192−5200.(50) Crowhurst, L.; Mawdsley, P. R.; Perez-Arlandis, J. M.; Salter, P.A.; Welton, T. Solvent-solute interactions in ionic liquids. Phys. Chem.Chem. Phys. 2003, 5, 2790−2794.(51) Dong, K.; Zhang, S.; Wang, D.; Yao, X. Hydrogen Bonds inImidazolium Ionic Liquids. J. Phys. Chem. A 2006, 110, 9775−9782.(52) Ferguson, L.; Scovazzo, P. Solubility, Diffusivity, andPermeability of Gases in Phosphonium-Based Room TemperatureIonic Liquids: Data and Correlations. Ind. Eng. Chem. Res. 2007, 46,1369−1374.(53) Simoni, L. D.; Ficke, L. E.; Lambert, C. A.; Stadtherr, M. A.;Brennecke, J. F. Measurement and Prediction of Vapor−LiquidEquilibrium of Aqueous 1-Ethyl-3-methylimidazolium-Based IonicLiquid Systems. Ind. Eng. Chem. Res. 2010, 49, 3893−3901.(54) Zhang, S.; Yuan, X.; Chen, Y.; Zhang, X. Solubilities of CO2 in1-Butyl-3-methylimidazolium Hexafluorophosphate and 1,1,3,3-Tetra-methylguanidium Lactate at Elevated Pressures. J. Chem. Eng. Data2005, 50, 1582−1585.(55) Zhang, J.; Zhang, S.; Dong, K.; Zhang, Y.; Shen, Y.; Lv, X.Supported Absorption of CO2 by Tetrabutylphosphonium AminoAcid Ionic Liquids. Chem.Eur. J. 2006, 12, 4021−4026.(56) Cadena, C.; Anthony, J. L.; Shah, J. K.; Morrow, T. I.;Brennecke, J. F.; Maginn, E. J. Why Is CO2 So Soluble in Imidazolium-Based Ionic Liquids? J. Am. Chem. Soc. 2004, 126, 5300−5308.(57) Carvalho, P. J.; Coutinho, J. A. P. On the Nonideality of CO2

Solutions in Ionic Liquids and Other Low Volatile Solvents. J. Phys.Chem. Lett. 2010, 1, 774−780.(58) Span, R.; Wagner, W. A New Equation of State for CarbonDioxide Covering the Fluid Region from the Triple-Point Temper-ature to 1100 K at Pressures up to 800 MPa. J. Phys. Chem. Ref. Data1996, 25, 1509−1596.(59) Scurto, A. M. High-pressure phase and chemical equilibria of β-diketone ligands and chelates with carbon dioxide. Ph.D. Thesis,University of Notre Dame, Notre Dame, IN, August 2002.(60) Ng, H.-J.; Robinson, D. B. Equilibrium-phase properties of thetoluene-carbon dioxide system. J. Chem. Eng. Data 1978, 23, 325−327.(61) Shiflett, M. B.; Yokozeki, A. Solubility of CO2 in RoomTemperature Ionic Liquid [hmim][Tf2N]. J. Phys. Chem. B 2007, 111,2070−2074.(62) Shariati, A.; Peters, C. J. High-pressure phase behavior ofsystems with ionic liquids: II. The binary system carbon dioxide+1-ethyl-3-methylimidazolium hexafluorophosphate. J. Supercrit. Fluids2004, 29, 43−48.(63) Shariati, A.; Peters, C. J. High-pressure phase behavior ofsystems with ionic liquids: Part III. The binary system carbon dioxide+ 1-hexyl-3-methylimidazolium hexafluorophosphate. J. Supercrit.Fluids 2004, 30, 139−144.(64) Alvarez, V. H.; Aznar, M. Thermodynamic modeling of vapor−liquid equilibrium of binary systems ionic liquid+supercritical {CO2 orCHF3} and ionic liquid + hydrocarbons using Peng−Robinsonequation of state. J. Chin. Inst. Chem. Eng. 2008, 39, 353−360.(65) Nebig, S.; Bolts, R.; Gmehling, J. Measurement of vapor−liquidequilibria (VLE) and excess enthalpies (HE) of binary systems with 1-alkyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide and pre-

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXK

diction of these properties and γ∞ using modified UNIFAC(Dortmund). Fluid Phase Equilib. 2007, 258, 168−178.(66) Chen, C.-C.; Simoni, L. D.; Brennecke, J. F.; Stadtherr, M. A.Correlation and Prediction of Phase Behavior of Organic Compoundsin Ionic Liquids Using the Nonrandom Two-Liquid Segment ActivityCoefficient Model. Ind. Eng. Chem. Res. 2008, 47, 7081−7093.(67) Doker, M.; Gmehling, J. Measurement and prediction of vapor−liquid equilibria of ternary systems containing ionic liquids. Fluid PhaseEquilib. 2005, 227, 255−266.(68) Kato, R.; Gmehling, J. Systems with ionic liquids: Measurementof VLE and γ∞ data and prediction of their thermodynamic behaviorusing original UNIFAC, mod. UNIFAC(Do) and COSMO-RS(Ol). J.Chem. Thermodyn. 2005, 37, 603−619.(69) Nebig, S.; Liebert, V.; Gmehling, J. Measurement and predictionof activity coefficients at infinite dilution (γ∞), vapor−liquid equilibria(VLE) and excess enthalpies (HE) of binary systems with 1,1-dialkyl-pyrrolidinium bis(trifluoromethylsulfonyl)imide using mod. UNIFAC(Dortmund). Fluid Phase Equilib. 2009, 277, 61−67.(70) Kato, R.; Gmehling, J. Measurement and correlation of vapor−liquid equilibria of binary systems containing the ionic liquids[EMIM][(CF3SO2)2N], [BMIM][(CF3SO2)2N], [MMIM]-[(CH3)2PO4] and oxygenated organic compounds respectivelywater. Fluid Phase Equilib. 2005, 231, 38−43.(71) Gurau, G.; Rodríguez, H.; Kelley, S. P.; Janiczek, P.; Kalb, R. S.;Rogers, R. D. Demonstration of Chemisorption of Carbon Dioxide in1,3-Dialkylimidazolium Acetate Ionic Liquids. Angew. Chem., Int. Ed.2011, 50, 12024−12026.(72) Janiczek, P.; Kalb, R. S.; Thonhauser, G.; Gamse, T. Carbondioxide absorption in a technical-scale-plant utilizing an imidazoliumbased ionic liquid. Sep. Purif. Technol. 2012, 97, 20−25.(73) Blath, J.; Deubler, N.; Hirth, T.; Schiestel, T. Chemisorption ofcarbon dioxide in imidazolium based ionic liquids with carboxylicanions. Chem. Eng. J. 2012, 181−182, 152−158.(74) Stevanovic, S.; Podgorsek, A.; Moura, L.; Santini, C. C.; Padua,A. A. H.; Costa Gomes, M. F. Absorption of carbon dioxide by ionicliquids with carboxylate anions. Int. J. Greenhouse Gas Control 2013, 17,78−88.(75) Carvalho, P. J.; Alvarez, V. H.; Marrucho, I. M.; Aznar, M.;Coutinho, J. A. P. High carbon dioxide solubilities in trihexylte-tradecylphosphonium-based ionic liquids. J. Supercrit. Fluids 2010, 52,258−265.(76) Park, Y. Specific interactions between phosphorus compoundsand carbon dioxide: Ab initio approach. J. Supercrit. Fluids 2005, 36,154−159.(77) Ramdin, M.; de Loos, T. W.; Vlugt, T. J. H. State-of-the-Art ofCO2 Capture with Ionic Liquids. Ind. Eng. Chem. Res. 2012, 51, 8149−8177.

Journal of Chemical & Engineering Data Article

dx.doi.org/10.1021/je400542b | J. Chem. Eng. Data XXXX, XXX, XXX−XXXL