Temperature and composition dependence of the density and viscosity of binary mixtures of water + ionic liquid

Article abstract of DOI:10.1021/je0602824

Density and viscosity were determined for binary mixtures of water and three ionic liquids: 1-ethyl-3-methylimidazolium ethylsulfate, 1-ethyl-3-methylimidazolium trifluoroacetate, and 1-ethyl-3-methylimidazolium trifluoromethanesulfonate. The experimental measurements of these properties were carried out at atmospheric pressure and temperatures from (278.15 to 348.15) K. The temperature dependence of density and viscosity for these systems can be described by an empirical second-order polynomial and by the Vogel - Fulcher - Tammann equation, respectively. Excess molar volumes and viscosity deviations were calculated and correlated by Redlich - Kister polynomial expansions. The latter correlations describe the variation of density and viscosity with composition. Comparison of the results for the three binary systems elucidates the influence of the anion on these physical properties.

Full text of DOI:10.1021/je0602824

J. Chem. Eng. Data 2006, 51, 2145-2155
2145
Temperature and Composition Dependence of the Density and Viscosity of Binary
Mixtures of Water + Ionic Liquid
He´ctor Rodr´ıguez† and Joan F. Brennecke*
Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556
Density and viscosity were determined for binary mixtures of water and three ionic liquids: 1-ethyl-3-
methylimidazolium ethylsulfate, 1-ethyl-3-methylimidazolium trifluoroacetate, and 1-ethyl-3-methylimidazolium
trifluoromethanesulfonate. The experimental measurements of these properties were carried out at atmospheric
pressure and temperatures from (278.15 to 348.15) K. The temperature dependence of density and viscosity for
these systems can be described by an empirical second-order polynomial and by the Vogel-Fulcher-Tammann
equation, respectively. Excess molar volumes and viscosity deviations were calculated and correlated by Redlich-
Kister polynomial expansions. The latter correlations describe the variation of density and viscosity with
composition. Comparison of the results for the three binary systems elucidates the influence of the anion on these
physical properties.
development of practical applications and design of processes
using water + IL systems. To date, a number of papers have
studied fundamental properties such as density or viscosity of
these binary systems: some of them report values for dilute
aqueous solutions of ILs9-11 or, in the other extreme, the effect
of water content in “pure” ILs or binary mixtures with low
fractions of water;8,12-15 whereas only some other works have
performed measurements covering the whole composition
range.16-22 Most of the papers in the latter group, however, focus
on ILs with either a halide or tetrafluoroborate as the anion,
which may lead to corrosion problems or to generation of HF
by decomposition.23 An exception is the work of Yang and co-
workers,10,19 which focused on the density of the binary system
water and 1-ethyl-3-methylimidazolium ethylsulfate ([emim]-
[EtSO4]).24
Introduction
The term ionic liquid (IL) has come to define, in recent years,
a family of substances entirely constituted by ions with melting
points below 373 K.1,2 A wide variety of ILs remain liquid at
room temperature and even at temperatures well below that.
Among their unique characteristics,2,3 the ability to solubilize
an enormous variety of compounds, a negligible vapor pressure
under common process conditions, a wide liquid range, and the
possibility to tailor their chemical structure at will to better
match a given target are particularly noteworthy. This interesting
set of properties has spurred the increasing attention drawn to
ILs by both industry and academia over the past decade. While
applications of ILs in a diverse range of fields, such as
electrochemistry, separation processes, synthesis, and catalysis,
have begun to be explored, much research must be done to fully
exploit these interesting fluids.
A major impetus for IL research is the possibility of using
them as substitutes for classical industrial solvents, most of
which are volatile organic compounds (VOCs).1 This replace-
ment would lead to a critical reduction in the emission of VOCs
to the atmosphere, currently a major source of pollution. Four
common strategies to avoid the use of conventional organic
solvents are solventless processes, water, supercritical fluids
(especially CO2), and ILs. Combinations of these options may
also be considered as environmentally favorable. In our research
group, for instance, we have investigated the use of ILs/
supercritical CO2 biphasic systems.4-7
In this work, accurate data at atmospheric pressure are
reported for the density and viscosity of mixtures of water and
three different ILs: [emim][EtSO4], 1-ethyl-3-methylimidazo-
lium trifluoroacetate ([emim][TFA]), and 1-ethyl-3-methylimi-
dazolium trifluoromethanesulfonate ([emim][OTf]) at temper-
atures from (278.15 to 348.15) K. The temperature dependence
of both properties is analyzed and correlated. Excess molar
volume and viscosity deviation, calculated from the experimental
data, are also correlated and used to characterize the influence
of the composition of the mixture on its density and viscosity.
Experimental Section
Similarly, processes that currently involve homogeneous
mixtures of water and a VOC solvent might be improved by
substitution of the VOC solvent with a hydrophilic IL. It has
been shown, for instance, that water can dramatically decrease
the viscosity of ILs, even when only present in small amounts.8
Thus, a mixture of water and a water-miscible IL might provide
an attractive low viscosity alternative to some organic solvents.
The knowledge of thermophysical properties of binary
mixtures of water plus hydrophilic ILs is needed for the
Chemicals. Two samples of [emim][EtSO4], with a nominal
purity greater than 98 %, were purchased from Solvent
Innovation; their water content, found to be (23 and 45) ppm,
was measured by Karl Fischer titration with a Metrohm 831
KF Coulometer.
Ethyltrifluoromethanesulfonate (Aldrich, 99 %) and 1-meth-
ylimidazole (Aldrich, 99 %, freshly redistilled over KOH) were
reacted at low temperature in a stirred round-bottomed flask
with a reflux condenser in the absence of any solvent to form
[emim][OTf]. A similar procedure can be found in the litera-
ture.25 Unreacted starting materials were removed by heating
under reduced pressure. The structure of the final product was
* Corresponding author. Telephone: (574)631-5847. Fax: (574)631-8366.
E-mail: jfb@nd.edu.
† Permanent address: Department of Chemical Engineering, University of
Santiago de Compostela, E-15782 Santiago de Compostela, Spain.
10.1021/je0602824 CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/10/2006

2146 Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006
Table 1. Experimental Density G, Excess Molar Volume V E, Dynamic Viscosity η, and Viscosity Deviation ∆η for the Binary System Water (1)
+ [emim][EtSO4] (2)
T/K
x1
278.15
288.15
298.15
308.15
318.15
328.15
338.15
348.15
F/g‚cm-3
1.23061
1.22621
1.21902
1.20692
1.18072
1.15490
1.10300
1.05427
1.01326
0.99406
0.0000
0.2112
0.4119
0.5939
0.7669
0.8487
0.9292
0.9683
0.9916
1.0000
1.25112
1.24659
1.23956
1.22774
1.20183
1.17563
1.12063
1.06686
1.02113
0.99999
1.24413
1.23979
1.23270
1.22080
1.19485
1.16883
1.11502
1.06325
1.01946
0.99913
1.23737
1.23299
1.22585
1.21387
1.18781
1.16192
1.10914
1.05904
1.01678
0.99707
1.22390
1.21945
1.21221
1.19997
1.17356
1.14777
1.09659
1.04901
1.00903
0.99023
1.21723
1.21274
1.20541
1.19301
1.16634
1.14052
1.08992
1.04329
1.00416
0.98571
1.21060
1.20605
1.19863
1.18603
1.15905
1.13314
1.08300
1.03715
0.99872
0.98056
1.20401
1.19940
1.19191
1.17905
1.15168
1.12570
1.07583
1.03061
0.99276
0.97484
V E/cm3‚mol-1
0.0000
0.2112
0.4119
0.5939
0.7669
0.8487
0.9292
0.9683
0.9916
1.0000
0.000
-0.211
-0.399
-0.524
-0.514
-0.450
-0.246
-0.058
-0.013
0.000
0.000
-0.215
-0.372
-0.470
-0.439
-0.371
-0.185
-0.028
-0.005
0.000
0.000
-0.195
-0.334
-0.415
-0.367
-0.300
-0.134
-0.003
0.001
0.000
0.000
-0.166
-0.278
-0.325
-0.249
-0.182
-0.054
0.034
0.000
-0.156
-0.253
-0.285
-0.196
-0.130
-0.019
0.050
0.000
-0.144
-0.231
-0.248
-0.146
-0.081
0.012
0.000
-0.136
-0.215
-0.213
-0.097
-0.037
0.043
-0.181
-0.304
-0.367
-0.306
-0.238
-0.091
0.017
0.065
0.079
0.005
0.000
0.009
0.013
0.017
0.000
0.020
0.000
0.000
0.000
0.000
η/mPa‚s
0.0000
0.2245
0.4084
0.5976
0.7675
0.9292
308
150
83
45
22
6
166
86
49
27
14
5
95
54
33
19
10
4
62
36
23
14
8
41
26
17
11
6
30
19
13
9
21
15
11
7
17
12
9
6
6
5
4
2
4
3
3
3
∆η/mPa‚s
0
0.0000
0.2245
0.4084
0.5976
0.7675
0.9292
0
-89
-100
-80
-51
-17
0
-43
-49
-40
-25
-8
0
-20
-24
-20
-13
-3
0
-6
-7
-6
-3
0
0
-4
-5
-4
-2
0
0
-2
-2
-2
-1
1
0
-1
-1
-1
0
-12
-14
-12
-7
-2
1
checked by NMR spectroscopy, and its water content was found
to be 96 ppm, measured by Karl Fischer titration.
Density Measurements. Densities were measured at atmo-
spheric pressure in a DMA 4500 Anton Paar oscillating U-tube
densitometer, which includes an automatic correction for the
viscosity of the sample. The apparatus is precise to within 1 ×
10-5 g‚cm-3, and the uncertainty of the measurements was
estimated to be better than ( 5 × 10-5 g‚cm-3. Two integrated
Pt 100 platinum thermometers provided good precision (( 0.01
K) in temperature control internally.
The procedure for the synthesis of [emim][TFA] was
analogous to that reported elsewhere.25 Equimolar amounts of
1-ethyl-3-methylimidazolium bromide (Fluka, 97.0 %), further
purified in our laboratory prior to use, and silver trifluoroacetate
(Aldrich, 98 %), used as received, were dissolved in water. Both
solutions were mixed and allowed to react for 24 h in a similar
setup to that described for the synthesis of [emim][OTf]. The
resulting solution was filtered to remove the AgBr precipitate,
and water was partially eliminated in a rotary evaporator. The
presence of bromide or silver anions in the product was tested,
respectively, by addition of AgNO3 and NaCl solutions to probe
vials containing an aliquot of the reaction product. According
to the precipitations observed in the tests, an additional amount
of one of the starting materials was added in order to neutralize
the excess of the other. The reaction and subsequent steps were
repeated until no precipitate was observed in either of the tests.
The purification was completed by drying the IL thoroughly
under high vacuum at moderate temperature for several days.
A water content of 128 ppm was assessed by Karl Fischer
titration. High-purity water, deionized through a Milli-Q Water
System, was used for all the experiments.
Viscosity Measurements. Viscosity measurements were car-
ried out in a Brookfield model DV-III Ultra Programmable (cone
and plate) rheometer, with a nominal uncertainty of ( 2 %.
Actual uncertainties for low viscosity samples with high water
contents are greater than this, as described in the next paragraph.
Temperature was controlled with a precision of ( 0.1 K by
means of a TC-602 bath thermostat with a Brookfield temper-
ature controller unit attached.
The measurement chamber for the viscometer is closed but
not hermetically sealed from the atmosphere, so the time a
sample spent in the chamber was kept to a minimum in order
to reduce moisture uptake by the IL or loss of water by
evaporation. However, the time required to ensure thermal
equilibrium at each measured temperature was unavoidable, so
variations in the concentration of the samples did occur, which
introduces an additional experimental error in the viscosity data.
Karl Fischer measurements on test samples before and after the
viscosity measurements showed a rise in water content for pure
IL samples and water loss for high water content mixtures. These
variations were particularly significant at high temperatures, with
water losses as great as 0.035 mass fraction observed. An
additional complication is that the viscometer is significantly
Mixtures of water and IL were prepared by mass using a
Mettler Toledo AE 160 balance, precise to within ( 1 × 10-4
g. Good mixing was ensured by magnetic stirring. All the
samples were prepared immediately prior to performing density
or viscosity measurements to avoid variations in composition
due to evaporation of water or pickup of water by the
hygroscopic IL.

Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006 2147
Table 2. Experimental Density G, Excess Molar Volume V E, Dynamic Viscosity η, and Viscosity Deviation ∆η for the Binary System Water (1)
+ [emim][OTf] (2)
T/K
x1
278.15
288.15
298.15
308.15
318.15
328.15
338.15
348.15
F/g‚cm-3
1.37522
1.36295
1.34693
1.32186
1.27510
1.23142
1.15294
1.08390
1.02224
0.99406
0.0000
0.2295
0.4310
0.6157
0.7825
0.8608
0.9353
0.9712
0.9924
1.0000
1.40052
1.38830
1.37238
1.34734
1.29991
1.25476
1.17190
1.09749
1.03035
0.99999
1.39204
1.37981
1.36387
1.33887
1.29174
1.24716
1.16597
1.09364
1.02864
0.99913
1.38360
1.37136
1.35539
1.33037
1.28347
1.23938
1.15965
1.08908
1.02588
0.99707
1.36690
1.35459
1.33849
1.31333
1.26662
1.22327
1.14588
1.07818
1.01786
0.99023
1.35863
1.34627
1.33007
1.30478
1.25804
1.21494
1.13849
1.07195
1.01283
0.98571
1.35043
1.33801
1.32168
1.29619
1.24935
1.20643
1.13076
1.06525
1.00721
0.98056
1.34230
1.32977
1.31329
1.28758
1.24055
1.19774
1.12273
1.05813
1.00105
0.97484
V E/cm3‚mol-1
0.0000
0.2295
0.4310
0.6157
0.7825
0.8608
0.9353
0.9712
0.9924
1.0000
0.000
0.104
0.000
0.135
0.000
0.162
0.000
0.000
0.210
0.260
0.230
0.174
0.131
0.067
0.028
0.008
0.000
0.000
0.232
0.302
0.287
0.237
0.188
0.105
0.049
0.013
0.000
0.000
0.252
0.343
0.346
0.300
0.243
0.142
0.069
0.018
0.000
0.000
0.276
0.385
0.404
0.363
0.299
0.179
0.088
0.024
0.000
0.187
0.062
0.120
0.170
0.216
-0.040
-0.123
-0.142
-0.124
-0.075
-0.018
0.000
0.037
0.106
0.170
-0.037
-0.061
-0.065
-0.043
-0.010
0.000
0.039
0.108
0.008
0.071
-0.016
-0.016
-0.003
0.000
0.027
0.007
0.002
0.000
η/mPa‚s
0.0000
0.2286
0.4324
0.6159
0.7830
0.9352
96
47
28
18
10
5
61
32
20
13
7
41
23
15
10
6
29
17
12
8
5
3
21
16
13
11
13
10
7
10
8
9
7
5
4
2
7
6
5
3
2
6
4
3
4
4
3
3
∆η/mPa‚s
0.0000
0.2286
0.4324
0.6159
0.7830
0.9352
0
-27
-27
-20
-12
-3
0
0
-9
-9
-7
-4
0
0
-6
-5
-4
-2
0
0
-4
-3
-2
-1
1
0
-2
-1
-1
0
0
-1
-1
0
0
-1
0
-15
-15
-11
-7
1
1
1
1
-1
1
1
less accurate (i.e., greater than ( 2 % uncertainty) for viscosities
below 5 mPa‚s. Of course, the high water content samples have
low viscosity, especially at high temperatures. Taking both these
factors into consideration, we estimate the uncertainty in the
measurements to be approximately ( 2 % for high viscosity
samples but as much as ( 2 mPa‚s for medium viscosity
samples (between 50 and 100 mPa‚s) and ( 1 mPa‚s for low
viscosity samples (below 50 mPa‚s), which represents a
considerably large relative error for the samples at high
temperature and with high water content.
are consistent with the pure IL used by Yang et al. containing
significant amounts of water, thus altering the true composition
of the mixtures. We do not have any explanation for the small
but significant discrepancy between our pure [emim][EtSO4]
data and that reported by Krummen et al.27 and Jacquemin et
al.28 except possible different nonvolatile impurities in their
samples or our samples from Solvent Innovation. We did repeat
our measurements with a sample synthesized in our laboratory,
with similar water content, and found values slightly higher (by
approximately 0.0014 g‚cm-3) than those presented here for the
Solvent Innovation sample, yet still a bit lower than the
Krummen et al.27 and Jacquemin et al.28 data. We found no
previous data as a function of temperature for [emim][OTf] or
[emim][TFA] or their mixtures with water. No comparison is
attempted with a few literature room temperature density values
since the water contents were not reported.
Results and Discussion
Experimental density F and dynamic viscosity η of the water
+ [emim][EtSO4], water + [emim][OTf], and water + [emim]-
[TFA] binary systems over the temperature range T ) (278.15
to 348.15) K are reported in Tables 1 to 3. The densities and
viscosities are highest for the pure ILs, with decreasing values
with increasing water content. As expected, both density and
viscosity decrease with increasing temperature. The density of
pure water agrees with literature data within experimental
error.26 Comparison with the viscosity of pure water was not
possible because the rheometer is not accurate at low viscosity.
Viscosity values at several temperatures were reported in the
literature for pure [emim][EtSO4] and [emim][OTf] by Jac-
quemin et al.28 and Seddon et al.,30 respectively, although using
ILs with not exactly the same water content. Their values were
found to be in reasonable agreement (within 6 % for the worst
case throughout the entire temperature range studied here) with
the values for the corresponding pure ILs in this paper.
Compared to literature, the density values for pure [emim]-
[EtSO4] are somewhat lower (by about 0.0045 g‚cm-3) than
those reported by Krummen et al.27 and Jacquemin et al.28 and
notably higher than those given by Yang et al.,19,29 who do not
report the water content of the IL. In the latter, measurements
of the density of mixtures water + [emim][EtSO4] are also
carried out, which are again found to be systematically lower
than the ones reported in the present work. These observations
Effect of Temperature on Density. Figure 1 shows the
experimental densities for the three binary systems studied, as
a function of temperature, for different compositions. As it can
be observed, the density decreases as both temperature and water
composition in the systems increase.
To the eye, the change in density with temperature for the
pure ILs in Figure 1 looks linear. This linear behavior has

2148 Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006
Table 3. Experimental Density G, Excess Molar Volume V E, Dynamic Viscosity η, and Viscosity Deviation ∆η for the Binary System Water (1)
+ [emim][TFA] (2)
T/K
x1
278.15
288.15
298.15
308.15
318.15
328.15
338.15
348.15
F/g‚cm-3
1.28285
1.27592
1.26729
1.25377
1.22366
1.19257
1.1304
0.0000
0.2050
0.3983
0.5808
0.7573
0.8420
0.9256
0.9667
0.9912
1.0000
1.30689
1.29984
1.29127
1.278
1.24838
1.21687
1.15147
1.08688
1.02668
0.99999
1.29881
1.29182
1.28325
1.26993
1.24021
1.20886
1.14467
1.08251
1.02493
0.99913
1.2908
1.27495
1.26804
1.25934
1.24567
1.2153
1.18427
1.12293
1.06608
1.01411
0.99023
1.26712
1.26021
1.25141
1.23755
1.20684
1.17586
1.1152
1.05966
1.00907
0.98571
1.25935
1.25242
1.24349
1.22941
1.19832
1.16733
1.10728
1.05283
1.00287
0.98056
1.25163
1.24466
1.23559
1.22123
1.18998
1.15869
1.09905
1.04556
0.99723
0.97484
1.28385
1.27526
1.26185
1.23195
1.20076
1.13765
1.07755
1.02215
0.99707
1.07206
1.0185
0.99406
V E/cm3‚mol-1
0.000
0.0000
0.2050
0.3983
0.5808
0.7573
0.8420
0.9256
0.9667
0.9912
1.0000
0.000
-0.112
-0.370
-0.651
-0.764
-0.698
-0.471
-0.236
-0.050
0.000
0.000
-0.095
-0.331
-0.588
-0.675
-0.604
-0.392
-0.194
-0.042
0.000
0.000
-0.080
-0.296
-0.532
-0.595
-0.523
-0.327
-0.160
-0.035
0.000
0.000
-0.055
-0.237
-0.435
-0.457
-0.382
-0.219
-0.104
-0.024
0.000
0.000
-0.045
-0.209
-0.388
-0.392
-0.319
-0.171
-0.079
-0.019
0.000
0.000
-0.034
-0.181
-0.343
-0.328
-0.257
-0.127
-0.056
-0.002
0.000
0.000
-0.024
-0.154
-0.297
-0.278
-0.199
-0.083
-0.032
-0.007
0.000
-0.066
-0.265
-0.482
-0.524
-0.449
-0.270
-0.130
-0.030
0.000
η/mPa‚s
0.0000
0.2042
0.3952
0.5808
0.7565
0.9256
78
56
40
25
16
6
48
35
26
17
11
5
32
24
18
12
8
23
17
14
9
6
3
17
13
10
8
5
3
13
10
8
6
5
10
8
7
6
4
9
7
6
5
4
2
4
3
2
∆η/mPa‚s
0.0000
0.2042
0.3952
0.5808
0.7565
0.9256
0
-7
-8
-8
-4
-1
0
-3
-4
-4
-2
0
0
-2
-2
-2
-1
1
0
-1
-1
-1
0
0
0
0
0
1
1
0
0
0
1
1
1
0
0
1
1
1
1
0
0
1
1
1
1
1
Table 4. Volume Expansivity r of Pure ILs for a 99 % Confidence
in Table S1 in the Supporting Information, where the magnitude
of k2 is an indication of the degree of curvature of density as a
function of temperature for each given composition. From the
standard deviation values given in Table S1, it is clear that the
quality of the correlation decreases with increasing water
concentration. However, even for pure water, the accuracy of
the fit is acceptable for our purposes. Nevertheless, it must be
noticed that the empirical correlation presented is not intended
to provide an accurate description of the density of pure water
itself, a task that is out of the scope of this work. The polynomial
correlations are plotted with solid lines in Figure 1, along with
the experimental data.
Interval
R/10-4 K-1
T/K
[emim][EtSO4]
[emim][OTf]
[emim][TFA]
278.15
288.15
298.15
308.15
318.15
328.15
338.15
348.15
5.51 ( 0.03
5.50 ( 0.03
5.49 ( 0.03
5.48 ( 0.03
5.47 ( 0.03
5.46 ( 0.03
5.45 ( 0.03
5.43 ( 0.03
6.087 ( 0.004
6.082 ( 0.005
6.076 ( 0.005
6.071 ( 0.005
6.065 ( 0.005
6.059 ( 0.005
6.052 ( 0.005
6.045 ( 0.006
6.201 ( 0.002
6.193 ( 0.002
6.185 ( 0.002
6.177 ( 0.003
6.168 ( 0.003
6.159 ( 0.003
6.149 ( 0.003
6.139 ( 0.003
already been reported by several authors for pure ILs.20,28 For
the ILs in this work, the observation of such behavior will be
discussed in more detail below in the section on volume
expansivity of pure ILs.
Volume ExpansiWity of the Pure ILs. For a pure fluid, the
volume expansivity R, also known as the coefficient of thermal
expansion, is a measure of how the volume changes with
temperature. R is defined as31
By increasing the composition of water in the systems, the
temperature dependence becomes distinctly nonlinear, especially
at high water content. This is not surprising since the density
of pure water is well-known to show nonlinear behavior with
temperature. A second-order polynomial was found to satisfac-
torily correlate, from an empirical perspective, the change of
density with temperature throughout the whole range of
composition:
1 ∂V
V ∂T p
1 ∂F
F ∂T p
R )
) -
(2)
(
)
(
)
where subscript p indicates constant pressure. It has already been
shown in Figure 1 that the densities of pure ILs appear to
decrease linearly (R 2 > 0.9999, standard deviation σ < 0.00017)
with temperature (i.e., (∂F/∂T)p is a constant). From elemental
calculus, it turns out that the right-hand term in eq 2 can be
expressed as -(∂ ln F/∂T)p. Not surprisingly, a plot of ln F as
a function of temperature can also be fit by a straight line, with
improved correlation coefficients (R 2 > 0.99999) and standard
deviations (σ < 0.00005). According to this second perspective,
F ) k2T 2 + k1T + k0
(1)
where T is for the absolute temperature and k2, k1, and k0 refer
to the fit coefficients. Least-square fits generated the values listed

Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006 2149
Figure 1. Density F for the water (1) + IL (2) binary systems as a function
of temperature at different approximated mass fractions of water: b, 0 %;
O, 2 %; 1, 5 %; 3, 10 %; 9, 20 %; 0, 30 %; [, 50 %; ], 70 %; 2, 90
%; 4, 100 %. Polynomial correlations are plotted as solid lines. IL: (a)
[emim][EtSO4]; (b) [emim][OTf]; (c) [emim][TFA].
Figure 2. Dynamic viscosity η for the water (1) + IL (2) binary systems
as a function of temperature at different approximated mass fractions of
water: b, 0 %; O, 2 %; 1, 5 %; 3, 10 %; 9, 20 %; 0, 50 %. The
correlations performed with the VFT equation are plotted as solid lines.
IL: (a) [emim][EtSO4]; (b) [emim][OTf]; (c) [emim][TFA].
(∂ln F/∂T)p is a constant. This seems to lead to conflicting results.
If one considers a plot of ln F versus T to be linear, then R is
a constant. If one considers a plot of F versus T to be linear,
then the magnitude of R increases with increasing temperature
because one must multiply the slope by the inverse of the
density. In the literature, both alternatives can be found for the
calculation of volume expansivities for ILs.20,28,29,32
A more careful examination reveals that the densities of the
pure ILs measured here actually do not increase linearly with
temperature. An analysis of the residuals indicates subtle but
clear curvature for all three ILs. A second-order polynomial,
of the type already used to fit the densities of the IL + water
mixtures, provides a much better fit. A statistical analysis by
means of the F-test showed that the addition of the quadratic
term provided a statistically significant improvement in the fit
at a significance level of 0.05 for the water + [emim][EtSO4]
system and at significance levels as low as 0.001 for the other
two binary systems. The fit parameters are shown in Table S1
in the Supporting Information. The partial derivative in the right-
hand side of eq 2 can be evaluated from the polynomial fits,
and the corresponding volume expansivity values can be
calculated. The results are summarized in Table 4.

2150 Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006
Figure 3. Ideal glass transition temperature T0 for the water (1) + IL (2)
systems, as obtained from the VFT equation correlations. IL: (a) [emim]-
[EtSO4]; (b) [emim][OTf]; (c) [emim][TFA].
Figure 4. Excess molar volume V E for the water (1) + IL (2) systems at
different temperatures: b, 278.15 K; O, 288.15 K; 1, 298.15 K; 3, 308.15
K; 9, 318.15 K; 0, 328.15 K; [, 338.15 K; ], 348.15 K. The solid lines
represent the corresponding correlations by the Redlich-Kister equation.
IL: (a) [emim][EtSO4]; (b) [emim][OTf]; (c) [emim][TFA].
The volume expansivities calculated from the polynomial fits
and eq 2 are almost constant, just decreasing slightly with
increasing temperature for all three ILs. Actually, the constant
values obtained by assuming that a plot of ln F versus
temperature is linear provide good estimates: 5.48 × 10-4 K-1
for [emim][EtSO4], 6.07 × 10-4 K-1 for [emim][OTf], and 6.17
× 10-4 K-1 for [emim][TFA]. Nonetheless, the values shown
in Table 4, which do decrease with increasing temperature, are
the best analysis of the experimental data. These comparisons
emphasize that it is important to carefully analyze the experi-
mental data in order to avoid corruption of the thermodynamic
concepts or misrepresentation of the experimental data. In
addition, if analyzed with one of the linear approximations, one
should not make major distinctions in whether R is increasing,
constant, or decreasing with temperature for these systems since
this is highly dependent on how the data are analyzed.
In general, values in Table 4 are within the common range
reported for ILs in the literature (roughly around 5 × 10-4 to 8
× 10-4 K-1), being somewhat lower than typical expansivities
for conventional molecular solvents.32,33 A specific comparison
of the results is possible for [emim][EtSO4] as three other
research groups have studied the density variation of this pure
IL with temperature. Our values for [emim][EtSO4] are certainly
close to the constant of 5.37 × 10-4 K-1 reported by Yang et

Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006 2151
Figure 6. Density F for the water (1) + IL (2) systems, as a function of
mole fraction of water, at different temperatures: b, 278.15 K; O, 288.15
K; 1, 298.15 K; 3, 308.15 K; 9, 318.15 K; 0, 328.15 K; 2, 338.15 K; 4,
348.15 K. The solid lines correspond to the fits derived from the Redlich-
Kister correlation of the excess molar volume V E. IL: (a) [emim][EtSO4];
(b) [emim][OTf]; (c) [emim][TFA].
Figure 5. Viscosity deviation ∆η for the water (1) + IL (2) systems at
different temperatures: b, 278.15 K; O, 288.15 K; 1, 298.15 K; 3, 308.15
K; 9, 318.15 K; 0, 328.15 K; 2, 338.15 K; 4, 348.15 K. The solid lines
represent the corresponding correlations by the Redlich-Kister equation.
IL: (a) [emim][EtSO4]; (b) [emim][OTf]; (c) [emim][TFA].
al.29 However, they are significantly higher than those reported
by Jacquemin et al.28: 4.82 × 10-4 K-1 at 293.15 K and 4.94
× 10-4 K-1 at 343.15 K. These authors report increasing volume
expansivity with increasing temperature, the opposite of the
trend we report, but as explained above, this is simply because
they fit F versus T with a straight line. As mentioned above,
our data show a clear convex curvature, which is well fit by a
second-order polynomial. A similar curvature is present in the
density data reported by Yang et al. and Krummen et al.27 for
[emim][EtSO4]. By fitting Krummen’s raw data to a second-
order polynomial and calculating the volume expansivity
according to eq 2, one obtains values of (5.62 to 5.47) × 10-4
K-1 at temperatures from (293.15 to 353.15) K, in good
agreement with our values.
Interestingly, the detailed analysis of the data performed here
indicates that the three ILs studied in this work actually show
decreasing volume expansivity with increasing temperature. This
is a very unusual behavior for fluids in general. Of course, it is
contrary to the results reported for other ILs,33,34 but this is not
surprising since assumptions of linearity in fitting the data can
affect the results, as explained above. Rebelo et al.17 speculated

2152 Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006
three ILs investigated here, which contain the rather different
[EtSO4]-, [OTf]-, and [TFA]- anions, exhibit this behavior.
Effect of Temperature on Viscosity. The values of viscosity
as a function of temperature for binary mixtures of different
compositions for the three water + IL systems are depicted in
Figure 2. The well-known Arrhenius-type equation has been
widely used to correlate variations of viscosity with temperature
in liquids:
Ea
η ) η exp -
(3)
(
)
∞
RT
where R is the universal gas constant and the characteristic
parameters are η∞ (viscosity at infinite temperature) and Ea
(“activation” energy). However, for ILs it has been reported35
that a better description is provided by the modified version36
of the Vogel-Fulcher-Tammann (VFT) equation,37-39 which
characterizes glass-forming liquids:
k
η ) AT0.5 exp
(4)
(
)
T - T0
The fit parameters are A, k, and T0. It was originally proposed
as an entirely empirical equation, but the development of the
free volume theory of Cohen and Turnbull36 and the configu-
rational entropy approach of Adam and Gibbs40 convey some
theoretical significance to the parameter T0. It is the “ideal glass
transition temperature”, a temperature below which the fluid
exists as an equilibrium glass where the mass-transporting
motions are frozen out.41 For kinetic reasons, T0 cannot be
reached in a finite time scale experiment; however, it should
be slightly below the experimental glass transition temperature
Tg.42
The VFT equation suitably correlates, as a function of the
temperature, not only the viscosities of the pure ILs but also
the viscosities of the mixtures for the binary systems throughout
the composition range. The significant improvement in the data
fit by the VFT equation over the Arrhenius equation can be
noticed by comparison of the relative standard deviations listed
in Table S2 in the Supporting Information, along with the
adjusted parameters for the correlations with each equation. This,
of course, is not unexpected since the VFT contains three
parameters compared to two for the Arrhenius equation. The
VFT fits are plotted with the experimental data in Figure 2.
Figure 3 shows T0 as a function of composition for the three
water + IL systems. Since the adjustable parameters are very
sensitive to small changes in the viscosity data, large variations
in the fit values of T0 occur simply as a result of the experimental
error. Therefore, error bars were estimated and are included in
the plot. For pure water, the value of T0 was obtained from
viscosity data taken from literature,26 correlated by means of
the VFT equation. T0 increases with increasing water concentra-
tion for all three water + IL systems, reaching a maximum at
about 0.8 mole fraction of water. While the large asymmetry
in molecular size and the complexity of the IL may complicate
the analysis, for systems of small molecules and ionic species
the most important factor in setting the magnitude of T0 (or Tg)
is the overall cohesive energy of the liquid mixture.41 Thus,
the plots in Figure 3 imply that the interactions between water
and the ILs are maximized in mixtures with an IL mole fraction
of approximately 0.20. While this may be inferring too much
meaning from a largely empirical equation, the similarity in the
behavior of T0 for the three systems is noteworthy.
Figure 7. Viscosity η for the water (1) + IL (2) systems, as a function of
mole fraction of water, at different temperatures: b, 278.15 K; O, 288.15
K; 1, 298.15 K; 3, 308.15 K; 9, 318.15 K; 0, 328.15 K; 2, 338.15 K; 4,
348.15 K. The solid lines correspond to the fits derived from the Redlich-
Kister correlation of the viscosity deviation ∆η. IL: (a) [emim][EtSO4];
(b) [emim][OTf]; (c) [emim][TFA].
about the possibility of finding ILs following such behavior,
which, in accord with thermodynamic relationships, implies the
unusual situation that their molar heat capacities will increase
with increasing pressure. Since volume expansivity must rise
to +∞ as the critical point of a substance is approached, a
minimum in R must appear at higher temperatures than those
investigated here, unless obscured by decomposition of the ILs.
While decreasing volume expansivity with increasing temper-
ature may not be general for all ILs, it is interesting that all
The viscosity data for all three pure ILs resulted in similar
values of T0, in the range of (150 to 170) K. Comparison with

Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006 2153
Figure 8. Pseudo-quantitative change of (a) density F and (b) viscosity η with temperature and composition for the water + IL binary systems. The left
plots correspond to the IL [emim][EtSO4]; the central plots correspond to [emim][OTf]; and the right plots correspond to [emim][TFA].
experimental Tg values is possible for [emim][EtSO4], with
literature Tg data ranging from (181 to 208) K.43,44 Experimental
discrepancies are no doubt due to varying concentrations of
impurities, including water. Nonetheless, the experimental Tg
values are somewhat greater than T0 calculated from the
viscosity data, consistent with the theoretical rationale for the
VFT equation.
The parameters A and k in eq 4 change smoothly with
composition for all three water + IL systems, as shown in Table
S2 in the Supporting Information. However, their analysis will
not be detailed here, as the physical meaning of these parameters
is not so clear and their values are strongly sensitive to the
choice of T0.
degree of the polynomial expansion. A third-order polynomial
was found to be the optimum for both properties. The fit
parameters are summarized in Tables S3 and S4 in the
Supporting Information along with the corresponding standard
deviations for the correlations. The values of V E and ∆η as
well as the Redlich-Kister fits are plotted in Figures 4 and 5.
The graphs of V E shown in Figure 4 indicate that mixtures
of water with [emim][EtSO4] and [emim][TFA] exhibit mostly
negative deviations from ideality. The magnitude of the excess
volume decreases with increasing temperature. Notable are the
small positive deviations from ideality for dilute aqueous
solutions of [emim][EtSO4], which are not described very well
by the Redlich-Kister correlation. In stark contrast is the
behavior of the system water + [emim][OTf]. It exhibits mostly
positive values of the excess volume, except at the lowest
temperatures, where V E is slightly positive for IL-rich composi-
tions and slightly negative for water-rich compositions. The
[OTf]- anion is known to be less hydrophilic than [EtSO4]- or
[TFA]-.8 Therefore, the probable explanation is that [emim]-
[EtSO4] and [emim][TFA] have stronger interactions with water,
resulting in negative excess volumes, while the weaker interac-
tions with [emim][OTf] are insufficient to cause volume
contraction. The decrease in the magnitude of the negative V E
values and the increase in the magnitude of the positive V E
values with increasing temperature can be attributed to the
decreasing importance of hydrogen bonding with increasing
temperature. Note that the slight positive values in the modeling
for water + [emim][TFA] are not corroborated by any experi-
mental data and should be taken lightly.
Effect of Composition on Density and Viscosity. The excess
molar volume, V E, and viscosity deviation, ∆η, were calculated
from the experimental data for all the water + IL mixtures
according to the following equations:
1
F
1
Fi
V E
)
xiMi
-
(5)
(6)
∑
(
)
∆η ) η - xiηi
∑
where F and η are the density and viscosity of the solution and
Fi and ηi refer to the density and viscosity of the pure component
i. For the application of eq 6, water viscosity data were taken
from reference literature26 because our viscometer has large
uncertainties for low viscosity samples, as explained in the
Experimental Section. The results are listed in Tables 1 to 3.
The excess volumes and viscosity deviations were fit for each
temperature with a Redlich-Kister polynomial:45
All of the excess molar volumes reported here are relatively
small in comparison to the molar volume. As a consequence, it
may be possible to correlate the molar volumes of water + IL
systems with a straight line.17 Although this may not be
satisfying conceptually, it may be a valuable tool in practice
for the quick calculation of densities of binary water + IL
systems.
m
Q ) xwxIL
Ai(xw - xIL)i
(7)
∑
i)0
where Q represents either the excess molar volume V E or the
viscosity deviation ∆η, Ai is the fit parameters, and m is the

2154 Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006
The water + [emim][EtSO4] and water + [emim][OTf]
systems show strong negative viscosity deviations at low
temperatures, as shown in Figure 5. Although they show similar
trends, the viscosity deviations for the water + [emim][TFA]
system are not significantly greater than the uncertainty in the
measurements. As the temperature increases, the deviations are
reduced and even reach slightly positive values. The largest
magnitude deviations occur for the system with the highest
viscositys[emim][EtSO4]. In general, the Redlich-Kister fit
provides a good description of the change in ∆η with composi-
tion for all three systems.
Using the correlation of the excess properties by the Redlich-
Kister polynomials, the absolute values of the densities and
viscosities of the binary system can be expressed as simple
functions of composition. By combining eq 7 with eqs 5 and 6,
analytical expressions are obtained for F and η. Both the
experimental data and the correlations are shown in Figures 6
and 7. Note how a few moles of IL added to pure water
dramatically increases the density of the mixtures, while the
most pronounced effect on viscosity occurs when adding water
to pure IL, especially at low temperatures.
is especially critical at low temperatures or low water concentra-
tions, where dramatic variations of viscosity occur.
The ILs incorporating the fluorinated [OTf]- and [TFA]-
anion lead to higher densities and lower viscosities than the
one with the non-fluorinated [EtSO4]- anion. Nevertheless,
economical and even environmental considerations should also
be taken into account before establishing the best IL for a given
application involving these kinds of binary mixtures.
Supporting Information Available:
Fit parameters and standard deviation for the empirical correlation
of density in the systems water + IL as a function of temperature
(Table S1); fit parameters of the Arrhenius equation and the VFT
equation, as well as their relative standard deviations, for the
correlation of viscosity as a function of temperature in the systems
water + IL (Table S2); coefficients of the Redlich-Kister equation
for the correlation of the excess molar volume of the systems water
+ IL, along with the standard deviations (Table S3); and coefficients
of the Redlich-Kister equation for the correlation of the viscosity
deviation of the systems water + IL, along with the standard
deviations (Table S4). This material is available free of charge via
the Internet at http://pubs.acs.org.
Combined Effect of Temperature and Composition. Above
we have presented experimental data and correlations of the
density and viscosity of the three water + IL systems as a
function of temperature and as a function of composition. If,
for instance, one analyzes the parameters in the temperature-
dependent fit as a function of composition, then it is possible
to combine the correlations and build a three-dimensional plot
of the physical property as a function of both temperature and
composition. Such plots are shown in Figure 8. They provide a
quick visualization of the combined influence of temperature
and composition on the density and viscosity for the three water
+ IL binary systems studied in this work. The scales are the
same for all three systems in order to allow an easy comparison
between the ILs. Although a mole basis is usually preferred for
thermodynamic analysis of mixtures, the significant difference
between the molecular weight of water and the formula weights
of the ILs makes presentation on a mass fraction basis of interest,
and this is what is shown in Figure 8. Note that the viscosity
plots only extend to a water mass fraction of 50 %, which is
the highest water composition investigated.
It is clear that composition has a greater influence than
temperature on density. For viscosity, dramatic decreases are
observed with the addition of small masses of water, especially
at low temperature. Hence, these graphs emphasize the extreme
sensitivity of pure IL viscosity to water content.8 From a
practical standpoint, this means that highly viscous solutions
of IL and water can be circumvented simply by avoiding
temperatures close to the freezing point of water or streams with
very low water content.
Literature Cited
(1) Rogers, R. D.; Seddon, K. R. Ionic liquidsssolvents of the future?
Science 2003, 302, 792-793.
(2) Seddon, K. R. Ionic liquids: a taste of the future. Nat. Mater. 2003,
2, 363-365.
(3) Brennecke, J. F.; Maginn, E. J. Ionic liquids: innovative fluids for
chemical processing. AIChE J. 2001, 47, 2384-2389.
(4) Blanchard, L. A.; Hancu, D.; Beckman, E. J.; Brennecke, J. F. Green
processing using ionic liquids and CO2. Nature 1999, 399, 28-29.
(5) Blanchard, L. A.; Brennecke, J. F. Recovery of organic products from
ionic liquids using supercritical carbon dioxide. Ind. Eng. Chem. Res.
2001, 40, 287-292.
(6) Blanchard, L. A.; Gu, Z.; Brennecke, J. F. High-pressure phase
behavior of ionic liquid/CO2 systems. J. Phys. Chem. B 2001, 105,
2437-2444.
(7) Scurto, A. M.; Aki, S. N. V. K.; Brennecke, J. F. CO2 as a separation
switch for ionic liquid/organic mixtures. J. Am. Chem. Soc. 2002, 124,
10276-10277.
(8) Seddon, K. R.; Stark, A.; Torres, M. J. Influence of chloride, water,
and organic solvents on the physical properties of ionic liquids. Pure
Appl. Chem. 2000, 72, 2275-2287.
(9) Gaillon, L.; Sirieix-Plenet, J.; Letellier, P. Volumetric study of binary
solvent mixtures constituted by amphiphilic ionic liquids at room
temperature (1-alkyl-3-methylimidazolium bromide) and water. J.
Solution Chem. 2004, 33, 1333-1347.
(10) Lu, X.-M.; Xu, W.-G.; Gui, J.-S.; Li, H.-W.; Yang, J. -Z. Volumetric
properties of room temperature ionic liquid. 1. The system {1-methyl-
3-ethylimidazolium ethyl sulfate + water} at temperature in the range
(278.15 to 333.15) K. J. Chem. Thermodyn. 2005, 37, 13-19.
(11) Zafarani-Moattar, M. T.; Shekaari, H. Apparent molar volume and
isentropic compressibility of ionic liquid 1-butyl-3-methylimidazolium
bromide in water, methanol, and ethanol at T ) (298.15 to 318.15)
K. J. Chem. Thermodyn. 2005, 37, 1029-1035.
(12) Fitchett, B. D.; Knepp, T. N.; Conboy, J. C. 1-Alkyl-3-methylimida-
zolium bis(perfluoroalkylsulfonyl)imide water-immiscible ionic liquids.
The effect of water on electrochemical and physical properties. J.
Electrochem. Soc. 2004, 151, E219-E225.
(13) Pandey, S.; Fletcher, K. A.; Baker, S. N.; Baker, G. A. Correlation
between the fluorescent response of microfluidity probes and the water
content and viscosity of ionic liquid and water mixtures. Analyst 2004,
129, 569-573.
(14) Widegren, J. A.; Laesecke, A.; Magee, J. W. The effect of dissolved
water on the viscosities of hydrophobic room-temperature ionic liquids.
Chem. Commun. 2005, 1610-1612.
Conclusions
The density and viscosity of binary mixtures of water +
[emim][EtSO4], water + [emim][OTf], and water + [emim]-
[TFA] were investigated at atmospheric pressure from (278.15
to 348.15) K. Both properties were found to decrease with an
increase in either temperature or in mole fraction of water. An
empirical second-order polynomial and the VFT equation were
used to correlate, respectively, density and viscosity as a function
of the absolute temperature. Reasonable correlations of the
composition dependence were achieved by fitting the excess
molar volumes and the viscosity deviations to third-order
Redlich-Kister polynomials. Viscosity is more sensitive than
density to changes in temperature or composition; this behavior
(15) Chakrabarty, D.; Chakraborty, A.; Seth, D.; Sarkar, N. Effect of water,
methanol, and acetonitrile on solvent relaxation and rotational
relaxation of coumarin 153 in neat 1-hexyl-3-methylimidazolium
hexafluorophosphate. J. Phys. Chem. A 2005, 109, 1764-1769.
(16) Zhang, S.; Li, X.; Chen, H.; Wang, J.; Zhang, J.; Zhang, M.
Determination of physical properties for the binary system of 1-ethyl-
3-methylimidazolium tetrafluoroborate + H2O. J. Chem. Eng. Data
2004, 49, 760-764.
(17) Rebelo, L. P. N.; Najdanovic-Visak, V.; Visak, Z. P.; Nunes. da Ponte,
M.; Szydlowski, J.; Cerdeirin˜a, C. A.; Troncoso, J.; Roman´ı, L.;

Journal of Chemical and Engineering Data, Vol. 51, No. 6, 2006 2155
Esperanc¸a, J. M. S. S.; Guedes, H. J. R.; de Sousa, H. C. A detailed
thermodynamic analysis of [C4mim][BF4] + water as a case study to
model ionic liquid aqueous solutions. Green Chem. 2004, 6, 369-
381.
(32) Gu, Z.; Brennecke, J. F. Volume expansivities and isothermal
compressibilities of imidazolium- and pyridinium-based ionic liquids.
J. Chem. Eng. Data 2002, 47, 339-345.
(33) Rebelo, L. P. N.; Najdanovic-Visak, V.; Gomes de Azevedo, R.;
Esperanc¸a, J. M. S. S.; Nunes da Ponte, M.; Guedes, H. J. R.; Visak,
Z. P.; de Sousa, H. C.; Szydlowski, J.; Canongia Lopes, J. N.; Cordeiro,
T. C. Phase behavior and thermodynamic properties of ionic liquids,
ionic liquid mixtures, and ionic liquid solutions. ACS Symp. Ser. 2005,
No. 901 (Ionic Liquids IIIA: Fundamentals, Progress, Challenges, and
Opportunities), 270-291.
(18) Xu, H.; Zhao, D.; Xu, P.; Liu, F.; Gao, G. Conductivity and viscosity
of 1-allyl-3-methyl-imidazolium chloride + water and + ethanol from
293.15 K to 333.15 K. J. Chem. Eng. Data 2005, 50, 133-135.
(19) Yang, J.-Z.; Lu, X.-M.; Gui, J.-S.; Xu, W.-G.; Li, H. -W. Volumetric
properties of room temperature ionic liquid. 2. The concentrated
aqueous solutions of {1-methyl-3-ethylimidazolium ethyl sulfate +
water} in a temperature range of 278.2 to 333.2 K. J. Chem.
Thermodyn. 2005, 37, 1250-1255.
(20) Go´mez, E.; Gonza´lez, B.; Dom´ınguez, AÄ .; Tojo, E.; Tojo, J. Dynamic
viscosities of a series of 1-alkyl-3-methylimidazolium chloride ionic
liquids and their binary mixtures with water at several temperatures.
J. Chem. Eng. Data 2006, 51, 696-701.
(21) Jarosik, A.; Krajewski, S. R.; Lewandowski, A.; Radzimski, P.
Conductivity of ionic liquids in mixtures. J. Mol. Liq. 2006, 123, 43-
50.
(22) Zhou, Q.; Wang, L.-S.; Chen, H.-P. Densities and viscosities of 1-butyl-
3-methylimidazolium tetrafluoroborate + H2O binary mixtures from
(303.15 to 353.15) K. J. Chem. Eng. Data 2006, 51, 905-908.
(23) Archer, D. G.; Widegren, J. A.; Kirklin, D. R.; Magee, J. W. Enthalpy
of solution of 1-octyl-3-methylimidazolium tetrafluoroborate in water
and in aqueous sodium fluoride. J. Chem. Eng. Data 2005, 50, 1484-
1491.
(24) Holbrey, J. D.; Reichert, W. M.; Swatloski, R. P.; Broker, G. A.; Pitner,
W. R.; Seddon, K. R.; Rogers, R. D. Efficient, halide free synthesis
of new, low cost ionic liquids: 1,3-dialkylimidazolium salts containing
methyl- and ethyl-sulfate anions. Green Chem. 2002, 4, 407-413.
(25) Bonhoˆte, P.; Dias, A.-P.; Armand, M.; Papageorgiou, N.; Kalyana-
sundaram, K.; Gra¨tzel, M. Hydrophobic, highly conductive ambient-
temperature molten salts. Inorg. Chem. 1996, 35, 1168-1178.
(Correction: Inorg. Chem. 1998, 37, 166).
(34) Marsh, K. N.; Boxall, J. A.; Lichtenthaler, R. Room temperature ionic
liquids and their mixturessa review. Fluid Phase Equilib. 2004, 219,
93-98.
(35) Wilkes, J. S. Properties of ionic liquid solvents for catalysis. J. Mol.
Catal. A 2004, 214, 11-17.
(36) Cohen, M. H.; Turnbull, D. Molecular transport in liquids and glasses.
J. Chem. Phys. 1959, 31, 1164-1169.
(37) Vogel, H. Das Temperatur-abha¨ngigkeitsgesetz der Viskosita¨t von
Flu¨ssigkeiten. Phys. Z. 1921, 22, 645-646.
(38) Fulcher, G. S. Analysis of recent measurements of the viscosity of
glasses. Am. Ceram. Soc. J. 1925, 8, 339-355.
(39) Tammann, G.; Hesse, W. Die Abha¨ngigkeit der Viscosita¨t von der
Temperatur bei unterku¨hlten Flu¨ssigkeiten. Z. Anorg. Allg. Chem. 1926,
156, 245-257.
(40) Adam, G.; Gibbs, J. H. On the temperature dependence of cooperative
relaxation properties in glass-forming liquids. J. Chem. Phys. 1965,
43, 139-146.
(41) Angell, C. A.; Moynihan, C. T. Transport processes in low-melting
molten salt systems. In Molten Salts: Characterization and Analysis;
Mamantov, G., Ed.; Marcel Dekker: New York, 1969.
(42) Gibbs, J. H.; DiMarzio, E. A. Nature of the glass transition and the
glassy state. J. Chem. Phys. 1958, 28, 373-383.
(43) Holbrey, J. D.; Reichert, W. M.; Swatloski, R. P.; Broker, G. A.; Pitner,
W. R.; Seddon, K. R.; Rogers, R. D. Efficient, halide free synthesis
of new, low cost ionic liquids: 1,3-dialkylimidazolium salts containing
methyl- and ethyl-sulfate anions. Green Chem. 2002, 4, 407-413.
(44) Strehmel, V.; Laschewsky, A.; Wetzel, H.; Go¨rnitz, E. Free radical
polymerization of n-butyl methacrylate in ionic liquids. Macromol-
ecules 2006, 39, 923-930.
(45) Redlich, O.; Kister, A. T. Thermodynamics of nonelectrolyte solutions.
Algebraic representation of thermodynamic properties and the clas-
sification of solutions. Ind. Eng. Chem. 1948, 40, 345-348.
(26) Riddick, J. A.; Bunger, W. B.; Sakano, T. K. Organic SolVents:
Physical Properties and Methods of Purification, 4th ed.; John Wiley
& Sons: New York, 1986.
(27) Krummen, M.; Wasserscheid, P.; Gmehling, J. Measurement of activity
coefficients at infinite dilution in ionic liquids using the dilutor
technique. J. Chem. Eng. Data 2002, 47, 1411-1417.
(28) Jacquemin, J.; Husson, P.; Padua, A. A. H.; Majer, V. Density and
viscosity of several pure and water-saturated ionic liquids. Green
Chem. 2006, 8, 172-180.
(29) Yang, J.-Z.; Lu, X.-M.; Gui, J.-S.; Xu, W. -G. A new theory for ionic
liquidssthe interstice model. Part 1. The density and surface tension
of ionic liquid EMISE. Green Chem. 2004, 6, 541-543.
(30) Seddon, K. R.; Stark, A.; Torres, M.-J. Viscosity and density of 1-alkyl-
3-methylimidazolium ionic liquids. ACS Symp. Ser. 2002, No. 819
(Clean Solvents: Alternative Media for Chemical Reactions and
Processing), 34-49.
(31) Smith, J. M.; Van Ness, H. C.; Abbott, M. M. Introduction to Chemical
Engineering Thermodynamics, 6th ed.; McGraw-Hill: New York,
2001.
Received for review June 20, 2006. Accepted August 25, 2006. We
acknowledge funding for this project from the U.S. Department of
Energy, under Grant FG02-05CH11294. H.R. is grateful to the
Ministerio de Educacio´n y Ciencia (Spain) for the FPI grant with
reference BES-2004-5311 under project PPQ2003-01326.
JE0602824