Conductivities of binary mixtures of ionic liquids with polar solvents

Article abstract of DOI:10.1021/je800468h

Data for the conductivity, κ, of selected binary mixtures of the ionic liquids [emim][BF₄], [bmim][BF₄], [bmim][PF ₆], [brmm][DCA], and [hmim][BF₄] with polar solvents (water, propylene carbonate, dimethyl-sulfoxide, methanol, dichloromethane) at 25 °C are reported. Additionally, mixture densities, p, were determined to convert κ into molar conductivity, A. The obtained results were fitted by appropriate interpolation formulas. Where possible, data were compared with information from the literature. Electrode polarization and sample purity, including [BF₄^-] hydrolysis, were considered as possible sources of errors in κ. The effect of viscosity on the accuracy of p and thus A was checked.

Full text of DOI:10.1021/je800468h

472
J. Chem. Eng. Data 2009, 54, 472–479
Conductivities of Binary Mixtures of Ionic Liquids with Polar Solvents†
Alexander Stoppa, Johannes Hunger, and Richard Buchner*
Institut fu¨r Physikalische and Theoretische Chemie, Universita¨t Regensburg, D-93040 Regensburg, Germany
Data for the conductivity, κ, of selected binary mixtures of the ionic liquids [emim][BF4], [bmim][BF4],
[bmim][PF6], [bmim][DCA], and [hmim][BF4] with polar solvents (water, propylene carbonate, dimethyl-
sulfoxide, methanol, dichloromethane) at 25 °C are reported. Additionally, mixture densities, F, were
determined to convert κ into molar conductivity, Λ. The obtained results were fitted by appropriate
interpolation formulas. Where possible, data were compared with information from the literature. Electrode
polarization and sample purity, including [BF-4 ] hydrolysis, were considered as possible sources of errors
in κ. The effect of viscosity on the accuracy of F and thus Λ was checked.
latter, measurements of AN mixtures prepared with a com-
mercial batch of [emim][BF4], further on designated as
[emim][BF4]#1, and with material synthesized after a newly
developed procedure, [emim][BF4]#2, were compared. Ad-
ditionally, the impact of [BF-4 ] hydrolysis was studied. The
effect of sample viscosity on the accuracy of F (and thus c and
Λ) determined by vibrating-tube densimetry is considered.
Introduction
Room-temperature ionic liquids (RTILs) are salts with a
melting point around or below ambient temperature. Generally
formed by bulky and asymmetric cations and/or anions, they
exhibit a variety of interesting properties, like negligible vapor
pressure, that can be tuned by appropriate choice of anion and
cation. This possibility to design (at least in principle) task
specific RTILs stimulated the search for applications in chem-
istry and chemical engineering.1-4
Experimental Section
Among the large variety of conceivable or already available
ionic liquids, those based on substituted imidazolium cations
are probably the most intensively studied. This is reflected by
a growing number of reviews dealing with the physicochemical
properties of such RTILs in the pure state and of their mixtures
with cosolvents.5-7 Surprisingly, systematic studies of the
transport properties of binary mixtures of RTILs and polar
solvents, like their conductivity, κ, are scarce. Only for aqueous
mixtures, a fair number of investigations was published
recently.8-12
In this paper, we report precise measurements of the
conductivity, κ, of several imidazolium-based ionic liquids and
their mixtures with polar solvents over the entire miscibility
range at 25 °C. Additionally, mixture densities, F, were
determined to calculate molar concentrations, c, and molar
conductivities, Λ, of the investigated samples.
The following systems were studied: 1-N-ethyl-3-N-meth-
ylimidazolium tetrafluoroborate ([emim][BF4]) + water (W) or
+ acetonitrile (AN); 1-N-butyl-3-N-methylimidazolium tet-
rafluoroborate ([bmim][BF4]) + W, + AN, + methanol
(MeOH), + propylene carbonate (PC), + dimethyl sulfoxide
(DMSO), and + dichloromethane (DCM); 1-N-butyl-3-N-
methylimidazolium hexafluorophosphate ([bmim][PF6]) + AN;
1-N-butyl-3-N-methylimidazolium dicyanamide ([bmim][DCA])
+ W and + AN; and 1-N-hexyl-3-N-methylimidazolium tet-
rafluoroborate ([hmim][BF4]) + W and + AN. The results were
compared with available literature data.
Materials. The following chemicals were used for preparing
the RTILs: 1-methylimidazole (MI, Merck & Carl Roth 99 %)
was distilled over KOH under reduced pressure, stored over 4
Å molecular sieves, and redistilled under reduced pressure prior
to use. 1-Bromoethane (Merck g 99 %), 1-chlorobutane (Merck
g 99 %), 1-bromobutane (Merck g 98 %), and 1-chlorohexane
(Merck g 99 %) were distilled prior to use. The salts AgBF4
(fluorochem, 99 %), NaBF4 (VWR Prolabo 98.6 %), and KPF6
(fluorochem, 99 %) were dried but otherwise used as received,
whereas sodium dicyanamide ([Na][DCA], Fluka g 96 %) was
recrystallized from methanol.
High-purity solvents were used throughout. Water was
purified with a Millipore MILLI-Q purification unit, yielding
batches with specific resistivity g 18 MΩ·cm-1. AN (Merck
g 99.9 %) was distilled over CaH2 and stored over 4 Å
molecular sieves. The GC purity of the used product was >
99.99 %, and coulometric Karl Fischer titration (Mitsubishi
Moisturemeter MCI CA-02) yielded mass fraction H2O <
50·10-6. MeOH (Merck g 99.9 %) was distilled over Mg/I2
yielding GC purity > 99.99 % and mass fraction H2O <
20 · 10-6. PC (Merck 99.5 %), DMSO (Merck 99.5 %), and
DCM (Acros 99.9 %) were stored over 4 Å molecular sieves.
Prior to use, these solvents had GC purities > 99.94 % for PC
and > 99.99 % for DMSO and DCM. Their water mass fractions
were < 20·10-6
.
The sample [emim][BF4]#1 purchased from IoLiTec had a
stated purity of > 98 %. It was dried and stored like the other
RTILs (see below) but otherwise used as received.
Synthesis. All RTILs were synthesized from MI via appropri-
ate halides (see below). Except for [emim][BF4], previously
published routes were followed.13-15
[emim][Br], [bmim][Cl], [bmim][Br], and [hmim][Cl] were
obtained by adding a slight molar excess, nRHal ≈ 1.1nMI, of
As possible reasons for the observed differences between
literature data for κ and the present results, effects of electrode
polarization and sample purity are discussed. To quantify the
* To whom correspondence should be addressed. E-mail: richard.buchner@
chemie.uni-regensburg.de.
† Part of the special issue “Robin H. Stokes Festschrift”.
10.1021/je800468h CCC: $40.75
2009 American Chemical Society
Published on Web 10/25/2008

Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009 473
Table 1. Conductivities, K, of the Investigated Neat Ionic Liquids at 25 °Ca
κ/S·m-1
[emim][BF4]
[bmim][BF4]
[bmim][PF6]
[bmim][DCA]
[hmim][BF4]
this work
literature
1.553 ( 0.008b
1.63 ( 0.088
1.399
0.353 ( 0.002
0.194 ( 0.00225
0.3526
0.1469 ( 0.0007
0.146 ( 0.00322
0.144 ( 0.00423
0.159 ( 0.00224
0.157 ( 0.00225
0.1527
1.139 ( 0.006
0.1228 ( 0.0006
1.131
0.1628
1.426
0.3627
1.528
0.4628
1.430
1.571 ( 0.232
a Standard uncertainties are given where available. b Sample [emim][BF4]#2.
Table 2. Investigated Mole Fractions, xIL, Corresponding Molar Concentrations, cIL, Densities, G, Conductivities, K, and Molar Conductivities,
Λ, of [bmim][BF4] + Acetonitrile Mixtures at 25 °Ca
b
10-3cIL
10-3cIL
F
Fb
κ
104Λ
104Λb
xIL
mol·m-3
mol·m-3
kg·m-3
kg·m-3
S·m-1
S m2 ·mol-1
S m2 ·mol-1
0.009414
0.01981
0.03142
0.04389
0.07179
0.1077
0.1538
0.2149
0.2966
0.3900
0.6126
0.7544
0.8943
1
0.1744
0.3588
0.5545
0.7538
1.162
1.622
2.123
2.666
3.239
3.741
4.549
4.893
5.157
5.319
0.1744
0.3588
0.5545
0.7538
1.162
1.622
2.123
2.666
3.239
3.741
4.548
4.892
5.155
5.316
792.87
809.82
827.13
844.54
879.61
792.87
809.81
827.12
844.54
879.61
1.629
2.71
3.51
4.12
4.88
5.19
5.04
4.50
3.62
93.4
75.4
63.3
54.7
42.0
32.0
23.8
16.89
11.16
7.27
93.4
75.4
63.3
54.7
42.0
32.0
23.8
16.89
11.16
7.27
918.10
959.09
918.10
959.08
1002.40
1047.32
1085.90
1146.22
1171.38
1190.63
1202.19
1002.38
1047.29
1085.85
1146.07
1171.11
1190.20
1201.64
2.72
1.285
0.802
0.500
0.353
2.82
2.82
1.639
0.970
0.664
1.640
0.970
0.664
a The standard uncertainty of F is 0.05 kg ·m-3; the relative standard uncertainty of κ and Λ is 0.5 %. b Corrected for the damping effect in vibrating
tube density measurements.
the appropriate alkyl halide (1-bromoethane, 1-chlorobutane,
1-bromobutane, or 1-chlorohexane) to a stirred solution (∼ 40
% v/v) of MI in AN under an atmosphere of dry nitrogen. The
formed 1-N-alkyl-3-N-methylimidazolium halides [emim][Br],
[bmim][Cl], and [bmim][Br] were recrystallized at least four
times from AN to give colorless products, whereas [hmim][Cl]
was washed twice with ethyl acetate (Merck p.a.).
[emim][BF4] was obtained via anion metathesis from equimo-
lar amounts of [emim][Br] and NaBF4 dissolved in acetonitrile.
After evaporation of the solvent at a vacuum line, dichlo-
romethane was added in excess. The precipitated NaBr was
separated by filtration, and DCM was subsequently removed
by distillation. This procedure yielded [emim][BF4] with a Br-
mass fraction of 0.016. By adding equimolar amounts of AgBF4
dissolved in MeOH, the excess Br- was precipitated as AgBr
and removed by filtration. Then, the mixture was kept overnight
at ca. -18 °C leading to phase separation into a yellowish
methanol phase and a colorless RTIL-rich phase, which was
isolated.
[bmim][BF4], [bmim][PF6], and [hmim][BF4] were obtained
via anion metathesis from equimolar amounts of the corre-
sponding 1-N-alkyl-3-N-methylimidazolium halides dissolved
in DCM and metal salts (NaBF4, KPF6 as aqueous solution).
The aqueous phase was extracted three times with DCM to
collect the RTILs in the organic phase, which was subsequently
washed thrice with small amounts of water to remove traces of
metal halides (NaCl, NaBr, KCl) before distilling off the solvent.
All preparation steps involving aqueous phases were performed
rapidly with materials cooled to ≈ 0 °C prior to use to minimize
the hydrolysis of [BF-4 ] and [PF6-].
precipitating NaCl filtered off. After removal of the solvent at
a vacuum line, this procedure was repeated, yielding a slightly
yellowish product.
Sample Preparation. All RTILs were dried at a high-vacuum
line (p < 10-8 bar) for 7 days at ≈ 40 °C prior to use. Water
mass fractions were always < 50·10-6. Potentiometric titration
of RTIL samples in aqueous solution against a standard solution
of AgNO3 (Carl Roth) yielded halide mass fractions < 20·10-6
for [bmim][BF4], [bmim][PF6], and [hmim][BF4], as well as <
0.5 % for [bmim][DCA]. Halide impurities could neither be
detected for the commercial nor for the synthesized sample of
[emim][BF4]. Except for [emim][BF4]#1, where a signal arising
from an acidic proton (≈ 0.01 mol fraction of impurity) was
observed at a chemical shift of ≈ 6.5·10-6, no contaminations
were detected with 1H NMR and 19F NMR (where applicable).
All dried RTILs were stored in an N2-filled glovebox. N2-
protection was also maintained when preparing the mixtures
and during all subsequent steps of sample handling, including
the measurements. Mixtures were prepared immediately before
use on an analytical balance without buoyancy correction and
thus have a standard uncertainty of about ( 0.2 %.
ConductiWity. Measurements were performed with the equip-
ment described by Barthel and co-workers,16,17 consisting of a
home-built precision thermostat stable to < 0.003 K in
combination with a cold source (Lauda Kryomat K 90 SW), a
manually balanced high-precision conductivity bridge, and a set
of five two-electrode capillary cells. The cells with cell constants
C in the range of (25 to 360) cm-1 were calibrated with aqueous
KCl.18
All measurements were carried out at (25 ( 0.01) (NIST
traceable Pt sensor and bridge, ASL) by recording the cell
resistance as a function of AC frequency, ν, between 120 Hz
and 10 kHz. To eliminate electrode polarization, the conductiv-
[bmim][DCA] was obtained by stirring equimolar amounts
of [bmim][Cl] and [Na][DCA] at ≈ 35 °C overnight. To separate
the ionic liquid, an excess of DCM was added and the

474 Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009
ity, κ, of each sample was obtained as κ ) C/R∞ from the
extrapolation R∞ ) limνf∞ R(ν) using the empirical function
R(ν) ) R∞ + A/νa; A is specific to the cell and a ≈ 0.5.19
Measurements of selected samples with different cells agreed
within 0.5 %. Thus, this value may be taken as a good estimate
for the relative uncertainty of the present κ and the derived molar
conductivities, Λ ) κ/cIL.
Density. To convert RTIL mole fractions, xIL, into molar
concentrations, cIL, the densities, F, of the mixtures were
determined at (25 ( 0.02) °C with a vibrating tube densimeter
(Anton Paar DMA 60, DMA 601HT). The instrument was
calibrated with degassed water and purified nitrogen at atmo-
spheric pressure, assuming densities from standard sources.20
The precision of the measurements was ( 0.001 kg ·m-3
.
Taking into account the sources of error (calibration, measure-
ment, purity of materials), we estimate the uncertainty of F to
be within ( 0.05 kg ·m-3
.
Results and Discussion
ConductiWities of Neat Ionic Liquids. Table 1 summarizes
the measured κ together with available literature data for the
investigated ionic liquids. Where possible, published uncertain-
ties were included. The conductivity of a sample depends on
the charge density (which is high in RTILs) and on the mobility
of the ions, which is coupled to viscosity via the Stokes-Einstein
relation.21 As a consequence of their large viscosity, pure RTILs
generally exhibit only moderate values of κ compared to their
binary mixtures with polar solvents (see below).
The literature survey reveals a considerable scatter of the
available κ for the pure RTILs. For each studied RTIL, our data
are usually in reasonable agreement with some of the published
values. Especially for [bmim][PF6], which is easily accessible
in high purity, our value isswithin the claimed errorssin
quantitative agreement with the data of Widegren et al.22 and
Kanakubo et al.,23 but it is significantly smaller than those
published by Li et al.24,25 For [emim][BF4], larger deviations
of up to 10 % may reflect the more demanding synthesis of
this RTIL, but differences of 30 % for [hmim][BF4] and even
45 % for [bmim][BF4] are surprising as the purification of these
RTILs is straightforward.
One might think that the obvious discrepancies appearing in
Table 1 reflect the purity of the samples used by the various
groups. However, the published contents of water and halide
impurities are generally small.9,22-28 Measurement of our first
[bmim][BF4] sample, which was prepared via the [bmim][Br]
route and contained 1.5·10-4 mass fraction Br- yielded κ )
0.360 S · m-1, whereas subsequent batches, prepared via
[bmim][Cl] and containing < 20·10-6 mass fraction Cl-, had
κ ) 0.353 S ·m-1. Thus, from the sample purities stated in the
literature, a scatter of κ in the order of (2 to 3) % would be
expected.
Further possible sources of error are sample handling and
the measurement of conductivity itself. Generally, commercial
instruments operating at a fixed AC frequency (usually 1 kHz)
were used for the literature data of Table 1. However, it is well-
known that the measured cell impedance depends on frequency
due to the double-layer capacitance of the electrodes, so that
precise determinations of κ require correction for this systematic
error.19 In addition to comparable sample purity, the correction
for electrode polarization applied by the groups of Widegren et
al.,22 Kanakubo et al.,23 and by us is certainly the major reason
for the excellent agreement of the corresponding κ values.
However, perusal of our data revealed that conductivities
corrected to ν-1 f 0 differed from κ determined at 1 kHz by
Figure 1. Conductivities, κ, of mixtures of (a) [emim][BF4]#2 (squares),
(b) [bmim][BF4] (circles), and [hmim][BF4] (triangles) with water at 25
°C. Data represented by filled symbols were determined in this work, and
lines represent fits with eq 2. Open symbols refer to the data of 0, ref 8;
square with an × in it, ref 9; O, ref 25. The observed phase boundaries (+)
for [hmim][BF4] + W are indicated.
1 % at most. Thus, electrode polarization cannot explain the
large deviations noted from Table 1.
It is well-known that the anions [BF-4 ] and [PF6-] are prone
to hydrolysis33-37 with one of the reaction products being HF,
which may attack the conductivity cell. To study the effect of
anion hydrolysis, κ was measured in a PTFE flask at 1 kHz as
a function of time, t, for a [bmim][BF4] + W mixture of xIL
)
0.008604 (see Supporting Information). Within six days, after
which the experiment was stopped, κ rose from 2.41 S·m-1 to
2.59 S ·m-1 approaching a plateau value (κ(∞) ≈ 1.1κ(t0)) in a
(probably pseudo-) first-order reaction with half-life τ ≈ 1.2
days. Elucidating the mechanism of [BF-4 ] hydrolysis was
beyond the scope of this study, but the rather moderate increase
of κ suggests that some equilibrium is reached well before all
[BF-4 ] is decomposed. Thus, [BF4-] and [PF-6 ] hydrolysis may
contribute to the scatter of κ values reported in Table 1 but
also cannot explain the obvious outliers for [emim][BF4],9
[bmim][BF4],25,28 and [hmim][BF4]28 if we assume that these
samples were sufficiently dry and properly stored.29 In any case,
with the measurement procedure followed in our experiments
(see above), errors in κ associated with the hydrolysis of [BF-4 ]
and [PF-6 ] remained well below 0.5 %.
Jarosik et al.9 and Leys et al.28 used impedance spectroscopy
in combination with parallel-plate capacitor cells to determine
κ. Possibly, such equipment is difficult to calibrate for high-
precision conductivity measurements. Notable in the investiga-

Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009 475
tion of Li et al.25 is the rather low density of 1182.2 kg·m-3
reported for [bmim][BF4] at 25 °C, suggesting major contami-
nations of the sample. In contrast, our value of F ) 1202.19
kg·m-3 is in good agreement with previously reported densities
Table 3. Investigated Mole Fractions, xIL, Corresponding Molar
Concentrations, cIL, Densities, G, Conductivities, K, and Molar
Conductivities, Λ, of Binary RTIL + Acetonitrile Mixtures at 25
°Ca
10-3cIL
F
κ
104
Λ
of 1201.420 kg ·m-3 38
,
1202 kg ·m-3 27
,
1201.34 kg ·m-3 39
,
xIL
mol·m-3
kg·m-3
S·m-1
S·m2 ·mol-1
(1201.29 and 1201.4) kg ·m-3 40
,
and 1202.298 kg ·m-3 41
.
ConductiWities of Binary Mixtures. The κ values for the
investigated binary mixtures, covering the entire miscibility
range, are listed in Tables 2 to 5. Selected results are displayed
in Figures 1 to 3.
[emim][BF4]#1
795.48
0.01082
0.01646
0.02252
0.04929
0.08059
0.1218
0.1736
0.2371
0.3264
0.4565
0.6520
0.8336
1
0.2014
0.3036
0.4114
0.8621
1.342
1.903
2.513
3.146
3.867
4.672
5.528
6.092
6.484
1.90
2.55
3.13
4.89
6.03
6.72
6.92
6.65
5.90
4.64
3.10
2.11
1.553
94.4
84.0
76.1
56.7
44.9
35.3
27.6
21.2
15.26
9.94
5.61
3.46
2.40
804.68
814.30
853.38
893.87
939.88
988.57
1038.16
1093.25
1153.25
1215.55
1255.87
1283.70
All investigated systems are completely miscible at 25 °C,
except for [hmim][BF4] + W. For this system, Wagner et al.37
determined an upper critical solution temperature of Tc ) 58.16
°C at the RTIL mass fraction of 0.4097 (xIL ≈ 0.047). At 25
°C, we obtained the phase boundaries for the RTIL-rich
IL
IL
(indicated by superscript “IL”; x ) 0.2750) and the W-rich
phases (superscript “W”; xWIL ) 0.0050) by comparing the
conductivities of the saturated phases (κIL ) 2.07 S ·m-1 and
κW ) 1.53 S ·m-1) with suitable extrapolations of κ ) f(xIL)
from the corresponding homogeneous regions (Figure 1b).
The dependence of conductivity on RTIL concentration
follows the typical pattern of concentrated electrolyte solutions:
after a rapid initial rise at low xIL due to the increasing number
of charge carriers, κ passes through a pronounced maximum at
xIL ≈ 0.1 to 0.25 (depending on the mixture) due to the counter-
balancing effect of the rapidly rising viscosity on ion mobility.
To reproduce such a concentration dependence, the empirical
Casteel-Amis42 equation is widely used. Generally, data fitting
is done on the molality scale, and obtained parameters are the
maximum conductivity, κmax, the corresponding concentration,
and shape parameters, a and b. For the present systems, the
Casteel-Amis equation can be reasonably applied in the mole-
fraction scale, that is
[emim][BF4]#2
791.26
0.008411
0.01528
0.02253
0.04942
0.1007
0.1433
0.2043
0.2778
0.3840
0.4888
0.6478
0.7590
0.8902
1
0.1571
0.2823
0.4113
0.8649
1.623
2.165
2.829
3.490
4.245
4.828
5.500
5.867
6.221
6.468
1.585
2.46
3.14
4.94
6.50
6.94
6.91
6.41
5.41
4.44
3.20
2.54
1.95
1.553
100.9
87.1
76.4
57.2
40.0
32.0
24.4
18.36
12.75
9.21
5.81
4.33
3.13
2.40
802.59
813.87
854.17
916.26
960.07
1012.50
1063.36
1120.11
1163.04
1211.72
1237.95
1263.01
1280.38
[bmim][PF6]
795.63
815.17
835.56
855.89
0.007571
0.01576
0.02489
0.03448
0.05838
0.08779
0.1261
0.1772
0.2521
0.3649
0.5591
0.7291
0.8729
1
0.1404
0.2863
0.4415
0.5970
0.9529
1.338
1.766
2.238
2.780
3.374
4.038
4.415
4.652
4.815
1.485
2.54
3.39
4.05
5.00
5.39
5.27
4.68
3.63
105.8
88.7
76.8
67.9
52.5
40.3
29.8
20.9
13.04
6.82
a
xIL
xIL - xmax
κ ) κmax
exp b(xIL - xmax)2 - a
(1)
901.77
950.71
(
)
[
]
xmax
xmax
1004.47
1062.70
1128.61
1199.71
1278.14
1321.97
1349.76
1368.32
where xmax is the RTIL mole fraction at κmax
.
The obtained fit parameters are summarized in Table S1 of
the Supporting Information together with the corresponding
regression coefficients, r2, and the standard deviations, σ, of
the fits. Generally, the four-parameter description of κ ) f(xIL)
with eq 1 is excellent, yielding r2 > 0.999 and σ < 4·10-2
S·m-1 for most mixtures. Equation 1 can even be applied for
[hmim][BF4] + W, where the data sets on both sides of the
miscibility gap can be well described with a single set of
parameters (xmax, κmax, a, b) albeit with xmax in the biphasic
region. However, for several mixtures, systematic deviations
around xmax are obvious (Figure S3 of the Supporting Informa-
tion).
2.30
0.995
0.469
0.253
0.1469
2.47
1.062
0.543
0.305
[hmim][BF4]
784.24
0.004117
0.008445
0.01744
0.03816
0.06486
0.09763
0.1399
0.1943
0.2712
0.3907
0.5902
0.7435
0.8889
1
0.07700
0.1560
0.3142
0.6489
1.028
0.1427
1.855
2.300
2.785
3.322
3.894
4.183
4.287
4.512
0.838
1.442
2.35
3.51
4.13
4.25
3.96
3.37
2.53
1.543
0.656
0.344
0.192
0.1228
108.8
92.5
74.7
54.2
40.2
29.8
21.4
14.67
9.08
4.65
791.63
806.26
836.42
869.76
903.86
939.67
975.94
1014.86
1056.74
1100.45
1122.12
1137.20
1146.34
A significantly better interpolation of the present conductivity
data can be obtained with the Pade´ (n, m) approximation
n
Ai(2xIL - 1)i-1
1.685
0.823
0.437
0.272
∑
1 +
i)1
κ ) xIL(1 - xIL)
+ C1xIL
(2)
m
Bj(2xIL - 1)j
∑
j)1
a The stardard uncertainty of F is 0.05 kg ·m-3; the relative standard
uncertainty of κ and Λ is 0.5 %.
at the cost of an increased number of adjustable parameters. In
eq 2, κ ) 0 is assumed for the pure solvent (valid within our
error limits for κ), whereas parameter C1 represents the
conductivity of the pure RTIL. In the fitting procedure, C1 was
adjusted but always found to be close to the measured data.
The best results were obtained by using a seven-parameter fit
with n ) 4 and m ) 2. The parameters Ai, Bi, and C1 obtained
by least-squares analysis are summarized in Table S2 of the
Supporting Information, together with the corresponding stan-
dard deviations of the fits, σ (r2 > 0.999 in all cases). The lines
in Figures 1 to 3 represent such fits.
The effect of sample purity on conductivity was checked by
comparing the commercial sample [emim][BF4]#1 with material
synthesized in our laboratory, [emim][BF4]#2, in mixtures with

476 Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009
Table 4. Investigated Mole Fractions, xIL, Corresponding Molar
Concentrations, cIL, Densitites, G, Conductivities, K, and Molar
Conductivities, Λ, of Binary RTIL + Water Mixtures at 25 °Ca
Table 5. Investigated Mole Fractions, xIL, Corresponding Molar
Concentrations, cIL, Densities, G, Conductivities, K, and Molar
Conductivities, Λ, of Binary [bmim][BF4] + Polar Solvent Mixtures
at 25 °Ca
10-3cIL
F
κ
104
Λ
10-3cIL
F
κ
104
Λ
xIL
mol·m-3
kg·m-3
S·m-1
S·m2 ·mol-1
xIL
mol·m-3
kg·m-3
S·m-1
S·m2 ·mol-1
[emim][BF4]#2
1007.69
1018.31
1040.22
1061.36
1087.31
1110.93
1136.45
1168.60
1198.61
1233.28
1283.70
MeOH
802.48
818.28
834.29
849.76
885.02
921.13
961.01
1002.48
1046.31
1094.85
1145.28
0.004746
0.01001
0.02226
0.03637
0.05733
0.08102
0.1150
0.1740
0.2564
0.4190
1
0.2534
0.5144
1.052
1.572
2.200
2.761
3.376
4.122
4.790
5.532
6.484
1.874
3.26
5.35
6.81
8.08
8.84
9.24
9.08
8.13
5.99
1.553
73.9
63.4
50.9
43.3
36.7
32.0
27.4
22.0
16.98
10.82
2.40
0.007409
0.01543
0.02435
0.03355
0.05717
0.08589
0.1242
0.1755
0.2480
0.3624
0.5583
0.1776
0.3603
0.5526
0.7396
1.173
1.624
2.127
2.662
3.237
3.877
4.556
0.867
1.451
1.97
2.40
3.30
3.81
4.18
4.26
3.92
3.09
1.760
48.8
40.2
35.6
32.4
28.1
23.4
19.7
15.99
12.11
7.98
3.86
[bmim][BF4]
1007.00
1013.47
1030.78
1047.81
1065.19
1084.47
1103.47
1124.77
1137.30
1154.46
1171.08
1177.86
1183.94
1190.14
1193.53
1198.74
1202.19
PC
0.005053
0.008751
0.01927
0.03304
0.04969
0.07385
0.1053
0.1566
0.1991
0.2824
0.4140
0.4919
0.5749
0.6838
0.7594
0.8931
1
0.2669
0.4471
0.9018
1.391
1.867
2.400
2.910
3.482
3.810
4.247
4.656
4.815
4.946
5.078
5.150
5.253
5.319
1.720
2.47
3.59
4.46
4.90
5.17
5.21
4.93
4.55
3.71
2.57
2.04
1.574
1.098
0.846
0.527
0.353
64.3
55.2
39.8
32.1
26.3
21.5
17.92
14.15
11.94
8.73
5.51
4.23
3.18
2.16
1.643
1.003
0.664
0.02318
0.04836
0.1028
0.1601
0.2277
0.3056
0.4070
0.5126
0.6192
0.7173
0.8497
0.2653
0.5380
1.078
1.583
2.107
2.634
3.219
3.733
4.174
4.523
4.930
1201.32
1202.51
1204.30
1205.43
1206.17
1206.51
1206.44
1206.03
1205.48
1204.39
1203.39
0.537
0.873
1.230
1.346
1.332
1.229
1.055
0.875
0.715
0.593
0.463
20.2
16.22
11.41
8.51
6.32
4.67
3.28
2.34
1.713
1.310
0.939
DMSO
1101.56
1106.21
1117.80
1128.31
1139.50
1150.71
1159.40
1171.71
1178.20
1184.27
1192.13
0.01925
0.03427
0.07837
0.1247
0.1866
0.2554
0.3251
0.4423
0.5246
0.6132
0.7494
0.2618
0.4556
0.9763
1.457
2.011
2.536
2.986
3.611
3.969
4.301
4.728
0.687
1.029
1.558
1.715
1.672
1.502
1.310
1.015
0.854
0.711
0.543
26.2
22.6
15.96
11.77
8.31
5.92
4.39
2.81
2.15
[bmim][DCA]
1003.22
1011.25
1019.63
1025.43
1031.05
1040.01
1047.31
1051.86
1055.38
1058.33
1059.12
0.009639
0.02791
0.05432
0.07995
0.1112
0.1933
0.3214
0.4686
0.6386
0.8742
1
0.4879
1.214
1.965
2.485
2.952
3.708
4.304
4.661
4.898
5.092
5.160
2.02
3.49
4.43
4.80
4.90
4.47
3.55
2.71
2.00
1.366
1.139
41.5
28.7
22.5
19.3
16.60
12.04
8.25
5.82
4.09
2.68
2.21
1.653
1.149
DCM
1316.60
1314.92
1313.40
1311.80
1304.97
1294.06
1284.29
1271.96
1261.43
1242.95
1230.33
1216.18
1211.19
1207.46
1202.19
0.007550
0.01921
0.02714
0.03878
0.07987
0.1395
0.1940
0.2674
0.3355
0.4787
0.6019
0.7730
0.8496
0.9107
1
0.1156
0.2883
0.4016
0.5627
1.083
1.726
2.219
2.773
3.199
3.902
4.360
4.846
5.024
5.152
5.319
0.0602
0.200
0.310
0.436
0.980
1.427
1.550
1.607
1.530
1.209
0.930
0.620
0.511
0.441
0.353
5.21
6.92
7.72
7.76
9.05
8.27
6.99
5.80
4.78
3.10
2.13
1.278
1.018
0.856
0.664
[hmim][BF4]
998.06b
0.0007116
0.001428
0.001482
0.002238
0.002960
0.002960
0.003642
0.004470
0.004747
0.2816
0.3140
0.3441
0.3813
0.4148
0.5139
0.5161
0.7069
1
0.03906
0.07778
0.08064
0.1208
0.1568
0.1586
0.1937
0.2356
0.2495
3.712
3.806
3.881
3.962
4.024
4.170
4.173
4.354
4.512
0.327
0.567
0.608
0.854
1.049
1.064
1.225
1.411
1.466
2.03
1.829
1.655
1.456
1.293
0.898
0.885
0.417
0.1228
83.8
72.9
75.4
70.7
66.9
67.1
63.3
59.9
58.8
5.48
4.81
4.27
3.68
3.21
2.16
2.12
0.958
0.272
999.52
999.64b
1001.24b
1002.63
1002.70b
1003.94b
1005.24
1005.61b
1113.63b
1116.76
1119.43b
1122.43
1124.75b
1130.58
1130.73b
1138.70b
1146.34
a The standard uncertainty of F is 0.05 kg ·m-3; the relative standard
uncertainty of κ and Λ is 0.5 %.
the claimed accuracy of our instrumentation. On the other hand,
systematically smaller conductivities were observed for the
commercial sample #1. This seems surprising in view of the
acid-proton signal detected by NMR for this sample but may
reflect a higher viscosity of batch #1.
a The standard uncertainty of F is 0.05 kg ·m-3; the relative standard
uncertainty of κ and Λ is 0.5 %. b Interpolated values.
Figure 1a compares the κ values obtained for [emim][BF4]#2
+ W in this study with results (κlit) published by Vila et al.8
and by Jarosik et al.9 At low RTIL contents, the agreement is
fairly good, and the present values for xmax and κmax are roughly
consistent with those of Jarosik et al.9 However, for xIL > 0.1
rather large relative deviations, up to δlit ) (κ - κlit)/κ ≈ 0.2,
AN. The results are summarized in Figure 2c as relative
deviations δfit ) (κfit - κ)/κfit of the experimental data, κ, from
the interpolations, κfit, obtained from fitting the data set for
[emim][BF4]#2 with eq 2. For sample #2, the random scatter
of κ around κfit was generally well below 0.5 % and thus below

Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009 477
Figure 3. Conductivities, κ, of [bmim][BF4] + DCM mixtures at 25 °C.
Data represented by filled symbols were determined in this work, and the
line shows the fit with eq 2. Open symbols refer to the data of ref 25.
Figure 4. Molar conductivities, Λ, of mixtures of acetonitrile with 9,
[emim][BF4]#2; b, [bmim][BF4]; and 2, [hmim][BF4] at 25 °C. Lines
represent fits with eq 3. Open symbols represent [bmim][BF4] data from
x, ref 24; O, ref 25; #, ref 51.
Figure 2. Conductivities, κ, of mixtures of (a) [bmim][PF6] (triangles), (b)
[bmim][BF4] (circles), and [emim][BF4]#2 (squares) with acetonitrile at
25 °C. Data represented by filled symbols were determined in this work,
and lines represent fits with eq 2. Open symbols refer to: open triangle
pointing right, X, ref 24; open triangle pointing left, O, ref 25. Panel (c)
shows the relative deviations, δfit, of the experimental conductivities of
samples [, [emim][BF4]#1 and 9, [emim][BF4]#2 from the fit of the latter
data set with eq 2.
Molar ConductiWities of Binary Systems. Tables 2 to 5
include the molar conductivities, Λ ) κ/cIL, of the investigated
mixtures. Typical examples are shown in Figures 4 and 5. The
data were fitted to the equation
Λ ) D - D c + D c ln c + D c
4 IL
(3)
√
2
1
IL
3
IL
IL
occur. As already discussed for the pure RTIL, these deviations
are possibly connected to the technique used by these authors.
The conductivity peak of Vila et al.8 is significantly smaller
and at a higher RTIL mole fraction so that at low xIL the present
κ exceed their data, whereas for RTIL-rich mixtures κ < κlit.
The only plausible explanation here is sample contamination
well beyond the claimed purity.
which corresponds to the truncated series expression derived
for dilute electrolyte solutions from the theory of Onsager,43,44
but without a theoretical meaning of the parameters Di, i )
1,..., 4. The obtained parameters and the corresponding r2 and
σ values are summarized in Table S3 of the Supporting
Information.
For [bmim][BF4] + W, a comparison with the data of Li et
al.24 (Figure 1b) is possible. Since already the conductivity of
pure [bmim][BF4] reported by these authors is significantly
smaller than the value of this investigation (see above), it is
not surprising that also for the mixtures κlit < κ with 0.2 j δlit
j 0.5. Similar deviations were observed for the systems
[bmim][BF4] + AN (Figure 2b) and [bmim][BF4]/DCM (Figure
3).24,25 On the other hand, for [bmim][PF6] + AN (Figure 2a),
the data of Li et al.25,24 are in reasonable agreement with the
present κ for xIL > 0.4 but deviate increasingly on further
dilution.
Reliable calculation of Λ requires accurate values for the
molar electrolyte concentration, cIL, and thus the density, F, of
the solution. Vibrating-tube densimetry is very precise. However,
the oscillation period of the vibrating tube depends on sample
viscosity, η, leading to systematic errors in F.45 Generally, this
effect is small; however, pure RTILs are often very viscous,
and Heintz et al. showed that here viscosity correction of the
density data is worthwhile.46
Unfortunately, viscosity data for RTIL mixtures are scarce
and published datasespecially for the pure RTILssoften scatter
considerably. Using the results of Wang et al.,47 we checked

478 Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009
with polar solvents (water, propylene carbonate, DMSO,
methanol, and dichloromethane) at 25 °C were reported. The
data for κ and Λ were compared with literature values where
possible. While our results for the pure RTILs conform
excellently with some of the published data, the agreement with
the limited number of mixture studies is poor.
As possible sources of error, electrode polarization and sample
purity (as stated in the quoted literature) were considered. [BF-4 ]
hydrolysis was discussed. The effect of viscosity on the accuracy
of F and thus Λ was checked. It was found that these effects
may explain some of the observed deviations of our results from
the literature. However, they cannot account for the very large
discrepancies seen in Figures 1 to 5. This suggests that some
of the data found in the literature are flawed by major
contaminations of the studied samples.
Acknowledgment
Figure 5. Molar conductivity, Λ, of [bmim][BF4] + DCM mixtures at 25
°C. Lines represent fits with eq 3 with additional term D5x2. Filled symbols
were determined in our laboratory, and open symbols refer to ref 25.
Dedicated to Professor Robin H. Stokes on the occasion of his 90th
birthday. The authors thank W. Kunz for access to laboratory
facilities and R. Neueder for helpful discussions.
the magnitude of the viscosity correction suggested by Heintz
et al.46 on F, c, and Λ for [bmim][BF4] + AN mixtures (Table
2). The used viscosities were considered reliable as the value
given by the authors for the pure RTIL, η ) 0.110308 Pa·s,47
is in good agreement with those published by several other
groups.25,40,41,48 For the other investigated mixtures, a viscosity
correction of F was not possible due to lacking η(xIL).
Comparison of columns 4 and 5 of Table 2 indicates that for
η J 6 mPa ·s, corresponding to xIL J 0.4, the effect of η on F
is larger than the claimed accuracy (( 0.05 kg·m-3) of our
density measurements. However, even for pure [bmim][BF4],
the systematic error in F and thus cIL does not exceed 0.05 %.
Since our accuracy in κ is only ( 0.5 %, viscosity correction
of the reported molar conductivities is negligible. Obviously,
this will not be true for a discussion of excess volumes and
related quantities.
For the investigated mixtures, it was found that in
generals[bmim][BF4] + DCM being the only exceptionsΛ
decreases continuously with increasing RTIL concentration
(Tables 2 to 5, Figure 4). Probably, this reflects decreasing ion
mobility induced by rising viscosity. Association effects may
contribute, but dielectric relaxation studies suggest that the
concentration of ion pairs in mixtures of RTILs with polar
solvents is small.49 This is probably different for [bmim][BF4]
+ DCM, where a pronounced maximum is observed for Λ at
cIL ≈ 1·103 mol·m-3 (xIL ≈ 0.08) (see Figure 5). To fit the
experimental data of this mixture, eq 3 was extended by the
term D5x2 (D5 ) 1.157·10-10 S·m8 ·mol-3). For [bmim][BF4]
+ DCM mixtures, strong ion association was observed with
dielectric spectroscopy at low cIL, but already at xIL ≈ 0.3 (cIL
≈ 3 · 103 mol · m-3) ion pairs are negligible.50 The observed
increase of Λ may thus indicate the redissociation of ion pairs
as a consequence of increasing ion-ion interactions which is
finally outweighed by decreasing ion mobility due to rising
viscosity.
Supporting Information Available:
Graphs of the time-dependent conductivity of hydrolyzing
[bmim][BF4] + W and of representative fits with eq 1. Tables of
the fit parameters for κ and Λ obtained with eqs 1 to 3. This material
is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited
(1) Welton, T. Room-temperature ionic liquids. Solvents for synthesis and
catalysis. Chem. ReV. 1999, 99, 2071–2083.
(2) Wasserscheid, P.; Welton, T. Ionic Liquids in Synthesis; Wiley-VCH:
Weinheim, 2003.
(3) Endres, F.; El Abedin, S. Z. Air and water stable ionic liquids in
physical chemistry. Phys. Chem. Chem. Phys. 2006, 8, 2101–2116.
(4) Paˆrvulescu, V. I.; Hardacre, C. Catalysis in ionic liquids. Chem. ReV.
2007, 107, 2615–2665.
(5) Marsh, K. N.; Boxall, J. A.; Lichtenthaler, R. Room temperature ionic
liquids and their mixtures s a review. Fluid Phase Equilib. 2004,
219, 93–98.
(6) MacFarlane, D. R.; Seddon, K. R. Ionic liquids s progress on the
fundamental issues. Aust. J. Chem. 2007, 60, 3–5.
(7) Weinga¨rtner, H. Understanding ionic liquids at the molecular level:
facts, problems, and controversies. Angew. Chem., Int. Ed. 2008, 47,
654–670.
(8) Vila, J.; Gine´s, P.; Rilo, E.; Cabeza, O.; Varela, L. M. Great increase
of the electrical conductivity of ionic liquids in aqueous solutions.
Fluid Phase Equilib. 2006, 247, 32–39.
(9) Jarosik, A.; Krajewski, S. R.; Lewandowski, A.; Radzimski, P.
Conductivity of ionic liquids in mixtures. J. Mol. Liq. 2006, 123, 43–
50.
(10) Liu, W.; Zhao, T.; Zhang, Y.; Wang, H.; Yu, M. The physical
properties of aqueous solutions of the ionic liquid [BMIM][BF4]. J.
Solution Chem. 2006, 35, 1337–1346.
(11) Comminges, C.; Barhdadi, R.; Laurent, M.; Troupel, M. Determination
of viscosity, ionic conductivity, and diffusion coefficients in some
binary systems: ionic liquids + molecular solvents. J. Chem. Eng.
Data 2006, 51, 680–685.
(12) Widegren, J. A.; Magee, J. W. Density, viscosity, speed of sound,
and electrolytic conductivity for the ionic liquid 1-hexyl-3-methylimi-
dazolium bis(trifluoromethylsulfonyl)imide and its mixtures with water.
J. Chem. Eng. Data 2007, 52, 2331–2338.
(13) Holbrey, J. D.; Seddon, K. R. The phase behaviour of 1-alkyl-3-
methylimidazolium tetrafluoroborates; ionic liquids and ionic liquid
crystals. J. Chem. Soc., Dalton Trans. 1999, 13, 2133–2139.
(14) Fredlake, C. P.; Crosthwaite, J. M.; Hert, D. G.; Aki, S. N.; Brennecke,
J. F. Thermophysical properties of imidazolium-based ionic liquids.
J. Chem. Eng. Data 2004, 49, 954–964.
Not unexpectedly, comparison of the present Λ values with
the limited number of literature data8,24,25,28,51 reveals significant
deviations (Figures 4 and 5). The possible reasons were already
discussed above.
(15) 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.
Conclusions
(16) Barthel, J.; Wachter R.; Gores H.-J. Temperature dependence of
conductance of electrolytes in nonaqueous solutions. Modern Aspects
of Electrochemistry; Conway, B. E., Bockris, J. O’M., Eds.; 1979;
Vol. 13, pp 1-79.
Conductivities, κ, densities, F, and molar conductivities, Λ,
of selected binary mixtures of the ionic liquids [emim][BF4],
[bmim][BF4], [bmim][PF6], [bmim][DCA], and [hmim][BF4]

Journal of Chemical & Engineering Data, Vol. 54, No. 2, 2009 479
(17) Barthel, J.; Graml, H.; Neueder, R.; Turq, P.; Bernard, O. Electrolyte
conductivity from infinite dilution to saturation. Curr. Top. Solution
Chem. 1994, 1, 223–239.
(18) Barthel, J.; Feuerlein, F.; Neueder, R.; Wachter, R. Calibration of
conductance cells at various temperatures. J. Solution Chem. 1980, 9,
209–219.
(36) Wasserscheid, P.; van Hal, R.; Boesmann, A. New, halogen-free ionic
liquids s synthesis, properties, and applications. Proc. Electrochem.
Soc. 2002, 19, 146–154.
(37) Wagner, M.; Stanga, O.; Schro¨er, W. Corresponding states analysis
of the critical points in binary solutions of room temperature ionic
liquids. Phys. Chem. Chem. Phys. 2003, 5, 3943–3950.
(19) Hoover, T. B. Conductance of potassium chloride in highly purified
N-mehylpropionamide from 20 to 40 °. J. Phys. Chem. 1964, 68, 876–
879.
(20) Lide, D. R. CRC Handbook of Chemistry and Physics; CRC Press:
Boca Raton, FL, 2004.
(21) Robinson, R. A.; Stokes, R. H. Electrolyte Solutions; Butterworths:
London, 1959.
(22) Widegren, J. A.; Saurer, E. M.; Marsh, K. N.; Magee, J. W. Electrolytic
conductivity of four imidazolium-based room-temperature ionic liquids
and the effect of a water impurity. J. Chem. Thermodyn. 2005, 37,
569–575.
(23) Kanakubo, M.; Harris, K. R.; Tsuchihashi, N.; Ibuki, K.; Ueno, M.
Effect of pressure on transport properties of the ionic liquid 1-butyl-
3-methylimidazolium hexafluorophosphate. J. Phys. Chem. B 2007,
111, 2062–2069.
(24) Li, W.; Han, B.; Tao, R.; Zhang, Z.; Zhang, J. Measurement and
correlation of the ionic conductivity of ionic liquid-molecular solvent
solutions. Chin. J. Chem. 2007, 25, 1349–1356.
(25) Li, W.; Zhang, Z.; Han, B.; Hu, S.; Xie, Y.; Yang, G. Effect of water
and organic solvents on the ionic dissociation of ionic liquids. J. Phys.
Chem. B 2007, 111, 6452–6456.
(26) Nishida, T.; Tashiro, Y.; Yamamoto, M. Physical and electrochemical
properties of 1-alkyl-3-methylimidazolium tetrafluoroborate for elec-
trolyte. J. Fluorine Chem. 2003, 120, 135–141.
(27) Tokuda, H.; Tsuzuki, S.; Susan, M. A. B. H.; Hayamizu, K.; Watanabe,
M. How ionic are room-temperature ionic liquids? An indicator of
the physicochemical properties. J. Phys. Chem. B 2006, 110, 19593–
19600.
(28) Leys, J.; Wu¨bbenhorst, M.; Menon, C. P.; Rajesh, R.; Thoen, J.;
Glorieux, C.; Nockemann, P.; Thijs, B.; Binnemans, K.; Longuemart,
S. Temperature dependence of the electrical conductivity of imida-
zolium ionic liquids. J. Chem. Phys. 2008, 128, 064509.
(29) The situation seems to be different for wet samples stored in glass
vessels. For a [bmim][BF4] + W mixture with xIL ) 0.008604, we
observed a strong increase of κ after two days (Figure S2 of the
Supporting Information).
(30) Fuller, J.; Carlin, R. T.; Osteryoung, R. A. The room temperature ionic
liquid 1-ethyl-3-methylimidazolium tetrafluoroborate: electrochemical
couples and physical properties. J. Electrochem. Soc. 1997, 144, 3881–
3886.
(31) Yoshida, Y.; Baba, O.; Saito, G. Ionic liquids based on dicyanamide
anion: influence of structural variations in cationic structures on ionic
conductivity. J. Phys. Chem. B 2007, 111, 4742–4749.
(32) Vila, J.; Gine´s, P.; Pico, J. M.; Franjo, C.; Jime´nez, E.; Varela, L. M.;
Cabeza, O. Temperature dependence of the electrical conductivity in
EMIM-based ionic liquids Evidence of Vogel-Tamman-Fulcher be-
havior. Fluid Phase Equilib. 2006, 242, 141–146.
(38) Zafarani-Moattar, M. T.; Shekaari, H. Application of Prigogine-Flory-
Patterson theory to excess molar volume and speed of sound of 1- n
-butyl-3-methylimidazolium hexafluorophosphate or 1-n-butyl-3-me-
thylimidazolium tetrafluoroborate in methanol and acetonitrile.
J. Chem. Thermodyn. 2006, 38, 1377–1384.
(39) Iglesias-Otero, M. A.; Troncoso, J.; Carballo, E.; Roman´ı, L. Density
and refractive index for binary systems of the ionic liquid [Bmim][BF4]
with methanol, 1,3-dichloropropane, and dimethyl carbonate. J.
Solution Chem. 2007, 36, 1219–1230.
(40) Harris, K. R.; Kanakubo, M.; Woolf, L. A. Temperature and pressure
dependence of the viscosity of the ionic liquid 1-butyl-3-methylimi-
dazolium tetrafluoroborate: viscosity and density relationships in ionic
liquids. J. Chem. Eng. Data 2007, 52, 2425–2430.
(41) Sanmamed, Y. A.; Gonzalez-Salagado, D.; Troncoso, J.; Cerdeirina,
C. A.; Romani, L. Viscosity-induced errors in the density determination
of room temperature ionic liquids using vibrating tube densitometry.
Fluid Phase Equilib. 2007, 252, 96–102.
(42) Casteel, J. F.; Amis, E. S. Specific conductance of concentrated
solutions of magnesium salts in water-ethanol system. J. Chem. Eng.
Data 1972, 17, 55–59.
(43) Onsager, L.; Fuoss, R. M. Irreversible processes in electrolytes.
Diffusion, conductance and viscous flow in arbitrary mixtures of strong
electrolytes. J. Phys. Chem. 1932, 36, 2689–2778.
(44) Barthel, J. M. G.; Krienke, H.; Kunz, W. Physical Chemistry of
Electrolyte Solutions - Modern Aspects; Steinkopf: Darmstadt, Springer:
New York, 1998.
(45) Kratky, O.; Leopold, H.; Stabinger, H. Density determination of liquids
and gases to an accuracy of 10-6 g/cm3, with a sample volume of
only 0.6 cm3. Z. Angew. Phys. 1969, 27, 273–277.
(46) Heintz, A.; Klasen, D.; Lehmann, J. K.; Wertz, C. Excess molar
volumes and liquid-liquid equilibria of the ionic liquid 1-methyl-3-
octyl-imidazolium tetrafluoroborate mixed with butan-1-ol and pentan-
1-ol. J. Solution Chem. 2005, 34, 1135–1144.
(47) Wang, J.; Tian, Y.; Zhao, Y.; Zhuo, K. A volumetric and viscosity
study for the mixtures of 1-n-butyl-3-methylimidazolium tetrafluo-
roborate ionic liquid with acetonitrile, dichloromethane, 2-butanone
and N, N - dimethylformamide. Green Chem. 2003, 5, 618–622.
(48) Van Valkenburg, M. E.; Vaughn, R. L.; Williams, M.; Wilkes, J. S.
Thermochemistry of ionic liquid heat-transfer fluids. Thermochim. Acta
2005, 425, 181–188.
(49) Stoppa, A.; Hunger, J. Hefter, G.; Buchner, R.,unpublished results.
(50) Hunger, J.; Stoppa, A.; Hefter, G.; Buchner, R. From ionic liquid to
electrolyte solution: dynamics of binary 1-N-butyl-3-N-methylimida-
zolium tetrafluoroborate + dichloromethane mixtures. J. Phys. Chem.
B 2008, 112, 12913–12919.
(51) Li, W.; Zhang, Z.; Zhang, J.; Han, B.; Wang, B.; Hou, M.; Xie, Y.
Micropolarity and aggregation behavior in ionic liquid + organic
solvent solutions. Fluid Phase Equilib. 2006, 248, 211–216.
(33) Villagra´n, C.; Deetlefs, M.; Pitner, W. R.; Hardacre, C. Quantification
of halide in ionic liquids using ion chromatography. Anal. Chem. 2004,
76, 2118–2123.
(34) Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Griffin, S. T.; Rogers,
R. D. Traditional extractants in nontraditional solvents: groups 1 and
2 extraction by crown ethers in room-temperature ionic liquids. Ind.
Eng. Chem. Res. 2000, 39, 3596–3604.
(35) Swatloski, R. P.; Holbrey, J. D.; Rogers, R. D. Ionic liquids are not
always green: hydrolysis of 1-butyl-3-methylimidazolium hexafluo-
rophosphate. Green Chem. 2003, 5, 361–363.
Received for review June 26, 2008. Accepted September 19, 2008.
Financial support by the Deutsche Forschungsgemeinschaft in the
framework of Priority Program SPP 1191, “Ionic Liquids”, is gratefully
acknowledged.
JE800468H