Distribution of a monovalent anion in various ionic liquid/water biphasic systems: Relationship of the distribution ratio of picrate ions with the aqueous solubility of ionic liquids

Article abstract of DOI:10.1021/je400726y

The distribution of picrate anions in various ionic liquid (IL)/water biphasic systems was investigated at 298.2 K. The ILs were 1-butyl-3- methylimidazolium hexafluorophosphate, 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide, 1-hexyl-3-methylimidazolium hexafluorophosphate, 1-hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl) amide, 1-methyl-3-octylimidazolium tetrafluoroborate, 1-methyl-3- octylimidazolium hexafluorophosphate, 1-methyl-3-octylimidazolium bis(trifluoromethanesulfonyl)amide, 1-methyl-3-octylimidazolium bis(pentafluoroethanesulfonyl)amide, 1-butyl-2,3-dimethylimidazolium bis(trifluoromethanesulfonyl)amide, 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)amide, methyltrioctylammonium bis(trifluoromethanesulfonyl)amide, 1-butylimidazolium bis(trifluoromethanesulfonyl)amide, and 1-butylpyrrolidinium bis(trifluoromethanesulfonyl)amide. The distribution ratios in dilute conditions (D) and the aqueous solubilities of the ILs (square root of the solubility product, K_sp^1/2) were determined. The extractability of the picrate anion generally increases with increasing hydrophobicity of the IL cation (C⁺) and increasing hydrophilicity of the IL anion (A ^-). For the ILs with different C⁺ but the same A ^-, the log D vs log K_sp^1/2 plot generally gives a linear relationship with a slope of -1; when the ILs have similar K _sp^1/2 values, the D value decreases in the C⁺ order, protic cations ?1,3-dialkylimidazolium cations > other cations. For the ILs comprising different A^- but the same C⁺, the log D versus log K_sp^1/2 plot is close to a linear line with a slope of 1. These regularities can be explained on the basis of the extraction mechanism including both the ion pair extraction with C⁺ and the ion exchange with A^-.

Full text of DOI:10.1021/je400726y

Article
pubs.acs.org/jced
Distribution of a Monovalent Anion in Various Ionic Liquid/Water
Biphasic Systems: Relationship of the Distribution Ratio of Picrate
Ions with the Aqueous Solubility of Ionic Liquids
Yuta Watanabe and Shoichi Katsuta*
Department of Chemistry, Graduate School of Science, Chiba University, 1-33 Yayoi-cho, Inage, Chiba 263-8522, Japan
ABSTRACT: The distribution of picrate anions in various
ionic liquid (IL)/water biphasic systems was investigated at
298.2 K. The ILs were 1-butyl-3-methylimidazolium hexa-
fluorophosphate, 1-butyl-3-methylimidazolium bis-
(trifluoromethanesulfonyl)amide, 1-hexyl-3-methylimidazo-
lium hexafluorophosphate, 1-hexyl-3-methylimidazolium bis-
(trifluoromethanesulfonyl)amide, 1-methyl-3-
octylimidazolium tetrafluoroborate, 1-methyl-3-octylimidazo-
lium hexafluorophosphate, 1-methyl-3-octylimidazolium bis-
(trifluoromethanesulfonyl)amide, 1-methyl-3-octylimidazo-
lium bis(pentafluoroethanesulfonyl)amide, 1-butyl-2,3-dime-
thylimidazolium bis(trifluoromethanesulfonyl)amide, 1-butyl-
1-methylpyrrolidinium bis(trifluoromethanesulfonyl)amide, methyltrioctylammonium bis(trifluoromethanesulfonyl)amide, 1-
butylimidazolium bis(trifluoromethanesulfonyl)amide, and 1-butylpyrrolidinium bis(trifluoromethanesulfonyl)amide. The
distribution ratios in dilute conditions (D°) and the aqueous solubilities of the ILs (square root of the solubility product,
Ksp1/2) were determined. The extractability of the picrate anion generally increases with increasing hydrophobicity of the IL
cation (C+) and increasing hydrophilicity of the IL anion (A−). For the ILs with different C+ but the same A−, the log D° vs log
Ksp1/2 plot generally gives a linear relationship with a slope of −1; when the ILs have similar Ksp1/2 values, the D° value decreases
in the C+ order, protic cations ≫1,3-dialkylimidazolium cations > other cations. For the ILs comprising different A− but the same
C+, the log D° versus log Ksp1/2 plot is close to a linear line with a slope of 1. These regularities can be explained on the basis of
the extraction mechanism including both the ion pair extraction with C+ and the ion exchange with A−.
In a recent study,11,12 we proposed an equilibrium model
including both ion pair extraction and ion exchange for the
extraction of various phenolate anions with ILs. In this model,
the target anion in the aqueous phase is extracted together with
INTRODUCTION
■
Liquid−liquid extraction is a simple and convenient method for
recovering, removing, or concentrating organic compounds in
aqueous solutions. However, the use of organic solvents which
are generally volatile, flammable, and harmful has been a
the IL cation in the aqueous phase and also extracted by
exchange with the IL anion in the IL phase. We derived
significant problem of this method. Recently, ionic liquids (ILs)
theoretical equations expressing the distribution ratio (D) of
have attracted increasing attention as replacements for
the target anion as a function of the aqueous solubility product
conventional organic solvents because they are almost non-
(Ksp) of the IL. The dependence of D on the amount of the
extracted anion was successively explained based on the
equations. The dependence of D on the kind of IL was,
volatile and hence nonflammable and harmless.1−4 In addition,
ILs have a characteristic property that they can extract ionic
species without the aid of additional counterions such as ion
pairing reagents.5−9 The mechanism of the ion extraction with
ILs is generally interpreted in terms of ion exchange. For
example, when the target of extraction is an anion, it is
extracted from an aqueous phase to an IL phase by exchange
with the component anion of the IL.5,6,9 On the basis of this
model, the IL composed of a less hydrophobic anion is more
efficient for extracting the target anion. On the other hand, the
importance of hydrophobicity of the component cation of the
IL is also emphasized because an imidazolium-based IL having a
longer alkyl chain exhibits higher extraction ability in the anion
extraction.8,10 There has been no quantitative explanation for
such “solvent effects” of ILs on the anion extraction.
however, not clearly explained because of not sufficient number
of ILs examined.
In this study, we have investigated the distribution of picrate
(2,4,6-trinitrophenolate) anions in various IL/water biphasic
systems for the purpose of understanding the solvent effects of
ILs on the anion extraction. The reason for using picrate as the
distributing anion is that this is chemically stable, is hard to be
protonated, and has both high extractability and high visible-
light absorption which allow us to easily measure the
Received: August 9, 2013
Accepted: February 19, 2014
Published: March 3, 2014
© 2014 American Chemical Society
696
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701

Journal of Chemical & Engineering Data
Article
Figure 1. Structural formulas of the component ions of ILs and their abbreviations.
distribution ratio. We chose 13 kinds of hydrophobic ILs as
follows: 1-butyl-3-methylimidazolium hexafluorophosphate
([BMIm][PF6]), 1-butyl-3-methylimidazolium bis-
(trifluoromethanesulfonyl)amide ([BMIm][NTf2]), 1-hexyl-3-
methylimidazolium hexafluorophosphate ([HMIm][PF6]), 1-
hexyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide
([HMIm][NTf2]), 1-methyl-3-octylimidazolium tetrafluorobo-
rate ([MOIm][BF4]), 1-methyl-3-octylimidazolium hexafluor-
ophosphate ([MOIm][PF6]), 1-methyl-3-octylimidazolium bis-
(trifluoromethanesulfonyl)amide ([MOIm][NTf2]), 1-methyl-
3-octylimidazolium bis(pentafluoroethanesulfonyl)amide
([MOIm][BETI]), 1-butyl-2,3-dimethylimidazolium bis-
(trifluoromethanesulfonyl)amide ([BM2Im][NTf2]), 1-butyl-1-
methylpyrrolidinium bis(trifluoromethanesulfonyl)amide
([BMPyr][NTf2 ]), methyltrioctylammonium bis-
(trifluoromethanesulfonyl)amide ([MTOA][NTf2]), 1-butyl-
imidazolium bis(trifluoromethanesulfonyl)amide ([BImH]-
[NTf2]), and 1-butylpyrrolidinium bis(trifluoromethane-
sulfonyl)amide ([BPyrH][NTf2]). The structural formulas of
the component ions are shown in Figure 1. The aqueous
solubilities of the ILs were also measured to determine the Ksp
values. The experimental data for [BMIm][PF6], [BMIm]-
[NTf2], [MOIm][BF4], [MOIm][NTf2], and [BMPyr][NTf2]
were cited from our previous paper.11
and w(Cl) < 6·10−6 for [MOIm][BETI]. Protic ILs, [BImH]-
[NTf2] and [BPyrH][NTf2], were prepared as follows. 1-
Butylimidazole (Sigma-Aldrich Co., St. Louis, MO; 98 %
purity) or 1-butylpyrrolidine (Sigma-Aldrich Co.; 98 % purity),
both of which were preliminarily purified by vacuum
distillation, was added slowly to a 1.2-fold molar excess of
HNTf2 (1,1,1-trifluoro-N-(trifluoromethylsulfonyl)-
methanesulfonamide; Kanto Chemical Co.; 99.0 % purity).
The mixture was washed 10 times with deionized water until
the pH of the aqueous phase became a constant value (yield, >
90 %). [HMIm][PF6], [MOIm][PF6], [BM2Im][NTf2], and
[MTOA][NTf2] were the same as used previously.13,14 Water
was distilled and further deionized with a Milli-Q Lab system
(Millipore, Billerica, MA). Dichloromethane (Kanto Chemical
Co.; guaranteed reagent grade) was purified by distillation.
Sodium picrate monohydrate (Kanto Chemical Co.; extra-pure
reagent grade) and other reagents (guaranteed reagent grade)
were used as received.
Solubility Measurements. The solubilities of [HMIm]-
[PF6], [HMIm][NTf2], [MOIm][PF6], [MOIm][BETI],
[BM2Im][NTf2], [BImH][NTf2], and [BPyrH][NTf2] were
determined by measuring the cation concentration in aqueous
solutions saturated with the ILs at (298.2 0.2) K as follows.
An aliquot of the IL-saturated aqueous solution was transferred
into a stoppered glass tube, to which sodium picrate and
sodium hydroxide were added so that their concentrations
became 5·10−2 mol·dm−3 and 1·10−2 mol·dm−3, respectively;
for the protic ILs, the pH of the aqueous picrate solution was
adjusted to 7.0 with 1·10−2 mol·dm−3 sodium dihydrogen
phosphate and 6·10−3 mol·dm−3 sodium hydroxide to suppress
deprotonation of the protic cations. Dichloromethane, whose
volume was equal to that of the aqueous solution, was further
added, and the biphasic mixture was stirred for 30 min. It was
preliminarily confirmed that the IL cation in the aqueous phase
was quantitatively extracted into dichloromethane together with
the picrate anion in a 1:1 molar ratio. The picrate concentration
in the dichloromethane phase was determined spectrophoto-
metrically (λmax = 367.0 nm, ε = 1.70·104 mol−1·dm3·cm−1 for
[HMIm]+;3 λmax = 367.8 nm, ε = 1.72·104 mol−1·dm3·cm−1 for
[MOIm]+;4 λmax = 375.0 nm, ε = (1.81 0.02)·104 mol−1·dm3·
cm−1 for [BM2Im]+; λmax = 348.6 nm, ε = (1.08 0.02)·104
EXPERIMENTAL SECTION
■
Materials. [HMIm][NTf2] was prepared by mixing equal
volumes of aqueous solutions of 4.3 mol·dm−3 [HMIm]Cl
(Tokyo Chemical Industry Co., Tokyo, Japan; > 98 % purity)
and 5.2 mol·dm−3 Li[NTf2] (Kanto Chemical Co., Tokyo,
Japan; > 99.7 % purity). The IL phase separated from the
aqueous phase was washed five times with deionized water. The
yield was almost quantitative (97 %). [MOIm][BETI] was
prepared in a similar manner from [MOIm]Cl (Wako Pure
Chemical Industries, Osaka, Japan; “for organic synthesis”
grade) and Li[BETI] (Kishida Chemical Co., Osaka, Japan; 99
% purity) in a yield of 95 %. The purities of the products were
checked by atomic absorption spectrophotometry for Li+ and
ion-selective potentiometry for Cl−; mass fractions w(Li) < 4·
10−7 and w(Cl) < 3·10−6 for [HMIm][NTf2]; w(Li) < 3·10−6
697
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701

Journal of Chemical & Engineering Data
Article
a
Table 1. Solubilities and Solubility Products of ILs in Water at 298.2 K and 0.1 MPa
solubility
solubility
Ksp
b
2
No.
1
IL
mole fraction
mol·dm−3
(mol·dm−3
5.42·10−3
)
c
c
c
[BMIm][PF6]
1.33·10−3
7.36·10−2
d
1.21·10−3
c
c
c
2
3
4
5
6
7
8
[BMIm][NTf2]
[HMIm][PF6]
[HMIm][NTf2]
[MOIm][BF4]
[MOIm][PF6]
[MOIm][NTf2]
[MOIm][BETI]
2.968·10−4
1.643·10−2
2.70·10−4
6.35·10−4
2.28·10−5
e
3.07·10−4
4.55·10−4 (3·10−6
)
)
2.52·10−2 (2·10−4
4.77·10−3 (2·10−5
)
d
4.34·10−4
8.62·10−5 (3·10−7
)
e
9.56·10−5
c
c
c
1.1·10−3
6.2·10−2
3.8·10−3
f
1.17·10−3
1.46·10−4 (1·10−6
)
8.09·10−3 (7·10−5
)
6.54·10−5
2.76·10−6
8.60·10−8
d
1.27·10−4
3.00·10−5
c
c
c
1.66·10−3
e
g
3.36·10−5 , 2.5·10−5
5.298·10−6 (8·10−9
)
2.932·10−4 (4·10−7
1.100·10−2 (4·10−5
)
)
g
2.0·10−5
9
[BM2Im][NTf2]
[BMPyr][NTf2]
1.987·10−4 (8·10−7
)
1.21·10−4
2.22·10−4
c
c
c
10
2.69·10−4
1.49·10−2
h
2.57·10−4
i
i
11
12
13
[MTOA][NTf2]
[BImH][NTf2]
[BPyrH][NTf2]
4.63·10−7
2.56·10−5
6.55·10−10
2.04·10−3
8.159·10−4 (6·10−7
)
4.519·10−2 (3·10−5
3.18·10−2 (1·10−4
)
5.74·10−4 (2·10−6
)
)
1.01·10−3
a
b
c
d
e
f
Values in parentheses are standard uncertainties. Calculated as the square of solubility. Reference 11. Reference 15. Reference 16. Reference 17
g
h
i
(295 K). Reference 18. Reference 19 (295 K). Reference 14.
mol−1·dm3·cm−1 for [BImH]+; λmax = 350.8 nm, ε = (1.601
0.006)·104 mol−1·dm3·cm−1 for [BPyrH]+). The solubility
values together with the literature values11,14−19 are summar-
ized in Table 1.
mass balance. The distribution ratio was calculated as D =
[Pic−]IL/[Pic−]W. The [Pic−]W and log D values obtained at
different [Pic−]W,Int conditions are listed in Table 2. The
extraction of Na+ was confirmed to be negligibly small by using
atomic absorption spectrophotometry.
Partition Experiments. An aqueous solution of sodium
picrate, whose concentration was (2.7·10−6 to 9.8·10−2) mol·
dm−3, was placed in a stoppered glass tube, together with a
water-saturated IL. Here, the volume of the viscous IL phase
was accurately evaluated from the mass by using the densities of
the ILs: 1.2937 g·cm−3 for [HMIm][PF6] (dry),20 1.370 g·cm−3
RESULTS AND DISCUSSION
■
Based on the theory previously reported,11 the following
equilibria can be considered regarding the extraction of picrate
anion (Pic−) from the aqueous phase to the IL phase.
for [HMIm][NTf2] (dry),21 (1.2323
0.0001) g·cm−3 for
Ion pair extraction:
[MOIm][PF6] (wet), (1.3966 0.0002) g·cm−3 for [MOIm]-
[BETI] (wet), 1.4159 g·cm−3 for [BM2Im][NTf2] (wet),
1.1113 g·cm−3 for [MTOA][NTf2] (wet),14 (1.4322
0.0002) g·cm−3 for [BImH][NTf2] (wet), and (1.3826
0.0001) g·cm−3 for [BPyrH][NTf2] (wet); the densities of
water-saturated [MOIm][PF6], [MOIm][BETI], [BM2Im]-
[NTf2], [BImH][NTf2], and [BPyrH][NTf2] were determined
in this study by triplicate measurements for each IL at (298.2
0.2) K with a DMA35n oscillating U-tube density meter (Anton
Paar, Graz, Austria) calibrated with pure water. The volume
ratio of the IL phase to the aqueous phase was adjusted to 1/
10, where the volume change of the IL phase upon its
dissolution into the aqueous phase was considered based on the
solubility data (Table 1). The biphasic mixture in the glass tube
was equilibrated by stirring with a magnetic stirrer for 1 h in a
Pic− + C+ ⇄ Pic− + C+
(1)
W
W
IL
IL
Ion exchange extraction:
−
−
Pic− + A ⇄ Pic− + A
(2)
W
IL
IL
W
Here, C+ and A− denote the IL component cation and anion,
respectively; subscripts “W” and “IL” represent the aqueous and
IL phases, respectively. In addition, there exists the dissolution
equilibrium of IL in the aqueous phase.
−
−
C+ + A ⇄ C+ + A
(3)
IL
IL
W
W
The equilibrium constants for eq 1 (ion pair extraction
constant, Kex,IP), eq 2 (ion exchange extraction constant,
K
ex,IE), and eq 3 (solubility product, Ksp) are defined as
Kex,IP = [Pic−]IL /[Pic−]W [C+]W
water bath thermostatted at (298.2
0.2) K. After phase
(4)
(5)
separation by centrifugation, the glass tube was allowed to stand
for 15 min more in the thermostatted water bath. The picrate
concentration in the aqueous phase ([Pic−]W) was determined
spectrophotometrically (λmax = 356.0 nm, ε = 1.45·104 cm−1·
mol−1·dm3). The picrate concentration in the IL phase
([Pic−]IL) was calculated from [Pic−]W and the initial aqueous
concentration of sodium picrate ([Pic−]W,Init) based on the
−
Kex,IE = [Pic−]IL[A ]W /[Pic−]W
−
Ksp = [C+]W [A ]W
(6)
Among these equilibrium constants, the following relationship
holds.
698
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701

Journal of Chemical & Engineering Data
Article
−
2
[A ]W = {Δ[Pic−]W + (Δ[Pic−]W + 4Ksp)1/2}/2
Table 2. Distribution Ratios and Aqueous Equilibrium
(10)
Concentrations of Picrate Anion at Different Initial
On the other hand, from eqs 4 and 5, the log D value of Pic−
can be expressed as follows.
a
Concentrations of Sodium Picrate at 298.2 K
[Pic−]W,Init/mmol·dm−3
IL = [HMIm][PF6]
[Pic−]W/mmol·dm−3
log D
log D = log Kex,IP + log[C+]W
(11)
2.11
0.114
2.238
2.243
2.238
−
log D = log Kex,IE − log[A ]W
Substituting [C+]W and [A−]W in eqs 11 and 12 by eqs 9 and
10, the following equations are obtained.
(12)
1.06
0.0564
0.0287
0.529
IL = [HMIm][NTf2]
0.211
0.0578
0.0286
0.0144
1.423
1.426
1.426
log D = log Kex,IP + log{−Δ[Pic−]W
0.106
0.0528
+ (Δ[Pic−]W + 4Ksp)1/2}/2
2
IL = [MOIm][PF6]
(13)
2.11
0.0374
0.0173
0.00908
2.747
2.781
2.760
log D = log Kex,IE − log{Δ[Pic−]W
1.06
0.529
+ (Δ[Pic−]W + 4Ksp)1/2}/2
2
(14)
IL = [MOIm][BETI]
0.0106
0.00644
0.00332
0.00169
0.814
0.778
0.754
These equations are essentially equal to each other, and it has
already been confirmed that they hold for the distribution of
Pic− in some nonprotic IL/water biphasic systems.11 In this
study, the applicability of the equations to the distribution of
Pic− in protic IL/water biphasic systems was examined. In
Figure 2, the log D value of Pic− between [BImH][NTf2] and
0.00531
0.00266
IL = [BM2Im][NTf2]
1.13
0.768
0.386
0.192
0.662
0.663
0.669
0.564
0.282
IL = [MTOA][NTf2]
2.11
0.431
1.592
1.988
2.225
1.06
0.0982
0.0296
0.528
IL = [BImH][NTf2]
98.3
6.29
1.465
1.628
1.746
1.796
1.809
1.820
1.817
1.821
49.0
2.26
19.6
0.697
0.249
0.120
0.0592
0.0294
0.0146
7.85
3.89
1.97
0.971
0.486
IL = [BPyrH][NTf2]
2.12
1.06
0.531
0.422
0.211
0.105
1.601
1.608
1.606
Figure 2. Distribution ratio of picrate anion in the [BImH][NTf2]/
water system as a function of the difference between the initial and
equilibrium concentrations of picrate in the aqueous phase. The solid
line is the regression curve based on eq 13 or 14.
a
The volume ratio of the IL phase to the aqueous phase is 1/10.
K
ex,IE/Kex,IP = Ksp
(7)
water is shown as a function of Δ[Pic−]W. A regression curve
shown in the figure is generated based on eq 13 or 14 by using
the Ksp value of [BImH][NTf2] (Table 1). A good agreement
between the experimental results and the regression curve is
found (correlation coefficient r = 0.996), proving that the
distribution in the protic IL/water system is also explained
based on the above model.
According to eq 13 or 14, the log D value becomes constant
when Δ[Pic−]W ≪ 2 Ksp1/2 and the log D value in such a dilute
condition (log D°) is expressed as follows.
The Ksp value can be determined as the square of the solubility,
assuming the ion activity coefficients to be unity and neglecting
any ion association. The values are shown in Table 1.
The aqueous concentrations of C+ and A− should be equal to
each other when no extraction occurs. Upon the extraction of
Pic−, the aqueous concentration of C+ decreases (eq 1) and
that of A− increases (eq 2). The difference between the
aqueous concentrations of A− and C+ corresponds to the
decreased amount of the aqueous concentration of Pic−
(Δ[Pic−]W = [Pic−]W,Init − [Pic−]W):
1/2
log D° = log Kex,IP + log Ksp
(15)
[A ]W − [C+]W = Δ[Pic−]W
−
(8)
1/2
log D° = log Kex,IE − log Ksp
(16)
From eqs 6 and 8, the following equations expressing the
aqueous concentrations of C+ and A− are derived.
From the data in Table 2, the Kex,IP and Kex,IE values were
determined by using eqs 13 and 14, respectively. The D° values
were calculated from eq 15 or 16. The logarithmic values of
[C+]W = {−Δ[Pic−]W + (Δ[Pic−]W + 4Ksp)1/2}/2
2
(9)
Kex,IP, Kex,IE, and D° are summarized in Table 3.
699
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701

Journal of Chemical & Engineering Data
Article
Table 3. Distribution Ratios of Picrate Anion in Diluted
a
Condition and Extraction Equilibrium Constants at 298.2 K
log (Kex,IP/mol−1
·
log (K /mol·
ex,IE b
b
No.
IL
dm3)
dm−3
)
log D°
c
c
1
2
[BMIm][PF6]
2.709
0.443
1.576
0.85
c
c
[BMIm]
[NTf2]
2.64
−0.93
3
4
5
6
7
8
9
[HMIm]
[PF6]
3.848 (0.004)
0.651 (0.004)
2.250
1.4290
3.76
[HMIm]
[NTf2]
3.7501 (0.0005)
−0.8920 (0.0005)
c
,
d
c,d
[MOIm]
[BF4]
4.97
2.55
[MOIm]
[PF6]
4.89 (0.01)
0.70 (0.01)
2.80
c
c
[MOIm]
[NTf2]
4.73
−0.83
1.95
[MOIm]
[BETI]
4.32 (0.02)
−2.75 (0.02)
0.78
1/2
Figure 3. Relationship between log D° and log Ksp for [NTf2]−-
based ILs (filled circles) and [PF6]−-based ILs (open circles) with
different component cations. The numbers of ILs correspond to those
in Tables 1 and 3. The broken lines are the regression lines with a
slope of −1 for the IL groups [1, 3, 6], [2, 4, 7], and [9, 10, 11].
[BM2Im]
[NTf2]
2.627 (0.001)
−1.290 (0.001)
0.669
0.54
c
,
d
c,d
10 [BMPyr]
[NTf2]
2.37
−1.29
11 [MTOA]
[NTf2]
8.09 (0.05)
−1.10 (0.05)
0.488 (0.004)
0.114 (0.002)
3.50
12 [BImH]
[NTf2]
3.179 (0.004)
3.110 (0.002)
1.834
1.612
protic cations (no. 12 and 13) show large upward deviations
from the correlation lines for the other [NTf2]−-based ILs.
These differences with the type of C+ correspond to the
magnitude of the Kex,IE value: the Kex,IE value is much larger for
the protic cations and slightly larger for the 1,3-dialkylimida-
zolium cations than for [BM2Im]+, [BMPyr]+, and [MTOA]+.
The variation of the ion exchange extractability with the type of
C+ can be explained in terms of “hydrogen-bond donating
ability” or “Lewis acidity” of the IL cations. The protic cation
has a positively charged hydrogen atom which should have
strong hydrogen bonding with the picrate anion. It is also
known that the 1,3-dialkylimidazolium cation has a stronger
interaction with the picrate anion through the C2−H atom, as
compared to [BM2Im]+ and tetraalkylammonium cations.13
Because the cation−anion interaction is stronger for the picrate
anion than for the [NTf2]− anion,13 the ion exchange
extractability of the picrate anion with the [NTf2]-based IL
increases with an increase of the hydrogen-bond donating
ability of the IL cation.
13 [BPyrH]
[NTf2]
a
b
1/2
In the condition that Δ[Pic−]W ≪ 2 Ksp
are standard uncertainties. Reference 11. Recalculated in ref 12.
.
d
Values in parentheses
c
In the ion pair extraction (eq 1), C+ is directly concerned
with the reaction as a reactant, whereas A− is a component of
the solvent. This fact is reflected in the Kex,IP values: the Kex,IP
value for the [NTf2]−-based or [PF6]−-based IL depends greatly
on the kind of C+, whereas the values for the ILs having the
same C+ (e.g., [MOIm]+) are generally comparable to each
other regardless of the kind of A−. In the ion exchange reaction
(eq 2), the reverse is true: A− is one of the reactants, whereas
C+ is a component of the solvent. Indeed, the Kex,IE values
depend largely on the kind of A− but not on the kind of C+.
The Kex,IP value for a given A− decreases in the following C+
order: [MTOA]+ > [MOIm]+ > [HMIm]+ > [BImH]+
≈
[BPyrH]+ > [BMIm]+ ≈ [BM2Im]+ > [BMPyr]+. It appears
that the ion pair extractability is generally greater with a more
hydrophobic C+, although there are some exceptional cases
such as the protic cations and [BM2Im]+. On the other hand,
the Kex,IE value for a given cation varies with A− in the order,
[BF4]− > [PF6]− > [NTf2]− > [BETI]−. The ion exchange
extractability is greater with a more hydrophilic A−.
From eq 15, the A− dependence of the D° value is expected
to be governed by that of the Ksp value because the Kex,IP
1/2
value is insensitive to the kind of A−. In Figure 4, the plot of log
D° versus log Ksp1/2 is shown for the [MOIm]+-based ILs with
different A−. A linear relationship with a slope of about 1 (1.3)
is observed as expected from eq 15. The slope higher than 1
comes from the slight variation of the Kex,IP value with A−, i.e.,
[BF4]− ≥ [PF6]− ≥ [NTf2]− > [BETI]−. This sequence seems
to be consistent with that of the Lewis acidity of the ILs; the
values of the Lewis acidity parameter, ETN, are available for the
ILs except for [MOIm][BETI], that is, 0.65, 0.633, and 0.627
for [MOIm][BF4], [MOIm][PF6], and [MOIm][NTf2],
respectively.22
From the relation of eq 16 and the fact that the Kex,IE value is
insensitive to the kind of C+, it is expected that the C+
1/2
dependence of the D° value is governed by that of the Ksp
value, i.e., the aqueous solubility of the IL. In Figure 3, the plots
of log D° versus log Ksp are shown for the [NTf2]−- and
1/2
[PF6]−-based ILs with different C+. A linear relation with a
slope of −1 is expected from eq 16 and actually observed for
each of the following IL groups: the [PF6]−-based ILs with the
1,3-dialkylimidazolium cations (no. 1, 3, and 6), the [NTf2]−-
based ILs with the 1,3-dialkylimidazolium cations (no. 2, 4, and
7), and the [NTf2]−-based ILs with [BM2Im]+, [BMPyr]+, and
[MTOA]+ (no. 9, 10, and 11). In the case of the [NTf2]−-based
ILs, the line for the 1,3-dialkylimidazolium cations is slightly
higher than that for [BM2Im]+, [BMPyr]+, and [MTOA]+. In
addition, the log D° values for the [NTf2]−-based ILs with the
CONCLUSION
■
The solvent effects of ILs on the distribution of an anion in
hydrophobic IL/water biphasic systems were studied by using
picrate as the distributing anion. The extractability of the anion
generally increases with increasing hydrophobicity of the IL
cation and increasing hydrophilicity of the IL anion. The
dependence of the distribution ratio on the kind of IL cation
700
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701

Journal of Chemical & Engineering Data
Article
and Extraction-Voltammetric Determination of Phenols Using Room
Temperature Ionic Liquid. Anal. Bioanal. Chem. 2005, 381, 464−470.
(7) Vijayaraghavan, R.; Vedaraman, N.; Surianarayanan, M.;
MacFarlane, D. R. Extraction and Recovery of Azo Dyes into an
Ionic Liquid. Talanta 2006, 69, 1059−1062.
(8) Pei, Y. C.; Wang, J. J.; Xuan, X. P.; Fan, J.; Fan, M. Factors
Affecting Ionic Liquids Based Removal of Anionic Dyes from Water.
Environ. Sci. Technol. 2007, 41, 5090−5095.
(9) Li, C.; Xin, B.; Xu, W.; Zhang, Q. Study on the Extraction of
Dyes into a Room-Temperature Ionic Liquid and Their Mechanisms.
J. Chem. Technol. Biotechnol. 2007, 82, 196−204.
(10) Katsuta, S.; Nakamura, K.; Kudo, Y.; Takeda, Y.; Kato, H.
Partition Behavior of Chlorophenols and Nitrophenols between
Hydrophobic Ionic Liquids and Water. J. Chem. Eng. Data 2011, 56,
4083−4089.
(11) Katsuta, S.; Nakamura, K.; Kudo, Y.; Takeda, Y. Mechanisms
and Rules of Anion Partition into Ionic Liquids: Phenolate Ions in
Ionic Liquid/Water Biphasic Systems. J. Phys. Chem. B 2012, 116,
852−859.
(12) Katsuta, S. Distribution Behavior of Neutral and Anionic
Compounds in Ionic Liquid/Water Biphasic Systems. Bunseki Kagaku
2013, 62, 297−315 (in Japanese).
(13) Katsuta, S.; Imai, K.; Kudo, Y.; Takeda, Y.; Seki, H.; Nakakoshi,
M. Ion Pair Formation of Alkylimidazolium Ionic Liquids in
Dichloromethane. J. Chem. Eng. Data 2008, 53, 1528−1532.
(14) Katsuta, S.; Yoshimoto, Y.; Okai, M.; Takeda, Y.; Bessho, K.
Selective Extraction of Palladium and Platinum from Hydrochloric
Acid Solutions by Trioctylammonium-Based Mixed Ionic Liquids. Ind.
Eng. Chem. Res. 2011, 50, 12735−12740.
(15) Freire, M. G.; Neves, C. M. S. S.; Carvalho, P. J.; Gardas, R. L.;
Fernandes, A. M.; Marrucho, I. M.; Santos, L. M. N. B. F.; Coutinho, J.
A. P. Mutual Solubilities of Water and Hydrophobic Ionic Liquids. J.
Phys. Chem. B 2007, 111, 13082−13089.
(16) Freire, M. G.; Carvalho, P. J.; Gardas, R. L.; Marrucho, I. M.;
Santos, L. M. N. B. F.; Coutinho, J. A. P. Mutual Solubilities of Water
and the [Cnmim][Tf2N] Hydrophobic Ionic Liquids. J. Phys. Chem. B
2008, 112, 1604−1610.
1/2
Figure 4. Relationship between log D° and log Ksp for [MOIm]+-
based ILs with different component anions. The numbers of ILs
correspond to those in Tables 1 and 3. The broken line is the
regression line with a slope of 1.
and anion can be quantitatively explained as a function of the
solubility of the IL (Ksp1/2 value). This explanation is based on
the model where both the ion pair extraction and the anion
exchange extraction proceed. Protic ILs and 1,3-dialkylimida-
zolium-based ILs are favorable to the anion extraction because
the component cations, particularly protic cations, have a
strong interaction with the target anion. The regularities found
in this study will be utilized for the prediction of the ability of
an IL to extract an anion and also for the design of effective
extraction systems with ILs.
(17) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. Solution
Thermodynamics of Imidazolium-Based Ionic Liquids and Water. J.
Phys. Chem. B 2001, 105, 10942−10949.
(18) Kakiuchi, T.; Tsujioka, N.; Kurita, S.; Iwami, Y. Phase-Boundary
Potential across the Nonpolarized Interface between the Room-
Temperature Molten Salt and Water. Electrochem. Commun. 2003, 5,
159−164.
(19) Alfassi, Z. B.; Huie, R. E.; Milman, B. L.; Neta, P. Electrospray
Ionization Mass Spectrometry of Ionic Liquids and Determination of
Their Solubility in Water. Anal. Bioanal. Chem. 2003, 377, 159−164.
(20) Pereiro, A. B.; Rodríguez, A. Experimental Liquid−Liquid
Equilibria of 1-Alkyl-3-methylimidazolium Hexafluorophosphate with
1-Alcohols. J. Chem. Eng. Data 2007, 52, 1408−1412.
(21) Kato, R.; Gmehling, J. Systems with Ionic Liquids: Measure-
ment of VLE and γ∞ Data and Prediction of Their Thermodynamic
Behavior Using Original UNIFAC, mod. UNIFAC(Do), and
COSMO-RS(Ol). J. Chem. Thermodynamics 2005, 37, 603−619.
(22) Jessop, P. G.; Jessop, D. A; Fu, D.; Phan, L. Solvatochromic
Parameters for Solvents of Interest in Green Chemistry. Green Chem.
2012, 14, 1245−1259.
AUTHOR INFORMATION
Corresponding Author
*E-mail: katsuta@faculty.chiba-u.jp.
Funding
This work was partially supported by a research grant from the
Futaba Electronics Memorial Foundation.
Notes
■
The authors declare no competing financial interest.
REFERENCES
■
(1) Huddleston, J. G.; Willauer, H. D.; Swatloski, R. P.; Visser, A. E.;
Rogers, R. D. Room Temperature Ionic Liquids as Novel Media for
“Clean” Liquid−Liquid Extraction. Chem. Commun. 1998, 1765−1766.
(2) Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Willauer, H. D.;
Huddleston, J. G.; Rogers, R. D. Room Temperature Ionic Liquids as
Replacements for Traditional Organic Solvents and Their Applications
towards “Green Chemistry” in Separation Processes. In Green
Industrial Applications of Ionic Liquids; Rogers, R. D., Seddon, K. R.,
Volkov, S., Eds.; Kluwer Academic Publishers: Dordrecht, The
Netherlands, 2003; pp 137−156.
(3) Han, X.; Armstrong, D. W. Ionic Liquids in Separation. Acc.
Chem. Res. 2007, 40, 1079−1086.
(4) Oppermann, S.; Stein, F.; Kragl, U. Ionic Liquids for Two-Phase
Systems and Their Application for Purification, Extraction and
Biocatalysis. Appl. Microbiol. Biotechnol. 2011, 89, 493−499.
(5) Carda-Broch, S.; Berthod, A.; Armstrong, D. W. Solvent
Properties of the 1-Butyl-3-methylimidazolium Hexafluorophosphate
Ionic Liquid. Anal. Bioanal. Chem. 2003, 375, 191−199.
(6) Khachatryan, K. S.; Smirnova, S. V.; Torocheshnikova, I. I.;
Shvedene, N. V.; Formanovsky, A. A.; Pletnev, I. V. Solvent Extraction
701
dx.doi.org/10.1021/je400726y | J. Chem. Eng. Data 2014, 59, 696−701