Extraction of cesium ions from aqueous solutions using calix[4]arene-bis(tert.octylbenzo-crown-6) in ionic liquids

Article abstract of DOI:10.1021/ac049949k

Solvent extraction of cesium ions from aqueous solution to hydrophobic ionic liquids without the introduction of an organophilic anion in the aqueous phase was demonstrated using calix[4]arene-bis(tert-octylbenzo-crown-6) (BOBCalixC6) as an extractant. The selectivity of this extraction process toward cesium ions and the use of a sacrificial cation exchanger (NaBPh ₄) to control loss of imidazolium cation to the aqueous solutions by ion exchange have been investigated.

Full text of DOI:10.1021/ac049949k

Anal. Chem. 2004, 76, 3078-3083
Ex t ra c t io n o f Ce s iu m Io n s fro m Aq u e o u s S o lu t io n s
Us in g Ca lix [4 ]a re n e -b is (t e rt -o c t ylb e n zo -c ro w n -6 ) in
Io n ic Liq u id s
†
,‡
‡
‡
§
Huim in Luo, She ng Da i,* Pe te r V. Bonne s e n, A. C. Buc ha na n, III, J ohn D. Holbre y,
§
,§
Nic hola s J . Bridge s , a nd Robin D. Roge rs *
Nuclear Science and Technology Division and Chemical Sciences Division, Oak Ridge National Laboratory,
Oak Ridge, Tennessee 37831, and Department of Chemistry and Center for Green Manufacturing, The University of
Alabama, Tuscaloosa, Alabama 35487
Solvent extraction of cesium ions from aqueous solution
to hydrophobic ionic liquids without the introduction of
an organophilic anion in the aqueous phase was demon-
strated using calix[4 ]arene-bis(ter t-octylbenzo-crown-6 )
(
BOBCalixC6 ) as an extractant. The selectivity of this
extraction process toward cesium ions and the use of a
sacrificial cation exchanger (NaBP h ) to control loss of
4
Fig u re 1 . Strctures of two most common ionic liquid cations.
imidazolium cation to the aqueous solutions by ion
exchange have been investigated.
organic vapors,6 and synthesizing novel materials.7,8 Several
reviews and monographs have recently appeared, providing
Ionic liquids (ILs) are attracting increased attention worldwide
because of the perceived environmental benefits.1 Ionic liquids
have been defined as salts which melt below 100 °C, and many
are known that melt below room temperature. The most com-
monly studied classes of ILs (Figure 1) are cationic N-alkylated
nitrogen heterocycles, such as N,N-dialkylimidazolium or N-
alkylpyridinium ions.1 These organic cations, which are relatively
large and asymmetric, as compared to simple inorganic cations,
general overviews of the subject.1-3
-3
In contrast to higher-temperature inorganic molten salts,
9
examples of ILs are known that are hydrophobic and, thus, form
a biphase with aqueous phases yet still retain ionic characteristics.1
These novel dual properties of hydrophobicity and high ionic
character have led to the investigation of ILs as unique separation
media for IL/ aqueous solvent extractions.10-14 Large distribution
-4
M
coefficients (D ) for the extraction of metal ions from aqueous
account for the low melting points of the salts. A variety of anions
solutions to ILs containing complexing extractants have been
observed. For example, whereas conventional solvent extraction
-
can be utilized to form ILs, including [BF
or other complex anions.1,2
4
]-, [PF
6
]-, [CF
3
SO
3
] ,
(
(
5) Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2 0 0 2 , 106,
315.
6) (a) Baker, G. A.; Baker, S. N.; McCleskey, T. M. Chem. Commun. 2 0 0 3 ,
2932. (b) Liang, C. D.; Yuan, C. Y.; Warmack, R. J.; Barnes, C. E.; Dai, S.
Anal. Chem. 2 0 0 2 , 74, 2172.
7) Nakashima, T.; Kimizuka, N. J. Am. Chem. Soc. 2 0 0 3 , 125, 6386.
8) Dai, S.; Ju, Y. H.; Gao, H. J.; Lin, J. S.; Pennycook, S. J.; Barnes, C. E. Chem.
Comm. 2 0 0 0 , 243.
Many of these ILs, unlike many conventional molecular
solvents currently in use, are nonflammable, chemically tunable,
and exert no detectable vapor pressure. These unique features
have led to their designation as “designer solvents” for use as
potential replacements for noxious volatile organic compounds
7
(
(
(
VOCs), which can contribute to air pollution and health problems
for process workers.1 The applications of ILs as solvents for
various catalytic reactions have been extensively explored,1-6 as
(9) Bonhote, P.; Dias, A.-P.; Papageorgiou, N.; Kalyanasundaram, K.; Gratzel,
M. Inorg. Chem. 1 9 9 6 , 35, 1168.
(
10) (a) Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Mayton, R.; Sheff, S.;
Wierzbicki, A.; Davis, J. H.; Rogers, R. D. Chem. Commun. 2 0 0 1 , 135. (b)
Visser, A. E.; Swatloski, R. P.; Reichert, W. M.; Griffin, S. T.; Rogers, R. D.
Ind. Eng. Chem. Res. 2 0 0 0 , 39, 3596. (c) Huddleston, J. G.; Willauer, H.
D.; Swatloski, R. P.; Visser, A. E.; Rogers, R. D. Chem. Commun. 1 9 9 8 ,
1765.
have ILs for membrane-based separation of CO
2
,4,5 sensing volatile
†
Nuclear Science and Technology Division, Oak Ridge National Laboratory.
Chemical Sciences Division, Oak Ridge National Laboratory.
The University of Alabama.
‡
§
(
(
(
(
1) (a) Ionic Liquids in Synthesis; Wasserscheid, P., Welton, T. Eds.; Wiley-
VCH: Weinheim, 2003. (b) Wasserscheid, P.; Keim, W. Angew Chem. Int.
Ed. 2 0 0 0 , 39, 3772. (c) Welton, T. Chem. Rev., 1 9 9 9 , 99, 2071.
2) Ionic Liquids: Industrial Applications to Green Chemistry; Rogers, R. D.,
Seddon, K. R. Eds.; ACS Symposium Series 818; American Chemical
Society: Washington, DC, 2002.
3) Ionic Liquids as Green Solvents: Progress and Prospects; Rogers, R. D., K. R.
Seddon, Eds.; ACS Symposium Series 856; American Chemical Society:
Washington, DC, 2003.
(11) Chun, S.; Dzyuba, S. V.; Bartsch, R. A. Anal. Chem. 2 0 0 1 , 73, 3737.
(12) Dietz, M. L.; Dzielawa, J. A. Chem. Commun. 2 0 0 1 , 2124.
(13) (a) Carda-Broch, S.; Berthod, A.; Armstrong, D. W. Anal. Bioanal. Chem.
2 0 0 3 , 375, 191. (b) Armstrong, D. W.; He, L. F.; Liu, Y. S. Anal. Chem.
1 9 9 9 , 71, 3873.
(14) (a) Makote, R.; Luo, H.; Dai, S. in Ultra Clean Solvents, Abraham, M. A.;
Moens, L. Eds.; ACS Symposium Series 819; American Chemical Society:
Washington, DC, 2002; p 26-33. (b) Dai, S.; Ju, Y. H.; Barnes, C. E. J. Chem.
Soc., Dalton Trans. 1 9 9 9 , 1201. (c) Dai, S.; Shin, Y.; Toth, L. M.; Barnes,
C. E. Inorg. Chem. 1 9 9 7 , 36, 4900. (d) Luo, H.; Dai, S.; Bonnesen, P. V.
Anal. Chem. 2 0 0 4 , 76, 2773-2779.
4) Bates, E. D.; Mayton, R. D.; Ntai, I.; Davis, J. H. J. Am. Chem. Soc. 2 0 0 2 ,
1
24, 926.
3078 Analytical Chemistry, Vol. 76, No. 11, June 1, 2004
10.1021/ac049949k CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/29/2004

of Sr2 using cis-dicyclohexano-18-crown-6 can deliver practical
+
D
M
values of <1, using ILs as extraction solvents delivered values
4
13
M
of D on the order of 10 . The enhanced distribution coefficients
can be attributed to the unique solvation properties of ILs for ionic
species, with synergistic ion-exchange processes playing an
important role.10b,12
Fig u re 2 . Structure of BOBCalixC6.
The solvation processes of these ionic complexes by conju-
gated ions in ILs are much more favorable than those in
conventional solvents.13 The successful extraction of Sr2+ by cis-
dicyclohexano-18-crown-6 also indicates that any interaction
between imidazolium cations of the ILs and the crown is limited
and weaker than the interaction between Sr2+ and the crown ether.
The use of calixarene crown ethers for cesium ion extraction
in IL media may greatly increase the cesium distribution ratio.
Thus, in this report, we extend our investigation of the solvent
extraction and ion exchange properties of ILs to the extraction of
cesium ions, utilizing the cation receptor BOBCalixC6. These
studies are directed primarily toward acquiring an understanding
of the mechanism of extraction in the complex IL solvents and
secondarily to determining if a Cs-removal technology utilizing
ILs is possible.
Imidazolium compounds have been known to interact with
dibenzo-18-crown-6 in the solid state.15,16
137Cesium is a major fission product in spent nuclear wastes.17-21
Its removal from these wastes is a key part of waste remediation
strategies.21 The selective removal of cesium ions from nuclear
wastes is essential to the safe and cost-effective production of
associated waste forms for superior postclosure performance in
EXPERIMENTAL SECTION
Materials and Methods. All chemicals and solvents were
reagent grade and were used without further purification unless
a repository. Cesium ion extraction using crown ethers has been
investigated by McDowell17 and Horwitz,18 as well as many
others.19 McDowell showed that the distribution ratio of cesium
noted otherwise. The ILs used in this work were 1-C
n
-3-meth-
-),
(n-hexyl), or
(n-octyl). These ILs were synthesized via metathesis reactions
+
ylimidazolium (C
where C ) C
n
mim ) bis[(trifluoromethyl)sulfonyl]imide (NTf
2
ion (DCs) from 0.1 M nitric acid solutions could be on the order
2
n
2
(ethyl), C
3
(n-propyl), C
4
(n-butyl), C
6
of 10 when the crown ether bis(tert-butylbenzo)-21-crown-7 was
C
8
used in combination with the organophilic anion didodecylnaph-
thalenesulfonic acid (in toluene solution).
as described in the literature.9,23 The details on synthesis and
characterization are given in the Supporting Information Section.
Aqueous solutions were prepared using deionized (DI) water with
a specific resistance of 18.0 MΩ-cm or greater. BOBCalixC6 was
obtained from IBC Advanced Technologies (American Fork, UT)
and was used as received (97% stated purity). H and 13C NMR
In the past decade, mono- and bis-crown-6 derivatives of calix-
[
4]arenes in the 1,3-alternate conformation have been shown to
possess both extremely high extractive strength for cesium and
4
excellent (generally exceeding 10 ) selectivity for cesium ion over
1
sodium ion.20 Without the aid of an organophilic anion, these
calixarene crown ethers are capable of extracting cesium from
both acidic and alkaline media with distribution ratios generally
exceeding unity, and sometimes as high as 100, depending on
3
spectra were obtained in CDCl with a Bruker Avance DRX 400
NMR spectrometer. Concentrations of Cs+ were determined via
a Dionex LC20 ion chromatograph equipped with an IonPac CS
1
2 analytical column. The solubility in water was determined by
+
the concentration of competitive cations (e.g., K ) in the aqueous
combining 0.5 mL of IL and 5 mL DI water and shaking for 60
min in a vibrating mixer. A 0.5-mL aliquot of the aqueous phase
was removed and diluted to 5 mL with DI water. The absorbance
at 211 nm was measured using a Varian UV-vis-NIR spectrom-
eter (model 5000). The solubility of each IL in water was calculated
by the comparison of the measured absorbance with that obtained
from dissolving a known amount (2-20 mg) of IL in 5 mL of DI
water. The water content of the ILs was measured using a
Metrohm 652 KF coulometer.
solution, the concentration of the calixarene crown, and the
polarity of the diluent. By comparison, under the same conditions,
the crown ether bis(tert-butylbenzo)-21-crown-7 gives DCs values
at least 2 orders of magnitude lower.20b A lipophilic derivative of
a calix[4]-bis-crown-6, calix[4]arene-bis(tert-octylbenzo-crown-6)
(
BOBCalixC6), as seen in Figure 2, is currently being investigated
for solvent extraction processes for removing cesium ion from
alkaline and acidic tank waste.21,22
Tracer studies were conducted at The University of Alabama
with the use of 137Cs purchased from Amersham (Arlington
Heights, IL) as the chloride salt and used as CsCl in 0.6 M HCl.
(
(
(
(
15) Kiviniemi, S.; Sillanp aj aj , A.; Nissinen, M.; Rissanen, K.; T a¨ ms a¨ , M. T.;
Pursiainen, J. Chem. Commun. 1 9 9 9 , 897.
16) Kiviniemi, S. Ph.D. Thesis, University of Oulu, Finland, 2001, ISBN 951-42-
5
996-3, Oulu University Press: Oulu, Finland, 2001.
17) McDowell, W. J.; Case, G. N.; McDonough, J. A.; Bartsch, R. A. Anal. Chem.
9 9 2 , 64, 3013.
An in-house 14C-labeled 1-butyl-3-methylimidazolium bromide ([C
mim]Br) was made through the reaction of 1-methylimidazole
Aldrich) with 14C 1-bromobutane (American Radiolabeled Chemi-
cals, St. Louis, MO), where only the â carbon is 14C-labeled. The
4
-
1
18) (a) Dietz, M. L.; Horwitz, E. P.; Rhoads, S.; Bartsch, R. A.; Krzykawski, J.
Solvent Extr. Ion Exch. 1 9 9 6 , 14, 1. (b) Dietz, M. L.; Horwitz, E. P.; Jensen,
M. P.; Rhoads, S.; Bartsch, R. A.; Palka, A.; Krzykawski, J.; Nam, J. Solvent
Extr. Ion Exch. 1 9 9 6 , 14, 357.
(
1
4
C-labeled [C
4
mim]Br was diluted in water and spiked into a
]. All gamma counting was conducted on
(
19) See, for example: (a) Gerow, I. H.; Davis, M. V. Sep. Sci. Technol. 1 9 7 9 ,
sample of [C mim][NTf
4
2
1
1
4, 395. (b) Gerow, I. H.; Smith, J. E.; Davis, M. W. Sep. Sci. Technol. 1 9 8 1 ,
6, 519. (c) Blasius, E.; Nilles, K.-H. Radiochim. Acta 1 9 8 4 , 35, 173. (d)
a Packard Cobra II automated gamma counter with a through-
bore 3-in. NaI(Tl) crystal, and beta counting was done on a
Schulz, W. W.; Bray, L. A. Sep. Sci. Technol. 1 9 8 7 , 22, 191.
(
(
20) (a) Dozol, J.-F., Casas, J.; Sastre, A. M. Sep. Sci. Technol. 1 9 9 5 , 30, 435.
(
b) Haverlock, T. J.; Bonnesen, P. V.; Sachleben, R. A.; Moyer, B. A.
(22) Leonard, R. A.; Conner, C.; Liberatore, M. W.; Sedlet, J.; Aase, S. B.;
Vandegrift, G. F.; Delmau, L. H.; Bonnesen, P. V.; Moyer, B. A. Sep. Sci.
Technol. 2 0 0 1 , 35, 743.
Radiochim. Acta 1 9 9 7 , 76, 103, and references cited.
21) Bonnesen, P. V.; Delmau, L. H.; Moyer, B. A.; Leonard, R. A. Solvent Extr.
Ion Exch. 2 0 0 0 , 18, 1079.
(23) Dzyuba, S. V.; Bartsch, R. A. ChemPhysChem 2 0 0 2 , 3, 161.
Analytical Chemistry, Vol. 76, No. 11, June 1, 2004 3079

Packard 1500 TR liquid scintillation counter with Packard Ultima
Gold scintillation cocktail.
Ta b le 1 . Dis t rib u t io n Ra t io s (DCs ) b e t w e e n t h e ILs a n d
.5 m M Cs NO3
2
Extraction Experiments. The extraction experiments con-
ducted at ORNL were performed in duplicate for each IL by
contacting 1 mL of IL containing various concentrations of
BOBCalixC6 with 10 mL of Cs+ aqueous solution (2.5 mM) for
n n 2
C in [C mim][NTf ]
[
BOBCalixC6] ethyl
propyl
butyl butylb hexyl octyl
a
c
c
CHCl c
(
mM)
C
2
C
3
C
4
C
4
C
6
C
8
3
0
0.084
1.61
4.68
15.3
nm
0.055
1.36
3.90
14.6
137
0.024 0.19 ud
ud
ud
6
0 min in a vibrating mixer. After centrifugation, the upper
1
.00
3.40
.71
3.6
0.77
3.67
0.95
2.92
0.56 0.45 ud
3.20 2.50 ud
aqueous phase was separated, and the concentration of Cs+ was
7
13.8
131
11.7
63.8
11.5
57.4 17.9
8.10 ud
0.034
determined by ion chromatography.
1
Radiotracer extraction experiments conducted at The Univer-
sity of Alabama were performed in duplicate, and errors between
duplicate runs were within 5%. The distributions were conducted
in glass vials with 0.5 mL of IL with variable BOBCalixC6
a
nm: not measured because concentration of BOBCalixC6 could
not be reached. b Work conducted using radiotracer technique. Al-
though there are differences between the data obtained via ion
chromatography and radiotracer, the general trends are identical. ud:
undetectable via ion chromatography.
c
3
concentrations and 5 mL of 2.5 mM CsNO . The solutions were
vortexed for 90 s, followed by centrifugation of 90 s, and were
repeated twice. Equal volumes of both phases were removed, and
radiochemical counting was conducted.
The use of UV spectra could not confirm the mass balance of
the imidazolium cation and the distribution of [C
4
mim] between
mim]Br was
the IL and aqueous phases. Thus, the 14C-labeled [C
4
used to confirm the UV data. Through the use of the radio-labeled
IL, the mass balance of the system was also confirmed.
RESULTS AND DISCUSSION
Extraction Results: The distribution of cesium at various
concentrations of BOBCalixC6 (Table 1 and Figure 3) in the ILs
used here are high. The only exception is for [C
2 2
mim][NTf ] at
high BOBCalixC6 concentration, for which the solubility of
-
2
BOBCalixC6 is limited to ∼10 M. This difference in solubility
of BOBCalixC6 can be attributed to the hydrophobic interaction
Fig u re 3 . Dependence of DCs on the concentration of BOBCalixC6
in the IL phase. The data labeled in the closed symbols are from
ORNL, and those in the open symbol are from UA.
(
or lack thereof) between the alkyl groups of the imidazolium
cations and the BOBCalixC6. The longer alkyl groups interact
more favorably with the hydrophobic BOBCalixC6, leading to
higher solubilities.
The distribution coefficients when using ion chromatography
for extraction of Cs+ are defined in eq 1 as
ties of the IL organic cations.10 The hydrophobicity is also inversely
related to the solubility of the corresponding 1-alkyl-3-methylimi-
+
dazolium cation ([C
n
mim] ) in the aqueous phase and, thus, its
+
(
C - C )
ion-exchange capability. Because [C
2
mim] is the most hydrophilic
i
f
volume of aqueous phase
volume of IL phase
DCs
)
(1)
cation studied here and is exchangeable with metal ions, the
distribution coefficient of Cs+ to [C
mim][NTf ] is the largest for
Cf
2
2
the ILs in the series tested. Accordingly, the selectivity of this
solvent is dominated by the hydrophobicity of the metal ions
extracted (i.e., their Hofmeister selectivity).24,25
i f
where C and C represent the initial and final concentrations of
Cs+ in the aqueous phase. The distribution ratio is calculated by
As seen in Table 1, the distribution coefficients (DCs) of Cs+
from water to ILs with BOBCalixC6 strongly depend on the
concentration of BOBCalixC6 in the ILs and increase with the
concentration of the extractant. This observation is very similar
to that noted in the extraction of Sr2 with cis-dicyclohexano-18-
crown-6 (DCH18C6) in ILs, indicating that the complexation of
Cs+ plays a key role in the partitioning processes.14
the above equation, since only the aqueous phase is measured
and a volume ratio is needed in the calculation to account for the
difference in volume between the two phases. Thus, a distribution
ratio for Cs+ greater than 1 (DCs > 1) represents an overall
+
preference of Cs+ to the IL phase. The values of D
M
were
measured in duplicate with uncertainty within 5%.
Radiotracer distribution ratios were calculated using eq 2.
From infinite dilution distribution studies as a function of nitric
[
[
CPM]IL
CPM]aq
3
acid concentrations in the aqueous phase, at [HNO ] e 0.1 M,
DCs
)
(2)
BOBCalixC6 shows virtually quantitative extraction of cesium. At
acid concentrations >0.1 M, an acid dependency occurs, lowering
the distribution ratio, as shown in Figure 4. A ligand dependency
Both phases were counted radiochemically using equal volumes.
3
from 1 M HNO of 1:1 was calculated as illustrated in Figure 5. It
As seen in Table 1, the DCs values to the ILs in the absence of
an extractant are very low. The small amount of Cs , which does
partition into the ILs, correlates with the observed hydrophobici-
(
(
24) Kavallieratos, K.; Moyer, B. A. Chem. Commun. 2 0 0 1 , 1620.
25) Lee, B.; Bao, L.-L.; Im, H.-J.; Dai, S.; Hagaman, E. W.; Lin, J. S.; Langmuir
2 0 0 3 , 19, 4246.
+
3080 Analytical Chemistry, Vol. 76, No. 11, June 1, 2004

Ta b le 2 . S e le c t ivit y o f [C4 m im ][NTf2 ] Ex t ra c t io n
P h a s e s Co n t a in in g BOBCa lix C6
[
BOBCalixC6]
mM)
aqueous
phasea
(
D
Cs
D
K
D
Na
D
Sr
DCs/ D
K
1
.06
Cs+ + K+ + Na+
3.60 0.239 ∼0
3.51 0.890
4.67 1.05
15.1
3.94
4.45
53.0
28.2
68.7
15.2
+
+
+
Cs + K
+
2+
Cs + K + Sr
∼0
Cs + K + Na+
+
+
371
228
576
572
289
7.00
8.08
8.39
37.7
48.1
43.1
∼0
7
.96
+
+
+
Cs + K
+
2+
Cs + K + Sr
∼0 ∼0
∼0
Cs + K + Na+
+
+
1
4.1
+
+
+
Cs + K
5.98
26.4
Cs + K + Sr2+ 1138
+
∼0
a
The initial concentration of each ion in aqueous phase was 0.77
mM.
Table 2 shows the competitive extraction results for Cs+ in
Fig u re 4 . Extraction of cesium at trace level concentrations (“infinite
dilution”) into [C4mim][NTf2] containing BOBCalixC6 at 7.71 mM as
a function of nitric acid concentration.
the presence of Na , K , and Sr2+. The distribution coefficients
+
+
for Na+ and Sr2+ using BOBCalixC6 in [C
mim][NTf ] were too
4
2
small to be detected by ion chromatography. This observation is
consistent with what has previously been observed for BOB-
CalixC6 and related calixarene-crown ethers in conventional
solvents.20b,22,27
Coextraction of K+ along with Cs+ has been observed,
however, using BOBCalixC6 in conventional solvents.21 The
concomitant extraction of K+ along with Cs+ is also observed in
the studies reported here (Table 2). The selectivity coefficients
determined under various conditions range from ∼3.94 to ∼68.7,
which are lower than the selectivities of ∼220 seen in organic
solvents, such as 1,2-dichloroethane.27 The lower selectivity of
cesium over potassium may be attributable to the greater polarity
of the ionic liquid than 1,2-dichloroethane (dielectric constant 10.19
at 25 °C).
Effects of Extractable Counteranions and Addition of the
4
Sacrificial Ion Exchanger NaBP h . Table 3 shows the DCs
values as a function of the type of counteranion present for Cs+.
From these data, it is clear that the distribution coefficients do
not change significantly with different anions at low concentrations
of BOBCalixC6. At BOBCalixC6 concentrations of 13.8 mM, the
Fig u re 5 . Ligand dependency of BOBCalixC6 in [C4mim][NTf2] from
2
1
M HNO3. Slope ) 1.004; R ) 0.928.
is known that one cesium in each crown cavity can complex two
cesium atoms per BOBCalixC6,27 but at infinite dilution (that is,
DCs values for CsNO and CsOAc (OAc ) acetate) are greater
3
than that for CsCl. However, the distribution coefficient for CsCl
with cesium concentrations below picomolar), the 1:1 monocom-
is the greatest at 20.0 mM BOBCalixC6.
This negligible dependence of the DCs values on a counteranion
can be partially rationalized by two models: (1) ion-exchange or
plexed BOBCalixC6 is favored.
The distribution of Cs+ to the IL phase was observed to
decrease with increasing chain length of the substituted alkyl
group (R) in the organic cation of the IL. This observation is
consistent with a mechanism that includes ion-exchange during
the extraction. The longer the alkyl chain is, the more hydrophobic
(
2) similar anion solubilities in the ILs. On the basis of the ion-
exchange model, only cations are involved in the extraction
process. Accordingly, the counteranions have little effect on the
D
Cs values. The second model requires similar solvation properties
the organic cation becomes, and accordingly, the less the IL will
of Cl , NO3-, and OAc- in ILs. From the present experiments, it
-
partition to the aqueous phase.26
Selectivity. The selectivity for Cs+ was investigated from
is not yet clear which is the dominant factor.
+
+
2+
Using an ion-exchange model, imidazolium cations are lost
from the IL to the aqueous phase during partitioning of metal
cations from the aqueous phase to the IL. Thus, the addition of a
sacrificial cationic species (e.g., Na , H ) to the IL phase that
will preferentially transfer to the aqueous phase should reduce
the loss of imidazolium cations. A sacrificial cation in the IL that
has no affinity toward BOBCalixC6 and that is more hydrophilic
than the IL’s imidazolium cation should result in an enhancement
aqueous solutions containing competitive Na , K , and Sr ions
4 2
in contact with BOBCalixC6-loaded [C mim][NTf ]. The selectivity
coefficient for Cs+ in the presence of competitor cations can be
+
+
obtained from the ratios of the DCs values to the corresponding
distribution coefficients of the competitive ions.
(
(
26) Dietz, M. L.; Dzielawa, J. A.; Laszak, I.; Young, B. A.; Jensen, M. P.; Green
Chem. 2 0 0 3 , 5, 682.
27) Sachleben, R. A.; Bonnesen, P. V.; Descazeaud, T.; Haverlock, T. J.; Urvoas,
A.; Moyer, B. A. Solvent Extr. Ion Exch. 1 9 9 9 , 17, 1445.
of DCs
.
Analytical Chemistry, Vol. 76, No. 11, June 1, 2004 3081

Ta b le 3 . Th e Effe c t s o f Ex t ra c t a b le Co u n t e ra n io n o n t h e Dis t rib u t io n Ra t io s (DCs ) fro m [C4 m im ][NTf2 ] P h a s e s
Co n t a in in g BOBCa lix C6 w it h a n d w it h o u t Na BP h 4 a
[
BOBCalixC6] (mM)
7.71
10.0
13.8
20.0
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
anion
-
NO3
Cl
OAc
10.9
12.2
12.8
9.76
11.4
13.4
22.4
22.6
24.9
21.9
22.6
22.6
97.6
75.3
120
92.7
66.6
85.4
629
956
925
608
940
798
-
-
a
The initial aqueous concentration of Cs+ was 2.5 mM with the specified anion. The solubility of NaBPh4 is also dependent on the water
contents of the original ILs.
Ta b le 4 . Lo s s o f RTIL Ca lc u la t e d fro m UV S p e c t ra a
[
BOBCalixC6] (mM)
7.71
10.0
13.8
20.0
RTIL loss
mM)
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
NaBPh4
0.0 M
NaBPh4
0.12 M
(
CsCl
CsOAc
20.7
20.9
15.6
15.6
20.5
20.7
15.7
15.6
21.2
20.7
15.8
15.8
21.0
21.1
15.7
16.0
a
The initial aqueous concentration of Cs+ was 2.5 mM with the specified anion.
NaBPh
4
were observed. The small changes that were observed
imply that high concentrations
in DCs with the addition of NaBPh
4
of Na+ may compete with Cs+ for the coordination with BOB-
CalixC6 or that the extraction of Cs+ is not totally through the
ion-exchange process.
Support for the substitution of [C
4
mim]+ by Na+ in the ion-
exchange process proposed above comes from the analysis of the
+
concentration of [C
4
mim] released to the corresponding aqueous
phases during extraction. As seen in Table 4 and Figure 6, the
addition of NaBPh decreases the loss of ILs by about 24%, as
4
determined by measuring the UV spectra of [C
corresponding aqueous phases.
4
mim]+ in the
Fig u re 6 . UV spectra of imidazolium (C4mim) cation in the aqueous
phase with and without the use of NaBPh4.
CONCLUSIONS
The work reported here demonstrates that ILs can be effective
media for liquid-liquid extraction, though care must be exercised
in developing a mechanistic understanding of the entire process.
Solutions of BOBCalixC6 in these ILs provide efficient extraction
of Cs+ cation from aqueous solutions under conditions that give
4
Sodium tetraphenylborate (NaBPh ) was added in our experi-
ments as such a sacrificial hydrophilic cationic exchanger. It is
-
known that in aqueous solutions, the BPh
4
anion forms insoluble
compounds with large cations28 such as Cs+ and imidazolium
negligible extraction with traditional organic solvents (e.g., 1,2-
cations and also enhances extraction of cesium. Therefore, the
dichloroethane).20b,29 The selectivity for extracting Cs+ over Na+
-
release of the BPh
4
anion from the imidazolium-based ILs to
and Sr2+ is quite high; however, the concomitant extraction of
aqueous solutions should be negligible in the presence of the
organic cations in aqueous phases. Furthermore, Na+ is very
hydrophilic so that its transport to aqueous phases from ILs is
highly thermodynamically favorable. In addition, our experiments
+
+
K along with Cs has been observed, and the selectivity is less
than in traditional organic solvents. Although the shorter-alkyl-
chain ILs have higher distribution coefficients, the solubilities of
BOBCalixC6 in the corresponding ILs are less. The higher
distribution coefficients for the shorter-alkyl-chain ILs can be
attributed to the ion-exchange capability of the less hydrophobic
imidazolium cations. Because BOBCalixC6 is very hydrophobic,
its solubilities in the longer-alkyl-chain ILs (more hydrophobic)
are greater. Therefore, a compromise has to be made in determin-
ing the optimum IL.
indicated that NaBPh
ILs.
4
could be readily dissolved in the studied
Table 3 compares the extraction results obtained using [C
mim][NTf ] containing various concentrations of BOBCalixC6
with and without 0.12 M NaBPh . As seen in Table 3, no significant
changes of the corresponding DCs values with the addition of
4
-
2
4
(
28) Dupont, J.; Suarez, P. A. Z.; De Souza, R. F.; Burrow, R. A.; Kintzinger, J. P.
(29) Haverlock, T. J.; Bonnesen, P. V.; Sachleben, R. A.; Moyer, B. A. J. Inclusion
Phenom. Macrocyclic Chem. 2 0 0 0 , 36, 21.
Chem.sEur. J. 2 0 0 0 , 6, 2377.
3082 Analytical Chemistry, Vol. 76, No. 11, June 1, 2004

Unlike traditional organic solvent extraction, the efficiency of
cesium-cation extraction from aqueous solutions into IL solutions
of BOBCalixC6 does not change significantly with different
counteranions (e.g., from nitrate to chloride to acetate). The
Laboratory, managed and operated by UT-Battelle, LLC, and
Grants DE-FG07-01ER63266 (UA) and DE-FG07-01ER63296 (UA).
The authors thank T.G. Haverlock for running some tracer
experiments at ORNL.
4
addition of NaBPh can decrease the loss of ILs by 24%, thus
SUPPORTING INFORMATION AVAILABLE
indicating at least a partial role for ion exchange in the observed
results. Further study of these ILs continues in our laboratories.
The details on synthesis and characterization of ILs are
available as Supporting Information. This material is available free
of charge via the Internet at http:/ / pubs.acs.org.
ACKNOWLEDGMENT
This research was supported by the Environmental Manage-
Received for review January 8, 2004. Accepted March 23,
ment Sciences Program, Office of Biological and Environmental
Research, U.S. Department of Energy and DOE NA-22 program,
under Contract DE-AC05-00OR22725 with Oak Ridge National
2004.
AC049949K
Analytical Chemistry, Vol. 76, No. 11, June 1, 2004 3083