Calix[4]pyrrole: A new ion-pair receptor as demonstrated by liquid-liquid extraction

Article abstract of DOI:10.1021/ja7102179

Solvent-extraction studies provide confirming evidence that meso-octamethylcalix[4]pyrrole acts as an ion-pair receptor for cesium chloride and cesium bromide in nitrobenzene solution. The stoichiometry of the interaction under extraction conditions from water to nitrobenzene was determined from plots of the cesium distribution ratios vs cesium salt and receptor concentration, indicating the formation of an ion-paired 1:1:1 cesium:calix[4]pyrrole:halide complex. The extraction results were modeled to evaluate the equilibria inherent to the solvent-extraction system, with either chloride or bromide. The binding energy between the halide anion and the calix[4]pyrrole was found to be about 7 kJ/mol larger for cesium chloride than for the cesium bromide. The ion-pairing free energies between the calix[4]pyrrole-halide complex and the cesium cation are nearly the same within experimental uncertainty for either halide, consistent with a structural model in which the Cs⁺ cation resides in the calix bowl. These results are unexpected since nitrobenzene is a polar solvent that generally leads to dissociated complexes in the organic phase when used as a diluent in extraction studies of univalent ions. Control studies involving nitrate revealed no evidence of ion pairing for CsNO₃ under conditions identical to those where it is observed for CsCl and CsBr.

Full text of DOI:10.1021/ja7102179

Published on Web 03/01/2008
Calix[4]pyrrole: A New Ion-Pair Receptor As Demonstrated by
Liquid-Liquid Extraction
Matthieu P. Wintergerst,† Tatiana G. Levitskaia,†,§ Bruce A. Moyer,*,†
Jonathan L. Sessler,*,‡ and Lætitia H. Delmau*,†
Chemical Separations Group, Chemical Sciences DiVision, Oak Ridge National Laboratory,
P.O. Box 2008, MS-6119, Oak Ridge, Tennessee 37831-6119, and Department of Chemistry and
Biochemistry, 1 UniVersity Station-A5300, The UniVersity of Texas at Austin,
Austin, Texas 78712-0165
Received November 10, 2007; E-mail: moyerba@ornl.gov; sessler@mail.utexas.edu; delmaulh@ornl.gov
Abstract: Solvent-extraction studies provide confirming evidence that meso-octamethylcalix[4]pyrrole acts
as an ion-pair receptor for cesium chloride and cesium bromide in nitrobenzene solution. The stoichiometry
of the interaction under extraction conditions from water to nitrobenzene was determined from plots of the
cesium distribution ratios vs cesium salt and receptor concentration, indicating the formation of an ion-
paired 1:1:1 cesium:calix[4]pyrrole:halide complex. The extraction results were modeled to evaluate the
equilibria inherent to the solvent-extraction system, with either chloride or bromide. The binding energy
between the halide anion and the calix[4]pyrrole was found to be about 7 kJ/mol larger for cesium chloride
than for the cesium bromide. The ion-pairing free energies between the calix[4]pyrrole-halide complex
and the cesium cation are nearly the same within experimental uncertainty for either halide, consistent
with a structural model in which the Cs+ cation resides in the calix bowl. These results are unexpected
since nitrobenzene is a polar solvent that generally leads to dissociated complexes in the organic phase
when used as a diluent in extraction studies of univalent ions. Control studies involving nitrate revealed no
evidence of ion pairing for CsNO3 under conditions identical to those where it is observed for CsCl and
CsBr.
Introduction
These multifunctional receptors are of inherent interest due to
their potential ability to achieve higher levels of ion-recognition
As coordination chemistry has evolved from its traditional
preoccupation with complexation of cations to a more compre-
hensive view that includes recognition of anions, it has become
of interest to study systems that can effect the simultaneous
binding of both cations and anions. This has provided an
incentive to prepare and study so-called ditopic receptors,
synthetic systems that for the most part incorporate individual
cation- and anion-recognition sites within a single framework.1
selectivity in a neutral complex that minimizes the effects of
ion solvation. However, these systems are also important
because they draw attention to the intriguing possibility that
ion-pairing effects,2 generally overlooked in the study of simple
cation or anion receptors, may be playing a more important role
than hitherto appreciated. Although suggested early on by Lehn,3
the importance of ion-pairing in regulating the recognition
properties of synthetic receptors is underscored by the recent
finding that the choice of counter cation affects the chloride
anion binding behavior of two well-studied “simple” anion
receptors, namely calix[4]pyrrole4 (meso-octamethylcalix[4]-
pyrrole; cf. Figure 1) and the open-chain heptapeptides of Gokel
and co-workers.5 Ion-pairing effects have also been noted in
solvent extraction-based separation,6 although not, to the best
of our knowledge, where an anion-selective host is involved in
a high-polarity solvent such as nitrobenzene. In this paper, we
report equilibrium data confirming that calix[4]pyrrole, an easy-
to-synthesize porphyrinogen analogue, acts as an ion-pair
† Oak Ridge National Laboratory.
‡ The University of Texas at Austin.
§ Current address: Radiochemical Sciences and Engineering, Pacific
Northwest National Laboratory, Richland, WA 99352.
(1) For reviews of ion-pair receptors, see: (a) Reetz, M. T. In Molecular
Recognition: Receptors for Molecular Guests; Vo¨gtle, F., Ed.; Pergamon
Press: Oxford, UK, 1996; pp 553-562. (b) Kirkovits, G. J.; Shriver, J.
A.; Gale, P. A.; Sessler, J. L. J. Incl. Phenom. Macrocycl. Chem. 2001,
41, 69-75. (c) Smith, B. D. In Macrocyclic Chemistry: Current Trends
and Future PerspectiVes; Gloe, K., Ed.; Springer: Dordrecht, 2005; pp
137-151. (d) Itsikson, N. A.; Zyryanov, G. V.; Chupakhin, O. N.; Matern,
A. I. Russ. Chem. ReV. 2005, 74 (8), 747-755. (e) Sessler, J. L.; Gale, P.
A.; Cho, W. S. Synthetic Anion Receptor Chemistry; Royal Society of
Chemistry: London, 2006; Chapter 6, pp 259-293. (f) Suksai, S.; Leeladee,
L.; Jainuknan, D.; Tuntulani, T.; Muangsin, N.; Chailapakul, O.; Kong-
saeree, P.; Pakavatchai, C. Tetrahedron Lett. 2005, 46, 2765-2769. (g)
Nabeshima, T.; Saiki, T.; Iwabuchi, J.; Akine, S. J. Am. Chem. Soc. 2005,
127, 5507-5511. (h) Gong, J.; Gibb, B. C. Chem. Commun. 2005, 1393-
1395. (i) Lankshear, M. D.; Cowley, A. R.; Beer, P. D. Chem. Commun.
2006, 612-614. (j) Stetson, C. M.; Nishikawa, S.; Purkiss, D. W.; Dalley,
N. K.; Bartsch, R. A. J. Phys. Org. Chem. 2005, 18, 1107-1115. For a
recent contribution to this area, see: (k) Carmetti, M.; Missinen, M.; Dalla
Cort, A.; Mandolini, L.; Rissanen, K. J. Am. Chem. Soc. 2007, 129, 3641-
3648. (l) Lankshear, M. D.; Dudley, I. M.; Chan, K.-M.; Beer, P. D. New
J. Chem. 2007, 31, 684-690. (m) Dehaen, W.; Gale, P. A.; Garc´ıa-Garrido,
S. E.; Kostermans, M.; Light, M. E. New J. Chem. 2007, 31, 691-696.
(2) For a general introduction to ion-pair effects, see: Marcus, Y.; Hefter, G.
Chem. ReV. 2006, 106, 4585-4621.
(3) Lehn, J.-M. Design of Organic Complexing Agents. Strategies Towards
Properties. In Alkali Metal Complexes with Organic Ligands; Dunitz, J.
D., Ed.; Structure and Bonding 16; Springer-Verlag: New York, 1973; pp
1-69.
(4) Sessler, J. L.; Gross, D. E.; Cho, W.-S.; Lynch, V. M.; Schmidtchen, F.
P.; Bates, G. W.; Light, M. E.; Gale, P. A. J. Am. Chem. Soc. 2006, 128,
12281-12288.
(5) Pajewski, R.; Ferdani, R.; Pajewska, J.; Li, R.; Gokel, G. W. J. Am. Chem.
Soc. 2005, 127, 18281-18295.
9
10.1021/ja7102179 CCC: $40.75 © 2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 4129-4139
4129

A R T I C L E S
Wintergerst et al.
thermochemical steps, as shown in Scheme 1. Each of these
steps may be independently evaluated. First, the cation and anion
partition from the aqueous phase into the organic phase. Second,
the calix[4]pyrrole undergoes a conformational “flipping” so
as to adopt the cone conformation that thereupon binds the
halide. Third, the cesium cation binds within the bowl-like cavity
that is created as the result of this conformational change. These
steps, which could occur concurrently, are expected to allow
the halide anion in question (Cl- or Br- in the present study)
to be bound via pyrrole-NH hydrogen-bonding interactions and
the Cs+ counter cation to be accommodated within the electron-
rich cavity via a combination of π-cation and dipole interactions.
To test and evaluate the above model and to ascertain more
generally whether calix[4]pyrrole was acting as a ditopic (i.e.,
ion-pair) receptor under conditions of solvent extraction, we
decided to use a highly polar diluent, nitrobenzene, as the
organic phase. This choice was made deliberately so as to reduce
the effects of solvent-mediated receptor anion + counter cation
ion-pairing, which might serve to mask the effect of concurrent
anion + cation recognition and extraction. In the case of
receptors that target only one ion, the nature of the complex
(dissociated or ion-paired) is usually dictated by the polarity of
the diluent (solvent or solvents) used to make up the organic
phase. Specifically, diluents of high dielectric constant will lead,
in the limit, to a fully dissociated system for univalent ions,
wherein two charged entities, the receptor ion complex and the
counter ion, are transferred into the organic phase as distinct
species. Under conditions of low solvent polarity, however,
effective ion-pairing may be encountered due to a strong
electrostatic attraction between the receptor ion complex and
the counter ion that is not disrupted extensively by the diluent.
Thus, one of the hallmarks of an effective ditopic receptor would
be an ability to extract paired cation-anion combinations into
a range of organic phases, including ones of high polarity.
Although a number of ditopic receptors have been reported in
recent years,1 to the best of our knowledge we are unaware of
any synthetic receptor that can be used to effect ion-pair
extraction of monovalent ions into highly polar organic phases,
such as nitrobenzene. Nitrobenzene is a well-studied water-
immiscible diluent of high net dielectric constant (ꢀ ) 34.6)
that, in our experience, stabilizes dissociated ions, including
those formed from all monotopic receptor ion + counter ion
combinations we have tested to date. Nonetheless, we have
found that calix[4]pyrrole (cf. Figure 1) acts as an ion-pair
receptor for CsCl and CsBr under conditions of aqueous-
nitrobenzene extraction. Both the CsCl and CsBr extraction
systems, where the ion-pairing effects were seen, were fully
defined by detailed thermodynamic analyses. This has permitted
quantification of all anion binding, ion-pair complexation, and
extraction parameters. It has also provided insights into the
structure of the receptor anion-cation complex that could not
be inferred from simple correlations of receptor and salt
concentration with extraction efficiency.
Figure 1. Compounds used in this study.
receptor for CsCl and CsBr, but not CsNO3, under conditions
of aqueous-nitrobenzene extraction.
meso-Octamethylcalix[4]pyrrole (“calix[4]pyrrole”; Figure 1)
has been studied extensively in recent years as an easy-to-
prepare anion receptor.7-11 Prompted by suggestive extraction
behavior, we have recently found that this system acts as a
ditopic receptor for certain salts, including CsCl and CsBr, in
the solid state.7 Specifically, X-ray diffraction analyses of the
CsCl and CsBr complexes revealed that the macrocycle adopts
a cone conformation that allows for hydrogen-bonding interac-
tions between the halide anion and the pyrrole NH protons while
providing an electron-rich cavity that serves as a receptor for
1
the cesium cation. Detailed H NMR spectroscopic analyses
provided qualitative support for the presence of weak, albeit
significant, cesium cation receptor interactions in organic
solution. These findings have led us to return to liquid-liquid
extraction to determine whether calix[4]pyrrole functions as a
ditopic receptor in cesium salt extraction and, if so, to quantify
the equilibrium constants corresponding to the key thermo-
chemical steps (see below).
Previous experimental and computer modeling studies8 have
been carried out on this calix[4]pyrrole system. On the basis of
these studies, focused on anion recognition, we consider that a
solvent-extraction process involving cesium salts, and particu-
larly halide salts, may be described conveniently by three
(6) Examples of anion effects in solvent extraction include: (a) Yamamoto,
Y.; Tarumoto, T.; Iwamoto, E. Chem. Lett. 1972, 3, 255-258. (b)
McDowell, W. J.; Shoun, R. R. Proceedings of the International SolVent
Extraction Conference (ISEC 77), Toronto, Ontario, Canada, Sept 9-16,
1977. (c) Lucas, B. H., Ritcey, G. M., Smith, H. W., Eds.; Canadian Institute
of Mining and Metallurgy: Montreal, Quebec, Canada, 1979; Vol. 1, pp
95-100. (d) Yakshin, V. V.; Abashkin, V. M.; Laskorin, B. N. Dokl. Akad.
Nauk SSSR 1980, 252, 373. (e) Hankins, M. G.; Bartsch, R. A.; Olsher, U.
SolVent Extr. Ion Exch. 1995, 13, 983-995. (f) Miyabe, K.; Taguch, S.;
Kasahara, I.; Got, K. J. Phys. Chem. B 2000, 104, 8481-8490. (g) Shi, X.
D.; Mullaugh, K. M.; Fettinger, J. C.; Jiang, Y.; Hofstadler, S. A.; Davis,
J. T. J. Am. Chem. Soc. 2003, 125, 10830-10841. (h) Levitskaia, T. G.;
Maya, L.; Van Berkel, G. J.; Moyer, B. A. Inorg. Chem. 2007, 46, 261-
272. (i) Mahoney, J. M.; Beatty, A. M.; Smith, B. D. Inorg. Chem. 2004,
43, 7617-7621.
(7) Custelcean, R.; Delmau, L. H.; Moyer, B. A.; Sessler, J. L.; Cho, W.-S.;
Gross, D.; Bates, G. W.; Brooks, S. J.; Light, M. E.; Gate, P. A. Angew.
Chem., Int. Ed. 2005, 44, 2537-2542.
(8) (a) Gale, P. A.; Sessler, J. L.; Kra´l, V.; Lynch, V. J. Am. Chem. Soc. 1996,
118, 5140-5141. (b) Wu, Y.-D.; Wang, D.-F.; Sessler, J. L. J. Org. Chem.
2001, 66, 3739-3746.
We also sought to find a cesium salt where the anion
recognition was sufficiently weak that the calix[4]pyrrole would
not adopt the cone conformation required for cesium cation
complexation. Under these latter conditions, it was expected
that calix[4]pyrrole would either show no effect on the extraction
process or, at best, display behavior consistent with that of a
monotopic anion receptor. Since both of these control scenarios
(9) Gale, P. A.; Sessler, J. L.; Kra´l, V. Chem. Commun. 1998, 1-8.
(10) (a) Bayer, A. Ber. Dtsch. Chem. Ges. 1886, 19, 2184-2185. (b) Dennstedt,
M.; Zimmerman, J. Ber. Dtsch. Chem. Ges. 1887, 20, 850-857. (c)
Rothemund, P.; Gage, C. L. J. Am. Chem. Soc. 1955, 77, 3340-3342.
(11) Levitskaia, T. G.; Marquez, M.; Sessler, J. L.; Shriver, J. A.; Vercouter,
T.; Moyer, B. A. Chem. Commun. 2003, 17, 2248-2249.
9
4130 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Calix[4]pyrrole: A New Ion-Pair Receptor
A R T I C L E S
Scheme 1. Proposed Steps in Cesium Salt Extraction
would give rise to extraction behavior characteristic of a fully
dissociated extraction system, namely an independence of the
extraction efficiency on salt concentration, they would stand in
marked contrast to the case where calix[4]pyrrole is acting as
a bona fide ditopic receptor. As detailed below, we found that
CsNO3 serves as the requisite control system when nitrobenzene
is used as the diluent; in the presence of calix[4]pyrrole, this
salt exhibits a complete independence of the extraction efficiency
on the salt concentration. In contrast, the complexes of CsCl
and CsBr were fully ion-paired in nitrobenzene, as demonstrated
by the linear dependence of the extraction efficiency on both
the salt and the calix[4]pyrrole concentrations.
Considering the associated system (eq 1) and assuming no
other extracted form of Cs+, [Cs+]org is equivalent to [Cs-
(X)(calix)p]org. Defining DCs ) [Cs+]org/[Cs+]aq ) [Cs(X)
(calix)p]org/[Cs+]aq and substituting into the standard expression
for the extraction constant for eq 1, one obtains, upon rear-
rangement,
γCsγXγpcalix
DCs ) K1[X-][calix] p
(3)
γCs(X)(calix)
For this purpose, the activity coefficients γcalix and γCs(X)(calix)
are assumed to be close to unity, a reasonable assumption
considering the low calix[4]pyrrole concentration in the organic
phase. Furthermore, if CsX is assumed to be the only electrolyte
in the system, the following expression can be obtained:
Results and Discussion
Equilibrium Analysis. In the case of calix[4]pyrrole, two
kinds of organic-phase complexes can be expected, depending
on the extent of ion-pairing, according to the following two
equilibria:
log (DCs) ) log(K1) + log(γ2Cs[Cs+]) +
γpcalix
p log([calix]) + log
(4)
Cs+ + X- + p calix h CsX(calix)p
(K1)
(K2)
(1)
(2)
(
)
γCs(X)(calix)
Cs+ + X- + p calix h Cs+ + (calix)pX-
Slope analysis of plots of DCs vs [calix] (commonly in log-
log scale) would allow determination of the calix[4]pyrrole
stoichiometry (eqs 1 and 4). In addition, a plot of log(DCs/γCs)
versus log(γCs[Cs+]) according to eq 4 would show a slope of
1 in the case of extraction via an ion-pair mechanism.
where X- ) Cl-, Br-, or NO3-, calix represents the calix[4]-
pyrrole, and p represents the binding stoichiometry. Over bars
indicate that species are present in the organic phase. In general,
these complexes and the associated equilibria would be expected
to reflect the nature of the counteranion and the polarity of the
diluent, as noted above.
If the extraction produces a dissociated system as defined by
eq 2, assuming the electroneutrality of the organic phase and
an appreciable concentration in this phase of only Cs+
and
(org)
(calix)pX-
as product species permits the following rear-
(org)
The first equilibrium (eq 1) corresponds to an extraction
through ion-pairing, whereas the second equilibrium (eq 2)
reflects an extraction via a dissociated system with the calix-
[4]pyrrole acting as the anion receptor. On the basis of previous
findings,11 a system controlled by eq 2 is expected for univalent
ions when using a polar diluent, such as nitrobenzene.
rangement of the standard extraction constant expression:
γoXrg[Cs+]
org
Cs
γ
K2 ) DCs
(5)
γaqγaXq[X-][calix]p
Cs
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4131

A R T I C L E S
Wintergerst et al.
Figure 2. Extraction of cesium nitrate, cesium chloride, and cesium bromide by nitrobenzene. Experimental findings and modeling results. Organic (O)
phase: nitrobenzene. Aqueous (A) phase: variable cesium nitrate, chloride, or bromide concentrations. O/A ) 1; T ) 25 °C.
Applying the concept of electroneutrality to the aqueous phase
and assuming that the activity coefficients are close to unity,
this equation becomes
Extraction by Nitrobenzene Alone. To determine the
“background” due to the cesium salt partitioning between the
aqueous phase and nitrobenzene, a series of control experiments
was carried out; these involved separate extractions of cesium
nitrate, cesium chloride, and cesium bromide using nitrobenzene
in the absence of any added extractant. Figure 2 shows that, as
expected for this polar diluent, nitrobenzene extracts cesium
salts as dissociated ions (near independence of DCs on aqueous
salt concentrations). Cesium distribution ratios also are very low,
but this is expected and in agreement with the high free energies
of partitioning for these ions (see Supporting Information, Table
S1).
D2Cs
[calix]p
1
2
K2 )
w log(DCs) ) (log(K2) + p log[calix]) (6)
In the case of extraction as a dissociated system, eq 6 shows
that the value of DCs is completely independent of the variable
[Cs+]aq when activity coefficients approach unity.
In summary, a simple slope analysis of various graphs permits
one to obtain considerable insight into the system in question.
For instance, given constant activity coefficients:
Extraction of Cesium Nitrate and Cesium Halide by Calix-
[4]pyrrole in Nitrobenzene. 1. Cesium Nitrate. As explained
in the Experimental Section (Supporting Information), a simple
slope analysis allows the nature of the extraction system
(dissociated ions vs ion-pairs) and its stoichiometry to be
determined. The variation of the ratio DCs versus the cesium
nitrate concentration in the aqueous phase (Figure 3) revealed
that the ratio is essentially independent of the cesium nitrate
concentration. (Note: The low values of DCs and the small
amount of cesium transferred into the organic phase allow the
cesium concentration in the aqueous phase at equilibrium to be
approximated by the initial cesium nitrate concentration.) This
independence is consistent with an extraction system that is
dissociated according to eq 6. Considering the same equation
and plotting log(DCs) versus log[calix[4]pyrrole] gives a straight
line, the slope of which is p/2, where p is the stoichiometric
coefficient corresponding to the equilibrium of eq 2. The
resulting plot is shown in Figure 4; since a slope of 1/2 is seen,
it is concluded that only one molecule of calix[4]pyrrole is
involved per nitrate anion in the complex. This stoichiometry
and the dissociated nature of the system meet the original
assumptions and expectations for CsNO3, namely that with this
salt, the calix[4]pyrrole does not act as an ion-pair receptor.
This is not surprising since there is no evidence that nitrate anion
binds strongly to the calix[4]pyrrole in organic media9 or can
help organize it in the cone-like conformation that would favor
Cs+ binding.
• A plot of log(DCs) versus log([CsX]) will generally give a
straight line, the slope of which will allow one to determine
the nature of the extraction process (i.e., slope ) 1 and 0 for
ion-pairing and dissociation, respectively).
• A plot of log(DCs) versus log([calix]) will give a straight
line with a slope of p or p/2 for ion-pairing and dissociation
processes, respectively, where p corresponds to the stoichio-
metric coefficient for the calix[4]pyrrole involved in the
extraction system equilibrium.
Once the stoichiometry of extraction is determined through
slope analysis, the associated thermodynamic values can be
derived by modeling the cesium extraction process according
to the equations presented above. Of particular interest are the
equilibrium constants corresponding to halide anion complex-
ation and ion-pairing with cesium. These values, referred to as
Kcompl and Kip, respectively, are defined as follows:
X- + calix h Xcalix-
(Kcompl
(Kip)
)
(7)
(8)
M+ + Xcalix- h MXcalix
A primary goal of this study was to determine both of these
values for calix[4]pyrrole under a set of well-defined conditions.
In point of fact, ion-pairing in the case of CsCl and CsBr proved
so strong that Kcompl could only be determined through indirect
means, as discussed below.
9
4132 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Calix[4]pyrrole: A New Ion-Pair Receptor
A R T I C L E S
Figure 3. Extraction of cesium nitrate by calyx[4]pyrrole. Experimental findings and modeling results. Organic phase: [calix[4]pyrrole] ) 25 mM in
nitrobenzene. Aqueous phase: variable cesium chloride concentrations. O/A ) 1, T ) 25 °C.
Figure 4. Extraction of cesium nitrate by calix[4]pyrrole. Organic phase: variable calix[4]pyrrole concentrations in nitrobenzene. Aqueous phase: [CsNO3]
) 0.6 M. O/A ) 1, T ) 25 °C.
2. Cesium Halides. Experiments identical to those described
above were run in the case of cesium chloride and cesium
bromide. Again, DCs was plotted as a function of the cesium
halide concentration in the aqueous phase, as well as against
the concentration of calix[4]pyrrole in the organic phase.
The use of activity corrections in the chloride or bromide
systems led to a significant difference in the overall shape of
the curve. Subject to their use, the dependence of DCs on the
salt concentration clearly exhibits a slope of one, providing
strong evidence for the existence of an ion-pair interaction
between the cesium cation and the anion/calix[4]pyrrole com-
plex. Further support for this conclusion came from an experi-
ment where the salt concentration was kept constant while the
concentration of calix[4]pyrrole was allowed to vary. Figure 7
shows that plots of DCs vs concentration produced in this way
also have a slope of one. On the basis of this slope analysis,
we conclude that a 1:1:1 ion-pair of cesium:halide:calix[4]-
pyrrole is formed in the nitrobenzene phase (eq 9). This
stoichiometry is consistent with the previous crystallographic
results.7
The resulting slope analyses, shown in Figure 5, reveal that
the cesium distribution ratios are dependent on the cesium halide
concentration. Moreover, the slopes of these plots are close to
one, especially at [CsX] g 0.1 M. Such a finding is thus fully
consistent with the existence of an ion-pair complex between
the calix[4]pyrrole and these two sets of salts.
However, before drawing definitive conclusions about the
slopes in these plots, it is important to take into account the
relevant activity coefficients. This was specifically done in the
case of cesium bromide, as shown in Figure 6. Here, the cesium
distribution ratio DCs is plotted versus the initial concentrations
of cesium bromide in the aqueous phase, as well as versus the
corresponding activities for comparison. The activity coefficients
were calculated (see Supporting Information) using the specific
ion interaction theory (SIT) or “extended Debye-Hu¨ckel
theory”.12
Cs+ + X- + calix h CsX(calix)
(K1)
(9)
(12) Grenthe, I.; Plyasunov, A. V.; Spahiu, K. Modeling in Aquatic Chemistry.
In Estimation of Medium Effects on Thermodynamic Data; OECD Publica-
tions: Paris, 1997; Chapter IX.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4133

A R T I C L E S
Wintergerst et al.
Figure 5. Extraction of cesium chloride or bromide by calix[4]pyrrole. Experimental findings and modeling results. Organic phase: [calix[4]pyrrole] ) 10
or 25 mM in nitrobenzene. Aqueous phase: variable cesium chloride or cesium bromide concentrations. O/A ) 1, T ) 25 °C.
Figure 6. Extraction of cesium bromide by calix[4]pyrrole in nitrobenzene. Change in curve shape upon applying activity corrections. Organic phase:
[calix[4]pyrrole] ) 25 mM in nitrobenzene. Aqueous phase: variable cesium bromide concentrations. O/A ) 1, T ) 25 °C.
Determination of the Thermodynamic Constants Govern-
ing the Dissociated System. While the slope analyses described
above serve to define the stoichiometry of the extraction process
in well-behaved systems, such as those comprised of calix[4]-
pyrrole and the three cesium salts considered in the present
study, it is important to appreciate that they do not serve to
define the thermodynamics of either the overall process or the
individual binding events. Given the unusual nature of the ion-
pairing observed in the case of CsCl and CsBr, and the paucity
of systems that have been subject to this level of detailed
analysis, it became of interest to map out the thermodynamics
of binding in the case of these latter two salts.
In accord with Scheme 1, and appreciating the additive nature
of the interactions involved, we thought it would prove easiest
to calculate the equilibrium constant corresponding to formation
of the ion-paired complex in nitrobenzene using a stepwise
approach. With such considerations in mind, the plan was to
determine first the formation constants corresponding to the
complexation of the dissociated halide anion and the calix[4]-
pyrrole receptor in the organic phase. Following this, we would
then derive those for the binding of the cesium cation to the
negatively charged calix[4]pyrrole-halide anion complexes
produced as the result of this first recognition process. In
principle, the first of these stepwise contributions to the overall
energetics associated with formation of the ion-pair complex
could be determined at vanishingly low concentrations of cesium
halide, since in this limit the salts in question would tend to
exist as fully dissociated species in the organic phase. Unfor-
tunately, conditions could not be found where this requirement
is met. Specifically, it proved impossible to reduce the concen-
tration of free cesium cation in the organic phase to a level low
enough to preclude cation binding while concurrently providing
anion concentrations sufficiently high to allow the desired calix-
[4]pyrrole-anion complexation equilibria to be measured ac-
curately via direct means (including through data modeling).
This is not surprising, given the very low DCs values that govern
the CsCl and CsBr + calix[4]pyrrole aqueous-nitrobenzene
extraction system.
9
4134 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Calix[4]pyrrole: A New Ion-Pair Receptor
A R T I C L E S
Figure 7. Extraction of cesium chloride or bromide by calix[4]pyrrole. Experimental findings and modeling results. Organic phase: variable calix[4]pyrrole
concentrations in nitrobenzene. Aqueous phase: [CsCl] ) 1 M, [CsBr] ) 0.3 or 1 M. O/A ) 1, T ) 25 °C.
Figure 8. Extraction of cesium chloride or bromide by BOBcalixC6 and calix[4]pyrrole. Experimental findings and modeling results. Organic phase:
[BOBcalixC6] ) 10 mM, [calix[4]pyrrole] ) 0 or 10 mM in nitrobenzene. Aqueous phase: variable cesium chloride or cesium bromide concentrations. O/A
) 1; T ) 25 °C.
Given this challenge, an alternative approach was employed.
It relies on the use of a cesium ligand to prevent ion-pairing
and thereby increase the DCs values. This approach, which
proved successful (vide infra), is predicated on the recognition
that a strong cation receptor would act to complex the cesium
in the organic phase and thus suppress ion-pairing over a wide
range of salt concentrations. In our experience, one of the best
receptors for the cesium cation in solvent extraction is the
crown-strapped calixarene known as BOBcalixC6 (Figure 1).13
Figure 8 shows the results of these experiments in the case
of cesium chloride and cesium bromide. As expected for a
dissociated system, at low concentrations, the DCs values
recorded in the presence of BOBcalixC6 were found to be
independent of the salt concentration; this was true both in the
presence and in the absence of calix[4]pyrrole. Analysis of the
data is described in the next section.
At higher concentrations, the cesium distribution ratio may
be seen to decrease. The expected reason for the decrease is
that, at high concentrations, the concentration of the cesium salt
in the organic phase can exceed that of the initial calix[4]pyrrole
extractant, making it and BOBcalixC6 unavailable for further
extraction. The organic phase becomes saturated or “loaded”
with cesium, which would give rise to a decrease in the DCs
value, as seen by experiment (cf. Figure 8) and confirmed by
equilibrium modeling (vide infra). Nevertheless, there remains
a concern that, at higher concentrations, potential interactions
between the free BOBcalixC6 and calix[4]pyrrole would be
favored. In order to determine whether the interactions between
BOBcalixC6 and calix[4]pyrrole were appreciable, a solution
containing 10 mM of both receptors was made up in deuterated
nitrobenzene. The use of deuterated nitrobenzene allowed the
1H NMR spectrum to be recorded. It was found that the
spectrum of the mixture was a perfect superposition of the
(13) (a) Sachleben, R. A.; Bonnesen, P. V.; Descazeaud, T.; Haverlock, T. J.;
Urvoas, A.; Moyer, B. A. SolVent Extr. Ion Exch. 1999, 17, 1445-1459.
(b) Moyer, B. A.; Sachleben, R. A.; Bonnesen, P. V.; Presley, D. J.
Calixarene Crown Ether Solvent Composition and Use Thereof for
Extraction of Cesium from Alkaline Waste Solutions. U.S. Patent 6,174,-
503, Jan 16, 2001.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4135

A R T I C L E S
Wintergerst et al.
Table 1. Comparison between the Partitioning Constants from the
Literature and from the Computer Modeling for Cesium Nitrate,
Cesium Chloride, and Cesium Bromide
Table 3. Equilibrium Constants K3 and K4 Obtained from the
Model for the System Cs+-BOBcalix/X--Calix[4]pyrrole and
Kcompl and K2 Calculated from These Valuesa
nitro
Gwpart
∆
Gw
(kJ mol-
f
part
f
nitro
X-
log K3
log K4
log Kcompl
log K2
∆
1
a
a
1
b
b
system
(kJ mol-
)
log Ks
)
log Ks
±
±
Cl-
Br-
-1.05 ( 0.01
0.33 ( 0.01
2.61 ( 0.01
2.76 ( 0.02
3.66 ( 0.02
2.43 ( 0.03
-5.84 ( 0.11
-5.87 ( 0.05
CsNO3/NB
CsCl/NB
CsBr/NB
42
53
47
-7.36
-9.28
-8.23
43.2 ( 0.4
54.2 ( 0.5
47.4 ( 0.1
-7.56 ( 0.07
-9.50 ( 0.09
-8.30 ( 0.02
a Data points used in the model are shown in Figure 8.
a Marcus, Y. Ion properties; Marcel Dekker: New York, 1997. b This
work. NB ) nitrobenzene.
Table 4. Complexation Constant K5 and KBOB between
BOBcalixC6 and Cesium Obtained from the Model and the
Calculations, Respectively, for the CsCl and CsBr Saltsa
system
log KBOB
Table 2. Equilibrium Constant K1 As Derived from the Modela
Cs+-BOBcalix from Cl-
Cs+-BOBcalix from Br-
8.23 ( 0.01
8.57 ( 0.01
system
log K1
CsCl/calix[4]pyrrole
CsBr/calix[4]pyrrole
-0.80 ( 0.01
-0.58 ( 0.02
a Data points used in the model are shown in Figure 8.
a Data points used in the model are shown in Figures 5 and 7.
The last stage of the computer modeling involved calculating
the formation constants, K2, according to eq 4. Here, as noted
above, the use of BOBcalixC6 allows this key thermodynamic
parameter to be derived in an indirect manner. According to
the two equilibria of eqs 11 and 12, the constant K2 may be
determined from the ratio K4Ks(/K3; this is because eq 4 is
equivalent to eq 12 - eq 11 + eq 10; again, this assumes there
are no interactions between BOBcalixC6 and calix[4]pyrrole
and that BOBcalixC6 binds Cs+ much more strongly than calix-
[4]pyrrole.
spectra for the individual components. This provides support
for the notions that the interactions between BOBcalixC6 and
calix[4]pyrrole in this solvent are not appreciable and that
BOBcalixC6 could be used to prevent the cesium cation from
binding to the calix[4]pyrrole-anion complex.
Modeling and Thermodynamic Data Sets. Data analysis
and equilibrium-constant determinations were performed using
the SXLSQI computer program.14 Parameters used in the
modeling program to account for nonideality and volume effects
are presented in Tables S4-S6. Computer modeling was
performed in three stages. In the first stage, the extractions of
cesium salts by nitrobenzene only were analyzed, with the goal
of determining the partitioning constants and the free energies
of transfer of the various salts between the aqueous phase and
the organic nitrobenzene phase as shown below:
M+ + X- + BOB h M‚BOB+ + X-
(K3) (11)
M+ + X- + BOB + calix h
M‚BOB+ + Xcalix-
(K4) (12)
The results obtained when BOBcalixC6 is used are shown
in Figure 8, with the formation constants derived in this way
being given in Table 3. These formation constants, in turn, can
be used in order to derive the values of the complexation
M+ + X- h M+ + X-
(Ks()
(10)
The values found experimentally were then compared to those
available in the literature. Figure 2 shows the best fits obtained
for the three cesium salts considered in this study. It reveals an
approximate independence of the DCs value from the salt
concentrations in the absence of calix[4]pyrrole and is thus
consistent with the proposal that extractions occurring under
these conditions involve fully dissociated species. The slight
variation of DCs seen in the plots is quantitatively explained by
activity effects. Likewise, the values in Table 1 serve to highlight
the remarkable consistency between the partitioning constants
from the literature and the refined fitting parameters calculated
via the present computer modeling.
The second stage of the analysis was aimed at modeling the
cesium distribution ratios produced by calix[4]pyrrole alone over
the concentration regime where ion-pairing is prevalent, with
the goal of calculating log K1 (see eq 3). The modeling of the
cesium chloride and the cesium bromide data is shown in
Figures 7 and 8, which correspond to studies where the ligand
and salt concentrations were varied, respectively. All plots
demonstrate the excellent agreement between the observed and
calculated data points. The refined formation constants obtained
from the model, including K1, are given in Table 2.
constants in nitrobenzene in the absence of ion-pairing (KBOB
as follows:
)
M+ + BOB h M‚BOB+
(KBOB
)
(13)
This equation is derived from
K3
KBOB
)
Ks(
Table 4 summarizes the KBOB values from the calculations
using cesium chloride or bromide with, of course, the same value
being expected in both cases. Thus, to the extent consistent
results were seen, it would provide support for the validity of
the model.
A congruence between the chloride and bromide cases was
seen, with the corresponding values of KBOB being reasonably
close. These values were used to obtain the main equilibrium
constants of interest, namely the halide-calix[4]pyrrole com-
plexation constants, Kcompl, and the ion-pairing constants, Kip,
defined by the equilibria summarized in eqs 1 and 2.
Considering eq 4 and the partitioning equilibrium (eq 10),
the complexation constants can be calculated as the ratio, Kcompl
) K2/Ks(. However, as discussed previously, in order to
determine K2 directly, a very flat line, corresponding to a strong
(14) Baes, C. F., Jr. SXLSQI, A program for Modeling SolVent Extraction
Systems; Report ORNL/TM-13604; Oak Ridge National Laboratory: Oak
Ridge, TN, Dec 7, 1998.
9
4136 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Calix[4]pyrrole: A New Ion-Pair Receptor
A R T I C L E S
anion.8,9,15 In addition to these receptor-selective effects, the
anion binding affinity can be attenuated by anion hydration in
the nitrobenzene phase. The hydration numbers of Cl- and Br-
in nitrobenzene are 4 and 2.1, respectively.16 Since this water
of hydration presumably becomes partially lost upon anion
complexation by calixpyrrole, the energetic cost is expected to
be greater in the case of chloride than bromide, and the net
effect is to attenuate any selectivity between Cl- and Br- that
would be expected on the basis of anion binding affinities alone
(i.e., KaCl > KaBr).
Table 5. Complexation Constants and Complexation Energies
Obtained from Calculation
∆
Gcompl
1
systema
log Kcompl
Kcompl
(kJ mol-
)
calix[4]pyrrole-Cl-
calix[4]pyrrole-Br-
3.66 ( 0.02
2.43 ( 0.03
4570 ( 209
270 ( 18
-20.9 ( 0.1
-13.9 ( 0.2
a The complexation constant was calculated from K3 and K4.
Table 6. Ion-Pairing Constants and Ion-Pairing Energies Obtained
from the Calculation
∆
Gip
5
1
systema
log Kip
Kip
×
10-
(kJ mol-
)
Conclusion
Cs+-calix[4]pyrrole/Cl- 5.04 ( 0.12 (1.09 ( 0.30) -28.8 ( 0.7
Cs+-calix[4]pyrrole/Br- 5.25 ( 0.07 (1.93 ( 0.29) -30.2 ( 0.6
The most noteworthy finding to emerge from this study is
that calix[4]pyrrole, a system well-appreciated for its anion
binding capabilities,9 behaves as a ditopic ligand for cesium
chloride and cesium bromide, forming a complex of 1:1:1
cesium:calix[4]pyrrole:halide stoichiometry, under conditions
of aqueous-nitrobenzene solvent extraction. Such a finding,
which is nearly unprecedented in the case of neutral extractants,
occurs in spite of the high polarity of the organic phase (i.e.,
nitrobenzene). This ability to promote ion-pair formation
appears, at least so far, to be limited to these two halide anions,
as both qualitative and semiquantitative analyses of other cesium
salts (e.g., CsNO3) revealed behavior that was consistent with
that of a fully dissociated extraction system.
Using experimental data and appropriate computer models,
the thermodynamic constants and binding energies for various
interactions of CsCl and CsBr could be determined. These
analyses revealed, as expected, that the chloride anion is
bound more strongly to meso-octamethylcalix[4]pyrrole than
bromide. Specifically, a complexation energy roughly 7 kJ/mol
higher was seen in the case of CsCl as compared to CsBr. The
binding energy for the interaction of the cesium cation with
the anionic complexes formed from these two salts was also
determined and found to be nearly independent of the nature
of the halide.
Depending on the size and the geometry of the extracted anion
(e.g., halide vs nitrate), the calix[4]pyrrole is or is not able to
complex the cesium cation and thus act as an ion-pair receptor.
Such anion dependence has already been shown in the solid
state, with the structures of various calix[4]pyrrole cesium cation
anion complexes being found to differ considerably depending
on the nature of the bound anion.7 In other words, cesium cation
binding is seen only in those cases where the anion in question
(e.g., Cl- or Br-) is able to support formation of the so-called
cone conformation of the calix[4]pyrrole.
It is not possible to determine the exact position of cesium
in the ion-pair complex formed in nitrobenzene from CsCl or
CsBr under the present solvent extraction conditions. However,
by analogy to what is seen in the solid state, we think it is very
likely that the cation lies in the bowl-shaped cavity of the
calixpyrrole. In fact, the similarities in the ion-pairing formation
constants for the chloride and bromide systems are most easily
rationalized in terms of such an assumption as opposed to
invoking other orientations, even ones where the bound anion
and cesium counter cation might be in closer proximity.
a The complexation constant was calculated from K3 and K4 and used to
calculate K2.
dissociated system, is necessary at low concentrations. Since
such a limiting scenario could not be achieved under the present
experimental conditions, BOBcalixC6 was used. Recognizing
that the complexation constant can be defined also as the ratio
Kcompl ) K4/K3, the equilibrium K2 can be calculated since Ks(
is known.
Thermodynamic Observations. Once the intermediate equi-
librium constants, K1 and K2, were calculated using the above
computer modeling approach, both constants of interest, namely
K
compl and Kip, could be calculated using the following expres-
sions:
K2
K4
K1
K2
Kcompl
)
)
and
Kip )
Ks( K3
According to the values derived via this analysis, summarized
in Tables 5 and 6, the chloride anion binds more strongly to
calix[4]pyrrole than does the bromide anion. This result is in
agreement with early solution phase analyses (carried out in
dichloromethane).8 It is also consistent with the crystallographic
data, which has established that the hydrogen bond distances
between the four NH groups and the bound halide anion are
shorter for chloride than bromide in the solid state (d ) 3.27-
3.32 and 3.45-3.52 Å for the relevant N‚‚‚Cl and N‚‚‚Br
interactions, respectively).7 The ion-pairing constants are very
high and nearly the same within experimental uncertainty. Tables
5 and 6 also list the complexation and ion-pairing energies. The
value of the complexation energy reflects at least several
energetic components, namely (1) the deformation energy
corresponding to conversion of the calix[4]pyrrole from the 1,3-
alternate conformation to the cone conformation, something that
is expected to depend at least in part on the specific anion bound,
and (2) the four hydrogen bond energies. Thus, the complexation
energy directly relates to the overall binding energy of the
halide-calix[4]pyrrole complex, with chloride and bromide
displaying a difference of 7 kJ/mol in this solvent. While
solubility considerations prevented an independent confirmation
of these values by more direct means, as noted above, a number
of previous studies, carried out in a range of solvents other than
nitrobenzene, have served to confirm that chloride anion is
bound far more strongly by calix[4]pyrrole than bromide
(15) (a) Sessler, J. L.; Gale, P. A.; Cho, W. S. Synthetic Anion Receptor
Chemistry; Royal Society of Chemistry: London, 2006; 413 pp. (b) Lee,
C.-H.; Na, H.-K.; Yoon, D.-W.; Won, D.-H.; Cho, W.-S.; Lynch, V. M.;
Shevchuk, S. V.; Sessler, J. L. J. Am. Chem. Soc. 2003, 125, 7301-7306.
(16) Osakai, T.; Ogata, A.; Ebina, K. J. Phys. Chem. B 1997, 101, 8341-8348.
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4137

A R T I C L E S
Wintergerst et al.
Additional evidence that actual coordination of Cs+ ion in
the bowl of the calixpyrrole is taking place lies in the
unexpectedly large magnitude of the ion-pairing constants, Kip
) 5.04 × 105 and 5.25 × 105 for Cs(calixpyrrole)+ with Cl-
and Br-, respectively (Table 6). As a most conservative
reference point, one might choose to compare these values of
Kip with those for the “naked” Cs+ and halide ions in the water-
saturated organic phase. Such values are not available in the
literature, as the salts are too sparingly soluble and extraction
is too feeble to achieve sufficient organic-phase concentrations
of the ions for appreciable pairing. From the experiment shown
in Figure 2, however, we estimate that ion-pairing could be
detected if Kip(CsCl) > 2 × 104 and Kip(CsBr) > 4 × 103 for Cs+
pairing with Cl- and Br-, respectively, if we assume the
modeling could “see” [CsX]org > 0.2[Cs+]org. On the other hand,
calculation for dry nitrobenzene using the Fuoss theoretical ion-
association treatment2 predicts Kip(CsX) values less than 100.
Although literature on ion-pairing in nitrobenzene is scarce, one
may compare with Kip(CsPicrate) ) 602-1510 for the extraction
of cesium picrate into nitrobenzene.17 It is known that complexed
salts in polar solvents often behave as strong electrolytes,17,18
at least for cation complexation, in accord with the ligand-
shielding effect.3 Indeed, we have seen in the extraction of Li+
by a crown ether in 1-octanol that ion-pairing with Cl- decreases
from 1.3 × 104 for the free ions in the water-saturated 1-octanol
to 5.4 × 103 for the complex with the anion.19 Given that the
effective charge of the halide is reduced by the hydrogen
bonding from the calixpyrrole, it thus seems likely that ion-
pairing of the Cs+ cation outside of the bowl of the calixpyr-
role-halide complex is too weak to explain the high ion-pair
association observed. It should be noted that the equilibrium
analysis is indicative of Cl-(calixpyrrole)Cs+ assembly forma-
tion. However, one can argue that, in this case, Cs+ inclusion
into the calixpyrrole cavity is driven not by electrostatic
attraction to the bound anion (definition for ion-pairing process)
but rather by favorable interactions with the electron-rich ligand
cavity. Unfortunately, equilibrium analysis does not provide
insight into the nature of the process. Rather, equilibrium
analysis quantifies Cs+ inclusion into the pre-organized calix-
like “bowl” formed by the anion-binding calixpyrrole cavity.
In this sense, formation of the Cl-(calixpyrrole)Cs+ species
could reflect the fact that Cs+ coordination to the calixpyrrole
bowl is more important in nitrobenzene than electrostatic ion-
pairing effects, something that would manifest in terms of out-
of-the cavity Cs+-Cl- interactions leading to the generation
of a contact ion-pair. This proposal is also supported by the
finding that the D values depend on [CsCl] over the entire
concentration range, and a positive slope is observed even at
very low CsCl concentrations in the organic phase (Figure 5),
conditions under which contact ion-pair effects are not expected
to be significant in the moderately polar solvent, nitrobenzene.
Further, ion-pairing between the Cs(BOB)+ complexed cation
and the test anions was not observed, even at high loading levels
within the nitrobenzene phase. We thus infer that Cs+ is bound
strongly to the anion-bound form of calixpyrrole via an effective
combination of electronic and spatial complementarity. More-
over, since Cs+ is only minimally hydrated in water-saturated
nitrobenzene (hydration number of 0.4),16 there is little or no
dehydration penalty to pay when it is bound within the calix
like bowl.
In summary, we have demonstrated that, under appropriately
chosen conditions, even a very simple macrocycle, such as the
easy-to-prepare system meso-octamethylcalix[4]pyrrole, can act
as both a cation (Cs+) and an anion (Cl- or Br-) receptor,
extracting such species concurrently into an organic phase as
the result of ion-pairing effects. This finding has important
implications in terms of the choice of systems that may be
potentially considered as viable extractants for solvent-based
separations systems. This is because, in the limit, very different
properties might be expected for a neutral receptor that acts as
a cation extractant or one that acts as an anion extractant vs a
system that serves to facilitate the transfer of a salt from an
aqueous to an organic phase to form a combined neutral
receptor-cation-anion ensemble. A linear dependence of the
extraction efficiency on both the salt and receptor concentrations
would be expected in the case of paired anion-cation extraction.
In contrast, an extraction efficiency independent from the salt
concentration and increasing only with the square root of the
receptor concentration is characteristic of a system where the
receptor interacts exclusively with one ion and the counter ion
is fully dissociated. These stark differences make the use of
receptors, such as the calix[4]pyrrole described in this report,
that favor the concurrent, paired extraction of both anions and
cations (i.e., an overall neutral ensemble) more attractive for
the simple reason that increasing the receptor concentration will
produce an extraction efficiency that scales more strongly with
concentration. Ion-pair extractant systems also offer practical
advantages in terms of stripping, or removal of the extracted
ions, where release after extraction is important, as discussed
in greater detail elsewhere.20
In a broader context, the present results are noteworthy
because they underscore the fact that ion-pairing effects can be
substantial, even in simple systems. In particular, the fact that
ion-pairing effects are seen in the case of Cs+, a charge-diffuse
cation that would be expected to interact strongly only with
specifically tailored hosts, and in a solvent (nitrobenzene) that
favors the formation of charge separated species, leads us to
propose that they may be more prevalent in other milieus than
hitherto considered, including in the all-important realm of
biological anion and cation recognition. Efforts are therefore
underway to explore the generality of the present findings and
to test the extent to which hitherto unrecognized ion-pairing
effects might be controlling the recognition characteristics of
other “pure” cation or anion receptor systems.
Acknowledgment. The research at ORNL was sponsored by
the Division of Chemical Sciences, Geosciences, and Bio-
sciences, Office of Basic Energy Sciences, U.S. Department of
Energy, under contract DE-AC05-00OR22725 with Oak Ridge
National Laboratory, managed and operated by UT-Battelle,
LLC. The work at UT Austin was supported by the National
Institutes of Health (grant no. GM 58907 to J.L.S.). We thank
(17) Danesi, P. R.; Meider-Gorican, H.; Chiarizia, R.; Scibona, G. J. Inorg.
Nucl. Chem. 1975, 37, 1479-1483.
(18) (a) Cox, B. G.; Schneider, H. P. G. Coordination and Transport Properties
of Macrocyclic Compounds in Solution; Elsevier: New York, 1992; p 200.
(b) Takeda, Y. The SolVent Extraction of Metal Ions by Crown Compounds;
Vogtle, F., Weber, E., Eds.; Topics in Current Chemistry 121; Springer-
Verlag: Berlin, 1984; pp 1-38.
(19) Sun, Y.; Chen, Z.; Cavenaugh, K. L.; Sachleben, R. A.; Moyer, B. A. J.
Phys. Chem. 1996, 100, 9500-9505.
(20) Delmau, L. H.; Bonnesen, P. V.; Moyer, B. A. Hydrometallurgy 2004, 72,
9-19.
9
4138 J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008

Calix[4]pyrrole: A New Ion-Pair Receptor
A R T I C L E S
Mr. Dustin Gross of The University of Texas at Austin for
providing the synthetic protocol used to prepare the calix[4]-
pyrrole used in this study.
interaction theory parameters used to calculate activity coef-
ficients, molecular weights of the constituents, Masson coef-
ficients of the various ions present in the systems, organic
solubility parameters (Hildebrand parameters) of the different
organic components, and Pitzer parameters for the interactions
between cations and anions. This material is available free of
charge via the Internet at http://pubs.acs.org.
Supporting Information Available: General experimental
section and Tables S1-S6, containing the following informa-
tion: partitioning energies and constants for the partitioning
between water and nitrobenzene for the different anions and
cations used in this work, radii of alkali cations, specific ion
JA7102179
9
J. AM. CHEM. SOC. VOL. 130, NO. 12, 2008 4139