Single side strapping: A new approach to fine tuning the anion recognition properties of calix[4]pyrroles

Article abstract of DOI:10.1021/ja029175u

Three calix[4]pyrroles bearing m-orcinol-derived diether straps of different lengths on one side of the tetrapyrrolic core have been synthesized and characterized. Structural information for an analogous diester bridged strapped system reported previously

Full text of DOI:10.1021/ja029175u

Published on Web 05/21/2003
Single Side Strapping: A New Approach to Fine Tuning the
Anion Recognition Properties of Calix[4]pyrroles
Chang-Hee Lee,*,† Hee-Kyung Na,† Dae-Wi Yoon,† Dong-Hoon Won,†
Won-Seob Cho,‡ Vincent M. Lynch,‡ Sergey V. Shevchuk,‡ and
Jonathan L. Sessler*,‡
Contribution from the Department of Chemistry Kangwon National UniVersity
Chun-Chon, 200-701 Korea, and Department of Chemistry and Biochemistry UniVersity of Texas
at Austin, Austin Texas 78712
Received October 31, 2002; E-mail: chhlee@kangwon.ac.kr
Abstract: Three calix[4]pyrroles bearing m-orcinol-derived diether straps of different lengths on one side
of the tetrapyrrolic core have been synthesized and characterized. Structural information for an analogous
diester bridged strapped system reported previously (Yoon, D. W.; Hwang, H.; Lee, C. H. Angew. Chem.,
Int. Ed. Engl. 2002, 41, 1757-1759) is also provided as are bromide and chloride anion affinities for all
four systems determined by Isothermal Titration Calorimetry (ITC) in acetonitrile. Although both sets of the
strapped calix[4]pyrroles displayed enhanced affinities for chloride and bromide anion, differences were
seen among the various receptors that support the conclusion that the anion binding ability of calixpyrrole-
type systems can be effectively tuned by modifying the length and nature of the bridging straps. In the
specific case of the diether systems, the largest chloride affinity was seen with the shortest strap, whereas
the largest affinity for bromide anion was recorded in the case of the longest strap. On the basis of these
findings, as well as supporting 1H NMR spectroscopic studies, it is postulated that not only cavity size per
se, but also the ability of the aryl portion of the strap to serve as a CH hydrogen bond donor site are
important in regulating the observed anion affinities.
Introduction
tions have included the synthesis of systems bearing electron
donating or withdrawing substituents in the â-pyrrolic positions
The synthesis of anion receptors possessing high affinity and
adequate selectivity for various targeted substrates represents
an ongoing challenge in the area of supramolecular chemistry.
Appreciated as being more difficult to achieve than cation
recognition, this challenge continues to attract the attention of
researchers within the molecular recognition community due
to the important role anions play in various biological systems.1
Among the various neutral anion receptors described in the
literature, calix[4]pyrroles, possess several attributes that make
them attractive. They are, for instance, easily produced, being
the product of the acid catalyzed condensation of pyrrole and
acetone (or other readily available ketones), and have been
shown to bind fluoride, chloride, and phosphate anion in organic
media. Because their anion recognition characteristics were first
reported by Sessler et al. in 1996,2 various modifications have
been made in an effort to tune the binding characteristics of
the parent, acetone-derived system 5. To date, these modifica-
6,3 covalently linked dimers,4 functionalized derivatives bearing
chromophore attached to one of the â-pyrrolic or meso-like
positions,5 as well as so-called deep cavity models (e.g., ref 7),
wherein bulky substituents are used to either lock the conforma-
tion of the inherently flexible calix[4]pyrrole core or provide
ancillary recognition motifs.6 Although remarkable enhance-
ments in the binding affinities and useful modulations of the
(2) (a) Gale, P. A.; Sessler, J. L.; Kral, V.; Lynch, V. J. Am. Chem. Soc. 1996,
118, 5140-5141. (b) Sessler, J. L.; Gale, P. A.; Genge, J. W. Chem. Eur.
J. 1998, 4, 1095-1099. (c) Sessler, J. L.; Gale, P. A. Calixpyrroles: Novel
Anion and Neutral Substrate Receptors. In The Porphyrin Handbook;
Kadish, K. M.; Smith, K. M., Guilard, R., Eds.; Academic Press: San Diego,
CA, Burlington, MA, 2000. (d) Sessler, J. L.; Anzenbacher Jr. P.; Miyaji,
H.; Jursikova, K.; Bleasdale, E. R.; Gale, P. A. Ind. Eng. Chem. Res. 2000,
39, 3471-3478. (e) Sessler, J. L.; Genge, J. W.; Gale, P. A.; Kral, V. ACS
Symp. Series. 2000, 757. (Calixarenes for Separations) 238-254. (f) Gale,
P. A.; Anzenbacher Jr. P.; Sessler, J. L. Coord. Chem. ReV. 2001, 222,
57-102. (g) Bucher, C.; Zimmerman, R. S.; Lynch, V.; Sessler, J. L. J.
Am. Chem. Soc. 2001, 123, 9716-9717.
(3) (a) Gale, P. A.; Sessler, J. L.; Allen, W. E.; Tvermoes, N. A.; Lynch, V.
Chem. Commun. 1997, 665-666. (b) Anzenbacher, P., Jr.; Try, A. C.;
Miyaji, H.; Juris´ıkova´, K.; Lynch, V. M.; Marquez, M.; Sessler, J. L. J.
Am. Chem. Soc. 2000, 122, 10 268-10 272. (c) Miyaji, H.and Sessler, J.
L. Supramolecular Chemistry 2001, 13, 661-669.
† Department of Chemistry Kangwon National University Chun-Chon.
‡ Department of Chemistry and Biochemistry University of Texas at
Austin.
(1) (a) Chakrabarti, P. J. Mol. Biol. 1993, 234, 463-482, (b) Van Kuijck, M.
A.; Van Aubel, R. A. M. H.; Busch, A. E.; Lang, F.; Russel, G. M.; Bindels,
R. J. M.; Van Os, C. H.; Deen, P. M. T. Proc. Natl. Acad. Sci. U. S. A.
1996, 93, 5401, (c) Calnan, B. J.; Tidor, B.; Biancalana, S.; Hudson, D.;
Frankel, A. D. Science 1991, 252, 1167.
(4) Sato, W.; Miyaji, H.; Sessler, J. L. Tetrahedron Lett. 2000, 41, 6731-
6736.
(5) (a)Miyaji, H.; Sato, W.; Sessler, J. L.; Lynch, V. M. Tetrahedron Lett.
2000, 41, 1369-1373. (b) Miyaji, H.; Sato, W.; Sessler, J. L. Angew. Chem.,
Int. Ed. 2000, 39, 1777-1779. (c) Anzenbacher Jr., P.; Jursikova, K.;
Sessler, J. L. J. Am. Chem. Soc. 2000, 122, 9350-9351.
9
10.1021/ja029175u CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 7301-7306
7301

A R T I C L E S
Lee et al.
inherent anion selectivities have been achieved as the result of
these efforts, further fine-tuning of these features would be
desirable.
Figure 1. Schematic representation of strapped calix[4]pyrroles showing
how the properties of the parent system can modified. Here, X- represents
an anion, G indicates an ancillary binding motif, and Probes refers to any
of a range of species that could be used to provide a readout of the putative
anion binding event.
Scheme 1. Retrosynthetic Analysis Showing Two Generalized
Approaches to Strapped Calix[4]pyrroles, Convergent and
Divergent (represented by the right and left arrows, respectively).
One way of achieving this goal, outlined in a recently
preliminary communication,7 would involve the construction of
systems wherein a bridging strap is used not only to pre-organize
the calix[4]pyrrole binding domain but also, in the limit, adjust
its inherent electronic features. In this article we report the design
and synthesis of several new calix[4]pyrrole-type anion receptors
based on this paradigm (compounds 1-3) and show how the
use of appropriately chosen, functionalized straps does indeed
allow the inherent anion affinities of calix[4]pyrroles to be
modulated significantly. Also reported is the solid-state structure
of the chloride complex of a diester strapped system, 4, reported
previously,7 as well as further details of the solution phase anion
binding characteristics of this latter first generation system.
In the design of the first generation receptor 4, a flexible strap
bearing an aromatic diester was used as the bridging element.
This strap was expected to allow for the isolation of the binding
site from solvent and encapsulation of a bound anionic substrate
target. It was also expected to provide additional hydrogen
bonding sites and thus allow for the specific modulation of the
inherent anion affinities. Although marked changes in the
chloride anion affinity relative to the parent, acetone-derived
calix[4]pyrrole 5 were observed, it was appreciated that by using
straps of different sizes or by introducing other functional groups
into the straps, further modifications in the inherent properties
of the system could be engendered, including ultimately the
production of sensor systems (Structure A, Figure 1).
Results and Discussions
In an effort to generate strapped systems represented by
generalized structure C in Scheme 1 (a schematic version of
structure A in Figure 1), two limiting approaches, convergent
and divergent, can be easily conceived. There are pros and cons
associated with each one. In principle, the coupling of a calix-
[4]pyrrole such as B with a functionalized strap should give
rise to the desired product C. However, the synthesis of
precursors represented by structure B is challenging since it
would likely require the generation and separation of isomeric
mixtures regardless of the specific synthetic route chosen (e.g.,
condensation of pyrrole with two different ketones or condensing
a preformed dipyrromethane with a ketone). Confounding the
problem is the fact that isolating and assigning the structure of
the desired cis, as opposed to trans, configuration in a product
such as B might not be easy. Given this, we considered that a
convergent approach, involving the intramolecular cyclization
of a precursor such as D with a ketone, would be more
efficacious. For the present study, we chose m-orcinol (3,5-
dihydroxytoluene) as the “keystone” of the strap as shown in
Scheme 2.
On a different level, it was recognized that by varying the
nature of the strap, fundamental insights into the relationship
between receptor structure, anion size and shape, and binding
affinities might be obtained. It was in an effort to address this
latter issue that the present study was undertaken. Specifically,
we sought to address how the use of phenyl-containing, ether
bridged straps of varying lengths would effect the chloride and
bromide anion binding properties of calix[4]pyrrole relative both
to the parent system 5 and the ester strapped material 4 whose
synthesis and binding properties were recently described.7
This choice represented a balance between precursor avail-
ability and the desire to have a strap that would enhance, or at
least not diminish, the solubility of the resulting product in
solvents such as dichloromethane and acetonitrile that have been
traditionally used to study calix[4]pyrroles. We were also keen
to explore whether CH hydrogen bonding interactions, consid-
ered to be important in the case of 4,6 could play a role in
regulating the halide anion affinities of the resulting products
1-3.
(6) (a) Anzenbacher, Jr., P.; Jursikova, K.; Lynch, V. M.; Gale, P. A.; Sessler,
J. L. J. Am. Chem. Soc. 1999, 121, 11 020-11 021. (b) Camiolo, S.; Gale,
P. A. Chem. Comm. 2000, 1129-1130.
(7) Yoon, D. W.; Hwang, H.; Lee, C. H. Angew. Chem., Int. Ed. Engl. 2002,
41, 1757-1759.
9
7302 J. AM. CHEM. SOC. VOL. 125, NO. 24, 2003

Single Side Strapping of Calix[4]pyrroles
A R T I C L E S
Scheme 2
The requisite precursor bromoketones, 8-10 were synthesized
in accord with literature procedures,8 and were subsequently
condensed with m-orcinol in DMF/K2CO3 to give the corre-
sponding diketones 12-14 in high yields. Treatment of these
diketones 12-14 with an of excess pyrrole in the presence of
1 equivalent of trifluoroacetic acid then afforded the corre-
sponding bisdipyrromethanes 15-17 as one of several conden-
sation products. Although the yields were moderate, purification
and isolation of the bisdipyrromethanes proved straightforward,
requiring simply column chromatography over silica gel.
Figure 2. Ortep (left) and CPK (right) representations of the chloride
complex of receptor (4) deduced from an X-ray diffraction analysis.
Hydrogen atoms and the countercation (triethylammonium ion) have been
omitted for clarity. The complex adopts a cone conformation and clearly
shows a hydrogen bond interaction between Ar-H and Cl- (CsCl distance
) 3.793 Å).
With the bisdipyrromethanes 15-17 in hand, the desired
strapped calix[4]pyrroles 1-3 were obtained in yields of 8-11%
by condensing with acetone in the presence of a catalytic amount
of BF3. Proton NMR spectra of all three new receptors were
consistent with the proposed structure and symmetry, with the
aromatic protons, in particular, being sufficiently well resolved
to permit a straightforward first-order structural analysis.
Attempts to study the anion binding properties of systems
1-3 were carried out in CD3CN using proton NMR spectros-
copy. No quantitative estimates of anion affinities could be made
on the basis of these studies due to the fact that extremely strong
binding with slow complexation/decomplexation kinetics was
observed. On the other hand, these spectroscopic studies did
provide good qualitative evidence for halide anion binding. For
example, titration of receptor 2 with chloride anion (studied in
the form of its tetrabutylammonium salt) gave rise to a
completely new set of signals, including ones for both (i) the
pyrrole NH protons at 11.11 ppm that were shifted to dramati-
cally lower field than what was observed in the absence of
anions (δ ) 8.11 ppm), (ii) the central aromatic CH proton, a
singlet at 6.27 ppm was shifted to lower field at 6.78 ppm, and
(iii) the methylene protons adjacent to the phenolic oxygen that
were shifted from 3.92 to 4.11 ppm. On the other hand, the
â-pyrrolic protons, which appeared originally at 5.78 ppm, were
found to be shifted upfield to 5.51 ppm as expected. Smaller
upfield shifts (ca. ∼0.05 ppm) were also observed in the case
of the two upper aromatic CH signals. This appearance of
completely new peaks, coupled with a corresponding disap-
pearance in the original signals, was fully complete by the time
ca. ∼1 equivalent of chloride anion had been added. Similar
trends were observed with the other two new receptors, 1 and
3, as well as when bromide anion was substituted for chloride
anion. Taken together, these observations are consistent with a
Efforts to obtain single crystals of 1-3, or their halide anion
complexes, suitable for X-ray diffraction analysis proved
unsuccessful. However, such crystals were obtained in the case
4 when crystallization was carried out in the presence of excess
triethylammonium chloride. The resulting structure, shown in
Figure 2, reveals a cone-like conformation for the calix[4]pyrrole
core and the presence of a chloride anion encapsulated within
the 3-dimensional binding cavity. The distance between the
central phenyl carbon atom and the chloride anion is 3.793 Å,
a finding that is consistent with the presence of a strong aryl
CH- - -Cl- hydrogen bond. The average distance between the
pyrrolic nitrogens and the bound chloride anion center is
3.285 ( 0.0005 Å. The phenyl group of the strap is tilted off
the perpendicular drawn through the RMS center of the four
calix[4]pyrrole nitrogen atoms by about 70° as the result,
perhaps, of a need to accommodate an interaction between the
chloride anion and the countercation (not shown in Figure 1).
Despite this deviation, it is clear that the use of a 5 atom linker
between the meso-like positions of an otherwise normal calix-
[4]pyrrole core and a capping tolyl group provides a cavity that
is suitable for anion recognition, at least in the solid state.
(8) Zhang, W.; Li, C. J. Org. Chem. 2000, 65, 5831-5833.
9
J. AM. CHEM. SOC. VOL. 125, NO. 24, 2003 7303

A R T I C L E S
Lee et al.
Table 2. Themodynamic Parameters Determined by ITC for the
Table 1. Association Constants for the Binding of Chloride and
Bromide by Compound 1-6 as Measured by ITC (isothermal
titration calorimetry) at 30 °C Using the Corresponding
Alkylammonium and Cryptand Saltsf
Binding of Various Anion Salts by the Indicated Receptors in Dry
Acetonitrile at 30 °Ca
host
guest
[H] (mM)
[G] (mM)
∆H
T∆S
∆G
TBA−Cl
K−Cryptand−Cl
TBA−Br
TBA−Cl
1
1
2
2
3
3
3
3
4
4
6
6
TBA-Cl
0.12
0.24
0.23
0.23
0.22
0.23
0.45
0.23
0.19
0.19
0.26
0.25
1.64
2.43
3.03
2.43
3.06
2.43
1.77
3.53
2.99
2.45
3.31
2.91
-12.65 -3.55 -9.10
-11.27 -2.07 -9.20
CH CN
3
DMSO
Cryptand-Cl
TBA-Cl
-7.43
-7.08
-7.73
-7.53
1.08 -8.51
1.60 -8.68
0.77 -8.50
1.14 -8.67
1
2
3
4
5
5
5
5
6
3 630 000
1 370 000
1 370 000
1 380 000
96 000b
4 270 000
1 810 000
1 810 000
1 620 000
190 000b
30 000
31 000
120 000
∼0
340 000
11 000
7400
Cryptand-Cl
TBA-Cl
Cryptand-Cl
TBA-Br
110 000a
-7.46 -0.45 -7.01
2 800b
3 400
-6.88
0.09 -6.97
140 000
1300
TBA-Cl
-9.17 -0.65 -8.52
-8.83 -0.22 -8.61
55 000c
Cryptand-Cl
TBA-Cl
5000d
1025e
1500
-7.78
-7.13
0.16 -7.94
1.03 -5.45
530 000
760 000
8500
Cryptand-Cl
a Data from Ref 7. bMeasured using tetraethylammonium as the counter
cation; data from ref 9. cMeasured in acetonitrile-d3 (0.5% v/v D2O) at 22
°C; association constant determined by ITC. dData from ref 3b; association
constant determined by 1H NMR spectroscopic titration. eData from ref
6b; association constant determined by 1H NMR spectroscopic titration.
The CH3CN was rigorously dried simply by passage over two columns of
activated molecular sieves using a solvent dispensing system ([H2O] < 10
ppm). The DMSO was spectral grade (Aldrich) but not otherwise dried.
Estimated error <10%. f TBA ) tetrabutylammonium; cryptand ) [2,2,2]-
cryptand.
a The error for the fitting is estimated to be < 0.08 kcal/mol.
drying. Inspection of this table underscores the obvious advan-
tage that accrues as the result of using a strap. Even as compared
to the previous “best” calix[4]pyrrole system, the octafluoro
derivative 6, a large increase in association constants is
engendered by the use of a “strap”.11 In the case of the
C-4(ether) 1, where the effect is most dramatic, an enhancement
factor of almost 7 relative to this earlier benchmark is achieved
in the case of chloride in acetonitrile. This means that well over
an order in magnitude in Ka increase has been engendered
relative to the parent system 5. In DMSO, the effect appears to
be even more dramatic, as a comparison of the Ka values for
the C-4(ether) 1 and normal calix[4]pyrrole 5 will confirm.
Although for the most part, the effect of the strap appears to
be a “simple” increase in anion affinity, in the case of the C-6
(ether) strap system 3, an appreciable increase in the bromide
anion affinity relative to the chloride anion affinity appears to
have been achieved. Presumably, this result reflects the more
open binding cavity present in this latter system and the fact
that its size better complements that of the larger bromide anion.
Although further studies, involving the synthesis and analysis
of other systems, will be required to substantiate such a
hypothesis, to the extent it stands, it leads to the suggestion
that the use of straps to define further the size and nature of the
binding pockets in elaborated calix[4]pyrroles provides a way
of fine-tuning both anion affinities and selectivities.
strong 1:1 binding motif and a rate of complexation-decom-
plexation that is slow on the NMR time scale.
Quantitative analyses of the solution phase chloride and
bromide anion binding properties of strapped compounds 1-3,
as well as the unstrapped “control” systems octamethylcalix-
[4]pyrrole 5 and octafluorocalix[4]pyrrole 6, were made using
isothermal titration calorimetry (ITC). The advantages of this
method for studying the anion recognition properties of the
parent calix[4]pyrrole 3 in acetonitrile, particularly in the case
when organic soluble salts of chloride and bromide are used
for analysis, have recently been presented by Schmidtchen.9 In
the case of Ka values that are on the order of 103 M-1 or less,
we have found that ITC gives Ka values that are concordant
with those obtained using more classic 1H NMR spectroscopic
titration methods. However, we have found that for Ka values
larger than this, ITC routinely gives rise to calculated affinity
constants that are larger, often substantially larger, than those
determined by 1H NMR spectroscopic means.10 We ascribe this
difference to limitations in the range of useful affinities that
1
Table 2 shows the thermodynamic parameters for selected
compounds involved in the chloride anion-receptor binding
process. An interesting conclusion that results from an inspection
of this table is that changing the countercation from an
alkylammonium to a potassium cryptand[2,2,2] complex did not
change the experimental enthalpy significantly in the case of
strapped calixpyrroles 1-3. Such an observation is consistent
with the two cations in question acting as essentially innocent
may be determined by H NMR spectroscopic means, rather
than to inherent problems associated with ITC as applied to
anion recognition. Indeed, we have found similar disparities in
calculated affinity constants when fluorescence-based, rather
1
than H NMR spectroscopic, methods are used to determine
large (i.e., > 103 M-1) Ka values.5,6a Moreover, we have found
that our own ITC measurements involving simple calix[4]pyrrole
gave rise to Ka values that are coincident with those of
Schmidtchen when essentially identical experimental protocols
were used (cf., Table 1). We thus feel confident that ITC permits
a relatively precise comparison between the compounds of this
study, even though the Ka values in question are exceptionally
high.
(11) Attempts to estimate the association constants of the strapped systems by
1H NMR methods using competition experiments proved unsuccessful.
Under conditions where an equimolar amount of the receptor 1 and
octafluorocalix[4]pyrrole 6 are mixed with a stiochiometric limiting amount
of TBACl (0.7-1.2 equivalents) in CD3CN, such experiments indicate that
there is almost no change is observed in the chemical shift of the NH protons
for 6, whereas a completely new sets of signals were observed for the NH
signals of 1. Upon adding >1.1 equiv of the anion, shifts in the resonance
ascribed to 6 began to be observed. Although such observations provide
important qualitative support for the notion that 6 is a stronger chloride
anion receptor than 1 under these experimental conditions, the large
differences in apparent affinity constants, coupled with the slow nature of
the exchange observed in the case of 6, precludes calculation of an
equilibrium constant for the exchange of chloride anion between 1 and 6.
This, in turn, makes it impossible to calculate a competition-derived Ka
value for 6.
Table 1 summarizes the equilibrium association constants
measured by ITC for the binding of chloride and bromide anion
to various calix[4]pyrroles as determined in dry acetonitrile or
in spectral grade DMSO that was not subject to any special
(9) Schmidtchen F. P. Org. Lett. 2002, 4, 431-434.
(10) Sessler, J. L.; An, D.; Cho, W.-S.; Lynch, V. Angew. Chem., in press.
9
7304 J. AM. CHEM. SOC. VOL. 125, NO. 24, 2003

Single Side Strapping of Calix[4]pyrroles
A R T I C L E S
observers.9 Another noteworthy feature highlighted by this table
is that receptors 4 and 1 display the largest enthalpy changes,
possibly due to the presence of an additional hydrogen bonding
motif provided by the CH proton of the bridging phenyl ring.
On the other hand, these species, wherein binding presumably
most strongly benefits from a cryptand effect, are characterized
by the most negative entropic terms. The resulting enthalpy-
entropy compensation, which appears to be structure, as opposed
to solvent,9 based serves to reduce somewhat the affinity values
of 1 and 4 below what might expect based on an extrapolation
of those measured for 2 and 3. Nonetheless, it is important to
appreciate that system 1, in particular, displays an absolute
chloride anion affinity that is remarkably high, exceeding by a
large margin the value seen for any other neutral pyrrole-based
anion binding receptor.
evaporated under reduced pressure. The remaining solid was purified
by column chromatography on silica gel (eluent: CH2Cl2/EtOAc )
19/1). Collection of the appropriate fraction and drying produced 12
1
in the form of a yellow solid (1.44 g; 83%). Mp 42-44 °C; H NMR
(CDCl3) δ 1.74-1.77 (m, 8H, CH2), 2.15 (s, 6H, CH3), 2.28 (s, 3H,
tolyl), 2.51 (m, 4H, CH2), 3.92(t, 4H, J ) 5.79 Hz, OCH2), 6.24 (m,
1H, Ar-H), 6.30(m, 2H, Ar-H). 13C NMR (CDCl3), δ 208.72, 160.02,
140.13, 107.65, 98.44, 67.41, 43.24, 29.92, 28.67, 21.79, 20.46. HRMS
(CI+): Calcd for C19H28O4 321.206585, Found 321.207465.
7-[3-Methyl-5-(6-oxo-heptyloxy)phenoxy]-2-heptanone (13). Or-
cinol (0.1 g, 0.81 mmol), K2CO3 (1.1 g, 8.05 mmol), and compound 9
(0.78 g, 4.03 mmol) were subject to reaction under conditions identical
to those used for the synthesis of 12. Column chromatography over
silica gel (eluent: CH2Cl2/EtOAc ) 19/1) gave product 13 in 78%
1
yield (0.22 g). Mp 48-49.5 °C; H NMR (CDCl3) δ 1.42-1.49 (m,
4H, CH2), 1.60-1.68 (m, 4H, CH2), 1.73-1.80 (m, 4H, CH2), 2.14 (s,
6H, CH2), 2.28 (s, 3H, tolyl-H), 2.46 (m, 4H, CH2), 3.91 (t, 4H, J )
6.40 Hz, OCH2), 6.25 (m, 1H, Ar-H), 6.30(m, 2H, Ar-H). 13C NMR
(CDCl3), δ 209.38, 160.50, 140.49, 108.01, 98.80, 67.94, 44.01, 30.33,
29.49, 26.09, 23.90, 22.18. HRMS (CI+): Calcd for C21H32O4
349.237885, Found 349.238106.
Conclusions
In summary, we have demonstrated that by strapping the
calix[4]pyrrole core, substantial increases in chloride and
bromide binding affinity in acetonitrile and DMSO may be
achieved. This approach also appears to allow for a modification
of the inherent anion selectivity of calix[4]pyrroles, in that the
larger strapped system show an increased affinity for bromide
relative to chloride, albeit at an absolute affinity value that is
still substantially reduced as compared to this latter anion. A
further benefit of the present strapping strategy is that it should
be amenable to future modification through, e.g., the introduction
of electron withdrawing atoms on the â-pyrrolic positions or
the use of straps that contain ancillary hydrogen bond donor
groups. Possibilities involving the use of straps containing built-
in chromophores are also attractive and may allow for the
construction of new sensors that exhibit dynamic ranges or
inherent selectivities that are superior to existing systems. Also
attractive is the use of modified strapped systems to create clefts
that might allow for the specific targeting of nonspherical anions
such as, e.g., oxalate or pyrophosphate, that have obvious
biological importance. Work along these lines is currently in
progress.
7-[3-Methyl-5-(6-oxo-heptyloxy)phenoxy]-2-heptanone (14). Or-
cinol (240 mg, 1.7 mmol), K2CO3 (2.4 g, 17 mmol), and compound 10
(1.06 g, 5.1 mmol) were subject to reaction under conditions identical
to those used for the synthesis of 12. Column chromatography over
silica gel (eluent: CH2Cl2/EtOAc ) 19/1), gave product 14 in 63%
yield (0.41 g). Mp 40-41 °C; 1H NMR (CDCl3) δ 6.31 (s, 2H, ArH),
6.26 (s, 1H, ArH), 3.91 (t, 4H, J ) 6.4 Hz, CH2), 2.44 (t, 4H, J ) 7.4
Hz, CH2), 2.28 (s, 3H, ArCH3), 2.14 (s, 6H, meso-CH3), 1.79-1.72
(m, 4H, CH2), 1.64-1.57 (m, 4H, CH2), 1.50-1.42 (m, 4H, CH2),
1.39-1.31 (m, 4H, CH2). 13C NMR (CDCl3), δ 209.54, 160.55, 140.46,
107.98, 98.81, 68.10, 44.05, 30.29, 29.52, 29.30, 26.29, 24.12, 22.19.
HRMS (CI+): Calcd for C23H36O4 377.269185, Found 377.269217.
1,3-Bis[5,5-di(pyrrol-2-yl)hexyloxy]-5-methylbenzene (15). To a
solution of pyrrole (4.3 mL, 62.4 mmol) and compound 12 (0.5 g, 1.56
mmol) was added TFA (120 µL, 1.56 mmol) and the solution was
stirred for 10 min at 60 °C. Then the reaction mixture was combined
with aqueous NaOH (0.1 N, 20 mL) and extracted with CH2Cl2. The
organic layer was dried over Na2SO4 and the solvent was removed in
vaccuo. The remaining oily solid was purified by column chromatog-
raphy over silica gel (eluent: CH2Cl2) to give 0.5 g of 15 (58% yield).
This product was then used directly to the next step without further
purification.
Experimental Section
1,3-Bis[6,6-di(pyrrol-2-yl)heptyloxy]-5-methylbenzene (16). Com-
pound 13 (1.0 g, 2.87 mmol), pyrrole (4 mL, 57.4 mmol), and TFA
(0.22 mL, 2.87 mmol) were reacted in accord with the procedure used
in synthesis of 15. Column chromatography over silica gel (eluent: CH2-
Cl2) gave product 16 in 60% yield (1.03 g, oily solid); 1H NMR (CDCl3)
δ 1.23-0.29 (m, 5H, CH2), 1.38-1.43 (m, 4H, CH2), 1.58 (s, 6H, CH3),
1.67-1.73 (m, 4H, CH2), 1.96-2.00 (m, 4H, CH2), 2.27 (s, 3H, CH3),
3.86 (t, 4H, J ) 6.42 Hz, CH2), 6.07-6.08 (m, 4H, pyrrole-H), 6.12-
6.14 (m, 4H, pyrrole-H), 6.22 (m, 1H, Ar-H), 6.28 (m, 2H, Ar-H),
6.62-6.64 (m, 4H, pyrrole-H), 7.77 (brs, 4H, NH). FAB MS Calcd
for C37H48N4O2 580.38, Found 580.30.
Proton NMR spectra (400 MHz, Bruker IFS 48) were recorded in
CDCl3 using TMS as the internal standard. High and Low resolution
FAB mass spectra were obtained on AUTO SPEC M-363 high-
resolution mass spectrometer. Column chromatography was performed
over silica gel (Merck, 230-400 mesh). Pyrrole was distilled at
atmospheric pressure from CaH2. Both CH2Cl2 and CHCl3 (reagent
grade) were distilled from K2CO3 to eliminate traces of acid. All other
reagents were obtained from Aldrich and used as received unless noted
otherwise. Compounds 8-10 were prepared by a Retro-Barbier
procedure.8 The ester-strapped receptor 4 was prepared using the
procedure reported recently.7 The acetonitrile used for the ITC studies
was dried simply by passage over two columns of activated molecular
sieves using a solvent dispensing system designed by J. C. Meyer
(“[H2O] < 10 ppm”), whereas the DMSO was spectral grade (Aldrich)
but used without further purification.
1,3-Bis[7,7-di(pyrrol-2-yl)octyloxy]-5-methylbenzene (17). Com-
pound 14 (0.40 g, 1.06 mmol), pyrrole (2.9 mL, 42.5 mmol) and TFA
(82 µL, 1.06 mmol) were reacted in accord with the procedure used in
synthesis of 15. Column chromatography over silica gel (eluent: CH2-
1
Cl2) gave desired product in 52% yield (0.33 g, oily solid). H NMR
6-[3-Methyl-5-(5-oxo-hexyloxy)phenoxy]-2-hexanone (12). A solu-
tion of DMF (150 mL), K2CO3 (7.46 g, 54.0 mmol), and orcinol (11,
0.67 g, 5.4 mmol) was stirred for 10 min at room temperature and
then compound 8 (4.8 g, 27.0 mmol) was added. The whole mixture
was heated to 60 °C and stirred for 30 min. The mixture was combined
with water (20 mL) and extracted with methylene chloride (2 × 50
mL). The organic layer was dried (Na2SO4) and the solvent was
(CDCl3) δ 7.75 (br s, 4H, NH), 6.63 (m, 4H, pyrrole-H), 6.29 (s, 2H,
Ar-H), 6.24 (s, 1H, Ar-H), 6.14-6.12 (m, 4H, pyrrole-H), 6.07 (m,
4H, pyrrole-H), 3.87 (t, 4H, J ) 6.40 Hz, CH2), 2.27 (s, 3H, Ar-
CH3), 1.98-1.94 (m, 4H, CH2), 1.74-1.67 (m, 4H, CH2), 1.58 (s, 6H,
CH3), 1.45-1.37 (m, 4H, CH2), 1.35-1.22 (m, 8H, CH2). CI-MS
Calcd for C39H52N4O2 608.41, Found 609 (M++1).
9
J. AM. CHEM. SOC. VOL. 125, NO. 24, 2003 7305

A R T I C L E S
Lee et al.
C4-Strapped calix[4]pyrrole (1). To a solution of 15 (0.5 g, 0.9
mmol) and acetone (100 mL) was added BF3‚OEt2 (34 µL, 0.26 mmol).
The resulting solution was stirred for 20 min at room temperature before
being combined with aqueous NaOH (0.1 N, 20 mL). The mixture
obtained as the result of this addition was extracted twice with CH2Cl2
and the organic layer was dried over Na2SO4. The solvent was removed
in vacuo and the resulting solid was purified by column chromatography
over silica gel (eluent: CH2Cl2). Recrystallization from methanol
afforded product 1 in pure form. Yield: 0.055 g (10%). Mp 213 °C
decomp.; 1H NMR (CDCl3) δ 1.38-1.44 (m, 4H, CH2), 1.47 (m, 12H,
CH2), 1.54(s, 6H, CH3) 1.73(m, 4H, CH2), 1.91 (m, 4H, CH2), 2.34 (s,
3H, tolyl), 4.04 (t, 4H, J ) 5.50 Hz, OCH2), 5.83 (m, 4H, pyrrole-H),
5.88 (m, 4H, pyrrole-H), 6.27 (m, 1H, Ar-H), 6.36 (m, 2H, Ar-H)
7.07 (br s, 4H, NH). 13C NMR (CDCl3), δ 160.16, 140.28, 138.07,
136.25, 107.35, 104.34, 102.83, 101.75, 67.80, 40.83, 39.14, 35.42,
30.98, 28.77, 28.46, 27.38, 22.14, 22.03. FAB-MS Calcd for C41H52N4O2
632.41, Found 632.59 (M+).
X-ray Experimental. Crystal structure analyses were measured on
a Nonius Kappa CCD diffractometer using a graphite monochromator
with MoKR radiation (λ ) 0.71073 Å). The data were collected at 153K
using an Oxford Cryostream low-temperature device. Data reduction
were performed using DENZO-SMN.12 The structure was solved by
direct methods using SIR9213 and refined by full-matrix least-squares
on F2 with anisotropic displacement parameters for the non-H atoms
using SHELXL-97.14 The hydrogen atoms were calculated in ideal
positions with isotropic displacement parameters set to 1.2 × Ueq of
the attached atom (1.5 × Ueq for methyl hydrogen atoms). The
hydrogen atoms bound to nitrogen were located in a ∆F map and refined
2
2 2
with isotropic displacement parameters. The function, Σw(|Fo| - |Fc| ) ,
was minimized, where w ) 1/[(σ(Fo))2 + (0.0317P)2 + (2.8372P)]
2
2
2
2
and P ) (|Fo| + 2|Fc| )/3. Rw(F2) ) {Σw(|Fo| - |Fc| )2/Σw(|Fo|)4}1/2
where w is the weight given each reflection. R(F) ) Σ(|Fo| - |Fc|)/
Σ|Fo|} for reflections with Fo > 4(σ(Fo)). S ) [Σ w(|Fo| - |Fc| )2/
(n - p)]1/2, where n is the number of reflections, and p is the number
of refined parameters. The data were corrected for secondary extinction
effects. The correction takes the form: Fcorr ) kFc/[1 + (3.1(6) ×
10-6)Fc2 λ3/(sin2θ)]0.25 where k is the overall scale factor. Neutral atom
scattering factors and values used to calculate the linear absorption
coefficient are from the International Tables for X-ray Crystallography
(1992).13
2
2
C5-strapped Calix[4]pyrrole (2). Compound 16 (0.2 g, 0.344
mmol), acetone (34 mL), and BF3‚OEt2 (12 µL, 0.1 mmol) were reacted
in accord with the procedure used for the synthesis of 1. Column
chromatography over silica gel (eluent: CH2Cl2) and recrystallization
from methanol gave the strapped calixpyrrole product (0.025 g; 11%)
in pure form. Mp 205 °C decomp.; 1H NMR (CDCl3) δ 1.22-1.36 (m,
6H, CH2), 1.46 (m, 12H, CH2), 1.49-1.50 (m, 16H, CH2), 1.76-1.83
(m, 8H, CH2), 2.33 (s, 3H, tolyl), 3.99 (t, 4H, J ) 6.11 Hz, OCH2),
5.84(m, 4H, pyrrole-H), 5.88 (m, 4H, pyrrole-H), 6.30 (s, 1H, Ar-H),
6.39 (s, 2H, Ar-H), 7.32 (br s, 4H, NH). 13C NMR (CDCl3), δ 160.27,
140.09, 138.16, 136.34, 107.75, 104.13, 102.69, 99.17, 67.73, 41.92,
39.19, 35.38, 31.18, 29.71, 28.78, 28.26, 27.25, 26.59, 24.92, 21.87.
FAB-MS Calcd for C43H56N4O2 660.44, Found 660.47 (M+).
C6-Strapped Calix[4]pyrrole (3). Compound 17 (0.33 g, 0.542
mmol), acetone (50 mL) and BF3‚OEt2 (21 µL, 0.16 mmol) were reacted
in accord with the procedure used for the synthesis of 1. Column
chromatography over silica gel (eluent: CH2Cl2) and recrystallization
from methanol gave the strapped calixpyrrole product in 8% yield (30
4‚(C2H5)3NHCl. C46H62N5O4Cl; Crystals grew as colorless prisms
and lathes. The data crystal was cut from a long needle and had
approximate dimensions 0.20 × 0.15 × 0.11 mm; triclinic, space group
P-1, a ) 9.5568(2) Å, b ) 11.0858(3) Å, c ) 20.6450(6) Å, R )
76.832(1)°, â ) 77.413(1)°, γ ) 84.170(1)°, V ) 2075.43(9) Å3, Z )
2, Fcalc ) 1.255 gcm-3, µ ) 0.142 mm-1, F(000) ) 844; a total of 461
frames of data were collected using ω-scans with a scan range of 1°
and a counting time of 123 s per frame. A total of 15232 were measured,
9255 unique (Rint ) 0.0812). The structure was refined on F2 to 0.156,
with R(F) equal to 0.0857 and a goodness of fit, S, ) 0.994.
1
mg). Mp 204 °C decomp.; H NMR (CDCl3) δ 7.13 (br s, 4H, NH),
Acknowledgment. This work was supported by a grant (R05-
2000-000-00077-0) from the Basic research program of the
Korea Science and Engineering Foundation and by the U.S.
National Institutes of Health (Grant No. GM 58907 to J.L.S.).
The Vascular System Research Center at KNU is acknowledged
for financial support.
6.39 (s, 1H, Ar-H), 6.37 (s, 2H, Ar-H), 5.86 (m, 4H, pyrrole-H),
5.85 (m, 4H, pyrrole-H), 4.01 (t, 4H, J ) 5.7 Hz, CH2), 2.33 (s, 3H,
Ar-CH3), 1.84-1.80 (m, 4H, CH2), 1.76-1.70 (m, 4H, CH2), 1.55 (s,
6H, CH3), 1.55-1.47 (m, 4H, CH2), 1.47 (s, 6H, CH3), 1.44 (s, 6H,
CH3), 1.40-1.33 (m, 4H, CH2), 1.24-1.19 (m, 4H, CH2). 13C NMR
(CDCl3), δ 160.32, 140.09, 138.16, 137.28, 106.74, 103.85, 102.83,
100.74, 67.41, 39.93, 38.86, 35.45, 35.31, 30.82, 28.15, 27.67, 27.63,
26.91, 24.57, 23.36, 22.05. HRMS (CI+) Calcd for C45H60N4O2
689.479453, Found 689.478151.
ITC and 1H NMR Spectroscopic Binding Studies. Isothermal
Titration Calorimetery (ITC) measurements were performed as fol-
lows: Solutions of the chosen receptor in rigorously dry acetonitrile
or spectral grade DMSO were made up so as to provide a receptor
concentration range of 0.1-1 mM. These solutions were then individu-
ally titrated with the appropriate alkylammonium or potassium cryptand
salts at 30 ( 0.01 °C, unless indicated otherwise. The original heat
pulses were normalized using reference titrations carried out using the
same salt solution but pure solvent, as opposed to a solution containing
the receptor. The values recorded in Table 1 are the average result of
at three separate titrations carried out at least two different concentra-
tions.
Supporting Information Available: Spectra, tables of ex-
perimental data, ITC plots of 1-6, and an X-ray crystallographic
data file. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA029175U
(12) Carter, Jr., C. W., Sweet, R. M., Eds. Macromolecular Crystallography,
Part A. [In: Methods Enzymol. 1997, 276].
(13) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. J. Appl.
Crystallogr. 1993, 26, 343-350.
(14) Sheldrick, G. M. SHELX 97. Program for the Refinement of Crystal
Structures; University of Gottingen: Germany, 1994, 13. Kasper, J. S.,
Lonsdale, K., Eds. International Tables for X-ray Crystallography. Vol.
C, Tables 4.2.6.8 and 6.1.1.4; Kluwer Academic Press: Boston, 1992.
9
7306 J. AM. CHEM. SOC. VOL. 125, NO. 24, 2003