Induction-driven stabilization of the anion-π interaction in electron-rich aromatics as the key to fluoride inclusion in imidazolium-cage receptors

Article abstract of DOI:10.1002/chem.201002105

Intermolecular interactions that involve aromatic rings are key processes in both chemical and biological recognition. It is common knowledge that the existence of anion-π interactions between anions and electron-deficient (π-acidic) aromatics indicates that electron-rich (π-basic) aromatics are expected to be repulsive to anions due to their electron-donating character. Here we report the first concrete theoretical and experimental evidence of the anion-π interaction between electron-rich alkylbenzene rings and a fluoride ion in CH₃CN. The cyclophane cavity bridged with three naphthoimidazolium groups selectively complexes a fluoride ion by means of a combination of anion-π interactions and (Ci?￡?H) ⁺?F^--type ionic hydrogen bonds. ¹H NMR, ¹⁹F NMR, and fluorescence spectra of 1 and 2 with fluoride ions are examined to show that only 2 can host a fluoride ion in the cavity between two alkylbenzene rings to form a sandwich complex. In addition, the cage compounds can serve as highly selective and ratiometric fluorescent sensors for a fluoride ion. With the addition of 1 equiv of F^-, a strongly increased fluorescence emission centered at 385 nm appears at the expense of the fluorescence emission of 2 centered at 474 nm. Finally, isothermal titration calorimetry (ITC) experiments were performed to obtain the binding constants of the compounds 1 and 2 with F^- as well as Gibbs free energy. The 2-F^- complex is more stable than the 1-F^- complex by 1.87 kcal mol^-1, which is attributable to the stronger anion-π interaction between F^- and triethylbenzene.

Full text of DOI:10.1002/chem.201002105

FULL PAPER
DOI: 10.1002/chem.201002105
Induction-Driven Stabilization of the Anion–p Interaction in Electron-Rich
Aromatics as the Key to Fluoride Inclusion in Imidazolium-Cage ACTHUNRGTNERNUG eceptors
Zhaochao Xu,[a, b, d] N. Jiten Singh,[c, d] Sook Kyung Kim,[a] David R. Spring,*[b]
Kwang S. Kim,*[c] and Juyoung Yoon*[a]
Abstract: Intermolecular interactions
that involve aromatic rings are key pro-
cesses in both chemical and biological
recognition. It is common knowledge
that the existence of anion–p interac-
tions between anions and electron-defi-
cient (p-acidic) aromatics indicates that
electron-rich (p-basic) aromatics are
expected to be repulsive to anions due
to their electron-donating character.
Here we report the first concrete theo-
retical and experimental evidence of
the anion–p interaction between elec-
tron-rich alkylbenzene rings and a fluo-
ride ion in CH3CN. The cyclophane
cavity bridged with three naphthoimi-
dazolium groups selectively complexes
a fluoride ion by means of a combina-
lective and ratiometric fluorescent sen-
sors for a fluoride ion. With the addi-
tion of 1 equiv of FÀ, a strongly in-
creased fluorescence emission centered
at 385 nm appears at the expense of
the fluorescence emission of 2 centered
at 474 nm. Finally, isothermal titration
calorimetry (ITC) experiments were
performed to obtain the binding con-
stants of the compounds 1 and 2 with
FÀ as well as Gibbs free energy. The 2-
FÀ complex is more stable than the 1-
FÀ complex by 1.87 kcalmolÀ1, which is
attributable to the stronger anion–p in-
teraction between FÀ and triethylben-
zene.
À
tion of anion–p interactions and (C
H)+···FÀ-type ionic hydrogen bonds.
1H NMR, 19F NMR, and fluorescence
spectra of 1 and 2 with fluoride ions
are examined to show that only 2 can
host a fluoride ion in the cavity be-
tween two alkylbenzene rings to form a
sandwich complex. In addition, the
cage compounds can serve as highly se-
Keywords: anions
·
electron-rich
aromatics · fluorides · hydrogen
bonds · pi interactions
Introduction
Anion–p interactions have become an actively investigated
branch of supramolecular chemistry in recent years.[1–5]
Schneider first used the term ꢀanion–p interaction,ꢁ which
has pioneered this new field.[6] The study of anion–p interac-
tions gained the momentum to grow as a new research field
with the initial ab initio studies of Frontera, Deyꢂ et al.,[7]
Mascal et al.,[8] and Alkorta et al.[9] These theoretical stud-
ies[7–12] have in turn greatly expedited experiments to pro-
vide evidence of anion–p interactions obtained in the solid
state[13–19] and in solution[20–23] accompanied by the applica-
tion of anion–p interactions in anion recognition and bind-
ing processes.[24–29]
[a] Dr. Z. Xu, Dr. S. K. Kim, Prof. J. Yoon
Department of Chemistry and Nanoscience
and Department of Bioinspired Science (WCU)
Ewha Womans University, Seoul, 120-750 (Korea)
Fax : (+82)232772384
E-mail: jyoon@ewha.ac.kr
[b] Dr. Z. Xu, Dr. D. R. Spring
Department of Chemistry, University of Cambridge
Cambridge, CB2 1EW (UK)
Fax : (+44)1223336362
E-mail: spring@ch.cam.ac.uk
Anion–p interactions have primarily been studied with
electron-deficient aromatics of positive quadrupole moment,
such as hexafluorobenzene, 1,3,5-trinitrobenzene, and 1,3,5-
triazine, because anion–p interactions that involve electron-
rich aromatics with a negative quadrupole moment have
generally been expected to be repulsive due to their elec-
tron-donating character.[7–12] The noncovalent force between
an anion and an electron-deficient aromatic ring has been
[c] Dr. N. J. Singh, Prof. K. S. Kim
Center for Superfunctional Materials, Department of Chemistry
Pohang University of Science and Technology (China)
Fax : (+82)542798137
E-mail: kim@postech.ac.kr
[d] Dr. Z. Xu, Dr. N. J. Singh
Contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002105.
Chem. Eur. J. 2011, 17, 1163 – 1170
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1163

mainly explained as a contribution of electrostatic attraction
between the anion and one of the positive ends of the aro-
matic quadrupole moment. On the other hand, it is interest-
ing to note that Clements and Lewis theoretically reported
that trichlorobenzene/triiodobenzene with negative quadru-
pole moments can have anion–p interactions.[30] Neverthe-
less, these compounds are still considered to be an electron-
deficient or non-electron-rich system, for which the electro-
static interaction with FÀ can be attractive. It is theoretically
natural to conclude that electron-rich alkylbenzenes that
have negative quadrupole moments would have repulsive
anion–p interactions (compound 1). To substantiate it, we
have studied an alkylbenzene with donating substituents
(compound 2). However, surprisingly, we have found that an
attractive anion–p interaction does exist for such systems
despite the common theoretical knowledge. It is worth men-
tioning here that Schneider had reported the interaction be-
tween a sulfate group and a benzene ring in a 4-[(4-sulfona-
tophenyl)methyl]benzenesulfonate and N-phenylaniline
complex.[5,6] The speculation of the anion–p interaction in
this system was, however, based solely on the binding free
energy without X-ray/NMR spectroscopic structural infor-
mation. Therefore, given that numerous data of anion–p in-
teractions are available for electron-deficient p systems, it is
desirable to have a solid example to provide both clear the-
oretical understanding and concrete experimental evidence
of the anion–p interaction for an electron-rich p system
based on well-characterized structures. Here, we report the
first theoretical and experimental structural evidence of the
anion–p interaction in solution between an electron-rich al-
kylbenzene ring and a fluoride ion. Furthermore, the anion–
p interactions, along with ionic hydrogen bonds, make com-
pound 2 a highly selective host for a fluoride ion.
Scheme 1. Structures of compounds 1–4.
+
À
À
three (C H) ···F -type ionic hydrogen bonds position the
fluoride ion in the cavity and between these two alkylben-
zene rings to form a sandwich complex. The nonrepulsive
anion–p interaction between the two electron-rich alkylben-
zene rings and the fluoride ion in CD3CN and DMSO was
1
deduced by 19F and H NMR spectroscopy as well as theo-
retical calculations. Compound 4 was prepared as a refer-
ence.
Reaction of 1H-naphtho
bromomethylbenzene derivatives
smoothly to give the 1,3,5-tris[(1H-naphthoACHTUNGTRENNUNG
ACHTUNGTRENNUNG
6
8
1-yl)methyl]benzene derivatives 7 and 9, respectively. Cyclo-
phanes 1 and 2 are easily synthesized in good yield by the
reaction of 1,3,5-tribromomethylbenzene derivatives with
Results and Discussion
Design and synthesis: Recently, Ballester et al. employed an
“enforced proximity” approach in calix[4]pyrrole that deriv-
atives bore axially substituted aryl groups on each of the
four meso carbon atoms to detect weak chloride–p interac-
tions in solution.[31] Among the various approaches to the
recognition of anions,[32,33] the imidazolium group can make
appropriate 1,3,5-tris[(1H-naphthoACHTGNUTERNNU[G 2,3-d]imidazol-1-yl)me-
thyl]benzene derivatives. For example, without recourse to
the use of high-dilution conditions, the triply bridged cyclo-
phane 2 was obtained in a good yield of 71% by treatment
of 1,3,5-tris(bromomethyl)-2,4,6-triethylbenzene (8) with
1,1’,1’’-(2,4,6-triethylbenzene-1,3,5-triyl)tris(methylene)ACHTUNGTRENNUNGtris-
+
À
À
a strong interaction with anions through (C H) ···X -type
ionic hydrogen bonds in which the charge–charge electro-
static interaction dominates.[34,35] The naphthoimidazolium
compound 3 (Scheme 1) was reported as a ratiometric fluo-
rescent chemosensor for a fluoride ion through the strong
(1H-naphthoACTHUNGETRNNU[G 2,3-d]imidazole) (9; 1 equiv) in acetone
(Scheme 2).[37,38] Compound 1 was synthesized by an analo-
gous route and obtained in 52% yield.
Computational: High-level ab initio coupled-cluster with
single and double and perturbative triple/complete basis set
(CCSD(T)/CBS)[39] results show that the interaction energies
of FÀ with benzene (Bz) and triethylbenzene (Et3Bz) are 0.7
and À0.9 kcalmolÀ1, respectively (Table 1). Energy compo-
nent analysis using symmetry-adapted perturbation theory
+
À
À
(C H) ···F -type ionic hydrogen bond, which resulted in a
blueshift in emission.[36] As shown in Scheme 1, the extreme-
ly rigid cage compounds 1 and 2 contain three imidazolium
groups inserted between two substituted benzene rings. The
2
1
2
À
three imidazolium C hydrogen atoms (H , C H) in 1 and 2
are directed toward the cyclophane cavity and are shielded
by magnetically anisotropic benzene rings that display an
unusual chemical shift at high field.[37] However, through an
“enforced proximity” approach, only in compound 2 can the
À
À
(SAPT)[40,41] reveals that the Bz F and Et3Bz F com-
plexes have large negative quadrupole moments, and their
electrostatic energies (ꢀ+6 kcalmolÀ1) and exchange ener-
gies (+4 and 6 kcalmolÀ1, respectively) are repulsive in
À
À
1164
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1163 – 1170

Anion–p Interaction
FULL PAPER
wiched system is highly stabi-
lized with a greatly enhanced
interaction energy (À4.3 kcal
molÀ1; cf. À1.5 kcalmolÀ1 of the
1:1 complex) and
a greatly
shortened distance (d) between
FÀ and the aromatic ring cent-
roid (2.87 ꢄ, which is reduced
from 3.08 ꢄ). This is because
the higher excess electron den-
sity migration from FÀ to two p
rings due to higher polarization
dramatically stabilizes the 2:1
complex relative to the 1:1
complex. In the case of the Bz-
FÀ-Bz complex, the FÀ ion is
still in a repulsive potential;
thus, this sandwiched complex
is not stabilized. We also ob-
served a similar trend in the
calculated geometrical and en-
Scheme 2. Synthesis of compounds 1, 2, and 4.
ergetic results using density
functional theory with empirical
Table 1. Interaction energies (DE in kcalmolÀ1) and distance from FÀ to
the aromatic ring centroid (d in ꢄ) of various benzene (Bz) derivatives
at the level of ab initio and density functional theories.[a]
dispersion corrections (DFT-D (TPSS-D/aVDZ)),[42] even
though the anion-induced polarization effect is weaker.
Hence, the utilization of Et3Bz as the tripodal base of a cage
structure allows accommodation of an FÀ anion inside the
cage, whereas such accommodation of a FÀ anion is not pos-
sible in the case of benzene as a tripodal base. With this the-
oretical understanding, we proceeded to check if FÀ can be
captured inside the cage of the rigid framework of com-
pound 2.
À
À
À
À
À
À8.8
À
À
À9.1
À
Complex
Bz F [Bz2 F ]
Et3Bz F [(Et3Bz)2 F ]
Q (HF/6-311G**)[b]
d (RIMP2/aVDZ)
d (TPSS/aVDZ)
3.27 [3.11]
3.11 [3.00]
0.7 [0.6]
0.7 [1.5]
3.08 [2.87]
2.95 [2.85]
À0.9 [À3.2]
À1.0 [À1.9]
DE
DE
ACHTUNGTRENNUNG
(CCSD(T)/CBS)[c]
ACHTUNGTRENNUNG
SAPT
ACHTUNGTRENNUNG
Eelectrostatic
Einduction**
5.7
À7.1
À1.6
3.7
6.2
À11.1
À2.1
6.1
[e]
[e]
Experimental evidence: The 1H and 19F NMR spectra as
well as fluorescence properties of 1 and 2 with fluoride ions
were examined to verify the inclusion of fluoride in the
cavity. As reference data, the NMR spectra and fluorescence
Edispersion**
[e]
Eexchange*
[a] Geometries are optimized at the RIMP2/aug-cc-PVDZACTHNUGRTNEUNG(aVDZ) and
DFT-D (TPSS-D/aVDZ) levels. See the Supporting Information for de-
tails of the calculations. [b] Quadrupole moments (Q, Dꢄ) of the uncom-
plexed aromatic compounds. [c] Calculated at the RIMP2/aVDZ basis set
superposition error-corrected geometries. [d] Details of the SAPT results
are in the Supporting Information. [e] Asterix is defined in the Support-
ing Information.
properties of 3 with FÀ were reported previously.[35] The (C
À
H)+···FÀ interaction blueshifts the fluorescence of 3 from
2
À
À
440 to 372 nm. The chemical shifts of C H and F change
from d=9.08 to 9.52 ppm and from d=À115.4 (free FÀ in
CD3CN) to À150.9 ppm, respectively. However, the electro-
static interaction between the imidazolium cation and FÀ
can only quench the fluorescence without a shift in emission,
which is supported by the titration test with 4 (Figure S1 in
nature. Meanwhile, the dispersion energies (ꢀÀ2 kcalmolÀ1)
are weakly attractive. The repulsive energy terms are mostly
compensated by the large attractive induction energy term
(À7 and À11 kcalmolÀ1, respectively). Therefore, although
the Supporting Information). This indicates the key role of
2
À
C
C
H in the interaction with fluoride. Therefore, the d of
H and the blueshift in the fluorescence emission can be
À
À
2
À
À
À
the Bz F complex is not yet stabilized, the Et3Bz F com-
plex is stabilized. Hence, despite the large negative quadru-
pole moment (ꢀÀ9 Dꢄ), the highly polarizable Et3Bz binds
FÀ with their strong anion-induced polarization by FÀ. This
type of interaction is quite different from that of the FÀ
complexes with positive quadrupole moments due to elec-
tron-deficient rings.
2
À
À
used to probe the hydrogen bond between C H and F .
1H NMR spectra of 1 and 2 upon the addition of fluoride
are shown in Figure 1. Tetrabutylammonium fluoride hy-
drate was used as the fluoride source. With the addition of
2
1
À
fluoride, the peaks of C H (H ) in 1 shifted downfield
slightly from d=6.82 to 7.30 ppm (Figure 1b). However, for
2
1
À
As the second p system is added to a complex in which
2, upon fluoride binding, the singlet peak of C H (H ) sub-
stantially shifted downfield (d=6.43 to 12.60–12.81 ppm for
2 in Figure 1e) and split into a doublet (J=84 Hz, spin–spin
À
the p system interacts with FÀ (i.e., as the Et3Bz F com-
À
À
À
plex changes to the (Et3Bz)2
F
complex), the latter sand-
Chem. Eur. J. 2011, 17, 1163 – 1170
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1165

J. Yoon, D. R. Spring, K. S. Kim et al.
Figure 1. 1H NMR spectra of 1 and 2 with fluoride in CD3CN at room
temperature. a) Compound 1 with 0 equiv of FÀ; b) 1 with 1.0 equiv of
FÀ; c) 2 with 0 equiv of FÀ; d) 2 with 0.5 equiv of FÀ; e) 2 with 1 equiv of
FÀ, 19F coupled; f) 2 with 1 equiv of FÀ, 19F decoupled. [2]=7.5 mm.
2
À
À
coupling between C H and F ), whereas the resonances of
H2,H5 significantly shifted upfield. The H NMR spectra of
1
2-FÀ with fluorine decoupling display the d of C H as a
2
À
broad singlet at d=12.71 ppm.
The fluorescence responses of 1 and 2 to various anions
were then examined. As shown in Figure 2a, with the addi-
tion of 1 equiv of FÀ, a strongly increased fluorescent emis-
sion centered at 385 nm appeared at the expense of the fluo-
rescent emission of 2 centered at 474 nm. Other tested
anions quenched the fluorescence without a shift in emis-
sion; the extent of quenching depends on the nature of the
anion. However, the addition of fluoride quenched the fluo-
rescence of 1 with the electrostatic interaction between the
imidazolium and fluoride anion (Figure 2b). These results
support the selective inclusion of FÀ in the cavity of 2
Figure 2. a) Fluorescence responses of
2 (10 mm) to various anions
(1 equiv). b) Fluorescence spectra of 1 (10 mm) in the presence of differ-
ent concentrations of FÀ (0 to 1 equiv) in CH3CN. c) Fluorescence spectra
of 2 (10 mm) in the presence of different concentrations of FÀ (0 to
1 equiv) in CH3CN. Inset: Ratiometric calibration curve I385/I474 as a func-
tion of the FÀ concentration.
+
À
À
through (C H) ···F ionic hydrogen bonds. The fluores-
cence titration experiment of 2 with FÀ showed the depend-
ence of the ratio of the emission intensities at 375 and
474 nm (I375/I474) on the concentration of FÀ (Figure 2c).
Thus, 2 could also be used for ratiometric fluorescent prob-
ing for FÀ (see the Supporting Information for details),
which means that changes in
À47.01–47.66 ppm due to the strong interaction between FÀ
and the two benzene rings) and split into a quartet (J=
84 Hz, spin–spin coupling between C H and F ). The cou-
2
À
À
the ratio of the emission inten-
sities at two wavelengths are
observed.
19F NMR spectroscopy was
carried out to explore the FÀ–p
interaction in solution. The res-
À
onances of PF6 appear as a
doublet between d=À70 and
À73 ppm. Upon the inclusion of
fluoride in the cavity of 2, as
shown in Figure 3a, the singlet
peak of free FÀ was substantial-
Figure 3. 19F NMR spectra of 1 and 2 with 1 equiv of fluoride in CD3CN at room temperature. a) Compound 2,
ly shifted upfield (d=À115.4 to 1H coupled; b) 2, 1H decoupled. [2]=7.5 mm; c) 1.
1166
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1163 – 1170

Anion–p Interaction
FULL PAPER
2
À
Table 2. Isothermal titration calorimetry (ITC)-determined binding con-
stants (ꢅ104 mÀ1) and thermodynamic quantities [kcalmolÀ1] for (1–2)-FÀ
complexes. Theoretically estimated free energies are in square brackets.
pling constant is the same as that of C H in Figure 1e. The
19F spectrum of 2-FÀ with proton decoupling displays the
chemical shift of FÀ at d=À47.34 ppm as a sharp singlet
(Figure 3b). The quartet shown in Figure 3a is consistent
Compound 1
Compound 2
n
Ka
1.89 (ꢀ2)[a]
1.07[a]
68.0
with the hypothesis that FÀ is in the center of the cavity and
6.62 (K1), 0.0335 (K2)[b]
À6.07 (DG1) [À5.59]
À2.41 (DG2) [À1.02]
2.93 (DH1), 16.36 (DH2)
9.00 (DS1), 18.77 (DS2)
2
À
interacts equally with the three C H groups. However, in
the mixture of 1-FÀ (1:1), the singlet peak of free FÀ is shift-
ed downfield from d=À115.4 to À138.05 ppm due to the
DG
À7.94 [À7.91]
DH
TDS
3.86
11.80
electrostatic interaction (Figure 3c). Further studies indicate
that FÀ can also interact with only two or one C H in the
2
À
[a] “One set of sites” binding model. [b] “Sequential binding sites”
model, 1/F=1:2.
À
À
À
cavity of 2 to form a benzene F benzene sandwich com-
plex and to display the FÀ–p interaction, which may depend
on the concentration of the 2-FÀcomplex. For example,
when the concentration of 2-FÀ is lower than 6.2 mm, the
triplet and doublet peaks for two and one (C H) ···F ionic
hydrogen bonds, respectively, are observed in the 19F NMR
spectra. Correspondingly, two groups of doublet peaks
appear in the 1H NMR spectra. Moreover, the 19F,1H
room temperature indicates the endothermic reorganization
of the solvent shells into a less-ordered form upon host–FÀ
complex formation. This is due to the fact that the reduced
charge of the host–FÀ complex by ion pairing results in the
substantial reduction of the dipole-charge interaction with
the less-ordered solvent. The complex is less solvated than
the sum of its free components (verified by the formation of
+
À
À
HOESY spectrum also confirms the strong contact between
FÀ and C H nuclei (see the Supporting Information). This
2
À
reflects the stepwise polarization induced by FÀ during the
course of FÀ capture from the outside to the center of the
cavity. After a period of time, the lower concentration
sample displayed the same NMR spectroscopic signals with
the higher concentration sample, indicating that the FÀ is in
precipitation in H NMR spectroscopic titration in which the
1
higher concentration of host was used in the ITC experi-
ment); thus, the release of solvent molecules to the bulk
leads to the entropic overcompensation of the unfavorable
positive DH of desolvation.[43,44]
the center of the cavity and interacts equally with the three
+
À
(C H) groups.
Molecular modeling: Furthermore, we have investigated the
structural isomers and energetics of the 1-FÀ and 2-FÀ com-
plexes using DFT-D (TPSS-D/aVDZ)[41,42] with the consider-
ation of solvent treatment using the conductor-like screening
model (COSMO).[45] The optimized geometries of 1-FÀ and
2-FÀ complexes show that the isomers with the FÀ ion out-
side/inside (external/internal isomers) the cage of 1/2 are
thermodynamically more favorable (Figure 4). The estimat-
ed binding free energies[34] of the 1-FÀ (ÀDG1, ÀDG2) and
2-FÀ complexes are respectively 5.58 kcalmolÀ1 for 1:1 bind-
ing mode, 1.02 kcalmolÀ1 for 1:2 binding mode, and
7.91 kcalmolÀ1, which are in good agreement with the ITC
To check the effect of solvent on the anion–p interaction,
the binding properties of 1 and 2 with FÀ in DMSO were in-
vestigated through H and 19F NMR spectroscopy as well as
1
fluorescence spectra. The upfield-shifted FÀ at d=
À49.20 ppm indicates the existence of an FÀ–p interaction in
DMSO for compound 2 (see the Supporting Information).
2
1
À
Correspondingly, the singlet peak of C H (H ) shifted
downfield and split into a doublet (spin–spin coupling be-
2
À
2
5
À
tween C H and F ), whereas the resonances of H ,H sig-
nificantly shifted upfield (see the Supporting Information).
However, a broad singlet at about d=À157.5 ppm in the
19F NMR spectra was also observed, which was mostly due
to the C F bond formation. Then the addition of F into
DMSO quenched the fluorescence of 2 with a slight blue-
shift in emission (see the Supporting Information).
2
À
À
Isothermal titration calorimetry (ITC) experiments: The
binding constants of compounds 1 and 2 with FÀ were ob-
tained by means of isothermal titration calorimetry (ITC)
experiments at 258C (Table 2). The ITC plots show that the
stoichiometry of 1-FÀ is 1:2, whereas 2 binds FÀ in a 1:1 pat-
tern. This also provides evidence that FÀ only approaches
imidazolium molecules outside the cavity of 1, whereas FÀ is
hosted in the cavity of 2 to interact with three imidazolium
molecules equally. In the case of 1, the first FÀ could be
bound with two imidazolium moieties and the second FÀ
with the third imidazolium moiety. Since triply charged imi-
dazolium host systems and FÀ anions are solvated very effec-
tively (as acetonitrile molecules around the hosts are strong-
ly polarized and well ordered), the positive DH value at
Figure 4. Thermodynamically favorable optimized structures of com-
plexes 1-FÀ (external isomer) and 2-FÀ (internal isomers) at the TPSS-D/
aVDZ level of theory in acetonitrile solvent. The transition-state struc-
ture (TS) 1-FÀ is approximately 10.5 kcalmolÀ1 higher than that of 1-FÀ
external structure, yet around 5.6 kcalmolÀ1 higher than that of 2-FÀ.
Chem. Eur. J. 2011, 17, 1163 – 1170
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1167

J. Yoon, D. R. Spring, K. S. Kim et al.
experimental values of 6.07, 2.41, and 7.94 kcalmolÀ1
(Table 2). The relative free-energy stabilization of the 2-FÀ
complex with respect to the 1-FÀ complex from the ITC
Synthesis of compound 7: NaH (300 mg, 7.5 mmol, 60% in mineral oil)
was added at 08C to a reaction mixture of 5[46] (300 mg, 1.8 mmol) in
THF (20 mL). After the reaction mixture was stirred for 20 min at 08C, 6
(210 mg, 0.59 mmol) was added. After additional stirring for 1 h at room
temperature, the reaction mixture was added to water (50 mL) and was
extracted with CHCl3. The organic layer was then separated, dried with
anhydrous sodium sulfate, and concentrated under reduced pressure. Pu-
rification by flash chromatography on silica gel (CH2Cl2/MeOH=100:1)
afforded 7 as a white solid (247 mg, 68%). 1H NMR (CDCl3, 400 MHz):
d=5.26 (s, 6H), 6.95 (s, 3H), 7.33–7.42 (m, 9H), 7.55 (d, J=8.0 Hz, 3H),
7.98 (d, J=8.0 Hz, 3H), 8.06 (s, 3H), 8.27 ppm (s, 3H); 13C NMR
(CDCl3, 62.5 MHz): d=48.28, 105.54, 117.83, 123.80, 124.81, 125.12,
127.39, 128.20, 128.55, 130.18, 130.55, 134.00, 137.98, 143.81, 146.86 ppm;
HRMS (FAB): m/z: calcd for C42H31N6: 619.2610 [M+H+]; found:
619.2602.
data is 1.9 kcalmolÀ1
;
the theoretical data gives
2.3 kcalmolÀ1. Additionally, the transition-energy barrier rel-
ative to the most stable isomer of the 1-FÀ and 2-FÀ com-
plexes with FÀ outside the cage toward the formation of
structures with FÀ inside the cage were calculated to be
around 10.5 and 5.6 kcalmolÀ1, respectively. Therefore, at
the experimental temperature (258C), the external isomer
+
À
of the 1-FÀ complex with two strong (C H) ···F ionic hy-
drogen bonds could cross the transition-energy barrier of
10.5 kcalmolÀ1 to form the internal isomer (calculated to be
equally stable as that of the external isomer due to en-
hanced interaction from the imidazolium moieties); howev-
er, this internal structure would be thermodynamically un-
favorable due to the unfavorable anion–p interaction. On
the contrary, in the case of the 2-FÀ complex, as the external
isomer is approximately 6 kcalmolÀ1 less stable than the in-
ternal isomer, the formation of the latter isomer would be
easily feasible by crossing the transition-energy barrier of
approximately 6 kcalmolÀ1. Therefore, FÀ is stabilized inside
À
Synthesis of compound 1: Acetone (20 mL) was added to a mixture of 6
(0.087 g, 0.24 mmol) and 7 (0.15 g, 0.24 mmol). The mixture was stirred at
room temperature for 20 h. The precipitate was filtered to give 1-Br3 as
white solid. The bromide salt was dissolved in DMF (10 mL). During the
dropwise addition of saturated aqueous KPF6 solution, a precipitate was
formed. After washing the precipitate several times with water, the de-
sired product was obtained as
a light yellow solid (146 mg, 52%).
1H NMR (CD3CN, 250 MHz): d=5.56 (s, 12H), 6.82 (s, 3H), 7.65 (dd,
6H), 7.75 (s, 6H), 8.16 (dd, 6H), 8.54 ppm (s, 6H); 13C NMR (CD3CN,
100 MHz): d=51.12, 111.87, 112.54, 128.52, 131.85, 132.84, 135.76, 136.06,
143.44 ppm; HRMS (FAB): m/z: calcd for C51H39F12N6P2: 1025.2520
[MÀPF6]+; found: 1025.2512.
+
À
À
the cage of 2 with the formation of (C H) ···F ionic hydro-
gen bonds, whereas the anion–p interaction, though small, is
key for the selectivity of FÀ.
Synthesis of compound 9: NaH (300 mg, 7.5 mmol, 60% in mineral oil)
was added at 08C to a reaction mixture of 5 (300 mg, 1.8 mmol) in THF
(20 mL). After the reaction mixture was stirred for 20 min at 08C, com-
pound 8[47] (250 mg, 0.57 mmol) was added. After additional stirring for
1 h at room temperature, the reaction mixture was added to water
(50 mL) and extracted with CHCl3. The organic layer was then separated,
dried over anhydrous sodium sulfate, and concentrated under reduced
pressure. Purification by flash chromatography on silica gel (CH2Cl2/
MeOH=100:1) afforded 9 (327 mg, 82%) as a white solid. 1H NMR
(CDCl3, 250 MHz) d=0.85 (t, J=6.6 Hz, 9H), 2.60 (m, 6H), 5.32 (s, 6H),
7.34 (m, 6H), 7.53 (s, 3H), 7.75 (s, 3H), 8.87 (t, J=8.0 Hz, 6H), 8.22 ppm
(s, 3H); 13C NMR (CDCl3, 62.5 MHz): d=15.58, 23.84, 43.07, 105.23,
117.98, 123.91, 124.98, 127.51, 128.70, 130.23, 130.43, 130.65, 134.57.
143.93, 144.84, 146.54 ppm; HRMS (FAB): m/z: calcd for C48H43N6:
703.3549 [M+H+]; found: 703.3552.
Conclusion
In summary, by using highly accurate ab initio calculations
of electron-rich aromatic model systems, we predict for the
first time that anion–p interactions are possible even for
electron-rich aromatic systems when the induction effect is
large enough to compensate for the electrostatic repulsion.
Such anion–p interactions have been demonstrated by ex-
perimentally verifying the inclusion of fluoride in the rigid
framework of trisimidazolium cage 2 in which electron-rich
triethylbenzene is a tripodal base. The unique inclusion pat-
Synthesis of compound 2: Acetone (20 mL) was added to a mixture of 8
(0.075 g, 0.17 mmol) and 9 (0.12 g, 0.17 mmol). The mixture was stirred at
room temperature for 20 h. The precipitate was filtered to give 2-Br3 as
1
white solid. The filtrate was concentrated to = volume and the new pre-
+
À
3
À
tern of fluoride with the formation of (C H) ···F ionic hy-
drogen bonds and the anion–p interaction between the elec-
tron-rich alkylbenzene and fluoride has been demonstrated
cipitate was filtered again. The total product was 144 mg in 74% yield.
1H NMR (CD3OD, 250 MHz): d=1.15 (s, 18H), 2.49 (s, 12H), 5.83 (s,
12H), 6.37 (s, 3H), 7.78 (d, 6H), 8.31 (d, 6H), 9.03 ppm (s, 6H);
13C NMR (CD3OD, 62.5 MHz): d=16.42, 24.63, 46.56, 113.23, 128.84,
129.81, 130.56, 132.88, 133.93, 150.65 ppm. The bromide salt was dis-
solved in methanol (20 mL). During the dropwise addition of saturated
aqueous KPF6 solution, precipitate was formed. After washing the pre-
cipitate several times with water, the desired product was obtained as a
light yellow solid (93 mg, 71%). 1H NMR (CD3CN, 250 MHz): d=1.32
(s, 18H), 2.48 (s, 12H), 5.76 (s, 12H), 6.43 (s, 3H), 7.82 (d, 6H), 8.34 (d,
6H), 8.79 ppm (s, 6H); 13C NMR (CD3CN, 62.5 MHz): d=14.98, 22.77,
45.15, 112.03, 127.49, 128.20, 128.28, 131.46, 131.75, 136.39, 149.07 ppm;
HRMS (FAB): m/z: calcd for C63H63F12N6P2: 1193.4398 [MÀPF6]+;
found: 1193.4392.
with fluorescent changes, 19F and H NMR spectroscopic ti-
1
trations, and theoretical calculations.
Experimental Section
Materials and methods: Unless otherwise noted, materials were obtained
from commercial suppliers and were used without further purification.
Flash chromatography was carried out on silica gel 60 (230–400 mesh
ASTM; Merck). Thin-layer chromatography (TLC) was carried out using
Merck 60 F254 plates with a thickness of 0.25 mm. Preparative TLC was
performed using Merck 60 F254 plates with a thickness of 1 mm. 1H, 13C,
and 19F NMR spectra were recorded using a Bruker instrument operating
at 250 or 400 MHz. Chemical shifts (d) were given in ppm and coupling
constants (J) in Hz. Fluorescence emission spectra were obtained using
an RF-5301/PC spectrofluorophotometer from Shimadzu.
Synthesis of compound 11: NaH (330 mg, 8.3 mmol, 60% in mineral oil)
was added at 08C to a reaction mixture of 10[45] (746 mg, 4.1 mmol) in
THF (20 mL). After the reaction mixture was stirred for 20 min at 08C,
iodomethane (960 mg, 6.3 mmol) was added. After additional stirring for
1 h at room temperature, the reaction mixture was added to water
(50 mL) and extracted with CHCl3. The organic layer was then separated,
dried over anhydrous magnesium sulfate, and concentrated under re-
duced pressure. Purification by flash chromatography on silica gel
1168
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1163 – 1170

Anion–p Interaction
FULL PAPER
(CH2Cl2/MeOH=100:1) afforded 11 (740 mg, 92%) as a pale yellow
solid. H NMR (CDCl3, 250 MHz): d=2.64 (s, 3H), 3.73 (s, 3H), 7.25 (m,
1H), 7.40 (m, 2H), 7.60 (m, 1H), 7.93 (m, 2H), 8.13 ppm (m, 1H);
13C NMR (CDCl3, 62.5 MHz): d=14.09, 29.62, 104.37, 115.28, 123.21,
123.98, 127.41, 128.31, 129.98, 132.53, 136.45, 142.71, 156.14 ppm; HRMS
(FAB): m/z: calcd for C13H13N2: 197.1079 [M+H+]; found: 197.1082.
WCU (R32-2008-000-10180-0), the EPB Center (2009-0063312), KISTI
(KSC-2008K08-0002), D.R.S. acknowledges the Herchel Smith Postdoc-
toral Fellowship Fund and the Newman Trust.
1
Synthesis of compound 4: A mixture of 11 (100 mg, 0.51 mmol) and
benzyl bromide (130 mg, 0.76 mmol) in acetonitrile (10 mL) was heated
at reflux for 24 h under N2. After cooling to the room temperature, the
precipitate was filtered and washed with cold CH2Cl2 to give 4 as bro-
mide salt. The bromide salt was dissolved in methanol (10 mL). During
the dropwise addition of aqueous saturated KFP6 solution, precipitate
was formed. After washing the precipitate with water several times, the
desired product was obtained as a white solid (165 mg, 75%). 1H NMR
(DMSO, 250 MHz): d=2.90 (s, 3H), 4.03 (s, 3H), 5.71 (s, 2H), 7.38 (m,
5H), 7.65 (m, 2H), 8.10 (m, 3H), 8.31 ppm (m, 1H); 13C NMR (DMSO,
62.5 MHz): d=10.71, 31.57, 48.43, 109.79, 109.86, 117.02, 117.66, 117.79,
126.42, 126.90, 127.79, 127.85, 128.36, 128.80, 129.99, 130.71, 130.94,
131.03, 133.14, 155.74 ppm; HRMS (FAB): m/z: calcd for C20H19N2:
[1] P. Gamez, T. J. Mooibroek, S. J. Teat, J. Reedijk, Acc. Chem. Res.
2007, 40, 435–444.
[2] B. P. Hay, V. S. Bryantsev, Chem. Commun. 2008, 2417–2428.
[3] B. L. Schottel, H. T. Chifotides, K. R. Dunbar, Chem. Soc. Rev. 2008,
37, 68–83.
[4] N. J. Singh, H. M. Lee, I.-C. Hwang, K. S. Kim, Supramol. Chem.
2007, 19, 321–332.
[5] J. L. Sessler, S. Camiolo, P. A. Gale, Coord. Chem. Rev. 2003, 240,
17–55.
[6] a) H.-J. Schneider, T. Schiestel, P. Zimmermann, J. Am. Chem. Soc.
1992, 114, 7698–7703; b) H.-J. Schneider, T. Schiestel, P. Zimmer-
mann, J. Phys. Org. Chem. 1993, 6, 590–594.
[7] D. QuiÇonero, C. Garau, C. Rotger, A. Frontera, P. Ballester, A.
Costa, P. M. Deyꢂ, Angew. Chem. 2002, 114, 3539–3542; Angew.
Chem. Int. Ed. 2002, 41, 3389–3392.
[8] M. Mascal, A. Armstrong, M. D. Bartberger, J. Am. Chem. Soc.
2002, 124, 6274–6276.
[9] I. Alkorta, I. Rozas, J. Elguero, J. Am. Chem. Soc. 2002, 124, 8593–
8598.
287.1543 [MÀPF6]+; found: 287.1540.
À
Calculation methods: Interaction energies of benzene F (BzÀFÀ) and
À
À
À
À
À
triethylbenzene F (Et3Bz F ) complexes are calculated on the opti-
mized geometries at the level of RIMP2/aug-cc-pVDZ (abbreviated as
aVDZ) theory with the basis-set superposition-error (BSSE) correction.
These optimized geometries were used for the further calculation of
RIMP2/aVTZ and CCSD(T)/aVDZ single-point energies with BSSE cor-
rections. The complete basis set (CBS) limit values for the RIMP2 inter-
action energies were evaluated based on the extrapolation method by ex-
ploiting the fact that the basis set error in the electron correlation energy
is proportional to NÀ3 for the aug-cc-pVNZ (or aVNZ) basis set.[39] The
CCSD(T)/CBS interaction energies were estimated by calculating the
sum of CCSD(T)/aVDZ interaction energies and the difference between
[10] C. Garau, A. Frontera, D. QuiÇonero, P. Ballester, A. Costa, P. M.
Deyꢂ, ChemPhysChem 2003, 4, 1344–1348.
[11] D. Kim, P. Tarakeshwar, K. S. Kim, J. Phys. Chem. A 2004, 108,
1250–1258.
[12] D. Y. Kim, N. J. Singh, J. W. Lee, K. S. Kim, J. Chem. Theory
Comput. 2008, 4, 1162–1169.
the RIMP2/CBS and RIMP2/aVDZ interaction energies. Meanwhile, the
[13] S. Demeshko, S. Dechert, F. Meyer, J. Am. Chem. Soc. 2004, 126,
4508–4509.
[14] P. de Hoog, P. Gamez, I. Mutikainen, U. Turpeinen, J. Reedijk,
Angew. Chem. 2004, 116, 5939–5941; Angew. Chem. Int. Ed. 2004,
43, 5815–5817.
[15] C. S. Campos-Fernꢆndez, B. L. Schottel, H. T. Chifotides, J. K. Bera,
J. Bacsa, J. M. Koomen, D. H. Russell, K. R. Dunbar, J. Am. Chem.
Soc. 2005, 127, 12909–12923.
[16] I. A. Guralꢁskiy, P. V. Solntsev, H. Krautscheid, K. V. Domasevitch,
Chem. Commun. 2006, 4808–4810.
[17] S. Furukawa, T. Okubo, S. Masaoka, D. Tanaka, H. Chang, S. Kita-
gawa, Angew. Chem. 2005, 117, 2760–2764; Angew. Chem. Int. Ed.
2005, 44, 2700–2704.
[18] C. Estarellas, M. C. Rotger, M. Capꢇ, D. QuiÇonero, A. Frontera,
A. Costa, P. M. Deyꢂ, Org. Lett. 2009, 11, 1987–1990.
[19] R. J. Gçtz, A. Robertazzi, I. Mutikainen, U. Turpeinen, P. Gamez, J.
Reedijk, Chem. Commun. 2008, 3384–3386.
[20] H. Maeda, A. Osuka, H. Furuta, J. Inclusion Phenom. Macrocyclic
Chem. 2004, 49, 33–36.
[21] Y. S. Rosokha, S. V. Lindeman, S. V. Rosokha, J. K. Kochi, Angew.
Chem. 2004, 116, 4750–4752; Angew. Chem. Int. Ed. 2004, 43, 4650–
4652.
À
CCSD(T)/CBS interaction energies of (Bz)2 F /
were estimated as the sum of the MP2/CBS interaction energy and twice
À A
F
complexes
À
À
À
À
À
the [CCSD(T)/CBS–MP2/CBS] energy for the respective Bz F /Et3Bz
FÀ complexes. Additionally, geometry optimization and energy calcula-
tions were also performed using density functional theory with empirical
dispersion correction at the TPSS-D/aVDZ level, which gives energy
values close to the CCSD(T)/CBS limit. This calculation for large systems
lends credence to the estimated CCSD(T)/CBS results for the large sys-
tems. More details are provided in the Supporting Information. The cal-
culation details of SAPT are in the Supporting Information.
Isothermal titration calorimetry experiments: Isothermal titration calo-
rimetry experiments were performed using a VP-ITC microcalorimeter
(Microcal Inc., Northampton, MA). All solutions were prepared in
HPLC-grade MeCN. For a typical ITC run, the instrument chamber
(1.4 mL) contained a solution of the host ([1]=0.1 mm, [2]=0.1 mm), and
a 2.0 mm solution of tetrabutylammonium fluoride (guest) was taken up
in a 300 mL injection syringe. The syringe was assembled into the cham-
ber for equilibration while stirring at 502 rpm. The chamber temperature
was set up to 258C. The injections were programmed at 10 mL each,
added over 20 s, and spaced 200 s apart. “Number of sites” (N), binding
constant (K in mÀ1), association enthalpy (DH in calmolÀ1), and associa-
tion entropy (DS in calmolÀ1 KÀ1) were obtained by fitting the titration
data using the “one set of sites” model algorithm provided in the Micro-
Cal Origin Software package (version 7.0). For compound 1, the curve-
fitting “sequential binding sites” model was also used.
[22] O. B. Berryman, F. Hof, M. J. Hynes, D. W. A. Johnson, Chem.
Commun. 2006, 506–508.
[23] R. M. Fairchild, K. T. Holman, J. Am. Chem. Soc. 2005, 127, 16364–
16365.
[24] V. Gorteau, G. Bollot, J. Mareda, A. Perez-Velasco, S. Matile, J.
Am. Chem. Soc. 2006, 128, 14788–14789.
[25] M. Mascal, I. Yakovlev, E. B. Nikitin, J. C. Fettinger, Angew. Chem.
2007, 119, 8938–8940; Angew. Chem. Int. Ed. 2007, 46, 8782–8784.
[26] A. P. Davis, D. N. Sheppard, B. D. Smith, Chem. Soc. Rev. 2007, 36,
348–357.
Acknowledgements
We thank Dr. Nick Bampos for helpful discussions. Y.J. acknowledges fi-
nancial support from the Basic Science Research Program (2010-
0018895), the Converging Research Center Program (2009-0093677), and
WCU (R31-2008-000-10010-0) through the National Research Founda-
tion; K.S.K. acknowledges the National Honor Scientist Program, GRL,
[27] L. A. Barrios, G. Aromi, A. Frontera, D. Quinonero, P. M. Deyꢂ, P.
Gamez, O. Roubeau, E. J. Shotton, S. J. Teat, Inorg. Chem. 2008, 47,
5873–5881.
[28] C. Hung, A. S. Singh, C. Chen, Y. Wen, S. Sun, Chem. Commun.
2009, 1511–1513.
Chem. Eur. J. 2011, 17, 1163 – 1170
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1169

J. Yoon, D. R. Spring, K. S. Kim et al.
[29] R. E. Dawson, A. Hennig, D. P. Weimann, D. Emery, V. Rauikumar,
J. Montenegro, T. Takeuchi, S. Gabutti, M. Mayor, J. Mareda, C. A.
Schalley, S. Matile, Nat. Chem. 2010, 2, 533–538.
[30] A. Clements, M. J. Lewis, J. Phys. Chem. A 2006, 110, 12705–12710.
[31] G. Gil-Ramꢈrez, E. C. Escudero-Adꢆn, J. Bene tBuchholz, P. Balles-
ter, Angew. Chem. 2008, 120, 4182–4186; Angew. Chem. Int. Ed.
2008, 47, 4114–4118.
117, 2959–2963; Angew. Chem. Int. Ed. 2005, 44, 2899–2903;
c) S. K. Kim, N. J. Singh, J. Kwon, I.-C. Hwang, S. J. Park, K. S. Kim,
J. Yoon, Tetrahedron 2006, 62, 6065–6072; d) V. Amendola, B. Co-
lasson, L. Fabbrizzi, M.-J. R. Douton, Chem. Eur. J. 2007, 13, 4988–
4997; e) C. Coll, R. Casasffls, R. Martꢈnez-MꢆÇez, M. D. Marcos, F.
Sancenꢇn, J. Soto, Angew. Chem. 2007, 119, 1705–1708; Angew.
Chem. Int. Ed. 2007, 46, 1675–1678; f) Z. Xu, N. J. Singh, J. Lim, J.
Pan, H. N. Kim, S. Park, K. S. Kim, J. Yoon, J. Am. Chem. Soc. 2009,
131, 15528–15533; g) X. Chen, S. Kang, M. J. Kim, J. Kim, Y. S.
Kim, H. Kim, B. Chi, S.-J. Kim, J. Y. Lee, J. Yoon, Angew. Chem.
2010, 122, 1448–1451; Angew. Chem. Int. Ed. 2010, 49, 1422–1425.
[36] Z. Xu, S. K. Kim, S. J. Han, G. Kociok-Kohn, T. D. James, J. Yoon,
Eur. J. Org. Chem. 2009, 3058–3065.
[37] M. V. Baker, M. J. Bosnich, C. C. Williams, B. W. Skelton, A. H.
White, Aust. J. Chem. 1999, 52, 823–825.
[38] C. E. Willans, K. M. Anderson, P. C. Junk, L. J. Barbour, J. W. Steed,
Chem. Commun. 2007, 3634–3636.
[39] S. K. Min, E. C. Lee, H. M. Lee, D. Y. Kim, D. Kim, K. S. Kim, J.
Comput. Chem. 2008, 29, 1208–1221.
[40] B. Jeziorski, R. Moszynski, K. Szalewicz, Chem. Rev. 1994, 94, 1887–
1930.
[41] N. J. Singh, S. K. Min, D. Y. Kim, K. S. Kim, J. Chem. Theory
Comput. 2009, 5, 515–529.
[42] a) J. M. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys. Rev.
Lett. 2003, 91, 146401; b) S. Grimme, J. Comput. Chem. 2006, 27,
1787–1799.
[43] M. Berger, F. P. Schmidtchen, Angew. Chem. 1998, 110, 2840–2842;
Angew. Chem. Int. Ed. 1998, 37, 2694–2696.
[44] M. Haj-Zaroubi, N. W. Mitzel, F. P. Schmidtchen, Angew. Chem.
2002, 114, 111–114; Angew. Chem. Int. Ed. 2002, 41, 104–107.
[45] A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans.2 1993, 799–
805.
[32] a) R. Martꢈnez-MꢆÇez, F. Sancanꢇn, Chem. Rev. 2003, 103, 4419–
4476; b) T. Gunnlaugsson, M. Glynn, G. M. Tocci, P. E. Kruger,
F. M. Pfeffer, Coord. Chem. Rev. 2006, 250, 3094–3117; c) C.-H.
Lee, H. Miyaji, D.-W. Yoon, J. L. Sessler, Chem. Commun. 2008,
24–34; d) T. Sakamoto, A. Ojida, I. Hamachi, Chem. Commun.
2009, 141–152; e) Z. Xu, X. Chen, H. N. Kim, J. Yoon, Chem. Soc.
Rev. 2010, 39, 127–137; f) C. Caltagirone, P. A. Gale, Chem. Soc.
Rev. 2009, 38, 520–563; g) H. Y. Gong, B. M. Rambo, E. Karnas,
V. M. Lynch, J. L. Sessler, Nat. Chem. 2010, 2, 406–409; h) X. Chen,
Y. Zhou, X. Peng, J. Yoon, Chem. Soc. Rev. 2010, 39, 2120–2135.
[33] a) Y. Kubo, T. Ishida, T. Minami and T. D. James, Chem. Lett. 2006,
35, 996–997; b) Y. Kubo, A. Kobayashi, T. Ishida, Y. Misawa, T. D.
James, Chem. Commun. 2005, 2846–2848; c) Z. Q. Guo, W. H. Zhu,
H. Tian, Macromolecules 2010, 43, 739–774; C. W. Chiu, F. P.
Gabbaꢉ, J. Am. Chem. Soc. 2006, 128, 14248–14249; d) S. K. Kim,
J. H. Bok, R. A. Bartsch, J. Y. Lee, J. S. Kim, Org. Lett. 2005, 7,
4839–4842; e) D. H. Lee, H. Y. Lee, K. H. Lee, J.-I. Hong, Chem.
Commun. 2001, 1188–1189; f) J. L. Sessler, W.-S. Cho, V. Lynch, V.
Krꢆl, Chem. Eur. J. 2002, 8, 1134–1143; g) S.-J. Hong, J. Yoo, D.-W.
Yoon, J. Yoon, J. S. Kim, C.-H. Lee, Chem. Asian J. 2010, 5, 768–
772; h) Q.-Q. Jiang, B. Darhkijav, H. Liu, F. Wang, Z. Li, Y.-B.
Jiang, Chem. Asian J. 2010, 5, 543–549; i) Y. Qu, J. Hua, H. Tian,
Org. Lett. 2010, 12, 3320–3323; j) E. J. Jun, K. M. K. Swamy, H.
Bang, S.-J. Kim, J. Yoon, Tetrahedron Lett. 2006, 47, 3103–3106.
[34] a) J. Yoon, S. K. Kim, N. J. Singh, K. S. Kim, Chem. Soc. Rev. 2006,
35, 355–360; b) Z. Xu, S. K. Kim, J. Yoon, Chem. Soc. Rev. 2010, 39,
1457–1466.
[46] A. Hijazi, W. Pfleiderer, Nucleosides Nucleotides 1984, 3, 549–557.
[47] A. Vacca, C. Nativi, M. Cacciarini, R. Pergoli, S. Roelens, J. Am.
Chem. Soc. 2004, 126, 16456–16465.
[35] a) S. K. Kim, B.-G. Kang, H. S. Koh, Y. J. Yoon, S. J. Jung, B. Jeong,
K.-D. Lee, J. Yoon, Org. Lett. 2004, 6, 4655–4658; b) K. Chellappan,
N. J. Singh, I. C. Hwang, J. W. Lee, K. S. Kim, Angew. Chem. 2005,
Received: July 23, 2010
Published online: December 13, 2010
1170
www.chemeurj.org
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1163 – 1170