Cycloalkane and alicyclic heterocycle complexation by new switchable resorcin[4]arene-based container molecules: NMR and ITC binding studies

Article abstract of DOI:10.1002/chem.201101861

The synthesis and structural characterization of novel, "molecular basket"-type bridged cavitands is reported. The resorcin[4]arene-based container molecules feature well-defined cavities that bind a wide variety of cycloalkanes and alicyclic heterocycles

Full text of DOI:10.1002/chem.201101861

DOI: 10.1002/chem.201101861
Cycloalkane and Alicyclic Heterocycle Complexation by New Switchable
Resorcin[4]arene-Based Container Molecules: NMR and ITC Binding
Studies**
Jens Hornung,[a] Daniel Fankhauser,[a] Laura D. Shirtcliff,[a, b] Antonia Praetorius,[a]
W. Bernd Schweizer,[a] and FranÅois Diederich*[a]
Abstract: The synthesis and structural
characterization of novel, “molecular
basket”-type bridged cavitands is re-
ported. The resorcin[4]arene-based
container molecules feature well-de-
fined cavities that bind a wide variety
of cycloalkanes and alicyclic heterocy-
cles. Association constants (Ka) of the
1:1 inclusion complexes were deter-
mined by both 1H NMR and isothermal
titration calorimetry (ITC). The ob-
tained Ka values in mesitylene ranged
from 1.7ꢀ102 mÀ1 for cycloheptane up
to 1.7ꢀ107 mÀ1 for morpholine. Host–
guest complexation by the molecular
baskets is generally driven by disper-
tions. The first crystal structure of a
cavitand-based molecular basket is re-
ported, providing precise information
on the geometry and volume of the
inner cavity in the solid state. Molecu-
lar dynamic (MD) simulations provid-
ed information on the size and confor-
mational preorganization of the cavity
in the presence of encapsulated guests.
The strongest binding of heterocyclic
guests, engaging in polar interactions
with the host, was observed at a cavity
filling volume of 63 Æ9%.
À
sion interactions, C H···p interactions
of the guests with the aromatic walls of
the cavity, and optimal cavity filling.
Correlations between NMR-based
structural data and binding affinities
support that the complexed heterocy-
clic guests undergo additional polar
À
À
C O···C=O, N H···p, and S···p interac-
Keywords: container molecules
·
controllable encapsulation · host–
guest systems · molecular recogni-
tion · supramolecular chemistry
Introduction
son to the original, open-top cavitands with four identical
quinoxaline flaps introduced and studied by Cram and co-
workers[3,4] as well as by others,[5–7] the acid (CF3COOH
(TFA)) concentration required for complete vase-to-kite
switching of the bridged cavitands is higher by one order of
magnitude.[2] The bridge spanning the diazaphthalimide side
walls enforces and rigidifies the vase conformation of the
cavitand, which increases host–guest binding strength.[2]
Open, rim-decorated resorcin[4]arene cavitands have also
been shown to form stable inclusion complexes;[8–10] howev-
er, their conformational flexibility remains high,[10] which
limits precise molecular recognition studies in the interior
cavity.
Bridged resorcin[4]arene-derived container molecules (“mo-
lecular baskets”) feature two diazaphthalimide side walls
connected by a rigid bridge in an anti-orientation and two
flexible quinoxaline flaps. These container molecules under-
go a switchable conformational change from a closed vase to
an open kite form by modulation of pH or temperature
(Figure 1).[1,2]
In the vase form, the inner cavity of the bridged resorcin-
[4]arene-derived cavitands is completely enclosed, allowing
stable host–guest complexation in solution.[1,2] In compari-
Here, we present the syntheses and properties of two new
switchable cavitands (1 and 2 in Figure 2) with rigidifying
para-xylylene bridges and different legs. The first crystal
structure of a cavitand-based molecular basket is reported,
providing precise information on the geometry and volume
of the inner cavity in the solid state. Host–guest studies
using cavitand 1 and a wide variety of alicyclic and alicyclic
heterocyclic compounds were performed by 1H NMR and
isothermal titration calorimetry (ITC) to determine the Ka
values of the formed inclusion complexes and to decipher
the interactions between bridged cavitand and guest mole-
cule. Heterocycles were of particular interest for exploration
of polar interactions between the complexed guests and the
surrounding host molecule. This study therefore comple-
[a] Dr. J. Hornung,+ D. Fankhauser,+ Dr. L. D. Shirtcliff, A. Praetorius,
Dr. W. B. Schweizer, Prof. F. Diederich
Laboratorium fꢁr Organische Chemie, ETH Zꢁrich
Hçnggerberg, HCI, CH-8093 Zꢁrich (Switzerland)
Fax : (+41)44-632-1109
E-mail: diederich@org.chem.ethz.ch
[b] Dr. L. D. Shirtcliff
Current address: Department of Chemistry
Oklahoma State University
107 Physical Science, Stillwater, OK 74074 (USA)
[+] Contributed equally to this work.
[**] ITC=Isothermal Titration Calorimetry.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201101861.
12362
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12362 – 12371

FULL PAPER
troscopic properties fully sup-
port the molecular structures
assigned to 1, 2, 11, and 13,
which are stable, colorless
solids, decomposing around
3008C (for the binding and
switching properties of tube 13,
see Figure 1SI and 2SI in the
Supporting Information).
Figure 1. Conformational vase–kite switching of a molecular basket-type container molecule.
X-ray crystal structure of mo-
lecular basket 2:
A crystal
structure for an open cavitand
with four quinoxaline walls and di-tert-butylbenzyl legs was
published in 2001 by our group.[15] Crystals suitable for X-
ray analysis were obtained from a solution of 2 in CH2Cl2/
heptane by slow evaporation of CH2Cl2. The crystal struc-
ture (space group: monoclinic P21) shows the cavitand in
the vase conformation (Figure 3). Strong disordering of
atoms in the di-tert-butylbenzyl legs, which was observed in
our earlier published structure as well,[15] rendered the loca-
tion and refinement of individual atom positions of some of
the leg atoms impossible. The structure from the X-ray ex-
periment showed a well-defined cavity, two complete di-tert-
butylbenzyl legs and a part of a third one. The completely
disordered atoms of two of the di-tert-butylbenzyl residues
could not be resolved and were therefore calculated and re-
fined as rigid body allowing the tert-butyl groups to rotate.
The guest molecule inside the cavity could be a CH2Cl2
molecule with the two chlorines aligned almost on an axis
passing through the phenyl ring of the bridge and the center
of the rim, positioned below and above the level of the qui-
noxaline N atoms. The CH2Cl2 carbon is fully disordered in
the crystal and could not be seen; the distance of 2.87 ꢃ be-
Figure 2. Structures of the bridged cavitands 1 and 2.
ments the previous investigations on apolar binding process-
es inside container molecules by Mecozzi and Rebek.[11]
Results and Discussion
Synthesis: The syntheses of molecular baskets 1 and 2 are
outlined in Scheme 1. Aldehyde 3 was synthesized in three
steps from commercially available 3,5-di-tert-butyltoluene,
avoiding previously used cyanide chemistry and improving
the overall yield from 25% to 42% compared to literature
procedures.[12,13] Aldehyde 4[14] was subjected to a Wittig re-
action, affording a mixture of cis- and trans-vinyl methyl
ethers 5a and 5b. The two isomers were isolated and char-
acterized separately, but reacted together in the next step to
afford aldehyde 3. Condensation of resorcinol with aldehyde
3, following standard procedures, afforded octol 6 in 33%
yield.[15] Bridging of 6 with 2,3-dichloroquinoxaline afforded
cavitand 7[12–16] from which two bridges were selectively re-
moved to afford anti-tetrol 8 (Scheme 1A).[13,17,18]
tween the two Cl atoms suggests the presence of a CH2Cl2
[19]
À
molecule (Cl Cl distance: 2.843 ꢃ). However, some un-
certainty remains, as the resolution of the guest is rather
low. The crystal packing shows a head-to-tail arrangement
forming infinite columns along the a axis (see Figure 3SI).
The neighboring columns have an antiparallel orientation
towards each other. We assume that the low resolution of
the crystal structure results from disorder, most probably
caused by a rotation of the molecules by 908 along the stack-
ing axes.
In contrast to our previously published work,[1,2] the
basket bridge was introduced in one single step. For this
transformation, the bis(dichlorodiazaphthalimide) 9 was pre-
pared by condensation of anhydride 10[17] and the HCl salt
of commercially available 1,4-bis(aminomethyl)benzene
(Scheme 1A).
Bridging moiety 9 was subsequently allowed to react with
tetrol 8 under basic conditions to afford cavitand 2 in 6%
isolated yield. In addition, the tube-type dimeric 11 was ob-
tained in 9% yield (Scheme 1B). Similarly, cavitand 1 was
synthesized in 23% isolated yield by reacting 9 with the
known tetrol 12 bearing n-hexyl legs.[13,18] Additionally, di-
meric 13 was obtained in 10% yield (Scheme 1B). The spec-
Host–guest studies by 1H NMR spectroscopy and ITC:
Complexation studies by 1H NMR spectroscopy and ITC
with the more soluble basket 1 were carried out in mesity-
lene at 303 K. Mesitylene is the solvent of choice as it does
not fit inside the cavity and therefore cannot compete for
binding.[7,20,21] However, molecular basket cavitands have
been shown to bind minute solvent impurities. Therefore,
the purities of commercially available deuterated and non-
deuterated mesitylene were examined by GC/MS. In
[D12]mesitylene, deuterated impurities were found. These
1
impurities are invisible in the H NMR spectra but poten-
tially compete with the investigated guests for interior bind-
Chem. Eur. J. 2011, 17, 12362 – 12371
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12363

F. Diederich et al.
Scheme 1. Synthesis of molecular baskets 1 and 2. a) (C6H5)3P+CH2OCH3 ClÀ, NaH, THF, 18 h, reflux; 80% (5a), (5b). b) HCOOH, CH2Cl2, 20 h,
258C; 69%. c) 3, HCl, EtOH, 5 days, 608C; 33%. d) 2,3-Dichloroquinoxaline, K2CO3, DMF, 48 h, 608C; 80%. e) CsF, catechol, DMF, 1 h, 808C; 46%.
f) Ac2O, sealed tube, 80 min, 1258C; 68%. g) 9, Et3N, DMF, 18 h, 708C; 6% (2), 9% (11); h) 9, DIPEA, DMF, 18 h, 858C; 23% (1), 10% (13).
DIPEA=diisopropylethylamine; DMF=N,N-dimethylformamide.
ing. By careful distillation, realized by the vendor, high-
quality [D12]mesitylene was obtained (see Figure 4SI). The
only remaining, very minor impurities are non-separable,
deuterated structural isomers of [D12]mesitylene, which
could possibly fit inside the cavity and compete with guests.
A variety of cycloalkanes and alicyclic heterocycles of dif-
ferent sizes were chosen as guests (Figure 4). In addition, we
attempted to investigate azetane, glutarimide, and morpho-
lin-3-one. All three substances, however, were not sufficient-
ly soluble in mesitylene to accurately determine association
constants.
hexyl resonances of host 1, a proper determination of the
free guest peak area was problematic. We overcame this
problem by using an internal standard (1,3,5-trimethoxyben-
zene) and evaluated the peak area of encapsulated guest sig-
nals against this standard. 1,3,5-Trimethoxybenzene is not a
1
competitive binder and has no overlapping H NMR signals.
1H NMR signals for all unbound guests in [D12]mesitylene
and for the bound guests that were analyzed by this method
are listed in Table 1SI.
1H NMR measurements of host 1 in CDCl3 indicated the
presence of a single host species, presumably forming a 1:1
host–guest complex with CDCl3. In [D12]mesitylene, howev-
er, three different sets of host signals were observed. We
speculate that the deuterated structural isomers of
[D12]mesitylene, which were found by GC/MS, possible fit in
the cavity and therefore produce different signals. When the
sample was heated to 423 K, only one set of signals was
Molecular basket host–guest exchange kinetics are slow
1
on the H NMR time scale at 303 K, and Ka values can be
determined from the ratio of the signals of free and encap-
sulated guest, with the latter appearing in the d range
beween À2 and À4 ppm. Due to the fact that a large
number of free guest resonances appear in the area of the
12364
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12362 – 12371

Host–Guest Complexes of Cycloalkanes and Alicyclic Heterocycles
FULL PAPER
host–guest complex could be readily assigned. As an exam-
1
ple, the H NMR spectrum of host 1 with encapsulated mor-
pholine (1.2 equiv) as guest (298 K, [D12]mesitylene) is
shown in Figure 6SI.
With 1H NMR spectroscopy, association constants can
only be accurately determined in the range of 10–104 mÀ1 [22]
.
ITC analysis, however can accurately determine association
constants between 104–107 mÀ1.[23] Determination of Ka
1
values above 104 mÀ1 by H NMR spectroscopy is limited by
the accuracy of the measured integrals, while the small evo-
lution of heat for encapsulation of guests with Ka values
below 104 mÀ1 prevents accurate ITC analysis. The results of
1H NMR spectroscopic and ITC binding studies with host 1
are summarized in Table 1.[24] Preliminary binding studies
with tube 13 are reported in the Supporting Information
(Figure 1SI).
Results for the complexation of cyclohexane and thietane
obtained by both methods were reproducible and compara-
ble among each other. However, Ka values from NMR data
were slightly smaller than from ITC measurements. This is
most likely due to the higher purity of the non-deuterated
mesitylene, as compared to the deuterated one: in the deu-
terated solvent, more solvent impurities compete with the
guest for the cavity binding site.
Figure 3. Crystal structure of molecular basket 2 at 100 K. Atoms are
shown with isotropic temperature factors at the 50% level.
present (Figure 5SI), as complexation/decomplexation be-
comes fast on the 1H NMR time scale. Additionally, after
addition of any guest (>1 equiv) at 303 K, one set of host
signals was observed as well. In this case, all signals of the
Guest exchange can be monitored by guest competition
experiments. This is shown in Figure 7SI for the displace-
Figure 4. Structures of guest molecules used in binding studies.
Chem. Eur. J. 2011, 17, 12362 – 12371
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12365

F. Diederich et al.
Table 1. Host–guest complexes with 1:1 stoichiometry formed by molecular basket 1 in mesitylene (ITC) and [D12]mesitylene (1H NMR spectroscopy) at
303 K. The volumes of free guest and the host cavities in the complexes, packing coefficients, and thermodynamic data determined by ITC analysis and
1H NMR spectroscopy are shown.
Complex
VGuest
VCavity
PCMD
Ka
[mÀ1
DH
[kJmolÀ1
DS
DG
[kJmolÀ1
Ka
[mÀ1
[a]
[g]
[b]
[ꢃ3]
[ꢃ3][f]
G
]
R
]
[JmolÀ1KÀ1
]
N
]
N
]
[c]
[c]
[c]
[c]
methylcyclopentane ꢀ 1
methylcyclohexane ꢀ 1
cyclopentane ꢀ 1
cyclohexane ꢀ 1
cycloheptane ꢀ 1
oxetane ꢀ 1
102.3
118.5
85.2
101.9
117.6
62.0
77.2
93.6
109.7
80.2
98.0
114.3
70.9
87.7
103.5
85.4
94.3
104.8
89.5
98.4
98.4
82.4
155.1 Æ9.1
172.8 Æ9.6[h]
141.2 Æ5.9
146.7 Æ5.9
157.1 Æ8.3
139.6 Æ6.5
140.0 Æ5.9
142.8 Æ6.0
151.8 Æ7.0
139.9 Æ6.0
142.9 Æ5.9
152.1 Æ7.5
139.5 Æ6.1
140.6 Æ6.1
148.6 Æ6.8
142.9 Æ6.1
143.5 Æ6.5
150.9 Æ7.2
142.5 Æ5.8
144.9 Æ6.3
145.5 Æ6.2
145.0 Æ7.3
165.6 Æ17.9[h]
186.8 Æ11.3[h]
0.66 Æ0.04
0.69 Æ0.04
0.60 Æ0.03
0.69 Æ0.03
0.75 Æ0.04
0.44 Æ0.02
0.55 Æ0.02
0.66 Æ0.03
0.72 Æ0.03
0.57 Æ0.02
0.69 Æ0.03
0.75 Æ0.04
0.51 Æ0.02
0.62 Æ0.03
0.70 Æ0.03
0.60 Æ0.03
0.66 Æ0.03
0.69 Æ0.03
0.63 Æ0.03
0.68 Æ0.03
0.68 Æ0.03
0.57 Æ0.03
0.59 Æ0.06
0.63 Æ0.04
–
–
–
–
–
–
–
–
(1.5Æ0.2)ꢀ103
[e]
[e]
[e]
[e]
[e]
–
(1.2Æ0.1)ꢀ105
À13.8 Æ0.2
51.5 Æ1.5
À29.4 Æ0.3
–
[d]
(2.3Æ0.5)ꢀ104
À12.8 Æ0.7
41.2 Æ3.8
À25.2 Æ0.6
(3.0Æ0.8)ꢀ103
(1.7Æ0.1)ꢀ102
(4.6Æ0.1)ꢀ103
[c]
[c]
[c]
[c]
–
–
–
–
–
–
–
–
[c]
[c]
[c]
[c]
[d]
oxolane ꢀ 1
(3.2Æ1.5)ꢀ105
(1.5Æ0.1)ꢀ106
(2.5Æ0.5)ꢀ105
(8.0Æ1.9)ꢀ105
(2.4Æ0.2)ꢀ106
À9.4 Æ0.5
À21.4 Æ0.3
À21.6 Æ1.9
À11.9 Æ0.2
À24.2 Æ0.3
73.5 Æ6.1
47.9 Æ1.0
31.7 Æ8.1
73.8 Æ3.2
42.2 Æ1.8
À31.6 Æ1.3
À35.9 Æ0.1
À31.2 Æ0.6
À34.2 Æ0.6
À37.0 Æ0.2
–
–
–
–
[d]
oxane ꢀ 1
[d]
oxepane ꢀ 1
[d]
azolane ꢀ 1
[d]
azinane ꢀ 1
–
[i]
[i]
[i]
[i]
azepane ꢀ 1
–
–
–
–
(1.8Æ1.0)ꢀ105
(3.9Æ1.2)ꢀ104
thietane ꢀ 1
(4.2Æ0.1)ꢀ104
(1.6Æ0.1)ꢀ106
(1.3Æ0.1)ꢀ105
(1.1Æ0.4)ꢀ107
(8.5Æ0.6)ꢀ106
(3.3Æ0.1)ꢀ105
(1.7Æ0.4)ꢀ107
(3.2Æ0.2)ꢀ106
(3.5Æ0.2)ꢀ106
(9.7Æ0.6)ꢀ105
À8.8 Æ0.1
59.7 Æ0.1
61.5 Æ1.8
53.6 Æ1.8
62.5 Æ3.1
54.4 Æ2.1
50.3 Æ0.1
57.0 Æ5.0
61.5 Æ0.1
52.2 Æ2.0
68.2 Æ0.9
À26.9 Æ0.1
À36.0 Æ0.1
À29.7 Æ0.1
À40.7 Æ1.0
À40.2 Æ0.2
À32.1 Æ0.1
À41.8 Æ0.7
À37.8 Æ0.1
À38.0 Æ0.1
À34.7 Æ0.2
[d]
thiolane ꢀ 1
À17.4 Æ0.4
À13.5 Æ0.4
À21.8 Æ2.2
À23.8 Æ0.4
À16.8 Æ0.1
À24.6 Æ0.5
À19.1 Æ0.1
À22.2 Æ0.5
À14.1 Æ0.1
–
–
–
–
–
–
–
–
[d]
thiane ꢀ 1
[d]
1,4-dioxane ꢀ 1
piperazine ꢀ 1
1,4-dithiane ꢀ 1
morpholine ꢀ 1
1,4-thioxane ꢀ 1
thiomorpholine ꢀ 1
benzene ꢀ 1
[d]
[d]
[d]
[d]
[d]
[d]
–
[c]
[c]
[c]
[c]
toluene ꢀ 1
97.0
117.2
–
–
–
–
–
–
–
–
(2.5Æ0.6)ꢀ103
[c]
[c]
[c]
[c]
m-xylene ꢀ 1
(2.2Æ0.3)ꢀ102
1
[a] Ka values determined by ITC (double to pentuple runs). [b] Ka values determined by H NMR spectroscopy (double to quadruple runs). [c] Evolution
1
of heat was too small for determination of an association constant by ITC. [d] Ka values are too high for determination by H NMR spectroscopy. [e] No
binding was observed. [f] Determined by MD calculations according to the literature.[2] [g] Calculated with DG=ÀRTlnKa. [h] Calculation was not con-
vergent. [i] Calculation not possible due to impossible curve fitting.
ment of the weaker guest cycloheptane by the stronger-bind-
ing morpholine, waiting approximately 15 min after addition
of the second guest before recording a new spectrum.
The detailed protocol for the ITC measurements is given
in the Experimental Section. Selected ITC isotherms for ti-
trations of a strongly binding guest (morpholine) and a
weaker binder (cyclohexane) are shown in Figures 8SI–11SI.
The full set of thermodynamic parameters obtained by ITC
is included in Table 1. The stoichiometry of the host–guest
complexes was determined by ITC to be 1:1 for all guests
included in this study.
A comparison of the measured association constants
shows that the most stable complexes (Ka values=106 to
107 mÀ1) are predominantly formed by alicyclic six-mem-
bered heterocycles. In comparison, the hydrocarbon cyclo-
hexane only gives Ka =2.3ꢀ104 mÀ1. Binding affinity of the
heterocycles drops strongly into the range of Ka =103–
104 mÀ1 by either increasing or decreasing the ring size. Opti-
mal volume occupancy (see below), dispersion interactions,
cannot explain the large preference of host 1 for heterocy-
clic guests over hydrocarbon guests.
The thermodynamic parameters calculated from the ex-
perimental ITC data show that binding of all guests is pro-
moted by both a favorable negative enthalpic term DH and
a favorable positive entropic term DS (Table 1). The origins
of the entropic driving force can be rationalized by the fol-
lowing points: i) The container is rigid and preorganized,
and guest complexation does not substantially restrict its
conformational mobility; ii) the favorable entropy of binding
results from the empty host cavity in mesitylene, which is
filled by the guest under gain of entropy of mixing;[1] iii) the
unbound heterocyclic guests in the bulk mesitylene solvent
might not have full freedom of rotation and translation.
Rather, they might adopt preferred, more ordered align-
ments (as is known for molecules dissolved in benzene and
reflected in the aromatic-solvent-induced shifts (ASIS))[25]
so that electronegative oxygen atoms avoid pointing into the
p surface; iv) guests with an H-bond donor and acceptor,
such as morpholine, undergo intermolecular association in
the apolar solvent.
À
and in particular C H···p interactions between the encapsu-
lated guest and the p surfaces in the walls, the resorcinarene
bowl, and the cap of the host clearly are major contributors
to the complexation strength. Compared to cyclic hydrocar-
An informative summary[26] of the results from the ITC
analysis is documented in Figure 5. It shows that the gain in
binding entropy is larger for smaller guests than for larger
guests. Smaller guests lose less translational and rotational
degrees of freedom upon complexation as they can still
À
bons, heterocyclic guests undergo stronger C H···p interac-
À
tions as a result of the higher polarization of the C H bonds
by the s-withdrawing heteroatoms. This alone, however,
12366
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12362 – 12371

Host–Guest Complexes of Cycloalkanes and Alicyclic Heterocycles
FULL PAPER
Figure 5. Thermodynamic quantities from ITC analysis at 303 K.
À
À
move inside the cavity. Accordingly, the enthalpic gain upon
their inclusion is small, as tight contacts are not established
(see azolane, oxolane, thietane).
their N H groups may undergo stabilizing N H···p interac-
tions.[29,30] Hydrogen bonding to the C=O groups in the con-
tainer walls is geometrically not possible.
Six-membered heterocycles are clearly among the most
potent binders. While entropic gains for their encapsulation
are less than in the case of the smaller analogs, they benefit
from a particularly favorable complexation enthalpy. The
complexation enthalpy for N,O-containing six-membered
guests is uniformly below DH <À20 kJmolÀ1, with morpho-
line topping the series at DH=À24.6 kJmolÀ1, whereas cy-
clohexane only binds with DH=À12.8 kJmolÀ1. We there-
fore analyzed which additional favorable interactions the
various heterocyclic guests could undergo with the sur-
rounding cavity walls.[2,27] To find the preferred orientation
of the heteroatoms inside the cavity, rotating frame Over-
hauser effect spectroscopy (ROESY) experiments were per-
formed (for examples, see Figure 12SI). We observed that
the oxygen atoms in cyclic ethers (oxetane, oxolane, oxane,
oxepane) point away from the electron-rich bowl towards
the upper part of the container. We explain this orientation-
al preference with additional fa-
Sulfur-containing guests (thietane, thiolane, thiane) also
point with their S atom towards the para-xylylene bridge,
presumably to gain favorable S···p interactions.[30] In the
complexes of morpholine and 1,4-thioxane, the oxygen of
the guest points towards the upper part of the container; an
À
orientation also adopted by the N H moiety of thiomorpho-
line. It is therefore reasonable to suggest that additional
À
À
polar contacts, such as orthogonal C O···C=O, N H···p, and
S···p interactions stabilize the complexes of heterocyclic
guests (Figure 6).[31] This is particularly reflected for six-
membered heterocyclic guests in the highly favorable values
for the complexation enthalpy DH and ultimately the com-
plexation free enthalpy DG.
In comparison with previous host–guest studies from our
group in less pure [D12]mesitylene, containing among others
0.26% m-xylene, we found that Ka values and hence DG
values are robust. However, our DH and DS values deter-
vorable, near-orthogonal dipo-
À
lar C O···C=O interactions be-
tween the guests and the dia-
zaphthalimide C=O groups.[28]
According to computer model-
ing, repulsive contacts of the O
atom with the para-xylylene
bridge are avoided. The same
observation was made with
cyclic amines (azolane, azinane,
Figure 6. Preferred encapsulation geometries of oxane (a), azinane (b), and thiane (c), supported by ROESY
experiments. Shown are the lowest-energy conformations of the complexes calculated using MacroModel 9.7
azepane). When pointing to-
wards the para-xylylene bridge, (OPLS 2005 force field, GB/SA solvation model for CHCl3).
Chem. Eur. J. 2011, 17, 12362 – 12371
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12367

F. Diederich et al.
VOIDOO to determine the average cavity volumes.[33,34] Fig-
ure 13SI shows the VOIDOO cavity volume calculations for
different host–guest complexes. Taking these volumes and
those of the guests (geometry optimization using PM3 in
SPARTAN[19] and subsequent volume determination by
VOIDOO), the PC values shown in Table 1 were calculated.
In the series of alicyclic heterocycles, the best binding is
established at a filling of about 63 Æ9% (Figure 8). This
value is significantly higher than the value of 55 Æ9% pub-
lished by Mecozzi and Rebek for the binding of liphophilic
guests in apolar hosts, which is driven mainly by van der
Waals interactions.[11] The Mecozzi–Rebek rule has held true
Figure 7. Structure of bridged cavitand 14, which was synthesized and in-
vestigated in previous studies.[2]
mined previously from van’t
Hoff plots (VT-NMR spectros-
copy) are not reliable.[2] This
can be attributed to the fact
that cavitand-based molecular
baskets are notorious for their
propensity to encapsulate even
minor
solvent
impurities.
These conclusions were de-
rived by performing additional
ITC studies with molecular
basket 14[2] in mesitylene of
better quality (Figure 7). The
suitability of van’t Hoff analy-
sis by VT-NMR spectroscopy
for the determination of ther-
modynamic
quantities
for
host–guest complexation by
dynamic container molecules
will require further investiga-
tion.
Figure 8. Diagram showing the free enthalpy for the binding of heterocycles vs. the PC (the ratio of guest
volume to host volume)[11] of the corresponding inclusion complexes.
Cavity occupancy by guest
molecules: Based on the X-ray
crystal structure, the free-cavity volume in 2 (and therefore
also in 1) was determined to be 135 ꢃ3. We have previously
shown in molecular dynamics simulations that molecular
baskets adapt their cavity size to the guest (“induced fit”) to
optimize the fraction of occupied space.[2] For optimal ac-
commodation of a smaller guest, the cavity volume con-
tracts, as compared to the “static” volume seen in the crystal
structure, while it expands to accept larger guests. We pro-
posed that molecular dynamics simulations provide an ap-
propriate way for determining packing coefficients PC, that
is, the ratio of guest volume to host volume.[11] All host–
guest complexes were submitted to molecular dynamics sim-
ulations using MacroModel 9.7 (OPLS 2005 force field, GB/
SA solvation model for CHCl3, 300 K, 1000 ps simulation
time with 1.5 fs time steps).[32] The n-hexyl legs were substi-
tuted with methyl legs for the calculations. A total of 1000
geometries from each of the resulting trajectories were ob-
tained for the host–guest complex, the guest was removed,
and the empty host structures submitted to the program
in studies of other apolar complexation processes in both
chemistry and biology.[35,36] We suggest that the increase in
the optimal PC in our series of complexes is due to the addi-
À
À
tional close polar contacts (C O···C=O, N H···p, and S···p)
established in the complexes of host 1 with heterocyclic
guests. As already mentioned by Rebek and Mecozzi,[11] the
optimal PC increases when polar contacts are established in
confined environments.
Conformational switching: One of the remarkable proper-
ties of the cavitand based molecular baskets is the ability to
reversibly change their conformation upon pH modula-
tion.[1,2] CF3COOD protonates the mildly basic nitrogen
atoms of the quinoxaline flaps, resulting in electrostatic re-
pulsion and a change from vase to kite conformation, as il-
lustrated in Figure 1. While complete vase-to-kite switching
of open cavitands with four quinoxaline walls occurs at
[CF3COOD] ꢁ 0.4m, bridged cavitands require a much
higher concentration of acid.
12368
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12362 – 12371

Host–Guest Complexes of Cycloalkanes and Alicyclic Heterocycles
FULL PAPER
1
Figure 9. a) Acid-induced vase–kite switching of cavitand 1 (2.9ꢀ10À3 m) with cyclopentane encapsulated (1.2 equiv) monitored by H NMR spectroscopy
(500 MHz, [D12]mesitylene, 303 K). b) Cutout from 1H NMR spectra (500 MHz, [D12]mesitylene, 303 K) of cyclopentane (1.2 equiv) encapsulated in 1
(2.9ꢀ10À3 m), upon acidification by CF3COOD and after neutralization with aqueous K2CO3. The dilution of the host–guest complex based on the experi-
ment is mathematically corrected.
The H NMR spectra for the H+-induced switching pro-
cess of bridged cavitand 1 (2.9ꢀ10À3 m) with encapsulated
cyclopentane (1.2 equiv) in [D12]mesitylene at 303 K is out-
lined in Figure 9. In the absence of acid, the signal for the
methine protons under the flexible quinoxaline flaps and
under the diazaphthalimide appear at d=5.95 and 5.67 ppm,
respectively. Upon addition of CF3COOD, the resonance
under the quinoxaline shifts upfield to d=5.15 ppm, where-
as the other methine resonance remains virtually unchanged
(Figure 9a). The entire spectrum at 0 and 9.4m CF3COOD
is shown in Figure 14SI.
most of the smaller- or larger-ring analogues. The measured
differences in association constants, up to five orders of
magnitude, can be explained by the steric fit of the guest
molecules in the cavity and the additional interactions of
the heterocyclic guests with the host. Optimal volume occu-
pancy was investigated by molecular dynamics simulations,
followed by volume calculations. The strongest complexa-
tion was observed at a PC of 63 Æ9%. Besides dispersion
1
À
interactions and in particular C H···p interactions between
the encapsulated guest and the p surfaces in the walls, the
capping bridge, and the resorcinarene bowl of the host, the
best fitting heterocycles undergo additional polar interac-
1
Figure 9b depicts the H NMR resonance of cyclopentane
(1.2 equiv) encapsulated in basket
1
(2.9ꢀ10À3 m) in
tions (C O···C=O, N H···p, and S···p) when bound to the
cavity. This raises the PC beyond the value of 55 Æ9% ob-
served for purely apolar complexation.[11] Preferred orienta-
tions of the heterocyclic guests in the cavity of cavitand 1
were identified by ROESY spectroscopy. Thermodynamic
data from ITC analysis show that binding is driven by both
a favorable enthalpic and entropic change. The latter is larg-
est for the guests that are too small to fully fill the cavity.
The enthalpic driving force is highest when additional polar
interactions are established. Guest binding is reversibly
switchable: upon addition of CF3COOD, the quinoxaline
flaps open up and the cargo is rapidly and quantitatively re-
leased from the host, whereas subsequent neutralization
fully re-establishes guest encapsulation.
À
À
[D12]mesitylene. After the addition of CF3COOD, the
amount of encapsulated cyclopentane decreases drastically.
At an acid concentration of 9.4m, no more complexed guest
is observed. Upon neutralization by means of extraction
with aqueous K2CO3 solution, the bridged cavitand again
becomes fully occupied by the guest (for the switching of
tube 13, see Figure 2SI).
Conclusion
Comprehensive host–guest complexation studies were per-
formed with the resorcin[4]arene-based molecular basket 1
as a host and a broad variety of alicyclic and alicyclic heter-
ocyclic guests. An X-ray crystal structure was obtained for 2,
a close analogue of 1, and provided first crystallographic
support for the postulated container-like structure in this
family of synthetic receptors. Association constants (Ka) for
the 1:1 complexes formed in mesitylene were measured with
Experimental Section
Materials and general methods: See the Supporting Information. This
section only includes the synthesis and characterization of the molecular
baskets 1 and 2 and their side products, tubelike dimeric 11 and 13; all
other syntheses are reported in the Supporting Information. The naming
of the compounds by the “phane nomenclature”[37] is also included in the
Supporting Information.
1
two different methods, H NMR spectroscopy and ITC anal-
ysis. The Ka values at 303 K range from 1.7ꢀ102 mÀ1 for cy-
cloheptane up to 1.7ꢀ107 mÀ1 for morpholine. In general, the
heterocyclic guests form the more stable complexes. Six-
membered-ring guests are much better encapsulated than
Molecular basket 1 and tube 13: A suspension of tetrol 12 (1.00 g,
0.928 mmol) and tetrachloride 9 (649 mg, 1.21 mmol) in DMF (880 mL)
Chem. Eur. J. 2011, 17, 12362 – 12371
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12369

F. Diederich et al.
was treated dropwise with DIPEA (713 mL, 4.18 mmol). The solution was
stirred at 858C for 18 h, cooled to 258C, and poured on ice-cold saturated
aqueous NaCl solution (2 L). The yellow-orange suspension was stirred
for 1 h. The mixture was extracted with CH2Cl2 (3ꢀ200 mL). The com-
bined organic extracts were washed with saturated aqueous NaCl, dried
over MgSO4, and evaporated in vacuo. Medium-pressure liquid chroma-
tography (MPLC; CH2Cl2/EtOAc 100:0 to 100:10 in 75 min,
50 mLminÀ1) afforded 1 (317 mg, 23%) and 13 (133 mg, 10%) as off-
white solids.
Basket 1: M.p.> 2958C (decomp); 1H NMR (400 MHz, CDCl3, 298 K):
d=0.90–0.94 (m, 12H), 1.31–1.50 (m, 32H), 2.19 (q, J=7.6 Hz, 4H), 2.29
(q, J=7.5 Hz, 4H), 4.60 (s, 4H), 5.53 (t, J=8.2 Hz, 2H), 5.75 (t, J=
8.3 Hz, 2H), 6.81 (s, 4H), 7.20 (s, 4H), 7.89–7.93 (m, 4H), 8.20 (s, 4H),
8.28–8.32 ppm (m, 4H); 13C NMR (100 MHz, CDCl3, 298 K): d=14.19,
14.21, 22.79, 22.80, 28.04, 28.09, 29.43, 29.49, 31.90, 31.95, 32.05, 33.17,
34.04, 34.20, 41.31, 118.97, 123.64, 129.18, 129.70, 129.92, 135.44, 135.73,
136.99, 139.77, 141.95, 152.22, 152.59, 152.83, 158.21, 161.25 ppm; IR
Tube 12: M.p.> 3008C (decomp); 1H NMR (400 MHz, CDCl3, 298 K):
d=1.29 (s, 144H), 3.68–3.75 (m, 16H), 4.92 (br s, 8H), 6.17 (br s, 4H),
6.32 (t, J=8.2 Hz, 4H), 6.84 (br s, 4H), 7.01 (br s, 8H), 7.14 (s, 8H), 7.16
(d, J=1.7 Hz, 8H), 7.23 (t, J=1.7 Hz, 4H), 7.24 (t, J=1.7 Hz, 4H), 7.61
(s, 8H), 7.64 (br s, 4H), 7.90–7.91 (m, 4H), 8.03–8.04 (m, 4H), 8.23–
8.25 ppm (m, 8H); 13C NMR (100 MHz, CDCl3, 298 K): d=31.54, 34.68,
34.84, 38.44, 38.75, 41.10, 108.33, 118.94, 120.38, 120.46, 122.77, 211.79,
124.35, 127.68, 127.76, 134.70, 135.64, 136.71, 137.59, 137.63, 139.71,
141.71, 150.75, 151.99, 152.28, 153.08, 158.67, 162.45 ppm (4 signals are
missing due to overlap or they are hidden in the noise. Due to conforma-
tional equilibria caused by the tilted phenyl bridge, signals are broad-
~
ened); IR (neat): n=2954 (m), 1739 (m), 1598 (w), 1480 (m), 1446 (w),
1412 (m), 1372 (s), 1330 (s), 1248 (w), 1223 (w), 1200 (s), 1157 (m), 1064
(w), 1030 (w), 895 (m), 864 (m), 755 (s), 710 (m), 666 cmÀ1 (w); HR-
+
MALDI-MS: m/z (%): calcd for
C
248H241N20O24
:
3882.8253, found:
3882.8349 (31) [M+H]+, 3883.8354 (69), 3884.8316 (99), 3885.8273 (100),
3886.8257 (79), 3887.8272 (55), 3888.8305 (31), 3889.8371 (13).
~
NMR binding studies: 1H NMR studies were performed with a Bruker
DRX 500 spectrometer. As solvent, specially purified [D12]mesitylene
from ARMAR, 99 Atom%D, was used. The purity was determined by
GC/MS (see Figure 4SI).
(neat): n=2925 (w), 1736 (m), 1481 (m), 1443 (w), 1411 (m), 1369 (s),
1329 (s), 1262 (m), 1220 (w), 1198 (s), 1157 (m), 1072 (m), 894 (m), 760
+
(m), 636 cmÀ1 (w); HR-MALDI-MS: m/z (%): calcd for C88H81N10O12
:
1469.6030, found: 1469.6031 (56, [M+H]+), 1470.6102 (100), 1471.6156
(70), 1472.6193 (33); elemental analysis calcd (%) for C88H80N10O12
(1469.66): C 71.92, H 5.49, N 9.53; found: C 71.44, H 5.68, N 9.48.
For the experiments, which were done in double to quadruple runs, a
stock solution of container molecule 1 in [D12]mesitylene was prepared
(about 3ꢀ10À3 m). A total of 0.6 equiv of 1,3,5-trimethoxybenzene in
[D12]mesitylene (about 2.5ꢀ10À2 m) as internal standard was added direct-
ly to the host stock solution. A total of 600 mL of host solution was trans-
ferred into new NMR tubes, and TMS was added.
Tube 13: M.p.> 3058C (decomp); 1H NMR (400 MHz, CDCl3, 298 K):
d=0.90–0.96 (m, 24H), 1.33–1.51 (m, 64H), 2.24–2.32 (m, 16H), 4.92 (br
s, 8H), 5.61 (br s, 4H), 5.71 (t, J=8.2 Hz, 4H), 6.87 (br s, 4H), 7.01 (br s,
8H), 7.26 (s, 8H), 7.64 (br s, 4H), 7.89–7.93 (m, 4H), 8.01–8.05 (m, 4H),
8.24 ppm (br s, 8H), (no assignment of the signals was carried out due to
very broad signals, high-temperature NMR measurements were carried
out but no sharpening of the lines in the aromatic area was observed);
13C NMR (100 MHz, CDCl3, 298 K): d=14.05, 22.65, 27.95, 29.34, 31.86,
32.30, 32.66, 34.14, 41.09, 118.90, 123.72, 127.70, 128.18, 128.46, 129.25,
129.76, 135.68, 136.95, 139.78, 141.76, 152.04, 152.27, 153.07, 158.68,
162.27 ppm (5 signals are missing due to overlap or they are hidden in
the noise. Due to conformational equilibria caused by the tilted phenyl
Stock solutions of guests in [D12]mesitylene were prepared (about 9ꢀ
10À1 m) and added twice to the host solution (0.4 equiv and 0.8 equiv)
during the measurement at 303 K.
The concentration of free and encapsulated guest was calculated from
the peak area of the internal standard and the encapsulated guest, con-
sidering the corresponding number of protons and concentration of inter-
nal standard as well as the total amount of added guest. Finally, the Ka
value was calculated according to Ka =[HG]/[H]ꢀ[G].
~
bridge, signals are broadened); IR (neat): n=2926 (m), 2856 (w), 1790
ITC analysis: ITC studies were performed using a commercial calorime-
ter MicroCal VP-ITC. As solvent, mesitylene from Acros, 99% extra
pure, was used. The purity was determined by GC/MS (Figure 4SI).
(w), 1738 (s), 1580 (w), 1482 (m), 1445 (w), 1410 (s), 1372 (s), 1327 (s),
1261 (m), 1220 (m), 1200 (s), 1155 (s), 1117 (m), 1080 (m), 942 (w), 895
(m), 764 (s), 735 (m), 617 cmÀ1 (w); HR-MALDI-MS: m/z (%): calcd for
C176H161N20O24+: 2938.1987, found: 2938.2024 (43) [M+H]+, 2939.1988
(89), 2490.1975 (100), 2941.2004 (81), 2942.2092 (49), 2943.2156 (24).
For the experiments, which were done in double to pentuple runs, 25 por-
tions of 10 mL of host solution (about 1ꢀ10À3 m to 4ꢀ10À3 m) were added
to a solution of the guest (about 0.1ꢀ10À3 m) at intervals of 240 s with the
first addition being only 2 mL. The power P that was needed to keep the
sample at 308C was monitored over time t. The heat of dilution for the
addition of host solution to pure mesitylene was measured for all used
concentrations and subtracted. The heat of dilution for the addition of
pure mesitylene into guest solution was measured as well. The measured
values were too small for an evaluation and were neglected.
Molecular basket 2 and tube 11: A suspension of tetrol 8 (650 mg,
0.423 mmol) and tetrachloride 9 (251 mg, 0.465 mmol) in DMF (170 mL)
was treated dropwise with Et3N (265 mL, 1.90 mmol). The resulting solu-
tion was stirred at 708C for 18 h, cooled to 258C, and poured on ice-cold
saturated aqueous NaCl solution (400 mL). The yellow-orange suspen-
sion was stirred for 1 h. The mixture was extracted with CH2Cl2 (2ꢀ
300 mL). The combined organic phases were washed with saturated
aqueous NaCl solution, dried over MgSO4, and evaporated in vacuo. The
yellow solid was taken up in CHCl3 and evaporated to dryness. The
crude product was purified by flash chromatography (SiO2; CH2Cl2/
EtOAc 100:0 to 100:10 in 75 min, 50 mLminÀ1) to afford monomeric 2
(51 mg, 6%) and dimeric 11 (77 mg, 9%) as colorless glasslike solids.
For evaluation of the data, the area of the peaks in the P/t diagram were
evaluated with Origin 7,[38] and the heat of dilution was subtracted. The
resulting sets of data points were then fitted with Origin 7 giving access
to the thermodynamic data and Ka values that are summarized in
Table 1.
Basket 2: M.p.> 3108C (decomp); 1H NMR (400 MHz, CDCl3, 298 K):
d=1.27 (s, 36H), 1.28 (s, 36H), 3.60 (d, J=8.2 Hz, 4H), 3.70 (d, J=
8.2 Hz, 4H), 4.60 (s, 4H), 6.13 (t, J=8.2 Hz, 2H), 6.33 (t, J=8.2 Hz, 2H),
6.82 (s, 4H), 7.07 (d, J=1.7 Hz, 4H), 7.19 (d, J=1.7 Hz, 4H), 7.21 (t, J=
1.7 Hz, 2H), 7.23 (t, J=1.7 Hz, 2H), 7.54 (s, 4H), 7.91–7.93 (m, 4H), 8.20
(s, 4H), 8.29–8.32 ppm (m, 4H); 13C NMR (100 MHz, CDCl3, 298 K): d=
31.64, 31.69, 34.59, 34.95, 34.97, 35.02, 38.11, 39.22, 41.33, 119.01, 120.48,
120.58, 122.85, 122.97, 124.32, 129.17, 129.77, 129.93, 135.45, 135.63,
136.91, 137.69, 137.89 139.78, 142.00, 150.83, 150.90, 152.22, 152.58,
MD calculations: MD calculations were performed using MacroModel
9.7 (molecular dynamics, OPLS 2005 force field, 300 K, Dt=1.5 fs, simu-
lation time 1000 ps).
X-ray analysis: Compound 2, C124H120N10O12 ꢀ CH2Cl2, Mr =2027.27. A
crystal of the size 0.33ꢀ0.27ꢀ0.14 mm was measured at 123 K on a
Bruker-Nonius Kappa-CCD with MoKa radiation, l=0.71073 ꢃ. Mono-
clinic space group P21, 1calcd =1.084 gcmÀ3, Z=2, a=14.2720(6) ꢃ, b=
24.8574(11) ꢃ, c=18.3511(9) ꢃ, b=108.598(2)8, V=6170.4(5) ꢃ3, m=
0.156 mmÀ1. Of the 18110 measured reflexions 13066 were independent
(Rint =0.08). The structure was solved by direct methods (SHELXS-
97),[39] and refined by full-matrix least-squares analysis (SHELXL-97).[39]
Heavily disordered structure with space group ambiguity. Solved and re-
fined in several space groups (orthorhombic C-centered, monoclinic P21/
m and P21) Best result in P21. Possible multiple twin (ca. 60:20:20). Struc-
ture solution depicts the well-defined cage, two complete diisopropyl-
~
152.80, 158.17, 161.21 ppm; IR (neat): n=2954 (m), 1738 (s), 1668 (w),
1599 (m), 1481 (m), 1443 (w), 1412 (s), 1363 (s), 1337 (s), 1265 (m), 1222
(m), 1199 (s), 1158 (s), 1074 (s), 1030 (w), 922 (m), 896 (m), 865 (m), 800
(w), 759 (s), 711 (m), 635 cmÀ1 (m); HR-MALDI-MS: m/z (%): calcd for
C124H121N10O12+: 1941.9160, found: 1941.9211 (54) [M+H]+, 1942.9333
(100), 1943.9370 (78), 1944.9410 (46), 1945.9455 (23).
12370
www.chemeurj.org
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 12362 – 12371

Host–Guest Complexes of Cycloalkanes and Alicyclic Heterocycles
FULL PAPER
phenyl fragments and part of a third one. The fourth substituent had to
be constructed completely. All substituents were included in the refine-
ment as rigid body allowing the isopropyl groups to rotate. Two Cl atoms
have been assigned to two electron peaks in the cavity assuming a
CH2Cl2 molecule with totally disordered C atom (not included). Final
R(F)=0.324, wR(F2)=0.642 for 10165 reflections with I> 2s(I).
[17] V. A. Azov, F. Diederich, Y. Lill, B. Hecht, Helv. Chim. Acta 2003,
86, 2149–2155.
[18] P. P. Castro, G. Zhao, G. A. Masangkay, C. Hernandez, L. M. Gutier-
rez-Tunstad, Org. Lett. 2004, 6, 333–336.
[19] Spartanꢅ08, Macintosh Version 1.0.3, Wavefunction Inc., 18401 Von
Karman Ave., Irvine, 2008.
[20] T. Heinz, D. M. Rudkevich, J. Rebek, Jr., Nature 1998, 394, 764–
766.
[21] P. Roncucci, L. Pirondini, G. Paderni, C. Massera, E. Dalcanale,
V. A. Azov, F. Diederich, Chem. Eur. J. 2006, 12, 4775–4784.
[22] L. Fielding, Tetrahedron 2000, 56, 6151–6170.
CCDC 797074 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[23] J. L. Sessler, D. E. Gross, W.-S. Cho, V. M. Lynch, F. P. Schmidtchen,
G. W. Bates, M. E. Light, P. A. Gale, J. Am. Chem. Soc. 2006, 128,
12281–12288.
Acknowledgements
[24] For ITC measurements on cavitands, see: a) F. Tancini, R. M. Ye-
beutchou, L. Pirondini, R. De Zorzi, S. Geremia, O. A. Scherman,
E. Dalcanale, Chem. Eur. J. 2010, 16, 14313–14321; b) F. Tancini, T.
Gottschalk, W. B. Schweizer, F. Diederich, E. Dalcanale, Chem. Eur.
J. 2010, 16, 7813–7819; c) H. Gan, B. C. Gibb, Supramol. Chem.
2010, 22, 808–814.
[25] F. Diederich, K. Dick, D. Griebel, Chem. Ber. 1985, 118, 3588–3619.
[26] E. Freire, Drug Discovery Today 2008, 13, 869–874.
[27] For polar effects in the complexation of adamantyl derivatives in
rim-functionalized open cavitands, see: R. J. Hooley, S. R. Shenoy, J.
Rebek, Jr., Org. Lett. 2008, 10, 5397–5400.
This work was supported by the European Union through the Marie-
Curie Research Training Network PRAIRIES, contract MRTN-CT-2006–
035810, the Swiss National Science Foundation, and the NCCR “Nano-
scale Science”. L.D.S. acknowledges a NSF-IRFP postdoctoral fellowship
from the US National Science Foundation. We thank Dr. M.-O. Ebert
(ETH) for assistance with NMR spectroscopy, P. Kꢄlin (ETH) for guid-
ance with the ITC measurements, and A. Dutly (ETH) for help with GC/
MS measurements. Prof. Gerhard Klebe (Universitꢄt Marburg) is ac-
knowledged for discussions about limitations of van’t Hoff plot analysis.
[28] R. Paulini, K. Mꢁller, F. Diederich, Angew. Chem. 2005, 117, 1820–
1839; Angew. Chem. Int. Ed. 2005, 44, 1788–1805.
[29] C. Allott, H. Adams, P. L. Bernad, Jr., C. A. Hunter, C. Rotger,
J. A. Thomas, Chem. Commun. 1998, 2449–2450.
[30] For reviews on interactions with aromatic rings, see: a) E. A. Meyer,
R. K. Castellano, F. Diederich, Angew. Chem. 2003, 115, 1244–1287;
Angew. Chem. Int. Ed. 2003, 42, 1210–1250; b) L. M. Salonen, M.
Ellermann, F. Diederich, Angew. Chem. 2011, 123, 4908–4944;
Angew. Chem. Int. Ed. 2011, 50, 4808–4842.
[31] For other reviews on the mechanisms of molecular recognition, see:
a) I. K. Mati, S. L. Cockroft, Chem. Soc. Rev. 2010, 39, 4195–4205;
b) H.-J. Schneider, Angew. Chem. 2009, 121, 3982–4036; Angew.
Chem. Int. Ed. 2009, 48, 3924–3977.
[1] T. Gottschalk, B. Jaun, F. Diederich, Angew. Chem. 2006, 119, 264–
268; Angew. Chem. Int. Ed. 2007, 46, 260–264.
[2] T. Gottschalk, P. D. Jarowski, F. Diederich, Tetrahedron 2008, 64,
8307–8317.
[3] J. R. Moran, S. Karbach, D. J. Cram, J. Am. Chem. Soc. 1982, 104,
5826–5828.
[4] D. J. Cram, J. M. Cram, Container Molecules and Their Guests, The
Royal Society of Chemistry, Cambridge, 1997.
[5] E. Dalcanale, P. Soncini, G. Bacchilega, F. Ugozzoli, J. Chem. Soc.
Chem., Commun. 1989, 500–502.
[6] S. Zampolli, P. Betti, I. Elmi, E. Dalcanale, Chem. Commun. 2007,
2790–2792.
[7] V. A. Azov, B. Jaun, F. Diederich, Helv. Chim. Acta 2004, 87, 449–
462.
[8] R. J. Hooley, H. Van Anda, J. Rebek, Jr., J. Am. Chem. Soc. 2006,
128, 3894–3895.
[9] E. Mann, J. Rebek, Jr., Tetrahedron 2008, 64, 8484–8487.
[10] L. D. Shirtcliff, H. Xu, F. Diederich, Eur. J. Org. Chem. 2010, 846–
855.
[11] S. Mecozzi, J. Rebek, Jr., Chem. Eur. J. 1998, 4, 1016–1022.
[12] M. Yoshifuji, K. Shimura, K. Toyota, Bull. Chem. Soc. Jpn. 1994, 67,
1980–1983.
[32] MacroModel, Version 9.7, Schrçdinger LLC, New York, 2009.
[33] G. J. Kleywegt, VOIDOO, version 3.3.4, Uppsala Software Factory,
Uppsala, 2007.
[34] T. Gottschalk, Dissertation, Eidgençssische Technische Hochschule
Zꢁrich (Switzerland), 2008, Number 18067.
[35] L. Pirondini, D. Bonifazi, B. Cantadori, P. Braiuca, M. Campagnolo,
R. De Zorzi, S. Geremia, F. Diederich, E. Dalcanale, Tetrahedron
2006, 62, 2008–2015.
[36] M. Zꢁrcher, T. Gottschalk, S. Meyer, D. Bur, F. Diederich, Chem-
MedChem 2008, 3, 237–240.
[13] V. A. Azov, P. J. Skinner, Y. Yamakoshi, P. Seiler, V. Gramlich, F.
Diederich, Helv. Chim. Acta 2003, 86, 3648–3670.
[14] M. J. Plater, S. Aiken, G. Bourhill, Tetrahedron 2002, 58, 2405–2413.
[15] P. J. Skinner, A. G. Cheetham, A. Beeby, V. Gramlich, F. Diederich,
Helv. Chim. Acta 2001, 84, 2146–2153.
[37] H. A. Favre, D. Hellwinkel, W. H. Powell, H. A. Smith, Jr., S. S.-C.
Tsay, Pure Appl. Chem. 2002, 74, 809–834.
[38] Origin 7, OriginLab Corporation, Northampton, 2002.
[39] G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112–122.
[16] M. G. J. Beets, W. Meerburg, H. Vanessen, Recl. Trav. Chim. Pays-
Bas 1959, 78, 570–585.
Received: June 17, 2011
Published online: September 21, 2011
Chem. Eur. J. 2011, 17, 12362 – 12371
ꢂ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12371