Resorcin[4]arene-based molecular baskets and water-soluble container molecules: Synthesis and 1H NMR Host-guest complexation studies

Article abstract of DOI:10.1002/ejoc.201402140

Resorcin[4]arene-based molecular baskets, with four free or methylene-bridged HO groups, and water-soluble container molecules bearing poly(ethylene glycol) chains of different lengths on the lower rim and cap have been synthesized. These cavitands, toppe

Full text of DOI:10.1002/ejoc.201402140

FULL PAPER
DOI: 10.1002/ejoc.201402140
Resorcin[4]arene-Based Molecular Baskets and Water-Soluble Container
1
Molecules: Synthesis and H NMR Host–Guest Complexation Studies
[a]
[a]
[a]
[a]
ˇ
Daniel Fankhauser, Dusan Kolarski, Wolfram R. Grüning, and François Diederich*
Keywords: Supramolecular chemistry / Host–guest systems / Molecular recognition
Resorcin[4]arene-based molecular baskets, with four free or
methylene-bridged HO groups, and water-soluble container
molecules bearing poly(ethylene glycol) chains of different
lengths on the lower rim and cap have been synthesized.
These cavitands, topped with p-xylylene bridges, feature
well-defined cavities capable of encapsulating heteroali-
cyclic guests. Association constants (Ka) were determined by
1H NMR spectroscopy for the organic-soluble molecular bas-
kets in CDCl3 and for the water-soluble container molecules
in D2O/CD3CN (2:1). Opposite guest selectivities were ob-
served in the two environments. Upon complexation, the
water-soluble hosts show changes in their 1H NMR spectra.
In the absence of guests, the p-xylylene bridge rotates
rapidly on the 1H NMR timescale, revealing a time-averaged
achiral C2v structure, whereas this rotation is hindered by
guest inclusion, resulting in spectra showing a racemic C2-
symmetric host, indicative of planar chirality.
Introduction
In 1982, Cram and co-workers introduced a new class of
synthetic organic receptors comprised of resorcin[4]arenes
bridged by quinoxaline flaps.[1] These compounds can be
switched from a closed vase to an open kite conformation
and inspired organic chemists to develop more complex
structures to serve as model systems for biological processes
by monitoring their host–guest complexation behavior.[2,3]
Studies in organic solvents[4–9] and in aqueous media[10–16]
have been reported, with water-soluble cavitands[17] being
of special interest for potential medical applications as re-
cently demonstrated by Hooley and co-workers.[18]
Herein we describe the synthesis and binding properties
of the molecular baskets 1 and 2 and the water-soluble con-
tainer molecules 3a,b (Figure 1). In earlier work, we ob-
served that a cleft-type resorcin[4]arene-based cavitand with
two flexible quinoxaline wall flaps and four free HO groups
in the octol-derived bottom selectively recognizes steroidal
substrates in CDCl3.[4] We have now modified this system
by introducing rigidly bridged wall flaps to generate the
new cavitands 1 and 2 with better defined cavities for host–
guest complexation.
Figure 1. Molecular baskets 1 and 2, and water-soluble container
molecules 3a,b.
bility is achieved through the covalent attachment of poly-
(ethylene glycol) (PEG) chains[16,19] in both the legs and the
top bridge of cavitands 3a,b. We chose PEG chains over
charged water-solubilizing groups such as ammonium,[20]
carboxylate,[12] phosphate,[21] or sulfate[20] salts to limit po-
tential interactions with the diverse guests studied in in-
clusion complexation. We describe how the PEG groups at-
tached to the bridge change the planar chirality properties
In a second approach, we report the synthesis of novel
water-soluble container molecules that are prevented by
bridges at the top of the molecule from undergoing undesir-
able dimerization in the kite form, which is frequently ob-
served in water for top-open systems.[10,11,15,16] Water solu-
1
of the molecule, as reflected in the H NMR spectra.[22–24]
Results and Discussion
[a] Laboratorium für Organische Chemie, ETH Zürich, HCI,
Vladimir-Prelog-Weg 3, 8093 Zürich, Switzerland
E-mail: diederich@org.chem.ethz.ch
Synthesis of the Receptors
The synthesis of the molecular baskets 1 and 2 is out-
lined in Scheme 1, A. Under high dilution conditions, octol
4[25,26] was treated with bridge 5[9] in the presence of DBU
http://www.diederich.chem.ethz.ch/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201402140.
Eur. J. Org. Chem. 2014, 3575–3583
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3575

D. Fankhauser, D. Kolarski, W. R. Grüning, F. Diederich
FULL PAPER
Scheme 1. Synthesis of target compounds 1, 2, and 3a,b. Reagents and conditions: a) 5,[9] DBU, DMA, MW, 140 °C, 1 h, 34%;
b) CH2ClBr, DBU, DMA, pressure tube, 80 °C, 2 d, 55%; c) 7, K2CO3, DMA, 60 °C, 20 h, 67% (8a), 96% (8b); d) CsF, catechol, DMF,
80 °C, 1.5 h, 59% (9a), 38% (9b); e) 10a or 10b, quinuclidine, 1,4-dioxane, 60 °C, 40 h, 3% (3a), 6% (3b). DBU = 1,8-diazabicyclo[5.4.0]-
undec-7-ene; DMA = N,N-dimethylacetamide; DMF = N,N-dimethylformamide; MW = microwave.
as base to afford tetrol basket 1 in 34% yield. The two pairs scopic data fully supporting their molecular structures (see
of neighboring phenolic groups in 1 were linked through a Figures 1SI–24SI in the Supporting Information).
methylene bridge by using CH2ClBr and DBU to yield bas-
ket 2 in 55% yield.[27–29]
For the synthesis of container molecules 3a,b (Scheme 1,
Host–Guest Binding Studies with Molecular Baskets 1 and 2
B), PEGylated octols 6a,b[16] were treated under basic con-
ditions with 2,3-dichloroquinoxaline (7) to give cavitands
Binding studies on molecular basket 1 were performed
8a,b.[1,27,30,31] Two of the side-flaps were then selectively re- by using 1H NMR spectroscopy in CDCl3 at 298 K. Several
moved by using CsF and catechol to afford tetrols 9a,b.[4,32] heteroalicyclic substrates as well as cyclohexane and the
Introduction of the PEGylated bridges 10a,b (for their syn- steroids progesterone and cortisone acetate[4] were used as
thesis, see the Supporting Information) afforded container guests (Figure 2). The side-open cavities of the molecular
molecules 3a,b in low yields of 3 and 6%, respectively.[9] All baskets are readily accessible to small guests, and host–
the compounds were fully characterized, with the spectro- guest exchange kinetics is fast on the 1H NMR timescale at
3576
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 3575–3583

Resorcinarene Molecular Baskets and Container Molecules
298 K. Association constants Ka (m–1) were obtained from available, or the guest is too small or too large, the cavity
binding titrations, following the complexation-induced has to distort, and 3) in the case of distortion, the C2v sym-
change in the chemical shifts of host protons.[33] Detailed metry of the host breaks down, previously chemically
protocols are given in the Exp. Sect.
equivalent protons become different, and more signals ap-
pear in the H NMR spectrum.
1
We concluded that host 1 in CDCl3 exists in two forms, a
C2v-symmetric species (C2v-1) and a C1-symmetric racemate
(C1-1). These two species are in equilibrium, transformed
by twisting a set of diaryl ether bonds (Figure 3, a), and
can be clearly differentiated by the resonances of the phen-
olic HO groups. Whereas C2v-1 shows one signal for all four
HO groups, C1-1 features four individual HO signals (Fig-
ure 3, b). By heating the sample to 323 K, only one HO
signal can be observed, which indicates fast equilibration on
1
the H NMR timescale (see Figure 25SI in the Supporting
Information).
Upon guest complexation in the titration studies, only
one C2v-symmetric host is observed, as illustrated in Fig-
ure 26SI for the titration with 1,4-dioxane as guest. The ob-
Figure 2. Guest molecules used for binding studies.
1H NMR analysis of molecular basket 1 in [D8]1,4-diox- tained titration curves (see Figures 27SI–54SI in the Sup-
ane at 298 K indicated the presence of a single host species, porting Information) for molecular basket 1 were evaluated
presumably forming a 1:1 inclusion complex with the sol- by using the software IGOR Pro.[34] The Ka values deter-
vent (see Figure 1SI, a in the Supporting Information). On mined by monitoring the chemical shift of the HO signal
the other hand, in CDCl3 at 298 K, two sets of host signals and the average Ka values obtained by monitoring the
were detected. We rationalized this observation as follows: chemical shifts of the aromatic lower rim and/or methine
1) The host cavity can never be empty, 2) if there is no guest protons are summarized in Table 1.
1
Figure 3. Model explaining the H NMR observations in CDCl3. a) Host 1 exists in a C2v- and a racemic C1-symmetric form, which are
1
in slow equilibrium on the H NMR timescale at room temperature and below, and are interconverted by twisting diaryl ether bonds.
b) 1H NMR spectrum (500 MHz, CDCl3, 283 K) of host 1 showing two sets of host signals. Whereas the C2v-symmetric conformer only
shows one signal for all HO groups, the C1-symmetric conformer generates four different HO resonances.
Eur. J. Org. Chem. 2014, 3575–3583
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3577

D. Fankhauser, D. Kolarski, W. R. Grüning, F. Diederich
FULL PAPER
Table 1. Results of 1H NMR titration studies with molecular basket phthalimide C=O groups, as shown in Figure 4. Geometries
1 as host in CDCl3 at 298 K. The Ka values obtained are deter-
showing these interactions were preferred over those in
which the guest engages in hydrogen-bonding interactions
with the phenolic HO groups.
mined from different host protons.
Guest
Ka [m–1][a]
Ka [m–1][b]
1,4-Dioxane[c]
1,4-Thioxane
1,4-Dithiane
Oxolane
Oxane
Thiolane
43.2Ϯ1.5[d]
15.9Ϯ0.2[e]
28.7Ϯ1.4[d]
12.2Ϯ1.3[e]
[f]
[f]
–
–
48.6Ϯ2.2[d]
38.6Ϯ2.3[d]
9.1Ϯ0.2[d]
9.0Ϯ0.3[d]
39.3Ϯ8.0[e]
27.5Ϯ1.4[e]
6.2Ϯ2.1[d]
8.0Ϯ1.3[d]
Thiane
[f]
[f]
Cyclohexane
Morpholine
Progesterone
Cortisone acetate
–
–
–
[g]
710Ϯ91[e]
[g]
–
43.7Ϯ2.2[d]
66.8Ϯ18.9[e]
134Ϯ15[d]
[a] Ka values determined by monitoring the chemical shift of the
HO signal. [b] Average Ka values determined by monitoring the
chemical shifts of the aromatic lower rim and/or methine protons.
[c] Reproducibility was checked in duplicate runs. [d] Standard de-
viation of titration curve-fitting. [e] Standard deviation of averaged
Ka values. [f] No binding was observed. [g] No observable HO sig-
nal for evaluation.
In the heteroalicyclic series, the Ka values range from
6.2 m–1 for thiolane up to 710 m–1 for morpholine. However,
the Ka values determined by monitoring the strongly shift-
ing HO signal are larger by a factor of 1.1–2.7 than the
average Ka values determined from the more weakly shifting
aromatic lower rim and/or methine proton resonances.
However, the relative order of binding constants for this set
of guests remains the same. A reasonable explanation for
the discrepancy between the two sets of data is the higher
sensitivity of HO protons due to hydrogen bonding,[35]
coming with the drawback of higher possible errors.
Oxygen-containing guests form more stable host–guest
complexes than guests containing sulfur, with the binding
strength increasing from 1,4-dithiane to 1,4-thioxane to 1,4-
dioxane. The steroids progesterone and cortisone acetate
show an affinity towards host 1. According to molecular
modeling studies conducted with moloc,[36] they are too big
to fit inside the rigidified cavity of 1, and presumably they
undergo side-on hydrogen-bonding interactions with the
phenolic HO groups. Morpholine shows the strongest bind-
ing but, due to its basicity, causes decomposition of host 1
over time. With cyclohexane as a guest without heteroatoms
capable of undergoing polar interactions, no binding was
observed.
Figure 4. Lowest-energy conformation of molecular basket 1 with
encapsulated 1,4-dioxane, calculated by using MacroModel 9.7[40]
(OPLS 2005 force field, GB/SA solvation model for CHCl3) and
Spartan ’14[41] (PM3). The host–guest complex is stabilized by
C–O···C=O interactions. For simplification, the hexyl legs have
been replaced by methyl groups.
We also performed 1H NMR binding titrations with mo-
lecular basket 2 in CDCl3 at 298 K using the same series of
guests. The pure cavitand showed only one set of signals in
CDCl3 (see Figure 55SI in the Supporting Information) as
a result of the rigidification caused by the introduction of
additional methylene bridges. No significant changes in
chemical shift were observed during the titrations
(ΔδϽ0.01 ppm), which indicates no or only very weak
host–guest inclusion complexation. Although the removal
of the hydrogen-bond-donating HO groups might contrib-
ute to the lack of complexation ability of host 2, we also
propose that the two additional methylene bridges enhance
the rigidity and reduce the adaptability of the receptor.
In addition to dispersion and C–H···π interactions,[37–39]
polar interactions contribute to the observed host–guest
complexation. We analyzed the complexation of 1 with 1,4-
dioxane by using moloc,[36] which suggested that temporary
Host–Guest Binding Studies with Water-Soluble Container
Molecule 3b
The attachment of PEG chains to our previously de-
hydrogen bonding of the guest to the phenolic HO groups scribed container molecules[9] enhanced their solubility in
of the host initiates the formation of the host–guest com- aqueous media. Although compound 3a with diethylene
plex. A subsequent conformational search and energy mini- glycol chains is soluble in 55:45 MeCN/water, host 3b with
mization of the complex using MacroModel (OPLS 2005 tetraethylene glycol chains readily dissolves in 33:67 MeCN/
force field)[40] followed by further energy minimizations water. Methanol is a less effective co-solvent, and 3a dis-
using PM3 in Spartan[41] revealed the preferential orienta- solves in 86:14 MeOH/water, whereas host 3b is soluble in
tion of the guest inside the cavity. According to this analy- 48:52 MeOH/water. The 1H NMR spectra of both cavitands
sis, 1,4-dioxane binds in a conformation that allows stabiliz- in D2O/CD3CN show sharp signals. Their methine protons,
ing orthogonal C–O···C=O interactions[42] with the diaza- which are diagnostic of the vase and kite conformations,
3578
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 3575–3583

Resorcinarene Molecular Baskets and Container Molecules
1
appear below 5 ppm, characteristic of a vase-like form. In
Binding studies by H NMR spectroscopy were carried
contrast, their spectra in D2O/MeOD show very broad sig- out with host 3b in D2O/CD3CN (2:1) at 298 K using barely
nals, which are not suitable for an accurate determination soluble cyclohexane and the heteroalicyclic guests shown in
of Ka values. We found no evidence in D2O/CD3CN for the Figure 2. With four wall flaps, guest access to the cavity is
frequently described host dimerizations in aqueous me- sterically hindered, and slow host–guest exchange kinetics
1
dia.[10,11,15,16] Only traces of dimeric 3a,b were detected by is observed on the H NMR timescale at 298 K. This en-
high-resolution matrix-assisted laser desorption/ionization abled the determination of Ka values by integrating the sig-
mass spectrometry (HR-MALDI-MS). Dimerization usu- nal intensities of the free and encapsulated guest.[33] Al-
ally involves association of the open kite conformation, though the signals of bound guests in host 3b appear in the
which cannot be adopted by our basket-type systems with range of 0.2 to –3.4 ppm and are easily integrable, many
their capping, PEG-garnered p-xylylene bridge.
signals of the free guests coincide with the resonances of the
Figure 5. Complexation of 1,4-dithiane by container 3b. a) Chemical structure of the receptor (left) and lowest-energy conformation of
the inclusion complex with PEG chains on the p-xylylene bridge and simplified methyl legs (right). MacroModel 9.7[40] (OPLS 2005 force
field, GB/SA solvation model for H2O) and Spartan ’14[41] (PM3) were used to calculate the structure. b) Monitoring of the complexation
process by 1H NMR spectroscopy [500 MHz, D2O/CD3CN (2:1), 298 K]. The bottom spectrum depicts the free host and the top spectrum
shows the complexation of 3b (0.5 mm) with 1,4-dithiane (1.2 equiv.). The spectrum of the free host reveals a time-averaged C2v-symmetric
structure with fast rotation of the PEGylated p-xylylene bridge. The top spectrum shows a solution of the complex, in which this rotation
1
is slow on the H NMR timescale, leading to a doubling of resonances in accord with a C2-symmetric host structure.
Eur. J. Org. Chem. 2014, 3575–3583
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3579

D. Fankhauser, D. Kolarski, W. R. Grüning, F. Diederich
FULL PAPER
PEG chains in the host, which makes a proper integration of signals in the free form, indicative of a time-averaged C2v
impossible. We circumvented this problem by using 1,3,5- symmetry, as the absence of guest and greater flexibility al-
trimethoxybenzene as internal standard, determining by in- low the PEG-substituted bridge ring to rotate freely on the
tegration only the concentration of the bound guest.[9] The 1H NMR timescale at 298 K. Upon guest inclusion, the res-
detailed procedure for the 1H NMR binding studies is given onances of the octol bowl protons, as well as of the CH2
in the Exp. Sect. All the chemical shifts of the free and groups in the p-xylylene bridge split into two sets of signals
encapsulated guests in D2O/CD3CN (2:1) at 298 K are (Figure 5 and Figures 57SI and 58SI in the Supporting In-
shown in Table 1SI in the Supporting Information.
formation). As a result of the rigidification and the cavity
1H NMR binding studies on 3b were performed at con- occupancy by the guest, the rotation of the PEG-substi-
stant host concentration and three different guest concen- tuted bridge ring is slowed down, thereby establishing the
trations by monitoring the signal ratio of the internal stan- planar chirality of the host, which now appears as a C2-
dard and the bound guest. The three individual values of symmetric structure. The latter is observed in the 1H NMR
Ka calculated per host–guest complex formed a sequence spectrum up to 348 K.
of decreasing magnitude, and slow host decomposition was
Container molecule 3a shows the vase form in the free
1
observed in the H NMR spectrum. Therefore we report state in D2O/CD3CN (45:55), and rotation of the p-xylylene
herein only the Ka values obtained from the first runs, bridge with its shorter PEG chains is fast on the NMR
which are summarized in Table 2. The binding study with time scale (see Figure 59SI in the Supporting Information).
1,4-dithiane is shown in Figure 5 (for the study with 1,4- Similarly to 3b, inclusion of an appropriate guest, such as
dioxane, see Figure 56SI in the Supporting Information). 1,4-dioxane, slows down the rotation of the bridge in 3a,
The guest resonances are strongly shifted upfield upon and the spectra of both hosts in [D8]1,4-dioxane (see Fig-
complexation, from 3.25 ppm in the free guest to –1.37 ppm ures 3SI, a and 4SI, a) show C2 symmetry as a result of
in the bound state.
planar chirality.
Table 2. 1H NMR binding studies with container molecule 3b as
host in D2O/CD3CN (2:1) at 298 K.
Conclusions
Guest
Ka [m–1]
Resorcin[4]arene-based molecular baskets with free and
methylene-bridged HO groups, as well as water-soluble con-
tainer molecules bearing PEG chains of different length as
legs and as substituents of the p-xylylene cap, have been
synthesized. Their host–guest complexation behavior was
1,4-Dioxane
1,4-Thioxane
1,4-Dithiane
Oxolane
Oxane
Cyclohexane
Morpholine
1952
2021
3724
[a]
–
941
200
1
[b]
investigated by H NMR binding studies using a variety of
–
small heteroalicyclic guests. Dispersion and C–H···π inter-
actions, in addition to polar interactions such as C–
O···C=O or S···π interactions, stabilize the host–guest com-
[a] No binding was observed. [b] Decomposition of host.
In general, the Ka values range from 200 m–1 for cyclo- plexes, in addition to solvophobic effects in aqueous solu-
hexane up to 3724 m–1 for 1,4-dithiane. In aqueous solution, tion. Complexation studies with molecular basket 1 in
the more soluble, polar oxygen-containing guests form less CDCl3 revealed better complexation of O-containing than
stable complexes than the less soluble sulfur-containing S-containing guests (Ka values: 1,4-dithianeϽ1,4-thiox-
guests, which have a higher tendency to partition from the aneϽ1,4-dioxane), whereas the investigations with the
aqueous phase into the less polar cavitand interior.[43] The water-soluble container molecule 3b in D2O/CD3CN (2:1)
binding strength increases from 1,4-dioxane (Ka = 1952 m–1) showed the reverse trend. With its higher capacity for guest
to 1,4-thioxane (Ka = 2021 m–1) to 1,4-dithiane (Ka
=
encapsulation and the aqueous solution favoring apolar
3724 m–1). In addition to the enhanced partitioning, S···π binding, the Ka values for the complexes of container 3b in
interactions[37,39] in the complex seem to contribute to the aqueous solution are greater than those measured for the
better binding in the complex, as also suggested by the com- complexes of the more open cavitand 1 in CDCl3. Complex-
puter simulations for bound 1,4-dithiane (Figure 5).
ation by molecular basket 1 in CDCl3 is fast on the 1H
Complexation affects the planar chirality properties of NMR timescale, whereas the encapsulation of guests by
the container molecules. Host 3b adopts a vase-like form in container molecule 3b is slow. The PEG-substituted p-xylyl-
both the free and bound state. However, the free vase is ene cap efficiently prevents dimerization of the container
much more flexible than the complex vase. In the free state, molecules in aqueous solution. Guest complexation by
the resonances of the methine protons below the two flexi- hosts 3a,b leads to changes in their 1H NMR spectra:
ble quinoxaline flaps appear at δ = 5.33 ppm. Upon guest Whereas the free hosts appear as time-averaged C2v-sym-
inclusion, the vase rigidifies and the methine resonance metric structures, complexes with 1,4-dioxane or 1,4-dithi-
shifts downfield to 6.11 ppm in the case of 1,4-dithiane ane appear as C2-symmetric, as guest inclusion rigidifies the
(Figure 5), whereas the methine protons under the bridged containers and slows down the rotation of the capping
wall flaps appear at the same position in both the unbound PEG-substituted p-xylylene rings, reinforcing the planar
and bound state, at δ = 5.97 ppm. The host shows one set chirality of the systems. In future work we intend to com-
3580
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 3575–3583

Resorcinarene Molecular Baskets and Container Molecules
33.06, 34.58, 37.03, 41.86, 100.22, 118.63, 122.61, 131.31, 136.36,
137.48, 140.69, 143.24, 152.92, 156.22, 158.63, 163.20 ppm. IR
bine the knowledge gained on rendering container resor-
cin[4]arene cavitands soluble with the recently reported re-
dox-triggered switching properties to develop new switch-
able receptors for use in aqueous solution.[44,45]
(ATR): ν = 2927 (m), 2856 (m), 1793 (w), 1735 (s), 1607 (w), 1578
˜
(w), 1538 (w), 1487 (m), 1439 (m), 1368 (s), 1337 (s), 1277 (m),
1198 (s), 1152 (m), 1138 (m), 1069 (m), 970 (s), 923 (m), 891 (m),
798 (m), 739 (m), 718 (m), 633 (s) cm–1. HRMS (MALDI-TOF,
DCTB): m/z (%)
= 1279.5150 (9) [M +
K]+ (calcd. for
Experimental Section
+
C74H76KN6O12 1279.5153), 1265.5477 (11), 1264.5442 (24),
1263.5410 (30) [M + Na]+ (calcd. for C74H76N6NaO12+ 1263.5413),
685.4356 (87), 437.1934 (100), 354.1699 (40).
General Details and Synthetic Procedures: Octol 4,[25] PEGylated
octol 6b,[11,16] and bridge 5[9] were prepared according to literature
procedures. 2,3-Dichloroquinoxaline (7) was bought from ABCR.
The experimental details for the synthesis and characterization of
compounds 1, 2, 3a, and 3b are described in this manuscript, for
the synthesis and characterization of compounds 6a, 8a,b, 9a,b, and
10a,b, see the Supporting Information. The atom numbering used
Container Molecule 3a: A solution of tetrol 9a (200 mg, 145 μmol)
in dry 1,4-dioxane (370 mL) was treated with a spatula of molecu-
lar sieves (3 Å) and a solution of quinuclidine (67.6 mg, 608 μmol)
in dry 1,4-dioxane (10 mL), and the mixture was heated to 60 °C.
A solution of bridge 10a (118 mg, 152 μmol) in dry 1,4-dioxane
(20 mL) was added over 40 min, and stirring was continued at
60 °C for 40 h. The mixture was filtered through a short plug of
SiO2, the filtrate treated with a spatula of silica (500 mg), evapo-
rated, and the crude purified by MPLC (SiO2; CH2Cl2/THF/
MeOH, 96:2:2 to 92:4:4 in 30 min, 92:4:4 for 20 min, 40 mLmin–1)
and HPLC (diol-phase; Nucl. 7 OH, Macherey–Nagel; n-hexane/
CH2Cl2/THF/MeOH, 75:25:2:2, 18 mLmin–1) to yield 3a (8 mg,
3%) as a yellow waxy solid. Rf = 0.22 (SiO2; CH2Cl2/THF/MeOH,
1
to assign the H NMR signals and the naming of the compounds
by phane nomenclature[46,47] are reported in the Supporting Infor-
mation.
Molecular Basket 1: A solution of octol 4[25] (25 mg, 30 μmol) and
arene-bridge 5[9] (16 mg, 30 μmol) in DMA (10 mL) was treated
with DBU (22.8 μL, 150 μmol) and stirred at 140 °C under MW
irradiation for 1 h. A spatula of SiO2 (200 mg) was added, the mix-
ture was evaporated (HV, 50 °C), and the crude purified through a
short FC plug (SiO2; CH2Cl2/EtOAc, 8:2) to yield 1 (15 mg, 34%)
as a slightly yellow solid. Rf = 0.70 (SiO2; CH2Cl2/EtOAc, 8:2),
1
92:4:4). H NMR (600 MHz, [D8]1,4-dioxane): δ = 1.55 [quint., J
= 8.2 Hz, 4 H, H2C(2Јl)], 1.68 [quint., J = 7.2 Hz, 4 H, H2C(2l)],
2.29 [q, J = 8.2 Hz, 4 H, H2C(1Јl)], 2.40–2.47 [m, 4 H, H2C(1l)],
3.25 [s, 6 H, H3C(15Јf)], 3.27 [s, 6 H, H3C(8Јl)], 3.28 [s, 6 H,
H3C(8l)], 3.41–3.88 [m, 56 H, H2C(3Јl–7Јl, 3l–7l, 11Јf–14Јf)], 4.29
and 5.08 [2d, J = 13.6 Hz, 4 H, H2C(4Јf)], 5.46 [t, J = 8.3 Hz, 2 H,
H–C(1r)], 5.61 [t, J = 8.2 Hz, 2 H, H–C(1Јr)], 6.75 [s, 2 H, H–
C(7Јf)], 7.34 and 7.39 [2s, 4 H, H–C(2Јr, 2r)], 7.81–7.83 and 7.87–
7.90 [2m, 4 H, H–C(4f, 5f)], 7.86 and 8.00 [2s, 4 H, H–C(5Јr, 5r)],
8.12–8.15 [m, 4 H, H–C(3f, 6f)] ppm. 13C NMR (150 MHz, [D8]1,4-
dioxane): δ = 27.98, 28.46, 28.74, 30.35, 34.47, 35.06, 36.54, 58.90,
58.91, 70.13, 70.74, 70.94, 70.97, 71.04, 71.14, 71.18, 71.20, 71.26,
71.31, 72.72, 72.75, 72.83, 118.46, 118.65, 119.17, 123.87, 125.06,
127.95, 129.05, 129.08, 130.62, 131.36, 136.29, 136.52, 136.80,
136.81, 140.24, 140.37, 143.14, 143.57, 151.61, 152.08, 152.38,
152.96, 153.15, 153.50, 153.78, 157.82, 157.90, 161.64, 164.01 ppm
1
m.p. Ͼ250 °C (decomp.). H NMR (500 MHz, [D8]1,4-dioxane): δ
= 0.89 [t, J = 6.9 Hz, 6 H, H3C(6Јl); partial overlap with next reso-
nance], 0.90 [t, J = 7.0 Hz, 6 H, H3C(6l)], 1.25–1.43 [m, 32 H,
H2C(2Јl–5Јl, 2l–5l)], 2.15 [q, J = 8.1 Hz, 4 H, H2C(1Јl)], 2.23 [q, J
= 7.8 Hz, 4 H, H2C(1l)], 4.36 [t, J = 7.8 Hz, 2 H, H–C(1r)], 4.73 [s,
4 H, H2C(4Јf)], 5.42 [t, J = 8.1 Hz, 2 H, H–C(1Јr)], 7.04 [s, 4 H, H–
C(5r)], 7.19 [s, 4 H, H–C(2r)], 7.33 [s, 4 H, H–C(6Јf)], 8.58 (s, 4 H,
HO) ppm. 13C NMR (125 MHz, [D8]1,4-dioxane): δ = 14.43, 23.36,
23.38, 28.76, 30.01, 30.07, 30.42, 32.66, 32.69, 33.52, 33.63, 34.05,
34.46, 41.67, 111.78, 124.13, 130.76, 130.97, 131.49, 137.48, 143.18,
153.00, 153.57, 158.82, 163.64 ppm (one signal missing due to over-
lap). IR (ATR): ν = 3337 (br, w), 2926 (m), 2856 (m), 1791 (w),
˜
1733 (s), 1613 (w), 1585 (w), 1489 (m), 1434 (m), 1370 (s), 1337 (s),
1280 (m), 1224 (m), 1199 (s), 1169 (m), 1132 (m), 1073 (s), 923 (m),
905 (m), 857 (m), 800 (m), 740 (m), 637 (m) cm–1. HRMS
(MALDI-TOF, DCTB): m/z (%) = 1255.5157 (20) [M + K]+ (calcd.
(two signals missing due to overlap). IR (ATR): ν = 3478 (br, w),
˜
2870 (m), 1793 (w), 1734 (m), 1657 (w), 1569 (w), 1511 (w), 1483
(m), 1444 (m), 1410 (s), 1364 (s), 1330 (s), 1261 (m), 1197 (s), 1160
(m), 1139 (s), 1088 (br, s), 1020 (m), 944 (m), 907 (m), 877 (m), 851
(m), 762 (m), 673 (m), 603 (m) cm–1. HRMS (MALDI-TOF,
+
for C72H76KN6O12 1255.5153), 1241.5484 (28), 1240.5451 (67),
1239.5415 (84) [M + Na]+ (calcd. for C72H76N6NaO12+ 1239.5413),
1219.5665 (34), 1218.5625 (75), 1217.5578 (100) [M + H]+ (calcd.
DCTB): m/z (%) = 4042.5299 (7) [2M + Na]+ (calcd. for
+
for C72H77N6O12 1217.5594), 1216.5517 (50) [M]+ (calcd. for
+
+
C
212H232N20NaO60 4042.5615), 2034.7848 (25), 2033.7816 (58),
C72H76N6O12 1216.5521), 663.4537 (72), 354.1700 (86).
2032.7777 (100), 2031.7734 (87) [M
+
Na]+ (calcd. for
Molecular Basket 2: A solution of 1 (97 mg, 80 μmol) in DMA
(12 mL) was treated with CH2ClBr (218 μL, 3.19 mmol) and DBU
(60.1 μL, 400 μmol) and stirred in a pressure tube at 80 °C for 2 d
[with extra CH2ClBr (218 μL, 3.19 mmol) after 24 h]. A spatula of
+
C106H116N10NaO30 2031.7751), 2011.7966 (24), 2010.7918 (48),
+
2009.7872 (70), 2008.7826 (54) [M]+ (calcd. for C106H116N10O30
2008.7853).
SiO2 (300 mg) was added, the mixture was evaporated (HV, 50 °C), Container Molecule 3b: A solution of tetrol 9b (200 mg, 115 μmol)
and the crude purified through a short FC plug (SiO2; CH2Cl2/
EtOAc, 9.5:0.5) to yield 2 (54 mg, 55%) as a slightly yellow solid.
Rf = 0.74 (SiO2; CH2Cl2/EtOAc, 95:5), m.p. Ͼ218 °C (decomp.).
1H NMR (500 MHz, [D8]1,4-dioxane): δ = 0.89 [t, J = 6.9 Hz, 6
in dry 1,4-dioxane (370 mL) was treated with a spatula of molecu-
lar sieves (3 Å) and a solution of quinuclidine (55.5 mg, 484 μmol)
in dry 1,4-dioxane (10 mL), and the mixture was heated to 60 °C.
A solution of bridge 10b (115 mg, 121 μmol) in dry 1,4-dioxane
H, H3C(6Јl)], 0.92 [t, J = 6.9 Hz, 6 H, H3C(6l)], 1.25–1.51 [m, 32 (20 mL) was added over 40 min, and stirring was continued at
H, H2C(2Јl–5Јl, 2l–5l)], 2.18 [q, J = 8.2 Hz, 4 H, H2C(1Јl)], 2.28 [q,
J = 8.0 Hz, 4 H, H2C(1l)], 4.05 and 5.75 [2d, J = 8.0 Hz, 4 H,
H2C(8r)], 4.79 [s, 4 H, H2C(4Јf); partial overlap with next reso-
nance], 4.80 [t, J = 8.0 Hz, 2 H, H–C(1r)], 5.56 [t, J = 8.2 Hz, 2 H,
H–C(1Јr)], 7.20 [s, 4 H, H–C(2r)], 7.22 [s, 4 H, H–C(5r)], 7.41 [s, 4 H,
60 °C for 40 h. The mixture was filtered through a short plug of
SiO2, the filtrate treated with a spatula of silica (500 mg), evapo-
rated, and the crude purified by MPLC (SiO2; CH2Cl2/THF/
MeOH, 94:3:3 to 90:5:5 in 30 min, 90:5:5 for 20 min, 40 mLmin–1)
and HPLC (CN-phase; LiChrospher 100 CN 5 μm, Merck; THF,
H–C(6Јf)] ppm. 13C NMR (125 MHz, [D8]1,4-dioxane): δ = 14.39, 18 mLmin–1) to yield 3b (18 mg, 6%) as a slightly yellow oil. Rf =
14.46, 23.35, 23.36, 28.54, 28.57, 30.11, 30.12, 30.40, 32.51, 32.72,
0.13 (SiO2; CH2Cl2/THF/MeOH, 90:5:5). 1H NMR (600 MHz,
Eur. J. Org. Chem. 2014, 3575–3583
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3581

D. Fankhauser, D. Kolarski, W. R. Grüning, F. Diederich
FULL PAPER
[D8]1,4-dioxane): δ = 1.55 [quint., J = 7.6 Hz, 4 H, H2C(2Јl)], 1.68
[quint., J = 6.6 Hz, 4 H, H2C(2l)], 2.31 [q, J = 7.6 Hz, 4 H,
H2C(1Јl)], 2.44–2.51 [m, 4 H, H2C(1l)], 3.26 [s, 12 H, H3C(12Јl, 12l)],
of the host. The complex was further investigated by using Macro-
Model 9.7[40] [conformational search, Monte–Carlo Multiple Mini-
mum (MCMM) algorithm, OPLS 2005 force field, GB/SA sol-
3.27 [s, 6 H, H3C(19Јf)], 3.42–3.88 [m, 104 H, H2C(3Јl–11Јl, 3l–11l, vation model for CHCl3, followed by energy minimization, Polak-
11Јf–18Јf)], 4.29 and 5.08 [2d, J = 13.7 Hz, 4 H, H2C(4Јf)], 5.47 [t, Ribière Conjugate Gradient (PRCG) algorithm]. Afterwards, the
J = 8.2 Hz, 2 H, H–C(1r)], 5.59 [t, J = 7.6 Hz, 2 H, H–C(1Јr)], 6.73 host–guest complex was further energy-minimized by using PM3
[s, 2 H, H–C(7Јf)], 7.42 and 7.46 [2s, 4 H, H–C(2Јr, 2r)], 7.82–7.85 in Spartan ‘14.[41] Complexation of container molecule 3b with en-
and 7.89–7.92 [2m, 4 H, H–C(4f, 5f)], 7.85 and 8.00 [2s, 4 H, H–
capsulated 1,4-dithiane was investigated by using Macro-
C(5Јr, 5r)], 8.13–8.15 [m, 4 H, H–C(3f, 6f)] ppm. 13C NMR Model 9.7[40] and Spartan ’14[41] as described above by using the
(150 MHz, [D8]1,4-dioxane): δ = 27.95, 28.46, 28.69, 30.27, 34.44,
35.07, 36.58, 58.90, 70.08, 70.79, 70.84, 70.91, 70.93, 71.03, 71.11,
71.25, 71.28, 71.30, 71.32, 71.45, 72.69, 118.39, 118.56, 119.02,
124.22, 125.36, 127.91, 129.04, 129.11, 130.74, 131.49, 136.46,
136.60, 136.85, 136.92, 140.22, 140.34, 143.14, 143.56, 151.52,
152.15, 152.43, 152.88, 153.07, 153.45, 153.73, 157.82, 157.89,
161.70, 164.07 ppm (15 signals missing due to overlap). IR (ATR):
GB/SA solvation model for H2O.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures and characterization of the precur-
sors to the container molecules, as well as NMR spectra for all
compounds; method for the 1H NMR assignments of the molecules
and applied phane nomenclature; additional tables and figures.
ν = 3507 (br, w), 2868 (m), 1793 (w), 1735 (m), 1571 (w), 1484 (m),
˜
1411 (m), 1365 (s), 1332 (s), 1254 (m), 1197 (m), 1095 (br, s), 944
(m), 851 (m), 761 (m), 603 (m) cm–1. HRMS (MALDI-TOF,
DCTB): m/z (%) = 5100.2919 (11) [2 M + Na]+ (calcd. for
Acknowledgments
This work was supported by a grant from the Swiss National Sci-
ence Foundation.
+
C260H328N20NaO84 5100.1906), 2563.0999 (38), 2562.0956 (76),
2561.0914 (100), 2560.0888 (63) [M
+
Na]+ (calcd. for
+
C130H164N10NaO42 2560.0897).
[1] J. R. Moran, S. Karbach, D. J. Cram, J. Am. Chem. Soc. 1982,
104, 5826–5828.
[2] D. J. Cram, Science 1983, 219, 1177–1183.
[3] D. J. Cram, J. M. Cram, Container Molecules and Their Guests,
The Royal Society of Chemistry, Cambridge, UK, 1997.
[4] M. Cacciarini, V. A. Azov, P. Seiler, H. Künzer, F. Diederich,
Chem. Commun. 2005, 5269–5271.
[5] T. Gottschalk, B. Jaun, F. Diederich, Angew. Chem. Int. Ed.
2007, 46, 260–264; Angew. Chem. 2007, 119, 264–268.
[6] T. Gottschalk, P. D. Jarowski, F. Diederich, Tetrahedron 2008,
64, 8307–8317.
[7] E. Mann, J. Rebek Jr., Tetrahedron 2008, 64, 8484–8487.
[8] L. D. Shirtcliff, H. Xu, F. Diederich, Eur. J. Org. Chem. 2010,
846–855.
NMR Titrations with Molecular Baskets 1 and 2: 1H NMR ti-
trations were performed at 298 K with a Bruker AV III 500 spec-
trometer (500 MHz) in CDCl3 (99.8 atom-% D, stored over molec-
ular sieves (4 Å) and filtered through basic Al2O3 before use) pur-
chased from Armar. For the experiments, host 1 or 2 was dissolved
in CDCl3 (about 1ϫ10–3 m) and an aliquot of this solution was
used to dissolve the guest (about 2.5ϫ10–1 m). During the titration,
the host solution was successively treated with guest solution by
syringe (about 11 additions) so that the host concentration re-
mained constant and the guest concentration gradually increased
1
(about 0.5–100 equiv.). H NMR spectra were recorded after each
addition and the change in chemical shift (δ) of different host pro-
tons was recorded as a function of guest concentration. The ti-
tration curves obtained were evaluated by using the software IGOR
Pro[34] to give the corresponding Ka values (Table 1).
[9] J. Hornung, D. Fankhauser, L. D. Shirtcliff, A. Praetorius,
W. B. Schweizer, F. Diederich, Chem. Eur. J. 2011, 17, 12362–
12371.
[10] T. Haino, D. M. Rudkevich, J. Rebek Jr., J. Am. Chem. Soc.
1999, 121, 11253–11254.
[11] T. Haino, D. M. Rudkevich, A. Shivanyuk, K. Rissanen, J. Re-
bek Jr., Chem. Eur. J. 2000, 6, 3797–3805.
[12] F. Hof, L. Trembleau, E. C. Ullrich, J. Rebek Jr., Angew. Chem.
Int. Ed. 2003, 42, 3150–3153; Angew. Chem. 2003, 115, 3258–
3261.
[13] S. M. Biros, E. C. Ullrich, F. Hof, L. Trembleau, J. Rebek Jr.,
J. Am. Chem. Soc. 2004, 126, 2870–2876.
[14] R. J. Hooley, H. J. Van Anda, J. Rebek Jr., J. Am. Chem. Soc.
2006, 128, 3894–3895.
[15] R. J. Hooley, H. J. Van Anda, J. Rebek Jr., J. Am. Chem. Soc.
2007, 129, 13464–13473.
NMR Binding Studies with Container Molecule 3b: 1H NMR bind-
ing studies were performed at 298 K with a Bruker AV III 500
spectrometer (500 MHz) in a mixture of D2O (99.8 atom-% D) and
CD3CN (99.8 atom-% D) purchased from Armar. For the experi-
ments, host 3b was dissolved in D2O/CD3CN (2:1, about
5ϫ10–4 m), an aliquot of this solution was used to dissolve the
guest (about 2.5ϫ10–2 m, depending on solubility) and another ali-
quot to dissolve 1,3,5-trimethoxybenzene (about 3ϫ10–2 m) as an
internal standard. During the binding studies, the host solution was
treated with a solution of the internal standard (about 1.0 equiv.)
and successively with the guest solution by syringe (0.8, 1.0, and
1.2 equiv.) so that the host concentration remained constant and
the guest concentration gradually increased. 1H NMR spectra were
recorded after each guest addition, and the concentrations of the
free and encapsulated guest were calculated by integration of the
peak areas of the internal standard and encapsulated guest, con-
sidering the corresponding number of protons and concentration
of the internal standard as well as the total amount of guest. The
Ka values were calculated according to Ka = [HG]/([H][G])
(Table 2).
[16] A. Lledó, J. Rebek Jr., Chem. Commun. 2010, 46, 8630–8632.
[17] S. M. Biros, J. Rebek Jr., Chem. Soc. Rev. 2007, 36, 93–104.
[18] Y.-J. Ghang, M. P. Schramm, F. Zhang, R. A. Acey, C. N. Da-
vid, E. H. Wilson, Y. Wang, Q. Cheng, R. J. Hooley, J. Am.
Chem. Soc. 2013, 135, 7090–7093.
[19] L. Sebo, F. Diederich, V. Gramlich, Helv. Chim. Acta 2000, 83,
93–113.
[20] L. Pirondini, M. Melegari, R. Pinalli, E. Dalcanale, Supramol.
Chem. 2007, 19, 67–74.
[21] A. R. Mezo, J. C. Sherman, J. Org. Chem. 1998, 63, 6824–6829.
[22] T. Ogoshi, R. Shiga, T. Yamagishi, Y. Nakamoto, J. Org. Chem.
2011, 76, 618–622.
[23] P. Schaal, H. Baars, G. Raabe, I. Atodiresei, C. Bolm, Adv.
Synth. Catal. 2013, 355, 2506–2512.
[24] M. Cakici, S. Bräse, Eur. J. Org. Chem. 2012, 6132–6135.
Computational Simulations of Host–Guest Complexation: The com-
plexation of molecular basket 1 with 1,4-dioxane was simulated by
using the MAB force field in moloc,[36] verifying the formation of
temporary hydrogen bonds between guest and phenolic HO groups
3582
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 3575–3583

Resorcinarene Molecular Baskets and Container Molecules
[25]
[37] E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. Int.
Ed. 2003, 42, 1210–1250; Angew. Chem. 2003, 115, 1244–1287.
[38] I. K. Mati, S. L. Cockroft, Chem. Soc. Rev. 2010, 39, 4195–
4205.
[39] L. M. Salonen, M. Ellermann, F. Diederich, Angew. Chem. Int.
Ed. 2011, 50, 4808–4842; Angew. Chem. 2011, 123, 4908–4944.
[40] MacroModel, version 9.7, Schrödinger LLC, New York, 2009.
[41] Spartan ’14, version 1.0.0, Wavefunction Inc., Irvine, CA, 2013.
[42] R. Paulini, K. Müller, F. Diederich, Angew. Chem. Int. Ed.
2005, 44, 1788–1805; Angew. Chem. 2005, 117, 1820–1839.
[43] C. L. D. Gibb, B. C. Gibb, J. Am. Chem. Soc. 2004, 126, 11408–
11409.
L. Abis, E. Dalcanale, A. Du Vosel, S. Spera, J. Org. Chem.
1988, 53, 5475–5479.
A. G. S. Högberg, J. Org. Chem. 1980, 45, 4498–4500.
V. A. Azov, P. J. Skinner, Y. Yamakoshi, P. Seiler, V. Gramlich,
F. Diederich, Helv. Chim. Acta 2003, 86, 3648–3670.
T. F. Al-Azemi, M. Vinodh, Tetrahedron 2011, 67, 2585–2590.
O. Hass, A. Schierholt, M. Jordan, A. Lützen, Synthesis 2006,
519–527.
[26]
[27]
[28]
[29]
[30]
E. Dalcanale, P. Soncini, G. Bacchilega, F. Ugozzoli, J. Chem.
Soc., Chem. Commun. 1989, 500–502.
[31] P. J. Skinner, A. G. Cheetham, A. Beeby, V. Gramlich, F. Died-
erich, Helv. Chim. Acta 2001, 84, 2146–2153.
[32] P. P. Castro, G. Zhao, G. A. Masangkay, C. Hernandez, L. M.
Gutierrez-Tunstad, Org. Lett. 2004, 6, 333–336.
[44] I. Pochorovski, C. Boudon, J.-P. Gisselbrecht, M.-O. Ebert,
W. B. Schweizer, F. Diederich, Angew. Chem. Int. Ed. 2012, 51,
262–266; Angew. Chem. 2012, 124, 269–273.
[33] L. Fielding, Tetrahedron 2000, 56, 6151–6170.
[34] IGOR Pro, version 6.12, WaveMetrics Inc., Lake Oswego, OR,
2009.
[35] H.-J. Schneider, Angew. Chem. Int. Ed. 2009, 48, 3924–3977;
Angew. Chem. 2009, 121, 3982–4036.
[36] a) P. R. Gerber, K. Müller, J. Comput.-Aided Mol. Des. 1995,
9, 251–268; b) P. R. Gerber, J. Comput.-Aided Mol. Des. 1998,
12, 37–51; c) Gerber Molecular Design, http://www.moloc.ch.
[45] I. Pochorovski, M.-O. Ebert, J.-P. Gisselbrecht, C. Boudon,
W. B. Schweizer, F. Diederich, J. Am. Chem. Soc. 2012, 134,
14702–14705.
[46] W. H. Powell, Pure Appl. Chem. 1998, 70, 1513–1545.
[47] H. A. Favre, D. Hellwinkel, W. H. Powell, H. A. Smith Jr.,
S. S.-C. Tsay, Pure Appl. Chem. 2002, 74, 809–834.
Received: February 24, 2014
Published Online: April 17, 2014
Eur. J. Org. Chem. 2014, 3575–3583
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3583