Evaluation of hydrogen-bond acceptors for redox-switchable resorcin[4]arene cavitands

Article abstract of DOI:10.1021/ja411429b

Various H-bond acceptor groups were evaluated for their propensity to induce conformational switching between the kite and vase forms of diquinone-diquinoxaline resorcin[4]arene cavitands upon redox interconversion. The H-bond acceptors were placed on the quinoxaline walls with the purpose of stabilizing the vase form only in the reduced hydroquinone state of the cavitand by forming H-bonds with the hydroquinone OH groups. Design guidelines for successful acceptors were derived. The carboxamide acceptor was shown to be the best candidate. Based on this moiety, a redox-switchable triptycene-based basket that can completely sterically encapsulate a guest in its closed vase conformation was prepared. The basket binds small molecule guests with association constants of up to 10⁴ M^-1 in mesitylene-d₁₂ and exhibits slow guest exchange kinetics with a half-life for guest release in the order of 10⁴ s.

Full text of DOI:10.1021/ja411429b

Article
pubs.acs.org/JACS
Evaluation of Hydrogen-Bond Acceptors for Redox-Switchable
Resorcin[4]arene Cavitands
Igor Pochorovski, Jovana Milic, Dusan Kolarski, Cornelius Gropp, W. Bernd Schweizer,
́
̌
and Franco̧ is Diederich*
Laboratorium fur Organische Chemie, ETH Zurich, Honggerberg, HCI, 8093 Zurich, Switzerland
̈
̈
̈
̈
S
* Supporting Information
ABSTRACT: Various H-bond acceptor groups were eval-
uated for their propensity to induce conformational switching
between the kite and vase forms of diquinone-diquinoxaline
resorcin[4]arene cavitands upon redox interconversion. The
H-bond acceptors were placed on the quinoxaline walls with
the purpose of stabilizing the vase form only in the reduced
hydroquinone state of the cavitand by forming H-bonds with
the hydroquinone OH groups. Design guidelines for successful
acceptors were derived. The carboxamide acceptor was shown
to be the best candidate. Based on this moiety, a redox-switchable triptycene-based basket that can completely sterically
encapsulate a guest in its closed vase conformation was prepared. The basket binds small molecule guests with association
constants of up to 104 M−1 in mesitylene-d12 and exhibits slow guest exchange kinetics with a half-life for guest release in the
order of 104 s.
basket red-7 exhibited high association constants (Ka in the
range of 102−104 M−1) for suitably sized guests in mesitylene-
d12 and slow guest release kinetics with a half-life τ1/2 in the
order of 104 s.
INTRODUCTION
■
Every rationally designed molecular machine,1 whose behavior
is studied and discerned, expands the boundaries of our
understanding of the molecular world and further transforms
the field of supramolecular chemistry from science to
engineering. The resorcin[4]arene cavitand system is an
intriguing platform for designing molecular machines, because
it can adopt two spatially distinct forms: an expanded “kite” and
a contracted “vase.”2 Switching between the two forms can be
employed to grab and release small molecules, enabling the
application of cavitands as molecular grippers.3 Besides the
ability to switch the properties of cavitands4 using stimuli such
as changes in temperature,2,5 pH,6 metal ion concentration,7 or
light,8 our recently introduced concept for the design of redox-
switchable cavitands9 (Figure 1A) is particularly promising for
the application of cavitands in electronic devices, as redox-
switchable derivatives have the potential to be interfaced with
electroactive metal surfaces.10
RESULTS AND DISCUSSION
■
Synthesis. The syntheses of cavitands ox-1 and ox-3−7
commenced with the preparation of wall precursors 8−13
(Scheme 1A). Precursor 815 and 9a9b were accessed according
to literature procedures. Precursor 10 was prepared from
diboronic ester 14.16 In the first step, 14 and 2-bromopyridine
were reacted in a Suzuki cross-coupling to yield benzothiadia-
zole 15. In the next steps, reductive sulfur extrusion,17
condensation of the resulting diamine with oxalic acid,18 and
chlorination of the resulting diol intermediate18 yielded
precursor 10. Precursor 11 was prepared from compound 16,
employing the same three-step procedure described above.
Compound 16, in turn, was obtained in a one-pot sequence
from 1-octylimidazole (17)19 via stannylation to stannane 18
and Stille coupling with dibromide 19. Precursors 12 and 13
were accessed from compound 2020 and pyrazines 2121 and
22,21 respectively. Toward precursor 9b, dibromide 19 was
reacted in a Pd-catalyzed aminocarbonylation reaction22 with
nBu2NH and CO, affording compound 23. Reductive sulfur
extrusion from 23,17 condensation of the resulting diamine with
oxalic acid,18 and chlorination of the resulting diol with thionyl
chloride18 then yielded precursor 9b.
The switching concept is based on H-bonding interactions
that stabilize the vase form and are only present in the reduced
hydroquinone state of the cavitand. Assuming that the H-bond
strength influences the switching ability and binding proper-
ties,9b,11 we sought to investigate various H-bond acceptor
groups12 on the basis of the ester- (1), pyridine- (2−4),
imidazole- (5), and carboxamide-based (6)13 cavitands (Figure
1B). The results of this study are presented herein. We
established that the carboxamide moiety in cavitand 6 is best
suited for assisting redox-induced conformational switching.
Guided by this finding, we prepared the redox-switchable
basket system 7, which can completely encapsulate guests in its
vase conformation due to the triptycene walls.14 The reduced
Received: November 9, 2013
Published: February 25, 2014
© 2014 American Chemical Society
3852
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
H2O. When the reductions were finished, the CDCl3 layers
were transferred into N2-filled NMR tubes containing ∼5 mg of
anhydrous MgSO4 (in order to remove traces of H2O) and
analyzed by 1H NMR. The oxidized states could be regenerated
upon exposing the cavitand solutions to air for 2−4 d. The
reduced basket red-7 was obtained from ox-7 through
reduction with Pd/C under a H2 atmosphere in THF.
Triptycenequinone-based basket red-7 is more stable toward
oxidation by air oxygen compared to the naphthoquinone-
based cavitands red-1−2 and red-4−6 (in the solid state for at
least 2 months). Oxidation of red-7 to ox-7 could, however, be
achieved by exposing a chloroform solution of red-7 to air for
∼7 d.
X-ray Structures of Wall Precursor 12 and Cavitand
ox-4. Crystals of precursor 12 suitable for X-ray analysis were
obtained by evaporation from a CHCl3/EtOH solution. The
molecular structure of 12 in the crystal is shown in Figure 2
(left). The torsion angle of the pyridine plane relative to the
main molecular plane is 78°, which would position the pyridine
acceptor in an optimal orientation for H-bonding with the
hydroquinone OH groups if the unit was incorporated into a
cavitand system. Crystals of cavitand ox-4 were obtained by
evaporation from a MeCN/CH2Cl2 solution (Figure 2, right).
The cavitand crystallized in the kite conformation, which is
typical for diquinone-diquinoxaline cavitands.9a The pyridine N
atoms are oriented away from the platform in order to avoid
lone pair repulsion with the N atoms of the quinoxaline
moieties. Although the kite form should have C2v symmetry, it
does not in the crystal; the pyridine moieties that are in the
vicinity of the quinone units adopt torsion angles of 154−158°
relative to the quinoxaline walls, while the ones that are further
away adopt torsion angles of 130−138°.
Figure 1. (A) Design concept of redox-switchable cavitands (X = H-
bond acceptor; the symbols and ★ represent methine protons). (B)
Target cavitands in this study.
●
Subsequently, we proceeded with the assemblies of cavitands
ox-1−7 (Scheme 1B). Cavitands ox-1 and ox-4−6 were
prepared through condensations of tetrol 249b,23 with
precursors 8−11, respectively. The synthesis of cavitand ox-2
required a different strategy, because the respective wall
precursor analogous to 8−13 is not easily accessible. Hence,
we planned to prepare cavitand ox-2 via nucleophilic
substitution reaction between the difluoropyrazine-substituted
cavitand 25, which was prepared from tetrol 24 and pyrazine
22, and pyridine-2-thiol (26).24 This reaction did not, however,
yield cavitand ox-2 directly. Instead, 26 caused the excision of
quinone walls to yield intermediate 27 (as revealed by MALDI-
MS). This observation means that the quinone walls of 25 are
more reactive toward nucleophilic attack than C−F bonds on
the pyrazine rings. Nevertheless, the desired cavitand ox-2 was
obtained by transforming intermediate 27 into 28 with an
excess of 26, and reattaching the quinone walls with compound
29. The synthesis of cavitand ox-3 was attempted via two
different routes: (i) reaction of compound 20 and cavitand 25
and (ii) reaction of tetrol 24 with dichloropyrazine 12 or
difluoropyrazine 13. However, neither of the routes was
successful. The outcome of route i was quinone excision
instead of substitution of the fluorine atoms, while in route ii
neither 12 nor 13 was reactive with tetrol 24. Presumably, the
electron-donating dihydropyrazine-N-atoms strongly reduce
the reactivity of the pyrazine C−Cl bonds (in 12) and the
C−F bonds (in 13) in nucleophilic substitution reactions.
Basket ox-7 was obtained by condensing precursor 9b with
tetrol 30.9b
Conformational Properties of Cavitands ox-1−2 and
1
ox-4−6, and red-1−2 and red-4−6 Studied by H NMR
Spectroscopy. The ability of the various H-bond acceptor
groups in cavitands ox-1−2 and ox-4−6 to assist conforma-
tional switching upon redox interconversion was investigated by
1
1H NMR spectroscopy. Figure 3 (left) shows the H NMR
spectra of the oxidized cavitand states in CDCl3. The adopted
conformations, kite or vase, are revealed by the characteristic
resonances of the methine protons (● and ★, denoted in
Figure 1A).2 Details about the assignment of the resonances (●
or ★) to the corresponding methine protons (the ones below
the quinone or quinoxaline walls) are presented in the
Supporting Information, section 3.3. For all cavitands, the
methine protons are located between 3.6 and 4.4 ppm,
indicating the presence of the kite conformations in the
oxidized cavitand states.2 This finding is in line with the fact
that (unsubstituted) diquinone-diquinoxaline cavitands favor
the kite over the vase form in chlorinated solvents.9a
1H NMR spectra of the reduced cavitands red-1−2 and red-
4−6, on the other hand, reveal pronounced differences (Figure
3, right). The ester-substituted red-1 is present in the kite form
(methine protons at 3.4 and 4.5 ppm), which means that
conformational switching upon redox interconversion does not
take place in that system. On the other hand, the methine
protons of the S-pyridine-substituted cavitand red-2 resonate at
5.4 and 5.6 ppm. Although these signals are broad, indicating a
dynamic structure, their position testifies that red-2 adopts the
vase conformation. Surprisingly, the 1H NMR spectra of
pyridine-substituted cavitand red-4 and imidazole-substituted
cavitand red-5 displayed very broad peaks, over a wider
temperature range, allowing no direct conclusion about the
With the oxidized cavitands in hand, we continued with
accessing their reduced redox states (Scheme 1C). Cavitands
red-1−2 and red-4−6 were obtained through reductions of ox-
1−2 and ox-4−6 with Na2S2O4 in biphasic mixtures of CDCl3/
3853
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
a
Scheme 1. Synthetic Routes Towards Cavitands ox-1−7 and red-1−7
a
(a) 2-Bromopyridine, K2CO3, [Pd(PPh3)4], toluene/H2O, 80 °C, 48 h; 71%. (b) (1) NaBH4, CoCl2·6 H2O, EtOH, reflux, 30 min; (2) (COOH)2,
4 M HCl, 100 °C, 15 h; (3) POCl3, DMF, 100 °C, 15 h; 69%. (c) (1) nBuLi, THF, −78 °C, 1 h; (2) nBu3SnCl, THF, 0→25 °C, 12 h. (d)
[Pd(PPh3)4], THF, reflux, 72 h; 21%. (e) (1) NaBH4, CoCl2·6 H2O, EtOH, reflux, 3 h; (2) (COOH)2, 4 M HCl, 100 °C, 12 h; (3) SOCl2, DMF/
(CHCl2)2, 100 °C, 4 h; 10%. (f) Cs2CO3, DMF, 120 °C, 16 h; 59%. (g) Cs2CO3, DMF, 90 °C, 5 h; 57%. (h) nOct2NH, CO (balloon), [Pd(OAc)2],
Xantphos, K3PO4, toluene, reflux, 3 d; 78%. (i) (1) NaBH4, CoCl2·6 H2O, EtOH, reflux, 6 h; (2) (COOH)2, EtOH/HCl, reflux, 4 d; (3) SOCl2,
DMF, (CH2Cl)2, 100 °C, 16 h; 40%. (j) Cs2CO3, THF, 50 °C, 3 h; 68%. (k) Cs2CO3, THF, reflux, 12 h; 28%. (l) Cs2CO3, DMF, 25 °C, 4 h; 27%.
(m) Cs2CO3, THF, reflux, 24 h; 85%. (n) Cs2CO3, DMF, 70 °C, 12 h. (o) Cs2CO3, DMF, 50 °C, 12 h. (p) K2CO3, DMF, 0→25 °C, 16 h; 50%. (q)
(1) 26, Cs2CO3, DMF, 50 °C, 9 h; (2) 29, Cs2CO3, DMF, 50 °C, 13 h; 9%. (r) Cs2CO3, THF, reflux, 16 h; 29%. (s) Na2S2O4, CDCl3/H2O, 60 °C, 3
h, quantitative. (t) Air, CDCl3, quantitative. (u) H2, Pd/C, THF, 25 °C, 3 h, quantitative. (v) Air, CDCl3, quantitative.
adopted conformations of these cavitands. Decompositions
were ruled out, as cavitands ox-4 and ox-5 fully regenerated
upon exposing the solutions of red-4 and red-5 to air for 1−3 d.
1
In contrast, the H NMR spectrum of red-6 displays sharp
3854
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
Figure 2. Molecular structures of wall precursor 12 and cavitand ox-4
in their respective crystals at 100 K. Crystals of 12 were obtained from
CHCl3/EtOH and those of ox-4 from CH2Cl2/MeCN. In case of ox-4,
solvent molecules, n-hexyl chains, and H-atoms are omitted for clarity.
Figure 4. Structures of cavitands red-1−2 and red-4−6 optimized at
the DFT B3LYP/6-31G(d) level of theory. Alkyl chains were
exchanged by methyl groups.
hydroquinone OH groups per cavitand undergo H-bonding
with the weak12 COOMe H-bond acceptors, while the other
two establish H-bonds with the ether oxygen atoms of the
cavitand backbone. On the other hand, in cavitand red-2 the
formation of all four H-bonds between the pyridine and
hydroquinone OH groups is possible. These observations are in
line with 1H NMR results, according to which red-1 adopts the
kite and red-2 the vase form (Figure 3, right). Apparently, the
formation of only two H-bonds is not sufficient to stabilize the
vase form but formation of all four is required. While all four H-
bonds can, in principle, be established in case of cavitands red-
4−6 (Figure 4), only red-6 clearly adopts a vase conformation
1
Figure 3. Sections of the 1H NMR spectra (298 K) of cavitands ox-1−
2 and ox-4−6, and red-1−2 and red-4−6 in CDCl3 (300 MHz). The
according to H NMR spectroscopy (Figure 3, right).
To understand why red-6 adopts the vase form while red-4
and red-5 do not, we performed dihedral angle (θ) scans
between the pyridine, imidazole, and carboxamide moieties,
respectively, and the quinoxaline moieties of hypothetical test
systems shown in Figure 5. The calculations were performed at
the MP2/cc-pVTZ//B3LYP/6-31+G(d,p)27 level of theory.
The global energy minima are located at dihedral angles of 140°
for the pyridine-wall model (which is consistent with the
dihedral angle range of 130−158° observed in the crystal
structure of ox-4, Figure 2), 130° for the imidazole-wall model,
and 120° for the carboxamide-wall model. However, formation
of intramolecular H-bonds in the hydroquinone cavitands is
structurally only possible for dihedral angles θ in the range of
60−100°. Turning the moieties into this desired θ range
requires at least 1.9 kcal mol−1 for the pyridine, 1.3 kcal mol−1
for the imidazole, and 0.3 kcal mol−1 for the carboxamide
moiety. Thus, in case of the pyridine and imidazole moieties,
the energetic cost for this process is presumably larger than that
of remaining in the optimal orientations and establishing
●
symbols
Figure 1A.
and ★ represent methine protons; for assignment, see
triplets for the methine protons located at 5.7 and 6.2 ppm,
indicating the presence of a rigid vase conformation.9b
Baskets ox-7 and red-7 exhibit the same conformational
properties as cavitands ox-6 and red-6; ox-7 is present in the
kite conformation in CDCl3 (methine protons between 3.8 and
4.2 ppm), while red-7 adopts a rigid vase form (methine
protons at 5.6 and 5.7 ppm as sharp triplets).25
Computational Studies. In order to rationalize the
conformational properties of cavitands red-1−2 and red-4−6
discussed above, we computed their optimized structures with
DFT at the B3LYP/6-31G(d) level of theory (Figure 4, alkyl
chains were exchanged by methyl groups).26 In the ester-based
cavitand red-1, only one COOMe group per bis(COOMe) unit
can participate in H-bonding with the hydroquinone OH
groups due to steric reasons. Consequently, only two
3855
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
Table 1. Association Constants Ka (M−1) between Baskets
ox-7 and red-7, and Various Guest Molecules (mesitylene-
a
d12, 298 K)
a
Determined by integration of the 1H NMR resonances of the relevant
species, relative to 1,3,5-trimethoxybenzene as internal standard.
Errors in Ka are estimated to be 20%. Ka values were corrected to
account for the presence of 0.4 equiv THF remaining after reduction
in the cavitand sample.
Figure 5. Dihedral angle scans for hypothetical test systems containing
the pyridine, imidazole, and carboxamide moieties, performed on the
MP2/cc-pVTZ//B3LYP/6-31+G(d,p) level of theory. The highlighted
gray area is the dihedral angle range, in which H-bonding to the
hydroquinone OH groups would be possible in the cavitand systems.
equilibrium host−guest concentration directly after guest
addition to red-7 and spectrum recording (half-lives of guest
uptake τ1/2 < 5 min), the larger cyclooctane exhibited binding
kinetics slow enough to be followed by NMR. The
concentration vs time curve of the host−guest complex formed
between red-7 (∼2 mM) and cyclooctane (∼0.25 equiv) is
illustrated in Figure 6 (left). The process followed first-order
intermolecular H-bonds with OH groups of other cavitand
molecules. On the other hand, the low barrier in the case of the
carboxamide moiety is smaller than the entropic costs of
intermolecular aggregations. This rationale is in line with the
1
complex H NMR spectra of red-4 and red-5 (Figure 3, right)
that could be explained by aggregate formation based on
intermolecular H-bonds,28 and the 1H NMR spectrum of red-6
that reveals a rigid vase conformation stabilized by intra-
molecular H-bonds.9b
Binding Properties of Baskets ox-7 and red-7. In our
earlier communication9b we showed that the top-open H-bond-
stabilized cavitand red-6 binds alicyclic guests with association
constants Ka in the range of 101−103 M−1, and guest release
rates with a half-life τ1/2 in the order of 10−2 s. With the basket
system ox-/red-7 we sought to investigate whether the sterically
congesting triptycene moiety can further enhance the
association constants and slow down guest release.
Figure 6. (Left) Binding kinetics of cyclooctane by red-7. (Right)
Release kinetics of cyclohexane from the host−guest complex with
red-7, induced by addition of excess 1,4-cyclohexanedione. Both data
sets were treated by a first-order kinetics law with the rate constants as
fit parameters.
The binding studies were performed in mesitylene-d12, as this
solvent is too large to compete with guests for the cavitand
binding site. In mesitylene-d12, ox-7 is present in the kite and
red-7 in the vase conformation. Thus, while red-7 was already
preorganized for binding, ox-7 switched from kite to vase upon
guest uptake (see Figures S6 and S7 in the Supporting
Information). One aspect complicating binding studies with
red-7 was that THF, which was used as solvent during the
reduction of ox-7 to red-7, remained tightly bound by the
cavitand.29 Difficulties to remove residual solvent from
cavitands with strong binding properties have already been
reported.6b,c,30 Therefore, binding studies with red-7 were
performed with samples containing 0.4 equiv of THF, whereby
the association constants were corrected for the presence of
THF as a competing guest.31 The association constants
between cavitands ox-7 and red-7 and various guests are
shown in Table 1. The Ka values of red-7 are significantly
higher than those of ox-7, by a factor of 102−103. Within the
cycloalkane series, red-7 shows the highest association constant
for cyclooctane. Thus, the cavity of red-7 is larger than that of
other reported cavitand-based basket systems, which show
maxima in Ka for cyclopentane or cyclohexane.6b,c,30 The largest
Ka value was measured for 1,4-cyclohexanedione, presumably
due to an optimal size complementarity.9b
kinetics, suggesting that the rate-determining step is governed
by cavitand opening to an extent at which cyclooctane can slip
in. The kinetic and thermodynamic parameters for cyclooctane
uptake were determined to be: kin = (5.4 0.3)10−4 s−1; τ1/2
=
21 1 min; ΔG⧧
= 21.9 kcal mol−1. The release kinetics
298 K
was studied on the example of cyclohexane by monitoring the
guest exchange with 1,4-cyclohexanedione − the guest with the
highest measured association constant. To a sample containing
red-7 (∼1 mM) and cyclohexane (∼1 equiv), an excess (30
equiv) of 1,4-cyclohexanedione was added. The concentration
vs time curve of the host−guest complex between cyclohexane
and red-7 is illustrated in Figure 6 (right). The exchange
process followed first-order kinetics, indicating that the rate-
determining step is unimolecular. As the uptake of 1,4-
cyclohexanedione was shown to be fast in separate binding
experiments (τ1/2 < 5 min), the rate-determining step for the
exchange process must be cyclohexane release. The kinetic and
thermodynamic parameters for cyclohexane release were
determined to be: kout = (5.3 0.2)10−5 s−1; τ1/2 = 216
7
min; ΔG⧧
= 23.3 kcal mol−1.
298 K
Binding Kinetics of Basket red-7. We investigated the
kinetics of guest binding and guest release of the triptycene-
derived basket system red-7. While smaller guests reached an
Guest-exchange mechanisms have been studied for a variety
of artificial host systems.32 The most likely mechanism for guest
exchange in cavitands presumably involves two steps:
3856
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
dissociation of the leaving guest, followed by association with
the incoming guest. The intermediate is presumably not empty
but filled with solvent molecules. The transition state is
represented by a cavitand structure that is open just to the
extent that allows guest release/uptake. The rate-determining
step therefore depends on how much the cavitand needs to
distort for guest release/uptake to take place, which, in turn,
strongly depends on the size of the leaving/incoming guest and
the H-bond stabilization energy. Thus, the slow binding
kinetics of basket red-7 is a consequence of both the congesting
influence of the triptycene moiety, and the strong hydrogen
bonds between the amide and hydroquinone units.
spectra. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
diederich@org.chem.ethz.ch
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by a grant from the Swiss National
Science Foundation (SNF). I.P. acknowledges the receipt of a
fellowship from the Fonds der Chemischen Industrie. J.M. and
D.K. were funded by the Fond za Mlade Talente Republike
Srbije. C.G. acknowledges the receipt of a fellowship from the
Studienstiftung des deutschen Volkes.
SUMMARY AND CONCLUSIONS
■
The goal of this work was to assess the propensity of various H-
bond acceptors in assisting conformational switching between
the kite and vase forms of diquinone cavitands. We established
that all four H-bond acceptors need to be involved in H-
bonding with the hydroquinone OH groups in order for the
vase form to become energetically favorable. In addition, the
acceptor moiety should not have too many conformational
degrees of freedom; otherwise the vase form becomes too
flexible. Another key design element is to ensure that the
optimal dihedral angle of the H-bond acceptor relative to the
quinoxaline wall lies within or close to the dihedral angle range
of 60−100° in which H-bond formation is sterically possible, or
that rotation into this range is accompanied only by a small
energetic penalty. A strong H-bond acceptor that is not well
preorganized for intramolecular H-bonding will likely rather
engage in intermolecular associations. We found that the
carboxamide moiety in cavitands red-6 and red-7 is an acceptor
that excellently satisfies these design guidelines.
REFERENCES
■
(1) (a) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Angew.
Chem., Int. Ed. 2000, 39, 3348. (b) Moonen, N.; Flood, A. H.;
́
Fernandez, J. M.; Stoddart, J. F. Top. Curr. Chem. 2005, 262, 99.
(c) Browne, W. R.; Feringa, B. L. Nature Nanotechnol. 2006, 1, 25.
(d) Kay, E. R.; Leigh, D. A.; Zerbetto, F. Angew. Chem., Int. Ed. 2007,
46, 72. (e) Coskun, A.; Banaszak, M.; Astumian, R. D.; Stoddart, J. F.;
Grzybowski, B. A. Chem. Soc. Rev. 2011, 41, 19.
(2) (a) Moran, J. R.; Karbach, S.; Cram, D. J. J. Am. Chem. Soc. 1982,
104, 5826. (b) Moran, J. R.; Ericson, J. L.; Dalcanale, E.; Bryant, J. A.;
Knobler, C. B.; Cram, D. J. J. Am. Chem. Soc. 1991, 113, 5707.
(3) Azov, V. A.; Beeby, A.; Cacciarini, M.; Cheetham, A. G.;
Diederich, F.; Frei, M.; Gimzewski, J. K.; Gramlich, V.; Hecht, B.; Jaun,
B.; Latychevskaia, T.; Lieb, A.; Lill, Y.; Marotti, F.; Schlegel, A.;
Schlittler, R. R.; Skinner, P. J.; Seiler, P.; Yamakoshi, Y. Adv. Funct.
Mater. 2006, 16, 147.
Closing the top of red-6 with the triptycene moiety resulted
in a redox-switchable basket 7 with enhanced binding
properties in the reduced state compared to the corresponding
top-open system.9b Association constants in the oxidized state
ox-7 were found to be in the range of 10−1−101 M−1, while
association constants in the range of 102−104 M−1 were
measured for the reduced cavitand red-7. Thereby, changing
the redox state of basket 7 modulates the association constants
by a factor of 102−103. This switchability of binding properties
might eventually allow cavitands of this type to be employed as
molecular grippers for nanorobotics. Closing the top of the
cavity has furthermore a dramatic impact on guest uptake and
release kinetics: the rate constant for cyclooctane uptake by
red-7 was kin = (5.4 0.3)10−4 s−1, while that for cyclohexane
release was kout = (5.3 0.2) × 10−5 s−1. This rate constant for
guest release is significantly lower than those reported for other
cavitand systems;6b,c,9b,11 so far, the record value is kout = 2.5 ×
10−3 s−1 for the release of a neutral guest (cyclohexane) from a
covalently top-bridged cavitand.6c Only the covalently bridged
carcerands and hemicarcerands can achieve even lower
values.32e,33
(4) Stimuli-responsive guest release has also been demonstrated in
metallosupramolecular systems, see: (a) Mal, P.; Schultz, D.; Beyeh,
K.; Rissanen, K.; Nitschke, J. R. Angew. Chem., Int. Ed. 2008, 47, 8297.
(b) Han, M.; Michel, R.; He, B.; Chen, Y.-S.; Stalke, D.; John, M.;
Clever, G. H. Angew. Chem., Int. Ed. 2012, 52, 1319. (c) Lewis, J. E.
M.; Gavey, E. L.; Cameron, S. A.; Crowley, J. D. Chem. Sci. 2012, 3,
778.
(5) (a) Azov, V. A.; Jaun, B.; Diederich, F. Helv. Chim. Acta 2004, 87,
449. (b) Roncucci, P.; Pirondini, L.; Paderni, G.; Massera, C.;
Dalcanale, E.; Azov, V. A.; Diederich, F. Chem.Eur. J. 2006, 12,
4775.
(6) (a) Skinner, P. J.; Cheetham, A. G.; Beeby, A.; Gramlich, V.;
Diederich, F. Helv. Chim. Acta 2001, 84, 2146. (b) Gottschalk, T.;
Jarowski, P. D.; Diederich, F. Tetrahedron 2008, 64, 8307.
(c) Gottschalk, T.; Jaun, B.; Diederich, F. Angew. Chem., Int. Ed.
2007, 46, 260.
(7) (a) Frei, M.; Marotti, F.; Diederich, F. Chem. Commun. 2004,
1362. (b) Durola, F.; Rebek, J., Jr. Angew. Chem., Int. Ed. 2010, 49,
3189.
(8) Berryman, O. B.; Sather, A. C.; Rebek, J., Jr. Chem. Commun.
2010, 47, 656.
(9) (a) Pochorovski, I.; Boudon, C.; Gisselbrecht, J.-P.; Ebert, M.-O.;
Schweizer, W. B.; Diederich, F. Angew. Chem., Int. Ed. 2012, 51, 262.
(b) Pochorovski, I.; Ebert, M.-O.; Gisselbrecht, J.-P.; Boudon, C.;
Schweizer, W. B.; Diederich, F. J. Am. Chem. Soc. 2012, 134, 14702.
(10) (a) Schierbaum, K. D.; Weiss, T.; Vanvelzen, E. U. T.;
Engbersen, J. F. J.; Reinhoudt, D. N.; Gopel, W. Science 1994, 265,
1413. (b) Vanvelzen, E. U. T.; Engbersen, J. F. J.; Reinhoudt, D. N. J.
Am. Chem. Soc. 1994, 116, 3597. (c) Huisman, B. H.; Rudkevich, D.
M.; van Veggel, F. C. J. M.; Reinhoudt, D. N. J. Am. Chem. Soc. 1996,
118, 3523. (d) Schonherr, H.; Beulen, M. W. J.; Bugler, J.; Huskens, J.;
van Veggel, F. C. J. M.; Reinhoudt, D. N.; Vancso, G. J. J. Am. Chem.
Soc. 2000, 122, 4963. (e) Yamakoshi, Y.; Schlittler, R. R.; Gimzewski, J.
K.; Diederich, F. J. Mater. Chem. 2001, 11, 2895. (f) Tsoi, S.; Griva, I.;
This work is an example of how a molecular system with a
desired function can be evolved, with the assistance of
molecular design, synthesis, 1H NMR spectroscopy, X-ray
crystallography, and computational studies, from a general
concept.
ASSOCIATED CONTENT
* Supporting Information
Synthetic procedures, characterization data, details on binding
studies, details on computational studies, X-ray data, NMR
■
S
3857
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858

Journal of the American Chemical Society
Article
Trammell, S. A.; Blum, A. S.; Schnur, J. M.; Lebedev, N. Acs Nano
2008, 2, 1289. (g) Kim, H. J.; Lee, M. H.; Mutihac, L.; Vicens, J.; Kim,
J. S. Chem. Soc. Rev. 2012, 41, 1173.
(11) (a) Rudkevich, D. M.; Hilmersson, G.; Rebek, J., Jr. J. Am. Chem.
Soc. 1998, 120, 12216. (b) Rudkevich, D. M.; Hilmersson, G.; Rebek,
J., Jr. J. Am. Chem. Soc. 1997, 119, 9911.
(12) Laurence, C.; Graton, J.; Berthelot, M.; Besseau, F.; Le Questel,
J.-Y.; Luco̧ n, M.; Ouvrard, C.; Planchat, A.; Renault, E. J. Org. Chem.
2010, 75, 4105.
(13) For an earlier communication on cavitand 6, see reference 9b.
(14) An analogous basket equipped with NBu2 instead of NOct2
groups (reported in reference 9b) exhibited poor solubility in the
reduced cavitand state. The NOct2 groups significantly improve the
solubility, allowing binding studies to be performed with ox-7 and red-
7.
(15) Tucci, F. C.; Rudkevich, D. M.; Rebek, J., Jr. Chem.Eur. J.
2000, 6, 1007.
(16) Zhang, M.; Tsao, H. N.; Pisula, W.; Yang, C.; Mishra, A. K.;
Mullen, K. J. Am. Chem. Soc. 2007, 129, 3472.
̈
(17) Neto, B. A. D.; Lopes, A. S.; Wust, M.; Costa, V. E. U.; Ebeling,
̈
G.; Dupont, J. Tetrahedron Lett. 2005, 46, 6843.
(18) Romer, D. R. J. Heterocycl. Chem. 2009, 46, 317.
(19) Serpell, C. J.; Kilah, N. L.; Costa, P. J.; Fel
Angew. Chem., Int. Ed. 2010, 49, 5322.
́
ix, V.; Beer, P. D.
(20) Gdaniec, M.; Bensemann, I.; Polonski, T. Acta Crystallogr., Sect.
C 2004, 60, o215.
(21) Allison, C. G.; Chambers, R. D.; MacBride, J. A. H.; Musgrave,
W. K. R. J. Chem. Soc. C 1970, 1023.
(22) Martinelli, J. R.; Watson, D. A.; Freckmann, D. M. M.; Barder,
T. E.; Buchwald, S. L. J. Org. Chem. 2008, 73, 7102.
(23) (a) Castro, P. P.; Zhao, G.; Masangkay, G. A.; Hernandez, C.;
Gutierrez-Tunstad, L. M. Org. Lett. 2004, 6, 333. (b) Azov, V. A.;
Diederich, F.; Lill, Y.; Hecht, B. Helv. Chim. Acta 2003, 86, 2149.
(24) This strategy was inspired by a report from Cram and co-
workers, in which the fluorine atoms of a difluoropyrazine-substituted
velcrand were substituted by N-, O, and S-nucleophiles: Cram, D. J.;
Choi, H. J.; Bryant, J. A.; Knobler, C. B. J. Am. Chem. Soc. 1992, 114,
7748.
1
(25) The H NMR spectra of ox-7 and red-7 are shown in Figures
S14 and S15 in the Supporting Information.
(26) For detailed information on computational studies, see section 3
of the Supporting Information.
̌
́ ̌
(27) (a) Riley, K. E.; Platts, J. A.; Rezac, J.; Hobza, P.; Hill, J. G. J.
Phys. Chem. A 2012, 116, 4159. (b) Jackson, N. E.; Savoie, B. M.;
Kohlstedt, K. L.; Olvera de la Cruz, M.; Schatz, G. C.; Chen, L. X.;
Ratner, M. A. J. Am. Chem. Soc. 2013, 135, 10475.
(28) Further evidence for intermolecular associations is provided by
the very broad IR absorption bands corresponding the OH groups in
the solution state (CDCl3) IR spectra of compounds red-4 and red-5.
(29) For example, drying the cavitand at high vacuum for 30 d was
1
not sufficient to remove residual THF: H NMR revealed that 1.1
equiv of THF was still present in the cavitand sample. Even after
coevaporation with mesitylene-d12 and subsequent drying at high
vacuum, 0.4 equiv of THF still remained in the sample.
(30) Hornung, J.; Fankhauser, D.; Shirtcliff, L. D.; Praetorius, A.;
Schweizer, W. B.; Diederich, F. Chem.Eur. J. 2011, 17, 12362.
(31) For more details, see section 4 of the Supporting Information.
(32) (a) Santamaria, J.; Martin, T.; Hilmersson, G.; Craig, S. L.;
Rebek, J., Jr. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 8344. (b) Craig, S.
L.; Lin, S.; Chen, J.; Rebek, J., Jr. J. Am. Chem. Soc. 2002, 124, 8780.
(c) Davis, A. V.; Raymond, K. N. J. Am. Chem. Soc. 2005, 127, 7912.
(d) Pluth, M. D.; Raymond, K. N. Chem. Soc. Rev. 2007, 36, 161.
(e) Rieth, S.; Hermann, K.; Wang, B.-Y.; Badjic, J. D. Chem. Soc. Rev.
2011, 40, 1609.
(33) (a) Warmuth, R.; Yoon, J. Acc. Chem. Res. 2001, 34, 95.
(b) Quan, M. L.; Cram, D. J. J. Am. Chem. Soc. 1991, 113, 2754.
3858
dx.doi.org/10.1021/ja411429b | J. Am. Chem. Soc. 2014, 136, 3852−3858