Hexameric resorcinarene capsule is a bronsted acid: Investigation and application to synthesis and catalysis

Article abstract of DOI:10.1021/ja4080375

Molecular capsules have attracted interest as simple enzyme mimetics and several examples of catalytic transformations in water-soluble metal-ligand based systems have been reported. This is not the case for hydrogen-bond based molecular capsules, which in contrast can be employed in organic solvents. We describe herein our investigations of such a system: The resorcin[4]arene hexamer is one of the largest hydrogen bond-based self-assembled capsules and has been studied intensively due to its ready availability. We present evidence that the capsule acts as a reasonably strong Bronsted acid (pK _a approximately 5.5-6). This finding explains the capsule's high affinity toward tertiary amines that are protonated and therefore encounter cation-π interactions inside the cavity. We were able to translate this finding into a first synthetic application: A highly substrate-selective Wittig reaction. We also report that this property renders the capsule an efficient enzyme-like catalyst for substrate selective diethyl acetal hydrolysis.

Full text of DOI:10.1021/ja4080375

Article
pubs.acs.org/JACS
Hexameric Resorcinarene Capsule is a Brønsted Acid: Investigation
and Application to Synthesis and Catalysis
Qi Zhang and Konrad Tiefenbacher*
Department Chemie, Technische Universitat Munchen, D-85747 Garching, Germany
̈
̈
S
* Supporting Information
ABSTRACT: Molecular capsules have attracted interest as
simple enzyme mimetics and several examples of catalytic
transformations in water-soluble metal−ligand based systems
have been reported. This is not the case for hydrogen-bond
based molecular capsules, which in contrast can be employed
in organic solvents. We describe herein our investigations of
such a system: The resorcin[4]arene hexamer is one of the
largest hydrogen bond-based self-assembled capsules and has been studied intensively due to its ready availability. We present
evidence that the capsule acts as a reasonably strong Brønsted acid (pKa approximately 5.5−6). This finding explains the capsule’s
high affinity toward tertiary amines that are protonated and therefore encounter cation-π interactions inside the cavity. We were
able to translate this finding into a first synthetic application: A highly substrate-selective Wittig reaction. We also report that this
property renders the capsule an efficient enzyme-like catalyst for substrate selective diethyl acetal hydrolysis.
reactions catalyzed inside such molecular flasks.5 In most cases,
INTRODUCTION
■
product inhibition prevents a catalytic turnover.3k,m
Nature’s capability to enzymatically catalyze reactions under
mild conditionsambient temperature or slightly higher
“body” temperature and necessarily without any precautions
against water and oxygenhas been fascinating synthetic
organic chemists for decades. Enzymes selectively isolate
suitable substrates inside a hydrophobic reaction pocket, adjust
them into the reactive orientation and/or conformation, alter or
enhance their reactivity by noncovalent or covalent interactions,
stabilize the transition state of the reaction, and finally expel the
product to complete the catalytic cycle.1 Additionally,
contributions to catalytic efficiency from quantum mechanical
tunneling, matching of pKa values in the transition state and
protein dynamics have been identified.1g
Naturally, numerous attempts have been made to mimic such
biological catalysts: Early examples of discrete entities included
functionalized cyclodextrins, cyclic porphyrin oligomers,
spherands, and cyclophanes, which were investigated among
others by the groups of Breslow, Sanders, Cram, and Diederich,
respectively.1a,b,2 Beside such covalently linked host structures,
noncovalently self-assembled structures also have been
developed. These molecular capsules spontaneously form in
solution via hydrogen bonds, metal−ligand interactions, or the
hydrophobic effect and were mainly investigated by the groups
of Rebek, Raymond, and Fujita, and Gibb, respectively.3 Since
such host structures self-assemble from smaller components,
their preparation usually requires less synthetic effort and,
additionally, they completely surround the encapsulated guests.
Therefore, attention has shifted toward these systems as
enzyme mimetics. The investigation of metal−ligand based
molecular capsules has yielded numerous examples of catalytic
transformation.4 Interestingly, this is not the case with
hydrogen-bonded capsules: There are only two examples of
We wanted to explore the possibility of utilizing hydrogen-
bonded molecular capsules as catalysts, not only because
examples are scarce but mainly due to the fact that such systems
are − in contrast to most metal−ligand capsules − soluble in
organic media. This should allow the use of reagents which are
not compatible with excess water. Since dozens of structurally
different hydrogen-bonded molecular capsules have been
described in the literature,3 we screened for the following two
properties: (1) ease of preparation: wider applications seem
conceivable only if the capsule can be readily synthesized in
large quantities; (2) large cavity size: to allow for experiments
with a variety of different guest sizes and to prevent excessive
binding of substrates (via strong contacts to multiple capsule
walls), which usually translates into strong binding of the
products (product inhibition). We identified the hexameric
resorcin[4]arene capsule I (Figure 1), which was reported by
the group of Atwood6 in 1997, as an ideal candidate. It
spontaneously forms in apolar solvents from six resorcin[4]-
arene units 1, which are readily prepared in multigram
quantities in a single step. And with an internal volume of
3
́
approximately 1400 Å, it also represents one of the largest
hydrogen-bonded molecular capsules.
The structure of I was elucidated by single crystal X-ray
analysis.6 In addition to the six resorcin[4]arene units, eight
water molecules participate in its formation and indeed also
proved essential to its formation in solution.7 The encapsula-
tion of different guests8 and the formation of the capsule itself
have been investigated extensively in solution.9 Also, the
Received: August 9, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
field due to the increased electron-density of the anionic
species, while the signals of the remaining part would be
unaffected. To investigate this phenomenon more closely, a
capsule solution was titrated with incremental amounts (0.2
equiv) of triethylamine (3) (Figure 2). Several changes can be
Figure 1. (a) Structure of resorcin[4]arene 1; (b) schematic
representation of the hexameric resorcin[4]arene capsule I, emphasiz-
ing the octahedral cavity space (blue); alkyl groups have been omitted
for clarity; (c) simplified symbolic representation of the hexameric
capsule; and (d) guest molecules first investigated for encapsulation in
I.
stability toward polar additives10 and the assembling properties
in the presence of alcohols has been studied.11 Besides NMR
spectroscopy, EPR spectroscopy12 and mass spectrometry13
have been used to analyze molecular capsule I. A recent review
also discusses the field of hexameric capsules.14
Figure 2. 1H NMR titration of capsule I (a) in water-saturated CDCl3
(3.3 mM) with various amounts of NEt3 (b−h). Three different
aromatic peaks at 7.20 ppm, indicating the encapsulation of different
guests, are observable and are highlighted by colored arrows. The
asterisk marks the shifted phenolic signals after deprotonation of the
capsule.
RESULTS AND DISCUSSION
■
Acidity of the Hexamer. Our investigations into this field
started with the reproduction of an intriguing observation8e,f,9c
described in the literature: Capsule I binds well both
tetraalkylammonium salts and trialkylamines. While the strong
interactions of alkylammonium compounds can be explained by
cation-π stabilization inside the aromatic cavity, the efficient
binding of amines was surprising to us. We reinvestigated this
issue and quantified the encapsulation by separately adding 0.5
equiv of quaternary ammonium salt 2 (Figure 1) and
triethylamine (3) to a solution of I in water-saturated CDCl3
(3.3 mM in capsule I). Indeed, in both cases, the respective
guests were completely encapsulated, as evidenced by
integration of the upfield-shifted guest signals in the 1H
NMR spectrum. This seemed even more interesting, as the
carbon analog of triethylamine (3), 3-ethylpentane (4),
completely resisted encapsulation under such conditions,
indicating that the amine functionality is responsible for the
observed difference. We speculated that the tertiary amine is
encapsulated so well only because it was protonated somehow.
Initial evidence in this direction and toward identifying the
origin of the proton was found in the 1H NMR spectrum of the
experiment with triethylamine (3). Integration of the phenolic
OH signals of assembly I at 9.75−9.31 ppm indicated that
approximately 40% had shifted to higher field (broad signal
4.9−3.6 ppm). Evidence that this new signal corresponded to
phenolic OH signals was obtained from NOESY-experiments:
Correlation to the neighboring ortho-aromatic proton and the
methine was observed (Supporting Information, SI, Figure 4).
A similar pattern is seen with the phenolic protons of assembly
I. If the capsule itself was acting as Brønsted acid, then the
addition of 0.5 equiv of amine 3 would indeed affect up to half
of the capsular assemblies and shift their OH signals to higher
1
observed in the H NMR spectra: (1) The integral of the
multiplet at 9.75−9.31 ppm, corresponding to the phenolic OH
signals of the capsule, decreases inversely with added amounts
of NEt3. (2) The aromatic signal of the capsule at 7.20 ppm
(meta to the phenol groups) is split; indicating in total three
different species (highlighted by arrows). These changes are
less pronounced in the other aromatic signal at ca. 6.11 ppm.
(3) The water peak (which before amine addition appears at
2.72 ppm) is gradually shifted to lower field (indicated with an
asterisk) and its integral gradually increases. Careful integration
of the broad signal (see SI Table 1) indicated that in fact the
phenolic signals at 9.75−9.31 ppm that had disappeared had
shifted to the broad peak. Its integral closely matches the
original water amount plus the shifted phenolic protons. This
indicates a fast exchange of protons between phenols and water,
which is not observed in the original capsule I. (4) Guest signal
peaks appear at −0.10 to −0.15 ppm. Integration indicates that
the added guest is completely encapsulated at all concen-
trations. Accordingly, no free guest signals can be observed in
1
the H NMR spectra.
These observations indicate protonation of the added amine
by capsule I. However, it was not clear if the hexameric capsular
structure was still intact as an anionic species after
deprotonation, although the strongly high field-shifted guest
signals (−0.10 to −0.15 ppm) indicated some form of capsular
assembly. DOSY spectroscopy has proven to be the ideal tool
to probe the size of the resorcinarene assembly as demonstrated
by the Cohen group.7,9a−d,10b Therefore, this technique was
used to determine the size of the respective species. The
B
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
observed diffusion value (0.24 0.01 × 10−5 cm2s−1; see SI
Figure 5) for a sample containing 1.4 equiv of triethylamine
(Figure 2, spectrum h; complete deprotonation of I) is in very
good agreement with the literature values of the hexamer I
reported by the Cohen group.7,9a−d,10b Therefore, a smaller
assemblyfor instance a dimer, which was observed in crystal
15
structures can be ruled out.
Taken together, the collected evidence is consistent with the
encapsulation of the amine 3 as a protonated species inside the
negatively charged capsule Idenoted here as HNEt3 @I−
+
(Figure 3). To some extent, a pairwise encapsulation of a
Figure 4. Estimation of the pKa value of I by addition of 0.5 equiv of
base to a solution of I in water-saturated CDCl3 (3.3 mM); pKa values
(measured in water) were taken from literature.16 aNo protonation
was observed.
Figure 3. Schematic representation of the protonation and
encapsulation of NEt3 (3) by I. The larger trioctadecylamine (5) is
protonated by I but cannot be accommodated inside the cavity.
Why is the hexameric capsule I acting as a relatively strong
acid? (1) The protonation leads to thermodynamically more
stable complexes due to cation-π interactions. (2) The anion
formed is stabilized well by delocalization. The negative charge
can be freely shifted between the 48 phenolic groups and the
eight water molecules of assembly I via proton migration (see
Figure 5). (3) Additionally, attractive Coulomb interactions
between the ammonium guest and the negatively charged
capsule could be involved.
protonated amine and regular amineinteracting via the
protonis likely, since we observe complete uptake of 1
equiv of NEt3 but not complete disappearance of the original
phenolic capsule I signals (e.g., Figure 2f). The pairwise
encapsulation of amines could also explain the splitting of the
aromatic signal at 7.20 ppm (Figure 2): The peak highlighted in
green corresponds to the original capsule I; the red arrow
+
marks the peak of HNEt3 @I−; and the yellow arrow refers to
+
assembly Et3N·HNEt3 @I−. Such a pairwise encapsulation of
amines was also observed in metal−ligand based capsules by
the groups of Bergman and Raymond.4j
To corroborate these findings, we tried to obtain direct
evidence for the protonation of the tertiary amine: The affected
methylene groups of NEt3 (3) next to the nitrogen atom were
strongly high field-shifted due to encapsulationconcealing
any possible low field shift caused by protonation. Therefore,
we studied the protonation using trioctadecylamine (5), which
is too large for encapsulation8f and where we therefore could
easily observe possible low field shifts of the affected methylene
1
groups in the H NMR spectrum. Consistent with protonation
of the amine, we observed a shift of the adjacent methylene
groups from 2.37 to 2.84 ppma value which is in good
agreement with the separately synthesized trifluoroacetic acid
salt (see SI Figure 7).
Figure 5. Section of the hydrogen bond seam of capsule I.
After having collected evidence that protonation of tertiary
amines is indeed occurring, we wanted to estimate the acidity of
assembly I. Therefore, we investigated its behavior toward
amines of decreasing basicity16 (Figure 4). Addition of 0.5
equiv of bases with pKa values ranging from 11−6.1 to a
solution of I in water-saturated CDCl3 (3.3 mM) resulted in
approximately 80% of protonation (as revealed by the residual
OH peak at 9.75−9.31 ppm, cf. Figure 2). As discussed before,
complete protonation is not observed, since a pairwise
encapsulation of a protonated and a regular amine is also
We next investigated if encapsulation of an ammonium guest
influences the acidity of I. We chose to synthesize the large
dialkyl aniline 6 (Figure 6) as a probe, which due to its pKa (cf.,
Figure 4) should not be protonated to a high extent and
therefore should allow the observation of increased and
decreased protonation. Additionally, due to its size it cannot
be encapsulated, and therefore, the addition of the ammonium
guest cannot change the encapsulation ratio (which could alter
the protonation equilibrium). While we observed 15
1%
protonation when adding 1 equiv of 6 to I in water-saturated
CDCl3 (3.3 mM), the degree of protonation was increased to
22 1% when the capsule was occupied by a tetrabutylammo-
nium guest (1.0 equiv of Bu4NBr was added to the solution of
I) under otherwise identical conditions. The increased acidity
of I in this case could stem from the energy gain from Coulomb
interactions between the tetrabutylammonium guest and the
negatively charged capsule. This observation was interesting in
observed (Et3N·HNEt3 @I−). Beginning with pyridine (pKa =
+
5.2), we observed a lower degree of protonation (53 1%),
which is further decreased to 23 2% in the case of aniline
(pKa = 4.6). Amines of lower basicity did not show any degree
1
of protonation as evidenced by H NMR spectroscopy. From
these results, we estimated the pKa of hexamer I as
approximately 5.5−6 (for calculations, see SI).17
C
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
For comparison, the carbon analog of 7, namely ethyl 3,3,3-
triphenylpropanoate (8, Figure 7a), which lacks the ability to
accept a proton, was also investigated: It did not yield any
detectable amounts of encapsulated species under similar
conditions, again emphasizing the role of protonation in the
binding of basic compounds. The encapsulated protonated
Wittig ylide 7 was tested for its reactivity toward aldehydes.
Even the addition of excess propanal (10 equiv) did not
produce any alkene product, although the uptake of propanal
into the capsule could be observed by the appearance of
characteristic signals (SI Figure 10). This lack of reactivity
further corroborates our understanding of the acid base
chemistry in this system.
We next synthesized a large Wittig ester 9 (Figure 7b), which
is not able to fit inside I. When 0.85 equiv of 9 were added to a
solution of I (1.0 equiv) in water-saturated CDCl3 (8.0 mM), as
expected, a high degree of protonation (approximately 90%)
and no encapsulation was observed via 1H NMR spectroscopy.
Addition of 1.5 equiv of propanal to this solution resulted in
formation of alkene 11 (66% yield after 16 h), indicating that
the protonation of the Wittig ylide outside of I was reversible
and did not prevent conversion to the alkene. Finally, we tested
if capsule I can act as a selector in the Wittig reaction of a
mixture of ylides 7 and 9 (Figure 7c): Addition of a ylide
mixture (0.85 equiv of 7, 0.85 equiv of 9) to a solution of I in
water-saturated CDCl3 (8.0 mM), followed by the addition of
propanal (1.5 equiv), led to the highly selective formation of
alkene 11 (74% isolated yield, E/Z = 14: 1). Only traces of
alkene 10 could be detected by 1H NMR spectroscopy and 72%
of ylide 7 could be recovered by basic column chromatography.
The selectivity imposed by capsule I (ratio of 10: 11 = 3: 97)
was in stark contrast to the unselective reaction without I under
otherwise identical conditions, which led to a 53:47-mixture of
10 and 11.
Figure 6. The protonation of aniline 6 by I was explored in the
presence and absence of the encapsulated tetrabutylammonium salt.
light of potential applications of I as an acidic catalyst, since
blocking the cavity with the ammonium guest could result in an
increased background reaction outside of the capsule during a
control reaction.
Synthetic Application via Size-Selective Protonation.
After having revealed the acidity of assembly I, we next tried to
translate these findings into first synthetic applications. We
explored the possibility of protonating a stabilized Wittig ylide,
which should have an appropriate pKa value of approximately
8−9.18 Indeed, the addition of 0.85 equiv of Wittig ethyl ester 7
(Figure 7a) to a solution of I (1.0 equiv) in water-saturated
CDCl3 (8.0 mM) resulted in complete protonation of the ylide
(as evidenced by the integral of the original phenolic OH
signals in the 1H NMR spectrum) and complete encapsulation.
Catalytic Size-Selective Hydrolysis of Acetals. After
having demonstrated that the acidity of assembly I can be
efficiently used to impose substrate selectivity on a stoichio-
metric reaction, we wanted to explore the potential use of I as a
selective enzyme-like catalyst. We decided to investigate the
hydrolysis of acetals, and after screening several different acetal
groups found that 1,1-diethoxyethane (12, R = methyl, Figure
8) was a suitable substrate: Addition of 10 equiv of acetal to a
solution of I in water-saturated CDCl3 (3.3 mM) resulted in
good conversion (85%) after 1 h at 25 °C. The control
experiment with inhibited catalyst (adding 2.4 equiv of the
guest Bu4NBr to the catalyst solution before addition of
substrate) only showed a slow background reaction (1% after 1
h). A second control experiment in water-saturated CDCl3
without added catalyst did not show any detectable conversion.
These first results were interesting, since they demonstrated
that a catalytic conversion is indeed possible with I and that the
reaction is taking place inside the cavity. Additionally, the
successful hydrolysis implies that water is able to enter the
cavity of I. It is likely that the eight water molecules that are
already bound at the surface of I (Figure 5) are enabling
hydrolysis.
We next investigated the substrate selectivity of the catalytic
diethyl acetal hydrolysis by varying the alkyl group R (Figure
8). 1,1-Diethoxypropane 13 (R = ethyl) showed comparable
results to 1,1-diethoxyethane 12 (R = methyl), giving
approximately 86% conversion after 60 min. When catalyst I
was blocked with a good guest (Bu4NBr), the hydrolysis of 13
was efficiently slowed down, giving only a weak background
Figure 7. (a) Encapsulation experiments with Wittig ylide 7 and
carbon analog 8; (b) reactivity test of Wittig ylides 7 and 9 in the
presence of I and EtCHO; and (c) comparison of the selective Wittig
reaction in the presence of I and the regular reaction in solution.
D
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
encapsulated aliphatic signals (characteristically shifted to the
region 0.6 to −2 ppm) in the various experiments. In Figure 8b,
these values are plotted versus the respective conversion after
10 min (initial rate). Although the larger acetals (especially 17
and 18) have more aliphatic protons (which potentially could
shift to the high field-region mentioned), we observe a negative
correlation of acetal size and alkyl integral. These findings point
to a faster encapsulation of the smaller guests, a phenomenon
which has been observed in other systems20 but has not been
translated into synthetic applications before. It seems that the
smaller acetals are cleaved much more rapidly due to more
efficient encapsulation.
Finally we explored the possibility of selectively hydrolyzing
one acetal in the presence of another. Indeed, the hydrolysis of
a mixture of 12 and 18 (10 equiv each) proceeded in a highly
selective fashion (Figure 9): After 60 min, the smaller acetal
Figure 9. Selective hydrolysis of acetal 12 in the presence of acetal 18
within catalyst I.
was hydrolyzed to a large extent (83%), while the other acetal
showed only 2% conversionresulting in a selectivity of 98:2
at 85% combined conversion. As a control experiment, we
wanted to replace capsule I with a regular acid of comparable
acidity. When utilizing acetic acid (pKa = 4.8)16 instead of I
under otherwise identical conditions, no conversion was
observed (monitored for 15 h), although its acidity is
approximately one magnitude higher than I. This observation
indicated the fundamental role of the cavity space in the
hydrolysis: We believe that it enables reactions under mild
conditions that are not possible in the solution phase by
stabilizing cationic intermediates and transition states in the
hydrolysis process. To reach comparable hydrolysis rates in
solution, we had to turn to the much stronger trifluoroacetic
acid (pKa = 0.2)16 and use it in large excess (4 equiv per acetal
as compared to 10 mol % I!). Not surprisingly, the hydrolysis
led to mixturesgiving 24% of 19 and 41% of 20 after 60 min
(ratio of 37:63). Thus, we could demonstrate selectivity
imposed by catalyst I in a reaction that is very hard to control
in the solution phase.
Figure 8. (a) Catalytic hydrolysis of various diethyl acetals inside I and
(b) comparison of the conversion after 10 min to encapsulated alkyl
signals (1H NMR region: 0.6 to −2 ppm).
reaction (dotted line in Figure 8), although we observed
increased acidity of I when occupied by the ammonium guest
(see Figure 6). This further demonstrates that the reaction is
indeed greatly accelerated inside the cavity of I. Interestingly,
the reaction rate slowed down considerably with longer alkyl
groups: 1,1-diethoxybutane 14 (R = propyl) gave 62%, 1,1-
diethoxypentane 15 (R = butyl) gave 44%, 1,1-diethoxyhexane
16 (R = pentyl) gave 30%, 1,1-diethoxyoctane 17 (R = heptyl)
gave 13% and 1,1-diethoxydodecane 18 (R = undecyl) gave
only 3.4% conversion after 60 min. The decreased hydrolysis
rate of the longer alkyl acetals cannot be explained by size
exclusion arguments. All of the tested acetals can be
accommodated well inside the cavity of I, as evidenced by
the occupation ratio. Even the largest tested acetal (1,1-
diethoxydodecane) only occupies 22% of the available space,
much less than the optimum of approximately 55%19 (see SI
Table 2). The remaining space would be filled with solvent
molecules. To explain the selectivity, we turned to the extent of
encapsulation of the different acetals. Unfortunately, we could
not determine the encapsulation ratio, since the characteristic
acetal-guest signals are not observable after encapsulation, due
to signal overlap. We therefore tried to utilize the encapsulated
alkyl signals of the acetal, which were shifted to the region of
0.6 to −2 ppm as measure of encapsulation. Since the acetals
utilized differ greatly in the number of alkyl protons, and more
importantly, it is not known which of the respective alkyl
protons are shifted into the observable region due to the
anisotropy of the capsule walls, the encapsulation ratio
remained elusive. We could only compare the integral of the
CONCLUSIONS
■
We have provided evidence that the hexameric capsule I is
acting as a reasonable strong acid (pKa ≈ 5.5−6). This finding
explains its good binding of tertiary amines, which are
protonated to give ammonium ions and then encounter strong
cation-π interactions inside the cavity. Further, we demon-
strated that the acidity of I can be translated into first synthetic
and even catalytic applications.21 The catalytic example
presented mimics the basic principle of operation in enzymes:
A suitable substrate is selectively isolated inside the reaction
pocket, its reactivity is enhanced by protonation, and the
cationic transition states likely stabilized by cation-π inter-
actions with the aromatic cavity. Finally, the product is released,
since it does not bind strongly to I, to complete the catalytic
E
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Raymond, K. N. Science 2007, 316, 85. (f) Hastings, C. J.; Fiedler, D.;
Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2008, 130, 10977.
(g) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc.
2008, 130, 11423. (h) Brown, C. J.; Bergman, R. G.; Raymond, K. N. J.
Am. Chem. Soc. 2009, 131, 17530. (i) Pluth, M. D.; Bergman, R. G.;
Raymond, K. N. Acc. Chem. Res. 2009, 42, 1650. (j) Pluth, M. D.;
Fiedler, D.; Mugridge, J. S.; Bergman, R. G.; Raymond, K. N. Proc.
Natl. Acad. Sci. U. S. A. 2009, 106, 10438. (k) Hastings, C. J.; Pluth, M.
D.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2010, 132,
6938. (l) Brown, C. J.; Miller, G. M.; Johnson, M. W.; Bergman, R. G.;
Raymond, K. N. J. Am. Chem. Soc. 2011, 133, 11964. (m) Hastings, C.
J.; Backlund, M. P.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2011, 50, 10570. (n) Wang, Z. J.; Brown, C. J.; Bergman, R. G.;
Raymond, K. N.; Toste, F. D. J. Am. Chem. Soc. 2011, 133, 7358.
(o) Hart-Cooper, W. M.; Clary, K. N.; Toste, F. D.; Bergman, R. G.;
Raymond, K. N. J. Am. Chem. Soc. 2012, 134, 17873. (p) Wang, Z. J.;
Clary, K. N.; Bergman, R. G.; Raymond, K. N.; Toste, F. D. Nature
Chem. 2013, 5, 100. (q) Ito, H.; Kusukawa, T.; Fujita, M. Chem. Lett.
2000, 598. (r) Yoshizawa, M.; Sato, N.; Fujita, M. Chem. Lett. 2005,
34, 1392. (s) Yoshizawa, M.; Tamura, M.; Fujita, M. Science 2006, 312,
251. (t) Murase, T.; Nishijima, Y.; Fujita, M. J. Am. Chem. Soc. 2012,
134, 162. (u) Merlau, M. L.; del Pilar Mejia, M.; Nguyen, S. T.; Hupp,
J. T. Angew. Chem., Int. Ed. 2001, 40, 4239. (v) Lee, S. J.; Cho, S.-H.;
Mulfort, K. L.; Tiede, D. M.; Hupp, J. T.; Nguyen, S. T. J. Am. Chem.
Soc. 2008, 130, 16828.
cycle. We are convinced that our findings will fuel further
applications of I as a reaction chamber and as an enzyme-like
catalyst.
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed experimental procedures, characterization data for new
compounds and relevant NMR spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
konrad.tiefenbacher@tum.de
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This project was supported by the “Fonds der Chemischen
Industrie” (Sachkostenzuschuss), the TUM Junior Fellow Fund
and the “Dr.-Ing. Leonhard-Lorenz-Stiftung”. The help of M.Sc.
Johannes Richers with graphical design is greatly acknowledged.
We thank PD Dr. Wolfgang Eisenreich and Dr. Christoph Gobl
̈
(both TU Munchen) for help with NMR-DOSY-measurements
̈
(5) (a) Kang, J.; Santamaría, J.; Hilmersson, G.; Rebek, J. J. Am.
Chem. Soc. 1998, 120, 7389. (b) Cavarzan, A.; Scarso, A.; Sgarbossa,
P.; Strukul, G.; Reek, J. N. H. J. Am. Chem. Soc. 2011, 133, 2848.
(6) MacGillivray, L. R.; Atwood, J. L. Nature 1997, 389, 469.
(7) Avram, L.; Cohen, Y. Org. Lett. 2002, 4, 4365.
and Prof. Leonard R. MacGillivray for providing crystallo-
graphic data of hexamer I.
REFERENCES
■
(8) (a) Shivanyuk, A.; Rebek, J. Proc. Natl. Acad. Sci. U. S. A. 2001,
98, 7662. (b) Shivanyuk, A.; Rebek, J. J. Chem. Commun. 2001, 2424.
(c) Palmer, L. C.; Shivanyuk, A.; Yamanaka, M.; Rebek, J. J. Chem.
Commun. 2005, 857. (d) Evan-Salem, T.; Baruch, I.; Avram, L.;
Cohen, Y.; Palmer, L. C.; Rebek, J. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 12296. (e) Avram, L.; Cohen, Y. J. Am. Chem. Soc. 2003, 125,
16180. (f) Slovak, S.; Cohen, Y. Supramol. Chem. 2010, 22, 803.
(g) Philip, I. E.; Kaifer, A. E. J. Am. Chem. Soc. 2002, 124, 12678.
(h) Philip, I.; Kaifer, A. E. J. Org. Chem. 2005, 70, 1558. (i) Bianchini,
G.; Scarso, A.; Sorella, G. L.; Strukul, G. Chem. Commun. 2012, 12082.
(9) (a) Avram, L.; Cohen, Y. J. Am. Chem. Soc. 2002, 124, 15148.
(b) Avram, L.; Cohen, Y. Org. Lett. 2003, 5, 1099. (c) Avram, L.;
Cohen, Y. J. Am. Chem. Soc. 2004, 126, 11556. (d) Avram, L.; Cohen,
Y. Org. Lett. 2008, 10, 1505. (e) Shivanyuk, A.; Rebek, J. J. Am. Chem.
Soc. 2003, 125, 3432. (f) Yamanaka, M.; Shivanyuk, A.; Rebek, J. J. Am.
Chem. Soc. 2004, 126, 2939. (g) Barrett, E. S.; Dale, T. J.; Rebek, J. J.
Am. Chem. Soc. 2007, 129, 3818. (h) Barrett, E. S.; Dale, T. J.; Rebek, J.
J. Am. Chem. Soc. 2008, 130, 2344.
(1) (a) Kirby, A. J. Angew. Chem., Int. Ed. 1996, 35, 707.
(b) Motherwell, W. B.; Bingham, M. J.; Six, Y. Tetrahedron 2001,
57, 4663. (c) Garcia-Viloca, M.; Gao, J.; Karplus, M.; Truhlar, D. G.
Science 2004, 303, 186. (d) Zhang, X.; Houk, K. N. Acc. Chem. Res.
2005, 38, 379. (e) Gao, J.; Ma, S.; Major, D. T.; Nam, K.; Pu, J.;
Truhlar, D. G. Chem. Rev. 2006, 106, 3188. (f) Warshel, A.; Sharma, P.
K.; Kato, M.; Xiang, Y.; Liu, H.; Olsson, M. H. M. Chem. Rev. 2006,
106, 3210. (g) Ringe, D.; Petsko, G. A. Science 2008, 320, 1428.
(h) Kamerlin, S. C. L.; Warshel, A. Proteins: Struct., Funct., Bioinf. 2010,
78, 1339.
(2) (a) Breslow, R.; Dong, S. D. Chem. Rev. 1998, 98, 1997.
(b) Sanders, J. K. M. Chem.−Eur. J. 1998, 4, 1378. (c) Vriezema, D.
M.; Comellas Aragones, M.; Elemans, J. A. A. W.; Cornelissen, J. J. L.
̀
M.; Rowan, A. E.; Nolte, R. J. M. Chem. Rev. 2005, 105, 1445.
(3) Selected examples: (a) Wyler, R.; de Mendoza, J.; Rebek, J.
Angew. Chem., Int. Ed. 1993, 32, 1699. (b) Heinz, T.; Rudkevich, D.
M.; Rebek, J. Nature 1998, 394, 764. (c) Caulder, D. L.; Powers, R. E.;
Parac, T. N.; Raymond, K. N. Angew. Chem., Int. Ed. 1998, 37, 1840.
(d) Davis, A. V.; Fiedler, D.; Ziegler, M.; Terpin, A.; Raymond, K. N. J.
Am. Chem. Soc. 2007, 129, 15354. (e) Takeda, N.; Umemoto, K.;
Yamaguchi, K.; Fujita, M. Nature 1999, 398, 794. (f) Umemoto, K.;
Tsukui, H.; Kusukawa, T.; Biradha, K.; Fujita, M. Angew. Chem., Int.
Ed. 2001, 40, 2620. (g) Gibb, C. L. D.; Gibb, B. C. J. Am. Chem. Soc.
2004, 126, 11408. (h) Kaanumalle, L. S.; Gibb, C. L. D.; Gibb, B. C.;
Ramamurthy, V. J. Am. Chem. Soc. 2004, 126, 14366. For reviews, see:
(i) Hof, F.; Craig, S. L.; Nuckolls, C.; Rebek, J. J. Angew. Chem., Int. Ed.
2002, 41, 1488. (j) Palmer, L. C.; Rebek, J. J. Org. Biomol. Chem. 2004,
2, 3051. (k) Koblenz, T. S.; Wassenaar, J.; Reek, J. N. H. Chem. Soc.
Rev. 2008, 37, 247. (l) Rebek, J. Acc. Chem. Res. 2009, 42, 1660.
(m) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed.
2009, 48, 3418. (n) Wiester, M. J.; Ulmann, P. A.; Mirkin, C. A. Angew.
Chem., Int. Ed. 2011, 50, 114. (o) Ajami, D.; Rebek, J. Acc. Chem. Res.
2012, 46, 990.
(10) (a) Atwood, J. L.; Barbour, L. J.; Jerga, A. Chem. Commun. 2001,
2376. (b) Avram, L.; Cohen, Y. Org. Lett. 2003, 5, 3329.
(11) (a) Ugono, O.; Holman, K. T. Chem. Commun. 2006, 2144.
(b) Schnatwinkel, B.; Stoll, I.; Mix, A.; Rekharsky, M. V.; Borovkov, V.
V.; Inoue, Y.; Mattay, J. Chem. Commun. 2008, 3873. (c) Slovak, S.;
Avram, L.; Cohen, Y. Angew. Chem., Int. Ed. 2010, 49, 428. (d) Slovak,
S.; Cohen, Y. Chem.−Eur. J. 2012, 18, 8515.
(12) (a) Mileo, E.; Yi, S.; Bhattacharya, P.; Kaifer, A. E. Angew. Chem.,
Int. Ed. 2009, 48, 5337. (b) Yi, S.; Mileo, E.; Kaifer, A. E. Org. Lett.
2009, 11, 5690.
(13) Beyeh, N. K.; Kogej, M.; Åhman, A.; Rissanen, K.; Schalley, C.
A. Angew. Chem., Int. Ed. 2006, 45, 5214.
(14) Avram, L.; Cohen, Y.; Rebek, J., Jr. Chem. Commun. 2011, 5368.
(15) (a) Murayama, K. Chem. Commun. 1998, 607. (b) N. Rose, K.
Chem. Commun. 1998, 407. (c) Shivanyuk, A.; Rissanen, K.;
Kolehmainen, E. Chem. Commun. 2000, 1107.
(4) (a) Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2004, 43, 6748. (b) Fiedler, D.; van Halbeek, H.; Bergman, R.
G.; Raymond, K. N. J. Am. Chem. Soc. 2006, 128, 10240. (c) Leung, D.
H.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2007, 129,
2746. (d) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Angew.
Chem., Int. Ed. 2007, 46, 8587. (e) Pluth, M. D.; Bergman, R. G.;
(16) W. P. Jencks, , J. R. Ionization Constants of Acids and Bases; CRC
Press: Cleveland, 1976; Vol. 1.
(17) The calculated pKa-value of I is based on the reported pKa-
values of pyridine and aniline, which were measured in water.
Therefore, the calculated pKa-value of I should be treated as relative
F
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
value (to compare with reported pKa-values measured in water) but
not as an absolute value.
(18) Zhang, X. M.; Bordwell, F. G. J. Am. Chem. Soc. 1994, 116, 968.
(19) Mecozzi, S.; Rebek, J. Chem.−Eur. J. 1998, 4, 1016.
(20) Jiang, W.; Ajami, D.; Rebek, J. J. Am. Chem. Soc. 2012, 134,
8070.
(21) During the preparation of the manuscript the isonitrile
hydration inside I at elevated temperatures was reported. Our findings
could explain this observation, since isonitrile hydration usually
requires an acidic catalyst. See: Bianchini, G.; La Sorella, G.; Canever,
N.; Scarso, A.; Strukul, G. Chem. Commun. 2013, 5322.
G
dx.doi.org/10.1021/ja4080375 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX