Supramolecular catalysis of unimolecular rearrangements: Substrate scope and mechanistic insights

Article abstract of DOI:10.1021/ja062329b

A cavity-containing metal-ligand assembly is employed as a catalytic host for the 3-aza Cope rearrangement of allyl enammonium cations. Upon binding, the rates of rearrangement are accelerated for all substrates studied, up to 850-fold. Activation parameters were measured for three enammonium cations in order to understand the origins of acceleration. Those parameters reveal that the supramolecular structure is able to reduce both the entropic and enthalpic barriers for rearrangement and is highly sensitive to small structural changes of the substrate. The space-restrictive cavity preferentially binds closely packed, preorganized substrate conformations, which resemble the conformations of the transition states. This hypothesis is also supported by quantitative NOE studies of two encapsulated substrates, which place the two reacting carbon atoms in close proximity. The capsule can act as a true catalyst, since release and hydrolysis facilitate catalytic turnover. The question of product hydrolysis was addressed through detailed kinetic studies. We conclude that the iminium product must dissociate from the cavity interior and the assembly exterior before hydroxide-mediated hydrolysis, and propose the intermediacy of a tight ion pair of the polyanionic host with the exiting product.

Full text of DOI:10.1021/ja062329b

Published on Web 07/19/2006
Supramolecular Catalysis of Unimolecular Rearrangements:
Substrate Scope and Mechanistic Insights
Dorothea Fiedler, Herman van Halbeek, Robert G. Bergman, and
Kenneth N. Raymond*
Contribution from the Department of Chemistry, UniVersity of California,
Berkeley, California 94720-1460
Received April 12, 2006; E-mail: raymond@socrates.berkeley.edu
Abstract: A cavity-containing metal-ligand assembly is employed as a catalytic host for the 3-aza Cope
rearrangement of allyl enammonium cations. Upon binding, the rates of rearrangement are accelerated for
all substrates studied, up to 850-fold. Activation parameters were measured for three enammonium cations
in order to understand the origins of acceleration. Those parameters reveal that the supramolecular structure
is able to reduce both the entropic and enthalpic barriers for rearrangement and is highly sensitive to small
structural changes of the substrate. The space-restrictive cavity preferentially binds closely packed,
preorganized substrate conformations, which resemble the conformations of the transition states. This
hypothesis is also supported by quantitative NOE studies of two encapsulated substrates, which place the
two reacting carbon atoms in close proximity. The capsule can act as a true catalyst, since release and
hydrolysis facilitate catalytic turnover. The question of product hydrolysis was addressed through detailed
kinetic studies. We conclude that the iminium product must dissociate from the cavity interior and the
assembly exterior before hydroxide-mediated hydrolysis, and propose the intermediacy of a tight ion pair
of the polyanionic host with the exiting product.
Introduction
exclusive supramolecular structure, and the utilization of weak
and reversible interactions between the components, such as
Inspired by nature’s ability to control and manipulate chemical
reactions by means of steric confinement and precisely posi-
tioned functional group interactions, synthetic chemists have
developed cavity-containing structures, which could potentially
fulfill similar functions. Lehn, Cram, and Pedersen introduced
crown ethers and cryptands as very simple enzyme mimics.1,2
Starting in the 1970s, significant advances were made by using
the naturally occurring cyclodextrins as molecular hosts for
chemical reactivity.3-9 However, as the scale and complexity
of the target reactions and their products increase, the preparation
of catalytic cavitands in a covalent fashion becomes very
difficult, involving costly multistep syntheses that often produce
only milligram yields of material.10-12 For these reasons, the
use of self-assembly is gaining popularity: large and complex
structures can be obtained through the self-assembly of relatively
simple subunits. These subunits are programmed to form one
hydrogen bonds or metal-ligand interactions, allows for the
self-correction and formation of the most favorable thermody-
namic product.13,14 By binding a reactive organometallic transi-
tion metal catalyst into such molecular containers, an active site
can be installed.15-20 Alternatively, the supramolecular host can
act as the catalyst itself. Bimolecular catalysis can occur when
two reactive partners are bound within the same capsule and
their relative encounter frequencies are thus increased.21-25
Phase-transfer catalysis can be achieved when the solubility
properties of the capsule exterior differ notably from those of
the interior environment.26
(13) Yeh, R. M.; Davis, A. V.; Raymond, K. N. In ComprehensiVe Coordination
Chemistry II; Meyer, T. J., Ed.; Elsevier Ltd.: Oxford, U.K., 2004; pp
327-355.
(14) Hof, F.; Craig, S. L.; Nuckolls, C.; Rebek, J., Jr. Angew. Chem., Int. Ed.
2002, 41, 1488-1508.
(15) Merlau, M. L.; Mejia, M. P.; Nguyen, S. T.; Hupp, J. T. Angew. Chem.,
Int. Ed. 2001, 40, 4239-4242.
(16) Slagt, V. F.; Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Angew. Chem., Int. Ed. 2001, 40, 4271-4274.
(1) Cram, D. J. Angew. Chem., Int. Ed. Engl. 1988, 27, 1009-1020.
(2) Lehn, J.-M. Angew. Chem., Int. Ed. Engl. 1988, 27, 89-112.
(3) Takahashi, K. Chem. ReV. 1998, 98, 2013-2034.
(17) Slagt, V. F.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Angew. Chem.,
Int. Ed. 2003, 42, 5619-5623.
(4) Breslow, R.; Guo, T. J. Am. Chem. Soc. 1988, 110, 5613-5617.
(5) Breslow, R.; Anslyn, E. J. Am. Chem. Soc. 1989, 111, 8931-8932.
(6) Breslow, R.; Schmuck, C. J. Am. Chem. Soc. 1996, 118, 6601-6606.
(7) Breslow, R.; Zhang, X.; Xu, R.; Maletic, M. J. Am. Chem. Soc. 1996, 118,
11678-11679.
(18) Slagt, V. F.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H.
J. Am. Chem. Soc. 2004, 126, 1526-1536.
(19) Slone, R. V.; Hupp, J. T. Inorg. Chem. 1997, 36, 5422-5423.
(20) Wilkinson, M. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H. Org. Biomol.
Chem. 2005, 3, 2371-2383.
(8) Breslow, R.; Huang, Y.; Zhang, X.; Yang, J. Proc. Natl. Acad. Sci. U.S.A.
1997, 94, 11156-11158.
(21) Kang, J. M.; Rebek, J., Jr. Nature 1997, 385, 50-52.
(22) Kang, J. M.; Santamaria, J.; Hilmersson, G.; Rebek, J., Jr. J. Am. Chem.
Soc. 1998, 120, 3650-3656.
(9) Breslow, R.; Dong, S. D. Chem. ReV. 1998, 98, 1997-2012.
(10) Coolen, H. K. A. C.; van Leeuwen, P. W. N. M.; Nolte, R. J. M. J. Am.
Chem. Soc. 1995, 117, 11906-11913.
(23) Chen, J.; Rebek, J., Jr. Org. Lett. 2002, 4, 327-329.
(24) Yoshizawa, M.; Takeyama, Y.; Kusukawa, T.; Fujita, M. Angew. Chem.,
Int. Ed. 2002, 41, 1347-1349.
(11) Clyde-Watson, Z.; Vidal-Ferran, A.; Twyman, L. J.; Nakash, M.; Sanders,
J. K. M. New J. Chem. 1998, 493-502.
(25) Yoshizawa, M.; Takeyama, Y.; Okano, T.; Fujita, M. J. Am. Chem. Soc.
2003, 125, 3243-3247.
(12) Mattei, F.; Diederich, F. HelV. Chim. Acta 1997, 80, 1555-1588.
9
10240
J. AM. CHEM. SOC. 2006, 128, 10240-10252
10.1021/ja062329b CCC: $33.50 © 2006 American Chemical Society

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
Figure 2. General reaction scheme of the 3-aza Cope rearrangement.
Starting from the enammonium cation A, [3,3] sigmatropic rearrangement
leads to iminium cation B, which then hydrolyzes to the aldehyde, C.
supramolecular host needed to be identified. The cationic 3-aza
Cope rearrangement seemed to be most suitable to be carried
out in the finite environment of the M4L6 assembly.36-38 The
3-aza Cope (or aza Claisen) reaction is a member of the [3,3]
class of sigmatropic rearrangements and occurs thermally in
N-allyl enamine systems with varying degrees of facility,
depending on structural features.36,37,39,40 Whereas neutral allylic
enamines rearrange to δ-ene imines at rather elevated temper-
atures (170-250 °C), the corresponding protonated, Lewis acid
coordinated, or quaternized molecules tend to rearrange at
considerably milder temperatures (20-120 °C),41 making these
reactions suitable for complex molecule syntheses. A catalytic
enantioselective variant, however, still remains to be devel-
oped.42,43
The substrates for the cationic 3-aza Cope rearrangement are
enammonium cations (A). These cations should bind to the
cavity interior because they closely resemble analogous am-
monium cations, such as NMe2Pr2+, which are known to
strongly associate with the assembly cavity. Sigmatropic rear-
rangement leads to iminium cations (B), which are subsequently
hydrolyzed to the corresponding γ,δ-unsaturated aldehydes (C)
(Figure 2). Since neutral molecules are only very weakly, if at
all, bound by the supramolecular host, we thereby hope to
circumvent the problem of product inhibition. The hydrolysis
step could potentially be followed by binding of more substrate,
enabling catalytic turnover.
A range of enammonium substrates varying in size, shape,
and substitution pattern was investigated (Table 1). All enam-
monium salts were synthesized by alkylation of N,N-dimeth-
ylisobutenylamine with the corresponding allyl bromides or allyl
tosylates, and were obtained as hygroscopic solids in very good
yields.
Encapsulation and Rate Acceleration. All of the substrates,
even the sterically more demanding 8 and 9, were encapsulated
by the metal-ligand assembly. This is most easily observed
by 1H NMR spectroscopy: shielding by the naphthalene moiety
of the ligand scaffold causes an upfield shift of the guest
resonances by 2-3 ppm, and placement into the chiral cavity
causes enantiotopic groups to become diastereotopic. Enam-
monium cation 1 (R1, R2, R3 ) H), for example, quantitatively
yielded the host-guest complex [1⊂Ga4L6]11-, as confirmed
Figure 1. CAChe model of the M4L6 supramolecular assembly, encapsulat-
ing an enammonium substrate. Left: View down the two-fold axis. Right:
View down the three-fold axis. The guest molecule is shown in space-
filling view.
The most common limitation in applying these host templates
as catalytic containers is product inhibition. Since the design
of the cavity relies on the shape complementarity with the
reaction transition state and/or reaction products, the latter tend
to associate with the host interior with similar or higher binding
strength, thereby preventing catalytic turnover. Despite these
difficulties, the number of catalytic scaffolds for bimolecular
reactions or phase-transfer reactions is increasing, but catalytic
cavities for unimolecular reactions are still very rare. We have
previously described the use of a metal-ligand assembly as a
catalytic host for unimolecular rearrangements.27 Here we
expand on our previous report, accounting for the origins of
the observed rate accelerations and providing mechanistic insight
into the separate steps involved in the catalytic cycle.
Results and Discussion
Identification of a Suitable Reaction: 3-Aza Cope Rear-
rangement. Raymond and co-workers have developed su-
pramolecular metal-ligand assemblies of M4L6 stoichiometry
(M ) Ga3+, Al3+, Fe3+, Ge4+, Ti4+, L ) 1,5-bis(2′,3′-
dihydroxybenzamido)naphthalene).28-31 In these structures, the
four metal atoms are located on the vertices of the tetrahedron,
while six bis-bidentate catecholamide ligands span the edges
(Figure 1). Though composed of achiral components, the overall
structure of the tetrahedron is chiral. The chirality is a result of
the tris-bidentate coordination at each metal center, leading to
either ∆ or Λ configuration, and due to strong mechanical
coupling between the metal centers through the rigid ligands,
the tetrahedra are homochiral (with either ∆,∆,∆,∆- or Λ,Λ,Λ,Λ-
configuration) and resolvable.32,33 The metal-ligand assemblies
have provided rich host-guest chemistry: a large variety of
monocationic guest molecules can be encapsulated into the
hydrophobic cavity, ranging from simple alkylammonium ions28
to cationic organometallic species.34,35
In pursuing supramolecular catalysis, a chemical transforma-
tion of a cationic substrate which is compatible with the
(26) Ito, H.; Kusukawa, T.; Fujita, M. Chem. Lett. 2000, 598-599.
(27) Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem., Int. Ed. 2004,
43, 6748-6751.
(34) Fiedler, D.; Leung, D.; Bergman, R. G.; Raymond, K. N. J. Am. Chem.
Soc. 2004, 126, 3674-3675.
(35) Leung, D.; Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2004, 43, 963-966.
(28) Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N. Angew. Chem.,
Int. Ed. 1998, 37, 1840-1843.
(36) Walters, M. A. J. Org. Chem. 1996, 61, 978-983.
(37) Przheval’skii, N. M.; Grandberg, I. I. Russ. Chem. ReV. 1987, 56, 477-
491.
(29) Caulder, D. L.; Raymond, K. N. J. Chem. Soc., Dalton Trans. 1999, 1185-
1200.
(30) Caulder, D. L.; Bru¨ckner, C.; Powers, R. E.; Ko¨nig, S.; Parac, T. N.; Leary,
J. A.; Raymond, K. N. J. Am. Chem. Soc. 2001, 123, 8923-8938.
(31) Davis, A. V.; Raymond, K. N. J. Am. Chem. Soc. 2005, 127, 7912-7919.
(32) Ziegler, M.; Davis, A. V.; Johnson, D. W.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2003, 42, 665-668.
(33) Terpin, A.; Ziegler, M.; Johnson, D. W.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2001, 40, 157-160.
(38) Elkik, E.; Francesch, C. Bull. Soc. Chim. 1968, 3, 903-910.
(39) Opitz, G. Justus Liebigs Ann. Chem. 1961, 122-132.
(40) Blechert, S. Synthesis 1989, 71-82.
(41) Nubbemeyer, U. Top. Curr. Chem. 2005, 244, 149-213.
(42) Enders, D.; Knopp, M.; Schiffers, R. Tetrahedron: Asymmetry 1996, 7,
1847-1882.
(43) Nubbemeyer, U. Synthesis 2003, 7, 961-1008.
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10241

A R T I C L E S
Fiedler et al.
Table 1. Synthesis of the Various Enammonium Salts
Table 2. Rate Constants for Free (kfree) and Encapsulated (kencaps
Rearrangements (Measured at 50 °C in D2O) and Their
Acceleration Factors
)
kfree
10-5 s-
kencaps
10-5 s-
1
1
compound
R1
R2
R3
(×
)
(
×
)
acceleration
1-Br
2-Br
3-Br
4-OTs
5-OTs
6-OTs
7-OTs
8-OTs
9-OTs
10-Br
H
H
H
Et
H
n-Pr
H
i-Pr
n-Bu
TMS
Me
H
H
H
Et
H
n-Pr
H
H
H
Me
3.49
7.61
3.17
1.50
4.04
1.69
0.37
3.97
0.033
6.3
16.3
5
26
141
90
150
44
854
56
Me
H
H
H
H
H
H
H
H
198
446
135
604
74.2
316
222
R1
R2
R3
X-
compound
H
H
H
Br
1-Br
Me
H
H
H
H
H
H
H
H
H
Et
H
n-Pr
H
i-Pr
n-Bu
TMS
Me
H
H
Et
H
n-Pr
H
H
H
Me
Br
Br
2-Br
3-Br
1.17
331
35
53
OTs
OTs
OTs
OTs
OTs
OTs
Br
4-OTs
5-OTs
6-OTs
7-OTs
8-OTs
9-OTs
10-Br
(3 vs 4, and 5 vs 6). Not only the size but also the shape of the
guests seem to be important in achieving optimal rate accelera-
tions. This phenomenon of shape selectivity is not unprecedented
for the M4L6 assembly, and aspects of this type of selectivity
have been discussed before.34,35
by NMR spectroscopy (Figure 3); the enantiotopic methyl
groups on the nitrogen become diastereotopic upon binding, and
a total of four upfield-shifted methyl resonances at 0.24, -0.03,
-0.60, and -1.08 ppm are displayed for the host-guest
complex. Additional confirmation of the 1:1 stoichiometry is
obtained by ESI-mass spectrometry, in which peaks correspond-
ing to a 1:1 host-guest complex are observed.
The rates of rearrangement were measured for the free and
encapsulated enammonium cations to explore potential changes
in substrate reactivity which may occur upon encapsulation. All
rearrangements displayed clean first-order kinetics at 50 °C in
buffered solution (pD ) 8.00). In all cases, though, the
encapsulated substrates rearranged much faster (Table 2). The
highest rate acceleration was observed for the isopropyl-
substituted enammonium cation 7; binding into the host interior
resulted in a rate increase by almost 3 orders of magnitude.
Interestingly, the trend of higher rate accelerations with an
increase in steric demand of the guest molecule is not
monotonic. There appears to be an “optimal fit” for substrate
7, whereas accelerations for the larger substrates 8 and 9 drop
off significantly. In addition, there are notable differences in
the acceleration of the trans- and cis-substituted stereoisomers
The observed rate enhancements might potentially be due to
the cavity’s more hydrophobic environment. Control experi-
ments, however, in which the free rearrangement of 1 was
monitored showed no significant solvent dependence on the rate
(kD O ) 3.49 × 10-5 s-1, kMeOD ) 3.62 × 10-5 s-1, kCDCl
)
2
3
3.84 × 10-5 s-1). In fact, the larger substrates 8 and 9 rearranged
more slowly in methanol than in aqueous solution. The prospect
that the host assembly’s negative charge was causing the rate
acceleration was ruled out by adding salt (2 M KCl) in the
absence of the assembly, which again did not result in a notable
increase in rate for the free rearrangement (kKCl ) 4.02 × 10-5
s-1).
Origins of Acceleration: Determination of Activation
Parameters. To investigate the origin of the observed rate
accelerations, the activation parameters were measured for
substrates 3, 4, and 7 for the free and the encapsulated
rearrangements. The obtained parameters for the free rearrange-
ment of substrate 3, for example, are ∆Hq ) 23.1((0.8) kcal/
mol and ∆Sq ) -8((2) eu. These values are similar to those
reported in the literature for related systems,44 and the highly
organized transition state required for the rearrangement is
mirrored in the negative entropy of activation. To ensure that
Figure 3. 1H NMR spectra of 1-Br (top) and [1⊂Ga4L6]11- (bottom), illustrating the effects of encapsulation.
9
10242 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
Table 3. Summary of Activation Parameters for 3, 4, and 7, Free
and Encapsulated
An illustration of the high sensitivity of the supramolecular
host to guest stereochemistry (and vice versa) are the differences
in chemical shift for the bound guest substrates 3 and 4 (Figure
4). The terminal methyl groups (Metrans) are separated by more
than 0.7 ppm for [3⊂Ga4L6]11- versus [4⊂Ga4L6]11- and
resonate at 0.21 and -0.52 ppm, respectively. For the unbound
3 and 4, the chemical shifts of these methyl groups are
essentially identical. This seems to imply that their averaged
positions within the host cavity differ significantly from each
other. A similar effect is experienced by the methylene protons
on the ethyl substituent (CH2Et); again, a large separation in
their chemical shifts can be observed (Figure 4).
Taken together, these results imply that the host assembly
can selectively bind preorganized, reactive conformations of the
substrates, and these “exclusive conformations” are dependent
on the size and shape of the entering guest substrate. The space-
restrictive host cavity allows for encapsulation of only tightly
packed conformers, closely resembling the conformations of the
transition states. The predisposed conformers, which have
already lost several rotational degrees of freedom, are selected
from an equilibrium mixture of all possible conformers, and
thus the entropic barrier for rearrangement decreases. Added
strain that is induced by squeezing the ground state into the
tight cavity is reflected by a lowered enthalpic barrier. This effect
of preorganization and induced strain becomes more significant
for the larger substrates, which fit more tightly in the host cavity.
If the optimal fit of the reactant transition state with the host
cavity is exceeded, however, rate accelerations drop off again,
as illustrated by substrates 8 and 9.
Evidence for Encapsulated Ground-State Preorganization.
The 2D NOESY spectra of [3⊂Ga4L6]11- and [4⊂Ga4L6]11-
corroborate the effect of preorganization into reactive conforma-
tions in the host-guest systems (Figure 5 and Supporting
Information). The encapsulated enammonium cation 4, for
example, displays strong NOE correlations between protons at
the two ends of the molecule, while unbound 4 shows no such
dipolar couplings between the distal alkyl chains. These
correlations would be anticipated for a tight, closely packed
conformation of the bound substrate.45
this negative entropy of activation was not an artifact of
solvation changes specific to the aqueous medium, the activation
parameters for 3 were also measured in C6D5Cl, again revealing
a negative entropy of activation (Table 3). The encapsulated
reaction of [3⊂Ga4L6]11- in water gave a very similar value
for the enthalpy of activation, ∆Hq ) 23.0((0.9) kcal/mol. The
entropy of activation, in contrast, differed remarkably by almost
10 eu, with ∆Sq ) +2((3) eu. The activation parameters are
summarized in Table 3.
To obtain more quantitative information about the exact
orientation of the bound guest molecules, NOE growth rates
for selected resonances in [3⊂Ga4L6]11- and [4⊂Ga4L6]11- were
determined. Assuming that the initial rate approximation is valid
(a linear NOE growth), the magnitude of an enhancement
between two spins A and B after a period τ will be proportional
to the cross-relaxation rate, which in turn depends on the
distance rAB-6. Using a known internal distance rXY as a
reference, and determining the NOE growth rates σ for that
Consideration of the activation parameters for the different
encapsulated substrates reveals that the reasons for supramo-
lecular catalysis are more complex than simply a reduction of
the entropy of activation, since different effects are observed
for substrates 3, 4, and 7. While the rate acceleration for the
encapsulated 3 was exclusively due to lowering the entropic
barrier, a smaller ∆Sq is also observed for 4 and 7, but for the
latter two the enthalpic barrier to rearrangement is also
decreased. It is possible that, for those two substrates, binding
into the narrow confines of the metal-ligand assembly induces
some strain on the bound molecules, thereby raising their
ground-state energies compared to those of the unbound
substrates. Even though the differences in the activation
parameters are small, they are certainly real. It is especially
interesting to note that the two stereoisomers 3 and 4 respond
differently to the effects of encapsulation.
reference and the unknown distance, the unknown distance rAB
1/6 46
can be calculated according to rAB ) rXY(σXY/σAB
)
.
Even
though the method bears several uncertainties, the success of
the approach relies on the insensitivity of the calculated rAB to
the experimental accuracy, due to the r1/6 dependence (a factor
of 2 error in the growth rate corresponds to only ∼10% error
in the distance measurement).
(44) Jolidon, S.; Hansen, H.-J. HelV. Chim. Acta 1977, 60, 978-1032.
(45) The short mixing time of 90 ms ensures that no correrlations due to spin
diffusion are observed.
(46) Neuhaus, D.; Williamson, M. P. The Nuclear OVerhauser Effect in
Structural and Confomational Analysis, 2nd ed.; VCH Publishers: New
York, 2001.
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10243

A R T I C L E S
Fiedler et al.
Figure 4. Comparison of the 1H NMR spectra of [3⊂Ga4L6]11- and [4⊂Ga4L6]11-. Significant differences in chemical shifts for the encapsulated guest can
be noted, implying a different orientation for the distinct substrates.
Figure 5. 2D NOESY spectrum of [4⊂Ga4L6]11- in a D2O/MeOD mixture (70:30) recorded at -10 °C, mixing time 90 ms. Indicated in red are selected
NOE correlations. The correlation between the methyl groups at the two distal ends of the molecule is a remarkable demonstration of the cavity’s enforcement
of a compressed and folded guest conformation.
molecules. One remarkable correlation is the NOE between
CH3h and CH3k (highlighted in red). This NOE is not observed
for the unbound substrate and again illustrates the close packing
of the bound guest molecule.
Quantification of the NOE growth rate yields a distance of
h
k
3.76 Å between the two methyl groups, CH3 and CH3 .
i
Similarly, the distance between Ha and CH3 was determined
to be 3.90 Å (Figures 6 and 7). In an analogous fashion, the
NOE growth rates for [4⊂Ga4L6]11- were translated into
distances as indicated in Figure 7.
Figure 6. NOE buildup curves for [4⊂Ga4L6]11-. Remarkable is the
k
correlation between CH3h and CH3 ; the two distal methyl groups must be
For both averaged conformations (of 3 and 4), the measured
distances illustrate the close proximity between the two distal
ends of the guest molecules, bringing the two bond-forming C
atoms into contact distances. In fact, employing MM3 modeling
and restricting several intramolecular distances to the calculated
held at close proximity.
Figure 6 depicts a plot of NOE cross-peak intensity vs mixing
time τ for a selected diagonal peak of [4⊂Ga4L6]11- and shows
the typical NOE cross-peak growth for large, slowly tumbling
9
10244 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
Figure 7. Derived intramolecular distances for [4⊂Ga4L6]11- (left) and
[3⊂Ga4L6]11- (right). Calculated distances are denoted in parentheses.
Distances between methyl groups refer to a pseudoatom located in the center
of the plane of the three hydrogen atoms.
values furnishes conformations of 3 and 4 in which the two
reacting C atoms are less than 4 Å apart. The NOE measure-
ments thus demonstrate that the metal-ligand assembly indeed
enforces the binding of preorganized, reactive conformations.
Catalytic Turnover. Since one of our main goals was the
design of catalytic scaffolds, the question arose as to whether
the M4L6 supramolecular assembly would also be able to
mediate the aza Cope rearrangement catalytically. The stoichio-
metric experiments had shown that, in all cases of the host-
mediated 3-aza Cope rearrangement, the iminium cations (B)
hydrolyzed to the corresponding aldehydes (C), leaving behind
an “empty” cavity.47 The empty cavity should be able to bind
more substrate and should therefore make it possible to carry
out the reaction under catalytic conditions. When the reaction
was conducted in the presence of 13 mol % catalyst relative to
enammonium substrate, truly catalytic behavior of the supramo-
lecular host was revealed. Increasing the catalyst loading from
13 to 27 to 40 mol % resulted in the expected rate acclerations;
the observed initial rate constants at 25 °C are k13 mol % ) 0.64
× 10-4 s-1, k27 mol % ) 1.17 × 10-4 s-1, and k40 mol % ) 1.80 ×
10-4 s-1 (Figure 8).
An inhibition experiment with NEt4+ further supports the idea
of the supramolecular assembly providing a catalytic cavity for
rearrangement. NEt4+ has a very high affinity for the assembly
interior and will essentially displace any other guest molecule.
Addition of 8 equiv of NEt4+ led to almost complete inhibition
of the catalytic activity of the supramolecular host (Figure 8).
Based on these results, the catalytic cycle illustrated in Figure
9 is proposed: (1) The restricted space of the host assembly
allows for the binding of only closely packed conformers of
enammonium substrates. (2) The rearrangement proceeds with
significant acceleration within the boundaries of the metal-
ligand assembly, due to lowered enthalpic and entropic barriers.
(3) Equilibration of the rearranged product with the bulk solution
is followed by hydrolysis to the corresponding aldehyde. This
enables catalytic turnover by regeneration of the empty assembly
which can bind additional substrate.
Figure 8. Initial rates for the catalytic 3-aza Cope rearrangement of
[3⊂Ga4L6]11-. Purple b, 40% catalyst loading; green (, 27% catalyst
loading; light blue 2, 13% catalyst loading; orange - -, 40% catalyst loading,
inhibited with 8 equiv of NEt4+; dark blue - -, uncatalyzed reaction. “sm”
denotes starting material, as determined by integration of the 1H NMR
signals.
other catalytic assemblies for different chemical transformations,
this issue needed to be examined.
The system of choice for these studies was host-guest
complex [10⊂Ga4L6]11-. Unlike most other bound enammonium
substrates, the rearrangement of 10 led to almost quantitative
formation of the iminium product 10i inside the host cavity at
pD 8.00 (Figure 10).48 That is, subsequent hydrolysis of
encapsulated 10i was slow compared to the initial rearrangement
step and therefore could be monitored over time. The decay of
[10i⊂Ga4L6]11- displayed clean first-order kinetics, and a rate
of hydrolysis could be obtained (Figure 11).
Other studies by Raymond and co-workers have established
that negatively charged molecules cannot enter the host cavity,
even transiently, due to strong electrostatic repulsion with the
highly negatively charged host structure.49 Hydrolysis of imi-
nium cations takes place at neutral pH but can also be catalyzed
by hydroxide ions.50 If hydrolysis of 10i proceeds exclusively
inside the protective host cavity, which prevents hydroxide ions
from entering, the rate of hydrolysis should be independent of
hydroxide concentration. The only nucleophile that can poten-
tially enter the inner space is water, and its concentration remains
constant when it is used as the solvent. If, however, iminium
dissociation precedes hydrolysis, a pH dependence is anticipated;
outside of the shielded cavity, the iminium ion can be attacked
by the hydroxide nucleophile.
Iminium Hydrolysis: Inside or Outside the Capsule? One
question that the proposed catalytic cycle did not address is that
of iminium hydrolysis. Does water enter the host cavity and
hydrolyze the iminium products to the corresponding aldehydes
in the interior, or is dissociation of the iminium products
necessary before hydrolysis can take place? With the goal of
optimizing the supramolecular catalyst and eventually designing
A series of kinetic runs was performed in buffered solutions
over a pD range from 6.5 to 12.8, and the data are summarized
(48) For several of the other substrates 2-9, iminium formation was also
detected, but product 10i is unique in that its rate of formation is
substantially greater than its rate of displacement from the cavity.
Presumably, 10i has a higher binding constant for the cavity interior
compared to those of the other iminium intermediates, which results in a
longer lifetime.
(47) “Empty” cavity in this case refers to the host assembly, incorporating 1
equiv of NMe4+. NMe4+ is a very weakly binding guest molecule and can
easily be replaced by the entering enammonium substrates.
(49) Leung, D.; Bergman, R. G.; Raymond, K. N., unpublished results.
(50) Hine, J.; Evangelista, R. A. J. Am. Chem. Soc. 1980, 102, 1649-1655.
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10245

A R T I C L E S
Fiedler et al.
Figure 9. Proposed catalytic cycle for the cationic 3-aza Cope rearrangement.
Figure 10. Rearrangement of [10⊂Ga4L6]11- leads to almost quantitative formation of the encapsulated iminium product [10i⊂Ga4L6]11-
.
in Figure 12. In the neutral pD range (pD 6-8), the rates of
hydrolysis are almost constant; presumably in this range water
acts as the nucleophile. At increasing pD, however, a depen-
dence on hydroxide concentration is observed up to a pD of
about 11, at which point the system seems to reach saturation.
The linear first-order dependence on hydroxide concentration
in the pD 9-10.5 range supports a mechanistic model which
envisions reversible iminium egression before hydrolysis. The
rate at saturation implies that iminium dissociation becomes rate
limiting, because hydrolysis becomes faster than re-encapsula-
tion. One surprising aspect of this rationale is that the rate of
guest egression did not compare very well to previously reported
rates of exchange for guest molecules of similar size and shape.51
For example, the rate constant for guest exchange at 50 °C for
for 10i, k323(10i) ) 0.002 s-1. Prior kinetic experiments also
demonstrated that rates of guest exchange were sensitive to the
concentration of “innocent” counterions, cations that did not
actively participate in guest exchange, such as NMe4+. Through-
out the previous experiments, the concentration of NMe4+ was
held constant (6 equiv used in all experiments).
To elucidate the role of NMe4+ in the rearrangement reaction
and iminum hydrolysis, variable [NMe4+] kinetics were con-
ducted at high pD (12.64) and constant ionic strength (0.5 M,
adjusted with KCl). The kinetics revealed a significant depen-
+
dence of iminium hydrolysis on NMe4 concentration (Figure
+
13). Increased NMe4 concentration led to faster hydrolysis,
and a first-order dependence on [NMe4+] was observed. At
higher concentrations of [NMe4+], saturation behavior was again
observed.
NMe2Pr2 is k323(NMe2Pr2+) ) 1020 s-1, which is about
+
500 000 times faster than the rate constant reached at saturation
The dependence of kobs on [NMe4+] can be interpreted in
two ways. If we assume an SN2-type bimolecular pathway for
guest exchange, added NMe4+ might assist guest displacement
(51) Davis, A. V.; Fiedler, D.; Seeber, G.; Zahl, A.; van Eldik, R.; Raymond,
K. N. J. Am. Chem. Soc. 2006, 128, 1324-1333.
9
10246 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
from the cavity interior, thus accelerating release of the iminium
cation into the surrounding basic solution. However, NMe4+ is
an extremely weakly binding guest, making this explanation
unlikely. In contrast, calorimetric studies have suggested a
significant interaction of NMe4 with the assembly exterior;52
+
in fact, the affinity of NMe4+ is higher for exterior than interior
binding sites (Kext ) 50 M-1, Kint ) 33 M-1). This exterior
interaction implies the intermediacy of a tight ion pair of the
tetrahedral host with the iminium cation, as illustrated in Figure
14. We suggest that, in the tight ion pair, 10i is closely associated
with the anionic host, not allowing hydroxide to attack the
iminium due to electrostatic repulsion. Only when 10i is released
into solution can hydroxide-catalyzed hydrolysis take place.
To determine which pathway was operating, the rate depen-
dence on [NBu4+] was studied. The tetrabutylammonium cation
is too large to fit into the cavity interior and therefore cannot
displace a guest molecule from that location. Exterior counterion
displacement, in contrast, is viable, since calorimetric studies
Figure 11. Disappearance of starting material [10⊂Ga4L6]11- (pink 9) is
accompanied by formation of [10i⊂Ga4L6]11-. Hydrolysis of [10i⊂Ga4L6]11-
(blue b) is slow and can be monitored over time. Concentrations are
measured relative to an internal standard.
indicate a strong exterior interaction of tetrabutylammonium
with the metal-ligand assembly.52 As was the case with NMe4
,
+
+
increasing concentrations of NBu4 led to faster rates of
hydrolysis, and a first order dependence on [NBu4+] was
observed (see Supporting Information).53 With this information,
the mechanistic model illustrated in Figure 14, which proposes
the intermediacy of a tight ion pair between the exiting iminium
cation and the assembly exterior, becomes more persuasive.
Additional proof for an intermediate that contains iminium
10i ion paired to the assembly exterior comes from a trapping
experiment with NEt4+, which is a very strongly binding guest
molecule that displaces almost all other ammonium cations.
Under the conditions of slowest iminium hydrolysis (pD 6.5
and no added counterions), about 1.5 equiv of NEt4+ was added
to a solution of K11[10i⊂Ga4L6]. An exchange reaction took
place within minutes and could be monitored by 1H NMR
spectroscopy. NEt4+ now occupies the cavity interior, and two
new broad resonances at 2.43 and 0.78 ppm emerge. Integration
of these peaks suggests that the resonance at 2.43 ppm
Figure 12. Plot of log[kobs] versus pD reveals a dependence on hydroxide
concentration for the hydrolysis reaction.
+
corresponds to the NMe2 methyl groups, whereas the broad
signal at 0.78 ppm accounts for the other four methyl groups
of 10i (see Supporting Information). The slight upfield shift of
those resonances and their broadness imply a tight ion-pairing
interaction with the assembly exterior. As anticipated, slow
hydrolysis of 10i takes place, and the resonances of the ion-
paired exterior iminium ion decay over time.
These observations of strong external interactions of the
exiting guest molecule with the host structure are further
substantiated by the findings of Leung et al.54 To determine
the rate for guest passage, several large phosphine ligands were
used to trap a cationic iridium guest species upon egression from
the cavity. Surprisingly, a neutral phosphine could trap the
iridium complex much faster than a trianionic phosphine. When
Figure 13. Kinetics with variable concentrations of NMe4+ reveal a first-
order dependence of the rate of iminium hydrolysis on NMe4+ concentration.
+
Figure 14. Tight ion pairing might be responsible for the dependence of iminium hydrolysis on the counterion NMe4
.
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10247

A R T I C L E S
Fiedler et al.
Figure 16. Rearrangement of K11[3⊂Ga4L6] (shown) and K11[4⊂Ga4L6]
(not shown) leads to almost quantitative formation of the corresponding
encapsulated iminium species. Since the iminium cation is chiral, the
formation of two diastereomers is observed.
disappearance of 10i is observed due to a background reaction.
This background reaction is the water-mediated hydrolysis of
10i, since water, as a neutral nucleophile, can approach and
intercept the ion-paired iminium cation (k5).
This argument is supported by the fact that the rates of
hydrolysis for K11[10i⊂Ga4L6] are independent of pD, in the
+
absence of exterior counterions NMe4 and NBu4+. The
observed rate constants for the disappearance of 10i are
essentially identical at pD 6.5 and 12.8 and correspond to the
background reaction rate k5. Even at elevated pD, hydroxide
cannot approach the ion-paired iminium ion and hydrolysis relies
on the neutral nucleophile water.
Chiral Induction in Sigmatropic Rearrangements. In the
absence of strongly binding counterions, the iminium intermedi-
ate 10i had a remarkably long lifetime inside (and bound
externally to) the supramolecular host (t1/2 ≈ 4 h). A similar
phenomenon was observed for the other substrates, 1-8. In the
absence of added NMe4+, the lifetime of the iminium intermedi-
ates was increased, and the formation of the iminium intermedi-
ates was fast compared to their rates of hydrolysis (except for
Figure 15. Overall proposed mechanism of iminium hydrolysis.
considering the possibility of an intermediate tight ion pair of
the cationic iridium complex with the host exterior, Leung et
al. were able to rationalize their findings: due to electrostatic
repulsion, the anionic phosphine cannot come close enough to
the iridium center when engaged in the ion pair, therefore
slowing the apparent rate of trapping compared to the neutral
phosphine trap.
The overall proposed mechanism is illustrated in Figure 15:
the encapsulated enammonium substrate 10 rearranges inside
the cavity interior to the iminium cation 10i (k1). The rear-
1
9), allowing for their detection by H NMR spectroscopy.55
Since the rearrangement of prochiral substrates 3-8 led to chiral
iminium products, the formation of two diastereomeric host-
guest complexes was the result. A representative 1H NMR
spectrum of the rearranged products of K11[3⊂Ga4L6] is shown
in Figure 16. The resonances for the two diastereomeric guests
are well resolved and allow for the quantification of the
diastereomeric excesses in these reactions.
Unfortunately, the degree of stereoselectivity in these reac-
tions is very low, and only for the rearrangement of K11-
[8⊂Ga4L6] was a notable discrimination between the two
diastereomeric transition states observed: one of the iminium
enantiomers is formed with a diastereomeric excess of ∼20%.
These results imply that, in principle, the resolved host
structure could serve as a stereoselective reaction vessel for
pericyclic reactions. The realization of catalytic enantioselective
pericyclic reactions has proven to be a challenge. In particular,
no good method for induction of stereoselectivity exists for
substrates that do not contain any coordinating groups (func-
+
rangement step is, as anticipated, independent of pD and NMe4
concentration (see Supporting Information). The product 10i can
reversibly dissociate from the interior to the assembly exterior,
where it is tightly ion paired (k2). In the presence of the
counterions NMe4+ and NBu4+, which have significant affinity
for the assembly exterior, 10i can be displaced from the exterior
and released into the surrounding solution (k3). Once 10i is
dissociated from the highly negatively charged cage, hydroxide
can attack the iminium ion, catalyzing its hydrolysis to the
corresponding aldehyde (k4). In the absence of NMe4+ or NBu4
breakup of the tight ion pair is very slow; in this case, the slow
+
,
(52) Michels, M.; Raymond, K. N., unpublished results.
+
(53) The strong exterior interactions of NBu4 with the supramolecular host
are also evidenced by the altered solubility properties of the host-guest
+
assembly. Addition of more than 2 equiv of NBu4 led to immediate
+
precipitation of the assembly, presumably due to coordination of the ions
to the exterior. The experimentally accessible concentration range for NBu4+
was therefore limited.
(55) In previous experiments, 6 equiv of NMe4 was present, resulting in a
more transient nature of the iminium products 2i-8i. We believe that
+
NMe4 is able to displace the iminium products from the cavity interior;
(54) Leung, D. H.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc., in
press.
concentration of the weakly bound 2i-8i can only build up in the absence
of competing guest molecules.
9
10248 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
cold petroleum ether. The organic phase was washed with 2 × 50 mL
of 50% saturated aqueous NaHCO3 and 1 × 50 mL of saturated aqueous
NaHCO3. The aqueous layers were combined and extracted with 1 ×
50 mL of petroleum ether, and the combined organic layers were dried
over K2CO3. After filtration, the volatile materials were removed by
rotary evaporation, leaving behind the allyl tosylates as colorless to
pale yellow oils in quantitative yields. The tosylates were characterized
tional groups with which the substrate can coordinate to a chiral
Lewis acid).43,56,57 The approach employed here thus constitutes
an attractive alternative: instead of employing complicated
chiral auxiliaries, chirality in these reactions is envisioned to
be transmitted by the chiral host environment. This system,
however, still needs to be optimized, and structural features that
will lead to higher degrees of recognition and stereoselectivity
need to be identified and introduced into the structures of either
host or guest.
1
by H NMR only and used without further purification. The tosylates
can be stored in a benzene matrix at -80 °C for several weeks.
Data for Allyl Tosylates, Prepared As Described Above. cis-2-
Pentenyl Tosylate. 1H NMR (500 MHz, CDCl3): δ 7.80 (d, 3J ) 8.0
Conclusion
3
Hz, 2H, OTs), 7.34 (d, J ) 8.0 Hz, 2H, OTs), 5.66 (m, 1H, dCH),
3
5.41 (m, 1H, dCH), 4.60 (d, J ) 7.5 Hz, 2H, OCH2), 2.45 (s, 3H,
Overall, our findings emphasize the ability of molecular
container compounds to provide a well-defined microenviron-
ment which can be utilized for the catalysis of unimolecular
rearrangements. By binding the substrates in reactive conforma-
tions, as evidenced by kinetic, activation parameter, and
quantitative NOE studies, the host assembly accelerates the rates
of rearrangement by up to 3 orders of magnitude. The cavity-
catalyzed rearrangement is followed by product release from
the cavity interior to an exterior binding site, where it can then
be hydrolyzed, and this hydrolysis step generates catalytic
turnover. With this, the large potential of supramolecular
assemblies as synthetically useful tools in organic chemistry
becomes apparent, and we can be certain that, with the multitude
of available supramolecular structures, many transformations
susceptible to supramolecular catalysis are to be discovered.
3
3
OTs), 1.99 (quint., J ) 7.5 Hz, 2H, CH2), 0.93 (t, J ) 7.5 Hz, 3H,
CH3).
1
trans-2-Hexenyl Tosylate. H NMR (500 MHz, CDCl3): δ 7.79
3
3
(d, J ) 8.0 Hz, 2H, OTs), 7.34 (d, J ) 8.0 Hz, 2H, OTs), 5.74 (dt,
3J ) 15.5 Hz, J ) 6.5 Hz, 1H, dCH), 5.45 (dt, J ) 15.5 Hz, J )
3
3
3
3
7.0 Hz, 1H, dCH), 4.49 (d, J ) 7.0 Hz, 2H, OCH2), 2.44 (s, 3H,
3
OTs), 1.97 (m, 2H, CH2), 1.36 (sext., J ) 7.5 Hz, 2H, CH2), 0.83 (t,
3J ) 7.5 Hz, 3H, CH3).
1
cis-2-Hexenyl Tosylate. H NMR (500 MHz, CDCl3): δ 7.79 (d,
3J ) 8.2 Hz, 2H, OTs), 7.34 (d, 3J ) 8.0 Hz, 2H, OTs), 5.66 (dt, 3J )
10.9 Hz, 3J ) 7.6 Hz, 1H, dCH), 5.45 (dtt, 3J ) 10.9 Hz, 3J ) 7.1 Hz,
4J ) 1.4 Hz, 1H, dCH), 4.59 (d, 3J ) 7.1 Hz, 2H, OCH2), 2.44 (s, 3H,
OTs), 1.94 (qd, 3J ) 7.5 Hz, 4J ) 1.3 Hz, 2H, CH2), 1.32 (sext., 3J )
3
7.4 Hz, 2H, CH2), 0.83 (t, J ) 7.4 Hz, 3H, CH3).
1
4-Methyl-2-trans-pentenyl Tosylate. H NMR (500 MHz, C6D6):
δ 7.76 (d, 3J ) 8.2 Hz, 2H, OTs), 6.67 (d, 3J ) 8.0 Hz, 2H, OTs), 5.32
3
3
3
Experimental Section
(dd, J ) 15.4 Hz, J ) 6.5 Hz, 1H, dCH), 5.18 (dtd, J ) 15.4 Hz,
3J ) 6.5 Hz, 4J ) 1.2 Hz, 1H, dCH), 4.32 (d, 3J ) 6.6 Hz, 2H, OCH2),
1.90 (sext., 3J ) 6.7 Hz, 1H, i-Pr), 1.82 (s, 3H, OTs), 0.70 (d, 3J ) 6.8
Hz, 6H, i-Pr).
General Considerations. All reagents were obtained from com-
mercial suppliers and used without further purification unless stated
otherwise. Anhydrous solvents were dried over activated alumina. K6-
(NMe4)5[NMe4⊂Ga4L6], K12[Ga4L6], Na12[Ga4L6], and N,N-dimethyl-
isobutenylamine were prepared according to the published proce-
dures.58,59 NMR spectra were obtained on Bruker Avance AV 400 or
DRX 500 MHz spectrometers. 1H chemical shifts are reported as δ in
ppm relative to residual protonated solvent resonances; 13C chemical
shifts are measured relative to solvent resonances and reported as δ in
ppm. Coupling constants are reported in hertz. IR spectra were recorded
on an Avatar 370 FT-IR instrument. Elemental analyses were performed
by the Microanalytical Laboratory in the College of Chemistry at the
University of California, Berkeley, and amounts of solvent contained
in the analyses were confirmed by 1H NMR spectroscopy. Electrospray
mass spectra were recorded on a triple-quadrupole VG Quatttro mass
spectrometer in the Mass Spectrometry Laboratory in the College of
Chemistry at the University of California, Berkeley. FAB spectra were
recorded at the University of California, Berkeley, Mass Spectrometry
facility; the symbol N+ denotes the parent ion of the enammonium
cation.
General Procedure for Tosylation Reactions. This is a modified
literature procedure:60 in a 500 mL Schlenk flask, 15.0 mM of the
appropriate allyl alcohol was combined with 80 mL of dry and degassed
THF, and the solution was cooled to -78 °C. To this solution was
added n-BuLi (9.34 mL, 15.0 mM, 1.6 M in hexanes). Tosyl chloride
(2.86 g, 15.0 mM) was added in a single portion, and the homogeneous
solution was put in the freezer at -80 °C for 2 days. The reaction was
worked up by diluting the reaction mixture at -78 °C with 150 mL of
1
trans-2-Heptenyl Tosylate. H NMR (500 MHz, CDCl3): δ 7.79
3
3
(d, J ) 8.3 Hz, 2H, OTs), 7.33 (d, J ) 8.1 Hz, 2H, OTs), 5.74 (dt,
3J ) 15.3 Hz, J ) 6.7 Hz, 1H, dCH), 5.44 (dtt, J ) 15.3 Hz, J )
3
3
3
4
3
6.7 Hz, J ) 1.4 Hz, 1H, dCH), 4.49 (d, J ) 6.7 Hz, 2H, OCH2),
2.44 (s, 3H, OTs), 1.99 (m, 2H, CH2), 1.27 (m, 4H, CH2), 0.86 (t, J
3
) 7.0 Hz, 3H, CH3).
trans-3-(Trimethylsilyl)allyl Tosylate. 1H NMR (500 MHz,
3
3
CDCl3): δ 7.80 (d, J ) 8.3 Hz, 2H, OTs), 7.34 (d, J ) 8.0 Hz, 2H,
3
OTs), 5.97-5.86 (m, 2H, dCH), 4.54 (d, J ) 3.9 Hz, 2H, OCH2),
2.45 (s, 3H, OTs), 0.02 (s, 9H, TMS).
General Procedure for Alkylation Reactions. This is a modified
literature procedure:38 under a moisture-free atmosphere, a cooled
solution (0 °C) of N,N-dimethylisobutenylamine (500 µL, 3.76 mmol)
in 5 mL of dry acetonitrile was combined with a cooled solution of
the appropriate allyl or propargyl bromide/tosylate (4.14 mmol) in 5
mL of acetonitrile via cannula. The reaction mixture was stirred at 0
°C for 48 h. All volatile materials were removed under reduced pressure,
leading to sticky, oily residues. The residues were washed with 5 ×
20 mL of dry Et2O, yielding the analytically pure products.
Data for Enammonium Salts, Prepared As Described Above.
[NMe2(allyl)(CHdC(CH3)2)]+Br- (1-Br). The product was isolated
as a very hygroscopic white powder in 94% yield (778 mg, 3.53 mmol).
3
1H NMR (500 MHz, CDCl3): δ 5.95 (s, br, 1H, dCH), 5.84 (ddt, J
) 16.9 Hz, 3J ) 9.9 Hz, 3J ) 7.1 Hz, 1H, dCH), 5.75 (dd, 3J ) 16.9
Hz, 2J ) 1.2 Hz, 1H, dCH2), 5.56 (dd, 3J ) 9.9 Hz, 2J ) 1.3 Hz, 1H,
3
dCH2), 4.50 (d, J ) 7.0 Hz, 2H, CH2), 3.50 (s, 6H, N(CH3)2), 1.92
4
4
(d, J ) 1.3 Hz, 3H, CH3), 1.74 (d, J ) 1.3 Hz, 3H, CH3). 13C{1H}
(125.8 MHz, CDCl3): δ 135.2 (dCH), 129.1 (dCH), 128.5 (Cquart),
124.8 (dCH2), 69.3 (CH2), 54.1 (N(CH3)2), 25.3 (CH3), 19.1 (CH3).
IR (neat): ν ) 3065, 3032, 3013, 2934, 1473, 1462, 1444, 1426, 1404,
1381, 1086, 1023 cm-1. LRFAB-MS (+): m/z 140 [N+]. Anal. Calcd
for C9H18BrN: C, 49.10; H, 8.24; N, 6.36. Found: C, 48.88; H, 8.39;
N, 6.12.
(56) Hiersemann, M.; Abraham, L. Eur. J. Org. Chem. 2002, 1461-1471.
(57) Abraham, L.; Czerwonka, R.; Hiersemann, M. Angew. Chem., Int. Ed. 2001,
40, 4700-4703.
(58) Caulder, D. L.; Powers, R. E.; Parac, T. N.; Raymond, K. N. Angew. Chem.,
Int. Ed. 1998, 37, 1840-1843.
(59) Ellenberger, M. R.; Dixon, D. A.; Farneth, W. E. J. Am Chem. Soc. 1981,
103, 5377-5382.
(60) Barta, N. S.; Cook, G. R.; Landis, M. S.; Stille, J. R. J. Org. Chem. 1992,
57, 7188-7194.
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10249

A R T I C L E S
Fiedler et al.
[NMe2(methallyl)(CHdC(CH3)2)]+Br- (2-Br). The product was
128.5 (OTs), 125.8 (OTs), 116.0 (dCH), 64.2 (CH2), 54.1 (N(CH3)2),
29.7 (CH2), 25.5 (OTs), 22.2, 21.2, 18.9, 13.6 (CH3 and CH2). IR
(neat): ν ) 3023, 2967, 1657, 1493.1458, 1210, 1191, 1118, 1033,
1010 cm-1. LRFAB-MS (+): m/z 182 [N+]. Anal. Calcd for C19H31-
NO3S: C, 64.55; H, 8.84; N, 3.96. Found: C, 64.29; H, 8.97; N, 4.20.
[NMe2(4-methyl-2-trans-pentenyl)(CHdC(CH3)2)]+OTs- (7-OTs).
The product was isolated as a hygroscopic white solid in 92% yield
(1.17 g, 3.46 mmol). 1H NMR (500 MHz, CDCl3): δ 7.77 (d, 3J ) 8.1
Hz, 2H, OTs), 7.13 (d, 3J ) 7.9 Hz, 2H, OTs), 6.08 (dd, 3J ) 15.3 Hz,
isolated as a very hygroscopic white powder in 91% yield (801 mg,
1
3.42 mmol). H NMR (500 MHz, CDCl3): δ 5.98 (s, br, 1H, dCH),
2
2
5.47 (d, J ) 1.3 Hz, 1H, dCH), 5.30 (d, J ) 1.3 Hz, 1H, dCH),
4
4.43 (s, 2H, CH2), 3.48 (s, 6H, N(CH3)2), 1.91 (d, J ) 1.4 Hz, 3H,
CH3), 1.77 (s, 3H, CH3), 1.71 (d, J ) 1.4 Hz, 3H, CH3). 13C{1H}
4
(125.8 MHz, CDCl3): δ 134.3 (dCH), 133.4 (dCH), 129.5 (Cquart),
127.2 (dCH), 72.2 (CH2), 54.5 (N(CH3)2), 25.3 (CH3), 22.9 (CH3),
19.1 (CH3). IR (neat): ν ) 3010, 2971, 2943, 2736, 1643, 1468, 1440,
1418, 1372, 1085, 1006 cm-1. LRFAB-MS (+): m/z 154 [N+]. Anal.
Calcd for C10H20BrN: C, 51.29; H, 8.61; N, 5.98. Found: C, 51.02;
H, 8.83; N, 5.80.
3
3J ) 6.8 Hz, 1H, dCH), 5.81 (s, br, 1H, dCH), 5.38 (dtd, J ) 15.4
3
4
3
Hz, J ) 7.6 Hz, J ) 1.2 Hz, 1H, dCH), 4.30 (d, J ) 7.4 Hz, 2H,
CH2), 3.46 (s, 6H, N(CH3)2), 2.35 (m, 1H, i-Pr), 2.33 (s, 3H, OTs),
4
4
[NMe2(trans-2-pentenyl)(CHdC(CH3)2)]+Br- (3-Br). The product
was isolated as a hygroscopic colorless oil in 95% yield (887 mg, 3.57
1.95 (d, J ) 1.3 Hz, 3H, CH3), 1.79 (d, J ) 1.4 Hz, 3H, CH3), 0.98
(d, J ) 6.8 Hz, 3H, CH3). 13C{1H} (125.8 MHz, CDCl3): δ 153.2
3
1
3
3
(dCH), 143.9 (OTs), 139.0 (OTs), 135.1 (dCH), 128.6 (Cquart), 128.5
(OTs), 125.9 (OTs), 113.7 (dCH), 69.7 (CH2), 53.9 (N(CH3)2), 31.4,
25.5 (OTs), 25.5, 21.6, 21.3, 19.0 (CH3 and CH). IR (neat): ν ) 3027,
2965, 2869, 1494, 1461, 1191, 1120, 1032, 1010 cm-1. LRFAB-MS
(+): m/z 182 [N+]. Anal. Calcd for C19H31NO3S: C, 64.55; H, 8.84;
N, 3.96. Found: C, 64.36; H, 8.99; N, 4.08.
mmol). H NMR (500 MHz, CDCl3): δ 6.20 (dt, J ) 15.3 Hz, J )
3
3
6.4 Hz, 1H, dCH), 5.87 (s, br, 1H, dCH), 5.40 (dt, J ) 15.3 Hz, J
) 7.6 Hz, 1H, dCH), 4.39 (d, J ) 7.4 Hz, 2H, CH2), 3.44 (s, 6H,
3
3
4
N(CH3)2), 2.02 (quint., J ) 7.2 Hz, 2H, CH2), 1.90 (d, J ) 1.3 Hz,
3H, CH3), 1.73 (d, 4J ) 1.3 Hz, 3H, CH3). 13C{1H} (125.8 MHz,
CDCl3): δ 148.1 (dCH), 134.9 (dCH), 128.5 (Cquart), 115.3 (dCH),
69.2 (CH2), 53.7 (N(CH3)2), 25.5 (CH3), 25.3 (CH2), 19.1 (CH3), 12.5
(CH3). LRFAB-MS (+): m/z 168 [N+]. IR (neat): ν ) 3010, 2963,
2935, 2874, 1665, 1462, 1453, 1375, 1086 cm-1. Anal. Calcd for C11H22-
BrN: C, 53.23; H, 8.93; N, 5.64. Found: C, 52.88; H, 9.23; N, 5.73.
[NMe2(cis-2-pentenyl)(CHdC(CH3)2)]+OTs- (4-OTs). The product
was isolated as a white hygroscopic powder in 94% yield (887 mg,
3.57 mmol). 1H NMR (500 MHz, CDCl3): δ 7.73 (d, 3J ) 8.1 Hz, 2H,
OTs), 7.10 (d, 3J ) 7.9 Hz, 2H, OTs), 5.92 (dt, 3J ) 10.9 Hz, 3J ) 7.7
[NMe2(trans-2-heptenyl)(CHdC(CH3)2)]+OTs- (8-OTs). The prod-
uct was isolated as a white solid in 90% yield (1.24 g, 3.38 mmol). 1H
NMR (500 MHz, CDCl3): δ 7.74 (d, 3J ) 8.0 Hz, 2H, OTs), 7.11 (d,
3
3
3J ) 7.9 Hz, 2H, OTs), 6.07 (dt, J ) 15.3 Hz, J ) 6.8 Hz, 1H, d
CH), 5.80 (s, br, 1H, dCH), 5.39 (dt, 3J ) 15.3 Hz, 3J ) 7.4 Hz, 1H,
dCH), 4.24 (d, 3J ) 7.4 Hz, 2H, CH2), 3.40 (s, 6H, N(CH3)2), 2.31 (s,
3
4
3H, OTs), 2.05 (q, J ) 7.0 Hz, 2H, CH2), 1.90 (d, J ) 1.2 Hz, 3H,
4
CH3), 1.75 (d, J ) 1.2 Hz, 3H, CH3), 1.35-1.20 (m, 4H, CH2), 0.86
(t, J ) 7.1 Hz, 3H, CH3). 13C{1H} (125.8 MHz, CDCl3): δ 146.7
3
3
3
Hz, 1H, dCH), 5.83 (s, br, 1H, dCH), 5.39 (dt, J ) 10.9 Hz, J )
7.8 Hz, 1H, dCH), 4.31 (d, 3J ) 7.7 Hz, 2H, CH2), 3.44 (s, 6H,
(dCH), 143.9 (OTs), 139.0 (OTs), 134.9 (dCH), 128.7 (Cquart), 128.5
(OTs), 125.8 (OTs), 116.5 (dCH), 65.8 (CH2), 53.7 (N(CH3)2), 32.3
(CH2), 30.5 (CH2), 25.5 (OTs), 22.2, 21.2, 18.9, 13.8 (CH3 and CH2).
LRFAB-MS (+): m/z 196 [N+]. Anal. Calcd for C20H33NO3S: C,
65.36; H, 9.05; N, 3.81. Found: C, 65.03; H, 9.20; N, 3.59.
[NMe2(trans-3-(trimethylsilyl)allyl)(CHdC(CH3)2)]+OTs- (9-
OTs). The product was isolated as a white powder in 95% yield (1.37
N(CH3)2), 2.30 (s, 3H, OTs), 2.02 (dquint., J ) 7.6 Hz, 4J ) 1.3 Hz,
3
4
4
2H, CH2), 1.91 (d, J ) 1.3 Hz, 3H, CH3), 1.74 (d, J ) 1.3 Hz, 3H,
CH3), 0.91 (t, J ) 7.5 Hz, 3H, CH3). 13C{1H} (125.8 MHz, CDCl3):
3
δ 145.0 (dCH), 143.9 (OTs), 139.0 (OTs), 134.9 (dCH), 128.7 (Cquart),
128.5 (OTs), 125.8 (OTs), 115.2 (dCH), 63.9 (CH2), 53.9 (N(CH3)2),
25.5 (OTs), 21.2, 21.1, 18.9, 13.5 (CH3 and CH2). IR (neat): ν ) 3028,
2966, 2934, 2872, 1483, 1461, 1205, 1184, 1119, 1032, 1010 cm-1
.
1
3
g, 3.57 mmol). H NMR (500 MHz, CDCl3): δ 7.79 (d, J ) 8.4 Hz,
LRFAB-MS (+): m/z 168 [N+]. Anal. Calcd for C18H29NO3S: C,
63.68; H, 8.61; N, 4.13. Found: C, 63.50; H, 8.79; N, 4.19.
2H, OTs), 7.14 (d, 3J ) 8.0 Hz, 2H, OTs), 6.42 (d, 3J ) 18.0 Hz, 1H,
dCH), 5.93 (dt, J ) 18.2 Hz, J ) 7.0 Hz, 1H, dCH), 5.86 (s, br,
1H, dCH), 4.40 (d, J ) 6.5 Hz, 2H, CH2), 3.53 (s, 6H, N(CH3)2),
3
3
3
[NMe2(trans-2-hexenyl)(CHdC(CH3)2)]+OTs- (5-OTs). The prod-
uct was isolated as a hygroscopic colorless oil in 88% yield (1.12 g,
3.31 mmol) which solidified upon standing in the freezer at -30 °C.
1H NMR (500 MHz, CDCl3): δ 7.73 (d, 3J ) 8.1 Hz, 2H, OTs), 7.11
4
4
2.33 (s, 3H, OTs), 1.97 (d, J ) 1.2 Hz, 3H, CH3), 1.82 (d, J ) 1.2
Hz, 3H, CH3), 0.02 (s, 9H, TMS). 13C{1H} (125.8 MHz, CDCl3): δ
147.7 (dCH), 143.8 (OTs), 139.1 (OTs), 135.0 (dCH), 131.1 (dCH),
128.7 (Cquart), 128.5 (OTs), 125.9 (OTs), 71.5 (CH2), 54.2 (N(CH3)2),
25.5 (OTs), 21.3, 19.0 (CH3), -1.8 (TMS). LRFAB-MS (+): m/z 212
[N+]. Anal. Calcd for C19H33NO3SSi: C, 59.49; H, 8.67; N, 3.65.
Found: C, 59.43; H, 8.77; N, 3.66.
3
3
3
(d, J ) 7.9 Hz, 2H, OTs), 6.06 (dt, J ) 15.3 Hz, J ) 6.9 Hz, 1H,
3
3
dCH), 5.81 (s, br, 1H, dCH), 5.40 (dt, J ) 15.3 Hz, J ) 7.4 Hz,
1H, dCH), 4.23 (d, J ) 7.4 Hz, 2H, CH2), 3.39 (s, 6H, N(CH3)2),
3
3
4
2.30 (s, 3H, OTs), 2.02 (quart., J ) 7.0 Hz, 2H, CH2), 1.90 (d, J )
1.4 Hz, 3H, CH3), 1.74 (d, 4J ) 1.4 Hz, 3H, CH3), 1.36 (sext., 3J ) 7.4
[NMe2(3,3-dimethylallyl)(CHdC(CH3)2)]+Br- (10-Br). The prod-
uct was isolated in 82% yield (765 mg, 3.08 mmol) as an off-white
sticky residue which solidified upon cooling to -30 °C. 1H NMR (500
Hz, 2H, CH2), 0.85 (t, J ) 7.3 Hz, 3H, CH3) ppm. 13C{1H} (125.8
3
MHz, CDCl3): δ 146.4 (dCH), 143.9 (OTs), 139.0 (OTs), 134.8 (d
CH), 128.7 (Cquart), 128.5 (OTs), 125.8 (OTs), 116.7 (dCH), 69.4 (CH2),
53.7 (N(CH3)2), 34.6 (CH2), 25.4 (OTs), 21.6, 21.2, 18.9, 13.6 (CH3
and CH2). IR (neat): ν ) 3030, 2959, 2928, 2871, 1665, 1455, 1378,
1192, 1118, 1033, 1011 cm-1. LRFAB-MS (+): m/z 182 [N+]. Anal.
Calcd for C19H31NO3S: C, 64.55; H, 8.84; N, 3.96. Found: C, 64.18;
H, 9.02; N, 4.16.
3
MHz, CDCl3): δ 5.87 (s, br, 1H, dCH), 5.26 (t, J ) 7.9 Hz, 1H,
dCH), 4.50 (d, 3J ) 7.9 Hz, 2H, CH2), 3.59 (s, 6H, N(CH3)2), 2.00 (d,
4J ) 1.3 Hz, 3H, CH3), 1.90 (s, 3H, CH3), 1.83 (s, 6H, 2 × CH3).
13C{1H} (125.8 MHz, CDCl3): δ 148.6 (dCH), 135.2 (dCH), 128.8
(Cquart), 111.3 (dCH), 65.6 (CH2), 53.8 (N(CH3)2), 26.4, 25.7, 19.5,
19.2 (4 × CH3). LRFAB-MS (+): m/z 168 [N+]. Anal. Calcd for
C11H22BrN‚1/3H2O: C, 51.91; H, 8.99; N, 5.51. Found: C, 51.82; H,
8.98; N, 5.55.
[NMe2(cis-2-hexenyl)(CHdC(CH3)2)]+OTs- (6-OTs). The product
was isolated as a hygroscopic white powder in 94% yield (1.20 g, 3.53
1
3
mmol). H NMR (500 MHz, CDCl3): δ 7.76 (d, J ) 8.1 Hz, 2H,
OTs), 7.12 (d, 3J ) 7.9 Hz, 2H, OTs), 5.97 (dt, 3J ) 11.0 Hz, 3J ) 7.6
Hz, 1H, dCH), 5.84 (s, br, 1H, dCH), 5.46 (m, 1H, dCH), 4.35 (d,
3J ) 7.7 Hz, 2H, CH2), 3.49 (s, 6H, N(CH3)2), 2.32 (s, 3H, OTs), 2.13
General Procedure for Encapsulation Reactions. K6(NMe4)5-
[NMe4⊂Ga4L6] (26.3 mg, 7.50 µmol) and the enammonium salt (7.50
µmol) were combined in a vial and dissolved in 500 µL of cold D2O.
The solution was transferred to an NMR tube and the spectrum recorded
within 5 min after dissolution. Due to the reactive nature of these host-
guest complexes, the complexes were only characterized in solution.
Representative mass spectrometry data for the allyl enammonium host-
guest complexes is given for host-guest complex Na11[1⊂Ga4L6]. The
3
4
4
(dq, J ) 7.3 Hz, J ) 1.2 Hz, 2H, CH2), 1.95 (d, J ) 1.3 Hz, 3H,
4
3
CH3), 1.78 (d, J ) 1.3 Hz, 3H, CH3), 1.36 (sext., J ) 7.3 Hz, 2H,
CH2), 0.86 (t, J ) 7.3 Hz, 3H, CH3). 13C{1H} (125.8 MHz, CDCl3):
3
δ 143.9 (OTs), 143.5 (dCH), 139.0 (OTs), 135.1 (dCH), 128.8 (Cquart),
9
10250 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006

Supramolecular Catalysis of Unimolecular Rearrangements
A R T I C L E S
utilization of Na+ as the counterion has the advantage of providing a
less complicated isotope pattern, thus leading to better signal intensities.
Data for Host-Guest Complexes, Prepared As Described Above.
(s, br, 3H, CH3, encaps.), -1.37 (m, br, 1H, CH2, encaps.), -1.42 (m,
br, 1H, CH2, encaps.), -1.53 (m, br, 1H, CH2, encaps.), -1.64 (m, br,
1H, CH2, encaps.).
K5(NMe4)6[7⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.47 (s, 12H,
N-H), 8.05 (s, br, 12H, Ar-H), 7.69 (s, br, 12H, Ar-H), 7.26 (s, br,
12H, Ar-H), 6.97 (s, br, 12H, Ar-H), 6.70 (s, br, 12H, Ar-H), 6.56
(s, br, 12H, Ar-H), 2.70 (m, br, 1H, dCH, encaps.), 2.52 (s, 72H,
NMe4, exterior), 1.84 (s, br, 1H, dCH, encaps.), 1.66 (m, br, 1H, d
CH, encaps.), 0.93 (s, br, 3H, CH3, encaps.), 0.85 (s, br, 3H, CH3,
encaps.), 0.69 (m, br, 1H, CH2, encaps.), 0.49 (s, br, 3H, CH3, encaps.),
-0.42 (m, br, 1H, CH2, encaps.), -1.32 (s, br, 3H, CH3, encaps.), -1.63
(s, br, 3H, CH3, encaps.), -1.71 (s, br, 3H, CH3, encaps.), -1.79 (m,
br, 1H, CH, encaps.).
1
K5(NMe4)6[1⊂Ga4L6]. H NMR (500 MHz, D2O): δ 13.50 (s, 12H,
3
N-H), 7.95 (s, br, 12H, Ar-H), 7.77 (s, br, 12H, Ar-H), 7.25 (d, J
) 7.9 Hz, 12H, Ar-H), 7.02 (t, 3J ) 8.0 Hz, 12H, Ar-H), 6.70 (d, 3J
) 7.2 Hz, 12H, Ar-H), 6.56 (t, 3J ) 7.7 Hz, 12H, Ar-H), 3.09 (d, 3J
) 7.2 Hz, 1H, dCH2, encaps.), 2.65 (s, 72H, NMe4, exterior), 2.44 (s,
br, 1H, dCH, encaps.), 2.32 (m, 1H, dCH, encaps.), 0.24 (s, 3H, CH3,
encaps.), 0.21 (m, 1H, CH2, encaps.), 0.05 (m, 1H, CH2, encaps.), -0.03
(s, 3H, CH3, encaps.), -0.60 (s, 3H, CH3, encaps.), -1.08 (s, 3H, CH3,
encaps.). ES(-)-MS (H2O/MeOH, 50:50) (( ) [Ga4L6],12-): m/z 605
[N+⊂( + 2Na+ + 4H+]5-, 610 [N+⊂( + 3Na+ + 3H+]5-, 614 [N+⊂(
+ 4Na+ + 2H+]5-, 619 [N+⊂( + 5Na+ + 1H+]5-, 623 [N+⊂( +
K5(NMe4)6[8⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.41 (s, 12H,
N-H), 7.94 (s, br, 12H, Ar-H), 7.77 (s, br, 12H, Ar-H), 7.24 (s, br,
12H, Ar-H), 6.99 (s, br, 12H, Ar-H), 6.70 (s. br, 12H, Ar-H), 6.56
(s, br, 12H, Ar-H), 2.50 (s, br, 72H, NMe4, exterior), 1.6 - (-1.4)
(many broad muliplets, not resolved, encapsulated substrate).
K5(NMe4)6[9⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.46 (s, 12H,
N-H), 7.93 (d, 3J ) 7.6 Hz, 12H, Ar-H), 7.88 (d, 3J ) 8.5 Hz, 12H,
Ar-H), 7.28 (d, 3J ) 7.4 Hz, 12H, Ar-H), 7.13 (t, 3J ) 8.1 Hz, 12H,
Ar-H), 6.75 (dd, 3J ) 7.3 Hz, 4J ) 1.3 Hz 12H, Ar-H), 6.60 (t, 3J )
7.8 Hz, 12H, Ar-H), 2.52 (s, br, 72H, NMe4, exterior), 3.3 - (-1.2)
(many broad resonances, not resolved, encapsulated substrate).
K5(NMe4)6[10⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.59 (s, 12H,
6Na+]5-, 764 [N+⊂( + 3Na+ + 4H+]4-, 769 [N+⊂( + 4Na+
+
,
3H+]4-, 774 [N+⊂( + 5Na+ + 2H+]4-, 780 [N+⊂( + 6Na+ + 1H+]4-
785 [N+⊂( + 7Na+]4-, 1017 [N+⊂( + 3Na+ + 5H+]3-, 1024 [N+⊂(
+ 4Na+ + 4H+]3-, 1032 [N+⊂( + 5Na+ + 3H+]3-, 1039 [N+⊂( +
6Na+ + 2H+]3-, 1046 [N+⊂( + 7Na+ + 1H+]3-, 1054 [N+⊂( +
8Na+]3-
.
K5(NMe4)6[2⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.44 (s, 12H,
3
N-H), 7.90 (s, br, 12H, Ar-H), 7.68 (s, br, 12H, Ar-H), 7.18 (d, J
) 8.0 Hz, 12H, Ar-H), 6.93 (t, 3J ) 7.8 Hz, 12H, Ar-H), 6.62 (d, 3J
3
) 7.0 Hz, 12H, Ar-H), 6.48 (t, J ) 7.5 Hz, 12H, Ar-H), 2.52 (s,
72H, NMe4, exterior), 2.38 (s, br, 1H, dCH, encaps.), 1.63 (s, br, 1H,
dCH, encaps.), 0.89 (s, br, 1H, dCH, encaps.), 0.34 (s, 3H, CH3,
encaps.), 0.21 (d, 3J ) 12.5 Hz, 1H, CH2, encaps.), 0.08 (d, 3J ) 12.5
Hz, 1H, CH2, encaps.), -0.24 (s, 3H, CH3, encaps.), -0.48 (s, 3H,
CH3, encaps.), -1.24 (s, 6H, 2 × CH3, encaps.).
3
N-H), 8.08 (s, br, 12H, Ar-H), 7.83 (s, br, 12H, Ar-H), 7.34 (d, J
) 8.0 Hz, 12H, Ar-H), 7.08 (t, 3J ) 7.9 Hz, 12H, Ar-H), 6.77 (d, 3J
3
) 7.2 Hz, 12H, Ar-H), 6.62 (t, J ) 7.6 Hz, 12H, Ar-H), 2.54 (s,
72H, NMe4, exterior), 1.72 (m, br, 1H, dCH, encaps.), 1.29 (s, 1H,
dCH, encaps.), 0.41 (s, 3H, CH3, encaps.), 0.35 (m, 2H, CH2, encaps.),
0.29 (s, 6H, 2 × CH3, encaps.), 0.16 (s, 3H, CH3, encaps.), -1.42 (s,
3H, CH3, encaps.), -1.51 (s, 3H, CH3, encaps.).
K5(NMe4)6[3⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.48 (s, 12H,
3
N-H), 7.91 (s, br, 12H, Ar-H), 7.63 (s, br, 12H, Ar-H), 7.17 (d, J
) 8.0 Hz, 12H, Ar-H), 6.89 (t, 3J ) 7.9 Hz, 12H, Ar-H), 6.61 (d, 3J
3
K5(NMe4)6[10i⊂Ga4L6]. 1H NMR (500 MHz, D2O, N-H not
) 7.5 Hz, 12H, Ar-H), 6.46 (t, J ) 7.5 Hz, 12H, Ar-H), 2.55 (s,
3
72H, NMe4, exterior), 2.31 (m, br, 1H, dCH, encaps.), 1.54 (s, br, 1H,
dCH, encaps.), 1.34 (m, br, 1H, dCH, encaps.), 0.68 (s, 3H, CH3,
encaps.), 0.38 (m, br, 1H, CH2, encaps.), 0.31 (s, 3H, CH3, encaps.),
0.22 (s, 3H, CH3, encaps.), -0.07 (m, br, 1H, CH2, encaps.), -0.42
(m, br, 1H, CH2, encaps.), -0.57 (m, br, 1H, CH2, encaps.), -1.07 (t,
br, 3H, CH3), -1.28 (s, 3H, CH3, encaps.).
observed due to H-D exchange): δ 8.12 (d, J ) 7.5 Hz, 12H, Ar-
3
3
H), 7.83 (d, J ) 8.4 Hz, 12H, Ar-H), 7.29 (d, J ) 8.3 Hz, 12H,
Ar-H), 7.04 (t, 3J ) 8.0 Hz, 12H, Ar-H), 6.76 (d, 3J ) 7.5 Hz, 12H,
Ar-H), 6.62 (t, 3J ) 7.6 Hz, 12H, Ar-H), 4.35 (s, 1H, dCH, encaps.),
3.91 (d, 3J ) 10.5 Hz, 1H, dCH, encaps.), 2.56 (s, 72H, NMe4,
exterior), 2.30 (m, 1H, dCH, encaps.), 1.67 (s, 3H, NMe2, encaps.),
0.86 (s, 3H, NMe2, encaps.), -1.86 (s, 3H, CH3, encaps.), -2.02 (s,
3H, CH3, encaps.), -2.20 (s, 3H, CH3, encaps.), -2.24 (s, 3H, CH3,
encaps.).
K5(NMe4)6[4⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.45 (s, 12H,
3
N-H), 7.93 (s, br, 12H, Ar-H), 7.65 (s, br, 12H, Ar-H), 7.18 (d, J
) 8.0 Hz, 12H, Ar-H), 6.89 (t, 3J ) 7.7 Hz, 12H, Ar-H), 6.61 (d, 3J
3
) 7.4 Hz, 12H, Ar-H), 6.47 (t, J ) 7.5 Hz, 12H, Ar-H), 2.82 (m,
Kinetic Runs of Enammonium Substrates. Kinetic runs were
performed in D2O on a Bruker AVB 400 or DRX 500 MHz
spectrometer. The concentration of all samples was 15 mM, and the
solutions were buffered with 150 mM phosphate buffer, adjusted to
pD 8.00. Sealed capillaries containing a D2O/dioxane mixture served
as internal standard for integration. Temperatures in the probe were
measured with an ethylene glycol standard.
br, 1H, dCH, encaps.), 2.56 (s, 72H, NMe4, exterior), 2.29 (s, br, 1H,
dCH, encaps.), 2.19 (m, br, 1H, dCH, encaps.), 0.73 (s, 3H, CH3,
encaps.), 0.36 (m, br, 1H, CH2, encaps.), 0.14 (m, 2H, CH2), -0.42 (s,
3H, CH3, encaps.), -1.03 (s, 3H, CH3, encaps.), -1.13 (t, br, 3H, CH3),
-1.32 (m, 2H, CH2, encaps.).
K5(NMe4)6[5⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.42 (s, 12H,
3
NOE Growth Rates. Samples of [3⊂Ga4L6]11- and [4⊂Ga4L6]11-
were prepared as 15 mM solutions in a D2O/MeOD mixture (70:30).
2D NOESY spectra were recorded at -10 °C on a Bruker Avance AV
500 spectrometer, with varying mixing times (20, 40, 65, 90, 120, 150,
200, 300, 500, 750, and 1000 ms). NOE cross-peak intensity was plotted
versus mixing time (τm). Using the geminal diastereotopic methylene
protons as a known internal reference distance rXY (1.75 Å), the NOE
growth rates σ for the unknown distances could be quantitated,
N-H), 7.88 (s, br, 12H, Ar-H), 7.65 (s, br, 12H, Ar-H), 7.15 (d, J
3
) 8.0 Hz, 12H, Ar-H), 6.87 (s, br, 12H, Ar-H), 6.61 (d, J ) 7.7
3
Hz, 12H, Ar-H), 6.46 (t, J ) 7.5 Hz, 12H, Ar-H), 3.11 (s, br, 1H,
dCH, encaps.), 2.54 (s, 72H, NMe4, exterior), 1.76 (s, br, 1H, dCH,
encaps.), 0.90 (s, 3H, CH3, encaps.), 0.38 (s, 3H, CH3, encaps.), 0.31
(m, br, 1H, CH2, encaps.), 0.16 (m, br, 1H, CH2, encaps.), -0.21 (s,
br, 3H, CH3, encaps.), -0.46 (m, br, 1H, CH2, encaps.), -0.89 (s, br,
4H, CH3 and CH2, encaps.), -1.14 (s, br, 3H, CH3, encaps.), -1.58
(m, br, 2H, CH2, encaps.).
according to rAB ) rXY(σXY/σAB) .
1/6 46 Methyl groups are treated as rapid
K5(NMe4)6[6⊂Ga4L6]. 1H NMR (500 MHz, D2O): δ 13.45 (s, 12H,
N-H), 8.03 (s, br, 12H, Ar-H), 7.73 (s, br, 12H, Ar-H), 7.26 (s, br,
12H, Ar-H), 6.96 (s, br, 12H, Ar-H), 6.70 (s, br, 12H, Ar-H), 6.57
(s, br, 12H, Ar-H), 3.58 (m, br, 1H, dCH, encaps.), 2.52 (s, 72H,
NMe4, exterior), 2.42 (m, br, 1H, dCH, encaps.), 1.37 (s, br, 1H, d
CH, encaps.), 0.79 (s, br, 3H, CH3, encaps.), 0.61 (s, br, 3H, CH3,
encaps.), 0.42 (m, br, 1H, CH2, encaps.), 0.14 (m, br, 1H, CH2, encaps.),
-0.33 (s, br, 3H, CH3, encaps.), -0.95 (s, br, 3H, CH3, encaps.), -1.11
rotors, and NOE growth rates for methyl groups are divided by 3 to
account for the precence of three correlating protons. Distances to
methyl groups are reported relative to a pseudoatom located in the center
of the plane of the three protons.
Variable pD Kinetics. All solutions were prepared in D2O contain-
ing 10 mM K5(NMe4)6[10⊂Ga4L6] and 100 mM of the different buffer
solutions. Solutions were buffered at pD 6.5, 7.7, 8.1, 8.3 (phosphate
buffer), 8.5, 8.6, 8.8, 9.1 (tris buffer), 9.7, 10.1, 10.3, 10.5, 10.6, 11.1
9
J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006 10251

A R T I C L E S
Fiedler et al.
(carbonate buffer), and 12.8 (KOD). The standard conversion for the
pH-calibrated electrode of pD ) pH + 0.4 was employed.61 The pD
was adjusted using either KOD (1 M in D2O) or DCl (1 M in D2O).
The pD was remeasured after each kinetic run and remained constant
for all samples. The pD was not corrected for temperature. Kinetic
runs were performed on a Bruker DRX 500 MHz spectrometer at 50
°C ((0.2 °C). Sealed capillaries containing a D2O/dioxane mixture
served as internal standard for integration. Temperatures in the probe
were measured before each kinetic run with an ethylene glycol standard.
a stock solution of K11[10⊂Ga4L6] (12.5 mM) was combined with 0-55
µL of a 0.10 M NBu4Cl solution (precipitation occurred at higher
temperatures when more than 55 µL of NBu4Cl was added). All samples
were adjusted to a total volume of 500 µL and transferred to an NMR
tube. Kinetic runs were performed on a Bruker DRX 500 MHz
spectrometer at 50 °C ((0.2 °C). Sealed capillaries containing a D2O/
dioxane mixture served as internal standard for integration. Temper-
atures in the probe were measured before each kinetic run with an
ethylene glycol standard.
+
Variable NMe4 Kinetics. All solutions were prepared using 0.01
Acknowledgment. The authors thank Dennis H. Leung, Dr.
Gojko Lalic, and Dr. Anna V. Davis for helpful discussions.
This work was supported by the Director, Office of Energy
Research, Office of Basic Energy Sciences, Chemical Sciences
Division, of the U.S. Department of Energy under contract DE-
AC03-7600098.
M KOD in D2O, and the overall ionic strength was kept at 0.5 M with
added KCl. For each sample, 300 µL of a stock solution of K11-
[10⊂Ga4L6] (16.67 mM) was combined with 0-100 µL of an NMe4-
Br solution (1.00 M) and 66-100 µL of a 2.5 M KCl solution, so that
the overall ionic strength of NMe4 and K+ amounted to 0.5 M. All
+
samples were brought to a total volume of 500 µL and transferred to
an NMR tube. Kinetic runs were performed on a Bruker DRX 500
MHz spectrometer at 50 °C ((0.2 °C). Sealed capillaries containing a
D2O/dioxane mixture served as internal standard for integration.
Temperatures in the probe were measured before each kinetic run with
an ethylene glycol standard.
Supporting Information Available: Eyring plots for 3 and
[3⊂Ga4L6]11-; 2D NOESY spectrum of [3⊂Ga4L6]11-; depen-
+
dence of hydrolysis on NBu4 concentration; displacement of
10i from the cavity interior with NEt4+; independence of the
initial rearrangement step on pD and NMe4+ concentration. This
material is available free of charge via the Internet at
http://pubs.acs.org.
+
Variable NBu4 Kinetics. All solutions were prepared using 0.5
M KCl in D2O, containing 0.01 M KOD. For each sample, 400 µL of
(61) Perrin, D. D.; Dempsey, B. Buffers for pH and Metal Ion Control; Chapman
and Hall: London, 1974.
JA062329B
9
10252 J. AM. CHEM. SOC. VOL. 128, NO. 31, 2006