Ajami and Rebek
JOCArticle
resorcinarene at the end of the capsule. Consequently, in
extended conformation C2 is closer to the aromatic walls
where the best CH-π interactions are possible. The next
carbon atom, C3, is located nearer the central axis and C4 is
again nearer the walls. In an extended conformation only
these last four carbon atoms are affected; C5 is in the region
of the electron-poor imide walls. The longer guests,
n-alkanes from C15 to C19, must undergo increasing com-
pression to fit within 1.24.1.26,27 As the alkane coils, the
number of methylene groups that fit into the aromatic zone
increases and gauche conformations force all protons
near the twisted walls. There they experience the chiral en-
vironment; for example, encapsulated n-nonadecane shows
six upfield shifted signals and all five sets of the geminal
hydrogens are diastereotopic.
FIGURE 3. Left: A top view of the cavitand part of 1.24.1
enantiomers minimized at B3LYP/3-21G. The walls of the capsule
are twisted and create an asymmetric magnetic environment. Right:
Calculated NICS values for the space inside using B3LYP/6-31G*.
Energetics. There are long-standing problems posed by
these reversible encapsulation complexes that can be traced
to the historical fact that we have never seen an empty
capsule. If there are no suitable guests to occupy the space,
the capsule simply does not assemble. Instead, some other
undefined aggregates are present that show broad, uninter-
pretable NMR spectra. The glycoluril itself in bulk mesity-
lene solution is also loosely organized in some hydrogen-
bonded arrays, again shown by a broad cluster of reso-
nances. Consequently, although it is possible to get relative
association constants through direct competition between
guests and derive thermodynamic parameters for guest ex-
change from these experiments, the absolute values are not
accessible: the resting state of the capsule alone is not known.
The energetic forces that drive the formation of the extended
assembly are not measured. The contributions that can
be identified must include the enthalpic benefits of a com-
plete set of hydrogen bonds to the components, including
those between the best donors (cavitand) and acceptors
(glycolurils). In the extended capsule the glycourils can have
the maximum number of hydrogen bonds with the minimum
number of partners. Another enthalpic benefit is the relaxa-
tion of the coiled guest to its extended conformation that
relieves gauche interactions. But during the unwinding of an
alkane guest, the C-H/π interactions within the aromatic
inner space must decrease and this desolvation of surfaces
creates a vacuum;clearly enthalpic liabilities. In the context
of proteins, this corresponds to ∼1.1 kcal/mol in free energy
for the creation of a cavity the size of a methylene group.32
The flexibility of the capsule’s hydrogen-bonded parts allows
some collapse around guests, so the empty space is not easily
estimated. From an entropic perspective, the first interpreta-
tion of the formation of this ordered, 6-component þ guest
assembly should be that it is extremely unfavorable. But on
reflection, any assembly (and even any molecular recogni-
tion event) liberates solvent molecules from the surfaces of
the solutes that become engaged. This is an entropic benefit,
but the number of such solvent molecules released in the
assembly process is difficult to even estimate, let alone
determine for organic solvents. (Again, there are figures
glycoluril-based capsules,28 and those involving asymmetric
elements outside capsules.29 The NMR spectrum of trans-7-
tetradecene (Figure 2) shows only 4 upfield shifted signals;
only 4 carbon atoms of this alkene can fit in the aromatic
envelope in an extended conformation. The methyl groups
are positioned at the tapered ends of the capsule where
the highest upfield shifts occur: Δδ is nearly -4.7 ppm!
In contrast, the vinyl hydrogen signals located near the
center of cavity are shifted downfield by 0.72 ppm.
Computational methods were applied to further explore
the induced magnetic shielding regions of the cavity. A map
of the induced magnetic shielding/deshielding for the inner
space was prepared by using nucleus independent chemical
shift (NICS)30 calculations at the B3LYP/6-31G* level
of density functional theory.31 These values are shown in
Figure 3 for coordinates along the central axis of 1.24.1 with
˚
spacing distances of 1 A. The calculated values are in good
agreement with the experimental results; the map can be used
to locate and monitor the position of a guest nucleus inside
the cavity.
The evidence of the capsule’s chirality is reflected in the
diastereotopic geminal hydrogen atoms of the encapsulated
guest, but how is the chirality in the middle of the capsule
transmitted to the aromatic subunits near the ends? The
answer lies in the hydrogen-bonding patterns of the spacers
with the imide walls. A carbonyl group of each imide wall is
left without a hydrogen bond donor. This asymmetric ar-
rangement causes a twist in the array of the imide walls that
changes the symmetry of the cavitands from C4v to C4, and
creates a chiral space, as shown in Figure 3. This twist is most
pronounced at the narrower middle of the extended capsule
to relieve the steric clashes of the unpaired carbonyl groups
with the adjacent imide panels.
Why do the geminal hydrogen signals of C2 and C4 in the
aliphatic guest show pronounced diastereotopic signals? The
methyl group of the alkane is positioned in the center of the
available for water, where the burial of hydrophobic, sol-
vent-accessible surface33 is ∼25 cal/mol for each A .) The
2
˚
€
(28) Lutzen, A.; Renslo, A. R.; Schalley, C. A.; O’Leary, B. M.; Rebek, J.,
Jr. J. Am. Chem. Soc. 1999, 121, 7455–7456.
guest molecules in mesitylene have the same C-H/π and
(29) Amaya, T.; Rebek, J., Jr. J. Am. Chem. Soc. 2004, 126, 6216–6217.
(30) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Hommes, N.
J. R. v. H. J. Am. Chem. Soc. 1996, 118, 6317–6318.
(31) Frisher, M. J. et al. Gaussian 03; Guassian, Inc., Pittsburgh, PA,
2002; see the Supporting Information for the full reference).
(32) Kellis, J. T., Jr.; Nyberg, K.; Sali, D.; Fersht, A. R. Nature 1988, 333,
784–786.
(33) Raschke, T. M.; Tsai, J.; Levitt, M. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 5965–5969.
J. Org. Chem. Vol. 74, No. 17, 2009 6587