Do deuteriums form stronger CH-π interactions?

Article abstract of DOI:10.1021/ja305788p

The D/H isotope effect for the CH-π interaction was studied experimentally and computationally. First, a series of molecular balances that are very sensitive to changes in the strength of the CH-π interactions in solution were designed. Balances with deuterated and non-deuterated alkyl groups were synthesized, and their intramolecular CH-π interactions were compared. The geometries of their intramolecular CH-π and CD-π interactions were characterized in the solid state by X-ray analysis, and the strength of each interaction was characterized in solution by the folded/unfolded ratio as measured by ¹H NMR spectra. Second, the relative strengths of the CH-π and CD-π interactions were also assessed computationally using dispersion-corrected DFT (PDE-D2/6-31+G*). No significant differencee was observed in either the experimental or theoretical studies, indicating that the D/H isotope effect for the CH-π interaction is either very small or nonexistent.

Full text of DOI:10.1021/ja305788p

Communication
pubs.acs.org/JACS
Do Deuteriums Form Stronger CH−π Interactions?
Chen Zhao,† Robert M. Parrish,‡ Mark D. Smith,† Perry J. Pellechia,† C. David Sherrill,‡
and Ken D. Shimizu*,†
†Department of Chemistry and Biochemistry, University of South Carolina, Columbia, South Carolina 29208, United States
‡School of Chemistry and Biochemistry and School of Computational Science and Engineering, Georgia Institute of Technology,
Atlanta, Georgia 30332, United States
S
* Supporting Information
ABSTRACT: The D/H isotope effect for the CH−π
interaction was studied experimentally and computation-
ally. First, a series of molecular balances that are very
sensitive to changes in the strength of the CH−π
interactions in solution were designed. Balances with
deuterated and non-deuterated alkyl groups were synthe-
sized, and their intramolecular CH−π interactions were
compared. The geometries of their intramolecular CH−π
and CD−π interactions were characterized in the solid
state by X-ray analysis, and the strength of each interaction
was characterized in solution by the folded/unfolded ratio
1
as measured by H NMR spectra. Second, the relative
strengths of the CH−π and CD−π interactions were also
assessed computationally using dispersion-corrected DFT
(PDE-D2/6-31+G*). No significant differencee was
observed in either the experimental or theoretical studies,
indicating that the D/H isotope effect for the CH−π
interaction is either very small or nonexistent.
Figure 1. (a) Scheme showing the unfolded ⇄ folded conformational
equilibrium of the molecular balances, which can be used to measure
changes in the strength of the intramolecular CH−π interaction in the
folded conformer. (b) Folded conformers of balances 1−3 designed to
form intramolecular CH−π interactions and the control balance 4.
environments of molecular capsules, which are very sensitive to
small differences in molecular volume. Thus, the observed
enhancements in the stability of deuterated guests could be due
to their reduced steric interactions arising from their shorter C−
D bonds, as opposed to stronger attractive CD−π interactions.
Therefore, the goal of this study was to study the D/H isotope
effect for CH−π interactions within less constrained environ-
ments in which steric interactions were minimized. First, an
experimental study was carried out using small-molecule model
systems. Second, a computational approach was carried out
applying density functional theory (DFT) to a methane−
benzene system. Both approaches found only minor differences
between CD−π and CH−π interactions.
First, differences in CH−π and CD−π interactions were
experimentally studied using molecular balances 1−3 (Figure 1).
The balances form intramolecular CH−π interactions within
relatively open environments with a minimum of steric
interactions. Therefore, these model systems are less susceptible
to repulsive interactions that could mask and attenuate the
CH−π interactions of interest. The strengths of these
interactions can be measured by following the folded/unfolded
equilibrium ratios. Because of restricted rotation around the
Caryl−Nimide single bond, each balance adopts distinct folded and
unfolded conformers. In the folded conformers, the o-methyl
H−π interactions are attractive non-covalent interactions
Cbetween the protons of an alkyl or aryl group and the π face
of an aromatic ring.1 Despite their weak nature, CH−π
interactions play crucial roles in supramolecular chemistry,2
asymmetric catalysis,3 and the folding of small molecules4 and
proteins.5 However, the study of CH−π interactions has been
difficult because of their weak strengths (0.5−2.5 kcal/mol)3 and
highly variable geometries. A potentially powerful method for
studying the interaction involves the use of D/H isotope effects,
which have been successfully applied to the study of other non-
covalent interactions.6 The presence of a pronounced D/H
isotope effect for the CH−π interactions could be used to verify
their formation and to probe their stability trends. The enhanced
CH−π interactions of deuterated molecules could also be used to
design better pharmaceuticals and asymmetric catalysts.
However, whether H and D form CH−π interactions with
different strengths remains unclear. Several studies have
observed significant deuterium isotope effects: Rebek and co-
workers7 and Iwata and co-workers8 found that deuterated
species formed stronger interactions within different molecular
capsules, and differences in the retention times of protic and
deuterated species were observed in chromatographic studies.9
Other studies have found little or no D/H isotope effect for the
CH−π interaction.10 A possible reason for these discrepancies is
that many of these studies were carried out within the confined
Received: June 14, 2012
Published: August 17, 2012
© 2012 American Chemical Society
14306
dx.doi.org/10.1021/ja305788p | J. Am. Chem. Soc. 2012, 134, 14306−14309

Journal of the American Chemical Society
Communication
Figure 3. Distance d and angles θ and α used to characterize 1−4.
Table 1. Values of d, θ, and α from the Crystal Structures of
Balances 1a−3a and the Folding Energies in Solution (ΔGfold
)
for Protic and Deuterated Balances 1−4 in CDCl3 at 25 °C
θ
α
ΔGfold,H
ΔGfold,D
ΔΔGfold
b
balance d (Å) (deg) (deg) (kcal/mol) (kcal/mol) (kcal/mol)
1
2
3
4
2.68
2.69
2.61
−
58
52
58
−
5
21
4
−0.10
0.84
−0.13
0.81
+0.03
+0.03
0.00
a
0.07
0.07
2
0.02
0.06
−0.04
a
d, θ, and α for 2a are averages of the values for the three unique
folded conformers in the unit cell.16 ΔΔGfold = ΔGfold,H − ΔGfold,D
.
b
the Supporting Information (SI)] and compared to that of its
protic counterpart 1a. The structures were nearly identical.
Although balances 1a−3a all formed intramolecular CH−π
interactions, the number (one hydrogen vs two), geometry, and
distance in these interactions varied considerably. The structural
parameters d, θ, and α used to compare the balances are shown in
Figure 3, and a comparison of the measurements from the crystal
structures of the balances are shown in Table 1. The “hinge”
angle θ defined by the succinimide and arene planes provides a
measure of how closely the o-methyl group is held against the
arene shelf. For example, balance 2a has the smallest θ, fixing the
o-methyl tightly against the arene shelf. This strain is evident
from the fact that the N-aryl group is pushed upward out of the
succinimide plane (α = +21°). In contrast, balances 1a and 3a
have larger θ values, positioning their o-methyl groups at more
optimal distances with less strain (α = +5 and +4°, respectively).
Next, the strengths of the CH−π interactions in balances 1−3
Figure 2. X-ray structures of the folded conformers of (a) 1a,14c (b) 2a,
(c) 3a,14a and (d) 4a. The bridgehead phenyl groups in 1a and 2a are
partially hidden for clarity. The unfolded conformers (not shown) were
also present in the crystal structures of 2a, 3a, and 4a.
groups (R) are perfectly positioned over the aromatic shelves to
form intramolecular CH−π interactions. In the unfolded
conformers, the o-methyl groups are held away from the
aromatic shelves and cannot form intramolecular CH−π
interactions. The two conformations are in slow exchange at
1
room temperature on the H NMR time scale. Therefore, the
folded/unfolded ratio can be accurately measured from the
integrations of the respective peaks.
Balances 1−3 provide a range of different CH−π interaction
geometries and environments, affording a comprehensive study
of the interaction. For example, balance 1 has a large
phenanthrene aromatic shelf, whereas balances 2 and 3 have
smaller benzene shelves. The geometry and steric interactions of
the o-methyl group are attenuated by subtle differences in the
bicyclic framework. Specifically, the different bridges −CO−,
−O−, and −(o-C6H4)− (Z in Figure 1a) on the back sides of the
balances attenuate the distance and steric interactions between
the methyl group and the aromatic shelf.11 Balance 4, which
cannot form a CH−π interaction because of the absence of an
aromatic shelf, was used as a control.
Balances 1−4 were synthesized via similar modular routes
allowing protic (1a−4a) and deuterated (1b−4b) forms to be
prepared.12 Protic balances 1a and 3a had been previously
described in the literature, and 3a had been used to study CH−π
interactions.13,14 The other six balances were new structures.
Initially, the structures of balances 1a−4a were verified and
characterized by X-ray structure analyses (Figure 2).15 The
crystal structures confirmed the formation of the endo bicyclic
framework. As expected, well-defined intramolecular CH−π
interactions were observed in the folded conformers of balances
1a−3a (Figure 2a−c).16 The methyl proton-to-arene plane
distances (d) in 1a−3a were all within the typical range for
CH−π interactions (2.6−3.0 Å).17 X-ray structure analysis also
confirmed the absence of an intramolecular CH−π interaction in
the folded conformation of control balance 4a (Figure 2d). The
X-ray structure of deuterated balance 1b was also examined [see
1
were measured in solution by H NMR spectroscopy. In each
case, separate peaks for the folded and unfolded conformers were
observed at room temperature, enabling facile measurement of
the folded/unfolded ratio. In particular, large upfield shifts were
observed for the o-methyl groups in the folded conformers,
consistent with the formation of CH−π interactions. The folded
methyl singlets of 1a−3a were shifted upfield by −2.08, −1.01,
and −1.04 ppm, respectively, compared with the peaks for the
unfolded methyl groups. By comparison, control balance 4a, in
which a CH−π interaction cannot form, had almost identical
chemical shifts for the folded and unfolded methyl protons (Δδ =
1
−0.03 ppm). The H NMR spectra of the deuterated balances
were identical to those of their protic counterparts except for the
absence of the deuterated o-methyl peaks.
Comparison of the folded/unfolded ratios for the protic
balances showed different strengths for their CH−π interactions
(Figure 4). These ratios were measured from the integration of
the singlets for the succinimide protons in the 1H NMR
spectra.19 As expected, control 4 had a folded/unfolded ratio of
nearly 1:1, suggesting that differences in dipole moment and
solvation of the conformers did not bias the folded/unfolded
ratios. Despite the presence of intramolecular CH−π inter-
actions in 1−3, only 1 displayed a preference for the folded
conformer. We attribute this to destabilizing repulsive steric
interactions that arise because the rigid bicyclic framework
positions the methyl group slightly too close to the arene shelf. As
14307
dx.doi.org/10.1021/ja305788p | J. Am. Chem. Soc. 2012, 134, 14306−14309

Journal of the American Chemical Society
Communication
for the CH−π interaction or that the effect was too small to be
measured accurately using our model systems. Another possible
explanation would be that the isotope effects for the attractive
CH−π and repulsive steric interactions perfectly canceled out in
all three balances. However, this possibility was deemed unlikely.
First, the attractive and repulsive isotope effects would have to
balance perfectly for all three model systems, despite the
differences in their geometries and conformational constraints.
Second, the repulsive steric interactions in balances 1−3 are very
small (<1.0 kcal/mol) and do not change significantly with small
differences in the lengths of the C−D and C−H bonds.20
To investigate further the isotope effect for the CH−π
interaction, computational studies were carried out using a
simple model system of methane interacting with benzene
(Figure 5). Under the Born−Oppenheimer approximation,
CH−π and CD−π interactions take place on the same electronic
potential energy surface. The primary difference should come
from the fact that a C−D stretch has a significantly lower
vibrational frequency than a C−H stretch; this would be reflected
as a difference in the corrections added to the electronic energies
to obtain the enthalpies. Secondarily, one could also consider
shifts due to the slight difference in the C−H and C−D bond
lengths. We investigated the order of magnitude of both possible
effects using DFT with the Perdew−Becke−Ernzerhof (PBE)
exchange−correlation functional.21 Because standard DFT does
not account for long-range dispersion interactions that are
important in intermolecular interactions, we added Grimme’s D2
empirical pairwise dispersion correction with the parameters
recommended for PBE;22 this approach is denoted PBE-D2.
Pople’s standard 6-31+G* basis set was used.
Figure 4. Folded/unfolded ratios for 1−4 in CDCl3 at 25 °C measured
by integration of the 1H NMR spectra with a 5% integration error.18
Table 2. Calculated ΔGfold Values (in kcal/mol) for Protic and
Deuterated Balances 1−3 in CDCl3 and Acetone-d6 at 25 °C
CDCl3
acetone-d6
ΔGfold,H
ΔGfold,D
ΔGfold,H
ΔGfold,D
1
2
3
−0.12 0.05
0.83 0.28
0.07 0.14
−0.14 0.11
0.80 0.19
0.08 0.38
−0.23 0.16
0.66 0.07
−0.26 0.37
0.65 0.07
−0.05 0.26
−0.06 0.10
predicted from the crystal structures, the repulsive interaction
was most evident in 2a, which also had the lowest folded/
unfolded ratio. The repulsive interactions complicate the
measurement of the absolute strengths of the CH−π interactions
but do not diminish the utility of 1−3 in measuring the isotope
effects for the CH−π interactions.
The geometry of the interaction model was optimized at the
PBE-D2/6-31+G* level. The C3v-symmetric model chosen was
not a potential energy minimum (the energy was lower when the
CH4 was tilted so one of the other hydrogens could also interact
with the benzene ring), but the associated small imaginary
vibrational frequencies should not significantly impact our
exploration of isotope effects (an alternative structure in which
the methane C−H bonds lie between benzene C−C bonds was
essentially isoenergetic and featured analogous small imaginary
frequencies). Replacing the interacting H of CH4 with D in the
complex led to an increase in binding energy of only 0.02 kcal/
mol due to mass-dependent changes in the 298 K enthalpy
correction (replacing all of the H’s in CH4 with D’s decreased the
binding energy by 0.01 kcal/mol).
A computational study by Houk and co-workers used this
same model system to estimate isotope effects in p-xylene-d3 as a
guest (along with CCl4 as a coguest) in a dimeric capsule
complex;7b they found a considerably larger effect due to
vibrational energy contributions. One major difference, however,
is that the host capsule in that study presented three aromatic
faces to interact with one methyl group of p-xylene, thus tripling
the size of the isotope effect in their additive estimates. Another is
that their MP2/6-311++G(d,p) level of theory predicted 3 times
the zero-point vibrational energy stabilization of CH3D/C6H6
versus CH4/C6H6. Our binding energy for this complex, 1.78
kcal/mol, is closer to coupled-cluster complete-basis-set
estimates of 1.45 kcal/mol23 than their value of 0.85 kcal/mol,
suggesting that our value for this quantity is likely to be more
reliable. Finally, our results also include finite-temperature effects
on the enthalpy that were neglected in the previous study, which
further reduce the computed isotope effect.
Differences in the strengths of the intramolecular CH−π and
CD−π interactions were assessed by comparison of the folded/
unfolded ratios and the corresponding folding energies (Table
1). The folding energies for protic and deuterated balances 1−3
were almost identical. The differences (ΔΔGfold) were very small
and within the error of the analysis ( 0.03 kcal/mol), which was
calculated on the basis of a conservative estimate of 5% for the
1H NMR integration error.18 The folding energies in acetone-d6
were also compared (see the SI). Again, nearly identical folding
energies were observed, with even smaller errors.
To confirm the above single-point measurements, more
comprehensive multipoint van’t Hoff analyses were carried out.
The folded/unfolded ratios for balances 1−3 were measured
over a range of temperatures (25−55 °C) in CDCl3 and acetone-
d6, and the ΔGfold values were calculated from the measured ΔH
and ΔS values (Table 2). This study led to the same conclusion
that the small differences in the ΔGfold values for the protic and
deuterated balances were well within the error of the analysis.
The above experimental studies found only small differences in
the strengths of the CH−π and CD−π interactions that were
smaller than the experimental error of the analyses. Therefore,
we concluded that either there was no deuterium isotope effect
Another possible source of D/H isotope effects for CH−π
interactions is a steric effect due to the fact that the vibrationally
Figure 5. Model benzene−methane system used in the computational
studies of the D/H isotope effect for the CH−π interaction.
14308
dx.doi.org/10.1021/ja305788p | J. Am. Chem. Soc. 2012, 134, 14306−14309

Journal of the American Chemical Society
Communication
Honda, K.; Nishio, M. Tetrahedron 2000, 56, 6185. (c) Tewari, A. K.;
Dubey, R. Bioorg. Med. Chem. 2008, 16, 126. (d) Ozawa, T.; Okazaki, K.;
Kitaura, K. J. Comput. Chem. 2011, 32, 2774. (e) Ozawa, T.; Okazaki, K.;
Kitaura, K. Bioorg. Med. Chem. 2011, 19, 5231.
averaged bond length of C−H is longer than that of C−D by
∼0.005 Å. To obtain a rough estimate the size of this steric effect
on the CH−π interactions, we computed interaction energies for
our methane−benzene model while stretching or compressing
the interacting methane C−H bond (Δr = 0.005 Å). The
distance R between the plane of the benzene and the methyl
carbon was set to 3.493 Å (the closest approach for 1a) and
varied by 0.2 Å. The results are plotted in Figure S9 in the SI.
The interaction between methane and benzene became more
favorable (ΔΔEint < 0) as the C−H bond was compressed
(mimicking C−D bonds). The effect was negligible (∼0.01 kcal/
mol or less) for R ≥ 3.493 Å, but it grew as the benzene came
closer to the methane (ΔR < 0), becoming as large as 0.04 kcal/
mol for ΔR = −0.2 Å and Δr = −0.005 Å. Hence, this “steric
isotope effect” is negligible for the present systems but may
become significant in other systems with closer CH−π contacts.
The steric origin of this effect was confirmed using our symmetry-
adapted perturbation theory (SAPT) program (Figure S27).24
In conclusion, D/H isotope effects were not observed for the
CH−π interaction within three very sensitive small-molecule
model systems containing different CH−π interaction geo-
metries and environments. The experimental results were
corroborated by theoretical calculations that compared the
interaction energies of methane and benzene. The experimental
and theoretical systems in this study were designed to minimize
steric interactions. Thus, previous reports of isotope effects were
probably due to other factors (e.g., the different sizes of CH3 and
CD3 groups placed within more confined environments) rather
than attenuation of the CH−π interaction.25 This steric
hypothesis was supported by the calculations, which showed
that differences in energy arose when the interacting groups were
brought closer than the optimal CH−π interaction distance.
While the lack of an isotope effect eliminates the possibility of
using deuteration to enhance the CH−π interaction, it validates
the use of deuteration for spectroscopic and labeling purposes, as
this introduces a minimal perturbation of the system.26
(3) Nishio, M. Tetrahedron 2005, 61, 6923.
(4) (a) Umezawa, Y.; Tsuboyama, S.; Takahashi, H.; Uzawa, J.; Nishio,
M. Bioorg. Med. Chem. 1999, 7, 2021. (b) Umezawa, Y.; Tsuboyama, S.;
Takahashi, H.; Uzawa, J.; Nishio, M. Tetrahedron 1999, 55, 10047.
(c) Takahashi, O.; Kohno, Y.; Nishio, M. Chem. Rev. 2010, 110, 6049.
(5) (a) Brandl, M.; Weiss, M. S.; Jabs, A.; Suhnel, J.; Hilgenfeld, R. J.
Mol. Biol. 2001, 307, 357. (b) Fernandez, M. D.; Canada, F. J.; Jimenez-
Barbero, J.; Cuevas, G. J. Am. Chem. Soc. 2005, 127, 7379.
(6) Wade, D. Chem.-Biol. Interact. 1999, 117, 191.
(7) (a) Rechavi, D.; Scarso, A.; Rebek, J., Jr. J. Am. Chem. Soc. 2004,
126, 7738. (b) Zhao, Y. L.; Houk, K. N.; Rechavi, D.; Scarso, A.; Rebek,
J., Jr. J. Am. Chem. Soc. 2004, 126, 11428.
(8) Haino, T.; Fukuta, K.; Iwamoto, H.; Iwata, S. Chem.Eur. J. 2009,
15, 13286.
(9) (a) Turowski, M.; Yamakawa, N.; Meller, J.; Kimata, K.; Ikegami,
T.; Hosoya, K.; Tanaka, N.; Thornton, E. R. J. Am. Chem. Soc. 2003, 125,
13836. (b) Iyer, S. S.; Zhang, Z. P.; Kellogg, G. E.; Karnes, H. T. J.
Chromatogr. Sci. 2004, 42, 383.
(10) (a) Liu, Y.; Warmuth, R. Org. Lett. 2007, 9, 2883. (b) Mugridge, J.
S.; Bergman, R. G.; Raymond, K. N. J. Am. Chem. Soc. 2010, 132, 1182.
(11) Bhayana, B.; Ams, M. R. J. Org. Chem. 2011, 76, 3594.
(12) (a) Carroll, W. R.; Pellechia, P.; Shimizu, K. D. Org. Lett. 2008, 10,
3547. (b) Carroll, W. R.; Zhao, C.; Smith, M. D.; Pellechia, P. J.;
Shimizu, K. D. Org. Lett. 2011, 13, 4320.
(13) (a) Eto, M.; Setoguchi, K.; Harada, A.; Sugiyama, E.; Harano, K.
Tetrahedron Lett. 1998, 39, 9751. (b) Yoshitake, Y.; Misaka, J.;
Setoguchi, K.; Abe, M.; Kawaji, T.; Eto, M.; Harano, K. J. Chem. Soc.,
Perkin Trans. 2 2002, 1611.
(14) (a) Grossmann, G.; Potrzebowski, M. J.; Olejniczak, S.;
Ziolkowska, N. E.; Bujacz, G. D.; Ciesielski, W.; Prezdo, W.; Nazarov,
V.; Golovko, V. New J. Chem. 2003, 27, 1095. (b) Srivastav, K. K.;
Verma, A. K.; Verma, S. M. Indian J. Chem., Sect. B 1993, 32, 1143.
(c) Kishikawa, K.; Yoshizaki, K.; Kohmoto, S.; Yamamoto, M.;
Yamaguchi, K.; Yamadab, K. J. Chem. Soc., Perkin Trans. 1 1997, 1233.
(15) The crystal structures of 1a and 3a were reported previously.14a,c
(16) The crystal structure of 2a contained three different folded
conformers that formed intramolecular CH−π interactions with slightly
different geometries and distances; only one is shown in Figure 1b.
(17) (a) Hobza, P.; Havlas, Z. Chem. Rev. 2000, 100, 4253. (b) Nishio,
M. CrystEngComm 2004, 6, 130.
(18) (a) Bauer, M.; Bertario, A.; Boccardi, G.; Fontaine, X.; Rao, R.;
Verrier, D. J. Pharm. Biomed. Anal. 1998, 17, 419. (b) Rizzo, V.; Pinciroli,
V. J. Pharm. Biomed. Anal. 2005, 38, 851. (c) Soininen, P.; Haarala, J.;
Vepsalainen, J.; Niemitz, M.; Laatikainen, R. Anal. Chim. Acta 2005, 542,
178.
(19) Integration of o-methyl groups gave similar folded/unfolded
ratios. However, o-methyl peak integrations could not be used for
comparisons because this peak is absent in the deuterated balances.
(20) A third reason that this explanation is unlikely is that it requires
the attractive CD−π interaction to be weaker than the CH−π
interaction. However, all reports that have observed deuterium isotope
effects for the CH−π interaction have found the opposite trend.
(21) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
3865.
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Experimental details, H and 13C NMR spectra, X-ray data
(CIF), and van’t Hoff plots for balances 1−4. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
shimizu@mail.chem.sc.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the National Science Foundation
(CHE 0911616 and CHE 1011360).
■
(22) Grimme, S. J. Comput. Chem. 2006, 27, 1787.
(23) Ringer, A. L.; Figgs, M. S.; Sinnokrot, M. O.; Sherrill, C. D. J. Phys.
Chem. A 2006, 110, 10822−10828.
REFERENCES
■
(1) (a) Nishio, M.; Hirota, M.; Umezawa, Y. The CH/π Interaction.
Evidence, Nature, and Consequences; Wiley-VCH: New York, 1998.
(b) Nishio, M.; Hirota, M. Tetrahedron 1989, 45, 7201. (c) Nishio, M.;
Umezawa, Y.; Honda, K.; Tsuboyama, S.; Suezawa, H. CrystEngComm
2009, 11, 1757. (d) Nishio, M. Phys. Chem. Chem. Phys. 2011, 13, 13873.
(e) Salonen, L. M.; Ellermann, M.; Diederich, F. Angew. Chem., Int. Ed.
2011, 50, 4808.
(24) Hohenstein, E. G.; Parrish, R. M.; Sherrill, C. D.; Turney, J. M.;
Schaefer, H. F., III. J. Chem. Phys. 2011, 135, No. 174107.
(25) (a) Turowski, M.; Yamakawa, N.; Meller, J.; Kimata, K.; Ikegami,
T.; Hosoya, K.; Tanaka, N.; Thornton, E. R. J. Am. Chem. Soc. 2003, 125,
13836. (b) Iyer, S. S.; Zhang, Z. P.; Kellogg, G. E.; Karnes, H. T. J.
Chromatogr. Sci. 2004, 42, 383.
(26) Ruschak, A. M.; Kay, L. E. J. Biomol. NMR 2010, 46, 75.
(2) (a) Nishio, M.; Umezawa, Y.; Hirota, M.; Takeuchi, Y. Tetrahedron
1995, 51, 8665. (b) Takahashi, H.; Tsuboyama, S.; Umezawa, Y.;
14309
dx.doi.org/10.1021/ja305788p | J. Am. Chem. Soc. 2012, 134, 14306−14309