Quantification of CH...π interactions: Implications on how substituent effects influence aromatic interactions

Article abstract of DOI:10.1002/chem.201001362

Attractive interactions between a substituted benzene ring and an α-substituted acetate group were determined experimentally by using the triptycene model system. The attractive interaction correlates well with the Hammett constants σ_m (R²=0.90), but correlates much better with the acidity of the α-protons (R²=0.98). A predominant CH...π interaction was found to control the conformational preference of model compounds 1a-g. Despite the predominance of the CH...π interaction in compounds 1a-g, a Hammett plot displays a fairly straight line for the substituent effect. These results show that when using Hammett plots in a simplified model system, a system designed to study the effect of X...π interactions could capture the X-H...π interaction instead.

Full text of DOI:10.1002/chem.201001362

FULL PAPER
DOI: 10.1002/chem.201001362
Quantification of CH···p Interactions: Implications on How Substituent
Effects Influence Aromatic Interactions
Benjamin W. Gung,*[a] Bright U. Emenike,[a] Michael Lewis,[b] and Kristin Kirschbaum[c]
Abstract: Attractive interactions between a substituted benzene ring and an a-sub-
stituted acetate group were determined experimentally by using the triptycene
Keywords: aromatic interactions ·
computational chemistry · Hammett
plots · substituent effects · tripty-
cenes
model system. The attractive interaction correlates well with the Hammett con-
stants sm (R2 =0.90), but correlates much better with the acidity of the a-protons
(R2 =0.98).
Introduction
most to least stable in one of the more recent studies are
calculated to be 1) parallel displaced (ꢀ2.74 kcalmolꢀ1),
2) edge-to-face (ꢀ2.39 kcalmolꢀ1), and 3) face-to-face
stacked (ꢀ1.99 kcalmolꢀ1) (Figure 1).[11]
Noncovalent interactions play an increasingly important role
in modern chemical research.[1–3] Understanding the mecha-
nisms involved in noncovalent interactions is important for
the design of new supramolecular systems and biomolecular
active agents.[4] Theoretical models of noncovalent interac-
tions have been constantly evolving and in general the ap-
proach has been to partition intermolecular interactions into
several fundamental physical forces.[5,6] A scheme by Hobza
and Mꢀller-Dethlefs partitioned noncovalent interaction
energy into five components: 1) electrostatic, 2) charge-
transfer, 3) dispersion, 4) ion-mediated, and 5) hydrophobic
interactions.[6] Compared with Morokumaꢁs original propos-
al,[5] electrostatic and charge-transfer interactions remained
as fundamental forces in the scheme, while dispersion was
added as an explicit term.[6]
Figure 1. The orientations of interacting benzene dimer: a) parallel dis-
placed, b) edge to face, and c) face to face.[11]
The magnitudes of individual nonbonded interactions in-
volving aromatic rings are provided mainly from high-level
computational studies.[7–14] For the simple system of benzene
dimers in the gas phase, the relative configurations from
The first theoretical construct that specifically deals with
aromatic interactions was proposed in 1990 by Hunter and
Sanders.[15] This model is based on electrostatic interactions
that describe the aromatic ring as a positively charged s
framework sandwiched between two regions of negatively
charged p-electron density.[15] The hydrogen atoms carry
partial positive charges and the p orbital carries partial neg-
ative charge. This model successfully explains the preferred
arrangements between aromatic rings with a direct and ef-
fective electrostatic argument. Experimental studies using
the 1,8-diarylnaphthalene system by Cozzi et al. support the
theoretical construct of the Hunter and Sanders model.[16,17]
Hobza and Mꢀller-Dethlefs recently pointed out that it is
enough to recognize the role of molecular quadrupoles in
aromatic molecules for electrostatic arguments.[6] Although
the simple electrostatic model for aromatic interactions has
[a] Prof. Dr. B. W. Gung, B. U. Emenike
Department of Chemistry and Biochemistry, Miami University
Oxford, Ohio 45056 (USA)
Fax : (+1)513-529-5715
E-mail: gungbw@muohio.edu
[b] Prof. Dr. M. Lewis
Department of Chemistry, Saint Louis University
Saint Louis, MO 63103 (USA)
[c] Prof. Dr. K. Kirschbaum
Department of Chemistry, University of Toledo
Toledo, Ohio 43606 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001362.
Chem. Eur. J. 2010, 16, 12357 – 12362
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12357

enjoyed acceptance for more than a decade, recent compu-
tational studies have raised the importance of dispersion
forces and indicated that simple electrostatic models do not
seem capable of explaining the energetic ordering of aro-
matic dimers with different substituents.[18] In general, Die-
derich et al. reported that experimental studies have empha-
sized electrostatic forces, whereas computational studies
have emphasized dispersion forces.[2]
place a partial positive charge on the hydrogen in the molec-
ꢀ
ular fragment X H. Thus, the results from this model study
ꢀ
appear to represent more of an X H···p interaction, rather
than an X···p interaction.
Results and Discussion
Along this line of debate, Wheeler and Houk have recent-
ly challenged the conventional concept of substituent effects
in aromatic interactions based on a simple computational
model.[19,20] It was suggested that substituents influence aro-
matic stacking interactions by interacting directly with an-
other aromatic ring, rather than through the polarization of
the arene. The computational results for substituent effects
gave similar Hammett plots for a benzene dimer model and
a simplified model system, despite the fact that the simple
model system was devoid of an aromatic ring (Scheme 1).
The results led to the conclusion that the substituent inter-
acts directly with the other aromatic ring as indicated by the
dashed lines between the substituent and the benzene ring
in Scheme 1.[19]
The 1,9-disubstituted triptycene system has been demon-
strated to be a valuable tool in our examination of arene–
arene interactions, 1a–g (Scheme 2).[24–26] The slow rotation
ꢀ
of the C9 benzyl group around the Csp3 Csp3 bond gives rise
to syn and anti conformations, which can be studied by
1H NMR spectroscopy (Scheme 3 and the Supporting Infor-
mation). A statistical 2:1 syn/anti ratio is expected when
there is no interaction between the C1 and the C9 groups.
Because the triptycene scaffold provides an otherwise iden-
tical environment for the syn and anti conformations, any
deviation from the statistical 2:1 syn/anti ratio indicates the
Scheme 1. Theoretical models used by Wheeler and Houk. a) Sandwich
benzene dimer with one benzene ring with a substituent; b) the model
that replaces the aromatic ring with a hydrogen atom. Both systems pro-
duced similar Hammett plots.
Scheme 2. a) The triptycene model 1a–g have been established in the
study of p–p interactions; b), c) the new models 2a–g and 3a–g were syn-
thesized for studying direct interactions between the substituent X and
the aromatic ring. Piv=pivaloyl; X=Me (a), OMe (b), H (c), Cl (d), Br
(e), I (f), CN (g); Y=Me (a), H (b), F (c), Cl (d), Br (e), CN (f), NO2
(g).
The traditional concept of p–p interactions tuned by sub-
stituents through electron withdrawal or donation to the ar-
omatic rings was suggested to be flawed. Previously, Rash-
kin and Waters also described an unexpected direct interac-
tion between substituents and the other aromatic ring.[21]
However, they gave no generalized extrapolation to other
systems. In a more recent theoretical study, Ringer and
Sherrill have reported that simple models that account for
only one type of interaction cannot capture the complexity
of substituent effects.[22] The understanding of how substitu-
ents influence p–p interactions is important in drug design
and in related fields.[2] We have prepared model systems at-
tempting to reproduce the direct interactions between sub-
stituents and the other aromatic ring in solution. Our initial
experimental results related to the direct interaction model
were reported recently although no clear conclusion was
reached.[23] Our further experimental results with model
compounds 2a–g and 3a–g show a surprisingly linear corre-
lation between calculated a-CH deprotonation energy and
the syn/anti ratios. This correlation indicates a predominant
CH···p interaction, rather than an X···p interaction as indi-
cated in Scheme 1.[19] Further examination of the simplified
model suggests that a dominant molecular dipole should
Scheme 3. a) Sketches to show the equilibrium between the syn and anti
conformations. b) Newman projections at C9 show the diastereotopic
protons in the syn conformation and the enantiotopic protons in the anti
conformation, respectively.
12358
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12357 – 12362

Quantification of CH···p Interactions
FULL PAPER
degree of preference for the interactions between the C1
and the C9 groups. In this study, we replaced the C1 ben-
zoate in previously used model system 1 with a-substituted
acetates and correlated the syn/anti ratios obtained from the
new models (2a–g, in which the substituent X is on the a-
substituted acetate group, and 3a–g, in which the substituent
Y is on the C9 aromatic ring; Scheme 2) with Hammett con-
stants sm and with the calculated deprotonation energy.
From the optimized conformations generated by Macro-
Model using the force field MMFFs, the distance between
the substituent X and the C9 aromatic ring is 4.1 ꢃ for the
new model (2d, X=Cl; see Figure 2b) and 5.6 ꢃ for the
previous model (1d, not shown). Therefore, direct interac-
tions between the substituent X and the C9 aromatic ring
should be better reflected in the new models (2a–g).
Scheme 4. Synthesis of the model compounds.
1
Figure 3. Temperature-dependent H NMR (300 MHz) spectra for the C9
benzyl and CH2Cl protons of model compound 2d.
the benzyl CH2 group appears centered at around d=
4.6 ppm and the CH2Cl group of compound 2d appears at
1
around d=3.3 ppm. The corresponding H NMR signals for
the anti conformation appear at around d=4.6 ppm as two
small singlets, one from the benzyl and the other from the
CH2Cl group. This difference in chemical shifts for the
CH2Cl group between the syn and the anti conformation is
an indication that the syn conformation places the CH2Cl
group in the “shielded” zone of the C9 aromatic ring, there-
fore, consistent with the proposed conformations.
A crystal suitable for X-ray structure analysis was ob-
tained for model compound 3e (Figure 4). The syn confor-
mation is also preferred in the solid state and is similar to
Figure 2. syn and anti conformations found by computer modeling (Mac-
roModel 9.0, MMFFs force field) for compound 2d.
The new series of triptycene compounds (2a–g and 3a–g)
were synthesized starting from the intermediates 4a–g, pre-
viously reported for the synthesis of model compounds 1a–
g.[26] Commercially available acid chlorides were used when-
ever possible. In situ generation of acid chloride was em-
ployed when the corresponding acid was available
(Scheme 4).
ꢀ
the optimized structures by MacroModel (Figure 2). The C
Br bond eclipses the carbonyl C=O bond, as shown in the
X-ray structure. This arrangement allows the Br substituent
to locate in the near van der Waals distance to assume a
direct interaction with the aromatic ring of the benzyl
The syn conformation shows two sets of AB quartets in
1
the H NMR spectra at temperatures below ꢀ208C, which is
illustrated in Figure 3 using compound 2d. The signal for
Chem. Eur. J. 2010, 16, 12357 – 12362
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12359

B. W. Gung et al.
the syn conformation. These observations are easy to ex-
ꢀ
plain with an X H···p interaction, namely, there is less at-
traction when the aromatic ring is electron deficient. But it
is difficult to rationalize with an X···p interaction. Because
electron-withdrawing groups reduce the p-electron density
of the aromatic ring in 3 and increase the acidity of the a-
CH in 2, the observed trend for the two series of model
compounds suggests a predominant CH···p interaction.[27]
One way to verify this hypothesis is to correlate the inter-
action free energy with the acidity of the a-CH protons in
compounds 2a–g. We were not able to find any reported
data on the acidity of a-substituted acetates and decided to
carry out a set of ab initio calculations.[28] The deprotonation
energy values for model compounds 2a–g were obtained by
Figure 4. X-ray structure of compound 3e. The syn conformation is also
preferred in the solid state with a parallel stacking arrangement between
the C9 arene and the C1 bromoacetate group. The nearest a-CH···arene
distance is around 3.0 ꢃ and a-Br···arene is around 4.0 ꢃ.
calculations at the MP2ACTHNUGRTENUNG(full)/6-311++G** level (except the
iodo atom, which was calculated by using MIDIX) with
group. Additionally, the a-CH proton is also close enough
to interact with the aromatic ring.
Equation (1).[29]
DE ¼ E ðesterÞꢀE ðenolateÞ
ð1Þ
The experimentally observed syn/anti ratios for com-
pounds 2a–g and 3a–g and the corresponding free energies
(calculated by using the equation: DG=ꢀRTlnKeq =
ꢀRTln(1/2·syn/anti ratio)) are compiled in Table 1. Elec-
The calculated energies related to acidity trend is given in
Table 2. An increase in acidity corresponds to a decreasing
value of DE. The acidity trends obtained for the a-substitut-
ed methyl acetates are MeCH2CO2MeꢁHCH2CO2Me<
MeOCH2CO2Me<ClCH2CO2MeꢁICH2CO2Meꢁ
BrCH2CO2Me !CNCH2CO2Me.
Table 1. The ratios of syn/anti isomers for model compounds 2a–g and
3a–g (experiments were performed in CDCl3).
Compound
Y
X
syn/anti ratio (ꢀ408C) DG[a] [kcalmolꢀ1
]
2c
2a
2d
2e
2 f
2b
2g
3a
3c
3d
3e
3 f
3g
H
H
H
H
H
H
H
Me
F
H
Me
Cl
Br
I
OMe
CN
Br
Br
Br
Br
Br
3.7
3.6
ꢀ0.27
ꢀ0.26
ꢀ0.86
ꢀ0.76
ꢀ0.74
ꢀ0.58
ꢀ1.70
ꢀ0.73
ꢀ0.52
ꢀ0.52
ꢀ0.54
ꢀ0.33
ꢀ0.24
Table 2. Calculated deprotonation energies of 2a–g at the MP2
311++G**.[a]
ACHTUNGTRENNUNG(full)/6-
14.0
11.2
10.6
7.4
93.5
9.9
6.2
6.2
6.7
4.1
Entry
X
E (ester) [a.u.]
E (enolate) [a.u.]
DE [kcalmolꢀ1
]
1
2
3
4
5
6
7
Me
MeO
H
Cl
Br
ꢀ307.0550875
ꢀ382.1218883
ꢀ267.8400638
ꢀ726.9390650
ꢀ2840.133197
ꢀ7154.372297
ꢀ359.9050864
ꢀ306.4436839
ꢀ381.5179183
ꢀ267.2296513
ꢀ726.3452216
ꢀ2839.5422679
ꢀ7153.7804367
ꢀ359.3485631
383.7
379.0
383.0
372.6
370.8
371.4
349.2
Cl
Br
CN
I
CN
[a] The calculations were performed by using Gaussian 03.[28]
NO2 Br
3.5
[a] Errors are estimated at ꢂ0.05 kcalmolꢀ1 (from an average of two
runs).
Once the interaction free energies and the acidity trends
were collected, the data were analyzed through correlation
with Hammett constants sm and with calculated deprotona-
tion energy values DE. The results are shown in Figure 5.
The free energies determined at ꢀ408C from the syn/anti
ratios were plotted against Hammett constants sm for com-
pounds 3a–g as shown in Figure 5a. Compounds 3a–g have
a fixed a-CH2Br group linked to C1 acetate and a variable
Y group on the C9 benzyl group. As the substituent Y be-
comes a stronger electron-withdrawing group, the attractive
interactions diminish. A linear free energy relationship with
R2 =0.97 was obtained. This is consistent with a predomi-
nant CH···p interaction. As the p-electron density of the ar-
omatic ring decreases, the CH···p interaction also decreases.
An alternative explanation would invoke direct Y···Br inter-
actions because the distance between the substituent Y and
the bromo atom is close enough to have a repulsive interac-
tion in the syn conformation. However, the fact that the at-
tron-withdrawing groups increase the syn/anti ratios for
model compounds 2a–g, in which the substituent is on the
a-substituted acetate group. The syn conformation is fa-
vored for all compounds even though steric repulsions are
possible in the syn arrangement. The highest syn/anti ratio is
observed for compound 2g, in which X=CN. The results
are consistent with a CH···p interaction in which electron-
withdrawing groups increase the acidity of the CH group,
which in turn increases the attractive interaction.
The opposite is true for model compounds 3a–g, that is,
electron-withdrawing groups decrease the syn/anti ratios.
Keep in mind that the substituent is on the C9 aromatic ring
for 3a–g, in which an electron-withdrawing group would
reduce the p-electron density. Compounds 3a and 3b, in
which Y=Me and H, show the highest syn/anti ratio, where-
as 3g, in which Y=NO2, displays the lowest preference for
12360
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12357 – 12362

Quantification of CH···p Interactions
FULL PAPER
drawing ability of the substituent. However, the correlation
is not nearly as good as when the substituent is on the ben-
zene ring, such as in compounds 3a–g (Figure 5a).
The free energies determined at ꢀ408C from the syn/anti
ratios were also plotted against the calculated deprotonation
energies in the gas phase (Figure 5c). An excellent linear
correlation with R2 =0.98 was obtained. This is by far the
strongest support for the predominance of the CH···p inter-
actions in the syn conformation of 2a–g. The Hammett con-
stants sm give an excellent correlation for compounds 3a–g
when the substituent is on the aromatic ring (Figure 5a) as
they should be. However, they also provide a fairly straight
line as in Figure 5b even though the substituent is not on an
aromatic ring. Our experimental results from model com-
pounds 2a–g and 3a–g made us reexamine the results re-
ported by Wheeler and Houk.[19] It appears that an attenuat-
ed (due to the computational constraint placed on the dis-
ꢀ
tance between the XH and the benzene ring) X H···p inter-
action has been captured by the simple model. As shown in
ꢀ
Scheme 1b, the partially positive hydrogen atom of the X H
group is pointed near the negative p system of the benzene
ꢀ
ring. The X H···p interaction appears to have resulted in
the similar Hammett plots obtained in the original report.[19]
Conclusion
Our experimental study has shown that a predominant
CH···p interaction controls the conformational preference of
model compounds 2a–g and 3a–g. Despite the predomi-
nance of the CH···p interaction in compounds 2a–g, a Ham-
mett plot displays a fairly straight line for the substituent
effect. These results show that when using Hammett plots in
a simplified model system one should be mindful of the limi-
tations. We initially set out to identify the proposed X···p in-
teractions, but we are now convinced that the predominant
interactions are of the CH···p type in 2a–g and 3a–g. The
model compounds 2a–g and 3a–g cannot exclude the CH···p
interactions from X···p interactions. Therefore they are not
ideal model compounds to study X···p interactions. Howev-
er, the same can be said about the simple model shown in
Scheme 1b. The benzene dimer is a prototype system for
studying aromatic interactions and extensive theoretical
studies have been reported on such systems.[30] The replace-
ment of one benzene ring in a benzene dimer system with a
hydrogen atom went one step further in simplifying real
world aromatic interactions. We caution against using such
simplifications when examining arene–arene interactions,
since our experimental results have shown that misleading
conclusions may be obtained. Instead, computational studies
should explore more realistic systems of aromatic interac-
tions and avoid generalization from studies of a simplified
model system.
Figure 5. a), b) Plots of free energies versus Hammett constants sm and
c) free energies versus calculated deprotonation energies, where a) is for
3a–g, b) is for 2a–g, and c) is for 2a–g.
tractions decrease with an increasingly electron-deficient ar-
omatic ring is inconsistent with an X(Br)···p interaction.
The syn/anti ratios for model compounds 2a–g vary from
3.6 to 93.5 when X changes from methyl to CN (see
Scheme 2 and Table 1). Compounds 2a–g have a fixed C9
benzyl group with various substituent X attached to the a-
carbon of the C1 acetate group. When X is a stronger elec-
tron-withdrawing group the syn/anti ratio increases rapidly.
This trend does not rule out a direct interaction between the
X group and the aromatic ring. However, it further supports
the notion that an attractive CH···p interaction is predomi-
nant in the syn conformations. Thus, an electron-withdraw-
ing group increases the acidity of the a-CH protons, which
in turn enhances the attractive CH···p interaction. To verify
that this is indeed the case, two correlations are made: one
uses Hammett constants (Figure 5b) and the other uses the
deprotonation energy (Figure 5c).
The free energies determined at ꢀ408C from the syn/anti
ratios are plotted against Hammett constants sm for com-
pounds 2a–g as shown in Figure 5b. A linear free energy re-
lationship with R2 =0.90 was obtained. This is consistent
with the fact that sm is a good indicator of the electron-with-
Chem. Eur. J. 2010, 16, 12357 – 12362
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12361

B. W. Gung et al.
[9] P. Hobza, H. L. Selzle, E. W. Schlag, J. Phys. Chem. 1996, 100,
18790–18794.
Experimental Section
[10] R. L. Jaffe, G. D. Smith, J. Chem. Phys. 1996, 105, 2780–2788.
[11] M. O. Sinnokrot, E. F. Valeev, C. D. Sherrill, J. Am. Chem. Soc.
2002, 124, 10887–10893.
[12] F. Tran, J. Weber, T. A. Wesolowski, Helv. Chim. Acta 2001, 84,
1489–1503.
[13] S. Tsuzuki, K. Honda, T. Uchimaru, M. Mikami, K. Tanabe, J. Am.
Chem. Soc. 2002, 124, 104–112.
[14] E. C. Lee, B. H. Hong, J. Y. Lee, J. C. Kim, D. Kim, Y. Kim, P. Tara-
keshwar, K. S. Kim, J. Am. Chem. Soc. 2005, 127, 4530–4537.
[15] C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5525–
5534.
[16] F. Cozzi, M. Cinquini, R. Annuziata, J. S. Siegel, J. Am. Chem. Soc.
1993, 115, 5330–5331.
[17] F. Cozzi, M. Cinquini, R. Annunziata, T. Dwyer, J. S. Siegel, J. Am.
Chem. Soc. 1992, 114, 5729–5733.
[18] M. O. Sinnokrot, C. D. Sherrill, J. Am. Chem. Soc. 2004, 126, 7690–
7697.
[19] S. E. Wheeler, K. N. Houk, J. Am. Chem. Soc. 2008, 130, 10854.
[20] S. E. Wheeler, K. N. Houk, J. Am. Chem. Soc. 2009, 131, 3126.
[21] M. J. Rashkin, M. L. Waters, J. Am. Chem. Soc. 2002, 124, 1860–
1861.
[22] A. L. Ringer, C. D. Sherrill, J. Am. Chem. Soc. 2009, 131, 4574.
[23] B. W. Gung, B. U. Emenike, C. N. Alverez, J. Rakovan, K. Kirsch-
baum, N. Jain, Tetrahedron Lett. 2010, 51, 1648–1650.
[24] B. W. Gung, F. Wekesa, C. L. Barnes, J. Org. Chem. 2008, 73, 1803–
1808.
[25] B. W. Gung, M. Patel, X. W. Xue, J. Org. Chem. 2005, 70, 10532–
10537.
[26] B. W. Gung, X. W. Xue, H. J. Reich, J. Org. Chem. 2005, 70, 3641–
3644.
[27] M. Nishio, M. Hirota and Y. Umezawa, The CH/pi Interaction: Evi-
dence, Nature and Consequences, Wiley-VCH, Weinheim, 1998.
[28] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
Representative procedure for the preparation of the model compounds
9-Phenyl-1-propionyl-4-trimethylacetyloxytriptycene (2a): Compound 4a
(100 mg, 0.208 mmol) was added to a solution of 4-dimethylaminopyri-
dine (DMAP; 5.3 mg, 0.042 mmol), pyridine (0.5 mL, 10.5 mmol), and
CH2Cl2 (2 mL) under a nitrogen atmosphere. After stirring the mixture
for 10 min at 08C, propionyl chloride (29 mg, 0.313 mmol) was added by
syringe. The reaction was monitored by TLC. After completion, the reac-
tion was warmed to room temperature, diluted with CH2Cl2, and
quenched with 1n HCl. The aqueous layer was extracted three times
with Et2O, and the combined organic extracts were washed with a 10%
aqueous solution of NaOH and brine. The organic extracts were dried
over MgSO4, filtered, concentrated, and purified by silica gel flash chro-
matography (5!10!20% EtOAc/hexanes) to yield a white solid (87 mg,
78%). M.p. 238–2408C; 1H NMR (500 MHz, CDCl3): d=0.99–1.01 (brs,
3H), 1.61 (s, 9H), 2.02–2.20 (brs, 2H), 4.68 (brs, 2H), 5.54 (s, 1H), 6.78–
7.39 ppm (m, 15H); 13C NMR (125 MHz, CDCl3): d=27.5, 34.7, 39.5,
48.7, 52.8, 119.8, 121.4, 123.7, 124.9, 125.6, 137.9, 129.7, 142.9, 143.9,
172.3, 176.5 ppm; LCMS: m/z: 539.3
Variable-temperature NMR spectroscopy experimental procedure: The
1H NMR spectra were recorded on a 300 MHz instrument with a varia-
ble-temperature probe. A 0.03m solution of the sample in a deuterated
solvent, such as chloroform, was placed in a high-quality NMR tube. All
samples were degassed by using a needle to bubble nitrogen through the
sample for about 1 min. The NMR tube was then capped and sealed with
parafilm. The sample tube was placed into the NMR probe and the air-
line to the probe was replaced with liquid-nitrogen transfer line. The de-
sired temperature was set on the variable-temperature unit and the
sample was allowed to equilibrate for 10–15 min at each set temperature.
Then the 1H NMR spectrum at each temperature was recorded. The
ratios of rotamers were obtained through the integrations of selected
peaks.
Acknowledgements
B.W.G. thanks the CFR of Miami University for a summer stipend. M.L.
thanks the National Center for Supercomputing Applications
(CHE050039N) for time on the Dell Intel 64 Linux Cluster. We thank
Laura K. E. Hardebeck and Selina Wireduaah for technical assistance
with the ab initio computations.
[1] J. M. Lehn, Supramolecular Chemistry: Concepts and Perspectives,
VCH, Weinheim, 1995, p. 271.
[2] E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. 2003,
115, 1244–1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250.
[3] J. Cerny, P. Hobza, Phys. Chem. Chem. Phys. 2007, 9, 5291–5303.
[4] H. J. Schneider, Angew. Chem. 2009, 121, 3982–4036; Angew. Chem.
Int. Ed. 2009, 48, 3924–3977.
[5] K. Morokuma, Acc. Chem. Res. 1977, 10, 294–300.
[6] K. Mꢀller-Dethlefs, P. Hobza, Chem. Rev. 2000, 100, 143–167.
[7] W. L. Jorgensen, D. L. Severance, J. Am. Chem. Soc. 1990, 112,
4768–4774.
[29] S. Damoun, W. Langenaeker, G. Vandewoude, P. Geerlings, J. Phys.
Chem. 1995, 99, 12151–12157.
[30] M. O. Sinnokrot, C. D. Sherrill, J. Phys. Chem. A 2006, 110, 10656–
10668.
Received: May 18, 2010
[8] G. D. Smith, R. L. Jaffe, J. Phys. Chem. 1996, 100, 9624–9630.
Published online: September 17, 2010
12362
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12357 – 12362