Structures and properties of molecular torsion balances to decipher the nature of substituent effects on the aromatic edge-to-face interaction

Article abstract of DOI:10.1002/chem.201304810

Various recent computational studies initiated this systematic re-investigation of substituent effects on aromatic edge-to-face interactions. Five series of Troeger base derived molecular torsion balances (MTBs), initially introduced by Wilcox and co-workers, showing an aromatic edge-to-face interaction in the folded, but not in the unfolded form, were synthesized. A fluorine atom or a trifluoromethyl group was introduced onto the edge ring in ortho-, meta-, and para-positions to the C-H group interacting with the face component. The substituents on the face component were varied from electron-donating to electron-withdrawing. Extensive X-ray crystallographic data allowed for a discussion on the conformational behavior of the torsional balances in the solid state. While most systems adopt the folded conformation, some were found to form supramolecular intercalative dimers, lacking the intramolecular edge-to-face interaction, which is compensated by the gain of aromatic π-stacking interactions between four aryl rings of the two molecular components. This dimerization does not take place in solution. The folding free enthalpy ΔG_fold of all torsion balances was determined by ¹H NMR measurements by using 10 mM solutions of samples in CDCl ₃ and C₆D₆. Only the ΔG_fold values of balances bearing an edge-ring substituent in ortho-position to the interacting C-H show a steep linear correlation with the Hammett parameter (σ_meta) of the face-component substituent. Thermodynamic analysis using vant Hoff plots revealed that the interaction is enthalpy-driven. The ΔG_fold values of the balances, in addition to partial charge calculations, suggest that increasing the polarization of the interacting C-H group makes a favorable contribution to the edge-to-face interaction. The largest contribution, however, seems to originate from local direct interactions between the substituent in ortho-position to the edge-ring C-H and the substituted face ring. Direct interactions dominate: The importance of local direct interactions in the edge-to-face interaction was examined with a large library of Troeger base derived molecular torsion balances. Modulation of the folding free enthalpy is only observed when edge-ring substituents are capable of directly interacting with the face-ring π system (see figure). Experimental results and theoretical calculations indicate that local direct interactions make the largest contribution to substituent effects in the edge-to-face aromatic interactions.

Full text of DOI:10.1002/chem.201304810

DOI: 10.1002/chem.201304810
Full Paper
&
Aromatic Interactions
Structures and Properties of Molecular Torsion Balances to
Decipher the Nature of Substituent Effects on the Aromatic Edge-
to-Face Interaction
Haraldur Gardarsson, W. Bernd Schweizer, Nils Trapp, and FranÅois Diederich*[a]
Abstract: Various recent computational studies initiated this
systematic re-investigation of substituent effects on aromatic
edge-to-face interactions. Five series of Trçger base derived
molecular torsion balances (MTBs), initially introduced by
Wilcox and co-workers, showing an aromatic edge-to-face
interaction in the folded, but not in the unfolded form, were
synthesized. A fluorine atom or a trifluoromethyl group was
introduced onto the edge ring in ortho-, meta-, and para-po-
sitions to the CÀH group interacting with the face compo-
nent. The substituents on the face component were varied
from electron-donating to electron-withdrawing. Extensive
X-ray crystallographic data allowed for a discussion on the
conformational behavior of the torsional balances in the
solid state. While most systems adopt the folded conforma-
tion, some were found to form supramolecular intercalative
dimers, lacking the intramolecular edge-to-face interaction,
which is compensated by the gain of aromatic p-stacking in-
teractions between four aryl rings of the two molecular com-
ponents. This dimerization does not take place in solution.
The folding free enthalpy DGfold of all torsion balances was
1
determined by H NMR measurements by using 10 mm solu-
tions of samples in CDCl3 and C6D6. Only the DGfold values of
balances bearing an edge-ring substituent in ortho-position
to the interacting CÀH show a steep linear correlation with
the Hammett parameter (smeta) of the face-component sub-
stituent. Thermodynamic analysis using van’t Hoff plots re-
vealed that the interaction is enthalpy-driven. The DGfold
values of the balances, in addition to partial charge calcula-
tions, suggest that increasing the polarization of the inter-
acting CÀH group makes a favorable contribution to the
edge-to-face interaction. The largest contribution, however,
seems to originate from local direct interactions between
the substituent in ortho-position to the edge-ring CÀH and
the substituted face ring.
Introduction
Noncovalent aromatic–aromatic interactions are pivotal in nu-
merous chemical and biological processes.[1] These interactions
play a key role in host–guest chemistry,[2] supramolecular self-
assembly,[3] stereoselective reactions,[4] chemoselective cataly-
sis,[5] or protein–ligand complexation[6] and influence the struc-
ture of biomolecules.[7] There are three major structural motifs
for aryl–aryl interactions: parallel-eclipsed, parallel-displaced,
and edge-to-face (Figure 1). It is accepted that substituents
affect the aryl–aryl interactions, but the exact nature of sub-
stituent effects remains a topic under investigation, both in
theoretical and experimental studies.[8]
Figure 1. Aryl–aryl ring interaction motifs.
1
around the biaryl bond is slow on the H NMR timescale, and
the ratio of the folded versus the unfolded conformation is
conveniently determined by integration of the signals of the
methyl group at C3, yielding the folding free enthalpy DGfold as
a measure for the aromatic interaction. The Pittsburgh group
studied substituent effects in CDCl3 by using an unsubstituted
phenyl ester edge ring and varying the face substituent in
para-position to the aniline N atom. A correlation between the
driving force for folding and the Hammett parameter (smeta) of
the face substituent was not observed. Together with investi-
gations on aliphatic edge moieties, this finding led the authors
to the suggestion that dispersion, rather than CÀH···p-type
electrostatic interactions between edge and face rings, contrib-
utes to the preference of the torsion balance for the folded
state.[9b,c]
In the 1990s, Wilcox and co-workers introduced the Trçger
base derived molecular torsion balance to investigate aromatic
edge-to-face interactions.[9] This system features, in its folded
state, an interaction between rings A and D (Figure 2), while
the interaction is absent in the unfolded form. Atropisomerism
[a] H. Gardarsson, Dr. W. B. Schweizer, Dr. N. Trapp, Prof. Dr. F. Diederich
Laboratory of Organic Chemistry, ETH Zurich
Hçnggerberg, HCI, 8093 Zurich (Switzerland)
Fax: (+41)44-632-1109
E-mail: diederich@org.chem.ethz.ch
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201304810.
Chem. Eur. J. 2014, 20, 1 – 10
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
ductive and field effects overwhelm the p-polariza-
tion effects. Sherrill and co-workers reported that
direct interactions[18] between substituents and the
opposing p-system are, at least partly, responsible for
substituent effects in parallel-eclipsed benzene
dimers and not the change of the p-electron density
in the rings.[15b,d,e] For the aromatic edge-to-face inter-
action, theoretical studies mostly agree that a stabiliz-
ing effect is achieved by the introduction of an elec-
Figure 2. Schematic representation of the equilibrium between the folded (right) and un- tron-withdrawing group on the edge ring, or the in-
folded (left) atropisomers of the Trçger base derived molecular torsion balance. The func-
tional groups examined in this study as well as the numbering system used, are shown.
troduction of an electron-donating group on the face
ring.[14e,15a,16b] In the majority of these computational
studies, the functional group is placed in the para-
With sets of new Trçger base derived molecular torsion bal-
ances, we have previously quantified the weak attractive or-
thogonal dipolar interaction between organofluorine and
amide groups[10] and extended these investigations to the en-
ergetics of dipolar interactions between orthogonal amide
groups.[11]
position of the edge CÀH which interacts with the face-ring
(Figure 3a). A theoretical investigation by Wheeler and Houk,
on the other hand, places the edge-ring functional group in
such a way that a direct interaction between this group and
the p-system of the face ring is possible, similar to the situa-
tion in the Trçger base derived torsion balances investigated
so far (Figure 3b, c).[14c]
During the course of our investigations on orthogonal dipo-
lar interactions involving organofluorine, we noticed a signifi-
cant substituent effect on the edge-to-face interaction in tor-
sion balances bearing a 4-(trifluoromethyl)phenyl ester as the
edge component and various electron-withdrawing (EWG) or
electron-donating groups (EDG) on the face component in
para-position to the anilino N atom.[10a] The folding free enthal-
py in C6D6 showed a steep linear correlation with the Hammett
parameter (smeta) of the respective substituents on the face
ring, with the DGfold values becoming more negative with in-
creasing electron-donating potency. We explained this finding
with a substantial electrostatic CÀH···pface interaction providing
the driving force for folding.[10a] These results were in stark con-
trast to the work by Wilcox and co-workers described
above.[9b,c] The different results between the two studies were
initially explained by Cockroft and Hunter by differences in sol-
vation effects in C6D6 and CDCl3.[12] A re-measurement of both
sets of torsion balances in both solvents confirmed all earlier
results and showed similar folding behavior in the two sol-
vents.[13] A steep linear free enthalpy relationship (LFER) was
observed for the torsion balances with an edge-trifluoromethyl
group in both C6D6 and CDCl3. No such relationship was
indeed observed for the unsubstituted phenyl ester as edge
component. Clearly, the additional CF3 substituent in ortho-po-
sition to the interacting edge-CÀH made all the difference. This
finding suggested the importance of re-investigating substitu-
ent effects on the aromatic edge-to-face interaction in a com-
prehensive experimental study.
Figure 3. Edge-to-face interaction motifs used in theoretical investigations. X
denotes any substituent.
Insight into the nature of aromatic edge-to-face interactions
using torsional balances can be further complicated by the
fact that local dipolar interactions between the dipoles of the
edge and the face ring in the folded state could also affect the
folding equilibrium. In a computational study, we recently
demonstrated the importance of dipole orientation for the in-
termolecular interaction between heteroarenes stacking on
peptide bonds: an antiparallel alignment of the dipoles is ener-
getically favorable, while parallel alignment is unfavorable.[19] In
their studies utilizing molecular zippers,[20] Hunter and co-work-
ers already alerted to the possible energetic contribution re-
sulting from interactions between the edge- and face-ring di-
poles.[20f,g]
For this re-investigation into the nature of aromatic edge-to-
face interactions, we synthesized 25 new molecular torsion bal-
ances and investigated their folding equilibria. Based on these
data, we analyze the energetic contributions of 1) the electro-
static interaction between edge-ring CÀH and the face-ring p-
system, 2) direct interaction of edge-ring substituents with the
substituted face ring, and 3) the interaction between the local
dipoles of substituted edge and face rings. Additionally, nu-
merous crystal structures are presented that illustrate a rather
remarkable flexibility of the Trçger base scaffold.[21]
Additional interest in such an experimental investigation
comes from recent computational studies on aryl–aryl interac-
tions by Wheeler and Houk,[14] Sherrill,[15] and others.[16] These
studies suggested that the earlier, purely electrostatics-based
Hunter–Sanders model[17] is not sufficient and does not ade-
quately describe aromatic interactions. Theoretical work by
Wheeler and Houk on substituted benzene dimers indicated
that all substituents, both electron-donating and electron-ac-
cepting, stabilize the parallel-eclipsed and parallel-displaced
benzene dimers.[14a,b] They concluded that through-space in-
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
rived torsion balances crystallized until today adopted the
folded form, clearly illustrating a favorable edge-to-face inter-
action in the solid state.[9b,10,13] In this study, we observed the
first examples of supramolecular dimers in which the edge-to-
face interaction is absent.
Results and Discussion
Synthesis and solid-state structures of molecular torsion bal-
ances
The synthesis of torsion balances (Æ)-1 to (Æ)-25 (Table 1) es-
sentially followed the protocols reported earlier.[9a] Synthetic
procedures and complete compound characterization are
Torsion balances (Æ)-1 (4’’-F/NH2) (the first substituent is the
edge substituent labeled with respect to the ester linker in po-
sition 1’’ and the second is the face-substituent in para-posi-
tion to the anilino N-atom) and (Æ)-4 (4’’-F/I) crystallize in the
folded state, depicting a classic edge-to-face interaction (Sup-
porting Information). The distance between the centroid of
ring D and the calculated plane of ring A for (Æ)-4 is approxi-
mately 4.9 ꢁ, which is similar to previously reported molecular
torsion balances.[9b,10,13] The crystal structure of (Æ)-5 (4’’-F/
NO2) also shows a favorable edge-to-face interaction between
rings A and D, while maintaining an interatomic distance be-
tween fluorine and nitro group that is larger than the sum of
the van der Waals radii (observed d(F···N) 3.34 ꢁ; sum of van
der Waals radii 3.02 ꢁ; Figure 4, top).[22]
Table 1. Folding free enthalpies DGfold for fluorinated and trifluoromethy-
lated molecular torsion balances.
[a]
[b]
R1
R2
smeta
DGfold [kJmolÀ1
]
C6D6
CDCl3
(Æ)-1
4’’-F
4’’-F
4’’-F
4’’-F
4’’-F
3’’-F
3’’-F
3’’-F
3’’-F
3’’-F
2’’-F
2’’-F
2’’-F
2’’-F
2’’-F
4’’-CF3
4’’-CF3
4’’-CF3
4’’-CF3
4’’-CF3
3’’-CF3
3’’-CF3
3’’-CF3
3’’-CF3
3’’-CF3
2’’-CF3
2’’-CF3
2’’-CF3
2’’-CF3
2’’-CF3
NH2
H
OH
I
NO2
NH2
H
OH
I
NO2
NH2
H
OH
I
NO2
NH2
H
OH
I
NO2
NH2
H
OH
I
NO2
NH2
H
OH
I
À0.160
0.000
0.121
0.352
0.710
À0.160
0.000
0.121
0.352
0.710
À0.160
0.000
0.121
0.352
0.710
À0.160
0.000
0.121
0.352
0.710
À0.160
0.000
0.121
0.352
0.710
À0.160
0.000
0.121
0.352
0.710
À3.48
À2.66
À2.81
À2.22
À0.52
À1.14
À1.04
À1.01
À1.05
À0.60
À1.75
À1.24
À1.62
À1.77
À1.23
À3.91
À3.47
À3.46
À1.52
À0.19
À1.00
À0.34
À0.67
À0.49
À0.58
À1.38
À0.89
À1.23
À1.45
À1.26
À2.01
À1.78
À1.85
À1.27
+0.01
À0.30
À0.34
À0.73
À0.48
À0.63
À0.64
À0.69
À1.08
À0.84
À1.05
À2.65
À2.41
À2.31
À0.61
+0.47
À0.23
À0.11
À0.38
À0.23
+0.41
À0.20
À0.32
À0.63
À0.82
À0.73
(Æ)-2
(Æ)-3
(Æ)-4
(Æ)-5
(Æ)-6
(Æ)-7
(Æ)-8
(Æ)-9
(Æ)-10
(Æ)-11
(Æ)-12
(Æ)-13
(Æ)-14
(Æ)-15
(Æ)-26[c]
(Æ)-27[c]
(Æ)-28[c]
(Æ)-29[c]
(Æ)-30[c]
(Æ)-16
(Æ)-17
(Æ)-18
(Æ)-19
(Æ)-20
(Æ)-21
(Æ)-22
(Æ)-23
(Æ)-24
(Æ)-25
NO2
[a] Hammett parameters based on the ionization of substituted benzoic
acids.[28] [b] Determined by integration of the line-fitted (100% Lorentz
functions) 1H NMR spectra of 10 mm solutions at 298 K. Uncertainty
Æ0.12 kJmolÀ1. [c] DGfold values previously reported.[10a,13]
given in the Supporting Information.[29] Additionally, data for
the already described systems (Æ)-26 to (Æ)-30 are included in
Table 1.[10a,13]
Figure 4. Top: ORTEP plot of a symmetry independent molecule of (Æ)-5 (4’’-
F/NO2). Atomic displacement parameters obtained at 100 K drawn at the
50% probability level. Arbitrary numbering. Bottom: The superimposed crys-
tal structures of (Æ)-5 (gray) and (Æ)-30 (green) (CCDC-239177).[10a] Magenta
dotted lines indicate the distance between the centroid of ring D and the
calculated plane of ring A. Black dotted lines indicate the closest F···N con-
tact.
Crystals of several molecular torsion balances suitable for X-
ray analysis were obtained by vapor diffusion of n-pentane
into solutions of the compounds in CH2Cl2. All Trçger base de-
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
When the crystal structures of
(Æ)-5 (4’’-F/NO2) and the previ-
ously reported balance (Æ)-30
(4’’-CF3/NO2)
(CCDC
code:
239177)[10a] are superimposed, it
becomes apparent that the
“hinge” angle (angle between
the calculated planes of rings A
and B) of the Trçger base scaf-
fold is quite variable (Figure 4,
bottom), as previously also re-
ported by Wilcox and co-work-
ers.[21] By changing from a hinge
angle of 1168 in (Æ)-30 to 978
for (Æ)-5, the interatomic dis-
tance between (the closest) fluo-
rine and the nitro N atom be-
comes similar in both balances
(3.34 ꢁ for (Æ)-5 and 3.36 ꢁ for
(Æ)-30). This becomes possible,
since ring D in (Æ)-5 moves sub-
stantially closer to ring A and is
no longer fully perpendicular to
this ring. In (Æ)-5, ring D is twist-
Figure 6. Dimerized structures of a) (Æ)-10 (3’’-F/NO2) and b) (Æ)-25 (2’’-CF3/NO2), and c) the parameters of the
parallel-displaced stacking interactions. Atomic displacement parameters of fluorine at 100 K drawn at 50% proba-
bility. Only the major conformer for (Æ)-10 is shown for clarity. The angle a is the angle between the calculated
planes of the interacting rings. Black dotted lines illustrate the favorable interactions in the dimerized structures.
Gray dotted lines indicate the closest F···N contact.
ed by approximately 208 relative to its position in (Æ)-30. As
a result of these conformational adjustments, the distance be-
tween the calculated centroid of ring D and the calculated
plane of ring A changes from 4.98 ꢁ ((Æ)-30) to 4.45 ꢁ ((Æ)-5)
(Figure 4, bottom). All symmetry-independent molecules of
(Æ)-5 show smaller inter-ring distances, as compared to (Æ)-30
(Supporting Information).
intermolecular parallel-displaced aromatic interactions, as
shown in Figure 6 (for the dimerized structure of (Æ)-15 see
the Supporting Information).
The extent of additional stabilization achieved by dimeriza-
tion, is best illustrated by (Æ)-25 (2’’-CF3/NO2). The dimer
causes a fluorine atom and the nitro group to come into a pre-
sumably destabilizing contact (observed d(F···N) 2.87 ꢁ; sum of
van der Waals radii 3.02 ꢁ),[22] which is overcome by the stabili-
zation from the parallel-displaced stacking interactions (Fig-
ure 6b). Interestingly, the isomeric molecular torsion balance
(Æ)-20 (3’’-CF3/NO2) crystallizes as a monomer in the “normal”
folded fashion (Supporting Information). Also, the crystal struc-
tures of previously reported tor-
The crystal structures of (Æ)-10 (3’’-F/NO2), (Æ)-15 (2’’-F/NO2),
and (Æ)-25 (2’’-CF3/NO2) do not show an edge-to-face interac-
tion, as supramolecular dimers form in the solid state (Figure 5
and the Supporting Information). In these dimers, the balances
lose the intramolecular edge-to-face interaction, but gain three
sional balances (Æ)-26 (4’’-CF3/
NH2), (Æ)-27 (4’’-CF3/H), and (Æ)-
31 (4’’-CF3/Br) all show the
folded atropisomer, with an in-
tramolecular edge-to-face inter-
action (Supporting Information).
Overall, this extensive X-ray
analysis confirms a substantial
flexibility in the Trçger base scaf-
fold of the torsional balances,
which in the folded form leads
to variation in the hinge angle
between its two aromatic
rings A and B. Rotation around
the bonds connecting the ester
carboxyl to ring C and edge
Figure 5. ORTEP plots of (Æ)-10 (3’’-F/NO2) and (Æ)-15 (2’’-F/NO2), left and right, respectively. Atomic displacement
parameters obtained at 100 K drawn at the 50% probability level. Arbitrary numbering. For (Æ)-10, the fluoro-
phenyl ring is disordered and has been refined over two positions. For (Æ)-15, the carboxylic group and the fluo-
rine are disordered and have been refined over two positions. Gray dotted lines indicate the distance between
the calculated centroid of ring D and the calculated plane of ring A.
ring D results in
a molecular
tweezer-type geometry and the
formation of an intercalative
supramolecular
dimer
with
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
a triple parallel-shifted aromatic p-stack, a hitherto unobserved
motif for stable crystal packing of Trçger base-derived torsional
balances.
1
Folding equilibrium constants determined by H NMR spec-
troscopy
In order to determine the folding equilibrium, the 1H NMR
spectra of each molecular torsion balance were measured as
10 mm solutions in CDCl3 and C6D6 at 298 K. Prior to integra-
tion, the C3-methyl group signals were line-fitted to a 100%
Lorentz function. Additional details on folding free enthalpy
extraction and error analysis can be found in the Supporting
Information. The folding free enthalpies (DGfold) are summar-
ized in Table 1.
To exclude the presence of supramolecular dimers in solu-
tion, the folding equilibrium of (Æ)-10 was measured in CDCl3
as a function of concentration (Figure S7 in the Supporting In-
formation). The equilibrium is essentially independent of con-
1
centration (between 0.5 and 100.0 mm), which, along with H-
19F HOESY experiments (Supporting Information), indicates no
dimerization in solution at the concentration used for this
study (10 mm). We also conducted a molecular dynamics (MD)
simulation (Figure S19 in the Supporting Information) for the
fluorine series of balances, which suggests that the edge
ring D is fully rotatable independent of the position of the F
substituent. The barrier for the rotation around the ester bond
to ring D was calculated for model systems 2’’-, 3’’-, and 4’’-flu-
orophenyl acetate at the wB97XD/6-311G+ +(2d,p) level of
theory.[23] While the rotational barriers for 3’’- and 4’’-substitut-
ed derivatives were below 10 kJmolÀ1, the barrier for the 2’’-
substituted compound was calculated as 47 kJmolÀ1 (Fig-
ure S20 in the Supporting Information), which is in agreement
with the rather unhindered rotation seen in the MD study.
To analyze the substituent effects, the folding free enthalpies
were plotted against the Hammett parameter (smeta) of the re-
spective face substituent (Figures 7 and 8). The folding free en-
thalpy of the 4’’-F series (Æ)-1 to (Æ)-5 shows a steep linear
correlation with the Hammett parameter, with DGfold values be-
coming more negative upon moving to electron-donating sub-
stituents (Figure 7), which was also reported for the 4’’-CF3
MTB series.[10a,13] At the time, this correlation was contributed
to an edge-to-face interaction strongly modulated by electro-
static contribution from the CÀH···pface contact. The folding
equilibria of 2’’-F and 3’’-F torsion balances ((Æ)-6 to (Æ)-15)
are nearly independent of the face-ring substituent (Figure 7).
All balances, however, that contain a hydrogen or iodine atom
on the face-ring, show DGfold values in C6D6 which are different
than expected. As predicted by theoretical work by Wheeler
and Houk,[14c] this is probably due to the small, for hydrogen,
and large, for iodine, dispersion contribution to the edge-to-
face interaction. The folding in the 2’’-F series is consistently
more favorable compared to the 3’’-F series. The same trend is
observed for the trifluoromethylated series (2’’- and 3’’-CF3).
The reasons for this are, as yet, unclear. The desolvation model
postulated by Hunter and Cockroft[12] cannot explain the afore-
mentioned results, as all molecular torsion balances now con-
Figure 7. Folding free enthalpies DGfold of molecular torsion balances (Æ)-
1 to (Æ)-15 in C6D6 (top) and CDCl3 (bottom) at 298 K plotted against the re-
spective Hammett parameter (smeta) of substituent R2.
tain a polarizing fluorine or a trifluoromethyl group on the
edge-ring. Recently, Cockroft and co-workers have examined
the desolvation of substituents.[24] Although desolvation could
be involved in the modulation of the folding free enthalpy
within a series (e.g., ((Æ)-1 to (Æ)-5), this approach is inade-
quate to describe the differences between the 2’’-, 3’’-, and 4’’-
substituted series, as they contain identical substituents.
The folding equilibrium in the 4’’-CF3 series is affected to
a greater extent by the nature of the face substituent than in
the 4’’-F series (Figure 8). Upon changing from (Æ)-30 (4’’-CF3/
NO2) to (Æ)-26 (4’’-CF3/NH2), the DGfold value in deuterated ben-
zene changes by À3.72 kJmolÀ1, whereas moving from (Æ)-5
(4’’-F/NO2) to (Æ)-1 (4’’-F/NH2), the free enthalpy difference
amounts to À2.96 kJmolÀ1. As aryl-bound fluorine exerts
a stronger polarization effect than an aryl-bound trifluorometh-
yl group on hydrogens ortho to the substituent (see below),
polarization alone is not enough to explain these folding data.
A stronger direct interaction between the CF3 group and the
face-ring p-system, which adds to the polarization effect, could
explain these results.
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 9. a) Calculated ESP (CHELPG) partial charges and the corresponding
dipole moments of trifluoromethylated and fluorinated phenyl acetates as
model systems for the edge-component. The dihedral angles (f) C2’’-C1’’-O-
C(O) and C1’’-O-C(O)-C1 were restricted to À117.38 and 174.48, respectively.
b) Calculated dipoles of 4-substituted (from top: NH2, H, OH, I, NO2) N,N-di-
methylanilines as model systems for the face-component in the torsional
balances. Level of theory MP2/aug-cc-pVDZ//wB97XD/6-311G+ +(2d,p), for
all structures except for the iodinated model system, where the MP2/
LANL2DZ//wB97XD/LANL2DZ level of theory was used. Two arrows indicate
the dipole of the hydroxy-containing model system after a 1808 rotation of
the hydroxy moiety. During optimization, parameters were restricted to
mimic the Trçger base backbone. a(C8’-C10a’-N11’)=175.88, a(C6a’-C10a’-
N11’)=121.38, a(C10a’-N11’-C13’)=110.28, a(C10a’-N11’-C12’)=113.48,
f(C6a’-C10a’-N11’-C13’)=À15.38, and f(C6a’-C10a’-N11’-C12’)=105.58.
Values were averaged from crystal structures in this manuscript showing an
edge-to-face interaction. c) Indicates the numbering system of the edge-
and face-components.
Figure 8. Folding free enthalpies DGfold of molecular torsion balances (Æ)-16
to (Æ)-30 in C6D6 (top) and CDCl3 (bottom) at 298 K plotted against the re-
spective Hammett parameter (smeta) of substituent R2. The data for the 4’’-
CF3 series have been taken from previous publications.[10a,13] The data point
for (Æ)-29 (4’’-CF3/I) has been shifted slightly from smeta =0.352 to 0.345, to
avoid complete overlap with the data point for (Æ)-24 (2’’-CF3/I).
Calculated dipoles and partial charges
We estimated the partial charges of the hydrogen atoms in the
F- and CF3-substituted edge ring D of the torsion balances by
performing calculations at the MP2/aug-cc-pVDZ//wB97XD/6-
311G+ +(2d,p) level of theory[23,25] for the respective constitu-
tionally isomeric fluorophenyl and trifluoromethylphenyl ace-
tates (Supporting Information) as model systems for the edge-
part of the molecular torsion balances. Using the ESP grid-
based method (CHELPG),[26] we obtained a good correlation
between calculated and experimental dipole moments (Sup-
porting Information), which had previously been reported
using other theoretical methods.[27] Figure 9a depicts the cal-
culated point charges for 5’’-H (for partial charges of all edge-
ring hydrogens, see Table S11 in the Supporting Information).
In both the F- and CF3-edge series, the extent of polarization
is as expected. The positive partial charge on 5’’-H is the high-
est for 4’’-substituted arenes, while the 2’’- and 3’’-substituted
derivatives show similar partial charges (Figure 9a). The calcu-
lated partial charges, however, do not explain, why the correla-
tion of DGfold values with the Hammett parameter is steeper in
the 4’’-CF3 than in the 4’’-F series. Clearly, the polarization of
5’’-H, which interacts with the face ring, is not the only contrib-
utor to the modulation of the folding free enthalpies by the
substituents on the face ring.
Two possible explanations can be offered. Firstly, there is
a local direct interaction between the edge-substituent at posi-
tion 4’’ and the face-component p-system, which is larger for
the more spacious 4’’-CF3 group than for fluorine. Secondly,
a substantial part of the substituent effect is due to interaction
between the local dipoles of the interacting rings (Figure 9a,
b).[19] The local dipole moments for models of the edge ring
and the face ring (4-substituted N,N-dimethylanilines) were cal-
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
culated at the MP2/aug-cc-pVDZ//wB97xd/6-311G+ +(2d,p)
level of theory, except for the iodinated compound, for which
MP2/LANL2DZ//wB97xd/LANL2DZ was used (Figure 9b). To get
a closer approximation of the dipoles in the molecular torsion
balances, several restraints were applied during optimization
(Figure 9). The model calculations suggest that in the substitut-
ed torsion balances with 4’’-F- or 4’’-CF3-substituents at the
edge ring, the interacting dipoles are aligned in an unfavora-
ble, nearly parallel fashion for the NO2-substituted face ring,
whereas they become increasingly orthogonal and therefore
more favorable when the face-ring is donor-substituted, in par-
ticular in the NH2 case. However, the fact that there is relatively
little change in the folding free enthalpy for the 2’’- and 3’’-
substituted torsion balances despite strongly different local
dipole–dipole interactions suggests that dipolar interactions
alone are not sufficient to explain the substituent effect. At
this point, our joint experimental/computational analysis
points to larger contributions to the substituent effects in aro-
matic edge-to-face interactions from local direct substituent in-
teractions and minor contributions from both the electrostatic
interactions between the polarized 5’’-H atom and the substi-
tuted face ring, and from local dipole–dipole interactions.
er when the face ring is unsubstituted ((Æ)-2), than for both
the amino- and hydroxy-substituted systems (Æ)-1 and (Æ)-3.
This could be due to the dispersion interactions mentioned
earlier, or indicative of a direct interaction between fluorine
and the substituents on ring A. The thermodynamic quantities
of 3’’- and 2’’-edge-ring fluorinated torsion balances do not
change with the nature (EWG or EDG) of the face-substituent.
These results further support the fact that the dominant factor
in the substituent effect seen for the 4’’-F series is the local
direct interaction between the edge substituent and the sub-
stituted face component.
Conclusion
We have synthesized 25 new molecular torsion balances and
examined their folding equilibrium in the ambitious attempt to
provide an experimental study on the nature of substituent ef-
fects on edge-to-face interactions, which have been addressed
in broad, recent computational analysis. We first investigated
in detail the structural properties of the balances in the solid
state and in the majority of the crystal structures observed
a preference for the folded conformation. Conformational flexi-
bility is visible in the Trçger base component, which can adopt
hinge angles between its two aromatic rings varying from 97
to 1168, in agreement with previous findings by the Wilcox
group.[21] Furthermore, some solid-state structures also reveal
different conformations of the carboxylated edge ring, shaping
a cleft for intercalative supramolecular dimer formation. This
dimerization however, is not observed in solution.
Folding analysis by 1H NMR spectroscopy clearly revealed
that only balances bearing a 4’’-edge-ring substituent show
a steep linear correlation with the Hammett parameter of the
face substituent. No such modulation of the driving force for
folding by face substituents was observed in the 2’’- and 3’’-
edge-substituted systems. Also, folding in the 2’’-substituted
series is consistently preferred over the folding in the 3’’-sub-
stituted series. These conclusions were further supported by
VT-NMR studies, followed by van’t Hoff analysis. To explain
these experimental results, theoretical calculations at a high
Thermodynamic analysis by variable-temperature NMR (VT-
NMR) measurements
All F-substituted molecular torsion balances ((Æ)-1 to (Æ)-15)
were subjected to variable-temperature NMR (VT-NMR) studies,
followed by van’t Hoff plot analysis, in order to extract the en-
thalpic (DH) and entropic term (ÀTDS) for the folding process
(Figure 10). For further details on the VT-NMR experiments and
van’t Hoff plots, see the Supporting Information.
As expected, the folding enthalpy for the 4’’-fluorinated
series ((Æ)-1 to (Æ)-5) becomes more favorable as the electron
density of the face-ring is increased (Figure 10, left). As the de-
grees of freedom are reduced upon folding, the entropic term
becomes increasingly unfavorable. The overall result is a small
gain in DGfold as the electron density of ring A is increased. In
the 4’’-edge-ring substituted series, the enthalpic gain is small-
Figure 10. Thermodynamic quantities from VT-NMR of molecular torsion balances (Æ)-1 to (Æ)-5 (left), (Æ)-6 to (Æ)-10 (middle), and (Æ)-11 to (Æ)-15 (right).
Black=DHfold; gray=ÀTDSfold; checkered=DGfold
.
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Lanzarotti, R. R. Biekofsky, D. A. Estrin, M. A. Marti, A. G. Turjanski, J.
Chem. Inf. Model. 2011, 51, 1623–1633; d) K. E. Riley, P. Hobza, Acc.
Chem. Res. 2013, 46, 927–936.
level of theory were carried out on model systems for the
edge- and face rings in the balances and based on the com-
bined experimental/computational data, we draw the follow-
ing conclusions.
[8] a) F. Cozzi, M. Cinquini, R. Annunziata, T. Dwyer, J. S. Siegel, J. Am. Chem.
Soc. 1992, 114, 5729–5733; b) F. Cozzi, M. Cinquini, R. Annunziata, J. S.
Siegel, J. Am. Chem. Soc. 1993, 115, 5330–5331; c) F. Cozzi, F. Ponzini, R.
Annunziata, M. Cinquini, J. S. Siegel, Angew. Chem. 1995, 107, 1092–
1094; Angew. Chem. Int. Ed. Engl. 1995, 34, 1019–1020; d) F. Cozzi, J. S.
Siegel, Pure Appl. Chem. 1995, 67, 683–689; e) M. J. Rashkin, M. L.
Waters, J. Am. Chem. Soc. 2002, 124, 1860–1861; f) B. W. Gung, M. Patel,
X. Xue, J. Org. Chem. 2005, 70, 10532–10537; g) B. W. Gung, X. Xue,
H. J. Reich, J. Org. Chem. 2005, 70, 3641–3644; h) X. Mei, C. Wolf, J. Org.
Chem. 2005, 70, 2299–2305; i) S. E. Snyder, P. I. Volkers, D. A. Engebret-
son, W. Lee, W. H. Pirkle, J. R. Carey, Org. Lett. 2007, 9, 2341–2343; j) F.
Cozzi, R. Annunziata, M. Benaglia, K. K. Baldridge, G. Aguirre, J. Estrada,
Y. Sritana-Anant, J. S. Siegel, Phys. Chem. Chem. Phys. 2008, 10, 2686–
2694; k) B. W. Gung, B. U. Emenike, C. N. Alvarez, J. Rakovan, K. Kirsch-
baum, N. Jain, Tetrahedron Lett. 2010, 51, 1648–1650; l) C. R. Martinez,
B. L. Iverson, Chem. Sci. 2012, 3, 2191–2201; m) S. E. Snyder, B.-S.
Huang, Y.-T. Chen, H.-S. Lin, J. R. Carey, Org. Lett. 2012, 14, 3442–3445;
n) J. L. Xia, S. H. Liu, F. Cozzi, M. Mancinelli, A. Mazzanti, Chem. Eur. J.
2012, 18, 3611–3620; o) W. B. Jennings, N. O’Connell, J. F. Malone, D. R.
Boyd, Org. Biomol. Chem. 2013, 11, 5278–5291.
[9] a) T. H. Webb, C. S. Wilcox, J. Org. Chem. 1990, 55, 363–365; b) S. Paliw-
al, S. Geib, C. S. Wilcox, J. Am. Chem. Soc. 1994, 116, 4497–4498; c) E.-i.
Kim, S. Paliwal, C. S. Wilcox, J. Am. Chem. Soc. 1998, 120, 11192–11193.
[10] a) F. Hof, D. M. Scofield, W. B. Schweizer, F. Diederich, Angew. Chem.
2004, 116, 5166–5169; Angew. Chem. Int. Ed. 2004, 43, 5056–5059;
b) F. R. Fischer, W. B. Schweizer, F. Diederich, Angew. Chem. 2007, 119,
8418–8421; Angew. Chem. Int. Ed. 2007, 46, 8270–8273.
[11] F. R. Fischer, P. A. Wood, F. H. Allen, F. Diederich, Proc. Natl. Acad. Sci.
USA 2008, 105, 17290–17294.
[12] a) S. L. Cockroft, C. A. Hunter, Chem. Commun. 2006, 3806–3808; b) S. L.
Cockroft, C. A. Hunter, Chem. Commun. 2009, 3961–3963.
1) The electrostatic component resulting from the edge CÀ
H···pface interaction is apparent in the comparison be-
tween the folding free enthalpies of the 4’’-substituted
series versus the 2’’- and 3’’-series, but does not seem the
major contributor to the driving force for folding.
2) A major part of DGfold presumably originates from direct
interactions of the 4’’-F and 4’’-CF3 on the edge ring with
the enhanced p-electron density of the face component
upon changing its substituent from NO2 to NH2. This is in
agreement with recent computational predictions.[14a,-
c,e–h,15b–f]
3) The interaction of the local dipoles of the edge- and face-
ring dipoles can be further invoked to explain the steep
linear correlation in the 4’’-substituted edge-ring series,
but the fact that the folding in the 2’’- and 3’’-F or the
corresponding CF3 series is largely independent of the
face substituent, casts doubt on the importance of this
local dipole–dipole interaction.
The folding equilibria of the described torsion balances sug-
gest that the modulation by the face substituent most
likely stems from local direct interactions between the 4’’-
substituent and the face component p-system and its
substituent.
[13] F. R. Fischer, W. B. Schweizer, F. Diederich, Chem. Commun. 2008, 4031–
4033.
[14] a) S. E. Wheeler, K. N. Houk, J. Am. Chem. Soc. 2008, 130, 10854–10855;
b) S. E. Wheeler, K. N. Houk, J. Chem. Theory Comput. 2009, 5, 2301–
2312; c) S. E. Wheeler, K. N. Houk, Mol. Phys. 2009, 107, 749–760; d) S. E.
Wheeler, A. J. McNeil, P. Mꢄller, T. M. Swager, K. N. Houk, J. Am. Chem.
Soc. 2010, 132, 3304–3311; e) R. K. Raju, J. W. G. Bloom, Y. An, S. E.
Wheeler, ChemPhysChem 2011, 12, 3116–3130; f) S. E. Wheeler, J. Am.
Chem. Soc. 2011, 133, 10262–10274; g) S. E. Wheeler, CrystEngComm
2012, 14, 6140–6145; h) S. E. Wheeler, Acc. Chem. Res. 2013, 46, 1029–
1038.
Acknowledgements
We acknowledge the ETH Research Council and F. Hoffmann-
La Roche Ltd., Basel for funding and Dr. Aaron D. Finke for
help with the X-ray crystal structures.
[15] a) M. O. Sinnokrot, C. D. Sherrill, J. Am. Chem. Soc. 2004, 126, 7690–
7697; b) A. L. Ringer, M. O. Sinnokrot, R. P. Lively, C. D. Sherrill, Chem.
Eur. J. 2006, 12, 3821–3828; c) M. O. Sinnokrot, C. D. Sherrill, J. Phys.
Chem. A 2006, 110, 10656–10668; d) S. A. Arnstein, C. D. Sherrill, Phys.
Chem. Chem. Phys. 2008, 10, 2646–2655; e) A. L. Ringer, C. D. Sherrill, J.
Am. Chem. Soc. 2009, 131, 4574–4575; f) E. G. Hohenstein, J. Duan,
C. D. Sherrill, J. Am. Chem. Soc. 2011, 133, 13244–13247; g) C. D. Sherrill,
Acc. Chem. Res. 2013, 46, 1020–1028.
[16] a) J. Ribas, E. Cubero, F. J. Luque, M. Orozco, J. Org. Chem. 2002, 67,
7057–7065; b) E. C. Lee, B. H. Hong, J. Y. Lee, J. C. Kim, D. Kim, Y. Kim, P.
Tarakeshwar, K. S. Kim, J. Am. Chem. Soc. 2005, 127, 4530–4537; c) K. E.
Riley, K. M. Merz Jr., J. Phys. Chem. B 2005, 109, 17752–17756; d) S. Tsu-
zuki, K. Honda, T. Uchimaru, M. Mikami, J. Chem. Phys. 2006, 125,
124304; e) S. Tsuzuki, T. Uchimaru, Curr. Org. Chem. 2006, 10, 745–762.
[17] a) C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112, 5525–
5534; b) C. A. Hunter, J. Singh, J. M. Thornton, J. Mol. Biol. 1991, 218,
837–846; c) C. A. Hunter, Angew. Chem. 1993, 105, 1653–1655; Angew.
Chem. Int. Ed. Engl. 1993, 32, 1584–1586; d) C. A. Hunter, Chem. Soc.
Rev. 1994, 23, 101–109; e) C. A. Hunter, K. R. Lawson, J. Perkins, C. J.
Urch, J. Chem. Soc. Perkin Trans. 2 2001, 651–669.
Keywords: aromatic edge-to-face interactions · dipole–dipole
interactions · fluorine · molecular torsion balance · substituent
effects · Trçger base
[1] a) E. A. Meyer, R. K. Castellano, F. Diederich, Angew. Chem. 2003, 115,
1244–1287; Angew. Chem. Int. Ed. 2003, 42, 1210–1250; b) L. M. Salo-
nen, M. Ellermann, F. Diederich, Angew. Chem. 2011, 123, 4908–4944;
Angew. Chem. Int. Ed. 2011, 50, 4808–4842.
[2] a) L. Mosca, P. Koutnꢂk, V. M. Lynch, G. V. Zyryanov, N. A. Esipenko, P. An-
zenbacher Jr., Cryst. Growth Des. 2012, 12, 6104–6109; b) F.-G. Klꢃrner,
T. Schrader, Acc. Chem. Res. 2013, 46, 967–978; c) H.-J. Schneider, Acc.
Chem. Res. 2013, 46, 1010–1019.
[3] a) C. G. Claessens, J. F. Stoddart, J. Phys. Org. Chem. 1997, 10, 254–272;
b) H.-J. Schneider, Angew. Chem. 2009, 121, 3982–4036; Angew. Chem.
Int. Ed. 2009, 48, 3924–3977.
[4] a) G. Santoni, M. Mba, M. Bonchio, W. A. Nugent, C. Zonta, G. Licini,
Chem. Eur. J. 2010, 16, 645–654; b) E. H. Krenske, K. N. Houk, Acc. Chem.
Res. 2013, 46, 979–989.
[5] H. Nakajima, M. Yasuda, R. Takeda, A. Baba, Angew. Chem. 2012, 124,
3933–3936; Angew. Chem. Int. Ed. 2012, 51, 3867–3870.
[6] A. Di Fenza, A. Heine, U. Koert, G. Klebe, ChemMedChem 2007, 2, 297–
308.
[7] a) S. K. Burley, G. A. Petsko, Adv. Protein Chem. 1988, 39, 125–189;
b) S. K. Burley, G. A. Petsko, Trends Biotechnol. 1989, 7, 354–359; c) E.
[18] In this study, we define as direct interaction the interaction between
edge substituents and the (substituted)face component. For various re-
finements of this definition see references [14a,c,e,15b,e,f].
[19] M. Harder, B. Kuhn, F. Diederich, ChemMedChem 2013, 8, 397–404.
[20] a) A. P. Bisson, F. J. Carver, C. A. Hunter, J. P. Waltho, J. Am. Chem. Soc.
1994, 116, 10292–10293; b) H. Adams, F. J. Carver, C. A. Hunter, J. C.
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Morales, E. M. Seward, Angew. Chem. 1996, 108, 1628–1631; Angew.
Chem. Int. Ed. Engl. 1996, 35, 1542–1544; c) F. J. Carver, C. A. Hunter,
E. M. Seward, Chem. Commun. 1998, 775–776; d) H. Adams, J.-L. J.
Blanco, G. Chessari, C. A. Hunter, C. M. R. Low, J. M. Sanderson, J. G.
Vinter, Chem. Eur. J. 2001, 7, 3494–3503; e) H. Adams, C. A. Hunter, K. R.
Lawson, J. Perkins, S. E. Spey, C. J. Urch, J. M. Sanderson, Chem. Eur. J.
2001, 7, 4863–4878; f) F. J. Carver, C. A. Hunter, P. S. Jones, D. J. Living-
stone, J. F. McCabe, E. M. Seward, P. Tiger, S. E. Spey, Chem. Eur. J. 2001,
7, 4854–4862; g) F. J. Carver, C. A. Hunter, D. J. Livingstone, J. F.
McCabe, E. M. Seward, Chem. Eur. J. 2002, 8, 2847–2859; h) G. Chessari,
C. A. Hunter, C. M. R. Low, M. J. Packer, J. G. Vinter, C. Zonta, Chem. Eur.
J. 2002, 8, 2860–2867; i) S. L. Cockroft, C. A. Hunter, K. R. Lawson, J. Per-
kins, C. J. Urch, J. Am. Chem. Soc. 2005, 127, 8594–8595; j) S. L. Cockroft,
C. A. Hunter, Chem. Soc. Rev. 2007, 36, 172–188; k) S. L. Cockroft, J. Per-
kins, C. Zonta, H. Adams, S. E. Spey, C. M. R. Low, J. G. Vinter, K. R.
Lawson, C. J. Urch, C. A. Hunter, Org. Biomol. Chem. 2007, 5, 1062–
1080.
[24] a) I. K. Mati, C. Adam, S. L. Cockroft, Chem. Sci. 2013, 4, 3965–3972;
b) K. B. Muchowska, C. Adam, I. K. Mati, S. L. Cockroft, J. Am. Chem. Soc.
2013, 135, 9976–9979; c) L. Yang, C. Adam, G. S. Nichol, S. L. Cockroft,
Nat. Chem. 2013, 5, 1006–1010.
[25] M. Head-Gordon, J. A. Pople, M. J. Frisch, Chem. Phys. Lett. 1988, 153,
503–506.
[26] C. M. Breneman, K. B. Wiberg, J. Comput. Chem. 1990, 11, 361–373.
[27] a) F. Martin, H. Zipse, J. Comput. Chem. 2005, 26, 97–105; b) G. S.
Maciel, E. Garcia, Chem. Phys. Lett. 2005, 409, 29–33.
[28] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195.
[29] CCDC-957749 ((Æ)-31), CCDC-957750 ((Æ)-27), CCDC-957751 ((Æ)-26),
CCDC-957752 ((Æ)-5), CCDC-957753 ((Æ)-10), CCDC-957754 ((Æ)-20),
CCDC-957755 ((Æ)-1), CCDC-957756 ((Æ)-4), CCDC-957757 ((Æ)-15), and
CCDC-957758 ((Æ)-25) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[21] I. Sucholeiki, V. Lynch, L. Phan, C. S. Wilcox, J. Org. Chem. 1988, 53, 98–
104.
Received: December 9, 2013
[22] A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
[23] J.-D. Chai, M. Head-Gordon, J. Chem. Phys. 2008, 128, 084106.
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Direct interactions dominate: The im-
portance of local direct interactions in
the edge-to-face interaction was exam-
ined with a large library of Trçger base
derived molecular torsion balances.
Modulation of the folding free enthalpy
is only observed when edge-ring sub-
stituents are capable of directly interact-
ing with the face-ring p system (see
figure). Experimental results and theo-
retical calculations indicate that local
direct interactions make the largest con-
tribution to substituent effects in the
edge-to-face aromatic interactions.
&
Aromatic Interactions
H. Gardarsson, W. B. Schweizer, N. Trapp,
F. Diederich*
&& – &&
Structures and Properties of Molecular
Torsion Balances to Decipher the
Nature of Substituent Effects on the
Aromatic Edge-to-Face Interaction
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!