Direct Experimental Evidence for Halogen–Aryl π Interactions in Solution from Molecular Torsion Balances

Article abstract of DOI:10.1002/anie.201700520

We dissected halogen–aryl π interactions experimentally using a bicyclic N-arylimide based molecular torsion balances system, which is based on the influence of the non-bonded interaction on the equilibria between folded and unfolded states. Through comparison of balances modulated by higher halogens with fluorine balances, we determined the magnitude of the halogen–aryl π interactions in our unimolecular systems to be larger than ?5.0 kJ mol^?1, which is comparable with the magnitude estimated in the biomolecular systems. Our study provides direct experimental evidence of halogen–aryl π interactions in solution, which until now have only been revealed in the solid state and evaluated theoretically by quantum-mechanical calculations.

Full text of DOI:10.1002/anie.201700520

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201700520
German Edition: DOI: 10.1002/ange.201700520
Molecular Balances
Direct Experimental Evidence for Halogen–Aryl p Interactions in
Solution from Molecular Torsion Balances
Han Sun+, Andrꢀ Horatscheck+, Vera Martos, Max Bartetzko, Ulrike Uhrig, Dieter Lentz,
Peter Schmieder,* and Marc Nazarꢀ*
Abstract: We dissected halogen–aryl p interactions experi-
mentally using a bicyclic N-arylimide based molecular torsion
balances system, which is based on the influence of the non-
bonded interaction on the equilibria between folded and
unfolded states. Through comparison of balances modulated
by higher halogens with fluorine balances, we determined the
magnitude of the halogen–aryl p interactions in our uni-
molecular systems to be larger than À5.0 kJmolÀ1, which is
comparable with the magnitude estimated in the biomolecular
systems. Our study provides direct experimental evidence of
halogen–aryl p interactions in solution, which until now have
only been revealed in the solid state and evaluated theoretically
by quantum-mechanical calculations.
regions as a result of the anisotropic charge distribution along
À
the C X axis, the s hole, whereas there is no such anisotropy
in the valence-shell charge concentration of fluorine.[2]
Although various theoretical investigations provided a good
basis for understanding halogen–aryl p interactions, there is
still no direct experimental evidence for such interactions in
solution, because in the proteinaceous environment a multi-
tude of contacts and synergistic interactions in protein–ligand
complexes renders the situation highly complex; in particular,
free-energy changes in protein–ligand systems are often
governed by entropic contributions, such as solvent effects.[3]
Nevertheless, direct experimental evidence and determina-
tion of the magnitude of halogen–aryl p interactions in
solution is essential for any rational approach to drug design
and would provide an experimental basis for predictions and
theoretical calculations.
H
alogen–aryl p interactions have received increasing atten-
tion with respect to molecular recognition in chemical and
biological systems, as a large number of studies have shown
that nonbonding halogen interactions contribute significantly
to high binding affinity in rational drug design and lead
optimization.[1] For a long time, however, only steric and
lipophilic contributions of halogens were considered in ligand
binding. The situation changed when different theoretical
investigations revealed that higher halogen atoms (Cl, Br, and
I) could form an attractive interaction with electronegative
In this study, we aimed to dissect the halogen–aryl
p interactions experimentally by using molecular torsion
balances, which are based on the influence of the nonbonding
interactions on the equilibria between folded and unfolded
states in a unimolecular system.[4] Molecular balances have
proven to be a reliable model system to quantify weak
noncovalent interactions, since this unimolecular system has
minimal entropic penalty associated with the intermolecular
association, and no other perturbations influence the overall
stability.[5] We adopted herein N-aryl imide based balances,
which were originally proposed by Shimizu and co-workers
and have been used to determine different nonbonding
interactions, such as CH–p, CD–p, cation–p, p–p, and
metal–p interactions.[6]
To verify the formation of the halogen–aryl p interactions,
we synthesized three N-aryl imide based balances with
increasing size of the aromatic shelf, which was benzene in
1, phenanthrene in 2, and pyrene in 3. Each balance was
modulated with different halogen atoms (F, Cl, Br, and I) at
the ortho position of the N-aryl group (Scheme 1). Despite
the fact that the observed free energy DG decomposes into an
enthalpic interaction term and an entropic contribution, we
expected higher halogens to form attractive interactions with
the aromatic p faces in balances 1–3, whereas fluorine does
not engage in such interactions because of its generally highly
electronegative character and very low polarizability.[2d,3b,7]
Ethylene balances 4 served as reference balances, as both van
der Waals and electrostatic interactions between halogens
and aromatic p faces are absent.
[*] Dr. H. Sun,[+] Dr. A. Horatscheck,[+] Dr. V. Martos, M. Sc. M. Bartetzko,
Dr. P. Schmieder, Dr. M. Nazarꢀ
Departments of Chemical Biology and Structural Biology
Leibniz-Institut f r Molekulare Pharmakologie (FMP)
Campus Berlin-Buch
Robert-Roessle-Strasse 10, 13125 Berlin (Germany)
E-mail: schmieder@fmp-berlin.de
nazare@fmp-berlin.de
Dr. A. Horatscheck[+]
Drug Discovery and Development Centre (H3D)
Department of Chemistry, University of Cape Town
Rondebosch 7701 (South Africa)
M. Sc. M. Bartetzko
Max-Planck-Institut fꢁr Kolloid- und Grenzflꢂchenforschung
Am Mꢁhlenberg 1, 14476 Potsdam (Germany)
Dr. U. Uhrig
European Molecular Biology Laboratory (EMBL)
Chemical Biology Core Facility
Meyerhofstrasse 1, 69117 Heidelberg (Germany)
Prof. Dr. D. Lentz
Institut fꢁr Chemie und Biochemie
Anorganische Chemie, Freie Universitꢂt Berlin
Fabeckstrasse 34–36, 14195 Berlin (Germany)
Determination of the conformations of both folded and
unfolded states in solution is challenging by X-ray crystallog-
raphy. Quite often only the crystal structure of the unfolded
conformation could be established.[6b] In this investigation, we
[+] These authors contributed equally.
Supporting information for this article can be found under:
https://doi.org/10.1002/anie.201700520.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 1. a,b) Conformations of the unfolded and folded states of 1-Cl,
as determined by quantum-mechanical calculations and RDC-based
NMR spectroscopy (Cl atom: yellow). c) X-ray crystallographic struc-
ture of an indole-2-carboxyamide in complex with human factor Xa
(PDB: 2BQW) showing the Cl–aryl p interaction observed in the
biological system.[11]
that separate peaks could be observed for both states. The
unfolded rotamer could be readily identified by the upfield
shift of the ortho hydrogen atom H-7 (numbering shown in
Figure 1 and Figure S1 of the Supporting Information) of the
halobenzene ring as a result of the ring-current effect of the
aromatic shelf. Furthermore, we computed the energy differ-
ences of folded and unfolded conformers by DFT calculations
at the M06-2X/cc-pVDZ level of theory,[12] which was also
used for geometry optimization (for iodinated compounds,
see the Supporting Information; see Figure S5–S12 for
optimized structures of all balances). Overall, the computed
energy differences between folded and unfolded conformers
are in a good agreement with the experimental DGfold values
for all balances 1–3 (Table 1), which suggests an overall low
entropic contribution to the conformational equilibria in the
balances with arene shelves.
As shown in Scheme 1, the weighting of the steric
repulsive interaction and the attractive halogen–aryl p inter-
action determines the population ratio of folded and unfolded
states. The experimental and theoretically calculated DGfold
values (Table 1) are positive for all aromatic balances, thus
indicating that repulsive steric energy between the halogen
atom and the aromatic shelf is dominant over other inter-
actions. To dissect the halogen–aryl p interaction from other
energy contributors, we compared the DGfold values of higher-
halogen balances with the fluorine balances, as theoretically
no significant fluorine–aryl interaction is expected.[2d,3b,7]. The
DGfold values of the arene-shelf balances modulated with
fluorine and higher halogens were first plotted within each
series against the van der Waals (vdW) radius (F: 1.47 ꢀ, Cl:
1.75 ꢀ, Br: 1.85 ꢀ, I: 1.98 ꢀ).[13] Interestingly, we found that
in the whole aromatic series (balances 1–3) the DGfold value
increased nearly linearly from chlorine to iodine (R > 0.98),
whereas the values for the fluorine balances deviated
significantly from the line (Figure 2a–c). This result strongly
suggests that different interactions are involved in fluorine
and higher-halogen balances.
Scheme 1. a) Conformational equilibrium of the unfolded and folded
states of the molecular torsion balances used for determining the
nonbonding halogen aromatic interactions. b) Molecular balances 1–4
with a benzene, phenanthrene, pyrene, and ethylene shelf, respectively.
decided to take advantage of the method based on residual
dipolar coupling (RDC) to study both conformations, as the
RDC is a powerful NMR parameter that provides long-range
structural information on internuclear vector orientation.[8]
RDC-enhanced NMR spectroscopy has emerged as an
important technique for the determination of the conforma-
tion and relative configuration of structurally challenging
systems.[9] In this study, we aligned 1-Cl in a cross-linked
1
polyacrylamide-based (PH) gel[10] and obtained eight DCH
values for each state (folded and unfolded). The excellent fit
between theoretically predicted RDCs for the conformations
calculated by DFT computations and the experimental values
(Q = 0.12 for both folded and unfolded states) confirmed the
quality of the DFT-computed structures and allowed us to
employ them as well as structures of other balances in the
energy calculations (Figure 1; see also Table S4 in the
Supporting Information).
For the folded state of 1-Cl, the distance between the
halogen atom and the arene shelf is 3.3 ꢀ (Figure 1b), which
is slightly lower than the value of 3.5 ꢀ at which the minimum
distance of the Cl–aryl p interaction was previously estab-
lished by MP2 calculations.[1d] Distances of 3.5–3.7 ꢀ between
the halogen and the centroid of the aromatic ring were
determined in a series of factor Xa ligand complexes (PDB:
2BQW, 4A7I, 2BOH, 2BMG), which were used extensively as
model systems to characterize the halogen–aryl p interactions
in the proteinaceous environment (Figure 1c).[11]
The free-energy difference between the observed folded
and unfolded states was determined from the ratio of the
corresponding rotamer peaks in the 1H NMR spectrum, as the
transition between the two conformations was slow enough
By using the linear regression of the DGfold values
obtained for higher-halogen balances, we extrapolated the
DGfold values for fluorine (denoted as DGfold,pred->F) under the
assumption that fluorine balances share the same interaction
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: Comparison of the folded/unfolded ratios at 288 K in CDCl3,
with fluorine, thus suggesting a contribution of attractive
experimentally determined DGfold values, and computed DEfold values
(from DFT calculations) for the balances with different aromatic shelves
and halogens.[a]
halogen–aryl interactions that results in an overall decrease in
DGfold in higher-halogen balances. The difference between the
predicted DGfold,pred->F value and experimental the DGfold
value for fluorine, denoted as DDG’, can be considered as
an approximate measure for the magnitude of the halogen–
aryl p interactions. The calculated DDG’ values are highly
consistent in all arene-shelf balances: À4.8 kJmolÀ1 in ben-
zene 1, À4.7 kJmolÀ1 in phenanthrene 2, and À5.1 kJmolÀ1 in
pyrene 3 (Figure 2a–c). Additionally, we excluded the
possibility that our observations rely on solvent effects by
measuring phenanthrene balances 2 in acetonitrile, which
resulted in an even larger DDG’ value of À6.5 kJmolÀ1 (see
Figure S13).
It is highly interesting that the DGfold values of higher-
halogen balances show a linear relationship with the atomic
radius. In particular, when the DGfold value of benzene
balance 1 was subtracted from those of 2 and 3, the DDGfold
values decreased linearly from fluorine to iodine for both
balance pairs (see Figure S14). The linear relationship
between the free-energy changes contributed by steric effects
and vdW radii was established by Taft and Charton in the
1952 and 1975.[14] The linear correlation observed in our
system between the DG value and the vdW radius can thus be
considered as a measure of repulsive energy. Nevertheless,
previous theoretical calculations suggested an increasing
trend of attractive halogen–aryl p interactions when going
from chlorine to iodine.[15] Consequently, it is possible that the
DG values associated with the attractive halogen–aryl p
interactions decrease linearly with the atomic size, or their
relationship has only a small nonlinear term. Hence, we
deduce that the magnitude of halo-
[b]
Balance
Keq
DGfold, exp
DEfold, calc
[kJmolÀ1
]
[kJmolÀ1
]
1 (benzene)
F
0.265
0.087
0.037
0.007
3.2
5.8
7.9
3.2
5.3
6.7
Cl
Br
I
11.8
12.6
2 (phenanthrene)
3 (pyrene)
F
0.340
0.469
0.295
0.130
2.6
1.8
2.9
4.9
4.1
1.6
0.4
4.8
Cl
Br
I
F
0.410
0.532
0.346
0.141
2.1
1.5
2.5
4.7
3.7
1.3
0.9
3.8
Cl
Br
I
4 (ethylene)
F
1.267
0.644
0.412
0.191
À0.6
1.1
2.1
À4.6
1.7
1.3
Cl
Br
I
4.0
4.2
[a] For error analysis, see Table S1 in the Supporting Information. [b] The
folded/unfolded ratio was determined by integration of 1H NMR signals
for both states.
characteristics as the higher-halogen balances. Considerably
lower predicted DGfold,pred->F values than the experimentally
determined values were found for all arene-shelf balances
gen–aryl p interactions for the
higher-halogen balances is larger
than the DDG’ values derived as
described above, as the DDG’ values
were predicted for fluorine balances
on the assumption that the fluorine
balances exhibit the same linear
relationship as the higher-halogen
balances.
To further corroborate the sig-
nificance of the halogen–aryl p in-
teractions observed in the arene-
shelf balances, we plotted the DGfold
values of ethylene balances modu-
lated with fluorine and higher hal-
ogens against the vdW radius (Fig-
ure 2d). Ethylene balances 4 were
chosen in this case as reference
balances, as both van der Waals
and halogen–aryl interactions are
supposed to be absent owing to the
lack of an aromatic p face. Indeed,
with a DDG’ value of À2.0 kJmolÀ1,
we observed a considerably smaller
Figure 2. Correlation plots of the experimental DGfold value against the vdW radius for a) benzene
balance 1, b) phenanthrene balance 2, c) pyrene balance 3, and d) ethylene balance 4. From the
linear fits obtained for the higher-halogen balances, values were predicted for the corresponding
fluorine balances (depicted in blue).
difference between the predicted
DGfold,pred->F value and the experi-
mentally determined value. Inter-
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
À
estingly, although the C F axis was not directed towards the
ethylene p surface in the ethylene balance (see Figures S11
and S12), a small DDG’ value, which was attributed to the
halogen–p interactions between higher halogens and the
p surface in ethylene, could still be identified. Furthermore,
we found that the magnitude of the halogen–aryl interactions
in all three aromatic balances were highly consistent,
Acknowledgements
We thank Brigitte Schlegel, Nils Trieloff, and Keven Mallow
for their excellent technical assistance, Barth-Jan van Rossum
for the graphical artwork, and Prof. Dr. Christoph A. Schalley
for helpful discussions.
À
although the s holes along the C X bond are in a more
favorable perpendicular geometry with respect to the aro-
matic p faces in the phenanthrene and pyrene balances as
compared to the benzene balances. On the basis of these two
observations, we hypothesize that halogen–aryl p interactions
are less directional than suggested by most theoretical
investigations,[15b,c] possibly as a result of additional induced
polarization of the halogen atom owing to contact with the
electron-rich p system, as suggested previously by DFT
calculations.[16] Furthermore, this finding is in good agreement
with previous surveys on protein databases showing that
halogen–aryl p interactions exhibit only a certain degree of
directionality in the biomolecular system.[1c,11a]
Conflict of interest
The authors declare no conflict of interest.
Keywords: halogen bonding · halogen–aryl p interactions ·
molecular balances · molecular recognition ·
residual dipolar coupling
[1] a) R. Wilcken, M. O. Zimmermann, A. Lange, A. C. Joerger,
F. M. Boeckler, J. Med. Chem. 2013, 56, 1363 – 1388; b) L. A.
Hardegger, B. Kuhn, B. Spinnler, L. Anselm, R. Ecabert, M.
Stihle, B. Gsell, R. Thoma, J. Diez, J. Benz, J.-M. Plancher, G.
Hartmann, D. W. Banner, W. Haap, F. Diederich, Angew. Chem.
Int. Ed. 2011, 50, 314 – 318; Angew. Chem. 2011, 123, 329 – 334;
c) P. Auffinger, F. A. Hays, E. Westhof, P. S. Ho, Proc. Natl.
Acad. Sci. USA 2004, 101, 16789 – 16794; d) H. Matter, M.
Nazarꢁ, S. Gꢂssregen in The Importance of Pi-Interactions in
Crystal Engineering (Eds.: E. R. T. Tiekink, J. Zukerman-
Schpector), Wiley, Chichester, 2012, pp. 187-132; e) P. M.
McTamney, S. E. Rokita, J. Am. Chem. Soc. 2009, 131, 14212 –
14213; f) M. Hammami, E. R. Uhmann, E. Maurer, A. Heine,
M. G. Utschow, G. Klebe, T. Steinmetzer, Chem. Med. Commun.
2012, 3, 807 – 813; g) Z. Xu, Y. Yang, Y. Liu, Y. Lu, K. Chen, W.
Zhu, J. Chem. Inf. Model. 2014, 54, 69 – 78.
[2] a) T. Clark, M. Hennemann, J. S. Murray, P. Politzer, J. Mol.
Model. 2007, 13, 291 – 296; b) H. Wang, W. Wang, W. J. Jin,
Chem. Rev. 2016, 116, 5072 – 5104; c) S. Gꢂssregen, H. Matter, G.
Hessler, M. Mꢂller, F. Schmidt, T. Clark, J. Chem. Inf. Model.
2012, 52, 2441 – 2453; d) G. Cavallo, P. Metrangolo, R. Milani, T.
Pilati, A. Priimagi, G. Resnati, G. Terraneo, Chem. Rev. 2016,
116, 2478 – 2601.
[3] a) M. Neeb, M. Betz, A. Heine, L. J. Barandun, C. Hohn, F.
Diederich, G. Klebe, J. Med. Chem. 2014, 57, 5566 – 5578; b) E.
Persch, O. Dumele, F. Diederich, Angew. Chem. Int. Ed. 2015,
54, 3290 – 3327; Angew. Chem. 2015, 127, 3341 – 3382; c) B.
Breiten, M. R. Lockett, W. Sherman, S. Fujita, M. Al-Sayah, H.
Lange, C. M. Bowers, A. Heroux, G. Krilov, G. M. Whitesides, J.
Am. Chem. Soc. 2013, 135, 15579 – 15584.
[4] a) T. H. Webb, C. S. Wilcox, J. Org. Chem. 1990, 55, 363 – 365;
b) S. Paliwal, 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; d) B. W. Gung, A. X. Xue, H. J.
Reich, J. Org. Chem. 2005, 70, 7232 – 7237.
Although it is impossible to directly determine the
halogen–aryl interactions in biological systems, comparison
of the affinities (Ki or IC50) of halogenated and nonhalo-
genated inhibitors toward an identical protein provides an
indirect measure of halogen–aryl interactions in protein
environments. In our previous study with fXa inhibitors, the
addition of chlorine to the ligand structure led to an estimated
change in the free energy of binding (DDG) of À10.5 and
À8.5 kJmolÀ1 in the indole-2-carboxamide and 3-oxybenz-
amide series, respectively.[11a] In contrast, a fluorine atom at
the same position had a significantly lower effect, with an
DDG value of only À3 kJmolÀ1. The difference in DDG
between chlorine- and fluorine-substituted ligands, À7.5 and
À5.0 kJmolÀ1 for the indole-2-carboxamide and 3-oxybenz-
amide series, respectively, can be considered as approximate
measures of halogen–aryl interactions in the fXa protein–
ligand system. In our current study, we determined the
magnitude of halogen–aryl interactions in the unimolecular
systems to be larger than À5.0 kJmolÀ1, which is in good
agreement with the experimentally estimated values in the
fXa systems.
In conclusion, we employed N-aryl imide based molecular
torsion balances to probe halogen–aryl interactions exper-
imentally. These interactions are difficult to assess in a pro-
teinaceous environment. The comparison of higher-halogen-
containing balances with fluorine-containing reference bal-
ances revealed a clear attractive contribution of the halogen–
aryl p interactions in solution. Until now these effects had
only been inferred from crystallographic evidence and
theoretically evaluated by quantum-mechanical calculations.
We estimated the magnitude of the halogen–aryl p interac-
tions in our unimolecular systems to be larger than
À5.0 kJmolÀ1. Our study provides an experimental basis for
theoretical investigations of halogen–aryl p interactions and
further enabled us to directly compare the magnitude of such
interactions with estimated values from affinity measure-
ments of protein–ligand systems.
[5] a) M. Oki, Acc. Chem. Res. 1990, 23, 351 – 356; b) I. K. Mati,
S. L. Cockroft, Chem. Soc. Rev. 2010, 39, 4195 – 4205; c) A. E.
Aliev, J. R. T. Arendorf, I. Pavlakos, R. B. Moreno, M. J. Porter,
H. S. Rzepa, W. B. Motherwell, Angew. Chem. Int. Ed. 2015, 54,
551; Angew. Chem. 2015, 127, 561; d) F. Hof, D. M. Scofield,
W. B. Schweizer, F. Diederich, Angew. Chem. Int. Ed. 2004, 43,
5056 – 5059; Angew. Chem. 2004, 116, 5166 – 5169; e) F. R.
Fischer, W. B. Schweizer, F. Diederich, Angew. Chem. Int. Ed.
2007, 46, 8270 – 8273; Angew. Chem. 2007, 119, 8418 – 8421;
f) F. R. Fischer, P. A. Wood, F. H. Allen, F. Diederich, Proc. Natl.
Acad. Sci. USA 2008, 105, 17290 – 17294.
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[6] a) W. R. Carroll, P. J. Pellechia, K. D. Shimizu, Org. Lett. 2008,
[11] a) H. Matter, M. Nazarꢁ, S. Gꢂssregen, D. W. Will, H. Schreuder,
10, 3547 – 3550; b) W. R. Carroll, C. Zhao, M. D. Smith, P. J.
Pellechia, K. D. Shimizu, Org. Lett. 2011, 13, 4320 – 4323; c) P. Li,
C. Zhao, M. D. Smith, K. D. Shimizu, J. Org. Chem. 2013, 78,
5303 – 5313; d) J. Hwang, P. Li, W. R. Carroll, P. J. Pellechia,
K. D. Shimizu, J. Am. Chem. Soc. 2014, 136, 14060 – 14067;
e) J. M. Maier, P. Li, J. Hwang, M. D. Smith, K. D. Shimizu, J.
Am. Chem. Soc. 2015, 137, 8014 – 8017.
A. Bauer, M. Urmann, K. Ritter, M. Wagner, V. Wehner, Angew.
Chem. Int. Ed. 2009, 48, 2911 – 2916; Angew. Chem. 2009, 121,
2955 – 2960; b) M. Nazarꢁ, D. W. Will, H. Matter, H. Schreuder,
K. Ritter, M. Urmann, M. Essrich, A. Bauer, M. Wagner, J.
Czech, M. Lorenz, V. Laux, V. Wehner, J. Med. Chem. 2005, 48,
4511 – 4525.
[12] A. Forni, S. Pieraccini, S. Rendine, F. Gabas, M. Sironi,
ChemPhysChem 2012, 13, 4224 – 4234.
[13] A. J. Bondi, Phys. Chem. 1964, 68, 441 – 451.
[14] a) R. W. Taft, J. Am. Chem. Soc. 1952, 74, 3120 – 3128; b) M.
Charton, J. Am. Chem. Soc. 1975, 97, 1552 – 1556; c) M. Charton,
J. Am. Chem. Soc. 1975, 97, 3691 – 3693; d) M. Charton, J. Am.
Chem. Soc. 1975, 97, 3694 – 3697.
[7] a) S. Sirimulla, J. B. Bailey, R. Vegesna, M. Narayan, J. Chem.
Inf. Model. 2013, 53, 2781 – 2791; b) T. M. Beale, M. G. Chud-
zinski, M. G. Sarwara, M. S. Taylor, Chem. Soc. Rev. 2013, 42,
1667 – 1680; c) B. W. Gung, J. C. Amicangelo, J. Org. Chem.
2006, 71, 9261 – 9270; d) K. E. Riley, K. M. Merz, J. Phys. Chem.
B 2005, 109, 17752 – 17756.
[8] F. Kramer, M. V. Deshmukh, H. Kessler, S. J. Glaser, Concepts
Magn. Reson. Part A 2004, 21A, 10 – 21.
[15] a) P. Politzer, J. S. Murray, T. Clark, Phys. Chem. Chem. Phys.
2013, 15, 11178 – 11179; b) S. Tsuzuki, A. Wakisaka, T. Ono, T.
ˇ
ˇ
[9] a) H. Sun, U. M. Reinscheid, E. L. Whitson, E. J. dꢃAuvergne,
C. M. Ireland, A. Navarro-Vꢄzquez, C. Griesinger, J. Am. Chem.
Soc. 2011, 133, 14629 – 14636; b) B. Bçttcher, V. Schmidts, J. A.
Raskatov, C. M. Thiele, Angew. Chem. Int. Ed. 2010, 49, 205 –
209; Angew. Chem. 2010, 122, 210 – 214; c) P. Trigo-MouriÇo,
M. C. de la Fuente, R. R. Gil, V. M. Sanchez-Pedregal, A.
Navarro-Vazquez, Chem. Eur. J. 2013, 19, 14989 – 14997.
[10] P. Haberz, J. Farjon, C. Griesinger, Angew. Chem. Int. Ed. 2005,
44, 427 – 429; Angew. Chem. 2005, 117, 431 – 433.
Sonoda, Chem. Eur. J. 2012, 18, 951 – 960; c) M. Kolꢄr, J. Hostas,
P. Hobza, Phys. Chem. Chem. Phys. 2014, 16, 9987 – 9996.
[16] N. Muzet, B. Guillot, C. Jelsch, E. Howard, C. Lecomte, Proc.
Natl. Acad. Sci. USA 2003, 100, 8742 – 8747.
Manuscript received: January 16, 2017
Final Article published: && &&, &&&&
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Molecular Balances
Weighty evidence: Molecular torsion bal-
ances containing an N-aryl imide and an
additional aromatic moiety provided
direct experimental evidence for halogen–
aryl p interactions in solution (see pic-
ture). The magnitude of the halogen–aryl
p interactions in the unimolecular sys-
tems described herein are found to be
H. Sun, A. Horatscheck, V. Martos,
M. Bartetzko, U. Uhrig, D. Lentz,
P. Schmieder,*
M. Nazarꢁ*
&&&& — &&&&
Direct Experimental Evidence for
Halogen–Aryl p Interactions in Solution
from Molecular Torsion Balances
larger than À5.0 kJmolÀ1
.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!