Stabilizing Fluorine–π Interactions

Article abstract of DOI:10.1002/anie.201702950

A series of N-arylimide molecular balances were designed to study and measure fluorine–aromatic (F–π) interactions. Fluorine substituents gave rise to increasingly more stabilizing interactions with more electron-deficient aromatic surfaces. The attractive F–π interaction is electrostatically driven and is stronger than other halogen–π interactions.

Full text of DOI:10.1002/anie.201702950

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201702950
German Edition: DOI: 10.1002/ange.201702950
Electrostatic Interactions
Stabilizing Fluorine–p Interactions
Ping Li, Josef M. Maier, Erik C. Vik, Christopher J. Yehl, Brent E. Dial, Amanda E. Rickher,
Mark D. Smith, Perry J. Pellechia, and Ken D. Shimizu*
Abstract: A series of N-arylimide molecular balances were
designed to study and measure fluorine–aromatic (F–p)
interactions. Fluorine substituents gave rise to increasingly
more stabilizing interactions with more electron-deficient
aromatic surfaces. The attractive F–p interaction is electro-
statically driven and is stronger than other halogen–p inter-
actions.
O
rganofluorine compounds[1] are widely used in synthesis,[2]
materials,[3] and medicine.[4] The high electronegativity and
small size of the fluorine atom endow organofluorine com-
pounds with unique noncovalent interactions,[5,6] chemical
stability,[7] and distinct conformational preferences.[8] For
example, F–p interactions have been shown to be capable
of controlling the regioselectivity of reactions of aromatic
rings.[9] However, the ability of C F bonds to form attractive
À
interactions with p-systems has been a subject of debate.[10]
Diederich and co-workers observed attractive interactions
Scheme 1. The equilibrium between the unfolded and folded isomers
of the N-arylimide atropisomeric molecular balances for quantitative
comparison of the electrostatic trends of F–p (1), CH–p (2), halogen–
p (3–5), and perfluoroalkyl–p (6) interactions.
[11]
À
=
between C F and C O p-systems experimentally and in
a database survey.[12] However, few studies[13] have examined
the interactions between organofluorides and aromatic sur-
faces (F–p interactions).[14] Therefore, the goal of this work
was to systematically measure the F–p interactions within
a series of N-arylimide “molecular balances”.[15] The ques-
tions addressed were: 1) Can fluorine and organofluorine
substituents form stabilizing interactions with aromatic sur-
faces? 2) What is the nature of the interaction? 3) Are F–p
interactions different from other halogen–p interactions?
The F–p interaction stability trends were measured using
a series of molecular balances 1a–1d (Scheme 1). Restricted
rotation of the N-aryl rotor generates distinct folded and
unfolded conformers in which an intramolecular interaction is
formed and broken. Thus variations in the arm–shelf inter-
action energies can be quantitatively measured by determin-
ing the folded–unfolded equilibrium. The N-arylimide molec-
ular balance model has been successfully employed to study
many noncovalent interactions, including aromatic stack-
ing,[16] CH–p,[17] heterocycle–p,[18] and metal–p interactions.[19]
In this work, a fluorine substituent (X = F) was affixed to the
rotor of balances containing a series of different aromatic
surfaces (1a–1d) of varying electrostatic potential. The
systematic incorporation of nitrogen atoms and positive
charges yielded seven aromatic shelves ranging from
“normal” (1b) to strongly electron-deficient (1c·H+ and
1d·H+).[20] The aromatic shelves had very similar steric
properties, which greatly simplified the analyses. Finally, to
examine the nature of the F–p interactions in 1, five
additional series of balances 2–6 were prepared with different
arms (CH3, Cl, Br, I, and CF3) and the same aromatic shelves
(Scheme 1).
The folding ratios of the molecular balances 1–6 were
1
determined by integration of their H or 19F NMR spectra in
CD2Cl2.[21] The folded and unfolded conformers were in slow
exchange at 238C, leading to distinct sets of peaks. The
reporter 5-methyl group provided easily measurable sets of
singlets at 2.1 and 1.7 ppm. Solution studies and crystal-
structure analysis confirmed that the 5-methyl group had
minimal influence on the folding equilibrium.
Molecular balance 1 gave rise to a wide range of folding
energies with the different aromatic shelves (Figure 1). Parent
molecular balance 1a formed a moderately destabilizing F–p
interaction (DG =+ 0.7 kcalmolÀ1). In contrast, the cationic
balances 1c·H+ and 1d·H+ gave rise to strongly stabilizing
F–p interactions (DG = À1.4 to À1.5 kcalmolÀ1). Overall, the
folding energy trends for 1a–1d were consistent with an
electrostatic interaction as the folded conformers became
increasingly more stabilized with more electron-poor aro-
matic shelves.[6,8a] Similar stability trends were also observed
in other organic solvents (see the Supporting Information,
Figure S11). These experimental trends mirrored computa-
[*] Dr. P. Li, J. M. Maier, E. C. Vik, C. J. Yehl, Dr. B. E. Dial, A. E. Rickher,
Dr. M. D. Smith, Dr. P. J. Pellechia, Prof. K. D. Shimizu
Department of Chemistry and Biochemistry
University of South Carolina
Columbia, SC 29208 (USA)
E-mail: shimizu@mail.chem.sc.edu
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201702950.
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Figure 1. Folding energies (in CD2Cl2) of the fluorine- (black) and CH3-
substituted (striped) molecular balances 1 and 2.
Figure 2. X-ray crystal structures of the folded fluorine-substituted
balance analogues 1’a (A), 1’b (B), 1’c (C), and 1’d (D). The intra-
molecular distances between the fluorine atoms and the aromatic
planes are highlighted (black lines).
tional predictions of the F–p interaction between CH3F and
various arenes.[22] For example, CH3F has been predicted to
form a slightly repulsive interaction with benzene but
a strongly attractive interaction with hexafluorobenzene.
The electrostatic nature of the F–p interaction was
evident from comparison with the molecular balances 2a–
2d, which formed intramolecular CH–p interactions[17a,b,d]
folding energies. These results suggest that the stabilizing F–p
interactions were not due to dipole–dipole interactions. A
possible reason is the nearly perpendicular geometry of the
that have only
a
minor electrostatic component[23]
À
(Figure 1). In contrast to the F–p molecular balances 1a–
1d, compounds 2a–2d showed little variation across the same
set of isomeric aromatic surfaces, providing support for the
dominant electrostatic component of the F–p interactions in
1. Comparison of the folding energies of molecular balances
1 and 2 provides confirmation of the attractive and stabilizing
nature of the F–p interactions. With a non-heterocyclic
aromatic surface (1a vs. 2a), the F–p interaction was slightly
destabilizing (DDG =+ 0.3 kcalmolÀ1) compared to the CH–
p interaction. However, with cationic aromatic surfaces
(1b·H+–1d·H+ vs. 2b·H+–2d·H+), the F–p interactions
became significantly more stabilizing (DDG = À1.5 kcal
molÀ1) than the CH–p interactions.
Next, the F–p interactions were characterized by X-ray
crystallography.[24] Molecular balances 1a–1d did not consis-
tently crystallize in the folded conformer. However, ana-
logues 1’a–1’d without the 5-methyl group crystallized as
mixtures of the folded and unfolded conformers.[25] Solution
studies confirmed that 1’a–1’d displayed analogous folding
energy trends as 1a–1d (Figure S2). In the crystal structures
of 1’a–1’d, the F atoms were positioned over the central rings
of the aromatic shelves (Figure 2). The short atom-to-plane
distances (3.0–3.1 ꢀ) are consistent with previous reports on
F–p interactions.[10d,14a,b]
C F bond relative to the aromatic surface in the crystal
structure (Figures 2). This perpendicular F–p geometry is
similar to that observed by Diederich and co-workers
between a C F bond and a carbonyl p-system.
[11]
À
Next, the hypothesis that the F–p interaction involves
attraction between the partial negative charge (dÀ) on F and
electropositive heterocyclic and cationic surfaces was
explored.[6,22] The folding energies of 1 were correlated with
the calculated electrostatic potentials (ESPs) of the seven
aromatic shelves (Figure 3). ESPs have been successfully
applied to study the electrostatic component of many non-
covalent aromatic interactions.[26] ESP values of an aromatic
surface are strongly correlated with the Hammett s parame-
ter but have the advantage that they can be applied to
heterocyclic and charged aromatic surfaces. The ESP values
at the central ring of the aromatic shelves (a–d, b·H+, c·H+,
and d·H+) were calculated at the B3LYP/6-31G* level of
theory using the truncated versions capped with methyl
groups. An excellent linear correlation was found between the
ESP values and the folding energies of 1 (Figure 3). Separate
trends were observed for the neutral and cationic aromatic
shelves owing to their drastically different ESPs. ESP analysis
also accurately predicted the similar folding energies of the
isomeric compounds 1c and 1d as well as 1c·H+ and 1d·H+.
The F–p stability trends in 1 were compared with other
halogen–p interactions using the Cl, Br, and I molecular
balances 3–5 (Figure 4).[14a,b] All of the halogen balances
showed similar trends with stronger stabilizing interactions
for electron-poorer aromatic surfaces.[27] However, the trend
was steeper for F–p balance 1, which is consistent with the
more negative atomic charge on the F atom.[28] Further
support for the strongly stabilizing nature of the F–p
interaction in 1 was provided by a similarly steep trend for
To investigate the possible role of dipole–dipole inter-
actions in the F–p interactions, the folding energies of
fluorine-substituted molecular balances with isomeric 4,7-
(1c) and 1,10-phenanthroline (1d) shelves were compared.
These heterocyclic shelves have similar electrostatic poten-
tials (see below) but have opposing dipoles relative to the
À
C F bond (Figure S13). The solution folding energies of 1c
and 1d were nearly identical (Æ 0.02 kcalmolÀ1). Similarly,
the protonated versions (1c·H+ and 1d·H+) had very similar
2
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
respect, the F–p interaction in 1 appears to be similar to an
anion–p-type interaction.[30] We recognize that the terminol-
ogy “F–p” is a convenient description of the interacting
groups but not an accurate description of the underlying basis
of the interaction. Further studies are currently underway to
quantify the dispersion, solvophobic, and steric components
of these F–p interactions.
Acknowledgements
We acknowledge support by the National Science Foundation
(CHE 1310139).
Conflict of interest
Figure 3. Correlation between the solution folding energies (in CD2Cl2)
of the fluorine-substituted balances 1a–1d and 1b·H+–1d·H+ and the
calculated center ESP values of truncated versions of the aromatic
shelves (a–d, b·H+, c·H+, and d·H+).
The authors declare no conflict of interest.
Keywords: electrostatic interactions · fluorine ·
F–p interactions · supramolecular chemistry · p interactions
[1] P. Kirsch, Modern Fluoroorganic Chemistry: Synthesis, Reac-
tivity, Applications, 2nd ed., Wiley-VCH, Weinheim, 2013.
[2] S. Fustero, A. Simꢁn-Fuentes, P. Barrio, G. Haufe, Chem. Rev.
2015, 115, 871 – 930.
[3] a) R. Berger, G. Resnati, P. Metrangolo, E. Weber, J. Hulliger,
Chem. Soc. Rev. 2011, 40, 3496 – 3508; b) K. Reichenbꢂcher, H. I.
Sꢃss, J. Hulliger, Chem. Soc. Rev. 2005, 34, 22 – 30.
[4] a) S. Purser, P. R. Moore, S. Swallow, V. Gouverneur, Chem. Soc.
Rev. 2008, 37, 320 – 330; b) E. P. Gillis, K. J. Eastman, M. D. Hill,
D. J. Donnelly, N. A. Meanwell, J. Med. Chem. 2015, 58, 8315 –
8359.
[5] a) N. C. Yoder, D. Yꢃksel, L. Dafik, K. Kumar, Curr. Opin.
Chem. Biol. 2006, 10, 576 – 583; b) C. Adam, L. Yang, S. L.
Cockroft, Angew. Chem. Int. Ed. 2015, 54, 1164 – 1167; Angew.
Chem. 2015, 127, 1180 – 1183; c) M. R. Ams, M. Fields, T.
Grabnic, B. G. Janesko, M. Zeller, R. Sheridan, A. Shay, J.
Org. Chem. 2015, 80, 7764 – 7769.
[6] D. OꢄHagan, Chem. Soc. Rev. 2008, 37, 308 – 319.
[7] B. E. Smart, J. Fluorine Chem. 2001, 109, 3 – 11.
[8] a) L. Hunter, Beilstein J. Org. Chem. 2010, 6, 38; b) L. E.
Zimmer, C. Sparr, R. Gilmour, Angew. Chem. Int. Ed. 2011, 50,
11860 – 11871; Angew. Chem. 2011, 123, 12062 – 12074.
[9] M. G. Holl, M. D. Struble, P. Singal, M. A. Siegler, T. Lectka,
Angew. Chem. Int. Ed. 2016, 55, 8266 – 8269; Angew. Chem.
2016, 128, 8406 – 8409.
[10] a) K. Mꢃller, C. Faeh, F. Diederich, Science 2007, 317, 1881 –
1886; b) P. Zhou, J. W. Zou, F. F. Tian, Z. C. Shang, J. Chem. Inf.
Model. 2009, 49, 2344 – 2355; c) R. A. Cormanich, R. Rittner, D.
OꢄHagan, M. Buhl, J. Comput. Chem. 2016, 37, 25 – 33; d) T. J.
Mooibroek, P. Gamez, J. Reedijk, CrystEngComm 2008, 10,
1501 – 1515.
[11] F. Hof, D. M. Scofield, W. B. Schweizer, F. Diederich, Angew.
Chem. Int. Ed. 2004, 43, 5056 – 5059; Angew. Chem. 2004, 116,
5166 – 5169.
Figure 4. Correlation of the solution folding energies (in CD2Cl2) of the
molecular balances 1 and 3–5 with F, Cl, Br, I and CF3 arms with the
calculated ESP values of the aromatic surfaces.
CF3 balance 6. The steep trend of 6 is even more remarkable
as the electrostatic F–p interaction must overcome the steric
repulsion of the large CF3 group, which is evident from its
least favorable (most positive) folding energies.
In conclusion, analysis of the six series of molecular
balances 1–6 (37 in total) has confirmed the ability of fluorine
atoms and fluorine-containing groups to form stabilizing
interactions with electron-poor aromatic surfaces. These F–p
interactions are consistent with an attractive electrostatic
interaction between an electronegative fluorine atom and
electron-deficient heterocyclic and cationic aromatic surfaces.
The F–p interactions are clearly different to other types of
halogen bond interactions.[29] For example, the halogen–p
interaction was strongest for the most electronegative fluo-
rine atom. In contrast, halogen bond interactions involving
lone pairs or sigma holes are the strongest with the most
polarizable and least electronegative halogen atoms. In this
[12] J. A. Olsen, D. W. Banner, P. Seiler, U. O. Sander, A. DꢄArcy, M.
Stihle, K. Mꢃller, F. Diederich, Angew. Chem. Int. Ed. 2003, 42,
2507 – 2511; Angew. Chem. 2003, 115, 2611 – 2615.
[13] a) H. Adams, S. Cockroft, C. Guardigli, C. A. Hunter, K. R.
Lawson, J. Perkins, S. E. Spey, C. J. Urch, R. Ford, ChemBio-
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
Chem 2004, 5, 657 – 665; b) C. D. Tatko, M. L. Waters, Org. Lett.
2004, 6, 3969 – 3972; c) M. T. Scerba, S. Bloom, N. Haselton, M.
Siegler, J. Jaffe, T. Lectka, J. Org. Chem. 2012, 77, 1605 – 1609;
d) W. B. Jennings, N. OꢄConnell, J. F. Malone, D. R. Boyd, Org.
Biomol. Chem. 2013, 11, 5278 – 5291; e) I. Pavlakos, T. Arif,
A. E. Aliev, W. B. Motherwell, G. J. Tizzard, S. J. Coles, Angew.
Chem. Int. Ed. 2015, 54, 8169 – 8174; Angew. Chem. 2015, 127,
8287 – 8292.
heterocyclic molecular balances in 10% (v/v) CF3CO2D/CD2Cl2
solution. The efficacy of this procedure was demonstrated by the
TFA titration experiments detailed in the Supporting Informa-
tion.
[21] The procedure for measuring the ratio of folded to unfolded
conformers in solution is detailed in the Supporting Information.
[22] S. Kawahara, S. Tsuzuki, T. Uchimaru, J. Phys. Chem. A 2004,
108, 6744 – 6749.
[14] a) I. Saraogi, V. G. Vijay, S. Das, K. Sekar, T. N. G. Row, Cryst.
Eng. 2003, 6, 69 – 77; b) M. D. Prasanna, T. N. Guru Row, Cryst.
Eng. 2000, 3, 135 – 154; c) H. R. Khavasi, N. Rahimi, Cryst.
Growth Des. 2017, 17, 834 – 845; d) H. Takezawa, T. Murase, G.
Resnati, P. Metrangolo, M. Fujita, J. Am. Chem. Soc. 2014, 136,
1786 – 1788; e) T. H. Chen, I. Popov, W. Kaveevivitchai, Y. C.
Chuang, Y. S. Chen, A. J. Jacobson, O. S. Miljanic, Angew. Chem.
Int. Ed. 2015, 54, 13902 – 13906; Angew. Chem. 2015, 127, 14108 –
14112.
[15] a) I. K. Mati, S. L. Cockroft, Chem. Soc. Rev. 2010, 39, 4195 –
4205; b) S. Paliwal, S. Geib, C. S. Wilcox, J. Am. Chem. Soc. 1994,
116, 4497 – 4498.
[16] a) J. Hwang, P. Li, W. R. Carroll, M. D. Smith, P. J. Pellechia,
K. D. Shimizu, J. Am. Chem. Soc. 2014, 136, 14060 – 14067; b) J.
Hwang, B. E. Dial, P. Li, M. E. Kozik, M. D. Smith, K. D.
Shimizu, Chem. Sci. 2015, 6, 4358 – 4364; c) J. Hwang, P. Li, M. D.
Smith, K. D. Shimizu, Angew. Chem. Int. Ed. 2016, 55, 8086 –
8089; Angew. Chem. 2016, 128, 8218 – 8221.
[23] a) S. Tsuzuki, K. Honda, T. Uchimaru, M. Mikami, A. Fujii, J.
Phys. Chem. A 2006, 110, 10163 – 10168; b) A. L. Ringer, M. S.
Figgs, M. O. Sinnokrot, C. D. Sherrill, J. Phys. Chem. A 2006, 110,
10822 – 10828; c) C. D. Sherrill, Acc. Chem. Res. 2013, 46, 1020 –
1028.
[24] CCDC 1538758– 1538766 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre.
[25] A good correlation was observed between the solid-state and
solution folded/unfolded ratios of 1’a–1’d (Figure S1).
[26] a) K. D. Shimizu, P. Li, J. Hwang in Aromatic Interactions:
Frontiers in Knowledge and Application, The Royal Society of
Chemistry, London, 2017, pp. 124 – 171; b) J. Hwang, P. Li, K. D.
Shimizu, Org. Biomol. Chem. 2017, 15, 1554 – 1564.
[27] Complete ESP correlation analysis of five halogen-substituted
molecular balances (1 and 3–6) with five types of aromatic
shelves (a–d, b·H+, c·H+, and d·H+) is provided in the Supporting
Information (Figure S3).
[17] a) C. Zhao, R. M. Parrish, M. D. Smith, P. J. Pellechia, C. D.
Sherrill, K. D. Shimizu, J. Am. Chem. Soc. 2012, 134, 14306 –
14309; b) C. Zhao, P. Li, M. D. Smith, P. J. Pellechia, K. D.
Shimizu, Org. Lett. 2014, 16, 3520 – 3523; c) P. Li, T. M. Parker, J.
Hwang, F. Deng, M. D. Smith, P. J. Pellechia, C. D. Sherrill, K. D.
Shimizu, Org. Lett. 2014, 16, 5064 – 5067; d) P. Li, J. Hwang, J. M.
Maier, C. Zhao, D. V. Kaborda, M. D. Smith, P. J. Pellechia,
K. D. Shimizu, Cryst. Growth Des. 2015, 15, 3561 – 3564; e) B. U.
Emenike, S. N. Bey, B. C. Bigelow, S. V. S. Chakravartula, Chem.
Sci. 2016, 7, 1401 – 1407.
[28] Additional crystallographic analysis suggested that the different
correlation slopes among halogen-arm balances were less likely
to be due to changes in the halogen–p geometry (Figure S4).
[29] a) M. Erdꢅlyi, Chem. Soc. Rev. 2012, 41, 3547 – 3557; b) G.
Cavallo, P. Metrangolo, R. Milani, T. Pilati, A. Priimagi, G.
Resnati, G. Terraneo, Chem. Rev. 2016, 116, 2478 – 2601.
[30] a) S. E. Wheeler, J. W. G. Bloom, Chem. Commun. 2014, 50,
11118 – 11121; b) O. B. Berryman, V. S. Bryantsev, D. P. Stay,
D. W. Johnson, B. P. Hay, J. Am. Chem. Soc. 2007, 129, 48 – 58;
c) P. Gamez, T. J. Mooibroek, S. J. Teat, J. Reedijk, Acc. Chem.
Res. 2007, 40, 435 – 444.
[18] P. Li, C. Zhao, M. D. Smith, K. D. Shimizu, J. Org. Chem. 2013,
78, 5303 – 5313.
[19] J. M. Maier, P. Li, J. Hwang, M. D. Smith, K. D. Shimizu, J. Am.
Chem. Soc. 2015, 137, 8014 – 8017.
[20] The singly protonated molecular balances (1b·H+, 1c·H+, and
1d·H+) were prepared in situ by dissolving the corresponding
Manuscript received: March 21, 2017
Final Article published: && &&, &&&&
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Electrostatic Interactions
Attractive fluorine: The F–p interaction
between a fluorine substituent and an
aromatic surface was measured by using
a series of molecular balances. This
interaction was found to be slightly
repulsive with electron-rich surfaces but
strongly attractive with electron-poor and
cationic surfaces.
P. Li, J. M. Maier, E. C. Vik, C. J. Yehl,
B. E. Dial, A. E. Rickher, M. D. Smith,
P. J. Pellechia,
K. D. Shimizu*
&&&& — &&&&
Stabilizing Fluorine–p Interactions
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!