Journal of the American Chemical Society
Article
solid state and clearly distinguish the meta protons of 1−3
(Figure S44). Corresponding DFT calculations of the
hyperfine couplings (Table S5) demonstrate that the aiso
coupling observed for each proton follows the Heller−
McConnell relationship and is not influenced by the nature
Table 1. Simulated EPR Parameters for 1−5
ortho
meta
para
a
a
a
iso
line width
b
iso
iso
a
simulation
giso
[MHz]
[MHz]
[MHz]
[Gauss]
1
2
2.00253
2.00240
2.00241
−
−
−
−
−
0.50
2.01
2.10
2.650
0.250
0.165
1.97
1.95
0.50
2.01
2.10
0.79
1.40
−
−
−
3(1H)
3(19F)
CONCLUSION
■
4
5
2.00239
2.00239
0.180
0.240
In conclusion, we have shown that, in contrast to the apparent
D5 symmetry of pentaarylcyclopentadienyl radicals in solution,
the single crystals contain only one distinct valence tautomer
which is “frozen out” and stabilized by the specific arrangement
in the crystal lattice. The unambiguous structural distortion of
the cyclopentadienyl cores is a unique observation of the
Jahn−Teller distortion in symmetrically substituted Cp radicals
and demonstrates that the distortion is unquestionably an
intrinsic energy-minimizing electronic effect rather than the
result of the substitution pattern.
a
b
All g values and hyperfine couplings are isotropic. Gaussian line-
widths measured peak-to-peak.
(HF) couplings, respectively. This corresponds to equivalent
1H HF couplings of the ortho and para protons of the Cp-
bonded phenyl ring substituents, following a Huckel-like spin
̈
density distribution.43−45 The weaker 1H couplings of the meta
substituents are not resolved as expected for the amount of
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
The observed hyperfine couplings of the substituents have a
cos2θ angular dependence between the π orbital overlap of the
Cp core and the phenyl π systems,27,44 following the Heller−
McConnell equation.48,49 Taking the average of the five Cp-
phenyl dihedral angles θ from 4 (57.9°) and 5 (51.3°) shows
that 5 exhibits a slightly smaller mean angle than 4, which
corresponds to a larger aiso value being observed in its EPR
spectrum (Table S6). Therefore, the EPR spectra of 4 and 5
and the observed HF couplings best correlate to the mean
angle between the planes defined by the phenyl groups and
cyclopentadiene ring. This is in good agreement with the
average radical geometrical parameters observed in the solid
state. Finally, such observations may occur from the rapid
reorientation of the radicals at room temperature or an
electronic valence equilibrium on the time scale of the EPR
experiment (10−7 s).19,50 Frozen solution EPR experiments of
1−5 reveal broadened S = 1/2 signals from anisotropic
hyperfine contributions with no resolvable hyperfine features
(Figure S43). It is noted that frozen solution EPR spectra 4
and 5 are slightly wider than 1−3 due to the larger hyperfine
contribution of the para protons of 4 and 5, as previously
discussed.
For para-OMe substituted radical 1, no resolved hyperfine
splittings are observed in the EPR spectrum and it may be
simply simulated as a broad S = 1/2 signal. The spectrum of 2
is well simulated with two groups of 10 protons originating
from the ortho (1.97 MHz) and meta (0.75 MHz) nuclei on
the phenyl rings. The EPR spectrum of 3 is even more highly
featured as additional HF splitting is introduced by the 19F
nuclei on the para-C6F5 group. The EPR spectrum of 3 is well
reproduced with two groups of 10 protons corresponding to
the ortho (1.95 MHz) and meta (1.40 MHz) phenyl positions.
Additionally, 15 equivalent 19F nuclei with 0.50 MHz HF
couplings are included, which originate from the ortho and
para positions of the para-C6F5 group, following the same spin
■
sı
Synthetic procedures and analytical data (NMR, IR,
EPR, and UV/vis spectra, computational details (PDF)
Accession Codes
tallographic data for this paper. These data can be obtained
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
Stephan Schulz − Institute for Inorganic Chemistry and Center
for Nanointegration Duisburg-Essen (Cenide), University of
George E. Cutsail III − Institute for Inorganic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany; Max
Planck Institute for Chemical Energy Conversion (CEC),
̈
Gebhard Haberhauer − Institute for Organic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany;
Authors
Yannick Schulte − Institute for Inorganic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany
Blaise L. Geoghegan − Institute for Inorganic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany; Max
Planck Institute for Chemical Energy Conversion (CEC),
̈
Christoph Helling − Institute for Inorganic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany
Christoph Wölper − Institute for Inorganic Chemistry,
University of Duisburg-Essen, 45117 Essen, Germany
density distribution and Huckel pattern as the phenyl groups.
̈
Despite the similar mean torsion angles in 2 (45.9°) and 3
(47.2°), the para-C6F5 substituent is highly electron with-
drawing and facilitates spin delocalization through F p-
orbitals,48 resulting in a much larger spin density observed at
the meta carbons of the phenyl ring, and an increase in
aiso(1Hmeta) from 2. Additional electron nuclear double
resonance spectroscopy measurements further reveal the
torsion angle dependence of the observed hyperfine in the
Complete contact information is available at:
12662
J. Am. Chem. Soc. 2021, 143, 12658−12664