Synthesis and Magnetic Properties of Trioxytriphenylamine Dimers in their Di(radical cationic) States

Article abstract of DOI:10.1002/chem.201703220

Three structural isomers of trioxytriphenylamine (TOT) dimers, 4,4′′′-bis(2,2′:6′,2“:6”,6-trioxytriphenylamine) (4), 3,3′′′-bis(2,2′:6′,2“:6”,6-trioxytriphenylamine) (5), and 3,4′′′-bis(2,2′:6′,2“:6”,6-trioxytriphenylamine) (6), have been prepared and their electronic and magnetic properties in their di(radical cationic) states have been investigated. These di(radical cationic) species can be handled under ambient conditions because of their high stability under aerated conditions even in solution. The X-ray crystal structure analysis demonstrated that the TOT moieties of all the di(radical cation)s have planar structures similar to that of the parent TOT radical cation 3⁺. The UV/Vis spectra of the di(radical cation)s show characteristic absorptions depending on the connecting pattern. Thus, in the long-wavelength region (600–900 nm), 4²⁺ exhibits strong and broad characteristic absorptions, whereas compounds 5²⁺ and 6²⁺ exhibit weak absorptions. Notably, in the 450–600 nm region, 5²⁺ displays very similar absorptions (with twice the intensity) to 3⁺, whereas small differences were observed for 6²⁺. Finally, we investigated in detail the magnetic properties of the corresponding di(radical cation)s by electron spin resonance spectroscopy and magnetic susceptibility measurements, which indicated intramolecular exchange interactions with a singlet ground state and a large singlet–triplet (S–T) gap for 4²⁺, a singlet ground state and a small S–T gap for 5²⁺, and a triplet ground state for 6²⁺.

Full text of DOI:10.1002/chem.201703220

DOI: 10.1002/chem.201703220
Full Paper
&
Radical Ions
Synthesis and Magnetic Properties of Trioxytriphenylamine
Dimers in their Di(radical cationic) States
Shuichi Suzuki,*[a, b] Nobuaki Tanaka,[a] Masatoshi Kozaki,[a] Daisuke Shiomi,[a]
Kazunobu Sato,[a] Takeji Takui,[a] and Keiji Okada*[a]
Abstract: Three structural isomers of trioxytriphenylamine
(TOT) dimers, 4,4’’’-bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (4),
3,3’’’-bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (5), and 3,4’’’-
bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (6), have been pre-
pared and their electronic and magnetic properties in their
di(radical cationic) states have been investigated. These
di(radical cationic) species can be handled under ambient
conditions because of their high stability under aerated con-
ditions even in solution. The X-ray crystal structure analysis
demonstrated that the TOT moieties of all the di(radical cat-
ion)s have planar structures similar to that of the parent TOT
radical cation 3+. The UV/Vis spectra of the di(radical cat-
ion)s show characteristic absorptions depending on the con-
necting pattern. Thus, in the long-wavelength region (600–
900 nm), 42+ exhibits strong and broad characteristic ab-
sorptions, whereas compounds 52+ and 62+ exhibit weak
absorptions. Notably, in the 450–600 nm region, 52+ displays
very similar absorptions (with twice the intensity) to 3+,
whereas small differences were observed for 62+. Finally, we
investigated in detail the magnetic properties of the corre-
sponding di(radical cation)s by electron spin resonance spec-
troscopy and magnetic susceptibility measurements, which
indicated intramolecular exchange interactions with a singlet
ground state and a large singlet–triplet (S–T) gap for 42+, a
singlet ground state and a small S–T gap for 52+, and a trip-
let ground state for 62+
.
Introduction
minobiphenyl (meta/meta-linked dication) and N,N,N’,N’-tetraa-
nisyl-3,4’-diaminobiphenyl (meta/para-linked dication) in the
solution state. They reported that the meta/meta-linked dicat-
ion is in a singlet spin state because of its disjointed character
and the meta/para-linked dication is predicted to be in a triplet
spin state because of its coextensive character.[7]
Arylamine derivatives, typical electron donors, have been in-
vestigated for their potential as components in organic semi-
conductors.[1] In addition, chemists have frequently synthesized
stable radical cations by electron oxidation of arylamine deriva-
tives[2] and investigated their magnetic behavior.[3] Arylamine-
based molecular assemblies and their oxidized species have at-
tracted considerable attention due to their unique p-electronic
systems.[4–9] For example, the partial oxidation of arylamine
dyads has drawn attention to intervalence p-electronic sys-
tems.[4] Arylamine derivatives can be employed as hole-trans-
porting materials applicable to Perovskite solar cells.[5] Owing
to recent progress in synthetic chemistry, various high- and
low-spin polyarylamine poly(radical cation)s have been investi-
gated in terms of their intermolecular spin–spin interac-
tions.[3b–d,6–9] For instance, Bushby et al. previously reported the
generation of di(radical cation)s of N,N,N’,N’-tetraanisyl-3,3’-dia-
[a] Dr. S. Suzuki, N. Tanaka, Prof. Dr. M. Kozaki, Prof. Dr. D. Shiomi,
Prof. Dr. K. Sato, Prof. Dr. T. Takui, Prof. Dr. K. Okada
Graduate School of Science, Osaka City University
3-3-138, Sugimoto, Sumiyoshi-ku, Osaka, Osaka 558–8585 (Japan)
E-mail: okadak@sci.osaka-cu.ac.jp
Of the polyarylamines to have been studied, para-substitut-
ed triphenylamines have been reported as stable radical cati-
ons.[3d,10,11] However, few attempts have been made to stabilize
triphenylamine radical cations without para substituents.[4b,12,13]
Hellwinkel and Melan designed triarylamine 1, which has bulky
dimethylmethylene bridges at the ortho positions and formed
[b] Dr. S. Suzuki
Present address: Graduate School of Engineering Science
Osaka University, Toyonaka, Osaka 560–8531 (Japan)
E-mail: suzuki-s@chem.es.osaka-u.ac.jp
Supporting information and the ORCID numbers for the authors of this arti-
cle can be found under https://doi.org/10.1002/chem.201703220.
Chem. Eur. J. 2017, 23, 16014 –16025
16014
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the radical cationic species 1+ in sulfuric acid; however, its
properties in the solid state have not been reported.[12,14] Our
research group has been continually interested in the physical
properties of triphenylamine radical cations without bulky sub-
stituents. In our previous studies we designed two oxygen-
bridged triphenylamines, 2,2’:6’,2“-dioxytriphenylamine (2)[13a,c]
and 2,2’:6’,2”:6“,6-trioxytriphenylamine (TOT, 3),[13b,c] the radical
cations of which, 2+ and 3+, are stable species under aerated
conditions both in solution and in the solid state.[13,15] We also
discovered their unique bulk magnetic properties with inter-
molecular contacts.[13c,15a] In this study we synthesized the posi-
tional isomers of the TOT dimer, namely 4,4’’’-bis(2,2’:6’,2”:6“,6-
trioxytriphenylamine) (4), 3,3’’’-bis(2,2’:6’,2”:6“,6-trioxytriphenyl-
amine) (5), and 3,4’’’-bis(2,2’:6’,2”:6“,6-trioxytriphenylamine) (6).
These electron-rich TOT dimers produced stable di(radical cat-
ion)s, 42+, 52+, and 62+, which were shown to have singlet, sin-
glet, and triplet ground states, respectively. In these studies,
structural analyses have been important to evaluate the ex-
change interactions (see below). To the best of our knowledge,
52+ and 62+ are the first isolated di(radical cation)s with meta/
meta- and meta/para-linked triphenylamine dyads to have
been structurally analyzed.
yellow solid by the homocoupling reaction of 11 using bis(1,5-
cyclooctadiene)nickel(0) ([Ni(cod)2]).[5a]
The synthesis of 5 is outlined in Scheme 2. Phenoxazine de-
rivative 12 was produced by the cross-coupling reaction of 8
with 2,6-difluoroaniline. The regioselective bromination of 12,
Scheme 2. Synthetic pathway to 5.
Results and Discussion
using N-bromosuccinimide (NBS), afforded monobromophe-
noxazine 13, which was demethylated to give 14. The cycliza-
tion of 14 by treatment with potassium carbonate in N,N-dime-
thylformamide led to 15. Finally, the homocoupling reaction of
15 afforded 5 as a yellow solid.[17]
The synthetic pathway to 6 is presented in Scheme 3. Phe-
noxazine derivative 9 was converted into the boronic acid de-
rivative 16. Subsequently, the Suzuki–Miyaura cross-coupling
of 16 with 13 produced bis(phenoxazine) derivative 17. Finally,
compound 17 was converted into compound 6 by demethyla-
tion and double cyclization under basic conditions in 38%
yield over two steps.
Synthesis of trioxytriphenylamine dimers
The synthetic pathways to 4–6 are illustrated in Schemes 1–3,
respectively. Compound 4 was prepared from bis(3-methoxy-
phenyl) ether (7; Scheme 1).[16] The dianionic species of 7 was
produced by using n-butyllithium and subsequent addition of
1,2-diiodoethane afforded the bis(iodophenyl) ether derivative
8. The cross-coupling reaction of 8 and 4-bromo-2,6-difluoroa-
niline with a catalytic amount of a palladium(0) complex af-
forded phenoxazine derivative 9, which was subsequently con-
verted into 10 by the removal of the methyl groups with tri-
bromoborane. Compound 10 was transformed into 11 by
treatment with potassium carbonate in N,N-dimethylforma-
mide at 808C.[5a,13a,b] Finally, compound 4 was obtained as a
Scheme 1. Synthetic pathway to 4.
Scheme 3. Synthetic pathway to 6.
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16015
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
Compounds 4–6 were characterized by H NMR spectrosco-
para connection, the para-substituted TOT moiety exhibits the
first oxidation wave in a lower region, and the meta-substitut-
ed TOT experiences only a small effect.
py and MS, even though their solubility in common organic
solvents was very poor. Fortunately, single crystals of 4 suitable
for X-ray crystal structure analysis were obtained by recrystalli-
zation from o-dichlorobenzene at 1508C.[18] The molecular
structure of 4 is shown in Figure 1. The TOT moieties show
slightly shallow bowl structures similar to the structure of the
parent TOT 3.[13b] The dihedral angle between the TOT moieties
in 4 is about 308.
Synthesis and structures of the di(radical cation)s
The pure di(radical cation)s of 4–6 were synthesized by chemi-
cal oxidation using 2 equivalents of tris(4-bromophenyl)amini-
À
um hexafluoroantimonate. The di(radical cation)s 42+·2SbF6
,
52+·2SbF6À, and 62+·2SbF6 were characterized by X-ray crystal
structure analysis,[19–21] MS, and electron spin resonance (ESR)
spectroscopy. In contrast to the neutral species, these di(radical
cation)s exhibit good solubility in acetonitrile and dichlorome-
thane. Furthermore, they are quite stable under aerated condi-
tions even in solution (see below).
À
Single crystals of the di(radical cation)s were obtained by re-
crystallization from suitable solvents (see the Experimental Sec-
tion). The molecular structures determined by X-ray crystal
structure analyses are illustrated in Figure 3. The TOT skeletons
in 42+ have planar structures (Figures 3a,b), in contrast to the
shallow bowl structures of the neutral compound 4 (Figure 1).
The CÀC bond length between the TOTs in 42+ is 1.466(5) Š,
which is shorter than that of the neutral species [1.483(4) Š],
and the C1ÀN1 and C19ÀN2 bond lengths [1.354(4) and
1.364(4) Š, respectively] are shorter than the corresponding
bond lengths in the neutral state [1.395(4) Š]. Additionally, the
dihedral angle between the TOT moieties (128) is smaller than
that observed in the neutral species (318). These changes are
consistent with a quinoidal structure in the di(radical cationic)
state. Thus, the intramolecular exchange interaction is consid-
ered to be antiferromagnetic with a large negative J value.
Figure 1. ORTEP views of 4: a) top, and b) side views (ellipsoids drawn at the
50% probability level). Hydrogen atoms are omitted for clarity.
Electrochemical studies
We next employed differential pulse voltammetry (DPV) to
give an insight into the electrochemical properties of com-
pounds 4–6 (Figure 2). The oxidation potentials for these com-
pounds were measured in 1:4 (v/v) o-dichlorobenzene/benzo-
nitrile in the presence of 0.1m tetra-n-butylammonium per-
Figure 2. Differential pulse voltammetry of: a) 4, b) 5, and c) 6 in o-dichloro-
benzene/benzonitrile (1:4, v/v) in the presence of 0.1m tetra-n-butylammoni-
um perchlorate at 708C.
chlorate at 708C. The voltammogram for compound 4 shows
two separate waves at +0.08 and +0.18 V (vs. Fc/Fc+), where-
as that of 5 shows a single wave at +0.09 V. Compound 6
shows a wave at +0.12 V with a shoulder at around +0.04 V.
These waves were assigned to the two sequential oxidations
of the two TOT moieties in 4–6 and are compatible with the
following considerations: 1) Because of the meta/meta linkage,
compound 5 approximates to two non-interacting TOTs with
an oxidation potential similar to that of the parent TOT (3,
+0.10 V); 2) compound 4 has a para/para connection that
allows strong spin and charge interactions and so the second
oxidation potential is higher than the first because of the cou-
lombic interaction in 4; 3) for compound 6, which has a meta/
Figure 3. ORTEP views of: a,b) 42+·2SbF6À, c,d) 52+·2SbF6À, and
e,f) 62+·2SbF6À (ellipsoids drawn at the 50% probability level): a,c,e) top and
b,d,f) side views. Counter ions, crystal solvents, and hydrogen atoms are
omitted for clarity.
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16016
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The TOT skeletons of 52+ and 62+ possess planar structures
similar to those of 42+ (Figure 3). The CÀC bond lengths be-
tween the two TOT moieties were determined to be 1.471(8) Š
for 52+ and 1.492(5) Š for 62+; these values are slightly longer
than that of 42+. The CÀN bond lengths of the TOT moieties
are similar to those of the TOT radical cation.[13b,c,15] The dihe-
dral angles between the TOT moieties in 52+ and 62+ are
around 61 and 418, respectively, which are larger than that of
42+ (128). This difference is attributed to the steric effect be-
tween the TOT moieties.
which exhibits weak absorption in the 600–900 nm region (Fig-
ure 4d).[13b] The observed spectrum of 42+ indicates the pres-
ence of a strong electronic interaction between the two TOT
radical-cation moieties, as observed in the case of other p-elec-
tronic systems of bis(triarylamine) di(radical cation)s.[7] The
À
spectrum of 52+·2SbF6 shows absorptions at lmax =398, 462,
and 490 nm and a broad absorption in the range 600–900 nm
(lmax =701 nm; Figure 4b), and is very similar to that of the
À
parent radical cation 3+·SbF6 (Figure 4d), but with almost
twice the intensity in the 600–900 nm region, which reflects
TOT monomer radical cations sometimes form a p-dimeric
pair with strongly antiferromagnetic interactions.13c However,
no p-dimeric pairs were observed in the crystalline structures
of the present di(radical cation)s (see Figures S1–S3 in the Sup-
porting Information). Thus, we could experimentally estimate
the intramolecular exchange interactions from their magnetic
susceptibilities (see below).
the meta/meta connection of the TOTs. The spectrum of com-
À
pound 62+·2SbF6 shows absorptions at lmax =399, 459, and
493 nm and a broad absorption in the 600–900 nm region (Fig-
ure 4c). Although the spectrum is similar to that of the parent
TOT radical cation 3+·SbF6À, a new broad absorption appears
at around 550 nm, which can be assigned to charge-transfer
absorption [lcalcd(TD-DFT)=569.4 nm, meta-substituted TOT!
para-substituted TOT, see the Supporting Information].[22,23]
UV/Vis/NIR spectra of the di(radical cation)s in acetonitrile
Electron spin resonance (ESR) spectra of the di(radical cat-
ion)s in butyronitrile
The UV/Vis/NIR absorption spectra of the di(radical cation)s re-
corded in acetonitrile are shown in Figure 4. These spectra ex-
hibit no change after 1 day even under aerated conditions,
which indicates the high stability of these di(radical cationic)
The X-band ESR spectra of 42+·2SbF6À, 52+·2SbF6À, and 62+
À
·2SbF6 in frozen butyronitrile at 110 K are shown in Figures 5–
7, respectively. For 42+·2SbF6À, a broad signal due to a mono-
radical impurity (0.5%) is observed in the center field
(332 mT).[24] In addition to the central signal, two pairs of
broad shoulder peaks, assigned to a thermally accessible triplet
species (Figure 5), are also observed. Because of their rather
À
species. The spectrum of 42+·2SbF6 shows strong absorptions
at lmax =704, 758, and 826 nm (Figure 4a). This spectrum is
quite different to that of the parent TOT radical cation 3+,
À
Figure 5. ESR spectrum of 42+·2SbF6 in a frozen butyronitrile matrix at
110 K (ca. 1 mm, microwave frequency=9.329435 GHz).
weak intensity, spin-forbidden transition signals could not be
observed. The zero-field splitting parameter jDj value was esti-
mated to be 2.91 mT from the difference between the outer-
most z-component signals (jD/hcj =0.00273 cmÀ1), which is
consistent with the calculated averaged distance (9.8 Š) in the
point dipole approximation (intramolecular N–N distance:
9.74 Š; see Figure S4 and Table S1 in the Supporting Informa-
tion).
For 52+·2SbF6À, slightly complicated ESR signals assigned to
randomly oriented triplet signals are observed in the outer re-
gions of the central broad signal (Figure 6a). In addition, a
spin-forbidden transition signal (DMs = Æ2) is clearly observed
in the g=4 region of the resonance field (Figure 6, inset). The
observed ESR spectrum in the g=2 region was satisfactorily re-
produced by two sets of spin-Hamiltonian parameters; one for
À
Figure 4. UV/Vis/NIR absorption spectra of: a) 42+·2SbF6À, b) 52+·2SbF6
c) 62+·2SbF6À, and d) 3+·SbF6À in acetonitrile.
,
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16017
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 7. Observed ESR spectrum of 62+·2SbF6À in a frozen butyronitrile
matrix at 110 K (ca. 1 mm, microwave frequency=9.316588 GHz; inset: ob-
served spectrum of a forbidden transition).
Figure 6. ESR spectrum of 52+·2SbF6À: a) observed spectrum in a frozen bu-
tyronitrile matrix at 110 K (ca. 1 mm, microwave frequency=9.311898 GHz;
inset: observed spectrum of a forbidden transition); b) simulated spectrum
of the triplet species attributed to 52+-A (S=1, gxx =2.0035, gyy =2.0036,
gzz =2.0036, jD/hcj =0.00632 cmÀ1, and jE/hcj =0.00059 cmÀ1); c) simulated
spectrum of the triplet species attributed to 52+-B (S=1, gxx =2.0043,
parameters. The jDj value, estimated from the difference be-
tween the outermost peaks, is 0.00380 cmÀ1, which suggests
an averaged distance between the spin centers of 8.8 Š (spin-
centered N–N distance: 8.84 Š; see Figure S4 and Table S4).
It is rather surprising to note that the point dipole method
provides reasonable averaged spin-center distances approxi-
mating the intramolecular N–N distances. As shown in Fig-
ure S6 in the Supporting Information, the spin density distribu-
tions of these di(radical cation)s are delocalized over the whole
of the molecular plane. The spin density distributions in each
TOT moiety of these di(radical cation)s are very similar to that
of the parent TOT 3+ with a D3h symmetric structure,[13b,c]
which leads to the coincidence of the spin-centered positions
at the nitrogen positions in the TOTs.
g
yy =2.0038, gzz =2.0061, jD/hcj =0.00578 cmÀ1, and jE/hcj =0.00029 cmÀ1).
The broad signal at around 332 mT is due to a monoradical impurity.
the triplet species 52+-A (S=1, gxx =2.0035, gyy =2.0036, gzz =
2.0036, jD/hcj =0.00632 cmÀ1, jE/hcj =0.00059 cmÀ1
;
Fig-
ure 6b) and the other for the triplet species 52+-B (S=1, gxx =
2.0043, gyy =2.0038, gzz =2.0061, jD/hcj =0.00578 cmÀ1, jE/hcj
=0.00029 cmÀ1; Figure 6c). These results suggest that there
are two different conformers in the frozen matrix. DFT calcula-
tions performed at the UB3LYP/6-31G(d,p) level of theory by
using Gaussian 09[22] suggest the existence of two local mini-
mum structures (see Figure S4 and Tables S2 and S3 in the
Supporting Information); one is nearly identical (the dihedral
angle is around 658 between the two TOT radical cation moiet-
ies with a cisoid conformation) to the structure determined by
X-ray crystal analysis, and the other is a transoid structure with
a different dihedral angle of around 1268. The cisoid conformer
has a shorter intramolecular N–N distance (7.72 Š) than the
transoid conformer (8.38 Š).
We also calculated the zero-field parameters considering the
spin densities at the UB3LYP/6-31(d) level of theory by using
the ORCA program package:[25] 42+: D/hc=À0.00670 cmÀ1
,
E/hc=À0.00039 cmÀ1
;
52+-A: D/hc=À0.00548 cmÀ1, E/hc=
À0.00015 cmÀ1
À0.00015 cmÀ1
;
;
52+-B:
62+
D/hc=À0.00486 cmÀ1,
D/hc=À0.00475 cmÀ1,
E/hc=
E/hc=
:
À0.00075 cmÀ1. Compound 42+ was predicted to have the
largest value of jD/hcj (0.00670 cmÀ1) in spite of the smallest
observed value (0.00273 cmÀ1). The disagreement is probably
due to an overestimation of the spin densities on the TOT-con-
necting para carbon atoms, which are strongly dependent on
the contribution of the quinoidal structure.
The zero-field splitting parameter E values for 52+-A and 52+
-B are very small (jD/Ej >10) but not zero. Although the point
dipole approximation is not a good method to apply to these
systems, we examined it for qualitative purpose to give an in-
sight into averaged spin-center distances (7.4 Š for 52+-A and
7.7 Š for 52+-B); the calculated values qualitatively accord with
the intramolecular N–N distances (7.72 Š for the cisoid confor-
mer and 8.38 Š for the transoid conformer), thus 52+-A is as-
signed to the cisoid conformer and 52+-B to the transoid con-
former. The conformer 52+-B was calculated to be more stable
than 52+-A with a small energy difference (DE=0.045 eV).
Highly complicated triplet ESR signals are observed for 62+
Magnetic properties of the di(radical cation)s
To obtain further insight into the exchange interactions (J, H=
À2JSaSb, in which Sa represents spin a and Sb denotes spin b),
we measured the magnetic susceptibilities of polycrystalline
À
samples of 42+·2SbF6À, 52+·2SbF6À, and 62+·2SbF6 by using a
SQUID (superconducting quantum interference device) magne-
tometer in the temperature range 298–1.9 K under an external
magnetic field of 0.1 T. The temperature dependence of the
paramagnetic susceptibilities of 42+·2SbF6À, 52+·2SbF6À, and
À
·2SbF6 (Figure 7). A spin-forbidden transition (DMs = Æ2) is
also clearly observed in the g=4 region of the resonance field
(Figure 7, inset). The plot of signal intensity of the spin-forbid-
den transition versus the reciprocal temperature afforded a
straight line, which indicates a triplet ground state (see Fig-
ure S5 in the Supporting Information). Although the observed
ESR spectrum of 62+ in Figure 7 is similar to that of 52+ and
may be reproduced by two sets of spin-Hamiltonian parame-
ters, we could not determine the unique solution of the triplet
À
62+·2SbF6 are displayed in Figures 8–10, respectively.
À
Compound 42+·2SbF6 (containing 0.5% monoradical impur-
ity, Figure 8 inset) was found to be almost diamagnetic and
showed very low cpT values (0.04 emuKmolÀ1) at 298 K
(Figure 8). The temperature dependence of the cpT values was
simulated by using the Bleaney–Bowers model [Eq. (1)].[26] The
results indicate a strong antiferromagnetic interaction (2J/kB =
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16018
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 10. Temperature dependence of cpT for 62+·2SbF6À. The dashed line
is a simulation based on a modified Bleaney–Bowers model with 2J/
kB = +140 K and q=À11.5 K.
Figure 8. Temperature dependence of cpT for 42+·2SbF6À. The dashed line is
a simulation based on the singlet–triplet model with 2J/kB =À1300 K. Inset:
expanded cpT–T plots showing a 0.5% monoradical impurity.
À1300 K), consistent with the quinoidal structure suggested by
strong ferromagnetic interaction and a weak antiferromagnetic
interaction. The observed cpT–T curve was analyzed by using a
modified Bleaney–Bowers model [Eq. (3)], which suggests a
moderately large ferromagnetic interaction of 2J/kB = +140 K
and an unspecified antiferromagnetic mean-field parameter of
q=À11.5 K. The large J/kB value was assigned to the intramo-
lecular interaction between the two spin centers, indicating
that 62+ is found as a triplet in the ground state. The small an-
tiferromagnetic interaction was assigned to intermolecular in-
teractions resulting from the contacts between the di(radical
cation)s observed in the crystal structure analysis. These results
are consistent with the ESR Curie plots (see Figure S5 in the
Supporting Information) and the theoretical calculations of the
exchange interactions (see below).
À
the crystal structure and ESR analyses for 42+·2SbF6 described
above.
2NAg2m2BT
1
cpT
kBT 3 exp À2J=kBT Š
À
The cpT value of 52+·2SbF6 at 298 K was determined to be
0.723 emuKmolÀ1, which is close to the theoretical value for a
magnetically independent system of two S=1= spins (0.375”
2
2=0.75 emuKmolÀ1 assuming g=2.0; Figure 9). The cpT value
2NAg2m2BT
3
cpT
kB T À q 3 exp À2J=kBT Š
Theoretical calculations for the di(radical cation)s
To investigate the above assignments of the magnetic interac-
tions, we next calculated by DFT at the UB3LYP/6-31(d,p) level
of theory using the Gaussian 09 program the optimized struc-
tures of the di(radical cation)s 42+, 52+, and 62+ by using the
crystal structures as the initial geometries (Table 1 and Ta-
bles S5 and S6 in the Supporting Information).[23] The calcula-
tions revealed that the total energies of the low-spin states of
42+ and 52+ are lower than those of the high-spin state, and
Figure 9. Temperature dependence of cpT for 52+·2SbF6À. The dashed line is
a simulation based on the Curie–Weiss model with q=À9 K.
was almost constant down to 100 K and then rapidly de-
creased with further decreases in temperature. The cpT–T curve
was simulated by using the Curie–Weiss model with an unspe-
cified antiferromagnetic mean-field parameter q=À9 K
[Eq. (2)], which is consistent with the calculated exchange in-
teraction (see below).
Table 1. Calculated total energies of the low-spin (LS) and high-spin (HS)
[a]
states and calculated exchange interactions for 42+, 52+, and 62+
.
Spin state
E [a.u.]
<S2 >
2Jintra_calcd/kB [K]
2NAg2m2B
3kB T À q
1
42+
52+
62+
LS
HS
LS
HS
LS
À1941.82226420
À1941.82334185
À1941.81232049
À1941.81234686
À1941.81758860
À1941.81739917
2.0236
0.9887
2.0254
1.0264
2.0275
1.0219
2
cpT
T
À658
À16.7
+118
The cpT value of 62+·2SbF6 at 298 K (0.795 emuKmolÀ1) is
À
higher than the theoretical value of two magnetically inde-
HS
pendent S=1= spins (0.375”2=0.75 emuKmolÀ1 assuming
2
[a] The structures were optimized by DFT at the UB3LYP/6-31G(d,p) level
of theory using the Gaussian 09 program and the crystal structures as the
initial geometries.
g=2.0; Figure 10). The cpT value gradually increased upon
cooling to 70 K, and subsequently decreased rapidly below
70 K. This behavior can be explained by the coexistence of a
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16019
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
procedure.[15] Tris(4-bromophenyl)aminium hexafluoroantimonate
(TBPA·SbF6) was prepared from tris(4-bromophenyl)amine and
silver hexafluoroantimonate according to the literature with some
the exchange interactions (Jintra_calcd) calculated by using an
equation proposed by Yamaguchi and co-worker were deter-
mined to be 2Jintra_calcd/kB =À658 K and À16.7 K for 42+ and 52+
, respectively [Eq. (4)].[27] On the other hand, the total energy
of the high-spin state of 62+ is lower than that of the low-spin
state, and the calculated exchange interaction was estimated
to be 2Jintra_calcd/kB = +118 K. Notably, the calculations of struc-
tures with different dihedral angles between the TOT moieties
suggest that the ground states do not change (see Tables S7–
S9). We also estimated the intermolecular spin–spin interac-
tions for the close contacts between the di(radical cation)s of
52+ and 62+ by theoretical calculations using the geometries
extracted from their crystal structures (Figure S7). The results
obtained suggest that the values of the magnetic interactions
are smaller than those of the intramolecular interactions.
À
modification.[3d] The parent mono(radical cation) 3+·SbF6 was pre-
pared from 3 by a method described in the literature using
TBPA·SbF6 as an oxidation agent.[13b] Dichloromethane (CH2Cl2),
N,N-dimethylformamide (DMF), benzonitrile (PhCN), acetonitrile
(CH3CN), and butyronitrile were dried and distilled over calcium hy-
dride. Tetrahydrofuran (THF), toluene, diethyl ether (Et2O), and ben-
zene were dried and distilled over sodium. o-Dichlorobenzene was
dried by passing through an activate alumina column prior to use.
The other chemicals were of commercial grade and used without
further purification.
1H and 13C NMR spectra were recorded on a Bruker Avance III 300,
Bruker Avance III 400, or JEOL JNM-ECZ 400 spectrometer. ESR
spectra were recorded on a Bruker ELEXSYS E500 spectrometer.
Melting points were measured by using a Yanaco-MP-J3 apparatus
and are not corrected. IR spectra were recorded on a JASCO FT/IR-
4600 spectrophotometer in KBr pellets. Absorption spectra were
recorded on a SHIMADZU UV-2500 spectrophotometer and a Hita-
chi U-3500L spectrometer. DART-MS and APCI-MS spectra were re-
corded on JEOL AccuTOF LC-plus 4G and Bruker micrOTOF II spec-
trometers, respectively. Differential pulse voltammetry measure-
ments were carried out with an ALS Electrochemical Analyzer
Model 610A instrument with glassy carbon as the working elec-
trode (WE), a platinum wire as the counter electrode (CE), and a sa-
turated calomel electrode (SCE) as the reference electrode (RE) in
o-dichlorobenzene/PhCN (1:4, v/v) containing 0.1m n-butylammo-
nium perchlorate at 708C. The redox potentials were calibrated
with the ferrocene/ferrocenium couple. The X-ray diffraction data
were collected by using a Rigaku Saturn 724 CCD detector with
graphite monochromated MoKa radiation. The magnetic suscepti-
bility measurements were performed by using an MPMS-XL Quan-
tum Design SQUID magnetometer.
ELS À EHS
4
J
< S2 >HS À < S2 >LS
The quite different jJj values of the di(radical cation)s can
be considered in terms of the SOMO of the parent TOT 3+.[13b]
The SOMO coefficient of 3+ is large and negligibly small at the
4- (para) and 3 (meta)-positions, respectively. From the view
point of connecting positions of the TOT dimers, 42+, 52+, and
62+ are categorized as conjugated, nonconjugated (disjoint),
and nonconjugated (coextensive) p-electronic systems, respec-
tively. These types of p-electron conjugation are consistent
with the jJj values of the di(radical cation)s.
Conclusions
We have designed and synthesized a series of structural iso-
mers of a dimeric system based on TOT, namely 4–6, and in-
vestigated their electrochemical properties. The voltammo-
grams of 4 and 6 show two splitting waves, whereas that of 5
shows a wave without splitting, which suggests differences in
the interactions between the two TOT moieties in the di(radi-
cal cationic) states. We also succeeded in synthesizing and iso-
lating the dimers in the di(radical cationic) states and elucidat-
ed their electronic and magnetic properties by UV/Vis/NIR, ESR
spectroscopy, as well as by SQUID measurements. These di(rad-
ical cation)s were found to be quite stable under aerated con-
ditions both in solution and in the solid state. In the di(radical
cationic) states, the TOT moieties have planar structures similar
to those of the parent TOT radical cation. The SQUID measure-
ments indicated that the intramolecular spin–spin interactions
in 42+, 52+, and 62+ are strongly antiferromagnetic, quite
weakly antiferromagnetic, and ferromagnetic, respectively.
These results create scope for further studies on polymer
poly(radical cationic) species with controllable spin states.
Synthesis and characterization
Bis(2-iodo-3-methoxyphenyl) ether (8): Under an inert atmos-
phere, compound 7 (10.0 g, 43.5 mmol) was placed in a 100-mL
two-necked flask. THF (40 mL) was added to the flask, which was
placed in an ice/water bath. A 2.6m hexane solution of n-butyllithi-
um (38.0 mL, 98.8 mmol) was slowly added to the THF solution.
After removing the ice/water bath, the mixture was stirred at room
temperature for 1 h. The flask was placed into an ice/water bath,
to the mixture was added a THF solution (40 mL) of 1,2-diiodo-
ethane (27.07 g, 96.0 mmol). After removing the ice/water bath,
the mixture was stirred at room temperature for 3 h. The mixture
was poured into a 10% aqueous solution of sodium hydrogen sul-
fite (NaHSO3) (400 mL). The organic layer was separated with
CH2Cl2 (150 mL”3) and washed successively with water (100 mL)
and a saturated aqueous sodium chloride (NaCl) solution (100 mL).
The organic layer was dried over sodium sulfate (Na2SO4), filtered,
and concentrated under reduced pressure. The resulting residue
was washed with ethanol (EtOH) and cold hexane to afford the de-
sired compound as
a colorless solid (yield: 14.53 g, 71%).
C14H12I2O3; MW 482.06; colorless solid; m.p. 166–1678C; 1H NMR
(300 MHz, CDCl3): d=7.22 (t, J=8.2 Hz, 2H), 6.62 (dd, J=8.2,
1.1 Hz, 2H), 6.42 (dd, J=8.2, 1.1 Hz, 2H), 3.92 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=160.08, 157.39, 129.83, 111.42, 106.21, 80.90,
56.67 ppm; IR (KBr): n=2967 (w), 2935 (w), 2833 (w), 1582 (s), 1461
(s), 1428 (s), 1293 (w), 1270 (w), 1238 (s), 1182 (w), 1085 (s), 1021
(m), 793 (m), 761 (m) cmÀ1; MS (DART+): m/z: 482.94 [M+H]+; ele-
mental analysis calcd (%) for C14H12I2O3: C 34.88, H 2.51, N 0.00;
found: C 34.73, H 2.51, N 0.00.
Experimental Section
General
All the reactions were performed under an inert atmosphere. Bis(3-
methoxyphenyl) ether (7) was prepared according to a literature
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16020
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
10-(4’-Bromo-2’,6’-difluorophenyl)-1,9-dimethoxy-10H-phenoxa-
zine (9): Under an inert atmosphere, compound (5.02 g,
tate was collected by filtration. The precipitate was washed with
water, EtOH, and hexane to afford 11 as a yellow solid (yield:
473 mg, 90% over two steps). C18H8BrNO3; MW 364.97; yellow
solid; m.p. 262–2638C; 1H NMR (400 MHz, C6D6): d=6.43 (s, 2H),
6.25–6.23 (m, 4H), 6.16 ppm (dd, J=6.3, 3.2 Hz, 2H); IR (KBr): n=
3078 (w), 1625 (m), 1598 (w), 1558 (w), 1518 (s), 1480 (s), 1478 (m),
1420 (w), 1347 (m), 1275 (w), 1257 (w), 1221 (w), 1205 (w), 1075
(w), 1061 (w), 1019 (s), 942 (w), 917 (w), 833 (m), 756 (s), 700 (m),
637 (w), 626 (w), 563 (w) cmÀ1; MS (APCI): m/z: 366.01 [M+H]+,
368.01 [M+2+H]+; elemental analysis calcd (%) for C18H8BrNO3: C
59.04, H 2.20, N 3.83; found: C 58.80, H 2.34, N 3.87. 13C NMR data
could not be obtained due to the low solubility of the product.
8
10.4 mmol), 4-bromo-2,6-difluoroaniline (2.14 g, 10.3 mmol),
sodium tert-butoxide (NaOtBu) (2.98 g, 31.1 mmol), and tris(benzyli-
deneacetone)palladium(0)–chloroform {[Pd2(dba)3]·CHCl3} (537 mg,
0.519 mmol) were placed in a 30-mL two-necked flask. Toluene
(10 mL) was added to the flask along with a 0.34m toluene solu-
tion of tri-tert-butylphosphine (4.6 mL, 1.56 mmol) and the mixture
was heated at reflux for 15 h. After cooling to room temperature,
the mixture was filtered through a pad of Celite. The filtrate was
concentrated under reduced pressure. The residue was separated
by silica gel column chromatography using hexane/CH2Cl2 (1:1, v/
v) as eluent to afford the desired product. After evaporation of the
solvent, the residue was washed with a minimum amount of
hexane/EtOH to form a precipitate. The precipitate was collected
by filtration to afford the desired compound as a colorless solid
(yield: 1.19 g, 26%). C20H14BrF2NO3; MW 434.24; colorless solid;
4,4’’’-Bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (4): In an argon
glove box, 11 (103 mg, 0.282 mmol), 2,2’-bipyridyl (65.7 mg,
0.423 mmol), bis(1,5-cyclooctadiene)nickel(0) {[Ni(cod)2]} (107 mg,
0.389 mmol), 1,5-cyclooctadiene (48 mg, 0.443 mmol), and THF
(18 mL) were placed in a pressure tube, which was then sealed.
The tube was placed in an oil bath at 708C and stirred for 20 h.
After cooling to room temperature, the resulting precipitate was
collected and successively washed with toluene, EtOH, and hexane.
The resulting solid (78 mg) was dissolved in hot o-dichlorobenzene
(80 mL) and insoluble materials were removed by filtration through
a filter paper. The filtrate was cooled to room temperature and the
resulting precipitate was collected by filtration. The precipitate was
then washed with EtOH and hexane to afford the desired com-
pound as a yellow solid (yield: 66 mg, 81%). C36H16N2O6; MW
572.53; yellow solid; m.p. >3008C; 1H NMR (600 MHz, [D4]o-di-
chlorobenzene, 1308C): d=6.58 (br., 4H), 6.52 (s, 4H), 6.39 (d, J=
8.1 Hz, 4H), 6.35 ppm (d, J=8.1 Hz, 4H); IR (KBr): n=3055 (w),
1620 (m), 1518 (s), 1480 (s), 1465 (m), 1350 (s), 1318 (w), 1290 (w),
1269 (s), 1066 (m), 1014 (s), 830 (w), 754 (m), 699 (m), 574
(w) cmÀ1; HRMS (APCI+): calcd for C36H16N2O6: 572.1003; found:
572.1005; MS (APCI+): m/z: 572.10 [M]+; elemental analysis calcd
(%) for C36H16N2O6·(C6H4Cl2)0.1: C 74.86, H 2.82, N 4.77; found: C
75.02, H 2.97, N 4.91. 13C NMR data could not be obtained due to
the low solubility of the product.
m.p. 255–2568C; 1H NMR (300 MHz, CDCl3): d=7.06 (d-like, JH,F
=
7.4 Hz, 2H), 6.91 (t, JH,H =8.3 Hz, 2H), 6.56 (dd, JH,H =8.3, 1.1 Hz,
2H), 6.44 (dd, JH,H =8.3, 1.1 Hz, 2H), 3.67 ppm (s, 6H); 13C NMR
(75 MHz, CDCl3): d=162.65 (dd, JC,F =257.5, 5.5 Hz), 152.05, 150.07,
124.92, 124.37 (t, JC,F =14.0 Hz), 122.81, 119.82 (t, JC,F =12.7 Hz),
115.56 (dd, JC,F =25.8, 3.0 Hz), 108.79, 106.54, 55.66 ppm; IR (KBr):
n=3070 (w), 2939 (w), 2837 (w), 1612 (m), 1597 (m), 1578 (m),
1493 (s), 1462 (s), 1439 (w), 1416 (m), 1308 (m), 1285 (s), 1246 (m),
1223 (m), 1097 (s), 1034 (m), 961 (w), 897 (w), 861 (w), 833 (w), 773
(m), 743 (w), 723 (w), 604 (w), 513 (w) cmÀ1; MS (DART+): m/z:
434.06 [M+H]+, 436.06 [M+2+H]+; elemental analysis calcd (%) for
C20H14BrF2NO3: C 55.32, H 3.25, N 3.23; found: C 55.18, H 3.34, N
3.37.
10-(4’-Bromo-2’,6’-difluorophenyl)-1,9-dihydroxy-10H-phenoxa-
zine (10): Under an inert atmosphere, compound 9 (602 mg,
1.39 mmol) was placed in a 30-mL two-necked flask. CH2Cl2 (30 mL)
was added to the flask. The mixture was cooled and stirred at
À788C. A 1.0m CH2Cl2 solution of boron tribromide (BBr3) (9.8 mL,
9.80 mmol) was added and the mixture was stirred for 36 h as the
temperature was allowed to rise to room temperature. The mixture
was then poured into water (40 mL) and extracted with CH2Cl2
(20 mL”3). The organic layer was successively washed with water
(100 mL) and a saturated aqueous NaCl solution (100 mL). The or-
ganic layer was dried over Na2SO4, filtered, and concentrated
under reduced pressure. The residue was subjected to silica gel
column chromatography using CH2Cl2 as eluent to afford the de-
sired compound as a crude material (596 mg, quant.). The com-
pound was used in the next reaction without further purification.
C18H10BrF2NO3; MW 404.98; colorless solid; m.p. 190–1918C;
1H NMR (300 MHz, CDCl3): d=7.18 (d-like, JH,F =7.9 Hz, 2H), 6.98 (t,
10-(2’,6’-Difluorophenyl)-1,9-dimethoxy-10H-phenoxazine (12):
Under an inert atmosphere, compound 8 (5.09 g, 10.6 mmol),
NaOtBu
(3.02 g,
31.4 mmol),
[Pd2(dba)3]·CHCl3
(539 mg,
0.520 mmol), tri-tert-butylphosphinium tetrafluoroborate (450 mg,
1.60 mmol), and toluene (10 mL) were placed in a 30-mL two-
necked flask. The mixture was stirred at room temperature for
5 min. A toluene solution (5 mL) of 2,6-difluoroaniline (2.01 g,
15.6 mmol) was added and the mixture was heated at reflux for
5 h. After cooling to room temperature, the mixture was passed
through a pad of Celite and the filtrate was concentrated under re-
duced pressure. The residue was subjected to silica gel column
chromatography using hexane/CH2Cl2 (1:1, v/v) as eluent to afford
the desired product, which was re-precipitated from a mixed sol-
vent of hexane/EtOH. The precipitate was collected by filtration to
afford the desired compound as a colorless solid (yield: 2.15 g,
57%). C20H15F2NO3; MW 355.34; colorless solid; m.p. 222–2238C;
1H NMR (300 MHz, CDCl3): d=7.15 (tt, JH,H =8.3, JH,F =6.0 Hz, 1H),
6.90 (t, JH,H =8.3 Hz, 2H), 6.86 (t-like, JH,H =JH,F =8.3 Hz, 2H), 6.56
(dd, JH,H =8.3, 1.3 Hz, 2H), 6.43 (dd, JH,H =8.3, 1.3 Hz, 2H), 3.64 ppm
J
H,H =8.2 Hz, 2H), 6.64 (dd, JH,H =8.2, 1.3 Hz, 2H), 6.57 (dd, JH,H =8.2,
1.3 Hz, 2H), 5.47 ppm (s, 2H); 13C NMR (75 MHz, CDCl3): d=161.09
(dd, JC,F =252.5, 5.5 Hz), 151.58, 149.50, 126.81, 122.88 (t, JC,F
=
14.3 Hz), 121.61 (t, JC,F =12.7 Hz), 120.72, 116.93 (dd, JC,F =25.0,
3.0 Hz), 111.38, 108.83 ppm; IR (KBr): n=3445 (br), 1617 (m), 1515
(w), 1478 (m), 1459 (m), 1421 (w), 1298 (w), 1230 (w), 1067 (w),
1022 (w), 835 (w), 775 (w), 723 (w) cmÀ1; MS (DART+): m/z: 406.03
[M+H]+, 408.03 [M+2+H]+; elemental analysis calcd (%) for
C18H10BrF2NO3: C 53.23, H 2.48, N 3.45; found: C 53.09, H 2.65, N
3.43.
(s, 6H); 13C NMR (75 MHz, CDCl3): d=162.73 (dd,
5.0 Hz), 152.16, 150.08, 127.76 (t, JC,F =10.4 Hz), 124.88 (t, JC,F
JC,F =253.6,
4-Bromo-2,2’:6’,2“:6”,6-trioxytriphenylamine (11): Under an inert
atmosphere, compound 10 (596 mg, crude 1.39 mmol) and potas-
sium carbonate (K2CO3) (1.00 g, 7.25 mmol) were placed in a 100-
mL round-bottomed flask. DMF (25 mL) was added to the flask and
the mixture was stirred at 1008C for 12 h. After cooling to room
temperature, water (75 mL) was added and the resulting precipi-
=
13.9 Hz), 124.69, 123.36, 111.48 (dd, JC,F =22.6, 2.7 Hz), 108.69,
106.58, 55.61 ppm; IR (KBr): n=3006 (w), 2943 (w), 2839 (w), 1614
(w), 1577 (w), 1488 (m), 1459 (s), 1436 (m), 1322 (m), 1287 (s), 1257
(m), 1239 (m), 1182 (w), 1096 (s), 1072 (w), 1002 (m), 796 (m), 767
(m), 729 (m), 711 (w), 592 (w) cmÀ1; MS (DART+): m/z: 356.14
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16021
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[M+H]+; elemental analysis calcd (%) for C20H15F2NO3: C 67.60, H
filtration and washed with water, EtOH, and hexane to afford 15 as
4.26, N 3.94; found: C 67.54, H 4.28, N 3.94.
a yellow solid (yield: 219 mg, 43% over three steps). C18H8BrNO3;
MW 364.97; yellow solid; m.p. 191–1928C; H NMR (400 MHz, C6D6):
1
4-Bromo-10-(2’,6’-difluorophenyl)-1,9-dimethoxy-10H-phenoxa-
zine (13): Under an inert atmosphere, compound 12 (300 mg,
0.844 mmol) N-bromosuccinimide (143 mg, 0.803 mmol), and
CH2Cl2 (10 mL) were placed in a 30-mL two-necked flask. The mix-
ture was stirred at room temperature for 14 h and then poured
into water and extracted with CH2Cl2 (10 mL”3). The organic layer
was washed successively with water (20 mL) and a saturated aque-
ous NaCl solution (20 mL). The organic layer was dried over
Na2SO4, filtered, and concentrated under reduced pressure. The
residue was subjected to silica gel column chromatography using
hexane/CH2Cl2 (3:1, v/v) as eluent to afford the desired compound
as a crude material (175 mg, crude yield: 48%). The compound
was used in the next reaction without further purification.
C20H14BrF2NO3; MW 434.24; colorless solid; 1H NMR (300 MHz,
CDCl3): d=7.17 (tt, JH,H =8.3, JH,F =6.0 Hz, 1H), 7.12 (d, JH,H =8.9 Hz,
1H), 6.93 (t, JH,H =8.3 Hz, 1H), 6.87 (t-like, JH,H =JH,F =8.3 Hz, 2H),
6.67 (dd, JH,H =8.3, 1.1 Hz, 1H), 6.46 (dd, JH,H =8.3, 1.1 Hz, 1H), 6.34
(d, JH,H =8.9 Hz, 1H), 3.65 (s, 3H), 3.63 ppm (s, 3H); 13C NMR
(75 MHz, CDCl3): d=162.65 (dd, JC,F =253.8, 5.1 Hz), 152.63, 151.46,
149.88, 147.05, 129.08 (t, JC,F =10.3 Hz), 127.92, 124.94, 124.77,
124.41 (t, JC,F =14.7 Hz), 122.94, 111.55 (dd, JC,F =22.7, 2.9 Hz),
109.06, 107.38, 107.09, 100.83, 55.77, 55.64 ppm; IR (KBr): n=3445
(br), 2834 (w), 1611 (w), 1584 (w), 4496 (s), 1487 (s), 1464 (m), 1437
(s), 1428 (s), 1285 (s), 1261 (w), 1231 (m), 1093 (s), 1004 (m), 796
(w), 786 (w), 772 (w), 760 (w), 716 (w) cmÀ1; MS (DART+): m/z:
434.06 [M+H]+, 436.06 [M+2+H]+; HRMS (DART+): calcd for
C20H15BrF2NO3 [M+H]+: 434.0198; found: 434.0192;
d=6.51 (d, J=8.8 Hz, 1H), 6.28–6.18 (m, 6H), 5.93 ppm (d, J=
8.8 Hz, 1H); IR (KBr): n=3443 (br), 2925 (w), 1623 (w), 1591 (w),
1518 (m), 1476 (s), 1349 (w), 1272 (w), 1096 (w), 1065 (w), 1027 (w),
755 (m), 700 (w) cmÀ1; MS (DART+): m/z: 366.02 [M+H]+, 368.02
[M+2+H]+; HRMS (ESI+): calcd for C18H8BrNO3: 364.9688; found:
364.9684.13C NMR data could not be obtained due to the low solu-
bility of the product.
3,3’’’-Bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (5): In an argon
glove box, 15 (100 mg, 0.274 mmol), 2,2’-bipyridyl (52.9 mg,
0.340 mmol), [Ni(cod)2] (90.0 mg, 0.327 mmol), 1,5-cyclooctadiene
(40 mg, 0.369 mmol), and THF (18 mL) were placed in a pressure
tube, which was then sealed. The tube was placed in an oil bath at
808C and stirred for 20 h. After cooling to room temperature, the
resulting precipitate was collected and washed successively with
toluene, EtOH, and hexane. The resulting solid (76 mg) was dis-
solved in hot o-dichlorobenzene (30 mL) and insoluble materials
were removed by filtration through a filter paper. The filtrate was
cooled to room temperature and the resulting precipitate was col-
lected by filtration. The precipitate was washed with EtOH and
hexane to afford the desired compound as a yellow solid (yield:
50 mg, 64%). C36H16N2O6; MW 572.53; yellow solid; m.p. >3008C;
1H NMR (400 MHz, [D4]o-dichlorobenzene): d=6.60 (d, J=8.7 Hz,
2H), 6.49 (t, J=8.0 Hz, 2H), 6.42–6.34 (m, 4H), 6.30–6.14 ppm (m,
8H); IR (KBr): n=3447 (br), 1624 (w), 1517 (m), 1482 (m), 1335 (w),
1270 (w), 1071 (w), 1021 (w), 751 (w), 697 (w) cmÀ1; MS (APCI+): m/
z: 572.10 [M]+; HRMS (APCI+): calcd for C36H16N2O6: 572.1003;
found:
572.1004;
elemental
analysis
calcd
(%)
for
4-Bromo-10-(2’,6’-difluorophenyl)-1,9-dihydroxy-10H-phenoxa-
zine (14): Under an inert atmosphere, compound 13 (287 mg,
crude 0.661 mmol) and CH2Cl2 (16 mL) ware placed in a 30-mL
two-necked flask. The mixture was cooled and stirred at À788C. A
1.0m CH2Cl2 solution of BBr3 (3.5 mL, 3.5 mmol) was added to the
mixture, and the solution was stirred for 12 h as the temperature
was allowed to rise to room temperature. The mixture was then
poured into water (40 mL) and extracted with CH2Cl2 (30 mL”3).
The organic layer was washed successively with water (20 mL) and
a saturated aqueous NaCl solution (20 mL). The organic layer was
dried over Na2SO4, filtered, and concentrated under reduced pres-
sure to give the desired compound as a crude material. The com-
pound was used in the next reaction without further purification
C36H16N2O6·(C6H4Cl2)0.1: C 74.86, H 2.82, N 4.77; found: C 74.67, H
2.89, N 4.76. 13C NMR data could not be obtained due to the low
solubility of the product.
10-(4’-Dihydroxyboryl-2’,6’-difluorophenyl)-1,9-dihydroxy-10H-
phenoxazine (16): Under an inert atmosphere, compound 9
(326 mg, 0.751 mmol) was placed in a 30-mL two-necked round-
bottomed flask. THF (12 mL) was added and the solution was
cooled and stirred at À788C. A 1.6m hexane solution of n-butyl-
lithium (0.65 mL, 1.04 mmol) was then slowly added to the THF so-
lution. After stirring the mixture for 1 h at À788C, triisopropyl
borate (0.24 mL, 1.04 mmol) was added to the flask. The mixture
was then stirred for 3 h as the temperature was allowed to rise to
room temperature. Water (20 mL) and 3m hydrochloric acid (2 mL)
were next added. The mixture was extracted with CH2Cl2 (10 mL”
3) and then washed successively with water (10 mL) and a saturat-
ed aqueous NaCl solution (10 mL). The organic layer was dried
over Na2SO4, filtered, and concentrated under reduced pressure.
The residue was purified by silica gel column chromatography with
CH2Cl2 and then ethyl acetate/methanol (EtOAc/MeOH) (10:1, v/v)
as eluent to afford the desired compound as a colorless solid
(yield: 242 mg, 81%). C20H16BF2NO5; MW 399.16; colorless solid;
m.p. 197–1988C; 1H NMR (400 MHz, [D6]DMSO): d=8.37 (s, 2H),
7.35 (d-like, JH,F =9.7 Hz, 2H), 6.95 (t, JH,H =8.3 Hz, 2H), 6.59 (dd,
(270 mg, quant.). C18H10BrF2NO3; MW 404.98; colorless solid; m.p.
1
225–2268C (decomp.); H NMR (300 MHz, CDCl3): d=7.30 (tt, JH,H
=
8.3, JH,F =6.2 Hz, 1H), 7.22 (d, JH,H =8.9 Hz, 1H), 7.05–6.99 (m, 3H),
6.78 (dd, JH,H =8.3, 1.3 Hz, 1H), 6.61 (dd, JH,H =8.3, 1.3 Hz, 1H), 6.52
(d, JH,H =8.9 Hz, 1H), 5.66 (s, 1H), 5.57 ppm (s, 1H); 13C NMR
(100 MHz, CDCl3): d=161.06 (dd,
JC,F =249.4, 4.0 Hz), 151.42,
149.55, 149.10, 148.63, 130.10, 129.78 (t, JC,F =11.0 Hz), 126.98,
122.96 (t, JC,F =14.3 Hz), 122.18, 120.60, 113.07 (dd, JC,F =21.7,
2.9 Hz), 112.05, 111.86, 109.12, 100.36 ppm; IR (KBr): n=3548 (m),
3534 (br), 1610 (m), 1575 (w), 1488 (s), 1473 (s), 1442 (s), 1341 (m),
1293 (w), 1280 (w), 1268 (w), 1242 (m), 1228 (w), 1201 (m), 1146
(w), 1119 (w), 1063 (w), 1033 (s), 995 (w), 931 (w), 812 (m), 792 (m),
775 (s), 733 (w), 710 (w), 465 (m) cmÀ1; MS (DART+): m/z: 406.03
[M+H]+, 408.03 [M+2+H]+; HRMS (DART+): calcd for
C18H11BrF2NO3 [M+H]+: 405.9885; found: 405.9885.
J
H,H =8.3, 1.2 Hz, 2H), 6.57 (dd, JH,H =8.3, 1.2 Hz, 2H), 3.54 ppm (s,
6H); MS (APCI): m/z: 400.11 [M+H]+; HRMS (APCI+): calcd for
C20H17BF2NO5 [M+H]+: 400.1165; found: 400.1165.
3,4’’’-Bis[10-(2’,6’-difluorophenyl)-1,9-dimethoxy-10H-phenoxa-
zine] (17): Under an inert atmosphere, compound 16 (84 mg,
0.210 mmol), compound 13 (100 mg, 0.230 mmol), and tetrakis(tri-
phenylphosphine)palladium(0) (24 mg, 0.021 mmol) were added to
a 30-mL round-bottomed flask. A degassed 1.0m Na2CO3 aqueous
solution (1.7 mL, 1.70 mmol), degassed EtOH (1.7 mL), and de-
gassed toluene (6 mL) were then added and the mixture was
3-Bromo-2,2’:6’,2“:6”,6-trioxytriphenylamine (15): Under an inert
atmosphere, compound 14 (270 mg, crude 0.661 mmol), K2CO3
(461 mg, 3.34 mmol), and DMF (11 mL) were placed in a 50-mL
round-bottomed flask. The mixture was stirred at 1008C for 12 h.
After cooling to room temperature, water (40 mL) was added to
the reaction mixture and the resulting precipitate was collected by
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16022
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
heated at reflux for 12 h. After cooling to room temperature, the
solvent was removed under reduced pressure. Toluene was added
to the residue and the insoluble material was removed by filtration
through a pad of Celite, which was washed with hot toluene. The
filtrates were combined and concentrated under reduced pressure
and the residue was purified by silica gel column chromatography
using toluene and then CH2Cl2 as eluent to afford the desired com-
pound as a colorless solid (yield: 128 mg, 86%). C40H28F4N2O6; MW
708.67; colorless solid; m.p. >3008C; 1H NMR (400 MHz,
[D6]DMSO): d=7.34 (tt-like, JH,H = JH,F = ca. 8 Hz, 1H), 7.23 (d-like,
for C36H16N2O6: 572.1003; found: 572.1001; elemental analysis calcd
(%) for C36H16N2O6·(C6H4Cl2)0.2: C 74.23, H 2.81, N 4.65; found: C
74.20, H 3.00, N 4.73. 13C NMR data could not be obtained due to
the low solubility of the product.
Synthesis of 42+·2SbF6À: Under an inert atmosphere, compound 4
(22.2 mg,
0.0387 mmol,
1.0 equiv.),
TBPA·SbF6
(58.6 mg,
0.0816 mmol), and CH2Cl2 (10 mL) were placed in a 30-mL two-
necked round-bottomed flask and the mixture was stirred at room
temperature for 1 h. The solvent was removed by evaporation
under reduced pressure and the residue was dissolved in a mini-
mal amount of CH3CN (ca. 10 mL) and then insoluble materials
were removed by filtration using a dismic filter. The filtrate was
placed in a small open bottle, which was placed in a larger bottle
containing Et2O. This larger bottle was capped and placed at room
temperature to allow crystals to form by slowly mixing Et2O vapor
with the CH3CN layer. The crystals were collected by filtration to
afford the desired compound as green crystals (yield: 30 mg, 74%).
C36H16N2O6·(SbF6)2; MW 1044.03; green solid; m.p. >3008C; IR
(KBr): n=3084 (w), 1645 (w), 1592 (s), 1546 (w), 1507 (m), 1494 (m),
1475 (m), 1437 (w), 1373 (w), 1332 (w), 1282 (m), 1273 (m), 1152
(w), 1074 (m), 1050 (w), 1023 (w), 867 (w), 785 (m), 734 (w), 660 (s),
555 (w) cmÀ1; MS (ESI+): m/z: 572.08 [C36H16N2O6]+; MS (ESIÀ): m/z:
J
H,F =9.9 Hz, 2H), 7.13 (d, JH,H =8.7 Hz, 1H), 7.07 (t-like, JH,H = JH,F =
ca. 8 Hz, 2H), 6.98 (t, JH,H = 8.3 Hz, 2H), 6.98 (d, JH,H = 8.3 Hz, 1H),
6.71 (d, JH,H =8.7 Hz, 1H), 6.66–6.59 (m, 5H), 6.48 (d, JH,H =8.3 Hz,
1H), 3.63 (s, 6H), 3.59 (s, 3H), 3.54 ppm (s, 3H); IR (KBr): n=2936
(w), 2839 (w), 1616 (w), 1590 (w), 1496 (m), 1461 (s), 1409 (w), 1285
(s), 1248 (m), 1112 (w), 1094 (s), 1029 (w), 1006 (w), 864 (w), 789
(w), 773 (w), 724 (w), 686 (w), 658 (w) cmÀ1; MS (DART+): m/z:
709.27 [M+H]+; HRMS (DART+): calcd for C40H29F4N2O6: [M+H]+:
709.1956; found: 709.1953. 13C NMR data could not be obtained
due to the low solubility of the product.
3,4’’’-Bis[10-(2’,6’-difluorophenyl)-1,9-dihydroxy-10H-phenoxa-
zine] (18): Under an inert atmosphere, compound 17 (128 mg,
0.181 mmol) and CH2Cl2 (80 mL) was placed in a 100-mL two-
necked flask. The solution was cooled and stirred at À788C. A
1.0m CH2Cl2 solution of BBr3 (2.8 mL, 2.8 mmol) was then added to
the flask and the mixture was stirred for 12 h as the temperature
was allowed to rise to room temperature. The mixture was poured
into water (50 mL) and the organic layer was separated with
CH2Cl2 (20 mL”3). The organic layer was successively washed with
water (20 mL) and a saturated aqueous NaCl solution (20 mL). The
organic layer was dried over Na2SO4, filtered, and concentrated
under reduced pressure to afford the desired compound as a
crude material (yield: 152 mg, crude quant.). The compound was
used in the next reaction without further purification. C36H20F4N2O6;
MW 652.56; colorless solid; m.p. 1738C (decomp.); 1H NMR
(300 MHz, CDCl3): d=7.34–7.23 (m, 3H), 7.10–6.96 (m, 6H), 6.69
(dd, JH,H =8.3, 1.3 Hz, 2H), 6.67–6.59 (m, 5H), 5.87 (br., 1H), 5.76
(br., 1H), 5.64 ppm (br., 1H); IR (KBr): n=3432 (br), 2927 (w), 1619
(s), 1483 (s), 1463 (s), 1410 (m), 1345 (w), 1302 (s), 1272 (m), 1243
(m), 1208 (w), 1096 (w), 1065 (w), 1028(s), 1004 (w), 778 (m), 727
(w), 572 (w) cmÀ1; MS (DART+): m/z: 653.20 [M+H]+; HRMS (APCI+
): calcd for C36H21F4N2O6 [M+H]+: 652.1330; found: 653.1331.
13C NMR data could not be obtained due to the low solubility of
the product.
234.90
[SbF6]À;
elemental
analysis
calcd
(%)
for
C36H16F12N2O6Sb2·CH3CN: C 42.06, H 1.77, N 3.87; found: C 42.36, H
2.09, N 3.75.
Synthesis of 52+·2SbF6À: Under an inert atmosphere, compound 5
(39.9 mg, 0.0699 mmol), TBPA·SbF6 (106 mg, 0.149 mmol), and
CH2Cl2 (20 mL) were placed in a 30-mL two-necked round-bot-
tomed flask. The mixture was stirred at room temperature for 1 h
and then the solvent was removed by evaporation under reduced
pressure. The residue was dissolved in a minimal amount of CH3CN
(ca. 20 mL) and then the insoluble materials were removed by fil-
tration using a dismic filter. Benzene was slowly added to the fil-
trate and the resulting precipitate was collected by filtration to
afford the desired compound as green crystals (yield: 45 mg, 62%).
C36H16N2O6·(SbF6)2; MW 1044.03; green solid; m.p. >3008C; IR
(KBr): n=3097 (w), 1614 (m), 1591 (s), 1580 (s), 1503 (s), 1327 (w),
1275 (s), 1088 (w), 1077 (w), 1064 (w), 1035 (m), 997 (w), 821 (w),
786 (m), 726 (w), 659 (s), 555 (w) cmÀ1; MS (ESI+): m/z: 286.05
[C36H16N2O6]2+, 572.08 [C36H16N2O6]+; MS (ESIÀ): m/z: 234.90 [SbF6]À;
elemental analysis calcd (%) for C36H16F12N2O6Sb2: C 41.42, H 1.54,
N 2.68; found: C 41.49, H 1.68, N 2.92.
Synthesis of 62+·2SbF6À: Under an inert atmosphere, compound 6
(35.4 mg, 0.0611 mmol), TBPA·SbF6 (88 mg, 0.123 mmol), and
CH2Cl2 (18 mL) were placed in a 50-mL two-necked round-bot-
tomed flask and the mixture was stirred at room temperature for
1 h. The solvent was removed by evaporation under reduced pres-
sure and the residue was dissolved in CH3CN (12 mL). The insoluble
materials were then removed by filtration using a dismic filter. The
filtrate was placed in a small open bottle, which was placed in a
larger bottle containing Et2O. The larger bottle was capped and
placed at room temperature. Crystals were formed by slowly
mixing Et2O vapor with the CH3CN layer. The crystals were collect-
ed by filtration to afford the desired compound as green crystals
(46 mg, 71%). C36H16N2O6·(SbF6)2; MW 1044.03; green solid; m.p.
>3008C; IR (KBr): n=3103 (w), 1613 (m), 1518 (w), 1502 (s), 1479
(w), 1328 (w), 1304 (w), 1272 (s), 1225 (w), 1077 (m), 1063 (w), 1032
3,4’’’-Bis(2,2’:6’,2“:6”,6-trioxytriphenylamine) (6): Under an inert
atmosphere, a crude compound 18 (152 mg, crude 0.181 mmol),
K2CO3 (301 mg, 2.18 mmol), and DMF (10 mL) were placed in a 50-
mL round-bottomed flask and the mixture was stirred at 1008C for
12 h. After cooling to room temperature, water (30 mL) was added
to the reaction mixture and the resulting precipitate was collected
by filtration. The precipitate was washed with water, EtOH, and
hexane. The resulting solid (91 mg) was dissolved in hot o-dichlor-
obenzene (20 mL) and insoluble materials were removed by filtra-
tion through a filter paper. The filtrate was cooled to room temper-
ature and the resulting precipitate was collected by filtration to
afford the desired compound as a yellow solid (yield: 39 mg, 38%
over two steps). C36H16N2O6; MW 572.53; yellow solid; m.p.
>3008C; 1H NMR (400 MHz, [D4]o-dichlorobenzene): d=6.60–6.46
(m, 5H), 6.43 (t, J=8.2 Hz, 1H), 6.33–6.22 ppm (m, 9H), overlap
with solvent peak, assumed to be 1H; IR (KBr): n=2924 (w), 1624
(w), 1597 (w), 1537 (m), 1518 (s), 1483 (s), 1465 (m), 1422 (w), 1378
(w), 1346 (m), 1307 (w), 1259 (s), 794 (w), 751 (s), 698 (m), 626 (w),
571 (w) cmÀ1; MS (APCI+): m/z: 572.10 [M]+; HRMS (APCI+): calcd
(m), 784 (m), 659 (s) cmÀ1; MS (ESI+): m/z: 286.05 [C36H16N2O6]2+
,
572.08 [C36H16N2O6]+; MS (ESIÀ): m/z: 234.90 [SbF6]À; elemental
analysis calcd (%) for C36H16F12N2O6Sb2·(CH3CN)1.5: C 42.37, H 1.87, N
4.43; found: C 42.42, H 1.98, N 4.19.
CCDC 1555772, 1555770, 1555771, and 1555773 contain the sup-
plementary crystallographic data for this paper. These data can be
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16023
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
A 2007, 111, 2951–2956; f) A. Ito, K. Tanaka, Pure Appl. Chem. 2010, 82,
979–989; g) T. Michinobu, J. Inui, H. Nishide, Polym. J. 2010, 42, 575–
582; h) Y. Yokoyama, D. Sakamaki, A. Ito, K. Tanaka, M. Shiro, Angew.
Chem. Int. Ed. 2012, 51, 9403–9406; Angew. Chem. 2012, 124, 9537–
9540; i) D. Sakamaki, A. Ito, K. Tanaka, K. Furukawa, T. Kato, M. Shiro,
Angew. Chem. Int. Ed. 2012, 51, 8281–8285; Angew. Chem. 2012, 124,
8406–8410; j) I. S. Peret¸eanu, T. J. J. Müller, Org. Biomol. Chem. 2013, 11,
5127–5135.
obtained free of charge from The Cambridge Crystallographic Data
Centre.
Acknowledgements
This work was partially supported by Grant-in-Aid for Scientific
Research on Innovative Areas “p-System Figuration: Control of
Electron and Structural Dynamism for Innovative Functions
(2601)” (JSPS KAKENHI Grant Number JP26102005 for S.S.) and
“Stimuli-responsive Chemical Species for the Creation of Func-
tional Molecules (2468)” (JSPS KAKENHI Grant Number
JP15H00956 for K.O.), a Grant-in-Aid for Scientific Research
(JSPS KAKENHI Grant Number JP26288041 for K.O. and
JP17K05783 for S.S.), and the Iketani Science and Technology
Foundation (for S.S.). The authors thank Dr. Rika Tanaka of the
X-ray Crystal Analysis Laboratory, Graduate School of Engineer-
ing Osaka City University for X-ray crystal structure analysis,
and Prof. Dr. Hiroyuki Miyake and Prof. Dr. Satoshi Shinoda of
the Graduate School of Science, Osaka City University for
measuring the UV/Vis/NIR spectra. The authors also thank Hiro
Matsui (Osaka City University) for the calculation of zero-field
parameters.
[7] a) R. J. Bushby, D. R. McGill, K. M. Ng, N. Taylor, Chem. Commun. 1996,
2641–2642; b) R. J. Bushby, D. R. McGill, K. M. Ng, N. Taylor, J. Chem.
Soc. Perkin Trans. 2 1997, 1406–1414.
[8] a) S. Zheng, S. Barlow, C. Risko, T. L. Kinnibrugh, V. N. Khrustalev, S. C.
Jones, M. Y. Antipin, N. M. Tucker, T. V. Timofeeva, V. Coropceanu, J.-L.
BrØdas, S. R. Marder, J. Am. Chem. Soc. 2006, 128, 1812–1817; b) X.
Zheng, X. Wang, Y. Qui, Y. Li, C. Zhou, Y. Sui, Y. Li, J. Ma, X. Wang, J. Am.
Chem. Soc. 2013, 135, 14912–14915; c) K. Kamada, S. Fuku-en, S. Mina-
mide, K. Ohta, R. Kishi, M. Nakano, H. Matsuzaki, H. Okamoto, H. Higa-
shikawa, K. Inoue, S. Kojima, Y. Yamamoto, J. Am. Chem. Soc. 2013, 135,
232–241; d) Y. Su, X. Wang, X. Zheng, Z. Zhang, Y. Song, Y. Sui, Y. Li, X.
Wang, Angew. Chem. Int. Ed. 2014, 53, 2857–2861; Angew. Chem. 2014,
126, 2901–2905.
[9] a) M. Bieri, S. Blankenburg, M. Kivala, C. A. Pignedoii, P. Ruffieux, K.
Müllen, R. Fasel, Chem. Commun. 2011, 47, 10239–10241; b) F. Schlütter,
F. Rossel, M. Kivala, V. Enkelmann, J.-P. Gisselbrecht, P. Ruffieux, R. Fasel,
K. Müllen, J. Am. Chem. Soc. 2013, 135, 4550–4557; c) K. Schmoltner, F.
Schlültter, M. Kivala, M. Baumgarten, S. Winkler, R. Trattnig, N. Koch, A.
Klug, E. J. W. List, K. Müllen, Polym. Chem. 2013, 4, 5337–5344.
[10] E. T. Seo, R. F. Nelson, J. M. Fritsch, L. S. Marcoux, D. W. Leedy, R. N.
Adams, J. Am. Chem. Soc. 1966, 88, 3498–3503.
[11] For instance, see: a) R. I. Walter, J. Am. Chem. Soc. 1955, 77, 5999–6002;
b) F. A. Bell, A. Ledwith, D. C. Sherrington, J. Chem. Soc. C 1969, 2719–
2720; c) G. M. Brown, G. R. Freeman, R. I. Walter, J. Am. Chem. Soc. 1977,
99, 6910–6915; d) O. Yurchenko, D. Freytag, L. zur Borg, R. Zentel, J.
Heinze, S. Ludwigs, J. Phys. Chem. B 2012, 116, 30–39.
Conflict of interest
The authors declare no conflict of interest.
[12] a) D. Hellwinkel, M. Melan, Chem. Ber. 1974, 107, 616–626; b) S. Bam-
berger, D. Hellwinkel, F. A. Neugebauer, Chem. Ber. 1975, 108, 2416–
2421.
Keywords: electronic structure · magnetic properties · radical
ions · structure elucidation
[13] a) M. Kuratsu, M. Kozaki, K. Okada, Chem. Lett. 2004, 33, 1174–1175;
b) M. Kuratsu, M. Kozaki, K. Okada, Angew. Chem. Int. Ed. 2005, 44,
4056–4058; Angew. Chem. 2005, 117, 4124–4126; c) M. Kuratsu, S.
Suzuki, M. Kozaki, D. Shiomi, K. Sato, T. Takui, K. Okada, Inorg. Chem.
2007, 46, 10153–10157.
[14] The radical cation 1+ was recently isolated by Wang and co-workers,
see ref. [7b].
[1] a) Y. Shirota, J. Mater. Chem. 2000, 10, 1–25; b) M. Thelakkat, Macromol.
Mater. Eng. 2002, 287, 442–461; c) Y. Shirota, H. Kageyama, Chem. Rev.
2007, 107, 953–1010; d) Y. Tao, C. Yang, J. Qin, Chem. Soc. Rev. 2011, 40,
2943–2970.
[2] a) N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910; b) E. Stec-
khan, Top. Curr. Chem. 1987, 142, 1–69.
[15] a) M. Kuratsu, S. Suzuki, M. Kozaki, D. Shiomi, K. Sato, T. Takui, T. Kanza-
wa, Y. Hosokoshi, X.-Z. Lan, Y. Miyazaki, A. Inaba, K. Okada, Chem. Asian
J. 2012, 7, 1604–1609; b) S. Suzuki, A. Nagata, M. Kuratsu, M. Kozaki, R.
Tanaka, D. Shiomi, K. Sugisaki, K. Toyota, K. Sato, T. Takui, K. Okada,
Angew. Chem. Int. Ed. 2012, 51, 3193–3197; Angew. Chem. 2012, 124,
3247–3251.
[16] A. G. Sergeev, J. F. Hartwig, Science 2011, 332, 439–443.
[17] So far, 3 has been synthesized from commercially available compounds
with more than 10 steps (see ref. [13b]).
[3] a) P. Bujak, I. Kulszewicz-Bajer, M. Zagorska, V. Maurel, I. Wielgus, A.
Pron, Chem. Soc. Rev. 2013, 42, 8895–8999; b) S. C. Blackstock, T. D.
Selby, in Magnetic Properties of Organic Materials (Ed.: P. M. Lahti),
Marcel Dekker, New York, 1999, chap. 9, pp. 165–178; c) R. J. Bushby, in
Magnetic Properties of Organic Materials (Ed.: P. M. Lahti), Marcel Dekker,
New York, 1999, chap. 10, pp. 179–195; d) S. Suzuki, K. Okada, in Or-
ganic Redox Systems (Ed.: T. Nishinaga), Wiley, New Jersey, 2016,
chap. 8, pp. 269–285.
[4] a) C. Lambert, G. Nçll, J. Am. Chem. Soc. 1999, 121, 8434–8442; b) Y.
Hirao, M. Urabe, A. Ito, K. Tanaka, Angew. Chem. Int. Ed. 2007, 46, 3300–
3303; Angew. Chem. 2007, 119, 3364–3367; c) C. Lambert, in Organic
Redox Systems (Ed.: T. Nishinaga), Wiley, New Jersey, 2016, chap. 7,
pp. 245–268.
[5] a) A. Wakamiya, H. Nishimura, T. Fukushima, F. Suzuki, A. Saeki, S. Seki, I.
Osaka, T. Sasamori, M. Murata, Y. Murata, H. Kaji, Angew. Chem. Int. Ed.
2014, 53, 5800–5804; Angew. Chem. 2014, 126, 5910–5914; b) H. Nishi-
mura, N. Ishida, A. Shimazaki, A. Wakamiya, A. Saeki, L. T. Scott, Y.
Murata, J. Am. Chem. Soc. 2015, 137, 15656–15659.
[6] a) K. Yoshizawa, A. Chano, A. Ito, K. Tanaka, T. Yamabe, H. Fujita, J. Ya-
mauchi, M. Shiro, J. Am. Chem. Soc. 1992, 114, 5994–5998; b) K. Yoshiza-
wa, H. Hatanaka, H. Ago, K. Tanaka, T. Yamane, Bull. Chem. Soc. Jpn.
1996, 69, 1417–1422; c) K. Sato, M. Yano, M. Furuichi, D. Shiomi, T.
Takui, K. Abe, K. Itoh, A. Higuchi, K. Katsuma, Y. Shirota, J. Am. Chem.
Soc. 1997, 119, 6607–6613; d) K. Sato, M. Furuichi, M. Yano, D. Shiomi,
K. Abe, T. Takui, K. Itoh, A. Higuchi, K. Katsuma, Y. Shirota, Synth. Met.
1997, 85, 1719–1720; e) Y. Hirao, H. Ino, A. Ito, K. Tanaka, J. Phys. Chem.
[18] Crystallographic data for 4: C36H16N2O6, triclinic, space group PÀ1 (#2),
a=3.720(3), b=15.890(15), c=19.630(19), a=84.72(2), b=86.55(2), g=
86.75(3)8, V=1151.8(19) Š3, Z=2, 1calcd =1.651 gcmÀ3, T=120 K, R1=
0.0763, Rw=0.2002, GOF=1.121 (CCDC: 1555772).
À
[19] Crystallographic
data
for
42+·2SbF6
:
C36H16N2O6·(SbF6)2·(C2H3N)·(C4H10O)0.5
,
monoclinic, space group C2/c
(#15), a=27.7283(11), b=17.3161(5), c=17.7859(7) Š, b=109.0532(15)8,
V=8072.05) Š3, Z=8, 1calcd =1.847 gcmÀ3, T=200 K, R1=0.0467, Rw=
0.1134, GOF=1.102 (CCDC: 1555770).
[20] Crystallographic data for 52+·2SbF6À: C18H8NO3·SbF6, monoclinic, space
group C2/c (#15), a=12.969(9), b=9.574(7), c=28.13(2) Š, b=
97.187(9)8, V=3465(4) Š3, Z=8, 1calcd =2.001 gcmÀ3
0.0639, Rw=0.1223, GOF=1.174 (CCDC: 1555771).
, T=150 K, R1=
[21] Crystallographic data for 62+·2SbF6À: C36H16N2O6·(SbF6)2·(C2H3N)1.5, mon-
oclinic, space group C2/c (#15), a=20.444(7), b=22.436(7), c=
17.411(6) Š, b=104.720(5)8, V=7724(4) Š3, Z=8, 1calcd =1.902 gcmÀ3
T=173 K, R1=0.0414, Rw=0.1087, GOF=1.055 (CCDC: 1555773).
,
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16024
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À
[22] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[24] We used the same sample of 42+·2SbF6 using its SQUID measurement.
[25] a) F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73–78; b) F. Neese, ORCA-
An Ab Initio, Density Functional and Semiempirical Program Package,
Version 2.9, Max Planck Institute for Bioinorganic Chemistry, Mulheim
(Germany), 2012.
[26] B. Bleaney, K. D. Bowers, Proc. R. Soc. London, Ser. A 1952, 214, 451–
465.
[27] K. Yamaguchi, T. Kawakami, Y. Takano, Y. Kitagawa, Y. Yamashita, H.
Fujita, Int. J. Quantum Chem. 2002, 90, 370–385.
Manuscript received: July 12, 2017
[23] The predominant contributing transitions are transition 1 [142b!147b
(c=0.62871)] and transition 2 [144b!147b (c=0.66776)]. Details of the
TD-DFT calculations are provided in the Supporting Information.
Accepted manuscript online: September 3, 2017
Version of record online: October 19, 2017
Chem. Eur. J. 2017, 23, 16014 –16025
www.chemeurj.org
16025
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim