Impact of diradical character on two-photon absorption: Bis(acridine) dimers synthesized from an allenic precursor

Article abstract of DOI:10.1021/ja308396a

The two-photon absorption (TPA) of a bis(acridine) dimer (8) having singlet diradical character in its ground state was found to be enhanced by more than 2 orders of magnitude as compared to its closed-shell counterpart (12), which has the same structural backbone and atom composition. The dimer, a tetracation species consisting of two connected acridinium cation moieties with high coplanarity, was obtained during our attempts to synthesize triplet carbenes by double oxidation of an allenic precursor (3b). High conjugation over the two aromatic rings connected by dimerization was revealed by X-ray analysis, and a small HOMO-LUMO gap was found in the visible-near-infrared one-photon absorption spectrum in solution and in the crystalline state, exhibiting that the ground state of 8 has singlet diradical nature. Ab initio molecular orbital calculations of the ground state also suggested that 8 has an intermediate diradical character (y) of 0.685. Interconversion between diradical tetracation dimer 8 and closed-shell dication dimer 12 was achieved by oxidation/reduction in good yields and was accompanied by formation of monoradical trication dimer 13 as an intermediate. TPA measurements at near-infrared wavelengths revealed that diradical dimer 8 has large TPA cross sections (3600 GM at 1200 nm), while closed-shell 12 has TPA cross sections of <21 GM. This result represents a straightforward comparison between the TPA activity of molecules with the same structural backbone and atom composition but with different degree of the diradical character, supporting the theoretical prediction that enhanced TPA intensity can be observed in the intermediate y region (0 < y < 1).

Full text of DOI:10.1021/ja308396a

Article
pubs.acs.org/JACS
Impact of Diradical Character on Two-Photon Absorption:
Bis(acridine) Dimers Synthesized from an Allenic Precursor
Kenji Kamada,*,† Shin-ichi Fuku-en,‡ Shu Minamide,†,§ Koji Ohta,† Ryohei Kishi,§ Masayoshi Nakano,§
Hiroyuki Matsuzaki,∥,⊥ Hiroshi Okamoto,∥ Hiroyuki Higashikawa,‡ Katsuya Inoue,‡ Satoshi Kojima,‡
and Yohsuke Yamamoto*,‡
†Research Institute for Ubiquitous Energy Devices, National Institute of Advanced Industrial Science and Technology (AIST), Ikeda,
Osaka 563-8577, Japan
‡Department of Chemistry, Graduate School of Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526,
Japan
§Department of Materials Engineering Science, Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka
560-8531, Japan
∥Department of Advanced Material Science, Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Chiba 277-8561,
Japan
S
* Supporting Information
ABSTRACT: The two-photon absorption (TPA) of a
bis(acridine) dimer (8) having singlet diradical character in
its ground state was found to be enhanced by more than 2
orders of magnitude as compared to its closed-shell counter-
part (12), which has the same structural backbone and atom
composition. The dimer, a tetracation species consisting of two
connected acridinium cation moieties with high coplanarity,
was obtained during our attempts to synthesize triplet carbenes
by double oxidation of an allenic precursor (3b). High
conjugation over the two aromatic rings connected by
dimerization was revealed by X-ray analysis, and a small HOMO−LUMO gap was found in the visible-near-infrared one-
photon absorption spectrum in solution and in the crystalline state, exhibiting that the ground state of 8 has singlet diradical
nature. Ab initio molecular orbital calculations of the ground state also suggested that 8 has an intermediate diradical character
(y) of 0.685. Interconversion between diradical tetracation dimer 8 and closed-shell dication dimer 12 was achieved by
oxidation/reduction in good yields and was accompanied by formation of monoradical trication dimer 13 as an intermediate.
TPA measurements at near-infrared wavelengths revealed that diradical dimer 8 has large TPA cross sections (3600 GM at 1200
nm), while closed-shell 12 has TPA cross sections of <21 GM. This result represents a straightforward comparison between the
TPA activity of molecules with the same structural backbone and atom composition but with different degree of the diradical
character, supporting the theoretical prediction that enhanced TPA intensity can be observed in the intermediate y region (0 < y
< 1).
intermolecular covalent bonds originating from the open-shell
character.6,7
1. INTRODUCTION
The concept of singlet diradicals is key to understanding the
nature of chemical bonds; thus, their electronic properties have
been investigated extensively.1 However, it has only been
several years since the singlet diradical nature of aromatic
hydrocarbons has attracted considerable attention because of
extensive progress in their theory, synthesis, and functionality.2
Prediction of the diradical nature of oligoacenes by Bendikov et
al.3 stimulated further theoretical4 and synthetic5 studies of
oligoacenes with longer π-conjugation. Thermodynamic
stabilization of the singlet diradical state was realized by
Kubo et al. through studies of the bis(phenalenyl) structure.6
The bis(phenalenyl) diradicaloids have significant singlet
diradical nature and exhibit the coexistence of intra- and
Clearly, molecules with partial singlet diradical nature in their
ground state, that is, singlet diradicaloids, are expected to have
interesting physical properties (electronic, optical, magnetic,
etc.) in addition to their structural uniqueness. Indeed, a
remarkable enhancement of the molecular third-order non-
linear optical (NLO) property, that is, second hyperpolariz-
ability (γ), was theoretically predicted in the intermediate bond
breaking region with partial singlet diradical character for the
H2 molecule.8 The ground-state electronic structure of singlet
diradicaloids is characterized by the diradical character, y, which
Received: August 30, 2012
Published: December 6, 2012
© 2012 American Chemical Society
232
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
is defined by the occupation number of the lowest unoccupied
natural orbital (LUNO) having antibonding nature. The y value
therefore represents the “degree” of the singlet diradical
character, that is, bond weakness (y = 0 for the closed-shell
and y = 1 for pure singlet diradicals), so that, in other words, 1
− y indicates an “effective bond order”. The theoretical
assessments of γ for singlet diradicaloid model molecules9a,b
revealed that γ has a bell-shape dependence as a function of the
diradical character y. This relationship has been theoretically
proven on the basis of the valence configuration interaction
two-site model9c,d not only in static γ but also in two-photon
absorption (TPA) cross section, which is the molecular
quantity proportional to the imaginary part of γ and an
important property for various potential applications.10 This
can be qualitatively understood by the perturbative formula of γ
≈ (μg,e1)2(μe1,e2)2/(Ee1)2(Ee2),9b,c as well as by the y depend-
ences of excitation energies (Ee1, Ee2) and transition moments
(μg,e1, μe1,e2) concerning the ground (g) and low-lying excited
states (e1, e2); that is, the increase in the y leads to the decrease
of μg,e1 and En, while to the increase of μe1,e2. The relationship
has also been exemplified using ab initio MO and density
functional theory calculations for model and real molecular
systems, for example, polycyclic aromatic hydrocarbons
including thermally stable bis(phenalenyl) systems,11a graphene
nanoflakes,11b and transition-metal systems.11c Originally, y is
defined in a purely theoretical manner;12 however, the
analytical formula to estimate the y value from experimentally
accessible quantities was proposed to fill the gap between
theoretical predictions and experimental observations obtained
from one- and two-photon absorption spectra as well as from
phosphorescence and electron spin resonance peaks.13
Experimental results revealed that bis(phenalenyl)-based
diradicaloids have the TPA cross sections of 1 or 2 orders of
magnitude larger than those of aromatic hydrocarbons that
have similar π-conjugation length, but less diradical character.14
However, such comparisons have left ambiguities because the
structures of the diradicaloid and closed-shell molecules are not
the same.
the oxidation of allene 3b gave rise to tetracation 8, an
unexpected product having partial singlet diradical character,
through demethylation and dimerization (Figure 1).18,19 Dimer
8 can have the resonance structures 8A, the closed-shell
quinoid, and diradical nonquinoidal form 8B, similar to the case
of Chichibabin’s hydrocarbon,20 a widely known example of a
diradicaloid (Figure 2). Moreover, chemical reduction of 8 gave
monoradical and closed-shell compounds with tricationic and
dicationic charges, respectively. This means that the redox
reaction can change the diradical character of the molecules
without changing the structural backbone. This property is ideal
for studying the relationship between the TPA property and
diradical character by direct comparison.
In this Article, we report on the synthesis, diradical nature,
and TPA properties of dimer 8 and its closed-shell (12) and
monoradical (13) counterparts. The diradical nature of dimer 8
is discussed on the basis of X-ray structural analysis, one-
photon absorption measurements, and theoretical calculations.
These results support that 8 has significant diradical nature with
an intermediate magnitude of y = 0.685. TPA measurements by
the femtosecond Z-scan method revealed that diradical dimer 8
has intense TPA (maximum TPA cross section σ(2) = 3600 GM
at 1200 nm, where 1 GM = 10−50 cm4 s molecule−1 photon−1).
On the other hand, the σ(2) of closed-shell counterpart 12 was
less than 20 GM. This high contrast in the value of σ(2) by more
than 2 orders of magnitudes, resulting from the change in the
value of y, clearly supports the theoretical prediction of
enhancement of σ(2) with intermediate diradical character y.
This is the first direct comparison, to the best of our
knowledge, between a singlet diradicaloid (8, y = 0.682) and
its closed-shell counterpart (12, y = 0.000), which have the
same backbone structure and atom composition (but different
charged state), although comparisons between molecules
having similar π-conjugation (but different backbones or
atom compositions) have been previously demonstrated.13,14,21
It is here noted that the introduction of positive charges tends
to mainly affect the contribution of type II virtual excitation
process (negative) to γ,22 while the enhancement of γ in the
intermediate diradical character region is found to be primarily
caused by that of type III (positive) process.9b,c Thus, such
differences in charged states are not predicted to provide a
strong influence on the diradical character dependences of γ in
the present compounds.
During our attempts to synthesize stable triplet carbenes for
other purposes, we serendipitously obtained a series of
compounds having different levels of singlet diradical character
with the same structural backbone. The previous approach for
preparing triplet carbenes was photolysis of diazo compound 1
to generate the relatively persistent triplet carbene 2 (Scheme
1).15,16 As structural elucidation of 2 in a pure state has not
been achieved,17 we tried a new strategy based on the two-
electron oxidation of allenic precursors (Scheme 2). We tested
the chemical oxidation of allenic compounds 3a, 3b, and 5, but
did not observe any triplet carbenes. However, we found that
The syntheses of diradical dimer 8 and its precursors are
presented in sections 2.1 and 2.2, followed by the evaluation of
the diradical nature of 8 in section 2.3. Derivation of its closed-
shell counterpart 12 and monoradical species 13 is described in
section 2.4, and plausible mechanisms of the reactions are
discussed in section 2.5. The one- and two-photon absorption
properties of 8, 12, and 13 are presented in sections 2.6 and
2.7, respectively. The conclusions are given in section 3. The
synthetic, experimental, and calculation procedures are
presented in the Supporting Information.
Scheme 1. Reported Method for the Synthesis of the Most
Persistent Triplet Carbene Known
2. RESULTS AND DISCUSSION
2.1. Synthesis of Allenic Precursor 3b. The allenic
precursor (3b) was prepared as shown in Scheme 3. The key
intermediate, acridone 10, was synthesized by a procedure
similar to that described previously.23 The methoxy-substituted
triarylamine (9) was synthesized by the Buchwald−Hartwig
cross-coupling reaction.24 Selective dilithiation of amine 9 upon
reaction with n-BuLi in refluxing n-hexane, followed by
treatment with methyl chloroformate, afforded the desired
233
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
Scheme 2. Our Strategy To Generate Stable Triplet Carbenes
Figure 1. Isolated dimeric species 8.
acridone 10 in 31% over three steps. Acridone 10 was easily
converted to the corresponding olefin 11 by the Wittig
reaction. The reaction25 of acridone 10 with olefin 11 in the
presence of Tf2O followed by treatment with DBU afforded the
desired allene 3b in 47% yield. Compound 3b showed a signal
characteristic of allenic carbons in the 13C NMR spectrum at
216.2 ppm. X-ray analysis confirmed the structure and revealed
the presence of a bent allene framework with an angle of
168.0(2)°, which must have resulted from steric crowding
caused by the two acridine units (Figure 3).
Figure 2. Chichibabin’s hydrocarbon.
Scheme 3. Synthesis of Allenic Precursor 3b
2.2. Oxidation of Allenic Precursor and Crystal
Structure of Isolated Dimer Species. The oxidation of 3b
with 2 equiv of (4-BrC6H4)3N•+SbCl6 in CH2Cl2 afforded a
−
dark purple solid. X-ray analysis of single crystals obtained by
recrystallization from CH3CN/ether revealed the compound to
be 8·4SbCl4. It can be rationalized that this compound resulted
from demethylation and dimerization of the reactant allene 3b
(Figure 4).26 A small amount of impurity that was later revealed
by UV−vis−NIR measurements to be trication monoradical 13,
resulting from incomplete oxidation, also formed and could not
be removed fully from 8·4SbCl4, even upon recrystallization.
Probably because of the presence of this species, which is
capable of undergoing rapid electron transfer with 8·4SbCl4, the
234
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
As for 8, the presence of four counteranions per molecule
implied that the cationic organic moiety was a tetracation. The
two connected acridinium units were found to form a highly
coplanar structure with the dihedral angles of the two
−
acridinium units being 0.2(1)° (X = SbCl4 ) and 2.0(1)° (X
−
= CHB11H5Cl6 ), and the sum of the bond angles around
−
nitrogen being 360.1(4)° (X = SbCl4 ) and 359.9(3)° (X =
−
CHB11H5Cl6 ). The bond lengths between the two acridinium
−
cation moieties were C2−C2* = 1.42(1) Å (X = SbCl4 ) and
−
1.419(6) Å (X = CHB11H5Cl6 ), and were longer than that of a
typical double bond (1.34 Å), but shorter than that of a typical
biphenyl single bond (1.48 Å). This suggests the contribution
of both the quinoidal form 8A with closed-shell structure and
the nonquinoidal form 8B with singlet diradical character, as
shown in Figure 1. A similar phenomenon was also observed in
the case of Chichibabin’s hydrocarbon20 (Figure 2) in which
the bond connecting the two phenylene moieties (1.448(4) Å)
was also longer than a typical double bond. This interesting X-
ray structure of 8 prompted us to study its singlet diradical
character in more detail.
Figure 3. Molecular structure of 3b. Hydrogen atoms, 4-t-BuC6H4
groups, and incorporated solvent molecules have been omitted for
clarity. Thermal ellipsoids are drawn at the 50% probability level.
Selected bond lengths [Å] and angles [deg] for 3b: C12−C26 =
1.311(4), C26−C38 = 1.317(3), C12−C26−C38 = 168.0(2).
2.3. Evaluation of the Singlet Diradical Character of
Isolated Dimer 8 Based upon X-ray Structure and
Computational Analysis. To establish a better understanding
of the ground state of dimer 8, single-point calculations of its
singlet and triplet states were performed at the BHandHLYP/6-
31G(d) level. The molecular geometry of dimer 8·4SbCl4
obtained from X-ray diffraction data was used excluding the
counteranions and incorporated solvent molecules. Surpris-
ingly, the most stable electronic state is that of the open-shell
singlet (OS), according to calculations by the broken-symmetry
(BS) DFT method. The closed-shell singlet (CS) solution
obtained by the spin-restricted DFT method is higher in energy
than the OS solution corrected by the approximate spin-
projection (AP) method27 using both the open- and the closed-
Figure 4. Molecular structure of 8·4SbCl4. Hydrogen atoms, 4-t-
BuC6H4 groups, counteranions, and incorporated solvent molecules
have been omitted for clarity. Thermal ellipsoids are drawn at the 50%
probability level. Selected bond lengths [Å] and angles [def] for
8·4SbCl4: N1−C5 = 1.336(10), C1−C2 = 1.405(12), C2−C3 =
1.441(12), C3−C4 = 1.379(10), C4−C5 = 1.410(12), C5−C13 =
1.411(12), C13−C1 = 1.366(10), C2−C2* = 1.420(10), C1−C2−
C2*−C3* = 0.19(12), C12−C14 = 1.374(12), C14−C37 =
1.516(12), C12−C14−C37 = 137.4(4).
shell solutions (ΔEAP
= −13.7 kcal/mol). The energy of
OS‑CS
the triplet state is also higher than that of the OS solution
corrected by the AP method (ΔEAPOS‑T = −8.9 kcal/mol). The
diradical character y of the dimer, estimated by the occupancy
of the lowest unoccupied natural orbital (LUNO) at the
BHandHLYP/6-31G(d) level, is 0.615, indicating that the
species has intermediate character between a pure singlet
diradical and a closed-shell species.12 The spin density
distribution in the open-shell singlet was also calculated at
the UBHandHLYP/6-31G(d) level (Figure 5), and it was
NMR spectrum was extensively broadened. A radical impurity
was detected by measuring the ESR spectrum of crude
8·4SbCl4 (see the Supporting Information). Nonetheless, the
single crystals obtained were sufficient for X-ray structural
analysis. In the hope that pure samples might be obtained,
−
−
counteranion exchange of SbCl4 with CHB11H5Cl6 was
carried out to furnish 8·4CHB11H5Cl6, which was also
structurally examined. Although 8·4CHB11H5Cl6 could not be
obtained in a pure state as desired, during the recrystallization
process, crystals with shapes different from those of
8·4CHB11H5Cl6 also fortuitously crystallized out, and they
were determined to be pure samples of 13·3CHB11H5Cl6. The
parent ion peak of 13 was observed at m/z 492.2300, which is
in good agreement with the peak expected for trication
C
100H92N4O83+ (calculated value: 492.2300) using HR-ESI-MS.
Figure 5. The full spin density distribution of 8 in the open-shell
singlet state. Only the central part of the structure, where the
distribution is concentrated, is shown. The calculations were
performed at the UBHandHLYP/6-31G(d) level of theory. Spin
densities are drawn at the 0.002 au threshold level.
1
The H NMR spectrum was silent, suggesting the existence of
an unpaired electron in the molecule. However, single crystals
suitable for crystallographic measurements could not be
obtained.
235
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
Table 1. Selected Bond Lengths (Å) for Tetracation Dimer 8·4SbCl4 from Its X-ray Structure and for the Optimized Structure of
Tetracation Dimer 8 Calculated at the RB3LYP/6-31G(d) and UB3LYP/6-31G(d) Levels of Theory
a
N−Cipso/Å
Cp1−Cp2/Å
Cα−Cβ/Å
avg Cipso−Co and Cm−Cp/Å
avg Co−Cm /Å
bond length difference
b
X-ray structure
closed-shell singlet
bd
1.336(10)
1.352
1.42(1)
1.428
1.451
1.474
1.374(2)
1.394
1.417(12)
1.429
1.373(10)
1.377
0.044(22)
0.052
bc
,
,
open-shell singlet
bd
1.363
1.402
1.420
1.384
0.036
,
triplet
1.374
1.410
1.412
1.390
0.022
a
b
Difference between the average of all Cipso−Co and Cm−Cp bond lengths and the average of Co−Cm bond lengths. Bond lengths are the average of
c
d
corresponding bonds of the two acridinium moieties. Calculated at the B3LYP/6-31G(d) level. Calculated at the UB3LYP/6-31G(d) level.
found that the largest spin densities are observed on the carbon
atoms connecting the planar dimeric core to the peripheral
acridinium substituents (formerly the allenic carbon atoms of
the precursor) and the nitrogen atoms. Thus, this species is best
described as a ca. 1:1 resonance hybrid of quinoidal structure
8A and diradical structure 8B.
species, redox reactions of the dimer were investigated.
Treatment of tetracation dimer 8·4SbCl4 with excess Zn
metal afforded dication dimer 12·2SbCl4 in 80% yield (Scheme
−
4). After counteranion exchange of SbCl4 for TfO−, the
Scheme 4. Redox Reaction of the Dimer Species
The open-shell character of tetracation dimer 8 was further
examined from the viewpoint of geometric structure. Geometry
optimizations of tetracation dimer 8 were performed at the
RB3LYP/6-31G(d) level for the singlet state, and at the
UB3LYP/6-31G(d) level for the BS singlet and triplet states.
The y value of the BS singlet structure (optimized at the
UB3LYP/6-31G(d) level) is 0.685 (at the UBHandHLYP/6-
31G(d) level), which is consistent with the LUNO occupation
of 0.615 calculated at the UBHandHLYP level on the full
structure based on X-ray analysis. The average bond lengths of
selected bonds and average bond length differences for
alternating bonds are summarized in Table 1 (see the
Supporting Information for details). The electronic structure
of 8 can be described by the resonance between the quinoidal
form 8A with bond alternation (closed-shell) and the
nonquinoidal form 8B with pure singlet diradical character.
We therefore focused on the bond length alternations of the
benzene ring fused to a furan ring in the quinoidal form 8A,
that is, Co−Cm versus Cipso−Co and Cm−Cp. The differences in
the averages of these bonds, that is, 0.044(22) Å (the X-ray
structure) and 0.036 Å (the optimized open-shell singlet
structure), are found to lie between those for the optimized
closed-shell (0.052 Å) and pure diradical triplet (0.022 Å)
structures, suggesting an intermediate diradical character in the
singlet ground state of 8.
To obtain more evidence for the singlet diradical nature of
the tetracation dimer, SQUID measurements of
8·4CHB11H5Cl6 were carried out. For singlet diradical species,
the excited triplet state is generally accessible, and this
transition process is observed as an increase in the population
of the magnetized state upon increase of temperature. Thus,
observations of such transitions in SQUID measurements can
be considered as indications of the presence of the open-shell
singlet ground state. However, in the case of 8, the excited
triplet species could not be observed (see the Supporting
Information). This could be due in part to the population of
the open-shell singlet diradical state, which was only about one-
half, and to a larger extent to the presence of a large singlet−
triplet gap as estimated by AP calculations of 8, in addition to
the inevitable intrinsic error associated with the measurement.
2.4. Redox Reactions of Dimer Species. To obtain a
deeper understanding of the electronic nature of the dimer
product, 12·2OTf, was treated with an excess amount of
AgOTf, and it was found that the tetracation dimer was
regenerated in the form of 8·4OTf in 78% isolated yield. It is
especially noteworthy that 8·4OTf obtained by this method was
free of 13 as an impurity, in contrast to the product of the
oxidation reaction of 3b. A CD2Cl2 solution of 8·4OTf in the
presence of excess AgOTf, which was used to guarantee
suppression of the formation of 13, still showed a partially
broadened 1H NMR spectrum at room temperature, suggesting
the existence of unpaired electrons (Figure 6). When the excess
AgOTf was removed, radical impurity 13 gradually formed.
1
Figure 6. VT H NMR of dimer 8·4OTf in the presence of excess
AgOTf in CD2Cl2. The aromatic region is given on the left, and the
signals of the methyl groups observed at 2.69 (6H) and 3.54 (12H)
ppm are shown on the right.
236
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
Thus, we could not use pure 8 for TPA experiments in section
2.8. The 1H NMR spectra of singlet diradical species are known
to become broad due to the accessibility of the excited triplet
state at higher temperatures. Because there was a possibility
that this phenomenon was operative for our compound, we
carried out variable temperature (VT) NMR measurements. By
measurements. Although tetracation species 8 was not included
in crude product at all, these results suggest that the
dimerization took place through demethylated intermediate
14·SbCl6. The formation of conjugated product 8 probably
takes place in a manner similar to the formation of biaryls by
oxidative coupling in general, which is facilitated by the
presence of electron-donating oxygen substituents.31 It should
be noted that the initial target, triplet carbene 4b, does not
seem to be involved in this reaction.
1
cooling the temperature to −50 °C, the H NMR spectrum in
CD2Cl2 became sharper. At present, we cannot leave out other
factors that might lead to this NMR behavior, but it is likely
that the reason stems from the singlet diradical origin of the
species. This is consistent with the singlet-to-triplet ratio (∼3 ×
106) of the excited triplet at room temperature estimated from
the calculated value of ΔEAPOS‑T = −8.9 kcal/mol mentioned in
the previous section.
2.6. One-Photon Absorption (OPA) Spectra of Dimeric
Species. Figure 7 shows the OPA spectrum of 8·4OTf in
2.5. Plausible Reaction Mechanism. Next, stepwise
oxidation of allene 3b was carried out to reveal the reaction
mechanism (Scheme 5). The oxidation of allene 3b with 1
Scheme 5. Plausible Reaction Mechanism
Figure 7. One-photon absorption spectra of dimers 8·4OTf, 12·2OTf,
and 13·3CHB11H5Cl6 in CH2Cl2.
CH2Cl2. This spectrum of 8 also shows absorption by trication
monoradical 13 (Scheme 4) because it was difficult to
completely remove 13 from the product of the oxidation of
allenic precursor 3. In contrast, pure 13 was obtained during
the recrystallization of 8·4CHB11H5Cl6, and the OPA spectrum
is also shown in Figure 7. The OPA spectrum of
13·3CHB11H5Cl6 has broad and intense peaks in the near-
infrared region (1449 and 1804 nm). These peaks originate
from transitions involving a singly occupied molecular orbital
(SOMO). Such near-infrared absorptions of radical ions were
previously observed in some aromatic compounds and have
been assigned to the SOMO → LUMO and HOMO → SOMO
excitations.32
Besides the near-infrared absorption peaks of the residual 13,
the intense peak at 1000 nm and the shoulder at 900 nm in the
spectrum for 8 in Figure 7 are attributed to inherent absorption
of tetracation diradical 8. Time-dependent DFT calculations at
the UB3LYP/6-31G(d) level based on the geometry from X-
ray analysis support that dimer 8 has two electron transitions in
the near-infrared region (982.1 nm, oscillator strength f =
0.1001, HOMO−7 → LUMO and 1221 nm, f = 0.0964,
HOMO → LUMO). Thus, the shoulder and the peak are
reasonably assigned to HOMO−7 → LUMO and HOMO →
LUMO transitions, respectively. The HOMO and LUMO are
mainly localized in the two central acridine units that form the
diradicaloid backbone structure (Figure 8), like in Chichiba-
bin’s hydrocarbon (Figure 2). Moreover, the small HOMO−
LUMO gaps are a typical feature for high singlet diradical
character in the ground state. Thus, these calculation results
also support the conclusion that the ground state of 8 has
singlet diradical character.
equiv of (4-BrC6H4)3N•+SbCl6 gave demethylated monomer
−
1
14·SbCl6 as confirmed by ESI−MS and H NMR. The parent
ion peak of 14·SbCl6 was observed at m/z 739.3518 (calcd for
+
C50H47O4N2 : 739.3530) using HR-ESI−MS. The identity of
the counteranion was confirmed with the negative mode. The
1H NMR spectrum of 14·SbCl6 in CDCl3 showed two Me
groups at δ 2.74 (3H) and 3.64 (6H), and two t-Bu groups at δ
1.45 (9H) and 1.51 (9H), indicating the unsymmetrical nature
of the structure of 14·SbCl6. These results suggest that
14·SbCl6 was formed by one-electron oxidation of 3b. This
would imply the elimination of a methyl radical from the initial
intermediate formed upon electron removal from 3b. However,
because we do not have definite proof of the quantitative
formation of 14·SbCl6, we cannot dismiss the possibility of a
−
second oxidation by counteranion SbCl6 to give an oxonium
intermediate that could easily extrude the methyl group upon
nucleophilic attack by a free chloride ion or a chlorine atom of
−
SbCl6 , which is known as a source of nucleophilic chloride
anions.28 This corresponds to the mechanism previously
assumed for the dimethoxythioxanthene system.29 SbCl6 is
−
known to oxidize ferrocene, I−1, and aromatic amines;
therefore, this assumption is not without reason.30 This type
of oxidation reaction would give rise to counteranions such as
30
−
SbCl4 and Sb2Cl82−, as observed here for 8·4SbCl4.
For the second oxidation step, 14·SbCl6 was treated with 1
−
equiv of (4-BrC6H)3N•+SbCl6 to afford trication monoradical
To ascertain the absorption spectrum of 8 more definitely,
we also measured the reflection spectrum on the (111) face for
species 13, whose identity was verified by UV−vis−NIR
237
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
Figure 8. Shapes of the α molecular orbitals of 8 calculated at the
UB3LYP/6-31G(d) level. The shapes are drawn at the 0.02 au
threshold level.
Figure 10. Spin density distribution of 8′ (a), and γ density
distributions (b) of 8′ (open-shell singlet) and 12′ (closed-shell
species). These densities are drawn at threshold values of 0.002 au for
(a) and 500 au for (b). Yellow and blue meshes in the γ density plots
represent the positive and negative regions, respectively.
a single crystal of pure 8·4SbCl4 with the polarization of light
parallel to the c-axis. Kramers−Kronig transformation of the
reflection spectrum gave the absorption spectrum of 8·4SbCl4
shown in Figure 9. The intense peaks at 952 and 1052 nm of
Static γ values were calculated by the finite-field method
using the total energies under the static electric field obtained at
the U(R)BHandHLYP/6-31G(d) level of approximation (for
details, see the Supporting Information). The results of the
calculations are summarized in Table 2. The z-axis component
Table 2. Diradical Character y and z-Axis Component of
Static γ (γzzzz) of Model Dimeric Species 8′ and 12′
Calculated at the U(R)BHandHLYP/6-31G(d) Level
compound
y
γzzzz/103 au
8′ (dication)
12′ (neutral)
0.605
0.0
3100
217
of static γ of diradical species 8′ (3100 × 103 au) was found to
be more than 10 times as large as that of closed-shell
counterpart 12′ (217 × 103 au). Such significant enhancement
of static γ in the diradical species supports our theoretical
prediction.9c Further, the amplitude of the γ value of 8′ is on
the same order as those of bis(phenalenyl) compounds that
have already been found to exhibit significant TPA cross
sections.11a,14
Figure 9. The absorption spectrum of a single crystal of 8·4SbCl4
obtained by Kramers−Kronig transformation of the reflection
spectrum.
the obtained spectrum show good agreement with the near-
infrared peaks in CH2Cl2, demonstrating that absorption peaks
at 900 and 1000 nm in the spectrum of the solution sample
definitely originate from 8.
To further clarify the spatial contributions of electrons to the
enhancement of γ in diradical species 8′, we applied γ density
analysis.22 The γ density represents the field-induced response
of charge density proportional to the cubic of the external
electric field amplitude. Positive and negative γ densities
represent the increase and decrease of charge densities related
to the third-order polarization, respectively. The sign of the
contribution to γ is positive when the direction from positive to
negative γ density coincides with the positive direction on the
coordinate axis, and the sign becomes negative in the opposite
case. The magnitude of the contribution associated with this
pair of γ densities is proportional to the distance between them.
The γ density distributions of these models are shown in Figure
10b. For open-shell species 8′, large positive (yellow) and
negative (blue) γ densities, dominated by π electrons, are found
to be well separated at the left- and right-hand sides of the
system, leading to significant positive contributions to γ. The
region with large positive and negative γ densities is found to be
well correlated to that with primary spin densities, the feature
of which indicates strong impact of diradical character on γ
enhancement. In sharp contrast, the γ density amplitudes of
closed-shell species 12′ are found to be negligible in the whole
region of the system. These results substantiate the significant
2.7. Diradical Character Dependence of Static Second
Hyperpolarizabilities (γ) of Model Species. Before
proceeding with TPA measurements, we theoretically examined
the impact of diradical character on the static γ for the dimeric
species. To focus on the diradical character effects in tetracation
dimer 8, we considered a simplified model (8′) composed of
two central acridines that form a diradicaloid backbone
structure. In the model, Acr+, OMe, and Ar of 8 are replaced
by hydrogen atoms; thus, the system is a dication. From
geometry optimization at the UB3LYP/6-31G(d) level, we
obtained a planar structure. Both diradical character (y = 0.605)
and spin density distribution of 8′ calculated at the
UBHandHLYP/6-31G(d) level are found to be very close to
those of original tetracation dimer 8 (see Figures 5 and 10a).
We also considered a two-electron reduced model (12′) of 8′,
which corresponds to closed-shell dication dimer 12. The
optimized structure of 12′ was found to be slightly nonplanar,
neutral, and closed shell (y = 0). As a result, these models are
expected to be useful for our purposes.
238
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
Figure 11. Open-aperture Z-scan traces (normalized transmittance TN versus the sample position z) of impure 8·4SbCl4 (a), 13·3CHB11H5Cl6 (b),
and 12·2SbCl4 (c) at wavelengths of 1200 nm (left), 1260 nm (middle), and 1330 nm (right). Solid curves are theoretical fits (see the Supporting
Information) to the recorded data (symbols). All samples are in CH2Cl2.
Table 3. Diradical Character y and Two-Photon Absorption Cross Section σ(2) of Dimeric Species 8, 12, and 13 at Different
Wavelengths
b
σ
(2)/GM
a
compound
y
1200 nm
1230 nm
1260 nm
1300 nm
1330 nm
c
8
0.685
3560 410
<9
2350 300
<21
1590 250
<11
455 66
396 52
12
0.000
0
0
d
13
−
<140
<110
<73
0
0
a
b
Calculated at the UBHandHLYP/6-31G(d) level. The structures were optimized by UB3LYP/6-31G(d). 1 GM = 10−50 cm4 s molecule−1
c
d
photon−1. Contribution of resulting 13 was subtracted for (2) (see the Supporting Information). Indefinable for monoradical species.
difference in γ amplitudes between 8′ and 12′ (see Table 2).
From these results, open-shell tetracation dimer 8 is expected
to exhibit significantly large TPA cross sections as compared to
closed-shell dicationic counterpart 12.
observed; the Z-scan trace showed an “M”-like shape, that is, a
concave curve overlapping a broad convex curve, suggesting
that TPA overlaps SA.
As pure 13 (the middle row in Figure 11) gave almost no
signal at the wavelengths of 1200 and 1260 nm, it can be
concluded that the strong TPA observed for impure 8 at these
wavelengths originates from 8 rather than 13. Meanwhile, a
strong SA signal was also observed for pure 13 at 1330 nm (the
middle-right), which can also contribute to the SA signal
observed for 8 at the same wavelength. Here, we ignore the
difference in counteranions, which are SbCl4− for the impure 8;
thus, the residual signals were from 13·3SbCl4 for 8, and
CHB11H5Cl6− for the pure 13. This simplification is reasonably
justified because the OPA spectra of 13 with different
counteranions were almost the same, except for minor spectral
shifts (the near-infrared peaks were 1461 and 1823 nm for the
2.8. Two-Photon Absorption (TPA) Measurements of
Dimeric Species. Following the results shown in previous
sections that tetracation dimer 8 has relatively high singlet
diradical character, TPA measurements of 8·4SbCl4, closed-
shell dication 12, and monoradical trication 13 were performed
by the open-aperture Z-scan method to clarify the relationship
between diradical character and TPA properties. The samples
for this study were prepared by dissolving the compounds in
CH2Cl2. Although we had pure samples of 12·2SbCl4 and
13·3CHB11H5Cl6, the available sample of 8·4SbCl4 contained
residual 13 due to the difficulty of the separation (as mentioned
in previous sections). Therefore, we conducted measurements
for the impure sample of 8 and separated the contribution of 8
from the impure solution by comparing the results to those
obtained for the solution of pure 13.
For the measurements, the excitation wavelength was
scanned from 1200 to 1300 nm, where the OPAs of the
samples are relatively small (Figure 7). Panels in the top row in
Figure 11 show typical Z-scan traces of the solution of impure
8. At 1260 nm, a clear concave curve centered at the focal point
(z = 61 mm) was observed (the top-middle panel), indicating
that strong TPA occurs at this wavelength. The opposite shape,
that is, a convex transmittance curve centered at the focal point,
was also observed at the longest wavelength (1330 nm, top-
right). This convex curve means that the OPA of the sample at
this wavelength decreased with increasing optical intensity by
the saturable absorption (SA) process. At the shortest
wavelength (1200 nm, top-left), the intermediate case was
−
impure solution with the counteranion of SbCl4 , and 1444 and
1826 nm for pure 13·3CHB11H5Cl6), suggesting that the choice
of counteranion does not have a significant influence on the
electronic states of 13. In comparison to 8 and 13, closed-shell
species 12 gave rather simple Z-scan traces (the bottom row in
Figure 11). The magnitudes of the SA signals in the recorded
traces always decreased as the wavelength increased and had no
indication of TPA.
Quantitative analysis of the recorded Z-scan traces was
performed by using the curve-fitting procedure with the
theoretical model assuming spatial and temporal Gaussian
profiles for the incident laser pulses (see the Supporting
Information). The observed traces were well reproduced with
the fitting curves (solid curves in Figure 11). In the procedure
to calculate the TPA cross section σ(2), we estimated the
concentration of 13 in the impure sample from the one-photon
239
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
absorption spectrum and then subtracted its contribution from
the total α(2) of the impure solution; however, the contribution
of 13 was found to be negligible as compared to that of 8.33
The TPA cross sections obtained for each dimeric species are
listed in Table 3. Clearly, diradical dimer 8 has much larger σ(2)
than the other species at each wavelength. The maximum value
was σ(2) = 3560 410 GM at 1200 nm, which is a relatively
large value for an organic chromophore. The resonance
enhancement by the HOMO → LUMO transition of OPA
may be involved in the large σ(2). In contrast, closed-shell
singlet 12 did not show any notable TPA; the σ(2) values are at
least 2 orders of magnitude smaller than those of diradical
dimer 8. This is the first direct comparison, to the best of our
knowledge, between a singlet diradicaloid (8, y = 0.682) and its
closed-shell counterpart (12, y = 0.000), which have the same
backbone structure and atom composition, although compar-
isons between molecules having similar π-conjugation (but
different backbones or atom compositions) have been
previously demonstrated.13,14,21 These results support our
theoretical prediction that molecules having intermediate
diradical character have large third-order nonlinearity including
TPA.9,11
procedures and analysis of the TPA measurements for an
impure sample of 8 mentioned in section 2.7. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
k.kamada@aist.go.jp; yyama@sci.hiroshima-u.ac.jp
Present Address
⊥Research Institute of Instrumentation Frontier, National
Institute of Advanced Industrial Science and Technology
(AIST), Tsukuba, Ibaraki 305-8565, Japan.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by two Grants-in-Aid for Scientific
Research on Priority Areas (nos. 14340199 and 17350021)
from the Ministry of Education, Culture, Sports, Science and
Technology, Japan. This work was also supported by a Grant-
in-Aid for Scientific Research (no. 21350011) from the Japan
Society for the Promotion of Science (JSPS). We are thankful
to the Natural Science Center for Basic Research and
Development (N-BARD), Hiroshima University, for combus-
tion analysis measurements on some of the new compounds.
We acknowledge the Research Center for Computational
Science, Okazaki, Japan, for machine time on their computer
system. We also thank Assistant Prof. Eigo Miyazaki and Prof.
Kazuo Takimiya of the Department of Applied Chemistry,
Graduate School of Engineering, Hiroshima University, for
UV−vis−NIR measurements and Prof. Manabu Abe of the
Department of Chemistry, Graduate School of Science,
Hiroshima University, for ESR measurements.
3. CONCLUSIONS
In summary, during our attempts to synthesize stable triplet
carbenes, a dimeric product 8 having singlet diradical nature in
the ground state was serendipitously obtained upon the
oxidation of an allenic precursor (3) bearing 1,8-dimethox-
−
yacridine moieties using 2 equiv of (4-BrC6H4)3N•+SbCl6 .
The diradical nature of 8 in the ground state was supported by
X-ray structural analysis and DFT calculations. Furthermore,
the corresponding dicationic dimer 12 was furnished upon
treatment of tetracation dimer 8 with Zn. Trication 13 was also
obtained as a byproduct of the oxidation of allenic precursor 3.
These three dimeric species, which have the same structural
backbone and atom composition, but different spin states,
enabled us to directly compare the dependence of TPA
properties on different diradical characters to test the
theoretical prediction that a molecule with intermediate
diradical character (0 < y < 1) has larger optical nonlinearity
than the corresponding closed-shell electronic structure (y =
0).9,11 Two-photon absorption measurements of the three
species showed significant differences depending on singlet
diradical character y. This is the first straightforward
demonstration supporting the theoretical prediction because
the same molecular backbone and atom composition are
retained across the compared species. Together with the
previous studies,13,14,21 the present results clearly show that
singlet diradical character with intermediate magnitude can be a
new molecular parameter for excellent TPA materials in the
near-infrared region. Moreover, it can be pointed out that the
present results suggest the possibility of switching the TPA
magnitude by using redox reactions, as reported for condensed-
ring fluorene derivatives.34 The use of redox reactions to switch
TPA properties may open new possibilities such as highly
sensitive detection of intermediate species of such reactions.
REFERENCES
■
(1) Salem, L.; Rowland, C. Angew. Chem., Int. Ed. Engl. 1972, 11, 92−
111.
(2) (a) Lambert, C. Angew. Chem., Int. Ed. 2011, 50, 1756−1758.
(b) Sun, Z.; Wu, J. J. Mater. Chem. 2012, 22, 4151−4160.
(3) Bendikov, M.; Duong, H. M.; Starkey, K.; Houk, K. N.; Carter, E.
A.; Wudl, F. J. Am. Chem. Soc. 2004, 126, 7416−7417.
(4) Bettinger, H. F. Pure Appl. Chem. 2010, 82, 905−915.
(5) (a) Tonshoff, C.; Bettinger, H. F. Angew. Chem., Int. Ed. 2010, 49,
4125−4128. (b) Purushothaman, B.; Bruzek, M.; Parkin, S. R.; Miller,
A.-F.; Anthony, J. E. Angew. Chem., Int. Ed. 2011, 50, 7013−7017.
(6) (a) Kubo, T.; Shimizu, A.; Sakamoto, M.; Uruichi, M.; Yakushi,
K.; Nakano, M.; Shimoi, D.; Sato, K.; Takui, T.; Morita, Y.; Nakatsuji,
K. Angew. Chem., Int. Ed. 2005, 44, 6564−6568. (b) Shimizu, A.;
Uruichi, M.; Yakushi, K.; Matsuzaki, H.; Okamoto, H.; Nakano, M.;
Hirao, Y.; Matsumoto, K.; Kurata, H.; Kubo, T. Angew. Chem., Int. Ed.
2009, 48, 5482−5486.
(7) Huang, J.; Kertesz, M. J. Am. Chem. Soc. 2007, 129, 1634−1643.
(8) Nakano, M.; Nagao, H.; Yamaguchi, K. Phys. Rev. A 1997, 55,
1503−1513.
(9) (a) Nakano, M.; Kishi, R.; Nitta, T.; Kubo, T.; Nakasuji, K.;
Kamada, K.; Ohta, K.; Champagne, B.; Botek, E.; Yamaguchi, K. J.
Phys. Chem. A 2005, 109, 885−891. (b) Nakano, M.; Kishi, R.; Ohta,
S.; Takebe, A.; Takahashi, H.; Furukawa, S.; Kubo, T.; Morita, Y.;
Nakasuji, K.; Yamaguchi, K.; Kamada, K.; Ohta, K.; Champagne, B.;
Botek, E. J. Chem. Phys. 2006, 125, 074113. (c) Nakano, M.; Kishi, R.;
Ohta, S.; Takahashi, H.; Kubo, T.; Kamada, K.; Ohta, K.; Botek, E.;
Champagne, B. Phys. Rev. Lett. 2007, 99, 033001. (d) Nakano, M.;
Yoneda, K.; Kishi, R.; Takahashi, T.; Kubo, T.; Kamada, K.; Ohta, K.;
Botek, E.; Champagne, B. J. Chem. Phys. 2009, 131, 114316.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic, experimental, and calculation procedures, character-
ization data including crystallographic measurements, geometry
optimizations of tetracation dimer 8, SQUID data mentioned in
section 2.3, and additional details related to the curve fitting
240
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241

Journal of the American Chemical Society
Article
(10) (a) He, G. S.; Tan, L.-S.; Zheng, Q.; Prasad, P. N. Chem. Rev.
2008, 108, 1245−330. (b) Pawlicki, M.; Collins, H. A.; Denning, R.
G.; Anderson, H. L. Angew. Chem., Int. Ed. 2009, 48, 3244−3266.
(11) (a) Nakano, M.; Kubo, T.; Kamada, K.; Ohta, K.; Kishi, R.;
Ohta, S.; Nakagawa, N.; Takahashi, H.; Furukawa, S.; Morita, Y.;
Nakasuji, K. Chem. Phys. Lett. 2006, 418, 142−147. (b) Nakano, M.;
Nagai, H.; Fukui, H.; Yoneda, K.; Kishi, R.; Takahashi, H.; Shimizu, A.;
Kubo, T.; Kamada, K.; Ohta, K.; Champagne, B.; Botek, E. Chem. Phys.
Lett. 2008, 467, 120−125. (c) Fukui, H.; Nakano, M.; Shigeta, Y.;
Champagne, B. J. Phys. Chem. Lett. 2011, 2, 2063−2066.
(12) (a) Yamaguchi, K. Instability in Chemical Bonds. In Self-
Consistent Field. Theory and Applications; Carbo, R., Klobukowski, M.,
Eds.; Elsevier: Amsterdam, 1990; pp 727−823. (b) Nakano, M.; Fukui,
H.; Minami, T.; Yoneda, K.; Shigeta, Y.; Kishi, R.; Champagne, B.;
Botek, E.; Kubo, T.; Ohta, K.; Kamada, K. Theor. Chem. Acc. 2011,
130, 711−724; 2011, 130, 725 (erratum).
(28) Stiddard, M. H. B.; Townsend, R. E. J. Chem. Soc. A 1969,
2355−2357.
(29) (a) Yamaguchi, T.; Yamamoto, Y.; Kinoshita, D.; Akiba, K.;
Zhang, Y.; Reed, C. A.; Hashizume, D.; Iwasaki, F. J. Am. Chem. Soc.
2008, 130, 6894−6895. (b) Yano, T.; Yamaguchi, T.; Yamamoto, Y.
Chem. Lett. 2009, 38, 794−795.
(30) Cowell, G. W.; Ledwith, A.; White, A. C.; Woods, H. J. J. Chem.
Soc. B 1970, 227−231.
(31) Kovacic, P.; Jones, M. B. Chem. Rev. 1987, 87, 357−379.
(32) (a) Ohashi, K.; Kubo, T.; Masui, T.; Yamamoto, K.; Nakasuji,
K.; Takui, T.; Kai, Y.; Murata, I. J. Am. Chem. Soc. 1998, 120, 2018−
2027. (b) Angliker, H.; Gerson, F.; Lopez, J.; Wirz, J. Chem. Phys. Lett.
1981, 81, 242−246. (c) Angliker, H.; Rommel, E.; Wirz, J. Chem. Phys.
Lett. 1982, 87, 208−212.
(33) See the Supporting Information.
(34) Cho, N.; Zhou, G.; Kamada, K.; Kim, R. H.; Ohta, K.; Jin, S.-H.;
Mullen, K.; Lee, K.-S. J. Mater. Chem. 2012, 22, 185−191.
̈
(13) Kamada, K.; Ohta, K.; Shimizu, A.; Kubo, T.; Kishi, R.;
Takahashi, H.; Botek, E.; Champagne, B.; Nakano, M. J. Phys. Chem.
Lett. 2010, 1, 937−940.
(14) Kamada, K.; Ohta, K.; Kubo, T.; Shimizu, A.; Morita, Y.;
Nakasuji, K.; Kishi, R.; Ohta, S.; Furukawa, S.; Takahashi, H.; Nakano,
M. Angew. Chem., Int. Ed. 2007, 46, 3544−3546.
(15) (a) Hirai, K.; Tomioka, H. J. Am. Chem. Soc. 1999, 121, 10213−
10214. (b) Itoh, T.; Nakata, Y.; Hirai, K.; Tomioka, H. J. Am. Chem.
Soc. 2006, 128, 957−967.
(16) Iwamoto, E.; Hirai, K.; Tomioka, H. J. Am. Chem. Soc. 2003,
125, 14664−14665.
(17) Kawano, M.; Hirai, K.; Tomioka, H.; Ohashi, Y. J. Am. Chem.
Soc. 2007, 129, 2383−2391.
(18) For the formation of cation radical dimers by oxidation, see:
(a) Oka, H. Org. Lett. 2010, 12, 448−451. (b) Terada, E.; Okamoto,
T.; Kozaki, M.; Masaki, M. E.; Shiomi, D.; Sato, K.; Takui, T.; Okada,
K. J. Org. Chem. 2005, 70, 10073−10081. (c) Zheng, S.; Barlow, S.;
Risko, C.; Kinnibrugh, T. L.; Khrustalev, V. N.; Jones, S. C.; Antipin,
́
M. Y.; Tucker, N. M.; Timofeeva, T. V.; Coropceanu, V.; Bredas, J.-L.;
Marder, S. R. J. Am. Chem. Soc. 2006, 128, 1812−1817. (d) Nishinaga,
T.; Tateno, M.; Fujii, M.; Fujita, W.; Takase, M.; Iyoda, M. Org. Lett.
2010, 12, 5374−5377. (e) Ito, A.; Urabe, M.; Tanaka, K. Angew.
Chem., Int. Ed. 2003, 42, 921−924.
(19) (a) Errede, L. A.; Landrum, B. F. J. Am. Chem. Soc. 1957, 79,
4952−4955. (b) Montgomery, L. K.; Huffman, J. C.; Jurczak, E. A.;
Gredze, M. P. J. Am. Chem. Soc. 1986, 108, 6004−6011. (c) Hiroto, S.;
Furukawa, K.; Shinokubo, H.; Osuka, A. J. Am. Chem. Soc. 2006, 128,
12380−12381. (d) Ishida, M.; Karasawa, S.; Uno, H.; Tani, F.; Naruta,
Y. Angew. Chem., Int. Ed. 2010, 49, 5906−5909. (e) Takahashi, T.;
Matsuoka, K.; Takimiya, K.; Otsubo, T.; Aso, Y. J. Am. Chem. Soc.
2005, 127, 8928−8929. (f) Kikuchi, A.; Iwahori, F.; Abe, J. J. Am.
Chem. Soc. 2004, 126, 6526−6527.
(20) Montgomery, L. K.; Huffman, J. C.; Jurczak, E. A.; Gredze, M. P.
J. Am. Chem. Soc. 1986, 108, 6004−6011.
(21) Ishida, M.; Shin, J.-Y.; Lim, J. M.; Lee, B. S.; Yoon, M.-C.; Koide,
T.; Sessler, J. L.; Osuka, A.; Kim, D. J. Am. Chem. Soc. 2011, 133,
15533−15544.
(22) Nakano, M.; Shigemoto, I.; Yamada, S.; Yamaguchi, K. J. Chem.
Phys. 1995, 103, 4175−4191.
(23) Yano, T.; Yamaguchi, T.; Yamamoto, Y. Chem. Lett. 2009, 38,
794−795.
(24) Ali, M. H.; Buchwald, S. J. Org. Chem. 2001, 66, 2560−2565.
(25) (a) Mass, G.; Singer, B. Z. Naturforsch. 1984, 39b, 1399.
(b) Mass, G.; Singer, B. Z. Naturforsch. 1985, 40b, 90.
−
(26) For the structure of dimeric SbCl4 , see: (a) Gerasimenko, A.
V.; Polishchuk, A. V.; Volkova, L. M.; Karaseva, E. T.; Karasev, V. E.
Russ. J. Coord. Chem. 2008, 34, 8−13. (b) Bott, S. G.; Brammer, L.;
Connelly, N. G.; Green, M.; Orpen, A. G.; Paxton, J. F.; Schaverien, C.
J.; Bristow, S.; Norman, N. C. J. Chem. Soc., Dalton Trans. 1990, 1957−
1969.
(27) (a) Yamaguchi, K.; Takahara, Y.; Fueno, T.; Houk, K. N. Theor.
Chim. Acta 1988, 73, 337−364. (b) Yamanaka, S.; Okumura, M.;
Nakano, M.; Yamaguchi, K. J. Mol. Struct. 1994, 310, 205−218.
241
dx.doi.org/10.1021/ja308396a | J. Am. Chem. Soc. 2013, 135, 232−241