A tris(carbene) constructed with stable triplet carbene units

Article abstract of DOI:10.1246/bcsj.78.2037

Bis[10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]diazomethane, which can generate a fairly stable triplet carbene that can survive for even a week in solution at room temperature, was shown to be stable enough to undergo a Sonogashira coupling reaction. Th

Full text of DOI:10.1246/bcsj.78.2037

ꢀ 2005 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 78, 2037–2050 (2005) 2037
A Tris(carbene) Constructed with Stable Triplet Carbene Units
ꢀ;1
ꢀ
Yoko Tsuchiya, Masayoshi Matsuno, Tetsuji Itoh, Katsuyuki Hirai, and Hideo Tomioka
Department of Chemistry for Materials, Faculty of Engineering, Mie University, Tsu, Mie 514-8507
1Instrumental Analysis Facilities, Life Science Research Center, Mie University, Tsu, Mie 514-8507
Received May 12, 2005; E-mail: tomioka@chem.mie-u.ac.jp
Bis[10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]diazomethane, which can generate a fairly stable triplet car-
bene that can survive for even a week in solution at room temperature, was shown to be stable enough to undergo a
Sonogashira coupling reaction. Thus, three units of the diazo compounds were connected through an ethynyl group
on 1,3,5-positions of benzene. The species generated by irradiation of the resulting tris(diazo) compound were charac-
terized by variable temperature ESR and SQUID measurements. The results indicated the generation of the correspond-
ing tris(carbene), that exhibits only weak exchange between carbene units. This fits the fact that dianthrylcarbenes
are connected to the central phenyl ring across nodal (inactive) sites. The anomalously high thermal stability and chemi-
cal inertness of the species were shown by variable temperature and time-resolved UV–vis spectroscopic studies. The
tris(carbenes) are shown to survive hours in solution at room temperature.
A great deal of work has been directed toward high-spin or-
ganic species, in which spins of unpaired electrons in 2p-orbitals
Results and Discussion
of light atoms, such as carbon, nitrogen, and oxygen, are mainly
used. Spin sources used for such studies are mostly thermo-
dynamically stable radicals.1,2 This is mainly because these rad-
icals are usually easy to handle. One of the potential problems
with these polyradicals is the weakness of the exchange cou-
pling between the neighboring radical centers. Thus, an increas-
ed probability of thermal population in the lower-spin states can-
not be avoided. Triplet carbenes are regarded as one of the most
effective spin sources, since the magnitude of the exchange cou-
pling between the neighboring centers is large.3,4 Moreover, the
photolytic production of polycarbenes is possible even in solid
solution at cryogenic temperatures if poly(diazo) precursors
are available. Actually, poly(diazo) compounds have been pre-
pared, and high-spin polycarbenes have been generated.4
However, those systems have two disadvantages that hinder
their further extension to usable magnetic materials. First, a
triplet carbene unit is highly unstable and lacks stability for
practical application under ambient conditions. Secondly,
diazo groups are also generally labile,5 and, hence, the diazo
compound cannot be used as a building block to prepare a
more complicated poly(diazo) compound.
Preparation of a Tris(diazo) Compound. The diazo unit
that we used in this study was bis[10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl]diazomethane (1-N2), which can generate a
fairly stable triplet carbene that survives even a week in solu-
tion at room temperature (Scheme 1).11,12
In order to attach a handle connecting this unit to a linker,
we first prepared the dianthryl diazo compound (2-N2), which
has a bromine group at the 3 position of the anthryl ring,
according to the procedures summarized in Scheme 2. Then,
this diazo compound was treated with trimethylsilylacetylene
in the presence of [PdCl2(Ph3P)2] and CuI at 40 ꢁC for 1
day. This treatment gave 3-N2(TMS). Deprotection13 of the
trimethylsilyl group of 3-N2(TMS) with NaOH proceeded
smoothly to give an ethynyl derivative 3-N2. A desired tris-
(diazo) compound (4-3N2) was then prepared by coupling
three equivalents of 3-N2 with 1,3,5-triiodobenzene14 under
mild Sonogashira coupling reaction conditions (Scheme 3).15
The characterization of the tris(diazo) compound (4-3N2)
1
was mainly based on H and 13C NMR and IR spectra. The
1H NMR spectrum of 4-3N2 showed two singlets (ꢀ 1.46 and
1.79 ppm) corresponding to the t-butyl and methyl groups,
respectively, and an aromatic singlet (ꢀ 7.54 ppm) in the ratio
of 18:12:1, in agreement with the calculated value. A strong
and sharp band at 2048 cmꢂ1 was observed in the IR spectrum.
However, molecular weight determination by using MS could
In order to overcome these difficulties, we have made great
efforts to stabilize a triplet carbene and succeeded in preparing
fairly stable ones that survived for days in solution at room
temperature.6 We also found that a diphenyldiazomethane pre-
pared to generate a persistent triplet carbene is also persistent
for the diphenyldiazo compound and, hence, can be further
modified, with the diazo group intact, into a more complicated
diazo compound.7–10
N2
hν
Thus, it is now potentially possible to prepare a stable poly-
carbene by connecting a stable carbene precursor diazo unit in
a proper manner. We wish to report here that a tris(diazo) com-
pound consisting of a dianthryl diazo compound generated a
fairly stable tris(carbene).
1-N2
1
Scheme 1.
Published on the web October 31, 2005; DOI 10.1246/bcsj.78.2037

2038 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
Br
MgBr
1.
1. n-BuLi
2.
OHC
Br Br2
Br
Br
2. H+/H2O
O
H
COOEt
Br
N2
OH
N
1.N2O4
1.HCl
2. t-BuOK
2.H2NCO2Et/
AgBF4
Br
Br
2-N2
Scheme 2.
N2
h
ν
Br
H
Br
2-N2
2
N2
N2
TMS
NaOH aq.
[PdCl2(Ph3P)2]
CuI
t-BuOH
Et3N
3-N2
H
TMS
3-N2(TMS)
N2
*
*
*
*
*
*
I
*
*
h
ν
I
I
[PdCl2(Ph3P)2]
CuI
*
*
*
*
*
*
*
Et3N
*
*
*
N2
*
*
*
*
*
N2
*
4-3N2
4
(n* - n) = (24 - 21) = 3
Scheme 3.
not be used in the present system since the diazo functional
groups could not survive ionization processes. One idea to
solve this problem is to convert all the diazo groups in 4-3N2
to a stable functional group and to determine the molecular
weight of the resulting trimeric product. This sounded reason-
able since there are several ways to quantitatively convert
diazo groups to other functional groups; such methods include
the treatment with hydrochloric acid to give the corresponding
chloride and photolysis in methanol to give the methyl ether as
a result of the O–H insertion of the photogenerated carbene
into the solvent.5,16 However, all attempts to convert 4-3N2
cleanly to a tris compound with a stable functional group were
unsuccessful. This is obviously due to the highly congested
nature of its reaction center.
Therefore, we decided to prepare an ‘‘authentic’’ sample.
Tris(ketone) (4-3O) was prepared following essentially the
same reaction procedure as outlined in Scheme 3 using 3-
bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl 10-(4-t-
butyl-2,6-dimethylphenyl)-9-anthryl ketone (2-O) as a starting
compound (Scheme 4).
Matrix-assisted laser desorption ionization time-of-flight
MS (MALDI-TOF MS) of the tris(ketone) (4-3O) afforded

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2039
O
O
2. NaOH aq.
1. H
TMS
t-BuOH
[PdCl2(Ph3P)2]
CuI
Et3N
Br
2-O
3-O
H
O
I
I
I
[PdCl2(Ph3P)2]
CuI
Et3N
O
O
4-3O
Scheme 4.
an Mþ ion peak at 2251.1432 (1:1 CH3OH/CHCl3, positive
mode). The spectrum for the Mþ ion cluster from the high-
resolution spectrum exhibited isotopic distributions essentially
identical with those simulated for C171H150O3 (Fig. S1). The
fully characterized tris(ketone) 4-3O was then used as a
‘‘standard’’ in the analysis of the diazo counterpart by GPC.
The elution volume of 4-3N2 was actually found to be in com-
plete agreement with that of 4-3O (Fig. S2).
(a)
77 K
95 K
(b)
(c)
77 K
240 K
ESR Studies. Irradiation (ꢁ > 350 nm) of a mono(diazo)
compound (2-N2) in 2-methyltetrahydrofuran (2-MTHF) at 77
K gave rise to a paramagnetic species (Fig. 1a) readily charac-
terized from its ESR spectrum as a derivative of a triplet car-
bene (32).12,17 The ESR signals were persistent at this low tem-
perature but disappeared irreversibly when the matrix temper-
ature was raised to room temperature and recooled to 77 K.
The ESR signals were analyzed in terms of zero-field splitting
(ZFS) parameters, which gave jDj ¼ 0:103 cmꢂ1 and jEj ¼
0:0006 cmꢂ1 (E=D ¼ 0:0058). These values are very similar
to those (jDj ¼ 0:102 cmꢂ1 and jEj ¼ 0:0008 cmꢂ1, E=D ¼
77 K
260 K
(d)
(e)
77 K
280 K
77 K
300 K
(f)
77 K
3
0:0078) observed for 1 obtained by irradiation of 1-N2.12
In order to examine the stability of 32, we carried out a
variable-temperature ESR study. The thermal stability of the
triplet carbenes could be estimated by gradually warming the
matrix containing triplet carbenes to a desired temperature;
the mixture was kept at this temperature for 5 min and
recooled again to 77 K to measure the signal. This procedure
can compensate any weakening of signals due to the Curie
law.18 When the 2-MTHF glass containing 32 was warmed
to around 95 K, a new set of triplet peaks (jDj ¼ 0:088
cmꢂ1 and jEj ¼ 0:0003 cmꢂ1) appeared at the expense of
the original peaks (Fig. 1b). These changes were not reversi-
ble; when the sample was cooled to 77 K, no change took
place, except that the signal intensity increased according to
the Curie law. Changes of this kind have often been observed
for sterically congested triplet diarylcarbenes and are usually
interpreted in terms of the geometrical relaxation of the car-
benes as the viscosity decreases upon warming.19,20 Significant
decay of the new set of signals was observed only at 280 K,
0
200
400
600
800
Magnetic Field / mT
Fig. 1. (a) ESR spectrum obtained by irradiation of mono-
diazo compound 2-N2 in 2-methyltetrahydrofuran at 77 K.
(b–f) ESR spectra observed at 77 K in 2-methyltetra-
hydrofuran after warming the matrix containing 2 to (b)
95, (c) 240, (d) 260, (e) 280, and (f) 300 K.
3
and the signals were observable even at 300 K. However,
the signals disappeared completely after the sample was kept
at room temperature for 15 min (the measurements were made
at 77 K). This result is to be compared with that observed for
31, whose ESR signals were observed even after standing for
three hours at room temperature.12b A marked decrease in
the thermal stability was thus noted once a bromo substituent
was introduced at 3 position.

2040 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
(a)
77 K
90 K
77 K
0
200
400
600
800
Magnetic Field / mT
(b)
(c)
240 K
77 K
= int 1
= int 2
= int 3
= int 4
280 K
77 K
(d)
(e)
0
0.02
0.04
0.06
0.08 0.10
T-1 / K-1
300 K
77 K
Fig. 3. Plots of the intensities of the ESR signals at 262.0
and 147.9 mT due to triplet species and at 324.5 and
310.4 mT due to a higher-spin species vs reciprocal of
temperature (10–85 K).
0
200
400
600
800
intensities of all the signals vs the reciprocal of the temperature
(Fig. 3) were linear.
Magnetic Field / mT
These observations suggest that all the signals are attributa-
ble to the ground states. It is possible that triplet species orig-
inate from monocarbene (formed due to incomplete photode-
composition), while high-spin species are formed as a result
of complete decomposition. Another possibility is that the four
states (singlet, triplet, quintet, and septet) originate from one
species, in which they are degenerate. However, the latter pos-
sibility is excluded by the variable-temperature study of ESR
signals, the ZFS parameters, and the wavelength dependence
of the spectrum (vide infra).
In order to examine the stability of those high-spin species,
we conducted a variable-temperature ESR study. Slight but
distinct shifts of the signals were observed along with an appa-
rent increase in the intensity when the matrix was warmed to
90 K. The ZFS parameters obtained for the triplet species
(jDj ¼ 0:086 cmꢂ1 and jEj ¼ 0:0006 cmꢂ1) slightly decreas-
ed, suggesting again that the species undergo geometrical
relaxation upon warming.19,20 No significant decay of the sig-
nals was observed until the matrix temperature was raised to
280 K, and the signals were observable even at 300 K. Anom-
alously high thermal stability was thus noted.
SQUID Measurement. In order to obtain more informa-
tion on the ground state multiplicity, we carried out magnetic
measurements. The 2-MTHF solution of tris(diazo) com-
pounds (4-3N2) was placed inside the sample compartment
of a superconducting quantum interference device (SQUID)
magnet/susceptometer and irradiated at 5–10 K with light
(ꢁ ¼ 488 nm) from an argon ion laser through an optical fiber.
The development of magnetization (M/emu) at 5 K in a con-
stant field of 5 kOe with the irradiation time for the tris(diazo)
compound was measured in situ and is shown in Fig. 4a. As the
irradiation time increased, the M values gradually increased
and reached a plateau after several hours. After the M values
reached a plateau, the magnetization values after irradiation,
Ma, were measured at 2.0, 3.0, and 5.0 K in a field range
of 0–50 kOe. The magnetization values of the sample before
Fig. 2. (a) ESR spectrum obtained by irradiation of tris(di-
azo) compound 4-3N2 in 2-methyltetrahydrofuran at 77 K.
(b–e) ESR spectra observed at 77 K in 2-methyltetra-
hydrofuran after warming the matrix containing 4 to (b)
90, (c) 240, (d) 280, and (e) 300 K.
Similar irradiation (ꢁ > 350 nm) of the tris(diazo) com-
pound 4-3N2 in a 2-methyltetrahydrofuran (2-MTHF) solid
solution at 77 K produced ESR signals (Fig. 2). The signals
at 147.9, 262.0, and 386.2 mT were similar to those observed
for triplet monocarbene 32 obtained by the photolysis of
2-N2 and were analyzed in terms of the zero-field splitting
(ZFS) parameters (jDj ¼ 0:106 cmꢂ1 and jEj ¼ 0:0008 cmꢂ1).
However, those at 310.4 and 324.5 mT were obviously differ-
ent from those of the monocarbene triplet and were attributable
to a higher spin state.
In order to assign the signals, we carried out computer sim-
ulations.21 The spectra simulated for S ¼ 1 (g ¼ 2:003, jDj ¼
0:024 cmꢂ1, and jEj ¼ 0:0003 cmꢂ1), S ¼ 2 (g ¼ 2:003, jDj ¼
0:009 cmꢂ1, and jEj ¼ 0:000 cmꢂ1), and S ¼ 3 (g ¼ 2:003,
jDj ¼ 0:009 cmꢂ1, and jEj ¼ 0:000 cmꢂ1) indicate that there
are only subtle differences among the three spectra (Fig. S3).
The observed peaks can thus be fitted to either 14 or 54 or
74. Thus, we could not confidently assign the signals. The pat-
tern of the signals did not change from the initial to the latter
stages of irradiation (Fig. S6), and all signals were observable
even at 4 K.
In order to identify the origin of those signals, we investigat-
ed the temperature dependency of the signal intensities. Thus,
a 2-MTHF matrix containing 4-3N2 was irradiated at 4 K, and
two signals (at 147.9 and 262.0 mT) due to triplet species and
another two (at 310.4 and 324.5 mT) ascribable to higher-spin
species were used for the study of the temperature dependence.
The intensities of all the signals decreased as the temperature
was raised. Reproducible signal intensities were obtained in
the temperature range of 10–85 K. The plots of the signal

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2041
× 10-6
1.0
0.8
0.6
0.4
0.2
(a)
-130
-135
-140
-145
-150
-155
S = 1.17
S = 2.0
S = 1.0
:M at 2 K, : M at 3 K, : M at 5 K
0.0
0
100
200
Time / min
300
400
0
5
10
15
20
25
H / T (kOe K-1)
× 10-3
(b)
Fig. 5. Plots of M/Ms vs H=T of the photoproduct from
tris(diazo) compound (4-3N2) measured at 2.0 ( ), 3.0
0.0
: M at 2 K, : M at 3 K, : M at 5 K
(
), and 5.0 ( ) K. The solid lines represent theoretical
curves with S ¼ 1:0, 1.17, and 2.0.
-0.5
-1.0
-1.5
2.0
1.5
1.0
+: Mb at 2 K, +: Mb at 3 K, +: Mb at 5 K;
×: Ma at 2 K, ×: Ma at 3 K, ×: Ma at 5 K
0
10
20
30
40
50
× 103
Magnetic Field / H
Fig. 4. (a) Plot of magnetization (M in emu) as a function
of irradiation time observed in the photolysis of tris(diazo)
compound 4-3N2 in 0.10 mM 2-methyltetrahydrofuran
matrix measured at 5.0 K and 5 kOe. (b) Field dependence
of the magnetization of the photoproduct from tris(diazo)
compound 4-3N2 in 0.10 mM 2-methyltetrahydrofuran
matrix measured at 2.0, 3.0, and 5.0 K. Ma and Mb refer
to the magnetization value after and before irradiation,
respectively, and M ¼ Ma ꢂ Mb.
0.5
χ
0
0
20
40
T / K
60
80
Fig. 6. Temperature dependence of the observed ꢃmolT for
the photoproduct from tris(diazo) compound (4-3N2)
obtained at a field of 0.1 T with a SQUID.
irradiation, Mb, were also measured under the same condi-
tions. The magnetization (M) due to the species generated by
photolysis was then obtained by subtracting the corresponding
values obtained before and after irradiation (Fig. 4b). Thus, the
effects of any paramagnetic impurities could be cancelled by
this treatment. The plots of the magnetization normalized by
the saturation magnetization (M/Ms) versus the temperature-
normalized magnetic field (H=T) were analyzed in terms of
the Brillouin function as follows:1c,4,22
tion magnetization can be fitted with S ¼ 1. The fact that the
observed value is slightly over one may suggest that the septet
and/or quintet states lies slightly below the triplet ones (vide
infra).
In order to examine magnetic interaction between the three
carbene units, we obtained the magnetic susceptibility data for
the photoproduct from 4-3N2 in the temperature range 2–80 K
at a constant field of 100 mT. The temperature dependences of
M ¼ Ma ꢂ Mb ¼ FNgJꢂBBðꢃÞ;
ð1Þ
where F is the photolysis factor of the diazo compound, N is
the number of the molecules, J is the quantum number for
the total angular momentum, ꢂB is the Bohr magneton, and
ꢃmolT are given in Fig. 6. The ꢃmolT value was approximately
1.4 emu K molꢂ1 at 80 K and remained almost constant in the
temperature range down to 5 K when the temperature was low-
ered; the value increased very slightly at lower temperature. If
there is no magnetic interaction between the three carbene
units of tris(carbene) 4, the ꢃT value should become to be
3.0 emu K molꢂ1. However, the experimental value is only
1.4 emu K molꢂ1, which is smaller than one half of 3.0. Name-
ly, the small ꢃT value indicates that <50% of carbene units
were generated under the irradiation conditions in the SQUID
measurement. Although the value at 10 K was smaller than a
theoretical value calculated for S ¼ 3, the observed slight
increase of the ꢃmolT value under 5 K indicates that the three
carbene units coupled to each other by a rather weak ferromag-
netic interaction.
´
g is the Lande g-factor. Since these carbenes are constituted
of light elements, the orbital angular momentum should be
negligible, and J can be replaced with the spin quantum
number S. The M/Ms versus H=T plots are shown in Fig. 5,
together with theoretical curves with S ¼ 1:0, 1.17, and 2.0.23
The observed data for the photoproduct from the tris(diazo)
compound (4-3N2) followed the theoretical curve with S ¼ 1.
The observed data were fitted with Eq. 1 with S ¼ 1:17 and
F ¼ 1:05. It is possible that, if the energy levels of triplet,
quintet, and septet states are degenerate and, hence, the ferro-
magnetic interaction among the three carbene units is nearly
zero, the plot of the magnetization normalized by the satura-

2042 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
458
1.0
0.5
0.0
(A)
(A)
(a) before irrad.
1.0
0.5
0.0
(a) before irrad.
(b) after irrad.
(c) 90 K
472
362
(b) after irrad.
(c) 90 K
483
364
477
(d) 100 K
250
400
600
800
250
400
600
800
Wavelength / nm
Wavelength / nm
458
1.0
472
(B)
(B)
362
100 K
300 K
1.0
0.5
0.0
90 K
364
386
408
0.5
0.0
300 K
250
400
600
800
250
400
600
800
Wavelength / nm
Wavelength / nm
Fig. 7. UV–vis spectra obtained by irradiation of monodi-
azo compound 2-N2. (A) (a) Spectra of 2-N2 in 2-methyl-
tetrahydrofuran at 77 K. (b) The same sample after irradi-
ation (ꢁ > 350 nm). (c–d) The same sample after thawing
to 90 and 100 K. (B) UV–vis spectral change measured at
40 K increment upon warming the same sample from 100
to 300 K.
Fig. 8. UV–vis spectra obtained by irradiation of tris(diazo)
compound 4-3N2. (A) (a) Spectra of 4-3N2 in 2-methyl-
tetrahydrofuran at 77 K. (b) The same sample after irradi-
ation (ꢁ > 350 nm). (c) The same sample after thawing to
90 K. (B) UV–vis spectral change measured at 35 K incre-
ment upon thawing the same sample from 90 to 300 K.
UV–Vis Spectroscopic Study in a Matrix at Low Tem-
perature. In order to examine the stability of those high-spin
species in solution at room temperature, one should know the
UV–vis spectroscopic feature of the species, since time-
resolved UV–vis spectroscopy is most conveniently employed
for such a study.6,24
disappearance of the original absorption due to 4-3N2 and
concurrent growth of sharp and strong absorption bands at
350, 362, 385, 430, and 483 nm (Fig. 8), which are very sim-
ilar to those of monocarbene. Since ESR signals ascribable to
tris(carbene) 4 are observed under identical conditions, the
absorption spectrum can be safely assigned to 4.
Irradiation of a mono(diazo) compound (2-N2) in a 2-MTHF
matrix at 77 K resulted in rapid disappearance of the original
absorption due to 2-N2 and concurrent growth of sharp and
strong absorption bands at 364, 385, 414, 438, and 477 nm
(Fig. 7). Since ESR signals ascribable to triplet carbene are
observed under identical conditions, the absorption spectrum
When the matrix was gradually warmed, the bands became
sharper, and, at a longer wavelength, i.e., 483 nm, the bands
shifted to 472 nm at around 90 K. This change was again
attributed to the geometrical change of the triplet carbene
associated with a decrease in the viscosity of the matrixes upon
annealing. Upon further warming, the band started to increase
with a slight shift to a shorter wavelength, and this change con-
tinued until the matrix temperature increased to 220 K. This
observation suggests that geometrical changes of tris(carbene)
4 are not completed until 220 K. This can be compared with
the fact that a similar change in the UV–vis spectra of mono-
carbene 2 is completed at around 100 K. The difference is
interpreted to indicate that the geometrical relaxation of tris-
(carbene) requires much more enhanced fluidness of the matrix
at higher temperature than that of monocarbene 2. This differ-
ence can also be interpreted to indicate that the molecular
motion associated with geometrical relaxation of three units
of a congested dianthrylcarbene is more complex than that
with a single unit.
3
can be safely assigned to triplet carbene 2. When the matrix
was gradually warmed, the bands became sharper, and at a
longer wavelength, i.e., 477 nm, the bands shifted to 458 nm
at around 90 K, and this new band increased up to 100 K. This
change can be attributed to the geometrical change of the trip-
let carbene associated with a change in the viscosity of the
matrixes upon annealing, as has been revealed in ESR experi-
ments. Upon further thawing, the band started to disappear
slowly at around 130 K but did not completely disappear even
at 300 K.
These results are again to be compared with that observed
3
for 1, which exhibited essentially identical absorption bands
(362, 384, 412, 438, and 475 nm) but distinctly greater thermal
stability.
The absorption bands began to disappear at 260 K and com-
pletely vanished at around 300 K. Anomalous thermal stability
was again noted.
Similar photolysis (ꢁ > 350 nm) of a tris(diazo) compound
(4-3N2) in a 2-MTHF matrix at 77 K resulted in the rapid

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2043
1.0
1.0
0.8
1.0
1.0
0.8
0.6
0.4
0.2
0
467
456
0.8
2k/
ε
l = 1.1 × 10-3 min-1
t1/2 = 15 h
2k/
t1/2 = 13 h
ε
l = 1.5 × 10-3 min-1
0.8
0.6
0.4
0.2
0
363
0.6
0.4
0.2
0.6
0.4
0.2
0
0
4000
8000
0
Time / min
0
2000
4000
6000
Time / min
300
400
500
600
700
800
Wavelength / nm
300
400
500
600
700
800
Wavelength / nm
Fig. 9. Absorption of transient products formed during irra-
diation of monodiazo compound 2-N2 in degassed ben-
zene at room temperature recorded 0, 5, 10, 20, 44, 92,
and 140 h after excitation. Inset shows the time course
of the absorption at 456 nm.
Fig. 10. Absorption of transient products formed during
irradiation of tris(diazo) compound 4-3N2 in degassed
benzene at room temperature recorded 0, 5, 10, 20, 29,
68, 91, and 98 h after excitation. Inset shows the time
course of the absorption at 467 nm.
Time-Resolved UV–Vis Spectroscopic Study in Solution
at Room Temperature. In order to determine the stability
of the present carbenes more accurately, we estimated the life-
time in a degassed benzene at room temperature, in which we
measured the lifetime of a series of sterically congested diaryl-
carbenes.6 The lifetime of the present carbenes was too long to
monitor with a laser flash photolysis technique, which is rou-
tinely used in such studies; therefore, it was more convenient
to use a conventional UV–vis spectroscopic method in this
case.
decayed very slowly in a second-order fashion (2k="l ¼ 1:5 ꢃ
10ꢂ3 minꢂ1). The approximate half-life (t1=2) of 4 was estimat-
ed to be 13 hours, indicating that the stability of a unit mono-
carbene is maintained in tris(carbene).
Spin State of Tris(carbene) 4. Ovchinnikov has derived a
method for predicting the ground-state multiplicity of alternant
hydrocarbons based on valence bond theory by counting star-
red and unstarred carbons.26 His conclusion is expressed in
terms of Eq. 2,
Brief irradiation of the mono(diazo) compound (2-N2) in
ꢁ
degassed benzene at 20 C with a 70–90 mJ, 308 nm pulse
ꢀ
S ¼ ðn ꢂ nÞ=2;
ð2Þ
from a XeCl excimer laser produced transient absorption bands
showing strong maxima at 363 and 456 nm, which appeared
coincidently with the pulse (Fig. 9). These transient absorption
bands are essentially the same as those observed in the photol-
ysis of 2-N2 in 2-MTHF matrix at 77 K, followed by warming
ꢀ
where n and n are the number of starred and unstarred car-
bons, respectively. When triplet carbenes, which have spin
quantum number S ¼ 1, are used as spin sources, Eq. 2 can
be rewritten as follows:
3
to 100 K. Thus, we assigned the transient product to 2 in its
ꢀ
S ¼ n ꢂ n:
ð3Þ
geometrically relaxed state. The absorption bands decayed
very slowly, and the transient bands did not disappear com-
pletely even after the material was kept for several hours under
these conditions. The decay was found to be of the second
order (2k="l ¼ 1:1 ꢃ 10ꢂ3 minꢂ1). In accordance with the
kinetic observation, the MS analysis of the spent solution indi-
cated the presence of a peak at 1531.72 m=z, corresponding to
the dimer of the carbene as a major peak (Fig. S4).25 The
approximate half-life (t1=2) of 32 was estimated to be 15 hours.
This is shorter-lived than 31, whose half-life is estimated to be
14.5 days,12b but still fairly stable for a triplet carbene.
If this method is applied to predict the ground-state multi-
plicity of tris(carbene) 4, in Scheme 3, the numbers of starred
and unstarred carbon are 24 and 21, respectively, and, hence,
S ¼ 3 is predicted.
However, the ground-state multiplicity of the tris(carbene)
should be examined in terms of the connectivity.27 The non-
bonding molecular orbitals (NBMOs) of 4 are studied by
means of the simple Huckel molecular orbital method, in
¨
which the atoms having nonzero NBMO coefficients are star-
red. To a first approximation, the spin densities are assumed to
be generated on the atoms with such nonzero coefficients.
However, the ethynyl groups in 4, which connect dianthrylcar-
bene to the central phenyl ring, are introduced on the anthryl
group at sites of very small spin density (orbital nodes). There-
fore, in 4, the coefficients in NBMOs in the atoms in the bridg-
ing group are nearly zero. In other words, the NBMOs are
mostly localized in three dianthrylcarbene moieties. The mag-
nitude of these spin interactions is thus expected to be very
small.
Similar irradiation of the tris(diazo) compound (4-3N2) also
produced transient absorption bands showing strong maxima at
362 and 467 nm (Fig. 10). Again, the transient absorption bands
were essentially the same as those observed for tris(carbene)
4 in its relaxed state. The fact that the molecular extinction
coefficient (" ¼ 4:83 ꢃ 104 molꢂ1 cmꢂ1 dm3) of this species
3
was roughly three times as large as that observed for 1 (" ¼
1:83 ꢃ 104 molꢂ1 cmꢂ1 dm3) indicates that the transient bands
are due to tris(carbene) 4 rather than to a monocarbene gener-
ated by incomplete decomposition of 4-3N2. This is further
supported by the finding that analysis of the spent solution by
IR spectroscopy did not indicate the presence of any character-
istic band due to the diazo group. Again, the transient bands
Connectivity across nodal (inactive) sites is typical for many
systems that are disjoint in the Borden/Davidson sense.28,29
Many of these systems have singlet ground states with low-
lying high spin states.30,31 Experimentally, in these systems,

2044 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
signals due to a higher-spin state are detected by ESR, but the
signal intensities deviate from the Curie law. However, in the
present case, the intensities of signals due to both triplet and
higher-spin state(s) decreased linearly with the reciprocal of
the temperature, in good agreement with the Curie law18 at
least up to 10 K. If one assumes that all signals originate from
tris(carbene), it could be interpreted that all states, including
singlet states, are degenerate or lie close to each other. In other
words, the exchanging coupling constant (J) is extremely
small or nearly zero.32 This may be partly because of small
spin densities in the dianthrylcarbene as a result of extensive
delocalization.12,33
The SQUID measurements also support the small interac-
tion. Namely, the slight increase of the ꢃmolT value under 5
K indicates that the three carbene units in the tris(carbene)
coupled by an extremely small ferromagnetic interaction.
Moreover, the fact that the observed spin quantum number is
slightly over unity may suggest that all the states of the tris-
(carbene) are almost degenerate, but that the septet and/or
quintet states lies slightly below the triplet ones.
Analogous spin systems are not unprecedented. For instance,
1,1,2,2-tetramethyl-1,2-di(3-nitrenophenyl)disilane showed
ESR signals of both triplet and quintet states, and the signal
intensities followed Curie’s law in the range of 10–64 K. This
is interpreted to indicate that the spin states are nearly degen-
erate with 0 > ꢀES-Q > ꢂ3:4 cal/mol.30c However, in this
system, ZFS parameters for the triplet state of bis(nitrene)
are not identical with those of triplet mononitrene.
It is to be noted here that, in the variable-temperature study
of ESR signals (Fig. 2), the signals at around 310–325 mT
started to decrease rather sharply at around 280 K and disap-
peared almost completely at 300 K, while the signals due to
triplet species remained mostly unchanged even at 280 K. This
suggests that triplet signals and signals at around 310–325 mT
are probably not arising from the same species. In accord with
this suggestion, we found that the intensities of the signals of
the two states depend on the wavelength of the irradiating
light. Thus, irradiation of 4-3N2 with a longer-wavelength
light (ꢁ > 450 nm) gave mainly signals due to the triplet spe-
cies, while irradiation with a shorter wavelength light (ꢁ >
350 nm) gave the signals of both states.
If tris(diazo) compound (4-N2) is decomposed in stepwise
manner, bis(diazo)monocarbene will be initially formed. Since
monocarbene 1 formed by photolysis of mono(diazo) com-
pound 1-N2 exhibits rather strong absorption bands around
450 nm, a longer wavelength than that of the precursor 1-N2,
the monocarbene is also expected to have strong absorption
bands in visible region where the diazo group has no remarka-
ble absorption band. Thus irradiation of tris(diazo) compound
(4-N2) with the wavelength longer than 450 nm is expected
to produce the monocarbene to a significant extent. On the other
hand, monocarbene 1 has no strong absorption band in the
region of 370–420 nm. Thus, irradiation of tris(diazo) com-
pound (4-N2) with the wavelength longer than 350 nm is
expected to decompose the monocarbene to generate bis- and
tris-carbenes. Therefore, it is reasonable that a mixed spectrum
of a monocarbene and high-spin species (a bis(carbene) and a
tris(carbene)) is observed upon >350 nm irradiation.
the triplet signals and the other ones are originating from
two different species. From ZFS parameters (E=D) for triplet
signals, it is mostly likely that the signals are due to triplet
monocarbene, not due to tris(carbene).33
The computer simulation did not discriminate between the
3
5
7
signals for triplet 4, quintet 4, and septet states 4 (Fig. S3).
It may be, then, that the signals around 310–325 mT in Fig. 2
consist of triplet, quintet, and septet states which are all nearly
degenerate.
Conclusion
The present investigation demonstrates that the preparation
of usable high-spin organic materials employing a stable triplet
carbene unit as a spin source is an achievable objective.
In the present system, the main reason why the high-spin
states are not exclusively generated and the S value is remark-
ably smaller than the theoretical one is that the connectivity was
wrong; however, it would be possible to generate septet tris-
(carbene) by connecting the anthryl carbene unit at either the
2 or the 4 position. All attempts so far to prepare the desired
tris(diazo) compounds by using the Sonogashira coupling reac-
tion, however, have been unsuccessful. This is partly because of
steric reasons, as molecular models suggest that connecting
three dianthryl units from these positions causes more steric
hindrance in the resulting tris(diazo) compounds than 4-3N2.
However, there are other ways to connect diazo units to a link-
er. One way to realize this is to introduce a bridging ligand,
which makes a polymeric chain structure by the ligation with
metal ions. For instance, diazo(di-4-pyridyl)methane formed
a 1:1 complex with bis(hexafluoroacetylacetonato)copper(II)
[Cu(hfac)2]. Upon photolysis, this complex generated a ferro-
magnetic chain with the apparent spin quantum number of 33
as a result of ferromagnetic interaction between the 3d spin
of Cu(II) and 2p spins of di(4-pyridyl)carbene.3,34 This suggests
that once pyridyl groups are introduced into a sterically con-
gested diazomethane, a precursor for a persistent triplet car-
bene, it may be possible to prepare a complex with metal ions
that can generate a persistent high-spin polycarbene.
The finding that a precursor diazo unit can basically be han-
dled as a building block to construct poly-diazo compounds
makes such a project very realistic. Thus, the building block
could very easily be introduced on properly designed ꢄ-topo-
logical linkers ad libitum. In other words, a poly-diazo com-
pound designed to generate a desired high-spin polycarbene
can be prepared with great ease. A starburst nona(diazo) com-
pound, for instance, was prepared by trimerization of aryl
ethynyl ketone, followed by oxidation, hydrazonation of the
resulting nona-ketone, and poly-oxidation.4 The method is
elegant, but it is not applicable to other more complicated sys-
tems, including sterically congested derivatives.
The study also reveals that persistent triplet carbenes, even
though they lose most of their typical reactivity as a triplet car-
bene, still retain electronic properties and act as a spin source
when aligned properly in the ꢄ-electron frameworks to gener-
ate a high-spin molecule, just as a prototypical triplet carbene
does.
Most importantly, polycarbenes consisting of a persistent
triplet carbene are also very persistent, surviving up to 300 K
and for up to a day in solution at room temperature. Moreover,
Judging from these observations, we think it is likely that

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2045
ꢁ
mL, 11.37 mmol) was added dropwise at 0 C, and the mixture
undesirable bimolecular side reactions, which generate chemi-
cal defects among the topologically controlled high-spin
(S ¼ 1) centers for extended spin alignment, are rigorously
avoided.
Although high-spin organic materials with the lifetime of a
day are still ephemeral for practical use, in the light of the fact
that a more persistent triplet carbene surviving for days under
ambient conditions is close to being realized, it will not be
long before a stable high-spin polycarbene is available.
was stirred at room temperature overnight. To this mixture, an
aqueous solution of 10% sodium hydroxide was added carefully.
The reaction mixture was extracted with carbon tetrachloride,
and the organic layer was washed with water, and dried over an-
hydrous sodium sulfate. After removal of the solvent, the resulting
crude product was washed with hexane to afford 9-bromo-10-(4-t-
butyl-2,6-dimethylphenyl)anthracene (3.7 g, 86%) as a yellow
crystal: mp 180 ꢁC. 1H NMR (CDCl3): ꢀ 1.44 (s, 9H), 1.72 (s,
6H), 7.24 (s, 2H), 7.35 (ddd, J ¼ 8:82, 6.43, 1.29 Hz, 2H), 7.48
(dd, J ¼ 8:82, 1.29 Hz, 2H), 7.57 (ddd, J ¼ 8:82, 6.43, 1.29 Hz,
2H), 8.60 (dd, J ¼ 8:82, 1.10 Hz, 2H). 13C NMR (CDCl3): ꢀ 20.4,
31.5, 34.5, 122.1, 124.5, 125.8, 126.5, 127.0, 128.0, 130.5, 130.6,
134.0, 136.8, 150.7. IR (KBr, cmꢂ1) 3065w, 2960s, 2951m,
2862m, 1605w, 1559w, 1540vw, 1507vw, 1437s, 1374vw, 1362m,
1341m, 1300w, 1256m, 1229vw, 1190w, 1148w, 1026m, 945s,
915m, 877w, 866w, 845w, 784w, 756vs, 652s, 605w, 577w. MS
m=z (relative intensity) 418 (M þ 2, 31.3), 416 (Mþ, 31.7), 338
(100), 323 (38.5); HRMS calcd for C26H25Br 416.1139, found
m=z 416.1125.
Experimental
1H NMR and 13C NMR spectra were
General Methods.
recorded on a JEOL JM-AL300FT/NMR spectrometer in CDCl3
with Me4Si as an internal reference. IR spectra were measured on
a JASCO FT/IR-410 spectrometer, and UV–vis spectra were
recorded on a JASCO CT-560 spectrophotometer. ESR spectra
were measured on a JEOL JES-TE200D ESR spectrometer. The
mass spectra were recorded on a JEOL JMS-600H mass spectrom-
eter and an AB Voyager-DE PRE MALDI-TOF mass spectrome-
ter. Gel permeation chromatography (GPC) was carried out on
a JASCO model HLC-01 instrument. The GPC column was a
Shodex H-2001. Thin-layer chromatography was carried out on
a Merck Kieselgel 60 PF254. Column chromatography was per-
formed on silica-gel (Kanto Chemical) for column chromatogra-
phy or ICN alumina (ICN Biomedicals) for dry column chroma-
tography.
(c) Preparation of 10-(4-t-Butyl-2,6-dimethylphenyl)anthra-
cene-9-carbaldehyde: To a solution of the bromide (2.0 g, 4.79
mmol) in anhydrous diethyl ether (10 mL) was added 2.6 M butyl-
lithium in hexane (2.2 mL, 5.75 mmol) at 0 ꢁC under nitrogen
atmosphere. Then the mixture was stirred for 2 h at 0 ꢁC. Absolute
DMF (1.12 mL, 14.37 mmol) was added to the lithiated mixture at
0
ꢁC, and the mixture was refluxed overnight. The mixture was
Preparation of [3-Bromo-10-(4-t-butyl-2,6-dimethylphenyl)-
9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]diazo-
methane (2-N2). (a) Preparation of 9-(4-t-Butyl-2,6-dimethyl-
phenyl)anthracene: To a solution of 4-t-butyl-2,6-dimethylbro-
mobenzene35 (5.0 g, 20.7 mmol) in anhydrous diethyl ether (40
mL) was added 1.4 M t-butyllithium in pentane (30.4 mL, 45.6
mmol) at ꢂ78 ꢁC under nitrogen atmosphere. The mixture was
stirred for 30 min at ꢂ78 ꢁC and allowed to warm up to room tem-
perature. To this mixture, a solution of the anthrone36 (4.0 g, 20.7
mmol) in anhydrous toluene (100 mL) was added, and the mixture
was refluxed overnight. To the mixture allowed to cool to room
temperature, a saturated ammonium chloride aqueous solution
was added carefully. The reaction mixture was extracted with
diethyl ether, and the organic layer was washed with water, and
dried over anhydrous sodium sulfate. A brown solid obtained after
concentration of the solution was chromatographed on a silica-
gel column eluted with hexane to afford a crude product, which
was purified by a short-path distillation (85 ꢁC/1.0 mmHg). 9-
(4-t-Butyl-2,6-dimethylphenyl)anthracene was obtained as a pale-
yellow solid in 53% yield (3.7 g): mp 167 ꢁC. 1H NMR (CDCl3): ꢀ
1.35 (s, 9H), 1.64 (s, 6H), 7.15 (s, 2H), 7.20 (ddd, J ¼ 8:45, 6.43,
1.10 Hz, 2H), 7.32 (ddd, J ¼ 8:45, 6.43, 1.29 Hz, 2H), 7.37 (d,
J ¼ 8:82 Hz, 2H), 7.93 (d, J ¼ 8:45 Hz, 2H), 8.35 (s, 1H).
13C NMR (CDCl3): ꢀ 20.4, 31.6, 34.5, 124.4, 125.1, 125.5, 126.0,
126.1, 128.6, 129.8, 131.6, 134.4, 136.0, 137.0, 150.3. IR (KBr,
cmꢂ1) 3051m, 3024w, 2961s, 2949s, 2921w, 2899w, 2862m,
1622vw, 1607w, 1575w, 1517w, 1478w, 1440m, 1408vw, 1374w,
1362m, 1302w, 1231w, 1189m, 1161vw, 1013m, 953w, 946m,
877s, 839s, 791m, 733vs, 652m, 606m, 561w, 513w. MS m=z
(relative intensity) 338 (Mþ, 100), 323 (41.2); HRMS calcd for
C26H26 338.2035, found m=z 338.2034.
allowed to cool to room temperature, and a saturated ammonium
chloride aqueous solution was added carefully. Insoluble material
was filtered and washed with water to give a crude aldehyde (1.4
g). The filtrate was extracted with diethyl ether, and the organic
layer was washed with water, dried over anhydrous sodium sulfate
and evaporated to leave a crude product (0.7 g). The combined
crude product was chromatographed on a silica-gel column eluted
with hexane/dichloromethane (1:1). 10-(4-t-Butyl-2,6-dimethyl-
phenyl)anthracene-9-carbaldehyde was obtained as a yellow solid
in 57% yield (1.0 g): mp 218 ꢁC. 1H NMR (CDCl3): ꢀ 1.44 (s, 9H),
1.72 (s, 6H), 7.24 (s, 2H), 7.38–7.43 (m, 2H), 7.56 (d, J ¼ 8:82 Hz,
2H), 7.64–7.69 (m, 2H), 9.04 (d, J ¼ 9:00 Hz, 2H), 11.60 (s, 1H).
13C NMR (CDCl3): ꢀ 20.3, 31.5, 34.5, 123.7, 124.59, 124.60,
125.9, 127.1, 128.8, 129.4, 131.9, 133.9, 136.3, 145.4, 151.0,
193.4. IR (KBr, cmꢂ1) 3066w, 3044w, 2960s, 2915w, 2864m,
2765w, 1680vs (ꢅC=O), 1607vw, 1556m, 1483w, 1438m, 1406vw,
1362w, 1270m, 1230vw, 1190m, 1152w, 1048m, 1029vw, 945m,
900w, 876w, 799w, 758m, 753m, 734w, 662m, 642vw, 602w,
569vw. MS m=z (relative intensity) 366 (Mþ, 100), 351 (36.9);
HRMS calcd for C27H26O 366.1984, found m=z 366.1986.
(d) Preparation of 2-Bromo-9-(4-t-butyl-2,6-dimethylphen-
yl)anthracene: To a solution of 4-t-butyl-2,6-dimethylbromo-
benzene (2.5 g, 10.3 mmol) in anhydrous diethyl ether (10 mL)
was added 1.4 M t-butyllithium in pentane (15.0 mL, 22.6 mmol)
at ꢂ78 ꢁC under nitrogen atmosphere. The mixture was stirred for
ꢁ
30 min at ꢂ78 C and allowed to warm up to room temperature.
To this mixture, a solution of 2-bromoanthrone37 (2.8 g, 10.3
mmol) in anhydrous toluene (100 mL) was added, and the mixture
was refluxed overnight. The mixture was allowed to cool to room
temperature, and a saturated ammonium chloride aqueous solution
was added carefully. The reaction mixture was extracted with
diethyl ether, and the organic layer was washed with water and
dried over anhydrous sodium sulfate. A brown solid obtained after
concentration of the solution was chromatographed on a silica-gel
column eluted with hexane to afford a crude product, which was
(b) Preparation of 9-Bromo-10-(4-t-butyl-2,6-dimethylphen-
yl)anthracene: A solution of the phenylanthracene (3.5 g, 10.34
mmol) in carbon tetrachloride (40 mL) was cooled to 0 ꢁC. To this
mixture, a carbon tetrachloride solution (3 mL) of bromine (0.56

2046 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
purified by a short-path distillation (85 ꢁC/1.0 mmHg). 2-Bromo-
9-(4-t-butyl-2,6-dimethylphenyl)anthracene was obtained as a
pale yellow solid in 38% yield (1.6 g): mp 188 ꢁC. 1H NMR
(CDCl3): ꢀ 1.45 (s, 9H), 1.72 (s, 6H), 7.24 (s, 2H), 7.30–7.51 (m,
4H), 7.63 (s, 1H), 7.92 (d, J ¼ 9:00 Hz, 1H), 8.03 (d, J ¼ 8:45 Hz,
1H), 8.43 (s, 1H). 13C NMR (CDCl3): ꢀ 20.4, 31.5, 34.5, 120.0,
124.6, 125.6, 126.1, 126.2, 126.3, 127.7, 128.6, 128.7, 129.8,
130.3, 130.4, 130.6, 131.7, 133.6, 135.5, 136.8, 150.6.
977w, 944s, 902m, 869m, 815m, 802m, 780s, 768s, 728vw, 706w,
676w, 664m, 626w, 610w, 588w, 569w, 556w, 526w, 511w. MS
m=z (relative intensity) 784 (M þ 2, 46.3), 783 (36.5), 782 (Mþ,
59.1), 445 (50.6), 443 (49.7), 366 (64.8), 365 (100); HRMS calcd
for C53H51BrO 782.3123, found m=z 782.3196.
(g) Preparation of [3-Bromo-10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-an-
thryl]chloromethane: Into a solution of the methanol ꢁ(600 mg,
0.77 mmol) in anhydrous benzene (20 mL) cooled to 0 C, a hy-
drogen chloride gas generated by adding hydrochloric acid (3 mL)
to sulfuric acid (4.5 mL)ꢁwas bubbled for 15 min and the mixture
was stirred for 1 h at 0 C. Removal of the solvent afforded [3-
bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl][10-(4-t-butyl-
2,6-dimethylphenyl)-9-anthryl]chloromethane (615 mg, quantita-
tively) as a brown viscous liquid as a crude product. This was used
without further purification since the chloride was found to be
rather unstable. 1H NMR (CDCl3): ꢀ 1.45 (brs, 18H), 1.77 (brs,
12H), 7.16–7.36 (m, 11H), 7.45–7.54 (m, 1H), 7.51–7.55 (m, 2H),
7.65 (d, J ¼ 1:42 Hz, 1H), 8.54–8.57 (m, 4H), 8.67 (s, 1H).
(h) Preparation of Ethyl [3-Bromo-10-(4-t-butyl-2,6-di-
methylphenyl)-9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-
anthryl]methylcarbamate: A solution of the chloromethane
(615 mg, 0.77 mmol) in anhydrous 1,4-dioxane (30 mL) was add-
ed to a molten mixture of silver tetrafluoroborate (204 mg, 1.12
(e) Preparation of 2,10-Dibromo-9-(4-t-butyl-2,6-dimethyl-
phenyl)anthracene:
A solution of 2-bromo-9-(4-t-butyl-2,6-
dimethylphenyl)anthracene (1.0 g, ꢁ2.40 mmol) in carbon tetra-
chloride (15 mL) was cooled to 0 C. To this mixture, a carbon
tetrachloride solution (2 mL) of bromine (0.13 mL, 2.64 mmol)
ꢁ
was added dropwise at 0 C, and the mixture was stirred at room
temperature overnight. To this mixture, an aqueous solution of
10% sodium hydroxide was added carefully. The reaction mixture
was extracted with carbon tetrachloride, and the organic layer was
washed with water and dried over anhydrous sodium sulfate. After
removal of the solvent, the resulting crude product was washed
with hexane to afford 2,10-dibromo-9-(4-t-butyl-2,6-dimethyl-
phenyl)anthracene (1.0 g, 87%) as a yellow solid: mp 210 ꢁC.
1H NMR (CDCl3): ꢀ 1.45 (s, 9H), 1.72 (s, 6H), 7.25 (s, 2H), 7.35–
7.45 (m, 2H), 7.58–7.63 (m, 2H), 7.64 (s, 1H), 8.49 (d, J ¼ 9:00
Hz, 1H), 8.57 (d, J ¼ 8:46 Hz, 1H). 13C NMR (CDCl3): ꢀ 20.4,
31.5, 34.5, 120.6, 122.5, 124.8, 126.51, 126.55, 127.5, 127.9,
128.1, 128.8, 130.1, 130.5, 130.7, 131.2, 131.3, 133.2, 136.2,
136.7, 151.0. IR (KBr, cmꢂ1) 3069w, 2958vs, 2916w, 2899w,
2860m, 1604m, 1546w, 1512vw, 1438s, 1374vw, 1361m, 1351w,
1323s, 1299w, 1255vw, 1228vw, 1186m, 1161vw, 1065m,
1026vw, 947vs, 915s, 874m, 849vw, 803s, 777s, 754vs, 719m,
664m, 652m, 579m, 511w. MS m=z (relative intensity) 498
(M þ 4, 50.3), 496 (M þ 2, 100), 494 (Mþ, 52.9); HRMS calcd
for C26H24Br2 494.0244, found m=z 494.0242.
ꢁ
mmol) and urethane (1.30 g, 15.4 mmol) at 60 C. After addition
was completed, the mixture was refluxed overnight. After cooling,
silver chloride was removed from the reaction mixture by filtra-
tion and washed with dichloromethane. The filtrate was washed
with water and the organic layer was dried over anhydrous sodium
sulfate and evaporated. The crude product was chromatographed
on a silica-gel column eluted with hexane/dichloromethane (1:1).
Ethyl [3-bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl][10-
(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]methylcarbamate was
obtained as an ocherous solid in 67% yield (440 mg): mp 187–
190 ꢁC. 1H NMR (CDCl3): ꢀ 1.25 (t, J ¼ 7:04 Hz, 3H), 1.44
(brs, 18H), 1.73 (brs, 12H), 4.20 (q, J ¼ 6:92 Hz, 2H), 5.94 (d,
J ¼ 7:72 Hz, 1H), 7.24 (s, 4H), 7.25–7.38 (m, 4H), 7.43–7.52
(m, 5H), 7.64 (d, J ¼ 1:84 Hz, 1H), 8.42–8.48 (m, 5H), 8.69
(d, J ¼ 6:80 Hz, 1H). 13C NMR (CDCl3): ꢀ 14.7, 20.3, 31.5,
34.5, 54.6, 61.4, 119.5, 124.2, 124.4, 124.7, 124.9, 125.5, 126.2,
126.6, 127.3, 127.4, 128.3, 128.7, 129.1, 130.0, 130.2, 130.6,
131.0, 133.0, 133.95, 134.01, 134.7, 136.8, 137.2, 138.2, 150.5,
150.7, 155.3. IR (KBr, cmꢂ1) 3455w (ꢅN{H), 3063w, 2963vs,
2867m, 1725vs (ꢅC=C), 1604m, 1570vw, 1495s, 1480s, 1443s,
1407w, 1393vw, 1376w, 1362m, 1324w, 1303w, 1276vw,
1228s, 1172vw, 1094vw, 1043m, 998vw, 944s, 900w, 868m,
816w, 803w, 763s, 754s, 703vw, 659m, 646vw, 606vw. MS m=z
(relative intensity) 855 (M þ 3, 52), 854 (M þ 2, 100), 853
(M þ 1, 70), 852 (Mþ, 92), 767 (67), 765 (60); HRMS calcd for
C56H55BrNO2 852.3411, found m=z 852.3357
(f) Preparation of [3-Bromo-10-(4-t-butyl-2,6-dimethylphen-
yl)-9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]-
methanol: To a solution of 2,10-dibromo-9-(4-t-butyl-2,6-di-
methylphenyl)anthracene (600 mg, 1.21 mmol) in anhydrous
diethyl ether (8 mL) was added 2.6 M butyllithium in hexane
(0.48 mL, 1.21 mmol) at 0 ꢁC under nitrogen atmosphere and
the mixture was stirred for 2 h at 0 ꢁC. A solution of 10-(4-t-
butyl-2,6-dimethylphenyl)anthracene-9-carbaldehyde (443 mg,
1.21 mmol) in anhydrous tetrahydrofuran (8 mL) was added to
the lithiated mixture at 0 ꢁC, and the mixture was heated overnight
ꢁ
at 40 C. The mixture was allowed to cool to room temperature,
and a saturated ammonium chloride solution was added carefully.
The reaction mixture was extracted with diethyl ether, and the
organic layer was washed with water, dried over anhydrous
sodium sulfate and evaporated. The crude product was recrystal-
lized from dichloromethane–hexane to give [3-bromo-10-(4-t-
butyl-2,6-dimethylphenyl)-9-anthryl][10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl]methanol (535 mg, 56%) as a yellow solid: mp
247–250 ꢁC. 1H NMR (CDCl3): ꢀ 1.448 (s, 9H), 1.452 (s, 9H),
1.76 (s, 6H), 1.77 (s, 6H), 2.97 (d, J ¼ 3:12 Hz, 1H), 7.19–7.28
(m, 7H), 7.227 (s, 2H), 7.233 (s, 2H), 7.45–7.55 (m, 3H), 7.65
(d, J ¼ 2:02 Hz, 1H), 8.52–8.57 (m, 4H), 8.67 (d, J ¼ 2:94 Hz,
1H). 13C NMR (CDCl3): ꢀ 20.3, 31.5, 31.6, 34.49, 34.51, 73.0,
119.6, 124.5, 124.7, 125.0, 125.3, 125.6, 126.0, 126.2, 127.1,
127.2, 127.8, 128.4, 128.9, 130.0, 130.2, 130.6, 133.8, 134.1,
134.7, 136.7, 136.8, 137.4, 138.5, 143.9, 144.1, 150.5, 150.7. IR
(KBr, cmꢂ1) 3519m (ꢅO{H), 3062w, 2960vs, 2921m, 2865m,
1604m, 1559w, 1540w, 1522w, 1507w, 1477m, 1457w, 1442s,
1407vw, 1393vw, 1362s, 1288vw, 1228m, 1191w, 1071m, 1042m,
(i) Preparation of [3-Bromo-10-(4-t-butyl-2,6-dimethylphen-
yl)-9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]di-
azomethane (2-N2): Some dinitrogen tetraoxide (1.7 g, 18.91
mmol) was bubbled into anhydrous carbon tetrachloride (10 mL)
ꢁ
which had been cooled to ꢂ20 C. After sodium acetate (3.1 g,
37.82 mmol) was added to this mixture, a solution of the carba-
mate (135 mg, 0.16 mmol) in anhydrous carbon tetrachloride (5
mL) was added at ꢂ20 ꢁC and the mixture was stirred for 2 h
at 0 ꢁC. The reaction mixture was then poured into crushed ice
and extracted with carbon tetrachloride. The organic layer was
washed with aqueous sodium hydrogencarbonate and water and
dried over anhydrous sodium sulfate. A brown solid obtained after

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2047
concentration of the solution was mixed with anhydrous tetra-
hydrofuran (5 mL) under nitrogen atmosphere and the mixture
was cooled to ꢂ20 ꢁC. Potassium t-butoxide (38 mg, 0.32 mmol)
was added and the mixture was stirred overnight at room temper-
ature. The reaction mixture was extracted with diethyl ether, and
the organic layer was washed with water and dried over anhydrous
sodium sulfate. After evaporation of the solvent, the residue was
purified by GPC (7 cycles, chloroform as eluent), and then
by TLC with hexane/dichloromethane (1:2) as eluent to give [3-
bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl][10-(4-t-butyl-
2,6-dimethylphenyl)-9-anthryl]diazomethane (2-N2) (51 mg, 40%)
as an orange viscous liquid. 1H NMR (CDCl3): ꢀ 1.45 (s, 9H),
1.46 (s, 9H), 1.78 (s, 12H), 7.27 (s, 4H), 7.30–7.43 (m, 7H),
7.53 (d, J ¼ 8:64 Hz, 1H), 7.57 (d, J ¼ 9:37 Hz, 2H), 7.69 (d, J ¼
2:02 Hz, 1H), 8.18 (d, J ¼ 9:55 Hz, 1H), 8.27 (d, J ¼ 7:90 Hz,
2H), 8.34 (d, J ¼ 8:64 Hz, 1H). 13C NMR (CDCl3): ꢀ 20.37,
20.40, 31.5, 31.6, 34.5, 56.4, 120.2, 124.5, 124.8, 125.3, 125.5,
126.2, 126.6, 127.3, 127.4, 128.3, 128.7, 129.1, 130.0, 130.20,
130.25, 130.3, 131.0, 131.3, 133.7, 134.4, 136.8, 136.9, 137.2,
138.3, 150.6, 150.8. IR (NaCl, cmꢂ1) 3062w, 2963s, 2865w,
2039vs (ꢅC=N2), 1604w, 1440m, 1409w, 1388w, 1362w, 1303w,
1228w, 1187w, 1028w, 967vw, 946s, 907s, 868w, 819w, 809vw,
784vw, 765m, 733s.
Preparation of Tris(diazo) Compound (4-3N2). (a) Prepa-
ration of [10-(4-t-Butyl-2,6-dimethylphenyl)-9-anthryl][3-tri-
methylsilylethynyl-10-(4-t-butyl-2,6-dimethylphenyl)-9-an-
thryl]diazomethane (3-N2(TMS)): The diazo compound (2-N2)
(30 mg, 0.038 mmol) and catalytic amounts of [PdCl2(Ph3P)2] and
copper(I) iodide were mixed with anhydrous tetrahydrofuran (0.7
mL) under nitrogen atmosphere. To this mixture, anhydrous tri-
ethylamine (0.3 mL) and trimethylsilylacetylene (0.05 mL, 0.30
mmol) were added successively, and the mixture was stirred over-
night at 40 ꢁC. After cooling, catalysts were removed from the
reaction mixture by filtration through a short alumina column
(hexane). After evaporation of the solvent, the residue was puri-
fied by GPC (overnight, chloroform as eluent) to afford [10-
(4-t-butyl-2,6-dimethylphenyl)-9-anthryl][3-trimethylsilylethynyl-
3302m (ꢅꢄC{H), 3062w, 2965vs, 2926m, 2866m, 2248w (ꢅCꢄC),
2039vs (ꢅC=N2), 1740w, 1636vw, 1606vw, 1439m, 1362m,
1281w, 1229m, 1191vw, 1138w, 1100w, 1029m, 944m, 908s,
869m, 824m, 765m, 735s, 651w.
(c) Preparation of Tris(diazo) Compound (4-3N2): The
diazomethane 3-N2 (35 mg, 0.047 mmol), 1,3,5-triiodobenzene
(6.5 mg, 0.016 mmol) and catalytic amounts of [PdCl2(Ph3P)2]
and copper(I) iodide were mixed with anhydrous tetrahydrofuran
(1 mL) under nitrogen atmosphere. To this mixture, anhydrous tri-
ethylamine (0.5 mL) was added and this combination was stirred
overnight at room temperature. Removal of the catalyst by filtra-
tion through a short alumina column, followed by removal of the
solvent, gave crude material which was purified by GPC (8 cycles,
chloroform as eluent) and TLC (CH2Cl2:hexane = 1:1) to give
tris(diazo) compound (4-3N2) (9 mg, 25%) as an orange viscous
liquid. 1H NMR (CDCl3): ꢀ 1.46 (brs, 54H), 1.79 (brs, 36H), 7.28
(brs, 12H), 7.30–7.41 (m, 21H), 7.51 (d, J ¼ 8:82 Hz, 3H), 7.54
(s, 3H), 7.58 (d, J ¼ 8:27 Hz, 6H), 7.72 (d, J ¼ 1:29 Hz, 3H),
8.24–8.35 (m, 12H). 13C NMR (CDCl3): ꢀ 20.37, 20.44, 31.6,
34.5, 56.4, 88.8, 91.3, 119.6, 123.9, 124.5, 124.7, 125.2, 125.4,
125.5, 125.8, 126.5, 127.0, 127.37, 127.44, 128.6, 129.3, 129.6,
129.8, 130.27, 130.35, 130.8, 130.9, 133.9, 134.4, 136.5, 136.9,
138.2, 138.3, 150.6, 150.7. IR (NaCl, cmꢂ1) 3064w, 2964s,
2865w, 2246vw (ꢅCꢄC), 2038vs (ꢅC=N2), 1654m, 1577w,
1559w, 1477w, 1457w, 1438w, 1362w, 1278w, 1228w, 1190w,
1129w, 1098w, 1028w, 1000w, 945m, 907s, 867w, 824w, 786w,
763m, 731s, 668w.
Preparation of Tris(ketone) (4-3O). (a) Preparation of 3-
Bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl
10-(4-t-
Butyl-2,6-dimethylphenyl)-9-anthryl Ketone (2-O): [3-bromo-
10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl][10-(4-t-butyl-2,6-di-
methylphenyl)-9-anthryl]methanol (250 mg, 0.33 mmol), manga-
nese(IV) oxide (295 mg, 3.40 mmol), and anhydrous sodium sulfate
(483 mg, 3.40 mmol) were mixed in anhydrous dichloromethane
(25 mL), and the mixture was stirred overnight at room tempera-
ture. The reaction mixture was filtered and the filtrated organic lay-
er was evaporated. The crude product was chromatographed on a
silica-gel column eluted with hexane/dichloromethane (2:1). 3-
Bromo-10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl 10-(4-t-butyl-
2,6-dimethylphenyl)-9-anthryl ketone 2-Oꢁwas obtained in 89%
10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl]diazomethane
3-
N2(TMS) (12.7 mg, 41%) as an orange viscous liquid. 1H NMR
(CDCl3): ꢀ 0.21 (s, 9H), 1.45 (s, 9H), 1.46 (s, 9H), 1.78 (brs, 12H),
7.27 (s, 2H), 7.28 (s, 2H), 7.31–7.37 (m, 7H), 7.47–7.57 (m, 3H),
7.69 (d, J ¼ 0:92 Hz, 1H), 8.24 (d, J ¼ 9:74 Hz, 1H), 8.27 (d, J ¼
9:16 Hz, 2H), 8.31 (d, J ¼ 8:64 Hz, 1H). IR (NaCl, cmꢂ1) 3063w,
2963s, 2866w, 2151w (ꢅCꢄC), 2038vs (ꢅC=N2), 1636w, 1477w,
1438w, 1362w, 1250m, 1228w, 1027w, 986w, 944w, 859s, 843s,
760s, 668w.
1
yield (230 mg) as a yellow solid: mp 248 C. H NMR (CDCl3):
ꢀ 1.45 (brs, 18H), 1.78 (brs, 12H), 7.27 (brs, 4H), 7.30–7.39 (m,
7H), 7.52–7.58 (m, 3H), 7.76 (s, 1H), 8.24 (d, J ¼ 9:74 Hz, 1H),
8.27 (d, J ¼ 9:16 Hz, 2H), 8.31 (d, J ¼ 8:64 Hz, 1H). 13C NMR
(CDCl3): ꢀ 20.3, 31.5, 34.5, 120.2, 124.6, 124.8, 125.3, 125.6,
126.2, 126.9, 127.5, 127.6, 128.1, 128.3, 129.6, 129.8, 129.9,
130.3, 130.6, 130.7, 133.4, 134.1, 136.1, 136.39, 136.43, 137.0,
140.7, 141.8, 150.9, 201.7. IR (NaCl, cmꢂ1) 3063w, 2961vs,
2921m, 2865m, 1641vs (ꢅC=O), 1605w, 1559m, 1521vw, 1477s,
1442vs, 1407w, 1393vw, 1377w, 1362m, 1340vw, 1281s,
1228w, 1188m, 1128s, 1100s, 1071vw, 1028m, 998vw, 946s,
902m, 870m, 823vs, 809m, 786s, 766m, 751s, 728w, 684vw,
666m, 606vw, 583w, 557vw, 521vw. MS m=z (EI) 782 (M þ 2,
99), 780 (Mþ, base). HRMS calcd for C53H49BrO, 780.2967, found
m=z 780.2944 (Mþ).
(b) Preparation of [3-Ethynyl-10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl][10-(4-t-butyl-2,6-dimethylphenyl)-9-an-
thryl]diazomethane (3-N2): To a solution of the diazomethane
3-N2(TMS) (12.7 mg, 0.015 mmol) in t-butyl alcohol (3 mL) was
added 10% aqueous solution (1 mL) of sodium hydroxide, and the
mixture was stirred overnight at room temperature. The reaction
mixture was extracted with diethyl ether; then the organic layer
was washed with water and dried over anhydrous sodium sulfate.
After evaporation of the solvent, the residue was purified by GPC
(overnight, chloroform as eluent) to afford [3-ethynyl-10-(4-t-
butyl-2,6-dimethylphenyl)-9-anthryl][10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl]diazomethane (3-N2). (11 mg, quantitatively) as
an orange viscous liquid. 1H NMR (CDCl3): ꢀ 1.45 (brs, 18H),
1.78 (brs, 12H), 3.08 (s, 1H), 7.27 (brs, 4H), 7.30–7.39 (m, 7H),
7.52–7.58 (m, 3H), 7.76 (s, 1H), 8.24 (d, J ¼ 9:74 Hz, 1H), 8.27
(b) Preparation of 10-(4-t-Butyl-2,6-dimethylphenyl)-9-
3-Trimethylsilylethynyl-10-(4-t-butyl-2,6-dimethyl-
anthryl
phenyl)-9-anthryl Ketone (3-O(TMS)): The ketone 2-O (100
mg, 0.13 mmol) and catalytic amounts of [PdCl2(Ph3P)2] and cop-
per(I) iodide were mixed with anhydrous tetrahydrofuran (1.0 mL)
under nitrogen atmosphere. To this mixture, anhydrous triethyl-
amine (0.3 mL) and trimethylsilylacetylene (0.15 mL, 0.90 mmol)
(d, J ¼ 9:16 Hz, 2H), 8.31 (d, J ¼ 8:64 Hz, 1H). IR (NaCl, cmꢂ1
)

2048 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
were added successively, and the mixture was stirred overnight at
40 ꢁC. After cooling, catalysts were removed from the reaction
mixture by filtration through a short alumina column (hexane).
After evaporation of the solvent, the residue was purified by
GPC (9 cycles, chloroform as eluent) to afford 10-(4-t-butyl-2,6-
dimethylphenyl)-9-anthryl 3-trimethylsilylethynyl-10-(4-t-butyl-
2,6-dimethylphenyl)-9-anthꢁryl ketone 3-O(TMS). (69.0 mg, 66%)
ESR Measurements. The diazo compound was dissolved in
2-methyltetrahydrofuran (10ꢂ3 M) and the solution was degassed
in a quartz cell by three freeze–degas–thaw cycles. The sample
was cooled in an optical transmission ESR cavity at 77 K and
irradiated with a Wacom 500 W Xe lamp using a Pyrex filter.
ESR spectra were measured on a JEOL JES TE 200 spectrometer
(X-band microwave unit, 100 kHz field modulation). The signal
positions were read by the use of a gaussmeter. The temperature
was controlled by a 9650 Microprocessor-based Digital Tempera-
ture Indicator/Controller, which provided the measurement accu-
racy within ꢅ0:1 K and the control ability within ꢅ0:2 K. Errors
in the measurements of component amplitudes did not exceed 5%;
the accuracy of the resonance fields determination was within
ꢅ0:5 mT.
1
as a yellow solid: mp 178 C. H NMR (CDCl3): ꢀ 0.21 (s, 9H),
1.45 (s, 9H), 1.46 (s, 9H), 1.76 (brs, 12H), 7.24–7.32 (m, 7H),
7.26 (s, 2H), 7.27 (s, 2H), 7.45 (dd, J ¼ 2:21, 6.43 Hz, 1H),
7.52–7.55 (m, 2H), 7.69 (brs, 1H), 8.23–8.26 (m, 4H). 13C NMR
(CDCl3): ꢀ 20.2, 20.3, 31.5, 34.51, 34.53, 95.9, 105.5, 120.1,
124.6, 124.7, 125.4, 125.5, 125.6, 126.0, 126.9, 127.4, 127.7,
128.98, 129.03, 129.6, 129.9, 130.0, 130.1, 130.27, 130.31,
133.8, 134.2, 136.2, 136.6, 136.7, 141.6, 141.7, 150.7, 150.9,
202.0. IR (NaCl, cmꢂ1) 3065w, 2963vs, 2867m, 2151m (ꢅCꢄC),
1637vs (ꢅC=O), 1560m, 1477m, 1451m, 1442m, 1362m, 1282m,
1250m, 1229w, 1185m, 1128s, 1098s, 1028w, 988w, 945m,
910m, 859s, 843s, 830w, 808m, 787m, 765m, 750m, 733s,
664w. MS m=z (relative intensity) 800 (M þ 2, 25.7), 799
(M þ 1, 64.5), 798 (Mþ, 100); HRMS calcd for C58H58OSi
798.4257, found m=z 798.4275.
(c) Preparation of 3-Ethynyl-10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl 10-(4-t-Butyl-2,6-dimethylphenyl)-9-anthryl
Ketone (3-O): To a solution of the ketone 3-O(TMS) (69.0
mg, 0.086 mmol) in t-butyl alcohol (12 mL) was added 10% aque-
ous solution (4 mL) of sodium hydroxide, and the mixture was
stirred overnight at room temperature. The reaction mixture was
extracted with diethyl ether, and organic layer was washed with
water and dried over anhydrous sodium sulfate. After evaporation
of the solvent, the residue was purified by GPC (7 cycles, chloro-
form as eluent) to afford 3-ethynyl-10-(4-t-butyl-2,6-dimethyl-
phenyl)-9-anthryl 10-(4-t-butyl-2,6-dimethylphenyl)-9-anthryl ke-
tone 3-O (62 mg, quantitatively) as a yellow viscous liquid.
1H NMR (CDCl3): ꢀ 1.45 (brs, 18H), 1.76 (brs, 12H), 3.08 (s, 1H),
7.26–7.31 (m, 11H), 7.48–7.55 (m, 3H), 7.73 (d, J ¼ 1:47 Hz,
1H), 8.22–8.26 (m, 4H). IR (NaCl, cmꢂ1) 1637vs (ꢅC=O).
Low-Temperature UV–Vis Spectra. Low-temperature spec-
tra at 77 K were obtained by using an Oxford variable-temperature
liquid-nitrogen cryostat (DN 1704) equipped with a quartz outer
window and a sapphire inner window. The sample was dissolved
in dry 2-MTHF, placed in a long-necked quartz cuvette of 1-mm
path length, and degassed thoroughly by repeated freeze–degas–
thaw cycles at a pressure near 10ꢂ5 Torr. The cuvette was flame-
sealed under reduced pressure, placed in the cryostat, and cooled
to 77 K. The sample was irradiated for several minutes in the spec-
trometer with a Halos 300-W high-pressure mercury lamp using a
Pyrex filter, and the spectral changes were recorded at appropriate
time intervals. The spectral changes upon thawing were also moni-
tored by carefully controlling the matrix temperature with an
Oxford Instrument Intelligent Temperature Controller (ITC 4).
Flash Photolysis. All flash measurements were made on a
Unisoku TSP-601 flash spectrometer. Three excitation light
sources were used, depending on the precursor absorption bands
and lifetime of the transient species. They were (i) a cylindrical
150-W Xe flash lamp (100 J/flash with 10-ms pulse duration);
(ii) a Quanta-Ray GCR-11 Nd:YAG laser (355 nm pulses of up
to 40 mJ/pulse and 5–6-ns duration; 266 nm pulses of up to 30
mJ/pulse and 4–5-ns duration); and (iii) Lamda Physik LEXTRA
XeCl excimer laser (308 nm pulses of up to 200 mJ/pulse and
17-ns duration). The beam shape and size were controlled by a
focal length cylindrical lens.
A Hamamatsu 150-W xenon short arc lamp (L 2195) was used
as the probe source, and the monitoring beam, guided using an
optical fiber scope, was arranged in an orientation perpendicular
to the excitation source. The probe beam was monitored with a
Hamamatsu R2949 photomultiplier tube through a Hamamatsu
S3701-512Q MOS linear image sensor (512 photodiodes used).
Timing of the excitation pulse, the probe beam, and the detection
system was achieved through an Iwatsu Model DS-8631 digital
synchro scope which was interfaced to a NEC 9801 RX2 comput-
er. This allowed for rapid processing and storage of the data and
provided printed graphic capabilities. Each trace was also dis-
played on a NEC CRT N5913U monitor.
(d) Preparation of Tris(ketone) (4-3O): The ketone 3-O
(55.2 mg, 0.076 mmol), 1,3,5-triiodobenzene (10.0 mg, 0.025
mmol) and catalytic amounts of [PdCl2(Ph3P)2] and copper(I)
iodide were mixed with anhydrous tetrahydrofuran (0.8 mL) under
nitrogen atmosphere. To this mixture, anhydrous triethylamine
(0.3 mL) was added and the mixture was stirred overnight at room
temperature. Removal of the catalyst by filtration through a short
alumina column, followed by removal of the solvent, gave crude
material which was purified by GPC (8 cycles, chloroform as elu-
ent) to give tri(ketone) compound 4-3O (26.0 mg, 46%) as a yel-
ꢁ
1
low solid: mp 279 C. H NMR (CDCl3): ꢀ 1.45 (brs, 54H), 1.77
(brs, 36H), 7.25–7.33 (m, 33H), 7.47 (dd, J ¼ 2:85, 7.07 Hz, 3H),
7.55 (s, 3H), 7.56 (dd, J ¼ 3:12, 7.12 Hz, 6H), 7.71 (d, J ¼ 1:10
Hz, 3H), 8.23–8.28 (m, 12H). 13C NMR (CDCl3): ꢀ 20.3, 31.5,
34.5, 88.9, 91.1, 119.7, 123.8, 124.6, 124.8, 125.4, 125.6, 125.8,
126.0, 126.9, 127.4, 127.8, 129.0, 129.4, 129.6, 129.9, 130.2,
130.4, 133.7, 134.2, 136.2, 136.5, 136.7, 141.7, 150.8, 201.9. IR
(NaCl, cmꢂ1) 3062w, 3012w, 2965vs, 2916m, 2866w, 2211w
(ꢅCꢄC), 1641vs (ꢅC=C), 1607m, 1579m, 1559m, 1521w, 1478m,
1442m, 1407w, 1378w, 1362m, 1282s, 1216m, 1190m, 1129s,
1098s, 1028w, 1000w, 945m, 869m, 827s, 813w, 787s, 757vs,
668w. MS m=z (relative intensity) 2255 (M þ 4, 33), 2254
(M þ 3, 57), 2253 (M þ 2, 85), 2252 (M þ 1, 100), 2251 (Mþ,
A sample was placed in a long-necked Pyrex tube which had a
side arm connected to a quartz fluorescence cuvette and degassed
using a minimum of four freeze–degas–thaw cycles at a pressure
near 10ꢂ5 Torr immediately prior to being flashed. The sample
system was flame-sealed under reduced pressure, and the solution
was transferred to the quartz cuvette, which was placed in the
sample chamber of the flash spectrometer. A cell holder block
of the sample chamber was equipped with a thermostat and
allowed to come to thermal equilibrium. The concentration of
the sample was adjusted so that it absorbed a significant portion
of the excitation light.
62); HRMS calcd for
2251.1432.
C171H150O3 2251.1580, found m=z

Y. Tsuchiya et al.
Bull. Chem. Soc. Jpn., 78, No. 11 (2005) 2049
SQUID Measurements. Magnetic susceptibility data were
obtained on a quantum design MPMS-2A superconducting quan-
tum interference device (SQUID) magnetometer/susceptometer.
Irradiation with light from an argon ion laser (488 nm, Omni-
chrome 543-150BS) through a flexible optical fiber which passes
through the inside of the SQUID sample holder was performed
inside the sample room of SQUID apparatus at 5–11 K. One
end of the optical fiber was located 40 mm above the sample cell
(capsule) and the other was attached to a coupler for the laser. The
bottom part of the capsule (6 ꢃ 10 mm2) without a cap was used
as a sample cell. A 50 mL portion of the sample solution (0.5 mM)
in 2-MTHF was placed in the cell which was held by a straw. The
irradiation was carried out until there is no further change of mag-
netization monitored at 5 K in a constant field of 5 kOe. The mag-
netization, Mb and Ma, before and after irradiation was measured
at 5 K in a field range 0–50 kOe. The plots of the magnetization
[M ¼ ðMa ꢂ MbÞ: F stands for a photolysis factor of diazo com-
pound] versus the magnetic field were analyzed in terms of
Brillouin functions.
K. Inoue, and H. Iwamura, Angew. Chem., Int. Ed. Engl., 32, 872
(1993). c) K. Matsuda, N. Nakamura, K. Takahashi, K. Inoue, N.
Koga, and H. Iwamura, J. Am. Chem. Soc., 117, 5550 (1995).
d) K. Matsuda, N. Nakamura, K. Inoue, N. Koga, and H. Iwamura,
Chem.—Eur. J., 2, 259 (1996). e) K. Matsuda, N. Nakamura, K.
Inoue, N. Koga, and H. Iwamura, Bull. Chem. Soc. Jpn., 69,
1483 (1996).
5
M. Regitz and G. Maas, ‘‘Diazo Compounds-Properties
and Synthesis,’’ Academic Press, Orland (1986).
a) H. Tomioka, Acc. Chem. Res., 30, 315 (1997). b) H.
6
Tomioka, ‘‘Advances in Carbene Chemistry,’’ ed by U. Brinker,
JAI Press, Greenwich, CT (1998), Vol. 2, pp. 175–214. c) H.
Tomioka, ‘‘Carbene Chemistry,’’ ed by G. Bertrand, Fontis Media,
Lansanne (2002), pp. 103–152.
7
H. Tomioka, M. Hattori, K. Hirai, K. Sato, D. Shiomi, T.
Takui, and K. Itoh, J. Am. Chem. Soc., 120, 1106 (1998).
T. Itoh, K. Hirai, and H. Tomioka, J. Am. Chem. Soc., 126,
1130 (2004).
8
9 Y. Ohtsuka, T. Itoh, K. Hirai, H. Tomioka, and T. Takui,
Org. Lett., 6, 847 (2004).
10 T. Maemura, Y. Ohtsuka, H. Wildt, K. Hirai, and H.
Tomioka, Eur. J. Org. Chem., 2004, 2991.
The authors are grateful to the Ministry of Education, Cul-
ture, Sports, Science and Technology of Japan for support of
this work through a Grant-in-Aid for Scientific Research for
Specially Promoted Research (No. 12002007). The supports
from the Mitsubishi Foundation and the Nagase Science and
Technology Foundation are also appreciated.
11 a) H. Tomioka, J. Nakijima, H. Mizuno, E. Iiba, and K.
Hirai, Can. J. Chem., 77, 1066 (1999). b) H. Itakura and H.
Tomioka, Org. Lett., 2, 2995 (2000). c) Y. Takahashi, M. Tomura,
K. Yoshida, S. Murata, and H. Tomioka, Angew. Chem., Int. Ed.,
39, 3478 (2000).
Supporting Information
12 a) H. Tomioka, E. Iwamoto, H. Itakura, and K. Hirai,
Nature, 412, 626 (2001). b) E. Iwamoto, K. Hirai, and H.
Tomioka, J. Am. Chem. Soc., 125, 14664 (2003).
13 D. T. Hurst and A. G. McInnes, Can. J. Chem., 43, 2004
(1965).
14 T. Sato, Bull. Chem. Soc. Jpn., 44, 2848 (1971).
15 K. Sonogashira, ‘‘Comprehensive Organic Synthesis,’’ ed
by B. M. Trost and I. Fleming, Pergamon Press, Oxford, U. K.
(1991), Vol. 3, pp. 521–549.
MALDI-TOF MS spectra of the tris(ketone) (Fig. S1); GPC
chart of tris(diazo) and tris(ketone) compounds (Fig. S2); Simulat-
ed ESR spectra of tris(carbene) (Fig. S3); MALDI-TOF MS spec-
tra of the spent solution of LFP of mono(diazo) and tris(diazo)
compounds (Figs. S4 and S5); ESR spectra obtained after irradia-
tion of tris(diazo) compound (Fig. S6). This material is available
free of charge on the web at http://www.csj.jp/journals/bcsj/.
16 For reviews of general reactions of carbenes, see: a) W.
Kirmse, ‘‘Carbene Chemistry,’’ 2nd ed, Academic Press, New
York (1971). b) ‘‘Carbenes,’’ ed by R. A. Moss and M. Jones, Jr.,
Wiley, New York (1973 and 1975), Vols. 1 and 2. c) ‘‘Carbene-
(oide), Carbine,’’ ed by M. Regitz, Thieme, Stuttgart (1989).
d) C. Wentrup, ‘‘Reactive Intermediates,’’ Wiley, New York
(1984), pp. 162–264.
17 See for reviews of the ESR spectra of triplet carbenes:
a) W. Sander, G. Bucher, and S. Wierlacher, Chem. Rev., 93,
1583 (1993). b) A. M. Trozzolo and E. Wasserman, ‘‘Carbenes,’’
ed by M. Jones, Jr. and R. A. Moss, Wiley, New York (1975),
Vol. 2, pp. 185–206.
18 a) M. S. Platz, ‘‘Diradicals,’’ ed by W. T. Borden, Wiley,
New York (1982), pp. 195–258. b) E. Wasserman and R. S.
Hutton, Acc. Chem. Res., 10, 27 (1977). c) R. Breslow, H. W.
Chang, and E. Wasserman, J. Am. Chem. Soc., 89, 1112 (1967).
19 a) H. Tukada, T. Sugawara, S. Murata, and H. Iwamura,
Tetrahedron Lett., 27, 235 (1986). b) A. S. Nazran, F. J. Gabe,
Y. LePage, D. J. Northcott, J. M. Park, and D. Griller, J. Am.
Chem. Soc., 105, 2912 (1983). c) A. S. Nazran and D. Griller,
J. Chem. Soc., Chem. Commun., 1983, 850. d) B. C. Gilbert,
D. Griller, and A. S. Nazran, J. Org. Chem., 50, 4738 (1985).
e) A. S. Nazran, F. L. Lee, Y. LePage, D. J. Northcott, J. M. Park,
and D. Griller, J. Phys. Chem., 88, 5251 (1984). f) H. Tomioka,
T. Watanabe, K. Hirai, K. Furukawa, T. Takui, and K. Itoh,
J. Am. Chem. Soc., 117, 6376 (1995).
References
1
a) H. Iwamura, Adv. Phys. Org. Chem., 26, 179 (1990).
b) D. A. Dougherty, Acc. Chem. Res., 24, 88 (1991). c) A. Rajca,
Chem. Rev., 94, 871 (1994).
2
a) O. Kahn, ‘‘Molecular Magnetism,’’ VCH Publishers Inc.,
Weinheim (1993). b) D. Gatteschi, Adv. Mater., 6, 635 (1994).
c) J. S. Miller and A. J. Epstein, Angew. Chem., Int. Ed. Engl.,
33, 385 (1994). d) J. S. Miller and A. J. Epstein, Chem. Eng. News,
73, 30 (1995). e) ‘‘Magnetism: A Supramolecular Function,’’ ed
by O. Kahn, NATO ASI Series C, Kluwer, Dordrecht (1996).
f) ‘‘Molecule-Based Magnetic Materials,’’ ed by M. M. Turnbull,
T. Sugimoto, and L. K. Thompson, ACS Symposium Series 644,
Washington, ACS (1996). g) D. Gatteschi, Curr. Opin. Solid State
Mater. Sci., 1, 192 (1996). h) J. S. Miller and A. J. Epstein, MRS
Bull., 25, 21 (2000). i) ‘‘Magnetism: Molecules to Materials II,’’
ed by J. S. Miller and M. Drillon, Wiley-VCH (2001). j) ‘‘Molecu-
lar Magnetism,’’ ed by K. Itoh and M. Kinoshita, Kodansha-
Gordon and Breach, Tokyo (2000).
3
a) N. Koga and H. Iwamura, ‘‘Carbene Chemistry,’’ ed by
G. Bertrand, Fontis Media, Lausanne (2002), pp. 271–296.
b) K. Matsuda, N. Nakamura, K. Takahashi, K. Inoue, N. Koga,
and H. Iwamura, ‘‘Molecule-Based Magnetic Materials,’’ ACS
Symposium Series 644, Washington, ACS (1996), p. 142.
4
a) N. Nakamura, K. Inoue, H. Iwamura, T. Fujioka, and
Y. Sawaki, J. Am. Chem. Soc., 114, 1484 (1992). b) N. Nakamura,

2050 Bull. Chem. Soc. Jpn., 78, No. 11 (2005)
Tris(carbene) Constructed with Stable Carbene
20 See for review: H. Tomioka, ‘‘Advances in Strained and
Interesting Organic Molecules,’’ ed by B. Halton, JAI Press,
Greenwich, CT (2000), Vol. 8, pp. 83–112.
21 a) Y. Teki, Thesis, Osaka City University, Osaka, Japan
(1985). b) Y. Teki, K. Takui, H. Yagi, K. Itoh, and H. Iwamura,
J. Chem. Phys., 83, 539 (1985). c) K. Takui, ‘‘Molecular Magnet-
ism in Organic-Based Materials,’’ ed by P. M. Lahti, Marcel
Dekker, New York (1999).
4587 (1977).
29 a) W. T. Borden, H. Iwamura, and J. A. Berson, Acc.
Chem. Res., 27, 109 (1994). b) W. T. Borden, ‘‘Diradicals,’’ ed
by W. T. Borden, Wiley-Interscience, New York (1982), pp. 1–12.
30 a) S. Murata and H. Iwamura, J. Am. Chem. Soc., 113,
5547 (1991). b) M. Matsumoto, T. Ishida, N. Koga, and H.
Iwamura, J. Am. Chem. Soc., 114, 9952 (1992). c) T. Doi, A. S.
Ichimura, N. Koga, and H. Iwamura, J. Am. Chem. Soc., 115,
8928 (1993).
31 a) C. Ling, M. Minato, P. M. Lahti, and H. van Willigen,
J. Am. Chem. Soc., 114, 9959 (1992). b) C. Ling and P. M. Lahti,
J. Am. Chem. Soc., 116, 8784 (1994).
32 K. Itoh, T. Takui, Y. Teki, K. Sato, D. Shiomi, T.
Kinoshita, T. Nakamura, T. Momose, T. Shida, T. Okuno, A.
Izuoka, T. Sugawara, T. Kaneko, E. Tsuchida, and H. Nishide,
Mol. Cryst. Liq. Cryst., 271, 129 (1995).
33 a) E. Wasserman, V. J. Kuck, W. A. Yager, R. S. Hutton,
F. D. Greene, V. P. Abegg, and N. M. Weinshenker, J. Am. Chem.
Soc., 93, 6355 (1971). b) D. J. Astles, M. Girard, D. Griller, R. J.
Kolt, and D. D. M. Wayner, J. Org. Chem., 53, 6053 (1988).
34 Y. Sano, M. Tanaka, N. Koga, K. Matsuda, H. Iwamura,
P. Rabu, and M. Drillon, J. Am. Chem. Soc., 119, 8246 (1997).
35 M. Tashiro and T. Yamato, J. Chem. Soc., Perkin Trans. 1,
1979, 176.
22 R. L. Carlin, ‘‘Magnetochemistry,’’ Springer-Verlag, Ber-
lin (1986).
23 Ms values obtained from the best fitted curves were used.
24 For reviews of laser flash photolysis study of carbenes, see:
a) R. S. Moss and N. J. Turro, ‘‘Kinetics and Spectroscopy of Car-
benes and Biradicals,’’ ed by M. S. Platz, Plenum Press, New York
(1990), pp. 213–238. b) M. S. Platz and V. M. Maloney, ‘‘Kinetics
and Spectroscopy of Carbenes and Biradicals,’’ ed by M. S. Platz,
Plenum Press, New York (1990), pp. 239–352. c) R. A. Moss,
‘‘Advances in Carbene Chemistry,’’ ed by U. H. Brinker, JAI
Press, Greenwich (1994), pp. 59–88. d) J. E. Jackson and M. S.
Platz, ‘‘Advances in Carbene Chemistry,’’ ed by U. H. Brinker,
JAI Press, Greenwich (1994), pp. 89–160.
25 Main decay pathway of those sterically congested diaryl-
carbenes in benzene is coupling at the carbenic center. However
as the carbenic center is more crowded, coupling at the aromatic
rings where unpaired electrons are delocalized becomes more
important process, see: Refs. 6, 11, and 12.
36 Y. Ogata, Y. Kosugi, and N. Nate, Tetrahedron, 27, 2705
(1971).
26 A. A. Ovchinnikov, Theor. Chim. Acta, 47, 297 (1978).
27 H. C. Longuet-Higgins, J. Chem. Phys., 18, 256 (1950).
28 W. T. Borden and E. R. Davidson, J. Am. Chem. Soc., 99,
37 E. D. Bergmann and E. Loewenthal, Bull. Soc. Chim. Fr.,
1952, 66.