A redox-active, amphoteric pyrogallolaldehyde derivative: Electrochemical characterization and schiff base formation for constructing multifunctional salphen complexes

Article abstract of DOI:10.1246/bcsj.20120316

We synthesized and determined the crystal structure of a novel pyrogallolaldehyde derivative with a 5-dimethylaminomethyl group. In solution, this compound underwent autoxidation to form unidentified blue species under ambient conditions. Electrochemical measurements revealed that addition of a base promoted the autoxidation, whereas addition of an acid effectively suppressed it. Using p-toluenesulfonic acid, we isolated a hydroxy-rich salphen-type Schiff base ligand as an ammonium salt of the acid. We obtained several transition-metal complexes of this ligand that retained its antioxidant stability. We discuss the spectroscopic and electrochemical properties of these complexes in comparison with analogous complexes without a 5-dimethylaminomethyl group.

Full text of DOI:10.1246/bcsj.20120316

698
Bull. Chem. Soc. Jpn. Vol. 86, No. 6, 698-706 (2013)
© 2013 The Chemical Society of Japan
Selected Papers
A Redox-Active, Amphoteric Pyrogallolaldehyde Derivative:
Electrochemical Characterization and Schiff Base Formation
for Constructing Multifunctional Salphen Complexes
Hajime Shingai, Hirohiko Houjou,* Isao Yoshikawa, and Koji Araki
Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505
Received November 13, 2012; E-mail: houjou@iis.u-tokyo.ac.jp
We synthesized and determined the crystal structure of a novel pyrogallolaldehyde derivative with a 5-dimethyl-
aminomethyl group. In solution, this compound underwent autoxidation to form unidentified blue species under ambient
conditions. Electrochemical measurements revealed that addition of a base promoted the autoxidation, whereas addition
of an acid effectively suppressed it. Using p-toluenesulfonic acid, we isolated a hydroxy-rich salphen-type Schiff base
ligand as an ammonium salt of the acid. We obtained several transition-metal complexes of this ligand that retained its
antioxidant stability. We discuss the spectroscopic and electrochemical properties of these complexes in comparison with
analogous complexes without a 5-dimethylaminomethyl group.
The electrochemistry of phenoxy-bearing compounds is of
tance of complexes with redox-active ligand(s), developing
a new component molecule for constructing multinuclear,
stimulus-responsive materials is a worthwhile challenge. For
this purpose, the coordination chemistry and redox activity of
the multifunctional ligand must be controlled.
To design a new redox-active pyrogallol derivative as a
component of salen/salphen ligands, we noted an interesting
concerted proton-electron-transfer protocol, by which the
resultant phenoxyl radical would be stabilized with a proximal
ammonium/iminium hydrogen.29-31 In this study, we syn-
thesized a pyrogallolaldehyde bearing an ammonium hydrogen
in the proximity of its phenol group as a prototype multi-
functional ligand, and here we describe its electrochemical
characterization, Schiff base formation, and subsequent metal-
complexation.
considerable importance for understanding biological catalytic
systems like ribonucleotide reductase,1 galactose oxidase,2 and
glyoxal oxidase.3 The phenoxyl radical and its complexes with
metals play an important role in the action of these enzymes,
and numerous complexes have been proposed as models for
the enzyme active sites.4-9 Those model complexes exhibit
interesting electrochemical phenomenon of valence tautomer-
ism, or redox isomerism,10-13 by virtue of a fascinating inter-
play between ligand-centered and metal-centered radical states.
The complexes with redox-active ligands have also received
attention because of their potential applications for nonlinear
optical materials,14 electroconductive materials,15 and chromo-
tropic materials,16 the properties of which can be controlled by
various external stimuli. Because the stability of the oxidized
species depends on the electron-donating character of ligands,
an hydroxy-rich aromatic core like pyrogallol is a good
candidate for the component of redox-active ligands.
Results and Discussion
Structure and Properties of the Aldehyde 1H3.
A
Recently, transition-metal complexes of Schiff base ligands,
including salen (salenH2: N,N¤-disalicylideneethylenediamine)
and salphen (salphenH2: N,N¤-disalicylidene-o-phenylenedi-
amine), have been extensively studied with a focus on the
above-mentioned properties related to valence tautomerism.17-20
Also, there have been several attempts to take advantage of the
radical-generating ability of hydroxy-rich salicylaldiminato
complexes to develop artificial nucleases.21,22 Hydroxy-rich
salens or salphens would also provide access to various
metallosupramolecular systems and metal-containing materi-
als.23-25 Several macrocyclic complexes composed of pyro-
catechualdehyde (2,3-dihydroxybenzaldehyde) and pyrogallol-
aldehyde (2,3,4-trihydroxybenzaldehyde) have been reported to
capture metal ions or guest molecules, resulting in fascinating
multicomponent systems.26-28 In view of the potential impor-
Mannich reaction of pyrogallolaldehyde afforded the corre-
sponding 5-(N,N¤-dimethylaminomethyl) derivative 1H3 in
good yield (Scheme 1). The 1-formyl group of 1H3 is relatively
inert, so undesired polymerization did not occur. Compound
1H3 was recrystallized from a THF-water system to give a
single crystal of X-ray quality. Interestingly, the melting point
CHO
OH
CHO
OH
Me2NH, (CH2O)n
MeOH-H2O
N
HO
HO
OH
OH
1H3
Scheme 1. Preparation of 2,3,4-trihydroxy-5-dimethylamino-
methylbenzaldehyde (1H3).
Published on the web June 15, 2013; doi:10.1246/bcsj.20120316

H. Shingai et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
699
Table 1. Selected Bond Lengtha) (¡) and HOMA Indexes
Derived from the Crystal Structures of 1H3 and Pyrogallol-
aldehyde
H
H
N
H
O
N
O
Pyrogallolaldehydeb)
O
OH
HO
OH
Bond
Unit 1
Unit 2
OH
OH
C1-C2
C2-C3
C3-C4
C4-C5
C5-C6
C6-C1
C1-C7
C2-O1
C3-O2
C4-O3
HOMAc)
1.417(4)
1.379(4)
1.424(4)
1.435(4)
1.373(4)
1.400(4)
1.430(4)
1.362(4)
1.368(3)
1.283(3)
0.790
1.406(4)
1.373(4)
1.432(4)
1.434(4)
1.369(4)
1.410(4)
1.426(5)
1.360(4)
1.371(3)
1.284(3)
0.767
1.418
1.375
1.399
1.396
1.373
1.405
1.426
1.355
1.368
zw
1H3
1H3
-e-
H
H
N
H
O
N
H
O
O
OH
O
OH
OH
OH
1.356
ox
1H3
0.905, 0.938
a) Standard deviations in parentheses. b) From Ref. 32. Bond
lengths are averaged over two asymmetric units in the crystal.
P
n
2
¡
n
c) HOMA ¼ 1 ꢀ
i¼1ðRopt ꢀ RiÞ , where ¡ = 257.7, and
Further oxidized product
R
opt = 1.388, as defined in Ref. 33.
Scheme 2. A presumptive reaction scheme for the autoxi-
dization of 1H3.
tionally, introduction of the 5-dimethylaminomethyl group
lengthened the C3-C4 distances (1.424 and 1.432 ¡) and the
C4-C5 distances (1.435 and 1.434 ¡) as compared to the
corresponding distances (1.399 and 1.396 ¡, respectively) in
pyrogallolaldehyde. We evaluated the overall distortion of
the benzene ring by means of the harmonic oscillator model
for aromaticity (HOMA) index,33 a geometry-based criterion
of aromaticity. The relatively low HOMA values (0.790 and
0.767) indicated a loss of aromaticity, suggesting that the C4-
O3 bond of 1H3 assumed keto-like character;34 therefore,
zw
the negative charge was delocalized over the benzene ring.
Mulliken population analysis (by Hartree-Fock/6-311G**
level of calculation) showed that on going from 1H3 to
1H3zw, the net charge at 4-O(H) group changes by ¹0.556e,
while that at 5-CH2N(H)(CH3)2 group changes by +0.689e.
The difference negative charge (¹0.134e) was found dis-
tributed over the benzene ring, especially at the carbon C5
(¹0.110e), and the formyl group (¹0.048e) (for detail,
Table S1). These changes in electronic distribution are reason-
ably in agreement with prediction from resonance structures.
As for packing structure, the asymmetric units formed a cyclic
tetramer in which the protonated amino groups, 3-OH groups,
and 4-O groups were connected through multiple hydrogen
bonds (Figure 2). A water molecule, though it was consid-
erably disordered, was found at every site where one cyclic
tetramer was connected with another (Figure S2, Supporting
Information).
The crystal of 1H3 was colorless to pale yellow, but a
DMF solution of the crystal rapidly turned blue under atmos-
pheric conditions. The UV-vis spectra of 1H3 in DMF showed
a broad absorption band around 600-700 nm (Figure 3). The
absorbance at 600 nm initially increased with time but began to
diminish gradually after 90 min (Figure 3, inset). In contrast,
the absorbance at 400 nm increased monotonically, and as
a result, the solution gradually turned yellow. Interestingly,
an aqueous solution of 1H3 was yellow from the beginning,
whereas a THF solution was almost colorless and remained
so throughout the time course of the experiment. Taking into
Figure 1. ORTEP drawing of 1H3. One of two asymmetric
units is shown, and a water molecule is omitted for clarity.
The atoms are numbered for comparison with the structural
data listed in Table 1.
of the crystalline sample (159-160 °C) was higher than that of
the as-prepared sample (146-147 °C) even though the 1H NMR
spectra of the two samples were identical to each other.
Furthermore, the IR spectrum of the recrystallized sample was
much different from that of the as-prepared sample (Figure S1,
Supporting Information), probably owing to differences in
hydrogen bonding and dissociation state. Analysis of the
crystal structure suggested that the proton of the 4-hydroxy
group was dissociated and that the nitrogen atom was proton-
ated (hereafter referred to as 1H3zw, also shown in Scheme 2,
vide infra). Figure 1 shows the ORTEP drawing of one of the
two independent asymmetric units, and Table 1 lists selected
bond lengths. In both asymmetric units, the C4-O3 distances
(1.283 and 1.284 ¡) were markedly shorter than the corre-
sponding distance (1.356 ¡) in pyrogallolaldehyde.32 Addi-

700
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
Redox-Active Amphoteric Pyrogallolaldehyde
(a)
These spectroscopic results imply that 1H3 easily underwent
autoxidation in solution to give the blue species. We confirmed
that the characteristic absorption at 600-700 nm was intensified
associatively with bulk electrolysis or addition of appropriate
oxidation reagents (Figures S4a and S4b). Although the spec-
tral features imply the presence of a phenoxy radical,7c,29,35,36
we could not detect any significant ESR signals after the
efforts of preparation under various conditions (e.g., in aceto-
nitrile with ferrocenium hexafluorophosphate at 77 K, 210K)
(Figure S4c). For the moment, we have not been able to
identify the blue species: while one candidate is a semiquinone
radical that is rapidly decomposed or undergoes further oxida-
tion to form a product such as an imine or orthoquinone,37 there
is some room for discussion about charge-transfer complexes
or some spin-coupled diamagnetic species. The formation of
the blue species was accelerated appreciably by the addition of
one equivalent of triethylamine, whereas the initial increase of
the absorption at 600 nm was suppressed by the addition of
p-toluene sulfonic acid (Figure S3, Supporting Information).
These results indicate that the progress of these reactions
was closely coupled with the association and dissociation of
protons. The proposed reaction scheme is given in Scheme 2.
Among the 1H NMR signals of 1H3 (in DMSO-d6), we
particularly noted three singlets (3.75, 6.92, and 9.59 ppm),
which were assigned to methylene, aromatic, and formyl
protons, respectively; these peaks were shifted to 4.21, 7.36,
and 9.86 ppm, respectively, by the addition of p-toluenesul-
fonic acid (Figure S5, Supporting Information). The substantial
change in the chemical shift of the methylene protons indi-
cates protonation of the -N(CH3)2 group, and the downfield
shift of the protons in the aromatic region indicates suppres-
sion of deprotonation of the 4-OH group. In contrast, the
addition of triethylamine resulted in only minor changes in
the chemical shifts of these protons (to 3.70, 6.89, and 9.55
ppm) relative to the shifts under neutral conditions. The chem-
ical shifts measured in D2O were 4.14, 7.07, and 9.59 ppm,
implying the protonation of the amino group, even taking
solvent difference into account. These results suggest that in
a polar solvent, 1H3 was in equilibrium with its zwitterionic
H
(b)
H
O
O
H
O
H
N
H
H
O
N
O
O
O
O
O
H
N
H
H
O
H
O
O
O
H
H
O
H
N
H
O
O
H
H
Figure 2. Hydrogen-bonding network in the cyclic tetramer
of 1H3zw; (a) stick model representation of the crystal
structure, and (b) drawing for showing topological relation.
1.2
ε
@400 nm
0.3
0.2
0.1
0.0
1.0
0.8
0.6
0.4
0.2
0.0
ε
@600 nm
zw
form 1H3 (Scheme 2).
The cyclic voltammogram of 1H3 in DMF exhibited several
irreversible peaks in the region of ¹1000 to +1000 mV (vs.
Ag+/Ag) (Figure 4). The intense anodic peak at 620 mV was
similar to that observed in the voltammogram of pyrogallol-
aldehyde (Figure S6). As is the case for other phenols, the
oxidation corresponding to this peak was irreversible, probably
owing to deprotonation of the oxidized form.29 The deproto-
nated form of pyrogallolaldehyde was reduced at 0 and ¹350
mV, but the corresponding peak (¹330 mV) for 1H3 was negli-
gibly small. Instead, a new cathodic peak and a new anodic
peak appeared at ¹820 mV and around ¹140 mV, respectively,
both of which had a similar current magnitude. Upon step-
wise addition of triethylamine, the anodic peak at ¹140 mV
developed (Figure 4). A similar phenomenon was observed for
pyrogallolaldehyde (Figure S6). The anodic peak grew with
appearance of a new peak at 350 mV, becoming saturated after
the addition of 2 equiv of amine (see inset). This result suggests
that this peak overlapped with the oxidation waves of the 4-
0
500
1000
1500
Time elapsed/min
ε
300
400
500
600
700
/ nm
800
900 1000
λ
Figure 3. UV-vis spectra of 1H3 in DMF (solid line), water
(dotted line), and THF (dashed line). The inset shows
the time dependence of the absorptivity (¾) of the DMF
solution.
account the results for some solvents (DMSO, acetonitrile,
methanol, and ethanol), we can summarize the solvent-
dependence that aprotic polar solvents promote the formation
of the blue species, while aquaous environment lead to another
path of reaction.
zw
dissociated form 1H3 and the 3,4-doubly-dissociated form

H. Shingai et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
Current (μA) @-140mV
701
100
30
20
10
0
Current (μA) @-140mV
200
100
0
200
100
0
80
60
40
20
0
2.00 equiv
1.50 equiv
1.00 equiv
0.50 equiv
0.00 equiv
0.00 equiv
0.50 equiv
1.00 equiv
0.0 1.0 2.0 3.0
0.0 1.0 2.0 3.0
[Et3N]/[1H3]
[TsOH]/[1H3]
100 mV s−1
100 mV s−1
-1500
-1000
-500
0
500
1000
-1500
-1000
-500
0
500
1000
Potential/mV (vs. Ag+/Ag)
Potential/mV (vs. Ag+/Ag)
Figure 5. Cyclic voltammograms of 1H3 with addition of
various amounts of p-toluenesulfonic acid.
Figure 4. Cyclic voltammograms of 1H3 with stepwise
addition of triethylamine.
of 1H3ox at this potential. Along with this change, the cathodic
peak at ¹820 mV gradually diminished, and another cathodic
peak developed at ¹360 mV. Like the cathodic peak at ¹350
mV of pyrogallolaldehyde, the latter peak appears to corre-
spond to reduction of the deprotonated semiquinone (or its
proton-transferred form). As a result, the profile of the cyclic
voltammogram of 1H3 became quite similar to that of
pyrogallolaldehyde when at least 1 equiv of acid was added.
On the basis of the above-mentioned results, we assumed
the proton-coupled redox scheme for 1H3 shown in Scheme 3.
Note that the structures of deprotonated forms have not been
unambiguously determined, because there are several pos-
sible dissociation states of the 2-OH, 3-OH, 4-OH, and 5-
CH2NH+(CH3)2 groups. At this stage, we can safely conclude
that under neutral to basic conditions, 1H3 underwent facile
oxidation to 1H3ox. The oxidized species underwent further
oxidation, coupled with proton dissociation, resulting in
considerable irreversibility in the cyclic voltammogram. The
electrochemical measurements indicate that the addition of
1 equiv of acid markedly stabilized 1H3 against oxidation;
these measurements were in agreement with the UV-vis results.
Scheme 3 was also validated by means of a series of cyclic
voltammetry experiments in selected scan ranges (Figure S7,
Supporting Information). Unless the potential exceeded +200
mV, the reduction peak at ¹360 mV did not appear. These
experiments also confirmed that the oxidation peak at +40
mV and the reduction peak at ¹820 mV were coupled with
each other.
CHO
OH
CHO
OH
-e-
N
H
N
H
HO
HO
(+0.62 V)
OH
OH
+
ox
1H4
1H4
+H+ -H+
+H+ -H+
CHO
OH
CHO
OH
-e-
N
N
HO
1H3
HO
(+0.62 V)
OH
OH
ox
1H3
CHO
OH
CHO
OH
-e-
N
H
N
H
O
O
(-0.14 V)
OH
OH
zw
ox
1H3
1H3
-
H
, -e
+H+ -H+
+H+ -H+
CHO
OH
CHO
OH
+e-
N
H
N
H
(-0.82 V)
O
O
O
O
ox
-
1H2
1H2
Schiff Base Ligand and Its Complexes. The reaction
of pyrogallolaldehyde with 4,5-dimethylphenylenediamine
afforded the 3,3¤,4,4¤-tetrahydroxy derivative of the salphen
ligand (H2Lb) in high yield (Scheme 4). However, a similar
reaction of 1H3 in chloroform-methanol (v/v = 1/1) resulted
in dark brown sticky material, in which we did not detect any
trace of the desired product. Use of nitrogen-bubbled solvents
significantly suppressed the generation of unidentifiable by-
products. In view of the above discussion, we suggest that in
the presence of molecular oxygen 1H3 underwent irreversible
oxidation owing to the basicity of the phenylenediamine, and
hence no Schiff base product was formed. When we performed
Scheme 3. Plausible redox scheme for 1H3.
¹
1H2 (for plausible structures, Scheme 3). Along with this
change, the anodic current at 620 mV diminished and the
cathodic current at ¹820 mV increased upon addition of the
amine, in agreement with the above interpretation.
In contrast, upon stepwise addition of p-toluenesulfonic
acid, the anodic peak at ¹140 mV gradually diminished
(Figure 5). The change in the anodic current became saturated
after the addition of 1 equiv of acid (see inset), suggesting that
protonation of the -N(CH3)2 group suppressed the generation

702
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
Redox-Active Amphoteric Pyrogallolaldehyde
TsH
3-OH
ArH
H2Lb
+
HO
HO
CHO
OH
4-OH
-NH
-N=CH-
×5
H2N
NH2
(a)
(b)
N
O
N
O
-N=CH-
M(OAc)2·nH2O
ArH
TsH
-NH
4-OH
M
×5
3-OH
(c)
11
HO
OH
HO
OH
10
9
8
7
Chemical Shift/ppm
MLb; M = Ni, Cu, Zn
Figure 6. 1H NMR spectra of functionalized salphen com-
plexes: (a) NiLa¢2TsOH, (b) NiLa¢2TsOH and additional
2 equiv of p-toluenesulfonic acid, and (c) ZnLa¢2TsOH.
Scheme 4. Synthesis of a pyrogallol-derived salphen ligand
and its metal complex.
and the signal at 9.1 ppm sharpened as well (Figure 6). Upon
addition of acid, the N-methylene and N-methyl protons
appeared as doublets with coupling constants of 5.4-5.5 Hz,
suggesting that the -NH signal appeared as a complicated
multiplet if the proton was firmly associated with the amino
group. Therefore, we assigned the broad signals at around
9 ppm observed for the nickel and zinc complexes to -NH
protons. Relatively sharp singlets at 8.75 ppm for the nickel
complex and at 7.86 ppm for the zinc complex were assigned to
3-OH protons; the sharpness of these peaks may have resulted
from intramolecular hydrogen bonding to the O-(Ni or Zn)
moiety. The corresponding peaks were observed at 9.36
ppm for NiLb and at 9.12 ppm for ZnLb. The chemical shift
differences between the complexes of the two metals may
have resulted from differences in the acidity of the OH groups.
Although these assignments are tentative, the acceptable
consistency of the spectra with the proposed salphen complex
shows that there was no trace of any by-products formed by
undesired coordination at the 3-OH or 4-OH group.
The UV-vis spectra of NiLa¢2TsOH, CuLa¢2TsOH, and
ZnLa¢2TsOH were compared with the spectra of the corre-
sponding NiLb, CuLb, and ZnLb complexes (Figure 7). Note
that 2 equiv of p-toluenesulfonic acid was added to the latter
complexes to achieve dissociation states similar to those of
the former. The spectral profiles of MLa¢2TsOH and MLb were
quite similar to each other for all three sets of complexes,
indicating that the electronic states of these complexes were
essentially identical. However, the intensities of the shoulder
bands (ca. 480 nm for nickel, ca. 460 nm for copper, and
ca. 450 nm for zinc) at the low-energy absorption edge were
slightly higher for the MLa complexes. We attributed these
bands to ³-³* transitions, mainly in the hydroxybenzaldehyde
moieties; to some extent, these bands assumed the character of
metal-to-ligand charge-transfer bands.38 Thus, it is conceivable
that slight modification of the acidity of the OH groups by
the -CH2N+H(CH3)2 group was responsible for the changes
in the spectra.
H2La·2TsOH
+
1H3
+
2TsOH
H2N
NH2
N
O
N
O
M(OAc)2·nH2O
SO3
NH
SO3
HN
M
HO
OH
HO
OH
MLa·2TsOH; M = Ni, Cu, Zn
Scheme 5. Synthesis of a functionalized salphen ligand and
its metal complex.
the same reaction in the presence of 1 equiv of p-toluenesul-
fonic acid with respect to 1H3, we obtained the 3,3¤,4,4¤-
tetrahydroxy-5,5¤-bis(N,N-dimethylaminomethyl) derivative of
salphen in 95% yield as its di-p-toluenesulfonic acid salt
(H2La¢2TsOH, Scheme 5). The 1H NMR chemical shift of
the methylene protons was 4.18 ppm, which indicates that the
amino group was protonated, like that of 1H4
The functionalized salphen ligand H2La¢2TsOH was allowed
to react separately with the acetates of nickel(II), copper(II),
and zinc(II). The reaction mixtures readily afforded crystalline
NiLa¢2TsOH, CuLa¢2TsOH, and ZnLa¢2TsOH, respectively,
+
.
1
in good yields. The H NMR chemical shifts of the methylene
(-CH2N(CH3)2) protons were 4.14 ppm for the nickel com-
plex and 4.12 ppm for the zinc complex, showing that these
complexes also existed in protonated form in solution. There
were two extremely broadened signals at 9.1 and 10.1 ppm
for the nickel complex and at 9.0 and 10.0 ppm for the zinc
complex. The peak areas indicated two protons for each peak,
and the peaks were assignable to either 4-OH or -NH protons.
Then, singlet peak at 8.75 ppm is attributable to 3-OH. When
an additional 2 equiv of p-toluenesulfonic acid was added to
NiLa¢2TsOH, the signal at 10.1 ppm became a sharp singlet,
To investigate the spectroscopic and electrochemical proper-
ties of the metal complexes in their neutral states (MLa), we
added 2 equiv of triethylamine to the solutions of MLa¢2TsOH.

H. Shingai et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
4.0
703
4.0
(a)
(a)
3.0
2.0
1.0
0.0
3.0
2.0
1.0
×10
ε
ε
0.0
300
400
500
600
700
4.0
λ
/ nm
(b)
(b)
200
100
0
3.0
2.0
1.0
(ii)
100 mV s−1
100 mV s−1
(i)
ε
0.0
4.0
-100
(c)
-500
0
500
Potential/mV (vs. Ag+/Ag)
3.0
2.0
1.0
0.0
Figure 8. (a) UV-vis absorption spectra and (b) cyclic
voltammograms of CuLa¢2TsOH before (dashed lines) and
after (solid lines) the addition of triethylamine. In (b),
scan regions were ¹700 to +500 mV for (i) and ¹700 to
¹100 mV for (ii). The baseline of voltammogram (ii) is
shifted by +100 ¯A for clarity.
ε
1H3 (Figure 4), because replacing the 2-OH proton with a
copper ion would have changed the redox potentials. In
contrast, we observed no remarkable changes in the absorp-
tion spectra of NiLa¢2TsOH and ZnLa¢2TsOH solutions when
2 equiv of triethylamine was added. However, the cyclic
voltammogram of NiLa¢2TsOH changed drastically upon
addition of the amine, as was observed for the copper complex
(Figure S8). Although ZnLa¢2TsOH gave a voltammogram
similar to that of CuLa¢2TsOH, it did not show any pronounced
change upon addition of base. At present, we have no clear
explanation for the difference in redox behavior dependent
on metal species. One possible hypothesis is that the redox
potential of the metal ion affects the stability of the oxidized
ligand moiety. This topic needs to be discussed with further
experiments in the future.
300
400
500
/ nm
600
λ
Figure 7. UV-vis spectra of MLa¢2TsOH (solid line) and
MLb (dashed line) complexes in DMSO with 2 equiv of
acid: (a) M = Ni, (b) M = Cu, and (c) M = Zn.
In the UV-vis absorption spectrum of CuLa¢2TsOH, the
absorptivity of the peak at 580 nm increased slightly (to
ca. 1000), whereas the absorptivity of the peak at 350-400 nm
was slightly diminished (Figure 8a). These changes imply the
formation of ligand-center oxidized species by autoxidation,
as was observed for 1H3 (Figure 3). Accordingly, we can
assume that under neutral to basic conditions, CuLa was easily
oxidized. This assumption was supported by cyclic voltammo-
grams (Figure 8b): under acidic conditions, we observed an
oxidation wave at +400 mV and a reduction wave at ¹500 mV,
whereas upon addition of triethylamine, a new oxidation wave
appeared at ¹300 mV together with a slight decrease of the
current at +400 mV. The peaks at ¹500 and ¹300 mV formed
a pseudoreversible redox wave, attributable to the reaction
shown in Scheme 4. It is conceivable that these two peaks
correspond to the peaks at ¹820 and ¹140 mV observed for
Conclusion
In summary, we synthesized a new amphoteric pyrogallol-
aldehyde (1H3) and investigated its electrochemical properties.
This aldehyde was redox active and underwent facile autoxi-
dation. Protonation of the side chains switched off the redox
activity and hence facilitated further reactions such as Schiff
base formation and metal complexation. We demonstrated the
preparation of functionalized salphen-type complexes of some

704
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
Redox-Active Amphoteric Pyrogallolaldehyde
transition metals, which had electronic states that were essen-
tially identical to those of the analogous hydroxy-rich salphen
complexes. It is notable that we could switch on the redox
function of the salphen complex by removing the proton on the
side chain, which enabled us to access some complexes with
oxidation locus on the ligand. Furthermore, we have a broad
choice of amines, including diamines and triamines having
various linker moieties, with which we will be able to construct
metallosupramolecules and metal-containing polymers.
4.18 (s, -CH2-, 4H), 7.10 (d, 3J = 7.8 Hz, TsH, 4H), 7.15
(s, ArH, 2H), 7.26 (s, ArH, 2H), 7.47 (d, 3J = 7.8 Hz, TsH, 4H),
8.78 (s, -N=CH-, 2H), 9.2-10.2 (br, -OH/-NH, 4H), 13.86
(brs, -OH, 2H), residual -OH signal (for 2H) was not observed
as a recognizable peak; 13C NMR: ¤ 19.2, 20.8, 42.1, 55.3,
109.3, 112.1, 120.5, 125.6, 127.7, 128.2, 132.6, 136.0, 137.9,
138.3, 145.5, 153.7, 162.5.
Synthesis of N,N¤-Bis(2,3,4-trihydroxybenzylidene)-4,5-
dimethyl-1,2-diaminobenzene (H2Lb). To a methanol solu-
tion (50 mL) of 3.09 g of 2,3,4-trihydroxybenzaldehyde (20
mmol) was added 1.36 g (10 mmol) of 4,5-dimethyl-1,2-phen-
ylenediamine. Within several hours the mixture gave a yellow
precipitate, which was collected by filtration and washed with
methanol (3.66 g, 90%). Mp 200-201 °C; Anal. Calcd for
C22H20N2O6 (408.40): C, 64.70; H, 4.94; N, 6.86%. Found: C,
64.12; H, 4.87; N, 6.81%. FAB MS(+) m/z = 409.1 (409.14
Experimental
General. All chemicals and solvents were purchased from
Tokyo Kasei Kogyo (TCI) and used without further purifica-
tion. UV-vis absorption spectra were measured for a 2 ©
10¹5 M solution of each solute with a JASCO V-630 spectro-
photometer. NMR spectra were recorded on a JEOL ECS-
1
¹1
400 instrument (400 MHz for H) for DMSO-d6 solution. IR
calcd for M + H+); IR (KBr): 1631 cm (¯C=N); 1H NMR:
3
spectra were recorded with a JEOL FT/IR-420 instrument.
Electrochemical measurements were performed by using a
conventional three-electrode cell with a glassy carbon working
electrode, a platinum wire electrode, and an Ag+/Ag reference
electrode. The samples were prepared as 5.0 mM solutions in
dry dimethylformamide with 0.1 M tetrabutylammonium hexa-
fluorophosphate as the supporting electrolyte. Cyclic voltam-
mograms were run by using a BAS 100B/W system. Solutions
were purged with nitrogen gas before each run. Melting points
were determined with an optical microscope equipped with a
¤ 2.28 (s, -CH3, 6H), 6.38 (d, J = 8.5 Hz, ArH, 2H), 6.93 (d,
3J = 8.5 Hz, ArH, 2H), 7.19 (s, ArH, 2H), 8.71 (s, -N=CH-,
2H), 9.12 (brs, -OH, 4H), 13.46 (brs, -OH, 2H); 13C NMR: ¤
19.1, 107.8, 112.7, 120.5, 124.1, 132.4, 135.4, 139.2, 150.3,
152.0, 162.8.
Synthesis of Divalent Metal Complexes of La (MLa¢
2TsOH; M = Ni, Cu, Zn). To a methanol solution (10 mL) of
H2La¢2TsOH (0.87 g, 1 mmol) was added a methanol solution
(10 mL) of the corresponding metal acetate hydrate (1 mmol).
Within several hours the mixture gave a precipitate, which was
collected by filtration and washed with diethyl ether.
NiLa¢2TsOH: Reddish brown solid (0.67 g, 73%). Mp
226-227 °C; Anal. Calcd for C42H48N4NiO12S2¢H2O (941.69):
C, 53.57; H, 5.35; N, 5.95%. Found: C, 53.60; H, 5.57; N,
5.84%. FAB MS(+) m/z = 579.7 (579.15 calcd for NiLa +
Linkam LK-600 temperature-variable stage; the heating rate
¹1
was 2 K min
.
Synthesis of 2,3,4-Trihydroxy-5-dimethylaminomethyl-
benzaldehyde (1H3). To 30 mL of methanol was added
1.08 g of aqueous dimethylamine (50%) (12 mmol for dimeth-
ylamine) and 0.36 g of paraformaldehyde (12 mmol for CH2O
unit). The mixture was stirred at 70 °C to obtain clear solution,
to which 1.54 g of 2,3,4-trihydroxybenzaldehyde (10 mmol)
was added. After 30 min of heating, the reaction mixture was
left at ambient temperature to grow colorless to pale yellow
precipitate. The product was collected by filtration, and washed
with methanol (1.54 g, 73%). Mp 146-147 °C; Anal. Calcd for
C10H13NO4 (211.22): C, 56.86; H, 6.20; N, 6.63%. Found: C,
57.25; H, 6.33; N, 6.61%. FAB MS(+): m/z = 212.2 (212.09
¹1
H+); IR (KBr): ¯ = 1626 cm (¯C=N), 1035, 1011, 817, 685,
¹
569 cm¹1 (TsO ); 1H NMR: ¤ 2.28 (s, Ar-CH3/Ts-CH3, 12H),
2.74 (s, N-CH3, 12H), 4.14 (s, -CH2-, 4H), 7.11 (s, ArH, 2H),
7.11 (d, 3J = 8.0 Hz, TsH, 4H), 7.47 (d, 3J = 8.0 Hz, TsH, 4H),
7.86 (s, ArH, 2H), 8.41 (s, -N=CH-, 2H), 8.75 (s, -OH, 2H),
9.00 (br, -NH, 2H), 10.1 (br, -OH, 2H); 13C NMR has not been
obtained because no signals were found for a concentrated
sample solution. For this concentrated sample, 1H NMR spectra
gave significantly broadened signals.
¹1
calcd for M + H+); IR (KBr): 1643 cm (¯C=O); 1H NMR:
CuLa¢2TsOH: Reddish brown solid (0.78 g, 84%). Mp
184-185 °C; Anal. Calcd for C42H48CuN4O12S2¢1.5H2O
(955.55): C, 52.79; H, 5.38; N, 5.86%. Found: C, 52.72; H,
5.53; N, 5.88%. FAB MS(+) m/z = 583.7 (584.17 calcd for
¤ 2.45 (s, -CH3, 6H), 3.80 (s, -CH2-, 2H), 6.8 (br, -OH), 6.92
(s, ArH, 1H), 9.59 (s, -CHO, 1H); 13C NMR: ¤ 42.8, 59.7,
111.9, 113.2, 124.8, 132.1, 148.8, 160.1, 191.0.
¹1
Synthesis of N,N¤-Bis(2,3,4-trihydroxy-5-dimethylamino-
methylbenzylidene)-4,5-dimethyl-1,2-diaminobenzene Di-p-
toluenesulfonic Acid Salt (H2La¢2TsOH). To a THF solution
(20 mL) of 1 (2.11 g, 10 mmol) and p-toluenesulfonic acid
(1.72 g, 10 mmol) was added 0.68 g (5 mmol) of 4,5-dimethyl-
1,2-phenylenediamine. Then the mixture separated an orange
oil, which later crystallized within several hours. The crys-
talline material was collected by filtration and washed with
THF (4.10 g, 95%). Mp 135-136 °C; Anal. Calcd for C42H50-
N4O12S2¢H2O (884.30): C, 57.00; H, 5.92; N, 6.33%. Found:
C, 57.09; H, 6.23; N, 6.21%. FAB MS(+): m/z = 523.8
CuLa + H+); IR (KBr): 1627 cm (¯C=N), 1035, 1011, 817,
¹1
¹
684, 569 cm (TsO ).
ZnLa¢2TsOH: Yellowish brown solid (0.70 g, 75%). Mp
>220 °C (decomp.); Anal. Calcd for C42H48N4O12S2Zn¢H2O
(948.39): C, 53.19; H, 5.31; N, 5.91%. Found: C, 53.04; H,
5.37; N, 5.75%. FAB MS(+) m/z = 585.2 (586.97 calcd for
¹1
ZnLa + H+); IR (KBr): 1617 cm (¯C=N), 1035, 1011, 817,
¹1
¹
1
686, 570 cm (TsO ); H NMR: ¤ 2.28 (s, ArH, 6H), 2.31
(s, TsH, 6H), 2.74 (s, N-CH3, 12H), 4.12 (s, -CH2-, 4H), 6.97
3
(s, ArH, 2H), 7.10 (d, 3J = 7.9 Hz, TsH, 4H), 7.47 (d, J = 7.9
Hz, TsH, 4H), 7.65 (s, ArH, 2H), 7.86 (brs, -OH, 2H), 8.88
(s, -N=CH-, 2H), 9.0 (br, -NH, 2H), 9.8 (br, -OH, 2H);
13C NMR: ¤ 19.6, 20.8, 41.9, 105.8, 112.4, 125.5, 125.6, 128.2,
135.4, 135.7, 136.9, 137.9, 145.5, 145.4, 161.6.
¹1
(523.26 calcd for H2La + H+); IR (KBr): 1625 cm (¯C=N),
¹1
¹
1034, 1009, 818, 682, 567 cm (TsO ); 1H NMR: ¤ 2.28
(s, Ts-CH3, 6H), 2.30 (s, Ar-CH3, 6H), 2.73 (s, N-CH3, 12H),

H. Shingai et al.
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
705
Synthesis of Divalent Metal Complexes of Lb (MLb;
M = Ni, Cu, Zn).
References
J. Stubbe, D. G. Nocera, C. S. Yee, M. C. Y. Chang, Chem.
Rev. 2003, 103, 2167.
To a THF solution (10 mL) of H2Lb
1
(0.41 g, 1 mmol) was added a methanol solution (10 mL) of the
corresponding metal acetate hydrate (1 mmol). Within several
hours the mixture gave a precipitate, which was collected by
filtration and washed with methanol.
NiLb: Reddish brown solid (0.41 g, 89%). Mp >300 °C
(decomp.); Anal. Calcd for C22H18N2NiO6¢H2O (483.10): C,
54.70; H, 4.17; N, 5.80%. Found: C, 54.78; H, 4.01; N, 5.77%.
FAB MS(+) m/z = 465.0 (465.05 calcd for NiLb + H+); IR
(KBr): 1626 cm¹1 (¯C=N); 1H NMR: ¤ 2.24 (s, -CH3, 6H), 6.26
(d, 3J = 8.8 Hz, ArH, 2H), 6.90 (d, 3J = 8.8 Hz, ArH, 2H), 7.79
(s, ArH, 2H), 8.06 (brs, -OH, 2H), 8.47 (s, -N=CH-, 2H), 9.36
(brs, -OH, 2H); 13C NMR spectrum has not been obtained due
to insufficient solubility.
2
3
J. W. Whittaker, Chem. Rev. 2003, 103, 2347.
a) M. M. Whittaker, P. J. Kersten, N. Nakamura, J. Sanders-
Loehr, E. S. Schweizer, J. W. Whittaker, J. Biol. Chem. 1996, 271,
681. b) M. M. Whittaker, P. J. Kersten, D. Cullen, J. W. Whittaker,
J. Biol. Chem. 1999, 274, 36226.
4
198, 3.
5
S. Itoh, M. Taki, S. Fukuzumi, Coord. Chem. Rev. 2000,
B. A. Jazdzewski, W. B. Tolman, Coord. Chem. Rev. 2000,
200-202, 633.
6
7
W. Kaim, Dalton Trans. 2003, 761.
a) D. Parker, E. S. Davies, C. Wilson, J. McMaster, Inorg.
Chim. Acta 2007, 360, 203. b) L. Benisvy, A. J. Blake, D.
Collison, E. S. Davies, C. D. Garner, E. J. L. McInnes, J.
McMaster, G. Whittaker, C. Wilson, Dalton Trans. 2003, 1975.
c) L. Benisvy, E. Bill, A. J. Blake, D. Collison, E. S. Davies, C. D.
Garner, G. McArdle, E. J. L. McInnes, J. McMaster, S. H. K. Ross,
C. Wilson, Dalton Trans. 2006, 258. d) L. Benisvy, E. Bill, A. J.
Blake, D. Collison, E. S. Davies, C. D. Garner, C. I. Guindy,
E. J. L. McInnes, G. McArdle, J. McMaster, C. Wilson, J.
Wolowska, Dalton Trans. 2004, 3647.
CuLb: Reddish brown solid (0.41 g, 87%). Mp >300 °C
(decomp.); Anal. Calcd for C22H18CuN2O6¢0.5H2O (478.94):
C, 55.17; H, 4.00; N, 5.85%. Found: C, 54.78; H, 3.79; N,
5.72%. FAB MS(+) m/z = 470.1 (470.05 calcd for CuLb +
¹1
H+), IR (KBr): 1620 cm (¯C=N).
ZnLb: Yellowish brown solid (0.46 g, 97%). Mp >300 °C
(decomp.); Anal. Calcd for C22H18N2ZnO6¢1.5H2O (498.80):
C, 52.97; H, 4.24; N, 5.62%. Found: C, 53.35; H, 4.01; N,
5.60%. FAB MS(+) m/z = 471.1 (471.05 calcd for ZnLb +
8
A. Mukherjee, F. Lloret, R. Mukherjee, Eur. J. Inorg.
Chem. 2010, 1032.
¹1
1
H+); IR (KBr): 1618 cm (¯C=N); H NMR: ¤ 2.29 (s, -CH3,
6H), 6.13 (d, 3J = 9.0 Hz, ArH, 2H), 6.77 (d, 3J = 9.0 Hz, ArH,
2H), 7.56 (brs, -OH, 2H), 7.63 (s, ArH, 2H), 8.84 (s, -N=CH-,
2H), 9.13 (brs, -OH, 2H); 13C NMR: ¤ 19.5, 105.4, 112.6,
116.5, 125.9, 135.1, 135.2, 136.9, 147.2, 160.3, 160.8.
Crystallography. For X-ray diffraction of single crystals,
data were collected on a MacScience DIP Labo Imaging
Plate diffractometer, ®(Cu K¡) = 1.5418 ¡. The structure was
solved by direct methods and expanded using Fourier tech-
niques. All calculations were performed with the crystallo-
graphic software package SHELX-97.39 Crystallographic data
have been deposited with Cambridge Crystallographic Data
Centre: Deposition numbers CCDC-900253 to compound
1H3. Copies of the data can be obtained free of charge via
http://www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
Cambridge Crystallographic Data Centre, 12, Union Road,
Cambridge, CB2 1EZ, U.K.; Fax: +44 1223 336033; e-mail:
deposit@ccdc.cam.ac.uk).
9
J. Müller, A. Kikuchi, E. Bill, T. Weyhermüller, P.
Hildebrandt, L. Ould-Moussa, K. Wieghardt, Inorg. Chim. Acta
2000, 297, 265.
10 C. G. Pierpont, Coord. Chem. Rev. 2001, 216-217, 99.
11 P. Chaudhuri, C. N. Verani, E. Bill, E. Bothe, T.
Weyhermüller, K. Wieghardt, J. Am. Chem. Soc. 2001, 123, 2213.
12 E. Evangelio, D. Ruiz-Molina, Eur. J. Inorg. Chem. 2005,
2957.
13 P. Gütlich, Y. Garcia, T. Woike, Coord. Chem. Rev. 2001,
219-221, 839.
14 O. Margeat, P. G. Lacroix, J. P. Costes, B. Donnadieu, C.
Lepetit, K. Nakatani, Inorg. Chem. 2004, 43, 4743.
15 T. J. O’Sullivan, B. Djukic, P. A. Dube, M. T. Lemaire,
Chem. Commun. 2009, 1903.
16 D. Kiriya, H.-C. Chang, K. Nakamura, D. Tanaka, K.
Yoneda, S. Kitagawa, Chem. Mater. 2009, 21, 1980.
17 a) O. Rotthaus, F. Thomas, O. Jarjayes, C. Philouze, E.
Saint-Aman, J.-L. Pierre, Chem.®Eur. J. 2006, 12, 6953. b) O.
Rotthaus, O. Jarjayes, F. Thomas, C. Philouze, E. Saint-Aman,
J.-L. Pierre, Dalton Trans. 2007, 889. c) O. Rotthaus, O. Jarjayes,
C. Philouze, C. P. Del Valle, F. Thomas, Dalton Trans. 2009, 1792.
d) H. Arora, C. Philouze, O. Jarjayes, F. Thomas, Dalton Trans.
2010, 39, 10088.
18 a) Y. Shimazaki, F. Tani, K. Fukui, Y. Naruta, O. Yamauchi,
J. Am. Chem. Soc. 2003, 125, 10512. b) Y. Shimazaki, T. Yajima,
F. Tani, S. Karasawa, K. Fukui, Y. Naruta, O. Yamauchi, J. Am.
Chem. Soc. 2007, 129, 2559. c) Y. Shimazaki, N. Arai, T. J. Dunn,
T. Yajima, F. Tani, C. F. Ramogida, T. Storr, Dalton Trans. 2011,
40, 2469.
Crystal Data of 1H3:
less plate, Mr = 215.72, monoclinic, a = 9.5940(3), b =
12.1230(4), c = 18.3890(6) ¡, ¢ = 92.511(2)°, V =
(C10H13NO4)2¢0.5H2O, color-
2136.74(12) ¡3, Dcalcd = 1.341 g cm¹3, T = 193 K, space group
P2/n (#13), Z = 4, ®(Cu K¡) = 8.88 cm¹1, 19461 reflections
measured and 3450 unique (2ªmax = 144.1°, Rint = 0.034)
which were used in all calculations. R = 0.073, Rw = 0.196.
The authors are grateful to Prof. Kazuyuki Ishii at the
University of Tokyo and Dr. Atsushi Okazawa at the University
of Tokyo for their kind cooperation on ESR measurements
and helpful discussion about organic radicals.
19 T. J. Dunn, C. F. Ramogida, C. Simmonds, A. Paterson,
E. W. Y. Wong, L. Chiang, Y. Shimazaki, T. Storr, Inorg. Chem.
2011, 50, 6746.
20 A. Kochem, O. Jarjayes, B. Baptiste, C. Philouze, H. Vezin,
K. Tsukidate, F. Tani, M. Orio, Y. Shimazaki, F. Thomas, Chem.®
Eur. J. 2012, 18, 1068.
Supporting Information
Supplementary spectra, crystallograms, voltammograms, and
computational data. This material is available free of charge on
the Web at: http://www.csj.jp/journals/bcsj/.
21 a) S. Routier, H. Vezin, E. Lamour, J.-L. Bernier, J.-P.

706
Bull. Chem. Soc. Jpn. Vol. 86, No. 6 (2013)
Redox-Active Amphoteric Pyrogallolaldehyde
Catteau, C. Bailly, Nucleic Acids Res. 1999, 27, 4160. b) E.
Lamour, S. Routier, J.-L. Bernier, J.-P. Catteau, C. Bailly, H. Vezin,
J. Am. Chem. Soc. 1999, 121, 1862. c) H. Vezin, E. Lamour, S.
Routier, F. Villain, C. Bailly, J.-L. Bernier, J. P. Catteau, J. Inorg.
Biochem. 2002, 92, 177.
110, 7024.
31 F. Thomas, H. Arora, C. Philouze, O. Jarjayes, Inorg. Chim.
Acta 2010, 363, 3122.
32 S. W. Ng, Acta Crystallogr., Sect. E: Struct. Rep. Online
2008, 64, o1565.
22 S. S. Tonde, A. S. Kumbhar, S. B. Padhye, R. J. Butcher,
J. Inorg. Biochem. 2006, 100, 51.
33 J. Kruszewski, T. M. Krygowski, Tetrahedron Lett. 1972,
13, 3839.
23 S. J. Wezenberg, A. W. Kleij, Angew. Chem., Int. Ed. 2008,
47, 2354.
34 H. Houjou, T. Motoyama, S. Banno, I. Yoshikawa, K.
Araki, J. Org. Chem. 2009, 74, 520.
24 S. Akine, J. Inclusion Phenom. Macrocyclic Chem. 2012,
72, 25.
35 T. Maki, Y. Araki, Y. Ishida, O. Onomura, Y. Matsumura,
J. Am. Chem. Soc. 2001, 123, 3371.
25 A. C. W. Leung, M. J. MacLachlan, J. Inorg. Organomet.
Polym. Mater. 2007, 17, 57.
26 S. Akine, T. Nabeshima, Dalton Trans. 2009, 10395.
27 J. Jiang, M. J. MacLachlan, Chem. Commun. 2009, 5695.
28 M. Narita, I. Yoon, M. Aoyagi, M. Goto, T. Shimizu, M.
Asakawa, Eur. J. Inorg. Chem. 2007, 4229.
29 a) T. F. Markle, I. J. Rhile, A. G. DiPasquale, J. M. Mayer,
Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 8185. b) I. J. Rhile, T. F.
Markle, H. Nagao, A. G. DiPasquale, O. P. Lam, M. A. Lockwood,
K. Rotter, J. M. Mayer, J. Am. Chem. Soc. 2006, 128, 6075.
30 J. L. Dempsey, J. R. Winkler, H. B. Gray, Chem. Rev. 2010,
36 F. Thomas, O. Jarjayes, H. Jamet, S. Hamman, E.
Saint-Aman, C. Duboc, J.-L. Pierre, Angew. Chem., Int. Ed.
2004, 43, 594.
37 E. R. Altwicker, Chem. Rev. 1967, 67, 475.
38 a) H. Houjou, T. Motoyama, K. Araki, Eur. J. Inorg. Chem.
2009, 533. b) H. Houjou, M. Ito, K. Araki, Inorg. Chem. 2009, 48,
10703. c) H. Houjou, M. Ito, K. Araki, Inorg. Chem. 2011, 50,
5298.
39 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found.
Crystallogr. 2008, 64, 112.