Metal-metal bonded diruthenium unit axial-capped by di-tert-butylphenolate: [Ru2(O2CCH3)2(t-Busal-R′py) 2]- (t-Busal-R′py2- = N-(R′-2-pyridyl)-2-oxido-3,5-di-tert-butylbenzylamin

Article abstract of DOI:10.1246/bcsj.79.612

Three diruthenium compounds, [Na(18-crown-6)(thf)₂][Ru ₂(O₂CCH₃)₂(t-Busalpy)₂] (1), [K(18-crown-6)(thf)₂][Ru₂(O₂CCH ₃)₂(t-Busal-4-Mepy

Full text of DOI:10.1246/bcsj.79.612

612 Bull. Chem. Soc. Jpn. Vol. 79, No. 4, 612–620 (2006)
Ó 2006 The Chemical Society of Japan
Metal–Metal Bonded Diruthenium Unit Axial-Capped
by Di-tert-butylphenolate: [Ru2(O2CCH3)2(t-Busal-R0py)2]ꢀ
(t-Busal-R0py2ꢀ = N-(R0-2-pyridyl)-2-oxido-3,5-di-tert-
butylbenzylaminato; R0 ¼ H, 4-Me, and 5-Me)
ꢀ1
Hitoshi Miyasaka, Toru Izawa,1 Shinya Takaishi,2 Kunihisa Sugimoto,3
Ken-ichi Sugiura,1 and Masahiro Yamashita2
1Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University,
1-1 Minamiohsawa, Hachioji, Tokyo 192-0397
2Department of Chemistry, Graduate School of Science, Tohoku University,
6-3 Aoba, Aramaki, Aoba-ku, Sendai 980-8578
3X-ray Research Laboratory, Rigaku Co., Ltd., 3-9-12 Matsubara-cho, Akishima, Tokyo 196-8666
Received October 20, 2005; E-mail: miyasaka@comp.metro-u.ac.jp
Three diruthenium compounds, [Na(18-crown-6)(thf)2][Ru2(O2CCH3)2(t-Busalpy)2] (1), [K(18-crown-6)(thf)2]-
[Ru2(O2CCH3)2(t-Busal-4-Mepy)2] (2), and [K(18-crown-6)(thf)(H2O)][K(18-crown-6)(thf)(MeO)][Ru2(O2CCH3)2-
(t-Busal-5-Mepy)2] (3) (18-crown-6 = 1,4,7,10,13,16-hexaoxocyclooctadecane; 18-crown-6-ether), have been synthe-
sized by the ligand substitution reaction of Ru2(O2CCH3)4Cl with newly prepared tridentate bridging/chelating ligands,
N-(2-pyridyl)-2-oxido-3,5-di-tert-butylbenzylaminato (t-Busalpy2ꢁ), N-(4-methyl-2-pyridyl)-2-oxido-3,5-di-tert-butyl-
benzylaminato (t-Busal-4-Mepy2ꢁ), and N-(5-methyl-2-pyridyl)-2-oxido-3,5-di-tert-butylbenzylaminato (t-Busal-5-
Mepy2ꢁ), respectively, and isolated using [Na(18-crown-6)]þ or [K(18-crown-6)]þ as the counter cation. The struc-
tural features of the anionic parts are very similar to those of the previously synthesized family of [Ru2(O2CCH3)2-
(5-Rsalpy)2]ꢁ (R ¼ H, Me, Cl, Br, and NO2); two acetate and two t-Busal-R0py2ꢁ ligands are respectively located
around the Ru2 unit in a trans fashion, where the t-Busal-R0py2ꢁ ligand acts as a tridentate ligand having both bridging
and chelating characters to form the M–M bridging/axial-chelating mode. Their Ru–Ru bonded core with an electronic
2
ꢀ
3
ꢀ
configuration of ꢀ2ꢁ4ꢂ ðꢁ ꢂ Þ is axial-capped by the di-t-butylphenolate groups of the t-Busal-R0py2ꢁ ligands, and
4þ
6þ
experiences multi-redox properties: a one-electron reduction to Ru2 and two one-electron oxidations to Ru2 fol-
lowed by Ru27þ, in addition to t-Busal-R0py2ꢁ ligand-centered redox. The strong electron-donating ability of the t-butyl
groups compared with the other R-groups leads to negative shifts of the redox potentials according to the Hammett law.
Since the discovery of the Re–Re quadruple bond in
1
N–N–O catecholate type (II), and N–N–O phenolate type (III).
The common aspect in these bridging/chelating ligands is the
existence of redox-active organic moieties corresponding to
[Re2Cl8]2ꢁ
,
paddlewheel-type metal–metal bonded com-
pounds have been extensively studied over the past three dec-
ades,2 principally from the viewpoint of the correlation be-
tween their core structures, including the metal–metal bond
and the electronic structures on a set of M–M frontier orbitals,
ꢀ
ꢀ
ꢀ
ꢀꢁ2ꢂꢂ ꢁ ꢀ . The most fascinating feature of paddlewheel-
2
type compounds is their multi-redox activity; the compounds
undergo both oxidation and reduction on the dimetal center
with a systematical change of the metal–metal bond length ac-
cording to the electron arrangement on the frontier orbitals, but
without a considerable structural change. Therefore, the elec-
tronic tuning of the metal–metal bonding core is an attractive
theme. Indeed, several studies have shown that the tuning is
realized by an electronic effect induced by remote substituents
on the surrounding bridging ligands, in which the Hammett
correlation has revealed the possibility of fine-tuning of the
M–M based redox potentials.3 The ligand-induced electronic
effect could also be seen in dimetal compounds with specific
bridging/chelating polydentate ligands. Such ligands are clas-
sified into three types in Chart 1: rigid N–N–N pyridyl type (I),
Chart 1.
Published on the web April 7, 2006; DOI: 10.1246/bcsj.79.612

H. Miyasaka et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
613
1,8-naphthyridine in I,4–6 2-oxoanilino group in II,7–9 and phen-
oxido group in III,10,11 with which their dimetal compounds un-
dergo not only metal-centered redox, but also ligand-centered
redox. Moreover, some of them revealed d–ꢁ electron-conju-
gated redox behavior.
drying agents and freshly distilled under a nitrogen atmosphere
before use.
Preparation of Protonated Ligands, t-BusalpyH2, t-Busal-4-
MepyH2, and t-Busal-5-MepyH2. An ethanol solution of the
corresponding Schiff-base compounds for the derivation of the
reduced forms, t-BusalpyH2, t-Busal-4-MepyH2, and t-Busal-5-
MepyH2, was prepared by refluxing 3,5-di-tert-butylsalicylalde-
hyde (4.69 g, 20 mmol) and 2-aminopyridine (1.88 g, 20 mmol), 2-
amino-4-picoline (2.16 g, 20 mmol), and 2-amino-5-picoline (2.16
g, 20 mmol), respectively, in ethanol (30 mL). Without isolating
the Schiff-base compound, to the ethanol solution was slowly add-
ed solid NaBH4 (756 mg, 20 mmol) at room temperature. During
this time, the yellow suspension became a colorless solution. This
solution was stirred overnight at room temperature. To the solu-
tion was slowly added water (100 mL) to form colorless block-
shaped microcrystals of the ligands. The crude sample was recrys-
tallized from an ethanol/water mixture. The crystalline sample
was filtered, washed with water, and dried in vacuo over P2O5.
Yield 1.9 g, 30% for t-BusalpyH2; 2.7 g, 41% for t-Busal-4-
MepyH2; 3.6 g, 55% for t-Busal-5-MepyH2. t-BusalpyH2: Anal.
Calcd for C20H28N2O: C, 76.88; H, 9.03; N, 8.97%. Found: C,
76.82; H, 8.97; N, 8.88%. IR (KBr, cmꢁ1): 3433 (m, NH), 2959
(s, CH), 2588 (w, br, OH), 1618 (s), 1570 (w), 1510 (s), 1439
We focus on diruthenium compounds with the type III li-
gand, in which the phenolate component is redox-active, even
in such dimetal compounds. The desired compounds have been
synthesized by the ligand substitution reaction of tetrakis(ace-
tato)diruthenium(II, III) with a family of redox-active triden-
tate ligands, N-(2-pyridyl)-2-oxido-R-benzylaminate (abbrevi-
ated as Rsalpy2ꢁ), to produce a bis-substituted type of [Ru2-
(O2CCH3)2(Rsalpy)2]ꢁ10 and a tris-substituted type of [Ru2(5-
Clsalpy)3]ꢁ.11 Among them, bis-substituted compounds were
first obtained by using 5-position-substituted ligands of type
III, where 5-R corresponded to H, Me, Cl, Br, and NO2 groups
to form a series of A[Ru2(O2CCH3)2(5-Rsalpy)2] (A = coun-
ter cation).10 In addition to metal-centered redox couplings
(Ru25þ/Ru24þ, Ru26þ/Ru25þ, and Ru27þ/Ru26þ), oxidations
of two 5-Rsalpy2ꢁ ligands in a compound were observed as
an irreversible two-electron coupling at relatively high poten-
tials of 0.70 (Me)–1.21 V (NO2) (vs Ag/Agþ) in accord with
the Hammett law. The observation of only one coupling for
the phenolate oxidations, unfortunately, suggests that the elec-
tronic interaction between the two phenoxido groups via the
Ru2 unit is almost negligible.
1
(m), 1236 (m), 1202 (w), 974 (w), 770 (m). H NMR (500 MHz,
CDCl3): ꢂ 11.69 (s, 1H, OH), 8.05 (m, 1H, Cpyr-H), 7.35 (m, 1H,
C
pyr-H), 7.27, (d, 1H, J ¼ 2:5 Hz, Cph-H), 7.07 (d, 1H, J ¼ 2:5 Hz,
Cph-H), 6.54 (t, 1H, J ¼ 6:1 Hz, Cpyr-H), 6.42 (d, 1H, J ¼ 8:6 Hz,
pyr-H), 5.07 (t, 1H, J ¼ 6:2 Hz, NH), 4.45 (2H, J ¼ 6:7 Hz, CH2),
C
Based on this work, we prepared three new type III
ligands, N-(2-pyridyl)-2-oxido-3,5-di-tert-butylbenzylaminate
(t-Busalpy2ꢁ), N-(4-methyl-2-pyridyl)-2-oxido-3,5-di-tert-butyl-
benzylaminate (t-Busal-4-Mepy2ꢁ), and N-(5-methyl-2-pyridyl)-
2-oxido-3,5-di-tert-butylbenzylaminate (t-Busal-5-Mepy2ꢁ),
and synthesized the corresponding diruthenium compounds:
[Na(18-crown-6)(thf)2][Ru2(O2CCH3)2(t-Busalpy)2] (1), [K-
(18-crown-6)(thf)2][Ru2(O2CCH3)2(t-Busal-4-Mepy)2] (2),
and [K(18-crown-6)(thf)(H2O)][K(18-crown-6)(thf)(MeO)]-
[Ru2(O2CCH3)2(t-Busal-5-Mepy)2] (3) (18-crown-6 = 1,4,7,-
10,13,16-hexaoxocyclooctadecane; 18-crown-6-ether). It is
well known that 3,5-di-tert-butylphenolate produces electro-
chemically a relatively persistent phenoxyl radical (reversible
couple) in its metal complexes because of the strong elec-
tron-donating effect of the tert-butyl groups and the ortho and
para substitutions of the parent phenolate, which provide reso-
nance stabilization.12 As expected, compounds 1 and 2 undergo
the ligand-based redox as two separated quasi-reversible cou-
ples with ꢀE1=2 ¼ 96 and 70 mV, respectively, in addition to
the metal-centered redox, indicating a weak electronic interac-
tion between the phenoxido groups via the Ru–Ru bond. By
contrast, compound 3 undergoes phenolate dissociation on
one of the two moieties during the metal-centered oxidation
1.43 (s, 9H, t-Bu), 1.30 (s, 9H, t-Bu). t-Busal-4-MepyH2: Anal.
Calcd for C21H30N2O: C, 77.26; H, 9.26; N, 8.58%. Found: C,
76.85; H, 9.22; N, 8.35%. IR (KBr, cmꢁ1): 3315 (m, NH), 2951
(s, CH), 2588 (w, br, OH), 1632 (s), 1539 (m), 1441 (m), 1238 (m),
1186 (w), 980 (w), 789 (m). 1H NMR (500 MHz, CDCl3): ꢂ 11.87
(s, 1H, OH), 7.92 (d, 1H, J ¼ 5:5 Hz, Cpyr-H), 7.26 (d, 1H, J ¼ 1:6
Hz, Cph-H), 7.06 (d, 1H, J ¼ 2:5 Hz, Cph-H), 6.39 (dd, 1H, J ¼
5:5 Hz, 0.9 Hz, Cpyr-H), 6.22 (s, 1H, Cpyr-H), 4.98 (t, 1H, J ¼ 6:4
Hz, NH), 4.43 (d, 2H, J ¼ 6:4 Hz, CH2), 2.18 (s, 3H, CH3), 1.43
(s, 9H, t-Bu), 1.29 (s, 9H, t-Bu). t-Busal-5-MepyH2: Anal. Calcd
for C21H30N2O: C, 77.26; H, 9.26; N, 8.58%. Found: C, 77.26; H,
9.34; N, 8.49%. IR (KBr, cmꢁ1): 3325 (s, NH), 2953 (s, CH), 2866
(m, CH), 2569 (m, br, OH), 1773 (w), 1632 (s), 1531 (s), 1481 (m),
1
1439 (s), 1360 (s), 1236 (s), 1161 (w), 881 (m), 822 (s). H NMR
(500 MHz, CDCl3): ꢂ 11.77 (s, 1H, OH), 7.88 (s, 1H, Cpry-H), 7.27
(d, 1H, J ¼ 2:5 Hz, Cph-H), 7.20 (m, 1H, Cpyr-H), 7.07 (d, 1H, J ¼
2:5 Hz, Cph-H), 6.35 (d, 1H, J ¼ 8:3 Hz, Cpyr-H), 4.94 (t, 1H, J ¼
6:4 Hz, NH), 4.42 (d, 2H, J ¼ 6:4 Hz, CH2), 2.15 (s, 3H, CH3),
1.45 (s, 9H, t-Bu), 1.31 (s, 9H, t-Bu).
Preparation of Diruthenium Compounds 1–3. A mixture of
t-BusalpyH2 (187 mg, 0.6 mmol) for 1, t-Busal-4-MepyH2 (196
mg, 0.6 mmol) for 2, or t-Busal-5-MepyH2 (196 mg, 0.6 mmol)
for 3 in 15 cm3 of methanol containing NaOH (48 mg, 1.2 mmol)
for 1, or KOH (67 mg, 1.2 mmol) for 2 and 3 was added to solid
Ru2(O2CCH3)4Cl (142 mg, 0.3 mmol), and the brown suspension
was stirred overnight at room temperature. During this time, the
brown suspension became a dark-green solution with some precip-
itates. Then, the solution was filtered to remove any insoluble sol-
ids, and the filtrate was evaporated under reduced pressure. The
green solid was dissolved again in 18 cm3 of THF containing
18-crown-6-ether (317 mg, 1.2 mmol) and the obtained solution
was stirred for 10 min at room temperature. The filtered solution
was added into a narrow Schlenk tube for diffusion with 18 cm3
of toluene to form rectangular green crystals. Other diffusion sol-
6þ
of Ru2 to Ru27þ. We describe herein the syntheses, struc-
tures, and electrochemistry of these compounds.
Experimental
Materials and Reagents. Syntheses of diruthenium com-
pounds were carried out under a dried nitrogen atmosphere. All
chemicals used for the syntheses were of reagent grade. The di-
ruthenium(II, III) compound Ru2(O2CCH3)4Cl was synthesized
according to a method in the literature.13 All solvents used for
the syntheses of the Ru2 compounds and their electrochemical
and spectroscopic studies were dried by refluxing over common

614 Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
Ru2 Unit Capped by Di-tert-butylphenolate
vents, such as n-hexane or diethyl ether, also resulted in the same
products, but the purity of the sample was higher with toluene, al-
though the yield seemed to be quite low. For 1, yield 66 mg, 16%.
Anal. Calcd for C64H98N4O14NaRu2: C, 55.99; H, 7.20; N, 4.08%.
Found: C, 55.41; H, 7.06; N, 3.91%. Mass spectral data (m=z): cal-
culated for C44H58N4O6Ru2 941.3; MALDI-TOF found 941.6,
886.9, 825.7 ([Ru2(O2CCH3)2(t-Busalpy)2]ꢁ, [Ru2(O2CCH3)-
(t-Busalpy)2], [Ru2(t-Busalpy)2]þ). IR (KBr, cmꢁ1): 3413 (w,
br), 2949 (m), 2903 (m), 1603 (m), 1468 (s), 1435 (s), 1352 (m),
refined anisotropically, whereas the hydrogen atoms were refined
isotropically. For full-matrix least-squares refinements based on
the I > 2:00ꢀðIÞ reflections and the observed reflections, the un-
weighted and weighted agreement factors of R1 ¼ ꢂjjFoj ꢁ
2
2
2
jFcjj=ꢂjFoj (I > 2:00ꢀðIÞ), R2 ¼ ꢂðFo ꢁ Fc Þ=ꢂFo (all data),
2
2
2
1=2
2
and Rw ¼ ½ꢂwðFo ꢁ Fc Þ =ꢂwðFoÞ ꢆ
(all data) were used.
The weighting scheme was based on counting statistics. All calcu-
lations were performed using the teXsan crystallographic software
package of Molecular Structure Corporation.17 The crystal data
and details of the structure determinations for 1–3 are summarized
in Table 1. Crystallographic data (excluding structure factors) for
the structures of 1–3 have been deposited at the Cambridge Data
Centre as supplementary publication Nos. CCDC-286435 for 1,
CCDC-286433 for 2, and CCDC-286434 for 3. 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, UK (fax:
+44 1223 336033; e-mail: deposit@ccdc.cam.ac.uk).
1286 (m), 1109 (s), 837 (w). UV–vis (ꢃmax/nm ("/Mꢁ1 cmꢁ1
)
in MeCN): 451 (5054), 592 (5573), 1046 (1304). For 2 [K(18-
ꢂ
crown-6)(CH3O)(H2O)], yield 90 mg, 19%. Anal. Calcd for
C79H131K2N4O22Ru2: C, 53.62; H, 7.46; N, 3.17%. Found: C,
53.82; H, 7.17; N, 3.08%. Mass spectral data (m=z): calculated
for C46H62N4O6Ru2 969.3; MALDI-TOF found 969.5, 914.9,
855.1 ([Ru2(O2CCH3)2(t-Busal-4-Mepy)2]ꢁ, [Ru2(O2CCH3)-
(t-Busal-4-Mepy)2], [Ru2(t-Busal-4-Mepy)2]þ). IR (KBr, cmꢁ1):
3385 (w, br), 2949 (s), 2901 (s), 1614 (s), 1462 (s), 1429 (s),
1352 (m), 1286 (m), 1109 (s), 964 (m), 827 (m). UV–vis (ꢃmax/nm
("/Mꢁ1 cmꢁ1) in MeCN): 294 (27531), 442 (4198), 591 (5106),
Results and Discussion
1052 (565). For 3, yield 50 mg, 9%. Anal. Calcd for C79H131
-
Syntheses and General Properties. The protonated tri-
dentate ligands that correspond to the amine species, t-Busal-
R0pyH2, were obtained by reducing the corresponding Schiff-
base compounds with NaBH4 in methanol. Completion of the
reduction was easily confirmed by a change in color of the
solution from light yellow to colorless. Crystallization in a
methanol/water mixture produced well-shaped crystals of t-
Busal-R0pyH2 (R0 ¼ H, 4-Me, and 5-Me). In the IR spectra, the
vibration of C=N (imine; ꢅ = ca. 1600 cmꢁ1) in Schiff-base
compounds was no longer detected in the reduced species. In
K2N4O22Ru2: C, 53.62; H, 7.46; N, 3.17%. Found: C, 53.33;
H, 7.28; N, 3.33%. Mass spectral data (m=z): calculated for
C46H62N4O6Ru2 969.3; MALDI-TOF found 969.3, 910.9, 852.6
([Ru2(O2CCH3)2(t-Busal-5-Mepy)2]ꢁ, [Ru2(O2CCH3)(t-Busal-5-
Mepy)2], [Ru2(t-Busal-5-Mepy)2]þ). IR (KBr, cmꢁ1): 3454 (w),
2949 (m), 2899 (m), 1618 (m), 1578 (m), 1481 (m), 1437 (m),
1400 (m), 1352 (m), 1288 (m), 1109 (s), 962 (m), 843 (w). UV–
vis (ꢃmax/nm ("/Mꢁ1 cmꢁ1) in MeCN): 282 (32957), 465 (5792),
609 (5873), 1041 (1755).
Physical Measurements.
Infrared spectra were measured
1
the H NMR spectra, a triplet corresponding to amino proton
on KBr disks with a Shimadzu FT-IR 8600 spectrophotometer.
MALDI-TOF mass spectra were acquired on a Kratos Analytical
KOMPACT PROBE. General magnetic susceptibility data for
ground samples were measured over the temperature range of
1.9–300 K using an MPMS-XL SQUID susceptometer (Quantum
Design, Inc.), where the applied magnetic fields were 1 T. Correc-
tions were applied for diamagnetism using Pascal’s constants14
and for vinyl capsule wrapping samples. The UV–vis spectrum
was measured in degassed acetonitrile in a septum-sealed cell us-
ing a Hitachi U-3500 spectrophotometer. Cyclic voltammograms
(CVs) and differential pulse voltammograms (DPVs) were record-
ed in 1,2-dichloroethane (tetra-n-butylammonium hexafluorophos-
phate [n-Bu4N(PF6)] = 0.2 M as supporting electrolyte) under a
nitrogen atmosphere with the use of an ALS/CHI Instruments
Electrochemical Analyzer (model CHI 600A). At the beginning
of measurement of the compounds, the CVs of 1,2-dichloroethane
with only supporting electrolyte were measured. To this solution
were added the compounds ([Compound] = 2 ꢃ 10ꢁ3 M) and
measured with one unit of a carbon working electrode, a Pt coun-
ter electrode, and a Ag/AgNO3 reference electrode. Finally, CV
potentials were justified by using the ferrocene/ferrocenium cou-
ple as an internal reference (Fc=Fcþ ¼ 0:194 V (ꢀE ¼ 123 mV)
in 1,2-dichloroethane vs Ag/AgNO3).
(–NH) was newly observed in the reduced species at 5.07 ppm
for t-BusalpyH2, 4.98 ppm for t-Busal-4-MepyH2, and 4.94
ppm for t-Busal-5-MepyH2.
Bis-substitution of the acetato ligands of the paddlewheel-
type diruthenium compound Ru2(O2CCH3)4Cl with t-Busal-
R0pyH2 was accomplished under a methanolic condition in a
1:2 molar ratio of Ru2(O2CCH3)4Cl:t-Busal-R0pyH2 in the
presence of NaOH or KOH as the deprotonation reagent. The
addition of 18-crown-6-ether to form the Naþ- or Kþ-crown-
ether cationic complex led to the successful crystallization of
1–3. Thus, compounds 1–3 are composed of the Naþ- or Kþ-
crown-ether cation and diruthenium anionic species, as shown
in Scheme 1. The stability of the anionic components, [Ru2-
(O2CCH3)2(t-Busal-R0py)2]ꢁ, was first checked by MALDI-
TOF mass spectroscopy in an experimental nujol medium (see
Experimental). The main signal assigned to [Ru2(O2CCH3)2-
(t-Busal-R0py)2]ꢁ was detected at 941.6 (m=z, 941.3), 969.5
(m=z, 969.3), and 969.3 (m=z, 969.3), respectively, for 1–3,
with two characteristic minor signals attributed to the ace-
tate-eliminated species of [Ru2(O2CCH3)(t-Busal-R0py)2] and
[Ru2(t-Busal-R0py)2]þ, proving the formation of the present
bis-substituted compounds. Because a confirmation of the
magnetic data is useful for determining the electronic state
of a diruthenium unit, the magnetic susceptibilities of 1–3 were
measured in the temperature range of 1.9–300 K. For all com-
pounds, the ꢆT products at 300 K were found in the range of
1.88–1.97 cm3 K molꢁ1. Then, upon cooling, they gradually
decreased to finally reach minimum values (0.39–1.06
cm3 K molꢁ1). The room temperature ꢆT values are very sim-
X-ray Crystallographic Analyses. Measurements were made
on a Rigaku CCD diffractometer for 1 and on a Rigaku Imaging
Plate diffractometer for 2 and 3 with graphite monochromated
ꢁ
Mo Kꢄ radiation (ꢃ ¼ 0:71069 A). The data were collected at
ꢁ170 ꢄ 1 ꢅC for all compounds. An empirical absorption correc-
tion based on azimuthal scans of several reflections was applied.
The structures were solved by direct methods (SIR92)15 and ex-
panded using Fourier techniques.16 The non-hydrogen atoms were

H. Miyasaka et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
615
Table 1. Crystallographic Data of 1–3
1
2 2THF
ꢂ
3
Empirical formula
Fw/g molꢁ1
Space group
T/ꢅC
C64H98N4O14NaRu2
1372.830
C74H118KN4O16Ru2
1561.207
C79H131K2N4O22Ru2
1769.459
ꢃ
P1 (#2)
ꢃ
P1 (#2)
C2=c (#15)
ꢁ150ð1Þ
ꢁ170ð1Þ
ꢁ170ð1Þ
ꢁ
ꢃ/A
0.7107
21.6776(6)
0.7107
0.7107
ꢁ
a/A
11.017(2)
13.406(3)
15.162(3)
115.019(9)
92.691(8)
104.948(8)
1929.2(6)
1
1.344
5.11
8332 (I > 2:00ꢀðIÞ)
439
1.361
9.6966(5)
13.1709(7)
18.212(1)
77.362(2)
77.797(3)
85.157(2)
2216.5(2)
1
1.325
5.04
8984 (I > 2:00ꢀðIÞ)
505
1.883
ꢁ
b/A
16.5661(4)
20.9809(7)
90
103.6283(4)
90
7322.4(4)
4
1.245
ꢁ
c/A
ꢄ/deg
ꢇ/deg
ꢈ/deg
ꢁ 3
V/A
Z
Dcalcd/g cmꢁ3
ꢉ(Mo Kꢄ)/cmꢁ1
Unique reflns.
No. variables
GOF
4.77
8381 (I > 2:00ꢀðIÞ)
382
0.999
R1a)
0.0758 (I > 2:00ꢀðIÞ)
0.0480 (I > 2:00ꢀðIÞ)
0.1445 (all data)
0.0750 (I > 2:00ꢀðIÞ)
0.2411 (all data)
wR2
0.2301 (all data)
2
2 1=2
2
2
2
2
2
2
a) R1 ¼ ꢂjjFoj ꢁ jFcjj=ꢂjFoj. b) wR2 ¼ ½ꢂwðFo ꢁ Fc Þ =ꢂwðFo Þ ꢆ . c) w ¼ 1=½ꢀ2Fo þ ðpðmaxðFo2; 0Þ þ 2Fc Þ=3Þ ꢆ
2
2
2
2
for 1, where p ¼ 0:11300 for 1, 0.08900 for 2, w ¼ 1=½ꢀ2Fo þ ð0:1000PÞ ꢆ, where P ¼ ðFo þ 2Fc Þ=3 for 3.
Scheme 1.
ilar to those observed in this class of diruthenium compounds;
for example, 2.00–2.31 cm3 K molꢁ1 for [Ru2(O2CR)4]þ and
Ru2(O2CR)4Cl, possessing an S ¼ 3=2 spin ground state.18–27
The decrease of the ꢆT product should be principally due to
the contribution of zero-field splitting (ZFS) arising from the
4B2u electronic ground state.28,29 To take into account the
decrease of ꢆT, best-fitting using the S ¼ 3=2 paramagnetic
Van Vleck approximation with the ZFS contribution (D)30
and the inter-molecular interaction (zJ)31 was performed, and
the best agreement between the model and the experiment

616 Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
Ru2 Unit Capped by Di-tert-butylphenolate
was obtained with the following parameters: g ¼ 2:05, D ¼
57:3 cmꢁ1, zJ ¼ ꢁ0:06 cmꢁ1 (R ¼ 0:9981) for 1; g ¼ 2:02,
D ¼ 64:0 cmꢁ1, zJ ¼ ꢁ0:06 cmꢁ1 (R ¼ 0:9990) for 2; g ¼
2:08, D ¼ 53:5 cmꢁ1, zJ ¼ ꢁ0:88 cmꢁ1 (R ¼ 0:9990) for 3
logues A[Ru2(O2CCH3)2(5-Rsalpy)2] (R ¼ H, Me, Cl, Br, and
ꢁ
NO2) (2.28–2.30 A), and follows the trend 5-NO2 < 5-Cl,
5-Br < H ꢇ 5-Me ꢇ 1 < 2, 3, due to the electron-donating
ability of remote substituents. The typical value of the Ru–
2
(R ¼ 1 ꢁ ꢂfðꢆTcalc ꢁ ꢆTobsÞ=ꢂðꢆTobsÞg ). All of the above pa-
rameters are in good agreement with previously reported values
for paddlewheel-type diruthenium compounds (D ꢇ 40{90
cmꢁ1).18–27,29 Consequently, the electronic configuration of
2
ꢀ
3
ꢀ
1–3 could be ꢀ2ꢁ4ꢂ ðꢁ ꢂ Þ .
The UV–vis spectra of 1–3 were recorded in MeCN under
N2 (Fig. 1 and Table 2). All compounds exhibited relatively
strong bands in the UV–vis region at around 300 nm (sh),
380 nm (sh), 450 nm (" ꢇ 5000{6000 Mꢁ1 cmꢁ1), and 600 nm
(" ꢇ 6000 Mꢁ1 cmꢁ1), the last one of which is understandable
because of the intense green color of 1–3. In addition, a weak
band at around 1050 nm (" ꢇ 1000{1800 Mꢁ1 cmꢁ1) was also
observed. The overall spectral feature is very similar to that of
the family of A[Ru2(O2CCH3)2(5-Rsalpy)2].10
Structures of 1–3. t-Busal-R0py2ꢁ acts as a dianionic tri-
dentate ligand containing aminopyridyl and phenolate moieties
that are connected to each other by a methylene group, where
the former could attach to the diruthenium unit and the latter
could flexibly coordinate to the axial positions of the Ru–Ru
bond. All solid-state structures of the anionic components were
thus confirmed to have a similar geometrical form. ORTEP
drawings of the diruthenium anionic components of 1–3 are
depicted in Fig. 2. Selected bond distances and angles around
the diruthenium unit are summarized in Table 3. In this type of
compound, the equatorial ligands are located around a diruthe-
nium unit in trans-fashion; however, this is rare and most cases
show localization in cis-fashion.6,7,9 The Ru–Ru bond distance
ꢁ
gives a typical value of 2.301(1) A for 1, 2.3114(4) A for 2, and
ꢁ
ꢁ
2.3091(4) A for 3, which seems to be similar to those of ana-
1
Comp. 1
Comp. 2
Comp. 3
0.8
0.6
0.4
0.2
0
400
600
800
1000 1200 1400
Wavelength / nm
Fig. 2. Structure drawings of an asymmetrical anionic unit
of 1 (a), 2 (b), and 3 (c).
Fig. 1. UV–vis–NIR spectra of 1–3 in MeCN.
Table 2. UV–Vis–NIR Spectral Data of 1–3 in MeCN under N2
max/nm ("/103 Mꢁ1 cmꢁ1
ꢃ
)
Compd
Band I
Band II
Band III
Band IV
Band V
1
2
3
300 (sh)
290 (sh)
282 (sh)
374 (sh)
370 (sh)
380 (sh)
451 (5.0)
441 (4.9)
465 (5.8)
592 (5.6)
591 (5.9)
609 (5.9)
1046 (1.3)
1052 (1.3)
1041 (1.8)

H. Miyasaka et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
617
ꢁ
Table 3. Relevant Bond Distances (A) and Angles (deg) of
1–3 with Estimated Standard Deviations in Parentheses
1
2 2THF
3
ꢂ
ꢁ
Distances/A
2.3114(4)
2.194(2)
2.056(2)
2.037(2)
2.075(2)
2.034(2)
ꢀa)
Ru(1)–Ru(1)
Ru(1)–O(1)
Ru(1)–O(2)
Ru(1)–O(3)
Ru(1)–N(1)
Ru(1)–N(2)
2.301(1)
2.178(4)
2.035(5)
2.065(5)
2.056(6)
2.026(7)
2.3091(4)
2.193(3)
2.052(2)
2.074(2)
2.070(3)
2.037(3)
ꢀ
ꢀ
Angles/deg
174.17(5)
87.22(6)
91.16(6)
91.09(7)
88.77(7)
88.60(8)
93.10(8)
92.98(9)
87.49(9)
178.02(7)
89.95(8)
94.33(9)
88.93(9)
86.78(9)
175.70(9)
ꢀ
Ru(1) –Ru(1)–O(2)
Ru(1) –Ru(1)–O(1)
ꢀ
Ru(1) –Ru(1)–O(3)
172.5(1)
91.6(1)
87.0(1)
90.4(2)
89.3(2)
95.3(2)
86.2(2)
92.8(2)
87.9(2)
178.1(2)
88.4(2)
87.8(2)
90.3(2)
93.5(2)
176.2(2)
172.83(7)
90.59(8)
87.74(8)
90.88(9)
89.15(9)
95.9(1)
85.82(10)
92.2(1)
88.2(1)
178.2(1)
90.4(1)
86.3(1)
88.98(10)
94.3(1)
176.7(1)
ꢀ
Ru(1) –Ru(1)–N(1)
ꢀ
ꢀ
ꢀ
Ru(1) –Ru(1)–N(2)
ꢀ
O(1)–Ru(1)–O(2)
O(1)–Ru(1)–O(3)
O(1)–Ru(1)–N(1)
ꢀ
ꢀ
O(1)–Ru(1)–N(2)
O(2)–Ru(1)–O(3)
O(2)–Ru(1)–N(1)
O(2)–Ru(1)–N(2)
ꢀ
ꢀ
ꢀ
O(3) –Ru(1)–N(2)
ꢀ
O(3) –Ru(1)–N(1)
ꢀ
N(1) –Ru(1)–N(2)
ꢀ
ꢀ
a) Symmetry operations ( ): ꢁx þ 1=2, ꢁy þ 1=2, ꢁz for 1;
ꢁx, ꢁy, ꢁz þ 1 for 2; ꢁx þ 2, ꢁy, ꢁz for 3.
5þ
Ru bond distance is one proof of the Ru2 center with the
2
ꢀ
3
electronic configuration of ꢀ2ꢁ4ꢂ ðꢁ ꢂ Þ .18–27 As mentioned
above, the aminopyridyl moiety of the t-Busal-R0py2ꢁ ligand
bridges the diruthenium unit with bond distances of (Ru–
ꢀ
ꢁ ꢁ
Npyridine av = 2.067 A and (Ru–Naminato)av = 2.032 A, in which
)
the latter is slightly shorter than the former. The bond distances
between Ru and carboxylate oxygen are found in the range of
ꢁ
2.035–2.074 A. Two phenolate moieties as pendant ligands cap
the axial positions of the Ru–Ru bond with distances and an-
ꢀ
ꢁ
gles of Ru(1)–O(1) = 2.178(4) A and O(1)–Ru(1)–Ru(1)
=
172.5(1)ꢅ for 1, Ruꢅ(1)–O(1) = 2.194(2) A and O(1)–Ru(1)–
ꢁ
ꢀ
ꢁ
Fig. 3. Cyclic and differential pulse voltammograms of 1–3
in 1,2-dichloroethane containing 0.2 M TBA(PF6) under
N2. The arrow in the DPV displays the sweep direction
from the rest potential.
Ru(1) = 174.17(5) for 2, and Ru(1)–O(1) = 2.193(3) A and
O(1)–Ru(1)–Ru(1) = 172.83(7)ꢅ for 3. The Ru–O distance
ꢀ
is slightly shorter than that in A[Ru2(O2CCH3)2(5-Rsalpy)2]
(R ¼ H, Me, Cl, Br, and NO2) with the trend of 5-NO2 ꢇ H >
5-Cl, 5-Br > 5-Me > 2, 3 > 1; the remote substituent effect
can also be seen here.
ꢁ1:6 V as a reversible wave (Ia=Ic ꢇ 1, ꢀEp ꢇ 90 mV), and
Electrochemistry of 1–3. The electronic behavior of 1–3
was confirmed by cyclic voltammetry (CV) and differential
pulse voltammetry (DPV) in 1,2-dichloroethane containing n-
Bu4N(PF6) as the supporting electrolyte (vs Ag/AgNO3). Both
kinds of voltammograms for 1–3 are shown in Fig. 3, and the
obtained electrochemical data are summarized in Table 4
together with those of the family of A[Ru2(O2CCH3)2(5-
Rsalpy)2] (R ¼ H, Cl, Br, Me, and NO2). All compounds re-
vealed rich redox chemistry and basically identical redox proc-
esses that included principally one reduction and four oxida-
tions. Some additional redox waves representative of struc-
ture-modified species were noted in 3 during measurements
(vide infra). The one-electron reduction was observed around
was assigned to a metal-based 1eꢁ reduction of Ru25þ/Ru2
.
4þ
For the oxidation processes, the first and second couples were
5þ
7þ
assigned to the metal-based ones, Ru26þ/Ru2 and Ru2
/
Ru26þ, respectively. These values were shifted more nega-
tively than the corresponding values for [Ru2(O2CCH3)2-
(5-Rsalpy)2]ꢁ. Indeed, the one-electron oxidation of 1 to
6þ
form the Ru2 species occured at ꢁ0:469 V, which was more
negative than ꢁ0:371 V of [Ru2(O2CCH3)2(5-Mesalpy)2]ꢁ,
ꢁ0:326 V of [Ru2(O2CCH3)2(salpy)2]ꢁ, or ꢁ0:238 V of [Ru2-
(O2CCH3)2(5-Clsalpy)2]ꢁ. The redox shifts should be due
to the ligand-substituent effect on the R-group in the [Ru2-
(O2CCH3)2(R-salpy)2]ꢁ family, where there could be a strong
electron-donating effect when R = 3,5-di-t-butyl group. There-

618 Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
Ru2 Unit Capped by Di-tert-butylphenolate
Table 4. Electrochemical Data of 1–3 Measured in 1,2-Dichloroethane Containing 0.2 M TBA(PF6) under N2 (V vs Ag/
AgNO3)a)
e)
A
B
C
D
L/Lꢁ
E
L/Lꢁ
ꢀE1=2(E–D)
mVd)
Kc
4þ
5þ
6þ
Compd
Ru25þ/Ru2
Ru26þ/Ru2
Ru27þ/Ru2
1
ꢁ1:576b)
ꢁ0:469b)
0.274b)
(0.272)c)
0.248b)
0.806f)
(0.747)c)
0.790f)
0.902f)
(0.843)c)
0.858f)
(0.806)c)
—
96
42
15
(ꢁ1:577)c)
ꢁ1:613b)
(ꢁ1:613)c)
ꢁ1:682b)
(ꢁ1:676)c)
ꢁ1:511b)
(ꢁ1:490)c)
ꢁ1:504b)
(ꢁ1:506)c)
ꢁ1:446b)
(ꢁ1:440)c)
ꢁ1:427b)
(ꢁ1:417)c)
ꢁ1:282b)
(ꢁ1:274)c)
(ꢁ0:471)c)
ꢁ0:491b)
(ꢁ0:494)c)
ꢁ0:506b)
(ꢁ0:509)c)
ꢁ0:326b)
(ꢁ0:316)c)
ꢁ0:371b)
(ꢁ0:381)c)
ꢁ0:238b)
(ꢁ0:254)c)
ꢁ0:227b)
(ꢁ0:218)c)
ꢁ0:029b)
(ꢁ0:029)c)
2
3
70
(0.247)c)
0.205f)
(0.736)c)
0.818
—
—
(0.144)c)
0.281b)
(0.818)c)
0.867b)
H
—
—
—
192
—
(0.252)c)
0.259b)
(0.810)c)
0.704b)
5-Meg)
5-Clg)
5-Brg)
0.864f)
(0.810)c)
—
135
—
(0.220)c)
0.355b)
(0.675)c)
0.982b)
(0.312)c)
0.361b)
(0.886)c)
0.977b)
—
—
—
—
(0.330)c)
0.585b)
(0.908)c)
1.205b)
g)
5-NO2
—
—
(0.394)c)
(1.082)c)
a) The ferrocene/ferrocenium couple, Fc/Fcþ ¼ 0:194 V (Ic=Ia ꢇ 1, ꢀEp ¼ 123 mV), was observed at the same con-
dition described in the Experimental Section in the text. b) Half-wave potentials (E1=2’s) from cyclic voltammograms.
The couples of A, B, and C were estimated because of their reversibility (Ic=Ia ꢇ 1, ꢀEp ꢇ 100 mV at a scan rate of
0.03 V sꢁ1). c) Peak potentials from differential pulse voltammograms (scan rate, 8 mV sꢁ1; pulse width, 60 ms; pulse
1{2
period, 120 ms). d) ꢀE1=2 were calculated by using the DPV peak half-height method of Richardson and Taube.32
e) Comproportionation constant (Kc), calculated by using the equation of Kc ¼ expðꢀE1=2=25:69Þ (at 298 K).32
f) Anode peaks because of their irreversibility. g) The electrochemical data have been measured at an acetonitrile solu-
tion quoted in Ref. 10b.
fore, the respective redox potentials were plotted as a func-
tion of the summation of the Hammett constants (ꢀ), and are
shown in Fig. 4a. The respective potentials exhibited a linear
relationship.
ration, as shown in Fig. 3. However, their values may include,
even if minimally, the electronic effect of the methyl group
attached to the pyridyl moiety of the ligands; i.e., the R0-sub-
stituent effect on the 4-position (trans vs pyridine N) in 2, and
the 5-position (trans vs pyridylamine) in 3. Considering the
Hammett substituent constants (ꢀ) of H, p-Me, and m-Me for
1–3, respectively, the respective redox potentials on the Ru2
center and the ligand (a to e in Fig. 3) have a linear relation-
ship as a function of 2ꢀ, where 2 is the number of substituents
per unit (Fig. 4b). Unfortunately, the difference in the compro-
portionation constants among 1, 2, and the R ¼ 5-Me species
could not be well understood from the relationship of substitu-
ent donation.
The third and fourth redox couples corresponded to the 1eꢁ
oxidation of individual t-Busal-R0py2ꢁ ligands. Interestingly,
when R ¼ H, 5-Cl, 5-Br, and 5-NO2 in R-salpy, the redox
for the corresponding ligand was observed as one couple of
two one-electron oxidations, and when R ¼ 5-Me, each oxida-
tion was detected separately with a peak-to-peak separation of
ꢀp ¼ 135 mV on DPV (CV did not clearly show the peak sep-
aration).10b The finding suggests that the electron-donating
ability of the methyl and t-butyl substituted groups affected
the HOMO/LUMO energy levels of not only the ligand, but
also the diruthenium center. From the DPV data, the peak-
to-peak separations between the third and fourth peaks of 1
As mentioned above, repeated redox cycling in the measur-
ing window led to the appearance of new waves around 0 and
ꢀ
waves were not formed when the measurements were repeated
0.5 V, especially as shown in Fig. 3c for 3 ( marked). These
and 2 were 100 and 76 mV, respectively. A more accurate cal-
1ox{2ox
culation of ꢀE1=2
was accomplished using the DPV
at a range lower than ꢁ0:3 V that includes redox reactions of
4þ
peak half-height method of Richardson and Taube to give
96 mV for 1 and 70 mV for 2.32 Therefore, these values lead to
a comproportionation constant, Kc, of 42 for 1 and 15 for 2,
for the redox reaction Kc ¼ expðꢀE1=2=25:69Þ at 298 K.33 This
constant could be related to the degree of electronic communi-
Ru25þ/Ru2 and Ru26þ/Ru25þ. Therefore, during oxidation
7þ
to Ru2 and/or redox reactions on the ligands, some of the
molecules would be modified to form axial-free species. Nev-
ertheless, since a redox on the t-Busal-5-Mepy2ꢁ ligand is ob-
served as a reversible wave, even at 0.8 V, the mono-capped
species, as shown in Scheme 2, is plausible. Although the
non-capped species, of which both phenoxido groups are coor-
dination-free, could be considered, the following results dis-
miss such an idea concerning the presence of the non-capped
species: (i) the half-height width of DPV peak at 0.8 V in 3
is approximately half of the corresponding values in 1 and 2,
indicating one-electron oxidation per the mono-capped spe-
7þ
cation between ligands, probably via the Ru2 unit. However,
the calculated values are basically small, so that the degree of
electronic communication between ligands seems to be low,
being Class I in the Robin–Day classification for electron
delocalization.33 The Kc values are slightly different among
1, 2, and the R ¼ 5-Me species. We cannot exclude the possi-
bility of estimation errors because of the indefinite peak sepa-

H. Miyasaka et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
619
(t-Busal-4-Mepy2ꢁ), and N-(5-methyl-2-pyridyl)-2-oxido-3,5-
di-tert-butylbenzylaminate (t-Busal-5-Mepy2ꢁ), have been pre-
pared, and the corresponding diruthenium compounds, [Na(18-
crown-6)(thf)2][Ru2(O2CCH3)2(t-Busalpy)2] (1), [K(18-crown-
6)(thf)2][Ru2(O2CCH3)2(t-Busal-4-Mepy)2] (2), and [K(18-
crown-6)(thf)(H2O)][K(18-crown-6)(thf)(MeO)][Ru2(O2CCH3)2-
(t-Busal-5-Mepy)2] (3), have been synthesized and isolated,
thanks to the large counter cation of the Na- or K-18-crown-
6-ether complex that stabilizes the crystallinity of the mate-
rials. Compounds 1–3 have a Ru–Ru bonded core with an
2
ꢀ
3
ꢀ
electronic configuration of ꢀ2ꢁ4ꢂ ðꢁ ꢂ Þ , which is axial-
capped by the di-t-butylphenoxido groups of the t-Busal-
R0py2ꢁ ligands. The electrochemistry of these compounds re-
veals multi-redox properties on both the Ru2-center and the
t-Busal-R0py2ꢁ ligands. Compared with the previously report-
ed family of [Ru2(O2CCH3)2(5-Rsalpy)2]ꢁ (R ¼ H, Me, Cl,
Br, and NO2), the strong electron-donating ability of the t-
butyl groups compared with the other R-groups leads to signif-
icant shifts of the redox potentials according to the Hammett
law. With the establishment of electronic communication be-
tween phenoxido groups via the Ru–Ru bond as our objective,
we observed a sign of electronic communication in 1 and 2, as
well as the R ¼ Me compound reported previously, albeit
small. In addition, the introduction of a methyl group to the
pyridine of the ligands (for 2 and 3) does not provide any
meaningful substituent effects for electronic communication.
A further ligand modification and exchange of metal ion to
control the HOMO/LUMO energy levels between the metal
center and the ligand are in progress.
This work was partially supported by CREST project, the
Japan Science and Technology Agency (JST) and a Grant-in-
Aid for Scientific Research from the Ministry of Education,
Culture, Sports, Science and Technology, Japan.
Fig. 4. Hammett plots of redox potentials (vs Ag/AgNO3)
on [Ru2(O2CCH3)2(5-Rsalpy)2]ꢁ (R ¼ H, Me, Cl, Br, and
NO2)10 and 1 (a), where ꢂꢀ is the summation of Hammett
constants for all R-substituents, and on 1–3 (b), where ꢀ is
the Hammett constant for Me group (R0) on t-Busal-R0py
and 2 is the number of substituents. For the referred poten-
tials, the peak potentials in differential pulse voltammo-
grams are used.
References
1
a) F. A. Cotton, C. B. Harris, Inorg. Chem. 1965, 4, 330.
b) F. A. Cotton, Inorg. Chem. 1965, 4, 334.
F. A. Cotton, R. A. Walton, Multiple Bonds between Metal
Atoms, Oxford University Press, Oxford, 1993.
a) C. Lin, J. D. Protasiewicz, E. T. Smith, T. Ren, J. Chem.
2
3
Soc., Chem. Commun. 1995, 2257. b) C. Lin, J. D. Protasiewicz,
E. T. Smith, T. Ren, Inorg. Chem. 1996, 35, 6422. c) C. Lin,
J. D. Protasiewicz, T. Ren, Inorg. Chem. 1996, 35, 7455. d) C.
Lin, T. Ren, E. J. Valente, J. D. Zubkowski, E. T. Smith, Chem.
Lett. 1997, 753. e) K. M. Carlson-Day, J. L. Eglin, C. Lin, L. T.
Smith, R. J. Staples, D. O. Wipf, Polyhedron 1999, 18, 817.
f) N. G. Connelly, P. R. G. Davis, E. E. Harry, P. Klangsinsirikul,
M. Venter, J. Chem. Soc., Dalton Trans. 2000, 2273. g) K. M.
Kadish, L. Wang, A. Thuriere, E. V. Caemelbecke, J. L. Bear,
Inorg. Chem. 2003, 42, 834.
Scheme 2.
cies, (ii) the oxidation coupling on free-phanolate group does
not appear at 0.8 V, rather showing a lower potential as a non-
reversible wave. It is however uncertain why only 3 shows
such a modification.
4
W. R. Tikkanen, E. Binamira-Soriaga, W. C. Kaska, P.
Ford, Inorg. Chem. 1984, 23, 141.
A. T. Baker, W. R. Tikkanen, W. C. Kaska, P. Ford, Inorg.
Chem. 1984, 23, 3254.
a) C. S. Campos-Fernandez, X. Ouyang, K. R. Dunbar,
5
´
6
Summary
´
Inorg. Chem. 2000, 39, 2432. b) C. S. Campos-Fernandez, L. M.
Thomson, J. R. Galan-Mascaros, X. Ouyang, K. R. Dunbar, Inorg.
Three new tridentate bridging/chelating ligands, N-(2-pyri-
dyl)-2-oxido-3,5-di-tert-butylbenzylaminate (t-Busalpy2ꢁ), N-
(4-methyl-2-pyridyl)-2-oxido-3,5-di-tert-butylbenzylaminate
´
´
Chem. 2002, 41, 1523.
7
L. Bear, Y. Li, B. Han, E. V. Caemelbecke, K. M. Kadish,

620 Bull. Chem. Soc. Jpn. Vol. 79, No. 4 (2006)
Inorg. Chem. 2001, 40, 182.
Ru2 Unit Capped by Di-tert-butylphenolate
´
´
26 M. C. Barral, R. Gonzalez-Prieto, R. Jumenez-Aparicio,
J. L. Priego, M. R. Torres, F. A. Urbanos, Eur. J. Inorg. Chem.
2003, 2339.
8
L. Bear, Y. Li, B. Han, E. V. Caemelbecke, K. M. Kadish,
Inorg. Chem. 1996, 35, 3053.
T. Ren, V. DeSilva, G. Zou, C. Lin, L. M. Daniels, C. F.
´
´
9
27 G. Arribas, M. C. Barral, R. Gonzalez-Prieto, R. Jimenez-
Aparicio, J. L. Priego, M. R. Torres, F. A. Urbanos, Inorg. Chem.
2005, 44, 5770.
Campana, J. C. Alvarez, Inorg. Chem. Commun. 1999, 2, 301.
10 a) H. Miyasaka, C. Kachi-Terajima, T. Ishii, M.
Yamashita, J. Chem. Soc., Dalton Trans. 2001, 1929. b) H.
Miyasaka, T. Izawa, K. Sugiura, M. Yamashita, Inorg. Chem.
2003, 42, 7683.
11 H. Miyasaka, K. Sugiura, M. Yamashita, Inorg. Chem.
Commun. 2003, 6, 1078.
12 P. Chaudhuri, K. Wieghardt, Prog. Inorg. Chem. 2001, 50,
151.
13 R. W. Mitchell, A. Spencer, G. Wilkinson, J. Chem. Soc.,
Dalton Trans. 1973, 846.
14 E. A. Boudreaux, L. N. Mulay, Theory and Applications of
Molecular Paramagnetism, John Wiley & Sons, New York, 1976,
pp. 491–495.
15 A. Altomare, M. C. Burla, M. Camalli, M. Cascarano, C.
Giacovazzo, A. Guagliardi, G. Polidori, J. Appl. Crystallogr.
1994, 27, 435.
28 J. G. Norman, Jr., G. E. Renzoni, D. A. Case, J. Am. Chem.
Soc. 1979, 101, 5256.
29 W.-Z. Chen, F. A. Cotton, N. S. Dalal, C. A. Murillo, C. M.
Ramsey, T. Ren, X. Wang, J. Am. Chem. Soc. 2005, 127, 12691.
30 If the main contributor to the decrease of the magnetic
moment is the ZFS effect, the following expressions for the mag-
netic susceptibility of S ¼ 3=2 are applicable:
Ng2ꢇ2 1 þ 9eꢁ2x
ꢆk ¼
;
ð1Þ
kBT 4ð1 þ eꢁ2x
Þ
3
4 þ ð1 ꢁ eꢁ2x
Þ
Ng2ꢇ2
kBT
x
ꢆ?
¼
;
ð2Þ
4ð1 þ eꢁ2x
Þ
where x is D=ðkBTÞ and D is the magnitude of the ZFS. The aver-
age molar magnetic susceptibility is given as follows:
16 P. T. Beurskens, G. Admiraal, G. Beurskens, W. P.
Bosman, R. de Gelder, R. Israel, J. M. M. Smits, The DIRDIF
Program System, Technical Report of the Crystallography Labora-
tory, University of Nijmegen, The Netherlands, 1994.
17 Crystal Structure Analysis Package, Molecular Structure
Corporation, The Woodlands, TX, 1985 and 1992.
18 M. A. S. Aquino, Coord. Chem. Rev. 1998, 170, 141.
ꢆk þ 2ꢆ?
ꢆtotal
¼
:
ð3Þ
3
The contributions of temperature-independent paramagnetism
(TIP) and impurities as Ru(III) or Ru2(III, III) species were ex-
cluded for simplicity, although those have been included in some
cases of Ru2(II, III) materials reported so far.
31 Using the mean-field approximation to treat the intermo-
lecular interactions, the following definition of the susceptibility
has been used:
´
19 R. Jimenez-Aparicio, F. A. Urbanos, J. M. Arrieta, Inorg.
Chem. 2001, 40, 613.
20 E. J. Beck, K. D. Drysdale, L. K. Thompson, L. Li, C. A.
Murphy, M. A. S. Aquino, Inorg. Chim. Acta 1998, 279, 121.
ꢆtotal
2zJ
Ng2ꢉ2B
ꢆ ¼
;
´
´
21 M. C. Barral, R. Jimenez-Aparicio, D. Perez-Quintanilla,
J. L. Priego, E. C. Royer, M. R. Torres, F. A. Urbanos, Inorg.
Chem. 2000, 39, 65.
1 ꢁ
ꢆtotal
22 F. D. Cukiernik, A.-M. Giroud-Godquin, P. Maldivi, J.-C.
Marchon, Inorg. Chim. Acta 1994, 215, 203.
23 J. Telser, R. S. Drago, Inorg. Chem. 1984, 23, 3114.
24 F. D. Cukiernik, D. Luneau, J.-C. Marchon, P. Maldivi,
Inorg. Chem. 1998, 37, 3698.
where z is the number of neighboring molecules that are interact-
ing with a molecule. See for example: a) B. E. Myers, L. Berger,
S. Friedberg, J. Appl. Phys. 1969, 40, 1149. b) C. J. O’Connor,
Prog. Inorg. Chem. 1982, 29, 203.
32 D. E. Richardson, H. Taube, Inorg. Chem. 1981, 20, 1278.
33 M. B. Robin, P. Day, Adv. Inorg. Chem. Radiochem. 1967,
10, 247.
´
´
25 H. Miyasaka, R. Clerac, C. S. Campos-Fernandez, K. R.
Dunbar, Inorg. Chem. 2001, 40, 1663.