Specific C-C coupling of the labile diruthenium bridging methylene complex, Cp2Ru2(μ-CH2)(CO)2(MeCN), with diazoalkanes (R2C=N2) leading to alkenyl complexes, Cp2Ru2(μ-CH=CR2)(μ-H)(CO)2, and alkenes, CH2=CR2

Article abstract of DOI:10.1016/S0022-328X(98)00775-X

Treatment of the labile μ-methylene species, Cp₂Ru₂(μ-CH₂)(CO)₂(MeCN) 1, with diazoalkanes results in C-C coupling of the alkylidene units. The reaction of monosubstituted diazoalkanes (N₂=CHR) and 4-t-butyldiazocyclohexane gives the μ-alkenyl-μ-hydride complexes, Cp₂Ru₂(μ-CH=CR₂)(μ-H)(CO)₂ 2, as sole products, whereas the reaction of disubstituted aryldiazoalkanes (N₂=CR¹R²) produces olefins (CH₂=CR¹R²) together with mixtures of unidentified organometallic compounds. Labeling experiments using 1-d₂ (μ-CD₂) and 1-¹³C (μ-¹³CH₂) reveal that the methylene bridge in 1 is converted to the α-CH and μ-H parts in 2 and the alkylidene part of diazoalkane is incorporated into the β-CR₂ part of the alkenyl bridge. The olefinic products should come from a η²-olefin intermediate, Cp₂Ru₂(η²CH₂=CHR)(CO) ₂ 7, resulting from coupling of the two alkylidene fragments within the coordination sphere of the diruthenium core. It is notable that C-H oxidative addition in 7 takes place prior to free rotation of the η²-olefin ligand as judged by the regiospecificity of the labeling experiments. Thus specific coupling of alkylidene units proceeds under mild conditions and this C-C coupling reaction serves as a model system for the carbon chain propagation step of surface-catalyzed reactions such as catalytic CO hydrogenation.

Full text of DOI:10.1016/S0022-328X(98)00775-X

Journal of Organometallic Chemistry 569 (1998) 71–83
Specific C–C coupling of the labile diruthenium bridging methylene
complex, Cp2Ru2(v-CH2)(CO)2(MeCN), with diazoalkanes (R2CꢀN2)
leading to alkenyl complexes, Cp2Ru2(v-CHꢀCR2)(v-H)(CO)2, and
1
alkenes, CH2ꢀCR2
Munetaka Akita *, Ruimao Hua, Selby A.R. Knox1, Yoshihiko Moro-oka1,
Sadahiro Nakanishi, Michael I. Yates
Research Laboratory of Resources Utilization, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan
Received 14 March 1998
Abstract
Treatment of the labile v-methylene species, Cp2Ru2(v-CH2)(CO)2(MeCN) 1, with diazoalkanes results in C–C coupling of the
alkylidene units. The reaction of monosubstituted diazoalkanes (N2ꢀCHR) and 4-t-butyldiazocyclohexane gives the v-alkenyl-v-
hydride complexes, Cp2Ru2(v-CHꢀCR2)(v-H)(CO)2 2, as sole products, whereas the reaction of disubstituted aryldiazoalkanes
(N2ꢀCR1R2) produces olefins (CH2ꢀCR1R2) together with mixtures of unidentified organometallic compounds. Labeling experi-
ments using 1-d2 (v-CD2) and 1-13C (v-13CH2) reveal that the methylene bridge in 1 is converted to the h-CH and v-H parts in
2 and the alkylidene part of diazoalkane is incorporated into the i-CR2 part of the alkenyl bridge. The olefinic products should
come from a p2-olefin intermediate, Cp2Ru2(p2CH2ꢀCHR)(CO)2 7, resulting from coupling of the two alkylidene fragments within
the coordination sphere of the diruthenium core. It is notable that C–H oxidative addition in 7 takes place prior to free rotation
of the p2-olefin ligand as judged by the regiospecificity of the labeling experiments. Thus specific coupling of alkylidene units
proceeds under mild conditions and this C–C coupling reaction serves as a model system for the carbon chain propagation step
of surface-catalyzed reactions such as catalytic CO hydrogenation. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: C–C coupling; Diruthenium; Alkenyl complexes
1. Introduction
products. Of the four elementary steps, steps 1, 2 and 4
have been successfully reproduced with model com-
plexes but a limited number of examples have been
reported for the third step.
During the last two decades much effort has been
focused on elucidation of mechanisms of catalytic CO
hydrogenation [1] using discrete transition metal com-
plexes [2]. The key features of the proposed mecha-
nisms can be summarized as follows: (1) dissociative
adsorption of CO leading to carbide intermediates; (2)
formation of CHX species via H-transfer; (3) C–C bond
formation involving the CHX species; (4) elimination of
In our laboratories, diruthenium bridging alkylidene
complexes, Cp2Ru2(v-CR2)(CO)3 and Cp2Ru2(v-
CR2)2(CO)2, have been the subject of recent research
activity [3,4]. In particular, the latter type of compound
is a system suitable for examination of the C–C cou-
pling process of alkylidene species. Although in-
tramolecular C–C bond formation was actually
observed upon thermolysis (Scheme 1), the selectivity
was not high and low yield mixtures of hydrocarbons
were obtained ([3]t, [4]a) (Scheme 1).
* Corresponding author. Fax: +81 45 9245226.
1 Dedicated to Professor Akira Nakamura in recognition of his
important contribution to organometallic chemistry.
0022-328X/98/$ - see front matter © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00775-X

72
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
Scheme 1.
transition metal coordination sphere ([3]q). We have
reported selective reactions of Cp2Ru2(v-
Polynuclear bridging alkylidene complexes (Mn(v-
CR2)) are studied extensively as models for surface-
bound alkylidene species [5]. As far as the structural
aspects are concerned, bridging alkylidene complexes
are recognized as better models for the surface-bound
species ([1]e) compared to mononuclear alkylidene com-
plexes [M=CR2]. But previous studies revealed that, in
general, polynuclear complexes were sluggish with re-
spect to C–C bond formation, whereas mononuclear
alkylidene complexes show versatile reactivities such as
stoichiometric and catalytic cyclopropanation, olefin
metathesis and acetylene polymerization [6].
CH2)(CO)2(MeCN) 1 bearing the lightly stabilizing
MeCN ligand. Addition of a substrate (e.g. allene or
hydrosilane) to 1 caused replacement of the MeCN
ligand to induce selective transformations (C–C cou-
pling ([3]s) and oxidative addition ([4]e), respectively)
(Scheme 2).
Herein we report details of the specific C–C coupling
reaction of 1 with diazoalkanes [8]. When a CR2 species
is incorporated into the diruthenium v-methylene sys-
tem in this way, the two alkylidenes readily couple with
The difference in the reactivities can be explained in
terms of extent of back donation to the vacant p orbital
of the CR2 fragment [7]. As can be seen from the above
chart, the overlap with the filled d orbitals is more
effective for dinuclear complexes and, therefore, their
M–C bonds bear much | bond character. In contrast,
less effective back donation in the mononuclear com-
plexes renders the alkylidene carbon atom electrophilic
enough to react with unsaturated hydrocarbons. A
typical C–C bond formation reaction of electrophilic
bridging methylene complexes is the migratory insertion
of a (cationic) alkyl-p1-alkylidene complex to give a
new homologated alkyl complex. This reaction is ana-
lyzed as a nucleophilic attack of the metal-bound car-
each other to give olefinic products under very mild
conditions.
2. Results and discussion
2.1. Reaction of the v-methylene complex
Cp2Ru2(v-CH2)(CO)2(NCMe) 1 with diazoalkanes
(N2ꢀCR2)
Reaction of the mono-v-methylene complex
Cp2Ru2(v-CH2)(CO)2(NCMe) 1 with a diazoalkane in
CH2Cl2 [9] proceeded smoothly at room temperature
(r.t.) to give olefinic products as summarized in Scheme
3. Upon addition of the diazoalkane to a CH2Cl2
solution of 1, N2 gas evolution was observed and the
reaction was completed within a few minutes. Separa-
tion by preparative TLC afforded three types of prod-
banion (R−) on the vacant
p orbital of the
p1-alkylidene ligand.
The disadvantages of v-alkylidene complexes can be
overcome through incorporation of a substrate into the

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
73
Scheme 2.
electron-withdrawing substituents, produced the brown
dinuclear adduct 4 which retained the N2 moiety, as
indicated by elemental analysis. The adduct 4 was
formed via replacement of the MeCN ligand. Although
it was anticipated that subsequent thermolysis might
induce removal of N2 and then C–C coupling, heating
a benzene-d6 solution of 4 at 60°C resulted in decompo-
sition giving a complicated mixture of unidentified
products.
Two notable reactions were observed for 2. In order
to remove the organic part from 2 its thermolysis was
examined. When a benzene solution of the SiMe3-
derivative 2c was heated for 2 h at 100°C, quantitative
isomerization to the v-vinyl-silyl complex 5 was ob-
served. Complex 5 was identified by comparison of its
spectral data with those of an authentic sample pre-
pared according to our reported method, i.e. interaction
ucts: the v-alkenyl-v-hydride complex 2 (CR2: CH2 (2a:
89%), CHMe (2b: 39%); CHSiMe3 (2c: 90%), CHPh
(2d: 42%); CHCOOEt (2e: 59%); 4-t-butylcyclohexyli-
dene (2f: 89%)), olefin (CR2: CPh2 (3g: 73%); CPh–
C(ꢀO)Ph: (3h: 90%); fluorenylidene (3i: 98%)) and the
adduct 4 (85%; CR2: C(COO)2CMe2 (diazoMeldrum’s
acid)), depending on the structure of diazoalkane. No
1
other products were detected by H-NMR monitoring
of reaction mixtures.
The v-alkenyl-v-hydride complexes 2 and olefins 3
result from C–C coupling of the methylene unit in 1
and the alkylidene part of external diazoalkane. Mono-
substituted diazoalkane (N2ꢀCHR) and dialkyl-substi-
tuted diazoalkane gave 2 in a specific manner, whereas
the reaction with diaryl-substituted diazoalkanes
(N2ꢀCR2) afforded 3 along with a mixture of uniden-
tified organometallic products [10]; no v-hydride-v-
alkenyl complex 2 could be detected by 1H-NMR
analysis of reaction mixtures.
of
a v-vinylidene species Cp2Ru2(v-CꢀCH2)(CO)2
(MeCN) with HSiMe3 ([4]f). Although other complexes
2 did not isomerize under similar reaction conditions,
thermolysis in the presence of CO or PPh3 gave a
mixture of unidentified dinuclear complexes.
Treatment of 1 with diazoMeldrum’s acid, a very
stable, commercially available diazoalkane with two
Scheme 3.

74
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
spectroscopic features were detected (2a:2a%=7:1 (see
The reaction with N2ꢀCHCOOEt afforded 2e as the
signals marked with asterisks in Fig. 1); 2b:2b%=3:1;
2d:2d%=3:1; 2e:2e%=3:1). The major component 2 was
assigned to the trans-isomer
sole organometallic product as mentioned above.
However, when the reaction mixture was left in
CH2Cl2 for 2 days, another v-CHꢀCHCOOEt com-
plex 6 (37%), without hydride resonance, was isolated
in addition to 2e (15%).
on the basis of the X-ray structure of the major iso-
mer of 2b (see below), and the minor component 2% to
the cis-isomer. No cis–trans interconversion was ob-
served for the isolated samples of 2 and 2% (a,b,d-se-
ries) even when heated at 100°C, whereas 2e slowly
converted to 2e% even at r.t. The large proton coupling
constants for the substituted alkenyls [11.7–13.6 Hz]
lead to assignment of trans configuration.
Complex 6 was characterized as the chloride deriva-
tive of 2e as confrmed by X-ray crystallography (see
below). The chloride complex 6 is predominantly
formed by halogenolysis of the hydride complex 2e in
CH2Cl2.
2.3. Molecular structures of v-alkenyl complexes 2b
and 6
Thus it is found that efficient C–C coupling takes
place on the diruthenium system. It should be noted
that the reaction conditions of the present C–C cou-
pling are much milder than those of thermolysis of the
The molecular structures of the v-ethenyl complex
2b and the chloro complex 6 are presented in Fig. 2
and Fig. 3, and selected structural parameters are
summarized in Table 2 and Table 3. The two Ru
centers are bridged by the v-p1:p2-alkenyl ligands.
One of the structural features is the configuration of
the Cp rings with respect to the Ru–Ru bond. Com-
plexes 2b and 6 adopt trans- and cis-configurations,
respectively. Structural characterization of closely re-
lated complexes, a stereoisomer of 2b with the cis-Cp-
configuration (2b%) ([3]i) and the methyl derivative of 6
([3]w), have been reported.
di-v-alkylidene
complexes
Cp2Ru2(v-CR2)2(CO)2
(Scheme 1).
2.2. Characterization of v-alkenyl-v-hydride complexes
Cp2Ru2(v-CHꢀCR2)(v-H)(CO)2 2
The v-alkenyl-v-hydride complexes 2 were readily
characterized on the basis of their NMR and IR spec-
1
tra (Table 1). As a typical example, a H-NMR spec-
trum of 2a is reproduced in Fig. 1a. The hydride
signal located at l −14.42 falls in the range of bridg-
ing hydride signals, and the three vinyl protons ob-
served at l 8.61 (dd), 4.41 (d), and 2.68 (d) are readily
assigned to Ha, Hb, and Hc, respectively, on the basis
of the coupling constants (Jab=7.8, Jac=12.7 Hz).
1
Separation of the H- and 13C-NMR signals is typical
for v-p1:p2-alkenyl ligands ([3]h) and IR spectra of 2
containing only terminal CO stretching vibrations
(1907–1983 cm−1 in CH2Cl2) also support the struc-
ture. Compound 2b and related compounds have been
already reported by us ([3]i). They were prepared by
hydride addition to the corresponding v-alkenyl
cations [Cp2Ru2(v-CRꢀCR2)(CO)3]+.
With regard to the configuration of the Cp rings,
only one isomer was observed for 2c. However, in the
case of 2a,b,d,e, minor isomers 2a%,b%,d%,e% with similar
They were prepared by very different methods.
In complex 2b the two CpRu(CO) fragments linked
by the Ru–Ru bond [2b: 2.830(1); cf. 6: 2.7815(6) A]
are bridged by a 4e- (v-alkenyl) and 2e-donor (v-H)
to attain the coordinatively saturated 34e-configura-
tion. Replacement of the v-H ligand by the Cl ligand
(6) causes a change in the coordination mode of one
of the two terminal CO ligands of 2b to give a bridg-
ing CO. Otherwise, a bridging Cl ligand acting as a
4e-donor would lead to a 36e-configuration.
˚

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
75
Table 1
Spectroscopic data for the diruthenium complexes 2 and 6
Complex
(R1)
1H-NMR
h-CH,
8.61
(dd, 12.7, 7.8) (d, 7.8)
9.09 4.30
(dd, 11.7, 7.8) (d, 7.8)
8.34
(d, 11.7)
8.41
(d, 11.9)
8.78
(d, 13.6)
9.36
13C-NMRa
IRb
i-CH
R1
v-H
Cp
h-CH
i-CH
R1
—
Cp
CO
2a
(H)
2a%
(H)
2b
(Me)
2b%
(Me)
2c
(SiMe3)
4.41
2.68
(d, 12.7)
3.09
(d, 11.7)
1.61
−14.42 4.42 144.4c
4.75 (d, 147)
−15.02 4.20
4.60
53.2
(t, 156)
83.5 202.4 1913
84.1 203.5 1961
3.54
−14.44 4.47 145.2
4.76 (d, 153)
−14.56 4.49
4.78
75.9
(d, 158)
28.9
(q, 127)
84.1 203.6 1954
84.3 206.9 1911
(dq, 11.7, 5.9) (d, 5.9)
3.72 1.64
(dq, 11.9, 6.0) (d, 6.0)
2.55
(d, 13.6)
4.64
0.68
−14.8
4.89 152.1c
5.06 (d, 149)
70.0
(d, 137)
77.9
−0.14
83.1 202.5 1958
84.8 205.7 1908
84.5 203.6 1958
85.2 206.9 1915
(9H, s)
7.2–7.0
(5H, m)
(q, 120)
146.2
2d
(Ph)
−14.43 4.64 161.9
(d, 12.5)
(d, 12.5)
4.80 (d, 162)
(d, 160)
139.0d
128.8d, 126 0d
2d%
9.89
4.74
−14.96 4.65
(Ph)
2e
(COOEt)
(d, 11.9)
10.04
(d, 11.7)
(d, 11.9)
3.55
(d, 11.7)
4.83
4.18
(2H, q, 6.8)
1.29 (t, 6.8)
−15.0
−14.6
4.91 154.8c
58.5
(d, 165)
60.0
(t, 147)
14.5 (q, 127)
83.9 202.3 1983
84.8 202.5 1919
1700
5.20 (d, 145)
2e%
(COOEt)
9.27
(d, 11.7)
3.63
(d, 11.7)
4.89
4.97
1973
1923
1700
2fe
8.48
—
2.4–2.1
(5H, m)
1.9–1.0
(4H, m)
0.92
(9H, s)
1.34–1.29
(3H, m)
−13.92 4.95 150.3
78.0
40.2
35.7
83.3 203.4 1956
85.9 206.7 1907
4.52 (d, 154)
(d, 119)
33.3 (t, 122)
25.0 (t, 127)
20.1 (q, 125)
174.3
60.5
(t, 149)
14.6
6
10.38
(d, 9.8)
4.30–4.09
(3H, m)
4.87 163.5c
5.32 (d, 168)
74.3
(d, 168)
87.7 197.6 1996
91.4 224.9 1842
1693
(COOEt)
(q, 127)
a 1H- and 13C-NMR spectra were observed at 400 and 100 MHz, respectively, at room temperature. Spectra were observed in C6D6 unless
otherwise stated. Chemical shifts are reported in ppm downfield from TMS. Signals unless otherwise stated are singlets, and multiplicity and
coupling constants (in Hz) are shown in parentheses.
b Recorded in CH2Cl2. The wavenumbers are reported in cm−1
c Observed in CDCl3.
.
d Doublet signals with J=ca. 160 Hz.
e CR1R2=4-t-butylcyclohexylidene.
As usually observed for dinuclear v-p1:p2-alkenyl
complexes, the bridging C1 atom is located almost
equidistant from the two Ru centers (2b: Ru1–C1:
2.027(5); Ru2–C1: 2.164(5); 6: Ru1–C1: 2.022(5);
the C–C coupling, labeling experiments using 1-d2
and 1-13C were conducted (Scheme 4). Treatment of
the v-CD2 complex (1-d2) with N2ꢀCH2 afforded 2a-
1
d2, a H-NMR spectrum of which is compared with
˚
that of 2a (Fig. 1). Disappearance of the hydride and
Ha signals and lack of coupling with the latter proton
clearly indicated that the methylene unit in dia-
zomethane was converted to the i-CH2 part of the
bridging ethenyl ligand and the v-CH2 moiety in 1
was incorporated into the hydride and the h-CH moi-
ety. The latter feature is confirmed by the 13C-labeling
experiment. The 13C-labeled v-methylene carbon atom
in 1-13C is converted to the h-carbon atom in 2a-13C.
Ru2–C1: 2.130(4) A) and the C1–C2 distances (2b:
˚
1.385(7); 6: 1.412(6) A) are slightly elongated (normal
˚
CꢀC distance: 1.34 A upon coordination to Ru2.
Other structural parameters fall in the normal ranges.
The configuration of the olefinic part in 6 is trans in
accord with the 1H-NMR analysis, as discussed
above.
2.4. Regiospecific C–C coupling re6ealed by labeling
experiments using 1-d2 (v-CD2) and 1-13C (v-13CH2)
1
As can be seen from the H-NMR spectra (Fig. 1), no
other isomer except 2a% was formed and, therefore,
In order to obtain an insight into the mechanism of
the transformation is found to be specific (Scheme 4).

76
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
Fig. 1. 1H-NMR spectra of 2a observed in C6D6 (r.t. at 400 MHz). (a) 2a obtained from 1 (peaks indicated by asterisks are for isomer 2a%). (b)
2-d2 obtained from 1-d2.
specificity of the labeling experiments (the second as-
pect: Scheme 4). Participation of the di-v-methylene
intermediate 8% can also be eliminated on the basis of
its low reactivity and selectivity with respect to C–C
coupling (Scheme 1). Thus we propose the p1-alkyli-
dene-v-methylene intermediate 8 as the key intermedi-
ate of the transformation. Because a p1-alkylidene
species may be reactive enough to couple with an-
other p1-bound ligand, migratory insertion of the p1-
alkylidene part into one of the two Ru–CH2 bonds
realizes the C–C coupling as discussed in the Section
1 (chart 1).
Similarly, reaction of 1-d2 with N2ꢀCHSiMe3 pro-
duced 2c-d2 where the hydride and the h-CH sites
were completely deuterated. Subsequent thermolysis
afforded 5-d2 with a i-CD2 group.
Thus the C–C coupling process (Scheme 3), like the
thermal isomerization, is a regiospecific transforma-
tion. The hydride and the h-CH moieties in 2 come
from the v-CH2 ligand in 1, and the alkylidene moi-
ety of the diazoalkane is incorporated into the i-
CHR part. In addition, thermolysis of 2c results in
the quantitative conversion of the v-H and h-CH
components into i-CH2 in 5.
In the resulting coordinatively unsaturated species
7, a C–H bond is so closely disposed to the coordina-
tively unsaturated site of the adjacent ruthenium cen-
ter that its oxidative addition produces the
v-hydride-v-alkenyl complex 2, whereas a bulky olefin
ligand (3) may be expelled from the coordination
sphere due to steric repulsion with the Cp2Ru2(CO)2
residue as observed in the reaction with diaryldia-
zoalkanes. The labeling experiments suggest that the
oxidative addition takes place prior to free rotation of
the p2-CH2ꢀCH2 ligand with respect to the Ru–CꢀC
midpoint axis. Otherwise, a mixture of 2a-d2 and its
regioisomer Cp2Ru2(v-H)(v-CHꢀCD2)(CO)2 would be
formed. This means that the C–H oxidative addition
is a very facile process when it is disposed to a reac-
tive site in an appropriate manner. This step can be
viewed as an example of ‘bimetallic activation’ [11],
2.5. Mechanism of C–C coupling
Plausible reaction pathways for the specific C–C
coupling process are summarized in Scheme 5, and
explain the following aspects: (1) formation of the
olefinic products 2 and 3; (2) specific transformation
of the v-CH2 moiety of 1 into the v-H and h-CH
parts in 2; (3) thermal isomerization of 2c into 5.
The first aspect suggests that a key intermediate
should be the p2-olefin species 7 arising from coupling
of the two CR2 fragments (the v-methylene ligand in
1 and the CR2 unit from external diazoalkane) within
the coordination sphere of the diruthenium unit. In
addition, the intermediate in the formation of 7
should have an unsymmetrical structure, because sym-
metrical ones such as 8% and 8%% can not explain the

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
77
Fig. 2. Molecular structure of 2b drawn at the 30% probability level.
A ‘vinylidene’ version of the present coupling reac-
tion has been reported by us [15]. Reaction of the
vinylidene analogue of 1 (10) with diazomethane gave
the v-allene complex 11
where one of the metal centers binds the substrate and
the other metal center serves as the activation site.
The isomerization of the SiMe3-derivative 2c into 5
clearly indicates that the C–H oxidative addition is
reversible. The reverse reaction (C–H reductive elimi-
nation) regenerates the p2-olefin intermediate 7 [12],
which is converted to the silyl complex 5 via C–Si
bond cleavage after prolonged heating. These results
reveal that the hydride complex resulting from the
C–H oxidative addition is the kinetic product and the
silyl complex resulting from the C–Si oxidative addi-
tion is the thermodynamic product. The kinetic
product should arise from oxidative addition of the
C–H bond, which is much more facile than that of
the sterically congested C–Si bond. The driving force
of the isomerization will be the thermodynamic stabil-
ity of the Ru–Si bond formed by the isomerization.
C–Si bond cleavage is not always a facile process [13]
but it can be achieved by means of ‘bimetallic activa-
tion’. We have reported another example of C–Si
bond activation in the silyl-v-methylene species,
Cp2Ru2(v-CH2)(X)(SiR3)(CO) (X=H or SiR3), where
reversible reductive elimination between the methylene
carbon atom and the silyl group occurs at ambient
temperature ([4]b,c).
[11]. Formation of complex 11 is explained in terms of
a mechanism similar to Scheme 5 with the intermedi-
ate stabilized by coordination of the two y bonds of
the allene molecule.
The isolation of the adduct of diazoMeldrum’s acid
(4) provided an opportunity to consider the interaction
modes of the diazoalkane with 1. Two resonance
structures 12A,B depicted in Scheme 6 are possible for
a diazoalkane and both forms can be coordinated to a
transition metal center either at the terminal nitrogen
atom or the h-carbon atom, where negative charge is
developed. In the case of diazoMeldrum’s acid, conju-
gation through the N–N–C(ester)2 y-linkage would
reduce the electron-density at the h-carbon atom.
Therefore reaction occurs at the N-terminus to give
the adduct 4 through the N-bonded form 13A. Judg-
ing from its stability, the coordination of the diazo
part of diazoMeldrum’s acid appears to be irreversible.
In other cases, coordination at the h-carbon atom
(13B) followed by elimination of N2 gives the p1-
alkylidene intermediate 8. If coordination at the N-ter-
minus is reversible, conversion from 13A to 8 is viable
by way of the C-bonded form 13B. Such bimodal
reactivity of diazoalkanes has often been noted [16].
A similar coupling of alkylidene species on a poly-
metallic system has been reported by Shapley and co-
workers [14]. Treatment of the labile triosmium
v-methylene species Os3(v-CH2)(CO)9(v-CO)(MeCN)
with N2ꢀCHR (R=H, SiMe3) furnished the v-vinyl-
v-hydride complexes Os3(v-CHꢀCHR)(v-H)(CO)10. A
labeling experiment using N2ꢀ13CH2, however, af-
forded a mixture of the regioisomers, v-13CHꢀCH2
and v-CHꢀ13CH2 species.

78
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
Fig. 3. Molecular structure of 6 drawn at the 30% probability level.
(v-alkylidene)(p1-alkylidene) intermediate 8 (Scheme 5)
3. Conclusion
and the bis(v-alkylidene) complex have the same com-
position ‘Cp2Ru2(CR2)2(CO)2’, their reactivity has
proved to be quite different. Species 8 undergoes C–C
specific coupling, whereas the bis(v-alkylidene) species
The present study has revealed that a bridging
methylene complex bearing a labile acetonitrile ligand,
Cp2Ru2(v-CH2)(CO)2(NCMe) 1, readily reacts with di-
azoalkane (N2ꢀCR2) to give the olefinic products 2 and
3 via C–C coupling of the bridging CH2 moiety with
the external CR2 source. Labeling experiments have
shown that the v-CH2 moiety in 1 is transferred into
the v-H and h-CH parts of the v-vinyl-v-hydride com-
plex 2 in a specific manner. It is also notable that the
efficiency of C–C coupling depends on the coordina-
tion mode of the alkylidene ligand. Although both the
Table 3
Selected structural parameters for 6a
˚
Bond lengths (A)
Ru1–Ru2
Ru1–C1
Ru1–C6
Ru1–C7
Ru2–C11
Ru2–C1
Ru2–C2
Ru2–C7
C3–O1
2.7815(6)
2.022(5)
1.885(5)
2.007(5)
2.431(2)
2.130(4)
2.240(5)
2.133(5)
1.206(6)
C3–O2
C4–O2
C6–O6
C7–O7
C1–C2
C2–C3
C4–C5
C1–H1
C2–H2
1.346(6)
1.434(6)
1.129(5)
1.155(5)
1.412(6)
1.465(6)
1.484(8)
1.07(4)
Table 2
Selected structural parameters for 2ba
0.93(4)
Bond angles (°)
Ru2–Ru1–C1
Ru2–Ru1–C6
Ru2–Ru1–C7
C1–Ru1–C6
C1–Ru1–C7
C6–Ru1–C7
Ru1–Ru2–C11
Ru1–Ru2–C1
Ru1–Ru2–C2
Ru1–Ru2–C7
C11–Ru2–C1
C11–Ru2–C2
C11–Ru2–C7
C1–Ru2–C2
C1–Ru2–C7
C2–Ru2–C7
˚
Bond lengths (A)
49.6(1)
93.3(1)
49.7(1)
89.8(2)
99.3(2)
91.5(2)
97.85(4)
46.3(1)
74.2(1)
45.9(1)
110.8(1)
84.2(1)
78.0(1)
37.6(2)
92.1(2)
112.8(2)
C3–O2–C4
Ru1–C2–Ru2
Ru1–C1–C2
Ru2–C1–C2
Ru2–C2–C1
Ru2–C2–C3
C1–C2–C3
O1–C3–O2
O1–C3–C2
O2–C3–C2
O2–C4–C5
Ru1–C6–06
Ru1–C7–Ru2
Ru1–C7–O7
Ru2–C7–O7
117.0(4)
84.1(2)
125.2(4)
75.4(3)
Ru1–Ru2
Ru1–C1
Ru2–C1
Ru2–C2
C1–C2
2.830(1)
2.027(5)
2.164(5)
2.281(5)
1.385(7)
1.503(7)
Ru1–C11–15
Ru2–C21–25
Ru1–C16
Ru2–C26
C16–O11
2.205–2.273(6)
2.201–2.245(6)
1.821(6)
1.860(6)
1.160(7)
67.0(3)
117.8(3)
122.6(5))
123.1(5)
127.4(5)
109.4(4)
108.5(5)
177.6(5)
84.4(2)
C2–C3
C26–O21
1.144(7)
Bond angles (°)
Ru2–Ru1–C1
Ru2–Ru1–C16 94.7(2)
49.6(1)
C2–Ru2–C26
Ru1–Cl–Ru2
Ru1–Cl–C2
C1–C2–C3
C1–C2–Ru2
C3–C2–Ru2
81.6(2)
84.9(2)
124.3(4)
123.1(5)
67.3(3)
117.2(3)
C1–Ru1–C16
Ru1–Ru2–C1
Ru1–Ru2–C2
88.4(2)
45.5(1)
71.9(1)
Ru1–Ru2–C26 91.5(2)
142.1(4)
133.3(4)
C1–Ru2–C2
C1–Ru2–C26
36.2(2)
105.5(2)
Ru1–C16–O11 179.8(6)
Ru2–C26–O21 177.4(6)
a
a
˚
˚
Bond lengths in A and bond angles in degrees.
Bond lengths in A and bond angles in degrees.

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
79
Scheme 4.
were treated with appropriate drying agents, distilled,
and stored under argon. Complex 1 ([3]q) and dia-
zoalkanes (N2ꢀCR1R2: CR1R2ꢀ CH2 [18], CHMe [19],
CHCOOEt [20], 4-t-butylcyclohexylidene [21], CHPh
[22], CPh2 [23], CPh(COPh) [23], fluorenylidene [23])
were prepared according to the reported methods. Sam-
ples of N2ꢀCHSiMe3 (10% hexane solution) and dia-
zoMeldrum’s acid were purchased from Tokyo Kasei
(TCI) and used as received. Preparative TLC was per-
formed on silica gel [silica gel 60 PF254 (Merck
1.007747)] and alumina [aluminum oxide 60 PF254 (Typ
produces a low yield mixture of products upon pyroly-
sis. The C–C bond formation in 1, however, is viewed
as a migratory insertion reaction at a mononuclear
center (Chart 1), with the other ruthenium center func-
tioning as a substituent of the v-methylene ligand to
support the dinuclear structure.
On the basis of the present results, a modified mech-
anism for catalytic CO hydrogenation, involving olefin
as a key intermediate, can be proposed. First, the
carbon chain can be propagated via coupling of two
alkylidene units (divalent carbon species) to give an
olefin intermediate. (In the classical Fischer–Tropsch
mechanism, alkyl–alkylidene coupling leads to a ho-
mologated alkyl). Secondly, olefin can be formed by
dissociation of the resulting olefin intermediate. (In the
classical mechanism, olefinic products are formed via
i-elimination of an alkyl intermediate). Finally, C–H
oxidative addition of the olefin species furnishes an
alkenyl intermediate, which was recently postulated as
an initiator of FT synthesis [17].
1
E) (Merck Art. 1103)]. H- and 13C-NMR spectra were
recorded on a JEOL EX400 (1H: 400; 13C: 100 MHz)
spectrometer. Solvents for NMR measurements con-
taining 0.5% TMS were dried over molecular sieves,
degassed, distilled under reduced pressure, and stored
under Ar. IR and FD-MS spectra were obtained on a
JASCO FT/JR 5300 spectrometer and a Hitachi M80
mass spectrometer, respectively. Elemental analyses
were performed in the Elemental Analysis Laboratory
of our laboratory.
4.2. Reaction of Cp2Ru2(v-CH2)(CO)2(NCMe) 1 with
4. Experimental section
diazomethane
4.1. General methods
To a benzene (2 ml) suspension of 1 (22.3 mg, 0.052
mmol) was added an ethereal (3 ml) solution of CH2N2
(ca. 0.50 mmol). After stirring for 20 min at r.t., the
reaction mixture was passed through a short alumina
All manipulations were carried out under an inert
atmosphere using standard Schlenk tube techniques.
Ether and hexanes (Na–K alloy) and CH2Cl2 (P2O5)

80
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
Scheme 5.
solution, and the ether layer was used without further
purification. To a THF suspension (2 ml) of 1 (75.7 mg,
0.171 mmol) was added to the ethereal diazoethane
solution at ambient temperature. Gas evolution (N2)
was noted and the solution color changed from brown
into dark brown. After 10 min of stirring the volatiles
were removed under reduced pressure and preparative
TLC separation (silica, eluted with hexane:ether=10:1)
gave two isomers: 2b (28.8 mg, 0.067 mmol, 39% yield)
and 2b% (4.4 mg, 0.010 mmol, 6%). Anal. Calc. for
C15H16O2Ru2: C, 41.85; H, 3.75. Found: C, 41.79; H,
3.69.
pad. The volatiles were removed under reduced pres-
sure and the residue was subjected to preparative TLC
separation (silica, eluted with ether:hexane=1:3). The
vl-ethenyl-v-hydride complex 2a (19.3 mg, 0.046 mmol,
89%) was obtained as yellow solid. 2a: Anal. Calc. for
C14H14O2Ru2: C, 40.38; H, 3.36. Found: C, 40.15; H,
3.31.
4.3. Reaction of 1 with diazoethane
An ethereal solution of diazoalkane was prepared by
treatment of a cold ethereal solution of N-ethyl-N-ni-
trosourea (210 mg, 1.79 mmol) with aqueous KOH
4.4. Reaction of 1 with trimethylsilyldiazomethane
A
mixture of 1 (32.0 mg, 0.072 mmol) and
Me3SiCHN2 (0.4 ml of n-hexane solution, ca. 10%, ca.
0.23 mmol) in benzene (2 ml) was stirred for 5 min at
r.t. After work up as described for 2a, complex 2c (31.8
mg, 0.072 mmol, 91%) was obtained as an orange solid.
2c: Anal. Calc. for C17H22O2Ru2Si: C, 41.80; H, 4.50.
Found: C, 41.36; H, 4.45.
4.5. Thermal isomerization of 2b
Heating a C6D6 solution of 2b for 2 h at 100°C gave
the isomer 5 in a quantitative yield. Complex 5 was
1
Scheme 6.
identified by comparison of a H-NMR spectrum with

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
81
that of an authentic sample prepared according to ref.
Table 4
Crystallographic data for 2b and 6
([4]f). 5: lH (C6D6) 7.82 (1H, dd, J=11.7 and 7.8 Hz,
CH), 4.69, 4.60 (5H×2, s×2, Cp2), 3.77 (1H, d, J=7.8
Hz, CH2), 2.24 (1H, d, J=11.7 Hz, CH2), IR(CH2Cl2):
Compound
Empirical formula
2b
6
C15H16O2Ru C17H17O4Ru2Cl
w(CꢁO) 1958, 1908 cm−1
.
2
Formula weight
Crystal system
Space group
430.43
Triclinic
P1
522.9
Monoclinic
P21/n
8.794(2)
13.826(5)
14.359(5)
4.6. Reaction of 1 with ethyl diazoacetate
˚
a (A)
8.421(3)
13.802(5)
7.353(2)
94.15(3)
98.94(3)
117.49(2)
739(1)
2
1.93
20.4
3–60
4619
A sample of ethyl diazoacetate (1.7 ml, O.5 M in
CH2Cl2, ca. 0.85 mmol) was added to a solution of 1
(126.0 mg, 0.28 mmol) in CH2Cl2 (3 ml). The mixture
was then stirred for 1 h at r.t. After removal of the
volatiles under reduced pressure, the residue was sub-
jected to preparative TLC separation (silica, eluted with
acetone:hexane=1:3). 2c (81.8 mg, 0.17 mmol, 59%)
was obtained as a yellow solid. 2c: Anal. Calc. for
C17H18O4Ru2: C,4 1.80; H, 3.68. Found: C, 42.10; H,
3.38.
˚
b (A)
˚
c (A)
h (°)
i (°)
k (°)
92.90(2)
3
˚
V (A )
1744(1)
4
Z
Dcalc. (g cm−3
)
1.99
19.0
5–50
3441
2631
v (cm−1
2q (°)
)
Number of data collected
Number of unique data with I\
3|(I)
Number of parameters refined
R
Rw
G-O-F
3479
When the reaction was carried under the same condi-
tions for 4 days, 2c (15%) end the chloride derivative 6
(37%) were obtained. 6: Anal. Calc. for C17H17O4Ru2Cl:
C, 39.01; H, 3.25. Found: C, 38.43; H, 3.02.
172
225
0.037
0.039
3.44
0.026
0.025
2.64
4.7. Reaction of 1 with 4-t-butyldiazocyclohexane
(2H, m, Ph), 7.2–7.0 (6H, m, Ph), 5.63, 5.32 s×2,
ꢀCH2). lC (CDCl3) 150.0 (s, ꢀC), 141.5 (s, ipso-Ph),
129.2 (d, J=160 Hz, Ph), 128.1 (d, J=160 Hz, Ph),
127.7 (d, J=160 Hz, Ph). MS (FD) 208 (M+).
To a THF suspension (2 ml) of 1 (25.0 mg, 0.056
mmol) was added a hexane solution of 4-t-butyldiazocy-
clohexane (0.1 M solution, 2 ml, 0.2 mmol). After
stirring for 20 min, the mixture was worked up as
described for 2b. 2f (yellow solid): Anal. Calc. for
C23H29O2Ru2: C, 51.19; H, 5.42. Found: C, 50.87; H,
5.39.
4.10. Reaction of 1 with diazofluorene
The reaction was carried out as described for 3b
starting from 1 (41.1 mg, 0.10 mmol) and diazofluorene
(57.0 mg, 0.29 mmol). 3c [25]: lH (CDCl3) 7.5–7.4,
7.3–7.1 (4H×2, m×2, aromatic protons), 5.78 (2H, s,
ꢀCH2).
4.8. Reaction of 1 with diphenyldiazomethane
To a benzene (2 ml) suspension of 1 (41.2 mg, 0.093
mmol), 2.5 equivalents of Ph2CN2 (0.5 M CH2Cl2 solu-
tion) was added. The resultant mixture was stirred for
30 min at r.t. After removal of the volatiles, the residue
was subjected to preparative TLC separation (silica,
eluted with CH2Cl2:hexane=1:2). From the colorless
band diphenylethylene 3a (12.1 mg, 0.067 mmol, 73%
yield) was obtained. 3a: lH (CDCl3) 7.37–7.27 (10H, m,
Ph), 5.48 (2H, s, CH2). lC (CDCl3) 150.0 (s, ꢀCPh2),
141.5 (s, ipso), 129.2, 128.1, 127.6 (d×3, J=160 Hz, Ph
signals), 114.3 (t, J=158 Hz, CH2).
4.11. Reaction of 1 with 5-diazoMeldrum’s acid
To a benzene (2 ml) suspension of 1 (42.3 mg, 0.095
mmol) was added diazoMeldrum’s acid (16.5 mg, 0.097
mmol). The solids dissolved to give a brown solution
and the resultant mixture was stirred for 30 min at r.t.
After removal of the volatiles under reduced pressure,
the residue was dissolved in acetone and subjected to
preparative TLC separation (alumina, eluted with ace-
tone:hexane=1:2). 4 (46.2 mg, 0.08 mmol, 85%) was
obtained as a brown solid. (the major isomer) lH
(CDCl3) 9.94, 9.13 (1H×2, s×2, v-CH2), 5.41, 5.39
(SH×2, s×2, Cp2), 1.64 (6H, s, CH3). lC (CDCl3)
230.9 (s, v-CO), 197.8 (s, CO), 160.1 (s, CO), 129.1 (t,
J=151 Hz, v-CH2), 103.5 (s, N–C), 92.7 (d, J=180
Hz, Cp), 91.2 (d, J=180 Hz, Cp). 68.7 (s, C), 26.4 (q,
J=132 Hz). IR(CH2Cl2): w(C–O) 1987, 1834, 1674
cm−1. Anal. Calc. for C19H10N2O6Ru2: C, 39.86; H,
3.15; N, 4.89. Found: C, 39.75; H,3.43, N, 4.34.
4.9. Reaction of 1 with phenylbenzoyldiazomethane
To a mixture of 1 (51.5 mg, 0.12 mmol) and phenyl-
benzoyldiazomethane (82.5 mg, 0.37 mmol) was added
THF (4 ml) and the resultant mixture was stirred for 1
h at r.t. After 1 h the volatiles were removed under
reduced pressure, and the product was quantified by
1H-NMR after addition of ferrocene (an internal stan-
dard). 3b [24]: lH (CDCl3) 7.9–7.8 (2H, m, Ph), 7.4–7.3

82
M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
[2] (a) C. Masters, Adv. Organomet. Chem. 19 (1979) 63. (b) E.L.
4.12. X-ray crystallography
Muetterties, T.N. Rhodin, E. Band, C.F. Brucker, W.R. Pretzer,
Chem. Rev. 79 (1979) 79. (c) C.K. Roofer-DePoorter, Chem.
Rev. 81 (1981) 447. (d) W.A. Herrmann, Angew. Chem. Int. Ed.
Engl. 21 (1982) 117. (e) A.R. Cutler, P.K. Hanna, J.C. Vites,
Chem. Rev. 88 (1988) 1363 and references cited therein. For
carbide complexes, (f) M. Tachikawa, E.L. Muetterties, Prog.
Inorg. Chem. 28 (1981) 203. (g) J.S. Bradley, Adv. Organomet.
Chem. 22 (1983) 1. (h) D.F. Shriver, M.J. Sailor, Acc. Chem.
Res. 21 (1988) 374. See also (i) W. Beck, B. Niemer, M. Wieser,
Angew. Chem. Int. Ed. Engl. 32 (1993) 923.
Single crystals of 2b and 6 were obtained by recrys-
tallization from THF-hexanes and CH2Cl2-hexane, re-
spectively, and mounted on glass fibers.
Diffraction measurements of 2b and 6 were made on
a Rigaku AFC5R automated fourcircle diffractometer
by using graphite-monochromated Mo–Kh radiation
˚
(u=0.71059 A). The unit cells were determined and
[3] (a) A.F. Dyke, S.A.R. Knox, P.J. Naish, A.G. Orpen, J. Chem.
Soc. Chem. Commun. (1980) 441. (b) A.F. Dyke, S.A.R. Knox,
P.J. Naish, A.G. Orpen, J. Chem. Soc. Chem. Commun. (1980)
804. (c) D.L. Davies, A.F. Dyke, A. Endesfelder, et al., J.
Organomet. Chem. 198 (1980) C43. (d) A.F. Dyke, S.A.R.
Knox, K.A. Mead, P. Woodward, J. Chem. Soc. Chem. Com-
mun. (1981) 861. (e) M. Cooke, D.L. Davies, J.E. Guerchais,
S.A.R. Knox, K.A. Mead, J. Roue´, P. Woodward, J. Chem. Soc.
Chem. Commun. (1981) 862. (f) P.Q. Adams, D.L. Davies, A.F.
Dyke, S.A.R. Knox, K.A. Mead, P. Woodward, J. Chem. Soc.
Chem. Commun. (1983) 222. (g) N.J. Ferrow, S.A.R. Knox,
M.J. Morris, A.G. Orpen, J. Chem. Soc. Chem. Commun. (1983)
234. (h) A.F. Dyke, S.A.R. Knox, M.J. Morris, P.J. Naish, J.
Chem. Chem. Soc. Dalton Trans. (1983) 1417. (i) R.E. Colborn,
A.F. Dyke, S.A.R. Knox, K.A. Mead, P. Woodward, J. Chem.
Chem. Soc. Dalton Trans. (1983) 2099. (j) R.E. Colborn, D.L.
Davies, A.F. Dyke, A. Endesfelder, et al., J. Chem. Soc. Dalton
Trans. (1983) 2661. (k) S.A.R. Knox, Pure Appl. Chem. 56
(1984) 81. (l) N.J. Ferrow, S.A.R. Knox, J. Chem. Soc. Chem.
Commun. (1984) 679. (m) D.L. Davis, B.P. Gracey, V. Guer-
chais, S.A.R. Knox, A.G. Orpen, J. Chem. Soc. Chem. Com-
mun. (1984) 841. (n) D.L. Davies, S.A.R. Knox, K.A. Mead,
M.J. Morris, P. Woodward, J. Chem. Soc. Dalton Trans. (1984)
2293. (o) N.G. Connelly, N.J. Ferrow, B.P. Gracey, S.A.R.
Knox, A.G. Orpen, J. Chem. Soc. Chem. Commun. (1985) 14.
(p) N.G. Connelly, N.J. Ferrow, S.A.R. Knox, K.A. Macpher-
son, A.G. Orpen, J. Chem. Soc. Chem. Commun. (1985) 16. (q)
N.M. Doherty, J.A.K. Howard, S.A.R. Knox, N.J. Terrill, M.I.
Yates, J. Chem. Soc. Chem. Commun. (1989) 638. (r) J.A.K.
Howard, S.A.R. Knox, N.J. Terrill, M.I. Yates, J. Chem. Soc.
Chem. Commun. (1989) 640. (s) M.J. Fildes, S.A.R. Knox, A.G.
Orpen, M.L. Turner, M.I. Yates, J. Chem. Soc. Chem. Com-
mun. (1989) 1680. (t) R.E. Colborn, D.L. Davies, A.F. Dyke,
S.A.R. Knox, K.A. Mead, A.G. Orpen, J. Chem. Soc. Dalton
Trans. (1989) 1799. (u) S.A.R. Knox, K.A. Macpherson, A.G.
Orpen, M.C. Rendle, J. Chem. Soc. Dalton Trans. (1989) 1807.
(v) G.C. Bruce, S.A.R. Knox, A.J. Phillips, J. Chem. Soc. Chem.
Commun. (1990) 716. (w) G.C. Bruce, B. Gangnus, S.E. Garbar,
S.A.R. Knox, A.G. Orpen, A.J. Philiiips, J. Chem. Soc. Chem.
Commun. (1990) 1360. (x) R.E. Colborn, A.F. Dyke, B.P.
Gracey, et al., J. Chem. Soc. Dalton Trans. (1990) 761. (y)
S.A.R. Knox, J. Organomet. Chem. 400 (1990) 255.
refined by a least-squares method using 20 independent
reflections (2u ca. 20°). Data were collected with a 2q-ꢀ
scan technique. If |(F)/F was more than 0.1, a scan
was repeated up to three times and the results were
added to the first scan. Three standard reflections were
monitored at every 150 measurements. The data pro-
cessing (data collection) was performed on a microvax
II computer. In the reduction of data, Lorentz and
polarization corrections were made. Empirical absorp-
tion corrections (C scan) were made. Crystallographic
data and the results of refinements are summarized in
Table 4.
Structure analysis was performed on an IRIS Indigo
computer by using the teXsan structure solving pro-
gram system obtained from the Rigaku Corporation,
Tokyo, Japan. Neutral scattering factors were obtained
from the standard source [26]. The function minimized
was
ꢀw(ꢁFoꢁ–ꢁFcꢁ)2
where
w=[|2(Fo)]−1
=
4F2o ·[|2(Fo2)]−1. The schemes for unweighted and
weighed agreement factors were as follows: R=ꢀꢂFoꢁ–
ꢁFcꢂ/ꢀꢁFoꢁ; Rw=[(ꢀw(ꢁFoꢁ–Fcꢂ)2/ꢀwFo2)]12. The structures
were solved by a combination of the direct methods
(SAPI91 and SHELXL 87) and Fourier synthesis
(DIRDIF). Unless otherwise stated, non-hydrogen
atoms were refined with anisotropic thermal parame-
ters, and hydrogen atoms except H1 and H2 of 6
(refined isotropically) were fixed at the calculated posi-
˚
tions (C–H=0.95 A) and were not refined.
Acknowledgements
We are grateful to the Ministry of Education, Sci-
ence, Sports and Culture of the Japanese Government
for financial support of this research. M. Akita thanks
the Royal Society of Chemistry for Journals Grants for
International Authors.
[4] (a) M. Akita, T. Oku, Y. Moro-oka, J. Chem. Soc. Chem.
Commun. (1992) 1031. (b) M. Akita, T. Oku, R. Hua, Y.
Moro-oka, J. Chem. Soc. Chem. Commun. (1993) 1670. (c) M.
Akita, R. Hua, Y. Moro-oka, J. Chem. Soc. Chem. Commun.
(1996) 541. (d) M. Akita, R. Hua, T. Oku, Y. Moro-oka,
Organometallics 15 (1996) 2548. (e) M. Akita, R. Hua, T. Oku,
M. Tanaka, Y. Moro-oka, Organometallics 15 (1996) 4162. (f)
R. Hua, M. Akita, Y. Moro-oka, Inorg. Chim. Acta 250 (1996)
177. (g) M. Akita, R. Hua, Y. Moro-oka, J. Organomet. Chem.
539 (1997) 205. (h) M. Akita, R. Hua, S. Nakanishi, M. Tanaka,
Y. Moro-oka, Organometallics 16 (1997) 5572.
References
[1] (a) J. Falbe, New Synthesis with Carbon Monoxide, Springer,
Berlin, 1980. (b) W. Keim, Catalysis in C1 Chemistry, D. Reidel,
Dordrecht, 1983. (c) R.B. Anderson, The Fischer-Tropsch Syn-
thesis, Academic Press, London, 1984. (d) H. Pines, The Chem-
istry of Catalytic Hydrocarbon Conversions, Academic Press,
London, 1981. (e) G.A. Somorjai, Introduction to Surface
Chemistry and Catalysis, Wiley, New York, 1994.
[5] (a) W.A. Herrmann, Adv. Organomet. Chem. 20 (1982) 159. (b)
W.A. Herrmann, J. Organomet. Chem. 250 (1983) 319. (c) I.M.
Saez, D.G. Andrews, P.M. Maitlis, Polyhedron 7 (1988) 827. (d)
R.J. Puddephatt, Polyhedron 7 (1988) 767.

M. Akita et al. / Journal of Organometallic Chemistry 569 (1998) 71–83
83
[6] (a) J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Princi-
ples and Applications of Organotransition Metal Chemistry,
University Science Books, Mill Valley, 1987. (b) M. Brookhart
and W. B. Studabaker, Chem. Rev. 87 (1987) 411. (c) E.W. Abel,
F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive
Organometallic Chemistry II, vol. 12, Pergamon, Oxford, 1995,
ch. 5.1–5.5. (d) B. Cornils, W.A. Herrmann, Applied Homoge-
neous Catalysis with Organometallic Compounds (2 vols.),
VCH, Weinheim, 1996. (e) G.W. Parshal, S.D. Ittel, Homoge-
neous Catalysis, 2nd ed., Wiley-Interscience, New York, 1992.
[7] (a) H. Berke, R. Hoffmann, J. Am. Chem. Soc. 100 (1978) 7224.
(b) B.E.R. Schilling, R. Hoffmann, D.L. Lichtenberger, J. Am.
Chem. Soc. 101 (1979) 585. (c) N.M. Kostic, R.F. Fenske, J.
Am. Chem. Soc. 104 (1982) 3879.
[12] The ease of the C–H reductive elimination (2“7) is dependent
on the substituent of the v-alkenyl part. The interconversion
between 2e and 2e% is explained in terms of free rotation of the
Ru–Ru bond in 7.
[13] See 4 (c,h) and references cited therein.
[14] J.R. Shapley, A.C. Sievert, M.R. Churchill, H.J. Wassermann, J.
Am. Chem. Soc. 103 (1981) 6975.
[15] G.S. Lewandos, N.M. Doherty, S.A.R. Knox, K.A. Macpher-
son, A.G. Orpen, Polyhedron 7 (1988) 837.
[16] See, for example, (a) K.D. Schramm, J.A. Ibers, J. Am. Chem.
Soc. 100 (1978) 2932. (b) G.L. Hillhouse, B.L. Haymore, W.A.
Herrmann, Inorg. Chem. 18 (1979) 2423. (c) L. Messerle, M.D.
Curtis, J. Am. Chem. Soc. 104 (1982) 889. (d) M.D. Curtis, L.
Messerle, J.J. D’Errico, W.M. Butler, M.S. Hay, Organometal-
lics 5 (1986) 2283. (e) A. Nakamura, T. Yoshida, M. Cowie, S.
Otsuka, J.A. Ibers, J. Am. Chem. Soc. 99 (1977) 2108.
[17] P.M. Maitlis, H.C. Long, R. Quyoum, M.L. Turner, Z.-Q.
Wang, J. Chem. Soc. Chem. Commun. (1996) 1.
[8] A part of the results were already reported in a preliminary
communication. M. Akita, R. Hua, S.A.R. Knox, Y. Moro-oka,
S. Nakanishi, M.I. Yates, J. Chem. Soc. Chem. Commun. (1997)
51.
[9] Complex 1 is soluble in CH2Cl2 but insoluble in other organic
solvents such as benzene and ether. However, the reaction with
diazoalkane is so fast that it makes the system homogeneous
within a few minutes.
[10] One of the organometallic products was tentatively assigned to a
dinitrogen complex Cp2Ru2(v-N2)(CO)2 on the basis of its spec-
[18] T.J. DeBoer, H.J. Backer, Org. Synth. Collect. 4 (1963) 250.
[19] A.F. Mckay, W.L. Ott, G.W. Taylor, M.N. Buchanan, J.F.
Crooker, Can. J. Res. Sect. B 28 (1950) 683.
[20] N.E. Searle, Org. Synth. Collect. vol. 4 (1963) 424.
[21] T.L. Holton, H. Shechter, J. Org. Chem. 60 (1995) 4725.
[22] D.G. Farnum, J. Org. Chem. 28 (1963) 870.
[23] C.D. Nenitzescu, E. Solomonica, Org. Synth. Collect. 2 (1943)
496.
[24] R.P. Kopinski, J. Pinhey, Aust. J. Chem. 36 (1983) 311.
[25] M. Neuenschwander, H.P. Farrni, H. Lehmann, R. Vo¨geli,
Chimia 28 (1974) 115.
tral data [lH 4.27, w(CO): 1953 cm−1
,
m/e=316 (for
Cp1201Ru2(N2)(CO)2)] but it could not be fully characterized due
to its thermal instability. It should arise from the trapping of an
eliminated highly coordinatively unsaturated species by the
evolved N2.
[11] See, for example, T. Takao, S. Yoshida, S. Suzuki, M. Tanaka,
Organometallics 14 (1995) 3856 and references cited therein.
[26] International Tables for X-Ray Crystallography, vol. 4, Kynoch
Press, Birmingham, 1975.
.