Oxidative addition of allylic substrates to coordinatively unsaturated ruthenium compounds, [Ru(η5-C5Me5)(η-amidinate)]: Preparation, structure elucidation, and catalysis of novel ruthenium (IV)-η3-allyl complexes

Article abstract of DOI:10.1246/bcsj.74.1927

Oxidative addition reactions of allylic halides, acetates, and carbonates with [Ru(η⁵-C₅Me₅)(η-amidinate)] [amidinate: ⁱPrNC(Me)=NⁱPr (1a), ¹BuNC(Ph)=N¹Bu (1b)], which shows signs of coordinative unsaturation, gave novel cationic π-allyl ruthenium(IV) species. The compounds [Ru(η³-allyl)(η⁵-C⁵Mes)(η ²-amidinate)]⁺X^- were isolated by anion exchange of the products (X = PF₆, BF₄, BPh₄), and were characterized by spectroscopic analysis. The crystallography of two of the [Ru(η³-allyl)(η⁵-C₅Mes)(η ²-amidinate)] ⁺X^- revealed a four-legged piano stool structure in which two nitrogen atoms in the amidinate ligand and two carbon atoms in the η³-allyl ligands occupy the positions of four legs; the orientation of the η³-allyl ligand was endo. Although cyclic voltammograms of the precursor, [Ru(η⁵-C₅Me₅)(η-amidinate)], indicated possible oxidative addition of organic halides other than allylic halides to [Ru(η⁵-C₅Me₅)(η-amidinate)], only allylic halides gave the corresponding Ru(IV) products. The importance of prior coordination of the carbon-carbon double bond of allylic substrates was evidenced by NMR observation of the intermediate in the reaction of 1a or 1b with allyl acetate. Addition of nucleophiles such as PhLi, dimethyl methylsodiomalonate, and piperidine to the [Ru(η³-allyl)(η⁵-C₅Me ₅)(η²-amidinate)]⁺X^- gave rise to allylation of these nucleophiles and regeneration of [Ru(η⁵-C₅Me₅)(η-amidinate)]. The reactions of allyl methyl carbonate with nucleophiles were also achieved by catalysis of either [Ru(η⁵C₅Me₅)(η-amidinate)] or [Ru(η³-allyl)(η⁵-C₅Me ₅)(η²-amidinate^)]+X^-.

Full text of DOI:10.1246/bcsj.74.1927

c. Jpn.
© 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 1927–1937 (2001) 1927
e Chemical
ciety of Ja-
n
e Chemical
ciety of Ja-
n
Oxidative Addition of Allylic Substrates to Coordinatively Unsaturated
η5
η
Ruthenium Compounds, [Ru( -C5Me5)( -amidinate)]: Preparation,
η3
Structure Elucidation, and Catalysis of Novel Ruthenium (Ⅳ)- -Allyl
Complexes
01
27
Ⅱ
Oxidative Addition to Ru (C5Me5)(amidinate)
H. Kondo et al.
37
Hideo Kondo,# Akira Kageyama,## Yoshitaka Yamaguchi,† Masa-aki Haga,†† Karl Kirchner,†††, ### and
SJA8
09-2673
098
Hideo Nagashima*
Institute of Advanced Material Study, Graduate School of Engineering Sciences, and CREST, Japan Science and
Technology Corporation (JST), Kyushu University, Kasuga, Fukuoka 816-8580.
01
01
.2
†Department of Materials Chemistry, Faculty of Engineering, Yokohama National University, Tokiwadai, Yokohama,
240-8501.
††Department of Applied Chemistry, Faculty of Science and Engineering, Chuo University, Kasuga, Bunkyo-ku, Tokyo
112-8551.
†††Institute of Inorganic Chemistry, Vienna University of Technology, Getreidemarkt 9, A-1060 Vienna, Austria
(Received March 21, 2001)
Oxidative addition reactions of allylic halides, acetates, and carbonates with [Ru(η5-C5Me5)(η-amidinate)] [amidi-
nate: iPrNC(Me)wNiPr (1a), tBuNC(Ph)wNtBu (1b)], which shows signs of coordinative unsaturation, gave novel cation-
ic π-allyl ruthenium(Ⅳ) species. The compounds [Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+X− were isolated by anion
exchange of the products (X = PF6, BF4, BPh4), and were characterized by spectroscopic analysis. The crystallography
of two of the [Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+X− revealed a four-legged piano stool structure in which two ni-
trogen atoms in the amidinate ligand and two carbon atoms in the η3-allyl ligands occupy the positions of four legs; the
orientation of the η3-allyl ligand was endo. Although cyclic voltammograms of the precursor, [Ru(η5-C5Me5)(η-amidi-
nate)], indicated possible oxidative addition of organic halides other than allylic halides to [Ru(η5-C5Me5)(η-amidinate)],
only allylic halides gave the corresponding Ru(Ⅳ) products. The importance of prior coordination of the carbon-carbon
double bond of allylic substrates was evidenced by NMR observation of the intermediate in the reaction of 1a or 1b with
allyl acetate. Addition of nucleophiles such as PhLi, dimethyl methylsodiomalonate, and piperidine to the [Ru(η3-al-
lyl)(η5-C5Me5)(η2-amidinate)]+X− gave rise to allylation of these nucleophiles and regeneration of [Ru(η5-C5Me5)(η-
amidinate)]. The reactions of allyl methyl carbonate with nucleophiles were also achieved by catalysis of either [Ru(η5-
C5Me5)(η-amidinate)] or [Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+X−.
Coordinatively unsaturated species1 have attracted the atten-
tion of organometallic chemists in relation to possible interme-
diates in transition metal-catalyzed organic reactions.2 In par-
ticular, the coordinatively unsaturated nature of certain ruthe-
nium complexes has been actively investigated, because ruthe-
nium complexes are generally faithful to the 18-electron rule,
and preparation and characterization of rare examples of isol-
able ruthenium complexes bearing 16 valence electrons con-
tributes to better understanding of factors to stabilize the reac-
tive ruthenium center.1a,c,3 Recent work has revealed the possi-
ble contribution of sterically hindered circumstances around
the metal center3a–m and donation of π-3a–l,n–s and σ-electrons3t–v
of the ligand to the vacant orbitals of the metal for the stabili-
zation of 16 electron half sandwich ruthenium complexes.
We have recently discovered a novel type of ruthenium
complexes formally bearing 16 valence electrons, in which
amidinato ligands4 play an important role in stabilizing the ru-
thenium center.5 These novel ruthenium-amidinate complexes
are assignable to [Ru(η5-C5Me5)(η-amidinate)] on the basis of
NMR analysis, and are considered to have 16 valence elec-
trons, because the amidinato ligand generally acts as a 4 elec-
tron ligand to Ru2+ 5a
Possible coordination of π-electrons in
.
the amidinato ligand as an additional stabilizing factor was
suggested by crystallographic analysis,5a and recently support-
ed by DFT calculations of [Ru(η5-C5H5)(η-HNwCH–NH)]6
and the molecular structure of [Ru2(η5-C5Me5)2Br(µ2:η-amidi-
nate)].5c The coordinatively unsaturated nature of [Ru(η5-
C5Me5)(η-amidinate)] was evidenced by high reactivity of this
complex with various two-electron donor ligands, leading to
formation of [Ru(η5-C5Me5)(η2-amidinate)(L)].5a The strong
π-donor property of [Ru(η5-C5Me5)(η-amidinate)] is suggest-
# CREST, Japan Science and Technology Coorporation (JST).
## Graduate School of Engineering Sciences, Kyushu University.
### Visiting Professor of Institute of Advanced Material Study,
Kyushu University, 2000 Jan. –Apr.

Ⅱ
1928 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Oxidative Addition to Ru (C5Me5)(amidinate)
ed from formation of stable [Ru(η5-C5Me5)(η2-amidinate)(L)],
where L is π-acceptor ligands such as CO and TCNE, and from
the very low νco absorption of [Ru(η5-C5Me5)(η2-amidi-
nate)(CO)] which has been observed.
ible oxidation wave was observed at Epa = 0.53 V (1a) and
0.63 V (1b), corresponding to the Ru(Ⅲ)/Ru(Ⅳ) process. A
higher Ru(Ⅱ)/Ru(Ⅲ) oxidation potential has been reported for
[Ru(η5-C5H5)Cl(PPh3)2], [Ru(η5-C5Me5)Cl(PPh3)2], [Ru(η5-
indenyl)Cl(PPh3)2], and [Ru(Tp)Cl(PPh3)2], in ClCH2CH2Cl
(Epa = 0.53–0.83 V, Epc = 0.40–0.71 V vs Ag/Ag+ at the scan
rate of 0.1 V/s) compared to that for 1a or 1b. In addition, no
peak corresponding to the oxidation of Ru(Ⅲ) to Ru(Ⅳ) was
seen within the potential window in these cases.11a The report-
ed examples for a Ru(Ⅲ)/Ru(Ⅳ) process are few.11b,c,d On the
other hand, in the case of the present complex, 1a or 1b , the
Ru(Ⅳ) state will be attained by the reaction with appropriate
oxidants. Organic halides are a possible candidate for the oxi-
dative reaction of 1a or 1b to Ru(Ⅳ) products; however, at-
tempted treatment of 1a or 1b with MeI or benzyl chloride
only afforded green paramagnetic products, presumably a ha-
logeno-Ru(Ⅲ) compound, showing several broad 1H resonanc-
es at −12 to 28 ppm. In sharp contrast, reaction of allyl ha-
lides with 1a actually gave the corresponding diamagnetic
Ru(Ⅳ) products.
Other typical reactions of coordinatively unsaturated com-
plexes are oxidative addition reactions.1a,3a–e,i,m,v It is well
known that the oxidative addition reactions are involved in var-
ious catalytic cycles,2 and numerous studies on the oxidative
addition of a wide variety of addenda have been undertaken.
As reported earlier,7a,b we discovered oxidative addition of al-
lylic halides to [Ru(η5-C5R5)X(L)2] (R = H, Me; L = PPh3,
CO) to afford [Ru(η3-allyl)(η5-C5R5)X2], which is a rare exam-
ple of a reaction involving change of the oxidation state of the
complex from Ru(Ⅱ) to Ru(Ⅳ). Other Ru(Ⅱ) or Ru(Ⅲ) precur-
sors, [Ru(η5-C5H5)Cl(COD)],8a,c [Ru(η5-C5Me5)Cl(COD)],8d
[Ru(η5-C5H5)(NCMe)3]+,8e [Ru(η5-C5Me5)Cl2]n,7c and [Ru(η5-
C5Me5)Cl]4,8b were later proved to be active towards the oxida-
tive addition of allylic substrates. In particular, a successful re-
action of [Ru(η5-C5Me5)Cl]4, which might produce coordina-
tively unsaturated “Ru(η5-C5Me5)Cl” species in solution,8b,9
indicates that appropriate coordinatively unsaturated Ru(Ⅱ)
compounds could be useful as precursors to produce Ru(Ⅳ)-
η3-allyl complexes by the oxidative addition of allylic sub-
strates.
Treatment of a violet pentane solution of 1a with allyl chlo-
ride at room temperature resulted in a dramatic color change of
the solution to yellow, from which yellow solids were precipi-
tated. NMR analysis revealed that the solids contained
[Ru(η3-CH2CHCH2)(η5-C5Me5)(η2-iPrNC(Me)wNiPr)]+Cl−
In this paper, we describe the reaction of [Ru(η5-C5Me5)(η-
amidinate)] with allylic substrates, leading to synthesis of nov-
el cationic Ru(Ⅳ) compounds: [Ru(η3-allyl)(η5-C5Me5)(η2-
amidinate)]+. The formed Ru(Ⅳ) products were reactive to-
wards nucleophiles, and catalytic nucleophilic substitution of
allylic substrates with several nucleophiles was achieved by
[Ru(η5-C5Me5)(η-amidinate)] or [Ru(η3-allyl)(η5-C5Me5)(η2-
amidinate)]+. A part of this paper appeared as a communica-
tion,5b and full details are given in this paper.
(2a) and
a
small amount of [Ru(η3-CH2CHCH2)(η5-
C5Me5)Cl2] (7)7b (2a:7 = 98:2). No paramagnetic by-product
was observable in the product. 1H NMR spectrum of 2a
showed three signals: two doublets (δ 2.32, J = 10.1 Hz, δ
4.03, J = 6.0 Hz) and a doublet of triplet (δ 4.94, J = 6.0, 10.1
Hz) in the integral ratio of 2:2:1, due to the typical symmetric
η3-allyl moiety. 1H resonances derived from the amidinate
ligand appeared as two doublets (δ 1.23 and 1.30, J = 7.0 Hz)
due to the diastereotopic methyl groups and a septet (δ 3.22, J
= 7.0 Hz) due to the methine proton of the isopropyl group.
The methyl group on the central carbon of the amidinato
Results and Discussion
η5
Oxidative Addition of Allylic Substrates to [Ru(
-
C5Me5)( -amidinate)]. Two ruthenium amidinates, [Ru(η5-
C5Me5)(η-iPrNC(Me)wNiPr)] (1a) and [Ru(η5-C5Me5)(η-
tBuNC(Ph)wNtBu)] (1b), were prepared according to the pro-
cedure reported earlier.5a These ruthenium amidinates are
good π-donors and are expected to undergo facile oxidation
(Table 1). In fact, cyclic voltammmograms of 1a or 1b in THF
showed a quasi-reversible one-electron oxidation wave (1a;
Epa = −0.32 V, Epc = −0.46 V, 1b; Epa = −0.24 V, Epc = −
0.43 V vs Ag/Ag+ at the scan rate of 0.1 V/s) which is assign-
able to the Ru(Ⅱ)/Ru(Ⅲ) oxidation process.10 Another irrevers-
ligand appeared as a singlet at δ 1.92. These H NMR data
1
η
suggest that 2a has a Cs-symmetric structure, as shown in
Scheme 1, which has a mirror plane containing the central car-
bon of the η3-allyl ligand, the methyl carbon and the central
carbon of the amidinate ligand, the ruthenium atom, and the
center of the C5Me5 ligand. The Cs-symmetric structure was
also supported by the appearance of nine 13C resonances. The
13C signal due to the Cp-ring carbon appeared at 104.9 ppm,
which was significantly downfield (ca. 35 ppm) compared with
that of 1a, and the location was similar to that of a typical
Ru(Ⅳ)-η3-allyl complex, 7.
Table 1. Reactions of [Ru(η5-C5Me5)(η-amidinate)] with
Allylic Halides
The chloro complex 2a was unstable in THF, acetone, or
dichloromethane, and gradually decomposed to intractable
products. This instability, which is presumably triggered by
nucleophilic attack of the chloride counter anion to the η3-allyl
or amidinate moiety, made separation of 2a from 7 difficult.
Exchange of the chloride anion of 2a by weakly coordinating
PF6 anion resulted in formation of [Ru(η3-CH2CHCH2)(η5-
Amidinate
Complex
Allylic Halide
Product (Yield, %)a)
R1
H
H
R2
H
X
1a
1b
1a
1a
1b
1a
Cl
Cl
Br
Cl
Cl
Cl
2a (98)
2b (95)
3a (77)
4a (98)
4b (96)
5b (70)
7 (2)
7 (5)
8 (15)
9 (2)
9 (4)
H
−
C5Me5)(η2-iPrNC(Me)wNiPr)]+PF6 (2a-PF6). Structure of
H
H
H
H
Ph
Me
Me
H
2a-PF6 was evidenced by similar spectral features of 2a-PF6 to
2a, which was supported by crystallographic determination of
its homologues, 2b-PF6 and 5a-PF6 (vide infra). The ORTEP
drawings of 2b-PF6 and 5a-PF6 are shown in Fig. 1, and se-
10 (20)
a) Determined by 1H NMR.

H. Kondo et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1929
Fig. 1. The ORTEP drawings of 2b-PF6 (left) and 5a-PF6 (right) with thermal ellipsoids drawn at the 50% probability level. PF6 an-
ions are omitted for clarity.
Table 2. Selected Bond Distances (Å) and Angles (°) for
2b-PF6 and 5a-PF6
2b-PF6
5a-PF6
Ru1–av.C1–5
Ru1–C11
Ru1–C12
Ru1–C13
Ru1–N1
Ru1–N2
N1–C14
N2–C14
C11–C12
C12–C13
2.263 (4)
2.193 (5)
2.132 (4)
2.206 (5)
2.128 (3)
2.125 (3)
1.345 (5)
1.333 (5)
1.385 (8)
1.379 (8)
2.251 (4)
2.179 (4)
2.172 (4)
2.301 (4)
2.104 (4)
2.083 (3)
1.334 (6)
1.318 (5)
1.421 (6)
1.391 (6)
N1–Ru1–N2
Ru1–N1–C14
Ru1–N2–C14
N1–C14–N2
C11–Ru1–C12
C11–Ru1–C13
C12–Ru1–C13
62.0 (1)
93.2 (2)
94.4 (2)
109.7 (3)
37.3 (2)
64.1 (2)
37.0 (2)
61.6 (1)
94.6 (3)
96.0 (3)
107.8 (4)
38.1 (2)
64.1 (2)
36.1 (2)
Scheme 1.
This is generally seen in coordinatively saturated ruthenium
amidinates such as [Ru(η5-C5Me5) (η2-amidinate)(CO)] (the
corresponding angle = 3.9(4)°). The average Ru–C (terminal
carbons of the allyl group) bond distances of 2b-PF6 [2.200(5)
Å] are similar to those reported for a Ru(Ⅳ)-allyl complex,
[Ru(η3-CH2CHCH2)(η5-C5Me5)Br2]7b or its analogues (2.20–
2.24 Å),8a,c,d while the average C–C distance of the π-allyl
ligand in 2b-PF6 (1.38 Å) is significantly shorter than that of
[Ru(η3-CH2CHCH2)(η5-C5Me5)Br2]7b or its analogues (1.41–
1.48 Å).8a,c,d These bond distances of 2b-PF6 are similar to
those seen in the two isomers of a half sandwich Ru(Ⅱ)-allyl
complex, [Ru(η3-CH2CMeCH2)(η5-C5H5)(CO)];12 the Ru–
C(terminal of allyl) and C–C(allyl)av distances are 2.20 and
1.40 Å, respectively.12a Orientation of the allyl group of 2b-
PF6 is endo, and variable temperature NMR studies showed
that there was no equilibrium with the corresponding exo-iso-
mer. This endo orientation is generally seen in half-sandwich
Ru(Ⅳ)–allyl complexes. We also determined the crystal struc-
lected bond distances and angles are summarized in Table 2.
Complex 2b-PF6 has a square-pyramidal structure with two
nitrogen atoms of the amidinato ligand and terminal carbons of
the η3-allyl ligand at the basal positions. The Ru–N [2.128(3)
and 2.125(3) Å] and average Ru–C(Cp) [2.263(4) Å] bonds of
2b-PF6 are longer than those of 1b [1b: Ru–N 2.073(3) Å, Ru–
C(Cp)av 2.158(4) Å],5a and these bond distances are similar to
those of a CO adduct of 1b, [Ru(η5-C5Me5) (η2-amidi-
nate)(CO)] [Ru–N 2.133(5) Å, Ru–Cav 2.240(7) Å].5a As re-
ported earlier, a plane consisting of the Ru atom and two nitro-
gen atoms in 1b makes an angle of 48.9(4)° with a plane of the
amidinate N–C–N moiety; this contributes to better interaction
of π-electrons of the amidinato ligand with the ruthenium cen-
ter to mitigate the coordinatively unsaturated nature.5a In sharp
contrast, the corresponding angle of 2.6(4)° in 2b-PF6 shows
that four atoms, N1, Ru1, N2, and C14 are almost in a plane.

Ⅱ
1930 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Oxidative Addition to Ru (C5Me5)(amidinate)
Scheme 4.
Scheme 2.
Scheme 5.
strates: the reaction of 1a with crotyl chloride gave a mixture
of [Ru(η5-C5Me5)Cl(η4-butadiene)] (12)13 and [Ru(η3-
MeCHCH2CH2)(η5-C5Me5)Cl2] (11)7b in 35 and 37% yields, as
shown in Scheme 4. Formation of 12 can be attributed to rapid
β-hydrogen elimination, which is one of the general features of
organoruthenium(Ⅱ) compounds.14 A possible reaction mech-
anism is shown in Scheme 5. Similarly, treatment of 1-ace-
toxy-2-butene or 3-acetoxy-1-butene with 1b in THF at room
Scheme 3.
ture of 5a-PF6, in which a phenyl group is bonded with a ter-
minus of the η3-allyl group. The phenyl group is located at the
syn-position of the allyl fragment similar to [Ru(η3-
CH2CHCHPh)(η5-C5Me5)Cl2].8d Other structural features of
5a-PF6 are analogous to those of 2b-PF6.
1
temperature for 1 h gave a mixture of compounds. The H
Other allylic halides summarized in Table 1 also reacted
with 1a or 1b to give the corresponding [Ru(η3-allyl)(η5-
C5Me5)(η2-amidinate)]+X− (X = Cl or Br) in good yields. In
all cases, some amounts of [Ru(η3-allyl)(η5-C5Me5)X2]7b,8d
were formed as the by-product. As shown in Schemes 2 and 3
as representative examples, treatment of [Ru(η3-allyl)(η5-
C5Me5)(η2-amidinate)]+Cl− with NH4PF6 in CHCl3 resulted in
successful anion exchange reactions to give the corresponding
PF6 salt. A BF4 analogue was also obtained using AgBF4 in-
stead of NH4PF6. The PF6 salts were alternatively obtainable
by treatment of 1a or 1b with allylic chlorides in THF in the
presence of NaPF6. The PF6 or BPh4 salts were available in
higher yields by the reactions with allylic acetates or allylic
carbonates, as shown in Table 3.
NMR spectrum of this mixture showed that they contained
products similar to 12; signals due to the η4-butadiene were
seen at δ 1.61 (overlapping with the C5Me5 signal), 3.52 (dd, J
= 1.6, 8.3 Hz, 2H), and 4.57 (m, 2H). It was proposed that ac-
tivation of 2-(trimethylsilylmethyl)-2-propenyl acetate by
Pd(0) species resulted in formation of a η3-allylpalladium in-
termediate, followed by attack of the resulting acetate anion to
the trimethylsilyl group to furnish a trimethylenemethane pal-
ladium species, as shown in Scheme 6.15 Although similar tri-
The following interesting observations were available from
the experiments which changed the structure of allylic sub-
Table 3. Reactions of [Ru(η5-C5Me5)(η-amidinate)] with
Allylic Acetates or Carbonates in the Presence of NaPF6 or
NaBPh4
Amidinate
Complex
Allylic Compound
Product
(Isolated Yield/%)
X
OAc
OAc
R1
H
H
R2
H
H
H
H
Me
Me
1a
1b
1a
1b
1a
1b
1a
1a
1a
2a-PF6 (90)
2b-PF6 (97)
2a-PF6 (90)
2b-PF6 (97)
4a-PF6 (90)
4b-PF6 (98)
5a-PF6 (96)
6a-PF6 (99)
6a-BPh4 (69)
OCO2Me
OCO2Me
OCO2Me
OCO2Me
OCO2Me
OAc
H
H
H
H
Ph
H
H
H
CH2SiMe3
CH2SiMe3
OAc
Scheme 6.

H. Kondo et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1931
perature in the presence of NaPF6.
Stoichiometric and Catalytic Transformation of Allylic
η5
η
Substrates by [Ru( -C5Me5)( -amidinate)]. Reactions of
η3-allyl complexes with nucleophiles are a well investigated is-
sue in organometallic chemistry.2,15–17 In particular, extensive
studies on η3-allyl palladium species revealed that stoichio-
metric as well as catalytic transformation of allylic substrates
was easily accomplished by a wide variety of nucleophiles.15–18
Mitsudo and co-workers reported analogous stoichiometric re-
actions of nucleophiles with η3-allyl ruthenium compounds as
well as ruthenium-catalyzed reactions of allylic carbonates
with nucleophiles.8d,19 Although the ruthenium-mediated reac-
tions are less versatile for application to various allylic sub-
strates or nucleophiles than the reactions using palladium com-
plexes, in some cases there exist interesting differences in cata-
lytic activity and product selectivity.8d,19 The η3-allyl com-
plexes presented in this paper are cationic, and are expected to
undergo facile reactions with nucleophiles. In fact, addition of
one equivalent of PhLi in ether to a THF solution of 2b-PF6 at
room temperature for 1 h afforded a mixture of 1-phenyl-2-
propene and 3-phenyl-1-propene in a ratio of 2:1 in 70% yield,
whereas attempted reactions of 2b-PF6 with ethylene (1 atm)
or benzaldehyde resulted in complete recovery of 2b-PF6.
Similar reaction of 2b-PF6 with an equimolar amount of dime-
thyl methylsodiomalonate in THF at 0 °C for 1 h furnished the
formation of dimethyl allylmethylmalonate in 93% yield. Re-
generation of 1b (74% yield) was confirmed from the NMR
spectrum of the reaction mixture. Treatment of 5a-PF6 with an
equimolar amount of piperidine at 0 °C for 1 h gave a mixture
of 1-phenyl-3-piperidino-1-propene and its HPF6 salt, and re-
covery of 1a was also observed (71%). By passing this mix-
ture through a short pad of Al2O3, 1-phenyl-3-piperidino-1-
propene was obtained in 97% yield (Scheme 9). The corre-
sponding regio-isomer, 3-phenyl-3-piperidino-1-propene, was
not formed.
Scheme 7.
methylenemethane formation may occur in the reaction of 1a
with 2-(trimethylsilylmethyl)-2-propenyl acetate, stable η3-al-
lyl complexes 6a-PF6 and 6a-BPh4 were formed in 98% and
69% yields, respectively (Scheme 6).
As noted earlier, the reaction of 1a or 1b with MeI or benzyl
chloride did not afford the corresponding diamagnetic Ru(Ⅳ)
alkyl complex. This suggests that involvement of coordination
of a carbon–carbon double bond in the allylic substrates may
play a crucial role in the oxidative addition reaction. Coordi-
nation of a carbon–carbon double bond of allyl acetate to 1a or
1b as the intermediates for the oxidative addition was evi-
denced by variable temperature 1H NMR, as shown in Scheme
7. In the spectra of a 1:1 mixture of 1a or 1b with allyl acetate
in THF-d8 at −80 °C, 1H and 13C resonances due to the
H2CwCH moiety appeared at the significantly more upfield re-
gion (∆δH = 3 ppm, ∆δc = 70 ppm), due to coordination to the
ruthenium center, than those derived from the uncoordinated
allyl acetate. The chemical shifts were similar to those ob-
served in the spectra of [Ru(η5-C5Me5)(η2-amidinate)(η2-
CH2wCH2)].5a As described earlier, two methyl groups in the
isopropyl group in 1a appeared as a single doublet due to the
C2v symmetry of the molecule. Upon coordination of ethylene,
they were observed as two doublets because of the Cs symme-
try. The lack of symmetry in the compound formed by coordi-
nation of allyl acetate to 1a made all of the methyl signals un-
equal; thus they appear as four doublets. These data are con-
sistent with the structure 13a. It is likely that the coordination
of the carbon–carbon double bond of allyl acetate facilitates
the carbon-oxygen bond cleavage, as shown in Scheme 8. In
fact, the cationic η3-allyl complex 2a-PF6 or 2b-PF6 was
formed quantitatively via the displacement of the OAc group
by PF6 upon warming the solution of 13a or 13b to room tem-
Similar to [Ru(η5-C5Me5)(OMe)]2 and other half-sand-
20
Scheme 8.
Scheme 9.

Ⅱ
1932 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Oxidative Addition to Ru (C5Me5)(amidinate)
Table 4. Catalytic Allylations of Piperidine or Dimethyl Methylmalonate with Allylic Methyl Carbonates
a)
Entry
H-Nu
R
Cat
1a
overall yield/%a)
1
2
3
4
Ph
Ph
Ph
Ph
> 99
> 99
> 99
> 99
(> 99
90
23
13
100
15
17
21
:
77
87
0
85
83)c)
79
1b
5a-PF6
:
:
:
:
Ruꢀb)
5
Me
1a
6
7
8
9
10
11
H
H
H
H
Ph
Me
1a
1b
2b-PF6
Ruꢀb)
1b
40
—
—
—
—
:
51 (100)d)
no reaction
no reaction
2 (18)e)
100
100
0
0
1b
12 (37)e)
:
Ⅱ
a) Determined by 1H NMR. b) Ru = [Ru(η5-C5Me5)(OMe)]2 (0.5 eq.). c) Reported by T. Mitsudo et al.,
see Ref. 8d. d) Reaction condition: room temperature, 4h. e) Reaction condition: 45 °C, 6 h.
wich Ru(Ⅱ) complexes reported by Mitsudo,8d,19 the amidina-
toruthenium(Ⅱ) complexes, 1a and 1b, act as the catalysts for
the allylation of various nucleophiles. As selected examples,
reactions of allyl methyl carbonates with piperidine and dime-
thyl methylmalonate were examined. As summarized in Table
4, the reactions of cinnamyl methyl carbonate or crotyl methyl
carbonate with piperidine in the presence of 5 mol% of 1a or
1b at 0 °C proceeded in good yields within 1 h. Catalytic ac-
tivity was similar to that of [Ru(η5-C5Me5)(OMe)]2. Piperi-
dine moiety was generally introduced favorably at the more
substituted position between two possible reaction points of
the allyl moiety by catalysis of either 1a, 1b, or [Ru(η5-
C5Me5)(OMe)]2. Interestingly, 5a-PF6 also catalyzed the reac-
tion of cinnamyl methyl carbonate with piperidine to give N-
cinnamylpiperidine as the single product. This regioselectivity
is similar to that obtained by the stoichiometric reaction de-
scribed above, but is apparently different from those available
by catalysis of 1a, 1b, or [Ru(η5-C5Me5)(OMe)]2.
A significant difference in reactivity of 1a or 1b from that of
[Ru(η5-C5Me5)(OMe)]2 was seen in the reaction of allyl meth-
yl carbonate with dimethyl methylmalonate; 1a or 1b gave the
corresponding product in moderate yields, whereas no reaction
took place with [Ru(η5-C5Me5)(OMe)]2. The reaction was
sensitive towards substituents of the allyl carbonates; the reac-
tions of crotyl methyl carbonate or cinnamyl methyl carbonate
were slower.
Ruthenium-catalyzed Carrole reactions shown in Scheme
10 were also achieved using 1a, 1b, or [Ru(η5-C5Me5)(OMe)]2
as the catalyst. Three products A–C were obtained by treat-
ment of allyl acetoacetate with the ruthenium catalyst (Eq. 1)
at 0 °C , while the reaction shown in Eq. 2 proceeded at 70 °C,
Scheme 10.
giving the desired product D and a small amount of enone E.
In these cases, little difference in rate and selectivity was seen
between 1a, or 1b, or [Ru(η5-C5Me5)(OMe)]2 as the catalyst.
ed by Tsuji and co-workers.18
As described above, we found that the reaction of 1a or 1b
Similar reactions catalyzed by palladium were well investigat-

H. Kondo et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1933
strates mediated by 1a, 1b, or [Ru(η3-allyl)(η5-C5Me5)(η2-
amidinate)]+ were achieved. Involvement of the oxidative ad-
dition of allylic substrates to 1a or 1b affording [Ru(η3-al-
lyl)(η5-C5Me5)(η2-amidinate)]+ and the subsequent reaction
with nucleophiles supports the reaction mechanism shown in
Scheme 11. The interesting effect of PF6 anion described in
the text requires further investigation.
Experimental
General Methods. All manipulations were carried out under
a dry argon atmosphere using the standard Schlenk tube technique
associated with a high-vacuum line and a glove box under a puri-
fied N2 atmosphere. All solvents were distilled over appropriate
drying reagents prior to use (Et2O, THF, pentane; Ph2CO/Na,
CH2Cl2, CHCl3; CaH2). An ether solution of PhLi was prepared
from PhBr and Li wire, and was titrated prior to use. Allylic com-
pounds and piperidine were distilled before use. [Ru(η5-
C5Me5)(η-amidinate)]
was
synthesized
from
[Ru(η5-
C5Me5)(OMe)]2 and lithium amidinate according to the procedure
reported previously.5a Column chromatography was performed
with Merck, neutral aluminum oxide 90 (70–230 mesh ASTM).
1H and 13C NMR spectra were recorded on JEOL Lambda 600 and
Lambda 400 spectrometers at ambient temperature unless other-
wise noted. 1H and 13C NMR chemical shifts are reported in ppm
relative to solvent resonances. All coupling constants (J) are re-
ported in Hz. IR spectra were recorded on a JASCO FT/IR-550
spectrometer. Elemental analyses were performed at the Analysis
Center in the Faculty of Science of Kyushu University.
Scheme 11.
with allylic carbonate afforded the oxidative adducts [Ru(η3-
allyl)(η5-C5Me5)(η2-amidinate)]+ in good yield. Stoichiomet-
ric reactions of [Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+ with
nucleophiles gave rise to allylation of these nucleophiles and
regeneration of [Ru(η5-C5Me5)(η-amidinate)]. Thus, a possi-
ble catalytic cycle can be illustrated as Scheme 11, in which
the oxidative adducts [Ru(η3-allyl)(η5-C5Me5)(η2-amidi-
nate)]+ are involved as key intermediates.23
η5
η
Reactions of Allylic Halides with [Ru( -C5Me5)( -amidi-
η3
nate)] (1a,1b). I. Preparation of Chloro-Ru(Ⅳ)- -allyl
Complexes. In a typical example, 1a (70 mg, 0.185 mmol) was
dissolved in dry pentane (ca. 10 mL) in a Schlenk tube. To the re-
sulting clear purple solution, allyl chloride (15 µL, 0.184 mmol)
was added at room temperature. The color of the solution imme-
diately changed to yellow, from which yellow precipitates were
formed. After 1 h, the solvent was removed under a reduced pres-
sure to give a yellow solid (84 mg) containing 2a and 7. Assign-
ment of 7 was performed according to the literature.7b Yields of
2a and 7 were determined by 1H NMR spectrometry on the basis
of the integrated intensity of internal standard (mesitylene). (2a:
98%, 7:2%). By a similar procedure, products described in Table
1 and Scheme 4, 2b, 3a, 4a, 4b, 5a, 8,7b 9,7b 10,8d 11,7b and 12,13
were obtained. These new halo-Ru(Ⅳ)-π-allyl complexes, 2a, 2b,
3a, 4a, 4b, and 5a, are unstable at room temperature, and attempts
to remove a small amount of 7, 8, 9, and 10 by chromatography
and fractional crystallization failed. Therefore, assignment of the
products was carried out by NMR spectroscopy, and full charac-
terization was made using PF6 or BF4 analogue obtained by the
following anion exchange reaction or by direct preparation from
2a or 2b.
Conclusion
Although several coordinatively unsaturated organorutheni-
um(Ⅱ) complexes bearing 16 valence electrons have success-
fully been isolated and there reactivity has been subjected to
study, none of them was referred to oxidative addition of allyl-
ic substrates. We discovered the oxidative addition of allylic
substrates to the isolable coordinatively unsaturated organoru-
thenium(Ⅱ) complex 1a or 1b, which offers a new cationic or-
ganoruthenium(Ⅳ) complex stabilized by a nitrogen-donor
ligand, [Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+. The strong
donor property of the ruthenium center of 1a or 1b derived
from the C5Me5 and amidinato ligands contributes to facile ox-
idative addition of allylic substrates. In contrast, the formed
[Ru(η3-allyl)(η5-C5Me5)(η2-amidinate)]+ is an acceptor, and
facilely reacted with nucleophiles. There are several examples
of organoruthenium(Ⅳ) compounds; however, only a few ex-
amples of the complexes bearing nitrogen donor ligand have
been reported to our knowledge.3v,21 Reaction of halogens
with certain Ru(Ⅱ) precursors bearing nitrogen donor ligands
reportedly resulted in formation of halogeno-Ru(Ⅲ) prod-
ucts.3t,22 As described earlier, the reaction of organic halides
with 1a or 1b also gave rise to formation of halogeno-Ru(Ⅲ)
species. Since the complexes of late transition metals in higher
oxidation states are one of the unexplored fields in organome-
tallic chemistry, we believe that these findings contribute to the
progress of this field.
II. Preparation of PF6 and BF4 Salts from Chloro-Ru(Ⅳ)-
-allyl Complexes. In a typical example, the mixture (65 mg)
η3
of 2a (0.140 mmol) and 7 was treated with NH4PF6 (23 mg, 0.141
mmol) in CHCl3 (ca. 10 mL) at −78 °C. The reaction mixture
was allowed to warm to room temperature and stirred at this tem-
perature for 1 h. After removal of the solvent in vacuo, purifica-
tion of the residue by column chromatography [alumina (0.5 cmΦ
× 1.5 cm); CH2Cl2] was followed by recrystallization from
CH2Cl2–Et2O to form 2a-PF6 (56 mg, 0.089 mmol, 64% yield;
yellow crystals). By a similar procedure, 2b-PF6, 2b-BF4, and
4b-PF6 were obtained from 2b and 4b.
Stoichiometric and catalytic transformations of allylic sub-

Ⅱ
1934 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Oxidative Addition to Ru (C5Me5)(amidinate)
2a: yellow microcrystals. 1H NMR (CDCl3) δ 1.23 (d, J =
7.0 Hz, 6H, CH(CH3)2), 1.30 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.79
(s, 15H, C5(CH3)5), 1.92 (s, 3H, CCH3), 2.32 (d, J = 10.1 Hz,, 2H,
anti-CH of the allyl group), 3.22 (sep, J = 7.0 Hz, 2H, CH(CH3)2),
4.03 (d, J = 6.0 Hz, 2H, syn-CH of the allyl group), 4.94 (dt, J =
6.0 Hz, 10.1, 1H, central-CH of the allyl group). 13C{1H}NMR
(CDCl3) δ 10.3 (C5(CH3)5), 22.5 (CCH3), 25.7, 26.1 (CH(CH3)2),
51.1 (CH(CH3)2), 60.8 (CH2 of the allyl group), 98.6 (CH of the
allyl group), 104.9 (C5(CH3)5), 174.3 (NCN).
2a-PF6: Orange crystals. Anal. Calcd for C21H37F6N2PRu: C,
44.76; H, 6.62; N, 4.97%. Found: C, 44.75; H, 6.58; N, 4.90%. 1H
NMR (in CDCl3) δ 1.22 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.29 (d, J
= 7.0 Hz, 6H, CH(CH3)2), 1.71 (s, 15H, C5(CH3)5), 1.90 (s, 3H,
CCH3), 2.09 (d, J = 10.3 Hz,, 2H, anti-CH of the allyl group),
3.20 (sep, J = 7.0 Hz, 2H, CH(CH3)2), 4.00 (d, J = 6.1 Hz,, 2H,
syn-CH of the allyl group), 4.93 (dt, J = 6.1 Hz,, 10.3, 1H, cen-
tral-CH of the allyl group). 13C{1H} NMR (CDCl3) δ 9.8
(C5(CH3)5), 22.4 (CCH3), 25.6 (CH(CH3)2), 25.9 (CH(CH3)2),
51.0 (CH(CH3)2), 60.2 (CH2 of the allyl group), 98.3 (CH of the
allyl group), 104.9 (C5(CH3)5), 174.3 (NCN). IR (KBr,) 3422,
2979, 2932, 1520, 1485, 1457, 1214, 1140, 839cm−1. mp: 138 °C
(dec.).
(CH of the allyl group), 106.7 (C5(CH3)5), 127.5, 127.6, 127.9,
129.9, 132.7, 138.6 (C6H5), 178.9 (NCN). IR (KBr) 3421, 3005,
2965, 1437, 1395, 1360, 1289, 2207, 1182, 1048, 954, 880, 802,
751, 718 cm−1. mp: 160 °C (dec.).
1
4a: Yellow microcrystals. H NMR (CDCl3) δ 1.20 (d, J =
7.0 Hz, 6H, CH(CH3)2), 1.24 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.71
(s, 15H, C5(CH3)5), 1.90 (s, 3H, CCH3), 2.07 (s, 3H, CH3 of the
methallyl group), 2.18 (s, 2H, anti-CH of the methallyl group),
3.13 (sep, J = 7.0 Hz, 2H, CH(CH3)2), 3.58 (s, 2H, syn-CH of the
methallyl group). 13C{1H} NMR (in CDCl3) δ 10.2 (C5(CH3)5),
17.6 (CH3 of the methallyl group), 22.8 (CCH3), 24.6 (CH(CH3)2),
26.0 (CH(CH3)2), 50.3 (CH(CH3)2), 58.3 (CH2 of the allyl group),
104.6 (C5(CH3)5), 113.1 (CH of the allyl group), 172.2 (NCN).
4a-PF6: Yellow microcrystals. Anal.
Calcd for
C22H39F6N2PRu: C, 45.75; H, 6.81; N, 4.85%. Found: C, 45.84;
H, 6.76; N, 4.82%. 1H NMR (CDCl3) δ 1.23 (d, J = 7.0 Hz, 6H,
CH(CH3)2), 1.27 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.68 (s, 15H,
C5(CH3)5), 1.92 (s, 3H, CCH3), 2.05 (s, 2H, anti-CH of the methal-
lyl group), 2.10 (s, 3H, CH3 of the methallyl group), 3.16 (sep, J
= 7.0 Hz, 2H, CH(CH3)2), 3.60 (s, 2H, syn-CH of the methallyl
group). 13C{1H} NMR (CDCl3) δ 9.8 (C5(CH3)5), 17.5 (CH3 of
the methallyl group), 22.8 (CCH3), 24.6 (CH(CH3)2), 26.0
(CH(CH3)2), 50.3 (CH(CH3)2), 58.0 (CH2 of the allyl group),
104.6 (C5(CH3)5), 113.0 (CH of the allyl group), 172.3 (NCN). IR
3a: Yellow microcrystals. 1H NMR (CDCl3) δ 1.21 (d, J =
7.0 Hz, 6H, CH(CH3)2), 1.28 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.76
(s, 15H, C5(CH3)5), 1.81 (s, 3H, CCH3), 2.25 (d, J = 10.1 Hz,, 2H,
anti-CH of the allyl group), 3.20 (sep, J = 7.0 Hz, 2H, CH(CH3)2),
4.01 (d, J = 6.0 Hz,, 2H, syn-CH of the allyl group), 4.92 (dt, J =
6.0 Hz,, 10.1, 1H, central-CH of the allyl group). 13C{1H} NMR
(CDCl3) δ 10.2 (C5(CH3)5), 22.4 (CCH3), 25.6 (CH(CH3)2), 25.9
(CH(CH3)2), 51.0 (CH(CH3)2), 60.6 (CH2 of the allyl group), 98.4
(CH of the allyl group), 104.8 (C5(CH3)5), 174.2 (NCN).
(KBr) 2981, 1519, 1489, 1384, 1337, 1213, 1138, 843, 557 cm−1
mp: 184 °C (dec.).
.
4b: Yellow microcrystals. 1H NMR (CDCl3) δ 0.91 (s, 18H,
C(CH3)3), 1.82 (s, 15H, C5(CH3)5), 2.40 (s, 2H, anti-CH of the
methallyl group), 2.41 (s, 3H, CH3 of the methallyl group), 4.06
(s, 2H, syn-CH of the methallyl group), 7.15–7.40 (m, 5H, C6H5).
13C{1H} NMR (in CDCl3) δ 11.3 (C5(CH3)5), 18.2 (CH3 of the
methallyl group), 35.6 (C5(CH3)5), 57.0 (C(CH3)3), 58.5 (CH2 of
the methallyl group), 106.1 (C5(CH3)5), 112.8 (central-C of the
methallyl group), 127.0, 127.3, 128.2, 129.8, 133.1, 137.9 (C6H5),
177.2 (NCN).
2b: Yellow microcrystals. 1H NMR (CDCl3) δ 0.95 (s, 18H,
C(CH3)3), 1.90 (s, 15H, C5(CH3)5), 2.51 (d, J = 10.4 Hz, 2H, anti-
CH of the allyl group), 4.59 (d, J = 6.3 Hz, 2H, syn-CH of the al-
lyl group), 5.36 (dt, J = 6.3 Hz, 10.4, 1H, central-CH of the allyl
group), 7.14 (m, 1H, C6H5), 7.24 (m, 1H, C6H5), 7.31 (m, 1H,
C6H5), 7.36 (m, 1H, C6H5), 7.44 (m, 1H, C6H5). 13C{1H} NMR (in
CDCl3) δ 11.4 (C5(CH3)5), 35.5 (C(CH3)3), 58.0 (C(CH3)3), 60.4
(CH2 of the allyl group), 97.4 (CH of the allyl group), 106.5
(C5(CH3)5), 127.5, 127.6, 127.8, 130.0, 132.6, 138.6 (C6H5), 179.2
(NCN).
4b-PF6: Orange crystals. Anal. Calcd for C29H45F6N2PRu:
C, 52.16; H, 6.79; N, 4.20%. Found: C, 52.31; H, 6.80; N, 4.26%.
1H NMR (CDCl3) δ 0.96 (s, 18H, C(CH3)3), 1.79 (s, 15H,
C5(CH3)5), 2.28 (s, 2H, anti-CH of the methallyl group), 2.46 (s,
3H, CH3 of the methallyl group), 4.06 (s, 2H, syn-CH of the meth-
allyl group), 7.21 (m, 1H, C6H5), 7.25 (m, 1H, C6H5), 7.32 (m, 1H,
C6H5), 7.34 (m, 1H, C6H5), 7.43 (m, 1H, C6H5). 13C{1H} NMR
(CDCl3) δ 11.0 (C5(CH3)5), 18.3 (CH3 of the methallyl group),
35.7 (C5(CH3)5), 57.1 (C(CH3)3), 58.1 (CH2 of the methallyl
group), 106.6 (C5(CH3)5), 112.5 (central-C of the methallyl
group), 127.2, 127.3, 128.4, 129.9, 133.4, 138.2 (C6H5), 177.1
2b-PF6: Orange crystals. Anal. Calcd for C28H43F6N2PRu: C,
1
51.45; H, 6.63; N, 4.29%. Found: C, 51.22; H, 6.62; N, 4.34%. H
NMR (CDCl3) δ 0.95 (s, 18H, C(CH3)3), 1.81 (s, 15H, C5(CH3)5)),
2.22 (d, J = 10.2 Hz,, 2H, anti-CH of the allyl group), 4.53 (d, J =
6.1 Hz,, 2H, syn-CH of the allyl group), 5.36 (dt, J = 6.1 Hz,,
10.2, 1H, central-CH of the allyl group), 7.16 (m, 1H, C6H5), 7.24
(m, 1H, C6H5), 7.32 (m, 1H, C6H5), 7.35 (m, 1H, C6H5), 7.44 (m,
1H, C6H5). 13C{1H} NMR (CDCl3) δ 10.9 (C5(CH3)5), 35.5
(C(CH3)3), 58.0 (C(CH3)3), 59.7 (CH2 of the allyl group), 97.2
(CH of the allyl group), 106.6 (C5(CH3)5), 127.4, 127.6, 127.8,
129.9, 132.8, 138.6 (C6H5), 178.9 (NCN). IR (KBr) 3437, 2966,
1442, 1208, 1183, 840 cm−1. mp: 164 °C (dec.).
(NCN). IR (KBr) 3437, 3001, 2966, 1439, 1207, 1183, 842 cm−1
.
mp: 184 °C (dec.).
5a: Yellow microcrystals. 1H NMR (CDCl3) δ 0.47 (d, J =
7.0 Hz, 3H, CH(CH3)2), 0.98 (d, J = 7.0 Hz, 3H, CH(CH3)2), 1.32
(d, J = 7.0 Hz, 3H, CH(CH3)2), 1.33 (d, J = 7.0 Hz, 3H,
CH(CH3)2), 1.83 (s, 3H, CCH3), 1.84 (s, 15H, C5(CH3)5), 2.27
(sep, J = 7.0 Hz, 1H, CH(CH3)2), 2.52 (d, J = 9.8 Hz, 1H, CHH
(anti)), 3.18 (sep, J = 7.0 Hz, 1H, CH(CH3)2), 4.16 (d, J = 6.1 Hz,
1H, CHH (syn)), 4.28 (d, J = 10.8 Hz, 1H, CHPh), 5.62 (ddd, J =
6.1, 9.8, 10.8 Hz, 1H, CH2CH), 7.29–7.42 (m, 5H, Ph). 13C{1H}
NMR (in CDCl3) δ 10.4 (C5(CH3)5), 23.6 (CCH3), 24.8, 25.7,
26.2, 26.7 (CH(CH3)2), 47.7, 50.4 (CH(CH3)2), 60.0 (CH2 of the
cinnamyl group), 86.0 (CHPh of the cinnamyl group), 95.6
(CH2CH of the cinnamyl group), 104.6 (C5(CH3)5), 128.6, 129.3,
129.6, 136.9 (Ph), 173.8 (NCN).
2b-BF4: Orange crystals. Anal. Calcd for C28H43BF4N2Ru: C,
56.47; H, 7.28; N, 4.70%. Found: C, 56.00; H, 7.24; N, 4.70%. 1H
NMR (CDCl3) δ 0.94 (s, 18H, C(CH3)3), 1.82 (s, 15H, C5(CH3)5),
2.42 (d, J = 10.4 Hz, 2H, anti-CH of the allyl group), 4.54 (d, J =
6.1, 2H, syn-CH of the allyl group), 5.36 (dt, J = 6.1, 10.4 Hz,
1H, central-CH of the allyl group), 7.15 (m, 1H, C6H5), 7.25 (m,
1H, C6H5), 7.31 (m, 1H, C6H5), 7.35 (m, 1H, C6H5), 7.43 (m, 1H,
C6H5).
(C(CH3)3), 58.0 (C(CH3)3), 59.7 (CH2 of the allyl group), 97.2
13C{1H} NMR (CDCl3) δ 11.0 (C5(CH3)5), 35.5
5a-PF6: Yellow microcrystals. Anal.
Calcd for

H. Kondo et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1935
C27H41F6N2PRu: C, 50.70; H, 6.46; N, 4.38%. Found: C, 50.21;
H, 6.40; N, 4.26%. 1H NMR (CDCl3) δ 0.45 (d, J = 7.0 Hz, 3H,
CH(CH3)2), 0.95 (d, J = 7.0 Hz, 3H, CH(CH3)2), 1.28 (d, J = 7.0
Hz, 3H, CH(CH3)2), 1.30 (d, J = 7.0 Hz, 3H, CH(CH3)2), 1.73 (s,
15H, C5(CH3)5), 1.81 (s, 3H, CCH3), 2.23 (sep, J = 7.0 Hz, 1H,
CH(CH3)2), 2.24 (d, J = 9.1 Hz, 1H, CHH (anti)), 3.13 (sep, J =
7.0 Hz, 1H, CH(CH3)2), 4.05 (d, J = 10.8 Hz, 1H, CHPh), 4.10 (d,
J = 6.1 Hz, 1H, CHH (syn)), 5.61 (ddd, J = 6.1, 9.1, 10.8 Hz, 1H,
CH2CH), 7.31–7.40 (m, 5H, Ph). 13C{1H} NMR (in CDCl3) δ 9.9
(C5(CH3)5), 23.4 (CCH3), 24.7, 25.5, 26.0, 26.6 (CH(CH3)2), 47.7,
50.3 (CH(CH3)2), 59.5 (CH2 of the cinnamyl group ), 85.2 (CHPh
of the cinnamyl group), 95.4 (CH2CH of the cinnamyl group),
104.6 (C5(CH3)5), 128.6, 129.3, 129.3, 137.0 (Ph), 173.9 (NCN).
IR (KBr) 2968, 2933, 1522, 1487, 1463, 1437, 1380, 1360, 1338,
1210, 1138, 839, 700 cm−1. mp: 142 °C (dec.).
Direct Preparation of PF6 and BPh4 Salts from 2a or 2b.
A mixture of 1a (42 mg, 0.111 mmol), allyl chloride (11 µL,
0.135 mmol) and NaPF6 (19 mg, 0.113 mmol) in THF (5 mL) was
stirred at room temperature for 1 h. After removal of the solvent
in vacuo, the residue was purified by column chromatography
[alumina (0.5 cmΦ ×1.0 cm); CH2Cl2]. Recrystallization of the
product from CH2Cl2–Et2O gave 2a-PF6 in 62% yield (39 mg; yel-
low crystal). Alternatively, 1a (70 mg, 0.185 mmol) was treated
with allyl acetate (20 µL, 0.185 mmol) in the presence of NaPF6
(31 mg, 0.185 mmol) in THF (10 mL) at room temperature. The
mixture was stirred at room temperature for 1 h. After removal of
the solvent in vacuo, the residue was extracted with several por-
tions of CH2Cl2 (10 mL) and the combined extracts were filtered
through Celite to remove insoluble materials. After removal of the
solvent in vacuo, analytically pure 2a-PF6 was obtained without
further purification (90% yield, 94 mg ). By a similar procedure,
products described in Table 3 were isolated.
NMR Observation of 13a and 13b. 1a (20 mg, 0.053 mmol)
was dissolved in THF-d8 (0.5 mL) in a 5 mm o.d. NMR tube. To
the resulting purple solution, allyl acetate (5.7 µL, 0.053 mmol)
was added at −78 °C. The color of solution changed to yellow
immediately. 1H, 13C{1H} and HMQC NMR spectra of 13a were
recorded at −80 °C. Using a similar procedure, 13b was detected.
13a: 1H NMR (THF-d8, −80 °C) δ 0.76 (d, J = 6.3 Hz, 3H,
CH(CH3)2), 0.90 (d, J = 6.3 Hz, 3H, CH(CH3)2), 1.27 (d, J = 6.3
Hz, 3H, CH(CH3)2), 1.30 (d, J = 6.3 Hz, 3H, CH(CH3)2), 1.60 (s,
15H, C5(CH3)5), 1.71 (s, 3H, CCH3 or CH3CO2), 1.99 (s, 3H,
CCH3 or CH3CO2), 2.45 (d, J=7.1 Hz, 1H, CHwCH2), 2.58 (m,
1H, CHwCH2), 2.91 (d, J = 9.9 Hz, 1H, CHwCH2), 3.22 (sep, J =
6.3 Hz, 1H, CH(CH3)2), 3.52 (sep, J = 6.3 Hz, 1H, CH(CH3)2),
3.80 (dd, J = 10.8, 10.8 Hz, 1H, CH2OAc), 4.75(dd, J = 2.6, 10.8
Hz, 1H, CH2OAc). 13C{1H} NMR (in THF-d8, −80 °C) δ 11.1
(C5(CH3)5), 17.6 (CCH3 or CH3CO2), 21.2 (CCH3 or CH3CO2),
24.9, 25.6, 26.6, 26.6 (CH(CH3)2), 47.8 (CH(CH3)2), 47.9
(CHwCH2), 51.4 (CH(CH3)2), 59.1 (CHwCH2), 70.8 (CH2OAc),
89.8 (C5(CH3)5), 167.1 (NCN or CH3CO2), 170.9 (NCN or
CH3CO2).
13b: 1H NMR (in THF-d8, −80 °C) δ 0.82 (s, 9H, C(CH3)3),
1.04 (s, 9H, C(CH3)3), 1.65 (s, 15H, C5(CH3)5), 2.03 (s, 3H,
CH3CO2), 2.60 (d, J = 7.5 Hz, 1H, CHwCH2), 3.03 (d, J = 9.8
Hz, 1H, CHwCH2), 3.13 (m, 1H, CHwCH2), 3.84 (t, J = 11.0 Hz,
1H, CH2OAc), 4.99 (dd, J = 3.7, 11.0 Hz, 1H, CH2OAc), 7.23–
7.36 (m, 5H, C6H5). 13C{1H} NMR (THF-d8, −80 °C) δ 10.7
(C5(CH3)5), 21.2 (CH3CO2), 35.4, 36.0 (C(CH3)3), 48.4
(CHwCH2), 53.8, 55.4 (C(CH3)3), 56.7 (CHwCH2), 70.9
(CH2OAc), 91.0 (C5(CH3)5), 127.4, 127.5, 128.9, 130.8, 134.6,
141.3 (C6H5), 170.4 (NCN or CH3CO2), 170.8 (NCN or CH3CO2).
Reactions of 2b-PF6 and 5a-PF6 with Nucleophiles. In
a
typical example, an ether solution of PhLi (0.27 mL, 0.54 M) was
added to the THF yellow solution (10 mL) of 2b-PF6 (70 mg, 0.15
mmol) at 0 °C, and the mixture was stirred at this temperature for
1 h. Color of the reaction mixture changed from yellow to black.
After removal of the solvent in vacuo, an oily residue was ob-
tained (103 mg). This residue contained 3-phenyl-1-propene and
1-phenyl-1-propene in a ratio of 2:1 in 70% yield. The yield and
6a-PF6: Yellow microcrystals. 1H NMR (CDCl3) δ 0.18 (s,
9H, Si(CH3)3) 1.23 (d, J = 7.0 Hz, 6H, CH(CH3)2), 1.25 (d, J =
7.0 Hz, 6H, CH(CH3)2), 1.66 (s, 2H, CH2SiMe3), 1.68 (s, 15H,
C5(CH3)5), 1.93 (s, 2H, CHH (anti)), 1.91 (s, 3H, CCH3), 3.17
(sep, J = 7.0 Hz, 2H, CH(CH3)2), 3.42 (s, 2H, CHH (syn)).
13C{1H} NMR (CDCl3) δ −1.7 (Si(CH3)3), 9.8 (C5(CH3)5), 22.4
(CH2SiMe3), 22.8 (CCH3), 25.0, 26.0 (CH(CH3)2), 50.2
(CH(CH3)2), 54.8 (C(CH2)2), 104.4 (C5(CH3)5), 118.6
(CH2C(CH2)2), 172.2 (s, NCN). 29Si{1H} NMR (CDCl3) δ 5.1 (s,
SiMe3). IR (KBr) 2977, 1511, 1487, 1465, 1383, 1343, 1330,
1251, 1212, 1139, 841 cm−1. mp: 170 °C (dec.). 6a-PF6 is unsta-
ble in the air, and elementary analysis of 6a-PF6 has not been ac-
complished. Therefore, full characterization was carried out using
the BPh4 analogue.
1
ratio were determined by H NMR spectrometry on the basis of
the integrated intensity of internal standard (mesitylene). The
spectral data were in accord with those of commercially available
authentic samples. Using a similar procedure, reaction of 2b-PF6
with a THF solution of Na[MeC(CO2Me)2] (0.63 M), which was
prepared from MeHC(CO2Me)2 and NaH before use, was carried
out, and afforded a mixture of dimethyl allylmethylmalonate and
1b in 93% and 74% yields, respectively. The reaction of 5a-PF6
with piperidine gave a mixture of 1-phenyl-3-piperidino-1-pro-
6a-BPh4: Yellow microcrystals. Anal.
Calcd for
1
C49H67BN2RuSi: C, 71.42; H, 8.19; N, 3.40%. Found: C, 70.77;
H, 8.18; N, 3.37%. 1H NMR (CDCl3) δ 0.21 (s, 9H, Si(CH3)3),
1.20 (d, J = 7.0 Hz, 12H, CH(CH3)2), 1.45 (s, 15H, C5(CH3)5),
1.63 (s, 2H, CH2SiMe3), 1.78 (s, 2H, CHH (anti)), 1.84 (s, 3H,
CCH3), 3.06 (sep, J = 7.0 Hz, 2H, CH(CH3)2), 3.31 (s, 2H, CHH
(syn)), 6.89 (m, 4H, B(C6H5)4), 7.04 (m, 8H, B(C6H5)4), 7.42 (m,
8H, B(C6H5)4). 13C{1H} NMR (CDCl3) δ −1.6 (Si(CH3)3, 9.9
(C5(CH3)5), 22.5 (CH2SiMe3), 22.7 (CCH3), 25.0, 26.0
(CH(CH3)2), 50.2 (CH(CH3)2), 54.8 (C(CH2)2), 104.3 (C5(CH3)5),
pene, its HPF6 salt, and 1a (71%), which were determined by H
NMR spectrometry. The spectral data of the HPF6 salt was in ac-
cord with those of the sample prepared by reaction with 1-phenyl-
3-piperidino-1-propene and aqueous HPF6. When the sample was
passed through a short pad of Al2O3 by elution with THF, 1-phe-
nyl-3-piperidino-1-propene was obtained in 97% yield.
General Procedure for the Catalytic Allylation of Piperidine
and MeHC(CO2Me)2 and Decarboxylative Allylation. In
a
typical procedure, a THF solution (0.5 mL) of 1a (9.5 mg, 0.025
mmol) was added to a mixture of cinnamyl carbonate (80 µL, 0.5
mmol), piperidine (50 µL, 0.5 mmol), and THF (0.5 mL) at 0 °C,
and the mixture was stirred at this temperature. After 1 h, the re-
sulting yellow mixture was transferred to the head of a short pad
of silica gel to remove metal species. Eluents available by Et2O
(80 mL) were concentrated in vacuo to give a mixture of 1-phenyl-
118.7 (CH2C(CH2)2), 121.5 (p-CH of B(C6H5)4), 125.4 (q, 2JBC
=
2.6 Hz, o-CH of B(C6H5)4), 136.3 (m-CH of B(C6H5)4), 164.2 (q,
1JBC = 49.3 Hz, ipso-C of B(C6H5)4), 172.3 (s, NCN). 29Si{1H}
NMR (CDCl3) δ 6.0 (s, SiMe3). IR (KBr) 3419, 3054, 3033,
2983, 1579, 1525, 1481, 1428, 1379, 1360, 1335, 1260, 1249,
1212, 1135, 853, 749, 732, 703, 610 cm−1. mp: 178 °C (dec.).

Ⅱ
1936 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Oxidative Addition to Ru (C5Me5)(amidinate)
3-piperidino-1-propene and 3-phenyl-3-piperidino-1-propene
(100 mg).8d Yields of these products were determined by 1H NMR
spectrometry on the basis of the integrated intensity of internal
standard (mesitylene). By a similar procedure, the products18,24 in
Table 4 and Scheme 10 were obtained.
Crystallographic data have been deposited at the CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK and copies can be obtained on re-
quest, free of charge, by quating the publication citation and depo-
sition numbers CCDC 146850 and 164869. The data are also de-
posited as Document No. 74048 at the Office of the Editor of Bull.
Chem. Soc. Jpn.
X-ray Data Collection, Solution, and Refinement of Struc-
ture. Single crystals of 2b-PF6 and 5a-PF6 were grown from a
mixture of CH2Cl2 and ether. Each X-ray crystallography mea-
surement was performed on a Rigaku AFC-7R four circle axis dif-
fraction meter in the case of 2b-PF6, and on a Rigaku RAXIS
RAPID imaging plate diffraction meter in the case of 5a-PF6 with
graphite monochromated Mo Kα radiation (λ = 0.71070 Å). The
data were collected at 293(2) K (2b-PF6) and 153(2) K (5a-PF6).
The structures were solved by direct method (SIR92)25 in the case
of 2b-PF6 and by Patterson method (DIRDIF99 PATTY)26 in the
case of 5a-PF6. Atomic coordinates and the anisotropic thermal
parameters of non-hydrogen atoms were refined by full-matrix
least squares on F2 (SHELXL97-2).27 The positions of all hydro-
gen atoms were calculated assuming idealized geometry. Several
cycles of a full-matrix least-squares refinement with anisotropic
temperature factors for non-hydrogen atoms led to final R1 and
wR2 values (I > 2σ(I)). All calculations were performed on a Pen-
tium computer. Crystallographic data are summarized in Table 5.
We are grateful to Dr. Kouki Matsubara (Kyushu Univ.) for
his help in the X-ray analyses. Part of this work was financial-
ly supported by the Japan Society for the Promotion of Science
(Grants-in-Aid for Scientific Research 10450343 and
13450374).
References
1
For recent reviews on coordinatively unsaturated complex-
es, and references cited therein: a) R. Poli, Chem. Rev., 96, 2135
(1996). b) R. Poli, Acc. Chem. Res., 30, 494, (1997). c) K. G.
Caulton, New. J. Chem., 18, 25 (1994).
2
a) R. H. Crabtree, “The Organometallic Chemistry of the
Transition Metals,” John Wiley & Sons, New York (1994), p. 206.
b) J. P. Collman, L. S. Hegedus, J. R. Norton, and R. G. Finke,
“Principles and Applications of Organotransition Metal Chemis-
try,” University Science Books, Mill Valley, CA (1987), p. 521.
3
For recent studies of 16 electron half sandwich rutheni-
Table 5. Crystallographic Date for 2b-PF6 and 5a-PF6
um(Ⅱ) complexes: Stabilization by both steric effect and π-donor
ligand; a) B. K. Campion, R. H. Heyn, and T. D. Tilley, J. Chem.
Soc., Chem. Commun., 1988, 278. b) B. K. Campion, R. H. Heyn,
and T. D. Tilley, J. Chem. Soc., Chem. Commun., 1992, 1201. c)
T. Arliguie, C. Border, B. Chaudret, J. Devillers, and R. Poilblanc,
Organometallics, 8, 1308 (1989). d) T. J. Johnson, J. C. Huffman,
and K. G. Caulton, J. Am. Chem. Soc., 114, 2725 (1992). e) T. J.
Johnson, P. S. Coan, and K. G. Caulton, Inorg. Chem., 32, 4594
(1993). f) C. C. Bickford, T. J. Johnson, E. R. Davidson, and K.
G. Caulton, Inorg. Chem., 33, 1080 (1994). g) T. J. Johnson, K.
Folting, W. E. Streib, J. D. Martin, J, C, Huffman, S. A. Jackson,
O. Eisenstein, and K. G. Caulton, Inorg. Chem., 34, 488 (1995).
h) E. Lindner,. M. Haustein, H. A. Mayer, K. Gierling, R. Fawzi,
and M. Steimann, Organometallics, 14, 2246 (1995). i) T. Braun,
M. Laubender, O. Gevert, and H. Werner, Chem. Ber./Recueil,
130, 559 (1997). j) J. Huang, H.-J. Schanz, E. D. Stevens, and S.
P. Nolan, Organometallics, 18, 2370 (1999). k) J. Huang, H.-J.
Schanz, E. D. Stevens, S. P. Nolan, and J. L. Petersen, Organome-
tallics, 18, 2370 (1999). l) M. J. Tenorio, M. C. Puerta, and P.
Valerga, J. Organomet. Chem., 609, 161 (2000). Stabilization by
steric effect; m) M. J. Tenorio, K. Mereiter, M. C. Puerta, and P.
Velerga, J. Am. Chem. Soc., 122, 11230 (2000). Stabilization by
π-donor ligand; n) K. Mashima, H. Kaneyoshi, S.-I. Kaneko, A.
Mikami, K. Tani, and A. Nakamura, Organometallics, 16, 1016
(1997). o) K. Mashima, S.-I. Kaneko, K. Tani, H. Kaneyoshi, and
A. Nakamura, J. Organomet. Chem., 545, 345 (1997). p) K-J.
Haack, S. Hashiguchi, A. Fujii, T. Ikariya, and R. Noyori, Angew.
Chem., Int. Ed. Engl., 36, 285 (1997). q) P. J. Bailey, K. J. Grant,
and S. Parsons, Organometallics, 17, 551 (1998). r) M.
Herberhold, H. Yan, and W. Milius, J. Organomet. Chem., 598,
142 (2000). s) M. Herberhold, H. Yan, W. Milius, and B.
Wrackmeyer, J. Organomet. Chem., 604, 170 (2000). Stabiliza-
tion by σ-donor ligand; t) C. Gemel, K. Mereiter, R. Schmid and
K. Kirchner, Organometallics, 16, 5601 (1997). u) C. Gemel, V.
N. Sapunov, K. Mereiter, M. Ferencic, R. Schmid, and K.
Kirchner, Inorg. Chim. Acta, 286, 114 (1999). v) C. Gemel, J. C.
Huffman, K. G. Caulton, K. Mauthner, and K. Kirchner, J. Orga-
Complex
2b-PF6
5a-PF6
Formula
Formula weight
Temp/K
Habit
Crystal size/mm
C28H43F6N2PRu
653.68
293(2)
Orange prism
0.34×0.25×
0.18
Orthorhombic
Pbca
C27H41F6N2PRu
639.66
153(2)
Orange Prism
0.50×0.50×
0.25
Monoclinic
P21/c
12.5817(12)
9.2301(9)
24.013(2)
90.762(3)
2788.4(5)
4
Crystal system
Space group
a/Å
b/Å
c/Å
31.771(6)
14.034(4)
13.366(5)
β/deg
V/Å3
5961(3)
8
Z
Dcalcd/gc m−3
Diffractometer
1.457
1.524
AFC-7R
(Rigaku)
Mo Kα
(0.71070)
graphite
0.637
RAXIS Rapid
(Rigaku)
Mo Kα
(0.71070)
graphite
0.679
Radiation(λ/Å)
Monochromator
µ
calcd/cm−1
F(000)
2704
1320
Scan type
ω−2θ
2.56 < θ < 27.50
6849
θ range/deg
No. of date
collected
No. of used date
(Fo > 2σ(Fo))
No. of variables
GOF
R1(I > 2σ(I))
wR2(I > 2σ(I))
R1(all data)
wR2(all data)
∆ρ max/e Å−3
1.70 < θ < 27.48
5347
4403
4441
343
334
1.037
0.0492
0.1418
0.0917
0.1620
0.833
1.091
0.0525
0.1212
0.0660
0.1304
0.571

H. Kondo et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1937
Jørgensen, Chem. Rev., 98, 1689 (1998).
nomet. Chem., 593, 342, (2000).
4
For reviews on amidinato ligands: a) J. Barker and M.
17 For review; J. Tuji, “Palladium Reagents and Catalysis,”
John Wiley & Sons, New York (1995).
18 a) I. Shimizu, T. Yamada, and J. Tsuji, Tetrahedoron Lett.,
21, 3199 (1980). b) T. Tsuda, Y. Chujo, S. Nishi, K. Tawara, and
T. Saegusa, J. Am. Chem. Soc., 102, 6381 (1980). c) J. Tsuji, M.
Nisar, I. Shimizu and I. Minami, Synthesis, 1984, 1009
19 a) S.-W. Zhang, T. Mitsudo, T. Kondo, and Y. Watanabe, J.
Organomet. Chem., 450, 197 (1993). b) Y. Morisaki, T. Kondo,
and T. Mistudo, Organometallics, 18, 4742 (1999). c) T. Kondo,
Y. Morisaki, S. Uenoyama, K. Wada, and T. Mitsudo, J. Am.
Chem. Soc., 121, 8657 (1999).
Kilner, Coord. Chem. Rev., 133, 219 (1994). b) F. T. Edelmann,
Coord. Chem. Rev., 137, 403 (1994).
5
a) Y. Yamaguchi and H. Nagashima, Organometallics, 19,
725 (2000). b) H. Kondo, Y. Yamagychi, and H. Nagashima,
Chem. Commun., 2000, 1075. c) H. Kondo, Y. Yamaguchi, and H.
Nagashima, J. Am. Chem. Soc., 123, 500 (2001).
6
DFT calculations were carried out using the GAUSSIAN
98 package on the B3LYP level of theory with SDD(Ru),
D95v(N), and D95*(C,H): K. Kirchner, unpublished results.
7
a) H. Nagashima, K. Mukai and K. Itoh, Organometallics,
3, 1314 (1984). b) H. Nagashima, K. Mukai, Y. Shiota, K.-I. Ara,
K. Itoh, H. Suzuki, N. Oshima, and Y. Moro-oka, Organometal-
lics, 4, 1314 (1985). c) H. Nagashima, K. Mukai, Y. Shiota, K.
Yamaguchi, K. Ara, T. Fukahori, H. Suzuki, M. Akita, Y. Moro-
oka, and K. Itoh, Organometallics, 9, 799 (1990).
20 U. Koelle and J. Kossakowski, Inorg. Synth., 29, 225
(1992).
21 A. N. Sahay, D. S. Pandey, and M. G. Walawalkar, J. Or-
ganomet. Chem., 613, 250 (2000).
22 For another example of Ru(Ⅲ) complexes being obtained
rather than the expected Ru(Ⅳ) complexes bearing N-donor
ligand: B. R. Manzano, F. A. Jalon, F. J. Lahoz, B. Chaudret, and
D. Montauzon, J. Chem. Soc., Dalton. Trans., 1992, 977.
23 There are several interesting features of this ruthenium-
amidinate catalyzed allylation, in contrast to those achieved by
Mitsudo’s catalyst ([Ru(η5-C5Me5)(OMe)]2). The first is success-
ful reaction of dimethyl methylmalonate, which cannot be accom-
plished by catalysis of [Ru(η5-C5Me5)(OMe)]2. The reaction cata-
lyzed by 2b-PF6 failed, but in the presence of Et3N (one equiv. to
allyl methyl carbonate) it proceeded to give the corresponding
product in 21% yield. These results suggest the importance of in
situ generation of dimethyl methylmalonate anion. Facile oxida-
tive addition of allyl methyl carbonate to 1a or 1b leads to forma-
tion of CO2 and MeO−, which may take part in facile generation
of the malonate anion. The second is the regioselectivity of the
products in the allylation of piperidine; in the reactions catalyzed
by 1a or 1b, the regioselectivity was similar to that available by
catalysis of [Ru(η5-C5Me5)(OMe)]2, whereas completely different
regioselectivity was available in the reactions catalyzed by 5a-
PF6, which is rather similar to the ratio obtained by the stoichio-
metric reaction shown in Scheme 9. Further investigation is re-
quired for the explanation of these results; however, it is worth-
while to point out that the ruthenium amidinate, 1a or 1b, might
act as the catalyst only at the initial stage of the catalytic reaction,
and the net catalytic species is [Ru(η5-C5Me5)(OMe)]2 produced
during the reaction. In fact, the reaction of 1a with allyl methyl
8
a) M. O. Albers, D. C. Liles, D. J. Robinson, A. Shaver,
and E. Singleton, Organometallics, 6, 2347 (1987). b) P. J. Fagan,
W. S. Mahoney, J. C. Calabrese, and I. D. Williams, Organometal-
lics, 9, 1843 (1990). c) K. Kirchner, K. Mereiter, A. Umfahrer,
and R. Schmid, Organometallics, 13, 1886 (1994). d) T. Kondo,
H. Ono, N. Satake, T. Mitsudo, and Y. Watanabe, Organometal-
lics, 14, 1945 (1995). e) E. Rüba, W. Simanko, K. Mauthner, K.
M. Soldouzi, C. Slugovc, K. Mereiter, R. Schmid, and K.
Kirchner, Organometallics, 18, 3843 (1999).
9
P. J. Fagan, M. D. Ward, and J. C. Calabrese, J. Am. Chem.
Soc., 111, 1698 (1989).
10 Cyclic voltammetric studies were carried out using a BAS
50B/W electrochemical analyzer under N2 in a glovebox at 0.1
Vs−1 in a THF solution (3 mL) containing (nBu)4N+PF6 (0.1 M) at
room temperature. A three-electrode cell was used which was
equipped with a platinum disk working electrode, a platinum wire
counter electrode, and a silver reference electrode comprised of a
silver wire in contact with 0.01 M AgNO3 and 0.2 M (nBu)4N+PF6
in acetonitrile.
11 a) T. Ando, M. Kamigaito, and M. Sawamoto, Macromole-
cules, 33, 5825 (2000). b) M. Haga, H. Sakamoto, and H. Suzuki,
J. Organometal. Chem., 377, C77 (1989). c) U. Koelle and J.
Kossakowski, Inorg. Chim. Acta, 164, 23 (1989). d) T. Rüther, U.
Englert, and U. Koelle, Inorg. Chem., 37, 4265 (1998).
12 a) L.-Y. Hsu, C. E. Nordman, D. H. Gibson, and W.-L.
Hsu, Organometallics, 8, 241 (1989). b) D. H. Gibson, W.-L.
Hsu, A. L. Steinmetz, and B. V. Johnson, J. Organomet. Chem.,
208, 89 (1981).
−
carbonate in the absence of PF6 gave a mixture of products in-
cluding [Ru(η5-C5Me5)(OMe)]2 (30% yield) at ambient tempera-
ture. It was also confirmed that 1a or 1b instantly reacted with
MeOH to afford [Ru(η5-C5Me5)(OMe)]2.
13 K. Masuda, H. Ohkita, S. Kurumatani, and K. Itoh, Orga-
nometallics, 12, 2221 (1993).
14 For example affording a framework of “[Ru(η5-C5Me5)(η4-
butadiene)]” via β-hydogen elimination from Ru(Ⅳ) complex,
see; H. Nagashima, Y. Michino, K. Ara, T. Fukahori, and K. Itoh,
J. Organomet. Chem., 406, 189 (1991).
24 T. Hayashi, M. Kawatsura, and Y. Uozumi, J. Am. Chem.
Soc., 120, 1681 (1998).
25 SIR92: A. Altomare, M. Cascarano, C. Giacovazzo, and A.
Guagliardi, J. Appl. Crystallogr., 26, 343 (1994).
26 PATTY: P.T. Beurskens, G. Admiraal, G. Beurskens, W.P.
Bosman, S. Garcia-Granda, R.O. Gould, J.M.M. Smits, and C.
Smykalla, (1992). The DIRDIF program system, Technical Re-
port of the Crystallography Laboratory, University of Nijmegen,
The Netherlands.
15 For review on trimethylenemethane complexes, and refer-
ences cited therein: a) B. M. Trost, Angew. Chem., Int. Ed. Engl.,
25, 1 (1986). b) P. J. Harrington, “Comprehensive Organometallic
Chemistry Ⅱ,” ed by E. W. Abel, F. G. A. Stone, G. Wilkinson, Per-
gamon, Oxford, U.K. (1995), Vol. 12.
16 For reviews on allylation reactions by using η3-allyl com-
plexes, and references cited there in; a) B. M. Trost and D. L. Van
Vranken, Chem. Rev., 96, 395 (1996). b) M. Johannsen and K. A.
27 SHELX97 package. Sheldrick, G. M. University of
Göttingen, Germany (1997).