Synthesis and molecular structure of [RuCl{C(=CHPh)OC(=O)CH2CH3}(CO)(PPh3) 2]: A real intermediate in ruthenium complex-catalyzed selective synthesis of a (Z)-enol ester

Article abstract of DOI:10.1016/S0022-328X(00)00027-9

Reaction of [RuCl(η²-O₂CCH₂CH₃)(CO)(PPh ₃)₂] (1) and phenylacetylene gives [RuCl{C(=CHPh)OC(=O)CH₂CH₃}(CO)(PPh₃) ₂] (2a). The X-ray structure analysis of 2a reveals that it includes a (Z)-enol ester-like 1-propanoyloxy-2-phenylethenyl-C¹,O ligand. In the catalytic addition of propanoic acid to phenylacetylene, the complex 2a acts as a real intermediate that gives (Z)-2-phenylethenyl propanoate, selectively. The presence of the free PPh₃ in the reaction mixture depresses formation of some dicarbonylruthenium species that catalytically produce (E)- and Markovnikov-type enol esters.

Full text of DOI:10.1016/S0022-328X(00)00027-9

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 601 (2000) 69–77
Synthesis and molecular structure of
[RuCl{C(ꢀCHPh)OC(ꢀO)CH2CH3}(CO)(PPh3)2]: a real
intermediate in ruthenium complex-catalyzed selective synthesis of
a (Z)-enol ester
Hiroyuki Kawano a,*, Yoshiko Masaki b, Takahiro Matsunaga b, Katsuma Hiraki b,
Masayoshi Onishi b, Taro Tsubomura c
a Graduate School of Marine Science and Engineering, Nagasaki Uni6ersity, Bunkyo-machi, Nagasaki 852–8521, Japan
b Department of Applied Chemistry, Faculty of Engineering, Nagasaki Uni6ersity, Bunkyo-machi, Nagasaki 852–8521, Japan
c Department of Industrial Chemistry, Faculty of Engineering, Seikei Uni6ersity, Kichijoji, Musashinoshi, Tokyo 180-8633, Japan
Received 4 October 1999; received in revised form 4 January 2000; accepted 14 January 2000
Abstract
Reaction of [RuCl(h2-O2CCH2CH3)(CO)(PPh3)2] (1) and phenylacetylene gives [RuCl{C(ꢀCHPh)OC(ꢀO)CH2CH3}-
(CO)(PPh3)2] (2a). The X-ray structure analysis of 2a reveals that it includes a (Z)-enol ester-like 1-propanoyloxy-2-phenylethenyl-
C1,O ligand. In the catalytic addition of propanoic acid to phenylacetylene, the complex 2a acts as a real intermediate that gives
(Z)-2-phenylethenyl propanoate, selectively. The presence of the free PPh3 in the reaction mixture depresses formation of some
dicarbonylruthenium species that catalytically produce (E)- and Markovnikov-type enol esters. © 2000 Elsevier Science S.A. All
rights reserved.
Keywords: Ruthenium complexes; Selective catalysis; Catalytic intermediate; Enol ester; X-ray structure
1. Introduction
catalytic system generally leads to the regioselective
formation [2,7]. However, in contrast, it is reported
rarely that the catalytic addition affords a (Z)-type enol
ester, regioselectively. Addition of benzoic acid onto
1-hexyne catalyzed by [Ru(h3-C4H7)2(dppe)] reported
by Dixneuf’s group [8] is the first example of the
selective synthesis of (Z)-enol esters. Very recently,
Matas et al. [9] have reported the second (Z)-selective
addition of ferrocenecarboxylic acid to phenylacetylene.
These two works, however, have shown no reason why
the (Z)-enol esters are produced predominantly in their
catalytic systems and gave no information about the
intermediary organometallic species.
From the viewpoint of organometallic complexes, on
the other hand, there have been some organoruthenium
complexes closely related to the catalytic formation of
enol esters reported. Daniel et al. [10] and Esteruelas et
al. [11] have independently synthesized a new class of
ruthenium(II) complexes having the enol ester-like 2-
substituted-1-acyloxyethenyl chelate ligands. In spite of
Direct addition of carboxylic acids to terminal alky-
nes with the aid of transition-metal complexes is a
powerful tool for preparing synthetically useful enol
esters [1] (see Scheme 1). In 1985, Mitsudo and Wata-
nabe’s group [2] and Dixneuf’s group [3] independently
achieved the direct addition under mild conditions us-
ing ruthenium complexes as catalysts. In most cases, the
catalytic addition promoted by the ruthenium com-
plexes affords the Markovnikov-type products, selec-
tively. Dixneuf and collaborators have applied the
ruthenium-catalyzed Markovnikov addition to the one-
pot syntheses of 2-acyloxy-1,3-dienes [4], enol formates
[5], and chiral 1,3-dioxolan-4-ones [6]. As for the (E)-
type products, the presence of phosphine ligands in the
* Corresponding author. Tel.: +81-95-8437281; fax: +81-95-
8476749.
E-mail address: hirokwn@net.nagasaki-u.ac.jp (H. Kawano)
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00027 -9

70
H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
Infrared spectra were recorded on a JASCO A-100
spectrometer using KBr tablets. GLPC analyses were
carried out on a Hitachi model 263-30 equipped with a
flame ionization detector and a 5 mmf×3 m stainless-
steel column (SE-30 or PEG 20M). NMR spectra were
obtained on a JEOL GX-400 spectrometer operating at
400 MHz for 1H, 101 MHz for 13C referenced to
Si(CH3)4, and at 162 MHz for 31P referenced to 85%
H3PO4 in water. Elemental analyses were performed on
a Yanaco MT-3 CHN Recorder and FABMS measure-
ments on a JEOL JMS-DX303 mass spectrometer at
the Center for Instrumental Analysis, Nagasaki
University.
Scheme 1. Transition-metal-catalyzed direct addition of a carboxylic
acid to a terminal alkyne producing the three types of enol esters.
2.1. Preparation of 1
Scheme 2. Production of 2a–c bearing the (Z)-enol ester-like chelate
ligand.
Preparation of 1 was carried out in a modified man-
ner according to the reported procedure [13]. A mixture
of 3 (1.06 g, 1.11 mmol) and propanoic acid (1.51 g,
20.4 mmol) in benzene (30 ml) was refluxed for 24 h
under nitrogen. After the reaction, the reaction mixture
was concentrated under a reduced pressure. A small
amount of the cis-isomer and propanoic acid were
removed from the crude product by silica-gel column
chromatography (eluent: CH2Cl2). The complex 1 was
isolated after CH2Cl2 was removed (yield 0.64 g, 75%).
Anal. Calc. for C40H35ClO3P2Ru: C, 63.03; H, 4.63.
Found: C, 63.32; H, 5.15%.
the structures of the chelating ligands, they have not
reported the catalytic aspects of their complexes.
Hence, there still exists a missing link between the
possible intermediary organometallic species and the
catalytic regioselective synthesis of (Z)-enol esters.
Here we report the reactions of a propanoatoruthen-
ium(II) complex, [RuCl(h2-O2CCH2CH3)(CO)(PPh3)2]
(1), with some terminal alkynes producing a series of
novel ruthenium(II) complexes, [RuCl{C(ꢀCHR)O-
C(ꢀO)CH2CH3}(CO)(PPh3)2] (R=Ph for 2a, R=
(CH2)4CH3 for 2b, and R=CH2OH for 2c) bearing
2 - substituted-1-propanoyloxyethenyl-C1,O-chelates,
namely, the (Z)-enol ester-like ligands (Scheme 2). The
molecular structure of 2a is determined by single-crystal
X-ray analysis. Moreover, the complex is found to be a
real intermediate in the catalytic synthesis of the (Z)-
enol ester; detailed studies reveal the missing link and
the overall reaction pathways of the catalytic addition
of propanoic acid to phenylacetylene producing the
(Z)-enol ester, regioselectively.
2.2. Preparation of
[RuCl{C(ꢀCHPh)OC(ꢀO)CH2CH3}(CO)(PPh3)2] (2a)
A mixture of 1 (214 mg, 0.28 mmol) and pheny-
lacetylene (0.47 g, 4.6 mmol) in benzene (20 ml) was
refluxed for 3 h under nitrogen. The yellow–orange
suspension turned smoothly into a solution. The reac-
tion mixture was concentrated to 1/4–1/5 of its original
volume under a reduced pressure. Addition of hexane
to the concentrated reaction mixture gave a crude 2a as
a yellow powder. Then the crude 2a was applied to the
silica-gel column chromatography using benzene–
dichloromethane (4:1) as eluent. Evaporation of the
eluent gave 2a in a pure form, as a dichloromethane
solvate (yield 186 mg, 63%). No satisfactory microanal-
ysis data have been obtained since the incorporated
dichloromethane is released readily even at room tem-
perature (r.t.). M.p. (dec.) 187–189°C. IR (cm−1):
2. Experimental
All experiments were performed under a dry nitrogen
atmosphere using standard Schlenk tube techniques.
However, any special precautions against air and mois-
ture were not taken in handling the complexes, since
most of the complexes were air-stable as solids and
stable for a short period in solution. All solvents were
dried and distilled over appropriate drying agents and
stored under nitrogen before use. The complex [Ru-
ClH(CO)(PPh3)3] (3) was prepared according to the
literature [12]. Phenylacetylene-d1 was prepared by
treating a solution of sodium phenylacetylide with D2O
in diethyl ether. All other reagents were purchased and
used without further purification unless otherwise
stated.
1
w(CꢁO) 1935 vs, w(CꢀO) 1630 vs, w(CꢀC) 1597 s. H-
NMR (CDCl3): l 0.64 (t, 3H, J(HH)=7.5 Hz, CH3),
1.80 (q, 2H, J(HH)=7.5 Hz, CH2), 4.41 (s, 1H, CH),
6.72 (d, 2H, J(HH)=7.3 Hz, o-H on the chelate lig-
and), 6.97 (t, 1H, J(HH)=7.3 Hz, p-H on the chelate
ligand), 7.10 (t, 2H, J(HH)=7.3 Hz, m-H on the
chelate ligand), 7.25 (m, 18H, p- and m-H on the PPh3),
7.61 (m, 12H, o-H on the PPh3). 13C{1H}-NMR
(CDCl3): l 8.3 (s, CH3), 25.5 (s, CH2), 121.1 (s, Cb),

H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
71
127.8 (s), 129.7 (s), 130.7 (t, J(CP)=21.5 Hz), 134.6 (s),
181.1 (s, CO2), 188.6 (t, J(CP)=13.6 Hz, Ca), 204.8 (t,
J(CP)=15.6 Hz, CꢁO). 31P{1H}-NMR (CDCl3): l 33.6
(s).
failure. The identification of 2c was achieved using IR
and NMR spectroscopic methods. IR (cm−1): w(CꢁO)
1
1935 vs, w(CꢀO) 1625 vs, w(CꢀC) 1600 sh. H-NMR
(CDCl3): l 0.63 (t, 3H, J(HH)=7.6 Hz, CH3), 1.84 (q,
2H, J(HH)=7.6 Hz, CH2), 3.50 (m, 2H, CH2), 3.75 (tt,
1H, J(HH)=7.0, J(HP)=1.9 Hz, CH), 7.36 (m, 18H,
p- and m-H on the PPh3), 7.71 (m, 12H, o-H on the
PPh3). 13C{1H}-NMR (CDCl3): l 8.2 (s, CH3), 25.4 (s,
CH2), 58.0 (s, CH2), 119.1 (s, Cb), 127.8 (s), 129.8 (s),
131.5 (t, J(CP)=21.5 Hz), 134.7 (s), 180.9 (s, CO2),
187.0 (t, J(CP)=13.6 Hz, Ca), 204.6 (t, J(CP)=15.6
Hz, CꢁO). 31P{1H}-NMR (CDCl3): l 32.8 (s).
2.3. Reaction of 1 with phenylacetylene-d1
A solution of 1 (101 mg, 0.13 mmol) and pheny-
lacetylene-d1 (0.78 g, 7.6 mmol) in benzene (20 ml) was
allowed to react under reflux. After 3 h, the solvent was
removed under a reduced pressure. The non-volatile
residue was dissolved to CDCl3 (0.5 ml) and applied to
NMR measurements. The 31P{1H}-NMR spectrum
showed that the entire starting complex was converted
2.6. Acidolysis of 2a producing the
dicarbonylruthenium(II) complexes
1
into 2a-d1 (l 33.6 (s)). The loss of the H-signal at l
4.41 and the splitting of the 13C{1H}-signal at l 121.1
(J(CD)=36 Hz) into three peaks indicated that the
i-carbon of the chelating ligand was deuterated. No
other significant change was found in the NMR
spectra.
A solution of 2a (44 mg, 0.051 mmol) and propanoic
acid (25 mg, 0.34 mmol) in benzene (5 ml) was allowed
to react in a sealed tube at 80°C. After 24 h, a small
portion of the mixture was analyzed by GLPC. The
GLPC analysis showed
a
peak of (Z)-PhCHꢀ
2.4. Preparation of
CHOC(ꢀO)CH2CH3; no peak due to the (E)- or
Markovnikov-type enol esters was detected. The rest of
the mixture was concentrated under a reduced pressure,
dissolved in C6D6 (0.5 ml) and applied to NMR analy-
sis. The 31P{1H}-NMR spectrum showed four singlets:
l 34.4 (1, 20%), 31.8 (14%), 24.1 (50%), and 15.4 (16%),
relative intensities shown in parentheses. The last three
signals were assigned to the following dicarbonylruthe-
nium(II) complexes, [Ru{h1-OC(ꢀO)C2H5}2(CO)2-
(PPh3)2] (4), [RuCl{h1-OC(ꢀO)C2H5}(CO)2(PPh3)2] (5),
and cct-[RuCl2(CO)2(PPh3)2] (6) [14], respectively. The
identification of the novel dicarbonyl complexes 4 and 5
was achieved by comparing their NMR data to those of
the authentic samples prepared according to the litera-
ture method [15]. The NMR spectroscopic data for 4,
1H-NMR (C6D6): l 0.87 (t, 6H, J(HH)=7.3 Hz, CH3),
1.83 (q, 4H, J(HH)=7.3 Hz, CH2), 7.00 (m, 6H, p-H),
7.09 (m, 12H, m-H), 8.04 (m, 12H, o-H). 31P{1H}-
NMR (C6D6): l 31.8 (s). The NMR spectroscopic data
[RuCl{C(ꢀCHC5H11)OC(ꢀO)CH2CH3}(CO)(PPh3)2]
(2b)
A mixture of 1 (75 mg, 0.10 mmol) and 1-heptyne (96
mg, 1.0 mmol) in benzene (25 ml) was refluxed for 72 h
under nitrogen. After the reaction was over, a similar
work-up procedure for 2a gave the complex 2b as a
yellow powder (yield 53 mg, 63%). Anal. Calc. for
C47H47ClO3P2Ru: C, 65.77; H, 5.52. Found: C, 65.65;
H, 5.80%. m.p. (dec.) 221–223°C. IR (cm−1): w(CꢁO)
1935 vs, w(CꢀO) 1635 vs, w(CꢀC) 1610 s. 1H-NMR
(CDCl3): l 0.54 (t, 3H, J(HH)=7.3 Hz, CH3), 0.81 (m,
2H, CH2), 0.82 (t, 3H, J(HH)=7.3 Hz, CH3), 0.98 (m,
2H, CH2), 1.16 (m, 2H, CH2), 1.63 (m, 4H, 2CH2), 3.51
(tt, 1H, J(HH)=5.8 Hz, J(HP)=2.2 Hz, CH), 7.34
(m, 18H, p- and m-H on the PPh3), 7.65 (m, 12H, o-H
on the PPh3). 13C{1H}-NMR (CDCl3): l 8.4 (s, CH3),
14.1 (s, CH3), 22.6 (s, CH2), 25.4 (s, CH2), 27.3 (s,
CH2), 29.1 (s, CH2), 31.8 (s, CH2), 121.3 (s, Cb), 127.7
(s), 129.6 (s), 131.6 (t, J(CP)=21.5 Hz), 134.6 (s), 179.0
(t, J(CP)=13.7 Hz, Ca), 180.6 (s, CO2), 205.4 (t,
J(CP)=13.7 Hz, CꢁO). 31P{1H}-NMR (CDCl3): l 33.9
(s).
1
for 5, H-NMR (C6D6): l 0.93 (t, 3H, J(HH)=7.7 Hz,
CH3), 1.94 (q, 2H, J(HH)=7.7 Hz, CH2), 6.97 (m, 6H,
p-H), 7.04 (m, 12H, m-H), 8.10 (m, 12H, o-H).
31P{1H}-NMR (C6D6): l 24.1 (s).
2.5. Reaction of 2-propyn-1-ol with 1 producing
[RuCl{C(ꢀCHCH2OH)OC(ꢀO)CH2CH3}(CO)(PPh3)2]
(2c)
2.7. Catalytic addition of propanoic acid to
phenylacetylene forming the enol esters
A mixture of a catalyst precursor (0.10 mmol),
propanoic acid (10 mmol), phenylacetylene (10 mmol),
triphenylphosphine (0.10 mmol, if required) and unde-
cane (an internal standard, 2.8 mmol) in benzene (10
ml) was placed in a reaction vessel equipped with a
reflux condenser and a rubber septum. The mixture was
refluxed under a nitrogen atmosphere. At appropriate
A mixture of 1 (76 mg, 0.10 mmol) and 2-propyn-1-
ol (30 mg, 0.54 mmol) in benzene (30 ml) was refluxed
for 20 h under nitrogen. Addition of hexane to the
concentrated reaction mixture gave a crude 2c as a
yellow powder (yield 50 mg). Unfortunately, any effort
to get the pure 2c out of the crude product resulted in

72
H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
intervals, a small portion of the reaction mixture was
sampled through the septum and applied to GLPC
analysis. Yields of the enol esters were given by means
of the GLPC analysis of the reaction mixture.
In order to follow the ruthenium-containing species
in the catalytic system, the catalyst precursor (0.02
mmol), propanoic acid (0.80 mmol), and pheny-
lacetylene (0.80 mmol) in C6D6 (0.6 ml) were sealed
under vacuum in an NMR tube. The reaction tube was
heated at 80°C. After an appropriate reaction time, the
reaction was stopped by cooling the tube in an ice bath,
then the NMR spectra of the reaction mixture were
measured at 30°C.
3. Results and discussion
3.1. Reaction between
[RuCl(p2-O2CCH2CH3)(CO)(PPh3)2] (1) and terminal
alkynes gi6ing 2-substituted-1-propanoyloxyethenyl-
C1,O-chelate ligands
The reactions of a propanoatoruthenium(II) com-
plex, [RuCl(h2-O2CCH2CH3)(CO)(PPh3)2] (1) with
some terminal alkynes gave a series of ruthenium(II)
complexes bearing enol ester-like 2-substituted-1-
propanoyloxyethenyl-C1,O-chelate ligands. The reac-
tion of 1 with phenylacetylene gave 1-propanoyloxy-
2-phenylethenylruthenium(II)
complex
[RuCl{C-
2.8. X-ray structure study of 2a·2CH2Cl2
(ꢀCHPh)OC(ꢀO)CH2CH3}(CO)(PPh3)2] (2a) as a yel-
1
low powder. The H-NMR spectrum of 2a contains a
Recrystallization of 2a from dichloromethane–hex-
ane solution afforded single crystals suitable for X-ray
diffraction structure analysis. Dichloromethane was in-
corporated as crystallizing solvent in the crystal. Crys-
tal data for 2a·2CH2Cl2: C48H41ClO3P2Ru·2CH2Cl2,
M=1034.19, monoclinic, space group P21/m (no. 11);
singlet at l 4.41 due to the vinylic proton. Additionally,
a characteristic set of 1H-signals of an ethyl group
indicates the presence of the propanoate moiety. The
vinyl carbon atoms appear at l 188.6 (triplet, J(CP)=
13.6 Hz, Ca) and 121.1 (Cb) in its 13C{1H}-NMR
spectrum. Although the latter signal should have been
observed as a triplet because of the PꢂC coupling, the
coupling constant is too small to be observed in com-
parison with the resolution of our spectrometer (3.91
Hz). Actually, J(CP) values for closely related com-
plexes have been reported to be 2–3 Hz by Esteruelas
,
a=9.805(1), b=22.688(2), c=11.406(2) A; i=
3
,
107.754(9)°; V=2416.4 A ; Z=2; yellow prism 0.35×
0.35×0.50 mm; v(Mo–Ka)=7.07 cm−1. The intensity
data were collected at 20°C on a Rigaku AFC5S dif-
fractometer with graphite-monochromated Mo–Ka ra-
,
diation [u(Mo–Ka)=0.71069 A]. The ꢀ–2q scan
1
et al. [11]. The H-13C COSY-NMR spectrum suggests
technique was applied with a maximum 2q value of
60.0°. Of the 7606 reflections that were collected, 7230
were unique (Rint=0.035). The intensities of three stan-
dard reflections, measured every 150 reflections
throughout the data collection, decayed by 2.18%, and
a linear correction factor was applied. The intensities
were corrected for Lorentz and polarization effects. An
empirical absorption correction based on azimuthal
scans of several reflections was applied (transmission
factors in the range 0.95–1.00). The structure was
solved by heavy-atom Patterson methods [16] and ex-
panded using Fourier techniques [17]. All non-hydro-
gen atoms were refined anisotropically. Hydrogen
atoms were included but not refined. Possible disorder
of C29 was neglected and it was positioned on the
crystallographic symmetry plane according to the re-
striction of the space group. The final structure appar-
ently included some errors arising from fixing C29 on
the symmetry plane. Nevertheless, the errors never
affected our discussion of the structure of the chela-
ting ligand so much. The final cycle of full-matrix
least-squares refinement was based on 3073 observed
reflections [I\3|(I)] and 304 variable parameters.
The function minimized was S w(ꢀFoꢀ−ꢀFcꢀ)2. Final R
and Rw values were 0.055 and 0.039, respectively
{R=SꢀꢀFoꢀ−ꢀFcꢀꢀ/SꢀFoꢀ, Rw=[S w(ꢀFoꢀ−ꢀFcꢀ)2/S wFo2]1/
2, where w−1=|2(Fo)+0.006(Fo)2}. Goodness-of-fit
factor was 1.68. All calculations were performed using
the TEXSAN [18] crystallographic software package.
that the vinyl proton (l 4.41) is attached to the Cb (l
121.1). The 31P{1H}-NMR spectrum shows a singlet at
l 33.6, indicating that the two PPh3 ligands are equiva-
lent and positioned mutually trans. The IR spectrum of
2a shows a w(CꢁO) vibration at 1935 cm−1, a w(CꢀO)
at 1630 and a w(CꢀC) at 1597.
Two other terminal alkynes, 1-heptyne and propargyl
alcohol reacted with 1 to give the corresponding com-
plexes, [RuCl{C(ꢀCHC5H11)OC(ꢀO)CH2CH3}(CO)(P-
Ph3)2] (2b) and [RuCl{C(ꢀCHCH2OH)OC(ꢀO)CH2
-CH3}(CO)(PPh3)2] (2c), respectively. These two com-
plexes show common features in their NMR and IR
spectroscopic data that indicate both 2b and 2c are the
same class as 2a. The signals of the ethyl group of the
propanoate moiety, the vinyl proton and the two vinyl
carbons give the formation of the chelating enol ester-
like ligand.
Unfortunately, reaction of 1 with propargylamine
gave a complicated mixture of products that were
difficult to characterize. The elementary analysis and
31P-NMR data of the mixture indicated a significant
loss of the phosphorus content of the products. The
nitrogen content, on the contrary, was larger than
expected. A similar behavior was observed for the
reaction of 1 and methyl propargyl sulfide. These re-
sults suggest that the exchange between the phosphine
ligands and the amine (and also sulfide) is favored more

H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
73
Table 1
,
Selected bond lengths (A) and angles (°) for 2a·2CH2Cl2
Bond lengths
RuꢂCl1
RuꢂP1
RuꢂO3
2.481(2)
2.394(1)
2.127(5)
2.007(8)
1.805(9)
1.156(9)
1.470(8)
1.328(10)
1.223(9)
1.46(1)
RuꢂC27
RuꢂC31
O1ꢂC31
O2ꢂC27
O2ꢂC28
O3ꢂC28
C20ꢂC26
C26ꢂC27
C28ꢂC29
C29ꢂC30
1.331(9)
1.51(1)
1.22(2)
Bond angles
Cl1ꢂRuꢂP1
Cl1ꢂRuꢂO3
Cl1ꢂRuꢂC27
Cl1ꢂRuꢂC31
P1ꢂRuꢂO3
P1ꢂRuꢂC27
P1ꢂRuꢂC31
O3ꢂRuꢂC27
O3ꢂRuꢂC31
C27ꢂRuꢂC31
C27ꢂO2ꢂC28
RuꢂO3ꢂC28
C27ꢂC26ꢂC20
RuꢂC27ꢂO2
RuꢂC27ꢂC26
O2ꢂC27ꢂC26
O2ꢂC28ꢂO3
O3ꢂC28ꢂC29
C28ꢂC29ꢂC30
RuꢂC31ꢂO1
90.11(5)
88.7(2)
167.6(2)
100.2(3)
89.06(5)
89.69(5)
90.90(5)
78.9(3)
Fig. 1. The molecular structure of 2a with the atom-numbering
scheme. Dichloromethane molecules are omitted. The atom numbers
in an asymmetric unit are indicated.
171.2(3)
92.2(3)
115.8(7)
112.9(6)
134.4(8)
110.6(5)
135.6(6)
113.8(7)
121.9(9)
123.9(9)
124(1)
than the formation of the chelating ligand. On the other
hand, treatment of 1 with diphenylacetylene, 1-phenyl-
1-butyne or 4-octyne, under toluene refluxing condi-
tions, resulted in the recovery of the starting complex.
The internal triple bond never reacted with the
propanoate 1.
176.4(8)
3.2. Molecular structure of 2a
,
(1.967(8) A) [11], and is shorter than those of the
The single-crystal X-ray analysis of 2a revealed an
octahedral arrangement of the ligands around the
ruthenium center. A perspective view of the molecular
structure of 2a is illustrated in Fig. 1. Selected bond
lengths and angles are listed in Table 1. There exists a
crystallographic symmetry plane including Ru atom in
the molecule of 2a. The CO, Cl and all non-hydrogen
atoms in the chelating 1-propanoyloxy-2-phenylethenyl-
C1,O ligand except for the terminal CH3 are positioned
on the plane. Two positions of the disordered methyl
group and those of two PPh3 ligands are symmetrical
about the same plane, respectively. The chelate ligand
apparently shows its (Z)-enol ester-like backbone.
Some characteristic bond lengths of the chelating
ligand inform us that the ligand is in resonance between
the structures A and B shown in Scheme 3. Owing to
the contribution of the resonance with the h1-
propanoato-h1-phenylethenylideneruthenium(II) struc-
ordinary RuꢂCsp single bonds found in [Ru(h5-
2
i
,
C5H5){C(ꢀCHPh)O Pr}(CO)(PPh3)] (2.103(6) A) [19],
[Ru{C(ꢀCHCO2CH3)CO2CH3}(CO)(NCCH3)2(PPh3)2]-
ClO4 (2.12(5) A) [20], and [Ru{CHꢀCHtBu}-
,
,
Cl(CO){(CH3)2Hpz}(PPh3)2] (2.063(7) A) [21]. Addi-
tionally, the RuꢂC27 distance is longer than most of the
RuꢀCa distances (1.75–1.90 A) of the reported ethenyli-
deneruthenium(II) complexes [22] according to its par-
tial double bonding character. The contribution of the
structure B also weakens the C27ꢂO2 bond; the bond
,
,
ture B, the RuꢂC27 distance (2.007(8) A) is similar
to those of the related [Ru(h5-C5H5){C(ꢀCHCO2-
,
CH3)OC(ꢀO)CH3}(PPh3)] (2.002(2) A) [10] and
Scheme 3. Two resonance structures of the 1-propanoyloxy-2-
[Ru{C(ꢀCHPh)OC(ꢀO)CH3}(CO)(acetone)(PiPr3)2]BF4
phenylethenyl-C1,O-chelate of 2a.

74
H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
,
distance (1.470(8) A) falls within the range of those
in [Ru(h5-C5H5){C(ꢀCHCO2CH3)OC(ꢀO)CH3}(PPh3)]
,
(1.493(2)
A)
and
[Ru{C(ꢀCHPh)OC(ꢀO)CH3}-
i
,
,
(CO)(acetone)(P Pr3)2]BF4 (1.466(9) A) and is ca. 0.1 A
longer than the CꢂO bond of the ethenyl ligand in
[Ru(h5-C5H5){C(ꢀCHPh)OiPr}(CO)(PPh3)]. The sp-like
nature of Ca is responsible for the unusual bond angles
in RuꢂC27ꢂC26 and O2ꢂC27ꢂC26. The value of the
former (135.6(6)°) is larger and that of the latter
(113.8(7)°) is smaller than 120° for an ideal sp2 carbon.
The phenyl group on the Cb carbon is positioned
trans to the ruthenium atom with respect to the exo-
cyclic CꢀC bond, avoiding the steric hindrance of the
two PPh3 and the CO ligands. No spectroscopic and
crystallographic evidence to support the presence of the
cis-isomer was observed for 2a. This is similar to the
case of [Ru{C(ꢀCHPh)OC(ꢀO)CH3}(CO)(acetone)-
(PiPr3)2]BF4 where the phenyl substituent on the CꢀC
bond is situated trans to the ruthenium, and is quite
contrasting with the fact that the sterically less demand-
ing [Ru(h5-C5H5){C(ꢀCHCO2CH3)OC(ꢀO)CH3}(PPh3)]
is observed as a mixture of the cis- and trans-isomers at
the CꢀC bond. Moreover, the CO2CH3 group of the
1-acetoxy-2-methoxycarbonylethenyl ligand is laid
dominantly cis to the CpRu moiety.
Scheme 4. Two possible routes for the (Z)-enol ester-like chelate
ligand.
3.3. CꢂO bond forming steps between the propanoate
and alkyne moieties
cleavage of the terminal CꢂH bond of the alkyne, being
consistent with the inertness of the internal alkynes to
the formation of the enol ester-like ligands.
As figured in the structure B in Scheme 3, the forma-
tion of the CꢂO bond is interpreted as a result of the
nucleophilic attack of the carboxylate oxygen onto the
Ca carbon of the ethenylidene ligand. Once the
ethenylidene ligand is formed on the ruthenium center,
the nucleophilic attack follows. Electron deficiency and
electrophilicity at the Ca of the ethenylidene ligand are
wel described in the literature [23]. Although the nucleo-
philicity of the carboxylate is somewhat weak, the
chelation of the ligand stabilizes the resulting five-mem-
bered metallacycle. The long C27ꢂO2 bond length may
reflect the weak nucleophilicity of the coordinating
carboxylate.
In the former route, the Ca of the resulting ethenyli-
dene ligand is attacked successively by the coordinating
carboxylate nucleophilically. The carboxylate ligand
need not be released from the ruthenium throughout
the formation of the chelate. In the latter route, car-
boxylic acid, from the coordination sphere, is a real
source of the proton, therefore the proton attached to
the terminal alkyne should be transferred to the car-
boxylate residue to release the free carboxylic acid in
contrast to the former route. Therefore the latter route
is practically impossible because of the acidity of the
terminal alkyne is too weak to liberate the free car-
boxylic acid from the carboxylate complex.
Reaction of 1 with PhCꢁCD (phenylacetylene-d1)
gave only 2-deutero-1-propanoyloxy-2-phenylethenyl-
ruthenium(II) complex (2a-d1). The terminal D com-
pletely moved onto the Cb of the ligand. This result
shows apparently that the source of the b-hydrogen of
the chelating ligand is the terminal hydrogen of the
alkyne. On the basis of this finding, two possible path-
ways are illustrated in Scheme 4, according to the
published mechanisms. One route includes a ruthe-
nium-assisted direct 1,2-hydrogen shift from the Ca to
the Cb of the terminal alkyne. This process is often
found in octahedral d6 complexes [24]. The other route
is b-protonation of an alkynyl intermediate derived
from the terminal alkyne. Both mechanisms require a
3.4. Acidolysis of 2a and ruthenium complex-catalyzed
(Z)-enol ester-selecti6e synthesis
The (Z)-enol ester-like backbone of the 1-propanoyl-
oxy-2-phenylethenyl-C1,O ligand suggests directly that
complex 2a is a possible catalyst for the (Z)-enol ester-
selective addition of propanoic acid to phenylacetylene.
As stated above, the propanoate complex 1 readily
reacts with phenylacetylene to give 2a. If 2a reacts
successively with propanoic acid to liberate (Z)-2-
phenylethenyl propanoate and to reproduce 1, a cata-
lytic cycle is completed to produce the (Z)-enol ester
selectively (see (Z)-enol ester manifold in Scheme 5).

H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
75
should be a result of disproportionation of the chloro-
propanoato complex 5. The acidolysis of 2a giving the
(Z)-enol ester and the dicarbonylchloro(h1-propano-
ato) complex 5 is very similar to those of the neutral
[RuCl{C(ꢀCHPh)OC(ꢀO)CH3}(CO)(PiPr3)2] and the
cationic
[Ru{C(ꢀCHPh)OC(ꢀO)CH3}(CO)(acetone)-
(PiPr3)2]+ reported by Esteruelas et al. [11]. The source
of the second carbonyl of these complexes seems to be
the carboxylic acid. There are some examples of the
ruthenium complexes-promoted decarbonylation of or-
ganic acids and esters although detail of its process is
unknown [25].
The liberation of the (Z)-enol ester in the acidolysis
of the complex 2a prompted us to try the addition of
propanoic acid to phenylacetylene catalyzed by 2a. In
order to compare the catalytic activity and selectivity,
the propanoato complex 1 and a hydridoruthenium(II)
complex [RuClH(CO)(PPh3)3] (3) were also used as a
catalyst or a catalyst precursor. Table 2 summarizes the
results of the catalytic addition. When 1 or 2a was used
as a catalyst precursor, the total yield of the enol esters
reached up to 99% within 24 h in refluxing benzene and
no dimer of phenylacetylene (diphenylbut-1-en-3-ynes)
that was often formed in the presence of a transition-
metal complex-catalyst [26] was recognized. As for the
selectivity, 1-phenylethenyl propanoate was formed to-
gether with (Z)-2-phenylethenyl propanoate preferen-
tially. On the other hand, catalytic addition using 3 as
a catalyst precursor was much slower than those using
1 and 2a; it afforded the enol esters in only 35% yield
after 72 h accompanied by a small amount of the
dimers. The rest of the acid and alkyne remained
non-reacted. Among the three isomers of the enol ester,
the (Z)-type was produced in 86% relative selectivity
using 3. Very interestingly, the presence of the addi-
tional PPh3 (run 2) suppressed the total yield of the
enol esters but increased the relative selectivity for the
(Z)-type enol ester up to 86% compared to those in the
run 1. These results are the third example of selective
(Z)-enol ester formation catalyzed by a ruthenium com-
plex following Dixneuf’s catalytic system [8], but are in
contrast to the case of the ferrocenecarboxylato com-
plex [Ru{h2-O2C(C5H4)Fe(C5H5)}Cl(CO)(PPh3)2] that
has been shown to be a very effective catalyst precursor
Scheme 5. The catalytic cycles for the selective addition of propanoic
acid to phenylacetylene.
Therefore the following detailed analysis of the acidoly-
sis of 2a is examined.
Treatment of 2a with propanoic acid in refluxing
benzene gave (Z)-2-phenylethenyl propanoate; the cis-
geometry between the phenyl group and the propanoy-
loxy group was retained in the acidolysis. The
production of the (Z)-enol ester through the acidolysis
of 2a implies that the ruthenium complex-catalyzed
synthesis of the enol ester can proceed via 2a. Neverthe-
less, it should be noted that the acidolysis of 2a does
not completely reproduce the propanoate 1. The 31P-
NMR spectrum of the reaction mixture showed not
only the signal of 1 (l 34.4) but additional three singlets
at l 31.8, 24.1 and 15.4. These three singlets were
assigned to a dicarbonylbis(h1-propanoato) complex
[Ru{h1-OC(ꢀO)C2H5}2(CO)2(PPh3)2] (4), and to a di-
carbonylchloro(h1-propanoato) complex [RuCl{h1-
OC(ꢀO)C2H5}(CO)2(PPh3)2] (5), and to
a
dicar-
bonyldichloro complex cct-[RuCl2(CO)2(PPh3)2] (6)
[14], respectively. The relative ratio of 1 based on the
integrated 31P peak area fell only 20% when the acidol-
ysis was completed. The relative ratios of three dicar-
bonyl species were 14% for 4, 50% for 5, and 16% for
6. The amounts of 4 and 6 were almost equal to each
other during the acidolysis. Apparently, formation of
the bis(propanoato) and the dichloro complexes 4 and 6
Table 2
Addition of propanoic acid to phenylacetylene catalyzed by ruthenium complexes
Reaction
Run
Relative ratio (%)
Complex
Time (h)
Yield (%)
(Z)-type
(E)-type
Markovnikov-type
1
2
3
4
1
24
72
24
72
\99
29
\99
35
35
86
42
86
7
4
7
9
58
10
51
5
1 a
2a
3
a Triphenylphosphine was added: [PPh3]/[Ru]=1.0.

76
H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
3.5. Conclusions: the intermediates in the catalysis and
the origin of the (Z)-enol ester-selecti6ity
The results of the acidolysis and the catalytic reac-
tions indicate that the (Z)-enol ester is produced pre-
dominantly by the acidolysis of 2a in the catalytic cycle.
The presence of free PPh3 in the catalytic system leads
the predominant formation of 2a from 1 and pheny-
lacetylene. After all, the presence of the free PPh3
causes the (Z)-selectivity in the catalysis. The most
plausible (Z)-selecting mechanism is as the followings:
the slow acidolysis of 2a by propanoic acid gives the
(Z)-enol ester and the propanoato complex 1. Once the
complex 1 is produced, it is converted competitively
into 2a or the three dicarbonyl species 4, 5, and 6 under
the reaction conditions. When the free PPh3 is present
in the reaction mixture, PPh3 suppresses the competitive
formation of the dicarbonyl species. Therefore, the two
intermediary species, 1 and 2a complete the catalytic
cycle producing the (Z)-enol ester. In the absence of the
free PPh3, the dicarbonyl species evolve after the acidol-
ysis, bring about another catalytic route including
them, and lead the production of the (E)- and
Markovnikov-type enol esters. As stated here, the
structure of the intermediate 2a and the origin of the
(Z)-selectivity of the catalytic system are clearly ratio-
nalized in this study.
Fig. 2. The 31P{1H}-NMR spectra of the ruthenium species in the
catalytic addition of propanoic acid to phenylacetylene: (a) complex,
3; reaction time, 0.5 h; (b) complex, 1; reaction time, 0.5 h; (c)
complex, 1; reaction time, 15 h. These spectra were observed in C6D6;
the 31P-chemical shift value of 2a (l 34.3) is slightly different from
that in CDCl3 (l 33.6).
in the (Z)-selective addition of ferrocenecarboxylic acid
to phenylacetylene [9].
The 31P-NMR spectra of the catalytic reaction mix-
tures are shown in Fig. 2. When 3 was employed for the
catalyst precursor, the 31P-signals of 3 completely
turned into those of 2a and free PPh3 within 30 min
despite the slow formation of the enol esters. Once 2a
appeared, any other ruthenium-containing species (the
propanoate 1, the starting 3 complex, and the dicar-
bonyl species 4, 5, and 6) was not observed throughout
the catalytic reaction. The presence of 2a as the only
observable complex apparently means the acidolysis of
2a is the rate-determining step of the catalytic cycle.
When 1 was used for the catalyst, in contrast, the
starting complex 1 disappeared within 30 min. The
reaction mixture contained 2a and three dicarbonyl-
ruthenium species 4, 5, and 6. The complex 2a also
disappeared when the catalytic reaction was completed,
and there remained only 4, 5, and 6. When 2a was used
for the catalyst precursor, the results were similar to
those for 1. The use of 3 means the presence of the
excess free PPh3 in the catalytic mixture. In the pres-
ence of the excess free PPh3, even 1 afforded the
complex 2a and the (Z)-enol ester; the formation of 4,
5, and 6 was suppressed. In other words, the presence
of the free PPh3 effectively suppresses the competitive
side-reaction leading to the dicarbonylruthenium spe-
cies 4, 5, and 6.
4. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC no. 113479 for 2a·2CH2Cl2.
Copies of this information may be obtained free of
charge from: the Director, CCDC, 12 Union Road,
Cambridge, CB2 1EZ, UK (Fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or WWW: http://
www.ccdc.cam.ac.uk).
Acknowledgements
A part of this work was financially supported by a
grant for one of the authors (H.K.) from the Founda-
tion ‘Hattori-Hokokai.’
References
[1] (a) T. Mitsudo, Y. Hori, Y. Watanabe, J. Organomet. Chem.
334 (1987) 157. (b) C. Bruneau, P.H. Dixneuf, Chem. Commun.
(Cambridge) (1997) 507 and references cited therein.
[2] (a) T. Mitsudo, Y. Hori, Y. Watanabe, J. Org. Chem. 50 (1985)
1566. (b) T. Mitsudo, Y. Hori, Y. Yamanaka, Y. Watanabe,
Tetrahedron Lett. 27 (1986) 2125. (c) T. Mitsudo, Y. Hori, Y.
Watanabe, J. Org. Chem. 52 (1987) 2230.

H. Kawano et al. / Journal of Organometallic Chemistry 601 (2000) 69–77
77
[16] F. Hai-Fu, SAPI-91, Structure Analysis Programs with Intelligent
Control, Rigaku Corporation, Tokyo, Japan, 1991.
[3] (a) C. Ruppin, P.H. Dixneuf, Tetrahedron Lett. 27 (1986) 6323.
(b) C. Ruppin, P.H. Dixneuf, S. Le´colier, Tetrahedron Lett. 29
(1988) 5365.
[4] K. Philippot, D. Devanne, P.H. Dixneuf, J. Chem. Soc. Chem.
Commun. (1990) 1199.
[5] M. Neveux, C. Bruneau, P.H. Dixneuf, J. Chem. Soc. Perkin
Trans. 1 (1991) 1197.
[6] M. Neveux, B. Seiller, F. Hagedorn, C. Bruneau, P.H. Dixneuf,
J. Organomet. Chem. 453 (1991) 133.
[17] P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S.
Garcia-Granda, R.O. Gould, J.M.M. Smits, C. Smykalla,
DIRDIF-92, The DIRDIF Program System, Technical Report of
the Crystallography Laboratory, University of Nijmegen, The
Netherlands, 1992.
[18] TEXSAN, Crystal Structure Analysis Package, Molecular Struc-
ture Corporation, 1985, 1992.
[19] M.I. Bruce, D.N. Duffy, M.G. Humphrey, A.G. Swincer, J.
Organomet. Chem. 282 (1985) 383.
[7] M. Rotem, Y. Shvo, J. Organomet. Chem. 448 (1993) 189.
[8] (a) H. Doucet, J. Ho¨fer, C. Bruneau, P.H. Dixneuf, J. Chem.
Soc. Chem. Commun. (1993) 850. (b) H. Doucet, B. Martin-
Vaca, C. Bruneau, P.H. Dixneuf, J. Org. Chem. 60 (1995) 7247.
[20] J. Lo´pez, A. Romero, A. Santos, A. Vegas, A.M. Echavarren, P.
Noheda, J. Organomet. Chem. 373 (1989) 249.
[21] A. Romero, A. Santos, A. Vegas, Organometallics 7 (1988) 1988.
[22] R. Le Lagadec, E. Roman, L. Toupet, U. Mu¨ller, P.H. Dixneuf,
Organometallics 13 (1994) 5030 and references cited therein.
[23] (a) N.M. Kostic, R.F. Fenske, Organometallics 1 (1982) 974. (b)
H. Werner, J. Wolf, R. Zolk, U. Schubert, Angew. Chem. Int.
Ed. Engl. 22 (1983) 981. (c) H. Werner, J. Wolf, G. Mu¨ller, C.
Kruger, Angew. Chem. Int. Ed. Engl. 23 (1984) 431.
[24] (a) M.I. Bruce, Chem. Rev. 91 (1991) 197. (b) Y. Wakatsuki, N.
Koga, H. Yamazaki, K. Morokuma, J. Am. Chem. Soc. 116
(1994) 8105.
´
[9] L. Matas, I. Moldes, J. Soler, J. Ros, A. Alvarez-Larena, J.F.
Piniella, Organometallics 17 (1998) 4551.
[10] T. Daniel, N. Mahr, T. Braun, H. Werner, Organometallics 12
(1993) 1475.
[11] M.A. Esteruelas, F.J. Lahoz, A.M. Lo´pez, E. On˜ate, L.A. Oro,
Organometallics 13 (1994) 1669.
[12] N. Ahmad, J.J. Levison, S.D. Robinson, M.F. Uttley, Inorg.
Synth. 15 (1974) 45.
[13] (a) S.D. Robinson, M.F. Uttley, J. Chem. Soc. Dalton Trans.
(1973) 1912. (b) K. Hiraki, A. Nonaka, T. Matsunaga, H.
Kawano, J. Organomet. Chem. 574 (1999) 121.
[14] D.W. Krassowski, J.H. Nelson, K.R. Brower, D. Hauenstein,
R.A. Jacobson, Inorg. Chem. 27 (1988) 4294.
[15] A. Dobson, S.D. Robinson, M.F. Uttley, Inorg. Synth. 17 (1977)
124.
[25] (a) J. Ropp, J. Am. Chem. Soc. 82 (1960) 842. (b) T. Yamamoto,
S. Miyashita, Y. Naito, S. Komiya, T. Ito, A. Yamamoto,
Organometallics
1 (1982) 808. (c) K. Hiraki, S. Kira, H.
Kawano, Bull. Chem. Soc. Jpn. 70 (1997) 1583.
[26] R. Mahe´, P.H. Dixneuf, S. Le´colier, Tetrahedron Lett. 27 (1986)
6333.
.