Reactivity of [Pd3(μ-OAc)3(μ,η2-MeSCHCO 2Et-C,S)3] in the presence of triphenylphosphine: A model of the early steps of the Pd/PR3-catalysed Heck reaction

Article abstract of DOI:10.1016/S0022-328X(00)00063-2

The stable complex [Pd(η¹-OAc)(η²-MeSCHCO₂Et-C,S)(PPh ₃)] (2) is readily formed by addition of triphenylphosphine to [Pd₃(μ-OAc)₃(μ,η²-MeSCHCO ₂Et-C,S)₃] (1:1 P:Pd) in acetone. It crystallizes in the monoclinic space group C2/c, with Z=8, a=145.08(2), b=109.82(1), c=306.33(3) pm and β=96.3(1)°, R(F)=0.026, R_w (F²)=0.0765. The coordination of a second triphenylphosphine to the palladium atom leads to [Pd(η¹-OAc){CH(SMe)CO₂Et}(PPh₃) ₂] (3). Whereas 2 appears to be remarkably stable, 3, after a slow reductive elimination, evolves mainly to [Pd(η²-MeSCHCO₂Et-C,S)(PPh₃) ₂]^-. Zero-valent palladium species can also be generated by electrochemical reduction of 3. The rate constant for the oxidative addition of iodobenzene to the electrogenerated palladium(0) species has been estimated.

Full text of DOI:10.1016/S0022-328X(00)00063-2

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 601 (2000) 201–210
Reactivity of [Pd3(m-OAc)3(m,h2-MeSCHCO2Et-C,S)3] in the
presence of triphenylphosphine: a model of the early steps of the
Pd/PR3-catalysed Heck reaction
Marino Basato* a,1, Barbara Sesto a, Marco Zecca* a,2, Giovanni Valle b,
Sabrina Antonello c, Flavio Maran* c,3
a Dipartimento di Chimica Inorganica, Metallorganica e Analitica,
and Centro di Studio sulla Stabilita` e Reatti6ita` dei Composti di Coordinazione, Uni6ersita` di Pado6a, 6ia Marzolo 1, I-35131 Padua, Italy
b Centro di Studio sui Biopolimeri, C.N.R., 6ia Marzolo 1, I-35131 Padua, Italy
c Dipartimento di Chimica Fisica, Uni6ersita` di Pado6a, 6ia Loredan 2, I-35131 Padua, Italy
Received 23 July 1999; accepted 17 January 2000
Abstract
The stable complex [Pd(h1-OAc)(h2-MeSCHCO2Et-C,S)(PPh3)] (2) is readily formed by addition of triphenylphosphine to
[Pd3(m-OAc)3(m,h2-MeSCHCO2Et-C,S)3] (1:1 P:Pd) in acetone. It crystallizes in the monoclinic space group C2/c, with Z=8,
a=145.08(2), b=109.82(1), c=306.33(3) pm and i=96.3(1)°, R(F)=0.026, Rw (F2)=0.0765. The coordination of a second
triphenylphosphine to the palladium atom leads to [Pd(h1-OAc){CH(SMe)CO2Et}(PPh3)2] (3). Whereas 2 appears to be
remarkably stable, 3, after a slow reductive elimination, evolves mainly to [Pd(h2-MeSCHCO2Et-C,S)(PPh3)2]−. Zero-valent
palladium species can also be generated by electrochemical reduction of 3. The rate constant for the oxidative addition of
iodobenzene to the electrogenerated palladium(0) species has been estimated. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Palladium complex; b-Oxothioether ligands; Phosphines; Oxidative addition; Reductive elimination; Aryl iodides
the reaction with bipyridine the sulfur ligand is dis-
placed from the metal coordination sphere and
1. Introduction
[Pd(OAc)2(bipy)] is the main product. By contrast, the
We have recently published the synthesis of the
reaction with b-diketones affords dimers or trimers with
mixed-sphere trimeric complex of palladium(II), [Pd3(m-
OAc)3(m,h2-MeSCHCO2Et-C,S)3] (1) [1]. This complex
is isostructural to palladium acetate and the three
bridging sulfur ligands lie on the same side of the Pd3
plane. The coordinated sulfur atoms can bind further to
electrophilic metal centres. For example, we have found
bridging sulfur ligands and chelate b-diketonates are
formed.
We report herein data on the reactivity of 1 with
triphenylphosphine. The presence of acetato ligands in
the coordination sphere of palladium(II) centres and
the close structural relationship to palladium acetate
make the reactivity of complex 1 towards PPh3 very
similar to that of [Pd(OAc)2]. The system [Pd(OAc)2]/
PR3 has been the subject of extensive investigation, as it
is commonly used to catalyse the Heck reaction [4].
Phosphines reduce [Pd(OAc)2] and the reaction affords
zero-valent palladium species along with the pertinent
phosphine oxide [5–7]. Electrochemical and 31P-NMR
studies have shown that, in the presence of an excess of
PPh3, the reactive anionic complex [Pd(h1-
OAc)(PPh3)2]− is in equilibrium with the coordinatively
that
1 reacts with either [Pd(NCMe)4](BF4)2 or
[Cu(NCMe)4]PF6 and behaves as an organometallic
ligand towards Pd(II) and Cu(I) in the relevant adducts
[2]. We have also investigated the reactivity of 1 with
bidentate nucleophiles, both neutral (bipyridine) and
monoanionic (acetylacetonate, benzoylacetonate) [3]. In
1 *Corresponding author. Fax: +39-049-8275223; e-mail:
basato@chin.unipd.it
2 *Corresponding author. E-mail: mzecca@chin.unipd.it
3 *Corresponding author. E-mail: f.maran@chfi.unipd.it
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00063-2

202
M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
saturated species [Pd(h1-OAc)(PPh3)3]− and readily un-
dergoes oxidative addition of iodobenzene [8]. These
reactions account for the activation of aryl halides in
the Heck reaction, according to a widely accepted
mechanism [2]. We have also found that 1 is reduced by
an excess amount of PPh3 and that the ensuing zero-va-
lent palladium species undergoes oxidative addition of
iodobenzene. The reduction is a stepwise process, in
which the C,S-chelate [2,9] monophosphino complex
(d, 3H, CH3S, 4JHP=4.6 Hz), ca. 3.5 (m, 2H,
CH3CH2), 3.70 (s, 1H, SCH), 7.43–7.59 (m, 15H,
C6H5). 13C-NMR (27°C; CDCl3): l=13.7 (CH3CH2),
22.1, 23.3 (CH3S and CH3CO2, 52.0 (CH), 60.7
(CH3CH2), 128.1–134.1 (C6H5), 169.5 (C(O)OEt),
177.8 (CH3CO2). 31P-NMR (27°C; CDCl3): l=31.77
1
(s, PPh3). H-NMR (27°C; [D7] DMF): l=0.82 (t, 3H,
3
CH3CH2, JHH=7.1 Hz), 1.68 (s, 3H, CH3CO2), 2.58
4
(d, 3H, CH3S, JHP=4.9 Hz), 3.56 (AB system m, 2H,
2
[Pd(h1-OAc)(h2-MeSCHCO2Et-C,S)(PPh3)]
(2)
is
CH3CH2, JAB=11.0 Hz), 4.15 (s, 1H, SCH), 7.50–
formed first. The intermediacy of palladium(0) species
was also supported by analysis of the voltammetric
behaviour of the 1/PPh3 system. In addition, electro-
chemistry provided relevant information on the effect
of added PPh3 on the reduction of 1, the oxidation of
the zero-valent palladium species and the reactivity of
the latter towards iodobenzene.
7.56 (m, 15H, C6H5). 31P-NMR (27°C; [D7] DMF):
l=32.99 (s, PPh3). FTIR (KBr disk): w(CO)=1705 (s,
ester); 1601, 1373 (s, acetate).
2.3. Reaction of [Pd3(v-OAc)3(v,p2-MeSCHCO2Et)3]
(1) with PPh3
The progress of the reaction in THF and DMF
1
solution was followed by means of FTIR and H- and
2. Experimental
31P-NMR spectroscopy, respectively. In the first case, 1
(0.25 g, 0.28 mmol) was dissolved in THF (20 cm3) and
PPh3 (0.44 g, 1.78 mmol, 2:1 P:Pd) was added to the
solution. Aliquot portions of the reaction mixture were
periodically analysed by FTIR in the 4000–1000 cm−1
range. As we first found (see Section 3) that under these
conditions half of the phosphine is consumed to form
complex 2, in the second case we used it directly as the
starting material. Thus, 2 (8.4 mg, 15 mmol) was trans-
ferred into a NMR tube and dissolved with 300 ml of
[D7] DMF, and PPh3 (3.9 mg, 15 mmol) was added to
the solution. The reaction was allowed to proceed in
the NMR tube at 27°C without stirring and in contact
with air. 1H- and 31P-NMR spectra of the reaction
mixture were periodically recorded. Apparently, no
solid precipitated from the solution, which remained
clear for at least 6 days: in particular, no metallic
palladium formed.
2.1. Reagents and apparatus
Palladium acetate (98%), ethyl (methylthio)acetate
(98%) and other products were supplied by Aldrich. All
of them except for triphenylphosphine, which was re-
crystallized from hot ethanol prior to use, were used as
received. Anhydrous solvents were purchased from
BDH and stored under inert atmosphere over molecu-
lar sieves. All the syntheses were carried out under inert
atmosphere with standard vacuum/Schlenck line tech-
niques. [Pd3(m-OAc)3(m,h2-MeSCHCO2Et)3] (1) was
synthesized as described in Ref. [3]. FTIR spectra were
recorded on either a Bruker IFS 66 Fourier Trans-
former or a Biorad FT S7 PC spectrophotometer.
NMR measurements were carried out with a Jeol FX
90Q Fourier transform spectrometer. Chemical shifts
were referenced to either TMS (1H-NMR and 13C-
NMR) or 75% aqueous H3PO4 (31P-NMR).
2.4. Reaction of [Pd3(v-OAc)3(v,p2-MeSCHCO2Et)3]
(1) with PhI in the presence of PPh3
2.2. Synthesis of
[Pd(p1-OAc)(p2-MeSCHCO2Et)(PPh3)] (2)
The reaction of 1 with iodobenzene (equimolar
amount vs. palladium) was studied in anhydrous ben-
zene in the presence of PPh3 and several bases. Typical
reaction conditions were as follows: 1 (0.16–0.20 g,
0.18–0.22 mmol) was dissolved in anhydrous benzene
and PPh3 (0.29–0.35 g, 1.08–1.34 mmol, P:Pd=2) was
added to the solution. Iodobenzene (62.2–75.9 cm3,
0.54–0.66 mmol, PhI:Pd=1) and the base were added
immediately thereafter to the solution. The base:Pd
PPh3 (0.26 g, 1 mmol) was added to a suspension of
1 (0.30 g, 0.33 mmol) in anhydrous acetone (30 cm3).
The suspension immediately cleared. In a few minutes
the colour of the solution turned to green and a solid
started to precipitate. The suspension was stirred for a
further 30 min and then most of the solvent was
removed under reduced pressure. A greenish solid pre-
cipitated and was recovered by filtration (0.40 g, 73%).
Recrystallization from dichloromethane–acetone af-
forded the light yellow complex 2 (60%). C25H27O4PPdS
(FW 560.93): Calc. C, 53.53; H, 4.85; S, 5.71. Found:
n
molar ratio was equal to 1 for Bu4NOAc and NPh3,
whereas it was equal to 2 for 1,8-dimethyl-aminonaph-
thalene (‘proton sponge’, PS). After stirring for several
days (3–7), the solvent was removed, mostly or com-
pletely, from the reaction mixtures under reduced pres-
sure. Upon treatment with anhydrous diethyl ether, the
1
C, 53.48; H, 4.78; S, 5.71%. H-NMR (27°C; CDCl3):
l=0.92 (t, 3H, CH3CH2), 1.67 (s, 3H, CH3CO2), 2.49

M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
203
obtained residues (oils or concentrated solutions)
yielded solid products, whose colours ranged from
brick red to brown. Typical analytical and spectro-
scopic data were as follows (reaction with NPh3). The
elemental analysis is not coherent with the formation of
pure [Pd(h1-OAc)(Ph)PPh3], which requires C, 61.86
and H, 4.59. The values found (C, 58.12; H, 4.26%) are
consistent with a mixture of [Pd(h1-OAc)(Ph)PPh3]2,
[PdI(Ph)PPh3]2 and [PdI(Ph)(PPh3)2] (7:4:4, mol; C,
58.01; H, 4.18). The expected ratio between aromatic
protons and acetato protons in this mixture is 14.
1H-NMR (27°C; CDCl3): l=1.67 (s, 3H, CH3CO2),
7.5–7.7 (m, 45H, PC18H15+C6H5), weak singlets also
at 0.58, 0.95, 1.20, 1.28 ppm. 31P-NMR (27°C; CDCl3):
l=22.94, 21.33, 20.46, 12.85.
When the reaction was performed in the presence of
nBu4NOAc, a second solid product spontaneously sepa-
rated within a few days from the solution recovered
after the separation of the first precipitate. After wash-
ing with ethyl acetate to remove triphenylphosphine
oxide, the solid was dried and characterized.
C26H23O2PPd ([Pd(h1-OAc)(Ph)PPh3], FW=504.8):
Calc.: C, 61.86; H, 4.59. Found: C, 59.26; H, 4.37%.
1H-NMR (27°C; CDCl3): l=1.67 (s, 3H, CH3CO2),
7.3–7.8 (m, 20H, PC18H15+C6H5). 31P-NMR (27°C;
CDCl3): l=28.8.
fore each measurement [10]. The gold (Goodfellow,
99.99%) disk electrode was prepared and activated in a
similar way. The reference electrode was a homemade
Ag/AgCl, calibrated after each experiment against the
ferrocene/ferricinium couple and then against the KCl
saturated calomel electrode, SCE (in DMF–0.1 M
TBAP, E°Fc/Fc+ =0.464 V vs. SCE). In the following all
of the potential values will be reported against SCE.
The counterelectrode was a 1 cm2 Pt plate. The voltam-
metric curves were obtained by using an EG&G-Parc
173 potentiostat, an EG&G-Parc 175 universal pro-
grammer, and a Nicolet 3091 12-bit resolution digital
oscilloscope. The feedback correction was applied in
order to minimize the ohmic drop between the working
and reference electrodes. In most of the experiments,
the cyclic voltammograms were recorded in a selected
potential range and for scan rates ranging from 0.1 to
200 V s−1 by the digital oscilloscope (digitalized 1
point/mV) and then transferred to a PC. The curves
were then analysed by the conventional voltammetric
criteria [11], using our own laboratory software. Digital
simulations of the cyclic voltammetry curves were per-
formed by using the DigiSim 2.1 software by Bioanalyt-
ical Systems Inc.
2.7. Crystal-structure determination of
[Pd(p1-OAc)(p2-MeSCHCO2Et)(PPh3)] (2)
2.5. Reaction of [Pd3(v-OAc)3(v,p2-MeSCHCO2Et)3]
(1) with MeI in the presence of PPh3
Suitable crystals for X-ray diffraction were obtained
upon slow evaporation of acetone solutions of 2. 4964
independent reflections up to q=28° were collected by
means of a Philips PW 1100 four-circle diffractometer
using the q–2q technique and the monochromatic Mo–
Ka radiation with u=71.07 pm. The positional
parameters for C, S, O and Pd atoms were determined
directly with SHELXS-86 and were refined on F2 (SHELX-
93) through least-squares cycles using anisotropic ther-
mal parameters. All of the hydrogen atoms were
identified in the difference Fourier map, constrained to
the parent site and refined with isotropic thermal
parameters. An idealized geometry was imposed on the
phenyl rings. The conventional final R(F) value is 0.026
(for the observed 4338 reflections with F\4|(F)),
whereas the Rw(F2) value is 0.0765 for all data (w=1/
[|2(F2)+(0.0499P)2], P=(Fo2 +2F2c)/3). Molecular for-
mula: C25H27O4PPdS (FW=560.93); crystal system:
Complex 1 (0.25 g, 0.28 mmol) was dissolved in
anhydrous toluene (10 cm3). PPh3 (0.65 g, 2.5 mmol,
P:Pd=3) was added to the solution and iodomethane
(55 ml, 0.84 mmol, MeI:Pd=1) immediately thereafter.
The colour of the solution immediately turned from red
to dark red. After 5 h under magnetic stirring, traces of
a solid were filtered off and an orange solid precipitated
upon addition of ethyl ether to the filtrate.
C23H24IO2PPdS ([PdI(MeSCHCO2Et)PPh3], FW=
628.81): Calc. C, 43.93; H, 3.85; S, 5.10.
C59H54IO2P3PdS
FW=1153.39): Calc.: C, 61.44; H, 4.72; S, 2.78.
Found: C, 51.80; H, 4.25; S, 3.75%. H-NMR (27°C;
([Pd{h1-CH(SMe)CO2Et}(PPh3)3]I,
1
CDCl3): l=0.91 (2t, 3H CH3CH2,), 1.96 (s, br, 1.2H,
4
CH3S), 2.35 (d, JHP=4.9 Hz, 1.8H, CH3S), ca. 3.5 (m,
2H, CH3CH2), 3.77 (s, br, 1H, CHS), 7.3–7.8 (ca. 30H,
Ph).
monoclinic; space group: C2/c; a, b, c (pm): 145.08(2),
3
,
109.82(1), 306.63(3), i (°)=96.3(1); V (A )=4851(3);
2.6. Electrochemistry
Z=8. Dcalc. (g cm−3)=1.536.
N,N-Dimethylformamide (Janssen, 99%) and tetra-
butylammonium perchlorate (TBAP, 99%, Fluka) were
purified as previously described [10]. Electrochemical
measurements were conducted in an all-glass cell and
under an argon atmosphere. The glassy carbon (Tokai
GC-20) disk electrode was prepared and activated be-
3. Results and discussion
Palladium acetate is the precursor of many catalytic
systems employed in organic synthesis [12]. The olefina-
tion of aryl halides (Heck reaction, Eq. (1)), in particu-

204
M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
lar, has been the subject of several recent studies
ands. In addition, the three acetato ligands and the
three sulfur ligands lie on opposite sides with respect to
the plane of the palladium atoms. The sulfur ligands
displayed a greater affinity to palladium(II) than ac-
etate. In fact, despite the much lower acidity of ethyl
(methylthio)acetate than acetic acid (the pKDa MF values
are 24.9 and 13.5, respectively) [15,16], half of the
acetates were displaced as acetic acid and substituted by
an equal number of MeSCHCO2Et− ligands.
[13,14].
(1)
One of the most interesting features is that this
reaction is rather insensitive to the nature of the reac-
tants, yielding a wide range of products. It is promoted
by palladium(II) acetate in the presence of an excess
amount of alkyl or aryl phosphines (P:Pd=2). In this
context, both the behaviour of palladium(II) acetate
and its reactivity with iodobenzene in the presence of
tertiary phosphines have been investigated by means of
electrochemical and spectroscopic (31P-NMR) tech-
niques [6–8]. These studies demonstrated the intermedi-
acy of zero-valent palladium species, such as
[Pd(h1-OAc)(PPh3)x]−, undergoing oxidative addition
of iodobenzene. The rate-determining step of this pro-
cess is the slow reductive elimination of AcOPPh+3 from
[Pd(h1-OAc)2(PPh3)2], which forms quickly upon coor-
dination of two phosphine molecules (Eqs. (2)–(4)):
In agreement with its structural analogy with palladi-
um(II) acetate, complex 1 readily reacts with phosphi-
nes. The presence of the sulfur ligand, however,
stabilizes the adducts with triphenylphosphine. In fact,
we obtained evidence for the stepwise reaction:
PPh
3
1/3[Pd(m-OAc)(m,h2-CS)]3 “
1
[Pd(h1-OAc)(h2-CS)(PPh3)]
(5)
2
CS=MeSCHCOOEt−
[Pd(h1-OAc)(h2-CS)(PPh3)]
+PPh
3
ꢀꢀꢀꢀꢁ[Pd(h1-OAc)(h1-CS)(PPh3)2]
(6)
−PPh
3
3
[Pd(OAc)2]+PPh3“[Pd(h1-OAc)2(PPh3)2]
(2)
The stabilization brought about by the sulfur ligand is
such that we were able to isolate complex 2 in a pure
form and obtain crystals suitable for XRD analysis
(Fig. 1).
PPh , H O
[Pd(h1-OAc)2(PPh3)2] ꢀꢀꢀ3ꢀꢀꢀꢀꢁ [Pd(h1-OAc)(PPh3)2]−
2
+
−HOAc, −H
+OꢂPPh3
(3)
[Pd(h1-OAc)(PPh3)2]−
The phosphine attacks the trimeric complex 1 and
disrupts its polynuclear structure. Although both the
acetato and the sulfur ligands are bidentate (O,O and
C,S, respectively) in the starting complex, only the
sulfur ligand is still bidentate in complex 2. This is a
consequence of the higher affinity of the soft sulfur
donor atoms towards the soft palladium(II) centres, in
comparison with the hard oxygen donor atoms. It is
remarkable that in this chelate mode the coordinated C
and S atoms also adopt the same absolute configuration
as observed in the starting bridged trimer. The main
structural feature of complex 2 is the presence of a
strained three-membered metallacycle. The oxygen and
phosphorus donors also lie in the plane of the metalla-
cycle. The structure is quite similar to that of [PdCl(h2-
CH2SMe)(PPh3)] (5) [9]. The bond distances (pm, Table
1) of the four donor atoms from the palladium atom in
2 are as follows: 203.5(3) (PdꢀC), 234.4(1) (PdꢀS),
226.9(1) (PdꢀP), 209.2(2) (PdꢀO). They are very close to
the values observed for 5: 204.2(6) (PdꢀC), 237.1(1)
(PdꢀS), 226.7(1) (PdꢀP) pm. The difference between the
PdꢀO distance and the PdꢀCl distance, 240.2(1) pm, is
the same as the difference in the covalent radii of
oxygen and chlorine, respectively, and reflects only the
different sizes of the two donor atoms. Also, the bond
angle values compare well with those observed in the
dimethylsulfide derivative. The SꢀPdꢀC angle, for in-
stance, is 46.72(6)° wide with respect to 45.4(3)° in 5.
The opposite angles, OꢀPdꢀP in 2 and ClꢀPdꢀP in 5,
−
+PhI−ꢀꢀIꢁ [Pd(h1-OAc)Ph(PPh3)2]
(4)
These steps are believed to be involved in the catalyst
formation and the activation of the aryl halides in the
Heck reaction. The complex [Pd3(m-OAc)3(m,h2-
MeSCHCO2Et)3] (1) bears a close structural relation-
ship with palladium(II) acetate. Whereas the latter has
trimeric molecules with six bridging acetato ligands,
complex 1 has three bridging acetato ligands and three
bridging C,S (methylthio)(ethoxycarbonyl)methyl lig-
Fig. 1. ORTEP view of [Pd(h1-OAc)(h2-MeSCHCO2Et)(PPh3)] (2).

M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
205
Table 1
ences its geometry around the palladium atom: in fact,
although the CꢀPdꢀS angle in the ring is equal to
46.4(3)°, the opposite SꢀPdꢀS angle, 85.7(1)°, is rela-
tively sharp (Scheme 1).
Selected bond distances (pm) and angles (°) for [Pd(h1-OAc)(h2-
MeSCHCO2Et)(PPh3)] (2)
Bond distances
PdꢀP
PdꢀO(3)
SꢀC(1)
O(1)ꢀC(2)
O(2)ꢀC(3)
O(4)ꢀC(5)
C(3)ꢀC(4)
226.9(1)
209.2(2)
175.9(2)
120.2(3)
145.5(3)
122.1(3)
146.3(5)
PdꢀS
PdꢀC(1)
SꢀC(7)
O(2)ꢀC(2)
O(3)ꢀC(5)
C(1)ꢀC(2)
C(5)ꢀC(6)
234.4(1)
203.5(3)
180.3(2)
134.5(2)
127.2(3)
147.3(3)
150.7(4)
The analytical and spectroscopic data are fully con-
sistent with the crystal structure of 2. The asymmetric
and symmetric stretching bands of the coordinated
acetate shifted to 1601 and 1373 cm−1 (1557 and 1414
in 1), respectively, whereas the ester carbonyl stretching
band was observed at 1705 cm−1, the same as in 1.
Several bands due to the presence of the phosphine
Bond angles
O(3)ꢀPdꢀC(1)
SꢀPdꢀO(3)
PꢀPdꢀO(3)
PdꢀSꢀC(7)
C(1)ꢀSꢀC(7)
PdꢀO(3)ꢀC(5)
SꢀC(1)ꢀC(2)
O(2)ꢀC(2)ꢀC(1)
O(1)ꢀC(2)ꢀO(2)
O(4)ꢀC(5)ꢀC(6)
were observed at 1582, 1480, 1435 and 1097 cm−1
.
159.37(8)
112.76(6)
96.76(6)
104.5(1)
102.4(1)
114.2(2)
117.6(2)
112.8(2)
123.5(2)
119.9(3)
SꢀPdꢀC(1)
PꢀPdꢀC(1)
PꢀPdꢀS
PdꢀSꢀC(1)
C(2)ꢀO(2)ꢀC(3)
PdꢀC(1)ꢀS
PdꢀC(1)ꢀC(2)
O(1)ꢀC(2)ꢀC(1)
O(3)ꢀC(5)ꢀO(4)
O(3)ꢀC(5)ꢀC(6)
46.72(6)
103.58(6)
150.18(2)
57.35(8)
116.2(2)
75.9(1)
116.5(2)
123.7(2)
124.9(2)
115.1(3)
More valuable information can be gathered from the
1H- and 31P{1H}-NMR spectra in solution. The most
1
interesting features of the H spectrum are the multi-
plicity of the signal due to the protons of the
methylthio group and of the signal from the protons of
the methylene in the ethoxy group. The former appears
as a doublet, due to the coupling (4JHP=4.6 Hz) with
the P nucleus of the phosphine in trans to the sulfur
donor atom. The same coupling was previously ob-
served in complex 5 (4JHP=4.0 Hz) [9]. A complex
multiplet arises from the methylene protons (ethoxy),
due to the geminal coupling between them and to the
vicinal coupling with the methyl protons. In fact, the
two methylene protons are not equivalent due to the
asymmetry of the methine carbon atom, which stems
from its coordination to the palladium atom. The anal-
2
ysis of the multiplet in [D7] DMF yielded the JHH and
Scheme 1.
3JHH values of 11.0 and 7.1 Hz, respectively. These are
practically the same as those observed in the protonic
spectra of complex 1, in which the signal of the corre-
sponding methylene shows the same hyperfine structure
[1]. These results are a clear evidence that in solution
the sulfur and methine carbon atoms of the ethyl
(methylthio)acetate anion are also both coordinated to
the palladium atom and that the phosphine is trans to
the former. The different mode of coordination of both
the acetate and the sulfur ligands are highlighted by the
chemical shifts of the relevant signals with respect to 1.
Thus, the protons of the h1 acetate resonate at 1.81
ppm versus 2.08 in complex 1. Also, the methyl and the
methylene protons of the ethoxy group resonate 0.3–
0.5 ppm upfield in comparison with 1 (0.92 and ca. 3.5
vs. 1.24 and 4.10, respectively). By contrast, the chemi-
cal shift of the CH3S hydrogen nuclei is ca. 0.4 ppm
downfield with respect to 1 (2.49 vs. 2.14). The proton
resonance of the coordinated methine strongly depends
on the solvent employed. It was observed at 3.70 ppm
in CDCl3 and at 4.14 ppm in [D7] DMF (3.63 ppm in
CDCl3 for 1). The difference between the 13C chemical
shifts of the methine carbon atom in 1 and 2, respec-
tively, are even more significant. In fact, whereas the
methine carbon atom in 1 resonates at 25.09 ppm
(CD2Cl4 [1]), its resonance was observed at 52.0 ppm
(CDCl3) in 2. The 31P{1H} spectrum is exceedingly
are, respectively, equal to 96.76(6) and 99.92(8)°. The
CꢀS distances in the three-membered rings are 175.9(2)
and 175.6(6) pm in 2 and 5, respectively: both these
values are lower than the respective SꢀCH3 bond
lengths, 180.3(2) and 180.7(7) pm. This suggests that
the CꢀS bond order in the palladacycle is higher than
one. No electronic delocalization, however, involves the
ethoxycarbonyl substituent, which maintains well-
defined single and double bonds.
The strongly distorted square-planar geometry
around the palladium atom is unaffected by substitu-
tion of the chloro ligand with the acetato and by
replacement of one hydrogen atom at the metal-bound
carbon atom with the bulkier ethoxycarbonyl group.
The geometry of the palladium(II) coordination sphere
is apparently dominated by the formation of the three-
membered palladacycle, at least in monomeric com-
plexes. We observed a slightly different situation in the
tetranuclear cationic complex [Pd4(m-OAc)3(m,h2-
CS)(m3,h2-CS)2(h2-CS)](BF4) (4) [2]. This compound
can be considered as a complex between 1, acting as a
bidentate S,S Lewis base, and the cationic, cyclic frag-
ment [Pd(h2-MeSCHCO2Et)]+. In this case, the con-
straint imposed by the relatively rigid structure of the
bidentate S,S ligand onto the palladacycle also influ-

206
M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
simple, with a single sharp resonance at 31.77 ppm
(32.99 ppm in [D7] DMF). This means that 2 does not
undergo any exchange reaction at room temperature.
The overall NMR picture demonstrates that complex 2
holds in solution the same structure as in the solid
state.
min after the addition of the phosphine to the starting
metal complex. After 4 h two bands, at 1202 and 1119
cm−1 and attributed to the presence of OꢂPPh3, were
detected. At the same time, the intensity of the bands of
the coordinated acetate decreased. After several hours
(overnight), the appearance of the band at 1732 cm−1
revealed the presence of the protonated sulfur ligand,
MeSCH2CO2Et. These results indicate that (i) the for-
mation of complex 2 and its further reaction with PPh3
(Eqs. (5) and (6)) are fast and that (ii) the reductive
elimination of AcOPPh3+ from 3, followed by the hy-
drolysis of the cation, takes place (Eqs. (7) and (8))
slowly. The latter observation is in agreement with the
results obtained with the [Pd(OAc)2]/PPh3 system,
where the same slow reductive elimination was ob-
served (first order in palladium concentration, t1/2=ca.
45 min, Pd:P=1:20, DMF) [7].
The formation and isolation of complex 2 is consid-
ered as particularly interesting. In fact, whereas the
reaction of Pd(OAc)2 with PPh3 to form [Pd(h1-
OAc)2(PPh3)2] is immediate [6], in our case the
monophosphino species is stabilised by the sulfur ligand
and could be isolated. The relatively high stability
arises from the chelate effect, which enhances the effec-
tive coordination capability of the sulfur donor atom
relative to that of the second phosphine. It is worth
noting that complex 2 is stable, despite the presence of
both one phosphine and one monohapto acetato, in cis
position, in the coordination sphere of palladium(II).
This is the same situation holding for [Pd(h1-
OAc)2(PPh3)2], the key precursor of the active catalyst
of the Heck reaction. However, whereas the latter
complex spontaneously gives reductive elimination, 2
does not.
[Pd(h1-OAc)(h1-CS)(PPh3)2]
“[Pd(h2-CS)(PPh3)2]− +AcOPPh3+
(7)
(8)
AcOPPh+3 +H2O“HOAc+OꢂPPh3+H+
Starting from complex 2, a second molecule of
triphenylphosphine can be coordinated to the palla-
dium centre. This is clearly demonstrated by H-NMR
Our NMR results are in full agreement with the above
picture. In fact, whereas after ca. 2 h only traces of
OꢂPPh3 were detectable by means of 31P-NMR, after
ca. 3.5 h the oxide amounted to about 10% of the total
phosphorus species in the reaction mixtures. After ca.
24 h, the triphenylphosphine oxide represented ca. 30%
of the total phosphorus, corresponding to the reduction
of ca. 60% of palladium(II) initially present. No further
formation of OꢂPPh3 was observed up to 122 h. The
occurrence of the reductive elimination was also confi-
rmed by the formation of significant amounts of
[Pd(O2)(PPh3)2], which is known to be formed by the
oxidative addition of dioxygen to some zero-valent
palladium species. This complex, which accounts for ca.
25% of the total phosphorus in solution after 122 h,
was detected by the 31P resonance at 33.49 ppm (33.42
1
in [D7] DMF: upon addition of one equivalent of
phosphine the coupling of the methylthio protons with
the phosphorus nucleus immediately disappeared. Their
chemical shift, in comparison with 2, is slightly upfield
(2.53 vs. 2.58 ppm) and that of the coordinated methine
0.20 ppm downfield (4.34 vs. 4.14 ppm). This pattern is
similar to that of [PdCl(h1-CH2SMe)(PPh3)2] (6) [17].
Accordingly, these results indicate that the second
phosphine displaces the sulfur donor atom from the
palladium
centre
and
that
[Pd(h1-OAc){h1-
CH(SMe)CO2Et}(PPh3)2] (3, Eq. (6)) is formed. The
relevant 31P{1H}-NMR spectrum in [D7] DMF at 27°C
shows a single, broad peak observed at 17.19 immedi-
ately after the addition of the phosphine, which sug-
gests the occurrence of fast phosphine exchange
equilibria. The reversible formation of cationic species
upon detachment of the acetato ligands can be ruled
out on the basis of the protonic spectrum of 3, although
it is known that complex 6 can lose reversibly the
chloro ligand [13].
n
ppm in a 0.3 M solution of Bu4NBF4 in DMF) [18].
The formation of [Pd(O2)(PPh3)2] suggests that its par-
ent zero-valent palladium species already contained at
least two phosphine molecules per palladium atom. It is
noteworthy that, no evidence for the formation of
metallic palladium was gained and all the reaction
mixtures always remained as clean solutions. The pro-
ton spectra of the reaction mixture (t\4 h) are rather
complex. Several resonances were observed in the
higher field region (lB4.5), but none of them could be
attributed to either the free sulfur ligand or the palladi-
um(II) complexes 1, 2 and 3. Since protons are sup-
Attempts to isolate a pure sample of 3 were unsuc-
cessful, due to the rapid equilibrium reactions involving
1–3 and to the progressive reduction of Pd(II) to Pd(0).
The reaction of 1 and of 2 with triphenylphosphine to
yield 3 was followed, respectively, by FTIR in THF
solution and 1H- and 31P{1H}-NMR spectroscopy in
DMF solution. The FTIR investigation on the disap-
pearance of 1 revealed a rapid change of the coordina-
tion mode of the acetato ligands. The typical band of
the monohapto acetate at 1609 cm−1 appeared within 5
posed
to
form
upon
hydrolysis
of
the
acetoxytriphenylphosphonium cation (Eqs. (7) and (8)),
it seems likely that the sulfur ligand was protonated
and that the observed resonances stemmed from coor-
dinated, neutral MeSCH2CO2Et.

M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
207
The reductive elimination stopped at about 50%
yield. One possible explanation is that as soon as
the first reaction product, [Pd(MeSCHCO2Et)(PPh3)]−
(7), is formed, it reacts with PPh3 to give
[Pd(MeSCHCO2Et)(PPh3)2]− (8) (Scheme 2), which
eventually transforms into [Pd(O2)(PPh3)2]. In fact, 7 is
expected to be unstable because of coordinative unsatu-
ration and the strain of the three-membered palladacy-
cle. The strain could be released by ring opening, but
an even less coordinated species would be obtained.
Therefore, 7 took up a further phosphine molecule. As
only two equivalents of phosphine per palladium were
present in the system, 7 competed effectively with 2 for
the additional phosphine molecule. This implies that
when about half of the total palladium has been re-
duced, 3 is essentially no longer present in solution and
the reductive elimination stopped. The electrochemical
data reported below fully support this hypothesis. The
addition of excess triphenylphosphine to a solution of 1
in anhydrous benzene (P:Pd=5:1) led to the formation
of [Pd(PPh3)4], which was isolated from the reaction
mixture.
Cyclic voltammetry allowed us to confirm the above
findings and provided additional information on both
the solution equilibria and the reactivity of zero-valent
palladium species. The voltammetric analysis was car-
ried out in DMF–0.1 M nBu4NClO4, using a glassy
carbon electrode. Although we found that the voltam-
metric pattern at a gold electrode was essentially the
same, the glassy carbon electrode proved to be superior
from the point of view of both the quality of the curves
and the reproducibility of the results. At a scan rate of
0.2 V s−1, the electroreduction of 1 is chemically
irreversible and the peak is located at −1.31 V. Start-
ing from 0.5 V s−1, however, the anodic peak corre-
sponding to the re-oxidation of the species
electrogenerated upon reduction of 1 is detectable dur-
ing the return scan. Digital simulation of the voltam-
metric curve led us to estimate the lifetime of the
reduction product to be ca. 0.1 s. By using the voltam-
metric data obtained at high scan rates, the standard
potential for the reduction of 1 could be calculated to
be −1.24 V. The complex was found to be stable in
solution, up to several hours. Upon addition of PPh3
(1:1 Pd:P), complex 2 was rapidly formed. In fact, the
same voltammetric pattern was observed by using an
authentic sample of 2. The voltammetric behaviour of 2
was quite different from that of 1. When scanning the
potential first in the negative going direction, a main
irreversible reduction peak at −1.46 V (at 0.2 V s−1
)
was observed (Fig. 2).
The peak was preceded by a small wave at ca. −1.1
V, the current of the wave decreasing slightly with
respect to the main peak when the scan rate was
increased. In fact, the two processes are linked. While
the main peak decreases upon stepwise addition of
PPh3, the wave increases accordingly. Finally, when 1
equivalent of PPh3 was added to the solution of 2, i.e.
under the conditions in which 3 is formed, a nearly
complete transformation of the two peaks was ob-
served: The original wave was now a fully developed
peak having the same height as the main peak of 2, but
a more positive peak potential, −1.06 V. The voltam-
metric pattern is in agreement with Eq. (6) being very
displaced to the right-hand side. In addition, it suggests
that there is an equilibrium in which 2 can exchange
PPh3 with another 2 molecule to form 3, although to a
very small extent.
Scheme 2.
Reduction of 2 produces irreversibly a palladium(0)
species that can be oxidised on the backward scan,
giving rise to a broad and irreversible peak that, at 0.2
V s−1, is located at 0.42 V (Fig. 2). The peak is
preceded by a much smaller peak at 0.0 V. Both peaks
can be observed only after reduction of 2, the reduction
wave of 2 being connected to the most positive oxida-
tion peak. Upon stepwise addition of PPh3, the oxida-
Fig. 2. Cyclic voltammetry curves for the reduction of 2.7 mM 2 in
DMF–0.1 M Bu4NClO4 at a glassy carbon electrode. The five curves
were recorded after addition of 0, 0.25, 0.50, 0.75, and 1 equivalent of
PPh3. The arrows indicate the corresponding variations of the reduc-
tion and oxidation peaks, 0.2 V s−1, 25°C.

208
M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
voltammetry experiment starting from −0.3 V, where
neither oxidation nor reduction may take place, and
scanning the potential first toward positive values. Fig.
3(a) also shows that by reversing the potential scan past
the oxidation peak and running a full cyclic voltamme-
try to include reduction of 3, an oxidation peak, having
the same potential as that of the peak that forms
spontaneously by chemical decay of 3, develops. Fig.
3(b) illustrates the evolution of the voltammetric pat-
tern during ca. 3 h.
While the cathodic peak decreases until the peak
current levels off, the current of the anodic peak in-
creases, reaches a maximum, and (not shown in the
figure) starts to decrease. The fact that the spontaneous
reduction of 3 is not quantitative but stops after ca. 3 h
and that decomposition of the palladium(0) species is
observed afterwards is completely coherent with the
FTIR and NMR findings. Again the picture coming
from these experiments is that the formation of the
palladium(0) species is followed by some equilibrium in
which a further PPh3 molecule is requested to stabilise
the low oxidation state of palladium. In the absence of
an excess of PPh3, the additional PPh3 molecule seems
to be provided by 3 itself, which transforms into 2,
leading to a dramatic decrease of the rate of the reduc-
tive elimination.
Cyclic voltammetry experiments were carried out to
study the reactivity of the zero-valent palladium species
toward PhI, at 25°C. This is a typical oxidative addition
previously used with another palladium(0) system and
thus considered as a good test to evaluate the relative
reactivity of these two electron-donating systems. The
experiments were carried out by voltammetric electro-
generation of the reactive zero-valent species, through
reduction of 3, first in the absence and then in the
presence of 20 equivalents of PhI. Fig. 4 shows very
convincingly that whereas at 0.2 V s−1 no oxidation
peak was observed in the presence of PhI, at 10 V s−1
the oxidation peak is reduced to ca. half of its original
value. At 200 V s−1 the oxidation peak is practically
the same as if the halide was not present in solution.
Analysis of the cyclic voltammograms obtained in the
0.1–200 V s−1 range, in comparison with the corre-
sponding digitally simulated curves, led us to estimate
the rate constant for the oxidative addition to be in the
Fig. 3. Cyclic voltammetry curves obtained in a solution containing
4.0 mM 3 in DMF–0.1 M Bu4NClO4 at 60°C. Plot (a) was recorded
a few minutes after taking the solution to 60°C. The curves in plot (b)
show, qualitatively, the evolution of the voltammetric pattern during
ca. 3 h. The arrows indicate the corresponding variations of the
reduction and oxidation peaks. For both plots the initial scan was
from −0.3 V to more positive potentials. Gold electrode, 0.2 V s−1
.
tion peaks undergo a similar pattern as that already
described for the negative going scan. While the broad
peak decreases, the smaller, less positive oxidation peak
develops into a sharp and irreversible peak located at
−0.04 V (at 0.2 V s−1). If more than 1 equivalent of
PPh3 is added to the solution of 2 (or more than 2
equivalents to a solution of 1), both this oxidation peak
and the main reduction peak (at −1.06 V) do not
change significantly, again pointing to the equilibrium
for the formation of 3 (Eq. (6)) being almost completely
displaced on the right-hand side.
As described above, complex 3 is more easily reduced
than 2, by ca. 0.4 V. Analogously, the reduction
product of 3 is more easily oxidised than that of 2 by
ca. 0.4 V. This points to a greater reactivity of 3 relative
to 2. In fact, whereas complex 2 appears to be stable in
solution, even after several hours, we found that addi-
tion of PPh3 leading to 3 leads to chemical instability of
the palladium(II) complex. Although the effect is small
at 25°C, if the solution is taken at 60°C, an oxidation
peak spontaneously forms at ca. 0 V in a few minutes
(Fig. 3(a)). This was ascertained by running the cyclic
range 1–2×102 M−1 −1. This value is comparable
s
with 140 M−1 s−1 reported for the corresponding
oxidative addition of PhI to [Pd(h1-OAc)(PPh3)2]− [8].
The latter value, however, was determined for [Pd(h1-
OAc)(PPh3)2]− generated from [Pd(OAc)2]+3PPh3.
Under these conditions, hydrogen ions form and this
has been envisaged as a factor enhancing the rate of
oxidative addition of PhI to palladium(0) [8]. On the
other hand, the value reported here pertains to electro-
generated palladium(0), i.e. to conditions under which
no hydrogen ions form. This would support the idea

M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
209
that the palladium(0) species obtained by electrochemi-
cal reduction of 3 is more reactive than [Pd(h1-
OAc)(PPh3)2]−. The electrochemical reduction of 3
should afford directly 8 (Scheme 3(a)), which in turn
could undergo at least two equilibrium reactions
(Scheme 4).
Complex 7 is the most likely product of electroreduc-
tion of 2 (Scheme 3(b)). As mentioned above, its oxida-
tion peak is observed at a more positive potential with
respect to 8. This suggests that it should be less reactive
in the oxidative addition of PhI than 8 (or 9). Under
these conditions, the intermediacy of a species without
the sulfur ligand can be ruled out. In fact, the protona-
tion of the ligand cannot occur and its release in
solution seems unlikely. On the other hand, 8 is coordi-
natively saturated and this would suggest that the active
palladium(0) species is 9. The presence of a s-bonded
carbon ligand would suggest even higher reactivity than
observed: kinetic constants as high as 104 M−1 s−1
have been reported for related s-aryldiphosphino–pal-
ladate(0) complexes, [Pd(Ar)(PPh3)2]− [19]. However, in
our case the expected effect of the carbon ligand is
likely mitigated by the low concentration of 9 in equi-
librium with the main species 8.
The oxidative addition of PhI was also observed
under conditions of chemical reduction. In the presence
of PPh3 (1:2 Pd:P) and one out of several bases
(nBu4NOAc, NPh3, PS; 1:1, 1:1, 2:1 base:Pd), the reac-
tion of PhI and 1 in anhydrous benzene gave mixtures
of acetato- and iodo-phenyl derivatives of palladium-
n
1
(II). In the presence of Bu4NOAc, the H-NMR data
show that the main product was [Pd(h1-OAc)-
(Ph)PPh3]2, although the elemental analysis indicated
that it was contaminated by some iodo derivative. The
presence of acetato ligands in the products of oxidative
addition of iodobenzene to palladium(0) has been al-
ready observed [8].
We also believe that in this case the anionic species 8
is the key Pd(0) intermediate. However, the presence of
hydrogen ions and acetic acid (side products of the
reductive elimination from 3) could lead to neutral
Pd(0) complexes containing the protonated thioether
ligand.
Fig. 4. Cyclic voltammetry curves for the reduction of 2.7 mM 3 in
DMF–0.1 M Bu4NClO4 at 0.2 (plot a) and 10 V s−1 (plot b). For
both plots, the curves were obtained first in the absence (— —) and
then in the presence (········) of 20 equivalents of PhI. Glassy carbon
electrode, 25°C.
Finally, we attempted to carry out the reaction of 1
with iodomethane, in the presence of PPh3 (3:1 P:Pd),
as an example of reaction with a different aliphatic
halide. However, in this case no base was added to the
reaction mixtures. The analytical and spectroscopic
data show the presence of [PdI(h2-MeSCHCO2Et)-
(PPh3)] and [Pd{h1-CH(SMe)CO2Et}(PPh3)3]I. The for-
mation of both is easily explained on the basis of an
electrophilic attack of MeI to the sulfur atom, followed
by AcO−/I− [20], I−/PPh3, PPh3/SMe exchange pro-
cesses.
Scheme 3.
4. Conclusions
[Pd3(m-OAc)3(m,h2-MeSCHCO2Et)3] (1) shows a sub-
stantially similar behaviour towards triphenylphosphine
in comparison with [Pd(OAc)2]. However, due to the
stabilization of palladium(II) arising from the chelate
Scheme 4.

210
M. Basato et al. / Journal of Organometallic Chemistry 601 (2000) 201–210
MeSCHCO2Et− ligand (chelate effect, effective compe-
tition of the sulfur donor atom with phosphine for
metal coordination) its reactivity towards reductive
elimination of AcOPPh+3 can be controlled. This allows
the synthesis of the monoacetato-monophosphino inter-
mediate [Pd(h1-OAc)(h2-MeSCHCO2Et)(PPh3)] (2)
from 1 and the determination of its crystal structure. By
contrast, [Pd(h1-OAc)2(PPh3)2] is formed instanta-
neously from [Pd(OAc)2]. Remarkably, complex 2 does
not undergo reductive elimination, although an acetato
ligand and a triphenylphosphine molecule are present,
in mutual cis position, in the coordination sphere of
palladium(II). This demonstrates that this is not a
sufficient condition for the reductive elimination to take
place. Elimination occurs only upon addition of an
equivalent of PPh3 to complex 2, leading to the forma-
tion of the intermediate palladium(II) diphosphino
complex [Pd(h1-OAc){h1-CH(SMe)CO2Et}(PPh3)2] (3).
The latter is unstable in DMF solution and gives a
spontaneous reductive elimination. However, this pro-
cess is slower than reported for the related complex
[Pd(h1-OAc)2(PPh3)2]. In the presence of PhI, a zero-va-
lent palladium species, most likely arising from 8, un-
dergoes fast oxidative addition. Conversely, no
oxidative addition was observed with MeI in the pres-
ence of triphenylphosphine. In this case, the exchange
between the acetato and the iodo ligand is faster than
reductive elimination and takes place instead.
Acknowledgements
We thank Mr A. Ravazzolo for skilled technical
assistance. S.A. and F.M. thank the Consiglio
Nazionale delle Ricerche for financial support.
References
[1] M. Basato, A. Grassi, G. Valle, Organometallics 14 (1995)
4439.
[2] M. Basato, C. Bertani, B. Sesto, M. Zecca, A. Grassi, G. Valle,
J. Organomet. Chem. 552 (1998) 277.
[3] M. Basato, C. Bertani, M. Zecca, A. Grassi, G. Valle, J.
Organomet. Chem. 575 (1999) 163.
[4] W.A. Herrmann, in: B. Cornils, W.A. Herrmann (Eds.), Ap-
plied Homogeneous Catalysis with Organometallic Compounds,
vol. II, VCH, Weinheim, 1996, p. 712.
[5] F. Ozawa, A. Kubo, T. Hayashi, Chem. Lett. (1992) 2177.
[6] C. Amatore, A. Jutand, M.A. M’Barki, Organometallics 11
(1992) 3009.
[7] C. Amatore, E. Carre´, A. Jutand, M.A. M’Barki, Organometal-
lics 14 (1995) 1818.
[8] C. Amatore, E. Carre´, A. Jutand, M.A. M’Barki, G. Meyer,
Organometallics 14 (1995) 5605.
[9] K. Miki, Y. Kai, N. Yasuoka, N. Kasai, J. Organomet. Chem.
135 (1977) 53.
[10] S. Antonello, M. Musumeci, D.D.M. Wayner, F. Maran, J.
Am. Chem. Soc. 119 (1997) 9541.
[11] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Funda-
mentals and Applications, Wiley, New York, 1980.
[12] J. Tsuji, Palladium Reagents and Catalysts, Wiley, Chichester,
1995.
[13] W.A. Herrmann, C. Brossmer, C.P. Reisinger, T.H. Riermeier,
8
K. Ofele, M. Beller, Chem. Eur. J. 3 (1997) 1357.
5. Supplementary material
[14] M. Beller, C.P. Reisinger, Eur. J. Inorg. Chem. (1998) 29.
[15] F.G. Bordwell, X.-M. Zhang, M.S. Alnajjar, J. Am. Chem.
Soc. 114 (1992) 7623.
Crystallographic data (excluding structure factors)
for the structure reported in this paper have been
deposited with the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC
113856. Copies of the data can be obtained free of
charge on application to 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).
[16] F. Maran, D. Celadon, M.G. Severin, E. Vianello, J. Am.
Chem. Soc. 113 (1991) 9320.
[17] G. Yoshida, H. Kurosawa, R. Okawara, J. Organomet. Chem.
113 (1976) 85.
[18] C. Amatore, A. Jutand, M.J. Medeiros, New J. Chem. 20
(1996) 1143.
[19] C. Amatore, E. Carre´, A. Jutand, H. Tanaka, Q. Ren, S. Torii,
Chem. Eur. J. 2 (1996) 957.
[20] H.D. McPherson, J.L. Wardell, Inorg. Chim. Acta 81 (1984)
39.
.