Chiral diaminocarbene palladium(II) complexes: Synthesis, reduction to Pd(0) and activity in the Mizoroki-Heck reaction as recyclable catalysts

Article abstract of DOI:10.1016/S0022-328X(03)00476-5

A preparation of chiral palladium(II) bis-diaminocarbene complexes was developed from the corresponding silver(I) diaminocarbenes and bis(acetonitrile)dichloropalladium(II). Crystal structure details of trans -diiodo bis[(4 R ,5 R )-1,3-dibenzyl-4,5-di-tert- butylimidazolin-2-ylidene] palladium(II) are presented. Mixed diaminocarbene-phosphine complexes were also prepared in two steps from the corresponding imidazolidinium salts and palladium acetate. The very high air, moisture and thermal stability of such complexes were confirmed. These two families of complexes were tested in the Mizoroki-Heck reaction. Their ability to be reduced into Pd(0) which undergoes oxidative addition with aryl halides was studied by cyclic voltammetry. The ability of different agents to generate Pd(0) complexes and the reactivity of these latters with aryl halides were also studied in situ. The obtained results allowed us to perform Heck reactions at temperatures starting from 40 °C in DMF, by activation of the palladium(II) catalyst precursor. Moreover, in the case of the bis-diaminocarbene complexes, the Pd(0) catalysts were totally re-oxidized during the work-up and the starting Pd(II) complexes could be quantitatively recovered.

Full text of DOI:10.1016/S0022-328X(03)00476-5

Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
www.elsevier.com/locate/jorganchem
Chiral diaminocarbene palladium(II) complexes: Synthesis, reduction
to Pd(0) and activity in the MizorokiꢀHeck reaction as recyclable
catalysts
/
a
Julien Pytkowicz , Sylvain Roland *, Pierre Mangeney , Gilbert Meyer ,
a,
a
b
b,
Anny Jutand *
a
e` me
Laboratoire de chimie organique, UMR 7611, Universit e´ Pierre et Marie Curie, 4, place Jussieu, tr. 44-45 2
b
Departement de chimie, UMR 8640, Ecole Normale Sup e´ rieure, 24, rue Lhomond, 75231 Paris Cedex 05, France
e´ t., 75252 Paris Cedex 05, France
Received 5 March 2003; received in revised form 20 May 2003; accepted 20 May 2003
Abstract
A preparation of chiral palladium(II) bis-diaminocarbene complexes was developed from the corresponding silver(I)
diaminocarbenes and bis(acetonitrile)dichloropalladium(II). Crystal structure details of trans-diiodo bis[(4R,5R)-1,3-dibenzyl-
,5-di-tert-butylimidazolin-2-ylidene] palladium(II) are presented. Mixed diaminocarbeneꢀphosphine complexes were also prepared
in two steps from the corresponding imidazolidinium salts and palladium acetate. The very high air, moisture and thermal stability
of such complexes were confirmed. These two families of complexes were tested in the MizorokiꢀHeck reaction. Their ability to be
4
/
/
reduced into Pd(0) which undergoes oxidative addition with aryl halides was studied by cyclic voltammetry. The ability of different
agents to generate Pd(0) complexes and the reactivity of these latters with aryl halides were also studied in situ. The obtained results
allowed us to perform Heck reactions at temperatures starting from 40 8C in DMF, by activation of the palladium(II) catalyst
precursor. Moreover, in the case of the bis-diaminocarbene complexes, the Pd(0) catalysts were totally re-oxidized during the work-
up and the starting Pd(II) complexes could be quantitatively recovered.
#
2003 Elsevier Science B.V. All rights reserved.
Keywords: N-heterocyclic carbenes; Palladium; Heck reaction; Cyclic voltammetry; Recyclable catalysts
1
. Introduction
reaction [5]. This group postulated that the Pd(II)
species, a stable catalyst precursor, might be activated
into Pd(0) for the coupling to occur. Since then, carbene
complexes of Pd(II) or Pd(0) have been widely used as
N-heterocyclic carbenes (NHC) have recently
emerged as an important family of ligands with strong
s-donor electronic properties [1ꢀ4]. In contrast to the
widely used phosphines complexes, most of the com-
plexes formed with these ligands are exceptionally stable
toward heat, air and moisture. Indeed, NHC are tightly
bound to the metal thereby avoiding decomposition
pathway or deposition of free (and inactive) metal under
catalytic conditions.
/
catalysts or catalyst precursors for Heck [6ꢀ
Suzuki [10,11,17,18,20,24ꢀ31] reactions, a-arylations of
ketones or amides [32ꢀ35], and other CÃC couplings
/23] or
/
/
/
[20,36ꢀ/40], as well as N-arylations of amines or imines
[
resolution of alcohols [46].
34,41ꢀ/44], allylic alkylations [45] and oxidative kinetic
Generally, Heck reactions conducted with these com-
plexes required high temperatures (higher than 120 8C)
and polar solvents. These drastic conditions, that
seemed to be necessary to get good conversions, have
highlighted the stability of many of these Pd species. For
instance, couplings of aryl chlorides could be performed
at 140 8C by stabilizing the active Pd(0) complex in ionic
liquids [7].
Herrmann et al. reported for the first time in 1995 the
use of palladium(II) NHC complexes in the Heck
*
Corresponding author. Tel.: ꢁ
567.
E-mail address: sroland@ccr.jussieu.fr (S. Roland).
/
33-1-4427-5567; fax: ꢁ33-1-4427-
/
7
0
022-328X/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00476-5

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
167
Several syntheses of chiral NHC complexes of palla-
dium have been reported in the literature [16,20,22,47ꢀ
9]. But, to our knowledge, the only example of
asymmetric catalysis with Pd carbene complexes, giving
good selectivities, was reported by Hartwig and co-
workers in 2001 in the a-arylation of amides [32] with
enantiomeric excesses up to 76%. However, maybe as a
consequence of the harsh conditions required for the
Heck reaction, only one example of an asymmetric
preparation of several imidazolin-2-ylidene copper com-
plexes which were found to be efficient catalysts for the
enantioselective conjugate addition of diethylzinc to
enones [52,53], affording enantioselectivities up to
88%. In addition to the fact that these silver(I) carbene
complexes are generally air-stable and easy to handle,
they could often act as effective carbene transfer agents
towards various transition metals, including palladium
[51]. This property was recently exploited by several
groups for the synthesis of palladium(II) imidazol-2-
ylidene complexes. [10,15,47].
/
4
version, with a low asymmetric induction (e.e.ꢂ
/8%),
was mentioned by Enders et al. [16].
The generation of the active species and its exact
nature and reactivity, was not precisely determined even
though, as cited above, Herrmann et al. reported
increased activities in the presence of a reducing agent
Thus, we first examined the ability of the chiral silver
complex 1 to act as a carbene transfer agent towards
Pd(II) complexes. Indeed, treatment of 1 by 0.5 equiva-
lent of PdCl (CH CN) in CH Cl at 20 8C, led after 4
2
3
2
2
2
[
5]. In order to develop an asymmetric version of the
h, to a mixture of the bis-carbene complexes 2a and 2b in
95% yield (Scheme 2).
Heck reaction with NHC complexes of palladium, it
would be of interest to be able to perform the reaction
under conditions as mild as possible. Therefore, the
determination of the limitant steps of the coupling
reaction is of great importance.
Another important point concerns the regeneration
and re-use of the catalyst. Polymer-supported carbene
complexes of Pd(II) as recyclable catalysts for the Heck
reaction were recently reported [19]. The possibility to
recycle non-supported palladium(II) carbene complexes
was also already considered by Enders et al. [16].
However, this opportunity was not widely studied in
homogeneous catalysis but could be of particular
interest in the case of chiral catalysts whose preparations
are often expensive and require multi-step syntheses.
We wish to report here the synthesis of several chiral
imidazolin-2-ylidene palladium(II) complexes, their ac-
These two complexes were easily isolated by filtration
of the silver salts and concentration. The yellow solid
obtained was air-stable and non-sensitive to water.
These complexes were also stable on silica gel and could
be separated by flash-chromatography. This purification
allowed the full characterization of each complex and
the determination of their ratio in the crude mixture by
NMR (2a/2bꢂ70:30). A trans configuration was attrib-
/
uted to the major isomer 2a since its NMR spectrum
was characteristic of a symmetrical complex (C sym-
2
1
metry). The H-NMR signals of 2b were representative
1
of a non-symmetrical complex. The C-NMR chemical
3
shift of the carbene carbon (C ÃPd) in the minor isomer
/
2b (198.4 ppm) was very similar to the one of 2a (198.6
ppm). Since the difference of chemical shifts of the
carbene carbon between trans or cis palladium diami-
nocarbene complexes was shown to be generally im-
portant [11], we assumed that 2b was the trans
diaminocarbene complex with one bromide and one
chloride on the palladium and not the cis complex.
tivity in the MizorokiꢀHeck reaction, their reduction
/
into Pd(0) and the re-oxidation of the Pd(0) species. We
also wish to present the possible use of some of these
Pd(II) complexes as recyclable chiral catalysts.
2
. Results and discussion
2
.1. Synthesis of bis-carbene palladium(II) complexes
Recently, we have reported the preparation of various
chiral silver(I) diaminocarbenes such as 1 [50], using the
method described by Wang and Lin [51] (Scheme 1).
These chiral complexes were used for the in situ
Scheme 1. Preparation of silver(I) diaminocarbene complexes.
Scheme 2. Preparation of Pd(II) bis-diaminocarbene complexes.

168
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
Although the separation of the complexes 2a and 2b
was necessary to determine their exact structure, it was
rather delicate and arduous. Therefore, the mixture of
Table 1
Crystallographic data for compound 2c
Compound
2c
2
a/2b was quantitatively transformed into the trans
Formula
Colour
C
50
H
68PdI
Yellow
Orthorhombric
2 4
N
diiodo complex 2c by treatment with an excess of
sodium iodide in CH Cl at 20 8C (Scheme 2). This
2
2
Crystal class
Space group
Z
halogen exchange at room temperature allowed the one
step synthesis of complex 2c in 87% yield from the silver
carbene 1 and PdCl (MeCN) , in the presence of an
1 1 1
P2 2 2
4
Unit cell parameters
˚
2
2
a (A)
˚
16.300 (12)
17.607 (9)
17.780 (9)
5103 (5)
excess of sodium iodide (Scheme 3).
b (A)
The structure of 2c was determined by X-ray spectro-
scopy (Fig. 1). The crystallographic data are shown in
Table 1. Selected bond lengths and angles are given in
Table 2. This trans bis-square planar complex of Pd(II)
shows a high degree of symmetry. The diaminocarbenes
˚
c (A)
3
Volume (A )
˚
Radiation type
˚
MoꢀKa
0.71069
1.41
/
Wavelength (A)
Density (g cmꢃ3)
ꢃ1
M (g mol
ꢃ1
)
1085.34
16.08
planes are both oriented of 69.38 and ꢃ68.18 towards
/
m (cm
)
the coordination plan of the palladium and the dihedral
angle between the two diaminocarbene subunits is of
Temperature (K)
Shape
Diffractometer
295
Parallelepiped
Enraf-Nonius Mach-3
4983
4
2
2.68. The palladiumꢀ
/
carbene bond length is of about
˚
.03 A.
Reflections measured
Independent reflections
4952
2.2. Synthesis of mixed phosphineꢀ
palladium(II) complexes
/carbene
phosphineꢀ
yield.
/carbene complex 6 was obtained in 82%
Herrmann et al. and Nolan and co-workers have
reported the superior activity of mixed phosphineꢀ
2.3. Catalytic activities of the palladium(II) carbene
complexes 2c and 6 in the Heck reaction
/
carbene complexes compared to their bis-carbene ana-
logues [20,21]. To synthesize such complexes we first
attempted, by analogy with the preparation of com-
plexes 2, to transfer only one diaminocarbene unit from
We first tested the activity and stability of 2c in the
classical Heck coupling of phenyl iodide with butyla-
crylate (Scheme 5) [54]. The reaction performed at
the silver complex
1
to one equivalent of
PdCl (CH CN) . Unfortunately, by this procedure, the
1
20 8C in DMF, in the presence of 2% mol of catalyst
2
3
2
2c, led after 16 h to a complete conversion into (E)-
bis-carbene 2a was obtained in poor yield together with
a mixture of non-characterized species. We turned then
towards the alternative methodology described by
Herrmann et al. [20]. The equimolar reaction of the
Table 2
˚
Selected bond lengths (A) and angles (8) for 2c
Compound
2c
imidazolinium salt 4 with Pd(OAc) in THF at 20 8C, in
2
the presence of potassium tert-butoxide and sodium
iodide (four equivalents), led to the stable and isolable
Pd(II) dimeric complex 5 (Scheme 4) which was
Bond lengths
Pd(1)Ã
Pd(1)Ã
N(1)Ã
N(101)Ã
N(3)ÃC(2)
N(103)ÃC(102)
/
C(2)
2.027(8)
2.038(8)
1.36(1)
1.34(1)
1.33(1)
1.32(1)
1.49(1)
1.477(11)
/
C(102)
/C(2)
obtained in 72% yield as a redꢀbrown crystalline solid.
/
/C(102)
This dimer was easily and rapidly cleaved by triphenyl-
phosphine at 20 8C in CH CN or CH Cl . The orange
/
3
2
2
/
N(1)Ã
/
C(5)
crystalline solid thus obtained was also air and moisture
stable and was meticulously purified by silica gel
chromatography to remove possible traces of PPh3
and other residual palladium complexes. The mixed
N(3)Ã
/
C(4)
Bond angles
C(2)ÃPd(1)Ã
N(1)ÃC(2)ÃN(3)
N(101)ÃC(102)Ã
Pd(1)ÃC(2)ÃN(1)
Pd(1)ÃC(2)ÃN(3)
C(2)ÃN(3)Ã
C(2)ÃN(1)Ã
C(2)ÃN(3)Ã
C(2)ÃN(1)Ã
/
/
C(102)
177.0(3)
108.4(7)
108.9(7)
125.5(5)
126.1(6)
113.8(7)
111.8(6)
122.0(7)
122.0(6)
/
/
/
/N(103)
/
/
/
/
/
/
C(4)
/
/
C(5)
/
/
C(14)
C(15)
/
/
Scheme 3. Preparation of complex 2c.

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
169
Fig. 1. ORTEP view of complex 2c.
butylcinnamate (Table 3, entries 1ꢀ
/
2). To test further
obtained without palladium metal deposit, demonstrat-
ing the stability of the catalyst in these conditions
the long-term stability of 2c, very small amounts of
catalyst were used (0.05 and 0.005% mol) in the presence
of either sodium acetate or diisopropylethylamine. After
(entries 3ꢀ5). Moreover, in the standard conditions
/
using 2% of catalyst 2c (21.7 mg), we were surprised to
1
detect by H-NMR the presence of the initial Pd(II)
9
6 h at 120 8C, turn-over numbers up to 14 600 were
catalyst in the crude mixture after work-up. More than
90% of 2c could be recovered by simple precipitation in
MeOH and drying. As mentioned before, Enders et al.
already reported the possibility to recover their bis-
carbene Pd(II) catalysts after the Heck reaction and they
suggested, that a Pd(IV) intermediate may be involved
in the catalytic cycle [16].
Heck reactions were then attempted at lower tem-
peratures. At 80 8C, still with phenyl iodide, the
reactivity decreased dramatically and only poor conver-
sions were obtained with both bases tested (Table 3,
entries 6 and 7).
Scheme 4. Preparation of the mixed phosphineꢀcarbene complex 6.
/
Scheme 5. Heck reactions with complexes 2c and 6.

170
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
Table 3
Heck reactions with complexes 2c and 6
a
Catalyst % mol
a
a
b
Entry
PhX
Base
T (8C)
t (h)
Conv. (%)
TON
50
1
2
3
4
5
6
7
2c 2%
2c 2%
PhI
PhI
PhI
PhI
PhI
PhI
PhI
iPr
AcONa
iPr EtN
2
EtN
120
120
120
120
120
80
16
16
96
96
96
16
16
100
100
93
50
1860
2c 0.05%
2c 0.05%
2c 0.005%
2c 2%
2
AcONa
AcONa
100
73
2000
14 600
iPr
2
EtN
1
2c 2%
AcONa
80
10
8
9
2c 2%
6 2%
6 2%
6 2%
PhBr
PhI
PhI
AcONa
AcONa
AcONa
120
120
80
16
16
16
16
2
96
5
1
1
0
1
PhI
iPr
2
EtN
80
0
a
1 mmol of PhX, 1.5 mmol of base, % mol of catalyst based on the amount of PhX.
Dodecane as internal reference.
b
The bis-carbene catalyst 2c was also ineffective, even
at 120 8C, to promote the coupling reaction of phenyl
bromide and butylacrylate (entry 8).
2.4. Reduction of palladium(II) complexes 2 and 6 to
palladium(0) and re-oxidation of the generated
palladium(0) complexes
Using the mixed NHCꢀphosphine complex 6, the
/
reaction proceeded rapidly at 120 8C with no traces of
Pd metal deposit (entry 9). However, contrary to the bis-
carbene complex 2c, the mixed complex 6 was not
recovered after the usual work-up but was only detected
2
.4.1. Electrochemical reduction of palladium(II)
complexes 2 and 6 to palladium(0) complexes:
determination of the rate constant of the oxidative
addition of aryl halides to the electrogenerated
palladium(0) complexes
1
as traces together with traces of 2c by H-NMR of the
crude reaction mixture.
At 80 8C, a very low, or no conversion, was obtained
with both bases (entries 10 and 11).
In order to explain the lack of reactivity of our
catalysts at 80 8C, even with PhI, and to understand how
a Pd(II) complex could be recovered at the end of the
reaction, we decided to investigate the reduction of
complexes 2 and 6 into Pd(0) and the re-oxidation of the
latter.
The electrochemical reduction of palladium(II) to
palladium(0) complexes allows the characterization of
the electrogenerated palladium(0) complexes in terms of
nucleophilic properties by the determination of their
oxidation potential, as well as the characterization of
their reactivity in oxidative addition of aryl halides by
the determination of the rate constants [55,56].
The electrochemical reduction of complexes 2a, 2c
and 6 (3 mM) was investigated in DMF. Whatever the
Fig. 2. (a) (*
.5 mm) with a scan rate vꢂ
bis(carbene)Pd(0) (3 mM) generated by the reduction of complex 2c in DMF at 20 8C: plot of the molar fraction x of the unreacted Pd(0) complex
/
) Reduction of complex 2c (3 mM) in DMF (containing n-Bu
4
NBF
4
, 0.3 M): cyclic voltammetry at a steady gold disk electrode (i.d.ꢂ
/
ꢃ1
0
/10 V s , at 20 8C. (---) In the presence of PhBr (3 mM). (b) Kinetics of the oxidative addition of PhBr (3 mM) with the
0
0
versus log([PhBr]/v) (xꢂ
/
i
ox/iox; iox: oxidation peak current of the electrogenerated Pd(0) in the presence of PhBr at O
the electrogenerated bis(carbene)Pd(0) at O in the absence of PhBr, at a given scan rate v, determined as in Fig. 2a.
1
; iox: oxidation peak current of
1

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
171
Table 4
Reduction peak potentials ER1 of 2a, 2c, 6 and oxidation peak potentials EO1 of the electrogenerated Pd(0) complexes
p
p
a
p
R1 (V versus SCE)
b
p
O1 (V versus SCE) electrogenerated Pd(0)
b
k (Mꢃ1 sꢃ1) PhI k (Mꢃ1 sꢃ1) PhBr
Entry Pd(II) complex
E
E
c
1
2
3
4
Bis(carbene)PdCl
Bis(carbene)PdI
(carbene)(PPh )PdI
(PPh PdCl [58]
2
2a
ꢃ
/
2.03
1.58
1.20
0.91
ꢃ
/
0.49
0.48
0.15
0.03
n.d.
1180
830
ꢀ
1180
2
/
2
2c
ꢃ
/
ꢃ
/
ꢀ
/
3
2
6
ꢃ
/
ꢃ
/
d
3
)
2
2
ꢃ/
ꢃ/
540
ꢀ/
Rate constant k of the oxidative addition of aryl halides to the electrogenerated Pd(0) (DMF, 20 8C).
a
[
Pd(II)]ꢂ
/
4 4
3 mM in DMF containing n-Bu NBF (0.3 M).
b
c
Peak potentials were determined at a steady gold disk electrode (i.d. 1 mm) versus a saturated calomel electrode at a scan rate of 0.5 V sꢃ1.
Not determined because the reduction potential of complex 2a is close to that of PhI (ꢃ2.1 V). Consequently, it is not possible to reduce 2a
/
without reducing PhI.
d
In THF.
scan rate (in the range 0.2ꢀ
/
100 V sꢃ1), the three
complexes exhibited a single irreversible reduction
ne)Pd(0) is more electron rich and consequently less
incline to coordinate anionic halides. Since the carbene
ligand is more electron donor than PPh , the electro-
peak R with similar reduction peak current intensities
1
3
(
Fig. 2a, Table 4). An irreversible oxidation peak O was
1
generated Pd(0) complexes ligated by carbene(s) were
found to be more easily oxidized (Table 4).
Pd(0) complexes ligated by the carbene being more
observed on the reverse scan which characterized the
complexes generated by the reduction of complexes 2a,
2
c and 6 at R (Fig. 2a, Table 4).
1
electron rich than those ligated by PPh , they should be
3
The determination of the absolute number n of
electron(s) [57] involved in the electrochemical reduction
more reactive in the oxidative addition of aryl halides
(ArX). The rate constant k of the oxidative addition of
the electrogenerated Pd(0) complex (Scheme 6) was
determined by performing cyclic voltammetry at various
scan rates v, taking the opportunity that the oxidation
of complex 2c at R revealed a two-electron reduction
1
(
nꢂ
/
1.890.1) in the time scale of the cyclic voltamme-
/
try. Consequently the electrochemical reduction of
palladium(II) complexes 2a, 2c and 6 generates palla-
dium(0) complexes characterized by their oxidation
current of Pd(0) at O
tion [55,56]. At low scan rate (vꢂ
is proportional to its concentra-
ꢃ1
1
/
0.2 V s ), the Pd(0)
peaks O . From the reduction potentials of the palla-
1
complex electrogenerated by the reduction of complex
2c (3 mM) had totally disappeared in the presence of
PhBr (3 mM) due to the oxidative addition, as attested
dium(II) complexes investigated here, one observes that,
for a given halide, the introduction of carbene ligands
onto the Pd(II) complexes resulted in more negative
potentials (Table 4, entries 1,4 and 2,3). This shows that
by the absence of the oxidation peak O
scan. When the scan rate was progressively increased,
the oxidation peak O of Pd(0) was progressively
restored (Fig. 2a, vꢂ
10 V sꢃ1). The time allotted to
on the reverse
1
the carbene ligand is more electron donor than PPh and
3
1
consequently the Pd(II) ligated by one or two carbene
ligands is less easily reduced than complexes ligated by
PPh . For a given ligand, the Pd(II) complex ligated by
/
the Pd(0) for its reaction with PhBr was that which was
spent between the generation of Pd(0) by reduction of
3
two iodides is more easily reduced than that ligated by
two chlorides, as already observed for PPh ligands [56].
the Pd(II) complex at R
1
and its oxidation at O
(e.g. in
1
Fig. 2a, tꢂDE/v with DEꢂ
/
/
2.4 V). This time was
shorter and shorter when the scan rate was increased
3
Even if we are comparing peak potentials and not
standard potentials, the great similarity observed for the
oxidation peak potentials of the Pd(0) generated by the
reduction of 2a and 2c (Table 4, entries 1,2) suggests
that a similar complex was generated: a neutral bis(car-
bene)Pd(0) (Scheme 6) not ligated by the halide [59], in
and the oxidation peak current of Pd(0) at O progres-
1
ꢃ1
sively increased. At 100 V s , the oxidation peak
current of the Pd(0) was 92% of that obtained in the
absence of PhBr, corresponding to a conversion of only
8%. Since the oxidation current is proportional to the
Pd(0) concentration, the molar fraction x of the
contrast to what is reported for PPh with formation of
3
ꢃ
anionic (PPh ) Pd(0)Y (Yꢂ
/
I, Br, Cl) complexes [56].
This is probably due to the fact that the bis(carbe-
electrogenerated Pd(0) which had not reacted with
0
3
2
PhBr (for a given scan rate v) was: xꢂ
/i
ox/iox (iox
:
oxidation peak current of the electrogenerated Pd(0) in
0
the presence of PhBr, i : oxidation peak current of the
ox
electrogenerated Pd(0) in the absence of PhBr). The plot
of x versus log([PhBr]/v) is shown in Fig. 2b. This allows
the determination of the scan rate v1/2 corresponding to
the half reaction (Fig. 2b) and then t1/2
rate constant k of the oxidative addition is determined
ꢂDE/v1/2. The
/
Scheme 6. Electrochemical reduction of Pd(II) complexes.

172
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
by the kinetic law: 1/xꢂ
PhBr]ꢂ[Pd(0)]).
The rate constants of the oxidative addition of aryl
/
kC tꢁ
/
1 (C ꢂ
/
3 mMꢂ
/
0
0
[
/
halides with the Pd(0) complexes generated by reduction
of complexes 2c and 6 are collected in Table 4. One
observes that the bis(carbene)Pd(0) is the most reactive
complex and that the substitution of one and then two
carbenes by PPh results in less and less reactive Pd(0)
3
complexes.
As a conclusion, the Pd(II) complexes ligated to the
carbene(s) are less easily reduced than those ligated by
PPh . The Pd(0) complexes ligated to the carbene(s) are
3
more easily oxidized than those ligated by PPh and thus
3
more reactive in oxidative additions with aryl halides. It
is worthwhile to note that PhBr undergoes the oxidative
addition at room temperature.
Taking into account the high reactivity of the
bis(carbene)Pd(0) complex in oxidative addition, it is
of great interest to introduce a chemical reducing agent
in any catalytic reaction, so that to decrease any
induction period in which the Pd(II) complex is trans-
formed into the active Pd(0) complex and to increase the
concentration of the active Pd(0) complex.
2
Scheme 8. Reduction of 2c by iPr EtN in the presence of PhBr.
the formation of the complex 2d from 2c gives evidence
of the reduction into Pd(0).
To confirm that 2d was really the bromoꢀiodo
/
complex, the dibromo complex 2e was quantitatively
prepared by treatment of 2c with an excess of NaBr in
DMF at 120 8C and fully characterized. The reaction of
the dibromo complex with one equivalent of NaI at
20 8C gave a mixture of three complexes 2c, 2d and 2e.
The complex 2d was not separable from the mixture. By
2
.4.2. In situ reduction of palladium(II) complexes 2 and
to palladium(0) complexes and re-oxidation of the
6
palladium(0) complexes
To evaluate the ability of the base to reduce 2c in situ
during the Heck reaction, 0.02 mmol of the complex was
dissolved in DMF. After addition of iPr EtN (1 mmol)
2
1
and PhBr (1 mmol), the mixture was heated for 16 h at
1
tatively recovered as a 71/29 mixture of 2c and 2d
comparison with the H-NMR spectrum of the diiodo
complex 2c and of the dibromo complex 2e, it was
possible to determine the signals corresponding to the
20 8C. After work-up, the Pd(II) complex was quanti-
(
Scheme 7).
The generation of 2d that results from an halogen
bromoꢀiodo complex, which chemical shifts were iden-
/
tical to those of 2d, obtained by the experiment
described in Scheme 7. The detection of complexes 2d
and 2e was used as a qualitative test to determine the
ability of various agents to reduce the Pd(II) complexes
exchange on the palladium indicated that bromide ions
were generated during the reaction. This is only possible
if 2c is first reduced into Pd(0) by the base (Scheme 8,
steps a and b), followed by oxidative addition of the
Pd(0) with PhBr (step c). The reduction of the resulting
complex, by a second equivalent of base, generated one
(
Scheme 9). The results are presented in Table 5. First of
all, we observed that the reduction of 2c by AcONa at
20 8C in DMF, seems to be more efficient than the
1
ꢁ
ꢃ
equivalent of iPr EtN H, Br and a palladium hydride
2
reduction by iPr NEt (entries 1 and 2). Indeed, with
2
(
step d) that was recycled into Pd(0) by reductive
AcONa, a 24/50/26 mixture of 2c, 2d and 2e was
obtained (entry 2). No halogen exchange occurred in
the absence of base at 120 8C (Table 5, entry 3).
The mechanism of the reduction of the Pd(II) com-
elimination (step e). During the work-up, the Pd(0)
complex might be easily re-oxidized by oxygen in the
ꢁ
ꢃ
presence of iodides or bromides (R NH , I or
ꢁ
R NH , Br ) to give 2c or 2d (step f) [30]. Therefore,
3
ꢃ
3
plex by iPr NEt could occur by an hydride transfer
2
Scheme 7. Reduction of 2c by iPr EtN in the presence of PhBr.
2
Scheme 9. Reduction of 2c in the presence of PhBr.

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
173
Table 5
Reduction of complex 2c and oxidation of the resulting Pd(0) complex
a
a
a
Entry
Reducing agent (equivalent)
Aryl halide
T (8C)
2c (%)
2d (%)
2e (%)
1
2
3
4
5
iPr EtN (50)
2
PhBr
PhBr
PhBr
PhBr
PhBr
120
120
120
80
71
24
29
50
0
0
26
0
AcONa (50)
ꢀ/
100
91
iPr EtN (50)
2
9
0
AcONa (50)
80
88
11
1
6
7
8
9
HCO
HCO
HCO
HCO
HCO
HCO
2
2
2
2
2
2
Na (10)
Na (10)
Na (1)
PhI
PhBr
80
80
80
20
80
80
100
11
45
85
100
6
0
42
43
15
0
0
47
12
0
PhBr
PhBr
Na (10)
Na (47.5)
Na (47.5)
1
1
0
1
1-Iodonaphtalene
1-Bromonaphtalene
0
54
40
a
1
Ratio determined by H-NMR after work-up. No loss of Pd: 2c gave quantitatively a mixture of 2c, 2d and 2e.
the oxidative addition may also be affected by the
decrease of the temperature. These first observations
could explain to a great extent the results obtained
above in the MizorokiꢀHeck reaction. The Pd(II) pre-
/
catalyst must be reduced in situ by the base for the
coupling reaction to occur. This operation is effective at
2
Scheme 10. Reduction of 2c by iPr EtN.
120 8C where good conversions were obtained, but not
at 80 8C (low conversions).
from the amine to the palladium, followed by reductive
elimination (Scheme 10) [54,60]. It is more delicate in the
case of sodium acetate since the usual mechanism
Then, we tested the use of sodium formate in order to
determine under which conditions this additive was able
to reduce the Pd(II) pre-catalyst. We demonstrated, that
at 80 8C in DMF, this reduction was much more
efficient than with the bases. Indeed, the use of ten
proposed for PPh ligands, involved, after substitution
3
of an halogen by an acetate, the reductive elimination
between the acetate and the ligand of the Pd(II) (Scheme
equivalents of HCO Na gave a 11/42/47 mixture of 2c,
2
1
and since no modified form of the carbene ligand was
1) [61]. Since our catalyst 2c was recovered unchanged,
2d and 2e (Table 5, entry 7). A decrease in the quantities
of HCO Na to one equivalent, also led to an important
2
1
detected by H-NMR of the crude, the reduction of the
amount of transformed catalysts (entry 8). At 20 8C, a
non-negligible amount of 2d (15%) was formed after 16
h (entry 9), suggesting that the oxidative addition may
occur at room temperature.
carbene complex by NaOAc remains ambiguous.
After 16 h at 80 8C, the reaction from 2c afforded only
2% of the modified complexes 2d and 2e using NaOAc
1
as the reducing agent and 9% of 2d using iPr EtN (Table
2
The reduction of the aryl bromide showed in Scheme
5
, entries 4 and 5). At this temperature, the reduction of
the Pd(II) complex by both bases seemed to be less
8
was confirmed by performing the reaction with 1-
bromonaphtalene (1 mmol), in the presence of 0.95
mmol of sodium formate (Scheme 12). After 16 h at
efficient than at 120 8C. It is noteworthy that the rate of
8
2
0 8C with 2% mol of 2c, naphthalene was obtained in
6% yield together with 74% of starting material (ratio
1
determined by H-NMR). The catalyst was recovered as
a mixture of 2c, 2d and 2e in a 6/40/54 ratio (Table 5,
entry 11). Surprisingly, 1-iodonaphtalene was reduced
within the same yield in these conditions. The dehalo-
Scheme 12. Reduction of 1-bromo and 1-iodonaphthalene with 2c and
Scheme 11. Reduction of 2c by AcONa.
2
HCO Na.

174
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
genation of aryl halides mediated by Pd/imidazolinium
salt system was reported by Nolan and co-workers [63].
In all the reactions described above (Section 2.4.2), we
never noticed precipitation of Pd black and the Pd(II)
catalyst was always quantitatively recovered by precipi-
tation with methanol. We also never observed the
complex formed in the oxidative addition of Pd(0)
with the aryl halide. This is probably due to a fast
5). It is noteworthy that, in all reactions using PhI
(entries 1ꢀ6), no precipitation of Pd metal was observed
and the complex 2c could be recovered by precipitation
and recycled.
However, the coupling of phenyl bromide tested in
/
various conditions failed. The use of 10% of HCO Na in
2
the presence of iPr EtN or Cs CO as the base, led to no
2
2
3
conversion at 80 8C. Moreover, the catalyst was not
stable in these conditions since an important palladium
black deposit was observed after 16 h (entries 7 and 8).
The catalyst could be stabilized by performing the
reaction in the presence of NaI [67]. In this case, no
degradation was observed but still without conversion
(entry 9). From these observations, we can assume that
the efficiency of the oxidative addition of the active
Pd(0) complex to phenyl bromide may be dramatically
altered by the presence of the alkene. Indeed, in the
presence sodium formate but without alkene, we have
demonstrated that the Pd(0) species generated was able
to undergo the oxidative addition with PhBr, even at
80 8C (Table 5, entries 4 and 5). In the same way, at
reduction of the (NHC) PhPdBr species, the bis-carbene
2
Pd(0) complex obtained being probably rapidly oxidized
into 2c, 2d or 2e during the work-up, and thus not
observed (Scheme 8, steps c and f). However, oxidative
addition products were isolated recently by Cavell et al.,
from a bis-carbene Pd(0) complex [64ꢀ66].
/
In the case of phenyl chloride, the oxidative addition
occurred very slowly at 80 8C in DMF since only 11% of
complex 2g was detected after 16 h (Scheme 13).
Moreover, at this temperature, the carbene ligands in
the (NHC) Pd(0) species formed, was reduced by the
2
excess of sodium formate to generate the aminal 7 and
palladium black which precipitates.
120 8C in the absence of sodium formate, only traces of
coupled product were observed with PhBr and butyla-
crylate (Table 3, entry 8) although we have shown that
the base is able to reduce 2c and that the oxidative
addition occurred in these conditions without alkene
(Table 5, entry 1). Therefore, we presume that the lack
of reactivity with PhBr in the coupling reaction with
butylacrylate is probably due to a competitive com-
plexation of the bis-carbene Pd(0) species by the alkene,
that inhibits the oxidative addition [62] and allows the
degradation of the carbene ligand by the hydride. We
can not exclude either that the low reactivity with PhBr
comes from the difficulty for the alkene to carry out the
substitution of the bromide which is present on the
palladium after the oxidative addition had occurred.
For the same reasons, the coupling of PhCl using 2c,
even in the presence of sodium formate proved to be
inefficient. The use of only 1.8% mol of formate for 2%
mol of 2c did not prevent from the degradation of the
catalyst and precipitation of Pd metal was observed
(entries 11,12).
2
conditions
.5. Optimisation of the Mizorokiꢀ
/
Heck reaction
With all these observations in hand, we tested again
the activity of the carbene complexes in the Mizorokiꢀ
/
Heck reaction, in the presence of reducing agents to
activate the Pd(II) pre-catalyst. The results are presented
in Table 6. First, when the coupling of PhI and
butylacrylate, using 2c at 80 8C, was not possible
without reducing agent (Table 3, entries 6 and 7), the
addition of 10% mol of sodium formate led to a
complete conversion into (E)-butylcinnamate (Table 6,
entry 1). Moreover, the reaction was also effective at
4
0 8C leading to 52% of conversion after 16 h (entry 2).
Prolonging the reaction time to 48 h afforded quantita-
tively the Heck coupled product (entry 3). We have
shown by cyclic voltammetry (Section 2.4.1), that the
oxidative addition of the Pd(0) bis-carbene complex to
PhI was fast even at 20 8C. However, no conversion was
obtained at this temperature with 10% of HCO Na
The mixed complex 6 gave also good results in the
coupling of phenyl iodide and butylacrylate in the
presence of a reducing agent. The use of 10% mol of
2
(entry 4). The stability of the catalyst, generated from 2c
in the presence of HCO Na, was demonstrated by
performing the Heck reaction with phenyl iodide at
2
HCO Na afforded quantitatively (E)-butylcinnamate
2
ꢃ3
1
3
20 8C, using 10 % mol of 2c. A turn-over number of
1 000 was obtained after 96 h at this temperature (entry
after 16 h at 80 8C, without precipitation of Pd metal
(entry 13). This reaction could be performed at 40 8C
with a complete conversion after 48 h (entry 14).
Surprisingly, the use of 4% mol of PPh with 2% of
3
the mixed complex 6, but without HCO Na, led to a
2
total conversion after 16 h at 80 8C (entry 15). We have
shown previously that the coupling reaction of PhI and
butylacrylate did not occur at 80 8C using 6 without
excess of PPh (Table 3, entries 10 and 11). The exact
role of the phosphine is not obvious. It might be either
3
Scheme 13. Reduction of 2c by HCO
2
Na in the presence of PhCl.

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
175
Table 6
Heck reactions with complexes 2c, 6 and 5 in the presence of reducing and stabilizing agents
a
a
a
a
b
Entry Catalyst (% mol)
PhX
Additives (% mol)
Base
T (8C) t (h) Conv. (%)
Precipitation of Pd black
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2c 2%
2c 2%
2c 2%
2c 2%
2c 0.001%
2c 2%
2c 2%
2c 2%
2c 2%
2c 2%
2c 2%
2c 2%
6 2%
PhI
PhI
PhI
PhI
PhI
PhI
HCO
HCO
HCO
HCO
HCO
2
2
2
2
2
Na (10)
Na (10)
Na (10)
Na (10)
Na (10)
iPr
iPr
iPr
2
2
2
EtN
EtN
EtN
80
40
16
16
48
16
96
16
16
16
16
16
16
16
16
48
16
16
16
16
16
16
16
16
100
52
100
2
No
No
No
No
40
NEt
3
20
d
iPr
iPr
iPr
2
2
2
EtN
120
80
31
0
No
PPh
3
(4)
EtN
EtN
No
Yes
Yes
No
PhBr
PhBr
PhBr
PhBr
PhCl
PhCl
PhI
HCO
HCO
HCO
2
Na (10)
Na (10)
Na (10)
80
80
2
2
2
Cs
2
CO
3
0
c
iPr
iPr
iPr
2
2
2
EtN
80
0
0
1
2
3
4
5
6
7
8
9
0
1
2
PPh
3
(4)
EtN
EtN
120
120
80
0
0
No
HCO
HCO
HCO
HCO
2
Na (1.8)
Na (1.8)
Na (10)
Na (10)
Yes
Yes
No
2
2
2
AcONa
0
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
iPr
2
2
2
2
2
2
2
2
2
2
EtN
EtN
EtN
EtN
EtN
EtN
EtN
EtN
EtN
EtN
80
40
100
100
100
0
6 2%
6 2%
PhI
PhI
No
No
PPh
HCO
3
(4)
80
80
6 2%
5 1%
PhBr
PhI
2
Na (10)
(4)
Yes
No
PPh
PPh
PPh
PPh
3
3
3
3
80
100
70
5 1%
5 1%
PhBr
PhBr
PhBr
PhBr
PhBr
(4)
(4)
120
120
100
60
Yes
Tiny
Yes
Yes
Tiny
c
96
5 1%
5 1%
5 1%
(4)
3
0
HCO
HCO
2
Na (10)
Na (10)
c
2
80
3
a
b
c
1
mmol of PhX, 1.5 mmol of base, 1.2 mmol of butylacrylate, % mol of catalyst and additives based on the amount of PhX.
Dodecane as internal reference.
The reaction was performed with 1 mmol of NaI.
The precipitation of Pd metal was not detectable.
d
to stabilise the active catalytic species (NHCPd(0)PPh3),
or to generate a more active species by ligand exchange,
or to reduce 6 into a Pd(0) species. To confirm one of
these hypotheses, a solution of 0.02 mmol of 6 was
By comparison, the use of 4% PPh without sodium
3
formate was totally ineffective to activate 2c in the same
conditions: no Heck coupled product was obtained from
PhI and butylacrylate at 80 8C (entry 6). The use of 4%
stirred with 0.08 mmol of PPh and 1 mmol of PhBr for
3
PPh was also ineffective in the coupling of PhBr using
2c at 120 8C (entry 10).
The coupling of phenyl bromide, in the presence of
3
1
6 h at 80 8C in DMF. After usual work-up, the analysis
1
of the crude by H-NMR revealed the presence of 6 and
of a new complex, in a 1:1 ratio. The same experiment
performed with PhI instead of PhBr led exclusively to 6.
Noteworthy, in these two experiments, we never de-
tected the bis-carbene complex 2c, this is probably an
indication of the absence of ligand exchange. The new
complex obtained in the first reaction was separated
HCO Na, was not effective with 6. At 80 8C, we only
2
observed the degradation of the catalyst with formation
of Pd black (entry 16). As previously reported by
Herrmann et al., the mixed phosphineꢀ
6 could be generated in situ starting from its dimeric
precursor 5 and PPh at 20 8C [20]. We attempted then
/
carbene complex
3
1
from 6 by silica gel chromatography. The H-NMR
the Heck reaction, with 1% mol of 5 and an excess of
PPh (4% mol), in order to generate 6 and to activate
spectrum of this isolated complex was characteristic of a
non-symmetrical complex [68]. However, when the
signals of the diaminocarbene ligand were effectively
3
this latter with the excess of phosphine. In such
conditions, the coupling of PhI occurred with a com-
plete conversion at 80 8C (entry 17). Moreover, this
procedure allowed the coupling of PhBr at 120 8C. A
conversion of 70% was obtained after 16 h (entry 18).
Nevertheless, precipitation of Pd metal was still ob-
served. The active catalytic species was stabilized in the
presence of an excess of NaI [67]. In this case, still at
present, those of PPh were not visible and it seemed
3
that the phosphine has been replaced by one molecule of
water. According to this spectrum, we suggested that
this compound could be either the (NHC)PdI (H O)
2
2
complex with the two iodides in a cis position or the
NHC)PdIBr(H O) complex. According to Scheme 8,
only the second complex could come from the reduction
(
2
120 8C and with 4% mol of PPh , a conversion of 96%
3
of 6 by PPh , followed by oxidative addition of the Pd(0)
3
was reached with little or no degradation of the catalyst
(entry 19). The fact that this reaction failed using
sodium formate as the reducing agent, is probably due
to a competitive reduction of the carbene ligand by the
with PhBr. Since we were not able to obtain supple-
mentary analyses of this complex [69], we could not
conclude on the exact role of PPh3.

176
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
excess of formate that destroys the catalyst (entry 16).
No conversion was obtained with PhBr at temperatures
4. Experimental
inferior to 120 8C using the dimer 5 and PPh (entry 20).
All experiments were carried out under argon. Sol-
vents were of analytical grade type and used without
special drying or distillation. NMR spectra were re-
corded on a Brucker ARX 400 or AC 200 Q instrument,
in CDCl or DMSO-d as the solvent. Optical rotations
3
After 16 h at 100 8C, we only observed precipitation of
Pd metal. Likely, in these conditions, the oxidative
addition with PhBr does not occur. This could be
ascribed, once more, to the competitive complexation
of the active Pd(0) species by the alkene. Finally, the
dimeric Pd(II) complex 5 was inactive in the Heck
reaction in the absence of phosphine (entry 21). In this
reaction, the carbene complex was rapidly degraded by
the formate and precipitation of Pd metal was observed
after 15 min at 20 8C. The Pd(0) species generated
3
6
were measured on a PerkinꢀElmer 343.
/
4.1. Synthesis of (2a) and (2b)
A mixture of N,N?-dibenzylimidazolin-2-ylidene sil-
ver(I) bromide
1
PdCl (CH CN) (130 mg, 0.5 mmol) in CH Cl was
(550 mg, 1.0 mmol) and
directly from 5 and HCO Na could be stabilized by an
excess of NaI even though no conversion was obtained
after 16 h at 80 8C (entry 22) Scheme 14.
2
2
3
2
2
2
stirred for 2 h at 20 8C. Discoloration of the solution
and apparition of a white precipitate of silver salt was
observed. The crude mixture was filtered over celite and
concentrated to give a 2/1 mixture of palladium com-
plexes 2a and 2b (430 mg, 95% yield). The two
complexes were separated by chromatography over
silica gel (cyclohexane/AcOEt; 95:5).
3
. Conclusion
Chiral bis-diaminocarbene complexes of palladiu-
m(II) are readily accessible from the corresponding
4.1.1. Complex (2a)
silver(I) diaminocarbene complexes and PdCl (MeCN) .
2
2
2
53mg (56%) of a white solid were obtained after
2
Mixed carbeneꢀphosphine complexes are also easily
/
0
chromatography (R 0.45). M.p. dec. 248 8C. [a] ꢂ
/
ꢁ
/
prepared from Pd(OAc) and the corresponding chiral
f
D
2
5
0.1 (c 0.38, CHCl ). Anal. Calc. for C H N Cl Pd
imidazolinium salts. These two families of complexes are
air and moisture stable and can be purified by silica gel
chromatography. These Pd(II) complexes are not easily
reduced into the active Pd(0) species but the generated
Pd(0) complexes undergo facile re-oxidation. In the
3 50 68 4 2
(
M ꢂ
/
902.43): C, 66.55; H, 7.60; N, 6.21. Found: C,
6.28; H, 7.67; N, 5.90. Exact MS (FABꢁ): 830.4533 for
2Cl], 865.4219 for C H N PdCl
W
6
/
ꢁ
C H N Pd [M ꢀ
/
50 68
4
50 68
4
ꢁ
1
[
M ꢀ
/
Cl]. H-NMR (400 MHz) d 0.76 (s, 36H), 3.06 (s,
4
7
H), 4.68 (d, 4H, J 14.7 Hz), 6.58 (d, 4H, J 14.7 Hz),
3
classical conditions of the MizorokiꢀHeck reaction,
/
1
.23ꢀ7.34 (m, 12H), 7.75 (d, 8H, J 7.4 Hz). C-NMR
/
the pre-catalyst is probably reduced by the base but
only at high temperature and coupling products are
exclusively obtained with phenyl iodide in DMF at
(
1
50 MHz) d 27.8, 36.2, 54.8, 69.3, 128.0, 128.5, 130.7,
N).
36.0, 198.6 (NÃ
/
C Ã
/
120 8C. The in situ reduction of the Pd(II) complexes by
addition of sodium formate allows the coupling reaction
4.1.2. Complex (2b)
1
22mg (25%) of a white solid were obtained after
2
of PhI and butylacrylate at 40 8C instead of 120 8C. The
0
chromatography (R 0.5). M.p. dec. 259 8C. [a] ꢂ
68 (c 0.21, CH Cl ). Anal. Calc. for C H N BrClPd
/
ꢁ
/
mixed carbeneꢀ
/
phosphine Pd(II) complex is more easily
reduced than the bis-carbene complex and can be
f
D
1
2 2 50 68 4
(
M ꢂ946.88): C, 63.42; H, 7.24; N, 5.92. Found: C,
/
activated by PPh . This activation allows the coupling
3
W
1
3.72; H, 7.31; N, 5.62. H-NMR (400 MHz) d 0.76 (s,
8H), 0.77 (s, 18H), 3.06 (s, 4H), 4.63 (d, 2H, J 14.7 Hz),
6
1
4
7
of PhBr with butylacrylate at 120 8C. The bis-carbene
complex 2c is particularly interesting since it is quanti-
tatively regenerated by re-oxidation of the active Pd(0)
species during the work-up. In summary, these com-
.68 (d, 2H, J 14.7 Hz), 6.58 (bd, 4H, J 14.7 Hz), 7.22ꢀ
/
1
.32 (m, 12H), 7.75 (d, 8H, J 7.4 Hz). C-NMR (50
3
MHz) d 27.7, 36.0, 36.2, 54.6, 56.1, 69.0, 69.2, 127.8,
plexes are very efficient catalysts for the MizorokiꢀHeck
/
1
8
28.3, 130.7, 135.7, 135.8, 198.4 (NÃ
/
C Ã
/
N). MS (ESI):
Br], 911.41 for
reaction with aryl iodides but their activity is lower for
the coupling of aryl bromides. The use of these chiral
palladium carbene complexes in asymmetric coupling
reactions is under investigation.
ꢁ
65.42 for C H N PdCl [M ꢀ
/
50 68
ꢁ
4
C H N PdBr [M ꢀ
/
Cl].
50 68
4
4
4
.2. Synthesis of (2c)
.2.1. Method A
A mixture of 2a and 2b (220 mg, 0.24 mmol) and
sodium iodide (ten equivalent) in DMF (5 ml) was
stirred for 12 h at 120 8C. The solution was cooled to
ambient temperature, diluted with Et O (30 ml) and
2
Scheme 14. Heck reactions with complexes 2c, 6 and 5 in the presence
of reducing agents.

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
177
washed with H O (4ꢄ
/
3 ml). The organic layer was
of the mixture by removal of the known signals of 2c
and 2e.
H-NMR (400 MHz) d 0.81 (s, 18H), 0.85 (s, 18H),
3.09 (m, 4H), 4.58 (d, 2H, J 14.7 Hz), 4.67 (d, 2H, J 14.7
Hz), 6.40 (d, 2H, J 14.7 Hz), 6.43 (d, 2H, J 14.7 Hz),
2
dried over MgSO , filtered and concentrated to give
4
1
complex 2c as a pale yellow solid (260 mg, 100% yield).
Purification over silica gel (cyclohexane/AcOEt; 95: 5;
R 0.55) afforded analytically pure 2c. The complex was
f
crystallised by slow evaporation in a mixture of CH Cl
2
7.2ꢀ
.5. Synthesis of (5)
Pd(OAc) (112 mg, 0.5 mmol), imidazolinium bro-
mide 4 (222 mg, 0.5 mmol), potassium tert-butoxide (67
mg, 0.6 mmol) and sodium iodide (300 mg, 2 mmol)
were dissolved in dry THF (40 ml) and the solution was
stirred at 20 8C for 5 h. After filtration and evaporation
to dryness, purification by silica gel chromatography
/
7.34 (m, 12H), 7.77 (d, 8H, J 7 Hz).
2
and cyclohexane.
4
4
.2.2. Method B
To a solution of PdCl (CH CN) (72 mg, 0.278 mmol)
in CH Cl (5 ml) were added the silver carbene 1 (306
2
3
2
2
2
2
mg, 0.555 mmol) and NaI (2.78 mmol). The mixture was
stirred for 20 h at 20 8C then filtered over celite and
concentrated. The crude solid was washed with MeOH
and dried to afford 264 mg (87%) of 2c as a pale yellowꢀ
green solid.
M.p. 307 8C. [a] ꢂ
/
(cyclohexane/AcOEt; 1: 1; R 0.9) afforded 5 as a redꢀ
/
f
2
D
0
/
ꢁ
5
/
52 (c 0.3, CHCl ). Exact MS
brown powder (130 mg, 72% yield).
20
3
ꢁ
(
FABꢁ
/
): 830.4516 for C H N Pd [M ꢀ
/2l], 957.3570
M.p. dec.ꢀ
/
280 8C, [a] ꢂ
/
ꢁ218 (c 0.07, CH Cl ).
/
0
68
4
D
2 2
ꢁ
for C H N PdCl [M ꢀ
/
l]. Anal. Calc.: C, 55.33; H,
Anal. Calc. for C H I N Pd (M ꢂ1445.56): C,
/
5
0
68
4
50 64 4
4
2
W
1
.32; N, 5.16; Found: C, 55.72; H, 6.65; N, 5.09. H-
6
NMR (400 MHz) d 0.87 (s, 36H), 3.12 (s, 4H), 4.60 (d,
41.54; H, 4.74; N, 3.88; Found: C, 41.50; H, 4.81; N,
1
3.70. H-NMR (400 MHz) d 0.80 (s, 18H), 0.84 (s, 18H),
3.23 (s, 2H), 3.29 (s, 2H), 4.60 (m, 4H), 6.42 (d, 2H, J
4
1
2
H, J 14.8 Hz), 6.30 (d, 4H, J 14.8 Hz), 7.25ꢀ
/
7.34 (m,
2H), 7.77 (d, 8H, J 7.4 Hz). C-NMR (100 MHz) d
1
3
14.7 Hz), 6.63 (d, 2H, J 14.7 Hz), 7.30ꢀ
/
7.46 (m, 16H),
7.81 (m, 8H, Haro). C-NMR (100 MHz) d 27.8,
28.0, 36.6, 36.9, 56.5, 56.8, 69.6, 70.8, 128.4, 128.7,
1
3
8.5, 37.2, 56.1, 69.8, 128.2, 128.5, 131.4, 135.8, 198.2
7.74ꢀ
/
(
NÃ
/
C Ã/N).
1
30.9, 131.2, 134.1, 134.5, 186.8 (NÃ
/
C Ã/N).
4
.3. Synthesis of (2e)
4.6. Synthesis of (6)
A mixture of 2a and 2b (351 mg, 0.723 mmol) and
sodium bromide (ten equivalents) in DMF (5 ml) was
stirred for 12 h at 120 8C. The solution was cooled to
To a solution of complex 5 (85 mg, 0.058 mmol) in
CH Cl (2 ml) was added triphenylphosphine (22 mg,
2
2
ambient temperature, diluted with Et O (30 ml) and
0.116 mmol). The solution was stirred at 20 8C for 30
min and concentrated. The orange solid obtained (107
mg) was purified by silica gel chromatography (cyclo-
2
washed with H O (4ꢄ3 ml). The organic layer was
/
2
dried over MgSO , filtered and concentrated to give
complex 2e as a pale yellow solid (360 mg, 100% yield
from 0.723 mmol of 1). Purification over silica gel
4
hexane/AcOEt; 95: 5; R 0.3) to give 6 as an orange solid
f
(93 mg, 82% yield).
2
0
(
cyclohexane/AcOEt; 95:5) afforded analytically pure 2e
as a white solid.
Anal. Calc. for C H N Br Pd (M ꢂ
M.p. 98 8C. [a] ꢂ
/
ꢁ120 (c 0.08, CH Cl ). Anal.
/
D
2 2
Calc. for C H N I PPd (M ꢂ985.07): C, 52.43; H,
/
4
3
50 2 2
2
W
1
5.01; N, 2.84; Found: C, 51.98; H, 5.42; N, 2.43. H-
/
991.33): C,
5
0
68
4
2
W
6
5
4
7
0.58; H, 6.91; N, 5.65; Found: C, 60.61; H, 6.91; N,
1
.59. H-NMR (400 MHz) d 0.78 (s, 36H), 3.05 (s, 4H),
NMR (400 MHz, CDCl ) d 0.91 (s, 18H), 3.24 (s, 2H),
3
4.63 (d, 2H, J 14.8 Hz), 6.17 (d, 2H, J 14.8 Hz), 7.23ꢀ
/
1
7.58 (m, 10H). H-NMR (400
MHz, DMSO-d ) d 0.85 (s, 18H), 3.35 (s, 2H), 4.65
.64 (d, 4H, J 14.8 Hz), 6.48 (d, 4H, J 14.8 Hz), 7.22ꢀ
/
7.37 (m, 15H), 7.54ꢀ
/
1
.36 (m, 12H), 7.75 (d, 8H, J 7 Hz). C-NMR (100
3
6
MHz) d 28.1, 36.6, 55.4, 69.6, 128.1, 128.6, 131.1, 136.0,
1
(d, 2H, J 14.9 Hz), 5.88 (d, 2H, J 14.9 Hz), 7.35ꢀ7.40
/
3
1
98.8 (NÃ
/
C Ã/N).
(m, 15H), 7.54ꢀ
/
7.58 (m, 5H), 7.72ꢀ7.74 (m, 5H). P-
/
1
NMR (162 MHz, DMSO-d ) d 15.68. C-NMR (50
3
6
1
.4. Determination of the H-NMR chemical shifts of
4
MHz, DMSO-d ) d 28.5, 37.5, 56.4, 70.7, 128.7, 128.9,
6
(
2d). Synthesis of a reference
129.0, 131.0, 131.3, 132.9, 133.8, 135.4, 135.6, 135.7,
N).
1
89.2 (NÃ
/
C Ã
/
9
9mg (0.1 mmol) of 2e and 15 mg (0.1 mmol) of NaI
in CH Cl (2 ml) were stirred for 20 h at 20 8C. The
4.7. Typical procedure for the in situ reduction and
oxidative addition with palladium(II) complex 2c
2
2
mixture was diluted with Et O (10 ml) and washed with
2
water (3ꢄ2 ml). The organic layer was dried over
/
MgSO , filtered and concentrated to give 102 mg of a
To a solution of 2c (21.7 mg, 0.02 mmol) in DMF (1
ml) were added the aryl halide (1 mmol) and the
appropriate amount of reducing agent. The mixture
4
1
mixture of 2c, 2d and 2e in a 8/40/52 ratio. The H-
NMR reference of 2d was determined from the spectrum

178
J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ179
/
was degassed under vacuum then stirred under argon at
the indicated temperature. After cooling, the solution
disk electrode (i.d. 0.5 mm for step durations uꢂ
/
100
and 200 ms) and cyclic voltammetry at a steady gold
ꢃ1
was diluted with Et O (10 ml), washed with water (4ꢄ
/
2
ml), dried over MgSO , filtered and concentrated. The
disk electrode (i.d. 25 mm, vꢂ
/
0.02 V s ) as reported in
2
Ref. [57]. The resulting reduction currents were com-
pared to the oxidation currents obtained in both
techniques for ferrocene (0.3 mM in DMF) as an
internal standard (monoelectronic oxidation).
4
catalyst was recovered almost quantitatively by precipi-
tation with MeOH. After drying, the residue was
1
analyzed by H-NMR.
4
.8. Typical procedure for the MizorokiꢀHeck reaction
/
5
. Supplementary material
To a solution of Pd(II) complex (0.02 mmol) in DMF
1 ml) were added, the reducing agent (as indicated in
Table 3), the aryl halide (1 mmol), the base (1.5 mmol)
(
Crystallographic data for the structure reported in
this paper have been deposited with the Cambridge
Crystallographic Data Centre, CCDC No 200814 for
compound 2c. Copies of this information may be
obtained free of charge from the Director, CCDC, 12
and butylacrylate (173 ml, 1.2 mmol). Dodecane (230 ml,
1
The mixture was degassed under vacuum then stirred
mmol) can be added as an internal reference for GC.
under argon at the indicated temperature. After cooling,
the solution was diluted with Et O (10 ml) and washed
Union Road, Cambridge CB2 1EZ, UK (Fax: ꢁ
1223-336033; e-mail: deposit@ccdc.cam.ac.uk or www:
http://www.ccdc.cam.ac.uk).
/
44-
2
with water (4ꢄ2 ml). The solution was analyzed by GC.
/
After drying over MgSO , filtration and concentration,
4
the catalyst, in the case of 2c, can be separated from the
coupling product by precipitation with MeOH. The
crude coupling product was purified by silica gel
Acknowledgements
chromatography (cyclohexane/AcOEt; 95: 5).
(
1
E)-Butyl-cinnamate: H-NMR (400 MHz) d 0.99 (t,
J.P. thanks the MESR for a grant. We thank C.G-D.
Centre de resolution de structures-Laboratoire de
chimie inorganique et mat e´ riaux mol e´ culaires, universit e´
(
3
2
H, J 7.4 Hz), 1.45 (tq, 2H, J 6.7 and 7.4 Hz), 1.71 (tt,
H, J 6.7 Hz), 4.23 (t, 2H, J 6.7 Hz), 6.47 (d, 1H, J 16
P. et M. Curie) for X-ray structures.
Hz), 2H), 7.40ꢀ
/
7.41 (m, 3H), 7.52ꢀ7.54 (m, 2H), 7.70 (d,
/
1
H, J 16 Hz).
References
4
.9. Typical procedure for the electrochemical reduction
of complex 2c and for the determination of the rate
constant of the oxidative addition of aryl halides with the
electrogenerated Pd(0)
[
1] D. Bourrissou, O. Guerret, F.P. Gabba ¨ı , G. Bertrand, Chem.
Rev. 100 (2000) 39.
[2] W.A. Herrmann, C. K o¨ cher, Angew. Chem. Int. Ed. Engl. 36
(
[3] W.A. Herrmann, Angew. Chem. Int. Ed. Engl. 41 (2002) 1290.
1997) 2162.
Cyclic voltammetry was performed with a wave-form
generator PAR Model 175. The voltammograms were
recorded with a digital oscilloscope Nicolet 3091.
Experiments were carried out in a three-electrode cell
connected to a Schlenk line. The counter electrode was a
[
[
4] L. Jafarpour, S.P. Nolan, Adv. Organomet. Chem. 46 (2001) 181.
5] W.A. Herrmann, M. Elison, J. Fischer, C. K o¨ cher, G.R.J. Artus,
Angew. Chem. Int. Ed. Engl. 34 (1995) 2371.
[
[
6] M.B. Andrus, C. Song, Org. Lett. 3 (2001) 3761.
7] K. Selvakumar, A. Zapf, M. Beller, Org. Lett. 4 (2002) 3031.
2
platinum wire of ca. 1 cm apparent surface area; the
[8] V. Cal o` , R. Del Sole, A. Nacci, E. Schingaro, F. Scordari, Eur. J.
Org. Chem. (2000) 869.
reference was a saturated calomel electrode (Radiometer
Analytical) separated from the solution by a bridge (3
ml) filled with 2 ml of DMF containing n-Bu NBF (0.3
[
9] V. Cal o` , A. Nacci, L. Lopez, N. Mannarini, Tetrahedron Lett. 41
(
2000) 8973.
4
4
[
10] D.S. McGuinness, K.J. Cavell, Organometallics 19 (2000) 741.
11] A.M. Magill, D.S. McGuinness, K.J. Cavell, G.J.P. Britovsek,
V.C. Gibson, A.J.P. White, D.J. Williams, A.H. White, B.W.
M). The working electrode was a steady gold disk
electrode whose diameter was adapted to the scan rate
[
ꢃ1
Skelton, J. Organomet. Chem. 617ꢀ618 (2001) 546.
/
(
i.d. 1 mm for 0.2B
/
v B
/5 V s
and i.d. 0.5 mm for
ꢃ1
[12] S. Grundemann, M. Albrecht, J.A. Loch, J.W. Faller, R.H.
Crabtree, Organometallics 20 (2001) 5485.
1 v B100 V s ). In a solution in 8 ml of DMF
containing n-Bu NBF (0.3 M) was added 26 mg (0.024
0B
/
/
4
4
[
13] J.A. Loch, M. Albrecht, E. Peris, J. Mata, J.W. Faller, R.H.
Crabtree, Organometallics 21 (2002) 700.
mmol, 3 mM) of complex 2c. The cyclic voltammetry
was performed at different scan rates (0.2Bv B100 V
) in the absence of PhBr and then in the presence of
/
/
[14] A.A.D. Tulloch, A.A. Danopoulos, R.P. Tooze, S.M. Cafferkey,
S. Kleihenz, M.B. Hursthouse, Chem. Comm. (2000) 1247.
ꢃ1
s
[
15] A.A.D. Tulloch, A.A. Danopoulos, G.J. Tizzard, S.J. Colles,
M.B. Hursthouse, R.S. Hay-Motherwell, W.B. Motherwell,
Chem. Comm. (2001) 1270.
PhBr (2.5 ml, 3 mM) with the same range of scan rates.
The absolute number of electron n involved in the
reduction of complex 2c (0.3 mM in DMF) was
determined using chronoamperometry at a steady gold
[
16] D. Enders, H. Gielen, G. Raabe, J. Runsink, J.H. Teles, Chem.
Ber. 129 (1996) 1483.

J. Pytkowicz et al. / Journal of Organometallic Chemistry 678 (2003) 166ꢀ
/179
179
[
[
[
17] V. C e´ sar, S. Bellemin-Laponnaz, L.H. Gade, Organometallics 21
2002) 5204.
[47] M.C. Perry, X. Cui, K. Burgess, Tetrahedron: Asymmetry 13
(2002) 1969.
(
18] W.A. Herrmann, C.-P. Reisinger, M. Spiegler, J. Organomet.
Chem. 557 (1998) 93.
[48] D. Enders, H. Gielen, J. Organomet. Chem. 617ꢀ618 (2001) 70.
/
[49] W.A. Herrmann, L.J. Goossen, M. Spiegler, Organometallics 17
19] J. Schwarz, V.P.W. B o¨ hm, M.G. Gardiner, M. Grosche, W.A.
Herrmann, W. Hieringer, G. Raudaschl-Sieber, Chem. Eur. J. 6
(1998) 2162.
[50] J. Pytkowicz, S. Roland, P. Mangeney, J. Organomet. Chem. 631
(2001) 157.
(2000) 1773.
[
20] W.A. Herrmann, V.P.W. B o¨ hm, C.W.K. Gst o¨ ttmayr, M.
Grosche, C.-P. Reisinger, T. Weskamp, J. Organomet. Chem.
[51] H.M.J. Wang, I.J.B. Lin, Organometallics 17 (1998) 972.
[52] J. Pytkowicz, S. Roland, P. Mangeney, Tetrahedron: Asymmetry
6
17ꢀ618 (2001) 616.
/
1
2 (2001) 2087.
53] A. Alexakis, C.L. Winn, F. Guillen, J. Pytkowicz, S. Roland, P.
[
[
21] C. Yang, H.M. Lee, S.P. Nolan, Org. Lett. 3 (2001) 1511.
22] D.S. Clyne, J. Jin, E. Genest, J.C. Gallucci, T.V. RajanBabu, Org.
Lett. 2 (2000) 1125.
[
Mangeney, Adv. Synth. Catal. 345 (2003) 1.
[
54] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.
55] C. Amatore, M. Azzabi, A. Jutand, J. Organomet. Chem. 363
[
[
[
[
23] L. Xu, W. Chen, J. Xiao, Organometallics 19 (2000) 1123.
24] M.B. Andrus, C. Song, J. Zhang. Org. Lett. 4 (2002) 2079.
25] A. F u¨ rstner, A. Leitner, Synlett 2 (2001) 290.
26] C.W.K. Gst o¨ ttmayr, V.P.W. B o¨ hm, E. Herdtweck, M. Grosche,
W.A. Herrmann, Angew. Chem. Int. Ed. Engl. 41 (2002) 1363.
27] C. Zhang, J. Huang, M.L. Trudell, S.P. Nolan, J. Org. Chem. 64
[
(
1989) C41ꢀ
56] C. Amatore, M. Azzabi, A. Jutand, J. Am. Chem. Soc. 113 (1991)
375.
/C45.
[
[
8
57] For the determination of the absolute number of electron see: C.
Amatore, M. Azzabi, P. Calas, A. Jutand, C. Lefrou, Y. Rollin, J.
Electroanal. Chem. 288 (1990) 45.
[
[
(1999) 3804.
28] G.A. Grasa, M.S. Viciu, J. Huang, C. Zhang, M.L. Trudell, S.P.
Nolan, Organometallics 21 (2002) 2866.
[
[
58] A. Jutand, A. Mosleh, J. Org. Chem. 62 (1997) 261.
59] The slight difference in potentials arises from the oxidation
[
[
29] C.J. Mathews, P.J. Smith, T. Welton, Chem. Comm. (2000) 1249.
30] C.J. Mathews, P.J. Smith, T. Welton, A.J.P. White, D.J.
Williams, Organometallics 20 (2001) 3848.
I
process which generates a Pd complex, prompted to react with
II
the halide ions generated in the former reduction of the Pd
complexes.
[
[
[
31] H. Yoshida, S. Sugiura, A. Kunai, Org. Lett. 4 (2002) 2767.
32] S. Lee, J.F. Hartwig, J. Org. Chem. 66 (2001) 3402.
33] M.S. Viciu, R.F. Germaneau, S.P. Nolan, Org. Lett. 4 (2002)
[
60] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles
and Applications of Organotransition Metal Chemistry, Second
ed., University Science Book, Mill Valley, California, 1987, p.
4
053.
7
25.
61] C. Amatore, A. Jutand, Acc. Chem. Res. 33 (2000) 314.
[
34] M.S. Viciu, R.F. Germaneau, O. Navarro-Fernandez, E.D.
Stevens, S.P. Nolan, Organometallics 21 (2002) 5470.
[
[
62] C. Amatore, E. Carr e´ , A. Jutand, Y. Medjour, Organometallics
[
[
[
35] T.Y. Zhang, H. Zhang, Tetrahedron Lett. 43 (2002) 193.
36] R.A. Batey, M. Shen, A.J. Lough, Org. Lett. 4 (2002) 1411.
37] S. Caddick, F.G.N. Cloke, G.K.B. Clentsmith, P.B. Hitchcock,
D. McKerrecher, L.R. Titcomb, M.R.V. Williams, J. Organomet.
2
1 (2002) 4540.
63] M.S. Viciu, G.A. Grasa, S.P. Nolan, Organometallics 105 (2001)
297.
[
7
[
64] D.S. McGuinness, K.J. Cavell, Organometallics 18 (1999) 1596.
65] D.S. McGuinness, K.J. Cavell, B.F. Yates, B.W. Skelton, A.H.
White, J. Am. Chem. Soc. 123 (2001) 8317.
Chem. 617ꢀ618 (2001) 635.
38] H.M. Lee, S.P. Nolan, Org. Lett. 14 (2000) 2053.
39] G.A. Grasa, S.P. Nolan, Org. Lett. 3 (2001) 119.
/
[
[
[
[
[
[
66] S. Caddick, F.G.N. Cloke, P.B. Hitchcock, J. Leonard,
A.K.de.K. Lewis, D. McKerrecher, L.R. Titcomb, Organometal-
lics 21 (2002) 4318.
40] C. Yang, S.P. Nolan, Organometallics 21 (2002) 1020.
41] G.A. Grasa, M.S. Viciu, J. Huang, S.P. Nolan, J. Org. Chem. 66
(
2001) 7729.
42] M.S. Viciu, R.M. Kissling, E.D. Stevens, S.P. Nolan, Org. Lett. 4
2002) 2229.
[
[
67] K. Fagnou, M. Lautens, Angew. Chem. Int. Ed. 41 (2002) 26.
1
68] H-NMR (CDCl
[
3
) d 0.75 (s, 9H), 0.82 (s, 9H), 1.43 (s, 1H), 1.56
(
[
[
43] J. Cheng, M.L. Trudell, Org. Lett. 3 (2001) 1371.
44] S.R. Stauffer, S. Lee, J.P. Stambulli, S.I. Hauck, J.F. Hartwig,
Org. Lett. 2 (2000) 1423.
(s, 1H), 3.16 (d, 1H, J 2.8 Hz), 3.21 (d, 1H, J 2.8 Hz), 4.54 (d, 1H,
J 14.9 Hz), 4.56 (d, 1H, J 14.6 Hz), 6.47 (d, 1H, J 14.9 Hz), 6.69
(d, 1H, J 14.6 Hz), 7.25-7.40 (m, 6H), 7.67 (d, 2H, J 7 Hz), 7.73
(d, 2H, J 7 Hz).
[
[
45] Y. Sato, T. Yoshino, M. Mori, Org. Lett. 5 (2003) 31.
46] D.R. Jensen, M.S. Sigman, Org. Lett. 5 (2003) 63.
[69] Less than 4 mg of complex were obtained after purification.