Allylic alkylation and amination using mixed (NHC)(phosphine) palladium complexes under biphasic conditions

Article abstract of DOI:10.1016/j.jorganchem.2007.10.012

A dramatic improvement of the catalytic activity was observed when a phosphine was added in allylic alkylation reactions catalyzed by (NHC)Pd(η³-C₃H₅)Cl complexes. Consequently, several palladium complexes, generated in situ from different NHC-silver complexes, [Pd(η³-C₃H₅)Cl]₂ and PPh₃, were tested in this reaction to evaluate their potential. High reaction rates and conversions could be obtained with this catalytic system in the alkylation of allylic acetates with dimethylmalonate, particularly under biphasic conditions using water/dichloromethane and KOH 1 M as the base. These conditions are experimentally more convenient and gave higher reaction rates than the classical anhydrous conditions (NaH/THF). In this system, the phosphine is essential since no conversion was obtained when it is not present. The steric hindrance of the carbene ligand has a great influence on the activity and the stability of the catalytic system. The best NHC ligands for this reaction are either 1-mesityl-3-methyl-imidazol-2-ylidene or 1-(2,6-diisopropylphenyl)-3-methyl-imidazol-2-ylidene which are less bulky among the NHC tested. These two ligands led in 5 min to a complete conversion at 20 °C. The Pd-catalyzed allylic amination reaction using (E)-1,3-diphenylprop-3-en-yl acetate and benzylamine was also tested with (NHC)(PPh₃)Pd complexes and under the biphasic conditions. This reaction was found to be slower than the alkylation with dimethylmalonate but a complete conversion could be reached in 6 h at 20 °C using K₂CO₃ 1 M as the base. NMR experiments indicated that mixed (NHC)(PPh₃)Pd complexes are formed in situ but their structure could not be established exactly.

Full text of DOI:10.1016/j.jorganchem.2007.10.012

Available online at www.sciencedirect.com
Journal of Organometallic Chemistry 692 (2007) 5754–5762
www.elsevier.com/locate/jorganchem
Allylic alkylation and amination using mixed
(NHC)(phosphine) palladium complexes under biphasic conditions
*
Alexandre Flahaut, Sylvain Roland , Pierre Mangeney
Universite´ Pierre et Marie Curie, Laboratoire de Chimie Organique, UMR 7611, Institut de Chimie Mole´culaire (FR 2769), 4, place Jussieu,
tr. 44-45 2e`me e´t., 75252 Paris Cedex 05, France
Received 27 July 2007; received in revised form 5 October 2007; accepted 5 October 2007
Available online 12 October 2007
Abstract
A dramatic improvement of the catalytic activity was observed when a phosphine was added in allylic alkylation reactions catalyzed
by (NHC)Pd(g3-C3H5)Cl complexes. Consequently, several palladium complexes, generated in situ from different NHC–silver com-
plexes, [Pd(g3-C3H5)Cl]2 and PPh3, were tested in this reaction to evaluate their potential. High reaction rates and conversions could
be obtained with this catalytic system in the alkylation of allylic acetates with dimethylmalonate, particularly under biphasic conditions
using water/dichloromethane and KOH 1 M as the base. These conditions are experimentally more convenient and gave higher reaction
rates than the classical anhydrous conditions (NaH/THF). In this system, the phosphine is essential since no conversion was obtained
when it is not present. The steric hindrance of the carbene ligand has a great influence on the activity and the stability of the catalytic
system. The best NHC ligands for this reaction are either 1-mesityl-3-methyl-imidazol-2-ylidene or 1-(2,6-diisopropylphenyl)-3-methyl-
imidazol-2-ylidene which are less bulky among the NHC tested. These two ligands led in 5 min to a complete conversion at 20 ꢀC. The
Pd-catalyzed allylic amination reaction using (E)-1,3-diphenylprop-3-en-yl acetate and benzylamine was also tested with (NHC)(PPh3)Pd
complexes and under the biphasic conditions. This reaction was found to be slower than the alkylation with dimethylmalonate but a
complete conversion could be reached in 6 h at 20 ꢀC using K2CO3 1 M as the base. NMR experiments indicated that mixed
(NHC)(PPh3)Pd complexes are formed in situ but their structure could not be established exactly.
ꢁ 2007 Elsevier B.V. All rights reserved.
Keywords: N-heterocyclic carbenes; Allylic alkylation; (NHC)(phosphine)palladium complexes; Biphasic conditions
1. Introduction
by Mori and Sato in 2003 [2]. Several enantioselective ver-
sions with chiral NHC ligands have been reported since,
with enantiomeric excesses up to 92% [3]. A first example
reported in 2005 by Douthwaite et al., describes the use of
mixed NHC–P palladium complexes in this reaction [3b].
Chiral bidentate NHC–aminophosphine and NHC–phos-
phoramidate ligands were tested by the authors that noticed
that the presence of the phosphine moiety increases the rate
of the allylic substitution reaction compared to the related
bis(NHC) or NHC–imino ligands. More recently, the use
of chiral bidentate NHC–phosphine ligands in the Pd-cata-
lyzed allylic amination reaction was also described. How-
ever, low reaction rates were observed with these ligands
compared to isostructural P–N ligands [3f]. We have previ-
ously described the preparation and structure of palladium
complexes bearing several types of chiral NHC [1g,3e,4].
Mixed (NHC)(phosphine) palladium complexes (NHC =
N-heterocyclic carbene) have been used with success in CC
or CN coupling reactions [1]. Their superior activity, com-
pared to their bis(NHC) or bis(phosphine) analogues, has
been demonstrated particularly in the Mizoroki-Heck,
Suzuki-Miyaura or Stille reaction [1a,1b,1c,1g,1j]. However,
the behaviour of such palladium complexes involving an
NHC and a phosphine ligand has been very little studied
in the Tsuji-Trost reaction. The first example of allylic alkyl-
ation catalyzed by NHC–palladium complexes was reported
*
Corresponding author. Tel.: +33 1 44275567; fax: +33 1 44277567.
E-mail address: sroland@ccr.jussieu.fr (S. Roland).
0022-328X/$ - see front matter ꢁ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.10.012

A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
5755
Chelating NHC–amino ligands were tested in the palladium-
catalyzed asymmetric allylic alkylation reaction of (E)-1,3-
diphenylprop-3-en-yl acetate with dimethyl malonate [3e].
However, although enantioselectivities up to 80% could be
obtained, low reaction rates have also been observed with
these ligands in the Tsuji-Trost reaction. We herein describe
our study concerning the development of an efficient and
convenient catalytic system for palladium catalyzed allyla-
tion reactions using in situ generated (NHC)(PPh3)Pd com-
plexes and biphasic conditions.
more significant in biphasic conditions using KOH 1 M as
the base and CH2Cl2 as the solvent. No reaction occurred
in the absence of PPh3 whereas a complete conversion was
reached after 15 min in the presence of 7 mol% of this addi-
tive (entries 3 and 4). In both cases, the addition of PPh3
led to a dramatic decrease of the enantioselectivity. We
assumed that this effect could be due to the replacement
of the chiral chelating amino side-chain on the palladium
by PPh3. In this case, mixed (NHC)(PPh3)Pd complexes
must be generated in situ leading to more active species
than the starting NHC–amino complexes. Interestingly,
the best conditions are achieved under the experimentally
more convenient biphasic conditions (KOH 1 M, 7 mol%
PPh3, CH2Cl2), which afforded 2 quantitatively in only
15 min (entry 4). Similar results were obtained using the
well-defined and isolated (NHC)Pd(g3-C3H5)Cl complex
4 [4a] (Scheme 2). With NaH as the base and in the absence
of PPh3, no conversion was observed after 2 h whereas
100% conversion was obtained after 1.5 h with 6 mol% of
phosphine. A complete conversion was reached in 15 min
under biphasic conditions.
2. Results and discussion
We noticed recently that the addition of PPh3 to pre-
formed NHC–amino palladium complexes dramatically
improved the efficiency of the catalytic system in the allylic
alkylation reaction (Scheme 1). The influence of PPh3 was
studied in the reaction of (E)-1,3-diphenylprop-3-en-yl ace-
tate 1 with dimethyl malonate, in anhydrous conditions or
in biphasic conditions [5]. As previously reported [3e], the
palladium complexes were generated in situ from the corre-
sponding NHC–silver complexes and [Pd(g3-C3H5)Cl]2.
The results obtained with silver complex 3 [3e] are reported
in Table 1.
In the absence of phosphine, the reaction afforded 2 in
33% conversion after 48 h at 20 ꢀC in THF, using NaH
as the base and anhydrous conditions (Table 1, entry 1).
When the same reaction was performed by adding
7 mol% of PPh3, a complete conversion was obtained after
1 h at 20 ꢀC (entry 2). The effect of the phosphine was even
Although low enantioselectivities are obtained, these
preliminary experiments seemed to demonstrate that this
catalytic system, probably involving mixed (NHC)(PPh3)Pd
complexes, is much more efficient than those previously
reported with NHC ligands [2,3]. Consequently, we decided
to investigate the scope of this system by testing the behav-
iour of several achiral NHC ligands. Five new NHC–silver
complexes 6a–e having an aromatic substituent (mesityl or
2,6-diisopropylphenyl) on one nitrogen atom of the hetero-
cycle and an alkyl group (more or less bulky) on the second
nitrogen atom, were synthesized from the corresponding
imidazolium salts 5a–e and Ag2O in CH2Cl2 [6]. The two sil-
ver complexes 6f–g, already known [7], were also prepared
to be tested (Scheme 3). The new silver complexes 6a–e were
characterized by NMR and HRMS. The elemental analyses
obtained for these complexes were not in accordance with
those calculated for the (NHC)AgX formulation. A minor
[Pd(η
3-C3H5)Cl]2, 3 mol %
MeO2C
Ph
CO2Me
Ph
OAc
Ph
NHC—AgCl 3, 8.5 mol %
Additive (PPh3)
dimethyl malonate (3 equiv)
Solvent/base, 20 °C
Ph
1
2
1
N
N
species is detectable for some complexes by H NMR. By
1
Ph
comparison of elemental analyses and H NMR spectra,
Ag
N
H
we assumed that in addition to the major complex, 6–10%
of a minor complex having probably the (NHC)2Ag+/Clꢀ
form (for 6a, 6b and 6e) or (NHC)2Ag+/Brꢀ form (for 6c
and 6d) is present. The majority of NHC silver complexes
with monodentate NHC ligands are known to adopt one
Cl
Ph
3
Et
Scheme 1. Allylique alkylation using in situ generated NHC–amino–Pd
complexes and PPh3.
Table 1
4, 3 mol %
PPh3 6 mol %
Influence of PPh3 in the allylic alkylation reaction using NHC–amino–Pd
complexes
2
1
Entry
Solvent/base
Additive
t (h)
Conversiona (ee)
dimethyl malonate (2 equiv)
N
N
1
2
3
4
a
THF/NaH
None
PPhb3
None
PPhb3
48
1
16
0.25
33% (76%)
>98% (10%)
0%
Pd
Conditions A or B
THF/NaH
4
Cl
CH2Cl2/KOHc
CH2Cl2/KOHc
A: NaH (2 equiv), THF,20 C, 90 min, convn >98% (ee 8%)
°
>98% (6%)
B: KOH 1M (5 equiv), CH2Cl
2 ,20
C, 15 min, convn >98% (ee 5%)
°
Determined by 1H NMR of the crude.
PPh3 7 mol%.
b
c
Scheme 2. Allylic alkylation using well-defined (NHC)Pd(g3-C3H5)Cl
complex 4 and PPh3.
KOH 1 M in H2O, 2 equiv.

5756
A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
Table 2
Ag2O, CH2Cl2
X
N
N
Allylic alkylation using (NHC)(PPh3)Pd complexes: influence of the NHC
ligand
R
Ar
N
N
R
Ar
AgX
Entry
NHC–AgX
t1 (min)
Convna (%)
t2
Convna (%)
5a: Ar = Mesityl, R = Me, X = I
6a (94%)
6b (70%)
6c (94%)
6d (96%)
6e (72%)
1
2
3b
4d
5
6
7
8
9
6a
5
5
5
5
5
5
5
5
5
5
>98
16
0
86
>98
82
56
88
45
48
–
–
16
0c
>98
–
>98
58c
>98
55c
53c
5b: Ar = 2,6-Diisopropylphenyl, R = Me, X = I
5c: Ar = Mes, R = Bn, X = Br
None
6a
6a
1.5 h
1.5 h
1.5 h
–
30 min
1.5 h
20 min
1.5 h
1.5 h
5d: Ar = 2,6-Diisopropylphenyl, R = Bn, X = Br
5e: Ar = 2,6-Diisopropylphenyl, R =
i-Pr, X = I
6b
6c
6d
6e
N
N
N
N
N
N
Bn
AgI
6f
6g
AgI
6b
10
a
AgBr
6a
1
Determined by H NMR. Di-tert-butyl-4,40-biphenyl was used as
6c
internal reference.
b
N
N
The reaction was performed in the absence of PPh3.
The same conversion was observed after 16 h.
The reaction was performed with 2.5 mol of PPh3.
N
N
Bn
c
d
AgBr
AgI
6e
6d
the catalytic system for each ligand, samples were system-
atically taken at 5 min and, if the reaction is not complete,
at 1.5 and 16 h. These samples were analysed by H NMR
and the conversion determined by comparison with the
internal reference.
N
N
N
N
1
Ag
Cl
AgCl
6g
6f
High reaction rates and conversions could be obtained
with this catalytic system if the appropriate NHC ligand
is used. A complete conversion was reached after 5 min
using the less bulky silver complexes 6a and 6b bearing a
methyl group on one of the two nitrogen atoms of the het-
erocycle (entries 1 and 5). These two complexes are without
ambiguity the most efficient. A first control experiment per-
formed with 6 mol% of PPh3 in the absence of silver com-
plex demonstrated that the carbene ligand was essential to
get high reaction rates and conversions. In these condi-
tions, only 16% conversion was measured after 1.5 h (entry
2). No formation of 2 was detected in a second control
experiment carried out with 6 mol% of the silver complex
6a but without PPh3 (entry 3). Consequently, as previously
observed (see Table 1, Schemes 1 and 2), both the carbene
ligand and the phosphine play an important role in the
reaction. Interestingly, the catalytic system was still effi-
cient when less than 1 equiv. of PPh3 was used. With a ratio
PPh3/Pd of 0.5 the reaction was slower but complete after
1.5 h (entries 4 and 1). The steric hindrance of the carbene
ligand has an important influence on the reaction rate and
the stability of the catalytic species. Silver complexes 6a
and 6b, which lead to very efficient palladium catalysts,
bear NHC ligands which are less bulky than the other
ligands depicted in Scheme 3. The replacement of the
methyl group on the nitrogen atom (complexes 6a/6b) by
a benzyl group (complexes 6c and 6d) significantly
decreased the reaction rate: 30 min was necessary for the
reaction to be complete with 6c, instead of 5 min with 6a
(entries 6 and 1). The results obtained with complexes 6c
and 6d allowed to distinguish the influence of the aryl
group of the ligand (mesityl or 2,6-diisopropylphenyl).
Using complex 6d with a more bulky 2,6-diisopropylphenyl
group instead of a mesityl group (complex 6c), a maximum
Scheme 3. Synthesis and structures of NHC–silver complexes.
of the (NHC)AgX or (NHC)2Ag+/AgX2ꢀ forms but the less
frequent structure (NHC)2Ag+/Xꢀ was also reported [7,8].
It has also been shown that the reaction of imidazolium
iodides with Ag2O in CH2Cl2 could produce NHC silver
complexes with chloride anion instead of iodide [8c].
In a first set of experiments, we examined the influence
of the NHC ligand in the allylic alkylation reaction of 1
with dimethyl malonate (Scheme 4) in the presence of
PPh3. All reactions were carried out in biphasic conditions.
The palladium complexes were generated in situ from silver
complexes 6a–g. In a typical procedure, a mixture of
[Pd(g3-C3H5)Cl]2 (0.01 mmol, 2.5 mol%) and silver com-
plex (6 mol%) in CH2Cl2 (1 mL) was stirred for 1 h at
20 ꢀC under argon in the absence of light. PPh3 (5 mol%)
was then added and the solution was stirred for 1 h at
the same temperature. To the mixture was added a solution
of acetate 1 (0.4 mmol, 1 equiv.) in CH2Cl2 (1 mL),
dimethyl malonate (2 equiv.), di-tert-butyl-4,40-biphenyl
(internal reference) and KOH 1 M (2 equiv.). The mixture
was stirred vigorously at 20 ꢀC. The reactions were fol-
1
lowed by T.L.C. and H NMR. The results are depicted
in Table 2. To compare the stability and the efficiency of
[Pd(η
3-C3H5)Cl]2, 2.5 mol %
NHC–AgX 6a-g, 6 mol %
PPh3 5 mol %
2
1
dimethyl malonate (2 equiv)
KOH 1M (2 equiv) / CH2Cl2, 20 °C
Scheme 4. Allylic alkylation of 1 using (NHC)(phosphine)Pd complexes
under biphasic conditions.

A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
5757
Table 3
of 58% conversion was reached after 1.5 h. This conversion
Allylic alkylation of acetates 1, 7–9 using (NHC)(PPh3)Pd complexes
is almost identical to the one observed after 5 min (56%).
With this complex, the catalytic system seems to become
rapidly inactive. The silver complex 6e with an NHC ligand
bearing a 2,6-diisopropylphenyl group and an isopropyl
group leads to a more active palladium complex than its
benzyl analogue 6d (entries 8 and 7). However, a prolonged
reaction time (20 min) was necessary to go to completion
by comparison with its methyl analogue 6b (entries 8 and
5). Finally, the more bulky silver complexes 6f and 6g gave
moderate conversions after 16 h (entries 9 and 10) and no
significant progress of the reaction was detected between
5 min and 16 h. This first set of experiments clearly shows
that the more active palladium catalysts are generated from
the less bulky silver complexes 6a and 6b. An increase of
the steric hindrance leads in the best cases to a decrease
of the reaction rate (silver complexes 6c/6e) but can also
lead to a rapid deactivation of the catalytic system (silver
complexes 6d, 6f and 6g) [9]. However, in contrast to our
results, the bulky NHC ligand of complex 6g was found
by Mori and Sato as being efficient in their conditions.
They could obtain 2 in almost quantitative yield (98%)
after 2 h in anhydrous conditions (THF, 50 ꢀC), using a
palladium complex generated in situ from the imidazolium
salt, Pd2dba3 Æ CHCl3 and Cs2CO3 [2].
The scope and limitation of the catalytic system were
examined by testing the allylic alkylation reaction of ace-
tates 7–9 (Fig. 1) with dimethyl malonate. The results,
depicted in Table 3, demonstrate that the steric hindrance
due to the substituents of the allylic acetates, is as impor-
tant as the one of the NHC ligand. For example, using
the palladium catalyst generated from silver complex 6c,
the reaction is faster with the acetates 7 and 8 than with
the more bulky acetate 1 (entries 1–3). For similar reasons,
the most bulky silver complex 6g afforded the allylated
products in low conversions from the acetates 1 and 7
(entries 4 and 5) but gave a complete conversion after
5 min with the less bulky acetate 8 (entry 6). To summarize,
with acetates 1, 7 and 8, high conversions and reaction
rates can be achieved if the appropriate silver complex is
chosen to minimize the steric hindrance. The unsymmetri-
cally substituted acetate 7 afforded exclusively the linear
product 10 (entries 2 and 5). A limitation of this catalytic
system was found when the reaction was performed with
the cyclic acetate 9 which led to low conversions. The best
result (26% conversion) was obtained with the less bulky
silver complex 6a using KOH 50% as the base (entry 7).
Entry
Acetate
NHC–AgX
Product
t
Convna (%)
1
2
3
4
5
6
7e
a
1
7
8
1
7
8
9
6c
6c
6c
6g
6g
6g
6a
2
30 min
5 min
5 min
16 h
1.5 h
5 min
16 h
>98
>98
>98
53b
10
11
2
10
11
12
30b,c,d
>98
26b
Determined by 1H NMR of the crude. Di-tert-butyl-4,40-biphenyl was
used as internal reference.
b
The remaining proportion corresponds to the starting acetate.
The linear product 10 was the sole addition product detectable in the
c
crude.
d
The same conversion was observed after 16 h.
KOH 50% (0.8 mL) was used.
e
We next investigated the reaction of benzylamine with
acetate 1. It was first reported that the allylic substitution
using NHC–Pd complexes seems to be inapplicable to
nitrogen nucleophiles [2b]. Contrary to this observation,
it was demonstrated in 2007 that bidentate NHC–P ligands
led to active palladium catalysts in the allylic amination
reaction. However, low reaction rates were observed in
the anhydrous conditions that have been used (THF,
40 ꢀC) [3f]. In a preliminary experiment, the reaction was
tested in the conditions developed for the alkylation, using
silver complex 6a and KOH 1 M as the base. We could
obtain the expected product 13 in 66% yield after 16 h at
20 ꢀC (Scheme 5). The amination reaction is slower than
the alkylation with dimethylmalonate and 15% of the ether
14 was obtained as a by-product. The conditions were opti-
mized by replacing KOH by K2CO3 (1 M) and by changing
the concentration in acetate 1 from 0.2 M in CH2Cl2 to
0.4 M. A complete conversion was then obtained after
6 h at ambient temperature and the homoallylic amine 13
was isolated in 95% yield.
Finally, we attempted to determine the exact nature of
the palladium species generated in situ. The experiments
were performed with silver complex 6a. The reaction of
6a with 0.5 equiv. of [Pd(g3-C3H5)Cl]2 in CH2Cl2 at
20 ꢀC afforded quantitatively, after filtration of the silver
[Pd(η
3-C3H5)Cl]2, 2.5 mol %
Ph
Ph
Ph
Ph
Ph
O
NHC–AgI 6a, 6 mol %
PPh3, 5 mol %
+
1
benzylamine (2 equiv)
base 1M/H2O (2 equiv)
CH2Cl2, 20 °C
NHBn
13
Ph
14
Ph
OAc
OAc
OAc
7
8
Base
Products
13
9
1
14
CO2Me
CO2Me
CO2Me
CO2Me
68 % 16 %
>98 %
KOH
K2CO3
CO2Me
Ph
CO2Me
10
11
12
Scheme 5. Amination of 1 using (NHC)(phosphine)Pd complexes and
Fig. 1. Allylic acetates and allylation products.
biphasic conditions.

5758
A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
involved [15]. The phosphine could also accelerate the
N
N
[Pd(
η
3-C3H5)Cl]2 (0.5 equiv)
Mes
nucleophile addition step: it was suggested that the low
activity of NHC–Pd complexes in the Tsuji-Trost reaction
(in the absence of phosphine) could be due to the strong r-
donating properties of the NHC leading to poorly electro-
philic cationic p-allyl palladium complexes [3b,16]. In this
case, the p-acidity of the phosphine could promote the sec-
ond step of the reaction by reducing the electron density at
the Pd center [17]. In our conditions, we attempted to
replace PPh3 by the more p-acidic P(OPh)3 ligand. In this
case, the alkylation reaction carried out with silver complex
6a and acetate 1 led to only 4% conversion.
(+ AgI)
6a
Pd
CH2Cl2, 20 °C, 2h
Cl
Quantitative
15
Scheme 6. Synthesis of well-defined (NHC)Pd(g3-C3H5)Cl complex 15.
salts, the palladium complex 15 that was fully characterized
(Scheme 6). The NMR characteristics of this complex are
in accordance with those reported in the literature for sim-
ilar complexes with monodentate NHC ligands [3c,4a,10].
This complex is stable in solution in CH2Cl2 and in
CD2Cl2.
3. Conclusion
To study the nature of the species formed in the presence
of phosphine, 1 equiv. of PPh3 was added in an NMR tube
to a solution of complex 15 in CD2Cl2. After the addition,
1H and 31P NMR spectra were recorded at different times.
The 1H NMR spectrum recorded at 19 ꢀC after 5 min
showed the complete disappearance of the starting complex
15 but exhibited mainly very broad signals, difficult to ana-
We have developed an efficient and convenient catalytic
system for the allylic alkylation and amination reaction
using mixed (NHC)(phosphine) palladium complexes and
biphasic conditions. The superior activity of NHC–P com-
plexes compared to other NHC–palladium complexes
seems to be even more significant in this reaction than in
Mizoroki-Heck or Suzuki-Miyaura reactions. We could
demonstrate that under these biphasic conditions, the pres-
ence of a phosphine is absolutely necessary to get active
NHC–palladium species and that the steric interactions
between the NHC ligand and the allylic acetate must be
minimized to get high reaction rates and conversions. Fur-
ther studies are ongoing to determine precisely the role of
the phosphine and to expand the use of this catalytic
system.
lyse. A similar spectrum was obtained after 30 min. The 31
P
NMR spectra recorded at 5 min and 30 min always exhibit
a single signal at 24.8 ppm. This indicates that PPh3 reacts
rapidly with 15 to form a single phosphine–Pd species, sta-
ble in solution in CD2Cl2 at room temperature [11]. The 31
P
resonance observed is in accordance with those reported in
the literature for mixed (NHC)(phosphine)PdII complexes
[1a,1g,1j,12]. A slow equilibration rate between several
forms could account for the broad resonances observed
1
by H NMR. To our knowledge, (NHC)(phosphine)Pd–
allyl complexes have been only isolated and characterized
as their g3-allyl cationic form by using noncoordinating
counterions [3f,13]. With a coordinating counterion such
as chloride, the presence of g3 and g1-allyl species must
be considered [14]. In addition, different g1-allyl species
could also be in equilibrium by cis–trans isomerization
4. Experimental
4.1. General considerations
All experiments were performed under argon using
standard Schlenk techniques unless stated otherwise.
Solvents were dried over the appropriate drying agent
and distilled under dinitrogen. Sodium benzophenone ketyl
(THF), CaH2 (CH2Cl2). Reagents were purchased from
Acros or Aldrich and used as received unless otherwise
stated. 1-Mesitylimidazole, 1-[2,6-(i-Pr)2C6H3]-imidazole
[18], 1-mesityl-3-methyl-imidazolium iodide 5a [19],
1,3-bis(mesityl)-imidazolin-2-ylidene silver chloride 6f [7]
and 1,3-bis(2,6-diisopropylphenyl)-imidazol-2-ylidene sil-
ver chloride 6g [7] were prepared according to reported
procedures. 1H NMR (400 MHz) and 13C NMR
(100 MHz) spectra were recorded on a Brucker ARX-400
1
[12] (Scheme 7). H NMR experiments performed at 30
and 42 ꢀC did not provide more precise information even
if some signals slightly sharpen upon increase of the tem-
perature. These experiments and the fact that both the
phosphine and the NHC are essential in the reaction con-
firm that mixed (NHC)(PPh3)Pd complexes are formed
in situ although their structure could not be determined.
The exact role of the phosphine remains to be deter-
mined. Its r-donating effect in association with the strong
r-donating effect of the NHC could favor the oxidative
addition (ionization) step if an associative mechanism is
Cl
N
N
N
N
N
N
N
N
Mes
Mes
Cl Pd
Mes
Mes
Ph3P Pd
PPh3
15
Cl
Pd
Ph3P Pd
Ph3P
Cl
PPh3
η3
η1
η1
η1
Scheme 7. Possible palladium species formed in the presence of triphenylphosphine.

A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
5759
spectrometer, in CDCl3, CD2Cl2 or DMSO-d6 as the sol-
vent. IR spectra were recorded on a Brucker Tensor 27
(pike) instrument and only the structurally most important
peaks are listed. Melting points are uncorrected and were
measured on a Stuart Scientific apparatus SMP3. The
NMR data for products 2 [20], 10 [21], 11 [22], 12
[23,24], 13 [24] and 14 [25] were in accordance with those
reported in the literature.
mg, 1 mmol), dichloromethane (1 mL) and benzyl bromide
(238 lL, 2 mmol). The mixture was stirred at 30 ꢀC (oil
bath) for 16 h. After cooling, the imidazolium salt was pre-
cipitated by several washings of the residue with pentane
(2 · 15 mL) and ether (2 · 10 mL). After drying under vac-
uum, a pale yellow solid was obtained (396 mg, 99%).
M.p. = 114–115 ꢀC. 1H NMR (400 MHz, CDCl3):
d = 10.65 (s, 1H), 7.75 (t, 1H, J = 1.7 Hz), 7.65–7.62 (m,
2H), 7.55 (t, 1H, J = 7.8 Hz), 7.46–7.41 (m, 3H), 7.31 (d,
2H, J = 7.8 Hz), 7.13 (t, 1H, J = 1.7 Hz), 6.09 (s, 2H),
2.28 (m, 2H, J = 6.8 Hz), 1.24 (d, 6H, J = 6.8 Hz), 1.14
(d, 6H, J = 6.8 Hz). 13C NMR (100 MHz, CDCl3):
d = 145.3, 138.1, 133.8, 131.9, 131.2, 129.4, 129.1, 124.6,
124.1, 123.1, 53.4, 28.7, 24.4, 24.0. HRMS: m/z calcd for
C22H27BrN2 319.21688, found 319.21642. Anal. Calc. for
C22H27BrN2 (Mw = 399.37): C, 66.16; H, 6.81; N, 7.01.
Found: C, 66.02; H, 7.00; N, 6.92%.
4.2. 1-(2,6-Diisopropylphenyl)-3-methyl imidazolium iodide
(5b)
An oven-dried screw-cap tube flushed with argon
was charged with 1-(2,6-diisopropylphenyl)-imidazole
(114 mg, 0.5 mmol), dichloromethane (1 mL) and iodo-
methane (62 lL, 1 mmol). The mixture was stirred at
30–35 ꢀC (oil bath) for 16 h. After cooling, the imidazolium
salt was precipitated by several washings of the residue
with pentane (2 · 15 mL) and ether (2 · 10 mL). After dry-
ing under vacuum, a white solid was obtained (181 mg,
98%). M.p. 81–82 ꢀC. 1H NMR (400 MHz, CDCl3):
d = 9.92 (s, 1H), 8.17 (s, 1H), 7.46 (t, 1H, J = 7.8 Hz),
7.22 (d, 2H, J = 7.8 Hz), 7.18 (s, 1H), 4.32 (s, 3H), 2.22
(m, 1H, J = 6.8 Hz), 1.14 (d, 3H, J = 6.8 Hz), 1.06 (d,
3H, J = 6.8 Hz). 13C NMR (100 MHz, CDCl3):
d = 145.3, 137.3, 131.9, 129.8, 125.3, 124.6, 124.2, 37.9,
4.5. 1-(2,6-Diisopropylphenyl)-3-isopropyl imidazolium
iodide (5e)
An oven-dried screw-cap tube flushed with argon was
charged with 1-(2,6-diisopropylphenyl)imidazole (228 mg,
1 mmol) and 2-iodopropane (1 mL, 10 mmol). The mixture
was stirred under argon at 50 ꢀC (oil bath) for 50 h. After
cooling, the solution was diluted with CH2Cl2 (20 mL)
and washed with aqueous Na2S2O3 (2 · 5 mL) then with
water (2 · 10 mL). The organic layer was dried over
MgSO4, filtered and concentrated. After drying under vac-
uum, a pale yellow hygroscopic solid was obtained
28.5,
24.4,
24.3.
HRMS:
m/z
calcd
for
C16H23N2243.18558, found 243.18549. Anal. Calc. for
C16H23IN2 (Mw = 370.27): C, 51.90; H, 6.26; N, 7.57.
Found: C, 51.61; H, 6.33; N, 7.41%.
1
(350 mg, 60%). M.p. = 207–208 ꢀC. H NMR (400 MHz,
CDCl3)
d 10.16 (t, 1H, J = 1.5 Hz), 8.08 (t, 1H,
4.3. 1-Mesityl-3-benzyl-imidazolium bromide (5c)
J = 1.8 Hz), 7.55 (t, 1H, J = 8 Hz), 7.32 (d, 2H,
J = 8 Hz), 7.29 (s, 1H), 5.63 (m, 1H, J = 6.6 Hz), 2.28
(m, 2H, J = 6.8 Hz), 1.72 (d, 3H, J = 6.6 Hz), 1.25 (d,
3H, J = 6.8 Hz), 1.17 (d, 3H, J = 6.8 Hz). 13C NMR
(100 MHz, CDCl3) d 145.2, 136.5, 131.8, 130.0, 124.7,
124.6, 121.4, 53.7, 28.7, 24.4, 24.2, 23.4, 23.4. HRMS:
m/z calcd for C18H27N2 271.21688, found 271.21668. Anal.
Calc. for (C18H27IN2)2(H2O): C, 53.08; H, 6.93; N, 6.88.
Found: C, 53.03; H, 6.78; N, 6.91%.
An oven-dried screw-cap tube flushed with argon was
charged with 1-mesityl-imidazole (186 mg, 1 mmol),
dichloromethane (1 mL) and benzyl bromide (238 lL,
2 mmol). The mixture was stirred at 30–35 ꢀC (oil bath)
for 16 h. After cooling, the imidazolium salt was precipi-
tated by several washings of the residue with pentane
(2 · 15 mL) and ether (2 · 10 mL). After drying under vac-
uum,
a white solid was obtained (360 mg, 100%).
M.p. = 237–238 ꢀC. 1H NMR (400 MHz, CDCl3):
d = 10.69 (s, 1H), 7.65–7.55 (m, 3H), 7.45–7.40 (m, 3H),
7.12 (s, 1H), 7.01 (s, 2H), 6.01 (s, 2H), 2.35 (s, 3H), 2.09
(s, 6H). 3C NMR (100 MHz, CDCl3): d = 141.2, 137.7,
134.1, 133.6, 130.7, 129.8, 129.3, 129.1, 123.2, 123.0, 53.3,
21.0, 17.6. HRMS: m/z calcd for C19H21N2 277.16993,
found 217.16950. Anal. Calc. for C19H21BrN2
(Mw = 357.29): C, 63.87; H, 5.92; N, 7.84. Found: C,
63.47; H, 5.91; N, 7.74%.
4.6. Preparation of NHC–silver(I) complexes. General
procedure
To a solution of imidazolium salt (1 mmol) in dry
CH2Cl2 (15 mL) was added Ag2O (0.55 mmol). The mix-
ture was stirred at 20 ꢀC for 16 h with exclusion of light, fil-
tered through celite and concentrated under reduced
pressure. The solid was washed with pentane and dried
under vacuum.
4.4. 1-(2,6-Diisopropylphenyl)-3-benzyl imidazolium
bromide (5d)
4.7. 1-Mesityl-3-methyl-imidazol-2-ylidene silver iodide (6a)
567 mg (94%) of a beige solid was obtained starting
An oven-dried screw-cap tube flushed with argon was
charged with 1-(2,6-diisopropylphenyl)-imidazole (228
from 500 mg (1.5 mmol) of 5a. M.p. = 82–83 ꢀC. 1H

5760
A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
NMR (400 MHz, CDCl3): d = 7.27 (s, 1H), 6.90 (s, 2H),
6.88 (m, 1H), 3.91 (s, 3H), 2.33 (s, 3H), 1.86 (s, 6H). 13C
NMR (100 MHz, CDCl3): d = 183.6, 139.2, 135.3, 134.7,
129.2, 122.5, 122.4, 38.9, 21.0, 17.7. HRMS: [(NHC)2Ag]
[AgI2] m/z calcd for C26H32N4Ag (cation) 507.1672, found
507.16763. Anal. Calc. for (C13H16AgIN2)10(C26H32-
AgClN4): C, 38.28; H, 3.95; N, 6.87. Found: C, 38.34; H,
3.96; N, 6.76%.
4.11. 1-(2,6-Diisopropylphenyl)-3-isopropyl-imidazol-2-
ylidene silver iodide (6e)
202 mg (72%) of a pale beige solid was obtained start-
ing from 150 mg (0.554 mmol) of 5e. M.p. 88–89 ꢀC. H
NMR (400 MHz, DMSO-d6, the complex is little soluble
1
in CDCl3 and CD2Cl2).
A
mixture of isomers
is observed, major isomer (ꢁ90%): d = 7.79 (d, 1H,
J = 1.5 Hz), 7,69 (d, 1H, J = 1.5 Hz), 7.57 (t, 1H,
J = 7.8 Hz), 7.38 (d, 2H, J = 7.8 Hz), 3.87 (m, 1H),
2.25–2.10 (m, 2H), 1.30–0.95 (m, 18H). 13C NMR
(100 MHz, CDCl3): d = 145.9, 135.1, 130.4, 124.1, 123.7,
53.9, 28.2, 24.4, 24.3, 24.2, 24.0, the carbene carbon was
not detected. HRMS: [(NHC)2Ag] [AgI2] m/z calcd for
C36H52N4Ag (cation) 647.32374, found 647.32238. Anal.
Calc. for (C18H26AgIN2)12(C36H52AgClN4): C, 44.29; H,
5.39; N, 5.74. Calc. for (C18H26AgIN2)12(C36H52AgIN4):
C, 44.26; H, 5.37; N, 5.74. Found: C, 44.33; H, 5.23; N,
5.57%.
4.8. 1-(2,6-Diisopropylphenyl)-3-methyl-imidazol-2-ylidene
silver iodide (6b)
90 mg (70%) of a white solid was obtained starting from
100 mg (0.27 mmol) of 5b. M.p. 94–95 ꢀC. 1H NMR
(400 MHz, CDCl3): d = 7.44 (t, 3H, J = 15.6 Hz), 7.25 (s,
1H), 7.23 (s, 1H), 7.21 (s, 1H), 6.96 (d, 1H), 3.93 (s, 3H),
2.34 (m, 2H, J = 6.8 Hz), 1.14 (d, 6H, J = 6.8 Hz), 1.09
(d, 6H, J = 6.8 Hz). 13C NMR (100 MHz, CDCl3):
d = 145.7, 134.5, 130.4, 124.1, 123.8, 38.9, 28.1, 24.6,
24.5, 24.3, the carbene carbon was not detected. HRMS:
[(NHC)2Ag] [AgI2] m/z calcd for C32H44N4Ag (cation)
4.12. Typical procedure for allylic alkylation reactions in
biphasic conditions
591.26114,
found
591.26137.
Anal.
Calc.
for
(C16H22AgIN2)9(C32H44AgClN4): C, 42.95; H, 4.96; N,
6.26. Found: C, 43.04; H, 4.96; N, 6.03%.
A mixture of [Pd(g3-C3H5)Cl]2 (3.6 mg, 0.01 mmol,
2.5 mol%) and silver complex (0.024 mmol, 6 mol%) in
CH2Cl2 (1 mL) was stirred for 1 h at 20 ꢀC under argon
in the absence of light. PPh3 (5.2 mg, 0.02 mmol, 5 mol%)
was then added and the solution was stirred for 1 h at
the same temperature. To the mixture was added a solution
4.9. 1-Mesityl-3-benzyl-imidazol-2-ylidene silver bromide
(6c)
345 mg (94%) of a beige solid was obtained starting
from 297 mg (0.79 mmol) of 5c. M.p. 64–65 ꢀC. 1H
NMR (400 MHz, CDCl3): d = 7.36 (m, 3H), 7.25 (m,
2H), 7.12 (s, 1H), 6.93 (m, 3H), 5.39 (s, 2H), 2.32 (s,
3H), 1.97 (s, 6H). 13C NMR (100 MHz, CDCl3):
d = 139.8, 135.5, 135.2, 134.6, 129.4, 129.2, 128.7, 127.6,
123.1, 121.0, 55.7, 21.0, 17.7, the carbene carbon was not
detected. HRMS: [(NHC)2Ag] [AgBr2] m/z calcd for
C38H40N4Ag (cation) 659.2298, found 659.2300. Anal.
Calc. for (C19H20AgBrN2)15(C38H40AgBrN4): C, 50.36;
H, 4.45; N, 6.18. Found: C, 50.53; H, 4.52; N, 6.14%.
of (E)-1,3-diphenylprop-3-en-yl acetate
1
(100 mg,
0.4 mmol, 1 equiv.) in CH2Cl2 (1 mL), dimethyl malonate
(91 lL, 0.8 mmol, 2 equiv.), di-tert-butyl-4,40-biphenyl
(34.8 mg, 0.33 equiv., 0.13 mmol, internal reference) and
KOH 1 M in H2O (0.8 mL, 2 equiv.). The mixture was stir-
red vigorously for the indicated time at 20 ꢀC. The mixture
was diluted with Et2O and the organic layer was separated.
The residue was extracted with Et2O. The combined
organic layers were filtered through a short pad of silicagel,
dried over Na2SO4 and concentrated. The conversion was
determined by 1H NMR of the residue by comparison with
the internal reference.
4.10. 1-(2,6-Diisopropylphenyl)-3-benzyl-imidazol-2-ylidene
silver bromide (6d)
4.13. Allylic amination reaction with benzylamine in biphasic
conditions
415 mg (96%) of an off-white solid was obtained start-
ing from 340 mg (0.85 mmol) of 5d. M.p. 78–79 ꢀC. H
1
A mixture of [Pd(g3-C3H5)Cl]2 (3.6 mg, 0.01 mmol,
2.5 mol%) and silver complex 6a (10.5 mg, 0.024 mmol,
6 mol%) in CH2Cl2 (0.5 mL) was stirred for 1 h at 20 ꢀC
under argon in the absence of light. PPh3 (5.2 mg,
0.02 mmol, 5 mol%) was then added and the solution was
stirred for 1 h at the same temperature. To the mixture
was added a solution of (E)-1,3-diphenylprop-3-en-yl ace-
tate 1 (100 mg, 0.4 mmol, 1 equiv.) in CH2Cl2 (0.5 mL),
benzylamine (87 lL, 0.8 mmol, 2 equiv.), di-tert-butyl-
4,40-biphenyl (34.8 mg, 0.33 equiv., 0.13 mmol, internal ref-
erence) and K2CO3 1 M in H2O (0.8 mL, 2 equiv.). The
mixture was stirred vigorously for 6 h at 20 ꢀC. The mix-
NMR (400 MHz, CDCl3): d = 7.51–7.27 (m, 8H), 7.13
(d, 1H, J = 1.5 Hz), 7.02 (d, 1H, J = 1.5 Hz), 5.46 (s,
2H), 2.41 (m, 2H, J = 6.8 Hz), 1.27 (d, 6H, J = 6.8 Hz),
1.14 (d, 6H, J = 6.8 Hz). 13C NMR (100 MHz, CD2Cl2)
d = 146.2, 136.2, 135.2, 130.8, 129.5, 128.9, 127.8 (broad),
124.9, 124.5, 121.8 (broad), 55.7 (broad), 28.7, 24.7, 24.3,
the carbene carbon was not detected. HRMS: [(NHC)2Ag]
[AgBr2] m/z calcd for C44H52N4Ag (cation) 743.32374,
found
743.32197.
Anal.
Calc.
for
(C22H26-
AgBrN2)13(C44H52AgBrN4): C, 53.52; H, 5.31; N, 5.67.
Found: C, 53.66; H, 5.38; N, 5.61%.

A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
5761
(l) J. Zong, J.-H. Xie, A.-E. Wang, W. Zhang, Q.-L. Zhou, Synlett
(2006) 1193.
[2] (a) Y. Sato, M. Mori, Org. Lett. 5 (2003) 31;
(b) Y. Sato, T. Yoshino, M. Mori, J. Organomet. Chem. 690 (2005)
5753.
[3] (a) L.G. Bonnet, R.E. Douthwaite, Organometallics 22 (2003)
ture was diluted with Et2O, solid NaCl was added and the
organic layer was separated. The residue was extracted
with Et2O. The combined organic layers were dried over
K2CO3, filtered through a short pad of silicagel and con-
1
centrated. A complete conversion was determined by H
4187;
NMR of the crude by comparison with the internal refer-
ence. The residue was purified by column chromatography
(SiO2, pentane/Et2O;9/1) to afford 114 mg (95%) of the
expected product 13 as a colorless oil.
(b) R. Hodgson, R.E. Douthwaite, J. Organomet. Chem. 690
(2005) 5822;
(c) S.-J. Li, J.-H. Zhong, Y.-G. Wang, Tetrahedron: Asymmetry
17 (2006) 1650;
´
(d) A. Ros, D. Monge, M. Alcarazo, E. Alvarez, J.M.
´
Lassaletta, R. Fernandez, Organometallics 25 (2006) 6039;
4.14. Synthesis of (NHC)Pd(allyl)Cl complex (15)
(e) A. Flahaut, S. Roland, P. Mangeney, Tetrahedron: Asym-
metry 18 (2007) 229;
A mixture of silver complex 6a (100 mg, 0.23 mmol) and
[Pd(g3-C3H5)Cl]2 (42 mg, 0.115 mmol) in degassed CH2Cl2
(5 mL) was stirred for 2 h at 20 ꢀC. The silver salts were
removed by filtration through a short pad of Celite. Evap-
oration of the CH2Cl2 and drying under vacuum affor-
ded quantitatively 92 mg of an off-white solid. M.p. =
179–180 ꢀC. 1H NMR (400 MHz, CDCl3) d = 7.10 (d,
1H, J = 1.8 Hz), 6.97 (s, 1H), 6.94 (s, 1H), 6.91 (d, 1H,
J = 1.8 Hz), 5.03 (m, 1H), 4.11 (dd, 1H, J = 7.4 and
2 Hz), 4.10 (s, 3H), 3.15 (d, 1H, J = 6.6 Hz), 3.03 (d,
1H, J = 13.4 Hz), 2.34 (s, 3H), 2.22 (s, 3H), 2.06 (s, 3H).
13C NMR (100 MHz, CD2Cl2): d = 181.3 (Ccarbene),
139.7, 136.4, 135.7, 135.3, 128.9, 128.6, 122.6, 122.2,
114.5 (CHallyl), 71.1 (CH2), 48.6 (CH2), 38.1 (N–CH3),
20.8 (CH3, Mes), 18.1 (CH3, Mes), 17.3 (CH3, Mes). Anal.
Calc. for C16H21ClN2Pd (Mw = 383.22): C, 50.15; H, 5.52;
N, 7.31. Found: C, 49.91; H, 5.39; N, 7.28%.
(f) F. Visentin, A. Togni, Organometallics 26 (2007) 3746.
[4] (a) S. Roland, M. Audouin, P. Mangeney, Organometallics 23 (2004)
3075;
(b) A. Flahaut, S. Roland, P. Mangeney, J. Organomet. Chem. 691
(2006) 3498.
[5] For Pd-catalyzed allylic alkylation reactions in aqueous or biphasic
conditions, see: (a) T. Hashizume, K. Yonehara, K. Ohe, S. Uemura,
J. Org. Chem. 65 (2000) 5197;
(b) C. Rabeyrin, C. Nguefack, D. Sinou, Tetrahedron Lett. 41 (2000)
7461;
(c) M. Nakoji, T. Kanayama, T. Okino, Y.J. Takemoto, Org. Lett. 3
(2001) 3329;
(d) Y. Uozumi, K. Shibatomi, J. Am. Chem. Soc. 123 (2001) 2919;
(e) G. Chen, Y. Deng, L. Gong, A. Mi, X. Cui, Y. Jiang, M.C.K.
Choi, A.S.C. Chan, Tetrahedron: Asymmetry 12 (2001) 1567;
(f) M. Nakoji, T. Kanayama, T. Okino, Y. Takemoto, J. Org. Chem.
67 (2002) 7418;
(g) K. Manabe, S. Kobayashi, Org. Lett. 5 (2003) 3241;
(h) H. Kinoshita, H. Shinokubo, K. Oshima, Org. Lett. 6 (2004)
4085;
(i) Y. Nakai, Y. Uozumi, Org. Lett. 7 (2005) 291;
(j) D. Madec, G. Prestat, E. Martini, P. Fristrup, G. Poli, P.-O.
Norrby, Org. Lett. 7 (2005) 995.
Acknowledgments
[6] H.M. Wang, I.J.B. Lin, Organometallics 17 (1998) 972.
CNRS and UPMC are acknowledged for the financial
support. A. Flahaut thanks Clariant France for a grant.
We thank E. Caytan for the NMR experiments with
triphenylphosphine.
´
[7] P. De Fremont, N.M. Scott, E.D. Stevens, T. Ramnial, O.C.
Lightbody, C.L.B. Macdonald, J.A.C. Clyburne, C.D. Abernethy,
S.P. Nolan, Organometallics 24 (2005) 6301.
[8] (a) C.P. Newman, G.J. Clarkson, J.P. Rourke, J. Organomet. Chem.
692 (2007) 4962;
(b) J.C. Garrison, W.J. Youngs, Chem. Rev. 105 (2005) 3978;
(c) I.J.B. Lin, C. Sekhar Vasam, Coord. Chem. Rev. 251 (2007)
642.
References
[9] A possible explanation for these results would be that the increase of
the steric hindrance could decrease the rate of the ionization step
(oxidative addition) and favor side-reactions that lead to inactive
species.
[10] (a) D.R. Jensen, M.S. Sigman, Org. Lett. 5 (2003) 63;
(b) M.S. Viciu, O. Navarro, R.F. Germaneau, R.A. Kelly III, W.
Sommer, N. Marion, E.D. Stevens, L. Cavallo, S.P. Nolan, Organo-
metallics 23 (2004) 1629.
[1] (a) W.A. Herrmann, V.P.W. Bo¨hm, C.W.K. Gsto¨ttmayr, M.
Grosche, C.-P. Reisinger, T. Weskamp, J. Organomet. Chem. 617–
618 (2001) 616;
(b) C. Yang, H.M. Lee, S.P. Nolan, Org. Lett. 3 (2001) 1511;
(c) L.R. Titcomb, S. Caddick, F.G.N. Cloke, D.J. Wilson, D.
McKerrecher, Chem. Commun. (2001) 1388;
(d) C.W.K. Gsto¨ttmayr, V.P.W. Bo¨hm, E. Herdtweck, M. Grosche,
W.A. Herrmann, Angew. Chem. Int. Ed. 41 (2002) 1363;
[11] In the procedure used for the Tsuji-Trost reactions, the silver salts are
not removed from the reaction media. When one equivalent of PPh3 is
added to the in situ formed complex 15, the precipitate of AgI
immediately disappears then rapidly reappears after 5 min stirring at
20 ꢀC. It seems that a soluble PPh3–AgI complex is formed first
followed by the transfer of PPh3 from silver to palladium with again
precipitation of AgI. Since the same results have been obtained using
isolated (NHC)Pd(g3-C3H5)Cl complexes or complexes generated
in situ from silver complexes and [Pd(g3-C3H5)Cl]2, it is likely that the
same palladium species are formed in the presence or in the absence of
silver salts.
(e) A. Furstner, G. Seidel, D. Kremzov, C.W. Lehmann, Organome-
¨
tallics 22 (2003) 907;
(f) N. Tsoureas, A.A. Danopoulos, A.A.D. Tulloch, M.E. Light,
Organometallics 22 (2003) 4750;
(g) J. Pytkowicz, S. Roland, P. Mangeney, G. Meyer, A. Jutand, J.
Organomet. Chem. 678 (2003) 166;
(h) H.M. Lee, J.Y. Zeng, C.-H. Hu, M.-T. Lee, Inorg. Chem. 43
(2004) 6822;
(i) A.-E. Wang, J.-H. Xie, L.-X. Wang, Q.-L. Zhou, Tetrahedron 61
(2005) 259;
(j) H. Turkmen, B. C¸ etinkaya, J. Organomet. Chem. 691 (2006) 3749;
¨
[12] H.V. Huynh, Y. Han, J.H.H. Ho, G.K. Tan, Organometallics 25
(2006) 3267.
(k) F.E. Hahn, M.C. Jahnke, T. Pape, Organometallics 25 (2006)
5927;

5762
A. Flahaut et al. / Journal of Organometallic Chemistry 692 (2007) 5754–5762
[13] N.T. Barczak, R.E. Grote, E.R. Jarvo, Organometallics 26 (2007)
4863.
[17] This effect must enhance the electrophilicity of the p-allylic interme-
diates and favor the regeneration of Pd0.
[18] J. Liu, J. Chen, J. Zhao, Y. Zhao, L. Li, H. Zhang, Synthesis (2003)
2661.
[19] C. Nieto-Oberhuber, C. Lopez, A.M. Echavarren, J. Am. Chem. Soc.
127 (2005) 6178.
[20] M. Widhalm, P. Wimmer, G. Klintschar, J. Organomet. Chem. 523
(1996) 167.
[14] Equilibrium between g3 and g1-allyl is known to occur in Pd–allyl
complexes. For example, see: (a) P.S. Pregosin, R. Salzmann, Coord.
Chem. Rev. 155 (1996) 35;
(b) P. Braunstein, F. Naud, A. Dedieu, M.-M. Rohmer, A. DeCian,
S.J. Rettig, Organometallics 20 (2001) 2966.
[15] We recently demonstrated, that (NHC)(PPh3)Pd0 complexes with not
bulky and strong r-donor NHC ligands react in oxidative additions
of aryl halides in associative mechanism. The intervention in the
oxidative addition of active species coming from the dissociation of
the phosphine or of the NHC was ruled out by electrochemistry
experiments: S. Roland, P. Mangeney, A. Jutand, Synlett (2006)
3088.
ˇ
´ˇ
ˇ
´
[21] S. Vyskocˇil, M. Smrcˇina, V. Hanusˇ, M. Polasek, P. Kocovsky, J.
Org. Chem. 63 (1998) 7738.
[22] S. Ma, B. Xu, B. Ni, J. Org. Chem. 65 (2000) 8532.
´
[23] J. Franzen, J.-E. Ba¨ckvall, J. Am. Chem. Soc. 125 (2003) 6056.
[24] D. Zhao, J. Sun, K. Ding, Chem. Eur. J. 10 (2004) 5952.
[25] P. Evans, P. Jonhson, R.J.K. Taylor, Eur. J. Org. Chem. (2006)
1740.
[16] R.E. Douthwaite, Coord. Chem. Rev. 251 (2007) 702.