Exploiting noninnocent (E,E)-dibenzylideneacetone (dba) effects in palladium(0)-mediated cross-coupling reactions: Modulation of the electronic properties of dba affects catalyst activity and stability in ligand and ligand-free reaction systems

Article abstract of DOI:10.1002/chem.200600473

The reactivity of palladium(0) complexes, [Pd₂ ⁰(dba-n,n′-Z)₃] (n,n′-Z = 4,4′-F; 4,4′-CF₃; 4,4′-H; 4,4′-MeO) and [Pd ⁰(dba-n,n′-Z)₂] (n,n′-Z = 4,4′-CF ₃; 4,4′-H; 3,3′,5,5′-OMe),

Full text of DOI:10.1002/chem.200600473

DOI: 10.1002/chem.200600473
Exploiting Noninnocent (E,E)-Dibenzylideneacetone (dba) Effects in
Palladium(0)-Mediated Cross-Coupling Reactions: Modulation of the
Electronic Properties of dba Affects Catalyst Activity and Stability in Ligand
and Ligand-Free Reaction Systems
[
a]
[a]
[a]
[a]
Ian J. S. Fairlamb,* Anant R. Kapdi, Adam F. Lee, Gerard P. McGlacken,
[
a]
[b]
Felix Weissburger, AndrØ H. M. de Vries, and
[
b]
Lizette Schmieder-van de Vondervoort
Abstract: The reactivity of palladi-
reaction with aryl halide in the first
committed step in the general catalytic
cycle(s) (oxidative addition). Decreas-
ing electron density on the aryl moiety
of the dba-n,n’-Z ligand stabilizes the
Pd –h -dba interaction, reducing cata-
lytic activity. The specific type of dba-
n,n’-Z ligand appears to also play a sta-
bilizing role in the catalytic cycle, pre-
venting Pd agglomeration, and increas-
ing the type of dba-n,n’-Z ligand con-
0
0
um(0) complexes, [Pd (dba-n,n’-Z) ]
trols the concentration of L Pd and
2
3
n
(
n,n’-Z=4,4’-F; 4,4’-CF ; 4,4’-H; 4,4’-
the rate of the oxidative addition step,
and not other intimate steps within the
catalytic cycle(s), for example, transme-
tallation (or carbopalladation) and re-
ductive elimination. The role of dba-
n,n’-Z ligands in Heck arylation is
more convoluted and dependant on the
alkene substrate employed, although
trends have emerged. Changes in the
structure of dba-n,n’-Z had a minimal
affect on Buchwald–Hartwig aryl ami-
nation processes. A secondary Michael
reaction of dba-n,n’-Z with amine and/
or base effectively lessens its interfer-
ence in the catalytic cycle.
3
0
MeO) and [Pd (dba-n,n’-Z) ] (n,n’-Z=
2
4
,4’-CF ; 4,4’-H; 3,3’,5,5’-OMe), used as
3
0
2
precursor catalysts with suitable donor
ligands (e.g. phosphines, N-heterocyclic
carbenes), has been correlated in sever-
al palladium(0)-mediated cross-cou-
pling processes. Increasing the electron
density on the aryl moiety of the dba-
n,n’-Z ligand increases the overall cata-
lytic activity in the majority of these
processes. This effect primarily derives
ing catalyst longevity. A subtle balance
0
therefore exists between the L Pd con-
n
centration (and the associated catalytic
activity) and catalyst longevity. Chang-
0
2
from destabilization of the L Pd –h -
n
dba interaction (in dp–p* synergic
bonding, n=1 or 2), which ultimately
Keywords: arylation
pling homogeneous catalysis
kinetics · palladium
·
CÀC cou-
·
·
increases the global concentration of
0
catalytically active L Pd available for
n
Introduction
design and development of highly active Pd catalysts/preca-
[2]
talysts for these processes, particularly in Hartwig–Buch-
[
3]
[4]
[5]
[6]
CarbonÀcarbon and carbonÀheteroatom bond-forming pro-
wald amination,
Heck,
Sonogashira,
Stille
and
0
[7]
cesses mediated by Pd are amongst the most important and
Suzuki–Miyaura reactions. These studies have traditionally
[
1]
powerful in synthesis, which has led global interest in the
focused on the design and optimisation of s-donor ligands,
0
II
which are then added in situ to various Pd and Pd precur-
[
8]
0
[9]
sors. A phosphine-free Pd precursor is the air-stable
[
a] Dr. I. J. S. Fairlamb, A. R. Kapdi, Dr. A. F. Lee, Dr. G. P. McGlacken,
F. Weissburger
Department of Chemistry, University of York
Heslington, York, YO10 5DD (UK)
Fax : (+44)1904-432-515
0
[10]
complex [Pd dba ] (dba=E,E-dibenzylideneacetone), first
prepared by Ishii and co-workers. Donor ligands (L) such
as phosphines, are routinely added to this precursor to gen-
3
[11]
0
2
erate complexes of the type “L Pd (h -dba)”. The impor-
A
H
R
U
G
2
E-mail: ijsf1@york.ac.uk
tance associated with dba dissociation from electron-rich
[
b] Dr. A. H. M. de Vries, L. Schmieder-van de Vondervoort
DSM Pharma Chemicals
0
Pd species containing monodentate phosphine ligands, has
[12]
been systematically investigated by Amatore and Jutand.
Advanced Synthesis, Catalysis & Development
PO Box 18, 6160 MD Geleen (The Netherlands)
Cyclic voltametric, chronoamperometry and UV spectro-
8750
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

FULL PAPER
scopic experiments elegantly demonstrate that the most re-
this unusual ligand/Pd ratio (n,n’-Z = 3,3’-NO ), see
2
0
active Pd species in the oxidative addition with aryl halides
Figure 1; the parent dba ligand will now be referred to as
dba-4,4’-Hto avoid confusion. In Suzuki–Miyaura cross-cou-
pling it was determined that electron-withdrawing substitu-
0
such as iodobenzene, is the lowest ligated complex “L Pd ”
n
[13]
(
n=1 or 2). Ligation by dba (a p-acid), which ultimately
0
2
[14]
stabilizes the Pd centre via h -alkene coordination,
di-
minishes the concentration of the active catalyst species
0
L Pd (Scheme 1). As well as lowering the concentration of
n
0
0
L Pd , dba (and other p-acids) may reduce Pd agglomera-
n
tion (to produce clusters/colloids).
0
Scheme 1. Ligation of dba to Pd which diminishes the concentration of
0
catalytically active Pd .
Crucially, there is an antagonistic effect between the high
0
Figure 1. Various binuclear [Pd
3).
x y
(dba-R) ] complexes (x=1 or 2; y=2 or
A
H
R
U
G
intrinsic reactivity of the L Pd complex and its concentra-
n
0
0
tion in the equilibrium between L Pd (dba) and L Pd .
A
H
R
U
G
n
n
Indeed, if L is very electron rich, as is the case for trialkyl-
0
phosphines or N-heterocyclic carbenes, the L Pd
A
H
R
U
(dba) com-
ents such as NO or CF , deactivate the active catalyst spe-
2 3
n
0
plex will be more stable because the dba may accept the
cies (L Pd ), whereas strongly donating substituents such as
n
electrons and consequently the equilibrium will be more
OMe, increase catalytic activity over that of the dba-4,4’-H
ligand. In collaboration with the group of Jutand, we have
established that the dba-n,n’-Z ligand principally influences
0
shifted towards L Pd
A
H
R
U
G
n
tivity.
[22]
One may interpret that dba is a non-spectator ligand in
the true sense of the catalytic cycle for cross-coupling reac-
tions. Clearly dba is a spectator ligand in stabilization of
the oxidative addition reaction with aryl iodides. Indeed,
considering PPh as ligand, the overall rate of the oxidative
3
addition was faster when Z was an electron-donating group,
primarily because the greater electron-donating capacity of
0
0
2
L Pd to give [L Pd (h -dba)], which controls catalyst activi-
A
H
R
U
G
n
n
0
ty and longevity. However, under certain conditions dba
could also be involved in intimate steps within the catalytic
the Z group, the lower the affinity of dba-4,4’-Z for Pd -
0
A
H
R
U
G
3
2
II
cycle. Cationic Pd complexes ligated by dba have been de-
A
H
R
U
G
3
2
tected using ESI-MS/MS techniques in intermolecular Heck
arylation, providing direct evidence that dba could affect
overall rates more so than in other cross-coupling reac-
tive addition step is rate-limiting, as is the case when aryl
chlorides are used as coupling partners. The work also built
on the insightful studies conducted by Scrivanti and co-
workers who demonstrated that the p-accepting ability of an
[15]
[16]
tions. Other alkenic p-acids, for example, fumaronitrile,
[17]
[18]
0
2
p-fluorostyrene
as well as several other alkenes,
are
alkene in (P,N)Pd (h -alkene) complexes determines its cata-
[23]
known to play significant roles in cross-coupling reactions, a
lytic reactivity and overall stability.
[19]
property that is arguably underutilized. In these reactions
it is believed that the accepting alkene ligand binds to and
removes electron density from the metal centre (Pd or Ni),
thereby facilitating an otherwise difficult reductive elimina-
In this study, we present evidence that the “dba effect” is
generic in several common palladium(0)-mediated process-
0
es. The catalytic activity of all the Pd complexes as primary
0
sources of Pd in combination with donor ligands (N-hetero-
3
3
tion step in C
A
C
H
T
R
E
U
N
G
(sp )ÀC
A
H
E
N
(sp ) cross-coupling.
cyclic carbenes or phosphines) has been assessed in Suzuki–
Miyaura cross-coupling, a-arylation of esters, intermolecular
Heck arylation and Buchwald–Hartwig amination. The
affect of the aryl substituent in dba-n,n’-Z on the reaction
rate and enantioselectivity is further assessed in an asym-
metric variant of the intermolecular Heck arylation.
In a recent communication our group has demonstrated
that the strength of Pd –h -dba coordination is affected by
the p-electron accepting ability of the dba ligand. Syner-
gic effects (dp–p* backbonding) appear to control the rate
of dissociation of the dba ligand from Pd . A series of dinu-
clear Pd complexes, possessing aryl-substituted dba ligands,
generically represented as Pd
been prepared by our group.
down into two distinct classes, [Pd (dba-n,n’-Z) ] and
0
2
[20]
0
0
0
A
H
R
U
G
x
y
[
20,21]
Results and Discussion
3
0
[
[
Pd (dba-n,n’-Z) ], and although a third class is presented as
Suzuki–Miyaura cross-coupling—The effect of donor ligand:
2
0
0
Pd (dba-n,n’-Z) ], there is only one complex that possesses
The pure Pd -precursor complexes were evaluated against a
2
2
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8751

I. J. S. Fairlamb et al.
highly active catalyst system recently reported by Caddick
and co-workers for Suzuki–Miyaura coupling. We selected
Table 1. Pseudo first-order rates for consumption of 1 in Suzuki–Miyaura
cross-coupling for different Pd precursor complexes.
0
[25]
[24]
0
À1
Entry
Pd Precursor
Rate/s
[Pd (dba-n,n’-Z)
the reaction of 4-chlorotoluene (1) with phenylboronic acid
0
0
(
(
(
2) catalyzed by [Pd (dba-4,4’-H) ] (3 mol%)/NHC ligand
A
H
R
U
G
2
]
2
3
À5
À5
À6
À6
À6
1
2
3
4,4’-H3.11”10
(Æ2.3”10
)
)
)
derived from imidazolium salt 4) (3 mol%) to give 3
Scheme 2). Based on the ligand/Pd ratio, the generation of
4,4’-CF
3
2.25”10 (Æ3.6”10
À5
3,3’,5,5’-OMe
6.19”10 (Æ6.9”10
0
2 3
Pd (dba-n,n’-Z)
À5
À5
À6
À6
À6
À6
À6
4
5
6
7
8
4,4’-H3.69”10
(Æ2.5”10
)
)
)
)
)
4,4’-CF
4,4’-F
3
2.85”10 (Æ2.1”10
À5
3.49”10 (Æ3.3”10
À5
4,4’-tBu
4,4’-OMe
3.83”10 (Æ3.0”10
À5
5.47”10 (Æ5.1”10
0
A
H
R
U
G
2 2
[Pd (dba-n,n’-Z) ]
9
3-NO
2
6.24”10 (Æ9.6”10À7
À6
)
Scheme 2. Benchmark reaction for assessing the effect of aryl substitution
in Pd precursor complexes containing dba-n,n’-Z ligands.
ent (entry 7) showed a slightly higher rate, however, the
0
0
error hinders a direct comparison with [Pd (dba-4,4’-H) ].
2
3
The electron-donating 4-methoxy substituent did give a sub-
0
0
a mono-NHC ligand Pd species 5 was expected, which
stantially higher rate (entry 8). The Pd precursor complex
0
would be stabilized by ligation with dba-4,4’-H(Scheme 3).
The initial oxidative addition with 1 was thus expected to in-
exhibiting the slowest rate was [Pd (dba-3,3’-NO ) ]
(entry 9).
2
2 2
[26]
This is by some margin the most p-accepting
volve dissociation of the dba ligand from 5 to generate 6
ligand of the series of dba-n,n’-Z ligands evaluated.
II
(
which is likely to be solvated), which finally gives Pd -inter-
In these experiments the rates are affected by the nature
of the substituent in the dba-n,n’-Z ligand. These observed
mediate 7.
[20]
effects are substrate independent. A Hammett plot, utilis-
ing spara values, illustrates that the substituent effect is non-
linear (Figure 2a). A linear relationship was established for
the effect of the substituent in dba-n,n’-Z ligands in the stoi-
Scheme 3. Synthesis of complex 7.
0
The reaction was used as a benchmark for all of the Pd -
precursor complexes. Each reaction was monitored by high-
resolution gas chromatography (HRGC) analysis, allowing
the kinetics to be determined (pseudo first-order rates for
consumption of 1) (Table 1).
The rate for the reaction depicted in Scheme 2 using
0
0
À5 À1
[
(
Pd (dba-4,4’-H) ] as the Pd precursor was 3.11”10
s
2
À6
Æ2.3”10 ) (entry 1). As expected, there is a small differ-
ence in catalytic activity for the different dba-4,4’-H/Pd
ratios (entries 1 and 4); increasing dba-4,4’-Hconcentration
should decrease catalytic activity. The electron-withdrawing
0
trifluoromethyl group reduces the rate in both [Pd (dba-4,4’-
0
2
CF ) ] and [Pd (dba-4,4’-CF ) ] complexes (entries 2 and 5).
3
2
3 3
0
On the other hand, [Pd (dba-3,3’,5,5’-OMe) ] (entry 3) sig-
nificantly increases the rate of reaction relative to [Pd (dba-
2
0
4
,4’-H) ]. The electron-withdrawing fluoro substituent in
2
Figure 2. a) Hammett plot (using spara values) for the reaction depicted in
0
0
[Pd (dba-4,4’-F) ] exhibits a similar rate as for [Pd (dba-
0
2
3
2
Scheme 1 mediated by [Pd (dba-4,4’-Z) ] complexes (Z=OMe, tBu, H,
2
3
+
4,4’-H) ] (compare entries 4 and 6). The tert-butyl substitu-
3
F or CF ); b) Hammett plot using s values.
3
8752
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

Cross-Coupling Reactions
FULL PAPER
chiometric oxidative addition reaction of PhI (2 mm) with
the well-known solution behaviour of [Pd (dba-4,4-H) ] and
2
3
0
0
[30]
the [Pd
A
C
H
T
R
E
U
N
G
(PPh ) ] complex generated from [Pd (dba-4,4’-Z) ]
related complexes, therefore oxidative addition of 8 with
3
2
2
3
[
22]
0
(
1 mm) and PPh (4 mm) in DMF at 208C (1=À1.7).
Pd to give 11 should also be promoted by the donating
3
0
2
We have not included [Pd (dba-3,3’-NO ) ] for compari-
properties of the dba-n,n’-Z ligands (Scheme 5).
2 2
son as this possesses one less dba-4,4’-Z ligand than the
0
[
Pd (dba-n,n’-Z) ] complexes. However, we can state that
2
3
the 3-NO group (log(k
/k ) = À0.772; s =0.71) would
2
3-NO2
H
m
lie significantly away from the curve (Figure 2). This indi-
cates that there could be a change in the rate determining
step as a function of the different substituents (from oxida-
tive addition to dba-3,3’-NO2 dissociation from Pd ). A
Hammett plot, utilizing the s values,
0
+
[27]
which takes into
Scheme 5.
account resonance stabilisation, exhibits a less pronounced
non-linear trend (Figure 2b), which indicates a direct corre-
lation of the para-substituent of dba-n,n’-Z with rate. The 3-
NO2 substituent (not depicted) lies away from the curve
Running the reactions at higher concentrations (0.33m
rather than 0.164m with respect to 8) results in the precipita-
tion of Pd black. The p-accepting properties of the dba-n,n’-
Z ligand must also play a vital role in these reactions, al-
though there appears to be a very fine balance. The differ-
+
when the s and s values are compared with log(k
/k ).
H
1
3-NO2
Suzuki–Miyaura cross-coupling—In a ligand-free catalyst
0
system: The Pd complexes were next evaluated in a ligand-
[28]
0
free system for the cross-coupling of the activated bromo-
-pyrone (8) with 4-fluorophenylboronic acid (9) to give 10
ence in reactivity seen between [Pd (dba-4,4’-F) ] and
2
3
0
2
[Pd (dba-4,4’-OMe) ] (entries 5 and 7) does not fit so readi-
2 3
[29]
(
Scheme 4).
Given the absence of an activating donor
ly into the overall model, as the latter complex would be ex-
pected to be more reactive.
0
The significant difference in rate between [Pd (dba-4,4’-
0
H) ] and [Pd (dba-4,4’-H) ] (entries 1 and 4), and lesser dif-
2
2
3
0
ferences seen for [Pd (dba-n,n’-Z) ]-type complexes (en-
2
3
tries 4–7), suggests that in the ligand-free system the global
concentration of dba-n,n’-Z ligand is a more dominant
factor than subtle modifications to the dba-n,n’-Z ligand
structure. This was confirmed by increasing the dba-4,4’-H
0
2
concentration in reactions mediated by [Pd (dba-4,4’-H) ];
3
Scheme 4. Benchmark reaction for a ligand-free type system.
the reaction profiles are shown in Figure 3. Higher concen-
trations of dba-4,4’-Hhave a dramatic affect on the rates. In
À4 À1
ligand (e.g. NHC or PR ligand) it was predicted that the
the presence of 2 mol% dba-4,4’-H, the rate is 3.65”10
s
3
À5)
dba-4,4’-Z ligand could act as a donor and/or acceptor. The
rates for the pseudo first order consumption of 8 are listed
in Table 2 (established by HRGC analysis).
(Æ7.2”10 . The rate is reduced more so in the presence of
À4 À1 À5
5 mol% dba-4,4’-H(2.63”10 s , Æ2.8”10 ).
These data indicate that the choice of the substituent on
the dba ligand influences the so-called ligand-free system,
0
particularly for [Pd (dba-n,n’-Z) ] complexes (entries 1–3).
2
0
Initial ligand dissociation from the [Pd (dba-n,n’-Z) ] com-
x
y
plexes (x=1 or 2; y=2 or 3) is assumed to be facile, given
Table 2. Pseudo first-order rates for the consumption of 8 in Suzuki–
0
Miyaura cross-coupling for different Pd precursor complexes.
0
À1
Entry
Pd Precursor
Rate/s
[Pd (dba-n,n’-Z)
0
A
H
R
U
G
2
]
À4
À4
À5
À5
À5
1
2
3
4,4’-H3.63”10
(Æ5.6”10
)
)
)
4,4’-CF
3
2.87”10 (Æ3.2”10
À4
3,3’,5,5’-OMe
6.34”10 (Æ1.6”10
0
A
H
R
U
G
2 3
[Pd (dba-n,n’-Z) ]
À4
À4
À5
À5
À5
À5
4
5
6
7
4,4’-H6.00”10
4,4’-F
4,4’-tBu
(Æ4.1”10
)
)
)
)
6.24”10 (Æ1.5”10
Figure 3. Increasing dba-4,4’-Hconcentration in the Suzuki–Miyaura re-
À4
6.20”10 (Æ1.2”10
0
action of 8 and 9 to give 10 mediated by [Pd
2
(dba-4,4’-H)
3
] (dashed lines
À4
4,4’-OMe
5.29”10 (Æ6.6”10
0
0
serve as a guide to the eye). *: [Pd
(dba-4,4’-H)
; *: [Pd
(dba-4,4’-H)
]
0
2
3
2
3
A
T
E
N
2 2
[Pd (dba-n,n’-Z) ]
0
0
+
2 mol% dba (based on Pd ); ~: [Pd
2
(dba-4,4’-H)
3
] + 5 mol% dba
À5
8
3-NO
2
5.75”10 (Æ7.5”10À6
)
0
(
based on Pd ).
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8753

I. J. S. Fairlamb et al.
So, although higher dba concentration is expected to
hinder Pd agglomeration (catalyst decomposition), it does
so at the expense of catalyst activity, blocking required coor-
Table 3. Pseudo first-order rates for the consumption of 12 in a-arylation
of 13 for different Pd -precursor complexes.
0
0
À1
Entry
Pd Precursor (Z)
Rate/s
(dba-n,n’-Z)
0
0
dination sites on Pd .
A
H
E
N
[Pd
2
3
]
À5
À5
À6
À7
À6
À6
À6
À6
1
2
3
4
5
6
4,4’-H4.57”10
(Æ3.8”10
)
)
)
)
)
)
4,4’-CF
3
1.29”10 (Æ7.9”10
a-Arylation of esters: The a-arylation of ester compounds
has emerged as a potentially useful tool for drug discovery,
À5
3,3’,5,5’-OMe
4,4’-F
4,4’-tBu
9.37”10 (Æ2.8”10
À5
3.80”10 (Æ4.8”10
[
31]
À5
amongst other applications.
The benchmark a-arylation
4.99”10 (Æ4.5”10
0
À5
reaction employed to evaluate the Pd precursor complexes
was that concerning cross-coupling of bromobenzene (12)
and tert-butyl acetate (13), using LHMDS as the base, to
give the a-arylated product 14 (Scheme 6).
4,4’-OMe
5.44”10 (Æ2.8”10
0
A
H
R
U
G
2 2
[Pd (dba-n,n’-Z) ]
7
3-NO
2
4.86”10 (Æ6.1”10À7
À6
)
0
CF ) ] and [Pd (dba-3,3’-NO ) ], respectively, is much
3
3
2
2 2
0
slower relative to [Pd (dba-4,4’-H) ]. Alternatively, there is
2
3
the possibility that the dba-n,n’-Z ligand is being consumed
where Z is CF or NO ), governed by a slow secondary rate
(
3
2
which will be affected by the nature of Z; reaction of the
ester enolates, or aryl halide, with the dba-n,n’-Z ligands has
not been found in these a-arylation reactions (by HRGC/
MS), nor have any derived products been isolated by
column chromatography.
Scheme 6. Benchmark a-arylation reaction.
0
0
The reaction is mediated by a 1:1 mixture of [Pd (dba-
Intermolecular Heck arylation: The Pd complexes were
4
,4’-H) ] and an NHC ligand (generated in situ from the imi-
next assessed in Heck arylation, using similar reaction con-
ditions to those reported by Fu and co-workers
(Scheme 7).
2
dazolium salt 4), reported by Hartwig and co-workers to
proceed in 87% isolated yield after 12 h. With this infor-
mation to hand, the reaction profiles and pseudo first-order
[32]
[33]
rates for the consumption of 12 were determined for
0
[
[
Pd (dba-n,n’-Z) ] (Z=H , F, C F, tBu and OMe) and
2
3
3
0
2
Pd (dba-3,3’-NO ) ] complexes (Figure 4 and Table 3).
2 2
Changing substituents in the dba-n,n’-Z ligands again af-
0
fects the rates. The catalyst precursors [Pd (dba-4,4’-CF ) ]
2
3 3
0
and [Pd (dba-3,3’-NO ) ], containing electron-withdrawing
2
2 2
Scheme 7. Benchmark Heck arylation of n-butyl acrylate 15 mediated by
a Pd/tBu P·HBF catalyst system.
ligands, both exhibit induction periods, which is intriguing.
At about 4–5 h, both reactions begin to turnover. The origin
3
4
of the induction period is unclear. It is tentatively proposed
0
2
that the reaction of the NHC ligands with [Pd (dba-4,4’-
Note that we employed the air-stable salt, tBu P·HBF in-
3 4
stead of the free ligand, tBu P. The kinetic profiles for the
3
reaction of 12 with n-butyl acrylate (15) are shown in
Figure 5; the rates are given in Table 4.
The kinetic profiles from the reaction of 12 with 15 to
0
give 16 exhibits broad rate differences for the Pd -precursor
complexes investigated. An induction period is seen for
0
[
Pd (dba-3,3’-NO ) ], whereas induction periods for other
2
2 2
0
Pd precursor complexes were less pronounced. Curiously,
the [Pd (dba-4,4’-tBu) ] complex is more deactivated, rela-
tive to [Pd (dba-4,4’-H) ,], in this reaction. The other com-
0
2
3
0
2
3
plexes fall into the same reactivity sequence mentioned
above. To give an idea of the difference in reactivity of all
0
the Pd
A
T
E
N
(dba-R) complexes, an identical reaction was con-
x
y
À5 À1
ducted using Pd
A
H
R
U
G
s
2
À6
(
Æ9.7”10 ) was determined. Although Pd
A
H
R
U
G
Figure 4. Reaction profiles from the reaction depicted in Scheme 6
0
always out perform Pd
A
H
R
U
G
x
y
0
(
[
[
dashed lines serve as a guide to the eye). *: [Pd
2
(dba-3,3’-NO
2
)
2
];
]; &:
(dba-3,3’,5,5’-
&
:
0
0
0
is sometimes seen with Pd
A
H
R
U
G
2
Pd
Pd
OMe)
2
(dba-4,4’-CF
(dba-4,4’-tBu)
].
3
)
3
]; ^: [Pd
]; *: [Pd
2
0
(dba-4,4’-F)
3
]; ~: [Pd
2
(dba-4,4’-H)
3
[28,37]
0
]; ~: [Pd0
free conditions),
this example clearly shows that the
2
3
2
(dba-4,4’-OMe)
3
2
3
rate is substantially faster using a Pd(OAc) precatalyst.
A
H
R
U
G
2
8754
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

Cross-Coupling Reactions
FULL PAPER
of examples, regardless of the structure of the aryl halide.
0
[
Pd (dba-3,3’,5,5’-OMe) ] gives the highest conversions,
2
3
0
whereas [Pd (dba-4,4’-CF) ] gives the lowest. In reaction
2
3
IV, that of p-fluorobromobenzene with 15, indicates that
0
[
Pd (dba-4,4’-H) ] is the most active precursor complex. For
2 3
reactions employing styrene as a substrate (reactions V–IX),
the reactivity sequence is as expected. However, the differ-
0
2
0
2
ences between [Pd (dba-4,4’-H) ] and [Pd (dba-3,3’,5,5’-
3
OMe) ] is relatively small, albeit still in favour of the latter.
3
0
The [Pd (dba-4,4’-CF) ] complex is on the other hand nota-
2
3
bly less active.
In this series, the results indicate that styrene, in excess
relative to dba-n,n’-Z, effectively competes with dba-n,n’-Z
0
ligation to Pd , thus playing a more dominant role in the cat-
Figure 5. Reaction profiles from the reaction depicted in Scheme 7
0
(
[
[
dashed lines serve as a guide to the eye). *: [Pd
2
0
(dba-3,3’-NO
(dba-4,4’-tBu)
2 3
2
)
2
];
&
:
alytic cycle than the more sterically hindered alkene 15. It
has been demonstrated by Jutand and Amatore that when-
ever the oxidative addition is not rate-determining, the de-
celerating effect of the alkene on this reaction is in favour
0
0
Pd
Pd
2
(dba-4,4’-CF
3
)
3
]; ~: [Pd
2
0
(dba-4,4’-H)
3
]; &: [Pd
]; *:
0
2
(dba-4,4’-OMe)
3
]; ~: [Pd
2
(dba-3,3’,5,5’-OMe) ].
3
[34]
of a higher efficiency in the catalytic cycle.
Effectively,
Table 4. Pseudo first-order rates for the consumption of 12 in Heck ary-
alkene coordination reduces the global concentration of
0
lation for different Pd precursor complexes.
“
0
L Pd ”, bringing the rate of the oxidative addition closer to
n
0
À1
Entry
Pd Precursor (Z)
Rate/s
(dba-n,n’-Z)
that of the carbopalladation step, that is, alkene nucleophilic
0
II
ACHTREUNG
A
H
E
N
[Pd
2
3
]
attack on the Pd
(aryl)halide complex formed in the oxida-
À5
À5
À6
À6
À6
À6
À6
1
2
3
4
5
4,4’-H3.34”10
(Æ1.3”10
)
)
)
)
)
tive addition step, which is usually the rate determining
4,4’-CF
3
2.04”10 (Æ7.6”10
[35]
À5
step. A mechanism is outlined in Scheme 8 which shows
the roles of the alkene ligands in the catalytic cycle (the al-
kenyl product may also play a role).
3,3’,5,5’-OMe
4,4’-tBu
4,4’-OMe
6.13”10 (Æ5.3”10
À5
3.11”10 (Æ1.4”10
À5
5.13”10 (Æ5.1”10
0
A
T
E
N
[Pd
2
(dba-n,n’-Z)
2
]
0
The reactive Pd -species I is assumed to be monoligate-
6
3-NO
2
4.12”10 (Æ6.2”10À7
À6
)
[13a]
d
by tBu P (solvated by dioxane), the concentration of
3
this species will be diminished by both the dba-n,n’-Z ligand
and the alkene reactant (styrene or n-butyl acrylate), to give
II and III, respectively. The concentration of II and III will
be determined by the strength of the Pd –h -alkene interac-
tion. Crucially, for monoligated species III direct oxidative
addition gives complex VII, which can either undergo carbo-
palladation to give V or alkene ligand loss to give IV. The
relative conversions seen in Figure 6, for styrene in particu-
lar, indicate that there is a delicate and subtle balance be-
To test the substrate dependence in Heck arylation using
0
0
0
0
2
the [Pd (dba-3,3’,5,5’-OMe) ] and [Pd (dba-4,4’-CF) ] Pd -
2
3
2
3
0
2
precursor complexes, relative to [Pd (dba-4,4’-H) ], a series
3
of reactions were carried out in parallel using the ASW
000 from ChemSpeed technology (Figure 6). In the first set
of reactions (I–III), using 15 as the alkene substrate, it is ap-
parent that the dba-4,4’-Z ligands play a role in the majority
2
0
0
0
Figure 6. Heck arylation mediated by A= [Pd
Scheme 7: tBu P·HBF
ternal standard (dihexyl ether); yield calculations include corrected response factors for reagents and products.
2
(dba-4,4’-H)
3
]; B = [Pd
2
(dba-3,3’,5,5’-OMe)
3
]; C = [Pd
2
(dba-4,4’-CF)
3
]. Reagents and conditions, see
3
4
(6 mol%) was used in the reactions employing styrene. Conversions were determined by HRGC/MS and calculated against an in-
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8755

I. J. S. Fairlamb et al.
Scheme 9. Heck arylation of methyl 2-acetamidoacrylate.
tions conducted in the presence of styrene, indicating that
methyl 2-acetamidoacrylate also plays a role in the catalytic
cycle (other than in carbopalladation).
Effects on the asymmetric intermolecular Heck arylation:
Having established the effects of the various dba-n,n’-Z li-
gands in intermolecular Heck arylation, we speculated that
the regio- and enantioselectivity could be affected in an
asymmetric variant, particularly if the dba-n,n’-Z ligand is
II
bound to the Pd intermediate undergoing carbopallada-
tion—the carbonÀcarbon bond forming step in the catalytic
cycle where the chiral centre is generated. The protocol se-
lected was one initially reported by Hayashi and co-work-
ers, but developed by Keay and co-workers. The reac-
tion of phenyltriflate (1 equiv) with 2,3-dihydrofuran
Scheme 8. Roles for alkene ligands in Heck arylation. Note: We expect
that all 14-electron intermediates will be stabilized by solvent (as more
stable 16-electron complexes); in equilibrium with dimeric Pd species or
[38]
[39]
[
36]
stabilized by suitable agostic interactions.
0
0
(
6 equiv), promoted by a [Pd (dba-4,4’-H) ] (3 mol% Pd )/
2 3
[40]
tween the equilibria of I and II and I and III; presumably II
and III are also in equilibrium. It can thus be proposed that
III is favoured when R=Ph and the dibenzylideneacetone
ligand is dba-3,3’,5,5’-OMe or dba-4,4’-H.
(R)-BINAP (3 mol%) catalyst system in the presence of
DIPEA (3 equiv) in dioxane at 508C for 7 d, was used as a
benchmark (Scheme 10). The major product is (R)-21a.
Minor products were expected, for example, (S)-21b and
21c. We were able to reproduce the results reported for
[Pd (dba-4,4’-H) ] (entry 1, Table 5). For [Pd (dba-4,4’-
2 3 2
0
0
[39]
We also evaluated [Pd (dba-4,4’-H) ], [Pd (dba-3,3’,5,5’-
2
3
2
0
0
0
OMe) ] and [Pd (dba-4,4’-CF) ] in intermolecular Heck ary-
3
2
3
lation of methyl 2-acetamido-
acrylate with 4-bromobiphenyl
and
4-fluoro-bromobenzene
(
reactions I and II, respective-
ly, Figure 7), using a protocol
developed by de Vries and co-
[37]
workers (Scheme 9).
The results exhibit a trend
which is similar to the reac-
Scheme 10. Benchmark reaction for asymmetric Heck arylation.
0
Table 5. Application of various Pd precursor complexes in the Heck ary-
[
a]
lation of 2,3-dihydrofuran 20.
0
Entry Pd Precursor
Yield/
%
ee/%
(R)-23a
ee/%
(S)-23b
Selectivity
for 23a/%
[
b]
[b]
[c]
(
Z)
0
A
H
R
U
G
2 3
[Pd (dba-n,n’-Z) ]
[
d]
1
2
3
4
5
4,4’-H
4,4’-CF
3,3’,5,5’-OMe
4,4’-F
51
42
63
69
63
63
61
65
68
62
21
18
22
41
20
95
93
94
95
94
3
4,4’-OMe
[
a] Yields and the enantiomeric excess (ee) values represent an average
from at least two separate runs. Yields given are after column chromatog-
raphy. [b] The enantiomeric excess was determined on a b-cyclodextrin
chiral column by HRGC. The ee values and product regioisomeric selec-
tivities were determined by analysis of the crude reaction products by
HRGC. [c] In all cases <1% of 21c was detected by HRGC. [d] The re-
0
Figure 7. Heck alkenylation reactions mediated by A = [Pd
2
(dba-4,4’-
]. Re-
agents and conditions, see Scheme 7: aryl bromide (1 equiv), 17
0
0
H)
3
]; B
=
[Pd
2
(dba-3,3’,5,5’-OMe)
3
]; C
=
[Pd
2 3
(dba-4,4’-CF)
0
[39]
(
1.1 equiv), (iPr)
2
NEt (1.15 equiv),
C
6
H
5
CH
2
NEt
3
Br (4.6 equiv), Pd
ported conversion to products is 56%: (R)-21a (52% yield, 66% ee);
(
0.30 mol% with respect to aryl bromide), NMP, at 1258C, 3 h.
(S)-21b (2% yield; 15% ee); 21c (1% yield).
8756
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

Cross-Coupling Reactions
FULL PAPER
[46]
0
CF) ] catalytic activity was reduced, whereas the enantiose-
tBu P·HBF
4
and pre-mixing the Pd precursor in the pres-
3
3
lectivity and product selectivity were near identical to
ence of the base (tBuONa), we were able to effectively
carry out the reactions under standard Schlenk techniques.
In the first reaction, reaction of PhBr (12) with aniline using
0
[
Pd (dba-4,4’-H) ] (entry 2, Table 5).
2
3
0
0
For [Pd (dba-3,3’,5,5’-OMe) ], [Pd (dba-4,4’-F) ] and
2
3
2
3
0
0
[
Pd (dba-4,4’-OMe) ] the conversions improved to 63, 69
1 mol% [Pd (dba-4,4’-H) ] and 0.8 mol% tBu P·HBF we
2
3
2
3
4
and 63%, respectively (entries 3–5, Table 5). The minor dif-
ferences in enantiomeric excess observed for (R)-21a pro-
vide indirect evidence that the dba-n,n’-Z ligands are not in-
volved in the carbopalladation
able to attain a quantitative yield of 24 in 10 min (entry 1,
0
Table 6). To discern any differences between the Pd -precur-
0
sor complexes, the concentration of [Pd (dba-4,4’-H) ] was
2
step of the catalytic cycle. The
0
[a]
Table 6. Application of various Pd precursor complexes in Buchwald–Hartwig aryl amination reactions.
higher enantiomeric excess of
S)-21b (41%), relative to the
other Pd -precursor complexes,
is worthy of note. The highest
enantiomeric excess recorded
[b]
[c]
Entry
Aryl bromide
Pd/mol%
t/min
Yield/%
(
0
A
H
R
N
2
]
0
4
,4’-H3,3
4,4’-CF
>99
3
1
1
10
>99
>99
for
(R)-21a
was
with
0
2
[Pd (dba-4,4’-F) ], and of the
3
0
0.2
.5
10
10
>99
10
>99
27
>99
22
series, exhibits the most
marked affect on the stereo-
chemical outcome. The higher
yield observed in the presence
of dba-4,4’-F suggests that this
ligand plays a greater stabiliz-
ing role, preventing the ag-
2
3
1
60
60
82
13
72
14
76
9
0
1
.5
glomeration/precipitation
of
60
94
97
98
Pd from the reaction over the
long reaction time (7 d); the
balance between catalyst activ-
ity and stability being the key
to higher yields in this particu-
lar process.
[a] Reagents and conditions, see Scheme 11. [b] The Pd/L ratio was maintained at 1:0.8 on lowering the overall
Pd concentration. [c] This is the yield after column chromatography (purity > 97%).
0
lowered to 0.5 and 0.2 mol% (the Pd/L ratio was main-
tained at 1:0.8). The yield dropped substantially going from
0.5 to 0.2 mol%, and some minor differences between
0
Buchwald–Hartwig amination: The Pd -catalysed amination
of aryl halides has emerged as an important method for the
[41]
0
0
0
synthesis of aryl amines.
It was anticipated that the in-
[Pd (dba-4,4’-H) ], [Pd (dba-3,3’,5,5’-OMe) ] and [Pd (dba-
2 2
0
volvement of dba-n,n’-Z ligands in the reaction of an aryl
halide with amine could be more complex than the other
processes. Amines, specifically anilines, are known to react
with dba-4,4’-Hand related compounds through a Michael
addition reaction, which inadvertently would reduce the
interference of dba-4,4’-Hin the catalytic process. Clearly, a
combination of amine and base would serve as a better nu-
cleophile to dba-4,4’-Hthan any individual reactions of
4,4’-CF ) ] were recorded at 0.2 mol% Pd . The reaction of
3 2
1-bromo-3,5-dimethylbenzene required 1 mol% Pd for high
yield (entry 2). At both 1 mol% and 0.5 mol% Pd , negligi-
ble differences were seen between the Pd -precursor com-
0
0
0
[42]
plexes. For the more activated substrate, methyl 4-bromo-
benzoate, the yields of the products were near identical.
These results strongly suggest that the dba-n,n’-Z ligand
plays an insignificant role under the conditions described by
Hartwigꢁs group. The isolation of small quantities of mono-
and bisaminated products of dba-4,4’-Z suggests that it is
consumed under the reaction conditions, although it should
be noted that on several occasions unreacted dba-4,4’-Z
ligand has also been recovered after careful column chroma-
tography. Overall, it appears that dba-n,n’-Z ligands play a
more innocent role in aryl amination processes.
[43]
amine or base. We selected the reaction conditions report-
ed by Hartwigꢁs group, employing the monophosphine
[44]
[45]
ligand, tBu P (Scheme 11).
3
On a general note, these reactions are more difficult to
conduct without the use of a dry-box (in our hands, strict
Schlenk techniques were essential and the results were
sometimes
variable).
However,
through
use
of
Conclusion
In this study we have developed a tuneable concept through
modulation of the electronic properties of the non-innocent
dibenzylideneacetone (dba) ligands and shown that this sig-
Scheme 11. Benchmark aryl amination reaction.
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8757

I. J. S. Fairlamb et al.
0
nificantly affects catalytic activity in several Pd -mediated
and as a consequence these ligands do not measurably con-
0
processes. Simple modifications to the ancillary dba-Z li-
gands surrounding the active Pd species causes dramatic
trol the concentration of catalytically active “L Pd ” (where
n
0
L is a monophosphine). We envisage that complexes con-
[47]
0
changes in catalytic performance. With the exception of
the aryl amination processes, all other reactions investigated
indicate that more electron rich dba-n,n’-Z ligands cause an
increase in catalyst activity. Electron deficient dba-n,n’-Z li-
gands reduce the rates of reaction in the majority of cases,
taining more electron deficient ligands, [Pd (dba-4,4’-F) ]
2
3
0
and [Pd (dba-4,4’-CF ) ] or variants thereof, will be used in
2
3 3
specialized applications.
the exception being in asymmetric Heck arylation, where
0
Experimental Section
[
Pd (dba-4,4’-F) ] gave the highest conversions and enantio-
2
3
selectivities. The 4-fluoro substituent can play two roles
electronically: the first effect would be deactivating based
on inductive arguments, and the second effect would be acti-
vating via mesomeric effects (as a strong p-donor). Given
the long reaction times for these reactions (7 d), the dba-
All solvents were dried using standard procedures prior to use and stored
under an argon atmosphere. Nitrogen gas was oxygen-free and was dried
immediately prior to use by passage through an 80 cm column containing
sodium hydroxide pellets and silica gel. Argon gas was used directly via
balloon transfer or on a Schlenk line. TLC analysis was performed rou-
tinely using Merck 5554 aluminum backed silica plates or Macherey–
Nagel polygram ALOX N/UV254 aluminum oxide coated plastic sheets.
Compounds were visualized using UV light (254 nm) and a basic aqueous
solution of potassium permanganate or acidic DNP (dinitrophenol hydra-
zine). GC parameters: Analysis was performed using a Varian CP-3800
GC equipped with a CP-8400 Autosampler. Separation was achieved
using a DB-1 column (30 m”0.32 mm, 0.25 mm film thickness) with carri-
4
,4’-F ligand will be expected to play a stabilizing role in
0
keeping Pd available for further reaction (avoiding agglom-
eration/precipitation). The extent of the substituent effect is
subtly dependent on individual reaction systems. For the
0
2
[
Pd (dba-3,3’,5,5’-OMe) ] complex, the subtle blend of do-
3
nating substituents which possess electron removal ability
the substituents being in the meta-position) appears to find
À1
er gas flow rate of 3 mLmin and a temperature ramp from 50 to 2508C
(
À1
at 208Cmin . The injection volume was 1 mL with a split ratio of 50. The
the right balance between catalytic activity and stability.
Generally, when electron-donating ligands are employed in
Suzuki–Miyaura coupling, a-arylation and Heck arylation,
response factors for the internal standard, substrates and products were
calculated using an appropriate calibration for this GC and column. The
0
0
Pd complexes, [Pd
2
(dba-n,n’-Z)
] (n,n’-Z=4,4’-CF
], have been reported previously
] (n,n’-Z=4,4’-CF ; 4,4’-H; 3,3’,5,5’-OMe) were formed from CH
] over 72 h, which were then dried under
3
] (n,n’-Z=4,4’-CF
3
; 4,4’-H; 4,4’-MeO;
0
4
,4’-tBu), [Pd (dba-n,n’-Z)
2
3
; 4,4’-H; 3,3’,5,5’-OMe) and
for example, tBu P or NHC ligands derived from 4, the dba-
0
[20,22]
0
3
[
Pd
2
(dba-3,3’-NO
2
)
2
[Pd
2
(dba-n,n’-
Cl
n,n’-Z ligands appear to play a critical role in diminishing
Z)
3
3
2
2
0
0
the concentration of the catalytically active species “L Pd ”.
solutions of [Pd (dba-n,n’-Z)
2
n
high vacuum (0.1 mmHg at 308C for 12 h). 1,3-Bis(2,6-diisopropylpheny-
l)imidazolium chloride (4) and tri-tert-butylphosphonium tetrafluoroboric
acid (tBu P·BF ) were purchased from Strem Chemicals. HRGC/MS or
3 4
In Heck arylation using Fuꢁs conditions, Pd(OAc) per-
A
H
R
U
G
2
0
formed considerably better than [Pd (dba-4,4’-H) ];
2
3
0
[
Pd (dba-3,3’,5,5’-OMe) ] shows slightly lesser activity than
2
3
LCMS was used throughout these studies to confirm the identity and
Pd
A
C
H
T
R
E
U
N
G
(OAc) under these specific conditions. A general point
purity of individual reaction components (substrates and products).
2
0
therefore emerges from this comparison: for some cross-
coupling reactions, certainly not all, Pd(OAc) is a superior
source of palladium. In due course we will report a full com-
[Pd (dba-3,3’,5,5’-OMe)
2
A
H
R
U
G
2
], also known as [Pd(dm-dba) ], is available com-
mercially through Sigma-Aldrich (Cat. No. 65693-3), and is a non-propri-
AHCTREUNG
2
etary material.
0
2
[20]
A
H
R
U
G
(dba-4,4’-F)
3
]: Following the reported procedure,
NaCl (1.64 g,
0
parative study of Pd(OAc) , [Pd (dba-4,4’-H) ] and [Pd-
A
H
R
U
G
2 2 3
2
2
(2.5 g, 14.6 mmol) in metha-
A
C
H
T
R
E
U
N
G
(nbe) ] as palladium sources for cross-coupling reactions
3
nol (70 mL) and stirred at ambient temperature under an inert atmos-
phere for 24 h. It was subsequently filtered through a plug of cotton wool
and concentrated in vacuo to approximately half its original volume. The
solution was warmed to 608C and then 1,5-bis(4’-fluorophenyl)penta-
(
nbe=norbornene).
Under the ligand-free Suzuki–Miyaura cross-coupling
conditions, for example, in the absence of an intentionally
added donor ligand, it was shown that the dba-n,n’-Z ligands
exhibit a marked affect on catalyst activity, an observation
that may be rationalized in part by consideration of the elec-
tronic properties of the ligands. It is conceded that the sub-
stituent effects here follow a less obvious trend than reac-
tions employing a donor ligand, which is an observation less
well understood at this stage. It is clear that dba-n,n’-Z con-
centration plays a more crucial role in the ligand-free reac-
tions.
(
1E,4E)-dien-3-one (11.9 g, 44 mmol) was added. The resulting mixture
was stirred at 608C for 15 min, then sodium acetate (7.23 g, 87.6 mmol)
was added, and the reaction cooled gradually to ambient temperature.
The mixture was stirred at ambient temperature for 2 h until a dark red
precipitate was observed. The precipitate was filtered and washed with
methanol (2”50 mL), water (2”50 mL), and finally acetone (2”3 mL).
The product was partially dried under suction and then the solid was
added to a Schlenk flask which was placed under a flow of nitrogen with
stirring overnight. This gave the title compound as a maroon/purple mi-
crocrystalline solid (3.81 g, 51.1%). M.p. 149–1518C; IR (CH
1
aromatic), 984 cm (# CHtrans); UV (THF, nm); 226 ( p–p*), 337 (n–
p*), 536 (d–d); ESI-MS: m/z: [M ] not detected, 775.4 [Pd
2
Cl
2
): n˜ =
652 m ((# C=O), 1607 vs
ACHTREUNG
(# C=C), 1585 (# C=C aromatic), 1567 (# C=C
À1
0
With these results in hand we propose that [Pd (dba-
+
+
2
dba
2
Na]
0
+
+
3
,3’,5,5’-OMe) ] or [Pd (dba-3,3’,5,5’-OMe) ] should replace
(70), 881.2 [Pd dba Na] (54), 987.1 [Pd dba Na] (100); elemental anal-
ysis calcd (%) for C51
3.78. This compound may be recrystallized from CH Cl /Et O 1:5 to give
2
2
3
3
2
4
2
0
0
the most commonly used Pd precursor complex, [Pd (dba-
,4’-H) ], which are both as easy to prepare. For Buch-
36 6 3 2
H F O Pd : C, 59.84, H3.54; found: C 59.53, H
2
[20]
2
2
2
4
3
0
[21]
[
Pd
General procedure for Suzuki–Miyaura cross-coupling in the presence of
,3-bis(2,6-diisopropylphenyl)imidazolium chloride (4) (general proce-
dure A): Phenylboronic acid (0.41 mmol), 4-chlorotoluene (0.45 mmol), 4
2 3 2 2
(dba-4,4’-F) ]·CH Cl (X-ray obtained).
wald–Hartwig aryl amination reactions, it essentially does
0
not matter what type of Pd -precursor complex is used, as
1
the dba-n,n’-Z ligand is ultimately consumed under the reac-
tion conditions (by base, amine or a combination of both),
0
2
(0.012 mmol, 3 mol%), [Pd (dba-4,4’-H) ] (0.012 mmol, 3 mol%), tetra-
8758
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

Cross-Coupling Reactions
FULL PAPER
butylammonium bromide (0.041 mmol, 10 mol%) and biphenyl (internal
standard, 0.04 mmol, 10 mol%) were added to a dry Schlenk tube. The
tube was evacuated through three cycles, whilst purging back with nitro-
gen or argon gas. Dry toluene (2 mL) was then added via cannula and
the mixture is stirred magnetically for 5 min (pre-incubation time=
rene and methyl 2-acetamidoacrylate. The following products exhibited
spectroscopic data equivalent to previously reported literature data: (E)-
[
49]
4-methoxycinnamic acid n-butyl ester,
(E)-4-fluoro-cinnamic acid n-
[
49]
[49]
[49]
butyl ester,
(E)-stilbene,
(E)-2-methoxystilbene,
bene, (E)-3-biphenyl-prop-2-enoic acid
n-butyl ester, 1-phenyl-2-(4’-biphenyl)-(Z)-ethene, (E)-2-(4’-benzoyl-
(E)-4-fluorostil-
[
49]
[
28]
[50]
1
0 min). The mixture was cooled to 08C (10 min) and then a solution of
[
51]
KOMe (1.23 mmol, 3 equiv) in dry MeOH(1 mL) was added dropwise
over 10 min (total pre-incubation time=0.5 h for all reactions). The reac-
tion was then warmed to 408C over ca. 5 min. The reaction was followed
by HRGC analysis after the temperature had reached 408C (except for
t=0). For each analysis point, a 50 mL sample was withdrawn from the
reaction mixture by gas-tight syringe and sample was quenched by pas-
sage through a small plug of silica (washed with acetone). Several data
points were chosen to see whether there was any time dependence effect
on post-quenched samples. For this reaction, there was non-observable
change in the distribution of starting materials/products which were left
standing at room temperature (~238C) over 24 h. The spectroscopic data
phenyl)-1-phenyl-ethene
and (E)-2-(4’-benzoylphenyl)-prop-2-enoic
[
52]
acid n-butyl ester.
Asymmetric intermolecular Heck arylation: Using a slightly modified
[
38,39]
procedure to that previously reported.
A Schlenk tube was charged
0
with [Pd (dba-4,4’-H) ] (3 mol% based on Pd) and (R)-BINAP
2
3
(3 mol%, 3.2 mmol) under inert atmosphere (N , back-filled three times),
2
followed by dioxane (2 mL). The mixture was stirred for one hour at
508C. After which time, iPr NEt, (117 mL, 0.642 mmol), PhOTf (36 mL,
2
0.212 mmol) and 2,3-dihydrofuran (81 mL, 1.07 mmol) were added. The
reaction was stirred at 508C under N2 for 7 d; monitored by HRGC anal-
ysis every 24 h (removal of 10 mL aliqouts). The product could be puri-
[
48]
for 4-methylbiphenyl (3) was in agreement with the literature.
fied by filtration through silica gel, washed with Et O (”3) and concen-
2
1
trated in vacuo to give the coupled product (>95% pure by HNMR
General procedure for Suzuki–Miyaura cross-coupling under ligand-free
conditions: 4-Fluorophenylboronic acid (0.41 mmol), 4-bromo-6-methyl-
spectroscopy). The spectroscopic data for the products is in agreement
[
38,39]
with the literature.
2
1
-pyrone (0.45 mmol), biphenyl (0.041 mmol, as the internal standard) in
m Na CO (1.0 mL) and THF (1.5 mL) were degassed via three freeze–
2
3
General procedure for the Buchwald–Hartwig amination reactions: Note
0
pump–thaw cycles. Then the Pd precursor complex (1 mol%) was added
under nitrogen gas. The resulting mixture was allowed to stir at 238C.
Analysis and quenching of the reaction was done in the same way as de-
tailed in procedure A. The spectroscopic data for 4-(4’-fluorophenyl)-6-
methyl-2-pyrone (10) was in agreement with the literature.
General procedure for a-arylation of esters: The procedure was slightly
that we did not use a dry-box for the preparation of these reactions as
[
45]
originally reported. We found that it was possible to conduct these re-
actions using Schlenk techniques providing that the following procedure
0
was followed: To a mixture of [Pd (dba-4,4’-H)
2
] (1 mol%), tBu
3
P·HBF
4
[
20]
(0.8 mol%) and tBuONa (1.5 mmol) in a Schlenk tube was added the
aryl bromide (1 mmol) under an inert atmosphere (N , back-filled three
2
[
32]
times), followed by dry degassed toluene (1.5 mL) via glass syringe (aryl
halide dissolves the plastic syringes). The mixture was allowed to stir for
15 min, and then hexadecane internal standard (0.1 mmol) and finally
aniline (1 mmol) was added (total pre-incubation time=0.25 h). The
progress of the reaction was monitored by HRGC. The product was puri-
fied by flash chromatography using ethyl acetate/hexane solvent mix-
tures. The spectroscopic data for all the amination products is in agree-
modified to that reported by Hartwig and co-workers: To a Schlenk
tube under an argon atmosphere, containing a magnetic stirrer bar, was
0
2
added 4 (1.0 mol%) and [Pd
(dba-4,4’-H)
3
] (0.5 mol%). LiHMDS (1m
solution in hexane) (2.3 mmol, 2.3 mL) was added to a separate reaction
flask containing tert-butyl acetate (1.1 mmol) which was externally cooled
to 08C. After 10 min, this solution was cannula transferred into the
0
Schlenk tube containing [Pd
2
(dba-4,4’-H)
3
]/4, and finally bromobenzene
[
45]
ment with the literature.
(
1.0 mmol) and hexadecane (0.1 mmol, as the internal standard) added.
The mixture was stirred at 258C for 15 h. Samples for HRGC were
quenched by addition of aqueous NH Cl and the organic components ex-
tracted into diethyl ether. The spectroscopic data for tert-butyl 2-phenyl-
4
[
32]
Acknowledgements
A
C
H
T
R
E
U
N
G
acetate 14 is in agreement with the literature.
General procedure for Heck arylation reactions (using Schlenk techni-
We are indebted to Professor Anny Jutand (Ecole Normale Supꢂrieure,
Paris, France) for many discussions regarding this and related studies. We
also thank Professor Stephen L. Buchwald (M.I.T., Cambridge, USA) for
discussions regarding dba effects in the aryl amination reactions. We ac-
knowledge Sigma–Aldrich for supporting this project in-kind (donation
of isotopically labelled compounds) and thank Johnson–Matthey for a
ques): The procedure was slightly modified to that reported by Fu and
[
33]
co-workers:
A Schlenk tube containing a magnetic stirrer bar was
CO (0.69 mmol), (tBu) P·HBF (0.022 mmol,
(dba-4,4’-H) ] (3 mol% based on Pd), under an atmos-
phere of argon. The pre-incubation time was 10 min. 4-Bromoanisole
0.63 mmol), n-butyl acrylate (1.2 mmol) and dihexyl ether (0.063 mmol,
charged with Cs
2
3
3
4
0
3
.6 mol%), [Pd
2
3
(
2
generous loan of PdCl . We thank the EPSRC (GR/S94926/01) for sup-
as the internal standard) were added by microsyringe, followed by dry
degassed dioxane (1.0 mL) via cannula transfer. The Schlenk tube was
sealed and heated to 1208C in an oil bath and stirred (600 rpm) for 12 h.
Analysis and quenching of the reaction was done in the same way as de-
tailed in procedure A. The spectroscopic data for (E)-butyl cinnamate 16
porting a post-doctoral research fellowship (G.P.M.). Miss Deborah Far-
rington is thanked for preliminary catalyst studies. I.J.S.F. thanks the
Royal Society for a University Research Fellowship and generous equip-
ment grant.
[
33]
is in agreement with the literature.
General procedure for Heck arylation reactions (for n-butyl acrylate): To
an automated synthesis robot (ASW 2000, Chemspeed), 32 double jack-
eted reaction vessels, each equipped with a condenser, were installed.
[1] J. Stetter, F. Lieb, Angew. Chem. 2000, 112, 1792–1812; Angew.
Chem. Int. Ed. 2000, 39, 1724–1744.
[2] a) A. Zapf, M. Beller, Chem. Commun. 2005, 431–440; b) I. J. S.
Fairlamb, Tetrahedron 2005, 61, 9661–9662.
Each vessel was charged with Cs
2
CO
3
(0.69 mmol), (tBu)
3 4
P·HBF
(
3.6 mol%), the vessels were closed and N
2
applied. The pre-incubation
[3] a) J. F. Hartwig, Acc. Chem. Res. 1998, 31, 852–860; b) J. F. Hartwig,
in Organopalladium Chemistry for Organic Synthesis, Vol. 1 (Ed.:
E.-i. Negishi), Wiley-Interscience, New York, 2002, p. 1051; c) B. H.
Yang, S. L. Buchwald, J. Organomet. Chem. 1999, 576, 125–146.
[4] a) N. J. Whitcombe, K. K. Hii, S. E. Gibson, Tetrahedron 2001, 57,
7449–7476; b) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000,
100, 3009–3066.
[5] a) K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975,
16, 4467–4470; b) R. R. Tykwinski, Angew. Chem. 2003, 115, 1604–
1606; Angew. Chem. Int. Ed. 2003, 42, 1566–1568; c) K. Sonoga-
shira, J. Organomet. Chem. 2002, 653, 46–49; d) K. Sonogashira, in
periods may vary by ca. 10 min between reactions. By using the fully au-
tomated syringe of the robot, each vessel was charged with dioxane
(
0
0.6 mL), 4-bromoanisole (0.63 mmol), [Pd
2
(dba-4,4’-H)
3
] as a solution in
dioxane (0.021m, 0.4 mL, 3 mol%; based on Pd), n-butyl acrylate
1.2 mmol) and dihexyl ether (0.063 mmol, as the internal standard). All
(
vessels were vortexed (at 600 rpm) and heated to 1208C for 15 h. At this
time, a sample from each vessel was removed (0.1 mL), which was diluted
with acetone, and analyzed by HRGC/MS (no further reaction was seen
in these samples at 258C over several hours by HRGC/MS analysis). An
identical procedure was conducted for the other alkenyl substrates, sty-
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8759

I. J. S. Fairlamb et al.
Organopalladium Chemistry for Organic Synthesis, Vol. 1 (Ed.: E.-i.
Negishi), Wiley-Interscience, New York, 2002, p. 493.
6] a) J. K. Stille, Angew. Chem. 1986, 98, 504–519; Angew. Chem. Int.
Ed. Engl. 1986, 25, 508–523; b) V. Farina, V. Krishnamurthy, W. J.
Scott, Org. React. 1997, 50, 1–120.
[24] a) K. Arentsen, S. Caddick, F. G. N. Cloke, A. P. Herring, P. B.
Hitchcock, Tetrahedron Lett. 2004, 45, 3511–3515; b) K. Arentsen,
S. Caddick, F. G. N. Cloke, Tetrahedron 2005, 61, 9710–9715.
[25] Selected reaction profiles can be found in ref. [20].
[
[
0
[26] It should be noted that we have not been able to prepare Pd com-
7] a) A. Suzuki, in Metal-Catalyzed Cross-Coupling Reactions (Eds.: F.
Diederich, P. J. Stang), VCH, Weinheim, 1998, p. 49; b) N. Miyaura,
A. Suzuki, Chem. Rev. 1995, 95, 2457–2483; c) A. Suzuki, J. Orga-
nomet. Chem. 1999, 576, 147–168; d) A. Zapf, in Transition Metals
for Organic Synthesis: Building Blocks and Fine Chemicals, Vol. 1
2
plexes containing the dba-4,4’-NO ligand for direct comparison (Pd
nanoparticles are produced exclusively).
[27] Y. Okamoto, H. C. Brown, J. Org. Chem. 1957, 22, 485–494.
[28] The phrase “ligand-free” is used when no additional ligand, such as
a phosphine, is added to the reaction, see: A. H. M. de Vries,
J. M. C. A. Mulders, J. H. M. Mommers, H. J. W. Henderickx, J. G.
de Vries, Org. Lett. 2003, 5, 3285–3288.
[29] We have employed this substrate for several cross-coupling process-
es, see: a) L. R. Marrison, J. M. Dickinson, I. J. S. Fairlamb, Bioorg.
Med. Chem. Lett. 2002, 12, 3509–3514; b) L. R. Marrison, J. M.
Dickinson, R. Ahmed, I. J. S. Fairlamb, Tetrahedron Lett. 2002, 43,
8853–8857; c) L. R. Marrison, J. M. Dickinson, R. Ahmed, I. J. S.
Fairlamb, Bioorg. Med. Chem. Lett. 2003, 13, 2667–2671; d) I. J. S.
Fairlamb, F.-J. Lu, J.-P. Schmidt, Synthesis 2003, 2564–2570; e) I. J. S.
Fairlamb, P. S. Bꢃuerlein, L. R. Marrison, J. M. Dickinson, Chem.
Commun. 2003, 632–633; f) E. H. Niemela, A. F. Lee, I. J. S. Fair-
lamb, Tetrahedron Lett. 2004, 45, 3593–3595; g) I. J. S. Fairlamb,
L. R. Marrison, J. Dickinson, F.-J. Lu, J.-P. Schmidt, Bioorg. Med.
(Eds.: M. Beller, C. Bolm), Wiley-VCH, Weinheim, 2004, p. 211.
[
8] A. F. Littke, G. C. Fu, Angew. Chem. 2002, 114, 4350–4386; Angew.
Chem. Int. Ed. 2002, 41, 4176–4211.
[
9] C. Amatore, A. Jutand, Coord. Chem. Rev. 1998, 178–180, 511–528,
and references therein.
0
0
[
10]
A
C
H
T
R
E
U
N
G
[Pd dba
3
]·CHCl
3
is commonly used as a precursor Pd catalyst. It is
0
important to note that CHCl
palladium cluster [Pd
3
reacts with [Pd dba
3
]/tBu
3
P to give a
3
4
A
H
E
N
(m -CH)
A
H
E
G
3 3 4
(m-Cl) (tBu P) ], an implication for
A
H
R
U
G
fair comparisons of new catalysts, see: a) A. D. Burrows, D. M. P.
Mingos, S. Menzer, R. Vilar, D. J. Williams, J. Chem. Soc. Dalton
Trans. 1995, 2107–2108; b) R. Vilar, S. E. Lawrence, D. M. P.
Mingos, D. J. Williams, Chem. Commun. 1997, 285–286; c) D. M. P.
Mingos, R. Vilar, J. Organomet. Chem. 1998, 557, 131–142.
[
11] Y. Takahashi, T. Ito, Y. Ishii, J. Chem. Soc. Chem. Commun. 1970,
065–1066. Various Pd/dba ratios are often quoted for these types
Chem. 2004, 12, 4285–4299.
[30] For a general report on the solution behaviour of Pd dba
0
1
3
com-
of complexes. It should be noted that in solution these complexes
are dimeric; the precise stoichiometry depends on whether the com-
plexes, see: H. Kawazura, H. Tanaka, K. Yamada, T. Takahashi, Y.
Ishii, Bull. Chem. Soc. Jpn. 1978, 51, 3466–3470.
plex has been recrystallized, to give [Pd
used as is, or vacuum dried to give [Pd dba
plex is [Pd dba ]·dba. This latter complex is commonly referred to as
Pddba , which is structurally incorrect based on previous solution H
and HNMR studies, see: R. Ukai, H. Kawazura, Y Ishii, J. J.
Bonnet, J. A. Ibers, J. Organomet. Chem. 1974, 65, 253–266; see
also ref. [30].
2
dba
3
]·solvent, and either
[31] a) D. A. Culkin, J. F. Hartwig, Acc. Chem. Res. 2003, 36, 234–245;
b) G. C. Lloyd-Jones, Angew. Chem. 2002, 114, 995–998; Angew.
Chem. Int. Ed. 2002, 41, 953–956.
[32] S. Lee, N. A. Beare, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123,
8410–8411; note that the saturated NHC variant of ligand 4 was
used in this report.
[33] A. F. littke, G. C. Fu, J. Org. Chem. 1999, 64, 10–11; note that aryl
chlorides and methyl acrylate were employed as substrates under
the reaction conditions described in this paper. The reason for em-
ploying aryl bromides was that significantly longer reaction times
were required for the less reactive aryl chlorides.
2
3
]. Otherwise, the com-
2
3
1
2
2
[
[
12] a) C. Amatore, A. Jutand, F. Khalil, M. A. M’Barki, L. Mottier, Or-
ganometallics 1993, 12, 3168–3178; b) C. Amatore, A. Jutand, G.
Meyer, Inorg. Chim. Acta 1998, 273, 76–84.
0
13] The importance of low ligated Pd complexes has been reviewed,
see: a) U. Christmann, R. Vilar, Angew. Chem. 2005, 117, 370–378;
[34] C. Amatore, E. Carrꢂ, A. Jutand, Y. Medjour, Organometallics 2002,
21, 4540–4545.
[35] C. Amatore, A. Jutand, J. Organomet. Chem. 1999, 576, 254–278.
[36] J. A. Casares, P. Espinet, G. Salas, Chem. Eur. J. 2002, 8, 4844–4853.
[37] C. E. Willans, J. M. C. A. Mulders, J. G. de Vries, A. H. M. de Vries,
J. Organomet. Chem. 2003, 687, 494–497.
[38] F. Ozawa, A. Kubo, T. Hayashi, J. Am. Chem. Soc. 1991, 113, 1417–
1419.
[39] N. G. Andersen, R. McDonald, B. A. Keay, Tetrahedron: Asymmetry
2001, 12, 263–269.
Angew. Chem. Int. Ed. 2005, 44, 366–374; for the in situ preparation
0
of mono-ligated “Ph
3
P-Pd ”, see: b) A. Beeby, S. Bettington, I. J. S.
Fairlamb, A. E. Goeta, A. R. Kapdi, A. L. Thompson, New J. Chem.
004, 28, 600–605.
n
14] It is well established that ligation of L Pd (n=1 or 2) by p-acidic li-
2
0
[
1
0
gands stabilises the electron-rich d palladium centre, see: J. J. M.
de Pater, D. S. Tromp, D. M. Tooke, A. L. Spek, B.-J. Deelman, G.
van Koten, C. J. Elsevier, Organometallics 2005, 24, 6411–6419, and
references therein.
[
15] A. A. Sabino, A. H. L. Machado, C. R. D. Correia, M. N. Eberlin,
Angew. Chem. 2004, 116, 2568–2572; Angew. Chem. Int. Ed. 2004,
[40] BINAP is 2,2’-bis(diphenylphosphino)-1,1’-binaphthalene; (R)-
BINAP was purchased from Strem Chemicals.
4
3, 2514–2518; corrigendum: A. A. Sabino, A. H. L. Machado,
[41] a) L. Jiang, S. L. Buchwald, in: Metal-Catalyzed Cross-Coupling Re-
actions (Eds.: A. de Meijere, F. Diederich), Wiley-VCH, Weinheim,
1998; b) B. Schlummer, U. Scholz, Adv. Synth. Catal. 2004, 346,
1599–1626; c) J. F. Hartwig, Acc. Chem. Res. 1998, 31, 852–860;
d) J. F. Hartwig, in Organopalladium Chemistry for Organic Synthe-
sis, Vol. 1 (Ed.: E.-i. Negishi), Wiley-Interscience, New York, 2002,
p. 1051; e) B. H. Yang, S. L. Buchwald, J. Organomet. Chem. 1999,
576, 125–146; f) S. L. Buchwald, C. Mauger, G. Mignani, U. Scholz,
Adv. Synth. Catal. 2006, 348, 23–39;
C. R. D. Correia, M. N. Eberlin, Angew. Chem. 2004, 116, 4489;
Angew. Chem. Int. Ed. 2004, 43, 4389.
[
[
[
16] R. Sustmann, J. Lau, M. Zipp, Tetrahedron 1986, 27, 5207–5210.
17] A. E. Jensen, P. Knochel, J. Org. Chem. 2002, 67, 79–85.
18] T. Yamamoto, A. Yamamoto, S. Ikeda, J. Am. Chem. Soc. 1971, 93,
3
350–3358.
[
[
19] F. Glorius, Angew. Chem. 2004, 116, 3444–3446; Angew. Chem. Int.
Ed. 2004, 43, 3364–3366.
20] I. J. S. Fairlamb, A. R. Kapdi, A. F. Lee, Org. Lett. 2004, 6, 4435–
[42] N. S. Kozlov, L. Pinegina, B. E. Tselishchev, Zh. Khim. 1968, Abstr.
No. 9, h230. We have been able to reproduce the reaction of aniline
with dba-4,4’-Hto give the bis-aminated 1,4-product.
4
438.
[
[
21] I. J. S. Fairlamb, A. R. Kapdi, A. C. Whitwood, unpublished results.
22] Y. Macꢂ, A. R. Kapdi, I. J. S. Fairlamb, A. Jutand, Organometallics
[43] S. Shekhar, P. Ryberg, J. F. Hartwig, J. S. Mathew, D. G. Blackmond,
E. R. Strieter, S. L. Buchwald, J. Am. Chem. Soc. 2006, 128, 3584–
3591.
2
006, 25, 1785–1800.
[
23] a) A. Scrivanti, V. Beghetto, U. Matteoli, S. Antonaroli, A. Marini,
0
F. Mandoj, R. Paolesse, B. Crociani, Tetrahedron Lett. 2004, 45,
[44] Oxidative additions of PhI on monodentate phosphine Pd com-
5
861–5864; b) A. Scrivanti, V. Beghetto, U. Matteoli, S. Antonaroli,
plexes usually involve dissociation of dba-4,4’-Hligand prior to oxi-
A. Marini, B. Crociani, Tetrahedron 2005, 61, 9752–9758.
dative addition, see ref. [35], and references therein.
8760
www.chemeurj.org
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8750 – 8761

Cross-Coupling Reactions
FULL PAPER
[
45] J. F. Hartwig, M. Kawatsura, S. I. Hauck, K. H. Shaugnessy, L. M.
Alcazar-Roman, J. Org. Chem. 1999, 64, 5575–5576.
[48] N. E. Leadbeater, S. Resouly, Tetrahedron 1999, 55, 11889–11894.
[49] R. Wang, B. Twamley, J. M. Shreeve, J. Org. Chem. 2006, 71, 426–
429.
[50] M. Gruber, S. Chouzier, K. Koehler, L. Djakovitch, Appl. Cat. A
Gen. 2004, 265, 161–169.
[51] D. Yang, Y. C. Chen, N. Y. Zhu, Org. Lett. 2004, 6, 1577–1580.
[52] M. Feuerstein, H. Doucet, M. Santelli, J. Org. Chem. 2001, 66,
5923–5925.
[
[
46] M. R. Netherton, G. C. Fu, Org. Lett. 2001, 3, 4295–4298.
47] In an elegant recent study, Nolan and co-workers have reported that
dramatic differences in catalytic activity are observed when subtle
structural changes are made to the terminal substituent in the allyl
II
scaffold of Pd precatalysts containing N-heterocyclic carbene li-
gands. This effect is associated with the ease of elimination of the al-
lylic moiety, which relates directly to catalyst activation, see: N.
Marion, O. Navarro, J. Mei, E. D. Stevens, N. M. Scott, S. P. Nolan,
J. Am. Chem. Soc. 2006, 128, 4101–4111.
Received: April 4, 2006
Published online: September 1, 2006
Chem. Eur. J. 2006, 12, 8750 – 8761
ꢀ 2006 Wiley-VCHVerlag GmbH& Co. KGaA, Weinheim
www.chemeurj.org
8761