Kinetic investigation of a PC(sp3)P pincer palladium (II) complex in the Heck reaction

Article abstract of DOI:10.1016/j.jorganchem.2005.06.014

An investigation of the kinetics of the Heck reaction between 4-iodoanisole and styrene catalysed by {cis-1,3-bis[(di-tert-butylphosphino)methyl]- cyclohexane} palladium (II) iodide (1) has been performed in DMF-d₇ solution. Based on mercury po

Full text of DOI:10.1016/j.jorganchem.2005.06.014

Journal of Organometallic Chemistry 690 (2005) 4197–4202
www.elsevier.com/locate/jorganchem
Note
Kinetic investigation of a PC(sp3)P pincer palladium (II) complex
in the Heck reaction
*
Patrik Nilsson, Ola F. Wendt
Department of Chemistry, Lund University, P.O. Box 124, S-22100 Lund, Sweden
Received 4 May 2005; received in revised form 12 June 2005; accepted 16 June 2005
Abstract
An investigation of the kinetics of the Heck reaction between 4-iodoanisole and styrene catalysed by {cis-1,3-bis[(di-tert-butyl-
phosphino)methyl]-cyclohexane} palladium (II) iodide (1) has been performed in DMF-d7 solution. Based on mercury poisoning
experiments a heterogeneous palladium catalyst formed from the PCsp3 P PdðIIÞ pre-catalyst is proposed. Saturation behaviour with
respect to the olefin concentration suggests a mechanism consisting of a pre-equilibrium association of the olefin followed by a rate
determining reaction with aryl halide. The equilibrium constant for the olefin association, K1, and the rate constant for the subse-
quent oxidative addition step, k2, were determined to (5.7 2.5) · 10ꢀ3 and 18.4 2.7 Mꢀ1 sꢀ1, respectively.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Palladium; Pincer complexes; Heck reaction; Kinetics; Heterogeneous catalysis
1. Introduction
Thus, they have been applied in a large number of inter-
esting reactions such as dehydrogenation [9], Karasch
addition [10,11], ketone reduction [12], asymmetric aldol
[13,14] and Heck reaction [8,15–18]. Recently, we also
reported on the application of PCP pincer complexes
in the Stille cross-coupling reaction [19].
In the large majority of cases the PCP backbone con-
sists of an aromatic ring and only a few complexes with
an aliphatic backbone, i.e., PCsp3 P pincer complexes,
have been reported. The catalytic activity of such
PCsp3 P pincer complexes has been investigated by Mil-
stein and co-workers [8] and in our laboratory [18].
The general conclusion has been that an increased elec-
tron density on the metal centre, caused by coordination
of an sp3 carbon instead of an sp2 carbon, increases the
catalytic activity [8,18,20].
Carbon–carbon bond coupling reactions have be-
come one of the most important tools in organic synthe-
sis [1]. Among these the palladium-catalysed vinylation
of aryl halides, the so-called Heck-reaction, occupies a
prominent position since it involves functionalisation
of a C–H bond [2].
Since the early work by Moulton and Shaw [3],
cyclometallated phosphine-based pincer ligands have
developed into powerful and important ligands for
organometallic catalysts in C–C bond forming reactions
[4–7]. In 1997, Milstein and co-workers [8] found that
PCP pincer complexes could be applied as catalysts in
the Heck reaction under aerobic conditions without deg-
radation of the catalyst. This observation has inspired
many others since pincer-based metal complexes seem
to strike a unique balance of stability vs. reactivity.
The mechanism of operation and the nature of the
ECE complexes in catalysis is the subject of some debate
[21]. Beside the classical Pd(0)/Pd(II) cycle [4,22–26]
several authors have reported the involvement of a
Pd(II)/Pd(IV) cycle [8,27,28]. Recently a study of a
*
Corresponding author. Tel.: +46 46 2228153; fax: +46 46 2224439.
E-mail address: ola.wendt@inorg.lu.se (O.F. Wendt).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.06.014

4198
P. Nilsson, O.F. Wendt / Journal of Organometallic Chemistry 690 (2005) 4197–4202
OCH
2.3. Preparation of trans-(4-methoxyphenyl)-2-
phenylethylene (2)
3
[1]
A J. Young NMR-tube was charged with 1.0 lL
(8.0 · 10ꢀ6 mol) 4-bromoanisole and 1.0 lL (8.7 ·
10ꢀ6 mol) styrene together with 1 mL DMF-d7. Approx-
imately one equivalent of complex 1 was dissolved in
0.500 mL DMF-d7 and this solution was added to the
NMR-tube. 9.7 mg (9.2 · 10ꢀ5 mol) of Na2CO3 was
added and the reaction solution was heated to 160 ꢁC.
After 2 h an iodo for bromo substitution in complex 1
(to give the corresponding bromo-complex) was com-
plete as indicated by 31P NMR spectroscopy. After
another 18 h at 160 ꢁC a total consumption of the or-
ganic starting material was seen and 1 mL 20% HCl
(aq) was added, leading to the formation of a white pre-
cipitatate. Extraction with diethylether and evaporation
under reduced pressure gave an off-white crystalline
product. The yield was 1.4 mg (6.6 · 10ꢀ6 mol, 83%)
and the NMR signals were in accordance with the liter-
+
I
OCH
3
NEt
Ph
3
Ph
(2)
Scheme 1.
ditopic double pincer palladacycle with SCS pincer
coordination was reported [29] and it was concluded
that the reaction most probably is homogeneous. On
the other hand, in our work on the Stille reaction [19],
we concluded that the reaction is heterogeneous and
that the PCP complexes only act as a source for catalyt-
ically active metallic palladium species. Eberhard
reached the same conclusion in his work on the Heck
reaction catalyzed by a series of PCP complexes with
phosphite functionalities [30].
Thus, there seems to be no unanimous picture of the
mechanistic role of pincer complexes in these reactions,
and we decided to investigate this feature, using the
PCsp3 P pincer complexes with their reported higher
activity and thermal robustness. Here, we report a
kinetic and mechanistic investigation of the catalytic
reaction between styrene and 4-iodoanisole in DMF-d7
solution using the PCsp3 P complex {cis-1,3-bis[(di-tert-
butylphosphino)methyl]cyclohexane}–palladium (II) io-
dide (1) as catalyst in the Heck reaction, cf. Scheme 1.
1
ature data for compound 2 [31]. H NMR (DMF-d7): d
3
3.84 (s, 3H, CH3), 7.00 (d, JH–H = 8.5 Hz, 2H, m-H,
Ph–OMe), 7.17 (d,3JH–H = 16.5 Hz, 1H, CH), 7.25
(t,3JH–H = 8.0 Hz, 1H, p-H, Ph), 7.27 (d,3JH–H
=
16.5 Hz, 1H, CH), 7.38 (t,3JH–H = 8.0 Hz, 2H, m-H,
Ph–OMe), 7.60 (d,3JH–H = 8.5 Hz, 2H, o-H, Ph), 7.61
3
(d, JH–H = 8.0 Hz, 2H, o-H, Ph).
2.4. Kinetic investigations
The kinetics of the catalytic reaction was studied
1
2. Experimental
using H NMR spectroscopy. All equipment was rinsed
with aqua regia prior to use. In a typical experiment, a J.
Young NMR-tube was loaded with 1, 4-iodoanisole,
styrene, Et3N, internal standard and solvent and placed
in an oil bath at 160 ꢁC. The reaction was monitored by
1H NMR spectroscopy. The product was not separated
and isolated, but characterized in situ. Ferrocene was
used as internal standard, and all reactions were studied
under pseudo-first-order conditions with an excess of the
olefin (0.05–1.05 mol dmꢀ3) with respect to the aryl io-
dide ((8.04–8.11) · 10ꢀ3 mol dmꢀ3). To assure catalytic
reaction conditions the concentration of the palladium
(II) complex was at least one order of magnitude less
((0.04–0.79) · 10ꢀ3 mol dmꢀ3) than that of styrene and
4-iodoanisole. Stock solutions of all reagents in
DMF-d7 were used, except for the Et3N, which was
administered as received. Inhibition experiments
were performed by adding approximately 0.1 g (5 ·
10ꢀ4 mol) of elemental mercury to the reaction mixture
[32,33]. In addition, experiments using cyclooctatetraene
(COT) as catalyst inhibitor were also performed [34].
The amount of COT was approximately in 5-fold excess
2.1. General procedures and materials
All experiments were carried out using standard high-
vacuum line or Schlenk techniques or in a glove box un-
der nitrogen. If nothing else is stated all commercially
available reagents were used as received from Aldrich.
Styrene was distilled from CaH2 and stored in the glove
box at ꢀ25 ꢁC to prevent polymerisation. The complex
{cis-1,3-bis[(di-tert-butylphosphino)methyl]cyclohex-
ane} palladium (II) iodide (1) was prepared as described
earlier [18]. All stock solutions were prepared and stored
under nitrogen.
2.2. NMR measurements
The NMR spectroscopic measurements were per-
formed in DMF-d7 unless otherwise stated. 1H, 13C
and 31P NMR spectra were recorded on a Varian Unity
INOVA 500 spectrometer working at 499.77 MHz (1H).
Chemical shifts are given in ppm downfield from TMS
using residual solvent peaks (1H-, 13C NMR) or
H3PO4 as reference. J. Young NMR-tubes were pur-
chased from J. Young (Scientific Glassware) Ltd.
(0.35 · 10ꢀ3 mol dmꢀ3
) relative to the amount of
the catalyst (0.07 · 10ꢀ3 mol dmꢀ3), and was added at
the start of the reaction. The 1H NMR signals were

P. Nilsson, O.F. Wendt / Journal of Organometallic Chemistry 690 (2005) 4197–4202
4199
integrated generating kinetic traces, which were fitted to
single exponentials using the KaleidaGraph software.
ganic bases as superior, but for the present reaction in
DMF, we found that triethylamine was the base of
choice, due to the higher reactivity observed and the
advantage of having a homogeneous reaction mixture
in the kinetic investigations.
3. Results and discussion
In no case was any reaction observed when 1 and
either of the substrates were mixed separately, i.e., both
substrates are required to activate the catalytic process.
Finally it was noted that these reactions require an inert
atmosphere; under ambient conditions the catalytic
activity disappeared, most probably because of decom-
position of the catalyst.
3.1. General observations
The b-substituted styrene 2 was isolated in 83% yield
in the coupling reaction between 4-bromoanisole and
styrene using 1 as catalyst. The only product observed
was the E-isomer, and there was no indication of any
other regio- or stereo isomer. We have earlier reported
that the coupling between PhI and styrene catalysed
by 1 (as the trifluoroacetate complex) has a regioselectiv-
ity of approximately 90% in favour of the E-isomer. The
NMR data of 2 reported in the literature are rather brief
and in agreement with our findings [31]. We were able to
3.2. Kinetics and mechanism
All kinetic experiments were performed with 4-iodo-
anisole and styrene in large excess compared to the pal-
ladium precursor at 160 ꢁC, cf. Scheme 1. The choice of
substrate enables the monitoring of the reaction by
1
assign all peaks in the H NMR spectrum and the trans
1
configuration is most clearly seen in the large olefinic
coupling constant (16.5 Hz), vide supra.
means of H NMR spectroscopy. Thus, the intensity
of the strong singlet at d 3.79, belonging to the methoxy
group of 4-iodoanisole, decreased as a new downfield (d
3.84) peak grew in with the formation of the reaction
product 2. Styrene was always used in at least ten-fold
excess over aryl halide and under these conditions the
reaction is finished within 15 h. The concentration of
both styrene and complex 1 was varied and the kinetic
traces were fitted to single exponentials, cf. Fig. 1, show-
ing that the reaction is first order in aryl halide and giv-
ing observed rate constants as a function of styrene and
palladium concentrations, cf. Figs. 2 and 3. All observed
pseudo first-order rate constants are given in Tables 1
and 2. Using the aryl halide in excess over styrene gave
very low conversion and as seen from Fig. 1(a) conver-
sion is far from 100% at lower styrene concentrations.
The addition of Hg(l) to the reaction mixture resulted
in almost total loss of catalytic activity. Hence, less than
20% of the 4-iodoanisole was converted to (4-methoxy-
phenyl)-2-phenylethylene after 8 h. Mercury is known to
The fate of complex 1 in the reaction above was fol-
lowed by 31P NMR spectroscopy. Thus, the single peak
(d 76.8) from 1 was totally converted to a new single
peak at higher field (d 75.0) within 3 h. No further
change was observed when the reaction was complete,
and we interpret this as an iodo for bromo substitution
at the metal centre. As expected, no such change of the
31P NMR peak was seen in the kinetic investigation,
where an aryl iodide was employed.
In order to find proper reaction conditions for the ki-
netic measurements different substrates and bases were
considered. A qualitative investigation of the reaction
conditions clearly shows that 4-iodoanisole is a more
reactive substrate than 4-bromoanisole, which is in line
with earlier results [35]. The influence from the use of
different bases is also noteworthy, but the results were
somewhat unexpected in view of what has been reported
[18,36,37]. In general, the literature results point to inor-
b
a
3
2
1
0
3
2
1
0
4×104
time/s
0
8×104
(s) as
0
8000
time/s
1.6×104
Fig. 1. Observed intensities of 4-iodoanisole (h) and
2
a ;
function of time at 160 ꢁC when: (a) [styrene] = 0.05 mol dmꢀ3
(b) [styrene] = 1.05 mol dmꢀ3. General reaction conditions were: [Pd] = 0.07 · 10ꢀ3 mol dmꢀ3; [4-iodoanisole] = 8.04 · 10ꢀ3 mol dmꢀ3; [ferro-
cene] = 2.76 · 10ꢀ3 mol dmꢀ3. Solid lines denote the best fit to a single exponential using KaleidaGraph.

4200
P. Nilsson, O.F. Wendt / Journal of Organometallic Chemistry 690 (2005) 4197–4202
Table 2
2
1.5
1
Reaction conditions and rate constants at 160 ꢁC in DMF-d7 solventa
103/M
104k1/sꢀ1
[Pd]
[Ar–X]b
[Olefin]c
[Base]d
Reactantb
Producte
0.04
0.07
0.39
0.79
8.04
8.04
8.04
8.04
524
524
524
524
144
144
144
144
10.4 1.2
8.58 0.49
9.84 0.31
11.1 2.6
7.83 1.66
7.55 1.35
25 10
31
3
0.5
0
Both constants for the reactant decay and the product formation are
displayed.
a
[Ferrocene] = 2.32 mM in all measurements.
Ar–X = reactant = 4-iodoanisole.
Olefin = styrene.
Base = NEt3.
b
0
0.4
0.8
1.2
c
[styrene]/mol dm -3
d
e
Fig. 2. Observed pseudo-first-order rate constants for the reaction in
Scheme 1 as a function of [styrene] in DMF-d7 solution at 160 ꢁC. The
values are reported as means of the reactant decay and the product
formation. Complete data and further reaction conditions are reported
Table 1. The solid line denotes the best fit to Eq. (1), where
Product = trans-(4-methoxyphenyl)-2-phenylethylene.
amalgamate Pd colloid [33] and to verify the Hg(l) poi-
soning experiments, also cross-experiments with addi-
tion of cyclooctatetraene (COT) were performed.
Agreement between the two tests provides a good indi-
cation that the results are valid [34]. Even though we
see a substantial inhibition with COT (cf. Fig. 4) we
interpret our results in favour of a heterogeneous rather
than a homogenous catalytic process. It is known that
COT can inhibit also heterogeneous catalysts (in this
case presumably by competing with styrene for active
sites) but any homogeneous palladium(0) compounds
are expected to react with COT and become completely
inactive [32]. Complete inhibition by 2 equiv. of a 1,4
diene has been reported for an aromatic PCP-complex
in the Heck reaction, but in that case no cross-experi-
ment with mercury was performed [38]. Recently, we
reported a mechanistic investigation on the Stille cross-
coupling that pointed strongly to a heterogeneous sys-
tem where the palladium particles were formed from
an aromatic PCP–Pd(II) pre-catalyst [19]. The observa-
tions on the current Heck system speak in favour of a
K1 = (5.7 2.5) · 10ꢀ3 Mꢀ1, and k2 = 18.4 2.7 Mꢀ1 sꢀ1
.
3
2
1
0
0
0.4
0.8
103 [Pd]/mol dm-3
Fig. 3. Observed pseudo-first-order rate constants for the reaction in
Scheme 1 as a function of [Pd] in DMF-d7 solution at 160 ꢁC. The
values are reported as means for the reactant decay and the product
formation. Complete data and further reaction conditions are reported
in Table 2.
3
2
1
0
Table 1
Reaction conditions and rate constants at 160 ꢁC in DMF-d7 solventa
103/M
104k1/sꢀ1
[Pd]
[Ar–X]b
[Olefin]c
[Base]d
Reactantb
Producte
0.07
0.07
0.07
0.07
0.07
8.04
8.04
8.04
8.04
8.04
50.0
105
210
524
1050
144
144
144
144
144
2.75 0.83
9.27 3.26
4.66 0.51
8.58 0.49
11.1 0.6
0.90 0.18
1.89 0.28
6.37 0.65
7.83 1.66
12.5 1.1
0
1.5×104
time/s
3×104
Both constants for the reactant decay and the product formation are
displayed.
a
[Ferrocene] = 2.76 mM in all measurements.
Ar–X = reactant = 4-iodoanisole.
Olefin = styrene.
Base = NEt3.
Fig. 4. Observed intensities of 4-iodoanisole as a function of time at
160 ꢁC in DMF-d7 with addition of Hg(l) (s), with addition of COT
(h) and without any additive (e). Reaction conditions:
[Pd] = 0.07 · 10ꢀ3 mol dmꢀ3; [styrene] = 0.524 mol dmꢀ3; [4-iodoani-
b
c
d
e
Product = trans-(4-methoxyphenyl)-2-phenylethylene.
sole] = 8.04 · 10ꢀ3 mol dmꢀ3; [ferrocene] = 2.76 · 10ꢀ3 mol dmꢀ3
.

P. Nilsson, O.F. Wendt / Journal of Organometallic Chemistry 690 (2005) 4197–4202
4201
similar interpretation and a heterogeneous form of pal-
ladium as the active catalyst seems most reasonable.
From Fig. 3, we can see that although the errors are
large the reaction rate is higher at high [Pd] indicating a
dependence in palladium that is probably first order
(vide infra). A heterogeneous catalytic cycle requires
activation of the pre-catalyst and we know that this hap-
pens only to a small extent; almost all of the PCP com-
plex is recovered after the reaction is over. Since the
initiation mechanism is currently unknown it cannot
be excluded that it is a varying degree of initiation from
run to run that gives rise to the kind of results observed
in Fig. 3. This deviation from linearity could then be
taken as support for a heterogeneous catalysis. Fig. 2
displays smaller errors and there is a saturation behav-
iour with respect to styrene concentration. Assuming a
heterogeneous catalytic system, a mechanism as shown
in Scheme 2 can be proposed. A fast pre-equilibrium fol-
lowed by an irreversible reaction with aryl halide will
give the following rate law:
Another possibility is that saturation of the active
sites with olefin prevents the catalyst from aggregating
and thus becoming inactive. It should be noted,
though, that in no cases do we see any formation of
palladium black.
It is clear that the kinetics (the aryl halide and olefin
concentration dependencies) are completely compatible
with a homogeneous catalytic cycle with initial revers-
ible olefin activation, as proposed by Jensen. However,
it fails to explain the inhibition experiments with mer-
cury and the erratic dependence on palladium concen-
tration. Also, contrary to our observation, in a
homogeneous scheme it is expected that 1 would react
with styrene in a stoichiometric reaction since we know
from the kinetics that this equilibrium can be pushed to
the right.
Eberhard has suggested that the decreased electronic
density on the metal centre expected in phosphinite PCP
complexes (compared to phosphines) will cause the com-
plex to decompose more easily and therefore lead to a
more active source of colloidal palladium [30]. In the
current reaction system, the introduction of an aliphatic
PCP ligand induces the opposite effect, i.e., the sp3 car-
bon donates electrons more readily towards the metal
centre as compared to the corresponding aromatic ones.
Still, the results in [18] suggest that we get a more active
catalyst compared to the aromatic phosphine system[8]
(but less active than the phosphinites [16,30]). It is thus
clear that we do not understand the initiation process of
these complexes, but it seems that less robust com-
pounds, such as 1 (which is more active under inert reac-
tion conditions) or the phosphinite complexes, give
more efficient catalytic process.
In conclusion we propose that 1 serves as a Pd(II)
pre-catalyst forming small amounts of a catalytically ac-
tive colloidal Pd species, although we have no direct evi-
dence as to the nature of the active catalyst. The kinetics
speak in favour of a heterogeneous mechanism consist-
ing of a pre-equilibrium step involving association of
the olefin followed by a rate determining oxidative addi-
tion step of the aryl halide.
K1k2½pdꢁ½styreneꢁ
rate ¼ kobs½ArXꢁ; kobs
¼
.
ð1Þ
1 þ K1½styreneꢁ
In comparison to the classical Heck mechanism this
means that we propose a reversed order of entry of the
substrates, where the olefin is activated before the aryl
halide [37]. If oxidative addition were to precede the ole-
fin coordination this could not explain the observed sat-
uration. Also, this is not a completely new notion and
has been proposed by both Shaw and Jensen in the con-
text of Pd(II)/Pd(IV) cycles [16,27].
The simple catalytic cycle cannot explain the inhibi-
tion (incomplete conversion) seen in runs where the
styrene concentration was low. This inhibition might
be explained by a competitive non-productive binding
of aryl halide at higher aryl halide/styrene ratios.
inhibition
high [ArX]
low [styrene]
Pd
HX
Acknowledgements
Ph
K1
We thank Ms. Ida Truedsson for helpful assistance in
the laboratory. Financial support from the Swedish Re-
search Council, the Crafoord Foundation, the Knut and
Alice Wallenberg Foundation and the Royal Physio-
graphic Society in Lund is gratefully acknowledged.
PtBu2
Pd I
PtBu2
(1)
Ph
fast
Pd
Ph
Ph
k2
Ar
X
Ar
References
Ar X
Pd
[1] F. Diederich, P.J. Stang (Eds.), Metal Catalysed Cross-coupling
Reactions, Wiley-VCH, New York, 1998.
Scheme 2.

4202
P. Nilsson, O.F. Wendt / Journal of Organometallic Chemistry 690 (2005) 4197–4202
[2] R.F. Heck, Comprehensive Organic Synthesis 1991, fourth ed.,
Pergamon Press, Oxford, United Kingdom, 1991, p. 833–863.
[3] C.J. Moulton, B.L. Shaw, J Chem. Soc., Dalton Trans. (1976)
1020.
[20] S. Gibson, D.F. Foster, G.R. Eastman, R.P. Tooze, D.J. Cole-
Hamilton, Chem. Commun. (2001) 779.
[21] N.J. Whitcombe, K.K. Hii, S.E. Gibson, Tetrahedron 57 (2001)
7449.
[4] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.
[5] A.F. Littke, G.C. Fu, Angew. Chem. Int. Ed. 41 (2002) 4176.
[6] R.H. Crabtree, Chem. Rev. 85 (1985) 245.
[7] M.E. van der Boom, D. Milstein, Chem. Rev. 103 (2003) 1759.
[8] M. Ohff, A. Ohff, M.E. van der Boom, D. Milstein, J. Am. Chem.
Soc. 119 (1997) 11687.
[22] R.B. Bedford, Chem. Commun. (2003) 1787.
[23] I.P. Beletskaya, A.N. Kashin, N.B. Karlstedt, A.V. Mitin, A.V.
Cheprakov, G.M. Kazankov, J. Organomet. Chem. 622 (2001) 89.
[24] C.S. Consorti, M.L. Zanini, S. Leal, G. Ebeling, J. Dupont, Org.
Lett. 5 (2003) 983.
[25] C. Rocaboy, J.A. Gladysz, New J. Chem. 27 (2003) 39.
[26] (a) J. Dupont, M. Pfeffer, J. Spencer, Eur. J. Inorg. Chem. (2001)
1917;
[9] M. Gupta, C. Hagen, W.C. Kaska, R.E. Cramer, C.M. Jensen, J.
Am. Chem. Soc. 119 (1997) 840.
[10] R.A. Gossage, L.A. van de Kuil, G. van Koten, Acc. Chem. Res.
31 (1998) 423.
(b) K. Yu, W. Sommer, M. Weck, C.W. Jones, J. Catal. 226
(2004) 101;
[11] L.A. van de Kuil, D.M. Grove, R.A. Gossage, J.W. Zwikker,
L.W. Jenneskens, W. Drenth, G. van Koten, Organometallics 16
(1997) 4985.
[12] P. Dani, T. Karlen, R.A. Gossage, S. Gladiali, G. van Koten,
Angew. Chem., Int. Ed. 39 (2000) 743.
(c) K. Yu, W. Sommer, J.M. Richardson, M. Weck, C.W. Jones,
Adv. Synth. Catal. 347 (2005) 161.
[27] B.L. Shaw, S.D. Perera, E.A. Staley, Chem. Commun. (1998)
1361.
[28] B.L. Shaw, New J. Chem. (1998) 77.
[13] J.M. Longmire, X. Zhang, M. Shang, Organometallics 17 (1998)
4374.
[29] M.A. Hossain, S. Lucarini, D. Powell, K. Bowman-James, Inorg.
Chem. 43 (2004) 7275.
[14] F. Gorla, A. Togni, L.M. Venanzi, A. Albinati, F. Lianza,
Organometallics 13 (1994) 1607.
[15] D. Morales-Morales, C. Grause, K. Kasaoka, R. Redon, R.E.
Cramer, C.M. Jensen, Inorg. Chim. Acta 300–302 (2000) 958.
[16] D. Morales-Morales, R. Redon, C. Yung, C.M. Jensen, Chem.
Commun. (2000) 1619.
[17] I.P. Beletskaya, A.V. Chuchurjukin, H.P. Dijkstra, G.P.M. van
Klink, G. van Koten, Tetrahedron Lett. 41 (2000) 1075.
[18] S. Sjo¨vall, O.F. Wendt, C. Andersson, J. Chem. Soc., Dalton
Trans. (2002) 1396.
[19] D. Olsson, P. Nilsson, M. El Masnaouy, O.F. Wendt, Dalton
Trans. (2005) 1924.
[30] M.R. Eberhard, Org. Lett. 6 (2004) 2125.
[31] T. Majima, S. Tojo, A. Ishida, S. Takamuku, J. Org. Chem. 61
(1996) 7793.
[32] D.R. Anton, R.H. Crabtree, Organometallics 2 (1983) 855.
[33] C. Paal, W. Hartmann, Ber. Dtsch. Chem. Ges. 51 (1918) 711.
[34] T. Malmstro¨m, C. Andersson, J. Mol. Catal. 116 (1997) 237.
[35] A. Jutand, A. Mosleh, Organometallics 14 (1995) 1810.
[36] W. Cabri, I. Candiani, S. DeBernardinis, F. Francalanci, S.
Penco, R. Santi, J. Org. Chem. 56 (1991) 5796.
[37] C. Amatore, A. Jutand, Acc. Chem. Res. 33 (2000) 314.
[38] K. Kiewel, Y. Liu, D.E. Bergbreiter, G.A. Sulikowski, Tetrahe-
dron Lett. 40 (1999) 8945.