Synthesis and structural features of α-acyloxy-(η3-allyl)palladium complexes

Article abstract of DOI:10.1016/j.jorganchem.2006.05.013

α-Acetoxy (η³-allyl)palladium complexes were prepared from acyloxy functionalized allylsilanes under mild conditions and in good isolated yields. The substituent and ligand effects of the acetoxy group on the palladium-allyl bonding were studied by X-ray diffraction. These studies show that the acetoxy group generates a strongly deformed bonding between the metal atom and the allyl moiety. This unsymmetrical bonding is modulated by the σ-donor/π-acceptor properties of the ligands. The ¹³C NMR studies indicated that the shift values correlate with the carbon-palladium bond lengths and the inductive effects of the acetoxy group.

Full text of DOI:10.1016/j.jorganchem.2006.05.013

Journal of Organometallic Chemistry 691 (2006) 3640–3645
www.elsevier.com/locate/jorganchem
Synthesis and structural features of
a-acyloxy-(g3-allyl)palladium complexes
a
b
a,
*
´
´
´
Johan Kjellgren , Mikael Kritikos , Kalman J. Szabo
a
Stockholm University, Arrhenius Laboratory, Department of Organic Chemistry, SE-106 91 Stockholm, Sweden
b
Biovitrum, Preclinical R&D, SE-112 76 Stockholm, Sweden
Received 10 February 2006; received in revised form 5 May 2006; accepted 11 May 2006
Available online 20 May 2006
Abstract
a-Acetoxy (g3-allyl)palladium complexes were prepared from acyloxy functionalized allylsilanes under mild conditions and in good
isolated yields. The substituent and ligand effects of the acetoxy group on the palladium–allyl bonding were studied by X-ray diffraction.
These studies show that the acetoxy group generates a strongly deformed bonding between the metal atom and the allyl moiety. This
unsymmetrical bonding is modulated by the r-donor/p-acceptor properties of the ligands. The 13C NMR studies indicated that the shift
values correlate with the carbon–palladium bond lengths and the inductive effects of the acetoxy group.
Ó 2006 Elsevier B.V. All rights reserved.
Keywords: Allyl; Palladium, Ligand effects; Carbon–metal bonding; X-ray structure
1. Introduction
structure of (g3-allyl)palladium complexes is indispensable
for development of new selective palladium-catalyzed
transformations.
Allylpalladium chemistry offers efficient and selective
preparative methods for synthesis of densely functionalized
allylic products [1–4]. In these reactions allylpalladium
complexes occur as catalytic intermediates, which are usu-
ally generated by the displacement of the allylic leaving
group (e.g. acetate, carbonate, carbamate or halide) by a
palladium(0) catalyst [1–8]. Alternatively, allylpalladium
intermediates can be formed by reaction of palladium(II)
catalysts with alkenes [9–13], dienes [14–17] and allylsilanes
[18–27]. Under appropriate conditions the resulted (g3-
allyl)palladium intermediates react with a wide range of
nucleophiles, such as malonates, enolates, and different
N- and O-nucleophiles [1–4]. The regiochemistry of the
nucleophilic attack on the (g3-allyl)palladium complexes
is mainly determined by the electronic and steric effects
of the allylic substituents [1–8,28,29]. Therefore, explicit
knowledge on the effects of the allylic substituents on the
In this paper we report our recent results on preparation
and structural studies of a-acyloxy (g3-allyl)palladium
complexes (1a–g). These types of complexes are reaction
intermediates in important palladium-catalyzed allylic sub-
stitution reactions [30–34], in which the regioselectivity of
the reaction is controlled by the electronic effects of the
acyloxy substituent. Interestingly, the nucleophilic attack
on a-acyloxy (g3-allyl)palladium intermediates usually
takes place at the acyloxy substituted carbon to give the
branched allylic product [30–33], while the regioselectivity
of the catalytic transformations via b-acyloxy (g3-allyl)pal-
ladium intermediates is reversed to give the linear allylic
product [14,28,35–38]. Although, several studies have
appeared on the b-substituent effects in (g3-allyl)palladium
complexes, the a-substituent effects received somewhat less
attention. Therefore, in the present study we concentrated
to the substituent effects of the acyloxy substituent on the
structure of (g3-allyl)palladium complexes using X-ray
crystallography and 13C NMR spectroscopy.
*
Corresponding author. Tel.: +46 86747485; fax: +46 8155908.
´
E-mail address: kalman@organ.su.se (K.J. Szabo).
0022-328X/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.05.013

J. Kjellgren et al. / Journal of Organometallic Chemistry 691 (2006) 3640–3645
3641
2. Results and discussion
mers probably differ only in the configuration of their
chloro bridges [44]. Phosphine complex 1b was prepared
from 1a by exchange of the chloro ligand to dppe using
AgBF4 (entry 2). The obtained complexes 1a–g proved to
be air- and thermo-stable, and therefore they could be puri-
fied by column chromatography. The solubility of chloro
complexes 1a and 1c–g is relatively low in common organic
solvents, and therefore their 13C NMR spectrum was
recorded in DMSO-d6. On the other hand, phosphine com-
plex 1b is easily soluble in most organic solvents, however,
it was decomposed in DMSO, and therefore a CDCl3 solu-
tion was used for determination of the NMR spectrum.
Comparison of the X-ray structure of 1a and 1b. In
order to determine the substituent effects of the acetoxy
substituent on the structure of chloride and dppe ligated
(g3-allyl)palladium complexes, we carried out single-crys-
tal X-ray diffraction measurements. In both complexes the
acetoxy group is co-planar with the allyl-plane (Fig. 1)
indicating a conjugative interaction between the acetoxy
group and the p-system of the allyl moiety. Inspection
of Fig. 1 reveals that the carbon–palladium bonds are sys-
tematically shorter in 1a than in 1b. This is a well-known
structural effect of the r-donor chloride anion on the
bonding structure of (g3-allyl)palladium complexes
As mentioned above there are two main strategies for
preparation of functionalized (g3-allyl)palladium com-
plexes by allylic displacement of the appropriate leaving
group by palladium. The first method is based on the reac-
tion of palladium(0) complexes with allyl acetates or chlo-
˚
rides [39,40] This method was employed by Akermark and
co-workers [41] to synthesize an a-acetoxy (g3-allyl)palla-
dium complex, which has been the only one described in
the literature so far. An alternative method for synthesis
of (g3-allyl)palladium complexes is based on use of palla-
dium(II) sources and functionalized allylsilanes [20,23,42].
We employed this latter method for synthesis of a-acyloxy
(g3-allyl)palladium complexes because of its efficiency and
high functional group tolerance. Accordingly, various allyl-
silanes 2a–2f, prepared using the method reported by
Panek and Sparks [43], were reacted with Li2[PdCl4] (3)
in THF to obtain chloro-dimer complexes 1a and 1c–g in
good yield (Scheme 1 and Table 1). The palladadesilylation
reaction of allylsilanes 2a–b and 2d–f was accomplished in
2–3 h under mild conditions. However, in the presence of
an electron supplying p-methoxy-benzoyl group (2c) for-
mation of the corresponding (g3-allyl)palladium complex
(1d) is relatively slow (entry 4). Interestingly, purification
of the crude-product obtained from the reaction of 2a
and 3 resulted in two isomeric forms in a ratio of 2.4–1.
As the 1H and 13C NMR spectrum of these forms are iden-
tical within 0.1 ppm and 0.9 ppm, respectively, the two iso-
˚
[37,45,46]. Interestingly, in 1a the Pd–C3 bond (2.137 A)
˚
˚
is longer than the Pd–C1 bond (2.074 A) by 0.06 A. On
the contrary, in phosphine complex 1b the Pd–C3 bond
˚
˚
(2.156 A) is shorter than the Pd–C1 bond (2.212 A) by
˚
0.06 A, and thus the Pd–C1 bond in 1b is longer than
the corresponding palladium–carbon bond in 1a by
˚
0.14 A. As a consequence of the unsymmetrical palla-
dium–allyl bonding in 1a and 1b, the oxygen atom of
the acetoxy group (O1) is much closer to palladium in
˚
˚
1a (2.963 A) than in 1b (3.205 A). The above described
perturbation of the acyloxy group on the palladium–allyl
bonding is very similar to the substituent effects of other
functionalities [28,37,38,45–47], however, the magnitude
observed for 1a and 1b is clearly the largest.
Scheme 1.
˚
Fig. 1. X-ray structure of Complexes 1a and 1b. Selected bond lengths (A) for 1a: Pd–C1, 2.074(5); Pd–C2, 2.079(4); Pd–C3, 2.137(4); C1–C2, 1.383(7);
C2–C3, 1.359(7); C1–O1, 1.405(8); C4–O1, 1.345(6); C4–O2, 1.194(5); Pd–Cl1, 2.4301(14); Pd–Cl1i, 2.3973(16) (symmetry code i = Àx, 1 À y, 1 À z).
˚
Selected bond lengths (A) for 1b: Pd–C1, 2.212(6); Pd–C2, 2.158(7); Pd–C3, 2.156(7); C1–C2, 1.390(8); C2–C3, 1.375(10); C1–O1, 1.410(8); C4–O1,
1.362(8); C4–O2, 1.177(9); Pd–P1, 2.2951(16); Pd–P2, 2.3066(18).

3642
J. Kjellgren et al. / Journal of Organometallic Chemistry 691 (2006) 3640–3645
Table 1
Synthesis of a-acyloxy (g3-allyl)palladium complexes
a
Method A: The reactions were conducted in THF using Li2[PdCl4] (3) as palladium source. Method B: The chloride to dppe ligand exchange was
mediated by AgBF4 in CHCl3.
b
Reaction temperature/reaction time (°C)/(h).
Isolated yield.
c
13C NMR shifts of the allylic carbons. It is well known
[48,49] that 13C NMR shift values of the allylic carbons
are strongly influenced by the substituent and ligand
effects. Accordingly, the acetoxy substituted C1 carbon in
1b (108.9 ppm) is less shielded (Table 2) than the corre-
sponding carbon in 1a (105.4 ppm). This effect is probably
due to the longer Pd–C1 bond in 1b than in 1a (Fig. 1),
which leads to less shielding by the electropositive palla-
dium on C1 in 1b than in 1a. Varying the acyl substituent
(Q) has a relatively weak effect on the chemical shift of C1.
The methoxy substitution of the benzoyl group has a
small but significant effect, as C1 in the methoxy benzoyl
derivative 1d (105.8 ppm) is somewhat more shielded than
the C1 in 1c (106.1 ppm). There is large difference (typically
43–44 ppm) in the shift values of C1 and C3 carbons,
because of the inductive effects of the acyloxy substituents.
Interestingly, substitution of C2 (1f) leads to a decrease of
the difference of the shift values of the allylic terminal car-
bons (37 ppm). It is particularly interesting to compare the
substituent effects of the methyl and acetoxy groups on the
terminal carbons in 1g. The shielding value at the methyl
substituted allylic terminus (C3) is 81.9 ppm, which is very
similar to the shift value observed for the simple methyl–
allyl complex analog [48] (81.5 ppm). On the other hand,

J. Kjellgren et al. / Journal of Organometallic Chemistry 691 (2006) 3640–3645
3643
Table 2
was dissolved in dry THF (5 mL) at 20 °C under argon
atmosphere, and then Li2PdCl4 (130 mg, 0.5 mmol) was
added. The resulting solution was stirred for the time given
in Table 1, and then the solvent was evaporated and the
crude product was purified by silica-gel chromatography
affording the products as pale-yellow powders. The
NMR-spectra of the products were recorded in DMSO-d6
except for complex 1b which was recorded in CDCl3.
(g3-1-Acetoxy-allyl)palladium chloride dimer (1a). This
complex was prepared according to method A. The crude
product was purified by silica-chromatography using
CHCl3 as eluent. On the column, two well separated yellow
bands were observed (fraction A and fraction B). The iso-
lated product ratio between fraction A and B was found to
be 2.4:1. NMR data for fraction A: 1H NMR: 2.04 (s, 3H),
3.63 (bs, 2H), 5.98 (q, 1H, J = 9.8, 14.0 Hz), 6.98 (d, 1H,
J = 9.0). 13C NMR: 21.71, 61.66, 104.14, 105.38, 168.47.
13C NMR shifts (d) of the allylic carbons (ppm)a
C1
C2
C3
1a
1b
1c
1d
1e
1f
1g
a
105.4
108.9
106.1
105.8
106.0
100.8
106.9
104.1
110.3
103.9
104.4
105.7
120.6
100.1
61.7
61.6
61.5
61.2
61.3
63.1
81.9
The spectrum of 1b is recorded in CDCl3, while rest of the spectra are
recorded in DMSO-d6.
the acetoxy substituted C1 (106.9 ppm) in 1g is much more
deshilded than C3 (81.9 ppm) indicating that the inductive
effect of the acetoxy group is much stronger than that of
the methyl group in the allyl moiety. This inductive effect
may explain the preferential attack at the acetoxy substi-
tuted carbon in palladium-catalyzed allylic substitution
reactions proceeding via a-acetoxy substituted (g3-
allyl)palladium complexes.
1
NMR data for fraction B: H NMR: 2.13 (s, 3H), 3.64
(bs, 2H), 6.05 (q, 1H, J = 9.9, 14.9), 7.04 (d, 1H, J =
9.1). 13C NMR: 21.70, 61.29, 103.86, 105.06, 168.47. Prep-
aration of crystals for X-ray diffraction: 20 mg of 1a (from
fraction A) was dissolved in 0.5 mL CH2Cl2 followed by
layering pentane (0.25 mL) on the surface of this solution;
and the resulting mixture was kept at 4 °C affording crys-
tals suitable for X-ray diffraction.
3. Conclusions
a-Acyloxy substituted (g3-allyl)palladium complexes (1)
can be simply and efficiently prepared by palladadesilyla-
tion of acetoxy allyl silanes. The structural analysis of 1a
and 1b clearly shows that the acetoxy substituent has a pro-
nounced substituent effect on the allyl moiety. In chloro-
complex 1a, the acetoxy substituted carbon (C1) interacts
stronger with palladium than the other terminal carbon
(C3). The ligand effects alter the carbon–metal bonding,
and thus in dppe complex 1b the substituted carbon (C1)
is less strongly bound to palladium than the unsubstituted
one (C3). The substituent effects of the acyloxy groups are
similar to other allylic functionalities, however, the substi-
tuent effects leads to a considerable deformation of the pal-
ladium–allyl bonding. The strong inductive effects of the
acyloxy substituents are clearly reflected by the 13C NMR
shifts of the complexes. This inductive effect varies rela-
tively weakly with the substituents of the acyloxy group.
The acetoxy substitution of the allylic terminal carbon
leads to a considerably larger deshilding of (by 25 ppm)
than the methyl substitution.
(g3-1-Benzoyloxy-allyl)palladium chloride dimer (1c).
This complex was prepared according to method A. The
crude product was purified by silica-chromatography using
1
CH2Cl2 as eluent. H NMR: 3.83 (bs, 2H), 6.28 (q, 1H,
J = 9.2, 15.1 Hz), 7.30 (d, 1H, J = 9.2 Hz), 7.59 (t, 2H,
J = 7.4 Hz), 7.74 (t, 1H, J = 7.6 Hz), 8.05 (d, 2H, J =
8.3 Hz). 13C NMR: 61.75, 104.08, 106.29, 129.39, 129.71,
130.61, 134.90, 163.85.
(g3-1-(4-Methoxy)benzoyloxy-allyl)palladium chloride
dimer (1d). This complex was prepared according to
method A. The crude product was purified by silica-chro-
matography using CH2Cl2:toluene (5:1) as eluent. 1H
NMR: 3.81 (bs, 1H), 3.89 (s, 3H + 1H), 6.23 (q, 1H,
J = 9.6, 14.2 Hz), 7.11 (d, 2H, J = 8.5 Hz), 7.28 (d, 1H,
J = 9.3 Hz), 8.00 (d, 2H, 8.5 Hz). 13C NMR: 61.53,
103.86, 106.07, 129.17, 129.49, 130.39, 134.68, 163.63.
(g3-1-Methyl carbonate-allyl)palladium chloride dimer
(1e). This complex was prepared according to method A.
The crude product was purified by silica-chromatography
1
using CH2Cl2:toluene (5:1) as eluent. H NMR: 3.74 (bs,
4. Experimental
2H), 3.81 (s, 3H), 6.09 (q, 1H, J = 9.7, 14.8 Hz), 6.92 (d,
1H, J = 8.8). 13C NMR: 56.49, 61.68, 105.74, 105.96,
153.31.
The THF used in the reactions was distilled over benzo-
phenone/Na prior to use. H NMR spectra were recorded
1
(g3-1-Acetoxy-2-methyl-allyl)palladium chloride dimer
(1f). This complex was prepared according to method A.
The crude product was purified by silica-chromatography
at either 300 or 400 MHz, 13C NMR spectra were recorded
at either 75.4 or 100.5 MHz (d(CDCl3) = 7.26 ppm and
77.0 ppm, d(DMSO-d6) = 2.54 ppm and 40.45 ppm). For
column chromatography silica gel (230-400 mesh) was
used. The eluent systems are given in volume:volume
ratios.
1
using CH2Cl2:ether (100:1) as eluent. H NMR: 2.15 (s,
3H + 3H), 3.34 (bs, 1H), 4.01 (bs, 1H), 6.91 (s, 1H). 13C
NMR: 18.00, 21.79, 63.11, 100.84, 120.63, 168.48.
(g3-1-Acetoxy-3-methyl-allyl)palladium chloride dimer
(1g). This complex was prepared according to method A.
The crude product was purified by silica-chromatography
General procedure for preparation of allylpalladium com-
plexes (Method A). The appropriate allylsilane (0.5 mmol)

3644
J. Kjellgren et al. / Journal of Organometallic Chemistry 691 (2006) 3640–3645
using CH2Cl2:ether (100:1) as eluent. 1H NMR: 1.53
(d, 3H, J = 6.5 Hz), 2.10 (s, 3H), 4.25 (m, 1H), 6.02 (dd,
1H, J = 8.8, 12.2 Hz), 6.90 (d, 1H, J = 8.6 Hz). 13C
NMR: 17.78, 21.79, 81.88, 100.11, 106.92, 168.79.
Appendix A. Supporting information
13C NMR spectra for complexes 1a–1g. Crystallo-
graphic data for the structural analysis has been deposited
with the Cambridge Crystallographic Data Centre, CCDC
No. 296820 and 296821 for compounds 1a and 1b, respec-
tively. Supplementary data associated with this article can
be found, in the online version, at doi:10.1016/
j.jorganchem.2006.05.013.
Preparation of complex 1b (Method B). Dimer 1a
(25 mg, 0.05 mmol, from fraction A) was dissolved in
CHCl3 (0.2 mL) and added slowly to a suspension of
diphenylphosphinoethane (dppe) (41 mg, 0.10 mmol) and
AgBF4 (20 g, 0.10 mmol) in 0.5 mL CHCl3 at 0 °C under
argon. The solution was stirred for 30 min at 0 °C, then
the AgCl precipitation was removed by centrifugation and
the supernatant was concentrated in vacuo yielding 1b
References
1
as a white powder in 95% yield. H NMR: 1.43 (s, 1H),
[1] J. Tsuji, Palladium Reagents and CatalysisNew Perspectives for the
21st Century, Wiley, Chichester, 2004.
[2] J. Tsuji, Perspectives in Organopalladium Chemistry for the 21st
Century, Elsevier, Amsterdam, 1999.
[3] J. Tsuji, Palladium Reagents and Catalysis: Innovations in Organic
Synthesis, Wiley, Chichester, 1995.
[4] E. Negishi, A.d. Meijre, Organopalladium Chemistry for Organic
Synthesis, Wiley, New York, 2002.
2.18 (m, 2H), 2.94 (m, 2H), 3.07 (t, 1H, J = 12.1 Hz), 4.96
(t, 1H, J = 7.7 Hz), 6.08 (q, 1H, J = 9.8, 16.9 Hz), 7.10 (t,
1H, J = 10.7 Hz), 7.16–7.86 (m, 20H). 13C NMR: 19.55,
26.64 (dd, J = 14.1, 31.8 Hz), 28.30 (dd, J = 13.5, 33.3
Hz), 61.64 (dd, 4.6, 26.9), 103.63 (dd, 8.0, 36.43 Hz), 125–
145 (aromatic C), 166.91. Preparation of crystals for X-
ray diffraction: Complex 1b (50 mg) was dissolved in CHCl3
(0.5 mL). To this solution pentane (0.25 mL) was layered
and the obtained mixture was kept at 4 °C to form crystals
suitable for X-ray diffraction.
Determination of the X-ray structure of 1a and 1b. The
intensities of the reflections were integrated with the STOE
software [50], and the numerical absorption correction was
performed with the programs X-RED and X-SHAPE. The struc-
ture was solved by direct methods (SHELXS-97 [51]) and
refined by full-matrix least-squares on F2 (SHELXL-97 [52]).
Crystal data for 1a: C10H14Cl2O4Pd2: M = 481.95.
[5] S.A. Godleski, in: B.M. Trost, I. Fleming (Eds.), Nucleophiles
with Allyl–Metal Complexes, vol. 4, Pergamon Press, New York,
1991.
[6] B.M. Trost, Acc. Chem. Res. 29 (1996) 355.
[7] B.M. Trost, D.L.V. Vranken, Chem. Rev. 96 (1996) 395.
[8] B.M. Trost, Acc. Chem. Res. 13 (1980) 385.
˚
[9] S. Hansson, A. Heumann, T. Rein, B. Akermark, J. Org. Chem. 55
(1990) 975.
[10] A.D. Ketley, J. Braatz Chem. Commun. (1968) 169.
[11] H.C. Volger, Recl. Trav. Chim. Pays Bas 88 (1969) 25.
[12] R. Huttel, H. Schmid, Chem. Ber. 101 (1968) 252.
¨
[13] B.M. Trost, P.E. Strege, Tetrahedron Lett. 15 (1974) 2603.
[14] J.-E. Ba¨ckvall, Metal-catalyzed Cross Coupling Reactions, VCH,
Weinheim, 1998.
[15] B.L. Shaw, Chem. Ind. (1962) 1190.
[16] S.D. Robinson, B.L. Shaw, J. Chem. Soc. (1963) 4806.
[17] J.M. Rowe, D.A. White, J. Chem. Soc. (A) (1967) 1451.
Orthorhombic space group, Pbca (No. 61), a = 7.638(2)
3
˚
˚
˚
˚
A, b = 7.797(2) A, c = 24.939(11) A, V = 1485.2(9) A ,
Z = 4, T = 293 K, Dcalc = 2.155 g cmÀ3, l (Mo Ka) =
2.78 mmÀ1, F(000) = 929, 5824 reflections measured, 762
unique (Rint = 0.151), 712 observed (I > 2r(I)), 96 parame-
´
´
[18] I. Macsari, K.J. Szabo, Tetrahedron Lett. 41 (2000) 1119.
´
´
[19] I. Macsari, K.J. Szabo, Chem. Eur. J. 7 (2001) 4097.
ters refined, absorption correction (numerical): Tmin/Tmax
=
[20] J.M. Kliegman, J. Organomet. Chem. 29 (1971) 73.
0.269/0.613. R1 = 0.0295 (I > 2r(I)), wR(F2) = 0.0754,
S = 1.12 (all data). Maximum/minimum residual electron
density: 0.81/À0.49.
[21] R.J.P. Corriu, N. Escudie, C. Guerin, Tetrahedron Lett. 271 (1984)
´
C7.
[22] K. Itoh, M. Fukui, Y. Kurachi, J. Chem. Soc., Chem. Commun.
(1977) 500.
Crystal data for 1b: C31H30BF4O2P2Pd: M = 689.72.
[23] T. Hayashi, M. Konishi, M. Kumada, J. Chem. Soc., Chem.
Commun. (1983) 736.
[24] K. Yamamoto, K. Shinohara, T. Ohuchi, M. Kumada, Tetrahedron
Lett. 15 (1974).
[25] T. Hayashi, M. Konishi, Y. Okamoto, K. Kabeta, M. Kumada, J.
Org. Chem. 51 (1986) 3772.
˚
Hexagonal space group, P61 (No. 169), a = 24.182(2) A,
3
˚
˚
c = 9.8114(6) A, V = 4968.6(8) A , Z = 6, T = 293 K,
Dcalc = 1.390 g cmÀ3, l(Mo Ka) = 0.705 mmÀ1, F(000) =
2115, 26301 reflections measured, 4147 unique (Rint
=
0.0977), 3221 observed (I > 2r(I)), 372 parameters refined,
absorption correction (numerical): Tmin/Tmax = 0.894/
0.917. R1 = 0.0461 (I > 2r(I)), wR(F2) = 0.1157, S = 1.04
(all data), absolute structure, Flack parameter = À0.04(6).
Maximum/minimum residual electron density: 0.72/À033.
Data collection: STOE-IPDS image plate diffractometer
[26] K. Fugami, O. Koichiro, U. Kiitro, N. Hitosi, Bull. Chem. Soc. Jpn.
60 (1987) 2509.
[27] S. Ogoshi, W. Yoshida, K. Ohe, S. Murai, Organometallics 12 (1993)
578.
´
[28] K.J. Szabo, Chem. Soc. Rev. 30 (2001) 136.
˚
[29] B. Akermark, S. Hansson, B. Krakenberger, A. Vitagliano, K.
Zetterberg, Organometallics 3 (1984) 679.
[30] B.M. Trost, J. Vercauteren, Tetrahedron Lett. 26 (1985) 131.
[31] X. Lu, Y. Huang, Tetrahedron Lett. 27 (1986) 1615.
[32] L.V. Dunkerton, J.M. Euske, A.J. Serino, Carbohyd. Res. 171 (1987)
89.
˚
using Mo Ka radiation (k = 0.71073 A).
Acknowledgement
´
[33] N. Vicart, B. Cazes, J. Gore, Tetrahedron Lett. 36 (1995) 535.
[34] S.A. Godleski, E.B. Villhauer, J. Org. Chem. 49 (1984) 2246.
[35] J.-E. Ba¨ckvall, J.-E. Nystro¨m, R.E. Nordberg, J. Am. Chem. Soc.
107 (1985) 3676.
This work was supported by the Swedish Natural Sci-
ence Research Council (VR).

J. Kjellgren et al. / Journal of Organometallic Chemistry 691 (2006) 3640–3645
3645
[36] J.E. Ba¨ckvall, S.E. Bystro¨m, R.E. Nordberg, J. Org. Chem. 49 (1984)
4619.
[44] J.-M. Rosset, M.P. Glenn, J.D. Cotton, A.C. Willis, C.H.L. Kennard,
K.A. Byriel, B.H. Riches, W. Kitching, Organometallics17 (1998)1968.
´
´
[45] K.J. Szabo, Organometallics 15 (1996) 1128.
[37] K.J. Szabo, Chem. Eur. J. 3 (1997) 592.
´
´
[46] K.J. Szabo, J. Am. Chem. Soc. 118 (1996) 7818.
[38] C. Jonasson, M. Kritikos, J.-E. Ba¨ckvall, K.J. Szabo, Chem. Eur. J. 6
´
(2000) 432.
[47] K.J. Szabo, E. Hupe, A.L.E. Larsson, Organometallics 16 (1997)
[39] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles and
Applications of Organotransition Metal Chemistry, Oxford Univer-
sity Press, Mill Valley, CA, 1987.
3779.
˚
[48] B. Akermark, B. Krakenberger, S. Hansson, A. Vitagliano, Organo-
metallics 6 (1987) 620.
[40] H. Kurosawa, H. Kajimaru, S. Ogoshi, H. Yoneda, K. Miki, N.
Kasai, S. Murai, I. Ikeda, J. Am. Chem. Soc. 114 (1992)
8417.
[49] M. Moreno-Manas, F. Pajuelo, T. Parella, R. Pleixats, Organomet-
˜
allics 16 (1997) 205.
[50] STOE, X-SHAPE revision 1.09. Crystal Optimisation for Numerical
Absorption Correction, Darmstadt, 1997.
˚
[41] M. Sjo¨gren, S. Hansson, P.-O. Norrby, B. Akermark, M.E. Cucci-
olito, A. Vitagliano, Organometallics 11 (1992) 3954.
[51] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1994) 467.
[52] G.M. Sheldrick, in: G.M. Sheldrick (Ed.), SHELX-97: Computer
Program for the Refinement of Crystal Structures, Release 97-2,
University of Go¨ttingen, Go¨ttingen, Germany, 1997.
´
´
[42] I. Macsari, K.J. Szabo, Organometallics 18 (1999) 701.
[43] J.S. Panek, M.A. Sparks, Tetrahedron Asymmetry
801.
1
(1990)