Using a tripod as a chiral chelating ligand: Chemical exchange between equivalent molecular structures in palladium catalysis with 1,1,1- tris(oxazolinyl)ethane ("trisox")

Article abstract of DOI:10.1002/chem.200700307

Threefold symmetrical chiral podands may simplify the stereochemistry of key catalytic intermediates for cases in which they only act as bidentate ligands. This applies to systems in which chemical exchange between the different κ²-coordinated

Full text of DOI:10.1002/chem.200700307

DOI: 10.1002/chem.200700307
Using a Tripod as a Chiral Chelating Ligand: Chemical Exchange Between
Equivalent Molecular Structures in Palladium Catalysis with 1,1,1-
Tris(oxazolinyl)ethane (“Trisox”)
Carole Foltz,[a] Markus Enders,[a] StØphane Bellemin-Laponnaz,*[b]
Hubert Wadepohl,[a] and Lutz H. Gade*[a]
Dedicated to Professor Peter Hofmann on the occasion of his 60th birthday
Abstract: Threefold symmetrical chiral
podands may simplify the stereochem-
istry of key catalytic intermediates for
cases in which they only act as biden-
tate ligands. This applies to systems in
which chemical exchange between the
different k2-coordinated forms takes
place and in which the non-coordinated
sidearm may play a direct or indirect
role at some earlier or later stage in
the catalytic cycle. Palladium(II)-cata-
lysed allylic substitutions provide ap-
propriate test reactions along these
lines. Aseries of neutral dichloropalla-
dium(II) complexes, [PdCl2(iPr-trisox)]
nuclei of the coordinated oxazoline
rings resonate at d=160–167 ppm and
appear as two singlets due to their dia-
stereotopicity, the signal assigned to
the dangling oxazoline “arm” is ob-
served at d=238–240 ppm. Variable-
temperature NMR studies along with a
systematic series of magnetisation
transfer experiments established ex-
change between ligating and non-ligat-
ing oxazoline rings. Reaction of [Pd-
C
ane) ligands with [Pd-
ACHTREUNG
A
N
with Ph-trisox in CH2Cl2 gave the cor-
responding allyl complex 2, for which
fast exchange between the three oxazo-
line heterocycles as well as between
the exo and endo diastereomers was
observed along with a very slow h3-h1-
h3 process of the allyl fragment (mag-
netisation transfer). Palladium(0) com-
plexes were prepared by reaction of
trisox derivatives or sidearm-function-
alised BOX (BOX=bis(oxazolinyl)-
(1a),
[PdCl2(Ph-trisox)]
(1b),
[PdCl2(Bn-trisox)] (1c) and [PdCl2(Ind-
trisox)] (1d) (trisox=1,1,1-tris(oxazol-
ACHTREUNG
ACHTREUNG
inter alia by 15N NMR spectroscopy.
Direct detection of the heteronuclei
without isotope enrichment and with
“normal” sample concentrations was
achieved with the aid of a cryogenically
cooled NMR probe on a 600 MHz
NMR spectrometer. Whereas the 15N
Keywords: asymmetric catalysis
·
chirality · NMR spectroscopy · pal-
ladium · trisoxazolines
[a] C. Foltz, Priv.-Doz. Dr. M. Enders, Prof. Dr. H. Wadepohl,
Prof. Dr. L. H. Gade
[b] Dr. S. Bellemin-Laponnaz
Institut de Chimie, UMR 7177
UniversitØ Louis Pasteur, 4 rue Blaise Pascal
67000 Strasbourg (France)
Anorganisch-Chemisches Institut, Universität Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
Fax : (+49)6221-545609
Fax : (+33)390-24-50-01
E-mail: lutz.gade@uni-hd.de
E-mail: bellemin@chimie.u-strasbg.fr
5994
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

FULL PAPER
Introduction
The exploitation of C3 chirality in the design of chiral ste-
reodirecting ligands for homogeneous catalytic transforma-
tions is currently the focus of intense research efforts and
considerable conceptual debate.[1–5] The most frequently
cited line of argument put forward to motivate this ap-
proach is based on the homotopicity of the “reactive” coor-
dination sites in octahedral complexes representing inter-
mediates in catalytic cycles (Figure 1).[1a,3b] The use of such
Figure 2. Dynamic exchange of k2-chelating tripods coordinated to a com-
*
plex fragment ( ).
cluded a detailed experimental study into this proposed be-
haviour.[10] We therefore decided to study a catalytic reac-
tion that also involves an active species with square planar
(i.e. non-deltahedral) coordination geometry, which is dia-
magnetic and less labile than the CuII complexes. Palladi-
Figure 1. “Static” k3-facial coordination of tripod ligands.
A
symmetrical stereodirecting ligands may reduce the number
of transition states and diastereomeric reaction intermedi-
ates. In favourable cases, this degeneration of alternative re-
action pathways may lead to high stereoselectivity in catalyt-
ic reactions and greatly simplifies the analysis of such trans-
formations.[1,2,6] Ligand design based on C3 chirality mostly
relates to facially coordinating tripodal ligands. Threefold
rotational symmetry represents the only possibility adapted
to this topology of ligation, in the same way that C2 symme-
try is related to simple chelation.[7,8]
test reactions along these lines and the expected types of in-
terconverting key intermediates (A, B and C) are represent-
ed in Figure 3.
The stereochemical points made above are illustrated for
the case of “static” k3-facial coordination of a symmetrical
chiral tripod in Figure 1. However, a threefold symmetrical
chiral podand may generate a similar simplification in the
stereochemistry of key catalytic intermediates for cases in
which it only acts a bidentate ligand in the stereoselectivity
determining step, in other words, for metal complexes with
a stereoelectronic preference for non-deltahedral coordina-
tion geometries. This is the case for systems in which chemi-
cal exchange between the different k2-coordinated forms
takes place and in which the non-coordinated sidearm may
play a direct or indirect role at some earlier or later stage in
the catalytic cycle. As is schematically shown in Figure 2,
such an exchange represents an equilibrium between identi-
cal species for a symmetrical tripod rather than between iso-
meric complexes, as would be the case for less symmetrical
species.
Figure 3. The three symmetry-related square-planar PdII complexes bear-
ing k2-chelating C3-chiral trisox ligands (top: axial views, bottom: side
views).
Since the first example of a catalytic asymmetric allylic al-
kylation in 1977, this process has undergone a considerable
development towards a practical synthetic tool.[11] New li-
gands of various types have been developed with numerous
substrates and nucleophiles.[12] Ligands that have been used
successfully in the palladium-catalysed asymmetric allylic al-
kylation include chiral phosphanes, P,N-chelate ligands or
nitrogen-based ligands, such as bisoxazolines.[13,14]
This work provides a systematic and comprehensive study
into the coordination chemistry of the “trisox” ligands
(trisox=1,1,1-tris(oxazolinyl)ethane) with palladium(II) and
palladium(0), the dynamic behaviour of the complexes in so-
lution and the use of both symmetrical and non-symmetrical
The test reaction: We have previously invoked this type of
(degenerate) exchange in our interpretation of asymmetric
CuII catalysed Mannich reactions with prochiral b-ketoesters
and related stereoselective transformations with 1,1,1-tris-
A
ligands.[9,10] However, the high substitutional lability of the
copper(II) complexes along with their paramagnetism pre-
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5995

L. H. Gade, S. Bellemin-Laponnaz et al.
tripods in palladium-catalysed allylic alkylations. By using
various C3 symmetric trisoxazoline ligands, in addition to
bis
oxazoline ligands that contain a “hetero”-sidearm,[15,16]
A
we demonstrate that the use of potentially tridentate ligands
in this reaction results in a rate enhancement and an in-
crease in enantioselectvity relative to corresponding cata-
lysts bearing purely bidentate stereodirecting ligands.
Results and Discussion
Synthesis and molecular structures of trisox–palladium(II)
complexes: Aseries of neutral dichloropalladium(II) com-
plexes was synthesised as depicted in Scheme 1. All four
1
Figure 4. 2D H-15N NMR correlated experiment (HMBC) from complex
1c in CDCl3. The directly recorded one-dimensional 15N NMR is dis-
played along the F1 axis.
The molecular structures of the complexes 1a and 1b in
the solid state are shown in Figures 5 and 6 and with select-
Scheme 1. Synthesis of palladium(II) complexes 1a–1d.
complexes, [PdCl2(iPr-trisox)] (1a), [PdCl2(Ph-trisox)] (1b),
[PdCl2(Bn-trisox)] (1c) and [PdCl2(Ind-trisox)] (1d) were
isolated as crystalline air-stable solids from the reaction of
the respective trisox derivative with [PdCl2(PhCN)2] in
G
CH2Cl2 at room temperature. Their analytical data con-
firmed the formulation, and the resonance patterns in the
1H, 13C and 15N NMR spectra recorded at 296 K are consis-
tent with a k2-coordination of the trisox ligands.
Notably, we obtained good quality 15N NMR spectra by
direct detection of the heteronuclei without isotope enrich-
ment and with “normal” sample concentrations with the aid
of a cryogenically cooled direct detection probe (Bruker
QNP-cryoprobeTM, for details see Experimental Section) on
a 600 MHz NMR spectrometer. It was possible completely
Figure 5. Molecular structure of [PdCl2(iPr-trisox)] (1a). Selected bond
À
À
À
lengths (Š) and angles (8): Pd N(1) 2.050(2), Pd N(2) 2.029(3), Pd
À
À
À
Cl(1) 2.2864(9), Pd Cl(2) 2.2825(8), N(1) C(3) 1.272(4), N(2) C(9)
À
1.278(4), N(3) C(15) 1.256(4); Cl(1)-Pd-N(1) 92.31(8), N(1)-Pd-N(2)
1
to assign the signals in H and 13C NMR spectra of the coor-
88.7(1), N(2)-Pd-Cl(2) 91.84(8), Cl(2)-Pd-Cl(1) 87.09(4).
dinated oxazolines and the non-coordinated oxazoline by
1
1
combined 2D H-15N and H-13C NMR experiments. To our
knowledge this is the first such application of 15N NMR
spectroscopy in the coordination chemistry of N-donor li-
gands. Whereas the 15N nuclei of the coordinated oxazoline
rings resonate at d=160–167 ppm and appear as two singlets
due to their diastereotopicity, the signal assigned to the dan-
gling oxazoline “arm” is observed at d=238–240 ppm (ext.
standard: liquid NH3). The 1H-15N correlated spectrum of
complex 1c is shown in Figure 4 with the directly recorded
one-dimensional 15N NMR spectrum displayed along the F1
axis. Given the ubiquity of N-donor ligands, we expect this
methodology, which is enabled by new spectrometer tech-
nology, to find wider applications.
ed bond lengths and angles given in the figure legend. In
both complexes the geometry around the metal centre is dis-
torted square planar, and this deformation is probably a
result of steric repulsion between chloro ligands and the iso-
propyl and phenyl substituents, respectively, of the oxazoline
rings.
The trisox ligands adopt bidentate coordination, with the
third oxazoline unit dangling, and the N-donor pointing
À
À
away from the metal centre. The Pd N and Pd Cl bond
lengths are in the range of those found for related struc-
tures.[17] As expected, the C=N bond length of the free oxa-
zoline is slightly shorter than those in the coordinated oxa-
5996
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
symmetry, are consistent with a fast exchange between ligat-
ing and non-ligating oxazoline rings.
To quantitatively study this exchange and to gain insight
into the microscopic mechanism of this process, a systematic
series of magnetisation-transfer experiments was carried
out. These allowed the direct monitoring of the exchange of
the protons between the coordinated oxazolines (protons a
and a’, -CHMe2) and the non-coordinated oxazoline (proton
b, -CHMe2) (Figure 7). In a series of experiments carried
out with 1a in 1,1,2,2-[D2]tetrachloroethane between 302
and 318 K, the protons of the coordinated oxazolines (a)
were selectively inverted with a shaped pulse, followed by
monitoring of the time evolution of the intensities in the
two sites a/a’ and b. The results of experiments in which
magnetisation of protons a was inverted are shown in
Figure 8.
Figure 6. Molecular structure of [PdCl2(Ph-trisox)] (1b). Selected bond
À
À
À
lengths (Š) and angles (8): Pd N(1) 2.030(2), Pd N(2) 2.041(2), Pd
À
À
À
Cl(1) 2.2915 (7), Pd Cl(2) 2.2606(8); N(1) C(1) 1.273(4), N(2) C(12)
À
1.283(4), N(3) C(21) 1.294(4); Cl(1)-Pd-N(1) 91.58(7), N(1)-Pd-N(2)
89.42(10), N(2)-Pd-Cl(2) 91.02(8), Cl(2)-Pd-Cl(1) 88.07(3).
zoline units (1.256(4) vs. 1.272(4) and 1.278(4) Š for 1a and
1.249(4) vs. 1.273(4) and 1.283(4) Š for 1b).
Solution dynamics of complexes 1a–1d: As indicated above,
the 1H NMR spectrum (recorded at 400 MHz) of 1a at
296 K is consistent with the molecular structure established
for the crystalline state. Separated sets of signals, which are
attributable to the protons of the three different isopropyl
groups of the oxazoline units, indicate the loss of local
threefold symmetry for the tripod ligand (Figure 7). The
same loss of degeneracy of the three oxazoline units is ob-
served in the 13C NMR spectra (75 MHz, 296 K) and the 15N
NMR spectrum (60 MHz, 296 K). Upon increasing the tem-
perature to 373 K coalescence occurs, and the two doublets
Figure 8. Time evolution of the magnetisation in the two sites (blue
traces, coordinated oxazolines a, a’; red traces, non-coordinated oxazoline
b) after selective inversion of the coordinated oxazolines resonance
(complex 1a in 1,1,2,2-[D2]tetrachloroethane). Data are shown in
pseudo-3D representation as a function of the temperature (8C).
a
for the -CH(CH3)2 isopropyl protons observed in the high-
A
temperature limiting spectrum, representing effective C3
For any given temperature T, the time dependence of the
magnetisation in either of the two sites (the previously in-
verted site and the one connected to it by chemical ex-
change) could be derived directly from the McConnell equa-
tions.[18] Numerical nonlinear least-squares fits of the theo-
retical curves to the experimental data gave the NMR spec-
troscopic rate constants kNMR(T) for the fluxional exchange.
Taking into account the statistical factors for the fluxional
exchange of the three sites,[19] kNMR(T) was then converted
into the chemical rate constant kchem(T) (Table 1).
The Eyring plot resulting from k values obtained from in-
version of the coordinated oxazoline sites is shown in
Figure 9. The analysis gave an enthalpy of activation for the
fluxional process of DH° =75.6Æ0.5 kJmolÀ1 and an entro-
py of activation DS° =14.0Æ1.5 JmolÀ1. Similarly, the indan-
yl complex 1d (Figure 10) was studied by using the same ex-
periment, and an enthalpy of activation for the fluxional
Figure 7. 1H NMR of complex [PdCl2(iPr-trisox)] (1a) in 1,1,2,2-
[D2]tetrachloroethane at 296 K (400 MHz).
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5997

L. H. Gade, S. Bellemin-Laponnaz et al.
Table 1. Rate constants kchem [sÀ1] for the fluxional process in complex 1a
Table 2. Activation parameters for the dynamic exchange of coordinated
in 1,1,2,2-[D2]tetrachloroethane (400 MHz).
and free oxazoline rings in complexes 1a and 1d.
T [K]
Proton a
Proton b
Combined data
1a
1d
302
304
306
308
310
312
314
316
318
3.19
3.96
4.77
5.58
6.87
8.19
9.57
11.52
13.71
2.63
3.23
3.90
4.74
5.78
6.98
8.46
10.07
11.93
2.91
3.59
4.34
5.16
6.32
7.58
9.01
10.79
12.82
DH° [kJmolÀ1
DS° [JmolÀ1
DG°(298 K) [kJmolÀ1
]
]
75.6Æ0.5
14.0Æ1.5
71.4Æ0.6
79.4Æ2.0
9.3Æ6.0
76.4Æ2.8
]
N
tensive body of mechanistic work carried out for substitu-
tions at square-planar PdII complexes,[20] we assume that the
interchange has slightly associative character (Ia mecha-
nism), which implies the transient formation of a pentacoor-
dinate palladium(II) complex in the transition state.
Synthesis and structural characterisation of an allyl–palladi-
um complexas a model for the key intermediate in trisox–
Pd-catalysed allylic substitutions: Since the reaction of pal-
ladium allyl derivatives with our reference tripod iPr-trisox
yielded oily products, we focused on the phenyl-substituted
trisoxazoline derivative Ph-trisox in order to obtain a crys-
talline model complex for the allyl intermediates in the cata-
lytic allylic substitution. Reaction of [Pd
N
N
(cod=cyclooctadiene) with Ph-trisox in CH2Cl2 gave the
corresponding complex 2 in 52% yield (Scheme 2).
The molecular structure of complex 2 is displayed in
Figure 11 and with selected bond lengths and angles given in
the figure legend. The palladium atom adopts a planar coor-
dination geometry with the p-allyl ligand and two of the
three oxazoline rings of the trisox ligand being coordinated,
and the nitrogen donors and the carbon termini of the allyl
defining an approximate square. As for the dichloropalladi-
um derivatives 1a–1d, the third oxazoline unit is dangling
with the N-donor pointing away from the palladium centre.
Figure 9. Eyring plot for the data from the last column of Table 1 (com-
plex 1a).
À
À
The Pd N and Pd C bond lengths are within the range
found for related complexes reported in the literature.[21]
The six-membered chelate ring adopts a slightly twisted
boat conformation. Notably, a disorder in the central C
atom of the p-allyl ligand is found in the crystal structure of
the complex with a relative occupancy of about 2:1. This in-
dicates a mixture of the two diastereomers in that ratio, re-
sulting from the reduction of the C2 symmetry of the coordi-
nated bisoxazoline due to the presence of the third uncoor-
dinated heterocycle. The orientation of the allyl group of
Figure 10. 1H NMR of complex [PdCl2(Ind-trisox)] (1d) in 1,1,2,2-
[D2]tetrachloroethane at 296 K (400 MHz).
process of DH° =79.4Æ2.0 kJmolÀ1 and an entropy of acti-
vation DS° =9.3Æ6.0 JmolÀ1 were obtained (Table 2).
Whereas the calculated enthalpy values are as expected
for a dynamic process showing the qualitative behaviour de-
scribed above, the small activa-
the major isomer is exo (exo being defined as the central C
H allylic bond pointing in the same direction as the axial
methyl group).
À
tion entropies indicate neither
an associative nor a dominant-
ly dissociative substitution
mechanism. Areasonable inti-
mate mechanism for the ex-
change between coordinating
and non-coordinating oxazo-
lines may resemble an inter-
change process. Given the ex-
Scheme 2. Synthesis of allyl–palladium(II) complex 2.
5998
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
tween the pairs of the syn or anti protons nor cross exchange
among them. As previously observed for other non-symmet-
rical allyl complexes, this is consistent with the stereochemi-
cal rigidity of the allyl–Pd fragment on the timescale of the
experiments; in other words, it is not affected by the inter-
change of the oxazoline coordination as well as the (possibly
concomitant) interconversion of exo and endo diastereo-
mers.[22]
1
To quantify the latter, a low-temperature H NMR study
was carried out. Coalescence of the trisoxazoline ligand res-
onances occurred at 198 K; however, the low-temperature-
limit was not attained at 190 K (in CD2Cl2). Although the
exchange of the three oxazolines was not completely frozen,
an estimate of the activation barrier DG° for the oxazoline
exchange of ca. 55 kJmolÀ1 may be derived. At low temper-
ature, the resonance pattern of the ligand is thus consistent
with the bidentate N,N-chelation observed in the solid state.
Notably, the signals of the allyl fragment begin to broaden
at 190 K, presumably owing to the slowing down of the exo/
endo exchange; however, complete decoalescence was not
obtained.
The stereochemical rigidity of the allyl–Pd fragment on
the VT-NMR timescale and the observation that the oxazo-
line exchange and the exo/endo isomerisation are associated
with similarly low activation barriers may indicate that the
last two are mechanistically coupled; that is, that the trisox
“walk-about” and the reorientation of the allyl–Pd unit
occur by the same route. Apossible pathway, based on a
stereospecific interchange of the oxazolines is proposed in
Scheme 3, which extends the proposed general exchange
pathway of Figure 3.
Figure 11. Molecular structure of [Pd(Ph-trisox)
(h3-allyl)]+ (2). For clari-
ty, only a single orientation of the allyl ligand is shown, corresponding to
A
À
the major isomer. Selected bond lengths (Š) and angles (8): Pd N(1)
À
À
À
À
2.104(2), Pd N(2) 2.102(2), Pd C(32) 2.120(3), Pd C(30) 2.131(3), N(1)
À
À
C(1) 1.279(3), N(2) C(10) 1.274(3), N(3) C(19) 1.264(3); C(32)-Pd-N(1)
102.16(9), N(1)-Pd-N(2) 88.14(7), N(2)-Pd-C(30) 100.85(10), C(32)-Pd-
C(30) 68.87(12).
1
In the H NMR spectrum of complex 2 at room tempera-
ture, only three well-defined signals for the oxazoline pro-
tons are observed indicating fast exchange between the
three heterocycles as well as between the exo and endo dia-
stereomers. Five signals are observed for the allyl moiety as
expected for a dissymmetrical p-allyl ligand (Figure 12), the
Figure 12. 1H NMR of complex [Pd(Ph-trisox)
at 296 K (400 MHz).
A
Scheme 3. Possible exchange pathway, based on a stereospecific inter-
change of the oxazolines which extends the proposed general exchange
pathway of Figure 3.
1
partial assignment being based on a H NOESY experiment.
In addition to the negative phase non-diagonal NOE cross-
peaks, EXSY signals between syn and anti protons were ob-
served, indicating slow exchange between h3-allyl and h1-
allyl forms.
Synthesis, structural characterisation and dynamic behaviour
of trisox-palladium(0) complexes: Having explored the coor-
dination chemistry with palladium(II), we then turned our
attention to zero-valent palladium complexes, which repre-
sent the other intermediates in the catalytic cycle of allylic
Aseries of variable-temperature (VT) 1H NMR experi-
ments performed in the range of 296 to 373 K indicated no
spectral changes for the allyl proton resonances, neither be-
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5999

L. H. Gade, S. Bellemin-Laponnaz et al.
substitutions. Compared to the extensive work on phos-
phane–h2-alkene palladium(0) complexes, there are only few
well defined [Pd(0)(h2-alkene)] complexes that contain an-
cillary nitrogen-donor ligands.[23] In particular, we note that
there is no report of structurally characterised Pd0 com-
plexes containing oxazoline-based ligands. Palladium(0)
compounds are generally either formed in situ by reduction
of a suitable palladium(II) precursor, or by starting from a
zero-valent palladium precursor complex containing labile
ligands. Attempts to reduce the [PdCl2(trisox)] complexes
A
described in the previous sections did not lead to the expect-
ed palladium(0) complexes; however, an alternative synthet-
ic route starting from zero-valent palladium complexes was
successful.
Anumber of palladium(0) complexes with different li-
gands (including a potentially tridentate pyridine-bisoxazo-
line ligand (vide infra)) and alkenes were synthesised
(Scheme 4) by using complexes of general formula [Pd-
Figure 13. Molecular structure of [Pd(iPr-trisox)(ma)] (3a). Selected
À
À
bond lengths (Š) and angles (8): Pd N(1) 2.127(5), Pd N(2) 2.115(4),
À
À
À
À
Pd C(1) 2.107(7), Pd C(2) 2.082(6), N(1) C(7) 1.261(7), N(2) C(13)
A
N
À
À
1.261(7), N(3) C(19) 1.254(9), C(1) C(2) 1.448(9); N(1)-Pd-N(2) 85.6(2),
N(2)-Pd-C(2) 116.9(2), C(1)-Pd-N(1) 117.6(2), C(1)-Pd-C(2) 40.4(2); tor-
sion angle C(3)-C(2)-C(1)-Pd À98.48.
dride or tetracyanoethylene) as precursor.[23a,24] Complexes
3a–e were obtained by substitution of the nbd ligand by the
respective tripod ligand in THF and were isolated as highly
air sensitive yellow powders in 50–75% yield.
The molecular structure of compound 3a is presented in
Figure 13 along with the principal geometric parameters. For
palladium(0) complexes, both a trigonal planar or a tetrahe-
dral molecular geometry would be expected in case of a
facial tridentate ligand. Complex 3a possesses Y-shaped
trigonal-planar geometry with the trisox ligand coordinated
in a bidentate fashion, whilst the third oxazoline unit is dan-
gling with the N-donor pointing away from the metal centre.
À
À
The Pd N (2.127(5) Š) and Pd C (2.082(6) Š) bond lengths
are in agreement with reported values for similar complexes
that contain bidentate nitrogen-based ligands such as
tBuDAB
(tBuDAB=di-tert-butyl-1,4-diazabutadiene).[25]
The N(1)-Pd-N(2) angle of 85.6(2)8 is greater than those re-
ported for related complexes (typically 77.2–77.58).[26] This
increase in bite angle is due to
the six-membered chelate ring
of
a bidentate trisoxazoline
with respect to the values of
five-membered rings described
in the literature. As expected,
upon coordination of the
alkene an elongation of the C=
C bond distance (1.448(9) Š)
with respect to the free alkene
(1.3322(9) Š) is observed.[27] It
is of interest to note that al-
though an equilibrium of two
diastereomers is observed in
solution at ambient tempera-
ture, the crystals employed for
the X-ray diffraction study of
3a consisted only of one dia-
stereomeric form.
In the crystals of the phenyl-
substituted complex 3b, the
other of the two diastereomers
is exclusively found. Apart
from that, the gross molecular
structure appears similar to
that of 3a. However, the quali-
ty of the data was only suffi-
Scheme 4. Palladium(0) complexes 3a–e (nbd=norbornadiene, ma=maleic anhydride, tcne=tetracyanoethy-
lene).
6000
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
the p-coordinated maleic anhydride may adopt two possible
orientations with respect to the pendant arm, which is either
1
an oxazoline or the CH2Py arm. By means of a VT H NMR
study the high-temperature limit for these exchange process-
es between a total of six diastereomers was attained and an
effective free enthalpy of activation for the overall process
of bisoxazoline/oxazolin–pyridine exchange of DG°
(315 K)
G
ꢀ63 kJmolÀ1 was estimated.
cient to unequivocally establish the molecular connectivity
and configuration, but did not allow a more detailed appre-
ciation of the structure.
Palladium-catalysed asymmetric allylic alkylation with tris-
A
action that is mechanistically very well understood and may
thus serve as a probe for studying the catalytic behaviour of
the new Pd complexes, the allylic alkylation of 1,3-diphenyl-
prop-2-enyl acetate substrate with dimethyl malonate as nu-
cleophile (in the presence of N,O-bis(trimethylsilyl)acet-
1H NMR spectra of all complexes recorded at 296 K rep-
resent a dynamic exchange regime for the trisox ligand as
well as the equilibrium between the diastereomers due to
the different orientation of the p-bonded maleic anhydride
ligand relative to the pendant oxazoline arm. Upon lowering
the temperature, these fluxional processes are frozen for
complexes 3a–c and two sets of resonances attributable to
the two diastereomers are observed. For complex 3b the
ratio was found to be 1:1.1 and the measured activation bar-
AHCTREUNG
the ligand to the palladium allyl chloride [{PdCl
precursor (Scheme 5).[30]
A
rier DG°(261 K)=55 kJmolÀ1. The isomer interconversion
G
may, in principle, be caused by either an olefin rotation or a
decoordination–coordination process that involves the
alkene and/or the ligand. The DG° values reported in the
literature for alkene rotation are usually around 60–
Scheme 5. Palladium-catalysed allylic alkylation of 1,3-diphenylprop-2-
enyl acetate.
70 kJmolÀ1 [28] however, energy barriers of 50 kJmolÀ1 have
;
also been noted.[29] In the case of complex 3b, it is therefore
difficult to conclude on the exact mechanism, since the
ligand may be involved in the process.
Pfaltz and co-workers previously investigated this reaction
with the well-established bisoxazoline (BOX) ligands as ste-
reodirecting ligands, and this particular system therefore
provided the point of reference for the catalytic study at
hand. By using 2.2 mol% of palladium and the iPr-BOX
ligand, an enantiomeric excess (ee) of 89% and 89% isolat-
ed yield were obtained after three days in good agreement
with Pfaltzꢁs results (Table 3, entry 1).[31] Under the same
conditions, the analogous catalyst with iPr-trisox as stereodi-
recting ligand gave an ee of 95% and 90% yield. As a simi-
lar comparative study with the other oxazoline derivatives
showed, the trisoxazoline-based catalysts generally induce a
better enantioselectivity compared to their bisoxazoline ana-
In the case of complex 3d, which contains tetracyanoeth-
ylene as p-bonded alkene, the resonance pattern of the oxa-
zoline protons is consistent with effective C3 symmetry and
thus rapid exchange between the three heterocycles. Lower-
ing the temperature leads to coalescence at 243 K and the
non-symmetrical low-temperature limiting spectrum at
198 K. Asimulation of the dynamic NMR spectra gave an
activation barrier of DG°
A
Whereas the trisox–Pd complexes 3a–3d contain three-
fold symmetrical tripods, complex 3e contains a bisoxazoline
ligand to which a 2-pyridylmethyl sidearm has been added.
This renders the ligand completely non-symmetrical and
thus—as indicated in Figure 2 for C1 symmetric tripods—
there are potentially three species, which differ in the way
the ligand is coordinated to the metal centre. Apart from an
isomer with two oxazoline units bound to the metal there
are two closely related but diastereomeric forms with one
oxazoline ring and the pyridine arm bound to the metal.
Table 3. Results of asymmetric allylic alkylation with various bisoxazo-
lines and trisoxazolines.[a]
1
Whereas the H NMR spectra recorded at 296 K represent
Entry
R
Yield [%]
ee [%]
Yield [%]
ee [%]
an intermediate dynamic regime, the exchange between the
different diastereomers was frozen out at 218 K. Notably, a
1:1 equilibrium mixture of the bisoxazoline complex on the
one hand and the two forms of the oxazolin–pyridine isomer
on the other are observed. Since the oxazolin–pyridine iso-
mers possess near-identical overlapping resonance patterns
they were treated as one species in the analysis of the dy-
namic process. The situation is complicated by the fact that
1
2
3
4
(S)-iPr
(R)-Ph
(S)-Bn
89
7
88
13
89
À72
83
À93
90
28
92
95
95
À88
88
À98
(4R,5S)-Ind
[a] Experimental conditions: catalytic precursor generated in situ from
[{PdCl
AHCTREUNG
508C for 1.5 h; catalysis carried out at room temperature; yields and ee
determined after 72 h.
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6001

L. H. Gade, S. Bellemin-Laponnaz et al.
logues, and this behaviour appears to be independent of the
substituent as shown on Table 3.
were synthesised by reaction of the monolithiated 1,1-bis-
(oxazolinyl)ethane with the appropriate electrophiles. Addi-
In a further comparative study, the catalytic conversions
with the BOX and trisox ligands from Table 3 were moni-
tored by gas chromatography (Figure 14). The most notable
observation is the rate acceleration with the tripods com-
pared to the BOX ligands for all substitution patterns. The
turn-over frequencies (TOFs) derived from the quasilinear
section in the conversion curves are displayed in Table 4.
tionally, we also synthesised several dually functionalised C2
symmetric systems that are readily obtained from the bis-
AHCTREUNG
marised in Table 5 (together with previously discussed iPr-
BOX/iPr-trisox couple, entries 1–2). In general, the highest
yields were obtained from ligands that contain a potentially
donating heteroatom as sidearm, with ligands that contain
nitrogen donors displaying slightly higher activity than
oxygen donors. The enantioselectivity of the product is also
affected by the nature of the sidearm. For example, methyl-
pyridyl as sidearm gave moderate to very low enantioselec-
tivity (entries 6 and 10), whilst the introduction of ketone
units appear to give more selective catalysts (entries 4 and
5). Ligands with no heteroatom-containing sidearms usually
give lower yields (entries 1,7–9) and moderate to good enan-
tiomeric excesses, depending on the bulkiness of the sub-
stituents.
Table 4. Turn-over frequencies TOF [hÀ1] derived from the conversion
curves shown in Figure 14.
TOF ratio
Trisox/BOX
R
(S)-iPr
(R)-Ph
(S)-Bn
1.37
0.05
0.91
0.19
5.02
0.2
3.73
12.2
3.7
4
4.1
64.2
(4R,5S)-Ind
Acomparative study of the catalytic conversion with two
representative non-symmetric side-arm-functinalised ligands
(-COtBu, entry 4 and -CH2Py, entry 6) was carried out and
compared with the results obtained for iPr-BOX and iPr-
trisox systems (Figure 15). Whereas the introduction of a do-
nating sidearm leads to rate enhancement, none of the cata-
lysts containing the non-symmetrical stereodirecting ligands
reach the activity obtained with the trisoxazoline ligand iPr-
trisox.
The mechanism of palladium-catalysed allylic alkylation is
generally thought to involve four steps.[12,13] After coordina-
tion of the substrate to the palladium(0) species, an oxida-
tive addition results in the formation of the p-allyl palladi-
The rate of the reaction is strongly dependent on the sub-
stituent of the respective ligand, with the iPr substituent
yielding the most active BOX-derivative, whereas the indan-
yl substituent leads to the highest rate for the trisox-based
catalysts. With iPr, Ph and Bn-based ligands, the TOFs
differ by a factor of four in favour of the tripod, whilst a
striking 64-fold acceleration was found for the indanyl-deriv-
ative!
To explore the role of the third oxazoline unit, a series of
modified bisoxazoline ligands containing potentially coordi-
nating or non-coordinating “sidearms” at the apical position
~
&
Figure 14. Comparison of the conversion curves for the Pd/trisoxazoline systems ( ) and the corresponding Pd/bisoxazoline systems ( ).
6002
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
Table 5. Results of asymmetric allylic alkylation of 1,3-diphenylprop-2-
enyl acetate with dimethylmalonate.[a]
Entry
Ligand
Yield [%]
ee [%]
1
89
89
2
3
90
67
95
91
Figure 15. Comparison of the conversion curves for the following catalyst
~
&
systems: Pd/(iPr-trisox) ( ), Pd/
A
ACHTREUNG
^
(+) and Pd/
ACHTREUNG
respectively).
which the product may be displaced by way of substitution
through another substrate molecule.
4
5
74
68
95
95
Since our catalytic systems were prepared in situ from
palladium(II) allyl chloride and the trisox ligand, the first
step in the formation of the active catalyst involves the re-
duction of the palladium(II) precursor by a nucleophilic
attack by the carbanionic species. The observation of an in-
duction period in the conversion curves is thus not surpris-
ing. Introduction of an additional donor function in the ste-
reodirecting ligand generally resulted not only in a rate en-
hancement, but also the reduction of this induction period.
The observed overall rate acceleration might be due to the
ability of the additional donating group to induce the forma-
tion of the palladium(0) species both in the initial genera-
tion of the active species as well as in the product/substrate
exchange step at the end of the catalytic cycle.
6
93
66
This mechanistic aspect, as well as the symmetry-related
simplification of the reaction network for the catalysts bear-
ing C3-chiral tripods, may be at the root of the superior per-
formance of the trisox systems.
7
8
15
51
86
94
Conclusion
In this work we have demonstrated the potential of C3-
chiral trisox ligands in enantionselective catalytic transfor-
mation in which the complex intermediate in the selectivity-
determining step does not adopt a deltahedral coordination
geometry and thus does not allow tripodal coordination of
the stereodirecting ligand. Instead, the presence of the addi-
tional donor function appears to play a role in the subse-
quent product/substrate exchange step as well as in the ini-
tial generation of the active catalyst. Since the key inter-
mediate contains a dicoordinate tripod with a dangling
ligand arm, with dynamic exchange between the donor func-
tions, the use of a C3-symmetric trisoxazoline significantly
reduces the number of interconverting isomers. This simpli-
fication of the catalytic reaction network appears to be man-
ifested in superior catalyst performance for such species.
Notably, the system with the least internal degrees of free-
9
63
84
93
10
3
[a] Experimental conditions: catalyst precursor generated in situ by stir-
ring [{PdCl
(h3-C3H5)}2] and ligand (1.1 mol%, ratio ligand/Pd=1.1) in
THF at 508C for 1.5 h; catalysis carried out at ambient temperature;
ACHTREUNG
yields and ee determined after a reaction time of 72 h.
ACHTREUNG
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6003

L. H. Gade, S. Bellemin-Laponnaz et al.
1,1-Bis
(0.9 mL, 2m in THF/pentane, 1.8 mmol) was added dropwise to a solution
of bis
[(4R,5S)-4,5-indanediyloxazolin-2-yl]methane (552 mg, 1.7 mmol)[36]
A
dom, the catalyst bearing the indanyl-trisox ligand displays
the highest activity and selectivity.[32]
AHCTREUNG
in THF (25 mL) at À788C. The brown solution of the anion was allowed
to warm to ambient temperature and stirred for an additional 0.5 h prior
to the addition of methyl trifluoromethanesulfonate (0.20 mL, 1.8 mmol).
The colourless solution was stirred for 12 h and was concentrated to dry-
ness. The residue was redissolved in dichloromethane (60 mL) and
washed with a satured aqueous solution of NH4Cl (10 mL) and brine
(10 mL). The organic extract was dried over Na2SO4 and concentrated in
vacuo to give a yellowish solid. Purification by flash chromatography
Experimental Section
All manipulations, except those indicated otherwise, were carried out
under an inert atmosphere of dry argon using standard Schlenk tech-
niques. Solvents were purified and dried by standard methods. The start-
ing materials (R)-phenylglycinol,[33] (S)-phenylalaninol,[33] 1,1-bis[(4R)-4-
phenyl-1,3-oxazolin-2-yl]ethane,[34] (4R)-4-phenyloxazoline,[35] (4S)-4-ben-
(hexane/EtOAc, 50:50) gave the desired product as
(340 mg, 59% yield). 1H NMR (400 MHz, CD2Cl2, 296 K): d=1.35 (d,
J=7.2 Hz, 3H; CH3), 3.03 (m, 2H; CH2Ind), 3.39 (m, 3H; CH2Ind
a white solid
zyloxazoline,[35] bis
(cod)
(h3-C3H5)]BF4,[37]
tcne)],[23a]
[(4R,5S)-4,5-indanediyloxazolin-2-yl]methane,[36] [Pd-
,
CHbridge), 5.30 (m, 2H; OCHoxa), 5.52 (d, J=8.0 Hz, 2H; NCHoxa), 7.28
(m, 6H; CHarom), 7.45 ppm (m, 2H; CHarom); 13C {1H NMR (100 MHz,
CD2Cl2, 296 K): d=14.6 (CH3), 33.9 (CHbridge), 39.7 (CH2Ind), 76.6
(NCHoxa), 83.2 (OCHoxa), 125.3, 127.2, 128.3 (Carom), 139.9, 141.9 (Cquat-
arom), 165.6 ppm (NCO); FT-IR (KBr): n˜ =1653 cmÀ1 (s, C=N); MS (EI):
m/z (%): 344.4 (81) [M+]; elemental analysis calcd (%) for C22H20N2O2:
C 76.72, H 5.85, N 8.13; found: C 76. 61, H 5.89, N 8.17.
G
E
[Pd(h2,h2-nbd)(h2-ma)],[23a] [Pd(h2,h2-nbd)(h2-
A
ACHTREUNG
2,2-bis[(4S)-4-isopropyloxazolin-2-yl]-1-(pyridin-2-yl)popa-
ne,[15a] 2,2-bis[(4S)-4-isopropyloxazolin-2-yl]-1-phenylpropane,[15a] and 2,2-
bis[(4S)-4-isopropyloxazolin-2-yl]-1,3-diphenylpropane,[38] as well as (4S)-
2-bromo-4-benzyloxazoline,[39]
(4R,5S)-2-bromo-4,5-indanediyloxazo-
line,[39] (4R)-2-bromo-4-phenyloxazoline[39] were synthesised according to
literature procedures. 2,2-Bis[(4S)-4-isopropyloxazolin-2-yl]-4,4-dimethyl-
pentan-3-one,
2,2-bis[(4S)-4-isopropyloxazolin-2-yl]-1-phenylpropan-1-
1,1,1-Tris
1.5m in pentane, 0.77 mmol) was added dropwise to a solution of 1,1-bis-
[(4R,5S)-4,5-indanediyloxazolin-2-yl]ethane (222 mg, 0.64 mmol) in tolu-
[(4R,5S)-4,5-indanediyloxazolin-2-yl]ethane: tBuLi (0.52 mL,
one, 2,2-bis[(4S)-4-isopropyloxazolin-2-yl]-1,3-di(naphth-2-yl)propane
and 2,2-bis[(4S)-4-isopropyloxazolin-2-yl]-1,3-di(pyrin-2-yl)propane were
obtained according to previously published protocols.[40] All other re-
agents were commercially available and were used without further purifi-
cation. 1H and 13C NMR spectra were recorded on a Bruker DRX 200
spectrometer at 200 MHz and 50 MHz, respectively, on a Bruker Avance
300 spectrometer at 300 MHz and 75 MHz, respectively, on a Bruker
Avance II 400 specrometer at 400 MHz and 100 MHz, respectively, and
on a Bruker Avance III 600 spectrometer at 600 MHz and 150 MHz, re-
spectively, and were referenced using the residual proton or 13C solvent
peak. 15N NMR spectra were recorded on a Bruker Avance III 600 spec-
trometer equipped with a cryogenically cooled direct detection probe
(QNP-cryoprobeTM, optimised for detection of 31P, 13C and 15N). The ex-
perimental parameters for the direct 15N detection were optimised with a
0.2m solution of a non-enriched N-ligand titanium complex. For routine
direct 15N NMR detection concentrations of not less than 0.1m are neces-
sary. Selected parameters are: Puls-program: inverse gated decoupled;
relaxation delay 6 s; 908-15N-puls (10 ms); pre-acquisition delay 400 ms;
time domain 64 K; sweep 500 ppm; acquisition time 1 s; number of accu-
mulations 5000. Processing: 60 points of backward linear prediction in
order to minimise base-line artefacts; exponential window function with
AHCTREUNG
ene (60 mL) at À858C. The resulting yellow solution was stirred for an
additional 30 min prior to the addition of a solution of (4R,5S)-2-bromo-
4,5-indanediyloxazoline (346 mg, 1.45 mmol)[39] in cold toluene (4 mL).
The solution was allowed to warm slowly to room temperature for 12 h
and then concentrated to remove the pentane and finally the Schlenk
tube was sealed. The stirred solution was heated at 608C for two days.
The resulting orange solution was evaporated to dryness. The residue was
redissolved in dichloromethane (100 mL) and washed with a satured
aqueous solution of NaHCO3 (10 mL) and brine (10 mL). The organic
extract was dried over Na2SO4 and concentrated in vacuo to give an
orange foam. Purification by flash chromatography (hexane/EtOAc,
80:20) gave the desired product as a yellowish solid (100 mg, 31% yield).
1H NMR (400 MHz, CD2Cl2, 296 K): d=1.60 (s, 3H; CH3), 3.04 (dd, J=
1.8, 18.0 Hz, 3H; CH2Ind), 3.33 (dd, J=7.2, 18.1 Hz, 3H; CH2Ind), 5.29
(ddd, J=2.1, 7.2, 8.0 Hz, 3H; OCHoxa), 5.49 (d, J=8.0 Hz, 3H; NCHoxa
7.25 (m, 9H; CHarom), 7.36 ppm (d, J=7.2 Hz, 3H; CHarom); 13C {1H}
NMR (100 MHz, CD2Cl2, 296 K): d=21.3 (CH3), 22.4 ((CH3)C(oxa)3),
)
ACHTREUNG
39.9 (CH2Ind), 76.9 (NCHoxa), 84.0 (OCHoxa), 125.5, 125.8, 127.5, 128.7
(Carom), 140.4, 141.9 (Cquat-arom), 164.7 ppm (NCO); FT-IR (KBr): n˜ =
1653 cmÀ1 (s, C=N); MS (FAB): m/z (%): 502,4 (100) [M+]; elemental
analysis calcd (%) for C32H27N3O3: C 76.63, H 5.43, N 8.38; found: C
76.50, H 5.47, N 8.45.
0.3 Hz line broadening factor. Infrared spectra were obtained on
a
Perkin–Elmer 1600 FT-IR spectrometer. Mass spectra and elemental
analysis were recorded by the analytical services of the Strasbourg and
Heidelberg chemistry departments.
1,1-Bis
A
Diethylmethylmalonate
Preparation of the trisoxligands and their precursors
(2.8 mL, 16.3 mmol) and (S)-phenylalaninol were added in a Schlenk
flask. NaH (10 mg, 0.25 mmol; 60% dispersion in mineral oil) was then
added under nitrogen to the flask which was sealed and heated to 1308C.
After 16 h, the ethanol was removed under vacuum to leave a white solid
pure enough to be used for the next step without further purification
(5.9 g, 95%). Asolution of TsCl (5.9 g, 30.8 mmol) in CH 2Cl2 (30 mL)
was slowly added to an ice-cooled solution of the dihydroxy diamide pre-
pared in the previous step (5.9 g, 15.4 mmol), triethylamine (17.2 mL,
123.2 mmol) and DMAP (188 mg, 1.54 mmol) in CH2Cl2 (200 mL). The
mixture was warmed to room temperature, was stirred for 3 d and was
washed with a saturated aqueous solution of NH4Cl and brine. The or-
ganic phase was dried over Na2SO4 and was concentrated in vacuo to
give a yellow oil. Purification by flash chromatography (CH2Cl2/MeOH/
Et3N, 97:3:1) gave the desired product as a colourless oil (2.4 g, 45%
yield). 1H NMR (200 MHz, CDCl3, 296 K) d=1.46 (d, J=7.2 Hz, 3H;
CH3), 2.67 (dd, J=8.5, 13.7 Hz, 2H; CH2Bn), 3.11 (dd, J=5.0, 13.7 Hz,
2H; CH2Bn), 3.48 (q, J=7.2 Hz, 1H; CHbridge), 4.00 (dd, J=7.4, 8.3 Hz,
2H; CH2oxa), 4.20 (m, 2H; CH2oxa), 4.42 (m, 2H; CHoxa), 7.23 ppm (m,
10H; CHarom); 13C {1H} NMR (50 MHz, CDCl3, 296 K): d=15.2 (CH3),
33.9 (CHbridge), 41.4 (CH2Bn), 67.2 (CHoxa), 72.1 (CH2oxa), 126.5, 128.5,
129.3 (Carom), 137.7 (Cquat-arom), 166.1 ppm (NCO); FT-IR (KBr): n˜ =
1661 cmÀ1 (s, C=N); MS (EI): m/z (%): 257.1 (100) [M+ÀCH2Ph] 348.2
1,1,1-Tris[(4R)-4-phenyloxazolin-2-yl]ethane: tBuLi (1.2 mL, 1.7m in
A
pentane, 2 mmol) was added dropwise to a solution of 1,1-bis[(4R)-4-phe-
nyloxazolin-2-yl]ethane (535 mg, 1.8 mmol)[34] in THF (60 mL) at À788C.
The resulting yellow solution was stirred for an additional 30 min prior to
the addition of (4R)-2-bromo-4-phenyloxazoline (1.2 equiv, 453 mg,
2 mmol).[39] The solution was allowed to warm slowly to room tempera-
ture over 12 h and then concentrated to remove the pentane and finally
the Schlenk tube was sealed. The stirred solution was heated at 708C for
5 d. The resulting orange solution was evaporated to dryness. The residue
was redissolved in dichloromethane (100 mL) and washed with water
(10 mL). The organic extract was dried over Na2SO4 and concentrated in
vacuo to give a yellow solid. Purification by crystallisation from CH2Cl2/
pentane gave the desired product (470 mg, 60% yield). 1H NMR
(200 MHz, CDCl3, 296 K): d=2.06 (s, 3H; CH3), 4.27 (dd, J=7.8, 8.3 Hz,
3H; CH2oxa), 4.76 (dd, J=8.3, 10.1 Hz, 3H; CH2oxa), 5.32 (dd, J=7.7,
10.1 Hz, 3H; CHoxa), 7.30 ppm (m, 15H; CHarom); 13C {1H} NMR
(50 MHz, CDCl3, 296 K): d=21.4 (CH3), 45.2 ((CH3)C(oxa)3), 69.7
G
(CHoxa), 76.0 (CH2oxa), 126.9, 127.6, 128.7 (Carom), 142.1 (Cquat-arom),
166.1 ppm (NCO); FT-IR (KBr): n˜ =1665 cmÀ1 (s, C=N); MS (EI): m/z
(%): 465.7 (92) [M+]; elemental analysis calcd (%) for C29H27N3O3: C
74.82, H 5.85, N 9.03; found: C 74.70, H 5.81, N 8.99.
6004
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
(50) [M+]; elemental analysis calcd (%) for C22H24N2O2: C 75.83, H 6.94,
N 8.04; found: C 75.62, H 6.99, N 7.96.
[M+À2Cl]; elemental analysis calcd (%) for C29H27Cl2N3O3Pd: C 54.18,
H 4.23, N 6.54; found: C 53.01, H 4.29, N 6.60.
(1,1,1-Tris
(1c): Yield: 68%. 1H NMR (600 MHz, CDCl3, 296 K): d=1.55 (s, 3H;
CH3), 2.37 (pseudo-t, J=11.2 Hz, 1H; CH2Bn C), 2.79 (dd, J=7.4,
13.9 Hz, 1H; CH2Bn, F), 2.93 (dd, J=8.5, 13.4 Hz, 1H; CH2Bn, C), 3.01
(dd, J=5.3, 13.8 Hz, 1H; CH2Bn, F), 3.44 (dd, J=1.4, 13.8 Hz, 1H;
CH2Bn, C), 3.82 (dd, J=1.6, 13.2 Hz, 1H; CH2Bn, C), 4.06 (pseudo-t, J=
8.1 Hz, 1H; CH2oxa, F), 4.17 (pseudo-t, J=8.6 Hz, 1H; CH2oxa, C), 4.26
(pseudo-t, J=8.8 Hz, 1H; CH2oxa, C), 4.30 (pseudo-t, J=9.1 Hz, 1H;
G
dichloride
1,1,1-Tris
tane, 2.6 mmol) was added dropwise to a solution of 1,1-bis[(4S)-4-benzyl-
oxazolin-2-yl]ethane (768 mg, 2.2 mmol) in THF (80 mL) at À788C. The
U
ACHTREUNG
resulting yellow solution was stirred for an additional 30 min prior to the
addition of (4S)-2-bromo-4-benzyloxazoline (1.2 equiv, 634 mg,
2.6 mmol).[39] The solution was allowed to warm slowly to room tempera-
ture over 12 h and then concentrated to remove the pentane; finally the
Schlenk tube was sealed. The stirred solution was heated at 708C for 3 d.
The resulting orange solution was evaporated to dryness. The residue was
redissolved in dichloromethane (100 mL) and washed with water
(10 mL). The organic extract was dried over Na2SO4 and concentrated in
vacuo to give an orange oil. Purification by flash chromatography
(CH2Cl2/MeOH/Et3N, 97:3:1) gave the desired product as a white solid
CH2oxa, F), 4.37 (dd, J=1.9, 8.9 Hz, 1H; CH2oxa, C), 4.49 (m, 1H; CHoxa
F), 4.54 (dd, J=1.6, 8.8 Hz, 1H, CH2oxa, C), 5.04 (m, 1H; CHoxa, C), 5.10
(m, 1H; CHoxa C), 7.19–7.24 (m, 4H; CHarom), 7.28–7.35 (m, 9H;
CHarom), 7.45 ppm (m, 2H; CHarom); 13C {1H} NMR (150 MHz, CDCl3,
296 K): d=20.9 (CH3), 39.2, 39.4 (CH2Bn C) 40.6 (CH2Bn F) 45.0
((CH3)C(oxa)3), 66.2 (CHoxa, C), 67.1 (CHoxa, F) 67.2 (CHoxa, C), 72.8,
,
,
,
,
1
AHCTREUNG
(620 mg, 56% yield). H NMR (400 MHz, CDCl3, 296 K): d=1.76 (s, 3H;
72.9 (CH2oxa, C), 73.3 (CH2oxa, F), 126.8, 127.1, 1267.3 (Carom, F), 128.6,
128.7, 128.8, 129.7, 129.8, 130.1 (Carom, C), 135.4, 135.9 (Cquat-arom, C), 136.5
(Cquat-arom, F), 161.0 (NCO, F), 166.6, 166.8 ppm (NCO, C); 15N (60 MHz,
CDCl3, 296 K): d=161.3 162.2 (N, C) 239.9 ppm (N, F); FT-IR (KBr): n˜ =
1655 cmÀ1 (s, C=N); HRMS (FAB): m/z (%): 613.151 (100) [M+À2Cl];
elemental analysis calcd (%) for C32H33Cl2N3O3Pd: C 56.11, H 4.86, N
6.13; found: C 56.04, H 4.80, N 6.19.
CH3), 2.68 (dd, J=8.5, 13.7 Hz, 3H; CH2Bn), 3.11 (dd, J=5.1, 13.7 Hz,
3H; CH2Bn), 4.07 (dd, J=6.9, 8.3 Hz, 3H; CH2oxa), 4.23 (dd, J=8.5,
9.0 Hz, 3H; CH2oxa), 4.46 (m, 3H; CHoxa), 7.24 ppm (m, 15H; CHarom);
13C {1H} NMR (100 MHz, CDCl3, 296 K): d=20.9 (CH3), 41.2 (CH2Bn),
44.6 ((CH3)C(oxa)3), 67.2 (CHoxa), 72.5 (CH2oxa), 126.4, 128.4, 129.4
G
(Carom), 137.7 (Cquat-arom), 165.0 ppm (NCO); FT-IR (KBr): n˜ =1664 cmÀ1
(s, C=N); MS (FAB): m/z (%): 508.5 (100) [M+]; elemental analysis
calcd (%) for C32H33N3O3: C 75.71, H 6.55, N 8.28; found: C 75.55, H
6.52, N 8.33.
(1,1,1-Tris
chloride
ACHTREUNG
[D2]tetrachloroethane, 296 K): d=1.50 (s, 3H; CH3), 2.09 (d, J=18.2 Hz,
1H; CH2Ind, C), 2.48 (d, J=18.6 Hz, 1H; CH2Ind, C), 3.03 (dd, J=6.8,
18.2 Hz, 1H; CH2Ind,C), 3.08 (dd, J=5.8, 17.8 Hz, 1H; CH2Ind, C), 3.37
(m, 2H; CH2Ind, F), 5.11 (ddd, J=1.4, 5.8, 7.2 Hz, 1H; OCHoxa, C), 5.15
(dd, J=5.7, 7.2 Hz, 1H; OCHoxa, C), 5.41 (dd, J=5.9, 7.3 Hz, 1H;
OCHoxa, F), 5.49 (d, J=7.6 Hz, 1H; NCHoxa, F), 6.25 (d, J=6.8 Hz, 1H;
NCHoxa, C), 6.26 (d, J=6.5 Hz, 1H; NCHoxa, C), 7.16–7.46 (m, 10H;
CHarom), 8.41 (m, 1H; CHarom), 8.50 ppm (m, 1H; CHarom); 13C {1H} NMR
(100 MHz, 1,1,2,2-[D2]tetrachloroethane, 296 K): d=20.8 (CH3), 37.3,
Preparation of the palladium complexes
[Pd
(trisox)Cl2] complexes—general procedure:[41] Bis(benzonitrile)palla-
G
dium(II) dichloride (0.142 mmol) and the required trisox derivative
(0.149 mmol) were dissolved in CH2Cl2 (1 mL). The reaction mixture was
stirred for 90 min at room temperature and pentane was added (8 mL) to
form an orange precipitate. The crude product was washed twice with
pentane (8 mL) and dried in vacuo to give the complex as an orange
powder. (In the following NMR notation C=coordinated oxazoline, F=
free oxazoline.)
38.8, (CH2Ind), 45.2 ((CH3)Coxa3), 73.3, 74,0 (NCHoxa, C), 76.4 (NCHoxa
,
(1,1,1-Tris[(S)-4-isopropyloxazolin-2-yl]ethane)palladium(II) dichloride
(1a): Yield: 85%. Crystallisation from CH2Cl2/Et2O gave orange crystals
suitable for X-ray diffraction. 1H NMR (300 MHz, CDCl3, 296 K): d=
F), 85.1, 86.7, (OCHoxa, C), 87.8, (OCHoxa, F), 124.9, 125.0, 125.3, 125.7,
127.6, 127.9, 128.0, 128.1, 128.3, 128.9, 129.7, 129.9 (Carom), 138.2, 138.5,
138.5, 138.6 (Cquat-arom, C), 139.5, 140.5 (Cquat-arom, F), 160.3 (NCO, F),
166.6, 167.6 ppm (NCO, C); 15N (60 MHz, 1,1,2,2-[D2]tetrachloroethane,
296 K): d=161.5 162.8 (N, C), 238.3 ppm (N, F); FT-IR (KBr): n˜ =1653
(s, C=N free oxazoline), 1649 cmÀ1 (s, C=N coordinated oxazoline);
HRMS (ESI): m/z (%): 702.035 (100) [M++Na] 644.077 [M+ÀCl]; ele-
mental analysis calcd (%) for C32H27Cl2N3O3Pd: C 56.61, H 4.01, N 6.19;
found: C 56.70, H 4.18, N 6.10.
0.74 (m, 6H; CH
J=6.8 Hz, 3H; CH
CH3apical), 2.87 (m, 2H; CH
3H; CH2oxa, C, CHoxa, F), 4.44 (m, 2H; CH2oxa, C), 4.66 (m, 1H; CHoxa
C), 4.81 ppm (m, 1H; CHoxa, C); 13C {1H} NMR (75 MHz, CDCl3, 296 K):
d=13.1, 13.6, 17.8, 18.3, 18.5, 18.6 (CH(CH3)2), 22.5 (CH3apical), 29.3, 29.7
(CH(CH3)2, C), 32.1 (CH(CH3)2, F), 45.2 ((CH3)C(oxa)3), 69.3, 69.6
A
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
,
AHCTREUNG
(h3-Allyl)(1,1,1-tris[(R)-4-phenyloxazolin-2-yl]ethane)palladium(II)
(2):[42] Ph-trisox (65 mg, 0.14 mmol) in dry CH2Cl2 (2 mL) was added to a
solution of [Pd(cod)
(h3-C3H5)]BF4 (47.3 mg, 0.14 mmol) in dry CH2Cl2
A
N
ACHTREUNG
(CH2oxa, C), 70.0, 70.4 (CHoxa, C), 71.5 (CH2oxa, F), 72.0 (CHoxa, F), 160.1
(NCO, F), 165.8, 166.7 ppm (NCO, C); 15N (60 MHz, CDCl3, 296 K): d=
160.2 161.2 (N, C) 239.9 ppm (N, F); FT-IR (KBr): n˜ =1660 (s, C=N free
oxazoline), 1650 cmÀ1 (s, C=N coordinated oxazoline); HRMS (ESI): m/z
(%): 564.084 (85) [M++Na], 1105.177 (100) [2M++Na]; elemental analy-
sis calcd (%) for C20H33Cl2N3O3Pd: C 44.42, H 6.15, N 7.77; found: C
44.28, H 6.27, N 7.67.
A
ACHTREUNG
(1 mL). The reaction mixture was stirred for 45 min, filtered through
Celite and washed with CH2Cl2 (2”1 mL). The solvents were evaporated
to give a white powder which was washed twice with pentane (5 mL) and
dried under vacuum to yield the complex (50 mg, 52%). Suitable crystals
for an X-ray diffraction study were obtained by slow diffusion of pentane
into a solution of the complex in CH2Cl2. 1H NMR (400 MHz, CDCl3,
296 K): d=1.89 (d, J=12.5 Hz, 1H; HAallyl), 2.22 (s, 3H; CH3), 2.52 (d,
J=12.6 Hz, 1H; HAallyl), 2.78 (dd, J=2.0, 7.0 Hz, 1H; HSallyl), 3.44 (d, J=
6.9 Hz, 1H; HAallyl), 4.40 (pseudo-t, J=8.3 Hz, 3H; CH2oxa), 4.89 (m, 1H,
HCallyl), 5.03 (dd, J=8.7, 10.4 Hz, 3H; CH2oxa), 5.47 (dd, J=7.9, 10.4 Hz,
3H; CHoxa), 7.33 ppm (m, 15H; CHarom); 13C {1H} NMR (100 MHz,
(1,1,1-Tris
A
dichloride
(1b): Yield: 89%. Crystallisation from CH2Cl2/pentane gave orange crys-
tals suitable for X-ray diffraction. 1H NMR (600 MHz, CDCl3, 296 K):
d=2.27 (s, 3H; CH3), 4.36 (pseudo-t, J=8.4 Hz, 1H; CH2oxa, F), 4.58–
4.61 (m, 2H; CH2oxa, C), 4.75 (pseudo-t, J=8.9 Hz, 1H; CH2oxa, C), 4.80
(pseudo-t, J=9.2 Hz, 1H; CH2oxa, C), 4.90 (dd, J=8.6, 10.2 Hz, 1H;
CH2oxa, F), 5.40 (dd, J=8.3, 10.1 Hz, 1H; CHoxa, F), 5.91 (dd, J=2.3,
9.0 Hz, 1H; CHoxa, C), 6.08 (dd, J=3.5, 9.6 Hz, 1H; CHoxa, C), 7.25–7.27
(m, 3H; CHarom), 7.32–7.40 ppm (m, 15H; CHarom); 13C {1H} NMR
CDCl3, 296 K): d=20.8 (CH3), 45.9 ((CH3)C(oxa)3), 61.1 (C1allyl, C3allyl),
N
71.6 (CHoxa), 77.0 (CH2oxa), 115.6 (C2allyl), 127.0, 128.6, 129.2 (Carom), 140.2
(Cquat-arom), 167.6 ppm (NCO); FT-IR (KBr): n˜ =1658 cmÀ1 (s, C=N); MS
(FAB): m/z (%): 612.1 (100) [M+ÀBF4]; elemental analysis calcd (%) for
C32H32N3O3PdBF4 with CH2Cl2: C 50.51, H 4.37, N 5.35; found: C 50.47,
H 4.32, N 5.41.
(150 MHz, CDCl3, 296 K): d=22.4 (CH3), 45.6 ((CH3)C
A
(CHoxa, C), 68.7 (CHoxa, C) 69.7 (CHoxa, F), 76.7 (CH2oxa, F), 76.9 (CH2oxa
,
F), 77.3 (CH2oxa, C), 77.4 (CH2oxa, C), 126.1, 126.6, 126.8 (Carom, F), 128.3,
128.4, 128.5, 128.9, 129.0, 129.1 (Carom, C), 139.4, 139.5 (Cquat-arom, C), 140.5
[Pd(ma)
ACHTREUNG
,
F), 161.8 (NCO, NC), 167.1, 168.0 ppm (NCO, C); 15N
in dry THF (3 mL) was added to a solution of [Pd0(h2,h2-nbd)
ACHTREUNG
(Cquat-arom
(60 MHz, CDCl3, 296 K): d=160.8 161.6 (N, C) 239.9 ppm (N, F); FT-IR
(KBr): n˜ =1655 cmÀ1 (s, C=N); HRMS (FAB): m/z (%): 570.103 (100)
(0.54 mmol) in dry THF (4 mL). After stirring for 15 min, the reaction
mixture was filtered through Celite and concentrated under reduced
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6005

L. H. Gade, S. Bellemin-Laponnaz et al.
pressure to 2 mL. Pentane (10 mL) was then added to the solution, after
which a yellow powder was obtained. The crude solid was washed with
pentane (3”7 mL) and dried in vacuo to give the complex as a yellow
powder.
CH2Ind), 5.40 (pseudo-t, J=6.6 Hz, 3H; OCHoxa), 5.66 (d, J=7.4 Hz, 3H;
NCHoxa) 7.28 (m, 3H; CHarom), 7.38 (m, 6H; CHarom), 7.73 ppm (m, 3H;
CHarom); 13C NMR (100 MHz, CD2Cl2, 296 K): d=20.6 (CH3), 38.6
(CH2Ind), 45.6 ((CH3)C(oxa)3), 77.8 (NCHoxa), 86.9 (OCHoxa), 114.6, 115.0
A
(Ctcne), 125.7, 126.2, 128.3, 129.8 (Carom), 139.3, 139.6 (Cquat-arom), 159.7 ppm
(NCO); 1H NMR (400 MHz, CD2Cl2, 203 K): d=1.54 (s, 3H; CH3), 1.88
(d, J=18.2 Hz, 1H, CH2Ind), 2.42 (d, J=18.3 Hz, 1H; CH2Ind), 3.01 (dd,
J=6.8, 18.4 Hz, 1H; CH2Ind), 3.18 (dd, J=5.2, 18.3 Hz, 1H; CH2Ind), 3.38
(d, J=18.4 Hz, 1H; CH2Ind), 3.46 (dd, J=5.7, 19.2 Hz, 1H; CH2Ind), 5.14
(pseudo-t, J=7.0 Hz, 1H; OCHoxa), 5.36 (pseudo-t, J=6.1 Hz, 1H;
OCHoxa), 5.53 (d, J=7.8 Hz, 1H; NCHoxa), 5.58 (d, J=7.0 Hz, 1H;
NCHoxa), 5.61 (pseudo-t, J=6.4 Hz, 1H; OCHoxa), 5.76 (d, J=7.2 Hz,
1H; NCHoxa), 7.35 (m, 10H; CHarom), 7.76 (d, J=7.8 Hz, 1H; CHarom);
7.87 ppm (d, J=7.5 Hz, 1H; CHarom); 13C NMR (100 MHz, CD2Cl2,
203 K): d=20.6 (CH3), 37.1, 37.4, 38.1, 38.4 (CH2Ind), 44.3 ((CH3)C-
(1,1,1-Tris[(S)-4-isopropyloxazolin-2-yl]ethane)(maleic anhydride)palladi-
um(0) (3a): Yield: 54%. Crystallisation from Et2O/pentane gave yellow
crystals suitable for an X-ray diffraction study. 1H NMR (300 MHz,
CDCl3, 296 K): d=0.81 (d, J=6.8 Hz, 9H; CH
ACHTREUNG
9H; CH(CH3)2), 1.85 (s, 3H; CH3apical), 2.23 (brm, 3H; CH
A
ACHTREUNG
(d, J=3.7 Hz, 1H; C4H2O3), 3.74 (d, J=3.7 Hz, 1H; C4H2O3), 4.20 (m,
6H; CH2oxa), 4.33 ppm (m, 3H; CHoxa); 13C {1H} NMR (75 MHz, CDCl3,
296 K): d=14.8 (CH
A
ACHTREUNG
A
ACHTREUNG
72.8 (CH2oxa), 164.3 (NCO), 172.3 (Cquat-ma), 172.8 ppm (Cquat-ma); FT-IR
(KBr): n˜ =1786, 1722 (C=O), 1660 (C=N, coordinated oxazoline),
1650 cmÀ1 (C=N, free oxazoline); HRMS (ESI): m/z (%): 590.162 (100)
[M++Na], 1159.329 (12) [2M++Na]; elemental analysis calcd (%) for
C24H35Cl2N3O6Pd: C 50.75, H 6.21, N 7.40; found: C 50.86, H 6.19, N
7.37;
A
113.8, 114.4 (Ctcne), 124.8, 125.1, 127.6, 129.5 (Carom), 137.5, 138.9 (Cquat-
arom), 160.2, 168.0 ppm (NCO); FT-IR (KBr): n˜ =1653 (C=N, coordinated
oxazoline), 1649 cmÀ1 (C=N, free oxazoline); HRMS (FAB): m/z (%)
607.115 (100) [M+Àtcne], 735.129 (50) [M+].
(1,1,1-Tris[(R)-4-phenyloxazolin-2-yl]ethane)(maleic anhydride)palladi-
1
um(0) (3b): Yield: 75%; H NMR (600 MHz, CD2Cl2, 296 K): d=2.16 (s,
Procedure for the palladium-catalysed allylic alkylation reaction: Ade-
gassed solution of [{Pd
(h3-C3H5)Cl}2] (6.4 mmol, 1.1 mol%) and ligand
U
3H; CH3), 2.92 (d, J=3.9 Hz, 1H; C4H2O3), 3.30 (d, J=3.8 Hz, 1H;
C4H2O3), 4.46 (brs, 3H; CH2oxa), 4.86 (pseudo-t, J=9.2 Hz, 3H; CH2oxa),
5.40 (dd, J=6.7, 10.1 Hz, 3H; CHoxa), 7.33–7.48 ppm (m, 15H; CHarom);
13C {1H} NMR (150 MHz, CD2Cl2, 296 K): d=22.2 (CH3), 38.9 (CHma),
(14.5 mmol, 2.5 mol%) in THF (0.7 mL) was stirred at 508C for 1.5 h.
After cooling down to room temperature, rac-1,3-diphenylprop-2-enyl
acetate (147 mg, 0.58 mmol) in THF (2.2 mL) dimethyl malonate
(0.2 mL, 1.75 mmol), BSA(0.4 mL, 1.75 mmol) and a few milligrams of
potassium acetate were added. The reaction mixture was stirred at room
temperature. After the desired reaction time the reaction mixture was di-
luted with CH2Cl2, washed with a saturated aqueous solution of ammoni-
um chloride and the organic extract was dried over Na2SO4. The residue
was purified by flash chromatography and the ee of the product was de-
termined by HPLC using a Daicel Chiralpak AD-H column. Yields and
ee values are the average of at least two corroborating runs. The absolute
configuration was assigned by comparing the optical rotation value with
literature data.[43]
41.4 (CHma), 45.6 ((CH3)C
ACHTREUNG
(CHoxa), 127.0, 127.2, 127.9, 128.7 , 129.1
AHCTREUNG
(NCO), 171.9 (Cquat-ma), 172.7 ppm (Cquat-ma); FT-IR (KBr) n˜ =1798, 1727
(C=O), 1662 (C=N, coordinated oxazoline), 1655 cmÀ1 (C=N, free oxazo-
line); HRMS (FAB): m/z (%): 571.114 (100) [M+Àma]; elemental analy-
sis calcd (%) for C33H29N3O6Pd: C 59.16, H 4.36, N 6.27; found: C 58.02,
H 4.32, N 6.35.
(1,1,1-Tris[(4S)-4-benzyloxazolin-2-yl]ethane)(maleic anhydride)palladi-
R
1
um(0) (3c): Yield: 56%; H NMR (400 MHz, CD2Cl2, 296 K): d=1.53 (s,
3H; CH3), 2.78 (dd, J=8.1, 13.2 Hz, 2H; CH2Bn, 1H; C4H2O3), 3.13
(brm, 2H; CH2Bn, 1H; C4H2O3) 3.87 (brm, 2H; CH2Bn), 4.14 (brm, 3H;
CH2oxa), 4.27 (pseudo-t, J=8.8 Hz, 3H; CH2oxa), 4.51 (m, 3H; CHoxa),
7.22–7.34 ppm (m, 15H; CHarom); 13C {1H} NMR (100 MHz, CD2Cl2,
296 K): d=21.8 (CH3), 39.8 (CHma), 40.5 (CH2Bn), 41.2 (CHma), 45.0
The comparative studies of the catalytic activities were conducted for
each experiment using palladium catalysts prepared in situ by reacting
the respective ligand with the [{Pd
(h3-C3H5)Cl}2] dimer at 508C for 1.5 h
R
in THF. The catalytic allylic alkylations were carried out at room temper-
ature. The progress of the reaction was monitored by measuring the ap-
pearance of the product by GC, using dodecane as internal standard.
((CH3)C(oxa)3), 68.2 (CHoxa), 72.8 (CH2oxa), 127.1, 128.9, 130.2 (Carom),
G
139.8 (Cquat-arom), 164.4 (NCO), 172.8 (Cquat-ma), 173.6 ppm (Cquat-ma); FT-IR
(KBr): n˜ =1793, 1724 (C=O), 1662 (C=N, coordinated oxazoline),
Crystal structure determinations: Single crystals the complexes 1a, 1b, 2,
3a and 3b were obtained by layering concentrated solutions in polar sol-
vents (dichloromethane or diethyl ether) with pentane or diethyl ether
and allowing slow diffusion at room temperature. Intensity data were col-
lected at low temperature on Nonius Kappa CCD (1a, 3a) and Bruker
Smart 1000 CCD (1b, 2, 3b) diffractometers. Crystals of 3b, although of
nice appearance, were invariably found to be multiples. Datasets from
two different crystals could only be partially de-twinned, resulting in un-
acceptably high Rint values. The structure could be solved but not refined
to a satisfactory level. Crystal data and experimental details are given in
Table 6.
1655 cmÀ1
(C=N,
free
oxazoline);
HRMS
(FAB):
m/z (%): 614.162 (100) [M+Àma].
U
G
hydride)palladium(0) (3e): Yield: 69%; 1H NMR (600 MHz, CD2Cl2,
296 K): d=0.48–0.53 (m, 2H; CH(CH3)2), 0.74–0.98 (m, 10H; CH-
(CH3)2), 1.71, 173 (s, 3H; CH3apical), 2.33 (brm, 1H; CH(CH3)2), 2.45
(brm, 1H; CH(CH3)2), 3.43 (d, J=15.2 Hz, 1H; CH2 Py), 3.52 (d, J=
AHCTREUNG
A
ACHTREUNG
ACHTREUNG
16.1 Hz, 1H; CH2 Py), 3.60–3.68 (m, 2H; C4H2O3), 4.02–4.32 (m, 6H;
CH2oxa, CHoxa), 7.11 (m, 2H; CHarom), 7.61 (m, 1H; CHarom), 8.41 ppm (m,
1H; CHarom); 13C {1H} NMR (150 MHz, CD2Cl2, 296 K): d=13.6,13.8,
18.4 (CH
N
G
Data were corrected for Lorentz, polarisation and absorption effects
(semiempirical[44] or empirical[45]). The structures were solved using heavy
atom or direct methods and refined by a full-matrix least-squares proce-
dure based on F2 with all measured unique reflections. All non-hydrogen
atoms were given anisotropic displacement parameters. Hydrogen atoms
were input at calculated positions and refined with a riding model. In the
structure of 2, dichloromethane was found as a solvent of crystallisation.
The calculations were performed by using the programs DIRDIF,[46]
SIR,[47] SHELXS-86,[48] SHELXL-97[49] and OpenMoleN.[50] Graphical
representations were drawn with XP.[51] Anisotropic displacement ellip-
soids are scaled to 25% probability.
(CHma), 43.9 (CH2Bn), 42.7 ((CH3)C
AHCTREUNG
(CH2oxa), 121.6, 122.8, 136.0 (Carom), 156.3 (Cquat-arom), 169.1 (NCO), 172.8,
173.8 ppm (Cquat-ma); FT-IR (KBr): n˜ =1794, 1723 (C=O), 1664 (C=N, co-
ordinated oxazoline), 1653 cmÀ1 (C=N free oxazoline); HRMS (FAB): m/
z (%): 449.131 (100) [M+Àma].
(1,1,1-Tris
ene)palladium(0) (3d): Ind-trisox (105 mg, 0.21 mmol) in THF (2 mL)
was added to a solution of [Pd(h2,h2-nbd)(h2-tcne)] (68 mg, 0.21 mmol) in
A
ACHTREUNG
AHCTREUNG
THF (2 mL). After stirring for 2 h, the reaction mixture was filtered
through Celite and concentrated under reduced pressure to 2 mL. Pen-
tane (10 mL) was then added to the solution, after which a yellow
powder was obtained. The crude solid was washed three times with pen-
tane (7 mL) and dried in vacuo to yield 70 mg (50%) of the complex as a
yellow powder. 1H NMR (400 MHz, CD2Cl2, 296 K): d=1.60 (s, 3H;
CH3), 2.77 (d, J=18 Hz, 3H; CH2Ind), 3.28 (dd, J=6.4, 18.3 Hz, 3H;
CCDC-633244 (1a), CCDC-633245 (1b), CCDC-633246 (2) and CCDC-
633247 (3a) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
6006
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008

Asymmetric Catalysis
FULL PAPER
Table 6. X-ray experimental data of the crystallographically characterised compounds.
1a
1b
2
3a
formula
Mr
C20H33Cl2N3O3Pd
540.81
C29H27Cl2N3O3Pd
642.84
C33H34BCl2F4N3O3Pd
784.74
C24H35N3O6Pd
567.96
T [K]
173
100(2)
100(2)
173
crystal system
space group
a [Š]
b [Š]
c [Š]
orthorhombic
P212121
8.7083(1)
10.8395(1)
24.8911(4)
monoclinic
P21
monoclinic
P21
9.1344(4)
20.6281(10)
9.3599(4)
111.4780(10)
1641.17(13)
2
monoclinic
P21
10.2954(10)
10.0926(10)
13.0348(13)
101.537(2)
1327.0(2)
10.4450(2)
10.0975(2)
14.0887(3)
90.161(5)
1485.91(5)
2
b [8]
V [Š3]
Z
2349.56(5)
4
1.53
2
1calcd [MgmÀ3
]
1.609
0.938
1.588
0.790
1.27
0.661
m [mmÀ1
]
1.042
F000
1112
652
796
588
crystal size [mm3]
q range [8]
0.10”0.06”0.06
2.5–30.02
À12ꢁhꢁ12
À15ꢁkꢁ15
À34ꢁlꢁ35
0.20”0.05”0.05
2.02–32.08
À15ꢁhꢁ14
À14ꢁkꢁ15
0ꢁlꢁ19
0.30”0.30”0.30
1.97–32.01
À13ꢁhꢁ12
À30ꢁkꢁ25
0ꢁlꢁ13
0.20”0.16”0.10
2.5–30.05
À14ꢁhꢁ14
À14ꢁkꢁ14,
À19ꢁlꢁ19
8556
index ranges
reflns collected
32759
15547
independent reflns [Rint
completeness to q=30.028 [%]
]
6871 [0.040]
99.9
8646 [0.0583]
95.6
8670 [0.0224]
100
4588 [0.040]
99.8
absorption correction
empirical (SHELXA)
semi-empirical
from equivalents
0.7458/0.6070
8646/1/344
1.084
semi-empirical
from equivalents
0.7974/0.7974
8670/8/43
empirical (SHELXA)
max/min transmission
0.928/0.901
4997/0/262
1.323
1.0000/0.9401
3495/1/306
1.966
data/restraints/parameters
goodness-of-fit on F2
1.072
final R indices [I>2s(I)]
R indices (all data)
Abs. struct. parameter
R1=0.033, wR2=0.037
R1=0.058, wR2=0.108
À0.02(3)
R1=0.0372, wR2=0.0688
R1=0.0560, wR2=0.0750
0.000(19)
1.430/À0.609
R1=0.0290, wR2=0.0748
R1=0.0303, wR2=0.0760
À0.021(15)
1.218/À0.603
R1=0.038, wR2=0.061
R1=0.051, wR2=0.120
À0.04(3)
À3
Res. density, max diff. peak/hole [eŠ
]
1.115/À0.304
0.895/À0.317
[4] Asymmetric catalysis with chiral aminoalcohols as ligands: a) W. A.
Nugent, R. L. Harlow, J. Am. Chem. Soc. 1994, 116, 6142; b) F. Di-
Furia, G. Licini, G. Modena, R. Motterle, W. A. Nugent, J. Org.
Chem. 1996, 61, 5175; c) H. Lütjens, P. Knochel, Tetrahedron: Asym-
metry 1994, 5, 1161; d) H. Lütjens, G. Wahl, F. Mçller, F. Knochel, J.
Sundermeyer, Organometallics 1997, 16, 5869; e) M. Cernerud, H.
Adolfsson, C. Moberg, Tetrahedron: Asymmetry 1997, 8, 2655;
f) W. A. Nugent, G. Licini, M. Bonchio, O. Bortolini, M. G. Finn,
B. W. McCleland, Pure Appl. Chem. 1998, 70, 1041; g) M. Bonchio,
G. Licini, F. Di Furia, S. Mantovani, G. Modena, W. A. Nugent, J.
Org. Chem. 1999, 64, 1326; h) M. Bonchio, G. Licini, G. Modena, O.
Bortolini, S. Moro, W. A. Nugent, J. Am. Chem. Soc. 1999, 121,
6258; i) M. Bonchio, O. Bortolini, G. Licini, S. Moro, W. A. Nugent,
Eur. J. Org. Chem. 2003, 507; j) M. G. Buonomenna, E. Drioli,
W. A. Nugent, L. J. Prins, P. Scrimin, G. Licini, Tetrahedron Lett.
2004, 45, 7515.
[5] Recent advances in the design of C3-chiral podands: a) G. Bring-
mann, M. Breuning, R.-M. Pfeifer, P. Schreiber, Tetrahedron: Asym-
metry 2003, 14, 2225; b) G. Bringmann, R.-M. Pfeifer, C. Rummey,
K. Hartner, M. Breuning, J. Org. Chem. 2003, 68, 6859; c) T. Fang,
D.-M. Du, S.-F. Lu, J. Xu, Org. Lett. 2005, 7, 2081; d) M. P. Castaldi,
S. E. Gibson, M. Rudd, A. J. P. White, Angew. Chem. 2005, 117,
3498; Angew. Chem. Int. Ed. 2005, 44, 3432; e) M. P. Castaldi, S. E.
Gibson, M. Rudd, A. J. P. White, Chem. Eur. J. 2006, 12, 138.
[6] For a discussion of the special case of five-coordinate complexes
bearing C3-chiral tripods, see: L. H. Gade, P. Renner, H. Memmler,
F. Fecher, C. H. Galka, M. Laubender, S. Radojevic, M. McPartlin,
J. W. Lauher, Chem. Eur. J. 2001, 7, 2563.
Acknowledgements
We thank the Deutsche Forschungsgemeinschaft (SFB 623), the CNRS
(France) and the Deutsch-Franzçsische Hochschule for funding. Support
by BASF (Ludwigshafen) and Degussa (Hanau) is gratefully acknowl-
edged. We also thank Dr A. De Cian and N. Gruber for some of the X-
Ray diffraction studies.
[1] a) C. Moberg, Angew. Chem. 1998, 110, 260; Angew. Chem. Int. Ed.
1998, 37, 248; b) M. C. Keyes, W. B. Tolman, Adv. Catal. Processes
1997, 2, 189.
[2] a) S. E. Gibson, M. P. Castaldi, Chem. Commun. 2006, 3045; b) S. E.
Gibson, M. P. Castaldi, Angew. Chem. 2006, 118, 4834; Angew.
Chem. Int. Ed. 2006, 45, 4718; c) C. Moberg, Angew. Chem. 2006,
118, 4838; Angew. Chem. Int. Ed. 2006, 45, 4721.
[3] Examples of the use of C3-chiral ligands in asymmetric catalysis:
a) H. Brunner, A. F. M. M. Rahman, Chem. Ber. 1984, 117, 710;
b) M. J. Burk, R. L. Harlow, Angew. Chem. 1990, 102, 1511; Angew.
Chem. Int. Ed. Engl. 1990, 29, 1467; c) M. J. Burk, J. E. Feaster,
R. L. Harlow, Tetrahedron: Asymmetry 1991, 2, 569; d) T. R. Ward,
L. M. Venanzi, A. Albinati, F. Lianza, T. Gerfin, V. Gramlich,
G. M. R. Tombo, Helv. Chim. Acta 1991, 74, 983; e) M. J. Baker, P. J.
Pringle, J. Chem. Soc. Chem. Commun. 1993, 314; f) H. Adolfsson,
K. Wärnmark, C. Moberg, J. Chem. Soc. Chem. Commun. 1992,
1054; g) H. Adolfsson, K. Nordstrçm, K. Wärnmark, C. Moberg,
Tetrahedron: Asymmetry 1996, 7, 1967; h) D. D. LeCloux, W. B.
Tolman, J. Am. Chem. Soc. 1993, 115, 1153; i) C. J. Tokar, P. B. Ket-
tler, W. B. Tolman, Organometallics 1992, 11, 2737; j) D. D. Lecloux,
C. J. Tokar, M. Osawa, R. P. Houser, M. C. Keyes, W. B. Tolman, Or-
ganometallics 1994, 13, 2855; k) K. Kawasaki, S. Tsumura, T. Katsu-
ki, Synlett 1995, 1245.
[7] H. B. Kagan, T.-P. Dang, J. Am. Chem. Soc. 1972, 94, 6429.
[8] A. von Zelewsky, Stereochemistry of Coordination Compounds,
Wiley, Chichester, 1996.
[9] a) S. Bellemin-Laponnaz, L. H. Gade, Chem. Commun. 2002, 1286;
b) S. Bellemin-Laponnaz, L. H. Gade, Angew. Chem. 2002, 114,
Chem. Eur. J. 2007, 13, 5994 – 6008
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
6007

L. H. Gade, S. Bellemin-Laponnaz et al.
3623; Angew. Chem. Int. Ed. 2002, 41, 3473; c) L. H. Gade, G. Mar-
coni, C. Dro, B. D. Ward, M. Poyatos, S. Bellemin-Laponnaz, H. Wa-
depohl, L. Sorace, G. Poneti, Chem. Eur. J. 2007, 13, 3058.
[10] C. Foltz, B. Stecker, G. Marconi, H. Wadepohl, S. Bellemin-Lapon-
naz, L. H. Gade, Chem. Commun. 2005, 5115.
[11] a) B. M. Trost, P. E. Strege, J. Am. Chem. Soc. 1977, 99, 1649; the
first asymmetric allylic alkylation was reported four years earlier:
b) B. M. Trost, T. J. Dietsche, J. Am. Chem. Soc. 1973, 95, 8200.
[12] a) B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996, 96, 395; b) A.
Heumann in Transition Metals for Organic Synthesis, 2nd ed. (Eds.:
M. Beller, C. Bolm), Wiley-VCH, Weinheim, 2004, p. 307; c) J.-F.
Paquin; M. Leutens, Compr. Asymmetric Catal. Suppl. 2004, 2, 73;
d) G. Helmchen, M. Ernst, G. Paradies, Pure Appl. Chem. 2004, 76,
495; e) T. Graening, H.-G. Schmalz, Angew. Chem. 2003, 115, 2684;
Angew. Chem. Int. Ed. 2003, 42, 2580; f) C. Moberg, U. Bremberg,
K. Hallman, M. Svensson, P.-O. Norrby, A. Hallberg, M. Larhed, I.
Csoregh, Pure Appl. Chem. 1999, 71, 1477.
L. Vicentini, A. Sessanta o Santi, E. Zangrando, S. Geremia, G.
Mestroni, Organometallics 1997, 16, 5064; g) A. M. Kluwer, T. S. Ko-
blenz, T. Jonischkeit, K. Woelk, C. J. Elsevier, J. Am. Chem. Soc.
2005, 127, 15470.
[24] K. Itoh, F. Ueda, K. Hirai, Y. Ishii, Chem. Lett. 1977, 877.
[25] a) D. D. Ellis, A. L. Spek, Acta Crystallogr. Sect. C 2001, 57, 235;
b) 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.
[26] a) T. Schleis, J. Heinemann, T. P. Spaniol, R. Mülhaupt, J. Okuda,
Inorg. Chem. Commun. 1998, 1, 431; b) M. L. Ferrara, F. Giordana,
I. Orabona, A. Panunzi, F. Ruffo, Eur. J. Inorg. Chem. 1999, 1939.
[27] M. Lutz, Acta Crystallogr. Sect. E 2001, 57, o1136.
[28] R. Fernandez-Galan, F. A. Jalon, B. R. Manzano, J. Rodriguez-de la
Fuente, Organometallics 1997, 16, 3758.
[29] R. Van Asselt, C. J. Elsevier, W. J. J. Smeets, A. L. Spek, Inorg.
Chem. 1994, 33, 1521.
[13] a) B. M. Trost in Catalytic Asymmetric Synthesis (Ed.: I. Ojima),
Wiley-VCH, Weinheim, 2000, Chapter 8E; b) B. M. Trost, M. R. Ma-
chacek, A. Aponick, Acc. Chem. Res. 2006, 39, 747.
[14] a) A. Togni, L. M. Venanzi, Angew. Chem. 1994, 106, 517; Angew.
Chem. Int. Ed. Engl. 1994, 33, 497; b) B. M. Trost, C. Lee, Catalytic
Asymmetric Synthesis, 2nd ed. (Ed.: I. Ojima), Wiley-VCH, Wein-
heim 2000; c) R. PrØtꢂt, G. C. Lloyd-Jones, A. Pfaltz, Pure Appl.
Chem. 1998, 70, 1035; d) T. Hayashi, J. Organomet. Chem. 1999, 576,
195; e) G. Helmchen, J. Organomet. Chem. 1999, 576, 203; f) G.
Helmchen, A. Pfaltz, Acc. Chem. Res. 2000, 33, 336.
[30] B. M. Trost, S. J. Brickner, J. Am. Chem. Soc. 1983, 105, 568.
[31] P. von Matt, G. C. Lloyd-Jones, A. B. E. Minidis, A. Pfaltz, L.
Macko, M. Neuburger, M. Zehnder, H. Rüeger, P. S. Pregosin, Helv.
Chim. Acta 1995, 78, 265.
[32] We have observed a similar behaviour for related allylic substitu-
tions which will be reported elsewhere.
[33] A. Abiko, S. Masamune, Tetrahedron Lett. 1992, 33, 5517.
[34] J. Bourguignon, U. Bremberg, G. Dupas, K. Hallman, L. Hagberg,
L. Hortala, V. Levacher, S. Lutsenko, E. Macedo, C. Moberg, G.
QuØguiner, F. Rahm, Tetrahedron 2003, 59, 9583.
[15] This sidearm strategy has been succesfully employed for the devel-
opment of efficient ligands for a range of copper-catalysed enantio-
selective reactions: a) J. Zhou, M. C. Ye, Y. Tang, J. Comb. Chem.
2004, 6, 301; b) Z.-Z. Huang, Y.-B. Kang, J. Zhou, M.-C. Ye, Y.
Tang, Org. Lett. 2004, 6, 1677; c) M.-C. Ye, J. Zhou, Y. Tang, J. Org.
Chem. 2006, 71, 3576.
[35] W. R. Leonard, J. L. Romine, A. I. Meyers, J. Org. Chem. 1991, 56,
1961.
[36] D. M. Barnes, J. Ji, M. G. Fickes, M. A. Fitzgerald, S. A. King, H. E.
Morton, F. A. Plagge, M. Preskill, S. H. Wagaw, S. J. Wittenberger, J.
Zhang, J. Am. Chem. Soc. 2002, 124, 13097.
[37] D. A. White, Inorg. Synth. 1972, 13, 55.
[16] For a review see: J. Zhou, Y. Tang, Chem. Soc. Rev. 2005, 34, 664.
[17] A. El Hatimi, M. Gꢃmez, S. Jansat, G. Muller, M. Font-Bardꢄa, X.
Solans, J. Chem. Soc. Dalton Trans. 1998, 4229.
[38] M. Honma, T. Sawada, Y. Fujisawa, M. Utsugi, H. Wanatabe, A.
Umino, T. Matsamura, T. Hagihara, M. Takano, M. Nakada, J. Am.
Chem. Soc. 2003, 125, 2860.
[18] H. M. McConnell, J. Chem. Phys. 1958, 28, 430.
[19] M. L. H. Green, L.-L. Wong, A. Sella, Organometallics 1992, 11,
2660.
[39] a) A. I. Meyers, K. A. Novachek, Tetrahedron Lett. 1996, 37, 1747;
b) C. Foltz, S. Bellemin-Laponnaz, L. H. Gade, unpublished results.
[40] M. Seitz, C. Capacchione, S. Bellemin-Laponnaz, H. Wadepohl,
B. D. Ward, L. H. Gade, Dalton Trans. 2006, 193.
[41] S. E. Denmark, R. A. Stavenger, A.-M. Faucher, J. P. Edwards, J.
Org. Chem. 1997, 62, 3375.
[20] a) J. B. Goddard, F. Basolo, Inorg. Chem. 1968, 7, 936; b) L. A. P.
Kane-Maguire, G. Thomas, J. Chem. Soc. Dalton Trans. 1975, 19,
1890. Studies published more recently include: c) T. Shi, L. I.
Elding, Inorg. Chem. 1996, 35, 735; d) T. Shi, L. I. Elding, Inorg.
Chem. 1997, 36, 528; e) Z. D. Bugarꢅic, G. Liehr, R. van Eldik, J.
Chem. Soc. Dalton Trans. 2002, 951.
[21] a) O. Hoarau, H. Ait-Haddou, J.-C. Daran, D. Cramailere, G. G. A.
Balavoine, Organometallics 1999, 18, 4718; b) M. Kehnder, M. Neu-
burger, P. von Matt, A. Pfaltz, Acta Crystallogr. Sect. C 1995, 51,
1109.
[42] D. Franco, M. Gomez, F. JimØnez, G. Muller, M. Rocamora, M. A.
Maestro, J. Mahia, Organometallics 2004, 23, 3197.
[43] J. Sprinz, G. Helmchen, Tetrahedron Lett. 1993, 34, 1769.
[44] G. M. Sheldrick, SADABS-2004/1, Bruker AXS, 2004.
[45] N. Walker, D. Stuart, Acta Crystallogr. Sect. A 1983, 39, 158.
[46] P. T. Beurskens, G. Beurskens, R. de Gelder, S. Garcia-Granda,
R. O. Gould, R. Israel, J. M. M. Smits, DIRDIF-99, University of
Nijmegen, The Netherlands, 1999.
[47] M. C. Burla, M. Camalli, G. Cascarano, C. Giacovazzo, G. Polidori,
R. Spagna, D. Viterbo, J. Appl. Crystallogr. 1989, 22, 389.
[48] G. M. Sheldrick, SHELXS-86, University of Gçttingen, 1986; G. M.
Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467.
[22] M. A. Pericas, C. Puigjamer, A. Riera, A. Vidal-Ferran, M. Gomez,
F. Jimenez, G. Muller, M. Rocamora, Chem. Eur. J. 2002, 8, 4164
and refs cited therein.
[23] a) A. M. Kluwer, C. J. Elsevier, M. Bühl, M. Lutz, A. L. Spek,
Angew. Chem. 2003, 115, 3625; Angew. Chem. Int. Ed. 2003, 42,
3501; b) B. Crociani, F. Di Bianca, P. Uguagliati, L. Canovese, A.
Berton, J. Chem. Soc. Dalton Trans. 1991, 71; c) K. J. Cavell, D. J.
Stufkens, K. Vrieze, Inorg. Chim. Acta 1981, 47, 67; d) M. W. van
Laren, C. J. Elsevier, Angew. Chem. 1999, 111, 3926; Angew. Chem.
Int. Ed. 1999, 38, 3715; e) R. A. Klein, P. Witte, R. Van Belzen, J.
Fraanje, K. Goubitz, M. Numan, H. Schenk, J. M. Ernsting, C. J.
Elsevier, Eur. J. Inorg. Chem. 1998, 319; f) B. Milani, A. Anzilutti,
[49] G. M. Sheldrick, SHELXL-97, University of Gçttingen, 1997.
[50] OpenMoleN, Interactive Intelligent Structure Solution, Nonius
B. V., Delft, 1997.
[51] SHELXTL, Bruker AXS GmbH, Karlsruhe, 1997.
Received: February 23, 2007
Published online: May 24, 2007
6008
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 5994 – 6008