Asymmetric allylic alkylation catalyzed by Pd(II)-complexes with (S)-BINPO, a hemilabile axially chiral P,O-heterodonor inducer

Article abstract of DOI:10.1016/j.tetasy.2004.03.028

A complex generated in situ from [Pd(η³-C₃H ₅)Cl]₂ and (S)-BINPO 1 is an active catalyst in the asymmetric allylic alkylation of 1,3-diphenyl-2-propenyl esters 6 via a malonate anion with ee's up to 81% being obtained. Rate and stereoselectivity of the reaction are dramatically influenced by the solvent. The Pd-complex [(S)-(BINPO)Pd(η³-1,3-diphenylallyl)] 5has been synthesized and its structure in the solid state determined by X-ray diffraction. The stoichiometric reaction of 5 with malonate anion affords the alkylated product of the same configuration as the one obtained in the catalytic reaction at room temperature in nearly identical enantiomeric purity (71%).

Full text of DOI:10.1016/j.tetasy.2004.03.028

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 1477–1485
Asymmetric allylic alkylation catalyzed by Pd(II)-complexes with
(S)-BINPO, a hemilabile axially chiral P,O-heterodonor inducer
a
b
b
Serafino Gladiali,a, Rossana Taras, Rosa M. Ceder, Merce Rocamora,
*
ꢀ
Guillermo Muller,b, Xavier Solans and Merce Font-Bardia
c
c
*
ꢀ
aDipartimento di Chimica, Universitꢀa di Sassari, Via Vienna 2, 07100 Sassari, Italy
bDepartament de Quimica Inorganica, Universitat de Barcelona, Diagonal 647, E-08028 Barcelona, Spain
cDepartament de Cristal.lografia, Mineralogia i Dipꢀosits Minerals, Universitat de Barcelona, Martꢁı i Franquꢀes s/n.,
08028 Barcelona, Spain
Received 18 February 2004; accepted 18 March 2004
Abstract—A complex generated in situ from [Pd(g3-C3H5)Cl]2 and (S)-BINPO 1 is an active catalyst in the asymmetric allylic
alkylation of 1,3-diphenyl-2-propenyl esters 6 via a malonate anion with ee’s up to 81% being obtained. Rate and stereoselectivity of
the reaction are dramatically influenced by the solvent. The Pd-complex [(S)-(BINPO)Pd(g3-1,3-diphenylallyl)] 5 has been syn-
thesized and its structure in the solid state determined by X-ray diffraction. The stoichiometric reaction of 5 with malonate anion
affords the alkylated product of the same configuration as the one obtained in the catalytic reaction at room temperature in nearly
identical enantiomeric purity (71%).
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
The P,O-association of donors is one of the most com-
monly exploited in the search for improved perfor-
mances of a complex in a catalytic reaction where a
dissociative step is critically involved at some stage of
the process. This concept is illustrated at its best by the
success encountered by the catalytic system utilized in
the SHOP process more than 25 years ago.2 Several
additional examples of significant value of this strategy
can be found in the recent literature.3
Bidentate P,X-heterodonor ligands where the chelate
coordination of a phosphino group (P) is supported by
a different heteroatom (X) display a peculiar catalytic
behaviour due to the presence of two donors of diverse
ligating properties. Within this class of derivatives,
mixed ligands with a phosphorus and an oxygen as
substitutionally inert and labile donors, respectively,
usually display a pronounced hemilabile character.1
The advantage of the use of hemilabile ligands is not
immediately apparent in the case of metal catalyzed
asymmetric reactions. In these processes, strong donors,
which tightly bind the metal seem better suited for
steering the complex towards the rigid conformational
arrays, which are considered appropriate for obtaining
high enantioselectivities. This is probably one of the
main reasons why the use of hemilabile ligands as chiral
auxiliaries has been overlooked for so long.
Easily dissociable donors may readily provide vacant
coordination sites around the metal. This fact may be
useful whenever dissociation of an ancillary ligand to
give a coordinatively unsaturated species is required for
the complex to be catalytically active. Furthermore, the
presence of a coordinative site in the proximity of the
metal may be helpful in stabilizing, and possibly per-
mitting the interception of otherwise labile short-lived
intermediates involved in the catalytic cycle.
Recently, however, an increased number of heterodonor
ligands have been designed for the use in asymmetric
catalysis with hemilability being more frequently
encountered in this field.4 Among chiral P,O-hemilabile
derivatives, the binaphthalene-scaffolded axially chiral
phosphinyl phosphane 1 (BINPO,5 BINAP monophos-
phine oxide) has attracted the attention of several
* Corresponding authors. Tel.: +34-934-021273; fax: +34-934-907725
(G.M.); fax: +39-079-229559 (S.G.); e-mail addresses: gladiali@
uniss.it; guillermo.muller@qi.ub.es
0957-4166/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.03.028

1478
S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
different research groups, which in recent years have
repeatedly reported both on synthetic and applicative
aspects of its chemistry.
catalyzed allylic alkylation of 1,3-diphenyl-2-propenyl
esters by a malonate anion.
BINPO provides moderate to good stereoselectivities in
a variety of metal catalyzed asymmetric processes. In a
range of addition reactions to styrene, for instance, the
ee’s fluctuated in between 70% (Pd-catalyzed hydro-
silylation)6 and 30% (hydroformylation with Pt–Sn cat-
alysts)7 and 40% (Rh-catalyzed hydroboration).8 Higher
stereoselectivities, up to 99%, have been more recently
reported in the asymmetric Diels–Alder reaction of
cyclopentadiene with methacrolein catalyzed by diaste-
reomerically pure (g6-cymene)Ru(BINPO) complexes.9
BINPO itself, in the absence of any transition metal, can
participate in promoting the nonenzymatic kinetic res-
olution of secondary alcohols.10
2. Results and discussion
2.1. Synthesis of BINPO
Although the first report on the isolation of enantiopure
BINPO dates back to 1992, when the (R)-isomer was
obtained as a co-product in the preparation of
Pd(0)[(R)-BINAP]2 from palladium(II) diacetate and
(R)-BINAP,16 the first preparation of any practical
value of this ligand was reported by some of us in 1998.6
Soon after, Grushin developed a simple and efficient
synthesis of phosphine–phosphine oxides which relies
on the Pd-catalyzed selective mono-oxidation of the
homotopic P atoms of C2-symmetrical bis-phosphines.
Application of this procedure to enantiopure BINAP
allows BINPO of the same configuration to be isolated
in some 80% yield.17
Over the course of these studies, several transition metal
complexes incorporating BINPO have been prepared
and structurally characterized. As a general trend, it has
been observed that BINPO displays a sharp propensity
towards bidentate binding to transition metal centres.
Some derivatives of Pd(II)6;11 and Rh(I)12 where BINPO
displays a ‘normal’ P,O-chelate coordination have been
isolated and characterized by NMR and X-ray diffrac-
tion. The P,O-chelate coordination of BINPO to (g6-
cymene)Ru-complexes is remarkable in that it proceeds
with complete stereoselectivity affording one single dia-
stereomeric complex.13
In our original methodology,6 racemic BINPO was
synthesized from 2,20-dihydroxy-1,10-binaphthalene 2
(R ¼ H; BINOL) through a four step reaction procedure
involving the Pd-catalyzed sequential substitution of the
homotopic triflate groups of the BINOL ditriflate 2
(R ¼ SO2CF3) by diphenylphosphine oxide (Scheme
1).18 Resolution of the racemate was accomplished by
complexation to an enantiopure chloride-bridged C,N-
cyclopalladated complex.
The inherent electronic disparity of the donors of the
chelating arms of BINPO is mainly responsible for the
selectivity shown by some BINPO complexes in
the oxidative addition of aryl iodide (Pd)11 and in the
insertion of tin chloride into the metal–chloride bond
(Pt).14 The formation of one single isomer in reactions
such as these may have important bearings in addressing
the selectivity of the reactions where these complexes are
involved as catalysts.
(DPE)NiCl2,Ph2P(O)H
OR
OR DABCO,DMF,100˚C
P(O)Ph2
PPh2
2
1
R=H;CF3SO2
Ph2P(O)H
Pd(OAc)2/L2
In addition to the P,O-chelate binding mode, BINPO
can display a different hapticity as has been noticed in
the coordination to Pd(0)11 and to some L2Ru(II)-
complexes (L2 ¼ bipy or phen).15 Therein the ligand acts
as a six-electron donor, adding to the P and O atoms, a
g2-arene coordination of the naphthyl group proximate
to the oxidized P atom.
Ph2P(O)H
Pd(OAc)2/L2
OSO2CF3 HSiCl3,Et3N
P(O)Ph2 130˚C,xylene
OSO2CF3
PPh2
Direct experimental evidence of the hemilabile behav-
iour of BINPO has only been collected in the case of
Pt(II)7 and Pd(II)11 derivatives where 31P NMR moni-
toring clearly shows that opening of the chelate ring
involving displacement of the oxygenated arm of BIN-
PO readily takes place in the presence of competitive
donors. The role, if any, of species such as these in the
catalytic reactions supported by these complexes, that is,
hydroformylation7 and hydrosilylation,6 is still to be
cleared.
3
Scheme 1.
The same route can also be adopted for the preparation
of enantiopure BINPO from enantiopure BINOL as we
routinely did in the following years when enantiopure
BINOL became more and more accessible.
For the preparation of BINPO congeners, we have now
devised a novel and more expedient route relying on a
Ni-catalyzed cascade reaction of the ditriflate 2 with a
diarylphosphine oxide, which allows the homotopic
triflate groups of the substrate to be sequentially
The results reported above stimulated us to expand the
scope of this hemilabile ligand in asymmetric catalysis.
Herein we report on the behaviour of BINPO in the Pd-

S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
1479
substituted in a single process affording the corre-
sponding mono-phosphine oxide in a fair to good iso-
lated yield.19
assigned to syn-hydrogen atom located in the carbon
trans to the coordinated phosphorus atom of the (S)-
1
BINPO ligand. The proposal was based on the H–13C
heterocorrelation and on reported values for similar
complexes.20 Although only two broad signals at 3.30
and 2.81 ppm were observed, the peaks in between 3.70–
2.40 ppm integrate three hydrogens corresponding to
one anti proton trans to the P atom and to both the
terminal protons in cis position respect to the P. This
provides an indication of the dynamic behaviour of
neutral compound 4 in solution. Accordingly, 2D
NOESY experiment shows exchange signals between all
terminal protons. This fact suggests that the two well
documented dynamic processes of Pd-allyl complexes
are operative:21;22 the first is the apparent rotation
around Pd-allylic bond exchanging anti–anti and syn–
syn protons on different terminal carbon atom; the sec-
ond being the p–r–p mechanism exchanging syn–anti
protons on the carbon atom that retains the r bond with
the palladium centre.
Although some details of this reaction are still to be
cleared, 31P NMR monitoring shows that the reaction
proceeds through the intermediate formation of the
phosphino triflate 3, which undergoes the fast substitu-
tion of the second triflate by diphenylphosphine oxide.
Small amounts of BINAP, arising from the reaction of 3
with diphenylphosphine are simultaneously formed.
Evidence has been collected that suggests 3 originates
from the reaction of the ditriflate 2 with diphenylphos-
phine and that this latter reagent is formed in the early
stages of the reaction by disproportionation of diphen-
ylphosphine oxide.
2.2. Synthesis of [Pd(g3-C3H5)Cl(BINPO-j1P,O)] com-
plex 4
The synthesis of the [Pd(g3-C3H5)Cl(BINPO-j1P,O)]
complex was carried out by reacting the appropriate
amount of [Pd((g3-C3H5)Cl]2 dimer with 2 equiv of the
enantiopure, (S)-BINPO 1 ligand in CH2Cl2 at room
temperature under a nitrogen atmosphere (Scheme 2). It
is apparent from the 31P NMR spectra of the isolated
neutral complex 4 that the ligand is bound to the metal
in a monodentate fashion. For free (S)-BINPO, 31P
NMR signals appear at d ¼ ꢀ16:3 ppm (PPh2) and
26.1 ppm (OPPh2), while for complex 4 two different
signals appear at 26.8 ppm (OPPh2) and 25.0 ppm
(PPh2). The 41.3 ppm downfield shift observed for the
PPh2 phosphorous atom and the nearly unchanged
position of the OPPh2 resonance provide a strong indi-
cation that only the PPh2 group is coordinated to the
metal as observed for other complexes of this ligand.7
13C NMR spectrum at 298 K shows three signals in the
allylic region that can be unambiguously assigned to the
three carbon atoms of the C3H5 ligand, central C2 at
117.9 ppm, terminal carbon C1 that appears as a doublet
at 78.6 ppm (JPC ¼ 31 Hz) and a broad singlet at
63.9 ppm assigned to C3. 13C–1H heterocorrelation al-
lowed us to assign the H anti trans to the phosphorus
atom to the broad signal at 3.3 ppm.
2.3. Variable temperature experiments
1H NMR variable temperature experiments have been
recorded within the range 223–323 K. A selection of the
1H NMR spectra in the allylic region is shown in Figure
1 showing the broadening of the signals on lowering the
temperature. At 238 K, the existence of two different
isomers in a ratio 1/1 was evidenced. There are two
isomers of complex 4 when considering the Pd(allyl)ClP
fragment as shown in Scheme 3. Extending Cahn, In-
gold, and Prelog rules to define the configuration in
these systems and considering the metal as the stereo-
genic centre, we can design (R)-4 or (S)-4 owing to the
clockwise or anticlockwise assignment of priorities as
described by Faller et al.22 The H central appears as two
1H NMR spectrum in CDCl3 at 298 K showed several
overlapping signals between 7.5 and 6.2 ppm corre-
sponding to the 32 aromatic protons of (S)-BINPO
ligand. Only
a well defined triplet at 6.48 ppm
(J ¼ 7:5 Hz) and a doublet at 6.24 ppm (J ¼ 8:5 Hz)
were observed in the aromatic region. At higher fields
the allylic protons showed a pseudoquintuplet at
5.28 ppm (J ¼ 10 Hz) assigned to the central allylic
hydrogen and a broad triplet at 4.42 ppm (J ¼ 7 Hz)
H2central
H1syn
H3syn
CH2Cl2, 25˚ C, 0.5 h
R = H
H1anti
H3anti
Cl
Pd
P
P=O
R
R
4
P(O)Ph2
PPh2
Pd
Cl
Cl
Pd
+
H2central
Ph
+
Ph
R
R
H1anti
Pd
H3anti
-
BF4
R = H, Ph
i) acetone, 25˚ C, 1 h; ii) TlBF4, 5 h
R = Ph
P
P=O
1
P
O
P
5
Scheme 2.

1480
S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
2.4. Reaction of 4 with TlBF4
Reaction of neutral complex 4 with a stoichiometric
amount of TlBF4 for 30 min at room temperature in
CDCl3 led to a suspension that decomposed after 1 h.
1
Nevertheless the reaction was followed by H and 31P
NMR spectroscopy. 31P NMR spectrum shows one
sharp signal at 26 ppm (PPh2) and a broad one centered
at 31.8 ppm (OPPh2). The 5 ppm shift downfield of the
OPPh2 phosphorous atom is indicative of the presumed
formation of a cationic complex through chelate coor-
dination of BINPO via P and O donor atoms; however
this species is not stable in these conditions. The
appearance of the broad signal at 31.8 ppm suggests that
the oxygen atom may be involved in a continuous
coordination–decoordination process. The 1H NMR
spectrum shows in the allylic region a pseudoquintuplet
at 5.35 ppm (Hcentral), a broad signal at 4.53 ppm
(Hsyn trans P), a broad triplet at 3.44 ppm (Hanti trans
P) and a very broad signal between 3.4–2.6 ppm (Hsyn
and Hanti cis P). The last very broad signals suggest a
rapid interconversion of the two protons at the cis-po-
sition with respect to the coordinated P atom, through
the p–r–p mechanism that is favoured when the Pd–C
bond trans to the P atom is open.
Figure 1. Selection of 1H NMR spectra at different temperatures of
compound 4.
H2central
H'1anti
H'3
H1syn
H3syn
anti
H'1syn
H'3syn
H1anti
H3anti
H'2central
P P=O
Cl Pd
Cl Pd
P
P=O
(R)-4
(S)-4
Scheme 3.
2.5. Synthesis of [Pd(g3-Ph2C3H5)(BINPO-j2P,O)]BF4
ionic complex 5
broad signals at 5.4 and 5.2 ppm and Hsyn at the trans
position to the phosphorous atom as a single signal at
4.4 ppm. Moreover, two broad signals at 3.4 and
3.1 ppm are clearly defined and can be assigned to the
The synthesis of 5 was carried out via reaction of 1,3-
diphenylallylpalladium chloride dimer with stoichiome-
tric amounts of (S)-BINPO and TlBF4 in acetone at
room temperature to produce [Pd(g3-1,3-diphenyl-
allyl)((S)-BINPO-j2P,O)]BF4 (Scheme 2). The pre-
parative reaction was monitored by 31P. No cleavage of
the chloro bridge took place in the absence of TlBF4.
1
H1anti and H10 anti based on the heterocorrelation H–
13C at 238 K. Terminal protons at the cis position to the
P atom appear at 3.5 (1H), 2.8 (2H) and 2.5 (1H) ppm.
2D COSY experiment shows that signals at 5.4, 4.4 and
3.1 ppm belong to one isomer, while signals at 5.2, 4.4
and 3.4 to the other one. 2D ROESY experiment at
238 K only shows exchange signals between H2central–
H20central and H1anti–H10anti suggesting that the pre-
dominant exchange mechanism between both isomers at
238 K may be a p–r–p process, proceeding by opening
the Pd–C allylic bond trans to the P atom. At this
temperature, Pd–allylic rotation probably occurs at
slower rates since exchange between H1anti and H30anti
is not observed.
Due to the presence of two isomers in solution, the 31P
NMR spectrum in CDCl3 of complex 5 showed four
signals, at 40.7 and 25.9 ppm for the major isomer and at
42.5 and 27.2 ppm for the minor one. It is readily
apparent from these chemical shifts that the diaryl-
phosphinyl and diarylphosphine groups are both coor-
dinated, with the signals at lower fields being assigned to
the OPPh2 and those at higher fields to the PPh2 groups,
respectively. 31P NMR in CD3COCD3 at room tem-
perature only shows one slightly broad signal at
20.0 ppm and a very broad signal at 40.0 ppm. This fact
suggests that the coordinating solvent exchanges rapidly
with the OPPh2 group in the coordination sphere,
leading to a broadening of the phosphorus resonance as
previously observed with Pt(II) complexes with BINPO
ligand.7
31P NMR was run at 238 K in CDCl3 solution. Three
different signals appear, one at 26.94 ppm assigned to
P(O)PPh2 not coordinated, that has the same chemical
shift as observed at 298 K and two signals of the same
intensity but lower than the first one at 25.73 and
25.4 ppm assigned to the P atom coordinated to the
metal (PPh2) of each one of the isomers.
1H NMR spectra of 5 in CDCl3 solution were also
consistent with the existence of two isomers in a 85/15
ratio (Scheme 4). The allylic protons of the major isomer
1H NMR spectra at 323 K was also run. The central
proton appears as a well defined quintuplet suggesting
equivalency of the four terminal allylic hydrogen atoms
but the signals of the terminal allylic protons appear as
broad signals between 4.8–2.5 ppm indicating rapid
interconversion between them, probably through both
p–r–p and Pd–allyl rotation mechanisms.
appeared as a doublet at 4.58 ppm (J
two pseudotriplets at 5.60 ppm (J
¼ 10:5 Hz) and
ðHHÞ
ðHHÞ
¼ 10:0 Hz) and
6.03 ppm (J
¼11.5 Hz). At 273 K, the pseudotriplet
ðHHÞ
at 5.60 was separated into a doublet of doublets,
J
¼ 8:7 and J ¼12.9 Hz, suggesting a location
ðPHÞ
ðHHÞ

S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
1481
Ph
O
Ph
P
P
O
Pd
Pd
Ph
Ph
(S)-5
(R)-5
15%
85%
Scheme 4.
trans to the P atom. Taking into account the values
of the coupling constants between vicinal protons and
reported data,23 we assigned the doublet at 4.58 ppm to
the terminal allylic H atom cis to the phosphorus, the
triplet at 6.03 ppm to the central allylic H atom and the
geometry of the 1,3-diphenylallyl system assumed to be
syn/syn.
The allylic protons of the minor isomer appeared as
broad signals centered at 4.14, 5.87 and 6.30 ppm.
According to their chemical shifts (similar to those of
the major isomer) we proposed a syn/syn structure for
the allyl group and assigned them to the H atom cis and
trans to the phosphorus, and central, respectively. The
analysis of the signals thus indicate that the two syn/syn
isomers differ in the orientation of the 1,3-diphenylallyl
group respect the Pd-(S)-BINPO fragment. As shown in
Scheme 4 and following Faller et al. nomenclature22 we
can design (R)-5 and (S)-5 each one of the isomers.
The aromatic region showed a complex pattern, with
two groups of multiplets between 6.43–7.60 and 8.08–
8.13 ppm and two well separated doublets at 5.78 and
8.60 ppm for the major isomer. COSY experiments
allow us to identify five contiguous aromatic protons of
the same phenyl ring (5.78, 6.51, 7.40, 8.10, 8.60 ppm).
Figure 2. Representation of the cationic fragment of 5 with the num-
bering scheme emphasizing the geometry of the coordination sphere.
3
ꢂ
Table 1. Selected bond lengths (A) and bond angles (°) for 5 [Pd(g -
Ph2C3H3)((S)-BINPO–P,O)]BF4ÆCH2Cl2
NOESY spectra at 500 MHz, in CDCl3 showed NOE
contacts between terminal allylic H trans P, central
allylic H and aromatic protons at 7.6 ppm (presumably
two ortho hydrogens of the phenyl allylic group trans to
P atom). Besides strong NOE contacts between the
two anti protons of the allyl group of the major isomer,
exchange signals between allylic protons of both isomers
are observed. The dynamic process was consistent with a
pseudorotation through the Pd–allylic bond. 13C NMR
spectrum shows chemical shifts similar to those reported
for P,N-ligands24 at 109.4 (central, s) ppm, 103.8 (trans
P, d, JPC ¼ 14 Hz) and 67.3 (cis P, s) ppm for the major
isomer.
Bond lengths
Bond angles
Pd–O
2.101(6)
2.3214(19)
2.113(9)
2.139(8)
2.253(7)
1.447(12)
1.364(12)
P(1)–Pd–O
98.46(16)
67.1(3)
Pd–P(1)
Pd–C(1)
Pd–C(2)
Pd–C(3)
C(1)–C(2)
C(2)–C(3)
C(1)–Pd–C(3)
C(1)–Pd–P(1)
C(3)–Pd–O
100.0(2)
94.4(3)
that the diphenylallyl group adopts a syn/syn configu-
ration and that the phenyl groups of the allyl ligand and
the naphthyl groups of the (S)-BINPO ligand are lo-
cated at the opposite side of the coordination plane as
described in Scheme 2. The formation enthalpy of both
diasteromers were studied by means of theoretical cal-
culations at PM3 (tm) level [255.138 Kcal/mol for the
(R)-5 isomer and 257.355 Kcal/mol for the (S)-5 one]
showing that the (R)-5 disposition was more stable. The
bond lengths are within the reported values,6 but the
Pd–P bond is longer than reported for similar complexes
with P,N-ligands.24a Furthermore, the C(1)–Pd bond
(2.113(9) A) was significantly shorter than the C(3)–Pd
ꢂ
bond [2.253(7) A] according to the stronger trans influ-
ence of the phosphorus atom compared with oxygen
atom. The terminal allylic carbons C(1) and C(3)
ꢂ
were 0.069(7) and 0.157(7) A below the P(1)–Pd–O
Crystals suitable for X-ray determinations were ob-
tained after crystallization from CH2Cl2–hexane. A view
of the allylic cation and the most representative
parameters of the structure are reported in Figure 2 and
Table 1.
The crystal contained discrete units of 5 in which (S)-
BINPO acts as a bidentate ligand providing an eight-
membered chelate ring with an irregular, partly boat-
like conformation. The palladium coordination sphere
was closely square planar (the distances of Pd, C(3),
C(1), O and P from the mean plane defined by these
ꢂ
ꢂ
atoms are small, less than 0.040 A). The structure shows

1482
S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
plane. The allyl group was bound rather asymmetrically
to the Pd-atom. The allylic C–C bond trans to P was
2.6. Catalytic allylic alkylation of 1,3-diphenyl-2-propen-
yl esters
ꢂ
closer to the double bond 1.364(12) A than the C–C
ꢂ
bond trans to O 1.447(12) A as reported for nonsubsti-
The Pd-catalyzed asymmetric allylic substitution of
1,3-diphenyl-2-propenyl esters 6 with the dimethyl
malonate–BSA–potassium acetate system was success-
fully carried out in the presence of the catalyst generated
in situ from [Pd(g3-C3H5)Cl]2 and (S)-BINPO (Scheme
5).
tuted allylic complexes.25 The chelate bond angle O–Pd–
P(1) was 98.46(16)° rather than 90.9(1)° as reported
in cyclometallated Pd(II) complexes with a BINPO
ligand.24b A similar trend was detected when comparing
the chelate angles of a cyclometallated Pd(II) complex
and a diphenylallyl Pd(II) complex with the same P,N-
ligand (1-methyl-2-diphenylphosphino-3-(10-isoquinolyl)-
indole).26
Rate and stereoselectivity of the reaction are strongly
dependent on the solvent (Tables 2 and 3) and for a
complete conversion to be achieved quite different
reaction times were required. Chlorinated solvents per-
formed far better than hydrocarbon or oxygenated
counterparts, chloroform being the most effective of the
list. The solvent effect was as pronounced as to be the
cause of a switch of handness from the (S)- to the (R)-
configured product in moving from the chlorinated
solvents to toluene (Table 2, entry 3).
The naphthyl rings of the binaphthyl framework showed
a torsion angle, C(28), C(37), C(38), C(47) of 73.916°.
A significant-stacking interaction involves one of the
phenyl substituents on each P atom and the phenyl ring,
which is more remote from the metal of each naph-
thyl group of the binaphthalene template. The dihedral
angles between the mean planes of the stacked aromatic
rings are 13.6(4)° and 11.9(5)° for the interaction
involving the naphthyl rings C(31)–C(36) and C(39)–
C(44) and the distances between the centre of the aro-
The stereoselectivities were almost identical for both the
acetate (R ¼ Me) and the carbonate (R ¼ OMe) esters
and improve slightly when decreasing the temperature in
the range 298–223 K. The best ee (81%) was obtained in
chloroform at 0 °C. At lower temperatures (ꢀ40 °C), the
reaction was sluggish while after 12 h, the catalyst
apparently underwent an irreversible deactivation.
Notably, in this reaction BINAP was by far less stereo-
selective (25–35% ee depending on the conditions)
and led to the alkylated product of opposite configura-
tion.28
ꢂ
matic rings being of 3.649 and 3.58 A, respectively.
Moreover the phenyl allylic group on C(1) and one of
the phenyl’s of the PPh2 substituent adopt a nearly
parallel stacked configuration. The dihedral angles
between the mean planes of the aromatic rings C(4)–
C(9) and C(16)–C(21) was 2.7(6)°. The distance between
ꢂ
the center of both rings was 3.876 A. This third inter-
action probably contributed to the adoption of an ideal
square planar geometry as reported.27
O
O
R
CH(COOCH3)2
(S)-BINPO, [Pd(η3-C3H5)Cl]2
CH2(COOCH3)2, BSA, AcOK
6
7
R = CH3; OCH3
Scheme 5.
Table 2. Allylic alkylation of 1,3-diphenylallyl acetate by in situ Pd/(S)-BINPO catalyst
O
[Pd(η3-C3H5)Cl]2; ligand; solvent
CH(COOCH3)2
Ph
O
CH3
CH2(COOCH3)2, BSA, AcOK
Ph
Ph
Ph
Solvent
T °C
t
Ee%
Configuration
THF
DMF
25
25
15 m
7 h
10
0
S
––
R
S
Toluene
ClCH2CH2Cl
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
25
25
1.5 h
40 m
20 m
1 h
6
48
47
50
51
51
72
81
32
35
25
0
S
S
ꢀ40
ꢀ50
25
2 h
S
2.5 h
1.5 h
4
S
S
CHCl3
CH2Cl2
THFa
0
25
S
a
3.2 h
70 h
R
R
25
a Chiral ligand (S)-BINAP.

S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
1483
Table 3. Allylic alkylation of 1,3-diphenylallyl carbonate by in situ Pd/(S)-BINPO catalyst
O
3-C3H5)Cl]2; ligand; solvent
CH2(COOCH3)2, BSA
CH(COOCH3)2
Ph
[Pd(η
O
OCH3
Ph
Ph
Ph
Solvent
T °C
t
Ee%
Configuration
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
THF
25
0
20 m
1.5 h
2 h
47
48
S
S
S
S
S
R
ꢀ20
ꢀ40
25
56
60
2.5 h
15 m
1.5 h
32
25.5
a
CH2Cl2
25
a Chiral ligand (S)-BINAP.
It is well known that the enantioselectivity in Pd-cata-
lyzed allylic alkylation with soft reagents is controlled by
nucleophilic attack at the more electrophilic terminal
allylic carbon atom in Pd(II) intermediate 5.29 As de-
scribed above, 5 is a 85:15 mixture of two isomers (R)-5
and (S)-5, the former being the major one. The stereo-
chemical outcome and the observed enantiomeric excess
[72% in (S)] can be rationalized if we assume that the
nucleophilic attack proceeds at similar rates onto the
allylic carbon of the two isomers trans to the P-donor.
This is the more electrophilic carbon as inferred by the
longer bond length C3–P and higher 13C chemical shift.
CHIRACEL OD column using hexane–i-PrOH, 99:1, as
eluent, at a flow rate of 0.3 cm3 minꢀ1 and pressure
10 bar. Elemental analyses were carried out by the Ser-
ꢀ
veis Cientifico-Tecnics de la Universitat de Barcelona in
an Eager 1108 microanalyzer.
3.2. Synthesis of (S)-BINPO 1
A solution of diphenylphosphinoxide (1.84 g, 9.1 mmol)
in DMF was added to a solution of NiCl2 (dppe) (0.1 g,
0.182 mmol) in DMF (5 mL) under nitrogen. The mix-
ture was stirred at 100 °C for 30 min, then a solution of
(S)-BINOL ditriflate (1 g, 1.82 mmol) and DABCO
(0,82 g, 7.32 mmol) in DMF (10 mL) were added. The
mixture was stirred at 100 °C for 12 h. After cooling to
room temperature, degassed water was added and the
resulting yellow solid filtered off. After washing several
times with water, the precipitate was dissolved in
CH2Cl2, dried and evaporated. The residue was purified
by flash chromatography on silica gel using petroleum
ether–ethyl acetate (3:2) as eluent to give BINPO (45%)
2.7. Stoichiometric allylic alkylation
In order to confirm the source of enantioselectivity, we
carried out the stoichiometric allylic alkylation of the
preformed complex 5. The reaction was run in CHCl3 at
room temperature with stoichiometric amounts of the
nucleophile under basic conditions as used for the cat-
alytic process. The change in colour was immediate,
indicating a fast reaction. The enantiomeric excess of the
alkylation product (71%) was close to the value ob-
served in the catalytic reaction in the same solvent. This
result provides further support to the reaction path
formulated above.
1
and BINAP (20%). BINPO: mp 230–233 °C; H NMR
(CDCl3): d(ppm) ¼ 6:63 (d, J ¼ 8:1 Hz, Ar, 1H), 6.72
(dt, J ¼ 1:2, 6.6 Hz, Ar, 1H), 6.81–7.43 (series of m, Ar,
23H), 7.60 (m, Ar, 3H), 7.69 (d, J ¼ 8:1 Hz, Ar, 1H),
7.74 (d, J ¼ 8:4 Hz, Ar, 1H), 7.82 (d, J ¼ 7:8 Hz, Ar,
1H), 7.93 (dd, J ¼ 2:4, 8.4 Hz, Ar, 1H); 31P NMR
(CDCl3): d(ppm) ¼ ꢀ14:25 (s), 28.51 (s); [a]2D0 ¼ ꢀ92:2
1
(c 1.02, CHCl3). BINAP: H NMR (CDCl3): d(ppm)
3. Experimental
¼ 7:0–8.0 (series of m, Ar, 23H), 31P NMR (CDCl3):
d(ppm) ¼ ꢀ13:04.
3.1. General methods
All compounds were prepared under a purified nitrogen
atmosphere using standard Schlenk and vacuum-line
3.3. Synthesis of [Pd(g3-C3H5)Cl(BINPO-j1P,O)] 4
techniques.
[Pd(g3-C3H5)(l-Cl)2]30
and
[Pd(g3-
A mixture of [Pd(g3-C3H5)(l-Cl)]2 (28 mg, 0.078 mmol)
and (S)-BINPO (100 mg, 0.156 mmol) was stirred in di-
chloromethane (10 cm3) under nitrogen for 1 h at room
temperature. The solvent was removed under reduced
pressure, affording a pale yellow solid. Recrystallization
from dichloromethane/diethylether (1/1). Yield 77 mg
(60%). Anal. Found: C, 67.99; H, 4.59. Calcd for
PdClC47H37OP2C, 68.71 ; H, 4.54. MS (FAB positive).
m=z 785 ([MꢀCl]þ). NMR data (CDCl3 298 K): 1H
(500 MHz): 8.04–6.09 ppm (32H aromatic); 5.28 ppm
(pq, J ¼ 10 Hz, 1H); 4.42 ppm (pt, J ¼ 7 Hz, 1H) ; 3.70–
2.40 ppm (3H); 31P NMR (101 MHz): 26.8 ppm (s);
31
Ph2C3H3)(l-Cl)]2 were prepared as described previ-
ously. NMR spectra were recorded on a Bruker DRX
500 (1H, standard SiMe4, 2D NMR experiments), Bru-
ker Unity 300 (13C, standard SiMe4) and Bruker DRX
250 (1H, 13C standard SiMe4, 31P standard H3PO4)
spectrometers in CDCl3 unless otherwise cited. Chemi-
cal shifts were reported downfield from standards. IR
spectra were recorded on a FTIR Nicolet Impact 400
spectrometers. FAB mass chromatograms were
obtained on a Fisons V6-Quattro instrument. Enantio-
meric excesses were determined by HPLC on a

1484
S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
25.0 ppm (s); 13C NMR (75 MHz): allylic carbons:
117.9 ppm (s), 78.6 ppm (d, J
¼ 31 Hz); 63.9 ppm (s).
given results.35 All H atoms were computed and refined,
using a riding model, with an isotropic temperature
factor equal to 1.2 time the equivalent temperature
ðPCÞ
factor of the atom which are linked. The final R (on F )
2
3.4. Synthesis of [Pd(g3-Ph2C3H3)(BINPO-j2O,P)]BF4 5
mixture of [Pd(g3-Ph2C3H3)(l-Cl)]2 (67 mg,
factor was 0.054, wR (on jF j ) ¼ 0.141 and goodness of
fit ¼ 0.950 for all observed reflections. The number of
refined parameters was 615. Max. shift/esd ¼ 0.0, Mean
shift/esd ¼ 0.0. Max. and min. peaks in final difference
A
0.1 mmol) and (S)-BINPO (130 mg, 0.2 mmol) was stir-
red in acetone (25 cm3) under nitrogen for 1 h. The
resulting suspension was treated with TlBF4 (58 mg,
0.2 mmol). After 5 h, the mixture was filtered through a
pad of Celite and the solution concentrated in vacuo to
leave a yellow solid. Recrystallization from dichloro-
methane–diethylether (1:5), gave 125 mg (61% yield) of a
synthesis were 0.899 and ꢀ0.538 eAꢀ3, respectively
ꢂ
(CCDC 231370).
3.6. General procedure for palladium-catalyzed allylic
alkylation
1
mixture of two isomers (85:15) followed by H NMR in
CDCl3. Crystals suitable for X-ray analysis were ob-
tained from CH2Cl2–hexane. Anal. Found: C, 66.57; H,
4.76. Calcd for C59H45BF4OP2PdÆCH2Cl2: C, 64.92; H,
4.27. MS (FAB positive). m=z 938 ([MꢀBF4]þ). NMR
data (CDCl3 298 K): 1H (500 MHz): major isomer:
A solution of [Pd(g3-C3H5)Cl]2 (2 mg, 0.01 mmol) and
the free ligand (13 mg; 0.02 mmol) in CH2Cl2 (1 mL) was
stirred at rt under nitrogen. After 30 min a solution of
1,3-diphenylprop-2-enyl acetate (50 mg, 0.2 mmol) in
CH2Cl2 (0.5 mL) was added. Dimethyl malonate (80 mg,
0.6 mmol), N,O-bis(trimethylsilyl)acetamide (BSA)
(122 mg, 0.6 mmol) dissolved in CH2Cl2 (1.5 mL) and
potassium acetate (0.6 mg, 0.006 mmol) were added to
the reaction mixture. Stirring continued at the required
temperature until complete conversion was attained.
The reaction mixture was diluted with ether (25 mL) and
washed with saturated NH4Cl. The organic phase was
dried over Na2SO4, concentrated under reduced pres-
sure and the residue purified by flash chromatography
(light petroleum–ether, 3:1) to give dimethyl[(S)-1,3-
diphenylprop-2-enyl]malonate. Ee’s were determined
from the integrals of the methoxy groups of (1,3-
diphenylprop-2-enyl)malonate, as split by the chiral
shift reagent europium tris[3-(trifluoromethylhydr-
oxymethylene)-(þ)-camphorate].
4.58 ppm [d, 3J
¼ 10:5, 1H], 5.60 ppm (pt,
ðHHÞ
J
6.03 ppm (pt,
¼ 10:0 Hz, 1H), 5.78 ppm [d, J3 ¼8.5 Hz 1H),
ðHHÞ
ðHHÞ
J ¼ 11:5 Hz, 1H), 6.43–7.60 ppm
(40H), 8.08–8.13 ppm (m, 2H), 8.60 ppm (d,
ðHHÞ
3J
¼ 7:5 Hz, 1H); minor isomer (allylic protons):
ðHHÞ
4.14 (br s), 5.87 (br s), 6.30 (br s), 31P NMR (101 MHz):
major isomer: 40.7 ppm (s, OPPh2), 25.9 ppm (s, PPh2),
minor isomer: 42.5 (s, OPPh2), 27.2 ppm (s, PPh2); 13C
NMR (63 MHz): allylic carbons major isomer:
109.4 ppm (s), 103.8 ppm (d, J
(s).
¼ 14 Hz); 67.3 ppm
ðP;CÞ
3.5. Crystallography
The crystallographic data are summarized in Table 1.
Yellow crystals of
[Pd(g3-Ph2C3H3)(BINPO-
j2O,P)]BF4 3 were obtained by slow diffusion of n-
pentane over dichloromethane solution of the ionic
complex. The complex crystallizes with one molecule of
dichloromethane.
3.7. Stoichiometric allylic alkylation
Ionic complex 3 (75 mg, 0.073 mmol), was dissolved in
0.7 mL of CHCl3 and stirred at room temperature under
nitrogen for 30 min. Dimethyl malonate (29 mg,
0.219 mmol), BSA [N,O-bis(trimethylsilyl)acetamide]
(45 mg, 0.219 mmol) and a catalytic amount of potas-
sium acetate were added with 2 mL of CHCl3. The
mixture was stirred at room temperature for 45 min. The
solution was then diluted with diethyl ether (8 mL) and
washed with an aqueous solution of NH4Br (10%) and
water (3 ꢁ 10 mL). The organic phase was dried over
anhydrous Na2SO4, filtered and the solvent removed
under reduced pressure. The product was purified by
column chromatography (SiO2; ethyl acetate–hexane,
1:5) to give dimethyl [(S)-1,3-diphenylprop-2-enyl]mal-
onate in 71% ee.
A prismatic crystal (0.1 ꢁ 0.1 ꢁ 0.2 mm) was selected and
mounted on a MAR345 diffractometer with a image
plate detector. Unit-cell parameters were determined
from automatic centering of 14381 reflections
(3< h <31) and refined by least-squares method. Inten-
sities were collected with graphite monochromatized
ꢂ
Mo-Ka radiation(k ¼ 0.71069 A). 22086 reflections were
measured in the range 2.43–25.00°. 4606 of which were
nonequivalent by symmetry (Rint (on I) ¼ 0.088). 3568
reflections were assumed as observed applying the con-
dition I > 2r(I). Lorentz-polarization but no absorption
corrections were made. The structure was solved by di-
rect methods, using the SHELXS computer program32
and refined by the full-matrix least-squares method with
SHELX 97 computer program,33 using 4606 reflections,
(very negative intensities were not assumed). The func-
Acknowledgements
2
2 2
tion
minimized
was
RwkFoj ꢀ jFcj j ,
where
We would like to thank the Spanish Ministerio de
ꢁ
Ciencia y Tecnologıa (BQU2001-3358) for its financial
support. Financial support from MIUR (PRIN-
2003033857) and from the University of Sassari is also
gratefully acknowledged by S.G.
w ¼ ½r2ðIÞ þ ð0:1217PÞ2ꢂꢀ1, and P ¼ ðjFoj þ 2jFcj Þ=3,
values of f , f 0 and f 00 were taken from Ref. 34. The
absolute configuration of the structure was defined from
the Flack coefficient, which it is equal to 0.07(5) for the
2
2

S. Gladiali et al. / Tetrahedron: Asymmetry 15 (2004) 1477–1485
1485
16. Ozawa, F.; Kubo, A.; Hayashi, T. Chem. Lett. 1992, 2177.
17. (a) Grushin, V. V. J. Am. Chem. Soc. 1999, 121, 5831; (b)
Grushin, V. V. Organometallics 2001, 20, 3950.
18. Kurz, L.; Lee, G.; Morgans, D., Jr.; Waldike, M. J.; Ward,
T. Tetrahedron Lett. 1990, 31, 1665.
19. Gladiali, S.; Pinna, S.; Taras, R., in press.
20. (a) Brown, J. M.; Hulmes, D. I.; Guiry, P. J. Tetrahedron
1994, 50, 443; (b) Brassat, I.; Englert, U.; Keim, W.;
Keitel, D. P.; Killat, S.; Suranna, G. P.; Wang, R. Inorg.
Chim. Acta 1998, 280, 150.
References and notes
1. Reviews: (a) Braunstein, P.; Naud, F. Angew. Chem., Int.
Ed. 2001, 40, 680; (b) Slone, C. S.; Weinberger, D. A.;
Mirkin, C. A. Prog. Inorg. Chem. 1999, 5, 233; (c) Bader,
A.; Lindner, E. Coord. Chem. Rev. 1991, 108, 27.
€
2. Keim, W.; Kowalt, F. H.; Goddard, R.; Kruger, C.
Angew. Chem., Int. Ed. Engl. 1978, 17, 466.
3. (a) Malinoski, J. M.; Brookhart, M. Organometallics 2003,
22, 5324, and references cited therein; (b) RajanBabu, T.
V.; Nomura, N.; Jin, J.; Nandi, M.; Park, H.; Sun, X. J.
Org. Chem. 2003, 68, 8431; (c) Mecking, S.; Keim, W.
Organometallics 1996, 15, 2650.
21. Sprinz, J.; Kiefer, M.; Helmchen, G.; Reggelin, M.;
Huttner, G.; Walter, A.; Zsolnai, L. Tetrahedron Lett.
1994, 35, 1523.
22. Faller, J. W.; Stokes-Huby, H. L.; Albrizzio, M. A. Helv.
Chim. Acta 2001, 84, 3031.
23. Baltzer, N.; Macko, L.; Schaffner, S.; Zehnder, M. Helv.
Chim. Acta 1996, 79, 803.
4. (a) Faller, J. W.; Patel, B. P.; Albrizzio, M. A.; Curtis, M.
Organometallics 1999, 18, 3096; (b) Faller, J. W.; Parr, J.
Organometallics 2000, 19, 3556, For further pertinent
references see Ref. 1a.
24. (a) Barbaro, P.; Pregosin, P. S.; Salzmann, R.; Albinati,
A.; Kenz, R. W. Organometallics 1995, 14, 5160; (b)
Pregosin, P. S.; Salzmann, R. Coord. Chem. Rev. 1996, 15,
35.
25. Schaffner, S.; Macko, L.; Neuburger, M.; Zehnder, M.
Helv. Chim. Acta 1997, 80, 463.
26. Claridge, T. D. W.; Long, J. M.; Brown, J. M.; Hibbs, D.;
Hurtshouse, M. B. Tetrahedron 1997, 53, 4035.
27. Magistrato, A.; Pregosin, P. S.; Albinat, A.; Rothlisberger,
U. Organometallics 2001, 20, 4178.
5. BINAP monophosphine oxide has been nicknamed either
BINAPO (we and others) or BINAP(O) (Grushin) or
BINPO (Faller). Since the acronym BINAPO has been
previously used for different binaphthalene-core ligands
(at least two more), to avoid (or to reduce) misunder-
standings, we would like to suggest the adoption of the
acronym BINPO for its univocal identification.
6. Gladiali, S.; Pulacchini, S.; Fabbri, D.; Manassero, M.;
Sansoni, M. Tetrahedron: Asymmetry 1998, 9, 391.
ꢀ
7. Gladiali, S.; Alberico, E.; Pulacchini, S.; Kollar, L. J. Mol.
28. Yamaguchi, M.; Shima, T.; Yamagishi, T.; Hida, M.
Tetrahedron Lett. 1990, 35, 5049.
Catal. 1999, 143, 155.
8. Gladiali, S.; Medici, S.; Kegl, T.; Kollar, L. Monatsh.
ꢀ
29. Leutenneger, U.; Umbricht, G.; Fahrni, C.; Matt, P. V.;
Pfaltz, A. Tetrahedron 1992, 48, 2143.
30. Tatsuna, Y.; Yoshida, T.; Otsuka, S. Inorg. Synth. 1990,
28, 342.
31. von Matt, P.; Lloyd-Jones, G. C.; Minidis, A. B. E.;
Pfaltz, A.; Macko, L.; Neuburger, M.; Zehnder, M.;
Chem. 2000, 131, 1351.
9. Faller, J. W.; Grimmond, B. J.; D’Alliessi, D. J. J. Am.
Chem. Soc. 2001, 123, 2525.
10. Sekar, G.; Nishiyama, H. J. Am. Chem. Soc. 2001, 123,
3603.
11. Marshall, W. J.; Grushin, V. V. Organometallics 2003, 22,
555.
12. Bunten, K. A.; Farrar, D. H.; Poe, A. J.; Lough, A.
Organometallics 2002, 21, 3344.
13. Faller, J. W.; Parr, J. Organometallics 2000, 19, 1829.
€
Ruegger, H.; Pregosin, P. S. Helv. Chim. Acta 1995, 78,
265.
32. Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.
33. Sheldrick, G. M., SHELXL 97. A computer program for
€
crystal structure refinment, University of Gottingen, 1997.
ꢀ
14. Kollar, L.; Gladiali, S.; Tenorio, M. J.; Weissensteiner, W.
J. Cluster Sci. 1998, 9, 321.
34. International Tables for X-ray Crystallography; Kynoch:
Birmingham, 1974; Vol. IV, pp 99, 100 and 149.
35. Flack, H. D. Acta Crystallogr., Sect. A. 1983, 39, 876.
15. Cyr, P. W.; Rettig, S. J.; Patrick, B. O.; James, B. R.
Organometallics 2002, 21, 4672.