Bulky monodentate phosphoramidites in palladium-catalyzed allylic alkylation reactions: Aspects of regioselectivity and enantioselectivity

Article abstract of DOI:10.1002/chem.200400154

A series of bulky monodentate phosphoramidite ligands, based on biphenol, BINOL and TADDOL backbones, have been employed in the Pd-catalysed allylic alkylation reaction. Reaction of disodium diethyl 2-methyl malonate with monosubstituted allylic substrates in the presence of palladium complexes of the phosphoramidite ligands proceeds smoothly at room temperature. The regioselectivities observed depend strongly on the leaving group and the geometry of the allylic starting compounds. Mono-coordination occurs when these ligands are ligated in [Pd(allyl)(X)] complexes (allyl=C₃H ₅, 1-CH₃C₃H₄, 1-C₆H ₅C₃H₄, 1,3-(C₆H₅) ₂C₃H₃; X=Cl, OAc). The solid-state structure determined by X-ray diffraction of [Pd(C₃H₅)(1)(Cl)] reveals a non-symmetric coordination of the allyl moiety, caused by the stronger trans influence of the phosphoramidite ligand relative to X^-. In all of these complexes, the syn,trans isomer is the major species present in solution. Because of fast isomerisation and high reactivity of the syn,cis complex, the major product formed upon alkylation is the linear product, especially for monosubstituted phenylallyl substrates in the presence of halide counterions. In the case of biphenol- and BINOL-based phosphoramidites, however, a strong memory effect is observed when 1-phenyl-2-propenyl acetate is employed as the substrate. In this case, nucleophilic attack competes effectively with the isomerisation of the transient cinnamylpalladium complexes. The asymmetric allylic alkylation of 1,3-diphenyl-2-propenyl acetate afforded the chiral product in up to 93% ee. Substrates with smaller substituents gave lower enantioselectivities. The observed stereoselectivity is explained in terms of a preferential rotation mechanism, in which the product is formed by attack on one of the isomers of the intermediate [Pd{1,3-(C₆H₅) ₂C₃H₃}(L)-(OAc)] complex.

Full text of DOI:10.1002/chem.200400154

FULL PAPER
Bulky Monodentate Phosphoramidites in Palladium-Catalyzed Allylic
Alkylation Reactions: Aspects of Regioselectivity and Enantioselectivity
[a]
[c]
[a]
[b]
Maarten D.K.Boele,
Paul C.J.Kamer,
Martin Lutz,[b] Anthony L.Spek,
[a]
[a]
Johannes G.de Vries, Piet W.N.M.van Leeuwen,*
and Gino P.F.van Strijdonck
Abstract: A series of bulky monoden-
tate phosphoramidite ligands, based on
biphenol, BINOL and TADDOL back-
bones, have been employed in the Pd-
catalysed allylic alkylation reaction.
Reaction of disodium diethyl 2-methyl
malonate with monosubstituted allylic
substrates in the presence of palladium
complexes of the phosphoramidite li-
gands proceeds smoothly at room tem-
perature. The regioselectivities ob-
served depend strongly on the leaving
group and the geometry of the allylic
starting compounds. Mono-coordina-
tion occurs when these ligands are li-
gated in [Pd(allyl)(X)] complexes
(allyl=C3H5, 1-CH3C3H4, 1-C6H5C3H4,
1,3-(C6H5)2C3H3; X=Cl, OAc). The
solid-state structure determined by X-
ray diffraction of [Pd(C3H5)(1)(Cl)] re-
based phosphoramidites, however, a
veals a non-symmetric coordination of
the allyl moiety, caused by the stronger
trans influence of the phosphoramidite
ligand relative to XÀ. In all of these
complexes, the syn,trans isomer is the
major species present in solution. Be-
cause of fast isomerisation and high re-
activity of the syn,cis complex, the
major product formed upon alkylation
is the linear product, especially for
monosubstituted phenylallyl substrates
in the presence of halide counterions.
In the case of biphenol- and BINOL-
strong memory effect is observed when
1-phenyl-2-propenyl acetate is em-
ployed as the substrate. In this case,
nucleophilic attack competes effective-
ly with the isomerisation of the transi-
ent cinnamylpalladium complexes. The
asymmetric allylic alkylation of 1,3-di-
phenyl-2-propenyl acetate afforded the
chiral product in up to 93% ee. Sub-
strates with smaller substituents gave
lower enantioselectivities. The ob-
served stereoselectivity is explained in
terms of a preferential rotation mecha-
nism, in which the product is formed
by attack on one of the isomers of the
intermediate [Pd{1,3-(C6H5)2C3H3}(L)-
(OAc)] complex.
Keywords: alkylation · asymmetric
catalysis · homogeneous catalysis ·
ligand design · palladium
Introduction
structure of the allyl substrate, ligand or nucleophile and the
nature of the solvent.[1–4]
Allylic-substitution reactions are currently among the most
important and widely studied catalytic reactions in organic
synthesis. Especially in the last decade, enormous progress
has been made in gaining insight in the factors that deter-
mine the outcome of the reaction, such as the metal, the
The vast majority of the studies reported apply palladium
as the metal catalyst of choice. A plethora of ligands has
been designed and employed, mainly with phosphorus and/
or nitrogen donor atoms. Especially the asymmetric allylic
alkylation has received broad attention (Scheme 1).[1–6] Ex-
cellent enantioselectivities have been obtained when C2
symmetrical diphosphines were used, as demonstrated by
[1,6]
[a] Dr. M. D. K. Boele, Dr. P. C. J. Kamer,
Prof. Dr. P. W. N. M. van Leeuwen, Dr. G. P. F. van Strijdonck
University of Amsterdam
Trost et al.
The observed stereoselectivities for these and
other systems can often be explained by the concept of a
“chiral pocket”.[7–10] The chiral induction stems from the se-
lective clockwise or anticlockwise rotation of the allyl
moiety in this pocket upon nucleophilic attack to form the
product h2-alkene complex.
vanꢀt Hoff Institute for Molecular Sciences
Nieuwe Achtergracht 166, 1018 WV Amsterdam (Netherlands)
Fax : (+31)20-525-5604
E-mail: pwnm@science.uva.nl
[b] Dr. M. Lutz, Prof. Dr. A. L. Spek
University of Utrecht
The potential of other classes of ligands has been recog-
À
nised in recent years. These include chiral P N ligands, as
Bijvoet Center for Biomolecular Research
Utrecht (Netherlands)
was demonstrated by Pfaltz and Helmchen, who employed
chiral phosphinooxazoline ligands possessing C1 symme-
try.[5,11–18] The difference in donor properties of the coordi-
nating phosphorus and nitrogen atoms offered the possibili-
[c] Prof. Dr. J. G. de Vries
DSM Research B.V., Postbus 18
6160 MD, Geleen (Netherlands)
6232
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200400154
Chem. Eur. J. 2004, 10, 6232 – 6246

6232 – 6246
of mixed donor ligands or by
the use of bulky, monodentate
ligands. We have described the
application of bulky, monoden-
tate phosphoramidite ligands in
the Suzuki and the Heck reac-
tion.[35] These ligands were
shown to have properties that
distinguish them from most li-
gands used to date. The most
Scheme 1. Asymmetric palladium-catalysed allylic alkylation of a symmetrically substituted allyl substrate.
ty of directing the site of nucleophilic attack to the allylic
carbon atom trans to phosphorus. This is caused by the
larger trans influence exerted by P relative to N, thereby
rendering the carbon atom trans to phosphorus more elec-
trophilic.[14,19–22] Other bidentate P N ligands, such as
À
planar, chiral ferrocenyl derivatives, have also been success-
fully employed, with high regio- and stereoselectivities.[23]
Scheme 3. Asymmetric palladium-catalysed allylic alkylation of an un-
symmetrically substituted allyl substrate and a monodentate ligand.
The number of chiral monodentate ligands that have been
studied in palladium-catalysed asymmetric allylic transfor-
mations is rather limited. The reports on application of
chiral monodentate ligands in the allylic alkylation deal
almost without exception with chiral phosphines.[24–32] In
most cases the enantioselectivities obtained are at best in
the same range as those observed for bidentate analogues.[26]
Among the most prominent examples of successful mono-
dentate ligands are the so-called MOP-type ligands
(Scheme 2), introduced by Hayashi and co-workers.[30,33] It
important feature is the bulkiness of the ligand, resulting in
enforced mono-coordination to palladium, which leads to
enhanced reaction rates. Because of these interesting com-
plexation characteristics, we decided to study bulky phos-
phoramidites in the palladium-catalysed allylic alkylation re-
action, both the non-asymmetric and asymmetric variant. In
the last few years a revival has taken place in the use of
chiral monodentate ligands for asymmetric reactions.[36] In
most cases, however, these ligands are relatively small and
more than one ligand coordinates to the metal centre. The
few early examples of the use of phosphoramidites as li-
gands include the hydroformylation.[37] An important break-
through demonstrating the value of chiral phosphoramidites
has been accomplished by Feringa and co-workers. They
successfully applied this class of ligands in the copper-cata-
lysed asymmetric 1,4-addition.[38] More recently, it has been
shown that monodentate phosphoramidites can function as
effective chiral ligands in other metal-catalysed reactions
such as hydrogenations,[39,40] hydrosilylations[41] and hydrovi-
nylations.[42] Despite the recent increase in interest for phos-
phoramidites as auxiliaries in catalysis, very few applications
in palladium-catalysed allylic subtitution have been reported
yet.[43] Several groups have exemplified the potential of
chiral phosphoramidites in copper-catalysed allylic alkyla-
tions, which proceed through a mechanism (predominantly
SN2’ from s-allyl complexes) different from that of the palla-
dium-catalysed reaction. The application of phosphorami-
dites in iridium-catalysed allylic alkylations has been report-
ed.[44] The reaction was proposed to proceed through s-allyl
complexes as well and resulted in enantioselectivities of up
to 93% with very high regioselectivity to the branched prod-
uct (>99%). In this paper, we present the results we ob-
tained in testing bulky phosphoramidites in palladium-cata-
lysed allylic alkylations of several allylic substrates.[35] The
ligand effects on regio- and enantioselectivity are described,
together with the observation of a large regiochemical
memory effect, and its implications for asymmetric transfor-
mations.
Scheme 2. [Pd((R)-MeO-MOP)]-catalysed regioselective and enantiose-
lective alkylation of 1-substituted cinnamyl derivatives.
was shown that MOP ligands give very selective and active
catalysts in the asymmetric Pd-catalysed reduction of allylic
esters with formic acid.[27,30] Application of these ligands in
allylic alkylation revealed several interesting features with
respect to the regioselectivity of unsymmetrically substituted
allyl groups (Scheme 3). With the use of MeO-MOP, the re-
action of 1-aryl-2-propenyl acetate derivatives with the
sodium salt of dimethyl 2-methyl malonate resulted in for-
mation of the branched product in 90% regioselectivity,
with an ee of 87% (Ar=4-CH3OC6H4, Scheme 2).[34]
Thus, directing the site of nucleophilic attack by a differ-
ence in donor properties could be accomplished by the use
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6233

P. W. N. M. van Leeuwen et al.
FULL PAPER
Results and Discussion
of the allyl moiety, can be formed in E (trans) or Z (cis) ge-
ometry. Attack on the substituted site of the allyl results in
formation of the chiral, branched product. The results for
the catalytic reaction of crotyl acetate ((E)-3-methyl-2-pro-
penyl acetate, 5b), cinnamyl acetate ((E)-3-phenyl-2-pro-
penyl acetate, 6b) and rac-a-vinyl benzylacetate (1-phenyl-
2-propenyl acetate, 7b) and the corresponding chlorides,
disodium diethyl 2-methyl malonate are shown in Table 1.
In all experiments, more than 95% conversion was ob-
tained within two hours. The re-
Ligand synthesis: Several bulky phosphoramidites, based on
bulky biphenol, binaphthol or TADDOL backbones were
prepared. These can be easily synthesised by published
methods[37,45,46] and possess modular properties, allowing
fine-tuning of their steric and electronic characteristics. An
overview of the ligands we applied in the present studies of
the palladium-catalysed allylic alkylation is shown here.
gioselectivity observed in the
allylic alkylation appears to be
largely dependent on the anion-
ic counterion present. For
crotyl chloride (5a), the
branched and linear E product
are formed in equal amounts
(43%), together with a minor
amount of the Z product. Upon
changing to crotyl acetate (5b),
the regioselectivity shifts dra-
matically towards the linear E
product (84%, entry 2). This
change in regioselectivity is ac-
companied by a drop in reac-
tion rate. When 20 equivalents
of tetra-n-butylammonium chlo-
ride are added, the high selec-
tivity towards the branched
product is restored, resulting in
an even slightly higher percent-
age of branched alkene (57%).
A
similar trend is observed
when the methyl substituent is
replaced by the larger phenyl
group, although in this case the
overall formation of the E linear
product is much more favoured
than for the crotyl substrates
(entries 5–7). No formation of
the Z alkene is observed for the
cinnamyl substrate (6). This is
in accordance with observations
in other studies.[19] The addition
of bromide results in an in-
crease of the amount of
branched product formed. Fur-
Allylic alkylation reactions:
Phosphoramidite 1 was tested
in the palladium-catalysed allyl-
ic alkylation reaction of differ-
ent monosubstituted allylic sub-
strates. From these reactants,
three different isomeric prod-
ucts can be formed (see
Scheme 4). The linear product,
arising from nucleophilic attack
on the nonsubstituted terminus Scheme 4. Allylic alkylation of monosubstituted allyl substrates.
6234
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
Table 1. Catalytic alkylation of monosubstituted allyl compounds 5–7
experiments. For the stoichiometric reactions, the regioselec-
tivity obtained when the crotyl complexes are applied is rel-
atively insensitive to the counterion present (Table 2, en-
tries 1–3). This is in sharp contrast to the catalytic reactions
(Table 1, entries 1–4), in which the presence of halide is re-
quired for formation of the branched alkene. Stoichiometric
reactions with the corresponding cinnamyl complexes result
in similar observations. Thus, the application of [Pd(cinna-
myl)(1)(OAc)] yields linear the E and branched product in
a 83:17 ratio (entry 5), whereas in the catalytic reaction the
branched product is formed in only 3% (see Table 1). The
results from the stoichiometric experiments with halide pres-
ent do not deviate significantly from those of the catalytic
reactions, resulting in high selectivity towards the linear
product.
The observed trends can be explained as follows. From
the principle of microscopic reversibility it follows that the
oxidative addition (being the reverse reaction of reductive
elimination) of allylic compounds to Pd0 containing mono-
dentate phosphorous ligands occurs in a cis fashion. Thus,
the initial complexes formed from crotyl or cinnamyl sub-
strates will have the substituent on the allyl moiety and the
ligand in a cis relationship (see Scheme 5).[50–53] Since in the
with disodium diethyl 2-methyl malonate using [{Pd(h3-C3H5)(OAc)2}2]/
1.[a]
Substrate
X
R
TOF[b] Linear Linear Branched
E [%]
Z [%]
[%]
1
2
3
5a
5b
5c
Cl
CH3
180
122
n.d.
174
284
56
43
84
58
40
91
97
99
33
84
14
8
11
3
0
0
0
0
0
43
8
31
57
9
3
1
67
16
OAc CH3
OTf CH3
OAc CH3
Cl C6H5
OAc C6H5
OAc C6H5
4[c] 5b
5
6
6a
6b
7[c] 6b
113
8
7b
OAc 1-C6H5 202
OAc 1-C6H5 307
9[c] 7b
[a] Reaction conditions: Pd/1/allylic substrate/nucleophile=1:1:100:200;
[Pd]=10mm in THF. Reaction time is 2 h. Selectivity after final conver-
sion, determined by GC with dihexyl ether as internal standard. [b] Initial
turnover frequency, determined after 5 mins. [c] 20 equivalents (to Pd) of
(nBu)4NCl added.
thermore, for the phenyl-substituted allyl substrate, the allyl-
ic substitution displays a large memory effect. Thus, the al-
kylation of a-vinylbenzyl acetate 7b yields a product mix-
ture in which 67% of the branched product is formed, com-
pared to only 3% starting from the isomeric linear substrate
(entries 6 and 8). This memory effect largely disappears
when extra halide is added to the reaction mixture (entry 9).
To gain more insight into the origin of the observed regio-
selectivity, stoichiometric allylic alkylation reactions were
also carried out. In the stoichiometric reactions, the complex
geometries have fully equilibrated to their thermodynamic
ratios prior to alkylation. Therefore the regioselectivity ob-
served in these alkylations will reflect the thermodynamic
composition of the [Pd(allyl)(1)(X)] complexes, provided
that the nucleophilic attack is faster than isomerisation of
the Pd–allyl complex. Recent research into the regioselectiv-
ity of stoichiometric allylic alkylation with complexes with
bidentate ligands has shown that the syn complex reacts
mainly towards the linear E product, with formation of
small quantities of the branched product. The isomeric anti
complex reacts to the branched product or the linear Z
product.[14,47–49] The reactions were performed by addition of
a large excess of disodium diethyl 2-methyl malonate to a
solution of the [Pd(allyl)(1)(X)] complex (X=Cl, OAc).
The results of the stoichiometric alkylation reactions are
shown in Table 2.
Scheme 5. Possible explanation for regioselectivity observed in palladi-
um-catalysed allylic alkylation in the presence of monodentate phosphor-
amidites.
The observed selectivities in the stoichiometric alkylation
differ significantly from those obtained from the catalytic
NMR spectra only the syn,trans complexes (b) are observed
(vide infra), we conclude that this is the most stable com-
plex geometry. Therefore, the initially formed syn,cis com-
plexes a will isomerise to the corresponding trans complex,
by means of a pseudorotation mechanism.[1,54–56] The ob-
served regioselectivity during catalysis will depend on the
rate of this isomerisation, the equilibrium constant of the
equilibrium C between a and b (Scheme 5) and the rate of
nucleophilic attack (A and B). It has been shown by several
groups that the addition of halide anions can accelerate dy-
namic processes in Pd(p-allyl) complexes considera-
bly.[47,57–59] Since attack of the nucleophile will mainly occur
on the allyl terminus trans to phosphorus (which has a much
larger trans influence than the anionic ligand),[60] the syn,-
Table 2. Stoichiometric alkylation of [Pd(3-RC3H4)(1)(X)] with disodium
diethyl 2-methyl malonate.[a]
Substrate
X
R
Linear
E [%]
Linear
Z [%]
Branched
[%]
1
2
5a
5b
5b
6a
6b
6b
Cl
CH3
CH3
CH3
C6H5
C6H5
C6H5
39
37
31
94
83
98
8
8
3
0
0
0
53
55
66
6
17
2
OAc
OAc
Cl
OAc
OAc
3[b]
4
5
6[b]
[a] Reaction conditions: Pd/nucleophile=1:50; [Pd]=10mm in THF. Se-
lectivity determined by GC. [b] 20 equivalents (to Pd) of (nBu)4NCl
added.
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6235

P. W. N. M. van Leeuwen et al.
FULL PAPER
trans complex b will mainly react to the branched product
(route B). The presence of halide counterions (Cl or Br) en-
hances the rate of isomerisation and therefore the amount
of branched product formed. The isomerisation C from the
reacts mainly to the branched product. Therefore, the
branched product is formed in excess when crotyl substrates
are employed in the presence of halide counterions. In con-
trast, when phenyl-substituted allyl substrates are used the
cis isomer reacts at a much higher rate than the trans com-
plex (kA @kB in Scheme 5), resulting in the formation of
mainly linear product, despite the fact that it is present in
only very small quantities under equilibrium conditions.
Only when the branched a-vinylbenzyl acetate 7b is em-
ployed in the absence of halide anions, a significant amount
of the branched product is formed, due to slow isomerisa-
tion.
initially formed cis complex
a in the case of the
[Pd(RC3H4)(1)(OAc)] complexes is slower compared to the
halide-containing intermediates, resulting in formation of a
larger amount of the linear (E) product through pathway A.
This is what is observed for crotyl substrates.
For the phenyl-substituted substrates, the situation is
slightly different. Similar to the crotyl substrates, the isomer-
isation C in the cinnamylpalladium complexes will be rela-
tively slow when OAcÀ is the counterion. This is confirmed
by the presence of the strong memory effect: when the 3-
phenyl-substituted substrate 7b is employed, which initially
forms the trans complex (b), a high selectivity towards the
branched product is observed. In contrast, the 1-acetoxy
substrate (6b) will form the cis isomer (a) and therefore
mainly reacts towards the linear product. The presence of
halide enhances the rate of isomerisation between the cis
and trans complexes drastically, resulting in faster isomerisa-
tion relative to nucleophilic attack. In this situation, the
linear (E) product is formed predominantly. However, since
only the trans complex is observed in NMR spectra, this im-
plies that for cinnamyl complexes the cis complex reacts
with higher rate than the corre-
Asymmetric allylic alkylations: It is clear that phosphorami-
dite ligands are a useful class of ligands for palladium-cata-
lysed allylic alkylation reactions. Moreover, the possibility
of regioselective nucleophilic attack on the substituted allyl
terminus enables the enantioselective formation of the
(chiral) branched product. Therefore, we tested chiral phos-
phoramidites in the asymmetric variant of the allylic alkyla-
tion reaction of both symmetric and nonsymmetric sub-
strates.
First the alkylation of 1,3-diphenyl-2-propenyl acetate (8)
with dimethyl malonate anion as the nucleophile (Scheme 6)
was tested. Thus, 1 mol% [Pd(C3H5)(m-OAc)]2 and two
sponding trans complex, result-
ing in the predominant forma-
tion of linear product. Being
higher in energy (and thus un-
favoured in the equilibrium
mixture), the energy barrier of
the cis complex for nucleophilic
attack is apparently lower than
Scheme 6. Asymmetric allylic alkylation of 1,3-diphenyl-2-propenyl acetate (8).
that of the trans complex, re-
sulting in higher reaction rates and higher selectivity to the
linear product isomer. Additionally, the increased steric hin-
drance of the phenyl group at the substituted allyl terminus
hampers attack of the nucleophile on this carbon atom. Nu-
cleophilic attack on this position is not favoured by electron-
ic factors either, which render the unsubstituted terminus
more electrophilic.[48] Therefore, a preference for the linear
product exists when employing cinnamyl substrates. Because
of the large trans influence of the phosphoramidite ligand,
this product will be formed through the cis isomer (route A
in Scheme 5).
These conclusions are supported by the observations in
the stoichiometric reactions. If the isomerisation of the
[Pd(cinnamyl)(1)(X)] (X=Cl, Br) complexes is very fast
compared to nucleophilic attack, the stoichiometric reac-
tions should not deviate significantly from the results ob-
tained in the catalytic reactions. This is indeed what we have
found (Tables 1 and 2). When no halide is present, more of
the branched product is formed, due to the slower cis/trans
isomerisation.
equivalents of ligand were stirred at room temperature for
15 minutes before the reaction was started. The nucleophile
was formed in situ by addition of BSA (N,O-bis(trimethylsi-
lyl) acetamide) and a catalytic amount of KOAc. The results
of the catalytic reactions are shown in Table 3.
Use of TADDOL-based phosphoramidites in the asym-
metric alkylation of 1,3-diphenyl-2-propenyl acetate resulted
in excellent yields and good enantioslectivity of the product
in most cases. Thus, employing (R,R)-2b as the chiral auxili-
ary, product 9 was obtained in 89% ee, with S configuration.
Lowering the reaction temperature improved the enantiose-
lectivity marginally, to 90% (entry 4), but the conversion
was incomplete in this case. Reaction with the correspond-
ing ligand with the opposite stereochemistry in the TAD-
DOLate backbone results in the same ee, with the opposite
enantiomer (R) formed in excess (entry 5). Changing the
structure of the backbone at the remote acetal moiety
(ligand 2i) resulted in a decrease of the enantioselectivity to
81% (entry 6). This can be explained by the fact that the cy-
clohexyl group slightly distorts the adjacent ring structure,
and hence affects the geometry of the aryl substituents. Var-
iation of the aryl moieties also has a large influence on the
stereochemical outcome of the reaction: both the 2-naph-
thyl- (2 f) and phenyl-containing (2e) ligands give rise to
In summary, halides enhance the rate of isomerisation in
[Pd(allyl)(1)(X)] complexes, resulting in more rapid equili-
bration of the isomeric complexes from the initially formed
syn,cis to the more stable syn,trans form. The latter isomer
6236
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
Table 3. Asymmetric allylic alkylation of 8 with dimethyl malonate in
the presence of phosphoramidite ligands 2–4.[a]
Ligand
T
[8C]
Yield[b]
[%]
ee (config)[c]
[%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15[d]
16[e]
4
25
25
25
0
65
n.d.
>99
87
>99
70
68 (R)
67 (R)
89 (S)
90 (S)
90 (R)
81 (S)
61 (S)
73 (S)
83 (S)
88 (S)
92 (S)
87 (S)
87 (R)
93 (R)
66
(R)-3b
(R,R)-2b
(R,R)-2b
(S,S)-2b
(R,R)-2i
(R,R)-2 f
(R,R)-2e
(R,R)-2 g
(R,R)-2a
(R,R,S)-2d
(R,R,R)-2c
(S,S,S)-2d
(S,S,R)-2c
(R,R)-2b
(R,R)-2b
25
25
25
25
25
25
25
25
25
25
25
25
86
>99
n.d.
n.d.
>99
>99
>99
98
Figure 1. Correlation between the enantiopurity of ligand 2b and the
enantioselectivity in the reaction of 8 to 9.
>99
34
dental linear behaviour for a bis-ligand complex remains
feasible.
85 (S)
[a] Reaction conditions: Pd/ligand/allyl/dimethyl malonate/BSA=
1:2:100:150:150 in CH2Cl2. Catalyst incubation time is 15 mins, reaction
time 16 h. [b] Based on conversion of the acetate, determined by GC
with dihexyl ether as internal standard; n.d.=not determined. [c] Deter-
mined by HPLC (Daicel OD). [d] Reaction carried out in THF with diso-
dium diethyl 2-methyl malonate as the nucleophile. [e] 20 equivalents (to
Pd) of (nBu)4NBr added.
As was demonstrated, bulky phosphoramidites can steer
the regioselectivity of monosubstituted allylic compounds
towards the chiral, branched product. Therefore, we also
probed these TADDOL-based ligands in the asymmetric al-
lylic alkylation of other prochiral, nonsymmetrically substi-
tuted substrates. The results of the reactions performed with
rac-1-phenyl-2-propenyl acetate (a-vinylbenzyl acetate 7b),
rac-1-phenyl-3-but-1-enyl acetate (1-methyl-3-phenylallyl
acetate 10) and cinnamyl acetate (6b) with dimethyl malo-
nate in dichloromethane at room temperature are shown in
Table 4.
lower eeꢀs than the 3,5-dimethylphenyl group (entries 7 and
8).
The structure of the amino group in the ligand has a
much less pronounced influence on the enantioselectivity
(entries 9–14). The piperazine-substituted ligand 2g gives
lower ee. In the case of the N,a-dimethylbenzylamino sub-
stituent, the possibility of chiral cooperativity effects arises.
Such effects, although not extremely large, have been ob-
served for these ligands in asymmetric intramolecular Heck
reactions.[35] In the asymmetric alkylations, a positive effect
is observed when the amine moiety and the TADDOLate
backbone have the opposite configuration (entry 11 and 14).
The latter determines the absolute configuration and enan-
tioselectivity of the reaction to a large extent (Table 3), but
a further increase to 93% ee is observed when ligand
(S,S,R)-2c is used (entry 14). Changing the nucleophile to
disodium diethyl 2-methyl malonate in THF has a strong
detrimental effect on the level of stereocontrol (entry 15).
The addition of extra halide anions lowers both the product
yield and ee (entry 16).
Table 4. Asymmetric allylic alkylation of 6b, 7b and 10b with dimethyl
malonate using phosphoramidite ligands 2b and 3b.[a]
Substrate
T
[8C]
Ligand
Yield[b]
[%]
Regio-
ee[d]
[%]
selectivity[c]
1
10b
25
(R,R)-2b
>95
>99:1
(11:12)
94:6
94:6
98:2
32
2
3
7b
7b
7b
6b
7b
25
0
25
25
25
(R,R)-2b
(R,R)-2b
(R,R)-2b
(R,R)-2b
(R)-3b
>99
79
>99
>99
>99
18
25
n.d.
n.d.
7
4[e]
5
98:2
26:74
6
[a] Reaction conditions: Pd/ligand/allyl substrate/dimethyl malonate/
BSA=1:2:100:150:150 in CH2Cl2. Catalyst incubation time is 15 mins, re-
action time 16 h. [b] Based on conversion of the acetate, determined by
GC with dihexyl ether as internal standard. [c] Lineair/branched ratio,
determined by GC. [d] Determined by HPLC (Daicel OD), n.d.=not de-
termined. [e] 20 equivalents (to Pd) of (nBu)4NBr added.
To gain insight into the coordination mode of the bulky
phosphoramidite ligands during the catalytic reaction, we
tested the dependence of the observed enantioselectivity in
the reaction of 8 with dimethyl malonate on the enantiopur-
ity of the ligand employed. Such relationships have been
shown to give important information on the nature of the
catalytically active species as bis-ligand complexes often
show nonlinear effects.[52,53] The results for ligand 2b are
shown in Figure 1. Within experimental error, the stereo-
chemical outcome of the reaction is linearly correlated to
the ee of the ligand. Therefore, a reaction mechanism in
which the active catalytic species has only one chiral ligand
coordinated to the Pd centre is most likely, as also observed
for the enantioselective Heck cyclisations,[35,61,62] but an acci-
As for most ligands reported,[1] the high level of enantio-
selectivity obtained for the 1,3-diphenylallyl substrate (8)
drops dramatically when allylic substrates containing smaller
substituents are employed. Thus, reaction of 10b with di-
methyl malonate in the presence of ligand (R,R)-2b results
in complete regioselective formation of the product stem-
ming from nucleophilic attack at the methyl-substituted allyl
terminus (11) in 32% ee (Scheme 7, Table 4 entry 1), com-
pared to 89% ee for diphenylallyl substrate 8. Moreover,
the strong regiochemical memory effect observed when
ligand 1 was employed for the alkylation of substrate 7b
(Table 1) is not found for the TADDOL-based chiral phos-
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6237

P. W. N. M. van Leeuwen et al.
FULL PAPER
logues and were therefore not
isolated.
For the nonchiral ligand 1,
the complex [Pd(C3H5)(1)(Cl)]
can exist as two enantiotopic
isomers (possible conformers
arising from different orienta-
tions within the ligand exclud-
ed): the complex with the allyl
“up” (also called exo) or, alter-
Scheme 7. Palladium-catalysed allylic alkylation with unsymmetrically substituted allyl substrates.
phoramidites (Table 4, entries 2–5). Apparently, for all non-
symmetrically substituted allyl substrates, the isomerisation
and/or the relative reactivity of the transient [Pd(allyl)(2)]
complexes largely favours reaction from the electronically
and sterically less stable isomer bearing the phenyl substitu-
ent cis to the bulky ligand. In the case of cinnamyl deriva-
tives this results in the formation of mainly linear (nonchi-
ral) product. When BINOL-based ligand (R)-3b is used, the
regiochemical selectivity for the branched product is higher
again. This ligand is expected to have steric properties that
are similar to those of ligand 1, because of the resemblance
of the backbones in both ligands. Nevertheless, the enantio-
selectivity observed for the reaction of (R)-3b is very low,
also relative to the enantioselectivity obtained for (R)-3b in
the alkylation of 1,3-diphenylallyl acetate 8 (7% vs 67% ee
for the latter compound). Possibly, the diminished steric
constraints reduce the steric interactions that discriminate
the two possible pathways upon allyl rotation. The reason
for this remarkable influence of
natively, “down” (endo).[55] These two isomers are enantio-
mers in the case of the C3H5 ligand and any other C2v sym-
metric allyl moiety. As expected, the 31P NMR spectrum
showed only one sharp resonance at d=140.1 ppm (CDCl3,
208C). Addition of an extra 0.5 equivalents of ligand did not
change the spectrum for the complex, but only resulted in
the appearance of the signal of uncoordinated phosphorami-
1
dite (153.3 ppm). Inspection of the corresponding H NMR
spectrum revealed that all allyl protons are inequivalent,
showing slightly broadened signals at room temperature.
This confirms that only one ligand coordinates to the Pd
centre. The broadening of the allyl signals can be explained
by slow syn/anti exchange through the well-known h3-h1-h3
mechanism (Scheme 8, R=H). The exchange is faster for
1
the allyl protons with the lowest shifts in the H NMR spec-
trum. Cooling down to À208C resulted in sharpening of the
allyl resonances, while the peak in the 31P NMR spectrum
broadened significantly. Phosphorus decoupling of the
the ligand backbone on the re-
gioselectivity (i.e. the memory
effect) is unclear at present.
Complex synthesis and charac-
terisation: To investigate the
coordination behaviour of the
bulky
phosphoramidites
in
more detail, we synthesised var-
ious [Pd(allyl)(L)(X)] (allyl=
C3H5,
(C6H5)2C3H3; L=bulky phos-
phoramidite, X=Cl, OAc)
C6H5C3H4,
1,3-
complexes and studied them
using NMR spectroscopy. The
chloride-containing complexes
could readily be obtained by re-
action of the ligand with 0.5
equivalents of the correspond-
ing [{Pd(allyl)(m-Cl)}2] dimer,
which yielded the desired com-
plexes
in
near-quantitative
yields. The acetate complexes
were prepared from the [{Pd(al-
lyl)(OAc)}2] dimer, which was
prepared from the chloro dimer
by reaction with AgOAc. These
complexes appeared to be less
stable than their chloro ana-
Scheme 8. Possible isomerisation mechanisms in [Pd(allyl)(L)(X)] complexes.
6238
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
proton spectrum resulted in the change of the two apparent
triplets stemming from the allyl moiety into two doublets.
Therefore, these two signals can be attributed to the protons
being trans to the phosphoramidite ligand. This implies that
the syn/anti exchange is faster for the protons situated cis to
the phosphorus donor. On further cooling, the phosphorus
resonance decoalesces into two broad singlets of approxi-
mately equal intensity (1:0.9 at À708C). This decoalescence
is accompanied by broadening of the isopropyl protons from
1
the amine moiety of the ligand in the H NMR spectrum.
The two observed different species at low temperature are
likely to arise from the two different orientations of the bi-
phenyl backbone of the ligand. When rotation around the
À
C C axis is restricted, the ligand becomes chiral due to atro-
pisomery, giving rise to two diastereomers in the palladium–
allyl complex.
We were able to grow crystals of [Pd(C3H5)(1)(Cl)] by
slow evaporation of a solution of the complex in hexanes.
X-ray crystal structure determination revealed that the com-
pound crystallises as a racemate in the centrosymmetric
space group P21/c. As a consequence of the disordered allyl
group both diastereomers (in the crystal no biphenyl rota-
tion will take place) are present in a 1:1 ratio. The structure
of the (S)-isomer (allyl up) is shown in Figure 2 and con-
firms the mono-coordination of the bulky ligand to the pal-
ladium centre. The nonsymmetric environment of the allyl
moiety is reflected in the bond lengths observed in the
solid-state structure (see Table 5). The allyl ligand was re-
Table 5. Selected distances [Š] and bond angles [8] from the crystal
structure of [Pd(C3H5)(1)(Cl)].[a]
À
À
Pd(1) C(1)
2.1809(19)
2.174(4)
2.159(4)
2.150(2)
2.2960(5)
Pd(1) Cl(1)
2.3598(5)
1.411(5)
1.383(5)
1.350(5)
1.437(5)
À
À
Pd(1) C(2)(R)
C(2)(R) C(1)
À
À
Pd(1) C(2)(S)
C(2)(R) C(3)
C(2)(S) C(1)
C(2)(S) C(3)
À
À
Pd(1) C(3)
À
À
Pd(1) P(1)
Cl(1)-Pd(1)-P(1)
P(1)-Pd(1)- C(3)
96.828(17)
103.62(6)
Cl(1)-Pd(1)-C(1)
92.09(6)
[a] C(3)(R) and C(3)(S) belong to two different disorder components.
Figure 2. Molecular structure of [(S)-Pd(C6H5C3H4)(1)(Cl)]: side-view
(top) and front-view (bottom).
fined with a disorder model. In this model the end carbon
atoms remain on the same position. The observed signifi-
cantly dissymmetric character of the allyl group is thus relia-
monodentate phosphines.[63] In addition, sterically the trans
site seems more favourable for attack, since there is virtually
no steric hindrance present on this side of the allyl, as can
be seen from the crystal structure. This effect will be even
more important for monosubstituted allyl substrates, since
the substituents on the newly formed sp3 carbon atom have
to bend backwards during the reaction, into the relative
empty space around the chloride ligand.
Complexes bearing unsymmetrically substituted allyl moi-
eties have a larger number of possible isomers than the un-
substituted ones. For the cinnamyl-containing palladium
complex for example, eight different isomers can be envis-
À
bly determined; the Pd C bond trans to the phosphorami-
dite ligand is longer than the one at the cis position
(2.1809(19) Š vs 2.150(2) Š respectively, see Table 5) and is
caused by the stronger trans influence of the phosphorus-
amidite relative to the chloride.[63] Due to the disorder of
À
the allyl group the C C distances are less accurately deter-
À
mined. The length of the C C bond within the allyl group
trans to phosphorus atom is 1.350(6) Š and shows signifi-
À
cantly more double bond character than the other C C
bond, which is 1.437(5) Š. Nucleophilic attack will take
place preferentially on the trans terminus, which is more
electrophilic. Remarkably, this difference is reversed (al-
though smaller) for the R isomer in the crystal structure.
This reversed order has been observed before in allyl com-
plexes bearing bidentate phosphinyloxazoline ligands[18] or
À
aged (again assuming fast rotation around the Pd P axis):
the syn,trans-, syn,cis-, anti,trans- and anti,syn complexes,
each having two enantiomers. However, in the case of non-
chiral phosphoramidite 1, the monosubstituted allyl com-
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6239

P. W. N. M. van Leeuwen et al.
FULL PAPER
plexes show a behaviour similar to the unsubstituted com-
plexes. [Pd(1-C6H5C3H4)(1)(Cl)] has four different allyl reso-
nances in the 1H NMR spectrum at room temperature. From
Origin of enantioselectivity: From the NMR studies of the
[Pd(1,3-diphenylallyl)((R,R)-2b)(Cl)] complex, it is conclud-
ed that the complex exists mainly as one isomer. Based on
steric considerations, we assume that the syn,syn geometry
of the 1,3-diphenylallyl moiety is largely favoured. In this
case, two isomeric complexes can be envisaged: one with
the chirality from the [Pd(allyl)(L*)(X)] complex with S
configuration and one with the opposite R configuration.
These complexes can interconvert (epimerise) through p ro-
tation (see Scheme 8). Semi-empirical modelling studies
(PM3(tm) level) did not enable us to determine which of
the diastereomers is thermodynamically the most stable and
thus corresponds to the one observed by NMR spectroscopy.
However, assuming that nucleophilic attack will take place
trans to the phosphorus atom,[13,15,17,64] the reactive complex
should be the one with the allyl fragment coordinated in an
R configuration when employing the (S,S)-TADDOLate
backbone (see Table 3, entry 4).
We explain the observed enantioselectivities in terms of
steric considerations. We assume a transition state that re-
sembles more the h2-alkene product complex rather than
the p-allyl starting compound. Upon nucleophilic attack, the
hybridisation of the carbon atom at which the new bond is
formed changes from sp2 to sp3. The coordination mode of
the allyl moiety changes from h3 to h2. To enable this, the
pre (R)-allyl has to rotate in a clockwise fashion for the
(Rax)-allyl complex, whereas the (Sax)-allyl complex will
À
À
the P H and H H couplings observed on the hydrogen
atom attached to the benzylic carbon atom (3J(P,H)=14 Hz,
3J(H,H)=15 Hz) in this complex we conclude that the pre-
fered geometry of the cinnamyl is trans with respect to the
phosphorus atom and syn to the central allyl hydrogen
atom. No other complexes are observed, as is confirmed by
31P NMR spectroscopy, which shows one signal at
141.4 ppm. This confirms the assumption that the allyl
moiety will orient itself in a fashion in which it experiences
the least steric hindrance, that is, trans,syn to the bulky
ligand. This is also the electronically most stable isomer.
The signals of the two protons on the nonsubstituted termi-
nus of the allyl ligand are very broad at room temperature,
indicating syn/anti exchange. No syn/anti isomerisation is ob-
served for the other side of the allyl over a temperature
range of À508C to +558C. The acetato complexes showed
similar behaviour. Upon addition of an excess of ligand to a
solution of [Pd(CH3C3H4)(1)(OAc)], prepared in situ, no
1
changes in the the H NMR spectrum were observed, and
only the signal of free 1 appeared in the 31P NMR spectrum.
Therefore we conclude that also in the [Pd(allyl)(L)(OAc)]
complexes only one phosphoramidite ligand is coordinated
to palladium.
Next, the behaviour of complexes bearing chiral bulky
phosphoramidites was studied. The [Pd(C3H5){(R,R)-
2b}(Cl)] complex appeared to be present as two diastereo-
mers, as indicated by the presence of two resonances in the
31P NMR spectrum (d=119.3 and 117.3 ppm, CD2Cl2,
À208C). The ratio of the two diastereomers is temperature
dependent, ranging from 2.7:1 at À508C to 1.5:1 at 208C.
On further increasing the temperature, coalescence starts to
occur. Unfortunately we were unable to determine the coa-
lescence temperature due to decomposition of the complex
above 608C (CDCl3). Also in the 1H NMR spectrum, the
presence of the two diastereomers was observed. Overlap of
the signals of the ligand and the allyl hampered complete as-
signment of all the resonances, but by means of 2D NMR
(GHSQC) spectroscopy we were able to confirm that both
species are p-allyl complexes. Because of the chirality of the
ligand, the two isomers stemming from the different orienta-
tion of the allyl fragment are diastereomers, one with the
allyl group in the endo position and one with it in the exo
show counterclockwise rotation giving the
(Scheme 9). Inspection of the complex structures obtained
S product
Scheme 9. Pathways of nucleophilic attack on the trans-termini yielding
the two enantiomers.
position.
The
analogous
complex
[Pd(1,3-
(C6H5)2C3H3){(R,R)-2b}(Cl)], bearing the 1,3-diphenylallyl
moiety, showed only one signal (d=119.4 ppm in CHCl3) in
the 31P NMR spectrum over a broad temperature range
(À60 to 408C). In the H NMR spectrum, only one isomer
from
modeling
reveals
that
in
the
[Pd(1,3-
(C6H5)2C3H3){(S,S)-2b}(Cl)] complex, the ligand adopts a
geometry in which the lower left side of the allyl group is
shielded by one of the aryl groups of the ligand backbone
(see Figure 3). This structure closely resembles the ligand
conformation found in the X-ray structure elucidated of the
related [{Pd(4-C6H4CN)(2)(m-Br)}2] complex.[35] Upon attack
on the S isomer, the counterclockwise rotation moves the
phenyl group at the sp2-hybridised carbon atom in close
proximity of this aryl group. In contrast, the clockwise rota-
tion (from the R complex) will result in relief of steric inter-
actions, since both sides of the product alkene rotate into
less occupied space around the metal centre. Therefore, we
1
was observed (> 95%) as judged from the presence of two
allyl signals (the third resonance is obscured by overlap with
signals of the ligand). No significant change was observed in
the 1H NMR spectrum upon variation of the temperature
either. This suggests the presence of only one major diaster-
eomer in solution for [Pd(1,3-(C6H5)2C3H3){(R,R)-2b}(Cl)].
Unfortunately, we were unable to obtain crystals suitable
for X-ray analysis of allyl complexes with 2 as the ligand to
confirm the complex geometry.
6240
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
from the results generally obtained with symmetrical biden-
tate ligands, especially with respect to regioselectivity. From
the studies on isolated complexes, it was concluded that in
the allylpalladium complexes only one ligand can coordinate
to the metal centre. Despite this monocoordination, high
enantioselectivities can be obtained with bulky chiral phos-
phoramidites and disubstituted allyllic substrates. The eeꢀs
are significantly lower for substrates with smaller substitu-
ents or non-symmetrically substituted allyl compounds.
Experimental Section
General remarks: All experiments were carried out under a purified ni-
trogen atmosphere by using standard Schlenk techniques unless noted
otherwise. Solvents were purchased from Acros and dried prior to use.
Toluene was distilled from sodium; THF, hexanes and diethyl ether from
sodium/benzophenone ketyl and dichloromethane from calcium hydride.
All amines employed were distilled from calcium hydride prior to use.
Phosphorus trichloride was distilled prior to use and stored at À208C.
Hexamethyl phosphoramide (HMPT), 1,8-bis(dimethylamino)naphtha-
lene, N,O-bis(trimethylsilyl) acetamide (BSA), cinnamyl chloride and di-
methyl malonate were purchased from Acros and used as received.
Other allylic substrates were prepared following standard procedures.[65]
The TADDOL-type backbones employed were synthesised according to
literature procedures.[66,67] (R)-3,3’-Trimethylsilyl-1,1’-binaphtyl-2,2’-diol
was synthesised as reported by Buisman et al.[68] N-Benzyl-(1R,2S)-nor-
ephedrine was synthesised by a literature procedure.[69] [{Pd(C3H5)(m-
Cl)}2], [{Pd(cinnamyl)(m-Cl)}2] and [{Pd(1,3-diphenylallyl)(m-Cl)}2] were
synthesised according to literature procedures.[70, 71] Disodium diethyl 2-
methyl malonate (0.5m in THF) was prepared from diethyl 2-methyl mal-
onate and NaH in THF at 08C. [{Pd(crotyl)(m-OAc)}2] was freshly pre-
pared from the corrseponding chloride-bridged dimer and AgOAc
(1.0 equiv) in the appropriate solvent, followed by filtration over Celite.
Figure 3. Modelled structure (PM3(tm)-level) of the two syn,syn [Pd{1,3-
(C6H5)2C3H3}{(S,S)-2b}(Cl)] complexes (left), schematically represented
when viewed along the Pd–allyl axis on the right (the plane represents
the aryl-fragment of the ligand blocking the allyl moiety upon rotation,
empty circles represent minor steric interactions).
expect the clockwise rotation to be more facile, resulting in
the formation of the R product, as is observed. This simple
steric model is supported by the fact that replacement of the
3,5-dimethylphenyl groups in the TADDOLate ligands by
other, less bulky aryl moieties results in a large decrease of
stereoinduction, because of decreased steric interactions
during the anticlockwise rotation of the alkene being
formed, lowering the enantiodiscrimination. In this model,
the amine functionality is not expected to have a large influ-
ence in the stereochemical outcome of the reaction, which is
confirmed by experiment.
Changing the substrate from 1,3-diphenylallyl to the
smaller 1-methyl-3-phenylallyl or even cinnamyl, also has a
large detrimental effect on the observed eeꢀs (Table 4). The
simple model presented above for the mechanism cannot
fully account for this observation, since one would expect
the enantioselectivity to be mainly dependent on the bulk of
the substituent on the side of the allyl group cis to phospho-
rus and rather insensitive towards substitution on the trans
carbon atom of the allyl moiety. Probably, the smaller bulk
of the substituents bending backwards on the carbon atom
on which the nucleophile attacks also plays an important
role in interacting with the chirality of the ligand and thus
preventing anticlockwise rotation.
NMR spectra were recorded on a Varian Mercury 300 (300.1 MHz) in
CDCl3 and are reported in ppm with tetramethylsilane (1H and 13C) and
85% H3PO4 (31P) as external standards. Thin-layer chromatography was
carried out on Macherey-Nagel SIL G/UV plates of Kieselgel 60.
Column chromatography was performed with silica 60 (SDS Chromagel,
70–200 mm). GC measurements were performed on a Shimadzu GC-17A
apparatus (split/splitless, equipped with a F.I.D. detector and a BPX35
column (internal diameter of 0.22 mm, film thickness 0.25 mm, carrier gas
70 kPa He)) and an Interscience HR GC Mega 2 apparatus (J&W Scien-
tific, DB-1 column, 30 m/inner diameter 0.32 mm/film thickness 3.0 mm,
carrier gas 70 kPa He, F.I.D. detector). GC/MS measurements (E.I. de-
tection) were performed on a HP 5890/5971 apparatus, equipped with a
ZB-5 column (5% cross-linked phenyl polysiloxane) with an internal di-
ameter of 0.25 mm and film thickness of 0.25 mm. Melting points were
measured on a Gallenkamp melting point apparatus and are reported un-
corrected. HPLC measurements for determination of enantiomeric ex-
cesses were performed using a Gilson apparatus equipped with a Dyna-
max UV-1 absorbance detector. High-resolution mass spectra were re-
corded at the Department of Mass Spectroscopy at the University of
Amsterdam using FAB+ ionisation on a JEOL JMS SX/SX102A four
sector mass spectrometer with 3-nitrobenzyl alcohol as a matrix. Elemen-
tal analyses were performed at the Department of Microanalysis at the
Rijksuniversiteit Groningen (The Netherlands).
(1R,7R)-4-Dimethylamino-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphen-
yl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-2a): (2R,3R)-
O-isopropylidene-1,1,4,4-tetra(3,5-dimethylphenyl)threitol
(300 mg,
0.52 mmol) was dried by dissolving the compound in toluene (5 mL) and
evaporating evaporating thoroughly to dryness (3 times). Next, the diol
was dissolved in toluene (20 mL). The solution was cooled to 08C and a
solution of HMPT(113 mL, 0.62 mmol) together with 1H-tetrazole (ca.
1 mg) in toluene (5 mL) was added dropwise. The mixture was warmed
to room temperature and stirred for 2 h and subsequently refluxed for
24 h. After removal of the solvent in vacuo the resulting foamy solid was
purified by column chromatography (hexanes/EtOAc/triethylamine, 95/5/
Conclusion
Bulky monodentate phosphoramidite ligands can successful-
ly be applied in the Pd-catalysed allylic alkylation reaction
with carbon nucleophiles. The results obtained with these li-
gands show interesting features that differ considerably
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6241

P. W. N. M. van Leeuwen et al.
FULL PAPER
2.5 v/v/v) to yield the product as a white powder. Yield: 165 mg (49%).
1H NMR (300 MHz, CDCl3): d=7.39 (s, 2H; Ar-H), 7.19 (s, 2H; Ar-H),
7.06 (s, 4H; Ar-H), 6.87 (s, 2H partial overlap; Ar-H), 6.86 (s, 2H partial
overlap; Ar-H) 5.08 (dd, J=3.3, 8.2 Hz, 1H; CH), 4.77 (d, J=8.2 Hz,
1H; CH), 2.73 (d, 3J(P,H)=10.4 Hz, 6H; NCH3), 2.29–2.26 (m, 24H;
ArCH3), 1.35 (s, 3H; OCCH3), 0.28 ppm (s, 3H; OCCH3); 13C{1H} NMR
(75.5 MHz, CDCl3): d=147.2, 146.9, 142.0, 137.4, 137.0, 136.7, 136.4,
129.3, 129.0, 129.0, 128.8, 127.1, 126.8, 126.7, 125.3, 125.1, 111.8, 83.2,
83.0, 82.7, 81.8, 81.3, 81.2 (the area 83.2–81.2 ppm contains some extra
J=8.3 Hz, 1H; OCH), 3.26 (dq, J=11.0 (3J(P,H)), 7.0 Hz, 4H;
N(CH2CH3)2), 2.30–2.26 (m, 24H; ArCH3), 1.42 (s, 3H; OCCH3), 1.17 (t,
J=7.0 Hz, 6H; N(CH2CH3)2), 0.25 ppm (s, 3H; OCCH3); 13C{1H} NMR
(75.5 MHz, CDCl3): d=147.6, 147.2, 142.4, 137.3, 136.9, 136.7, 136.4,
129.2, 129.0, 128.8, 128.8, 127.2, 126.8, 125.3, 111.4, 83.5, 83.1, 82.8, 81.4,
81.3, 81.2, 39.3 (d, 2J(P,C)=22.0 Hz), 28.0, 25.6, 21.8, 15.7, 15.6 ppm;
31P{1H} NMR (121.5 MHz, CDCl3): d=141.5 ppm; HRMS (FAB+): m/z
calcd for C43H55NO4P: 680.3869; found: 680.3856 [M+H]+; elemental
analysis calcd (%) for C43H54NO4P: C 75.96, H 8.01, N 2.06; found: C
75.86, H 8.22, N 2.20.
resonances probably due to P C coupling.) 35.5 (d, 2J(P,C)=19.5 Hz),
À
27.9, 25.7, 21.9, 21.8 ppm; 31P{1H} NMR (121.5 MHz, CDCl3): d=
139.7 ppm; HRMS: m/z calcd for C41H51NO4P: 652.3556; found: 652.3557
[MÀH]+; elemental analysis calcd (%) for C41H50NO4P: C 75.55, H 7.73,
N 2.15; found: C 75.17, H 7.84, N 2.19.
(1S,7S)-4-(N-Methyl-N-(R)-1-phenylethylamino)-9,9-dimethyl-2,2,6,6-
tetra(3,5-dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]de-
cane ((S,S,R)-2c): Prepared analogously to the synthesis of (R,R)-2b
from
(1S,7S)-4-chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
(1S,7S)-4-Dimethylamino-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((S,S)-2a): This com-
pound was prepared as described above for (R,R)-2a from 2S,3S-O-iso-
propylidene-1,1,4,4-tetra(3,5-dimethylphenyl)threitol (500 mg, 0.86 mmol)
and HMPT(188 mL, 1.03 mmol). Yield: 310 mg (55%); 1H NMR
(300 MHz, CDCl3): d=7.44 (s, 2H; Ar-H), 7.24 (s, 2H; Ar-H), 7.11 (s,
4H; Ar-H), 6.91 (s, 3H partial overlap; Ar-H), 6.88 (s, 1H partial over-
lap; Ar-H) 5.12 (dd, J=3.3, 8.2 Hz, 1H; CH), 4.81 (d, J=8.2 Hz, 1H;
CH), 2.78 (d, 3J(P,H)=10.5 Hz, 6H; N(CH3)2), 2.33–2.30 (m, 24H;
ArCH3), 1.40 (s, 3H; OCCH3), 0.33 ppm (s, 3H; OCCH3); 31P{1H} NMR
(121.5 MHz, CDCl3): d=139.7 ppm. HRMS: m/z calcd for C41H51NO4P:
652.3556; found: 652.3533 [M+H]+; elemental analysis calcd (%) for
C41H50NO4P: C 75.55, H 7.73, N 2.15; found: C 75.62, H 7.92, N 2.21.
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane (2.6 mmol) and (R)-(+)-
N,a-dimethylbenzylamine (0.47 mL, 3.2 mmol). The crude product was
purified by extraction with hexanes (2”10 mL), followed by quick
column chromatography (hexanes/triethylamine 95/5 v/v) to give a white
foam as the product. Yield: 510 mg (27%); 1H NMR (300 MHz): d=
7.60–6.80 (m, 17H; Ar-H), 5.14 (dd, J=8.2, 3.2 Hz, 1H; OCH), 4.70 (m,
1H; NCH), 4.65 (d, J=8.2 Hz, 1H; OCH), 2.61 (d, J=6.0 Hz, 3H;
NCH3), 2.34–2.18 (m, 24H; ArCH3), 1.53 (d, J=6.9 Hz, 3H; NCHCH3),
1.47 (s, 3H; OCCH3), 0.50 (s, 3H; OCCH3); 13C{1H} (75.5 MHz, C6D6)
NMR: d=148.7, 148.2, 148.1, 144.0, 144.0, 143.6, 143.1, 138.0, 137.6,
137.5, 137.0, 129.9, 129.6, 129.5, 128.9, 127.8, 127.7, 127.3, 126.2, 126.1,
112.1, 84.8, 83.7, 83.4, 82.6, 82.5, 82.5, 56.6 (d, 2J(P,C)=35.4 Hz), 29.8,
28.5, 27.6, 26.3, 26.0, 22.1, 22.0 (m), 19.5, 19.4, 12.0 ppm; 31P{1H} NMR
(121 MHz): d=140.9 ppm; HRMS (FAB+): m/z calcd for C48H57NO4P:
742.4025; found 742.3993 [MÀH]+; elemental analysis calcd (%) for
C48H56NO4P: C 77.70, H 7.61, N 1.89; found: C 77.59, H 7.52, N 1.81.
(1R,7R)-4-Chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane:
2R,3R-O-isopropyl-
idene-1,1,4,4-tetra(3,5-dimethylphenyl)threitol (3.50 g, 6.0 mmol) was
dried by dissolving the compound in toluene (15 mL) and evaporating
thoroughly to dryness (3 times). The diol was dissolved in toluene
(60 mL) together with triethylamine (1.3 mL, 9 mmol) and cooled to
À408C. Subsequently, a solution of phosphorus trichloride (0.66 mL,
7.5 mmol) and triethylamine (1.3 mL, 9 mmol) in toluene (20 mL) was
added dropwise. After complete addition, the cooling bath was removed
and the cloudy mixture slowly warmed to room temperature and stirred
overnight. Next, the suspension was refluxed for 2 h and filtered under
strict nitrogen atmosphere to remove the salts (Et3N·HCl) formed. The
resulting solution was evaporated in vacuo to yield a slightly yellow mois-
ture-sensitive powder as the product. This was not purified further, but
used immediately in the following reactions.
(1R,7R)-4-(N-Methyl-N-(R)-1-phenylethylamino)-9,9-dimethyl-2,2,6,6-
tetra(3,5-dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]de-
cane ((R,R,R)-2c): This compound was prepared similarly to the synthe-
sis of (S,S,R)-2c from (1R,7R)-4-Chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-di-
methylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane.
Yield:
42%; 1H NMR (300 MHz, CDCl3): d=7.48–7.04 (m, 13H; Ar-H), 6.90–
6.69 (m, 4H; Ar-H), 5.14 (dd, J=8.3, 3.1 Hz, 1H; OCH), 4.85–4.72 (m,
1H; NCH(Ph)(Me)), 4.69 (d, J=8.3 Hz, 1H; OCH), 2.57 (d, 3J(P,H)=
6.9 Hz, 3H; NCH3), 2.32–2.23 (m, 24H; Ar-CH3), 1.58 (d, J=7.2 Hz, 3H;
NCHCH3), 1.42 (s, 3H; OCCH3), 0.28 ppm (s, 3H; OCCH3); 13C{1H}
NMR (75.5 MHz, CDCl3): d=147.5, 147.2, 143.2, 143.2, 142.6, 142.1,
137.4, 136.9, 136.8, 136.4, 129.3, 129.0, 128.9, 128.9, 128.3, 127.7, 127.2,
126.8, 125.3, 111.5, 83.4, 82.9, 82.7, 81.6, 81.5, 81.4, 55.3 (d, 2J(P,C)=
35.4 Hz), 28.0, 26.5, 26.4, 25.7, 21.9, 21.8, 18.4, 18.4, 12.0 ppm; 31P{1H}
NMR (121.5 MHz, CDCl3): d=140.1 ppm; HRMS (FAB+): m/z calcd for
C48H57NO4P: 742.4025; found: 742.4025 [MÀH]+; elemental analysis
calcd (%) for C48H56NO4P: C 77.70, H 7.61, N 1.89; found: C 77.02, H
7.87, N 1.95.
(1R,7R)-4-Diethylamino-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-2b): A solution
of diethylamine (0.54 mL, 5.3 mmol) and triethylamine (0.73 mL,
5.3 mmol) in toluene (20 mL) was added dropwise to a solution of
(1R,7R)-4-chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-3,5,8,10-
tetraoxa-4-phosphabicyclo[5.3.0]decane (2.3 g, 3.5 mmol) in toluene
(75 mL) at 08C. After complete addition, the mixture was allowed to
warm to room temperature and stirred for 2 h and subsequently stirred
at 808C during 4 h. The resulting suspension was filtered under nitrogen
and evaporated in vacuo. The resulting off-white powder was purified by
column chromatography (hexanes/triethylamine 95/5 v/v). Yield: 2.10 g
(88%) of white semi-crystalline powder; 1H NMR (300 MHz, CDCl3):
d=7.51 (s, 2H; Ar-H), 7.29 (s, 2H; Ar-H), 7.15 (s, 4H; Ar-H), 6.92–6.88
(m, 4H; Ar-H), 5.15 (dd, J=8.3, 3.5 Hz, 1H; OCH), 4.75 (d, J=8.3 Hz,
1H; OCH), 3.33 (dq, J=11.0 (3J(P,H)), 7.0 Hz, 4H; N(CH2CH3)2), 2.35–
2.33 (m, 24H; ArCH3), 1.50 (s, 3H; OCCH3), 1.25 (t, J=7.0 Hz, 6H;
N(CH2CH3)2), 0.33 ppm (s, 3H; OCCH3); 13C{1H} (75.5 MHz, CDCl3)
NMR: d=147.6, 147.2, 142.3, 137.3, 136.9, 136.7, 136.4, 129.2, 129.0,
128.9, 128.8, 127.2, 126.8, 126.7, 125.3, 111.4, 83.5, 83.1, 82.8, 81.4, 81.3,
81.2, 39.5, 39.2, 28.0, 27.2, 25.6, 21.9, 21.9, 15.7 ppm; 31P{1H} (121.5 MHz,
CDCl3) NMR: d=141.5 ppm; HRMS (FAB+): m/z calcd for
C43H55NO4P: 680.3869; found: 680.3856 [MÀH]+; elemental analysis
calcd (%) for C43H54NO4P: C 75.96, H 8.01, N 2.06; found: C 75.83, H
7.92, N 1.98.
(1R,7R)-4-(N-Methyl-N-(S)-1-phenylethylamino)-9,9-dimethyl-2,2,6,6-
tetra(3,5-dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]de-
cane ((R,R,S)-2c): This compound was prepared similarly to the synthe-
sis of (S,S,R)-2c from (1R,7R)-4-chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-di-
methylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane
(0.81 mmol) and (S)-(À)-N,a-dimethylbenzylamine (141 mL; 0.97 mmol).
1
Yield: 268 mg (45%); H NMR (300 MHz, C6D6): d=8.00 (s, 2H; Ar-H),
7.71 (s, 2H; Ar-H), 7.58 (s, 4H; Ar-H), 7.46 (s, 1H; Ar-H), 7.44 (s, 1H;
Ar-H), 7.22–7.07 (m, 4H; Ar-H), 6.77–6.70 (m, 4H; Ar-H), 5.84 (dd, J=
8.4, 3.9 Hz, 1H; OCH), 5.34 (d, 3J(H,H)=8.4 Hz, 1H; OCH), 4.84–4.75
(m, 1H; NCH(Me)(Ph)), 2.81 (d, 3J(P,H)=6.3 Hz, 3H; NCH3), 2.18–2.07
(m, 24H; Ar-CH3), 1.55 (s, 3H; OCCH3), 1.48 (d, 3J(H,H)=6.9 Hz 3H;
NCH(CH3)(Ph)), 0.49 ppm (s, 3H; OCCH3); 13C{1H} NMR (75.5 MHz,
C6D6): d=148.7, 148.1, 148.1, 144.0, 143.9, 143.5, 143.5, 143.1, 138.0,
137.6, 137.4, 137.4, 137.4, 137.0, 129.9, 129.6, 129.5, 128.9, 128.0, 127.8,
127.7, 127.4, 126.2 (partial overlap with residual solvent signal), 112.1,
84.7, 83.7, 83.4, 82.6, 82.5, 82.5, 56.6 (d, 2J(P,C)=35.4 Hz), 35.3 (m), 29.8,
28.5, 27.6 (m), 26.3, 26.0, 22.0 (m), 19.5, 12.0 ppm; 31P{1H} NMR
(121.5 MHz, C6D6): d=140.8 ppm; HRMS (FAB+): m/z calcd for
C48H57NO4P: 742.4025; found: 742.3996 [MÀH]+; elemental analysis
calcd (%) for C48H56NO4P: C 77.70, H 7.61, N 1.89; found: C 77.68, H
7.54, N 1.94.
(1S,7S)-4-Diethylamino-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((S,S)-2b): This com-
pound was prepared as described for (R,R)-2b. 1H NMR (300 MHz,
CDCl3): d=7.43 (s, 2H; Ar-H), 7.21 (s, 2H; Ar-H), 7.05 (s, 4H; Ar-H),
6.86–6.82 (m, 4H; Ar-H), 5.08 (dd, J=8.3, 3.5 Hz, 1H; OCH), 4.67 (d,
6242
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
(1S,7S)-4-(N-Methyl-N-(S)-1-phenylethylamino)-9,9-dimethyl-2,2,6,6-
tetra(3,5-dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]de-
cane ((S,S,S)-2c): This compound was prepared similarly to the synthesis
of (S,S,R)-2c from (1S,7S)-4-Chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-di-
methylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane
Ar-CH3), 1.43 (s, 3H; OCCH3), 0.29 ppm (s, 3H; OCCH3); 13C{1H} NMR
(75.5 MHz, C6D6): d=148.2, 147.9, 147.9, 143.1, 143.0, 138.1, 137.7, 137.4,
137.0, 130.0, 129.8, 129.7, 129.5, 127.8, 127.8, 126.0, 112.7, 83.9, 83.7, 83.6,
83.3, 82.5, 82.4, 68.2 (d, J=4.9 Hz), 45.0 (d, J=17.1 Hz), 35.3, 29.7, 28.3,
27.6, 26.5, 22.0, 21.8, 21.2, 19.3, 12.0 ppm; 31P{1H} NMR (121.5 MHz,
C6D6): d=138.8 ppm; HRMS (FAB+): m/z calcd for C43H53NO5P:
694.3661; found: 694.3670 [MÀH]+; elemental analysis calcd (%) for
C43H52NO5P: C 74.43, H 7.55, N 2.02; found: C 74.28, H 7.46, N 1.94.
(0.81 mmol) and (S)-(À)-N,a-dimethylbenzylamine (141 mL, 0.97 mmol).
1
Yield: 261 mg (43%); H NMR (300 MHz, C6D6): d=7.99 (s, 2H; Ar-H),
7.73 (s, 1H; Ar-H), 7.63 (s, 2H; Ar-H), 7.60 (s, 2H; Ar-H), 7.42 (d, J=
7.7 Hz, 2H; Ar-H), 7.22–7.11 (m, 4H; Ar-H), 6.79–6.70 (m, 4H; Ar-H),
5.83 (dd, J=8.3, 3.8 Hz, 1H; OCH), 5.44 (d, 3J(H,H)=8.3 Hz, 1H;
OCH), 4.91–4.83 (m, 1H; NCH), 2.77 (d, 3J(P,C)=7.1 Hz, 3H; NCH3),
2.18–2.07 (m, 24H; Ar-CH3), 1.53 (d, 3J(H,H)=7.1 Hz, 3H;
NCH(CH3)(Ph)), 1.52 (s, 3H; OCCH3), 0.55 ppm (s, 3H; OCCH3);
13C{1H} NMR (75.5 MHz, C6D6): d=148.6, 148.1, 143.7, 143.6, 143.5,
143.1, 138.1, 137.6, 137.4, 137.0, 130.0, 129.8, 129.6, 129.4, 127.9, 127.3,
126.1, 112.3, 84.3, 83.8, 83.5, 83.1, 82.5, 82.4, 56.0 (d, 2J(P,C)=34.2 Hz),
35.3, 29.8, 28.4, 27.6, 27.6, 27.3, 27.2, 26.4, 26.0, 23.1, 22.0 (m), 19.0,
12.0 ppm; 31P{1H} NMR (121.5 MHz, C6D6): d=142.0 ppm; HRMS
(FAB+): m/z calcd for C48H57NO4P: 742.4025; found: 742.4016 [MÀH]+;
elemental analysis calcd (%) for C48H56NO4P: C 77.70, H 7.61, N 1.89;
found: C 77.83, H 7.68, N 1.84.
(1R,7R)-4-N-(1-Aza-4-(N’-methyl)azacyclohexyl)-9,9-dimethyl-2,2,6,6-
tetra(3,5-dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]de-
cane ((R,R)-2g): This compound was prepared analogously to the syn-
thesis of (R,R)-2b from (1R,7R)-4-chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-
dimethylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane
(1.30 mmol) and 1-methylpiperazine (173 mL, 156 mg, 1.56 mmol). The
crude product was purified by quick column chromatography (hexanes/
triethylamine 95/5 v/v) to give a white foam as the product. Yield:
510 mg (27%); 13C{1H} NMR (75.5 MHz, C6D6): d=148.4, 148.0, 148.0,
143.3, 143.2, 138.0, 137.6, 137.3, 129.9, 129.8, 129.7, 129.4, 127.8, 127.8,
126.1, 112.6, 84.0, 84.1, 83.7, 83.5, 83.4, 82.5, 82.4, 56.7, 47.1 (d, J=
7.3 Hz), 44.8 (d, J=19.5 Hz), 35.3, 29.8, 28.3, 27.6, 26.5, 26.0, 21.9 (m),
12.0 ppm; HRMS (FAB+): m/z calcd for C44H56N2O4P: 707.3978; found:
707.3986 [MÀH]+; elemental analysis calcd (%) for C44H55N2O4P: C
74.76, H 7.84, N 3.96; found: C 74.83, H 8.04, N 3.66.
(1R,7R)-4-(Diisopropylamino)-9,9-dimethyl-2,2,6,6-tetraphenyl-3,5,8,10-
tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-2e): This compound was
prepared analogously to (R,R)-2b from (2R,3R)-O-isopropylidene-
1,1,4,4-tetraphenylthreitol. Its observed spectroscopic characteristics
matched previously reported data.[45]
(R)-O,O’-(1,1’)-Dinaphthyl-2,2’-diyl-3,3’-di(trimethylsilyl)phosphorchlori-
dite: (R)-3,3’-trimethylsilyl-1,1’-binaphtyl-2,2’-diol (1.00 g, 2.31 mmol) was
azeotropically dried by dissolving it in anhydrous toluene (10 mL) fol-
lowed by evaporation in vacuo of the solvent (2 times). The dry starting
material was then dissolved in toluene (25 mL). PCl3 (242 mL, 381 mg,
2.77 mmol) and triethylamine (1.3 mL, 0.93 g, 9.2 mmol) were dissolved
in toluene (50 mL) and were added dropwise to the diol solution at 08C.
After stirring overnight, the resulting suspension was stirred at 508C for
1 h. After cooling to room temperature, the suspension was filtered and
the solvent evaporated to give a yellow, air-sensitive powder (31P{1H}
NMR: d=176 ppm). This compound was not characterised further, but
used immediately in further reactions.
(1R,7R)-4-Diethylamino-9,9-cyclohexyl-2,2,6,6-tetra(3,5-dimethylphen-
yl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-2i): This com-
pound was prepared analogously to (R,R)-2b from the corresponding
TADDOL backbone. 1H NMR (300 MHz, CDCl3): d=7.40 (s, 2H; Ar-
H), 7.23 (s, 2H; Ar-H), 7.15 (s, 2H; Ar-H), 7.04 (s, 2H; Ar-H), 6.84–6.80
(m, 4H; Ar-H), 4.97 (dd, J=8.5, 3.5 Hz, 1H; OCH), 4.68 (d, 1H; J=
8.5 Hz, OCH), 3.22 (m, 4H; N(CH2CH3)2), 2.28–2.25 (m, 24H; ArCH3),
1.56–1.49 (m, 4H; OCCH2), 1.28–1.12 (m, 4H; CH2), 1.15 (t, 6H; J=
6.9 Hz, N(CH2CH3)2), 0.55–0.45 (m, 1H; CH), 0.35–0.24 ppm (m, 1H;
CH); 13C{1H} (75.5 MHz, CDCl3) NMR: d=147.6, 147.4, 142.5, 142.1,
137.2, 136.8, 136.5, 136.2, 129.1, 128.8, 128.7, 128.6, 127.1, 126.9, 126.8,
125.4, 125.2, 112.1, 82.7, 82.4, 81.9, 81.5, 81.4, 41.9, 39.1 (d, J=22.0 Hz),
37.5, 35.6, 25.4, 24.6, 24.3, 21.9, 15.6, 15.5, 11.4 ppm; 31P{1H} NMR
(121.5 MHz, CDCl3): d=141.4 ppm; HRMS (FAB+): m/z calcd for
C46H59NO4P: m/z 720.4182; found: 720.4182 [MÀH]+; elemental analysis
calcd (%) for C46H58NO4P: C 76.74, H 8.12, N 1.95; found: C 76.65, H
8.06, N 1.88.
O,O’-(1,1’)-Dinaphthyl-2,2’-diyl-3,3’-di(trimethylsilyl)-N,N-diethylphos-
phoramidite
((R)-3a):
(R)-O,O’-(1,1’)-Dinaphthyl-2,2’-diyl-3,3’-
di(trimethylsilyl)phosphorchloridite (546 mg, 1.10 mmol) was dissolved in
toluene (25 mL) and cooled to 08C; then triethylamine (0.23 mL,
1.65 mmol) was added. Next,
a solution of diethylamine (0.17 mL,
1.65 mmol) in toluene (20 mL) was added dropwise. After complete addi-
tion, the solution was warmed to room temperature and stirred for 2 h.
Subsequently the cloudy mixture was stirred overnight at 758C. After
cooling to room temperature the resulting suspension was filtered and all
volatiles evaporated in vacuo. The crude product was washed with hex-
anes (2”15 mL) and purified by column chromatography (SiO2, eluens
EtOAc/Et3N/PE=5:5:90) to afford the product as a white powder. Yield:
110 mg (19%). Precipitation from the hexane washing fractions yielded
another 100 mg pure product. 1H NMR (300 MHz, CDCl3): d=8.04 (d,
2H; J=12.3 Hz, Ar-H), 7.89 (d, J=8.1 Hz, 2H; Ar-H), 7.39–7.05 (m, 6H;
Ar-H), 2.94 (br, 4H; NCH2CH3), 1.02 (br, 6H; NCH2C H3), 0.47 (s, 9H;
Si(CH3)3), 0.41 ppm (s, 9H; Si(CH3)3); 13C{1H} (75.5 MHz, CDCl3): d=
154.1, 154.0, 137.0, 136.8, 134.3, 134.1, 132.9, 132.5, 131.0, 130.2, 128.5,
128.4, 127.1, 127.0, 126.4, 126.3, 124.6, 124.3, 122.9, 122.8, 121.3, 40.6 (m),
29.9, 15.7, 0.3, 0.2 ppm; 31P{1H} (121.5 MHz, CDCl3): d=150.2 ppm;
HRMS: m/z calcd for C30H38NO2PSi2: 532.2257; found: 532.2274
[MÀH]+; elemental analysis calcd (%) for C30H38NO2PSi2: C 67.76,
H 7.20, N 2.63; found: C 67.47, H 7.42, N 2.67.
(1R,7R)-4-Diethylamino-9,9-dimethyl-2,2,6,6-tetra(2-naphthyl)-3,5,8,10-
tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-2 f): The compound was
prepared analogously to the 3,5-(CH3)C6H3-substituted ligand ((R,R)-
2b). Purification was done by column chromatography (SiO2) twice using
hexanes/Et3N/CH2Cl2 (92.5/5.0/2.5 v/v) to yield the compound as a white
solid (49% yield). 1H NMR (300 MHz, CDCl3): d=8.60 (s, 1H; Ar-H),
8.21 (s, 1H; Ar-H), 8.12 (s, 1H; Ar-H), 8.05 (s, 1H; Ar-H), 7.91–7.41 (m,
24H; Ar-H), 5.53 (dd, J=8.5, 3.6 Hz, 1H; OCH), 5.12 (d, J=8.5 Hz, 1H;
OCH), 3.37–3.28 (m, 4H; N(CH2CH3), 1.41 (s, 3H; OCCH3), 1.21 (t, J=
6.9 Hz, 6H; N(CH2CH3)), 0.28 ppm (s, 3H; OCCH3); 13C{1H} NMR
(75.5 MHz, CD2Cl2): d=145.1, 144.3, 144.3, 140.2, 139.8, 133.3–133.0
(overlapping), 129.0–126.5 (overlapping), 112.5, 83.6, 82.4, 82.1, 82.0,
46.8, 44.5, 39.8, 39.5, 28.1, 25.9, 15.8 ppm; 31P{1H} NMR (121.5 MHz): d=
141.9 ppm; HRMS (FAB+): m/z calcd for C51H47NO4P: 768.3243; found:
768.3235 [MÀH]+; elemental analysis calcd (%) for C51H46NO4P: C
79.77, H 6.04, N 1.82; found: C 79.85, H 6.12, N 1.78.
O,O’-(1,1’-Dinaphthyl-2,2’-diyl-3,3’-di(trimethylsilyl))-N-methyl-N-(R)-1-
phenylethylphosphoramidite ((R)-3b): This compound was prepared
analogously to (R)-3a from (R)-O,O’-(1,1’)-dinaphthyl-2,2’-diyl-3,3’-di-
(trimethylsilyl)phosphorchloridite (546 mg, 1.10 mmol) and (R)-(+)-N,a-
dimethylbenzylamine (0.22 g, 1.65 mmol). Purified by column chromatog-
raphy (SiO2, eluents EtOAc/Et3N/PE=2.5:5:92.5). Yield: 251 mg (38%)
of a white powder. 1H NMR (300 MHz, CDCl3): d=8.06 (d, 2H; J=
4.8 Hz, Ar-H), 7.88 (dd, J=8.3, 1.1 Hz, 2H; Ar-H), 7.46–7.01 (m, 11H;
Ar-H), 4.90 (m, 1H; NCH(Me)(Ph)), 2.13 (d, J=4.3 Hz, 3H; NCH3),
1.63 (d, J=7.0 Hz, 3H; NCH(CH3)), 0.44 (s, 9H; Si(CH3)3), 0.43 ppm (s,
9H; Si(CH3)3); 13C{1H} (75.5 MHz, CDCl3): d=153.5 (d, J=1.5 Hz),
142.0, 136.7, 16.6, 133.9, 133.7, 132.3, 131.9, 130.5, 129.8, 128.0, 127.8,
127.4, 126.8, 126.7, 126.5, 126.0, 124.2, 124.1, 122.5, 122.5, 121.1, 121.1,
(1R,7R)-4-N-(1-Aza-4-oxacyclohexyl)-9,9-dimethyl-2,2,6,6-tetra(3,5-di-
methylphenyl)-3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane ((R,R)-
2h): This compound was prepared analogously to the synthesis of (R,R)-
2b from (1R,7R)-4-chloro-9,9-dimethyl-2,2,6,6-tetra(3,5-dimethylphenyl)-
3,5,8,10-tetraoxa-4-phosphabicyclo[5.3.0]decane (2.6 mmol) and morpho-
line (164 mL, 164 mg, 1.88 mmol). The crude product was purified by
quick column chromatography (hexanes/triethylamine 95/5 v/v) to give a
white foam as the product. Yield: 510 mg (27%); 1H NMR (300 MHz,
CDCl3): d=7.37 (s, 2H; Ar-H), 7.20 (s, 2H; Ar-H), 7.03 (s, 2H; Ar-H),
6.99 (s, 2H; Ar-H), 6.87–6.83 (m, 4H; Ar-H), 5.07 (dd, J=8.4, 3.6 Hz,
1H; OCH), 4.73 (d, 1H; 8.3 Hz), 3.74–3.68 (m, 4H; N(CH2CH2)2O),
3.35–3.33 (m, 2H; CH2O), 3.21–3.18 (m, 2H; CH2O), 2.28–2.26 (m, 24H;
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6243

P. W. N. M. van Leeuwen et al.
FULL PAPER
56.1, 55.5, 45.3, 29.5, 27.5, 19.1, 19.0, 0.3, À0.2, À0.2 ppm; 31P{1H}
(121.5 MHz, CDCl3): d=145.6 ppm; HRMS: m/z calcd for
C35H40NO2PSi2: 594.2414; found: 594.2415 [MÀH]+; elemental analysis
calcd (%) for C35H40NO2PSi2: C 70.79, H 6.79, N 2.36; found: C 70.49, H
7.04, N 2.45.
0.36 mmol). Dichloromethane (10 mL) was added and the resulting mix-
ture was stirred for 30 min at room temperature. Evaporation of the sol-
vent gave the product complex as a yellow powder in quantitative yield.
1H NMR (08C, CD2Cl2): d=7.30 (brs, 5H; (cinnamyl)Ar-H), 6.96 (brs,
2H; (L)Ar-H), 6.80 (brs, 2H; (L)Ar-H), 5.50 (apparent double t, J=
12.6, 9.6 Hz, 1H; (cinnamyl)-Hmeso), 4.91 (apparent t, J=14.1 Hz, 1H;
(cinnamyl)-H3), 4.25 (heptet, J=6.5 Hz, 2H; N(CHMe2)2), 3.80 (s, 6H;
OCH3), 3.24 (brd, J=6.5 Hz, 1H; (cinnamyl)-Hsyn), 2.16 (brd, 1H; (cin-
namyl)-Hanti), 1.56 (s, 9H; C(CH3)3), 1.48 (s, 9H; C(CH3)3), 1.28 ppm (d,
J=6.5 Hz, 6H; N(CH(CH3)2)2); 13C{1H} NMR: d=153.74, 142.34, 140.33,
134.51 (d, J=9.7 Hz), 130.37, 127.10, 126.71, 126.54, 126.49, 113.19,
111.82, 109.47 (d, J=9.0 Hz), 54.13, 53.91, 47.69 (d, J=12.8 Hz), 34.30,
30.13, 23.23 ppm; 31P NMR (CD2Cl2): d=141.4 ppm; HRMS (FAB+): m/z
calcd for C37H51NO4PPd: 710.2590; found: 710.2601 [MÀCl]+; elemental
analysis calcd (%) for C37H51ClNO4PPd: C 59.52, H 6.88, N 1.88; found:
C 59.61, H 6.82, N 1.93.
3-Benzyl-2-(2,6-di-tert-butylphenoxy)-4-(S)-methyl-5-(R)-phenyl-1,3,2-ox-
azaphospholidine
(4):
N-Benzyl-(1R,2S)-norephedrine
(340 mg,
1.40 mmol) was dissolved in anhydrous toluene (40 mL). Separately, 2,6-
di(tert-butyl)phenoxyphosphordichloridite (433 mg, 1.40 mmol)[37] was
dissolved in anhydrous toluene (40 mL). Both solutions were slowly
dropwise added, simultaneously, to a solution of triethylamine (3 mL) in
toluene (40 mL) at À408C. Care was taken to ensure that the rate of ad-
dition of both reactant-containing solutions was as equal as possible.
After complete addition, the resulting reaction mixture was warmed to
room temperature and stirred for 48 h. The formed ammonium salts were
filtered and the solvent removed in vacuo. This resulted in a yellowish
powder as the crude product. Pure compound was obtained by crystalli-
sation from diethyl ether, followed by washing of the formed crystals
with cold diethyl ether (5 mL) and acetonitrile (2”5 mL). Yield: 365 mg
(55%); 1H NMR (300 MHz, CDCl3): d=7.38–7.34 (m, 12H; Ar-H), 6.99
(t, J=7.7 Hz, 1H; Ar-H), 5.13 (d, J=6.0 Hz, 1H; CH(Ph)O), 4.60 (dd,
J=15.1, 7.9 Hz, 1H; NC(H)(H)(Ph)), 4.09 (dd, J=23.2, 7.9 Hz, 1H;
NC(H)(H)(Ph)), 3.31 (m, 1H; CH(Me)N), 1.61 (s, 18H; C(CH3)3),
1.00 ppm (d, J=6.8 Hz, 3H; CH(CH3)N); 13C NMR (75.5 MHz, CDCl3):
d=152.1 (d, J=11.0 Hz), 143.7, 139.3, 138.2, 128.9, 128.3, 128.1, 128.1,
127.9, 127.6, 126.7, 126.3, 122.7, 86.0 (d, J=12.2 Hz), 55.2, 51.0 (d, J=
30.5 Hz), 35.8, 32.4, 15.6 ppm; 31P NMR (121.5 MHz, CDCl3): d=
154.9 ppm; HRMS: m/z calcd for C30H38NO2P: 476.2718; found: 476.2746
[MÀH]+; elemental analysis calcd (%) for C30H38NO2P: C 75.76, H 8.05,
N 2.95; found: C 75.58, H 8.07, N 3.06.
[Pd(C3H5){(R,R)-2b}(Cl)]: This compound was prepared as described for
[Pd(C3H5)(1)(Cl)]. 13C{1H} NMR (CD2Cl2, mixture of isomers): d=
144.12, 142.18, 142.11, 141.67, 138.38, 138.07, 137.58, 136.34, 129.73,
128.92, 128.80, 127.46, 127.05, 126.87, 126.37, 125.63, 125.10, 118.06,
117.87, 114.92, 89.33, 89.16, 86.61, 79.87, 79.23, 78.99, 78.52, 65.89, 62.82,
57.16, 55.12, 41.06, 27.16, 26.89, 26.67, 26.54, 21.46, 21.39, 21.22, 15.33,
14.91 ppm; 31P NMR (MHz, À208C, CD2Cl2): d=119.25 (major isomer),
117.32 ppm (minor isomer); HRMS (FAB+): m/z calcd for
C46H59NO4PPd: 826.3216; found: 826.3198 [MÀCl]+; elemental analysis
calcd (%) for C46H59ClNO4PPd: C 64.04, H 6.89, N 1.62; found: C 64.10,
H 6.82, N 1.58.
[Pd(1,3-diphenylallyl){(R,R)-2b}(Cl)]:
A Schlenk vessel was charged
with [Pd(1,3-diphenylallyl)(m-Cl)]2 (91 mg, 0.18 mmol) and phosphorami-
dite (R,R)-2b (172 mg, 0.36 mmol). Dichloromethane (10 mL) was added
and the resulting mixture was stirred for 30 min at room temperature.
Evaporation of the solvent gave the product complex as an orange/
yellow microcrystalline powder in quantitative yield. The compound de-
composed slowly in solution. 1H NMR (300 MHz, CDCl3): d=7.60–6.78
(m, 22H; Ar-H), 6.19 (t, J=12.9 Hz, 1H), 5.90 (d, J=7.7 Hz, 1H), 5.12
(d, J=7.5 Hz, 1H), 4.30 (apparent t, J=13.0 Hz, 1H), 2.92–2.82 (m, 3H),
2.50–2.17 (m, 27H), 1.59–1.51 (m, 2H), 0.63 (s, 3H), 0.50 (s, 3H), 0.45–
0.41 ppm (m, 6H); 13C{1H} NMR (CD2Cl2): d=144.62, 142.07, 141.68,
141.62, 139.50, 138.51, 137.42, 136.35, 135.97, 130.72, 129.53, 129.25,
128.77, 128.25, 128.20, 127.97, 127.84, 127.20, 125.73, 114.82, 107.27,
107.18, 94.40, 94.08, 90.06, 89.87, 85.93, 79.24, 78.44, 78.41, 71.39, 71.33,
39.82 (d, J=9.3 Hz), 26.87, 26.54, 21.83, 21.50, 21.39, 21.17, 13.83 ppm;
31P NMR: d=119.4 ppm.
[Pd(C3H5)(1)(Cl)]: A Schlenk vessel was charged with [{Pd(allyl)(m-Cl)}2]
(111 mg, 0.30 mmol) and phosphoramidite 1 (293 mg, 0.60 mmol). Next,
dichloromethane (15 mL) was added and the resulting mixture was stir-
red for 30 min at room temperature. Evaporation of the solvent gave the
product complex as a pale yellow microcrystalline powder in quantitative
yield. Crystals suitable for an X-ray analysis were grown by slow evapo-
ration of a solution of the compound in hexanes. 1H NMR (300 MHz,
CD2Cl2, 08C): d=6.92 (brs, 2H; (L)Ar-H), 6.78 (brd, 2H; (L)Ar-H),
5.18–5.09 (m, 1H; (allyl)-Hmeso), 4.41 (apparent t, J=8.8 Hz, 1H; (allyl)-
H3,syn), 4.24–4.18 (m, 2H; N(CHMe2)2), 3.80 (s, 6H; OCH3), 3.38 (appa-
rent t, J=14.0 Hz, 1H; (allyl)-H3,anti), 3.24 (brd, J=5.7 Hz, 1H; (allyl)-
H1,anti), 1.94 (d, J=12.2 Hz, 1H; (allyl)-H1,syn) 1.47 (s, 9H; C(CH3)3), 1.42
(s, 9H; C(CH3)3), 1.30 ppm (d, J=6.7 Hz, 6H; N(CH(CH3)2)2); 13C{1H}
NMR(75.5 MHz,CD2Cl2, 08C): d=157.47, 146.14, 143.85 (d, J=5.0 Hz),
133.4, 132.96, 119.78 (d, J=9.7 Hz), 116.75, 115.59, 81.44 (d, J=44.9 Hz),
61.49, 57.73, 50.76 (d, J=12.1 Hz), 37.78, 33.32, 26.59 ppm; 31P NMR
(121.5 MHz, CD2Cl2): d=140.1 ppm; HRMS (FAB+): m/z calcd for
C31H47NO4PPd: 634.2277; found: 634.2297 [MÀCl]+; elemental analysis
calcd (%) for C31H47ClNO4PPd: C 55.53, H 7.07, N 2.09; found: C 55.39,
H 7.12, N 2.03.
Allylic alkylation reactions
General procedure: A flame-dried Schlenk vessel was charged with fresh-
ly prepared stock-solutions in THF (1 mL per experiment) containing
[{Pd(crotyl)(m-OAc)}2] (2.0 mg, 0.005 mmol) and ligand
1 (4.9 mg,
0.010 mmol, 2 equiv). Subsequently, the allylic substrate (1.0 mmol) and
dihexyl ether (50 mL), as internal standard for GC measurements, were
added and the reaction mixture was stirred for 15 min at room tempera-
ture. A solution of disodium diethyl 2-methyl malonate (4 mL, 2.0 mmol,
0.5m) was added and the solution was stirred at room temperature. Ali-
quots were taken from the reaction mixture at certain time intervals, di-
luted with diethyl ether, washed with saturated aqueous ammonium chlo-
ride solution, dried over MgSO4 and analysed by GC.
[Pd(crotyl)(1)(Cl)]: A Schlenk vessel was charged with [{Pd(crotyl)(m-
Cl)}2] (37 mg, 0.09 mmol) and phosphoramidite (92 mg, 0.18 mmol).
Next, dichloromethane (5 mL) was added and the resulting mixture was
stirred for 30 min at room temperature. The dichloromethane was re-
moved in vacuo and the resulting solid washed with two portions of di-
ethyl ether (5 mL). After drying under vacuum the product complex was
Asymmetric allylic alkylation reactions
obtained as
a beige microcrystalline powder in quantitative yield.
1H NMR (300 MHz, CD2Cl2): d=6.96 (brs, 2H; Ar-H), 6.78 (brs, 2H;
Ar-H), 4.93 (brq, J=10.4 Hz, 1H; allyl-Hmeso), 4.33–4.24 (m, 2H;
N(CHMe2)2), 4.16–4.05 (m, 1H; allyl-H), 3.80 (s, 6H; OCH3), 3.06 (brs,
1H; allyl-H), 1.86 (brd, J=10.7 Hz, 1H; allyl-H), 1.66–1.32 ppm (m,
33H; L+allyl-CH3); 13C{1H} NMR (CD2Cl2): d=155.83, 144.51 (d, J=
6.1 Hz), 142.22, (d, J=4.9 Hz), 131.9 (br), 131.5 (br), 117.25 (d, J=
8.5 Hz), 115.11, 113.98, 100.04, 99.54, 66.20, 56.10, 49.20 (d, J=13.4 Hz),
36.17, 31.76, 25.01, 17.69 ppm (d, J=7.3 Hz); 31P NMR (CD2Cl2): d=
143.6 ppm; HRMS (FAB+): m/z calcd for C32H49NO4PPd: 648.2434;
found: 648.2430 [MÀCl]+; elemental analysis calcd (%) for
C32H49ClNO4PPd: C 56.15, H 7.21, N 2.05; found: C 56.12, H 7.20, N
2.01.
General procedure: A flame-dried Schlenk vessel was charged with
[{Pd(allyl)(m-OAc)}2] (2.0 mg, 0.005 mmol) and ligand (0.020 mmol,
4 equiv) through freshly prepared stock-solutions in CH2Cl2 (2 mL per
experiment). Subsequently, the allylic substrate (1.0 mmol) and dihexyl
ether (50 mL), as internal standard for GC measurements were added,
followed by dimethyl malonate (171 mL, 1.5 mmol). This mixture was stir-
red at room temperature for 15 minutes. The reaction was started by ad-
dition of N,O-bis(trimethylsilyl) acetamide (371 mL, 1.5 mmol) and
KOAc (1 mg). The reaction was monitored by GC and TLC. After the
desired reaction time, the mixture was diluted with Et2O (10 mL),
washed with saturated ammonium chloride solution (5 mL) and dried
over MgSO4. Evaporation of the solvent gave crude product which was
purified by flash column chromatography (SiO2, eluents: EtOAc/hex-
anes=1:3) to yield a colourless oil. The enantiomeric excess was deter-
[Pd(cinnamyl)(1)(Cl)]: A Schlenk vessel was charged with [{Pd(cinna-
myl)(m-Cl)}2] (91 mg, 0.18 mmol) and phosphoramidite (172 mg,
6244
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246

Phosphoramidites
6232 – 6246
mined by chiral HPLC (Daicel OD, n-hexane/2-propanol=99.5:0.5, flow
0.5 mLminÀ1, tR (R)=35.4, tR (S)=38.7 min, l=254 nm for the product
from 1,3-diphenylallyl acetate).
[12] P. von Matt, A. Pfaltz, Angew. Chem. 1993, 105, 614; Angew. Chem.
Int. Ed. Engl. 1993, 32, 566.
[13] J. Sprinz, M. Kiefer, G. Helmchen, M. Reggelin, G. Huttner, O.
Walter, L. Zsolnai, Tetrahedron Lett. 1994, 35, 1523.
[14] R. Prꢂtꢃt, A. Pfaltz, Angew. Chem. 1998, 110, 337; Angew. Chem.
Int. Ed. 1998, 37, 323.
[15] P. von Matt, G. C. Lloyd-Jones, A. B. F. Minidis, A. Pfaltz, L. Macko,
M. Neuburger, M. Zehnder, H. Ruegger, P. S. Pregosin, Helv. Chim.
Acta 1995, 78, 265.
[16] S. Kudis, G. Helmchen, Angew. Chem. 1998, 110, 3210; Angew.
Chem. Int. Ed. 1998, 37, 3047.
[17] J. M. Brown, D. I. Hulmes, P. J. Guiry, Tetrahedron 1994, 50, 4493.
[18] M. Kollmar, B. Goldfuss, M. Reggelin, F. Rominger, G. Helmchen,
Chem. Eur. J. 2001, 7, 4913.
[19] R. J. van Haaren, C. J. M. Druijven, G. P. F. van Strijdonck, H. Oev-
ering, J. N. H. Reek, P. C. J. Kamer, P. W. N. M. van Leeuwen, J.
Chem. Soc. Dalton Trans. 2000, 1549, and references therein.
[20] M. Widhalm, U. Nettekoven, H. Kalchhauser, K. Mereiter, M. J.
Calhorda, V. Fꢂlix, Organometallics 2002, 21, 315, and references
herein.
Computational details: All calculations were performed on an SG work-
station using the commercially available SPARTAN program (ver-
sion 5.0.3.). The geometry optimisations were carried out on semi-empiri-
cal (PM3(tm)) level after optimisation using molecular mechanics (Sybyl
force field), for the [Pd(allyl)(L)(Cl)] complexes, in which L is the struc-
ture of the ligand without the substituents on the acetal moiety in the
backbone. The product h2-alkene complexes were modelled by using
À
X-ray
CH(COOH)2 as the newly coupled fragment.
crystal
structure
determination
of
[Pd(C3H3)(1)(Cl)]:
C31H47ClNO4PPd, Mr =670.52, pale yellow block, 0.30”0.30”0.15 mm3.
Monoclinic crystal system, space group P21/c (no. 14). Cell parameters:
a=10.9862(1), b=16.6159(1), c=18.1533(2) Š, b=104.4576(3)8, V=
3208.86(5) Š3; Z=4, 1=1.388 gcmÀ3; 41792 reflections were measured
on a Nonius KappaCCD diffractometer with rotating anode and MoKa ra-
diation (graphite monochromator, l=0.71073 Š) at a temperature of
150 K. An absorption correction based on multiple measured reflections
was applied (m=0.75 mmÀ1, correction range 0.80–0.89). The reflections
[21] L. Xiao, W. Weissensteiner, K. Mereiter, M. Widhalm, J. Org. Chem.
2002, 67, 2206.
[72]
were merged using the program SORTAV, resulting in 7358 unique re-
flections (Rint =0.0493), of which 6093 were observed [I>2s(I)]. The
structure was solved with automated Patterson methods using the pro-
gram DIRDIF,[73] and refined with the program SHELXL97[74] against F2
of all reflections up to a resolution of (sin#/l)max =0.65 ŠÀ1. Non-hydro-
gen atoms were refined freely with anisotropic displacement parameters,
hydrogen atoms were refined as rigid groups. The allyl ligand was
rotationally disordered over two conformations and was refined with a
disorder model. There were 374 refined parameters, 0 restraints. R
(obsd reflns): R1=0.0276, wR2=0.0711; R (all data): R1=0.0370, wR2=
0.0762. Weighting scheme w=1/[s2(Fo2)+(0.0367P)2 +1.1877P], where
P=(F2o +2F2c)/3. GoF=1.066. Residual electron density between À0.66
and 0.71 eŠÀ3. The drawings, structure calculations and checking for
higher symmetry was performed with the program PLATON.[75] CCDC-
247629 contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the Cambridge Crystallographic Data Centre,
12 Union Road, Cambridge CB21EZ, UK; fax: (+44)1223-336-033; or
deposit@ccdc.cam.uk).
[22] A. Togni, U. Burckhardt, V. Gramlich, P. S. Pregosin, R. Salzmann,
J. Am. Chem. Soc. 1996, 118, 1031.
[23] see for example: S.-L. You, X.-Z. Zhu, Y.-M. Luo, X.-L. Hou, L.-X.
Dai, J. Am. Chem. Soc. 2001, 123, 7471.
[24] M. Ogasawara, K. Yoshida, T. Hayashi, Organometallics 2001, 20,
3913.
[25] Y. Chen, M. D. Smith, K. D. Shimizu, Tetrahedron Lett. 2001, 42,
7185.
[26] X. Dai, S. Virgil, Tetrahedron Lett. 1999, 40, 1245.
[27] K. Fuji, M. Sakurai, T. Kinoshita, T. Kawabata, Tetrahedron Lett.
1998, 39, 6323.
[28] C.-D. Graf, C. Malan, K. Harms, P. Knochel, J. Org. Chem. 1999, 64,
5581.
[29] Y. Hamada, N. Seto, Y. Takayanagi, T. Nakano, O. Hara, Tetrahe-
dron Lett. 1999, 40, 7791.
[30] T. Hayashi, Acc. Chem. Res. 2000, 33, 354.
[31] H. Kodama, T. Taiji, T. Ohta, I. Furukawa,Tetrahedron: Asymmetry
2000, 11, 4009.
[32] S. G. Nelson, M. A. Hilfiker, Org. Lett. 1999, 1, 1379.
[33] T. Hayashi, J. Organomet. Chem. 1999, 576, 195.
[34] T. Hayashi, M. Kawatsura, Y. Uozomi, Chem. Commun. 1997, 561.
[35] a) M. D. K. Boele, Ph.D. thesis, 2002, Amsterdam. b) G. P. F. van
Strijdonck, M. D. K. Boele, P. C. J. Kamer, J. G. de Vries, P. W. N. M.
van Leeuwen, Eur. J. Inorg. Chem. 1999, 1073.
Acknowledgement
[36] for example, in hydrogenation reactions: H.-U. Blaser, C. Malan, B.
Pugin, F. Spindler, H. Steiner, M. Stud, Adv. Synth. Catal. 2003, 345,
103.
[37] A. van Rooy, D. Burgers, P. C. J. Kamer, P. W. N. M. van Leeuwen,
Recl. Trav. Chim. Pays-Bas 1996, 115, 492.
The research described in this paper was financially supported by DSM
Research (MDKB), the Netherlands Ministry of Economic Affairs, grant
EETK 97107 (GPFvS) and NWO-CW (ML, ALS).
[38] a) B. L. Feringa, Acc. Chem. Res. 2001, 34, 504. b) H. Zhou, W. H.
Wang, Y. Fu, J. H. Xie, W. J. Shi, L. X. Wang, Q. L. Zhou , J. Org.
Chem. 2003, 68, 1582.
[1] B. M. Trost, D. L. van Vranken, Chem. Rev. 1996, 96, 395, and refer-
ences herein.
[2] B. M. Trost, C. Lee, in Catalytic Asymmetric Synthesis, 2nd ed. (Ed.:
I. Ojima), Wiley-VCH, New York, 2000, p. 593.
[3] J. Tsuji, in Palladium Reagents and Catalysts, Wiley, New York,
1995, p. 290.
[4] C. G. Frost, J. Howarth, J. M. J. Williams, Tetrahedron: Asymmetry
1992, 3, 1089.
[5] G. Helmchen, A. Pfaltz, Acc. Chem. Res. 2000, 33, 336.
[6] B. M. Trost, Acc. Chem. Res. 1996, 29, 355.
[39] a) M. van den Berg, A. J. Minnaard, E. P. Schudde, J. van Esch,
A. H. M. de Vries, J. G. de Vries, B. L. Feringa, J. Am. Chem. Soc.
2000, 122, 11539. b) M. van den Berg, A. J. Minnaard, R. M. Haak,
M. Leeman, E. P. Schudde, A. Meetsma, B. L. Feringa, A. H. M.
de Vries, C. E. P. Maljaars, C. E. Willans, D. Hyett, J. A. F. Boogers,
H. J. W. Henderickx, J. G. de Vries, Adv. Synth. catal. 2003, 345, 308.
[40] a) Y. Fu, J.-H. Xie, A.-G. Hu, H. Zhou, L.-X. Wang, Q.-L. Zhou,
Chem. Commun. 2002, 480. b) H. M. Huang, Z. Zheng, H. L. Luo,
C. M. Ba, X. Q. Hu, H. L. Chen, Org. Lett. 2003, 5, 4137.
[41] J. F. Jensen, B. Y. Svendsen, T. V. la Cour, H. L. Pedersen, M. Jo-
hannsen, J. Am. Chem. Soc. 2002, 124, 4558.
[42] G. Francio, F. Faraone, W. Leitner, J. Am. Chem. Soc. 2002, 124,
736.
[43] A. I. Polosukhin, O. G. Bondarev, S. E. Lyubimov, A. V. Korostylev,
K. A. Lyssenko, V. A. Davankov, K. N. Gavrilov, Tetrahedron:
Asymmetry 2001, 12, 2197.
[7] B. M. Trost, M. G. Organ, G. A. OꢀDoherty, J. Am. Chem. Soc. 1995,
117, 9662.
[8] P. E. Blçchl, A. Togni, Organometallics 1996, 15, 4125.
[9] P. Dierkes, S. Ramdeehul, L. Barloy, A. de Cian, J. Fischer, P. C. J.
Kamer, P. W. N. M. van Leeuwen, J. A. Osborn, Angew. Chem. 1998,
110, 3299; Angew. Chem. Int. Ed. 1998, 37, 3116.
[10] B. M. Trost, F. D. Toste,J. Am. Chem. Soc. 1999, 121, 4545.
[11] A. Pfaltz, M. Lautens, in Comprehensive Asymmetric Catalysis,
Vol. 2 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer,
New York, 1999, p. 833.
[44] a) H. C. Volger, K. Vrieze, J. W. F. M. Lemmers, A. P. Praat and
P. W. N. M. van Leeuwen, Inorg. Chim. Acta 1970, 4,435. b) B. Bar-
Chem. Eur. J. 2004, 10, 6232 – 6246
www.chemeurj.org
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6245

P. W. N. M. van Leeuwen et al.
FULL PAPER
tels, G. Helmchen, Chem. Commun. 1999, 741; for examples involv-
ing phosphate ligands: c) R. Takeuchi, M. Kashio, J. Am. Chem.
Soc. 1998, 120, 8647. d) R. Takeuchi, M. Kashio, Angew. Chem.
1997, 109, 268; Angew. Chem. Int. Ed. Engl. 1997, 36, 263.
[45] A. Alexakis, J. Burton, J. Vastra, C. Benhaim, X. Fournioux, A. van
den Heuvel, J.-M. Leveque, F. Maze, S. Rosset, Eur. J. Org. Chem.
2000, 4011.
[60] See, for example, reference [44c,d].
[61] C. Girard, H. B. Kagan, Angew. Chem. 1998, 110, 3088; Angew.
Chem. Int. Ed. 1998, 37, 2922.
[62] D. G. Blackmond, Acc. Chem. Res. 2000, 33, 402.
[63] J. Albert, R. Bosque, J. M. Cadena, S. Delgado, J. Granell, G.
Muller, J. I. Ordinas, M. F. Bardia, X. Solans, Chem. Eur. J. 2002, 8,
2279.
[46] L. A. Arnold, R. Imbos, A. Mandoli, A. H. M. de Vries, R. Naasz,
B. L. Feringa, Tetrahedron 2000, 56, 2865.
[64] F. Robert, F. Delbecq, C. Nguefack, D. Sinou, Eur. J. Inorg. Chem.
2000, 351.
[47] M. P. T. Sjçgren, S. Hansson, B. Škermark, A. Vitagliano, Organo-
metallics 1994, 13, 1963.
[65] P. R. Auburn, P. B. Mackenzie, B. Bosnich, J. Am. Chem. Soc. 1985,
107, 2033.
[48] V. Branchadell, M. Moreno-Manas, F. Pajuelo, R. Pleixats, Organo-
metallics 1999, 18, 4934.
[49] R. J. Van Haaren, H. Oevering, B. B. Coussens, G. P. F. van Strij-
donck, J. N. H. Reek, P. C. J. Kamer, P. W. N. M. van Leeuwen, Eur.
J. Inorg. Chem. 1999, 1237.
[50] B. Goldfuss, U. Kazmaier, Tetrahedron 2000, 56, 6493.
[51] H. Kurosawa, H. Kajimaru, S. Ogoshi, H. Yoneda, K. Miki, N.
Kasai, I. Ikeda, J. Am. Chem. Soc. 1992, 114, 8417.
[52] T. Yamamoto, O. Saito, A. Yamamoto, J. Am. Chem. Soc. 1981, 103,
5600.
[66] A. D. Beck, B. Bastani, D. A. Plattner, W. Petter, D. Seebach, H.
Braunschweiger, P. Gysi, L. La Vecchia, Chimia 1991, 45, 238.
[67] D. Seebach, R. Dahinden, R. E. Marti, A. K. Beck, D. A. Plattner,
F. N. M. Kuhnle, J. Org. Chem. 1995, 60, 1788.
[68] G. J. H. Buisman, L. A. van der Veen, A. Klootwijk, W. G. J. de Lan-
ge. P. C. J. Kamer, P. W. N. M. van Leeuwen, D. Vogt, Organometal-
lics 1997, 16, 2929.
[69] D. G. I. Petra, P. C. J. Kamer, P. W. N. M. van Leeuwen, K. Goubitz,
A. M. van Loon, J. G. de Vries, H. E. Schoemaker, Eur. J. Inorg.
Chem. 1999, 2335.
[53] T. Yamamoto, M. Akimoto, O. Saito, A. Yamamoto, Organometal-
lics 1986, 5, 1559.
[54] S. Hansson, P. A. Norrby, M. P. T. Sjogren, B. Škermark, M. E. Cuc-
ciolito, F. Giordano, A. Vitagliano, Organometallics 1993, 12, 4940.
[55] M. Kawatsura, Y. Uozumi, M. Ogasawara, T. Hayashi, Tetrahedron
2000, 56, 2247.
[56] For an alternative dissociative pathway see: L. Granberg, J. E. Bꢄck-
vall, J. Am. Chem. Soc. 1992, 114, 6858.
[57] R. Kuwano, Y. Ito, J. Am. Chem. Soc. 1999, 121, 3236.
[58] K. Vrieze, P. W. N. M. van Leeuwen, Studies of Dynamic Organome-
tallic Compounds of the Transition Metals by means of Nuclear Mag-
netic Resonance, Vol. 14, Wiley, New York, 1971.
[70] W. D. Dent, R. Long, A. J. Wilkinson, J. Chem. Soc. 1964, 1585.
[71] R. J. van Haaren, Ph.D. thesis, 2002, Amsterdam.
[72] R. H. Blessing, J. Appl. Crystallogr. 1997, 30, 421.
[73] P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, S.
Garcia-Granda, R. O. Gould, J. M. M. Smits, C. Smykalla, The
DIRDIF97 program system, technical report of the Crystallography
Laboratory, University of Nijmegen (The Netherlands), 1997.
[74] G. M. Sheldrick, SHELXL-97, program for crystal structure refine-
ment, University of Gçttingen (Germany), 1997.
[75] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7.
Received: February 16, 2004
Revised: July 13, 2004
Published online: October 29, 2004
[59] K. Vrieze, P. W. N. M. van Leeuwen, Dynamic Nuclear Magnetic
Resonance Spectroscopy, Academic Press, New York, 1975.
6246
ꢁ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 6232 – 6246