New Dihydroxy Bis(Oxazoline) Ligands for the Palladium-Catalyzed Asymmetric Allylic Alkylation: Experimental Investigations of the Origin of the Reversal of the Enantioselectivity

Article abstract of DOI:10.1002/chem.200204649

The origin of the reversal of the enantioselectivity in the palladium-catalyzed allylic alkylation of rac-1,3-diphenyl-2-propenyl acetate with dimethyl malonate anion using chiral dihydroxy bis(oxazoline) "BO" ligands derived from (1S,2S)-(+)-2-amino-1-ph

Full text of DOI:10.1002/chem.200204649

FULL PAPER
New Dihydroxy Bis(Oxazoline) Ligands for the Palladium-Catalyzed
Asymmetric Allylic Alkylation: Experimental Investigations of the Origin of
the Reversal of the Enantioselectivity
Hassan AÔt-Haddou,*[a] Olivier Hoarau,[b] Dimitri Cramailÿre,[b] Frÿdÿric Pezet,[b] Jean-
Claude Daran,[b] and Gilbert G. A. Balavoine[b]
Abstract: The origin of the reversal of
the enantioselectivity in the palladium-
catalyzed allylic alkylation of rac-1,3-
diphenyl-2-propenyl acetate with di-
methyl malonate anion using chiral di-
hydroxy bis(oxazoline) ™BO∫ligands
derived from (1S,2S)-(+)-2-amino-1-
phenyl-1,3-propanediol was investigat-
ed. To determine the structural effects
of the dihydroxy BO ligand on this
unique phenomenon, new homochiral
dihydroxy BO ligands were prepared
from l-threonine and l-serine and
were assessed in the transformation.
The results obtained with these novel
BO ligands, compared with the one ob-
tained by using the dihydroxy BO li-
gands derived from (1S,2S)-(+)-2-
amino-1-phenyl-1,3-propanediol, reveal
that the reversal in the enantioselectivi-
ty observed with the dihydroxy BO
ligand depends on the structure of the
ligand. The effect of different bases
used to generate the dimethyl malo-
nate anion was also examined. The re-
sults are discussed in terms of the inter-
action of one hydroxy group in the in-
termediate p-allyl palladium complex
with the dimethyl malonate anion.
Keywords: allylic alkylation ¥ asym-
metric catalysis ¥ enantioselectivity ¥
N ligands ¥ palladium
Introduction
product with a 90% ee, while the dihydroxy BO 1b led un-
expectedly to the S product with a 92% ee (Scheme 1).
Based on X-ray analysis of the palladium p-allyl complex-
es[2b] we assumed that an interaction of the nucleophile with
one of the hydroxy groups located on the side chain of the
oxazoline ring of ligand 1b could reverse the enantioselec-
tion. However, this reversal of the enantioselectivity could
also be explained by the presence of the o-protecting group.[3]
To ascertain whether this reversal of enantioselectivity is due
to steric hindrance or to the interaction between the OH
group and the nucleophile, we decided to design new BOs
with benzoyl or hydroxy groups on their side chains. Herein,
we describe the synthesis of new BO ligands derived from l-
threonine and l-serine (Scheme 2), with hydroxy or benzoyl
groups on the side chain, and report their behavior in the Pd-
catalyzed enantioselective allylic alkylation of rac-1,3-diphen-
yl-2-propenyl acetate with dimethyl malonate (DMM). The
role of the base was also studied to determine its impact on
the observed shift of the enantioselectivity.
We have recently been interested in the use of a new set of
chiral bis(oxazoline)[1,2] ™BO∫ 1 in the asymmetric transi-
tion-metal-catalyzed reactions. We have reported that a high
level of enantioselectivity (up to 92% ee) was obtained in
the palladium-catalyzed allylic alkylation of rac-1,3-diphen-
yl-2-propenyl acetate with dimethyl malonate anion. We
have shown that the ligand structure has a dramatic impact
on both the sense of the chiral induction and the level of
enantioselectivity. Although ligands 1a and 1b possess iden-
tical chiral scaffolds, they produce opposite enantiomers of
the alkylated product. The dibenzoyl BO 1a gave the R
[a] Dr. H. AÔt-Haddou
Department of Chemistry, University of North Carolina at Chapel
Hill, Chapel Hill, NC 27599 (USA)
and
Zyvex Corporation, 1321 North Plano Road, Richardson, Texas
75081 (USA)
Fax: (972)235-7882
E-mail: hassan@zyvex.com
Results and Discussion
[b] Dr. O. Hoarau, D. Cramailÿre, Dr. F. Pezet, Dr. J.-C. Daran,
Pr. G. G. A. Balavoine
Ligands 2 and 3 (Scheme 2) were synthesized because we
believed that they would impart good information on the
role played by steric hindrance in influencing enantioselec-
Laboratoire de Chimie de Coordination CNRS, UPR 8241, 205 rte de
Narbonne, 31077, Toulouse Cedex 4 (France)
699
Chem. Eur. J. 2004, 10, 699 707
DOI: 10.1002/chem.200204649
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
1 through the monoprotected
aminodiol intermediate from
(1S,2S)-(+)-2-amino-1-phenyl-
1,3-propanediol (Scheme 1)[2a]
and used the same approach to
synthesize (R,R)-BO ligands 2
and (R,S)-BO ligands 3 from l-
threonine and (R)-BO ligands 4
and 5 from l-serine. For the
synthesis of BO 2 and 3, l-
threonine methyl ester was
transformed to monoprotected
aminodiols
(2R,3R)-9
and
(2R,3S)-13 by using two differ-
ent pathways (Scheme 3). Con-
densation of l-threonine with
ethyl benzimidate hydrochlo-
ride led to the (S,R)-oxazoline
7 in excellent yield.[4]
Scheme 1. Synthesis of BO ligand 1 and stereochemistry of the sodium dimethyl malonate (NuÀ) attack on the
p-allyl palladium complex using BO 1a (R=COPh) and BO 1b (R=H).
Treatment of 7 with LiAlH4
in diethyl ether,[5] followed by
selective hydrolysis of the oxazoline ring with HCl in THF[6]
gave the monoprotected aminodiol (R,R)-9 in 81% yield in
two steps and an overall yield of 77%. Compound (2R,3S)-
13 was prepared in four steps with 83% overall yield from
l-threonine. Reaction of l-threonine methyl ester with the
benzoyl chloride yielded the amide 10, which underwent
cyclization in the presence of thionyl chloride to form the
oxazoline product 11 with inversion of configuration at
carbon C3.[7] Treatment of 11 with LiAlH4 in diethyl ether
followed by a selective cleavage of the oxazoline ring under
acidic conditions afforded (2R,3S)-13 in a high yield. Com-
pounds 9 and 13 were then independently used to synthesize
2 and 3, respectively (Scheme 4). The coupling of the mo-
noprotected aminodiols (R,R)-9 and (R,S)-13 with 2,2’-dime-
thylpropane-1,3-dioyl chloride[8] gave the dihydroxy dia-
mides 14a and 15a in high yields (95 and 94%, respective-
ly).[2a] These dihydroxy diamides were transformed to their
corresponding dichloro diamides 14b and 15b by treatment
with SOCl2. Refluxing 14b and 15b in toluene in the pres-
ence of triethylamine produced 2a in 85% yield and 3a in
90% yield, respectively.[9] Saponification of the dibenzoyl
derivatives 2a and 3a with sodium hydroxide in methanol
gave the dihydroxy BOs 2b in 88% and 3b in 95% yield.
Scheme 2. BO ligands 2 and 3 derived from l-threonine and BO 4 and 5
derived from l-serine.
tivity in the palladium-catalyzed allylic alkylation. As can be
seen from Scheme 2, ligands 2 and 3 are diastereomers of
each other and differ from ligands 1 in that they posses
methyl, rather than phenyl substituents at the stereocenter
on the side chain. Thus, the results obtained by the use of
BO ligands 2 and 3 in the palladium-catalyzed allylic alkyla-
tion of rac-1,3-diphenyl-2-propenyl acetate with dimethyl
malonate compared to the BO 1 ones (Scheme 1) will be
very informative on the role of the steric hindrance and the
effect of the relative stereo-
chemistry of the stereogenic
center located on the side
chain. On the other hand, li-
gands 4 and 5 contain no ste-
reocenter on the side chain and
were synthesized so that we
could determine whether a ste-
reocenter at this position was
necessary for the control of the
chiral induction sense.
Synthesis of BO ligands 2 and
3: We had previously reported
Scheme 3. Synthesis of the monoprotected aminodiols 9 and 13: a) Ethyl benzimidate hydrochloride, Et3N,
CH2Cl2, room temperature, 48 h; b) LiAlH4, Et2O, reflux; c) THF, 2m HCl, room temperature; d) benzoyl
the synthesis of (S,S)-BO ligand chloride, Et3N, CH2Cl2; e) thionyl chloride, 08C, 16 h.
700
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 699 707

Dihydroxy Bis(Oxazoline) Ligands for the Pd-Catalyzed Asymmetric Allylic Alkylation
699 707
pane-1,3-dioyl chloride provid-
ed the diamide 19a in 90%
yield, and 19b in 89% yield.
19a and 19b were then convert-
ed to 4 and 5 through separate
synthetic routes. The dihydroxy
diamide 19a was then trans-
formed to the bis(oxazoline) 4a
by the cyclization of its corre-
sponding dichloride diamide
19a (Cl) in toluene in the pres-
ence of triethylamine.[8] BO 4a
was then saponified with
sodium hydroxide in methanol
to give 4b in good yield. The
cyclization of the diamide 19b
under acidic conditions[10b] led
to the bis(oxazoline) 5a, which
Scheme 4. Synthesis of BO (R,R)-2 and (R,S)-3 from the monoprotected aminodiols 9 and 13: a) Et3N,
CH2Cl2; b) thionyl chloride, reflux, 6 h; c) toluene, Et3N, reflux 6 or 16 h; d) 1% sodium hydroxide in metha-
nol, room temperature.
was in turn saponified to give the dihydroxy BO 5b.
BO ligands from l-serine:[10] The key intermediates for the
synthesis of 4 and 5, monoprotected aminodiols 18a and
18b, were obtained in four steps from l-serine methyl ester
(Scheme 5). Condensation of l-serine methyl ester and ethyl
benzimidate in dichloromethane in the presence of triethyla-
mine at room temperature resulted in the formation of 2-
phenyloxazoline 16.[11] This oxazoline ester was then trans-
formed to its corresponding hydroxyl oxazoline, either by
treatment with diisobutylaluminum hydride (DIBAH)[6]
giving 17a in 91% yield, or by reaction with methyl magne-
sium iodide[12] leading to 17b in 78% yield. The selective
cleavage of the oxazoline ring by HCl in THF led to the mo-
noprotected aminodiols 18a and 18b in quasi-quantitative
yields. Condensation of 18a and 18b with 2,2’-dimethyl pro-
Asymmetric allylic alkylation: Allylic alkylation of rac-1,3-
diphenyl-2-propenyl acetate 20 was performed in dichloro-
methane at 368C in the presence of a (p-allyl)-palladium
ligand complex generated in situ from 1 mol% of bis[(p-al-
lyl)palladium chloride] and 4 mol% of the appropriate BO
ligand. The nucleophile was generated from dimethyl malo-
nate (DMM) in the presence of sodium hydride [Eq.
(1)].[13,14] The data collected is summarized in Table 1.
Our attention was focused on comparing the sense of the
chiral induction produced by different BO ligands possess-
ing the same chiral scaffold to extract the structural futures
that affect the regiochemistry of the nucleophilic attack on
the palladium allyl intermediate. We had previously report-
ed that with sodium hydride as the base, (S,S)-BO 1a led to
Scheme 5. Synthesis of BO 4 and 5 from l-serine: a) DIBAH, 17a; b) MeMgI, Et2O, reflux, 17b; c) 2m HCl, room temperature, 18a and 18b; d) Et3N,
ClCOC(Me)2COCl, CH2Cl2, 08C to room temperature, 19a and 19b; e) SOCl2, reflux, 6 h; f) toluene, Et3N, reflux; g) 1% NaOH in methanol, room tem-
perature; h) methanesulfonic acid, CH2Cl2, reflux, see reference [10b].
701
Chem. Eur. J. 2004, 10, 699 707
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

H. AÔt-Haddou et al.
FULL PAPER
Table 1. Palladium-catalyzed enantioselective allylic alkylation of rac-20 using BO 1 5 and sodium hydride
ring in BO 4. The same result
was obtained with the BO li-
gands 5a/5b (Table 1, entries 9
and 10). The levels of the enan-
siolectivity obtained by BO 4
and BO 5 are similar to each
other showing that the increase
of the steric hindrance at the C-
5 position of the oxazoline ring
has no significant impact on the
asymmetric induction.[17] More
importantly, these results ob-
tained with BO 4 and 5 prove
clearly that a specific structure
of the dihydroxy BO ligand is
required to induce a reversal of
the sense of the chiral induc-
tion. Hence, the presence of a
stereogenic center on the side
procedure.[a]
Run
L
t [h]
ee[b] [%]
Config.[c]
Yield[d] [%]
1
2
3
4
5
6
7
8
9
(S,S)-1a
(S,S)-1b
(R,R)-2a
(R,R)-2b
(R,S)-3a
(R,S)-3b
(R)-4a
(R)-4b
(R)-5a
(R)-5b
24
2
90
92
68
63
97
72
80
80
85
77
R
S
S
R
S
S
S
S
S
S
98
98
90
90
95
24
24
16
24
15
24
144
72
90
100
100
80
chain indeed has
a crucial
10
88
impact on the selectivity. We
further investigated the effect
that the base [Eq. (2)] had on
the enantioselectivity and the
reactivity. First, a soft anionic
nucleophile was produced by
using DMM in the presence of
N,O-bis(trimethylsilyl)aceta-
[a] rac-20 (1 equiv), DMM (3 equiv), NaH (3 equiv), [{Pd(C3H5)Cl}2] (1% mol), L=Ligand (4% mol), CH2Cl2,
368C. [b] The ee values were determined by HPLC using a chiral column (Pharmacir 7C, flow rate 0.7
mLminÀ1, n-BuOH/n-hexane 1:9). [c] The absolute stereochemistry of the product was determined by compar-
ison of the optical rotation with the literature values.[15] [d] Yields refer to purified product after column chro-
matography.
(R)-21, whereas (S,S)-BO 1b gave (S)-21, with high degrees
of enantioselectivity in both cases (Table 1, entries 1 and
2).[16] The level of enantioselectivity was lower with (R,R)-
BO 2a and 2b, but we observed parallel behavior concern-
ing the chiral induction sense, since BO ligand 2a gave (S)-
21, whereas BO 2b led to (R)-21 (Table 1, entries 3 and 4).
The difference in the level of induction between (S,S)-BO 1
and (R,R)-BO 2 is presumably to be due to the difference in
the steric environment of BO 1 and BO 2 (Ph versus Me).
In stark contrast, dihydroxy (R,S)-BO 3b, a diastereoisomer
of dihydroxy BO 2b, did not reverse the chiral induction of
21, and produce the same enantiomer of 21 as ligand 3a,
albeit with lower ee (Table 1, entries 4 and 5). The results
obtained with (R,S)-BO 3 point out the huge impact of the
relative stereochemistry of the stereogenic center located on
the side chain on both the sense of the chiral induction and
the regioselectivity of the nucleophilic attack on the palladi-
um p-allyl intermediate. The behavior of BO 2 and BO 3 in
this transformation clearly demonstrates that the reversal of
the enantioselectivity occurs when the two stereogenic cen-
ters (the one on the oxazoline ring and the one on the side
chain) of the dihydroxy BO ligand have the same absolute
configuration. This phenomena is presumably due to a more
pronounced steric interaction between the oxazoline sub-
stituents and one of the phenyl substituents on the allyl
moiety. No reversal of the enantioselectivity occurred upon
employing BO ligands 4a/4b, with the S product being ob-
tained with 80% ee in both cases (Table 1, entries 7 and 8).
The lower enantioselectivities obtained with 4 with respect
to 3 could be explained by the lack of a stereocenter with a
defined stereochemistry on the side chain of the oxazoline
mide (BSA) and a catalytic amount of potassium acetate
(KOAc).[18,19] To our surprise, by switching from sodium di-
methyl malonate (Table 1, entry 2) to BSA/KOAc (Table 2,
entry 2) with ligand 1b, the chirality of the reaction product
is inverted ((R) instead of (S)). The same trend was also ob-
served with BO 2b, but with a lower level of enantiomeric
excess under these conditions (Table 1, entry 4, and Table 2,
entry 4). Additionally, very high levels of enantioselectivity
(up to >99%) were obtained with BO 3a and 3b (Table 2,
entries 5 and 6), although the sense of asymmetric induction
did not reverse. A similar trend was also observed with BO
4 (Table 2, entries 7 and 8).
In our previous study,[2b] we demonstrated that the use of
BO 1b under NaH conditions proceed via a change in the
stereochemical attack of the nucleophile on the palladium
p-allyl complex in direct contrast with BO 1a (Scheme 1).
An interaction between one of the hydroxy groups located
on the side chain of the oxazoline ring was assumed to
govern the regioselectivity of the attack of the nucleophile
on the palladium p-allyl intermediate.[20] Using the BSA
procedure suggests that such an interaction was suppressed
under these reactions conditions. We propose that BSA
works not only as a base, but also as a silylating agent for
the dimethyl malonate salt to produce a ketene silyl acetal
[Eq. (3)].[21] To confirm this hypothesis we decided to use a
pre-formed ketene silyl acetal as a masked nucleophile in
this reaction. Silyl ketene acetals have been successfully
used in the allylic alkylations as alternative nucleophiles to
[22]
the ™harder∫anionic species.
Thus, the use of ketene silyl
acetal of dimethyl malonate gave, under our conditions, the
R product with both BO ligands 1a and 1b in 80 and 78%
702
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 699 707

Dihydroxy Bis(Oxazoline) Ligands for the Pd-Catalyzed Asymmetric Allylic Alkylation
699 707
Table 2. Results of enantioselective allylic alkylation of rac-20 using BO 1 4 and BSA procedure.[a]
Run
Base
L
t[h]
ee[b] [%]
Config.[c]
Yield[d] [%]
1
2
3
4
5
6
7
8
9
KOAc
KOAc
KOAc
KOAc
KOAc
KOAc
KOAc
KOAc
NBu4F
NBu4F
(S,S)-1a
(S,S)-1b
(R,R)-2a
(R,R)-2b
(R,S)-3a
(R,S)-3b
(R)-4a
(R)-4b
(S,S)-1a
(S,S)-1b
72
72
72
72
44
90
91
55
50
98
>99
78
81
65
52
R
R
S
S
S
S
S
S
R
S
75
70
70
81
98
95
75
72
80
75
44
124
124
36
10
36
[a] rac-20 (1 equiv), DMM (2 equiv), BSA (2 equiv), KOAc (3% mol), NBu4F (2 equiv), [{Pd(C3H5)Cl}2] (1% mol), L=Ligand (4% mol), CH2Cl2, 368C.
[b] The ee values were determined by HPLC using a chiral column (Pharmacir 7C, flow rate 0.7 mLminÀ1, nBuOH/n-hexane 1:9). [c] The absolute ster-
eochemistry of the product was determined by comparison of the optical rotation with the literature values.[15] [d] Yields refer to purified product after
column chromatography.
ee confirming that BSA is probably acting as a silylating re-
agent of the nucleophile.
confirm the impact that the interaction of the nucleophile
with one of the hydroxy groups plays on the regioselection
of the nucleophilic attack on the p-allyl intermediate. Thus,
with the sodium dimethyl malonate salt, the selectivity ob-
served with the dihydroxy BO 1b and BO 2b ligands origi-
nates from a hydrogen bond interaction between the hy-
droxy group and the nucleophile. As shown in Scheme 6,
two different pathways are suggested to rationalize the nu-
cleophilic attack of the sodium malonate on the p-allylpalla-
dium intermediate including BO ligand 1b. The attack of
the nucleophile on C3 yielding the S product is likely fa-
vored by the hydrogen bond involving either the hydroxy
group close to C3, which could drags the nucleophile to
attack this carbon on the side opposite the palladium, or by
the H bond of the nucleophile with the hydroxy group close
to C1, which prevents the attack on C1 by increasing the
steric hindrance at this position which in turn will favor the
attack of the nucleophile on C3 from the opposite side to
the palladium. However, at this stage, we have no experi-
mental evidence to favor any of these two pathways.
The stereochemistry of the nucleophilic attack observed
when BO 1b was used with ketene silyl acetal or employing
the BSA/KOAc procedure is may be due to the reaction
mode of the silylated nucleophile with the palladium p-allyl
intermediate. It was proposed that this reaction probably
proceeds by the nucleophilic attack of ketene silyl acetal on
the p-allyl complex through an interaction of p-orbital be-
tween the p-allyl and the carbon-carbon double bond of
ketene silyl acetal.[22c] This reaction pathway prevents any
interaction between the nucleophile and the side chain of
the dihydroxy BO ligand in the p-allyl intermediate. Inter-
estingly, the opposite trend was observed when BSA was
employed with an equimolar amount of tetrabulylammoniun
fluoride (TBAF), in which the fluorine trapped the trimethyl
silyl group (Table 2, entries 7 and 8).[23] These results clearly
In summary, this study provides a clear explanation of the
shift of the enantioselectivity observed in the Pd-catalyzed
allylic alkylation of rac-1,3-diphenyl-2-propenyl acetate with
the sodium dimethyl malonate anion in the presence of BO
ligands 1b and 2b. The comparison of the behavior of the
dihydroxy BO ligands obtained from l-serine and l-threo-
nine and BO 1b revealed the importance of the structure of
Scheme 6. Change of the stereochemistry of the nucleophilic attack caused by the OH -Nu hydrogen bonding.
703
Chem. Eur. J. 2004, 10, 699 707
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

H. AÔt-Haddou et al.
FULL PAPER
layer was dried over MgSO4, filtered and evaporated to dryness. The
product was purified by chromatography on silica gel (ethyl acetate) to
give 8 (14.13 g, 74 mmol) as a white solid. Yield=85%. M.p. 97 988C;
[a]2D0 =+80.7 (c=1.2 in CHCl3); 1H NMR (CDCl3): d=1.55 (d, J=6.3
Hz, 3H; CH3), 3.75 (dd, J=4.0 Hz and 11.4 Hz, 1H; NÀCH), 4.05 (m,
2H; CH2ÀOH), 4.90 (m, 3H; OÀCH3), 7.50 (m, 3H; ArÀH), 8.00 ppm
(m, 2H; ArÀH); 13C NMR (CDCl3): d=20.5, 63.2, 74.8, 78.0, 127.1,
128.0, 128.2, 131.0, 131.2, 164.7 ppm; elemental analysis calcd (%) for
C11H13NO2 (191.23): C 69.09, H 6.85, N 7.32; found: C 68.80, H 6.85, N
7.30.
the ligand in the selectivity of this reaction. We have deter-
mined that the presence of a hydroxyl group on the side
chain of the oxazoline ring is not in itself sufficient to
change the stereochemistry of the nucleophilic attack on the
palladium p-allyl intermediate. A defined structure of the
dihydroxy BO ligand is needed to orient the nucleophilic
attack. Both stereogenic centers (the one on the oxazoline
ring and the one on the side chain bearing the hydroxyl
group) must have the same absolute configuration for true
influence on the sense of the chiral induction. The investiga-
tion of the effect of the base used to generate the nucleo-
phile was very revealing. The reversal of the enantioselectiv-
ity observed using 1b and 2b employing NaH as the base
was suppressed when BSA/KOAc was used as the base
system. We have also confirmed that BSA acts as silylating
agent for the dimethyl malonate anion under the reaction
conditions. These results demonstrate the impact exerted by
the H bond interaction between the hydroxyl group and the
nucleophile on the chiral sense of induction, which confirms
our previously postulated hypothesis concerning the origin
of the shift of the enantioselectivity observed with BO 1b
ligand. This study prompted us to develop a series of BO li-
gands, which are able to impart very high levels of asymmet-
ric induction. We are currently exploring the possibility of
employing these ligands in other asymmetric catalytic pro-
cesses.
(2R,3R)-2-amino-3-benzoyloxybutan-1-ol (9): To a solution of the hy-
droxy oxazoline 8 (11.7 g, 61.4 mmol) in THF (400 mL) a 2m aqueous
solution of HCl (45.0 mL) was added. The mixture was stirred for 24 h at
room temperature. Evaporation of the solvent and drying under reduced
pressure left the monoprotected aminodiol 9 (14.8 g, 60.0 mmol) as a
white solid, which was used without further purification in the next step.
Yield=99%. M.p. 144 1458C; [a]2D0 =À35.7 (c=1.7 in CH3OH); 1H
NMR (CDCl3): d=1.55 (d, J=6.3 Hz, 3H; CH3), 3.60 (m, 1H; CHÀ
NH2), 3.90 (m, 2H; CH2ÀOH), 5.50 (dq, J=6.3 Hz and 7.4 Hz, 1H; CHÀ
CH3), 7.55 (m, 2H; ArÀH), 7.75 (m, 1H; ArÀH), 8.20 ppm (m, 2H; ArÀ
H); 13C NMR (CDCl3): d=16.6, 56.4, 58.7, 68.6, 128.0, 128.9, 129.8, 133.0,
165.5 ppm; MS (CI, NH3): m/z (%): 210 (100) [M + H].
Methyl (2R,3S)-N-2-benzoylamide-3-hydroxybutanoate (10): A solution
of benzoyl chloride (11.2 mL, 96.6 mmol) in CH2Cl2 (100 mL) was added
dropwise to a mixture of l-threonine methyl ester (18.0 g, 106.0 mmol)
and triethylamine (27.1 mL, 212.0 mmol) in CH2Cl2 (200 mL) at 08C. The
reaction was stirred for 12 h at 08C and quenched with water. After
warming to room temperature, the organic layer was separated, dried
over MgSO4, filtered and concentrated in vacuum to give the amide 10
(23.6 g, 99.5 mmol) as a white solid. Yield=95%. M.p. 95 968C; [a]D20
=
+22.6 (c=1.0 in CHCl3); 1H NMR (CDCl3) d=1.30 (d, J=6.4 Hz, 3H;
CHÀCH3), 2.74 (s, 1H; OH),=3.75 (s, 3H; COOCH3), 4.45 (dq, J=6.4
Hz and 4.0 Hz, 1H; CHÀOH), 4.80 (dd, J=3.5 Hz and 6.3 Hz, 1H; CHÀ
NH), 7.05 (d, J=8.7Hz, 1H; NH), 7.45 (m, 3H; ArÀH), 7.80 ppm (m,
2H; ArÀH); 13C NMR (CDCl3): d=19.7, 52.1, 57.9, 67.4, 126.9, 127.9,
129.5, 131.6, 132.7, 133.1, 168.2, 171.0 ppm; elemental analysis calcd (%)
for C12H15NO4 (237.25): C 60.75, H 6.37, N 5.90; found: C 60.80, H 6.05,
N 5.70.
Experimental Section
All reactions were carried out under an inert argon atmosphere. Melting
points were taken on a Stuart scientific apparatus and are uncorrected.
Nuclear magnetic resonance spectra were recorded on Bruker AM-250
(250 MHz) and AC-200 (200 MHz) spectrometers for 1H and 13C, and
chemical shifts are reported in ppm downfield from Me4Si. Optical rota-
tions were measured on Perkin-Elmer 241 MC. DCI mass spectra were
recorded with a quadripolar Nermag R 10-10H instrument. Elemental
analyses were performed by LCC (Laboratoire de Chimie de Coordina-
tion) Microanalytical Service. Column chromatography was performed
using Merck alumina (70 230 mesh ASTM), deactivated with 8% of
water or Merck silica gel (230 400 mesh). High pressure liquid chroma-
tography (HPLC) analyses were performed with an analytical chiral
chromatography column (Pharmacir 7C, 4.6 mmx 250 mm) eluted with
nBuOH/n-hexane (10 : 90); flow rate 0.7 mLminÀ1 using a Waters 600
pump and a Waters 486 detector operating at 254 nm.
(4S,5S)-4-carboxymethoxy-5-methyl-2-phenyloxazoline (11):
A
large
excess of thionyl chloride (56.0 mL) was added to a solution of amide 10
(19.8 g, 83.0 mmol) in CH2Cl2 (200 mL) at 08C. The resulting mixture
was stirred at the same temperature overnight. Excess thionyl chloride
was removed by distillation under reduced pressure. The residue was dis-
solved in CH2Cl2 (100 mL), washed with a 10% Na2CO3 aqueous solution
(100 mL) and water (3î100 mL). The organic layer was dried over
MgSO4, filtered and concentrated under reduced pressure. The crude
mixture was purified by chromatography on silica gel (ethyl acetate/pen-
tane; 3:2) to afford the oxazoline 11 (17.8 g, 81.0 mmol) as a clear oil.
Yield=98%. [a]2D0 =+69.4 (c=8.5 in EtOH). 1H NMR (CDCl3): d=1.35
(d, J=6 Hz, 3H; CHÀCH3), 3.75 (s, 3H; COOCH3), 5.00 (m, 2H; CHÀ
CH), 7.50 (m, 3H; ArÀH), 7.95 (m, 2H; ArÀH); elemental analysis calcd
(%) for C12H13NO3 (219.24): C 65.74, H 5.98, N 6.39; found: C 65.50, H
6.25, N 6.43.
(4S,5R)-4-carbomethoxy-5-methyl-2-phenyloxazoline (7): l-Threonine
methyl ester (33.46 g, 197.0 mmol) and ethyl benzimidate hydrochloride
(36.52 g, 197 mmol) were mixed in CH2Cl2 (200 mL). Triethylamine (27.6
mL, 197.0 mmol) was added and the mixture was stirred for 24 h. The in-
soluble salts were filtered off and the filtrate was washed with an
NaHCO3 saturated aqueous solution, dried over MgSO4 and filtered.
Evaporation of the solvent afforded 7 (39.98 g, 182.0 mmol) as a clear oil.
(4R,5S)-4-hydroxymethyl-5-methyl-2-phenyloxazoline (12): Following
the procedure used for the synthesis of alcohol 8, ester 11 (14.0 g, 64.1
mmol) in of diethyl ether (200 mL) was reacted with LiAlH4 (2.5 g, 64.1
mmol) in diethyl ether (150 mL) to give, after purification by chromatog-
raphy on silica gel (ethyl acetate), 12 (11.0 g, 57.7 mmol) as a white solid.
Yield=90%. M.p. 98 998C; [a]2D0 =+105.5 (c=1.0 in CHCl3); 1H NMR
(CDCl3): d=1.45 (d, J=6.6 Hz, 3H; CHÀCH3), 3.65 (s, 1H; OH), 3.75
(dd, J=5.6 Hz and 6.3 Hz, 2H; CH2ÀOH), 4.30 (ddd, J=5.1 Hz, 4.7 Hz
and 3.9 Hz, 1H; CH-N), 4.90 (dq, J=6.6 Hz and 3.9Hz, 1H; CHÀO),
7.40 (m, 3H; ArÀH), 7.80 ppm (m, 2H; ArÀH); 13C NMR (CDCl3): d=
14.7, 61.1, 69.1, 78.2, 127.3, 128.0, 131.2, 164.4 ppm; MS (CI, NH3): m/z
(%): 192 [100, M + H]; elemental analysis calcd (%) for C11H13NO2
(191.23): C 69.09, H 6.85, N 7.32; found: C 68.85, H 6.60, N 7.10.
1
Yield=92%. [a]2D0 =+96.7 (c=1.6 in CHCl3); H NMR (CDCl3): d=1.50
(d, J=6.3 Hz, 3H; CH3), 3.80 (s, 3H; OÀCH3), 4.50 (d, J=7.4 Hz, 1H;
CHÀCOOCH3), 5.00 (dq, J=7.4 Hz and 6.3 Hz, 1H; CHÀCH3), 7.45 (m,
3H; ArÀH), 8.00 ppm (m, 2H; ArÀH); 13C NMR (CDCl3): d=20.7, 52.3,
74.8, 78.5, 126.6, 126.9, 128.0, 129.2, 130.9, 131.5, 165.2, 171.3 ppm; ele-
mental analysis calcd (%) for C12H13NO3 (219.25): C 65.74, H 5.98, N
6.39; found: C 65.85, H 5.80, N 6.25.
(4R,5R)-4-hydroxymethyl-5-methyl-2-phenyloxazoline (8):A solution of
oxazoline ester 7 (19.7 g, 90 mmol) in diethyl ether (200 mL) was added
dropwise to a cold suspension of LiAlH4 (3.54 g, 90.0 mmol) in diethyl
ether (250 mL). The mixture was stirred for 4 h at 0®C. The reaction was
quenched with water and the insoluble salts were filtered off. The organic
(2R,3S)-2-amino-3-benzoyloxybutan-1-ol (13): Following the procedure
used for the preparation of 9, the oxazoline 12 (11.7 g, 61.4 mmol) was al-
lowed to react with aqueous HCl2m (45.0 mL, 90.0 mmol) in THF (400
mL) to give 13 (14.9g, 60.85 mmol) as a colorless oil. Yield=99%. M.p.
704
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 699 707

Dihydroxy Bis(Oxazoline) Ligands for the Pd-Catalyzed Asymmetric Allylic Alkylation
699 707
125 1278C; [a]2D0 =À45.5 (c=1.25 in CH3OH); 1H NMR (CDCl3): d=
1.55 (d, J=6.4 Hz, 3H; CHÀCH3), 3.70 (m, 1H; CHÀNH2), 3.90 (dd, J=
4.2 Hz and 7.5 Hz, 1H; CH2ÀOH), 4.10 (dd, J=4.2 Hz and 7.2 Hz, 1H;
CH2ÀOH), 5.50 (dq, J=6.4 Hz and 7.5 Hz, 1H; CHÀO), 7.70 (m, 3H;
ArÀH), 8.15 ppm (m, 2H; ArÀH); 13C NMR (CDCl3): d=15.5, 56.6, 58.3,
69.3, 128.7, 129.9, 133.7, 166.1 ppm; elemental analysis calcd (%) for
C11H16NO3Cl (245.70): C 53.77, H 6.56, N 5.70; found: C 53.50, H 6.80, N
5.45.
(R,R)-2,2-Bis-[2(-1-phenylcarboxy-1-ethyl)-1,3-oxazolinyl)]propane (2a):
A solution of bisamide 14b (7.76 g, 14.40 mmol) and dry triethylamine
(10.6 mL, 140.0 mmol) in dry toluene (120 mL) was heated to reflux for
4 h. The solution mixture was allowed to cool to room temperature and
was diluted with ethyl acetate (100 mL). The resulting mixture was
washed with saturated NaHCO3 (2î50 mL) and the aqueous phase was
extracted with ethyl acetate (3î50 mL). The organic layers were com-
bined and washed with brine (100 mL), dried over Na2SO4 and concen-
trated under reduced pressure. Purification of the residue by column
chromatography on deactivated alumina (ethyl acetate/pentane; 3:7) af-
forded the bis(oxazoline) 2a ( 5.80 g, 12.1 mmol) as a colorless viscous
oil. Yield=85%. [a]2D0 =À64.1 (c=1.5 in CHCl3); 1H NMR (CDCl3): d=
1.40 (d, J=6.0 Hz, 6H; CHÀCH3), 1.60 (s, 6H; CÀCH3), 4.30 (m, 4H;
OÀCH2), 4.45 (m, 2H; CHÀN), 5.30 (m, 2H; CHÀOCOPh), 7.50 (m, 6H;
ArÀH), 8.05 ppm (m, 4H; ArÀH); 13C NMR (CDCl3): d=15.2, 24.4, 38.8,
68.7, 68.9, 71.3, 165.8, 170.5 ppm; elemental analysis calcd (%) for
C27H30N2O6 (478.55): C 67.77, H 6.32, N 5.85; found: C 67.60, H 6.20, N
6.05.
(2R,3R)-N,N’-Bis-[(3-phenylcarboxy-3-methyl-1-hydroxypropyl)]-2,2-1,3-
propanamide (14a): Triethylamine (16.2 mL, 115.6 mmol) was added
dropwise to a solution of dimethyl malonic chloride (4.85 g, 29.0 mmol)
in CH2Cl2 (50 mL) at 08C. The mixture was then added to a suspension
of 9 (9.5 g, 30.8 mmol) in CH2Cl2 (150 mL). The reaction mixture was
stirred at room temperature for 3 h. The solution was then concentrated
under reduced pressure. The crude product was triturated in acetone
(200 mL) and the insoluble salts were filtered off and the filtrate was
concentrated. The residue was purified by chromatography on silica gel
(ethyl acetate) to afford the bis-amide 14a (14.18 g, 27.6 mmol) as a col-
orless oil. Yield=95%. [a]2D0 =+40.0 (c=1.0 in CHCl3); 1H NMR
(CDCl3): d=3.0 (d, J=6.2 Hz, 6H; CHÀCH3), 1.35 (s, 6H; CÀCH3), 3.60
(m, 4H; CH2ÀOH), 4.20 (m, 2H; NÀCH), 5.40 (m, 2H; CHÀOCOPh),
6.90 (d, J=9.1 Hz, 1H; NH), 7.40 (m, 6H; ArÀH), 8.00 ppm (m, 4H;
ArÀH); 13C NMR (CDCl3): d=17.3, 23.3, 49.8, 55.4, 61.9, 69.8, 128.3,
129.5, 129.6, 133.1, 166.0, 174.0 ppm; elemental analysis calcd (%) for
C27H34N2O8 (514.58): C 63.02, H 6.66, N 5.44; found: C 63.60, H 6.62, N
5.40.
(2R,3R)-2,2-Bis-[2(1-hydroxy-1-ethyl)-1,3-oxazolinyl)]propane (2b):
A
solution of NaOH (200 mg, 3.09 mmol) in methanol (1 mL) and THF (7
mL) was added to a solution of bis(oxazoline) 2a (500 mg, 1.05 mmol) in
THF (5 mL). After the mixture had been stirred for 2 h at room tempera-
ture, the solvents were removed under reduced pressure. The crude resi-
due was dissolved in CH2Cl2 (10 mL) and washed with water (2î5 mL).
The organic layer was dried over Na2SO4, filtered and concentrated
under reduced pressure. The residue was dissolved in ether (10 mL) and
pentane was added slowly to induce precipitation of the desired product.
The white precipitate was recovered by filtration to yield bis(oxazoline)
2b (248 mg, 0.92 mmol). Yield=88%. M.p. 115 1168C; [a]2D0 =À18.3 (c=
1.5 in CH3OH); 1H NMR (CDCl3): d=1.30 (d, J=6 Hz, 6H; CHÀCH3),
1.50 (s, 6H; CÀCH3), 3.60 (m, 2H; CHÀN), 4.10 (m, 2H; CHÀO), 4.30
ppm (m, 4H; CH2ÀO); 13C NMR (CDCl3): d=19.1, 23.3, 39.1, 69.5, 70.1,
70.6, 171.1 ppm; elemental analysis calcd (%) for C13H22N2O4 (270.33): C
57.76, H 8.20, N 10.36; found: C 57.50, H 8.30, N 10.70.
(2R,3S)-N,N’-Bis-[(3-phenylcarboxy-3-methyl-1-hydroxypropyl)]-2,2-1,3-
propanamide (15a): Following the procedure used for the synthesis of
the bisamide 14a, a solution of triethylamine (16.2 mL, 115.6 mmol) and
dimethyl malonic chloride (4.8 g, 29.0 mmol) in CH2Cl2 was reacted with
a suspension of amino ester 13 (9.5 g, 30.8 mmol) in CH2Cl2 to give, after
work up and purification by chromatography on silica gel (ethyl acetate),
the bisamide 15a (14.0 g, 27.26 mmol) as a white solid. Yield=94%. M.p.
85 868C; [a]2D0 =+3.0 (c=1.0 in CHCl3); 1H NMR (CDCl3): d=1.40 (d,
J=7.5 Hz, 6H; CHÀCH3), 1.55 (s, 6H; CÀCH3), 2.95 (s, 2H; OH), 3.70
(m, 4H; CH2ÀOH), 4.25 (m, 2H; CHÀNH), 5.20 (m, 2H; CHÀOCOPh),
7.10 (d, J=7.5Hz, 2H; NH), 7.40 7.65 (m, 6H; ArÀH), 8.05 ppm (m,
4H; ArÀH); 13C NMR (CDCl3): d=17.1, 23.6, 50.1, 54.8, 60.9, 70.3,
128.4, 129.6, 133.2, 166.3, 174.1 ppm; elemental analysis calcd (%) for
C27H34N2O8 (514.58): C 63.02, H 6.66, N 5.44; found: C 63.50, H 6.40, N
5.40.
(2R,3S)-2,2-Bis-[2(-1-phenylcarboxy-1-ethyl)-1,3-oxazolinyl)]propane
(3a): Following the procedure used for the synthesis of 2a, reaction bis-
amide 15b (7.76 g, 14.4 mmol) with dry triethylamine (10.6 mL, 140.0
mmol) in toluene at reflux gave after purification by chromatography on
alumina (ethyl acetate/pentane; 3:7) the bis(oxazoline) 3a (6.14 g, 12.8
mmol) as a white powder. Yield=90%. M.p. 144 1458C; [a]2D0 =À43.8
1
(c=1.0 in CHCl3); H NMR (CDCl3) d 1.35 (m, 12H; C-CH3; CHÀCH3),
4.20 (m, 6H; CH2ÀO; CHÀN), 5.20 (m, 2H; CHÀOCOPh), 7.25 7.60 (m,
6H; ArÀH), 8.00 ppm (m, 4H; ArÀH); 13C NMR (CDCl3): d=17.2, 24.2,
38.5, 69.0, 69.9, 72.0, 128.2, 129.5, 130.2, 132.9, 165.5, 170.5 ppm; elemen-
tal analysis calcd (%) for C27H30N2O6 (478.55): C 67.77, H 6.32, N 5.85;
found: C 68.00, H 6.30, N 5.80.
(2R,3R)-N,N’-Bis-[(3-phenylcarboxy-3-methyl-1-chloropropyl)]-2,2-1,3-
propanamide (14b): To a suspension of bisamide 14a (9.65 g, 19.2 mmol)
in 1,2-dichloroethane (180 mL), freshly distilled thionyl chloride (13.9
mL, 192 mmol) was added. The resulting mixture was kept at reflux for
3 h. The solvent and the excess of thionyl chloride were removed by dis-
tillation under reduced pressure. The residue was dissolved in CH2Cl2
(30 mL), washed with 2m K2CO3 (20 mL), with water (20 mL), with brine
(20 mL), and dried over sodium sulfate. After filtration, the organic layer
was concentrated under reduced pressure to afford the dichloride bis-
amide 14b (10.3 g, 18.6 mmol) as a white solid. Yield=97%. M.p. 115
1168C; [a]2D0 =+24.2 (c=1.3 in CHCl3); 1H NMR (CDCl3): d=1.30 (d,
J=6.4 Hz, 6H; CHÀCH3), 1.40 (s, 6H; CÀCH3), 3.60 (m, 4H; CH2ÀCl),
4.40 (m, 2H; CHÀNH), 5.50 (dq, J=6.4 Hz and 4.8 Hz, 2H; CHÀ
OCOPh), 7.20 (d, J=8.7 Hz, 2H; NH), 7.40 (m, 4H; Ar-H), 7.60 (m,
2H; ArÀH) , 8.00 ppm (m, 4H; ArÀH); 13C NMR (CDCl3): d=17.3,
23.8, 43.9, 49.4, 54.0, 69.6, 128.4, 129.4, 129.5, 133.3, 165.7, 174.0 ppm; ele-
mental analysis calcd (%) for C27H32N2O6Cl2 (551.47): C 58.81, H 5.85, N
5.08; found: C 58.40, H 5.52, N 5.05.
(2R,3S)-2,2-Bis-[2(1-hydroxy-1-ethyl)-1,3-oxazolinyl)]propane (3b): The
bis(oxazoline) 3a (500 mg, 1.05 mmol) was added in one portion to a solu-
tion of NaOH (10 mg, 0.16 mmol) in methanol (5 mL). The reaction mix-
ture was stirred at room temperature for 2 h and the methanol was re-
moved under reduced pressure. The crude residue was dissolved in
CH2Cl2 (10 mL) and washed with water (2î5 mL). The organic layer was
dried over Na2SO4, filtered and concentrated under reduced pressure.
The residue was dissolved in ether (10 mL) and pentane was added
slowly to induce precipitation of the desired product. The white precipi-
tate was recovered by filtration to yield bis(oxazoline) 3b (270 mg, 0.99
mmol) as white solid. Yield=95%. M.p. 141 1438C; [a]2D0 =À75.0 (c=1.0
in CHCl3); 1H NMR (CDCl3): d=1.05 (d, J=6 Hz, 6H; CHÀCH3), 1.60
(s, 6H; CH3ÀC), 4.10 (m, 4H; CH2ÀO), 4.35 (m, 2H; CHÀN), 4.60 ppm
(m, 2H; CHÀOCOPh); 13C NMR (CDCl3): d=18.6, 23.4, 39.2, 67.8, 68.2,
71.3, 171.4 ppm; elemental analysis calcd (%) for C13H22N2O4 (270.33): C
57.76, H 8.20, N 10.36; found: C 57.29, H 7.90, N 10.10.
(2R,3S)-N,N’-Bis-[(3-phenylcarboxy-3-methyl-1-chloropropyl)]-2,2-1,3-
propanamide (15b): Following the procedure used for the synthesis of bis-
amide 14b, reaction of bisamide 15a (9.65 g, 19.2 mmol) and thionyl
chloride (13.9 mL, 192 mmol) afforded the bis-amide 15b (9.8 g, 18.2
mmol) as a white solid. Yield=95%. This compound was used in the
(S)-4-carbomethoxy-2-phenyloxazoline (16): Triethylamine (10.9 mL,
62.8 mmol) was added slowly (15 minutes) to a solution of ethyl benzimi-
date hydrochloride (11.6 g, 62.8 mmol) in CH2Cl2 (150 mL). The reaction
mixture was stirred at room temperature for 30 min and l-serine methyl
ester hydrochloride (13.5 g, 79.6 mmol) was added by portion. The result-
ing mixture was stirred for 48 h at room temperature. The solution was
concentrated under reduced pressure. The residue was dissolved in ace-
tone (250 mL), the insoluble salts were filtered off, and the filtrate was
concentrated. Recrystallization of the crude material from diethyl ether
1
next step without any further purification. H NMR (CDCl3): d=1.35 (d,
J=6.5 Hz, 6H; CHÀCH3), 1.50 (s, 6H; CÀCH3), 3.75 (m, 4H; CH2ÀCl),
4.50 (ddd, J=4.3 Hz, 4.5 Hz and 7.0 Hz, 2H; CHÀNH), 5.25 (dq, J=6.5
Hz and 7 Hz, 2H; CHÀOCOPh), 7.10 (d, J=9 Hz, 2H; NH), 7.25 7.60
(m, 6H; ArÀH), 8.00 ppm (m, 4H; Ar-H); 13C NMR (CDCl3): d=17.0,
23.8, 44.0, 49.6, 53.1, 70.0, 128.4, 129.5, 133.2, 165.5, 173.2 ppm.
705
Chem. Eur. J. 2004, 10, 699 707
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

H. AÔt-Haddou et al.
FULL PAPER
gave the oxazoline 16 (11.0 g, 53.6 mmol). Yield=86%. 1H NMR
(CDCl3): d=4.60 (dd, J=10.7 Hz and 8.7 Hz, 1H; OCH), 4,70 (dd, J=8.7
Hz and 8.0 Hz, 1H; OCH), 4.93 (dd, J=10.5 Hz and 8.0 Hz, 1H; OCH ),
7.56 7.38 (m, 3H; ArÀH), 8.00 ppm (d, 2H; J=7.2 Hz); 13C NMR
(CDCl3): d=53.0, 68.3, 69.8, 127.3, 129.3, 129.6, 131.5, 167.0, 171.5 ppm.
(4R)-2,2-Bis-[2-(hydroxymethyl-1,3-oxazolinyl)]propane (4b): Bis(oxazo-
line) 4a (0.36 g, 0.83 mmol) was added in one portion to a solution of
NaOH (1% wt) in methanol (10 mL). The reaction mixture was stirred
for 30 min at room temperature, then quenched with water (2 mL) and
concentrated. The residue was then purified by column chromatography
on silica gel (ethyl acetate) to afford the dihydroxy bis(oxazoline) 4b
(0.19 g, 0.81 mmol) as colorless oil. Yield=97%. [a]2D0 =À70.5 (c=1.2 in
CH2Cl2); 1H NMR (CDCl3): d=1.46 (s, 6H; CH3), 3.43 (dd, J=2.8 Hz
and 11.7 Hz, 2H), 3.69 (dd, J=2.8 Hz and 2.8 Hz, 2H), 4.15 ppm (m,
6H); 13C NMR (CDCl3): d=23.5, 39.4, 63.8, 70.0, 70.1, 170.7 ppm; MS
(CI, NH3): m/z (%): 243 (100) [M + H]; elemental analysis calcd (%)
for C11H18N2O4 (242.27): C 54.53, H 7.49, N 11.56; found: C 54.05, H
7.54, N 8.24.
(R)-4-hydroxymethyl-2-phenyloxazoline (17a): A solution of diisobutyl-
aluminum hydride (142 mL, 1m in THF) was added over a period of 2 h
to a cold (08C) solution of the oxazoline 16 (9.70 g, 47.3 mmol) in THF
(200 mL). The resulting mixture was stirred continuously for an addition-
al 2 h. The mixture was quenched with a saturated sodium tartrate aque-
ous solution (300 mL) and stirred vigorously for 4 h at room temperature.
The reaction mixture was extracted with ethyl acetate (4î50 mL). The
combined organic layers were dried over Na2SO4, filtered and concentrat-
ed under reduced pressure. The residue was purified by column chroma-
tography on silica gel (ethyl acetate) to afford the hydroxy oxazoline 17a
(7.65 g, 43.70 mmol). Yield=91%. 1H NMR (CDCl3): d=3.60 3.80 (m,
1H), 3.80 3.93 (m, 2H), 4.47 4.60 (m, 2H), 7.50 (m, 2H), 7.65 (m, 1H),
8.10 ppm (dd, J=8.3 Hz and 1.2 Hz, 2H); 13C NMR (CDCl3): d=55.9,
62.7, 65.9, 132.2, 133.5, 137.2, 170.0 ppm.
(4S)-[(2-hydroxyethyl-2-methyl]-2 phenyloxazoline (17b): A solution of
methyl magnesium iodide (32.5 mL, 3m in diethyl ether) was added drop-
wise by canula to a refluxed and vigorously stirred solution of 16 (8.0 g,
38.0 mmol) in diethyl ether (250 mL). The reaction mixture was stirred at
reflux for further 3 h and then quenched carefully with saturated NH4Cl
aqueous solution (200 mL). The organic layer was separated, and the
aqueous layer was extracted with diethyl ether (2î100 mL). The com-
bined organic layers were dried over MgSO4, filtered, and concentrated
under reduced pressure to afford 17b (6.22 g, 29.6 mmol). Yield=78%.
[a]2D0 =+64.4 (c=1.0 in CH2Cl2); 1H NMR (CDCl3): d=1.17 (s, 3H;
CH3), 1.34 (s, 3H; CH3), 2.02 (s, 1H; OH), 4.20 (dd, J=10.2 Hz and 8
Hz, 1H), 4.34 (dd, J=8.0 Hz and 8.3 Hz, 1H), 4.43 (dd, J=8,3 Hz and
10.2 Hz, 1H), 7.24 7.48 (m, 3H; ArÀH ), 7.94 ppm (m, 2H; ArÀH ); 13C
NMR (CDCl3): d=26.0, 68.7, 71.5, 75.6, 122.8, 128.2, 128.3, 131.4, 164.5
ppm; MS (CI, NH3): m/z (%): 206 (100) [M + H]; elemental analysis
calcd (%) for C12H15NO2 (205.25): C 70.22, H 7.37, N 6.82; found: C
70.64, H 7.02, N 6.25.
(R)-2-amino-3-benzoyloxy-propan-1-ol (18a): Following the procedure
described for the synthesis of 9, the reaction of the oxazoline 17 (12.0 g,
47.2 mmol) in THF (400 mL) and aqueous HCl (45 mL, 2m) gave the
monoprotected aminodiol 18a (9.22 g, 47.20 mmol) as a white solid. yield
>99%. This material was used in the next step without any farther puri-
fication. 1H NMR (CDCl3): d=3.80 3.60 (m, 1H), 3.93 3.80 (m, 2H),
4.60 4.47 (m, 2H), 7.65 (m, 1H), 7.50 (m, 2H), 8.30 ppm (dd, J=8.3 Hz
and 1.2 Hz, 2H); MS (CI, NH3): m/z (%): 195 (100) [M + H].
(2R)-N,N’-Bis-[3-phenylcarboxy-1-hydroxypropyl]-2,2-dimethyl-1,3-pro-
panediamide (19a(OH)): Following the procedure described for the syn-
thesis of the bisamide 14a, triethylamine (7.0 mL, 50.0 mmol), dimethyl
malonic chloride (4.85 g, 29.0 mmol) and amino ester 18a (4.8 g, 25.0
mmol) were reacted to give, after purification by column chromatogra-
phy on silica gel (ethyl acetate/methanol; 95:5), the bisamide 19a(OH)
(S)-4-phenylcarboxy-3-amino-2-methylbutan-1-ol (18b): Following the
procedure described for the synthesis of 13, the oxazoline 17b (4.8 g,
23.4 mmol) was stirred in a mixture of THF (400 mL) and aqueous HCl
(45 mL, 2m) to yield 18b (6.0 g, 23.0 mmol). This product was used in the
next step without further purification. Yield=97%. 1H NMR (CDCl3):
d=1.27 (s, 3H; CH3), 1.36 (s, 3H; CH3), 3.71 (s, 1H; OH), 4.43 (m, 2H;
CH2), 4.54 (m, 1H; CH), 7.34 (dd, J=7.3 Hz and 7.5 Hz, 2H; ArÀHmeta),
7.48 (dd, J=7.5 Hz and 7.3 Hz, 1H; ArÀHpara), 8.12 ppm (d, J=7.5 Hz,
2H; ArÀHortho); 13C NMR (CDCl3): d=23.8, 28.1, 59.7, 62.0, 70.2, 128.4,
130.2, 133.5, 166.1 ppm; MS (CI, NH3): m/z (%): 224 (100) [M + H].
(5.4 g, 11.3 mmol) as a white solid. Yield=90%. M.p. 85 878C; [a]D20
=
+16,6 (c=1.1 in CH2Cl2); 1H NMR (CDCl3): d=1.40 (s, 6H; CH3), 3.34
(s, 2H; OH ), 3.66 (dd, J=12.1 Hz and 4.8 Hz, 2H), 3.75 (dd, J=12.1 Hz
and 3.9 Hz, 2H), 4.34 (m, 2H; CHN), 4.40 (m, 4H; CHOH), 7.00 (d, J=
7.7 Hz, NH), 7.20 7.40 ppm (m, 10H; ArÀH); 13C NMR (CDCl3): d=
23.5, 49.8, 51.0, 61.5, 63.3, 128.4, 129.5, 130.0, 133.7, 166.8 , 174.2 ppm;
MS (CI, NH3): m/z (%): 540 (25) [M + NH4+], 487 (100) [M + H]; ele-
mental analysis calcd (%) for C25H30N2O8 (486.52): C 61.72, H 6.22, N
5.76; found: C 61.85, H 6.02, N 5.74.
(3S)-2,2-Bis-[4-phenylcarboxy-2-methyl-2-hydroxybutyl]-2,2-dimethyl-
1,3-propanediamide (19b): Following the procedure described for the
synthesis of bisamide 14a, the addition of a solution of triethylamine
(7.0 mL, 50.0 mmol) and dimethyl malonic chloride (4.85 g, 29.0 mmol) in
CH2Cl2 to the amino ester 18b (6.5 g, 25.0 mmol) led, after purification
by column chromatography on silica gel (ethyl acetate/methanol; 95:5),
to 19b (6.0 g, 11.0 mmol) as a white solid. Yield=89%. M.p. 75 778C;
[a]2D0 =+24,6 (c=1.1 in CH2Cl2); 1H NMR (CDCl3): d=1.22 (s, 6H;
CH3), 1.28 (s, 6H; CH3), 1.34 (s, 6H; CH3), 2.95 (s, 2H; OH), 4.10 (ddd,
J=9.0 Hz, 5.4 Hz and 6.3 Hz, 2H; CHN), 4.46 (d, J=5.4 Hz, 1H;
CH2O), 4.47 (d, J=6.3 Hz, 1H; CH2O), 7.08 (d, J=9 Hz, 2H; NH ), 7.38
(m, 4H; ArÀH ), 7.41 (m, 2H; ArÀH ), 7.94 ppm (m, 4H; ArÀH); 13C
NMR (CDCl3): d=23.6, 26.5, 27.5, 49.6, 56.6, 63.5, 71.7, 128.3, 129.5,
133.1, 166.5, 173.4 ppm; MS (CI, NH3): m/z (%): 543 (100) [M + H]; ele-
mental analysis calcd (%) for C29H38N2O8 (542.62): C 64.19, H 7.06, N
5.16; found: C 64.57, H 6.84, N 5.52.
(2R)-N,N’-Bis-[3-phenylcarboxy-1-chloropropyl]-2,2-dimethyl-1,3-pro-
panediamide (19a(Cl)): Following the procedure used for the synthesis
of the bisamide 14b, the bisamide 19a(OH) (5.0 g, 10.3 mmol) was al-
lowed to react with thionyl chloride (13.9 mL, 192.0 mmol) to yield the
bisamide 19a(Cl) (5.0 g, 9.6 mmol) as a white solid. The residue was puri-
fied by flash chromatography on silica gel (ethyl acetate). Yield=94%.
[a]2D0 =+6,1 (c=0.5 in CH2Cl2); 1H NMR (CDCl3): d=1.47 (s, 6H; CH3,
3.66 (dd, J=11.2 Hz and 6.0 Hz, 2H; CHCl), 3,72 (dd, J=11.2 Hz and
6.0 Hz, 2H; CHCl), 4.39 (m, 2H; CHN), 4.53 (m, 4H; CH2O), 7.18 (d,
J=6.3 Hz, 2H; NH), 7.48 (m, 4H; ArÀH ), 7.57 (m, 2H; Ar-H ), 7.99
ppm (m, 4H; ArÀH ); 13C NMR (CDCl3): d=23.8, 43.6, 49.5, 49.5, 63.1,
128.5, 129.3, 129.7, 133.4, 166.4, 173.4 ppm; elemental analysis calcd (%)
for C25H28Cl2N2O6 (523.41): C 57.37, H 5.39, N 5.35; found: C 57.43, H
5.68, N 5.92.
(4R)-2,2-Bis-[2-(3,3-dimethyl-4-phenylcarboxymethyl-1,3-oxazolinyl)]-
(4R)-2,2-Bis-[2-(phenylcarboxymethyl-1,3-oxazolinyl)]propane (4a):
A
propane (5a): In
a 500 mL one-neck round bottom flask bisamide
solution of bisamide 19a(Cl) (0.48 g, 0.90 mmol) and dry triethylamine
(3.0 mL, 21.6 mmol) in toluene (30 mL) was refluxed for 4 h. The mixture
was diluted with saturated NaHCO3 (50 mL) and extracted with ethyl
acetate (3î50 mL). The combined organic layers were washed with
brine, dried over Na2SO4, filtered and concentrated. The residue was pu-
rified by chromatography on deactivated alumina (ethyl acetate) to
afford the bis(oxazoline) 4a (0.30 g, 0.66 mmol) as colorless viscous oil.
Yield=73%. [a]2D0 =À91.2 (c=1.8 in CH2Cl2); 1H NMR (CDCl3): d=
1.52 (s, 6H; CH3), 4.20 4.60 (m, 10H), 7.30 7.60 (m, 6H; ArÀH), 7.95
ppm (m, 4H; Ar-H); 13C NMR (CDCl3): d=24.4, 38.6, 64.8 , 65.5, 70.0,
128.3, 129.6, 129.7, 133.0, 166.1, 170.7 ppm; MS (CI, NH3); m/z (%): 451
(100) [M + H]; elemental analysis calcd (%) for C25H26N2O6 (450.49): C
66.66, H 5.82, N 6.22, found: C 66.86, H 5.62, N 6.02.
19a(OH) (0.86 g, 1.59 mmol) was dissolved in CH2Cl2 (150 mL). The
flask was then fitted with a Soxhlet apparatus containing a cartridge with
3 g of CaH2. Methane sulfonic acid (0.52 mL, 7.95 mmol) was added to
the solution and the mixture was refluxed for 5 h. After cooling to room
temperature, the reaction mixture was washed with saturated NaHCO3
(2î50 mL). The organic layer was separated, dried over MgSO4, filtered
and concentrated under reduced pressure. The residue was purified by
column chromatography on silica gel (ethyl acetate) to afford 5a (0.62 g,
1.22 mmol) as colorless viscous oil. Yield=78%. [a]2D0 =À61.0 (c=1.35 in
CH2Cl2); 1H NMR (CDCl3): d=1.36 (s, 6H; CH3), 1.42 (s, 6H; CH3),
1.45 (s, 6H; CH3), 4.00 (dd, J=4.5 Hz and 7.4 Hz, 2H), 4.34 (dd, J=11.6
Hz and 7.4 Hz, 2H), 4.46 (dd, J=11.6 Hz and 4.5 Hz, 2H), 7.40 (m, 4H;
ArÀH ), 7.53 (m, 2H; ArÀH), 7.98 ppm (m, 4H; ArÀH ); 13C NMR
706
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 699 707

Dihydroxy Bis(Oxazoline) Ligands for the Pd-Catalyzed Asymmetric Allylic Alkylation
699 707
(CDCl3): d=21.3, 23.7, 28.7, 35.6, 63.6, 71.6, 85.9, 128.3, 129.5, 129.8,
133.1, 166.0, 169.3 ppm; MS (CI, NH3): m/z (%): 507 (100) [M + H]; ele-
mental analysis calcd (%) for C29H34N2O6 (506.59): C 68.76, H 6.76, N
5.53; found: C 68.35, H 6.90, N 5.65.
[4] S. Cossu, G. Giancomelli, S. Conti, M. Falorni, Tetrahedron 1994, 50,
5083.
[5] H. LÿvÜque, C. Malhiac, D. Speybrouck, Y. Combret, V. Guerpin, J.-
C. Combret, Bull. Soc. Fr. 1995, 132, 801.
[6] A. I. Meyers, W. Schmidt, M. J. McKennon, Synthesis 1993, 250.
[7] D. F. Elliot, J. Chem. Soc. 1950, 62.
[8] C. C. Price, E. L. Eliel, R. J. Convery, J. Org. Chem. 1957, 347.
[9] S. E. Denmark, N. Nakajima, O. J. C. Nicaise, A. M. Faucher, J. P.
Edwards, J. Org. Chem. 1995, 60, 4884.
[10] For previous use of l-serine in the preparation of chiral bis(oxazo-
line) ligands see: a) D. Muller, G. Umbricht, B. Weber, A. Pfaltz,
Helv. Chim. Acta 1991, 74, 232; b) E. J. Corey, K. Ishihara, Tetrahe-
dron Lett. 1992, 33, 6807; c) R. Tokunoh, H. Tomiyama, M. Sodeo-
ka, M. Shibasaki, Tetrahedron Lett. 1996, 37, 2449.
(4R)-2,2-Bis-[2-(3,3-dimethyl-4-hydroxymethyl-1,3-oxazolinyl)]propane
(5b): Following the procedure described for the synthesis of bis(oxazo-
line) 4b, the reaction of bis(oxazoline) 5a (0.35 g, 0.71 mmol) with
NaOH (1% wt) in methanol (10 mL) yielded 5b (0.20 g, 0.69 mmol).
Yield=96%. [a]2D0 =À79.2 (c=1.0 in CH2Cl2); 1H NMR (CDCl3): d=
1.34 (s, 6H; CH3), 1.44 (s, 6H; CH3), 1.47 (s, 6H; CH3), 3.70 ppm (m,
8H); 13C NMR (CD3OD): d=21.6, 24.3, 29.4, 62.5, 75.5, 88.4, 171.3 ppm;
MS (CI, NH3): m/z (%): 299 (100) [M + H]; elemental analysis calcd
(%) for C15H26N2O4 (298.38): C 60.38, H 8.78, N 9.39; found: C 60.25, H
8.75, N 9.10.
[11] J. F. Hansen, C. S. Cooper, J. Org. Chem. 1976, 41, 3219.
[12] D. Muller, G. Umbricht, B. Weber, A. Pfaltz, Helv. Chim. Acta 1991,
74, 232.
Alkylation of (E)-1,3-diphenyl-2-propenyl acetate with dimethyl malo-
nate: general procedure: Dimethyl malonate (0.170 mL, 1.5 mmol) and
two equivalents of the appropriate base system were mixed in dichloro-
methane (3 mL) under argon. In a separate flask, the chiral BO ligand
(0.02 mmol), [{Pd(h3-C3H5)Cl}2] (1.80 mg, 0.005 mmol) and racemic (E)-
1,3-diphenyl-2-propenyl-1-acetate 20 (126 mg, 0.5 mmol) in dichlorome-
thane (2 mL) were stirred for 20 minutes, and then added dropwise to
the above mixture. The resulting mixture was refluxed and the reaction
was monitored by thin layer chromatography on silica plate (ethyl ace-
tate/pentane; 15:85). After consumption of (E)-1,3-diphenyl-2-propenyl-
1-acetate 20, the reaction mixture was diluted with a NH4Cl saturated
aqueous solution (10 mL) and washed with dichloromethane (2î10 mL).
The organic layers were assembled, dried over MgSO4, filtered and con-
centrated under reduced pressure. The crude was purified by flash chro-
matography (ethyl acetate/pentane; 15:85) to yield 21. The enantiomeric
excess was determined by HPLC analysis with a chiral analytical column.
1H NMR (CDCl3, for 21): d=3.53 (s, 3H), 3.70 (s, 3H), 3.95 (d, J=11
Hz, 1H), 4.27 (dd, J=11 Hz, 8Hz, 1H), 6.32 (dd, J=15 Hz, 8 Hz, 1H),
6.48 (d, J=15 Hz, 8 Hz, 1H), 7.15 7.44 ppm (m, 10H). [a]20 =+19.2 (c=
1.30 in CHCl3) (+)-(R). HPLC Pharmacir 7C (nBuOH/n-hexane; 10:90;
0.7 mLminÀ1): tR (S) 10.66 min. and tR (R) 11.88 min.
[13] For reviews, see: a) C. G. Frost, J. Howarth, J. M. J. Williams, Tetra-
hedron:Asymmetry 1992, 3, 1089; b) J. Tsuji, Palladium Reagents
and Catalysis:Innovations in Organic Synthesis , Wiley, New York,
1995; c) B. M. Trost, Acc. Chem. Res. 1996, 29, 355; d) T. Hayashi in
Catalytic Asymmetric Synthesis, (Ed.: I. Ojima), Weinheim, 1993,
pp. 325; e) O. Reiser, Angew. Chem. 1993, 105, 576; Angew. Chem.
Int. Ed. Engl. 1993, 32, 547; f) B. M. Trost, D. L. van Vranken,
Chem. Rev. 1996, 96, 395; g) R. Noyori, Asymmetric Catalysis in Or-
ganic Synthesis, Wiley, New York, 1994; g) A. Pfaltz, M. Lautens, in
Comprehensive Asymmetric Catalysis, Vol. 2 (Eds.: E. N. Jacobson,
A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999, Chapter 24.
[14] For previous use of chiral bis(oxazoline) ligands in the palladium-
catalyzed asymmetric allylic alkylation, see: a) D. Muller, G. Um-
bricht, B. Weber, A. Pfaltz, Helv. Chim. Acta 1991, 74, 232; b) P.
von Matt, G. C. Lloyd-Jones, A. B. E. Minidis, A. Pfaltz, L. Macko,
M. Neuburger, M. Zehnder, H. Ruegger, P. S. Pregosin, Helv. Chim.
Acta 1995, 78, 265; c) W. Zhang, T. Kida, Y. Nakatsuji, I. Ikeda, Tet-
rahedron Lett. 1996, 37, 7995; d) M. Gomez, S. Jansat, G. Muller,
M. A. Maestro, J. Mahia, M. Font-Bardia, X. Solans, J. Chem. Soc
Dalton Trans. 2001, 1432; e) M. Gomez, S. Jansat, G. Muller, M. A.
Maestro, J. Mahia, Organometallics 2002, 21, 1077.
[15] a) T. Hayashi, A. Yamamoto, T. Hagihara, Y. Ito, Tetrahedron Lett.
1986, 27, 191; b) T. Hayashi, A. Yamamoto, Y. Ito, E. Nishioka, H.
Miura, K. Yanagi, J. Am. Chem. Soc. 1989, 111, 6301; d) P. von Matt,
A. Pfaltz, Angew. Chem. 1993, 105, 614; Angew. Chem. Int. Ed.
Engl. 1993, 32, 566.
[16] No solvent effect on this shift in the enantioselectivity was observed.
[17] For the effect of substituents at the C5 of the oxazoline ring in the
palladium-catalyzed allylic alkylation, see: M. A. Peric‡s, C. Puigja-
ner, A. Riera, A. Vidal-Ferran, M. GÛmez, F. Jeminÿz, G. Muller,
M. Rocamora, Chem. Eur. J. 2002, 8, 4164.
[18] B. M. Trost, D. J. Murphy, Organometallics 1985, 4, 1143.
[19] The verification of other bases such as nBuLi, HK, and tBuMgCl
did not show any effect on the direction of chiral induction, but af-
fected the level of enantioselectivity. For example, in the presence
of nBuLi as a base BO 1a afforded the R product in 53% ee and 1b
led to the S product in 28% ee.
[20] a) M. Sawamura, Y. Ito, Chem. Rev. 1992, 92, 857; b) T. Hayashi, K.
Kenhira, T. Hagihara, M. Kumada, J. Org. Chem. 1988, 53, 113; c) T.
Hayashi, A. Yamamoto, Y. Ito, E. Nishioka, H. Miura, K. Yanagi, J.
Am. Chem. Soc. 1989, 111, 6301.
[21] The silylation of the dihydroxy BO was ruled out because it is un-
likely that this reaction could occurs under the BSA/KOAc proce-
dure. Efforts were made for the preparation of this disylilated BO
ligand. This disilylated BO was found to be very instable under the
BSA/KOAc conditions.
[22] a) J. Tsuji, K. Takahashi, I. Minami, I. Shimazu, Tetrahedron Lett.
1984, 25, 4783; b) C. Carfagna, R. Galarini, A. Musco, R. Santi, J.
Org. Chem. 1991, 56, 3924; c) A. Saitoh, K. Achiwa, K. Tanaka, T.
Morimoto, J. Org. Chem. 2000, 65, 4227.
Acknowlegment
The authors are grateful to the CNRS for financial support of this work
and to the ™Ministõre de l’Enseignement Superieur et de la Recherche∫
for doctoral fellowships (H.O.).
[1] For the recent use of bis(oxazoline) ligands in asymmetric catalysis
see, for example: a) A. K. Ghosh, P. Mathivanan, J. Cappiello, Tetra-
hedron:Asymmetry 1998, 9, 1; b) J. S. Johnson, D. A. Evans, Acc.
Chem. Res. 2000, 33, 325; c) D. W. McMillan, N. Varga, B. A Redd,
C. King, J. Org. Chem. 2000, 65, 2532; d) S. E. Denmark, C. M. Stiff,
J. Org. Chem. 2000, 65, 5875; e) S. Nakamura, R. Nakagawa, Y. Wa-
tanabe, T. Toru, J. Am. Chem. Soc. 2000, 122, 11340; f) K. R. Knud-
sen, T. Risgaard, N. Nishiwaki, K. V. Gothelf, K. A. J˘rgensen, J.
Am. Chem. Soc. 2001, 123, 5843; g) D. L. Davies, J. Fawcett, S. A.
Garratt, D. R. Russell, Organometallics 2001, 20, 3029.
[2] a) O. Hoarau, H. AÔt-Haddou, M. Castro, G. G. A. Balavoine, Tetra-
hedron:Asymmetry 1997, 8, 3755; b) O. Hoarau, H. AÔt-Haddou, J.-
C. Daran, D. Cramailere, G. G. A. Balavoine, Organometallics 1999,
18, 4718; c) S. Iwasa, F. Takezawa, Y. Tsuchiya, H. Nishiyama,
Chem. Commun. 2001, 59; d) S. Iwasa, S. Tsushima, T. Shimada, H.
Nishiyama, Tetrahedron Lett. 2001, 42, 6715; e) M. Schinnerl, M.
Seitz, A. Kaiser, O. Reiser, Org. Lett. 2001, 3, 4259.
[3] For the reversal in enantioselectivity with bis(oxazoline) metal cata-
lysts depending on the oxazoline ring substituent see: a) D. A.
Evans, J. S. Johnson, C. S. Burgey, K. R. Campos, Tetrahedron Lett.
1999, 40, 2879; b) M. Johannsen, K. A. J˘rgensen, J. Org. Chem.
1995, 60, 5757; c) M. Johannsen, S. Yao, K. A. J˘rgensen, Chem.
Commun. 1997, 2169; d) M. Johannsen, S. Yao, A. Graven, K. A.
J˘rgensen, Pure Appl. Chem. 1998, 70, 1117.
[23] S. R. Gilbertson, C.-W. T. Chang, J. Org. Chem. 1998, 63, 8424.
Received: December 6, 2002
Revised: June 16, 2003 [F4649]
707
Chem. Eur. J. 2004, 10, 699 707
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim