Synthesis of new sulfur-containing oxazoline ligands and their use in palladium-catalyzed allylic substitution

Article abstract of DOI:10.1016/S0957-4166(02)00829-7

New chiral sulfur-containing ligands have been prepared and fully characterized. Their structure includes a dibenzothiophene or benzothiophene ring as backbone, where the sulfur atom is enclosed in a strong π-donor structure. The chirality was introduced

Full text of DOI:10.1016/S0957-4166(02)00829-7

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 339–346
Synthesis of new sulfur-containing oxazoline ligands and their
use in palladium-catalyzed allylic substitution
Arnaud Voituriez and Emmanuelle Schulz*
Laboratoire de Catalyse Mole´culaire, Institut de Chimie Mole´culaire d’Orsay, Universite´ Paris-Sud,
UPRESA CNRS 8075 Baˆt 420, 91405 Orsay cedex, France
Received 31 October 2002; accepted 4 December 2002
Abstract—New chiral sulfur-containing ligands have been prepared and fully characterized. Their structure includes a dibenzo-
thiophene or benzothiophene ring as backbone, where the sulfur atom is enclosed in a strong p-donor structure. The chirality was
introduced by oxazoline moieties placed near the sulfur atom. These ligands have been successfully tested in asymmetric
palladium-catalyzed allylic substitutions leading to products with ee of up to 77%. © 2003 Elsevier Science Ltd. All rights
reserved.
1. Introduction
ric catalytic CꢀC bond formations. For example,
Sannicolo` et al. have described the preparation of lig-
ands based on chiral atropoisomeric structures ((R)-(+)-
BITIANP 1,4 and (+)-TMBTP 2,5 Scheme 1) and their
use in enantioselective Heck reactions.6,7
The preparation of new and efficient enantiopure lig-
ands providing a chiral environment to metals for
asymmetric catalysis is nowadays a great concern in
modern organic chemistry. Highly selective and active
catalysts have already been prepared, some of them
being used at the industrial scale, mostly in hydrogena-
tion reactions.1 A lot of work has already been pub-
lished and is still in progress to optimize transforma-
tions involving CꢀC bond formations. Recent reviews2,3
clearly show that nitrogen-containing ligands (and espe-
cially those with oxazoline moieties) are particularly
good candidates to achieve, for example, allylic substi-
tutions, cyclopropanations, Diels–Alder reactions, aldol
or Michael-type additions.
Pd-catalyzed allylic substitutions, i.e. Tsuji–Trost reac-
tions,8 have been intensively studied with sulfur-con-
taining ligands over the past few years. Anderson et
al.9,10 developed a series of sulfur–imine mixed donor
chiral ligands (see structure 3, Scheme 1) which gave up
to 94% enantiomeric excess in this reaction. Sulfur-pyr-
idine compounds (4, Scheme 1) were prepared by
Chelucci11 and enantioselectivities of up to 83% were
obtained. Williams prepared various S,N-ligands
(structures 512–7, Scheme 2) containing oxazoline func-
tionalities and sulfur as an auxiliary donor ligand,
providing enantioselectivities of 40–96% ee.13 Bryce
synthesized chiral oxazolines linked to tetrathiaful-
Furthermore, the efficient use of sulfur-containing lig-
ands has been recently reported in numerous asymmet-
Scheme 1.
* Corresponding author. Fax: 33 (0) 1 69154680; e-mail: emmaschulz@icmo.u-psud.fr
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00829-7

340
A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
Scheme 2.
valene14 and chiral ferrocenyl-oxazolines incorporating
thioether units (see structure 8, Scheme 2).15 These
redox-active liganding systems were successfully used in
palladium-catalyzed allylic substitution reactions (up to
93% ee).
efficient and enantioselective in Diels–Alder reactions.19
Such complexes promoted moreover asymmetric
nitrone cycloadditions20 and 1,3-dipolar cycloadditions
of diazoalkane.21 Only recently, DBFOX/Ph was tested
as a chiral ligand to prepare a palladium complex and
its catalytic performance was studied in allylic substitu-
tion reactions.22 This system proved to be inactive and
the authors assumed that the steric hindrance between
the chiral ligand and the allyl group was responsible for
the instability of the complex. The well-known affinity
of palladium towards sulfur justified the choice of
testing DBT-BOx’s and their derivatives in Pd-cata-
lyzed allylic substitutions.
Based upon the previous work, we attempted to intro-
duce new sulfur-containing oxazoline compounds as
ligands for asymmetric catalysis. We report herein their
synthesis as well as our first promising results with
these chiral ligands for the enantioselective palladium-
catalyzed allylic substitution.16
2. Results and discussion
2.1. Synthesis of 4,6-dibenzothiophenediyl-2,2%-bis(4-
alkyloxazolines)—DBT-BOx’s
We have chosen dibenzothiophene and benzothiophene
as backbones, where the sulfur atom is part of a strong
p-donor structure. These skeletons are of particular
interest since their aromatic structure opens the door to
a great variety of synthetic transformations, on various
positions of the aromatic rings, according to the proce-
dure used.
DBT-BOx ligands have been prepared following classi-
cal procedures to obtain bis(oxazolines) (Scheme 4).
Dibenzothiophene was bislithiated23 using 4 equiv. of
n-butyllithium
and
tetramethylethylenediamine
(TMEDA) in refluxing cyclohexane and subsequently
dicarboxylated24 with dry ice to give 4,6-dibenzothio-
phenedicarboxylic acid 12 with 88% yield. The corres-
ponding diamide 14 was obtained via the acid dichlo-
ride 13, and cyclization was then performed by activa-
tion with diethylaminosulfur trifluoride (DAST).25 The
Furthermore different possible sites of chelation are
offered by these ligands, i.e., N,N%- or N,S-type liga-
tion. C2-symmetric bis(oxazolines) based on the diben-
zothiophene structure, (DBT-BOx, 9, Scheme 3),17 have
also been synthesized to allow potential trans-chelating
tridentate ligands. DBT-MOx 10 and BT-MOx 11
(Scheme 3) provide different possibilities: the former
can afford a six-membered chelate with the metal,
whereas the latter may yield a five-membered chelate.
Moreover, the structural differences of both heteroaro-
matic rings were supposed to induce various electronic
effects on the complexes. Kanemasa et al.18 have
described an analogous ligand derived from dibenzo-
furan and substituted by two (R)-phenyl-bis(oxazoli-
nes) (DBFOX/Ph). With various transition-metal(II)
perchlorates, this ligand proved to be particularly
use of (S)-(+)-2-amino-3-methyl-1-butanol (L-valinol)
allowed the preparation of DBT-BOx(iPr) 9a in 52%
yield for the whole procedure starting from dicarboxylic
acid 12. DBT-BOx(Ph) 9b was similarly synthesized in
41% yield from (R)-(−)-2-phenylglycinol and DBT-
BOx(tBu) 9c was obtained in 37% yield from (S)-tert-
leucinol.
2.2. Synthesis of 4-dibenzothiophenyl-2-(4-alkyloxa-
zolines)—DBT-MOx’s
A similar strategy was used to obtain different DBT-
MOx derivatives (Scheme 5). The synthesis of 4-diben
Scheme 3.

A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
341
Scheme 4.
Scheme 5.
zothiophene carboxylic acid could be performed under
less severe conditions than those used to synthesize the
corresponding diacid. The monoacid 15 was indeed
obtained in 75% yield by preparing the monolithiated
species with n-BuLi (2 equiv.) in THF at 0°C and
subsequently quenching the solution with dry ice.26 The
formation of mono-oxazoline dibenzothiophene deriva-
tives followed then the above-mentioned procedure by
using 1.1 equiv. of the desired amino-alcohol. Starting
from 4-dibenzothiophene carboxylic acid, DBT-
MOx(iPr) 10a was thus obtained in 22% yield for the
total steps. DBT-MOx(Ph) 10b was similarly synthe-
sized in 34% yield.
Scheme 6.
the transformation of the 1,3-diphenylallyl system with
dimethyl malonate (Scheme 7). This reaction allows
indeed an easy comparison and analysis of the results
due to the symmetry of the substrate. Allylic substitu-
tion of rac-1,3-diphenyl-2-propenyl acetate 17 was
firstly investigated under various reaction conditions
with DBT-BOx(iPr) 9a as chiral ligand. The C2-symme-
try of this ligand resulted in catalytic systems with
restricted numbers of diastereomeric transition states
and the analysis of the interactions responsible for
enantioselection was thus facilitated.
2.3. Synthesis of 2-benzothiophenyl-2-(4-
alkyloxazoline)s—BT-MOx’s
Interested in testing five-membered chelating N,S-lig-
ands, we chose the commercially available thianaph-
thene-2-carboxylic acid 16 as precursor. From this
backbone, BT-MOx(iPr) 11a could be analogously
obtained in 54% yield, BT-MOx(Ph) 11b in 64% yield
and BT-MOx(tBu) 11c was prepared in 74% yield
(Scheme 6).
We have previously reported16 that DBT-BOx 9a, con-
trary to its dibenzofuran analogue,22 induced a high
level of enantioselectivity (77% e.e., Table 1, entry 1)
and activity in the transformation of rac-17, when the
nucleophile was prepared in refluxing dichloromethane
from BSA (N,O-bis(trimethylsilyl)acetamide) using
2.4. Pd-catalyzed allylic substitution
To evaluate the selectivity of a new chiral ligand for
allylic substitutions, the reaction usually performed is

342
A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
Scheme 7.
allylpalladium chloride dimer as catalyst precursor.
Under these conditions, (R)-18 was isolated in 90%
yield.
sible for the observed configuration. Regarding the
steric hindrance generated by the isopropyl-substituent
on the oxazoline group, the substrate was thus placed
on the square-planar palladium complex, as depicted in
Scheme 8. In such a context, the intermediate A would
be favored over the alternative B, where the steric
hindrance is greater between the substituent of the
oxazoline ring and the phenyl group of the substrate.
Analogous ligands 9b and 9c (Table 1, entries 2 and 3)
were both less efficient and selective when the transfor-
mation was performed under these optimized condi-
tions. Furthermore, we noticed that DBT-BOx 9a
containing chiral fragments with (S)-configuration led
to the substitution-product 18 with (R)-configuration as
the major isomer. This is in contrast with results
obtained by using other C2-symmetric bis(oxazoli-
nes),28–30 where (S)-configuration in the chiral ligand
led to the (S)-18 product as the major product. Our
DBT-BOx system, as potential tridentate chelate, may
also act as a bidentate or monodentate ligand, which
complicates the explanation of the results obtained.
However, compared with the inefficiency of the diben-
zofuran analogue to perform this palladium-catalyzed
reaction, we assume a probable participation of the
sulfur atom for stabilizing the complex involved in the
selective transformation.
The incoming nucleophile is supposed to attack on the
opposite face of the p-allyl system, i.e. on the less-steri-
cally hindered face. The regiochemistry of attack has
been demonstrated to be influenced by electronic fac-
tors with bidentate ligands containing donor atoms
with different properties.10,12,31 Dibenzothiophene is an
electron-rich compound, and thus acts as a p-donor.32
It is assumed that this electronic information would be
transferred via the trans effect to the allyl moiety. In
our particular case, the nucleophile would thus prefer-
entially attack the allylic system on the carbon possess-
ing a greater positive charge character, i.e. on the
carbon situated trans to the nitrogen atom of the
oxazoline moiety (see Scheme 8, for DBT-MOx 10a).
DBT-MOx 10a led indeed to (S)-18 as the major enan-
tiomer. On the contrary, and as expected by the above-
mentioned mechanism, DBT-MOx 10b possessing a
phenyl group with (R)-configuration on the oxazoline,
allowed the preparation of 18 with a similar enan-
tiomeric excess (34%, Table 2, entry 5) but with (R)-
configuration. Furthermore, we tested the influence of
the ratio ligand versus palladium on the enantiomeric
excess, and we observed in the case of 10a an increase
in the ee value with the quantity of ligand introduced,
accompanied with a maximum activity for a ratio of
2:1. This indicated that this ligand was probably not
strongly bounded to the metal. However, when present
in a large excess (Table 2, entry 4) DBT-MOx 10a acted
to poison the catalyst, since the conversion only
reached 56% in nearly 5 days.
DBT-MOx compounds were then tested as ligands for
the palladium-catalyzed allylic substitution of rac-17
under the same conditions as those used for DBT-BOx
(Table 2). These N,S-chelating ligands do not contain a
C2-symmetry axis and are able to form six-membered
chelates with the palladium atom.
The transformation of rac-17 was first realized with 10a
in dichloromethane, with a ligand:palladium ratio of
1:1. This catalytic system allowed the preparation of the
desired product 18, with a lower activity than that of
the analogous bis(oxazoline) system 9a. Furthermore, it
was noticed that, contrary to DBT-BOx(iPr) 9a, the
monooxazoline 10a led predominantly to the (S)-enan-
tiomer of 18, with 30% ee (Table 2, entry 1). We
assumed both steric and electronic effects to be respon-
Finally, the Tsuji–Trost reaction was tested in the pres-
ence of the BT-Mox ligands 11 as sulfur-nitrogen-
Table 1. Allylic substitution of rac-17 in the presence of
DBT-BOx ligandsa
chelates, leading to
a
five-membered ring on
coordination with the palladium centre, in an attempt
to achieve selectivity control by reducing the chelate
ring-size. The results are summarized in Table 3. How-
ever, the use of ligands 11 lead to catalysts with slightly
lower activity than DBT-MOx’s.
Entry
Ligand
t (h)
Conversionb
(% of 18)
e.e.c (% of 18)
1
2
3
9a
9b
9c
70
120
70
100
39
50
77 (R)
28 (S)
51 (R)
Again, it could be demonstrated that these N,S-ligands
are not strongly bound to the metal centre, since the
enantioselectivity value increased with the ratio ligand
11a/palladium introduced at the beginning of the reac-
tion, from 0% for a 1/1 ratio to a maximum of 35% ee
for a 4/1 ratio (Table 3, entries 1–3).
a 2.5 mol% [(h3-C3H5)PdCl]2, 3 equiv. of dimethylmalonate, 3 equiv.
of BSA, 2 mol% of KOAc, CH2Cl2, 40°C.
b Determined by GC analysis.
c Determined by HPLC (Whelk O1) and absolute configuration deter-
mined by comparison with literature values.27

A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
Table 2. Allylic substitution of rac-17 in the presence of DBT-MOx ligandsa
343
Entry
Ligand
L*/Pd
t (h)
Conversionb (% of 18)
e.e.c (% of 18)
1
2
3
4
5
10a
10a
10a
10a
10b
1/1
2/1
4/1
6/1
2/1
111
88
98
117
112
54
100
100
56
30 (S)
32 (S)
37 (S)
41 (S)
34 (R)
80
a 2.5 mol% [(h3-C3H5)PdCl]2, 3 equiv. of dimethyl malonate, 3 equiv. of BSA, 2 mol% of KOAc, CH2Cl2, 40°C.
b Determined by GC analysis.
c Determined by HPLC (Whelk O1) and absolute configuration determined by comparison with literature values.27
Scheme 8.
Table 3. Allylic substitution of rac-17 in the presence of BT-MOx ligandsa
Entry
Ligand
L*/Pd
t (h)
Conversionb (% of 18)
e.e.c (% of 18)
1
2
3
4
5
11a
11a
11a
11b
11c
1/1
2/1
4/1
2/1
2/1
111
115
118
111
115
74
100
30
71
78
0
13 (R)
35 (R)
30 (S)
13 (S)
a 2.5 mol% [(h3-C3H5)PdCl]2, 3 equiv. of dimethylmalonate, 3 equiv. of BSA, 2 mol% of KOAc, CH2Cl2, 40°C.
b Determined by GC analysis.
c Determined by HPLC (Whelk O1) and absolute configuration determined by comparison with literature values.27
The direction of the enantioselectivity was more
difficult to describe and rationalize since BT-MOx 11a
and BT-MOx 11c, possessing both a substituent with
(S)-configuration on the oxazoline moiety, led to (R)-
18 (or (S)-18)), respectively, as the major enantiomer,
albeit with low enantioselectivity (Table 3, entries 2 and
5). We assume that the ability of (S)-BT-MOx(tBu) 11c
to preferentially stabilize one of the diastereomeric
complexes is too low to allow predominant stereodiffer-
ention. (R)-BT-MOx 11b led also to (S)-18 as major
enantiomer (Table 3, entry 4). The enantioselectivity of
the reaction performed with these ligands is not
explainable by a mechanism as proposed in Scheme 8.
Such effects have already been reported by Chelucci et
al.33 who described the use of oxazolinylpyridines and
acridininyloxazolines as ligands for the same transfor-
mation. The introduction of a benzo-fused substituent
on the pyridine ring not containing the chiral backbone
resulted, also in their case, in the switch of the expected
chiral sense of enantioselection.
3. Conclusion
We have prepared a variety of new sulfur-containing
oxazoline(s) with benzothiophene or dibenzothiophene
as backbones. The compounds have been fully charac-
terized and tested as ligands in the asymmetric palla-
dium-catalyzed substitution of rac-1,3-diphenyl-
2-propenyl acetate with dimethylmalonate. (S,S)-DBT-
BOx(iPr) 9a, as the most active and selective ligand of
this series, allowed the preparation of (R)-18 with 77%
e.e. It is worth mentioning that the analogous dibenzo-
furan–bisoxazoline ligand was not successfully used for

344
A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
this transformation. Furthermore, and to the best of
our knowledge, other reported bisoxazolines led to
reversal selectivities as those here obtained. We thus
assume that the sulfur atom of the dibenzothiophene
backbone probably participates in the stabilization of
the sterically favored Pd-complex. DBT-MOx’s, as six-
membered chelate ring-forming ligands proved to be
more active and selective than their five-membered
chelate ring-forming counterparts of the BT-MOx
series. These ligands displayed lower efficiency and
selectivity than the DBT-BOx’s. Other catalytic enan-
tioselective reactions using these ligands are currently in
progress in our laboratory.
mmol, 2.2 equiv.) in dry CHCl3 (15 mL). This mixture
was stirred at rt overnight. Usual work up procedure
(aq. NH4Cl and CH2Cl2) yielded dibenzothiophene-4,6-
diamide 14 as a brown solid, used in the cyclization
step without purification.
To 14 (3.8 mmol) in dry CH2Cl2 (40 mL) at −78°C, was
slowly added (diethylamino)sulfur trifluoride DAST
(10.0 mol, 2.6 equiv.). After stirring 4 h at −78°C and
overnight at rt, a solution of 14% aq. NH4OH was
added. Usual work up (CH2Cl2) yielded DBT-BOx
ligands 9 as solids, which were recrystallized several
times in cold pentane.
4.1.1.
propyloxazoline) (DBT-BOx(iPr)), 9a. Obtained from
(S)-(+)-2-amino-3-methyl-1-butanol ( -Valinol) in 52%
(S,S)-4,6-Dibenzothiophenediyl-2,2%-bis(4-iso-
4. Experimental
L
yield from dibenzothiophene-4,6-dicarboxylic acid 12 as
Solvents were purified by standard techniques: CH2Cl2,
CHCl3, cyclohexane and diethylether were dried over
CaH2 and THF over sodium and benzophenone. H
a light brown solid: mp 194–196°C; [h]2D0=−74.3 (c
1
2.28, CHCl3); H NMR (250 MHz, CDCl3) l 8.27 (dd,
1
2H, J=7.8 Hz, 1.0 Hz), 7.99 (dd, 2H, J=7.3 Hz, 1.0
Hz), 7.50 (dd, 2H, J=7.8 Hz, 7.3 Hz), 4.51 (m, 2H),
4.19 (m, 4H), 1.84 (m, 2H), 1.17 (d, 6H, J=6.8 Hz),
1.03 (d, 6H, J=6.8 Hz); 13C NMR (62.5 MHz, CDCl3)
l 162.1, 141.7, 135.9, 127.2, 123.9 123.7, 122.5, 73.4,
70.8, 33.5, 19.1, 19.0; HRMS calcd for C24H26N2O2S
406.1704, found 406.1715.
and 13C Nuclear Magnetic Resonance (NMR) spectra
were recorded on a Brucker AM-200 and a AC-250
Fourier Transform spectrometer and obtained in chlo-
roform-d. Chemical shifts are reported in parts per
million (ppm), and coupling constants are reported in
Hertz (Hz). Optical rotations were measured with a
Perkin–Elmer 341 Polarimeter. Conversions were deter-
mined by GC analysis (DB-1 column, 10 m×0.32 mm
Ø). Enantiomeric excesses were measured by HPLC
(Whelk 01 column, hexane/isopropanol: 98/2, flow
rate=1 mL/min, u=254 nm). High Resolution Mass
Spectra were obtained with a Finningan-MAT-95-5-S
apparatus. The amino-alcohols were purchased from
Acros and thianaphthene-2-carboxylic acid 16 is com-
mercially available from Aldrich.
4.1.2. (R,R)-4,6-Dibenzothiophenediyl-2,2%-bis(4-pheny-
loxazoline) (DBT-BOx(Ph)), 9b. Obtained from (R)-(−)-
2-phenylglycinol (2-amino-2-phenylethanol) in 41%
yield from 12 as a light brown solid: mp 230–232°C;
[h]2D0=−247.7 (c 1.29, CHCl3); 1H NMR (200 MHz,
CDCl3) l 8.36 (dd, 2H, J=7.8 Hz, 1.0 Hz), 8.18 (dd,
2H, J=7.3 Hz, 1.0 Hz), 7.58 (dd, 2H, J=7.8 Hz, 7.3
Hz), 7.50–7.20 (m, 10H), 5.59 (dd, 2H, J=8.3 Hz, 8.3
Hz), 4.89 (dd, 2H, J=8.3 Hz, 8.3 Hz), 4.33 (dd, 2H,
J=8.3 Hz, 8.3 Hz); 13C NMR (62.5 MHz, CDCl3) l
163.8, 142.4, 141.4, 136.0, 128.6, 128.0, 127.4, 126.7,
124.2, 124.1, 122.3, 74.9, 70.1; HRMS calcd for
C30H22N2O2S 474.1402, found 474.1402.
4.1. General synthesis of DBT-BOx ligands, 9
Dibenzothiophene-4,6-dicarboxylic acid 12 was pre-
pared according to a procedure described by Macaud et
1
al.24 and obtained in 88% yield: H NMR (250 MHz,
DMSO-d6) l 8.71 (d, 2H, J=7.8 Hz), 8.20 (d, 2H,
J=7.3 Hz), 7.67 (dd, 2H, J=7.3 Hz, 7.8 Hz); 13C NMR
(62.5 MHz, DMSO-d6) l 167.3, 141.7, 135.6, 129.5,
126.3, 125.0, 124.8.
4.1.3.
(S,S)-4,6-Dibenzothiophenediyl-2,2%-bis(4-tert-
butyloxazoline) (DBT-BOx(tBu)), 9c. Obtained from
(S)-tert-Leucinol (2-amino-3,3-dimethyl-1-butanol) in
37% yield from 12 as a yellow solid: mp 183–185°C;
[h]2D0=−49.4 (c 1.06, CHCl3); 1H NMR (200 MHz,
CDCl3) l 8.27 (dd, 2H, J=7.8 Hz, 1.0 Hz), 7.98 (dd,
2H, J=7.3 Hz, 1.0 Hz), 7.50 (dd, 2H, J=7.8 Hz, 7.3
Hz), 4.41 (m, 2H), 4.26 (m, 4H), 1.03 (s, 18H); 13C
NMR (62.5 MHz, CDCl3) l 162.4, 142.4, 136.4, 127.6,
124.3, 124.1, 123.0, 77.3, 69.0, 34.4, 26.4; HRMS calcd
for C26H30N2O2S 434.2029, found 434.2028.
Diacid 12 (4.0 g, 14.7 mmol) was suspended in dry
CHCl3 (55 ml), thionyl chloride (32 mL, 0.4 mol, 30
equiv.) and DMF (1 drop) were added at rt. The
mixture was refluxed 2.5 h, and then stirred at rt
overnight. The resulting mixture was filtered, and the
solid washed with CHCl3. After drying in vacuum,
dibenzothiophene-4,6-dicarbonyl dichloride 13 was
obtained as a white powder (3.8 g, 13.5 mmol, 92%).
This compound was used in the following reaction
without further purification.
4.2. General synthesis of DBT-MOx ligands, 10
Dibenzothiophene-4-carboxylic acid 15 was prepared
according to a procedure described by Tye et al.26 and
obtained in 75% yield, mp 257–259°C. (lit. mp (AcOH)
261–263°C, 75% yield). This monoacid 15 (2.0 g, 8.8
mmol) was suspended in dry CHCl3 (35 mL), thionyl
To a suspension of 13 (2.0 g, 6.5 mmol) in dry CHCl3
(50 mL) at 0°C, was slowly added a solution containing
the commercially available chiral amino-alcohol (14.9
mmol, 2.3 equiv.) and triethylamine (2.0 mL, 14.3

A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
345
chloride (19.2 mL, 0.3 mol, 30 equiv.) and DMF (1
drop) were added at rt. The mixture was refluxed 2.5 h,
and then stirred at rt overnight. The resulting mixture
was filtered, and the solid washed with CHCl3. After
drying in vacuum, dibenzothiophene-4-carbonyl chlo-
ride was obtained as a white powder (2.01 g, 8.1 mmol,
93%). This compound was used in the following reac-
tion without further purification.
stirred at rt overnight. The resulting mixture was
filtered, and the solid washed with CHCl3. After drying
in vacuum, thianaphtene-2-carbonyl chloride was
obtained as a white powder (1.6 g, 8.1 mmol, 97%).
This compound was used in the following reaction
without further purification.
To a suspension of the monoacid chloride (1.6 g, 8.1
mmol) in dry CHCl3 (50 mL) at 0°C, was slowly added
a solution containing the commercially available chiral
amino-alcohol (9.0 mmol, 1.1 equiv.) and triethylamine
(1.25 mL, 9.0 mmol, 1.1 equiv.) in dry CHCl3 (15 mL).
This mixture was stirred overnight at rt. Usual work up
procedure (aq. NH4Cl and CH2Cl2) yielded thianapht-
ene-4-amide as a brown solid, used in the cyclization
step without purification.
To a suspension of the monoacid chloride (1.0 g, 4.1
mmol) in dry CHCl3 (30 mL) at 0°C, was slowly added
a solution containing the commercially available chiral
amino-alcohol (4.5 mmol, 1.1 equiv.) and triethylamine
(0.6 mL, 4.5 mmol, 1.1 equiv.) in dry CHCl3 (10 mL).
This mixture was stirred overnight at rt. Usual work up
procedure (aq. NH4Cl and CH2Cl2) yielded dibenzoth-
iophene-4-amide as a brown solid, used directly in the
cyclization step without purification.
To the monoamide (8.1 mmol) in dry CH2Cl2 (80 mL)
at −78°C, was slowly added (diethylamino)sulfur tri-
fluoride DAST (0.01 mol, 1.3 equiv.). After stirring 4 h
at −78°C and overnight at rt, a solution of 14% aq.
NH4OH was added. Usual work up (CH2Cl2) yielded
BT-MOx ligands 11 as solids, which were recrystallized
several times in cold pentane.
To the monoamide (1.2 g, 3.5 mmol) in dry CH2Cl2 (35
mL) at −78°C, was slowly added (diethylamino)sulfur
trifluoride DAST (0.6 mL, 4.5 mmol, 1.3 equiv.). After
stirring 4 h at −78°C and overnight at rt, a solution of
14% aq. NH4OH was added. Usual work up (CH2Cl2)
yielded DBT-MOx ligands 10 as brown solids, which
were recrystallized several times in cold pentane.
4.3.1. (S)-2-Benzo[b]thiophen-2-yl-4-isopropyloxazoline
(BT-MOx(iPr)), 11a. Obtained from (S)-(+)-2-amino-3-
4.2.1. (S)-2-Dibenzothiophen-4-yl-4-isopropyloxazoline
(DBT-MOx(iPr)), 10a. Obtained from (S)-(+)-2-amino-
methyl-1-butanol (L-Valinol) in 54% yield from thi-
anaphthene-2-carboxylic acid 16 as a white solid; mp
1
80–82°C; [h]2D0=−57.6 (c 0.99, CHCl3); H NMR (200
3-methyl-1-butanol (
L
-Valinol) in 22% yield from
dibenzothiophene-4-carboxylic acid 15 as a light brown
solid: mp 88–90°C; [h]2D0=−28.3 (c 0.86, CHCl3); 1H
NMR (250 MHz, CDCl3) l 8.26 (dd, 1H, J=7.8 Hz,
1.0 Hz), 8.16 (m, 1H), 7.98 (dd, 1H, J=7.3 Hz, 1.0 Hz),
7.89 (m, 1H), 7.52 (dd, 1H, J=7.8 Hz, 7.3 Hz), 7.48–
7.40 (m, 2H), 4.48 (m, 1H), 4.21 (m, 2H), 1.88 (m, 1H),
1.13 (d, 3H, J=6.5 Hz), 1.01 (d, 3H, J=6.5 Hz); 13C
NMR (62.5 MHz, CDCl3) l 162.8, 142.3, 140.2, 137.2,
135.5, 127.7, 127.6, 124.9, 124.7, 124.6, 123.3, 123.0,
122.1, 73.8, 71.1, 34.0, 19.6, 19.2; HRMS calcd for
C18H17NOS 295.1027, found 295.1031.
MHz, CDCl3) l 7.85–7.72 (m, 3H), 7.45–7.38 (m, 2H),
4.50–4.35 (m, 1H), 4.20–4.05 (m, 2H), 1.96–1.79 (m,
1H), 1.02 (d, 3H, J=6.8 Hz), 0.92 (d, 3H, J=6.8 Hz);
13C NMR (62.5 MHz, CDCl3) l 160.1, 141.9, 139.8,
131.2, 127.8, 126.7, 125.4, 125.4, 123.2, 73.6, 71.3, 33.4,
19.6, 18.7; HRMS (ESI) calcd for C14H15NOS (M+H+)
246.0949, found 246.0953.
4.3.2.
(R)-2-Benzo[b]thiophen-2-yl-4-phenyloxazoline
(BT-MOx(Ph)), 11b. Obtained from (R)-(−)-2-phenyl-
glycinol (2-amino-2-phenylethanol) in 64% yield from
16 as a white solid; mp 90–92°C; [h]2D0=+6.6 (c 0.99,
4.2.2.
(R)-2-Dibenzothiophen-4-yl-4-phenyloxazoline
1
CHCl3); H NMR (250 MHz, CDCl3) l 7.95–7.75 (m,
(DBT-MOx(Ph)), 10b. Obtained from (R)-(−)-2-phenyl-
glycinol (2-amino-2-phenylethanol) in 34% yield from
15 as a brown solid: mp 110–112°C; [h]2D0=−190.1 (c
3H), 7.50–7.25 (m, 7H), 5.40 (dd, 1H, J=9.8 Hz, 8.3
Hz), 4.82 (dd, 1H, J=8.3 Hz, 9.8 Hz), 4.30 (dd, 1H,
J=8.3 Hz, 8.3 Hz); 13C NMR (62.5 MHz, CDCl3) l
161.4, 142.5, 142.0, 139.7, 130.7, 129.4, 128.4, 127.4,
126.9, 125.5, 125.4, 123.2, 76.1, 71.0; HRMS (ESI)
calcd for C17H14NOS (M+H+) 280.0801, found
280.0796.
1
1.03, CHCl3); H NMR (200 MHz, CDCl3) l 8.29 (d,
1H, J=7.8 Hz), 8.16 (m, 1H), 8.10 (d, 1H, J=7.8 Hz),
7.90 (m, 1H), 7.60–7.30 (m, 8H), 5.64 (dd, 1H, J=8.3
Hz, 8.3 Hz), 4.86 (dd, 1H, J=8.3 Hz, 8.3 Hz), 4.31 (dd,
1H, J=8.3 Hz, 8.3 Hz); 13C NMR (62.5 MHz, CDCl3)
l 164.4, 143.1, 142.1, 140.3, 137.3, 135.4, 129.4, 128.2,
128.0, 127.6, 127.2, 125.0, 124.9, 124.7, 123.3, 122.6,
122.1, 75.5, 70.9; HRMS calcd for C21H15NOS
329.0875, found 329.0874.
4.3.3. (S)-2-Benzo[b]thiophen-2-yl-4-tert-butyloxazoline
(BT-MOx(tBu)), 11c. Obtained from (S)-tert-Leucinol
(2-amino-3,3-dimethyl-1-butanol) in 74% yield from 16
as a light brown solid; mp 75°C; [h]2D0=−58.7 (c 1.01,
1
4.3. General synthesis of BT-MOx ligands, 11
CHCl3); H NMR (250 MHz, CDCl3) l 7.85–7.70 (m,
3H), 7.40–7.25 (m, 2H), 4.40–4.20 (m, 2H), 4.05 (m,
1H), 0.95 (s, 9H); 13C NMR (62.5 MHz, CDCl3) l
160.3, 141.8, 139.5, 130.2, 128.4, 126.7, 125.4, 125.2,
122.9, 76.3, 70.1, 34.6, 26.3; HRMS calcd for
C15H17NOS 259.1026, found 259.1030.
Thianaphthene-2-carboxylic acid 16 (1.5 g, 8.4 mmol)
was suspended in dry CHCl3 (30 mL), thionyl chloride
(18.5 mL, 0.25 mol, 30 equiv.) and DMF (1 drop) were
added at rt. The mixture was refluxed 2.5 h, and then

346
A. Voituriez, E. Schulz / Tetrahedron: Asymmetry 14 (2003) 339–346
4.4. General procedure for palladium-catalyzed allylic
alkylation of rac-(E)-1,3-diphenyl-prop-2-enyl acetate
11. Chelucci, G.; Cabras, M. A. Tetrahedron: Asymmetry
1996, 7, 965–966.
12. Frost, C. G.; Williams, J. M. J. Tetrahedron Lett. 1993,
34, 2015–2018.
13. Allen, J. V.; Coote, S. J.; Dawson, G. J.; Frost, C. G.;
Martin, C. J.; Williams, J. M. J. J. Chem. Soc., Perkin
Trans. 1 1994, 2065–2072.
14. Chesney, A.; Bryce, M. R. Tetrahedron: Asymmetry 1996,
7, 3247–3254.
15. Chesney, A.; Bryce, M. R.; Chubb, R. W. J.; Batsanov,
A. S.; Howard, J. A. K. Tetrahedron: Asymmetry 1997, 8,
2337–2346.
The ligand (10 mol%), [(h3-C3H5)PdCl]2 (2.5 mol%) and
KOAc (2 mol%) were dissolved in dry CH2Cl2 (1.5
mL). The reaction mixture was degassed in three freeze-
evacuate-thaw cycles and stirred at 40°C for 15 min
under argon atmosphere. BSA (N,O-bis(trimethylsi-
lyl)acetamide) (3 equiv.) and dimethyl malonate (3
equiv.) were added, and the resultant solution was
stirred at 40°C for 15 min. Whereupon, rac-(E)-1,3-
diphenylprop-2-enyl acetate (1 equiv.) was added, and
the reaction mixture was degassed in three freeze–evac-
uate–thaw cycles before the reaction vessel was sealed,
and stirred at 40°C under argon atmosphere for 3 days.
It was then diluted with diethyl ether and washed with
a saturated aqueous NH4Cl solution. The organic phase
was separated, dried (MgSO4) and the solvent removed
in vacuo. Purification by column chromatography
(cyclohexane/ethyl acetate: 9/1), resulted in a clear oil.
The enantiomeric excess was determined by chiral
HPLC (Whelk 01 column). The absolute configuration
was determined by polarometric measurements and
compared to the literature values.27
16. Voituriez, A.; Fiaud, J.-C.; Schulz, E. Tetrahedron Lett.
2002, 43, 4907–4909.
17. In the course of our studies, a Japanese patent was
published that claimed the synthesis of analogous com-
pounds and their use as ligands for the copper-induced
cyclopropanation of olefins. Ikehira, H. (Sumitomo
Chemical Co., Ltd., Japan) Jpn. Kokai Tokkyo Koho
(2001) JP 2001139573.
18. Kanemasa, S.; Oderaotoshi, Y.; Yamamoto, H.; Tanaka,
J.; Wada, E.; Curran, D. P. J. Org. Chem. 1997, 62,
6454–6455.
19. Kanemasa, S.; Oderaotoshi, Y.; Sakaguchi, S.-I.;
Yamamoto, H.; Tanaka, J.; Wada, E.; Curran, D. P. J.
Am. Chem. Soc. 1998, 120, 3074–3088.
20. Kanemasa, S.; Oderaotoshi, Y.; Tanaka, J.; Wada, E. J.
Acknowledgements
Am. Chem. Soc. 1998, 120, 12355–12356.
21. Kanemasa, S.; Kanai, T. J. Am. Chem. Soc. 2000, 122,
10710–10711.
The authors are indebted to J. Collin and J.-C. Fiaud
for helpful discussions. They thank the CNRS and the
MENESR for financial support.
22. Go´mez, M.; Jansat, S.; Muller, G.; Maestro, M. A.;
Mah´ıa, J. Organometallics 2002, 21, 1077–1087.
23. Kuehm-Caube`re, C.; Adach-Becker, S.; Fort, Y.;
Caube`re, P. Tetrahedron 1996, 52, 9087–9092.
24. Macaud, M.; Milenkovic, A.; Schulz, E.; Lemaire, M.;
Vrinat, M. J. Catal. 2000, 193, 255–263.
References
25. Iserloh, U.; Curran, D. P.; Kanemasa, S. Tetrahedron:
1. Blaser, H. U.; Spindler, F.; Studer, M. Appl. Catal. A:
General 2001, 22, 119–143.
Asymmetry 1999, 10, 2417–2428.
26. Tye, H.; Eldred, C.; Wills, M. J. Chem. Soc., Perkin
Trans. 1 1998, 457–465.
2. Fache, F.; Schulz, E.; Tommasino, M. L.; Lemaire, M.
Chem. Rev. 2000, 100, 2159–2231.
27. Leutenegger, U.; Umbricht, G.; Fahrni, C.; von Matt, P.;
Pfaltz, A. Tetrahedron 1992, 48, 2143–2156.
28. Mu¨ller, D.; Umbricht, G.; Weber, B.; Pfaltz, A. Helv.
Chim. Acta 1991, 74, 232–240.
29. von Matt, P.; Lloyd-Jones, G. C.; Minidis, A. B. E.;
Pfaltz, A.; Macko, L.; Neuburger, M.; Zehnder, M.;
Ru¨egger, H.; Pregosin, P. S. Helv. Chim. Acta 1995, 78,
265–284.
30. Chelucci, G.; Deriu, S.; Pinna, G. A.; Saba, A.; Valenti,
R. Tetrahedron: Asymmetry 1999, 10, 3803–3809.
31. Bremberg, U.; Rahm, F.; Moberg, C. Tetrahedron: Asym-
metry 1998, 9, 3437–3443.
32. Milenkovic, A.; Schulz, E.; Meille, V.; Loffreda, D.;
Forissier, M.; Vrinat, M.; Sautet, P.; Lemaire, M. Energy
and Fuels 1999, 13, 881–887.
3. Ghosh, A. K.; Mathivanan, P.; Cappiello, J. Tetrahedron:
Asymmetry 1998, 9, 1–45.
4. Benincori, T.; Brenna, E.; Sannicolo`, F.; Trimarco, L.;
Antognazza, P.; Cesarotti, E.; Demartin, F.; Pilati, T. J.
Org. Chem. 1996, 61, 6244–6251.
5. Benincori, T.; Cesarotti, E.; P¨ıccolo, O.; Sannicolo`, F. J.
Org. Chem. 2000, 65, 2043–2047.
6. Tietze, L. F.; Thede, K.; Sannicolo`, F. Chem. Commun.
1999, 1811–1812.
7. Tietze, L. F.; Thede, K.; Schimpf, R.; Sannicolo`, F.
Chem. Commun. 2000, 583–584.
8. Trost, B. M.; Van Vranken, D. L. Chem. Rev. 1996, 96,
395–422.
9. Anderson, J. C.; James, D. S.; Mathias, J. P. Tetrahedron:
Asymmetry 1998, 9, 753–756.
10. Adams, H.; Anderson, J. C.; Cubbon, R.; James, D. S.;
Mathias, J. P. J. Org. Chem. 1999, 64, 8256–8262.
33. Chelucci, G.; Pinna, G. A.; Saba, A.; Valenti, R. Tetra-
hedron: Asymmetry 2000, 11, 4027–4036.