Furanoside thioether-phosphinite ligands for Pd-catalyzed asymmetric allylic substitution reactions: Scope and limitations

Article abstract of DOI:10.1016/j.jorganchem.2005.11.024

A series of readily available thioether-phosphinite ligands has been tested in the Pd-catalyzed allylic substitution reactions of several acyclic and cyclic allylic substrates (S1-S7). This series of ligands have been designed to uncover their important s

Full text of DOI:10.1016/j.jorganchem.2005.11.024

Journal of Organometallic Chemistry 691 (2006) 2257–2262
www.elsevier.com/locate/jorganchem
Furanoside thioether–phosphinite ligands for Pd-catalyzed
asymmetric allylic substitution reactions: Scope and limitations
*
*
Montserrat Di e´ guez , Oscar P a` mies , Carmen Claver
Departament de Qu ı´ mica F ı´ sica i Inorg a` nica, Universitat Rovira i Virgili, C/ Marcel l ı´ Domingo s/n, 43007 Tarragona, Spain
Received 30 September 2005; received in revised form 8 November 2005; accepted 8 November 2005
Available online 5 January 2006
Abstract
A series of readily available thioether–phosphinite ligands has been tested in the Pd-catalyzed allylic substitution reactions of several
acyclic and cyclic allylic substrates (S1–S7). This series of ligands have been designed to uncover their important structural features and
to determine the scope of the thioether–phosphinite ligands in these catalytic reactions. Systematic variation of the electronic and steric
properties at the thioether moiety provide useful information about the ligand parameters that control the enantiodiscrimination. By
carefully selecting the ligand parameters, good enantioselectivities with high activities were obtained for hindered linear substrates S1
and S2 (eeÕs up to 95%) and for unhindered cyclic substrates S4 and S5 (eeÕs up to 91%).
ꢀ
2005 Elsevier B.V. All rights reserved.
Keywords: Palladium; Asymmetric catalysis; Allylic alkylation; Phosphinite–thioether ligands; Carbohydrate
1
. Experimental
red for 30 min. Subsequently, a solution of substrate
0.5 mmol) in dichloromethane (1.5 mL), dimethyl malon-
ate (171 lL, 1.5 mmol), N,O-bis(trimethylsilyl)-acetamide
370 lL, 1.5 mmol) and KOAc (5 mg) were added. The
(
1
.1. General considerations
(
All reactions were carried out using standard Schlenk
reaction mixture was stirred at room temperature. After
the desired reaction time, the reaction mixture was diluted
with Et O (5 mL) and a saturated NH Cl (aq) (25 mL) was
techniques under an atmosphere of argon. Solvents were
purified and dried by standard procedures. Thioether–
phosphinite ligands 1–7 were prepared using previously
described methods [1]. Racemic substrates S1–S7 were pre-
pared as previously reported [2–5]. NMR spectra were
recorded using a 400 MHz spectrometer. Chemical shifts
2
4
added. The mixture was extracted with Et O (3 · 10 mL)
2
and the extract dried over MgSO . For substrate S1, con-
4
1
version was measured by H NMR and enantiomeric
excess was determined by HPLC (Chiralcel-OD, 0.5% 2-
1
13
are relative to that of SiMe ( H and C) as internal stan-
propanol/hexane, flow 0.5 mL/min). For substrate S2, con-
4
3
1
1
dard or H PO ( P) as external standard.
version was measured by H NMR and enantiomeric
3
4
1
excess was determined by H NMR using Eu(hfc)3 as
1.2. Typical procedure of allylic alkylation of disubstituted
linear substrates S1–S3
resolving agent. For substrate S3, conversion and enantio-
meric excess were determined by GC.
A
degassed solution of [Pd(p-C H )Cl] (0.9 mg,
1.3. Typical procedure of allylic alkylation of cyclic
substrates S4 and S5
3
5
2
0
.0025 mmol) and the corresponding thioether–phosphi-
nite (0.0055 mmol) in dichloromethane (0.5 mL) was stir-
A
degassed solution of [Pd(p-C H )Cl] (0.9 mg,
3 5 2
*
0.0025 mmol) and the corresponding thioether–phosphi-
nite ligand (0.0055 mmol) in dichloromethane (0.5 mL)
Corresponding author.
E-mail address: dieguez@quimica.urv.es (M. Di e´ guez).
0
022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.11.024

2258
M. Di e´ guez et al. / Journal of Organometallic Chemistry 691 (2006) 2257–2262
was stirred for 30 min. Subsequently, a solution of sub-
strate (0.5 mmol) in dichloromethane (1.5 mL), dimethyl
malonate (171 lL, 1.5 mmol), N,O-bis(trimethylsilyl)-acet-
amide (370 lL, 1.5 mmol) and KOAc (5 mg) were added.
The reaction mixture was stirred at room temperature.
After the desired reaction time, the reaction mixture was
diluted with Et O (5 mL) and a saturated NH Cl (aq)
have played a dominant role among the heterodonor
ligands. To a lesser extent, phosphorus–thioether ligands
have also demonstrated their potential utility in Pd-cata-
lyzed asymmetric allylic substitution [7,8]. In this context
only two families of thioether–phosphinite ligands have
been developed for this process [1a,8]. Evans and cowork-
ers have therefore designed the first family of thioether–
phosphinite ligands that proved to be effective [8]. Their
results indicated a remarkable steric effect of the thioether
moiety on enantioselectivity. More recently, we have
reported the first carbohydrate-based thioether–phosphi-
nite ligands for the Pd-catalyzed asymmetric allylic alkyl-
ation reactions (ligands 1–3, Fig. 1) [1a]. Results were
only satisfactory for the allylic substitution of rac-1,3-
diphenyl-3-acetoxyprop-1-ene (eeÕs up to 93%) while
enantioselectivities were low for the cyclic substrate rac-3-
acetoxycyclohexene (eeÕs up to 51%) [1a]. Furthermore, in
contrast to EvansÕ work an unclear effect of the thioether
substituents on enantioselectivity were observed. In light
of this, in this paper we expand the previous study to other
furanoside thioether–phosphinite ligands (Fig. 1, ligands
4–7 [9]) in which the steric and electronic properties of
the thioether moiety were systematically varied. This
allowed us to investigate the possibility of an steric and/
or electronic effect of the sulfur substituent in enantioselec-
tivity. Therefore, with ligands 1–4 and 7 we mainly investi-
gated the steric effect on enantioselectivity, while with
ligands 1, 5 and 6 we studied how the electronic properties
affected enantioselectivity. This study will therefore provide
more information about the parameters that control
enantioselectivity and, at the same time, may help us to
increase the efficiency of our catalytic system. We also
extended the study to other types of substrates with differ-
ent steric properties (S1–S7) to determine the scope of this
type of ligands.
2
4
(
25 mL) was added. The mixture was extracted with Et O
2
(
3 · 10 mL) and the extract dried over MgSO . For sub-
4
strate S4, conversion and enantiomeric excess were deter-
mined by GC using a FS-Cyclodex b-I/P 25 m column,
internal diameter 0.2 mm, film thickness 0.33 mm, carrier
gas: 100 kPa He, F.I.D. detector. For substrate S5, conver-
sion was determined by GC and enantiomeric excess was
1
determined by H NMR using Eu(hfc) .
3
1.4. Typical procedure of allylic alkylation of
monosubstituted linear substrates S6 and S7
A
degassed solution of [Pd(p-C H )Cl] (1.8 mg,
3 5 2
0
.005 mmol) and the corresponding thioether–phosphinite
ligand (0.011 mmol) in dichloromethane (0.5 mL) was stir-
red for 30 min. Subsequently, a solution of substrate
(
0.5 mmol) in dichloromethane (1.5 mL), dimethyl malon-
ate (171 lL, 1.5 mmol), N,O-bis(trimethylsilyl)-acetamide
370 lL, 1.5 mmol) and KOAc (5 mg) were added. The
(
reaction mixture was stirred at room temperature. After
the desired reaction time, the reaction mixture was diluted
with Et O (5 mL) and a saturated NH Cl (aq) (25 mL) was
2
4
added. The mixture was extracted with Et O (3 · 10 mL)
2
and the extract dried over MgSO . Solvent was removed
4
1
and conversion and regioselectivity were measured by H
NMR. To determine the ee by HPLC (Chiralcel-OJ, 3%
2
-propanol/hexane, flow 0.7 mL/min), a sample was fil-
tered over basic alumina using dichloromethane as the
eluent.
3. Results and discussion
2
. Introduction
3.1. Allylic alkylation of disubstituted linear substrates
Palladium-catalyzed allylic substitution is a useful syn-
thetic method for the formation of carbon–carbon and car-
bon–heteroatom bonds [6]. Most of the chiral ligands
developed for asymmetric allylic substitution are mixed
bidentate donor ligands (such as P–N, P–S and N–S) [6].
The efficiency of this type of hard–soft heterodonor ligands
has been mainly attributed to the different electronic effects
of the donor atoms. Mixed phosphorus–nitrogen ligands
In this section, we report the use of the chiral thioether–
phosphinite ligands 1–7 in the Pd-catalyzed allylic alkyl-
ation (Eq. (1)) of three linear substrates with different
steric properties: rac-1,3-diphenyl-3-acetoxyprop-1-ene S1
(widely used as a model substrate), rac-(E)-ethyl-2,5-
dimethyl-3-hex-4-enylcarbonate S2 and rac-1,3-dimethyl-
3-acetoxyprop-1-ene S3. In all the cases, the catalysts were
Ph
Ph
P
1
R = Ph
R = Me
5
6
7
R = 4-Me-C H
6 4
R
S
O
O
2
3
4
R = 4-CF -C H
3 6 4
i
R = Pr
R = 2,6-di-Me-C H
6 3
O
O
t
R = Bu
Fig. 1. Thioether–phosphinite ligands 1–7.

M. Di e´ guez et al. / Journal of Organometallic Chemistry 691 (2006) 2257–2262
2259
generated in situ from p-allyl-palladium chloride dimer
Pd(p-C H )Cl] and the corresponding ligand. The nucle-
ophile was generated from dimethyl malonate in the pres-
Table 2
a
Pd-catalyzed allylic alkylation of S1 with ligands 1–7
[
3
5
2
b
c
Entry
Ligand
% Conv. (min)
% ee
ence of N,O-bis(trimethylsilyl)-acetamide (BSA).
1
2
3
4
5
6
7
1
2
3
4
5
6
7
4
90 (30)
91 (5)
47 (S)
61 (S)
86 (S)
87 (S)
59 (S)
39 (S)
68 (S)
95 (S)
X
CH (COOMe) / BSA
CH(COOMe)2
2
2
100 (30)
100 (15)
100 (30)
100 (30)
100 (30)
48 (120)
R
R
R
R
[
Pd(π-C H )Cl] / 1 - 7
3 5 2
S1 R= Ph; X= OAc
S2 R = Pr; X= OCOOEt
S3 R= Me; X= OAc
8
9
R= Ph
R= Pr
10 R= Me
i
i
d
8
a
0
.5 mol% [Pd(p-C
3
H
5
)Cl]
(COOMe)
BSA), KOAc (5 mg), room temperature and CH
2
,
1.1 mol% ligand, room temperature,
and N,O-bis(trimethylsilyl)acetamide
Cl as solvent.
H NMR. Reaction time in minutes shown in
ð1Þ
3
(
0 min; 3 equiv of CH
2
2
2
2
b
1
Measured by
parentheses.
3
1
.1.1. Allylic alkylation of rac-1,3-diphenyl-3-acetoxyprop-
-ene S1 (Eq. (1))
c
Determined by HPLC (Chiralcel OD). Absolute configuration drawn
in parentheses.
We first investigated the Pd-catalyzed allylic substitution
of rac-1,3-diphenyl-3-acetoxyprop-1-ene S1 with dimethyl
d
Reaction carried out at ꢀ5 ꢁC.
malonate using the chiral thioether–phosphinite ligands
Using ligands 1, 5 and 6, with different substituents at
the para positions of the thiophenyl group, we studied
how the electronic properties of the ligand affects enanti-
oselectivity. The results indicated that enantioselectivity
was best when electron-donating thioether substituents
were present. The use of ligand 5 with a methyl substituent
in the para position of the phenyl thioether moiety there-
fore provided 8 in 59% ee (entry 5), while ligand 6 with
an electron-withdrawing group yielded 8 in only 39% ee
1
–7 (Eq. (1)).
Preliminary investigations into the solvent effect, ligand-
to-palladium ratio and source of base with the new ligand 4
provided the same trends as those with the previously
tested ligands 1–3 [1a] in substrate S1 (Table 1). The
trade-off between enantioselectivities and reaction rates
was therefore optimum with dichloromethane, a ligand-
to-palladium ratio of 1.1 and KOAc as base.
The results of using the rest of the furanoside thioether–
phosphinite ligands (1–7) under the optimized conditions
are showed in Table 2. Our results show that the enantio-
meric excesses are dependent on both the electronic and
steric properties of the substituents in the thioether moiety.
This behavior contrasts with the results obtained with the
EvansÕ thioether–phosphinite ligands, for which enantiose-
lectivities were only affected by the steric properties of the
thioether moiety [8b].
(entry 6).
With ligands 1–4 and 7 we mainly studied how the steric
properties of the ligand affects the product outcome. If we
compare the results obtained with these ligands we can
conclude that enantioselectivities were best when bulky thi-
oether substituents were present (entries 1–4 and 7). Thus,
the best enantioselectivities were obtained with ligands 3
and 4 that contains bulky isopropyl and tert-butyl groups
in the thioether moiety. The fact that ligand 2, with a small
methyl substituent, provided higher enantioselectivity than
ligand 1, with a phenyl substituent, can be explained by the
electronic effect previously mentioned.
Table 1
Effect of the solvent, ligand-to-palladium ratio and source of base in the
a
Pd-catalyzed allylic alkylation of S1 using ligand 4
To sum up, the best enantioselectivities were obtained
with the previously tested ligand 3 [1a] and the new ligand
b
c
Entry
Solvent
CH Cl
DMF
THF
Base
% Conv. (min)
% ee
4
, that contain electron-donating and bulky substituents at
1
2
3
4
5
6
7
2
2
KOAc
KOAc
KOAc
KOAc
100 (15)
100 (5)
55 (30)
29 (30)
72 (20)
96 (15)
92 (30)
98 (15)
100 (20)
87 (S)
78 (S)
83 (S)
87 (S)
86 (S)
84 (S)
82 (S)
79 (S)
86 (S)
the thioether moiety.
We also studied how the temperature affected the out-
come of the reaction using ligand 4. Lowering the reaction
temperature to ꢀ5 ꢁC, enantioselectivity increased to up to
Toluene
CH
CH
CH
CH
CH
2
2
2
2
2
Cl
Cl
Cl
Cl
Cl
2
2
2
2
2
K
2
CO
NaOAc
Li CO
3
95% (S) (entries 8 vs. 4).
2
3
d
8
e
KOAc
KOAc
9
3.1.2. Allylic alkylation of rac-(E)-ethyl-2,5-dimethyl-3-
hex-4-enylcarbonate S2 (Eq. (1))
We also evaluated the thioether–phosphinite ligands 1–7
in the allylic substitution process of S2 using dimethyl mal-
onate as nucleophile (Eq. (1)). This substrate is more steri-
cally demanding than the previously used substrate S1
a
0
.5 mol% [Pd(p-C
3
H
5
)Cl]
(COOMe)
BSA), 5 mg of the corresponding base, room temperature.
2
,
1.1 mol% ligand, room temperature,
3
(
0 min; 3 equiv of CH
2
2
and N,O-bis(trimethylsilyl)acetamide
b
1
Measured by
parentheses.
H NMR. Reaction time in minutes shown in
c
Determined by HPLC (Chiralcel OD). Absolute configuration drawn
in parentheses.
[
6,8b]. The most remarkable results are shown in Table 3.
d
In general, they follow the same trends as for the allylic
alkylation of S1. However, the enantiomeric excesses were
L/Pd = 2.
L/Pd = 0.9.
e

2260
M. Di e´ guez et al. / Journal of Organometallic Chemistry 691 (2006) 2257–2262
Table 3
tions directed to increase the steric bulk of the ligand could
provided better results.
a
Pd-catalyzed allylic alkylation of S2 with ligands 1–7
b
c
Entry
Ligand
% Conv. (h)
% ee
1
2
3
4
5
6
7
a
1
2
3
4
5
6
7
100 (18)
100 (18)
100 (18)
100 (18)
100 (18)
94 (18)
52 (R)
65 (R)
88 (R)
90 (R)
63 (R)
45 (R)
68 (R)
3.2. Allylic alkylation of cyclic substrates
As for the unhindered substrate S3, enantioselectivity in
cyclic substrates is difficult to control, mainly because of
the presence of less sterically syn substituents. These syn
substituents are thought to play a crucial role in the enan-
tioselection observed with acyclic substrates in the corre-
sponding Pd-allyl intermediate [6]. To obtain high eeÕs,
the ligand must create a small chiral pocket (the chiral cav-
ity where the allyl is embedded) around the metal center [6].
In this section, we report the use of the chiral thioe-
ther–phosphinite ligands 1–7 in the Pd-catalyzed allylic
alkylation of two cyclic substrates (Eq. (2)): rac-3-acet-
oxycyclohexene S4 (which is widely used as a model
substrate) and rac-3-acetoxycycloheptene S5.
100 (18)
0
.5 mol% [Pd(p-C
3
H
5
)Cl]
2
,
1.1 mol% ligand, room temperature,
and N,O-bis(trimethylsilyl)acetamide
Cl as solvent.
Measured by H NMR. Reaction time in hours shown in parentheses.
30 min; 3 equiv of CH
2
(COOMe)
2
(
BSA), KOAc (5 mg), room temperature and CH
2
2
b
1
c
1
3
Enantiomeric excesses determined by H NMR using Eu(hfc) .
Absolute configuration drawn in parentheses.
slightly higher (eeÕs up to 90% at room temperature). As
expected, the activities were lower than in the alkylation
reaction of S1 [6,8b]. Again, the catalyst precursor contain-
ing the thioether–phosphinite ligands 3 and 4 provided the
best enantioselectivity (entries 3 and 4). The stereoselectiv-
ity of the alkylation of S2 was the same as for the alkyl-
ation reaction of S1, though the CIP descriptor was
inverted due to the change in priority of the groups.
OAc
CH(COOMe)2
*
CH (COOMe) / BSA
2
2
[
Pd(π-C H )Cl] / 1 - 7
3 5 2
ð2Þ
n
n
S4 n=1
S5 n=2
11 n=1
12 n=2
3
1
.1.3. Allylic alkylation of rac-1,3-dimethyl-3-acetoxyprop-
-ene S3 (Eq. (1))
We also screened ligands 1–7 in the allylic alkylation of
We initially studied the allylic alkylation of rac-3-acet-
oxycyclohexene S4 using ligands 1–7. Preliminary investi-
gations into the solvent effect, ligand-to-palladium ratio
and source of base provided the same trends as those with
the previously tested linear substrate S1. The trade-off
between enantioselectivities and reaction rates was there-
fore optimum with dichloromethane, a ligand-to-palladium
ratio of 1.1 and KOAc as base. The results of using the
thioether–phosphinite ligands 1–7 under the optimized
conditions are showed Table 5.
the linear substrate S3 (Eq. (1)). This substrate is less steri-
cally demanding than the previously used substrates S1 and
S2. Enantioselectivity for S3 is therefore more difficult to
control than with hindered substrates such as S1 and S2
[
6]. The results of using the thioether–phosphinite ligands
under the optimized conditions are summarized in Table
. The general trends that controlled enantioselectivity
4
were different from those that controlled S1 and S2, being
the steric effect more predominantly. Therefore, the best
enantioselectivity was obtained with ligand 4. Unfortu-
nately, as observed with the EvansÕ thioether–phosphinite
ligands [8b], our ligand systems also proved to be inade-
quate in terms of enantioselectivities. So, further modifica-
Table 5
a
Pd-catalyzed allylic alkylation of S4 and S5 with ligands 1–7
b
c
Entry
Substrate
Ligand
% Conv. (min)
% ee
1
2
3
4
5
6
7
S4
S4
S4
S4
S4
S4
S4
S4
S4
S5
S5
1
2
3
4
5
6
7
3
4
4
4
98 (30)
100 (30)
100 (30)
50 (30)
32 (30)
8 (60)
12 (30)
98 (90)
12 (360)
98 (60)
10 (360)
4 (R)
21 (R)
41 (R)
75 (S)
28 (R)
2 (R)
24 (R)
51 (R)
89 (S)
79 (S)
91 (S)
Table 4
a
Pd-catalyzed allylic alkylation of S3 with ligands 1–7
b
c
Entry
Ligand
% Conv. (min)
% ee
1
2
3
4
5
6
7
a
1
2
3
4
5
6
7
100 (30)
100 (15)
100 (15)
100 (15)
100 (30)
100 (30)
100 (30)
7 (R)
4 (R)
18 (R)
33 (R)
9 (R)
d
8
e
9
10
11
e
5 (R)
13 (R)
a
0
.5 mol% [Pd(p-C
3
H
5
)Cl]
2
,
1.1 mol% ligand, room temperature,
and N,O-bis(trimethylsilyl)acetamide
3
0 min; 3 equiv of CH
2
(COOMe)
2
0
.5 mol% [Pd(p-C
0 min; 3 equiv of CH
BSA), KOAc (5 mg), room temperature and CH
Measured by GC. Reaction time in minutes shown in parentheses.
Enantiomeric excesses determined by GC. Absolute configuration
3
H
5
)Cl]
2
,
1.1 mol% ligand, room temperature,
and N,O-bis(trimethylsilyl)acetamide
Cl as solvent.
(BSA), KOAc (5 mg), room temperature and CH Cl as solvent.
2
2
b
3
2
(COOMe)
2
Measured by GC. Reaction time in minutes shown in parentheses.
Enantiomeric excesses determined by GC. Absolute configuration
c
(
2
2
b
drawn in parentheses.
c
d
Reaction carried out at 0 ꢁC.
Reaction carried out at ꢀ20 ꢁC.
e
drawn in parentheses.

M. Di e´ guez et al. / Journal of Organometallic Chemistry 691 (2006) 2257–2262
2261
As observed for substrates S1 and S2, enantioselectivi-
ties were affected by the electronic and steric properties
of the substituents at the thioether moiety. However, the
effect of these parameters was different. Enantioselectivity
was therefore best only with ligand 4, containing a bulky
tert-butyl thioether substituent. It is interesting to note that
ligand 4 affords the opposite sense of induction to the rest
of ligands tested (entries 4 vs. 1–3 and 5–7). Therefore with
the new ligand 4, we were able to increase enantioselectivity
from the previously reported 51% [1a] ee to 89% ee (entries
sense of enantioselectivity was also observed. Thus, ligands
containing electron-withdrawing thioether groups favor
the (S) enantiomer (entries 1, 6 and 7), while ligands con-
taining electron-donor groups favor the formation of the
(R) enantiomer (entries 2–5).
In contrast, regioselectivity was mainly dependent on
the electronic properties of the thioether moiety and the
0
size of the substrate substituent (R ). Electron-withdrawing
substituents in the thioether moiety of the ligand [11] (entry
0
6 vs. 5) along with bulky R substituents in the substrate
8
vs. 9). This result is amongst the best reported for cyclic
(entry 4 vs. 8) therefore favored the formation of the
desired branched product.
substrates [6,8b].
Ligand 4 were also effective (eeÕs up to 91%) in the allylic
alkylation of the seven-membered ring substrate S5 (entries
4. Conclusions
1
0 and 11).
A series of readily available thioether–phosphinite
ligands has been tested in the Pd-catalyzed allylic substitu-
tion reactions of several acyclic and cyclic allylic substrates
3
.3. Allylic substitution of monosubstituted linear substrates
(S1–S7). This series of ligands have been designed to
Finally, we also examined the regio- and stereoselective
uncover their important structural features and to deter-
mine the scope of the thioether–phosphinite ligands in
these catalytic reactions. Systematic variation of the elec-
tronic and steric properties at the thioether moiety provide
useful information about the ligand parameters that con-
trol the enantiodiscrimination. Particularly, for the hin-
dered disubstituted acyclic substrates S1 and S2, we
found that enantioselectivities depended on the steric and
electronic properties of the substituent of the thioether
moiety. Good enantioselectivies with high activities were
obtained with ligands 3 and 4, containing bulky and elec-
tron-donating isopropyl and tert-butyl substituents, respec-
tively (eeÕs up to 95%). For the unhindered linear substrate
S3, enantioselectivities were mainly affected by the steric
properties of the thioether moiety. Enantioselectivity was
therefore best using ligand 4 that contains the more bulky
sulfur substituent (eeÕs up to 33%). For the unhindered cyc-
lic substrates S4 and S5, enantioselectivites followed simi-
lar trends as those for the alkylation of substrates S1 and
S2, but an interesting steric effect on the sense of enantiose-
lectivity was observed. Therefore, with the new ligand 4 we
were able to increase enantioselectivities until 89% ee and
91% ee for substrates S4 and S5, respectively. For the mon-
osubstituted acyclic substrates, these ligands proved to be
inadequate in terms of regioselectivities. However, we
obtained good enantioselectivty by carefully selecting the
electronic and steric properties of the thioether moiety
allylic alkylation of 1-(1-naphthyl)allyl acetate S6 and 1-
phenylallyl acetate S7 with dimethyl malonate. For both
substrates, the development of highly regio- and enantio-
selective Pd-catalysts is still a challenge [10]. The results
obtained with the furanoside thioether–phosphinite
ligands 1–7 are summarized in Table 6. Unfortunately,
the regioselectivity for the branched products was not
high. However, good enantioselectivities can be obtained
(
eeÕs up to 79% for substrate S6 and 74% for substrate
S7).
From the results presented in Table 6, we can conclude
that enantioselectivity was affected on both electronic and
steric properties of the substituents at the thioether moiety.
Enantioselectivities were therefore best when bulky tert-
butyl groups in the thioether moiety (ligand 4) were present
(
entries 4 and 8). In addition, an electronic effect on the
Table 6
a
Selected results for the Pd-catalyzed allylic alkylation of S6 and S7
OAc
CH (COOMe)2 / BSA
CH(COOMe)2
+
2
R'
CH(COOMe)2
R'
R'
[
Pd(π-C3H5)Cl]2 / L
S6 R'= 1-Naphthyl
S7 R'= Phenyl
b
l
b
b
c
Entry
Substrate
Ligand
% Conv. (t/h)
b/l
% ee
1
2
3
4
5
6
7
8
a
S6
S6
S6
S6
S6
S6
S6
S7
1
2
3
4
5
6
7
4
100 (2)
100 (2)
100 (2)
100 (2)
100 (2)
100 (2)
100 (2)
100 (2)
50/50
34/66
31/69
30/70
30/70
60/40
50/50
7/93
4 (R)
11 (S)
34 (S)
79 (S)
4 (S)
8 (R)
8 (S)
74 (S)
(eeÕs up to 79%).
References
[
1] (a) E. Guimet, M. Di e´ guez, A. Ruiz, C. Claver, Tetrahedron:
1
mol% [Pd(p-C
3
H
5
)Cl]
2
, 2.2 mol% ligand, room temperature, 30 min;
Asymmetry 16 (2005) 959;
3
equiv of CH (COOMe)
2
2
and N,O-bis(trimethylsilyl)acetamide (BSA),
(b) M. Di e´ guez, O. P a` mies, C. Claver, Tetrahedron: Asymmetry 16
(2005) 3877.
[2] P.R. Auburn, P.B. Mackenzie, B. Bosnich, J. Am. Chem. Soc. 107
KOAc (5 mg), room temperature and CH
2
Cl
Conversion percentage and linear-to-branched ratio determined by H
NMR.
2
as solvent.
b
1
(1985) 2033.
c
Enantiomeric excesses determined by HPLC on a Chiralcel-OJ col-
umn. Absolute configuration drawn in parentheses.
[3] C. Jia, P. M u¨ ller, H. Mimoun, J. Mol. Cat. A: Chem. 101 (1995)
127.

2262
M. Di e´ guez et al. / Journal of Organometallic Chemistry 691 (2006) 2257–2262
[
[
4] J. Lehman, G.C. Lloyd-Jones, Tetrahedron 51 (1995) 8863. [8] (a) D.A. Evans, J.R. Campos, J.R. Tedrow, F.E. Michael, M.R.
5] T. Hayashi, A. Yamamoto, Y. Ito, E. Nishioka, H. Miura, K.
Yanagi, J. Am. Chem. Soc. 111 (1989) 6301.
Gagn e´ , J. Org. Chem. 64 (1999) 2994;
(b) D.A. Evans, J.R. Campos, J.R. Tedrow, F.E. Michael, M.R.
Gagn e´ , J. Am. Chem. Soc. 122 (2000) 7905.
[
6] For representative reviews, see (a) J. Tsuji, Palladium Reagents and
Catalysis, Innovations in Organic Synthesis, Wiley, New York, 1995;
[9] Ligands 4–7 have recently been tested in the Rh-catalyzed hydrosyl-
ilation of ketones, see Ref. [1b].
[10] (a) For recent successful applications of Pd-catalysts, see S.-L. You,
X.-Z. Zhu, Y.-M. Luo, X.-L. Hou, L.-X. Dai, J. Am. Chem. Soc. 123
(2001) 7471;
(
(
(
(
b) B.M. Trost, D.L. van Vranken, Chem. Rev. 96 (1996) 395;
c) M. Johannsen, K.A. Jorgensen, Chem. Rev. 98 (1998) 1689;
d) A. Pfaltz, M. Lautens, in: E.N. Jacobsen, A. Pfaltz, H. Yamamoto
Eds.), Comprehensive Asymmetric Catalysis, vol. 2, Springer-Verlag,
Berlin, 1999 (Chapter 24);
(b) R. Pr e´ t oˆ t, A. Pfaltz, Angew. Chem. Int. Ed. 37 (1998) 323;
(c) R. Hilgraf, A. Pfaltz, Synlett (1999) 1814;
(d) O. P a` mies, M. Di e´ guez, C. Claver, J. Am. Chem. Soc. 127 (2005)
3646.
(
(
e) B.M. Trost, M.L. Crawley, Chem. Rev. 103 (2003) 2921;
f) G. Helmchen, A. Pfaltz, Acc. Chem. Res. 33 (2000) 336.
[
7] (a) A.M. Masdeu-Bult o´ , M. Di e´ guez, E. Martin, M. G o´ mez,
Coordin. Chem. Rev. 242 (2003) 159;
[11] As reported in the literature for P/N ligands the introduction of
electron-withdrawing substituents in the ligand render the Pd center
(
b) D. Enders, R. Peters, R. Lochtman, G. Raabe, J. Runsink, J.W.
Bats, Eur. J. Org. Chem. (2000) 3399;
c) A. Albinati, P.S. Pregosin, K. Wick, Organometallics 15 (1996)
419;
d) H. Nakano, Y. Okuyama, M. Yanagida, H. Hongo, J. Org.
Chem. 66 (2001) 620.
N
more electrophilic enhancing the S 1 character and therefore the
formation of the branched product is favored. See for example: (a)
Ref. [10];
(b) R. Pr e´ t oˆ t, G.C. Loyd-Jones, A. Pfaltz, Pure Appl. Chem. 70
(1998) 1035.
(
2
(