Novel and efficient chiral sulfideoxathiane ligands for palladium-catalyzed asymmetric allylic alkylation

Article abstract of DOI:10.1039/b211031h

Easily prepared, chiral sulfideoxathiane ligands are described which give excellent enantioselectivity (up to 99% ee) in the Pd-catalyzed allylic alkylation of 1,3-diphenyl-2-propenyl acetate with a range of alkyl malonate nucleophiles.

Full text of DOI:10.1039/b211031h

Novel and efficient chiral sulfideoxathiane ligands for
palladium-catalyzed asymmetric allylic alkylation†
Yuko Okuyama,a Hiroto Nakano,*a Kouichi Takahashi,a Hiroshi Hongoa and Chizuko Kabuto*b
a Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981-8558, Japan.
E-mail: hnakano@tohoku-pharm.ac.jp; Fax: 81 22 275 2013; Tel: 81 22 234 4181
b Instrumental Analysis Center for Chemistry, Graduate School of Science, Tohoku University, Aoba,
Aramaki, Aoba-ku, Sendai 980-8578, Japan. E-mail: kabuto@kiki.chem.tohoku.ac.jp
Received (in Cambridge, UK) 8th November 2002, Accepted 2nd January 2003
First published as an Advance Article on the web 17th January 2003
Easily prepared, chiral sulfideoxathiane ligands are de-
scribed which give excellent enantioselectivity (up to 99% ee)
in the Pd-catalyzed allylic alkylation of 1,3-diphenyl-
2-propenyl acetate with a range of alkyl malonate nucleo-
philes.
product 11a; the results are summarized in Table 1. Initially,
chiral ligands 5–8 (2 mol%) were tested at room temperature.
The ligands showed excellent reactivity, but the enantioselectiv-
ity greatly depended on the individual ligand structure. Thus,
ligands 5 and 7 containing a linking phenylthio moiety gave the
corresponding product 11a in low enantiopurity (5: 57% ee, 7:
49% ee), whereas ligands 6 and 8, incorporating a bulkier linked
2,6-dimethylphenylthio moiety, brought about high asymmetric
induction (6: 94% ee, 8: 75% ee) (entries 1–4). In particular,
chiral ligand 6 afforded 11a in high levels of enantiomeric
excess (94% ee) at room temperature, while reaction at 0 °C
improved enantioselectivity to 98% ee (entry 5).
Carbon–carbon bond formation is one of the most important
reactions in synthetic organic chemistry. One useful and
popular method is palladium-catalyzed allylation,1 and asym-
metric versions of this reaction have also been extensively
studied over the last decade.1 Strategies for controlling
enantioselectivity in Pd-catalyzed asymmetric reactions have
depended on the design and application of chiral ligands.
Although many of the efficient homo- and hetero-donor chiral
ligands such as N–N (e.g. bisoxazolines2), P–P (e.g. Trost’s P–P
ligands3), N–P (phosphinooxazoline4), and S–P (Evans S–P
ligands and our phosphinooxathianes5) types have been ex-
ploited and utilized, the S–S type ligand has not, in spite of
having advantages such as lower cost, toxicity and oxidation
potential. To the best of our knowledge, only one example
employing C2-symmetric S–S type ligands in the allylic
alkylation has been reported,6 by Gómez and co-workers, but
this only afforded modest asymmetric induction (up to 81% ee)
owing to the donor sites being insufficiently different for
discrimination between both terminal allylic carbons in the
intermediate.6 We planned to synthesize the asymmetric S–S
type ligands 5–8 having a borneol backbone because the ligand
can be prepared easily from the reactions of mercaptoisoborneol
or mercaptoborneol with phenylthiobenzaldehydes and because
the lack of C2-symmetry in the ligand may give rise to more than
one intermediate complex whose reactivities determine the
enantioselection. Herein, we wish to report that the easily
prepared S–S type sulfideoxathiane ligand 6 showed dramatic
reactivity and enantioselectivity (up to 99% ee) in all cases of
the Pd-catalyzed allylic alkylation of 1,3-diphenyl-2-propenyl
acetate 9 with dimethyl and dialkyl methylmalonate nucleo-
philes 10a–c. This is the first time that the allylic alkylation has
been catalyzed with excellent enantioselectivity by a chiral
homo-donor S–S type ligand.
Scheme 1
Table 1 Asymmetic Pd-catalyzed allylation of acetate 9
Ligand
Entrya (mol%)
Temp./°C
(Time/h)
Yieldc
(%)
Eed (%)
(Config.f)
R1
R2
1
2
3
4
5 (2)
6 (2)
7 (2)
8 (2)
6 (2)
6 (5)
6 (1)
6 (0.5)
6 (2)
6 (2)
H
H
H
H
H
H
H
H
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
rt (12)
rt (15)
rt (9)
rt (13)
0 (48)
0 (48)
0 (120)
0 (144)
0 (48)
0 (48)
100
100
100
100
92
100
40
38
57 (R)
94 (R)
49 (S)
75 (S)
98 (R)
93 (R)
94 (R)
92 (R)
96e (S)
99e (S)
The requisite chiral ligands 5–8 were easily prepared by the
condensation of commercially available (1S)-(2)-10-mercap-
toisoborneol 1 or (1S)-(2)-10-mercaptoborneol 2 with 2-(phe-
nylthio)- or 2-(2,6-dimethylphenylthio)benzaldehydes (3 and
4)7 in good yields (86–97%) (Scheme 1). In all four cases (5–8),
the assigned stereochemistry at the a-position of the 1,3-ox-
athiane ring was determined by NOE difference spectra
(NOEDS). NOE enhancement was observed between the
hydrogen at the a-position and the hydrogen at the b-position
when the a- and b-positions were irradiated, respectively
(Scheme 1).5b–d
5
6b
7b
8b
9
Me
Me
96
100
10
3
a Molar ratio for entries 1–5 and 9, 10: [PdCl(h -C3H5)]2 (0.01 equiv.),
malonates (3 equiv.), N,O-bis(trimethylsilyl)acetamide (BSA) (3 equiv.),
potassium acetate (0.02 equiv.), 9 (1 equiv.), ligands 5–8 (0.02 equiv.).
3
b Molar ratio for entries 6–8: [PdCl(h -C3H5)]2 (5 mol%: 0.025 equiv., 0.1
mol%: 0.005 equiv., 0.5 mol%: 0.0025 equiv.), dimethyl malonate (3
equiv.), BSA (3 equiv.), potassium acetate (0.02 equiv.), 9 (1 equiv.), ligand
6 (5 mol%: 0.05 equiv., 1 mol%: 0.0105 equiv., 0.5 mol%: 0.005 equiv.).
c Isolated yields. d Determined by chiral HPLC using a Daicel OD-H
column. e Determined by chiral HPLC using a Daicel (OD-H + OD-H)
column. f R or S configuration based on the specific rotation with literature
data.1e,f
The Pd-catalyzed allylic alkylation of 1,3-diphenyl-2-prope-
nyl acetate 9 with dimethyl malonate 10a using chiral ligands
3
5–8 was examined in the presence of [PdCl(h -C3H5)]2 and
N,O-bis(trimethylsilyl)acetamide (BSA)8 to give the allylation
† Electronic supplementary information (ESI) available: experimental
details. See http://www.rsc.org/suppdata/cc/b2/b211031h/
524
CHEM. COMMUN., 2003, 524–525
This journal is © The Royal Society of Chemistry 2003

To optimize reaction conditions, we next examined the effect
of the molar ratio of ligand 6 at 0 °C. The use of 5 mol% of 6
brought about a slight decrease in enantioselectivity (93% ee)
(entry 6). At low catalytic loadings (1 mol% and 0.5 mol%) the
reactions gave good levels of enantioselectivity (1 mol%: 94%
ee and 0.5 mol%: 92% ee), albeit in low chemical yields (entries
7 and 8). From these results, the most effective set of reaction
conditions was given by 2 mol% of ligand 6 at 0 °C.
We also examined the reactions of acetate 9 with bulkier
dimethyl- and diethyl methylmalonates 10b and 10c as
nucleophiles under the optimized reaction conditions. The
reaction with 10b gave the corresponding product 11b in
satisfactory enantiomeric excess (96% ee) and the chemical
yield (96%) (entry 9). Further, the bulkiest malonate 10c
achieved near complete stereocontrol (99% ee) with quantita-
tive yield to give the product 11c, which has been difficult to
secure in high optical purity.4a,b
Fig. 1 Optimized structures of (a) 12-A1 and (b) 14-A1.
proceeds via selective conformers. Furthermore, the calcula-
tions for the final palladium–olefin complexes of ligand 6 show
that 14-A1 is preferred by more than 4 kcal mol21 over 14-A2.
Thus, MO calculations give a rationale for high ee for 6 and low
ee for 5 and the optimized structures (Fig. 1) indicate that the
steric hindrance of dimethyl groups attached to the phenyl ring
of 6 controls the conformation.
In conclusion, the developed sulfideoxathiane ligand 6 was
prepared easily in one step and showed dramatic reactivity and
enantioselectivity for the allylic alkylation of acetate 9 with
three kinds of malonates (96–100%, 96–99% ee), comparable to
the results of the Evans group.5a As another advantage, the
ligand 6 is considerably stable in air and may be superior for
practical use to ligands containing the phosphorus atom.
Finally we examined semi-empirical MO calculations9 in
order to explain the remarkable difference of the enantiose-
lectivity between ligands 5 (R = H) and 6 (R = Me). A reaction
mechanism for the Pd-catalyzed allylic alkylation was proposed
similar to the case of Evans et al.5a Scheme 2 shows the possible
models for ligands, palladium p-allyl complexes, and palla-
dium–olefin complexes. For ligands 5 and 6, two isomers of
each (5-A, 5-B, 6-A and 6-B) are considered due to the
orientation of the phenyl substituent. For the next palladium p-
allyl complexes in 6, a total of four isomers, 12-A1 and 12-A2
from 6-A, and 12-B1 and 12-B2 from 6-B are considered due to
the orientation of the p-allyl moiety. Geometry optimizations
show that 6-A is preferred over 6-B by about 2 kcal mol21 in
energy, and in palladium p-allyl complexes 12-A1 is preferred
by about 2 kcal mol21 over the others. In contrast, the two
conformers 5-A and 5-B of ligand 5 show the same in energy
and also no essential difference is shown between two
palladium p-allyl complexes 13-B1 and 13-B2 with the lowest
energy. These results give support that the reaction of ligand 6
Notes and references
1 For recent reviews, see: (a) B. M. Trost, Chem. Pharm. Bull., 2002, 50,
1; (b) G. Helmchen and A. Pfaltz, Acc. Chem. Res., 2000, 33, 336; (c)
Catalytic Asymmetric Synthesis, ed. I. Ojima, Wiley-VCH, New York,
2000 pp. 802–856. For selected examples, see: (d) L. Xiao, W.
Weissensteiner, K. Mereiter and M. Widhalm, J. Org. Chem., 2002, 67,
2206; (e) C. J. Anderson, R. J. Curbon and J. D. Harling, Tetrahedron:
Asymmetry, 2001, 12, 923; (f) T. Mino, M. Shiotsuki, N. Yamamoto, T.
Suenaga, M. Sakamoto, T. Fujita and M. Yamashita, J. Org. Chem.,
2001, 66, 1795; (g) T. Mino, Y. Tanaka, M. Sakamoto and T. Fujita,
Tetrahedron: Asymmetry, 2001, 12, 2435; (h) D.-R. Hou, J. H.
Reibenspies and K. Burgess, J. Org. Chem., 2001, 66, 206; (i) Y.
Okuyama, H. Nakano and H. Hongo, Tetrahedron: Asymmetry, 2000, 11,
1193; (j) S. R. Gilbertson and D. Xie, Angew. Chem., Int. Ed., 1999, 38,
2750; (k) B. M. Trost, A. C. Krueger, R. C. Bunt and J. Zambrano, J. Am.
Chem. Soc., 1996, 118, 6520.
2 A. Pfaltz, Acc. Chem. Res., 1993, 26, 339.
3 B. M. Trost, Acc. Chem. Res, 1996, 29, 355 and references cited therein;
S.-G. Lee, C. W. Lim, C. E. Song, K. M. Kim and C. H. Jun, J. Org.
Chem., 1999, 64, 4445.
4 (a) P. A. Evans and T. A. Brandt, Org. Lett., 1999, 1, 1563; (b) H.
Steinhagen, M. Reggelin and G. Helmchen, Angew. Chem., Int. Ed. Engl.,
1997, 36, 2108; (c) R. Prétoˆt and A. Pfaltz, Angew. Chem., Int. Ed., 1998,
37, 323 and references cited therein.
5 (a) D. A. Evans, K. R. Campos, J. S. Tedrow, F. E. Michael and M. R.
Gagné, J. Am. Chem. Soc., 2000, 122, 7905; (b) H. Nakano, Y. Okuyama
and H. Hongo, Tetrahedron Lett., 2000, 41, 4615; (c) H. Nakano, Y.
Okuyama, M. Yanagida and H. Hongo, J. Org. Chem., 2001, 66, 620; (d)
H. Nakano, J. Yokoyama, R. Fujita and H. Hongo, Tetrahedron Lett.,
2002, 43, 7761.
6 S. Jansat, M. Gómez, G. Muller, M. Diéguez, A. Aghmiz, C. Claver, A.
M. Masdeu-Bultó, L. Flores-Santos, E. Martin, M. A. Maestro and J.
Mahía, Tetrahedron: Asymmetry, 2001, 12, 1469.
7 S. Ohno, H. Shimizu, T. Kataoka and M. Hori, Chem. Pharm. Bull., 1984,
32, 3471.
8 B. M. Trost and D. J. Murphy, Organometallics, 1985, 4, 1143.
9 AM1 optimization was carried out using WINMOPAC 3.5 version
(Fujistu inc.) and PM 3 optimization was done using Mac Spartan 02
(Wave function inc.). DHf energies obtained from MO calculations and
stereo views of optimized structures are presented in Supporting
Information†.
Scheme 2
CHEM. COMMUN., 2003, 524–525
525