Enantioselecitve elimination of Pd-H from η3-allylpalladium- tol BINAP complexes. Evidence of Syn elimination pathway

Article abstract of DOI:10.1246/cl.2007.1338

The palladium-catalyzed elimination of the bicyclic cis-acetate lb using Pd catalyst and (R)-p-Tol BINAP gave the (S)-diene 2 with 58% ee. Bicyclic η³-allylpalladium complex 3a with (R)-p-Tol BINAP, considered as an intermediate in the catalyti

Full text of DOI:10.1246/cl.2007.1338

1338
Chemistry Letters Vol.36, No.11 (2007)
Enantioselecitve Elimination of Pd–H from ꢀ3-Allylpalladium–Tol BINAP Complexes.
Evidence of Syn Elimination Pathway
Ryohei Ogawa, Yusuke Shigemori, Koichi Uehara, Jiro Sano, Takayuki Nakajima, and Isao ShimizuÃ
Department of Applied Chemistry, School of Science and Engineering, Waseda University,
3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555
(Received July 23, 2007; CL-070776; E-mail: shimizui@waseda.jp)
-
PF6
The palladium-catalyzed elimination of the bicyclic cis-ace-
tate 1b using Pd catalyst and (R)-p-Tol BINAP gave the (S)-di-
ene 2 with 58% ee. Bicyclic ꢀ3-allylpalladium complex 3a with
(R)-p-Tol BINAP, considered as an intermediate in the catalytic
reaction from 1b, was prepared. Enantioselectivity in decompo-
sition of 3a is dependent on the reaction conditions. The thermal
decomposition of 3a without base gave the (S)-2 with 70% ee.
However, the decomposition of 3a in the presence of excess base
gave (R)-2 with 58% ee. Syn elimination from 3a was found
to proceed preferentially from the decomposition results of the
deuterium-labeled complexes.
a
b
Pd+
Pd
Cl
P
P
( )-4
2
5
3a: (R)-p-Tol BINAP
3b: Dppp
Scheme 2. Synthesis of [Pd(ꢀ3-C11H17)(diphosphine)]PF6. Re-
agents and conditions: (a) PdCl2(PhCN)2, CHCl3, reflux, 23%
(b) diphosphine, AgPF6, CH2Cl2, 73% (for 3a), 80% (for 3b).
Table 1. Thermal decomposition of complex 3a5
Base
Additive
1,4-Dioxane, 100 oC
+
3a
Asymmetric allylic alkylation catalyzed by palladium com-
plexes bearing chiral ligands is a useful synthetic method. The
direction of nucleophilic attack caused by the desymmerrization
of allylic moiety by the coordination of chiral ligands is ex-
plained.1 Although numerous studies of the palladium-catalyzed
allylic alkylation have appeared, enantioselective elimination
from allylic compounds to optically active 1,3-dienes has scarce-
ly been explored.2 Several years ago, we reported that the reac-
tion of bicyclic trans-allylic carbonate 1a in the presence of cat-
alytic amounts of [(1-Me-allyl)PdCl]2 and (R)-BINAP gave the
bicyclic diene (S)-2 with 86% ee (Scheme 1).3 In our continuous
studies, the catalytic reaction was carried out using 1b under
similar conditions as 1a to give the same S isomer 2 in 58%
ee. The same stereochemical outcome from the opposite config-
uration of starting allylic substrates suggests that equilibration of
ꢀ3-allylpalladium intermediate proceeded prior to the elimina-
tion as shown in Scheme 1. In order to elucidate the intermediate
of the enantioselective elimination reactions, we have prepared
{Pd(ꢀ3-C11H17)[(R)-p-Tol BINAP]}PF6 complex 3a, and de-
composition reaction of 3a was investigated to gain insight into
precise mechanisms of the elimination reactions.
(S)-2
(R)-2
Base
/equiv.
Additive Time
/h
Entry
Yield/%a ee/%b
/equiv.
1
2
3
4
5
4
0.4
6
79
85
76
92
58
18 (S)
70 (S)
8 (R)
58 (R)
8 (S)
LiCl (10)
Et3N (1)
Et3N (10)
1
Et3N (10) LiCl (10) 18
aIsolated yield. bEnantiomeric excess was determined by
GLC using a chiral column.
AgPF6 in CH2Cl2 at room temperature to give 3a (73% yield)
or 3b (80% yield), respectively, after recrystallization from a
mixture of hexane–CH2Cl2 (Scheme 2). The NMR spectrum
of 3a shows unsymmetrical feature of the allylic moiety in the
complex 3a, although that of the complex 3b with achiral phos-
phine has CS symmetrical structure.6
The decomposition of 3a was carried out under various con-
ditions. As shown in Table 1, the enantioselection and enantio-
selectivity were dependent on the reaction conditions. The ther-
mal decomposition of 3a at 100 ꢀC in 1,4-dioxane without a base
and an additive gave the (S)-diene 2 in 79% yield with low enan-
tioselectivity (18% ee). When LiCl was added, the elimination
proceeded with high enantioselectivity to give (S)-2 (70% ee,
85% yield). Interestingly when one equivalent of Et3N was add-
ed, the opposite enantiomer (R)-2 was obtained but low enantio-
selectivity (8% ee). Furthermore, when excess Et3N (10 equiv.)
was used, the decomposition proceeded smoothly to give (R)-2
with considerable enantioselectivity (58% ee, 92% yield).
Although 3a was one of plausible intermediates from 1b, the
stereochemical results of the decomposition of 3a using Et3N
were not in accordance with the enantioselection for (S)-2 in
the catalytic reaction starting with 1b.
At first the bicyclic ꢀ3-allylpalladium complexes, which
correspond to the trans-ꢀ3-allylpalladium intermediate obtained
directly from 1b, were prepared. Treatment of the complex 54
with the diphosphine ligands, (R)-p-Tol BINAP or Dppp, and
[(1-Me-allyl)PdCl]2
(R)-BINAP
Et3N (5 mol %)
1,4-Dioxane
100 oC, 2 h
OCO2Me
( )-1a
PdLn
[(1-Me-allyl)PdCl]2
(R)-p-Tol BINAP
(S)-2
Et3N (1 equiv.)
1,4-Dioxane
100 oC, 22 h
61%, 78% ee from 1a
48%, 58% ee from 1b
PdLn
OAc
( )-1b
Takacs and co-workers reported the palladium-catalyzed
elimination of allylic compounds proceeded via specific base-
Scheme 1. Palladium-catalyzed enantioselective elimination of
bicyclic allylic compounds 1a and 1b.3
Copyright Ó 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.11 (2007)
1339
Table 2. Catalytic elimination reaction of (Æ)-1b5
Neutral condition
H
H
[(1-Me-allyl)PdCl]2 (2.5 mol %)
(R)-p-Tol BINAP (10 mol %)
-HPdLn
-HPdLn
H
H
Pd+
*
Base (1 equiv.)
Tol
Tol
Tol
P
P
1,4-Dioxane, 100 oC
OAc
Tol
(S)-2
H
H
(S)-2
( )-1b
H
Entry
Base
Time/h
Yield/%a
% eeb
H
Pd+
Tol
H
Tol
P
P
H
1
2
6
4
22
4
27
77
48
91
89
31
58
78
To l
Tol
*
3a
H
Tol
Tol
H
K2CO3
Et3N
Et3N
Pd+
*
Tol
P
P
3
4c
Tol
Basic condition
(R)-2
aIsolated yield. bEnantiomeric excess was determined by
GLC using a chiral column. cAn equivalent of LiCl was
added.
Scheme 3. The relationship between the direction and reaction
conitions in the elimination of Pd–H from 3a to (S)- or (R)-2.
considered to be very small, which is in accordance with small
kinetic isotopic effects in ꢂ-hydride elimination.11 Furthermore,
when the reaction of 3ac was carried out with LiCl, syn elimina-
tion proceeded predominantly. In the reaction under conditions
in Table 3, syn elimination is preferential from the ꢀ3-allylpalla-
dium complexes, although the catalytic reaction is known to
proceed with anti elimination.
In conclusion, the palladium-catalyzed elimination reaction
of 1b was carried out smoothly with Et3N and LiCl to give (S)-2
with high enantiomeric excess. Enantioselection in decomposi-
tion of ꢀ3-allylpalladium complexes 3a with Et3N was opposed
to that without the base. The decomposition of 3a took place
in syn elimination pathway with or without a base (Scheme 3).
Table 3. Decomposition of complexes 3aa, 3ab, and 3ac
-
PF6
R
H
Additive (1 equiv.)
+
R2
R1
1,4-Dioxane
100 oC
H
Pd+
D
L
2a: R2 = D
2b: R2 = H
L
2c
3aa: R = H, R1 = D, L = (R)-p-Tol BINAP
3ab: R = D, R1 = H, L = (R)-p-Tol BINAP
3ac: R = D, R1 = H, L = (S)-p-Tol BINAP
Entry Substrate Additive Time/h Yield/%a 2a/2b/2cb
1
2
3
3aa
3ab
3ac
Et3N
Et3N
LiCl
1
1
0.5
68
60
99
10:48:42
48:10:42
61:8:31
This work was supported by the program of Initiatives for
Attractive Education in Graduate Schools (b043) from JSPS-
MEXT and Waseda University Grant for Special Research
Projects (2006A-063).
aIsolated yield. bThe ratio of 2a:2b:2c was caluculated by
1H NMR spectra and GLC using a chiral column.
promoted anti elimination of Pd–H from the ꢀ3-allylpalladium
intermediate.7 However, stereochemical studies using character-
ized ꢀ3-allylpalladium complexes relevant to the catalytic reac-
tion have not been reported. We carried out the elimination of 1b
using 1:4 [(1-Me-allyl)PdCl]2:(R)-p-Tol BINAP catalyst, and
(S)-2 was obtained mainly in all cases (Table 2). The reaction
was very slow without a base, however, high enantioselectivity
was obtained (Entry 1; 89% ee). When the reaction was carried
out adding K2CO3, the reaction proceeded smoothly, however,
the enantioselectivity decreased (31% ee, 77% yield). When
Et3N was used instead of K2CO3, the yield of 2 decreased, but
the enantioselectivity increased (58% ee, 48% yield). Further-
more when LiCl was added, the reaction proceeded smoothly
to give (S)-2 with high enantioselectivity (78% ee, 91% yield).8
It is noteworthy that the (S) isomer formed as a major product in
each catalytic reaction of 1b, evenwhen Et3N was used, whereas
(R)-2 was obtained in the decomposition of 3a (Entry 4 in
Table 1).
In order to elicit direct evidence for the involvement wheth-
er syn or anti elimination of Pd–H occurs from the ꢀ3-allylpalla-
dium intermediates, we prepared the deuterium-labeled ꢀ3-allyl-
palladium complexes with (R)- or (S)-p-Tol BINAP, 3aa, 3ab,
and 3ac, from the 2ꢁ or 2ꢂ duterio-1-octalin, (R)-2ꢁ-d-4a and
(R)-2ꢂ-d-4b. Decomposition of 3aa was carried out under sim-
ilar conditions as shown in Table 3.9 The ratio of 2a:2b was
10:48, which indicates that the syn H(D) to palladium atom
was picked up preferentially. The elimination was also examined
with 3ab, and the same syn:anti elimination ratio was ob-
served.10 These results indicated that the isotope effects were
References and Notes
1
For reviews, see: a) B. M. Trost, D. L. Van Vranken, Chem. Rev. 1996,
96, 395. b) J. M. Williams, Synlett 1996, 705. c) T. Hayashi, J. Orga-
nomet. Chem. 1999, 576, 195. d) G. Helmchen, J. Organomet. Chem.
1999, 576, 203.
2
a) T. Hayashi, K. Kishi, Y. Uozumi, Tetrahedron: Asymmetry 1991, 2,
195. b) E. B. Koroleva, P. G. Andersson, Tetrahedron: Asymmetry
1996, 7, 2467. c) P. G. Andersson, S. Schab, Organometallics 1995,
14, 1.
3
4
5
6
I. Shimizu, Y. Matsumoto, K. Shoji, T. Ono, A. Satake, A. Yamamoto,
Tetrahedron Lett. 1996, 37, 7115.
The complex 5 was prepared according to the Jones’s method. D. N.
Jones, S. D. Knox, J. Chem. Soc., Chem. Commun. 1975, 165.
Isomerization between (S)- and (R)-2 was scarcely observed under
these condition.
Each allylic proton of the ꢀ3-allylpalladium complex 3a appeared at
4.77 and 3.64 ppm respectively, whereas those of 3b appeared at
4.17 ppm in the 1H NMR spectra. The two phosphorus nuclei in (R)-
p-Tol BINAP appeared at 22.6 and 21.4 ppm, whereas 3b showed
the single chemical shift at 8.3 ppm in the 31P{1H} NMR spectrum.
a) J. M. Takacs, E. C. Lawson, F. Clement, J. Am. Chem. Soc. 1997,
119, 5956. b) J. M. Takacs, F. Clement, J. Zhu, S. Chandramouli, X.
Gong, J. Am. Chem. Soc. 1997, 119, 5804.
7
8
9
Halide effect in the elimination is not unambiguous. For a review, see:
K. Fagnou, M. Lautens, Angew. Chem., Int. Ed. 2002, 41, 26.
The ratios of 2a:2b:2c were calculated by 1H NMR spectra and GLC
using a chiral column.
10 The large deuterium effect in anti elimination is described in Ref. 6a.
11 a) F. Ozawa, T. Ito, A. Yamamoto, J. Am. Chem. Soc. 1980, 102, 6457.
b) R. L. Brainard, G. M. Whitesides, Organometallics 1985, 4, 1550.
c) R. Romeo, G. Alibrandi, L. M. Scolaro, Inorg. Chem. 1993, 32,
4688. d) J. Evans, J. Schwartz, P. W. Urquhart, J. Organomet. Chem.
1974, 81, C37.
Published on the web (Advance View) October 13, 2007; doi:10.1246/cl.2007.1338