Cobalt-catalyzed allylic substitution reaction of allylic ethers with phenyl and trimethylsilylmethyl grignard reagents

Article abstract of DOI:10.1246/cl.2004.832

Treatment of cinnamyl methyl ether with phenylmagnesium bromide in ether in the presence of CoCl₂[1,5-bis(diphenylphosphino)pentane] affords 1,3-diphenylpropene in good yield. Similar allylic substitution reaction with trimethylsilylmethylmagne

Full text of DOI:10.1246/cl.2004.832

8
32
Chemistry Letters Vol.33, No.7 (2004)
Cobalt-Catalyzed Allylic Substitution Reaction of Allylic Ethers
with Phenyl and Trimethylsilylmethyl Grignard Reagents
Ã
Keiya Mizutani, Hideki Yorimitsu, and Koichiro Oshima
Department of Material Chemistry, Graduate School of Engineering, Kyoto University,
Kyoto-daigaku Katsura, Nishikyo-ku, Kyoto 615-8510
(Received April 6, 2004; CL-040375)
Treatment of cinnamyl methyl ether with phenylmagnesium
Reactions at ambient temperature decreased the yield of 2.
The choice of the solvent was essential to obtain 2 in satisfactory
yield. Reaction in THF gave rise to very low conversion of 1. It
is worth noting that treatment of cinnamyl bromide under similar
conditions gave a mixture of dimeric compounds such as 1,6-di-
phenyl-1,5-hexadiene and 3,4-diphenyl-1,5-hexadiene in addi-
tion to a trace of 2. The formation of the dimeric products sug-
gests that single electron transfer from a cobalt complex yields
the cinnamyl radical intermediate. Starting from branched ether
3, 2 was obtained in comparable yield (Eq 1).
bromide in ether in the presence of CoCl2[1,5-bis(diphenylphos-
phino)pentane] affords 1,3-diphenylpropene in good yield. Sim-
ilar allylic substitution reaction with trimethylsilylmethylmag-
nesium chloride proceeded smoothly to yield homoallylsilanes.
ꢀ,ꢁ-Unsaturated aldehyde acetal also underwent allylic substitu-
tion.
Palladium-catalyzed allylic substitution is recognized as one
of the most useful reactions catalyzed by transition metal com-
CoCl (dpppen)(5mol %)
2
1
plexes. On the other hand, cobalt-catalyzed allylic substitutions
have scarcely been studied, where enolates were employed as
nucleophiles.2 We have been interested in cobalt–phosphine
OMe
PhMgBr (2.0 equiv.)
2
(1)
Ph
ether, reflux, 16 h, 60%
,3
3
4
complexes-catalyzed reactions. During the course of our study,
Treatment of (E)-2-octenyl methyl ether (4) with phenyl-
magnesium bromide in the presence of CoCl2(dpppen) afforded
a mixture of the corresponding coupling products 5, 6, and 7 in
very poor yields (Table 2, Entry 1). Under the reaction condi-
tions, a part of 5 was transformed into 6. Without any phosphine
ligand, coupling products were obtained in moderate yield (En-
try 2). Instead, triphenylphosphine proved to be the best ligand
for the reaction of 4 (Entry 4). Other monodentate phosphine li-
gands were inferior to triphenylphosphine (Entries 5-8).
we felt that the combination of cobalt salts, phosphine ligands,
and Grignard reagents deserves further investigation and that al-
lylic substitution reaction is a suitable probe. Here we report the
reaction of allylic ether with phenyl and trimethylsilylmethyl
Grignard reagents in the presence of cobalt–phosphine com-
plexes.
The reaction of cinnamyl methyl ether (1) with phenyl
Grignard reagent was first examined (Table 1). In the previous
report on the cross-coupling reaction of alkyl halide with
ArMgBr, 1,2-bis(diphenylphosphino)ethane (DPPE) was the
Under the reaction conditions of Entry 4 in Table 2, 8 was
also converted into 5 and 7 (Eq 2). No isomerization of 5 into
4
a
choice of the ligand. For the allylic substitution reaction, DPPE
was not suitable and linear product 2 was obtained in only 15%
yield (Entry 1). A variety of ligands were surveyed and 1,5-bis-
6
was observed in this reaction.
CoCl2(PPh3)2 (5 mol %)
PhMgBr (2.0 equiv.)
OMe
(
diphenylphosphino)pentane (DPPPEN) was found to be the
5
+
7
(2)
most effective for this reaction. 3,3-Diphenyl-1-propene was
not detected in every experiment.
ether, reflux, 16 h
n-C H
5
11
7
4% (50/50)
8
Table 2. Reaction of (E)-2-octenyl ether 4 with PhMgBr
Table 1. Reaction of cinnamyl ether 1 with PhMgBr
R
Ph
CoCl (ligand) (5 mol %)
CoCl (ligand) (5 mol %)
2
2
PhMgBr (2.0 equiv.)
PhMgBr (2.0 equiv.)
5
Ph
OMe
Ph
Ph
R
OMe
Ph
ether, reflux, 16 h
ether, reflux, 16 h
R
Ph
1
2
4
R
R = n-C H
5
11
6
7
a
Entry
Ligand
Yield /%
a
Entry
Ligand
Yield /%
5/6/7
b
1
2
3
4
5
6
7
8
DPPE
None
PPh3 (10 mol %)
DPPM
DPPP
DPPB
DPPPEN
DPPH
15
29
30
24
27
50
72
58
1
2
3
4
5
6
7
DPPPEN
—
DPPE
12
47
32
78
39
49
trace
41
N.D.
58/10/32
10/53/37
36/7/57
66/<1/33
42/6/52
N.D.b
c
PPh3
c
c
P(2-MeC6H4)3
P(4-MeC6H4)3
c
P[3,5-(CF3)2C6H3]3
c
aLigands DPPM–DPPH represent Ph2P(CH2)nPPh2, n = 1,
DPPM; n = 2, DPPE; n = 3, DPPP; n = 4, DPPB; n = 5,
DPPPEN; n = 6, DPPH.
8
a
P(4-MeOC H )
6
31/16/53
4 3
b
Combined yield of 5, 6, and 7. Not determined.
10 mol % of monodentate phosphines were used.
c
Copyright Ó 2004 The Chemical Society of Japan

Chemistry Letters Vol.33, No.7 (2004)
833
Allylic substitution reaction with Me3SiCH2MgCl was fac-
ile. Treatment of 1 with Me3SiCH2MgCl in the presence of
CoCl (dpph) (5 mol %)
2
Me SiCH MgCl (1.5 equiv.)
OMe
3
2
ꢀ
14
(8)
CoCl2(dpph) for 14 h at 20 C afforded a linear product 9 in
5
SiMe3
ether, 20 °C, 35 h
Ph
Ph
99% yield (Eq 3). Whereas the choice of the ligand was crucial
1
7 85%
to establish phenylation, ligandless CoCl2 and CoCl2(dppb) also
catalyzed the trimethylsilylmethylation to afford 9 in 92 and
CoCl (dpph) (5 mol %)
2
SiMe3
Me SiCH MgCl (3.0 equiv.)
98% yields, respectively. Reactions of 3 with Me3SiCH2MgCl
3
2
1
4
(9)
SiMe3
afforded 9 in excellent yield. On the other hand, alkyl-substitut-
ed allylic ethers 4 and 8 were converted into mixtures of re-
gioisomers 10 and 11 (Eq 4). The reaction required a higher tem-
perature to complete within a satisfactory reaction time.
Trimethylsilylmethylation of branched ether 8 afforded higher
yields of 10 and 11 than the reaction of 4.
ether, reflux, 48 h
1
8 94%
In summary, we have demonstrated allylic substitution reac-
tion of allylic ethers with Grignard reagents under cobalt catal-
ysis. The cobalt-catalyzed reaction with phenyl Grignard reagent
proved to be a function of a substrate as well as of solvent and
ligand. To attain high yields in the phenylation reaction, inten-
sive tunings of variants are essential. In contrast, introduction
of a trimethylsilylmethyl group was facile and clean under cobalt
catalysis.
Co catalyst (5 mol %)
Me SiCH MgCl (2.0 equiv.)
3
2
1
or 3
Ph
(3)
SiMe3
ether, 20 °C, 14 h
9
9
9% CoCl (dpph) from 1
98% CoCl (dppb) from 1
2
2
9
3% CoCl (dpph) from 3
92% CoCl2 from 1
2
References and Notes
n-C H
5
11
1
a) J. Tsuji, Acc. Chem. Res., 2, 144 (1969). b) J. Tsuji,
Pure Appl. Chem., 54, 197 (1982). c) B. M. Trost and
D. L. Van Vranken, Chem. Rev., 96, 395 (1996). d) J. Tsuji,
J. Organomet. Chem., 300, 281 (1986).
SiMe3
CoCl (dpph) (5 mol %)
2
10
Me SiCH MgCl (2.0 equiv.)
3
2
(4)
4
or 8
+
SiMe3
ether, reflux, 16 h
2
3
4
a) B. Bhatia, M. M. Reddy, and J. Iqbal, Tetrahedron Lett.,
34, 6301 (1993). b) P. A. Grieco, W. J. DuBay, and L. J.
Todd, Tetrahedron Lett., 37, 8707 (1996).
Cobalt-mediated direct electrochemical coupling of aryl hal-
ides with allylic esters was reported. P. Gomes, C. Gosmini,
and J. P e´ richon, J. Org. Chem., 68, 1142 (2003).
a) K. Wakabayashi, H. Yorimitsu, and K. Oshima, J. Am.
Chem. Soc., 123, 5374 (2001). b) Y. Ikeda, H. Yorimitsu,
and K. Oshima, J. Am. Chem. Soc., 124, 6514 (2002). c) T.
Fujioka, T. Nakamura, H. Yorimitsu, and K. Oshima, Org.
Lett., 4, 2257 (2002). d) T. Tsuji, H. Yorimitsu, and K.
Oshima, Angew. Chem., Int. Ed., 41, 4137 (2002). e) K.
Mizutani, H. Shinokubo, and K. Oshima, Org. Lett., 5,
3959 (2003).
5
9% (10/11 = 63/37) from 4 n-C H
5
11
1
9
2% (10/11 = 82/18) from 8
1
Treatment of acetals 12 and 14 with phenyl Grignard re-
agent in the presence of CoCl2(dpppen) afforded the correspond-
ing monophenylated allylic ethers 13 and 15, respectively (Eqs 5
and 6). The dimethyl and phenyl groups would interfere with
second phenylation.
CoCl (dpppen) (5 mol %)
2
OEt
OEt
OEt
Ph
PhMgBr (3.0 equiv.)
(
5)
ether, reflux, 35 h
1
2
13 90%
CoCl (dpppen) (5 mol %)
2
OMe
OMe
OMe
PhMgBr (3.0 equiv.)
(6)
5
Experimental Procedure: Anhydrous CoCl2 (7 mg,
Ph
Ph
Ph
5 62%
ether, reflux, 35 h
0
.05 mmol) was placed in a 50-mL two-necked flask and
heated with a hair dryer in vacuo for 3 min. DPPH (27 mg,
.06 mmol) and ether (1 mL) were sequentially added under
1
4
1
In contrast to the reaction with phenyl Grignard reagent, bis-
0
trimethylsilylmethylation occurred in the reaction of 12 with
three equimolar amounts of the Grignard reagent in refluxing
ether (Eq 7). Interestingly, in the reaction of 14, we could control
the distribution of the product by changing the amount of the
Grignard reagent and reaction time (Eqs 8 and 9). Reaction with
argon. After the mixture was stirred for 30 min to obtain blue
suspension, cinnamyl methyl ether (1, 0.15 g, 1.0 mmol) and
Me3SiCH2MgCl (1.0 M in ether, 2.0 mL, 2.0 mmol) were
ꢀ
successively added to the reaction mixture at 0 C. After be-
ꢀ
ing stirred for 14 h at 20 C, the reaction mixture was poured
1.5 equimolar amounts of the Grignard reagent at ambient tem-
into saturated NH4Cl solution. The products were extracted
with ethyl acetate (20 mL Â 3) and the combined organic
layer was dried over sodium sulfate and concentrated. Silica
gel column purification of the crude product provided 9
perature for 35 h afforded monosubstituted product 17 in 85%
yield. On the other hand, use of three equimolar amounts of
the Grignard reagent resulted in generation of doubly substituted
product 18 in 94% yield after 48 h reaction at reflux.
(
0.20 g, 0.99 mmol) in 99% yield as colorless oil.
CoCl (dpph) (5mol %)
2
SiMe3
SiMe3
Me SiCH MgCl (3.0 equiv.)
3
2
1
2
(7)
ether, reflux, 24 h
16 61%
Published on the web (Advance View) June 7, 2004; DOI 10.1246/cl.2004.832