M. Jean et al. / Tetrahedron Letters 50 (2009) 6546–6548
6547
Table 1
5 mol%
O
Palladium-catalyzed arylation of 2-methyl-1-propen-1-yl acetate 2
R1
PdCl2[(o-Tol)3P]2
R1
OAc
ArBr
+
Ar
O
Method A or B
Me
Me Me
Ar CHO
Bu3SnOMe (2 equiv)
DMSO, 100 °C
R2
R2
OAc
Me
ArBr
+
+ Bu3SnOMe
+
Ar
DMSO, 100 °C, 14 h
Me
2
Me
1
3
4
5
(2 equiv)
(2 equiv)
Scheme 1. Pd-catalyzed acylation of arylbromides.
Entry
Aryl bromide 1
Methoda
Ratio 4/5b
Yieldc (%)
Br
we set out to examine the influence of the phosphine on the regi-
oselectivity of this palladium-catalyzed reaction. We compared the
coupling reaction of 2-methyl-1-propen-1-yl acetate 2 with vari-
ous aryl bromides 1 using either tri-tert-butylphosphine (method
A) or tri(o-tolyl)phosphine (method B). The results are shown in
Table 1.
1
2
A
B
95/5
5/95
68
57
Br
3
4
A
B
97/3
3/97
75
75d
As can be seen the reaction afforded low to good yields of cou-
pling products. The formation of the aldehyde 4 or the aryl ketone
5 appears to be both dependent on the aryl bromide and the
phosphine. As previously reported,9 in the presence of tri(o-
tolyl)phosphine (Method B), the aryl ketone 5 was obtained as
the major product (ratio 5:4 >90/10 except with 4-nitro-bromo-
benzene, entry 24). Using the tri-tert-butylphosphine (method A),
a complete inversion of selectivity was observed in most cases (ra-
tio 4:5 >80/20, entries 1, 3, 5, 9 and 13). For the 4-nitro-bromoben-
zene, the conversion and the yield are too low to conclude (entry
23). On the other hand it is noteworthy that the regioselectivity
of the coupling reaction in the presence of the tri-tert-butylphos-
phine is very sensitive to the steric hindrance. A slight decrease
of the selectivity of the formation of aldehyde 4 was observed with
Br
5
6
A
B
80/20
3/97
70
76d
Br
7
8
A
B
70/30
3/97
36
75
d
MeO
Br
9
10
A
B
92/8
3/97
60
51
OMe
a-bromonaphthalene, o-methoxy-bromobenzene, and o-methyl-
bromobenzene (entries 5, 9, and 17). In the case of more sterically
hindered aromatic derivatives as 2-bromo-N,N-dimethylaniline
and o,o0-dimethyl-bromobenzene the aryl ketones 5 were achieved
in high selectivities (ratio 5:4 >97/3).
Br
11
12
A
B
55/45
3/97
48
42
Me2N
Aware of the influence of ligand on the outcome of the reaction,
the tricyclohexylphosphine was also used instead of P(t-Bu)3.
Br
Unfortunately, as Miura and coll. noticed in the a-arylation of alde-
hydes with aryl bromides,2a we did not observe any coupling reac-
tion between the b-bromonaphthalene and 2.
13
14
A
B
97/3
10/90
78
69
NMe2
Based on our previous work,9 two concurrent mechanisms can
be considered. In the presence of tri(o-tolyl)phosphine (cycle B)
the aryl ketone 5 was produced via the addition of the aryl group
to a ketene whereas in the presence of tri-tert-butylphosphine (cy-
cle A) the aldehyde 4 was obtained as the major product following
Br
15
16
A
B
3/97
3/97
79
75
NMe2
the classical catalytic cycle of a-arylation of aldehydes (Scheme 2).
Hartwig and Roy reported that reductive elimination of aryl halide
is made easy by the addition of tri-tert-butylphosphine to a dimeric
tri(o-tolyl)phosphine arylpalladium (II) halide complexe.10 Conse-
quently, the difference of reactivity in our reaction would be due
to a faster reductive elimination with P(t-Bu)3 than with P(o-
Tol)3 after the formation of the intermediate A. Moreover, the bulk-
iest phosphine11 P(o-Tol)3 helps the b-H-elimination to give the
palladium-ketene intermediate B. This is also consistent with the
preferential formation of the aryl ketone 5 when sterically hin-
dered ortho-substituted aryl bromides are used.
Br
17
18
A
B
65/35
3/97
64
41
Me
Me
Br
19
20
A
B
3/97
3/97
57
43
Me
Having established that the tri-tert-butylphosphine leads to the
-arylation product, we turned our attention to the reactivity of
aryl bromides in the presence of vinyl acetate and monosubsti-
tuted vinylic acetates. Reaction of b-bromonaphthalene with vinyl
acetate under the method A reaction conditions (Scheme 3) affor-
Br
21
22
A
B
Inextricable mixture
a
3/97
62
MeO2C
O2N
Br
ded a complex mixture from which the a,b-unsaturated aldehyde 6
23
24
A
B
97/3
45/55
21
39
was isolated in 28% yield (in the presence of P(o-Tol)3, method B,9
the corresponding aryl ketone was isolated in 41% yield). The alde-
hyde 6 was probably formed through a palladium-catalyzed cou-
pling reaction followed by enolization of the aldehyde-
aldolization–crotonization cascade reactions. The use of 3-phe-
nyl-1-propen-1-yl acetate instead of vinyl acetate also provided a
a
Method A: 5 mol % PdCl2(CH3CN)2, 10 mol % P(t-Bu)3. Method B: 5 mol %
PdCl2[P(o-Tol)3]2.
Ratio determined by 1H NMR from the crude mixture.
Isolated yield (average of two runs).
b
c
d
See Ref. 9.