However, when the conversion exceeds 50%, the catalytic
transformation of 2a is faster in the presence of 3 than in
the absence of it. This indicates that accumulation of allyl
boronate 8 inhibits the catalytic process.
It is well-known that homocoupling product 7 readily
6
Figure 1. Palladium-catalyzed transformation of 2a in the presence
b) and absence (9) of aldehyde 3.
7
a,b
forms from bis-allylpalladium complexes.
Therefore,
(
appearance of the allyl-allyl coupling product (7) in the
absence of aldehyde 3 indicates that a bis-allylpalladium
intermediate (10) is generated under catalytic conditions (eq
).
The possible intermediacy of bis-allylpalladium complex
0 in the above-presented catalytic process (Scheme 1) is
usually accompanied by formation of the allyl-allyl coupling
product 7.
3
1
also indicated by the reactions involving imine 4 (entries
9
-12). In contrast to aldehyde 3, sulfon-imine 4 cannot be
allylboronated with 8 in the absence of palladium catalyst
(eq 2). Accordingly, the boron atom does not exert an internal
Lewis acid reactivity as in the reactions with aldehyde types
It is reasonable to assume that the first step of the above-
presented allylic substitution reaction (Scheme 1) is pal-
ladium-catalyzed formation of the transient allyl boronate.
However, it is interesting to note that allyl-allyl coupling
products (such as 7) were not observed in any of the reactions
presented in Table 1. It is well established that allyl boronates
5
c
of substrates. On the other hand, we have found that in the
presence of Pd (dba) catalyst, allyl boronate 8 smoothly
2
3
reacts with 4 affording 6a. A similar reactivity was reported
for palladium-catalyzed allylic substitution of trialkyl-allyl
3
,8
undergo direct allylboration with aldehydes and other
stannanes with aldehydes and imines. In these reactions,
the bis-allypalladium intermediate (10) is formed from allyl
stannanes and 9 (cf. eq 3).
electrophiles.5
b-d
In these reactions, the carbonyl group of
the aldehyde coordinates to the boron atom, which reacts as
5c
an internal Lewis acid due to its low-lying vacant p
Indeed, allyl boronate 8 smoothly reacted with aldehyde 3
eq 2) under the applied reaction conditions (Scheme 1) in
π
-orbital.
It was shown that bis-allylpalladium complexes readily
react with electrophiles under catalytic conditions.3 Al-
though the catalytic allylic substitution via bis-allylpalladium
complexes is a well-known reaction for trialkyl-allyl stan-
nanes, as far as we know, this is the first study reporting a
similar reactivity for allyl boronates.
,7,8
(
the absence of palladium catalyst. Accordingly, the catalytic
allylic substitution of allyl acetates (2a-h) with aldehyde 3
can be considered as a palladium-catalyzed formation of
transient allyl boronates followed by direct coupling with
the electrophile (i.e., 3).
The high stereoselectivity is a synthetically useful feature
of the presented allylic substitution reaction. The reactions
with aldehyde 3 selectively provide (entries 2-8) the anti
diastereomer. This stereoselectivity can be explained by the
selective formation of the trans isomer of the allyl boronate
intermediate in the initial palladium-catalyzed step. The
subsequent reaction of the trans-allyl boronates with alde-
5c
hydes leads to formation of the anti diastereomer. The major
diastereomer formed in the reactions with sulfon-imine 4 has
syn configuration. A similar syn diastereoselectivity was
reported for the formation of 6b from 4 and cinnamyl
Performing the reaction as a one-pot sequence has several
advantages. (1) Isolation of the reactive allyl boronate
intermediates (entries 5-8) can be avoided. (2) Allyl-allyl
coupling does not take place in the presence of electrophiles.
9a
bromide in indium- and zinc-mediated reactions. Although
understanding the mechanism of the stereochemistry requires
further studies, we believe that the nature of the steric and
(
3) The palladium-catalyzed conversion of allyl acetate (2a)
is faster in the presence of 3 than in the parent reaction (eq
). To further explore feature 3, we have studied the rate of
1
(
7) (a) Nakamura, H.; Bao, M.; Yamamoto, Y. Angew. Chem., Int. Ed.
001, 40, 3208. (b) Goliaszewski, A.; Schwartz, J. Tetrahedron 1985, 41,
5779.
consumption of 2a in the presence and in the absence of
2
6
aldehyde 3 (Figure 1). In the beginning of the reaction (t <
(
8) (a) Nakamura, H.; Iwama, H.; Yamamoto, Y. J. Am. Chem. Soc. 1996,
10 min) the rate of conversion of 2a does not depend on 3.
1
18, 6641.(b) Nakamura, H.; Aoyagi, K.; Shim, J.-G., Yamamoto, Y. J.
Am. Chem. Soc. 2001, 123, 372. (c) Solin, N.; Narayan, S.; Szab o´ , K. J. J.
Org. Chem. 2001, 66, 1686. (d) Solin, N.; Narayan, S.; Szab o´ , K. J. Org.
Lett. 2001, 3, 909.
(
6) These reactions were performed in the presence of 1 and a catalytic
amount of Pd2(dba)3 in DMSO-d6 at 50 °C.
Org. Lett., Vol. 5, No. 17, 2003
3067