Communications
iodoalkynylation took place to give 3y in excellent yields
(Table 1, entry 25). The use of norbornenoic acid 1g resulted
in the exclusive formation of the 3-alkynyl derivative 3x
(Table 1, entry 24), thus indicating that the nucleophilic attack
occurred prior to the rearrangement.[8] Unfortunately, the
reaction of haloalkynes with open-chain alkenes, such as 4-
octene, only afforded a mixture of products, and the reaction
was unsuccessful with terminal alkenes.
The reaction between a bromoalkyne and cyclooctene is
equally interesting, since the latter is more flexible compared
to the strained norbornene derivatives. We predicted that the
reaction should lead to a 2-propynyl bromide derivative
(Scheme 5, path a). However, when the reaction was per-
Scheme 5. Bromoalkynylation of cyclooctene.
Scheme 4. Norbornene derivatives and haloalkynes. Substrates 1b, 1c,
1e, and 1g are mixtures of endo and exo isomers (see the Supporting
Information).
formed under conditions similar to those described above, we
found that this reaction resulted in the formation of a four-
membered ring by a [2+2] cycloaddition reaction (Scheme 5,
path b). To the best of our knowledge, the [2+2] cycloaddition
of alkynes and monocyclic alkenes continues to represent a
challenge.[9] This approach represents another utilization of
haloalkynes for carbocycle formations that employ palladium
catalysis.
(Table 1, entries 17 and 18). On the other hand, the carbonyl
substituent in norbornene 1e led to an obvious selectivity for
2,5,7-substituted norbornane 3s (Table 1, entry 19). To our
surprise, the use of 4-bromo-2-phenylbut-3-yn-2-ol (2n)
resulted in the formation of the dehydration product 3w in
67% yield (Table 1, entry 23). Trimethylsilylethynyl bromide
(2l) also added onto norbornene to afford 2-bromo-7-
alkynylnorbornane 3u in good yield (Table 1, entry 21), thus
showing the mildness and robustness of our method.[7] Alkyl
alkynyl bromides such as n-pentyl bromide (2m) can also
undergo this transformation in good yields (Table 1, entry 22).
In this case, simple filtration through a silica gel plug was
sufficient to remove the residual catalyst and provide product
3v. Interestingly, the use of 1,4-bis(2-bromoethynyl)benzene
(2p) resulted in the formation of the corresponding diyne
product 3z as white crystals in 73% yield (Table 1, entry 26).
The molecular structure of 3z was established by X-ray
crystallography (Figure 1), which enabled us to confirm the
reaction nature as a 2,7-addition process. We also extended
this reaction to phenylethynyl iodide (2o) and found that the
As shown in Scheme 6, aromatic alkynyl bromides with
either electron-donating or electron-withdrawing groups
attached to the benzene rings were able to smoothly undergo
a [2+2] cycloaddition with cyclooctene, and generated the
corresponding products in moderate to good yields. The
reaction tolerated a variety of substituents including Cl, Br, F,
and OMe groups. The steric hindrance associated with the
alkynyl bromide also affected the yield of the reaction, as the
introduction of an o-trifluoromethyl group onto the alkynyl
bromide lowered the conversion of the haloalkyne (Scheme 6,
5o). We also extended this reaction to phenylethynyl iodide
(2o) and found that the [2+2] cycloaddition took place to give
5n in good yield. Other alkenes such as cycloheptene were
also subjected to this reaction, however, the yields decreased
and the main products were inseparable from Alder–ene by-
products.[9e] Furthermore, it was found that reactions with
cyclododecene were completely ineffective. These observa-
tions indicate that the ring size of the alkene plays a major
role in the formation of the desired product. Finally, the
reaction of bromoalkyne 2e with cyclopentene under typical
palladium-catalyzed cross-coupling conditions gave enynes
6a and 6b in 87% yield, thus suggesting that the reaction
involves the insertion of the alkene to an alkynyl palladium
species rather than a palladium cyclopentene intermediate
(Scheme 7a).[9]
To demonstrate the synthetic utility of cyclobutenyl
bromide, we showed that the Pd/Cu-catalyzed coupling of
5b with phenylacetylene gave enyne 7a in 71% yield.
Figure 1. X-ray crystal structure of 3z. Thermal ellipsoids set at 50%
probability.
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 6341 –6345