TABLE 3. Preparation of Furans 9 and 10 from 1 and Methyl
Vinyl Ketonea
(entry 5). Other alkenes, such as acrylaldehyde, cyclohexenone,
ethyl crotonate, tert-butyl methacrylate, styrene, cyclohexene,
or 3,4-dihydro-2H-pyran, were not successful.
The corresponding reactions with other epoxides 1 and
electron-deficient alkenes 4a-c provided uniformly good yields
of 3-vinylfuran derivatives 5–7 (entries 6–20), which display
structural diversity stemming from the choice of the olefin
activating group as well as from the convergent assembly of
the starting esters.9 Therefore, in common with other Pd-
catalyzed applications of intermediates 3, this cyclization/Heck-
type coupling sequence succeeds in the formation of densely
substituted functionalized furans10 in a highly efficient manner,
with complete regiocontrol, and without interference from the
competing Pd(II)-catalyzed cycloisomerization process.
As noticed above, under the standard cyclization/coupling
conditions, the reaction with methyl vinyl ketone departed from
the general trend in that it provided 10a as major product in
addition to the expected 9a (entry 5, Table 2). The tendency of
R,ꢀ-unsaturated ketones (and aldehydes) to give unwanted
hydroarylation (conjugate addition) products with aryl halides
or triflates, under typical Pd(0)-catalyzed Heck reaction condi-
tions, is well-known.11,12 In fact, that particular reactivity of
methyl vinyl ketone (and related conjugated ketones and
aldehydes) has already been exploited with nucleophile-tethered
(oxygen- or nitrogen-based) unsaturated derivatives in Pd(II)-
catalyzed cyclization/conjugate-addition reactions.8c,13 However,
in the reported cases a relatively high acidity of the nucleophile
(pH e 5) has been found to be a requirement for adequate
reactivity, whereas the use of relatively nonacidic alcohol
nucleophiles has not been described so far.14 Therefore, the
possibility of taking synthetic advantage of this divergent
behavior to perform alternatively Heck- or conjugate-addition-
type coupling sequences in the context of furan synthesis was
explored next. The corresponding results are displayed in Table
3. It was found that performing the reaction with Pd(OAc)2 as
catalyst in the absence of Et3N, under either air or Ar
atmospheres, completely suppressed the formation of Heck
products 9 (entries 1–6). On the other hand, a high Heck
1
conditionsb
t (h)b 9, 10c
yieldd
53
52
65
50
70
52
60
57
1
2
3
4
5
6
7
8
9
1a Pd(OAc)2, LiCl, air
1a Pd(OAc)2, LiCl, Ar
1b Pd(OAc)2, LiCl, Ar
1c Pd(OAc)2, LiCl, Ar
1d Pd(OAc)2, LiCl, Ar
1f Pd(OAc)2, LiCl, Ar
2
5
2
22
2
23
26
52
40
3
10a
10a
10b
10c
10d
10f
9a
1a Pd(PPh3)4, Ag2CO3, Et3N, air
1c Pd(PPh3)4, Ag2CO3, Et3N, air
1f Pd(PPh3)4, Ag2CO3, Et3N, air
9c
9f, 10f 66, 18
10e 1f Pd(OAc)2, LiCl, air
9f, 10f (21), (21)
a Amounts of reagents relative to 1: SmI2 (2.25 equiv), palladium catalyst
(5 mol%), H2O (1 equiv), LiCl (4 equiv), Et3N (5 equiv), Ag2CO3 (2
equiv), methyl vinyl ketone (15 equiv). b Conditions and reaction time for
the Pd-catalyzed step. c Cycloisomerization products 11 were also obtained
(<10% yield) in entries 1–6. d Isolated yield (%) of pure product. Alt-
ernatively, within parentheses, yield calculated by 1H NMR with 3,4-
dimethoxyphenylacetonitrile as internal standard. e Reaction run from
isolated 3f.
SCHEME 3. Catalytic Cycle for Formation of 5-10 from 3
(9) (a) Aurrecoechea, J. M.; Pérez, E. Tetrahedron 2004, 60, 4139–4149.
(b) Aurrecoechea, J. M.; Pérez, E. Tetrahedron Lett. 2003, 44, 3263–3266. (c)
Aurrecoechea, J. M.; Pérez, E.; Solay, M. J. Org. Chem. 2001, 66, 564–569.
(10) Leading references into the synthesis of tetrasubstituted furans: (a) Zhan,
Z. P.; Wang, S. P.; Cai, X. B.; Liu, H. J.; Yu, J. L.; Cui, Y. Y. AdV. Synth.
Catal. 2007, 349, 2097–2102. (b) Dudnik, A. S.; Gevorgyan, V. Angew. Chem.,
Int. Ed. 2007, 46, 5195–5197. (c) Sanz, R.; Miguel, D.; Martínez, A.; Alvarez-
Gutiérrez, J. M.; Rodríguez, F. Org. Lett. 2007, 9, 727–730. (d) Shindo, M.;
Yoshimura, Y.; Hayashi, M.; Soejima, H.; Yoshikawa, T.; Matsumoto, K.;
Shishido, K Org. Lett. 2007, 9, 1963–1966. (e) Watanabe, S.-I.; Miura, Y.;
Iwamura, T.; Nagasawa, H.; Kataoka, T. Tetrahedron Lett. 2007, 48, 813–816.
(f) Dey, S.; Nandi, D.; Pradhan, P. K.; Giri, V. S.; Jaisankar, P. Tetrahedron
Lett. 2007, 48, 2573–2575. (g) Miyagawa, T.; Satoh, T. Tetrahedron Lett. 2007,
48, 4849–4853, and references cited therein.
selectivity was restored by using Ag2CO3 in place of LiCl under
otherwise standard conditions (entries 7–9). Again, running the
reactions without isolation of alcohols 3 proved advantageous,
as shown by the experiment in entry 10 where an isolated
alcohol 3f was submitted to the conditions of entry 1 (expected
to give 10f selectively), resulting in a mixture of 9f and 10f
with no selectivity. Therefore, a simple switch of reaction
conditions provides access to either series of products 9 or 10
selectively. Furthermore, ethyl acrylate and methyl vinyl ketone
give noticeably different results under identical reaction condi-
tions (compare entry 1 of Table 1 with entry 1 of Table 3).
Thus, the former follows the Heck pathway exclusively whereas
the latter affords only a conjugate-addition product.
(11) Cacchi, S.; Fabrizi, G.; Goggiamani, A. ArkiVoc 2003, 8, 58–66, and
references cited herein.
(12) For geometrically unbiased acyclic systems, conditions have been devised
that provide hydroarylation products predominantly: (a) Cacchi, S.; Palmieri,
G. Synthesis 1984, 575–577. (b) Amorese, A.; Arcadi, A.; Bernocchi, E.; Cacchi,
S.; Cerrini, S.; Fedeli, W.; Ortar, G. Tetrahedron 1989, 45, 813–828. (c) Cacchi,
S. Pure Appl. Chem. 1990, 62, 713–22. (d) Arcadi, A.; Cacchi, S.; Fabrizi, G.;
Marinelli, F.; Pace, P. Tetrahedron 1996, 52, 6983–6996.
(13) (a) Lei, A.; Lu, X. Org. Lett. 2000, 2, 2699–2702. (b) Wang, Z.; Zhang,
Z.; Lu, X. Organometallics 2000, 19, 775–780. (c) Liu, G. S.; Lu, X. Y. Org.
Lett. 2001, 3, 3879–3882. (d) Zhang, Z. G.; Lu, X. Y.; Xu, Z. R.; Zhang, Q. H.;
Han, X. L. Organometallics 2001, 20, 3724–3728. (e) Liu, G. S.; Lu, X. Y.
Tetrahedron Lett. 2003, 44, 127–130. (f) Shen, Z. M.; Lu, X. Y. Tetrahedron
2006, 62, 10896–10899.
A mechanistic rationale for formation of 5–10 is provided in
Scheme 3. Thus, after the initial Pd(II)-promoted cyclization,
intermediate 13 undergoes carbopalladation followed by either
Pd-H elimination (eventually leading to 5–9) or protonation
(14) For a related Ru-catalyzed reaction, see: (a) Trost, B. M.; Pinkerton,
A. B. J. Am. Chem. Soc. 2002, 124, 7376–7389. For a mechanistic variant in a
similarly Ru-catalyzed reaction, see: (b) Trost, B. M.; Pinkerton, A. B. J. Am.
Chem. Soc. 1999, 121, 10842–10843. (c) Trost, B. M.; Pinkerton, A. B.; Seidel,
M. J. Am. Chem. Soc. 2001, 123, 12466–12476.
3652 J. Org. Chem. Vol. 73, No. 9, 2008