path I),4 cine-substitution/intramolecular Michael type
addition of 2-nitrofurans with β-dicarbonyl compounds
(Scheme 1, path II),5 rhodium-catalyzed 1,3-dipolar cyclo-
addition of diazo-carbonyls/iodonium ylides to furans
(Scheme 1, path III),6 and palladium-catalyzed dearoma-
tization of furans via electrocyclic ring-closure (Scheme 1,
path IV).7
Afterward, we examined other inorganic bases, such as
CuO and Al2O3, both of which led to 5a in a low yield
(Table S1, entries 3À4). However, with K2CO3 as the base,
5a was obtained in 84% yield (Table S1, entry 5). The
transformation did not occur in the absence of base
(Table S1, entry 6). Then, we examined the influence of
diverse organic bases on the reaction, such as pyridine,
DABCO, piperidine, DBU, DIPEA, DMAP, and NEt3
(Table S1, entries 7À13). It was observed that most of the
organic bases could promote the reaction. Among these,
NEt3 proved to be the best, where 5a was isolated in 87%
yield (Table S1, entry 13). However, only a low yield was
obtained in other solvents (EtOH, CH2Cl2, CH3CN,
DMF, and DMSO) (Table S1, entries 14À18). Unfortu-
nately, the targeted product was not detected in AcOH
(Table S1, entry 19).
Scheme 1. Methods for the Synthesis of 3a,6a-Dihydrofuro-
[2,3-b]furans
With the optimized conditions in hand, the scope of the
unsymmetrical 1,4-enediones was investigated. The results
are shown in Scheme 2. It is noteworthy that both electron-
rich and -deficient substrates could provide the desired
products smoothly in good to excellent yields (64À96%).
The 1,4-enediones, bearing substituted groups such as
4-Me, 4-OMe, and 4-NO2 on the phenyl rings, reacted
with 4 smoothly to afford the corresponding products
(72À85%, 5b, 5c, and 5g). Good to excellent yields were
also obtained for halo-substituted substrates (75À96%,
5dÀf, 5i, and 5j). Among them, a 2,4-dichloro substituted
substrate showed the best result, with a yield of 96% (3i).
Sterically hindered R-naphthyl and β-naphthyl also af-
forded the corresponding desired products 5n and 5o in
75% and 79% yields, respectively. To our delight, the
substrates with a heteroaryl group for R1, such as 2-furyl,
3-thienyl, and 2-benzofuryl, delivered the products suc-
cessfully in moderate to excellent yields (64À86%, 5kÀm).
When R2 = CH3, 4-NO2C6H4, 3,4,5-(MeO)3C6H2, and
2-furyl, the reaction could also be tolerated in 65À76%
yields (5pÀs). When R3 = CN, COPh, and COCH3, good
to excellent yields have been achieved (86À91%, 5tÀv).
Moreover, when R2 = CH3 = R3 COCH3 (5w), the
reaction performed providing the desired product cleanly
in 76% yield. Moreover, the hydroxyl group substituted
substrate could not afford the desired product (5h). The
structure of 5f and 5g were further determined by X-ray
crystallographic analysis (see SI).
In our previous work, we found that 1,3-dicarbonyl
compounds could react with methyl ketones easily to
afford unsymmetrical 1,4-enediones.8 On the basis of the
facile access to unsymmetrical 1,4-enediones, we consid-
eredusingthesematerialsasusefulprecursorstosynthesize
diverse heterocyclic compounds.9 Fortunately, the fused
ring 3a,6a-dihydrofuro[2,3-b]furans were obtained by the
reaction of unsymmetrical 1,4-enediones with malono-
nitrile (Scheme 1b). To the best of our knowledge,
the transformation of 3a,6a-dihydrofuro[2,3-b]furans via
the novel bicyclizition reaction has not been reported to
date yet.
To optimize the reaction conditions, we attempted to
treat ethyl 2-benzoyl-4-oxo-4-phenylbut-2-enoate (3a)
with malononitrile (4) under different reaction conditions.
Various bases and solvents were examined at room tem-
perature, as shown in Table S1 (see Supporting Informa-
tion (SI)). To our delight, the reaction of 3a (1.0 mmol)
with 4 (1.0 mmol) and MgO (1.0 mmol) performed well to
give 5a in 80% yield in CH3OH in 2 h (Table S1, entry 1).
After discovering a novel bicyclization reaction (Scheme 3b),
we began to explore a more efficient approach to syn-
thesize the 3a,6a-dihydrofuro[2,3-b]furan derivatives from
readily available starting materials. We discovered that
methyl ketone 1 could be transformed into compound 3
via a domino reaction, which involved iodination,10
Kornblum oxidation,11 and Knoevenagel condensation12
(Scheme 3a). Based on this, we aimed to create a reaction
chain consisting of the domino reaction I and domino
(6) For selected examples, see: (a) Schank, K.; Lick, C. Synthesis
1983, 392–395. (b) Pirrung, M. C.; Zhang, J.; McPhail, A. T.
J. Org. Chem. 1991, 56, 6269–6271. (c) McCarthy, N.; McKervey,
M. A.; Ye, T.; McCann, M.; Murphy, E.; Doyle, M. P. Tetrahedron
Lett. 1992, 33, 5983–5986. (d) Pirrung, M. C.; Zhang, J. Tetrahedron
€
Lett. 1992, 33, 5987–5990. (e) Muller, P.; Fernandez, D. Helv. Chim.
Acta 1995, 78, 947–958. (f) Davies, H. M. L.; Calvo, R. L. Tetrahedron
Lett. 1997, 38, 5623–5626.
(7) Yin, B.-L.; Zeng, G.-H.; Cai, C.-B.; Ji, F.-H.; Huang, L.; Li,
Z.-R.; Jiang, H.-F. Org. Lett. 2012, 14, 616–619.
(8) Gao, M.; Yang, Y.; Wu, Y.-D.; Deng, C.; Cao, L.-P.; Meng,
X.-G.; Wu, A.-X. Org. Lett. 2010, 12, 1856–1859.
(10) Yin, G. D.; Gao, M.; She, N. F.; Hu, S. L.; Wu, A. X.; Pan, Y. J.
Synthesis 2007, 3113–3116.
(11) Kornblum, N.; Powers, J. W.; Anderson, G. J.; Jones, W. J.;
Larson, H. O.; Levand, O.; Weaver, W. M. J. Am. Chem. Soc. 1957, 79,
6562–6562.
(9) (a) Yang, Y.; Gao, M.; Wu, L.-M.; Deng, C.; Zhang, D.-X.; Gao,
Y.; Zhu, Y.-P.; Wu, A.-X. Tetrahedron 2011, 67, 5142–5149. (b) Gao,
M.; Yang, Y.; Wu, Y.-D.; Deng, C.; Shu, W.-M.; Zhang, D.-X.; Cao,
L.-P.; She, N.-F.; Wu, A.-X. Org. Lett. 2010, 12, 4026–4029.
(12) Knoevenagel, E. Chem. Dtsch. Ber. Ges. 1896, 29, 172–174.
Org. Lett., Vol. 15, No. 3, 2013
457