1h, with a methylenedioxy substitution on the benzene ring,
could afford the benzo[b]furan product 2h in an excellent
94% yield by the method. By replacing the meta-methoxy
group with a bromo group in 1a, we found that substrate 1i
could also give the desired 2i in good yield. However, we
were disappointed to find that substrate 4a (Figure 2), an
analogue of 1i by exchanging the position of the methoxy
and the bromo group, yielded no benzo[b]furan product under
the same conditions. Gratifyingly, substrate 1j, structurely
differing from compound 4a by replacing the methyl group
with a phenyl group, gave the cyclized product 2j in
relatively lower yield, with the cost of a longer reaction time
and a higher reaction temperature. Further reducing one
methoxy group on the benzene ring in substrate 1a led us to
discover that the solely para-methoxy-substituted 1k fur-
nished the desired 2k in a poor 30% yield after the reaction
mixture was refluxed for 5 h, while the meta-methoxy-
substituted compound 4b (Figure 2) was unreactive under
the same conditions. Interestingly, when the methyl group
in 1k was replaced with a phenyl group, we were pleased to
see that benzo[b]furan 2l could be obtained in moderate yield,
although at the expense of a higher reaction temperature and
a longer reaction time. On the basis of the above experimental
facts, we tentatively conclude that a para-methoxy group is
more determinant than a meta-methoxy group in the substrate
for the ring closure to occur. Furthermore, when R1 was a
Figure 2. Other models that failed to afford benzo[b]furans.
the formation of a series of N-alkoxyindole-3-carbonitriles
via a novel FeCl3-mediated intramolecular aromatic C-N
bond formation.5 Inspired by this finding, we were interested
in investigating the construction of benzo[b]furan skeletons
by a similar intramolecular oxidative C-O coupling of R-aryl
ketonitriles 1, which are readily available by the condensation
of substituted benzyl cyanides with carboxylic esters. To
initiate our study, the unsubstituted R-aryl ketonitrile 3a
(Figure 2) was chosen to test the feasibility of the ring closure
mediated by FeCl3. However, it was found that substrate 3a
was inert, and no desired benzo[b]furan compound was
achieved when applying various oxidants under many
conditions.6 A further screen of the substrate pattern led us
to discover that substrate 1a, with 3,4-dimethoxy substituted
on the benzene ring, could be conveniently converted to the
desired benzo[b]furan 2a in 70% yield when oxidized by
FeCl3 in DCE at room temperature (entry 1, Table 1). This
result implies that the presence of electron-donating methoxy
groups on the benzene ring of the substrate was crucial for
the oxidative intramolecular cyclization to occur.
Table 2. Further Variations of E groups in Substrate 1a
Since the importance of an electron-rich substrate is
obvious, we first sought to probe the substrate scope of this
FeCl3-mediated oxidative reaction by changing the R1 groups
while keeping the 3,4-dimethoxy benzene moiety intact. The
results listed in Table 1 demonstrated that when R1 was a
relatively long-chained propyl group (entry 2, Table 1), a
bulkier tert-butyl group (entry 3, Table 1), or benzyl group
(entry 4, Table 1), the substrates could undergo ring closure
to give the desired corresponding benzo[b]furan products in
moderate yields, although these substrates needed relatively
longer reaction times. For the substrates with R1 being an
aryl group, the reactions also furnished the desired cyclized
products in acceptable yields, with shorter reaction times
(entries 5-7, Table 1).
Next, we decided to investigate the substitution variations
of the electron-rich substrates. It was observed that substrate
(4) (a) Brasche, G.; Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47,
1932. (b) Chen, X.; Hao, X.-S.; Goodhue, C. E.; Yu, J.-Q. J. Am. Chem.
Soc. 2006, 128, 6790. (c) Inamoto, K.; Saito, T.; Katsuno, M.; Sakamoto,
T.; Hiroya, K. Org. Lett. 2007, 9, 2931. (d) Thu, H.-Y.; Yu, W.-Y.; Che,
C.-M. J. Am. Chem. Soc. 2006, 128, 9048. (e) Tsang, W. C. P.; Zheng, N.;
Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 14560. (f) Ueda, S.;
Nagasawa, H. Angew. Chem., Int. Ed. 2008, 47, 6411.
(5) Du, Y.; Chang, J.; Reiner, J.; Zhao, K. J. Org. Chem. 2008, 73,
2007.
(6) (a) No desired product achieved when using FeCl3 as oxidant and
carrying out the reaction in DCE at rt to reflux. (b) Other oxidative
conditions: K3Fe(CN)6/MeCN, Cu(OAc)2/AcOH, Mn(OAc)3/AcOH, CAN/
CH3CN, and MnO2/CH2Cl2 were also ineffective for this conversion, since
substrate 3a would dimerize under these conditions. For such homocoupling
reactions, see: De Jongh, H. A. P.; De Jonge, C. R. H. I.; Mijs, W. J J.
Org. Chem. 1971, 36, 3160, and references cited therein.
a Conditions: See Table 1. b Isolated yields after silica gel chromatography.
c Reaction occurred at 40 °C. d Complex mixture, 30% of 1t recovered.
4980
Org. Lett., Vol. 11, No. 21, 2009