4952
J. Chauhan, S. Fletcher / Tetrahedron Letters 53 (2012) 4951–4954
nitroacylbenzenes, respectively.8 The classical combination of Sn
or SnCl2 in concentrated HCl to effect the transformation of
2-nitroacylbenzenes into 2,1-benzisoxazoles is harsh and many
moieties, particularly acid–labile groups, are expected to be incom-
patible with such conditions. Milder approaches to the synthesis of
2,1-benzisoxazoles would be welcomed. In this Letter, we investi-
gate the scope and limitations of the SnCl2Á2H2O-mediated tandem
reduction–heterocyclization of 2-nitroacylbenzenes under neutral
conditions in organic solvents.
The conditions for the reductive heterocyclization reaction of
substrate 1 (Table 1) were first optimized. With 3 equiv of
SnCl2Á2H2O in EtOAc at 50 °C, complete consumption of the start-
ing material was realized after 1 h. However, the yield of 2 was
only 57%, with the mass balance identified as aniline 3. Reducing
the temperature to 25 °C led to a reduction in the conversion of 1
to 2 with about 70% of the starting material recovered. Signifi-
cantly, there was no aniline (3) produced in the reaction, suggest-
ing heat is necessary to accomplish the undesired over-reduction.
Enlisting MeOH as a co-solvent allowed the reaction to proceed at
room temperature with complete consumption of the starting
material over 20 h with 3 equiv of SnCl2Á2H2O, delivering benzis-
oxazole 2 in 89% yield. Presumably, the benefit of MeOH was its
ability to promote the dissolution of SnCl2Á2H2O. Mechanistically,
this reaction requires just 2 equiv of SnCl2Á2H2O (reduction of ni-
tro to hydroxylamine). Attempts to reduce the amount of toxic tin
waste produced, however, resulted in lower yields of 2, either
through incomplete reactions at room temperature or through
over-reduction at 50 °C to afford a mixture of benzisoxazole 2
and aniline 3. Therefore, using the optimized conditions of
Instead, the only product isolated was 2-aminobenzamide (5o),
in which the nitro group had been completely reduced to an amine
and the cyano group had undergone an effective hydration reaction
to a carboxamide. It is known that the cyano group is inert to our
reaction conditions.8 Furthermore, treatment of 4-nitrobenzonitri-
le with SnCl2Á2H2O under our optimized conditions led to the ex-
pected 4-aminobenzonitrile.
It is surmised that 5o was produced from a chain reaction of
three events: reduction, heterocyclization, and reductive ring
opening;9 our postulated mechanism is shown in Scheme 2. First,
the usual partial reduction of the nitro group generates the corre-
sponding hydroxylamine, which then attacks the nitrile carbon in a
favoured 5-exo-dig cyclization to furnish benzo[c]isoxazol-3(1H)-
imine, which is in tautomeric equilibrium with the anticipated 3-
amino-2,1-benzisoxazole. The ‘‘leaving group’’ here is the C=N
p-
bond that can be re-formed, which is in contrast to the irreversible
expulsion of water observed with aldehyde and ketone substrates.
In the imine tautomer, the heterocycle is no longer aromatic and is
more readily reduced by SnCl2Á2H2O, which accounts for the isola-
tion of only 5o under our reaction conditions. Attempts to isolate
the postulated 3-amino-2,1-benzisoxazole through the use of less
SnCl2Á2H2O, lower temperatures, and shorter reaction times proved
unsuccessful. This overall reduction–neighbouring group hydra-
tion reaction is in contrast to a number of publications that report
the expected 2-aminobenzonitrile derivatives from a range of 2-
nitrobenzonitrile substrates under both neutral8 and acidic10
(HCl) conditions with SnCl2Á2H2O. Apparently, the unique combi-
nation of our reaction conditions and specific substrate 4o allowed
the intramolecular attack of the cyano group by the intermediate
hydroxylamine moiety to outcompete the reduction of the hydrox-
ylamine to the amine. We intend to investigate this chemistry fur-
ther to determine the stereo-electronic factors that govern the
outcome of the reaction of 2-nitrobenzonitrile substrates with
SnCl2Á2H2O.
3 equiv of SnCl2Á2H2O in
a 1:1 mixture of EtOAc/MeOH at
room temperature, the generality of this chemistry was next
investigated.
The results of our investigation are shown in Table 2. By com-
paring entries 1, 9, and 12, it is revealed that aldehydes and ke-
tones are equally susceptible to the reductive heterocyclization
reaction conditions. Of particular note are entries 5, 7, and 8, which
demonstrate that our mild reaction conditions are compatible with
the acid–labile tert-butyl ether, tert-butyl ester, and Boc carbamate
groups, respectively. In addition to aldehydes and ketones, we
were curious to see if our methodology could be applied to other
acyl-containing groups. 2-Nitrobenzoic acid (4l) did not generate
the expected 3-hydroxy-2,1-benzisoxazole, however, and instead
delivered the complete reduction product 5l; the low yield was
presumably due to difficulty in the work-up of the amino acid.
Likewise, treatment of esters 4m and 4n with SnCl2Á2H2O fur-
nished anilines 5m and 5n, respectively, in good to excellent yields.
It is unclear at this stage if these products were generated through
intermediary 3-hydroxy- (in the case of 5l), 3-methoxy- (5n), or 3-
tert-butoxy- (5n) 2,1-benzisoxazoles. On the other hand, the reac-
tion of 2-nitrobenzonitrile (4o) did not deliver the expected 3-ami-
no-2,1-benzisoxazole or the alternative 2-aminobenzonitrile.
In conclusion, we have investigated the scope of the SnCl2Á
2H2O-mediated reductive heterocyclization of 2-nitroacylbenz-
enes to their corresponding 2,1-benzisoxazoles under essentially
neutral conditions. Aldehydes and ketones proved excellent sub-
strates, and a wide variety of functional groups, including com-
monly used acid–labile protecting groups, were tolerated.
A
limitation of this chemistry appears to reside in the application
to substrates hosting an acid, ester, or nitrile, in place of an alde-
hyde/ketone, ortho to the nitro group, which would be expected
to furnish the corresponding 3-hydroxy-, 3-alkoxy-, or 3-amino-
2,1-benzisoxazoles, respectively; instead, complete reduction of
the nitro function to the amino group was observed. In such cases,
it is likely that the reaction mechanism proceeds via the expected
3-substituted-2,1-benzisoxazole but, through their abilities to tau-
tomerize to structures of reduced aromaticities, they are more
prone to further reduction. Nevertheless, the mildness of the meth-
odology reported herein suggests that it will prove a useful alter-
Table 1
Reductive heterocyclization reaction conditions for conversion of 1 into 2
SnCl2Á2H2O (equiv)
Solvent
Temperature (°C)
Time (h)
Yielda (%)
1
2
3
3
1
2
2.5
3
2
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc/MeOH (1:1)
EtOAc/MeOH (1:1)
EtOAc/MeOH (1:1)
EtOAc/MeOH (1:1)
EtOAc/MeOH (1:1)
50
50
50
25
25
25
25
25
50
1
1
1
20
20
20
20
20
1
22
45
57
31
47
66
80
89
63
a
Isolated yield after work-up and silica gel flash column chromatography.
Chauhan, Jay
