Synthesis of 2-aryl-2H-indazoles by base catalysed reaction of 2-nitrobenzyl triphenylphosphonium bromide and aryl isocyanates

Article abstract of DOI:10.1016/S0040-4039(00)01757-3

The synthesis of a series of 2-aryl-2H-indazoles is reported. These compounds are obtained in moderate to good yield by reaction of 2-nitrobenzyl triphenylphosphonium bromide with aryl isocyanates, catalysed by sodium hydride or DBU. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01757-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 9893–9897
Synthesis of 2-aryl-2H-indazoles by base catalysed reaction
of 2-nitrobenzyl triphenylphosphonium bromide and aryl
isocyanates
Abutariq Taher, Sweta Ladwa, Srinath Thirumalai Rajan and George W. Weaver*
Department of Chemistry, Loughborough University, Loughborough LE11 3TU, UK
Received 15 August 2000; accepted 4 October 2000
Abstract
The synthesis of a series of 2-aryl-2H-indazoles is reported. These compounds are obtained in moderate
to good yield by reaction of 2-nitrobenzyl triphenylphosphonium bromide with aryl isocyanates, catalysed
by sodium hydride or DBU. © 2000 Elsevier Science Ltd. All rights reserved.
As part of our studies on developing new heterocyclisation reactions,1 we have been
investigating reactions where the nitrogen atom of a nitro substituent on an aromatic or
heteroaromatic ring undergoes transformation into a new heterocyclic ring nitrogen. We have
shown that nitroimidazolyl phosphoramidates undergo reaction with aryl isocyanates to form
imidazo fused [1,2,3]triazoles,2 and now wish to report our findings on the reaction of
nitrobenzyl phosphonium salts with aryl isocyanates (Scheme 1). We have found that 2-
nitrobenzyl triphenylphosphonium bromide reacts smoothly with aryl isocyanates in the pres-
ence of a base, such as sodium hydride or DBU, to form 2-aryl-2H-indazoles in moderate to
good yield; the nitro group nitrogen being transformed into the indazole N-1 atom.
Scheme 1.
Indazoles are pharmacologically important compounds and the indazole ring system forms
the basis of a number of drug molecules including granisetron,3 a 5-HT3 receptor antagonist
* Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01757-3

9894
used as an anti-emetic in cancer chemotherapy, and benzydamine,4 an anti-inflammatory agent.
Indazoles5 can be synthesised by a number of methods, usually involving formation of the NꢀN
bond as the key step. One of the now classical methods6 for their synthesis is the deoxygenation
of 2-nitroanils 3 with triethyl phosphite, developed by Cadogan and Mackie, shown in Scheme
2.
Scheme 2.
More recently Song and Yee7 have used a palladium catalysed cyclisation of bromobenzyl aryl
hydrazines 4 (Scheme 3) to synthesise indazoles, in this case forming a CꢀN bond as the key ring
closing step. Our route to indazoles also employs a nitro compound with readily available
nitrobenzyl triphenylphosphonium bromide acting as the starting material. Deprotonation to
form the purple ylide can be easily be effected with a range of bases. We have found DBU to
be the best base to use in this reaction. Treatment of the ylide with an aryl isocyanate and
heating the mixture at 60°C in acetonitrile then affords the 2-aryl indazoles† shown in Table 1.
Scheme 3.
Table 1
Indazoles 2 synthesised†
Ar
Yield (%)
a
b
c
d
e
f
4-MeOC6H4
4-PhOC6H4
576,10
45
2-MeO-3-MeC6H3
4-EtO2CꢀC6H4
4-FꢀC6H4
3,4,5-(MeO)3C6H2
3-NCꢀC6H4
40
6011,12
5511,13
62
g
5311
The reaction works for aryl isocyanates with either electron withdrawing, or electron donating
substituents. Sodium hydride can also be employed as the base, although in this case the
indazoles were accompanied by small amounts of 2-nitrotoluene as a by-product. We believe this
is formed by traces of sodium hydroxide in the hydride reagent which attack the phosphonium
† New compounds exhibited satisfactory analytical and spectroscopic data.

9895
cation forming a pentacoordinate phosphorane. This can then cleave to form triphenylphos-
phine oxide, and the nitrotoluene anion, which is subsequently protonated during work-up.
The reaction to form the indazole ring involves loss of triphenylphosphine oxide and carbon
dioxide, and several mechanisms can be considered to transform the nitro compound into the
indazole nucleus. We have considered two possible pathways as the most likely that could lead
to closure of the pyrazole ring. In the first case a Wittig reaction may occur between the ylide
5 and the isocyanate to form a ketimine such as 6 (Scheme 4). This could undergo attack by an
oxygen atom of the adjacent nitro group leading to 7. Ring opening and reclosure through the
aryl-substituted nitrogen would generate the cinnoline-N-oxide 8. A second ring opening would
produce the ketene 9, which may be attacked by the oxygen atom of the neighbouring azoxy
group to form 10. Loss of carbon dioxide would then produce the azo-carbene intermediate 11,
which would readily undergo electrocyclic ring closure to form the 2-indazole. Such a mecha-
nism is closely related to that proposed by Rees8 to account for the formation of phenyl
benzotriazole in the thermolysis of 2-nitrophenyl carbodiimide.
Scheme 4.
Alternatively a mechanism involving acylation of a nitro group oxygen atom may operate to
form an acyl nitronate intermediate of the type 13 (Scheme 5). Aliphatic nitro compounds are
known to undergo acylation by isocyanates in the dehydration reaction, to form nitrile oxides.9

9896
We can find no reports on the acylation of aromatic nitro compounds, and aryl isocyanates are
inert to nitrobenzene and 2-nitrotoluene up to 150°C. It is possible that the aryl nitro substituent
is more nucleophilic, due to delocalisation of negative charge from the ylide carbon, as shown
by resonance structure 12. Delocalisation is supported by the intense purple colour of the ylide.
The acyl nitronate could then be converted into the indazole as shown in Scheme 5. Cyclisation
of the carbamate nitrogen onto the methine carbon would produce the seven-membered ring
intermediate 14. Triphenylphosphine oxide and carbon dioxide could then be lost from the
bridged structure 15 to generate the indazole ring. Loss of these two compounds may equally
well be separate events. At present we have no evidence to support either of the two pathways
and further work is underway to determine the mechanism of this reaction which will be
reported in due course.
Scheme 5.
In summary we believe this reaction represents a convenient way to synthesise indazole
derivatives substituted at the 2-position with a variety of aromatic groups.
Acknowledgements
We thank the EPSRC for a studentship (to A.T.), and Loughborough University and Pfizer
Central Research for financial support.

9897
References
1. Taher, A.; Slawin, A. M. Z.; Weaver, G. W. Tetrahedron Lett. 1999, 40, 8157.
2. Taher, A.; Eichenseher, S.; Weaver, G. W. Tetrahedron Lett. 2000, 41, 9889.
3. Bermudez, J.; Fake, C. S.; Joiner, G. F.; Joiner, K. A.; King, F. D.; Miner, W. D.; Sanger, G. J. J. Med. Chem.
1990, 33, 1924.
4. Boehm, R.; Hirschelmann, R. Pharmazie 1980, 4, 232.
5. For a review on indazoles, see: Elgue´ro, J. In Comprehensive Heterocyclic Chemistry; Potts, K. T, Ed.; Pergamon:
Oxford, 1984; Vol. 5, p. 167; Elgue´ro, J. In Comprehensive Heterocyclic Chemistry II; Shinkai, I., Ed.; Elsevier
Science: Oxford, 1996; Vol. 3, p. 1.
6. Cadogan, J. I. G.; Cameron-Wood, M.; Mackie R. K.; Searle, R. J. G. J. Chem. Soc. 1965, 4831.
7. Song, J. J.; Yee, N. K. Org. Lett. 2000, 2, 519.
8. Houghton, P. G.; Pipe D. F.; Rees, C. W. J. Chem. Soc., Chem Commun. 1979, 771.
9. For a discussion of nitrile oxide generation, see: Torssell, K. B. G. Nitrile Oxides, Nitrones and Nitronates in
Organic Synthesis; VCH: Weinheim, 1988; p. 55.
10. Frontana-Uribe, B. A.; Moinet, C. Tetrahedron 1998, 54, 3197.
11. Picciola, G.; Ravenna, F.; Carenini, G.; Gentili, P.; Riva, M. Farmaco Ed. Sci. 1981, 36, 1037.
12. Armour, M.-A.; Cadogan, J. I. G.; Grace, D. S. B. J. Chem. Soc., Perkin Trans 2 1975, 1185.
13. Elgue´ro, J.; Estopa´, C.; Ilavsky, D. J. Chem. Res. 1981, 4237.
.