A new redox-denitration reaction of aromatic nitro compounds

Article abstract of DOI:10.1039/a910290f

When 4,4-dimethyl-1-(2-nitrophenyl)pyrazolidin-3-one 9 is heated in pyridine containing pyridine hydrochloride it is transformed into 4,4- dimethyl-1-phenylpyrazolidin-3,5-dione 10 in which the methylene group has been oxidised and the nitro group has dis

Full text of DOI:10.1039/a910290f

A new redox-denitration reaction of aromatic nitro compounds
Charles W. Rees*a and Siu C. Tsoib
a Department of Chemistry, Imperial College of Science, Technology and Medicine, London, UK SW7 2AY.
E-mail: c.rees@ic.ac.uk
b Kodak European Research and Development, Headstone Drive, Harrow, Middlesex, UK HA1 4TY
Received (in Liverpool, UK) 23rd December 1999, Accepted 3rd February 2000
When 4,4-dimethyl-1-(2-nitrophenyl)pyrazolidin-3-one 9 is
heated in pyridine containing pyridine hydrochloride it is
transformed
into
4,4-dimethyl-1-phenylpyrazolidin-
3,5-dione 10 in which the methylene group has been oxidised
and the nitro group has disappeared; two further examples
of the same reaction, of pyrazolidinones 4 and 7, are reported
together with a mechanistic rationalisation of this curious
reaction.
New reactions of aromatic nitro groups are rare. We recently
discovered one during an unsuccessful attempt to synthesise the
1-arylpyrazolidin-3-one 4 from 4,5-dimethoxy-2-nitrophenyl-
hydrazine 1 and 3-chloropivaloyl chloride 2. After an initial
reaction in cold pyridine which gave 3,† mp 163 °C (isolable in
90% yield), the reaction mixture was heated under reflux for 24
h. From this complex reaction we could not isolate any 4, but
only a rather low yield of the 1-arylpyrazolidin-3,5-dione 5,†
mp 189–191 °C (23%), in which the nitro group has been
replaced by hydrogen and the pyrazolidinone methylene group
oxidised to a carbonyl group (Scheme 1). Compound 5 was
synthesised independently from 3,4-dimethoxyphenylhydra-
zine and dimethylmalonyl dichloride. We assume that 1 and 2
are converted via 3 into the desired pyrazolidinone 4, but this
has reacted further to give 5. On heating preformed 3 in pyridine
for 20 h, 5 was again formed as the major product. This
unexpected conversion of 4 into 5 is a novel redox reaction in
which the nitro group has presumably oxidised the adjacent
methylene group and has undergone, most unusually, complete
cleavage from the aromatic ring with overall loss of the
elements of nitroxyl, HNO.
Scheme 2 Reagents and conditions: i, pyridine, 5 °C, N2, 1 h, then room
temp., 1 h, then 115 °C, 2 h, ii; pyridine 115 °C, N2, 20 h.
We performed the same reaction (Scheme 2) with the
monomethoxyphenylhydrazine 6 and acid chloride 2 in cold
pyridine, to give the acylic hydrazide intermediate (TLC),
followed by brief heating (2 h) to give the expected 1-arylpyr-
azolidin-3-one 7 as the major product. After extended heating
(20 h) a mixture of 7 (25%) and the denitrated pyrazolidin-
3,5-dione 8 (19%) was isolated. The formation of the mono-
methoxy product 8 is distinctly slower than for the dimethoxy
product 5.
Finally we treated 2-nitrophenylhydrazine with acid chloride
2 in cold pyridine and then at 90 °C for 11 h to give the
nitrophenylpyrazolidin-3-one 9,† mp 196–198 °C (43%)
(Scheme 3). When pure compound 9 was heated in pyridine for
up to 5 days there was very little reaction,‡ but when heated in
pyridine containing pyridine hydrochloride (1 equiv.) to
simulate the conditions of the reactions of 1 and 6, the overall
loss of HNO again occurred to give 1-phenylpyrazolidin-
3,5-dione 10 (mp 180 °C, lit.1 180–182 °C) (23%) after 20 h.
Product 10 was also synthesised from phenylhydrazine and
dimethylmalonyl dichloride.
This new acid-catalysed redox-denitration reaction could
thus be general for the conversion of 1-(2-nitroaryl)pyrazolidin-
3-ones, like 4, 7 and 9, into 1-arylpyrazolidin-3,5-diones, like 5,
8 and 10. A possible mechanism is outlined in Scheme 4, for the
simplest case. Aromatic nitro groups are well known to interact
in thermal, photolytic and catalysed reactions with a range of
ortho substituents,2 often being reduced to nitroso, azo, azoxy
or amino groups, but very rarely with concomitant cleavage of
the aryl–nitrogen bond.3 Since the methylene group would be
activated towards oxidation by the adjacent pyrazolidinone
Scheme 1 Reagents and conditions: i, pyridine, 5 °C, N2, 1 h; ii, pyridine,
115 °C, N2, 24 h.
Scheme 3 Reagents and conditions: i, pyridine, 50 °C, N2, 1 h, then 90 °C,
11 h; ii, pyridine, py·HCl (1 equiv.), 115 °C, N2, 20 h.
DOI: 10.1039/a910290f
Chem. Commun., 2000, 415–416
This journal is © The Royal Society of Chemistry 2000
415

outcome may depend critically upon the particular structural
features of the starting pyrazolidinones. Since protonation at the
benzene 2-position must presumably be involved, the reaction
should be acid-catalysed (as observed), inter- or intra-molec-
ularly. It could occur, for example, in the conjugate acid 16 of
15 as shown (arrows in 16). In this mechanism (Scheme 4)
various steps would be facilitated by the electron releasing
groups in the mono- and di-methoxy compounds, in accord with
the observed relative rates of reaction. Further work is required
to establish the scope and mechanism of this new reaction.
We are grateful to Kodak Ltd. for permission to publish this
work and to Dr N. E. Milner for support and encouragement.
Notes and references
† The structures of all new compounds are based upon IR, MS, LCMS,
HRMS and 1H NMR spectroscopy.
‡ Some intractable material containing traces of 10 (mass spectrometry)
was formed.
Scheme 4
1 W. H. Pirkle and P. L. Gravel, J. Org. Chem., 1978, 43, 808.
2 P. N. Preston and G. Tennant, Chem. Rev., 1972, 72, 627.
3 R. Fielden, O. Meth-Cohn and H. Suschitzky, Tetrahedron Lett., 1970,
1229 report such a reaction in the formation of a by-product in the
thermolysis of 1-diethylamino-2-nitrobenzene but the loss of the nitro
group is not accompanied by oxidation of an adjacent methylene
group.
4 O. Meth-Cohn and H. Suschitzky, Adv. Heterocycl. Chem., 1972, 14,
211; O. Meth-Cohn, Adv. Heterocycl. Chem., 1996, 65, 1.
5 P. A. S. Smith and G. E. Hein, J. Am. Chem. Soc., 1960, 82, 5731; M. N.
Hughes, Q. Rev. Chem. Soc., 1968, 22, 1
nitrogen atom, it seems likely that the first step would be an
intramolecular hydride transfer to the nitro group (arrows in 11).
This is exactly analogous to that proposed for tertiary amines,
when the amine replaces our heterocyclic ring (the tert-amino
effect).4 This would give the iminium ion 12 which could
collapse to the nitroso compound 13 or its tricyclic tautomer 14;
13 and 14 would be in equilibrium with each other and with the
hydroxylamine 15.
The final and most unusual step would then be the loss of
nitroxyl, HNO,5 from 15 to give the isolated product 10; this
Communication a910290f
416
Chem. Commun., 2000, 415–416