A practical and chemoselective reduction of nitroarenes to anilines using activated iron

Article abstract of DOI:10.1002/adsc.200404236

Reduction of nitroarenes to anilines using activated iron generated by Fe/HCl or Zn/FeSO₄ is described. A variety of functional groups such as alkyne, ketone, enone, nitrile, lactone, and aromatic halide are well tolerated under these conditions.

Full text of DOI:10.1002/adsc.200404236

COMMUNICATIONS
A Practical and Chemoselective Reduction of Nitroarenes to
Anilines Using Activated Iron
Yugang Liu,* Yansong Lu, Mahavir Prashad,* Oljan Repi cˇ , Thomas J. Blacklock
Process Research and Development, Department of Chemical and Analytical Development, Novartis Institute for
Biomedical Research, One Health Plaza, East Hanover, New Jersey 07936, USA
Fax: (þ1)-973-781-2188, e-mail: yugang.liu@pharma.novartis.com
Received: July 22, 2004; Accepted: December 6, 2004
Abstract: Reduction of nitroarenes to anilines using
We elected to use iron as the reducing agent because it
is cheap and readily available, and known to tolerate a
variety of functionalities. Furthermore, palladium cata-
lyst-based conditions often afford difficult to remove
residues. To our disappointment, reduction of (A) using
Fe-AcOH was slow, gave poor conversion (37%) and
produced several by-products, one of which (C, 7%) re-
sulted from reduction of both the nitro group and the
propargyl group. This by-product was difficult to re-
move by either silica gel chromatography or by crystal-
lization of a variety of salts. We next studied this reduc-
activated iron generated by Fe/HCl or Zn/FeSO is
4
described. A variety of functional groups such as al-
kyne, ketone, enone, nitrile, lactone, and aromatic
halide are well tolerated under these conditions.
Keywords: anilines, chemoselectivity, iron, nitroar-
enes, reduction, zinc
tion using Fe-NH Cl. However, the reduction was in-
4
Numerous methods have been reported in the literature complete even at reflux in ethanol. One interesting ob-
[
1]
for the reduction of nitroarenes to anilines. The most servation was that the conversion of the starting materi-
[
2]
commonly used methods utilize iron, such as Fe- al was not consistent from run to run, and only in one
NH Cl, Fe-AcOH, Fe-HCl, Fe-CaCl , FeCl -Zn-DMF- case when additional NH Cl was added after the reac-
4
2
3
4
[
3]
H O, FeS-NH Cl-CH OH-H O, etc., and zinc, such tion mixture was heated at reflux for 16 h, was a com-
2
4
3
2
as Zn-AcOH, Zn-NaOH, Zn-NH Cl, Zn-CaCl , etc. plete conversion observed. However, we were unable
4
2
Less commonly used are methods utilizing indium-ultra- to reproduce these results. We reasoned that although
[
4]
[5]
sound and indium-dichlorotitanocene. Reductions prolonged heating of the substrate with Fe-NH Cl did
4
[
6]
using catalytic hydrogenation are also well known. Se- not improve the conversion, the surface of the iron
lective methods for the reduction of the nitro group in might have been altered (perhaps cleaned) with pro-
the presence of other reducible functional groups are longed heating, thereby activating the iron, and the ad-
highly desirable. In a development program, we needed dition of more NH Cl triggered complete reduction of
4
to find an efficient method for the reduction of an aro- the substrate. If this hypothesis was correct, then pre-
matic nitro group in (A) to the corresponding aniline treatment of iron using astronger acid such as HCl might
(
B) that would tolerate an aromatic ketone and a prop- activate the iron surface more efficiently, and the reac-
argyl functional group and that was suitable for a large tion could be made reproducible. Thus, a suspension of
scale process in the pilot plant.
iron powder (5.0 equivs.) in ethanol was heated with
concentrated HCl (0.5 equivs.) at 658C for 2 h, and the
mixture was cooled to 558C. A solution of NH Cl solu-
4
tion was added, followed by the addition of the nitroar-
ene (1.0 equiv.) in portions. The reaction was exothermic
and was complete in 1–3 h. The corresponding aniline
product was obtained in excellent yield with <1.0% of
the allyl by-product. Normally, NH Cl solution was nec-
4
essary for the reaction to go to completion, however, for
reactive substrates such as 2-bromo-5-chloro-nitroben-
zene, the reduction proceeded smoothly to completion
even without the use of NH Cl. Because concentrated
4
HCl was consumed during the initial activation stage,
the reaction medium was virtually neutral, which was
in contrast to those methods involving Fe/HCl where
acid-sensitive functional groups did not survive. By sim-
Adv. Synth. Catal. 2005, 347, 217–219
DOI: 10.1002/adsc.200404236
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
217

COMMUNICATIONS
Yugang Liu et al.
Table 1. Reduction of nitroarenes to anilines.
ply activating iron in situ with concentrated HCl, we with zinc powder in the presence of NH Cl solution in
4
have discovered a convenient, highly chemoselective ethanol. We found that the reduction worked very
and reproducible reduction of nitroarenes to the corre- well, and similar results were obtained under these con-
sponding anilines. Interestingly, other acids such as ditions. The results obtained using both methods are
AcOH, H SO and H PO did not provide the same ef- summarized in Table 1. An examination of Table 1 sug-
2
4
3
4
fect as HCl.
gested that a variety of functional groups were well tol-
[7]
To further test the idea that an activated form of iron erated under these conditions.
was indeed operational in the reduction of nitroarenes In conclusion, reduction of nitroarenes to anilines us-
to the corresponding anilines, we generated an activated ing activated iron generated by Fe/HCl or Zn/FeSO is
4
form of iron in another way by treating ferrous sulfate described. Avariety of functional groups such as alkyne,
218
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2005, 347, 217–219

Chemoselective Reduction of Nitroarenes to Anilines
COMMUNICATIONS
ketone, enone, nitrile, lactone, and aromatic halide are References and Notes
well tolerated under these conditions.
[
1] G. W. Kabalka, R. S. Varma, Reduction of Nitro and Ni-
troso Compounds, in: Comprehensive Organic Synthesis,
st
1
edn., (Eds.: B. M. Trost, I. Fleming), Pergamon Press,
Experimental Section
All the compounds gave satisfactory spectral and analytical
data. Compounds 2a–d, 2f–i are cataloged chemicals. Com-
pounds 2e, 2j, and 2k are known in the literature.
Oxford, 1991, Vol. 8, p. 363.
[
2] a) C. Macleod, G. J. McKiernan, E. J. Guthrie, L. J. Farru-
gia, D. W. Hamprecht, J. Macritchie, R. C. Hartley, J. Org.
Chem. 2003, 68, 387–401; b) K. Ramadas, N. Srinivasan,
Synth. Commun. 1992, 22, 3189–3195; c) A. Perzyna, C.
Marty, M. Facompr e´ , J.-F. Goossens, N. Pommery, P. Col-
son, C. Houssier, R. Houssin, J.-P. H e´ nichart, C. Bailly, J.
Med. Chem. 2002, 45, 5809–5812; d) S. Wang, Z. Li, W.
Hua, Synth. Commun. 2002, 32, 3339–3345; e) T. Messeri,
G. Pentassuglia, R. D. Fabio, Tetrahedron Lett. 2001, 42,
3227–3230; f) D. G. Desai, S. S. Swami, S. K. Dabhade,
M. G. Ghagare, Synth. Commun. 2001, 31, 1249–1251;
g) D. G. Desai, S. S. Swami, S. B. Hapase, Synth. Commun.
1999, 29, 1033–1036; h) L. Wang, P. Li, Z. Wu, J. Yan, M.
Wang, Y. Ding, Synthesis 2003, 2001–2004; i) W. I. Awad,
S. S. M. Hassan, M. T. M. Zaki, Anal. Chem. 1972, 44, 9
[8]
General Procedure (Method A)
A 250-mL, 3-necked, round-bottomed flask was charged with
ethanol (80 mL). Iron powder (250 mmol, <10 mm) was added
in portions under efficient stirring, followed by concentrated
HCl (25 mmol). The suspension was stirred at 658C for 2 h
and was then cooled to 55–608C over a period of 10 min.
Then 25% aqueous ammonium chloride solution (40 mL)
was added. The nitroarene (50 mmol) was added in portions
(
exothermic) over a period of 30 min while maintaining the in-
ternal temperature at 65–808C. The reaction mixture was stir-
red at 55–658C for an additional 1–3 h and was cooled to
1
1–915.
408C. Ethanol (100 mL) and Celite (20 g) were added subse-
[
3] a) J. M. Matthews, M. N. Greco, L. R. Hecker, W. J. Hoek-
stra, P. Andrade-Gordon, L. de Garavilla, K. T. Demarest,
E. Ericson, J. W. Gunnet, W. Hageman, R. Look, J. B.
Moore, B. E. Maryanoff, Bioorg. & Med. Chem. Lett.
quently. The reaction mixture was filtered over a pad of Celite
(
(
15 g) with suction. The filter cake was washed with EtOH
100 mL), and the filtrate was concentrated under reduced
pressure to give a residue. To the residue, isopropyl acetate
120 mL) and saturated NaHCO (50 mL) were added. The bi-
2
003, 13, 753–756; b) Y. Kim, N.-H. Nam, Y.-J. You, B.-
(
3
Z. Ahn, Bioorg. & Med. Chem. Lett. 2002, 12, 719–722;
c) J. P. Edwards, L. Zhi, C. L. F. Pooley, C. M. Tegley,
S. J. West, M.-W. Wang, M. M. Gottardis, C. Pathirana,
W. T. Schrader, T. K. Jones, J. Med. Chem. 1998, 41,
phasic mixture was stirred at 20–258C, and the organic layer
was separated. The organic layer was washed with brine (1ꢀ
30 mL) and dried over sodium sulfate. The solvent was re-
moved under reduced pressure to afford the crude product.
The purity of the crude product was usually >97%.
2
779–2785; d) R. Neidlein, D. Christen, Helv. Chim.
Acta 1986, 69, 1623–1626.
[
[
4] J. G. Lee, K. I. Choi, H.Yeong Koh, Y. Kim, Y. Kang, Y.
Seo Cho, Synthesis 2001, 81–84.
5] B. W. Yoo, S. J. Lee, B. S. Yoo, K. I. Choi, J. H. Kim, Synth.
Commun. 2002, 32, 2489–2493.
General Procedure (Method B)
A 100-mL, 3-necked flask was charged with a nitroarene
(
(
10 mmol) and ethanol (50 mL). Iron sulfate heptahydrate
30 mmol), water (9 mL) and ammonium chloride (80 mmol)
[6] a) D. C. Gowda, B. Mahesh, Synth. Commun. 2000, 30,
639–3644; b) M. C. Kimber, A. C. Try, L. Painter,
M. M. Harding, P. Turner, J. Org. Chem. 2000, 65, 3042–
046.
7] Intermediates such as nitrosoarenes have been observed
3
were added subsequently with efficient stirring. Then zinc pow-
der (30 mmol, <10 mm) was added to the reaction mixture.
The reaction was exothermic and the internal temperature
rose to 30–358C in ~3 min. The reaction mixture was heated
to an internal temperature of 508C and stirred for an additional
3
[
[2i]
(
ref. ), but we did not observe these intermediates prob-
ably because they were more reactive than the corre-
sponding nitroarenes under our reaction conditions.
8] Compound 2e: F. K. Kurbanov, K. D. Allaberganov, A. B.
Kuchkarov, Izvestiya Vysshikh Uchebnykh Zavedenii,
Khimiya I Khimicheskaya Tekhnologiya 1976, 19, 653;
compound 2j: P. P. Joshi, T. R. Ingle, B. V. Bhide, J. Indian
Chem. Soc. 1960, 37, 479; compound 2k: Z. Lin, C. M. Teg-
ley, E. A. Kallel, K. B. Marschke, D. E. Mais, M. Gottar-
dis, T. K. Jones, J. Med. Chem. 1998, 41, 291.
1
–3 h. It was then cooled to room temperature and filtered
over a pad of Celite (5 g) with suction. The filter cake was wash-
ed with EtOH (70 mL), and the filtrate was concentrated under
reduced pressure to give a residue. To the residue, isopropyl
acetate (70 mL) and 25% aqueous ammonium chloride
[
(
30 mL) solution were added. The biphasic mixture was stirred
at room temperature for 5 min and the organic layer was sepa-
rated. The organic layer was washed with water (30 mL), satu-
rated NaHCO (30 mL), and brine (30 mL), and dried over so-
3
dium sulfate. The solvents were removed under reduced pres-
sure to afford the crude product. The purity of the crude prod-
uct was usually >97%.
Adv. Synth. Catal. 2005, 347, 217–219
asc.wiley-vch.de
ꢀ 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
219