Cobalt- and iron-catalyzed redox condensation of o-substituted nitrobenzenes with alkylamines: A step- and redox-economical synthesis of diazaheterocycles

Article abstract of DOI:10.1021/ol403064z

A wide variety of functionalized 2-aryl benzimidazoles can be prepared by a solvent-free cobalt- or iron-catalyzed redox condensation of 2-nitroanilines and benzylamines. The cascade including benzylamine oxidation, nitro reduction, condensation, and aromatization occurs without any added reducing or oxidizing agent. The method can be extended to other alkylamines as reducing components or 2-nitrobenzamides as oxidizing components when using an iron/sulfur catalyst to afford various diazaheterocycles.

Full text of DOI:10.1021/ol403064z

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Cobalt- and Iron-Catalyzed Redox
Condensation of o‑Substituted
Nitrobenzenes with Alkylamines:
A Step- and Redox-Economical
Synthesis of Diazaheterocycles
Thanh Binh Nguyen,* Julie Le Bescont, Ludmila Ermolenko, and Ali Al-Mourabit
Centre de Recherche de Gif-sur-Yvette, Institut de Chimie des Substances Naturelles,
CNRS, 91198 Gif-sur-Yvette Cedex, France
nguyen@icsn.cnrs-gif.fr
Received October 24, 2013
ABSTRACT
A wide variety of functionalized 2-aryl benzimidazoles can be prepared by a solvent-free cobalt- or iron-catalyzed redox condensation of
2-nitroanilines and benzylamines. The cascade including benzylamine oxidation, nitro reduction, condensation, and aromatization occurs without
any added reducing or oxidizing agent. The method can be extended to other alkylamines as reducing components or 2-nitrobenzamides as
oxidizing components when using an iron/sulfur catalyst to afford various diazaheterocycles.
Redox condensation of the nitro group with other
reducing components provides a direct, step-, atom-, and
redox-economical approach to nitrogen containing com-
pounds, including amides and aza-heterocycles. Most of
the reported methods for redox condensation of the nitro
group employ expensive metals and/or ligands1 with a
large excess of the reducing components or external oxidiz-
ing agents.2 In this context, we are interested in developing
such methods using low-cost and readily available simple
salts of the first-row transition metals in association with
ligands within reach. Recently, we reported ironꢀsulfur
catalyzed methods for redox condensation between
o-nitroanilines and 2- or 4-methylhetarenes3 or 2-phene-
thylamines4 as an efficient route to aza-heterocycles. In
continuing our study, we report here a general method for
iron- and cobalt-catalyzed redox condensation between
o-substituted nitrobenzenes and amines.
Our initial efforts were focused on identifying the cata-
lytic activity of the first row transition metal salts. For this
(4) Nguyen, T. B.; Retailleau, P.; Al-Mourabit, A. Org. Lett. 2013,
15, 5238.
(5) For some selected recent examples, see: (a) Li, J.; Benard, S.;
Neuville, L.; Zhu, J. Org. Lett. 2012, 14, 5980. (b) Bastug, G.; Eviolitte,
C.; Marko, I. E. Org. Lett. 2012, 14, 3502. (d) Pizzetti, M.; De Luca, E.;
Petricci, E.; Porcheddu, A.; Taddei, M. Adv. Synth. Catal. 2012, 354,
2453. (e) Cano, R.; Ramon, D. J.; Yus, M. J. Org. Chem. 2011, 76, 654.
(f) Matsushita, H.; Lee, S.; Joung, M.; Clapham, B.; Janda, K. D.
Tetrahedron Lett. 2004, 45, 313. (g) Brasche, G.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2008, 47, 1932. (h) Nguyen, T. B.; Ermolenko, L.; Al-
Mourabit, A. Heterocycles 2012, 86, 555. (i) Huang, J.; He, Y.; Wang,
Y.; Zhu, Q. Chem.;Eur. J. 2012, 18, 13964. (j) Peng, J.; Ye, M.; Zong,
C.; Hu, F.; Feng, L.; Wang, X.; Wang, Y.; Chen, C. J. Org. Chem. 2011,
76, 716. (k) Lewis, J. C.; Berman, A. M.; Bergman, R. G.; Ellman, J. A.
J. Am. Chem. Soc. 2008, 130, 2493. (l) Shen, M.; Driver, T. G. Org. Lett.
2008, 10, 3367. (m) Hubbard, J. W.; Piegols, A. M.; Soederberg, B. C. G.
Tetrahedron 2007, 63, 7077. (n) Nguyen, T. B.; Ermolenko, L.; Dean,
W. A.; Al-Mourabit, A. Org. Lett. 2012, 14, 5948. (o) Nguyen, T. B.;
Ermolenko, L.; Al-Mourabit, A. Green Chem. 2013, 15, 2713.
(1) Pd/XantPhos: (a) Xie, Y.; Liu, S.; Liu, Y.; Wen, Y.; Deng, G. Org.
Lett. 2012, 14, 1692. Ru(acac)3/dppe: (b) Liu, Y.; Chen, W.; Feng, C.;
Deng, G. J. Chem.;Asian J. 2011, 6, 1142. Au: (c) Tang, C. H.; He, L.;
Liu, Y. M.; Cao, Y.; He, H. Y.; Fan, K. N. Chem.;Eur. J. 2011, 17,
7172. dppf: (d) Wang, H.; Cao, X.; Xiao, F.; Liu, S.; Deng, G. Org. Lett.
2013, 15, 4900. (e) Wu, M.; Hu, X.; Liu, J.; Liao, Y.; Deng, G. Org. Lett.
2012, 14, 2722. Ir/Pd complex: (f) Zanardi, A.; Mata, J. A.; Peris, E.
Chem.;Eur. J. 2010, 16, 10502.
(2) (a) Xiao, F.; Liu, Y.; Tang, C.; Deng, G. Org. Lett. 2012, 14, 984.
(3) Nguyen, T. B.; Ermolenko, L.; Al-Mourabit, A. J. Am. Chem.
Soc. 2013, 135, 118.
r
10.1021/ol403064z
XXXX American Chemical Society

purpose, the redox condensation of o-nitroaniline with
benzylamine was chosen as a model reaction. This reac-
tion has been reported under uncatalyzed conditions
at high reaction temperatures (>210 °C) to furnish
2-phenylbenzimidazole5 in low yields.6
temperatures required to afford good yields. It should be
noted that the N-debenzylation was not observed for the
case of 1i.
All the screened reactions were carried out at 120 °C,
with 3 equiv of benzylamine under solvent-free conditions
and an argon atmosphere. As can be seen from the data
shown in Table 1, in the absence of catalysts, no trace of the
condensed product 3aa could be detected and both starting
materials were recovered unchanged. In the presence of a
catalyst (2 mol %), while the reaction catalyzed by man-
ganese- (entry 2), nickel- (entry 7), and copper- (entry 8)
halides proceeded sluggishly, cobalt- (entries 5ꢀ6) and
iron- (entries 3ꢀ4) halides were all competent to provide
cleanly the desired 2-phenylbenzimidazole 3aa product.
Table 2. Cobalt- or Iron-Catalyzed Redox Condensation of
o-Nitroaniline 1a with Benzylamine 2a
Table 1. Redox Condensation of o-Nitroaniline 1a with
Benzylamine 2aa
entry
catalyst
conversion (%)c
1
2
3
4
5
6
7
8
ꢀ
0
MnCl2 4H2O
20
85
92
85
95
5
3
FeCl2 4H2O
3
3
3
FeCl3 6H2O
CoCl2 6H2O
CoBr2 xH2Ob
3
NiCl2 6H2O
3
CuCl
15
a Reaction conditions: o-nitroaniline 1a (5 mmol), benzylamine 2
(15 mmol), metal salt (2 mol %, 0.1 mmol). b Catalyst CoBr2 xH2O
a Reaction conditions: 1a (5 mmol), 2a (15 mmol), metal salt (2 mol %,
3
(18.5% Co). c Catalyst FeCl3 6H2O.
0.1 mmol). b 18.5% Co. c Deteremined by 1H NMR.
3
We then investigated the scope of the redox condensa-
tion between o-nitroaniline 1a and various primary benzy-
lamines 2. The results are summarized in Table 2. All
the benzylamines examined gave the condensed products
in moderate-to-good yields (entries 1ꢀ6). No significant
difference in catalytic activity between Co and Fe was
found. In the case where benzylamine is o-substituted by a
chlorine atom, the yield is somewhat lower, possibly due to
the steric hindrance (entry 6).
Next, a series of o-nitroanilines were then subjected to
the selected optimized conditions with benzylamine 2a
(Table 3). o-Nitroanilines bearing methyl, chlorine, bro-
mine, and methoxy substituents at different positions are
all effective substrates although a lower yield was observed
for 1g (entry 6) wherein a methoxy group at the para
position to the nitro group could probably strongly
influence the redox property of the nitro group. The
reaction with N-substituted o-nitroanilines 1hꢀi (entries
7ꢀ8) proceeded also efficiently despite the higher reaction
A plausible mechanism for the present transformation
shown in Scheme 1 could involve the following steps: (a)
the initial oxidation of benzylamine 2a into benzaldimine
which is rapidly transformed into N-benzylbenzaldimine 4
via a transamination reaction.7 This step also yielded the
metal complex in reduced form that could participate in
the next reduction step, (b) the nitro group.
Although the reduction of a nitro group into an amino
group is a complex six-electron transfer cascade process,
under no circumstances could we observe the formation of
partially reduced intermediates but fully reduced product
5 is readily detected in trace quantities in the crude reac-
tion mixture. The process was achieved by step (c) which
(7) For selected examples of oxidation of benzylamines into
N-benzylbenzaldimines, see: (a) Hu, Z.; Kerton, F. M. Org. Biomol.
Chem. 2012, 10, 1618. (b) Huang, B.; Tian, H.; Lin, S.; Xie, M.; Yu, X.;
Xu, Q. Tetrahedron Lett. 2013, 54, 2861. (c) Aschwanden, L.; Mallat, T.;
Krumeich, F.; Baiker, A. J. Mol. Catal. A: Chem. 2009, 309, 57. (d) Zhu,
B.; Lazar, M.; Trewyna, B. G.; Angelici, R. J. J. Catal. 2008, 260. (e)
Grirrane, A.; Corma, A.; Garcia, H. J. Catal. 2009, 264, 138. (f) Guo, H.;
€
Kemell, M.; Al-Hunaiti, A.; Rautiainen, S.; Leskela, M.; Repo, T. Catal.
(6) (a) Nishioka, H.; Ohmori, Y.; Iba, Y.; Tsuda, E.; Harayama, T.
Heterocycles 2004, 64, 193. (b) Rao, C. V. C.; Reddy, K. K.; Rao,
N. V. S. Indian J. Chem. B 1980, 19, 655.
Commun. 2011, 12, 1260. (g) Wendlandt, A. E; Stahl, S. S. Org. Lett.
2012, 14, 2850. (h) Zhang, E.; Tian, H.; Xu, S.; Yu, X.; Xu, Q. Org. Lett.
2013, 15, 2704.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 3. Cobalt- or Iron-Catalyzed Redox Condensation of
o-Nitroaniline 1 with Benzylamine 2aa
Table 4. Fe/S-Catalyzed Redox Condensation of o-Nitroaniline
1a with Amines and o-Nitrobenzamides 1j,k with Benzylaminesa
a Reaction conditions: o-nitroaniline 1 (5 mmol), benzylamine 2a
(15 mmol), metal salt (2ꢀ5 mol %). b Catalyst CoBr2 xH2O (Co assay
3
18.5% Co) (2 mol %). c Catalyst FeCl3 6H2O (2 mol %). d Catalyst
FeCl3 6H2O (5 mol %).
3
3
includes formation of the final benzimidazole 3aa pro-
duct by a cascade condensation 4 þ 5 f 6, oxidationꢀ
aromatization 6 f 3aa.
Scheme 1. Proposed Mechanism
a Reaction conditions: 1a,j,k(5 mmol), amine 2(15 mmol), FeCl3 6H2O
(5 mol %, 0.25 mmol), S (20 mol %, 1 mmol, 32 mg) unless otherwise noted.
3
b 1a (2.5 mmol), 2g (5 mmol). c Conversion by 1H NMR.
benzylamines (Table 4). To solve this handicap and in
connection with our previous results obtained with the
Fe/S catalyst,3,4 we decided to study the applicability of the
present methodology using the iron/sulfur catalyst gener-
ated in situ from hexahydrated ferric chloride (5 mol %)
and elemental sulfur (20 mol %).8
It is noteworthy that both hydrated CoBr2 and FeCl3
failed to promote efficiently the redox condensation
of o-nitroaniline 1a with amines other than primary
Org. Lett., Vol. XX, No. XX, XXXX
C

To our delight, this catalyst displayed a high efficiency
in promoting the reaction not only with unsubstituted
benzylamine 2a (entry 1) but also with various amines
(entries 2ꢀ8). The condensed product 3aa could be formed
from different N-substituted benzylamines.
could be applied successfully to a wide range of benzyla-
mines with high functional group tolerance. As a general
trend, reaction with the N-substitued substrate 1k required
a higher temperature.
The differences in catalytic activity observed with the
two Co and Fe halides and Fe/S can be attributed to the
differences in their electrochemical properties. As clusters
with multiple iron and sulfur atoms in different oxidation
states, the Fe/S catalyst is expected to behave as an
electronically more flexible capacitor which is capable of
storing and releasing electrons for redox reactions with a
wide range of redox potentials.
In conclusion, we have demonstrated that the redox
condensation reaction between o-nitroanilines and benzy-
lamines can be achieved smoothly by using iron- and
cobalt-halide salts. The scope of the reaction can be
extended to other alkylamines as reducing reaction com-
ponents or 2-nitrobenzamides as oxidizing components
when using an iron/sulfur catalyst. Overall, our procedure
is highlighted by its simple, straightforward, step- and
redox-economical nature combined with the use of readily
available, inexpensive, and nontoxic catalysts. Our future
study will focus on the mechanistic understanding and on
the application of this redox condensation methodology
to the synthesis of biologically interesting molecules.
While the reaction with N-methylbenzylamine 2a0 pro-
ceeded smoothly, the condensation with more hindered
analogues 2a00 and 2a000 required higher temperatures
(entries 3, 5). Amines other than benzylamines such as
n-octylamine (2g, entry 6), di-n-butylamine (2h, entry 7),
triethylamine (2i, entry 8), and tri-n-propylamine (2j,
entry 9) could also condense with 1a to provide the
corresponding benzimidazoles 3agꢀ3aj. Although the for-
mation of quinoxaline product 3ag0 was the major reaction
pathway in the reaction of 2g (entry 6), this type of
condensation remained only a minor side reaction in the
cases of secondary and tertiary amines 2hꢀj. While the
explanation for this observation remained unknown, some
factors could strongly modify the product distribution,
including steric hindrance of amines 2, molar ratio of the
starting materials, and reaction temperature. An investiga-
tion of these parameters is underway in our laboratory.
As final trials to establish the scope of o-substituted
nitrobenzene using the Fe/S catalyst, we condensed
o-nitrobenzamide 1j,k with benzylamines to furnish the
corresponding quinazolin-4(3H)-ones.9 Representative
examples given in entries 12ꢀ16 showed that the reaction
Supporting Information Available. Experimental de-
tails and characterization data. This material is available
free of charge via the Internet at http://pubs.acs.org.
(8) Control experiments using only elemetal sulfur as a catalyst
(20 mol %) resulted in less than 5% of the condensed heterocyclic
products 3.
(9) Hydrated CoBr2 or FeCl3 or elemental sulfur failed to promote
this redox condensation.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX