The dilemma between acid and base catalysis in the synthesis of benzimidazole from: O -phenylenediamine and carbon dioxide

Article abstract of DOI:10.1039/c9cc06156h

The tandem synthesis of benzimidazole and other azoles can be achived by the N-formylation of ortho-substituted anilines followed by a cyclization reaction. However, CO₂-based N-formylations with hydrosilane reducing agents are base catalyzed whereas the cyclization reaction is acid catalyzed. The mismatch in catalytic conditions means that only one of the steps can be catalyzed in a single pot reaction. While the N-formylation reaction is frequently the target of catalyst development, the cyclization reaction requires comparably much harsher reaction conditions. Identification of these difficulties lead us to the development of a one-pot, two-step synthesis of benzimidazole under mild reaction conditions employing acid catalysts.

Full text of DOI:10.1039/c9cc06156h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
The dilemma between acid and base catalysis
in the synthesis of benzimidazole from
o-phenylenediamine and carbon dioxide†‡
Cite this: Chem. Commun., 2019,
55, 13089
Received 8th August 2019,
Accepted 8th October 2019
Martin Hulla,
Simon Nussbaum, Alexy R. Bonnin and Paul J. Dyson
*
DOI: 10.1039/c9cc06156h
rsc.li/chemcomm
The tandem synthesis of benzimidazole and other azoles can be
achived by the N-formylation of ortho-substituted anilines followed
by a cyclization reaction. However, CO2-based N-formylations with
hydrosilane reducing agents are base catalyzed whereas the cyclization
reaction is acid catalyzed. The mismatch in catalytic conditions means
that only one of the steps can be catalyzed in a single pot reaction.
While the N-formylation reaction is frequently the target of catalyst
development, the cyclization reaction requires comparably much
harsher reaction conditions. Identification of these difficulties lead us
to the development of a one-pot, two-step synthesis of benzimidazole
under mild reaction conditions employing acid catalysts.
Fig. 1 (a) N-formylation of amines with CO2 and hydrosilane reducing
agents. (b) Cascade synthesis of benzimidazole, oxazole and thiazole,
where X = NH, O or S respectively.
and cyclization reaction in order to identify why such harsh reaction
conditions are necessary. The synthesis of benzimidazole from
Carbon dioxide is an attractive C1 building block for chemical o-phenylenediamine and CO2 was selected as the model reaction
synthesis.1–4 However, its inherent thermodynamic stability5 and studied by in situ 1H NMR spectroscopy as well as synthetic
limits the reactions of CO2 to those involving high energy methods.
substrates,6,7 strong nucleophiles8–11 or equally strong reducing
The reaction was proposed to proceed by the N-formylation
agents.12–14 For all other applications catalysts are required with of o-phenylenediamine to N-(2-aminophenyl)formamide, followed
catalysts for reductive C–N bond forming reactions of amines by its cyclization to benzimidazole, accompanied by the elimination
1
with CO2 having advanced considerably.15–18 The N-formylation of water (Fig. 1b).25–28 Monitoring the reaction in situ by H NMR
reaction with hydrosilane reducing agents, in particular, has spectroscopy in DMSO-d6 showed that even at 25 1C and in the
reached the stage where it can be performed with organic19,20 or absence of a catalyst o-phenylenediamine rapidly N-formylates
simple salt catalysts21–24 under ambient conditions (Fig. 1a).
to N-(2-aminophenyl)formamide. However, the subsequent cycliza-
In addition, the N-formylation reaction with N-heterocyclic tion reaction is slow (Fig. 2). Addition of an N-formylation catalyst,
carbene (NHC) catalysts has been extended to the tandem synthesis such as tetra-n-butylammonium acetate ([TBA][OAc]),29 slightly
of benzimidazole and related cyclic compounds (Fig. 1b).25 accelerates the N-formylation reaction but has no significant
Subsequently, the synthesis of various azoles prepared using effect on the cyclization of N-(2-aminophenyl)formamide to
this methodology have been reported with a number of benzimidazole, its yield remained below 10% even after 13 h.
N-formylation catalysts including 1-butyl-3-methylimidazolium
Since, monitoring of the reaction by 1H NMR spectrscopy
acetate ([BMIm][OAc]),26 1,5-diazabicyclo[4.3.0]non-5-ene (DBN)27 in situ requires a slightly elevated CO2 pressure (5 bars), which
and B(C6F5)3.28 However, unlike the N-formylation reaction, promote the N-formylation reaction and may hinder the cycli-
temperatures of 60–150 1C and elevated pressures up to 50 bars zation, we performed a series of ex situ experiments with the
are required. In addition, excess hydrosilane is required in all reported N-formylation catalysts [TBA][OAc],29 Cs2CO3,22 1,5,7-
28
cases.25–28 Hence, we decided to study the tandem N-formylation triazabicyclo[4.4.0]dec-5-ene (TBD)30 and B(C6F5)3 under ambient
conditions (Table 1).
´
´ ´
Institute of Chemistry and Chemical Engineering, Ecole Polytechnique Federale de
Lausanne (EPFL), CH-1015 Lausanne, Switzerland. E-mail: pjd@epfl.ch
† Dedicated to Prof. Robin Perutz on the occasion of his 70th birthday.
‡ Electronic supplementary information (ESI) available: NMR spectra and synthetic
procedures. See DOI: 10.1039/c9cc06156h
The ex situ experiments performed under ambient conditions
1
confirm the trend from the in situ H NMR spectroscopic study
at 5 bar of CO2 and concur with earlier studies.22,29–31 The
N-formylation reaction proceeds catalyst free in DMSO31
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13089--13092 | 13089

View Article Online
Communication
ChemComm
Fig. 2 In situ 1H NMR spectroscopic monitoring of the N-formylation and
cyclization of o-phenylenediamine to N-(2-aminophenyl)formamide and
benzimidazole respectively. Reaction conditions: o-phenylenediamine
(0.25 mmol), phenylsilane (0.25 mmol), [TBA][OAc] catalyst (0.025 mmol),
CO2 (5 bar), DMSO-d6 (1.5 mL), 25 1C, 13 h.
Fig. 3 Sequential reduction of CO2 with hydrosilanes and possible
products resulting from reactions at elevated temperatures with one
equivalent of o-phenylenediamine.
Table 1 N-formylation of o-phenylenediamine to N-(2-aminophenyl)-
formamide and its cyclization to benzimidazole
formoxysilanes, silylacetals, methoxysilanes and eventually to
silylethers and methane in the presence of basic catalysts
(Fig. 3).33,34 While formoxysilanes are necessary intermediates
along the N-formylation pathway,29,35,36 silylacetals lead to the
undesirable formation of aminals and N-methylamines (Fig. 3).37–39
Furthermore, such sequential reductions decrease the quantity of
available hydrosilane potentially leading to the need for having it in
excess. These side reactions can be suppressed at elevated reaction
pressures,39 leading to the elevated reaction temperatures and
pressures frequently observed in the synthesis of benzimidazole.
Alternatively, the sequential reduction of CO2 appears to be
suppressed by the use of less reactive hydrosilanes such as
polymethylhydrosiloxane (PMHS).25 Using DMSO instead of a
base catalyst, which accelerates not only the N-formylation
reaction but the entire sequence of CO2 reductions, leads to
good yields and N-formylation selectivity even at elevated reaction
temperatures. In DMSO complete N-formylation of o-phenylene-
diamine can be achieved under ambient reaction conditions in
24 h or at 70 1C in 6 h with partial cyclization to benzimidazole
(Fig. 4). If only one equivalent of phenylsilane is used no
undesirable side reactions are observed. The cyclization reaction
should then be the priority for catalyst development, which
would enable an efficient tandem synthesis of benzimidazole
and other azols under mild reaction conditions.
Entry
Catalyst
Solvent
Yield 2 (%)
Yield 3 (%)
1
2
3
4
5
6
7
8
9
—
DMSO
DMSO
DMSO
DMSO
DMSO
THF
THF
THF
THF
THF
58
83a
83a
84a
19
0
84
0
10
0
4
12
5
10
1
0
13
0
[TBA][OAc]
Cs2CO3
TBD
B(C6F5)3
—
[TBA][OAc]
b
Cs2CO3
TBD
B(C6F5)3
0
0
10
Reaction conditions: o-phenylenediamine (0.25 mmol), phenylsilane
(0.25 mmol), catalysts (0.025 mmol), CO2 (1 bar), DMSO or THF (0.5 mL),
23 1C, 5 h. The reaction yields were determined by 1H NMR spectroscopy
using dibromomethane as an internal standard. An average of two runs
is reported. Additionally B5% of o-phenylenediformamide was
obtained. Not fully soluble i.e. o10 mol% of catalyst in solution.
a
b
(Table 1, entry 1) and is accelerated by basic catalysts (Table 1,
entries 2–4).22,29,30,32 However, the effect of the basic N-formylation
catalysts on the cyclization reaction is minimal and the increased
concentration of benzimidazole can be attributed to the increased
concentration of the N-(2-aminophenyl)formamide intermediate
rather than a true catalytic effect towards the cyclization reaction.
28
In order to test the cyclization reaction we prepared the N-(2-
aminophenyl)formamide intermediate and attempted to cyclize
Contrary to a previous hypothesis, the acid catalyst B(C6F5)3
appears to hinder the N-formylation reaction (Table 1, entry 5), at
least under the ambient conditions in DMSO employed here. In
addition, no reaction occurred in THF, where a base catalyst is
necessary (Table 1, entries 6, 7 and 9). Hence, it would appear,
at least in DMSO, that the synthesis of benzimidazole from
o-phenylenediamine and CO2 is limited by the cyclization
reaction and not by the N-formylation of the starting material,
which tends to be the target of catalysis.25–27
Fig. 4 Tandem N-formylation and cyclization of o-phenylenediamine.
Reaction conditions: o-phenylenediamine (0.25 mmol), phenylsilane
(0.25 mmol), CO2 (1 bar), DMSO (0.5 mL), 70 1C, 6 h. The reaction yields
were determined by 1H NMR spectroscopy using dibromomethane as an
The cyclization reaction is promoted at elevated reaction
temperatures.25 However, elevated reaction temperatures also
promote the sequential reduction of CO2 with hydrosilanes to internal standard.
13090 | Chem. Commun., 2019, 55, 13089--13092
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Communication
Table 2 Cyclization of N-(2-aminophenyl)formamide to benzimidazole
Fig. 5 Catalysts required for the individual steps of benzimidazole synthesis
by the tandem N-formylation-cyclization reaction of o-phenylenediamine
with CO2 and hydrosilanes.
Entry
Catalyst
Temp. (1C)
Yield (%)
1
2
3
4
5
6
7
8
—
70
70
70
70
70
70
70
70
70
23
23
23
23
23
23
23
23
3
3
6
[TBA][OAc]
HCOOH
HCOOHa
AlCl3
Al(OH)3
The cyclization of N-(2-aminophenyl)formamide is acid
catalyzed, whereas its synthesis from o-phenylenediamine,
CO2 and a hydrosilane is base catalyzed (Fig. 5).29,32,35 The
mismatch in base and acid catalysts results in the frequently
observed harsh reaction conditions for the tandem synthesis of
azoles. If base is used to promote the N-formylation, elevated
reaction temperatures are required for the subsequent cyclization
to proceed and elevated CO2 pressures must also be applied to
suppress undesirable side reactions. Acid catalysts accelerate the
cyclization, but partly inhibit the N-formylation reaction (Table 1,
entries 1 and 5), which then requires equally harsh reaction
conditions to proceed.
71a
29
5
b
b
BBr3
100
32
78
1
3
2
3
1
31
1
8
B(OH)3
B(C6F5)3
—
[TBA][OAc]
HCOOH
AlCl3
Al(OH)3
BBr3
B(OH)3
B(C6F5)3
9
10
11
12
13
14
15
16
17
b
b
In DMSO the N-formylation reaction proceeds under catalyst-
free conditions31 and the tandem reaction is limited only by the
cyclization of the N-formylated intermediate. The tandem synthesis
of benzimidazole can then be achieved by a two-step procedure in a
single pot, where an acid catalyst is added after the completion of
the N-formylation reaction (Fig. 6a). Optimization of the reaction
conditions in DMSO at 70 1C, 1 bar of CO2 and with phenylsilane as
the reducing agent allows full N-formylation of o-phenylenediamine
to be achieved in 6 h without the need of excess reducing agent. The
cyclization of the N-(2-aminophenyl)formamide intermediate to
benzimidazole is then completed in the presence of 10 mol% of
BBr3 in less than 2 h. Under these conditions no significant
amounts of side products were observed. In comparison, a single
step reaction under acidic conditions with 10 mol% of BBr3 resulted
in almost complete inhibition of the N-formylation reaction and the
recovery of 98% of o-phenylenediamine starting material and
benzimidazole yield of only 2%.
Reaction conditions: N-(2-aminophenyl)formamide (0.25 mmol),
catalyst (10 mol%), DMSO (0.5 mL), 23–70 1C, 5 h. The reaction yields
were determined by 1H NMR spectroscopy using dibromomethane as
an internal standard. An average of two runs is reported. a 200 mol% of
b
HCOOH. Not fully soluble i.e. o10 mol% of catalyst in solution.
it (Table 2). In the absence of a catalyst even at 70 1C the
cyclization reaction is slow with benzimidazole obtained in
only 3% yield (Table 2, entry 1). Addition of an N-formylation
catalyst, e.g. [TBA][OAc], does not promote the reaction and
yield of benzimidazole remains unchanged (Table 2, entry 2),
which confirms the limited effect of base catalysts on the
cyclization step of the tandem reaction. Notably, the observed
benzimidazole yields of 3% are lower than observed under
comparable reaction temperatures starting from o-phenylene-
diamine, CO2 and phenylsilane (Fig. 4). Nevertheless, under the
N-formylation reaction conditions including CO2 and phenyl-
silane only the second N-formylation to o-phenylenediformamide
took place further supporting the preference of N-formylation over
the cyclization reaction.
Considering that cyclization reactions are frequently acid
catalyzed and that formic acid forms as the N-formylation
reaction side product from the reaction of formoxysilane with
water and/or silanol,29 we decided to test a few acids for their
catalytic activity in the synthesis of benzimidazole from N-(2-
aminophenyl)formamide (Table 2). The presence of formic acid
indeed slightly accelerates the cyclization reaction (Table 2,
entries 1 and 3). However, large quantities are necessary to have a
significant effect on the reaction (Table 2, entry 4). Nevertheless,
these results explain the higher benzimidazole yield observed
from the N-formylation of o-phenylenediame in comparison to
the cyclization of the N-(2-aminophenyl)formamide intermediate.
Lewis acids such as AlCl3, B(C6F5)3 or BBr3 were found to be more
efficient catalysts with benzimidazole yields of 29, 78 and 100%,
respectively (Table 2, entries 5, 7 and 9). Nevertheless, at 23 1C
only BBr3 and B(C6F5)3 demonstrated significant catalytic activity
(Table 2, entries 15 and 17).
In order to demonstrate that the reaction can be extended to
the synthesis of other azoles and cyclic products, we have
synthesized benzoxazole (82% yield), benzothiazole (73% yield)
and dihydroquinazoline (85% yield) from 2-aminophenol,
2-aminothiophenol and 2-aminobenzylamine respectively using
the same one-pot, two-step reaction setup and reaction
Fig. 6 (a) Two step synthesis of bezimidazole from o-phenylenediamine,
CO2 and phenylsilane with BBr3 acid catalyst (10 mol%) added after 6 h
reaction. The reaction yield was determined by 1H NMR spectroscopy
using dibromomethane as an internal standard. An average of two runs is
reported. (b) Synthesis of benzoxazole, benzothiazole and dihydroquinazoline
under the same reaction conditions.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13089--13092 | 13091

View Article Online
Communication
ChemComm
8 G. W. Ebert, W. L. Juda, R. H. Kosakowski, B. Ma, L. Dong, K. E.
Cummings, M. V. B. Phelps, A. E. Mostafa and L. Jianyuan, J. Org.
Chem., 2005, 70, 4314–4317.
9 J. Wu, X. Yang, Z. He, X. Mao, T. A. Hatton and T. F. Jamison, Angew.
Chem., 2014, 126, 8556–8560.
conditions (Fig. 6b). While not optimized for these substrates,
the high yields indicate that the process is adaptable to other
tandem syntheses of cyclic products via amine N-formylation
followed by acid catalyzed cyclization.
´
10 A. Correa and R. Martın, Angew. Chem., Int. Ed., 2009, 48, 6201–6204.
The first N-formylation step in the synthesis of benzimida-
zole from o-phenylenediamine, CO2 and hydrosilane, is base
catalyzed. However, base catalysts appear to have no effect on the
subsequent cyclization step, which uncatalyzed requires elevated
reaction temperatures. Elevation of the reaction temperatures
leads to the formation of undesirable side products, which can be
suppressed by elevated reaction pressures, the use of less reactive
hydrosilanes or by the absence of basic N-formylation catalysts
(all of which inhibit sequential CO2 reduction reactions). Never-
theless, this leads to the harsh reaction conditions frequently
observed in the tandem synthesis of azoles from ortho-substituted
anilines. The second cyclization step of the N-(2-aminophenyl)-
formamide intermediate is acid catalyzed. However, acids such as
B(C6F5)3 and BBr3 at least partly inhibit the N-formylation reaction,
which results in the need for equally harsh reaction conditions. A
two-step reaction appears to be the optimal solution, where the
N-formylation is performed catalyst free in DMSO followed by
acidification of the reaction mixture, which promotes the cyclization
of the N-formylated intermediate. Such a two-step process leads to
excellent yields under mild reaction conditions.
11 G. Zadel and E. Breitmaier, Angew. Chem., Int. Ed. Engl., 1992, 31,
1035–1036.
12 R. F. Nystrom, W. H. Yanko and W. Brown, J. Am. Chem. Soc., 1948,
70, 441.
13 T. Wartik and R. K. Pearson, J. Am. Chem. Soc., 1955, 77, 1075.
14 I. Knopf and C. C. Cummins, Organometallics, 2015, 34, 1601–1603.
15 J. R. Cabrero-Antonino, A. R. Ortiz and M. Beller, Angew. Chem., Int. Ed.,
2018, 12820–12838.
16 X.-F. Liu, X.-Y. Li, C. Qiao and L.-N. He, Synlett, 2018, 548–555.
17 A. Tlili, E. Blondiaux, X. Frogneux and T. Cantat, Green Chem., 2015,
17, 157–168.
18 Y. Li, X. Cui, K. Dong, K. Junge and M. Beller, ACS Catal., 2017, 7,
1077–1086.
19 G. Li, J. Chen, D.-Y. Zhu, Y. Chen and J.-B. Xia, Adv. Synth. Catal.,
2018, 360, 2364–2369.
20 C. C. Chong and R. Kinjo, Angew. Chem., Int. Ed., 2015, 54, 12116–12120.
21 M. Hulla, F. D. Bobbink, S. Das and P. J. Dyson, ChemCatChem,
2016, 8, 3338–3342.
22 C. Fang, C. Lu, M. Liu, Y. Zhu, Y. Fu and B.-L. Lin, ACS Catal., 2016,
6, 7876–7881.
23 X.-F. Liu, X.-Y. Li, C. Qiao, H.-C. Fu and L.-N. He, Angew. Chem., Int. Ed.,
2017, 56, 7425–7429.
24 C. Xie, J. Song, H. Wu, B. Zhou, C. Wu and B. Han, ACS Sustainable
Chem. Eng., 2017, 5, 7086–7092.
25 O. Jacquet, D. N. C. Gomes, M. Ephritikhine and T. Cantat, Chem-
CatChem, 2013, 5, 117–120.
26 X. Gao, B. Yu, Z. Yang, Y. Zhao, H. Zhang, L. Hao, B. Han and Z. Liu,
ACS Catal., 2015, 5, 6648–6652.
27 X. Gao, B. Yu, Y. Zhao, L. Hao and Z. Liu, RSC Adv., 2014, 4, 56957–56960.
28 Z. Zhang, Q. Sun, C. Xia and W. Sun, Org. Lett., 2016, 18, 6316–6319.
29 M. Hulla, G. Laurenczy and P. J. Dyson, ACS Catal., 2018, 8,
10619–10630.
We believe that a similar optimization is possible for all
tandem syntheses of azole type compounds from ortho-substituted
anilines.
´
30 C. D. N. Gomes, O. Jacquet, C. Villiers, P. Thuery, M. Ephritikhine
Conflicts of interest
and T. Cantat, Angew. Chem., Int. Ed., 2012, 51, 187–190.
31 H. Lv, Q. Xing, C. Yue, Z. Lei and F. Li, Chem. Commun., 2016, 52,
6545–6548.
32 M. Hulla, D. Ortiz, S. Katsyuba, D. Vasyliev and P. J. Dyson, Chem. –
Eur. J., 2019, 25, 11074–11079.
There are no conflicts to declare.
33 S. N. Riduan, Y. Zhang and J. Y. Ying, Angew. Chem., Int. Ed., 2009,
48, 3322–3325.
Notes and references
´
1 J. Artz, T. E. Mu¨ller, K. Thenert, J. Kleinekorte, R. Meys, A. Sternberg, 34 F. J. Fernandez-Alvarez and L. A. Oro, ChemCatChem, 2018, 10,
A. Bardow and W. Leitner, Chem. Rev., 2018, 118, 434–504.
2 S. Wang and C. Xi, Chem. Soc. Rev., 2019, 48, 382–404.
4783–4796.
35 Q. Zhou and Y. Li, J. Am. Chem. Soc., 2015, 137, 10182–10189.
3 J. Klankermayer, S. Wesselbaum, K. Beydoun and W. Leitner, Angew. 36 X.-F. Liu, R. Ma, C. Qiao, H. Cao and L.-N. He, Chem. – Eur. J., 2016,
Chem., Int. Ed., 2016, 55, 7296–7343. 22, 16489–16493.
4 Q.-W. Song, Z.-H. Zhou and L.-N. He, Green Chem., 2017, 19, 3707–3728. 37 R. L. Nicholls, J. A. McManus, C. M. Rayner, J. A. Morales-Serna,
5 J. Pacansky, U. Wahlgren and P. S. Bagus, J. Chem. Phys., 1975, 62,
2740–2744.
6 R. R. Shaikh, S. Pornpraprom and V. D’Elia, ACS Catal., 2018, 8,
419–450.
A. J. P. White and B. N. Nguyen, ACS Catal., 2018, 8, 3678–3687.
38 X.-F. Liu, C. Qiao, X.-Y. Li and L.-N. He, Green Chem., 2017, 19,
1726–1731.
39 M.-Y. Wang, N. Wang, X.-F. Liu, C. Qiao and L.-N. He, Green Chem.,
2018, 20, 1564–1570.
7 F. D. Bobbink and P. J. Dyson, J. Catal., 2016, 343, 52–61.
13092 | Chem. Commun., 2019, 55, 13089--13092
This journal is ©The Royal Society of Chemistry 2019