From N-(dinitrophenyl) amino acids to benzimidazole N-oxides. Synthesis, kinetics and mechanism

Article abstract of DOI:10.1002/poc.1017

Several benzimidazole N-oxide derivatives were synthesized in very good yields by heating under reflux the corresponding N-(2,4- or 2,6-dinitrophenyl) amino acid derivative with NaOH in 60% 1,4-dioxane-H₂O. The N-oxides obtained from glycine and α- and β-alanine derivatives lost the carboxylic group. The observed rate constant for the reaction of N-(2,4-dinitrophenyl) glycine (2a) in 10% 1,4-dioxane-H₂O to give 5-nitro-1H-benzimidazole-3-oxide (4a) is first order on [NaOH]; the second-order rate constant is k_N=1.7 × 10^-3 M^-1 S^-1. The mechanism proposed includes the formation of an N-alkylidene 2-nitrosoaniline-type intermediate as the rate-determining step. Copyright

Full text of DOI:10.1002/poc.1017

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2006; 19: 187–195
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.1017
From N-(dinitrophenyl) amino acids to benzimidazole
N-oxides. Synthesis, kinetics and mechanism
Elba I. Buj a´ n* and Mar ı´ a Laura Salum
Instituto de Investigaciones en Fisicoqu ı´ mica de C o´ rdoba (INFIQC), Dpto. de Qu ı´ mica Org a´ nica, Facultad de Ciencias Qu ´ı micas,
Universidad Nacional de C o´ rdoba, Ciudad Universitaria, X5000HUA, C o´ rdoba, Argentina
Received 8 September 2005; revised 21 November 2005; accepted 30 November 2005
ABSTRACT: Several benzimidazole N-oxide derivatives were synthesized in very good yields by heating under reflux
the corresponding N-(2,4- or 2,6-dinitrophenyl) amino acid derivative with NaOH in 60% 1,4-dioxane–H O. The
2
N-oxides obtained from glycine and ꢀ- and ꢁ-alanine derivatives lost the carboxylic group. The observed rate constant
for the reaction of N-(2,4-dinitrophenyl) glycine (2a) in 10% 1,4-dioxane–H O to give 5-nitro-1H-benzimidazole-3-
2
ꢁ3
ꢁ1 ꢁ1
oxide (4a) is first order on [NaOH]; the second-order rate constant is k ¼ 1.7 ꢀ 10
M . The mechanism
s
N
proposed includes the formation of an N-alkylidene 2-nitrosoaniline-type intermediate as the rate-determining step.
Copyright # 2006 John Wiley & Sons, Ltd.
KEYWORDS: benzimidazole N-oxides; synthesis; kinetics; mechanism
INTRODUCTION
some cases, substitution of the amino group leading to
1
3,14
phenol competes with the cyclization reaction. The
yield of the N-oxide depends on the substituents on the
Compounds derived from benzimidazole and benzimida-
zole N-oxide exhibit a wide range of biological properties,
1
3,14
aromatic ring and the concentration of the base.
1
including insecticide and herbicide activity, antiviral,
To determine the scope of the method with regard to
the effect of an electron-withdrawing substituent on the
alkylic chain, we carried out a study on the reaction of
substrates 1a–1d, 2a and 2b and the results are reported
here.
2
antifungal and antibacterial activity, antitumoral and
3,4
antimicrobial activity, anti-ulcerative, antihypertensive
5
and antihistaminic activity, antihelmintic activity in
5
veterinarian medicine and also anti HIV-1 activity.
6
Benzimidazole N-oxides are not synthesized by direct
oxidation of benzimidazoles, thus it is important to
develop versatile synthethic methods to obtain a variety
of benzimidazole N-oxide derivatives. Thermal and
photochemical cyclizations of 2-nitroaniline derivatives
having electron-withdrawing substituents ꢀ to the amino
7–11
group are known to give benzimidazole N-oxides,
but the reaction failed to obtain 7-nitro-substituted
12
compounds.
We have reported previously on the synthesis of
7-substituted benzimidazole N-oxides by cyclization
of several N-alkyl-2-nitroaniline derivatives.
13,14
The
cyclization reaction requires two ortho substituents to
the amino group and the presence of an NH proton. In
*
Correspondence to: E. I. Buj a´ n, Dpto. de Qu ´ı mica Org a´ nica, Facultad
de Ciencias Qu ´ı micas, Universidad Nacional de C o´ rdoba, Ciudad
Universitaria, X5000HUA, C o´ rdoba, Argentina.
E-mail: elba@mail.fcq.unc.edu.ar
Contract/grant sponsor: Secretar ´ı a de Ciencia y Tecnolog ´ı a, UNC;
Contract/grant number: 123/04.
RESULTS AND DISCUSSION
Synthesis
Contract/grant sponsor: Agencia C o´ rdoba Ciencia; Contract/grant
number: 161/01.
Contract/grant sponsor: Agencia Nacional de Promoci o´ n Cient ´ı fica y
Tecnol o´ gica (FONCYT); Contract/grant number: PICT 2000 06-
When compounds 1a–1d, 2a and 2b were heated at
reflux in 60% (v/v) 1,4-dioxane–H O with 0.2 M NaOH,
they afforded the corresponding benzimidazole N-oxides
2
0
8036.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

´
E. I. BUJAN AND M. L. SALUM
1
88
a
Table 1. Synthesis of benzimidazole N-oxides 3 and 4
b
Yield (%)
b
Product Yield (%)
Amino acid
[NaOH] (M)
Time (min)
Product
pKa1
pKa2
1
1
1
1
2
2
2
2
2
a
b
c
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.01
20
15
10
20
10
60
120
300
2400
3a
3b
3b
3c
4a
4a
4b
70
90
79
83
71
84
88
2.20 ꢂ 0.03
2.41 ꢂ 0.04
6.22 ꢂ 0.01
6.83 ꢂ 0.04
c
5a
5a
5a
<5
10
7
d
a
2.40 ꢂ 0.03
2.23 ꢂ 0.05
7.18 ꢂ 0.02e
5.90 ꢂ 0.01
d
f
a
b
c
c
c
2.53 ꢂ 0.02
6.56 ꢂ 0.02
5b
5b
5b
13.5
94
c,g
6
a
2
In 60% 1,4-dioxane–H O at reflux.
b
c
d
e
f
Percentage yield of isolated product.
1
Quantified by H NMR.
Literature value of 2.2 (Ref. 9).
Literature value of 5.9 (Ref. 9).
ꢃ
Temperature ¼ 25 C.
g
Unreacted 2c was also found.
3
a–3c, 4a and 4b in very good yields. The results are
and deprotonation of NH in benzimidazole N-oxide
(pK ).
a2
14,15
summarized in Table 1. It is important to notice that 1d,
which has the carboxylic acid in ꢂ position with respect
to the amino group, gave the N-oxide derivative 3c, which
retained the carboxylic group, whereas compounds 1a–
The cyclization reaction leading to the formation of N-
oxide 3a or 4a was the only reaction observed with
glycine derivatives 1a and 2a (Scheme 1). Substitution
of the amino group to give 2,6-dinitrophenol (5a) or 2,4-
dinitrophenol (5b) competes with cyclization in the re-
action of compounds 1b–1d and 2b (Schemes 2 and 3).
Formation of the substitution product and cyclization
product was also reported for the reaction of N-n-butyl-
1
c, 2a and 2b, which have the carboxylic group in ꢀ or ꢁ
position, gave the cyclization product with elimination of
the COOH group.
1
3
2,4,6-trinitroaniline (6),
fluoromethylaniline (7a) and N-n-propyl-2,6-dinitro-4-
N-n-butyl-2,6-dinitro-4-tri-
1
4
trifluoromethylaniline (7b) in 60% 1,4-dioxane–H O
2
at reflux. On the other hand, N-(2,4-dinitrophenyl)
The two pK values of the benzimidazole N-oxides 3a–
a
3d, 4a and 4b were determined by spectrophotometric
titration and are summarized in Table 1; they are similar to
9
those reported before for 4a and other related N-oxi-
13,14
des.
These pK values correspond to deprotonation of
a
Scheme 1
þ
NH in conjugate acid of benzimidazole N-oxide (pK )
2 a1
Scheme 2
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

FROM N-(DINITROPHENYL) AMINO ACIDS TO BENZIMIDAZOLE N-OXIDES
189
Scheme 3
The formation of spiro complexes is a process ther-
modynamically and kinetically favored over that of
analogous complexes arising from intermolecular pro-
1
7,18
cesses.
ments where N displaces O and others where O displaces
There are also reports on Smiles rearrange-
19
N. The presence of the carboxylic group in substrates
b–1d enables the formation of the spiro Meisenheimer
1
complexes 8b–8d; the phenol could be obtained from
these complexes via C—N bond-breaking and hydrolysis
of the intermediate ester.
Scheme 4
ꢁ
-alanine (2c) gave only 2,4-dinitrophenol (5b) even with
.01 M NaOH (Scheme 4). In previously studied reac-
0
tions it was found that the relative yield of N-oxide/
phenol increases as the HO concentration decreases.
16
ꢁ
Effect of COOH group
The presence of a carboxylic group in the alkylic sub-
stituent on the amino nitrogen has an important effect on
the reaction. With the 2,4-dinitro derivatives, although N-
n-butyl-2,4-dinitroaniline (2e) gave only 2,4-dinitrophe-
13,16
nol (5b) (Scheme 4), 2a gave only N-oxide 4a
Scheme 1) and 2b gave N-oxide 4b as the main product
(
together with a minor amount of phenol 5b (Scheme 2).
In order to obtain the cyclization product, proton abstrac-
tion from the CH group is proposed in the generally
The five-membered complex 8b is expected to be more
stable than the others and to display a greater neighbor-
20
ing group effect. It is well known that the rate of
18
2
11,12
accepted mechanism for these reactions.
The pre-
sence of an electron-withdrawing substituent close to the
reaction center facilitates this process. As the distance of
the carboxylic group from the reaction center increases,
the stabilization of the negative charge diminishes and
only the substitution reaction takes place, as in 2c
cyclization increases when geminal dialkyl groups are
located in the alkyl chain, in-between the neighbouring
20
group and the reaction center. Thus the absence of an
electron-donating CH group in 1a may favor cyclization
to give N-oxide over substitution via spiro complex
3
(
Scheme 4).
On the other hand, although the N-alkyl-2,6-dinitro
formation.
derivatives, like 1e in 60% dioxane–H O, gave only the
2
N-oxides irrespective of the length of the alkylic
in the case of compounds 1b–1d the sub-
Kinetics of N-oxide formation
13,14,16
chain,
stitution reaction competes with N-oxide formation to
give phenol 5a (Schemes 2 and 3) in yields of 5–10%.
The fact that a substitution reaction was found with
compounds 1b–1d indicates that a parallel pathway for
the substitution of the amino group may operate when a
carboxylic group is present in the alkylic chain. One
possibility is the intramolecular reaction shown in 8b–8d.
We undertook a kinetic study of the reaction of 2a in 10%
ꢃ
dioxane–H O at 25 C with 0.01–1.00 M NaOH, which
2
leads to the formation of N-oxide 4a. The formation of 4a
was first order on [NaOH] with a second-order rate
ꢁ
3
ꢁ1 ꢁ1
constant k ¼ 1.57 ꢀ 10
M s (Table 2 and Fig. 1).
This rate constant can be compared with the value of
N
the observed rate constant for the reaction of 2e in 20%
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

´
E. I. BUJAN AND M. L. SALUM
1
90
Table 2. Rate constants for the reaction of N-(2,4-dinitro-
ꢃ
phenyl) glycine (2a) with NaOH at 25 C in 10%
a
1
,4-dioxane–H2O
4
ꢁ1
)
[NaOH] (M)
kobs ꢀ 10 (s
0
0
0
0
1
.204
.407
.610
.814
.001
2.95 ꢂ 0.02
5.88 ꢂ 0.03
8.47 ꢂ 0.05
12.9 ꢂ 0.02
16.3 ꢂ 0.2
a
ꢁ5
Ionic strength I ¼ 1 M (NaCl when needed); [2a]
0
¼ 6.93 ꢀ 10 M.
1
Figure 2. 200 MHz H NMR spectra of: (a) 9 recorded in
0% dioxane–d -D O at room temperature, (b) 15 min after
6
8
2
ꢁ2
addition of 0.2 M NaOD (c) at infinity; [9] ¼ 3.08 ꢀ 10
M
0
alkyl chain, as suggested above for the reaction of
compounds 1b–1d.
NMR studies
Figure 1. Plot of k
from 2a at 25 C in 10% 1,4-dioxane–H O. Ionic strength
vs. [NaOH] for the formation of 4a
obs
The reaction of N-(2,4,6-trinitrophenyl) glycine (9) was
1
followed by H NMR in 60% dioxane–d -D O at room
ꢃ
2
ꢁ
5
I ¼ 1 M (NaCl); [2a] ¼ 7.00 ꢀ 10
M
8
2
0
temperature with 0.2 M NaOD; only the formation of
,7-dinitro-1H-benzimidazole-3-oxide (10) was detected
Scheme 1). The NMR spectrum of 9 has a single peak in
5
(
1
,4-dioxane–H O, which gave only the substitution pro-
2
the aromatic region at ꢃ ¼ 9.16 (Fig. 2a). In the spectrum
taken 15 min after the addition of NaOD (Fig. 2b), three
signals attributed to N-oxide 10 by comparison with the
spectrum taken at infinite (Fig. 2c) appeared at 8.67 (d),
8.52 (d) and 7.94 (d), along with signals at ꢃ ¼ 9.00 (d),
8.97, 8.85 (d), 6.19 (d), 6.15 (d) and 5.26 which may be
attributed to the presence of scomplexes formed by
ꢁ7
ꢁ1 16
duct 5b, and for 1 M NaOH it is k ¼ 8 ꢀ 10 s ;
P
substrate 2a is about 2000 times more reactive than 2e.
These results are consistent with the proposed mechan-
ism for the cyclization reaction that requires deprotona-
11,12
tion of the CH ꢀ to the amino group, because it is
well known that a carboxylate group decreases the acidity
of an ꢀ carbon by about 25 pK units.
a
2
21
17
addition of nucleophiles to unsubstituted ring positions
22
and the dianion of 9. The addition of nucleophiles to
unsubstituted ring positions of trinitrobenzene derivatives
1
7,22,23
Kinetics of phenol formation
is a well-documented reaction.
this type are formed in the hydrolysis reaction of 1-
amino-2,4-dinitro
25
Intermediates of
24
We also performed a kinetic study of the reaction of 2c to
ꢃ
and 1-amino-2,4,6-trinitroben-
zenes with pyrrolidine, piperidine or morpholine as
the amino group.
give phenol 5b in 20% 1,4-dioxane–H O at 25 C with
2
1
M NaOH. The observed rate constant for the formation
ꢁ
6
ꢁ1
1
We followed the reaction of 2a by H NMR with NaOD
of 5b is k ¼ 2.2 ꢀ 10 s , which is 2.75 times higher
P
16
than that for the formation of the same phenol from 2e.
These results support the assumption that, along with the
expected aryl nucleophilic substitution pathway proposed
in 10% 1,4-dioxane–d -D O at room temperature. When
8
2
ꢁ2
the initial concentration of 2a was 4.8 ꢀ 10 M and
NaOD was 1 M the spectrum of the solution at the end
of the reaction indicates that an unknown product was
formed along with N-oxide 4a. This was confirmed by the
16
for the formation of 5b from 1e, a parallel reaction
pathway is available when there is a COOH group in the
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

FROM N-(DINITROPHENYL) AMINO ACIDS TO BENZIMIDAZOLE N-OXIDES
191
UV–visible spectrum. When the reaction was conducted
1
with 0.2 M NaOD the only product observed by H NMR
mechanism shown in Scheme 5 is the one that better
represents the complete set of data.
and UV–visible spectrophotometry was 4a, as was found
ꢁ
We suggest the formation of intermediate 12 in the
reaction pathway to benzimidazole N-oxide on the follow-
ing basis: the reduction of nitro to nitroso groups of
benzene derivatives in basic solution is a well-documented
5
in the kinetic study with [2a] ¼ 7.00 ꢀ 10 M. These
0
results may indicate that the formation of the unknown
product depends on substrate concentration as well as on
27
[
formation of azoxybenzene derivatives in the reaction of
NaOH]. Smith and co-workers reported previously the
reaction; the reaction of formaldehyde with N-methyl-
4-nitro-2-nitrosoaniline in 9 N sulfuric acid gave 1-
1
2,26
7
methyl-5-nitro-1H-benzimidazole-3-oxide; 2-aryl-5-ni-
some substituted o-nitroanilines in basic media.
further attempts to identify the product were made.
No
trobenzimidazole-3-oxides were formed by reaction of
4-nitro-2-nitrosoaniline and aromatic aldehydes in acid
28
media; reductive cyclization of N-benzylidene-2-ni-
troanilines prepared from o-nitroaniline and benzalde-
hydes is described as a synthethic route to substituted
Mechanism of N-oxide formation
5
Considering the results of the kinetic study of the reaction
of 2a in 10% 1,4-dioxane–H O, and the mechanism
benzimidazoles; and in the reactions of some ester
and nitrile derivatives of N-(o-nitrophenyl) glycine with
and without N-alkyl substituents, structures of the type of
12 were proposed as intermediates in benzimidazole
2
16
previously proposed, we suggest the mechanism shown
in Scheme 5 for the formation of N-oxides 4a and 4b.
The first step of the mechanism is the ionization of
substrate. The absence of the formation of N-oxide in the
reactions of N-butyl-N-methyl-2,6-dinitroaniline in 10%
12
N-oxide formation.
Assuming that the steps after cyclization are faster than
the others and considering 12 as a steady-state intermedi-
ate, kobs for the mechanism of Scheme 5 is given by
16
and 60% 1,4-dioxane–H O, N,N-di-n-propyl-2,6-dini-
2
14
tro-4-trifluoromethylaniline in 60% 1,4-dioxane–H O
2
ꢁ
and N,N-diethyl-2,4-dinitro-6-trifluoromethylaniline in
K k k ½HO ꢄ
1
2 3
11
kobs
¼
ð1Þ
basic ethanol is a good indication of the need to
generate a negative charge over the nitrogen of the amino
group in order to give the cyclization product. However, it
may be that for compounds 2a and 2b, which bear a
carboxylate group in the lateral chain, this is not true. A
reviewer suggested that an alternative mechanism could
involve ionization of the ꢀ-carbon of compounds 2a and
ꢁ
k ½HO ꢄ þ k
ꢁ
2
3
ꢁ
If k>> k [HO ], Eqn (1) is reduced to Eqn (2), which
3
ꢁ2
has the same mathematical form of the experimentally
observed expression for the reaction of 2a in 10% 1,4-
dioxane–H O. This suggests that the rate-determining
step for the reaction is the formation of intermediate 12
2
2b as the first step in the mechanism and then cyclization
through intramolecular attack of this carbanion on the
ꢁ
kobs ¼ K k ½HO ꢄ
ð2Þ
1
2
nitrogen of the o-nitro group. However, we think that the
Scheme 5
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

´
E. I. BUJAN AND M. L. SALUM
1
92
It was observed that compounds with a COOH group ꢀ
Nicolet 5SXC FT-IR spectrometer and UV–visible spec-
tra and kinetic measurements were recorded on a
Shimadzu UV 2101 spectrophotometer. Thin-layer chro-
matography (TLC) was carried out using silica gel
60 F254 (Macherey-Nagel) and spots were visualized by
UV. Silica gel 60 mesh size (0.063–0.200 nm, Merck)
was used for column chromatography.
or ꢁ to NH (1a–1c, 2a and 2b) lost the carboxylate to give
the N-oxides (3a, 3b, 4a and 4b) whereas that with
COOH ꢂ to NH (1d) retained the COOH (3c). This could
be explained by the mechanism proposed in Scheme 5;
loss of CO is posible in those compounds in which the
2
8
carboxy group is adjacent to an N-oxide system where
the negative charge generated could be stabilized by
resonance. In the case of compound 1c, which has the
COOH group ꢁ to NH, we suggest that CO elimination
2
Determination of pKa
could occur after the formation of compound 15, leading
to the formation of a stable anion.
The pK values of compounds 3a–3d, 4a and 4b were
a
determined by spectrophotometric titration in 10% (v/v)
ꢃ
1,4-dioxane–H O at 25 C. Solutions for the various pH
2
ranges were prepared using HCl (pH 0.7–3.0), acetic
acid–sodium hydroxide (pH 3.5–5.5) and Na HPO –
KH PO (pH 5.8–8.0).
2
4
2
4
Synthesis of N-dinitrophenyl amino acids
Compounds 1a–1d, 2a–2c and 9 were synthesized and
8
isolated by the method previously described.
Conclusions
N-(2,6-Dinitrophenyl) glycine (1a). Yellow solid, m.p.
1
The method described here allowed the synthesis of 7-
nitro-1H-benzimidazole N-oxide derivatives that were
not formed by previously described methods. Besides,
the presence of an electron-withdrawing group in the
alkylic chain ꢀ to the NH that facilitates proton abstrac-
ꢃ
12
ꢃ
1
72–174 C (lit. 173–175 C), yield 60%. H NMR
12
(CDCl þ (CD ) SO, 200 MHz) ꢃ: 3.65 (d, 2H,
3
3 2
J¼ 4.38 Hz, CH ), 6.76 (t, 1H, J ¼ 8.4 Hz, Ar), 8.14
2
1
3
(
(
1
(
d, 2H, J¼ 8.4 Hz, Ar) and 8.99.
C
NMR
CDCl þ (CD ) SO, 50 MHz) ꢃ: 46.81, 114.87, 132.14,
3
3 2
tion from the CH group enables cyclization in the
2
ꢁ
1
38.15, 139.10 and 170.50. IR (KBr, ꢄ
cm ): 1723.9
max
absence of a second substituent in the aromatic ring ortho
to the amino group that was not possible in previously
þ
þ
s). MS m/z: 241.8 ([M þ 1] , 2.2), 240.75 (M , 16.25),
13
196 (100), 168, 138, 105, 91 and 75. HRMS (EI): m/z calcd.
for C H N O : 241.0335; found: 241.0338.
studied N-alkyl derivatives. The results presented give
support to the mechanism proposed before for N-oxide
8
7 3 6
16
formation.
N-(2,6-Dinitrophenyl) ꢀ-alanine (1b). Yellow solid, m.p.
ꢃ
1
1
2
37–139 C, yield 70%. H NMR (CDCl þ (CD ) SO,
3
3
3 2
00 MHz) ꢃ: 1.46 (d, 3H, J ¼ 6.94 Hz, CH ), 4.02 (m, 1H,
EXPERIMENTAL
Materials
J ¼ 7.3 Hz, J ¼ 6.94 Hz, CH), 6.86 (t, 1H, J ¼ 7.68 Hz,
J ¼ 8.76 Hz, Ar), 8.18 (d, 2H, J ¼ 8.04 Hz, Ar) and 8.66
1
3
(d, 1H, J ¼ 7.68 Hz, NH). C NMR (CDCl þ (CD ) SO,
3
3 2
50 MHz) ꢃ: 18.70, 53.42, 115.65, 131.90, 138.37, 139.39
and 173.89. IR (KBr, ꢄ
29
ꢁ1
1
,4-Dioxane was purified as described previously.
cm ): 1734.1 (s), 1713.7 (s).
max
þ
Water purified in a Millipore Milli-Q apparatus was
used throughout. All of the inorganic reagents were of
analytical-reagent grade and were used without further
purification. The NaOH concentrations are expressed in
MS m/z: 255.0 (M , 14.18), 209.9 (100), 163.9, 134.0,
117.0 and 106.0. HRMS (EI): m/z calcd. for C H N O :
255.0491; found: 255.0496.
9
9 3 6
terms of total solvent volume (1,4-dioxane–H O). Melt-
2
N-(2,6-Dinitrophenyl) ꢁ-alanine (1c). Orange solid, m.p.
ꢃ
1
ing points were determined on an electrothermal appara-
137–139 C, yield 88%. H NMR (CDCl þ (CD ) SO,
2 2
3
3 2
13
tus. Proton (200 MHz) and C (50 MHz) NMR spectra
were recorded on a Bruker ACE 200 spectrometer.
Chemical shifts in CDCl þ (CD ) SO are reported as ꢃ
200 MHz) ꢃ: 2.65 (t, 2H, J ¼ 5.84 Hz, CH CH ), 3.26 (q,
2H, J ¼ 5.86 Hz, CH ), 6.78 (t, 2H, J ¼ 8.4 Hz, Ar), 8.17
2
1
3
3
3 2
(d, 2H, J ¼ 8.4 Hz, Ar) and 8.57 (s, 1H, NH). C NMR
values downfield from (CH ) Si and those in (CD ) SO
(CDCl þ (CD ) SO, 50 MHz) ꢃ: 34.01, 42.44, 114.35,
3
4
3 2
3
3 2
ꢁ
cm ):
max
1
are referenced to the solvent signal. The mass spectra (EI)
and HRMS (EI) were done at Unidad de Espectrometr ´ı a
de Masas, Universidad de Santiago de Compostela,
Spain. Infrared (IR) spectra in KBr were obtained on a
132.06, 138.02, 139.02 and 172.95. IR (KBr, ꢄ
þ
1734.1 (s). MS m/z: 254.9 (M , 57.33), 196.0, 191.9,
175.9 (100), 133.0 and 105.0. HRMS (EI): m/z calcd. for
C N N O : 255.0491; found: 255.0487.
9
9 3 6
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

FROM N-(DINITROPHENYL) AMINO ACIDS TO BENZIMIDAZOLE N-OXIDES
193
N-(2,6-Dinitrophenyl) aminobutyric acid (1d). Yellow
Synthesis of benzimidazole N-oxides;
general procedure
ꢃ
1
solid, m.p. 174–175 C, yield 76.6%. H NMR
(
6
CDCl þ (CD ) SO, 200 MHz) ꢃ: 1.99 (m, 2H, J ¼
3
3 2
.94 Hz, CH CH ), 2.37 (t, 2H, J ¼ 7.12 Hz,
The corresponding N-dinitrophenyl amino acid derivative
was heated at reflux in 60% (v/v) 1,4-dioxane–H O with
2
2
CH CH CH ), 3.05 (td 2H, J ¼ 4.74 Hz, J ¼ 2.2 Hz,
2
2
2
2
CH ), 6.80 (t, 2H, J¼ 8.04 Hz, Ar), 8.18 (d, 2H, J ¼ 8.04
NaOH at the concentration indicated in Table 1. The
reaction was followed by TLC; when no reactant was
observed, the heating was stopped; the solution was
cooled to room temperature and acidified to pH < 5
with 3.2 M HCl.
2
13
Hz, Ar) and 8.30 (s, 1H, NH).
(
C
NMR
CDCl þ (CD ) SO, 50 MHz) ꢃ: 24.71, 30.59, 45.46,
3
3 2
1
13.84, 131.63, 137.29, 139.31 and 173.62. IR (KBr,
ꢁ1
þ
ꢄmax cm ): 1708.6 (s). MS m/z: 271.2 ([M þ 1] , 3.22),
þ
2
(
70.1 (M , 27.39), 154.0, 87.1 (100) and 85.1. HRMS
CI): m/z calcd. for C N N O : 270.0726; found:
7-Nitro-1H-benzimidazole-3-oxide (3a). A solution pre-
pared by dissolving 105 mg (0.436 mmol) of 1a in 50 ml
of solvent was heated at reflux with NaOH (0.2 M) for
20 min. The reaction mixture was acidified to pH 4 and
the solvent was evaporated in a rotavapor. The salts were
separated from the product by column chromatography
on alumina (Aluminumoxid G type E, Merck) by eluting
with acetone and acetone–methanol (90:10). The N-oxide
was isolated as an orange powder that decomposed at
10 12 3 6
2
70.0724.
N-(2,4-Dinitrophenyl) glycine (2a). Yellow solid, m.p.
ꢃ
1
1
2
1
89–191 C, yield 83%. H NMR (CDCl þ (CD ) SO,
3
3 2
00 MHz) ꢃ: 4.24 (d, 2H, J ¼ 5.48 Hz, CH ) 6.98 (d,
2
H, J ¼ 9.5 Hz, Ar), 8.29 (dd, 1H, J ¼ 2.56, 6.94 Hz,
1
3
Ar) and 9.04 (d, 1H, J ¼ 2.56 Hz, Ar). C NMR
(
1
CDCl þ (CD ) SO, 50 MHz) ꢃ: 43.79, 114.14, 122.66,
3
3 2
ꢃ
29.12, 129.47, 135.03, 146.59 and 168.99. IR
198 C.
1
ꢁ
1
(
(
KBr, ꢄ
[M þ 1] , 4.5), 240.75 (M , 55.09), 195.9 (100),
cm ): 1718.8 (s). MS m/z: 241.75
þ
H NMR ((CD ) SO, 200 MHz) ꢃ: 7.14 (t, 1H, J¼
max
þ
3 2
7.8 Hz, Ar), 7.61 (d, 2H, J¼ 8.04 Hz, Ar), 7.89 (d, 1H,
13
168.0, 138.0, 105.0, 91.0 and 75.0. HRMS (EI): m/z
calcd. for C N N O : 241.0335; found: 241.0334.
J¼ 8.04 Hz, Ar) and 8.22 (s, 1H, Ar). C NMR ((CD ) SO,
3
2
50 MHz) ꢃ: 116.58, 117.52, 119.49, 132.10, 135.33, 137.89
and 142.53. HRMS (FAB) m/z: calcd. for C H N O :
178.0247; found: 178.0254.
8
7 3 6
7
4 3 3
N-(2,4-Dinitrophenyl) ꢀ-alanine (2b). Yellow solid, m.p.
ꢃ
1
1
2
1
78–180 C, yield 73%. H NMR (CDCl þ (CD ) SO,
3
3 2
00 MHz) ꢃ: 1.66 (d, 3H, J ¼ 6.94 Hz, CH ), 4.37 (m,
2-Methyl-7-nitro-1H-benzimidazole-3-oxide (3b) from
1b. A solution prepared by dissolving 199.9 mg
(0.784 mmol) of 1b in 100 ml of solvent was heated at
reflux with NaOH (0.2 M) for 15 min. The reaction
mixture was acidified to pH 3 and filtrated with an
3
H, CH), 6.87 (d, 1H, J ¼ 9.88 Hz, Ar), 8.28 (dd, 2H,
J ¼ 7.3; 2.2 Hz, Ar), 8.99 (d, 1H, NH) and 9.14
13
(
5
1
d, 1H, J ¼ 2.4 Hz, Ar). C NMR (CDCl þ (CD ) SO,
3
3 2
0 MHz) ꢃ: 18.24, 51.47, 114.22, 124.30, 130.36, 130.93,
36.40, 147.07 and 173.24. IR (KBr, ꢄ
ꢁ
1
cm ): 1708.6
ionic interchange resin (DIAION) with H O, MeOH
2
max
þ
þ
(
2
(
s). MS m/z: 256.9 ([M þ 1] , 0.41), 255.8 (M , 2.68),
and acetone. The products were treated with acetone,
in which the phenol is completely soluble and the
N-oxide precipitates. The N-oxide was isolated by
09.9 (100), 164.0, 149.0, 118.0, 91.0 and 75.0. HRMS
EI): m/z calcd. for C H N O : 255.0491; found:
9
9 3 6
2
55.0496.
vacuum filtration as a yellow solid that decomposed at
ꢃ
186 C.
1
N-(2,4-Dinitrophenyl) ꢁ-alanine (2c). Yellow solid, m.p.
1
2
H NMR ((CD ) SO, 200 MHz) ꢃ: 2.58 (s, 3H, CH ),
3 2 3
ꢃ
1
46–148 C, yield 81%. H NMR (CDCl þ (CD ) SO,
7.37 (t, 1H, J ¼ 8.04 Hz, Ar), 7.84 (d, 1H, J ¼ 8.04 Hz,
3
3 2
13
00 MHz) ꢃ: 2.74 (t, 2H, CH CH ), 3.74 (q, 2H,
2
Ar) and 7.98 (d, 1H, J ¼ 8.04 Hz, Ar). C NMR
2
J ¼ 6.22 Hz, CH ), 7.00 (d, 1H, J ¼ 7.04 Hz, Ar), 8.29
((CD ) SO, 50 MHz) ꢃ: 12.16, 115.58, 117.92, 120.75,
3 2
2
þ
(
dd, 2H, J ¼ 2.56, 7.3 Hz, Ar), 8.87 (s, 1H, NH) and 9.06
131.21, 135.17, 137.44 and 152.26. MS m/z: 192.95(M ,
43.16), 150.9 (100), 133.0 and 103.88. HRMS (EI): m/z
13
(
d, 1H, J ¼ 2.56 Hz, Ar). C NMR (CDCl þ (CD ) SO,
3
3 2
5
0 MHz) ꢃ: 33.01, 38.86, 113.73, 123.92, 130.09, 130.23,
35.67, 147.94 and 172.79. IR (KBr, ꢄ
calcd. for C H N O : 193.0487; found: 193.0493.
7 3 3
8
ꢁ
1
cm ): 1723.9
1
max
þ
þ
(
1
1
s). MS m/z: 256.9 ([M þ 2] , 0.41), 255.8 ([M þ 1] ,
2-Methyl-7-nitro-1H-benzimidazole-3-oxide (3b) from
1c. solution prepared by dissolving 204 mg
þ
2.68), 254.8 (M , 100), 195.9, 190.9, 175.9, 149.9,
19.95, 104.0 and 91.95. HRMS (EI): m/z calcd. for
A
(0.80 mmol) of 1c in 100 ml of solvent was heated at
reflux with NaOH (0.2 M) for 10 min. The reaction
mixture was acidified to pH 5 and filtrated with an ionic
C H N O : 255.0491; found: 255.0491.
9 9 3 6
1
N-(2,4,6-Trinitrophenyl) glycine (9). Orange solid.
H
interchange resin (DIAION) with H O, MeOH and acet-
2
NMR (CDCl þ (CD ) SO, 200 MHz) ꢃ: 3.63 (d, 2H,
one. The products were isolated by column chromato-
graphy on silica gel by eluting with variable amounts of
acetone–n-PrOH. The N-oxide was a yellow solid that
3
3 2
13
C
J ¼ 4.02 Hz, CH ), 8.9 (s, 2H, Ar) and 9.8 (s, 1H).
2
NMR (CDCl þ (CD ) SO, 50 MHz) ꢃ: 45.98, 125.86,
3
3 2
ꢃ
1
32.38, 135.38, 140.77 and 168.50.
decomposed at 184 C.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

´
E. I. BUJAN AND M. L. SALUM
1
94
1
H NMR ((CD ) SO, 200 MHz) ꢃ: 2.58 (s, 3H, CH ),
2
Reaction of N-(2,4-dinitrophenyl) ꢁ-alanine (2c). A solu-
3
3
7
.36 (t, 1H, J ¼ 8.04 Hz, Ar), 7.83 (d, 1H, J ¼ 8.04 Hz,
tion prepared by dissolving 103.2 mg (0.405 mmol) of 2c
in 50 ml of solvent was heated at reflux with NaOH
(0.2 M) for 6 h. The reaction mixture was acidified to
pH 2 and the solvent was evaporated in a rotavapor. The
salts were separated from the product by column chro-
matography on celite by eluting with acetone. After
evaporation a yellow solid was isolated and identified
13
Ar) and 7.98 (d, 1H, J ¼ 8.04 Hz, Ar). C NMR
(
(CD ) SO, 50 MHz) ꢃ: 12.14, 115.50, 118.00, 120.92,
3 2
þ
1
31.18, 135.07, 137.50 and 152.21. MS m/z: 192.9 (M ,
1.2), 177.0, 150.95 (100) and 103.05. HRMS (EI): m/z
3
calcd. for C H N O : 193.0487; found: 193.0491.
8
7 3 3
1
2
-Ethylcarboxylic-7-nitro-1H-benzimidazole-3-oxide (3c).
as 2,4-dinitrophenol (5b) by comparison of the H NMR
spectrum with that of an authentic sample.
A solution prepared by dissolving 200 mg (0.743 mmol)
of 1d in 100 ml of solvent was heated at reflux with
NaOH (0.2 M) for 20 min. The reaction mixture was
acidified to pH 4. The solvent was evaporated untill the
products precipitated. The N-oxide was a yellow solid
Because it is known that N-oxide yield increases at a
a solution prepared by
13,14,16
lower base concentration,
dissolving 101.1 mg (0.396 mmol) of 2c in 50 ml of
solvent was heated at reflux with 0.01 M NaOH for
40.5 h. After work-up, the solid obtained was identified
ꢃ
that decomposed at 196 C.
1
1
H NMR ((CD ) SO, 200 MHz) ꢃ: 2.85 (t, 2H,
3
and quantified by H NMR as a mixture of unreacted
substrate and phenol 5b.
2
J ¼ 7.32 Hz, CH ), 3.16 (t, 2H, J ¼ 7.32 Hz, CH CH ),
2
2
2
1
7
8
.41 (t, 1H, J ¼ 7.82 Hz, Ar), 7.94 (dd, 2H, J ¼
H NMR ((CD ) SO þ ((CH ) SO, 200 MHz) ꢃ: 2.74
3
2
3 2
13
.0, 17.1 Hz, Ar) and 12.22 (s, 1H, NH). C NMR
(t, 2H, CH ), 3.74 (q, 2H, J ¼ 6.22 Hz, CH ) 7.00 (d, 1H,
2
2
(
(CD ) SO, 50 MHz) ꢃ: 21.08, 30.19, 115.39, 118.22,
3
J ¼ 9.5 Hz, Ar), 8.29 (dd, 1H, J ¼ 2.56 Hz, Ar), 8.43
(dd, 1H, J ¼ 2.96 Hz, Ar), 8.86 (s, 1H, NH), 9.01 (d,
1H, J ¼ 2.96 Hz, Ar) and 9.12 (d, 1H, J ¼ 2.96 Hz, Ar).
2
1
21.40, 130.94, 135.04, 137.81, 154.17 and 173.12. MS
þ
þ
m/z: 251.9 ([M þ 1] , 4.88), 250.85 (M , 67.99) 234.8,
05.85, 189.9 (100), 176.9, 142.95, 116.0 and 101.0.
HRMS (EI): m/z calcd. for C H N O : 251.0542; found:
2
10 9 3 5
2
51.0536.
Kinetic procedures
5
pared by dissolving 50 mg (0.207 mmol) of 2a in 25 ml of
-Nitro-1H-benzimidazole-3-oxide (4a). A solution pre-
Reactions were initiated by adding the substrate dis-
solved in 1,4-dioxane to a solution containing all the
other constituents. The total 1,4-dioxane concentration
was 10% or 20% v/v; the reaction temperature was
solvent was heated at reflux with NaOH (0.2 M) for
1
0 min. Only the N-oxide was formed; it precipitated
upon acidification of the solution with HCl and was
separated by vacuum filtration as a yellow solid that
ꢃ
25 ꢂ 0.01 C. The ionic strength was kept constant at
1 M by adding NaCl as compensating electrolyte when
needed.
ꢃ
9
decomposed at 215 C (lit. 269 C).
ꢃ
1
H NMR ((CD ) SO, 200 MHz) ꢃ: 7.81 (d, 1H,
2
All kinetic runs were carried out under pseudo-first-
order conditions, with substrate concentrations of about
3
J ¼ 9.12 Hz, Ar), 8.07 (d, 1H, J ¼ 8.8 Hz, Ar), 8.34 (s,
þ
ꢁ5
1
1
H, Ar) and 8.71 (s, 1H, Ar). MS m/z: 178.9 (M , 100),
63.0, 149.0, 133.0, 110.0, 105.0, 89.0 and 78.0. HRMS
7.00–7.15 ꢀ 10 M. The reaction of 2a was followed by
measuring the decrease in absorbance of the reaction
mixture at 360 nm, which is the wavelength maximum of
the substrate. The spectra of the solutions at infinity were
compared with solutions of the corresponding N-oxide
under the same reaction conditions. The reaction of 2c
was followed by measuring the decrease in absorbance of
3 ml aliquots of the reaction solution taken at different
reaction times and poured over 1 ml of 3 M HCl at
358 nm, which is the wavelength maximum of the sub-
strate.
(
EI): m/z calcd. for C H N O : 179.0331; found:
7 4 3 3
1
79.0336.
-Methyl-5-nitro-1H-benzimidazole-3-oxide (4b). A solu-
tion prepared by dissolving 199.7 mg (0.783 mmol) of 2b
2
in 100 ml of solvent was heated at reflux with NaOH
(0.2 M) for 2 h. The N-oxide precipitated upon acidification
of the solution with HCl and was separated by vacuum
ꢃ
filtration. It was a yellow solid that decomposed at 199 C.
The filtrate was evaporated to dryness, and acetone was
added to precipitate the salts. The acetone was evaporated
giving a mixture of phenol 5b and N-oxide 4b.
Acknowledgments
1
H NMR ((CD ) SO, 200 MHz) ꢃ: 2.58 (s, 3H, CH )
3
2
3
7
6
.70 (d, 1H, J ¼ 9.12 Hz, Ar) 8.06 (dd, 1H, J ¼ 2.2,
This research was supported in part by Consejo Nacional
de Investigaciones Cient ´ı ficas y T e´ cnicas, Argentina
(CONICET), Agencia C o´ rdoba Ciencia, Agencia Nacio-
nal de Promoci o´ n Cient ´ı fica y Tecnol o´ gica (FONCYT)
and Secretar ´ı a de Ciencia y Tecnolog ´ı a, Universidad
Nacional de C o´ rdoba, Argentina. M.L.S. is a CONICET
doctoral fellow.
1
3
.6 Hz, Ar) and 8.27 (d, 1H, J ¼ 2.2 Hz, Ar). C NMR
(
CDCl , 50 MHz) ꢃ: 12.24, 104.98, 116.98, 118.95,
3
1
1
1
31.32, 142.26, 142.34 and 153.85. MS m/z:
þ
þ
93.85([M þ 1] , 12.26), 192.9 (M , 100), 177.0,
63.0, 147.0, 130.0, 104.0, 89.0 and 78.0. HRMS (EI):
m/z calcd. for C H N O : 193.0487; found: 193.0490.
8 7 3 3
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195

FROM N-(DINITROPHENYL) AMINO ACIDS TO BENZIMIDAZOLE N-OXIDES
195
1
1
6. Buj a´ n EI, Ca n˜ as AI, de Rossi RH. J. Chem. Soc., Perkin Trans. 2
REFERENCES
2
001; 1973–1977.
7. Terrier F. Nucleophilic Aromatic Displacement: The Influence of
the Nitro Group. VCH Publishers: New York, 1991; Chapt. 2.
1
. Kobayashi M, Uneyama K. J. Org. Chem. 1996; 61: 3902–
3
. Cundy DJ, Sympson GW. Aust. J. Chem. 1996; 49: 199–203.
. Nohara F, Nishii M, Ogawa K, Isono K, Ubukata M, Fujii T, Itaya
T, Saito T. Tetrahedron Lett. 1987; 28: 1287–1290.
905.
18. Bernasconi CF, Gandler JR. J. Org. Chem. 1977; 42: 3387–
3393.
2
3
19. Terrier F. Nucleophilic Aromatic Displacement: The Influence of
the Nitro Group. VCH Publishers: New York, 1991; 226.
20. Ruff F, Csizmadia IG. Organic Reactions. Equilibria, Kinetics
and Mechanism. Elsevier: Amsterdam, 1994; 428.
21. Ruff F, Csizmadia IG. Organic Reactions. Equilibria, Kinetics
and Mechanism. Elsevier: Amsterdam, 1994; 324.
22. Crampton MR, Gold V. J. Chem. Soc. (B) 1966; 893–900.
23. Crampton MR, Willison MJ. J. Chem. Soc., Chem Commun.
1973; 215–216; Gibson B, Crampton MR. J. Chem. Soc., Perkin
Trans. II 1979; 648–652; Crampton MR, Lord SD. J. Chem. Soc.,
Perkin Trans 2 1997; 369–376.
4
5
. Fujii T, Itaya T, Ogawa K. Heterocycles 1997; 44: 573–592.
. Yang D, Fokas D, Li J, Yu L, Baldino CM. Synthesis 2005;
4
7–56.
6
. Gardiner JM, Loyns CR, Schwalbe CH, Barrett GC, Lowe PR.
Tetrahedron 1995; 51: 4101–4110.
. Neadle DJ, Pollitt RJ. J. Chem. Soc. (C) 1969; 2127–2130.
. Goudie RS, Preston PN. J. Chem. Soc. (C) 1971; 1139–1142.
. Ljublinskaya LA, Stepanov VM. Tetrahedron Lett. 1971; 12:
7
8
9
4
511–4514.
1
0. Harvey IW, McFarlane MD, Moody DJ, Smith DM. J. Chem.
Soc., Perkin Trans I 1988; 681–689.
24. Buj a´ n de Vargas E, Remedi MV, de Rossi RH. J. Phys. Org.
Chem. 1995; 8: 113–120.
1
1
1. Szczecinski P, Bartusik D. J. Chem. Res. (S) 2002; 84–85.
2. Collins Cafiero PA, French CS, McFarlane MD, Mackie RK,
Smith DM. J. Chem. Soc. Perkin Trans. 1 1997; 1375–1384.
3. Buj a´ n de Vargas E, Ca n˜ as AI. Tetrahedron Lett. 1996; 37: 767–
25. Buj a´ n EI, Remedi MV, de Rossi RH. J. Chem. Soc., Perkin Trans. 2
2000; 969–975.
26. Collins PA, McFarlane MD, Mackie RK, Smith DM. Tetrahedron
1992; 48: 7887–7898.
1
7
70.
27. Paradisi C, Scorrano G. Acc. Chem. Res. 1999; 32: 958–968.
28. Russell DW. J. Med. Chem. 1967; 10: 984–985.
29. de Rossi RH, de Vargas EB. J. Am. Chem. Soc. 1981; 103: 1533–
1540.
1
1
4. Buj a´ n EI, Salum ML. Can. J. Chem. 2004; 82: 1322–1327.
5. Chua SO, Cook MJ, Katritzky AR. J. Chem. Soc.(B) 1971;
2
350–2355.
Copyright # 2006 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2006; 19: 187–195