Investigation of electrochemically induced Michael addition reactions. Oxidation of some dihydroxybenzene derivatives in the presence of azide ion

Article abstract of DOI:10.1016/j.tet.2009.04.015

In aqueous solution containing azide ion as a nucleophile, electrochemical oxidation of hydroquinone and some dihydroxybenzoic acids have been studied using cyclic voltammetry and controlled-potential coulometry. The voltammetric data show that electroche

Full text of DOI:10.1016/j.tet.2009.04.015

Tetrahedron 65 (2009) 4742–4750
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Investigation of electrochemically induced Michael addition reactions. Oxidation
of some dihydroxybenzene derivatives in the presence of azide ion
*
Davood Nematollahi , Hosain Khoshsafar
Faculty of Chemistry, Bu-Ali-Sina University, P.O. Box 65174, Hamadan, Iran
a r t i c l e i n f o
a b s t r a c t
Article history:
In aqueous solution containing azide ion as a nucleophile, electrochemical oxidation of hydroquinone
and some dihydroxybenzoic acids have been studied using cyclic voltammetry and controlled-potential
coulometry. The voltammetric data show that electrochemically generated para and ortho-benzoqui-
nones participate in Michael addition reactions with azide ions to form the corresponding diazido or
diaminobenzoquinones. In this work, we have proposed various mechanisms for the electrode process
and we report an efficient and one-pot method for the synthesis of 2,5-diazido-1,4-benzoquinone,
2,5-diamino-1,4-benzoquinone, 4,5-diamino-1,2-benzoquinone, and 2,3-diamino-5,6-dioxocyclohexa-
1,3-dienecarboxylic acid based on the Michael reaction of electrochemically generated ortho and para-
benzoquinones with azide ion in an undivided cell using an environmentally friendly reagent-less
method in ambient conditions. An estimation of the observed homogeneous rate constant (kobs) of the
reaction of electrochemically generated para-benzoquinone with azide ion by the digital simulation
method is also presented.
Received 22 January 2009
Received in revised form 18 March 2009
Accepted 2 April 2009
Available online 10 April 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
nucleophiles,25–32 herein we wish to describe a one-pot and
straightforward protocol for the synthesis of some diazido and
The conjugate addition reaction of azides to
a
,
b
-unsaturated
diaminobenzoquinones (ortho and para) via electrochemical oxi-
dation of hydroquinone and some dihydroxybenzoic acids in the
presence of azide ions. The reaction proceeds in a single step with an
environmentally friendly reagent-less method in aqueous solution
with high atom economy in ambient conditions in an undivided cell
using a graphite electrode. Furthermore, the observed homoge-
neous rate constant (kobs) of the intramolecular reaction of electro-
chemically generated para-benzoquinones with azide ions has been
estimated by the digital simulation of the cyclic voltammograms.
carbonyl compounds is a useful method in synthetic organic
chemistry, providing efficient access to a variety of natural products
or compounds with pronounced biological activity.1,2 Also, the
chemistry of azides has been the subject of many investigations
because of their importance in preparative heterocyclic chemistry.3
In this context, azidoquinones constitute a synthetically versatile
and readily available class of compounds. Depending upon their
substitution pattern and the reaction conditions, a variety of very
specific and high yielding transformations can be accomplished.4–14
On the other hand, amino derivatives of quinones are important
building blocks in the synthesis of a variety of natural products,
medicinal compounds15–21 with antitumor and antimalarial activi-
ties.22,23 The importance of these compounds has motivated us to
study the electrochemical oxidation of catechol in the presence of
azide ion and to synthesize 4,5-diamino-o-benzoquinone.24 In
a continuation of our efforts to develop the synthesis of azido and
amino derivatives of quinones and to study the kinetics of the re-
action of quinones with azide ions, and following our experience in
electrochemical synthesis of organic compounds based on the oxi-
dation of catechols and hydroquinones in the presence of
2. Results and discussion
2.1. Electrochemical oxidation of hydroquinone
Figure 1, curve a, shows the voltammetric curve obtained for the
oxidation of hydroquinone (1) (1 mM) in water containing phos-
phate buffer (c¼0.2 M, pH¼6.0) at a glassy carbon electrode. In the
studied potential range, a well-defined voltammetric curve is
obtained that has an anodic (A1) and the corresponding cathodic
(C1) peaks. These peaks correspond to the oxidation of hydroqui-
none (1) to p-benzoquinone (1ox) and vice versa within a reversible
two-electron process.26,29,31 The oxidation of hydroquinone (1) in
the presence of azide ion as a nucleophile was studied in some
detail. Figure 1, curve b shows the cyclic voltammogram obtained
* Corresponding author. Tel.: þ98 811 8282807; fax: þ98 811 8257407.
E-mail address: nemat@basu.ac.ir (D. Nematollahi).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.04.015

D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
4743
A1
A1
6
4
37 µA
a
b
2
b
c
a
C1
0
-2
A2
A3
C1
d
-4
-0.25
C3
C2
-0.05
0.15
E/V vs. SCE
0.35
0.55
e
f
Figure 1. Cyclic voltammograms of 1 mM hydroquinone (1): (a) in the absence (b) in
the presence of 30 mM of azide ion in water containing phosphate buffer (c¼0.2 M,
pH¼6.0) at glassy carbon electrode. Scan rate: 10 mV sꢀ1; t¼25ꢁ1 ꢂC.
for a 1 mM solution of 1 in the presence of 30 mM of azide ion.
Under these conditions, the voltammogram exhibits anodic and
cathodic peaks A1 and C1, respectively. The comparison of A1 and C1
peaks in the absence and presence of azide ion shows an increase in
peak A1 and a decrease in peak C1. This indicates the reactivity of
electrochemically generated p-benzoquinone toward azide ion. The
existence of a subsequent chemical reaction is supported by the
following evidence: (a) increasing of peak A1 current, which could
be related to increasing the apparent number of electrons (napp). (b)
Decreasing of peak C1 current during the reverse scan, which could
indicate that p-benzoquinone 1ox formed at the surface of elec-
trode is removed by chemical reaction with azide ion. In this case,
the presence of the cathodic peak C1 strongly depends on the po-
tential sweep rate. Thus, for the highest sweep rate employed
a well-defined cathodic peak C1 is observed. For lower sweep rates,
the peak current ratio (IpC1/IpA1) is less than one. It is about 0.7 for
a sweep rate of 10 mV sꢀ1, and it increases when the sweep rate
increases. A similar situation is observed when the azide ion to
hydroquinone (1) concentration ratio is decreased.
Controlled-potential coulometry was performed in aqueous
solution containing 0.2 mmol of 1 and 1.5 mmol of azide ion in an
undivided cell at 0.2 V (vs SCE). Monitoring of the progress of the
electrolysis was carried out by cyclic voltammetry (Fig. 2). As
shown, during coulometry, in parallel with the decrease in height of
the anodic peak A1 and its cathodic counterpart (C1), new anodic
(A2 and A3), and cathodic counterpart peaks (C2 and C3) appear and
the height of them increases. At the end of the coulometry all an-
odic and cathodic peaks disappear and only anodic and cathodic
peaks A3 and C3 remain. These peaks are related to the redox re-
action of 3/3ox couple. The anodic peak A1 disappears when the
charge consumption becomes about 6eꢀ per molecule of 1. These
observations allow us to propose the pathway (an ECECE mecha-
nism) in Scheme 1 for the electrochemical oxidation of 1 in the
presence of azide ion. According to our results, it seems that the
Michael addition reaction of azide ion to p-benzoquinone 1ox is
C1
C2
C3
-0.6
-0.3
0
E/V vs. SCE
0.3
0.6
Figure 2. Cyclic voltammograms of 0.2 mmol hydroquinone (1) in the presence of
1.5 mmol of azide ion in water containing phosphate buffer (c¼0.2 M, pH¼6.0) at
glassy carbon electrode during controlled-potential coulometry at 0.2 V versus SCE.
After consumption of: (a) 0, (b) 15, (c) 30, (d) 60, (e) 100, and (f) 120 C.
OH
OH
O
-2e- -2H+
-
+N3
+H+
N3
OH
OH
O
1ox
1
2
OH
O
OH
N3
-2e- -2H+
N3
-
+N3
+H+
N3
N3
OH
O
OH
2ox
2
3
OH
O
N
N3
3-2e- -2H+
N3
N3
OH
O
3
3ox
Scheme 1.

4744
D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
Table 1
faster than other side reactions and leads to azidohydroquinone 2.
The oxidation of compound 2 is easier than the oxidation of hy-
droquinone 1 by virtue of the presence of an electron-donating
group. In the next step, p-benzoquinone 2ox, via a Michael reaction,
is converted into diazidohydroquinone 3. Further oxidation con-
verts diazidohydroquinone 3 into the final product 3ox.
The calculated kobs based on an EC0 mechanism for reaction of p-benzoquinone 1ox
with azide ion
pH
1.50
1.65ꢁ0.18
1.75
1.50ꢁ0.29
2.00
1.36ꢁ0.28
2.25
1.21ꢁ0.25
2.50
1.10ꢁ0.25
kobs (Mꢀ1 sꢀ1
)
a
a
For n¼3 at scan rates; 10, 15, and 25 mV sꢀ1
.
Accordingly, the anodic peak A1 pertains to the oxidation of hy-
droquinone 1 to the p-benzoquinone 1ox. Obviously, the cathodic
peak C1 corresponds to the reduction of the p-benzoquinone 1ox.
The new anodic peaks A2 and A3 can be related to electrochemical
oxidation of azidohydroquinone 2 and diazidohydroquinone 3 tothe
p-benzoquinones 2ox and 3ox, respectively. Clearly, the cathodic
peaks C2 and C3 correspond to the reduction of the p-benzoquinones
2ox and 3ox, respectively. The presence of electron-donating
group(s) on p-benzoquinones 2ox and 3ox reduces the reactivity of
these p-benzoquinones toward Michael reactions, so the rate of
reactions of these p-benzoquinones with azide ions are low and the
effects of these reactions are not observed in obtained cyclic vol-
tammograms because of the time scale of the cyclic voltammograms
and the times of these chemical reactions are not comparable.
However, in controlled-potential coulometry with a longer time
scale, the effect of these reactions is observed as an increase in the
number of electrons per molecule (n¼6) and appearance of new
anodic and cathodic peaks (A2, A3, C2, and C3) in cyclic voltammo-
grams obtained during coulometry (Fig. 2).
Couladouros and co-workers in the reaction of hydrazoic acid with
naphthoquinone.1 In this direction, in acidic media, the simulation
was performed based on an EC0 (catalytic) mechanism. The calcu-
lated values of the observed homogeneous rate constant (kobs) for
reaction of p-benzoquinone 1ox with an azide ion have shown in
Table 1.
Figure 3 shows the plot of observed homogeneous rate constant
(kobs) as a function of pH. As shown, the kobs decreases linearly with
increasing pH and reaches zero at pH¼4.5. Here is a competition
between catalytic and addition (Michael) reactions. When azide
ions are in protonated form (hydrazoic acid, pKa¼4.7234) the main
reaction is catalytic reaction (Scheme 2), while in less acidic, neu-
tral, and basic solutions, the preferable reaction is Michael addition
reaction (Scheme 1). So, for estimation of the observed homoge-
neous rate constant (kobs) of the reaction of electrochemically
generated p-benzoquinone 1ox with azide ion, we simulated cyclic
voltammograms of hydroquinone (1) (1 mM) in the presence of
azide ions (60 mM) at pH¼6.0 based on an ECE mechanism (Fig. 4).
The calculated value of the kobs is 0.21ꢁ0.02 Mꢀ1 sꢀ1
.
2.2. Kinetic evaluation
The electrochemical oxidation of hydroquinone (1) in the
presence of azide ions was tested by digital simulation. The simu-
lation was carried out assuming semi-infinite one-dimensional
diffusion and planar electrode geometry. The experimental pa-
rameters entered for digital simulation consisted of the following:
Estart, Eswitch, Eend, t¼25 ꢂC and analytical concentration of hydro-
2.3. Electrochemical oxidation of 2,5-dihydroxybenzoic acid
The electrochemical oxidation of 1 mM solution of 2,5-dihy-
droxybenzoic acid (4) in water containing phosphate buffer
(c¼0.2 M, pH¼6.0) at a bare glassy carbon electrode has been
studied using cyclic voltammetry (Fig. 5, curve a). The voltammo-
gram shows one anodic (A1) and corresponding cathodic peak (C1),
at 0.22 V and 0.16 V versus a standard calomel electrode (SCE),
respectively, which corresponds to the transformation of 2,5-
dihydroxybenzoic acid (4) to p-benzoquinone 4ox and vice versa
within a reversible two-electron process (Scheme 3).26 Figure 5
quinone (1) and azide ion. The transfer coefficient (a) was assumed
to be 0.5 and the formal potential was obtained experimentally as
the midpoint potential between the anodic and cathodic peaks
(Emid). The heterogeneous rate constant (0.002 cmꢀ1) for oxidation
of hydroquinone (1) was estimated by use of an experimental
working curve.33 All these parameters were kept constant
throughout the fitting of the digitally simulated voltammogram to
the experimental data. The parameter kobs was allowed to change
through the fitting processes. In order to estimation of the observed
homogeneous rate constant (kobs) of the reaction of electrochemi-
cally generated p-benzoquinone 1ox with azide ion, we studied the
electrochemical oxidation of hydroquinone (1) in the presence of
azide ion at various pHs. The results indicate an increase in anodic
peak current A1 against a decrease in current of cathodic peak C1,
with decreasing pH. As previously reported these observations
confirm the homogeneous catalytic pathway for the electro-
chemical oxidation of 1 in the presence of azide ion in acidic media
(Scheme 2).24 This type of reaction, also previously reported by
1.6
kobs = -0.556pH + 2.476
R = 0.9987
1.2
0.8
0.4
0
OH
O
-2e- -2H+
OH
O
1ox
1
O
OH
2HN3
+ 3N2
1.5
2.5
3.5
4.5
O
OH
pH
1ox
1
Scheme 2.
Figure 3. Linear relations between kobs and pH.

D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
4745
10
8
OH
OH
O
A1
COOH
COOH
-
COOH
-2e- -2H+
+N3
+H+
N3
OH
OH
O
5
4ox
4
6
4
2
0
OH
O
a
COOH
COOH
-N2
Intramolecular
Oxidation-Reduction
N3
H2N
OH
O
5
6
b
O
O
NH2
+N2 +CO2
COOH
-
+H+
+N3
Schmidt Rearrangement
H2N
H2N
O
O
7
6
-2
-4
-6
Scheme 3.
Besides, in electrochemical oxidation of 2,5-dihydroxybenzoic
acid (4) in the presence of azide ion, proportional to the increasing
of the potential sweep rate and in parallel with the decrease in
current of C2, the current of C1 increases (Fig. 6). Under these
conditions, the peak current ratios (IpC1/IpA1) and (IpC2/IpC1) versus
scan rate for a mixture of 2,5-dihydroxybenzoic acid (4) and azide
ion shows the reactivity of 4ox toward azide ion, appearing as an
increase in the peak current ratio IpC1/IpA1 and decrease in the peak
current ratio IpC2/IpC1 at higher scan rates. A comparable condition
is observed when the azide ion to 4ox concentration ratio is de-
creased. Also, under these conditions, the current function for peak
A1 (IpA1/v1/2) decreases with increasing scan rate. According to the
obtained electrochemical data and the spectroscopic data of the
final product, we propose the mechanism shown in Scheme 3 for
the electrochemical oxidation of 4 in the presence of azide ions.
According to our results, the Michael addition reaction of azide
ion to p-benzoquinone 4ox leads to azidohydroquinone 5. In the
C1
-0.3
0
0.3
E/V vs. SCE
0.6
Figure 4. Experimental (curve a) and simulated (curve b) cyclic voltammograms of
1 mM hydroquinone (1) in the presence of 60 mM azide ion at glassy carbon electrode,
in phosphate buffer solution (c¼0.2 M, pH¼6.0). Scan rate: 25 mV sꢀ1; t¼25ꢁ1 ꢂC.
(curves b and c) shows cyclic voltammogram obtained for a 1 mM
solution of 4 in the presence of 15 mM azide ion. The voltammo-
gram exhibits two cathodic peaks C1 (0.16 V vs SCE) and C2 (0.05 V
vs SCE). In the second cycle, a new anodic peak (A2) appears with an
Ep value of 0.10 V versus SCE. The comparison of Figure 5 (curve b)
with Figure 1, curve b shows the higher reactivity of Michael ac-
ceptor 4ox, because of the presence of carboxylic acid group as an
electron-withdrawn group in structure of 4ox.
A1
A1
b
9
9
c
5
1
a
5
A2
1
-3
-7
-3
C2
C1
C1
-7
-0.15
-0.15
0.1
E/V vs. SCE
0.35
0.1
E/V vs. SCE
0.35
Figure 5. Cyclic voltammograms of: (a) 1 mM 2,5-dihydroxybenzoic acid (4), (b) and (c) first and second cycles of 1 mM 2,5-dihydroxybenzoic acid (4) in the presence of 15 mM
azide ion, respectively, at glassy carbon electrode, in aqueous solution containing 0.2 M phosphate buffer (pH 6.0). Scan rate: 25 mV sꢀ1; t¼25ꢁ1 ꢂC.

4746
D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
50
40
30
20
10
0
H
O
O
O
O
OH
OH
N
N
N
O
A1
OH
-
+N3
+H+
OH
N
H2N
H2N
O
H2N
N
N
O
O
O
6
6a
6b
-N2
O
O
H
N
NH2
-CO2
g
a
O H
H2N
H2N
O
O
6c
7
Scheme 5.
The formation of diamino-p-benzoquinone 7 via Schmidt rear-
rangement2,35,36 is showed in Scheme 5.
The difference in behavior of electrochemical oxidation of hy-
droquinone (1) and 2,5-dihydroxybenzoic acid (4) in the presence
of azide ion can be attributed to (1) the enol–keto equilibrium in
the case of 5 (in comparison with 2) arising from intramolecular
hydrogen bonding (Scheme 6), (2) the easy loss of dinitrogen from
the 5c due to the more electron-withdrawing character of para-
quinone ring because of the presence of the carboxyl group with
electro-withdrawing character. In other words, it seems that elec-
tro-withdrawing group, i.e., –COOH in benzene ring makes –N3
group more easier to mediate intramolecular oxidation–reduction
to form –NH2, compared to hydroquinone where final product is
diazidohydroquinone rather than diaminohydroquinone.
-10
-20
-30
C2
C1
-0.2
0
0.2
E/V vs. SCE
0.4
For estimation of the observed homogeneous rate constant
(kobs) of the reaction of electrochemically generated p-benzoqui-
none 4ox with azide ion, we simulated cyclic voltammograms of
2,5-dihydroxybenzoic acid (4) (1 mM) in the presence of azide ions
(15 mM) at pH¼6.0, based on an ECE mechanism (Fig. 7). The cal-
culated value of the kobs is 1.93ꢁ0.05 Mꢀ1 sꢀ1. The comparison of
this number with the observed homogeneous rate constant (kobs) of
hydroquinone (1) (0.21ꢁ0.02), shows the presence of electron-
withdrawing carboxylic group on the hydroquinone ring causes an
Figure 6. Typical cyclic voltammograms of 1 mM 2,5-dihydroxybenzoic acid (4) in the
presence of 15 mM azide ion at a glassy carbon electrode, in aqueous solution con-
taining 0.2 M phosphate buffer (pH 6.0). Scan rates from (a) to (g) are: 10, 25, 50, 100,
200, 300, and 400 mV sꢀ1, respectively. t¼25ꢁ1 ꢂC.
next step, azidohydroquinone 5, via an overall intramolecular oxi-
dation–reduction reaction,1 converts into amino-p-benzoquinone
6. Next addition of azide ion and Schmidt rearrangement35 converts
6 into diamino-p-benzoquinone 7 as the final product. Conse-
quently, the anodic peak A1 pertains to the oxidation of 4 to the p-
benzoquinone 4ox. Clearly, peak C1 corresponds to the reduction of
4ox. The new anodic peak A2 can be related to oxidation of 5 to p-
benzoquinone 5ox and the cathodic peak C2 is counterpart of it.
Since the rate of intramolecular oxidation–reduction reaction and
Schmidt reaction are slow, the effects of these reactions are not
observed in the time scale of the cyclic voltammograms.
Formation of amino-p-benzoquinone 6 under intramolecular
oxidation–reduction reaction conditions can be rationalized as
shown in Scheme 4. According to this mechanism, 5c is converted
into 5d by loss of N2. Compound 5d is then readily transformed into
the stable tautomeric form 6.
increase in kobs
.
2.4. Electrochemical oxidation of 3,4-dihydroxybenzoic acid
The cyclic voltammogram of a 1 mM solution of 3,4-dihydroxy-
benzoic acid (8) in aqueous solution containing 0.2 M phosphate
buffer (pH 6.0) is shown in Figure 8 (curve a). The voltammogram
shows one anodic (A1) and a corresponding cathodic peak (C1),
which corresponds to the transformation of 3,4-dihydroxybenzoic
acid (8) to o-benzoquinone 8ox and vice versa within a quasi-re-
versible two-electron process.26,28 Figure 8 (curve b) shows the
cyclic voltammogram obtained for a 1 mM solution of 8 in the
presence of 15 mM azide ion in same condition. The voltammo-
gram exhibits one anodic peak A1 and three cathodic peaks (C1, C2,
and C3). In the second cycle, the voltammogram shows that, parallel
to the decrease in current of A1 and the shift of its potential in
a positive direction, two new anodic peaks (A2 and A3) appear at
less positive potentials (Fig. 8 (curve c)).
O
OH
O
H
H
H
H
COOH
H
H
N
COOH
H
COOH
H
H
N
H
N
N
N
N
N
N
N
N
O
OH
O
5a
5b
5
+H+
H
H
O
O
O
O
O
O
O
H
H
H
H
COOH
H
H
HN
COOH
H
COOH
H
OH
-H+
-N2
O
H
H
H
N3
H2N
H
N3
NH
N
OH
5
O
OH
5
O
O
6
5d
Scheme 4.
5c
Scheme 6.

D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
4747
A1
A1
9
8
6
Scan rate
7
20 mV/s
Scan rate
15 mV/s
a
b
a
b
5
3
4
2
1
0
-1
-2
-3
-5
C2
C1
C2
C1
-4
-0.2
0
0.2
E/V vs. SCE
0.4
-0.2
0
0.2
E/V vs. SCE
0.4
Figure 7. Experimental (curves a) and simulated (curves b) cyclic voltammograms of 1 mM 2,5-dihydroxybenzoic acid (4) in the presence of 15 mM azide ion at glassy carbon
electrode, in phosphate buffer solution (c¼0.2 M, pH¼6.0). Scan rate: 15 and 20 mV sꢀ1; t¼25ꢁ1 ꢂC.
The positive shift of the peak A1 in the presence of azide ion is
due to the formation of a thin film of product at the surface of the
electrode, inhibiting to a certain extent, the performance of the
electrode process. Also, it is seen that proportional to the increasing
of the potential sweep rate, parallel to the increase in height of the
C1 the height of C2 and C3 decreases (Fig. 9). A similar situation is
observed when the azide ion to 8 concentration ratio is decreased.
According to these data and the spectroscopic data of the final
product, which is as same as obtained product in electrochemical
oxidation of catechol in the presence of azide ions,24 we propose
the mechanism shown in Scheme 7 for the electrochemical oxi-
dation of 8 in the presence of azide ions.
According to proposed mechanism, the anodic peak A1 pertains
to the oxidation of 8 to the o-benzoquinone 8ox and cathodic peak
C1 is its counterpart. The new anodic peaks A2 and A3 can be related
to electrochemical oxidation of 9 (to 9ox) and 10red, respectively.
Intermediates 9ox and 10red are anodic and cathodic counterparts
of 9 and 10, respectively. Clearly, the cathodic peaks C2 and C3
A1
A1
9
6
3
14
11
b
a
8
c
5
0
-3
-6
2
-1
-4
A2
A3
C1
C3
C2
C1
-0.4
-0.1
0.2
E/V vs. SCE
0.5
-0.4
-0.1
0.2
E/V vs. SCE
0.5
Figure 8. Cyclic voltammograms of: (a) 1 mM 3,4-dihydroxybenzoic acid (8), (b) and (c) first and second cycles of 1 mM 3,4-dihydroxybenzoic acid (8) in the presence of 15 mM
azide ion, respectively, at glassy carbon electrode, in aqueous solution containing 0.2 M phosphate buffer (pH 6.0). Scan rate: 25 mV sꢀ1; t¼25ꢁ1 ꢂC.

4748
D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
A1
A1
23
28
18
8
18
13
8
c
b
a
b
a
3
-2
C3
C2
-2
-7
C4
C3
C1
C2
C1
-12
-0.4
0.1
E/V vs.SCE
0.6
-0.4
-0.1
0.2
E/V vs. SCE
0.5
Figure 10. Cyclic voltammograms of 1 mM 2,3-dihydroxybenzoic acid (12): (a) in the
absence of azide ion, (b) in the presence of 15 mM of azide ion, at a glassy carbon
electrode, in aqueous solution containing 0.2 M phosphate buffer (pH 6.0). Scan rate:
25 mV sꢀ1; t¼25ꢁ1 ꢂC.
Figure 9. Typical cyclic voltammograms of 1 mM 3,4-dihydroxybenzoic acid (8) in the
presence of 15 mM azide ion at a glassy carbon electrode, in aqueous solution con-
taining 0.2 M phosphate buffer (pH 6.0). Scan rates from (a) to (c) are: 25, 100, and
200 mV sꢀ1, respectively. t¼25ꢁ1 ꢂC.
into the related o-benzoquinone (5,6-dioxocyclohexa-1,3-dien-
ecarboxylic acid, 12ox) and vice versa within a quasi-reversible
two-electron process.28 Figure 10 (curve b) shows the cyclic vol-
tammogram obtained for a 1 mM solution of 12 in the presence of
15 mM azide ion at 25 mV sꢀ1. The voltammogram exhibits three
cathodic peaks C2 (0.09 V vs SCE), C3 (ꢀ0.12 V vs SCE) and C4
(ꢀ0.28 V vs SCE). For more data, the influence of the potential
sweep rate on the shape of cyclic voltammograms of a solution of
12 in the presence of azide ion has been studied (Fig.11). The results
show that proportional to the increasing of the potential sweep
rate, in the beginning, the cathodic peaks C4 and C3 disappears
(curves a and b) and then the cathodic peak C1 appears and in
parallel with the decrease in current of C2, the height of it increases
(Fig. 11 curves c–f). A comparable condition is observed when the
azide ion to 12 concentration ratio is decreased. These voltam-
metric data in addition to spectroscopic data (IR, 1H NMR, 13C NMR,
and MS) of the final product cause us to propose the following
mechanism for electrochemical oxidation of 2,3-dihydroxybenzoic
acid (12) in the presence of azide ions (Scheme 8).
correspond to the reduction of the o-benzoquinones 9ox and 10,
respectively. Because, the rate of Schmidt reaction is slow, the effect
of it is not observed in the time scale of the cyclic voltammograms.
The comparison of Figure 8 (curve b) with Figure 5, curve
b shows the higher reactivity of Michael acceptor 8ox, due to the
more reactivity of ortho-quinone ring in comparison with para-
quinone ring.
2.5. Electrochemical oxidation of 2,3-dihydroxybenzoic acid
Cyclic voltammograms of 2,3-dihydroxybenzoic acid (12) in the
absence and in the presence of azide ion in water containing
phosphate buffer (c¼0.2 M, pH¼6.0) at a bare glassy carbon elec-
trode are shown in Figure 10. The cyclic voltammogram of 12 in the
absence of azide ion (curve a) shows one anodic peak (A1) at 0.29 V
and the corresponding cathodic peak (C1) at 0.22 V, which corre-
sponds to the transformation of 2,3-dihydroxybenzoic acid (12)
According to the proposed mechanism, the cathodic peaks C1,
C2, C3, and C4 can be related to the reduction of intermediates 12ox,
13ox (counterpart of 13), 14, and 15ox (counterpart of 15),
respectively.
HOOC
O
O
HOOC
N3
OH
OH
HOOC
OH
OH
-
-2e- -2H+
+N3
+H+
8ox
8
9
HOOC
H2N
O
HOOC
N3
OH
OH
-N2
3. Conclusions
Intramolecular
Oxidation-Reduction
O
10
9
The present results complete the previous reports on the
anodic oxidation of catechol in the presence of azide ion.24 In this
report, the mechanism of the reaction of electrochemically
generated para and ortho-benzoquinones with azide ions is in-
vestigated, and a method of general applicability for their high-
yield synthesis under mild experimental conditions is provided.
HOOC
H2N
O
O
H2N
H2N
O
- +H+
+N3
+N2 +CO2
O
Schmidt Rearrangement
10
11
Scheme 7.

D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
4749
A1
A1
95
25
20
15
10
5
f
75
b
e
55
a
35
d
c
15
-5
0
-5
-25
-45
C4
C3
C2 C1
C2
-10
-0.4
-0.1
0.2
E/V vs. SCE
0.5
-0.4
-0.1
0.2
E/V vs. SCE
0.5
Figure 11. Typical cyclic voltammograms of 1 mM 2,3-dihydroxybenzoic acid (12) in the presence of 15 mM azide ion at a glassy carbon electrode, in aqueous solution containing
0.2 M phosphate buffer (pH 6.0). Scan rates from (a) to (f) are: 10, 50, 100, 200, 400, and 1000 mV sꢀ1, respectively. t¼25ꢁ1 ꢂC.
We observed an interesting diversity in the mechanism of elec-
trochemical oxidation of hydroquinone (1) and other para and
ortho dihydroxybenzoic acid derivatives (4, 8, and 12) in the
presence of azide ion. In the case of hydroquinone (1), the final
product is diazido-p-benzoquinone 3ox, whereas in the cases of
2,5-dihydroxybenzoic acid (4) and 3,4-dihydroxybenzoic acid (8),
the final products (7 and 11) are diaminobenzoquinones that were
obtained after intermolecular Michael addition reaction, intra-
molecular oxidation–reduction and Schmidt reactions. Also, in the
case of 2,3-dihydroxybenzoic acid (12), the final product, which is
diamino-o-benzoquinone 16, was obtained after two successive
intermolecular Michael addition reactions and intramolecular
oxidation–reduction. The difference in behavior of electrochemical
oxidation of hydroquinone (1) with other para and ortho dihy-
droxybenzoic acid derivatives (4, 8, and 12) in the presence of
azide ion is attributed to the enol–keto equilibrium due to intra-
molecular hydrogen bonding as well as the easy loss of dinitrogen
due to the more electron-withdrawing character of para (because
of the presence of carboxyl group) and ortho-quinone rings. Fi-
nally, we can state that the addition of azide anion to hydroqui-
none (1) and other para and ortho dihydroxybenzoic acid
derivatives (4, 8, and 12) is an interesting reaction that provides
direct access to diazido-p-benzoquinone and diaminobenzo-
quinones, respectively.
4. Experimental
4.1. Electrodes, electrochemical instruments, and chemicals
Cyclic voltammetry, controlled-potential coulometry and pre-
parative electrolysis were performed using a Behpajoh model BHP-
2062 potentiostat/galvanostat. The working electrode used in the
voltammetry experiments was a glassy carbon disc (1.8 mm2 area)
and platinum wire was used as a counter electrode. The working
electrode used in controlled-potential coulometry and macroscale
electrolysis was an assembly of four graphite rods (31 cm2) and
a large stain less steely gauze constitutes the counter electrode. The
working electrode potentials were measured versus SCE (all elec-
trodes from AZAR electrode). Hydroquinone, 2,5-dihydroxybenzoic
acid, 3,4-dihydroxybenzoic acid, and 2,3-dihydroxybenzoic acid
were reagent-grade materials from Aldrich and sodium azide,
phosphoric acid, acetic acid, ammonia, etc. were of pro-analysis
grade from E. Merck. These chemicals were used without further
purification. The homogeneous rate constants were estimated by
analyzing the cyclic voltammetric responses, using the DigiElch
simulation software.37 An excellent fit between the experimental
and simulated data was obtained over this range of experimental
conditions for the following kinetic parameter values.
4.2. Electroorganic synthesis of 3ox, 7, 11, and 16
COOH
OH
COOH
COOH
OH
A solution (ca. 80 mL) of aqueous phosphate buffer solution
(c¼0.2, pH 6.0) containing 3 mmol of hydroquinone (or 2,5-dihy-
droxybenzoic acid, 3,4-dihydroxybenzoic acid, 2,3-dihydroxy-
benzoic acid) and 20 mmol of sodium azide was electrolyzed at
potential of peak A1 in an undivided cell equipped with a carbon
anode and a large stain less steely gauze as cathode, at 25 ꢂC. The
electrolysis was terminated when the current decayed to 5% of its
original value. The process was interrupted during the electrolysis
and the carbon anode was washed in acetone in order to reactivate
it. For the synthesis of 3ox, 7, and 11 at the end of electrolysis, cell
was placed in refrigerator overnight. The solid precipitated was
collected by filtration and was washed several times with water.
The synthesis of 16 performed under similar conditions as for the
3ox, 7, and 11, but at the end of electrolysis, after evaporation of
solvent, the product was purified by column chromatography (silica
N3
O
-2e- -2H+
-
+N3
+H+
OH
O
OH
12ox
13
12
COOH
COOH
H2N
N3
OH
H2N
O
-
-N2
Intramolecular
Oxidation-Reduction
+N3
+H+
OH
O
15
14
COOH
H2N
H2N
Scheme 8.
O
-N2
Intramolecular
Oxidation-Reduction
O
16

4750
D. Nematollahi, H. Khoshsafar / Tetrahedron 65 (2009) 4742–4750
gel, dichloromethane/ethanol (80:20)). The isolated yields of 3ox, 7,
11, and 16 are 67%, 71%, 73%, and 70%, respectively.
Center of Excellence in Development of Chemical Methods
(CEDCM) for support this work.
4.3. Characteristics of products
References and notes
1. Couladouros, E. A.; Plyta, Z. F.; Haroutounian, S. A.; Papageorgiou, V. P. J. Org.
Chem. 1997, 62, 6.
2. Scriven, E. F.; Turnbull, K. Chem. Rev. 1988, 88, 297.
4.3.1. 2,5-Diazido-1,4-benzoquinone (3ox)
The product 3ox was obtained as orange precipitate, (67% yield),
mp 96–68 ꢂC (dec) (lit. 93–9413). IR (Nujol) 2137 and 2093 (azide),
1651 (quinone carbonyl), 1573, 1463, 1377, 1277, 1213, 868, 761,
3. Labbe, G. Chem. Rev. 1968, 69, 345.
4. Weyler, W.; Pearce, D. S., Jr.; Moore, H. W. J. Am. Chem. Soc. 1973, 95, 2603.
5. Moore, H. W.; Decker, O. H. W. Chem. Rev. 1986, 86, 821.
6. Germeraad, P.; Moore, H. W. J. Org. Chern. 1974, 39, 774.
7. Weyler, W.; Duncan, W. G., Jr.; Moore, H. W. J. Am. Chem. Soc. 1975, 97, 6178.
8. Pearce, D. S., Jr.; Locke, M. J.; Moore, H. W. J. Am. Chem. Soc. 1975, 97, 6181.
9. Germeraad, P.; Weyler, W., Jr.; Moore, H. W. J. Org. Chem. 1974, 39, 781.
10. Cajipe, G. J. B.; Landen, G.; Semler, B.; Moore, H. W. J. Org. Chem. 1975, 40,
3874.
11. Naruta, Y.; Nagai, N.; Arita, Y.; Maruyama, K. J. Org. Chem. 1987, 52, 3956.
12. Chow, K.; Moore, H. W. J. Org. Chem. 1990, 55, 370.
13. Moore, H. W.; Shelden, H. R.; Shellhamer, D. F. J. Org. Chem. 1969, 34, 1999.
14. Moore, H. W.; Shelden, H. R.; Deters, D. W.; Wikholm, R. J. J. Am. Chem. Soc. 1970,
92, 1675.
722 cmꢀ1
etone-d6, 22.5 MHz):
;
1H NMR (acetone-d6, 90 MHz):
116.0, 146.1, 181.5.
d
6.28 (s); 13C NMR (ac-
d
4.3.2. 2,5-Diamino-1,4-benzoquinone (7)
The product 7 was obtained as purple precipitate, (71% yield),
mp >380 ꢂC. IR (KBr) 3352, 3120, 1663, 1534, 1418, 1257, 1061, 843,
749, 681 cmꢀ1; 1H NMR (DMSO-d6, 90 MHz):
d
5.32 (s, 2H), 7.32 (br
4H); 13C NMR (DMSO-d6, 75 MHz):
d 116.1, 150.2, 178.5. MS (m/z)
ꢃ
(relative intensity) 138 (Mþ , 100), 111 (16), 70 (19).
15. Boger, D. L.; Duff, S. R.; Panek, J. S.; Yasuda, M. J. J. Org. Chem. 1985, 50, 5782.
16. Boger, D. L.; Duff, S. R.; Panek, J. S.; Yasuda, M. J. J. Org. Chem. 1985, 50, 5790.
17. Renault, J.; Giorgi-Renault, S.; Baron, M.; Mailliet, P.; Paoletti, C.; Cros, S.; Voisin,
E. J. J. Med. Chem. 1983, 26, 1715.
18. Kende, S.; Ebetino, F. H. Tetrahedron Lett. 1984, 25, 923.
19. Nagaoka, H.; Kishi, Y. Tetrahedron 1981, 37, 3873.
20. Thomson, R. H. Naturally Occurring Quinones, 2nd ed.; Academic: New York, NY,
1971.
21. Lancini, G.; Zanichelli, W. In Antibiotics; Perlman, D., Ed.; Academic: New York,
NY, 1971; pp 531–600.
4.3.3. 4,5-Diamino-1,2-benzoquinone (11)
The product 11 was obtained as violet precipitate, (73% yield),
mp >300 ꢂC (lit. >300 ꢂC24). IR (KBr) 3431, 3066, 1722, 1684, 1577,
1532, 1465, 1294, 831, 670 cmꢀ1 1H NMR (DMSO-d6, 90 MHz):
;
d
5.32 (s, 2H), 7.44 (br 4H); 13C NMR (DMSO-d6, 22.5 MHz):
153.4, 178.8.
d 98.6,
22. Lin, T. S.; Xu, S. P.; Zhu, L. Y.; Divo, A.; Sartorelli, A. J. Med. Chem. 1991, 34, 1634.
23. Huang, Z. D.; Chen, Y. N.; Menon, K.; Teicher, B. A. J. Med. Chem. 1993, 36,
1797.
24. Nematollahi, D.; Afkhami, A.; Tammari, E.; Shariatmanesh, T.; Hesari, M.;
Shojaeifard, M. Chem. Commun. 2007, 162.
25. Nematollahi, D.; Tammari, E. J. Org. Chem. 2005, 70, 7769.
26. Nematollahi, D.; Rafiee, M. Green Chem. 2005, 7, 638.
27. Nematollahi, D.; Habbibi, D.; Rahmati, M.; Rafiee, M. J. Org. Chem. 2004, 69,
2637.
28. Nematollahi, D.; Goodarzi, H. J. Org. Chem. 2002, 67, 5036.
29. Hosseiny Davarani, S. S.; Nematollahi, D.; Mashkouri Najafi, N.; Masoumi, L.;
Ramyar, S. J. Org. Chem. 2006, 71, 2139.
30. Nematollahi, D.; Workentin, M. S.; Tammari, E. Chem. Commun. 2006, 1631.
31. Nematollahi, D.; Amani, A.; Tammari, E. J. Org. Chem. 2007, 72, 3646.
32. Nematollahi, D.; Shayani-jam, H. J. Org. Chem. 2008, 73, 3428.
33. Greef, R.; Peat, R.; Peter, L. M.; Pletcher, D.; Robinson, J. Instrumental Methods in
Electrochemistry; Ellis Horwood: New York, NY, 1990; p 189.
34. Bjerrum, J.; Schwarzenbach, G.; Sillen, L. G. Stability Constants of Metal-ion
Complexes; Chemical Society: London, 1958.
35. Schmidt, K. F. Ber. 1924, 57, 704.
36. Patai, S. The Chemistry of the Azido Group; Wiley: New York, NY, 1971.
37. Rudolph, M. J. Electroanal. Chem. 2002, 529, 97 Also, see: http://www.digielch.de.
4.3.4. 2,3-Diamino-5,6-dioxocyclohexa-1,3-dienecarboxylic
acid (16)
The product 16 was obtained as dark violet precipitate, (70%
yield), mp >350 ꢂC. IR (KBr) 3179,1680,1623,1558,1470,1406, 1113,
822, 701 cmꢀ1; 1H NMR (DMSO-d6, 90 MHz):
d 5.63 (s, 1H), 7.77 (br
2H, this peak disappears in the presence of D2O), 9.5 (br about 2H,
this peak disappears in the presence of D2O), 11.16 (br 1H, this peak
disappears in the presence of D2O); 13C NMR (DMSO-d6, 22.5 MHz):
d
96.6, 102.2, 150.9, 160.7, 169.7, 176.7, 179.1. MS (m/z) (relative in-
ꢃ
tensity) 182 (Mþ , 1.5), 138 (3.9), 135 (5.2), 68 (100).
Acknowledgements
We would like to thank Dr. M. Rudolph for his cyclic voltam-
mogram digital simulation software (DigiElch SB) and the authors
acknowledge the Bu-Ali Sina University Research Council and