Organic Letters
Letter
On the basis of the aforementioned mechanistic rationale
(Scheme 1a), we realized that the 2-azo-benzonitrile (ene-ene-
yne) scaffold 3 can easily be accessed from the corresponding
azido indazole derivative 2 via a denitrogenative ring
fragmentation−coarctate reaction (Scheme 1b). Traditional
methods for the azidation consistently require a stoichiometric
amount of oxidants, including peroxydisulfates, hypervalent
iodines, or high-valent metals, a high temperature, and a
prolonged reaction time.9 In addition, azidotrimethylsilane
(TMSN3) is the most commonly used azide source, which is
toxic, volatile, and relatively expensive. As a part of our
ongoing research program on the development of reaction
methodologies based on redox processes,10 we envisioned a
direct electrochemical azidation of the indazole derivatives to
produce 2 by using inexpensive sodium azide. This economical
and environmentally benign approach is a new addition to the
available methods for the functionalization of indazoles11 and
also bypasses the multistep synthesis of azido derivatives
involving harsh oxidants and toxic reagents.
Recently, organo-electrochemistry has realized a renaissance
for redox-mediated transformations,12 which has opened a new
platform in organic synthesis from an environment-friendly
and economic perspective. The use of electricity concedes the
avoidance of stoichiometric redox reagents and also restricts
the use of toxic reagents and the generation of unwanted
byproducts.13 In 2017, Lin first established an electrochemical
approach for the diazidation of alkenes from sodium azide.14
Herein we report the successful implementation of the
electrogenerated azide radical in the C3−H azidation of
various 2H-indazoles and a follow-up fragmentation to access
the conjugated 2-azo-benzonitrile derivatives. Advantageously,
this undivided cell electrolysis proceeds under external-
chemical-oxidant-free condition at room temperature and
under air, which supports its sustainability.
formed in 66% yield. Further optimization revealed that the
MeCN/TFA mixture furnished better results, and 3a was
1
obtained in 81% H NMR yield and in 78% isolated yield.
Presumably, trifluoroacetic acid (TFA) facilitates the solubility
of the reagents and also enhances the cathodic proton
reduction.
Significantly lower yields of 3a were obtained in the MeCN/
H2O (entry 3) mixture and also in the dry MeCN (entry 4).
Either raising the current flow to 16 mA or lowering it to 5 mA
also resulted in a decreased reaction yield (entries 5 and 6). In
the absence of the manganese catalyst, a 40% yield of 3a was
detected, most likely through the direct anodic oxidation of
azide (entry 7), but only 5 mol % MnBr2 assists in efficiently
reinforcing the oxidation of azide and improves the yield.
However, another manganese salt displayed lower efficiency
(entry 8). As expected, no desired product was obtained
without running the current (entry 9).
To disclose the utility of the developed method, the
substrate scope with various N-2-substituted 2H-indazoles
was surveyed (Table 2). In the beginning, the effects of
different N-2-substituted 2H-indazoles were explored. 2-Aryl-
2H-indazoles bearing electron-donating ortho-, meta-, and
para-substituted alkyl (−Me, −nBu, −tBu) and alkoxy (−OMe)
groups in the phenyl ring efficiently provided the correspond-
ing products in good yields (3a−3h). The reaction with
halogen substituents such as para-F, para-Cl, and meta-Cl in
the phenyl ring occurred smoothly for this transformation,
furnishing moderate to good yields of the expected products
(3i−3k). In addition, electron-deficient arene- (−CN- and
−COMe-substituted) and heteroarene-bearing 2H-indazoles
were found to be compatible with this reaction protocol to give
the corresponding azo-ene-yne products (3l−3o). Naphthyl-
substituted indazole derivative gave rise to the desired product
3p in 72% yield. Moreover, aliphatic-substituted 2H-indazole
at the N-2 position responded well in this system to provide a
good yield (3q−3r).
Our investigation was initiated by using 2-(p-tolyl)-2H-
indazole (1a) and sodium azide as the model substrates in an
undivided cell equipped with a graphite anode and a platinum
cathode (Table 1). When the electrolysis was regulated at a
constant current of 10 mA in MeCN/AcOH solvent mixture
and LiClO4 as the electrolyte in the presence of MnBr2 (5 mol
%) catalyst, the desired 2-azo-benzonitrile product 3a was
Next, we evaluated differently substituted arene moieties in
2H-indazoles to justify the general applicability of the
developed method (Table 3). The electron-rich, 5,6-
dimethoxy-substituted 2H-indazole conversantly delivered the
expected rearrangement products in very good yields (3s, 3t).
Single-crystal X-ray analysis of 3s was determined to support
the structure. Different 5-halo-substituted (−F, −Cl) arenes
were well tolerated and afforded the product in excellent yields
(3u−3y). The gram-scale reaction of 3u proceeded to give
72% yield, which supports the scalability and applicability of
this protocol.
a
Table 1. Optimization of the Reaction Conditions
To manifest the colossal applicability of the current
protocol, one of the 2-azo-benzonitrile compounds (3u) was
synthetically converted to different valuable derivatives
(Scheme 2). Selective Grignard addition to the azo
functionality was performed at a lower temperature without
affecting the nitrile group to form 4u in good yield. When the
temperature was increased and an excess amount of Grignard
reagent was used, addition to both azo and nitrile
functionalities was observed to deliver the substituted ketone
5u in excellent yield. In both cases, the alkyl group
regioselectivity was added to the more electron-deficient N
atom, which was confirmed by a nuclear Overhauser effect
(NOE) experiment. (See the SI for details.) Nucleophilic
aromatic substitution of the fluorine atom delivered 6u without
reacting with the other functionalities. Finally, a selective
reduction of the azo group through transfer hydrogenation in
b
entry
deviation from the standard condition
yield (%)
66
81 (78)
20
10
70
72
40
58
0
1
2
3
4
5
6
7
8
9
MeCN/AcOH (12:1) as the solvent
none
MeCN/H2O (12:1) as the solvent
MeCN as the solvent
5 mA for 5 h
16 mA for 1.5 h
without catalyst
Mn(OAc)2 instead of MnBr2
without current
c
a
Reaction conditions: 1a (0.2 mmol), NaN3 (0.6 mmol), MnBr2 (5
mol %), LiClO4 (0.1 M) in MeCN/TFA (20:1, 6 mL), C anode, Pt
cathode, undivided cell, constant current = 10 mA, at rt under air for
b
c
2.5 h. NMR yield. Isolated yield.
1743
Org. Lett. 2021, 23, 1742−1747