Electrochemical reduction of nitrotriazoles in aqueous media as an approach to the synthesis of new green energetic materials

Article abstract of DOI:10.1039/c1nj20770a

The synthesis of new azo and azoxy compounds via electrochemical reduction of nitrotriazoles has been investigated in aqueous media, using nitrotriazolone (NTO) and nitrotriazole (NTr) as representative substrates. Reduction of NTO produces mainly solid azoxytriazolone (AZTO), with azotriazolone (azoTO) and aminotriazolone (ATO) as minor products, while 3-hydroxylaminotriazole is the major product formed from NTr. AZTO and azoTO are of interest as new green high-nitrogen compounds for use as insensitive high explosives (IHE). The effect of varying reaction conditions such as pH and substrate concentration has been evaluated, and a mechanism is proposed accounting for the experimental observations. In particular, the ratio of azoxy to azo in the solid product is influenced by pH and temperature, and the minor product ATO is formed not via direct reduction of NTO but via a novel thermal disproportionation reaction of the hydrazotriazolone intermediate. Conditions of high substrate concentration and low cell temperature maximise the azoxy yield and minimise the formation of minor products. Results indicate that this green electrosynthetic approach may be generally useful for the synthesis of new azoxy and azo triazoles from suitable substrates.

Full text of DOI:10.1039/c1nj20770a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 2894–2901
www.rsc.org/njc
PAPER
Electrochemical reduction of nitrotriazoles in aqueous media as an
approach to the synthesis of new green energetic materialsw
Lynne Wallace,* Christie J. Underwood, Anthony I. Day and Damian P. Buck
Received (in Victoria, Australia) 5th September 2011, Accepted 3rd October 2011
DOI: 10.1039/c1nj20770a
The synthesis of new azo and azoxy compounds via electrochemical reduction of nitrotriazoles
has been investigated in aqueous media, using nitrotriazolone (NTO) and nitrotriazole (NTr) as
representative substrates. Reduction of NTO produces mainly solid azoxytriazolone (AZTO), with
azotriazolone (azoTO) and aminotriazolone (ATO) as minor products, while
3-hydroxylaminotriazole is the major product formed from NTr. AZTO and azoTO are of
interest as new green high-nitrogen compounds for use as insensitive high explosives (IHE). The
effect of varying reaction conditions such as pH and substrate concentration has been evaluated,
and a mechanism is proposed accounting for the experimental observations. In particular, the
ratio of azoxy to azo in the solid product is influenced by pH and temperature, and the minor
product ATO is formed not via direct reduction of NTO but via a novel thermal
disproportionation reaction of the hydrazotriazolone intermediate. Conditions of high substrate
concentration and low cell temperature maximise the azoxy yield and minimise the formation of
minor products. Results indicate that this green electrosynthetic approach may be generally useful
for the synthesis of new azoxy and azo triazoles from suitable substrates.
other incompletely oxidised explosive residues. High nitrogen
Introduction
compounds are used, or are being considered for use, in many
applications; for example, as new IHE,14,18 as cleaner, cooler
Green methodologies are increasingly being introduced in the
field of energetic materials.1–4 In addition to the international
burning propellants,15 and also in some civilian applications,
such as gas-generating compounds in vehicle airbags.2 Azo,
move towards insensitive high explosives (IHE), which are
safer, less sensitive materials that retain the performance of
azoxy and hydrazo-based systems comprise one of the more
conventional explosives,5 there is now a growing realisation
promising classes of high nitrogen compounds,14–19 however
that new explosives also need to be more environmentally
the current synthetic procedures for these generally utilise
friendly. An increasing amount of research has therefore been
strong chemical oxidising or reducing agents, such as potassium
focused on initiatives to mitigate the environmental impact of
permanganate, N-bromosuccinimide, concentrated sulfuric acid
and hydrazine monohydrate.14–16
the production and use of energetic materials. These approaches
range from developing greener methods of synthesis for existing
explosives6–8 and treating manufacturing waste,3,9 through
are increasingly favoured as they are often considered to be
Electrochemical methods of synthesis and waste remediation
studies of the environmental fate of explosives,10,11 to the design
economical and environmentally friendly.20–22 The use of electrons
and synthesis of safer and greener end products.1,2,4,12–19 One
as reagents avoids introduction of potentially toxic oxidising
promising approach is to replace some conventional explosives
or reducing agents, reaction conditions are generally mild, and
with high-nitrogen compounds. These materials contain a higher
the electrodes can be considered as heterogeneous catalysts,
proportion of nitrogen by mass, compared to conventional
readily re-used. Previously, we reported on the formation of
explosives, and they derive their energy output from this factor,
the new high nitrogen compound azoxytriazolone (AZTO),
rather than via the oxidation of fuel elements (carbon, hydrogen)
which is produced by electrochemical reduction of nitro-
triazolone (NTO) in acidic aqueous solution.23 NTO is a
by oxygen. Since nitrogen gas (N2) is the major product of
explosion, high-nitrogen compounds burn more cleanly than other
new insensitive high explosive (IHE) that shows promise as a
organic explosives, producing less carbon monoxide, soot and
safer replacement for standard explosive materials in several
applications.24 One issue with its manufacture has been the
School of Physical, Environmental and Mathematical Sciences,
University of New South Wales at the Australian Defence Force
solubility of NTO means that the wastewater cannot be treated
Academy, PO Box 7916, Canberra BC ACT 2610, Australia
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c1nj20770a
problem of treating the wastewater, since the very high aqueous
by conventional means. In previous work, we investigated the
direct electrolysis of aqueous NTO solutions as a means of
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
2894 New J. Chem., 2011, 35, 2894–2901

View Article Online
removing the soluble organic material from waste streams.23,25
This research led to the discovery of AZTO, which precipitates
from solution in good yield.23 The structure and formula of
AZTO indicated that it also had the potential to be used as an
IHE, and preliminary results have confirmed that it responds
as a typical IHE in small-scale tests.26 For a typical AZTO
sample, the sensitivity to impact and electrostatic discharge
(ESD) was identical to that measured for NTO27 (impact Figure of
Insensitivity = 100, ESD ignition at 4.5 J not 0.45 J) and the
friction sensitivity was lower (>360 N compared to 252 N for
NTO). In particular, AZTO exhibited greater thermal stability than
the parent NTO, as is often the case for azoxy and azo materials.14
Thus, the formation of AZTO from waste NTO solutions could
prove to be a very economical method of remediation.
The reduction of NTO to form AZTO is an interesting
reaction in itself. While it is well know that the electrochemical
reduction of nitrobenzene species can produce azoxybenzenes
in some cases (via coupling of the initial nitroso and hydroxy-
lamine electrode products),28 to date there are very few
examples of this type of electrosynthesis in 5-membered
heterocyclic systems.29
Fig. 1 Structures of (a) NTO (5-nitro-1,2,4-triazol-3-one); (b) NTr
(3-nitro-1,2,4-triazole) and possible reduction products (R = triazolone,
triazole): (c) nitroso, (d) hydroxylamine, (e) azoxy, (f) hydrazo, (g) azo,
(h) amine.
which would be expected to hydrolyse rapidly in the aqueous
medium. For both NTO and NTr, HPLC analysis showed that
all substrate was consumed within 5 h.
In the case of AZTO, the efficiency of azoxy formation
appeared to be related at least partly to the very low solubility
of the product, which precipitated readily from aqueous
solution. NMR spectroscopy showed that, despite appearing
to be microanalytically pure, AZTO samples contained an
unidentified minor species that could not be removed by
recrystallisation.
Identification of reduction products
In the case of NTO, the HPLC chromatogram of the filtered
product solution showed a strong peak at 7.0 min corres-
ponding to aminotriazolone (ATO), which was formed in 14%
yield. A second, very weak, peak occurred at 3.1 min. This was
assigned as hydroxylaminotriazolone on the basis of the UV
absorbance profile and the trend in retention times seen for the
triazole derivatives (nitro > amine > hydroxylamine, see
below). This species was present in very low amounts
(o0.5%). There was also a small peak at 0.85 min due to
nitrate ion, consistent with some oxidative mineralisation
occurring at the counter electrode.
In the present work we report a comprehensive study of the
mechanism of this novel electrochemical reaction and the
factors influencing product distribution. This is particularly
important in optimising the yield of the desired products. A
full understanding of these factors will assist in the potential
industrial application of a very economical synthesis, and the
results are also relevant to laboratory syntheses of potential
new high-nitrogen compounds. The minor component present
in AZTO samples has been identified as azotriazolone
(azoTO), another new high nitrogen compound, and we have
found that experimental conditions can favour this species to
the extent that it becomes the major component of the solid
product. In particular we wished to ascertain whether it was
possible to avoid or limit formation of azoTO in the cell, and
maximise the yield of AZTO. Some experiments were carried
out on nitrotriazole (NTr), a compound structurally similar to
NTO and containing a non-exchangeable proton, in order to
assist in spectroscopic assignments and to assess the wider
applicability of this electrosynthetic reaction in triazole systems.
Fig. 1 shows the structures of NTO and NTr, and possible
reduction products.
The chromatogram of the final filtered NTr solution showed
a peak at 2.9 min due to 3-hydroxylaminotriazole, formed in
81% yield, and traces of a peak at 4.8 min due to aminotriazole
(o0.5% yield). The results of typical HPLC analyses are shown
in Fig. 2.
The solid product of NTO reduction has previously been
identified as azoxytriazolone (AZTO), with a small amount of
another species also evident in the NMR spectra.23 AZTO has
significant solubility only in DMSO or neat sulfuric acid. The
minor component (15%) in the precipitate was identified as
azotriazolone (azoTO), having been isolated in pure form via
reduction of AZTO (see below), and fully characterised.
AZTO and azoTO have close/overlapping peaks in both
1H and 13C NMR spectra in DMSO-d6. Trace amounts
(o3%) of hydrazotriazolone (see below) were usually also evident
by NMR; this species could be removed by recrystallisation from
DMSO-water if necessary. This process does not remove azoTO,
which tends to be found in slightly greater proportions in
recrystallised material.
Results and discussion
Upon reduction at À1.2 V vs. SCE in 0.1 mol LÀ1 H2SO4,
NTO solutions turned first green, then yellow, before a thick
yellow precipitate was formed. NTr solutions turned light blue
initially, then pale yellow, and a small amount of yellowish-
white precipitate was formed. The green (NTO) and blue
(NTr) colours observed in the initial stages of reduction could
be due to intermediate nitroso species, or to anion radicals,
The precipitate from NTr reduction had more limited
solubility than the NTO product (AZTO and azoTO); neat
H2SO4 was the only suitable solvent for this material, with
D2SO4 being used for NMR characterisation. The major
component of this precipitate was identified as azotriazole,
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2894–2901 2895

View Article Online
The unassigned consumed substrate (‘‘other’’) is attributed to
a combination of: (1) loss to counter electrode compartment;
(2) low concentrations of dissolved species (azo, azoxy and
hydrazo) that are not observed by HPLC; and (3) losses
through washing of precipitates. Loss of substrate (or dissolved
reduction products) can occur through exposure to the counter
electrode via diffusion, which is unavoidable. Previous results
indicate this can occur at levels of 10–20% in anodic oxidation
with a substrate concentration of 0.05 mol LÀ1).25 UV-vis
spectroscopy indicates that AZTO/azoTO is present in product
solutions at concentrations representing o2% of NTO
consumed under standard conditions. Since these products
are not observed by HPLC it seems that they have a strong
affinity for the column and are not readily eluted. Oxidative
mineralisation is therefore the principal fate of substrate in the
unaccounted ‘‘other’’ category, though under certain conditions
(see below) more significant concentrations of dissolved azo,
azoxy or hydrazo species could also be present.
Electrochemical reduction of NTO. In a ‘typical’ electrolysis,
NTO (10 g LÀ1) was reduced at À1.2 V in 0.1 mol LÀ1 H2SO4
at room temperature (E20 1C), to give the products described
in the preceding section. Under these conditions (Fig. 3A1),
ATO accounts for 14% of the NTO consumed, and the
precipitate accounts for a further 69% (solid contains AZTO
and azoTO in 5.7 : 1 ratio).
Fig. 2 HPLC analysis of the reduction of (a) NTO (circles) to form
ATO (squares) and (b) NTr (diamonds) to form hydroxylaminotriazole
(triangles). Solid azoxy and azo species removed via filtration prior to
analysis.
Effect of cell potential. As further reduction of AZTO was
considered to be the most likely source of azoTO, the minor
component in the solid product, it was necessary to establish
whether this could be prevented or limited by use of a lower
cell potential. Cyclic voltammetry has shown that reduction of
NTO occurs at À0.38 V in acidic solution,25 well below the cell
potential of À1.2 V typically used in our experiments. No
further peaks were observed in the voltammogram before
À1.4 V, where the solvent front began. However, when a lower
applied potential of À0.4 V was used in the bulk electrolysis, a
slightly higher proportion of azoTO was found in the cell
precipitate (AZTO : azoTO = 4.3 : 1, Fig. 3A2). The reduction
proceeded more slowly at the lower voltage and, importantly,
ATO was formed in solution to approximately the same extent.
This result suggests ATO is not formed via further reduction of
by comparison with an authentic sample prepared by a chemical
method.15 The minor component, which gave rise to two
1
CH peaks in the H NMR spectrum at very similar chemical
shift to those for azotriazole, was assigned as azoxytriazole, with
the ratio of azo : azoxy being approximately 60 : 40. The 13C
spectrum also showed one set of two major peaks and another
of four minor peaks, with similar chemical shift to those for
AZTO and azoTO. The IR spectrum of the precipitate gave peaks
at 1365 and 1338 cmÀ1, consistent with the azo/azoxy functionality,
with the 1365 cmÀ1 peak being more intense in samples containing
a higher proportion of the azoxy compound.
Effect of electrochemical reaction conditions on product
distribution
Since the azo and azoxy species derived from both NTO and
NTr are of potential interest as energetic materials, a more
comprehensive investigation was then carried out to determine
the effect of different reaction conditions on the product
distribution. In particular, we wished to assess whether it
was possible to obtain pure samples of AZTO and/or azoTO
by control of the cell parameters. It was also important to
ascertain the mechanism by which the azo species is formed,
and the source of ATO in the NTO reductions.
Aqueous solutions of NTO or NTr were reduced under
different conditions of applied potential, cell temperature,
substrate concentration and solution pH. Reactions were
followed by HPLC, and any solid products were collected
and characterised by NMR spectroscopy. The results are
summarised in Fig. 3.
Fig. 3 Fate of substrate (A) NTO and (B) NTr upon reductive
electrolysis under different cell conditions: (1) Standard conditions
(0.077 M substrate in 0.1 mol LÀ1 H2SO4, T E 20 1C, À1.2 V);
(2) Applied potential = À0.4 V; (3) Cell temperature = 5–10 1C;
(4) [Substrate] = 0.0077 mol LÀ1; (5) Supporting electrolyte =
0.1 mol LÀ1 Na2SO4.
c
2896 New J. Chem., 2011, 35, 2894–2901
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
and possibly hydrazotriazolone may then contribute more
significantly to the ‘‘other’’ category.
Effect of increased cell pH. In electrolyses carried out at
higher pH (initial pH = 2, final pH = 6–7), with 0.1 mol LÀ1
Na2SO4 as electrolyte (Fig. 3A5), azoTO again became the
dominant component of the solid product (1 : 1.6 AZTO :
azoTO) and ATO was formed to a greater extent (39% based
on starting NTO). The overall yield of solid product (24%)
was much lower than in the acid experiments.
Electrochemical reduction of NTr. Bulk electroreduction
experiments were also carried out on solutions of NTr, to assess
whether the observations made for NTO were applicable to
other nitroazoles. The results are shown in Fig. 3B1–B5.
Reduction of NTr occurred efficiently and at a similar rate
to reduction of NTO, but in this case the major product of
reduction was identified as hydroxylaminotriazole, a soluble species
which was isolated as the sulfate salt and fully characterised.
This observation is in agreement with the work of Kokkinidis
et al.,30 who reported on the voltammetric behaviour of NTr
on platinum-modified electrodes. In the present case, a small
amount of precipitate was also formed, comprising azo-and
azoxytriazole (3%).
Fig. 4 Cyclic voltammograms of (a) AZTO and (b) NTO in
DMSO (0.1 mol LÀ1 TBABF4. Dotted lines show the effect of added
0.1 mol LÀ1 H2SO4(aq).
the hydroxylamine intermediate, since the reduction of hydroxy-
lamine to amine normally occurs at a potential considerably more
negative than that for reduction of nitro to nitroso/hydroxy-
lamine.28 In addition, this result shows that further reduction of
AZTO to azoTO cannot be eliminated by use of a lower cell
potential.
The effect of reaction conditions on the precipitate mirrored
the results obtained for NTO: use of a lower temperature gave
a higher yield of precipitate and increased the proportion of
azoxy to azo (B3); and electrolysis at higher pH also increased
the amount of azoxytriazole in the precipitate (B5). However,
in all cases the hydroxylamine was the major product of
reduction. Aminotriazole was produced only in low amounts.
This is consistent with the conclusion drawn for NTO: that
further reduction of hydroxylamine directly to amine does not
occur at the cell potentials used in this study. It also accords
with previously reported voltammetric results,23 which show
no reductions for NTr between À0.4 and À1.4 V. As observed
for NTO, lower substrate concentration (B4) and higher
pH (B5) resulted in slightly higher concentrations of the amine.
Cyclic voltammetry of both AZTO and NTO in DMSO
(Fig. 4) showed that the reduction peaks occurred at very
similar potentials of À0.76 and À0.72 V vs. Ag/AgCl. (DMSO
is the only suitable solvent in which a measurable concen-
tration of AZTO could be achieved). To provide some insight
into the likely behaviour in aqueous acid, several drops of
0.1 mol LÀ1 H2SO4 were added to the voltammetric solutions.
This caused an anodic shift in both peaks but the potentials
remained very close (À0.67 and À0.61 V respectively).
This supports the observations found in the bulk electrolysis
experiments: during the reduction of NTO, further reduction
of AZTO cannot be avoided by control of the cell potential.
Electrochemical reduction of AZTO. To obtain further
information on the mechanism of further reduction of AZTO,
and the likely source of ATO, bulk reduction experiments were
carried out on AZTO (samples containing E15% azoTO)
in 0.1 mol LÀ1 Na2SO4(aq). In these experiments, the initial
solution was adjusted to pH 7–8 by addition of concentrated
NaOH(aq) to achieve sufficient solubility of the starting AZTO.
At alkaline pH, AZTO dissolves fully due to deprotonation.
1H NMR spectroscopy in DMSO showed that this crude
product consisted of a small amount of azoTO and another
major product displaying 3 proton environments of equal
intensity. This second species was present in variable proportions
in different batches, and in one case was formed exclusively.
This allowed the species to be identified as hydrazotriazolone,
confirmed by microanalysis and mass spectrometry. When the
NMR solution was heated strongly, hydrazotriazolone peaks
decreased, while the amount of azoTO increased and new
peaks appeared due to the formation of ATO. This raised the
possibility that ATO was produced from hydrazotriazolone
via a disproportionation reaction (eqn (1)).
Effect of lower cell temperature. We then considered that
cooling the cell might enhance precipitation of solid AZTO
component, thereby inhibiting further reduction. When the
bulk reduction was carried out at 5–10 1C on an ice bath, it
was found that the formation of azoTO was indeed inhibited,
with the precipitate comprising AZTO : azoTO in 19 : 1 ratio
(Fig. 3A3). The overall yield of precipitate was higher (87%)
than obtained at room temperature, and much less ATO was
formed in solution (1%).
Effect of substrate concentration. With a much lower NTO
concentration of 1 g LÀ1 (Fig. 3A4), ATO was formed in
solution to a greater extent (24%). The precipitate was formed
in lower yield (36%) and contained a much higher proportion
of azoTO (1 : 5.7 AZTO : azoTO). Under these conditions,
dissolved AZTO represents a much greater proportion of the
total generated, so that further reduction can become more
significant. At these concentrations, dissolved AZTO, azoTO
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2894–2901 2897

View Article Online
2 R–NH–NH–R - R–NQN–R + 2 R–NH2
(1)
To test this hypothesis, reduction of AZTO was repeated and
the final aqueous solution was divided into two equal portions:
one was immediately acidified to pH E 2 (giving precipitate X,
0.116g), and the other was heated to 80 1C for ca. 5 min before
being cooled and acidified (giving precipitate Y, 0.056 g).
NMR spectroscopy showed that ppt X comprised hydrazo-
triazolone and azoTO in a 9.6 : 1 ratio, while ppt Y consisted
of only azoTO. The molar ratio of hydrazotriazolone in ppt X
to azoTO in ppt Y was 2 : 1 as expected from eqn (1). HPLC
analysis showed that in the filtrate of the heated portion
(filtrate Y) the concentration of ATO had increased almost
5-fold compared to the unheated portion. However for both
portions, the molar ratio of ATO : azoTO was approximately
2 : 1, again as expected from eqn (1). Almost identical product
ratios were obtained when the above thermal procedure was
carried out under an inert atmosphere, demonstrating that
aerial oxidation of hydrazotriazolone is not a significant
reaction pathway here.
Scheme 1 Proposed mechanism for reduction of nitrotriazoles, showing
possible major reaction pathways. Dotted arrows represent possible
minor pathways. Blue arrows correspond to electron transfer steps, and
red arrows to thermal reactions.
reaction to produce azoTO and ATO. This reaction results
in a much lower yield of solid product, since soluble ATO is
formed in up to 40% yield. The results indicate that the
disproportionation reaction of hydrazotriazolone is the only
source of ATO, and the major source of azoTO, in the
reduction of NTO. It is possible that some direct reduction
of AZTO to azoTO also occurs to a limited extent.
Thermal disproportionation of hydrazobenzenes to form
azo compounds and amines has been reported previously,31–33
but this appears to be the first example of such a transformation
involving 5-membered heterocyclic systems. The results suggest
this thermal reaction is the source of ATO and the principal, if
not only, source of azoTO in the reduction of NTO, with
hydrazotriazolone being the major product of reduction of
AZTO in alkaline solution. Since the hydrazotriazolone inter-
mediate is not a major component of precipitates formed in
acid electrolyses, but both azoTO and ATO are produced in
significant amounts, it appears that the disproportionation
rate may be enhanced in acid solution (H+ catalysis). Fig. 3
shows that the molar ratio of ATO : azoTO is generally also
around 2 : 1 in room temperature acid solution (A1, A2, A4),
consistent with this hypothesis.
On reduction of NTr, the major product formed is hydroxy-
laminotriazole, with very low yields of the azoxy or azo species
being formed. This indicates that the coupling of the initial
electrode products, the nitroso and hydroxylamine derivatives,
is much less efficient in this case. In nitrobenzenes, the coupling
reaction occurs between the protonated nitroso derivative and
the unprotonated hydroxylamine.28 The efficiency of this step
depends on the concentrations achieved by both of these
species, and is therefore related to the pKa values for both.
NTr has a much higher pKa value (6.0534) for the ring NH
protons than NTO (3.7635). It therefore appears likely that the
pKa values for the relevant nitroso and/or hydroxylamine
groups are less favourable for coupling in the case of NTr.
Although in this case the amine is formed to only a small
extent, the higher concentrations of this species observed
under conditions of dilute substrate (Fig. 3B4) and higher
solution pH (Fig. 3B5) follow the same trend as observed in
the NTO reactions, implying that a similar mechanism is likely
to operate here.
Proposed reaction mechanism
The results obtained can be explained in terms of the reaction
mechanism presented in Scheme 1.
In acidic solution, the major product formed on reduction
of NTO is the azoxy derivative, AZTO. Thus, the coupling of
nitroso and hydroxylamine derivatives is clearly very efficient
in this case. Further reduction of AZTO can occur to some
extent, producing azoTO and ATO as the final products, but
this is limited by the low solubility of AZTO. If the cell is
cooled, precipitation of solid AZTO is enhanced and the
extent of further reduction is minimised by the lower solution
concentration of AZTO. It is not possible to prevent further
reduction by control of cell voltage, as the reduction potentials
of NTO and AZTO are very similar.
These results suggest that while the electrochemical
formation of azo- and/or azoxytriazoles is possible, the efficiency
depends on the particular substrate. In the case of NTO, the
coupling reaction between the nitroso and hydroxylamine
derivatives is highly favoured, but the corresponding species
for NTr react only slowly.
At higher pH, further reduction of AZTO occurs to a much
greater extent. This is probably due to its increased solubility,
especially in the vicinity of the working electrode where the
local pH is likely to be higher than the bulk solution.28 Any
conditions that increase the proportion of AZTO in solution
(increased temperature, pH or decreased substrate concen-
tration) serve to increase the extent of further reduction. Upon
reduction of AZTO, the main electrode product is hydrazotria-
zolone, which then undergoes a thermal disproportionation
Current efficiency of AZTO synthesis
In an electrosynthesis the current efficiency of the reaction is
an important economical factor. Table 1 shows the current
efficiency of AZTO electrosynthesis at different applied cell
potentials, expressed as the charge passed in forming AZTO as
a percentage of the total charge passed. The current efficiency
for the total solid product (AZTO plus azoTO) is also shown.
c
2898 New J. Chem., 2011, 35, 2894–2901
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
At higher cell potentials the product is formed more quickly
as expected, but the current efficiency decreases. Optimum
current efficiency was achieved with an applied cell potential of
À0.6 V (vs. SCE) and at low temperature (5–10 1C). Under
these conditions the yield of solid AZTO was 71%, formed
with a good current efficiency of 71%. Based on the total yield
of solid product, the current efficiency here is 80%.
or hydroxylaminotriazole could be utilised in the form of ionic
liquids; for example, ionic liquid triazolammonium salts
are currently of interest as desensitising additives for some
explosives formulations.36
Experimental
Materials and methods
NTO was supplied as an aqueous solution (10 g LÀ1), by Defence
Science and Technology Organisation (DSTO, Edinburgh, SA,
Australia). 3-nitro-1,2,4-triazole, 3-amino-1,2,4-triazole, palladium
charcoal, sulfuric acid (98%), sulfuric acid-d2, DMSO-d6, HPLC-
grade acetonitrile, spectrophotometric grade trifluoroacetic acid
(TFA) and Na2SO4Á10H2O were obtained from Sigma-Aldrich,
and high-purity argon from BOC gases. Other reagents were
obtained from Ajax fine chemicals. A sample of azotriazole was
prepared by oxidation of aminotriazole, in an adaptation of the
method of Sivabalan et al.15
Conclusions
Electrosynthesis could be a valuable and environmentally
friendly approach to the synthesis of new high nitrogen
compounds as green energetic materials, using clean metho-
dology, mild conditions and aqueous media. In particular, the
electrochemical technique may prove to be a highly economical
method of treating NTO waste. Three different species (AZTO,
azoTO and ATO) are produced from the reaction, all of which
have useful energetic properties. Reaction conditions can be
chosen to maximise the yield of AZTO (high NTO concen-
tration, low temperature). In a slightly cooled cell the conversion
of NTO to AZTO occurs with very high efficiency. AzoTO and
hydrazotriazolone can be produced by direct reduction of
AZTO. These materials have potential for use as IHE by
themselves, or might act as substrates for nitration to produce
gas-generating solids.
Electrochemistry: Preparative-scale electrolyses were carried
out using a potentiostat that was built in-house, which had
a Æ3 V compliance voltage and 10 A current maximum.
A double carbon plate working electrode was used (dimensions
2.5 Â 4.5 cm), with a platinum basket counter electrode (gauze
dimensions 1.2 cm  4.7 cm) and a saturated calomel reference
electrode (SCE). The counter electrode was housed in a separate
compartment separated by a porous frit. Cyclic voltammetry was
carried out using an eDAQ EA161 potentiostat operated via an
eDAQ ED401 e-corder. A glassy carbon working electrode,
platinum wire counter electrode and Ag/AgCl reference electrode
were used, and the supporting electrolyte was 0.1 mol LÀ1
Na2SO4.
The mechanism of the reduction follows one of the standard
pathways expected for nitrobenzenes, but to date there are
very few examples of electrochemical azo/azoxy synthesis in
5-membered heterocycles—we are aware of only two other
examples.29 The final product distribution is strongly influenced
by substrate concentration, reaction pH and temperature. In
the case of NTO, the high yield of AZTO is due partly to the
low solubility of this species, which limits further reduction to
hydrazo and azo derivatives. The NTr experiments show that
the method can be applied to other triazoles, with a similar
overall reaction mechanism being observed. However the yield
of the desired azo and azoxy products is highly dependent on the
particular substrate. Current work in our laboratory is focussed
on the electrochemical reduction of further heteroaromatic nitro
compounds to form new azo and azoxy derivatives, also under
aqueous conditions.
Chromatography: HPLC analyses were performed on a
Shimadzu VP series instrument with SPD-M10AVP diode array
detector. A Hypercarb 5 mm column (100 mm  3 mm) from
Thermo Electron Corporation was used. The method was
adapted from that used by Le Campion et al.37 The elution
gradient consisted of: t0–5min: 1% aqueous TFA; t10–20: 15%
acetonitrile in 1% aqueous TFA; t25–30: 1% aqueous TFA.
Samples were extracted during electrolysis and left to stand
for at least 24 h to allow any solid material to settle, before
HPLC analysis was conducted on the diluted aliquots. Standard
solutions were using in quantifying NTO (retention time
12.1 min), ATO (7.0 min), NTr (11.0 min), 3-aminotriazole
(4.8 min) and 3-hydroxylaminotriazole (2.9 min). Hydroxy-
laminotriazolone (3.1 min) was estimated by assuming a
response factor similar to that of NTO and ATO (average value).
Molar absorptivities of NTO and ATO are similar at 200 nm in
aqueous acid (respectively 3900 and 3400 molÀ1 L cmÀ1).
FTIR spectra were collected as KBr disks on a Shimadzu
This work has identified the new high-nitrogen compounds
azotriazolone and hydrazotriazolone, in addition to the
previously reported AZTO. These materials have potential for
use as energetic materials or as substrates for production of
new energetic materials.12 In addition, salts of aminotriazolone
Table 1 Current efficiency of electrosynthesis under various cell
conditions
V (vs.
SCE)
t99
% Yield
% Yield
azoTO
CEb per mol AZTO
(plus azo) formed
IRPrestige-21 FTIR spectrometer, at a resolution of 4 cmÀ1
.
T/1C (min)a AZTO
NMR spectra were measured on a Varian Unityplus-400
À0.4
À0.6
À0.6
À1.2
À1.2
20–25 600
20–25 380
5–10 480
20–25 200
5–10 420
51
59
71
58
82
12
10
11
10
4
51% (61%)
57% (65%)
71% (80%)
42% (48%)
52% (53%)
spectrometer.
Preparative scale electrolyses
In typical reactions, 250-mL solutions of the substrate (NTO
or NTr) in either 0.1 mol LÀ1 H2SO4 or 0.1 mol LÀ1 Na2SO4
were reduced for 5–8 h depending on time required for
consumption of substrate. Some reactions were monitored
a
b
Time for >99% consumption of NTO. Current efficiency, based
on 6 Faraday per mol AZTO formed, and 10 Faraday per mol azoTO
formed (via hydrazotriazolone, see Scheme 1).
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2894–2901 2899

View Article Online
by HPLC. NTO was used as supplied (10 g LÀ1, 0.077 mol LÀ1),
and NTr solutions were also prepared to 0.077 mol LÀ1
25.5, H, 3.2, N, 59.4%. IR spectrum: nmax/cmÀ1 3111, 3022,
2875, 2777, 1475, 1365*, 1339, 1273, 1175, 1117, 1030, 972,
908, 768, 741, 648, 577, 436. *This peak was stronger in
samples containing a higher proportion of the azoxy component.
dH (400 MHz, D2SO4, DMSO-d6): azoxytriazole: 8.85 (1H, s,
CH), 8.98 (1H, s, CH); azotriazole: 8.95 (2H, s, CH). dC
(400 MHz, D2SO4, DMSO-d6): azoxytriazole: 156.4, 153.4,
149.0, 146.2; azotriazole: 162.1, 148.0.
.
Precipitates formed during electrolysis were collected by
filtration, washed with water and ethanol and dried at 80 1C.
Electrosynthesis of azoxy-1,2,4-triazole-3-one (AZTO). The
initially colourless NTO solutions turned green within 15 min
and then yellow. AZTO was produced as a thick yellow
precipitate. Yields: typically 65–90%, with higher yields
obtained at low cell temperature (5–10 1C). Mp > 300 1C.
Microanalysis (for a sample comprising 85% AZTO and 15%
azoTO by NMR): Found: C, 22.3; H, 2.1; N, 52.3%; Calc for
(C4H4N8O2.85Á0.5H2O): C, 22.3; H, 2.3; N, 52.0%. IR spectrum:
Electrosynthesis of 1,2,4-triazol-3-hydroxylammonium sulfate.
The acidic filtrate from electrolysis of NTr (above) was evapo-
rated to near dryness, producing a pale yellow crystalline solid,
which was collected by filtration and washed with a little cold
water. The crude product (1.03 g, 36%) was recrystallised
several times from water-ethanol, washed with ethanol and
dried at 80 1C overnight. Mp 170–172 1C dec. Microanalysis:
Found: C, 16.3; H, 3.7; N, 38.6%. Calc for (C2H5N4O)2SO4: C,
16.1; H, 3.4; N, 37.6%. IR spectrum: nmax/cmÀ1 3242, 3127,
2872, 2690,1662, 1510, 1369, 1335, 1256, 1161, 1111, 1086, 1065,
980, 945, 878, 830, 768, 716, 698, 677, 636, 611, 548, 507. dH
(400 MHz, DMSO-d6): 8.03 (1H, s, CH). dC (400 MHz,
DMSO-d6): 157.85, 144.39. Mass spectrum (ESI-HR) MH+
m/z 101.046.
n
max/cmÀ1 3050, 2805, 1665, 1575, 1536, 1466, 1438, 1361, 1314,
1258, 1200, 1132, 1122, 1028, 1013, 997, 808, 785, 735, 693, 664.
dH (400 MHz, DMSO-d6): 12.08 (1H, br, NH), 12.44 (1H, br,
NH), 12.48 (1H, br, NH), 13.2 (1H, vbr, NH). Small peak at
12.74 ppm corresponds to azoTO (see below). dC (400 MHz,
DMSO-d6): 144.41, 145.96, 153.67, 154.32. Smaller peaks at
154.38 and 154.25 ppm correspond to azoTO (see below).
Electrosynthesis of hydrazo-1,2,4-triazol-3-one. AZTO (0.5 g)
was suspended in 0.1 mol LÀ1 Na2SO4 (200 mL), and concen-
trated NaOH(aq) was added to pH8, when most substrate had
dissolved. The solution was electrolysed at À1.2 V for 8 h. The
pH of the solution increased steadily, and was periodically
adjusted to EpH8 by dropwise addition of conc. H2SO4. The
final orange-yellow solution was acidified to pH2 by addition of
conc. H2SO4, giving a thick yellow precipitate (60–70% yield).
Typically, this solid product comprised hydrazotriazolone plus
azoTO (azoTO present at 10–30% by NMR), but a high-purity
sample of hydrazotriazolone was obtained in one instance,
allowing full characterisation of this species. Mp 180 1C dec.
Microanalysis: Found: C, 21.4; H, 3.05; N, 50.55%; Calc for
C4H6N8O2:1.25H2O: C, 21.8; H, 3.9; N, 50.8%. IR spectrum:
Synthesis of 1,2,4-triazol-3-one-5-amine (ATO). This material
was initially prepared by chemical reduction of NTO, via an
adaptation of the method described by Le Campion et al.37
and was also obtained via reduction of AZTO using a similar
procedure. Pd/C was added to either (a) a solution of NTO in
water/methanol (1.5 : 1), or (b) a suspension of AZTO in
water/methanol (1.6 : 1; the solid gradually dissolved). The
vessel was then placed under an atmosphere of H2 gas
(1 atm) and the mixture shaken for several hours. The final
mixture was filtered through celite and the solvent evaporated.
The crude product was recrystallised from ethanol-water.
The product of NTO reduction was yellow, as described in
reference 37, and the 1H NMR spectrum (not previously
reported) showed that it contained small amounts of azoTO
and hydrazotriazolone, in addition to the major peaks (listed
below). The product of AZTO reduction was white, and no
impurities were evident by 1H NMR spectroscopy. The IR
spectra of the two batches were identical. The melting point
for the white solid was higher than reported in reference
37 (240–245 1C)37 but agreed well with the value given in an
earlier report of the material, prepared via a different method.38
Characterisation data are given for the white solid ATO. Yield
(from AZTO) = 47%. Mp 285 1C (dec), (lit.38 286–290 1C
(dec). Microanalysis: Found: C, 23.4, H, 4.2, N, 56.6%. Calc
n
max/cmÀ1 3200, 3094, 3030, 2864, 1734, 1618, 1466, 1443,
1292, 1128, 1053, 1024, 793, 712, 664, 583, 509. dH (400 MHz,
DMSO-d6): 7.97 (2H, br, NH–NH), 10.40 (2H, br, NH), 10.87
(2H, br, NH). dC (400 MHz, DMSO-d6): 149.47, 155.21. Mass
spectrum (ESI): MH+ m/z 199.1.
Electrosynthesis of azo-1,2,4-triazol-3-one (azoTO). The
above procedure for electrosynthesis of hydrazotriazolone
was followed. The final yellow solution was heated to 80 1C
for 5 min, allowed to cool, and acidified to pH2 with conc.
H2SO4. This produced azoTO as a yellow precipitate in 25%
yield. Mp > 300 1C. Microanalysis: Found: C, 24.3; H, 2.05;
N, 57.1%. Calc for C4H4N8O2: C, 24.5; H, 2.1; N, 57.1%. IR
spectrum: nmax/cmÀ1 3040, 2955, 2831, 1670, 1541, 1470, 1428,
1293, 1119, 1041, 1016, 809, 743, 699, 670. dH (400 MHz,
DMSO-d6): 12.44 (br, 2H, NH),12.74 (br, 2H, NH). dC
(400 MHz, DMSO-d6): 154.28,154.40.
for C2H4N4O: C, 24.0, H, 4.0, N, 56.0%. IR spectrum: nmax
/
cmÀ1 3397, 3188, 2914, 2826, 2671, 1694, 1647, 1522, 1420,
1354, 1302, 1148, 1040, 1017, 853, 800, 750, 702, 675, 573. dH
(400 MHz, DMSO-d6) 5.25 (2H, br, NH2), 10.04 (1H, br, NH),
10.27 (1H, br, NH). dC (400 MHz, DMSO-d6) 147.9, 155.2
(lit.37 147.8, 155.1). Mass spectrum: M+ m/z = 100.0.
Electrosynthesis of azoxy-1,2,4-triazole/azo-1,2,4-triazole.
A pale yellowish-white precipitate was formed during electrolysis
of NTr. Yields were 3–20%, depending on conditions (see text).
Precipitate comprised a mixture of azotriazole and azoxytriazole
and very poor solubility prevented separation of these. Mp >
300 1C. Microanalysis on sample comprising 60 : 40 azo : azoxy:
Found: C, 24.9, H, 2.9, N, 60.3%. Calc for C4H4N8O0.4ÁH2O: C,
Acknowledgements
The Microanalytical and Mass Spectroscopy units of the
Research School of Chemistry at the Australian National
University. DSTO for provision of NTO solutions. The Royal
c
2900 New J. Chem., 2011, 35, 2894–2901
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Australian Navy for support for SBLT C.J. Underwood. This
work was funded under the UNSW@ADFA Defence-Related
Research Funding Scheme.
19 M. A. Hiskey, N. Goldman and J. R. Stine, J. Energ. Mater., 1998,
16, 119–127.
20 B. A. Frontana-Uribe, R. D. Little, J. G. Ibanez, A. Palma and
R. Vasquez-Medrano, Green Chem., 2010, 12, 2099–2119.
21 J. Paul Chen, S.-Y. Chang and Y.-T. Hung, in Handbook of
Environmental Engineering, ed. L. K. Wang, Y. Hung and N. K.
Shammas, Humana Press, 2005, vol. 3, pp. 359–378.
22 K. Scott, in Handbook of Green Science and Technology, ed. J. H.
Clark and D. J. Macquarrie, Wiley-Blackwell, 2002, pp. 433–465.
23 M. Cronin, A. Day and L. Wallace, J. Hazard. Mater., 2007, 149,
527–531.
24 M. W. Smith and M. D. Cliff, NTO-based explosives formulations:
A technology review, DSTO technical report, DSTO-TR-0796,
1999, http://www.dsto.defence.gov.au/corporate/reports/DSTO-
TR-0796.pdf, accessed 7 April 2011.
References
1 M. Talawar, R. Sivabalan, H. Muthurajan, A. K. Sikder,
B. R. Gandhe and A. Subhananda Rao, J. Hazard. Mater.,
2009, 161, 589–607.
2 G. Steinhauser and T. M. Klapotke, Angew. Chem., Int. Ed., 2008,
47, 3330–3347.
¨
3 R. K. Pandey, S. N. Asthana, B. Bhattacharya, I. Tiwari and
V. S. Ghole, J. Environ. Sci. Eng., 2007, 49, 195–202.
4 J. Giles, Nature, 2004, 427, 580.
25 L. Wallace, M. P. Cronin, A. I. Day and D. P. Buck, Environ. Sci.
Technol., 2009, 43, 1993–1998.
26 C. J. Underwood, C. Wall, A. I. Day, A. Provatas and L. Wallace,
Proceedings of PARARI 2009 (8th Australian Explosives
Ordnance Conference), 2009.
27 Explosive Hazard Data Sheet, Number 182, Defence Science and
Technology Organisation, Australia. NTO Fine Grade, 2002.
28 H. Lund, in Organic Electrochemistry, ed. M.M. Baizer, Marcel
Dekker Inc. NY, 1973, pp. 315–345.
29 A. Tallec, R. Hazard, J. Suwinski and P. Wagner, Pol. J. Chem.,
2000, 74, 1177–1183.
5 NATO Munitions Safety Information and Analysis Centre
(MSIAC), http://www.msiac.nato.int, accessed 27 June 2011.
6 H. Qian, D. Liu and C. Lv, Lett. Org. Chem., 2011, 8, 184–187.
7 D. M. Badgujar, M. B. Talawar, A. N. Asthana and
P. P. Mahulikar, J. Sci. Ind. Res., 2007, 66, 250–251.
8 R. W. Miller and S. P. Philbin, Tetrahedron, 1997, 53, 4371–4386.
9 K. Ayoub, E. D. van Hullebusch, M. Cassir and A. Bermond,
J. Hazard. Mater., 2010, 178, 10–28.
10 J. C. Pennington and J. M. Brannon, Thermochim. Acta, 2002, 384,
163–172.
11 W. G. Palmer, M. J. Small, J. C. Dacre and J. C. Eaton, in Organic
Energetic Compounds, ed. P.L. Marinkas, Nova Science Publishers
Inc., 1996, pp. 289–372.
30 G. Kokkinidis, K. Hasiotis and D. Sazou, Electrochim. Acta, 1990,
35, 1957–1964.
31 E. S. Rhee and H. J. Shine, J. Am. Chem. Soc., 1986, 108,
1000–1006.
32 P. N. Gupta and A. Raina, Bull. Electrochem., 1988, 4, 989–992.
33 T. M. Florence, J. Electroanal. Chem. Interfac. Electrochem., 1974,
52, 115–132.
34 T. P. Kofman, Russ. J. Org. Chem., 2001, 37, 1158–1168.
35 A. I. Vokin, L. V. Sherstyannikova, L. V. Kanitskaya,
K. A. Abzaeva, V. A. Lopyrev and V. K. Turchaninov, Russ. J.
Gen. Chem., 2001, 71, 1708–1716.
12 D. M. Badgujar, M. B. Talawar, S. N. Asthana and
P. P. Mahulikar, J. Hazard. Mater., 2008, 151, 289–305.
13 M. H. V. Huynh, M. A. Hiskey, T. J. Meyer and M. Wetzler, Proc.
Natl. Acad. Sci. U. S. A., 2006, 103, 5409–5412.
14 J. C. Oxley, J. L. Smith and J. S. Moran, J. Energ. Mater., 2009,
27, 63–93.
15 R. Sivabalan, M. Anniyappan, S. J. Pawar, M. B. Talawar,
G. M. Gore, S. Venugopalan and B. R. Gandhe, J. Hazard.
Mater., 2006, 137, 672–680.
36 A. Chowdhury and S. T. Thynell, Propellants, Explosives, Pyrotechnics,
2010, 35, 572–581.
37 L. Le Campion, S. de Suzzoni-Dezard, N. Robic, A. Vandais,
P. Varenne, J. P. Noel and J. Ouazzani, J. Labelled Compd.
Radiopharm., 1999, 42, 1203–1213.
16 M. B. Talawar, R. Sivabalan, N. Senthilkumar, G. Prabhu and
S. N. Asthana, J. Hazard. Mater., 2004, 113, 11–25.
17 D. E. Chavez, M. A. Hiskey and R. D. Gilardi, Angew. Chem., Int.
Ed., 2000, 39, 1791–1793.
18 D. Chavez, L. Hill, M. Hiskey and S. Kinkead, J. Energ. Mater.,
2000, 18, 219–236.
38 L. E. A. Godfrey and F. Kurzer, J. Chem. Soc., 1960, 3437–3444.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2894–2901 2901