Novel illustrations of the specific reactivity of 1,1-diamino-2,2-dinitroethene (DADNE) leading to new unexpected compounds

Article abstract of DOI:10.1016/j.tet.2006.11.031

1,1-Diamino-2,2-dinitroethene (DADNE) was further investigated and evaluated in oxidation and azidation reactions. DADNE gave new unexpected products as a result of its specific reactivity as previously observed. The X-ray structure determination was the key of success in this work enabling us to perfectly characterise the new products and argue about the reaction mechanisms as well. Once again, the nucleophilic gem-dinitrocarbon of DADNE seemed to be directly involved in these transformations. Attempts to change the regioselectivity were performed by modifying the experimental conditions.

Full text of DOI:10.1016/j.tet.2006.11.031

Tetrahedron 63 (2007) 953–959
Novel illustrations of the specific reactivity of 1,1-diamino-2,2-
dinitroethene (DADNE) leading to new unexpected compounds
*
ꢀ
ꢀ
Gregoire Herve and Guy Jacob
´
Laboratoire de Nouvelles Molecules, SNPE Materiaux Energetiques, SNPE Group, Centre de recherches du Bouchet,
´
´
9, rue Lavoisier, 91710 Vert-le-Petit, France
Received 28 September 2006; revised 2 November 2006; accepted 10 November 2006
Available online 8 December 2006
Abstract—1,1-Diamino-2,2-dinitroethene (DADNE) was further investigated and evaluated in oxidation and azidation reactions. DADNE
gave new unexpected products as a result of its specific reactivity as previously observed. The X-ray structure determination was the key
of success in this work enabling us to perfectly characterise the new products and argue about the reaction mechanisms as well. Once again,
the nucleophilic gem-dinitrocarbon of DADNE seemed to be directly involved in these transformations. Attempts to change the regioselec-
tivity were performed by modifying the experimental conditions.
Ó 2006 Elsevier Ltd. All rights reserved.
H
N
O
1. Introduction
NO2
NO2
H2N
H2N
NO2
E
N
E+
+
CH CO
NO2
NO2
3
E
NO2
Munition vulnerability is an important strategic issue. In the
past several serious accidents and chain explosions onboard
of military ships were reported. Consequently, huge efforts
were undertaken to enhance the vulnerability standards of
weapons. Energetic material researches have been focused
on the improvement of material and molecule sensitivity.
5-Oxo-3-nitro-1,2,4-triazole (NTO) and 1,3,5-triamino-
2,4,6-trinitrobenzene (TATB) were found to be sufficiently
insensitive to fulfil the new Armed Forces needs (i.e., French
MURAT concept). Chemists currently design and optimise
energetic molecule structures in order to find the best com-
promise between insensitiveness and performances. For in-
stance, nitro and azide are known explosophore functions
enhancing the potential energy of energetic molecules,
whereas hydrogen bonding induced by polarised NH groups
stabilise the chemical structure decreasing the sensitivity.1
More recently, 1,1-diamino-2,2-dinitroethene DADNE 1
(also called FOX-7) was prepared as a new efficient insensi-
tive molecule.2 Our work of interest consisted in modifying
and optimising the DADNE structure in order to enhance its
performances in respect of its insensitiveness.
H2N
H2N
1
E = Br, Cl, NO2
Scheme 1. The reactivity of DADNE 1.
Most studied electrophiles react rapidly with DADNE 1 at
room temperature to give the di-substituted derivatives. As
a nucleophile, the gem-dinitrocarbon can react with halogen-
ium and nitronium electrophiles. Both reactions are then
followed by a fast attack on the amidino moiety. Another
DADNE reactivity was also observed since one amino group
was directly attacked by a potent electrophilic complex in-
duced under vigorous conditions (acylation catalysed by
a Lewis acid in refluxing acetyl chloride).3 These experimen-
tal results indicate that the real DADNE structure 2 is far
from the assumption of enamine 1 as commonly represented
in the literature, but rather closer to a resonance structure
such as compound 2 in accordance with the crystallographic
data6 (see Scheme 2).
H
N+
NO2
-
NO2
H2N
H2N
NO2
NO2
H
We firstly started examining the chemical behaviour of
DADNE3 extending the previous works.4,5 Two chemical
behaviours of DADNE were indeed determined depending
on the nature of electrophiles involved. The DADNE reactiv-
ity can be illustrated according to the following Scheme 1.
H2N
1
2
Scheme 2. DADNE and one of its more ‘real’ structural representations.
For convenience, DADNE is still represented in the paper by
the unpolarised structure 1.
Keywords: Azidation; Peroxide oxidation; Tetrazole; Electrophilic reac-
tions; Nitro compounds.
Further works on DADNE deal with oxidation and azidation
reactions leading to new unexpected products. They are all
* Corresponding author. Tel.: +33 1 64 99 11 81; fax: +33 1 64 99 11 75;
e-mail: g.herve@snpe.com
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.11.031

954
G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
characterised by X-ray analysis. The mechanisms are also
discussed taking into consideration the established reactivity
of DADNE.3
2. Results and discussion
2.1. Oxidation of DADNE with peroxide reagents
Several works report efficient oxidation methods of deacti-
vated primary amines producing a mixture of nitro, azo
and azoxy products in the furazane series (Scheme 3).7a–d
Selectivity depends on the nature of the substrate, the oxidant
strength and on the employed experimental conditions.8,9
Sheremetev suggested a correlation between the ionisation
potentials of deactivated primary amines and the oxidative
capacity of the mixtures so that a wide variety of nitrofura-
zanes, bearing both electron-withdrawing and electron-
donating groups, have been prepared in this way.9 In order
to reach an azoxy derivative of DADNE, we chose to carry
out the DADNE oxidation in a mixture of diluted hydrogen
peroxide (30% in H2O2) and concentrated sulfuric acid
(Scheme 4) as previously described in the furazane series.7a
Figure 1. X-ray crystal structure of amidinoformic acid 5.
between both the nitrogens. They have almost a double
˚
˚
bond character (1.31 A instead of 1.25 A for an exact double
˚
bond). The C–C bond is an exact single bond (1.53 A) and
˚
the carboxylate C]O bond has a standard value (1.25 A).
The zwitterionic structure is a result of the acid–base
amidinoformic acid properties as the amidino moiety is
sufficiently basic to abstract the acidic proton. The IR
spectrum displayed two characteristic peaks. The amidinium
moiety (CNH+2) absorbs at 1715 cmꢀ1, which is very close to
the frequency of formamidinium chloride (1717 cmꢀ1). The
ionised carboxyl group absorbs at 1645 cmꢀ1 and fairly
flanks standard values reported in the amino-acid series.13
This value is also close to the strong vibrations of sodium
formate carbonyl (1611 and 1632 cmꢀ1).
Awhite insoluble product rapidly precipitated in the mixture.
Analyses were difficult to undertake since the product had
averylow solubility in all solvents including water. However,
we succeeded in preparing suitable crystals for X-ray analy-
sis. Surprisingly, unexpected amidinoformic acid 5 was iden-
tified as the oxidised product. This compound was firstly
synthesised by Yoshimura et al. from potassium thiooxami-
date and ammonia.10 Wieland and Seeliger then suggested
other synthetic ways to access amidinoformic acid from
dithioacetic acid sulfide and from hydroxyacetonitrile.11 Fi-
nally, Nii et al. reported a new synthesis of amidinoformic
acids using benzyl cyanoformates.12 In spite of many reports,
no full analytical characterisation of amidinoformic acid was
performed. No X-ray and full IR data were reported.10–12 Fig-
ure 1 displays the X-ray structure of amidinoformic acid 5.
We obviously assumed that the nucleophilic gem-dinitro-
carbon was attacked instead of amino moieties as initially ex-
pected. We therefore attempted to modify the regioselectivity
by treating DADNE 1 with a non-aqueous and highly-con-
centrated pertrifluoroacetic acid solution. However, despite
using more severe oxidation conditions, no change was ob-
served, amidino acid 5 being isolated again.
Such an unexpected result prompted us to investigate the
oxidation mechanism. The literature regarding the con-
version of a nitrocarbon into a carboxyl derivative is
As an obvious zwitterionic molecule, the two C–NH2 bonds
˚
(1.31 A) are similar due to a full p-electron delocalisation
–
N+
O
H2N
NH2
NH2
N
H2N
NO2
N
N
H2N
H2N
NH2
H2O2
N
N
+
+
H2SO4
O
N
N
N
N
N
N
N
N
N
N
O
O
O
O
O
Scheme 3. Oxidation of diaminofurazane in hydrogen peroxide.
–
N+
O
O2N
O2N
NH2
N
–
N+
O
NO2
NO2
O2N
N
NO2
NO2
N
N+
O2N
H2N
H2N
NO2
NO2
30% H2O2
H SO
–
H2N
O
2
4
H
N+
O
1
–
H
H2N
O
5
Scheme 4. Oxidation of DADNE in diluted hydrogen peroxide and concentrated sulfuric acid.

G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
955
well-documented.14 The well-known Nef reaction consists in
an acid-catalysed hydrolysis of nitro compounds to give the
corresponding carbonyl derivatives. The mechanism was
shown to proceed via an addition of H2O onto the protonated
nitro group leading to the coloured a-hydroxynitroso inter-
mediate. Since the reaction occurred in a non-aqueous pertri-
fluoroacetic acid solution (see Scheme 5) and assuming that
the gem-dinitrocarbon was nucleophilic,3 a transfer of hy-
droxylium ‘HO+’ onto the gem-dinitrocarbon was a more
likely and valuable explanation. The primary amine oxida-
tion by peroxide reagents is a well-known process giving
mixtures of nitro, azo and azoxy products.7a,15 Several hy-
droxylium transfers are required to reach these compounds.7a
In this reaction, the involved hydroxylium electrophile was
suspected to attack the nucleophilic gem-dinitrocarbon of
DADNE 1 (as in nitrations and halogenations).3 Therefore,
we suggested the following mechanism (Scheme 6).
cycloaddition on the carbon–carbon bond was observed.
Surprisingly, the reaction of 1 and trimethylsilylazide
(TMSN3) gave an unexpected white product. However,
once again no cycloaddition occurred. IR spectrum dis-
played NH or/and NH2 absorption bands, whereas 14N
NMR analysis revealed no evidence of nitro groups. The
product formula (C2N6H4) was given thanks to elementary
analyses and mass spectrometry. Being an insoluble product,
the full characterisation could not be completed by X-ray de-
termination. The new structure 6 was only perfectly assigned
after an efficient derivatisation work (Scheme 7).
X-ray determinations of derivatives 7 and 8a, obtained after
alkaline hydrolysis and subsequent methylation, provided an
unambiguous proof of structure 6. Compounds 8a and 8b as
a mixture of regioisomers (4:1, mol:mol) were separated
from the crude mixture by selective crystallisation. The re-
gioselectivity of the alkylation depends on the substituent
on C(5), and also on the exact form (neutral or ionised) in
which the substrate reacts with the alkylating agent.17 The
derivatives 7 and 8a have standard N–N tetrazole bonds
H
H
N+
O
O
–
H2N
H2N
NO2
NO2
CF3CO3
H
˚
ranging from 1.31 to 1.35 A (Fig. 2). Both structures also
have standard C]O and C(O)–NH2 bonds, which respec-
H2N
1
5
˚
tively measure 1.23 and 1.32 A close to usual standard
values. Both ionic and neutral structures have the same
average N–N cyclic bond lengths. No significant structural
Scheme 5. Oxidation of DADNE in concentrated pertrifluoroacetic acid
mixture.
-
NO2
-
H
H
NO2
H
H
H2N
H2N
NO2
NO2
N+
O
O
H
N+
O
O
H
N
–
H
N+
O
-H+
-ROH
NO2
NO2
H
-HNO2
H
HO O+
-N2O3
+
H2N
NO2
H2N
R
ONO
H2N
H2N
1
R = H or CF3CO
3
4
4bis
5
Scheme 6. Suggested oxidation mechanism of DADNE. The reaction is presented as an acid-catalysed electrophilic reaction.
1-Amidino-2-hydroxy-2,2-dinitroethane 3 can be seen as the
first intermediate. A subsequent HNO2 elimination followed
by a nitro–nitrite rearrangement afforded amidinoformic
acid 5. Nitrite nucleophilic substitution is known to produce
hydroxyl derivatives as the ambident nitrite ion can react ac-
cording to both N- and O-attack.16a The O-attack leads to the
nitrite ester, which breaks downvery rapidly to the final prod-
uct delivering dinitrogen trioxide (N2O3).16a Broxton re-
ported that in the deficient aromatic series the O-attack was
20 times greater than the N-attack. Many nitrite O-attacks
giving hydroxyl products are exemplified in the litera-
ture.16b–e So we supposed that equilibrium between 4 and
4bis did exist due to the presence of the ambident nitrite
counter-ion to further produce amidino acid 5 by delivering
N2O3.
differences between the two tetrazole cycles are observed.
Nevertheless, we can notice a slight longer N(4)–C(5)
bond in tetrazolate 7 compared to tetrazole 8a due to a de-
localisation of the anionic charge.
As a confirmation of the structure 6, a disulfilimine by-
product 9 slowly crystallised from the crude reaction mix-
ture (see X-ray structure in Fig. 3). This compound was
formed by condensation of 6 with two molecules of dime-
thylsulfoxide (used as a solvent).
At a first glance, we assumed after a rough evaluation of the
X-ray data that the disulfilimine by-product might have a
simple and symmetrical structure as represented in Figure 4.
Yet experimental X-ray observations (i.e., bond lengths)
could not match the structure 9a.
2.2. The particular reaction of DADNE and TMSN3
˚
The exocyclic C(5)–C(6) bond is single (1.49 A) and the het-
The reactivity of DADNE 1 was also previously evaluated
towards dipolar reactants in presence of unsubstituted and
substituted azides,3 but all attempts failed and no 1,3-
erocycle is clearly a usual tetrazole. All cyclic tetrazole
bonds are fairly similar to those of tetrazoles 7 and 8a as
described above (see Fig. 2). The N(1)–N(2), N(3)–N(4),
O
N
H
O
N
N
N
O
N
N
NH2
NH2
O2N
O2N
N
N
N
N
N
N
MeI
N
N
TMSN3
DMSO
KOH
H2O
+
DMSO
N
N
H
N
-
NH2
NH2
NH2
NH2
Me
K+
Me
1
6
7
8a
8b
Scheme 7. Derivatisation work for complete identification of tetrazole 6.

956
G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
N(1)
C(8)
0
N(2)
N(1)
0
C(6)
N(2)
N(3)
C(5)
N(3)
C(6)
C(5)
N(4)
N(7)
N(4)
N(7)
Figure 2. X-ray structures of tetrazoles 7 and 8a.
N
S+
S
N
N
N
S
N
N
N
N
S+
N
N
-
N
N
-
Scheme 8. Representation of disulfilimine 9.
observed with other electrophiles.3 But to the best of our
knowledge no electrophilic behaviour of TMSN3 has been
reported before in the literature. The only example
described by Olah consists in an activated system involving
trimethylsilylazide and a super acid to induce a potent
aminating [NH2N2]+ reagent.18 Our gentle experimental
conditions are definitely far from those described by Olah.
A deeper bibliographic investigation on azides let us to
consider trimethylsilylazide as an ambivalent reagent as
nitrogen atoms have both electron-donating and electron-
withdrawing properties.19 These linear compounds can be
represented by two mesomeric structures. Resonance of I
and II, with equal contributions, leads to bond order 1.5
and 2.5 for Na–Nb and Nb–Nc, respectively (Scheme 9).
Figure 3. A disulfilimine by-product characterised by X-ray analysis.
N
N+ N–
N– N+
N
N
N
S
S
N
N
N
N
R
R
I
II
Scheme 9. Resonance structures of substituted azides.
9a
According to Scheme 9 resonance II can be seen as both an
acceptor and an electron-donator reagent. Since the amino
moiety of DADNE was not affected, we assumed a likely
interaction between the DADNE gem-dinitrocarbon and
TMSN3 (see Scheme 10). Thus, the remaining negative
charge located on the azide function could lead to a rapid
elimination of NOꢀ2 . The final expected product formed
in this process could be, after several eliminations,
Figure 4. Symmetrical disulfilimine structure 9a as an unsuitable
representation.
N(1)–C(5) and N(4)–C(5) bonds range from 1.33 to
˚
1.34 A, whereas the shortest ‘double-character’ N(2)–N(3)
˚
bond has a 1.32 A length. It is noteworthy that the two
C–N bonds of the disulfilimine moiety have almost the
˚
same length (1.33–1.34 A) indicating a full electron delocal-
isation between both the nitrogens N(7) and N(8). A full de-
localisation is also allowed onto the two sulfur atoms (both
˚
N–S bonds measure 1.66–1.67 A). Consequently, the real di-
NO2
H
N
H
N
NO2
NO2
H
N
sulfilimine structure can be represented by the two zwitter-
ionic mesomers as represented in Scheme 8.
nitro
elimination
?
CN
N
N
N
H2N
H2N
H2N
N
N
Si
N
Si
This surprising result prompted us to investigate the mecha-
nism of the reaction. We first speculated an interaction be-
tween the nucleophilic gem-dinitrocarbon and TMSN3 as
-
Scheme 10. Amidinoacetonitrile as a possible precursor of product 6?

G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
957
amidinoacetonitrile (Scheme 10). Therefore, we decided to
pursue our experimental works to check this hypothesis by
several means. We first tried to detect amidinoacetonitrile
by following the reaction by IR, but no significant nitrile ab-
sorption band was noticed.20
grades and were purchased from Aldrich. DADNE was pre-
pared as previously described by Latypov et al.2 IR spectra
were recorded on a Nicolet Avatar 320 FTIR instrument in
dry KBr pellets. NMR spectra were recorded with a Bruker
Avance 400 spectrometer fitted with a 10 mm broadband
1
ATM probe. Chemical shifts were referred to TMS for H
No acidic activation of TMSN3 was successfully achieved
and no yield improvement was observed either.21 Anhydrous
runs were carried out to prevent hydrolysis of TMSN3, but
no yield improvement was reached. A longer reaction time
was required to obtain 6.22 A run with a large excess of
hydrazoic acid (HN3) in the presence of DADNE 1 was
also attempted, but no trace of 6 was detected, even after
a one-week reaction. This indicates that HN3 cannot be
solely involved in this transformation.
and 13C and to nitromethane for 15N. Mass spectra (MS)
were recorded by using a Nermag R10-10H mass spectro-
meter by electronic impact (IE, 70 eV), positive chemical
ionisation (CI+, NH3) and negative chemical ionisation
(CIꢀ, NH3). Elementary analyses were performed with an
NA2500 ThermoElectron Corporation apparatus. Sensitiv-
ities to impact (CSI) and to friction (CSF) were determined
using the Julius Peter apparatus. It was shown (see below)
that compounds 5–7 are insensitive. These values can be
compared to 2,4,6-trinitrotoluene (TNT) as a reference com-
pound (CSI¼30 J and CSF¼300 N).
All of these trials are proof that the mechanism is not trivial
and worth being investigated in further depth. More than
solving a particular and interesting problem, there is a funda-
mental reactivity issue around the role and the chemical
behaviour of trimethylsilylazide.
4.2. X-ray crystallography
Compounds 5, 7, 8a and 9 were analysed by J. Marrot at
the University of Versailles (France). Colourless crystals
were glued to a glass fibre. Intensity data were collected at
room temperature with a Siemens SMART diffractometer
equipped with a CCD two-dimensional detector [l Mo
3. Conclusion
˚
These further works on DADNE highlighted once again
the particular reactivity of such a compound. The X-ray
structures unambiguously characterised the novel and unex-
pected compounds obtained from oxidation and azidation
of DADNE. As previously demonstrated,3 we reasonably
suspect for both reactions an obvious contribution of
the DADNE gem-dinitrocarbon in these transformations.
TMSN3 was suspected to behave in a different way as usually
observed. TheDADNE regioselectivity has not been success-
fully modified by varying the experimental conditions, as
already noticed with harder activated electrophiles [Cl]+ in
the presence of DADNE.23
Ka¼0.71073 A].
Slightly more than one hemisphere of data was collected
in 1271 frames with u scans (width of 0.30ꢁ and exposure
time of 30 s per frame (7), 10 s per frame (5), (8a) and
(9)). Data reduction was performed with SAINT software.
Data were corrected for Lorentz and polarisation effects,
and a semi-empirical absorption correction based on
symmetry equivalent reflections was applied by using
the SADABS program.24 Lattice parameters were obtained
from least-squares analysis of all reflections. The structure
was solved by direct method and refined by full matrix
least-squares, based on F2, using the SHELX-TL software
package.25 All non-hydrogen atoms were refined with
anisotropic displacement parameters. All hydrogen atoms
were located with geometrical restraints in the riding
mode.
4. Experimental section
4.1. General
Melting points were determined by using a DSC823 Mettler
Toledo apparatus (8 ꢁC/min). Reagents had commercial
CCDC 294504, 620086, 620087, 620088 contain the supple-
mentary crystallographic data. These data can be obtained
3
Volume (A )
˚
Compd
Crystal
size (mm)
Crystal
system
Space
group
Unit cell
dimensions
Unit cell
dimensions
Z
Density
1.737
ꢁ
˚
(lengths, A)
(angles,
)
5
0,10 0,08 0,02
0,75 0,30 0,10
0,73 0,42 0,24
0,48 0,22 0,08
Triclinic
P-1
a¼3.5490(3),
b¼5.2822(4),
c¼9.4294(6)
a¼80.332(3),
b¼87.856(4),
g¼75.043(3)
168.35(2)
525.31(2)
579.60(5)
1069.17(9)
2
7
Monoclinic
Monoclinic
Monoclinic
P21/c
P21/c
P21/c
a¼3.7277(1),
b¼21.0790(3),
c¼7.2773(2)
a¼90,
4
4
4
1.912
1.457
1.443
b¼113.268(2),
g¼90
8a
9
a¼4.2299(2),
b¼21.4700(5),
c¼6.9942(4)
a¼90,
b¼114.148(2),
g¼90
a¼7.4196(4),
b¼13.5605(6),
c¼10.6268(5)
a¼90,
b¼90.425(1),
g¼90

958
G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
free of charge from the Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
the red residue and a small amount of sodium bisulfite was
added to remove iodine. A white solid precipitated and
was collected by filtration. The solid was washed with water
(twice), dried over P2O5 and further triturated with acetone.
The remaining salts were removed by filtration. Evaporation
of acetone afforded 0.30 g of pure 1-methyltetrazole 8a. The
aqueous filtrate was left for 2 days under ambient air and
another portion of 1-methyltetrazole 8a was filtered and
washed with iced water (0.30 g). Both fractions were com-
bined to afford the pure product 8a as a white solid with
33% yield. Mp 164 ꢁC (lit. 164 ꢁC).26 1H NMR (acetone)
d¼7.13 (br s, 1H), 7.44 (br s, 1H), 4.36 (3H, s) ppm. 13C
NMR (acetone) d¼148.3, 157.8, 36.5 ppm. 15N NMR (ace-
tone) d¼ꢀ280, ꢀ236, ꢀ51, ꢀ4, 7 ppm. IR (KBr) 3348,
3192, 2963, 2880, 2798, 1714, 1624, 1508, 1459, 1401,
1374, 1228, 1195, 952, 722, 704, 598, 526 cmꢀ1. MS (CI+,
NH3): m/z¼128 (100), 145 (68); MS (CIꢀ, NH3): m/z¼126
(100); Elem. analysis calcd (%): C 28.3, H 3.9, N 55.1;
found: C 28.1, H 3.8, N 53.6. The combined filtrates were
further extracted with ethyl acetate (3ꢂ30 mL). The organic
layers were combined, dried over MgSO4 and concentrated
under vacuum. A pale yellow product (0.72 g) was isolated
Caution: DADNE is an insensitive compound, nevertheless
it has a high potential energy and must be handled with
care and proper shielding.
4.3. Amidinoformic acid (5)
DADNE 1 (400 mg, 2.70 mmol) was rapidly added to
a stirred solution containing 30% hydrogen peroxide (8 g)
and concentrated sulfuric acid (4.4 g) at room temperature.
The mixture was kept at room temperature for 18 h and
poured into iced water (60 mL). The precipitate was col-
lected by filtration. It was washed twice with cold water
and dried over P2O5. A white insoluble solid was obtained
1
(80% yield). Mp 275 ꢁC. H NMR (DMSO) d¼9.34 (2s,
1H, 1H) ppm. 13C NMR (DMSO) d¼164.7, 166.3 ppm. IR
(KBr) 3354, 3323, 3025, 1715 (C]N), 1645 (C]O),
1491, 1130, 1106, 885, 853, 826, 757, 671, 538, 466 cmꢀ1
.
MS (CI+, NH3): m/z¼107 (100) [MH+18]+; MS (CIꢀ,
NH3): m/z¼106 (37) [MH+18]ꢀ; Elem. analysis calcd (%):
C 27.3, H 4.5, N 31.8; found: C 26.3, H 4.1, N 30.8.
CSI>50.1 J; CSF>353 N.
1
and analysed by H NMR. A mixture of 1-methyltetrazole
8a and 2-methyltetrazole 8b was thus obtained in a 4:1 molar
ratio.
4.4. 5-Amidino-tetrazole (6)
Acknowledgements
DADNE 1 (28.5 g, 0.193 mol) and trimethylsilylazide
(208.0 g, 1.806 mol) were stirred in DMSO (360 mL) for
3–5 days at ambient temperature. The mixture was filtered
and washed with a small amount of DMSO. The crude
5-amidino-tetrazole 6 was washed with water (twice) and
with acetone. 5-Amidino-tetrazole (4.68 g) was obtained
(22%) as a white solid (DSC: decomposition started from
260 ꢁC, onset: 323 ꢁC). 1H NMR (DMSO) d¼8.96 (br
s) ppm. 13C NMR (DMSO) d¼152.8, 156.7 ppm. IR (KBr)
3223, 3051, 1704, 1579, 1474, 1391, 1370, 1122, 787,
757, 689, 520, 458 cmꢀ1. MS (IE): m/z¼112 (5); MS (CI+,
NH3): m/z¼113 (100), 225 (7); Elem. analysis calcd (%):
C 21.4, H 3.6, N 75.0; found: C 21.0, H 3.8, 71.3.
CSI>50.1 J; CSF>353 N.
We are grateful for the financial support of the French De-
ꢀ ˆ
fence representative DGA. We would like to thank Jerome
Marrot for his precious and very helpful crystallographic
works, and Christian Porter for his linguistic help.
References and notes
1. Pagoria, P. F.; Lee, G. S.; Mitchell, A. R.; Schmidt, R. D.
Thermochim. Acta 2002, 384, 187–204.
2. Latypov, N. V.; Bergman, J.; Langlet, A.; Wellmar, U.; Bemm,
U. Tetrahedron 1998, 54, 11525–11536.
3. Herve, G.; Jacob, G.; Latypov, N. Tetrahedron 2005, 61, 6743–
6748.
ꢀ
4.5. Potassium tetrazole-5-carboxamide salt (7)
4. Sandberg C.; Latypov N. V.; Goede P.; Tryman R.; Bellamy A.
Acid–Base Characteristics of FOX-7 and its Mono-hydrazo
Analogue. Proceedings of the Fifth Seminar ‘New Trends in
Research of Energetic Materials’; Pardubice, Prague, 2002.
5. Bellamy, A. J.; Latypov, N. V.; Goede, P. J. Chem. Res.,
Miniprint 2002, 641–644.
5-Amidino-tetrazole (0.49 g, 4.4 mmol) was heated in re-
fluxing potassium hydroxide solution (0.25 g, 4.5 mmol,
25 mL) until completely dissolved (40 min). The mixture
was evaporated to dryness under vacuum. The residue was
dried over P2O5 and washed with THF. A white solid was
6. Bemm, U.; Ostmark, H. Acta Crystallogr. 1998, C54, 1997–
1999.
1
yielded (84%). H NMR (DMSO) d¼7.13 (br s, 1H), 7.44
(br s, 1H) ppm. 13C NMR (DMSO) d¼157.7, 162.8 ppm.
15N NMR (D2O) d¼ꢀ278 (weak signal), ꢀ68, ꢀ1 ppm;
Elem. analysis calcd (%): C 15.9, H 1.3, N 46.3; found: C
15.1, H 1.5, N 43.2. CSI>50.1 J; CSF>353 N.
7. (a) Solodyuk, G. D.; Boldyrev, M. D.; Gidaspov, B. V.;
Nikolaev, V. D. Zh. Org. Khim. 1981, 17, 861–865; (b)
Gunasekaran, A.; Trudell, M. L.; Boyer, J. H. Heteroat.
Chem. 1994, 5, 441; (c) Gunasekaran, A.; Boyer, J. H.
Heteroat. Chem. 1993, 4, 521; (d) Sheremetev, A. B.;
Novikova, T. S.; Mel’nikova, T. M.; Khmel’nitskii, L. I. Bull.
Acad. Sci. USSR, Div. Chem. Sci. 1990, 39, 1073.
8. Sheremetev, A. B.; Makhova, N. N. Adv. Heterocycl. Chem.
2001, 78, 65.
9. Novikova, T. S.; Melnikova, T. M.; Kulagina, O. V.;
Aleksandrova, N. S.; Sheremetev, A. B.; Pivina, T. S.;
Khmelnitskii, L. I.; Novikov, S. S. Mendeleev Commun. 1994,
138.
4.6. 1-Methyltetrazole-5-carboxamide (8a)
Potassium tetrazole-5-carboxamide salt
7
(1.83 g,
12.2 mmol) was dissolved in 100 mL of DMSO at 75–
80 ꢁC. Iodomethane (5 mL, 11.4 g, 80.3 mmol) was quickly
introduced with a syringe. The mixture became dark (iodine
was formed) and was stirred for 30 min. DMSO was distilled
(bp 60 ꢁC) under vacuum. Water (30 mL) was poured onto

G. Herveꢀ, G. Jacob / Tetrahedron 63 (2007) 953–959
959
10. Yoshimura, J.; Fujimori, K.; Sugiyama, Y.; Ando, H. Bull.
Chem. Soc. Jpn. 1971, 44, 3131–3136.
11. Wieland, T.; Seeliger, A. Liebigs Ann. Chem. 1976, 820–
823.
19. Reiser, A.; Wagner, H. M. The Chemistry of Azido Group;
Patai, S., Ed.; Wiley: London, 1971; pp 4–5.
20. Only few broad weak bands around 2200 cmꢀ1 in the nitrile
characteristic absorption range were observed.
12. Nii, Y.; Okano, K.; Kobayashi, S.; Ohno, M. Tetrahedron 1979,
27, 2517–2520.
21. When the reaction was run for 5 days in anhydrous DMSO with
1 equiv of acetic acid the yield was somewhat lower (11%) and
even much lower when a Lewis acid such as Hf(OTf)4 was used
(4%). The isolated product 6 was identified by IR and com-
pared to an authentic product.
22. The same reaction in freshly distilled DMSO was run for 10
days. After 1–2 days no formation of 6 was observed. The mix-
ture was kept for one additional week to recover 6 with 17%
yield.
13. Lenormand. J. Chim. Phys. 1946, 43, 327.
14. Hawthorne, M. F. J. Am. Chem. Soc. 1957, 79, 2510–2515.
15. (a) Thiel, W. R.; Krohn, K. Chem.—Eur. J. 2002, 8, 1049–1058;
(b) Rozen, S.; Kol, M. J. Org. Chem. 1992, 57, 7342–7344.
16. (a) Broxton, T. J. Aust. J. Chem. 1981, 34, 969–979; (b)
Rosenblatt, D. H.; Dennis, W. H.; Goodin, R. D. J. Am.
Chem. Soc. 1973, 95, 2133–2136; (c) Zilberman, J.; Ioffe,
D.; Gozlan, I. Synthesis 1992, 659–660; (d) Broxton, T. J.;
Muir, D. M.; Parker, A. J. J. Org. Chem. 1975, 40, 2037; (e)
Broxton, T. J.; Muir, D. M.; Parker, A. J. J. Org. Chem. 1975,
40, 3230–3233.
23. In all runs, only 1-chloro-1,1-dinitro-2-(N-chloroamidino)-eth-
ꢀ
ane was recovered. Herve, G.; Jacob, G. Unpublished results.
24. Sheldrick, G. M. SADABS, Program for Scaling and
€
Correction of Area Detector Data; University of Gottingen:
€
17. Koldobskii, G. I.; Ostrovskii, V. A. Russ. Chem. Rev. 1994, 63,
797–814.
Gottingen, Germany, 1997.
25. Blessing, R. H. Acta Crystallogr., Sect. A 1995, 51, 33.
€
18. Olah, G. A.; Ernst, T. D. J. Org. Chem. 1989, 54, 1203–
1204.
26. Lobbecke, S.; Pfeil, A.; Krause, H. H.; Sauer, J.; Holland, U.
Propellants, Explos., Pyrotech. 1999, 24, 168–175.