High Explosives and Propellants by Fordham
High Explosives and Propellants by Fordham
High Explosives and Propellants by Fordham
Editorial Board:
P. Day · X. Duan · T. J. Meyer
G. Parkin · H. W. Roesky · J.-P. Sauvage
Structure and Bonding
Series Editor: D. M. P. Mingos
Recently Published and Forthcoming Volumes
With contributions by
A. J. Bellamy · E. F. C. Byrd · R. D. Chapman · H. Gao
T. M. Klapötke · W. D. Mattson · D. T. Meshri · B. M. Rice
J. M. Shreeve · R. P. Singh · S. Zeman
123
The series Structure and Bonding publishes critical reviews on topics of research concerned with
chemical structure and bonding. The scope of the series spans the entire Periodic Table. It focuses
attention on new and developing areas of modern structural and theoretical chemistry such as na-
nostructures, molecular electronics, designed molecular solids, surfaces, metal clusters and supra-
molecular structures. Physical and spectroscopic techniques used to determine, examine and model
structures fall within the purview of Structure and Bonding to the extent that the focus is on the
scientific results obtained and not on specialist information concerning the techniques themselves.
Issues associated with the development of bonding models and generalizations that illuminate the
reactivity pathways and rates of chemical processes are also relevant.
As a rule, contributions are specially commissioned. The editors and publishers will, however, always
be pleased to receive suggestions and supplementary information. Papers are accepted for Structure
and Bonding in English.
In references Structure and Bonding is abbeviated Struct Bond and is cited as a journal.
This work is subject to copyright. All rights are reserved, whether the whole or part of the material
is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broad-
casting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of
this publication or parts thereof is permitted only under the provisions of the German Copyright Law
of September 9, 1965, in its current version, and permission for use must always be obtained from
Springer. Violations are liable for prosecution under the German Copyright Law.
Springer is a part of Springer Science+Business Media
springer.com
c Springer-Verlag Berlin Heidelberg 2007
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence
of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
Cover design: Design & Production GmbH, Heidelberg
Typesetting and Production: LE-TEX Jelonek, Schmidt & Vöckler GbR, Leipzig
Printed on acid-free paper 02/3180 YL – 5 4 3 2 1 0
Series Editor
Prof. D. Michael P. Mingos
Principal
St. Edmund Hall
Oxford OX1 4AR, UK
[email protected]
Volume Editor
Thomas M. Klapötke
Dept. Chemie und Biochemie
Universität München
Butenandstr. 5–13
81377 München, Germany
[email protected]
Editorial Board
Prof. Peter Day Prof. Gerard Parkin
Director and Fullerian Professor Department of Chemistry (Box 3115)
of Chemistry Columbia University
The Royal Institution of Great Britain 3000 Broadway
21 Albermarle Street New York, New York 10027, USA
London W1X 4BS, UK [email protected]
[email protected]
Prof. Herbert W. Roesky
Prof. Xue Duan
Institut for Anorganic Chemistry
Director University of Göttingen
State Key Laboratory Tammannstr. 4
of Chemical Resource Engineering 37077 Göttingen, Germany
Beijing University of Chemical Technology [email protected]
15 Bei San Huan Dong Lu
Beijing 100029, P.R. China Prof. Jean-Pierre Sauvage
[email protected]
Faculté de Chimie
Prof. Thomas J. Meyer Laboratoires de Chimie
Organo-Minérale
Department of Chemistry Université Louis Pasteur
Campus Box 3290 4, rue Blaise Pascal
Venable and Kenan Laboratories 67070 Strasbourg Cedex, France
The University of North Carolina [email protected]
and Chapel Hill
Chapel Hill, NC 27599-3290, USA
[email protected]
Structure and Bonding
Also Available Electronically
For all customers who have a standing order to Structure and Bonding, we
offer the electronic version via SpringerLink free of charge. Please contact
your librarian who can receive a password or free access to the full articles by
registering at:
springerlink.com
If you do not have a subscription, you can still view the tables of contents of the
volumes and the abstract of each article by going to the SpringerLink Home-
page, clicking on “Browse by Online Libraries”, then “Chemical Sciences”, and
finally choose Structure and Bonding.
Thomas M. Klapötke
Contents
FOX-7 (1,1-Diamino-2,2-dinitroethene)
A. J. Bellamy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Nitrogen-Rich Heterocycles
R. P. Singh · H. Gao · D. T. Meshri · J. M. Shreeve . . . . . . . . . . . . . 35
FOX-7 (1,1-Diamino-2,2-dinitroethene)
Anthony J. Bellamy
Cranfield University, Defence Academy of the United Kingdom, Shrivenham,
Swindon SN6 8LA, UK
a.j.bellamy@cranfield.ac.uk
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 Historical Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
9 Isomers of FOX-7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Abstract In this paper, the discovery and synthesis of the explosive 1,1-diamino-2,2-
dinitroethene (FOX-7) are described, together with an account of its structural, spectro-
scopic, and explosive properties. The chemical reactivity of FOX-7 towards nucleophilic
substitution (transamination), electrophilic substitution, and acid-base properties is ex-
plored, as is its thermal behavior (phase transformations and thermal decomposition).
The molecular structure and physical properties of FOX-7 are compared with those of its
three isomers (as yet unsynthesized), as derived by theoretical calculations. Finally, the
physical properties of FOX-7 are compared to those of various energetic molecules that
are structurally related to FOX-7.
1
Introduction
For more than half a century, the main explosive component of artillery
shells and other explosive devices has been RDX (1,3,5-trinitro-1,3,5-
2 A.J. Bellamy
triazacyclohexane, 1). While RDX was not the most energetic material avail-
able during most of this period, it was relatively easy to produce, in a single
step, from commonly available reagents, namely hexamethylenetetramine
(hexamine), nitric acid, and, in most processes, acetic anhydride. However,
RDX exhibits several unsatisfactory properties when used as a high explo-
sive. It is sensitive to impact, friction, electrostatic discharge and heat, and
it is this sensitivity that has been the cause of many very serious accidents
where ordnance has either suffered direct impact by enemy fire or been sub-
jected to heating in fires. Consequently, during the past 25 years there has
been a search for replacement explosive materials that are at least as en-
ergetic as RDX but at the same time are much less sensitive to external
stimuli. One such potential replacement is the simple molecule 1,1-diamino-
2,2-dinitroethene, also known as FOX-7 (2).
2
Historical Development
nitroimidazole (11), but when 80–100% sulphuric acid was used, the prod-
ucts were parabanic acid and 2-dinitromethylene-4,4-dinitroimidazolidin-
5-one (12). The latter separated from the reaction medium and was fil-
tered off. It was found to be thermally unstable and lost the elements of
N2 O3 to form 2-dinitromethyleneimidazolidine-4,5-dione (13). The dione
was subsequently hydrolyzed with aqueous ammonia to give 1,1-diamino-
2,2-dinitroethene and oxalate (Scheme 2). A number of control experiments
showed that 2-methyl-4-nitroimidazole was not an intermediate in the forma-
tion of 2-dinitromethylene-4,4-dinitroimidazolidin-5-one, but that 2-methyl-
4,5-dihydro(1H)imidazol-4-one (14) probably was.
A more direct (but related) route to 1,1-diamino-2,2-dinitroethene was
also reported in the same paper [2]. This involved nitration of the methanol
adduct of 2-methylimidazolidine-4,5-dione (15, synthesized from acetami-
dine hydrochloride and diethyl oxalate, followed by recrystallization from
methanol). This gave 2-dinitromethyleneimidazolidine-4,5-dione (13, 67%
yield), which, on hydrolysis with aqueous ammonia, gave 1,1-diamino-2,2-
dinitroethene (87%; 37% overall from acetamidine hydrochloride, Scheme 3).
Further development work [3, 4] has shown that nitration of the analogous
4,6-dihydroxy-2-methylpyrimidine [16 ≡ 2-methylpyrimidine-4,6(1H,5H)-
dione], which is commercially available, gives a better overall yield of 1,1-di-
amino-2,2-dinitroethene than 2-methylimidazolidine-4,5-dione (Scheme 4).
In this case, the nitrated intermediate that separates is 2-dinitromethylene-
5,5-dinitrodihydropyrimidine-4,6(1H,5H)-dione (17). To effect hydrolysis to
1,1-diamino-2,2-dinitroethene, the unfiltered slurry may be simply added to
water and left for several hours. The product separates out as it is formed
and is finally filtered off and washed. However, in view of the need to re-
cycle the nitrating acids, it is preferable to filter off the solid intermediate,
leaving a controlled amount of sulfuric acid on the solid, and adding the acid-
damp solid to water for hydrolysis [5]. Using this modification and optimized
conditions, the yield of 1,1-diamino-2,2-dinitroethene is over 90%.
The C-5 fragment of the nitrated intermediate (when 4,6-dihydroxy-2-
methylpyrimidine is the starting material) is released during the hydrolysis as
dinitromethane. If desired, this may be extracted into diethyl ether from the
filtrate and precipitated as its potassium salt, or alternatively it may be left to
FOX-7 (1,1-Diamino-2,2-dinitroethene) 5
3
Structural Properties of FOX-7
Fig. 1 FOX-7
and C – NO2 bonds (1.42 and 1.39 Å) and the essentially planar structure of
the molecule as a whole (average C – C torsion angle 4.8◦ ). Two strong hydro-
gen bonds exist between NH and NO on both sides of the molecule. There
are four molecules per unit cell whose dimensions are: a 6.941 Å, b 6.569 Å,
c 11.315 Å, β 90.55◦ , V 515.9 Å3 , Dx 1.907 Mg m–3 .
The supramolecular arrangement of molecules within the crystal is in the
form of wave-shaped infinite layers (Fig. 2 [10]), akin to that observed for
graphite and triaminotrinitrobenzene (TATB), with intra-layer N – H...O – N
hydrogen bonding and inter-layer van der Waals interactions. The distance
between layers is ∼ 3 Å cf. the layer separation in graphite is ∼ 3.5 Å, while
that in TATB is ∼ 6 Å. The extensive network of hydrogen bonding is probably
the reason why 1,1-diamino-2,2-dinitroethene is only slightly soluble in many
common solvents (acetone, acetic acid, ethyl acetate, nitromethane, acetoni-
trile – all < 0.5 g/100ml at 20 ◦ C), the greatest solubility being exhibited in
N,N-dimethylformamide, N-methylpyrrolidin-2-one and dimethylsulphox-
α –→ β –→◦ γ .
113 – 116 ◦ C 173 C
Three other studies, two of them being chronologically earlier than the above
work [15–18], report somewhat different observations. Here the tempera-
tures given for the first two transitions are markedly lower, and one further
transition is observed. The first phase transition (designated α → β) is here
reported to occur at 90 ◦ C and was only detected by X-ray diffraction; it is not
visible by DSC. Again a small decrease in density (1.885 → 1.803 ≡ increase
in volume) is observed. The second transition (designated β → γ ) occurs
at 115 ◦ C (DSC endotherm at 117 ◦ C), and a further transition (designated
γ → δ) occurs at 155 ◦ C (DSC endotherm around 155 ◦ C). The reverse tran-
sitions (γ → β and β → α) occur on cooling, but while the β → α transition
is instantaneous, the γ → β (and hence α) transition may take days or even
weeks. The second of these studies [16] also demonstrated how particle size
can affect the temperature and the rate at which phase transitions occur. For
example it was reported that the first transition temperature can vary between
∼ 78 ◦ C (for very fine particles) and ∼ 95 ◦ C (for coarse particles), but the
rate (∼ 3 h for the complete transition) is unaffected by particle size, while
the reverse is true for the second transition temperature (116 ◦ C) i.e. the tran-
sition temperature is not affected by particle size, but the rate is (10 h and
14 h for fine and coarse particles respectively). The average enthalpy changes
involved in the second and third transitions (those visible by DSC) were re-
ported as 21.6 and 17.1 J g–1 respectively [17, 18].
Although the FOX-7 samples used in the first two of these three stud-
ies [15, 16] were stated to have been recrystallized from dimethylsulphoxide,
the behavior described is similar to that of unrecrystallized FOX-7. A broad
DSC endotherm at 90–110 ◦ C is the normal observation for unprocessed
FOX-7 (Latypov, personal communication), and this is replaced by an en-
dotherm at 113 ◦ C on recrystallization. The TG analysis reproduced in one of
the studies [16] shows no mass loss i.e. loss of solvent, in the region of 160 ◦ C,
FOX-7 (1,1-Diamino-2,2-dinitroethene) 9
which would have been expected if the material had been recrystallized, again
suggesting that unrecrystallized material might have been used. The origin,
but not the history of the sample used in the more recent study [17, 18] is
reported. It may or may not have been recrystallized at source.
In the absence of more definite information on the pre-treatment of the
samples used in the earlier work [15, 16], one is tempted to conclude that
the differences in the results reported stem from the differences between the
crystallographic nature of unrecrystallized and recrystallized material. Un-
recrystallized material will have separated from a rather different medium
(acidic aqueous medium) to that used in a recrystallized sample, and at a dif-
ferent rate. Its internal structure may therefore be somewhat different. At the
present time (2006) some confusion exists regarding which phase change, if
any, occurs below 100 ◦ C, but the transition observed at around 115 ◦ C in all
studies [11, 12, 14–18] appears to be the α → β change (but designated β → γ
in [15–18]).
4
Thermal Decomposition of FOX-7
was confirmed by HPLC and powder X-ray diffraction. The authors’ sug-
gested explanation for these somewhat surprising results is that the original
sample of FOX-7 may have consisted of two phases, part crystalline (detected
by XRD) and part amorphous (not detected by XRD), and it is the amorph-
ous phase that decomposes during the first DSC peak leaving the original
crystalline part. They point out that the relative amounts of these two phases
should vary with the previous processing conditions. Their explanation is
supported by the observation that recrystallization of the heat treated mate-
rial gave FOX-7 which exhibited the normal phase changes on DTA analysis,
and the double step mass loss at 200–300 ◦ C on TG analysis.
The actual products formed during decomposition in both decomposition
steps have been identified [17, 18] by FTIR as CO2 , HCN, N2 O, NO2 , HOCN,
H2 O and NO, while HNO2 and HCOOH are produced in the second decom-
position only. The apparent activation energies for the two steps are reported
as 238.3 and 322.4 kJ mol–1 , respectively.
FOX-7 (1,1-Diamino-2,2-dinitroethene) 11
5
Spectroscopic Properties of FOX-7
Most of the common spectroscopic data for FOX-7 were reported in the ori-
ginal FOI publication [2]. Nuclear Magnetic Resonance (NMR) spectroscopic
data were obtained in dimethyl sulphoxide (DMSO–d6 ). The 1 H spectrum
(Fig. 4a) shows a single broad peak (at 8.77ppm) for all 4 NH nuclei, while
the 13 C spectrum (Fig. 4b) exhibits 2 peaks at 128.5 [C(NO2 )2 ] and 158.8 ppm
[C(NH2 )2 ] [2], the former being significantly less intense (can be as low as
10–15% relative intensity) than the latter. Carbon atoms with nitro groups
attached normally give low intensity peaks. The relative peak positions re-
flect the strong polarization of the electrons of the C = C bond towards the
carbon with nitro groups attached, causing greater shielding of that carbon
relative to that of the carbon bearing the amino groups. In the 15 N spectrum
absorptions occur at – 24.0 (NO2 ) and – 276.6 ppm (NH2 ) [21].
The infrared spectrum of solid FOX-7 (KBr disc, Fig. 5) exhibits [2] peaks
at 3417 (NH2 ), 3315 (NH2 ), 3200 (NH2 ), 1638 (NH2 ), 1524 (NO2 ), 1471, 1400,
1356 (NO2 ), 1234, 1176, 1143, 1027, 752, 652, 521, 462 cm–1 . The Raman spec-
trum of solid FOX-7 (neat) is shown in Fig. 6 [10].
FOX-7’s ultraviolet spectrum (in water) exhibits three maxima, 210
(ε9500), 278 (ε5000) and 346 nm (ε11 300) (Latypov, unpublished results).
The mass spectrum (70 eV) exhibits a very strong molecular peak at
m/z 148 (M+ ). A more detailed study [16] reported that the intensity of the
parent ion peak indicated that this species was more stable than those en-
countered with other energetic molecules e.g. RDX, TNT. At both 20 and
70 eV there were peaks at m/z 148, 130 and 18, indicative of loss of H2 O, as
also observed with TATB.
6
Explosive Properties of FOX-7
Molecular formula C2 H4 N4 O4 C3 H6 N6 O6
Molecular mass 148 222
Density (g cm–3 ) 1.88 1.82
Enthalpy of formation – 130 [23] + 67 [23]
(kJ mol–1 )
Oxygen balance (to CO2 , %) – 21.6 – 21.6
Decomposition temperature 219–223 [24] 195–199 [24]
(ARC onset, ◦ C)
Detonation velocity 8870 [22] 8930 [22]
(experimental, m s–1 )
Detonation pressure 33.96 [22] 34.63 [22]
(calculated, GPa)
Impact sensitivity
Dropweight test (BAM, 2 kg, cm) 126 [16] 38 [16]
Rotter Impact (F of I) 110–140 [7] 80 [7]
Friction sensitivity
(Julius–Petri, > 35 kp) > 350 [16] 120 [16]
Electrostatic Discharge Test 4.5 J ignition, 4.5 J ignition,
0.45 J no ignition [7] 0.45 J no ignition [7]
FOX-7: 1,1-diamino-2,2-dinitroethene
RDX: 1,3,5-trinitro-1,3,5-triazacyclohexane
and + 67 kJ mol–1, respectively] were selected because they were both from
the same reference [23].
In order to estimate the amount of energy available for release on detona-
tion, we need to apply the Wilson–Kistiakowsky rules [26], which state that
(for an explosive with an oxygen balance not below – 40%; FOX-7 and RDX,
with oxygen balances of – 21.6%, both fall within this group):
(i) carbon atoms are converted to CO
(ii) any remaining oxygen is used to convert hydrogen atoms to H2 O
(iii) any oxygen remaining after (ii) is satisfied is used to convert CO to CO2
(iv) all nitrogen atoms are converted to N2 .
Thus, applying these rules to FOX-7 and RDX, the following ratios of detona-
tion products are predicted:
◦
∆Hdet (RDX) = 3∆Hf◦ (CO) + 3∆Hf◦ (H2 O) – ∆Hf◦ (RDX)
= 3 × (– 110) + 3 × (– 242) – (+ 67)
= – 1123 kJ mol–1 = – 5.06 MJ kg–1 = – 9.21 MJ dm–3 .
An experimental value for ∆Hdet ◦ (FOX-7) is – 4.86 ± 0.06 MJ kg–1 [27], signifi-
cantly larger than the above calculated value. Selection of a less negative value
for ∆Hf◦ (FOX-7) in the above calculation would have resulted in a calcu-
lated ∆Hdet◦ (FOX-7) closer to the experimental value. An experimental value
◦
for ∆Hdet (RDX-wax 94/6), using the same technique, is – 5.28 MJ kg–1 [28].
The calculated value for ∆Hdet ◦ (RDX) would have been closer to the experi-
mental value if a more positive value for ∆Hf◦ (RDX) had been chosen. As
already stated, the values of ∆Hf◦ chosen for the calculations were selected
because they were both from the same reference and were therefore assumed
to be mutually consistent. The experimental values for ∆Hdet ◦ suggest that this
clear that on a mole basis only about half the amount of energy is released
by FOX-7 compared to RDX. However, it is volume that limits the mass of
energetic material that can be contained within a piece of ordnance, and on
the basis of volume (via mass and then density) ∼ 21% less energy is re-
leased by FOX-7 relative to an equal volume of RDX. [Using the experimental
values for ∆Hdet ◦ instead, FOX-7 would release only 5% less energy than an
Other reported testing has included the range of tests that are normally con-
ducted in order to ascertain the suitability of an energetic material for service.
This includes production of crystals of the desired size and shape, large-scale
sensitivity and thermal stability tests, and compatibility with other compo-
nents in an explosive formulation. The overall picture from these tests is that
the future for FOX-7 as a replacement for RDX seems very promising.
7
Chemical Reactions of FOX-7
The location of two electron-donating, amino groups on one carbon, and two
electron-withdrawing, nitro groups on the other carbon should make 1,1-
diamino-2,2-dinitroethene a good example of a push-pull alkene [31], with
strong polarization of the double bond (Scheme 5).
The main consequences of this polarization are (i) the carbon bearing the
amino groups (C-1) should be relatively electron deficient and susceptible to
attack by nucleophiles, (ii) the H atoms should be more acidic than those of
simple amines, and (iii) the amine N atoms should be poor nucleophiles. All
of these aspects have been demonstrated to be present in the chemistry of 1,1-
diamino-2,2-dinitroethene.
Simple nitro-enamines have been reported to undergo transamina-
tion reactions via an addition-elimination mechanism. For example, 1-di-
methylamino-2-nitroethene reacts with amines to release dimethylamine,
accompanied by the formation of a new nitro-enamine (Scheme 6 where
R1 R2 NH = piperidine, morpholine and pyrrolidine) [32, 33].
1,1-Diamino-2,2-dinitroethene, which is a double nitro-enamine, reacts
with primary alkylamines in N-methylpyrrolidin-2-one (NMP) at 90 ◦ C to
give both single and double substitution products [34–36], with loss of am-
monia. 3-Aminopentane reacts quantitatively to give the mono-substituted
FOX-7 (1,1-Diamino-2,2-dinitroethene) 17
(Scheme 11), and thus exhibits another aspect of the reactions of 1,1-
diamino-2,2-dinitroethene.
Indeed, this is a general reaction with common bases like KOH. The
potassium salt of 1,1-diamino-2,2-dinitroethene may be isolated as a white,
crystalline solid on treatment with cold aqueous KOH, and the free 1,1-
diamino-2,2-dinitroethene may be regenerated on acidification. The pKa of
1,1-diamino-2,2-dinitroethene was found [36, 37] to be approximately 10.6,
the conjugate base existing in the pH range 11–13 (Scheme 12). Both 1,1-
diamino-2,2-dinitroethene and its conjugate base may be detected by UV and
13 C NMR spectroscopy.
philic attack by HO– at C-1, and generates the potassium salt of dini-
tromethane (Scheme 14).
The observation by Baum et al. [1] that the ammonium salt of cyano-
dinitromethane (8) and not 1,1-diamino-2,2-dinitroethene is formed when
1,1-diiodo-2,2-dinitroethene is treated with ammonia (Scheme 1), prompted
a study of the behavior of ammonium cyanodinitromethanide (8) and 1,1-
diamino-2,2-dinitroethene in liquid ammonia, in the hope that ammonium
cyanodinitromethanide might isomerize to 1,1-diamino-2,2-dinitroethene
(Scheme 15) (Bellamy et al., unpublished results).
8
Alternative Attempts to Synthesize FOX-7
At the present time (2006) the preferred starting material for the manu-
facture of FOX-7 is 4,6-dihydroxy-2-methylpyrimidine, despite the fact that
for every 10 kg of FOX-7 there is > 5 kg of unwanted dinitromethane pro-
duced as a by-product and two equivalents of nitric acid are wasted. It would
obviously be preferable if these disadvantages could be avoided and con-
sequently the nitration of a number of other, closely related heterocyclic
systems has been studied as potential alternative precursors [40–42]. These
were chosen because they were unable or unlikely to nitrate at position 5. Per-
haps the most obvious candidate to investigate was 6-methyl-1,3,5-triazine-
2,4(1H,3H)-dione (20), since this molecule has nitrogen instead of carbon
between the carbonyl groups.
When 6-methyl-1,3,5-triazine-2,4(1H,3H)-dione (20) was nitrated under
the conditions normally used for the 2-methylpyrimidine derivative (mixed
acid, 25 ◦ C), there was no evidence for nitration even after several days.
When the temperature was raised to 80 ◦ C, nitration did occur, but not to
give the anticipated product. After 20 h at 80 ◦ C, a quantitative yield of 2,4,6-
trihydroxy-1,3,5-triazine (21, cyanuric acid) was obtained, accompanied by
tetranitromethane (Scheme 22). Thus it appears that nitration has occurred
initially as anticipated, but then the nitration has progressed further with
cleavage of the original C – CH3 bond and full nitration of the methyl car-
bon. This cleavage is analogous to that which occurs in the nitration of
2-methylimidazole (9, Scheme 2), where parabanic acid (10) is a major prod-
uct. It is likely that cleavage to parabanic acid also occurs to some extent
in the nitration of 2-methylimidazolidine-4,5-dione (15, Scheme 3) since the
yield of the intermediate 13 is only 67%. It is possible that cleavage does not
occur in the methylpyrimidine case because the tetra-nitrated derivative 17
(Scheme 4) separates from the nitration medium. When the methyltriazine 20
was nitrated with nitric acid alone at 20 ◦ C, cyanuric acid was still the product
isolated.
nitration at the methyl carbon, but in this case the product, 2,4-dimethoxy-
6-trinitromethyl-1,3,5-triazine (23), had not undergone C – C bond cleavage.
Reductive removal of one of the nitro groups (KI/MeOH) did produce the
potassium salt of 2,4-dimethoxy-6-dinitromethyl-1,3,5-triazine (24) but this
could not be hydrolyzed to FOX-7, cyanuric acid again being the final product
(Scheme 23).
9
Isomers of FOX-7
While FOX-7 (2) shows great promise as a secondary explosive, it is not ideal
and it is interesting to compare its properties with other, structurally related
molecules. How does it compare for example with its isomers, trans- (32) and
cis-(33) 1,2-diamino-1,2-dinitroethene and 1,3-dinitro-1,3-diazacyclobutane
(34)? Would they be more or less energetic than FOX-7? Would they have
more desirable physical properties, e.g., higher density, greater thermal sta-
FOX-7 (1,1-Diamino-2,2-dinitroethene) 25
In the gas phase, the heats of formation are – 4, + 4 and + 63 kJ mol–1 for
FOX-7, trans- and cis-diaminodinitroethene, respectively, which, by means of
the calculated heats of sublimation, translates to – 113, – 80 and – 29 kJ mol–1
respectively for the solids. FOX-7 is thus calculated to be the solid of lowest
energy. Conversely, since all three molecules are isomeric, FOX-7 would lib-
erate the least amount of energy on decomposition (detonation) to CO, H2 O
and N2 . The heats of such a reaction are calculated to be – 592, – 630 and
– 680 kJ mol–1 , respectively, roughly half that emitted by RDX (experimental
value – 1134 kJ mol–1). However, on a kJ g–1 basis, which is the more import-
ant parameter for energetic performance, cis-diaminodinitroethene is much
closer to RDX (– 4.59 vs – 5.11 kJ g–1 ).
Regarding the molecular geometries, all of the bonds attached to each car-
bon are essentially coplanar in all three diaminodinitroethenes. The bonds
to the oxygen-bearing nitrogen atoms are also coplanar, but those to the
hydrogen-bearing nitrogen atoms are somewhat pyramidal. For FOX-7 and
the trans-isomer, the molecules are almost planar, but the cis-isomer has one
nitro group and one amino group which are rotated out of the plane in order
to facilitate weak hydrogen bonding and to avoid steric interactions between
neighboring syn nitro groups. FOX-7 has two relatively strong (short, both
1.77 Å) intramolecular hydrogen bonds, the trans-isomer has four weaker
(longer, two each at 1.90 and 2.08 Å) ones, while the cis-isomer also has four
26 A.J. Bellamy
hydrogen bonds but these are very weak (2.38–2.94 Å). The “push-pull” effect
exerted by the electron donating amino groups and the electron withdrawing
nitro groups, is, as expected, strongest in FOX-7 and least in the cis-isomer,
where non-planarity of one nitro group and one amino group reduces the
orbital overlap.
Two further theoretical studies of the isomers of diaminodinitroethene [45,
46] gave similar results to those found by the Politzer group. The study by
Sorescu et al. [45] was particularly detailed and gave good agreement with the
parameters calculated by Politzer et al. for molecules in vacuo. In the case of
FOX-7, the study was also extended to the crystalline state and these results
were in accord with the previously reported crystallographic data [8, 9]. This
agreement adds confidence to the results obtained for the other two isomers.
However, calculation of the lattice parameters for FOX-7 at 25 K intervals over
the temperature range 25–450 K failed to detect the experimentally observed
phase change (α → β) in the region of 373–393 K.
A comparison of the various bond lengths, bond angles and crystallo-
graphic data for FOX-7 as calculated by the studies of Politzer et al. [43]
and Sorescu et al. [45], and those found experimentally [8, 9, 14] is given in
Table 2. The largest error appears to be in the estimation of the length, and
hence strength, of the intramolecular hydrogen bonds. The more recent study
is somewhat better in this respect, but still underestimates the length. How-
ever there is some divergence on the crystallographically determined length
also.
One further theoretical study [30] on FOX-7 alone, in addition to using
CHEETAH to calculate the detonation velocity and detonation pressure for
FOX-7, also reported the experimentally determined performance of both
FOX-7 [velocity of detonation 8870 m s–1 , detonation pressure 34.0 GPa, 91%
performance compared to RDX] and RDX [8930 m s–1 and 35.64 GPa, respec-
tively].
Returning to the third, un-synthesized isomer of FOX-7, 1,3-dinitro-1,3-
diazacyclobutane (34), the four-membered ring analogue of RDX, its mo-
lecular geometry and heat of formation have been calculated by the Politzer
group [47]. The bond lengths were computed to be: C – C 1.466 Å, N – N
1.355 Å, N – O 1.191 Å; the bond angles were: N – C – N 87.3◦ , C – N – C
92.7◦ , C – N – N 119.9◦ , N – N – O 116.4◦ , O – N – O 127.2◦ ; and the dihedral
angles were: N – C – N – N 127.1◦ , C – N – N – O 34.6◦ and 147.7◦ . The heat
of formation at 25 ◦ C was calculated to be + 218 kJ mol–1, 26 kJ mol–1 greater
than that for RDX. 1,3-Dinitro-1,3-diazacyclobutane therefore contains con-
siderably more energy than FOX-7 (– 113 kJ mol–1 ) and its other isomers
(trans – 80 kJ mol–1, cis – 29 kJ mol–1 ). These data enable one to estimate the
heat of reaction to form CO, H2 O and N2 (detonation) as about – 922 kJ mol–1
[≡ – 6.23 kJ g–1 vs – 5.28 kJ g–1 (experimental) for RDX]. Thus, from an en-
ergy release viewpoint 1,3-dinitro-1,3-diazacyclobutane would appear to be
a worthwhile target molecule for synthesis. However it is most probable that
FOX-7 (1,1-Diamino-2,2-dinitroethene) 27
Table 2 Measured and computed bond lengths, bond angles and crystallographic data of
FOX-7
Interatomic
distances (Å)
C2 – N3 1.399 1.389 1.400 1.424 1.410
C2 – N4 1.426 1.417 1.418 1.424 1.423
C1 – C2 1.456 1.460 1.446 1.426 1.465
C1 – N1 1.319 1.309 1.310 1.339 1.331
C1 – N2 1.325 1.320 1.324 1.339 1.336
N3 – O2 1.249 1.243 1.247 1.251 1.271
N3 – O1 1.252 1.254 1.239 1.219 1.262
N4 – O4 1.242 1.237 1.237 1.251 1.262
N4 – O3 1.242 1.238 1.231 1.219 1.264
O2. . .H2 1.97 1.87 1.96 1.77 1.88
O4. . .H3 2.03 2.04 1.98 1.77 1.92
Bond angles (◦ )
N3 – C2 – N4 116.3 116.5 115.9 – 116.7
N3 – C2 – C1 123.9 124.1 123.7 – 123.5
N4 – C2 – C1 119.8 119.4 120.3 – 119.9
N1 – C1 – N2 118.4 118.2 118.2 – 118.2
N1 – C1 – C2 120.8 120.7 121.3 – 121.0
N2 – C1 – C2 120.7 121.1 120.4 – 120.7
O1 – N3 – O2 120.9 120.3 121.0 – 120.7
O2 – N3 – C2 118.9 119.1 118.6 – 119.0
O1 – N3 – C2 120.1 120.5 120.4 – 120.7
O3 – N4 – O4 121.0 121.7 120.7 – 120.6
O4 – N4 – C2 118.6 118.4 118.3 – 118.8
O3 – N4 – C2 120.4 119.9 121.0 – 120.6
Crystallographic data
a (Å) 6.941 6.921 6.940 – 6.978
b (Å) 6.569 6.552 6.637 – 6.776
c (Å) 11.315 11.274 11.341 – 11.336
β (◦ ) 90.55 90.06 90.61 – 90.80
Z 4 4 4 – 4
V (Å3 ) 515.9 511.2 522.3 – 536.0
it would be less thermally and chemically stable than either its other isomers
or RDX. Nevertheless one must never allow tentative theoretical arguments to
persuade one not to attempt to synthesize a promising molecule.
Other properties of 1,3-dinitro-1,3-diazacyclobutane (34) have been pre-
dicted in a series of publications by Zeman et al. [48–52]. By using data
collected for a wide range of energetic nitramines and other polynitro com-
28 A.J. Bellamy
10
Compounds Structurally Related to FOX-7
Table 3 Comparison of the physical properties of FOX-7 and structurally related energetic
molecules
References
1. Baum K, Bigelow SS, Nguyen NV, Archibald TA, Gilardi R, Flippen-Anderson JL (1992)
J Org Chem 57:235
2. Latypov NV, Bergman J, Langlet A, Wellmar U, Bemm U (1998) Tetrahedron 54:11525
3. Latypov NV, Langlet A, Wellmar U (1999) WO Patent 99/03818
4. Latypov NV, Langlet A, Wellmar U, Goede P, Bergman J (2001) Insensitive Munitions
and Energetic Materials Technology Symposium (NDIA) Bordeaux, France, p 620
5. Latypov NV, Johansson M, Holmgren E, Sizova EV, Sizov VV, Bellamy AJ (2007) Org
Proc Res Dev 11:56
6. Astrat’ev AA, Dashko DV, Marshin AY, Stepanov AI, Urazgil’deev NA (2001) Russian
J Org Chem 37:729
7. Lockert IJ (2001) DSTO-TR-1238
8. Bemm U, Östmark H (1997) Acta Cryst C54:1997
9. Gilardi R (1999) Cambridge Crystallographic Data Centre, CCDC 127539
10. Klapötke TM (2006) Ludwig-Maximilian University, Munich, Germany
11. Kempa PB, Herrmann M, Molina Metzger FJ, Thome V, Kjellström A, Latypov N
(2004) International Annual Conference of ICT, Karlsruhe, Germany, 71:1–15
12. Kempa PB, Herrmann M (2005) Part Part Syst Char 22:418
13. Bellamy AJ, Brzoska E (2003) J Ener Mater 21:43
14. Evers J, Klapötke TM, Mayer P, Oehlinger G, Welch J (2006) Inorg Chem 45:4996
15. Bemm U, Eriksson L (2001) Insensitive Munitions and Energetic Materials Technol-
ogy Symposium (NDIA), Bordeaux, France, p 775
16. Östmark H, Bergmann H, Bemm U, Goede P, Holmgren E, Johannson M, Langlet A,
Latypov NV, Pettersson A, Pettersson M-L, Wingborg N (2001) International Annual
Conference of ICT, Karlsruhe, Germany, 26:1–21
17. Burnham AK, Weese RK, Wang R, Kwok QSM, Jones DEG (2005) International An-
nual Conference of ICT, Karlsruhe, Germany, 150:1–12
18. Burnham AK, Weese RK, Wang R, Kwok QSM, Jones DEG (2005) Proc North Am
Therm Anal SOC CONF 33:51
19. Sinditskii VP, Levshenkov AI, Egorshev VY, Serushkin VV (2003) Proceedings of the
30th International Pyrotechnics Seminar, St Malo, France, p 299
32 A.J. Bellamy
20. Garmasheva NV, Chemagina IV, Filin VP, Kazakova MB, Loboiko BG (2004) New
Trends in Research of Energetic Materials. 7th International Seminar. Pardubice,
Czech Republic, p 116
21. Ticmanis U, Kaiser M, Pantel G, Fuhr I, Teipel U (2004) International Annual Confer-
ence of ICT, Karlsruhe, Germany, 70:1–13
22. Janzon B, Bergman H, Eldstäter C, Lamnevik, Östmark H (2002) 20th International
Symposium on Ballistics. Orlando, FL, p 686
23. Matyushin YN, Afanas’ev GT, Lebedev VP, Manov MN, Pepekin VI (2003) Interna-
tional Annual Conference of ICT, Karlsruhe, Germany, 119:1–13
24. Bergman H, Östmark H, Pettersson A, Pettersson M-L, Bemm U, Hihkiö M (1999)
NDIA Conference, Tampa, FL
25. Fraunhofer Institut for Chemische Technologie (2004) ICT Database of Thermochem-
ical Values, Version 7.0
26. Akhavan J (2004) The Chemistry of Explosives, 2nd edn. Royal Soc Chem, p 78
27. Trzcinski WA, Cudzilo S, Chylek Z, Szymanczyk L (2006) International Annual Con-
ference of ICT, Karlsruhe, Germany, 57:1–10
28. Cudzilo S, Trebinski R, Trzcinski WA, Wolanski P (1998) International Annual Con-
ference of ICT, Karlsruhe, Germany, 150:1–8
29. Wild R, Teipel U (2004) International Annual Conference of ICT, Karlsruhe, Germany,
69:1–9
30. Cai H-Q, Shu Y-J, Yu W-F, Zeng G-Y, Cheng B-B (2004) Ener Mater 12:124
31. Rajappa S (1981) Tetrahedron 37:1453, (section 4.3)
32. Marchetti L, Passalacqua V (1967) Ann Chim 57:1266
33. Fetell AI, Feuer H (1978) J Org Chem 43:1238
34. Bellamy AJ, Latypov NV, Goede P (2002) J Chem Res (S)257
35. Bellamy AJ, Latypov NV, Goede P (2002) J Chem Research (M)0641–0661
36. Bellamy AJ, Goede P, Sandberg C, Latypov NV (2002) International Annual Confer-
ence of ICT, Karlsruhe, Germany, 3:1–9
37. Sandberg C, Latypov N, Goede P, Tryman R, Bellamy AJ (2002) New Trends in Re-
search of Energetic Materials. 5th International Seminar. Pardubice, Czech Republic,
p 292
38. Hervé G, Jacob G, Latypov N (2005) Tetrahedron 61:6743
39. Baum K, Nguyen NV (1992) J Org Chem 57:3026
40. Bellamy AJ, Latypov NV, Goede P (2003) J Chem Res (S)529
41. Bellamy AJ, Latypov NV, Goede P (2003) J Chem Res (M)0943–0958
42. Bellamy AJ, Latypov NV, Goede P (2004) New Trends in Research of Energetic Mate-
rials. 7th International Seminar. Pardubice, Czech Republic, p 75
43. Politzer P, Concha MC, Grice ME, Murray JS, Lane P, Habibollazadeh D (1998) J Mol
Struc 452:75
44. Politzer P, Concha MC, Grice ME, Murray JS, Lane P, Habibollazadeh D (1998) Compu-
tational investigation of the structures and relative stabilities of amino/nitro derivatives
of ethylene. J Mol Struc 452:77, reprinted with permission from Elsevier
45. Sorescu DC, Boatz JA, Thompson DL (2001) J Phys Chem A 105:5010
46. Ji G-F, Xiao H-M, Dong H-S, Gong X-D, Li J-S, Wang Z-Y (2001) Acta Chim Sin 59:39
47. Grice ME, Habibollahzadeh D, Politzer P (1994) J Chem Phys 100:4706
48. Zeman S (1997) Themochim Acta 302:11
49. Zeman S (1999) J Ener Mater 17:305
50. Zeman S (1999) Themochim Acta 333:121
51. Zeman S, Krupka M (2003) Propell Explos Pyrotech 28:301
FOX-7 (1,1-Diamino-2,2-dinitroethene) 33
Nitrogen-Rich Heterocycles
Rajendra P. Singh1 · Haixiang Gao2 · Dayal T. Meshri1 ·
Jean’ne M. Shreeve2 (u)
1 Advance Research Chemicals, Inc., 1110 W. Keystone Avenue, Catoosa, OK 74015, USA
2 Department of Chemistry, University of Idaho, Moscow, ID 83844-2343, USA
[email protected]
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2 Triazolium-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.1 1,2,3- and 1,2,4-Triazolium Heterocycles . . . . . . . . . . . . . . . . . . . 37
2.2 Triazolium Heterocycles Containing Amino Substituents . . . . . . . . . . 38
2.3 Triazolium Heterocycles Containing Azido Substituents . . . . . . . . . . . 41
2.4 Triazolium Heterocycles Containing Fluoroalkyl Substituents . . . . . . . . 45
2.5 1,2,4-Triazoliun Azolate Heterocycles . . . . . . . . . . . . . . . . . . . . . 50
3 Tetrazolium-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . 52
3.1 Tetrazolium Heterocycles Containing Amino and Azido Substituents . . . . 52
3.2 Tetrazolium Heterocycles Containing 5,5 -Azotetrazolate Anions . . . . . . 56
3.3 Bistetrazolate Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4 Urotropinium-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . 63
5 Tetrazine-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . 65
6 Azetidinium-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . 70
7 Picrate-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
8 Imidazolium-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . 74
9 Nitrocubane-Based Heterocycles . . . . . . . . . . . . . . . . . . . . . . . . 75
10 Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
1
Introduction
2
Triazolium-Based Heterocycles
2.1
1,2,3- and 1,2,4-Triazolium Heterocycles
Scheme 1
Table 1 Yields, thermal characteristics, and densities of 1,2,4- and 1,2,3-triazolium salts
Tm, melting point, Td, thermal degradation temperature, ∆Hf◦ standard heat of formation
(calculated by Gaussian 03)
showed that significant hydrogen bonding exists between the perchlorate an-
ion and the protonated 1,2,4-triazolium ring, which likely contributes to its
relatively high density. Interestingly, the 1,2,3-triazolium salts are in general
less dense, which suggests less efficient packing in the crystal.
2.2
Triazolium Heterocycles Containing Amino Substituents
Scheme 2
40 R.P. Singh et al.
2.3
Triazolium Heterocycles Containing Azido Substituents
Scheme 3
Scheme 4
44 R.P. Singh et al.
Scheme 5
2.4
Triazolium Heterocycles Containing Fluoroalkyl Substituents
Scheme 6
46 R.P. Singh et al.
Scheme 7
Nitrogen-Rich Heterocycles 47
sis of 24a–n with various metal salts gave excellent yields of salts which
include the bis(trifluoromethanesulfonyl)amide salts of 1-methyl-4-(3,3,3-
trifluoropropyl)-1,2,4-triazolium (25a), 1-butyl-4-(3,3,3-trifluoropropyl)-1,2,
4-triazolium (25b), 1-heptyl-4-(3,3,3-trifluoropropyl)-1,2,4-triazolium (25c),
1-decyl-4-(3,3,3-trifluoropropyl)-1,2,4-triazolium (25d), 1-butyl-4-(1H,1H,2H,
2H-perfluorohexyl)-1,2,4-triazolium (25e), 1-heptyl-4-(1H,1H,2H,2H-perflu-
orohexyl)-1,2,4-triazolium (25f), 1-decyl-4-(1H,1H,2H,2H-perfluorohexyl)-1,
2,4-triazolium (25g), 1-methyl-4-(1H,1H,2H,2H-perfluorooctyl)-1,2,4-triazol-
ium (25h), 1-butyl-4-(1H,1H,2H,2H-perfluorooctyl)-1,2,4-triazolium (25i),
1-heptyl-4-(1H,1H,2H,2H-perfluorooctyl)-1,2,4-triazolium (25j); the hex-
afluorophosphate salts of 1-butyl-4-(1H,1H,2H,2H-perfluorohexyl)-1,2,4-
triazolium (25k), 1-heptyl-4-(1H,1H,2H,2H-perfluorohexyl)-1,2,4-triazolium
(25l), 1-butyl-4-(1H,1H,2H,2H- perfluorooctyl)-1,2,4-triazolium (25m); and
the trifluoromethane-sulfonate salt of 1-butyl-4-(1H,1H,2H,2H-perfluoro-
hexyl)-1,2,4-triazolium (25n) (Scheme 7) [41, 42]. Salts containing longer
alkyl and polyfluoroalkyl substituents have lower melting points because of
the less efficient packing in the solid. The densities were found to be higher
with elongation of the fluoroalkyl substituent and with higher fluorine con-
centration. Cations with bulkier alkyl groups resulted in salts of lower density.
A high-yield, efficient procedure to synthesize functionalized alkyl/fluoro-
alkyl-containing triazolium quaternary salts and triazolium ionic liquids led
to triazolium cations that contained covalently bound anionic sites, such as,
fluorocarboxy, sulfonate, fluorohomoallylic, and fluoroalkanol groups [43,
44]. These were converted by metathetical reactions with fluorine-containing
anions to low-melting salts (26a,b, 27, 28a,b, 29c,d, 30) (Scheme 8) [43]. These
functionalized triazolium salts exhibit many characteristics associated with
excellent ionic liquids with respect to stability in air and water, and when
heated. All are liquids at 25 ◦ C.
A synthetic route to 1-alkyl-1,2,4-triazolium 4-nitroimides was developed
based on alkylation of the metal salts of 4-nitramino-1,2,4-triazole with
halo- and dihaloalkanes (Scheme 9) [44]. These salts are 1-methyl-1,2,4-
triazolum 4-nitroimide (31a), 1-ethyl-1,2,4-triazolum 4-nitroimide (31b),
1-(2-oxapropyl)-1,2,4-triazolium 4-nitroimide (32a), 1-ethoxycarbonylme-
thyl-1,2,4-triazolium 4-nitroimide (32b), 1-methoxymethyl-1,2,4-triazolium
4-nitroimide (33a), 1-(4-fluoro-4,4-dinitro-2-oxabutyl)-1,2,4-triazolium 4-ni-
troimide (33b), 1-(2,3-dihydroxypropyl)-1,2,4-triazolium 4-nitroimide (34),
bis(nitroimido-1,2,4-triazolium-1-yl)ethane (35), 1,3-bis(4-nitroimido-1,2,4-
triazolium-1-yl)-2-oxapropane (36), and 1,3-bis(4-nitroimido-1,2,4-triazol-
ium-1-yl)-3-oxapentane (37).
These salts are solid and were characterized by IR, NMR and elemen-
tal analysis. In the IR spectra the characteristic absorption bands due to
an N-nitroimido group bonded to a heterocycle are observed at 1280–1300
and 1390–1415 cm–1 . The nitramino group that would have shown absorp-
tion bands at 1550–1620 cm–1 is absent. In the 1 H NMR spectra, the signals
48 R.P. Singh et al.
Scheme 8
for the protons of the triazole ring are nonequivalent and are shifted down-
field relative to the signals for the protons in the starting material which
is evidence that supports the imide structure. This conclusion is supported
by the 13 C NMR data. However, no thermal or physical properties were
reported.
Nitrogen-Rich Heterocycles 49
Scheme 9
50 R.P. Singh et al.
2.5
1,2,4-Triazoliun Azolate Heterocycles
Scheme 10
3
Tetrazolium-Based Heterocycles
3.1
Tetrazolium Heterocycles Containing Amino and Azido Substituents
Aminotetrazoles have a high nitrogen content and, despite their large posi-
tive enthalpies of formation [55], are thermally stable [56]. They are, there-
fore, prospective high energy materials. The synthesis of various tetrazolium
salts containing amino substituents on the ring via reactions of 1-amino-5-
methyl tetrazole, and 2-amino-5-methyl tetrazole with iodomethane resulted
in quaternary salts (44) and (46). Metathesis with AgNO3 or AgClO4 formed
1-amino-4,5-dimethyltetrazolium nitrate (45a) or perchlorate (45b), and 2-
amino-4,5-dimethyltetrazolium nitrate (47a) or perchlorate (47b) in high
yields (Scheme 11) [30, 31]. Elemental analyses, and IR, multinuclear NMR,
and mass spectral analyses support the syntheses (Table 5).
A high yield synthesis of 5-amino-tetrazolium nitrate (48) is obtained from
the reaction of 5-amino-tetrazole with nitric acid (Scheme 12) [57]. Its char-
acterization was accomplished through IR, Raman, and multinuclear (1 H,
13 C, 15 N) NMR spectroscopy, and DSC. Bomb calorimetry, sensitivity meas-
urements, and ab initio calculations were also performed. From the combined
Nitrogen-Rich Heterocycles 53
Scheme 11
Scheme 12
well as 45a, 45b, and 47a melt at < 100 ◦ C which qualifies them as energetic
ionic liquids.
These salts were characterized by IR, Raman and NMR spectroscopy, mass
spectrometry, elemental analysis, X-ray, and initial safety testing (impact and
friction sensitivity). Low impact sensitivities were demonstrated. Densities
and thermochemical characteristics of substituted amino, amino-methyl, and
polymethyl tetrazolium salts are summarized in Table 5. All of these new
salts exhibit thermal stabilities > 170 ◦ C based on DSC/TGA studies (except
the azide). The densities of 1-amino-4,5-dimethyl tetrazolium perchlorate
(45b) and 1-methyl-4,5-diamino tetrazolium dinitramide (50b) are markedly
higher than the others.
Synthesis and characterization of 1-nitrotetrazolato-2-nitro-2-azapropane
(52) and 1,4-dimethyl-5-aminotetrazolium 5-nitrotetrazolate (53) are re-
ported in high yields (Scheme 13). Compounds 52 and 53 are white powders
and were characterized by single crystal X-ray analysis. Compound 52 ex-
plodes when thrown in a flame. Both 52 and 53 have low impact sensitiv-
ity [64, 65].
Nitrogen-rich energetic heterocycles, N-nitroso- (57a,c) and N-nitramino-
tetrazole (58a–c) were synthesized from the corresponding aminotetrazoles
(56a–c) by the direct nitration with acetic anhydride/HNO3 (Scheme 14).
The conversion of 57a,c with peroxytrifluoroacetic acid yielded the corres-
ponding nitraamines in high yields (58a: 82%, 58c: 80%). Compounds 56a,c,
57a,c, 58a–c were characterized by single crystal X-ray analysis. The calcu-
lated heats of formation for 58a and 58c were found to be positive (11.7
and 357.0 kJ mol–1 , respectively) and the calculated detonation velocity of
5988 ms–1 (58a) is similar to the values of TNT and nitroglycerin [66].
The chemistry of 5-azidotetrazole, 5-azido-1-phenyltetrazole, and tetra-
zolylpentazole has been reported. The 5-pentazolyl and 5-azido derivatives
Nitrogen-Rich Heterocycles 55
Scheme 13
Scheme 14
(60a,b) of tetrazole include large amounts of nitrogen at 90.6 and 88.3 weight
percent. While the pentazole compound decomposes at low temperature,
5-azido tetrazole is stable at room temperature but extremely sensitive (and
should be synthesized in only small amounts). 5-Azido-1-phenyl tetrazole
(60b) is found to be stable and characterized by single crystal X-ray analysis
(Scheme 15) [63, 67].
Scheme 15
56 R.P. Singh et al.
3.2
Tetrazolium Heterocycles Containing 5,5 -Azotetrazolate Anions
In the 1890s, Thiele first prepared sodium 5,5 -azotetrazolate (61) from oxi-
dation of 5-amino-tetrazole with potassium permanganate [68–71]. Reaction
of strontium and barium chlorides with an aqueous solution of 61 gave
strontium 5,5 -azotetrazolate (62a) and barium 5,5 -azotetrazolate (62b), re-
spectively [72]. Both salts were insoluble in water. Barium 5,5 -azotetrazolate
(62b), was used with various metal sulfates to produce the corresponding
5,5 -azotetrazolate salts (63a–d, 64a,b, and 65a–f), in water (Scheme 16).
These metal salts often contain water of crystallization, and loss of wa-
ter has been observed during storage. Upon loss of water, the sensitivity
Scheme 16
Nitrogen-Rich Heterocycles 57
Scheme 17
Scheme 18
3.3
Bistetrazolate Heterocycles
Although there are several reports of 5,5 -bistetrazolate [82–87] and imino-
bis(5-tetrazolate) [88–90] salts, most are described in the patent literature.
Some processes have been developed for preparing various 5,5 -tetrazolate
salts, which were very useful as slightly toxic and easy to handle gas-
generating agents for air bags and as high molecular weight foaming
agents. In methanol at reflux, bistriazole or iminobis(5-tetrazole) read-
ily quaternized 4-amino-1,2,4-triazole to prepare [bis(4-amino-triazolium)]
5,5 -bistetrazolate (71a) or [bis(4-amino-triazolium)] iminobis(5-tetrazolate)
(71b) (Scheme 19) [81].
Scheme 19
Nitrogen-Rich Heterocycles 61
Scheme 20
62 R.P. Singh et al.
Compound Y No Tm a Td b d ∆ f Hm c
◦C ◦C g/cm3 kJ/mol
a melting point
b thermal decomposition temperature
c standard heat of formation (calculated by Gaussian 03)
Nitrogen-Rich Heterocycles 63
4
Urotropinium-Based Heterocycles
Scheme 21
64 R.P. Singh et al.
were handled easily, were insensitive to air or light, and were soluble in po-
lar organic solvents. They were characterized by analytical and spectroscopic
(IR, Raman, 1 H, 13 C, 14 N NMR) methods, and X-ray diffraction techniques.
The X-ray crystal structure of 87 (monoclinic, space group P21 /C) con-
sisted of a urotropinium cation linked to a planar nitrate group via hydrogen
bridges. The structures of 88a–c showed separation of anions and cations
without contact. The azide ion in 88a (monoclinic, space group P21 /m) is
linear; the dinitramide ion in 88b (monoclinic, space group P21 ) is asymmet-
ric; and the azotetrazolate anion in 88c (monoclinic, space group C2 /m) is
planar [93].
The syntheses of urotropinium and N-methylurotropinium salts were ex-
tended to include several other energetic organic and inorganic anions, such
as 3,5-dinitro-pyrazolate, 4,5-dinitro-imidazolate, 3,5-dinitro-1,2,4-triazolate,
5-nitro-tetrazolate, perchlorate, nitrate, and azide [94]. In methanol solu-
tion, urotropine was found to react readily with 3,5-dinitro-pyrazole, 4,5-
dinitro-imidazole, 3,5-dinitro-1,2,4-triazole, and 5-nitro-tetrazole to form
urotropinium 3,5-dinitro-pyrazolate (89a), 4,5-dinitro-imidazolate (89b), 3,5-
dinitro-1,2,4-triazolate (89c), and 5-nitro-tetrazolate (89d). Reaction of sil-
ver salts of 3,5-dinitro-1,2,4-triazole, nitric, perchloric, and hydrofluoric acid
with N-methylurotropinium iodide led to N-methylurotropinium 3,5-dinitro-
1,2,-triazolate (90a), nitrate (90b), perchlorate (90c), and fluoride (90d)
Scheme 22
Nitrogen-Rich Heterocycles 65
Table 8 Phase transition, decomposition temperature, density, nitrogen content (N), oxy-
gen coefficient (α), and thermochemical results for urotropinium salts at 298.2 K
5
Tetrazine-Based Heterocycles
There has been considerable interest in the study of the reactivity and prop-
erties of various tetrazine derivatives. The 1,2,4,5-tetrazine ring system is
electroactive and has a high electron affinity. Tetrazines possess high posi-
tive heats of formation and large crystal densities—properties important in
energetic materials applications. Additionally, they seem to be insensitive to
destructive stimuli such as friction, impact, and electrostatic discharge.
The synthesis and properties of various ionic 1,2,4,5-tetrazine explo-
sives and energetic materials including 3,6-diguanidino-1,2,4,5-tetrazine
diperchlorate (92a), 3,6-diguanidino-1,2,4,5-tetrazine dinitrate (92b), 3,6-di-
guanidino-1,2,4,5-tetrazine-1,4-di-N-oxide diperchlorate (93a), 3,6-diguanid-
ino-1,2,4,5-tetrazine-1,4-di-N-oxide dinitrate (93b), disilver 3,6-bis-nitrogua-
nyl tetrazine (94), bistriamino-guanidinium 3,6-bis-nitro-guanyl tetrazine
(95), 3,6-dihydrazino-1,2,4,5-tetrazine diperchlorate (96a), 3,6-dihydrazino-
1,2,4,5-tetrazine dinitrate (96b), 3,6-dihydrazino-1,2,4,5-tetrazine bis-(di-
nitramide) (96c), and 3,6-dihydrazino-1,2,4,5-tetrazine bis-dinitroimidazolate
(97) were reported (Fig. 5) [95–98].
66 R.P. Singh et al.
Energetic salts 92a,b, 93a,b, and 96a,b were synthesized by the reaction of
nitric acid and perchloric acid with 3,6-diguanidino-1,2,4,5-tetrazine, 3,6-di-
guanidino-1,2,4,5-tetrazine-1,4-di-N-oxide, and 3,6-dihydrazino-1,2,4,5-te-
trazine, respectively, whereas 96c was made by the reaction of dinitramide
with 3,6-dihydrazino-1,2,4,5-tetrazine.
Synthesis of 94 and 95 involved the reaction of the disodium salt of 3,6-bis-
nitroguanyl tetrazine with AgNO3 and triaminoguanidinium hydrochloride,
respectively.
Poly-rho tests, which are single shot experiments to determine detona-
tion velocity as a function of density, were performed on 92a,b. When 92a
is formulated with 5 weight percent of Kel-F 800 binder, a maximum pellet
density of 1.79 g/cm3 was obtained. At this density, the detonation velocity
measured was 8.07 km/s. The dinitrate derivative (92b) was formulated with
3 weight percent Estane binder and 3 weight percent nitro-plasticizer. At
a maximum pellet density of 1.60 g/cm3 , a detonation velocity of 7.31 km/s
was obtained. Salt 93b was found to be unstable based on DSC analysis,
whereas salt 93a showed improved thermal stability but less than 93a [95].
The crystal density of 95 was reported to be 1.61 g/cm3 , and the heat of for-
mation was 300 ± 2 kcal/mol [97]. The densities of 96a–c, 97 are reported
to be in the range of 1.80 to 1.96 g/cm3 [96]. When these salts were heated
Nitrogen-Rich Heterocycles 67
over the temperature range 40–500 ◦ C at a scan rate of 20 ◦ C/min, 96a–c left
no residue, whereas 97 left an orange powder. In DSC studies, 96a, 96c ex-
hibited their major exotherm between 152 and 164 ◦ C, respectively. Salt 96b
exhibited an exotherm at 190 ◦ C, which is about 30 ◦ C higher than 96a, 96c.
Salt 97 showed the highest exotherm at 220 ◦ C [96]. All tetrazine-based salts
seem to have interesting explosive performance and extraordinary combus-
tion properties [95–98].
The synthesis and properties of several high-nitrogen materials with 3-
amino-6-nitroamino-tetrazine (ANAT) as the anion are reported (Scheme 23).
These salts (99, 100, 103a,b) provide a new and easy approach to highly en-
ergetic salts. All these salts are relatively dense (> 1.55 g/cm3 ) and exhibit
good thermal stability (Td > 148 ◦ C). The calculated detonation velocities and
detonation pressures are comparable to those of explosives such as Tetryl,
PETN, TATB, and RDX. A combination of theoretical and empirical calcu-
lations shows that all these salts have high molar enthalpies of formation
(Table 9) [99].
Scheme 23
Scheme 24
Scheme 25
70 R.P. Singh et al.
Scheme 26
6
Azetidinium-Based Heterocycles
the salts were subjected to small-scale thermal and sensitivity tests [111].
dinitramide ion in 114. The possible reason for these dramatic differences
in the conformations observed for 113c and 114 is claimed to be due to the
different symmetries for this ion found in the two structures, as well as the
absence of hydrogen bonding interaction in 114 [112]. Dehydration of 3,3-di-
nitroazetidinium nitrate with acetic anhydride provided an alternate route
for the synthesis of 1,3,3-trinitroazetidine [111]. The synthesis and charac-
terization of 15 N-labeled isomers of 3,3-dinitroazetidinium nitrate were also
reported [113].
7
Picrate-Based Heterocycles
While anhydrous picric acid is unstable, and its impact and friction sensi-
tivities are higher than those of trinitrotoluene, many organic and inorganic
picrates have been reported [9]. The picrate anion, when combined with high-
nitrogen azolium cations, formed energetic salts with rather high positive
heats of formation. By taking advantage of the acidic properties of picric
acid, energetic salts containing picrate and dipicrate anions have been pre-
pared [114]. Scheme 27 depicts synthetic routes to salts composed of azolium
cations with monopictrate as the anion, including the following picrate
salts: 1,2,4-triazolium (115a), 1-methyl-1,2,4-triazolium (115b), 4-amino-
1,2,4-triazolium (116a), and 5-amino-tetrazolium (116b); and 1,2,4-triazolyl-
3-ammonium (117), 1,4-dimethyl-1,2,4-triazolium (118a), and 1,4-dimethyl-
3-azido-1,2,4-triazolium (118b). In Scheme 28 are shown the synthetic routes
to the energetic bisazolium dipicrate salts. These dipicrate salts are 1,1 -
methylenebisimidazolium (119), 1,1 -methylene bis(3-methylimidazolium)
(120a), and 1,1 -methylene bis(4-methyltriazolium) (120b) [114].
Triazolium or substituted triazolium picrates were first prepared via direct
reaction of the triazole with picric acid in methanol or with silver picrate after
72 R.P. Singh et al.
Scheme 27
Scheme 28
Table 11 Physical characteristics and thermochemical data of picrate and dipicrate salts
NcOd];
e standard enthalpy of formation (calculated by Gaussian 03);
f X-ray (calc);
g triaminoguanidinium azotetrazolate
74 R.P. Singh et al.
places it between picric acid (1.77 cm/g3 ) and HMX (1.90 g/cm3 ). Unexpect-
edly protonation in 116a occurred at one of the ring-nitrogen atoms rather
than on the N-amino group. However, in contrast, in the reaction of picric
acid with the C-amino triazole, a high-melting ammonium salt (117) was
formed.
8
Imidazolium-Based Heterocycles
9
Nitrocubane-Based Heterocycles
Scheme 29
76 R.P. Singh et al.
10
Miscellaneous
Scheme 30
Scheme 31
Scheme 32
studied. Reaction of 132a–c with sodium azide in acetonitrile led to the for-
mation of 133a–c in very good yields (Scheme 32) [122, 123].
Compounds 133a and 133b are liquids, while 133c is a solid at room
temperature. Each was characterized by NMR, HRMS and IR spectroscopy.
Compound 133a is highly friction sensitive and is known to detonate. How-
ever, 133b and 133c are substantially more stable. The densities of 133b and
133c are 1.67 and 1.79 g/cm3 , respectively [122, 123]. Compound 133a has
been recently fully characterized and its single crystal X-ray structure deter-
mined [123].
Energetic heterocyclic salts containing SF5 -propyl substituents on im-
idazole, 4-amino-1,2,4-triazole (134a–c, 135a–d) and pyridine (136) moi-
eties were synthesized (Scheme 33). These salts have moderate densities
(> 1.4 g/cm3 ). The calculated heats of formation for dinitramide-containing
salts are higher than perchlorate and nitrate salts (Table 12) [124].
Reduction of 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane
(137) with stannous chloride dihydrate is reported to give a mixture of
isomeric 2,4,6,8,10- and 2,4,6,8,12-pentanitro-2,4,6,8,10,12-hexaazaisowurt-
78 R.P. Singh et al.
Scheme 33
Cation Anion No Tm a Td b dd ∆f Hm ◦ d
◦C ◦C g/cm3 kJ/mol
a melting point
b thermal degradation temperature
c density, gas pycnometer
d standard heat of formation (calculated by Gaussian 03)
Scheme 34
Scheme 35
11
Summary and Outlook
References
1. Gregory CE (1984) Explosives for North American Engineers, vol 5. Trans Tech Pub-
lications, Clausthal-Zellerfeld, Germany
2. Davis TL (1943) The Chemistry of Powder and Explosives, vol 2. Wiley, New York
3. Cook MA (1958) The Science of High Explosives. Reinhold, New York
4. Akhaven J (2005) Explosives and Propellants. In: Seidel A (ed) Encyclopedia of
Chemical Technology. Wiley, Hoboken, N.J.
5. Urbanski T (1964) Chemistry and Technology of Explosives, vol 3. Pergamon, Ox-
ford
6. Bailey A, Murray SG (1989) Explosives, Propellent and Pyrotechnics. Brassey’s, Ox-
ford
7. Urbanski T (1984) Chemistry and Technology of Explosives, vol 4. Pergamon, Ox-
ford, p 202
8. Agrawal JP (1998) Prog Energy Combust Sci 24:1–30
9. Singh G, Kapoor IPS, Mannan SM, Kaur J (2000) J Hazard Mater A79:1–18
10. Fried LE, Manaa MR, Pagoria PF, Simpson RL (2001) Ann Rev Mater Res 31:291–
321
11. Singh RP, Verma RD, Meshri DT, Shreeve JM (2006) Angew Chem 45:3584–3601
12. Shlyapochnikov VA, Tafipolsky MA, Tokmakov IV, Baskir ES, Anikin OV, Strelen-
ko YA, Luk’yanov OA, Tartakovsky VA (2001) J Molec Struct 559:147–166
13. Pagoria PF, Lee GS, Mitchell AR, Schmidt RD (2002) Thermochim Acta 384:187–204
14. Lesnikovich AI, Ivashkevich OA, Levchik SV, Balabanovich AI, Gaponik PN, Ku-
lak AA (2002) Thermochim Acta 388:233–251
15. Sikder AK, Sikder NJ (2004) J Hazard Mater A112:1–15
16. Bottaro J (2005) Ideas to Expand Thinking About New Energetic Materials. In:
Shaw RW, Brill TB, Thompson DL (eds) Adv Ser Phys Chem, vol 16. World Scientific,
Singapore, pp 473–501
17. Feuer H, Nielsen AT (1990) Nitro Compounds. VCH, New York
18. Nielsen AT (1995) Nitrocarbons. VCH, New York
19. Köhler J, Meyer R (1991) Explosivstoffe, 7th Ed. Wiley, Weinheim, Germany
20. Köhler J, Meyer R (1998) Explosivstoffe, 9th Ed. Wiley, Weinheim, Germany
21. Klapötke TM, Krumm B, Holl G, Kaiser M (2000) Energetic Materials: Modeling
of Phenomena, Experimental Characterization, Environmental Engineering. Fraun-
hofer Institut, Karlsruhe, Germany
22. Simpson RL, Urtiew PA, Ornellas DL, Moody GL, Scribner KS, Hofman DM (1997)
Propell Explos Pyrotech 22:249–255 and references cited therein
23. Krause HH (2005) New Energetic Materials. In: Teipal U (ed) Energetic Materials.
Wiley, Weinheim, Germany
24. Hiskey MA, Chavez DE, Naud DL, Son SF, Berghout HL, Bolme CA (2000) Proc Int
Pyrotech Semin 27:3–14
25. Pedley JB (1994) Thermochemical Data and Structure of Organic Compounds, vol I.
Thermodynamic Research Center, College Station, TX
26. Jiminez P, Roux MV, Turrion CJ (1998) Chem Thermodyn 21:759–764
27. Drake GW, Hawkins T, Brand A, Hall L, McKay M, Vij A, Ismail I (2003) Propell
Explos Pyrotech 28(4):174–180
28. Drake GW (2003) US Patent 6 509 473 B1
29. Schmidt MW, Gordon MS, Boatz JA (2005) J Phys Chem A 109:7285–7295
30. Xue H, Arritt SW, Twamley B, Shreeve JM (2004) Inorg Chem 43:7972–7977
31. Xue H, Gao Y, Twamley B, Shreeve JM (2005) Chem Mater 17:191–198
Nitrogen-Rich Heterocycles 81
32. Drake GW, Hawkins T, Tollison K, Hall L, Vij A, Sobaski S (2005) In: Rogers RD,
Seddon KR (eds) Ionic Liquids-III A: Fundamentals, Progress, Challenges, and Op-
portunities. ACS Symposium Series 902. ACS, Washington, DC, pp 259–302
33. Trohalaki S, Pachter R, Drake GW, Hawkins T (2005) Energ Fuel 19:279–284
34. Kaplan G, Drake G, Tollison K, Hall L, Hawkins T (2005) J Heterocycl Chem 42:19–27
35. Ostrovskii VA, Pevzner MS, Kofman TP, Tselinskii IV (1999) Targets Heterocycl Syst
3:467–526
36. Denault CC, Marx PC, Takimoto HH (1968) J Chem Eng Data 13:514–516
37. Xue H, Shreeve JM (2005) Adv Mater 17:2142–2146
38. Buscemi S, Pace A, Pibiri I, Vivona N, Spinelli D (2003) J Org Chem 68:605–608
39. Funabiki K, Noma N, Shibata K (1999) J Chem Res Synop pp 300–301, and references
cited therein
40. Xue H, Twamley B, Shreeve JM (2004) J Org Chem 69:1397–1400
41. Mirzaei YR, Shreeve JM (2003) Synthesis 24–26
42. Mirzaei YR, Shreeve JM (2002) J Org Chem 67:9340–9345
43. Mirzaei YR, Xue H, Shreeve JM (2004) Inorg Chem 43:361–367
44. Shitov OP, Korolev VL, Bogdanov VS, Tartakovsky VA (2003) Russ Chem Bull Int Ed
52:695–699
45. Egashira M, Scrosati B, Armand M, Beranger S, Michot C (2003) Electrochem Solid
State Lett 6(4):A71–A73
46. Katritzky A, Singh S, Kirichenko K, Holbrey JD, Smiglak M, Reichert MW, Rogers RD
(2005) Chem Commun 868–870
47. Sitzmann MI (1978) J Org Chem 43:3389–3391
48. Kofman TP, Paketina EA (1997) Russ J Org Chem (Eng Trans) 33:1125–1132
49. Chernyshev VM, Zemlyakov ND, Il’in VB, Taranushich VA (2000) Zh Prikl Khim
73:791–793
50. Ogihara W, Yoshizawa M, Ohno H (2004) Chem Lett 33(8):1022–1023
51. Xue H, Gao Y, Twamley B, Shreeve JM (2005) Inorg Chem 44:5068–5072
52. Martin AR, Yallop HJ (1958) Trans Faraday Soc 54:257–267
53. Ostrovskii VA, Koldobskii GI, Shirokova NP, Poplavskii VS (1981) Khim, Grterotsikl
Soedin 4:559–562
54. Krakovskii IM, MolChanova MS, Evtushenko AV, Shlyapochnikova VA (1998) Russ
Chem Bull 47:1266–1273
55. Kozyro AA, Simirsky VV, Krasulin AP, Sevruk VM, Kabo GJ, Gopanik ML, Grig-
otiev YV (1990) Zh Fiz Khim 64:656–661
56. Gao A, Oyumi Y, Brill TB (1991) Combust Flame 83:345–352
57. von Denffer M, Klapötke TM, Kramer G, Spieß G, Welch JM, Heeb G (2005) Propell
Explos Pyrotech 30:191–195
58. Klapötke TM, Mayer P, Schulz A, Weigand JJ (2005) J Am Chem Soc 127:2032–2033
59. Gálvez-Ruiz JC, Holl G, Karaghiosoff K, Klapötke TM, Löhnwitz K, Mayer P, Nöth H,
Polborn K, Rohbogner CJ, Suter M, Weigand JJ (2005) Inorg Chem 44:4237–4253
60. Fisher G, Holl G, Klapötke TM, Weigand JJ (2005) Thermochim Acta 437:168–178
61. Geith J, Klapötke TM, Weigand J (2004) Propell Explos Pyrotech 29:3–8
62. Hammerl A, Klapötke TM, Nöth H, Warchhold M, Holl G (2003) Propell Explos
Pyrotech 28:165–173
63. Hammerl A, Klapötke TM, Mayer P, Weigand JJ, Holl G (2005) Propell Explos Py-
rotech 30:17–26
64. Klapötke TM, Mayer P, Verma V (2006) Propell Explos Pyrotech 31:263–268
65. Klapötke TM, Karaghiosoff K, Mayer P, Penger A, Welch JM (2006) Propell Explos
Pyrotech 31:188–195
82 R.P. Singh et al.
66. Karaghiosoff K, Klapötke TM, Mayer P, Piotrowski H, Polborn K, Willer RL, Wei-
gand JJ (2006) J Org Chem 71:1295–1305
67. Hammerl A, Klapötke TM (2002) Inorg Chem 41:906–912
68. Thiele J (1892) Liebigs Ann 270:54–63
69. Thiele J, Marais JT (1893) Justus Liebigs Ann Chem 273:144–160
70. Thiele J (1893) Ber Dtsch Chem Ges 26:2645–2646
71. Thiele J (1898) Justus Liebigs Ann Chem 303:57–75
72. Hammerl A, Holl G, Klapötke TM, Mayer P, Noth H, Piotrowski H, Warchhold M
(2002) Eur J Inorg Chem 834–845
73. Hiskey MA, Goldman N, Stine JR (1998) J Energ Mater 16:119–127
74. Peng Y, Wong C (1999) US Patent 5 877 300 CA 130:196656
75. Hammerl A, Klapötke TM, Nöth H, Warchhod M, Holl G, Kaiser M, Ticmanis U
(2001) Inorg Chem 3570–3575
76. Hammerl A, Holl G, Kaiser M, Klapötke TM, Mayer P, Piotrowski H, Vogt M (2001)
Naturforschung 847–856
77. Hammerl A, Holl G, Kaiser M, Klapötke TM, Mayer P, Nöth H, Piotrowski H, Suter M
(2001) Naturforschung 857–870
78. Klapotke TM, Holl G (2001) Green Chem G75–G76
79. Hammerl A, Hiskey MA, Holl G, Klapötke TM, Polborn K, Stierstorfer J, Weigand J
(2005) Chem Mater 17:3784–3793
80. Tappan BC, Ali AN, Son SF (2006) Propell Explos Pyrotech 31:163–168
81. Ye C, Xiao J-C, Twamley B, Shreeve JM (2005) Chem Commun 2750–2752
82. Hyoda S, Kita M, Sawada H, Nemugaki S, Otsuka S, Miyawaki Y, Ogawa T, Kubo Y
(2000) US Patent 6 040 453, CA: 132:207845
83. Hyoda S, Kita M, Sugino A, Ueta T, Sato K (2001) EU Patent 1 162 198, CA: 136:20077
84. Hyoda S, Kita M, Swada H, Nemugaki S, Ueta T, Satoh K, Otsuka S, Miyawaki Y,
Taniguchi H (2000) EU Patent Appl, EP 1 016 662, CA: 133:74021
85. Torii S, Tsuyama M, Miyawaki Y, Kubo Y, Ogawa T (2000) Jpn Kokai Tokkyo Koho,
JP 2000281662, CA 133:252440
86. Friedrich M, Gálvez-Ruiz JC, Klapötke TM, Mayer P, Weber P, Weigand JJ (2005) In-
org Chem 44:8044–8052
87. Hammerl A, Holl G, Kaiser M, Klapötke TM, Piotrowski H (2003) Z Anorg Allg Chem
629:2117–2121
88. Kita M, Ueda T (2004) Jpn Kokai Tokkyo Koho, JP 2004067544, CA 140:217643
89. Naud DL, Hiskey MA (2003) US Patent Appl Publ, 2 003 060 634, CA:138:255236
90. Highsmith TK, Hajik RM, Wardle RB, Lund GK, Blau RJ (1995) US Patent 5 468 866
91. Gao Y, Ye C, Twamley B, Shreeve JM (2006) Chem-Eur J 12:9010
92. Foss ME, Hirst EL, Jones JKN, Springall HD, Thomas AT, Urbanski T (1950) J Chem
Soc 624–628
93. Ang H-G, Fraenk W, Karaghiosoff K, Klapötke TM, Nöth H, Sprott J, Sutter M,
Vogt M, Warchhold MZ (2002) Z Anorg Allg Chem 628:2901–2906
94. Xue H, Twamley B, Shreeve JM (2006) Eur J Inorg Chem 2959–2965
95. Chavez DE, Hiskey MA, Naud Di (2004) Propell Explos Pyrotech 29:209–215
96. Oxley JC, Smith JL, Chen H (2002) Thermochim Acta 91–99
97. Chavez DE, Hiskey MA, Gilardi RD (2004) Org Lett 6:2889–2891
98. Chavez DE, Tappan BC, Hiskey MA, Son SF, Harry H, Montoya D, Hagelberg S (2005)
Propell Explos Pyrotech 30:412–417
99. Gao H, Wang R, Twamley B, Hiskey MA, Shreeve JM (2006) Chem Commun 4007–
4009
100. Kerth J, Löbbecke S (2002) Propell Explos Pyrotech 27:111–118
Nitrogen-Rich Heterocycles 83
101. Chavez DE, Hiskey MA, Gilardi RD (2000) Angew Chem Int Ed 39:1791–1793
102. Huynh MHV, Hiskey MA, Chavez DE, Naud DL, Gilardi RD (2005) J Am Chem Soc
127:12537–12543
103. Huynh MHV, Hiskey MA, Archuleta JG, Roemer EL, Gilardi R, Chavez DE, Naud DL,
Gilardi RD (2004) Angew Chem Int Ed 43:5658–5661
104. Miller DR, Swenson DC, Gillan EG (2004) J Am Chem Soc 126:5372–5373
105. Huynh MHV, Hiskey MA, Hartline EL, Montoya DP, Gilardi R (2004) Angew Chem
Int Ed 43:4924–4928
106. Huynh MHV, Hiskey MA, Pollard CJ, Montoya DP, Hartline EL, Gilardi R (2004)
J Energ Mater 22:217–229
107. Frumkin AE, Churakov AM, Strelenko YA, Kachala VV, Tartakovsky VA (1999) Org
Lett 1:721–724
108. Archibald TG, Gilardi R, Baum K, George C (1990) J Org Chem 55:2920–2924
109. Coburn MD, Hiskey MA, Archibald TG (1997) Waste Manag 17:143–146
110. Hiskey MA, Coburn MD, Mitchell MA, Benicewicz BC (1992) J Heterocycl Chem
29:1855–1856
111. Hiskey MA, Stincipher MM, Brown JE (1993) J Energ Mater 11:157–165
112. Gilardi RD, Butcher RJ (1998) J Chem Crystallogr 28(3):163–169
113. Coburn MD, Hiskey MA, Oxley JC, Smith JL, Zheng W, Rogers E (1998) J Energ
Mater 16(2):73–99
114. Jin C-M, Ye C, Piekarski C, Twamley B, Shreeve JM (2005) Eur J Inorg Chem 3760–
3767
115. Zhang M-X, Eaton PE, Gilardi R (2000) Angew Chem Int Ed 39(2):404–401
116. Chung G, Schmidt MW, Gordon MS (2000) J Phys Chem A 104:5647–5650
117. Fau S, Bartlett RJ (2001) J Phys Chem A 105:4096–4106
118. Nguyen NT (2003) Coordinat Chem Rev 244:93–113
119. Christe KO, Dixon DA, McLemore D, Wilson WW, Sheehy JA, Boatz JA (2000) J Fluor
Chem 101:151–153
120. Jones CB, Haiges R, Schroer T, Christe KO (2006) Angew Chem Int Ed 45:4981–4984
121. Ju Y, Kumar D, Verma RS (2006) J Org Chem 71:6697–6700
122. Muralidharan K, Omotowa BA, Twamley B, Piekarski C, Shreeve JM (2005) Chem
Commun 5193–5195
123. Göbel M, Karaghiosoff K, Klapötke TM (2006) Angew Chem Int Ed 45:6037–6040
124. Gao H, Ye C, Winter RW, Gard GL, Sitzmann ME, Shreeve JM (2006) Eur J Inorg
Chem 3221–3226
125. Lukyanov OA, Shykova NI (2004) Russ Chem Bull Int Ed 53:566–568
126. Gyeong S, Cho JR, Goh EM, Kim J-K (2005) Propell Explos Pyrotech 30:445–449
Struct Bond (2007) 125: 85–121
DOI 10.1007/430_2007_057
© Springer-Verlag Berlin Heidelberg
Published online: 12 May 2007
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
2 Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Abstract The possibility of new high explosives based on nitrogen-rich tetrazole build-
ing blocks is discussed. The expected advantages include gaseous products, high heats of
formation, high propulsive/expolosive power, high specific impulse, and high flame tem-
peratures. In addition, these new explosives do not have the toxicity and environmental
activity of currently used organo-nitro explosives. The synthesis and characteristics of
a series of neutral tetrazole compounds are looked at as well as the neutral nitramine,
dinitrobiuret.
Abbreviations
BTNA Bistrinitroethylamine
DAT 1,5-Diaminotetrazole
DNB Dinitrobiuret
EI energetic ingredient
NTNAP 1-Nitrotetrazolato-2-nitro-2-azapropane
TNE Trinitroethanol
MMTHT 1-Methyl-5-(1-methyl-2-(2,2,2-trinitroethyl)hydrazinyl)-1H-tetrazole
MTHTE 2-(5-(1-methyl-2-(2,2,2-trinitroethyl)hydrazinyl)-1H-tetrazol-1-yl)ethanol
86 T.M. Klapötke
1
Introduction
2
Strategies
Fig. 2 Examples for EM that (i) show oxidation of the carbon backbone (left) [12], (ii)
possess ring or cage strain (center) [13], or (iii) have very high positive heats of formation
(right) [14]
hepta- and octanitrocubanes possess very high densities and enhance the
energies utilizing substantial cage strain and therefore belong to the sec-
ond class of explosives. Members of the third class of compounds are
5,5 -azotetrazolate salts, which show the desired remarkable insensitivity to
electrostatic discharge, friction, and impact while having a very high (posi-
tive) heat of formation.
Nitrogen is unique amongst all other elements of the periodic table in
so far that the bond energy per two-electron bond increases from a sin-
gle over a double to a triple bond (Fig. 3). For carbon the situation is the
opposite and one might expect acetylene (H – C ≡ C – H) to polymerize in
an exothermic reaction whereas N2 (N ≡ N) is more stable than any other
polynitrogen species [15–18]. One may ask what the difference between for-
mally sp-hybridized nitrogen in N2 and sp-hybridized carbon in HCCH is.
Indeed the bond situation is very similar and only an NBO analysis can help
to explain the difference:
N≡N σ(N – N) 64% s 36% p
σ– LP(N) 34% s 66% p
H–C≡C–H σ(C – C) 49% s 51% p
σ(C – H)C 45% s 55% p
H 100% s
Whereas in acetylene the C – C σ bond is indeed essentially made up of sp-
hybrid orbitals, in N2 the σ bond orbitals have two-third s and only one-third
p character and form therefore a much stronger bond than the carbon atoms
do in HCCH.
With respect to all-nitrogen species (Nx ), there are only a very few species
known that have been isolated and characterized (Table 1).
Quite recently, researchers from the Max Plank Institute for Chemistry
and the Geophysical Laboratory at the Carnegie Institution of Washington re-
ported the polymerization of nitrogen in sodium azide [16–18]. With increas-
ing pressure on sodium azide, the sample became completely opaque above
New Nitrogen-Rich High Explosives 89
Fig. 3 Bond energies (in kcal mol–1 ) for single, double and triple bonds (left) and per two-
electron bond (right) [15]
Neutral compounds
N2 Dinitrogen Thermodynamic stable form of nitrogen, [15]
not energetic
cg-Nx Cubic polynitrogen Prepared from N2 in diamond anvil [16–18]
cell at 200 K and 110 GPa; metastable only
above 42 GPa
Ionic species
N3 – Azide anion Suitable energetic anion (or covalent azide [15]
group R-NNN)
N5 + Pentanitrogen cation Metastable salts only with large counter [19–21]
anions (Sb2 F11 – etc.)
N5 – Pentanitrogen anion Only detected in the gas phase (mass [22, 23]
spectrometry)
Fig. 4 cg-N structure. Each N atom is connected to three neighbors by three single co-
valent bonds. The primitive cell is shown on the left, and an extended structure of the
polymeric N is shown on the right
Fig. 5 Structure of N5 – and some computationally known (but not yet synthesized) neu-
tral polynitrogen species [24–30]
Fig. 6 Nitro (A), nitrato (B), nitramino (C), and nitrimino (D) groups
3
Neutral Tetrazole Compounds
3.1
1,5-Diaminotetrazole (DAT)
Before Raap, Gaponik et al. [37] improved the synthesis reported by Stolle
et al. [38]. By reacting thiosemicarbazide with lead oxide and sodium azide in
a CO2 atmosphere, a carbodiimide intermediate is formed and reacts in situ
with HN3 to lead to 1,5-DAT (Fig. 10). Unfortunately, this reaction leads to
large amounts of lead azide as the side-product, which makes this synthesis
problematic for an industrial scale.
The most recent approach to the synthesis of DAT was made at LMU
Munich [39, 40] and eliminates the formation of highly explosive lead azide
94 T.M. Klapötke
3.2
1-Nitrotetrazolato-2-nitro-2-azapropane (NTNAP)
Fig. 13 Molecular structure of NTNAP in the crystalline state. Displacement ellipsoids are
shown at the 50% probability level
trum the methyl (– CH3 , 3.55 ppm) and methylene (– CH2 –, 6.85 ppm) pro-
tons show singlet resonances clearly shifted, as expected, to low field.
The DSC plot of NTNAP was recorded with a heating rate of 2 ◦ C min–1
(Fig. 14). The covalent compound is thermally reasonably stable, melts with-
out decomposition at 100 ◦ C, and shows decomposition with an onset of
about 180 ◦ C. Therefore, the compound possesses properties that may make
it suitable for melt-casting processes.
The sensitivity and detonation parameters of NTNAP are summarized in
Table 2.
96 T.M. Klapötke
NTNAP
3.3
N-Trinitroethyl Derivatives of Nitrogen Containing Compounds
Sum formula C5 H9 N9 O6
Formula weight [g mol–1 ] 291.21
TMD [g cm–3 ] 1.63
Tdecomposition, onset (DSC, 2 K min–1 ) [◦ C] 82.5
– ∆c U [cal g–1 ] 2936
Impact sensitivity [J] > 30
Friction sensitivity [N] 108 (visible flame)
Oxygen balance [%] – 46.7
– ∆f U [kJ kg–1 ] + 1889
QV [kJ kg–1 ] – 6368
Tex [K] 4404
P [kbar] 277
D [m s–1 ] 8307
V0 [L kg–1 ] 783
Fig. 16 ORTEP representation of the molecular structure of TNE in the crystalline state.
The thermal ellipsoids are shown at the 50% probability level
Fig. 17 ORTEP representation of the molecular structure of BTNA in the crystalline state.
The thermal ellipsoids are shown at the 50% probability level
3.4
Nitrated Aminotetrazoles
method [75]. In the chiral space group P21 21 21 the Friedel pairs were merged.
Selected data from the X-ray data collection and refinements are given in
Table 5. Further information regarding the crystal structure determinations
has been deposited with the Cambridge Crystallographic Data Centre [76] as
supplementary publication numbers 635164 (2), 635163 (5), and 635160 (6).
Compound 2 was characterized as a dibasic acid with pKa values of 2.5 and
6.1 [60]. Previously it was not possible to locate the hydrogen atoms and it was
assumed that the more acid hydrogen is located at the nitramino group [77].
However, using single crystal diffraction both hydrogen atoms were located
on the tetrazole ring at nitrogen atoms N1 and N4. The 15 N NMR spectra
shows only four signals, which suggests the same connectivity present in so-
lution. DFT calculations show that this isomer is the one calculated lowest in
energy even in the gas phase. The molecular unit can be seen in Fig. 21.
The geometry of the tetrazole ring in 2 can be compared with that of
5-aminotetrazole monohydrate (5-AT) [78]. The bond lengths in 2 are about
1.0 Å shorter than in 5-AT, with the shortest distance in 2 between the
atoms N2 and N3 with 1.278(1) Å and the longest between N1 and N2
with 1.358(1) Å. The C1 – N5 bond length is 1.341(1) Å, which is closer to
a C = N double bond (1.28 Å) than a C – N single bond (1.46 Å), whereas
the nitramine bond N5 – N6 is considerably longer (1.363(1) Å). The angles
in 2 differ from those in 5-AT. The N4 – C1 – N1 angle in 2 (103.9(1)◦ ) is
smaller than in 5-AT (107.9(1)◦ ), which can be explained using the VSEPR
102 T.M. Klapötke
2 5 6
Formula CH2 N6 O2 C2 H4 N6 O2 C2 H4 N6 O2
Form. weight [g mol–1 ] 130.09 144.11 144.11
Crystal system Monoclinic Orthorhombic Monoclinic
Space Group P21 /c (14) P21 21 21 (19) P21 /c (14)
Color/habit Colorless cuboids Colorless rods Colorless blocks
Size [mm] 0.18 × 0.13 × 0.08 0.19 × 0.16 × 0.08 0.24 × 0.18 × 0.15
a [Å] 9.4010(3) 6.6140(1) 9.5278(9)
b [Å] 5.4918(1) 8.5672(2) 7.7308(7)
c [Å] 9.3150(3) 19.2473(4) 8.4598(9)
α [◦ ] 90.0 90.0 90.0
β [◦ ] 105.762(3) 90.0 112.875(9)
γ [◦ ] 90.0 90.0 90.0
V [Å3 ] 462.84(2) 1090.62(4) 574.1(1)
Z 4 8 4
ρcalc [g cm–3 ] 1.867 1.755 1.667
µ [mm–1 ] 0.169 0.153 0.145
F(000) 264 592 296
λMoKα [Å] 0.71073 0.71073 0.71073
T [K] 100 100 200
θ min, max [◦ ] 4.3, 32.1 3.7, 32.1 4.6, 26.0
Dataset – 13 : 13; – 8 : 8; – 9 : 9; – 12 : 12; – 11 : 11; – 9 : 9;
– 13 : 13 – 28 : 27 – 10 : 10
Reflections collected 6375 15881 5625
Independent reflections 1537 2104 1128
Rint 0.037 0.034 0.034
Observed reflections 1050 1616 1087
No. parameters 90 213 107
R1 (obs) 0.0343 0.0297 0.0440
wR2 (all data) 0.0960 0.0696 0.0950
GooF 1.00 0.99 1.21
Weighting scheme 0.05710 0.04220 0.03930 0.16880
Resd. Density [e/ Å3 ] – 0.23, 0.31 – 0.32, 0.21 – 0.22, 0.17
Device type Oxford Oxford Oxford
Xcalibur3 CCD Xcalibur3 CCD Xcalibur3 CCD
Solution SIR-92 SIR-92 SIR-92
Refinement SHELXL-97 SHELXL-97 SHELXL-97
Absorption correction None Multiscan Multiscan
CCDC 635164 635163 635160
Fig. 21 Left: View of the molecular structure of 2 with thermal ellipsoid drawn at
the 50% probability level. Hydrogen atoms are shown as small spheres of arbitrary
radii. Right: H-bonding in 2 (N1 – H1 – N2i : D – H 0.89(2) Å, H – A 2.11(2) Å, D – A
2.948(1) Å, D – H – A 156 (1)◦ ; N4 – H4 – N5ii : 0.98(2) Å, 1.83(2) Å, 2.850(2) Å, 175 (2)◦ ;
N4 – H4 – O2ii : 0.98(2) Å, 2.48(2) Å, 3.146(1) Å, 125 (1)◦ ; N4 – H4 – N6ii : 0.98(2) Å,
2.58(2) Å, 3.4748(14) Å, 153 (1)◦ ; ((i) 1 – x, 0.5 + y, 1.5 – z (ii) 2 – x, – 0.5 + y, 1.5 – z))
group also shows only slight derivations from the planarity (torsion angle
C1 – N5 – N5 – O1 1.8(1)◦ ), and is stabilized via a intramolecular hydrogen
bond between N1 – H1 – O1. Bond lengths for the non-hydrogen atoms in 2
are given in Table 6. Each 5-nitriminotetrazole molecule participates in two
strong intermolecular hydrogen bonds, as is illustrated in Fig. 21. The strong
hydrogen bonds appear to be the most important reason for the high density
of 1.867 g cm–3 .
Accordingly, the densities of 5 and 6 decrease because of the presence of
the methyl groups. The molecular unit of 5 (Fig. 22) shows a similar con-
104 T.M. Klapötke
Fig. 22 Left: View of the molecular structure of 5, representing the half of the
asymmetric unit. Thermal ellipsoid are drawn at the 50% probability level and hy-
drogen atoms are shown as small spheres of arbitrary radii. Right: H-bonding in 5
(N4 – H4 – N11i : D – H 0.93(2) Å, H – A 1.96(2) Å, D – A 2.848(2) Å, D – H – A 158 (2)◦ ;
N4 – H4 – O4: 0.93(19) Å, 2.59(2) Å, 3.098(2) Å, 115.3(15)◦ ; N10 – H10 – O1i : 0.80(2) Å,
2.10(2) Å, 2.874(2) Å, 163 (2)◦ ; ((i) – x, 0.5 + y, 0.5 – z))
the vibrations expected from comparison with similar compounds in the lit-
erature [80, 81]. The N – NO2 groups result in the strong absorptions in the
1280–1300 (νsym (NO2 )) and 1560–1620 (νasym (NO2 )) cm–1 regions as well as
to a weak band at 945–970 (ν(N – N)) cm–1 [82]. The IR and Raman spec-
tra of compounds 2, 5, and 6 show further characteristic absorption bands
in the regions listed below: 3250–3100 cm–1 [ν(N – H)], 3000–2850 [ν(C – H)
5, 6], 1680–1550 [ν(N – H)], 1550–1350 [ν, tetrazole ring, νas (CH3 )], ∼ 1380
[ν(CH3 )], 1350–700 [ν(N1 – C1 – N4), ν(N – N), ν(N – H), ν, tetrazole ring],
< 700 [νδoop (N – H)].
106 T.M. Klapötke
Fig. 23 Left: View of the molecular structure of 6. Thermal ellipsoid are drawn at the
50% probability level and hydrogen atoms are shown as small spheres of arbitrary radii.
Right: H-bonding in 6 (N5 – H1 – N4i : D – H 0.86(2) Å, H – A 2.13(2) Å, D – A 2.965(2) Å,
D – H – A 163 (2)◦ ; N5 – H1 – O1ii : 0,86(2), 2.66(2), 3.070(2), 111.0(18) ((i) – x, – y, 2 – z;
(ii) x, 0.5 – y, 0.5 + z))
Fig. 24 Solid state Raman spectra of compounds 2 (bottom), 5 (middle), and 6 (top)
Table 8 15 N NMR and 13 C NMR chemical shifts and 15 N–1 H coupling constants
Fig. 25 DSC plot (endo up) of compounds 2, 5, and 6 (5 ◦ C min–1 ). Tonset : 2 122 ◦ C;
5 125 ◦ C; 6 123 ◦ C
friction (< 8 N) and since the value is comparable with lead azide, it should be
considered to be a primary explosive, and should therefore only be handled
with appropriate precautions. Compound 5 is moderately sensitive towards
impact (< 12.5 J) and friction (< 160 N). However, 6 shows increased sensitiv-
ities comparing to 5 (impact: < 3.0 J, friction: < 145 N). Accordingly, 5 and 6
fall into the group of compounds described as “sensitive”.
The heats of combustion for the compounds 2, 5, and 6 were determined
experimentally, using a Parr 1356 bomb calorimeter (static jacket) equipped
with a Parr 1108CL oxygen bomb [87]. To ensure better combustion, the
samples (ca. 200 mg) were pressed with a defined amount of benzoic acid
(ca. 800 mg) to form a tablet and a Parr 45C10 alloy fuse wire was used
for ignition. In all measurements, a correction of 2.3 cal cm–1 wire burned
has been applied, and the bomb was examined for evidence of noncom-
busted carbon after each run. A Parr 1755 printer was furnished with the
Parr 1356 calorimeter to produce a permanent record of all activities within
the calorimeter. The reported values are the average of three separate meas-
urements. The calorimeter was calibrated by combustion of certified benzoic
acid (SRM, 39i, NIST) in an oxygen atmosphere at a pressure of 3.05 MPa. The
experimental results of the constant volume combustion energy (∆c U) of the
salts are summarized in Table 9. The standard molar enthalpy of combustion
(∆c H ◦ ) was derived from ∆c H ◦ = ∆c U + ∆nRT (∆n = ∆ni (products, g) –
∆ni (reactants, g);∆ni is the total molar amount of gases in the products or re-
actants). The enthalpy of formation, ∆f H ◦ , for each of the salts was calculated
at 298.15 K using Hess’ law and the following combustion reactions:
2 5 6
Formula CH2 N6 O2 C2 H4 N6 O2 C2 H4 N6 O2
Molecular mass 130.09 144.11 144.11
Impact sensitivity [J]a 1.5 12.5 3.0
Friction sensitivity [N]b 8 160 145
Electrical discharge No reaction No reaction No reaction
N [%]c 64.61 58.32 58.32
Ω [%]d – 12.30 – 44.41 – 44.41
Combustion Yes Yes Yes
Tdec [◦ C]e 122 125 122
Density [g cm–3 ]f 1.87 1.76 1.67
– ∆c U [cal g–1 ]g 1750 2700 2902
– ∆c H ◦ [kJ mol–1 ]h 944 1619 1740
◦ [kJ mol–1 ]i
∆ f Hm 264 260 380
∆dec Hm◦ [J g–1 ]j 2638 1685 2158
– ∆E Um◦ [J g–1 ]k 5326 5235 5998
TE [K] l 4309 3824 4283
P [kbar]m 363 295 289
D [m s–1 ]n 9173 8433 8434
Gas volume [mL g–1 ]o 404 395 413
a BAM methods, insensitive > 40 J, less sensitive ≥ 35 J, sensitive ≥ 4 J,
very sensitive ≤ 3 J [85, 86];
b BAM methods, insensitive > 360 N, less sensitive = 360 N, sensitive < 360 N to > 80 N,
very sensitive ≤ 80 N, extremely sensitive ≤ 10 N [85, 86];
c Nitrogen content;
d Oxygen balance;
e Decomposition temperature from DSC (β = 5 ◦ C);
f Estimated from X-ray diffraction;
g Experimental (constant volume) combustion energy;
h Experimental molar enthalpy of combustion;
i Molar enthalpy of formation;
j Experimental enthalpy of decomposition using DSC;
k Energy of explosion, EXPLO5 V5.02;
l Explosion temperature, EXPLO5 V5.02;
m Detonation pressure, EXPLO5 V5.02;
n Detonation velocity, EXPLO5 V5.02;
o Assuming only gaseous products, EXPLO5 V5.02
detonation pressure of 289 kbar in spite of the highest positive heat of forma-
tion due to its low density.
In order to determine the thermal decomposition products, the com-
pounds were heated in a evacuated steel tube for ca. 30 s at a temperature
of 250 ◦ C and the gaseous products were transferred into an evacuated gas
IR cell. Figure Fig. 26 shows the gas phase IR spectra of the decomposition
products of 2, 5, and 6.
The thermal decomposition of 2 results in the formation of only two main
products, which could be identified using IR spectroscopy, namely CO2 [94]
and CO [95]. In addition trace amounts of HCN and CH4 were visible in
the gas-phase IR spectrum, however, no evidence for the formation of water
vapor was found. In the methylated compounds, many more decomposition
products were observed using gas-phase IR spectroscopy. Besides CO2 and
CO, larger amounts of CH4 [96] and HCN were found in the decomposition of
5 and 6, in comparison with 2. In contrast to 5, where bigger amounts of ex-
pected NH3 were detected, the thermal decomposition of 6 shows only traces
of NH3 but moderate amounts of N2 O [97].
The electrostatic potentials were illustrated after computing a optimal
geometry at the B3LYP/6-31G∗ level of theory using the program package
HyperChem 7.52 [98]. Figure 27 shows the 0.001 electron bohr–3 3D isosur-
face of electron density for 2, 5, and 6. In these diagrams a Jorgensen–Salem
3.5
1,6-Dimethyl-5-nitraminotetrazole
Fig. 29 1,6-Dimethyl-5-nitraminotetrazole
114 T.M. Klapötke
4
Neutral Nitramine Compounds
4.1
Dinitrobiuret (DNB)
The LMU Munich research group has also been looking at tetrazole-free
nitrogen-rich compounds that contain oxidizing groups such as nitramine
functionalities. In this context, the preparation and structural characteriza-
tion of dinitrobiuret (DNB) (Fig. 31) was carried out [109, 110]. The high
chemical and thermal stability of DNB and the determined critical diameter
of 6 mm for DNB (Figs. 32, 33) in the Koenen test (steel shell test) is compa-
rable to the values reported for HMX (8 mm), RDX (8 mm), or PETN (6 mm)
and prompted us to obtain the thermodynamic data and detonation pressures
and velocities for DNB in a combined experimental and theoretical study.
Fig. 33 Steel sleeve test of DNB (10 g, 6 mm) before (left) and after test (right)
05-C-0027 and the Bundeswehr Research Institute for Materials, Explosives, Fuels and
Lubricants (WIWEB) under contract nos. E/E210/4D004/X5143 and E/E210/7D002/4F088
is gratefully acknowledged. The work presented in this chapter would not have been pos-
sible without the help of several outstanding coworkers in the LMU research group. The
author especially thanks Prof. Konstantin Karaghiosoff, Dr. habil. Margaret-Jane Craw-
ford, Dr. Jan J. Weigand, Dipl.-Chem. Michael Göbel and Dipl.-Chem. Jörg Stierstorfer
for their help and support. Ms. Carmen Nowak is thanked for drawing all structures and
preparing the figures and Ms. Irene Scheckenbach for her help with a difficult manuscript.
References
1. Klapötke TM, Holl G (2001) Green Chem 3:G75
2. Klapötke TM, Holl G (2002) Chem Aust, p 11
3. Kanekar P, Dautpure P, Sarnaik S (2003) Ind J Exp Biol 41:991
4. Robidoux PY, Gong P, Sarrazin M, Bardai G, Paquet L, Hawari J, Dubois C, Suna-
hara GI (2004) Can Ecotoxicol Environ Safe 58:300
5. Robidoux PY, Sunahara GI, Savard K, Berthelot Y, Dodard S, Martel M, Gong P,
Hawari J (2004) Can Environ Toxicol Chem 23:1026
6. Simini M, Checkai RT, Kuperman RG, Phillips CT, Kolakowski JE, Kurnas CW, Suna-
hara GI (2003) Pedobiologia 47:657
7. Robidoux PY, Hawari J, Bardai G, Paquet L, Ampleman G, Thiboutot S, Sunahara GI
(2002) Can Arch Environ Contam Toxicol 43:379
8. Steevens JA, Duke BM, Lotufo GR, Bridges TS (2002) Environ Toxicol Chem 21:1475
9. Pennington JC, Brannon JM (2002) Thermochim Act 384:163
10. Fournier D, Halasz A, Spain J, Spanggord RJ, Bottaro JC, Hawari J (2004) Appl
Environment Microbiol 70:1123
11. Doherty RM (2003) In: De Luca LT, Galfetti L, Pesce-Rodriguez RA (eds) Novel en-
ergetic materials and applications. Proceedings of the 9th IWCP, Lerici, La Specia,
Italy
12. Karaghiosoff K, Klapötke TM, Michailovski A, Nöth H, Suter M (2003) Prop Explos
Pyrotech 28:1
13. Klapötke TM, Krumm B, Holl G, Kaiser M (1999) Proc of 30th int annual conference
of ICT, June 29–July 2, Karlsruhe, Germany, p 120
14. Hammerl A, Klapötke TM, Nöth H, Warchhold M, Holl M, Kaiser M (2001) Inorg
Chem 40:3570
15. Klapötke TM (2007) Nichtmetallchemie. In: Riedel E (ed) Moderne Anorganische
Chemie, 3rd edn. Walter de Gruyter, Berlin
16. Eremets MI, Gavriliuk AG, Serebryanaya NR, Trojan IA, Dzivenko DA, Boehler R,
Mao HK, Hemley RJ (2004) J Chem Phys 121:11296
17. Eremets MI, Gavriliuk AG, Trojan IA, Dzivenko DA, Boehler R (2004) Nat Mater
3:558
18. Eremets MI, Popov MY, Trojan IA, Denisov VN, Boehler R, Hemley RJ (2004) J Chem
Phys 120:10618
19. Christe KO, Wilson WW, Sheehy JA, Boatz JA (1999) Angew Chem Int Ed 38:2004
20. Vij A, Wilson WW, Vij V, Tham FS, Sheehy JA, Christe KO (2001) J Am Chem Soc
123:6308
21. Klapötke TM (1999) Angew Chem 111:2694
22. Schroer T, Haiges R, Schneider S, Christe KO (2005) Chem Comm, p 1607
New Nitrogen-Rich High Explosives 119
23. Vij A, Pavlovich JG, Wilson WW, Vij V, Christe KO (2002) Angew Chem Int Ed
41:3051
24. Lauderdale WJ, Stanton JF, Bartlett RJ (1992) J Phys Chem 96:1173
25. Perera SA, Bartlett RJ (1999) Chem Phys Lett 314:381
26. Tobita M, Bartlett RJ (2001) J Phys Chem A 105:4107
27. Glukhovtsev MN, Jiao H, Schleyer PVR (1996) Inorg Chem 35:7124
28. Glukhovtsev MN, Schleyer PVR (1992) Chem Phys Lett 198:547
29. Klapötke TM (2000) J Mol Struct (THEOCHEM) 499:99
30. Klapötke TM, Harcourt RD (2001) J Mol Struct (THEOCHEM) 541:237
31. Schmidt MW, Gordon MS, Boatz JA (2005) J Phys Chem A 109:7285
32. Wang R, Gao H, Ye C, Twamley B, Shreeve JM (2007) Inorg Chem 46:932
33. Geith J, Klapötke TM, Weigand JJ, Holl H (2004) Prop Explos Pyrotech 29:3
34. Klapötke TM, Mayer P, Schulz A, Weigand JJ (2005) J Am Chem Soc 127:2032
35. Fischer G, Holl G, Klapötke TM, Weigand JJ (2005) Thermochim Acta 437:168
36. Raap R (1969) Can J Chem 47:3677
37. Gaponik PN, Karavai VP (1984) Khim Geterotsikl Soedin 12:1683
38. Stolle R, Netz H, Kramer O, Rothschild S, Erbe E, Schick O (1933) J Prak Chem 138:1
39. Gálvez-Ruiz JC, Holl G, Karaghiosoff K, Klapötke TM, Löhnwitz K, Mayer P, Nöth H,
Polborn K, Rohbogner CJ, Suter M, Weigand JJ (2005) Inorg Chem 44:4237
40. Darwich C, Klapötke TM (2006) New trends in research of energetic materials. Pro-
ceedings of the 9th seminar, Pardubice, Czech Republic, p 551
41. Gálvez-Ruiz JC, Holl G, Karaghiosoff K, Klapötke TM, Löhnwitz K, Mayer P, Nöth N,
Polborn K, Rohbogner CJ, Suter M, Weigand JJ (2005) Inorg Chem 44:5192
42. Boese R, Klapötke TM, Mayer P, Verma V (2006) Prop Explos Pyrotech 31:263
43. Murray WM, Sauer WC (Arthur D, Little Inc) (1961) US Patent 3006957
44. Marans NS, Zelinski RP (1950) J Am Chem Soc 72:5329
45. Göbel M, Klapötke TM (2007) New trends in research of energetic materials. Pro-
ceedings of the 10th seminar, Pardubice, Czech Republic, p L13
46. Adam D, Karaghiosoff K, Holl G, Kaiser M, Klapötke TM (2002) Prop Explos Py-
rotech 27:7
47. Karaghiosoff K, Klapötke TM, Michailovski A, Nöth H, Suter M (2003) Prop Explos
Pyrotech 28:1
48. Klapötke TM, Mayer P, Verma V (2006) Prop Explos Pyrotech 31:263
49. Deal WE (1957) J Chem Phys 27:796
50. Mader CL (1963) Report LA-2900: Fortran BKW code for computing the detonation
properties of explosives. Los Alomos Scientific Laboratory, NM
51. Urbanski T (1985) Chemistry and technology of explsoives. Pergamon, England
52. Astakhov AM, Vasilev AD, Molokeev MS, Revenko VA, Stepanov RS (2005) Russ
J Org Chem 41:910
53. Klapötke TM, Stierstorfer J (2007) New trends in research of energetic materials.
Proceedings of the 10th seminar, Pardubice, Czech Republic, p 35
54. Bryden JH (1953) Acta Cryst 6:669
55. Karahiosoff K, Klapötke TM, Mayer P, Piotrowski H, Polborn K, Willer RL, Wei-
gand JJ (2005) J Org Chem 71:1295
56. Tappan BC, Beal RW, Brill TB (2002) Thermochim Act 288:227
57. Tappan BC, Incarnito CD, Rheingold AL, Brill TB (2002) Thermochim Act 384:113
58. Brill TB, Tappan BC, Beal RW (2001) New trends in research of energetic materials.
Proceedings of the 4th seminar, Pardubice, Czech Rep, p 17
59. Thiele J (1892) Ann 270:1
60. Herbst RM, Garrison JA (1953) J Org Chem 18:941
120 T.M. Klapötke
61. Lieber E, Sherman E, Henry RA, Cohen J (1951) J Am Chem Soc 73:2327
62. Astachov AM, Nefedo AA, Vasiliev AD, Kruglyakova LA, Dyugaev KP, Stepanov RS
(2005) Proc of 36th int annual conference of ICT, Jun 28–July 1, Karlsruhe, Germany,
p 113
63. Lieber E, Sherman E, Henry RA, Cohen J (1951) J Am Chem Soc 73:2327
64. O’Connor TE, Fleming G, Reilly J (1949) J Soc Chem Ind (London) 68:309
65. Mayants AG, Klimenko VS, Erina VV, Pyreseva KG, Gordeichuk SS, Leibzon VN,
Kuz’min VS, Burtsev UN (1991) Khim Geterot Soed 8:1067
66. Göbel M, Klapötke TM, Mayer E (2006) Z Anorg Allg Chem 2632:1043
67. Thiele J (1892) Ann 270:1
68. Henry RA, Finnegan WG (1954) J Am Chem Soc 76:923
69. Henry RA, Finnegan WG (1956) J Am Chem Soc 78:411
70. Oxford Diffraction (2005) CrysAlis CCD, Version 1.171.27p5 beta (release 01-04-2005
CrysAlis171.NET)
71. Oxford Diffraction (2005) CrysAlis RED, Version 1.171.27p5 beta (release 01-04-2005
CrysAlis171.NET)
72. Altomare A, Cascarano G, Giacovazzo C, Guagliardi A (1993) SIR-92, A program for
crystal structure solution. J App Cryst 26:343
73. Sheldrick GM (1994) SHELXL97 program for the refinement of crystal structures.
University of Göttingen, Germany
74. Spek AL (1999) PLATON, a multipurpose crystallographic tool. Utrecht University,
Utrecht, The Netherlands
75. Oxford Diffraction (2005) SCALE3 ABSPACK (1.0.4,gui:1.0.3) (C)
76. Cambridge Crystallographic Data Centre (2007) http://www.ccdc.cam.ac.uk/, last
visited: 30 Mar 2007
77. Lieber E, Patinkin T (1951) J Am Chem Soc 73:1792
78. Bray DD, White JG (1979) Acta Cryst B 35:3089
79. Riedel E (1999) Anorganische Chemie, 4th edn. Walter de Gruyter, Berlin, p 134
80. Lieber E, Levering DR, Patterson LJ (1951) Anal Chem 23:1594
81. Weigand JJ (2005) Dissertation, Ludwig Maximilian University Munich
82. Daszkiewicz Z, Nowakowska EM, Preźdo WW, Kyzioł JB (1995) Pol J Chem 69:1437
83. PerkinElmer (2007) http://www.perkinelmer.com, last visited: 30 Mar 2007
84. Linseis (2007) http://www.linseis.com, last visited: 30 Mar 2007
85. United Nations Economic Commission for Europe (2005) UN recommendations
on the transport of dangerous goods, 14th edn. http://www.unece.org/trans/danger/
publi/unrec/rev14/14files_.html, last visited: 30 Mar 2007
86. Bundesanstalt für Materialforschung und -prüfung (2007) http://www.bam.de, last
visited: 30 Mar 2007
87. Parr Instrument Company (2007) http://www.parrinst.com, last visited: 30 Mar 2007
88. West RC, Selby SM (eds) (1967–1968) Handbook of chemistry and physics, 48th edn.
CRC, Cleveland, OH
89. McEwan WS, Rigg MW (1951) J Am Chem Soc 73:4725
90. Ostrovskii VA, Pevzner MS, Kofman TP, Tselinskii IV (1999) Targets Heterocycl Syst
3:467
91. Suceska M (1999) Proc of 30th int annual conference of ICT, June 29–July 2, Karl-
sruhe, Germany, p 50
92. Suceska M (2001) EXPLO5.V2: computer program for calculation of detonation pa-
rameters. Proc of 32nd int annual conference of ICT, July 3–6, Karlsruhe, German,
p 110
93. Suceska M (1991) Prop Explos Pyrotech 16:197
New Nitrogen-Rich High Explosives 121
94. Mecke R, Langenbucher F (1965) Infrared spectra. Heyden, London, Serial no. 6
95. Shimanouchi T (1972) Tables of molecular vibrational frequencies consolidated,
vol II. J Phys Chem Ref Data 6:993
96. National Institue of Standards and Technology (2007) Vibrational energy search
http://webbook.nist.gov/chemistry/vib-ser.html, last visited: 30 Mar 2007
97. Nakamoto K (1986) Infrared and Raman Spectra of inorganic and coordination
compounds, 4th edn. Wiley, New York
98. Hypercube (2002) HyperChem 7.52: Molecular visualization and simulation pro-
gram package. Hypercube, Gainsville, FL
99. Murray JS, Lane P, Politzer P (1995) Mol Phys 85:1
100. Murray JS, Lane P, Politzer P (1998) Mol Phys 93:187
101. Politzer P, Murray JS (1999) Computational characterization of energetic materials.
In: Maksic ZB, Orville-Thomas WJ (eds) Pauling’s legacy: modern modelling of the
chemical bond. Theor Comput Chem 6:347
102. Politzer P, Murray JS, Seminario JM, Lane P, Grice ME, Concha MC (2001) J Mol
Struct (THEOCHEM) 573:1
103. Rice BM, Chabalowski CF, Adams GF, Mowrey RC, Page M (1991) Chem Phys Lett
184:335
104. Rice BM, Hare JJ (2002) J Phys Chem A 106:1770
105. Rice BM, Sahu S, Owens FJ (2002) J Mol Struct (THEOCHEM) 583:69
106. Rice BM (2005) Adv Ser Phys Chem 16:33
107. Systag (2007) Process development and safety http://www.systag.ch, last visited: 30
Mar 2007
108. Karaghiosoff K, Klapötke TM, Mayer P, Piotrowski H, Polborn K, Willer RL, Wei-
gand JJ (2005) J Org Chem 71:1295
109. Geith J, Klapötke TM, Weigand JJ, Holl G (2004) Prop Explos Pyrotech 29:3
110. Geith J, Holl G, Klapötke TM, Weigand JJ (2004) Combust Flame 139:358
Struct Bond (2007) 125: 123–151
DOI 10.1007/430_2007_058
© Springer-Verlag Berlin Heidelberg
Published online: 12 May 2007
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
2 Historical Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
3 A New Hope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
3.1 Cyclic Difluoramino-nitramines . . . . . . . . . . . . . . . . . . . . . . . . 126
3.1.1 HNFX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
3.1.2 4,4-Bis(difluoramino)-1-nitropiperidine . . . . . . . . . . . . . . . . . . . . 131
3.1.3 RNFX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
3.1.4 TNFX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
3.1.5 Properties and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . 134
3.2 Electrophilic Difluoramination and Its Products . . . . . . . . . . . . . . . 135
3.2.1 Syntheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.2.2 Thermochemical and Other Properties . . . . . . . . . . . . . . . . . . . . 137
3.3 Other Organic Difluoramines . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3.3.1 Polymers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3.3.2 Plasticizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
3.3.3 Miscellaneous Organic Derivatives . . . . . . . . . . . . . . . . . . . . . . . 140
4 New Synthetic Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . 143
5 Theoretical Treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Abstract This article reviews recent developments in synthesis and characterizations of or-
ganic difluoramine (NF2 ) derivatives since the 1980s. Specific classes of chemical compounds
are covered: cyclic difluoramino-nitramines, especially gem-bis(difluoramino)-substituted
analogs of conventional nitramines; products from the relatively new transformation of elec-
trophilic difluoramination; other classes of difluoramines such as polymers, plasticizers,
and compounds not otherwise categorized. Also covered are aspects of other recently de-
veloped synthetic methodology to prepare compounds in this general class, as well as purely
theoretical treatments of the chemistry and properties of this class.
Abbreviations
BNMO 3,3-Bis(nitratomethyl)oxetane
HMX Octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine
124 R.D. Chapman
HNFX 3,3,7,7-Tetrakis(difluoramino)octahydro-1,5-dinitro-1,5-diazocine
NMMO 3-Methyl-3-(nitratomethyl)oxetane
PETN Pentaerythrityl tetranitrate
RDX Hexahydro-1,3,5-trinitro-1,3,5-triazine
RNFX 5,5-Bis(difluoramino)hexahydro-1,3-dinitropyrimidine
TNFX 3,3-Bis(difluoramino)octahydro-1,5,7,7-tetranitro-1,5-diazocine
1
Introduction
1 This review will use the colloquial American English term “difluoramino” rather than the (admit-
tedly more systematic) equivalent “difluoroamino”, which is more commonly used in non-English
publications and English translations of non-English (especially Russian) publications.
Organic Difluoramine Derivatives 125
ume additivity method [12], NF2 has a substituent density of 2.303 g cm–3
(52.0035 amu/37.504 Å3 ) vs. an NO2 density of 2.166 g cm–3 (46.005 amu/
35.265 Å3 ).
2
Historical Development
3
A New Hope
in the late 1980s, Zheng et al. (Xi’an Modern Chemistry Institute, China) al-
luded, in a conference proceeding, to preparing at least one member of this
class [28]: “For example, the difluoramination of cyclic derivatives of 1,3-
diaminoacetone (the substituents on the nitrogen atoms were –NO2 ... etc.)
is easier to be carried out than that of the chain derivatives.” However, no
specific example was described, and subsequent journal reports [29, 30] of
other chemistry from this conference proceeding made no mention of cyclic
derivatives.
This section will review a variety of experimental results arising from this
renaissance period in difluoramine chemistry.
3.1
Cyclic Difluoramino-nitramines
3.1.1
HNFX
Synthesis
In the first synthesis of HNFX, the use of a nitramine reactant, 5-(difluor-
amino)-3,7-dinitro-9-oxa-3,7-diazabicyclo[3.3.1]nonan-1-ol (2), in a typi-
cal difluoramination reaction (difluoramine–difluorosulfamic acid–sulfuric
acid) produced HNFX in only ∼ 1% yield [34] (Scheme 1).
Scheme 3 Scalable preparation of HNFX (1): protolytic N-denosylation plus simple nitra-
tion
Chapman et al. [36] had suggested that solvents with less ozone-depleting
potential than CFCl3 could be used, and a preliminary demonstration of
difluoramination with dichloromethane solvent was described. Adolph and
Stern more recently chose n-pentane as an example of an alternative cosolvent
(with CFCl3 ) for this purpose [38].
Organic Difluoramine Derivatives 129
Structure
X-ray crystallographic analysis of HNFX recrystallized from a variety of
solvent systems showed an interesting, unpredicted feature (Fig. 1): chan-
nels, with a threefold axis of symmetry surrounded by HNFX molecules,
passing through each unit cell [36]. Analysis sometimes showed the pres-
ence of disordered mass due to recrystallization solvent(s), though solvent-
free crystals can be prepared by driving out solvent (e.g., by heating under
vacuum).
Fig. 1 Crystal packing of HNFX in the trigonal crystal form; empty channels occur along
threefold axes
a density of 2.027 g cm–3 recently predicted by Ammon [39] from its pos-
sible crystal packings. However, cyclic nitramines, especially those incorpo-
rating eight-membered rings, are well known to exhibit polymorphism [40,
41]: HMX has four known polymorphs of different densities. The phe-
nomenon of solvent inclusion observed in HNFX crystals is also qualitatively
similar to that found in the first polymorph of hexanitrohexaazaisowurtz-
itane (α-CL-20) [42], which is the lowest-density form. With the promise
of preparing a higher-density polymorph of HNFX, which would dramat-
ically improve its attractiveness as a munitions ingredient, several tech-
nical approaches have been undertaken to effect such a phase transition.
Peiris monitored HNFX for phase transitions during compression in a di-
amond anvil cell at pressures up to 5 GPa but observed none [43]. Nicol
and coworkers subjected HNFX to pressures up to 30 GPa in a diamond
anvil cell and studied it by synchrotron radiation X-ray diffraction and Ra-
man and infrared spectroscopy at ambient temperature, also without ob-
serving evidence of a polymorphic transition [44]. More recently, a com-
bined experimental and modeling study of the vibrational spectra of solid
HNFX was reported [45]. Kalyon and coworkers have performed a system-
atic study of recrystallizations of HNFX with several different solvent sys-
tems [46], similarly without evidence of a new polymorph, though they
independently computationally corroborate the feasible existence of a higher-
density form [47]. While these failures to achieve formation of a new poly-
morph of HNFX might indicate that its manifest density is that observed
in the crystals prepared to date, they may instead reflect the recognized
poor understanding, in general, of how to predict and prepare new stable
polymorphs [48].
Hazard Properties
Sensitivity measurements on HNFX (the “polymorph” tentatively called α-
HNFX) show it to be comparable to CL-20 and PETN on impact but more
sensitive by friction and electrostatic discharge [49]. Impact sensitivity (Ex-
plosives Research Laboratory impact tester, modified Type 12 tooling, 2.5-
kg weight): HNFX h50 = 8 cm; CL-20 ∼ 8 cm; PETN = 10–13 cm; RDX =
16–17 cm. Friction sensitivity (Allegany Ballistics Laboratory test, per MIL-
STD-1751, threshold initiation value): HNFX fired at 72 lbf; PETN low-fires
ranged 159–316 lbf. Electrostatic sensitivity: HNFX fired at 0.05 J; PETN
showed 10/10 no-fires at 0.25 J. DSC (2 ◦ C min–1 per ASTM E537) onset/peak:
221/231 ◦ C.
However, it is also well known that sensitivities of the polymorphs of an-
other eight-membered-ring nitramine, HMX, correlate inversely with dens-
ity [50–52]: the lowest-density δ form exhibits the highest sensitivity. If this
correlation transfers to HNFX, a higher-density “β-HNFX” would exhibit
lower sensitivity than “α-HNFX”.
Organic Difluoramine Derivatives 131
3.1.2
4,4-Bis(difluoramino)-1-nitropiperidine
3.1.3
RNFX
3.1.4
TNFX
Another early conceptual target member of this class was one in which one
each of two nitramine groups of HMX was replaced by a gem-bis(difluor-
amino) grouping and a gem-dinitro grouping. This would provide a prod-
uct of better oxygen balance than HNFX but still some NF2 content. The
introduction of asymmetric components into a 1,5-diazocine substrate suit-
able for the incorporation of three different energetic functionalities was
a formidable challenge, which was undertaken by Axenrod et al. [61]; the suc-
cessful synthetic sequence is shown below (Scheme 5). The ultimate nitrolysis
step proved inordinately difficult because of steric and electronic deactivating
effects of C(NF2 )2 and C(NO2 )2 . Thus, the reaction required not only a su-
peracid (trifluoromethanesulfonic acid) as a medium for nitric acid in order
to generate protonitronium reagent, but also the addition of a strong Lewis
acid, antimony pentafluoride, in order to increase the superacid strength to
generate a higher concentration of NO2 H2+ . Upon its successful preparation,
the new product (12) was named TNFX (a tetranitro derivative in the series of
NF2 -modified cyclic nitramine explosives).
3.1.5
Properties and Applications
3.2
Electrophilic Difluoramination and Its Products
3.2.1
Syntheses
3.2.2
Thermochemical and Other Properties
Pepekin’s first report [74, 75] of products in this class gave experimen-
tal values of thermodynamic properties such as heats of formation. Also
prior to the synthetic methodology descriptions [80, 81], Litvinov et al. [76,
77] reported various physical properties (melting point and/or boiling
point, density) of several members of this class, including a particularly
dense and energetic one: bis[2-(difluoramino)-2,2-dinitroethyl]nitramine,
[NF2 C(NO2 )2 CH2 ]2 N–NO2 (13). This compound had an attractive dens-
ity of 2.045 g cm–3 , a melting point of 103 ◦ C, and thermal stability bet-
ter than bis(2,2,2-trinitroethyl)nitramine. However, the authors made an
assessment of these members of this new class: “Nevertheless, it should
be noted that the considered compounds exhibit a rather high sensitiv-
ity to the action of external factors, and thus have no practical appli-
cation” [76, 77]. Indeed, Shchetinin [92, 93] later reported that bis[2-(di-
fluoramino)-2,2-dinitroethyl]nitramine (13) had a sensitivity comparable to
the primary explosive lead azide: the critical pressure of explosion excita-
tion measured by the disintegrating shell method was pDS cr = 0.42 ± 0.02 for
DS
[NF2 C(NO2 )2 CH2 ]2 N–NO2 (13) compared to pcr = 0.38 ± 0.03 for lead azide;
pDS DS
cr = 0.93 ± 0.03 for PETN; and pcr = 1.15 ± 0.03 for RDX.
Nazin and coworkers measured thermal decomposition of α-(difluor-
amino)polynitroalkanes and concluded that (difluoramino)dinitroalkanes
differ little in their stability from their nitro analogs, trinitroalkanes [94, 95].
Bis[2-(difluoramino)-2,2-dinitroethyl]nitramine (13) was not one of the ex-
amples in this study, however.
Pepekin [96] measured and reported experimental detonation parameters
and the metal acceleration ability of bis[2-(difluoramino)-2,2-dinitroethyl]-
nitramine (13), as well as compositions of it with HMX and with aluminum,
and compared them to these parameters for HMX as well as formulations of
bis(2,2,2-trinitroethyl)nitramine. The highest metal acceleration ability was
offered by a 60 : 40 composition of 13 with HMX, which performed at 110.8%
of HMX. Pure 13 also exhibited a higher detonation velocity than HMX. Lit-
erature involving this class of compound has mostly subsided in recent years,
though there is another report by Pepekin and Gubin [97, 98] that 13 ex-
hibits propellant performance that is 105.2% of that of HMX, measured by the
“M-40 technique” (acceleration of steel plates 40 mm in diameter and 4 mm
thick accelerated from the end of a cylindrical 40 × 40 mm charge placed in
a thick-walled steel shell). Hexanitrobenzene is reported as having a perform-
ance 106% of that of HMX.
138 R.D. Chapman
3.3
Other Organic Difluoramines
3.3.1
Polymers
3.3.2
Plasticizers
with PETN [107]. The mixture compressed in a cubic zirconia anvil cell was
heated by absorption of a single laser pulse (Nd : YAG frequency-doubled
laser at 532 nm), followed by recooling of the sample. Products observed at
1.7 GPa and 50 K included O=B–F, identified by infrared absorptions.
3,3-Bis(difluoramino)-1,5-dinitratopentane (17), as a potentially useful
plasticizer, was prepared by Adolph and Trivedi [108] by conventional di-
fluoramination of 1,5-bis(trifluoroacetoxy)-3-pentanone followed by transes-
terification with methanol, and then nitration with mixed acid (Scheme 13).
3.3.3
Miscellaneous Organic Derivatives
Synthesis
Fokin et al. directly fluorinated several aminofurazans (Scheme 14) to pre-
pare 3-(difluoramino)furazans [109, 110]. Similarly, certain N-alkyl-5-amino-
tetrazoles were amenable to direct fluorination to 5-(difluoramino)tetrazole
derivatives [111, 112]. NF2 -containing 1,2,4-oxadiazoles were prepared by
Shreeve and coworkers by reactions of sodium 5-[(difluoraminomethyl)di-
fluoromethyl]tetrazolate with oxalyl chloride [113].
1,1,2,2-Tetrakis[N-(difluoraminomethyl)nitramino]ethane (18) was first
reported by Zheng et al. in a 1985 conference [114], and its chemistry
(Scheme 15) was described further in later publications [28–30].
An interesting example of a unique product of difluoramination of
a heterocyclic imide carbonyl group appeared in the context of a report
by Gilardi and coworkers of structural features among a group of gem-
bis(difluoramino)-substituted organics [115]: 1-{3-[5,5-bis(difluoramino)-
2-oxopyrrolidinyl]-2-oxopropyl}pyrrolidine-2,5-dione (19). Imide carbonyls
are usually not subject to difluoramination under typical reaction conditions.
Organic Difluoramine Derivatives 141
Thermochemistry
Thermochemical characterizations of several organic difluoramines prepared
during the Project Principia era were reported throughout the 1960s and
1970s and will not be reviewed here. Studies of the specific class of products
of electrophilic difluoramination were reviewed above (Sect. 3.2.2).
More recently, two significant papers by Fokin, Grebennikov, Manelis, and
Nazin [120–123] reported kinetic studies of thermal decomposition of a good
number of organic difluoramine derivatives. The structures were chosen in
order to ascertain inductive and steric effects on thermal decomposition, as
well as effects of conjugation within the reactant and the presence of an α-
hydrogen (relative to NF2 ). The authors report that internal alkyl-substituted
difluoramines (i.e., those lacking α-hydrogen) are more stable than analogous
nitro compounds, owing to higher activation energies and somewhat smaller
Organic Difluoramine Derivatives 143
4
New Synthetic Methodology
low (31%) for ketone difluoramination, and its yield by the new transform-
ation maximized at about 10%. 1,1-Bis(difluoramino)propane’s yield (∼ 40%)
was more typical of conventional difluoraminations.
5
Theoretical Treatments
ory, Hartree–Fock, and MP2 methods (6-31G∗ and 6-311G∗∗ basis sets), as
well as semiempirical MO methods MNDO, AM1, and PM3 for the former
chemical system. The semiempirical methods did not produce accurate and
reliable results for heats of formation. Interactions between difluoramino
substituents were determined. Ju and coworkers also calculated (using DFT
and semiempirical methods) heats of formation of a variety of difluoramino
pentaerythrityl derivatives [156], including some that had been considered
by Archibald and Manser [106] as prospective plasticizers, for example,
a tetrakis(difluoramino) analog of PETN, C(CH2 NF2 )4 , as a theoretical struc-
ture of interest.
References
1. Furin GG, Fainzilberg AA (1999) Russ Chem Rev 68:653
2. Taylor SD, Kotoris CC, Hum G (1999) Tetrahedron 55:12431
3. Li Y (2001) Huaxue Shiji [Chemical Reagents] 23:143
4. Colburn CB (1965) Endeavour 24:138
5. Fokin AV, Kosyrev YuM (1966) Russ Chem Rev 35:791
6. Freeman JP (1967) Inorg Chim Acta Rev 1:65
7. Ruff JK (1967) Chem Rev 67:665
8. Klapötke TM (2006) J Fluorine Chem 127:679
9. Politzer P, Lane P (1997) Adv Mol Struct Res 3:269
10. Brown RC, Kolb CE, Yetter RA, Dryer FL, Rabitz H (1995) Combust Flame 101:221
11. Yetter RA, Dryer FL, Rabitz H, Brown RC, Kolb CE (1998) Combust Flame 112:387
12. Ammon HL (2001) Struct Chem 12:205
13. Ruff O, Giese M (1936) Ber Dtsch Chem Ges B 69B:598
14. Davenas A (2003) J Propul Power 19:1108
15. Holzmann RT (ed) (1966) Advanced propellant chemistry. American Chemical Soci-
ety, Washington, DC
16. Fokin AV, Kosyrev YuM (1970) Zh Vses Ova im D I Mendeleeva 15:81
17. Fokin AV, Kosyrev YuM (1970) Mendeleev Chem J 15:109
18. Baum K (1971) Intra-Sci Chem Rep 5:69
19. Fokin AV, Studnev YuN, Kuznetsova LD (1976) Reakts Metody Issled Org Soedin 24:7
20. Fokin AV, Studnev YuN (1982) Izv Akad Nauk SSSR Ser Khim 1812
21. Fokin AV, Studnev YuN (1982) Bull Acad Sci USSR Div Chem Sci 31:1609
22. Fokin AV, Kosyrev YuM, Shevchenko VI (1982) Izv Akad Nauk SSSR Ser Khim 1831
23. Fokin AV, Kosyrev YuM, Shevchenko VI (1982) Bull Acad Sci USSR Div Chem Sci
31:1626
24. Urbański T (1984) Chemistry and technology of explosives, vol 4, chap 9. Pergamon,
Oxford
25. Hanefeld W (1990) N,N-Dihalogen-amine. In: Klamann D (ed) Methoden der or-
ganischen chemie (Houben-Weyl), vol E16a, organische Stickstoff-Verbindungen I,
Part 2. Georg Thieme, Stuttgart, p 893
26. Miller RS (1996) Mater Res Soc Symp Proc 418:3
27. Chapman RD, Archibald TG, Baum K (1989) Research in energetic compounds.
Report ONR-7-1 (interim). Fluorochem, Azusa, CA; NTIS accession number
Organic Difluoramine Derivatives 147
50. Cady HH, Smith LC (1962) Studies on the polymorphs of HMX. Report LAMS-2652.
Los Alamos Scientific Laboratory, Los Alamos, NM
51. Licht HH (1971) In: Hansson J (ed) Proceedings of symposium on chemical prob-
lems connected with the stability of explosives. Sektionen för Detonik och Förbrän-
ning, Sundbyberg, Sweden, p 168
52. McCrone WC (2001) Microscope 49:47
53. Zhang J, Oxley J, Smith J, Bedford C, Chapman R (2000) J Mass Spec 35:841
54. Oxley JC, Smith JL, Zhang J, Bedford C (2001) J Phys Chem A 105:579
55. Kuo KK, Young G (2002) Proc Combust Inst 29:2947
56. Chapman RD, Nguyen BV (2001) US Patent 6 310 204
57. Emmons WD, Freeman JP (1955) J Am Chem Soc 77:6061
58. Luk’yanov OA, Ternikova TV (1983) Izv Akad Nauk SSSR Ser Khim 667
59. Luk’yanov OA, Ternikova TV (1983) Bull Acad Sci USSR Div Chem Sci 32:605
60. Garcia J, González J, Segura R, Urpí F, Vilarrasa J (1984) J Org Chem 49:3322
61. Axenrod T, Guan X-P, Sun J, Qi L, Chapman RD, Gilardi RD (2001) Tetrahedron Lett
42:2621
62. Park Y-D, Cho S-D, Kim J-J, Kim H-K, Kweon D-H, Lee S-G, Yoon Y-J (2006) J Hete-
rocycl Chem 43:519
63. Gilardi R (2004) Polytypical polymorphs occurring in an energetic material. In: Vir-
tual course on polymorphism: diversity amidst similarity, International School of
Crystallography – 35th course, Erice, Sicily, Italy; http://erice2004.docking.org/
vcourse/polymorph/15tue/1830-Gilardi/Gilardi.ppt, last visited: 20 April 2007
64. Dunn BP (2001) Rocket engine specific impulse program. http://dunnspace.com/
isp.htm, last visited: 20 April 2007
65. Hong W-L, Tian D-Y, Liu J-H, Wang F (2001) Guti Huojian Jishu [J Solid Rocket
Technol] 24:41
66. Hong W-L, Tian D-Y, Liu J-H, Wang F (2001) Chem Abstr 135:275005
67. Politzer P, Murray JS, Grice ME, Sjoberg P (1991) Computer-aided design of mono-
propellants. In: Olah GA, Squire DR (eds) Chemistry of energetic materials, chap 4.
Academic, San Diego, CA
68. Feng Z (2000) Huaxue Jinzhan [Prog Chem] 12:171
69. Feng Z (2000) Chem Abstr 134:88422
70. Tan G-H (2000) 2 [Trends in Explosives]
71. Li S-W, Zhao F-Q, Yuan C, Luo Y, Gao Y (2002) Guti Huojian Jishu [J Solid Rocket
Technol] 25:36
72. Li S-W, Zhao F-Q, Yuan C, Luo Y, Gao Y (2002) Chem Abstr 138:6183
73. Kim YG (2004) BK21 Program in Chemical Engineering, Seoul National Uni-
versity, Newsletter 5(1):22, http://web.archive.org/web/20041112214526/http://bk21-
chem-eng.snu.ac.kr /news/newsletter/2004_spring/5-1.pdf, last visited: 23 April 2007
74. Pepekin VI (1994) Khim Fiz 13:42
75. Pepekin VI (1994) Chem Phys Rep 13:67
76. Litvinov BV, Fainzil’berg AA, Pepekin VI, Smirnov SP, Loboiko BG, Shevelev SA,
Nazin GM (1994) Dokl Akad Nauk 336:67
77. Litvinov BV, Fainzil’berg AA, Pepekin VI, Smirnov SP, Loboiko BG, Shevelev SA,
Nazin GM (1994) Dokl Chem 336:86
78. Dalinger IL, Vinogradov VM, Shevelev SA, Kuz’min VS (1996) Mendeleev Commun
6:13
79. Dalinger IL, Vinogradov VM, Shevelev SA, Kuz’min VS, Arnautova EA, Pivina TS
(1998) Propellants Explos Pyrotech 23:212
Organic Difluoramine Derivatives 149
80. Fokin AV, Studnev YuN, Kuznetsova LD (1996) Dokl Akad Nauk 346:358
81. Fokin AV, Studnev YuN, Kuznetsova LD (1996) Dokl Chem 346:22
82. Fokin AV, Studnev YuN, Rapkin AI, Kuznetsova LD (1996) Izv Akad Nauk Ser Khim
2689
83. Fokin AV, Studnev YuN, Rapkin AI, Kuznetsova LD (1996) Russ Chem Bull 45:2547
84. Fokin AV, Studnev YuN, Kuznetsova LD (1996) Izv Akad Nauk Ser Khim 2056
85. Fokin AV, Studnev YuN, Kuznetsova LD (1996) Russ Chem Bull 45:1952
86. Fokin AV, Studnev YuN, Stolyarov VP, Valiev RSh (1999) Izv Akad Nauk Ser Khim
130
87. Fokin AV, Studnev YuN, Stolyarov VP, Valiev RSh (1999) Russ Chem Bull 48:131
88. Khisamutdinov GKh, Slovetsky VI, Golub YuM, Shevelev SA, Fainzil’berg AA (1997)
Izv Akad Nauk Ser Khim 338
89. Khisamutdinov GKh, Slovetsky VI, Golub YuM, Shevelev SA, Fainzil’berg AA (1997)
Russ Chem Bull 46:324
90. Khisamutdinov GKh, Shevelev SA (2001) Izv Akad Nauk Ser Khim 706
91. Khisamutdinov GKh, Shevelev SA (2001) Russ Chem Bull 50:736
92. Shchetinin VG (1999) Fiz Goreniya Vzryva 35:116
93. Shchetinin VG (1999) Combust Explos Shock Waves (Engl Transl) 35:570
94. Grebennikov VN, Nazin GM, Manelis GB (1995) Izv Akad Nauk Ser Khim 649
95. Grebennikov VN, Nazin GM, Manelis GB (1995) Russ Chem Bull 44:628
96. Pepekin V (1996) Development of high-efficiency energetic explosives. 27th interna-
tional annual conference of ICT, 25–28 June 1996, Karlsruhe, Germany,
97. Pepekin VI, Gubin SA (2007) Fiz Goreniya Vzryva 43:99
98. Pepekin VI, Gubin SA (2007) Combust Explos Shock Waves (Engl Transl) 43:84
99. Jiang Z, Xu G, Chen Y (1990) 21st International Annual Conference of ICT, Karls-
ruhe, Germany
100. Gilardi R, Evans RN, Manser GE (2003) Acta Crystallogr Sect E Struct Rep Online
E59:o2032
101. Archibald TG, Manser GE, Immoos JE (1993) US Patent 5 272 249
102. Archibald TG, Manser GE, Immoos JE (1995) US Patent 5 420 311
103. Solomun T, Schimanski A, Sturm H, Illenberger E (2005) Macromolecules 38:4231
104. Grakauskas V, Baum K (1970) J Org Chem 35:1545
105. Archibald TG, Manser GE (1994) PCT Patent WO 94 /05 643
106. Archibald TG, Manser GE (1998) US Patent 5 789 617
107. Rice JK, Russell TP (1996) Mater Res Soc Symp Proc 418:373
108. Adolph HG, Trivedi NJ (2001) US Patent 6 325 876
109. Fokin AV, Tselinskii IV, Mel’nikova SF, Vergizov SN, Studnev YuN, Stolyarov VP,
Il’in SS (1986) Izv Akad Nauk SSSR Ser Khim 2086
110. Fokin AV, Tselinskii IV, Mel’nikova SF, Vergizov SN, Studnev YuN, Stolyarov VP,
Il’in SS (1986) Bull Acad Sci USSR Div Chem Sci 35:1901
111. Fokin AV, Studnev YuN, Stolyarov VP, Mel’nikov AA (2000) Izv Akad Nauk Ser Khim
950
112. Fokin AV, Studnev YuN, Stolyarov VP, Mel’nikov AA (2000) Russ Chem Bull 49:949
113. John EO, Kirchmeier RL, Shreeve JM (1990) J Fluorine Chem 47:333
114. Zheng YY, Zhou JZ, Huang TJ, Zhou DL, Zhang MN (1985) J Fluorine Chem 29:216
115. Butcher RJ, Gilardi R, Baum K, Trivedi NJ (2002) Thermochim Acta 384:219
116. Fokin AV, Voronkov AN, Timofeenko IA (1978) Izv Akad Nauk SSSR Ser Khim 2644
117. Fokin AV, Voronkov AN, Timofeenko IA (1978) Bull Acad Sci USSR Div Chem Sci
27:2366
150 R.D. Chapman
118. Prakash GKS, Bae C, Kroll M, Olah GA (2002) J Fluorine Chem 117:103
119. Majumder U, Armantrout JR, Williams RV, Shreeve JM (2002) J Org Chem 67:8435
120. Fokin AV, Grebennikov VN, Manelis GB, Nazin GM (1993) Dokl Akad Nauk 332:
735
121. Fokin AV, Grebennikov VN, Manelis GB, Nazin GM (1993) Dokl Chem 332:247
122. Grebennikov VN, Manelis GB, Nazin GM, Fokin AV (1994) Izv Akad Nauk Ser Khim
336
123. Grebennikov VN, Manelis GB, Nazin GM, Fokin AV (1994) Russ Chem Bull 43:315
124. Manelis GB, Nazin GM, Rubtsov YuI, Strunin VA (2003) Thermal decomposition and
combustion of explosives and propellants. Taylor & Francis, London, chap 7
125. Park J, Chakraborty D, Jamindar S, Xia WS, Lin MC, Bedford C (2002) Thermochim
Acta 384:101
126. Nazin GM, Prokudin VG, Manelis GB (2000) Izv Akad Nauk Ser Khim 231
127. Nazin GM, Prokudin VG, Manelis GB (2000) Russ Chem Bull 49:234
128. Dubikhin VV, Karnauh GE, Lagodzinskaya GV, Matveev VG, Nazin GM, Prokudin
GV (2003) 34th International Annual Conference of ICT, 24–27 June 2003, Karlsruhe,
Germany
129. Matveev VG, Nazin GM (2003) Kinet Catal 44:735
130. Chapman RD, Kreutzberger CB, Welker MF (1995) Difluoramino energetic mate-
rials. JANNAF novel ingredients for liquid and solid propellants specialist session,
4 April 1995, Pasadena, CA; http://stinet.dtic.mil/cgi-bin/GetTRDoc?AD=A451187,
last visited: 20 April 2007
131. Chapman RD, Yee RY, Gilardi RD, Pinkerton AA (1996) Chemistry of difluo-
ramine derivatives. 15th annual working group institute on synthesis of high energy
density materials, 12–13 June 1996,Picatinny Arsenal, NJ; http://stinet.dtic.mil/cgi-
bin/GetTRDoc?AD=A451459, last visited: 20 April 2007
132. Haiges R, Wagner R, Boatz JA, Yousufuddin M, Etzkorn M, Prakash GKS, Christe KO,
Chapman RD, Welker MF, Kreutzberger CB (2006) Angew Chem Int Ed Engl 45:5179
133. Singh RP, Shreeve JM (2001) Chem Commun, p 1196
134. Graham WH, Parker CO (1963) J Org Chem 28:850
135. Fokin AV, Kosyrev YuM, Shevchenko VI, Potarina TM (1971) USSR Patent 299 505
136. Fokin AV, Kosyrev YuM, Shevchenko VI, Potarina TM (1971) Chem Abstr 75:76265
137. Prakash GKS, Etzkorn M, Olah GA, Christe KO, Schneider S, Vij A (2002) Chem
Commun, p 1712
138. Fokin AV, Kosyrev YuM, Makarov VA, Novoselov NP (1969) Dokl Akad Nauk SSSR
186:112
139. Fokin AV, Kosyrev YuM, Makarov VA, Novoselov NP (1969) Proc Acad Sci USSR
Chem Sect 186:350
140. Chapman RD, Davis MC, Gilardi R (2003) Synth Commun 33:4173
141. Inel Y (1976) Bogazici Univ Derg Temel Bilimler Kim 4/5:9
142. Inel Y (1977) Chem Abstr 92:170261
143. Leroy G, Sana M, Wilante C, Peeters D, Bourasseau S (1989) J Mol Struct:
THEOCHEM 187:251
144. Sana M, Leroy G, Wilante C, Peeters D, Bourasseau S (1991) J Energ Mater 9:137
145. Sjoberg P, Murray JS, Brinck T, Politzer P (1990) Can J Chem 68:1440
146. Haeberlein M, Murray JS, Brinck T, Politzer P (1992) Can J Chem 70:2209
147. Politzer P, Lane P, Grice ME, Concha MC, Redfern PC (1995) J Mol Struct:
THEOCHEM 338:249
148. Politzer P, Lane P, Sjoberg P, Grice ME, Shechter H (1995) Struct Chem 6:217
Organic Difluoramine Derivatives 151
149. Politzer P, Murray JS, Grice ME (1996) Mater Res Soc Symp Proc 418:55
150. Politzer P, Lane P (1996) J Mol Struct: THEOCHEM 388:51
151. Politzer P, Lane P, Grice ME (1996) J Mol Struct: THEOCHEM 365:89
152. Politzer P, Murray JS, Seminario JM, Lane P, Grice ME, Concha MC (2001) J Mol
Struct: THEOCHEM 573:1
153. Politzer P, Lane P, Concha MC (2003) Computational approaches to heats of for-
mation. In: Politzer P, Murray JS (eds) Energetic materials, part 1. Decomposition,
crystal and molecular properties, chap 9. Elsevier, Amsterdam
154. Ju X-H, Li Y-M, Xiao H-M (2005) J Phys Chem A 109:934
155. Ju X-H, Ji G-F, Xiao H-M (2006) Chem Phys 326:395
156. Fan X-W, Ju X-H, Xiao H-M, Qiu L (2006) J Mol Struct: THEOCHEM 801:55
Struct Bond (2007) 125: 153–194
DOI 10.1007/430_2006_053
© Springer-Verlag Berlin Heidelberg
Published online: 18 January 2007
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Abbreviations
A7 Strukturbericht designation for α-arsenic crystal lattice structure
B3LYP Becke 3-parameter hybrid density functional using non-local correla-
tion provided by Lee, Yang and Parr
BP Black phosphorus crystal lattice structure. Strukturbericht designation
is A17
CCSD(T) Coupled cluster with single, double and perturbative triple excitations
CG Cubic gauche
CHNO Carbon-hydrogen-nitrogen-oxygenal
154 B.M. Rice et al.
1
Introduction
The development of accurate models and simulations of energetic materi-
als (EM) has been aggressively pursued within the EM research community
since the advent of computational capabilities. Numerous analytic and com-
putational models that predict the performance of an EM in a variety of
applications exist, but most have a significant dependence on empirical data
that previously could be obtained only through measurement. Due to the
cost and time associated with collection of such empirical data (i.e., synthesis
Computational Aspects of Nitrogen-Rich HEDMs 155
and testing), the EM community has recognized that for purposes of screen-
ing new materials, its dependence on such information must be reduced.
Therefore, great attention has been given towards developing physics-based
atomistic models for use in EM research and has resulted in a dramatic evo-
lution of methods and applications of these to EM.
A variety of models based on atomistic simulation methods have been de-
veloped to predict key properties of an EM that are used to assess potential
performance in a weapon or its sensitivity to impact [1, 2]. While showing
a measure of success, these models still include some empiricism and have
been developed for conventional explosives, most of which are composed
of carbon, hydrogen, nitrogen, and oxygen (CHNO). Therefore, one cannot
immediately conclude that the same models and methods will be applica-
ble to the new class of high-nitrogen molecular and ionic crystals that are
showing great promise as high energy density materials (HEDMs), and thus
are limited in their use as screening tools. Also, concern for potential en-
vironmental hazards associated with the use and synthesis of EMs calls for
the development of methods that will predict the environmental impact of
any new material being considered for synthesis. Unfortunately, there are
only a few methods that exist for such a purpose, both for CHNO and high-
nitrogen materials, and significant efforts in development and assessment of
such methods are still required. Efforts are being made to address all of these
limitations, and will be discussed herein.
Atomistic methods in EM research are not limited to use as tools
for screening purposes only. There are several examples where quantum
mechanical characterizations of isolated molecules or elementary reac-
tions are used to augment experimental information of newly synthesized
high-nitrogen condensed phase materials [3–13]. Quantum mechanical ap-
proaches are also used to identify novel forms of nitrogen in which structural
energy is stored, such as exotic all-nitrogen molecular species and high pres-
sure polymorphic phases of solid nitrogen. Several examples will be discussed
in this chapter. Atomistic simulation methods can also be used to predict
probable reaction mechanisms and for exploring the dynamic response of
a material to an initiating event. Simulations for the latter process using mo-
lecular dynamics (MD) methods, however, have been limited to conventional
CHNO explosives and to our knowledge, only two have been performed for
triazolium-based compounds, with a focus on predicting physical proper-
ties [14, 15]. Since there are no widespread applications of this methodology
to high-nitrogen systems, these types of simulations will not be discussed in
this chapter.
Section 2 will describe the various computational approaches used to de-
velop predictive procedures for high-nitrogen solids. Section 3 will describe
applications of these procedures to existing high-nitrogen HEDMs (both
molecular and ionic crystals). Section 4 will be devoted to a discussion of
predictions of novel all- or high-nitrogen species, followed by Sect. 5, in
156 B.M. Rice et al.
2
Computational Methods
(i.e., DFT using modest basis sets) have been used for predicting molecular
properties associated with screening since these can be calculated for large
molecules rapidly yet provide sufficiently accurate results. Atomistic simula-
tion of novel high pressure phases of nitrogen has been dominated by solid
state DFT treatments, the only computationally tractable quantum mechan-
ical method for the condensed phase. These calculations will be reviewed
in Sect. 5. Although there are numerous classical molecular simulations of
highly compressed nitrogen, none of the interaction potentials used in those
simulations correctly model phase transitions occurring in the nitrogen sys-
tem and will not be discussed herein.
QM methods are also used in the third category of atomistic simula-
tion methods we will include in this discussion. We focus on quantitative
structure activity relationship/quantitative structure property relationship
(QSAR/QSPR), a computational chemistry methodology that is extremely
popular within the pharmaceutical community for new drug design. This ap-
proach establishes mathematical correlations between molecular descriptors
(an inherent property of the chemical system) and various physico-chemical
properties and behavior for various classes of compounds. An extensive
compilation and description of over two thousand molecular descriptors
are provided by Todeschini and Consonni [20]. While many of the descrip-
tors are empirically derived or obtained from experimental information,
many of them can be generated using quantum mechanical methods [21].
QSPR/QSAR methods using conventional and quantum-mechanical molecu-
lar descriptors have been used successfully to predict certain key properties of
EMs for conventional CHNO materials [1] and applications will be described
in the next section. Drawbacks to this method are its empiricism and reliance
on experimental information; predictive capability of this method is not en-
sured for systems that are outside of the data set to which the QSPRs were
parameterized. As with all empirical models, there is a possibility that a QSPR
will perform poorly for dissimilar chemical systems. To our knowledge, this
approach has not been widely applied to high-nitrogen materials.
The theoretical methods that will be discussed in this chapter represent
only a subset of the various atomistic simulation methods used in compu-
tational materials research, and we refer the interested reader to the various
comprehensive reviews on each method [22–30]. For the purposes of this
chapter, however, we will highlight important points associated with various
theories in application to the high nitrogen materials.
3
Prediction of Properties of High-Nitrogen Solids (Neutrals, Ionics)
3.1
Performance Properties
Special attention has been given to the prediction of two properties that are
used to provide an initial assessment of the potential performance of a mate-
rial in a gun or warhead: the heat of formation and the density of the material.
A variety of computational chemistry methods to accurately predict such per-
formance properties [1, 31] exists, but applications to EM have been limited
almost exclusively to conventional CHNO explosives. Unfortunately, there has
been no similar wide-scale application of these computational methods to
high nitrogen compounds. Additionally, there have not been extensive ap-
plications to ionic crystals (either high-nitrogen or CHNO salts), since the
majority of EMs are neutral molecular organic crystals. Previous tools that
have been developed to predict either of these properties for CHNO energetic
neutral crystals are not suited for ionic materials, as will be described here-
after. Further, applications of existing tools that work well for CHNO neutral
molecular crystals have shown inconsistent (and sometimes poor) behavior
when applied to high nitrogen materials. Thus, the existing methods must be
reassessed and modified to accommodate the high nitrogen systems, and al-
ternative methods should also be explored. In this section, we will present
some of these alternatives.
3.1.1
Crystal Densities: Neutral Molecular Crystals
We will begin this section with a discussion of the prediction of crystal den-
sities for neutral molecular crystals. Several approaches have been used to
predict crystal densities without a priori knowledge of the system. The first,
and most sophisticated, is that of ab initio crystal prediction, in which the
crystal microstructure is determined using information about a single, iso-
lated molecule. While a number of procedures have been developed and
assessed [32], all methods generally follow a three-step computational ap-
proach. The first step corresponds to generating a three-dimensional model
of the molecule (the packing moiety) that will be used to construct candidate
crystals of different symmetries. In the next step hypothetical crystal struc-
tures using the molecular models are created; the contents of the structures
are dependent upon the crystal symmetry being explored and the orienta-
Computational Aspects of Nitrogen-Rich HEDMs 159
tion of the packing moiety. In the third step the energy of each hypothetical
crystal is minimized with respect to lattice and molecular orientation pa-
rameters. After generating a large series of candidate crystal structures, the
crystals can then be ranked (usually by energy or density). Although it has
been shown to be a very useful predictive methodology, it suffers from certain
limitations. The majority of computational methods assume that the crystal
structure with the lowest lattice energy corresponds to the thermodynamic-
ally favored structure rather than the structure with the lowest free energy.
This assumption effectively ignores entropic and vibrational enthalpic con-
tributions to the free energy and does not consider kinetic factors associated
with crystal growth, such as solvent effects and crystallization conditions.
Another of the major limitations of this method of crystal density predic-
tion is its reliance on a description of interatomic interactions. As in any
atomistic modeling procedure, the quality of the result is dependent on the
accuracy of the description of the interatomic forces. Finally, current methods
require significant improvement in their search methods used to generate the
candidate crystals, particularly for systems in which the asymmetric unit of
a crystal (Z ) is greater than one [32]. However, the utility of the proced-
ure in EM research has been demonstrated, in spite of the aforementioned
assumptions and limitations [33, 34]. A survey study [34] was performed in
which 174 CHNO molecular crystals whose molecules contained functional
groups common to CHNO explosives (i.e., nitramines, nitroaliphatics, ni-
trate esters, and nitroaromatics) were subjected to the method of ab initio
crystal prediction using a transferable CHNO interaction potential [19] and
the method developed by Ammon and co-workers(MOLPAK/WMIN) [33].
The systems chosen were restricted to crystalline space groups and systems
that could be treated by MOLPAK/WMIN (i.e., Z ≤ 1). The study showed
that for 85% of the chemical systems simulated, the method and model pro-
duced a crystal structure whose lattice parameters and contents of the crystal
(i.e., molecular positions and orientations) matched the experimental crys-
tal. Additionally, approximately 75% of these were the low energy structures
of all possible candidates generated in the crystal prediction process. Pre-
dicted densities (calculated at 0 K) had a root-mean-square (rms) deviation
from experiment of 3%. Inclusion of thermal effects is expected to bring these
results more in line with the experimental values. This study showed that
the method of ab initio crystal prediction is suitable to predict crystal den-
sities of a new EM (provided the description of the interatomic interactions
are reasonably accurate). However, because it utilizes an empirically derived
interatomic model that was parameterized to CHNO systems, it cannot be as-
sumed that the model will adequately predict structures of systems that are
dissimilar from the original set to which it was parameterized. We demon-
strate this in an application of the same interaction potential [19] and the
method of ab initio crystal prediction to six high-nitrogen neutral molecu-
lar crystals synthesized by Klapötke [35]. The results were unsatisfactory; in
160 B.M. Rice et al.
three cases the experimental crystal was not found among the various poly-
morphs that were generated in the calculations. For the remaining three,
two of the structures identified as matching the experimental crystals were
significantly higher in energy than those of other polymorphs generated in
the calculations. Furthermore, the contents of the unit cells and the cell di-
mensions had unacceptably large deviations from the measured values. This
exercise clearly demonstrated a need for an interaction potential to treat such
high-nitrogen compounds. Ammon has recently refined the default inter-
atomic interaction potential used in the MOLPAK/WMIN suite of programs
to describe high-nitrogen systems [36]. Application of the MOLPAK/WMIN
methods and this newly refined interatomic potential produces densities of
eight high nitrogen crystals within a few percent. These results are given in
Table 1. Further assessments of this model and method for high-nitrogen sys-
tems are being performed as experimental information is being collected.
Ab initio crystal prediction provides a density value and important in-
formation about the positions of the atoms in the unit cell, invaluable in-
formation that can be used to construct molecular models for use in other
simulations or might be useful in analysis of experimental results of dynamic
response (e.g., directional shock sensitivity). Unfortunately, the methods are
computationally intensive and analysis of the results is not trivial. For assess-
ment of potential performance, often the user only needs the value of the
crystal density at room conditions. For such calculations, QSAR/QPSR ap-
proaches are extremely attractive due to their ease of use and speed. Several
QSPR-type approaches have been used to predict condensed phase densities
of molecular organic systems [37–45]; some of which are included in EM re-
search. In some of the QSPR-type approaches, a property called “molecular
Table 1 Crystal densities predicted using molecular volumes within the 0.001 a.u. isosur-
face of electron density
3.1.2
Crystal Densities: Ionic Molecular Crystals
In principle, the various methods used for crystal density prediction of neu-
tral molecular crystals should be applicable to ionic systems. Indeed, group
additivity schemes have been developed for ionic crystals [53, 54]. Unfortu-
nately, the ionic partners that make up the formula unit in the ionic crystal
introduce complications in implementing some of the other methods. The
main challenge is properly defining the arrangement of the ionic partners
164 B.M. Rice et al.
where M denotes the cation and X denotes the anion. This method of formula
unit volume prediction was applied to 34 high-nitrogen ionic salts provided
by Klapötke [35]. In this application, the ionic volume is assumed to be that
contained within the 0.001 a.u. isosurface of electron density of the ion calcu-
lated at the B3LYP/6-31G∗∗ level. Two sets of ionic volumes were generated.
The first set were calculated using molecular structures consistent with the
experimental crystal and the second used equilibrium geometries resulting
from a B3LYP/6-31G∗∗ geometry optimization. Unlike what we observed in
our earlier application of this method to neutral CHNO molecular crystals,
the formula unit volumes calculated using the experimental structures were
in closer agreement with experimental values than those using the optimized
structures. Upon examination of the optimized geometries of the cationic and
anionic components of the ionic volumes, we found the following: for those
ions that had no hydrogen atoms, the differences between the volumes of ions
assuming the experimental structures with those using the optimized struc-
tures were very small (fractions of Å3 ). The largest differences were for ions
that contained hydrogen atoms. Also, the magnitude of the difference in vol-
ume is directly proportional to the number of hydrogen atoms in the ion.
This difference appears to be due to differences in the X – H (X = C, N) bond
distances in the optimized and experimental structures. The B3LYP/6-31G∗∗
X – H bond distances are ∼ 0.1 Å larger than those reported for the experi-
mental structures.
Computational Aspects of Nitrogen-Rich HEDMs 165
Fig. 2 Comparison of formula unit volumes of 34 high-nitrogen ionic crystals (g/cc) using
experimental volume, predicted volumes using the experimental structures, predicted un-
corrected volumes using optimized structures, and predicted volumes using optimized
structures, corrected for the number of hydrogens (see text)
166 B.M. Rice et al.
3.1.3
Solid Phase Heats of Formation: Neutrals
It can be argued that the solid phase heats of formation of a crystal can be cal-
culated using highly accurate solid state quantum mechanical methods. How-
ever, computational obstacles rule out this approach. First, the only computa-
tionally feasible quantum mechanical approach to calculate the lattice energy
of molecular crystals at this time is density functional theory. DFT does not
adequately describe dispersion interactions, which are the main components
of the binding energy in such crystals at ambient conditions [59, 60]. Sec-
ondly, this approach requires knowledge of the crystalline environment, and
thus requires empirical information, which works against the overall goal to
reduce dependence on such information through modeling. Solid phase heats
of formation can be predicted using QSPR approaches [61–64], but their pre-
dictive capability might be limited to the chemical systems to which they were
Computational Aspects of Nitrogen-Rich HEDMs 167
heats of formation. The results and experimental values for the resulting solid
phase heats of formation are shown in Fig. 4.
As evident in Fig. 4, the theoretical results are in reasonable agreement
with each other for all compounds. Additionally, all calculated values are
in agreement with the experiment except for Compound 2 (1-Methyl-5-
(methylnitramino)-1H-tetrazole [4]). This suggests that either re-measure-
ment of the heat of formation of Compound 2 should be performed or that
there is some aspect of this system that is not captured by the quantum
mechanical calculations. We emphasize that the two quantum mechanical ap-
proaches (atom/group equivalents versus G3 methods) for prediction of the
gas phase heats of formation are sufficiently dissimilar that it is extremely
unlikely that the error is within this portion of Eq. 1. There is a possibility,
of course, that the heat of sublimation is poorly predicted for this system.
That would not, however, explain the large discrepancy between the theoret-
ical values and the measured result for Compound 2. Average and rms percent
deviations were calculated using the seven other high-nitrogen compounds.
Average percent deviations of the predictions from experiment are approxi-
mately the same for the group-equivalent and G3(MP2) approaches (these
overestimate, on average, the experimental value by 5.3 and 5.6%, respec-
tively). The average percent deviation of the G3(MP2)//B3LYP approach is
Fig. 4 Solid phase heats of formation for high-nitrogen crystals. Experimental values are
taken from [4] and [223]. “1999 Atom Equiv” denotes calculations using the atom equiva-
lent method described in [68]. “2006 Group Equiv” denotes calculations using the group
equivalent method described in [67]. G3(MP2) and G3(MP2)//B3LYP denote calculations
using variants of the G3 method [78, 79], respectively
170 B.M. Rice et al.
larger (– 7.9 kcal/mol) and underestimates the experimental value. The rms
percent deviations from experiment are comparable for the group equivalent
and G3 approaches, with the group equivalent and G3(MP2) having the best
overall agreement with experiment (rms % deviation is ∼ 9.5%).
Of particular interest to us are the consistently close agreement between
the group-equivalent method [67] and the G3(MP2) values. We remind the
reader that the group-equivalent method was fitted to a smaller and more
chemically specific set of experimental data (CHNO compounds with func-
tional groups common to explosives) whereas the G3 parameters were fitted
to a more general and substantially larger set of chemical compounds. Ad-
ditionally, the group-equivalent approach requires considerably less compu-
tational resources. Since this good agreement between the two approaches
has only been demonstrated for eight compounds, further investigation is
required to determine whether this trend holds for a larger number of high-
nitrogen compounds.
3.1.4
Solid Phase Heats of Formation: Ionic Crystals
As for the neutral crystals, we are not aware of any QSPRs developed to pre-
dict the solid phase heats of formation of high nitrogen salts, probably due
to the relative scarcity of such experimental data required for establishing the
correlations.
Since the crystalline binding energy in an ionic crystal is dominated by
electrostatic interactions that are orders of magnitude larger than those in
a neutral molecular crystal, the same scheme used for predicting the solid
phase heat of formation for a neutral molecular crystal (Eq. 5) cannot be
applied to ionics. Rather, the heat of formation of ionic compounds is deter-
mined using Born–Haber cycles [65] in which a series of reactions involving
the ionic components are employed to produce the overall final result. Unfor-
tunately, error is always introduced when quantum mechanical calculations
are used in evaluating the reactions. Therefore, care must be taken in the
choice of reactions used in the cycle.
The simplest of cycles might consist of three steps in which formation en-
ergies for each of the components can be calculated.
M(g) + X(g) → M+(g) + X–(g) → M+ X–(solid) . (6)
In this simple scheme, the first step requires the evaluation of the heats of for-
mation of the neutral form of the charged components. The formation ener-
gies for the ionic components are then determined by evaluating the electron
affinity or ionization energy and adding these to the heats of formation of
the neutral moieties. The third step of the cycle requires the evaluation of the
lattice enthalpy. Chemical reactions that generate the gas phase ionic compo-
nents in Step 2 can also be used. Also, an alternative procedure for generating
Computational Aspects of Nitrogen-Rich HEDMs 171
the gas phase formation energies of the ions directly is presented by Beau-
camp et al. [81]. In this work, gas phase heats of formation for the ions were
determined using the method of atom equivalents. The atom equivalents were
derived for a series of neutral compounds [82] and applied to five ammonium
salts. These values and cohesive energies for these systems were used to gen-
erate solid phase heats of formation. The results are in reasonable agreement
with the experiment, and further study on a larger number of high-nitrogen
salts should be performed to determine the suitability of this method.
Unfortunately, there are limited (if any) experimental data that can be used
to assess the quality of the calculations associated with each individual step in
the aforementioned cycle. We are limited to comparing the overall final result
with the experimental value of the heat of formation of the ionic crystal. This
makes establishing the sources of errors in this cycle complicated.
It is possible to introduce significant error in the first step of this cycle.
The magnitude of the error introduced in a quantum mechanical prediction
of any reaction (including atomization) is dependent on the choice of reaction
(i.e., the types and number of bonds that are broken or formed). The most
accurate quantum mechanical approach for predicting the gas phase heat of
formation of a new compound is to use isodesmic reactions [83]. This ap-
proach works very well if there are reliable heat of formation values for the
products of the isodesmic reaction. If there is more than one possible reaction
that will lead to the gas-phase ions in the cycle, it is possible that the quan-
tum mechanical predictions will produce different results that are dependent
on which reaction is chosen [84]. Also, generation of the gas-phase ions in
Step 2 might require a sequence of reactions rather than a single reaction;
this would introduce further error. Also, while there might be many different
possible reactions leading to the charged moieties of interest, corresponding
experimental data of the reaction products must be available. In lieu of that,
Gx procedures [72] (or atom/group equivalent methods, if applicable to the
system) can be used to predict the heats of formation of the various reaction
products, but again, error is introduced.
To avoid the complications associated with identifying reaction sequences
that can be used to produce gas phase formation energies of the ionic species,
we will illustrate predictions of the solid phase heats of formation of a few
high-nitrogen salts using the Born–Haber cycle given in Eq. 6. In this, gas
phase heats of formation of the neutral forms of the charged species are first
calculated (Step 1), and are followed by calculations of the electron affinity or
ionization energy (Step 2) to generate the heats of formation of the charged
moieties in the gas phase. The main drawback to this approach is the require-
ment that the neutral form of the charged species must be a minimum on the
potential energy surface (PES). If it is not, then the Born–Haber cycle would
require inclusion of reactions leading to the ionic moieties in Step 2.
We were encouraged by the good agreement between experimental and
predicted values (using the G3 method) of gas phase heats of formation for
172 B.M. Rice et al.
Fig. 5 Molecular structures of CH9 N6 before (upper half of the figure) and after (lower
half of the figure) ionization. Structures depicted in the left-most portion of the figure
are top-down views of the moieties; structures in the right-most portions are side views
of the moieties
tion for the CH9 N6 moiety calculated at the B3LYP/6-31+G∗∗ level; it is clear
that in some cases, structural relaxation would be a significant contributor to
the final value of the ionization energy.
To determine the effect of structural relaxation on the ionization en-
ergy, we have calculated the relaxed and vertical ionization energies for
the 20 high-nitrogen cations and 6 anions contained within various high-
nitrogen salts synthesized by Klapötke [35]. The results are shown in Fig. 6.
We found the effect to be particularly pronounced for the ionization ener-
gies; half of the compounds had structural relaxation energies on the order of
50 kcal/mol.
The final step in the evaluation of the Born–Haber cycle given in Eq. 6 is
the determination of the lattice enthalpy, a measure of the energy required to
dissociate the ionic crystal into its gaseous ions. The magnitude of this energy
is quite large compared to the weaker intermolecular cohesive energies asso-
ciated with organic molecular crystals (which are due mainly to van der Waals
interactions), since it is a result of numerous long-range electrostatic interac-
tions of ionic partners within the crystal. This value can be directly calculated
by adding up all interatomic interactions within the crystal lattice if the pos-
itions of all atoms in the crystal are known and a reasonable description
174 B.M. Rice et al.
Fig. 6 Vertical and relaxed ionization energies for 20 cations and 6 anions used in high-
nitrogen ionic crystals synthesized by Klapötke [35]
of the interatomic interactions is available (see, for example [58, 81, 85, 86]).
Unfortunately, these two conditions cannot always be met. Therefore, numer-
ous empirical schemes have been introduced to predict this quantity without
knowledge of the crystal structure. One of the most widely used methods
was developed by Kapustinskii, in which the lattice energy of a salt is esti-
mated using values of the ionic radii [87]. Extensions to this approach have
been made to better predict lattice energies of complex ionic crystals [88, 89].
Another method uses lattice energies calculated using point charges and ex-
perimentally crystallographic structures for a larger number of systems to
generate a QSPR for cohesion energies in salts [85]. Politzer and Murray ap-
plied the GIPF approach to generate GIPF relations between lattice energies of
ammonium, sodium and potassium cations with various anions [90]. A dif-
ferent QSPR-type method developed by Jenkins et al. has been applied to
several high-nitrogen ionic crystals [58, 91]. This approach correlates the in-
verse cube root of the volume of the formula unit of an ionic crystal Mp Xq
with its lattice potential energy Upot (Mp Xq ). The lattice potential energy is
related to the lattice enthalpy as follows:
∆HL = Upot (Mp Xq ) + [p(nM /2 – 2) + q(nX /2 – 2)]RT , (7)
where the values of nM and nX are dependent on whether the ions are
monatomic, linear polyatomic, or non-linear polyatomic. Jenkins et al. [58,
91] parameterized the functional form for Upot using lattice potential ener-
Computational Aspects of Nitrogen-Rich HEDMs 175
gies from [92]. These data consist of values derived from thermochemical
cycles using experimental information and those from full scale calculations
of the lattice potential energy of a crystal. In these calculations, crystal struc-
tural data (atomic positions) are specified, and an interatomic interaction
potential (van der Waals terms plus Coulombic interactions, with partial
charges assigned to nuclei) are used to produce the lattice energies. The in-
formation used in the parameterization was limited to salts containing alkali
metal and alkaline earth cations (Li, Na, K, Rb, Cs, Mg, Ca, Sr, Ba). Gold-
schmidt radii of the cations (r+ ) were used to define the cation volumes
(V+ = 4/3πr+3 ). Anion volumes were determined first by calculating the mo-
lecular volume of the alkaline earth/alkali metal salts (using experimental
values, Vunit cell /Nunit ) and subtracting the volumes of the cations.
A possible source of error associated with this approach might be due to
the variable reliability of the fitting data [92]. Another source of error asso-
ciated with this approach when applied to high-nitrogen salts is the types
of salts used in the parameterization. In this parameterization of the QSPR,
the cations of the salts were limited to alkaline metal or alkaline earth atoms
(small and spherical volumes). The anion volumes were derived from experi-
mental information about the salts and the values assigned to the alkali metal
or alkaline earth cations. It is not known whether this method is transferable
to salts that have significantly different chemical compositions, and for which
the cations are substantially different in shape and size (such as some of the
new high-nitrogen salts).
A rough assessment of the Jenkins approach for predicting lattice energy
for high-nitrogen salts can be performed through direct calculations assum-
ing the experimental structure and reasonable interaction potentials (such as
that developed by Ammon, described earlier). At this time, we have not at-
tempted nor are we aware of such an assessment. We will, however, use the
Jenkins approach for purposes of illustrating prediction of the heat of for-
mation of ionic crystals, specifically, three recently synthesized high-nitrogen
salts [10]. These are presented and compared with experimental and other
theoretical predictions in Table 2. The theoretical result generated by Gálvez–
Ruiz et al. [10] invoked a Born–Haber cycle using reactions along with proton
affinities, whereas we utilize the procedure outlined above. Additionally, the
Gálvez–Ruiz et al. calculations used the G2 method [93], which has differ-
ent empirical corrections and approaches than the G3 methods we used.
Note that the two theoretical approaches, which would have distinctly differ-
ent sources of error, are very close in value for the three compounds. They
both, however, overshoot the experimental values by 30–40 kcal/mol. Since
the common element between the two theoretical methods is the Jenkins ap-
proach [58, 91] for predicting lattice energies, it is possible that the major
discrepancy between experiment and theory is within this step. However, be-
fore this conclusion can be made, the Jenkins method should be evaluated as
suggested above.
3.2
Prediction of Vulnerability and Environmental Hazard
weight impact tests, a crude and rapid method used to qualitatively assess
the sensitivity of a material to impact. This test is notoriously inaccurate
and its results are strongly dependent on the conditions under which the
experiments are performed. Therefore, correlations made with the results
could incorporate flawed measurements. Further, it is well established that
the sensitivity of a material to accidental initiation is influenced by material
morphology, something that cannot be captured with such simple correla-
tions. Finally, there is never a guarantee that a QSPR developed for one series
of compounds is transferable to a different chemical family. Since several
correlations developed for conventional CHNO explosives are not transfer-
able across chemical families, it is not expected that such correlations would
be maintained for high-nitrogen HEDMs. We note, however, that two stud-
ies [5, 6] applied a QSPR-type method to assess impact sensitivity using
quantum mechanically calculated electrostatic potentials [102, 159] to sev-
eral high-nitrogen solids and showed that the correlations were maintained.
However, since these methods were developed for CHNO explosives, similar
applications to a larger number of high-nitrogen systems must be performed
before it is concluded that such a correlation is indicative of sensitivity to
impact for these systems. Clearly, advances in development of predictive
methodologies in this area are needed.
4
Novel Polynitrogen Species
5
Novel High Pressure Phases of Nitrogen
sought, but it has only been recently that such was achieved [199–203]. In
addition to these non-molecular phases of nitrogen, several molecular phases
over wide temperature and pressure ranges have been experimentally deter-
mined [193–198, 204, 205].
Theoretical studies, on the other hand, have predicted a variety of non-
molecular structures at pressures for which only molecular phases have been
observed. These include a monatomic simple-cubic phase [206–208], a semi-
metallic arsenic structure A7 [206], a metallic, simple-tetragonal phase [206,
209], and the cubic gauche (CG) structure [205, 207, 210–213], for which re-
cent experimental evidence has been given [203]. Results from theoretical
calculations also predict other polymeric forms of nitrogen, as will be de-
scribed hereafter.
The disparity between the theoretical predictions and experimental veri-
fications can be attributed to several factors, including large energy barriers
that might inhibit any transition leading to the polymeric forms upon com-
pression of the molecular crystals. These can be explored using theoretical
methods, as will be shown here. This section will describe detailed, ab initio
descriptions of the electronic structure of high-pressure phases of nitrogen,
and ab initio quantum molecular dynamics (QMD) calculations, in which
integration of the classical equations of motion using quantum mechanical
forces produce time-dependent atomic positions and velocities. The latter cal-
culations have been used to identify stable structures through quenching of
materials simulated at high temperatures and pressures. QMD calculations
are also used to establish ranges of metastability. This section will be limited
to discussions of quantum mechanical calculations, with a focus on density
functional theory treatments of these systems, and will not include discus-
sions of classical molecular simulations of high-pressure phases of molecular
nitrogen.
It was generally assumed that pressure-induced transformation of molecu-
lar nitrogen would progress first to a chain-like structure, then to a layered
structure and finally to an bulk, or extended solid, structure. Early theoret-
ical work, therefore, focused on establishing the simple chain, layered, and
bulk structures known from other Group V elements, including A7 [206], and
black phosphorus (BP) [207]. Early calculations of the various forms of non-
molecular nitrogen predicted that the layered A7 and BP structures are lower
in energy than the chain and bulk structures [206]. Later calculations pre-
dicted that the transition from molecular nitrogen would progress directly
to the bulk CG structure, which is unique to nitrogen, bypassing formation
of layered and chain structures [207, 210–213]. The CG form of nitrogen is
a fully coordinated three-dimension structure, and is an analog to the dia-
mond structure of carbon. Calculations predict it to be substantially lower in
energy than the other non-molecular phases A7 and BP. Another chain struc-
ture was recently predicted that is energetically competitive with CG and will
be detailed hereafter.
180 B.M. Rice et al.
Fig. 7 Diagrams of various forms of compressed nitrogen. The structures are, from left to
right, cubic gauche (CG), the hexagonally packed chain (HPC), ε-nitrogen, and Hexagonal
N6 rings
Computational Aspects of Nitrogen-Rich HEDMs 181
are very similar to those of the chain. It is conceivable that such molecules
could be packed into the crystalline state, perhaps at near ambient conditions.
Figure 8 shows the energy curves for the HPC and CG structures. The
energy for the chain structure is slightly lower than CG at pressures well
above the transition pressure (i.e., V < 5 Å3 ), but is approximately the same
or slightly higher than CG for the pressure range near the transition pressure.
The difference in the energies of these structures is within the error of the
GGA approximations, making the HPC structure practically degenerate in en-
ergy with the CG structure between approximately 5 and 6.5 Å3 . At pressures
less than the transition pressure the HPC is lower in energy than CG.
In the HPC phase, each atom has valence electrons symmetrically bonded
to two other atoms. For such a case, it is expected that the bands at the Fermi
energy would be π-bands, similar to those in graphite. However, the nearly
one-dimensional nature of the HPC chain structure should lead to a large,
metallic, nearly flat Fermi surface unlike the semi-metallic Fermi surface in
graphite. Indeed, the chain structure is metallic in nature as evidenced by the
band structure in Fig. 9. The band structure of an identical single isolated
chain is provided for comparison. Only the bands along the major axis of the
chain are shown, as directions normal to the major axis are uninteresting, as
expected for chains. We explored the possibility of a Peierls distortion leading
to an insulator [215], but this is not found in the tests on distorted structures
Fig. 8 Energy versus volume curves for the HPC, CG, ε, amorphous molecular and
amorphous nonmolecular phases of nitrogen
Computational Aspects of Nitrogen-Rich HEDMs 183
Fig. 9 Band structure of the HPC structure (left-hand side) and a single isolated chain
(right-hand side). The Fermi energy is represented by the solid dark line
reported above. Presumably the stability of the metallic nature of the HPC
phase is due to the fact that the system is not truly one-dimensional.
A metallic form of nitrogen is of great interest because of the possibil-
ity of superconductivity at a high transition temperature in such a phase.
This possibility is suggested both by the small mass and by the large valence
of nitrogen. Indeed, one the highest temperature elemental superconductor
known is sulfur at high pressure [216–218], first predicted by theory [219].
The small mass of nitrogen implies a large Debye temperature, and the large
number of valence electrons of nitrogen implies a large carrier density in
a metallic phase. Both of these properties will tend to enhance the transition
temperature. The electron–phonon interactions should be large for this elem-
ent, as for other first row elements. On the one hand, the electron–phonon
interaction is responsible for the instability of structures like the simple cubic;
on the other hand, if the stable structure is metallic, then the electron-phonon
interaction could lead to a high superconducting transition temperature Tc .
Thus nitrogen can be an example of the competition between high Tc and
structural instabilities [206]. Such a question invites further study.
An interesting molecular phase that was identified during the exploration
of the conformational space of nitrogen was that of hexagonal N6 , shown
184 B.M. Rice et al.
on the right side of Fig. 7. While this is a molecular phase, it is not a di-
atomic molecular phase, and thus has the potential for large energy release
upon transition to the lower-energy diatomic molecular phase. When the
ε structure is minimized at ∼ 60 GPa, this non-diatomic molecular phase ap-
pears. The ε structure transitions into a hexagonal-close-packed structure
that consists of a unit cell containing one N6 and one N2 molecule, with
the N2 molecule centered coaxially between the N6 molecules on alternating
levels. The transition to this phase creates hexagonal N6 from the disk-like
molecules forming columns. The sphere-like molecules remain molecular
and are centered between two coaxial N6 molecules. These columns of N6
are packed hexagonally, with adjacent columns offset along the axis of the
column by a third the distance between N6 in the column. This offset al-
ternates direction going clockwise for each of the six neighboring columns.
This structure has an energy that is equivalent to the molecular ε struc-
ture at the same volume and is a local energy minimum between 18 and
145 GPa. Since this is a relatively high-energy structure, no further stability
testing was performed. At the lower end of the pressure range, the hexagonal
structure transforms back into the ε molecular structure. Interestingly, as
the pressure on this structure is increased beyond 145 GPa, the N6 molecules
begin to bond to each other at adjacent corners of the hexagons, forming
a three dimensional structure not entirely dissimilar to the cubic gauche
structure while maintaining the encapsulated N2 molecules. While the en-
ergy of this structure is clearly above the other structures shown, it may
still be physically observable if the transition barriers are large enough. It
may be possible to obtain this phase by heating then cooling the ε phase
at pressures below 100 GPa. Thus, this structure may be a candidate for
the θ phase [202]. In addition, N6 rings have been observed by Vogler
et al. [220].
A second series of calculations were performed that were different in
spirit than those described heretofore, in which the ε-form of molecular ni-
trogen was compressed to generate new structures. In this series, atomic
nitrogen gas at high temperature and pressure was rapidly cooled at con-
stant pressure. These calculations used QMD to produce the quench. The
quenches of the hot atomic nitrogen gas produced, as expected, amorph-
ous final states, but the final pressure at the end of the quench determined
whether the final state would be molecular or a connected network of ni-
trogen. The molecular phase was only found below 100 GPa. At 100 GPa and
above, the quenched configurations were interconnected networks, some of
these would have a few diatomic nitrogen molecules, but all other nitrogen
atoms are connected by a path of bonds to the other atom in the structure.
The transition pressure of 100 GPa agrees well with that obtained from analy-
sis of experimental results [200] (∼ 100 GPa). In addition Raman studies of
nitrogen have found evidence for the nitrogen diatomic bond destabilization
at about 100 GPa [197, 220].
Computational Aspects of Nitrogen-Rich HEDMs 185
Fig. 10 Left-hand side: A typical amorphous structure, quenched to 200 GPa. The bonds
that cross the periodic boundaries are not shown. Right-hand side: Passivated amorphous
nitrogen structure. The hydrogen atoms are shown in dark gray and the nitrogen in light
gray
186 B.M. Rice et al.
6
Concluding Remarks
References
1. Politzer P, Murray JS (eds) (2003) Energetic Materials: Part 1. Decomposition, Crys-
tal and Molecular Properties (Theoretical and Computational Chemistry). Elsevier
Science, Amsterdam
2. Politzer P, Murray JS (eds) (2003) Energetic Materials: Part 2. Detonation, Combus-
tion (Theoretical and Computational Chemistry). Elsevier Science, Amsterdam
3. Drake GW, Hawkins TW, Boatz J, Hall L, Ashwani V (2005) Propellants, Explos, Py-
rotech 30:156
188 B.M. Rice et al.
216. Kometani S, Eremets M, Shimizu K, Kobayashi M, Amaya K (1997) J Phys Soc Jpn
66:2564
217. Struzhkin VV, Hemley RJ, Mao HK, Timofeev YA (1997) Nature 390:382
218. Gregoryanz E, Struzhkin VV, Hemley RJ, Eremets MI, Mao H-K, Timofeev YA (2002)
Phys Rev B 65:064504
219. Zakharov O, Cohen ML (1995) Phys Rev B 52:12 572
220. Vogler A, Wright RE, Kunkely H (1980) Angew Chem Int Ed 19:717
221. Ciezak JA (2006) US Army Research Laboratory, private communication
222. Hemley RJ (2002) The Carnegie Institution Geophysical Laboratory, private commu-
nication
223. Huynh MHV, Hiskey MA, Chavez DE, Naud DL, Gilardi RD (2005) J Am Chem Soc
127:12537
224. Hammerl A, Holl G, Klapötke TM, Mayer P, Noth H, Piotrowski H, Warchhold M
(2002) Eur J Inorg Chem 4:834
225. Hang M, Huynh V, Hiskey MA, Archuleta JG, Roemer EL, Gilardi R (2004) Angew
Chem Int Ed 43:5658
226. Patra S, Sarkar B, Ghumaan S, Fiedler J, Kaim W, Lahiri GK (2004) Inorg Chem
43:6108
Struct Bond (2007) 125: 195–271
DOI 10.1007/430_2006_052
© Springer-Verlag Berlin Heidelberg
Published online: 18 January 2007
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
5 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Abstract The article presents a survey of the development trends in studies of sensitiv-
ity (initiation reactivity) of energetic materials (EMs) over the last nine years, focusing
mainly on impact and shock sensitivities. Attention is given to the initiation by heat,
laser, electrostatic discharge, impact and shock, including the influence of hydrostatic
compression, crystal defects, molecular structure and desensitizing admixtures on the
initiation reactivity. Problems of the initiation of nitromethane are examined with a spe-
cial accent. It is stated that one of the best-developed theories for such studies is
Dlott’s Model of the Multiphonon Up-Pumping. Also significant is the model based on
Non-Equilibrium Zeldovich–von Neuman–Döring theory. Very important are approaches
devised by Politzer and Murray, updated by Price et al. as a hybrid model of prediction
of the impact sensitivity of CHNO explosives. The physical organic chemistry approach
to the sensitivity problem (POC model) is discussed with special emphasis. In this way it
has been found that the electron structure and close neighborhood of the primarily leav-
ing nitro group are dominant factors in the initiation by shock, electric spark and heat of
polynitro compounds.
Abbreviations
1,3-DNB 1,3-dinitrobenzene
1,5-DNN 1,5-dinitronaphthalene
1,8-DNN 1,8-dinitronaphthalene
AcAn 1,9-diacetoxy-2,4,6,8-tetranitro-2,4,6,8-tetraazanonane
BITNT 3,3 -dimethyl-2,2 ,4,4 ,6,6 -hexanitro-1,1 -biphenyl
BTX 5,7-dinitro-1-(2,4,6-trinitrophenyl)-1H-1,2,3-benzotriazole
CPX 1,3-dinitroimidazolidine
CTB 2-chloro-1,3,5-trinitrobenzene
DADN 1,5-diacetyl-3,7-dinitro-1,3,5,7-tetrazocane
DATB 2,4-diamino-1,3,5-trinitrobenzene
DCTB 2,4-dichloro-1,3,5-trinitrobenzene
Sensitivities of High Energy Compounds 197
Decagen 1,3,5,7,9-pentanitro-1,3,5,7,9-pentazecane1
Digen 1-nitro-1-azaethylene2
DIPAM 3,3 -diamino-2,2 ,4,4 ,6,6 -hexanitro-1,1 -biphenyl
DIPS 1,3,5-trinitro-2-[(2,4,6-trinitrophenyl)thio]benzene
DIPSO 1,3,5-trinitro-2-[(2,4,6-trinitrophenyl)sulfonyl]benzene
DMDIPS 2,4,6-trinitro-3-[(3-methyl-2,4,6-trinitrophenyl)thio]toluene
DMEDNA 2,5-dinitro-2,5-diazahexane
DMNA 2-nitro-2-azapropane
DMNO 2,5-dinitro-2,5-diazahexane-3,4-dione
DNAt 4,6-dinitro-2,1-benzoisoxazole (4,6-dinitroanthranil)
DNBF 4,6-dinitro-2,1,3-benyoxadiayol-1-ium-1-olate (4,6-dinitrobezofuroxane)
DNDAH 3,5-dinitro-3,5-diazaheptane
DNDC 1,4-dinitropiperazine
DODECA 2,2 ,2 ,2 ,4,4 ,4 ,4 ,6,6 ,6 ,6 -dodecanitro-[1,3 ,1 ,1 ]quaterphenyl
DPE 1,3,5-trinitro-2-[2-(2,4,6-trinitrophenyl)ethyl]benzene
DPM 1,3,5-trinitro-2-(2,4,6-trinitrobenzyl)benzene
DPT 3,7-dinitro-1,3,5,7-tetraazabicyclo[3.3.1]nonane
HMX 1,3,5,7-tetranitro-1,3,5,7-tetrazocane (octogen)
HNB 2,2 ,4,4 ,6,6 -hexanitro-1,1 -biphenyl
HNIW 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane
HNO N,N - bis(2,4,6-trinitrophenyl)ethanediamide
HNS 1,3,5-trinitro-2-[(E)-2-(2,4,6-trinitrophenyl)vinyl]benzene
HOMO 1,3,5-trinitro-1,3,5-triazepane
MADNBF 7-amino-4,6-dinitro-2,1,3-benzoxadiazol-1-ium-1-olate
MDN 1,1-dinitro-1-azaethane
NONA 2,2 ,2 ,4,4 ,4 ,6,6 ,6 -nonanitro[1,1 ,3 ,1 ]-terpheny
NTFA N,N-bis(2,4-dinitrophenyl)-2,4,6-trinitroaniline
OHMX 2,4,6,8-tetranitro-2,4,6,8-tetraazanonane
ONT 2,4,6,4 ,6 ,2 ,4 ,6 -octanitro[1,1 ,3 ,1 ]-terphenyl
ORDX 2,4,6-trinitro-2,4,6-triazaheptane
PA 2,4,6-trinitrophenol
PAM 2,4,6-trinitroaniline
PDNBTO 4,6-dinitro-2-(2,4,6-trinitrophenyl)-2H-1,2,3-benzotriazol-1-ium-1-olate
PETN penterythritol tetranitrate
PYX N,N -bis(2,4,6-trinitrophenyl)-3,5-dinitropyridine-2,6-diamine
RDX 1,3,5-trinitro-1,3,5-triazinane (hexogen)
TACOT-Z 1,3,7,9-tetranitro[1,2,3]benzotriazolo[2,1-a]-[1,2,3]benzotriazole-5,11-diium-
6,12-diide
TATB 2,4,6-triamino-1,3,5-trinitrobenzene
TCTB 2,4,6-trichloro-1,3,5-trinitrobenzene
TENN 1,4,5,8-tetranitronaphthalene
TENPO 1,3,7,9-tetranitrophenoxazine
Tetrogen 1,3-dinitro-1,3-diazetidine3
TETRYL N-methyl-N,2,4,6-tetranitroaniline
TEX 4,10-dinitro-2,6,8,12-tetraoxa-4,10-diazaisowurtzitane
TMPM N,N ,N -tris(3-methyl-2,4,6-trinitrophenyl)-[1,3,5]triazine-2,4,6-triamine
1 It has not been synthesized yet, for a prediction of some its properties see [16, 98, 148, 176].
2 It has not been synthesized yet, for a prediction of some of its properties see [16, 98, 148, 176].
3 It has not been synthesized yet, for a prediction of some of its properties see [16, 98, 148, 176].
198 S. Zeman
TNA 2-methoxy-1,3,5-trinitrobenzene
TNAD trans-1,4,5,8-tetranitrodecahydro-pyrazino[2,3-b]pyrazine
TNADEC 2,4,7,9-tetranitro-2,4,7,9-tetraazadecane
TNAZ 1,3,3-trinitroazetidine
TNB 1,3,5-trinitrobenzene
TNBA 2,4,6-trinitrobenzoic acid
TNCr 3-hydroxi-2,4,6-trinitrotoluene
TNMs 2,4,6-trimethyl-1,3,5-trinitrobenzene
TNN 1,4,5-trinitronaphthalene
TNPMO 1,3,7,9-tetranitrophenothiazine-5-oxide
TNPTD 1,3,7,9-tetranitrophenothiazine-5,5-dioxide
TNR 2,4-dihydroxi-1,3,5-trinitrobenzene
TNT 2,4,6-trinitrotoluene
TNX 2,4-dimethyl-1,3,5-trinitrobenzene
TPM N,N ,N -tris(2,4,6-trinitrophenyl)-[1,3,5]triazine-2,4,6-triamine
TPT 2,4,6-tris(2,4,6-trinitrophenyl)-[1,3,5]triazine
1
Introduction
2
Primary Splitting of EM Molecules
2.1
Low-Temperature Thermal Decomposition
The basic problem of defining the kinetics and mechanism of thermal decom-
position of energetic materials lies in the strong dependence of the cor-
responding kinetic parameters on temperature, pressure, and construction
materials in contact with the sample decomposed [9]. Hence the mutual com-
patibility of results obtained from thermal analyses of energetic materials
using different methods and/or different types of apparatus of different ori-
gin is very rare [9, 10]. So far the most reliable results in this area have
included both theoretical and practical findings obtained by Russian scien-
tists on the basis of their manometric method (see [10, 11] and citations
herein). The data obtained by this method are known to correspond to the
non-autocatalyzed stage of thermal decomposition of the given material (i.e.
to molecular structure [10]), and also to the absolute values of the corres-
ponding Arrhenius parameters. The mechanisms of primary unimolecular
fragmentation in low-temperature thermal decomposition (up to 600 K) of
technologically attractive organic polynitro and polynitroso compounds in
condensed state, specified mostly by the Russian authors, can be divided into
the following basic classes:
2.1.1
Quantum Chemical Simulations of the Thermal Decomposition
bonds [25]; it was shown that these compounds may be initiated by the
cleavage of both C – NO2 and N – N2 bonds. The given conclusion does
not agree with reality, because derivatives of ortho-nitroazidobenzenes pri-
marily give derivatives of benzofuroxanes by thermal decomposition [11],
therefore, only cleavage of the N – N2 bond in the azido group might be
important here. The same methodology was used to obtain the decom-
position activation energies, Ea , of 1,2,3-, 1,2,4-, and 1,3,5-trinitrobenzenes
(TNB) and their chloro derivatives [26]. The authors did not compare these
results with experimental data. The same is true of analogous theoret-
ical study of polymethyl derivatives of 1,3,5-trinitrobenzene and dinitro-
toluene [27]. In this case, the authors calculated the Ea values for homolysis
of the weakest bond C – NO2 and for hydrogen transfer from the methyl
group to the oxygen atom of the ortho-nitro group [26]. The splitting dis-
cussed does not correspond with the real course of the process, while the
hydrogen transfer partially corresponds with Scheme 2. Energy of this trans-
fer correlates to the impact sensitivity of studied polymethyl derivatives of
TNB [27].
By using the B3LYP level of density functional theory, possible reaction
pathways of 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazawurtzitane (HNIW)
in the gas phase have been investigated [32]. In accordance with the findings
presented in [11–18], the homolysis primarily concerns the N – NO2 bond in
the 2-position of the molecule. The same result was obtained by UHF-SCF-
PM3 MO calculation [33].
Research based on quantum-chemical simulation of thermal decompos-
ition of solid-state RDX using the Hartree–Fock method (combined with
the cluster and periodic models) [15] showed that the dissociation energy
of the N – NO2 bond strongly depends on the environment of the molecule.
The equatorial bond of the molecules inside a crystal is sensitive to the
crystal field and is characterized by a notable increase in energy compared
to the gas-phase dissociation energy. A significantly lowered energy bar-
rier (by 8–15 kcal mol–1 ) is obtained for a molecule placed near the sur-
face [15].
Methods of quantum mechanics (quantum chemistry) were utilized to
predict the products of thermal decomposition of triacetone triperoxide
(TATP) [30], i.e. oxygen, acetone and unstable acetone monoperoxide. By
means of a ReaxFF monomolecular molecular-dynamic cookoff simulation
it was found [31] that the initiation of TATP decomposition in the con-
densed phase is dominated by a monomolecular process [31]. This simulation
demonstrates that this initiation is entropy-driven, since the initial reaction is
almost energy-neutral [31].
Sensitivities of High Energy Compounds 203
2.2
High-Temperature Thermal Decomposition
It turns out that the primary fragmentation mechanism under these condi-
tions is entirely different from that of the low-temperature variant. However,
the results of molecular-dynamic simulations depend considerably on the
starting conditions set for a given simulation [39].
2.3
Laser Initiation
In the past nine years, two review articles have been published about laser
ignition and initiation [40, 41]. In this section, facts are presented that are
connected with studies of initiation reactivity of explosives.
The energy density threshold of a beam in laser ignition is 1 kW cm–2 with
a laser pulse over 1 µs. The majority of research work in the field of EMs
studies using lasers utilizes the quantum generators based on neodymium
glass or ruby, which operate in the mono-pulse regime [42]. The applied pulse
times vary from fractions of a second to several tens of nanoseconds [42].
The intensity of laser irradiation can be high (up to 109 W cm–2 or even
higher) [42]. The distribution of density of absorbed energy in irradiated EM
is determined by its optical characteristics. The chemical changes in a laser-
204 S. Zeman
irradiated EM start in hot spots [42, 43]. The hot spots can be represented
by structural non-homogeneities in the given EM (point defects, clusters of
vacancies, interfacial boundaries), chemical admixtures etc. [42, 43]. Crystal
defects can also be formed by action of a massive light wave. With sufficiently
high energy of the initiation impulse, the number of hot spots can increase
to such an extent that the reaction appears as quasi-homogeneous [43]. If
the supplied energy fails to exceed the critical value needed to produce det-
onation, then pre-explosion phenomena (such as luminescence, formation of
caverns) can be observed.
Luminescence and conductivity of crystals of heavy metal azides after their
irradiation were successfully used in studies of their pre-explosion state [43–
50, 52]. The authors of [49] summarized their results in the following way:
The explosive decomposition of these azides is composed of two stages: a pre-
explosive stage and the stage of the explosion products dispersion. The role
of the initiating pulse is to create the necessary density of free electrons.
At the pre-explosive stage the actual processes are of chain nature and pro-
vide a smaller portion of energy-release (up to 20%), which is connected
with the non-radiative trapping and the recombination of electrons and holes,
multiplying by chain mechanism. The objective of this stage is to provide
the primary heating (up to 600 K) and the dispergation of the model, that
means to set conditions for the second stage of the product dispersion. At
this stage the majority of the energy connected with the exothermic reac-
tion 2N3 → 3N2 is released. The reaction is more preferable on the surface
or in the gas phase than inside a crystal. This model of the process of chain
multiplication of holes due to their intraband transitions into various energy
states [49] was also supported by quantum-chemical calculations using the
Hartree–Fock method [45]. The said model, which is consistent with cur-
rently available experimental evidence, can be presented by the following
quasi-chemical reaction [49]:
h + A → N3 0 + A (3a)
N3 0 + N3 – → N6 – (3b)
N6 – + Ag + → N6 0 + Ag 0 (3c)
N6 0 → N6 – + h∗ (3d)
h∗ → h + h + e (3e)
h + A + N3 – + Ag + → 2h + e + N6 – + Ag 0 , (3f)
where h and h∗ are thermal and hot holes, respectively. Here Eq. 3a repre-
sents hole trapping by a cation vacancy, Eq. 3b reconstruction of the centre
as a result of the chemical reaction with the formation of a quasi-local hole
state in the valence band, Eq. 3c hole detrapping from the reconstructed
centre, Eq. 3d bearer multiplication as a result of impact ionization by hot
holes and Eq. 3e reconstruction of a local state in the band gap. Accord-
Sensitivities of High Energy Compounds 205
ing to final Eq. 3f, an extra electron-hole pair is generated at the expense of
the decomposition of lattice fragments next to A (the formation of N6 – and
Ag0 ) [49].
According to the authors of paper [50], however, within the framework of
the above-mentioned model it is difficult to explain the dependence of the
ignition energy threshold on the excitation pulse parameters and excitation
type as well as the occurrence of the induction period. They suppose [50]
that the kinetics of initial processes of heavy metal azide decomposition
upon exposure to laser pulses, pulsed electron beams, and individual mi-
croparticles have demonstrated that they possess some features in common.
Severe local deformation of the lattice initiate reactions between lattice an-
ions, and energy of ∼ 10 eV is liberated as a result of each individual reaction
act. When a high density of such reactions is observed simultaneously in
a small volume, the resultant heat liberation in this region can be sufficient for
the avalanche process development culminating in the formation of a shock
wave with a pressure > 1010 Pa and explosive decomposition of the given
sample [50].
One of the ways of initiation by a laser beam consists of its action on
EM placed in a metal foil sheath. The authors of [52] point out that wrap-
ping of silver azide crystals in aluminium foil of 10 µm thickness results in
disappearance of cathode luminescence, but has virtually no effect on the
kinetics of explosion fluorescence (that corresponds with the kinetics of ex-
plosive decomposition in that part of the sample in which this decomposition
is initiated). The authors of paper [53] have found that surface roughness and
a thin metal layer are very effective at enhancing absorption of laser energy
and produce a very strong shock wave in ambient media. They have suc-
ceeded in the detonation of a very slender PETN powder column almost in the
steady propagation mode. A calculation of the laser initiation of thin PETN
(density 1.0 g cm–3 ), using one-dimensional hydrodynamic code to study of
the effect of the initiation methods on the initiation process, is a topic of
paper [54]. Three types of the initiation are modelled: utilizing direct absorp-
tion of laser energy, impact of the laser driven-flyer plate, and ablation of
the aluminium film. In the last case the initial pressure in aluminium is ex-
tremely high, after 0.05 mm propagation of shock wave, the pressure at the
shock front falls to 4 GPa. During propagation to intermediate particles, the
decomposition of PETN can not affect the shock front, so the pressure de-
crease continues [54].
A presentation of direct experimental evidence for the initial fast decom-
positions of the more powerful military explosives, HMX, RDX and TNAZ, as
well as of a non-explosive molecule, dimethyl nitramine (DMNA), is a topic
of paper [55]. Experimental data indicate that electronic excitation, in the
gas phase, of all the nitramines studied, with 248 nm, 15 nanosecond pulses
from the excimer laser leads to the formation of electronically excited NO2
(2 B2 ), resulting from the primary N – N bond scission in the electronically
206 S. Zeman
2.4
Electrostatic Discharge Initiation
This initiation has not attracted much attention in the past. Nevertheless, the
electric spark sensitivity is an important characteristic of explosives, espe-
cially primers and propellants. Important results were published by Skiner
Sensitivities of High Energy Compounds 207
2.5
Effect of the Hydrostatic Compression
2.6
Effects of Crystal Defects
dependence of the rate of process, which follows the Arrhenius law in the
limit. However, among the parameters of the Arrhenius equation also values,
that determine the temperature dependence of generating of dislocations, are
predominant. The concentration of dislocations and degree of transformation
depend exponentially on the distance from the front of reaction propaga-
tion. This means that the width of the zone of chemical reaction increases
with temperature more weakly as the velocity of propagation of the reaction
front [11]. In this light a paper by Kunz et al. [59, 65] can be mentioned: on
the basis of their study of laser pulse initiation of azides [54] they suggest
that chemical decomposition in crystals with a low density of dislocations
develops noticeably more slowly than that in the highly defective samples,
showing a strong correlation between density of dislocations in the crystal
and chemical reactivity [59, 65]. Kuklja and Kunz [66] have experimentally
determined configurations for edge-type dislocations with the Burgers vector
(001) in crystalline cyclotrimethylene trinitramine (RDX) and pentaetythri-
tol tetranitrate (PETN), and theoretically simulated them. It is shown that
a shear stress, induced by the dislocations, produces local electronic states
in the fundamental band gap of the crystal. These states are mainly formed
by molecular orbitals of critical bonds (which are the N – NO2 group in RDX
and the O – NO2 group in PETN) responsible for the stability of the materi-
als. Kuklja et al. [67] also studied structural and electronic features of shear
strain in the molecular crystal of 1,1-diamino-2,2-dinitroethylene (FOX-7)
and investigated its effect on decomposition of the material. The authors
demonstrate that shear strain lowers the decomposition barrier and narrows
the band gap of the solid and thus facilitates thermal chemistry in molecular
materials. The dynamic behaviour of the band gap contains a rich variety of
information about the details of the decomposition processes at the initiation
stage.
It is generally known that if the thermal effect of exothermic decompos-
ition of a given EM exceeds the thermal losses of this process, then the given
material will explode. In this context, Armstrong’s review article [64] presents
linear relationships between the size of the above-mentioned centres of ini-
tiation of reaction (generally, defects of crystal lattice that are also called
“hot spots”) in RDX and PETN crystals and reciprocal critical temperature
of these explosives. Armstrong also states [63] that local concentration of
strain energy within the relatively more stable energetic crystals is required to
induce intramolecular decomposition, relating to (intermolecular-type) hy-
drogen and (intermolecular) nitrogen/oxygen bond breakages. Dislocation
pile-ups provide an explanation of crystal size effects in energetic crystal
decomposition. Such molecular crystals give enhancement of local heating
from plastic work because of their insulating character [63].
According to Coffey [68, 69] an initiation of chemical reaction in explosive
crystals occurs due to the local energy dissipated in the crystals during plastic
flow arising from shock or impact. The dislocations created during the flow
210 S. Zeman
are held responsible for producing lattice and molecular distortions within
the crystals. A distorted lattice potential is developed through which the dis-
locations move by quantum tunnelling [69]. These processes together with
energy localization are likely to be responsible for detonation. If somehow the
plastic flow in the explosive crystals is stopped, even during full detonation,
the energy dissipation due to plastic flow and the chemical energy release also
stop regardless of the applied pressure (this is Dead Pressing) [68]. Coffey
developed the theory of plastic flow in detail and summarized his results in
a chapter [68] of monograph [2].
A significant role during initiation is ascribed to dislocations also in the
theory of electronic initiation developed by Kuklja et al [15, 45, 59, 65, 70–72].
The authors used ab initio simulation (Hartree–Fock level) [15, 45, 59, 65, 70–
72]. They found [70] that low formation energy characterizes defects in the
solid RDX. This indicates that a significant number of vacancies should be
present in the crystal even at low temperatures. Narrowing of the funda-
mental gap (about 1 eV) is caused by the presence of vacancies in the ma-
terial. This fact demonstrates anisotropy of the RDX crystal with respect to
the vacancy distribution [70]. In another paper [72], Kuklja et al. deal with
structure and properties of molecular vacancy dimers in the RDX crystal.
The three nearest-neighbour vacancy dimers have positive binding energies,
i.e. vacancy interactions are attractive. Higher vacancy association may also
be energetically favoured. High RDX anisotropy manifests in the different
binding energies and optical gap narrowing for the different mutual orien-
tations of vacancy dimers [72]. These authors also stated [71] that a key
role belongs to the edges of dislocations, which promote dramatic changes
in the electronic structure, primarily a reduction of the optical gap due to
the splitting off local electronic states from both valence and conduction
bands.
The crystal lattice defects also include cracks, either present in the EM
crystal or produced in it during initiation of ongoing decomposition. Holmes
et al. [73] presented a model for time and spatial dependences of the “heating
of molecular vibrations” and the possible initiation of chemical reaction from
heat dissipated in the vicinity of a propagating crack in a molecular crys-
tal. Energy from a moving crack tip is released as phonons in proximity to
the crack. Initially, the phonons and the molecular vibrations are not in ther-
mal equilibrium. Subsequently, there is a competition between excitation of
molecular vibrations by multiphonon up-pumping and diffusion of phonons
from the crack region. If the coupling between the locally hot phonon bath
and the molecular vibrations is sufficiently large, a transitory high vibrational
temperature will be achieved prior to eventual thermal equilibration with the
bulk of the crystal. The authors found that the peak vibrational temperature
can be sufficiently high for a significant time period for chemical reaction to
occur. This temperature change is connected with a production of chemical
reaction in EM [73].
Sensitivities of High Energy Compounds 211
3
Impact and Shock Sensitivities
3.1
Model of the Multiphonon Up-pumping in Energetic Materials
3.2
Impact Sensitivity
3.2.1
Low Velocity Initiation
In the case of low velocity initiation Dlott [74] arrived at a preliminary con-
clusion that the sensitivity to this impulse might increase in materials charac-
terized by fast up-pumping (a connection between fast up-pumping and poor
thermal conductivity was suggested in this case). In Dlott’s opinion [1] the
phonon generation and up-pumping under conditions of LVI could be inves-
tigated, where one obstacle is accurately describing the localized generation
of phonons under LVI conditions. McNesby and Coffey developed a simple
theory [75] for predicting impact sensitivities in crystalline explosives from
Raman vibrational spectra measured at room temperature. They suggest that
the initial rate of energy transfer, from states within a low-energy phonon and
near-phonon to a state within higher vibrational levels, plays a role in deter-
mining impact sensitivities of the explosives examined in their paper [75].
Shuji Ye et al. [76] estimated frequencies of the normal mode of 12 differ-
ent individual explosives by means of density functional theory calculations
at the B3LYP/6-31G(d) level of theory and evaluated the number of doorway
modes (i.e. the lower frequency vibrations with significant bandwidths) in
the three doorway regions. They compared the number of doorway modes
with the impact sensitivities (from the hammer test). A qualitative correlation
found indicates that the number of vibrational states in the doorway region
can be an important parameter for the intrinsic impact sensitivity of solid
molecular explosives.
3.2.2
Aspects of the Characteristics of Molecular Structure
tween log h50% and the heat of detonation, obtained by B3LYP/6-31G∗ level
of theory (no carbon monoxide path), was published by Edwards et al. [86].
These authors also published relationships between drop height h50% and
the lowest unoccupied orbital molecular energy (B3LYP/6-31G∗ ), between
the said height and the highest occupied molecular orbital energy (B3LYP/6-
31G∗ ), and between the said height and other quantities, all that without any
substantiation or other interpretation.
The PM3-SCF-MO calculation was used to study the impact sensitivity
of 5-substituted tetrazoles and their salts [87]. The authors found a parallel
relationship between the activation energy and the sensitivity and a good lin-
ear relationship between the experimental impact sensitivity of metal salts
and the reaction activation energy. However, no parallel relationship was
found between activation energies and the bond orders between nitrogen
atoms in positions 1 and 2 [87]. This bond should be primarily split during
thermal decomposition of the said heterocycles [11]. The relative magni-
tude of sensitivity for most tetrazole compounds is determined mainly by the
activation energy of the opening ring in the thermal decomposition. The sen-
sitivity found order for 5-substituted tetrazoles was as follows (order of sub-
stituents): CH3 ≈ Ph < NH2 < NHNO2 < Tz < Cl < NO2 < N3 < N2 – [87]. From
the same workplace also comes UHF MO PM3 and PM1 calculation of pyrol-
ysis mechanism of nitromethane, methyl nitrate, nitroguanidine, nitramine,
methylnitramine and dimethylnitramine [89]. The mechanisms of primary
homolysis of the given compounds stand in accordance with experimen-
tal findings [11]. The found activation energies of this process are consis-
tent with the experimental values of impact sensitivity [89]. The authors
found a parallel linear relationship between the bond orders of N – NO2
bonds in the molecules of studied nitramines and the activation energies
of their initiation reactions breaking these bonds [89]. Türker investigated
structure-impact sensitivity of some substituted 1,3,5-trinitrobenzenes by
using a topological approach [88]. He found that the sensitivity increases with
decreasing of a defined upper bound of the deepest lying molecular orbital
energy, or the fourth coefficient of the secular polynomial of the molecular
graph of the iso-conjugate hydrocarbon system within the Hückel MO frame-
work [88].
3.2.3
Specification of Centres of Initiation Reactivity in a Molecule
negative QNO2 being connected with the lowest sensitivity [90, 91]. Accord-
ing to the authors’ opinion the compound may be sensitive (h50% ≤ 0.4 m for
a hammer weight of 2.5 kg) when its nitro group has a lower negative charge
than about 0.23 e. The method is applicable to almost all nitro compounds in
contrast to the methods of length of the C – NO2 , N – NO2 , or O – NO2 , mo-
lecular electrostatic potential, and oxygen balance [90, 91]. However, in some
cases the selection of the most reactive groups in a molecule differs from ex-
perimental findings: especially in the cases of TNT and 2,4,6-trinitroaniline,
the authors choose for the correlation the most negative QNO2 values, i.e. the
nitro groups at the 4-position of the given polynitro arenes, which contradicts
experimental findings (Scheme 2). Nevertheless, these are seminal papers in-
spiring further activities in this area of specific chemical (initiation) reactivity
of EMs.
3.2.4
Effect of Desensitizing Admixtures
3.2.5
Methods of Prediction
From among the prediction methods of impact sensitivity, three methods can
be mentioned here. Vaullerin and Espagnacq discuss a criterion that aids in
such screening for all families of EMs [94]. This criterion is based on the
216 S. Zeman
3.3
Initiation by Shock
3.3.1
Physical Kinetics Model
In 1998 one of Wolker’s most recent papers [97] argues for his theory of phys-
ical kinetics of detonation. The theory is described as a non-equilibrium,
non-thermal process elicited by the exceedingly high kinetic energy of shock
and/or in the detonation front. This process leads to massive mechanical frac-
ture of covalent bonds at and near to the detonation front. According to this
theory, the reaction rate is determined by the medium vibrational energy
of free electron pairs (radicals) formed by the fracture of a molecule within
a time period of 10–14 to 10–12 s [40]. This theory can be considered as an ex-
ample of views of some physicists on chemical reactions, but in its time it was
a considerable contribution to development of detonation theory.
3.3.2
“Accumulation Mechanism” Application
3.3.3
The Multidimensional Reactive Flow Models (NEZND)
3.3.4
Application of Raman Spectroscopy to Study
Like McNesby and Coffey in the case of impact sensitivity [75], Koshi
et al. [105] have also used Raman spectra (but only for very low temperatures)
in the characterization of relationships between the shock/impact sensitivi-
ties and molecular structure. The authors start from Dlott’s ideas (see above)
of the transition of energy of excited lattice vibrations (phonons) to molecu-
lar vibrations before bond breaking can occur. This energy transfer process
is expected to be the rate-determining step for the explosion. The rate of this
transfer has been evaluated by the authors on the basis of a simple theory
coupled with experimental measurements of line width of Raman spectra at
0 K (the line was obtained by extrapolation of the spectra recorded within
the temperature interval of 3.8–300.0 K). A good correlation was found be-
tween the sensitivities derived from drop-hammer tests and the rate of energy
transfer. However, no correlation was obtained for the data of TNT (for its pri-
mary splitting see Scheme 2) [105]. This is understandable since the authors
did not include it among those EMs whose primary step of initiation consists
of homolysis of C – NO2 bonds (1,3-dinitrobenzene), N – NO2 bonds (HMX,
RDX, NQ) and O – NO2 bonds (PETN and propyl nitrate). The authors have
compared also the rates of energy transfer with wedge test results (as a meas-
ure of the shock sensitivity) [105]. In this case, loading pressures were very
high, and the effect of pressure on this transfer has to be considered. Taking
the increase of anharmonic coupling into account, the rates of energy trans-
fer behind the shock can be compared with the induction distance in wedge
tests [105]. Apart from the above-mentioned effect of chemical mechanism of
primary fragmentation, there is a clear difference between the initiation by
impact and that by shock in this case; the authors [105] do not point out these
two important facts.
Sensitivities of High Energy Compounds 219
3.3.5
The Excitonic Model
3.3.6
Influence of Crystal Characteristics
White et al. [109] report that crystalline structure has little effect on the initi-
ation of EMs by a shock wave. van der Heijden et al. [110], using the example
of RDX, HMX and HNIW, state that the following crystal parameters play
a role in determining the sensitivity towards a shock stimulus: (a) internal
product quality, (b) mean particle size, (c) surface smoothness/shape of the
explosive particle. Like the impact sensitivity, also the shock sensitivity is af-
fected by the density defect content (dislocations, grain boundaries, voids,
impurities, inclusions). The shock initiation tests with HMX, for example,
have clearly demonstrated a relationship between the average crystal density
and shock initiation pressure. The findings from this area are very important
for practice, because a modification of quality and crystal shape of, for ex-
ample, RDX can give a product with increased resistance to impact and shock
(I-RDX [110]).
The initiation depends on crystal orientation in energetic materials such
as PETN [109, 111, 112]. Using the anharmonic potentials [74], Jindal and
Dlott have calculated the directional compressibility, the mode Grüneisen pa-
rameters, and the bulk Grüneisen parameters along different crystallographic
directions for naphthalene [111]. They show that the temperature increase is
quite a bit greater for a shock of given pressure along the a or b axes than
along the c axis, mainly because the compressibility along the c axis is smaller.
With the use of thermochemical data for EMs it is shown that the temperature
increase anisotropy is large enough to lead to reaction rates which differ by
several orders of magnitude, when typical initiating shocks are incident along
different crystallographic directions [111]. PETN is sensitive along the shock
propagation perpendicular to the (113) plane, but highly insensitive along the
shock propagation perpendicular to the (102) plane [112].
3.3.7
Effect of Desensitizing Admixtures
3.3.8
Some Types of Chemical Behaviour of Energetic Materials
Under Shock Wave
3.3.9
Problems of Nitromethane Initiation
molecular course of this decomposition [11]). Paper [130] discusses both pos-
sible ways and intermolecular mechanisms of formation of the said acids. The
mentioned transfer of hydrogen atom can also be evoked by amines [130],
whose presence sensitizes both primary and secondary nitroparaffins to stim-
uli [129, 131]. In the case of nitromethane sensitized with amines, Gruzdkov
and Gupta suggested the following mechanism of initiation [132]:
As shown, the first stage should be, according to the authors’ opinion, an
acid-base equilibrium. This conclusion was made on the basis of applica-
tion of time-resolved optical spectroscopic techniques and action of shock
(12–17 GP) upon an NM layer of 0.2 mm thickness. Similar conclusion was
arrived at by Woods et al. [133] on the basis of laser-initiated decompos-
ition of single aerosol particles of NM and NM-diethyl amine mixture in the
source region of a time-of-flight spectrometer, followed by vacuum UV laser
ionization (products identification by mass spectroscopy). In comparison to
the above-mentioned enhancement of detonation parameters, an influence of
amines on the failure diameter of nitromethane [129] is better known and
more studied [127, 136]. Also this effect is connected with the increase in re-
action rate in the reaction zone of detonation (i.e. with reduction of the zone
length) due to influence of amines on nitromethane [136]. This fact might
indicate that the primary fragments entering the reaction zone of detona-
tion wave in the case of neat NM are different from those in the case of NM
sensitized by addition of amines. It has been known for quite a long time
that the concentration of the nitromethyl aci-anion (CH2 NO2 – ) in the NM
is increased not only with the addition of bases [130] but also by applica-
tion of static high pressure [134] (see also the above-mentioned paper [123]).
However, evidence for an increase in concentration of the anion in neat ni-
tromethane under shock conditions (dynamic high pressure) has been found
lacking [135].
Sensitivity of NM is also increased by additives that are inert to it and
which can represent active EMs (RDX, ammonium nitrate) [137], or by
non-active hard powders (aluminium, copper, fumed silica, boron carbide
etc.) [131, 137]. The amounts of the additives are 3 and 40% w/w in [131]
and [136], respectively. The sensitivity of the mixtures is higher than that of
neat NM [131, 137]; however, for example, the sensitivity increase achieved
by addition of 3% w/w of boron carbide is less efficient than the effect of
the same amount of added diethylenetriamine [131]. The increase can be
due [131] to difference in the number density of each in the mixture (dif-
ference in the number of available hot spots) or to an impedance mismatch
difference, or to a function of both variables.
Sensitivities of High Energy Compounds 225
3.4
Relation Between Performance and Sensitivity of EMs
4
Sensitivity of EMs from the Standpoint of Physical Organic Chemistry
(POC Model)
azides [43–50]. Also the development in the area of new individual EMs
mostly advances with orientation to organic polynitro compounds [51, 84,
140–143], and more recently it is intensively directed also to organic high ni-
trogen EMs (see, e.g., [142–144]). Thus, this special area of organic chemistry
plays a highly significant part in the study and development of EMs. In this
sense, studies of sensitivity of EMs, mainly organic polynitro and polynitroso
compounds, are also conducted at the workplace of the author of this review
article.
4.1
Approach to Solution of the Initiation Reactivity
It is a generally accepted idea that nitro groups represent the primary cause
of sensitivity (initiation reactivity) of polynitro compounds (see, e.g., [6–
20, 59, 90, 91]). Sensitivity of a given polynitro compound to stimuli results
from the type of bonds to nitro groups in the molecule, their number, the type
and intensity of their interaction with the rest of the molecule, and intermole-
cular interactions. Recently relationships have been found between outputs of
non-isothermal differential thermal analysis (DTA, i.e. low-temperature ther-
molysis) and characteristics of detonation of polynitro arenes [145, 146] and
also nitramines, nitrosamines and nitrate esters [146]. A more detailed an-
alysis of these results for polynitro arenes shows that their classification in
the sense of the relationships found is given primarily by (a) steric conditions
and (b) electron configuration in the ground state of the reaction centre of
the molecule [145, 146]. In this context, the term reaction centre means the
grouping of atoms and/or functional groups in the molecule whose primary
chemical changes initiate decomposition of this molecule (e.g., see Schemes 2,
3a, 3b, 7 and 8). At the same time, the said facts represent one of the basic
principles of approach of organic chemistry to dealing with reactivity prob-
lems in general.
4.1.1
Methods of Specification of the Reaction Centre in Molecule
Both the electron configuration and steric situation in the reaction centre of
the molecule can be characterized from the standpoint of organic chemistry
on the basis of:
• NMR chemical shifts of key atoms of the reaction centre of the molecule
(semi-empirical determination of the nitro groups that are first to react in
the initiation) [6, 7, 10, 16, 98, 148],
• direct correlation of reaction characteristics with electron charges at the
nitrogen atoms of these first-reacting nitro groups [7, 12, 152, 160, 161] or
with net charges of these nitro groups [90, 91, 160].
Sensitivities of High Energy Compounds 227
For the first time, the approach given was demonstrated on a group of sec-
ondary nitramines [7, 10, 98], because these compounds have relatively sim-
ple molecular structures, and the mechanism of their primary homolysis
is well known [11, 14] (see also Scheme 1). On the basis of these findings,
the approach was extended to the application of 13 C NMR [6, 16, 148] and
15 N NMR [150] chemical shifts in polynitro arenes. It can be argued that in
4.1.2
Selection of Sensitivity (Reaction) Characteristics of EMs
4.2
Initiation by Heat
Fig. 1 Relationship between the activation energies, Ea , resulting from the Russian mano-
metric method, and the 15 N NMR chemical shifts of nitrogen atoms of the nitro groups
(non-isochronous molecule HNIW correlates through the shift of the nitro–nitrogen atom
in the position 2)—taken from [7]
Sensitivities of High Energy Compounds 229
Figure 1 shows that in the case of HMX the effect of the liquid phase on
its thermolysis should be absent even if the thermolysis is realized within
a temperature range above 553 K (see discussion in [10]). An analogous de-
pendence for polynitro arenes is presented in Fig. 2 [151].
Figure 3 presents the relationship between Ea /R (from Eq. 7) and elec-
tronic charges, qN , at nitrogen atoms of the first-reacting nitro groups for
nitramines taken from [152]. The discussion of this graph confirmed the dif-
ferences between linear (line D) and cyclic nitramines (lines A, B and C) [152]
from the standpoint of the initiation energy transfer to the reaction centre.
The quantum chemical calculation itself revealed non-equality of nitrogen
atoms in nitro groups from the point-of-view of their potential participa-
tion in primary splitting processes of initiation [152], which was only further
clearly confirmed by the dependences in Fig. 4. A similar approach applied
to highly thermally stable polynitro arenes resulted in the relationships be-
tween Ea /R and qN values depicted in Fig. 4 [160]. Also in this case one can
clearly see very close molecular-structure dependence and, in the cases of
TNB and HNB, also effect of state of aggregation (both these polynitro arenes
decompose in the liquid state).
On the basis of relationship between electron charges, qN , at the nitrogen
atoms of the first-reacting nitro groups and DTA onsets, TD , of exothermic
decomposition of polynitro arenes in Fig. 5, a set of these substances are di-
Fig. 3 Relationship between the slopes of the Kissinger relation Eq. 7, Ea R–1 , and Mul-
liken B3LYP/6-31G∗∗ charges, qN , at the nitrogen atoms of the primarily reacting nitro
groups in nitramino groupings in the molecule (given in parenthesis are the positions in
the molecule which correlate with the non-isochronous molecules). Taken from [152]
Fig. 4 Relationship between the slopes of the Kissinger relation Eq. 7, Ea R–1 , and Mulliken
B3LYP/6-31G∗∗ charges, qN , at the nitrogen atoms of the primarily reacting nitro groups
in molecules of high-thermostable polynitro arenes (decomposition of TNB and HNB in
the liquid state; given in parenthesis are the f positions in the molecule). Taken from [160]
Sensitivities of High Energy Compounds 231
Fig. 5 Relationship between the qN values of the nitrogen atoms of the primarily react-
ing groups and onsets, TD , of exothermic decomposition. Codes of the genuine polynitro
arenes are written in boldface type. Taken from [161]
vided into several logical subsets [161]: one with the dominant stabilizing
effect of the crystal lattice (line D), one in which the effects of the mechan-
ism of primary homolysis (lines A, C, E) dominates, and one characterized
by combination of both above-mentioned effects (line B). An addition of sol-
vent (TNB) to substances of class D will suppress this stabilizing effect: TENN
and NONA move to class C (with primary homolysis of the C – NO2 bond),
TATB moves to class E (splitting in the sense of Scheme 2) and the stability of
TPT decreases so markedly that it does not correlate with any of the classes in
Fig. 5.
From the results in paper [6] it follows that a molecule of energetic ma-
terial can contain more than one potential reaction centre of initiation. As
an example, we can give N,N ,N -tris(3-methyl-2,4,6-trinitrophenyl)-1,3,5-
triazine-2,4,6-triamine (TMPM) with potential reaction centres in the vicin-
ity of nitrogen atoms N9 (position 6 in trinitrophenyl group) and N12
(position 2 in the same group) [6]; the data of the second reaction cen-
tre mentioned approach those of substances of class A, which means that
the primary splitting by heat should go by the classic trinitrotoluene mech-
anism (see also Scheme 8). The existence of this type of relationship (i.e.
Figure 5) might be connected with the electrostatic interaction of “instan-
taneous point dipoles” at the reaction centre of their molecules (discussion
of this problem in [161] is unfolded from the Einstein–Nernst relation-
ship [162]).
232 S. Zeman
4.3
Impact Sensitivity
4.3.1
Impact Sensitivity as the First Reaction
Fig. 6 Relationship between the drop energy, Edr , of “the first reaction” and 15 N NMR
chemical shifts, δA , of the aza nitrogen atoms carrying the nitro groups that are first to
react in the initiation reaction (given in brackets are the positions in the molecule). Taken
from [16] and complemented by data for Tetrogen, HNIW and TEX [6]
Sensitivities of High Energy Compounds 233
the molecule, and the extent to which the nitrogen lone pair is involved in π-
bonding with the NO2 group. As the conformation and size of the molecule
plays a dominant role in the intermolecular interactions in the correspond-
ing crystal, the interactions should have a significant influence on the impact
sensitivity.
This hypothesis is supported by Fig. 7 (taken from papers [6, 16]), which
represents a relationship between the Edr values and heats of fusion, ∆Hm,tr ,
of nitramines (the heats of fusion ∆Hm,tr , defined in the sense of mono-
graph [164] as the sum of heat of melting and heats of all polymorph tran-
sitions). As the heat represents the work needed for formation of defects in
the crystal lattice, the relationships found seem to stand in accordance with
the ideas about the decisive role of dislocations and plastic deformations of
crystal played in the initiation of energetic materials (Sect. 2.6).
Likewise, for nitramines a relationship exists between Edr values and
13 C NMR chemical shifts, δ , of the most reactive nitro group “bearers”
C
(i.e. carbon atoms in the reaction centres of the molecule) in polynitro
arenes [148].
As shown in Figs. 8a and 8b the relationship has once, again, the same
form as that for nitramines. The more complex molecular structure of
polynitro arenes (particularly the mesomeric and steric effects) and the
two presumed basic mechanisms of primary splitting, as compared with
nitramines, result in a more varied assortment of shapes of this relation-
Fig. 7 Relationship between the drop energy, Edr , of the first reaction and the heat of
fusion, ∆Hm,tr , of nitramines; taken from [16] and complemented by data for TEX [6]
234 S. Zeman
Fig. 8 a Relationships between the impact sensitivity (drop energy of the first reaction,
Edr ) and 13 C NMR chemical shifts of the carbon atoms which bears the primarily re-
acting nitro groups in the polynitro arenes studied (compounds with primary thermal
decomposition via a five-, six- or seven-membered transition state or aci-form) [148].
b Relationships between the impact sensitivity (drop energy of the first reaction, Edr )
and 13 C NMR chemical shifts of the carbon atoms which bears the primarily reacting ni-
tro groups in the “genuine” polynitro arenes (group I) and compounds (group V) with
primary thermal decomposition via a five- or six-membered transition state [148]
Sensitivities of High Energy Compounds 235
Fig. 9 Semilogarithmic relationships between the impact sensitivity (drop energy of the
first reaction, Edr ) and heat of fusion (∆Hm,tr ) of polynitro arenes [148]
4.3.2
Impact Sensitivity Detected by Sound
In this case, the impact sensitivity data were mostly taken from the literature.
Figure 10 represents the relationship between the impact sensitivity detected
by sound and 15 N NMR chemical shifts of nitramines [6]. This relationship, in
comparison with Fig. 6, is mathematically different. The difference could be
interpreted by the different mechanisms of transfer of drop energy to the re-
action centre of the molecule in the case of the first reaction as compared with
the impact sensitivity detected by sound. This difference is evident also in
the relationship between the sensitivity and the ∆Hm,tr values (Fig. 11 taken
from [147]).
The same shape of relationship as shown in Fig. 9 is valid for polynitro
arenes in Fig. 12: here the impact sensitivity detected by sound correlates with
13 C NMR chemical shifts of the bearers of the most reactive nitro groups
in the molecule. According to this diagram, they are the nitro groups in 2-
positions of the DATB, TNX, TNR and BITNT. These groups are also most
sterically crowded [6]. However, in the N, N -bis(2,4,6-trinitrophenyl)-3,5-
dinitropyridine-2,6-diamine (PYX) molecule there exist two potential reac-
tion centres, namely that with participation of the nitro group at position 2
(Eq. 2) in the picryl group and that in position 3 (Eq. 1) of the pyridine ring
(see also Scheme 7).
The dependences given for polynitro arenes in Fig. 13 are analogous to
those given for nitramines in Fig. 10. They are discussed in [147]; we should
Sensitivities of High Energy Compounds 237
Fig. 10 Relationship between the drop energy, Edr , (impact sensitivity detected by sound)
and 15 N NMR chemical shifts, δA , of the aza nitrogen atoms carrying the first reacting
nitro groups in the molecules of nitramines studied [6]
Fig. 11 Relationship between the drop energy, Edr , (impact sensitivity detected by sound)
and heat of fusion, ∆Hm,tr , of nitramines; taken from [147]
draw attention here to the effect of the introduction of the hydroxyl group into
the TNB molecule (line D), amino group (line A), alkyl groups (line B), or
“doubling” of trinitrophenyls (line C).
238 S. Zeman
Fig. 12 Relationship between the energy, Edr , (impact sensitivity detected by sound)
and 13 C NMR chemical shifts, δC , of the carbon atoms of the aromatic system carry-
ing the most reactive nitro groups (the numbers in brackets denote the positions in the
molecule) [165]
Fig. 13 Logarithmic relationships between the impact sensitivity, Edr , (impact sensitivity
detected by sound) and heat of fusion, ∆Hm,tr , of polynitro arenes [147]
Fig. 14 Relationship between impact sensitivity and net charges of primarily reacting
nitro groups in molecules of highly thermally stable polynitro arenes (the numbers in
brackets denote the positions in the molecule) [160]
densed analogues (i.e. NTFA, TPT and TACOT-Z). The data of TATB do not
correlate with either group due to the above-mentioned singularity of pri-
mary fission of its molecule.
4.3.3
Allocation of Polynitro Compounds
on the Basis of their Impact Sensitivities
Fig. 15 Relationships between the drop energies of the impact sensitivity detected by
sound and the drop energies of the first reaction [155]
4.4
Initiation by Shock
Fig. 16 Relationship between the square of the detonation velocity, D2 , and 15 N NMR
chemical shifts of nitrogen atoms of the primarily reacting nitro groups [6, 7, 98]
Fig. 17 Relationships between the heats of explosion, Qreal , and 15 N NMR chemical shifts
of the nitrogen atoms of the primarily reacting nitro groups in polynitro arenes [150]
represents substances with primary homolysis of the C – NO2 bond, and the
C – S bond in the case of DMDIPS.
4.5
Relationship Between Sensitivity and Heat of Explosion
As already stated above, Rice and Hare [85] found an exponential depen-
dence, and Edwards et al. [86] found a semi-logarithmic dependence between
impact sensitivity and heat of explosion, Q, without trying to interpret these
facts. Also reported is a relationship between the ratio Q.DH –1 and logarithm
of drop height h50% [84] (in this case DH stands for the dissociation energy of
the weakest bond of the nitro group in the molecule).
From what has been said so far, and from the published papers [6, 7, 9, 14,
16, 98, 145, 151–153] it follows that there exist logical relationships between
the characteristics of low-temperature thermal decomposition and those of
initiation and detonation, respectively. The homolytic character of primary
fission in both the detonation and low-temperature thermal decompositions
of energetic materials (for relevant quotations, see [9]) was a motive for Ze-
man to use the Evans–Polanyi–Semenov equation (E–P–S) [168] to study the
chemical micro-mechanism governing initiation of energetic materials [9].
A relationship formally similar to the E–P–S equation can also be obtained
by mutual comparison of the dependences shown in Figs. 2 and 17; it has the
following form:
Ea = α · Q + β , (8)
and application of the relationship between detonation velocities and explo-
sion heats Q [138] in the form
D2
Q= 2 , (9)
2 γ –1
where γ is the polytropy coefficient, transforms Eq. 8 into the following form
Ea = a · D2 + b , (10)
which can also be derived by mutual comparison of the dependences in
Figs. 1 and 16. Equations 8 and 10 were called a modified E–P–S equation by
Zeman [9]. The original E–P–S describes a relationship between activation
energies E of most substitution reactions of free radicals and corresponding
heats of reaction ∆H of narrow sets of substance structures. The equation
shows that the strength of the bond being split is a decisive factor in the given
reaction. In both Eqs. 8 and 10 the energy, E, can be the activation energy of
thermal decomposition, Ea [7, 9, 98], the slope Ea · R–1 of the Kissinger rela-
tionship Eq. 7 [153, 169, 170], the energy of electric spark, EES [175, 176, 179],
drop energy, Ep [160], it may be substituted by the charge, qN , at the nitrogen
Sensitivities of High Energy Compounds 243
atom of the most reactive nitro group [12] in the molecule, by the net charge
of this nitro group [160], or by half-wave polarographic potential [159]. The
following text presents several typical examples.
Figure 18 presents the relationship (Eq. 8) for nitramines, in which the Ea
values for cyclic structures correspond to their decomposition in the solid
state.
The validity of Eq. 8 was also successfully verified for the thermal decom-
position of inorganic azides and fulminates [9]—see Fig. 19. For azides
the literature gives activation energy values, Ea , for thermal decomposition
within various temperature ranges. In the sense of Eq. 8, however, only those
Ea values correlate that correspond to the lowest experimental temperature
ranges [9]. This is a very important finding.
Figure 20 has been taken from [9] and presents Eq. 8 in which the Ea
values resulted from the Russian manometric method for polynitro arenes
and Qreal values were calculated according to the Pepekin et al. semi-empirical
method [157]. It must be stated that the Ea value of TATB, obtained on the
basis of the Russian manometric method (i.e. in vacuum, see [151]) in a tem-
perature region above its hypothetical melting point [171], correlates with the
data for thermal decomposition in the liquid state (this Ea value, which is very
close to the heat of sublimation of TATB, is discussed in [171]). The data of
TATB obtained from DSC (see [151], i.e. from measurements at atmospheric
pressure) logically correlate with DATB and PAM data in Fig. 20.
Fig. 18 Modified E–P–S equation (Eq. 8) for the relationship between the activation
energies, Ea , corresponding to the outputs of the Russian manometric method for low-
temperature thermolysis of nitramines, and the real heats of explosion, Qreal [9]
244 S. Zeman
Fig. 19 Modified E–P–S equation (Eq. 8) for azides and fulminates [9]. For azides the
literature gives activation energy values, Ea , for thermal decomposition within various
temperature ranges; however, the Ea values that correlate correspond only to the lowest
experimental temperature ranges [9]
Fig. 20 Graphical representation of the modified E–P–S equation (Eq. 8) [9] for polynitro
arenes with a hydrogen atom at the γ -position with respect to the nitro group (Scheme 2);
exceptional are the polychlorinated derivatives of 1,3,5-trinitrobenzene—TNB (i. e. CTB,
DCTB, and TCTB) whose thermal decomposition, however, represents a certain analogy
with the decomposition of polymethyl derivatives of TNB [9]
Sensitivities of High Energy Compounds 245
Fig. 21 Graphical representation of the modified E–P–S relation (Eq. 10) between the
slope, Ea R–1 , of the Kissinger relationship (Eq. 7) and the square of the detonation vel-
ocity of nitramines [172]
246 S. Zeman
Fig. 22 Graphical representation of the E–P–S relation (Eq. 8) between the heat of explo-
sion and the impact sensitivity of highly thermally stable polynitro arenes [160]
Fig. 23 Relationship between the square of the detonation velocity and the impact sen-
sitivity of highly thermally stable polynitro arenes [160]. Data for TATB do not correlate
due to discrepant primary fission (Scheme 2)
low 30% w/w. In particular, the group around line B indicates that a critical
amount of admixtures of nitrate esters or nitramines exists for the mixtures
(i.e. about 30% by wt.): the decomposition of explosive mixture is not influ-
Sensitivities of High Energy Compounds 247
Fig. 24 Graphical representation of the E–P–S relation (Eq. 10) between the squares of the
experimental detonation rates of explosive mixtures and the slopes, Ea R–1 , of Kissingers
relationship (Eq. 7)—taken from [153]. Black points correspond to dynamites with a ni-
tric ester content of over 30 wt %. Oxidizing systems of all of these explosives are based
on ammonium nitrate (AN) and its mixtures with sodium nitrate (SN), potassium nitrate
(PN), calcium nitrate (CN), aluminium nitrate (AlN), lithium nitrate (LiN), and sodium
chloride–sodium nitrate (NaCl/SN), nickel(II) nitrate (NiN), copper(II) nitrate (CuN) and
cobalt(II) nitrate (CoN)
enced by the physical state (i.e. solution or crystalline) of the oxidizing system
of thus fortified mixtures only in the case when the content of the admixtures
is above the critical value. With the line B also data of commercial gelatinous
explosives correlate whose nitrate esters content exceeds of 30% (black points
in Fig. 24).
This effect is analogous to the well-known influence of the shock pres-
sure on initiation and propagation of the detonation in which the critical
pressure plays a decisive role [101]. If the initiation pressure is lower than
the critical value, then the initiation and growth of detonation themselves
will depend on the micro-structure of the given charge (hot spots). If the
pressure is higher than the critical value, then the initiation will depend
only on the chemical composition of the charge [101]. The pressure dur-
ing the detonation process is also given by the amount and reactivity of
the molecular fragments which come into the reaction zone of the deto-
nation wave. It was stated that in studied explosive mixtures the thermal
reactivity of the oxidizing system and/or its mixture with a high explosive
(for example: AN increases the thermal reactivity of RDX) replaces the pri-
mary thermal reactivity of explosophore groups in the individual energetic
materials [153].
248 S. Zeman
4.6
Electric Spark Sensitivity
Two different instruments are available for studies of electric spark sensitiv-
ity [6, 8, 61, 174–180] at the University of Pardubice: they differ by configura-
tion of electrodes and structure of circuit [6, 8, 174–180]. Scheme 4 presents
the spark gap of the older instrument with code RDAD; Scheme 5 depicts the
newer one with code ESZ KTTV.
The values of measurements carried out with the instrument RDAD were
comparable with those of LANL in Los Alamos [174]. In the papers [175, 176]
the electric spark sensitivities of 23 polynitro arenes and 13 nitramines were
determined as the spark energy, EES , required for 50 per cent initiation prob-
ability. The relationship between the EES values and the squares of detonation
Sensitivities of High Energy Compounds 249
Fig. 25 Graphical representation of the E–P–S relation (Eq. 10) between the electric spark
sensitivity and the square of the detonation velocity for compounds which thermolyze via
a five- or six-membered transition state involving a direct participation of an ortho–nitro
group (Schemes 2 and 3b) [175]
250 S. Zeman
Attention was also focused on the relationship between the EES values and im-
pact sensitivity, expressed as drop energies, Ed , of the first reaction. It was
concluded that, depending on intermolecular interaction factors in crystals of
energetic materials, the mechanism of impact energy transition to the reac-
tion centre of their molecule can differ from that of transition of energy of
electric spark [178]. In an attempt specify the reaction centre in molecules
of polynitro arenes by means of the 13 C NMR chemical shifts of bearers of
primarily leaving nitro groups logical dependences (Fig. 26) were observed
especially in the case of derivatives whose primary reacting nitro group is
conjugated with the rest of the molecule (Scheme 2) [6].
According to Fig. 26 it was confirmed that the most reactive nitro groups
in the TMPM molecule are bound at the 2-position, those in the DMDIPS
molecule at the 6-position, those in the BITNT molecule at the 2-position, and
those in PYX at the 3(5)-position of the pyridine ring [6].
Xiao et al. [179], using the DFT-B3LYP6-31G∗ approach, re-evaluated the
results of [175–177] for nitramines. They confirmed the linear relationship
between D2 and EES values and, in addition, specified another linear rela-
tionship between the logarithm of detonation pressure and EES values. Both
relationships show, that increasing performance is connected with increasing
electric spark sensitivity in technologically attractive nitramines [179].
In comparison with the instrument RDAD, the measurements carried out
with the instrument ESZ KTTV give EES values lower by as much as two
orders of magnitude. The outputs obtained from the two instruments for
nitramines are compared in Fig. 27 [8].
Figure 27 indicates that the initiation mechanisms in the spark gaps of
the two instruments are different. With regard to the considerably higher EES
Fig. 26 Relationship between the spark energy, EES , and 13 C NMR chemical shifts, δC , of
the carbon atoms of the aromatic system carrying the most reactive nitro groups (the
numbers in brackets denote the positions in the molecule) in polynitro arenes containing
amino, hydroxy and alkyl substituents [6]
Sensitivities of High Energy Compounds 251
Fig. 27 Relationship between the outputs from ESZ KTTV and RDAD instruments (see
also Schemes 4 and 5)
Fig. 28 Relationship between the EES values from the ESZ KTTV instrument and the heats
of fusion of nitramines [8]
Fig. 29 Electric spark sensitivity of mixtures of the RDX, HMX and HNIW with crushed
glass
Sensitivities of High Energy Compounds 253
case the decrease in the number of inter-grain points per volume unit. How-
ever, also dislocations in the crystals should have some effect on this type
of EMs. This theory was also confirmed by the electric spark sensitivity of
some nitramines with admixtures of fine particles of hard additives (crushed
glass, diamond powder [180]). The results are documented, for example, by
Fig. 29.
Artificial introduction of fine and hard particles into crystalline nitramines
decreases the electric spark sensitivity of the resulting mixtures, i.e. the
particles behave as a flegmatizing additive, here the number of inter-grain
contact points of the nitramine grains is decreased by incorporation of the
foreign particles [180].
4.7
Application of Polarography
4.8
Inequality of Nitramine and Nitro Groupings in Molecules of EMs
From what has been written so far it follows that the initiation of nitramines
by impact should be dominated by the key role of the aza-atoms carrying the
primarily leaving nitro groups [6, 7]. The dominant factor in the initiation by
shock, electric spark, and in low-temperature thermolysis should be the elec-
tronic structure and surroundings of the primarily leaving nitro group [6, 7].
For nitramine molecules, whose nitrogen atoms are not isochronous, the said
fact is documented in Scheme 6.
In the case of the 2,4,6,8-tetranitro-2,4,6,8-tetraazanonane (OHMX) mol-
ecule, the primary reactivity of inner nitramino groups (positions 4, 6) was
confirmed by Kohno et al. [163] using molecular-dynamic simulation. Their
paper tends to be quoted without mentioning one of its important merits,
i.e. the selectivity in the initiation reactivity of nitramino groups [163]: “in
connection with the action of impact or shock on a nitramine crystal, the in-
254 S. Zeman
Scheme 6 Summary of the findings concerning initiation reactivities, taken from [7], can
be presented as follows: the dominating reactivity of the “inner” nitramino groups of
the OHMX molecule in impact and shock was confirmed by molecular dynamics simu-
lation [163]; the most reactive nitramino group in the HOMO molecule is that at the
1-position and that in HNIW at position 2 (this is the longest N – N bond in the molecule
of the
-polymorph—see [7])
4.9
Several Reaction Centres in a Molecule
The same approach to the TMPM molecule (Scheme 8), i.e. taking into
account the data in Figs. 5 and 26 with application of the same computa-
tion apparatus showed [165] that primary participation of the nitro group
in position 2 together with the methyl group in position 3 during initia-
tion processes in this molecule should be a unique reaction centre (i.e. only
Channel III).
4.10
Comparison of Splitting of Polynitro Arenes by Heat and by Shock
Fig. 31 LC-UV chromatograms of thermally exposed PYX at 367 ◦ C (A) and of shock-
exposed PYX (B) [183]; here the following intermediates are: PA 2,4,6-trinitrophenol,
PAM 2,4,6-trinitroaniline, DPA N-(2,4,6-trinitrophenyl)-2,4,6-trinitroaniline and DNBz
4,6-dinitrobenzofurazane
Scheme 9 Mutual relationship of the main products of the incomplete initiation of PAM,
DAP, PYX, and TPM by heat and shock; here Pi- is 2,4,6-trinitrophenyl [183]
Sensitivities of High Energy Compounds 259
4.11
Survey of Results from Application of Physical Organic Chemistry
(Outputs of POC Model)
5
Comments
It has already been mentioned that the study of energetic systems by theor-
etical methods has accelerated dramatically over the past 16 years. This study
has provided a considerable insight into the understanding of some factors af-
fecting their behaviour, but, on the other hand, it also has brought a great deal
of opacity and guesswork. The main problem is that quantum chemical calcu-
Sensitivities of High Energy Compounds 261
lations and simulations start from certain opinions held by their authors, and
the results correspond to this fact.
The discrepancy between some starting ideas of pro quantum chemical
simulation of the thermal decomposition and practical findings from this
area of chemistry of polynitro compounds is pointed out in Sect. 2.1.1 for con-
crete cases. It is illogical, for example, to consider the dissociation energy of
C – NO2 to be a stability criterion, if experimental evidence proves the par-
ticipation of an oxygen atom of the nitro group in the primary fragmentation
process (Schemes 2, 7–9 but also [11] and [166]). What has been said is also
relevant to the trends that try to describe molecular-structure dependences
of sensitivity connected with various mechanisms of primary fragmentation
by means of a universal relationship, as mentioned in Sects. 3.2.2 and also
3.3.4. The large variety of intramolecular and thus connected intermolecular
interactions, especially in polynitro arenes and azaarenes, makes these de-
pendences relatively varied (in particular, see Sect. 4.5). Therefore, it should
be stated that, for example, the well-known correlations by Politzer and Mur-
ray (Sect. 3.2.2 and [8] and [81]) must be restricted to specific classes (i.e.
nitroaromatics, nitroheterocycles and nitramines). This is also the reason
why Price and Hare [85] encounter problems when trying to describe by a sin-
gle relationship the sensitivities of a larger set of structurally heterogeneous
individual EMs.
Whereas in the area of simulation of high-temperature thermal decompos-
ition of RDX and HMX there are no distinct differences between individual
authors’ results (Sect. 2.2), in the area of simulation of the hydrostatic com-
pression effect again some of the outputs disagree with practical findings.
Thus, for example, the author of paper [60], dealing with uniform and uni-
axial compression of the nitromethane crystal, calls attention to the distinct
difference between his results and those by Kuklja and Kunz [59]. It is ax-
iomatic because dynamic pressure is under consideration in the paper [60].
However, the latter authors using simulation of an effect of a hydrostatic
compression on the electronic structure of RDX, obtained very good agree-
ment with both experimental results from the action of static pressure on this
nitramine [61] (Sect. 2.5) and the findings about primary fragmentation of
this molecule (homolysis of the N – NO2 bond [11, 14]). The heterogeneity of
views of the fragmentation mechanism under extreme conditions is well doc-
umented by the example of initiation of nitromethane (Sect. 3.3.9), which is
a favoured model nitro compound in this field.
Studies of the sensitivity of EMs are mostly directed to impact and
shock sensitivities (Sect. 3). A primary fragmentation of the given EM un-
til adiabatic compression of a thin layer of its molecules by impact or shock
wave is the unambiguously dominant view in this case. Physical Kinetics
Model [97] was a typical representative of this view of mechanical split-
ting of the molecule at any deliberate bonds between its atoms by shock
wave. One of the best developed models is Dlott’s Model of the Multiphonon
262 S. Zeman
Up-Pumping [1, 74]. As already mentioned in Sect. 3.1, the author consulted
a number of specialists in advance about his ideas. However, in this way
he introduced into his model the ideas of primary fragmentation of EMs
under extreme conditions. Nevertheless, he enriched the theory of initiation
of explosive transformations of EMs by introducing significant ideas about
transfer of initiation impulse to molecules in the crystal lattice of these ma-
terials. Experimental verification of this model is absent to date. In Dlott’s
opinion [1, 74] the initiation by impact is a particular case of initiation by
mechanical impulses (Low Velocity Initiation). Also [105] (Sect. 3.3.4) clearly
documents the difference between the initiations by impact and by shock,
though the authors do not call attention to this fact. Both the cited facts agree
with study results of initiation reactivity of polynitro compounds by means of
physical organic chemistry (Sects. 4.7 and 4.11), where the initiation by im-
pact is treated separately from the set of initiations by shock, electric spark
and heat.
Important to the study of sensitivity of individual EMs are the findings
obtained by Politzer and Murray [77–82]. Their surface electrostatic poten-
tial sharply differentiates between organic compounds of explosive nature
and the non-explosive ones. Politzer and Murray’s approach [81] delimi-
tates the part of the EM molecule in which primary fragmentation can take
place by action of impulse. A similar result was also obtained by Rice and
Hare [85], who improved the method devised by Politzer and Murray. Both
of these approaches give results that agree relatively well with the outputs
from the POC model; however, the last mentioned method provides much
more detailed information. It is a disadvantage that the Politzer and Murray
and Rice and Hare approaches do not consider experimental findings from
kinetic studies and the mechanism of low-temperature thermolysis of indi-
vidual EMs [9, 11, 14, 151].
Interesting is the Multidimensional Reactive Flow model developed by
Tarver et al. [5, 103, 104]; it is based on the Non-Equilibrium Zeldovich–
von Neuman–Döring theory. This model starts from the primary chemical
changes occurring in the adiabatically compressed thin layer of molecules of
the given EM and multiphonon up-pumping due to shock, but in the math-
ematical description it works with experimental data of thermal explosion
of EM [5, 103, 104]; it considers the induction period of initiation of detona-
tion. However, the induction period of the EM decomposition in front of the
detonation wave makes the front kinetically unstable and pulsating [101].
The results of the Excitonic model by Kuklja and Kunz [15, 59, 65–67]
are in concordance with experimental findings obtained by low-temperature
decomposition of polynitro compounds. The outputs from this model also
correspond with results from the application of physical organic chemistry
to the study of initiation reactivity of EMs (POC model). Both these models
are also supported by the results obtained by Raikova and Likholatov [118],
who found that the Arrhenius parameters in both the detonation transform-
Sensitivities of High Energy Compounds 263
6
Conclusions
Acknowledgements The author is indebted to Prof. Zdenìk Friedl from the Chemical Fac-
ulty of the Brno University of Technology for his participation in solving some partial
problems of the initiation reactivity of EMs, to Assoc. Prof. Pavel Vávra from the Institute
of Energetic Materials and to Assoc. Prof. Josef Panchártek from the Department of Or-
ganic Chemistry, both from the University of Pardubice, for their valuable remarks on this
paper. This material is based upon work supported partly by the Ministry of Industry and
Trade of the Czech Republic as part of its Research project STRATECH No. FC-M2/05/00
and partly by the Ministry of Education, Youth and Sports of the Czech Republic as part
of its research project No. MSM 0021627501.
266 S. Zeman
References
1. Dlott DD (2005) Multi-phonon up-pumping in energetic materials. In: Shaw RW,
Brill TB, Thompson DL (eds) Overviews of recent research on energetic materials.
Adv Ser Phys Chem, vol 16. World Sci Publ Co Ptc Ltd., Singapore, p 303
2. Politzer P, Murray J (eds) (2003) Theor Comp Chem, vol 13. Energetic materials, Part
2, Detonation, combustion. Elsevier, Amsterdam
3. Fried LE, Manaa MR (2005) Modeling the reactions of energetic materials in the
condensed phase. In: Shaw RW, Brill TB, Thompson DL (eds) Overviews of recent
research on energetic materials. Adv Ser Phys Chem, vol 16. World Sci Publ Co Ptc
Ltd., Singapore, p 275
4. Rice BM (2005) Application of theoretical chemistry in assesing energetic mate-
rials for performance and sensitivity. In: Shaw RW, Brill TB, Thompson DL (eds)
Overviews of recent research on energetic materials. Adv Ser Phys Chem, vol 16.
World Sci Publ Co Ptc Ltd., Singapore, p 335
5. Tarver CM, Manaa RM (2004) Chemistry of detonation waves in condensed phase
explosives. In: Manaa RM (ed) Chemistry at extreme conditions. Elsevier, Amster-
dam, p 495
6. Zeman S (2003) A study of chemical micro-mechanisms of initiation of organic
polynitro compounds. In: Politzer P, Murray J (eds) Theor Comp Chemistry, vol 13,
Energetic materials, Part 2, Detonation, combustion. Elsevier, Amsterdam, p 25
7. Zeman S (2006) J Hazard Mater A132:155
8. Zeman S, Friedl Z, Kočí J, Pelikán V, Majzlík J (2006) Cent Eur J Energ Mater 3(3):27
9. Zeman S (2002) Thermochim Acta 384:137
10. Zeman S (1999) Thermochim Acta 333:121
11. Manelis GB, Nazin GM, Rubtsov YI, Strunin VA (1996) Thermal decomposition and
combustion of explosives and powders. Izdat Nauka, Moscow
12. Zeman S, Huczala R, Friedl Z (2002) J Energet Mater 20:53
13. Maksimov YY, Kogut EN (1979) Tr Mosk, Khim-Tekhnol Inst Mendeleeva 104:30
14. Shu Y, Korssounskii BL, Nazin GM (2004) Russ Chem Rev 73(3):293
15. Kuklja MM (2001) J Phys Chem B105:10159
16. Zeman S (2000) Propellants Explos Pyrotech 25:66
17. Chakraborty D, Muller RP, Dasgupta S, Goddard WA III (2000) J Phys Chem
A104:2261
18. Zhang S, Nguyen HN, Truong TN (2003) J Phys Chem A107:2981
19. Lur’e BA, Svetlov BS (1967) Tr Mosk, Khim-Tekhnol Inst Mendeleeva 53:40
20. Nazin GM, Prokudin VG, Manelis GB (2000) Russ Chem Bull 49(2):234
21. Park J, Chakraborty D, Jamindra S, Xia WS, Lin MC, Bedford C (2002) Thermochim
Acta 384:101
22. Sorescu DC, Alavi S, Thompson DL (2005) Theoretical and computational studies
of energetic salts. In: Manaa MR (ed) Chemistry at extreme conditions. Elsevier,
Amsterdam, p 431
23. Oxley JC, Smith JL, Chen H (2002) Thermochim Acta 384:91
24. Loebbecke S, Schuppler H, Schweikert W (2003) J Thermal Anal Calorim 72:453
25. Gong XD, Xiao HM, Dong HS (1998) Chin J Chem 16(4):311
26. Gong XD, Xiao HM, Dong HS (1999) Chem Res Chin Univ 15(2):152
27. Gu ZM, Fan JF, Xiao HM, Dong HS (2000) Chem Res Chin Univ 16:21
28. Xiao H, Fan J, Gu Z, Dong H (1998) Chem Phys 226:15
29. Fan J, Gu Z, Xiao H, Dong H (1998) J Phys Org Chem 11:177
Sensitivities of High Energy Compounds 267
53. Nagayama K, Inou K, Nakahara M (2001) In: Chiba A, Tanimura S, Hokamoto K (eds)
Impact Engineering and Application. Elsevier, Amsterdam, p 515
54. Kubota S, Nagayama K, Shimada H, Matsui K (2001) In: Chiba A, Tanimura S,
Hokamoto K (eds) Impact Engineering and Application. Elsevier, New York, p 521
55. Capellos C (1998) In: Proc 11th Detonation Symp, Snowmass Village, Colorado, p 3
56. Yu H, Hambir SA, Dlott DD (2006) Ultrafast dynamics of nanotechnology energetic
materials. In: Thadhani NN, Armstrong RW, Gash AE, Wilson WH (eds) Multifunc-
tional energetic materials. Mat Res Soc Symp Proc 896:71
57. Skinner D, Olson D, Block-Bolten A (1997) Propellants Explos Pyrotech 23:34
58. Sorescu DC, Rice BM, Thompson DL (1999) J Phys Chem B103:6783
59. Kuklja MM, Kunz AB (2000) J Appl Phys 87(1):2215
60. Manaa MR (2003) Initiation and decomposition mechanism of energetic materi-
als. In: Politzer P, Murray J (eds) Theoretical and computational chemistry, Vol. 13,
Energetic materials, Part 2, Detonation, combustion. Elsevier, Amsterdam, p 71
61. Goto N, Yamawaki H, Tonokura K, Wakabayashi K, Yoshida M, Koshi M (2004) Mat
Sci Forum, Vols. 465–466. Trans Tech Publ, Switzerland, p 189
62. Rice BM, Hare J, Pai SV, Mattson W, Krasko G, Trevino SF, Sorescu DC, Thomp-
son DL (2001) Khim Fiz 20(5):9
63. Armstrong RW, Elban WL (2004) Dislocation in energetic materials. In: Nabar-
ro FRN, Hirth JP (eds) Dislocation in solids. Elsevier, Amsterdam, p 404
64. Armstrong RW, Elban WL (2006) Mater Sci Technol 22:381
65. Kunz AB, Kuklja MM, Botcher TR, Russell TP (2002) Thermochim Acta 384:279
66. Kuklja MM, Kunz B (2001) J Appl Phys 89:4962
67. Kuklja MM, Rashkeev SN, Zerilli FJ (2006) Appl Phys Lett 89:071904
68. Coffey CS (2003) Initiation due to plastic deformation from shock or impact. In:
Politzer P, Murray J (eds) Theoretical and computational chemistry, vol 13, Energetic
materials, Part 2, Detonation, combustion. Elsevier, Amsterdam, p 101
69. Coffey CS (1998) Khim Fiz 17(1):4
70. Kuklja MM, Kunz AB (2000) J Phys Chem 61:35
71. Kuklja MM, Stefanovich EV, Kunz AB (2000) J Chem Phys 112(7):3417
72. Kuklja MM, Kunz AB (1999) J Phys Chem B103:8424
73. Holmes W, Francis RS, Fayer MD (1999) J Chem Phys 110(7):3576
74. Dlott DD (2003) Fast molecular processes in energetic materials. In: Politzer P, Mur-
ray J (eds) Theoretical and computational chemistry, vol 13, Energetic materials,
Part 2, Detonation, combustion. Elsevier, Amsterdam, p 125
75. McNesby KL, Coffey CS (1997) J Phys Chem B101:3097
76. Ye S, Tonokura K, Koshi M (2003) Combust Flame 132:240
77. Alper HE, Abu-Awwad F, Politzer P (1999) J Phys Chem B103:9738
78. Politzer P, Murray JS, Seminario JM, Lane P, Grice ME, Concha MC (2001) J Mol
Struct (Theochem) 573:1
79. Politzer P, Boyd S (2002) Struct Chem 13(2):105
80. Murray JS, Lane P, Politzer P (1998) Mol Phys 93(2):187
81. Politzer P, Murray JS (2003) Sensitivity correlations. In: Politzer P, Murray J (eds)
Theoretical and computational chemistry, vol 13, Energetic materials, Part 2, Det-
onation, combustion. Elsevier, Amsterdam, p 5
82. Politzer P, Lane P, Concha MC (2005) Computtional determination of the energet-
ics of boron and aluminium combustion reaction. In: Manaa RM (ed) Chemistry at
extreme conditions. Elsevier, Amsterdam, p 473
83. Rice BM, Sahu S, Owens FJ (2002) J Mol Struct (Theochem) 583:69
Sensitivities of High Energy Compounds 269
84. Fried LE, Manaa MR, Pagoria PF, Simpson RL (2001) Annu Rev Mater Res 31:291
85. Rice BM, Hare JJ (2002) J Phys Chem A106:1770
86. Edwars J, Eybl C, Johnson B (2004) Int J Quant Chem 100:713
87. Chen ZX, Xiao H, Yang S (1999) Chem Phys 250:243
88. Türker L (2005) J Mol Struct (Theochem) 725:85
89. Xiao MM, Fan JF, Gong XD (1997) Propellants Explos Pyrotech 22:360
90. Zhang C, Shu Y, Huang Y, Zhao X, Dong H (2005) J Phys Chem B109:8978
91. Zhang C, Shu Y, Wang X (2005) Energ Mater 23:107
92. Xiao HM, Li JS, Dong HS (2000) Acta Chim Sin 58(3):297
93. Loginov NP, Surkova SN (2006) Fiz Goreniya Vzryva 42(1):100
94. Vaullerin M, Espagnacq A (1998) Propellants Explos Pyrotech 23:237
95. Cho SG, No KT, Goh EM, Kim JK, Shin JH, Joo YD, Seong S (2005) Bull Korean Chem
Soc 26(3):399
96. Keshavarsz MH, Jaafari M (2006) Propellants Explos Pyrotech 31:216
97. Wolker FE (1998) Khim Fiz 17(1):25
98. Zeman S (1999) J Energ Mater 17:305
99. Klimenko VY, Yakoventsev MA, Dremin AN (1993) Khim Fizika 12:671
100. Klimenko VY (1998) Khim Fizika 17:11
101. Dremin AN (1999) Toward detonation theory. Springer, Berlin Heidelberg New York
102. Hong X, Hill JR, Dlot DD (1995) Vibrational Energy Transfer in High Explosives: Ni-
tromethane. In: Brill TB, Russell TP, Tao VC, Wardl RB (eds) Mat Res Soc Symp Proc
418:357
103. Tarver CM (1997) J Phys Chem A101:4845
104. Tarver CM, Utriew PA, Forbes JW (2001) Khim Fiz 20(3):38
105. Koshi M, Ye S, Widijaja J, Tonokura K (2001) Estimation of shock sensitivity based
on molecular properties. In: Chiba A, Tanimura S, Hokamoto K (eds) Impact engin-
eering and application. Elsevier, Amsterdam, p 175
106. Kuklja MM (2001) Electronic excitations in initiation of chemistry in molecular
solids. In: Bulatov V, Cleri F, Colombo L, Lewi L, Mousseau N (eds) Advances in ma-
terials theory and modeling, bridging over ultiple-length and time scales. Mat Res
Soc Symp Proc 677:AA2.4.1
107. Mathieu D, Martin P, La Hargue J-P (2005) Phys Scripta T118:171
108. Méreau R, Mathieu D, Elstner M, Frauenheim T (2004) Phys Rev B69:104101
109. White CT, Swanson DR, Robertson DH (2001) Molecular dynamics simulations of
detonations. In: Dressler RA (ed) Chemical dynamics in extreme environments.
World Sci., Singapore, p 547
110. van der Hejden AEDM, Bouma RHB, van der Steen AC (2004) Propellants Explos
Pyrotech 29:304
111. Jindal VK, Dlott DD (1998) J Appl Phys 83(5):5203
112. Yoo CS, Holmes NC, Souers PC, Wu CJ, Ree FH, Dick JJ (2000) J Appl Phys 88(01):70
113. Rice BM, Mattson W, Trevino SF (1998) Phys Rev E57(1):5106
114. Engelke R, Blais NC, Sheffield SA, Sander RK (2001) J Phys Chem A105:6955
115. Andreev SG (1998) Fiz Khim 17(1):55
116. Wu CJ, Ree FH, Yoo CS (2004) Propellants Explos Pyrotech 29:296
117. Dreger ZA, Gruzdkov YA, Gupta YM (2002) J Phys Chem B106:247
118. Raikova VM, Likholatov EA (2005) Propellants Explos Pyrotech 30:250
119. Hu WF, He TJ, Chen DM, Liu FC (2002) J Phys Chem A106(32):7294
120. Nguen MT, Le HT, Hajgato B, Veszpremi T, Lin MC (2003) J Phys Chem A107:4286
121. Arenas JF, Otero JC, Pealez D, Soto J (2003) J Chem Phys 119(15):7814
270 S. Zeman
122. Reed EJ, Joannopoulos JD, Laurence E (2000) Phys Rev B: Condens Matter Mater
Phys 62:16500
123. Margetis D, Kaxiras E, Elstner M, Frauenheim T, Manaa MR (2002) J Chem Phys
117:788
124. Zhang XX, Bauer SH (1997) J Phys Chem B101:8717
125. Glaborg P, Bendsen AB, Miller JA (1999) Int J Chem Kinet 31:591
126. Winey JM, Gupta YM (1997) J Phys Chem B101:10733
127. Blais NC, Engelke R, Sheffield SA (1997) J Phys Chem A101:8285
128. Bouyer V, Darbord I, Herve P, Baudin G, Le Gallic C, Clément F, Chavent G (2006)
Combust Flame 144:139
129. Leiber CO (2003) Assesment of safety and risk with a microscopic model of detona-
tion. Elsevier, Amsterdam
130. Bharatam PV, Lammertsma K (2003) Nitro-aci-nitro tautomerizm in high-energetic
compounds. In: Politzer P, Murray J (eds) Theoretical and computational chemistry,
vol 13, Energetic materials, Part 1, Detonation, combustion. Elsevier, Amsterdam,
p 61
131. Ferm EN (1999) Propellants Explos Pyrotech 24:1
132. Gruzdkov YA, Gupta YM (1998) J Phys Chem A102:2322
133. Woods E III, Dessiaterik Y, Miller RE, Baer T (2001) J Phys Chem A105:8273
134. Engelke R, Schiferel D, Storm CB, Earl WI (1988) J Phys Chem 92:6815
135. Nabatov SS, Zakushev VV, Dremin AN (1975) Fiz Goreniya Vryva 11:300
136. Engelke R, Sheffield SA, Stacy HL, Quintana JP (2005) Phys Fluids 17:096102
137. Cudzilo S, Trzcinski WA (1999) Pol J Tech Phys 40(2):223
138. Andreev SG, Babkin AV, Baum FN, Imkhovik NA, Kobylkin IF, Kolpakov VI, La-
dov SV, Odintsov VA, Orlenko LP, Okhitin VN, Selivanov VV, Solovev VS, Stanyuko-
vich KP, Chelyshev VP, Shekhte BI (2002) Fizika vzryva, Tom 1 (Physics of explosion,
vol 1). Fizmatlit, Moscow
139. Licht H-H (2000) Propellants Explos Pyrotech 25:126
140. Yang S, Yue S (2003) HanNeg CaiLiao 11(4):231
141. Agrawal JP (1998) Progr Energy Combust Sci 24:1
142. Pagoria PF, Lee GS, Mitchell AR, Schmidt RD (2002) Thermochim Acta 384:187
143. Chapman RD, Wilson WS, Fronabarger JW, Merwin LH, Ostrom GS (2002) Ther-
mochim Acta 384:229
144. Talawar MB, Sivabalan R, Asthana SN, Singh H (2005) Fiz Goreniya Vzryva 41(3):29
145. Zeman S (1980) Thermochim Acta 41:199
146. Zeman S (1998) In: Proc 29th Int Annual Conf on ICT, Karlsruhe, p 141/1
147. Zeman S, Krupka M (2003) Propellants Explos Pyrotech 28:301
148. Zeman S, Krupka M (2003) Propellants Explos Pyrotech 28:249
149. Clawson JS, Strohmeier M, Stueber D, Orend AM, Baric DH, Asay BA, Hiskey MA,
Pugmire RJ, Grant DM (2002) J Phys Chem A106:6352
150. Zeman S, Pešková M (2005) Cent Eur J Energ Mater 2(3):71
151. Zeman S (1997) Thermochim Acta 290:199
152. Zeman S, Friedl Z (2004) J Thermal Anal Calorim 77:217
153. Zeman S, Kohlíček P, Maranda M (2003) Thermochim Acta 398:185
154. Kissinger HE (1957) Anal Chem 29:1702
155. Zeman S (2003) Propellants Explos Pyrotech 28:308
156. Kamlet MJ, Jacobs SJ (1968) J Chem Phys 48:23
157. Pepekin VI, Makhov NM, Lebedev YA (1977) Dokl Akad Nauk SSSR 230:852
158. TITAN (2001) v.1.0.8. Wavefunction. Schrödinger, USA
Sensitivities of High Energy Compounds 271
Dalal NS, Gunaydin-Sen O, Bussmann-Holder A (2007) Experimental Evidence for the Coex-
istence of Order/Disorder and Displacive Behavior of Hydrogen-Bonded Ferroelectrics
and Antiferroelectrics. 124: 23–50
Dalal NS, see Bussmann-Holder A (2007) 124: 1–21
Daul CA, see Atanasov M (2003) 106: 97–125
Day P (2003) Whereof Man Cannot Speak: Some Scientific Vocabulary of Michael Faraday
and Klixbüll Jørgensen 106: 7–18
Deeth RJ (2004) Computational Bioinorganic Chemistry 113: 37–69
Delahaye S, see Hauser A (2003) 106: 81–96
Deng S, Simon A, Köhler J (2005) Pairing Mechanisms Viewed from Physics and Chemistry
114: 103–141
Di Bartolo B, see Bowlby BE (2003) 106: 191–208
Di Bilio AJ, see Contreras RR (2003) 106: 71–79
Ding Z, see Bard AJ (2005) 118: 1–57
Dovesi R, see Corà F (2004) 113: 171–232
Duan X, see He J (2005) 119: 89–119
Duan X, see Li F (2005) 119: 193–223
Flamigni L, Heitz V, Sauvage J-P (2006) Porphyrin Rotaxanes and Catenanes: Copper(I)-
Templated Synthesis and Photoinduced Processes 121: 217–261
276 Author Index Volumes 101–125
Lein M, Frunzke J, Frenking G (2003) Christian Klixbüll Jørgensen and the Nature of the
Chemical Bond in HArF 106: 181–191
Leroux F, see Taviot-Gueho C (2005) 119: 121–159
Lever ABP, Gorelesky SI (2004) Ruthenium Complexes of Non-Innocent Ligands; Aspects of
Charge Transfer Spectroscopy 107: 77–114
Li B, see He J (2005) 119: 89–119
Li F, Duan X (2005) Applications of Layered Double Hydroxides 119: 193–223
Liebau F, see Santamaría-Pérez D (2005) 118: 79–135
Linton DJ, Wheatley AEH (2003) The Synthesis and Structural Properties of Aluminium
Oxide, Hydroxide and Organooxide Compounds 105: 67–139
Lo KK-W (2007) Luminescent Transition Metal Complexes as Biological Labels and Probes.
123: 205–245
Lux B, see Haubner R (2002) 102: 1–46
Mossin S, Weihe H (2003) Average One-Center Two-Electron Exchange Integrals and Ex-
change Interactions 106: 173–180
Murakami S, see Morita M (2004) 107: 115–144
Müller E, see Roewer G (2002) 101: 59–136
Müller KA (2005) Essential Heterogeneities in Hole-Doped Cuprate Superconductors 114:
1–11
Müller KA, see Bussmann-Holder A (2005) 114: 367–386
Myung N, see Bard AJ (2005) 118: 1–57
Nazeeruddin MK, Grätzel M (2007) Transition Metal Complexes for Photovoltaic and Light
Emitting Applications. 123: 113–175
Nguyen YHL, see Contakes SM (2007) 123: 177–203
Nishibori E, see Takata M (2004) 109: 59–84
Nishihara H, see Kume S (2007) 123: 79–112
Nolet MC, Beaulac R, Boulanger AM, Reber C (2004) Allowed and Forbidden d-d Bands in
Octa-hedral Coordination Compounds: Intensity Borrowing and Interference Dips in
Absorption Spectra 107: 145–158
Reinen D, Atanasov M (2004) The Angular Overlap Model and Vibronic Coupling in Treating
s-p and d-s Mixing – a DFT Study 107: 159–178
Reisfeld R (2003) Rare Earth Ions: Their Spectroscopy of Cryptates and Related Complexes
in Glasses 106: 209–237
Renz F, see Gütlich P (2004) 107: 27–76
Reyes M, see Contreras RR (2003) 106: 71–79
Ricciardi G, see Rosa A (2004) 112: 49–116
Rice BM, Byrd EFC, Mattson WD (2007) Computational Aspects of Nitrogen-Rich HEDMs.
125: 153–194
Riesen H (2004) Progress in Hole-Burning Spectroscopy of Coordination Compounds 107:
179–205
Robaszkiewicz S, see Micnas R (2005) 114: 13–69
Roewer G, Herzog U, Trommer K, Müller E, Frühauf S (2002) Silicon Carbide – A Survey of
Synthetic Approaches, Properties and Applications 101: 59–136
Rosa A, Ricciardi G, Gritsenko O, Baerends EJ (2004) Excitation Energies of Metal Complexes
with Time-dependent Density Functional Theory 112: 49–116
Rudolf P, see Golden MS (2004) 109: 201–229
Ruiz E (2004) Theoretical Study of the Exchange Coupling in Large Polynuclear Transition
Metal Complexes Using DFT Methods 113: 71–102
Williams GR, Khan AI, O’Hare D (2005) Mechanistic and Kinetic Studies of Guest Ion
Intercalation into Layered Double Hydroxides Using Time-resolved, In-situ X-ray Powder
Diffraction 119: 161–192
de Wolff FA, see Berend K (2003) 104: 1–58
Woodley SM (2004) Prediction of Crystal Structures Using Evolutionary Algorithms and
Related Techniques 110: 95–132
Xantheas SS (2005) Interaction Potentials for Water from Accurate Cluster Calculations 116:
119–148