Crystal Engineering of Energetic Materials: Co-crystals of Ethylenedinitramine (EDNA) with Modified Performance and Improved Chemical Stability

Article abstract of DOI:10.1002/chem.201501721

In the area of energetic materials, co-crystallization is emerging as a new technology for modifying or enhancing the properties of existing energetic substances. Ethylenedinitramine (EDNA) is a known energetic material which requires attention partly due to its chemical instability originating with its two highly acidic protons. In order to stabilize EDNA, a co-crystallization approach targeting the acidic protons using a series of co-crystallizing agents with suitable hydrogen-bond acceptors was employed. Fifteen attempted co-crystallizations resulted in eight successful outcomes and six of these were crystallographically characterized and all showed evidence of hydrogen bonds to the intended protons. Calculated detonation properties and experimental thermal and impact data for the co-crystals were obtained and compared with those of pure EDNA. The co-crystal of EDNA and 1,2-bis(4-pyridyl)ethylene was recognized as a more thermally stable alternative to EDNA while the co-crystal of EDNA and pyrazine N,N′-dioxide showed comparable detonation strengths (and much improved chemical stability) compared with that of EDNA. The co-crystals EDNA:4,4′-bipyridine and EDNA:pyrazine N,N′-dioxide were found to be about 50 % less impact sensitive than EDNA, all of which illustrate how co-crystallizations can be utilized for successfully modifying specific aspects of energetic materials. The taming of EDNA: Ethylenedinitramine (EDNA) is a known energetic material the chemical instability of which originates with its two highly acidic protons. Co-crystallizations of EDNA demonstrate that the acidic protons can be targeted with suitable hydrogen-bond acceptors, which improves chemical stability and offers an avenue for modulating thermal properties, impact sensitivity, and overall performance of the material (see picture).

Full text of DOI:10.1002/chem.201501721

DOI: 10.1002/chem.201501721
Full Paper
&
Energetic Materials |Hot Paper|
Crystal Engineering of Energetic Materials: Co-crystals of
Ethylenedinitramine (EDNA) with Modified Performance and
Improved Chemical Stability
Christer B. Aakerçy,* Tharanga K. Wijethunga, and John Desper[a]
Abstract: In the area of energetic materials, co-crystallization
is emerging as a new technology for modifying or enhanc-
ing the properties of existing energetic substances. Ethylene-
dinitramine (EDNA) is a known energetic material which re-
quires attention partly due to its chemical instability origi-
nating with its two highly acidic protons. In order to stabilize
EDNA, a co-crystallization approach targeting the acidic pro-
tons using a series of co-crystallizing agents with suitable
hydrogen-bond acceptors was employed. Fifteen attempted
co-crystallizations resulted in eight successful outcomes and
six of these were crystallographically characterized and all
showed evidence of hydrogen bonds to the intended pro-
tons. Calculated detonation properties and experimental
thermal and impact data for the co-crystals were obtained
and compared with those of pure EDNA. The co-crystal of
EDNA and 1,2-bis(4-pyridyl)ethylene was recognized as
a more thermally stable alternative to EDNA while the co-
crystal of EDNA and pyrazine N,N’-dioxide showed compara-
ble detonation strengths (and much improved chemical sta-
bility) compared with that of EDNA. The co-crystals
EDNA:4,4’-bipyridine and EDNA:pyrazine N,N’-dioxide were
found to be about 50% less impact sensitive than EDNA, all
of which illustrate how co-crystallizations can be utilized for
successfully modifying specific aspects of energetic materi-
als.
ficult.[14] An alternative avenue may be found through the use
of co-crystal technology[15] wherein a known energetic material
is combined with either an energetic or a non-energetic com-
pound[16] via noncovalent interactions within a crystalline
framework.[17] Co-crystallizations can provide means for altering
a range of properties that are a function of the nature of the
solid state, such as density, mechanical and thermal stability,
and solubility which can all facilitate the pursuit of safer or
more versatile energetic materials.[18] More generally, co-crystal-
lizations may also offer an avenue for stabilizing liquids and
other volatile chemicals[19] thereby lowering the risk of expo-
sure through inhalation of toxic energetic materials[20] (many
nitro-containing compounds are toxic). All in all, a co-crystal
version of an energetic material can, in certain circumstances,
be more useful because of superior chemical stability and
shelf-life even though it may have slightly lower energetic per-
formance.
Introduction
Energetic materials contain substantial amounts of stored
energy that can be released rapidly,[1] and representatives of
energetic materials include explosives, propellants and pyro-
technics.[2] Even though a large number of different energetic
compounds are known in armaments, mining, fireworks and
space explorations,[3] many have drawbacks leading to restric-
tions or limitations of their use.[4] Common problems include
excessive sensitivity to impact, heat, shock or friction[5] (e.g.,
RDX[6] and CL-20[7]), as well as undesirable reactivity towards
different environmental factors (e.g., nitramine explosives[8]),
and high toxicity.[9] Highly powerful energetic materials inher-
ently display high sensitivity and/or high reactivity leading to
a power-safety contradiction[10] and, consequently, designing
powerful energetic materials with improved safety and stability
is a major challenge.[11]
The pursuit of new energetic materials typically follows one
of two approaches. First, a completely new molecule or salt
can be prepared[12] in order to uncover a material with the re-
quired properties such as high power, low sensitivity and low
reactivity expressed in an integrated manner,[13] but optimizing
all these properties within a finite time scale is enormously dif-
Ethylenedinitramine (EDNA) also known as Haleite[21]
(Figure 1), is described as the first entirely American high ex-
plosive.[22] EDNA combines the properties of a high explosive
like TNT and an initiating agent like mercury fulminate, pos-
sessing a high brisance, comparatively low impact sensitivity
and high heat sensitivity. This compound is applied as a secon-
dary explosive, booster explosive and in the preparation of
cast explosives such as ednatol.[23]
[a] Prof. C. B. Aakerçy, T. K. Wijethunga, Dr. J. Desper
Department of Chemistry, Kansas State University
Manhattan, KS, 66506 (USA)
An important reason behind some difficulties in the use of
EDNA is its relatively high chemical reactivity, a common draw-
back associated with many nitramine-based explosives.[24]
Overly acidic compounds are often prone to chemical instabili-
E-mail: aakeroy@ksu.edu
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501721.
Chem. Eur. J. 2015, 21, 11029 – 11037
11029
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. Chemical structure of EDNA.
ty and hygroscopicity,[25] which can create problems with trans-
port, processing, and storage. For example, tetranitropyrole is
unstable at ambient temperatures[26] and dinitrotriazole is deli-
quescent,[27] and both contain an acidic NÀH moiety. The
highly acidic picric acid, another well-known explosive, is cor-
rosive and reacts readily with metals to form shock-sensitive
salts. EDNA is also highly
Figure 2. Part of the crystal structure of EDNA with hydrogen bonds be-
tween nitro oxygen atoms and acidic amine protons.[33]
acidic[28] (pKa1 5.31 and pKa2
6.64),[29] due to the ease of de-
localization, facilitated by the
nitro groups, of the remaining
negative charge upon deproto-
nation; thus EDNA is corrosive
and can react with metals and
metal salts[30] to produce new
materials with unpredictable
properties.
For
example,
CuIIEDNA, is easily synthesized
and has low impact sensitivity[31]
whereas PbIIEDNA displays supe-
rior
properties/performance
compared to EDNA itself.[32]
Scheme 1. Hydrogen-bond acceptors used in this study.
In this study, we describe the
use of co-crystallizations for al-
tering the physical properties
and stability of EDNA while retaining some valuable character-
istics associated with an energetic material. The goals of this
study are to establish: 1) if EDNA is a suitable target for co-
crystal synthesis; 2) if the acidic protons can be engaged (and
essentially protected) by hydrogen bonding to suitable co-for-
mers; 3) if any explosive property of the resulting materials can
be modulated in a useful manner; and 4) if the chemical stabil-
ity of EDNA can be improved through co-crystallizations.
A close look at the crystal structure of EDNA[33] (Figure 2) re-
veals that these acidic protons are interacting with nitro
oxygen atoms which are considered to be poor hydrogen-
bond acceptors.
is likely to produce infinite supramolecular chains. Second, the
N-oxides are likely to form the same hydrogen-bonded motifs
as the parent N-heterocycles but with a better oxygen balance
which would be beneficial from a performance perspective.
Results
Initial co-crystal screening was carried out by combining EDNA
and the acceptors (Scheme 1) in stoichiometric ratios and the
products obtained were analyzed with both IR spectroscopy
and DSC. The solvent of choice was methanol as both EDNA
and the acceptors showed comparable solubility in alcohols.
Eight of fifteen attempts yielded co-crystals as indicated by
both IR and DSC (see the Supporting Information) data. Al-
though vibrational spectroscopy provides information about
whether a co-crystal has formed or not, it does not reveal to
what extent the amine protons on EDNA participate in hydro-
gen bonding. To address this issue, crystallographic data are
needed and six of the eight new co-crystals were characterized
using single-crystal X-ray diffraction.
With this in mind we selected fifteen hydrogen-bond accept-
ors that potentially would compete successfully with the nitro
groups for the attention of the acidic protons (Scheme 1), in-
cluding N-based acceptors (A1–A7, A11 and A12) O-based ac-
ceptors (A8–A10) and mixed N- and O-based acceptors (A13–
A15). These 15 acceptors can be further categorized as sym-
metric ditopic (A1–A10), monotopic (A11 and A12) and dis-
symmetric ditopic (A13 to A15). The reason for selecting pri-
marily ditopic acceptors (all but A11 and A12) was first of all
that we wanted to aim for some degree of structural consis-
tency; a combination of a ditopic donor and a ditopic acceptor
Crystallographic data for the six co-crystals (EDNA:A3,
EDNA:A4, EDNA:A5, EDNA:A7, EDNA:A8 and EDNA:A12) are
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11030
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
compared to their molecular conformations in the
crystal structure of EDNA by itself (Figure 2).
Table 1. Hydrogen bond geometries for six crystal structures.
Co-crystal
EDNA:A3
EDNA:A4
DÀH···A
DÀH [Š]
H···A [Š]
D···A [Š]
DÀH···A [8]
169.6(15)
176.9(15)
164(2)
169.2(17)
167.5(19)
175.0(13)
159.4(16)
176.4(16)
The N-C-C-N torsion angle for EDNA in EDNA:A3 is
288, while it is 1808 for pure EDNA. In order to find
the energy difference between the two orientations,
a molecular structure optimization was subsequently
performed. A plot of energy versus torsion angle
(Figure 4) shows that the energy difference between
the two conformations is about 24 kJmolÀ1, favoring
the molecular structure observed in the EDNA crystal
structure. It is reasonable to assume that the relative-
ly small energy differences can readily be compensat-
ed for by the new NÀH···N(py) hydrogen bonds that
are observed in the structure of EDNA:A3.
N(12)-H(12)···N(21)
N(15)-H(15)···N(31)#1
N121-H121···N211
N122-H122···N212
N(12)-H(12)···N(21)
N(12)-H(12)···N(21)
N(12)-H(12)···O(21)
N(12)-H(12)···N(21)
0.95(18)
1.5(18)
0.83(2)
0.97(2)
0.89(2)
0.92(15)
0.90(18)
0.92(17)
1.85(19)
1.74(18)
1.98(2)
1.82(2)
1.87(2)
1.93(15)
1.86(18)
1.89(17)
2.787(19)
2.788(19)
2.780(2)
2.770(2)
2.747(2)
2.841(12)
2.720(13)
2.818(17)
EDNA:A5
EDNA:A7
EDNA:A8
EDNA:A12
EDNA:A3 #1: Àx+1, yÀ1,Àz+1/2; EDNA:A4 #1: Àx+1,Ày+1,Àz+2; #2: Àx+
2,Ày+2,Àz+1; #3: Àx+2,Ày+1,Àz+1; #4: Àx+1,Ày,Àz+2; EDNA:A5 #1: Àx+
1,Ày+1,Àz+1; #2: Àx,Ày,Àz; EDNA:A7 #1: Àx,Ày+1,Àz; #2: Àx+2, Ày+1,Àz+1;
EDNA:A8 #1: Àx,Ày,Àz+1; #2: Àx+1, Ày+1,Àz; EDNA:A12 #1: Àx+1/2,Ày+3/
2,Àz+1.
The crystal structure of EDNA:A4 (Figure 5) reveals
the expected NÀH···N(py) hydrogen bonds and the
reported in the Supporting Information and hydrogen bond
geometries are summarized in Table 1.
two components are again present in a 1:1 ratio. In this struc-
ture the EDNA molecules adopt a 1808 torsion angle of the N-
C-C-N backbone (the molecule is located on a crystallographic
special position).
The crystal structure of EDNA:A3 (Figure 3) reveals a 1:1 sto-
ichiometry where the two components are connected via pri-
mary NÀH···N(py) hydrogen bonds. As intended, the pyridine-
based acceptor sites have provided a successful replacement
for the NÀH···O(nitro) hydrogen bonds present in the crystal
structure of EDNA itself.
The crystal structure of EDNA:A5 (Figure 6) shares the main
structural features with those found in the structure of
EDNA:A4 with a 1:1 stoichiometry, structure directing NÀ
H···N(py) hydrogen bonds resulting in infinite chains, and
a 1808 torsion angle of the N-C-C-N backbone of EDNA.
Although 4,4’-azopyridine, the co-former in EDNA:A7, does
contain two additional nitrogen atoms that, in theory, could
engage with the acidic protons of EDNA, the two p-sites are
substantially better acceptors[34] (Figure 7) and consequently,
NÀH···N(py) driven hydrogen-bonded chains dominate the
crystal structure, similar to what was found in both EDNA:A4
and EDNA:A5. The stoichiometry again is 1:1 which is dictated
by the fact that EDNA has two strong donors which are
matched by the two main N(py)-acceptor sites on A7.
Somewhat surprisingly though, the molecular geometry of
EDNA molecules in the EDNA:A3 lattice is very different when
Figure 3. Primary interactions in the EDNA:A3 co-crystal.
Figure 5. Primary hydrogen bonds in the EDNA:A4 co-crystal.
Figure 6. Part of the infinite hydrogen-bonded chain in the EDNA:A5 co-
crystal.
Figure 4. Relative energies of EDNA as a function of the N-C-C-N torsion
angle (changed from 08 to 3608 with 308 increments). The observed torsion
angle in the solid of pure EDNA is 1808 whereas it is 288 in the co-crystal
EDNA:A3.
Figure 7. Primary hydrogen bonds in the EDNA:A7 co-crystal.
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11031
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The crystal structure of EDNA:A8 (Figure 8) shows a 1:1 stoi-
chiometry, and infinite hydrogen bonds generated by NÀH···O-
N(py) chain formation between the oxygen atoms of N-oxide
and the acidic NÀH protons of EDNA. This structure illustrates
that EDNA is capable of forming co-crystals not only with N-
based acceptors but also with O-based acceptors via the acidic
protons.
Figure 10. DSC trace for EDNA:A7 co-crystal.
Figure 8. Primary interactions in the EDNA:A8 co-crystal.
Performance analysis
The detonation properties of the six co-crystals were calculated
using experimental densities obtained from the single-crystal
structure determinations. Detonation velocities and pressures
were calculated using nitrogen equivalents of different detona-
tion products,[36] and oxygen balances were calculated using
well-established methods.[37] Table 3 summarizes the detona-
tion properties for the six co-crystals. The oxygen balance of
an explosive is the mass of oxygen either in excess or in deficit
of that required for the complete oxidation of the carbon and
Figure 9. Main hydrogen bonds in the EDNA:A12 co-crystal.
The final crystal structure ob-
tained in this series was with
a monotopic hydrogen-bond ac-
Table 2. Thermal properties of co-crystals.
ceptor, A12, and consequently
the stoichiometry in this case is
1:2 (EDNA:A12; Figure 9). The
chains that resulted when ditop-
ic acceptors were combined
with EDNA is now replaced with
a discrete trimeric motif.
Compound/ Co-former melting Co-crystal melting Melting
Decomposition Energy released at
point [8C]
co-crystal
point [8C]
point [8C]
enthalpy [JgÀ1
]
decomposition [JgÀ1
]
EDNA
–
180
146
129
–
118
154
156
94
215Æ5
145Æ11
133Æ14
–
119 Æ5
160Æ28
150Æ7
198Æ2
121Æ4
186
189
191
198
194
165
172
203
168
1971Æ18
1055Æ11
985Æ3
EDNA:A3
EDNA:A4
EDNA:A5
EDNA:A6
EDNA:A7
EDNA:A8
EDNA:A11
EDNA:A12
112–114
110–112
148–152
53–56
106–110
285–289
69–73
955Æ23
769Æ31
1242Æ15
1447Æ14
1214Æ37
982Æ18
96–99
133
Thermal analysis
Melting enthalpies and decomposition enthalpies are reported as an average from two trials for each com-
pound with the calculated standard deviation.
The thermal analysis of the eight
co-crystals was mainly focused
on three aspects; melting tem-
perature, decomposition temper-
ature, and enthalpy of decompo-
sition, and the results were com-
pared with that of EDNA. As an
example, Figure 10 shows the
DSC trace for EDNA:A7.
The results are summarized in
Table 2 (see the Supporting
Information for DSC curves).
Pure EDNA has a melting point
of 1808C and decomposes at
Table 3. Detonation properties of co-crystals.
Compound/
co-crystal
Density
Detonation velocity[a]
Detonation
pressure[a] [GPa]
Oxygen
balance[a]
[gcmÀ3
]
[kmsÀ1
]
EDNA
1.65 (RT)
1.489 (120 K)
1.452 (RT)
7.890
6.178
26.72
15.24
À32
À141
EDNA:A3
1.407 (120 K)
1.355 (RT)
1.460 (120 K)
1.434 (RT)
1.529 (120 K)
1.68 (120 K)
1.64 (RT)
5.967
6.030
13.70
14.32
À158
À154
EDNA:A4
EDNA:A5
EDNA:A7
EDNA:A8
EDNA:A12
6.504
7.256
17.21
22.73
À129
À67
1868C
a
accompanied
by
1971(Æ18) JgÀ1 release of
1.432 (120 K)
5.948
13.75
À173
energy (for comparison the cor-
responding energy release for
HMX is 1987 JgÀ1[35]).
[a] All the velocity and pressure calculations for the co-crystals were carried out using low temperature densi-
ties while for EDNA the calculations were conducted using room temperature density.
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11032
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
hydrogen atoms present in the explosive.[38] A negative bal-
ance means that there is an insufficient amount of oxygen for
complete combustion. EDNA has an oxygen balance of À32%
and the co-crystals of EDNA presented here all have a more
negative oxygen balance with EDNA:A8 at À67% being the
least unfavorable (for comparison TNT has an oxygen balance
of À74%).
The impact sensitivities of EDNA, EDNA:A3, and EDNA:A8
were examined using conventional drop-weight measurements
and calculated using Bruceton analysis. The H50 values are
99(Æ1), 140(Æ1), and 138(Æ1) cm, respectively, where a higher
value indicates reduced sensitivity.
Figure 11. Co-crystal and co-former melting points.
Discussion
In all six crystal structures the targeted acidic NÀH protons did
form hydrogen bonds with the acceptors resulting in success-
ful co-crystallization. In five of the six structures, EDNA:A3
being the exception, the EDNA molecule adopted the most fa-
vorable molecular geometry with 1808 torsion angle of the N-
C-C-N backbone. Another supramolecular synthetic target in
this study was to achieve some structural consistency through
the use of ditopic hydrogen-bond acceptors. In four of the five
crystal structures of EDNA with ditopic acceptors (A3, A4, A5,
A8) the result was an infinite hydrogen-bonded chain of alter-
nating EDNA:co-former components, and the zig–zag shaped
chain found in the crystal structure of EDNA:A3 was due to
the unusual conformation of the backbone of the EDNA mole-
cules in that particular solid.
molecular structure and properties even in heteromeric solids
as long as the primary structural features are somewhat consis-
tent from one compound to the next.
Despite the lack of strong correlation between the two vari-
ables (which should not be expected), the general trend indi-
cates that the co-former exercises an influence on the solid
form of EDNA that can be estimated reasonably well.
Density is a major contributor to the explosive properties of
an energetic material.[39] Therefore, we examined whether
there was any correlation between the density of the pure co-
former and the density of the corresponding EDNA co-crystal
thereof (Figure 12 and Figure 13). The analysis was carried out
only for the co-crystals where the room temperature densities
are reported for the co-formers.
EDNA:A8 has a comparable density (1.64 gcmÀ3 vs. EDNA
1.65 gcmÀ3). As was the case with the melting points of the
EDNA co-crystals, there is, generally speaking, an increase of
density of the co-crystal as the density of the co-former goes
up (EDNA:A5 is a slight anomaly) and the big improvement is
seen in the co-crystal with an N-oxide based co-former, A8.
The use of N-oxides as co-formers serves two purposes
when it comes to designing and synthesizing energetic co-
crystals. The oxygen balance is obviously benefitting by
moving from a parent N-heterocycle to the corresponding N-
oxide, but the density can also improve.[40] In order to examine
how general this second point is, we compared densities of
Out of fifteen co-crystallization attempts eight formed co-
crystals successfully. In order to explain the reason behind the
successes and failures, we analyzed the electrostatic potentials
of each acceptor atom. By looking at the electrostatic potential
values of N-based acceptors it was evident that the higher
value of electrostatic potential on the acceptor lead to co-crys-
tal formation, where A1 and A2 being the molecules with least
potential, did not show any evidence of successful interactions.
But this explanation only seems to be valid with N-based ac-
ceptors only. When considering the O-based and mixed ac-
ceptors, it seems even the higher potential on the acceptor
atoms are not successful in forming hydrogen bonds with
EDNA.
The melting points for six of the eight co-crystals of EDNA
fall between those of the two pure components of each co-
crystal (Figure 11). EDNA:A8 has a lower melting point than
both initial compounds while EDNA:A5 has a decomposition
temperature approximately twenty degrees higher than the
melting point of EDNA, and about ten degrees above its de-
composition temperature. As mentioned previously, a drawback
with EDNA is a relatively high sensitivity to heat, which means
that the co-crystal EDNA:A5 offers an improvement in this re-
spect.
The relationship between molecular structure and melting
point of the corresponding solid is governed by a delicate bal-
ance between numerous intermolecular interactions and it is
largely unpredictable. However, a systematic co-crystallization
approach, can provide some trends and correlations between
Figure 12. Co-crystal and co-former densities at room temperature.
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11033
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 15. A plot of packing coefficient versus density in six EDNA co-crys-
tals.
Figure 13. Correlation between co-crystal and co-former densities at room
temperature.
beyond nitrogen and oxygen were present in any of the co-
formers). To verify this hypothesis we plotted the two parame-
ters against each other (Figure 15).
As shown, the packing coefficient provides an excellent indi-
cation of density of the resulting lattice.
Among the eight co-crystals EDNA:A8 gives the highest en-
thalpy change upon decomposition (as determined by DSC)
which is likely related to the influence of the co-former, A8,
which has a more favorable oxygen balance than any of the
co-formers used in this study.[42] The DSC analysis also shows
that there is a relatively large temperature difference between
the melting and decomposition temperature in EDNA:A6 and
EDNA:A11 compared to the same parameter in pure EDNA.
This means that the window of stability beyond the melting
point can be substantially improved upon co-crystallization. It
is particularly interesting to note that the molten phase of
EDNA:A11 shows a stability beyond that of the decomposition
temperature of EDNA itself which indicates that some critical
intermolecular interactions remain intact in the liquid phase.
EDNA is considered to be an energetic compound with rela-
tively low sensitivity to impact and friction and this is a proper-
ty that helps to determine if a material is suitable as a primary
or secondary explosive as well as what kind of precautions are
needed for storage and handling. Any incorporation of EDNA
within a new crystalline material is likely to impact sensitivity,
broadly speaking, as well as any property governed by the
nature of the solid state. We examined the impact sensitivity
of two co-crystals in this series with particularly favorable ther-
mal stability (EDNA:A3) and energetic properties (EDNA:A8)
and the two co-crystals were significantly less impact sensitive
than EDNA itself. The co-formers in the lattice can essentially
be viewed as acting as buffers or barriers between the inher-
ently unstable energetic molecules which consequently im-
proves the sensitivity.
Figure 14. Densities of N-heterocyclic compounds and their corresponding
mono- or bis-N-oxides.
ten N-heterocyclic compounds and their corresponding mono-
and/or bis-N-oxides[41] (Figure 14).
Although this data set may not be very large, the general
trend is nevertheless clear and unambiguous. In each case, the
oxidized species displays a higher density (as determined by
single-crystal X-ray diffraction) than the parent N-heterocycle
and the increases range from a few percent to almost 50% in
the case of pyrazine versus pyrazine bis-N-oxide. The types of
compounds in this series also span a large range and include
molecules decorated with a variety of functional groups.
Against this background, it is not surprising then that the
EDNA co-crystal with a density that is comparable to EDNA
itself involves pyrazine bis-N-oxide as the co-former. Further-
more, it may be particularly worthwhile to examine N-oxides
as co-formers in future strategies geared towards enhancing
the density of heteromeric energetic materials.
The acidity of pure EDNA makes it corrosive and prone to
react with metals which can restrict processability and shelf-
life. Furthermore, the sensitivity of the products formed can
create a completely new set of unpredictable and undesirable
In this systematic approach, structural consistency was a de-
sired outcome as we could then also expect a positive correla-
tion between packing coefficient and density (no heavy atoms
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11034
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
challenges. One reason for co-crystallizing EDNA with mole-
cules capable of binding to the acidic protons was to intro-
duce a “supramolecular protecting group” that might reduce
the chemical reactivity of EDNA thereby making it more stable.
In order to test the validity of this approach, we carried out
a simple qualitative test where EDNA and EDNA:A8, respec-
tively, were sprinkled in a slurried form (with a few drops of
methanol) directly onto strips of copper metal (Figure 16).
bility since the structural consistency throughout this series un-
earthed some correlations between molecular structure/prop-
erty of the pure co-former and the physical properties per-
formance of the resulting energetic co-crystal. This clearly sug-
gests that systematic co-crystallizations may allow us to fine-
tune properties that are important for storage, handling, and
processing, with minimal negative impact on the eventual en-
ergetic behavior of the targeted substance.
Experimental Section
All the precursors, solvents and acceptors A1–A6 and A11 were
purchased from commercial sources and were used without further
purification. EDNA was synthesized following a reported proce-
dure.[43] A8–A10 and A13–A15 were synthesized following previ-
ously reported procedures.[44] A7[45] and A12[46] were also synthe-
1
sized following reported procedures. H NMR spectra were record-
ed on a Varian Unity plus 400 MHz spectrometer. Infrared spectra
were recorded with a Nicolet 380 FT-IR. Differential scanning calo-
rimetry (DSC) measurements were carried out on a Q20, TA Instru-
ments using Tzero aluminum pans under nitrogen (50.0 mLminÀ1
purge at a heating rate of 58CminÀ1 over a range of 20–3008C.
)
Standard protocols for handling energetic materials were followed
at all times.[47] All synthesis and handling of EDNA was carried out
inside a fume hood and nonmetallic spatulas were used when
measuring EDNA. Kevlar gloves, laboratory coat and goggles were
worn at all times. Mechanical actions involving scratching or scrap-
ing were avoided and EDNA was synthesized only in small
amounts.
Figure 16. A slurry of EDNA (top left) and EDNA:A8 (top right), in contact
with copper metal. Notable corrosion of the metal was observed after 20 h
in the case of pure EDNA (bottom left) but no visible change to the copper
metal took place with EDNA:A8 under the same conditions (bottom right).
Synthesis of EDNA
A round-bottom flask was equipped with a stir bar and 7 mL of
90% nitric acid (fuming nitric acid) under nitrogen atmosphere
and cooled to 08C. To this, 2.0 g of ethylene urea (2-imidazolidone)
was added portion-wise while keeping the temperature below
208C. The reaction mixture was stirred under these conditions for
1 h and subsequently allowed to warm to room temperature. The
reaction mixture was then quenched with iced water and the pre-
cipitate that formed was collected by filtration.
The corrosive nature of EDNA is clearly manifested even
after a relatively short amount of time (20 h) at ambient tem-
perature, whereas the EDNA:A8 solid form does not display
any noticeable degrading of the copper metal. Much more
work needs to be done in order to accurately quantify the
damaging effect that the EDNA–metal system is experiencing
and we also intend to look more broadly at the potential dif-
ferences that EDNA and co-crystals thereof may have on mate-
rials that are likely to be used in processing, storage and for-
mulation of energetic materials of this type.
The precipitate was added to about 50 mL of boiling water in
small portions and the temperature of water was maintained until
no more gas evolution was seen. Then the aqueous solution was
cooled in an ice bath and the precipitated desired product was col-
lected by filtration as a white crystalline powder in 55% yield. Melt-
ing point 178–1808C (reported 174–1788C);[48] 1H NMR (dH; DMSO,
400 MHz): 12.12 (sbr, 2H), 3.59 ppm (s, 4H).
Conclusions
Theoretical calculations
A systematic co-crystallization study of EDNA has demonstrat-
ed that the acidic protons in the energetic material can be suc-
cessfully targeted with suitable hydrogen-bond acceptors. Six
of the eight co-crystals synthesized were characterized using
single-crystal diffraction and the outcome was predictable
supramolecular motifs based upon NÀH···N and NÀH···O struc-
ture-directing hydrogen bonds. The co-formers also act as
“supramolecular protecting groups” resulting in a reduced
chemical instability/corrosiveness which is otherwise detrimen-
tal to the storage and processability of EDNA. Thermal proper-
ties, impact sensitivity, and detonation velocities and pressure
could also be modified and altered with a degree of predicta-
Geometry optimizations of EDNA at various fixed torsion angles
were carried out with DFT calculations at the B3LYP/6-31G* level of
theory using Spartan’08.[49] The same level of theory was used in
calculating the electrostatic potentials on the acceptors (see the
Supporting Information).
Synthesis and characterization of co-crystals
EDNA was combined with the 15 acceptors in stoichiometric ratios
and dissolved in a minimum amount of methanol and kept in vials
for slow evaporation in order to obtain co-crystals suitable for
single crystal X-ray diffraction (see the Supporting Information for
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11035
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[9] D. Fischer, T. M. Klapçtke, D. G. Piercey, J. Stierstorfer, Chem. Eur. J. 2013,
19, 4602–4613.
[10] a) P. Vishnoi, M. G. Walawalkar, R. Murugavel, Cryst. Growth Des. 2014,
14, 5668–5673; b) Y. Ma, A. Zhang, C. Zhang, D. Jiang, Y. Zhu, C. Zhang,
Cryst. Growth Des. 2014, 14, 4703–4713; c) H. Li, Y. Shu, S. Gao, L. Chen,
Q. Ma, X. Ju, J. Mol. Model. 2013, 19, 4909–4917.
[11] a) Y. Ma, A. Zhang, X. Xue, D. Jiang, Y. Zhu, C. Zhang, Cryst. Growth Des.
2014, 14, 6101–6114; b) X. Liu, Z. Su, W. Ji, S. Chen, Q. Wei, G. Xie, X.
Yang, S. Gao, J. Phys. Chem. C 2014, 118, 23487–23498; c) J. P. Shen,
X. H. Duan, Q. P. Luo, Y. Zhou, Q. Bao, Y. J. Ma, C. H. Pei, Cryst. Growth
Des. 2011, 11, 1759–1765.
[12] A. P. Agrawal, Prog. Energy Combust. Sci. 1998, 24, 1–30.
[13] T. M. Klapçtke, D. G. Piercey, J. Stierstorfer, Eur. J. Inorg. Chem. 2013,
1509–1517.
X-ray experimental details). Once the crystals were obtained they
were analyzed using IR spectroscopy and DSC to identify potential
co-crystals (see the Supporting Information). From the initial
screening with IR and DSC eight out of 15 experiments produced
co-crystals. Out of the eight co-crystals obtained, six yielded crys-
tals suitable for single-crystal diffraction: EDNA:A3, EDNA:A4,
EDNA:A5, EDNA:A7, EDNA:A8 and EDNA:A12. Thermal properties
of all eight co-crystals were obtained using DSC data and detona-
tion properties were calculated for the six compounds for which
experimental densities were available from crystallographic data.
For the co-crystals EDNA:A3, EDNA:A4, EDNA:A5 and EDNA:A8
room temperature unit cells and room temperature densities were
determined.
[14] a) O. Bolton, A. J. Matzger, Angew. Chem. Int. Ed. 2011, 50, 8960–8963;
Angew. Chem. 2011, 123, 9122–9125; b) Z. Yang, H. Li, H. Huang, X.
Zhou, J. Li, F. Nie, Propellants Explos. Pyrotech. 2013, 38, 495–501.
[15] a) S. R. Anderson, D. J. am Ende, J. S. Salan, P. Samuels, Propellants
Explos. Pyrotech. 2014, 39, 637–640; b) H. R. Li, Y. J. Shu, C. Song, L.
Chen, R. J. Xu, X. H. Ju, Chin. Chem. Lett. 2014, 25, 783–786.
[16] C. Guo, H. Zhang, X. Wang, X. Liu, J. Sun, J. Mater. Sci. 2013, 48, 1351–
1357.
[17] K. B. Landenberger, O. Bolton, A. J. Matzger, Angew. Chem. Int. Ed. 2013,
52, 6468–6471; Angew. Chem. 2013, 125, 6596–6599.
[18] C. Zhang, Z. Yang, X. Zhou, C. Zhang, Y. Ma, J. Xu, Q. Zhang, F. Nie, H. Li,
Cryst. Growth Des. 2014, 14, 3923–3928.
[19] C. B. Aakerçy, T. K. Wijethunga, J. Benton, J. Desper, Chem. Commun.
2015, 51, 2425–2428.
[20] H. Xue, G. Ye, B. Twamley, J. M. Shreeve, Chem. Mater. 2005, 17, 191–
198.
[21] G. C. Hale, US Patent 2.011.578, 1935, Dover, N.J. .
[22] United States Department of the Army Technical Manual, Military Explo-
sives, Department of the Army, 1989.
Impact sensitivity
Impact sensitivities of EDNA, EDNA:A3 and EDNA:A8 were mea-
sured by impact drop testing. The tests were carried out using an
apparatus designed to accommodate small amounts of material[50]
and was calibrated against tetranitromethane (H50 39(Æ1) cm). Ap-
proximately 0.2 mg of each sample was sealed into aluminum DSC
pans. A freefalling 5 lb weight was used to strike the samples from
different heights measured in centimeters, which subsequently al-
lowed us to calculate the H50, the height at which the material ex-
hibited a 50% chance of detonation using value using Bruceton
analysis.[51] It must be stated that this is a nonstandard testing ap-
paratus and the results obtained are useful for relative compari-
sons only.
[23] B. T. Fedoroff, O. E. Sheffield, Encyclopedia of Explosives and Related
Items, Vol. 6, Picatinny Arsenal, 1974.
[24] M. S. Kettner, T. M. Klapçtke, T. G. Müller, M. Suc’eska, Eur. J. Inorg. Chem.
2014, 4756–4771.
[25] a) H. S. Jadhav, M. B. Talawar, R. Sivabalan, D. D. Dhavale, S. N. Asthana,
V. N. Krishnamurthy, J. Hazard. Mater. 2007, 143, 192–197; b) G. HervØ,
C. Roussel, H. Graindorge, Angew. Chem. Int. Ed. 2010, 49, 3177–3181;
Angew. Chem. 2010, 122, 3245–3249.
Acknowledgements
This material is based upon work supported by, or in part by,
the US Army Research Laboratory and the US Army Research
Office under contract/grant number W911NF-13-1-0387.
[26] J. C. Hinshaw, W. W. Edwards, W. W. George, J. Gilardi, Heterocycl. Chem.
1992, 29, 1721–1724.
Keywords: co-crystals
·
crystal engineering
·
energetic
[27] L. I. Bagal, M. S. Pevzner, A. N. Frolov, N. I. Sheludyakova, Khim. Geterot-
sikl. Soedin. 1970, 558–562.
materials · ethylenedinitramine · hydrogen bonds
[28] R. Meyer, J. Kçhler, A. Homburg, Explosives, Wiley-VCH, Weinheim, 2008.
[29] V. P. Ivshin, T. N. Ivshina, L. G. Smirnova, M. G. Ponomareva, Zh. Org.
Khim. 1984, 20, 7–13.
[30] M. Robert, J. Pachman, Primary Explosives, Springer, Heidelberg, 2013.
[31] R. W. Millar, Lead-Free Initiator Materials for Small Electro-explosive De-
vices for Medium Caliber Munitions, QineiQ/FST/CR032702/1.1, 2003.
[32] A. O. Franz, US 2708623A, 1955, Olin Mathieson Chemical Corporation,
Virginia.
[1] a) D. Guo, Q. An, W. A. Goddard, S. V. Zybin, F. Huang, J. Phys. Chem. C
2014, 118, 30202–30208; b) T. Yan, K. Wang, X. Tan, J. Liu, B. Liu, B. Zou,
J. Phys. Chem. C 2014, 118, 22960–22967.
[2] a) J. H. Urbelis, J. A. Swift, Cryst. Growth Des. 2014, 14, 1642–1649;
b) K. B. Landenberger, A. J. Matzger, Cryst. Growth Des. 2010, 10, 5341–
5347.
[3] a) D. I. A. Millar, H. E. Maynard-Casely, D. R. Allan, A. S. Cumming, A. R.
Lennie, A. J. Mackay, I. D. H. Oswald, C. C. Tang, C. R. Pulham, CrystEng-
Comm 2012, 14, 3742–3749; b) C. Wei, H. Huang, X. Duan, C. Peia, Pro-
pellants Explos. Pyrotech. 2011, 36, 416–423; c) C. Ma, Z. Liu, Q. Yao,
Can. J. Chem. 2014, 92, 803–808.
[4] a) C. He, J. Zhang, D. A. Parrish, J. M. Shreeve, J. Mater. Chem. A 2013, 1,
2863–2868; b) C. J. Underwood, C. Wall, A. Provatas, L. Wallace, New J.
Chem. 2012, 36, 2613–2617.
[5] a) K. B. Landenberger, A. J. Matzger, Cryst. Growth Des. 2012, 12, 3603–
3609; b) G. H. Tao, Y. Guo, Y. H. Joo, B. Twamley, J. M. Shreeve, J. Mater.
Chem. 2008, 18, 5524–5530.
[6] G. Chen, W. Shi, M. Xia, W. Lei, F. Wang, X. Gong, J. Mol. Model. 2014,
20, 2326.
[7] Y. Wang, Z. Yang, H. Li, X. Zhou, Q. Zhang, J. Wang, Y. Liu, Propellants
Explos. Pyrotech. 2014, 39, 590–596.
[8] a) M. M. Qasim, B. Moore, L. Taylor, P. Honea, L. Gorb, J. Leszczynski, Int.
J. Mol. Sci. 2007, 8, 1234–1264; b) V. K. Balakrishnan, A. Halasz, J.
Hawari, Environ. Sci. Technol. 2003, 37, 1838–1843; c) J. Zhang, C. He,
D. A. Parrish, J. M. Shreeve, Chem. Eur. J. 2013, 19, 8929–8936.
[33] J. W. Turley, Acta Crystallogr. Sect. B 1968, 24, 942–946.
[34] C. B. Aakerçy, S. Panikkattu, B. DeHaven, J. Desper, CrystEngComm 2013,
15, 463–470.
ˇ
ˇ
´
[35] Q. L. Yan, S. Zeman, J. Selesovsky, R. Svoboda, A. Elbeih, J. Therm. Anal.
Calorim. 2013, 111, 1419–1430.
[36] a) B. Yan, N. N. Zhao, H. X. Ma, J. R. Song, F. Q. Zhao, R. Z. S. Hu, S. Afr. J.
Chem. 2013, 66, 136–139; b) M. J. Kamlet, H. Hurwitz, J. Chem. Phys.
1968, 48, 3685–3692; c) M. J. Kamlet, S. J. Jacobs, J. Chem. Phys. 1968,
48, 23–35.
[37] J. Barton, Dust Explosion Prevention and Protection, IChemE, Rugby,
2002.
[38] P. Patnaik, A Comprehensive Guide to the Hazardous Properties of Chemi-
cal Substances, Wiley, New York, 2007.
[39] S. Li, Q. Li, R. Li, R. Liu, K. Yang, B. Liu, B. Zou, J. Phys. Chem. C 2014,
118, 23443–23450.
[40] a) H. Wei, H. Gao, J. M. Shreeve, Chem. Eur. J. 2014, 20, 16943–16952;
b) P. F. Pagoria, G. S. Lee, A. R. Mitchell, R. D. Schmidt, Thermochim. Acta
2002, 384, 187–204.
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11036
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[41] CSD ConQuest 1.17, 2013 Cambridge Crystallographic Data Centre, UK.
[42] D. M. Badgujar, M. B. Talawar, S. N. Asthana, P. P. Mahulikar, J. Hazard.
Mater. 2008, 151, 289–305.
[43] T. K. Highsmith, J. M. Hanks, S. P. Velarde, J. C. Battaro, WO02060881,
2002, Alliant Techsystems Inc, Virginia.
[44] C. B. Aakerçy, T. K. Wijethunga, J. Desper, CrystEngComm 2014, 16, 28–
31.
[45] N. Iranpoor, H. Firouzabadi, D. Khalili, S. Motevalli, J. Org. Chem. 2008,
73, 4882–4887.
[48] L. W. Kissinger, M. J. Schwartz, Org. Chem. 1958, 23, 1342–1344.
[49] Spartan’08, Wavefunction, Inc.
[50] O. Bolton, L. R. Simke, P. F. Pagoria, A. J. Matzger, Cryst. Growth Des.
2012, 12, 4311–4314.
[51] a) K. B. Landenberger, O. Bolton, A. J. Matzger, J. Am. Chem. Soc. 2015,
137, 5074–5079; b) L. R. Simpson, M. F. Foltz, LLNL Small-Scale Drop-
Hammer Impact Sensitivity Test, Lawrence Livennore National Laborato-
ry.
[46] N. R. Ayyangar, S. N. Naik, K. V. Srinivasan, Tetrahedron Lett. 1989, 30,
7253–7256.
Received: May 2, 2015
[47] T. M. Klapçtke, B. Krumm, F. X. Steemann, G. Steinhauser, Safety Sci.
2010, 48, 28–34.
Published online on June 26, 2015
Chem. Eur. J. 2015, 21, 11029 – 11037
www.chemeurj.org
11037
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim