Effect of counter-ion on packing and crystal density of 5,5′-(3,3′-bi[1,2,4-oxadiazole]-5,5′-diyl)bis(1H-tetrazol-1-olate) with five different cations

In energetic materials, the crystal density is an important parameter that affects the performance of the material. When making ionic energetic materials, the choice of counter-ion can have detrimental or beneficial effects on the packing, and therefore the density, of the resulting energetic crystal. Presented herein are a series of five ionic energetic crystals, all containing the 5,5′-(3,3′-bi[1,2,4-oxadiazole]-5,5′-diyl)bis(1H-tetrazol-1-olate) dianion.


Chemical context
One of the critical parameters directly related to the performance of an energetic material, specifically its detonation velocity, is the density of the material (Ma et al., 2014;Akhavan, 2011). This is an important consideration when designing energetic materials that incorporate counter-ions into their structures, since these counter-ions can, through supramolecular interactions, aid or disrupt effective packing of the molecule in question. Presented herein are the structures of a single energetic molecule, 5,5 0 -(3,3 0 -bi[1,2,4-oxadiazole]-5,5 0 -diyl)bis(1H-tetrazol-1-olate), as salts of five different cations: hydrazinium (1), hydroxylammonium (2) (Pagoria et al., 2017, included for comparison), dimethylammonium (3), 5-amino-1H-tetrazol-4-ium (4), and aminoguanidinium (5). As a result of the variety of cation structures and intermolecular interactions, each exhibits subtly different crystal packing, which affects the resulting density. The molecule of interest, however, only exhibits minor changes in bond distances depending on the cation.
In each structure, the oxadiazole oxygen atoms are on opposite sides. For 1, 2, 3, and 5 (Figs. 1-3, 5), the oxadiazole rings are coplanar with one another, with the N8-C9-C9 0 -N8 0 torsion angles constrained to 180 . Only slight deviation from coplanarity is seen in 4 (Fig. 4), with the N8-C9-C11-N12 torsion angles measuring 179.34 (16) . Coincidently, 4 is the only structure in which the primary molecule does not reside on an inversion center. For all structures, except 3, the tetrazolate ring is oriented such that the oxygen atoms of the oxadiazole and tetrazolate are on opposite sides, although 4 has a minor component of disorder      In all five structures, the tetrazolate C-N and N-N bond distances [ranging from 1.328 (5) to 1.351 (2) Å and 1.3170 (17) to 1.3455 (16) Å , respectively] suggest a delocalized aromatic system rather than discrete single and double bonds (Allen et al., 1987). The oxadiazole N-O, C-O, and C-N bond distances, however, suggest discrete single and double bonds. The N-O and C-O bonds range from 1.4033 (16) to 1.4115 (14) Å and 1.3391 (18) to 1.3468 (18) Å , respectively, suggesting single bonds between these atoms. The C-N bond opposite the oxygen atom ranges from 1.3671 (16) to 1.3755 (19) Å , also indicative of a single bond. The remaining C-N bonds range from 1.294 (2) to 1.309 (2) Å , typical for double bonds between these atoms. The central oxadiazole-oxadiazole C-C bond [ranging from 1.459 (3) to 1.465 (4) Å ] and the C-C bonds linking the oxadiazole rings to the tetrazolate rings [ranging from 1.432 (2) to 1.447 (2) Å ] are typical for C-C single bonds between non-fused heterocycles (Allen et al., 1987).

Supramolecular features
Packing of the energetic molecules will be described in four terms, following the example in Ma et al. (2014): sheet-like (with all molecules parallel to one another), wavelike (with two molecular planes that are not parallel to one another, but without intermolecular crossing), crossing (same as wavelike but with intermolecular crossing), and mixing (with molecular planes that do not fit in the prior three categories).

Figure 7
(a) Wavelike packing of 2 as seen down the c-axis, showing the opposing columns of dianion with hydroxylammonium occupying the space between the columns, and (b) view highlighting the hydrogen-bonding network (intermolecular contacts) between hydroxylammonium cation and the dianions. [Symmetry codes: (i) x À 1, y, z; (ii) x À 1, Ày + 3 2 , z + 1 2 ; (iii) x, y, z + 1.] below. The oxadiazole ring resides over the tetrazolate-oxadiazole C-C bond in the dianions above and below. The void space between the dianion columns is occupied by dimethylammonium ions, located within the plane of the molecules in an up-down arrangement. Two dimethylammonium ions are positioned between the sheets, forming hydrogen bonds between the NH 2 group and the tetrazolate oxygen atoms of dianions in neighboring sheets (Fig. 8b, Table 3). The tetrazolate ring engages in a staggeredinteraction with the oxadiazole rings of the neighboring dianion, at centroid C6/O7/N8/C9/N10 -centroid N1-N4/C5 distances of 3.51 (2) and 3.99 (2) Å (the latter distance to the inversion-related oxadiazole of the same dianion).
Structure 5, space group P2 1 /n, packs in a mixing pattern, with columns containing stacked sheets consisting of the dianion coplanar with two aminoguanidinium cations (Fig. 10a). Neighboring columns of sheets are rotated by 67 with respect to one another as a result of the hydrogen bonding of the amino group of the cation with the oxygen atom of a neighboring oxadiazole ring. In fact, it is the  Table 4 Hydrogen-bond geometry (Å , ) for 4.
As demonstrated above, it is the hydrogen-bonding networks that establish the crystal packing exhibited in each example, withand anion-interactions occurring if packing allows. As shown in Table 6, the densities of the crystals increase in the order 3 < 6 < 1 < 5 < 2. Unsurprisingly, the dimethylammonium, with minimal hydrogen bonding, non-interacting substituents, and a poor steric match for the dianion, is the least dense of the structures shown. Aminoguandinium, despite significant hydrogen bonding, exhibits a lower density as well, likely due to the directionality of the hydrogen-bond donors, which directs packing of the dianions into less efficient arrangements. Hydrazinium benefits from extensive hydrogen bonding, but the orientation of the hydrazinium directs the dianions into slightly less efficient packing than the hydroxylammonium cation, preventing the staggering of the columns that allows for improved space occupation. The 5-amino-1H-tetrazol-4-ium cation, with the second-highest density, packs very efficiently, in extended sheets with extensive hydrogen bonding, losing out to the hydroxylammonium cation likely only due to the included water molecules needed to fill in gaps between the dianions and cations. Hydroxylammonium exhibits the most efficient, highest-density packing due to the directing influence and strong hydrogen-bond donating ability of the hydroxyl group, which forms a short hydrogen bond and directs the columns into a staggered arrangement, fitting the dianions slightly closer together at the point where neighboring columns meet. The range of densities, from 1.544 to 1.873 g cm À1 , shows the significant effect that matching the hydrogen-bonding abilities and sterics of the counter-ion to the primary energetic ion has on efficient packing and, by extension, the expected performance of these ionic energetics.
Compound 1: Dihydrate 8 (0.15 g, 0.44 mmol) was added to a 20 ml vial with water (1.5 ml) and a stir bar. Hydrazine hydrate (45 ml, 0.93 mmol) was added to the reaction mixture and heated until dissolved. Stirring was discontinued, the stir bar was removed, and the solution was allowed to cool slowly providing crystals of 1.
Compound 3: In a round-bottom flask, fitted with a drying tube, was suspended chloroxime 6 (967 mg, 3.3 mmol) in dimethylformamide (DMF) (10 ml, anhydrous), which was then cooled in an ice-water bath. Sodium azide (472 mg, 7.26 mmol) was added in portions with stirring, and the reaction was allowed to warm to room temperature. Additional DMF (10 ml) was added to the creamy mixture, and after 1.5 h, the solids went into solution. At this point, complete formation of the diazidoxime was assumed, and cyclization to 1 proceeded as follows. A 1:1 mixture of diethyl ether/dioxane was added to the reaction mixture (100 ml total volume, anhydrous), and the solution was cooled to 273 K with an ice bath. HBr or Cl 2 gas was bubbled into the reaction at which time the temperature increased to 298 K. Gas was added until the reaction temperature returned to approximately 278 K, and the vessel was then stoppered and allowed to stir for 22 h. The voluminous, white precipitate that formed (hygroscopic dimethylamonium bromide) was separated by vacuum filtration, and the filtrate was allowed to evaporate from a crystallizing dish. Upon evaporation, a white solid (3) in a yellow oil remained. The solid was separated from the oil by vacuum filtration (535 mg). 3 was crystallized by heating in minimal water and slow cooling.
Compound 4: Dihydrate 7 (0.15 g, 0.44 mmol) was added to a 20 ml vial with water (1.5 ml) and a stir bar. 5-Aminotetrazole (0.10 g, 1.2 mmol) was added to the mixture, which was then heated with stirring until dissolved. Stirring was discontinued, the stir bar was removed, and the solution was allowed to cool slowly providing crystals of 4.
Compound 5: Dihydrate 7 (0.15 g, 0.44 mmol) was added to a 20 ml vial with water (1.5 ml) and a stir bar. Aminoguanidinium H 2 CO 3 (0.24 g, 1.8 mmol) was added to the mixture, which was then heated with stirring until dissolved. During dissolution, gas evolved, the solution became clear, followed by the formation of a tan precipitate. Heating was continued until complete dissolution, followed the removal of the stir bar, and slow cooling to provide crystals of 5.

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 7. In 5, the tetrazolate ring (N1, N2, N3, N4, C5, O1) is disordered over two positons (A and B) due to a 180 rotation in the terminal tetrazole rings. The disorder has the relative ratio of 90.7 (5) Scheme depicting synthesis pathways for the included structures.

Bis(hydrazinium) 5,5′-(3,3′-bi[1,2,4-oxadiazole]-5,5′-diyl)bis(1H-tetrazol-1-olate) (1)
Crystal data Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )

Bis(hydroxyammonium) 5,5′-(3,3′-bi[1,2,4-oxadiazole]-5,5′-diyl)bis(1H-tetrazol-1-olate) (2)
Crystal data Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.  Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq N1 0.59547 (17)    where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.37 e Å −3 Δρ min = −0.32 e Å −3 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.