2-Nitr­imino-1-nitro­imidazolidine

The mol­ecule of the title compound, C₃H₅N₅O₄, has a planar geometry. Due to the presence of a second nitro group, the lengths of the two conjugated C-N bonds are different. Nevertheless, the conjugation of the nitr­imine group spreads to the nitr­amine group. Intermolecular N-H...O bonds connect mol­ecules into dimers in the crystal structure.

S1. Comment

Nitrimines, of which the title compound, (I), is an example, are of interest as highly energetic compounds (McKay, 1952). According to McKay & Wright (1948), (I) has high explosive characteristics. The efficiency of (I) is 1.3 times greater than that of TNT (trinitrotoluene) in a ballistic-mortar test and 1.5 times greater according to the Trauzl-block test. In spite of the moderate value of the oxygen balance (−41.1%), the brisance (?) of (I) is comparable with that of the more balanced (−21.6%) well known explosive compound RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine). At the same time, compound (I) is 2.8 times more sensitive to impact and 1.6 times more sensitive to friction than RDX. At first, one supposes that in limits of one class of high energetic compounds, the sensitivity to mechanical influence increases with increasing oxygen balance (Kamlet, 1976). RDX and (I) are both N-nitro compounds. In connection with this, the question arises as to why compound (I) with the worse oxygen balance shows comparative brisance and essentially higher impact sensitivity than RDX.

At present, the most widespread theory is the hot-spot theory of explosion initiation by impact (Kamlet, 1976; Dubovik, 1986). In agreement with this theory, an impact produces so-called `hot spots', where the thermal decomposition of an explosive happens. Many different physical and chemical factors influence the impact sensitivity: a heat yielded at decomposition, heat capacity, heat conduction, crystal size and shape, crystal lattice energy, and so on. Nevertheless, the thermal stability of an explosive (that is, mechanism and kinetics of initial thermal decomposition) plays a main determining role. The thermal decomposition of N-nitro compounds starts with homolitic breaking of the least stable N—NO₂ bond (Manelis et al., 1996). Investigation of the thermal decomposition of compound (I) showed that its thermal stability is essentially less than that of RDX; this agrees with the relative impact sensitivity of both explosives (Astachov et al., 2002). It is unlikely that the N—NO₂ bond of the nitrimine group is responsible for the low thermal stability and the high sensitivity of (I). For instance, one of best known energetic nitrimines, nitroguanidine, has comparatively high thermal stability (McKay, 1952; Volk, 1985; Oyumi et al., 1987; Liu et al., 1989) and is one of less impact sensitive explosive of IHE class (insensitive high explosives) (Doherty & Simpson, 1997). Therefore, one may suppose that the N—NO₂ bond of the second nitramine group triggers thermal decomposition in (I). One may expect that the bond will be longer and, consequently, its strength will be less than the strength of analogous bonds in RDX. It is therefore important to study the structure of (I).

The structure of (I) is similar to the structures of other nitrimines (Bryden et al., 1956; Choi, 1981; Nordenson, 1981a,b; Nordenson & Hvoslef, 1981; Rice et al., 1984; Oyumi et al., 1987; Gao et al., 1991; Astachov et al., 2001; Vasiliev et al., 2001; Allen, 2002). The molecular conformation of (I) is close to being planar (Fig. 1). Deviations from the least-squares plane through the non-H atoms are 0.078 (2) Å (r.m.s.) and 0.202 (2) Å (maximum). There is the intramolecular hydrogen bond O1···H1 in the nitrimine fragment of the molecule. Because of the delocalization of the π-electron density over the nitrimine fragment, the C—N, N—N and N—O bond lengths possess values intermediate between the values characteristic for corresponding single and double bonds (Table 1). The C1—N2 bond [1.322 (3) Å] that is technically `double', is, in fact, slightly longer than the C1—N3 bond [1.308 (3) Å] that is technically `single'. A strong electron-acceptor substituent, viz. the nitro group on atom N4, decreases the electron density on this atom. The possibility of its participation in conjugation with the nitrimine group is diminished and, as a consequence, the C1—N4 bond length is increased [1.379 (3) Å]. A similar situation takes place in other nitrimines which have electron-acceptor substituents, e.g. 1-methyl-2-nitro-1-nitrosoguanidine (Nordenson & Hvoslef, 1981; Rice et al., 1984) and nitroguanyl azide (Vasiliev et al., 2001). In these compounds, the analogous C—N bond length is 1.389–1.408 Å. These increased values agree with the ease of nucleophilic replacement reaction proceeding observed in experiments with this compounds (McKay & Wright, 1947; McKay, 1949, 1952; Meen & Wright, 1952; Scott et al., 1956). Judging by the planar geometry of the molecule of (I) and the value of the N4—N5 bond length [1.373 (3) Å], atom N4 participates in conjugation not only with the nitroguanyl group but also with the nitro group. Therefore, the nitrimine conjugation spreads to the nitro group in (I). Methylene groups are not involved in any conjugation. Bond lengths N3—C2 [1.455 (3) Å] and N4—C3 [1.460 (3) Å] are close to values observed in compounds with single C—N bonds (Allen, 2002), in particular, i2-nitriminoimidazolidine (Nordenson, 1981b). The C2—C3 bond length [1.498 (4) Å] is slightly shortened in comparison with an ordinary C—C single bond. This is probably due to some strain in imidazolidine cycle in (I). Two weak hydrogen bonds, N3—H1···O1, connect the molecules as dimers in the crystal (Fig. 2).

The values of N—NO₂ bond lengths in (I) do not exceed values of analogous bonds in RDX [1.351 (3), 1.392 (3) and 1.398 (3) Å; Choi & Prince, 1972]. Also, because of the conjugation in a dinitroguanyl fragment, the N4—N5 bond length [1.373 (3) Å] in (I) is smaller than in RDX. The supposition about the increased value of the N4—N5 bond length has not been confirmed. Therefore, one cannot explain the essential difference in thermal stability and impact sensitivity of (I) and RDX based on the chemical-bond strength in the molecules. Knowledge of the molecular structure does not help either. Nevertheless, the investigation allows one of the most probable factors to be excluded from consideration and to search for other possible reasons for the high impact sensitivity of (I).

The presence of firm intermolecular hydrogen bonds, which absorb an energy to be brought, is a necessary condition of low sensitivity of an explosive to external influences. One can suppose that just because of network of intermolecular hydrogen bonds, the nitroguanidine is a low sensitive explosive (Oyumi et al., 1987). At the same time, compound (I) has no such network (Fig. 2). Moreover, it is possible that the primary reaction of thermal decomposition is not the N4—N5 bond breaking but the breaking of the C1—N4 bond with the imidazolidine cycle opening in (I). An analysis of the above-mentioned factors and other factors will be reported elsewhere.

S2. Experimental

Caution: the title compound should be treated as a dangerous explosive! Compound (I) was synthesized as described by McKay & Wright (1948). Single crystals were obtained by crystallization from a solution in ethanol.

S3. Refinement

H atoms were found in a difference Fourier map and refined in the isotropic approximation as riding atoms, with displacement parameters equal to 1.2 times those of the parent atom. The determination of the absolute structure was not carried out because of the absence of strong anomalous scatterers.

Computing details

Data collection: KM-4 Software (Kuma, 1991); cell refinement: KM-4 Software; data reduction: DATARED in KM-4 Software; program(s) used to solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: SHELXTL (Sheldrick, 1995); software used to prepare material for publication: SHELXL97.

Figures

	Fig. 1. The molecule of (I), showing the atomic numbering scheme and displacement ellipsoids at the 50% probability level.
	Fig. 2. The molecular arrangement of (I) in the crystal. Dashed lines indicate hydrogen bonds.

2-Nitrimino-1-nitroimidazolidine

Crystal data

C3H5N5O4	Dx = 1.747 Mg m^−3
Mr = 175.12	Cu Kα radiation, λ = 1.5418 Å
Tetragonal, P42bc	Cell parameters from 24 reflections
Hall symbol: P 4c -2ab	θ = 23–28°
a = 12.897 (1) Å	µ = 1.41 mm^−1
c = 8.0078 (8) Å	T = 293 K
V = 1332.0 (2) Å^3	Ellipsoidal, colourless
Z = 8	0.36 × 0.33 × 0.30 mm
F(000) = 720

Data collection

Kuma KM-4 diffractometer	651 reflections with I > 2σ(I)
Radiation source: fine-focus sealed tube	Rint = 0.030
Graphite monochromator	θmax = 79.8°, θmin = 4.9°
θ/2θ scans	h = 0→15
Absorption correction: ψ scan (XPREP; Bruker, 1997)	k = −15→16
Tmin = 0.605, Tmax = 0.661	l = 0→10
847 measured reflections	2 standard reflections every 50 reflections
769 independent reflections	intensity decay: variation 0.8%

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: inferred from neighbouring sites
R[F^2 > 2σ(F^2)] = 0.029	H-atom parameters constrained
wR(F^2) = 0.079	w = 1/[σ^2(Fo^2) + (0.0455P)^2 + 0.1821P] where P = (Fo^2 + 2Fc^2)/3
S = 1.01	(Δ/σ)max = 0.009
769 reflections	Δρmax = 0.17 e Å^−3
110 parameters	Δρmin = −0.13 e Å^−3
1 restraint	Extinction correction: SHELXL97, Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
Primary atom site location: structure-invariant direct methods	Extinction coefficient: 0.0151 (10)

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Refinement. Refinement of F² against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F², conventional R-factors R are based on F, with F set to zero for negative F². The threshold expression of F² > σ(F²) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F² 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 (Ų)

	x	y	z	Uiso*/Ueq
N1	0.05645 (16)	0.82687 (17)	0.5585 (3)	0.0462 (5)
N2	0.14476 (14)	0.83921 (15)	0.4675 (3)	0.0406 (5)
N3	0.17532 (15)	1.01035 (16)	0.5790 (3)	0.0423 (5)
H1	0.1200	1.0173	0.6379	0.051*
N4	0.28962 (15)	0.94380 (16)	0.4149 (3)	0.0428 (5)
N5	0.34175 (14)	0.87781 (17)	0.3097 (3)	0.0458 (5)
C1	0.19573 (16)	0.92678 (16)	0.4926 (3)	0.0335 (4)
C2	0.2542 (2)	1.09067 (19)	0.5670 (4)	0.0530 (7)
H2	0.2869	1.1026	0.6744	0.064*
H3	0.2246	1.1553	0.5274	0.064*
C3	0.3303 (2)	1.0479 (2)	0.4435 (4)	0.0492 (7)
H4	0.3310	1.0883	0.3414	0.059*
H5	0.3998	1.0457	0.4897	0.059*
O1	0.02909 (15)	0.88779 (16)	0.6706 (3)	0.0592 (6)
O2	0.0081 (2)	0.7477 (2)	0.5279 (4)	0.0867 (10)
O3	0.30739 (16)	0.79207 (16)	0.2844 (3)	0.0633 (6)
O4	0.42167 (16)	0.9132 (2)	0.2528 (4)	0.0693 (7)

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
N1	0.0391 (10)	0.0473 (11)	0.0522 (12)	−0.0062 (8)	0.0036 (10)	0.0087 (10)
N2	0.0369 (9)	0.0403 (10)	0.0446 (10)	−0.0071 (8)	0.0041 (9)	−0.0022 (9)
N3	0.0420 (10)	0.0381 (11)	0.0467 (10)	0.0044 (8)	0.0065 (10)	−0.0026 (10)
N4	0.0367 (10)	0.0441 (11)	0.0477 (11)	−0.0073 (8)	0.0103 (9)	−0.0094 (9)
N5	0.0351 (10)	0.0555 (12)	0.0469 (12)	0.0029 (9)	0.0037 (9)	−0.0071 (10)
C1	0.0321 (9)	0.0354 (10)	0.0328 (10)	0.0017 (8)	0.0002 (9)	0.0024 (9)
C2	0.0575 (15)	0.0366 (11)	0.0649 (18)	−0.0093 (10)	0.0018 (15)	−0.0076 (13)
C3	0.0519 (14)	0.0406 (12)	0.0551 (16)	−0.0131 (11)	0.0026 (13)	0.0007 (11)
O1	0.0497 (11)	0.0550 (11)	0.0730 (14)	0.0049 (8)	0.0245 (10)	0.0057 (11)
O2	0.0741 (15)	0.0850 (16)	0.101 (2)	−0.0459 (14)	0.0193 (15)	−0.0138 (15)
O3	0.0534 (11)	0.0564 (11)	0.0799 (16)	0.0012 (8)	0.0113 (11)	−0.0268 (11)
O4	0.0440 (12)	0.0833 (16)	0.0805 (16)	−0.0043 (8)	0.0251 (11)	−0.0083 (13)

Geometric parameters (Å, º)

N1—O2	1.221 (3)	N4—C3	1.460 (3)
N1—O1	1.244 (3)	N5—O3	1.208 (3)
N1—N2	1.361 (3)	N5—O4	1.216 (3)
N2—C1	1.322 (3)	C2—C3	1.498 (4)
N3—C1	1.308 (3)	C2—H2	0.9700
N3—C2	1.455 (3)	C2—H3	0.9700
N3—H1	0.8600	C3—H4	0.9700
N4—N5	1.373 (3)	C3—H5	0.9700
N4—C1	1.379 (3)
O2—N1—O1	121.9 (2)	N3—C1—N4	106.5 (2)
O2—N1—N2	114.7 (2)	N2—C1—N4	120.2 (2)
O1—N1—N2	123.4 (2)	N3—C2—C3	103.9 (2)
C1—N2—N1	115.7 (2)	N3—C2—H2	111.0
C1—N3—C2	114.3 (2)	C3—C2—H2	111.0
C1—N3—H1	122.9	N3—C2—H3	111.0
C2—N3—H1	122.9	C3—C2—H3	111.0
N5—N4—C1	127.5 (2)	H2—C2—H3	109.0
N5—N4—C3	119.4 (2)	N4—C3—C2	101.9 (2)
C1—N4—C3	113.0 (2)	N4—C3—H4	111.4
O3—N5—O4	126.3 (2)	C2—C3—H4	111.4
O3—N5—N4	119.4 (2)	N4—C3—H5	111.4
O4—N5—N4	114.4 (2)	C2—C3—H5	111.4
N3—C1—N2	133.2 (2)	H4—C3—H5	109.2

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N3—H1···O1	0.86	2.06	2.568 (2)	117
N3—H1···O1^i	0.86	2.29	3.035 (2)	144

Symmetry code: (i) −x, −y+2, z.