The twofold inter­penetrated three-connected three-dimensional (10,3)-net in 2-amino­ethene-1,1,2-tricarbonitrile

In the crystal structure of the title compound, C₅H₂N₄, each mol­ecule is linked by N-H...N[triple bond]C hydrogen bonds to four other mol­ecules, thus forming a network that can be described as a twofold inter­penetrated three-connected three-dimensional (10,3)-net. The inter­penetrated nets are related by (010) translation. If only inter­molecular hydrogen bonds are taken into account, these nets can be considered as independent. However, the inter­actions between the cyano groups from different nets indicate mutual assistance of the two nets during their formation.

S1. Comment

Enaminonitriles are used as precursors in the synthesis of a wide variety of heterocyclic systems, such as pharmaceuticals, fungicides and solvatochromic dyes (Erian, 1993), as well as of various coordination compounds (Sidorov et al., 1998 or 1999). Therefore, investigations of the preparation and properties of the simplest enaminonitriles used for further chemical transformations form a major part of modern chemistry. We present here a novel and very efficient approach (94% yield) to the synthesis of 2-aminoethene-1,1,2-tricarbontrile, (I). The crystallographic study of (I) extends our systematic investigation of the structural chemistry of enaminonitriles (Nasakin et al., 1991; Tafeenko et al., 1994a,b,c,d; Bogdan et al., 1996a,b), as well as of molecules containing the dicyanoethylene fragment (Nasakin et al., 1992; Tafeenko et al., 2003, 2005; Tafeenko, Nikolaev et al., 2004; Tafeenko, Peschar et al., 2004). Fig. 1 shows the molecular structure of (I) with the atom-labelling scheme, while the molecular dimensions are listed in Table 1. Particularly notable are the very short distance to the amine group (C2—N4), the lengthening of the adjacent C1═C2 double bond, and the inequality of the C—C bonds that link the cyano groups to the ethylene moiety. All of these features? result from a strong π conjugation of the amine group with the two cyano groups attached to atom C1, whereas the –C3/N1 cyano group does not take part in such conjugation. [In terms of energy, π–π C2—C3 bonding is not advantageous, as it would result in either unpaired electrons or a large (+2) positive charge on atom N4 of the molecule.] Additional evidence of this conjugation is the planarity of the amine group [the angle sum around atom N4 is 360 (2)°]. Previous studies of enaminonitriles showed that the triple-bond distance in the cyano group is not sensitive to the action of different substituents, and this insensitivity is also seen in (I) (Table 1). Meanwhile, if we knew which of the cyano groups acquired additional negative charge we could understand the preferences in hydrogen bonding between the amine and cyano groups in the solid state. Atomic Mulliken charges were calculated with GAUSSIAN98 (Frisch et al., 1998) at the B3LYP/6–311G** level (see Table 3). To elucidate how the amine group affects the charge distribution, the calculation was also carried out for 1,2,2-tricyanoethylene, (II). The data indicate that in (I) the negative charges on atoms N2 and N3 increase, while that on atom N1 decreases. The different charges correspond to the fact that only atoms N2 and N3 participate in hydrogen bonds (Table 2 and Fig. 2).

Previously, we concluded that the prevailing type of crystal packing of enaminonitriles includes formation of hydrogen-bonded centrosymmetric dimers (27 structures out of a total of 38), whereas hydrogen-bonded chains and other types of aggregation (which usually occur if a molecule of enaminonitrile contains other functional groups, such as hydroxy or ketone groups, or if a crystal contains solvent molecules, which can lock the H atoms of the amine group) are relatively rare (Tafeenko et al., 1991). On the basis of this observation and quantum chemical calculations, we suggested that the high probability of dimer formation is provided not only by hydrogen bonding between amine and cyano groups, but also by dipole–dipole interactions between the cyano groups; all dimer-containing structures exhibited short distances (2.9–3.3 Å) between the cyano? groups. Clearly, cyano–cyano interaction very often plays a decisive role in the form of molecular aggregation in these solids (e.g. see Tafeenko et al., 2005).

Preferential dimer formation was confirmed by a search of the Cambridge Structural Database (CSD; Version 5.22; Allen, 2002). Under consideration were only the structures of organic compounds that had no disorder and refined to R < 0.075. Data visualization was performed with Mercury (Version 1.2; Bruno et al., 2002). Among 105 retrieved structures, 65 (62%) contain dimers, 17 (16%) contain chains and 23 (22%) adopt other types of aggregation. All dimeric structures have centrosymmetric space groups, viz. P-1 (34 structures), P2₁/(n,c) (24) and C2/c (seven). Among the crystals that contain chains, eight structures (50%) have non-centrosymmetric space groups, viz. P2₁2₁2₁ (four structures), Pna2₁ (two), and P2₁ and Pca2₁ (one each).

There are several approaches to the analysis of hydrogen-bonding patterns, but each of them provides an incomplete description of the picture observed in (I). According to Kuleshova & Zorky (1980), this pattern could be described as F₄⁴(6), thereby expressing that the six-molecule building blocks form a hydrogen-bonded framework, where every molecule is connected to four others. The building block could be described as R₆⁶(32) (Etter et al., 1990; Bernstein et al., 1995). According to Batten & Robson (1998), the network in (I) can be assigned as an interpenetrating three-connected three-dimensional (10,3)-net. The motif (Fig. 2) consists of ten nodes; five nodes are atoms N4 and five are atoms C1. Each node is connected to two neighbors in a one-dimensional planar zigzag chain. The third connection at every node is to a node belonging to a zigzag chain running underneath at an interplanar angle of 61° to the original chain (Fig. 2). This structure conforms to the definition of a (10,3)-b-system. The second interpenetrating net originates from the first by a (0, 1, 0) translation. The interpenetration of the two nets is depicted in Fig. 3 (for clarity, only ring-forming atoms are shown). Alternatively, the network of (I) could be represented as parallel helices along the c axis with opposite handedness, which conform to the definition of a (10,3)-a-net. Fig. 4 shows how these helices penetrate each other as a resutl of interaction of the cyano groups not involved in hydrogen bonding (–C3/N1) with the cyano groups taking part in hydrogen bonding (–C4/N2). The N1···C4(x,1 + y,z) and N1···C4(1 - x,1 - y,-1/2 + z) distances are 3.257 (2) and 3.282 Å, respectively. In other words, we can conclude that the two nets are independent if we take into account only hydrogen bonds, but each of these nets is also formed as a result of mutual assistance and copolarization of cyano groups from different nets.

S2. Experimental

Tetracyanoethylene (0.64 g, 5 mmol) was added to a mixture of ammonium acetate (0.97 g, 12.5 mmol) and dioxane (5 ml). After 20 min of stirring, surplus ammonium acetate was filtered off. The solution was evaporated in vacuo, and the solid residue was triturated with hexane and filtered off to yield (I) (0.55 g, 4.7 mmol, 94%). Recrystallization from acetonitrile afforded yellow crystals of (I), which were suitable for X-ray analysis.

S3. Refinement

The positions of the H atoms were determined from a Fourier difference map, and their coordinates and isotropic displacement parameters were refined freely. Owing to the absence of any significant anomalous scatterers in the molecule, the absolute configuration of the molecule was assigned arbitrarily.

Computing details

Data collection: CAD-4 Software (Enraf–Nonius, 1989); cell refinement: CAD-4 Software; data reduction: XCAD4 (Harms & Wocadlo, 1995); program(s) used to solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: DIAMOND (Brandenburg, 2000) and ORTEP-3 (Farrugia, 1997); software used to prepare material for publication: SHELXL97.

Figures

	Fig. 1. The atom-labeling scheme of C5H2N4. Displacement ellipsoids are drawn at 50% probability level.
	Fig. 2. The arrangement of the R6^6(32) building blocks in the crystal. Every molecule is connected to four others by N4—H1···N2^i, N4—H2···N3^ii, N2···H1^iii—N4^iii and N3···H2^iv—N4^iv hydrogen bonds. Adjacent molecules in chains running along the a axis [e.g. the molecule at (x, y, z) and the related molecule with symmetry code i] make a dihedral angle of 61°, whereas molecules at (x, y, z) and related molecules with symmetry codes ii and iv are coplanar. [Symmetry codes: (i) 1/2 + x, -y, z; (ii) 3/2 - x, 1 + y, 1/2 + z; (iii) 1/2 + x, -y, z; (iv) 3/2 - x, -1 + y, 1/2 + z.]
	Fig. 3. The mutual interpenetration of two nets (shown as dark and light shaded balls) occurs as a result of hydrogen bonding (N4—H2)^*···N3^**. All atoms not involved in ring formation have been omitted for clarity. [Symmetry codes: (i) 3/2 - x, -1 + y, -1/2 + z; (ii) x, y, -1 + z; (iii) x, -1 + y, -1 + z; (iv) 3/2 - x, y, -1.5 + z.]
	Fig. 4. An illustration of how molecules of the `light' net form, on the one hand, their own helices and at the same time, owing to interaction between cyano groups, promote the formation of the `dark' helices. As a first approximation, the structure could be represented as follows: atom N1^i (dark) interacts with atom C4^ii (light) in the molecule with symmetry code ii, which forms the six-molecule building block (see Fig. 2) through atoms N2^ii and H1^ii; thereby the `light' molecule net is assembled within the `dark' net. Inside the `dark' helix, `light' atoms N1^v and N1^iii interact with `dark' atoms C4^ix and C4^viii, respectively, as well as with C4^v from the `light' net. Concurrently, molecule ii through atoms N3^ii and H2^ii forms two `light' helices with opposite handedness in which `dark' molecules are embedded. [Symmetry codes: (i) -1/2 + x, -y, 1 + z; (ii) -1/2 + x, -1 - y, 1 + z; (iii) x, -1 + y, 1 + z; (iv) 1.5 - x, y, 1/2 + z; (v) 1 - x, -y, 1/2 + z; (vi) 1/2 - x, y, 1/2 + z; (vii) -1 + x, -1 + y, 1 + z; (viii) x, y, 1 + z; (ix) 1 - x, -1 - y, 1/2 + z.]

2-aminoethene-1,1,2-tricarbonitrile

Crystal data

C5H2N4	Dx = 1.428 Mg m^−3
Mr = 118.11	Melting point: 152 K
Orthorhombic, Pca21	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: P 2c -2ac	Cell parameters from 25 reflections
a = 13.3146 (8) Å	θ = 15–19°
b = 5.7488 (4) Å	µ = 0.10 mm^−1
c = 7.1774 (6) Å	T = 295 K
V = 549.38 (7) Å^3	Prism, yellow
Z = 4	0.20 × 0.15 × 0.10 mm
F(000) = 240

Data collection

Enraf–Nonius CAD-4 diffractometr	Rint = 0.000
Radiation source: fine-focus sealed tube	θmax = 30.0°, θmin = 3.1°
Graphite monochromator	h = 0→18
non–profiled ω scans	k = 0→8
860 measured reflections	l = 0→10
860 independent reflections	2 standard reflections every 120 min
814 reflections with I > 2σ(I)	intensity decay: none

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.032	All H-atom parameters refined
wR(F^2) = 0.085	w = 1/[σ^2(Fo^2) + (0.0487P)^2 + 0.0401P] where P = (Fo^2 + 2Fc^2)/3
S = 1.07	(Δ/σ)max < 0.001
860 reflections	Δρmax = 0.12 e Å^−3
91 parameters	Δρmin = −0.18 e Å^−3
1 restraint	Extinction correction: SHELXL, Fc^*=kFc[1+0.001xFc^2λ^3/sin(2θ)]^-1/4
Primary atom site location: structure-invariant direct methods	Extinction coefficient: 0.0083 (17)

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
C1	0.61177 (11)	0.0533 (3)	0.8632 (3)	0.0289 (3)
C2	0.64949 (12)	0.2448 (3)	0.7749 (2)	0.0295 (3)
C3	0.57640 (13)	0.4060 (3)	0.6949 (3)	0.0349 (4)
C4	0.50661 (13)	0.0116 (3)	0.8659 (3)	0.0327 (3)
C5	0.67394 (12)	−0.1152 (3)	0.9537 (2)	0.0312 (3)
N1	0.51892 (15)	0.5274 (3)	0.6314 (4)	0.0535 (5)
N2	0.42326 (11)	−0.0304 (3)	0.8714 (3)	0.0461 (4)
N3	0.71926 (12)	−0.2530 (3)	1.0304 (3)	0.0430 (4)
N4	0.74543 (13)	0.2994 (2)	0.7520 (3)	0.0355 (3)
H1	0.7908 (15)	0.207 (4)	0.799 (4)	0.041 (6)*
H2	0.761 (2)	0.425 (4)	0.698 (5)	0.062 (8)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
C1	0.0299 (6)	0.0247 (6)	0.0320 (7)	0.0047 (5)	0.0003 (7)	−0.0001 (7)
C2	0.0346 (7)	0.0270 (6)	0.0268 (6)	0.0027 (6)	0.0002 (6)	−0.0037 (6)
C3	0.0429 (8)	0.0269 (7)	0.0350 (8)	0.0031 (6)	−0.0008 (7)	0.0017 (7)
C4	0.0327 (7)	0.0305 (6)	0.0347 (8)	0.0004 (5)	−0.0029 (8)	0.0014 (7)
C5	0.0309 (7)	0.0288 (7)	0.0340 (8)	−0.0011 (6)	0.0011 (6)	−0.0015 (6)
N1	0.0626 (13)	0.0425 (9)	0.0553 (11)	0.0156 (9)	−0.0121 (10)	0.0040 (9)
N2	0.0354 (7)	0.0462 (8)	0.0566 (10)	−0.0019 (7)	−0.0022 (9)	0.0034 (9)
N3	0.0435 (8)	0.0367 (7)	0.0487 (9)	0.0088 (6)	−0.0025 (8)	0.0113 (7)
N4	0.0328 (6)	0.0311 (6)	0.0426 (8)	−0.0009 (6)	0.0004 (7)	0.0071 (7)

Geometric parameters (Å, º)

N4—C2	1.326 (2)	N3—C5	1.138 (2)
N4—H1	0.87 (2)	C2—C1	1.366 (2)
N4—H2	0.85 (3)	C2—C3	1.462 (2)
N1—C3	1.131 (3)	C1—C4	1.421 (2)
N2—C4	1.136 (2)	C1—C5	1.430 (2)
C2—N4—H1	118.5 (14)	C2—C1—C4	120.34 (15)
C2—N4—H2	120.0 (19)	C2—C1—C5	122.94 (14)
H1—N4—H2	121 (2)	C4—C1—C5	116.72 (15)
N4—C2—C1	127.07 (15)	N1—C3—C2	178.7 (2)
N4—C2—C3	116.27 (15)	N2—C4—C1	177.2 (2)
C1—C2—C3	116.65 (14)	N3—C5—C1	176.56 (19)

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
N4—H1···N2^i	0.87 (2)	2.10 (2)	2.955 (2)	166 (2)
N4—H2···N3^ii	0.85 (3)	2.22 (3)	3.061 (2)	171 (3)

Symmetry codes: (i) x+1/2, −y, z; (ii) −x+3/2, y+1, z−1/2.