Comment

Intramolecular cyclization of 2-amino-Z-chalcones is nowadays one of the most expeditious methods for the synthesis of the biologically significant 2-aryl-2,3-dihydroquinolin-4(1H)-ones (Ahmed & van Lier, 2006, 2007; Low, Cobo, Cuervo et al., 2004). We report here the molecular and supramolecular structure of the title compound, (I), as a new example of this class of compound. We have recently reported the structures of the analogues (II)-(VI) (see scheme), which all exhibit completely different modes of supramolecular aggregation (Low et al., 2002; Low, Cobo, Nogueras et al., 2004). In compound (I), the supramolecular structure proves to be different from all examples previously studied.

Compound (I) crystallizes with Z' = 2 in the space group P. The two independent molecules (Fig. 1) are both nearly planar, as shown by the leading torsion angles (Table 1), with the sole exception of the five-membered rings, which both adopt envelope conformations folded across the OO vectors. The ring-puckering parameters (Cremer & Pople, 1975) for the atom sequences O1n, Cn2, On3, Cn3a, Cn7a (where n = 1 or 2) are 212.2 (6)° when n = 1 and 30.6 (5)° when n = 2, so that the two molecules within the selected asymmetric unit are approximately mirror images of one another.

The bond distances in the two independent molecules are very similar (Table 1) and they provide evidence for significant bond fixation within the aminoaryl rings. For example, the Cn3a-Cn4 and Cn7-Cn7a bonds (where n = 1 or 2) are all short, while the Cn5-Cn6 and Cn6-Cn7 bonds are all long. In addition, the Cn6-Cn8 bonds are short for their type, while the Cn8-On8 bonds are long. This pattern of behaviour closely mimics that in compounds (II)-(VI) and indicates the importance of the charge-separated form A as an important contributor to the overall molecular-electronic structure, alongside the delocalized form B.

In each of the two molecules of (I), there is an intramolecular N-HO hydrogen bond (Table 2). The actions of the intermolecular hydrogen bonds generate two entirely different substructures, one composed solely of type 1 molecules and other entirely of type 2 molecules. In the type 1 substructure, amino atom N15 in the type 1 molecule at (x, y, z) acts as hydrogen-bond donor to pyridyl atom N114 in the type 1 molecule at (x, 1 + y, 1 + z), so generating by translation a C(11) chain (Bernstein et al., 1995) running parallel to the [011] direction. In addition, aryl atom C116 at (x, y, z) acts as hydrogen-bond donor to carbonyl atom O18 in the type 1 molecule at (1 - x, 2 - y, 1 - z), so generating by inversion an R₂²(14) ring centred at (, 1, ). The combination of these two motifs then generates a chain of edge-fused rings along [011], with R₂²(14) rings centred at (, n, n - ) (n = zero or integer), and R₄⁴(20) rings, alternatively described as R₆⁴(16) rings if the intramolecular hydrogen bond is included, centred at (, n + , n) (n = zero or integer) (Fig. 2).

In contrast with the one-dimensional substructure formed by the type 1 molecules, the substructure built from type 2 molecules is two-dimensional. Atom N25 in the type 2 molecule at (x, y, z) acts as hydrogen-bond donor to pyridyl atom N214 in the type 2 molecule at (x, 1 + y, 1 + z), so forming by translation a C(11) chain along [011], exactly analogous to that formed by the type 1 molecules. In addition, atom C22 at (x, y, z) acts as hydrogen-bond donor to carbonyl atom O28 in the type 2 molecule at (-1 + x, y, z), so generating by translation a C(8) chain running parallel to the [100] direction. The combination of the [100] and [011] chains generates a sheet lying parallel to (01) and containing equal numbers of S(6) and R₅⁴(33) rings (Fig. 3).

Thus, the two types of molecule in compound (I) form substructures which are entirely different. There are no direction-specific interactions between adjacent chains of type 1 molecules, or between adjacent sheets of type 2 molecules, nor are there any direction-specific interactions between the two different substructures. Instead, the type 1 chains simply lie between pairs of type 2 sheets.

Two other members of this series also crystallize with Z' = 2. In compound (V) (Low et al., 2002), the two independent molecules are linked by N-HO hydrogen bonds to form cyclic centrosymmetric tetramers containing two molecules of each type, and the tetramers are further linked into chains by C-HO hydrogen bonds. Compound (III) exhibits concomitant polymorphism as it crystallizes from solution in dimethylformamide as a mixture of monoclinic (Z' = 1) and triclinic (Z' = 2) crystals (Low, Cobo, Nogueras et al., 2004). In the polymorph having Z' = 2, one type of molecule forms a simple chain and the other type is pendent from it. Neither of compounds (III) and (V) with Z' = 2 contains two distinct substructures analogous to those found here in compound (I).

For none of compounds (I)-(V), which differ only in the identity of a single substitution in an aryl ring, could the supramolecular structure of any single example be reliably predicted from a detailed knowledge of the supramolecular structures of the remainder. Not even the type of direction-specific intermolecular interaction manifest in each structure is readily predictable. The wide range of supramolecular structures observed in this series, combined on the one hand with the occurrence of two completely distinct and independent substructures formed by the two independent molecules in compound (I) and, on the other, with the occurrence of concomitant polymorphism in compound (III), together present a very keen challenge to the attempted prediction from first principles of the crystal structures, dominated by weak direction-specific intermolecular forces, of simple molecular compounds, an endeavour where convincing success is still elusive (Lommerse et al., 2000; Motherwell et al., 2002; Day et al., 2005).

Figure 1 The two independent molecules of compound (I), showing the atom-labelling scheme for (a) a type 1 molecule and (b) a type 2 molecule. Displacement ellipsoids are drawn at the 30% probability level and H atoms are shown as small spheres of arbitrary radii.

Figure 2 A stereoview of part of the crystal structure of compound (I), showing the formation of a chain of edge-fused rings along [011] containing only type 1 molecules.

Figure 3 A stereoview of part of the crystal structure of compound (I), showing the formation of a sheet parallel to (01) containing only type 2 molecules.

Experimental

A solution in ethanol (10 ml) of 6-amino-3,4-methylenedioxyacetophenone (2.8 mmol), pyridine-4-carbaldehyde (2.8 mmol) and aqueous sodium hydroxide solution (0.5 ml of a 20% solution) was stirred at room temperature for 3 h. The precipitate thus formed was collected by filtration and washed with ethanol, yielding compound (I) as an orange solid (yield 70%; m.p. 469 K). MS (70 eV) m/e (%): 268 (50, [M]⁺), 190 (100, [M - C₅H₄N]⁺). Crystals suitable for single-crystal X-ray diffraction were grown from a solution in ethanol.

Crystal data

C15H12N2O3 Mr = 268.27 Triclinic, a = 10.1044 (3) Å b = 11.7533 (5) Å c = 12.1655 (5) Å = 117.232 (2)° = 97.480 (3)° = 99.585 (2)° V = 1231.11 (9) Å3 Z = 4 Dx = 1.447 Mg m-3 Mo K radiation = 0.10 mm-1 T = 120 (2) K Lath, orange 0.44 × 0.38 × 0.07 mm

Data collection

Bruker-Nonius KappaCCD area-detector diffractometer and scans Absorption correction: multi-scan (SADABS; Sheldrick, 2003) Tmin = 0.967, Tmax = 0.993 28373 measured reflections 5675 independent reflections 3599 reflections with I > 2(I) Rint = 0.067 max = 27.7°

Refinement

Refinement on F2 R[F2 > 2(F2)] = 0.059 wR(F2) = 0.169 S = 1.03 5675 reflections 361 parameters H-atom parameters constrained w = 1/[2(Fo2) + (0.0873P)2 + 0.3836P] where P = (Fo2 + 2Fc2)/3 (/)max < 0.001 max = 0.25 e Å-3 min = -0.30 e Å-3

Table 1 Selected geometric parameters (Å, °) C13a-C14 1.364 (3)   C14-C15 1.427 (3) C15-C16 1.428 (3) C16-C17 1.425 (3) C17-C17a 1.357 (3) C17a-C13a 1.390 (3) C15-N15 1.356 (3) C16-C18 1.459 (3) C18-O18 1.242 (3) C23a-C24 1.353 (3) C24-C25 1.426 (3) C25-C26 1.428 (3) C26-C27 1.426 (3) C27-C27a 1.354 (3) C27a-C23a 1.394 (3) C25-N25 1.351 (3) C26-C28 1.467 (3) C28-O28 1.242 (2) C15-C16-C18-C19 176.55 (19) C16-C18-C19-C110 -152.2 (2) C18-C19-C110-C111 -179.3 (2) C19-C110-C111-C112 9.0 (4) C25-C26-C28-C29 171.38 (18) C26-C28-C29-C210 165.6 (2) C28-C29-C210-C211 179.56 (19) C29-C210-C211-C212 2.8 (3)

Table 2 Hydrogen-bond geometry (Å, °) D-HA D-H HA DA D-HA N15-H15AO18 0.96 1.90 2.657 (3) 134 N15-H15BN114i 0.96 2.09 3.054 (3) 179 C116-H116O18ii 0.95 2.55 3.430 (3) 154 N25-H25AO28 0.96 1.94 2.662 (2) 130 N25-H25BN214i 0.96 2.04 2.996 (3) 176 C22-H22AO28iii 0.99 2.41 3.270 (3) 145 Symmetry codes: (i) x, y+1, z+1; (ii) -x+1, -y+2, -z+1; (iii) x-1, y, z.

Crystals of compound (I) are triclinic. The space group P was selected and confirmed by the structure analysis. All H atoms were located in difference maps and then treated as riding atoms, with C-H = 0.95 (aromatic and alkenic) or 0.99 Å (CH₂) and N-H = 0.96 Å, and with U_iso(H) = 1.2U_eq(C,N).

Data collection: COLLECT (Nonius, 1999); cell refinement: DENZO (Otwinowski & Minor, 1997) and COLLECT; data reduction: DENZO and COLLECT; program(s) used to solve structure: OSCAIL (McArdle, 2003) and SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: OSCAIL and SHELXL97 (Sheldrick, 1997); molecular graphics: PLATON (Spek, 2003); software used to prepare material for publication: SHELXL97 and PRPKAPPA (Ferguson, 1999).

Supplementary data for this paper are available from the IUCr electronic archives (Reference: GG3063 ). Services for accessing these data are described at the back of the journal.