|Home | About | Journals | Submit | Contact Us | Français|
In (2RS,4SR)-7-chloro-2-exo-(2-chloro-6-fluorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, C16H12Cl2FNO, (I), molecules are linked into chains by a single C—Hπ(arene) hydrogen bond. (2RS,4SR)-2-exo-(2-Chloro-6-fluorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, C16H13ClFNO, (II), is isomorphous with compound (I) but not strictly isostructural with it, as the hydrogen-bonded chains in (II) are linked into sheets by an aromatic π–π stacking interaction. The molecules of (2RS,4SR)-7-methyl-2-exo-(4-methylphenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, C18H19NO, (III), are linked into sheets by a combination of C—HN and C—Hπ(arene) hydrogen bonds. (2S,4R)-2-exo-(2-Chlorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, C16H14ClNO, (IV), crystallizes as a single enantiomer and the molecules are linked into a three-dimensional framework structure by a combination of one C—HO hydrogen bond and three C—Hπ(arene) hydrogen bonds.
We report here the structures of four new substituted 2-aryl-1,4-epoxytetrahydro-1-benzazepines, namely (2RS,4SR)-7-chloro-2-exo-(2-chloro-6-fluorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, (I), (2RS,4SR)-2-exo-(2-chloro-6-fluorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, (II), (2RS,4SR)-7-methyl-2-exo-(4-methylphenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, (III), and (2S,4R)-2-exo-(2-chlorophenyl)-2,3,4,5-tetrahydro-1H-1,4-epoxy-1-benzazepine, (IV) (Fig. 1 ). The work reported here is a continuation of our structural study (Acosta et al., 2008 ; Blanco et al., 2008 ; Gómez et al., 2008 ) of 2-substituted 1,4-epoxytetrahydro-1-benzazepines, which included two close analogues of the present compounds, namely compounds (V) and (VI). Compounds (I)–(IV) were all synthesized using our previously reported synthetic approach (Gómez Ayala et al., 2006 ), with the eventual aim of identifying structurally novel antiparasitic compounds which are active against Trypanosoma cruzi and Leishmania chagasi parasites (Palma et al., 2009 ).
Compounds (I)–(III) all crystallize as racemates, while the crystals of (IV) contain only a single enantiomer, viz. (2S,4R), in the crystal selected for data collection. Given the racemic nature of (I)–(III) and the absence of any reagent in the synthetic procedure likely to be able to provide enantiomeric selectivity, it seems probable that (IV) is, in fact, produced as a mixture of (2S,4R) and (2R,4S) enantiomers, but that it happens to crystallize as a conglomerate rather than as a racemate. In this connection, it is interesting to note that, while (V) crystallizes as a racemate in the space group Pna21, (VI) crystallizes as a single enantiomer in the space group P212121 (Gómez et al., 2008 ).
Compounds (I) and (II), which differ only in the presence of the 7-chloro substituent in (I), are isomorphous, with similar unit-cell dimensions and similar atomic coordinates for the corresponding atoms. However, they are not strictly isostructural (Acosta et al., 2009 ), as the direction-specific intermolecular interactions in the two crystal structures are subtly different, as discussed below. Although pairs of analogous compounds carrying, respectively, a methyl or a chloro substituent at equivalent sites are not infrequently isomorphous, no such relationship is evident for (III) and (V), which crystallize, respectively, in the space groups P21/n and Pna21 and which exhibit entirely different modes of supramolecular aggregation.
The ring-puckering parameters (Cremer & Pople, 1975 ) for (I)–(IV) are collected in Table 1 , along with those for (V) and (VI) (Gómez et al., 2008 ) for comparison. All six compounds exhibit very similar shapes for the fused heterocyclic ring system. The five-membered ring component in each of (I), (II), (V) and (VI) adopts a nearly perfect half-chair conformation, for which the idealized value of the puckering angle ϕ is (36k + 18)°, where k represents an integer; the conformations in (III) and (IV) are intermediate between half-chair and envelope forms, for which the idealized value of ϕ is 36k°. The six-membered ring components all adopt conformations closer to the half-chair form, for which the ideal values of the ring-puckering angles are θ = 50.8° and ϕ = (60k + 30)°, than to the envelope conformation, where the ideal values of the puckering angles are θ = 54.7° and ϕ = 60k°.
The supramolecular aggregation in (I)–(IV) is dominated by C—HO, C—HN and C—Hπ(arene) hydrogen bonds, augmented by aromatic π–π stacking interactions in (II) only. There are short intermolecular C—HF contacts in (I) and (II) and an intermolecular C—HCl contact in (IV). None of these contacts is likely to be of structural significance, firstly because the C—H bonds involved are of low acidity, and secondly because it has been well established that F and Cl atoms when bound to C atoms are extremely poor acceptors of hydrogen bonds, even from donors such as O or N (Aakeröy et al., 1999 ; Brammer et al., 2001 ; Howard et al., 1996 ; Thallapally & Nangia, 2001 ). Similarly, in the intermolecular C—HN contacts in (I) and (II), involving a C—H bond of low acidity, the HN distances are probably too long for these contacts to be of structural significance.
On this basis, the molecules of (I) are linked by just a single C—Hπ(arene) hydrogen bond (Table 2 ) to form a simple chain running parallel to the  direction (Fig. 2 ), with no direction-specific interactions between adjacent chains. Entirely analogous chains are formed in (II), but these chains are now linked into sheets by a π–π stacking interaction. The fused aryl rings in the molecules at (x, y, z) and (1 − x, 1 − y, −z) are strictly parallel, with an interplanar spacing of 3.426 (2) Å, a ring-centroid separation of 3.810 (2) Å and a ring-centroid offset of 1.667 (2) Å. The effect of this interaction is to link the chains parallel to  into a sheet lying parallel to (101) (Fig. 3 ).
The crystal structure of (III) also contains sheets, but these are built solely from two hydrogen bonds, one each of the C—HN and C—Hπ(arene) types (Table 2 ). The C—Hπ(arene) interaction links a pair of molecules into a cyclic centrosymmetric dimer centred at (, 1, 0), and this dimer can be regarded as the building block for the sheet formation. The C—HN hydrogen bond links the reference dimer centred at (, 1, 0) directly to four other such dimers, viz. those centred at (0, , −), (0, , −), (1, , ) and (1, , ), and propagation of these two interactions then generates a hydrogen-bonded sheet lying parallel to (10) (Fig. 4 ).
Four hydrogen bonds, one of the C—HO type and three of the C—Hπ(arene) type, combine to link the molecules of (IV) into a single three-dimensional framework. The formation of the framework is most readily analysed in terms of two independent two-dimensional substructures. The three hydrogen bonds involving atoms C4, C6 and C8 as the donors combine to generate a sheet lying parallel to (001) (Fig. 5 ), while the two hydrogen bonds having C4 and C23 as the donors combine to form a sheet parallel to (100) (Fig. 6 ). The combination of the (100) and (001) sheets is sufficient to generate a three-dimensional structure.
It is of interest briefly to compare the aggregation in (I)–(IV) with that in the related compounds (V) and (VI) (Gómez et al., 2008 ). In (V), the molecules are linked by a combination of C—HO and C—HN hydrogen bonds to form a chain of edge-fused (12) (Bernstein et al., 1995 ) rings, while in (VI), a combination of two C—HO hydrogen bonds and one C—Hπ(arene) hydrogen bond generates a three-dimensional framework structure. In the course of the present work, we have also investigated (VII), which crystallizes in the space group P with Z′ = 2, but we have been unable to refine this below R = 0.11. However, it is clear that the two independent molecules within the asymmetric unit are linked by one C—HO hydrogen bond and one C—HN hydrogen bond to form an (8) motif, but that there are no further direction-specific interactions between the molecules. Thus, despite the very close constitutional, configurational and conformational similarity between the molecules of compounds (I)–(VII), no two of these compounds exhibits the same pattern of supramolecular aggregation.
For the preparation of compounds (I)–(IV), sodium tungstate dihydrate, Na2WO4·2H2O (10 mol%), followed by 30% aqueous hydrogen peroxide solution (added dropwise, 0.30 mol) were added to a stirred solution of the appropriately substituted 2-allyl-N-benzylaniline (0.10 mol) in methanol (34 ml) for (I), (III) and (IV), or in a mixture of methanol (34 ml) and nitromethane (3.4 ml) for (II). The resulting mixtures were then stirred at ambient temperature for periods ranging from 30 to 100 h. Each mixture was filtered and the solvent removed under reduced pressure. Toluene (40 ml) for compounds (I), (III) and (IV) or ethyl acetate (40 ml) for (II) was added to the solid residue and the resulting solution was heated to ca 353 K for periods ranging from 6 to 8 h. After cooling each solution to ambient temperature, the solvent was removed under reduced pressure and the crude products were purified by chromatography on silica gel using heptane–ethyl acetate (compositions ranged from 90:1 to 60:1 v/v) as eluant. Crystallization from heptane gave colourless crystals suitable for single-crystal X-ray diffraction. For (I), m.p. 429–430 K, yield 50%; MS (70 eV) m/z (%): 323 (M +, 35Cl, 17), 306 (3), 294 (1), 280 (1), 164 (3), 138 (100), 125 (6), 111 (4). For (II), m.p. 433–434 K, yield 65%; MS (70 eV) m/z (%): 289 (M +, 35Cl, 32), 272 (7), 260 (1), 246 (3), 130 (4), 104 (100), 91 (10), 77 (10). For (III), m.p. 374–376 K, yield 45%; MS (70 eV) m/z (%): 265 (M +, 35), 248 (18), 222 (10), 207 (5), 146 (7), 132 (23), 118 (100), 103 (12), 91 (30), 77 (18), 65 (9), 51 (5). For (IV), m.p. 385–387 K, yield 63%; MS (70 eV) m/z (%): 271 (M +, 35Cl, 50), 254 (19), 242 (2), 228 (2), 130 (4), 104 (100), 91 (25), 77 (26).
All H atoms were located in difference maps and then treated as riding atoms in geometrically idealized positions, with C—H distances of 0.95 (aromatic), 0.98 (CH3), 0.99 (CH2) or 1.00 Å (aliphatic CH), and with U iso(H) = kU eq(C), where k = 1.5 for the methyl groups, which were permitted to rotate but not to tilt, and k = 1.2 for all other H atoms. For (IV), the absolute configuration of the molecules in the crystal selected for data collection was established as (2S,4R) by means of the Flack (1983 ) x parameter of 0.07 (6) and the Hooft y parameter (Hooft et al., 2008 ) of 0.03 (4). Accordingly, the configuration of the reference molecules in the racemic compounds (I)–(III) was set to be S at C2, and on this basis all three compounds have configuration R at C4 for the reference molecules, so that the overall configuration for each of (I)–(III) is (2SR,4RS).
For all compounds, data collection: COLLECT (Hooft, 1999 ); cell refinement: DIRAX/LSQ (Duisenberg et al., 2000 ); data reduction: EVALCCD (Duisenberg et al., 2003 ); program(s) used to solve structure: SIR2004 (Burla et al., 2005 ); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008 ); molecular graphics: PLATON (Spek, 2009 ); software used to prepare material for publication: SHELXL97 and PLATON.
Crystal structure: contains datablocks global, I, II, III, IV, VII. DOI: 10.1107/S0108270109030339/fa3200sup1.cif
Structure factors: contains datablocks I. DOI: 10.1107/S0108270109030339/fa3200Isup2.hkl
Structure factors: contains datablocks II. DOI: 10.1107/S0108270109030339/fa3200IIsup3.hkl
Structure factors: contains datablocks III. DOI: 10.1107/S0108270109030339/fa3200IIIsup4.hkl
The authors thank ‘Servicios Técnicos de Investigación of Universidad de Jaén’ and the staff for data collection. JC thanks the Consejería de Innovación, Ciencia y Empresa (Junta de Andalucía, Spain), the Universidad de Jaén (project reference UJA_07_16_33) and Ministerio de Ciencia e Innovación (project reference SAF2008-04685-C02-02) for financial support. AP, SLG and CMS thanks COLCIENCIAS for financial support (grant No. 1102-408-20563).
Supplementary data for this paper are available from the IUCr electronic archives (Reference: FA3200). Services for accessing these data are described at the back of the journal.