Comment

The photochemistry of phthalimides has been studied extensively and has been reviewed by Kanaoka (1978), Coyle (1984) and Oelgemöller & Griesbeck (2002). Schwack (1987) has reported the photo-induced para-cycloaddition of cyclohexene to N-trichloromethylthio-, N-methyl- and N-phenylphthalimides. Suau et al. (1989) have reported the ortho- and para-photocycloaddition of 3-methoxy-N-methylphthalimide to 1-hexene and Kubo et al. (1989) have reported analogous ortho- and para-cycloadditions of N-methylphthalimide to allyltrimethylsilane. In each case, the para-cycloaddition products are structurally analogous to the title compound, (I). However, the structures were only elucidated by spectroscopic means and lack stereochemical certainty. The determination of the structure of (I) presented here was undertaken in the context of a study of the photochemistry of N-benzoylphthalimide but is clearly of significance in relation to the analogous compounds.

The molecule of (I) is shown in Fig. 1. Selected bond lengths and angles are given in Table 1. The bond lengths, along with those of the phenyl group R1 defined by C16-C21 in the range 1.361 (6)-1.389 (5) Å, are not unusual excepting, perhaps, the C2-C3 and C6-C7 bond lengths of 1.493 (5) and 1.481 (5) Å, respectively. Likewise, with the sole exception of the angle C9-C10-C14 of 134.1 (3)°, the bond angles, including the internal angles of the phenyl group in the range 117.9 (3)-121.1 (4)°, are as expected. The cyclohexane ring, R3, defined by C2-C7, adopts the chair conformation, with puckering parameters (Cremer & Pople, 1975) Q = 0.564 (4) Å, = 168.5 (5) and = 151 (2)°. The dihedral angle between the least-squares planes of phenyl group R1 (r.m.s. displacement = 0.0006 Å) and five-membered ring R2, defined by C1/C10/C13-C14/N1 (r.m.s. displacement = 0.0143 Å) is 61.97 (15)°. Atom O3 is displaced from the least-squares planes of R1 and R2 by 0.145 (7) and 1.134 (6) Å, respectively. The packing of the molecules of (I) creates layers parallel to (02) (Fig. 2) in such a way as to generate the first two C-H interactions given in Table 2 (shown as dashed lines in Fig. 2). The only contact between the layers, other than van der Waals interactions, is the third, longer, C-H contact given in Table 2.

The racemic nature of (I), a prerequisite for the refinement of the structure in the centrosymmetric space group P2₁/c, is a natural consequence of the manner in which the compound has been formed from achiral reactants. In principle, given that the unsymmetrical 1,4-addition across the aromatic ring must of necessity be cis, there are four possible racemic products, two involving trans ring junctions at C2-C7 and two involving cis junctions at C2-C7. Formation of the single unsymmetrical product, (I), suggests a favoured approach by the cyclohexene to the excited phthalimide, possibly involving minimization of steric interactions between the N-benzoylimide and cyclohexene rings in the transition state. The stereochemistry at the C2-C7 ring junction is the outcome of overall trans addition across the cyclohexene double bond.

Figure 1 A view of (I). Displacement ellipsoids are drawn at the 30% probability level and H atoms are shown as small circles of arbitrary radii.

Figure 2 A layer of molecules of (I). Displacement ellipsoids are drawn at the 20% probability level and H atoms involved in C-H.. contacts (dashed lines) are shown as small circles of arbitrary radii. [Symmetry codes (i) 1 - x, 1 - y, 1 - z; (ii) 1 + x,  - y,  +  z; (iv) 2 - x, y - ,  - z; (v) x - 1,  - y, z - ; (vi) -x, y - ,  - z.]

Experimental

Compound (I) was one of the products of irradiation for 40 h of N-benzoylphthalimide (2.90 g, 11.5 mmol) and cyclohexene (19.60 g, 239.0 mmol) in dichloromethane (300 ml) by a 400 W medium-pressure mercury vapour lamp fitted with a Pyrex filter. After removal of solvents under vacuum three products (previously detected by thin-layer chromatography) were isolated by means of a Chromatotron and a 4 mm silica plate with a mixture of dichloromethane and light petroleum (b.p. 313-333 K) (2:98 increased stepwise to 60:40) as eluant to yield: (i) recovered N-benzoylphthalimide (2.75 g); (ii) a mixture of minor products as a colourless oil (12 mg); (iii) compound (I), a white crystalline solid [160 mg, 80%; m.p. 411-413 K (from chloroform/light petroleum, b.p. 363-373 K)], _max (MeCN): 251 ( 20,208 dm³ mol^-1 cm^-1); _max 2929 (aliphatic CH), 1717 and 1694 (C=O), 1297 and 1252 cm^-1; _H (270 MHz, CDCl₃): 7.89-7.47 (5H, m, ArH), 7.10 (1H, d, J 6.0 Hz, vinylic H), 6.82 (1H, d of d, J 6.0 Hz, J 7.0 Hz, vinylic H), 6.16 (1H, d, J 7.0 Hz, vinylic H), 3.76 (1H, t, J 6.0 Hz), 2.10-1.11 (10H, m, cyclohexane derived moiety); _C (67.8 MHz, CDCl₃): 173.0, 167.1, 162.3 (carbonyl C), 143.3, 141.7, 137.3, 134.8, 131.9, 130.4, 128.8, 123.5 (aromatic and vinylic C), 56.2, 52.5, 50.7, 45.7, 32.7, 30.2, 27.7 and 27.4 (aliphatic C); analysis found: C 75.3, H 5.8, N 3.9%; C₂₁H₁₉NO₃ requires: C 75.7, H 5.8, N 4.2%; m/e: 333 (1), 265 (47), 264 (31), 252 (56), 105 (100), 77 (63) and 67 (45%).

Crystal data

C21H19NO3 Mr = 333.37 Monoclinic, P21/c a = 8.111 (3) Å b = 12.999 (7) Å c = 16.256 (5) Å = 100.76 (3)° V = 1683.8 (12) Å3 Z = 4 Dx = 1.315 Mg m-3 Mo K radiation Cell parameters from 14 reflections = 11.0-13.0° = 0.09 mm-1 T = 298 (2) K Block, colourless 0.60 × 0.40 × 0.26 mm

Data collection

Nicolet P3 four-circle diffractometer -2 scans Absorption correction: none 3882 measured reflections 3882 independent reflections 1880 reflections with I > 2(I) max = 30.1° h = 0 11 k = 0 18 l = -22 22 2 standard reflections every 50 reflections intensity decay: none

Refinement

Refinement on F2 R[F2 > 2(F2)] = 0.088 wR(F2) = 0.192 S = 1.03 3882 reflections 226 parameters H-atom parameters constrained w = 1/[2(Fo2) + (0.0735P)2] where P = (Fo2 + 2Fc2)/3 (/)max < 0.001 max = 0.32 e Å-3 min = -0.23 e Å-3

Table 1 Selected geometric parameters (Å, °) N1-C13 1.408 (4) N1-C14 1.425 (4) N1-C15 1.444 (4) O1-C13 1.203 (4) O2-C14 1.200 (4) O3-C15 1.194 (4) C1-C11 1.500 (5) C1-C13 1.504 (5) C1-C10 1.504 (4) C1-C2 1.571 (4) C2-C3 1.493 (5) C2-C7 1.514 (5) C6-C7 1.481 (5) C7-C8 1.584 (5) C8-C9 1.506 (5) C8-C12 1.519 (5) C9-C10 1.330 (4) C10-C14 1.462 (5) C11-C12 1.323 (5) C15-C16 1.473 (5) C13-N1-C14 113.0 (3) C13-N1-C15 121.3 (3) C14-N1-C15 125.0 (3) C11-C1-C13 118.0 (3) C11-C1-C10 108.2 (3) C13-C1-C10 103.4 (3) C11-C1-C2 109.4 (3) C13-C1-C2 114.3 (3) C10-C1-C2 102.0 (3) C3-C2-C7 110.4 (3) C3-C2-C1 124.6 (3) C7-C2-C1 107.4 (3) C6-C7-C2 110.8 (3) C6-C7-C8 122.4 (3) C2-C7-C8 109.0 (3) C9-C8-C12 108.3 (3) C9-C8-C7 109.1 (3) C12-C8-C7 100.8 (3) C10-C9-C8 112.1 (3) C9-C10-C14 134.1 (3) C9-C10-C1 115.4 (3) C14-C10-C1 110.3 (3) C12-C11-C1 113.2 (3) C11-C12-C8 114.5 (3) O1-C13-N1 124.0 (3) O1-C13-C1 127.9 (3) N1-C13-C1 108.1 (3) O2-C14-N1 123.9 (3) O2-C14-C10 130.9 (3) N1-C14-C10 105.1 (3) O3-C15-N1 118.4 (3) O3-C15-C16 123.8 (3) N1-C15-C16 117.7 (3) C11-C1-C2-C3 -86.5 (4) C13-C1-C2-C3 48.3 (5) C10-C1-C2-C3 159.1 (4) C11-C1-C2-C7 44.9 (4) C13-C1-C2-C7 179.7 (3) C10-C1-C2-C7 -69.5 (3) C1-C2-C3-C4 -171.5 (3) C5-C6-C7-C8 -170.5 (4) C3-C2-C7-C6 -65.8 (4) C1-C2-C7-C6 155.5 (3) C3-C2-C7-C8 156.7 (3) C1-C2-C7-C8 18.0 (4) C6-C7-C8-C9 -87.4 (4) C2-C7-C8-C9 44.1 (4) C6-C7-C8-C12 158.8 (4) C2-C7-C8-C12 -69.7 (4) C8-C9-C10-C14 178.1 (3) C8-C9-C10-C1 3.3 (4) C13-C1-C10-C9 179.5 (3) C13-C1-C10-C14 3.4 (3) C1-C11-C12-C8 3.4 (4) C9-C8-C12-C11 -54.9 (4) C7-C8-C12-C11 59.5 (4)

Table 2 Geometry (Å,°) of C-H contacts in (I) C-HCga C-H HCg Hperpb c C-HCg CCg C6-H6ACg1i 0.97 2.80 2.69 16 149 3.68 C6-H6BCg1ii 0.97 3.03 2.93 15 135 3.79 C4-H4BCg2iii 0.97 3.34 3.28 11 120 3.92 Notes: (a) Cg1 and Cg2 are the centroids of the rings defined by C16-C21 and C1/C10/C13-C14/N1, respectively; (b) Hperp is the perpendicular distance of the H atom from the mean plane of the ring; (c) is the angle at hydrogen between Hperp and HCg. Symmetry codes (i) 1-x,1-y,1-z; (ii) ; (iii) 1+ x,y,z.

The incompleteness (84.9% complete for _full = 25°) of the mid-1980s data set upon which this refinement is based is due to the suppression, during data reduction and contrary to current practice, of reflections with intensities measured as negative. As a consequence, the omissions are scattered throughout the data set although they are more prevalent at high . In the final stages of refinement, H atoms were introduced in calculated positions with C-H set at 0.93, 0.97 and 0.98 Å for aryl/alkene, methylene and tertiary H atoms, respectively, and refined with a riding model, with U_iso(H) = 1.2U_eq(C) in all cases.

Data collection: Nicolet P3 Software (Nicolet, 1980); cell refinement: Nicolet P3 Software; data reduction: RDNIC (Howie, 1980); program(s) used to solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: ORTEP-3 for Windows (Farrugia, 1997); software used to prepare material for publication: SHELXL97 and PLATON (Spek, 2003).