2-(2-Methylbenzoyl)benzoic acid: catemeric hydrogen bonding in a γ-keto acid1

The crystal structure of the title compound, C15H12O3, displays catemeric aggregation involving O—H⋯O hydrogen bonds progressing from the carboxyl group of one molecule to the ketone O atom of another glide-related neighbor. The molecule is twisted, with the toluene 80.61 (3)° out of plane with respect to the phenyl group of the benzoic acid. The acid group makes a dihedral angle of 13.79 (14)° with the attached phenyl ring. The molecules are achiral, but the space group glide planes create alternating conformational chirality in the chain units. The four hydrogen-bonding chains progress along [001] in an A—A—B—B pattern (right-to-left versus left-to-right), and are related to each other by the center of symmetry at (0.5, 0.5, 0.5) in the chosen cell. There is one close contact (2.54 Å) between a phenyl H atom and the acid carbonyl from a symmetry-related molecule.

The crystal structure of the title compound, C 15 H 12 O 3 , displays catemeric aggregation involving O-HÁ Á ÁO hydrogen bonds progressing from the carboxyl group of one molecule to the ketone O atom of another glide-related neighbor. The molecule is twisted, with the toluene 80.61 (3) out of plane with respect to the phenyl group of the benzoic acid. The acid group makes a dihedral angle of 13.79 (14) with the attached phenyl ring. The molecules are achiral, but the space group glide planes create alternating conformational chirality in the chain units. The four hydrogen-bonding chains progress along [001] in an A-A-B-B pattern (right-to-left versus left-toright), and are related to each other by the center of symmetry at (0.5, 0.5, 0.5) in the chosen cell. There is one close contact (2.54 Å ) between a phenyl H atom and the acid carbonyl from a symmetry-related molecule.

Comment
This carboxyl-to-ketone catemer (I) is an achiral molecule which generates heterochiral chains due to the glide-related chain members. Normally, these types of catemers involve homochiral chains having either a screw or translational internal relationship (Hickmott et al., 1985;Abell et al., 1991;Kashyap et al., 1995). Rarer are heterochiral chains resulting from glide-plane symmetry, which is the case here (Watson et al., 1990;Barcon et al., 1998Barcon et al., , 2002Thompson et al., 1998). The structures of 2-(4-methylbenzoyl)benzoic acid and its hydrate have been published: HOFGAK (Degen & Bolte, 1999) is the p-toluene analog of (I) and MIXTOD (Song, et al., 2008) is the hydrated p-toluene analog [Cambridge Structural Database (CSD, Version 5.28, update of Nov., 2006;Allen, 2002)]. Both of these molecules crystallize as centrosymmetric dimers of the acid groups (Borthwick, 1980), but the hydrated version has two water molecules inserted between the acid dimers. In HOFGAK, the dihedral angle between the toluene and the benzoic acid = 89.8°, and the acid is coplanar [0.00°] with the phenyl ring. In MIXTOD, these same angles are 69.5 and 25.8°, respectively. In (I), the toluene and the ketone are essentially coplanar [dihedral angle = 1.91 (7)°], but in HOFGAK and MIXTOD, this same angle is 20.6 and 10.8°, respectively. Centrosymmetrically-related H-bonded chains can be seen by comparing the A molecule in Fig. 2a with the D molecule in Fig. 2b; the B and C molecules are also centrosymmetrically related.
We characterize the geometry of H bonding to carbonyls using a combination of H···O=C angle and H···O=C-C torsion angle. These describe the approach of the acid H atom to the carbonyl O in terms of its deviation from, respectively, C=O axiality (ideal = 120°) and planarity with the carbonyl (ideal = 0°). In (I), the values for these two angles are 116.5 (5) and 10.0 (6)°.

Experimental
1.20 g of Mg and 1 ml of 1-bromo-2-methylbenzene were added to 17 ml of dry diethyl ether, followed by the gradual addition of 4.625 ml of 1-bromo-2-methylbenzene. The resulting Grignard reagent was added over a period of 30 min to a stirred suspension of 6.2 g of phthalic anhydride in 35 ml of dry benzene and 15 ml of dry diethyl ether. The pale orange mixture was stirred for 30 minutes, refluxed for 10 minutes, allowed to stir overnight and refluxed for a further 10 minutes. 7 ml of HCl was added along with a small quantity of ice water. The resulting solution was distilled to dryness.
The glassy product was then dissolved in aqueous KOH and filtered through Celite. The filtrate was then acidified with HCl and extracted with ether into several fractions. The fractions not yielding crystals melting between 100-110°C were discarded, and the remaining fractions were combined. These were then passed through an alumina column and the solvent allowed to concentrate, yielding pale yellow crystals. These were then recrystallized from 8.6 ml of methanol and 5 mL of water. The resulting fine needle-like crystals were then recrystallized again from pure ethanol at room temperature, yielding colourless block crystals. (See Newman & McCleary, 1941).

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 1. All H atoms for (I) were found in electron density difference maps. The hydroxyl H was refined. The methyl H atoms were put in ideally staggered positions with C-H distances of 0.98 Å and U iso (H) = 1.5U eq (C). The aromatic Hs were placed in geometrically idealized positions and constrained to ride on their parent C atoms with C-H distance of 0.95 Å and U iso (H) = 1.2U eq (C).

Figure 1
A view of the asymmetric unit of (I) with its numbering scheme. Displacement ellipsoids are drawn at a 40% probability level for non-H atoms.    where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.24 e Å −3 Δρ min = −0.17 e Å −3 Extinction correction: SHELXTL (Sheldrick, 2008b), Fc * =kFc[1+0.001xFc 2 λ 3 /sin(2θ)] -1/4 Extinction coefficient: 0.00145 (14) Special details Experimental. 'crystal mounted on a Cryoloop using Paratone-N′ 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 F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.