Redetermination of 2,6-dimethoxybenzoic acid

The crystal structure of the title compound, C9H10O4, was first reported by Swaminathan, Vimala & Lotter [Acta Cryst. (1976), B32, 1897–1900]. It has been re-examined, improving the precision of the derived geometric parameters. The asymmetric unit comprises a non-planar independent molecule, as the methoxy substituents force the carboxy group to be twisted away from the plane of the aromatic ring by 56.12 (9)°. Due to the antiplanar conformation adopted by the OH group, the molecular components do not form the conventional dimeric units, but are associated in the crystal in chains stabilized by linear O—H⋯O hydrogen bonds, involving the OH groups and the carbonyl O atoms, which form C(3) motifs.

The crystal structure of the title compound, C 9 H 10 O 4 , was first reported by Swaminathan, Vimala & Lotter [Acta Cryst. (1976), B32, 1897-1900. It has been re-examined, improving the precision of the derived geometric parameters. The asymmetric unit comprises a non-planar independent molecule, as the methoxy substituents force the carboxy group to be twisted away from the plane of the aromatic ring by 56.12 (9) . Due to the antiplanar conformation adopted by the OH group, the molecular components do not form the conventional dimeric units, but are associated in the crystal in chains stabilized by linear O-HÁ Á ÁO hydrogen bonds, involving the OH groups and the carbonyl O atoms, which form C(3) motifs.
Redetermination of 2,6-dimethoxybenzoic acid G. Portalone* Comment 2,6-Dimethoxybenzoic acid was determined some 30 years ago (Swaminathan et al., 1976) but the final refinement was carried only to R=0.15 and no atomic coordinates were provided. Subsequently, a new X-ray structure determination was reported (Bryan & White, 1982). In this study, 775 unique reflections were collected at ambient temperature on an automatic diffractometer using Cu Kα radiation. Data were corrected for Lp effects, but not for absorption [µ(Cu Kα)= 94 mm -1 ]. 708 reflections having values significantly above background were used in the block-diagonal least-squares refinement. The final calculations, carried out on a fairly small data set, led to R = 0.035 for 158 refined parameters, a data-to-parameter ratio of 4.5, the maximum shift-to-error in the final cycle being equal to 1/4, and standard deviations of 0.005Å in C-C bond lengths and 0.4° in bond angles.
The asymmetric unit of (I) comprises a non-planar independent molecule, as the methoxy substituents force the carboxy group to be twisted away from the plane of the aromatic ring by 56.12 (9)° (Fig. 1). The values of bond lengths and bond angles are consistent with that reported in the previous determination (Bryan & White, 1982) with the exception of the geometrical parameters of the carboxy group. Analysis of the crystal packing of (I), (Table 1, Fig. 2) shows that the molecular components do not form the conventional dimeric units observed in monocarboylic acids (Leiserowitz, 1976). The structure is stabilized by intermolecular O-H···O interactions of descriptor C(3) (Etter et al., 1990;Bernstein et al., 1995;Motherwell et al., 1999) (Table 1 having the OH group in the unusual antiplanar conformation (Gopalakrishna & Cartz, 1972;Byriel et al., 1991;Chen et al., 2007). For these compounds, the antiplanar conformation is favoured by the formation of intramolecular hydrogen bonding.
Experimental 2,6-Dimethoxybenzoic acid (0.1 mmol, Sigma Aldrich at 99% purity) was dissolved in ethanol (95%, 9 mL) and gently heated under reflux for 1 h. After cooling the solution to an ambient temperature, crystals suitable for single-crystal X-ray diffraction were grown by slow evaporation of the solvent after two days.

Refinement
All H atoms were found in a difference map and then treated as riding atoms, with C-H = 0.97 (phenyl) and 0.96-0.98 Å (methyl); their U iso values were kept equal to 1.2U eq (C, phenyl). and to 1.5U eq (C, methyl). The remaining H atom of the carboxy group was freely refined. In the absence of significant anomalous scattering in this light-atom study, Friedel pairs were merged.  Fig. 1. The molecular structure of (I) showing the atom-labelling scheme. Displacements ellipsoids are at the 50% probability level.   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 Rfactors(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.