2-Phenoxybenzoic acid at room temperature

Benzoic acid is a compound that has an elegant simplicity to its molecular structure, but its derivatives display an enormous complexity and diversity of molecular structures. The latest version of the Cambridge Structural Database (CSD, Version 5.26; Allen, 2002) contains 1883 structures with a benzoic acid derivative existing in the crystal structure as an isolated molecule; this does not include structures in which the molecules are either deprotonated or coordinated to metal ions. By contrast, the simple and readily available title compound, (I), is only observed in four crystal structures in the CSD, and in all of these it serves as a ligand. The 3and 4-phenoxybenzoic acid structures are observed even less frequently, with zero and one structures of these compounds, respectively. Possibly the most closely related structure available in the CSD is that of 2-(2-carboxyphenoxy)benzoic acid (CSD refcode MIGPAT; Field & Venkataraman, 2002), which differs only by the presence of an extra carboxylic acid group on the second benzene ring.

In the crystal structure of the title compound, C 13 H 10 O 3 , the molecules form classical hydrogen-bonded carboxylic acid dimers [OÁ Á ÁO = 2.651 (2) Å ]. These dimers are linked by C-HÁ Á Á andinteractions to give a three-dimensional network.

Comment
Benzoic acid is a compound that has an elegant simplicity to its molecular structure, but its derivatives display an enormous complexity and diversity of molecular structures. The latest version of the Cambridge Structural Database (CSD, Version 5.26; Allen, 2002) contains 1883 structures with a benzoic acid derivative existing in the crystal structure as an isolated molecule; this does not include structures in which the molecules are either deprotonated or coordinated to metal ions. By contrast, the simple and readily available title compound, (I), is only observed in four crystal structures in the CSD, and in all of these it serves as a ligand. The 3-and 4-phenoxybenzoic acid structures are observed even less frequently, with zero and one structures of these compounds, respectively. Possibly the most closely related structure available in the CSD is that of 2-(2-carboxyphenoxy)benzoic acid (CSD refcode MIGPAT; Field & Venkataraman, 2002), which differs only by the presence of an extra carboxylic acid group on the second benzene ring.
The molecular geometry observed in the structure of (I) is mostly unremarkable, with the principal features of note being the prolate displacement ellipsoid of atom O10, which is consistent with a large vibration perpendicular to the plane of the benzoic acid fragment (Fig. 1). This motion is not obviously propagated in the second benzene ring; in this portion, the displacement ellipsoids are surprisingly close to spherical, although large. These observations are most likely due to the combination of three movements: a typical in-plane rotational movement around the ring, the translational movement observed for O10 in the plane of this ring and perpendicular to the O10-C11 bond vector, and a rotational movement around the O10-C11 bond vector. The average C-C bond length in this ring is slightly short, at 1.36 Å ; this bond shortening can also be attributed to the effect of large thermal libration. The normals to the planes of the two benzene rings are nearly perpendicular, at 89.8 (2) .
The molecules of (I) assemble to form a classical hydrogenbonded dimer, in which the C7-O9 and C7-O8 bond lengths in the carboxylic acid group of 1.223 (2) and 1.3015 (18) Å , respectively, indicate a well ordered hydrogen bond. This is supported by the lack of H-atom disorder observed in the Fourier difference map (calculated with the program MAPVIEW, part of the WinGX suite; Farrugia, 1999) through the dimer group (Fig. 2). The single crystallographically unique hydrogen bond, viz. O8-H1Á Á ÁO9 i [symmetry code: (i) 3 À x, 1 À y, Àz], exhibits a typical OÁ Á ÁO separation for benzoic acid dimers of 2.651 (2) Å . The remainder of the contacts lie outside the sum of the van der Waals radii of the two atoms involved, but these very weak interactions can still be used to describe the remainder of the structure. The dimers assemble into extended ribbons through C-HÁ Á Á interactions of 3.658 Å for C4-H4Á Á ÁC12 ii [symmetry code: (ii) x, y À 1, z] (Fig. 3a), and these ribbons form stacks defined by a contact of 3.446 Å between atoms C13 and C16 iii [symmetry code: (iii) x À 1, y, z] (Fig. 3b). The stacks pack together with C-HÁ Á Á interactions of 3.697 Å for C12-H12Á Á ÁC5 iv [symmetry code: (iv) x À 1, y + 1, z] (Fig. 4).
The most striking difference between the molecular structure presented here and that of MIGPAT (Field & Venkata- Fourier difference map section through the carboxylic acid dimer plane defined by atoms C7, O8, O9, C7 i , O8 i and O9 i [symmetry code: (i) 3 À x, 1 À y, Àz]. There is clearly only a single peak associated with each hydrogen bond, corresponding to an ordered H atom.

Figure 1
A drawing of the molecule of (I), showing the atomic numbering scheme. Ellipsoids for non-H atoms are shown at the 30% probability level. All H atoms take their number from the parent C atom, except for H1.

Figure 3
Packing plots of (I), illustrating the principal contacts in the structure. (top) The hydrogen-bond dimers link together into ribbons via C-HÁ Á Á contacts (shown in yellow). (bottom) These ribbons stack due tointeractions (shown in pale green). raman, 2002) is the geometry of the carboxylic acid group. In the title compound, it is clear from the bond lengths that the C O double bond is C7 O9, involving the O atom closest to the ether group. By contrast, the shorter C-O bond in MIGPAT is that further from the ether O atom, although the difference between the two bond lengths is much less than we report here. As the two chemically different carboxylic acids in MIGPAT are crystallographically identical, it is possible that there is some correlated structural disorder between the C-O and C O bonds; this might explain the very similar C-O bond lengths in MIGPAT.

D-HÁ
H atoms were positioned geometrically and refined as riding groups, with C-H = 1.0 Å and U iso (H) = 1.2U eq (C), except for atom H1, which was located in a Fourier difference map and refined with an O-H distance restraint of 0.90 (5) Å and a fixed U iso (H) = 0.05 Å 2 .
Data collection: APEX2 (Bruker, 2005); cell refinement: APEX2; data reduction: APEX2; program(s) used to solve structure: SIR92 (Altomare et al., 1994); program(s) used to refine structure: CRYS-TALS (Betteridge et al., 2003); molecular graphics: ORTEP3 for Windows (Farrugia, 1997) and MERCURY (Bruno et al., 2002); software used to prepare material for publication: CRYSTALS. This paper is the result of an optional undergraduate class project entitled 'Frontiers of Crystallography', designed to show some of the sort of research that can be undertaken in crystallography. The data collection, structure solution, refinement and post-refinement analysis of the unknown title structure were all undertaken in parallel by the undergraduate students, who are all co-authors, and the collated information has resulted in this paper.