2,3-O-Isopropylidene-3-C-phenylerythrofuranose

The title compound, C13H16O, comprises two fused five-membered rings. Each ring has an envelope conformation, with the ether O atom in the furanose ring, and the CMe2 atom in the acetonide ring as the flap atoms. In the crystal, centrosymmetrically related molecules associate via hydroxy–ether O—H⋯O hydrogen bonds and the resulting dimers are linked into a supramolecular chain with a flattened topology via C—H⋯Ohydroxy contacts, and aligned in the a-axis direction.

The title compound, C 13 H 16 O, comprises two fused fivemembered rings. Each ring has an envelope conformation, with the ether O atom in the furanose ring, and the CMe 2 atom in the acetonide ring as the flap atoms. In the crystal, centrosymmetrically related molecules associate via hydroxyether O-HÁ Á ÁO hydrogen bonds and the resulting dimers are linked into a supramolecular chain with a flattened topology via C-HÁ Á ÁO hydroxy contacts, and aligned in the a-axis direction.

Comment
The dihydroxyation of monocyclic and bicyclic 1,2-dioxines has provided a new route for the stereoselective synthesis of a diverse range of carbohydrates and related compounds (Pedersen et al., 2009;Robinson et al., 2006;Robinson et al., 2009;Valente et al., 2009). During the course of these studies, the title compound, (I), was obtained by the Co(II)-mediated ring-opening of the precursor 1,2-dioxane, post dihydroxyation. The reactions of Co(II) complexes with endoperoxides have been well documented (Boyd et al., 1980;Sutbeyaz et al., 1988;Greatrex et al., 2003;Greatrex & Taylor, 2005).
The molecular structure of (I), Fig. 1, comprises two fused five-membered rings linked at the C3-C4 bond. Each of the five-membered rings adopts an envelope conformation, on atom O1 for the furanose (O1, C2-C5) ring, and on atom C6 for the acetonide (O3, O4, C3, C4, C6) ring. When viewed down the C3-C4 axis, the O1 atom lies above the plane through the four remaining atoms, away from the phenyl substituent and the C6 atom lies below the plane, being orientated in the same direction as the phenyl ring. In the crystal structure centrosymmetrically related pairs of molecules associate via O-H···O hydrogen bonds to form an eight-membered {···OCOH} 2 synthon, Table 1 and Fig. 2. The dimers are linked into a supramolecular chain via C-H···O contacts and ten-membered {···OH···OCH} 2 synthons, Table 1. The resulting chain comprising alternating eight-and ten-membered synthons has a flattened topology, Fig. 2, and is aligned along the a axis.

Experimental
For full synthetic procedures and characterization data see Pedersen et al. (2009) andRobinson et al. (2009). To a stirred solution of Co(salen) 2 (17 mg, 0.05 mmol) in THF (5 ml) at ambient temperature was added (3aR,7aS)-3a-phenyl-tetrahydro-2,2-dimethyl-[1,3]dioxolo[4,5-d][1,2]dioxine (501 mg, 2.12 mmol), and the reaction left to stir until complete by TLC (~16 h). All volatiles were removed in vacuo giving a crude mixture of regioisomers in a 40:60 ratio. The isomers were fully separated by flash chromatography giving a combined total yield of 496 mg (99%). Compound (I) was isolated as a colourless solid (198 mg), and the pure material was recrystallized from a slowly evaporating 1:1 mixture of dichloromethane/heptane to give colourless prisms, m. pt. 424-425 K. The compound was found to exist solely in its cyclic hemi-acetal form(s) both as a solid indicated by IR (absence of carbonyl signal), and in CDCl 3 solution which revealed a 90:10 ratio of anomers.

Refinement
Carbon-bound H-atoms were placed in calculated positions (C-H 0.95-1.00 Å) and were included in the refinement in the riding model approximation with U iso (H) set to 1.2-1.5U eq (C). The O-bound H-atom was located in a difference Fourier map and was refined with an O-H restraint of 0.840±0.001 Å, and with U iso (H) = 1.5U eq (O).
supplementary materials sup-2 Figures   Fig. 1. Molecular structure of (I) showing atom-labelling scheme and displacement ellipsoids at the 35% 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.