(+)-(1S,5R,10S)-11,11-Dimethyl-4-oxatricyclo[8.4.0.01,5]tetradecane-3,12-dione

The title compound, C15H22O3, was prepared via amino-acid-promoted Robinson annulation followed by tandem Pd/C-mediated hydrogenation and oxidative cyclization. This product was instrumental in determining the feasibility of a stereocontrolled hydrogenation in which the directing hydroxyl group is adjacent to the 6–7-ring network and its olefinic component. The asymmetric unit consists of a single molecule with normal geometric parameters. The absolute configuration was assigned based on the known enantiomeric prescursor. Intermolecular C—H⋯O interactions link each molecule with four neighboring molecules.

The title compound, C 15 H 22 O 3 , was prepared via amino-acidpromoted Robinson annulation followed by tandem Pd/Cmediated hydrogenation and oxidative cyclization. This product was instrumental in determining the feasibility of a stereocontrolled hydrogenation in which the directing hydroxyl group is adjacent to the 6-7-ring network and its olefinic component. The asymmetric unit consists of a single molecule with normal geometric parameters. The absolute configuration was assigned based on the known enantiomeric prescursor. Intermolecular C-HÁ Á ÁO interactions link each molecule with four neighboring molecules.

Experimental
Crystal data

Comment
The capability of L-amino acids to promote the enantioselective intramolecular aldolization of prochiral substrates (I) and (II) has projected the related Hajos-Parrish (III) and Wieland-Miescher ketones (IV) into favored positions as starting materials for targeted synthesis (see Fig. 1). Most notably, the selection of these particular synthons has resulted in the rather direct preparation of numerous terpenoids and steroids (Inomata et al., 2005).
More recently, the discovery has been made that comparable asymmetric Robinson annulation involving (V) and (VII) is accompanied by a striking crossover in enantioselectivity (Nagamine et al., 2007). When the 1,3-cyclohexanedione (V) is involved, the S enantiomer defined by (VI) continues to be formed predominantly. On the other hand, progression to the seven-membered triketone homolog (VII) results in the kinetically favored generation of the R product (VIII) (see Fig. 2).
As a result, our desire to involve 6-7 fused bicyclic systems of type (VIII) as synthetic intermediates now mandates that each ensuing step involving the introduction of a new stereogenic center be carefully evaluated. The present report details such an example.
The hindered nature of the double bond in (IX) causes this intermediate to be unreactive to a broad range of hydrogenation conditions. However, recourse to the use of 10% palladium on carbon in methanol at 550 psi leads to saturation of the olefinic linkage with concomitant loss of the acetonide functionality. The chromatographically inseparable nature of (X) and (XI) was overcome by efficient (93% overall) two-step oxidative cyclization to generate (XII) and (XIII), the ratio of which was shown by NMR analysis to be 56:44 (see Fig. 3). Identification of the less dominant, highly-crystalline product as the trans-fused isomer (XIII) was realised by X-ray crystallography, as shown in Fig. 4. The level of production of (XIII) provides suggestive indication that hydroxyl-directed hydrogenation is unable to operate at the heightened levels customarily observed (Brown, 1987;Crabtree & Davis, 1986;Peng et al., 2004;Stork & Kahne, 1983).
The bond distances in (XIII) are in agreement with those that were selected in the critical evaluation of structures in the Cambridge data base (Allen et al., 1987). The presence of intermolecular CH-O hydrogen bonds is indicated by short H to O distances (2.58Å to 2.70 Å) between the observed O1 and O2 positions and calculated H positions (Taylor & Kennard, 1982;Steiner & Saenger, 1992;Desiraju & Steiner, 1999). Each molecule H-bonds with four adjacent molecules, as shown in Fig. 5, with contact distances and angles given in the table of hydrogen bonds.

Experimental
A suspension of (IX) (20 mg) and 10% Pd-C (2 mg) in methanol (1 ml) was pressurized to 550 psi of hydrogen gas in an autoclave and stirred for 15 h at rt. After filtration through Celite and solvent evaporation, the residue was chromatographed on silica gel to afford 11 mg of an inseparable mixture of (X) and (XI). This mixture was dissolved in THF (0.5 ml) and saturated NaHCO 3 solution (0.5 ml), cooled to 0 °C, treated with NaIO 4 (48 mg), and stirred in the cold for 3 h. The mixture was extracted with ethyl acetate and the combined organic layers were dried and evaporated. The residue was dissolved in benzene (1 ml), and Ag 2 CO 3 on Celite (48 mg) was introduced. After being heated at reflux for 2 h, the mixture was filtered supplementary materials sup-2 through a Celite pad and the filtrate was evaporated under reduced pressure. Chromatographic purification was performed on silica gel to afford (XII) as a colorless oil (5 mg) and (XIII) as colorless crystals (4 mg) displaying a melting point of 155.5-156 °C after recrystallization from ethyl acetate.

Refinement
The intensity statistics are non-centrosymmetric and the systematic absences restrict the space group possibilities to P6 1 or P6 5 . The correct enantiomer was chosen based on the known chiral centers at atoms C1 and C5. For the methyl groups, the hydrogen atoms were added at calculated positions using a riding model with C-H = 0.98Å and U iso (H)=1.5*U eq (C). The torsion angle, which defines the orientation of the methyl group about the C-C bond, was refined. The remaining hydrogen atoms were included at calculated positions using a riding model with C-H = 0.99Å and U iso (H)=1.2*U eq (C).

Special details
Experimental. The data collection crystal was a clear, colorless chunk, which was cut from a cluster of crystals. Initial examination of the diffraction pattern on a Nonius Kappa CCD diffractometer indicated a trigonal or hexagonal crystal system. All work was done at 150 K using an Oxford Cryosystems Cryostream Cooler. Omega scans with a frame width of 1.0 degree were used for data collection. Data integration was done with DENZO (Otwinowski & Minor, 1997) and scaling and merging of the data was done with SCALE-PACK (Otwinowski & Minor, 1997).
The Laue group was determined to be 6/m by XPREP (Bruker Nonius, 2003). The intensity statistics are non-centrosymmetric and the systematic absences restrict the space group possibilities to P6 1 or P6 5 . The structure was solved by the direct methods procedure in SHELXS86 (Sheldrick, 1990). Full-matrix least-squares refinements based on F 2 were performed in SHELXL97 (Sheldrick, 1997), as incorporated in the WinGX package (Farrugia, 1999). The correct enantiomer was chosen based on the known chiral centers at atoms C(1) and C(5). 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 > 2sigma(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.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq C1  (2)