(3R,6S,7aS)-3-Phenyl-6-(phenylsulfanyl)perhydropyrrolo[1,2-c]oxazol-5-one

Molecules of the title compound [systematic name: (2R,5S,7S)-2-phenyl-7-phenylsulfanyl-1-aza-3-oxabicyclo[3.3.0]octan-8-one], C18H17NO2S, form high quality crystals even though they are only packed using C—H⋯O(carbonyl) and weak C—H⋯S interactions. The dihedral angle between the aromatic rings is 85.53 (5)°. The fused rings adopt envelope and twist conformations.

lographic study of the title cis(endo-) adduct 2 (Scheme) so that structural inferences made on the basis of nOe experiments could be corroborated (see bottom right entry in fig. 3).
The molecular packing is provided by mainly weak C-H···O interactions (entries 1-3, Table 1, involving a bifurcated O) as partly shown in Figure 2. The weak, just significant, C-H···S interactions have been observed before contributing to dimer formation in CUQXUH with a C-H···S angles of 140° (McCarthy et al., 1999).
These crystals were of superb quality, confirmed by the final agreement indices and difference density maps, which raises the issue of how many of the weak intermolecular interactions may just be fortuitous, with packing largely determined by the overall molecular shape and van der Waal's forces.

Experimental
To a solution of the lactam 1 (19.7 g, 97.0 mmol) in dry THF (180 ml) was added dropwise LiHMDS (1 M solution in hexanes; 116 ml, 116 mmol) at -78 °C and the solution was stirred for further 60 min before a concentrated solution of diphenylsulfide (25.4 g, 116 mmol) in dry THF (25 ml) was added rapidly in one single portion at the same temperature.
After being stirred for 3 h the reaction mixture was quenched at -78 °C with saturated aqueous ammonium chloride solution (50 ml) and allowed to warm to ambient temperature before being diluted with ethyl acetate (400 ml). The separated organic phase was washed with water and brine, dried over MgSO 4 and concentrated to give a crude oil, the composition of which was determined by HPLC analysis. Based on this analysis the two diastereomeric phenylthioethers 2 and 3 were formed in supplementary materials sup-2 40% and 49% yield, respectively. In addition, the reaction provided the bis(phenylthio)-adduct 4 in 9% as the by-product and 2% of the starting material remained unreacted. The crude product was finally fractionated by flash column chromatography (silica gel, 10% and 20% ethyl acetate/petroleum ether) to yield the three adducts 2, 3 and 4.

Refinement
A total of 9 reflections within 2θ 68° were omitted from refinement as either outliers or partially screened by the backstop.
All H atoms bound to carbon were constrained to their expected geometries (C-H 0.95, 0.99 & 1.00 Å). All H atoms were refined with U iso 1.2 times that of the U eq of their parent atom.

Fig. 3. Preparation of the title compound.
(2R,5S,7S)-2-phenyl-7-phenylsulfanyl-1-aza-3-oxabicyclo[3.3.0]octan-8-one 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.