(3R,4S)-3-Methyl-4-phenyl-2-[(R)-1-phenylethyl]-3,4-dihydroisoquinolin-2-ium tetrafluoridoborate

The title salt, C24H24N+·BF4 −, is one of two possible diastereoisomers having a different configuration of the asymmetric centre in the α-phenylethyl substituent, whose absolute configuration was established to be R. The two phenyl substituents of the cation have a cofacial orientation, albeit with a long centroid–centroid separation of 4.129 (3) Å. The crystal structure exhibits numerous C—H⋯F contacts between counter-ions, with the tetrafluoridoborate anion surrounded by five iminium cations.


Comment
In recent years, much effort has been devoted to the development of organocatalytic processes that afford metal-free procedures (Adam et al., 2001), such as asymmetric epoxidation catalysed by chiral iminium salts. It has been proved that iminium salts are effective catalysts at low loadings for enantioselective epoxidation using Oxone® (2KHSO 5 .KHSO 4 .K 2 SO 4 ) as the stoichiometric oxidant.
As a part of our interest in catalytic epoxidation by dihydroisoquinolinium-derived iminium salts (Bohé et al., 1999), we report herein the synthesis from the corresponding (3R,4S)-3-methyl-4-phenyl-3,4-dihydroisoquinoleine (Davies et al., 1988), by N-alkylation with (1-bromoethyl)benzene, and the crystal structure determination of the title compound. It was isolated as one of the two diastereoisomeric products; it is a new dihydroisoquinolinium-derived iminium salt containing an asymmetric centre in an exocyclic substituent at the nitrogen atom. The X-ray analyses confirmed the 2D NMR data and allowed us to define the absolute configuration of the exocyclic substituent at the nitrogen atom to be R (Fig. 1). The tetrahydroisoquinoline unit is substituted by methyl group in position 3, a phenyl substituent in position 4, both in axial conformation, and a (1-phenylethyl) group at the nitrogen atom (Fig.1). The six-heteromembered ring adopts a screwboat conformation (rather than a half-chair one as previously described), as indicated by puckering analysis [Q = 0.474 (6) Å, θ = 113.8 (7)°, φ = 93.9 (8) °] (Cremer & Pople, 1975). The two phenyl rings in position C4 and C18 are almost facing each other with a dihedral angle of 19.3 (4)° but with a rather long centroid-centroid distance of 4.129 (3)Å. In the crystal of the iminium salt, significant contacts between cationic species are uniquely mediated by BF4-anions, each of them being surrounded by five cations (Fig. 2). The tetrafluoridoborate anions are involved in intensive thermal motion, thus some B-F bond lengths and angles [range from 1.281 (10) to 1.327 (10) Å and from 98.9 (9) to 116.6 (8)°, respectively] deviate significantly from their standard values.
Two diastereoisomers were obtained in 1:1 ratio. These diastereoisomers were separated by column chromatography.
Each compound was treated by 1 equiv of AgBF4 in acetonitrile. Filtration and concentration in vacuo afforded a white solid.
Only one diastereoisomer (the title compound) was successfully recrystallized. Crystals were grown by placing a solution of this dastereoisomer (45mg) in CH 3 COCH 3 (0.5 ml) at the bottom of a test tube, then carefully covering it with pure hexane (50 ml).The test tube was covered and left undisturbed. Colorless crystals of the title compound appeared after several days. [α] D 24 = -16.1 (c 0.4; CHCl 3 ), m.p. 447 K.
The systematic absences permitted P4 1 2 1 2 and P4 3 2 1 2 as possible space groups, but in the absence of significant resonant scattering, it was not possible to distinguish between these enantiomeric space groups and the Friedel-equivalent reflections were merged. P4 1 2 1 2 was selected because the enantiomer has been assigned by reference to unchanging chiral centres in the synthetic procedure.    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 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.