In
Patent US4849521
they describe a stereospecific synthesis of cis-3-substituted fentanyls. First they prepare an imine from the unsubstituted piperidone, and then deprotonate that in the 3 position with LDA. Next they alkylate that position with an alkylhalogen, to get a 3-subst. imine. Finally they reduce the imine with lithium triethylborohydride to get the cis compound (and a little bit of the trans also).
(The search function isn't working, I hope noone has posted this before.)
But that seems a bit hard. Suppose you already had 3-methylpiperidone and reduced an imine of that with lithium triethylborohydride, would you get the cis-variant then also, or is it the first steps that make that possible? They mention that the hydride is to crowded to attack on the other side, so to me it seems like it would work?
From the patent:
EXAMPLE 1
This example illustrates the preparation of 1-benzyl-3-(cis-methyl)-4-[N-(2-fluorophenyl)amino]piperidine. A solution of 1-benzyl-4-piperidone, 15.20 gms (80.31 mmol), 2-fluoroaniline, 9.02 gms (81.17 mmol) and p-toluenesulfonic acid monohydrate, 0.48 gms in 200 ml of toluene was refluxed overnight under argon collecting water in a Dean-Stark trap. After 18 hours the theoretical amount of water had separated and the trap was drained several times to distill off 100 mls of toluene. The reaction was cooled under argon and diluted with 40 mls anhydrous THF. The deep orange solution of crude imine 1 was used "as is" for next step.
LDA was prepared under an atmosphere of argon by adding freshly distilled diisopropyl amine, 15 ml (107 mmol) dropwise to a cold (0 DEG C.) solution of 55 mls of 1.6M n-butyllithium (88 mmol) in hexane in 60 mls of anhydrous THF. After 15 minutes at 0 DEG C. the LDA solution was cooled to -78 DEG C. in an isopropanol-dry ice slush bath. The solution of imine 1 (80.31 mmol) was added dropwise with stirring. The reaction was stirred at -78 DEG C. for 1 hour followed by rapid addition of methyl iodide, 12.9 gms (90.88 mmol). The reaction was stirred at -78 DEG C. for 15 minutes followed by warming to room temperature. After 10 minutes at -78 DEG C. a precipitate formed which redissolved as the reaction warmed up. After stirring at room temperature for 2 hours the reaction was cooled to 0 DEG C. in an ice bath and 120 mls 1.0M Lithium triethlyborohydride in THF was added dropwise via syringe. The reaction was stirred overnight (16 hours) warming to room temperature. The reaction was then cooled back down to 0 DEG C. and 50 mls of water was added slowly dropwise (exotherm and vigorous gas evolution). The quenched reaction mixture was concentrated to a pasty oil which was dissolved in 200 mls toluene and washed with 100 mls of water. The toluene layer was separated, dried over anhydrous sodium sulfate and concentrated to give a crude mixture of cis and trans diamine. Analytical LC showed the cis/trans diamine ratio to be 92.2:7.8. This crude oil was flash chromatographed on Silica 60, 230-400 mesh (400 gms) eluting with 1:10 EtOAc/Hex with 0.1% ammonium hydroxide added to give 9.74 gms pure cis diamine (40.6%) shown below and 1.57 gms cis and trans diammine mixture. Total isolated yield of cis and trans diammine was 47.2%.
