A solution of 2.7 g. of the decahydro compound in 25 ml. of 85% phosphoric acid was refluxed for 70 hours and then poured onto ice. The aqueous phase was extracted with ether and the product then salted out with potassium carbonate. It was taken up in ether, dried and distilled to give 1.0 g. of pale yellow oil boiling at 130-132 Ñ at 0.7mm. The odor was quite similar to several derivatives of the l-azabicyclo[3.3.l]nonanering system which have been prepared by one of us.
Anal. Calcd. for CllH23N: C, 84.59; H,9.60; N,5.80. Found: C,84.33; H,9.50; N, 5.51.
Failure to form a crystalline derivative (picrate, hydrochloride, picrolonate) indicated that the product was a mixture. A test showed high analgesic activity for the mixture. The product decolorized permanganate readily suggesting the presence of 1-benzyl-2-methyl-Δ5-octahydroisoquinoline. The material was chromatographed by pouring a solution of 0.484 g. in 3.7 ml. of low boiling petroleum ether onto the top of 30 g. of aluminum oxide in a 50-ml. buret. The column was eluted with 20-1111. portions of low boiling petroleum ether to which was added 0, 0, 0, l, 3, 5, 5 and 5 ml. of ether, respectively. The last three solvents eluted 0.10, 0.12 and 0.12 g. The latter two samples crystallize; m.p. 50-54 Ñ. They gave a picrate melting a t 172-174 , not depressed when mixed with a sample of Dr. Grewe's N-methylmorphinane picrate kindly furnished to us by Dr. Nolte. The first fraction gave a picrate which, after one recrystallization, softened at 171' and melted a t 178'. It gave a large melting point depression with N-methylmorphinane picrate and was not further investigated.
© Collected Ruslan Anatolievich Kiper, burewestnik@mail.ru