Check this out:
Syntheses of 3-methylamino-a-phenylpropane and y-methylamino-a-phenylbutane.
Actually, "3-methylamino-a-phenylpropane" is listed as "B-methylamino-a-phenylpropane" in the abstract. Lugh: What makes you think that the "B" = "3"?
Ibid. 164-7. -PhCH2CHMeOH
Notice the middle carbon atom with one hydrogen atom attached. Carbon has the ability to create four bonds. No more. The carbon atom bonded directly to the phenyl group is also bonded to two hydrogen atoms and to the second carbon atom, which for the second carbon atom also represents one bond. The second carbon atom is bonded to one hydrogen atom, and also to the two carbon atoms--the first and third, to be exact--located on either side of it. THE SECOND CARBON ATOM IS THE ONLY ONE IN THE PROPANE SIDE CHAIN THAT CANNOT ACCOUNT FOR ONE OF ITS BONDS. Furthermore, the "OH" group cannot belong to the "Me" group (methyl group) as the Me group already has three hydrogen atoms along with the second carbon atom accounting for its four available bonds.
(XXII) (16.3 g.) at 0° treated with 13.5 g. PBr3 dropwise in 1.5 hrs., let stand overnight at room temp., the product poured in ice water, washed with 5% NaOH and water, shaken with 1 ml. coned. H2SO4, and distd. gave 12.6 g. PhCH2CHMeBr
Same thing here. The Br atom must be connected to the second carbon atom as it is the only carbon atom that cannot account for one of its available bonds.
(XXIII), b4 88-9°; 13.6 g.XXII in 60 ml. C5H5N at below 10° treated with 23.7 g. pMcC6H4S02C1, stirred 5 hrs. at 10-15°, the product poured in ice water, and the ester washed with petr. yielded 27.5 g. PhCH2CHMeO3SC6H4Me-p
Here it just looks as if the authors used toluenesulfonylchloride to create an ester from the phenylisopropanol.
(XXIV), m. 91-3°. XXIII (10 g.), 13.2 g. la, and 50 ml. HCONMe2 heated 5 hrs. at , 100° and 2 hrs. at 130-40°, the HCONMe2 removed in vacuo, and the residue extd. with Et20 gave 10.7 g. of a neutral oil (A) consisting of a mixt. of N-(9-phenylisopropyl)saccharin (XXV) and XVIII (R = PhCH2CHMe) (XXVI), sirup;
This part was typical of the advanced Gabriel synthesis using DMF as the solvent. Apparantly, the DMF helps to increase the reaction rate (probably via an SN2 reaction) and lower the reaction temperature, thus minimizing the formation of byproducts. Also, in the abstract, the saccharin intermediate is written as "N-(B-phenylisopropyl)saccharin" which seems very similar to B-phenylisopropylamine. Furthermore, why use "phenylisopropyl" prefix if not to indicate that the saccharin molecule is attached to the second central carbon atom on the propane side chain?
6 g. A boiled 2 hrs. on a water bath with 2 g. - NaOH, 2 ml. water, and 20 ml. MeOH, the McOH removed in vacuo, the residue with water poured in HCl at 0° gave 5.3 g. rapidly sepg. o-HO2CC6fI4SO2NHCHMeCH2Ph (XXVII), sirup, and 0.4 g. slowly crystg o-H02CC6H4S02NH2, m. 154-6°.
Apparently, aqueous NaOH hydrolysis of the alkylated saccharin intermediate allows for the cleavage of part of the saccharin molecule, protecting one side of the nitrogen atom, while leaving the other side available for further alkylation.
XXVH (5.3 g.) in 20 ml. 10% NaOH was treated with 1 g. NaHCO3 and 6 g. Me2SO4, kept 30 min. at 0°, 1 hr. at 50-60° and 30 min. at 70-80°,
This step adds a methyl group to the nitrogen atom, thus making it the precursor to the methylamino-phenylpropane.
the ester on the bottom washed with dil. NaOH, boiled 2, hrs. with 1.6 g. NaOH, 2 ml. water and 10 ml. EtOH acidified with HCl, the EtOH removed in vacuo, the residue (4 g.) boiled 10 hrs. at 130-40° with 100 ml. coned. HCI,
This part describes the hydrolysis of the final alkylated saccharin artifact to yield the desired monomethylated alkylated secondary amine; namely, B-methylamino-a-phenylpropane, better known as methamphetamine(?).
made alk. and the product steam distd. in dil. HC1 to obtain 1.2 g. (30% from XXIII) MeNHCHMeCH2Ph.HCI,
This part is definitely confusing because the way the atoms are all arranged, I really can't make heads or tails of the molecule's structure.
leaves, m. 131-3°; picrate, needles, m. 122°.
This is where I pretty much determined that the product of this reaction--of "the saccharin method"--was actually methamphetamine.
The m.p. of d,l-methamphetamine-HCl is about 132-135 C.; the m.p. reported for the product above is 131-133 C. Almost right on the money. As for methamphetamine picrate, it's listed as 128 C. Only six degrees away from what is reported above, 122 C.
It would appear that in order to resolve this mystery, the actual Journal of the Pharmaceutical Society of Japan articles will have to be fetched by someone who has access to them. No doubt, somewhere within the pages of the original articles--likely along with valuable and useful references to past Gabriel syntheses--will be found diagrams of the molecules in question that will help determine once and for all whether or not "B-methylamino-a-phenylpropane" is indeed synonymous with "methamphetamine".