The main problem with this method is that one gets l-Meth, to be more precise one gets from slightly enriched l-Meth to predominantly l-Meth depending on how long one lets it crystallize out. The more time the more l-Meth is in the cristalline mass. When good stirring is employed the result will be even more l-meth in the crystals (in relation of course).
The good thing s that one gets pretty decent d-Meth fromm the mother liquor.
It does not matter if one uses l-tartaric or d-tartaric or d/l tartaric - it will always be l-Meth what crystallizes out most. It is completely different from Amphetamine btw. cannot be compared at all. I actually understood what is really going on only a short time ago and I fully understand why somebody might believe I am wrong, but well, I am not.
As I try not to criticize when I cannot offer an alternative, here it is:
7.84 g (0.0525 mol) racemic N-methylamphetamine and 9.73 g (0.0525 mol) racemic N-methylamphetamine hydrochloride were dissolved in 70 ml of abs. ethanol and 7.91 g (0.0525 mol) R,R tartaric acid (L-(+)-tartaric acid) in 56 ml of abs. ethanol.
Strong stirring for twenty two hours. room temperature, then the mixture is cooled in the fridge, the formed salt is filtered out and the mother liquor is worked up by A/B extraction and recrystallisation to yield almost exclusively d-Meth.
Its all in the literature. Just not in one place but a bit obfusicated.
Good luck
