title="View this image">

[/url]
>[1] <html><body><p>
J. Am. Chem. Soc. 74, 763 (1952)</p></body></html> (
https://www.thevespiary.org/rhodium/Rhodium/pdf/ritter.reaction.safrole.pdf)
>[2] <html><body><p>
Journal of Clandestine Laboratory Investigating Chemists Association</p></body></html> (
https://www.thevespiary.org/rhodium/Rhodium/chemistry/clic.html)
[/quote]
The above result of the Ritter rxn w/ safrole MDisoquinoline(image) If HCN is used instead of CH3CN the methyl group beta to the aromatic ring wouldn't be there, and if the imine was reduced.It might,just might cleave off the aromatic ring under "reverse Mannich disconnection reduction" conditions eg; Heat formic acid/formamide similar to the cleavage of tetrahydro-B-carbolines.Kind of a reverse Pictet-Spengler Rxn. = P/S
for MDMA

The odd thing is in general when a benzyl group is used as amino protection.As in benzphetamine,hydrogenation cleaves it off as toluene giving methedrine.
As a starting material 3,4-MD-benzylamine with HOCH2CHXCH3--then Ac amino protection -> 3,4-MD-ArCH2N(Ac)CH(CH2OH)CH3, then Friedel-Crafts subs the hydroxyl w/ -H20; for the MD-tetrahydroisoquinoline.That might cleave also under reverse Mannich rxn for MDMA.
Where the first example given by Pictet-Spengler which used a unsubstitued ring PEA.
For the tetrahydroisoquinoline from PEA and formaldehyde.
Later investigators couldn't reproduce P/S results with PEA.Actually, the P/S rxn is a Mannich rxn.. P/S reported this one year before Mannich reported the Mannich rxn.Thus, in reality P/S invented the Mannich rxn.P/S's first reported rxn wasn't repeatable by later researchers.
I have a lot to say about this scheme,and will go into it later when time permits.
It's one of my pet peeves.