A bisulfite adduct wouldn't form using the organic product from the reaction of benzylmagnesium chloride and acetonitrile using the procedure from JOC 51 5338-41 (1986) & 52 3901-4 (1987) 

After getting lost with "metal/acid nitroreductions" was  investigated "the legendary classic route" to amphetamine via grignard from Benzyl chloride (or cyanide) and acetonitrile with NaBH4.
The yields for the ketone are not good enough but some one said high yields in the amine formation.
Practically is needed a flame or microwave to dry the glassware, molecular sieves to dry the ether, and a ultrasonic bath to kick start the reaction of the grignard reagent.
A good condensor is recomended to keep the solvent inside the flask and inert gas or a calcium chloride tube to keep out moisture in the reaction.
After addition of acetonitrile, is commented a white precipitate is formed. and it can be isolated by filtration or reduced with NaBH4 in a "one pot synth".

On the paper sounds great!!! Not impossibles reagents like nitroethane, with a ultrasonic bath in the Rodhium´s archives succes in grignard reaction are obtained even with not dry-ether and wet weather, and the possibility of one pot way from precursor to final product make of this reaction a powerfull tool to keep in mind.


And finally some one (I can´t find where i read it) propose one method to form the "Piperonyl-Magnesium" reagent from piperonil chloride.

if Any one had experience or recomendations on this routes, is welcome.
I prefer ask here before open a new thread wich would end in nothing

I like grignard!!!