I would be useful to purify compounds cause then you would have starting materials. Like magnesium to make a gringard reagent.

Like if you could split apart diphenylhydramine, then you would have a alkyl amine.

Diphenylhydramine + something ---------> (phenyl)2OH + [(CH3)2]N(CH2CH3)

The [(CH3)2]N(CH2CH3) might be able to be formed into a gringard with a purified magnesium suppliment someshow that could be used to add to a carboxylic acid derivative to from another amine that would be closer to the product. It is a stupid idea, but a synthetic pathroutes are always useful.