This has been published a few times but after using TFSE I havn´t found it mentioned here.
Some benzaldehydes are really tricky to condensate with nitromethane, nitroethane and particulary 1-nitropropane, the usual way. Shulgin´s favourite method using AcONH4 and AcOH on the steambath gives in many instances crappy yield or nothing. Equimolar amounts of the benzaldehyde and nitroalkane, 10% n-butylamine or cyclohexylamine in the dark for a week or two can work, but who can handle the waiting? I can´t.
This is a twitch that can work.
E.g., 100mmol of your favourite benzaldehyde and 100mmol n-butylamine (actually most aliphatic amines with 1-4 carbons work) is allowed to react in refluxing toluene. Water is removed with a Dean-Stark water trap. When 1,8ml water has been removed the toluene is distilled off. The resudual aldimine is combined with 1,1 equivalents nitroalkane in 50-100ml glacial acetic acid @ 90-100 deg C for 30-60min. When the aldimine has been consumed (TLC), the acetic acid and excess nitroalkane is removed in a rotovap and the residual oil is allowed to crystallise in MeOH @ 5 deg C.
This is just written from memory. I´ll come back with references and a few recipies.