The following is adapted from J. Org. Chem., 47, 4327
A mixture of 2-bromosafrole (24.1g, .1mol), hexadecyltributylphosphonium bromide(5.1g, .01mol), sodium azide (16.2g, .25mol), and water (50ml) is magnetically stirred at 80'C for 24 hours. Flask is cooled and phases are separated in sep funnel. The dark reddish organic layer is diluted with toluene (20ml) and poured back into flask and stirring is restarted. A solution of sodium borohydride (11.7g, .3mol) in water (30ml) is carefully added through an addition funnel over 30 minutes as temp is raised to 80'C and held there for 16h. Contents of flask are cooled and poured into a sep funnel. Aqueous layer is sepped off and organics are washed w/ 50ml dH20. Three 40ml portions of 4M HCl (approx 10%) are used to extract MDA base from organic layer. Combine all acid extracts and neutralize with 50% NaOH soln. Golden beads of MDA base will fall out of soln. Use two portions of 50ml toluene to isolate base. At this point one of two things can be done. If you are a greedy shoddy chemist you will dry the toluene extracts with anhydrous MgSO4 then bubble with HCl gas to isolate a relatively pure white MDA hydrochloride (8-12g) OR if you have any pride in your chemistry skills and product, toluene will be evaporated down to a orange oil and vacuum distilled using standard methods to yield 6.5-11.0g of water white pristine MDA base. Distilled base is most easily crystallized by adding about 6 times volume of dry IPA and neutralizing w/ conc. HCl. Place this in the freezer over night to find glorious white crystals of the hydrochloride. At this point not all of the salt has precipitated and it won't no matter how long you leave it in the freezer, so add the same volume of dry acetone to ppt. entire yield (6-12g).
Notes:
1) 2-Bromosafrole was prepared using process developed by Fester. HCl(g) is bubbled into a soln of 48%(aq)HBr, HOAc and PURE DISTILLED safrole. The procedure is archived on rRhodiums site and runs exactly as Fester claims, believe it or not!
2) Yield of product is totally dependant on reaction time/temp of first step. If literature citing is examined yields above 85% are reported for all primary alkyl bromides. Problem with lower yield here is the fact that we are dealing with a secondary bromide. This introduces the possibility of occurrence of two yield killing side reactions. Elemination of -Br resulting in isosafrole then elemination of azide may occur resulting in more isosafrole. Perhaps reducing the temp and extending rxn time will slow rate of elemination resulting in higher yield of alkylazide. Someone willing to try?
3) Hexadecyltributylphosphonium bromide is very expensive. Refs stating Aliquat 336 is a cheaper alternative PTC are on the Hive.