Dr. Shulgin, in his seminal work PiHKAL, uses an interesting method to form the versatile ketone 3,4-methylenedioxy- phenyl-2-propanone (MDP-2-P) from isosafrole. For what it's worth, I've done some investigation into the process, even going so far as to get the original text (a difficult to obtain article from a Japanese pharmacology journal), to see if any improvements to the process can be made, as it's a dismal and low-yield one (usually ~61%). A Journal of Forensic Sciences article came in handy for this. This article (ref at end) discusses modifications to uncommon processes for making P-2-P, but, a little gem was found in it that is equally applicable to making MDP-2-P, and does more than just improve the yield. But before that is discussed, other mods I have experimentally derived which apply to the beginning of the process will be discussed as well as a little review of how it's "supposed" to be done. In the normal scheme of things, one adds 32.4g isosafrole dissolved in 120mL of acetone to a stirring solution of 150g 80% formic acid and 32g of 30% hydrogen peroxide at a slow enough rate such that the temperature does not climb above 40C. Attempting to increase the amount of isosafrole processed (with a corresponding increase in the other reagents) generally results in diminishing returns in yield, up to the point (around 50g isosafrole) where the amount of byproducts is so great, that purification at the end by washing with 10% NaOH results in a homogenous mixture, instead of the expected two-phase one (ie - toss it all out). As well, it takes upwards of an hour to add all of the isosafrole solution to the performic acid solution, even with ice bath cooling. The first modification to the process involves preparing the performic acid ahead of time in a glass beaker immersed in plenty of ice and water. Let this solution stir for 15 minutes so that it is chilled to less than 5C. Also prepare the isosafrole/acetone solution and let it chill as well, though it need only be less than 15C or so. Add the isosafrole/acetone solution to the performic acid solution *quickly* in 20mL aliquots with vigorous stirring. The temperature should rapidly rise to 23C and hold there. All of the isosafrole solution should be added within 2 minutes. Let the mixture stir for the requisite 16 hours as indicated in PiHKAL. At the end of this time, add chilled 25% sodium hydroxide solution to the beaker *carefully* while monitoring pH with a meter. It will take several hundred mL of the solution to neutralize the acid/glycol mix. Watch for the pH to settle at around 8.2. The glycol will separate out as a nearly clear layer on top of the red aqueous layer (the colors are sometimes reversed). Separate the layers in a sep. funnel, and extract the aqueous layer with 2x50mL of methylene chloride. Combine the glycol layer with the extracts, then strip off the methylene chloride. It is okay to leave some methylene chloride in the flask, as will be seen. Next add the requisite 60mL of methanol and 360g of 15% sulfuric acid solution and bring to a *gentle* reflux, but don't put the condenser on the flask until after it starts boiling. This will easily drive off any remaining methylene chloride. Hold at reflux for 3 hours. Finally, allow the flask to cool by stirring in the air for about 20 minutes, then add 25% sodium hydroxide solution until the pH is slightly alkaline. A reddish/black, seemingly homogeneous system will result. Pour into a sep funnel and extract with 1x100mL, 1x50mL, and 2x25mL of methylene chloride. Combine the extracts. Strip off the solvent without vacuum, then apply vacuum to flask off any remaining solvent and to distill the ketone (150C @ 10mm Hg). In several test cases where these modifications were employed, and vacuum distillation of the ketone was used to purify it, the yield of MDP-2-P was 25.2g, instead of 20.6g (for 32.4g of isosafrole processed). While not a spectacular increase in yield, the savings in time and the elimination of the step where the particularly corrosive glycol ester/formic acid solution is concentrated under vacuum (a great destroyer of vacuum pumps) makes these mods well worth the consideration. References
|