Aum_Shanti said:
Didn't know that PH measurements had to be that accurate. For that I would need a new PH meter. I have such an el cheapo not really accurate thing...
Aum_Shanti, Jees, pH-meters don't need to be that accurate if you're willing to sacrifice a small amount of alkaloids to a 'mixed fraction'. Let me explain: first you'll notice a drop in pH at a pH not much above 7. If you follow VDS's experimental technique in paragraph 2.3, the depression would start at around 7,2. Actually, it doesn't matter at all at what pH the depression starts, the fact that it depresses is the main focus. Then, after having added a certain amount of base (you can get an order of magnitude from the protocol in VDS's article), you will see the pH rising again. Starting from lets's say 6,3 to around 7,0. At this point, most of the harmine has precipitated. Between 6,5 and 7,5 only 10% of the harmine needs further basification. And between 7,0 end 7,5, this amount gets completely negligable. This can be clearly observed on graph 3.3.3.
So this means that even if your pH-meter is accurate to only 0,5pH, you'll be able to selectively precipitate between 90 and 99% of your harmine (that could have been precipitated considering inevitable loss of about 5% through the reprecipitation-process) if you stop basifying at pH 7,0 (confidence interval 6,5-7,5). So first precipitation will start, and most likely yor meter will first have turned to 7,0ph before showing 6,5. If not, no problem, just continue basifiying slowly under constant stirring and from the moment the meter reads 7,0, stop the addition of base. Wait until the pH reads 6,5 again under occasional stirring. As long as the pH returns to 6,5, keep adding base in small increments until the pH stabilizes for more than an hour at 7,0. Filter of harmine, add more base to bring the pH to 9 or 10 and filter off your DHH. Optionally, you can isolate a mixed fraction by slowly adding base to solution stabilized at pH 7 until a definite amount of precipitation starts and a depression of 0,5pH or more is seen.
If your meter is accurate to 0,1pH, the amount of 'mixed fraction' that is not separated and mainly consists of DHH, can be kept to 2% (page 19 of article). In fact, the separation becomes so complete, that one can skip the isolation of this mixed fraction completely. Then one just stops at about 7,5 (or as Jees suggested: 0,5pH higher than the starting pH of the depression), filters of the harmine fraction and base further to precipitate the DHH. For all practical reasons, this is the way to go. However, if you don't want any trace amount of harmine in your DHH, go and isolate the mixed fraction (between pH 7,5 and the first sign of precipitation + pH-depression). This won't take more than an extra 10 minutes.
When reading the article, I thought about using 'accurate' pH-strips (0,2pH). Well, I ordered some but found out that the color of the alkaloid-solution strongly influenced and deteriorated the reading of the pH-value. I also tried to dissolve some relevant pH-indicator (nitrazine and thymol blue) into the solution. Only to find out that the indicator was caught in the precipitating alkaloids. Now that I think about it, it might just work if you only add the indicator near the expected end-point. Then there's practically no more alakaloid to 'wash out' the indicator during the accompanying strong pH rise. Hmm, interesting...
BTW, Aum-Shanti, this pH-specific separation-process is very fast. I did my third in about half an hour. And it should be effective as well, given the peculiar pH-curve, the microscopical characteristics and VDS's narrow melting points. We'll get definite evidence on the effectiveness of this technique when the results of the LC/MS come in.
So, all this just to let you know that even a cheap pH-meter with a minimal accuracy of 0,5pH is sufficient for effective and efficient separation of harmine + DHH. A meter accurate to 0,1pH provides more control over the process and even further increases yield by minimising the mixed fraction. Good luck!