A search of the site did not reveal anyone having attempted to obtain harmine / harmaline sulfate crystals by a modified manske crystallization procedure. Here I report that performing a manske using epsom salts (magnesium sulfate heptahydrate) in place of salt does not work.

Freebase P. harmala ß-carbolines were prepared and dissolved in a slight excess of vinegar. This was diluted to obtain 50 ml of a 2% solution of alkaloid acetates. The solution was heated and 5 grams of epsom salts were dissolved in. The solution was cooled to 5°C and no crystals formed. The solution was again heated and another 5 gram quantity of epsom salts dissolved in. Being cooled again to 5°C still no crystals formed.

Oh well, it was worth a shot

Thanks Auxin

So glad you posted the results. Are there any other salts you would consider experimenting with?

I've been toying with other ideas. I have quite a few options in my lab space but I'd prefer to keep it to common (and cheap) materials. Hence this trial.
The sulfates would have been groovie since they form square crystals rather than needle crystals, easier to clean.

I think one future direction with harmala salts will be to find ones that easily make sizable crystals for manual separation.
Harmine, harmaline, and tetrahydroharmine salts would form pure, independent crystals out of a mixed solution. If the crystal forms were distinctly different one could just pick the THH crystals out of a mixed batch with harmaline, for instance. I got that idea from the original manske harmine/harmaline separation procedure which exploits not only their pKa but also the crystalline shapes of the freebases ('The Alkaloids: Chemistry and Physiology, Volume 2' pg. 393) as well as from Lois Pasteur who once separated tartaric acid isomers by hand based on the optical rotation of each crystal

Fascinating stuff-- and useful for someone like me, who is not so great at separating harmaline from harmine....

The Epsom salts 'Manske' won't work as the NaCl Manske relies on the low solubility of the harmala HCl. No chloride, no precipitate. Using MgCl2 might work. The higher solubility of magnesium chloride will provide more chloride ions to drive a possible precipitation but the magnesium ions might also combine with the harmala alkaloids through their greater Lewis acidity, thus disrupting the process. Only an experiment will tell.

Magnesium chloride is a by-product of producing sodium sulfate using magnesium sulfate and sodium chloride. Sodium sulfate crystallises out and magnesium chloride remains in solution.


Not (initially) the optical rotation but the physical appearance. He examined each crystal using a magnifying glass and separated the crystals into the two mirror image forms using tweezers. Growing big crystals obviously will have made his work rather easier.

It's fortunate that tartaric acid crystallises in such a way that the chirality of the molecule is in a sense observable on a macro level. https://en.wikipedia.org/wiki/Louis_Pasteur#Molecular_asymmetry

Well you cant quite blame me for trying tho.
To the best of my knowledge there is no collection of known harmala salts solubilities available so the sulfate salt solubility was an unknown.
If I'm wrong and there is a collection of established salt properties somewhere please let me know before I start fiddling around in redundancy making harmine glutamates, tartates, citrates, salicylates, etc. (I wouldnt make those for modified manskes, of course, but their crystalline forms and solubilities would be interesting to know for other applications, or the all important 'shits and giggles', naturally).