Purifying an ethanolic extract of Artemisia annua

[DrDoom]

Hello everyone,

 

This is my first post here. I am trying to purify artemisinin from an ethanolic extract of Artemisia annua (sweet wormwood), and I ran into the the typical problem encountered by biochemists doing this procedure: removing the plant waxes without losing artemisinin. Most of the literature I have come across recomends using petroleum ether or hexane to defat ethanolic plant extracts. This would not work in my case because artemisinin happens to be very soluble in both of those solvents (in fact, hexane is used for industrial-scale production of artemisinin). I read a tip somewhere about using molten parafin wax, but that just removes the fats (as far as I know), not the wax. Am I wrong about this? Also, does the molten paraffin work well as a defatting agent? I read somewhere that heavier waxes aren't easily extracted by ethanol, is this true? Would defatting work with very cold petroleum ether (maybe the artemisinin won't be soluble enough at that temperature)?

 

Here is the rough plan of attack:

 

• Filter ethanolic extract
• Remove pigments with activated carbon
• Defat with molten paraffin wax
• Liquid-liquid extraction (either hot petroleum ether or isopropanol)
• Crystallization
• Adsorption chromatography (column with diatomite, elute with?)
• If necessary, wash with activated carbon again
• Recrystallize

Any comments on how to improve this?

 

Thanks for your time, I really appreciate any kind of feedback.

 

Sincerely,

Dr. Doom

[mcrnotts]

Hi Dr Doom,

 

Just stumbled upon your post. I suppose this reply might now be too late to help you, but...

 

You state that the solubility of artemisinin in petroleum ether/hexane is very high but actually this is not the case; it shows the lowest solubility of all solvents that are reported in the extraction of artemisinin. You can find the data here: Nti-Gyabaah, J., Gbewonyo, K., Chiew, Y.C., 2010. Solubility of Artemisinin in Different Single and Binary Solvent Mixtures Between (284.15 and 323.15) K and NRTL Interaction Parameters. Journal of Chemical and Engineering Data 55, 3356-3363. That said, the co-extracts that are extracted into the hydrocarbon solvent can have a huge impact on artemisinin solubility and I have observed almost 10x the concentration in an extract than the saturation value for the pure solvent.

 

I'm also not sure that you will get a defined interface if you were to try L/L extraction of ethanol with hexane. I have a feeling I tried something similar in the past (but when the extract was initially produced using hexane) and saw complete miscibility due to the complexity of the mixture. An interface would only form on the addition of a reasonable amount of water. Even if they aren't fully miscible, there is a chance of an emulsion that would be difficult to deal with.

 

What is the thinking behind diatomite in regards to what it removes? The industrial approach is often to use silica gel. I think I have only seen diatomite mentioned once, in "A novel purification method of artemisinin from Artemisia annua". The process developed in this paper is... unusual. The final product purity (98%) is insufficient for sale and inclusion in ACTs (with a requirement of at least 99% from buyers), the number of processing stages is almost triple the industrial approach, the solvent volumes that must be cycled through boiling and refrigeration are uneconomical and the yield of 60% is no improvement over current mass-scale. I'm curious about what you are trying to achieve. Is this for a large scale extraction or smaller scale simply for laboratory purposes? Perhaps because of the low artemisinin content of leaves, the process mentioned in this paper could be useful in the laboratory in the absence of sufficient leaf supplies I suppose.

 

Finally, a study into ethanolic extraction has already been published - "Assessing the Technical and Economic Viability of the Ethanolic

Extraction of Artemisia Annua" is freely available via the MMV.

 

Hopefully that might be of some use anyway.