Hello,
I'm a professional chemist, currently working in an HPLC lab until I can find a synthesis job. I am familiar with extractions, and have read many of the extraction techniques on the nexus over the years. However, I am primarily a synthetic chemist, and do not know all of the theories involved in extractions. I've found that the nexus is pretty much at the forefront of plant extraction chemistry, and thus turn here for assistance.

I am currently analyzing several plant extracts prepared by the local biology department for several extremely similar chemical compounds. I've mostly figured out how to get baseline resolution on them, and tried analyzing the extracts they gave me. Needless to say, they were primarily crap. The few peaks of interest are buried under a multitude of fats and presumably alkaloids to a ridiculous extent. I found out that the biologists were extracting things by literally soaking the plant in methanol, and giving me the resulting tincture. The compounds of interest are isoflavones and 7-O-glucosides of isoflavones, with varying hydroxy substituents.

I'm leaning towards a defatting step. However, I am not entirely certain if isoflavones can be acidified and basified as DMT, nor if that would help due to the many hydroxy groups. Can anyone shed some light on extracting these from assorted plant lipids and alkaloids?
Thank you

Stick it through a flash column?

This is I think the eventual plan, but for now I'm dealing with ~100ul extract sizes. Considering I need 10ul just to have the compounds register on my UHPLC, flash column is not even remotely possible. That's why I'm trying to cut more of the crap out during the extraction process. I'd need most of the sample as is to be able to reliably run results through a mass spec.

The other concern, and the reason I can't do, say, a microcolumn filtration, is that a good chunk of the fats are overlapping one of the peaks of interest. They would go through the column and muck up the sample anyway, assuming I didn't lose everything to the silica.

the compounds you're investigating (isoflavones etc) will be soluble in MeOH, so acidification and wash with an NP solvent should get rid of a bit..also, application of boiling in a strong acidic solution will congeal and break down a lot of un-needed compounds too

I too know the pain of working instrument analysis with too small amount of material.

Flash column would definitely be ideal, maybe ask those biologists to get you more material.

You could try drying them and then rinsing them with dichloromethane, and then pipetting it off, that might be enough to remove the fats but leave most of the flavinoids, worst case you can always recover it from the dichloromethane

I'm faced with a similar conundrum at the moment (how to work out CBD in a plant oil matrix?), at least you have access to GC/MS.

Looking through options, there's solid-phase microextraction to consider, Kryptos.

I'm stuck with kitchen-grade attempts at quantitative TLC, I think (sorry to partially thread-jack, will have to post my own!)