Hi, I'm a new member and this is a long-winded post. The length will, I hope, provide evidence I'm not some nutter and that I consider this fascinating and legitimate research.
I had my first crack at a cactus extraction several weeks back. I used the limonene food-safe tek. I tried to do everything by the book, though I began with fresh San Pedro cactus (not pre-dried powder). Pulled the thorns pulled, sliced off the waxy layer. cored and diced the flesh, and oven dried on as low a setting as I could manage with the oven door open. A portion of the prepared, wet cactus was put through a couple of freeze-thaw cycles before drying.
This reduced about 1.6 kg of wet material and to about 79g of dried. That seems like a low yield, and I may well have over-dried some of it. The dried material was run through a food processor to pulverize into a chunky powder.
After mixing the dried cactus with water, ca. 20g of CaOH, and 250ml of limonene, I hit my first snag. The result was a complete mess of a mixture that could not be separated. It was like wet, green mud and went right through the filter of the French coffee press.
The only way I could think of to get the liquid out was to put the whole mess into a piece of t-shirt fabric and let it drip out (A friend gets the juice out of tiny wild grapes this way). This utterly failed.
I resorted to forcing the liquid out by twisting the fabric and squeezing, a laborious, time-consuming and messy process with a significant loss of liquid into the fabric. (If I’d been using xylene I would have gassed the whole city block.)
I let this first limonene pull settle and it separated into a coloured but clear limonene layer and a cloudy aqueous layer. I separated them with a sep funnel and put the aqueous layer aside. Since the mescaline ought to be in freebase form at this point, I figure the aqueous layer should be mostly unwanted junk. Please challenge my assumption if you think I'm wrong.
The next two limonene pulls went according to plan, with coloured and mostly clear limonene decanted from the cactus mash using the coffee press.
I did three vinegar extracts of each limonene pull, pooled them and evaporated the vinegar using a fan. The result with a thin, tawny brown, waxy, somewhat sticky layer of 415 mg impure mescaline acetate (I presume).

Then I did three carbonated mineral water extracts of each limonene pull, after shaking vigorously in a plastic jug. There was considerable pressure created which distended the sides of the jug, a good sign I believe. Note: I used "designer" bottled water, not just generic soda water, because it had more carbonate or bicarbonate. I now realize all the other dissolved mineral salts might have complicated things.
I pooled the carbonated water extracts and evaporated again using a fan. The result was 4.6 grams of off-white crystalline material scraped from the interior surface of glass baking dish, presumable impure mescaline carbonate (or bicarbonate?)

Finally, the water from the first botched pull was also evaporated using a fan and yielded about 1.5 oz of amber goo of God knows what.

What I’d like is as much pure mescaline as possible, in any salt form, but I’m not really sure what I’ve got or how to clean it?
Bottom line questions: Should I use cold methyl ethyl ketone (MEK) to wash the acetate, and cold 99% (isopropanol) IPA to wash the carbonate, or are there better alternatives? A food safe process would be nice but is not required. And what about the amber goo? Junk, or potential goldmine?
Your advice and comments would be greatly appreciated.
By the way, MEK, as people have pointed out, is trivial to find this in the US in any hardware store. Not so in Canada, but I did find some in a place the specializes in plastics where I guess it’s used to clean up fibreglass resin. Cheap and no questions asked. Also found xylene (unused in this procedure) in a paint store, after much trucking around. Limonene and CaOH I got from the US. PM if you want details.