I've heard someone on here say DMT is much easier to extract, and I don't agree. But maybe it's a matter of preference.

In DMT A/B extractions you typically use NP solvents or (in dry teks) pulls with alcohol, acetone or similar. The thing I don't like about pulls is that I don't know how many times to pull, how much solvent to use etc., then how many times to wash / salt the goodies out or otherwise get them out of the solvent. I don't know when to stop and whether I'm discarding any goodies.

In simple A/B harmala extraction you base the acidic soup, the freebase alkaloids crash out and you collect them all in one go. To purify them further you may need to repeat certain steps, but repetition only serves refinement, not yield maximization (such as repetitive pulls in the case of DMT). I find this elegant.
Ok, one could argue that doing further Manske precipitation (if you choose to include Manske) on the Manske supernatant reduced in volume to push more goodies out is a repetitive step that serves yield maximization, but there is an exponential relationship here (i.e. after k manskes you have <= f^k goodies left in the supernatant for some 0 < f < 1) rather than a linear one.

What makes DMT so different that the same approach can't be used (or can it?)? Does fb DMT get suspended in the basic soup and take prohibitively long to settle? Does it form crystals so small that they pass through the filter? Does it get stuck in the plant gunk?

Is there a way to avoid pulls and stick to filtering and decanting, even if it takes longer (like weeks or even months of waiting)? Are there plant sources that are better candidates for that due to lower gunk content, e.g. MHRB?

You can throw a little bit of science to answer all these questions.

For example, you can determine a partitioning coefficient for various solvents, temperatures, and salt concentrations. From this, you can calculate the near optimal amount of solvent and the number of pulls.

You can also simulate the extraction using already extracted DMT, just get an idea of how much is extracted.

Usually I take it more as an art than science and fiddle with parameters, observing results and ten go by my feeling. I believe most people do it like that.

I have these book, one of which (I don't remember at the moment) has a chapter of liquid-liquid extraction, discussing solvent selection, partitioning coefficients, calculating matter transfer, salting out and more:

https://www.amazon.com/A...y-Leonard/dp/1138494143/
https://www.amazon.com/E...a-Cranwell/dp/1119952387

But you can definitely find more about that, these are basic principles taught in chemistry, physical chemistry, organic chemistry, chemical engineering and such. There is a huge interest in optimizing these processes in industry so there are methods on how to do that.

I would search and learn more about liquid-liquid extraction, salting out, fractional crystallization and more.

When I started reading chemistry books, I found 90% of guesswork can be eliminated, I just have to beat myself up to learn more.

Must be something about properties of the soup as dmt fb really can crash out and settle, when vinegar nps saltings are made basic again.
Once I have tried harmala-style tek on very concentrated mhrb acid soup. After basification, it turned milky as normally, but did not turn back to almost black colour (as would normally happen). It seemed that dmt crashed out (it was visually similar to the same step with harmala), but did not settle at the bottom even after 1 day.
I had to dilute it again, after this it turned black again, and continue with nps pulls.

The logKow of harmine, for example, is about 4, while DMT is about 2. Since the logKow is a logarithm, each change in one whole number represents an order of magnitude of change. So harmine's logKow of 4 is two orders of magnitude greater, or 100 times less water soluble than DMT, with a logKow of 2.

This means DMT is more polar and has a higher affinity for polar molecules, so it hangs in that water even if a polar solution were saturated with the very popular polar duo - NaCl or table salt.

Freebasing the DmT makes it less polar and more likely to crash out of solution, but keep in mind that sodium hydroxide (NaOH) is similar to salt in that it reduces the "room" for less polar molecules, because the water molecules are so taken by Na+'s super polar sexiness, and don't even have time for the moderately attractive DMT molecules, which then get pushed out as precipitates.

I know I've read that people have issues when the initial acid solution is reduced to too small a volume, and getting unwanted DMT precipitates after basifying.

Doing all that filtering with a basic solution though? Sounds like work.

Unless you only brought the pH to 2+pKa, which I think would be about pH 11.5. Not finger-melting!

One might think that a concentrated water based extract could be basified, almost in a STB fashion, diluted with 95% ethanol and Freeze precipitated. But I can almost guarantee it won't work with sodium carbonate. Maybe if more so. car. was dissolved, and NaCl also used to help precipitation?

Or even a sodium carbonate/NaCl solution that has enough salt so as to keep from freezing?

Would be interested to know what the pKa of freebase DMT is.