I have what I consider to be a good intuitive understanding of why these solvents work together and understand completely why VideoEditor wants to heat to 105°C. While this method does precipitate the pseudo HCl in a solution of naptha and xylene, remember that the gakks are present in the solution with the pseudo.
In the microwave xylene clean technique, I was polishing up some previously extracted pseudo HCl. The bulk of the gakk was gone, and the method was used as a substitute for the first of two recrystalizations. The xylene could hold the gakk because there was not much gakk in the mix to start with. I cautioned that it worked with pseudo that was fairly clean already.
The other point to keep in mind is when the pseudo precipitates out of the solution. It precipitates out before anything else does. That means if there is a bigger gakk load, you want that gakk held in solution as long as you can. Keeping the solvent hot does that. VideoEditor figured out at by the time the naptha/xylene mix got to 105°C the alcohol would be gone but the solvent hot enough to hold the gakk in solution while the pseudo HCl was filtered out. You MUST filter while HOT. Get the pseudo out of the solution before gakk starts coming out, and rinse what might be in there with MEK.
This technique seems to me to be very well thought out. It seems to work very well. I really suggest to each person interested in this method to do it as written a few times and get a feel for it before you attempt to improve on it.
There is, in a way, a philosophy to this method that appeals to me-- a little bit of Zen, perhaps.
I have been fooled before with new techniques and discovered after my inital enthusiasm that they were not quite as effective as I first thought. Straight to Bee was like this-- PEG was still present in the pseudo unless it was taken out before the a/b. The tyvek membrane microwave extraction seemed very promising but far from universal. PEG in particular seemed to come over with the pseudo-- and PEG at the time was the new problem in the pills. The high yield of pseudo from this method gives me some pause and will until we have final results to compare. That is not a knock on this approach at all. Its is perhaps caution based on past experience. My hunch is this one works as advertised. I think we will know soon enough. I also think we should at least find out if it works as written before we start modifying it.