elfspice-- This method works because at this temperature about the only thing that has precipitated in a filterable form is the pseudoephedrine HCl. Drop the temp ten degrees and gakks start precipitating too. Most of the inactives present and filterable or that attach to the pseudo rinse off with acetone or MEK rinses of the pseudo. If you do not filter the solution hot, you will have more than pseudo in the mix. Filtering while hot has a lot to do with this method working.
VideoEditor settled on the 105C temperature by experimenting and finding his pseudo was cleaner when filtered at that temp. At that temp the solvents xylene and naptha are still holding some inactives in solution which fall out at lower temperatures. You see them fall out of the solution as it cools. At 105C the pseudo has preciptiated in the solvents, but the others that will have not.
If you bring the solution up to temperature too slowly, either by not using enough heat, or by using too small a hotplate to get the volume of solvent hot enough fast enough, you will have an excess of naptha evaporate before it reaches its boiling point and before all the alcohol has evaporated. I noticed this when I tried to increase the size of the extraction and it was taking too long. The volume of solvent was far too small when I was through, and I found PEG in my pseudo. Hot naptha will dissolve PEG from the pill mass, but if you evaporate too much of the naptha before you precipitate out the pseudo, it has to go somewhere. Some of it goes with the pseudo. Boil the alcohol off too slowly in this method-- by not using enough heat to bring the temperature to 90C fast enough-- and the volume of naptha can drop too low before the pseudo precipitates. Naptha evaporates pretty fast even when it isn't at it boiling point, and when you simmer a large extraction by this method over too small a hot plate, enough naptha can evaporate to leave the solution too crowded with other gakks to contain them all. Adding room temp naptha slows down the boiling off of the alcohol, but gives naptha a friendly place to hide while the pseudo precipitates as the alcohol boils off. Using the right size hotplate, or adding more of the alky/natpha/pseudo solution to the pot with the xylene as it heats up may solve the problem.
I have had this happen on two or three occasions; each time I had a contaminant in the pseudo I did not have when I brought the temperature up faster. when I got peg in the final product I realized the reason was that I let the fluid volume of naptha fall too low to contain it even at temperature. That at least is the way it seems to me. I noticed the volume of the xylene/naptha remaining looked too low, and noted the time it took to get the solution hot enough to drive off the alcohol was common to all such less-than-clean extractions.
If what you filter out of the solution is a mix of crystals of different substances, saving the alcohol has been a false economy.
There is likely a good use to be put to the poor solubility of really cold alcohol. I am not sure this is it.