I tried to purify IPA contaminated with HCl, xylene and water.

These are the actions I took:

1) neutralization with excess sodium carbonate
2) filtering
3) simple distillation
4) drying with magnesium sulfate (this should remove water and break the possible IPA-xylene azeotrope, right?)
5) fractional distillation (this should separate IPA and xylene)

Unfortunately, in the last step, the thermometer shows constant 79.6 °C (slowly raising to 80.0 as half of the alcohol is distilled) instead of 82.6 °C, which would be the correct b.p. for IPA.

The product still somewhat smells of xylene even though I am using very long Vigreux column and removed the initial low boiling fraction.

Is there a way to remove the xylene? I've read that aromatic hydrocarbons (BTX) can be destroyed by sulfonation, i.e. adding conc. sulfuric acid, but I haven't found a proper way to do it and I don't want to cause explosion...

Another tips I found is to treat the alcohol with sodium bicarbonate and activated charcoal to remove smells. This works for ehtanol, I am not sure about IPA though...

Sorry for bothering with this, as this is probably unrelevant to Nexus, though some people who might have mixed they NPS with alcohol in some extraction might find this useful.

I will try Sciencemadness as well, I really should finally create an account there...

That's a bit tricky. You might be able to distil some of it out by adding ~50% water and pulling it out as a ternary azeotrope but I haven't looked up the data for that yet. The xylene would co-distil with a fair proportion of the IPA but you might be left with 91% IPA in the distillation pot.

A brief scan of the data suggests adding a dash of formic acid may help.

Xylene has a very strong odour so it's likely there's only a tiny amount present - ~0.01% ?

Thanks for the insights. I looked for the azeotrope over and under and it seems maybe only m-xylene forms a ternary azeotrope (I don't remember the source).

I consulted Sciencemadness forum and they said it should be possible to separate the two components with very careful fractional distillation (which is what I did, yet IPA still smells). There will probably be always some xylene as it might get easily volatilized and thus pass through the apparatus with IPA. And even trace amounts, as you said, would still make the alcohol smelly.

All in all, I probably won't mix IPA and xylene in the future. Or any alcohol and nonpolar solvent, hopefully.

It was of my insterest since this mixture seems to be commonly used for MDMA precipitation (see BrightStar's synthesis). Not that I am intersted in it, but when I did the re-extraction of MDMA from blueberry juice, I used the same final step (precipitation of MDMA.HCl from IPA/xylene mixture). I thought this would be a nice way of purification M from lots of adulterants (basically a mini A/B re-extraction) but having two dirty and hard-to-purify solvents in the end might not be the way - except maybe if they would be used solely for the purpose of MDMA purification. The acetone/IPA method (BOOFF tek) seems to be much more practical.