Some of the freebase is dissolved in toluene/xylene and its dryness asserted by boiling the solution for a moment. The gassing setup generates HCl from H2SO4 and NaCl and leads it through a H2SO4 bubbler, a CaCl drier and finally into a gas dispersion tube which is immersed into the solution. This worked once and gave hydrochloride crystals which could be filtered over a buchner. They were indeed found to be vastly hygroscopic, on the first attempt they drew water from the air and they were sucked through the buchner within a minute. Another attempt succeeded when the buchner was covered with plastic wrap which vacuumized and compressed the filtercake. The cake was then further dried to absolute dryness in a desiccator. However, numerous attempts at gassing already failed at the gassing stage. At the beginning of the gassing process a precipitate is visible for a brief moment, but then the solvent/base mixture seems to absorb more HCl than necessary to give the salt, resulting in a fuming acidic solution which gets a red tint more often than not. At this point it was never possible to get it to precipitate back, not by freezing and not by evaporation. Evaporation in this case seems to yield decomposition products. Sometimes they have a ketonic smell to them, a slight hint of benzaldehyde and clearly there is something which seems to boil higher than even the xylene (hydrochloride melts at 152°C, so when we're boiling off something with a BP > 152°C the salt is in the melt). Therefore I recognized the gassing procedure as too unreliable. The only clue why this failed numerous times is that the HCl gas wasn't yet dry enough which could yield a "super-acid" (w/w conc. > 37%) which hydrolyses the amphetamine and yields decomposition (or chlorinated) products which could have further absorbed HCl and/or irreversibly dissolved the hydrochloride salt.
