how to separate gamma-butyrolactone from N-methyl-2-pyrrolidone

Since these two chemicals have different freezing temps (pyrrolidone is $-11~\mathrm{^\circ F}$ [$-24~\mathrm{^\circ C}$] and $\unicode[Times]{x3b3}$-butyrolactone freezes (also called melting) at $-46~\mathrm{^\circ F}$ [$-43~\mathrm{^\circ C}$]), could a person wanting to separate these just immerse them in liquid nitrogen (or something else?) to get the desired chemical to freeze, then filter off the unfrozen butyrolactone? Other routes seem quite difficult for a non chemist. Any objections or other routes? Anything you can add? Even a step by process will be much appreciated.

• I do wonder why you are interested in the separation of GHB as a non-chemist. – Jan Oct 16 '15 at 17:19
• Can you add info on the relative fraction of each in the mixture as well as which one you want to recover pure? Both of those factors have significant impact on what would be a useful method to apply. – Andrew Feb 2 at 14:22

Cud [sic] a person wanting to separate these just emerse [sic] them into liquid nitro [sic] or any [sic] to get the desired chemical to freeze then filter off the unfrozen butyrolacton [sic]?

With liquid nitrogen this would not be possible because liquid nitrogen condenses/boils at -195 °C. It would freeze both the pyrrolidone and the lactone.

• Well. Then say what would a person use to separate these two compounds that would not require distillation and adding of chemicals. Freezing the undesirable must be possible ? – josh Mar 8 '15 at 22:58
• I'm not sure which compound (both?) you need to recover. If you want to recover just the NMP, then you could probably separate it from the GBL by hydrolyzing the GLB with a strong base, for 4-hydroxybutyrate salts, and then doing a solvent extraction with chloroform or hexanes to separate the NMP from the GBL-derived salts. Then the NMP can be re-isolated by evaporation of the chloroform or hexanes. – Curt F. Mar 9 '15 at 0:20
• My answer originally said LN2 "freezes" but of course I meant "condenses". I'm talking about the boiling point not the freezing point so I am editing my answer. – Curt F. Mar 9 '15 at 19:57

As Curt F. has already pointed out, liquid nitrogen is an overkill and won't help here.

You might want to try one of the following cooling mixtures - values are taken from (http://www.chemieunterricht.de/dc2/tip/08_98.htm):

• 81 g $\ce{CaCl2*6H2O}$ + 100 g ice : -21,5 °C (this might not be enough)
• 66 g $\ce{NaBr}$ + 100 g ice : -28 °C
• 105 g ethanol + 100 g ice : -30 °C

to enrich NMP in ther solid fraction.