Sciencemadness Discussion Board

Steamdistillation of GBL

Organikum - 24-1-2007 at 11:58

GBL readily distilles over with steam, thats wellknown. But how is the ratio, say how much percent w/w is GBL and how much water with a steamdistillation at atmospheric pressure?

Couldnt find this. Anybody knows?

And when we are at it: The best way to extract the GBL quantitativly from the water? Distilling off an azeotrope of benzene or chloroform/water comes to mind but that would be pretty tedious with the amounts of water involved.

Any suggestions how to do it or where to look?

regards
/ORG

Nicodem - 24-1-2007 at 12:12

Quote:
Originally posted by Organikum
The best way to extract the GBL quantitativly from the water?

Perhaps saturating the solution with common salt, extracting with CH2Cl2 a couple of times, drying over Na2SO4 (or MgSO4) and vacuum stripping off the solvent (vacuum distilling the remaining liquid, preferably with a column, would be in order to remove the solvent remains and other potential impurities).

Misanthropy - 26-1-2007 at 11:42

Would N-Methyl-2-pyrrolidone come over so readily as GBL this way when in the mix? I think so.

Also, what exactly are the byproducts of NaOH+NMP (at ~200C or however hot the usual equimolar NaOH/GBL becomes)?

[Edited on 26-1-2007 by Misanthropy]

Nicodem - 26-1-2007 at 12:09

I don't think NMP comes over with steam at any appreciably extent. It certainly can not steam distill since it is miscelable with water in all ratios. I'm not aware if it forms any azeotrope with water like gamma-butyrolactone does. There can be no steam distillation effect in the distillation of monophasic liquids, there are only special cases of azeotropes that do not conform to steam distillation rules.

At very harsh conditions, NaOH should hydrolyze NMP to sodium 4-methylaminobutyrate.

Misanthropy - 26-1-2007 at 12:14

Thank you. :-)

Furch - 27-1-2007 at 17:16

A little off topic perhaps, but concerning the separation of GBL and NMP, I've learned that it can be done by refluxing concentrated NaOH (20-25%), which hydrolyzes both substances as mentioned above, and after this, the mixture is made slightly acidic. This will cause the Na-GHB to esterify again, but not the 4-methylaminobuturate, as amides are not as easily formed as esters... Even if it's an intramolecular reaction.

Having done this, one should have a mixture of sodium 4-methylaminobuturate, GBL and water... Just distill the off the GBL.

Maybe you had already figured this out by yourself though... In that case, sorry for being redundant.


- Furch

unionised - 28-1-2007 at 05:14

"It certainly can not steam distill since it is miscelable with water in all ratios."
Pardon?

Nicodem - 28-1-2007 at 09:27

I was referring to the basic difference between the steam distillation and azeotropic distillation. The ideal model of steam distillation requires the two components to be completely immiscible. In such ideal biphasic model, one can calculate the exact composition of the distillate. Of course, practical models are not ideal since usually the compound to be distilled is at least partially soluble in water and generally interact with water molecules to some degree. But in the case where the mixture to be distilled is monophasic (like GBL/H2O), one can in no way talk about steam distillation but should talk about azeotropic distillation instead. I realize this makes little difference to anybody here, but when trying to calculate the boiling temperature or the distillate composition using steam distillation theory, it does make a lot of difference.

chemrox - 30-1-2007 at 00:11

What is the role of alkali metal ion in GBL?

unionised - 30-1-2007 at 12:24

"What is the role of alkali metal ion in GBL? "
Gamma butyrolactone C4H6O2 doesn't have an alkali metal in it to have any role.

AFAIK, it's only azeotropic distillation if you distill an azeotrope.
I'm pretty sure that if you boil a mix of NMP and water at least some of the NMP will distill over with the steam and I'd call that steam distillation (though I accept this is probably a matter of deffinition). I don't know if NMP forms an azeotrope with water but I will take your word for it that it doesn't so I can't see how you can call this azeotropic.
I realise that the miscibillity makes life rather difficult when it comes to doing the calculations, but I could still make an educated guess at what you would get by distilling a mixture of water and NMP. For a dilute solution the boiling point would be near 100 C and I think that gives a vapour pressure of NMP of about 26mmHg (data cribbed from the net). With that and Raoult's "law" you could make some sort of guess. Clearly this is much more difficult to calculate than the 2 phase system and it almost certainly gives a worse yield than a comparable 2 phase system.
BTW, does gbl disolve in hot saturated salt solution?
If not then I think you will get a better yield (ie less water in the product) by saturating the solution before distilling.

Organikum - 2-2-2007 at 01:02

Nicodem is of course right, its an azeotrope not a steamdistillation. For some reason I dont know I carry this assoziation GBL<->steamdistillation in my head for quite some time and I just never came to question it. Thanks for putting me right.

In the GABA -> GHB synthesis as on Rhodiums it is told that the azeotrope GBL/water is about 1/10. But it is unclear if he refers in this case to GBL or to the final Na-GHB.
Doesnt anybody have an azeotrope database which includes GBL and can look this up? I have not or have not found it in my resources.

Nicodem: I tried salting out GBL from water with NaCl and it didnt work. It is possible to get two layers as long enough Na-GHB is present but with only tablesalt not. But the general approach is promising I will post a procedure as soon I have worked it out.

thanks so far.

PS:Is NMP a health risk anyways? I read it being useful as food additive for clearing beeer and stuff?

Furch - 2-2-2007 at 11:51

Organikum,

I've got no data to back it up, but somehow I get the feeling you should try salting out with an acidic salt instead of a neutral one. Perhaps NaHSO4 or the like. This would, naturally, push the equilibrium towards GBL, thus prohibiting it from forming GHB in the water solution which it certainly does, to a considerable extent at neutral pH (whether it does in brine I don't know though). I can't help but feel that the GBL <-> GHB equlibrium is somewhat responsible for GBL's miscibility with water.

All I can say about NMP is that over time you will become pretty delirious from having it in your blood stream... So I personally definitely consider it to be a health risk, if continuously exposed to it for a longer period of time. I've tried to find information on its potential toxicity too, but found none... However considering its natural GABA similar metabolite I think it's pretty safe to say that it somehow interfers with the natural GABA transmission.

I hope you don't find my reflections too reduntant, and pardon me for being somewhat subjective in my motivations ;)

Sincerely,
Furch

Nicodem - 2-2-2007 at 12:28

Quote:
Originally posted by Organikum
Nicodem: I tried salting out GBL from water with NaCl and it didnt work. It is possible to get two layers as long enough Na-GHB is present but with only tablesalt not. But the general approach is promising I will post a procedure as soon I have worked it out.

I was not really expecting it would form two phases by saturating with salt. I was more like suggesting making the GBL in the distillate extractable with CH2Cl2 by NaCl saturation. Plain extraction of such a diluted aqueous GBL with CH2Cl2 would be very inefficient and would require numerous extractions. If the ionic strength of such a solution is increased (by saturating with NaCl) then the least polar component (GBL) is way more easily extracted into the non polar solvent (CH2Cl2 or ethyl acetate). In other words you increase the logP of GBL in your biphasic system. The logP(octanol/water) for GBL is -0.64, meaning that in two equal volumes of water in octanol, a sample of GBL will partitionate so that the concentration of GBL will be 4.36 times higher in the water phase. Even though CH2Cl2 and ethyl acetate should have a considerably better partition coefficient for GBL you can still see that it would take too many extractions to extract a useful amount of GBL from its water solution. You would need a continuous extractor or hydrolyze GBL with NaOH, evaporate the water, acidify with an equivalent of H2SO4 and distill off the reformed GBL or improving the partition coeficient by saturation with salt as suggested.
Quote:
PS:Is NMP a health risk anyways? I read it being useful as food additive for clearing beeer and stuff?

As you can check in the MSDS for NMP, its toxicity is pretty low. In the body it should get hydrolyzed to the relatively innocuous N-methyl-GABA and GABA which have bioactivity parallel to some of GBL's bioactivity anyway. NMP's toxicity is probably more related to high dose interference with the liver or the like, so I would assume a couple of mg/kg causes no harm whatsoever. It is anyway several times less toxic than GBL .
Unless NMP forms an azeotrope with water or GBL you can be sure that its amount in the distillate will be irrelevant (I never ever heard of such an azeotrope – which however does not mean it does not exist!). If you calculate using the Raoult's law (normal distillation rules), like Unionised suggests, you can see that there will be very little NMP distilling over, certainly negligible if you use a moderate size distillation column.

Nicodem - 3-2-2007 at 03:55

Quote:
Originally posted by Furch
I've got no data to back it up, but somehow I get the feeling you should try salting out with an acidic salt instead of a neutral one. Perhaps NaHSO4 or the like. This would, naturally, push the equilibrium towards GBL, thus prohibiting it from forming GHB in the water solution which it certainly does, to a considerable extent at neutral pH (whether it does in brine I don't know though). I can't help but feel that the GBL <-> GHB equlibrium is somewhat responsible for GBL's miscibility with water.

As with all esters there certainly exists such equilibrium, but both its kinetics and equilibrium constant prevents it to be either fast or leaned toward GHB. The kinetics of esters hydrolysis in neutral pH at room temperature are slow for most esters and since GBL is a lactone its equilibrium constant greatly favors GBL over GHB unless the pH is basic (the deprotonated GHB has no electrophilic –COOH group anymore and thus can not cyclisize anymore).
The issue here is about the logP which indicates the partitioning equilibrium for GBL in the extraction process. Its logP is very low (-0.64). This is not because of any such equilibrium as you say, since 4-hydroxybutanoic acid might have a higher logP given that it is not cyclic anymore even though it gains two H-bond donor groups (the logP calculated with Chem3D is -0.32; with KowWin is -0.4; with MDL QSAR is -0.78). Namely the reason for the low logP of GBL is the classical one thought in the school with the example of THF vs. diethyl ether. THF is fully miscelable with water while diethyl ether is not even though they are nearly identical by molecular weight and chemical type (THF has logP=0.53; Et2O has logP=0.85). The difference is in the exposed lone electron pairs of THF which are therefore more prone to H-bond with H2O molecules. The same analogy could be drawn from comparing GBL (logP=-0.64) vs. ethyl acetate (logP=0.68). The difference (almost 10-times lowering of the partition coefficient) is even much more striking given that GBL has two H-bond acceptors.

[Edited on 3-2-2007 by Nicodem]

Organikum - 4-2-2007 at 02:37

I was well aware that the NaCl was only to aide extraction, but when two layers formed as long Na-GHB was present is large amounts I got this idea that salting out with a more hygroscopic salt might be an option. Epsom salts for example.

The Ullmann states that the equilibrium of GBL/GHB is 100%/0% at 0°C and 80%/20% at 100°C. Somewhere else it is told that the equilibrium depends on the amount of water present and that an acidic environment shifts it to GBL.

It definitivly takes some time until the equilibrium is established. After dissolving some pure GBL in the tenfold amount of tapwater the taste stays unchanged for several hours at roomtemperature, overnight the taste changes though - and not to the better :o

A personal remark: Despite what is written allover the internet GBL is as drug preferable over GHB and be it alone for the immense amounts of sodium coming with the GHB. Everybody who wants to try GHB (as a salt) is strictly advised first to check if he/she is not salt-sensitive. 30% of the population are (as am I) and if so, Na-GHB is a VERY bad idea, also a mixture of K and Na-GHB makes no difference then.

The experiments here aim at GBL as a solvent though, gladly high-purity GBL is available in the EU, but at a price.

/ORG

Nicodem - 4-2-2007 at 13:32

Quote:
Originally posted by Organikum
I was well aware that the NaCl was only to aide extraction, but when two layers formed as long Na-GHB was present is large amounts I got this idea that salting out with a more hygroscopic salt might be an option. Epsom salts for example.

That is not a bad idea and I was about to suggest CaCl2 in my previous post, but then changed my mind due to the irrationality of using anything else but the cheap table salt. But if you have lots of CaCl2 to play with then use it since it possibly the best almost neutral and cheap salt that can hydrate enormous amount of water molecules in its solutions (unlike MgSO4 which is five times less soluble and hydrates less when in solution).

Misanthropy - 9-2-2007 at 22:52

If of any interest; references from the DEA's Microgram publication regarding GBL solutions... from here: http://www.dea.gov/programs/forensicsci/microgram/mg0304/mg0...

# Chappell JS. The non equilibrium aqueous solution chemistry of gamma-hydroxybutyric acid. Journal of the Clandestine Laboratory Investigating Chemists Association 2002;12(4):20.

# Ciolino LA, Mesmer MZ, Satzger RD, Machal AC, McCauley HA, Mohrhaus AS. The chemical interconversion of GHB and GBL: Forensic issues and implications. Journal of Forensic Sciences 2001;46(6):1315.

# Ciolino LA, Mesmer MZ. Bridging the gap between GHB and GBL: Forensic issues of interconversion. Proceedings of the American Academy of Forensic Sciences 2000;6:44.

Maja - 25-9-2011 at 04:41

So after sandmeyer reaction is done with gaba you can add like a few liters of water and distill out all the in 10% GBL / 90% water mix without any tedious/expensive/toxic solvent extractions ? What would you suggest for purifying it ? AFAIK some nitrosamines with various nitrogen oxides are formed in reaction as byproducts ? How pure GBL distilled out in azeotrope with water straight from post-reaction mixture is ? Suggest please.

aaparatuss - 25-9-2011 at 12:41

It has been shown that adding the molar equivalent of KOH in minimum water to the post reaction and then boiling it dryness is rather direct and loss limiting.

the mixed chlorides and butyrate can be separated with methanol

synsol - 12-7-2020 at 15:22

Is there more information on this somewhere? Or can someone elaborate on the use of KOH post reaction and subsequent seperation by methanol?