Headpace-methylation will H2SO4 in vial damage GC?

[MSCHemist]

Hi all

I have been reading about in-situ esterfication via headspace GC (free acids). It looks very useful. The typical protocol is to use either NaHSO4 or H2SO4 and add the alcohol you wish to esterfy with and do HS-GC. I am planning to use total vaporization HS GC/MS.

http://www.umb.no/statisk/nordost/ulf.pdf

This article says I can use 1 N H2SO4 (actualy 0.4N final) in the headspace vial. Do you think H2SO4 can damage the instrument? It also seems the esterfication proceeds in the presence of water (unlike fames).

This is the original reference
http://www.sciencedirect.com/science?_o ... xcerpt.pdf

[rb6banjo]

I'm surprised that the reaction would go in an aqueous matrix. In esterification reactions, water is a product. According to LeChatlier's Principle, that's a bad situation as having a "boat load" of reaction product present will push the reaction back to starting materials. I'd have to see it work to believe it.

[MSCHemist]

Well I intend to try it. It doesn't look like having 0.4 or so molar sulfuric acid in the vial should get into the instrument or anything but when I read the article about having concentrated acid in the headspace vial it made warning bells go off in my head. That should be more a problem with HCl than H2SO4.

Bruno Kolb even illustrated it in several places in his headspace book so it has to be a fairly reputable procedure.

[rb6banjo]

Excellent. With static headspace you won't likely contaminate your system. I am curious as to whether it will work reliably. Please let us know how it turns out.

[MSCHemist]

I tried it with the total evaporative technique

I made a 1000 ppm standard of lactic, succininc, fumaric, citric, malic, tartaric, and ascorbic (long shot) in water.

Then I added 2.5ul standard, 2.5ul of 1N H2SO4, and 2ul methanol in a 9.2ml headspace vial and did TV at 120 deg C. I saw succinic and fumaric but no others. Overnight I am going to try 85 deg standard headspace method with 1ml standard, 1ml H2SO4 and 0.5ml methanol as in the website.

Since I work in a flavors company there should be some bisulfate arround all also try it with that.

[Peter Apps]

Surely you need to add a large excess of methanol and lots of acid (or something else to suck up the water) - you have to drive the equilbrium towards esters.

Peter

[MSCHemist]

The method says 1 part sample, 1 part sat NaHSO4, 0.5 parts MeOH.

[MSCHemist]

SO far not great success. I was able to see lactic, succinic and fumaric using 1N H2SO4. This method looks like it was deloped for microbial identification in broth cultures by their acid production typically for very high levels. I wonder if it is kind of a hack method. even though the esterfication is very inefficient they go with it anyways. SOme of the chromatograms on the first link have butric acid alongside methyl butyrate. Still Kolb illustrates it in several different places in his book.

I currently use isobutyl chloroformate for fatty acids as only isobutyl acetate from acetic acid elutes clear of the reaction pyridine then ethyl chloroformate for fumaric, succinic, lactic, malic, and citric as di and hydroxy acids behave badly with isobutyl chloroformate because of the poor solubility of the isobutanol in the aqueous mix. They have a lot of side reaction with ECF as well and I can't do tartaric at all, citric poor repsonse, and fumaric isomerizes about 50% to maleic. It would be nice to have one method to do all acids together.

Now that I have had a chance to read it better it seems it is a slow reaction and should be allowed to thermostate 45min minimum preferable 1.5 hour. I'll do some more work on it next week. Kolb says it is fairly high yeild >90%.

[Peter Apps]

Maybe the total evaporation is the problem - do there reactions work in the vapour phase ??

Peter

[MSCHemist]

That is what I thought as well. Why add a nonvolatile catalyst when the analytes and solvent are in the vapor phase but all the articles claim TV is the prefered method especially for the more nonvolatile esters. I'll try it some more next well.

[MSCHemist]

I played with it a bit today. I made ~12% H2SO4/methanol solution from methanol and 70% H2SO4 and got real nice peaks for lactic, fumaric, and succinic acids, an infinatesmal peak (like 2-3% the area of the others) of malic, one of them forms 2-methyl furoate I would suppose the ascorbic acid, and no sign orf tartaric, citric, nor ascorbic. I am not quite sure the reason for the specficity for those acids (why lactic and not citric nor malic).

I think I am going to end up sticking with the ethyl chloroformate despite its flaws.

[Peter Apps]

I played with it a bit today. I made ~12% H2SO4/methanol solution from methanol and 70% H2SO4 and got real nice peaks for lactic, fumaric, and succinic acids, an infinatesmal peak (like 2-3% the area of the others) of malic, one of them forms 2-methyl furoate I would suppose the ascorbic acid, and no sign orf tartaric, citric, nor ascorbic. I am not quite sure the reason for the specficity for those acids (why lactic and not citric nor malic).

I think I am going to end up sticking with the ethyl chloroformate despite its flaws.

Good to know that it works for some of them. Was this still with total evaporation ? - could be something to do with vapour pressures of the acids ?

Peter

[MSCHemist]

Yep I use 9.2ml headspace vials on my Tekmar 7000 so I add 5ul of aqueous standard and 5ul of the methanol/H2SO4 and did 1 1.5 hour incubation at 120 deg C with the valve oven and transferline at 150 deg C. Kolb spends about a page talking about it in the section on derivatization headpspace. The only other method he mentions is dimethyl sulfate which I don't really want to have around in my lab.

I'm always looking for a simple esterfication technique that is repeatable, cheap, the derivatization reagents don't go bad, and does all the nonvolatile acids, and is ideally water insensitive. So far ethyl and isobutyl (for acetic propionic and butyric) chloroformate are the only things I've found that fit the bill. Some work with heptafluorobutyl chlroformate looks promising but the reagent is not available commerically yet at least not from any vendors I've looked at.

[RogerBardsley]

Slide 5 has a DMF peak. I did not see DMF mentioned in the sample prep. Does anybody have an idea where the DMF is from?

[MSCHemist]

I tried it today using volatile fatty acids and isobutanol (I had a standard in isobutanol for isobutyl chloroformate) and it worked well no sign of unreacted acids. I am now going to try it using methanol and cranking up the headspace temp to 185 as I saw reduced area counts after valeric though I was able to see it all the way to myristic.