Fatty acid esterfication and aqueous samples

[MSCHemist]

I am not sure if this should be here or sample prep but here it goes.

I finally found a universal esterication tecnique. The trouble is the esterfication is water sensitive as always though somewhat tolerant to small ammounts of water.
I am using the mild methanolysis in this article
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2817593/ only I am using isopropanol for isopropyl esters rather than methyl esters as the small acids like acetic are less volatile and the isopropyl esters can be done on a db-5 column and not tail badly like I saw with the smaller and hydroxy methyl esters. I've already tested it on lactic, citric, succinic, fumaric, malic, and tartaric and it does them too.

A brief summary of the protocol: 15% HCl in isopropanol is prepared from 35% aqueous HCl and isopropanol. In the reaction mix 2 ml (0.3 ml of the 15% HCl and 1.7ml of sample/standard in isopropanol) the final water content is only 2% so it is insignicant. I heat it in a closed tube for 2 hours at 100 deg C. Then I add saturated NaCl with saturated NaHCO3 to neutralize the acid. I then take the Isopropanol layer and inject it.

What would you suggest for aqueous samples? I can't blow it down or I'll loose the small volatile acids. I could perhaps acidify and salt them into isopropanol though I'd probably have low recovery of the hydroxy and diacids.

[Consumer Products Guy]

Occasionally we've made IPA, n-propyl, and butyl esters of fatty acids.

We've also made methyl esters from liquid soap consumer products, which may be 75% water, without noticing any issues.

[MSCHemist]

How much water do you end up with in the final reaction though? The linked paper that I am using noted decreased reaction efficiency when they used higher ammounts of HCl presumably due to increased water content from the original 35% HCl.

[Consumer Products Guy]

How much water do you end up with in the final reaction though?

I have no real idea. Typically we esterify with 10% sulfuric acid in methanol or similar level of BF3-methanol.

Typical for us for liquid soaps (not the same as liquid "soaps" that are actually synthetic anionic surfactants) would be a couple of grams sample in a 100 ml volumetric flask (so at least 1 gram of water), about 25 ml of the methanolic reagent, steam bath 5 minutes, allow to cool, then add a few ml of extraction solvent such as hexane and swirl, then add saturated sodium chloride to part-way up the neck, stopper and shake. Allow layers to separate, and withdraw some of the top into an autosampler vial. Sometimes the hexane layer will contain suspended stuff, so OK to filter it.

[Peter Apps]

Making methyl esters of free acids in wild dog pee and anal gland secretion I got much better results with concentrated sulphuric acid as catalyst than with 35% hydrochloric. I put this down to the sulphuric containing less water in the first place, and being a dehydrating agent in its own right. This was with extraction into hexane.

The extraction into hexane might be significant - MSChemist is injecting the isopropanol layer, which with a wet sample probably contains some water - maybe enough to hydrolyse some of the esters when it hits the hot inlet ??

Also MSChemist neutralises the acid afterwards, and CPG and I did not.

Peter

[MSCHemist]

I am concerned about injecting HCl into the GC which is why I am adding 1ml of water saturated with bicarb and NaCl. Because of isopropanol's broad solubility I can't extract with hexanes or anything else. I am going with IPA as methyl esters are to volatile for the short chains fatty acids and to polar to do on a db5 or db1 column (a lot of the small ones and hydroxy acids tail badly with methyl esters).

I am using 35%HCl and in the final reaction mix I am taking it down to 1.2% which leaves ~2.2% water.

I might try sulfuric acid in place of HCl.

The other issue I have is tartaric acid cannot seem to be coaxed to dissolve in isopropanol. Citric takes some heat to do so. So I had to make those standards in water but I made them at 10% so I take 100ul and add it to the 10ml reaction vial of isopropanol/HCl. I am seeing up to 10% free fatty acid mainly of branched chain acids (they seem to be harder to esterfy).

Yet another issue is the procedure saponifies so it is not a free acids method per se.

Overall I like my original method of using isobutyl chloroformate to esterfy the simple acids as it is a phase transfer reaction and is insensitive to water. The main difficulty is pyridine is necessary as a catalyst and isobutyl acetate is just barely past it so there are sometimes peak shape artifacts (broadening). Also there is a huge peak in the middle of the chromatogram of disobityl carbonate where NaHCO3 is added to neutralize the isobutyl chloroformate.

I like this HCl/IPA procedure because it does the nonvolatile fatty acids including the hydroxy acids which so far is the only method that does tartaric and citric well.

I like the HCl/methanol version of the procure for FAMES for fat analysis.

I like the headspace esterfication method for the very short chain fatty acids acetic and I haven't really tried formic. Shimadzu is having a poster of it at Pittcon so I will check it out.

So I may just keep my isobutyl chloroformate method for free fatty acids and Use the isopropanol HCl for the Kreb's cycle acids.

I just thought it would be nice to have a do all the carboxylic acids at once protocol.