Hi

I need some input from people who have done acids analysis by GC. I have a product stream that will contain water, cyclohexane/ol/one, adipic acid, succinic acid and maybe a few other acids - plus solvent. I've read of examples where
a) they esterify the acids before GC analysis (yuck)
b) they use ion chromatography for the adipic

Can someone suggest a column that might do this separation directly? Is column degradation from the acids the problem here? I see that agilent sells wax/ FFAP columns that separate adipic - but perhaps they are not robust? why bother esterifying?

thanks!
karin

Most GC column stationary phases are nonpolar, and acids are obviously polar compounds. If you try to analyzed acids on a nonpolar column, DB-1 or DB-5, you will have poor peak shapes which can be difficult to integrate. The peaks are usually triangular (shark fin), and often get wider with increasing concentration, not taller. If the acids are esterified, the polarity is lower, the volatility is often higher, and the chromatography is better.

Esterification is not that difficult, and it is the best way to get good chromatography.

I have used the FFAP columns for some analyses of low molecular weight organic acids (I'm talking less than 200 MW here), and it worked well for the purpose. Generally, the more polar the stationary phase, the less stable it is. There was significant instability when these columns were first developed, which resulted in the lack of attention to further refinement, especially when compared to DB-1 or DB-5 columns. With some refinements in the production process, columns manufactured today are generally more stable than earlier productions. However, I think the current upper temperature limit on a DB-FFAP is still limited to 250C. I can't comment on long term stability, as I used the FFAP column on analyses that were fairly infrequent (a few samples every couple months). If I remember the chromatogram correctly, there were some problem separations with some of the components that were not acids. This was not a concern to me at the time as I was not interested in everything in the sample, but it may pose a problem if you have a number of nonpolar analytes in the sample that are of interest to you. You might still be able to get away with a good separation if your sample isn't that complex.
I also agree with Ron. Esterification adds an extra step, but the process then makes the sample amenable to analysis with the more stable DB-1 and DB-5 columns. If your analysis is going to be a routine one, I would go with esterification.