I am trying to separate and detect small mono carboxylic acids which also have keto groups (pyruvate, hydroxypyuvate and glyoxylate). I am derivatising the keto groups with phenylhydrazine to enable UV detection. I am trying to adapt from methods published in the literature, which use RP C18 with 5% ethanol in 25 mM potassium phosphate buffer pH 6. At higher ethanol concentrations the metabolites are not retained long enough and I have tried varying the pH between 5.8 and 7 to improve resolution with no success. Any suggestions?

Most carboxylic acids have pKa around 4-5. If you want to increase retention try a mobile phase with a pH around 2. This will protonate the carboxylic acids and sould increase their reversed phase retention. TFA or a phosphate buffer should work.

You might also want to try some of the new "aqua" columns that contain an embedded amide group. Some of these are more retentive for small organic acids. They can also be run at 100% aqueous for poorly retained acids. Good luck.

Emma, try Tom's suggestion first. Alternatively you can use ion-pairing chromatography with 10-50 mM tetrabutylammonium sulphate as ion-pairing agent, in a 10-20 mM phsphate buffer of pH=6.0 - 7.0, C18 column and Methanol as organic solvent.
I've used it for separation of malic, maleic and fumaric. The problem is the usual one for rp columns with ion-pairing mobile phase, column selectivity is irreversively modified so that you shouldn't use it for other applications.
Good luck

Bio-Rad used to market a range of resin-based ion-exchange type columns for this sort of analysis. I think some worked on the principal of classical anion exchange and others on the principle of ion exclusion. The eluents necessary to elute organic acids were in some cases just simple dilute inorganic acids enabling at least some organic acids to be detected with UV at low wavelength (eg 210nm) without any derivatisation, although I don't know what sensitivity is required for your application. It might be worth checking this method out.

Hi, David:

What is the principle of ion exclusion and its application area? Could you or someone else explain it a little bit for me?

Many thanks in advance.

You can check either BioRad or Dionex web sites for more detail (in Dionex-speak, it's called "ion chromatography exclusion"). Most textbooks on ion chromatography will also have a discussion of the technique.

Briefly, though, the principle is that a strong acid cation exchanger essentially has fairly concentrated negative charge *inside* the resin beads. If you introduce a mobile negative charge (sample anions), it will tend to stay outside the beads (maximum entropy says that negative charge will tend to me maximally dispersed). Negatively charged species (swtrong acids) thus have only the interstitial volume accessible to them and elute from the column quickly (before t0). Neutral species have the entire column volume (interstitial and intraparticle) accessible and elute from the column at t0. Partially charged species (weak acids) elute in between. The mobile phases are usually dilute acids (e.g., sulfuric). Retention can be controlled to some extent by adjusting the mobile phase concentration so as to change the degree of ionization of the analytes.

In practice, species that can penetrate the bead can be additionally retained by a reversed-phase partition mechanism, so the same system can be used to separate small alcohols, for example.

-- Tom Jupille / LC Resources

In ion-exclusion chromatography of organic acids, the stationary phase carries a fixed negative charge (ie the same charge as analyte anions). At a fixed low pH, weak organic acids have different fractional negative charge dependent on their pKa. Those carrying a full negative charge will be eluted first, since they will tend to be repelled from the pores of the packing. Those with only a small negative charge (or neutral species) can enter the pores and interact with the surface eg by hydrophobic type interactions, dependent on the nature of the stationary phase, and will have longer retention times. This is a very simplistic explanation-I am not a great expert in this area, but I think organic acid analysis is probably the most common application of the technique. I am sure there must be others. Maybe someone else will volunteer more details.

Tom-almost a simultaneous posting! I didn't see your reply first. However, to emphasise the point you raised last. In the Bio-Rad applications I was referring to, many of the acids elute well after the column t(o) value indicating a significant contribution (in this particular example)of an additional (hydrophobic?) separation mechanism.

Tom Jupille: a very lucid and concise explanation of ion exclusion chromatography. I've tried to explain it before without fully understanding it myself. Thanks.

If you pick a mobile phase that does not absorb, for example acetonitrile and dilute phosphoric acid for the separation, it is not necessary to derivatize these or other acids for detection. For example, we have detected less than 1 ppm acetic acid at 205 nm with a 10 uL injection using a modern fixed wavelength detector and a good lamp. (If you run at these low wavelengths you will observe the begining of lamp death much earlier than at longer wavelength) Acetic acid has a molar extinction coefficient of only about 15! If the keto group is conjugated with the carbonyl of the acid function, the extinction coefficient will be vastly greater, with corresponding greater sensitivity.

We usally did these sorts of acids on one of the "aqueous" columns with acetonitrile and very dilute phsophoric acid as the elutent. On one occasion it was necessay to use ion pairing to get a separation of a keto acid from another compound of interest. Very pure ion pairing reagent will be required if you need to detect at low wavelength. Waters was the only source of ion pairing reagent we found that would work reliably at low wavelenghts.

Ion exclusion is a good approach to try for these and other acids, for example Biorad 87H. Use dilute phosphoric acid as the eluent and UV detection. However, ion exclusion columns are not as efficient as the reversed phase columns, and this can sometimes be a problem.

If keto/enol equilibria are possible with the compounds you are dealing with you will need to "freeze" the compound in one form or the other, or run under conditions where the equilibration is rapid, or you may get broad peaks.