Tailing problem

[Sarahr]

Hi
I am analysing open ocean seawater samples for D and L amino acids using pre-column derivatisaton with N-Iso-L-butyryl-L-cysteine and O-pthaldialdehyde. We're using an Agilent 1100 Series HPLC system with a Thermo Electron Corporation BDS Hypersil C18 column (only been fitted for a month). The solvent program employed uses acetonitrile, methanol, 23 mM sodium acetate buffer (pH 6) and HPLC grade water.
All of a sudden the four peaks that come off first (asparctic and glutamic acid) are tailing really badly. We have washed the column numerous times, made fresh reagents etc. yet the resolution is still bad. Could this problem be due to the salt in my samples or maybe metal contamination? In which case i hear that EDTA can be used. if so does anyone have a method for this cleanup?
Cheers

[HW Mueller]

As stated before, I very much admire anybody who gets somethng useful out of the OPA method, or even anybody who even tries it. So rather than trying to figure out which of the zillion possibilities went wrong here, my suggestion is to try FMOC.

[Uwe Neue]

If indeed the problem is only with the first few peaks in the chromatogram, I would attribute it to an injection bandspreading problem, and not to the column.

[Mark Tracy]

I agree with Uwe about bandspreading. I suspect you are injecting a huge volume to get the detection limits you need. At the pH of the derivatization, the Asp and Glu derivatives are triply negative anions and probably have very little retention even in aqueous conditions. Try a pH adjustment after the derivatization is complete to reduce the ionization.

This assay needs to resolve D from L and therefore needs a chiral derivatizing chemistry, hence the chiral thiol. FMOC is achiral and won't do. FLEC is a chiral version of FMOC and might work. As for the OPA method, most of the bad rap on it has to do with mercaptoethanol as the nucleophile; almost any other thiol gives a more stable derivative, stinks less, and is less toxic.

[HW Mueller]

Mark, do the other thiols give a single derivative??

[Mark Tracy]

For a mono-amine, the answer is yes, OPA-thiol gives one derivative. The catch comes with polyamines (lysine, cystine, etc.) where multiple derivatization is a problem for every chemistry, not just OPA-thiol. The well-known instability of the mercaptoethanol derivative can also lead to multiple peaks.

Usually, I prefer post-column derivatization after ion-exchange (I used to support this application for Pickering Laboratories), but in sarahr's situation, pre-column is the only way to go.