My analyte has a RT of 5 min at 1.3 ml/min on Zorbax SB aq (5 um, 150x4.6 mm). The MP is 2-5% organic and 50 mM trifluoroacetic acid. However, an interference peak is very close to the analyte one. Changing Methanol to acetonitrile does NOT resolve them. Is it a good idea to use trichloroacetic acid instead of trifluoroacetic acid? The analyte is a base with pKa 2.3.

Thanks a lot in advance.

Does your column resist this strong acidity? Is your pKa really a pKb?
Anyway, it seems wise to use a good buffer at higher pH for your aqu. part. TFA and TCA are highly dissociated and, therefore, not the best buffers.
It has been claimed that TCA attacks stainless steel (via Cl-?).

A pKa of 2.3 suggests that you have an acid group, which is partially ionised with that amount of TFA. TCA may not be available in the purity you require/can afford. Suggest you re-charaterize your analyte, look at other acidifiers, and up the organic content a little, reduce the flow rate to 0.8 ml/min and try THF/ACN/MeOH in combination. TFA can form ion pairs.
Good luck
Che

The analyte is a nucleoside analogue with a pKa 2.3 for the base. It has flourescence only under very low pH, which excludes buffers with pH >2.0. As organic is in very low concentrations, it seems to me that the type of organics does NOT affect the resolution/selectivity very much. I know only Zorbax SB-Aq columns that can be used under pH 2 with nearly 100% aq. MP. My quess is that the only possibility is to switch TFA to TCA or perchloric acid. I made a few trials and got excellent results with TCA. My conceern now is, why is TCA seldom used in HPLC? DOes it attack stainless steel as Mueller mentioned? If it was Cl- that caused the problem, then a well degassed MP should be OK with stainless steel. Without oxygen, Cl- can NOT attack stainless steel. Any further suggestions are highly appreciated.

Chloride ions do attack austenetic stainless steel. The process is called chloride stress corrosion and is an interstitial attack by chloride ions is regions with high residual stress ( bent tubing, etc.). This process is greatly accelerated under acidic or high temp conditions. Chloride stress corrosion is a well understood process and is detailed in any undergraduate text on material science.

In nuclear power plants if chloride ions are detected in the coolant systems it is called a chloride casualty and special mitigation prodecures are executed immediately.

In an HPLC I wouldn't be paranoid about chloride ions but I wouldn't use acidic chloride solutions unless necessary.

Actually what I would use is trichloroacetic acid (50 mM in water), which is of 99% purity as its C of A indicates. There might be some chloride and other impurities and degradation products there, but sure in quite low concentrations. Once I read, chloride attackes stainless steel only in presence of air or oxygen. So the possibility for chloride to attack stainless steel is small.

What other possible disadvantages are there? Destroy packing materials? Etc.

Ok, you have a "base" not a base.
TCA has a very strong and broad absorption (UV). It is surprising that you have enough light left, even at 50mM, to excite your analyte. What Ex wavelenghth do you use, what´s the concentration of your substance, its absorption coefficient?
I have never seen TCA used in a MP. If you go ahead with this please observe very closely (contact times, etc.) what happens and report on it.
Or, you may do the chromatogrphy as suggested by participants and adjust your pH post-column.

Thanks a lot to Dr. Mueller!

Ex. 260 nm bandwidth 18 nm

What is the absorbance of trichloroacetic acid below 250 nm?

Fluka for their "BioChemika MicroSelect": 260nm: A=1.6; 280nm: A=0.15 for 0.5M in H2O. In spite of the factor 10x I am surprised, especially, since one could expect the Cl of TCA, etc., to quench the fluorescence (not only absorb the UV).

The fluorescence with TCA is about the same as with TCA.
Does TFA quence fluorescence for Ex 260 nm and Em 390 nm?

Mant thanks to Herrn Prof. Dr. Mueller and all other respondants!