In my laboratory we have seen an unusual phenomenon several times. The most recent example is a YMC C18 column with methanol:buffer mobile phase and 240 nm detection. The buffer is tetrabutyl ammonium bromide pH-3 with acetic acid. The flow rate is about 0.7 ml/minute on Waters equipment. The void volume baseline disruption is about 4 minutes. In addition to the void volume we see a short fat peak elute at about 8 minutes which interferes with our assay. This peak is in our buffer, mobile phase, and samples diluted in buffer. The peak did not go away when we changed our needlewash from 50:50 methanol:water to 10:90 methanol water. In addition, if we inject acetonitrile or 25 ul of air we see large peaks at 8 minutes. I feel sure this is a mechanical issue but I am lost as to a solution. If I once knew the answer to this I have forgotten this. I would eagerly like to hear all comments because this is a tough one for me.

Sounds like you solved your problem: Dissolved gas in your injection material. Degas (vacuum, rotary evaporator) these and try again. Let us know what happens. This is especially common with samples that are concentrated via N2 stream.

I would like to point two other possible reasons for your trouble. Since TBAB is an Ion Pairing Reagent for anionic (acidic) compounds, who knows your C-18 column is retained some of your compounds very badly and it is eluting slowly with whatever you inject? If that is the case change the cloumn to a new column or any other C-18 you may have in the lab and see this ghost peak is still coming.
The 2nd thing is increase the purge time of your auto sampler and purge the injector several times to remove possible contamination from the needle.
Another remedy is to wash your flow cell in the UV detector with various solvents dependding on your test copmound (who knows there is some compounds are sticked into your flow cell?). If everyhting fails and if it is mechanical I would call Waters rep since we dont have any other alternatives left.

Good Luck!

Ananda

In regards to the previous message, extra purging of the AS is unlikly to help. If the contamination is from the AS, the frits are the probable culprits, extra washing seldom helps once they have become contaminated (they can be replaced). Washing the detector flow cell would be a total waste of time. Isocratic MP, washing a contaminatnt out of a flow cell at exactly the same time every injection, no way. Try the degassing solution first.

Thank you for all of your input. It looks like the culprit is dissolved oxygen. The peak is removed by sparging of the diluent and by the addition of sodium metabisulfite. This phenomena is unknown to me and I don't have a scientific rationale for this. Why don't we see this in all methods? The wavelength is relatively high (240nm). This is also a drug product related substances method therefore there is no nitrogen blow down or anything. What is happening. I would appreciate all theories because the level of dissolved gases should be the same as any method in my laboratory but we typically don't see this behavior. I am completely stumped for an explanation.

Responses like this make participation fun and worthwhile!

Your observations corresponds exactly to our experience: Gas peaks are highly unpredictable. In the analysis of cortisol and ouabain they usually were right on top of the analyte. Sometimes we have had trouble with the degasing of the sample (prior to injection), other times there was no degasing necessary over extented periods. Even saturation with N2 or O2 does not necessarilly produce a gas peak, though the probability to get it is raised sharply. Though we are stumped also, it is understandable when one considers that the solubility of gases in liquids has a strong dependance on a myriad of factors. There is one consolation: Gas peaks are usually easily recognized by their broadness.

I've seen a dissolved O2 peak using H2O as sample diluent and organic-free buffer on a YMC ODS-AQ, 220 nm. k' was ~ 5-7 , and yes, it was very broad and of course, eluted under peak of interest! This was discovered the day I forgot to turn my online degasser on, and the O2 peak vanished since mobile phase was at similar saturation level to diluent. I was able to create this peak by putting a vial into a pressurized O2 chamber and injecting immediately. Peak was huge and diminished quickly w/ next injections. I also found that its retention time increased w/ increasing column temp, as one could guess since gasses are less soluble in MP at higher temps. We lived with it instead of trying to degass diluent.

we requested for bromide in sodium metabisulfite by ion chromatograph method suggested is HPLC. Please get the desired information if available