From time to time I get a huge peak in the middle of my chromatogram with my waters hplc/pda. The spectrum is just a wavy line like a cross-section of a lake with a strong wind blowing. ANybody got an idea what it is ? It does not happen every sample.

Hmm. Something with that spectrum sounds like air in the flow cell. That shouldn't happen in the middle of a chromatogram, the beginning of the run is more likely.

Try to collect it. If what you collect is full of air bubbles, you have your answer.

You can check the spectrum of your collected sample with a cuvette reading absorbance detector, if you have one. You are not likely to see the spectrum in your collected sample once you tap the air bubbles out of the cuvette.

Most likely you have very retentive impurity in one of your samples. This impurity elutes few injections after one that it is belong to. Since peak is big you may be overloading UV detector which would explain the wavy spectrum.

Unless this is a compound that has absorbance at all wavelengths, overloading the detector will not produce the spectrum being seen. Air is the likely culprit. Depending on your mobile phase, gradient and conditions, you may be seeing outgassing in the flow cell. Some more info on the mobile phase would help (composition, gradient, flow rate, degassing technique, etc.) but you may want to look at the tubing on the outlet side of the detector. What do you have on there right now? If you have fairly wide bore tubing, try putting about 12" of 0.009" ID tubing first to keep a little back pressure on the cell and prevent outgassing. I'm assuming you're running analytical scale (1-2 mL/min), don't put 0.009" tubing on a prep system.

Try extracting a specta on the leading (or trailing) edge of this large peak. If you get a non wavy (normal) looking spectra, you are dealing with the problem outlined by Zelechonok. If it looks wavy, air might be the culprit (Hugh and RC).

Hi there, when air is injected onto our system (Waters system with Alltima C18 column), we get an air peak at approx 9min. If there is not sufficient solution in the sample vials, sample plus air will be injected. We use the 717plus Waters autosampler.

Hope that helps

Greg

What do you mean by wavy spectra? Just noise or a sine type wave?
This is what we have seen (sundry examples):
-Bubbles formed downflow from the column cause spikes in the chrom (no specs taken, but this should have spikes as well).
-Bubbles formed in the detector cell can produce a sawtooth baseline (builtup then sudden disappearance, no specs taken).
-Dissolved gases (or a very fine gas foam, ie milky...) produce a broad peak, can be anywhere, depending on?? Spectra ~horizontal (can be noisy and of very high absorbance) except near 200nm where it has a strong downslope from an even higher absorbance (in our detector actually similar in shape to a spec just taken after zeroing the detector, only at a higher level, that is, absorption). Chrom baseline is raised by the same amt.
-Cell which has been flushed with gas has the same shape spectrum (~horizontal above 220nm, at raised absortion level, chrom baseline raised by same amount).
(If this was due to pure scattering, the absorbance should rise toward longer wavelength (1/Lambda^4)).

I am seeing "spikes" on my CEC chromatogram. Could anyone tell me what happenned? I am running on a Waters 2690 with W2996 dioarray detector.

Thanks