Dear All,

I have a problem with one of my assays, and I'm stressed out....I am running a isocratic system with 40:60 acetonitrile: phosphate buffer mixutre, pH 3.3 and using a reversed-phase C18 column. The UV detection is at 216nm. My peaks of interests are at around 5.2 and 6.7 minutes respectively. Recently, there is an unknown peak at around 5.4 minutes that is appearing all the time. It's peak height is not constant. I have changed the guard column and column. I also thought it may be system contamination, and I have washed the system thoroughly and also change the tubings. However, the pump broke down in that system, and I have switched to another system. Still this peak appears. It is also appearing when I inject pure water, and all the pure standards.

The problem is that when I analyze my samples which are occuring at low concentrations, this peak co-elute (or elute right next to) with my peak of interest and it is effecting the integration.

All the professionals out there, please help!

I was thinking, is this an absorbance problem? But I did not have this problem before.....

Please help! Thank you!

The lambda max for molecular oxygen is 220nm. You could be seeing an "air peak".You could try injecting 1 or 2 microliters of air and seeing if this peak grows.

If you are using an inline vacuum degasser or helium sparge the concentration of air in the mobile phase will be kept low while air will resaturate with the sample. Some autosamplers use a small air plug to contain the sample and limit diffusion and help sweep the sample on to the column.

Also check any vent lines from the autosampler sometimes they get rerouted and can siphon mobile phase when in one position then air gets injected when the injection is made.

Good luck.

Dear Tom,

Thank you for your help. I have tried injecting 1 or 2 microliters of air into the system and there is no peak.

What else do you think I can try? By the way, my colleague was the one who had the splitting problem and who has asked for your advice too!

Thanks!

In our experience it is difficult to reproduce air peaks by injecting air. Saturating sample with oxygen behaved equally equivocal.
Air peaks are usually relatively broad. . . .
What happens if you collect it and re-inject? Can you take a spectrum?
How far are you from your dead time? Could it be that this is a light scattering peak? (Is that what you meant by "absorbance problem"?).
The solution to the latter problem would be to lower the organic modifyer to get higher RT.

Tom, is this 220nm for dissolved oxygen, is it possibly also scattering? I am sure to have seen peaks due to nitrogen, thatīs why I donīt concentrate HPLC samples with a stream of nitrogen anymore.

Thanks for your suggestions. The particular peak is not broad at all. It is appearing in almost all kinds of matrices -- injection of pure water, mobile phase, 20mM phosphoric acid (because I use it as back extraction) and pure saline. It is quite far from the dead time as the peak comes out after my peak of interest. I have also changed my organic modifier (30% acetonitrile vs. 40% acetonitrile)--the peak's RT shifts, but it's still there. Sample preparation has not been done to the pure standards, so it has not undergone nitrogen streams.

I do not have this problem before, and I can't think of anything which I have done differently. It's depressing to see 3 peaks when you are injecting a pure standard of only 2 compounds.

I'll try collecting it and re-inject. (Do you mean collecting it after going thru the column? Please advise)

Thank you for all your valuable suggestions.

What is your dead time (or what are your column dimensions and what is the flowrate)?
Does the peak shift as expected (rule of three) due to a 10% lowering of ACN? Does it stay between your analytes?
You should collect the peak as close to the detector outlet as possible when you see it come through. You can reinject the whole fraction if you remove most of the ACN with a rotary evaporator. You can also do (or have done) MS, UV, etc. specs on a fraction.
And please! Let us know how the problem was solved.
If this is a real peak than your syringe, injection valve, etc. is simply contaminated.

It sounds as though something is contaiminated - did you say that you still get the peak when you inject blank? Have you washed your needle?

To add to Helen's question about the blank, are you standards and reagents still pure? I've seen people spend a lot of time checking the instrument when it was actually a degraded standard.

do you have access to a manual injector ? if so, plumb it in and inject w/ a syringe. I have seen the vials used cause a leaching effect into the sample

Lots of good ideas. I throw in a couple more. If the problem goes away by plumbing in a manual injector. You could try replacing the injector rotor seal and needle seat if you are using an HP 1100 like P. Good luck and let us know how it turns out.

Thanks for all of your good ideas.

HW, I am using an old model column (not the new ones in the market) -- a Waters Radial Pak Cartridge column (NovaPak C18, 4 micron, 8x100mm)with a Guard Column insert. The flow rate is 1ml/min.

Yes, I do think the peak shifted due to the ACN change, and it still stays between my analytes. However, the peak height (intensity) is not constant.

Helen, the peak is appearing even if I changed from the Waters Millenium system to the HP 1100 system. Everything has been washed and columns/guard columns have been changed to new ones.

Anonymous, my standards and reagents are still pure because I have made new standards and re-injected the old standards as well. Also, when I check my previous chromatograms of the old standards, the peak was not there. But now, even if I inject the old standards, this peak comes up too.

Anonymous, I'll try to borrow a manual injector from the central lab and inject the standards into the system.

Thanks for all your good ideas. I'll give them all a try. Please do throw in more suggestions!

have you looked at your ACN? Just because it says HPLC on the bottle does not mean that it is without impurities. I have noticed some differences between HPLC grade ACN from different suppliers. Following much trial end error we have now settled on a single supplier who seems to provide the "cleanest" solvents, even though the costs are marginally higher

Last anonymous, how does dirty mobile phase produce a peak? One can imagine that an equilibrium disturbance, like the injection, kicks some accumulated dirt off the column, or solvent of the sample, but then one would not expect the peak to change its RT with org. modifier.
Initial anonymous: You are not sure the peak moves when changing the ACN amount? Maybe your peak is not due to "dirt" in your injection after all? Maybe last anonymous has a point then? Of course, you apparently donīt need dirt on the column to get a disturbance peak.

Have you tried already to inject a sample without a seal onto the cap, because its possible that some ingredients of the seal in the screwcap are eluating.

Dear All,

I can't borrow a manual injector! But I'll try all the other ideas. Thanks for all your help. I'll get back to you all if I can find a solution. (Hopefully)

Thank you!

Just another idea:
Did you check the vials? For certain analysises we rinse our vials befor use.