HPLC-ECD peak with 0 µl injection

[artsjeroen]

When injecting blank samples (ultra-pure water) on my HPLC-ECD system, I get a peak with 4.5 min. retention (injection dip is at 2.5 min.). At first I thought my ultra-pure water might be contaminated, however injecting 0 µl gives the same peak. The peak size does not increase when injecting 100 µl.

The needle wash consists of 30% ACN in water. The seal wash is 10% IPA in water. I don't expect that one of those solvents gives a peak with that much retention. What could be causing this peak?

[tom jupille]

Dissolved oxygen ? Try injecting degassed DI water and see if that makes a difference.

[artsjeroen]

Never ever seen that peak before, so I doubt it is dissolved oxygen. And the peak also appears when 0 µl is injected, so dissolved oxygen would cause noise on the baseline and not appear as a symmetric peak. Correct me if I'm wrong.

The HPLC-system has had service last week, the injector seals and selenoid valves have been replaced. But I don't see what could be causing the peak or how to get rid of it.

[artsjeroen]

The peak is still giving trouble. It is visible in both ECD and UV methods. I have rinsed the complete system with magic mix (MeOH/ACN/IPA/UPW 25/25/25/25) to no avail. The peak does not decrease in size after multiple injections, and it is present when injecting 0 µl. I've changed the needle wash and seal wash multiple times with fresh solvents. But nothing seems to solve this problem.

[sbrand]

I remember a case about such a ghost peak coming from a contaminated degasser. Did you checked this? What is the injection principle of the autosampler (Loop or inline Split loop)? What happens if you run an injection without moving the inject valve?

[tom jupille]

Another thought: how flow-sensitive is your detector (easy to check, let it stabilize at a given flow, then cut the flow in half). If it is very flow sensitive, you may just be seeing the effect of the momentary flow interruption from the injection.

[M Farooq]

When there is a large pressure difference during the injection process i.e. when the rheodyne switches its position from inject to load position, you may see a peak which appears later than the injection peak. Since injection volume etc. don't make any difference, it implies that it is a flow related artifact. Note the pressure difference before and during injection. Do you see a dip in a pressure profile as well?

[artsjeroen]

Thanks for all the tips!

The peak is still giving trouble. It is visible in both ECD and UV methods. I have rinsed the complete system with magic mix (MeOH/ACN/IPA/UPW 25/25/25/25) to no avail. The peak does not decrease in size after multiple injections, and it is present when injecting 0 µl. I've changed the needle wash and seal wash multiple times with fresh solvents. But nothing seems to solve this problem.

My fault, I've not seen the peak in an UV method yet. The HPLC system is a Waters 2695 Alliance with a 2465 ECD detector.

I've tried the following:
If I use our other HPLC-ECD system (Shimadzu Prominence with Antec Decade II ECD) with exactly the same mobile phase / post column phase, column and flowcell, no peak is visible. So the problem must be system specific.

I've connected the ECD directly to the in-line filter, so the injector does not take part of the system, then no peak is visible in the chromatogram. So the peak must be introduced in the injector part.

The detector is flow sensitive, it is a VT-03 Au flowcell which needs a high pH to measure my component. So I need a post column NaOH-solution addition. But I've never seen such peak after injecting samples.

A Waters service engineer is looking at it right now, I hope he can find the cause. I will let it know.

[artsjeroen]

The seal pack and valves have been replaced again. And we've prepared all solutions fresh (needle wash/seal wash/mobile phase/post column) but the peak remains in the chromatogram.

The service engineer has no clue where it is coming from.

I will test the system with an UV detector later this week.

[ravirocks86]

Are you able to find a root cause for the problem? I am facing same kind of situation with UV detector.