Tailing problem on Acquity UPLC/TQD

[mtorrey]

I am having this frustrating tailing problem I can't resolve. I have replaced mobil phase and the column. I have replace probe capillary and sample cone. No change. The tailing gets progressively worse. Every injection. Every sample, including unextracted standard untouched by human hands for months.

My peaks were perfectly symmetrical when I established the method, no changes to the method has been done. I checked all parameters, they are the same as when the method was established.

The method is for Pentobarbital. Internal standard is Pentobarbital-D5.

Mobil phase is 5 mM Ammonium Acetate, pH 10:ACN, isocratic at 30:70. Flow rate .350 mL/min.

Column is Aquity 2.1 x 50 mm BEH 1.7.

Detection is MRM, ESI- mode.

Strange thing is the tricyclic method I've been working on still has symmetrical peaks. Yet if I use the column I use for tricyclics (Acquity sheild RP) the Pento peaks still tail.

To me this keeps pointing to the chemistry but I thought BEH columns was immune to this sort of tailing.

Could a dirty or contaminated source cause tailing like this? Sensitivity is not effected that much though. It all points to the LC side but I'm wondering if there could be something wrong on the detector end.

TIA
Mike

[Uwe Neue]

What is the retention of your pentobarbital peak on the BEH C18 column and on the BEH Shield RP18 column. What is the retention of the tricyclics? Were the operating conditions for the tricyclics and the pentobarbital the same? If not, please specify!

[mtorrey]

What is the retention of your pentobarbital peak on the BEH C18 column and on the BEH Shield RP18 column. What is the retention of the tricyclics? Were the operating conditions for the tricyclics and the pentobarbital the same? If not, please specify!

The retention time on the 50 mm BEH C18 is 0.52 mins with a flow rate of 0.350. The BEH Shield RP18 is 100 mm so I had to increase the flow rate to 0.5 to get the peak out in time as the run time is only 1.2 minutes. At flow rate of 0.5 on the 100 mm BEH Shield the peak comes out at 0.7 minutes. But it is still broad and tailing.

The tricyclic method is also isocratic using the same mobil phase (30:70) at a flow rate of 0.35. But they come out at 2 to 3 minutes. It is also ESI+ while the Pento is ESI-. The tricyclics do tail a little which I assume is due to the amine groups. But they do not broaden like the Pento is doing.

[Uwe Neue]

You have a bad case of extra-column bandspreading. The peaks with little retention give you more tailing than the peaks that are well retained. You retention factor for pentobarbitol is about 0.53 to 0.59. The reason that the tricyclics look better is that they are retained more, which covers the extra-column bandspreading better.

Honestly, at such a low retention factor I am not surprised that extra-column bandspreading is getting you. There can be many, many different causes for this. How large is your injection volume? What is the sample dissolved in? Since you are doing MS, you may have a mile of tubing between the LC column and the detector. Was it always rather bad, or has it gotten worse after you changed something in the system?

The easiest way to fix this is to change the solvent composition just a bit (35:65) and, if you want to keep about the same run time, to increase the flow rate. The peak will look much better...

[mtorrey]

Service came in and found that the sample valve is bad. That is what was causing the excessive tailing. He replaced it and now my peak is as before. Also have a bad pump seal but that wasn't causing a problem since I'm doing an isocratic gradient.

But he says something in my sample is hanging on and blocking the front of the column because the pressure doesn't return to baseline after injection. So he is giving my procedure to the application specialists at Waters to look at and see what it could be.

Mike

[mtorrey]

Ok we found out the ZnSO4 was precipitating out in the mobil phase (ammonium acetate). So that ruined my sample loop and two columns. But the ZnSO4 improves the chromatography, more recovery and less tailing. So either I find a different aqueas phase compatable with ZnSO4 and the TQD or make due with lousy looking peaks. I could go to SPE but I'm doing that already using GC/MS and the whole point was to cut down on extraction time.

Mike

[Kostas Petritis]

I do not really see how ZnSO4 in your sample improves your chromatography, unless it is present in huge quantities... You might want to start with high aqueous mobile phase in order to keep ZnSO4 in solution and get rid of it as it will be eluted in the void volume (assuming that the rest of your compounds will be highly retained, also do not aquire during that time to avoid sending it in the MS), then make a step gradient to your present conditions... It would be good if you experiment at what mobile phase composition the ZnSO4 stays in solution...

[mtorrey]

I don't either. I assume because it aids in precitation and without it there are proteins left that are causing interference. The problem is with the TQD I don't get a lot of choices in what to use for a buffer. I may just have to live with the tailing.

Mike