TQ LCMS Peak (tr) wander 4 min to 3.5 min

[NordicJuha]

Dear all,

I recently joined a bioanalytical team, where LC-MS/MS is used.

I noticed that they run 6 standard curves for each analytical run (using 1 as std curve and 5 other as QC), and normally reject the first 2 (approximately 20 samples), as these are used to calibrate the system, whereafter the system becomes stable. I now also see this phenomenon on the instrument I work on (Sciex 6500+ TQ LC-MS). This affects the reproducebility of my data and reliability on the system.

It is very accepted and normalized in the team that this is how you do things, not only on this instrument. I used to work in another team on Waters Xevo TQ UPLC-MS, and this was not something that I have had problems with.

I then tried to calibrate the system for 30 min prior to running the first samples, but same phenomenon is observed.

Now i also noticed that it takes time for the analyte retention time to settle. It wanders from 4 min to 3.5 min over the course of 40 injections.

I run with MRM, column Acquity UPLC CSH Flupro-phenyl 1,7 µm 2.1x 100mm.
60 degrees celsius for column temp.

I run gradient elution (Mobile phase A: 0.1% formic acid, B: 95% Acetonitrile +0.1% formic acid).

0 min--> 1 min: Fraction B: 10%
1 min --> 4.5 min: Fraction B: 50%
4.51 min --> 5.5 min: Fraction B: 100%
5.51 min --> 7.00 min: fraction B: 10%

Flow: 0.6 mL/min.
Injection volume: 10 µL

[lmh]

For that size of column, 1 minute is a very short re-equilibration time. Also, from the way you write the gradient, it is not possible to tell whether you have the instrument spending 1 minute pumping 10% B to re-equilibrate, or whether you have a gradient over 1 minute falling to 10% at the end, which will leave it barely equilibrated.

If your column is not fully re-equilibrated, the retention time of the analytes on making the next injection will not be the same as it would be were you to equilibrate the column for longer.

This is not necessarily a bad thing. Provided it is reproducible, and the peak shape is acceptable, it's okay. Sometimes speed of assay is more important than full equilibration. But it does mean that you should not expect the first run in a batch to be the same as the remainder. This is entirely normal.

It also means you should not expect the method to move from instrument to instrument without significant changes. The reason is that different instruments will take a different length of time to prepare for the next injection between runs (a time which contributes to column re-equilibration) and will have different delay volumes between the pump and the column. Therefore if you have a method in which the column is only just equilibrated enough, on a different system you may find that it's not equilibrated enough (peak suddenly emerges ridiculously early, or badly shaped), or it's much better equilibrated (peak is significantly later).