UPLC Injection reproducility Issues

[mnrevoy]

Hello,

We are currently in the pre-validation stages of a new UPLC gradient method which uses 0.1% TFA, MeOH and ACN in the mobile phase.
Recently, we have noticed that the standard injections are no longer as reproducible as we would like. For example, in the last run, I ran 5 injections of each level of standard and control. For one of the control levels, the first injection appeared blank, the second was ~50% of expected, while the last 3 injections all quantified correctly. This is not the first time we've seen this type of issue. It has me concerned that the unknown samples might not be injecting their full amount either, therefore lowering their reportable values.
Does anyone have any idea as to what our issue might be?
Both the standards and the unknown samples have been diluted in 1-propanol.
I would appreciate any / all help!

[tom jupille]

If I had to guess, I'd say the autosampler sucked air for some reason on the first injection, the problem cleared during the second and all was well for injections 3-5. Without knowing a lot more -- and actually watching the system in action, it's hard to say:
- the needle cored the first septum?
- the wash solution wasn't degassed?
- the syringe hung up?
- an improperly tightened fitting allowed air to be aspirated?

[lmh]

sometimes silly things like a bubble stuck at the end of an insert, or in the bottom of a high-recovery (pointy-bottomed) vial???

[danko]

Is the described observation reproducible?
To me it sounds like strong substance adsorption to the stationary phase.
The cure is column saturation with the actual substance - fx injecting standard solution several times until the peak aria becomes constant.

Best Regards

[danko]

Is the described observation reproducible?
To me it sounds like strong substance adsorption to the stationary phase.
The cure is column saturation with the actual substance - fx injecting standard solution several times until the peak aria becomes constant.

Best Regards

[BostonFSE]

HI
Is this an acquity? 1260? Ultimate 3000?

Possible Air in sample on first 2 injections. Run a series of blanks and overlay their profile to see if they are reproducible.
Also, could be a partially clogged needle, which unclogged itself after a couple injections.
For this issue, if its a hardware problem, its likely needle, needle seat or inj rotor.
When was the last PM?