Mysterious shift in concentration after LCMS goes idle

[jsupica]

Running an 10 analyte Opiate method in dilute urine with deuterated internal standards on a Sciex API 3200 w/ Shimidzu LC20ADxr pumps. Mobile phase A is .1%formic acid and 10mMol ammonium acetate in water and Mobile phase B is .1% formic acid in acetonitrile. Column is a Restek DB C18 3um 100x2.1mm

I was running an experiment with a production high control that had been hydrolyzed 5 days previous and took an hour long break part way through the runs and the LCMS went idle. Later when I charted the concentration of Oxymorphone, the factors that I was testing had no effect (crap) but when I ran the samples in time order, there was a distinct upward shift in concentration from ~2800ng/ml to ~3100ng/ml after the hour long break!

I recreated this shift on another system by letting the LC go idle and keeping the MS active. But then...I tried keeping both the LC and MS active but not running any samples for an hour and got a similar shift in concentration.

Has anyone seen or heard of anything similar? We're just now getting in to seriously studying it but it has us baffled.

[carlo.annaratone]

Might be evaporation of the buffer? Do you use closed caps on the solvent bottles?

[jsupica]

Carlo,

We do use caps on the solvent bottles. The before and after tests we ran were all out of the same high control vial too.

[James_Ball]

Are you using an internal standard?

It can be that something is slightly precipitating on the inner cone orifice and causing a slightly higher vacuum which can shift sensitivity upwards or downwards if there is enough to block the analytes from entering the analyzer. I have a 3200 and with some methods I have to keep an eye on the analyzer vacuum as it will shift from say 3.2 to 3.0 as buffer accumulates on the cone. Running higher flow between the cones can help keep them clean.

Internal standard will help compensate for any sensitivity drift in the analyzer.

[jsupica]

James,

Yes, we are using an internal standard. We have had the same issues with having the curtain gas set too low on earlier instruments but corrected that some time back. It's a big production environment we're working in so the blanket rule is if we have an instrument that gets to 3.0 we clean it.

We haven't looked for the shift before and after cleaning though...I smell a new study! I can't figure out why it would be only Oxymorphone. It elutes right in the middle of the run, has its own IS, and isn't especially heavy or light.