Internal standard increases, analytes decrease in area......

[rickwg]

I've read the forums for a while and decided that I'd post the problem in the hopes that someone has had a similar experience.

I'm using a micromass quattro LCZ older instrument I know but it's what I have. 2795 waters HPLC on the front end and I've done an injector test and there is little to no injector problem. Running MRM experiments.

When the DMPK group runs samples, ~300 a night, mixture of qc, analyte with internal standard, washes...etc., the internal standard starts out at an area count of, say, 4000, and ends up at a area count of 7000 or so, while in the same samples the analytes start at around 6000 and drop to about 4000 or so. QC samples Bear this out and have same response trends.

Anyone ever seen this before? IS going up while Analytes going down over time?

Just had the instrument PM'd but it was doing it before. The sensitivity is up now and that's good. Using a diverter valve so no goop getting in my source(at least not like before). the run in 5 minutes long and everything comes out "chromatographically" nice, no peak overlaps. (~20s separation or so)

Driving me crazy and any help would be appreciated.

[Uwe Neue]

Could be a combination of ion suppression for the analyte with ion-enhancement for your internal standard. Both may depend on your sample cleanup technique and on accumulation of interferences on your column.

Question: have you tried a brief column wash in the middle of the run to see if this brings you back to the initial response pattern?

[rickwg]

Yes, we do periodic column washes in the batches and, no it doesn't bring back the response. Later, after the run is over response will come back but falls off again after running more samples.

I have mentioned using a spe type preparation to the dmpk group.
thanks for your reply.

I'm doing a run right now with several injections out of the same vial. This vial contains 5 compounds and the areas for these compounds seem to vary indepentantly of each other.
Say cmpd one will give good reproducibility while compound 3,5 vary in area and or some other combination of varaition and stability.
Can't figure it out. should be all the same as it is from the same vial.

[Uwe Neue]

One possibility is ion-enhancement of your internal standard from phospholipids accumulating on your column. There have been several posts on this board on this subject. This is why I suggested the washing protocol. However, the removal of such compounds may depend on the detail of the washing protocol.

Independent of an interpretation: if you recover the response after the run is over, i.e. with a new run, I would carefully examine what you do with the column between the runs, and use this protocol as your washing protocol.

[gpronger]

In GCMS, with more labile compounds; a phenomena I've seen is the slow deactivation of the column as a batch progresses; i.e., injection to injection, active sites are coated so that as the run progresses, sensitivity for a particular compound group improves. The typical solution is to start with a few injections of the problematic analytes at a relatively high concentration, deactivating the column for the actual analytical work.