Decrease in sensitivity in Agilent GCTQMS 7890B/7000C

[Meggie1983]

During one week I have observed a decrease in sensitivity of our GC triple quadrupole. The response of the standard measured yesterday is 1/3 of the response of the same standard measured 1 week ago.
I have eliminated problems with the GC injection part and changing the septa, liner, gold seal, syringe did not change the response. Baking out the column and the mass spec did not improve the response either.
Please note, I have cleaned the source 1 month ago and the tune passes without a problem.
However, the repeller voltage decreased from 10.8 to 4.5 in this time span and I not sure if this would lead to decreased sensitivity.
Any suggestions to the source of the problem. Any information would be much appreciated.

thank you
M

[dr_bahram1977]

First of all, check your EMV. if you've decreased EMV, then your problem resides in MSD and mainly in Ion source.
I think you'd better to clean Ion source, then check for results again.

[James_Ball]

If you run the autotune on a regular basis you have to remember that it is optimizing the response in the 219 mass range. If your compound of interest is above or below this, then the response will change as the balance of high and low masses change with each new tune.

It is better to use an injection of a known compound to verify the tuning in a daily basis than to run the autotune every day. Something similar to BFB or DFTPP that is used in the EPA methods would probably work. If you begin to see changes in the mass balance in the spectra then you know you need to retune or clean the source, if you get the same responses each day then you know everything is still clean and working correctly.

Changing the repeller voltage will change the sensitivity of the instrument, and will change it different amounts at different masses, so that could be some of your problem.

[Meggie1983]

Hello James,
Thank you for the suggestion - I will definitely try it.
I would like to bring your attention to one more issue with the instrument. With each sequence we have retention time shifts to the left.
I am looking at PCB analysis in SPME extracts and have a nice MRM method with quite large number of groups for this purpose.
Whenever running a sequence with extracts and standards and QA/QCs, with time I have retention time shifts such that the peaks a partially cut off from the group.
This is causing troubles with compound quantification. After all, I do need to adjust the time of specific groups before next analysis.
I am not sure if this is caused by the GC - potentially a leak which I am not able to track or due to dirty samples. The instrument is checked for leaks with a Restek leak detector and any external leaks at the inlet or to the MS transfer line are always eliminated. This is also confirmed with the air leak check.
Even though we analyze SPME extracts, the extraction lens and repeller are covered with a purple and brown layer and I do need to clean the source more often.
Still, not convinced that it would cause retention time shifts.
Please note, we have a new column but the same story happened with the old column as well as with the new one.

Any suggestions are appreciated
M

[James_Ball]

Have you tried the Retention Time Locking feature in the software? It is located in the Method pull down menu. Essentially you will assign some peaks that are used for monitoring retention times and the instrument will adjust the flow slightly to keep them in the proper times. You usually run the check at the beginning of the run, let it correct then proceed with analysis. This will correct for any flow problems or if you have clipped a little column from the front end for cleaning purposes.

Alternatively if you do clip column you can go into the column tab in the method setup window and use the calibrate column button to correct for the amount you have removed and this will help keep the retention times in place. Another thing I do with my MRM segments is I put the transitions from the last eluting peak into the next segment and the first eluting peak into the previous segment so that if the retention times so shift slightly you will still get a good peak from the border line compounds.

The purple/brown color you get on the ion lenses can be a sign that you have a small air leak, which can cause oxidation that will give a blue/bronze/purple color to the metal of the lenses. Go into manual tune and turn on the filament but leave the calibration gas valve closed and monitor 10-200 amu and spray some Dustoff around the connections and side plate seals of the MS and look for a spike in the ions. Depends on the molecular weight of the compound used but is normally between 50-85 amu. If you have a leak you will see a spike in these ions when you spray near the leak. This will reveal vacuum leaks that you can't find with the leak detector.

This normally works, but I have to admit that currently I have air and water getting into my 7000 and I just can't find where it is coming from. It almost has to be upstream of the GC so I am now working my way back to the tanks that are outside to see if I can find it.

[Meggie1983]

Dear James,

Thank you so much for all these valuable suggestions!
I will definitely try the retention time locking. Until now I was using the column calibration.
I am so happy that we had a chance to exchange thoughts.

Best wishes,
Meggie