Very low abundances for PFTBA fragments during GC-MS tune

[rekuci]

We have an HP 6890 GC/5973 MS that hasn't been used for over a month. On attempting an autotune, abundances for the perfluorotributylamine (PFTBA) standard fragments are unusually low. For example, abundance of the mass 69 fragment is in the 100s range instead of the 100000s range. Autotune errors have been "unable to achieve constant peak widths" - probably because the peaks are so small relative to the noise level, with the 502 fragment peak looking particularly terrible. Also had the error that the EM voltage was maxed out on attempting to autotune.

I followed directions for cleaning the ion source, put it back together again, and still have the same problem. Haven't messed around with manual tune much, but on first glance it doesn't look like anything will get abundances to where they should be.

This is my first time using MS, does anyone know what the problem is? Is this the photomultiplier tube gone bad?

[vballchemist]

First thing is to check the vial that holds the PFTBA. Be sure this is full.
After that, do some manual tuning to try to enhance the major ions. The electron multipliers for a 5973 last about a year based on usage. There is a shelf life on most of them also, so a new multiplier is probably indicated.

[Peggsy]

I agree with vballchemist, in that the first thing to do is to check the level of the PFTBA in the calibration vial, and refill if necessary.

To help out otherwise, what EM is being used for the autotune?? In addition both the AMU offset and the AMU gain (slope in the Mathieu stability diagram) will influence peak widths, and the influence of these on abundance of the calibration ions can be manually checked, again as mentioned by vballchemist.

Let us know hopw you go about troubleshooting this please.

Regards
Greg

[rekuci]

The PTFBA level looks fine. When I tried an autotune again, the abundances were way up (although I'm not 100% certain they weren't the other times too, since I didn't watch all of the autotune which takes nearly half an hour before it gives up)

But the mass 502 peak still looks terrible (it's barely there), and the autotune gives up saying it can't achieve constant peak widths. I tried manual tuning, and can get the abundances way up with the ion focus and entrance lens position, and also the EM voltage. But 502 never really improves, its abundance increases but it looks like little more than noise. I have a coursebook (I didn't go to the course...I just have the course notes) that says "poor ion focus results in poor high mass response", but no amount of changing the ion focus changes the nature of the peak. I tried changing AMU gain and offset too.

Can someone recommend some sort of algorithm for manual tuning, because there are so many parameters that it's impossible to handle! All these course notes say is to "continue adjusting the parameters until the desired effect is achieved", but why would I be able to do this better than the autotune program? I don't have any special needs that require manual tuning, and find it hard to believe that the program isn't adjusting all of these parameters and optimizing correctly if it could.

[WK]

Rekuci,
How old is your EM? What was the last tune voltage it had?
Was the instrument switched off for the month?
The first things to address are:
1) replace EM
2) Cleaning the MS
Try to do this one at a time.
If you had a spare EM then you would have replaced it by now - so I guess you need to order one to find out.
WK

[rekuci]

Nope...no spare EM. Our chemist says it was last replaced about 3-4 years ago, so he wouldn't be surprised if it were at the end of its usefulness now. It sounds like something you can purchase and install yourself, so this may be the thing to do next.

But get this...in Agilent's tech support document about non-constant MS peak widths:

"*****WARNING*****
IF MASS SPEC IS MODEL 5971,72 OR 73, MAKE SURE SOURCE is not
baked OUT IN OVEN AFTER CLEANING. THIS COULD RESULT IN SOURCE
BEING OXIDIZED BEFORE IT IS REINSTALLED INTO MASS SPEC.
This is true for other sources as well, but is especially detrimental to the above mentioned sources. When using volatile solvents during the ultrasonic portion of the source cleaning, the solvents are virtually evaporated by the time the parts are laid out, so oven drying does
not buy you that much time and could cause you to have to clean the Source again because of oxidation."

This is very interesting, since I cleaned out the ion source following THEIR OWN REFERENCE VIDEO THAT SAID TO OVEN DRY THE SOURCE PARTS.

Guess the next thing to do now is to redo the entire cleaning all over again, since the messy noisy 502 peak does look like a dirty source.

[WK]

Rekuci,
Agilent would not be the only manufacturer with apparently conflicting procedures. There will always be an element of "live and learn".
I am not familiar with Agilent "source" but my experience of this site leads me to understand that you need to handle them very carefully.
3-4 years seems quite long for a well used EM. Its worth replacing the EM to see what effect it has before you clean the source again. It should be easy to install.
Happy MS-ing
WK

[james little]

could you just go back to an old tune and check the response for air in the system. That way wouldn't depend on the delivery of the tuning compound.

Another thing to try if you can go back to an old tune, inject something by GC-MS and see what the sensitivity is like..

[rekuci]

The source was cleaned again without oven drying, a bit better but still bad. We're ordering a new EM tube. I'll repost once we get it and install it(in the government, this could take awhile...) In the meantime, back to hopelessly attempting to get this light gas analysis to work in our 5890 with TCD...

[JI2002]

Oven drying the ion source parts is ok, but not baking. The oven temperature should be kept at about 60C.

If the MS worked well a month ago, I doubt it's the EM problem, especially if you have good responses of other ions. Like James Little suggested, check for leaks by doing a scan, look for m/z 18, 32, and 28. The responses of theses ions should be less than 5% of m/z 69. If there is no leak, check the flow rate of the carrier gas, depending on your column dimention, carrier gas flow should be less than 1 ml/min. Check those two things before replacing EM.