GC/MS fluctuations, normalization, comparability of data

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

8 posts Page 1 of 1
Dear GC/MS-Users

for the next years I am plannig to analyse the headspace of many samples by SPME-GC/MS (PerkinElmer SQ8). For about twenty targets we would like to perform relative quantifications (I measure only the peak areas and compare them among the samples).
We will receive the samples in bunches during the next few years, and they have to be measured shortly after we have received them. Therefore, the GC/MS will be re-tuned, switched off, cleaned etc. all along these measurements. This will produce sensitivity fluctuations of the mass spectrometer.
Do you think using an internal standard (4-heptanone or 1,2-dichlorobenzene) for data normalization is enough to take the fluctuations into account?
What is your experience with such fluctuations? How do you deal with it?

Many thanks for your advice in advance!
Kind regards
Christian
I would recommend adopting a method that helps you determine what fluctuations are acceptable and what would require recalibration or retuning. If these are volatile compounds you may want to consider EPA method SW-846 8260D. A quick google search will help you find the document. It will have information on initial calibration, continuing calibration, even tuning guidance for your mass spec.
Regards,

Christian
Depending on what repeatability you require throughtout the period, I would be tempted to run a set of external standards (preferably of your analytes of interest) after each major maintenance (source clean etc) as well as internal standards.

Peter
Peter Apps
A lot will depend on how dirty the samples will be, but if it is only headspace then they should be fairly clean. If the vacuum is kept turned on and the analyzer up to temperature all the time then even over that amount of time there would not be much fluctuations and the internal standard will do quite well to compensate for any such changes.

I have Agilent 5973s that can run for two years using purge and trap as the inlet without needing to be cleaned, and they run hundreds of drinking water samples each month. If you don't have samples, then it is better to make an idle method that flows nitrogen or hydrogen carrier to save money on helium and keep the instrument running the whole time. Venting and shutdowns are not good on mass specs over time. If the sample load is very small you can probably even go 1-2 years between changes of the rough pump oil, just keep an eye on it to make sure it stays full.
The past is there to guide us into the future, not to dwell in.
Dear all.

Thanks for the valuable feedbacks!
So far, we kept the MS running without shutdowns for cleaning or any new tunings.
However, at the moment we see large fluctuations of the area of our regularly analyzed ISTDs. We are examining several possible causes.
1. The ISTD-solutions can not be stored.
Solution: the ISTD solutions must be made freshly.
2. A dirty SPME fibers causes fluctuations.
Solution: We prolong the fiber bake out before each run.
3. The calibration is not stable. Is that possible? How often do I have to calibrate with PFTBA?
Solution: we calibrate before each sample series.
4. A dirty liner causes fluctuation.
Solution: We change the liner.

Any other ideas?
Kind regards, Christian
Peak area's will change depending on the tuning of the instrument. GC/MS by SPME tends to have quite a bit of variability. Ideally you would have internal standards similar to each group of compounds you are running.

At what time frame are you seeing the fluctuations? Within a sequence? How much are you seeing? What is your istd? How are you spiking and storing the samples after?

Test old lots of ISTD against new lots to determine storage degradation.

I highly recommend you get known standards for at least a subset of your analytes of interest or similar compounds if that is not possible. At the very least that will give you an idea of how your recovery looks that day.
ChrisT wrote:
Dear all.

Thanks for the valuable feedbacks!
So far, we kept the MS running without shutdowns for cleaning or any new tunings.
However, at the moment we see large fluctuations of the area of our regularly analyzed ISTDs. We are examining several possible causes.
1. The ISTD-solutions can not be stored.
Solution: the ISTD solutions must be made freshly.
2. A dirty SPME fibers causes fluctuations.
Solution: We prolong the fiber bake out before each run.
3. The calibration is not stable. Is that possible? How often do I have to calibrate with PFTBA?
Solution: we calibrate before each sample series.

4. A dirty liner causes fluctuation.
Solution: We change the liner.

Any other ideas?
Kind regards, Christian


The problem with doing an autotune with PFTBA is the instrument does not hit the exact same response ratio between all masses each time. It only looks for a good fit within the acceptable window of ratios, which can cause quite a bit of difference if you are comparing m/z 45 with m/z 250 for example. One time it may be that 45 is 75% of 250, and the next time it may be 50% then the next time it may be 95%, all depending on the acceptance window in the tune algorithm.

We use the EPA method of injecting either Bromofluorobenzene or DFTPP (decafluorotriphenylphosphine) and use those to show the instrument is still within acceptable tune ratios instead of running the autotune. That way it doesn't alter the lens settings between runs, and we only retune if there are large shifts.
The past is there to guide us into the future, not to dwell in.
It sounds like you'd be leaning on the ISTD pretty hard... Anything that effects the IS and your analytes in exactly the same way will be mitigated. Anything that effects the IS and your analytes differently will become an easy-to-miss source of error. I would be worried about that. I make too many stupid mistakes, and have too many stupid things happen to my instruments, to rely on the IS alone to take care of everything.

You mentioned changing the liner. If you're only using one IS compound, and no external standards, how will you even know that the liner or column is dirty? Some of the internal standards used in EPA methods, for example, are not terribly inconvenienced by a dirty liner - even when they are dirty enough to obliterate some of the target analytes. You could try to choose an IS that has the exact same susceptibility to break down as your targets do... but that assumes your targets all behave the same way.

As far as keeping an eye on the tune, maybe if you pick an IS that has a large variety of ions that you can look at, you can compare their ratios in the way bromoflorobenzene is used in EPA 8260. Heck, maybe use BFB as your IS. It's cheap and it's not going to be in your samples.
8 posts Page 1 of 1

Who is online

In total there are 19 users online :: 0 registered, 0 hidden and 19 guests (based on users active over the past 5 minutes)
Most users ever online was 894 on Tue Dec 21, 2021 5:59 pm

Users browsing this forum: No registered users and 19 guests

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry