In purge-and-trap GC/MS: Using a standard 20 mL headspace vial (glass, Sil/PTFE crimp cap) for the low bp (-128°C) volatile compound CF4 the peak I obtain at the corresponding retention time has the characteristic masses 50 and 69 as expected. Now I add the same standard in equal or higher amounts to an 80 mL vessel (steel, viton seal and sil/PTFE headspace crimp caps) I suddenly get a peak at the good retention time for CF4 but the masses have changed to a weired gauß-kind of distribution over m/Z 55-69 with highest mass at 65. The blank from the same container shows no peak. What could possibly be a reason for that. I expect it to be a chromatographic or detection problem - or what kind of errors can I make in handling??? I welcome any suggestion very much!
A first guess is that the steel of the 80 mL vessel is coated and the Tetrafluoromethane dissolves the coating.
maybe you'll get a similiar MS when using CCl4?
Thanks for your comment - unfortunately I do not have CCl4, but SF6 (bp -64°C) behaves similarly. The steel is not coated and used for grinding hard minerals like quartz. I tried using the steel vessel in a dilution step of the gas standard. When I took the diluted gas out of the steel and transferd it to a glass headspace for measuring the masses detected were correct again. It's nuts! Also CF4 is quite an inert gas...
I'm not sure I've got this straight - using CF4 with your normal glass system, you get a normal retention time and a "normal" MS pattern for CF4. Using CF4 with a larger, steel vessel, you get the normal retention time, but an unusual fragmentation pattern. Using the steel vessel alone, you don't get any peak in the chromatogram, and of course, no fragmentation pattern. Is that it?
There's results that show the fragmentation of CF4 changes in the presences of nitrous oxide or molecular oxygen (see attached). I wonder if your larger container allows enough air to be admitted into the ionization source for this to be an issue.
If my understanding is correct, it is not a chromatographic issue - you get the correct retention.
One times I received SF6 fragmentation pattern because of helium contamination with SF6.
Did you checked this possibility?
Yours
Wojciech Draminski
Using 80ml vessel in place of 20ml vessel, your instrument may inject larger amount of CF4, and you possibly see the results of ion-molecule reactions in the MS source. Are the intensities of the spectra from the 80 ml experiment similar to those from the 20 ml experiment? In the spectra from the 80 ml experiment, are there ions with higher masses, such as m/z 157,138, 119 (2 CF4-nF)?
If you get mass 50 and 69 along with 65 then we can conclude that there is some contamination in the sample. Depends on intensity of mass in mass spectrum we can judge that your desire compund is present or not. If the intensity of 65 mass is maximum and intensity of 50 and 69 is very less then contamination is more in your sample but the desire compound is present in it.
Dr. M.R Tiwari (SICART)
Ines, The gaussian-like distribution of masses make me think that your ion volume contain too much sample. This could explain the very braod distribution of the mass peak. the m/z=65 could not be the "real" mass of the ion/ions inside the mass spectrometer. Between 50 and 69 have you got distinct masses or a very broad single peak centered at 65?. If you have good resolution in that range (peaks at let's say 51-52-53-....and so on) the problem could not be due to overloading of the sample in the mass spectrometer and the problem could be related to some other contaminating substance that entered the mass spectrometer during the change of the headspace vial.
Ines, my comment is substantially similar to that of Aladar Bencsath.
Dear all
i have Finingan GC-MS with headsapce with xcalibur sofware , some timies when i make sequance and run this sequnace there are error massage appear ( quanserve. error) I welcome any suggestion very much to slove this eror
I wounder whether you have stored your already processeed sample in a glass vail. Sometimes this distorts the peak.I know this from my experience with GC/MS. Also you can use hexane as a solvent. Try to use the standard at different concentration prepared from the same stock solution.
Thank you all for valuable comments so far. @John Ford - I think that could potentially be a problem as I have more relatively more oxygen after prugen the steel than the glass vessel. Thank you for the paper. I think I will try to measure CF4 and SF6 under argon instead of under ambient air. @ Markus Hollmann - good idea. I am not sure if I did it correctly. There is now a file with some images of chromatograms and spectra to illustrate my problem on my home - shared documents @Mary Kimani - thank you for your suggestion buit at the moment I work with pure gases, not standard solutions
@Bruce Freeman: Thank you - unforunatelly, your link did not work with on my computer. Is this a Firefox problem or did the link have an error?
@ John Ford - on the other hand, I am not dealing with CF4 plasmas at high enery - 70eV in the Ion trap is used for ionization and to my knowledge there is no plasma created. But I do not know too much about plasmas.
SOLUTION OF PROBLEM:
Dear all, thank you for your contributions. Thanks to the suggestions and ideas you gave I could now resolve the problem - special thanks to John Ford, who identified the problem correctly. I now fill the steel container with helium prior to injecting my gas standard. The large vessel volume also contributed higher amounts of atmospheric oxygen and nitrous oxides, which created the chemical ionization/plasma conditions that altered the characteristic masses for my compound. Unter helium (or argon) I am also able to get CF4 with characteristic masses of 50 and 69 from the steel chamber.
- Badges
- Science method