Quantitative analysis using GC/MS

[simonwei]

Hi,

I have some stupid questions about quantification by GC/MS. I'm analyzing chloroform vapor concentration in nitrogen matrix. Should I use the area counts of a single ion or the TIC for quantitative analysis? If a single ion should be used, which ion should I choose, the most abundant one?
Is fragmentation a function of analyte concentration? If so how can GC/MS do quantitative analysis? I'm asking because I observed the ratio of chloroform fragment peaks, say, area counts of m/z= 83 over that of m/z=35, varies at various chloroform concentrations.
Thanks.

simonwei

[Don_Hilton]

I prefer to use a characteristic ion out of the analyte of interest. In using the TIC, you get signals assing noise from all the other stuff that may be there.

As far as which ion: Pick the strongetst ion that does not show interference from coeluting peaks. You may want to run several samples just to see what else could be in the background.

Some additions to my original reply to address the last part of the question:

If ion ratios are changing at various concentrations one of several things is hapening:
1) you have a sufficiently strong signal that you have begun to saturate the detector -- For this, you need to either use a split injection or use a weaker ion for quantitation.
2) You have a coeluting contaminant -- Look at ion traces and see if you have some ions that look like they have different retetntion times than others.
3) If you are taking ratiios from peak areas, you could have integration issues - which can be a problem with either very large or very small peaks.
Note that spectra will change as you change ion source temperature or column flow rate. You can have variation if you do not use consistant operating conditions for the mass spectrometer.