[jj74]

Hi all,
I have to evacuate the sensitivity of a gascromatograph. VIM defines sensitivity as “quotient of the change in the indication of a measuring system and the corresponding change in the value of the quantity being measured”.
I have a Shimadzu instrument with TCD that I use to evaluate parts per million of impurities (nitrogen and oxygen especially) in Helium 6.0 (99.9999%). The integration is done with Shimadzu software CLASS-VP that gives the peak area.
I’d like to know if this procedure is correct:
I inject a gas mixture with a known certificated composition of: He with N₂ 5ppm. It gives me a peak area for N₂ of 8654 counts (the numbers that I give here are are random since I didn’t test this method yet )
Then I inject another gas mixture He with N₂ 7.5ppm, the area is 11025.
According to the definition above I have:
Change in indication of measuring system / change in the value of the quantity measured =
= (11025 - 8654) / (7.5 - 5) = 948 Area counts / ppm N₂.
Is this correct ? is 948 what I can call the sensitivity of my Shimadzu for nitrogen ?
Hope someone can help me out

[marquis]

Out of curiousity, are you using molecular seive columns for this work?

Haven't seen that definition of sensitivity.  The one we used related to signal to noise ratio.  The sensitivity limit was where the signal to noise ratio of the peak(s) being analyzed to the baseling noise was 3:1.  The detection limit was where the signal to baseline noise ratio of the peak(s) being analyzed was 10:1. So theoretically, you could detect something but not be able to quantitate it.

Your calculation may be a mathematical reduction of the signal to noise ratio.   The instrument company salesman or technical representative could probably give you more information.

Hope this helps.

[Don Shelly]

I agree with Marquis.  You really don't know what your sensitivity is until you inject a standard of known mass and look at the signal to noise ratio.

Don

[jj74]

thanks for the replies. Seems to me that signal / noise is the best description of sensitivity (I'll throw VIM in the garbage  ), since I'm preparing a sort of Logbook for all the instruments in our Lab, this definition is more useful to define the performance of the GC. I'll have to do that for FPD, FID, FID with methanizer, HADD, DID, TCD and amplified TCD and I suppose that the method of evaluate it is always the same, right ?
By the way, I made a mistake in my original post; for the ppm of N2 in He I use a more suitable DID detector with molecular sieve.