Setting up calibration on GC data analysis

[choiar]

Hello guys,

I am working on setting up calibration curves for fatty acids absolute quantification. One thing that I try to understand how calibration works is "response". Response in calibration is the area of the peak of interest as the manual says. However, I observed that the response of IS in the calibration (approximately 80000) is not the same number that I see in the percent report section of "correct area of IS" (approximately 10000000). Anyone can help me to understand this problem? I believe these different area of IS affects my absolute quantification since I see the different values of fatty acids in using calibration and manual calculation.

To give you more details:

1. Correct area in percent report section of IS is generated by clicking percent report under chromatogram in data analysis program.
2. To generate calibration curves, run samples (contained different concentrations of fatty acids) and use Autosetup calibration (they added automatically response in the calibration) and I manually added concentration for each response.

I really appreciate any help I can from experts!

Arum

[MichaelVW]

Can you post a picture/screencap of where you're getting the two IS values that don't match?

[choiar]

I am trying to post pictures here but not successful. How can I add pictures here? Sorry I am new in this forum.

Thanks,

Arum

[kubowicz.tomasz]

Hello

viewtopic.php?f=2&t=12659

Regards

Tomasz Kubowicz

[choiar]

Here are the pictures.



My internal standard comes at 32 min and you can see that "area" in the percentage report and in the calibration are very different. Now I am doubting setting up calibration is wrong.. Could you explain to me how you set up calibration with easy level words since I have this project for my Master degree but I am not chemistry graduate?

Thanks for any comments that I can try it and make this thing work.

Arum

[MichaelVW]

Looks to me like the percent report calculates the height of the internal standard in the TIC (Total Ion Chromatogram) window. But your calibration is based on your target ion, so the counts for that ion will be a fraction of the total counts of the sum of every ion. In other words, your calibration compares the response of your IS's main ion to the response of each of your your target analytes' main ions.

You can see in the cal table there is a dropdown menu where you can choose to do it that way or based on the TICs. Maybe the more experienced people here can explain why you'd do the latter, but to me it seems to defeat the purpose of a mass spec...

So basically, I don't think you have a problem. Try printing a report but insted of percent, select "summary" - I bet your IS will appear at the top of the page with the same number of counts as in your cal table.

[choiar]

Thanks Michael! It makes sense I will try that. Do you prepare to use TIC instead of target ion setting? When I was setting up the calibration (a rough version) I found that it is bizarre that there is no place to put MW of FAME ( mother ions) since my compounds of interests are fatty acids so their target ions are very similar although their retention time is different. One question I have is that does a level that you put in in the calibration setting affect the absolute quantification? I didn't think that affects the quantification since I put the concentration for each level of interest compound.

Thanks!

[MichaelVW]

I always use target ions to quantitate. I run dirty samples looking for environmental pollutants, so there are lots of 'unknowns' in the matrix with the compounds I'm looking for. If you use target ions, you don't have to worry so much about other compounds coeluting with what you are looking for - as long as they don't have similar ions. For example, often I'm looking for PAHes in samples that have lots of straight chain hydrocarbons - so many that they dwarf the PAH peaks - but I'm still able to get good results.

I am not familar with the Autosetup feature of the software but maybe it automatically selects the best ion for each analyte. Under "Edit Compounds" you can check to see which ion it is using for each compound. You also should look at each of the compounds you calibrated for and under the "Calibration" tab make sure you have a good looking curve (low RSD).

You'll also want to play around with the Qedit function in the software; it lets you see how the software has integrated your peaks, change them if necessary, and see how many counts (of your target ion) each peak has.

It doesn't matter what you put as the level ID, as long as you give the software the concentration. You could have level 1 be 20 ppm and level 2 be 1 ppm and skip a few and have level 6 be 0.1 ppm.

Once you have that all figured out, it might be good to calculate a few concentrations out by hand using the RF and see if you get the same answer as the software.