When comparing my results with those from other labs, I am finding that my results are more heavily weighted toward polyunsaturated fats. I was hoping to fix this by calibrating with theoretical response factors (TRF) but AOCS Ce 1h-05 has TRF increasing with a degree of unsaturation. So that ResponseFactor=(Actual/FID Response), or Actual=FID*ResponseFactor, would only serve to more heavily weight my results toward polyunsaturated.
I have the Supelco 37 FAME mix and my chromatogram offers slightly inflated values in the region of interest, is there a way to calibrate using only this mix?
I am not an expert on this topic, but I run a different std for my fame analysis... are you quantifying your results with an internal std? When I looked up Supleco 37 FAME Mix, it said you can get a different peak order depending on what column you use, here it is below.
http://www.sigmaaldrich.com/technical-documents/articles/reporter-us/supelco-slb-il100.html
Thank you for your replies. I am using SP-2560. 100m x 25mm x 20micron. I get excellent separation, and I have more than just that standard so it is not a peak identification problem. It seems to be a calibration problem.
I will try an IS and see what happens.
I would certainly suggest using an internal standard. Irrespective of this, if you are using split injection, you might check your injector and detector temperatures. If they are not high enough you can get some mass discrimination and lower responses for the longer chain saturates.
The supelco mix is indicated for qualitative analysis. I suggest that you use individual standards and an internal standard, since the number of carbones influence the signal size by FID detection, and then you can normalize this. I hope that it helps.
"When comparing my results with those from other labs" . Are you sure you have the same sample, What is the range of concentration %, ppm, ppt? do you purifie, concentrate, make extraction? Do you use the same column?
I thind you would must to take in account the following factors:
Check state and adjust of the septum of the injector.
Temperature, concentration of sample, and time of injection.
Time in minutes of the chromatogram, and their temperature.
The composition of Supelco 37 FAME mix could to change with the temperature of Injector and Detector.
I have made a 3 level calibration table using the supelco 37 fame mix based on mass (in ng) by varying injection volume. I am using temperature (detector and inlet) and ramp conditions identical to one of the other labs. They use rx2560 I am using sp2560.
You don't use the same column. Restrek and Supelco are two different manufacturers. In my experience between DB5 and HP5 i have observed different retention time for a same product. It is not a problem because your standard calibration will consider the différences of response factor.
I would go for my own response factor (not theoretical), using internal standard and a calibration curve with individual FAMEs. As far as I remember in our chromatographic conditions, response factors were not only dependent upon unsaturation, but also on chain length, so that response factor were almost twice higher for long chain PUFAs than for short chain SFAs.
I just perceive me that you calibrate by varying injection volume. Try to inject uniform volume in split or splitless mode. With different concentrations, you must have a linear calibration curve with different response's factors between each compound. i agree with Jorge Ruiz, keep your own respose factor which is caracteristic of your apparatus.
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training