non-linearity in FID

[rsester]

I'm running total petroleum hydrocarbons and I've been losing linearity in the heavy oil range over time. Diesel range and surrogates are nearly perfect, but the linear range for oil is narrowing.

Does anyone have any suggestions?

thanks
rsester@ncalabs.com

[Peter Apps]

Hi Rachel

The likely cause depends whether you are losing the top or the botton end of the calibration line.

The FID itself has a linear range of 6 orders of magnitude, so the problem probably lies somewhere else. My guess would be inlet discrimination against the heavy boilers.

Check out the article by John Hinshaw in LC.GC Europe July 2002 for info on sources of non-linearity.

Peter

[chhubert]

I guess the liner is soiled with high boiling residue after many injection of dirty samples. Usually high boiling compounds like to stick to these residue in the liner or column front. Did you change the liner (and gold seal if you use Agilent GC) and cut say 30cm of cap column from the inlet? I think you can get your high boiling compound back after changing these.

[rsester]

Thanks, you guys. I changed the gold seal and cut 30 cm off the front of the column. The higher end of my calibration for heavy compounds is still not as high as it should be.

I did find John Hinshaw's article, and am checking it out. In the meantime, maybe you could answer two related questions:

1. I was using a Restek splitless liner with wool. I accidentally ordered a gooseneck splitless liner with wool and decided to try it. Would that change the linearity of the response for heavy compounds?

2. How does the condition/age of the FID jet affect the linearity of the response? Or does it affect it at all?

Thanks. I'm very excited about having a way to get my questions answered.

Rachel

[Peter Apps]

Hi Rachel

Changing inlet liner design might change the discimination against high boilers.

You say that you are using a splitless liner, but are you doing splitless injections ? What are your analytical conditions - temperatures, flow rates, column length, i.d. and phase, and the injection settings for the autosampler (which are often not optimal) etc, without this information we are trying to troubleshoot by guesswork.

Did you see any improvement when you changed the gold seal and clipped the column ?

If there is a problem in the FID it will affect all the analytes equally - the only place that you can get discrimination is in the inlet.

Peter