Repeatability issues with FID/PFPD analyzing fuel

[JHZR2]

Dear forum,

I have performed a serach on numerous relevant topics, and have not been able to gather the information that i require, so I would appreciate some insight.

I am running a Varian 3800 GC, with an 8400 autosampler and 1177 injector. The system is set up to inject, elute through a 30m sulfur analysis column (sulfur analysis is the key functionality of this GC), and the splits via a glass 'wye' to an FID and PFPD.

Currently, we operate with no solvent plug. We inject 1uL and operate with splits between zero and 100 (we have verified that we are doing splitless correctly, and have the correct method modifications to adapt), depending upon the sulfur concentration in the fuel, which is the limiting factor, due to saturation of the PFPD response.

The major issue is repeatability at any split ratio. There are differences in total peak area from 30-70%, and this is with both fuel and binary mixtures, prepared to analyze how the GC is operating.

We have a slow autoinjector pull up, and inject slow for splitless, and relatively slow for split. (0.2 and 5 uL respectively).

So, background aside, there are major repeatability issues. We believe that it has to do with the glass splitter post-column, as we have theoretically mitigated the issues with syringe draw-up speed, and have verified injector parameters for our various split and splitless operations. As a result, we have a few questions:

1) is a solvent plug generally necessary when injecting jet fuel (C9-C16 range, typically)?
2) what is a good solvent to use for a jet fuel analysis?
3) what would you reccomend for a draw-up and injection speed when using jet fuels in split and splitless injections?
4) do you have any other reccomendations in order to be sure that our deviation is minimized? We are currently removing the y-splitter to determine if direct plumbing to a detector solves thee issues. All the same, the questions above still apply, as we typically have been injecting fuel without said plug.

Any insight would be most appreciated.

Thank you for your time,

JMH

[AICMM]

JHZR2,

Couple questions, couple comments. Is the variability both FID and PFPD? Are you getting variability at different concentrations at one split or are you trying to vary the split for different concentrations. The former should work the latter I have had great difficulty with. What liner are you using?

Regarding solvent plug, no you should not need one. Dichloromethane, pentane, even hexane would be appropriate solvents for jet fuel. The splitter could be an issue for sulfur (reactive) but you should still get good reproducability on the FID in that scenario. I have generally had good luck with splitters although most were on the front instead of the back. I am not sure you should be using a slow injection in split mode since you are relying on high injector flow rate to sweep inlet. A slow injection and split seem counter productive especially if you do not re-focus on the head of the column.

Best regards.