Headspace analysis of tolueen in unleaded gasoline

[Happymadness]

Hi I am a student in the Netherlands. I am studing at the Hogeschool Leiden (Uneverity of aplied sience Leiden).
The current project where i worked on is the GC analysis of toluene in gasoline with headspace on a Trace 1300GC with the FID detector.

I take 3,5mL of gasoline and put it in a 5,0 mL vial and 200µL N-butylformate as a internal standard.
This is headed for a houre in a waterbath at 60 dagree celcius.

Than there get 5µL of the headspace injected, the method was the same as http://www.restek.com/chromatogram/view/GC_PC01071 but the colum was the Rtx 5 ms, 30 meter.

The chromatograms are good but when the same sample is injected multiple times, and corrected with the internal standard, the corrected area of al the peaks are difrent, even the rates of the peaks are diffrent.

My teacher doesn't know how to get stable peak rates and the only thing i could find on the internet and some books was the use of a headspace autosampler, but we haven't one so this is not a solution, what else coold i try?

[Peter Apps]

Inject headspace from a vial at room temperature - heating the sample to 60 C causes problems with condensation in the syringe that you use to take the sample.

Peter

[rb6banjo]

How much toluene do you expect in the gasoline? This seems like a job for a direct injection to me. This is a tough sample for headspace analysis - especially at 60 °C.

[AICMM]

You might consider a much lower volume of gasoline in the sample. Toluene might typically be at 10-20% in a gasoline sample so 3,5 mL is significant quantity of toluene. Furthermore, you do not say if you are running split or splitless, the latter probably not appropriate for this analysis.

And, yes, a direct injection with a big split might be more appropriate....

Best regards,

AICMM