I have problems with the response factors of fluor-chemicals. I use an HP-5890 with an HP-1 cappillary column and FID. When injecting a standard solution (10 times), the first 2 or 4 times the response factor is higher (or lower, depending on the compound)and then stabelizes. I use decane as internal standard solution. This peak-area is the one that changes from a lower area to a higher area.
Can you help me explaining why? and what can I do about it?
Thanks!

It's tough to diagnose a problem like this without seeing the data. I'm not sure what you mean by fluor-chemicals, but will assume you're working with things like freons.

I have never worked with florinated compounds, so I don't know whether they might behave oddly in an FID. An FID works by creating ions, and perhaps flourine somehow interferes. Someone with more knowledge in this are will have to chime in about this.

My first rule of thumb for GC is that a leak (somewhere in the system) can cause any symptom. My second rule of thumb is that quantitation problems usually arise due to the injection. That said, there is some small possibility remaining that you're seeing irreversable adsorption.

The irreversable adsorption hypothesis will explain symptoms in which, with repeated injections, the peak area rises then levels off. This does not sound quite like your situation. Furthermore, it is not a likely scenario on an HP-1 column.

More likely are injection problems. What is you injection mode -- split, splitless, direct or whatever? What split ratio (if applicable)? What injection volume? What sample solvent? Are you using a silanized inlet liner? Is there any packing in the liner? How often do you change septa and syringes?

Some of those questions are loaded -- I'm really falling back on my first rule of thumb -- assuming a leak.

Bruce,
I am new to GC and I know now that I gave to few data to analyse the problem.
First, with F-chemicals I mean hydro-carbons with fluor in the structure. I use an automated injector, split injection(25:1),injection temperature is 200°C and injection volume is 1µl.

My sample solvent is MTBE (Methyl t-butyl ether). I use a straight, not sylanized, liner with deactivated glass-wool. When I did these tests I had just changed septum and liner and conditioned the GC.

The problem is, when I inject 10 times my standard solution (ISS = decane, X1 and X2),the area of decane decreases until it becomes stabel. X1 is normaly distributed and the area of X2 increases until it stabelizes. (this after 6 injections). Statistically decane and X2 decrease and increase auto-correlated.

I have checked for leaks(there were no leaks), and replaced the syringes. I am running a test now.

Thank you for your help.

Changing the syringe might help. Sometimes people don't realize how often syringes must be changed. You can tell one is worn and in need of change if you can see a short black smudge near the top of the barrel. This discoloration is due to metal wearing off the plunger onto the glass. It is a reliable indicator of syringe wear and likely leakage past the plunger.

Otherwise, the only potential problem I see is the very high injection port temperature. For MTBE solvent containing fluoro-hydrocarbons solutes, I doubt you need a temperature anywhere near that high. I'd suggest you try dropping it to 125 C or less.

The reason for doing this is that flash evaporation of the solvent is often detrimental to the separation. Sample can be blown back into the upstream plumbing and out the sample purge vent. However, since you are using a split injection, I don't see that as a likely problem.

What sort of syringe wash and fill routine have you programed? I typically use something like this: Wash 5 times with sample (discarding the sample), then fill and empty the syringe five times to get rid of air. Are you using anything like that?

Changing the syringe indeed helped !
Inspecting the syringe there was a discoloration near the top of the barrel like you said.
Thank you verry much for your help.

Caroline