I am using splitless liner for split injections.

[zimanli]

Does it affect my reproduciblity of peak area and quantitation? I have a really bad %RSD now (27%). The split ratio I used is 100:1. Could any one tell me the reason?

[stappg]

What is your solvent? Check its expanded volume. If the liner volume is too small for your chosen solvent and temperature then you may have a problem. Also check obvious stuff like syringe operation, needle blockage and port septa. I have used the same liners without problems but volumes are critical. Good luck.
Peter

[stappg]

Ooops, just thought of something else....................
Your using a fairly large split ratio (100:1) and I think Agilent supply a liner with "little glass notches" on the bottom edge where the liner touches the gold seal. These allow extra flow of waste gases during split mode. If you can, reduce your injection volume to 1 uL or less. Again - good luck.
Peter

[Consumer Products Guy]

Yes, the wrong liner can definitely affect your results. We had a contract manufacturer purchase and get a computerized GC set up by the supplier's engineers for one specific application about ten years ago. This was an internal standard procedure and the reproducibility out of the same vial was terrible. Engineers felt it was inlet issue, kept changing the liner out. After a week, I figured it out long distance, faxed the them a page out of catalog showing liners photos, asked them to tell me which one it was. Then I called the engineers (who had the split liner, but kept replacing with what the GC came with) and had them run over and install the correct liner. The written test procedure clearly stated that it was a split injection, but over long distances......

[Peter Apps]

In principal a "splitless" liner should work for split injections. If you go the other way; using a narrow bore split liner for splitless you can run into trouble.

The problem is more likely to be something to do with temperatures and injection technique; splitless injections are more rubust than split because the sample has longer to evaporate and get on to the column. Please post sufficient details of your instrument conditions to save us having to tropubleshoot by guesswork !

Peter

[zimanli]

Thank you very much!
Stappg, I injected 50uL gas sample with 250 uL gas tight syringe. The liner is 900uL splitless liner. I was trying to measure vapor pressure of some solid sample. We put solid in a little bottle with septum cap and inject gas sample into GC/MS.

Consumer products guy, when you mention long distance, does that mean splitless liner has a long distance injection? I did not understand this part. You mean I can not use splitless liner for split injection, right?

Peter, the temperature I used is 60 C to 200 C with 20 C/min.

I actually had ~8% RSD for liquid injection with internal standard method. I think there is some thing wrong with my injection technique or the gas tight syringe, but the liner is also suspectible.

Thanks a lot!

[Peter Apps]

In the light of what we now know, the poor repeatability almost certainly has nothing to do with the inlet liner.

To get repeatable peak areas for injections of equilibrium headspace (which is what you get when you seal a solid sample into a bottle) you need to thermostat the sample to + 0.1 degrees C and have the syringe at a fixed temperature somewhat higher than the equilibrium temperature of the sample. You need to equilibrate your sample for at least an hour. The harsh truth is that it is nearly impossible to get good repeatability of peak areas by manual headspace injections.

Comparing the repeatability that you got with liquid injections including an internal standard against what you get with gas injections with no internal standard is not valid.

The temperatures that you give look like the temperature programme for the column, but we are talking about the inlet - what is the temperature of the inlet ? If the temperatures that you gave really are the for the inlet, why are you programming the temperature ?

Peter

[Consumer Products Guy]

zimanli - I just mean I was about 1500 miles from that GC location; so it was tougher proble-solving than if I was in the same room as the GC. And sometimes it's difficult to know what others "don't know".

[zimanli]

the inlet temperature is 200 C, which is about 18 C lower than the boiling point of my solid.

I think I expected liquid injection with internal standard method would have %RSD about 1% or at least 2-4%, because we get rid of injection error.

Peter, I think what you said is absolutely helpful. Thanks.