Bad peak shape when switching to headspace

[MestizoJoe]

I was using liquid injection to detect DMF using DMSO as sample solvent and had nice peak shape using liquid injection.

I needed to switch to Headspace GC for irrelevant reasons and now my peak shape is bad. The GC conditions are the same. The only change is that now I'm using the headspace to inject. I'm guessing the solution is very simple, but so far I've tried

oven temp = 100C
Loop temp = 150C
loop transfer temp = 160C
vial equilibration time = 10min

loop fill time = 0.1min also tried 0.15min
loop equilibration time = 0.05min also tried 0.3min
changing injection time = 0.5min also tried 0.2min

my split ratio for head space is 0.1 which i think is similar to 1:9.

[Ron]

A split ratio of 0.1 is basically a splitless injeciton. I don't know what the total flow through the loop is, but you are probably doing a very slow injection, spreading out the injection band. Headspace is a concentration technique, so if you doing headspace using the same concentration you used for liquid injections you are probably overloading the column.

[Peter Apps]

Try setting the split to 10:1, how you do this will depend on what hardware you have, and how it is connected together.

Posting a chromatogram will help - instructions in a sticky at the top of the LC page.

Peter

[PSUalum]

consider a 1mmID straight-through split liner to minimize band broadening, if not already done so. This should clean them up nicely.