I am currently developing a headspace gas chromatography method for the analysis of residual solvents. Although I have found many good resources regarding HS-GC method development approaches, I have had a difficult time finding information about the transfer line and loop temperature settings. I am using DMSO as the sample solvent and the headspace oven temperature is 100 C. Is it important to keep the transfer line and loop temperature greater than the boiling point of DMSO (189 C) or does it just need to be slightly higher than the headspace oven temperature? I am concerned about sample degradation at high loop and transfer line temperatures but I also don't want DMSO to condense in the line and loop. Any information about this topic will be much appreciated.

I have used a temperature of 110°C for the transfer line quite successfully for DMAc and DMSO for trace analysis without problems using fused silica coated parts. Assuming you are not using caustics in your work the FS coated components will work well. I used mine for more than 4 years before others took over the instrument. Reproducible results with DMSO requires only a 80°C HS oven. Even at that temperature you are forming artifacts from the DMSO (caused by trace water and oxygen present in air and sample?). They increase with higher oven temperatures and increased heating time. Also there is a greater chance of sample side reactions occurring. There will be DMSO in small amounts present in the components (rotor if you have a unit with a valve, seals if you have a timed injection unit) but it will not affect your analysis, but your blank will have the DMSO present. The PE HS-40 has the least carry-over I have seen. (No I do not work for PE or ever have)

Part of my work was published in Analytical Chemistry in June 1997. I was injecting w/o split which increases the carry-over effect, but I measured residual solvents at a level of less than 1ng to 1000ng per sample linearly.

Good luck with your method development.

Rodney George

Thanks for the information, I will give 110C a shot. I am actually seeing many artifacts presumably from interaction between the the DMSO and my sample which is why I am interested in reducing temperature where possible.

Are you using a headspace unit which allows shaking during heating? Agitation during the equilibration time will allow a significantly shorter equilibration time, and possibly a lowere equilibration temperature. The transfer line and loop temperatures in headspace should be set above the sample temperature by at least 10C, but do not have to be above the boiling point of the solvent.