cis-1,2 dichloroethene split/ghost peaks

[unhpian]

I'm running a PE Autosystem with a PE Turbomatrix headspace sampler. We have a 0.25 mm x 30 m RTX-VMS column on an ECD with a splitless injector and a 1mm liner. We run chlorinated ethenes - and the more chlorinated compounds (TCE, PCE) are okay and generate a big response, but we're having lots of problems with cis-1,2-DCE. The response is very, very low and we get intermittent split/ghost peaks- where the cDCE peak will split into two completely different peaks 0.5 min apart....but, this only happens about 50% of the time. I've checked the column insertion distance, and it is within specs. We're running the ECD at 350, and the injector at 250 c.
Any ideas?

[sassman]

Sounds like degradation to me. Especially since the other compounds are not affected. You can try decreasing the injector temp and possibly using a higher carrier flow rate to minimize the potential for degradation. Pulsed high pressure injection and/or a liner designed for easily degradable compounds might also help.

[AICMM]

unhpian,

My limited experience has been that dichloro compounds have very poor ECD response. Gets better with trichlor and booming for tetrachloro. The split peak is probably an artifact of ionization competing with electron capture for that compound depending on the level analyzed (higher level, more ionization possible.) Usually spoken of as a W peak. One trick for poor capturing compounds is to run the detector cooler (Detectors for Capillary Chrom., Hill/McMinn) rather than hotter especially since you are dealing with volatiles.

Another option is to use a PID or AID since the ionization potential for this compound is 9.6 e.V. Leaves you high and dry on the dichloroalkanes though, since these typically have much higher IP's.

Best regards.