Chromatography Forum

Hi everyone,

I am using a Agilent 6850 Network GC with a splitless injector to analysis for residue ethylene oxide testing. I currently calibrate my GC using a range of ethylene oxide in water standards (1ppm - 100ppm). These standards are made according to ISO 10993-7. I have purchased a range of ethylene oxide gas standards in Nitrogen at between 1ppm - 100ppm. I am having problems with the calibration of my GC using the new gas standards. I can't get good linearity, my results are all over the shop is it something to do with the injection volume or should the injector be a split and if so what should the ratio be.

All and any help is welcome.

Are you manually injecting using a gas-tight syringe?

In my experience, manual injection of gas stds is highly dependent on technique and developing a uniform injection action greatly improves results. Also NEVER vary the injection volume , always inject a uniform volume.

I've not had a lot of experience with gas injection valves, so I can't help if that's how your injecting your samples.

Hi JGK,

I am manually injecting the standards. Should I use a split or splitless injection??

Best headspace results are made using an automated headspace system. That said: you may need to make your injections more consistent, or live with your results.

Phardy,

It has been my experience that doing gas syringe injections, especially larger volumes, with a splitless injector (or packed port) causes all sorts of problems. When the 6850 goes splitless it essentially acts like a packed port.

Thus, I would suggest you either cut down to a very small injection volume or introduce a small split flow (2:1 or 4:1 perhaps.) With a small split flow, the pneumatics configuration should be able to handle the pressure pulse of a syringe injection a bit better.

Also, keep in mind that your gas standard is probably mole/mole (most are) so you probably won't get an exact match between water concentration (which is usually mass/volume) and gas standard.

Best regards.

Hi, I have some the same problem, can you said me how did you to resolve the problem
Which parameter did you using to improve the anslysis

Thankyou

Gustavo