Headspace analysis of gases - how to do calibration

[Karin Elisabeth]

Dear All,

I'm going to analyse gases evolving from a resin system that is curing thermally with headspace GC/MS analysis. The curing will take place in the glass vial.

My problem is how I should do the calibration? Some of the gases I will analyse is N2, O2, CO, CO2, methane. Because of the matrix effect I guess I cant just dissolve the gases in another non-curing "liquid".
I am not familiar with the headspace technique as you all already have figured out

I would very much appreciate your feedback.

[Don_Hilton]

How you calibrate depends on how you are going to run the analysis. I would assume that you are going to take a gas sample from the sealed container and inject mixture into the GC? If so, you can purchase gas mixtures and use them for calibration. It has been a few years since I've looked for these, but I remember Scott Specialty Gasses and Matheson Tri Gas as a suppliers for these kinds of standards. I am sure that there are others as well.

And your method of injection - direct or via a sampling loop will dictate your method of handling the calibration gas.

Let me know if this information is helpful.

[GasMan]

Which headspace sampler are you using?

Gasman

[Karin Elisabeth]

I am going to do the curing of the resin in the glass vial in the heated agitator. After a certain time period the autosampler will inject a sample from the gas phase on to the GC. My method of injection will be an injection in a liner or on-column injection (I don't know which injection is the best right now).

I have a Varian GC 3800 with a CombiPAL autosampler (with a gastight syringe and a headspace agitator that can be heated).

Thank you so much for your answers. I will look for a gasmixture. And if you have any more tips for me it will be very appreciated.

[larkl]

If you calibrate with the gas mixture alone, you may be biased due to partitioning. Since some of the analytes may stay in the solid phase when the samples are run. You would be better off if you could inject the gas mixtures into a vial containing the same volume of resin. Of course, the solid would need to be degassed. Or you can assume zero partitioning by using a very small amount of resin for each test, but you'd have to have sufficient signal for this to work.

[Peter Apps]

Hi Karin

The critical question is what is it that you want to find out ? If you only want to know how much of each gas the resin gives off you can calibrate using a gas mixture in a vial (getting it accrurately into the vial might be a challenge - if I was doing this work I would rather have a valve and loop injector that I could fill direct from a cylinder). If you want to know how much gas the resin produces you need to take into account the gas that stays in the resin, and then you need to calibrate inthe presence of resin. You can do this by known additions (the physical problems of adding gas to a vial remain) without having to degas the resin but the equilibrium times are likely to be extremely long unles the reesin is spread as a very thin film at the bottom of the vial.

Peter

[Karin Elisabeth]

Thank you so much for all your answers.

As you wrote Peter Apps I will only consider how much gas is evolving from the resin.
I don't have a valve and a loop today but I will look in to it no that I know that it might be the best way to do it. In the first trials I have to buy gasmixture and try to get them in the glasvial in a proper way (with some type of flowmeter I guess?)

How is the best way to do the injection if I don't have a loop? Is it with a gastight syringe in the liner that is connected to the column or is it on-column?

[Peter Apps]

Hi Karin

It is not so much getting the gas into the vial that is the proble, it is keeping it there while you get the cap crimped on wihtout it exchnaging gasses with the air by diffusion.

If you do not mind doing a bit of plumbing you could try a dummy vial made out of metal with an inlet and outlet tube to flush it with calibration gas, and a septum on the top. For a quick try out you could put an inlet and outlet line through the septum of an ordinary glass vial - with the CTC the vial is not pressurized in order to take a sample, so you could ive with some tiny leaks for the suck it and see phase of the development. Just be careful to route the lines where the syringe canno thit them, and you will probably have to lock the agitator cover back out of the way.

Peter

[Karin Elisabeth]

Hi Peter,

A good ide with the dummy vial. I understand how you mean but I must say although I'm quit handy I think this would be to difficult for me .

What do you think about the injection if I don't have a loop? Is it okey to do it like a normal headspace injection onto the column via a liner?

I think I have to do a bit of a reading about analysis of gases. I think it is a bit challenging, but fun

[Peter Apps]

Hi Karin

Injection with an empty double gooseneck splitless liner should work fine.

How are you going to get the air out of the vial when you have first loaded it with resin ??

Peter

[Karin Elisabeth]

Thank you so much for your concern.

I was thinking of cap it with a metal/teflon septa and flush it with argon. One needle in and one needle out, and just before stop flushing take the outlet needle out so I get a slight overpressure.
Or how should you do it? Without the plumbing

[Peter Apps]

That should work. It is probably a good idea to run some tests to see how much argon you have to flush through to get the air background low enough.

Peter

[larkl]

re- getting the cal gas into the vial, we routinely just use a gas tight syringe to inject a known volume (at atmospheric pressure) through the septum into the vial.

re- Getting the sample into the GC. We've used a heated syringe and manully injected from the heated vial. It works, but a bit more tedious than an autosampler.

[Don_Hilton]

For the calibration gas, I would look for something that can either be evacuated and filled or flushed without using up too much of the calibration mixture. The folks who make the calibration gas mixtures may sell something.

If not, you can use something like a small Tedar bag, which can be evauated and filled. Or, you can construct some kind of flushable cell with a septum on it for sampling the gas. You would have gas flow through the cell to purge it and then sample with a syringe.