GC, MS or GC/MS for gas analysis at trace levels ?

[lgallet]

Hello,

I am working in an academic research group working on molten salts applications for nuclear reactors. We are now conducting 3 experiments needing a gas analyzer.

In each of those experiments, we want to analyze the composition of the gas phase of our containers. The gas phase will be Argon with traces of other compounds, probably at ppm levels like H2 and isotopes, HF, F2, CF4, CH4, and probably carbon oxydes.

We are now looking for information on what technology we should use, particulary between GC, MS, GC/MS. I found it difficult to see clearly a difference between those technology for this type of application.

I am novice in this kind of technology, and looking for advices from people that are more used to GC and MS (you guys!). What would be the pros and cons for this type of application, for those devices?

Thank you very much for your time

[Peter Apps]

Welcome to the forum.

That is a very challenging list of target analytes - you are not going to be able to buy a gas analyzer that can do them off the shelf, and some of them (isotopes, fluorine compounds) require specialist equipment and expertise. You will do much better to collaborate with someone who is already set up for these analyses than to try to do it yourself.

Peter

[BMU_VMW]

since they are in gas-fase, there are not to much analytes so they don't need separation it might be possible to skip the gc-part en use e.g. a syft-ms??

[lgallet]

Thank you very much for both of your answers.

Indeed it is very challenging, I think the best solution would be to target 2 or 3 different gas analyzers that are more specific. But the problem is that even if a gas analyzer can detect well a specific type of compound, there are still the other ones that could compromise the analysis I guess.

[mckrause]

A RGA (residual gas analyzer, or MS) will only get you some of these compounds. You will definitely need other detectors. Most semiconductor manufacturers do this type of analysis routinely on their LPCVD systems, so there are off-the-shelf applications that you can obtain for somewhere around the gross domestic product of small countries. The setup and calibration is beyond complex; I spent years getting systems to function while I was at Motorola. I very strongly encourage you to seek external expertise.