Troubles with GC-17A + FID

[Noxi]

Hi everyone,

I am a PhD student in New Caledonia (first year) and I try to restart a GC-17A -Shimadzu- to measure H2S and CH4 in natural gas (soil fluxes). I admit to having a lot of troubles with it. Maybe someone here will help me, I hope.



I use two different circuits:
- the first one with a capillary column GasPro 0.32x30m connected to a 5380 PFPD, which I use to measure H2S (Channel 2). These part seems working well with a standard of H2S.


- the second circuit passes through a column Porapak 80/100 (6ft*2) and then into a column Molecular Sieve 5A 80/100 (8ft) and finally in the FID (Channel 1).
The problem is whatever the gas I inject in the GC (Air, CH4, H2S) (or the volume of injection), I obtained the same signal on the FID. I don’t know what to do. You can find in attached the print-screen of the different chromatograms obtained.


I also put a diagram of my circuit. To simplify things, the TCD and the methaniser were removed from the flow path. Sample loops are 2ml and 100µl.



The method I use is that the engineer who installed the GC created for what I wanted to do.
Manual injection, injector 150 ° C, split, split ratio: 5.0
42 kpa pressure, total flow 7 ml / min
column flow 1 ml / min
Column 80 ° C, 220 ° C FID

10 Port Valve:
0.00 Position 0 Load
0.01 Position 1 injection
3.00 Position 0 Load

The FID is new and has never been used, so I think it works. Maybe I have a problem with my method, but I do not know where ..

Does someone could bring me some reflection tracks to solve this mystery, please?
Thanks in advance for any help,
Adrien

[rb6banjo]

The top part of your system is very complex. You have many places were dead volume can be added. This is a fancy way to say that your analyte peaks will be broadened, degrading the beautiful separation you have achieved on your columns. Is the FID chromatogram that you show collected after passing the sample through the molecular sieve or is that just through the porapak column? It would be interesting to see what happens in the flame if you don't divert to the molecular sieve.

I haven't used a methanizer in my career but it appears that there might be a problem of poisoning the catalyst with H2S:

http://en.wikipedia.org/wiki/Methanizer

[Noxi]

Hi rb6banjo, thanks for your response,

The chromatogram that I showed was collected after the sample is passed through the Porapak column and the molecular sieve column (Both are in series).
Infortunately I didn't had the time to make an other analyse without the molecular sieve column today, I'll try to do it tomorrow morning.


The top diagram is a bit complicated it's true, I simplified the circuit by removing the 6 port valve, the methaniser and the TCD. Here the new diagram.


My columns date 2003, they have been used very few times, but I wonder if they are not outdated..

Thank you for the link about methaniser, I now know that I will change the catalyst if I want to use it again

[rb6banjo]

I didn't have trouble following the diagram. My point is that there are many fittings. Each fitting is a place where your chromatography will degrade.

The age of the columns probably won't matter much unless they've been really abused in storage (high temperatures, no flow of carrier gas, etc.). Whenever I have to resurrect a system like this, I start as simply as possible. For yours, I think I'd choose methane as a model system. Bypass the molsieve until it looks good on the porapak. If all is not well coming out of the porapak, it's not going to get better later. Then, add the molsieve. Make sure you can make methane work well then add something else. Air might be a good (and easy) choice so that you can make sure it all looks good as you increase the complexity. You'll probably just be frustrated if you try to "shoot for the moon" out of the gate.

[Noxi]

I'm still here! Thank you for the advice. I removed the columns, and I test by simplifying the system (It's a bit long, Christmas break..).

[rb6banjo]

Hang in there. I am on Christmas break too.

[Noxi]

Hi,
Back after a short break. I disconnected the 6-ports valve, and bypassed the molsieve column. I then tried different samples: air, methane 100ppm. Nothing appeared on the chromatogram which stays stable. I think it's normal for air because CO2 is not seen by FID. But for methane...

Adding the molsieve column. Air and methane sample... I get the same results I've shown before.

Maybe there is a problem with the columns ? They remained for a long time without no carrier gas flow inside (maybe 3 or 4 years).. I'll buy a new one and see if it solves the problem. At least I'll be sure.

Thanks a lot for your help