Page 1 of 1

Gas Sample Injection

Posted: Fri Feb 08, 2019 8:09 pm
by Samuel_Brister
I am currently working to build a mixed gas permeation instrument with an Agilent 6890N GC at the end for analysis of the permeate gas (H2, O2, N2, CO, CO2, and CH4) as my senior engineering design project at University. I will be using the split/splitless inlet with a 1.0 mm ID Straight Inlet Liner, a J&W Select for Permanent Gases/CO2 GC Column, 7 inch cage column, and analysis will be done with the TCD.

Most of the resources that I have found only discuss introducing liquid samples to the GC. How can I connect my Mixed Gas Instrument to the GC and control injection of samples?


Thank you!

Re: Gas Sample Injection

Posted: Fri Feb 08, 2019 8:45 pm
by James_Ball
Probably the easiest way would be to use a switching valve and a 1-2ml sample loop. Let the effluent of the permeate flow through the valve and loop then switch the loop into the carrier flow and allow the carrier to sweep a few loop volumes onto the column, then reset for next sample.

Re: Gas Sample Injection

Posted: Fri Feb 08, 2019 9:01 pm
by Samuel_Brister
Thank you for the quick response!

Could you explain what you mean by "Let the effluent of the permeate flow through the valve and loop"?

Re: Gas Sample Injection

Posted: Fri Feb 08, 2019 9:27 pm
by Samuel_Brister
James_Ball wrote:
Probably the easiest way would be to use a switching valve and a 1-2ml sample loop. Let the effluent of the permeate flow through the valve and loop then switch the loop into the carrier flow and allow the carrier to sweep a few loop volumes onto the column, then reset for next sample.


I am new to the GC world, so I may be have just misunderstood. How can I connect my mixed gas instrument to the switching valves? What is the physical connection that I will have to implement?

Re: Gas Sample Injection

Posted: Fri Feb 08, 2019 10:55 pm
by James_Ball
I may have misunderstood how you are generating the permeate gasses. If you have a machine that is generating the gasses that need to be sampled you would connect tubing from the outlet of the machine that generates the gasses into the inlet of the sampling loop. The gas would flow through the sampling loop and out to waste or wherever you collect it, and at the interval you wish to sample you just switch the valve so the carrier sweeps the loop into the inlet while the gas flow from the machine goes through the bypass side and out the normal exit as when loading the loop.

If you are sampling the permeate into a syringe or tedlar sampling bag, then you attach that to the inlet of the sampling loop, flush the loop with enough gas to sweep the loop a few times, such as if you collect into a 10ml gas tight syringe, you push about 5ml through a 1-2ml loop, then you switch the valve so the carrier flows through the loop and into the inlet.

This is the type of valve I am thinking of, fairly simple setup.

https://www.restek.com/Sample-Handling/ ... lves-Loops

http://www.supelco.com.tw/A-9%20%20%20316-317.pdf

Re: Gas Sample Injection

Posted: Mon Feb 11, 2019 9:24 pm
by AICMM
Mr. Brister,

Sounds similar to another application I have worked with. One thing you might need to consider is some gas flow running on the permeate side to carry the gases you are interested in, over to the gas sampling valve on the GC that GasMan is suggesting. You could use a small stream of helium or argon, depending on your carrier and the peak for the "transmission" gas would disappear into the carrier of your GC.

Remember, you will be faced with the issue of TCD sensitivity to hydrogen versus all the other components on you list so you might not be able to catch the whole list with every run.

Best regards,

AICMM

Re: Gas Sample Injection

Posted: Tue Feb 12, 2019 7:02 pm
by Samuel_Brister
AICMM,

That is exactly what we have planned, we may switch to argon or nitrogen as the carrier when testing H2 for greater sensitivity. I am still unsure what pressures and concentrations that the instrument will be running for both the carrier gas and the permeate gases, making it difficult to make a choice of the sample loop volume and the number of valve ports that I would need.

I believe that the ShinCarbon ST micropacked column would be more suitable for my application due to cost constraints, I saw this post viewtopic.php?t=16466 that discusses someone else using this column with the valve configuration found here;
https://www.vici.com/support/app/st_app.php
Is this diagram applicable to my use case?
I do not believe that I will need to seperate all of the gas components in one run, we will be purchasing a tank of a dual gas mixture from airgas for testing so only two components and the carrier gas will need to be seperated during each run.

Thanks,
Samuel Brister

Re: Gas Sample Injection

Posted: Wed Feb 13, 2019 7:23 pm
by AICMM
Mr. Brister,

Permeate pressure and carrier pressure are relatively independent of each other since the sample loop is out of line until injection. You might get a bit of baseline disturbance early but should not be a huge issue.

Remember that a ShinCarbon does not fully separate O2 and N2 which is not a huge deal in some cases. Remember as well that ShinCarbon will elute (eventually) water and mess up subsequent chromatograms until you bake it off. This limits the number of consecutive injections you can make if your stream is wet. All that said, I like that column a lot!

Finally, keep in mind the dilution factor of known sample carrier stream flow and analyzed components.

Best regards,

AICMM