Helium concentration determination

Chromatography Forum: GC Archives: Helium concentration determination
Top of pagePrevious messageNext messageBottom of pageLink to this message  By Eric on Wednesday, February 7, 2001 - 03:42 pm:

I need to analyze gas samples that contain a mixture of He/Ar, or He/N2 and determine the %He in the mixture. I have been sending my results to two seperate labs. One uses a Mass Spectrometer and the other a gas chromatograph. The difference in the results from the two labs is about 6%[He], with the mass spec. being low compared to the GC.

Any idea why there would be such a large difference?

Does anyone know of a GC that can determine %He in a mixture of Ar or N2?

I tend to trust the mass spec. as when it was used to test the %He in a certified premixed gas is matched the results of a third lab within 1%.

Thanks for the help. I am quite new to GC testing.

Eric


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Friday, February 9, 2001 - 10:00 am:

You may be able to perform the analysis using a Carboxen 1006 GC column (Supleco) and sub-ambient oven cooling (you may need to go as low as 0 or -10C). We have been able to separte O2 and N2 with such a system. We used He carrier gas and a thermal conductivity detector. The TCD is not as sensitive as a MS or FID. Its sensitivity depends on the difference in thermal conductivity between the carrier gas and the analyte. At 6% He, I believe nitrogen carrier gas will provided a reasonable signal. Good luck.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Spencer on Friday, February 9, 2001 - 09:56 pm:

Any conventional GC can do what you want it to do. Use a gas sampling valve and a molecular sieves column with TCD detection. Choose your carrier gas wisely.

That said, we have chosen to do fixed gas determinations routinely in our lab with an MTI gas analyzer (now sold by Agilent under the name MicroGC or something like that). The petroleum industry uses these GCs extensively because they are high speed, portable, sensitive, and parallel (up to 4 simultaneous injections onto 4 separate columns in the MicroGC). In fact, this instrument is usually generically called a refinery gas analyzer. Basically, its a small, multi-oven/column instrument with solid state injectors and detectors loaded with short columns with conventional GC phases. The column used for the type of determination you are interested in is usually a molecular sieves column. If you're looking for He in Ar, you use H2 or N2 as a carrier. If you are looking for He in N2, use Ar or H2 as a carrier. Or some variation of the above. These instruments excel at fast, fairly accurate, precise light gas determinations.

What perhaps may be your problem is differing calibration methods used by the labs you are sending your samples to. I would check with both labs and ask them for their methods and/or validation type information.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Bill Jenko on Wednesday, February 14, 2001 - 08:20 pm:

You should be able to determine He in either N2 or Ar using any standard GC with TCD and gas sampling valve to about +/- 1/2% RSD routinely. The process analyzer folks (Siemens AAI or ABB Analytics) do it routinely. Similarly equipped, any major brand of lab GC will do the trick as well.

Just one point of disagreement with Spencer: I do not recommend you use Hydrogen as a carrier gas for determination of Helium in any gas sample (or vise-versa, H2 in He carrier). Although this can be done with great care, the thermal conductivity of mixtures of He and H2 is wildly non-linear and leads to well-known problems of response and funny negative peaks and "W" shaped peaks. I would bet this is the problem with your current GC analysis, i.e. you are using H2 carrier and are assuming a linear response for He. In gas analysis, a 6% error is massive. Best discussion I ever read of this phenominon is in a very old Varian technical note ("AT-Works #12 or 15 I think, by Brian Thompson). As I say, measurement of He in H2 carrier can be and is often done, but N2 or Argon carriers will work better with nice linear response curves(obviously you won't see N2 in N2 carrier, nor Ar if you use Ar carrier) and since Helium has a vastly different thermal conductivity from either Argon or Nitrogen, the sensitivity will be great in either case.

If you absolutely insist on seeing peaks for He, Ar and N2, you will need some other carrier, such as CO2, Methane, or Neon. You can also use Helium carrier, which will give nice peaks for N2 and Ar, and do the Helium by difference. This will work very nicely for He concentrations of about 5% or higher.

Beware of H2 in your sample, as it will measure as Helium. You need about 15 ft. of 1/8" Molecular Sieve 13X to separate He and H2 at 25 C, if memory serves, or about 20 ft. of Porapak or Hayesep Q to do that separation.

See Brian Thompson's book: "Fundamentals of Gas Analysis by Gas Chromatography" still in print and available from Varian I hope, pages 72-77 in particular.


Posting is currently disabled in this topic. Contact your discussion moderator for more information.