Hydrogen Detection on TCD

Chromatography Forum: GC Archives: Hydrogen Detection on TCD
Top of pagePrevious messageNext messageBottom of pageLink to this message  By kyaccato on Tuesday, August 7, 2001 - 06:18 pm:

I'm trying to set up 2 systems in house and would like feedback from anyone w/ experience in hydrogen analysis by GC.

System a)
CO, CO2, H2, Kr - TCD, He carrier gas

This method works fine - except that my hydrogen peak is super small! It is 70% of my gas sample - but is less than 10% of the GC peak areas...
I expected lower sensitivity for hydrogen - but this is really bad - is this reasonable or am I missing something?

System b)
very low levels of hydrogen (1% and to 100ppm)
with very high sensitivity requirements

My plan here was to run a molecular sieve w/ argon as carrier gas - however, b/c I have only 1 inlet - I need to use helium as the carrier through the inlet and argon as the carrier on the auxilliary gas supply (peak trapping valve. Does anyone have experience w/ optimizing Helium separation from Hydrogen - I will definitely need baseline analysis here - and I see nothing on this in the column manufacturer litterature on Plot Mole Sieve columns...

thanks,
karin


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Kuba on Wednesday, August 8, 2001 - 02:04 am:

I deal with analysis of gases especially of gaseous mixtures. I have no experience with pure hydrogen. But it is very strange that your hydrogen peak is small it should be large for 70% of hydrogen in mixtures. Try to inject pure hydrogen and compare it to your peak being analysed. What chromatograph do you use?


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Mike D on Wednesday, August 8, 2001 - 05:50 am:

I have occasionally analysed hydrogen/helium mixtures. Standard texts warn that hydrogen response with helium carrier is small or anomalous. The type of response must depend on TCD design. With my Perkin-Elmer Autosystem I can't see less than 2% hydrogen with He carrier. Now as to H2/He mixtures I found the separation on 25 m x 0.32mm Mol Sieve 5A PLOT column (50oC, 16 psi nitrogen carrier)was OK to measure either component down to 0.5%. He eluted in 1.18 mins and H2 in 1.27 mins. Slightly better resolution might be available at 30oC but to get the full performance of the capillary PLOT you must avoid overloading. My samples were equivalent to no more 2-5 microlitres of gas on the column and of course this makes trace analysis difficult.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By jclark on Wednesday, August 8, 2001 - 12:15 pm:

You should be able to "see" 1% H2 using He carrier gas on a 5A MolSieve with TC detection under the most optimal conditions. More realistic is the use of N2 as carrier gas with a TCD. Thermal conductivities of H2, He, CO and N2 are 41.6, 34.8, 5.6 and 5.8, respectively.
In GC Troubleshooting column, John Hinshaw notes peculiarities that occur with H2 and He carrier: LC-GC 8(2).


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Kuba on Thursday, August 9, 2001 - 02:57 am:

Using N2 as a carrier you could have problems with separation between H2 and He.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By RBora on Thursday, August 9, 2001 - 08:32 am:

A 5A Mol Sieve will separate H2 and He nicely using N2 as a carrier gas. I do this analysis on a somewhat routine basis.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Anonymous on Thursday, August 9, 2001 - 08:42 am:

We have had better sensitivity for low levels of hydrogen using the Valco PDID than can be obtained with a TCD. You may want to consider that for the second, high sensitivity system.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Kuba on Friday, August 10, 2001 - 02:48 am:

DID I very sensitive but expensive. I use TCD detector to determine ppm's of argon and nitrogen in oxygen using helium as a carier and it is not a problem to obtain level 100 ppm's for both so determining hydrogen at level of 100 ppms it shouldn't be a problem.


Top of pagePrevious messageNext messageBottom of pageLink to this message  By Bruce Freeman on Friday, August 10, 2001 - 10:59 am:

Commenting on the first message by Karin,
Measurement of hydrogen in helium goes way back. Back in the 1960's, I think, it was demonstrated that hydrogen gives an anomolous response on the TCD. One would expect to get a negative peak from hydrogen because it's TC is greater than that of He. However, if I recall correctly, small amounts will give a positive peak, whereas large amounts give a negative peak. The cross-over point (zero response) was somewhere around 6% H2 in He, so such a mixture can be used as carrier gas to give all negative peaks, regardless of amount of hydrogen present.

Small peaks would be expected if the TC of hydrogen is similar to that of He. Frankly I haven't used a TCD for so long I just don't remember whether this was a problem. (TCD response factors might be published in a handbook of chromatography.)

For system a, you don't say what sort of column you're using. I haven't kept up with available columns for such separations, but in the past we used columns packed with 5A molecular sieve. Such columns were incompatable with carbon dioxide, which was irreversably adsorbed.

System b: I simply don't understand what you mean by "I have only 1 inlet - I need to use helium as the carrier through the inlet and argon as the carrier on the auxilliary gas supply" Please clarify why you can't just use argon as carrier.


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