Page 1 of 1

problem with gases detection

Posted: Tue Jan 24, 2012 2:53 pm
by MariaPC
Hello everyone!

I'm Maria and I'm new in the forum. I find it very interesting! :)

I have a problem since I need to add Kr and H2O in my system and Kr interferes with the CH4 signal so I cannot get separated peaks in the chromatogram while I cannot detect the H2O.
I usually follow CH4 in the FID but it is also detected in the TCD so it's there where Kr and CH4 are solaped.
My carrier gas is Ar.
So the common problems are Kr is solaped with CH4, O2 with CO, He with H2 (in the TCD).

I'm currently using a Hayesep D column 3m, 1/8" ID, 100/120 mesh, I thought in getting a longer one but I'm not sure if that gonna help.
I tried changing the method but it didn't help.

Do you have any idea if another column would be better? or what parameters could be changed in the method to get all the compounds separated?

Thanks in advance,

Maria

Re: problem with gases detection

Posted: Tue Jan 24, 2012 7:26 pm
by chromatographer1
Maria,

You are trying to do the impossible.

To measure water D is not the best packing. Use R in Hayesep or Porapak.

You must have a very inert column tubing and sample loop to measure water. Plus the temperatures required are higher than you should use to separate the fixed gases (90 to 110°C). Water elutes before propane and after ethane.

Krypton should elute with nitrogen on a D column, not methane, unless your methane and nitrogen separation are not good to begin with. To separate Kr from N2 you need a 2 ft Carboxen 1000 60/80 mesh packed column.

Oxygen and nitrogen do not separate on a D column unless extremely low temperatures are involved.
Mole sieve 5A or 13X in 3m lengths or a 15ft Carboxen 1000 60/80 mesh column will separate them at 40-60°C

To separate hydrogen and helium you need a 6 to 8 meter 1mm ID D packed column or a 3 meter 2mm ID Carboxen 1000 at 40°C

You are trying to separate fixed gases that require more than one packing to be separated well.

good luck,

Rod

Re: problem with gases detection

Posted: Wed Jan 25, 2012 1:16 am
by MariaPC
Hii, maybe better use mass spectrometer...Hopefully I could get better results!
Thank you Rod!

Maria