GC analysis for O2 and H2 using GC-TCD

[Ellina]

Hi,

I have a problem during running my GC analysis. In my analysis I have to used two type of gases such as O2 and H2 in different method. I've used 2% H2 balance N2 and 20%O2 balance N2. My RT for O2 is about 0.625 and H2 is 0.782. My question now why when I'm inject the gas H2 standards in O2 method it can detect the O2? Sometimes the area become very big. I think supposed in O2 method cannot detect any H2. Actually I used H2 as a purging method to remove O2 in our molecular sieve. Is there any method to separate this gases? I already do the method development but still get this results. Hope anybody can help me

[GasMan]

Can you give what type of Molecular Sieve you are using. What is the length of the column and what temperature and flows are you using. The order of elution on a molecular sieve column will be hydrogen, oxygen and then nitrogen.

The choice of carrier gas is also important for hydrogen analysis. At the 2% level for hydrogen, you should be using nitrogen or argon as carrier gas, this will give you lower sensitivity for oxygen, but at 20% this should be no problem.

Gasman

[Ellina]

Dear Gasman,

I'm using GS-Q column which have 10 m length, 0.53 mm ID. My oven temperature is 40 Deg C and my flow rate is 4.0 for H2 and 5.0 for O2. Molecular sieve that I give before is not for the column. Actually like this, we used CMS molecular sieve to trap the O2 gas. We allowed 20% O2 into the vessel which have CMS molecular sieve to trap the gas before enter to the GC. The molecular sieve only retain certain limit to adsorb and after that O2 will go to our GC. After that we change the O2 cylinder with H2 cylinder to purge out O2 from the molecular sieve. My problem now, maybe three reading have O2 peak and after that don't have anymore because have another gas but now, O2 peak can detect after the three reading.

[GasMan]

Ellina,

I would not recommend the GC-Q column for the separation of hydrogen and oxygen with nitrogen present. The GC-Q column will not separate nitrogen and oxygen. For these gases, to get complete separation, you will have to use a molecular sieve column.

It is not quite clear to me what you are tring to do. Are you passing a gas through this CMS sieve ( I presume that CMS stands for Carbon Molecular Sieve) to a sampling valve on your GC and then injecting the sample. What carrier gas are you using.

Gasman

[Ellina]

Dear Gasman,

I don't want to detect nitrogen peak..I just want to detect oxygen and hydrogen only.
Yes, I pass through the gas from gas cylinder into vessel which is contain CMS and then the gas will go to the GC column. I used oxygen method to detect the hydrogen. How I want to separate this peak because the RT is quite near? What is the suitable method for GC chemstation? I always have a problem like this. When do the calibration is ok, after I run using the vessel the RT is changed. Why this happened?

[chromatographer1]

Ellina

I appreciate you are seeking help.

I have to agree with GASMAN that your procedure does not make sense as you have described it here. Would you please restate your procedure and explain in more detail how you process your gas sample and your chromatography parameters?

Your passing a gas sample through carbon sieve is confusing and the purpose for doing so is not clear. You will not get much separation from oxygen and hydrogen using a porous polymer column, especially using a short capillary column. It appears that you are using oxygen as a carrier gas. Really?

Gasman and others would give you advice but it is difficult based on your post. Please clarify the details of what you are measuring and how you are performing your test. Thanks.

best wishes,

Rod

[Ellina]

Dear Rod,

Maybe you are confuse what I mean the molecular sieve that I used. It is not type of column that I used in GC. The CMS that I mean is molecular sieve like beads which have cylindrical shape in white color. The function of this molecular sieve is to adsorb oxygen gas in the vessel before enter into the injector. My GC have valve gas sampling. My carrier gas is nitrogen. I don't know what is wrong because the hydrogen and oxygen peak is combined together. The area of the peak become broad. My oven temperature is 40 Deg C.

My application is like this:

gas cylinder -> vessel (CMS) -> GC injector -> column -> detector

[chromatographer1]

The molecular sieve you pass your gas through may absorb some of the oxygen in your sample but it will also absorb all the other gases you wish to measure, including hydrogen, nitrogen, methane, carbon monoxide, carbon dioxide, and any hydrocarbons present. It will absorb these gases until the sieve is saturated. Then it will allow the gases to pass to your valving injector where they can be injected onto your column.

I assume the sieve is not already saturated already with any of the individual gases you have mentioned.

This analysis procedure does not make sense to me.

If you want to measure hydrogen and oxygen, why do you pass your sample through a sieve that may capture a portion of the sample content?

Normally a sieve is only used to remove water, hydrocarbons, or sulfur gases from a gas stream.

Why you wish to partially trap hydrogen and oxygen on a sieve (I assume it is an alumino-silicate sieve since it is white, a 13x, or 5A, or 4A) is not obvious to me.

I wish I could offer you advice but I do not understand your procedure or the measurement you wish to achieve following the steps you take in your description of your method.

best wishes,

Rod

[Ellina]

Hi,

My analysis is like that. We want to trap the oxygen using the molecular sieve. We want to how long the gas will adsorb by using the molecular sieve. Hydrogen gas is used to purged the molecular sieve (to clean from oxygen). My problem now how to separate the peak of hydrogen and oxygen. It will combined together in the chromatograph. I cannot differentiate the compound. What is the suitable method that can separate this compound.

[chromatographer1]

OK. I am with you now.

To separate oxygen and hydrogen you will require a PACKED porous polymer column, using almost any of the porous polymers but typically Hayesep D 100/120 mesh, using a length of 9 meters. I would suggest a 2mm ID.

Alternately, you can order a 3m Molecular Sieve 13X packed column to do the same analysis (see Supelco catalog).

Temperature of your oven can be ambient or up to 80°C as a suggested maximum.

40°C is a typical oven temperature.

I hope this answers your question.

You are welcome to contact Supelco technical service for additional advice.

best wishes,

Rod

[Ellina]

Thank you Rod.

I'm using Agilent 6890N GC TCD. Is that suitable to used another brands of column? I only used capillary column before this and never used packed column. How I want to install this type of column?

[chromatographer1]

Yes, you can use other brands of columns in your GC.

Since I assume you are using a 6 port sampling valve you will be able to connect a packed column to it with the proper connectors.

You may require adapters to connect to your detector. You should contact your local Agilent technical service agent for assistance.

Supelco has taken over Agilent's packed column business officially. I would suggest you order your packed column from Supelco (yes I do work for Supelco).

best wishes,

Rod

[Ellina]

Thank you so much Rod for the information

[GasMan]

Ellina,

If you can give me the serial numer of the 6890, I can probably find out what you had installed at the time of delivery of your GC. From this, we can advise you on what to order to work with packed columns.

Gasman