Oh, where to begin? MolSieve 5A and CO2

[wave_man]

Hi all,

I am new to the group and fairly new to GC techniques. I occasionally listen to good advice so give it a shot!

I have an Agilent MicroGC with three columns:
A- MolSieve 5A with PLOT U Pre-Column
B- MolSieve 5A with PLOT U Pre-Column
C- PLOT Q Column

TCD's on all.

We run Argon in A channel, Helium in B and C. We have a lot of hydrogen to detect is the reason for A and Argon (hydrogen reforming research for fuel cell applications).

I currently have the GC in Agilent's shop to add a fourth column, an OV-1 to detect C2+ compounds as a result of incomplete heterogeneous catalysis of longer hydrocarbons (ultra-low sulfur diesel, biodiesel, kerosene, etc.).

I have been working with a gas supplier to get a set of calibration gases for the new column. I wanted the gases to give some new light to my old cal gas set, so I put in some stuff that I knew would not elute discretely in the OV-1, and the guys at Mesa said, hey, this stuff is going to co-elute, and did you know the CO2 is going to foul up your MolSieve 5A? Fact is, I didn't.

My old cal gases are like so:

Code: Select all

       cal1      cal2     cal3
H2   50%     89.1%  0
CO2 20.1%  5.1%    0 
CO   10ppm 202ppm40ppm
CH4 199ppm1028ppm 0
N2   bal.       bal.      bal.

And the gas out of a reformer typically has between 1 and 15% CO, and between 5 and 25% CO2, so these were good values to dial in the range. Now I am reading how the MolSieve 5A is going to get toasted by the CO2. I run 24-48hr bake-outs at about 190C religiously. Is this enough to desorb the CO2 that I will be adsorbing from my samples?

My new cal gases are like so (pending changes after the PO, which I can do):

Code: Select all

                    cal4          cal5          cal6          cal7
H2                                20%         40%         10%
CO                5%           10%         1%           3%
CO2              25%         30%         10%          15%
CH4              1%           3%           5%            0.5%
ETHANE         1000ppm   0               200ppm   0
ETHYLENE       0               1000ppm                  200ppm
PROPANE       1000ppm                  200ppm
ALLENE                          1000ppm                  200ppm
BUTANE         1000ppm                  200ppm
1,3 butadiene                 1000ppm                  200ppm
1-BUTENE      1000ppm                  200ppm
HEXANE                         1000ppm                  200ppm
BENZENE       1000ppm                  200ppm
TOLUENE       1000ppm                  200ppm
N2                 balance     balance   balance      balance

Am I making a big mistake with the CO2? Is my bake-out helping at all? Should I be doing something different? I realize that what I calibrate with, given the relative frequency of events, is not nearly as important as what I am sampling. Do people scrub out the CO2 to protect their columns? I suppose one could...

Sorry about the crappy looking tables, I am new to forum-typing too.
It ain't quite LaTeX.

Help!

wave_man

[chromatographer1]

Hi Wave,

First, relax about toasting your MS5A PLOT.

CO2 should elute readily with a bakeout of a few hours. Check with Agilent to make sure I am correct. If I am not, shame on me and Agilent, too.

The hydrocarbons if placed on the sieve will take longer to remove. I would expect a 24 hour bakeout to be sufficient.

Do not heat the column much hotter or you might damage the sieve. Heat damages the sieve with time as another factor. More heat, more time, more damage.

By all means, contact Agilent and ask them the questions you are posing here. They can tell you if you might be damaging your hardware or columns from the proposed treatment.

best wishes,

Rod

[Bruce Hamilton]

My limited experience with a MS5A PLOT column and a similar reaction mixture was that CO2 required 90 - 120 mins to elute at 190C.

However, if the mixture also included water vapour, the removal of both water and CO2 from the MS5A became rather complex, and CO2 had a nasty habit of eluting whenever it felt like it.

It seemed as though the reduced capacity of the PLOT ( compared to a packed column that we used previously ), made the CO2 retention more variable.

I overcame the problem by increasing the head pressure / column flow rate during a several hour overnight conditioning run.

I agree with Rod, talk to the Agilent experts, they should have some pointers.

Please keep having fun,

Bruce Hamilton

[chromatographer1]

To give you an idea how long it takes to elute CO2 from a 3m SS 2mm ID 60/80 mesh MS5A packed column I offer the following chromatogram on an Agilent 6890 with a µTCD. The packing is a special preparation of MS5A from Supelco.

This chromatogram shows fixed gases (1-H2, 2-O2, 3-N2, 4-Methane, 5-CO, 6-CO2) at 150°C using He carrier at 30cc/min. (0.75mL of nitrogen containing 2500 ppm H2, 1000ppm O2, 250 ppm methane, 375 ppm CO, 500 ppm CO2)



best wishes,

Rodney George
Senior Research and Development Scientist
Gas Separations Research
Supelco
595 North Harrison Road
Bellefonte, PA 16823

814-359-5737 voice
814-359-5459 fax
rodney.george@sial.com

[GasMan]

You should check what you have as an injection module for your channel with the 5A column. It is possible to get a backflush module which has a Porapak PLOT U column as a precolumn. This would prevent the CO2 from going onto the Mol Sieve 5A column.

Gasman

[GasMan]

wave-man,

I re-read your original message. You have no problem, as you state that you have the PLOT U columns as a precolumn, so you can backflush the CO2 and hydrocarbons before they get onto the Mole Sieve column.

Gasman