Chromatography Forum

Help!

I've been running my GC for sometime now. I am using Molecular Sieve 5A and Poropak Q to separate my peaks. Before, I can detect peaks of N2 and O2 (that is when i inject air to calibrate) but now, it suddenly disappeared. But I am still getting one big peak and another negative peak.

Do i need to replace my column?

Give us a bit more detail. I figure you are running gas samples, but how are you injecting? syringe, headspace sampler, sampling loop?

And, have you done any maintanace on the instrument between the time things were working and then they stopped. Even to changing a gas cylinder...

And, have you checked the system for leaks?

It's an on line GC. I have a valve that redirects the flow of my gas to the TCD if I want to do my sampling.

I haven't done anything. I noticed that my peaks becomes smaller until one day they eventually disappeared.

Maybe check for leaks again?

The fact that you have two columns suggests that this is not a simple analyzer. The Porapak Q column is there to stop certain compounds from going into the Mol Sieve column. I would guess that the valve is a 10 port valve which is configured to inject the gas sample and at some point in the run, backflush the Porapak column. You could have valve timing problems.

You need to tell us what valve is installed and how it is configured. The GC supplier should have given you a diagram for your GC. Is this a rotary valve or a diaphragm valve.

Gasman

I suspect you may have introduced some water into your mol sieve column. This can affect the separation. Have you tried baking the column at a higher temperature to see if that helps?

Can I detect water in my chromatograph? Will it show in form of a peak?
Though I am thinking that it would take huge amount of water before I can ruin my mol sieve.

But will try to bake my column.

The fact that you have two columns suggests that this is not a simple analyzer. The Porapak Q column is there to stop certain compounds from going into the Mol Sieve column. I would guess that the valve is a 10 port valve which is configured to inject the gas sample and at some point in the run, backflush the Porapak column. You could have valve timing problems.

You need to tell us what valve is installed and how it is configured. The GC supplier should have given you a diagram for your GC. Is this a rotary valve or a diaphragm valve.

Gasman

It's a six way valve that redirects the flow of the gases to my GC. Porapak Q is used to determine CO2 and xylene while mol sieve is for O2 and CO. I've injected air just to check the separation in my mol sieve. But since there is none, I am really troubled.

Are you getting results from the Porapak column? (I'm trying to figure out how far back in the system things are falling apart.) Also, are we correct that the sample first flows throug the Porapak column and then through the Mol Sieve - as in from one column to the next?

Are you getting results from the Porapak column? (I'm trying to figure out how far back in the system things are falling apart.) Also, are we correct that the sample first flows throug the Porapak column and then through the Mol Sieve - as in from one column to the next?

I used to get CO2 peaks from porapak. But now it seems that they have disappeared also.

My gases flow to a porapak Q, then to a TCD detector, then to Mol sieve then to another TCD detector then out to the vent.

If I guess correctly the only valve in this system is the 6-port valve at the inlet? If so, it would appear that the sample is not getting to the first detector -- or the first column is bad. Past espericned with other GC systems says that 80% of the time, the problem is in the inlet system - no matter what kind of system. (I've not touched process GC's, but I think it is still a safe bet.) Check for leaks and for proper opertion of the 6 port valve. Be sure that there are no leaks within the valve or to the outside of the valve and be sure that no path in the valve is plugged. And be sure there are no leaks around any fittings going from the valve into the first column and to the first detector.