how often should we re-condition Porapak Q column

[cheq]

Hello,

would anyone know how often should we recondition a Porapak Q column for the GC. I see a dual peak on the nitrogen region for the first couple of hours after the GC is swtich on each time when i pass helium or ppm level of nitrogen through. There might be ppm level of leak on the line feeding to the GC, but i would not expect the dual peak to appear.

would reconditioning the column eliminate the 2nd peak? this really annoys me.

thanks,
george

[chromatographer1]

Reconditioning the column should be of no help whatsoever with a split peak. The packing is spherical porous plastic beads (a dense styrofoam might be a closer description). Unless you have contaminated the beads with (to use a scientific term ) 'junk', there is little use in cooking out the column, it only accelerates the normal shrinkage of the beads which can produce voids and even more split or ever more increasing broader peaks.

IF, it is a column cause, then there is a void, a partial blockage, or an excessive dead space in the column or system.

But as usual, sample content, temperatures, amount injected and how it is injected all have bearing on the correct deduction of the problem.

Your question is like asking. What causes my car to 'stumble' when I drive it under 25 mph. There are a number of causes but without more information, all you will get is guess.

best wishes,

Rod

[cheq]

Thanks Rod,

It is a Varian 450-GC. I have a MS 5A and a porapak Q column in there. what i am doing now is bypass the MS column and just run the sample through the porapak. The GC injects samples from a SS tube connect to the sample inlet of the sample valve.

I always operate the oven at 80C with TCD 200C and filament temp 220C. I set the range to be 0.05 to get the most sensitive out of that possible.

I have notice this when i was calibrating the GC with pure CO2, but then i thought it might be some minor N2 content in the CO2 cylinder.

I have been using this GC to get CH4/CO2 seprated perfectly with TCD and FID. Now when I try to get N2/CO2 separated, the 2nd peak is really my problem.

I have also tried to ignore the 2nd peak and just take the first one as N2. Then I run pure He through to get the 'base' area. so when i have real sample injected, i will just substract this 'base' area. It seems to give me reasonable results. But I just want to sort out the 2nd peak issue.

Best regards,
George

Reconditioning the column should be of no help whatsoever with a split peak. The packing is spherical porous plastic beads (a dense styrofoam might be a closer description). Unless you have contaminated the beads with (to use a scientific term ) 'junk', there is little use in cooking out the column, it only accelerates the normal shrinkage of the beads which can produce voids and even more split or ever more increasing broader peaks.

IF, it is a column cause, then there is a void, a partial blockage, or an excessive dead space in the column or system.

But as usual, sample content, temperatures, amount injected and how it is injected all have bearing on the correct deduction of the problem.

Your question is like asking. What causes my car to 'stumble' when I drive it under 25 mph. There are a number of causes but without more information, all you will get is guess.

best wishes,

Rod

[chromatographer1]

A chromatogram would be so helpful.

But I think from your description of the problem (and I may be wrong) is that you are seeing an upset 'peak' close to your N2 peak. With helium sample with helium carrier you should be seeing NOTHING, no peaks at all, even with a TCD, UNLESS due to the change in pressure during the injection, a drop caused by the large? sample loop being at ambient pressure, you get a slight lowering of carrier gas flow for a few seconds, and this causes a 'peak' in the chromatogram. It often can occur about the time the fixed gases elute from a porous polymer column.

With a FID you never see this artifact, but with a TCD you will.

I hope that helps. Of course, you might have a leaking rotor in your sample valve, as well as an incomplete flushing out of the sample lines and sample loop, as another positibility of a cause.

best wishes,

Rod

[cheq]

Yes, you are right. there is an upset 'peak' always after the N2 peak. and there is alway a bIG peak at 0.3s which i assume it's due to valve switching from auto-injecting valve? (i dont know how to upload image here. but N2 normally ellute at 0.9s with CO2 at 2s.) so i think the 2nd peak may come from the valve switching as well?

And the column is acually is Hayesep Q not Porapak Q. But i think they are similar, as re-conditioning won't help?

A chromatogram would be so helpful.

But I think from your description of the problem (and I may be wrong) is that you are seeing an upset 'peak' close to your N2 peak. With helium sample with helium carrier you should be seeing NOTHING, no peaks at all, even with a TCD, UNLESS due to the change in pressure during the injection, a drop caused by the large? sample loop being at ambient pressure, you get a slight lowering of carrier gas flow for a few seconds, and this causes a 'peak' in the chromatogram. It often can occur about the time the fixed gases elute from a porous polymer column.

With a FID you never see this artifact, but with a TCD you will.

I hope that helps. Of course, you might have a leaking rotor in your sample valve, as well as an incomplete flushing out of the sample lines and sample loop, as another positibility of a cause.

best wishes,

Rod

[chromatographer1]

Under LC section this is the directions for inserting a link to a picture into your post.

The advice for Porapak Q applies also to Hayesep Q.

For anyone who wants to embed a graphic (such as a chromatogram) in a post, here's the procedure.

1. Convert your chromatogram to an image.

2. Resize it so that it's not too large (I would suggest no more than about 400 x 400 pixels) and save it in one of the following three formats: .jpg, .gif, or .png

3. Upload the image to a web server or hosting site which is accessible via the internet. There are lots of free image hosting services kicking around. One that I have tried is:

http://tinypic.com

4. Type the URL into your message, highlight it, and click on the "Img" button. (If you use tinypic, you can simply copy the "IMG" link from there after you have uploaded the image).

best wishes,

Rod

[cheq]

Thanks, Rod. Here is the chromatography



As you can see, there is an upset 'peak' always after the N2 peak. and there is always a bIG peak at 0.3s which i assume it's due to valve switching from auto-injecting valve? (i don't know how to upload image here. but N2 is normally eluted at 0.9s with CO2 at 2s.) I just have no idea what to do with the 2nd peak there...Also, as after a few hours, the 2nd peak will shrink and eventually disappear, which I don't have an explanation to either.

Regards,
George

Under LC section this is the directions for inserting a link to a picture into your post.

The advice for Porapak Q applies also to Hayesep Q.

For anyone who wants to embed a graphic (such as a chromatogram) in a post, here's the procedure.

1. Convert your chromatogram to an image.

2. Resize it so that it's not too large (I would suggest no more than about 400 x 400 pixels) and save it in one of the following three formats: .jpg, .gif, or .png

3. Upload the image to a web server or hosting site which is accessible via the internet. There are lots of free image hosting services kicking around. One that I have tried is:

http://tinypic.com

4. Type the URL into your message, highlight it, and click on the "Img" button. (If you use tinypic, you can simply copy the "IMG" link from there after you have uploaded the image).

best wishes,

Rod

[chromatographer1]

It looks like a channeling of the packing in the system. With enough pressure pulses the packing settles.

You can condition the column by slowly increasing oven temp by 1 degree per min and hold at 180C for overnight. Don't exceed the temperature limit of your rotor, and bring down the temp to normal before operating your sampling valve.

See if that helps.

Good luck,

Rod