GC 5890 FID/Purge and Trap Baseline Drift Problem

[FIDOttawa]

I am having trouble with my GC 5890 Chromatographs which I am running with a Tekmar 3000 purge and Trap and a 16 port Autosampler. I seem to be having a baseline drift problem which is now causing me trouble as I can no longer get my blanks to be less than my MDL.

It seems to be a consistent drift, but it's a new problem I did not use to have and am having a really hard time finding the source. I have tried changing my liner, septum, column and trap. If anyone has any suggestions I am very open to hear them. During the drift it is on a constant temperature ramp so I don't think that would be making it drift up and down the way it is.

The baseline stays stable before and after the run. It is only during the run it appears to drift.

The peak at about 16 min is a surrogate but other than that this is a clean sample.

[Peter Apps]

If what you have a problem with is the small peaks then you are looking at contamination, not baseline drift (baseline drift goes up or down slowly and stays there.

Do you have scrubbers on your carrier gas, and when did you last change them ?

Peter

[FIDOttawa]

I don't think it's the small peaks. I just started a run on the GC alone not using the purge and trap system. Here is what I got:



This drifting is enough to cause me issues for integrating blanks. So I'm guess its something in my GC and not the concentrator.

As for the scrubbers, no I do not have one installed but those peaks at the end are actually not in the range I am looking at so I am not sure that would fix my issue.

[Peter Apps]

Your first step needs to be to get a gas scrubber on your carrier gas line. It may take a while for the internal plumbing of the instruments to clean up, so the sooner the better. Check also for leaks - you may be getting stationary phase bleed from oxygen or water diffusing into the carrier gas.

Peter

[Bigbear]

Detector issue? What temp do you run your detector? When I was using a 5890 fid for headspace ETOH I would some times get condensed water in the FID. All the water also "crudded" up the collecter. Try cleanning the FID.

[Yama001]

That baseline looks what you may get from a system that has had a *lot* of water purged through the whole thing. Perhaps a sample that foamed enough to desorb large amount of water and surfactants. It seems you have changed out much of the flow path, but I wonder if the was still a lot of crap help up in the inlet.

[FIDOttawa]

Well, I have taken the whole FID apart cleaned and replaced it. I have also changed my Gas Clean Filter. I trimmed both ends of the column and increased my detector temperature from 250 to 280 degrees. I also checked on leaks in the whole system and double checked my flow rates. I am going to run clean water through my autosampler all night and we will see what the results give me in the morning.

In the last at least 6 months, I have not had any bad/foaming samples so unless water condensation is getting through my trap to the GC inlet I don't think that's what has caused my problems. I also changed my inlet insert and septum just 2 weeks ago and since nothing pierces my septum it should all be pretty good still.

Still open to suggestions if anyone has any and I will post how things went over night, tomorrow.

Thanks

[FIDOttawa]

Still not doing any better at all. Going to try an old column back in and see if that changes anything. I am running out of things to change.

There is where I am now run on a clean port

[rb6banjo]

Is that an injection made by the P&T device or just you pushing the start button on the GC (bypassing the trap)?

[FIDOttawa]

Is that an injection made by the P&T device or just you pushing the start button on the GC (bypassing the trap)?

The last picture and the first one was an injection by the purge and trap of organic free water only (with Surrogate in the first picture). The middle picture is bypassing the P&T.

[Peter Apps]

You have ghost peaks - the wide hump at about 12 min, contaminants -the sharp peaks between 12 and 23 min, and column bleed - the spiky hump at the end.

Two things to check - check the septum purge flow with an independent flow meter at the purge outlet, and check for leaks everywhere in the system using a leak seeker.

First trouble-shoot the GC by not using the purge and trap.

Peter

[rb6banjo]

I'd run the column temperature up to the maximum in your method for 30 minutes or so. Then, I'd cool the oven down and do a run where you push the start button the front of the GC to start it. This bypasses the injection of the contents of the trap. If it's clean (slight column bleed at highest temperatures), then you know the contamination is coming from the P&T. Once you know the origin of the contamination, you can figure out how to get it out of your system.

[Peter Apps]

Don't take the column temperature up until you are sure that the system is leak tight. If you have a leak the oxygen and water that diffuses in from the air will degrade the column a lot faster than you can bake off the resulting bleed.

Peter

[FIDOttawa]

I baked out the column yesterday. I am not sure how to check my septum purge flow as I am not running a split system. My sample goes directly onto the column from the purge and trap. I know my trouble is in the GC alone. This is an image of a run from just pressing start on the GC alone.

I checked the whole system for leaks and have found nothing. I also know it is not the new column I installed because I put an old one back in and am getting the exact same response.

[Peter Apps]

Presumably the connection form P&T to column is via the GC inlet, because in your first post you wrote; "I have tried changing my liner, septum, .....". Most (all?) capillary GC inlets have a septum purge - check your operators manual for how to set it and where to measure the flow. If the septum purge is off or set too low you will see the symptoms that your system suffers from. NB that the septum purge and the split purge are two separate flows.

How did you check for leaks ?

Peter