GC-ECD baseline not stable

[Irina]

Hi there,
I have a problem with my ECD unit (GC-ECD -2010 from Shimadzu). the baseline is not stable. I did changed the column and let it conditioning for a day. But after that the baseline is still shifting at about 1000uV/min. When I tried to inject a hexane blank it will rise up with temperature. The column is ZB-1701 and the temperature was set at 260 C degrees, the ECD at 300.
I did search for old topics but didn't find one. If I missed it please redirect my post.

Thank you!

Irina

[syringe]

Hi Irina

Try to divide the problem. Cool down the detector and remove the column from the detector side and put a blank nut (I am not familiar with Shimadzu connections but surely you can do that).
Then bring the GC to your working parameters. In that way you will test only the detector. Sometimes leaks at connections or at makeup entrance are responsible of that response.
If you get a stable signal, do research the inlet, flow supply, etc.
Regards,
syringe

[Irina]

Hi Irina

Try to divide the problem. Cool down the detector and remove the column from the detector side and put a blank nut (I am not familiar with Shimadzu connections but surely you can do that).
Then bring the GC to your working parameters. In that way you will test only the detector. Sometimes leaks at connections or at makeup entrance are responsible of that response.
If you get a stable signal, do research the inlet, flow supply, etc.
Regards,
syringe

When I put the blank nut it will gave me a straight line. I changed the septum and the glass insert, but when I started the GC it will gave me a lot of peaks and from time to time (not an exact amount of minutes - it varies) it will jump very high and then get back. I do not know how to attach a chromatogram to show how it looks

[syringe]

I assume you got chromatographic peaks (not electrical noise).
With the detector base capped, if you get peaks when running the oven program is evidence of contamination at detector base. Rise the oven and the detector temperatures higher than normal program to clean those spots. Monitor the signal. When you see it passing over a max and turning to lower values, cool down the oven and detector and wait to stabilization.
Then test the whole system.

There are other possible reasons to your problem: an air leak, a damaged tubing having a fissure, contamination in makeup line that release the contamination when heated.

[Irina]

They seemed to be chromatographic peaks, but from time to time it will jump higher than the chromatographic peaks stay for 1-2 seconds and drop down fast at the initial value. it looks like a reversed U. I will try again tomorrow to check for damaged tubing. If it'a an air leak is somewhere from the line because I closed the air cylinder and seems no difference.

Thank you for the reply

Regards,
Irina

[syringe]

Hi
What is the air for? The ECD doesn´t run with air...
ECD uses only Nitrogen. An option is Argon/Methane but not too common.
Tell about the general condition of the GC. How old is it?
Air leaks from the atmosphere introduce oxygen into the detector gas. The ECD is strongly affected by oxygen, showing a peak if it is temporarily present or rising highly the baseline in case is more or less constant in its presence.

[Irina]

Hi
I also have an FID on the same GC. It's a GC-2010 from shimadzu 7 years old and the ECD was changed in july 2013 and it wasn't used to much. I am analysing organochlorine pesticides on it and last year I did not had too many samples.

Regards,
Irina

[syringe]

What about the inverted U response?
Did you try to clean the ECD base with thermal cycles?
The response goes it down or remain at same level after several cycles?

[Irina]

Hi,

I did not solve the problem with my ECD, but I did had some work to do on GC-MS and I've ignored it.

Syringe, you asked me if it looks like chromatographic peaks or electrical noise. I think the signal looks more like electrical noise (I'm not sure how it should look). I am very stressed out because I have to make it run in like a week.

Should I isolate the ECD with a blank ferrule and bake it at maximum temperature or what else can I do?

Regards,
Irina

[whdees]

try baking out the ECD with methane running through it, and make sure the column insertion distance is correct - too far, and it can cause an erratic baseline. Also, if you have a different column, you can try that, you may have gotten a bad column (I doubt it, but it is worth eliminating as a possibility)

EDIT; I meant to say Hydrogen, not methane...

[Irina]

try baking out the ECD with methane running through it, and make sure the column insertion distance is correct - too far, and it can cause an erratic baseline. Also, if you have a different column, you can try that, you may have gotten a bad column (I doubt it, but it is worth eliminating as a possibility)

EDIT; I meant to say Hydrogen, not methane...

instead of helium? I'm not sure if I can switch from helium, but I will try and then I will let you know what the result is

thank you

[whdees]

try baking out the ECD with methane running through it, and make sure the column insertion distance is correct - too far, and it can cause an erratic baseline. Also, if you have a different column, you can try that, you may have gotten a bad column (I doubt it, but it is worth eliminating as a possibility)

EDIT; I meant to say Hydrogen, not methane...

instead of helium? I'm not sure if I can switch from helium, but I will try and then I will let you know what the result is

thank you

run the hydrogen in as your makeup gas not your carrier gas, elevate the ECD temp by about 50 C higher than normal, and let it cook out over nite. *IF* the ECD is contaminated, that should clean it out. I'm also wondering about the column... can you put the old column back in and see if the problem goes away?

[andres escobar]

Sorry for my intrusion but can you tell me more about the bakeout of the micro-ECD with Hydrogen as make-up gas instead of Ar-Me 5.
I am having lot of troubles in the stabilization of the signal-noise of my micro-ECD, and already tried the bakeout two times, with 400 C for around 20 hours, and the signal is still in 3800 Hz.
I am suspecting a Make-up gas contamination, but i can not compare with a new cylinder because it has not arrived to the lab.
In the meanwhile i would like to try with the bake out with hydrogen.

Constant makeup of 60 ml/min for hydrogen all night????? oven in 40° C ??
thx