Chromatography Forum

Dear all,

this is my problem. My initial FID signal is around 0,85 mV. For sample prep. I use methanol and after methanol is out of the detector my signal (and thus baseline) drops to around 0,70 which destroys my baseline and puts it bellow 0 mV. Chromatogram is normal but the problem is that its lowered for a significant amount so some peaks are "drowned" and I don't get correct results. Please check picture below and if someone has some experience with this please share it with me. It will help alot. Thanks in advance.

Important -- these are not same samples but just two chromatograms - good and bad:

good-->

bad -->

Some years ago, I found something similar... I observed a drop of baseline after the solvent peak.
In our case the problem was a broken ferrule in the flame tip.
I suggest you dismounting the jet from the FID and change the ferrule or simply check that the jet connection closes really tight.

This might also be connected to something that changes in the first minute or so, but that is not necessarily connected to the methanol - split vents opening and gas savers spring to mind. How does the baseline look if you do a dummy injection with no sample or solvent ?

As a work around while you sort out the real reason, try programming a detector autozero at 2 min or so.

Peter

Thank you for your replys. I "injected" blank sample and line is perfectly straight.

Does that mean that it is definitly the solvent or the reason could be elswhere?

suggestions?

Hi
The blank you did does suggest there is no intereference from the split vent opening/closing or the gas saver kicking in, make sure you zoom in on the chromatogram to double check this though.
It could be a flame stability issue and to that end I suggest you check for bits of ferrules and also check your gas flows are correct.
What carrier gas do you use out of interest?

I use Nitrogen as a carrier. Air and hydrogen for FID. Before this problem arose I used gas cylinders. After I switched to gas generator (for air and nitrogen) this problem started. Generator was installed by service crew.

First I thought obviously problem is the gas generator. After that I switched back to gas cylinders and the problem remained.

Do you think this could be contamination on gas tubing leading to gas chromatographer? or the problem is concentrated on FID?

Hi
No I dont think it is a tubing contamination problem, that would be pretty constant noise or spikes. What GC have you got?
Do you have trouble lighting the flame? Also check the hydrogen and air flows, usually hydrogen is 10% of Air. Do you use a make up gas?

Do you have gas scrubbers, especially for moisture, on the nitrogen just upstream of the GC ?

Peter

I don't use make up gas. GC is PerkinElmer Autosystem XL.

I use direct nitrogen from nitrogen cylinder (99,9995 %) which has no sort of gas scrubber. I just have on cylinder installed one gas pressure regulator.

It is a fact that this problem occurred after installing of PeakScientific ANG250 gas generator for air and nitrogen. But also after changing back to gas cylinders the problem remained.

I would suggest a thorough cleaning of the injector and detector.

Methanol is a wonderful cleansing agent for contaminated parts in a GC. Any oxidation in the detector is especially a target for the scrubbing nature of the solvent.

I would perform multiple injections of methanol and see if the baseline drop continues or 'levels' out.

I have seen detectors that were contaminated from silicon deposits (column bleed) lower their signal after a large methanol peak passed.

This may be the issue.

best wishes,

Rod