Chromatography Forum

After a week long normal operation of our GC FID analyzer, which gives positive peak with CH4 elution, we then started to get negative CH4 peak (dip) at the same retention time. Nothing has been touched, including elecronics and wiring. Anyone has a clue as to what might be the cause? Thanks.

Edward

Sounds like a software problem. Polarity has been reversed in your method for your data collection.

Are other hydrocarbon peaks also negative?

Are you using an older GC which has a dual FID and are using packed columns?

Methane burning in a FID will not give a negative signal.

And just because you did not change anything does not mean someone else was fooling around with your instrument.

best wishes,

Rod

Rod,

Thanks for your reply.

The GC under test has one FID with single packed column. Carrier gas is pure N2. The span gas is ppm CH4 in N2 or CO2. The negative peak was surely from CH4 elution, since it has the same retention time and it disappears if we use pure N2 as sample gas. About 2 month ago, we saw the same issue on this unit with other type of span gas, when elution peaks of all 2 compounds became negative. We never see the same kind of negative elution peak before with other units. This unit is a new one under development.

The strange thing is that, amplitude of the negative peak was changing sometimes if we tilted the unit. We thought some condensed water from FID exhausted vapor might cause this. However, this guess seems not true after we worked on FID associated parts.

Edward

Sounds like you have a contamination problem. If methane interferes with the current across the electrodes you would have to have a large signal to begin with. I doubt this is the situation.

I suspect you have a coelution of a non-burning analyte, perhaps water that is giving you a negative signal. There are many possibilities.

Your GC is contaminated. Find the source and get it cleared up.

best wishes,

Rod

Just wondering if this issue was ever resolved for eyan. I am dealing with the same issue....negative peak just before methane. We are using an older instrument with dual FID detectors, packed column, He as the carrier, and using sample loops for injection. From the previous string, it sounds like either a dirty detector, or possibly water vapor?
thanks,
Jane

Whilst not dismissing it - You tilted the GC as part of the trouble shooting procedure? !



Can you give more information on your sample/solvent?

Ralph

No, I didn't tilt the GC....that was someone else! But whatever works....

Our samples are headspace gasses from culture incubations, so they are either mostly air, with elevated concentrations of CO2 and CH4, or they may have been sparged with N2. We are actually running them through a TCD (for CO2) and then the FID (for CH4). They do pass through Mg perchlorate as a drying agent before injection, but we do see water peaks on the TCD output.

Jane

My reason for not dismissing it is that everything you do that causes a change can give a clue. (I know of a consumer product that failed its viscosity spec every full moon. The cause was that the plant was next to the coast and the artesian well water used increased its salinity with high tides )

I don't dismiss anything! I have worked with instruments for many years, and I know it sometimes takes unorthodox approaches to make things work. They can be very temperamental.

I some times see blip peaks when the 1/4" graphite/vespel ferrule on my capillary adapter for the 5890-FID has a leak. I am considering just swaging the adapter on with a steel swagelok ferrule.

Thanks; I'm not using a capillary column, but a leak is still a possibility. So far I have not found one. I think it is boiling down to a contaminated detector.

If the FID is in series with the TCD i.e. the effluent from the TCD passes through the FID then check whether the negative peaks happen when the vials have air in them, and the positive peaks with nitrogen.

Do you still get negative peaks on the FID if the filament in the TCD is cold ?

Peter

Yes, the FID is in series with the TCD, and the negative peak happens with samples that have air in them. I have not tried running them with the TCD cold....good idea. Thanks!

Yes, the FID is in series with the TCD, and the negative peak happens with samples that have air in them. I have not tried running them with the TCD cold....good idea. Thanks!

My suspicion is that the TCD filament is oxidising the methane, so you get a pulse of CO2 and water through the flame.

Peter