how to obtain lower signal for FID

[cella]

Dear All,

Anybody can tell how to get a lower signal for my GC/dualFID. The signal I am getting is @ 50 for both detectors. I think I am supposed to get 20 or lower. This is a refurbished GC we acquired some months ago. I am used to ECD and MS detectors but not with FIDs.

I will really appreciate any suggestions from you..

Thanks,

cella

[Consumer Products Guy]

Inject less sample?

Increase split ratio?

More information please. But only if you really want help. This relates to most posters here..

[cella]

Dear Consumer Product Guide,

I've checked our method, the split ratio is at 1:1. I am injecting 2ul of sample with a split flow of 10ml/min. I am using a Rtx-5, 15m x 0.32mmID column. My detector temp is at 325C and my oven final temp is 310C. I was advised by our service engineer to extinguished the flame during bake-out and standby operations. I only ignite the flame before I start the analysis. Am I doing it wrong?

Thanks for the reply.

cella

[Consumer Products Guy]

Dear Consumer Product Guide,

I've checked our method, the split ratio is at 1:1. I am injecting 2ul of sample with a split flow of 10ml/min. I am using a Rtx-5, 15m x 0.32mmID column. My detector temp is at 325C and my oven final temp is 310C. I was advised by our service engineer to extinguished the flame during bake-out and standby operations. I only ignite the flame before I start the analysis. Am I doing it wrong?

Thanks for the reply.

cella

I've not heard of a 1:1 split. Typically with a column of that dimension we'd have a split flow of maybe 30 ml/min, maybe a 25:1 split. Personally, I'd leave my FID lit during "bake-out" situations. Why do you think you signal is too high, what do your peak shapes look like, and how is your peak area reproducibility?

[GasMan]

Cella,

It might be a good idea to give the manufactuer and model number, as there are some times small tips that are unique to each manufactuer.

I am assuming the the '50' is the signal in pA that you have on your FID's as baseline signal. This is a measure of how clean your 'GC system' is, note the word GC system. You are looking at the total system, and if you wish to bring the number down, you will need to isolate where the signal is coming from. However, this number may not be too bad, and depends on your samples. If you are doing trace analysis, then it could be an issue as it will affect the signal to noise ratio, but if you are in percent levels, then bringing the number down below 50 will probably not make an effect on your results.

If you wish to bring it down, I would proceed as follows.

First disconnect the column from the FID and plug the inlet of the detector. You can use a ferrule without a hole or use a paperclip with a new clean ferrule. If the signal does not come down, then this would show that the detector has a problem. Things that can cause a problem are as follows.

1. Hydrogen, air or makeup gases could contain impurities. Install hydrocarbon traps in these lines.

2. Jet is contaminated. Clean or replace.

3. The lines bringing in the hydrogen, air and makeup gases could be contaminated by a previous user. If you heat up these lines with a hot air gun or hair dryer and see an increase in signal, this is a sign that they are contaminated. You may be able to remove the contamination by prolonged heating, but the best way would be to replace with new lines.

4. If you turn the flame off, and still have a high signal, then you either have an electrometer problem or dirty insulation giving a leakage path to ground, but this is dependent on manufacturer.

If the signal does come down when you disconnect the column, then this would mean that you could have a problem with the carrier gas, column or inlet.

Gasman

[Peter Apps]

Hi Cella

In addition to the suggestions above, check the actual flow of each gas to the FID - do not rely on the settings on the GC, measure them at the detector with a flow meter. I am assuming that the "50" is the baseline, not actually the signal but I need to know the units, probably pA, uV or mV. Depending on what hardware you have you might be able to auto-zero electronically, but this is poor practise if a high background is due to a problem with the system.

Peter

[cella]

Hi Guys,

Thank you for all your replies.

My GC is an HP5890series II. I did left the flame lit during bake-out and adjusted my split ratio, 25:1 but the signal is still the same (by the way, i believe it's in pA)

Gasman,

We analyze DRO at ppm levels.
You may be right about the detector jet being dirty or the gas lines. I have new gas filters/purifiers, however i used old copper tubes for my gas lines. And i don't think the jets are new either, though I still need to confirm that from our vendor. Is it normal to have black, soot-like materials on the column ends inserted on the jet? I tried wiping it with lint-free tissue and the material seemed to rub-off. It appears to be on the outside only.

I also disconnected the columns, plugged the detectors then re-lit the flame, the signal is still high. The signal drops to zero whenever the flame is out, which is good right?

I guess i need to replace/clean my jet; i might as well replace my gas lines. unfortunately, i will have to wait until our supplies arrived from the US (i am in the philippines, and we buy from US suppliers). That will be within two weeks.

Peter,

I did checked the actual flow of my gasses before I started, and I still do from time to time. My carrier gas is @ 10; make-up @ 25; H2 @ 25; Air @ 300ml/min. I don't know about zeroing the signal electronically, but i do column compensation.

Thanks a lot guys. I will clean my jet for the meantime then see what happens. I might just continue analysis with this condition.

cella

[Peter Apps]

From memory the flows on HP/Agilent FIDs are 40, 40 and 400 ml/min for hydrogen, make up and air respectively but I doubt that the flows you have will cause a high baseline. Someone who is currently using a 5890 can probably correct me.

Is the baseline stable ? If it is you will probably find that you can use the instrument exactly as it is becuase whatever signal you get from the analytes will just add to the background. A high and unstable background is a more serious problem.

It is years since I used a 5890, but I'm sure that somewhere on the front panel is a zero button. If not you can perhaps zero the detector from the software if you are running software control.

What is "column compensation" ?

Soot inside the jet sounds bad, and you probably have dirt on the collector electrode as well (the one above and around the flame) so while you are cleaning the jet it will be a good to give that a clean also.

Peter

[Spuzzin]

Peter,

Column compensation is an automatic way of subtracting column bleed. GC does a run without an injection and saves the data which then automatically gets subtracted from every subsequent run. I've played around with it but generally found it unreliable.

You are correct about the FID flows.


Cella,

I'd be inclined to take your FID apart and give it a good clean. Definately replace the jet if it is sooty.

Rich

[GasMan]

Cella,

The fact that you get a black sooty material on the outside of the column end that goes into the FID, is not a good sign. It suggests that that the inside of the FID, the part below where the jet seals, could be very dirty and contaminated. I would agree with the posting above, that you should disassemble the FID and give it a good cleaning.

The fact that the signal goes down to 0 when the flame is turned off, is good sign.

Be careful using the zero function on the 5890. If you use it now with the baseline at 50, it will subtract the 50pA from the signal. If you clean your detector and lets say that the signal comes down to 10pA, you will not see peaks below 50pA unless you re-zero your signal.

Gasman

[AICMM]

cella,

You probably don't need to replace your jet unless you are having trouble getting the detector lit. You certainly need to clean the detector as everyone has noted. One of the troubling things about the 5890 FID (and probably the later versions as well) is heat management which, in my opinion, is very poor in HP 5890's. If you are using DCM as a solvent you are creating HCL and the top of the FID tower is often not hot enough to prevent water from condensing and then HCL from condensing. So, clean your FID and then raise the temperature another 20 or 30 C minimum. In the past I have installed auxiliary heaters on the 5890 FID just to maintain the upper portion of the detector at a temperature high enough to prevent condensation.

Best regards.

[CE Instruments]

The way I read the responses I would just like the original poster to confirm that the column has not been pushed too far into the detector. If the column goes through the jet into the flame, you get a high background current and the tip of the column burns leaving a black deposit. Just my 2c

[thohry]

Per my experience, if you use a zeroair, the signal level of an FID reduces rapidly. About 10 or 15.

[aldehyde]

With high quality gases, traps, and a clean FID it is very possible to get a signal between 1.0-5.0.

I see discoloration on columns that have been in a hot FID all the time its not necessarily a sign of contamination.

Replace the jet and the tubing going from your gas to the FID (I would put money on the tubing being the problem).