Problems with Methane Analysis

[wbouma]

Hello all.

I am using an Agilent 7890 to analyze greenhouse gas (CH4, CO2, and N2O) flux from soil air samples. I am having problems with my methane analysis on the FID with a 'build up' of something over the course of my run (typically over 100 samples) where the baseline will gradually increase both in signal and noise and causing problems with the integration portion of the software (Chemstation).

Here is a link to some screenshots which illustrate the problem that I am having.

I am using packed columns ( PP-Q, 80/100) for this application. My system has got a TCD and FID in series. So my samples goes through the column into the TCD for CO2 then on to the FID for CH4 analysis. N2O detection is off the back inlet and runs in tandem but independently from the CO2/CH4 analysis. The parameters for this method are as follows:

Oven temp:85
Inlet temp: 200
Inlet pressure: 20 psi

TCD temp: 200
N2: 40

FID temp: 300
H2: 30
Air: 400
make up (N2): 1

In a nutshell...if I am running say 100 samples, the first 65-68 samples appear to be just fine…nice smooth baseline, nice peak with no integration issues (Fig A). Around about 70 injections (Fig B), the baseline starts to increase both in signal (say from 2 pA to 3.5 to eventually 7 or 8 pA) and in noise.

If I bake off and ramp the oven up to 130C at the end of the run…the signal will calm back down to the starting point. But the next run of 100 the same thing happens. This is very repeatable. And as you can see from the screenshots…this is a nightmare when it comes to integration!

I have tried several different temperature programs to add oven ramps after every sample and none of these things have helped. I have swapped out columns and tried a new one but it happens regardless. I have moisture traps on all of my gases. I have not tried a different type of column...is this something that might be worth exploring??

My suspicious is that it is moisture or something else coming in from the samples taken in the field. I have done a series of different runs...the first 100 standard CH4 samples. There was no change in the baseline. The other run was 100 laboratory air samples. Again, no change in the baseline. This again makes me think that it is coming in from the field samples and not something wrong with the FID. Whether it is moisture or not...I have yet to determine.

If anyone has any recommendations or experience with this issue I would appreciate some advice on troubleshooting/alleviating the problem.

Thanks again and please let me know if you need any additional info.

[chromatographer1]

You are getting something onto the column which takes a while to get through the PP column beads.

It is probably larger than 6 carbons in size, and may be 12 or more. Of course if you are using metal columns then it could also be a free acid or an amine. Lots of choices.

It is not water, as water would not cause a baseline rise, and since you are running at 85°C your water peak should be about 10 to 20 (?) times the retention of the methane.

I would run 50 samples, then bake the column the required time and temperature, and continue running samples.

Good luck,

Rod