Atmospheric pressure compensation for gas analysis

[Johnny Rod]

Does anyone use compensation during a run for changes in atmospheric pressure? If so how do you measure it? The ideal would be something electronic that could be incorporated into the calculations, think that might be a bit of a pipe dream though!

[antonk]

IMO every modern GC measures atmospheric pressure to maintain precise difference between inlet and outlet (controlling flow or pressure as you have selected in method). Some models even show you room temperature and atmospheric pressure.
As far i remember for gas analysis reproducibility problem it that you fill sampling loop with gas under different pressure. So the task is make loop-filling controllable and reproducible. Simplest way - set with manual pressure controller desirable pressure (some % above ambient) and use hydro-lock (flask with water or another liquid) on spare end of sampling channel. Heat sample loop above ambient. Flush sample loop within longer time period.

[GasMan]

What are the GC's that you are using?

Gasman

[Johnny Rod]

Agilent 7820 and Varian 430, both with 2m x 1/8" mol sieve 5A and TCD detection

I did notice that the 430 reports a figure for atmospheric pressure but it's not mentioned in the manual anywhere, I'll have to look at the Agilent, as this is news to me, thanks Anton.

We generally use either a fixed gas flow through the loop or inject at zero flow, the latter technique requires more skill and a cold (ambient) loop but yields tighter precision (this is required for EP gas assays), however is also prone to atmospheric pressure fluctuations affecting peak areas.

[GasMan]

I am not that familiar with the 7820. The 7890 does allow you to read the atmospheric pressure, but this is done via the keyboard which I believe is not present on the 7820. How are you calculating your results? If you are measuring percent, does it really matter if the atmospheric pressure changes. I only see that you need the pressure if you need the amount of the gas as an absolute value.

Gasman

[Johnny Rod]

There is a software keyboard suppleid by Agilent for the 7820, looks like a 6890/7890 keyboard. Lords knows why, but it's there and it works well, will have to have a root around in it.

We have to do an external standard assay rather than area% so yes the variations in area counts do matter. The target precision is 0.10% RSD from three injections, then at the end a recovery injection/back standard which has to give no more than 0.15%RSD when combined with the calibration injections (it's actually mroe complex than this, believe it or not). Days when the weather is changing can banjax a run, but if we could correct for this then maybe we could get better reliability, and longer runs. Now if the RSDs fail you just have to start again, thankfully most things can wait till tomorrow if the air pressure is wildly on the move, but it's not like any of us has time to waste doing duff runs after all.

[GasMan]

On the 7890, to see atmospheric pressure, you would proceedas follows:

Click on the blue Service Mode key. Scroll down to Diagnostics\Electronics\Pneumatics. You will then see the Atmospheric pressure. I do not know if this is available on the 7820. Are you using ChemStation?

Gasman

[Johnny Rod]

It's the same on the 7820. We're using Openlab Chemstation ed. standalone workstation, and Galaxie on the Varian though I have a Compass CD that is waiting for me to get my thumb out.

[GasMan]

It is possible to have a unit installed on the 7820 known as a Pneumatics Control Module (PCM). This module has two channels. One channel can operate either as a flow controller or a forward pressure controller. The second channel can work either as a forward pressure regulator or a back pressure regulator. If you connect the second channel to the sample OUTLET line of your gas sampling valve and configure this as a backpressure controller, you can then set the pressure to be 5 psi, for example, in the sample loop. This would then be independent of atmospheric pressure.

Gasman

[Johnny Rod]

Hi Gasman, that's brilliant. Would I need say a tee off to allow purging of the lines/loop then shut this off to control pressure before injection? Or if it purges straight throughthe loop into the PCM will that dump the excess pressure by itself?

[GasMan]

A back pressure controller will allow the pressure to build up in the loop until it reaches the set point. The controlling valve will then open and allow the gas to flow out to vent. Two questions.

1. What pressure is the gas at that wish to analyze. Can you regulate the pressure?

2. Does your gas contain any corrosive gases.

Gasman

[Johnny Rod]

The gases are almost always in a cylinder at least 10bar, but we can step down either with a regulator or just by metering the flow. Nothing corrosive, maybe some flammable stuff, mostly air components.

[GasMan]

One tip to get this working correctly. You MUST have a restriction between the regulator on your calibration/sample cylinder and the sample inlet line to the GSV. I am now running with a 2 meter length of 250 micron tubing. I need 5 psi in the sample loop and with 10 psi on the regulator and the system is working fine. You can use the Agilent flow calculator to get the flow that you need.

Gasman