Recovery reported from GC-TCD

[NRCan6]

Hi all,

I am wondering whether someone could give an answer from a theoretical perspective. The following paragraph is from a confidential source.

-For natural gas GC-TCD analysis

"The Thermal Conductivity Detector (TCD) information can be used to calculate the gas recovery of the sample injected into the column in mole%.
The partial pressure of the injected sample into the GC loop determines the recovery through the column. Total recovery should be 100 ± 3%."

My questions are: Why the recovery is 100± 3%, not 100%? Is it real that partial pressure determines the recovery through the column?

Thank you very much!

[chromatographer1]

That statement is a untrue sorta-true statement.

Nat. Gas samples tend to approximate the behavior of ideal gases. If each component had the same response to a TCD (they don't) and if the components acted ideally (they don't) then the partial pressure % of each component could be expressed in mole %.

Do you know why they are called the IDEAL GAS LAWS ?

If gases acted ideally then exact numbers could be generated from their behavior. But guess what? In the REAL world (not some fantasy world) gases don't cooperate and behave ideally.

Now you can find samples where they ALMOST act ideally (helium and hydrogen balances, for example) but nat. gas analysis is usually a money value determination based on BTU content. Ideal for demonstrating that over a range, natural gas DOES NOT ACT ideally. That is why calibration standards are made and corrections are imposed for non-ideal behavior.

Remember for samples injected at atmospheric pressure...... that can vary in absolute pressure. That means the change in weather means a change in molar sample size. Also, with comparing labs, the difference in altitude also changes the sample size in moles, which of course affects the measurement.

See why calibration stds are mandatory? And don't forget, water vapor is not usually measured directly in natural gas analysis. Do you have an idea how much water vapor can vary in natural gas samples?

best wishes,

Rod