how to determine the response factor from TCD

[manson]

We have applied response factors (Dietz, 1967) to determine the mol% of our components for quantitative results.
but we need to know the response factor of H2. is this formula work?
H2(TR)/H2(area)=O2(TR)/O2(area)
if not, what's a appropriate way to measure that?


thanks for your answer ! it confuse me a lot!!

[Foobs]

I've run standards on my TCD and have found that sometimes the response factors are different from Dietz's. On carbon monoxide and nitrogen, it says they should be the same and they are, but on other components, Dietz will say something is 20% lower than N2 and our standards give a different variation. Also, our experience is that the response factor does vary with amount somewhat.

I don't know what TR means, so I don't know what to say to your equation. If you're using helium as a carrier gas, our experience is that the hydrogen response factor / nitrogen response factor is about .013, but I'd really recommend running your own standards.

[manson]

Foobs,

sorry TR is thermal response from Dietz
i don't understand how you estimate your hydrogen response factor and nitrogen's.
could you explain more ?

my products is H2 ,CO,CO2,O2 and methanol and water.
i would like to know how to quantify them from GC equipped with TCD.
we use the thermal response from Dietz and calculate the mol% of our products. but we try to calculate the H2 thermal response .

[Foobs]

Ah... I'm used to using RF (response factor = area / amount)

We calculate our response factors by injecting knowns onto the GC. A pure gas is nice in that you can be awfully certain that the given composition is right , but may be misleading if your actual compositions are well below that.

The rigorous process would be to calculate moles of each component injected in your standard (PV = nRT, n(i) = n * Ci). Area / n(i) = RF. In our work, we try to get a relationship between Area and RF (ideally, RF would be a constant for an individual component), and then apply that to the unknown samples.

If you use an internal standard in your work, you might want to norm all of your response factors to that component. We use nitrogen for that, so we often do relative response factors.

We use the RF by taking the area in the unknown sample, using the RF to get moles, and then either the ideal gas law or the total number of moles analyzed to get concentration. Actually, we use an injection valve, so we just go directly between concentration and area, but that's another story

I hope that helps