Question on GC calibration basics

[sokkabs]

Hi folks,

I'm a newbie to GC and working currently on H2 gas detection. I've developed a method for calibrating standard H2 gas (1% and 3%, balance N2, vol%) and got repeatable area for duplicate injections (10 ul).

My question is, how would I report this vol% of H2 in mol or umol (micromol). I need this to generate a calibration plot and compare this results with published data.

My starting point of course is using PV = nRT relationship.

Assumption:

P = 1 atm (since injection at atmospheric pressure)
V = 10 ul (is this correct? or 1% of 10 ul)
n = need to calculate
R = 0.08205 L.atm/mol.K
T = 298 K

I'm ended up with 0.4 micromoles. I'm not sure this number is right?

I really appreciate your time and help for the clarification.

[Alexandre]

If you need to compare with specification use the same units of concentration as in it.
Keep good control of all you P,T&V parameters for calibr. standards and samples, they better be the same.
Use good certified std, evaluate bias, use uncertainties for accurate comparison.
And don’t forget the basic metrology rule – method should fit the purposes. There is no point to create expensive high accuracy and precision method if your client needs only rough idea.

------------------------
"If your experiment needs statistics, you ought to have done a better experiment." Rutherford

[Peter Apps]

If I understand correctly what it is that you need to do then this is very straightforward; as long as you have a close enough approximation to ideal gas behaviour (which you do) then the volume fraction is the same as the mole fraction. In round numbers 1 mole of gas occupies 22.4 l, you have 1% of 10 ul of hydrogen, 0.1 ul as a fraction of 22.4 l is the moles of hydrogen.

Peter

[Johnny Rod]

I think Peter's right, volume and mole fraction are pretty much the same. Also I'm not sure why you want to work out the number of moles in the injection? You probably want the concentration in the sample gas?

[sokkabs]

Thanks folks for your time and input.

@ Johnny: Yes, I would like to know the moles of H2 produced (from my experiment) based on the calibration plot I make with 1 to 3% H2 standards

@ Peter: Based on your suggestion, the no. of moles turns out to be 4 nanomoles in the injection. I guess this is very low number for 1% H2 (may be I'm wrong)



I'm attaching a figure from a published data where it shows both H2% as well as micromol. The procedure for these measurements are explained as below (from the paper)

"The amount of photogenerated H2 was detected and quantified by headspace gas analysis with an Agilent 7890A Series gas chromatograph (GC) equipped with a 5 Å molecular sieve column (N2 carrier gas at a flow rate of 3.5 mL min−1) held isothermally at 40 °C, and a thermal conductivity detector. The instrument was calibrated with various known amounts of H2 in a mixture of CH4 (2%) in N2, and the linearity and stability of the instrument and method were checked regularly. The total irradiation time for each light experiment was 4 h, with 10 μL samples of the headspace gas removed twice for GC analysis after 0.5, 1, 1.5, 2, 3, and 4 h."

I'm following exactly these conditions except my standard H2 gas doesn't have 2% methane.

Am I completely missing something in the calculations?

Thanks for your time again.

[sokkabs]

@ Alexandre: Thanks for the info. I'll maintain the conditions carefully.

[Johnny Rod]

If you want the moles of hydrogen from your experiment then the moles in the GC don't mean anything. You want the concentration (%mol) then multiply by the flow rate through your experiment (either flow rate or total flow depending on how you're doing things).

The GC method is fine, make sure the run is long enough to see the methane come off before you do your next injection. What diameter GC column is it?