FID vs. TCD - part 2

[HPLC_Help]

Hi, all -

I posted questions last month regarding issues we were seeing with our TCD vs. FID. At the time it was plumbing issues, and we got them straightened out.

Now I have a more theoretical question (or 2 or 3 ...)

1) Can H2 be measured on an FID detector?

2) Also - Someone told me (a non-Analytical team member) that - in off-gas samples from a fermentation process - we could "measure the volatiles and report as methane". I don't know what this means. Wouldn't - ideally - all the volatiles elute separately, and thus be quantified separately? Any idea what he means? I plan to ask the guy more about what he means, but since I'm such a GC novice, I wanted to get a little background first. (I hope).

3) If the answer to (2) is "yes, it can be done," can it be done on a TCD? (I assume the detection limit for TCD is not nearly as low as that of an FID).

Thanks, all, as usual!

H_H

[chromatographer1]

1 - NO

2 - yes, if you give the response factor for all of the unknowns to be the same as methane; Which is not correct but could be used as an approximation for trending studies. Propane has a different response to a FID than methane for example (or a TCD as well).


Rod

[chromatographer1]

oh, and by the way,

If you use a pHID detector it will be much closer to an equal molar response for all analytes, and you can measure H2 at the same time.

See AIC or Valco or Gow-Mac for their detectors and pricing.

Rod

[thohry]

For FID detection, all the alkanes have the same response according to the number of carbon atoms in the molecule. For example, methane response is about 1/3 of that of propane if they are in the same mole concentration.

[Peter Apps]

The equal response carbon counter behaviour of the FID is one of those myths that grows up because everyone wants it to be true, then everyone assumes that it is true and proceeds accordingly, and the errors of a few percent that arise are put down to inlet discrimination or whatever.

I did an extensive literature review of FID response factors as the first stage of a feasibility for its use in metrological primary methods - methods that do not need a standard of the target analyte for calibration. If the FID was a carbon counter it would be possible to calibrate an analysis for e.g. octane by calibrating with e.g. decane (or even better with PAHs which can be prepared at very high purities by zone refining). Sadly for that liitle plan the differences in response factor, even for different members of homologous series, were too high for such an application.

Peter

[AICMM]

HPLC_Help,

3) No. TCD has significantly different response factors (more so than the FID) so measuring total hydrocarbons this way would not be appropriate.

Best regards,

AICMM

[HPLC_Help]

Thank you everyone for your help. I appreciate it.

Regarding methane on an FID, and assuming equal response factors (or anyway, equal enough to look at trends) ... I would still have to know which peaks are -anes, and which peaks aren't, correct? Wouldn't I have to inject some known analytes onto the system, so that I know which ones to measure (however approximate it may be) with the methane?

H_H

[Peter Apps]

"Any idea what he means?"

Before we start speculating, why not go and ask the one person who should really know - the "non-Analytical team member". I have a sneaking suspicion that he/she might not be very clear about it themsleves.

Peter

[HPLC_Help]

Thanks, Peter. He's returning from vacation this week, and I do plan on asking. I just wanted a little more background information for myself before discussing it with him.

Thanks for all your help, everyone!

H_H