Question about switching from helium to hydrogen carrier gas

[gmfan730]

Will switching from helium carrier to hydrogen carrier affect the relative response factors? I'm using FIDs and TCDs in the lab. I understand the effect it'll have on retention times, but I can't seem to find a clear answer in regard to response factors. The main thing I'm concerned about is whether or not I'll have to redo calibrations for a given method or if I can just keep the same response factors I was using with helium carrier. Thanks!

[chromatographer1]

The TCD response of an analyte depends upon the difference of its conductivity to that of the carrier gas. If He and H2 had the same conductivity then the response of an analyte would be the same. However, this is not true, although they are similar they are not identical. Yes, make new determinations of response factors.

If using H2 as a carrier gas for a GC with a FID then if the carrier change changes the flame composition then the response of an analyte will also change. Changing an inert gas with a fuel gas should make a difference, no? It may not be much, but you have to check their values and determine for yourself if response factors should be modified.

best wishes,

Rod

[gmfan730]

Thanks for the quick response! But wouldn't the analyte responses all change proportionally to each other? Or not necessarily? If they did change proportionally to each other (e.g. the peaks are all 50% larger with a change in carrier gas), the normalized mole fractions of each component wouldn't change, which means a recalibration would not be necessary. Thanks again for the input.

[Peter Apps]

This all depends on how accurate you need your results to be, what compounds you are using as standards and what you have as analytes, and mundane things like column diameter and flow rate - for example is you just switch a megabore column from helium to hydrogen you will see a dramatic change in FID response.

I am slightly surprised that you are not running some kind of calibration with each batch of samples.

Peter

[mckrause]

Depends on how you've calibrated. In an FID the relative ionization efficiencies will track - you will probably still hold relative molar ratios, if you are truly normalizing on mole fraction rather than area%. Some of it depends on your column - if you are using packed or megabore then you will definitely have to recalibrate.

TCD is a completely different beast - definitely recalibrate. There is no effective way to calculate thermal loss, since it is highly dependent upon the condition of the thermistor at the time of use. You should run routine calibrations if you want to get good data out of a TCD.

[GasMan]

You should be aware that some compounds in your sample may react with the hydrogen. I have seen cases of oxygen reacting with the hydrogen carrier gas when in contact with hot metal.

Gasman

[Peter Apps]

Depends on how you've calibrated. In an FID the relative ionization efficiencies will track - you will probably still hold relative molar ratios, if you are truly normalizing on mole fraction rather than area%to a good approimation an FID is a carbon counter, so peak area is proprtional to mass not to moles - one mole of C12 molecules will give double the response of one mole of C6 molecules. Some of it depends on your column - if you are using packed or megabore then you will definitely have to recalibrate.

TCD is a completely different beast - definitely recalibrate. There is no effective way to calculate thermal loss, since it is highly dependent upon the condition of the thermistor at the time of use. You should run routine calibrations if you want to get good data out of a TCD.

[gmfan730]

Thanks again for all the responses. In our lab, we analyze reactor effluents (on- and offline) to evaluate catalyst performance for organic syntheses. We have dozens and dozens of methods in place to accommodate all our chemisties and are continually developing new ones. We have measures in place to check calibrations from time to time and they actually stay remarkably consistent over time (for the most part). I guess the bottom line, based on your input, is that a whole re-working of the methods/calibrations will be necessary if we make the switch to H2. Might be a tad time consuming!

[Peter Apps]

How accurate do your analyses have to be ? Once you have the answer to that you can run a few representative compounds on both gasses to see if anything changes by more than your required accuracy. If not, which is likely for the FID, then just carry on as usual with hydrogen instead of helium.

Peter

[jdezeeuw]

Response factor based on sensitivity of detectors should not make a big deal with FID using any type of carrier gas.

Your biggest difference will be that the eluting peaks will be 2x narrower and therefore 2x higher. You may consider to inject 2x less and get even less system contamination. FID is pretty linear so I won't worry about response factors. normally with a calibration standard, this is also a non issue.

make sure that you take full advantage of using hydrogen as the carrier gas by also adjusting the program rates. I see many people forget that.
see some examples at: http://blog.restek.com/?p=3520

cheers
jaap