Carrier gas

[kartoffel]

Hi everybody,

i would like to know if anybody of you uses H2 or N2 as an alternative mobile phase in GC/MS? Are there any serious problems using these gases instead of helium? And i dont mean the possible safety problems with H2, i mean problems in separation or in lower analyt responses.

I've never seen a lab that uses N2 as carrier gas, why is that? Are there any problems concerning MS? I've just read that responses are lower about a factor of 25 with N2 (and about a factor of 2 with H2). I could imagine that responses are lower for H2 because the higher gas velocity causes a lower vacuum and so the ions collide more often in the MS, but i don't see a reason for the lower responses with N2. Any suggestions?

Thanks

[DR]

Consult the VanDeempter plots for these various carriers...
http://www.restek.com/Landing-Pages/Content/gen_B008
sums it up nicely.

N2 - narrow range of high efficiency at low flow rates, H2 is most efficient at higher flow rates. He is best at flow rates a bit lower than H2's best efficiency.

Higher optimal linear flow rate = faster analysis, other things being held equal.

IE - using N2 means much longer run times, H2 is fastest.

[kartoffel]

Thank you for your answer.
I see that the Van Deemter function of N2 isn't optimal for GC applications, whereas H2 should be very good. Although not many labs seem to use it. Is there some kind of old fashion mentality like "we always used helium and we won't change that" or why do Analysts doesn't move to H2 as carrier gas for GC-MS, considering that it is way cheaper than helium.

[Steve Reimer]

both Agilent and Thermo have presentations / webinars available that show the issues associated with hydrogen as a carrier gas in GC/MS.
While H2 is great for the GC part it isn't so good for the MS part. Not only is the ionization energy for H2 lower but it is harder to pump out and more reactive. This means that the MS has to be inert, the flow rate has to be lower (even with a turbo pump) and still sensitivity will suffer.
It can be done but it isn't as easy as you would like. The next versions of EPA SW-846 Method 8260 and 8270 have hydrogen carrier gas listed with some conditions.

[Forse]

Is there some kind of old fashion mentality like "we always used helium and we won't change that"

Yes. At least from what I have experienced.

Changing over to hydrogen will take some man hours if you analyse a lot of different things. My current lab could save 2-3 thousand USD a year if we could change to hydrogen, but if this means I will use a few weeks experimenting with method development and what not, all savings for the next five years will be gone.

[James_Ball]

I have been working a couple years on trying to transition to H2 for our GC/MS applications. If you are trying to do EPA work you will have problems passing tune parameters with H2 carrier. BFB for volatiles I have gotten very close to passing but you must drop your flow rate down to near 0.2-0.3 ml/min to get it to pass because of the reaction that causes the m/z 96/95 ratio to be too high. H2 also pumps out better if you have an oil diffusion pump instead of a turbo pump for high vacuum.

For semivolatiles I have gotten DFTPP to pass pretty well, but it does require several days of baking out the system with a high flow of H2 before it will work. You get what looks like a hydrocarbon pattern when scanning in manual tune and it comes from the H2 cleaning out any and all residual oils from the manufacture of the tubing, valves, columns and any where else it can come from. Even a brand new system will have this. I was able to tune our Agilent 7000 QQQ to pass 525 and 8270 tune criteria on H2 but the requirement that m/z 68 and 70 be less than 2% of 69 is the difficult part, as those ions seem to always be present in the background once you switch to H2.

With H2 your vacuum reading will be higher, but some of this comes from the fact that the ion gauge tube is calibrated normally to N2 and for the same real pressure it will give different readings for He, H2 and N2. You can't really take advantage of the high flow capability of H2 in a mass spec system, but it still gives faster runs at the same ml/min as He since the linear velocity is higher for H2 than He at the same volumetric flow rate.

Nitrogen has always been shunned for capillary GC because it needs a low volumetric flow rate, but I have read new papers that show that nitrogen can work well in this application. One of our guys here is running N2 for our pesticide analysis and it works great. But I tried it with GC/MS and it really kills the sensitivity. If you think about it, the N2 is probably going to be soaking up a lot of the energy from the filaments that H2 and He would not since it is such a large molecule compared to those. I got great chromatography and separation using N2, but at least 10x lower sensitivity.

If you are not held to EPA tuning restrictions then I would definitely say go with H2 as a carrier.

For other research, any time I have a day or two that my 7000QQQ is free, I have been playing around with blended gasses to see if I can make the tunes pass. With two columns installed and connected just before entering the MS I can flow H2 in one and He in the other and experiment. Sometimes I can get tunes to pass with He carrier and a small H2 flow and this seems to improve linearity quite a bit. I hope to eventually get to run straight H2 on everything but that is still a ways off. I hope EPA will catch up soon and alter the tune requirements enough that it can be used.

[James_Ball]

both Agilent and Thermo have presentations / webinars available that show the issues associated with hydrogen as a carrier gas in GC/MS.
While H2 is great for the GC part it isn't so good for the MS part. Not only is the ionization energy for H2 lower but it is harder to pump out and more reactive. This means that the MS has to be inert, the flow rate has to be lower (even with a turbo pump) and still sensitivity will suffer.
It can be done but it isn't as easy as you would like. The next versions of EPA SW-846 Method 8260 and 8270 have hydrogen carrier gas listed with some conditions.

I was just reviewing 8260B and 8270D and noticed a caveat in each that is interesting and could possibly help with using H2 as carrier. 8260B section 7.3.1.2 and 8270D section 11.3.1.2 mention using either the criteria listed in Table 3 for tuning specs or specs that are published by the manufacturer. If the major instrument manufacturers could reach an agreement or at least individually come up with tuning criteria that accommodate the problems associated with using H2 carrier, would that become acceptable in the regulated setting? If m/z ratio for 96/95 in Volatiles BFB tuning was changed from 5-9% to something like 5-20% then passing for 8260B would be fairly easy, just as increasing m/z 68 and 70 vs 69 from 0-2% up to 0-10% would make 8270D work well too.

Seems like a simple change to us, but for a government agency this is equivalent to moving Mt Everest to South Africa using a spoon and teacup. The only aspect of the methods that would be effected at all would be Tentatively Identified Compounds when searching unknowns against NIST spectra, but I may do that for one sample per year, the rest are all identified versus the mass ratios generated by my calibration standards.

[kartoffel]

We're not bound to these EPA tune restrictions, we get some material samples (different polymers) this year and have to investigate PAH in it.
I read about this baking out process when switching to H2, we will also do this before messuring any samples.

[dr_bahram1977]

James has replied in detail. Just wanna say that N2 is not for MS at all. you can't check for air leaks, you loos sensitivity, and filaments will die in days.
I think He will be carrier choice for next 10 years. maybe later, H2 can replace He if price rises 10X higher.