Peter is right. “This [the choice of helium vs. hydrogen] has been discussed repeatedly and at length”. Check, for example, my previous postings.
Let me approach the question from a little different angle. I will consider only purely chromatographic factors. To simplify the comparison and to compare apples with apples, I will compare the use of helium and hydrogen under the following conditions:
a) in the same column
b) at optimal gas flow rates for each gas (for example, in 0.25 mm columns, it is 2 mL/min for helium and 2.5 mL/min for hydrogen; additional information on carrier gas optimal flow rate can be found in my previous postings; more info can be found in a recently published book [1]).
c) at optimal heating rate for each gas which is typically close to 10 ºC per hold-up time (also discussed in my previous postings).
To further simplify the comparison, I will also limit it to the most typical case of thin film columns (stationary phase film thickness has minor effect on column plate height).
Under these conditions, the choice of the carrier gas has no effect on the peak resolution – the resolution of each peak pair in GC analysis with hydrogen is exactly the same as that with helium.
*************************************
Before going directly to the comparison, it is worth mentioning that the difference between column operations with helium and hydrogen depends on the degree of the gas decompression along the column.
In relatively short wide bore columns (like 30 m – 0.53 mm) operating with outlet at atmospheric pressure, pressure drop along the column is much smaller than outlet pressure. As a result, there is almost no change in the pressure along the column and the gas decompression is weak (there is almost no gas decompression, i.e there is almost no change in the gas density along the column). These types of operations are typically used for GC analyses of relatively simple mixtures.
On the other hand, GC analyses of relatively complex mixtures use relatively long and relatively narrow bore columns. In these types of GC analyses and in all GC-MS operations (vacuum at the column outlet), the column inlet pressure is much larger than the outlet pressure. Large difference in the inlet and outlet pressure causes strong gas decompression (significant change in the gas density) along the column.
*************************************
Following are at least four advantages of hydrogen over helium as a carrier gas in GC.
1. The time of GC run with hydrogen is 20% to 40% shorter than that with helium. Actual difference in the run times depends on the degree of the gas decompression along the column.
1.a. When the gas decompression is weak, analysis time is inversely proportional to gas flow rate which, in turn, is proportional to diffusivities of analytes in the gas [1]. As mentioned earlier in this posting, optimal flow rate of helium is 20 % lower than that of hydrogen. Therefore, the run time of analysis using hydrogen is 20 % shorter than the run time of analysis using helium.
1.b. When the gas decompression is strong, analysis time is proportional to square root of the ratio (gas viscosity)/(analyte diffusivity in the gas) [1] which for hydrogen is about 40 % lower than for helium. Therefore, the run time of analysis using hydrogen is 40 % shorter than the run time of analysis using helium.
1.c. In intermediate cases where the gas decompression is noticeable, but not strong, the run time of GC analysis with hydrogen is somewhere between 20 % to 40 % lower than that for helium. Formulae for the intermediate cases can be found in ref. [1].
2. Faster analysis with hydrogen causes all peaks in it to be narrower. This reduces (improves) detection limit of low concentration analytes. The improvement is greater in the case of the strong gas decompression where the peaks in the analysis with hydrogen are 40% narrower than they are in the analysis with helium (see item 1.b).
3. Column pressure drop (the difference between the inlet and outlet pressure) is for hydrogen 45 % lower than for helium at low gas decompression and 25 % lower than for helium at strong gas decompression [1]. Since the pressure difference is practically important only when the pressure is relatively high, only the latter difference is worth remembering, that is, when the column pressure is relatively high, it is for hydrogen lower than for helium by about 25 %.
4. Helium with its small molecules can permeate through the silica column walls. This can reduce the column flow, increase hold-up time and cause unpredictable behavior of GC analysis. According to the published data [2], the hold-up time can rapidly increase with the column temperature above 250 ºC and can become several times larger than its expected value without the permeation. Longer columns should be more susceptible to this phenomenon than the shorter ones.
There is probably not enough experimental data to definitively confirm practical significance of this phenomenon, many good chromatographers disregard it, and, until recently, I accepted their judgment. However, several months ago, I participated in a series of experiments targeting the hold-up time measurements in a wide temperature range. We used helium as a carrier gas. The data went crazy in unexplainable way until we realized some similarity of our results with the ones described in ref. [2]. We replaced helium with hydrogen and all problems went away. Unfortunately, we did not have time to go back to helium and evaluate the phenomenon of its permeation as such. (Somebody is interested?). However, I no longer disregard its practical significance. And, if that is the case, then
High temperature GC analyses (above 300 ºC) with helium as a carrier gas are a suspect and should be avoided at least until the phenomenon is better understood.
****************************************
A Myth
It is widely assumed that minimal plate height of hydrogen is lower than that of helium. If this was true, it would mean that the best resolution obtainable in the same column with hydrogen is better than the resolution obtainable with helium. This assumption has no experimental evidence and contradicts to theory. In a previous posting
viewtopic.php?f=2&t=9370&p=44410#p44410, I pointed out that all known claims of lower plate height of hydrogen are based on a single source – the experimental curves originally published in a 1979 Hewlett-Packard brochure [3] where the difference between the lowest plate heights of hydrogen and helium was well within the measurement errors and the errors in construction of the curves.
Conclusion. There are many good chromatographic reasons to use hydrogen over helium in GC. However, the best resolution obtainable in the same column is not one of them.
************************************
References
[1] L. M. Blumberg, Temperature-Programmed Gas Chromatography, Wiley-VCH, 2010.
[2] J.E.Cahill, D.H.Tracy, Journal of High Resolution Chromatography 21 (1998), pp. 531-539.
[3]R.R.Freeman, High Resolution Gas Chromatography, Hewlett-Packard, 1979.
