In response to analyses of H₂ of some 25,000ppm reported by Landis and Hofstra in their paper "Ore genesis constraints on the Idaho cobalt belt from fluid inclusion gas, noble gas isotope and ion ratio analyses", Economic Geology Sept.-Oct. 2012, V107, #6, P1189, I submitted a discussion paper to query the reliability of analysis of H₂ by mass spectrometry. This was published in Economic Geology August 2013, V108 #5 p1211, together with a reply from Landis and Hofstra. This discussion and reply is here (as a pdf file).

The intent of the discussion was to bring attention to many unexplained or incorrectly explained analyses of hydrogen within aqueous analytes by mass spectrometry (MS) and to suggest that confirmation analyses using a more robust analytical technique such as Laser Raman analyses should be performed before relying on mass spectrometric data for H₂.

Although there may actually be H₂ in the fluids analysed by Landis and Hofstra, they have chosen not to provide independent verification by Laser Raman Spectroscopy and the issue of unexpected and potentially erroneous MS analyses of H₂ reported by many authors has not been dealt with. The problem of suspiciously high H₂ results by MS was identified by D. Norman  and F. Sawkins (Analysis of volatiles in fluid inclusions by mass spectrometry, Chemical Geology, V61, #4, March 1987) who suggested that the H₂ was the product of reaction of CH₄ and H₂O within the inclusions during analysis. But this reaction is thermodynamically improbable as previously discussed here.  Landis and Hofstra provide confirmation in their reply that this chemical reaction is not the cause of high H₂ results. Unfortunately this erroneous proposed mechanism seems to have been accepted and prevented proper further evaluation of the H₂ analytical problem to find the real cause of the spurious results.

Landis and Hofstra refer to "background" levels of H₂ in the mass spectrometer. This is precisely the problem. Why should there be any background H₂ in the spectrometer? Why do residual gas analyses of supposedly empty vacuum chambers ( shown here, and here, and here) consistently show high levels of H₂ at mass 2 when we are quite sure that H₂ is not a background gas in the atmosphere? It seems that it is assumed that H₂ comes from the walls of the instrument, but surely this is an obfuscation to avoid the absence of a proper explanation. Note that despite the very careful analysis of the cause of all the peaks in this spectrogram, there is no explanation of the peak at mass 2 for H₂, merely a statement that it is caused by hydrogen. But why should there be any hydrogen in residual atmospheric air? (The website owners have not responded to a request for an explanation.)

The extremely high analyses of H₂ by mass spectroscopy in fluid inclusion volatiles in this sample of proterozoic quartz is also discussed here.


The proterozoic aged quartz sample analysed here was collected from surface outcrop as hand sized fragments in 1980 and crushed to <400 micron size and stored with only a thin plastic cover without even being airtight. The approximately 30 years of storage under atmospheric conditions alone is enough for diffusion to seriously deplete any hydrogen which may have been present in the fluid inclusions. It is inconceivable that the H₂ reported in this analysis, supposedly even more than the CO₂, originated from the trapped inclusion volatiles.

These problems are indicative of an unexplained instrumental issue. Landis and Hofstra dismiss this as "fundamental mass spectrometry", but I suggest that there is indeed a problem with fundamental mass spectrometry when the presence of water in the analyte is ignored. Landis and Hofstra state that they "analyze mixtures of individual gas species with N₂ or Ar to document their relative ionization efficiencies and fragmentation in the source of our quadrupole mass spectrometer." But they do not mention if this is done in a vacuum system saturated with water, which is the normal condition for fluid inclusion volatiles analysis. As discussed here, the presence of water in the vacuum could cause significant interference with H₂ analyses. Correction coefficients determined in a dry vacuum may not be appropriate for analyses under wet conditions because of H⁺ ion impacts with the H₂O adsorbed on all the apparatus electrodes, potentially generating H₂ from water alone.

Summary

Mass spectrometric analysis of H₂ in fluid inclusions has long been recognized as problematical by various authors. Analyses of residual gas in mass spectrometer vacuums also routinely shows the presence of H₂ at mass 2, but is only explained as "background" when it is unclear why there should be any background of H₂ in residual atmospheric air. I suggest that the problem is caused by the generation of H₂ from water in the electron impact ionizer as explained here. It is incorrect to merely ignore the presence of H₂O adsorbed onto the electrode surfaces of the instrument. Positive ions from the ionizer are attracted to these water coated electrode surfaces, which are negatively charged, where the high energy impacts can potentially create additional species with higher mass, while conventional MS theory assumes that only fragments with lower mass are produced in the ionizer.

This effect, which generates H₂ (mass 2) from water in the analyte of mass spectrometers, explains the H₂ levels in residual gas instead of presuming it to be unexplained "background" and also explains the many incorrect H₂ analyses of fluid inclusion volatiles.

It is unsafe to trust H₂ analyses of fluid inclusion volatiles performed by mass spectrometry as H₂ is a spurious by-product of the electron impact ionizer when operating under "wet" conditions.

Back to hydrogen analysis index page