Accelerated communication

Received: 18 February 2013 Revised: 6 May 2013 Accepted: 7 May 2013 Published online in Wiley Online Library

(wileyonlinelibrary.com) DOI 10.1002/jms.3232

Indirect hydrogen analysis by gas chromatography

coupled to mass spectrometry (GC–MS)
V. Varlet,* F. Smith and M. Augsburger
Gas chromatography (GC) is an analytical tool very useful to investigate the composition of gaseous mixtures. The different
gases are separated by specific columns but, if hydrogen (H2) is present in the sample, its detection can be performed by a
thermal conductivity detector or a helium ionization detector. Indeed, coupled to GC, no other detector can perform this
detection except the expensive atomic emission detector. Based on the detection and analysis of H2 isotopes by low-pressure
chemical ionization mass spectrometry (MS), a new method for H2 detection by GC coupled to MS with an electron ionization
ion source and a quadrupole analyser is presented. The presence of H2 in a gaseous mixture could easily be put in evidence by
the monitoring of the molecular ion of the protonated carrier gas. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: hydrogen; helium; gas chromatography–mass spectrometry; thermal conductivity detector; carrier gas protonation

Introduction As a result, GC–TCD remains the unique analytical tool to de-

Hydrogen (H2) is one of the main putrefactive gases. It was found in
human bodies in cadaveric alteration in forensic studies in a range
going from 0.5 to 50%.[1,2] A similar range of concentrations was
also noticed in rabbits[3] and whales cadavers.[4,5] Therefore, the
detection of this gas is of high importance in forensic toxicology.
Due to its physicochemical properties (explosive), its detection is
also crucial in environmental and exposure monitorings.
H2 analysis is possible by mass spectrometry (MS). MS was
employed to detect H2 isotopes. Some specific analysers[6] and
high-resolution systems have given the best results. However,
when coupled to a prior gas chromatographic separation, H2
analysis is not possible, especially because no carrier gas lighter
than H2 is available. Indeed, according to the mass spectrometer
system, it has already been shown that gas chromatography
coupled to MS (GC–MS) analysis of a lighter gas than the carrier
gas was not possible.[7] Thus, GC–MS analysis of helium (He)
was possible using H2 as carrier gas whereas using nitrogen
(N2), it displays surprising negative peaks with a quadrupole mass
spectrometer.
Consequently, GC analysis of H2 is only possible with a ther-
mal conductivity detector (TCD), a helium ionization detector
(HID) and an atomic emission detector (AED). For the analysis
of most gases, He is the carrier gas of choice because of the
large difference of thermal conductivity between He and
other gases which enhances the TCD sensitivity. However,
He and H2 thermal conductivity are not very different. Therefore,
N2 can be used as an alternative but if a double TCD and MS
detection is monitored, the MS chromatography with N2 as
carrier gas is less optimal than with He. Nevertheless, H2 TCD
detection with He as carrier gas is possible but complex because
the thermal conductivity of the He–H2 binary mixture is not a
simple monotonic increasing function of H2 concentration and
goes through a positive local minimum at low concentrations
of H2 before increasing (in negative values) to the pure H2
concentration (100 %).[8]
914
tect H2 in a gaseous mixture. However, another GC detection of
H2 is possible using MS. Based on the detection and analysis of
H2 isotopes by low-pressure chemical ionization MS, it is
possible to perform an indirect detection of H2 with an electron
ionization (EI) ion source. Analysis of H2 isotopes by low-pressure
chemical ionization involves the detection of protonated reagent
gas ions (such as argon, neon, He or N2) rather than the direct
detection of 1H+2 and 2H+2 at masses below the scan limits of most
quadrupole GC–MS systems.[9] Gas protonation of H2, He and O2
in a N2 purged atmosphere was already studied with a quadru-
pole ion trap mass spectrometer.[10] It has shown that H2, H2O
and N2 can easily be protonated given ion signals at m/z 3 (H+3 ),
19 (H3O+) and 29 (N2H+). However, O2, and especially He, did
not exhibit any protonation. Similarly, having an EI ion source
and a quadrupole mass spectrometer, we have put in evidence
H2 because of the protonation of the carrier gas. A positive MS
signal appeared at the expected retention time of the gas used
as carrier gas when analysed as analyte with another carrier gas.
To our knowledge, this is the first time that this phenomenon is
reported with an EI ion source. Protonation of carrier gas by
hydrogen in GC–MS analysis is also described for the first time.
Material and methods
Reagents
All gases (N2, H2 and He) were obtained from Carbagas
(Lausanne, Switzerland). Putrefaction gases were sampled from
* Correspondence to: Vincent Varlet, Forensic Toxicology and Chemistry Unit,
University Center of Legal Medicine Lausanne - Geneva, CH-1011 Lausanne,
Switzerland. E-mail: vincent.varlet@chuv.ch
Forensic Toxicology and Chemistry Unit, University Center of Legal Medicine
Lausanne - Geneva, CH-1011 Lausanne, Switzerland

J. Mass Spectrom. 2013, 48, 914–918 Copyright © 2013 John Wiley & Sons, Ltd.
Indirect hydrogen analysis
human bodies at the University Center of Legal Medicine of
Lausanne (Switzerland).
Analytical equipment
An Agilent 6890N GC (Agilent Technologies, Palo Alto, CA)
combined to a headspace gas autosampler and equipped with
an Agilent Select Permanent Gases column was used. This col-
umn is specially designed for gas analysis and is constituted of
two capillary columns set in parallel: a molecular sieve 5 Å PLOT
capillary column (10 m  0.32 mm) and a Porabond Q (50 m 
0.53 mm). The temperature program was 40  C, held during
5 min. Injector (splitless mode) was set at 100  C and interface
MS temperature at 230  C. He and N2 were employed as carrier
gas. The detection was performed by an Agilent 5973 mass spec-
trometer (Agilent Technologies, Palo Alto, CA), operating in the EI
mode at 70 eV and with a TCD (Agilent Technologies, Palo Alto,
CA) maintained at 170  C. The end of the Select Permanent Gases
column was connected to a three-way valve enabling the gases
to be directed to TCD or MS detectors. The valve was operated
manually to connect the GC effluent to TCD or MS.
Results and discussion
Chromatographic system
The chromatographic system was used to perform a gas screening
in forensic studies. This equipment allows in the same run the
Figure 1. Chromatographic design for hydrogen and other gas analysis.
Figure 2. MS chromatogram (Total Ion Current) of gases in right heart ventricule gas mixture sampled from a human body in advanced cadaveric
alteration.
J. Mass Spectrom. 2013, 48, 914–918 Copyright © 2013 John Wiley & Sons, Ltd.
separation of permanent gases such as H2, He, argon, oxygen
(O2), N2, methane (CH4) and carbon monoxide (CO) on the
molecular sieve 5 Å PLOT capillary column, and the separation of
CH4, hydrogen sulfide (H2S), carbon dioxide (CO2) and heavier
volatile hydrocarbons on the Porabond Q (with no separation of
air constituents) (Fig. 1). Moreover, the double detection (MS and
TCD) allows to get the individual and precise concentration of a
gas in a gaseous mixture (MS quantification) and to get the individ-
ual proportion of this gas (percentage) in the total gaseous mixture
(TCD quantification). However, TCD allows the analysis of H2
whereas MS cannot theoretically permit this measurement.
This chromatographic system was applied to the gas screening
of gaseous samples from human bodies in cadaveric alteration.
Figure 2 displays the MS chromatogram result of right heart
ventricule gas mixture sampled from a human body in advanced
cadaveric alteration. Important amounts of CO2 were noticed
followed by N2 and O2 concentrations and finally, traces of CH4,
ammonia (NH3) and H2S. H2 was not identified on this MS
chromatogram where its signal was expected just before O2
signal retention time.
GC–MS H2 identification with N2 as carrier gas
Figure 3 displays the MS chromatogram of a pure H2 pulse in GC
injector. Unexpected peaks at the retention times of N2 on
Molecular sieve and Porabond Q are noticeable. Blank or N2
injections in same chromatographic conditions have shown no
baseline variation. Air injection caused a chromatographic signal
915
wileyonlinelibrary.com/journal/jms

Figure 3. MS chromatogram (Total Ion Current) of pure hydrogen pulse in GC injector (N2 as carrier gas).
only for O2, showing that N2 of air cannot be put in evidence
when N2 is used as carrier gas (Fig. 4). Focusing on the first peak
obtained in Fig. 3, it appears that the positive MS signal (Fig. 5(A))
corresponds to a negative TCD signal (Fig. 5(B)) specific to H2
when injected in important quantity.[11] The mass spectrum of
this signal exhibits a molecular ion at m/z = 29, with a second
ion at m/z = 14. This mass spectrum corresponds to the mass
spectrum of N2 except that the molecular ion is at m/z = 29
instead of m/z = 28. It appears that the carrier gas (N2) has been
protonated by H2 eluting from the chromatographic system
and has given a signal at the expected retention time of the
gas when not used as carrier gas, but with a molecular ion
corresponding to m/z + H+, herein m/z = 28 + H+ = 29.
GC–MS H2 identification with He as carrier gas
Figure 6 displays the MS chromatogram of a H2 mixture in air.
Unexpected peaks at the retention times of He on Molecular
sieve and Porabond Q are noticeable. Blank or He injections in
the same chromatographic conditions have shown no baseline
variation (Fig. 7). Positive MS signals were obtained (Fig. 6(A)).
916
Figure 4. Air injection (N2 + O2) with N2 as carrier gas.
wileyonlinelibrary.com/journal/jms Copyright © 2013 John Wiley & Sons, Ltd.
V. Varlet, F. Smith and M. Augsburger
Extract ion chromatograms and mass spectrum have shown a
main ion at m/z = 5 (Fig. 6(B) and (C)). This mass spectrum corre-
sponds to the mass spectrum of He except that the molecular ion
is at m/z = 5 instead of m/z = 4. It appears that the carrier gas (He)
has been protonated by H2 eluting from the chromatographic
system and has given a signal at the expected retention time of
the gas when not used as carrier gas, but with a molecular ion
corresponding to m/z + H+, herein m/z = 4 + H+ = 5.
Whatever the carrier gas, the eluting H2 protonates the carrier
gas resulting in a signal at the retention time of the gas when not
used as carrier gas. This unexpected reaction of chemical ioniza-
tion of the carrier gas by an eluting compound using an EI ion
source allows a reliable detection of H2 by GC–MS. TCD is still
important for quantification, but this result permits to better
investigate MS chromatograms of injected gaseous mixtures.
Therefore, without TCD detector, H2 cannot be identified in
gaseous samples where its presence is expected such as in a
heart ventricule gas mixture sampled from a human body in
advanced cadaveric alteration (Fig. 2). However, knowing the
behaviour of H2 with the carrier gas, it is possible to identify H2
by GC–MS. Performing an extracted ion chromatogram of the
J. Mass Spectrom. 2013, 48, 914–918
Indirect hydrogen analysis

Figure 5. Pulse hydrogen in GC injector (N2 as carrier gas), A : MS signal (zoom from Fig. 5), B : TCD signal (shift of 0.20 min between A and B is due to
the chromatographic design), C : Mass spectrum obtained for H2 peak.

Figure 6. Mixture of hydrogen in air (He as carrier gas), A : MS chromatogram of hydrogen mixture in air, B : Extracted ion chromatogram from MS
917

chromatogram A at m/z = 5, C : Mass spectrum obtained for unexpected peaks.

J. Mass Spectrom. 2013, 48, 914–918 Copyright © 2013 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jms

Figure 7. Blank injection (no injection) MS chromatogram (He as carrier gas).
Figure 8. Extracted ion chromatogram (m/z = 5) from MS chromatogram of Fig. 2 (He as carrier gas).
MS chromatogram at m/z of protonated carrier gas, a positive signal
of the carrier gas should indicate the presence of H2 as shown in
Fig. 8. However, unexpected double peaks were identified: proton-
ation of He by H2 is noticeable, but another signal is displayed at
O2 retention time. The cause of this phenomenon remains unknown.
Conclusion
An unusual reaction of carrier gas protonation by H2 was noticed
during H2 analysis by GC–MS/TCD with an EI ion source. This
reaction could be very useful in GC–MS gas screening and was
confirmed by two carrier gases (He and N2). Indeed, the presence
of H2 in a gaseous mixture could be put in evidence easily by the
monitoring of the molecular ion of protonated carrier gas. It gives
the opportunity to detect H2 with another GC detector and not to
be obliged to be equipped with TCD, HID or AED.
Acknowledgement
The authors greatly thank Kjetil Johnsen (Johnsen Consulting &
Services) for his chromatographic expertise during scientific discussions.
References
[1] T. Bajanowski, A. West, B. Brinkmann. Proof of fatal air embolism. Int.
J. Legal Med. 1998, 111, 208–211.
[2] T. Bajanowski, H. Köhler, A. DuChesne, E. Koops, B. Brinkmann. Proof of
918
fatal air embolism after exhumation. Int. J. Legal Med. 1998, 112, 2–7.
wileyonlinelibrary.com/journal/jms Copyright © 2013 John Wiley & Sons, Ltd.
V. Varlet, F. Smith and M. Augsburger
[3] Y. Bernaldo de Quirós, O. González-Díaz, A. Møllerløkken,
A.O. Brubakk, A. Hjelde, P. Saavedra, A. Fernandez. Differentation
at autopsy between in vivo gas embolism and putrefaction using
gas composition analysis. Int. J. Legal Med. 2013. DOI: 10.1007/
s00414-012-0783-6.
[4] Y. Bernaldo de Quirós, O. González-Díaz, P. Saavedra, M. Arbelo,
E. Sierra, S. Sacchini, P.D. Jepson, S. Mazzariol, G. Di Guardo,
A. Fernandez. Methodology for in situ gas sampling, transport
and laboratory analysis of gases from stranded cetaceans. Sci.
Rep. 2011, 1, 1–10.
[5] Y. Bernaldo de Quirós, O. González-Díaz, M. Arbelo, E. Sierra,
S. Sacchini, A. Fernández. Decompression vs. decomposition: dis-
tribution, amount, and gas composition of bubbles in stranded
marine mammals. Front Physiol. 2012, 3. DOI: 10.3389/
fphys.2012.00177.
[6] J. Y. Li, L. Shi, S. L. Hu. Methodological study on measurement of
hydrogen abundance in hydrogen isotopes system by low resolu-
tion mass spectrometry. Mass Spectrom. Lett. 2011, 2, 1–7.
[7] F. Musshoff, L. Hagemeier, K. Kirschbaum, B. Madea. Two cases of
suicide by asphyxiation due to helium and argon. Forensic Sci.
Int. 2012. DOI: 10.1016/j.forsciint.2012.08.049.
[8] K. Snavely, B. Subramaniam. Thermal conductivity detector analysis
of hydrogen using helium carrier gas and HayesepW D columns.
J. Chromatogr. Sci. 1998, 36, 191–196.
[9] A. R. Mack, Y. Liu, W. Luo, L. Liu, J. F. Parcher. Detection and analysis
of hydrogen isotopes and spin isomers by low-pressure chemical
ionization mass spectrometry. Anal. Chem. 1994, 66, 115–118.
[10] A. K. Ottens C. Richard Arkin, T. P. Griffin, P. T. Palmer, W.W., Harrison.
Ion-molecule reactions in quadrupole ion trap mass spectrometry:
implications for lightweight gas analysis. Int. J. Mass Spectrom. 2005,
243, 31–39.
[11] R. Villalobos, G. R. Nuss. Measurement of hydrogen in pro-
cess streams by gas chromatography. ISA Trans. 1965, 4,
281–286.
J. Mass Spectrom. 2013, 48, 914–918