Chemical Geology 575 (2021) 120098

Contents lists available at ScienceDirect

Chemical Geology
journal homepage: www.elsevier.com/locate/chemgeo

Hydrogen and hydrocarbons associated with the Neoarchean Frog’s Leg

Gold Camp, Yilgarn Craton, Western Australia
Christopher J. Boreham a, *, Jacob H. Sohn a, Nicholas Cox b, Jodi Williams b, Ziqing Hong a,
Mark A. Kendrick c
a
Geoscience Australia, PO Box 378, Canberra, ACT, 2601, Australia
b
Evolution Mining, 175 Liverpool Street, Sydney, NSW 2000, Australia
c
School of Earth and Environmental Sciences, The University of Queensland, St Lucia, Qld 4072, Australia

A R T I C L E I N F O A B S T R A C T

Keywords: High concentrations of hydrogen (H2) and methane (CH4) have been detected in shallow open-hole exploration
Australian abiogenic natural gas wells surrounding the Neoarchean Frog’s Leg gold camp in the Eastern Goldfields Superterrane, Yilgarn Craton,
Molecular hydrogen Western Australia. After corrections for modern air contaminants and excess nitrogen (N2) the Boomer deposit
Carbon and Hydrogen and noble gas isotopes
gases contain: 19.9–68.7 mol% H2; 28.7–76.9 mol% CH4; 0.47–1.6 mol% heavier hydrocarbons (C2–C5), which
Gold
follow an Anderson-Schulz-Flory distribution; 0.11–3.3 mol% carbon dioxide (CO2); and 0.69–1.87 mol% helium
Archean
(He). The isotopic composition of the gas components was further investigated: helium has 3He/4He ratios of
1.82–3.33 × 10− 8 (from 0.013 to 0.024 RA, where RA is the atmospheric value) indicating a purely crustal origin;
H2 has δ2H between − 781.3‰ and − 759.5‰; and CH4 has δ13C between − 20.3‰ and − 2.42‰ and δ2H be­
tween − 382.5‰ and − 342.2‰. The C2–C5 gaseous hydrocarbons are commonly depleted in 13C (up to 22.75‰
for ethane) and enriched in 2H (up to 117.3‰ for iso-butane) compared to methane, while a carbon isotope
reversal is observed between methane and ethane. The molecular and isotopic characteristics of the gas are
consistent with 1) H2 generation controlled by radiolytic reactions within basic and felsic igneous rocks and by
redox reactions including serpentinization in mafic-ultramafic lithologies, and 2) methane and C2+ gaseous
hydrocarbons produced by interaction of H2 with a single carbon source (mainly CO2) most likely in Fischer-
Tropsch type reactions. Using the H2–CH4 hydrogen isotopic fractionations as a geothermometer, isotopic
equilibrium temperatures are calculated between 42 ◦ C and 57 ◦ C, which based on the modern geotherm rep­
resents a depth of ~1.4–2.7 km, corresponding to the Gleesons Basalt beneath the Frog’s Leg gold camp. The free
gases sampled from exploration wells in the Boomer deposit, Frog’s Leg gold camp represent the first positive
identification of abiogenic natural gas in Australia, with gas compositions similar to abiogenic gases previously
reported from overseas Archean cratons. Interestingly, CH4-rich gases preserved in fluid inclusions associated
with some of the region’s lode gold deposits have distinctly different isotopic characteristics, suggesting that
methane production has occurred by different pathways over a period from the Neoarchean to present day in the
Yilgarn Craton.

1. Introduction associated with reactive Fe-rich rocks including mafic and ultramafic
lithologies and ironstones. Seeps of abiogenic CH4 and H2 are known in
Most natural gas is rich in methane (CH4) and dominantly biogenic in ophiolites and contribute to modern vents formed above peridotite on
origin, meaning that it is produced either by microbes (microbial gas; the seafloor and in ancient cratons (Reeves and Fiebig, 2020 and
Nandi and Sengupta, 1998; Formolo, 2010) or by thermal maturation of reference therein). Abiogenic methane can also be trapped in fluid in­
organic matter (thermogenic gas; Tissot and Welte, 1984; Hunt, 1996). clusions found in alkaline magmatic rocks and in some hydrothermal
However, abiogenic hydrocarbons (predominately CH4) associated with settings (ibid). In a few cases, abiogenically formed gases are also trap­
unusually high concentrations of molecular hydrogen (H2) are found ped in H2-rich (>10 mol% H2) gas fields in sedimentary basins (Neal and

* Corresponding author.
E-mail address: [email protected] (C.J. Boreham).

https://doi.org/10.1016/j.chemgeo.2021.120098
Received 17 December 2020; Received in revised form 26 January 2021; Accepted 27 January 2021
Available online 11 February 2021
0009-2541/Crown Copyright © 2021 Published by Elsevier B.V. All rights reserved.
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Stranger, 1983; Coveney Jr. et al., 1987; Smith et al., 2005; Zgonnik, scale, gold mineralisation is strongly associated with the greenstones,
2020 and references therein). but is also hosted by sediments, granitoids and mafic–ultramafic
It is important to identify and characterise new occurrences of greenstones at the deposit scale (Fig. 1; Golding et al., 1990; Groves
abiogenic CH4 and H2 from several perspectives. Plentiful natural et al., 1995; Hagemann and Cassidy, 2000). The major phase of gold
abiogenic CH4 and H2 represent a source of energy for industrial usage. mineralisation occurred between 2640 and 2630 Ma (Cassidy et al.,
Both gases can support chemotrophic ecosystems and might have 2006; Huston et al., 2012). The gold-bearing ore fluids are usually
contributed to the development of early life on Earth (Ménez, 2020 and attributed to either a metamorphic or magmatic source (Thébaud and
references therein) and other planets in our Solar System (Glein and Miller, 2019 and references therein) and a variety of precipitation
Zolotov, 2020 and reference therein). Abiogenic CH4 and H2 might in­ mechanisms including fluid mixing, fluid-rock reaction (e.g. in iron-rich
fluence crustal redox chemistry and have even been suggested as a lithology), phase separation, and vapour condensation have been pro­
possible control on the localisation of gold mineralisation in the Yilgarn posed (Huston et al., 2012).
Craton (Kendrick et al., 2011). The possible commercial significance of Pressurised (~1.4 MPa) natural gas rich in CH4 and nitrogen (N2)
abiogenic CH4 and H2 has been speculated for some time (Goebel et al., was reported at depths of 210 m in a borehole at the township of Kal­
1984; Smith et al., 2005). A recent estimate of abiogenic H2 production goorlie in the Eastern Goldfields Superterrane of the Yilgarn Craton as
from continental and oceanic crust (down to 10 km), water radiolysis, early as 1905 (Simpson, 1912). Methane and N2 have also been reported
volcanism and volcanic degassing (Truche et al., 2020) at 2.8 M-tonnes in fluid inclusions associated with sulphide minerals and gold mineral­
(3.1*1010 m3) per year, is a fraction of the industrial production of >70 isation in the Conqueror and Junction gold deposits in the Eastern
M-tonnes of H2 (8.4*1011 m3) in 2018 (IEA, 2019). Nonetheless, com­ Goldfield Superterrane (Fig. 1a; Polito et al., 2001; Kendrick et al.,
bined data from the Western and Eastern (former Soviet block) literature 2011). Gas discharges associated with mineralised zones are common­
suggest that global H2 occurrences may have been under-estimated place, and H2 is often present in gold mines (Nivin, 2019; Zgonnik, 2020
(Zgonnik, 2020). Furthermore, the recent documentation of a H2-rich, and references therein). Therefore, improving our understanding of the
over-pressurised gas field (97.4 mol% H2) from the shallow Bougou-1 possible origins of modern and ancient CH4 is critical to improve gold
well, Taoudeni Basin, Mali (Prinzhofer et al., 2018) that is used for mineralisation models that involve abiogenic methane (Walshe et al.,
small-scale electricity generation, attests to the niche potential of com­ 2006; Kendrick et al., 2011). Furthermore, if a genetic link between
mercial H2 exploitation. As a result of its potential and the development abiogenic CH4 and gold can be established, detection of near-surface
of new H2 exploration strategies (Dugamin et al., 2019; Zgonnik, 2020), free gases (CH4 and/or H2) could become an aid to gold exploration
abiogenic CH4 and H2 is now becoming an increasingly attractive and help identify deeper buried gold mineralisation (Polito et al., 2002)
exploration target. or gas reservoirs in the Eastern Goldfields Superterrane.
Abiogenic hydrocarbon fields have not yet been positively identified This study investigated free gases from recent shallow exploration
in Australia (Etiope and Whiticar, 2019). However, gas fields globally wells from the Boomer deposit in the Frog’s Leg gold camp (Fig. 1a). The
with >10 mol% H2 (Coveney Jr. et al., 1987) most likely have contri­ gases were analysed for their molecular composition and carbon,
butions from abiogenic sources. In Australia, these include: i) gas seeps hydrogen, nitrogen and noble gas stable isotopic systematics, enabling
with ~75% H2 and ~ 25% CH4 in the Cobar copper mine, New South their abiogenic origin to be unambiguously fingerprinted and direct
Wales (Stack and Rabone, 2018); ii) historic reports of wells with gases comparisons to be made with noble gas data available for CH4-rich fluid
that have up to 89% H2 air-free drilled in the Yorke Peninsula and on inclusions (Kendrick et al., 2011). Important lithologies in the Frog’s Leg
Kangaroo Island of South Australia (Ward, 1917, 1932, 1933, 1944; gold camp include: 1) the White Flag Formation (the lower formation of
Woolnough, 1934); iii) water bores of the Great Artesian Basin in New the Black Flag Group) containing intermediate volcanoclastic rocks,
South Wales (Woolnough, 1934); and iv) petroleum explorations wells andesite (upper part), sandstone, siltstone and shale (lower part), 2)
with up to 38.2 mol% H2 in Golden Beach 1, offshore Victoria, (http://ge Powder Gabbro containing coarse-grained leucogabbro, and 3) the
oscience-web.s3-website-ap-southeast-2.amazonaws.com/we plagioclase-phyric basalt of the Victorious Basalt and massive tholeiitic
ll/goldenbeach1.htm, accessed 25 February 2021), up to 95.2 mol% H2 basalt of the Bent Tree Basalt (Fig. 1b). The ultramafic Kambalda
in Meda 1, Canning Basin, Western Australia,(https://wapims.dmp.wa. komatiite is a minor lithology (see Supplementary material: Appendix
gov.au/WAPIMS/Search/Wells, accessed 25 February 2021), and 11 S1 for more detailed local geology). The gold deposits in the Frog’s Leg
mol% H2 in Mt. Kitty 1, Amadeus Basin, Northern Territory (McInnes gold camp are hosted by the major K2 shear zone (Fig. 1), which pro­
et al., 2017). The question of whether these limited occurrences of H2- vides an obvious fluid conduit from lithologies in the lower crust (see
rich natural gases (based on gas chromatographic analysis only) resulted Section 4.4).
from natural geological sources or are drilling artefacts (see Boreham
et al., 2021 and references therein for an overview of H2 sources) re­ 2. Samples and methodology
mains unresolved. Past petroleum exploration activities in Australia
have been focussed on sedimentary basins with little penetration into Gases sampled from ten of Evolution Mining’s exploration drill holes
fractured basement rocks where abiogenic gas is most likely to originate. encompass the Boomer deposit in the Frog’s Leg gold camp (Fig. 1). The
Therefore, new natural H2 and abiogenic hydrocarbons might yet be sampling of the gas was opportunistic and secondary to the mineral
found in depositional settings previously considered unsuitable for pe­ exploration purpose of the holes, as such, well head pressures etc. were
troleum exploration (Coveney Jr. et al., 1987). This paper reports on H2- not ascertained due to this not being part of normal diamond drilling
, helium (He)- and CH4-rich gases associated with mafic–ultramafic practice. The gases were analysed for major gas species including H2,
rocks in gold mineralised zones of the Archean Yilgarn Craton, Western He, CH4, CO2, N2, O2 and C2–C5 hydrocarbons. Based on appropriate
Australia. concentrations, samples were selected for carbon, hydrogen and nitro­
gen isotopic analysis of the individual gas components. The gases
1.1. Yilgarn Craton and associated gases recovered from three of the drill holes were also analysed isotopically for
noble gases. In addition, fluid inclusions (H2O-CO2 and CH4) in two
The Archean Yilgarn Craton of Western Australia is a granite- quartz veins from the ~2.65 Ga Conqueror deposit, St Ives Goldfield
greenstone belt that has been metamorphosed at greenschist to (Fig. 1; Huston et al., 2012; Witt et al., 2020), which have been analysed
amphibolite facies conditions. Granites make up around 65% of the for noble gases previously (Kendrick et al., 2011), were analysed for
Yilgarn crust (Fig. 1) and are mostly of the tonalite-trondhjemite- major gas species and their C- and H-isotopes. This was undertaken to
granodiorite series (Huston et al., 2012). The mafic and ultramafic enable a comparison of multiple geochemical tracers between modern
greenstones include both basalt flows and komatiites. At the regional and ancient CH4-bearing fluids in the Eastern Goldfields Superterrane.

2
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 1. Location map of the Yilgarn Craton and Eastern Goldfield Superterrane, specific gold deposits discussed in the text, stratigraphy and surface geology in the
surrounds of the Frog’s Leg gold camp. The base map (top) is modified from Champion and Sheraton (1997) while the stratigraphy and gold deposit positions within
the Kundana Goldfield are from Cooke et al. (2017). The diagonal red ‘cross-section’ represents the position of the surface expression of the bottom vertical section in
Fig. 13. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Analytical methods are reported in the Appendix.

1.33
6.90
3.07

53.0
11.2

2.46
The Boomer exploration boreholes were drilled to depths of 145 m to

n-C5
250 m into the Black Flag Group (Fig. 1b). The rocks penetrated
comprise felsic volcanoclastic intercalated with sedimentary rocks and

1.68
9.70
3.57

30.1
15.3

3.98
cut by various igneous units including basalt dykes, andesites and
i-C5
dacites (see Supplementary Material: Appendix S1). The wells were
drilled between late November 2019 to early January 2020 and the gas
10.6
53.4
22.9
1.31
1.03
4.49
5.09
79.4

33.6
samples were collected immediately after drilling, once rods were
n-C4

ppm

removed from and prior to the grouting of the hole. The well’s PVC
collar was sealed with a thick plastic sample bag and allowed to sit for
30 min. The cylinder valve at the surface end of a 2 m length of 1/8”
0.340
0.338
4.33
19.9
7.66

1.23
1.67
26.6

10.3
i-C4

teflon tubing (previously been fed down the collar) and attached to an
evacuated (0.01 kPa) Swagelok stainless steel cylinder (~150 mL ca­
pacity) was opened for 30 s. The valve was then closed and the tube
231.9

315.7

119.6
42.8

87.2
5.55
3.25
18.3
26.7

removed from the collar. Well depth readings (Table 1) are relative to
C3

the height of the PVC collar casing at ~15–30 cm above ground surface
depending on ground relief.
0.030
0.153
0.062
0.003
0.002
0.012
0.021
0.246

0.106
C2

3. Results
18.193

The molar volumes (mol%) of Air, H2, He, hydrocarbon species

1.858
8.918
4.733
0.123
0.052
0.519
0.803

5.905

(C1–C5), CO2, and excess N2 (N2excess) in gases from each of the 10 drill
C1

holes are reported in Table 1. Table S1 presents the data on an air-free

and on an air- plus N2excess-free basis. We note that the observed air
42.731
1.266
5.437
8.126
0.221
0.100
0.498
1.123

0.001
1.530

contamination of 30.7 to 92.7 mol% does not significantly affect our

H2

ability to measure the relative abundances of non-atmospheric compo­

nents (see Nivin, 2019). The nitrogen isotope (δ15N) compositions of N2
in two Boomer gases are reported in the footnote of Table 1 and are close
0.047
0.238
0.152
0.003
0.003
0.020
0.033
0.931

0.081

to that of air. However, the carbon and hydrogen isotopic composition of

He

mol%

excess N2 = N2-total − 3.73*O2, however O2 and Ar are not resolved under GC conditions so N2excess is a minimum value.

the H2 and individual C1–C5 gaseous hydrocarbons given in Table 2 are

not influenced by air contamination. Noble gas data was obtained for
0.06
0.06
0.07
0.11
0.03
0.05
0.10
0.08
0.05
0.07
CO2

three samples least affected by air contamination (Table 3 and Table S2).
The major gas and isotope compositions of the fluid inclusions are given
in Table 4.
excess N12

The Boomer deposit natural gases have H2 contents ranging

11.70

17.28
13.29
17.59
21.24

0.001–42.73 mol% and up to 18.19 mol% CH4 with the highest H2 and
7.24
9.56

7.05
7.28
9.71

CH4 contents in EVDD0083 (Table 1). These H2 concentrations are

within the overall range of Australian biogenic gases (Fig. 2a). However,
the air-free abundances of 19.9–61.9 mol% H2 (Table S1) are unusually
Air (N2 + O2)

high and the Boomer gases are further distinguished from typical
Australian thermogenic gases by higher C1/(C2 + C3) ratios and lower
85.03
77.92
77.28
82.26
86.52
81.31
76.67
30.73
92.66
82.58

CH4/H2 ratios (Fig. 2b). The concentration of H2 is strongly correlated

with He giving an average H2/He ratio of 38 (Fig. 3a; correlation coef­
ficient r2 = 0.98) and moderately correlated with CH4 (Fig. 3b; r2 =
121.26409
121.26419
121.26393
121.26343
121.26389
121.26392
121.26385
121.26397
121.26425
121.26441
Longitude

0.74), suggesting an intimate association between He, H2 and CH4. In

contrast, correlations are neither preserved between H2 and air-free N2
(N2excess; Table S1), which the latter gas likely has a modern air
component (see below) nor between H2 and CO2, which the latter gas is
a very minor component (Fig. 3b).
30.76148
30.76197
30.76103
30.76201
30.76108
30.76216
30.76112
30.76131
30.76255
30.76307

The Boomer gases contain up to 0.29 mol% wet gases (C2–C5). The
Latitude

relative abundances of the gaseous hydrocarbons (C1–n-C5) are shown

graphically in an Anderson-Schulz-Flory (ASF) plot (Anderson, 1984) in
−
−
−
−
−
−
−
−
−
−

δ N N2 = 0.22 ‰ for 3 and − 0.10 ‰ for 4.

which the normalised (ƩC1–n-C5 = 1) abundance of each gas component

is shown on a log-scale (Fig. 4). The gas components show a strong
190.14
210.60
145.10
255.10
150.00
245.00
175.20
170.13
250.00
245.00
Depth
Molecular composition of Boomer gases.

negative linear relationship between the C1–n-C5 fractions (r2 between

m

0.943 and 0.966). However, the linear correlation is much improved

(0.984 < r2 < 0.998) when only the C2–n-C5 components are considered.
EVDD00832,3

In addition to H2 and hydrocarbons, many of the samples contain

EVDD00804
EVDD0064
EVDD0071
EVDD0075
EVDD0067
EVDD0076
EVDD0073
EVDD0077

EVDD0093

substantial excess N2excess, with abundances ranging from 7.05 to 21.2

mol% (Table 1), or from 10.2 to 99.7 mol% on an air-free basis (Table
Well

neo-C5 = 0.343 ppm.

S1). The samples also contain up to 0.11 mol% CO2 and 0.93 mol% He
(Table 1). Trace amounts of unsaturated hydrocarbons ethylene (C2H4)
and acetylene (C2H2) are found in EVDD0067 with ethylene/ethane and
20,209,001
20,209,002
20,209,003
20,209,004
20,209,005
20,209,006
20,209,007
20,209,008
20,209,009
20,209,010
SampleID

acetylene/ethane ratios of 0.0065 and 0.071, respectively. Propylene

Table 1

(C3H6) was detected in EVDD0067, EVDD0073, EVDD0075 and

3,4 15

EVDD0093 with propylene/propane ratios of 0.31, 0.051, 0.012 and

1
2

4
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Table 2
Carbon and hydrogen isotopes of Boomer gases.
Well δ13C C1 δ13C C2 δ13C C3 δ13C i- δ13C n- δ13C i- δ13C n- δ13C δ2H C1 δ2H C2 δ2H C3 δ2H i-C4 δ2H n- δ2H H2
C4 C4 C5 C5 CO2 C4

EVDD0064 − 6.19 − 22.99 − 12.69 − 367.4 − 760.8

EVDD0071 − 5.23 − 24.11 − 20.14 − 21.28 − 18.81 − 17.68 − 15.71 − 13.47 − 369.6 − 370.4 − 344.1 − 273.8 − 344.8 − 762.7
EVDD0075 − 11.36 − 24.59 − 13.78 − 374.9 − 369.1 − 759.5
EVDD0067 − 6.49 − 15.22
EVDD0076 − 18.25 − 11.65
EVDD0073 − 20.30 − 25.21 − 11.39 − 378.8 − 781.3
EVDD0077 − 14.02 − 15.20 − 370.4 − 767.7
EVDD0083 − 2.42 − 25.18 − 20.53 − 20.93 − 17.89 − 17.06 − 14.08 − 14.60 − 382.5 − 365.3 − 331.9 − 265.2 − 335.1 − 760.9
EVDD0080 − 14.48
EVDD0093 − 9.62 − 23.60 − 21.07 − 13.75 − 342.3 − 371.6 − 760.5

Table 3
Noble gas isotope data for Boomer gases.
3 4
Well He He R/R1A RC/R2A 20
Ne Ne 40
Ar Ar Ar*3 84
Kr Kr 132
Xe Xe

pcc/cc μcc/cc μcc/cc μcc/cc μcc/cc μcc/cc μcc/cc ncc/cc ncc/cc ncc/cc ncc/cc

ASW
EVDD0071 99.92 2996.31 0.0241 0.0237 5.082 5.62 8276.81 8309.42 160.10 994.40 1747.63 53.95 200.55
EVDD0075 68.42 1996.34 0.0248 0.0234 10.804 11.94 8652.69 8687.20 74.74 992.96 1745.10 50.04 186.03
EVDD0083 248.12 13,621.14 0.0132 0.0131 2.295 2.54 4575.91 4592.92 346.56 299.89 527.04 11.42 42.47
1
R/RA = Rsample/Rair where Rsample = 3He/4He sample and RA = 1.384 × 10− 6.
2
RC/RA = [[R/RA*(He/Ne)]− [0.985*(He/Ne)ASW]]/[(He/Ne)-(He/Ne)ASW] where ASW (air saturated water) = 0.288 for clear air.
3
Ar* = Ar x (1− 296/(40Ar/36Ar)).

n-butane = 70‰; Fig. 6). Moreover, the hydrogen isotopes of the C2–C5
Table 4
wet gases are the same amongst the Boomer gases within experimental
Fluid inclusion gas data for Conqueror deposit, St Ives Goldfield.
error. The variation in hydrogen isotopes between C1 to C4 in the
Sample Depth C1/ δ13C C1 δ13C CO2 δ2H C1 C1–CO2 Boomer gases is broadly similar to the Australian non-biodegraded,
CO2 temp**
thermogenic natural gases shown by the dashed line in Fig. 6. In com­
m ‰ ◦
C
Q5 695 0.05# − 35.31 − 3.71 − 172.8 157 parison to the hydrocarbon gases, H2 in the Boomer gases has excep­
Q6 389 1.38 − 39.04 − 10.78 − 136.8 250 tionally low δ2H of − 760 to − 781‰ (Table 2).
Junction 0.01 to − 50.5 to − 10.4 90 to 220 The H2- and CH4-rich Boomer gas samples (EVDD0071, EVDD0075
gold 11.42 − 32.8 to − 0.6
and EVDD0083) are strongly enriched in He (0.15–0.93 mol% He;
deposit*
Table 1) compared to other noble gases and they are dominated by
*
Polito (1999); radiogenic 4He with 3He/4He ratios of 0.013–0.024 RA that are typical of
#
C1/C2 = 66.2. the continental crust (Table 3; RA denotes the atmospheric 3He/4He ratio
**
isotopic eqilibrium temperature calculated from Bottinga (1969). of 1.4 × 10− 6; Porcelli et al., 2002). In addition, the Boomer gases have
argon isotope ratios (40Ar/36Ar) ratios of 298.1–319.6 that are only
0.013, respectively. Sample EDV0080 is distinguished as the only sam­ slightly higher than the atmospheric ratio of 296 (non-atmospheric,
ple that is wholly dominated by air and excess N2excess, with H2 close to radiogenic 40Ar is denoted 40Ar*) and calculated 4He/40Ar* ratios of
the detection limit and hydrocarbons below their detection limits 18.7–39.3 that greatly exceed the average crustal production ratio of ~3
(Table 1). (Ballentine et al., 2002) (Table 3). The isotopic compositions of neon
The carbon (δ13C) and hydrogen (δ2H) isotopes of the individual (Ne), krypton (Kr) and xenon (Xe) are very close to, or indistinguishable
C1–C5 gaseous hydrocarbons are plotted in Figs. 5 and 6, respectively. from air. However, the elemental ratios of noble gases, 20Ne/36Ar,
Methane has δ13C of − 2.42‰ to − 20.30‰ compared to much lower 84
Kr/36Ar and 132Xe/36Ar ratios vary between air and air-saturated
values of − 22.99‰ to − 25.21‰ in ethane representing a carbon isotope water (Table S2). Sample EVDD0083 has 84Kr/36Ar and 132Xe/36Ar
reversal (Δ13Cmethane – ethane) of 4.9‰ to 22.8‰ (Fig. 5). Propane, bu­ isotope ratios close to air, whereas samples EVDD0071 and EVDD0075
tanes and pentanes have progressively higher δ13C than ethane (Fig. 5). preserve 20Ne/36Ar, 84Kr/36Ar and 132Xe/36Ar ratios closer to air-
The branched (iso; i-) C4 and C5 isomers are slightly depleted in 13C saturated water than air (Fig. 7).
compared to their respective normal (n-) isomers (Fig. 5). The Boomer After correcting for air contamination, excess N2 is the dominant gas
gases contrast with Australian non-biodegraded, thermogenic natural component (Table 1). Excess N2 is inversely correlated with He, H2, and
gases, which typically show increasing enrichment in 13C with C1–C5 hydrocarbons, meaning that gases with high N2excess have low He,
increasing carbon number (dashed line in Fig. 5). There are relative H2 and C1–C5 hydrocarbons (Fig. 3a for He). However, there is no
weak linear correlations between methane mol% and δ13C of methane obvious relationship between N2excess and either CO2 or air (Table S1).
(r2 = 0.43), and between C1/(C2 + C3) and δ13C of methane (r2 = 0.39) The δ15N of N2 in the gases varies from 0.22‰ in sample EVDD0083,
(Tables 1 and 2; see also Fig. 8a). which is least affected by air contamination (30.7 mol%) to a value of
The δ2H of the methane in the Boomer gases ranges from − 340‰ to − 0.10‰ in sample EVDD0080, which was affected by air contamination
− 382‰ that is more variable than the δ2H of the ethane which varies by to the greatest degree (92.7 mol%; Table 1). Considering the experi­
only 6‰ around − 368‰ (Fig. 6; Table 2). From ethane to propane there mental error is ±0.1‰ at the 1σ level, both determinations are very close
is enrichment in 2H while n-butane has a similar δ2H to propane, and i- to air but the least contaminated sample might be slightly enriched in
butane shows a larger enrichment in 2H relative to n-butane (Δ2Hi-butane - 15
N. If the excess N2 in this sample had a wholly indigenous origin, then

5
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 2. Plot of air-free compositions of a) helium mol% versus hydrogen mol%

and excess nitrogen mol%, and b) hydrogen mol% versus methane mol% and
carbon dioxide mol%.

based on mixing 7.03 mol% N2excess and 23.99 mol% N2air, the measured
δ15N value of 0.22± 0.10‰ indicates an indigenous N2 δ15N of +0.97 ±
Fig. 3. Plot of a) frequency of hydrogen mol% ranges in Australian biogenic
0.44‰ (1σ). However, it is likely that at least a component of the
natural gases analysed at Geoscience Australia, and b) mol% ratio of methane/
measured N2excess was derived from air, because some of the O2 in the air hydrogen versus mol% ratio of methane/(ethane + ethane) for Boomer gases
contaminant would have been consumed by reaction with minerals and Australian thermogenic natural gases where the δ2H H2 was measured
exposed during drilling. Therefore, the δ15N of the indigenous well gas (Boreham et al., 2021).
cannot be confidently resolved from the current samples and an accurate
determination of δ15N of indigneous N2 awaits analysis of gas samples
with much lower levels of air contamination. 4. Discussion
The gases extracted from the fluid inclusions in the quartz samples
from the Conqueror deposit in the St Ives Goldfield have much lower The geochemical data reported show striking differences between
CH4/CO2 ratios than the Boomer well gases from the Frog’s Leg gold the Boomer gases and typical biogenic natural gases elsewhere in
camp. This difference partly reflects mixing of gases from different Australia. Several features including the high H2 content (Tables 1 and
generations of CO2-rich and CH4-rich fluid inclusions in complex vein S1) together with high C1/(C2 + C3) ratios (Fig. 2), the positive linear
samples recording multiple generations of growth and fluid entrapment. correlation between H2 and He (Fig. 3), extremely negative δ2H of the H2
However, the methane component separated from the fluid inclusion (Table 2) and the hydrocarbon C-isotope reversal (Fig. 5) suggest that
samples has δ13CCH4 of − 35 to − 39‰ and δ 2HCH4 of − 137 to − 173‰ the Boomer gases have an abiogenic origin. This conclusion is further
(Table 4) that are quite different to the δ13CCH4 values (− 2 to − 21‰) supported by plots of δ13CCH4 versus C1/(C2 + C3), δ2HCH4 and δ13CCO2,
and δ 2HCH4 (− 342 to − 382‰) determined for methane in the Boomer which are commonly used to distinguish natural gases of different origin
gases (Figs. 5 and 6). (Fig. 8a, b and c, respectively). The Boomer gases share similarities with
the reported composition of gases dissolved in aqueous fluids in the

6
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 4. Anderson-Schulz-Flory plot (Schulz, 1999 and references therein) of the Fig. 6. Hydrogen isotopic composition (δ2H ‰) of the individual C1–C5 gaseous
carbon number of the C1–n-C5 normal gaseous hydrocarbons versus the log- hydrocarbons for the Boomer gases. The dashed line is purely pictorial and
normalised proportion of the individual C1–n-C5 components. designed to represent the hydrogen isotopic profile of Australian non-
biodegraded, thermogenic gas. It is based on i) δ2Hmethane being subtracted
from each C1–C5 gas component to give a Δ2H (Boreham and Davies, 2020), ii)
minimum values used for each gas component (e.g. Δ2Hmethane = 0‰), then iii)
a further 260‰ subtracted from each gas component in order to position the
curve on the same scale as the Boomer gases.

4.1. Origin of the Boomer gas components

4.1.1. Hydrogen
Many geological processes can produce natural hydrogen (see
Boreham et al., 2021 and references therein). For low pressure gases
with low flow rates, as is the case for the Boomer gases, there is the
possibility of H2 being generated as an artefact of the drilling operation.
However, H2-rich (Boreham et al., 2021) and CH4 (Simpson, 1912)
natural gases have been reported at other mining operations in the
Eastern Goldfields Superterrane. These independent reports of volumi­
nous H2- or CH4-rich gas released by different drilling operations suggest
the H2 is an indigenous component of well gases in the area and not a
drilling related artefact. The dominant geological sources of H2 in
crustal basement are considered to be 1) radiolytic H2 produced by
radiolysis of water as a result of the radioactive decay of U, Th and K;
and 2) H2 produced during hydration reactions, such as serpentiniza­
tion, as a result of oxidation of ferrous to ferric iron (i.e. 2FeO(rock) +
Fig. 5. Carbon isotopic composition (δ13C ‰) of the individual C1–C5 gaseous H2O → Fe2O3(rock) + H2; 3Fe2+(rock) + 4H2O → Fe3O4(rock) + 6H+ + H2;
hydrocarbons for the Boomer gases. The dashed line is purely pictorial and 3FeO(rock) + H2O → Fe3O4(rock) + H2 (Boyd et al., 1973; Vovk, 1987;
designed to represent the carbon isotopic profile of Australian non- Dubessy et al., 1989; Holland, 2002; Smith et al., 2005; Lin et al., 2005a
biodegraded, thermogenic gas. It is based on i) δ13Cmethane being subtracted and b;Sleep and Bird, 2007; Sherwood Lollar et al., 2014; Bouquet et al.,
from each C1–C5 gas component to give a Δ13C (Boreham and Davies, 2020), ii) 2017; Parnell and Blamey, 2017; Truche et al., 2018; Warr et al., 2019;
minimum values are for each gas component (e.g. Δ13Cmethane = 0‰), then iii) a
Zgonnik, 2020).
further 10‰ subtracted from each gas component in order to position the curve
The relative importance of these two natural abiogenic H2 sources
on the same scale as the Boomer gases.
has been evaluated using noble gases in a number of shield environ­
ments (Warr et al., 2019). This is possible because the noble gases 4He
Precambrian Shields in Canada and South Africa, where the dissolved and 40Ar* have crustal production rates determined by the concentra­
gases comprise up to 50 mol% H2, 30 mol% He and 6 mol% Ar (Warr tions of their radioactive parent elements only. In contrast, the pro­
et al., 2019). In the following Sections 4.1–4.3 we use the isotopic duction of radiolytic H2 is dependent upon the concentrations of U, Th
composition of H2 and the relative abundances of H2, 4He and 40Ar* and K and the amount of water present, which is controlled by crustal
(radiogenic 40Ar) to further evaluate the production pathways of the porosity (Eqs. 1–7).
abiogenic H2, CH4 and other gaseous components in the Boomer gases of )
the Frog’s Leg gold camp, Yilgarn Craton. The Boomer gases, which 4
He atoms kg− 1 yr− 1 = 3.2422 × 109 [U ppm] + 7.71 × 108 [Th ppm] (1)
represent modern fluids, are compared to the CH4-rich gold-related fluid
inclusions from the Conqueror deposit, St Ives Goldfield, which repre­ 40
Ar atoms kg− 1 yr− 1
= 1.022 × 105 [K ppm] (2)
sent Neoarchean-sourced fluids.

7
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

[ ]
ptH2 molecules kg− 1 yr− 1 = 0.0096 × ƩDα eV kg− 1 yr− 1 with 0.5 vol% porosity are still within the range reported for the Boomer
[ ] [
+ 0.006 × ƩDβ eV kg− 1 yr− 1 + 0.004 × ƩDλ eV kg− 1 yr− 1
]
(3) gases (19–75), but the uncertainty allows the possibility that a part of
the H2 in the Boomer gases was generated by serpentinization reactions.
[ ( / ) ] If a significant portion of the the Boomer gases migrated from deep
ƩDi eV kg− 1 yr− 1 = ƩEi eV kg− 1 yr− 1 × porosity × ρfluid ρrock × Si
/[ ( / ) ] sources with porosities as low as ~0.2 vol% (depth ~ 10 km), serpen­
pt 1 + porosity × ρfluid ρrock × Si (4) tinization at the depth of H-isotopic equilibration (i.e. 1.4–2.7 km; see
Section 4.2) would be implicated as a significant H2 source. Further­
ƩEα eV kg− 1 yr− 1
= 3.81 × 1014 [Th ppm] + 1.36 × 1015 [U ppm] (5) more, if CH4 was produced by reaction of H2 with CO2 (below), most of
the CH4 in the Boomer gases must represent H2 produced by serpenti­
ƩEβ eVkg− 1 yr− 1 =4.88×1015 [Kwt.%]+1.69×1014 [Thppm]+9.11×1014 [Uppm] nization reactions. This follows from the uniformly higher (H2 + 2 ×
(6) CH4)/He ratios (72–157; Table S1) of the Boomer gases compared to
crustal H2/4He production ratios (Tables 5 and S3) wherein CH4 was
ƩEλ eVkg− 1 yr− 1 =1.52×1015 [Kwt.%]+3.00×1014 [Thppm]+7.05×1014 [Uppm] likely produced by the reaction of two H2 molecules with a C species (e.
(7) g. Warr et al., 2019).
Eqs. 1 and 2 are empirically derived approximations for modern Since serpentinization is a significant reaction pathway to natural
production rates (see Ballentine and Burnard, 2002). The derivation of H2, a brief summary of the reaction mechanism follows. Applied to
Eqs. 3–7 is described in Tarnas et al. (2018, 2020) building on the work mafic–ultramafic rocks, serpentinization involves the hydration of MgFe
of Hofmann (1992) and Adamiec and Aitken (1998). The equations are olivine and MgFe pyroxenes (Fe2+) with the production of MgFe
simplified here by combining multiple constants in common units. Note serpentine, MgFe brucite, magnetite and H2 (Fe3+ residing mainly in
that E = energy released by radioactive decay; D = dose received by pore magnetite; Evans et al., 2013) and is considered the main process
waters; porosity is a fraction, ρ is density (units cancel out: ultramafic
3000 kg m− 3; granite 2600 kg m− 3; fluid 1000 kg m− 3) and Si is the
stopping power (dependent on the retarding force acting on the charged
particles due to the interaction with matter) where i denotes α, β or λ
particles and Sα = 1.5, Sβ = 1.25 and Sλ = 1.14 (Adamiec and Aitken,
1998; Tarnas et al., 2018).
Rates of 4He, 40Ar* and radiolytic H2 production estimated with eqs.
1–7, and corresponding 4He/40Ar* and H2/4He production ratios are
given for ultramafic and granitic lithologies in the Eastern Goldfields
Superterrane in Table 5 (calculated using Table S3). The production rate
of radiolytic H2 from ultramafic rocks is ~18 (mean) to ~24 (median)
times lower than that from granitic rocks because of the much lower
radioelement concentrations in ultramafic lithologies (Table 5). The
ultramafic and granitic lithologies in the Eastern Goldfields Super­
terrane have similar mean (U + Th)/K ratios of 5.3–9.4*10− 4 that give
similar expected mean 4He/40Ar* production ratios of ~8–11, which are
significantly lower than the Boomer gases with average 4He/40Ar* of 28
± 10 (Table 5). The enrichment of the Boomer gases in 4He relative to
40
Ar* can be explained because 4He is released from most crustal min­
erals at much lower temperatures than 40Ar* and as a result low tem­
perature (<200 ◦ C) groundwaters typically have elevated 4He/40Ar*
greater than the production ratio (e.g. Ballentine et al., 2002).
In contrast with the 4He/40Ar* ratio, the H2/4He ratio of the gases is
more likely to reflect the crustal production ratio. This is because 4He is
released to fluids by alpha recoil at low temperature (Ballentine et al.,
2002) and radiolytic H2 will be produced within the pore fluids. The
calculated H2/4He production ratio of the crust is fairly insensitive to
uncertainty in U or Th concentration, which influences production of
both H2 and 4He. However, it is highly sensitive to porosity, which
controls H2 production through limiting the water present. Based on
porosities of 0.9–1.2 volume percentage (vol%) estimated using the
generalised relationship between porosity and depth for basement rocks
formulised by Sherwood Lollar et al. (2014), eqs.1 and 3 predict that
ultramafic or granitic rocks at depths of 1.4–2.7 km (see Section 4.4)
would have overlapping H2/He production ratios of 44–76 (Table 5). In
comparison, the Boomer gases have a similar range of measured H2/He
ratios of 19–75, with an average of 37 ± 18 (Tables 1 and 5). This can be
interpreted as evidence that radiolytic H2 production accounts for the
majority of the H2 in Boomer gases. However, it is important to note that Fig. 7. Plots showing the relative abundances of radiogenic 4He and atmo­
spheric noble gases in gases from Boomer exploration drill holes: a) noble gas
this inference assumes that the gases were generated at depths of
fractionation values (F-values) relative to air (F of air = 1), showing the gases
~1.4–2.7 km, which is equivalent to the depth of hydrogen isotopic
are strongly enriched in 4He relative to air-saturated water or air and all the
equilibration between H2 and CH4 (Sections 4.2 and 4.4). If H2 and 4He samples have non-radiogenic noble gas isotope abundance ratios intermediate
originated from deeper regions of the crust, which are dominated by of air and air-saturated water, and b) a plot of 84Kr/36Ar versus 132Xe/36Ar
granite with lower porosities, the calculated H2/4He production ratios showing the Boomer gases together with air and various air-saturated waters.
would be lower (Table S3). The H2/4He production ratios of a granite Samples EVDD0071 and EVDD0075 plot closer to 25 ◦ C air-saturated waters
than air. ASW = air saturated water; SW = seawater.

8
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Table 5
Production of 4He, 40Ar* and radiolytic H2 in ultramafics and granites of the Eastern Goldfield Superterrain and values measured in the Boomer gases.
Concentrationsa Porosity Production Ratesb Prod. Ratios
4 40 4
K2O Th U He Ar* H2 He/40Ar* H2/4He
− 1 − 1
wt% ppm ppm % Atoms or molecules kg yr

Ultramafic
Mean 0.36 1.3 0.43 0.9 – 1.2 2.4 × 109 3.1 × 108 1.1 – 1.4 × 1011 7.8 45.3 – 60.3
Median 0.12 0.6 0.14 0.9 – 1.2 9.0 × 108 1.0 × 108 0.40 – 0.53 × 1011 8.8 43.8 – 58.4
Granitic
Mean 3.2 20.4 4.5 0.9 – 1.2 3.0 × 1010 2.7 × 109 1.4 – 1.9 × 1012 11.2 47.1 – 62.7
Median 3.1 11 2.5 0.9 – 1.2 1.7 × 1010 2.6 × 109 0.95 – 1.3 × 1012 6.3 57.2 – 76.2
Boomer gas (Average ± 1 s.d.) 28 ± 10 37 ± 18
a
Mean and median concentrations of K2O, Th and U downloaded from https://services.ga.gov.au/gis/inorganic-geochemistry/wms?request=GetFeature&ser
vice=WFS&version=1.1.0&outputFormat=csv&typeName=geochemistry:Samples. Production rates of radiogenic 4He and 40Ar* depend on the concentration of
the radioelements only. Production of radiolytic H2 depends on the concentrations of radioelements and water, which is controlled by porosity (see text and eqs. 1–7).
The range of production ratios given corresponds to the range of porosity.
b
Calculated using the approach of Tarnas et al., 2018, 2020. Note: using updated dose-rates (Gy yr− 1) in Guérin et al. (2011) has only a small effect of − 1.5% to 8.6%
difference on H2 production rates.

leading to abiogenic H2 (Evans et al., 2013; McCollom, 2013; Schrenk in rock-dominated environments where oxidising meteoric water
et al., 2013; Worman et al., 2016; Etiope and Whiticar, 2019; Hao et al., recharge is restricted (Abrajano et al., 1990) and low silica activity
2020). Serpentinization can occur at temperatures from just above (Katayama et al., 2010). The discharged aqueous fluids change from
ambient to 330–400 ◦ C, depending on pressure (McCollom and Seewald, highly alkaline to moderately acidic with increasing temperature
2013; Evans et al., 2013; Mayhew et al., 2013). Low temperature (<100 (McCollom and Seewald, 2013; Schrenk et al., 2013; Murray et al.,
◦
C) H2 production is likely associated with a spinel reactant (e.g. 2020), with high temperature fluids having highly elevated H2 and CH4
M2+(M3+)2O4; magnetite Fe3O4 and chromite FeCr2O4) and a cubic concentrations (Schrenk et al., 2013). Serpentinization of ultramafic
crystal structure. The spinel can be either as an impurity in the initial rocks generate higher concentration of H2 than basalt–water reactions
mineral or the serpentinization product (Mayhew et al., 2013). More­ (Hao et al., 2020). The H2 yield is directly proportional to the Fe/Mg
over, in laboratory experiments zinc and nickel oxides show catalytic ratio in the reactant minerals (McCollom and Seewald, 2013). In gran­
effects (Mayhew et al., 2013; Martinez et al., 2019). Critical factors for ites, oxidation of Fe2+ in biotite can also produce H2 (Murray et al.,
serpentinization are environments with highly reducing conditions i.e. 2020). However, the lower proportion of ferrous minerals in felsic rocks
below the low O2 fugacity/high H2 activity associated with the FMQ compared to basic rocks makes granites less effective sources of H2 than
buffer (Shock et al., 2019 and reference therein), likely to be maintained serpentinization of mafic–ultramafic rocks.

Fig. 9. Hydrogen isotopic range for H2 reported in different depositional environments and rock types (modified after Hao et al., 2020 and including data from
Thiessen et al., 2009 for ‘CO2 corrosion’) together with the δ2H range of H2 in the Boomer natural gases. δ2H range of Australian topped oil from Schimmelmann et al.
(2004) and of methane from gas data used in Boreham and Davies (2020).

9
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

The Boomer gases have unusually negative δ2HH2 values (− 760‰ < the methane and minor wet gas products are thermodynamically fav­
δ2HH2 < − 781‰; Table 2) that are within the range reported for H2 oured (Milesi et al., 2016). However, the kinetic inhibition to methane
produced by serpentinization reactions (Fig. 9; Hao et al., 2020 and (and wet gas; see Section 4.1.3) generation (Milesi et al., 2016 and
references therein). Although taken alone the unusually low δ2HH2 is not references therein) necessitates catalysis by transition metals (Fe, Ni, Co,
diagnostic of serpentinization because radiolytic H2 with initially higher Cr with contributions from Rh and Ru) (McCollom, 2013 and references
δ2HH2 could acquire similarly 2H-depleted values by re-equilibration therein; Zhang et al., 2013; Etiope and Ionescu, 2015; Etiope et al., 2017;
with low temperature groundwater, meaning it is also consistent with Etiope and Whiticar, 2019). However, Rh is not considered geologically
a mixed origin of H2 from radiolysis and serpentinization. Alternative important due to it trace concentration (Etiope and Ionescu, 2015).
microbial sources are considered unlikely because microbial oxidative Microporous hydrous titanium, niobium and zirconium silicalites and
H2 generation is usually accompanied by H2-consuming microbes (i.e. to zeolites are also considered catalytic (Nivin, 2009). Relatively low CO2
maintain H2 concentrations and ppm levels; Gregory et al., 2019 and reactant/CH4 product ratios are associated with laboratory FTT re­
references therein) and there is little isotopic evidence for methano­ actions using awaruite (Fe-Ni alloy) (Horita and Berndt, 1999), Ni sul­
genesis (see Sections 4.1.2 and 4.4). phides (Lazar et al., 2012) and Co-bearing magnetite (Ji et al., 2008). Ru
catalysts supported on TiO2 and Al2O3 (analogue for Ru-chromite)
4.1.2. Methane and carbon dioxide promotes reaction at very low temperatures of <100 ◦ C (Etiope and
The enrichment of CH4 in 13C over a relatively wide range of δ13C Ionescu, 2015), likely via a mechanism involving co-absorbed gases and
(− 20.30‰ to − 2.42‰) while maintaining a rather constant hydrogen a Ru-C intermediate (Thampi et al., 1987).
isotopic composition (δ2HCH4 = − 369.4 ± 12.1‰) reflects the redox FTT reaction associated with the Boomer gases would need to occur
conditions and the initial carbon (biogenic and/or inorganic) and at low temperature (Section 4.2) where the initial δ13CCO2 will be no
hydrogen (H2 and/or water) sources, as well as post-generative pro­ more enriched in 13C than the product δ13CCH4 (− 2.42‰; Table 2). In the
cesses (e.g. equilibration and mixing – see Sections 4.2 and 4.4 for laboratory, there is a strong kinetic isotope effect between CO2 and CH4
further discussion). on Ru-supported Sabatier reaction at temperatures <100 ◦ C. There, the
The carbon and hydrogen isotopes of methane in the Boomer gases initially generated methane is highly depleted in 13C becoming pro­
are clearly placed within the abiogenic 13C-2H isotope field in Fig. 8b. In gressively enriched in 13C with continuing reaction in a closed system,
contrast, methane from the Conqueror fluid inclusion plots toward the resulting in the lowest carbon isotope difference of ∆13CCH4-CO2 = − 36‰
edge of the abiogenic field overlapping typical thermogenic source obtained at 90 ◦ C after 30 days (Etiope and Ionescu, 2015). Near com­
signatures (Table 4; Fig. 8b and Fig. 8c; see also Section 4.3). This plete reduction of CO2 is also proposed to explain the 13C-enriched
pronounced difference might indicate CH4 in Neoarchean fluid in­ abiogenic methane found in the Lost City hydrothermal field (Proskur­
clusions has a distinctly different origin to CH4 in the modern Boomer owski et al., 2008). In the Boomer gases, a similar Rayleigh fractionation
gases (Section 4.4). However, the original δ13CCH4 of the Boomer gases acting on a limiting CO2 source (i.e. slow release from carbonate; see
and fluid inclusions have also been influenced by isotopic equilibration below) would result in 13C enriched CH4. For the hydrogen isotopes,
at different temperatures (see Section 4.2). Abiogenic and/or microbial Etiope and Ionescu (2015) observed methane initially depleted in 2H
methane oxidation could again be considered (as it has for the CH4–CO2 (∆2HCH4-H2 ~ 240‰ at 90 ◦ C after 7 days), and the authors expected
carbon isotope couplet, but rejected below), but then the residual methane to become increasingly enriched in 2H as the reaction proceeds
methane should likely become highly enriched in 2H with increasing and the isotopic fractionation to decrease to near zero with increasing
oxidation (Miller et al., 2018), which is not observed with δ2HCH4 temperature, in accord with an aqueous FTT reaction at around 250 ◦ C
remaining relatively constant. McCollom et al., 2010). The large ∆2HCH4-H2 (~400‰) in the Boomer
It is not appropriate for an a priori assumption of carbon isotopic gases is consistent with reaction temperatures <90 ◦ C (see Section 4.2).
equilibrium between CH4 and CO2. However, a range of equilibrium For CO2, its rate of release from the carbonate is similar to its con­
temperatures may explain the variability in the δ13CCH4 in the Boomer sumption by hydrogenation and where the released CO2 has a similar
gases, as was suggested to explain the 13C-depleted methane in fluid δ13C to the associated carbonate (Ohmoto and Rye, 1979). This would
inclusion gases from the Junction gold deposit (Fig. 1) (Polito, 1999). allow product CH4 gas to be free to migrate and to contain excess H2
Nevertheless, the lack of a correlation between δ13CCH4 and 13CCO2 when carbon is limiting (Vacquand et al., 2018). Another possibility is
(Fig. 8c) or an accompanying wide range in δ13CCO2 in the Boomer gases methane generation from a 13C-enriched graphite source (i.e. C + 2H2 =
voids this mechanism. By the same reasoning, a closed system under CH4) associated with past hydrothermal fluids, which can show high
kinetic control with a Rayleigh 13C-isotopic fractionation between CH4 enrichment in 13C for fluid-deposited graphite (Ray, 2009; Luque et al.,
and CO2 (single source) cannot be viable for the Boomer gases. Further 2012). Initially 13C-enriched graphite is also possible from the direct
excluded is oxidation of methane to CO2 by abiotic and microbial action carbonate methanation, CaCO3 + 4H2 → Ca(OH)2 + CH4 + H2O, with a
(Kiyosu and Krouse, 1989), which should show a strong correlation carbon saturated CH4-rich fluid precipitating graphite (Vitale Brovarone
between δ13CCH4 and δ13CCO2, which again is not observed. Addition­ et al., 2017). Most reactions involving graphite necessitate temperature
ally, microbial methanogensis alone does not lead to high enrichments > 200 ◦ C and up to 1000 ◦ C (Milesi et al., 2015; Etiope and Whiticar,
in 13C in methane. Miller et al. (2018) used an isolated culture of 2019), where the lower temperatures are favoured by disordered
hydrogenotrophic Methanobacterium sp. from hyperalkaline subsurface graphite and chemical impurities of heteroatoms (NSO) and unsatura­
fluids in the Samail ophiolite to show that the most 13C-enriched tions (Beyssac and Rumble, 2014). Although Vacquand et al. (2018 and
methane was still 28‰ lighter compared to the source carbonate. Using reference therein) indicate reactions can occur at much lower temper­
the δ13CCH4 of the Boomer gases this would translate to +8.4 < atures, the calculated cool formation temperatures based on isotopic
δ13Ccarbonate < +26.4. However, only one carbonate from the Eastern equilibrium (see Section 4.2) make graphite an unlikely carbon source
Goldfields Superterrane was found to have δ13Ccarbonate at the lower end for the Boomer gases.
of this carbon isotopic range, for which a subsequent re-analysis could The underlying reasons for the abiogenic isotopic demarcation still
not reproduce the high 13C enrichment (Walshe et al., 2006). The above remain unclear with crustal versus mantle feedstock and fractionation
evidence supports methane and carbon dioxide in the Boomer gases between CO2 and CH4 are deemed important (Etiope and Sherwood,
having no genetic relationship. 2013). In a multi-element study of reduced gas seeps occurring in ul­
The most likely mechanism for methane formation is from a C-H-O trabasic to basic rocks outcropping in terrestrial ophiolitic complexes
source consumed in a FTT reaction, specifically CO2(g) + 4H2(g) ↔ CH4 across the globe, Vacquand et al. (2018) grouped terrestrial abiogenic
(g) + 2H2O (Sabatier reaction). In the serpentinization of mafic- gases into four types: H2-rich, H2-CH4, N2-H2-CH4 and N2-rich based on
ultramafic rocks (e.g. peridotite) coupled with the Sabatier reaction, water chemistry, gas composition and C, H and noble gas isotopes (all

10
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 8. Plot of a) ratio of mol% methane / mol% (ethane + propane) versus the
carbon isotopic composition (δ13C ‰) of methane, modified after Milkov and
Etiope (2018). Boomer gases in open circles; b) δ13C versus δ2H of CH4 modified
after Milkov and Etiope (2018). Boomer gases in open circles. Fluid inclusion
gases from the Conqueror gold deposit (Kendrick et al., 2011) in filled circles.
Abiogenic methane from Vacquand et al. (2018): high-H2 (filled diamond), H2-
CH4 (filled triangle) and H2-CH4-N2 (filled circle). The crosses represent data in
Milkov and Etiope (2018) used to define the abiogenic methane genetic field for
δ13C-CH4 > − 20‰ and δ2H-CH4 < − 200‰ with data from Boschetti et al.
(2013), Botz et al. (1996), D’Alessandro et al. (2018a), D’Alessandro et al.
(2018b), Etiope et al. (2013), Kelley and Früh-Green (1999), Kietäväinen et al.
(2017), Marques et al. (2018), Miller et al. (2016) and Zeng et al. (2013); and c)
carbon isotopic composition (δ13C ‰) of methane versus the carbon isotopic
composition (δ13C ‰) of carbon dioxide, modified after Milkov and Etiope
(2018). CR - CO2 reduction, F - methyl-type fermentation, EMT - early mature
thermogenic gas, OA- oil-associated gas, LMT- late mature thermogenic gas.
Boomer gases in open circles. Fluid inclusion gases from the Conqueror Au-
deposit (Kendrick et al., 2011) in filled circles. (For interpretation of the ref­
erences to colour in this figure legend, the reader is referred to the web version
of this article.)

except for the N2-rich type are plotted in Fig. 8b). Abiogenic methane
was considered ‘primary’ (H inherited directly from H2O) when formed
directly during serpentinization in the presence of carbon and is ther­
modynamically favoured (Milesi et al., 2016) and under C-limitations
(CO2 and/or graphite) 13C-enriched methane is generated. Abiogenic
methane was considered ‘secondary’ when H is inherited from H2. The
high-H2 type abiogenic gases are considered ‘primary’ with δ2HCH4 >
− 200‰, while the H2-rich and N2-H2-CH4 abiogenic gases have δ2HCH4
< − 200‰ and are considered ‘secondary’ (Vacquand et al., 2018).
Boomer methane has ‘heavy’ δ13CCH4 > − 20‰ and ‘light’ δ2HCH4 <
− 200‰, thus sharing a similar isotopic signature with those other
abiogenic gases within the same dual isotopic range (Fig. 8b). Hence,
Boomer methane is considered ‘secondary’ within the Vacquand
schema, although 2H-depletion will follow with H2–CH4 thermodynamic
(H-isotopic) equilibrium at low temperatures (see Section 4.2). Never­
theless, caution should still be applied with this genetic classification
since the H2-CH4 type gases detailed in Vacquand et al. (2018) were
generated from deeper, hotter sources and the absence of an aqueous
transport phase may preclude further chemical and isotopic equilibrium
during gas phase migration to the surface (Vacquand et al., 2018). A
direct primary (H2O) serpentinization source for methane has been
evoked previously (Oze and Sharma, 2005; Suda et al., 2014) via an
unknown mechanism, while Etiope and Ionescu (2015) support sec­
ondary methane with H2 from serpentinization (peridototes) migrating
to an undersaturated rock (chromitites) involved in the Sabatier reac­
tion. Another variant is methanation at the site of serpentinization
involving spinel impurities within the initial serpentinizing rock or
reduction of a migrating CO2 and local H2 catalysed by the magnetite
product at ≤100 ◦ C in the laboratory (Mayhew et al., 2013). Addition­
ally, methane generation via retrograde metamorphism of a graphite
bearing rock (i.e. talc + calcite + C + water → dolomite + quartz + CH4)
is unlikely since CO2 is limited and high reaction temperatures are
required (Holloway, 1984). However, based on the analysis of Etiope
and Whiticar (2019) the distinction between primary and secondary
abiogenic gases is arbitrary given it is unlikely that there is a direct 1-
step conversion of an oxidised carbon source to methane in the ser­
pentinization reaction. Moreover, the ability of a FTT reaction to
generate hydrocarbons under geologically relevant conditions still needs
further laboratory verification (Barbier et al., 2020).

4.1.3. Wet gases

The Boomer gases are classified as wet gases since the %C1/(C1–C5) is
<97 mol% (Tissot and Welte, 1984). The C1/(C2 + C3) ratio in the
(caption on next column)
Boomer gases varies from 23 to 67, typical of abiogenic gas (Etiope and
Ionescu, 2015 and references therein). In aqueous FTT experiments
designed to mimic geological conditions (i.e. involving iron, nickel,

11
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

chromium or cobalt catalysts at 200 ◦ C) methane only was generated then becomes enriched in 13C as does δ13Cn-C4 and δ13Cn-C5. Selective
(McCollom, 2013 and references therein). More recently, Etiope and mild biodegradation of the C3+ hydrocarbon gases can show a similar
Ionescu (2015) found C2–C5 wet gases were generated from a gas phase trend in the C1–C5 linear alkane gases (Boreham and Edwards, 2008 and
Sabatier reaction with Ru catalysts at <100 ◦ C with C1/(C2 + C3) references therein) but microbial alteration is not supported here since
generally <100 and suggested that Ru-enriched chromites, specifically the corresponding H-isotopes, which should show even larger enrich­
platinum group elements-enriched veins in altered chromites, are good ments in 2H (Boreham et al., 2008) is not observed. Such partial carbon
source rocks for liberating abiogenic hydrocarbons in land-based ser­ isotope reversal is common in thermogenic, abiogenic, experimental
pentinization. The Boomer gases show a weak positive linear correlation FTT, and radiolysis studies (Boreham and Davies, 2020 and references
(r2 = 0.39) between δ13CCH4 and the proportion of C1/(C2 + C3), therein) and may reflect a preferential approach to C-isotopic equilib­
although the most 13C-enriched methane in EVDD0083 is also the driest rium (Thiagarajan et al., 2020a). Therefore, the observation of carbon
(Fig. 8a) and has the lowest air contamination (Table 1). The ASF plot of isotope reversals, in isolation, is a relatively insensitive source differ­
the C2–n-C5 gaseous hydrocarbons (Fig. 4) shows a strong linear corre­ entiation tool. The Boomer gases show, on average, an increasing
lation (r2 = 0.994–0.998) becoming weaker with the inclusion of C1 (r2 enrichment in 2H with increasing linear alkane carbon number (Fig. 6).
= 0.943–0.966). The combination of a near perfect linear negative slope This normal H-isotope profile is common amongst abiogenic seep gases
and its steepness is considered a characteristic of abiogenic gas (Nivin, (Boreham and Davies, 2020 and references therein) and points to a chain
2019 and references therein). Proskurowski et al. (2008) and Nivin propagation mechanism where the preference for 12C-1H over 12C-2H
(2009) also observed a similar correlation where the methane position bond cleavage leads to increasing 1H depletion at each polymerisation
on an ASF plot is offset relative to the straight-chained wet gases in the step (Sherwood Lollar et al., 2002) as well as approaching hydrogen
Lost City Hydrothermal field and the Lovozerskii alkaline massif, isotopic equilibrium (Reeves et al., 2012). The rate of hydrogen isotopic
respectively. This may indicate that there is another methane source not equilibrium between H2O and the C2–C5 gaseous hydrocarbons is
associated with wet gases (see Section 4.4 for further discussion). The considered faster than with CH4 and with C3+ alkanes faster than ethane
similarity in the carbon isotopes of the C2–C5 (Fig. 5) and hydrogen (Reeves et al., 2012). The hydrogen isotopic fractionation αCx/H2O =
isotopes of the C2–C4 gaseous hydrocarbons (Fig. 6) indicate a common (δ2HCx + 1000) / (δ2HH2O + 1000) where x = 2–4 are rather constant for
source for these wet gases. The trace amounts of unsaturated hydro­ the Boomer gases with an average αCx/H2O of 0.65, 0.68 and 0.68 for
carbons found in some Boomer gases support a radiolytic contribution ethane, propane and n-butane, respectively, using δ2HH2O = − 25‰ (i.e.
(Boreham and Davies, 2020 and references therein) and may represent midpoint ±5‰ of low salinity Yilgarn groundwaters; Turner et al.,
remnants of facile catalytic hydrogenation of intermediates (Shock et al., 1993). Although, these are much larger fractionations than predicted by
2019), perhaps even catalysed by gold (Fujitani et al., 2009). Alterna­ theory, experimental determined H-isotopic fractionations are also
tively, oxidative coupling of methane (OCM; Lehmann and Baerns, larger than those from theory (Reeves et al., 2012), suggesting that
1992; Sinev, 2003 and references therein) is a possible pathway to the under these conditions, the C2–C4 are approaching H-isotopic equilib­
unsaturated hydrocarbons, as well as meteoric fluid contaminant from rium (Thiagarajan et al., 2020a). For iso-butane, the smaller H-isotopic
soil gas oxidation of longer chained hydrocarbons (Gerdenich and fractionation between gas and water (i.e. larger α; average αi-C4/H2O
Hanel, 1986). Currently, the source of the trace amounts of unsaturated 0.75) is in accord with a closer approach to H-isotopic equilibrium,
hydrocarbons is unknown. driven by weaker a C–H bond at the secondary carbon atom (Xiao, 2001
Chain polymerisations reactions can produce wet gases from a and references therein) and potentially a fast rate of H-exchange in
methane and/or CO2 precursor associated with reactions involving FTT branched versus linear alkanes. Furthermore, the trace C2 and C3 un­
(Etiope and Sherwood, 2013 and references therein). Further, mobile saturated hydrocarbons in the Boomer gases should help to accelerate H-
methyl radical intermediates can combine during methane radiolysis isotopic exchange (Reeves et al., 2012).
(Boreham and Davies, 2020 and references therein) and/or thermo­ To better visualise the variation of C- and H- isotopic profiles of the
chemical OCM involving alkali metal and lanthanide oxides commonly methane, ethane and propane, Figs. 10, 11 and 12 show the relationship
associated with carbonates at high (experimental) temperatures (>650 between the isotopic differences between δ13CCH4 and δ13CC2 versus
◦
C; Lunsford, 1990; Sinev, 2003 and references therein). OCM can δ13CC2 and δ13CC3, δ2HCH4 and δ2HC2 versus δ2HC2 and δ2HC3, and
potentially utilise O2 from water radiolysis and also as an independent δ13CCH4 and δ13CC2 versus δ2HCH4 and δ2HC2, respectively. The Boomer
source of high temperature H2 (Lehmann and Baerns, 1992). gases are plotted in each figure together with genetic fields that define
Where measured, polymerisation reactions impart carbon and the range in Australian non-biodegraded, thermogenic gases, natural
hydrogen isotopic fractionations at each chain extension step (Boreham gases with complete isotope reversals in sedimentary basins, abiogenic
and Davies, 2020 and references therein) and for FTT δ13C and δ2H gases associated with geological seeps (serpentinization, hydrothermal,
depend on the precursor, catalyst, extent of reaction, chemical envi­ peralkaline systems) and experimental FFT synthesis; for the individual
ronment and reaction mechanism (Etiope and Sherwood, 2013; Zhang studies see Boreham and Davies (2020), and references therein. The
et al., 2013; Boreham and Davies, 2020). FTT reactions have been the Boomer gases show only carbon isotope reversal between 13C-enriched
most extensively studied (Anderson, 1984) and are controlled by kinetic methane and 13C-depleted ethane, which is not a diagnostic feature as it
reactions resulting in non-equilibrium distributions of molecular spe­ overlaps with other thermogenic and abiogenic gases (Fig. 10).
cies, structures of the carbon chains and likely carbon number distri­ Although published data is limited for hydrogen isotopes, there is
bution (Anderson et al., 1951). Mechanistically, amongst a variety of considerable overlap between thermogenic and abiogenic gas sources
possibilities for FTT synthesis (Anderson, 1984 Table 1 p176) C–C bond with the Boomer gases, which show a normal hydrogen isotope profile
formation is via alkyl migration to a surface methylene both bound to with increasing enrichment in 2H with increasing carbon number (i.e. no
the metal catalyst (Bertoncini et al., 2009), as well as chain growth by isotope reversal) (Fig. 6 and Fig. 11). On the other hand, the dual isotope
olefin insertion (methyl-substituted species from the insertion of pro­ plot in Fig. 12 shows discrimination of Australian non-biodegraded,
pylene into the growing chain; Anderson, 1984 p. 190; Schulz et al., thermogenic natural gases from experimental FTT and seeps. Interest­
1994) and where the polymerization (chain growth/propagation) pro­ ingly, the C1–C2 C- and H-isotopic differences of the Boomer gases have
cess is favoured over termination via H transfer for paraffin elimination the closest values to those observed in the radiolysis of methane
(Schulz, 1999 and references therein). In the sub-surface, gas-solid FTT (Boreham and Davies, 2020).
reactions are favoured (Etiope et al., 2017).
The Boomer gases show partial carbon isotope reversals with ethane 4.2. Isotopic equilibrium and the CH4–H2 geothermometer
the only homologue that is highly depleted in 13C compared to the
preceding carbon number (i.e. δ13CC2 < δ13CCH4) (Fig. 5). The δ13CC3 It is certain that the observed δ2HH2 and δ2HCH4 do not reflect a

12
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 10. Plot of the carbon isotopic difference (Δ13C

‰) between methane and ethane versus the carbon
isotopic difference (Δ13C ‰) between ethane and
propane. The genetic fields for the FTT, Australian
thermogenic gas and abiogenic seeps encompass the
data presented in Boreham and Davies (2020 and
references therein). The curved line represents the
experimental data from Zhang et al. (2013) for FTT
reaction between CO2 and H2 using Fe/Ni metal ion
catalyst at 400 ◦ C from 2 h to 60 h in a closed system
and from 700 ◦ C → 250 ◦ C.

primary source–product relationship involving a FTT reaction wherein Hagemann and Cassidy, 2000). Nevertheless, to obtain C-H-O H-isotopic
the product is expected to be similar or depleted in 2H compared to the equilibrium at such relatively low temperatures would require an
reactant (Etiope and Ionescu, 2015). In the Boomer gases, H-isotopic effective catalyst (Lyon and Hulston, 1984; Giggenbach, 1997; Fiebig
exchange between CH4, H2 and H2O is more likely. Therefore, further et al., 2007) and is in accord with experimental data and theoretical
insights into the source and mechanism of formation within the C-H-O models (Kietäväinen et al., 2017 and references therein).
system is achieved using the isotopic fractionation factors associated Estimated temperatures based on the thermodynamic equilibrium
with thermodynamic equilibrium to derive an equilibrium temperature between C1–C3 gaseous hydrocarbons are >200 ◦ C and > 400 ◦ C using
i.e. Horibe and Craig (1995) for CH4–H2–H2O and in Bottinga (1969) for the C1/C2 ratio (Darling, 1998) and a combination of C1/C2 and C2/C3
CH4–CO2. The rate of isotopic exchange decreases in the order H2–H2O (Tassi et al., 2007) ratios, respectively. Hence, thermodynamic chemical
(hydrogen) > CH4–H2 (hydrogen) > CH4–H2O (hydrogen) > CH4–CO2 equilibrium was not attained, indicating kinetic processes predomi­
(carbon) (Hulson, 1977; Suda et al., 2014). nately control the synthesis of Boomer gaseous hydrocarbons together
For the CH4–H2 system at hydrogen isotopic equilibrium, the with the mixing of different hydrocarbon sources (see Section 4.4).
hydrogen isotopic difference between CH4 and H2 requires a calculated Boomer gases have − 15.22‰ < δ13CCO2 < − 11.30‰, consistent with
equilibrium temperature of 42.2 ◦ C for EVDD0093 with the largest a mixed inorganic, atmospheric and organic source. The end-member
isotopic difference (Δ2HCH4-H2 = 418.1‰) and 57.0 ◦ C for EVDD0083 organic CO2 source is likely to be soil gas with δ13CCO2 = − 22‰
with the smallest isotopic difference (Δ2HCH4-H2 = 378.4‰). Note (Turner et al., 1993). For a CH4–CO2 geothermometer under thermo­
EVDD0073 is an outlier and has been excluded from further analysis. dynamic equilibrium control, the premise is that C-isotopic exchange is
The δ2HH2O of water in the system is not known. Nevertheless, using the slower than H-isotopic exchange (Lyon and Hulston, 1984) and CH4 is
lowest and highest equilibration temperatures of 42 and 57 ◦ C from the depleted in 13C compared CO2 (Bottinga, 1969). The latter condition is
CH4–H2 system, together with the δ2HH2 of the Boomer gases, the not generally met with the Boomer gases (Table 2 and Fig. 8c) so
calculated δ2HH2O range from − 39‰ to − 68‰ at 42.2 ◦ C and − 6‰ to calculation of CH4–CO2 equilibrium temperatures become unrealistic.
+29‰ at 57.0 ◦ C. This wide range in calculated δ2HH2O overlaps with However, the fluid inclusion gases show δ13CCH4 values that are
the δ2H of local groundwaters, which vary from about − 39 up to +19‰ considerably more negative than the δ13CCO2 values in the Junction
(Turner et al., 1993) and also overlaps fluid inclusion waters that record (Polito, 1999) and Conqueror (Fig. 1; Table 4) gold deposits. In the
δ2HH2O ranging from about − 10‰ to − 102‰ depending on the source Junction deposit, isotopic re-equilibration between CH4 and CO2 inside
of the fluid and equilibration temperature (e.g. Ojala et al., 1995; the fluid inclusions occurred between 90 ◦ C and 220 ◦ C, well below the

13
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 11. Plot of the hydrogen isotopic difference (Δ2H ‰) between methane and ethane versus the hydrogen isotopic difference (Δ2H ‰) between of ethane and
propane, modified after Boreham and Davies (2020). The genetic fields for the FTT, Australian thermogenic gas and abiogenic seeps encompass the data presented in
Boreham and Davies (2020 and references therein).

340–385 ◦ C entrapment temperatures (Polito, 1999). At Conqueror, fluid inclusions and the loss of H2 could lead to oxidation of CH4 and
calculated isotopic equilibration temperature of 157 ◦ C and 250 ◦ C other components in the fluid inclusion (e.g. Mavrogenes and Bodnar,
(Table 4) are similar to or lower than the fluid inclusion homogenisation 1994).
temperatures obtained from microthermometry (Kendrick et al., 2011). CH4 has a high abundance in fluid inclusions from the Junction and
This poses the question, in general, whether the carbon isotopes of Conqueror gold deposits in the Eastern Goldfields Superterrane. Polito
methane in fluid inclusions or even free gas re-equilibrate to varying et al. (2001) documented CH4-rich (>10 mol% CH4) and CH4-dominated
end-points different from their formation temperatures more often than (>50 mol% CH4) ± ethane fluid inclusions in quartz at the margins veins
previously acknowledged (e.g. Thiagarajan et al., 2020a), governed by in the Junction deposit. Kendrick et al. (2011) documented rare liquid
the availability of catalytic sites and the internal fluid chemistry CH4 fluid inclusions associated with reduced pyrrhotite-bearing alter­
(Katayama et al., 2010). Nevertheless, the disconnect with calculated ation in the Conqueror deposit and analysed quartz samples dominated
isotopic C-H-O equilibrium temperatures here indicates that the fluid by CH4-rich vapour inclusions. Mixing of different fluid inclusion pop­
inclusion and Boomer gases have different origins in space and time, ulations during bulk analysis complicates comparison of CH4 in fluid
although potentially formed by a similar mechanism (Section 4.4). inclusions and well gas data. However, the CH4 in the fluid inclusions
are isotopically different to the CH4 in the Boomer well gases. For
4.3. Comparison of fluid inclusion and Boomer gases example, the methane separated from the fluid inclusion gases has much
lower δ13CCH4 values of − 35.3 to − 39.0‰ and much higher δ 2HCH4
Fluid inclusions associated with orogenic gold mineralisation in the values of − 137 to − 173‰ compared to the Boomer gas values of − 2.4 to
Yilgarn Craton and elsewhere typically trap low to high salinity (<5 wt% − 20.3‰ (δ13CCH4) and − 342 to − 379‰ (δ 2HCH4) (Tables 2 and 4).
NaCl equivalent) mixed H2O-CO2 fluids (Hagemann and Cassidy, 2000; Previous work demonstrated fluids trapped in sulphides associated with
Ridley and Diamond, 2000; Polito et al., 2001; Mernagh et al., 2007). CH4-rich fluid inclusion assemblages have 3He/4He ratios of 0.006 to
Fluid unmixing can lead to entrapment of pure H2O or pure CO2 fluid 0.015 RA(Kendrick et al., 2011) that indicate a crustal origin similar to
inclusions. However, the dominant gold-related fluids are often esti­ the Boomer gases (Table 3). However, these fluid inclusions also had
mated to have had an original CO2 fraction of ~30%. In addition to CO2, exceptionally radiogenic noble gas isotope signatures with unusually
gold deposits are commonly associated with reduced carbon (graphite) high 40Ar/36Ar ratios of up to ~50,000, 21Ne/22Ne ratios of up to 0.27
and CH4 is frequently present as a trace component of fluid inclusions. and 136Xe/130Xe ratios of up to 2.7 that are very different to the near
Due to its small size and high diffusivity H2 is not readily preserved in atmospheric ratios determined for the Boomer gases (Table S2). In

14
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 12. Plot of the carbon isotopic difference (Δ13C ‰) between methane and ethane versus the hydrogen isotopic difference (Δ2H ‰) between methane and ethane,
modified after Boreham and Davies (2020). The genetic fields for the FTT, Australian thermogenic gas and abiogenic seeps encompass the data presented in Boreham
and Davies (2020 and references therein).

addition, whereas the fluid inclusions have 4He/40Ar* ratios of 1–10 that explain the variation and zonation of δ13C of carbonates in the Eastern
are in the range of modern production ratios, the Boomer gases are Goldfield Superterrane (Golding et al., 1990).
significantly higher in 4He/40Ar* ratios (28 ± 10) relative to the The new data show that CH4-rich fluid inclusions from the Conqueror
calculated production ratios (6–11) (Table 5). deposit have distinctly different stable isotope and noble gas isotope
The CH4 and CO2 in fluid inclusions from gold deposits have a signatures compared to the modern day Boomer gases (Tables 1–4). The
number of possible sources and the balance between these components Boomer gases have elevated 4He/40Ar* ratios that favour a low-
might be altered by fluid-rock reactions at the site of gold deposition temperature origin that is broadly consistent with the calculated
(Ceplecha and Wall, 1976; Bottrell et al., 1988; Naden and Shepherd, Δ2HCH4-H2 equilibration temperatures of 42–57 ◦ C (Section 4.2). The
1989; Salier et al., 2005; Kendrick et al., 2011; Polito et al., 2001). Based inferred low temperature is also consistent with the involvement of
on the more strongly radiogenic isotope signature of noble gases in CH4 modern ground waters, which is indicated by: the near atmospheric
40
fluid inclusions compared to CO2 fluid inclusions, Kendrick et al. (2011) Ar/36Ar signature of the Boomer gases; 84Kr/36Ar and 132Xe/36Ar
proposed that gold mineralisation in the St Ives gold camp was localised characteristic of air saturated water; calculated δ2HH2O in the range of
by redox reactions between relatively oxidised gold-sulphur bearing modern groundwater (Section 4.2); and δ13CCO2 of − 11.3 to − 15.2‰;
fluids interacting with deeply-sourced abiogenic methane. This model is the latter is consistent with a mixed inorganic, atmospheric and organic
largely consistent with the gold mineralisation model of Walshe et al. source typical of soil gas. In contrast, the fluid inclusions have 4He/40Ar*
(2006), but the noble gas data require that CH4 had a crustal (not ratios in the range of Archaean production ratios that imply a temper­
mantle) source; possible sources of the abiogenic CH4 were suggested as ature above the nominal closure temperature of Ar in the crust (e.g.
serpentinization reactions or magmatic fluids (Kendrick et al., 2011). In >200–250 ◦ C), which is consistent with other gold-related fluids trapped
contrast, CH4 trapped in fluid inclusions in the Junction Au-deposit at typical temperatures of 300–400 ◦ C (e.g. Polito et al., 2001). As
(Fig. 1) was interpreted to have formed locally through a FTT re­ suggested by Kendrick et al. (2011) the strongly radiogenic noble gas
actions involving reduction of CO2 by H2 in the presence of Group VIII isotope signatures in CH4-rich fluid inclusions from the Conqueror de­
metals in their native or oxidised form (Polito et al., 2001). The localised posit are most easily explained by a deep fluid source isolated from at­
reduction of CO2 to CH4 in a reduced environment was also proposed to mospheric input. The high concentrations of radiogenic 4He in the fluid

15
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

inclusions imply the original presence of much greater quantities of a 13C-enriched abiogenic wet gas (±H2). A thermogenic gas source is
hydrolytic H2, but H2 is not preserved in fluid inclusions. The CH4 possibly from carbonaceous organic matter in either the nearby Cente­
present could have been generated from hydrolytic and nary Shale outcropping within 3 km east of the Frog’s Leg gold mine or
serpentenization-derived H2 in a similar way as in the modern Boomer locally in the Halley Shale (Fig. 13). Metamorphism will most likely lead
gases. Therefore, our data indicate that similar fluid-rock reactions (e.g. to a thermogenic dry gas with (δ13CCH4 ≤ − 20.2‰) favouring option 2
serpentinization) involving H2O-rich fluids and hydrolytic H2 have led over option 1. Further, the lack of a strong correlation between C1/C2 +
to CH4 generation in the Yilgarn over a period of ~2.65 Ga. The isotopic C3) ratio and δ13CCH4 would disfavour mixing of a 13C-enriched abio­
differences between fluid inclusions and Boomer gases can be largely genic dry gas with a 13C-depleted thermogenic wetter gas, although
explained because hydrolytic H2 and abiogenic CH4 were generated more complex mixing options may need to be considered when data
from deeply sourced magmatic or metamorphic fluids in the Neoarchean density improves. Having said this, option 3 is favoured here and a
and are generated from lower temperature groundwaters today. conceptual model for gas generation and migration at Frog’s Leg Camp
is depicted in Fig. 13. Here, stronger gas flows (i.e. higher relative
4.4. Pathways to gas generation and migration in the surrounds of the concentration of hydrocarbon gases; Table 1) are associated with gas
Frog’s Leg Gold Camp having higher 13C-enrichment in methane.
Gas generation and migration are likely from deeper sources than the
An abiogenic origin is deemed to be the most likely source of gas Boomer sections (Fig. 13). The wide range in H2 and hydrocarbon gas
since gas first flowed from a local Kalgoorlie gold mine in the early contents from trace to major in the well headspace likely reflects dif­
1900s. Then, the gas and graphite schists were considered to be inti­ ferences in proximity to advective major faults and the development of
mately connected with shear zones in the greenstones being the conduits local migration fracture networks (Etiope, 2015), rather than variations
for the migration of volcanic gas and with graphite being deposited in source rock chemistry where catalytic metals can be erratically
during methane oxidation via the reduction of ferric to ferrous minerals distributed (Golding et al., 1990). Assuming an average surface tem­
(Simpson, 1912). Such an inorganic mechanism, coincident with the perature of 25 ◦ C and 85 ◦ C at 5 km depth (Kennett and Blewett, 2012)
start of conventional petroleum extraction in Australia, is insightful, has results in depths of ~1.4–2.7 km for temperatures of 42.2–57.0 ◦ C
shaped and still holds many characteristics of current thinking on the (Section 4.2), consistent with isotopic equilibration in the Gleesons
mechanism(s) of abiogenic gas generation. Basalt. The ultramafics of the Hampton Formation, directly underlie the
In the Eastern Goldfields Superterrane, as in other ore deposits Gleesons Basalt, and represent an additional possible site for serpenti­
worldwide, there is likely to be three distinct gas morphologies: nization reactions. Subsurface at the Boomer deposit, water radiolysis,
occluded (mainly in fluid inclusions) gas, diffusively dispersed gas, and serpentinization and FTT synthesis contribute to gas generation process
free gas (Nivin, 2019). At the Boomer wells, free gas is the major of H2 and hydrocarbons. Advective carrier gas focussed along the shear
morphological sample type with variable input from fluid inclusion gas zone faults would entrain radiogenic 4He released from wall-rock min­
(see below). The relatively shallow depth and low formation tempera­ erals by recoil and diffusion (Ballentine et al., 2002; Etiope, 2015). Local
ture ranges for the generation of abiogenic CH4 (Section 4.2) coincides granites can also contribute to radiogenic 4He and H2 production from
with H2 at its lowest solubility (Lopez-Lazaro et al., 2019). With co- water radiolysis and from biotite alteration within fracture zones
generation, H2 is likely to be predominately in a gaseous state with (Table 5; see Sections 4.1 and 4.2). Our results suggests radiolysis of
advective migration of micro bubbles (Etiope, 2015). Regionally, ul­ water is a major contributor to the H2 inventory. Radiation-induced
tramafic rocks of the Kambalda Komatiite, although only a modest host polymerisation of methane, both within the basic and felsic rocks, is
rock for gold mineralisation, are characterised by intense development another possible contributor to the C2–C5 wet gases. Further, gas mixing
of serpentine in proximal to distal hydrothermal alteration zones with can occur with a pervasive dry gas with 13C-depleted methane released
emplaced Cr-rich minerals (e.g. fuchsite) proximal to gold-ore veins by decrepitation of fluid inclusions associated with the gold minerali­
(Duuring et al., 2004; Nichols and Hagemann, 2014). sation (Table 4; Polito, 1999; Polito et al., 2002), most likely released
Locally, the Boomer gases predominately reflect post-mineralisation into microfractures caused by progressive relief of highly stress/strained
processes influenced by the emplacement of geochemically available rock environments (Nivin, 2019). This independent abiogenic methane
reactants. From the evidence presented above, microbial methane pro­ source lends support to the suggestion that shallow abiogenic gas seeps
duction and utilisation are unlikely. The H2- and CH4-rich Boomer gases may not be from local processes as previously documented but are
are consistent with having an abiogenic origin via low temperature migrated from deeper abiogenic gas previously trapped in fluid in­
inorganic chemical synthesis involving gas-water-rock interactions clusions (McDermott et al., 2015; Grozeva et al., 2020).
complemented with radiolytic H2 production. The latter is supported by To further explore the pervasive nature of and sources for natural
the strong positive linear correlation between H2 and 4He (RC/RA (abiogenic?) gases associated with the Eastern Goldfields Superterrane
showing a purely crustal origin Table 3), and where radioactive decay and Australian ore deposits in general, additional gases will need to be
facilitates water hydrolysis and polymerisation of methane (Boreham collected, preferentially air-free, and analysed from a variety of mineral
and Davies, 2020 and references therein). ore-forming environments.
Given that the H-isotopes of methane are relatively constant, the
wide range in δ13C of methane needs further explanation. Simplicity 4.5. Geochemical guidelines for abiogenic gas generation
dictates a single abiogenic source where the carbon isotopic composition
of methane is responding to a limited supply of CO2 (e.g. H2 is always in Guidelines used to distinguish abiogenic from biogenic (especially
excess) and to the rate of the inorganic reactions; most positive δ13CCH4 high maturity thermogenic and admixtures) natural gases is evolving
occur where CO2 is almost completely exhausted (i.e. a Rayleigh and in doing so becoming increasingly blurred (Milkov and Etiope,
distillation model). However, the lack of accompanying 13C enrichment 2018; Etiope and Whiticar, 2019; Fiebig et al., 2019; Reeves and Fiebig,
in CO2 indicates a more complex process. The cause of the C- and H- 2020). The application of interdisciplinary skills and their associated
isotope profiles for methane could potentially be admixtures of gases, analytical techniques will become essential in progressing these guide­
albeit the different gases having the same δ2HCH4. Considering mixing lines. There is a heightened awareness of microbial inputs (Bradley and
involving abiogenic and/or thermogenic sources there are only three Summons, 2010; Etiope and Whiticar, 2019; Gregory et al., 2019;
reasonable options: 1) a 13C-depleted thermogenic wet gas (±H2) mixed Ménez, 2020). Nevertheless, 13C-enriched methane (i.e. δ13CCH4 >
with a 13C enriched abiogenic dry gas (±H2), 2) high maturity 13C- − 10‰) is most likely devoid of microbial influence. Kinetic isotope
depleted thermogenic dry gas (±H2) mixed with a 13C-enriched abio­ fractionation (δ13CCH4 - δ13CCO2) as low as − 10‰ has been found in a
genic wet gas (±H2), and 3) a 13C-depleted abiogenic dry gas (±H2) and cultivated chemolithoautotroph at 122 ◦ C and 40 Mpa (Takai et al.,

16
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Fig. 13. Conceptual model for gas generation and migration at the Frog’s Leg gold camp and surrounds. Top cross-section modifies the seismic line 99AGS-Y3 line
representation in Drummond et al. (2004), 18 km SE of the Frog’s Leg mine, and with mafic–ultramafic rocks dominating the top 6 km and granites below 6 km. The
bottom cross-section represents the local stratigraphy surrounding the Frog’s Leg gold camp. (For interpretation of the references to colour in this figure legend, the
reader is referred to the web version of this article.)

2008). The carbon isotopic fractionation is more pressure than tem­ and source attribution, and challenge past views on low temperature
perature sensitive and, by analogy, a Boomer gas with δ13CCH4 = − 2.4‰ sources for many abiogenic gas seeps. However, sampling protocols
from methanogensis would require a CO2 source with δ13CCO2 = +8.4‰ need to be more stringent for accurate results from these intramolecular,
but at unrealistic depth of >10 km. Clumped isotopes of methane and trace multi-isotope techniques (Douglas et al., 2017). Moreover, in the
ethane (Clog et al., 2018; Eldridge et al., 2019; Young, 2019 and ref­ application of geothermometers based on CO2-CH4-H2-H2O isotope
erences therein) and position-specific isotopes of propane and butane systematics (e.g. Bottinga, 1969, Horibe and Craig, 1995) and apparent
isomers (Julien et al., 2020 and references therein) at the intramolecular temperatures derived from the natural abundance clumped isotopes of
trace levels offer new evidence on equilibrium versus kinetics control methane (i.e. analyses of the CH4 isotopologues 13CH3D and 12CH2D2; D

17
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

= deuterium = 2H) and potentially molecular hydrogen (i.e. D2; Popa based on their molecular compositional relationships together with
et al., 2019), the interpreter may need to decide which of these ap­ integration of the 13C-enriched and 2H-depleted methane, the C- and H-
proaches gives the greatest certainty and whether equilibrium versus isotope profiles of the wet gases, the highly 2H-depleted H2 and
disequilibrium is evident. Concordance between apparent temperatures hydrogen isotopic equilibration temperatures. The minimum H2–CH4 H-
from both Δ13CH3D and Δ12CH2D2 is deemed prerequesite for confi­ isotopic equilibrium temperatures of between 42 ◦ C and 57 ◦ C corre­
dence in the geological temperature of formation or re-equilibration spond to calculated depths of ~1.4–2.7 km, suggesting equilibration
(Thiagarajan et al., 2020b). On the other hand, near equilibrium within the Gleeson Basalt beneath the Frog’s Leg gold camp. The hy­
Δ13CH3D composition and highly depleted Δ12CH2D2 is thought to be a drocarbons in the Boomer gases most likely formed during low tem­
signature for secondary cracking of oil and wet gas (Dong et al., 2019). perature serpentinization of mafic–ultramafic rocks where a limiting
As important is the intimate knowledge of the geological setting, past source of gaseous CO2 (from carbonate dissolution) is available at the
and present; a multidiscipline approach involving the latest technolog­ catalytic reaction site. H2 as the co-reactant could be produced almost
ical advances will result in a deeper consideration of abiogenic versus entirely by water radiolysis in the mafic/ultramafic lithologies under­
biogenic sources. lying the Boomer deposit. However, if co-sourcing of radiolytic H2 and
He was at a greater crustal depth than the H2–CH4 H-isotopic equilib­
4.6. Summary of geological factors rium depth then the accompanying lower H2/4He ratio necessitates an
increased contribution from serpentinization to H2 in the Boomer gases.
The interaction of hydrocarbon and non-hydrocarbon gases within Fischer-Tropsch type reactions facilitate the synthesis of hydrocarbons
hydrothermal fluids may have been a primary control on gold miner­ where radiolytic process involving methane polymerisation also likely
alisation and involves gas re-equilibration, unmixing and redox re­ contribute.
actions within the fluids and associated with wall-rocks. CO2 and H2O, Supplementary data to this article can be found online at https://doi.
either from magmatic sources or CO2 produced by carbonate decom­ org/10.1016/j.chemgeo.2021.120098.
position, are probably the main components within fluid inclusions in
many Yilgarn orogenic gold deposits. There are common co-occurrences Author contributions
of CH4 (± higher gaseous hydrocarbons) and N2 and fewer with He and
H2 trapped in Yilgarn fluid inclusions. Methane in fluid inclusions is C.J.B conceptualized the study and together with M.A.K. wrote the
depleted in 13C and enriched in 2H and is most likely in isotopic equi­ paper. J.H.S and Z.H. performed the formal analyses available at Geo­
librium with CO2 and H2 where reaction rate, hence re-equilibrium science Australia. J.W. managed the gas collection and N.C. wrote the
temperatures, is controlled by the composition of the internal fluid local geology surrounding the Boomer deposit wells.
and availability of internal catalyst. However, the carbon and hydrogen
isotopes of methane could have been modified from their initial source Declaration of Competing Interest
values, as a result of mixing and re-equilibrating with independently
sourced CO2. The distinctly different noble gas and stable isotope sig­ The authors have no conflicts of interest to declare.
natures of the Conqueror fluid inclusions and Boomer gases indicates
independent origins. The Boomer gases with air-saturated water abun­ Acknowledgements
dance ratios of non-radiogenic noble gases favours a relatively modern
origin for these gases. This is consistent with the mineable gold deposits Andrew Engelbrecht, Dom Osbourne and Mick Pocock at Evolution
in the Eastern Goldfields Superterrane having experienced uplift, Mining are thanked for logistics and collection of the gas samples. Our
erosion and having been close to the surface for up to ~2.5 billion years. gratitude goes to Junhong Chen and Neel Jinadasa, Isotope of Organic
However, we suggest post-gold mineralisation gas generation processes Geochemistry Laboratory, Geoscience Australia for additional QC/QA of
are still active today that rely on the chemical and mineralogical con­ the data. C.J.B. is grateful to the inorganic geochemistry and commodity
ditions set down during the Neoarchean mineralisation process. experts, David Champion and David Huston, for sharing their insights
into Yilgarn gold mineralisation. Alexei Milkov is thanked for providing
5. Conclusions selected data associated with Milkov and Etiope (2018). Kind thanks go
to Bianca Reese for drafting the figures. M.A.K. gratefully acknowledges
The Boomer gases are the first positive identification of an Australian Oliver Warr for helpful discussions. Dianne Edwards and Emmanuelle
abiogenic natural gas and we encourage the quest for future discoveries Grosjean provided valuable feedback in their internal reviews at Geo­
of abiotic hydrocarbons in different geological settings within Australia, science Australia. The thoughtful reviews and suggestions from Laurent
providing the baseline for estimating Australia’s He and H2 inferred Truche and an anonymous reviewer are much appreciated. Finally, C.J.
resource potential (Boreham et al., 2018; Boreham et al., 2021). B. wishes to give special thanks to Dhilip Kumar and the Elsevier pro­
In the surrounds of the Evolution Mining’s Frog’s Leg underground duction team for their assistance and patience. C.J.B, J.H.S, Z.H. publish
mine, the Boomer deposit exploration wells are progressively filled with with the permission of the CEO, Geoscience.
flammable gas and at variable rates, giving gas-at-surface from almost
pure air down to 30 mol% air contamination when sampled immediately Appendix
after drilling. Nitrogen is in excess and is still the dominant gas after air
correction. However, this is due to an O2 deficiency most likely caused Molecular composition
by oxidation of newly revealed, reduced mineral surfaces. Hence, a large
proportion of excess N2 is also attributed to air with a minor 15N The molar volumes (mol%) of the gas species were determined at
enriched indigenous N2 component. CO2 is in minor amounts consistent Geoscience Australia using an Agilent 6890 GC equipped with a com­
with a highly reduced natural gas. The carbon isotopic composition of bination of packed columns and two thermal conductivity detectors
the CO2 is likely a mixture of a 13C-depleted biogenic CO2 source mixed (TCD) with a detection limit of 0.001 mol%. Six 1/8” stainless steel
with atmospheric and local inorganic CO2 sources. The Boomer gases are packed columns were used to achieve the baseline separations of indi­
H2-rich with commonly sub-ordinate methane associated with minor C2- vidual gases: 2′ 12% UCW 982 on PAW 80/100 mesh, 15′ 25% DC200 on
C5 wet gases; the latter having attained a strong Anderson-Schulz-Flory PAW 80/100 mesh; 10’ HayeSep Q 80/100 mesh; 10′ molecular sieve
relationship. Helium isotope ratios indicate a pure crustal source. The 13X 45/60 mesh; 8′ molecular sieve 5 Å 60/80 mesh, and 3’ HayeSep Q
strong H2–CH4–He correlation suggests a genetic relationship. 80/100mesh. Two injectors (front with a 0.5 mL valve loop and back
A post-gold mineralisation abiogenic source for the Boomer gases is with a 2 mL valve loop) were used to introduce the gas sample into the

18
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

GC system. The injectors were at a constant temperature of 150 ◦ C and with δ2H = − 190.7‰ was injected immediately before the gas sample.
the total carrier gas flows were 27 mL/min for the front injector and 28 Additionally, a certified H2 gas standard with δ2H = − 470‰ (99.99 mol
mL/min for the back injector. GC oven temperature remained %; prepared by Alphagaz and distributed by Airgas) was employed for
isothermal at 90 ◦ C and both TCDs were set at 150 ◦ C. One TCD (front) isotopically light H2. All analyses were run as a minimum in duplicate
measured C1–C6+ alkanes, carbon dioxide (CO2), oxygen (O2) and N2 with a standard deviation less than 5‰. The δ2H values are reported
using a He carrier gas, while an argon (Ar) carrier gas (back) was used relative to VSMOW.
for He and H2. Calibration was carried out against two gravimetric gas For the hydrogen isotopic analysis of H2 in the natural gases, the GC-
standards from BOC: 1) a synthetic natural gas reference standard of TC-IRMS setup above was used but with the He flow rate reduced to 1.0
C1− C6+ alkanes, CO2 and N2; and 2) 0.5 mol% He and H2 reference mL/min, split ratio of 1:200 and injection volume up to 1 mL. The py­
standard in an Ar balance. All the components in a sample are calculated rolysis interface was set at 200 ◦ C to minimise interference from the
in mol% by accounting response factors with gas standards. Atmo­ elution of the following (partially pyrolysed) methane peak. For
spheric N2 and O2 introduced before instrumental analysis was sub­ example, using a laboratory H2 gas standard mixed with pure methane
tracted using a blank correction. The standard error of measured molar (99.999 mol%; CoreGas), the δ2H of the hydrogen gas was within
percentage was ±5% for major components. experimental error (< 5‰) of the pure laboratory H2 standard for syn­
Wet gas components that were below the TCD detection limit were thetic mixtures of H2:CH4 at 1:9 and 1:99.
quantified using a gas chromatography fitted with capillary column and Nitrogen isotopic composition of N2 was undertaken using the GC-C-
a flame ionisation detector. Detector response calibration was done by IRMS setup above. The GC was held isothermal at 30 ◦ C. The combustion
injecting varying volumes (5–200 μL; split ratio 25:1) of a gravimetric interface of the GC Isolink was kept at 400 ◦ C during the analysis. The
gas standard (BOC) of C1–C5 gaseous saturated hydrocarbons (He bal­ analytical performance was measured against a nitrogen reference ma­
ance) of known concentrations around 2000 ppm v/v. terial (room air) with a value of δ15N = 0.0‰ and internal N2 reference
‘pulses’ with a value of δ15N = − 2.58‰. All analyses were run as a
Carbon, hydrogen and nitrogen isotopes of C1–C5 gaseous hydrocarbons, minimum in duplicate with a maximum experimental error (1σ) of
CO2, H2 and N2 0.1‰. The δ15N values are reported relative to N2air.

Carbon isotopes of C1–C5 hydrocarbons and CO2 (Table 2) were

obtained at Geoscience Australia by GC-Combustion-Isotope Ratio Mass Fluid Inclusion gas
Spectrometry (GC-C-IRMS) using an Ultra Trace GC (Thermo Scientific)
coupled with GC Isolink combustion (set at 1000 ◦ C) and Conflo IV At Geoscience Australia, two quartz veins (Q5 and Q6) from the
interface and linked to a Thermo MAT 253 stable isotope ratio mass Conqueror deposit, St Ives Goldfield (Kendrick et al., 2011) were ana­
spectrometer. The hydrogen isotopes of C1–C4 hydrocarbons were lysed for fluid inclusion gas. Quartz veins were crushed to pass through a
determined using a GC-Thermal Conversion-IRMS (GC-TC-IRMS), with a 1.4 mm metal sieve and collected upon the surface of a 0.3 mm metal
Thermo Finnigan Trace Ultra gas chromatograph and GC IsoLink sieve. To remove surface contamination and adsorbed organics, the
equipped with a pyrolysis furnace (set at 1420 ◦ C), which was interfaced 0.3–1.4 mm sized material was soxhlet extracted with 9:1 dichloro­
on-line via GC Combustion IV to a Thermo Scientific Delta V Advantage methane:methanol for 48 h then dried overnight in an oven at 70 ◦ C. The
Stable Isotope Ratio Mass Spectrometer. Before analysis, background fluid inclusion gases were released by mechanical crushing of 0.5–2.0 g
scans and zero enrichment were performed to monitor the day-to-day of the cleaned material under a flow of He carrier gas. In two separate
performance of the GC-TC-IRMS system as a routine daily procedure. crushing procedures, the isotopic compositions of the released gases
The H3+ factor was calculated and it was stable and within a desired were analysed on-line for carbon isotopes in one experiment and
range (<10) which ensured the linearity of results with responses within hydrogen isotopes in the other experiment using GC-C-IRMS and GC-TC-
0.5 to 10 V. IRMS, respectively (see above), and detailed in Boreham et al. (2020).
Following the method in Boreham and Edwards (2008) and Boreham
et al. (2008), the hydrocarbon gas components were resolved using a
Poraplot Q fused silica column (25 m × 0.32 mm ID; Agilent) with a He Noble gas isotopes
carrier at a constant flow rate of 1.2 mL/min. The injector temperature
was maintained at 150 ◦ C. The gas sample was manually injected with The three gases (EVDD0071, EVDD0075 and EVDD0083) with the
an air-tight syringe. For methane analysis, a range of 30–300 μL of lowest air contamination were analysed for noble gas isotopes at Smart
natural gas at atmospheric pressure was injected with a split ratio of Gas Inc., USA. Analysis was undertaken using the method detailed in
25:1 and the GC held at 30 ◦ C. For the C2–C5 wet gases and CO2, analysis Darrah et al. (2015) and Schroder et al. (2020), with major gases (CH4,
was performed with split ratio and injection volume varying from 25:1 CO2, H2, He, N2 and O2) determined by gas chromatography together
to 20:1 and 50 μL to 10 mL, respectively. The actual combination of with quadrupole mass spectrometry for noble gas isotopes of He, Ne, Ar,
these settings for each analysis can be determined based on the peak Kr and Xe. Analyses were deemed acceptable when complementary gas
height(s) of the sort component(s) from the preceding analysis of the mol% was within 3% of the values between analytical procedures. The
sample. Following the elution of ethane, the GC oven temperature was air content measured in the USA (not reported here) is much lower than
increased to 230 ◦ C at 20 ◦ C/min and then held isothermal for 10 min. compared to air measured at Geoscience Australia (Table 1). This major
Cryogenic cold trapping using liquid nitrogen was used for narrowing difference is attributed to the loss of O2 by chemical reaction with the
the chromatographic band and improving the detection limit for the metal surface of the cold sealed 3/8′′ x 12′′ refrigerator grade Cu tube
gaseous components of C1–C3 and CO2. after gas transfer at Geoscience Australia.
For carbon isotopes, performance was measured against an external
methane reference material with a δ13C of − 41.78‰ and internal CO2 References
reference ‘pulses’ with δ13C of − 33.48‰. All analyses were run as a
Abrajano, T.A., Surchio, N.C., Kennedy, B.M., Lyon, G.L., Muehlenbachs, K., Bohlke, J.K.,
minimum in duplicate with a maximum experimental error (1σ) of
1990. Geochemistry of reduced gas related to serpentinization of the Zambales
±0.3‰. The δ13C values are reported relative to VPDB. The NGS 3 gas ophiolite, Philippines. Appl. Geochem. 5, 625–630. https://doi.org/10.1016/0883-
with known δ2H methane = − 175.95‰ was used to monitor the whole 2927(90)90060-I.
GC-TC-IRMS instrument performance. A H2 laboratory reference gas Adamiec, G., Aitken, M., 1998. Dose-rate conversion factors: update. Ancient TL 16,
37–50.
(99.999 mol%; Coregas) with δ2H = − 170.0‰ was used for instrument Anderson, R.B., 1984. The Fischer-Tropsch Synthesis. Academic Press Inc., Harcourt
set-up while an external laboratory H2 standard (99.999 mol%; BOC) Brace Javanovich, Publishers.

19
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Anderson, R.B., Friedel, R.A., Storch, H.H., 1951. Fischer-Tropsch reaction mechanism D’Alessandro, W., Yücem, G., Italiano, F., Bellomo, S., Gülbay, A.H., Yasin, D.U.,
involving stepwise growth of carbon chain. J. Chem. Phys. 19, 313–319. https://doi. Gagliano, A.L., 2018b. Large compositional differences in the gases released from the
org/10.1063/1.1748201. Kizildag ophiolitic body Turkey evidences of prevailingly abiogenic origin. Mar. Pet.
Ballentine, C.J., Burnard, P.G., 2002. Production, release and transport of noble gases in Geol. 89, 174–184. https://doi.org/10.1016/j.marpetgeo.2016.12.017.
the continental crust. Rev. Mineral. Geochem. 47, 481–538. https://doi.org/ Darling, 1998. Hydrothermal hydrocarbon gases: 1, genesis and geothermometry. Appl.
10.2138/rmg.2002.47.12. Geochem. 13, 815–824. https://doi.org/10.1016/S0883-2927(98)00013-4.
Ballentine, C.J., Burgess, R., Marty, B., 2002. Tracing fluid origin, transport and Darrah, T.H., Jackson, R.B., Vengosh, A., Warner, N.R., Whyte, C.J., Walsh, T.B.,
Interaction in the crust. Rev. Mineral. Geochem. 47, 539–614. https://doi.org/ Kondash, A.J., Poreda, R.J., 2015. The evolution of Devonian hydrocarbon gases in
10.2138/rmg.2002.47.13. shallow aquifers of the northern Appalachian Basin: insights from integrating noble
Barbier, S., Huang, F., Andreani, M., Tao, R., Hao, J., Eleish, A., Prabhu, A., Minhas, O., gas and hydrocarbon geochemistry. Geochim. Cosmochim. Acta 170, 321–355.
Fontaine, K., Fox, P., Daniel, I., 2020. A review of H2, CH4, and hydrocarbon https://doi.org/10.1016/j.gca.2015.09.006.
formation in experimental serpentinization using network analysis. Frontiers in Dong, G., Xie, H., Formolo, M., Lawson, M., Sessions, A., Eiler, J., 2019. Methane
Earth Science 8, 209. https://doi.org/10.3389/feart.2020.00209. clumped isotope constraints on the equilibrium and kinetics of unconventional
Bertoncini, F., Marian, M.C., Brodusch, N., Esnault, S., 2009. Unravelling molecular gases. In: Abstract at Goldschmidt 2019 Conference, Barcelona, Spain, August
composition of products from cobalt catalysed Fischer-Tropsch reaction by 18–23, 2019. https://goldschmidtabstracts.info/2019/835.pdf.
comprehensive gas chromatography: methodology and application. Oil Gas Sci. Douglas, P.M.J., Stolper, D.A., Eiler, J.M., Sessions, A.L., Lawson, M., Shuai, Y.,
Technol. - Rev. IFP64, 79–90. Bishop, A., Podlaha, O.G., Ferreira, A.A., Santos Neto, E.A., Niemann, M., Steen, A.S.,
Beyssac, O., Rumble, D., 2014. Graphitic carbon: a ubiquitous, diverse, and useful Huang, L., Chimiak, L., Valentine, D.L., Fiebig, J., Luhmannm, A.J., Seyfried Jr., W.
geomaterial. Elements 10, 415–420. https://doi.org/10.2113/gselements.10.6.415. E., Etiope, G., Schoell, M., Inskeep, W.P., Moran, J.J., Kitchen, N., 2017. Methane
Boreham, C.J., Davies, J.B., 2020. Carbon and hydrogen isotopes of the wet gases clumped isotopes: progress and potential for a new isotopic tracer. Org. Geochem.
produced by gamma-ray-induced polymerisation of methane: insights into 113, 262–282. https://doi.org/10.1016/j.orggeochem.2017.07.016.
radiogenic mechanism and natural gas formation. Radiat. Phys. Chem. 168, 108546. Drummond, B.E., Hobbs, B.J., Hobbs, R.W., Goleby, B.R., 2004. Crustal fluids in tectonic
https://doi.org/10.1016/j.radphyschem.2019.108546. evolution and mineral systems: evidence from the Yilgarn Craton. In: Barnicoat, A.
Boreham, C.J., Edwards, D.S., 2008. Abundance and carbon isotopic composition of neo- C., Korsch, R.J. (Eds.), Predictive Mineral Discovery Cooperative Research Centre –
pentane in Australian natural gases. Org. Geochem. 39, 550–566. https://doi.org/ Extended Abstracts from the June 2004 Conference, pp. 33–38. Geoscience
10.1016/j.orggeochem.2007.11.004. Australia, Record 2004/09.
Boreham, C.J., Edwards, D.S., Hope, J.M., Chen, J., Hong, Z., 2008. Carbon and Dubessy, J., Poty, B., Ramboz, C., 1989. Advances in C-O-H-N-S fluid geochemistry based
hydrogen isotopes of neo-pentane for biodegraded natural gas correlation. Org. on micro-Raman spectrometric analysis of fluid inclusions. Eur. J. Mineral. 1,
Geochem. 39, 1483–1486. https://doi.org/10.1016/j.orggeochem.2008.06.010. 517–534.
Boreham, C.J., Edwards, D.S., Poreda, R.J., Darrah, T.H., Zhu, R., Grosjean, E., Main, P., Dugamin, E., Truche, L., Donzé, F.V., 2019. Natural hydrogen exploration guide,
Waltenberg, K., Henson, P.A., 2018. Helium in the Australian liquefied natural gas Geonum Ed.: ISRN GEONUM-NST–2019-01–ENG, 16p.
economy. APPEA J. 58, 209–237. https://doi.org/10.1071/AJ17049. Duuring, P., Hagemann, S.G., Cassidy, K.F., Johnson, C.A., 2004. Hydrothermal
Boreham, C.J., Edwards, D.S., Sohn, J.H., Palatty, P., Chen, J.H., Mory, A.J., 2020. Gas alteration, ore fluid characteristics, and gold depositional processes along a
systems in the onshore Canning Basin as revealed by gas trapped in fluid inclusions. trondhjemite-komatiite contact at Tarmoola, Western Australia. Econ. Geol. 99,
In: Czarnota, K., et al. (Eds.), Exploring for the Future: Extended Abstracts. 423–451. https://doi.org/10.2113/gsecongeo.99.3.423.
Geoscience Australia, Canberra, pp. 1–4. http://pid.geoscience.gov.au/dataset/ga/ Eldridge, D.L., Korol, R., Llyod, M.K., Turner, A.C., Webb, M.A., Miller III, T.F.,
135207. Stopler, D.A., 2019. Comparison of experimental vs theoretical abundances of
13
Boreham, C.J., Edwards, D.S., Czado, K., Rollett, N., Wang, L., Van Der Wielen, S., CH3D and 12CH2D2 for isotopically equilibrated systems from 1 to 500 ◦ C. ACS
Champion, D., Blewett, R., Feitz, A., Henson, P., 2021. Hydrogen in Australian Earth Space Chem. 3, 2747–2764. https://doi.org/10.1021/
natural gas: occurrences, sources and resource. APPEA J. 61 https://doi.org/ acsearthspacechem.9b00244.
10.1071/AJ2044 (in press). Etiope, G., 2015. Natural gas seepage – the Earth’s hydrocarbon degassing. Springer
Boschetti, T., Etiope, G., Toscani, L., 2013. Abiotic methane in the hyperalkaline springs International Publishing AG Switzerland. https://doi.org/10.1007/978-3-319-
of Genova, Italy. Proc. Earth Planet. Sci. 7, 248–251. https://doi.org/10.1016/j. 14601-0, 191 pp.
proeps.2013.02.004. Etiope, G., Ionescu, A., 2015. Low-temperature catalytic CO2 hydrogenation with
Bottinga, Y., 1969. Calculated fractionation factors for carbon and hydrogen isotope geological quantities of ruthenium: a possible abiotic CH4 source in chromitite-rich
exchange in the system calcite–carbon dioxide–graphite–methane–hydrogen–water serpentinized rocks. Geofluids 15, 438–452. https://doi.org/10.1111/gfl.12106.
vapor. Geochim. Cosmochim. Acta 33, 49–64. https://doi.org/10.1016/0016-7037 Etiope, G., Sherwood, Lollar B., 2013. Abiotic methane on Earth. Rev. Geophys. 51,
(69)90092-1. 276–299. https://doi.org/10.1002/rog.20011.
Bottrell, S.H., Shepherd, T.J., Yardley, B.W.D., Bubessy, J., 1988. A fluid inclusion model Etiope, G., Whiticar, M.J., 2019. Abiotic methane in continental ultramafic rock systems:
for the genesis of the ores of the Dolgellau Gold Belt, North Wales. J. Geol. Soc. Lond. towards a genetic model. Appl. Geochem. 102, 139–152. https://doi.org/10.1016/j.
145, 139–145. https://doi.org/10.1144/gsjgs.145.1.0139. apgeochem.2019.01.012.
Botz, R., Stüben, D., Winckler, G., Bayer, R., Schmitt, M., Faber, E., 1996. Hydrothermal Etiope, G., Drobniak, A., Schimmelmann, A., 2013. Natural seepage of shale gas and the
gases offshore Milos Island, Greece. Chem. Geol. 30, 161–173. https://doi.org/ origin of “eternal flames” in the Northern Appalachian Basin, USA. Mar. Pet. Geol.
10.1016/0009-2541(96)00023-X. 43, 178–186. https://doi.org/10.1016/j.marpetgeo.2013.02.009.
Bouquet, A., Glein, C.R., Wyrick, D., Waite, J.H., 2017. Alternative energy: production of Etiope, G., Samardžić, N., Grassa, F., Hrvatović, H., Miošić, N., Skopljak, F., 2017.
H2 by radiolysis of water in the rocky cores of icy bodies. Astrophys. J. Lett. 840, L8. Methane and hydrogen in hyperalkaline groundwaters of the serpentinized Dinaride
https://doi.org/10.3847/2041-8213/aa6d56. ophiolite belt. Bosnia Herzegovina. Appl. Geochem. 84, 286–296. https://doi.org/
Boyd, A.W., Willis, C., Miller, O.A., 1973. The radiolysis of water vapor and very high 10.1016/j.apgeochem.2017.07.006.
dose rates. I. Hydrogen yields from H2O, H2O-HCl, and H2O-HBr mixture. Can. J. Evans, B.W., Hattori, K., Baronnet, A., 2013. Serpentinite: what, why, where? Elements
Chem. 51, 4048–4055. 9, 99–106. https://doi.org/10.2113/gselements.9.2.99.
Bradley, A.S., Summons, R.E., 2010. Multiple origins of methane at the Lost City Fiebig, J., Woodland, A.B., Spangenberg, J., Oschmann, W., 2007. Natural evidence for
Hydrothermal Field. Earth Planet. Sci. Lett. 297, 34–41. https://doi.org/10.1016/j. rapid abiogenic hydrothermal generation of CH4. Geochim. Cosmochim. Acta 71,
epsl.2010.05.034. 3028–3039. https://doi.org/10.1016/j.gca.2007.04.010.
Cassidy, K.F., Champion, D.C., Krapež, B., Barley, M.E., Brown, S.J.A., Blewett, R.S., Fiebig, J., Stefánsson, A., Ricci, A., Tassi, F., Viveiros, F., Silva, C., Lopez, T.M.,
Groenewald, P.B., Tyler, I.M., 2006. A revised geological framework for the Yilgarn Schreiber, C., Hofmann, S., Mountain, B.W., 2019. Abiogenesis not required to
Craton, Western Australia. Western Australia Geological Survey, Record 2006/8, 8p. explain the origin of volcanic hydrothermal hydrocarbons. Geochem. Perspect. Lett.
Ceplecha, J.C., Wall, V.J., 1976. Chewton goldfield and Wattle Gully mine: a model for 11, 23–27. https://doi.org/10.7185/geochemlet.
gold-quartz mineralization in slate belts. Bull. Austral. Soc. Explor. Geophys. 7, 40. Formolo, M., 2010. The microbial production of methane and other volatile
Champion, D.C., Sheraton, J.W., 1997. Geochemistry and Nd isotope systematics of hydrocarbons. In: Handbook of Hydrocarbon and Lipid Microbiology – Formation
Archaean granites of the Eastern Goldfields, Yilgarn Craton, Australia: implications and Location, pp. 113–126. https://doi.org/10.1007/978-3-540-77587-4_6.
for crustal growth processes. Precambrian Res. 83, 109–132. https://doi.org/ Fujitani, T., Nakamura, I., Akita, T., Okumura, M., Haruta, M., 2009. Hydrogen
10.1016/S0301-9268(97)00007-7. dissociation in gold clusters. Angew. Chem. Int. Ed. 49, 9515–9518. https://doi.org/
Clog, M., Lawson, M., Peterson, B., Ferreira, A.A., Santos Neto, E.V., Eiler, J.M., 2018. 10.1002/anie.200905380.
A reconnaissance study of 13C–13C clumping in ethane from natural gas. Geochim. Gerdenich, M.J., Hanel, I.V., Kesler, S.E., 1986. Alkene/alkane hydrocarbon gas ratios in
Cosmochim. Acta 223, 229–244. https://doi.org/10.1016/j.gca.2017.12.004. fluid inclusions from Archean and Tertiary gold deposits. Geol. Soc. Am. Abstract
Cooke, D.P., Ramsey, T., Miller, J.M., Bath, A.B., 2017. Kundana goldfield. In: Phillips, G. Programs 18, 611 (No 109560).
N. (Ed.), Australian Ore Deposits. The Australasian Institute of Mining and Giggenbach, W.F., 1997. Relative importance of thermodynamic and kinetic processes in
Metallurgy, Melbourne, pp. 199–205. governing the chemical and isotopic composition of carbon gases in high-heatflow
Coveney Jr., R.M., Goebel, E.D., Zeller, E.J., Dreschhoff, G.A.M., Angino, E.E., 1987. sedimentary basins. Geochim. Cosmochim. Acta 61, 3763–3785. https://doi.org/
Serpentinization and the origin of hydrogen gas in Kansas. AAPG Bull. 71, 39–48. 10.1016/S0016-7037(97)00171-3.
https://doi.org/10.1306/94886D3F-1704-11D7-8645000102C1865D. Glein, C.R., Zolotov, M.Yu., 2020. Hydrogen, hydrocarbons, and habitability across the
D’Alessandro, W., Daskalopoulou, K., Calabrese, S., Bellomo, S., 2018a. Water chemistry Solar System. Elements 16, 47–51. https://doi.org/10.2138/gselements.16.1.47.
and abiogenic methane content of a hyperalkaline spring related to serpentinization Goebel, E.D., Coveney, R.M., Angino, E.E., Zeller, E.J., Dreschhoff, G.A.M., 1984.
in the Argolida ophiolite Ermioni, Greece. Mar. Pet. Geol. 89, 185–193. https://doi. Geology, composition, isotopes of naturally occurring H2/N2 rich gas from wells near
org/10.1016/j.marpetgeo.2017.01.028. Junction City, Kans. Oil Gas J. 82, 215–222.

20
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Golding, S.D., Clark, M.E., Keele, R.A., Wilson, A.F., Keays, R.R., 1990. Geochemistry of Lehmann, L., Baerns, M., 1992. Mechanistic aspects of the oxidative coupling of methane
Archaean epigenetic gold deposits in the Eastern Goldfields Province, Western over a NaOH/CaO catalyst. Catal. Today 13, 265–272. https://doi.org/10.1016/
Australia. In: Herbert, H.K., Ho, S.E. (Eds.), Stable Isotopes and Fluid Processes in 0920-5861(92)80150-L.
Mineralisation, p.23. Geology Department and University Extension. University of Lin, L.-H., Hall, J., Lippmann-Pipke, J., Ward, J.A., Sherwood Lollar, B., DeFlaun, M.,
Western Australia, Nedlands. Rothmel, M., Moser, D., Gihring, T.M., 2005a. Radiolytic H2 in continental crust:
Gregory, S.P., Barnett, M.J., Field, L.P., Milodowski, A.E., 2019. Subsurface microbial nuclear power for deep subsurface microbial communities. Geochem. Geophys.
hydrogen cycling: natural occurrence and implications for industry. Microorganisms GeoSyst. 6, Q07003 https://doi.org/10.1029/2004GC000907.
7, 53. https://doi.org/10.3390/microorganisms7020053. Lin, L.-H., Slater, G.F., Sherwood Lollar, B., Lacrampe-Couloume, G., Onstott, T.C.,
Groves, D.I., Ridley, J.R., Bloem, E.M.J., Gebre-Mariam, M., Hagemann, S.G., Hronsky, J. 2005b. The yield and isotopic composition of radiolytic H2, a potential energy source
M.A., Knight, J.T., McNaughton, N.J., Ojala, J., Vielreicher, R.M., McCuaig, T.C., for the deep subsurface biosphere. Geochim. Cosmochim. Acta 69, 893–903. https://
Holyland, P.W., 1995. Lode-gold deposits of the Yilgarn block: products of Late doi.org/10.1016/j.gca.2004.07.032.
Archaean crustal-scale overpressured hydrothermal systems. Geol. Soc. Lond., Spec. Lopez-Lazaro, C., Bachaud, P., Moretti, I., Ferrando, N., 2019. Predicting the phase
Publ. 95, 155–172. https://doi.org/10.1144/GSL.SP.1995.095.01.10. behavior of hydrogen in NaCl brines by molecular simulation for geological
Grozeva, N.G., Klein, F., Seewald, J.S., Sylva, S.P., 2020. Chemical and isotopic analyses applications. BSGF - Earth Sci. Bull. 190, 7. https://doi.org/10.1051/bsgf/2019008.
of hydrocarbon-bearing fluid inclusions in olivine-rich rocks. Phil. Trans. R. Soc. A Lunsford, J.H., 1990. The catalytic conversion of methane to higher hydrocarbons. Catal.
378, 20180431. https://doi.org/10.1098/rsta.2018.0431. Today 6, 235–259. https://doi.org/10.1016/0920-5861(90)85004-8.
Guérin, G., Mercier, N., Adamiec, G., 2011. Dose-rate conversion factors: update. Ancient Luque, F.J., Crespo-Feo, E., Barrenechea, J.F., Ortega, L., 2012. Carbon isotopes of
TL 29, 5–9. https://www.aber.ac.uk/en/media/departmental/dges/ancie graphite: implications on fluid history. Geosci. Front. 3, 197–207. https://doi.org/
nttl/pdf/vol29no1/atl-issue29-1.pdf#page=9. (Accessed 5 February 2021). 10.1016/j.gsf.2011.11.006.
Hagemann, S.G., Cassidy, K.F., 2000. The geodynamics of world-class gold deposits: Lyon, G.L., Hulston, J.R., 1984. Carbon and hydrogen isotopic compositions of New
characteristics, space-time distribution and origins. In: Hagemann, S.G., Brown, P.E. Zealand geothermal gases. Geochim. Cosmochim. Acta 48, 1161–1171. https://doi.
(Eds.), Gold in 2000, Reviews in Economic Geology, vol. 13, pp. 9–68. org/10.1016/0016-7037(84)90052-8.
Hao, Y., Pang, Z., Tian, J., Wang, Y., Li, Z., Li, L., 2020. Origin and evolution of Marques, J.M., Etiope, G., Neves, M.O., Carreira, P.M., Rocha, C., Vance, S.D.,
hydrogen-rich gas discharges from a hot spring in the eastern coastal area of China. Christensen, L., Miller, A.Z., Suzuki, S., 2018. Linking serpentinization,
Chem. Geol. 538, 119477. https://doi.org/10.1016/j.chemgeo.2020.119477. hyperalkaline mineral waters and abiotic methane production in continental
Hofmann, B.A., 1992. Isolated reduction phenomena in red beds: a result of porewater peridotites: an integrated hydrogeological-biogeochemical model from the Cabeço
radiolysis? In: Kharaka, Y.K., Maest, A.S. (Eds.), Water-rock interaction. Proceedings de Vide CH4-rich aquifer Portugal. Appl. Geochem. 96, 287–301. https://doi.org/
of the 7th international Symposium on Water-Rock Interaction, Park City, Utah 10.1016/j.apgeochem.2018.07.011.
13–18 July 1992. A.A. Balkema, Rotterdam, pp. 503–506. Martinez, I., Geymond, U., Guelard, J., Eka]mbas, P., Chardin, M., Sissmann, O., 2019.
Holland, H.D., 2002. Volcanic gases, black smokers, and the Great Oxidation Event. The role of metallic cation in H2 production during hydrothermal reactions. In
Geochim. Cosmochim. Acta 66, 3811–3826. https://doi.org/10.1016/S0016-7037 ‘Abstract at Goldschmidt 2019 Conference, Barcelona, Spain, August 18–23, 2019’.
(02)00950-X. https://goldschmidtabstracts.info/2019/2173.pdf.
Holloway, J.R., 1984. Graphite-CH4-H2O-CO2 equilibria at low-grade metamorphic Mavrogenes, J.A., Bodnar, R.J., 1994. Hydrogen movement into and out of fluid
conditions. Geology 12, 455–458. https://doi.org/10.1130/0091-7613(1984) inclusions in quartz: experimental evidence and geologic implications. Geochim.
12<455:GEALMC>2.0.CO;2. Cosmochim. Acta 58, 141–148. https://doi.org/10.1016/0016-7037(94)90452-9.
Horibe, Y., Craig, H., 1995. D/H fractionation in the system methane–hydrogen–water. Mayhew, L.E., Ellison, E.T., McCollom, T.M., Trainor, T.P., Templeton, A.S., 2013.
Geochim. Cosmochim. Acta 59, 5209–5217. https://doi.org/10.1016/0016-7037 Hydrogen generation from low-temperature water–rock reactions. Nat. Geosci. 6,
(95)00391-6. 478–483. https://doi.org/10.1038/NGEO1825.
Horita, J., Berndt, M.E., 1999. Abiogenic methane formation and isotopic fractionation McCollom, T.M., 2013. Laboratory simulations of abiotic hydrocarbon formation in
under hydrothermal conditions. Science 285, 1055–1057. https://doi.org/10.1126/ Earth’s deep subsurface. Rev. Mineral. Geochem. 75, 467–494. https://doi.org/
science.285.5430.1055. 10.2138/rmg.2013.75.15.
Hulson, J.R., 1977. Isotope work applied to geothermal systems at the Institute of McCollom, T.M., Seewald, J.S., 2013. Serpentinites, hydrogen and life. Elements 9,
Nuclear Sciences, New Zealand. Geothermics 5, 89–96. https://doi.org/10.1016/ 129–134. https://doi.org/10.2113/gselements.9.2.129.
0375-6505(77)90013-X. McCollom, T.M., Sherwood Lollar, B., Lacrampe-Couloume, G., Seewald, J.S., 2010. The
Hunt, J.M., 1996. Petroleum Geochemistry and Geology. W.H. Freeman and Co, New influence of carbon source on abiotic organic matter synthesis and carbon isotope
York, 743 pp. fractionation under hydrothermal conditions. Geochim. Cosmochim. Acta 74,
Huston, D.L., Blewett, R.S., Skirrow, R., McQueen, A., Wang, J., Jaques, L., Waters, D., 2717–2740. https://doi.org/10.1016/j.gca.2010.02.008.
Bierlein, F., 2012. Foundations of wealth—Australia’s major mineral provinces. In: McDermott, J.M., Seewald, J.S., German, C.R., Sylva, Sean P., 2015. Pathways for abiotic
Blewett, R.S. (Ed.), Shaping a Nation: Geology of Australia. Geoscience Australia and organic synthesis at submarine hydrothermal fields. Proc. Natl. Acad. Sci. U. S. A. 25,
ANU E Press, Canberra, pp. 381–431. http://press-files.anu.edu.au/downloads/press 7668–7672. www.pnas.org/cgi/doi/10.1073/pnas.1506295112.
/p194981/pdf/chapter-8.pdf. McInnes, B., Darrah, T., Hilton, R., Boreham, C., Edwards, D., Gusterhuber, J., 2017.
IEA, 2019. The Future of Hydrogen. Report prepared by the IEA for the G20, Japan, Investigations into the highest reported He concentration in a natural gas sample: Mt
pp. 1–199. Available at. https://www.iea.org/reports/the-future-of-hydrogen. Kitty, Amadeus Basin, Northern Territory. In: Abstract at Goldschmidt 2017
Ji, F., Zhou, H., Yang, Q., 2008. The abiotic formation of hydrocarbons from dissolved Conference, Paris, France, August 13–18, 2017. https://goldschmidt.info/2017/abst
CO2 under hydrothermal conditions with cobalt-bearing magnetite. Orig. Life Evol. racts/abstractView?id=2017003781.
Biosph. 38, 117–125. https://doi.org/10.1007/s11084-008-9124-7. Ménez, B., 2020. Abiotic hydrogen and methane: fuels for life. Elements 16, 39–46.
Julien, M., Goldman, M.J., Liu, C., Horita, J., Boreham, C.J., Yamada, K., Green, W.H., https://doi.org/10.2138/gselements.16.1.39.
Yoshida, N., Gilbert, A., 2020. Intramolecular 13C isotope distributions of butane Mernagh, T., Bastrakov, E.N., Zaw, K., Wygralak, A.S., Wyborn, L.A.I., 2007. Comparison
from natural gases. Chem. Geol. 541, 119571. https://doi.org/10.1016/j. of fluid inclusion data and mineralization processes for Australian orogenic gold and
chemgeo.2020.119571. intrusion related gold systems. Acta Petrol. Sin. 23, 21–32.
Katayama, I., Kurosaki, I., Hirauchi, K., 2010. Low silica activity for hydrogen generation Milesi, V., Guyot, F., Brunet, F., Richard, L., Recham, N., Benedetti, M., Dairou, J.,
during serpentinization: an example of natural serpentinites in the Mineoka ophiolite Prinzhofer, A., 2015. Formation of CO2, H2 and condensed carbon from siderite
complex, central Japan. Earth Planet. Sci. Lett. 298, 199–204. https://doi.org/ dissolution in the 200–300 ◦ C range and at 50 MPa. Geochim. Cosmochim. Acta 154,
10.1016/j.epsl.2010.07.045. 201–211. https://doi.org/10.1016/j.gca.2015.01.015.
Kelley, D.S., Früh-Green, L., 1999. Abiogenic methane in deep-seated mid-ocean ridge Milesi, V., McCollom, T.M., Guyot, F., 2016. Thermodynamic constraints on the
environments: insights from stable isotope analyses. J. Geophys. Res. 104, formation of condensed carbon from serpentinization fluids. Geochim. Cosmochim.
10439–10460. https://doi.org/10.1029/1999JB900058. Acta 189, 391–403. https://doi.org/10.1016/j.gca.2016.06.006.
Kendrick, M.A., Honda, M., Walshe, J., Petersen, K., 2011. Fluid sources and the role of Milkov, A.V., Etiope, G., 2018. Revised genetic diagrams for natural gases based on a
abiogenic-CH4 in Archean gold mineralization: constraints from noble gases and global dataset of >20,000 samples. Org. Geochem. 125, 109–120. https://doi.org/
halogens. Precambrian Res. 189, 313–327. https://doi.org/10.1016/j. 10.1016/j.orggeochem.2018.09.002.
precamres.2011.07.015. Miller, H.M., Matter, J.M., Kelemen, P., Ellison, E.T., Conrad, M.E., Fierer, N.,
Kennett, B.L.N., Blewett, R.S., 2012. Lithospheric framework of Australia. Episodes 35, Ruchala, T., Tominaga, M., Templeton, A.S., 2016. Modern water/rock reactions in
9–22. Oman hyperalkaline peridotite aquifers and implications for microbial habitability.
Kietäväinen, R., Ahonen, L., Niinikoski, P., Nykänen, H., Kukkonen, I.T., 2017. Abiotic Geochim. Cosmochim. Acta 179, 217–241. https://doi.org/10.1016/j.
and biotic controls on methane formation down to 2.5 km depth within the gca.2016.01.033.
Precambrian Fennoscandian Shield. Geochim. Cosmochim. Acta 202, 124–145. Miller, H.M., Chaudhry, N., Conrad, M.E., Bill, M., Kopf, S.H., Templeton, A.S., 2018.
https://doi.org/10.1016/j.gca.2016.12.020. Large carbon isotope variability during methanogenesis under alkaline conditions.
Kiyosu, Y., Krouse, H.R., 1989. Carbon isotope effect during abiogenic oxidation of Geochim. Cosmochim. Acta 237, 18–31. https://doi.org/10.1016/j.
methane. Earth Planet. Sci. Lett. 95, 302–306. https://doi.org/10.1016/0012-821X gca.2018.06.007.
(89)90105-2. Murray, J., Clément, A., Fritz, B., Schmittbuhl, J., Bordmann, V., Fleury, J.M., 2020.
Lazar, C., McCollom, T.M., Manning, C.E., 2012. Abiogenic methanogenesis during Abiotic hydrogen generation from biotite-rich granite: a case study of the Soultz-
experimental komatiite serpentinization: implications for the evolution of the early sous-Forêts geothermal site, France. Appl. Geochem. 119, 104631. https://doi.org/
Precambrian atmosphere. Chem. Geol. 326–327, 102–112. https://doi.org/10.1016/ 10.1016/j.apgeochem.2020.104631.
j.chemgeo.2012.07.019. Naden, J., Shepherd, T.J., 1989. Role of methane and carbon dioxide in gold deposition.
Nature 342, 793–795. https://doi.org/10.1038/342793a0.

21
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

Nandi, R., Sengupta, S., 1998. Microbial production of hydrogen: an overview. Crit. Rev. Sherwood Lollar, B., Westgate, T.D., Ward, J.A., Slater, G.F., Lacrampe-Couloume, G.,
Microbiol. 24, 61–84. https://doi.org/10.1080/10408419891294181. 2002. Abiogenic formation of alkanes in the earth’s crust as a minor source of global
Neal, C., Stranger, G., 1983. Hydrogen generation from mantle source rocks in Oman. hydrocarbon reservoirs. Nature 416, 522–524. https://doi.org/10.1038/416522a.
Earth Planet. Sci. Lett. 66, 315–320. https://doi.org/10.1016/0012-821X(83)90144- Shock, E., Bockisch, C., Estrade, C., Fecteua, K., Gould, I.R., Hartnett, H., Johnson, K.,
9. Robinson, K., Scipp, J., Williams, L., 2019. Earth as organic chemist. In: Orcutt, B.,
Nichols, S.J., Hagemann, S.G., 2014. Structural and hydrothermal alteration evidence for Daniel, I., Dasgupta, R. (Eds.), Deep Carbon: Past to Present. Cambridge University
two gold mineralisation events at the New Celebration gold deposits in Western Press, Cambridge, UK, pp. 415–445. https://doi.org/10.1017/9781108677950.
Australia. Aust. J. Earth Sci. 61, 113–141. https://doi.org/10.1080/ Simpson, E.S., 1912. Natural gas in the Boulder Mines. In: Simpson, E.S., Gibson, C.G.
08120099.2014.879072. (Eds.), Contributions to the study of the geology and ore deposits of Kalgoorlie: East
Nivin, V.A., 2009. Molecular-mass distribution of saturated hydrocarbons in gas of the Coolgardie Goldfield Part 1, 42, pp. 158–160. Geol. Surv. Western Australia Bull.
Lovozerskii nepheline-syenite massif. Dokl. Earth Sci. 429, 1580–1582. https://doi. Sinev, M.Yu., 2003. Free radicals in catalytic oxidation of light alkanes: kinetic and
org/10.1134/S1028334X09090359. thermochemical aspects. J. Catal. 216, 468–476. https://doi.org/10.1016/S0021-
Nivin, V.A., 2019. Occurrence forms, composition, distribution, origin and potential 9517(02)00116-1.
hazard of natural hydrogen–hydrocarbon gases in ore deposits of the Khibiny and Sleep, N.H., Bird, D.K., 2007. Niches of the pre-photosynthetic biosphere and geologic
Lovozero Massifs: a review. Minerals 9, 535. https://doi.org/10.3390/min9090535. preservation of Earth’s earliest ecology. Geobiology 5, 101–117. https://doi.org/
Ohmoto, H., Rye, R.O., 1979. Isotopes of sulfur and carbon. In: Barnes, H.L. (Ed.), 10.1111/j.1472-4669.2007.00105.x.
Geochemistry of hydrothermal ore deposits. John Wiley and Sons, New York, Smith, N.J.P., Shephard, T.J., Styles, M.T., Williams, G.M., 2005. Hydrogen exploration:
pp. 509–567. a review of global hydrogen accumulations and implications for prospective areas in
Ojala, V.J., Groves, D.I., Ridley, J.R., 1995. Hydrogen isotope fractionation factors NW Europe. In: Doré, A.G., Vining, B.A. (Eds.), Petroleum Geology: North-West
between hydrous minerals and ore fluids at low temperatures: evidence from the Europe and Global Perspectives – Proceedings of the 6th Petroleum Geology
Granny Smith gold deposit, Western Australia. Mineral. Deposita 30, 328–331. Conference, Geological Society, London, Petroleum Geology Conference series, vol.
https://doi.org/10.1007/BF00196369. 6, pp. 349–358. https://doi.org/10.1144/0060349.
Oze, C., Sharma, M., 2005. Have olivine, will gas: serpentinization and the abiogenic Stack, S., Rabone, H., 2018. Cobar Management Pty Ltd: the CSA mine annual
production of methane on Mars. Geophys. Res. Lett. 32, L10203 https://doi.org/ environmental management report. AEMR 2017. http://www.csamine.com.au/u
10.1029/2005GL022691. ploads/51687/ufiles/CSA-AEMR_2017_-_Final_opt.pdf.
Parnell, J., Blamey, N., 2017. Hydrogen from radiolysis of aqueous fluid inclusions Suda, K., Ueno, Y., Yoshizaki, M., Nakamura, H., Kurokawa, K., Nishiyama, E.,
during diagenesis. Minerals 7, 130. https://doi.org/10.3390/min7080130. Yoshino, K., Hongoh, Y., Kawachi, K., Omori, S., Yamada, K., Yoshida, N.,
Polito, P.A., 1999. Exploration implications predicted by the distribution of Maruyama, S., 2014. Origin of methane in serpentinite-hosted hydrothermal
carbon–oxygen–hydrogen gases above and within the Junction gold deposit, systems: The CH4–H2–H2O hydrogen isotope systematics of the Hakuba Happo hot
Kambalda, Western Australia. PhD thesis, University of Adelaide, Adelaide spring. Earth Planet. Sci. Lett. 386, 112–125. https://doi.org/10.1016/j.
(unpubl.). http://hdl.handle.net/2440/19421. epsl.2013.11.001.
Polito, P.A., Bone, Y., Clarke, J.D.A., Mernagh, T.P., 2001. Compositional zoning of fluid Takai, K., Nakamura, K., Toki, T., Tsunogai, U., Miyazaki, M., Miyazaki, J., Hirayama, H.,
inclusions in the Archaean Junction gold deposit, Western Australia: a process of Nakagawa, S., Nunoura, T., Horikoshi, K., 2008. Cell proliferation at 122 ◦ C and
fluid – wall-rock interaction? Aust. J. Earth Sci. 48, 833–855. https://doi.org/ isotopically heavy CH4 production by a hyperthermophilic methanogen under high-
10.1046/j.1440-0952.2001.00903.x. pressure cultivation. Proc. Natl. Acad. Sci. U. S. A. 31, 10949–10954. www.pnas.org
Polito, P.A., Clarhe, J.D.A., Bone, Y., Viellenave, J., 2002. A CO2–O2–light /cgi/doi/10.1073/pnas.0712334105.
hydrocarbon–soil-gas anomaly above the Junction orogenic gold deposit: a potential, Tarnas, J.D., Mustard, J.F., Sherwood Lollar, B., Bramble, M.S., Cannon, K.M.,
alternative exploration technique. Geochem. Explor. Environ. Anal. 2, 333–344. Palumbo, A.M., Plesa, A.-C., 2018. Radiolytic H2 production on Noachian Mars:
https://doi.org/10.1144/1467-787302-035. implications for habitability and atmospheric warming. Earth Planet. Sci. Lett. 502,
Popa, M.E., Paul, D., Janssen, C., Röckmann, 2019. H2 clumped isotope measurements at 133–145. https://doi.org/10.1016/j.epsl.2018.09.001.
natural isotopic abundances. Rapid Commun. Mass Spectrom. 33, 239–251. https:// Tarnas, J.D., Mustard, J.F., Sherwood Lollar, B., Bramble, M.S., Cannon, K.M.,
doi.org/10.1002/rcm.8323. Palumbo, A.M., Plesa, A.-C., 2020. Corrigendum to “Radiolytic H2 production on
Porcelli, D., Ballentine, C.J., Wieler, R., 2002. An overview of noble gas geochemistry Noachian Mars: implications for habitability and atmospheric warming” [Earth
and cosmochemistry. Rev. Mineral. Geochem. 47, 1–19. https://doi.org/10.2138/ Planet.Sci. Lett. 502 (2018b) 133–145]. Earth Planet. Sci. Lett. 530, 115983. https://
rmg.2002.47.1. doi.org/10.1016/j.epsl.2019.115983.
Prinzhofer, A., Cissé, S.S.T., Diallo, A.B., 2018. Discovery of a large accumulation of Tassi, F., Vaselli, O., Capaccioni, B., Montegrossi, G., Barahona, F., Caprai, A., 2007.
natural hydrogen in Bourakebougou Mali. Int. J. Hydrog. Energy 43, 19315–19326. Scrubbing process and chemical equilibria controlling the composition of light
https://doi.org/10.1016/j.ijhydene.2018.08.193. hydrocarbons in natural gas discharges: an example from the geothermal fields of El
Proskurowski, G., Lilley, M.D., Seewald, J.S., Früh-Green, G.L., Olson, E.J., Lupton, J.E., Salvador. Geochem. Geophys. Geosyst. 8, Q05008 https://doi.org/10.1029/
Sylva, S.P., Kelley, D.S., 2008. Abiogenic hydrocarbon production at Lost City 2006GC001487.
hydrothermal field. Science 319, 604–607. https://doi.org/10.1126/ Thampi, K.R., Kiwi, J., Grätzel, M., 1987. Methanation and photo-methanation of carbon
science.1151194. dioxide at room temperature and atmospheric pressure. Nature 327, 506–508.
Ray, J.S., 2009. Carbon isotopic variations in fluid-deposited graphite: evidence for https://doi.org/10.1038/327506a0.
multicomponent Rayleigh isotopic fractionation. Int. Geol. Rev. 51, 45–57. https:// Thébaud, N., Miller, J., 2019. Orogenic gold deposits in the Yilgarn craton: a review.
doi.org/10.1080/00206810802625057. World Gold 2019 Conference 11–13 September, Perth, Australia. Australasian
Reeves, E.P., Fiebig, J., 2020. Abiotic synthesis of methane and organic compounds in Institute of Mining and Metallurgy Publication Series 6, 96–100.
Earth’s lithosphere. Elements 16, 25–31. https://doi.org/10.2138/ Thiagarajan, N., Kitchen, N., Xie, H., Ponton, C., Lawson, M., Formolo, M., Eiler, J.,
gselements.16.1.25. 2020a. Identifying thermogenic and microbial methane in deep water Gulf of Mexico
Reeves, E.P., Seewald, J.S., Sylva, S.P., 2012. Hydrogen isotope exchange between n- reservoirs. Geochim. Cosmochim. Acta 275, 188–208. https://doi.org/10.1016/j.
alkanes and water under hydrothermal conditions. Geochim. Cosmochim. Acta 77, gca.2020.02.016.
582–599. https://doi.org/10.1016/j.gca.2011.10.008. Thiagarajan, N., Xie, H., Ponton, C., Kitchen, N., Peterson, B., Lawson, M., Formolo, M.,
Ridley, J.R., Diamond, W., 2000. Fluid chemistry of orogenic lode gold deposits and Xiao, Y., Eiler, J., 2020b. Isotopic evidence for quasi-equilibrium chemistry in
implications for genetic models. In: Gold in 2000 (Hagemann, S.G. and Brown, P.E., thermally mature natural gases. Proc. Natl. Acad. Sci. U. S. A. 117, 3989–3995. www
Eds.), Gold in 2000. Rev. Econ. Geol. 13, 141–162. .pnas.org/cgi/doi/10.1073/pnas.1906507117.
Salier, B.P., Groves, D.I., McNaughton, N.J., Fletcher, I.R., 2005. Geochronological and Thiessen, O., Mørkved, P.T., Johansen, H., van Grass, G.W., 2009. Uncommon gases with
stable isotope evidence for widespread orogenic gold mineralization from a deep- high hydrogen concentrations found in well annuli on Troll A platform. In: Book of
seated fluid source at ca 2.65 Ga in the Laverton Gold Province, Western Australia. Abstracts the 24th International Meeting on Organic Geochemistry (IMOG 2009),
Econ. Geol. 100, 1363–1388. https://doi.org/10.2113/gsecongeo.100.7.1363. September 6–11, 2009, Bremen, Germany. Abstract for Oral-79, p. 92.
Schimmelmann, A., Sessions, A.L., Boreham, C.J., Edwards, D.S., Logan, G.A., Tissot, B.P., Welte, D.H., 1984. Petroleum Formation and Occurrence. Springer-Verlag,
Summons, R.E., 2004. D/H ratios in terrestrially sourced petroleum systems. Org. Berlin.
Geochem. 35, 1169–1195. https://doi.org/10.1016/j.orggeochem.2004.05.006. Truche, L., Joubert, G., Dargent, M., Martz, P., Cathelineau, T., Quirt, D., 2018. Clay
Schrenk, M.O., Brazelton, W.J., Lang, S.Q., 2013. Serpentinization, carbon, and deep life. minerals trap hydrogen in the Earth’s crust: evidence from the Cigar Lake uranium
Rev. Mineral. Geochem. 75, 575–606. https://doi.org/10.2138/rmg.2013.75.18. deposit, Athabasca. Earth Planet. Sci. Lett. 493, 186–197. https://doi.org/10.1016/j.
Schroder, I., de Caritat, P., Wallace, L., Trihey, J., Boreham, C., Tobin, J., English, P., epsl.2018.04.038.
Sohn, J., Palatty, P., Coghlan, R. Maher, Main, P., Thorose, M.D., Long, I., Byass, J., Truche, L., McCollom, T.M., Martinez, I., 2020. Hydrogen and abiotic hydrocarbons:
Jinadasa, N., 2020. Northern Australia Hydrogeochemical Survey: Final Data molecules that change the world. Elements 16, 13–18. https://doi.org/10.2138/
Release and Hydrogeochemical Atlas for Exploring for the Future. Record 2020/15. gselements.16.1.13.
Geoscience Australia, Canberra. https://doi.org/10.11636/Record.2020.015. Turner, J., Rosen, M., Milligan, N., Sklash, M., Townley, L., 1993. Groundwater recharge
Schulz, H., 1999. Short history and present trends of Fischer-Tropsch synthesis. Appl. studies in the Kalgoorlie region: results of research carried out as MERIWA Project
Catal. A 186, 3–12. https://doi.org/10.1016/S0926-860X(99)00160-X. No M146 at the CSIRO Division of Water Resources Perth, AGC - Woodward-Clyde
Schulz, H., Vein Steen, E., Claeys, M., 1994. Selectivity and mechanism of Fischer Pty Ltd and Department of Earth Sciences, University of Windsor, Ontario. Minerals
Tropsch synthesis with iron and cobalt catalysts. Surf. Stud. Sci. Catalysis 81, Research Institute of Western Australia, East Perth, WA. https://nla.gov.au/nla.
455–460. https://doi.org/10.1016/S0167-2991(08)63911-7. obj-641047866/view.
Sherwood Lollar, B., Onstott, T.C., Lacrampe-Couloume, G., Ballentine, C.J., 2014. The Vacquand, C., Deville, E., Beaumont, V., Guyot, F., Sissmann, O., Pillot, D., Srcilla, C.,
contribution of the Precambrian continental lithosphere to global H2 production. Prinzhofer, A., 2018. Reduced gas seepages in ophiolitic complexes: Evidences for
Nature 516, 379–382. https://doi.org/10.1038/nature14017.

22
C.J. Boreham et al. Chemical Geology 575 (2021) 120098

multiple origins of the H2-CH4-N2 gas mixtures. Geochim. Cosmochim. Acta 223, Shield environments. Chem. Geol. 530, 119322. https://doi.org/10.1016/j.
437–461. https://doi.org/10.1016/j.gca.2017.12.018. chemgeo.2019.119322.
Vitale Brovarone, A., Martinez, I., Elmaleh, A., Compagnoni, R., Chaditeau, C., Witt, W.K., Cassidy, K.F., Lu, Y.-J., Hagemann, S.G., 2020. The tectonic setting and
Ferraris, C., Esteve, I., 2017. Massive production of abiotic methane during evolution of the 2.7 Ga Kalgoorlie–Kurnalpi Rift, a world-class Archean gold
subduction evidenced in metamorphosed ophicarbonates from the Italian Alps. Nat. province. Mineral. Deposita 55, 601–631. https://doi.org/10.1007/s00126-017-
Commun. 8, 14134. https://doi.org/10.1038/ncomms14134. 0778-9.
Vovk, I.F., 1987. Radiolytic salt enrichment and brines in the crystalline basement of the Woolnough, W.G., 1934. Natural gas in Australia and New Guinea. Bull. Am. Assoc. Pet.
East European Platform. In: Fritz, P., Frape, S.K. (Eds.), Saline Water and Gases in Geol. 18, 226–242. https://doi.org/10.1306/3D932C06-16B1-11D7-
Crystalline Rocks, Geological Association of Canada Special Paper, vol. 33, 8645000102C1865D.
pp. 197–210. Worman, S.L., Pratson, L.F., Karson, J.A., Klein, E.M., 2016. Global rate and distribution
Walshe, J., Neumayr, P., Halley, S., 2006. Fluids in the system. In: Walshe, J., of H2 gas produced by serpentinization within oceanic lithosphere. Geophys. Res.
Neumayr, P., Petersen, K. (Eds.), Scale-Integrated, Architectural and Geodynamic Lett. 43, 6435–6443. https://doi.org/10.1002/2016GL069066.
Controls on Alteration and Geochemistry of Gold Systems in the Eastern Goldfields Xiao, Y., 2001. Modeling the kinetics and mechanisms of petroleum and natural gas
Province, Yilgarn Craton. Minerals and Energy Research Institute OF Western generation: a first principles approach. Rev. Mineral. Geochem. 42, 383–436.
Australia, Project M358 Report, 256, pp. 29–72. https://doi.org/10.2138/rmg.2001.42.11.
Ward, L.K., 1917. Report on the prospects of obtaining supplies of Petroleum by Boring Young, E.D., 2019. A two-dimensional perspective on CH4-isotope clumping. In:
in the vicinity of Robe and elsewhere in the South-Eastern portion of South Australia. Orcutt, B.N., Daneil, I., Dasgupta, R. (Eds.), Deep Carbon: Past to Present. Cambridge
A Review of Mining Operations in the State of South Australia during the half-year University Press, Cambridge, pp. 388–414.
ended December 31st, 1916 25, 45–53. Zeng, H., Li, J., Huo, Q., 2013. A review of alkane gas geochemistry in the
Ward, L.K., 1932. The search for oil – notes by the Government geologist. South Xujiaweizifault-depression, Songliao Basin. Mar. Pet. Geol. 43, 284–296. https://doi.
Australian Department of Mines Mining Review for the half-year ended December org/10.1016/j.marpetgeo.2013.01.009.
31st 1931. 55, pp. 39–44. Zgonnik, V., 2020. The occurrence and geoscience of natural hydrogen: a comprehensive
Ward, L.K., 1933. Inflammable gases occluded in the Pre-Paleozoic rocks of South review. Earth Sci. Rev. 203, 103140. https://doi.org/10.1016/j.
Australia. Trans. Proc. R Soc. South Austral. 57, 42–47. earscirev.2020.103140.
Ward, L.K., 1944. The search for oil in South Australia. Geol. Surv. South Australia Bull. Zhang, S., Mi, J., He, K., 2013. Synthesis of hydrocarbon gases from four different carbon
22, 41p. sources and hydrogen gas using a gold-tube system by Fischer–Tropsch method.
Warr, O., Giunta, T., Ballentine, C.J., Sherwood Lollar, B., 2019. Mechanisms and rates of Chem. Geol. 349–350, 27–35. https://doi.org/10.1016/j.chemgeo.2013.03.016.
4
He, 40Ar, and H2 production and accumulation in fracture fluids in Precambrian

23