Figures and Tables

top

Introduction

The 3.49 Ga Dresser Formation in the Pilbara Craton, Western Australia has been widely considered to preserve palaeontological and geochemical evidence of Earth’s earliest life (e.g., Buick et al., 1981

Buick, R., Dunlop, J.S.R., Groves, D.I. (1981) Stromatolite recognition in ancient rocks: an appraisal of irregularly laminated structures in an Early Archaean chert-barite unit from North Pole, Western Australia. Alcheringa 5, 161–181.

; Ueno et al., 2006

Ueno, Y., Yamada, K., Yoshida, N., Maruyama, S., Isozaki, Y. (2006) Evidence from fluid inclusions for microbial methanogenesis in the early Archaean era. Nature 440, 516–519.

; Shen et al., 2009

Shen, Y., Farquhar, J., Masterson, A., Kaufman, A.J., Buick, R. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth and Planetary Science Letters 279, 383–391.

; Noffke et al., 2013

Noffke, N., Christian, D., Wacey, D., Hazen, R.M. (2013) Microbially Induced Sedimentary Structures Recording an Ancient Ecosystem in the ca. 3.48 Billion-Year-Old Dresser Formation, Pilbara, Western Australia. Astrobiology 13, 1103–1124.

). The multiple sulfur isotopes of sulfate and sulfide in the Dresser Formation have been measured by both bulk and in situ analytical methods. Results of bulk methods support microbial sulfate reduction (MSR; Ueno et al., 2008

Ueno, Y., Ono, S., Rumble, D., Maruyama, S. (2008) Quadruple sulfur isotope analysis of ca. 3.5 Ga Dresser Formation: New evidence for microbial sulfate reduction in the early Archean. Geochimica et Cosmochimica Acta 72, 5675–5691.

; Shen et al., 2009

Shen, Y., Farquhar, J., Masterson, A., Kaufman, A.J., Buick, R. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth and Planetary Science Letters 279, 383–391.

) while in situ measurements support elemental sulfur disproportionation (Philippot et al., 2007

Philippot, P., Van Zuilen, M., Lepot, K., Thomazo, C., Farquhar, J., Van Kranendonk, M.J. (2007) Early Archaean Microorganisms Preferred Elemental Sulfur, Not Sulfate. Science 317, 1534–1537.

). The differences in these data sets are profound. Although recent in situ studies have confirmed separately the δ³⁴S–Δ³³S positively correlated Δ³³S-negative (Baumgartner et al., 2020

Baumgartner, R.J., Caruso, S., Fiorentini, M.L., Van Kranendonk, M.J., Martin, L., Jeon, H., Pagès, A., Wacey, D. (2020) Sulfidization of 3.48 billion-year-old stromatolites of the Dresser Formation, Pilbara Craton: Constraints from in-situ sulfur isotope analysis of pyrite. Chemical Geology 538, 119488.

) and Δ³³S-positive (Wacey et al., 2015

Wacey, D., Noffke, N., Cliff, J., Barley, M.E., Farquhar, J. (2015) Micro-scale quadruple sulfur isotope analysis of pyrite from the ∼3480 Ma Dresser Formation: New insights into sulfur cycling on the early Earth. Precambrian Research 258, 24–35.

) pyrites, the relationship between the two groups remains unknown.

The chert-barite unit of Dresser Formation was generated by a complex hydrothermal system, thus the pyrites are likely to preserve multiple generations resulting from multiple stages of hydrothermal events (Van Kranendonk et al., 2008

Van Kranendonk, M.J., Philippot, P., Lepot, K., Bodorkos, S., Pirajno, F. (2008) Geological setting of Earth’s oldest fossils in the ca. 3.5 Ga Dresser Formation, Pilbara Craton, Western Australia. Precambrian Research 167, 93–124.

). This study established the generations of pyrite formation, and then measured the multiple sulfur isotopes in situ. We exploited the in situ capabilities of SHRIMP-SI allowing ³⁶S measurements in pyrite with a precision at a similar level to conventional bulk methods. In this way, we uniquely identified the sulfur isotopic composition of each generation of pyrites, and provided constraints on their formation pathways.

top

Geological Setting

The 3.49 Ga (Tessalina et al., 2010

Tessalina, S.G., Bourdon, B., Van Kranendonk, M., Birck, J.L., Philippot, P. (2010) Influence of Hadean crust evident in basalts and cherts from the Pilbara Craton. Nature Geoscience 3, 214–217.

) Dresser Formation crops out as a ring of the North Pole Dome in the Pilbara Craton. It consists of metabasalt interlayered with three chert horizons (Ueno et al., 2008

Ueno, Y., Ono, S., Rumble, D., Maruyama, S. (2008) Quadruple sulfur isotope analysis of ca. 3.5 Ga Dresser Formation: New evidence for microbial sulfate reduction in the early Archean. Geochimica et Cosmochimica Acta 72, 5675–5691.

). The lowermost chert hosts abundant barite, referred to as the chert-barite unit, and is composed of chert, barite, carbonate, sandstone and conglomerate. Below this unit are thousands of silica ± sulfides (pyrite and sphalerite) ± organic matter veins and barite veins intruding and transecting the underlying komatiitic basalt. The chert-barite unit and the veins are generally considered to be formed in a volcanic caldera, with sedimentation accompanied by volcanism, hydrothermal circulation and precipitation, and tectonic activity (Van Kranendonk et al., 2008

Van Kranendonk, M.J., Philippot, P., Lepot, K., Bodorkos, S., Pirajno, F. (2008) Geological setting of Earth’s oldest fossils in the ca. 3.5 Ga Dresser Formation, Pilbara Craton, Western Australia. Precambrian Research 167, 93–124.

). The hydrothermal fluids ascended along growth faults and deposited as concordant layers in a feeder-deposit manner. Detailed geological setting is in Supplementary Information (SI).

top

Samples and Methods

The investigated samples are from two drill cores: PDP2b and PDP2c. Detailed sample descriptions are in Table S-1. The drill core sections were made into polished mounts, which were immersed in sodium hypochlorite solution (8–12.5 % NaOCl) to reveal the internal textures of pyrites. The principle of pyrite etching is in SI. Appropriate pyrites were selected for subsequent BSE (Back Scattered Electron) imaging, EDS (Energy Dispersive X-ray Spectroscopy) and Raman spectroscopy analyses. Then the pyrites were measured for multiple sulfur isotopes using SHRIMP-SI. Detailed methods are in SI.

top

Results

Pyrites occur as tiny grains aligned in an array within barite (Py-i), pyrite laminae (Py-ii), individual grains (IG) or clusters associated with silica veinlets/beads in barite (within veinlet, Py-iii; on the wall of veinlet, Py-iv; adjacent to both barite and chert, Py-v), IG in barite with silica veinlets/beads (Py-vi), IG in chert with barite crystal fans (Py-vii), and IG in silica-barite-carbonate veins in hydrothermally altered komatiitic basalt (Py-viii) (Fig. S-2). Some pyrites show core-rim textures (portions of Py-i to vi and Py-viii) while the others are homogeneous (Figs. 1, S-3). The sulfur isotopic composition exhibits evident dependence on the internal texture and petrography of pyrite. The cores show a tendency towards positive δ³⁴S and Δ³³S, whereas the rims have more negative δ³⁴S and Δ³³S (Fig. S-8). Py-ii are mostly core-rim textured and display both negative and positive δ³⁴S, Δ³³S and Δ³⁶S, whereas Py-iii to vi are characterised by negative δ³⁴S and Δ³³S and positive Δ³⁶S, with only a few exceptions. Py-vii show characteristic positive δ³⁴S and Δ³³S, and Py-viii exhibit negative Δ³³S and positive δ³⁴S (Table 1, Fig. S-9). Detailed petrography, internal texture, and sulfur isotopic compositions are in SI.

Full size image

Pyrite type	Description		δ34S (‰)	Δ33S (‰)	Δ36S (‰)
Py-i	Pyrites occur as tiny grains aligned in an array within barite.
Py-ii	Pyrite laminae.		−10.40 to 2.12	−1.19 to 1.46	−2.14 to 1.11
Py-iii	Individual grains or clusters associated with silica veinlets/beads in barite	within veinlet	−21.45 to −0.79	−1.38 to 1.50	−0.31 to 1.87
Py-iv		on the wall of veinlet
Py-v		adjacent to both barite and chert
Py-vi	Individual grains in barite with silica veinlets/beads		−18.99 to −1.54	−1.11 to 0.10	−0.16 to 1.20
Py-vii	Individual grains in chert with barite crystal fans		2.81 to 5.09	2.52 to 3.62	−2.92 to −2.63
Py-viii	Individual grains in silica-barite-carbonate vein in hydrothermally altered komatiitic basalt		1.67 to 2.39	−1.01 to −0.67

top

Discussion

The core-rim relationship combined with sulfur isotopic composition reveals two main generations (Δ³³S-positive, euhedral to subhedral, locally porous G1 and Δ³³S-negative, anhedral G2) in the core-rim textured pyrites. The majority have Δ³³S-positive cores and Δ³³S-negative rims, indicating that Δ³³S-positive precede Δ³³S-negative pyrites. Δ³³S-0 (close to 0, considering uncertainties) pyrites are widespread, and can be formed via the sulfide pathway (Fe²⁺ + S^2– = [FeS], [FeS] + H₂S = FeS₂ + H₂; Rickard and Luther, 1997

Rickard, D., Luther, G.W. (1997) Kinetics of pyrite formation by the H₂S oxidation of iron (II) monosulfide in aqueous solutions between 25 and 125 °C: The mechanism. Geochimica et Cosmochimica Acta 61, 135–147.

) with magmatic H₂S abundant in the Dresser hydrothermal system throughout the multiple stages of hydrothermal events (Ueno et al., 2008

Ueno, Y., Ono, S., Rumble, D., Maruyama, S. (2008) Quadruple sulfur isotope analysis of ca. 3.5 Ga Dresser Formation: New evidence for microbial sulfate reduction in the early Archean. Geochimica et Cosmochimica Acta 72, 5675–5691.

and references therein). Δ³³S-0 pyrites can be synchronous with Δ³³S-positive and Δ³³S-negative pyrites, because both Δ³³S-positive and Δ³³S-negative pyrites show evident positive correlations between δ³⁴S and Δ³³S (Fig. 2b), indicating two separate mixing trends between two respective end members, with one being Δ³³S-0. Minor pyrites show Δ³³S-negative cores and rims, suggesting two stages of G2.

Full size image

Some Py-ii show two additional growth zones prior to G1 and G2 (Fig. 2a). The earlier zones (G1′) are tiny porous grains, the Δ³³S of which is beyond the resolution ability of SHRIMP spots. However, the textures of these pyrites are similar to the cores of core-rim textured Py-i (Fig. 1a) that have been shown to be Δ³³S-positive (Philippot et al., 2007

Philippot, P., Van Zuilen, M., Lepot, K., Thomazo, C., Farquhar, J., Van Kranendonk, M.J. (2007) Early Archaean Microorganisms Preferred Elemental Sulfur, Not Sulfate. Science 317, 1534–1537.

). The later zones (G2′) are partly oscillatory zoned and porous, and these have been measured by Baumgartner et al. (2020)

Baumgartner, R.J., Caruso, S., Fiorentini, M.L., Van Kranendonk, M.J., Martin, L., Jeon, H., Pagès, A., Wacey, D. (2020) Sulfidization of 3.48 billion-year-old stromatolites of the Dresser Formation, Pilbara Craton: Constraints from in-situ sulfur isotope analysis of pyrite. Chemical Geology 538, 119488.

and are mostly Δ³³S-negative. One Py-ii (Fig. S-3d) has a Δ³³S-negative core (oscillatory zoned) and Δ³³S-positive rim, confirming G2′ precedes G1. Collectively, four generations are recorded in core-rim textured pyrites: Δ³³S-positive G1′, Δ³³S-negative G2′ (Baumgartner et al., 2020

Baumgartner, R.J., Caruso, S., Fiorentini, M.L., Van Kranendonk, M.J., Martin, L., Jeon, H., Pagès, A., Wacey, D. (2020) Sulfidization of 3.48 billion-year-old stromatolites of the Dresser Formation, Pilbara Craton: Constraints from in-situ sulfur isotope analysis of pyrite. Chemical Geology 538, 119488.

), Δ³³S-positive G1 and Δ³³S-negative G2 (this study). The multiple generations of pyrites in the same sample can be attributed to multiple pulses of hydrothermal fluids, with the later generation overgrown on the earlier generation.

The pyrites with no internal texture (Py-ii to vi) have the same petrography as the core-rim textured counterparts, and show similar multiple sulfur isotope systematics to the cores and rims, thus the homogeneous pyrites (Py-ii to vi) are also categorised into the four generations based on Δ³³S and internal texture. The homogeneous Py-ii to vi lack the characteristic textures of G1′ and G2′, thus are all classified to G1 and G2. Δ³³S-positive Py-vii are Ni-free and not porous, distinct from G1′ while similar to G1. The negative Δ³³S and positive δ³⁴S of Py-viii are similar to some G2′.

Photochemical reactions of gaseous sulfur species have been widely proposed to produce S-MIF, and the products are Δ³³S-positive elemental sulfur and Δ³³S-negative sulfate (Ono, 2017

Ono, S. (2017) Photochemistry of Sulfur Dioxide and the Origin of Mass-Independent Isotope Fractionation in Earth’s Atmosphere. Annual Review of Earth and Planetary Sciences 45, 301–329.

and references therein). Δ³³S-positive G1 and Δ³³S-negative barites yield a line of Δ³⁶S = −0.98Δ³³S + 0.05 (Fig. S-10a). The slope is close to that for products of SO₂ photolysis under 193 nm UV (Farquhar et al., 2001

Farquhar, J., Savarino, J., Airieau, S., Thiemens, M.H. (2001) Observation of wavelength-sensitive mass-independent sulfur isotope effects during SO₂ photolysis: Implications for the early atmosphere. Journal of Geophysical Research 106, 32829–32839.

). Thus the photochemical products sulfate and elemental sulfur are likely to have been mostly sequestered in barites and G1 pyrites via reactions with Ba²⁺-rich fluids and Fe²⁺-, S²⁻-rich fluids, respectively, with the magnitudes diluted by magmatic and/or Δ³³S-opposite sign sulfur species in the atmosphere/water mass (detailed explanation in SI).

G1 yield a line of Δ³⁶S = −0.86Δ³³S − 0.20 (Fig. S-10b) close to the Archean Reference Array (ARA). Some G1 show slight deviations from ARA, both upwards and downwards (one analysis exhibits a downward offset of >1 ‰), suggesting possible microbial elemental sulfur disproportionation or sulfate reduction, respectively (Johnston et al., 2007

Johnston, D.T., Farquhar, J., Canfield, D.E. (2007) Sulfur isotope insights into microbial sulfate reduction: When microbes meet models. Geochimica et Cosmochimica Acta 71, 3929–3947.

). However, neither is likely since microbial activities prefer ³²S over ³⁴S (i.e.  negative δ³⁴S) while G1 have positive δ³⁴S.

G2 are most likely formed via the sulfide pathway, with S^2– derived from sulfate reduction. In addition to the sulfate in fluids and water mass, barite is also a possible sulfate source. G2 are closely associated with barites (e.g., G2 occur within barites, in the silica veinlets of barites, or contain barite residue), and have similar Δ³³S despite smaller magnitudes. The solubility of barite can be increased by NaCl (Shi et al., 2012

Shi, W., Kan, A.T., Fan, C., Tomson, M.B. (2012) Solubility of Barite up to 250 °C and 1500 bar in up to 6 m NaCl Solution. Industrial and Engineering Chemistry Research 51, 3119–3128.

) and silica (Cui et al., 2019

Cui, H., Zhong, R., Xie, Y., Yuan, X., Liu, W., Brugger, J., Yu, C. (2019) Forming sulfate- and REE-rich fluids in the presence of quartz. Geology 48, 145–148.

) that are present in the Dresser hydrothermal fluids (Harris et al., 2009

Harris, A.C., White, N.C., McPhie, J., Bull, S.W., Line, M.A., Skrzeczynski, R., Mernagh, T.P., Tosdal, R.M. (2009) Early Archean Hot Springs above Epithermal Veins, North Pole, Western Australia: New Insights from Fluid Inclusion Microanalysis. Economic Geology 104, 793–814.

). Therefore, G2 can be derived from barite reduction, with Fe²⁺ and reductants brought in by hydrothermal fluids penetrated into barites as indicated by the widespread silica veinlets in barites. Δ³³S of G2 can have been diluted by (1) the involvement of magmatic H₂S/HS^–/S^2– during pyrite formation via the sulfide pathway as indicated by the positive correlation between negative δ³⁴S and Δ³³S (Fig. 2b), and (2) the positive Δ³³S induced by sulfate reduction (thermochemical sulfate reduction, TSR; Oduro et al., 2011

Oduro, H., Harms, B., Sintim, H.O., Kaufman, A.J., Cody, G., Farquhar, J. (2011) Evidence of magnetic isotope effects during thermochemical sulfate reduction. Proceedings of the National Academy of Sciences 108, 17635–17638.

; MSR, Johnston et al., 2007

Johnston, D.T., Farquhar, J., Canfield, D.E. (2007) Sulfur isotope insights into microbial sulfate reduction: When microbes meet models. Geochimica et Cosmochimica Acta 71, 3929–3947.

).

The essential effects of MSR are the depletions in ³⁴S and downward shifts in Δ³⁶S/Δ³³S (i.e.  lower (Δ³³S + Δ³⁶S)² values) relative to the initial sulfate (Johnston et al., 2007

Johnston, D.T., Farquhar, J., Canfield, D.E. (2007) Sulfur isotope insights into microbial sulfate reduction: When microbes meet models. Geochimica et Cosmochimica Acta 71, 3929–3947.

). Previous bulk analyses prefer MSR based on such signatures (Ueno et al., 2008

Ueno, Y., Ono, S., Rumble, D., Maruyama, S. (2008) Quadruple sulfur isotope analysis of ca. 3.5 Ga Dresser Formation: New evidence for microbial sulfate reduction in the early Archean. Geochimica et Cosmochimica Acta 72, 5675–5691.

; Shen et al., 2009

Shen, Y., Farquhar, J., Masterson, A., Kaufman, A.J., Buick, R. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth and Planetary Science Letters 279, 383–391.

). Admittedly, some G2 show negative (Δ³³S + Δ³⁶S) (−0.03 to −0.67 ‰) and δ³⁴S fractionation of −7 to −26 ‰ from sulfate (the composition of sulfate is represented by that of barite), with portions plotted within the MSR area defined by culture experiments with sulfate reducers (2–36 °C; Figs. 2c, 3). However, the negative (Δ³³S + Δ³⁶S) of the MSR area are relative to (Δ³³S + Δ³⁶S) of 0 for initial sulfate. Bulk barite data are homogeneous with a restricted range of (Δ³³S + Δ³⁶S) near 0 (Ueno et al., 2008

Ueno, Y., Ono, S., Rumble, D., Maruyama, S. (2008) Quadruple sulfur isotope analysis of ca. 3.5 Ga Dresser Formation: New evidence for microbial sulfate reduction in the early Archean. Geochimica et Cosmochimica Acta 72, 5675–5691.

; Shen et al., 2009

Shen, Y., Farquhar, J., Masterson, A., Kaufman, A.J., Buick, R. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth and Planetary Science Letters 279, 383–391.

), while in situ analyses results are variable with (Δ³³S + Δ³⁶S) ranging from −0.60 to 0.92 ‰ (Muller et al., 2016

Muller, É., Philippot, P., Rollion-Bard, C., Cartigny, P. (2016) Multiple sulfur-isotope signatures in Archean sulfates and their implications for the chemistry and dynamics of the early atmosphere. Proceedings of the National Academy of Sciences 113, 7432–7437.

; Fig. 3). The (Δ³³S + Δ³⁶S) range of G2 is similar to that of barites (in situ data, Fig. 3), and no significant downward shifts of G2 relative to initial barite are observed. Additionally, barites and G2 pyrites yield a line of Δ³⁶S = −0.99Δ³³S − 0.02 close to ARA while distinct from the Biological Fractionation Line (BFL; Fig. S-10c). Therefore, MSR signal in G2 is weak, although we do not exclude its presence because two data points of Shen et al. (2009)

Shen, Y., Farquhar, J., Masterson, A., Kaufman, A.J., Buick, R. (2009) Evaluating the role of microbial sulfate reduction in the early Archean using quadruple isotope systematics. Earth and Planetary Science Letters 279, 383–391.

show negative (Δ³³S + Δ³⁶S) even relative to the lowest (Δ³³S + Δ³⁶S) value of in situ barite data.

Full size image

Furthermore, some G2 show positive (Δ³³S + Δ³⁶S) with larger magnitudes of positive shifts in Δ³³S than negative shifts in Δ³⁶S, which has not been observed in MSR but has been discovered in TSR (Figs. 2c, 3). TSR can induce positive and negative (Δ³³S + Δ³⁶S) depending on reductant and temperature (Oduro et al., 2011

Oduro, H., Harms, B., Sintim, H.O., Kaufman, A.J., Cody, G., Farquhar, J. (2011) Evidence of magnetic isotope effects during thermochemical sulfate reduction. Proceedings of the National Academy of Sciences 108, 17635–17638.

), in addition to significant depletions in δ³⁴S relative to initial sulfate (Ohmoto and Goldhaber, 1997

Ohmoto, H., Goldhaber, M.B. (1997) Sulfur and carbon isotopes. In: Barnes, H.L. (Ed.) Geochemistry of Hydrothermal Ore Deposits. Wiley, New York, 517–611.

). Therefore, the S²⁻ of G2 can also be derived from TSR apart from MSR. TSR could be in operation 3.49 Ga ago in the Dresser hydrothermal system. This is consistent with the highest temperature of 110 °C where sulfate reducers (discovered so far) can live (Jørgensen et al., 1992

Jørgensen, B.B., Isaksen, M.F., Jannasch, H.W. (1992) Bacterial Sulfate Reduction Above 100 °C in Deep-Sea Hydrothermal Vent Sediments. Science 258, 1756–1757.

) while the temperature of the hydrothermal fluids associated with the Dresser Formation ranges from 300 °C at depth to 120 °C near the palaeosurface (Harris et al., 2009

Harris, A.C., White, N.C., McPhie, J., Bull, S.W., Line, M.A., Skrzeczynski, R., Mernagh, T.P., Tosdal, R.M. (2009) Early Archean Hot Springs above Epithermal Veins, North Pole, Western Australia: New Insights from Fluid Inclusion Microanalysis. Economic Geology 104, 793–814.

).

top

Conclusions

A combination of sodium hypochlorite etching, BSE imaging and multiple sulfur isotopes reveals two main generations of pyrites associated with the barites of the 3.49 Ga Dresser Formation: Δ³³S-positive G1 and δ³⁴S- and Δ³³S-negative G2. Diluted sulfate and elemental sulfur produced by photochemical reactions were sequestered in barites and G1, respectively. The positive δ³⁴S of G1 indicates an abiological genesis. G2 were formed via the sulfide pathway, with the sulfur derived from magmatic H₂S and S²⁻ produced by thermochemical and possible microbial sulfate reduction.
²(Δ³³S + Δ³⁶S) is derived from Δ³⁶S ≈ −Δ³³S (ARA), i.e.  Δ³³S + Δ³⁶S is 0 for data on ARA, but positive and negative for data above and below ARA, respectively.

top

Acknowledgements

We sincerely thank the Perth Core Library for approving the access to drill cores and publishing the results. This work was supported by the Australian Research Council DP140103393 to TRI. The constructive comments from Dr. Boswell Wing and two anonymous reviewers are sincerely appreciated, which have improved this manuscript significantly.

Editor: Karim Benzerara

top

References

top

Supplementary Information

The Supplementary Information includes:

• 1. Geological Setting
• 2. Samples
• 3. Methods
• 4. Results
• 5. The Dependence of Multiple Sulfur Isotopic Compositions on Pyrite Texture and Petrography
• 6. The Dilution of S-MIF in Original Photochemical Products
• 7. Pyrites with Positive Δ³³S and Negative δ³⁴S
• Tables S-1 to S-6
• Figures S-1 to S-11
• Supplementary Information References

Download Tables S-4 to S-6 (Excel).

Download the Supplementary Information (PDF).