Comment on "Iron Isotope Constraints on the Archean and Paleoproterozoic Ocean Redox State"
Kosei E. Yamaguchi
Hiroshi Ohmoto
Rouxel
et al. (Reports, 18 February 2005, p. 1088) argued that
changes in the iron isotopic composition of sedimentary sulfides
reflect changes in the oxidation state of the atmosphere-ocean
system between 2.3 and 1.8 million years ago. We show that misinterpretations
of the origins of these minerals undermine their conclusions.
Agency for Marine-Earth Science and Technology (JAMSTEC),
2-15 Natsushima, Yokosuka 237-0061, Japan.
NASA Astrobiology
Institute, USA.
NASA Astrobiology
Institute, USA.
Astrobiology Research Center and
Department of Geosciences, The Pennsylvania State University,
University Park, PA 16802, USA.
To whom correspondence should be addressed. E-mail: kosei{at}jamstec.go.jp
Rouxel et al. (1) recently reported the iron isotopic composition (
56Fe values) of Fe sulfides (diagenetic pyrites) in black shales and showed that the values fall in distinct ranges depending on the geologic age of the shale: mostly negative [–3.5 to 0.5 per mil ()] before 
2.3 billion years ago (Ga); mostly positive (–0.5 to +1.0) between 2.3 and 1.7 Ga; and near zero (–0.5 to +0.2) after 1.5 Ga. Based on these stages and simple models for isotopic fractionation during Fe mineralization, they concluded that: (i) Diagenetic pyrites older than 1.7 Ga used the dissolved Fe2+ in ocean water, whereas younger pyrites used the Fe from Fe silicates and Fe oxides buried in the sediments. (ii) The oceans were Fe2+-rich and anoxic, and the atmosphere was anoxic before 2.3 Ga. (iii) The oceans were stratified (Fe2+-poor, oxygenated shallow water and Fe2+-rich, anoxic deep water), but the atmosphere was oxic during 2.3 to 1.7 Ga. (iv) Both the oceans and the atmosphere have been fully oxygenated since 1.5 Ga. We point to several problems in Rouxel et al.'s interpretations linking the Fe isotope data to the redox history of the oceans and atmosphere.
First, Rouxel et al. (1) suggested that the 
56Fe values of Fe2+ that formed diagenetic pyrites became negative (–3.5 to 0) as a result of the precipitation of very large amounts (50% to >90%) of Fe2+ as 56Fe-enriched Fe oxides (hematite and magnetite) in banded iron formations (BIFs). They noted that 50% precipitation of Fe oxides is similar to the estimates of Fe sink in BIFs based on phosphorus adsorption (2). However, more than 90% removal of Fe (1) is far beyond this estimate and is therefore unrealistic.
The second shortcoming of the Rouxel et al. model concerns the age relationships between the black shales and Fe oxides. The analyzed black shales (from the Jeerinah, Mt. McRae, and Gamohaan Formations) deposited in basins before, not after, the depositions of large BIFs (i.e., the Marra Mamba, Brockman, and Kuruman IFs, respectively) (3, 4). Therefore, BIF deposition cannot be the reason for negative and variable 56Fe values for pyrites in the older black shales.
Third, Kakegawa et al. (5) reported various features of extensive dissolution/reprecipitation of pyrites during diagenesis of the Mt. McRae shales studied in (1). Based on fluid inclusion analyses, Haruna et al. (6) concluded that the fluids involved in these processes reached temperatures between 150°C and 200°C. Variable and negative 56Fe values of diagenetic pyrites in the Archean black shales were, therefore, most likely caused by the dissolution/reprecipitation of pyrite (i.e., redox recycling of Fe), which was facilitated by locally discharged submarine hydrothermal fluids. This process must have greatly affected Fe isotope compositions of the pyrite crystals, which discourages the notion of Rouxel et al. that pyrite in the black shales recorded the Fe isotope signature of the global ocean at the time of black shale deposition. The 56Fe variations in diagenetic/hydrothermal pyrites cannot correlate with the 56Fe variation of Fe2+ in the overlying seawater.
Furthermore, diagenetic pyrites with negative 56Fe values (as low as –2) are actually common in modern marine sediments that accumulated under an oxic to suboxic water column such as the Monterey, Santa Barbara, and Santa Monica Basins and the Baja Mats (7, 8). Yamaguchi et al. (9) previously noted the similarity in the 56Fe values of pyrite between modern and Archean sedimentary rocks [figure 11 in (9)]. Both the Fe and S isotopic compositions of syngenetic and diagenetic pyrite crystals in black shales likely represent only the local geochemical conditions of sedimentary basins and cannot be extrapolated to the global ocean.
Fourth, Rouxel et al. (1) propose that the 56Fe values of pyrites in the 2.3 to 1.7 Ga black shales became positive because they attained isotopic equilibrium with Fe2+ in the ocean water (56Fe = 0). This is a poor argument, considering that they admitted in note 20 in (1) that "the fractionation of pyrite is poorly constrained from –0.3 to 1.0 relative to dissolved Fe(II)."
To correctly interpret the Fe isotope data of pyrite and to better constrain Fe isotope geochemistry during sedimentary diagenesis, further experimental studies must be carried out to determine the equilibrium/kinetic fractionation factors among various Fe phases. Specific attention should be directed to observing Fe2+aq, FeS, and FeS2 in various pathways during sulfide formation (i.e., Fe3+
Fe2+FeSFeS2) involving sulfate reduction, sulfide oxidation, and sulfur disproportionation reactions by microorganisms.
References
- 1. O. J. Rouxel, A. Bekker, K. J. Edwards, Science 307, 1088 (2005).[Abstract/Free Full Text]
- 2. C. J. Bjerrum, D. E. Canfield, Nature 417, 159 (2002). [CrossRef] [Medline]
- 3. Geological Survey of Western Australia, Geology and Mineral Resources of Western Australia: Western Australia Geological Survey, Memoir 3 (Geological Survey of Western Australia, Perth, 1990).
- 4. South African Committee for Stratigraphy (SACS), Stratigraphy of South Africa, Part I: Lithostratigraphy of the Republic of South Africa, South West Africa/Namibia and the Republics of Bophuthatswana, Transkei, and Venda (Comp. L. E. Kent), Geological Society of South Africa, Handbook 9 (Geological Society of South Africa, Pretoria, 1980).
- 5. T. Kakegawa, H. Kawai, H. Ohmoto, Geochim. Cosmochim. Acta 62, 3205 (1999).
- 6. M. Haruna et al., Resour. Geol. 53, 75 (2003).
- 7. S. Severmann et al., Eos 85, Fall Mtg. Suppl., abstract V51A-0521 (2004).
- 8. S. Severmann, J. McManus, C. M. Johnson, B. L. Beard, Eos 84, Ocean Sci. Mtg. Suppl., abstract OS31L-09 (2003).
- 9. K. E. Yamaguchi, C. M. Johnson, B. L. Beard, H. Ohmoto, Chem. Geol. 218, 135 (2005).
Received for publication 1 August 2005. Accepted for publication 7 December 2005.
