Biological Fe oxidation controlled deposition of banded iron formation in the ca. 3770 Ma Isua Supracrustal Belt (West Greenland)

Abstract

The redox balance of the Archean atmosphere–ocean system is among the most significant uncertainties in our understanding of the earliest history of Earth's surface zone. Most workers agree that oxygen did not constitute a significant proportion of the atmosphere until after ca. 2.45Ga, after the Great Oxidation Event, but there is less agreement on when O2 production began, and how this may have been consumed by reduced species such as Fe(II) in the oceans. The Fe redox cycle through time has been traced using banded iron formations (BIFs), and Fe isotopes are increasingly used to constrain the conditions of Earth's paleoenvironments, including the pathways of formation of BIFs. Iron isotope analyses of BIFs from the 3.7 to 3.8Ga Isua Supracrustal Belt (ISB), obtained by micro-sampling of magnetite-rich layers and conventional analysis, as well as by in situ femtosecond laser ablation (fs-LA-ICP-MS), indicate a consistently narrow range of non-zero δ56Fe values. Analysis of magnetite by fs-LA-ICP-MS allows for precise and accurate micron-scale analyses without the problems of orientation effects that are associated with secondary ion mass spectrometry (SIMS) analyses. Magnetite δ56Fe values range from +0.4‰ to +1.1‰ among different bands, but within individual layers magnetite grains are mostly homogeneous. Although these BIFs have been metamorphosed to amphibolite-facies, the metamorphism can neither explain the range in Fe isotope compositions across bands, nor that between hand samples. The isotopic compositions therefore reflect “primary”, low-temperature sedimentary values. The positive δ56Fe values measured from the ISB magnetites are best explained by deposition of Fe(III)-oxides produced by partial oxidation of Fe(II)-rich ocean water. A dispersion/reaction model, which accounts for rates of hydrothermal Fe(II)aq input, rates of oxidation, and rates of Fe(OH)3 settling suggests exceptionally low O2 contents, <0.001% of modern O2 contents in the photic zone. Such low levels suggest an anoxygenic pathway is more likely, and the data can be well modeled by anoxygenic photosynthetic Fe(II) oxidation. Comparison of the Fe isotope data from the Isua BIFs with those from the 2.5Ga BIFs from the Hamersley and Transvaal basins (Australia and South Africa, respectively) suggests a striking difference in Fe sources and pathways. The 2.5Ga magnetite facies BIFs of Australia and South Africa have δ56Fe values that range from −1.2‰ to +1.2‰ over small scales, and are on average close to 0‰, which is significantly lower than those reported here from the Isua BIFs. The wide range in Fe isotope compositions for the Hamersley and Transvaal BIFs, in concert with C and O isotope data, have been interpreted to reflect bacterial dissimilatory Fe(III) reduction (DIR). The absence of low δ56Fe values in the Isua BIFs, as well as the lack of fine-scale isotopic heterogeneity, may indicate formation prior to widespread DIR.