nai

Project Progress

Evolution of Earth’s early atmosphere is critical to understanding the geologic history of the Earth, as well as the origin and evolution of life (e.g., Canfield, 2005). The occurrence of ferric oxides (hematite) in 3.4 Ga rocks from the Marble Bar area, Western Australia, has played an important role in interpreting the amount of free oxygen in the atmosphere of the ancient Earth. One of the key issues of using iron oxide minerals in Archean rocks to infer high O₂ level in early Earth’s atmosphere is the age of the oxidation, and, if demonstrably ancient, the mechanism of oxidation. The Archean Biosphere Drilling Project (ABDP) was initiated to collect samples below weathering profiles in the hope of avoiding post-Archean weathering that may have affected near-surface samples. The first drill core (ABDP-1) was located at Marble Bar, and recovered both the hematite-bearing Marble Bar Chert (MBC) (Figure 1), as well as the Apex Basalt, from depths greater than 100 m. In Project 1C, we show that U-Th-Pb isotope geochronology of the Apex Basalt indicates oxidation of that unit by recent weathering, and not an ancient oxidation event in the Archean. Hoashi et al. (2009) studied the Marble Bar Chert, and argued that hematite is a primary mineral formed in the Archean, which they in turn interpreted to reflect high atmospheric O₂ in the early Archean.

We revisited the same ABDP-1 drill core studied by Hoashi et al. (2009), using U-Th-Pb geochronology and Fe isotope geochemistry. Our results indicate that the Marble Bar Chert was deposited from a Fe-rich, but U-poor ocean. The δ⁵⁶Fe values of hematite in MBC range between 1.71‰ and 2.63‰, which are the most positive δ⁵⁶Fe dataset that has ever been reported from natural rocks (Figure 2). Hydrothermal sources of aqueous Fe²⁺ should have had a δ⁵⁶Fe value of around 0‰ (Johnson et al., 2008), and therefore the very high δ⁵⁶Fe values of hematite in MBC do not support the hypothesis proposed by Hoashi et al. (2009) that the ocean water at 3.4 Ga was oxygenated. If Archean ocean water was oxygenated, Fe²⁺ from hydrothermal vents would have been quantitatively oxidized, producing little to zero Fe isotope fractionation between Fe-oxide and aqueous Fe²⁺. In contrast, the high δ⁵⁶Fe values of hematite in MBC requires partial oxidation of aqueous Fe²⁺, and, given the ~3‰ fractionation between oxides and aqueous Fe²⁺ (Johnson et al., 2002), less than 10% Fe oxidation occurred. This suggests that the oxidant was extremely limited.

In addition, U concentrations of the MBC samples are over one order of magnitude lower than those of Phanerozoic cherts (Figure 3), indicating lower U content of Archean ocean water than modern ocean water. This also indicates low O₂-level in Archean atmosphere because U is mobilized in oxidized form. We therefore conclude that the mechanism of oxidation was not interaction with free oxygen, but may have involved anoxygenic photosynthetic Fe(II) oxidation.