A multistage origin for Neoarchean layered hematite-magnetite iron formation from the Weld Range, Yilgarn Craton, Western Australia

The origins of iron formations remain somewhat enigmatic despite much progress over the past decades. This is due in part to continuing research demonstrating that these deposits do not have a single origin but rather can be formed under various and variable conditions. This study describes a formation pathway for an Algoma-type banded iron formation (BIF) from the 2.75 billion-year-old Weld Range of Western Australia, based on petrographic and Fe isotope analyses of drill core samples. The BIF is composed of alternating layers of jaspilitic and Fe-poor chert, magnetite layers of varying thickness and abundance, and rare pyrite. Petrographic analysis indicates the minerals formed in the order hematite, magnetite, and then pyrite, with the latter two clearly replacive of “primary” bedded hematite. Hematite δ⁵⁶Fe values from all portions of the drill core are positive and essentially identical (δ⁵⁶Fe = 0.61 ± 0.05‰). Magnetite δ⁵⁶Fe values are positive, as well (δ⁵⁶Fe = 0.51 ± 0.07‰) but vary systematically with total magnetite content of the core samples. Pyrite δ⁵⁶Fe values are positive and higher than both hematite and magnetite (δ⁵⁶Fe = 1.03 ± 0.05‰). Combined, these analyses reveal a multistage formational model for the Weld Range BIF. Initial deposition occurred by the partial oxidation of hydrothermally-sourced aqueous Fe(II) to form a Fe(III)-Si co-precipitate, probably by anoxygenic photosynthetic iron oxidizers. This material would have been converted to hematite under equilibrium conditions with excess aqueous Fe(II). Magnetite formed by a later intrusion of reducing fluids, mostly along bedding planes, but also as cross-cutting veins. This produced large quantities of magnetite layers parallel, and sub-parallel, to the original bedding, but that do not represent a primary depositional feature. Pyrite was produced in some portions of the unit, likely under equilibrium conditions, by later infiltration of reducing fluids containing sulfide. The entirely positive δ⁵⁶Fe values for all Fe phases in the Weld Range BIF stands in stark contrast to the more massive Superior-type BIFs of the Paleoproterozoic Hamersley and Transvaal basins of Western Australia and South Africa, respectively, which have δ⁵⁶Fe values that average close to 0‰ but vary over nearly the entire range of δ⁵⁶Fe values measured in nature (δ⁵⁶Fe = −2.5 to +2.6‰). The Weld Range BIF δ⁵⁶Fe values are similar to those of older Algoma-type BIFs such as the Paleoarchean Isua BIF of Greenland but are less positive on average. This overall trend of decreasing average δ⁵⁶Fe values through time likely records an overall increasing trend of oxygenation of the Archean surface ocean from 3.8 to 2.45 billion years ago.