Interwoven Elements: The Influence of Dissolved Silica, Persistent Dissolved Iron, and Oxidative Mechanism on the Genesis of Banded Iron Formations

Abstract

Banded iron formations (BIFs) are iron- and silica-rich sedimentary rocks deposited from the Archean and Paleoproterozoic oceans between ~3.8 and ~1.8 billion years ago. Because BIFs derive from seawater precipitates, the original iron minerals in BIFs may provide information about the biogeochemistry of Earth’s early oceans and the evolution of planetary redox state. However, the nature of primary BIF minerals and the processes controlling their formation remain poorly understood, limiting our ability to reconstruct past ocean biogeochemistry from these rocks. In particular, while BIFs precipitated from seawater rich in dissolved ferrous iron (Fe(II)aq) and silica (Siaq), the field has yet to recognize the influences that persistent Fe(II)aq and Siaq have in determining the Fe redox state and mineralogy of primary precipitates. Fe(II)aq and Siaq may also mediate processes that modify minerals soon after their formation, but existing research has not analyzed precipitates with sufficient temporal resolution to capture these early changes.

In this dissertation, I test how Fe(II)aq and Siaq influence the formation and transformation of Fe(III)-bearing minerals formed during microbial and abiotic Fe(II) oxidation. First, I investigated how minerals formed by Fe(II)-oxidizing photosynthesis varied as a function of dissolved silica. The hydrous Fe(III) oxyhydroxide ferrihydrite was the primary precipitate formed at all Si levels tested, with small amounts of more crystalline Fe(III) oxyhydroxides (goethite and lepidocrocite) only observed at the lowest Siaq loadings. Residual Fe(II)aq catalyzed the transformation of ferrihydrite to lepidocrocite and goethite, although dissolved silica impeded this process by protecting ferrihydrite from Fe(II)aq sorption and electron transfer. These findings show that the initial products of iron oxidation can evolve in the presence of remaining Fe(II)aq, with the extent of these transformations dependent on Siaq loadings.

Next, I examined whether a continuous supply of Fe(II)aq to iron-oxidizing microbes would favor the formation of mixed-valent Fe minerals, which contain both Fe(II) and Fe(III), in a system without silica. Despite an excess of Fe(II)aq relative to Fe(III), phototrophic Fe(II) oxidation only produced the Fe(III) oxyhydroxide goethite. This finding challenges the idea that the overall Fe redox state of precipitated phases must reflect that of the source solution. Here, fully ferric minerals formed even with substantial Fe(II)aq present, suggesting that the local chemical environment may be more important than bulk solution chemistry in determining which minerals nucleate and grow during Fe(II)aq oxidation.

Finally, I developed a novel continuous flow reactor to maintain constant levels of both Fe(II)aq and Siaq during abiotic Fe(II) oxidation by trace dissolved oxygen. The first precipitates formed were poorly crystalline Fe(II,III)-silicates, but interactions with dissolved oxygen and Fe(II)aq partially transformed these phases into the Fe(II,III) hydroxy salt green rust within hours. This study is the first to document this specific mineral sequence and suggests that Fe(II,III)-silicates could have been primary BIF phases formed in the Archean water column. However, this possibility requires that initial Fe(II,III)-silicates be exported to the seafloor before significant transformation occurs, highlighting the importance of constraining the relative rates of these secondary processes.

Overall, these results challenge our understanding of the information provided from authigenic BIF minerals. There has been a long-held assumption that the earliest-identified BIF minerals closely resemble the initial mineral phases formed in ancient seawater. However, Fe(II)aq and Siaq likely played an important and previously underestimated role in mediating rapid transformations of these seawater precipitates prior to deposition.