[H-RE13] Resource Geology

[HRE13-P10] Geochemical behavior of Fe and Mg in the Moodies Banded Iron Formations,Barberton Greenstone Belt, South Africa

*RYOHEI SUGIURA¹, Tsubasa Otake², Yoko Ohtomo², Tsutomu Sato², Ki-Cheol Shin³ (1.Graduate school of Engineering, Hokkaido University, 2.Facility of Engineering, Hokkaido University, 3.Research Institute for Humanity and Nature)

Keywords:isotope, sedimentary rock, Archean, primary precipitate, carbonate

Iron (Fe) and Magnesium (Mg) isotopes have recently been widely applied to chemical sedimentary rocks, since they are abundantly contained in authigenic minerals, such as carbonates and silicates, especially those formed in the Precambrian. In this study, whole rock Fe and Mg isotope compositions of the 3.2 Ga Moodies Banded Iron Formation (BIF), Barberton Greenstone Belt, South Africa were analyzed to investigate the geochemical behaviors of Fe and Mg in Archean sediments deposited in a shallow ocean environment, which may provide a new constraint on BIF-forming processes on the early Earth before the Great Oxidation Event. The result showed that 4 rock-types samples (i.e., Magnetite-rich siltstone, Carbonate-rich siltstone, Sandy-siltstone, Jaspilite) yielded variable Fe and Mg isotopic ratios (δ⁵⁶Fe = -0.58 to +0.60‰; δ²⁶Mg = -1.76 to +0.07‰). The whole rock δ⁵⁶Fe values in Magnetite-rich and Carbonate-rich siltstones tend to decrease with decreasing the Fe/Ti ratios as well as shallowing the depositional environment. No significant difference was observed in the trend between the two siltstones. These trends suggest that the variation in the Fe isotope ratios record chemical precipitation processes of iron that were derived from a deep-sea hydrothermally system. During the chemical precipitation processes, primary Fe mineral incorporated preferentially the heavier Fe isotope, causing enrichment of the lighter Fe isotope in the residual fluid with following a Rayleigh fractionation-type model. Previous experimental results suggest that only oxidative precipitation of ferrous iron as Fe(lll)-oxides (e.g., ferrihydrite), not Fe(ll)-silicate, in the seawater column can explain the variation in the whole rock δ⁵⁶Fe values observed in this study. On the other hand, the whole rock δ²⁶Mg values in 4 rock-types samples are linearly correlated to Mg distributions between carbonates (i.e., ankerite and Mg-siderite) and silicates (i.e., biotite and chlorite), Mg_carbonate / (Mg_carbonate+ Mg_silicate), which were obtained from the results of sequential extraction. This suggests that the whole rock δ²⁶Mg values are dictated by the mixing of the two end-member authigenic minerals. This linearity further suggests that δ²⁶Mg values of the end-member carbonates (δ²⁶Mg_carbonate) and silicates (δ²⁶Mg_silicate) were homogeneous among all the samples, regardless the rock-type or carbonate content. Because Mg isotope fractionation during the carbonate formation is estimated to be ~ -1.5‰, which is large enough to cause a significant change in δ²⁶Mg value in the remaining fluid in a closed-system, the homogeneity of δ²⁶Mg_carbonate implies that Mg was precipitated as carbonates in an open environment near the seafloor.