Uranium Addition and Loss in Serpentinites: The Potential Role of Iron Oxides

Abstract

Fluid-mobile uranium serves as a tracer for water-rock alteration in serpentinized mantle rocks, which constitute an important uranium reservoir. However, the mechanism for uranium addition, where uranium is hosted, and the stability of the uranium enriched material during subduction is not settled. We use geochemistry data from marine (mid-ocean ridge and fracture zone, fore-arc, fore-arc muds) and subaerially exposed (subducted, obducted, orogenic) serpentinites to show that uranium enrichment varies systematically with tectonic setting and depth from the seafloor. Only the upper ∼100 m of drilled and dredged serpentinites from marine settings contain ≥0.1 μg/g uranium enrichment, which does not correlate with the degree of serpentinization. Other settings (deeper marine samples, subducted, obducted and/or orogenic serpentinites) do not show the same degree of uranium enrichment, suggesting that uranium was lost or never gained. We use these data to argue that uranium addition requires oxidizing—bearing fluids and that uranium enrichment can be used as an indicator of these conditions. To understand where uranium is hosted, we show that uranium is enriched in samples with bulk rock Fe³⁺/Fe_tot ≥ 0.6 and appears to be buffered at a maximum uranium concentration of ∼1 μg/g. We explore potential mineral hosts for uranium in highly enriched serpentinites and propose that Fe-(oxyhydr)oxides (e.g., hematite, goethite) formed during weathering and/or carbonation could be under-appreciated hosts for uranium in serpentinized systems. We use these results to explore implications for uranium cycling and uranium isotope fractionation during subduction.