Proposal: Resources for sustainable energy: ore formation by percolative reactive flow

The transition to Net Zero will require a significant increase in the supply of critical metals, exceeding current production rates. However, ore formation mechanisms in magmatic systems are poorly understood, which limits the efficiency of mineral exploration. Layered igneous intrusions, which exhibit compositional and mineralogical layering across a range of length scales, often contain laterally extensive layers of monomineralic metal oxides, and are therefore major sources of critical metals; layered intrusions are the world's largest sources of chromium, vanadium and the platinum group elements [1].

The Bushveld Intrusion, South Africa, is the world's largest layered intrusion and contains around 25

magnetitite layers (monomineralic magnetite, Fe3O4) [2], which will be the focus of this study. Currently, the dominant models for the formation of ore layers within layered igneous intrusions are in situ crystallisation [3] or crystal settling [4] within a long-lived, liquid-dominated 'magma tank'. However there is a lack of evidence for magma tanks: geophysical imaging suggests that the shallow crust underlying present-day volcanic systems has low melt contents [5].

Furthermore, it is improbable ore layers crystallised directly from magma: to crystallise the Bushveld's layers of monomineralic chromite, mass balance calculations require a mafic intrusion several kilometres thick, for which there is no evidence [6].

This PhD aims to test the hypothesis that ore layers in layered intrusions form via percolative reactive flow. Many features predicted by Korzhinskii's theory of metasomatic zoning are observed in the Bushveld's magnetitite layers: there are large chemical potential gradients at the interfaces between layers within a layered intrusion and the Bushveld magnetitite layers are 90% 1. magnetite (near-monomineralic), up to 6 m thick and have sharp basal contacts with the underlying silicate rocks.

However, Korzhinskii's theory doesn't fully describe the Bushveld's magnetitite layers: these layers generally have diffuse top contacts which grade into overlying silicate rocks containing abundant disseminated magnetite (Figure 2) and the theory only describes the formation of single metasomatic layers, whereas the Bushveld exhibits repetitive layering. The Bushveld's magnetitite layers are economically significant sources of vanadium and titanium, with the Bushveld accounting for approximately 45% of global reserves of vanadium [10].

It is therefore important to study the behaviour of trace elements within metasomatic layers to understand why these become concentrated, and where within individual magnetitite layers trace elements become enriched