Metallurgy at Extreme Conditions: Molten Iron-Alloy Constraints on the Light Elements in Earth's Core

One of the outstanding mysteries in the Earth sciences is the composition of the core. We know from seismic and cosmochemical constraints that the core is made of a nearly pure iron (~95% Fe + 5% Ni), and that the inner core is solid and the outer core is molten. However, based on our knowledge of the behavior of molten iron at the extreme pressure and temperature conditions of the core, it is apparent that there must be some other light element or elements dissolved in the molten outer core as well. It is thought that the light element is related to convection in the outer core and is therefore important for spawning the Earth's magnetic field. The nature and abundance of the light element will also determine the kinds of reactions that might occur at the boundary between mantle silicate and the molten metal core. For the last half-century the primary candidates for the light elements in the core have included H, O, S, C, and Si. We will never be able to sample the core directly, hollywood movies notwithstanding, so other approaches are required to deduce the identity of the light elements. Basically, the approach has been to try and determine which elements can dissolve into molten iron at core conditions using experiment and theory. A perusal of the vast literature on this subject reveals that individual elements and cocktails of elements have come into, out of, and back into favor with time. Different experimental and theoretical approaches often lead to very different interpretations as to the identity of the light elements. Here we propose a method for deducing the light element in the core that relies on a combination of experiment, thermodynamic modeling, and seismic observations. Seismic data constrain the velocity at which compressional waves can move through molten iron as well as the density. They can also detect whether the core liquid has separated into more than one liquid (immiscibility). In principle, if one knows the same properties for various iron alloy - light element mixtures, one can deduce the composition of the core. The seismic observations are available. An internally consistent model for the properties of molten alloys at core conditions is not. However, thermodynamic relationships allow the physical properties to be determined through the equation of state of molten alloys. The parameters required to develop the thermodynamic model can be deduced through the melting curves of iron - light element alloy compositions. Here, we are proposing to make measurements of the melting curves of two-component (binary) alloys such as FeO, Fe3C, FeS, FeH and FeSi in order to derive the quantities required for the thermodynamic model. We have developed robust techniques in our lab for measuring melting points to very high pressures and temperatures using the laser-heated diamond anvil cell. Further, we have developed an exciting and novel new X-ray imaging technique with which we can measure directly the minimum melting compositions (eutectics) in iron - light element systems. These data further help constrain the thermodynamic models. In summary, we will use an experimental approach to measure how iron - light element alloys melt and from this data we will develop a multi-component thermodynamic model that will allow us to predict the seismic wave velocities and density of a wide range of possible core liquids. We will then compare the model with actual observations to deduce the identity of the elusive light elements in the molten outer core.