Skip to main content

Fluid Silicates at Extreme Conditions and the Magma Ocean

Final Report Summary - MOLTENEARTH (Fluid Silicates at Extreme Conditions and the Magma Ocean)

The goal of MoltenEarth is to predict the material behavior of fluid silicates that drove the evolution of the magma ocean, the formation of the first atmosphere, and reaction with the core. What happened when the originally molten Earth began to cool and crystallise? Did the crystals float or sink? This is an important question because Earth materials melt incongruently: like sea-ice, the crystal has a different composition than the liquid from which it froze. The relative motion of liquid and crystal then determines the vector of chemical evolution and whether chemically distinct fossil remains of the magma ocean may exist in the Earth today.
We have found that the density of crystals and liquids at high pressure are very similar. This means that crystal/liquid buoyancy is determined by the partitioning of the most abundant heavy element: iron. The influence of iron on the density of liquids and crystals depends on its magnetic state: is it high spin, as it is in familiar magnetic materials at low pressure, such as magnetite, or is it low spin, a non-magnetic, higher density state, that exists only at high pressure deep in the magma ocean. We have determined the nature of the high spin to low spin crossover for the first time from first principles quantum mechanical simulations. The crossover in silicate liquids occurs over a much broader pressure range than had been anticipated before. We have also determined the extent to which iron partitions into the liquid phase, thus increasing its density and further supporting crystal buoyancy.
We find that liquids deep in the Earth today are neutrally buoyant: essentially the same density as coexisting solids. This means that they may remain trapped at depth for a long time. This picture explains previously enigmatic seismic observations of 1000 km-scale chemically distinct structures in the deep mantle, which we propose are remnants of magma ocean solidification.
We also find that the electrical conductivity of silicate and oxide liquids is larger than what had been suspected. Silicate liquids in the deep Earth may couple with the magnetic field in a variety of ways: by influencing Earth’s rotation and the length of day and by altering the shape of the magnetic field as it makes its way from the core to the surface where we can observe it. Our results suggest that the magma ocean may have produced its own magnetic field early in Earth’s history.