4:45 PM - 5:00 PM
[SIT18-11] The Compostion of the Martian Core
Keywords:Mars, Core, Magnetism, Machine Learning, Molecular Dynamics, Iron Alloys
The InSight mission is beginning to reveal fundamental constraints on Mars' deep interior structure. The most striking results are associated with the Martian core. Inversions from seismic receiver functions and geodesy suggest that the core is both it is large and has a low density when compared to that of Earth. This reduced density leads some to speculate that the Martian core must contain more 'light elements' than Earth.
By considering geophysical data from the recent insight mission, we perform ab initio simulations to explore the range of possible chemical compositions. Explicitly we seek to constrain the Fe-S-Si-O-H-C liquid system using spin-polarised calculations with geophysical and cosmo-chemical constraints. Calculating the physical properties of iron-bearing systems (particularly in the case of the liquids) is far from straightforward as it is likely that, even at the conditions in the core of Mercury (up to 40 GPa and 4000 K), magnetism in the form of local atomic moments may alter the properties of both solid and liquid iron alloys. Indeed, pure solid iron is magnetic at these conditions, and magnetism significantly influences sub-solidus density and phase stability.
Our preliminary simulations of Fe-S-Si-O-H-C iron show finite local moments, which result in a reduced density compared with non-spin-polarised simulations, modify physical properties such as density, bulk-sound velocity and heat capacity. In total, we explore over fifty different iron alloys, where a machine-learning equation of state to used to interpolate across composition, pressure and temperature. Findings from this study are directly applicable to planetary cores more extensively. Furthermore, the methods developed and implemented apply more broadly to high-pressure physics and Earth science.
By considering geophysical data from the recent insight mission, we perform ab initio simulations to explore the range of possible chemical compositions. Explicitly we seek to constrain the Fe-S-Si-O-H-C liquid system using spin-polarised calculations with geophysical and cosmo-chemical constraints. Calculating the physical properties of iron-bearing systems (particularly in the case of the liquids) is far from straightforward as it is likely that, even at the conditions in the core of Mercury (up to 40 GPa and 4000 K), magnetism in the form of local atomic moments may alter the properties of both solid and liquid iron alloys. Indeed, pure solid iron is magnetic at these conditions, and magnetism significantly influences sub-solidus density and phase stability.
Our preliminary simulations of Fe-S-Si-O-H-C iron show finite local moments, which result in a reduced density compared with non-spin-polarised simulations, modify physical properties such as density, bulk-sound velocity and heat capacity. In total, we explore over fifty different iron alloys, where a machine-learning equation of state to used to interpolate across composition, pressure and temperature. Findings from this study are directly applicable to planetary cores more extensively. Furthermore, the methods developed and implemented apply more broadly to high-pressure physics and Earth science.