Abstract

In this seminar, I will present several laboratory experiments, complemented by numerical simulations, that aim at modelling iron snow: What can we learn about the settling and remelting of iron crystals in the planetary core of Ganymede (a natural satellite of Jupiter) which nourish a compositional convection at the origin of the planet’s magnetic field?

As Ganymede gradually releases heat in space and cools down, its iron-rich core gradually solidifies. This solidification leads to the formation of pure iron crystals at the core periphery, that subsequently sink deep in the core due to gravity. The settling dynamics of a cloud of crystals is modelled experimentally with particle clouds of sub-millimetric glass spheres settling in water. These clouds grow with depth due to the entrainment of ambient water into the clouds. The canonical model of entrainment by Morton et al. 1956 would predict their growth rate to be unaffected by the particles’ size. Yet, their growth rate is maximum for a specific particle size. This optimum originates from the partial decoupling between the settling particles and the flow, as explained with 3D two-way coupled Eulerian simulations.

When iron crystals ultimately reach very large temperatures deep in Ganymede’s core, they remelt. Their dense molten snow drives a compositional convection that is assumed vigorous enough to power a magnetic field through dynamo. To test this assumption, experiments are conducted where sugar grains (aka the iron snow flakes) are continuously sieved above a water tank (aka the deep convective core). The size of grains controls particle-scale interactions with the flow, with a critical influence on the length scales and velocity scale of convection, on the laminar/turbulent nature of the flow, and on the depth where sugar grains fully dissolve—with paramount implications for the emergence of dynamo in Ganymede.