Deep penetration of molten iron into the mantle caused by a morphological instability

Abstract

The core-mantle boundary of Earth is a region where iron-rich liquids interact with oxides and silicates in the mantle. Iron enrichment may occur at the bottom of the mantle, leading to low seismic-wave velocities and high electrical conductivity, but plausible physical processes of iron enrichment have not been suggested. Diffusion-controlled iron enrichment is inefficient because it is too slow, although the diffusion can be fast enough along grain boundaries for some elements. More fundamentally, experimental studies and geophysical observations show that the core is under-saturated with oxygen, implying that the mantle next to the core should be depleted in FeO. Here we show that (Mg,Fe)O in contact with iron-rich liquids leads to a morphological instability, causing blobs of iron-rich liquid to penetrate the oxide. This morphological instability is generated by the chemical potential gradient between two materials when they are not in bulk chemical equilibrium, and should be a common process in Earth's interior. Iron-rich melt could be transported 50 to 100 kilometres away from the core-mantle boundary by this mechanism, providing an explanation for the iron-rich regions in the mantle.

Fate of MgSiO 3 melts at core–mantle boundary conditions

“…There is a clear contrast between our data and those derived from Brillouin measurements on the MgSiO 3 glass (18). We found a density of 4.27 ± 0.2 g/cm 3 at 33 GPa compared with 3.5 ± 0.01 g/cm 3 from the Brillouin data. The bulk moduli also exhibit a large discrepancy: K T0 = 16.9 ± 3.2 GPa in our study compared with K T0 = 78.4 ± 0.6 GPa determined from the Brillouin data (18).…”

“…A fourth-order Birch-Murnaghan equation of state reproduces our experimental data over the entire pressure regime of the mantle. At the core-mantle boundary (CMB) pressure, the density of MgSiO 3 glass is 5.48 ± 0.18 g/cm 3 , which is only 1.6% lower than that of MgSiO 3 bridgmanite at 5.57 g/cm 3 , i.e., they are the same within the uncertainty. Taking into account the partitioning of iron into the melt, we conclude that melts are denser than the surrounding solid phases in the lowermost mantle and that melts will be trapped above the CMB.…”

Continent-sized anomalous zones with low seismic velocity at the base of Earth's mantle

“…Alternatively, liquid Fe might migrate to the mantle due to the chemical gradient caused by the suggested FeO depletion in the lowermost mantle, but such a process might not build structures taller than 100 km (ref. 90). A further uncertainty arises from the possible transport of volatiles, such as water and carbon, which may impact the chemical interaction between the lowermost mantle and the core 80 .…”

Core–Mantle Interactions

“…Chemical reactions between the core and mantle have been proposed as a mechanism to incorporate iron alloys into the base of the mantle (Knittle and Jeanloz, 1989). The iron alloy may be a reaction product ( Jeanloz, 1990) or a result of incorporating core material directly into the mantle Kanda and Stevenson, 2006;Otsuka and Karato, 2012;Petford et al, 2005;Poirier and LeMouel, 1992). A new highpressure phase of MgSiO 3 (Murakami et al, 2004;Oganov and Ono, 2004) may open other possibilities for high electrical conductivities in the lower mantle (Ohta et al, 2010;Ono et al, 2006).…”

“…A variety of mechanisms have been examined to explain (or refute) such a conducting layer at the base of the mantle Kanda and Stevenson, 2006;Knittle and Jeanloz, 1991;Otsuka and Karato, 2012;Petford et al, 2005;Poirier and LeMouel, 1992). While improvements in nutation theory (Koot et al, 2008;Mathews et al, 2002) and observations continue to support the existence of a conducting layer, other sources of dissipation have also been proposed.…”