CMBmelt

Seismological, geodynamical and mineral physics observations favor the presence of chemical heterogeneity in the lowermost mantle boundary layer. They are thought to be associated with residues from the core formation process, segregation of dense subducted slab components, effects of partial melting, and/or chemical interactions between the core and mantle. However, their interpretations and their modeling are dependent on our knowledge of the physical properties of materials at relevant pressure and temperature conditions. In particular, the ability to generate and retain partial melt at these depths is central to this question. The presence of high-pressure melts will have consequences for the dynamics of the deepest part of the Earth, the heat flow across the core mantle boundary (thereafter referred to as CMB), the partitioning of trace elements, the budget of rare gases, as well as the lifetime of a dense basal liquid layer. Surprisingly, phase diagrams for partial melting of mantle minerals in the D” region and at the vicinity of the CMB are almost completely unknown. Moreover, the time scale for the crystallization of a basal magma ocean depends critically on the accurate knowledge of phase diagram of the lowermost mantle. As shown by Labrosse et al (2007), it is proportional to the temperature difference between the highest and the lowest liquidus value and to the chemical composition difference between phases at liquidus and solidus at these extreme pressure and temperature conditions. Both of these parameters are very poorly constrained at the moment. One of the goals of the present proposal is to provide a realistic phase diagram in order to provide more accurate models of coupled thermal and chemical evolution of the mantle. At present, experimental evidences that a deep melt layer could maintain a relatively large isentropic temperature gradient relative to the liquidus temperature gradient (and hence guarantee that crystallization should proceed downward with a residual liquid trapped on the core surface) are indeed shock compression studies (Mosenfelder et al., 2007, 2009) carried out on samples with forsterite or MgSiO3 composition. However, these studies are focused on simple systems that are not fully representative of the Earth’s mantle, and are unable to yield the nature of the liquid nor the nature of the phase diagram (i.e., which phase at solidus or liquidus). Therefore, we propose an experimental mineral physics program aiming to constrain: (1) the nature of phases at the solidus and liquidus in the vicinity of the CMB, relevant phase diagrams and upper bounds on CMB temperatures (2) the partitioning of some specific trace elements associated with these deep partial melting processes, in relation to geochemical signatures recorded at oceanic island basalts (OIBs) (3) the thermal and chemical evolution of these melts and the consequences for the geochemical and heat budget, and a better constraint of the heat flux across the CMB. It has long been considered unrealistic to carry out experiments on liquids at extreme pressure and temperature, i.e. in a diamond-anvil cell. However, tremendous progress has been made recently with in situ experiments carried out at synchrotron sources on liquids; and there have also been advances in analytical tools that now work with spatial and analytical resolutions suited to the typical 100-nanometer scale of our high pressure phases. Both of these advances, combined with the possibility to detect and trap liquids in a diamond-anvil cell, make this experimental enterprise possible. Therefore, this ANR proposal could now achieve a significant progress in the understanding of the deep melt formation processes. We propose to address these questions with both experiments and appropriate thermodynamic and thermal modeling, in order to shed light on one of the most intriguing regions of the Earth, i.e., D” and the core mantle boundary.