Basalts core/mantle boundary

Can we be sure that the iron meteorites are indeed fragments of cores Since no differentiated asteroid has yet been visited by a spacecraft, we rely on circumstantial evidence. Some M-type asteroids have spectral characteristics expected from exposed metallic cores (Tholen, 1989), while others exhibit basaltic surfaces, a hallmark of global differentiation. Although olivine-rich mantles should dominate the volume of differentiated asteroids, there is an enigmatic lack of olivine-rich asteroids (and meteorites) that could represent mantle material (Burbine et al., 1996). Until we visit an asteroid with parts of a core-mantle boundary exposed, our best evidence supporting a core origin is detailed smdies of iron meteorites. 

Seismic scatterers within the lower mantle are more likely to represent chemical than thermal heterogeneities, with subducted slab material (especially the basaltic crustal component thereof) constituting a likely candidate. The very base of the mantle, nearest the core-mantle boundary, may also be characterized by significant major-element chemical heterogeneity. 

In this present version of the model the D" layer is thought to have originated very early in Earth history, as an early, incompatible element- and metal-rich basaltic crust, enriched during late accretion (4,540-4,000 Ma) with chondritic material. There is support from Nd-and Hf-isotopes for the existence of this very early differentiate of the mantle (see Sections 3.2.3.1 and 3.2.3.2). This crust, when subducted, had a bulk density which exceeded that of the mantle and numerical modeling experiments confirm that it would have stabilized at the core-mantle boundary (Davies, 2006). 

Different authors have variously located primitive mantle in the present-day lower mantle beneath the 660 km discontinuity, in a deep layer beneath a 1,600 km discontinuity, in the D" layer at the core-mantle boundary, or as "blobs" within the lower mantle (Becker et al., 1999). It is arguable whether any modern basalts have been derived from such a primitive reservoir, and it is possible that such a reservoir does not exist. In fact models of whole mantle convection, such as that of Helffrich and Wood (2001), in which the majority of the mantle has been processed through the subduction process, render the preservation of primitive mantle most unlikely. 

---