Melting of mantle and core materials at very high pressures

Abstract
Melting experiments on the most important mantle and core materials (Mg-Si-perovskite, magnesiowustite, iron and its sulphur and oxygen compounds) are reviewed and possible sources of controversies with results from other laboratories are discussed. The iron phase diagram, measured in laser-heated diamond cells to 2 Mbar, may be reconciled with shock sound velocity measurements if one allows a correction of about 1000 K for the overshoot of equilibrium conditions during shock. The data suggest a new, yet unidentified, phase for the inner core and an inner-core temperature of about 5000 K. The addition of the light elements oxygen and sulphur does not seem to effect the melting temperature of pure iron at very high pressure. The melting curves of the major lower-mantle minerals Mg-Si-perovskite and magnesiowustite are in sharp contrast to previous estimates. Magnesiowustite is the low melting phase of the lower mantle but its melting temperature at the core-mantle boundary of about 5000 K is still about twice the average mantle temperature. Even a large temperature increment at the base of the lower mantle, estimated to be at least 1300 K, will not result in melting in the core-mantle boundary region.