Dm. Sherman, STABILITY OF POSSIBLE FE-FES AND FE-FEO ALLOY PHASES AT HIGH-PRESSUREAND THE COMPOSITION OF THE EARTHS CORE, Earth and planetary science letters, 132(1-4), 1995, pp. 87-98
First-principles density functional calculations (beyond the local den
sity approximation) are used to predict the equations of state (EOS) a
nd formation energies of Fe-FeO and Fe-FeS alloys under the pressures
of the Earth's core. The accuracy of the static calculations is demons
trated from predicted equations of state and phase transitions of Fe,
FeO and FeS. As indicated by the formation energies of Fe3O and Fe4O,
solid solution between Fe and FeO remains energetically unfavorable up
to core pressures. The instabilities are so large that no reasonable
entropy term could stabilize an Fe-FeO solid solution at core temperat
ures. In contrast, solid solution between Fe and FeS becomes favored a
t core pressures as indicated by the formation energy of Fe3S. To the
extent that the Earth's inner core is not dense enough to be pure iron
, it follows that the inner core is most likely an Fe-FeS alloy rather
than an Fe-FeO alloy. This, however, requires that the melting point
of FeS fall below that of Fe at core pressures. The much lower density
of the outer core may reflect either the width of the ''phase loop''
in the Fe-FeS binary or presence of an additional light element which
cannot be incorporated into solid iron. Even if an additional light el
ement is present in the outer core, the Earth must be enriched in sulf
ur relative to potassium. The K/S ratio of the Earth must reflect the
segregation of the core as an Fe-FeS eutectic during the early differe
ntiation of the Earth.