It is demonstrated that the olivine flotation hypothesis is unconvincing for silicate liquids which compress according to the law of corresponding states. Although an olivine-liquid density crossover is invariably computed, it is located at P where equilibrium between the two phases is an assumption which is either in doubt or demonstrably in error (preceding abstract). For a liquid lherzolite composition, the density is calculated to be the same as the density of olivine at approx 15 + or - 2 GPa. However, this is likely to be an 'apparent' olivine-liquid density crossover rather than a real one because it is within this P range that olivine and liquid lherzolite become unstable; olivine is replaced by majorite as the stable liquidus phase, and olivine is transformed to its high-density modified-spinel polymorph. For a high-Mg komatiite composition, the density comparison is unambiguously erroneous because olivine is not in equilibrium with komatiite at 15 GPa; olivine is replaced by garnet as the liquidus phase in the 6-7 GPa range, and again, the modified spinel polymorph of olivine is likely to be stable at approx 15 GPa. The olivine hypothesis is supported for silicate liquids which compress in violation of the law of corresponding states. For komatiite and lherzolite liquids, this elastic law predicts that a real olivine-liquid density crossover will exist in the 5-6 GPa range.-J.M.H.
|Original language||English (US)|
|Number of pages||12|
|State||Published - 1987|
All Science Journal Classification (ASJC) codes
- Environmental Science(all)
- Earth and Planetary Sciences(all)