Predictions of mineral assemblages in planetary interiors

Abstract

The mineral assemblages expected in planetary interiors and their variations with depth can be predicted from estimates of planetary composition by using known mineral compatibilities in the model 5ystem CaO-MgO-Al_2O_3-SiO_2. In general, the available estimates of bulk composition of the terrestrial planets suggest that these planets can be divided into two groups based on their predicted mineral assemblages. The terrestrial, venusian, and lunar bulk compositions are expected to transform from plagioclase lherzolite to spinel lherzolite to garnet lherzolite with increasing pressure. In contrast, the martian and mercurian interiors are expected to transform from spinel-plagioclase wehrlite to spinel lherzolite to (spinel)-garnet wehrlite with increasing pressure. These assemblages exert a major influence on the compositions of liquids produced by melting of these planetary interiors, on the solidus temperatures, and consequently on the nature of planetary differentiation and on the types of magmas extruded at planetary surfaces. If melting occurs at intermediate pressures where all of the planetary mantles are expected to be spinel lherzolites, similarities between the phase equilibria controlling the compositions of the magmas produced in all of these planets can be anticipated. Superimposed on these similarities in the magmas, however, will be differences in Fe/(Fe+ Mg) ratio and volatile content inherited from the planetary compositions. At higher and lower pressures than the stability field of the spinel lherzolite assemblage, melts produced in Mercury and Mars can be expected to be more aluminous and undersaturated than melts produced on the other planets.

Additional details