Abstract

The thermodynamic theory describing the partitioning of trace elements between crystals and silicate melt implies that partition coefficients should depend on the major-element composition of the melt from two different causes, namely (1) the activity coefficient of the trace-element oxide component in the melt, and (2) the activities of all the major-element components needed to balance the trace-element substitution in the crystal (the “stoichiometric control”). Partition coefficients are also expected to vary with the composition of the crystal, and temperature and pressure. Because these variables cannot be controlled independently in direct crystal/melt partitioning studies, it has not been possible to disentangle their effects, or to determine their relative importance. In order to explore the effects of melt composition on activity coefficients of trace-element oxide components, the activity coefficients of five such components, MoO 2, MoO 3, FeO, NiO and CoO, were measured in 18 different melt compositions in the system CaO–MgO–Al 2O 3–SiO 2 plus one composition in CaO–MgO–Al 2O 3–SiO 2–TiO 2 at 1400 °C, by equilibration with the metal under controlled oxygen fugacity. MoO 2 and MoO 3 are expected to have geochemical properties similar to the High Field Strength Elements (HFSEs). The activity coefficients of MoO 2 and MoO 3 vary by factors of 20 and 60, respectively, over the range of compositions investigated. Their variation is highly correlated, and mainly depends on the amount of CaO in the melt, suggesting the influence of CaMoO 3 and CaMoO 4 complexes. The analogy between Mo and HFSEs implies that melt composition can be expected to have an important influence on HFSE partition coefficients. The activity coefficients of FeO, NiO and CoO vary by a factor of two over the same range of melt compositions, but show no simple dependence on any particular major-element oxide component. However, the activity coefficients of all three components are very highly correlated with each other. This means that the effect of melt composition can be largely eliminated if the ratios of two activity coefficients are taken, as, for example, when two-element distribution coefficients are used.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call