Kinetics of chemical exchange during melting of planetary interiors

Abstract

Experimental and numerical modeling studies are used to place constraints on the kinetics of chemical exchange during partial melting within the mantles of the terrestrial planets. Chapter 1 presents experiments on the diffusion rates of La, Ce, Nd, Dy and Yb in diopside at pressures of 0 to 2.5 GPa and temperatures of 1050 to 1450 'C. The results demonstrate a large variation in diffusivity among the rare earth elements, with the diffusion coefficient for La a factor of -35 smaller than for Yb at a given temperature and pressure. Chapter 2 presents experiments on the diffusion of Sm, Dy and Yb in pyrope at 2.8 GPa and 1200-1450 "C. No significant difference in diffusivity is found among these elements, and their absolute diffusion rates are similar to those of the heavy rare earth elements in diopside at the same pressure and temperature. Chapter 3 presents a numerical model for diffusion-controlled fractionation of trace elements during adiabatic decompression melting of a polyphase solid. The model is used to simulate the fractionation of rare earth elements between solid and melt during partial melting of Earth's upper mantle. Diffusion is found to exert a strong control on the evolution of the system at conditions typical of melting beneath ocean spreading centers, leading to less efficient fractionation of the rare earth elements than under conditions of local chemical equilibrium. Chapter 4 presents experiments on the diffusion of U and Th in diopside at I atm pressure. Uranium and thorium are found to diffuse at similar rates, and diffusive fractionation between these elements is therefore unlikely to be significant during partial melting in Earth's upper mantle. Thorium and radium may be diffusively fractionated, however, enhancing the production of 22 Ra/230Th radioactive disequilibrium during partial melting while inhibiting chromatographic fractionation during melt transport. Chapter 5 presents phase equilibrium and dissolution kinetics experiments that constrain hypotheses for the origin of lunar high-Ti ultramafic glasses. The experimental results demonstrate that assimilation of ilmenite-bearing cumulates is not a viable mechanism for production of the high-Ti glasses. It is proposed that the source of the high-Ti ultramafic glasses formed by shallow level mixing and reaction of late-stage magma ocean liquids with underlying olivine-orthopyroxene cumulates, followed by sinking of these dense hybrid materials into the lunar mantle.