Elemental and isotopic fractionation of Type B calcium-, aluminum-rich inclusions: Experiments, theoretical considerations, and constraints on their thermal evolution

Experiments exposing Type B calcium-, aluminum-rich inclusion (CAI)-like melts at high temperatures to high vacuum or reducing hydrogen-rich gas mixtures were used to determine the rates and consequences of elemental and isotopic fractionation by evaporation. Silicon and magnesium were found to evaporate much faster than calcium and aluminum, and the resulting residual liquid trajectories in composition space are reproduced via a thermodynamic model for the saturation vapor pressure of the evaporating species. Isotopic fractionations associated with evaporation were measured for magnesium. The resulting relationship between fraction of magnesium lost and enrichment of the residue in the heavy isotopes of magnesium follows a Rayleigh fractionation curve with a fractionation factor that is close to, but not exactly, the theoretically expected value. The rate of evaporation is found to be a strong function of temperature, oxygen fugacity, and melt composition, which can be understood and modeled in terms of the dependence of the saturation vapor pressures on these variables. The relationship between evaporation rate, which we measure, and calculated saturation vapor involves empirical evaporation coefficients that we find to be significantly less than one (similar to0.1). Analytical and numerical models are used to characterize how diffusion in both the melt and in the surrounding gas affects evaporation rates and the degree of chemical and isotopic fractionation. The experimental data and theoretical considerations are combined to give a parameterization of the rates and consequences of evaporation of Type B CAI-like liquids, which is then used to translate the measured isotopic fractionation of Type B CAIs into constraints on their thermal history. Cooling rates of the order of 10degreesC per hour are indicated. Copyright (C) 2002 Elsevier Science Ltd.