A review of experimental studies of crystal/liquid trace element partitioning

Available data on silicate/liquid, phosphate/liquid and oxide/liquid trace element partition coefficients from experimental studies show an encouraging degree of consistency, although much more work is required before a set of coefficients of wide-ranging application to planetary problems can be compiled. The complex dependences of most coefficients on bulk chemical composition (and liquid structure) are important aspects which remain to be fully resolved. Further determinations of coefficients for a number of elements for phosphates, zircon, spinels, amphiboles and garnets are especially desirable and investigations into the effects of pressure and volatiles on the magnitudes of partition coefficients are also needed. There is a major discrepancy among existing data regarding the upper concentration limits of Henry's Law dilute solution behavior under different experimental conditions. Data obtained at 1 atmosphere under dry conditions can apparently be reconciled with data from highpressure, H₂O-saturated experiments only if Henry's Law limits are themselves functions of variables such as pressure, temperature, H₂O activity and chemical composition. Further experiments (including studies bearing on the role of defect substitution at very low trace element concentrations) are required to resolve this question.