Thorium-Uranium fractionation as an indicator of petrogenetic processes

Abstract

A mean Th/U ratio, ~4, seems to characterize most terrestrial, lunar, and meteoritic igneous materials and major patterns of lead isotopic evolution develop principally in systems with Th/U about 3.7-4.0. Some important crustal igneous subsystems show systematic deviations from these values. Such fractionation generally is attributed to the geochemical behavior of the large lithophile actinide ions in various enriched minor phases. This probably is true in highly differentiated series with higher Th and U levels (>8 and 2 ppm). A different fractionation mechanism may be more important in some lower concentration systems. Isotope dilution mass spectrometric studies of U and Th in diverse igneous feldspar separates reveal significant U and Th partitioning into them with drastic fractionation of Th/U (values 0.2- 3.0). Alkali feldspars in granitic rocks display partition coefficients (feldspar/total rock) of about 0.005-0.02 for U and 0.001-0.004 for Th; Th/U~0.5-3.0. In plagioclase in gabbros, tonalites, and granodiorites with lower Th and U, the coefficients appear larger (U~0.1-0.4, Th~0.04-0.2) but Th/U appears lower (Th/U~0.2-2.0) than in K-feldspar. Limited data suggests some pyraxenes and other major minerals may also fractionate Th/U to lower ratios. Differences in ionic radius and uranium oxidation states may contribute to the undefined fractionation mechanism. Processes of primary differentiation (fractional crystallization, partial melting) involving gabbroic systems seem to reflect this fractionation. Oceanic tholeiites, end massive gabbroic complexes display generally low Th/U values. Their lead isotopes reflect reservoirs with more normal Th/U ratios. This suggests limits on the number of fractionation cycles, mantle mixing, and/or reservoir dimensions for primary basaltic systems. Th and U and their associated lead isotope systems can be used with lanthanide R.E.E. to assist in developmet of petrogenetic models.