K-U Studies of Silica-Rich Inclusions in the Shaw Chondrite

Abstract

The K/U can be regarded as a "planetary constant" which is invariant during magmatic processes but which differs for cosmochemical reasons between planets. This assumption is universal in all thermal history calculations for planets. K-SiO_2-rich inclusions are found in Shaw which many authors believe is a chondrite which has been subjected to partial melting (Taylor et al., 1979; Rambaldi and Lamimer, 1976). Although the origin of these inclusions is not well understood, it is possible that they represent the first melts or magmatic fluids produced in the formation of planets. Thus it is of interest to see if U and Th have followed K into these liquids. For the case of Shaw some evidence of K/REE fractionation already exists (Rambaldi and Lamimer, 1976). Six polished sections of Shaw with affixed mica fission track detectors were irradiated with ~ 2 x 10^(18)/cm^2 thermal neutrons. Excellent fission track images were obtained with no evidence for any significant contamination. (Random scans on our most-studied section gave 6 ppb U.) The fission track distributions show a high degree of localization. In one section, mapped in great detail, 20-30 large (> 50 micron) fission track localizations, can all be accounted for by whitlockite and chloroapatite. The whitlockite U concentrations (300-700 ppb) are variable, but typical for chondrites. The phosphate grains serve as fiducial points, allowing accurate location of the melt inclusions on the mica track detector (maximum position error= 20 microns). In many cases no localizations of tracks are found (U contents < 10 ppb) corresponding to the K-Si-rich inclusions, but in 7/25 cases localizations are found with U concentrations up to ~ 300 ppb. The inclusions are small (usually< 20μ), and there are many other small track localizations in this size range which have no obvious sources, thus some of these 7 cases may be accidental. However, track mapping at 13 random fields of view showed only 2 localizations (track density 3-4 times surroundings) within 20 microns. The small track localizations of unknown origin can be explained by a combination of buried sources (within 10 microns of surface), local contamination (which can never be ruled out) or localization of U on grain boundaries or cleavage planes in major phases. There is no correlation of U content with inclusion chemistry (K), size, opaque mineralogy, or petrographic location. Many of the inclusions are within large (hundreds of microns) olivine grains, including some that seem U-rich. In two of these cases the track localizations stand out from an almost blank background and match the location and size of the inclusion. It seems inescapable that the tracks do arise from the inclusions in these cases. It may be that these inclusions were a preferential site of contamination during sample preparation or by terrestrial weathering, but it is also possible that U-bearing phases only occasionally participated in the partial melting process. It may be significant that the inclusions contain no P, thus this U reservoir has not participated. Regardless, the important result is that in most cases K was mobilized to a much higher degree than U in the Shaw partial melting event. (The inclusion KIU is at least 10x bulk chondrites.) Although the inclusions are relatively Fe-rich, the concentration of K and Si and the exclusion of U and rare-earths follows the chemical systematics of immiscible silicate melts (Watson, 1976; Ryerson and Hess, 1978). To the extent that Shaw is representative of very small degrees of partial melting in planets, K and U appear to be fractionated.

Files

Additional details