147SM-143ND AND 176LU-176HF IN EUCRITES AND THE DIFFERENTIATION OF THE HED PARENT BODY

Abstract

The 147Sm-143Nd and 176Lu-176Hf systematics of 21 whole-rock eucrites, including five cumulates, have been investigated by MC-ICP-MS. A statistically significant Sm-Nd isochron was obtained on 18 samples with an age of 4464+/-75 Ma (MSWD=1.26) and an intercept of 0.50680+/-0.00010. This age clearly is controlled by the cumulate eucrites. The 21 basaltic and cumulate eucrites together do not form a statistically significant Lu-Hf isochron. Basaltic eucrites, however, form an errorchron with an age of 4604+/-39 Ma (MSWD=4.52), which becomes an acceptable isochron if the error on the Lu/Hf ratio is doubled with respect to the laboratory estimate. The initial 176Hf/177Hf is 0.27966+/-0.00002. Three of the cumulate eucrites regressed together further yield a statistically significant age of 4470+/-22 Ma, indistinguishable from their Sm-Nd age. We therefore conclude that cumulate eucrites are ~100 Myr younger than basaltic eucrites and basaltic eucrites are close in age to the Solar System. There is no evidence in the present data to support a decay constant of 176Lu significantly different from 1.93 10-11 yr-1 [Sguigna, Can. J. Phys., 60 (1982) 361-364]. The broad range of variation and strong correlation of Lu/Hf and Sm/Nd ratios require that eucrites are partial melts and not residual magmas. The relative Lu/Hf and Sm/Nd fractionation is controlled by the plagioclase to mafic mineral ratio in the residue. The steep correlation between these two ratios is explained by the enhancement of plagioclase saturation in the low-gravity field of the eucrite parent body. Basaltic eucrites probably formed as large-degree melts of a chondritic source. The strong Lu/Hf and Sm/Hf fractionation observed in cumulates clearly reflects the presence of residual ilmenite during melting: our preferred interpretation is that cumulates were impregnated by small-degree melts of ilmenite-bearing gabbros. Since pressure on the eucrite parent body never reaches the garnet stability field, this observation questions the ubiquitous presence of residual garnet in the mantle sources of magmas formed on larger planets such as the Moon, Mars, and possibly Earth.