Constraints on the Moon's origin from the partitioning behaviour of tungsten

Abstract

The Moon, Earth and eucritic meteorites display W/La ratios which correspond to depletions of W relative to chondrites by factors of ∼19 (ref. 1). This depletion can be attributed to metal/silicate fractionation. Ordinary chondrites and Cl chondrites with widely varying metal contents and oxidation states have similar W/La ratios, suggesting that nebular fractionations were unimportant. Therefore, assuming that both W and La are refractory and incompatible, the depletions of W relative to La in the Earth, Moon and eucrites are presumably due to separation of metal from silicate during planetary-scale igneous events. For the Moon, Rammensee and Wänke² concluded that any geophysically reasonable metallic lunar core would be too small to explain the large lunar depletion of W. It was therefore suggested² that the Moon acquired its depletion of W in a non-lunar igneous event, supporting the hypothesis that the Moon formed by fission from the proto-Earth, the upper portion of which was depleted in W due to terrestrial core formation^3–5. We show here that this conclusion is implicity based on a special case, namely equilibrium between metal and silicate phases with total melting of silicates. In the geologically more general case where metal fractionates from silicate at relatively low degrees of partial melting, the incompatible nature of W may negate this conclusion. A geophysically plausible metallic core can account for the depletion of W in lunar surface and, presumably, lunar mantle rocks. The low W/La ratio in the Moon cannot be used as unconditional evidence for a terrestrial origin of the Moon.