ACCRETION AND EARLY DEGASSING OF THE EARTH - CONSTRAINTS FROM PU-U-I-XE ISOTOPIC SYSTEMATICS

A review of problems related to Xe isotopic abundances in meteorites and terrestrial materials leads to four postulates which should be taken into account to build a model of the Earth's accretion and early evolution. 1. The pre-planetary accretion time scale was shorter than the I-129 half-life, 17 Ma, so the initial ratio of I-129/I-127 had not been decreased considerably when planetary accretion started; therefore, this must also be the case for the Pu-244 abundance. 2. The initial relative abundance of involatile refractory Pu-244 in proto-planetary materials should be the same as in chondrites, that is, Pu-244/U-238 = 0.0068; this value corresponds to initial Pu-244 congruent-to 0.30 ppb in the bulk silicate earth. In contrast, I is a highly volatile element; its initial abundance, accretion history and even the present-day mean concentrations in principal terrestrial reservoirs are poorly known. 3.There is much less fission Xe in the upper mantle, crust, and atmosphere than is predictable from the fission of Pu-244 (Xe(Pu)) based on the above argument. Therefore, Xe(Pu) has been mainly released from these reservoirs. 4. A mechanism for Xe(Pu) escape from the complementary upper mantle-crust-atmosphere reservoirs, for example, atmospheric escape via collisions of a growing Earth with large embryos and/or hydrodynamic hydrogen flux, etc., operated during the Earth's accretion. These postulates have been used as a background for a balance model of homogeneous Earth accretion which envisages: growth of the Earth due to accumulation of planetesimals; fractionation inside the Earth and segregation of the core; degassing via collision and fractionation; and escape of volatiles from the atmosphere. During the post-accretion terrestrial history, the processes described by the model are continuous fractionation, degassing and recycling of the upper mantle and crust. The lower mantle is considered as an isolated reservoir. Depending on the scenario invoked, the accretion time scale varies within the limits of 50 - 200 Ma. In the light of recent experimental data, the latter value is inferred to the most realistic version which explains a high Xe(U)/Xe(Pu) ratio in the upper mantle. Contrary to previous suggestions, the I-129-Xe-129 subsystem is considered to be meaningless with regard to the terrestrial accretion time scale. The terrestrial inventory of Xe-129(I) is controlled by the initial abundance of volatile elements (including I and Xe) in proto-terrestrial materials and the subsequent degassing history of the Earth. The residence time of a volatile element (eg., Xe) in the bulk mantle (bm) during accretion, [t(Xe)bm], is approximated by the ratio of [t(Xe)bm] almost-equal-to m(bm)(t)/phi(bm,mf) less-than-or-equal-to 10 Ma, where m(bm)(t) is the mantle mass, and phi(bm,mf) is the rate of metal/silicate fractionation, which provided segregation of the core; phi(bm,mf) is determined by involatile siderophile element abundances in the upper mantle. This relationship implies a link between the abundance of involatile siderophile and volatile incompatible elements. A short [t(Xe)bm] reflects a high degassing rate due to extremely high phi(bm,mf) almost-equal-to 10(20) g/year. A small ratio of the atmospheric amount of Xe over the total amount of this gas in prototerrestrial materials, less-than-or-equal-to 0.01, is in accord with the process of Xe escape and fractionation in the primary Earth atmosphere.