Geochemical constraints on mantle dynamics in the Hadean

Abstract

Mantle convection is intimately linked with terrestrial degassing through decompression melting, magma generation. Hence tracers of mantle degassing through time shed light on mantle dynamics from the Hadean to Present. Xenon isotopes provide an exceptional set of such tracers: 129Xe was produced by the now extinct β-decay of 129I with T 1 / 2 = 15.7 Myr, 131−136Xe were produced by the now extinct spontaneous fission of 244Pu with T 1 / 2 = 82 Myr, and 131−136Xe are produced by the extant spontaneous fission of 238U which decays with T 1 / 2 = 4.45 Gyr. We report new data on the isotopic composition of xenon in the deep mantle, as sampled by the Kola (Russia) mantle plume. According to the relative abundances of radiogenic/nucleogenic/fissiogenic noble gas isotopes, the 244Pu contribution is estimated to be 33–60% of total fission 136Xe. This value is significantly lower than what is expected for a chondritic reservoir (96%), and is equal to, or higher than, the previously proposed best estimate for the MORB mantle. We show that both MORB and plume sources have lost ≥ 99% fission Xe produced by the decay of 244Pu. In comparison, the mantle has lost only ∼50% of radiogenic argon produced by 40K ( T 1 / 2 = 1.25 Gyr). Therefore, the major loss of fissiogenic Xe isotopes from 244Pu took place before quantitative accumulation of noble gas isotopes produced by long–lived radioactivities. As constrained by the half-life of 244Pu, such loss post-dated the major episodes of terrestrial accretion and differentiation that took place within the first 100 Myr, and occurred mostly during the Hadean. The xenon record of early mantle dynamics is consistent with extensive loss of volatile elements in the first 100 Myr ASSC, during terrestrial accretion and magma ocean episodes, but also indicates global mantle differentiation that affected both the source of mantle plumes and that of mid-ocean ridge basalts for several hundreds of Myr (e.g., 400–700 Myr ASSC). During this period, the loss rate of volatile elements was at least one order of magnitude higher than at Present, implying a much more thermally active Earth during the Hadean. In contrast, Mars became an essentially static planet during this time period. Our view may also imply that the atmosphere was open for time intervals longer than those generally thought.