Evolution of volatile species in the earth’s mantle: A view from xenology

Abstract

To understand the scales and chronology of processes governing the evolution of terrestrial gas species, the constraints from 244Pu–238U–129I–Xe systematics are crucial and should be included in any model related to gas loss/gain by the Earth and to gas redistribution among terrestrial reservoirs.Reliable constraints can be derived from meteoritic and terrestrial abundances of the highly refractory lithophile incompatible parent isotopes 244Pu (half life τ244=80.0Myr) and 238U (τ238=4468Myr). Both isotopes produce heavy Xe isotopes by fission; different relative yields for the Xe isotopes allow contributions of Xe(Pu) and Xe(U) to be distinguished. It is also useful to consider the 129I–129Xe(I) systematics (τ129=15.6Myr) even though iodine is a highly volatile element and its terrestrial abundance is less well known. The parent isotopes, for which the initial (at the time of formation of the solar system, 4.567Gyr ago) abundances are known from investigations of meteorites and ancient terrestrial zircons, yield the closed-system (subscript CLOS) present-day ratios of [136Xe(Pu)/136Xe(U)]CLOS=28 and [129Xe(I)/136Xe(Pu)]CLOS=110, much exceeding values observed in the depleted heterogeneous mantle reservoir (DMR): [136Xe(Pu)/136Xe(U)]DMR⩽3 and [129Xe (I)/136Xe(Pu)]DMR⩽60 (Pepin and Porcelli, 2006). Also, the present-day amounts of 129Xe(I) in the mantle (∼0.01Tmol) and in the atmosphere (0.278Tmol) are well below the total value produced by decay of 129I (129Xe (I)=35Tmol). These relationships between the closed system and the observed values show loss of early-produced Xe isotopes occurred not only from the DMR, but also from the Earth–atmosphere system as a whole.Using abundances of the parent and daughter isotopes within the framework of a simple one-mantle-reservoir degassing model we conclude: (1) the present day mantle is a severely degassed reservoir, so that only<10−3 of the initially available amount of stable Xe atoms (e.g., 130Xe) has survived 4.567Gyr of degassing. This low retention parameter is practically model-independent, as any solution is governed by the requirement of almost total 136Xe(Pu) loss from the mantle. (2) The degassing rate as a function of time appears to be the most reliable constraint on mantle convection in the past. To ensure intense early degassing, the rate of mantle convection during the Hadean era must exceed the present day value by up to a factor of ∼100. These two issues also follow from the study of two mantle reservoir models: if primordial and early-produced species were added into the convecting mantle from a hypothetical early-formed reservoir, the mantle itself must be degassed to a higher rate than that predicted by the one-mantle-reservoir model. This is in contrast to the very model dependent issue (3): the one-mantle-reservoir degassing model predicts a rather late time for atmosphere closure to Xe loss, between 3.5 and 4Gyr ago (possibly even extending to the Archean).Several recent mantle evolution models (based on U–Th–He and K–Ar isotope systematics) postulate a low overall degassing of the Earth and are inconsistent with these results from mantle xenology.