[S-GC38] Solid Earth Geochemistry, Cosmochemistry

[SGC38-01] Reevaluation of the highly siderophile element composition of the primitive mantle by applying the pyrolite model

*Tetsu Kogiso¹, Akira Ishikawa², Norikatsu Akizawa³ (1.Kyoto University, 2.Institute of Science Tokyo, 3.Hiroshima University)

Keywords:primitive mantle, peridotite, chemical evolution of the mantle, ophiolite

Highly siderophile elements (HSE [Ru, Rh, Pd, Re, Os, Ir, Pt, Au]) in the Earth are concentrated in the metallic core, and their abundances in the silicate mantle are extremely low (µg/g level). However, due to their minute quantities, the HSE concentrations in the mantle are sensitive to processes involving metallic phases. Therefore, they are pivotal in understanding the processes like core-mantle segregation and subsequent core-mantle interaction since the formation of the Earth. Accurate estimation of the HSE composition of the mantle at the time of separation from the core (primitive mantle) is of paramount importance in constraining the starting point of the chemical evolution of the mantle. Estimates of the HSE composition of the primitive mantle have been based on the composition data of fertile peridotites, which are enriched in melt components, such as Al₂O₃ (e.g., Becker et al., 2006; Fischer-Gödde et al., 2011). This is predicated on the premise that the composition of peridotite that experienced less partial melting is closer to that of the primitive mantle. However, it is widely recognized that many fertile peridotites have undergone the process referred to as “refertilization”, addition of mafic melts to refractory peridotites that have experienced depletion in melt components. This modification renders them unsuitable for reliable estimation of the primitive mantle composition. Another approach to estimate the HSE composition of the primitive mantle involves the application of the “pyrolite model” (Ringwood, 1962), which hypothesize that the composition of the primitive mantle is that of a mixture of refractory peridotite and melt extracted from the mantle. However, refractory peridotites generally exhibit a substantial compositional variability in HSE. This observation indicates that the HSE of refractory peridotites are susceptible to the influence of secondary alteration processes, such as metasomatism. Consequently, in order to estimate the HSE composition of the primordial mantle using the pyrolite model, it is necessary to use data from refractory peridotites that have been minimally affected by secondary alteration processes. In this study, we attempted to estimate the HSE composition of the primitive mantle by applying the pyrolite model, using HSE composition data for refractory peridotites from the Timor-Tanimbar ophiolite (TTO) in eastern Indonesia, which have been minimally affected by secondary alteration. The HSE composition of the melt component extracted from the mantle was calculated using the HSE data of the most prevalent mantle-derived melt, mid-ocean ridge basalt (MORB). The calculated HSE concentrations of the mixture of TTO refractory peridotites and MORB (TTO pyrolite) is within ±35% of those of the primitive mantle estimated by Becker and co-workers (B-PM). The relative abundances of HSE of TTO pyrolite are comparable to those of B-PM, exhibiting a nearly chondritic composition, yet they lack the substantial enrichment of Ru and Pd as observed in B-PM. Furthermore, TTO pyrolite is notably enriched in Re: its Re/Os and Re/Ir ratios exceed 1.5 times those of CI chondrite (B-PM stands at approximately 1.1 times). Given that the composition of B-PM is estimated based on the composition of fertile peridotites affected by refertilization, the differences between TTO pyrolite and B-PM would be due to refertilization. This is in contrast to the fact that the compositions of fertile peridotites are similar to that of the primitive mantle in terms of major elements. If TTO pyrolite actually reflects the HSE composition of the primitive mantle, then the Re/Os of the bulk silicate Earth would exhibit significantly higher values than that of the CI chondrite. This would necessitate a comprehensive re-evaluation of the conventional model for the Os isotope evolution in the mantle.