Post-Archean formation of the lithospheric mantle in the central Siberian craton: Re-Os and PGE study of peridotite xenoliths from the Udachnaya kimberlite
Résumé
The formation age of the Siberian cratonic mantle is not well established. Re–Os data on various mantle-derived materials brought up by kimberlite magmas have shown that it contains Archean components, but the reported ages range broadly (3.4 to <1 Ga). We report Re–Os isotope and PGE concentration data for a suite of 29 fresh, well-characterized xenoliths from the Udachnaya-East kimberlite representing all major peridotite rock types and a large part of the cratonic mantle profile. Several xenoliths with very low Os contents (<0.3 ppb) and/or high Re/Os ratios are not suitable for age estimates. The Os (and Ir) depletions are common in cpx-bearing spinel harzburgites and coarse garnet harzburgites, but are not found in deformed, high-T peridotites. Twenty refractory (Al2O3 0.1–1.6%) peridotites yield TRD ages from 0.9 to 2.2 Ga. TRD for a subset of six high-Mg# (0.92–0.93), low-T (⩽930 °C) spinel harzburgites and a single garnet harzburgite yield a narrow range from 2.0 to 2.2 Ga with an average of 2.1 ± 0.1 Ga, which we consider the best estimate for the age of the melting event that initially formed the lithospheric mantle beneath Udachnaya. The TRD estimates for less refractory (Mg# 0.907–0.919) deformed garnet peridotites show a greater range and are generally lower (0.9–2.0 Ga; average 1.54 ± 0.28 Ga) apparently due to the effects of melt metasomatism on the initial melting residues. The predominant part of the mantle in the central Siberian craton formed in the Paleoproterozoic and not in the Archean, unlike cratons in southern Africa and North America. Minor older components reported earlier from Udachnaya may be fragments of pre-existing lithosphere trapped during stacking of melting residues formed about 2 Ga ago. We argue that the formation of cratonic lithospheric mantle, with common high-Mg# (⩾0.92) and opx-enriched peridotites, was not limited to the Archean as previously thought, but continued in the Paleoproterozoic, i.e. that asthenospheric mantle was generally hot enough to experience high-degree melting on a large scale and thus sustain the “Archean” tectonic regime till much later in the Earth’s history.