Search

The transfer of chalcophile metals across the continental lithosphere has been traditionally modeled based on their chemical equilibrium partitioning in sulfide liquids and silicate magmas. Here, we report a suite of Ni-Fe-Cu sulfide droplets across a trans-lithospheric magmatic network linking the subcontinental lithospheric mantle to the overlying continental crust. Petrographic characteristics and numerical calculations both support that the sulfide droplets were mechanically scavenged from the mantle source during partial melting and transported upwards by alkaline magmas rising through the continental lithosphere. Nanoscale investigation by high-resolution transmission electron microscopy (HR-TEM) documents the presence of galena (PbS) nanoinclusions within the sulfide droplets that are involved in the mantle-to-crust magma route. The galena nanoinclusions show a range of microstructural features that are inconsistent with a derivation of PbS by exsolution from the solid products of the Ni-Fe-Cu sulfide liquid. It is argued that galena nanoinclusions crystallized from a precursor Pb(-Cu)-rich nanomelt, which was originally immiscible within the sulfide liquid even at Pb concentrations largely below those required for attaining galena saturation. We suggest that evidence of immiscibility between metal-rich nanomelts and sulfide liquids during magma transport would disrupt the classical way by which metal flux and ore genesis are interpreted, hinting for mechanical transfer of nanophases as a key mechanism for sourcing the amounts of mantle-derived metals that can be concentrated in the crust.

Positive Eu and Sr anomalies in primitive lavas and melt inclusions are commonly interpreted as reflecting the presence of plagioclase-bearing oceanic crust in their source. These “ghost plagioclase” signatures document the impact of subduction recycling on the heterogeneity of the convective mantle (marble-cake model). Yet, their interpretations rely on the paradigmatic assumption that partial melts are generated in chemical equilibrium with their source, which is questioned by theoretical and experimental considerations. Diffusion rates of trace elements in mantle minerals are relatively slow compared to melt transport rates, suggesting that disequilibrium is the rule rather than the exception during mantle melting. Nonetheless, in the absence of direct field evidence, the paradigm has remained. Here we report the first natural evidence of Eu anomalies developed kinetically in mantle peridotite clinopyroxene [1]. Extreme REE heterogeneity was observed in high-temperature (up to 1165°C) basalt-hosted xenoliths. Single <1-mm clinopyroxene crystals exhibit strong core-to-rim LREE fractionation accompanied by transient Eu anomalies [(Eu/Eu*)N = 0.8-1.8]. Eu anomalies first increase with LREE enrichment and then markedly decreases with further enrichment towards the rims, which cannot be explained by chromatographic re-equilibration and require a kinetic process. Numerical modelling shows that diffusional fractionation between clinopyroxene and melt both devoid of Eu anomalies can generate the observed Eu anomalies without requiring plagioclase. We demonstrate that such anomalies may develop where low-degree, LREE-enriched melts percolate in the presence of small amounts of fast-diffusing Eu2+, which is compatible with temperatures, redox conditions and transport timescales in MORB and OIB sources. We particularly envisage that this is likely to occur in the deep, volatile-induced melting regime beneath mid-ocean ridges. In the absence of converging lines of evidence

The formation of primary arc basalts during subduction is attributed to the high-pressure dehydration of serpentinite (deserpentinisation), which generates fluids that flux and melt the overlying mantle wedge. These basalts are more oxidised than their mid-ocean ridge counterparts. At the slab surface of current subduction zones, these deserpentinisation fluids are intrinsically oxidised but, owing to the low sulphur content of subducted serpentinite, they only result in a low mantle-wedge oxidation rate, which cannot account for the oxidised source of arc basalts. Through a combination of natural observations and thermodynamic modelling, we show that infiltration of sediment-derived fluids can alter the oxidation capacity of deserpentinisation slab fluids, providing an explanation for the oxidised nature of arc volcanism. The deserpentinisation oxidation capacity is mainly influenced by the stability and abundance of dissolved oxidised aqueous species of redox-sensitive elements, notably sulphate, rather than the oxidation state of the sediment. Infiltration of CH4-bearing fluids derived from graphite-bearing sediment reduces the oxidant capacity of deserpentinisation fluids, leading to a relatively low fO2 observed in natural metaperidotite. On the other hand, infiltration of sulphate-CO2-bearing, sediment-derived fluids results in a high oxidation capacity of deserpentinisation fluids in cold and hot subduction zones, leading to a global mantle-wedge oxidation rate of 3.5 km3·yr¿1. Therefore, these slab fluids can oxidize the mantle wedge at a rate similar to that of arc basalt generation, explaining the oxidised nature of arc volcanism. The findings suggest that sediment-derived fluids (and potentially also fluids derived from the Altered Oceanic Crust) can significantly modulate the oxidation capacity of deserpentinisation slab fluids, highlighting the importance of slab fluid-mixing in the volatile recycling in the subduction zone system.

The major apatite ores from the Central High-Atlas Mesozoic alkaline intrusive complexes (Morocco) have been investigated through integrated field, petrographic, and geochemical studies (Ouabid et al., 2021). The apatite deposits occur as veins ¿few mm to several cm thick¿, spatially associated with syenite bodies. Gem-quality crystals of apatite¿up to 15 cm in size (e.g., Ouabid and Raji, 2023)¿, are associated with albite, clinopyroxene (hedenbergite¿augite¿diopside), amphibole (hornblende¿edenite¿actinolite), K-feldspar, quartz, magnetite, titanite, epidote, prehnite, and calcite. Two types of apatite have been distinguished according to their halogen contents including F- and Cl-apatites. Both types are enriched in rare earth elements (0.5-2 wt.% ¿REE) and have major¿trace element contents consistent with a magmatic origin, corroborated by the chemical similarity with accessory apatite in the host alkaline intrusions, as well as with other reference igneous apatites. Though, hydrothermal fluids percolating through country sedimentary rocks were also implicated, as revealed by the striking Na enrichment observed in the Cl-rich apatites, likely inherited from Triassic evaporites, showing close spatial relationships with the alkaline intrusions.

Evidence of Paleogene subduction in the Western Mediterranean is scattered throughout high-pressure (HP) events in polymetamorphic basement terrains of disputed and differing provenance exposed along the margins of deep basins formed during Miocene to Quaternary slab retreat. The reconstruction of the original subduction geometry is often hindered by pervasive Miocene high-temperature (HT) ductile overprinting of earlier HP events. Here, we report rutile U¿Pb ages, EBSD-determined microstructures and thermodynamic modeling in eclogites from the upper-plate domain of the Betics (Ojén nappe, Alpujárride complex). These data show that the Ojén eclogite peak mineral assemblage formed during HP (1.3 ±0.15 GPa) and HT (700 ± 25 °C) ductile flow in the early Oligocene (30 ± 2.5 Ma). We interpret this event as a collisional phase linking Ojén eclogites to the Paleogene subduction of the western Tethys. Contrasting P¿T gradients of eclogites and arc volcanism between Corsica¿Calabria and Kabylies¿Betics point to a highly segmented orogenic system. NW¿SE-oriented transform faults separated a continental collision domain to the South from an oceanic subduction to the NE, with two subducting slabs ¿Calabrian and Algerian¿ retreating respectively towards the E and SE. Early Eocene inversion of the European and Maghrebian margins and ensuing early Oligocene collision resulted in (U)HP¿HT metamorphism in the Alpujárride and Kabylies. This orogen underwent extensional collapse in the late Oligocene-Aquitanian, before further shortening and the intracrustal emplacement of the Ronda peridotite. The upper plate Ojén eclogites and orogenic peridotites drifted westwards following the retreating Betics¿Rif Tethys slab in the early Miocene, after transform collapse that induced HP subduction of the SE Iberian margin below the Alpujárrides.