Search

International audience; Rare earth element (REE) enrichments in carbonatites are often described as resulting from late magmatic-hydrothermal or supergene processes. However, magmatic pre-enrichment linked to the igneous processes at the origin of carbonatites are likely to contribute to the REE fertilisation. Experimental constraints reveals that immiscibility processes between carbonate and silicate melts can lead to both REE enrichments and depletions in carbonatites making the magmatic processes controlling REE enrichments unclear.We link REE contents of carbonatites to the magmatic stage at which carbonatites are separated from silicate magma in their course of differentiation. We present results of experiments made at pressure and temperature conditions of alkaline magmas and associated carbonatites differentiation (0.2 to 1.5 GPa; 725 to 975 °C; FMQ to FMQ +2.5), simultaneously addressing crystal fractionation of alkaline magmas and immiscibility between carbonate (calcio-carbonate type) and silicate melts (nephelinite to phonolite type). The experimental data shows that the degree of differentiation, controlling the chemical composition of alkaline melts, is a key factor ruling the REE concentration of the coexisting immiscible carbonate melts. In order to predict carbonate melt REE enrichments during alkaline magma differentiation, we performed a parameterisation of experimental data on immiscible silicate and carbonate melts, based exclusively on the silica content, the alumina saturation index and the alkali/alkaline-earth elements ratio of silicate melts. This parameterisation is applied to more than 1600 geochemical data of silicate magmas from various alkaline provinces (East African Rift, Canary and Cape Verde Islands) and show that REE concentrations of their potential coeval carbonatite melts can reach concentration ranges similar to those of highly REE enriched carbonatites (∑REE > 30 000 ppm) by immiscibility with phonolitic/phono-trachytic melt compositions, while more primitive alkaline magmas can only be immiscible with carbonatites that are not significantly enriched in REE.

International audience; Magma ascending from the storage region toward the surface may crystallize small and rapidly-grown crystals, designated as microlites. Upon decompression, the rapid changes of the microlite textures, such as number density and crystal size, directly impact the rheology of the magma in the volcanic conduit, and eventually control the effusive vs. explosive style of the subaerial eruption. This is of prime importance for volcanic risk assessment involving highly viscous silica-rich magmas that are prone to fragmentation. To this aim, we present a theoretical modeling of plagioclase nucleation and growth in rhyolitic-like melts, based on previous decompression experiments. Thus, the modeling is valid for plagioclase crystallization in rhyolitic melts decompressed at 875 • C from 200 MPa to final pressures of 50, 75, or 100 MPa, which represents effective undercooling (T eff) of 110, 80 and 55 • C, respectively. Our results of nucleation-rate calculation using the Classical Nucleation Theory (CNT; case of homogeneous nucleation) and values of crystal-melt interfacial energy (σ) either from literature or empirically calculated, strongly disagree with previously-determined experimental nucleation rates. By inverting the CNT calculation using the experimentally-determined nucleation rates, we propose plagioclase-liquid interfacial energies from 0.041 to 0.059 J.m −2 with T eff increasing from 55 to 110 • C, which is about 2-3 times lower than σ determined empirically and macroscopically. We modeled plagioclase growth rates by atom diffusion in melt (following Fick's second law adapted to multicomponent systems), considering a crystal-melt interface advancing over time (as crystal grows) and a chemically-closed finite reservoir (no component supply). The component limiting plagioclase growth has been determined to be CaO, for which we calculated diffusion coefficients from 10 −14 to 10 −15 m 2 /s (conditions of a silicic melt, 875 • C, and H 2 O saturation pressures from 50 to 100 MPa). The overall good agreement between the model and the experimental growth laws validates diffusion as the main process controlling isothermal decompression-induced plagioclase growth under moderate T eff < 110 • C. Combining both calculations, crystal number density and size, may be relevant to predict plagioclase microlite textures in ascending rhyolitic magmas, which may now be incorporated in conduit flow models to assess the magma rheological behavior that controls eruption dynamics.

International audience; Decompression melting of the upper mantle produces magmas and volcanism at the Earth's surface. Experimental petrology demonstrates that the presence of CO2 and H2O enhances peridotite melting anywhere within the upper mantle down to approximately 200–300 km depth. The presence of mantle melts with compositions ranging from carbonate-rich to silicate-rich unavoidably affects the geophysical signals retrieved from Earth's mantle. Geochemical investigations of erupted intraplate magmas along with geophysical surveys allow for constraining the nature and volume of primary melts, and a sound formalism is required to integrate these diverse datasets into a realistic model for the upper mantle including melting processes. Here, we introduce MAGLAB, a model developed to calculate the composition and volume fraction of melts in the upper mantle, together with the corresponding electrical conductivity of partially molten mantle peridotites at realistic pressure-temperature conditions and volatile contents. We use MAGLAB to show how the compositions of intraplate magmas relate to variations in lithosphere thickness. Progressive partial melting of a homogeneous peridotitic mantle source can in theory create the diversity of compositions observed among the spectrum of intraplate magma types, with kimberlite melts beneath thick continental shields, alkaline magmas such as melilitite, nephelinite and basanite beneath thinner continents and relatively old plus thick oceanic lithospheres, and ‘regular’ basalts beneath the youngest and thinnest oceanic lithospheres as well as beneath significantly thinned continental lithospheres. MAGLAB calculations support recent experimental findings about the role of H2O in the upper mantle on producing primary kimberlitic melts in addition to CO2. We demonstrate the robustness of MAGLAB calculations by reproducing the compositions of erupted melts as well as associated mantle electrical conductivities beneath the Society hotspot in the Pacific Ocean. A comparison of our simulations with magnetotelluric surveys at various oceanic settings shows that the heterogeneities in electrical conductivity of Earth's upper mantle are related to variations in volatile content via the presence of small (generally <

International audience; In paleo-accretionary prisms and the shallow metamorphic domains of orogens, circulating fluids trapped in inclusions are commonly composed of a mixture of salt water and methane, producing two types of fluid inclusions: methane-bearing aqueous and methane-rich gaseous fluid inclusions. In such geological settings, where multiple stages of deformation, veining and fluid influx are prevalent, textural relationships between aqueous and gaseous inclusions are often ambiguous, preventing the microthermometric determination of fluid trapping pressure and temperature conditions. To assess the P-T conditions of deep circulating fluids from the Hyuga unit of the Shimanto paleo-accretionary prism on Kyushu, Japan, we have developed a new computational code, applicable to the H2O-CH4-NaCl system, which allows the characterization of CH4-bearing aqueous inclusions using only the temperatures of their phase transitions estimated by microthermometry: Tmi, the melting temperature of ice; Thyd, the melting temperature of gas hydrate and Th,aq, homogenization temperature. This thermodynamic modeling calculates the bulk density and composition of aqueous inclusions, as well as their P-T isochoric paths in a P-T diagram with an estimated precision of approximatively 10 %. We use this computational tool to reconstruct the entrapment P-T conditions of aqueous inclusions in the Hyuga unit, and we show that these aqueous inclusions cannot be cogenetic with methane gaseous inclusions present in the same rocks. As a result, we propose that pulses of a high-pressure, methane-rich fluid transiently percolated through a rock wetted by a lower-pressure aqueous fluid. By coupling microthermometric results with petrological data, we infer that the exhumation of the Hyuga unit from the peak metamorphic conditions was nearly isothermal and ended up under a very hot geothermal gradient. In subduction or collision zones, modeling aqueous fluid inclusions in the ternary H2O-CH4-NaCl system and not simply in the binary H2O-NaCl is necessary, as the addition of even a small amount of methane to the water raises significantly the isochores to higher pressures. Our new code provides therefore the possibility to estimate precisely the pressure conditions of fluids circulating at depth.

International audience; We present here new experimental data on H2O-CO2 solubility in mafic melts with variable chemical compositions (alkali basalt, lamproite and kamafugite) that extend the existing database. We show that potassium and calcium-rich melts can dissolve ∼ 1 wt% CO2 at 3500 bar (350 MPa) and 1200°C, whereas conventional models predict solubilities of 0.2-0.5 wt%, under similar P-T conditions. These new data, together with those in the literature, stress the fundamental control of melt chemical composition on CO2 solubility. We present a semi-empirical H2O-CO2 solubility model for mafic melts, which employs simplified concepts of gas-melt thermodynamics coupled with a parameterization of both chemical composition and structure of the silicate melt. The model is calibrated on a selected database consisting of 289 experiments with 44 different mafic compositions. Statistical analyses of the experimental data indicate that, in mafic melts, the chemical composition and therefore the structure of the melt plays a fundamental role in CO2 solubility. CO2 solubility strongly depends on the amount of non-bridging oxygen per oxygen (NBO/O) in the melt, but the nature of the cation bonded to NBO is also critical. Alkalis (Na+K) bonded to NBO result in a strong enhancement of CO2 solubility, whereas Ca has a more moderate effect. Mg and Fe bonded to NBO have the weakest effect on CO2 solubility. Finally, we modelled the effect of water and concluded that H2O dissolution in the melt enhances CO2 solubility most likely by triggering NBO formation. In contrast with CO2 but in agreement with earlier findings, H2O solubility in mafic melts is negligibly affected by melt composition and structure: it only shows a weak correlation with NBO/O.

International audience; A novel experimental setup was used to measure in-situ variations of electrical conductivity (EC) during deformation in torsion (simple shear) at 300 MPa confining pressure and temperatures between 873 and 1473 K. This setup is designed to test if deformation of partially molten systems can produce electrical anisotropy. The motivation for this study comes from the observation that the Lithosphere–Asthenosphere Boundary (LAB) at mid-ocean ridges and in particular at the East Pacific Rise is strongly electrically anisotropic. In an initial set of calibration experiments, the variation of EC with temperature (873–1473 K) was determined for Carrara marble, Åheim dunite and basalt-bearing olivine aggregates. EC was then monitored during deformation experiments at 1473 K and measured in the frequency range between 6 MHz and 1 Hz. The electrical response of the different materials tested as a function of frequency, changes significantly depending on the presence, absence, proportion and distribution of melt contained in the specimen. Melt-free samples show a single conduction mechanism whereas melt-bearing samples display two conduction mechanisms linked in series, reflecting the contribution of isolated and connected melt. Impedance was measured along the sample radius, in a direction parallel to the shear gradient inherent in torsion experiments. During the tests, increasing values of the impedance measured suggest that the long range melt connectivity decreases radially, and melt drains from low to high shear stress regions. The conductivity, calculated from impedance measurements, is low and comparable to values measured along mid-ocean ridges. We suggest that electrical anisotropy of the LAB reflects an alternation of melt-enriched and melt-depleted channels elongated in the spreading direction possibly induced by spreading velocity gradients along the ridge. This implies that the observed electrical anisotropy reveals larger scale processes than strain-induced generation of crystallographic preferred orientations. Such large-scale processes could influence the distribution of seamounts and chemical variations of mid-ocean ridge basalts.