Final Report Summary - GOBMEUM (Generation of Basaltic Magmas in the Earth’s Upper Mantle) The project focused on the determination of melting phase diagrams of mantle peridotite in the Earth’s uppermost mantle (0.7 to 3 GPa, corresponding to depths of about 20-90 km in the Earth). Such melting phase diagrams are essential for our understanding of how melting occurs inside our planet, and its live expression in the form of volcanic eruptions at the bottom of the ocean floors, where tectonic plates spread apart from each other. To get a grip on such issues is significant, since partial melting in the Earth’s mantle is the only recognizable process that has generated continents, shallow-and-deep ocean floors, atmosphere, and hydrosphere. Voluminous basaltic magmas erupt at mid-oceanic ridges (mid-ocean ridge basalts, MORB) as a consequence of mantle upwelling and melting beneath spreading plates. Every year, millions of tones of magma erupt at such sites. For instance, oceanic crust covers is >60% of the surface of the Earth, while MORB are >75% of total volcanic rocks (volume erupted/year). Volumetrically less significant (about 10 times smaller than MORB) are magmatic eruptions at oceanic 'hot-spot' (intraplate) locations such as Hawaii and Iceland (ocean-island basalts, OIB). However, because the geochemistry of OIB lavas is distinct from MORB, both have great petrogenetic significance and carry important information about the chemical and physical properties/dynamics of the mantle. Hence, to understand how mantle melts, and generates basaltic magmas, is a first-order task. In fact, process of mantle melting can actually be extended to all of the oceanic crust, which necessarily includes the igneous (magmatic) part. This is because, not only since the latter is volumetrically the most significant (about 6 times the volume of MORB, about 60-80% of the total global Cenozoic production of magma/year), but also because it carries crucial information about mantle processes and crust accretion that basalts, in general, do not.At the time of the project termination date, the first work package of the GoBMEUM project was completed with 47 experiments performed in a piston cylinder apparatus at the Bayerisches Geoinstitut (BGI), Germany. These melting experiments were performed under anhydrous conditions using glasses as starting material. The starting glasses (materials) were synthesized in the systems CaO-MgO-Al2O3-SiO2-FeO-Cr2O3 (CMAS-Fe-Cr; 2 different starting compositions; 30 experiments) and CaO-MgO-Al2O3-SiO2-Na2O-FeO (CMASNF; 4 different starting compositions; 17 experiments) systems in 1 atm furnace at a temperature of 1600°C. Prior to these experiments, Dr Keshav performed thermodynamic computations, using MELTS, pMELTS, and his own optimized database. In pressure-temperature projections, these computations revealed clear differences between the different models. While MELTS and pMELTS predict a negative plagioclase-spinel peridotite solidus transition slope, meaning that there would be no melting, Dr Keshav's own computations suggest a positive slope, suggesting that melting would continue, if an adiabat were to intersect this slope. And this is exactly what we see in nature, that is, at mid-ocean ridges - basaltic magmas do erupt. These thermodynamic models were done in collaboration with Dr. Gudmundur Gudfinnsson (Reykjavik) and Dr. Max Tirone (Bochum). The results of the experiments conducted for this project in the CMAS-Fe-Cr system confirmed Dr Keshav's thermodynamic predictions, i.e. that the plagioclase-spinel peridotite solidus transition has a positive Clapeyron slope (dP/dT = L/Tdv, where P is pressure, T is temperature, L is the specific latent heat, and v is volume), in contrast to MELTS and pMELTS that predict a negative slope. While computations and experiments in the system CMAS-Fe-Cr show internal consistency, the same cannot be said about what might be happening in the system CMASNF, which spans the plagioclase-spinel solidus transition (thus far studied in these two systems). Theoretically, we are discovering that sodium (Na) and its interaction with calcium (Ca) in plagioclase, in terms of activity-composition (a-X) relations, has a major effect on how the plagioclase-spinel peridotite solidus transition might look in P-T space, and in the composition space encompassing CMASNF. This is because, at the moment, there is no available fundamental macro-thermodynamic theory on interaction of Na with Ca. There are quite a few theoretical models (including THERMOCALC and PERPLEX) on this interaction aspect, but we are finding out that they all predict different scenarios. The project stopped when Dr Keshav was refining his approach to go beyond the standard models of thermodynamics, and work with statistical mechanics on the interaction between Na and Ca in plagioclase.
