Dynamics of supersaturated magmatic systems
Whether expressed on the surface as lava or crystallised in the form of plutons far below the surface, magmas appear to be systematically multiphase. The different phases can be crystalline, gaseous, or liquid, and all three forms may possibly coexist. The appearance of crystals or gas bubbles depends on their ease of nucleation in the liquid magma. This nucleation is always subject to a delay that will naturally induce a supersaturation of the magma. We wish to study the simultaneous effect of this supersaturation on the exsolution of gases (Figure A)
Fig. A Bubbles in a basaltic glass, revealed by tomography (each side measures approximately 6 mm), and on the growth of minerals (Figure B).
Our study combines an experimental approach with natural targets. Several approaches will be used to study the process of nucleation : Experiments on programmed decompression (collaboration with D. Laporte, LMV Clermont-Ferrand) will be conducted on silicate liquids containing He and Ar. The exsolved gases will be analysed, bubble by bubble, by laser extraction at CRPG to determine the respective roles played by solubility and diffusivity during degassing. The chemical imbalance measured between a bubble and the glass encasing it should provide new time constraints on the degassing process that triggers volcanic eruptions. This approach will lead to very strong interactions with the theme Cycles, Atmosphere, Climate, which aims to describe, in both a qualitative and quantitative manner, cryptic volcanic degassing in order to quantify the volcanic contribution of He, Ar and CO2 to the atmosphere.
The delay in nucleation of certain minerals (including plagioclase, pyroxene, and others) will be measured in simplified synthetic systems at one atmosphere to quantify the effect of overheating on the time and temperature of the appearance of these crystalline phases. The magnitude of the delay, which is proportional to the supersaturation created within the magma, translates into rapid growth textures (typically skeletal or dendritic). Additionally, because nucleation and growth are time-dependent processes, we will study the competition between these two processes in magmatic rocks to provide kinetic constraints on the presence of several generations of minerals of the same phase. The experimental suppression of the nucleation of second generation phases will be used to determine the metastable extensions of liquidus surfaces. This suppression can be achieved either by superheating the liquid relative to the theoretical temperature of appearance of the second phase or by reducing the volume of liquid in which the nucleation should occur. For this purpose, synthetic or natural magmatic inclusions will be used. The development of such diagrams of metastable phases will provide real degrees of supersaturation (or supercooling) and not just simple approximations, as is currently the case with thermodynamic models.
The crystal skeleton of an olivine trapping magma inclusions ; fine dendrites are observed at the four corners of the crystal.
This experimental approach has to be combined with observation of natural objects, and two natural targets have already been identified. The first target involves the volcanic stacks of the Bushveld (4200 m of lavas, from rhyodacitic to andesitic). Most of these lavas show synchronous vesiculation during each eruption (sometimes in two successive phases), with textures indicating rapid growth of minerals. HHeterogeneous matrix textures are also present which may have resulted from the formation of a second immiscible phase. Our objective is to reconstruct the thermal history and kinetics of all identified instances of supersaturation (gas or liquid) and their petrogenetic settings. The second target is involves acid-alkaline magma mixing (mingling). Although, in most cases, the textures are microgranular (millimetre sized crystals), there are a few examples of oriented growth that has produced large (several cm) crystals. The goal is to understand, using natural examples, the role played by supersaturation (related to ΔT for magmas) in the nucleation and growth of these crystals.
Tracing the evolution of magmatic fluids with vitreous inclusions
The processes of partial melting of the mantle and transfer of magmas to the surface are crucial in terrestrial differentiation. However, these processes are difficult to quantify because the magmas are generally chemically disrupted by assimilation and fractional crystallisation before emplacement. Vitreous inclusions correspond to droplets of liquid magma that were trapped at different stages of magmatic evolution. These inclusions are, therefore, valuable objects for the study of magmas, at all stages from their creation to their emplacement, and of the eruptive mechanisms and the dynamics of the processes that affect them. They also allow the problem of interaction of the magma with the mantle or the crust during transfer to be addressed. The ion microprobe allows in situ analysis of the concentrations and isotopic compositions of many elements and compounds (H2O, CO2, F, Cl, Li, B, REE and HFSE, δD, δ7 li, δ18O, δ13C, δ34S, and others) in these vitreous inclusions and is therefore a particularly suitable tool for the characterisation of processes at different stages of magmatic evolution. We will focus on the following aspects : Partial melting in different geodynamic settings, including continental and oceanic rifts, oceanic islands and subduction zones, with the aim of describing the mechanisms of the partial melting and the contribution of recycled oceanic crust and/or sediment as a source for the magmas. Interactions with the crust during the percolation of magmas in order to quantify the processes of assimilation and fractional crystallisation. Magma chambers and eruptive processes, to study the behaviour of volatile elements (H2O, CO2, and others), whose abundance plays a critical role in the eruptive style, and to determine the residence time of magmas at different levels using the diffusion profiles of elements in minerals (in collaboration with N. Métrich, IPGP).
The origin of the primordial chemical and isotopic heterogeneities of the terrestrial mantle can be understood by studying the magmatism of hot spots that were initiated at different mantle depths. From this perspective, we will determine the concentrations and isotopic signatures of some siderophile elements in ocean island basalts (OIBs) produced at different depths of melting. These elements could play a major role in the chemical composition of the core-mantle interface and in the deep mantle ; the goal will be to determine whether there are elemental and isotopic variations recorded in OIBs and to deduce the processes operating in the deep mantle. These heterogeneities are also characterised by significant variations in the isotopic composition of helium. The minimum spatial scale of mantle heterogeneity is necessarily greater than the distance of helium diffusion. A good knowledge of the diffusivity of helium will allow us to constrain geochemically the spatial scale of heterogeneity of the mantle. In collaboration with J. Badro (IPGP) and J. Duprat (ORSAY), we propose to determine the rate of diffusion of noble gases given the conditions and composition of the Earth’s mantle.