Tahar Hammouda

High-pressure melting of carbonated eclogite and experimental constraints on carbon recycling and storage in the mantle

Abstract High-pressure experiments (5–10 GPa, corresponding to approximately 150–300 km depth in the mantle) have been conducted on a basalt+calcite mixture in order to constrain the fate of carbonates carried on subducted ocean floor. At 5 GPa, carbonate breakdown occurs between 1100 and 1150°C, and coincides with silicate melting. At 6.5 GPa and above, only carbonatitic melts were produced and the solidus temperature is located below 1000°C. Liquid immiscibility is observed at the transition from silicate to carbonate melting (6 GPa and 1300°C). The carbonatitic solidus in the eclogite is located 4 GPa higher in pressure than in the peridotitic system. This is due to the difference of silicate mineralogies involved in carbonation reactions. In addition, carbonatites produced in the present study are calcium-rich (Ca/(Ca+Fe+Mg) ca. 0.80), in striking contrast to those produced by melting of carbonated peridotite (Ca/(Ca+Fe+Mg) ca. 0.50). Carbonated eclogite should therefore be considered as a potential source for the most abundant carbonatite type worldwide, but it is stressed that carbonatitic magmatism could be a multistage process. Compared to pressure–temperature paths of subducting slabs, the present results suggest that carbonates will most likely be removed from the slab before reaching 300 km, and are unlikely to be introduced by subduction either in the transition zone or in the lower mantle. Therefore, the deep carbon cycle appears to be restricted to the upper mantle (300 km or shallower depths). Carbonate-enriched portions located in cooler parts of the slab (fractures) could allow for oxidized carbon introduction to deeper mantle regions, but more experiments at higher pressures are necessary to evaluate this hypothesis. Because carbonatite production from carbonated eclogites occurs in the diamond stability field, the present experimental results lend further support to recent models of diamond formation involving carbonated melts in the mantle.

Ultrafast mantle impregnation by carbonatite melts

Carbonatitic melts are often invoked as responsible for metasomatism in the mantle because of their unique chemical and physical properties. Here we report on infiltration experiments demonstrating that such melts can percolate very quickly in polycrystalline olivine. Carbonatites can travel over several millimeters in one hour and the infiltration rate is kinetically controlled by cation diffusion in the melt. The observed rates are several orders of magnitude higher than those previously found for basalt infiltration in mantle lithologies. Infiltration proceeds by a dissolution-precipitation mechanism wherein porosity is created in the dunite by dissolution of olivine at grain edges. This reaction is accompanied by forsterite reprecipitation in the carbonatite reservoir. Such a mechanism would likely favor chemical exchange between melt and matrix during percolation. We propose a migration model combining infiltration and compaction by which carbonatite melts can travel upward in the mantle over hundreds to thousands of meters on time scales of 0.1–1 m.y.

ISOTOPIC EQUILIBRATION DURING PARTIAL MELTING : AN EXPERIMENTAL TEST OF THE BEHAVIOUR OF SR

Abstract Experiments using isotopically enriched, Sr doped minerals designed to test for isotopic equilibrium between source and melt during partial melting reveal that 87 Sr/ 86 Sr ratios of the liquids are primarily determined by the proportions of minerals consumed and vary with the advancement of the melting reaction. The experiments were performed at 1 atm on model crustal assemblages composed of pairs of natural plagioclase (An 68 ; 87 Sr/ 86 Sr= 0.701 ) and synthetic fluorphlogopite doped with 90 ppm Sr having 87 Sr/ 86 Sr= 4.2 . SIMS traverses showed that during the initial stages of the reaction, liquids are isotopically zoned. All the analyzed melts have 87 Sr/ 86 Sr markedly higher than that of the bulk starting assemblage (i.e. the source), because of the faster melting rate of fluorphlogopite. At 1200°C and 1250°C melting occurs above a critical temperature, wherein the dissolution rates of the crystals are controlled by diffusion of species in the melt, and reactants and reaction products are out of isotopic equilibrium. This is due to faster melt-crystal boundary migration when compared to Sr diffusion in the crystals. Equilibration is possible only if melting stops. Calculations show that total equilibration between melt and residue by Sr tracer diffusion in the crystals takes 10 4 –10 6 yr (for temperature and grain size ranging, respectively, from 800°C to 1000°C, and 0.1 to 1 cm). When compared to the proposed residence time of crustal magmas at their sources, this result strongly suggests that magmas that do not reflect the bulk isotopic characteristics of their source regions can be produced.

Comparison of solid-state crystallization of boron polymorphs at ambient and high pressures

Here we report the systematic study of solid-state phase transformations between boron polymorphs: α -B12, β -B106, γ -B28, T-B52 and amorphous boron (am-B). It is evident that the Ostwald rule of stages plays an important role during phase transformations not only of amorphous boron, but also of crystalline forms. We have observed the crystallization of tetragonal boron T-B52 from amorphous phase of high purity (99.99%), which, however, cannot be easily distinguished from B50C2 boron compound. Many factors influence the transformations of amorphous phase, and it is possible to observe not only well-known am-B → α -B12 and am-B → β -B106 transformations, but also am-B → T-B52, never reported so far. At ∼14 GPa, the crystallization order becomes β -B106→α -B12→γ -B28, while at ∼11 GPa the intermediate crystallization of T-B52 still was observed. This unambiguously indicates that α -B12 is more thermodynamically stable than β -B106 at high pressures (HPs) and renders possible to transform, at least partially, common β -phase of high purity into α -B12 at very HPs and moderate temperatures (below 1600 K), i.e. outside the domain of its stability.

Poisson's Ratio and the Glass Network Topology - Relevance to High Pressure Densification and Indentation Behavior

Although Poissons ratio (ν) is a macroscopic elastic parameter it depends much on the fine details of the atomic packing. Glasses exhibit a wide range of values for  from 0.1 to 0.4 which correlate to the glass network polymerisation degree, hence reproducing at the atomic scale what is observed in cellular materials at the macroscopic scale[1]. As for pure oxide glasses, we found in various multi-component glasses built on ionic-, covalent- or Van der Waals bonds that an increase of Poisson’s ratio corresponds to a decrease of the atomic network crosslink degree[2]. Noteworthy, an extension of this analysis to the case of metallic glasses correlate the recently proposed cluster-like network structure for these glasses[3,4]. A general feature is that a highly cross-linked atomic network results in a glass with a low atomic packing density (large free volume fraction), as exemplified with the case of amorphous silica. The lower the atomic packing density is and the larger the volume change the glass experiences under high pressure (1 to 25 GPa). Indentation experiments with sharp indenters (such as the Vickers one) give birth to hydrostatic stresses of the same order of magnitude and thus induce glass densification. There is hence a direct correlation between ν (reflecting the packing density) and the indentation behavior[5].

Control of redox state and Sr isotopic composition of granitic magmas: a critical evaluation of the role of source rocks

The current underlying assumption in most geochemical studies of granitic rocks is that granitic magmas reflect their source regions. However, the mechanisms by which source rocks control the intensive and compositional parameters of the magmas remain poorly known. Recent experimental data are used to evaluate the ‘source rock model’ and to discuss controls of (1) redox states and (2) the Sr isotopic compositions of granitic magmas. Experimental studies have been performed in parallel on biotite-muscovite and tourmaline-muscovite leucogranites from the High Himalayas. Results under reducing conditions ( = FMQ – 0·5) at 4 kbar and variable suggest that the tourmaline-muscovite granite evolved under progressively more oxidising conditions during crystallisation, up to values more than four log units above the FMQ buffer. Leucogranite magmas thus provide an example of the control of redox conditions by post-segregation rather than by partial melting processes. Other experiments designed to test the mechanisms of isotopic equilibration of Sr during partial melting of a model crustal assemblage show that kinetic factors can dominate the isotopic signature in the case of source rocks not previously homogenised during an earlier metamorphic event. The possibility is therefore raised that partial melts may not necessarily reflect the Sr isotopic composition of their sources, weakening in a fundamental way the source rock model.

Role of iron and reducing conditions on the stability of dolomite + coesite between 4.25 and 6 GPa – a potential mechanism for diamond formation during subduction

We have investigated the effect of iron and oxygen fugacity on dolomite + coesite stability during subduction. For redox conditions buffered by the assemblage itself, the presence of iron (Fe/(Fe + Mg) ca . 0.4) lowers the decarbonation reaction by about 200 °C at 4.25 GPa and by 300 °C at 5.5 GPa, compared to the iron-free reaction. Clinopyroxene and CO 2 form by a decarbonation reaction similar to the iron-free system. At low temperature, however, graphite replaces CO 2 through redox interactions with iron in the carbonate. Melting occurs approximately 100 °C above decarbonation and a carbonatitic melt is produced. In a second series of experiments, we imposed lower oxygen fugacity by adding molybdenum, in order to study the potential redox mechanisms in contact with the peridotitic mantle. In these samples, we observe systematic carbon reduction producing the assemblage clinopyroxene + graphite. Our results show that the stability of dolomite + coesite on a subduction path is limited by redox interactions, in addition to pressure and temperature. As our experiments were run in the stability field of diamond, we also demonstrate that diamond may form from dolomite + coesite during subduction. Chemical diffusion of iron and oxygen in the slab and at the slab/mantle interface appears to be a key parameter to determine at what pressure–temperature conditions this may happen.

First field-scale occurrence of Si-Al-Na–rich low-degree partial melts from the upper mantle

Felsic dikes intruding an orogenic lherzolite body from the North Pyrenean zone (France) show great compositional similarities to some of the Si-Al-Na(K)–rich glasses found as microscopic inclusions in peridotite xenoliths. As shown by Sr-Nd isotopes, these silicic magmas originated from a mantle reservoir that was moderately depleted on a time- integrated basis. Their Si-Al-Na–rich bulk compositions match closely those predicted by experimental studies relevant to near-solidus melting of mantle peridotite. Specifically, the dikes are closely similar to low-degree melts from sodic harzburgite. On this basis, an olivine-orthopyroxene source that had been metasomatized by addition of sodic-carbonatite melts is inferred, as also suggested by peculiar trace element features (extreme fractionation of the light rare earth elements, high Nb contents). This new occurrence lends support to the existence of high-silica liquids in the mantle and provides evidence for their efficient extraction to form relatively large batches.

Compact multianvil device for in situ studies at high pressures and temperatures

We describe a recently developed device for in situ studies at pressures up to 25 GPa and temperatures up to 2300 K. The system consists of a 450 ton V7 Paris-Edinburgh press combined with a Stony Brook ‘T-cup’ multianvil stage. Such a compact large-volume set-up has a total mass of 100 kg only and can be readily used on most synchrotron radiation facilities. The optimization of the set-up by off-line tests is detailed, and we present some X-ray diffraction results which demonstrate the potential of the technique.

An experimental investigation of radon diffusion in an anhydrous andesitic melt at atmospheric pressure: implications for radon degassing from erupting magmas

Abstract We have measured the diffusivity of radon in a dry natural andesitic melt at atmospheric pressure and at temperatures ranging between 1350°C and 1500°C. The procedure is based on the determination by gamma-ray spectrometry of radon losses undergone by a disk-shaped melted sample during heating. A theoretical treatment of radon diffusion allowed us to relate these losses to diffusivities and values of D Rn can be described by the following Arrhenius equation: D Rn = D 0 · exp(− E a / RT ), with D 0 = 1.5 10 4 m 2 · s −1 and E a = 466 kJ · mol −1 . Our results confirm that the larger the ionic radius of the noble gas, the higher the frequency factor D 0 and the activation energy E a . At a temperature of 1225 ± 25°C the diffusivity of radon approaches that of other heavy noble gases (log D Rn = −12.0 ± 0.1 where D Rn is in m 2 · s −1 ). From the present study it is demonstrated that neither diffusion processes, which are too slow and balanced by radioactive ingrowth of radon from its parent, nor exsolution of pure radon bubbles, because of its exceedingly low abundance in the magma, can be responsible for the complete radon depletion observed in erupted lavas. The exsolution of water and other major gas phases acting as carriers of trace gaseous species appears to be the main cause for radon as well as other noble gases degassing.