Didier Laporte

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.

Experimental and theoretical constraints on melt distribution in crustal sources: the effect of crystalline anisotropy on melt interconnectivity

In partially molten systems, the equilibrium distribution of melt at the grain scale is governed by the principle of interfacial energy minimization. In ideal sources (i.e. partially molten rocks that are monomineralic, have single-valued solid-liquid and solid-solid interfacial energies, and are subject to hydrostatic stress) the wetting angle Θ is known to be a unique characteristic which specifies the melt configuration for a given melt fraction. Crustal rocks cannot be modelled as ideal sources because of their polymineralic nature, the moderate to high anisotropy of interfacial energies which characterizes common refractory minerals, and the possible presence of a crystallographic preferred orientation. That partially molten crustal rocks depart from ideal sources is documented by a series of high-P, high-T experiments illustrating the textural relationships of biotite and amphibole with silicic melts. The melt distributions observed in these experiments differ significantly from those expected in ideal sources: (1) crystal-melt interfaces are commonly planar, rational faces rather than smoothly curved, irrational surfaces; and (2) the concept of a unique wetting angle does not hold as shown in the biotite-silicic melt system. These textural features demonstrate that anisotropy of crystal-melt interfacial energy is a factor of primary importance in modelling the grain-scale distribution of partial melts. The petrological implications of our study are the following: 1. (1) At high degrees of anisotropy and low melt fractions, melt is predicted to form isolated, plane-faced pockets at grain corners. The overall shape of these pockets, and therefore the value of the connectivity threshold Φc, are expected to be very sensitive to the ratio of solid-solid to solid-liquid interfacial energies, γssγsl (Φc is the melt fraction at which melt interconnectivity is established). Melt pockets with low volume-to-surface ratio, and low (but non-constant) wetting angles should prevail at high γssγsl, resulting in very low values of Φc (≤1 to a few vol%). Higher values of Φc, a high volume-to-surface ratio of melt pockets, and high wetting angles are expected at low γssγsl. 2. (2) The wetting angle at hornblende-hornblende-melt junctions, at 1200 MPa-975°C, is 25°. A review of existing data indicates that quartz-melt and feldspar-melt wetting angles are also low to moderate (12–60°). A very low value of Φc should, therefore, be the general rule during crustal anatexis. In particular, a connectivity threshold lower than 3–4 vol% is predicted for partially molten amphibolite. 3. (3) In biotite-rich rock-types, such as melanosomes in migmatites, the combination of a pronounced crystalline anisotropy and a marked preferred orientation of mica flakes leads to a very low permeability (normal to layering). Biotite-rich melanosomes should therefore impede chemical interactions between neighbouring leucosomes and mesosomes.

Kinetics of bubble nucleation in a rhyolitic melt: an experimental study of the effect of ascent rate

In order to characterize the effect of ascent rate on the kinetics of bubble nucleation in a rhyolitic magma, we performed three series of experiments decompressed at rates of either 1000, 167, or 27.8 kPa/s. The experiments were carried out in an externally heated pressure vessel at 800°C and in the pressure range 260–59 MPa; the starting material was a crystal-free and bubble-free rhyolitic glass containing 7.0 wt% dissolved H2O. In all the decompression experiments, homogeneous bubble nucleation began at 90±2 MPa, that is, ≈150 MPa below the water saturation pressure of the silicate liquid, 240 MPa. The degree of supersaturation ΔPHoN required to trigger homogeneous bubble nucleation was almost independent of decompression rate (ΔPHoN is the difference between the saturation pressure and the nucleation pressure): nucleation pressure decreased by ≤3 MPa for a 36-fold increase in decompression rate. These results are in good agreement with the classical theory of nucleation assuming a rhyolite–H2O surface tension of 0.106 N m−1. Our major experimental finding is that, after a short nucleation event, the nucleation rate dropped and the bubble number density N reached a stationary value that was strongly sensitive to decompression rate: 6.8 mm−3 at 27.8 kPa/s, 470 mm−3 at 167 kPa/s, and 5800 mm−3 at 1000 kPa/s. The smaller value of N at low decompression rate was compensated by a larger mean bubble size, so that, at a given pressure, vesicularity was almost independent of decompression rate. The experimental values of N can be reproduced within a factor 0.3–1.4 using a relationship derived from published numerical simulations of vesiculation in ascending magmas. The nucleation behavior in our experiments is dictated by a competition between bubble nucleation and diffusive bubble growth, which depletes in water the surrounding liquid and therefore reduces the degree of volatile supersaturation. Once a critical value of N is attained, diffusive bubble growth can keep pace with decompression and prevent the degree of volatile supersaturation in the liquid to increase with decreasing pressure. The strong correlation between bubble number density and decompression rate has fundamental volcanological implications. If we extrapolate the experimental data to the typical ascent rates of silicic magmas, we obtain bubble number densities (10−3–101 mm−3 for homogeneous nucleation; ≈10−1–103 mm−3 for heterogeneous nucleation) that are orders of magnitude smaller than those measured in most natural pumices. We therefore propose that the large values of N in silicic pumices may be due to two successive nucleation events: (1) a first event, which occurs relatively deep in the volcanic conduit and which yields a moderate number of bubbles; and (2) a second nucleation event, yielding a very large number of small bubbles, and presumably related to the dramatic increase of decompression rate that precedes fragmentation. The small bubble number densities associated with homogeneous nucleation suggest that a strong departure from equilibrium degassing should be the rule even at slow ascent rates.

Wetting behavior of partial melts during crustal anatexis: the distribution of hydrous silicic melts in polycrystalline aggregates of quartz

The equilibrium distribution of hydrous silicic melts in polycrystalline aggregates of quartz was characterized in a series of partial melting and melt distribution experiments in the systems quartz-albite-orthoclase-H2O and quartz-anorthite-H2O, at 650 to 1000 MPa and 800 to 900° C. Near-equilibrium textures in these experiments are characterized by very low quartz-quartz-melt wetting angles, and by a substantial number of thin melt films along grain boundaries. Wetting angles in the H2O-saturated experiments are as follows: 18° at 800° C-1000 MPa, and 12° at 900° C-1000 MPa in the granitic system; 18° at 850° C-650 MPa, 15° at 900° C-650 MPa, and 15° at 900° C-1000 MPa in the quartzanorthite system. In the granitic system at 900° C-1000 MPa, a decrease of H2O content in melt from ∼17 wt% (at saturation) to ∼6 wt%, results in a slight increase of wetting angle from 12° to 16°. These low wetting angles — and the observation that many grain boundaries are wetted by melt films-indicate that the ratio of quartz-quartz to quartz-melt interfacial energies (γss/γs1) is high: ∼2. Secondary electron imaging of fracture surfaces of melt-poor samples provided a three-dimensional insight into the geometry of melt; at low melt fraction, melt forms an interconnected network of channels along grain edges, as predicted for isotropic systems with wetting angles below 60°. This high-permeability geometry suggests that the segregation of granitic melts is not as sluggish as previously anticipated; simple compaction calculations for a permeability range of 10-12 to 10-9 m2 indicate that segregation may operate at low to moderate melt fractions (below 30 vol. %), within relatively short time-scales, i.e., ≤105 to 106 years. Quartzmelt textures show significant deviations from the equilibrium geometries predicted for isotropic partially molten systems. The most consistent deviation is the pervasive development of crystallographically-controlled, planar faces of quartz; these faces provide definitive evidence for non-isotropic quartz-melt surface energy. For most silicates other than quartz, the grain-scale distribution of partial melts deviates even more significantly from equilibrium distributions in isotropic systems; accordingly, in order to describe adequately melt distributions in most natural source regions, the equilibrium model should be modified to account for anisotropy of solid-liquid interfacial energy.

Physics and chemistry of partially molten rocks

Preface N. Bagdassarov, et al. List of Contributors. 1. Rheology of Partially Molten Rocks D.L. Kohlstedt, et al. 2. Anelastic and Viscoelastic Behaviour of Partially Molten Rocks and Lavas N. Bagdassarov. 3. Constraints on the Melt Distribution in Anisotropic Polycrystalline Aggregates Undergoing Grain Growth U.H. Faul. 4. The Grain-Scale Distribution of Silicate, Carbonate and Metallosulfide Partial Melts: A Review of Theory and Experiments D. Laporte, A. Provost. 5. Partial Melting and Melt Segregation in a Convecting Mantle H. Schmeling. 6. A Fractionation Model for Hydrous Calcalkaline Plutons and the Heat Budget During Fractional Crystallisation and Assimilation L. Matile, et al. 7. Migmatitic Gabbros from a Shallow-Level Metamorphic Contact Aureole, Fuerteventura Basal Complex, Canary Islands. Role of Deformation in Melt Segregation A. Hobson, et al. 8. Thin Amorphous Intergranular Layers at Mineral Interfaces in Xenoliths: the Early Stage of Melting R. Wirth, L. Franz. Index.

Experimental study of homogeneous bubble nucleation in rhyolitic magmas

We present a comprehensive experimental study of homogeneous bubble nucleation in rhyolitic magmas. It was undertaken by performing a series of isothermal (800°C) rapid decompression experiments using crystal-free rhyolitic liquids containing from 4 to 7 wt% H2O and various CO2 contents of up to 820 ppm. No homogeneous nucleation was observed in samples containing 4.5 wt% H2O and 150–200 MPa), and that the pressure of homogeneous nucleation will be strongly sensitive to the multicomponent volatile composition of the liquid.

Direct Observation of Near-Equilibrium Pore Geometry in Synthetic Quartzites at 600°-800°C and 2-10.5 Kbar

A primary factor affecting geochemical transport and bulk-rock physical properties in the mid- and lower continental crust is the grain-scale distribution of fluid-filled porosity. Of particular significance is the continuity of the fluid phase in the direction of potential transport. SEM images reveal that

Significance of low symmetry fabrics in magmatic rocks

H_{2}O-CO_{2}-NaCl