Shock Melting Research Articles

Two populations of carbonate in ALH84001: geochemical evidence for discrimination and genesis

We present major and trace-element, oxygen isotope, textural, and structural data for carbonates and related phases in the SNC meteorite ALH84001. These data document the existence of at least two distinct carbonate populations: one composed of finely zoned, chemically and isotopically heterogeneous concretions of magnesio-siderite with distinct white magnesite rims, and a second composed of relatively homogeneous, isotopically and compositionally simple domains of ankeritic carbonate and intimately intergrown glass and fine-grained pyroxene. We suggest on the basis of textural evidence and geochemical systematics that the first population consists of low-temperature aqueous precipitates, and the second is produced by shock melting of the first. Values of δ18O and Sr/Ca ratios are correlated with one another in magnesio-siderite concretions; the trend formed by these data is consistent with the predicted relationship for inorganic precipitation of carbonate from a solution of constant composition between temperatures of ∼190°C (for concretion cores) to 20°C (for magnesite-rich concretion rims). Given the assumptions inherent in this temperature estimate, the aqueous fluid parental to carbonate concretions is constrained to have a δ18O of −5‰ VSMOW (significantly mass fractionated compared with expected juvenile martian volatiles) and minor-element abundances broadly similar to terrestrial seawater.

Shock-related mineralogical features and P-T history of the Suizhou L6 chondrite

The Suizhou meteorite, classified as an L6 chondrite, contains weakly shocked olivine and pyroxene, but almost all the plagioclase in the meteorite was melted and transformed into maskelynite during shock metamorphism. Chromite was heavily fragmented and granulated, and many tiny chromite fragments were incorporated into the molten plagioclase as inclusions. Metal and troilite show no obvious intragranular textures, but many tiny rounded FeNi metal grains were deposited in the intersecting joints of planar fractures in olivine and pyroxene. A few very thin shock melt veins occur in the Suizhou meteorite, which contain abundant high-pressure phases, including coarse-grained ringwoodite, majorite, (Na,Ca)AlSi3O8-hollandite and fine-grained liquidus majorite-pyrope garnet. The shock features of this meteorite match shock stages 3 to 5, while the presence of ringwoodite in Suizhou veins is considered to appear at stage 6. It is estimated that the Suizhou meteorite experienced a shock pressure and shock temperature of up to 22 GPa and 1000°C, respectively. The shearing friction along veins raised the temperature within the veins. Shock-induced pressure and temperature in the shock veins attained 22 GPa and 1900°C. Therefore, the actual shock level of the Suizhou meteorite could correspond to stage 3–4. A longer duration of the shock pressure and temperature regime in the Suizhou meteorite plays an important role in the pervasive melting of plagioclase in the unmelted part of the meteorite, as well as in the formation of abundant high-pressure phases in the very thin shock-melt veins. It appears that maskelynite cannot be used as the sole criteria for evaluating the shock stage of shock-metamorphosed chondrites.

Sound Velocities in Porous Iron Shocked to 177 GPa and theImplications for Shock Melting

Sound velocities in shock-loaded solids are not only important to determinebulk moduli of solids at high pressures, but are also crucial to inform theshock melting of solids upon loading. In this letter, we first report on shockmelting of porous solids at high pressures by measuring sound velocities inthe porous iron of average density 6.90 g/cm3 in the pressure range of110-180 GPa. The measured sound velocity softens at pressures from 122 to156 Gpa, which may be attributed to shock melting of the porous iron.

Microdistributions and petrogenetic implications of rare earth elements in polymict ureilites

Abstract— Polymict ureilites contain various mineral and lithic clasts not observed in monomict ureilites, including plagioclase, enstatite, feldspathic melt clasts and dark inclusions. This paper investigates the microdistributions and petrogenetic implications of rare earth elements (REEs) in three polymict ureilites (Elephant Moraine (EET) 83309, EET 87720 and North Haig), focusing particularly on the mineral and lithic clasts not found in monomict ureilites.As in monomict ureilites, olivine and pyroxene are the major heavy (H)REE carriers in polymict ureilites. They have light (L)REE‐depleted patterns with little variation in REE abundances, despite large differences in major element compositions.The textural and REE characteristics of feldspathic melt clasts in the three polymict ureilites indicate that they are most likely shocked melt that sampled the basaltic components associated with ureilites on their parent body. Simple REE modeling shows that the most common melt clasts in polymict ureilites can be produced by 20–30% partial melting of chondritic material, leaving behind a ureilitic residue. The plagioclase clasts, as well as some of the high‐Ca pyroxene grains, probably represent plagioclase‐pyroxene rock types on the ureilite parent body. However, the variety of REE patterns in both plagioclase and melt clasts cannot be the result of a single igneous differentiation event. Multiple processes, probably including shock melting and different sources, are required to account for all the REE characteristics observed in lithic and mineral clasts.The C‐rich matrix in polymict ureilites is LREE‐enriched, like that in monomict ureilites. The occurrence of Ce anomalies in C‐rich matrix, dark inclusions and the presence of the hydration product, iddingsite, imply significant terrestrial weathering.A search for 26Mg excesses, from the radioactive decay of 26Al, in the polymict ureilite EET 83309 was negative.

NaAlSi3O8-hollandite and other high-pressure minerals in the shock melt veins of the Suizhou meteorite

The Suizhou L6 chondrite contains a few very thin shock melt veins of 0.02–0.09 mm in width. In spite of small width of the veins, shock-induced high-pressure phases, such as coarse-grained NaAlSi3O8-hollandite, ringwoodite, majorite and fine-grained matrix majorite-pyropess have been discovered in these veins. NaAlSi3O8-hollandite, the high-pressure phase of plagioclase, in Suizhou shock veins occurs as a single phase mineral, no silicate glassy phase, such as albitic glass, was incorporated with it. The presence of above-mentioned high-pressure phases constrains the high pressure (up to 23–24 GPa) and high temperature (up to 1900–2000°C) regime in Suizhou shock veins, and indicates that the duration of high-pressure regime in the veins should be long enough (a few seconds) for phase transformation and crystallization of minerals under pressure. The discovery of the first natural single-phase crystalline NaAlSi3O8-hollandite in Suizhou meteorite is of important significance in understanding the Earth’s mantle geochemistry.

A comparative study of naturally and experimentally shocked chondrites

Samples of the Jilin H5 chondrite were experimentally shock-loaded at the peak pressures of 12, 27, 39, 53, 78, 83, 93, and 133 GPa. The aim of this study is to compare experimentally shock-induced phenomena with those in naturally shocked chondrites and to test the feasibility of experimentally calibrating naturally induced shock phenomena in H- and L-chondrites. Planar fractures, mosaicism, brecciation in olivine and pyroxene, as well as transformation of plagioclase into diaplectic glass were observed in the Jilin samples shocked at pressures lower than 53 GPa. Shock-induced chondritic melts were first obtained at P>78 GPa and more than 60% of the whole-rock melting was achieved at P∼133 GPa, and that shock-induced silicate melt consists of quenched microcrystalline olivine and pyroxene, metal, troilite and vesicular glass. No high-pressure phases were observed in any of the experimentally shocked samples, neither in the deformed nor in the molten regions. Deformation features in Jilin samples shock-loaded below 53 GPa are comparable to those found in H- and L-chondrites. The mineral assemblages in the molten regions in the shocked Jilin samples are also comparable to those encountered in the heavily shocked Yanzhuang (H6) and some Antarctic H-chondrites, but differ considerably from those found in heavily shocked Sixiangkou and many other L6 chondrites. Shock melt veins in L6 chondrites contain high-pressure polymorphs of olivine, pyroxene, plagioclase and high-pressure liquidus phases, whereas shock melt veins in heavily shocked H-chondrites contain mainly low-pressure mineral assemblages. The differences in the mineral constituents of shock melt veins in L- and H-chondrites clearly indicate differences in the shock histories of these meteorites. While crystallization in the shock melt veins in L-chondrites took place at high pressures, crystallization in shock-induced melt in most H-chondrites took place after decompression. It is evident that the thickness and abundance of shock melt veins and size of melt regions is not necessarily a quantitative measure of the degree of shock. The duration of the high-pressure regime, the time of the cooling and the P–T regime during the crystallization path, and the post-shock temperatures are stringent parameters that control the evolution of the shock-induced melt. So, scaling from shock experiments on millimeter-sized samples to natural shock features on kilometer-sized asteroids poses considerable problems in quantifying the P–T conditions during natural shock events on asteroids.

Dhofar 081: A new lunar highland meteorite

Abstract— The lunar meteorite Dhofar 081, found as a single fragment of 174 g in the Dhofar region of Oman, is a shocked feldspathic fragmental highland breccia dominated by anorthosite‐rich lithic and mineral clasts embedded into a fine‐grained mostly shock melted clastic matrix. Major mineral phases in the bulk rock are Ca‐rich plagioclase (An96.5–99.5), pyroxene (FS21.9–46.2Wo3.0–41.4), and olivine (Fa29.3–47.8); accessory phases include Fe‐Ni metal, ilmenite, and Ti‐Cr‐rich spinel. Dhofar 081 contains subordinate crystalline fragments of large anorthosites, intersertal impact‐melt rocks, microporphyritic impact‐melt breccias, dark fine‐grained impact‐melt breccias, large cataclastic feldspars, and irregularly shaped brown glass clasts. Mafic components are rare and no genuine regolith components were found in the sections studied. Minerals in Dhofar 081 show homogeneously distributed shock features: intergranular recrystallization, strong fracturing and mosaicism in feldspar as well as a high density of mostly irregular fractures in pyroxene and olivine. Localized impact melting caused by one or several impacts led to a strong lithification. Based on these effects an equilibration shock pressure of about 15–20 GPa is estimated for the strongest shock event in Dhofar 081. Devitrification of the “glassy” material in the rock indicates thermal annealing after shock melting suggesting that the 15–20 GPa shock event predated the ejection event. According to the concentrations of implanted solar noble gases Dhofar 081 represents a polymict clastic breccia deposit with possibly a minor regolith component. A similar noble gas record of Dhofar 081 and MacAlpine Hills 88104/05 suggests the possibility of a source crater pairing of both meteorites. As indicated by noble gas measurements pairing of Dhofar 081 with the other lunar meteorites found in Oman, Dhofar 025 and Dhofar 026, is unlikely.

Characteristics in naturally and experimentally shocked chondrites: A clue to P-T conditions of impacted asteroids

The aim of this study is to compare the experimentally shock-induced features with those in naturally shocked chondrites and to test the feasibility of experimentally calibrating naturally induced features in shocked H- and L-chondrites. Samples of the Jilin chondrite (H5) were experimentally shock-loaded at the following peak pressures: 12, 27, 39, 53, 78, 83, 93 and 133 GPa respectively. Chondritic melts were first obtained at P ＞78 GPa and more than 60% melting was achieved at P ～133 GPa. No high-pressure phases were observed in any of the shocked samples, neither in the deformed nor in the molten regions. Textural relations and mineral assemblages of the shocked samples are comparable to those encountered in the heavily shocked H-chondrite Yanzhuang but differ considerably from those found in heavily shocked L6 chondrites. Shock melt veins in L6 chondrites contain high-pressure polymorphs of olivine and pyroxene and high pressure liquidus phases. Scaling from shock experiments on millimeter-sized samples to natural shock features on kilometer-sized asteroids poses considerable problems in quantifying the P-T conditions during natural shock events on asteroids.

Structure, phase composition, and properties of an Al-Hf alloy subjected to isentropic spherical shock waves

The dimensions of characteristic zones of transformations related to the high-strain-rate deformation of the alloy in the solid state and its melting in stress waves, as well as partial evaporation of the shocked melt during unloading into the central cavity that is formed behind the front of a spherically divergent shock wave have been determined. The chosen regime of the explosive loading of the sphere led to melting at the isentrope and at the adiabat in shock waves in various zones along the radius, and to the subsequent formation of two zones with cast structures differing in the grain size, degree of supersaturation of the α-solid solution, and the size and shape of the precipitated aluminide particles. In the zone that underwent melting and high-rate solidification, metastable aluminides Al 3 Hf were revealed inside the adiabatic shear bands. The main features of the evolution of the structures of the matrix and the aluminides in the field of the solid-state transformation were revealed. As a result of high-rate loading and subsequent unloading, the brittle phase was shown to undergo fragmentation and fracture, while the aluminum matrix acquires the structure of a hot-worked material.

Microtektites on Mars: Volume and Texture of Distal Impact Ejecta Deposits

Microtektites, small blobs of ejecta formed in the shock melt and vapor plume of an impact, can be dispersed far from the source crater only if the impact is violent enough for the ejecta plume to pierce the atmosphere; they are therefore formed in far smaller (and more numerous) impact events on Mars than on Venus and Earth, which have thicker atmospheres. Microtektite abundances from the Chicxulub and Bosumtwi craters on Earth suggest that the volume of this material is ∼5 × 10 −5 D c 3.74 km 3, with D c the crater diameter in kilometers, similar to the observed volumes of the dark parabolic ejecta deposits on Venus. Corresponding volumes on Mars are ∼2.5 × smaller, but even so this result implies that even only a 15-km crater can produce a layer of microtektites with a global average thickness on Mars of 40 microtektites per square centimeter. I use a trajectory code and a thermal model to show that these particles are easily dispersed globally on Mars and that micrometeoroids of the same size will be unmelted by reentry heating. The uniform size and glassy texture of microtektites may allow such ejecta layers to be identified by the remote arm cameras on Mars landers, particularly in the polar layered terrain where they may be preserved against abrasion.

Formation of feldspathic and metallic melts by shock in enstatite chondrite Reckling Peak A80259

Abstract— The enstatite chondrite reckling peak (rkp) a80259 contains feldspathic glass, kamacite, troilite, and unusual sets of parallel fine‐grained enstatite prisms that formed by rapid cooling of shock melts. Metallic Fe,Ni and troilite occur as spherical inclusions in feldspathic glass, reflecting the immiscible Fe‐Ni‐S and feldspathic melts generated during the impact. The Fe‐Ni‐S and feldspathic liquids were injected into fractures in coarse‐grained enstatite and cooled rapidly, resulting in thin (≤ 10 μm) semicontinuous to discontinuous veins and inclusion trails in host enstatite. Whole‐rock melt veins characteristic of heavily shocked ordinary chondrites are conspicuously absent. Raman spectroscopy shows that the feldspathic material is a glass. Elevated MgO and SiO2 contents of the glass indicate that some enstatite and silica were incorporated in the feldspathic melt. Metallic Fe,Ni globules are enclosed by sulfide and exhibit Nienrichment along their margins characteristic of rapid crystallization from a Fe‐Ni‐S liquid. Metal enclosed by sulfide is higher in Si and P than metal in feldspathic glass and enstatite, possibly indicating lower O fugacities in metal/sulfide than in silicate domains. Fine‐grained, elongate enstatite prisms in troilite or feldspathic glass crystallized from local pyroxene melts that formed along precursor grain boundaries, but most of the enstatite in the target rock remained solid during the impact and occurs as deformed, coarsegrained crystals with lower CaO, Al2O3, and FeO than the fine‐grained enstatite. Reckling Peak A80259 represents an intermediate stage of shock melting between unmelted E chondrites and whole‐rock shock melts and melt breccias documented by previous workers. The shock petrogenesis of RKPA80259 reflects the extensive impact processing of the enstatite chondrite parent bodies relative to those of other chondrite types.

Hydrocode modeling of oblique impacts: The fate of the projectile

Abstract— All impacts are oblique to some degree. Only rarely do projectiles strike a planetary surface (near) vertically. The effects of an oblique impact event on the target are well known, producing craters that appear circular even for low impact angles (>15° with respect to the surface). However, we still have much to learn about the fate of the projectile, especially in oblique impact events. This work investigates the effect of angle of impact on the projectile.Sandia National Laboratories' three‐dimensional hydrocode CTH was used for a series of high‐resolution simulations (50 cells per projectile radius) with varying angle of impact. Simulations were carried out for impacts at 90, 60, 45, 30, and 15° from the horizontal, while keeping projectile size (5 km in radius), type (dunite), and impact velocity (20 km/s) constant.The three‐dimensional hydrocode simulations presented here show that in oblique impacts the distribution of shock pressure inside the projectile (and in the target as well) is highly complex, possessing only bilateral symmetry, even for a spherical projectile. Available experimental data suggest that only the vertical component of the impact velocity plays a role in an impact. If this were correct, simple theoretical considerations indicate that shock pressure, temperature, and energy would depend on sin2θ, where θ is the angle of impact (measured from the horizontal). However, our numerical simulations show that the mean shock pressure in the projectile is better fit by a sin θ dependence, whereas shock temperature and energy depend on sin3/2 θ. This demonstrates that in impact events the shock wave is the result of complex processes that cannot be described by simple empirical rules. The mass of shock melt or vapor in the projectile decreases drastically for low impact angles as a result of the weakening of the shock for decreasing impact angles. In particular, for asteroidal impacts the amount of projectile vaporized is always limited to a small fraction of the projectile mass. In cometary impacts, however, most of the projectile is vaporized even at low impact angles.In the oblique impact simulations a large fraction of the projectile material retains a net downrange motion. In agreement with experimental work, the simulations show that for low impact angles (30 and 15°), a downrange focusing of projectile material occurs, and a significant amount of it travels at velocities larger than the escape velocity of Earth.

Light rare earth element enrichments in ureilites: A detailed ion microprobe study

Abstract— This paper explores the possible origin of the light rare earth element (LREE) enrichments observed in some ureilites, a question that has both petrogenetic and chronologic implications for this group of achondritic meteorites. Rare earth element and other selected elemental abundances were measured in situ in 14 thin sections representing 11 different ureilites. The spatial microdistributions of REEs in C‐rich matrix areas of the three ureilites with the most striking V‐shaped whole‐rock REE patterns (Kenna, Goalpara, and Novo Urei) were investigated using the ion imaging capability of the ion microprobe.All olivines and clinopyroxenes measured have LREE‐depleted patterns with little variation in REE abundances, despite large differences in their major element compositions from ureilite to ureilite. Furthermore, we searched for but did not find any minor mineral phases that carry LREEs. The only exception is one Ti‐rich area (∼20μm) in Lewis Cliff (LEW) 85400 with a major element composition similar to that of titanite; REE abundances in this area are high, ranging from La ≅ 400 × CI to Lu ≅ 40 × CI. In contrast, all ion microprobe analyses of C‐rich matrix in Kenna, Goalpara, and Novo Urei revealed large LREE enrichments. In addition, C‐rich matrix areas in the three polymict ureilites, Elephant Moraine (EET) 83309, EET 87720, and North Haig, which have less pronounced V‐shaped whole‐rock REE patterns, show smaller but distinct LREE‐enrichments. The C‐rich matrix in Antarctic ureilites tends to have much lower LREE concentrations than the matrix in non‐Antarctic ureilites. There is no obvious association of the LREEs with other major or minor elements in the C‐rich areas. Ion images further show that the LREE enrichments are homogeneously distributed on a microscale in most C‐rich matrix areas of Kenna, Goalpara, and Novo Urei. These observations suggest that the LREEs in ureilites most probably are absorbed on the surface of fine‐grained amorphous graphite in the C‐rich matrix. It is unlikely that the LREE enrichments are due to shock melts or are the products of metasomatism on the ureilite parent body. We favor LREE introduction by terrestrial contamination.

Martian soil component in impact glasses in a Martian meteorite.

Chemical compositions of impact melt glass veins, called Lithology C (Lith C) in Martian meteorite EET79001 were determined by electron microprobe analysis. A large enrichment of S, and significant enrichments of Al, Ca, and Na were observed in Lith C glass compared to Lithology A (Lith A). The S enrichment is due to mixing of plagioclase- enriched Lith A material with Martian soil, either prior to or during impact on Mars. A mixture of 87% Lith A, 7% plagioclase, and 6% Martian soil reproduces the average elemental abundances observed in Lith C. Shock melting of such a mixture of plagioclase-enriched, fine-grained Lith A host rock and Martian soil could yield large excesses of S (observed in this study) and Martian atmospheric noble gases (found by Bogard et al., 1983) in Lith C. These mixing proportions can be used to constrain the elemental abundance of phosphorus in Martian soil.

Experimental shock metamorphism of the Murchison CM carbonaceous chondrite

A series of shock-recovery experiments were carried out on the Murchison CM carbonaceous chondrite by using a single-stage propellant gun. The Murchison samples were shocked in nine experiments at peak pressures from 4 to 49 GPa. The recovered samples were studied in detail by using an optical microscope, a scanning electron microscope and an electron-probe microanalyzer. Chondrules are flattened in the plane of the shock front at 4 to 30 GPa. The mean aspect ratio of chondrules increases from 1.17 to 1.57 roughly in proportion to the intensity of shock pressure up to ∼25 GPa. At 25 to 30 GPa, the mean aspect ratio does not increase further, and chondrules show increasingly more random orientations and degrade their preferred orientations, and at ∼35 GPa, they are extensively disrupted. Most coarse grains of olivine and pyroxene are irregularly fractured, fracture density increases with increasing shock pressure and at ∼30 GPa almost all are thoroughly fractured with subgrains of <1 to 5 μm in size. At ∼20 GPa, subparallel fractures begin to form in the matrix in directions roughly perpendicular to the compression axis and their densities increase with pressure, especially dramatically at 25 to 30 GPa; thus, the sample is increasingly comminuted and becomes fragile. Local shock melting occurs as melt veins and pockets at 20 to 30 GPa. Fracture-filling veins of fine grains of matrix are also produced at 25 to 30 GPa. The melts and the fine grains seem to result mainly from frictional heating due to displacement along fractures. At ∼35 GPa, melting occurs pervasively throughout the matrix. The melts are mainly produced from the matrix; however, they are consistently more enriched in Fe, S, and Ca, which indicates that these elements are selectively incorporated into the melts. The melts contain tiny spherules of Fe-Ni metal, Fe sulfide, and numerous vesicles. At 49 GPa, the matrix is totally melted and coarse grains of olivine are partially melted. The melts contain much larger vesicles (50–300 μm in diameter) than those in the samples shocked at lower pressures, which indicates that much more intense devolatilization and gas expansion took place. For the purpose of comparing shock thermal effects between the experimentally shocked samples and naturally shocked targets (surface materials in the Murchison parent body), we calculated internal energy increase for compression by multiple shock wave reflections (experimental case) and for compression by a single shock wave (natural case). The results suggest that postshock thermal effects observed at each experiment may be attained by impact on the natural targets at a considerably lower shock pressure than the peak shock pressure. From the results of our experiments and calculations, we conclude that if the Murchison parent body were shocked on the surface at pressures higher than ∼25 GPa, shocked material would probably undergo drastic increase in the degree of comminution and simultaneous generation of strong expansive forces on pressure release. Thus the results support the hypothesis of Scott et al. (1992) that volatile-rich carbonaceous chondrites shocked above 20 to 30 GPa escaped from the parent body and formed particles that are too small to survive as meteorites.

Shock melting of the canyon diablo impactor: constraints from nickel-59 contents and numerical modeling

Two main types of material survive from the Canyon Diablo impactor, which produced Meteor Crater in Arizona: iron meteorites, which did not melt during the impact; and spheroids, which did. Ultrasensitive measurements using accelerator mass spectrometry show that the meteorites contain about seven times as much nickel-59 as the spheroids. Lower average nickel-59 contents in the spheroids indicate that they typically came from 0.5 to 1 meter deeper in the impactor than did the meteorites. Numerical modeling for an impact velocity of 20 kilometers per second shows that a shell 1.5 to 2 meters thick, corresponding to 16 percent of the projectile volume, remained solid on the rear surface; that most of the projectile melted; and that little, if any, vaporized.

Molecular-dynamicssimulations for the shock Hugoniot meltings of Cu, Pd and Pt

We perform molecular-dynamics simulations to study the shock melting of transition metals such as Cu, Pd and Pt on the basis of an embedded-atom method. Using the coupling-constant integration method, the Gibbs free energies of the crystalline and liquid phases are calculated as a function of pressure and temperature, so that the melting curves are obtained, up to 3-4 Mbar. We find the melting properties near zero pressure to be in good agreement with experiments. For each metal, we compare the melting curve with the Hugoniot equations of state for the solid and liquid phases and determine the melting region of the shock Hugoniot. The Hugoniot melting of Cu is found to begin at 1.9 Mbar and to end at 2.25 Mbar, in good agreement with experimental data. During the shock compression, we find that the pressure region where the Hugoniot melting occurs increases almost linearly with increase of the ionic mass.

Exogenic and endogenic breccias: a discussion of major problematics

Pseudotachylitic breccias, impact melt rock, (ultra)cataclasites, and even ultramylonites may macroscopically be of very similar appearance. Even detailed optical microscopic work may not suffice to identify the actual type of melt rock studied. The difficulties to distinguish different tectonically produced breccias (fault rocks) and impact breccias (for which no uniformly accepted nomenclature exists) are examined, and the various formational processes (friction melting or shock melting), at the macro- and micro-scales, are discussed. Special emphasis is placed on the occurrence of different and multiple breccia types within impact structures, where not only impact-induced friction melt and impact melt may occur, but where tectonically produced fault rocks that are not related to the impact event may further complicate the breccia situation. Melting on shatter cone surfaces and its likely formation, the proposed origin as distal impact breccia of certain diamictites, and problems with carbonate breccias and caused by metamorphic overprint on breccias are discussed. It is cautioned not too easily to reach generalized conclusions, for example with regard to the usage of the A- and B-type pseudotachylite terminology. Areas in need of further breccia studies – if possible in close collaboration between impact and tectonic researchers – are emphasized.

New Experimental Constraints on the Nature of D″

Recent findings of 'Ultra-Low Velocity Zones Near the Core-Mantle Boundary' (1) have been interpreted as being due to partial melting. This interpretation is in agreement with our latest measurements of the solidus temperatures of a realistic multi-component system in the laser-heated diamond cell at lower mantle conditions. The previous measurements of the melting temperatures of the possible end-member component of the lower mantle Mg-Si-perovskite, magnesiowiistite, and Ca-Si-perovskite led to melting temperatures that were much higher than the core temperature, which is estimated from the melting measurements of iron and its oxygen and sulphur components to be around 4000 K at the core-mantle boundary. It was previously assumed, that Mg-Siperovskite, the major component in the lower mantle is near the eutectic composition. However, the melting behaviour of the lower mantle components is much more complicated than previously assumed: the perovskites having very steep melting curves, and magnesio~a4istite a shallow melting curve. Thus, the eutectic composition is likely to change significantly in the pressure range of the lower mantle (24-134 GPa). For the measurement of the solidus at high pressure a new method was developed. To verify the onset of melting we monitored the surface topography of the recovered samples using atomic force and scanning electron microscopy. Data were taken to about 60 GPa and a smooth extrapolation yields a solidus temperature at the core-mantle boundary of slightly above 4000 K. This is in very good agreement with a recent shock melting measurement (2) on olivine near 130 GPa. The data not only allow melting of mantle material in the near vicinity of the core but have further implications on the chemical composition: Our melting data imply that the composition of the partial melt is most likely magnesiowfistite-rich. This material is ahnost 2% more dense than the major component Mg-Siperovskite. This density contrast and partial melting may cause chemical segregation at the bottom of the Max-Planck Institut •r Chemie, 55020 Mainz, Germany

Melting curve of aluminum in a diamond cell to 0.8 Mbar: implications for iron

The melting curve of aluminum was measured in a laser-heated diamond cell up to a pressure of 0.8 Mbar in order to test the agreement between this technique and shock wave measurements, which has been lacking in the case for iron. At this pressure, which is over an order of magnitude higher than in previous experiments [1, 2], the melting temperature is 3800 K, comparable to that measured for iron at 2 Mbar [3]. The present results for aluminum extrapolate smoothly to the previous melting measurements in a multi-anvil apparatus to 60 kbar and to the calculated shock melting point of 4750 K at 1.25 Mbar. They are also in excellent agreement with theoretical calculations. A review of the shock data reported for Al, Ta and Mo, close-packed metals, in which a break in the sound velocity-pressure curve is used to determine the melting pressure, shows that the change in velocity at melting is about 10% for all three metals. In the case of iron, the sound velocity data have been used to infer two transitions: a solid-solid transition at 2.0 Mbar and melting at 2.4 Mbar, each of these transitions having about a 5% change in sound velocity. It is unlikely that a phase transition between close-packed cubic structures will have a 5% velocity change, the same as is found in the melting transition. We therefore suggest that for iron there exists only a single transition, starting at 2.0 Mbar, a region of incomplete shock melting between 2 and 2.4 Mbar, and a total change in sound velocity of about 10%, which is closer to the value of the other metals studied. This interpretation introduces a very good agreement between the shock melting results of Brown and McQueen [4] and diamond cell measurements for iron [3] which has up to now been lacking.