Siderophile-rich particles in the melt rocks at the E. Clearwater impact structure,Quebec: Their characteristics and relationship to the impacting body

The composition of the sampled melt rocks at the 22 km diameter E. Clearwater impact structure indicates the presence of ～8% C-1 material. The meteoritic component is fractionated with refractory siderophiles, up to 30 times C-1 abundances, concentrated in ten to hundred micron-sized, magnetic particles. These particles consist of the Ni-sulphide, millerite, and what is assumed to be a mixture of refractory silicates and magnetite with grain sizes of <1 μm. The larger particles have a core-rim structure with millerite and occasionally very minor galena and possibly pentlandite in the core. An origin as a combination of altered meteoritic metal and condensed meteoritic silicate is favored for the origin of the siderophile-rich particles. If 8% meteoritic material is taken as the average meteoritic contamination in the melt, then the E. Clearwater projectile may have impacted with a velocity of 17 km s^?1. Peak shock pressures would have been of the order of 300 GPa, sufficient to vaporize the silicate component but only melt the metal component of the projectile. As the meteoritic material was being driven down into vaporized/ melted target rocks during the initial stages of impact, the melted Fe, Ni metal underwent oxidation, Fe was removed, and meteoritic silicate material recondensed on the cooler, essentially Ni metal. As cavity excavation proceeded, these Ni metal, silicate-oxide particles were incorporated in the melt, their refractory nature prevented thermal digestion and sulphur in the melt reacted with the metal to produce millerite on final equilibration. If this hypothesis is correct, it suggests that the E. Clearwater projectile was a C-2 or C-3 chondrite, both of which are compatible with the trace element composition of the melt rocks. Clearwater Lake is a twin impact structure formed by an asteroid pair. It is still not clear, however, what type of projectile formed the 32 km diameter western structure, where the surface melt rocks contain no identifiable meteoritic signature.     

FeNi metal in impact melt rocks of Lake Lappajärvi,Finland

Particles of FeNi metal ranging in size from a few mto 0.1 mm in diameter are a distinctive feature of the shock-generated melt rocks of Lake Lappajärvi. Irondeficient iron sulphide, pyrrhotite, is nearly always associated with the metal. Accessory phases are FeNiCo-sulphides, a phosphorus-bearing mineral (probably apatite), iron-rich aluminosilicates, and ilmenite. The content of Ni in the metal varies from less than 2% to more than 10%. There is evidence that the iron formed from molten globules which crystallized slowly and adjusted to low temperatures by solid state diffusion. The Ni concentration in the largest metal particles which appear to be least affected by alteration processes, and hence may represent the composition of the parent metal phase of the projectile, is, however, lower than in any reported meteoritic iron.     

Identification of the projectile at the Brent crater,and further considerations of projectile types at terrestrial craters

Impact melt samples from drill hole B1-59 at the 3.8 km diameter Brent crater (Ontario) have been analysed for siderophile trace elements indicative of meteoritic contamination. Samples from the basal melt zone at 823–857 m depth are enriched in Ir, Os, Pd, Ni, Co, Cr and Se over basement, with the abundance pattern suggesting a chondritic projectile for Brent. From a Ni-Cr correlation of 10 melt samples an L or LL chondrite is inferred. The contribution of an ultramafic country rock (alnoite) in the melt is too small to significantly influence its $\text{Ni}\text{Cr}$ ratio. Glass-rich breccias from the allochthonous breccias filling the crater also contain a meteoritic component. Interelement ratios (e.g. $\text{Ni}\text{Cr}$) are, however, fractionated relative to the melt zone samples. This, as well as the low Au content of all Brent samples, is probably a product of alteration.Additional data on impact melts from the 65 km diameter crater Manicouagan still did not reveal a meteoritic component, as also for the Mistastin crater (28 km diameter) where Cr analyses set an upper limit of 1% of an achondritic projectile component in the melt. Irghizites (tektite like glasses) from the Zhamanshin impact structure have been found to contain high Ni and Co concentrations, and our data show that Ir is also enriched. It is however not possible to define the projectile-type. Enrichment of an Ivory Coast tektite in Ir is confirmed. There are large differences in siderophile element concentrations among tektites, with otherwise similar chemical composition.There are now four known craters formed by chondrites (Clearwater East, Lapparjärvi, Wanapitei, and Brent), with Brent being the smallest of these. For smaller craters the projectiles appear to be limited to iron or stony-iron meteorites, because of atmospheric destruction of relatively small stony meteorites. It appears, however, that all major classes of meteorites are represented among the projectiles at terrestrial impact craters.     

Meteoritic material at five large impact craters

Sixteen crater samples were analyzed by radiochemical neutron activation analysis for Ge, Ir, Ni, Os, Pd and Re. Two impact melt rock samples from Clearwater East (22 km) showed strong, uniform enrichments in all elements except Ge, corresponding to 7.4% C1 chondrite material. Interelement ratios suggest that the meteorite was a C1 (or C2) chondrite, not an iron, stony iron, or chondrite of another type. An Ivory Coast tektite (related to the 10 km Bosumtwi crater) was enriched in Ir + Os and Ni to about 0.04 and 1.6% of C1 chondrite levels, but in the absence of data on country rocks, the meteorite cannot yet be characterized.Impact melt rock samples from Clearwater West (32km), Manicouagan (70km), and Mistastin (28 km) showed no detectable meteoritic component. Upper limits, as Cl chondrite equivalent, were Os ≤ 2 × 10^?3% (~0.01 ppb), Ni ≤ 2 × 10^?1% (~20ppm). Possible causes are high impact velocity and/or a chemically inconspicuous meteorite (achondrite, Ir,Os-poor iron or stony iron). However, a more likely reason is that some fraction of the impact melt remains meteorite-free, especially at craters with central peaks.Clearwater East is the first terrestrial impact crater found to be associated with a stony meteorite. Apparently the consistent absence of stony projectiles at small craters (< 1 km diameter) reflects their destruction in the atmosphere, as proposed by Öpik.     

The Lappajärvi meteorite crater,Finland: petrography,Rb-Sr,major and trace element geochemistry of the impact melt and basement rocks

Impact melt lithologies of the 77 m.y. old Finnish meteorite crater Lappajärvi as well as the Precambrian target rocks have been studied in detail, to identify and characterize different impact melt types (clast-poor, clast-rich, suevitic melt) and to study their chemical (major and trace elements) and isotopic (Rb-Sr) compositions in comparison to the composition of the target rocks.The Rb-Sr system of the whole melt body—including the suevitic melt—is shown to have been reequilibrated by the impact by extensive turbulent mixing of the various melted or vaporized target rocks. Chemical interactions (exchange of alkali elements, ⁸⁷Sr-redistribution) between feldspar clasts and impact melt surrounding them are the result of thermal metamorphism following the incorporation of target rock fragments of various degrees of shock metamorphism into the superheated melt. Exchange reactions between clasts and melt are determined by thermal activation, but the degree of shock metamorphism in the clasts plays an important role, too.Major and trace element distributions in impact melt and basement rocks indicate that the Lappajärvi melt body chemically is extremely homogeneous. Even volatile elements (such as Zn and Cu) were not strongly fractionated. Comparison of the abundances of siderophile elements in the impact melt (e.g., 118–177 ppm Cr, 195–340 ppm Ni, 6–12 ppb Ir) and calculated target rock mixture (79% mica schist, 11% granite-pegmatite, 10% amphibolite) (e.g., 85.6 ppm Cr, 54.8 ppm Ni, 0.5 ppb Ir) revealed the chondritic nature (C or H chondritic) of the meteoritic projectile. Less than 2% of the meteorite can be detected in the coherent melt, whereas the suevitic melt is uncontaminated by the projectile.     

Carbon-bearing phases in shock-induced melt zones of the Elga meteorite

The mineralogy and texture of shock-induced melt veinlets and melt pockets in silicate inclusions in the Elga IIE iron meteorite have been studied by reflected-light optical microscopy, EMPA, SEM, Raman spectroscopy and TEM. The results suggest that Elga experienced two discrete impact events. The earlier event involved the collision of a metallic projectile with a silicate target and resulted in partial melting and recrystallization of the silicate material, forming schreibersite and oxide rims between the metal and silicate. The later impact event resulted in melt pockets in the silicate inclusions and was associated with fragmentation, melting, and brecciation of the rims and displacement of some fragments into the melt pockets. These fragments are shown to contain carbon-bearing phases: siderite and amorphous sp ² carbon, which form carbon–oxide, siderite–oxide, and siderite–schreibersite associations. The fact that the carbon-bearing fragments are spatially constrained to shock breccia and melt zones indicates that these fragments are genetically related to the impact process and that their carbon-bearing phases are of cosmic origin.     

Investigation and simulation of metallic spherules from lunar soils

Metallic spherules selected from the Apollo 11, 12, 14, 15 and 16 sites were studied by optical techniques as well as the electron probe and scanning electron microscope. In addition, metallic spherules of similar composition were produced experimentally. The structure of the metallic lunar spherules indicates an origin by solidification of molten globules of metal. The experimentally produced spherules have external morphologies, metallographic structures and solidification rates (7 × 10² to 10⁶ ° C/sec) similar to the lunar spherules which have rapidly solidified. The majority of the lunar spherules are, however, either more slowly cooled or have been reheated in place with the lunar fragmental rocks, glass or soil. The heavy meteorite bombardment of the highlands is strongly reflected by the evidence of reheating and/or slow cooling of a majority of Apollo 14 and 16 spherules.The metallic spherules are probably produced from both lunar and meteoritic sources. Impact processes cause localized shock melting of metallic (and non-metallic) constituents at metal-sulfide phase interfaces in surface rocks and in the meteoritic projectile. The major source of metallic spherules is the metal phase present in the lunar rocks and soil. The large variation in spherule bulk compositions is attributed to the different meteoritic projectiles bombarding the Moon, metal phases of differing compositions in the lunar soils and rocks and to the experimental results which indicate that high S, high P alloys form two immiscible liquids when melted.     

On the track of the elusive Sudbury impact: geochemical evidence for a chondrite or comet bolide

Siderophile and lithophile trace element data for 69 samples from the Sudbury impact crater fill (Onaping Formation) and quartz diorite offset dikes help constrain the sources of the established moderately elevated platinum group element signature associated with the impact structure. The siderophile element distribution of the crater fill requires contributions from bulk continental crust, mafic rocks and a chondritic component. A mantle component is absent, but the involvement of mid to lower crust is implied. After considering post‐impact hydrothermal alteration, melt heterogeneity and mafic target admixture, the projectile elemental ratios were determined on a more robust data subset. Chondrite discrimination diagrams of these ratios identify an ordinary or enstatite chondrite as the most probable source of meteoritic material in the Sudbury crater fill. However, the relative and absolute siderophile element distributions within the impact structure as well as bolide size models are congruent with the bolide being a comet that had a chondritic refractory component.     

Identification of the projectile component in the impact structures Rochechouart, France and Sääksjärvi, Finland: Implications for the impactor population for the earth

A set of 11 impact melt rock samples from the Rochechouart impact structure, France and nine impact melt rock samples from Sääksjärvi impact structure, Finland were studied for their major and trace element compositions, including the abundances of the platinum group elements. The main goal of this study was to identify the projectile type(s) responsible for the formation of these two impact structures. The results confirmed previous studies that suggested extraterrestrial contamination in both the Rochechouart and Sääksjärvi impact melt rocks. The projectile types found for Rochechouart and Sääksjärvi are quite similar, and compatible with the composition of non-magmatic iron meteorites (IA and IIIC). This interpretation is based on: identical platinum group element patterns as well as peculiar Ni/Cr, Ni/Ir and Cr/Ir ratios, which can be explained by mixing of the different components of non-magmatic iron meteorites. This result indicates that, besides ordinary chondrites, non-magmatic iron may be among the most common material impacting the Earth, as they also represent the majority of the projectiles for craters smaller that 1.5 km. The abundance of non-magmatic irons as projectiles as well as their composition (olivine, pyroxene and iron) supports the assumption that a fraction of the S-type asteroids could by related to this type of material.     

Trace elements in natural metallic iron from Disko Island,Greenland

The largest occurrence of natural metallic iron on Earth is on the island of Disko, Greenland. Metallic iron is found there in a variety of different types, from small metal particles in basalts to large meter-sized blocks. We have studied three different types of metallic iron: small metal spherules (< 300 m) in basaltic magma; larger metal grains (300 m-3 mm), often composed of aggregates of smaller particles, in similar host rocks; and massive iron lumps (up to several tons). Analytical data for 13 siderophile elements in samples from these three types are presented. All metals analysed have a distinctly crustal pattern of siderophile elements. High Co/Ni, Re/Ir or W/Ir ratios clearly demonstrate that a meteoritic origin for the metallic iron must be excluded. Since the Co/Ni and Re/Ir ratios are approximately chondritic in the upper mantle of the Earth, a mantle origin for the Disko metals can also be ruled out. This supports earlier petrological and geological evidence that the metallic iron was formed through reduction of basaltic magma by carbon derived from Tertiary shales and coals. Significant differences in absolute and relative abundances of siderophile elements occur among the three kinds of metals. The strongly siderophile elements (e.g. Ir, Re, Ni) increase in concentration from the small metal spherules through the larger grains to the massive iron lumps. The contents of less strongly siderophile elements (P, W, Ga) decrease in the same sequence. Evidence is presented that the small metal spherules are formed by in situ reduction. Larger iron metal grains and massive iron lumps are composed of small spherules, accumulated by gravitational settling in a magma reservoir. These metal cumulates have extracted highly siderophile elements from a larger volume of basaltic melt.Part of a Ph.D. thesis by W. Klöck     

The composition and origin of metal in howardites

Howardites can be divided into two main groups, Ni-rich (>350ppm Ni) and Ni-poor (<150ppm Ni). In the Ni-rich group Ni occurs principally in metal grains associated with melt rocks and is largely derived from projectiles which caused the melting. The metal in Bununu, Kapoeta and Malvern melt rocks plots in the meteoritic Ni-Co range and in Bununu and Kapoeta is enriched in P. By contrast, most metal grains in primary lithic and crystal clasts in howardites are Ni-poor and plot mainly in the composition field of pristine lunar anorthosite metal. However, there are variations in the abundance and exact composition of primary metal from howardite to howardite and each therefore represents a discrete source region. The matrix metal in Bholgati, Bununu, and Kapoeta shows the diversity of compositions expected in a polymict breccia, with compositions plotting in and between the anorthositic and meteoritic Ni-Co fields. Other howardites show a more limited range of matrix metal compositions, because of limited metal-bearing clasts.Petersburg differs from other howardites in several ways. The metal in primary clasts has a unique $\text{Ni}\text{Co}$ ratio of about 40, which indicates derivation from a different reservoir from other howardite primary clasts. The metal in the matrix consists of large grains intergrown with silicates with compositions clustering tightly at 3.3% Ni, 0.2% Co. This is interpreted as equilibration, possibly as the result of deeper burial for Petersburg than for other howardites.     

Spherules from the Tunguska event site: Could they originate from the Tunguska Cosmic Body?

Soil layers at the Tunguska event site may have accumulated infalling extraterrestrial matter derived from the Tunguska Cosmic Body (TCB). Using mineralogical, textural, and chemical criteria, a set of metal and silicate spherules of probably cosmic origin was identified in the collection of spherules and rounded particles recovered from sites with high concentrations of magnetic spherules on a terrace above the floodplain of the Chunya River. The metal spherules consist of Ni(Cr)-bearing wüstite and magnetite with Ni-rich metal inclusions. The silicate spherules are glassy, cryptocrystalline, barred, and porphyritic melted micrometeorites, some of which contain metal droplets. The number of spherules counted in our samples is higher than the background level, indicating the possible presence of the TCB material.     

Processes of Replacement by Melt at Interaction between Refractory Materials and Industrially Produced Melts

In a number of industries (ferrous and nonferrous metallurgy, glass-making and silicate-producing technologies), interaction between refractory materials with melts results in sequences of reaction zonation (reaction columns) that show all principal features of diffusion-controlled metasomatic zoning. However, in contrast to the latter, reaction melt is generated together with crystalline phases in the rear zones of the columns. This melt is neither mechanically displaced melt that affects the refractory materials, nor produced by melting. The process generating this melt is most adequately defined as replacement by melt. The principal characteristics of the zoning are discussed below with reference to the corrosion of chromite–periclase refractory materials with melted slag in nickel-producing metallurgy. Similarities between the relations observed under different conditions and in different systems and the evolutionary dynamics of the process, specifics of melt generation and changes in its composition in the zones are demonstrated below with the use of data on other technologies and their experimental modeling. The mechanism of melt replacement is applicable to describing natural reaction processes of magma interaction with host rocks (magmatic replacement), with the following unobvious implications. (1) It is reasonable to expect that the minerals of the rocks should host melt inclusions. (2) It is reasonable to expect that certain minerals should be found in two distinct populations: (i) those in equilibrium with melt in the reaction column and (ii) those crystallizing from the cooling melt. (3) Two or more zones of the column can consist of the same minerals, but their proportions should be different. (4) Plastic deformations in the rear zones of the column (magmatic replacement) should be associated with brittle ones in the pristine host rocks and frontal (metasomatic) zones. (5) In contrast to the rocks of metasomatic columns, the material of magmatic-replacement zones can flow through fractures cutting across the host metasomatic rocks and thereby intersect the outer metasomatic zones.     

Rb-Sr dating of the impact melt from East Clearwater,Quebec

The Rb-Sr composition of eight melt rock and three basement samples from the East Clearwater impact structure, Quebec, and two basement samples from the West Clearwater structure has been determined. The whole rock ⁸⁷Sr/⁸⁶Sr ratios of the melt samples, 0.7167–0.7253, are within the range of the basement samples, 0.7054–0.7322, and provide further evidence that the melt rocks represent shock-melted basement. A mineral isochron obtained from a relatively coarse grained melt rock gives an age of 287±26 Ma for the crystallization age of the melt. This is equivalent to K-Ar whole-rock ages of 285±30 Ma and 300±30 Ma and a Rb-Sr age of 266±15 Ma obtained on melt rocks from West Clearwater and confirms the previously generally held assumption that the East and West Clearwater structures resulted from the simultaneous impact of two bodies at 285–300 Ma ago.Contribution from the Earth Physics Branch No. 909     

Origin of rapidly solidified metal-troilite grains in chondrites and iron meteorites

Inclusions of troilite and metallic Fe,Ni 0.2–4 mm in size with a dendritic or cellular texture were observed in 12 ordinary chondrites. Cooling rates in the interval 1400?950°C calculated from the spacing of secondary dendrite arms or cell widths and published experimental data range from 10^?7 to 10⁴°C/sec. In 8 of these chondrites, which are breccias containing some normal slow-cooled metal grains, the inclusions solidified before they were incorporated into the breccias. Their cooling rates of 1–300 °C/sec indicate cooling by radiation, or by conduction in contact with cold silicate or hot silicate volumes only 6–40 mm in size. This is quantitative evidence that these inclusions and their associated clasts were melted on the surface of a parent body (by impact), and were not formed at depth from an internally derived melt. In Ramsdorf, Rose City and Shaw, which show extensive reheating to ? 1000°C, Fe-FeS textures in melted areas are coarser and indicate cooling rates of 10^?1 to 10^?4°C/sec during solidification. This metal may have solidified inside hot silicate volumes that were 10–300 cm in size. As Shaw and Rose City are breccias of unmelted and melted material, their melted metal did not necessarily cool through 1000°C within a few m of the surface. Shock-melted, fine-grained, irregular intergrowths of metal and troilite formed in situ in many irons and some chondrites by rapid solidification at cooling rates of ? 10⁵°C/sec. Their kamacite and taenite compositions may result from annealing at ~250°C of metallic glass or exceedingly fine-grained quench products.     

Petrogenesis of municipal solid waste combustion bottom ash

A petrographic study was conducted on a suite of bottom ash particles from 3 different modern municipal solid waste combustors. The object of the study was to evaluate the mineralogical characteristics and formation process of the bottom ash by using standard geological techniques of light microscopy, electron microscopy, and X-ray microanalysis. This information was subsequently used to model the bottom ash petrogenesis based upon an examination of the mineralogy, melt structure, and composition of the ash.Bottom ash can be divided into two major groups: 1) refractory waste products and 2) melt products. The refractory waste products consist largely of rock and mineral fragments, various waste metals, and unmelted glass shards. The melt products consist of two distinct glasses: 1) isotropic glass, and 2) opaque glass. Complex silicate minerals are precipitated from and are abundant in the isotropic glass whereas both metal oxide and silicate minerals are precipitated from the opaque glass.The isotropic and opaque glasses formed simultaneously in different locations on the combustor grate. The contrast in melting (liquidus) temperatures shown by these glasses suggests that the isotropic melts were produced at localized hot spots (1500°C to 1650°C) and the opaque melts formed at cold spots (1150°C to 1400°C) on the grate. This could be the result of heterogeneous distribution of combustible municipal solid waste on the grate or from localized hot spots where air is introduced through the grates. In some instances the two glasses then had the opportunity to variably mix with each other. Fe-oxides represent waste metal fragments that were assimilated by melting and later recrystallized.Bottom ash is produced via a co-mingled two melt system that forms melilite-bearing, alkaline, volcanic-like rocks. The great similarity of the bottom ash residues between these 3 different MSW combustors suggests that, despite variable combustor designs and heterogeneous waste feed, high temperature combustion of MSW produces bottom ash of fairly uniform composition and structure that formed via the petrogenetic process described above. Alterations to the combustion process or implementations of secondary treatment technologies may render the bottom ash residue into a more environmentally stable material better suited for aggregate or long term secure disposal in landfills.     

Spatial variations in the geochemistry of quaternary lavas across the Sunda arc in Java and Bali

Since Mesozoic time, Java and Bali have formed part of an evolving system of island arcs comprising the Sunda arc of Indonesia. The present tectonic setting is relatively simple with subduction occurring at the Java Trench to the south. A north-dipping Benioff seismic zone delineates an underthrust lithospheric slab to depths of approximately 600 km beneath the Java Sea. Quaternary lavas of the normal island arc association range from tholeiites to high-K calc-alkaline lavas over Benioff zone depths from 120–250 km, respectively. More abundant calc-alkaline lavas lie between these extremes. High-K alkaline lavas are found over Benioff zone depths in excess of 300 km.Both within and between these groups of rocks there are consistent spatial variations in the observed geochemistry. For approximately 200 rocks, incompatible elements such as K, Rb, Cs, Sr, Ba, light REE, U and Th show an increase in abundance of almost an order of magnitude with increasing depth to the seismic zone. Abundances of compatible elements show little consistent variation and trace elements such as Ni, Co, Cr, and Sc are characteristically depleted except in some of the alkaline lavas. Major element abundances in rocks of the normal island arc association show little variation, except for K and P, which both increase in abundance across the arc and Al, which shows a relative decrease.The major and trace element data are inconsistent with the derivation of the analyzed rocks by partial melting of the crustal component of the subducted lithosphere. On the other hand, low Ni abundances (20 ppm) in the basalts suggest that most of the lavas are fractionated and few if any represent primary mantle-derived melts. The spatial variations in the geochemistry of erupted lavas across Java and Bali are best explained by a combination of two processes: melting of a geochemically zoned mantle source and smaller degrees of partial melting of that material at progressively greater depths. Primary tholeiitic magmas could be formed by 20–25% melting at depths of 30–40 km, primary high-K calc-alkaline magmas by 5–15% melting at 40–60 km depth, and primary alkaline magmas by 5% melting at depths of 80–90 km. The geochemical zoning in the mantle, which is also manifested by increasing ⁸⁷Sr/⁸⁶Sr ratios in lavas across the arc, is interpreted to result from the addition of a small melt fraction derived from the crustal component of the subducted lithosphere.     

The melt rocks of the Boltysh impact crater,Ukraine, USSR

The 100±12 m.y., 25 km diameter Boltysh impact crater was formed in Precambrian granites and granite gneisses of the Ukrainian Shield. The crater deposits have undergone minimal post-impact erosion and it is possible to study a complete vertical section of the underlying 200 m thick melt sheet. The melt rocks, as sampled in two drill holes, can be subdivided into two major textural classes: microcrystalline and glassy. The microcrystalline melt rocks form an uppermost and two lowermost units, with the glassy variety occupying the middle of the melt sheet. The microcrystalline units contain 25% zoned plagioclase phenocrysts set in a microcrystalline matrix of intergrown alkali feldspar and quartz. Pyroxene has been replaced by sheet-silicates. Mineral and lithic clasts make up 5–15% and show varying degrees of shock and resorption. The glassy melt rocks are characterized by 10–30% zoned plagioclase and 5–10% orthopyroxene set in a fresh to partially devitrified glassy matrix. Clast content is <5%. Chemically, the melt rocks are relatively homogeneous and correspond to a mixture of Kirovograd granites and gneisses in the ratio of 5 to 1, with Ni, Ir and Cr showing slight enrichments over the target rocks. There are minor differences in the Fe₂O₃/FeO ratio and the alkalis between the microcrystalline and glassy varieties. The increase in matrix crystallinity at the upper and lower contacts is contrary to observations at other impact melt sheets, where greater matrix crystallinity occurs in the interiors of the melt sheets. One possible explanation is that the melt matrix was originally glassy throughout, due to its high SiO₂ content, and the microcrystalline matrix is the result of extensive devitrification involving minor alkali exchange with circulating ground-waters.Contribution from the Geological Survey of Canada 40986     

Chemistry and thermal history of metal particles in Luna 20 soils

Individual metal particles from Luna 20 thin sections 521, 513 and 514 as well as several small metallic inclusions in silicate particles from Luna 20 thin sections 501 and 502 were examined using optical microscopy and the electron microprobe. All the metallic particles and inclusions analyzed are of meteoritic Co-Ni content as are most of the metallic particles from the Fra Mauro and the Apollo 16 highlands sites. It is proposed that most of the metal at these 3 sites had its origin in the meteoritic projectiles that bombarded and accumulated in the early lunar crust. It is apparent that the metallic particles and some of the metallic inclusions in the Luna 20 soil have been subjected to reheating on the Moon and this process has removed any evidence of the original meteoritic microstructure of the metal.     

Generation, mobilization and crystallization of impact-induced alkali-rich melts in granitic target rocks: Evidence from the Araguainha impact structure, central Brazil

This paper provides important insights into the generation, extraction and crystallization of clast-laden impact melt rocks from the Araguainha impact structure, central Brazil. Despite the mixed nature of the Araguainha target rocks (comprising a 2 km thick sequence of sedimentary rocks and underlying granitic basement), the exposed melt bodies are characterised by an alkali-rich granitic matrix embedding mineral and rock fragments derived only from the target granite. The melt rocks occur in the form of a massive impact melt sheet overlying the eroded central uplift structure, and as melt veins in the granite of the core of the central uplift. Bulk-rock major and trace element data (including platinum group elements) indicate that the precursor melts were generated locally, principally by partial melting of the target granite, without any contribution from the sedimentary sequence or the projectile. The dense network of melt veins was formed in isolation, by selective melting of plagioclase and alkali feldspar within the granite target. Plagioclase and alkali feldspar melted discretely and congruently, producing domains in the matrix of the melt veins, which closely match the stoichiometry of these minerals. The compositionally discrete initial melt phases migrated through a dense network of microfractures before being assembled into larger melt veins. Freezing of the melt veins was substantially fast, and the melt components were quenched in the form of alkali-feldspar and plagioclase schlieren in the matrix of the melt veins. The overlying impact melt rock is, in contrast, characterised by a granophyric matrix consisting of albite, sanidine, quartz, biotite and chlorite. In this case, melt components appear to have been more mobile and to have mixed completely to form a granitic parental melt. We relate the melting of the minerals to post-shock temperatures that exceeded the melting point of feldspars.