Projectile-target interaction and rapid, high-temperature geochemical processes in impact melts

The collision of solid bodies at cosmic encounter velocities produces a variety of distinct physical, morphological, mineralogical, and petrological characteristics that are unique to hypervelocity impacts. Among these is melting of the projectile and substantial parts of the target it strikes. The resulting impact melts are typically intimate mixtures of melts from different precursor lithologies that initially occupied different stratigraphic positions in the crater’s melt zone; at small craters, projectile-derived melts may contribute substantially to the resulting melt mixture. These melts are mixed with each other during crater formation, bringing into contact melts that might have quite diverse physicochemical properties that reflect the different precursor lithologies from which they formed. This thesis investigates these processes by presenting a novel and so-far unique combination of petrologic observations on pristine natural impact glasses and impact melt rocks with results from specifically tailored laser irradiation and hypervelocity impact experiments. This thesis is initially concerned with an in-situ reconstruction of the interface between projectile and target materials in impact experiments that involved impact conditions similar to those during impacts of metallic micrometeoroids into regoliths of asteroids in the main belt. It is shown that under these impact conditions, substantial amounts of projectile melt remain as a continuous melt coating within the crater, and that projectile-coated, heterogeneous melt particles are produced that have a layered structure manifested in distinct layers of decreasing shock metamorphism (ranging from complete melting to below Hugoniot elastic limit; Chapter 5).These melt particles essentially sample the floor of the transient crater at early time steps of crater formation (before the transient crater reaches its final dimensions; Chapter 6), and, thus, preserve the original interface between projectile and target. Processes occurring along this interface are documented and discussed, and it is concluded that impacts of millimeter-size, metallic projectiles into asteroidal regoliths at typical impact velocities in the main belt result in qualitatively similar melt particles, but also that such planetary surfaces might accrete considerable amounts of foreign (i.e. non-endogenic) material. If such melt particles are retained within such regoliths, it is likely that their projectile component might influence surface reflectance spectra obtained from such surfaces. To further explore the chemical interaction between impact melts of diverse composition and structure, the thesis is then concerned with presenting and discussing a novel experimental approach capable of quasi-instantaneously producing, on macroscopic scales, melts and vapors from natural planetary materials by means of direct, continuous-wave laser irradiation (Chapter 7). These experiments simulate post-shock pressure–temperature conditions of hypervelocity impacts in the 4 to 20 km/s range, and experimental products (silicate glasses) are shown to be petrologically and thermodynamically similar to true impact melts (quenched to impact glasses) formed from similar starting materials. This is achieved by “matching” the entropy gains of the laser-generated melts to the entropy gains associated with the thermodynamic states produced in hypervelocity impacts at specific velocities. Thus, macroscopic volumes of melt can be produced that thermodynamically as well as petrologically resemble true impact melts formed at that specific impact velocity. The rest of the thesis is then concerned with investigating petrogenetic processes in impact melts that are so-far only poorly constrained. Chapters 8 and 9 investigate unmixing of silicate impact melts of diverse chemical composition and cooling history due to sensu stricto liquid immiscibility. This represents not only the first comprehensive study of silicate liquid immiscibility in terrestrial impact melts and experimental analogs, but also the first clear evidence of silicate liquid immiscibility in meteorites. Textural evidence of silicate-melt unmixing is presented, and it is shown that the compositions of the conjugate immiscible liquids (Si-rich and Fe-rich) are consistent with phase separation in high- or low-temperature two-liquid fields in the common petrologic (e.g., basaltic) system. Moreover, major-element partition coefficients are correlated with the degree of polymerization of the Si-rich melt. Hence, major-element partitioning between the conjugate liquids as well as the resulting emulsion textures are similar to those known from tholeiitic basalts, lunar basalts, and experimental analogs. However, in impact melts, the high-temperature, “quasi-binary” miscibility gaps (e.g., silica–ferrous oxide; T > 1700 °C) might be encountered that are inaccessible to endogenous magmatic systems. The characteristics of impact melt inhomogeneity produced by melt unmixing in a miscibility gap are then compared to impact melt inhomogeneity caused by incomplete homogenization of different (miscible or immiscible) impact melts, and petrographic tools are presented to distinguish between these two. Chapter 10 presents results from laser irradiation experiments aimed at constraining the fate of carbonates (calcite, dolomite) entrained in superheated silicate melts. Carbonate decomposition by coexisting silicate impact melt is shown to be extremely fast (tens of seconds), and results in contamination of silicate melt with carbonate-derived calcium oxide (and magnesium oxide in the case of dolomite) and release of carbon dioxide at the silicate melt–carbonate interface. Several, partially transient processes are shown to occur at this interface that are largely similar to the formation of calcic skarns during contact metamorphism. It is suggested that upon pressure release, “protracted” decomposition of carbonates by heat influx from coexisting silicate impact melt is an important, if not dominant, process during impact melting of mixed silicate–carbonate targets. However, a combination of these experimental findings, findings from previous studies, and consideration of the phase diagrams of calcite and quartz suggests that, as envisaged by a number of studies, carbonate impact melts are indeed readily produced during adiabatic decompression. Eventually, a working model is presented that suggests that hypervelocity impacts into mixed silicate–carbonate targets may involve both melting and decomposition of carbonates in specific parts of an impact crater, thus bringing together seemingly contrasting previous findings from impactites from terrestrial impact structures. The working model also suggests that impact-induced carbon dioxide degassing of carbonate-bearing targets during hypervelocity impact is more complex than previously thought, which has implications for estimating the net amount of carbon dioxide that is discharged into the atmosphere immediately after such impacts