¹Shigeko Togashi, ²Noriko T. Kita, ¹Akihiko Tomiya, ^1,3Yuichi Morishita
Geochimica et Cosmochimica Acta (in Press) Link to Article [http://doi.org/10.1016/j.gca.2017.04.031]
¹Geological Survey of Japan, AIST, Central 7, Higashi 1-1-1, Tsukuba 305-8567, Japan
²Department of Geoscience, University of Wisconsin-Madison, 1215 W Dayton Street Madison, WI 53706-1692, USA
³Department of Geoscience, Shizuoka University, 836 Ohya, Suruga-ku, Shizuoka 422-8529, Japan
Copyright Elsevier

The compositions of host magmas of ferroan anorthosites (FAN-host magmas) were estimated from secondary ion mass spectrometry analyses of plagioclase in lunar highland rocks. The evolution of the magmas was investigated by considering phase relations based on the MELTS algorithm and by re-examining partition coefficients for trace elements between plagioclase and melts. Data little affected by post-magmatic processes were selected by using plagioclase with relatively primitive Sc and Co contents. The FAN-host magma contained 90–174 ppm Sr, 40–119 ppm Ba and 0.5–1.3% TiO2, and had sub-chondritic Sr/Ba and Ti/Ba ratios. It is difficult to account for the formation of FAN-host magma on the basis of magma evolution processes of previously proposed bulk silicate Moon models with chondritic ratios for refractory elements at global scale. Therefore, the source of the FAN-host magma must have had primordial sub-chondritic Sr/Ba and Ti/Ba ratios. The FAN-host magmas were consistent in refractory elements with the estimated host mafic magma for feldspathic crust based on lunar meteorites, and some very-low-Ti mare rocks from lunar meteorites. Here, we propose an alternative bulk silicate Moon model (the cBSM model), which is enriched in crustal components of proto-bodies relative to the present whole Earth–Moon system.

^1,2Lukáš Ackerman, ²Tomáš Magna, ¹Karel Žák, ¹Roman Skála, ¹Šárka Jonášová, ³Jiří Mizera, ³Zdeněk Řanda
Geochimica et Cosmochimica Acta (in Press) Link to Article [http://doi.org/10.1016/j.gca.2017.04.028]
¹Institute of Geology, The Czech Academy of Sciences, Rozvojová 269, CZ-165 00 Prague 6 – Lysolaje, Czech Republic
²Czech Geological Survey, Klárov 3, CZ-118 21 Prague 1, Czech Republic
³Nuclear Physics Institute, The Czech Academy of Sciences, Hlavní 130, CZ-250 68 Husinec-Řež, Czech Republic
Copyright Elsevier

Impact processes are natural phenomena that contribute to a variety of physico–chemical mechanisms over an extreme range of shock pressures and temperatures, otherwise seldomly achieved in the Earth’s crust through other processes. Under these extreme conditions with transient temperatures and pressures ≥3,000K and ≥100 GPa, followed by their rapid decrease, the behavior of elements has remained poorly understood. Distal glassy ejecta (tektites) were produced in early phases of contact between the Earth’s surface and an impacting body. Here we provide evidence for a complex behavior of Os and other highly siderophile elements (HSE; Ir, Ru, Pt, Pd, and Re) during tektite production related to a hyper-velocity impact that formed the Ries structure in Germany. Instead of simple mixing between the surface materials, which are thought to form the major source of central European tektites (moldavites), and impactor matter, the patterns of HSE contents and 187Re/188Os – 187Os/188Os ratios in moldavites, target sediments and impact-related breccias (suevites) can be explained by several sequential and/or contemporary processes. These involve (i) evaporative loss of partially oxidized HSE from the overheated tektite melt, (ii) mixing of target-derived and impactor-derived HSE vapor (plasma) phases, and (iii) early (high-temperature) condensation of a part of the mixed vapor phase back to silicate melt droplets. An almost complete loss of terrestrial Os from the tektite melt and its replacement with extra-terrestrial Os are indicated by low 187Os/188Os ratios in tektites (<0.163) relative to precursor materials (>0.69). This is paralleled by a co-variation between Os and Ni contents in tektites but not in suevites formed later in the impact process.