GEOCHEMISTRY OF DARWIN IMPACT GLASS AND TARGET ROCKS

We have analyzed the major and trace element composition of 18 Darwin glass samples and 7 target rocks (sandstones, shales, and a quartzite) from the Darwin crater area. On the basis of our data, and using statistical methods, 3 chemically distinct groups of Darwin glass were identified: A (low Fe, Al = LFe,Al, or average Darwin glass group), B (HFe,Al group), and C (HMg,Na group). The glasses of group C also show anomalous enrichments of several elements e.g., Cr, Mn, Co, and Ni. Electron microprobe studies show that the glasses are inhomogeneous on the micrometer scale, which is typical for impact glasses. The geochemistry of all glasses is very similar to terrestrial sediments and thus supports the impact origin model. We have performed mixing calculations which show that in general Darwin glass can be formed by melting and mixing local target rocks. The best fit is obtained for a mixture of 30% quartzite, 60% shale B3-DG, and 10% shale B1-DG. Some major element contents do not agree exactly, which is most probably due to the limited selection of target rocks that were available for our study. The analyses and mixing models demonstrate that volatile elements (e.g., Zn, Ga, Sb, and the alkalies) have been lost during production of the impact glasses, which can be expected because of the high formation temperature. We have furthermore tried to explain the enrichments of Cr, Mn, Co, and Ni in group C glasses by contributions from a non-sedimentary source, e.g., ultrabasic rocks, or from the impacting body. None of the mixtures provides a satisfactory fit. Darwin glass does not show any significant Ir enrichments. Admixture of material from iron meteorites gives too high Fe, Co, and Ni, and too low Cr and Mn contents. Chondritic contaminations would yield Ir abundances in the glass that are several orders of magnitude above the observed levels. Better fits are obtained for an achondritic contamination, but again give excess Ir. An ultrabasic contribution gives better results, except for higher Mg, but no such rocks are known from the target area. Thus, at the present time, we are not able to explain the enrichments of Cr, Mn, Co, and Ni in glasses of group C in a satisfactory way. © 1990.