THE CRETACEOUS-PALEOGENE (K-PG) BOUNDARY SECTION AT SHELL CREEK, ALABAMA: ARGUMENTS FOR A PRIMARY IMPACT ORIGIN OF THE CALCITE SPHERULES

The about 50 to 75-cm-thick Chicxulub event bed at Shell Creek, Alabama, contains at its base an up to 25-cm-thick sand bed rich in impact spherules (King and Petruny 2008, GSA Spec Pap 437). These spherules, between about 250 µm and a few mm in size, consist of a calcite core, lined by a silicate shell varying in thickness, shape, mineralogical composition and color. Some of the spherules are hollow, and some show in situ collapse yet the shell structure is always similar. These impact spherules are generally interpreted as altered silicate glass spherules with a diagenetic laminated wall of smectite with the “sparry calcite core” representing a secondary infilling after the complete dissolution of the original silicate glass. Yet some observations do not match this interpretation: (i) Spherules frequently include smaller calcite spherules, (ii) spherules also consist completely of clay minerals, and (iii) internal bubbles are always lined by a 10 to 20 µm thick layer of smectite. Therefore, we interpret these very delicate internal structures as primary features. They match identical structures documented, for example in the Chicxulub event bed at La Lajilla, Mexico. There, the partly hydrated silicate glass spherules also contain bubbles filled with calcite or large cores of sparry calcite. Analogous to the Shell Creek section, these cores may be rich in bubbles, consist of clay minerals, and show several concave features at the contact to the outer silicate shell. Compared to the Shell Creek Chicxulub event bed, the ratio of silicate shell to calcite core is much higher for the La Lajilla spherules; this feature may reflect compositional differences in certain volumes of the ejecta plume. In conclusion, we interpret these impact spheroids as condensed or shock-melted calcite droplets that obtained a thin silicate melt shell in the expanding ejecta plume. The various internal structures (e.g., bubbles filled with smectite) may reflect liquid immiscibility features between silicate and carbonate melts. During deposition on the sea surface, most spherules were already well below the glass softening temperature but some display plastic deformation and welding. Alteration of the spherules was most probably restricted to the silicate glass – the calcite is part of the ejecta.