Search for r-process Pu-244 in the K-Pg boundary layer

The K-Pg (Cretaceous–Paleogene) boundary at 66 Ma marks one of five major mass extinctions in Earth’s fossil history. Based on strong enrichments of the platinum-group elements in the boundary layer, Alvarez et al. [1], in 1980, suggested that the impact of a large asteroid was responsible for the K/Pg event.
Earlier, other possible causes for the mass extinction, e.g., a nearby supernova(SN)-explosion, were also discussed, and indeed also Alvarez et al. initially considered this option to explain the high Ir concentration. However, to explain the observed Ir content, the distance for a SN would have to be less than one light-year. To exclude the SN option for the K-Pg event, they searched for a specific long-lived radionuclide, 244Pu, which has a half-life of 81 Myr and does not exist naturally on Earth. Assuming that this radionuclide is predominantly produced and ejected in SNe, its presence could indicate a nearby SN. No 244Pu at required levels was detected, leaving an impact as the most plausible cause (which was later confirmed by the discovery of shocked minerals and also a source crater, Chicxulub). However, since 1980, strong evidence evolved that the heavy r-process elements, e.g., actinides such as 244Pu, are produced in rare explosive events (ca. 1000 times less frequent than common type II core-collapse SNe in the galaxy) [2]. Neutron star mergers are potential candidates or rare subsets of SNe. Thus, the common core-collapse SNe might not have contributed significantly to actinide nucleosynthesis for the past few 100 Myr. This assumption agrees also with recent observations following the gravitational-wave event GW170817 [3]. Furthermore, by searching deep-sea archives for interstellar signatures we confirmed recently that nucleosynthesis yields of 244Pu are much lower (possibly a factor of 100) than expected if SNe dominate heavy isotope r-process nucleosynthesis [4-6]. However, the detection of a significant 244Pu influx above background into these terrestrial archives suggests the possibility of a nearby explosive event within the past few hundred millions years, possibly from a rare event. Thus, a small r-process contribution to actinide production from SNe is still a possibility. In general, site and frequency of r-process events are still strongly debated [2]. Thus, in contrast to the assumption of Alvarez et al. [1], it is not clear that non-detection of 244Pu excludes a nearby supernova explosion at 66 Ma. Despite the overwhelming evidence for an asteroid impact, a new method for direct atom counting has emerged with superior detection efficiency for 244Pu: since the original work by Alvarez et al. in 1980, the 244Pu detection-sensitivity has improved by more than a factor of a million by applying the method of Accelerator Mass Spectrometry (AMS) [5,7,8]. This enormous gain in abundance sensitivity prompted us to reinvestigate the 244Pu concentration in the K-Pg boundary layers. Here we present first results for a set of samples covering this transition period from the Cretaceous to the Paleogene.