In this project, Dr. Steven J. Desch of Arizona State University will undertake a theoretical study of how young stars are contaminated by ejected material from nearby supernova explosions. Supernova contamination of forming planetary systems has many observable consequences for the elemental abundances of stars. The contamination determines the inventory of short-lived radionuclides such as Aluminum-26 or Iron-60 in young stellar systems, and is a major driver of planetary evolution and the delivery of bioessential volatiles to Earth-like planets.
To determine the effects of supernova contamination, Dr. Desch and his students will numerically model the interaction of supernova ejecta with protoplanetary disks and with fragments of molecular clouds on the verge of collapse. They will also create a computational tool to allow exploration of dust dynamics and thermal evolution in these and other astrophysical flows.
The proposed research connects research and education by advancing discovery and understanding while at the same time educating undergraduate and graduate students, including one named student from an underrepresented group. The proposed research will create a platform on which to expose a younger generation to a multidisciplinary scientific process. The investigators will incorporate research developments from this proposal into the classroom as part of a recently revamped School of Earth and Space Exploration.
Our grant from the NSF funded us to use numerical simulations to assess different models of how supernova material could have been injected into our Solar System as it was forming. Isotopic studies of meteorites tell us that our Solar System contained such radioactive isotopes as aluminum 26 (half life 0.7 million years), calcium 41 (half life 0.1 million years) and iron 60 (half life 2.3 million years). Because their half lives are shorter than the time for stars to form, they must have been created soon before the Solar System formed and then injected into the dusty gas from which planets later formed. The most likely source of these new isotopes is thought to be a supernova, the collapse and explosion of a massive star. We tested two models for this injection process: injection of supernova material into the gas of a nearby (2 to 4 parsecs) molecular cloud, as stars are forming; and injection of supernova material into a nearby (0.1 to 2 parsecs) protoplanetary disk around an already-formed star. We were motivated to study injection into an already formed disk by Hubble Space Telescope (HST) images so-called "H II regions". These are astrophysical environments in which the massive (> 8 solar masses) stars that will go supernova are found, and they are characterized by low-density, ionized gas near the massive star, surrounded at a distance of 2 to 4 parsecs by denser molecular clouds. WIthin these H II regions, HST has observed thousands of protostars with disks (like the one our planets formed from), at distances 0.1 to 1 parsec from the massive stars. Our numerical simulations have shown that the supernova will not destroy these disks despite their proximity, but also that supernova gas does not efficiently mix into the disk. However, if large (> 1 micron diameter) dust grains condensed from the supernova ejecta, they would not be swept around the disks with the rest of the supernova gas, and they could deposit supernova material into the disk. We demonstrated that this has the potential to explain the abundance of isotopes like aluminum 26 in our Solar System; however, the disk would have to be close (0.1 parsecs) from the supernova, and just-formed disks are typically found 2 to 4 parsecs away. We synthesized results from the astrophysics literature to show that supernova ejecta are generally clumpy, like HST images of the supernova remnant Cassiopeia A. We showed that injection of such a supernova "bullet" into a disk at several parsecs could still explain the abundances of aluminum 26, etc., but we also assessed the probability that this happened to our Solar System at less than 1 in a thousand: plausible but not likely. We next investigated injection of supernova material into the molecular cloud that surrounds the H II region. This gas is known to actively form stars even many millions of years into the H II region's evolution, after the first (and most massive) stars have formed. When one of these massive stars explodes as a supernova after 4 million years, the ejecta can interact directly with star-forming gas. We conducted state-of-the-art numerical simulations involving about half a million CPU-hours on the 2000-node supercomputer cluster Saguaro at Arizona State University, using the FLASH hydrodynamics code. We found that if a supernova explodes symmetrically, the ejecta are unlikely to penetrate the cloud, but if the supernova ejecta are clumpy (as we had previously argued), then the supernova bullets do penetrate effectively. The calculations were complicated by cooling instabilities in the gas (as the gas cools it gets denser, allowing it to cool faster), which tend to fragment the bullets in chaotic, unpredicatble ways. But a robust result is that supernova bullets penetrate about half a parsec into the cloud, carving out a channel and depositing supernova material along that channel. Solar systems forming later from that gas should contain about 1 part in 10,000 supernova material, in accord with the estimates from meteorites. Because each supernova bullet samples a different zone within the exploding star, it is difficult to predict how much of each isotope the Sun should have been born with, but reasonable assumptions yield the meteoritic abundances of aluminum 26, etc. Moreover, we showed that the probability of the Sun forming in this manner to be > 10%, indicating this may be a universal mechanism in forming solar systems. Our results have implications for how quickly stars in galaxies acquire heavy elements, for determining from elemental or isotopic anomalies whether a star formed in an H II region or not, and clarifies many details about the origin of the Sun and planets in particular. In future work we will try to constrain the mass of the supernova progenitor star, and determine exactly what zones within the star contaminated our solar system.
