The early solar system contained short-lived radionuclides such as 60 Fe (t 1/2 = 1.5 Myr) whose most likely source was a nearby supernova. Previous models of solar system formation considered a supernova shock that triggered the collapse of the Sun's nascent molecular cloud. We advocate an alternative hypothesis, that the solar system's protoplanetary disk had already formed when a very close (<1 pc) supernova injected radioactive material directly into the disk. We conduct the first numerical simulations designed to answer two questions related to this hypothesis: Will the disk be destroyed by such a close supernova, and will any of the ejecta be mixed into the disk? Our simulations demonstrate that the disk does not absorb enough momentum from the shock to escape the protostar to which it is bound. Only low amounts (<1%) of mass loss occur, due to stripping by Kelvin-Helmholtz instabilities across the top of the disk, which also mix into the disk about 1% of the intercepted ejecta. These low efficiencies of destruction and injection are due to the fact that the high disk pressures prevent the ejecta from penetrating far into the disk before stalling. Injection of gas-phase ejecta is too inefficient to be consistent with the abundances of radionuclides inferred from meteorites. On the other hand, the radionuclides found in meteorites would have condensed into dust grains in the supernova ejecta, and we argue that such grains will be injected directly into the disk with nearly 100% efficiency. The meteoritic abundances of the short-lived radionuclides such as 60 Fe therefore are consistent with injection of grains condensed from the ejecta of a nearby (<1 pc) supernova, into an already formed protoplanetary disk.

No abstract

No abstract

No abstract

It has been proposed that some meteorites, CB and CH chondrites, contain material formed as a result of a protoplanetary collision during accretion. Their melt droplets (chondrules) and FeNi metal are proposed to have formed by evaporation and condensation in the resulting impact plume. We observe that the skeletal olivine (SO) chondrules in CB b chondrites have a blebby texture and an enrichment in refractory elements not found in normal chondrules. Because the texture requires complete melting, their maximum liquidus temperature of 1928 K represents a minimum temperature for the putative plume. Dynamic crystallization experiments show that the SO texture can be created only by brief reheating episodes during crystallization, giving a partial dissolution of olivine. The ejecta plume formed in a smoothed particle hydrodynamics simulation served as the basis for 3D modeling with the adaptive mesh refinement code FLASH4.3. Tracer particles that move with the fluid cells are used to measure the in situ cooling rates. Their cooling rates are ~10,000 K hr −1 briefly at peak temperature and, in the densest regions of the plume, ~100 K hr −1 for 1400-1600 K. A small fraction of cells is seen to be heating at any one time, with heating spikes explained by the compression of parcels of gas in a heterogeneous patchy plume. These temperature fluctuations are comparable to those required in crystallization experiments. For the first time, we find an agreement between experiments and models that supports the plume model specifically for the formation of CB b chondrules.