The Progenitor Dependence of Core-collapse Supernovae from Three-dimensional Simulations with Progenitor Models of 12–40 M_⊙
We present a first study of the progenitor star dependence of the three-dimensional (3D) neutrino mechanism of core-collapse supernovae. We employ full 3D general-relativistic multi-group neutrino radiation-hydrodynamics and simulate the postbounce evolutions of progenitors with zero-age main sequence masses of 12, 15, 20, 27, and 40 M_⊙. All progenitors, with the exception of the 12 M_⊙ star, experience shock runaway by the end of their simulations. In most cases, a strongly asymmetric explosion will result. We find three qualitatively distinct evolutions that suggest a complex dependence of explosion dynamics on progenitor density structure, neutrino heating, and 3D flow. (1) Progenitors with massive cores, shallow density profiles, and high post-core-bounce accretion rates experience very strong neutrino heating and neutrino-driven turbulent convection, leading to early shock runaway. Accretion continues at a high rate, likely leading to black hole formation. (2) Intermediate progenitors experience neutrino-driven, turbulence-aided explosions triggered by the arrival of density discontinuities at the shock. These occur typically at the silicon/silicon–oxygen shell boundary. (3) Progenitors with small cores and density profiles without strong discontinuities experience shock recession and develop the 3D standing-accretion shock instability (SASI). Shock runaway ensues late, once declining accretion rate, SASI, and neutrino-driven convection create favorable conditions. These differences in explosion times and dynamics result in a non-monotonic relationship between progenitor and compact remnant mass.
Additional Information© 2018 American Astronomical Society. Received 2017 December 4. Accepted 2018 January 19. Published 2018 February 26. We acknowledge helpful discussions with D. Radice, H. Nagakura, Y. Suwa, K. Kiuchi, M. Shibata, J. Takeda, H.-T. Janka, R. Bollig, M. Obergaulinger, S. Couch, E. O'Connor, P. Mösta, and K. S. Thorne. This research was partially supported by NSF grants CAREER PHY-1151197, PHY-1404569, OAC-1550514, and AST-1333520, and the Sherman Fairchild Foundation. The simulations took over a year to complete and were carried out on O(10,000) compute cores of NSF/NCSA Blue Waters (PRAC ACI-1440083, OCI-0725070 and ACI-1238993, and Illinois allocation baov), of Edison at the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231, and of the Texas Advanced Computing Center Stampede and Stampede2 clusters under NSF XSEDE allocation TG-PHY100033. This article has been assigned Yukawa Institute for Theoretical Physics report No. YITP-17-122.
Published - Ott_2018_ApJL_855_L3.pdf
Submitted - 1712.01304.pdf