Explodability matters: how realistic neutrino-driven explosions change explosive nucleosynthesis yields

Abstract

Explosive nucleosynthesis is affected by many uncertainties, particularly regarding assumptions and prescriptions adopted during the evolution of the star. Moreover, simple explosion models are often used in the literature, which can introduce large errors in the assumed explosion energy and mass cut. In this paper, our goal is to analyze the explosion properties and nucleosynthesis of a large range of progenitors from three different stellar evolution codes: FRANEC, KEPLER, and MESA. In particular, we will show the differences between the neutrino-driven explosions simulated in this work with the much simpler bomb and piston models that are typically widely used in the literature. We will then focus on the impact of different explodabilities and different explosion dynamics on the nucleosynthetic yields. We adopt the neutrino-driven core-collapse supernova explosion code GR1D+, i.e. a spherically symmetric model with state-of-the-art microphysics and neutrino transport and a time-dependent mixing-length model for neutrino-driven convection. We carry out explosions up to several seconds after bounce, and then calculate the nucleosynthetic yields with the post-processing code SkyNet. We find that our 1D+ simulations yield explosion energies and remnant masses in agreement with observations of type II-P, IIb, and Ib supernovae, as well as with the most recent 3D simulations of the explosion. We provide a complete set of yields for all the stars simulated, including rotating, low-metallicity, and binary progenitors. Finally, we find that piston and bomb models, compared to more realistic neutrino-driven explosions, can artificially increase the production of Fe-peak elements, whereas the different explodability tends to cause discrepancies in the lighter elements.

Figures (27)

• Figure 1: Comparison of explosion energies and ejected $\ce{^56Ni}$ masses obtained from the explosion simulations in this work (indicated by stars) with estimates based on hydrodynamical or semi-analytical modeling of observed light curves. The shaded circles and squares represent simulations of stripped (i.e., with less than $0.01 M_\odot$ of hydrogen in the envelope) and ultra-stripped (i.e., with less than $0.01 M_\odot$ of helium in the envelope) progenitors, respectively. The brown and yellow bands are the semi-analytical fits based on the light-curve models of Pejcha2015_typeIIP_fit using the calibrations of Litvinova1985_light_curve_params and Popov1993_AnModel_IIp, respectively. The magenta line is a fit presented by Fang2025_typeII_light_curve_modeling to the 2D simulations from Bruenn2016_expl_en and to the 3D simulations of Burrows2024_long3D_kicks_spins, which is in good agreement with our 1D+ simulations. The cyan line is a fit derived by Fang2025_typeII_light_curve_modeling based on observations of 32 type II SNe, where instead of assuming a fixed wind model, they used a model grid with different hydrogen envelope masses, which lessens the tension between simulations and observations.
• Figure 2: This figure shows a clear correlation between the explosion energy and the gravitational mass of a cold neutron star at the end of the simulation, where different colors refer to the different sets of simulations described in Section \ref{['sec:methods']}. A similar correlation is also found in 3D simulations Burrows2024_Phys_correlations.
• Figure 3: The abundances for selected isotopes after the explosive nucleosynthesis are shown for three selected progenitors from LC18 (left) and WH07 (middle and right). The solid lines show the results from the explosion simulated in this work with GR1D, the dashed lines show the bomb model of LC18, and the dotted lines show the piston model of WH07. The progenitor on the left was chosen to be the one for which the mass cut derived with GR1D is closest to the one in the original work. The progenitor in the middle was chosen instead to be the one for which the explosion energy derived with GR1D is the most different from the one in the original work of 1.4 B. The progenitor on the right was chosen instead to be the one for which the explosion energy derived with GR1D is closest to the one in the original work of 1.4 B. As one can see, the left and middle panels are quite similar, showing that what causes the main difference in explosion dynamics is the explosion energy, regardless of whether a piston or a bomb is used. When the explosion energies are similar (i.e. the right panel), the differences are smaller.
• Figure 4: Explodability of progenitor stars from WH07 as obtained in this work (upper panel) and by Curtis2019_PUSHIII_nucleosynthesis (bottom panel. Successful explosions, defined as simulations for which the shock is successfully revived and crosses 500 km, are shown as green bands. Failed explosions are shown as black bands.
• Figure 5: Explodability of progenitor stars from LC18 as obtained in this work, shown as a function of initial rotational velocity at the beginning of the Main Sequence, and metallicity. Successful explosions, defined as simulations for which the shock is successfully revived and crosses 500 km, are shown as green bands. Failed explosions are shown as black bands.