[Paper Review] The Nickel Mass Distribution of Stripped-Envelope Supernovae: Implications for Additional Power Sources

Niloufar Afsariardchi, M. R. Drout+4 more|arXiv (Cornell University)|Sep 14, 2020

Gamma-ray bursts and supernovaePhysics and Astronomy166 references53 citations

TL;DR

This study re-evaluates 56Ni masses in 27 stripped-envelope supernovae (SESNe) by modeling light-curve tails, demonstrating that Arnett’s rule overestimates MNi by a factor of ~2. Using an improved analytic model (Khatami & Kasen 2019), the authors calibrate the β parameter observationally and find that ~7–50% of peak luminosity in SESNe likely arises from additional power sources such as shock cooling or magnetar spin-down, resolving a long-standing discrepancy with simulations and reinforcing that SESNe have significantly higher MNi than Type II SNe.

ABSTRACT

We perform a systematic study of the $^{56}$Ni mass ($M_{ m Ni}$) of 27 stripped envelope supernovae (SESNe) by modeling their light-curve tails, highlighting that use of ``Arnett's rule'' overestimates $M_{ m Ni}$ for SESN by a factor of $\sim$2. Recently, \citet{Khatami2019} presented a new model relating the peak time ($t_{ m p}$) and luminosity ($L_{ m p}$) of a radioactive-powered SN to its $M_{ m Ni}$ that addresses several limitations of Arnett-like models, but depends on a dimensionless parameter, $\beta$. Using observed $t_{ m p}$, $L_{ m p}$, and tail-measured $M_{ m Ni}$ values for 27 SESN, we observationally calibrate $\beta$ for the first time. Despite scatter, we demonstrate that the model of \citet{Khatami2019} with empirically-calibrated $\beta$ values provides significantly improved measurements of $M_{ m Ni}$ when only photospheric data is available. However, these observationally-constrained $\beta$ values are systematically lower than those inferred from numerical simulations, primarily because the observed sample has significantly higher (0.2-0.4 dex) $L_{ m p}$ for a given $M_{ m Ni}$. While effects due to composition, mixing, and asymmetry can increase $L_{ m p}$ current models cannot explain the systematically low $\beta$ values. However, the discrepancy can be alleviated if $\sim$7--50\% of $L_{ m p}$ for the observed sample originates from sources other than $^{56}$Ni. Either shock cooling or magnetar spin-down could provide the requisite luminosity. Finally, we find that even with our improved measurements, the $M_{ m Ni}$ values of SESN are still a factor of $\sim$3 larger than those of hydrogen-rich Type II SN, indicating that these supernovae are inherently different in terms of their progenitor initial mass distributions or explosion mechanisms.

Motivation & Objective

• To re-evaluate 56Ni masses in stripped-envelope supernovae (SESNe) using improved light-curve modeling to address systematic overestimation by Arnett’s rule.
• To observationally calibrate the β parameter in the Khatami & Kasen (2019) analytic model, which accounts for delayed energy deposition relative to observed luminosity.
• To investigate the origin of the discrepancy between observed β values and those inferred from numerical simulations, particularly given the observed sample’s higher peak luminosity for a given 56Ni mass.
• To assess whether additional power sources—such as shock cooling or magnetar spin-down—can explain the systematically lower β values and higher luminosity observed in SESNe.
• To determine whether the observed discrepancy in 56Ni masses between SESNe and H-rich Type II SNe persists with improved measurements, and to explore its implications for progenitor mass distributions and explosion mechanisms.

Proposed method

• Model the radioactive decay tail of 27 SESNe light curves to derive precise 56Ni masses (MNi), using a 1D radiative transfer approach with time-dependent energy deposition.
• Apply the Khatami & Kasen (2019) analytic model, which relates peak luminosity (Lp), peak time (tp), and MNi via a dimensionless parameter β to account for delayed energy deposition.
• Use observed tp, Lp, and tail-derived MNi values to observationally calibrate β for each SN, yielding a sample of empirically constrained β values.
• Compare the observed β values to those predicted by numerical simulations (e.g., Dessart et al. 2016; Ertl et al. 2019) to quantify the discrepancy.
• Assess the impact of composition, mixing, and asymmetry on β and peak luminosity using theoretical models and sensitivity tests.
• Evaluate the viability of additional power sources—shock cooling and magnetar spin-down—by calculating the required luminosity and timescale to explain the observed luminosity excess.

Experimental results

Research questions

• RQ1 Does Arnett’s rule systematically overestimate 56Ni masses in stripped-envelope supernovae, and by how much?
• RQ2 Can the Khatami & Kasen (2019) model with observationally calibrated β values provide more accurate 56Ni mass estimates than Arnett’s rule when only photospheric data is available?
• RQ3 Why are the empirically derived β values systematically lower than those from numerical simulations, despite similar physical inputs?
• RQ4 To what extent can additional power sources such as shock cooling or magnetar spin-down explain the observed luminosity excess and low β values in SESNe?
• RQ5 Does the observed 56Ni mass distribution in SESNe remain significantly higher than in H-rich Type II SNe when using improved tail-based measurements?

Key findings

• The 56Ni masses derived from light-curve tails range from 0.03 M⊙ to 0.57 M⊙, with a median of 0.08 M⊙, and Type Ic-BL SNe show a higher median MNi of 0.15 M⊙.
• Arnett’s rule overestimates 56Ni masses by a factor of approximately 2 compared to tail-based measurements, a discrepancy larger than previously reported in numerical simulations.
• The observationally calibrated β values range from 0.0 to 1.71, with a median of 0.70, and exhibit a standard deviation of 0.34, indicating significant scatter across the sample.
• The Khatami & Kasen (2019) model with median calibrated β values yields significantly improved 56Ni mass estimates compared to Arnett’s rule when only photospheric data is available.
• The observed sample exhibits peak luminosities that are 0.3–0.4 dex higher than simulated models for the same 56Ni mass, which explains the systematically lower β values in observations.
• The discrepancy between observed and simulated β values can be resolved if an additional power source contributes between 7% and 50% of the peak luminosity, corresponding to 2.5×10⁴¹ to 5.5×10⁴² erg s⁻¹, with both shock cooling and magnetar spin-down being viable candidates.