The Search for Stable Nickel: Investigating the Origins of Type Ia Supernovae with Late-time NIR Spectroscopy from the Carnegie Supernova Project-II

1. Problem Statement

Type Ia supernovae (SNe Ia) are critical "standardizable candles" for cosmology, yet their progenitor systems (single-degenerate vs. double-degenerate) and explosion mechanisms (e.g., Delayed Detonation vs. Helium Detonation) remain debated. A key discriminator is the production of stable iron-group elements (IGEs), specifically stable Nickel-58 ($^{58}\text{Ni}$).

• Physical Constraint: Producing stable $^{58}\text{Ni}$ requires sufficiently high central densities and temperatures, conditions typically predicted for near-Chandrasekhar-mass ($M_{\text{Ch}}$) white dwarfs but often lacking in sub-Chandrasekhar-mass ($\text{sub-}M_{\text{Ch}}$) models.
• Observational Challenge: While radioactive $^{56}\text{Ni}$ decays quickly, stable $^{58}\text{Ni}$ persists. Detecting it requires observing the innermost regions of the supernova ejecta once they become optically thin.
  • Optical: The optical $[\text{Ni II}] \lambda 7378$ Å line is heavily blended with $[\text{Fe II}]$ and $[\text{Ca II}]$ lines, making accurate parameter extraction difficult.
  • Mid-Infrared (MIR): While space telescopes (Spitzer, JWST) can observe MIR Ni lines, sample sizes are currently very small, and atomic data uncertainties exist.
  • Near-Infrared (NIR): The NIR $[\text{Ni II}] \lambda 1.939$ $\mu$m line is theoretically the strongest and best-isolated Ni feature. However, ground-based observations are hindered by strong telluric (atmospheric) water vapor absorption in the 1.8–1.95 $\mu$m region.

Goal: To overcome these challenges by utilizing a large, homogeneous, high signal-to-noise (S/N) sample of late-time NIR spectra to detect the $[\text{Ni II}] \lambda 1.939$ $\mu$m line, thereby constraining progenitor masses and explosion mechanisms.

2. Methodology

The study utilizes data from the Carnegie Supernova Project-II (CSP-II), focusing on low-redshift ($z \lesssim 0.04$) SNe Ia.

• Dataset:
  • Sample: 79 telluric-corrected NIR spectra of 22 SNe Ia.
  • Time Coverage: Phases ranging from +50 to +505 days post-peak brightness (covering both "transitional" and "nebular" phases).
  • Instrumentation: Observations were obtained using the FIRE spectrograph on the 6.5m Magellan-Baade Telescope (Las Campanas Observatory).
  • New Data: Includes 31 previously unpublished nebular-phase spectra.
• Data Reduction & Correction:
  • All spectra were reduced using the firehose pipeline.
  • Telluric Correction: A critical step involving the observation of A0V standard stars and the use of the xtellcor tool to remove atmospheric water vapor absorption, enabling analysis of the 1.9 $\mu$m region.
  • Quality Control: A subsample of 58 spectra from 18 SNe Ia was selected based on a high S/N ($\geq 5$) in the 1.8–2.1 $\mu$m region.
• Analysis Technique:
  • Multi-Gaussian Fitting: The 1.9 $\mu$m region was fitted with two Gaussian profiles plus a flat continuum:
    1. Red-side Gaussian: Represents the broad feature dominated by $[\text{Co III}] \lambda 2.0028$ $\mu$m (and potentially $[\text{Fe II}]$).
    2. Blue-side Gaussian: Represents the $[\text{Ni II}] \lambda 1.939$ $\mu$m line.
  • Detection Metric: The authors introduced the Gaussian Peak Ratio ($r$), defined as the peak flux of the blue-side Gaussian divided by the red-side Gaussian.
    • Threshold: A detection is confirmed if $r \geq 0.5$. This ensures the Ni feature is sufficiently isolated from blending to measure velocity and width accurately.
  • Subtyping: SNe were categorized using the color-stretch parameter ($s_{BV}$), distinguishing between subluminous 86G-like objects ($s_{BV} < 0.75$) and normal-bright SNe Ia ($s_{BV} > 0.75$).

3. Key Results

• Detection Statistics:
  • Stable Ni was detected in 8 out of 18 high-S/N SNe Ia ($r \geq 0.5$).
  • Subluminous 86G-like SNe: Showed the earliest and strongest detections. SN 2011iv and SN 2015bp (both 86G-like) exhibited detections as early as +54 days with high $r$ values ($>1.0$).
  • Normal-bright SNe: Generally showed no detectable isolated Ni features until $\sim +150$ days. Early detections in normal SNe (e.g., SN 2013ct at +78 days) were likely contaminated by permitted lines or blending.
• Kinematics:
  • Velocity: The average measured line velocity for stable Ni is $\sim 1200$ km s$^{-1}$. Most detections are redshifted, though the two 86G-like objects showed blueshifts.
  • Line Width: The average Full Width at Half Maximum (FWHM) is $\leq 3500$ km s$^{-1}$.
  • Implication: The low velocities and narrow widths indicate that stable $^{58}\text{Ni}$ is confined to the innermost, low-velocity regions of the ejecta.
• Statistical Distinction:
  • A Kolmogorov-Smirnov (KS) test revealed that the distribution of Gaussian Peak Ratios for 86G-like SNe is statistically distinct from normal-bright SNe, suggesting different underlying physical populations or evolution timescales.
• Line Identification:
  • The red-side Gaussian is identified as primarily $[\text{Co III}] \lambda 2.0028$ $\mu$m, which decays over time, causing the relative strength of the Ni line ($r$) to increase in the nebular phase.
  • Comparison with JWST MIR data confirms the identification of the NIR feature as stable Ni.

4. Significance and Implications

The findings provide strong observational constraints on SN Ia explosion theories:

1. Progenitor Mass and Central Density:

   • The presence of stable $^{58}\text{Ni}$ in the innermost regions implies high central densities during the explosion.
   • This favors near-Chandrasekhar-mass ($M_{\text{Ch}}$) progenitors, as $\text{sub-}M_{\text{Ch}}$ models typically fail to produce significant stable Ni or predict it at higher velocities.
   • Surprising Finding: The strongest Ni detections occurred in subluminous 86G-like SNe. This suggests that these objects, despite their low luminosity, likely originate from high-mass ($M_{\text{Ch}}$) progenitors rather than low-mass ones, challenging the simple assumption that subluminosity always equates to low progenitor mass.

2. Explosion Mechanism:

   • The narrow line widths and central confinement of Ni argue against large-scale mixing or violent merger scenarios that would disperse stable material to high velocities.
   • The results align best with Delayed Detonation (DDT) models of $M_{\text{Ch}}$ white dwarfs, where the deflagration phase occurs in the high-density core, producing stable IGEs.
   • The data disfavors Helium Detonation (Double Detonation) scenarios for these specific objects, as those models often predict little to no stable Ni or different velocity distributions.

3. Observational Strategy for Future Surveys:

   • The study establishes a "time-evolution" strategy for future NIR follow-up of SNe Ia discovered by LSST or the Roman Space Telescope:
     • Subluminous (86G-like): Observe NIR spectra at +50 to +100 days to catch early Ni emergence.
     • Normal-bright: Observe NIR spectra after +150 days to allow the ejecta to become optically thin enough to reveal the inner Ni.

Conclusion

This paper presents the largest homogeneous sample of late-time NIR spectra for SNe Ia to date. By successfully detecting the $[\text{Ni II}] \lambda 1.939$ $\mu$m line using a novel Gaussian Peak Ratio metric, the authors demonstrate that stable nickel is centrally located and kinematically cold. This provides compelling evidence that at least a subset of subluminous SNe Ia (86G-like) are the result of high-mass ($M_{\text{Ch}}$) Delayed Detonation explosions, offering a new pathway to disentangle the diversity of SN Ia progenitors and improve the precision of cosmological measurements.