Supernovae (SNe) provide crucial information about stellar evolution and nucleosynthesis through their late-time (nebular) spectra. In the nebular phase (≳100 days post-maximum light), the expanding, less opaque ejecta allow observation of forbidden emission lines, revealing the composition, geometry, kinematics, and ionization state. Type Ia supernovae (SNe Ia), the thermonuclear explosions of white dwarfs (WDs), are particularly important for cosmology, but their progenitor systems and explosion mechanisms remain unclear. While most SNe Ia are relatively uniform, a growing number of peculiar subtypes, such as "03fg-like" (or "super-Chandrasekhar") SNe Ia, exhibit unique features that offer insights into their origins. These peculiar SNe Ia are characterized by high luminosities, broad light curves, and low-ionization nebular spectra. Explanations for these SNe include single WDs with additional support from rotation, mergers of two WDs with a total mass exceeding the Chandrasekhar limit, and core-degenerate scenarios involving a near-Chandrasekhar mass WD exploding within an AGB star's envelope. Violent mergers of two sub-Chandrasekhar mass WDs are another proposed explanation, where dynamical interaction disrupts the secondary WD and detonates the primary, leading to characteristic features like enhanced IME production, lower IGE levels, centrally located IMEs, and a low late-time ionization state. JWST observations, particularly in the near-infrared (NIR) and mid-infrared (MIR), are crucial for accessing emission lines from IMEs and distinguishing between progenitor scenarios. This study focuses on SN 2022pul, a peculiar 03fg-like SN Ia, using combined ground-based and JWST data to investigate its origin and explore the violent merger hypothesis.

The paper reviews various models proposed to explain the origins of Type Ia supernovae (SNe Ia), particularly the peculiar 03fg-like or super-Chandrasekhar events. These models include single-degenerate scenarios involving a near-Chandrasekhar mass white dwarf (WD) exploding within a common envelope with an AGB star; double-degenerate scenarios involving the merger of two WDs, potentially leading to a violent merger; and a range of variations within these scenarios. The authors highlight the challenges in distinguishing between these models using observational data and emphasize the importance of detailed nebular spectroscopy to uncover crucial clues. Existing models are discussed, including violent merger models that are shown to predict characteristics that are consistent with the peculiar 03fg-like supernovae, such as enhanced production of intermediate-mass elements (IMEs) compared to iron-group elements (IGEs). The limitations of previous studies and the need for more comprehensive spectral coverage, particularly in the infrared, are discussed as motivation for this research.

The research employed a combined approach using both ground-based and JWST observations of SN 2022pul. The spectral data spans a continuous wavelength range from 0.4 to 14 µm, including the first mid-infrared spectrum of an 03fg-like SN Ia. Data reduction involved subtracting a blackbody component from the 2.5–14 µm region to account for the dust continuum contribution. The analysis focused on the dust-subtracted spectral lines. Emission line profiles were fit throughout the optical, NIR and MIR using a combination of Gaussian and asymmetric non-Gaussian line profiles. The relative line strengths were initially constrained by an existing violent merger model (MERGER model) computed by Blondin et al. (2023), allowing the relative amplitudes between ions to vary as fitting parameters. Asymmetric profiles were modeled to account for asymmetries in the ejecta. The study included a direct comparison of SN 2022pul's spectral properties with those of SN 2021aefx, a spectroscopically regular SN Ia, to highlight distinctive features. The analysis involved detailed fitting of numerous spectral lines, allowing the determination of the distribution and composition of various elements within the ejecta (IGEs, IMEs, O, and Ne). The initial 56Ni mass was estimated from the integrated spectral flux, taking into account the energy deposition from 56Co decay and the fraction of decay energy absorbed by the ejecta. The observed spectral features were then compared to predictions from different SN Ia models (including violent merger, delayed detonation, and others) to evaluate the best-fitting model.

The combined ground-based and JWST observations of SN 2022pul in the nebular phase revealed several key findings that deviate significantly from typical SN Ia characteristics. These include: 1. **Low Mean Ionization State:** SN 2022pul shows a significantly lower mean ionization state compared to normal SNe Ia like SN 2021aefx, indicative of a denser ejecta. 2. **Asymmetric Emission Line Profiles:** The emission lines exhibit strong asymmetries, with IMEs (S, Ar, Ca) showing a blueshifted peak and IGEs (Fe, Co, Ni) a redshifted peak, suggesting large-scale asymmetry in the ejecta distribution. 3. **Strong IME Emission:** The strength of IME emission lines ([Ar II] 6.98 µm, [Ar III] 8.99 µm, [Ca II] λλ7291, 7324, [S IV] 10.51 µm) is remarkably higher than in SN 2021aefx, exceeding that of IGEs. 4. **Weak IGE Emission:** The intensity of IGE emission lines is relatively weak in comparison to the IMEs and consistent with a low ionization state, suggesting less overall Fe production than in typical SN Ia. 5. **First Unambiguous Detection of Neon:** The spectrum shows the first clear detection of [Ne II] 12.81 µm in a SN Ia, with a strong, broad, and centrally peaked profile. 6. **Narrow Oxygen Emission:** The [O I] λλ6300, 6364 feature is unexpectedly narrow, and centrally peaked, indicating oxygen in the inner ejecta region. 7. **Nickel-56 Mass Estimation:** The estimated 56Ni mass of 0.66 ± 0.17 M⊙ is comparable to that of normal SNe Ia, but the unusual distribution of elements suggests a different origin. 8. **Modified Merger Model:** A modified violent WD-WD merger model that includes ejecta clumping and additional mass in the innermost region (≲2000 km s−1) provides a better match to the observed spectral features than other existing SN Ia models. This modification particularly improves the match to [O I] and the overall IGE vs. IME ratios. The distinctive spectral characteristics of SN 2022pul strongly support the violent merger scenario for this SN Ia subclass.

The findings of this study strongly favor the violent merger model for SN 2022pul. The presence of centrally located oxygen and neon, combined with the strong IME emission and the highly asymmetric line profiles, provides compelling evidence that aligns well with the predictions of violent WD-WD merger simulations. The modified merger model, which includes ejecta clumping and additional mass in the inner region, offers a better fit to the observed spectrum, improving the agreement particularly in the optical iron emission and the narrow oxygen emission. The analysis directly contrasts SN 2022pul's characteristics with those of a spectroscopically normal SN Ia, highlighting the uniqueness of this event. The violent merger model accounts for the large-scale ejecta asymmetries observed between the IMEs and IGEs, the central location of the narrow oxygen and broad neon features, and the overall low ionization state. This research significantly strengthens the case for violent WD-WD mergers as a significant channel for producing a subset of peculiar SNe Ia.

This paper provides strong evidence supporting the violent merger scenario for SN 2022pul, a peculiar 03fg-like Type Ia supernova. The unique spectral features observed, including the first unambiguous detection of neon in a SN Ia and the highly asymmetric distribution of elements, align well with the predictions of violent WD-WD merger models. Modifications to existing models, incorporating ejecta clumping and additional central mass, further enhance the agreement. Future research should focus on expanding the parameter space of violent merger simulations, exploring the diversity of possible outcomes, and obtaining higher-resolution spectroscopic data to constrain the model parameters more precisely. Spectropolarimetry could provide additional insights into the three-dimensional structure of the ejecta.

The analysis relies on a number of simplifying assumptions, including axial symmetry in the 2D projection of the ejecta and the accuracy of existing theoretical models. The exact physical processes leading to clumping are not fully understood and may affect the interpretation of certain line profiles. While the modified merger model provides a good fit, further refinement might be needed to perfectly match all observed features. Additionally, the connection between the violent merger model and the inferred presence of a dense C/O rich circumstellar medium (CSM) requires further investigation.