🔭 astrophysics

Elemental Abundances from Off-center Carbon Burning in Accreting CO White Dwarfs: Implications for SN 2021yfj-like events

This study demonstrates that off-center carbon burning in accreting CO white dwarfs can generate sufficient silicon, sulfur, and argon to explain the circumstellar material observed in SN 2021yfj-like events, thereby supporting the double white dwarf merger scenario as a viable progenitor channel.

Original authors: Chengyuan Wu, Dongdong Liu, Takashi J. Moriya, Zhengwei Liu, Heran Xiong, Bo Wang

Published 2026-06-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Chengyuan Wu, Dongdong Liu, Takashi J. Moriya, Zhengwei Liu, Heran Xiong, Bo Wang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Mystery of SN 2021yfj: A Cosmic Crime Scene

Imagine a supernova (a giant exploding star) called SN 2021yfj as a crime scene. When astronomers looked at the "debris" (the light and gas) from this explosion, they found something strange. Usually, exploding stars leave behind clouds of hydrogen or helium. But this one left behind a cloud rich in Silicon, Sulfur, and Argon.

It's like walking into a room after a party and finding it filled with expensive champagne and caviar, but no pizza or soda. The ingredients don't match the usual recipe for a star explosion. The big question was: How did this star get so much Silicon and Sulfur before it blew up?

The Suspect: A Double-Acting Star System

The paper investigates a specific theory: that this explosion came from a double white dwarf system.

Think of a white dwarf as a dead star that has stopped burning its fuel, like a cooling ember. In this scenario, we have two of them (or a dead star and a helium star) dancing around each other.

The Feeder: One star is a "He-star" (a star made mostly of helium). It is losing mass, like a leaking balloon.
The Eater: The other star is a Carbon-Oxygen (CO) white dwarf. It is greedily swallowing the helium from its partner.

The Cooking Process: Off-Center Carbon Burning

As the CO white dwarf eats the helium, it doesn't just sit there. The helium piles up on its surface and gets squeezed so hard that it turns into Carbon and Oxygen. This is like adding more dough to a pizza that's already in the oven.

Eventually, the pressure gets so high that the Carbon inside the star starts burning again, but not in the center. It starts burning in a ring around the center. The authors call this "off-center carbon burning."

Think of it like a campfire that starts in a ring around the middle of a log, rather than in the very center. As this ring of fire burns inward, it cooks the material, turning simple elements into heavier, more complex ones like Silicon and Sulfur.

The Experiment: Simulating the Recipe

The researchers wanted to know if this "cooking process" could actually produce the exact amount of Silicon and Sulfur found in SN 2021yfj.

Previous attempts: Earlier scientists tried to simulate this by assuming the star ate helium at a steady, constant speed (like a machine feeding a constant stream of sand).
This paper's approach: The authors realized that in real life, stars don't eat steadily. Sometimes they gulp, sometimes they sip. They used a more realistic simulation where the "eating speed" changes over time, just like a real binary star system would behave.

They also tested different "recipes" for the star's starting ingredients. They asked: What if the star started with more Carbon? What if it started with less?

The Results: A Close Match

Here is what they found:

It Works: The simulation showed that when a white dwarf eats helium at a realistic, changing rate, the "off-center fire" does indeed cook up a lot of Silicon and Sulfur.
The Secret Ingredient: The amount of Silicon produced depends heavily on how much Carbon the star started with.
- If the star started with more Carbon, it cooked up more Silicon and Sulfur.
- If it started with less Carbon, it made less.
The Comparison: When they compared their "cooked" results to the actual debris from SN 2021yfj, it was a good match.
- Their models produced the right mix of Silicon, Sulfur, and Argon.
- There were some small differences (the model made a bit too much Magnesium and not quite enough Calcium), but the authors note that Calcium wasn't even clearly seen in the real explosion, so this isn't a deal-breaker.

The Grand Finale: The Merger

The paper suggests a dramatic ending to this story. After the white dwarf has eaten enough helium and cooked up a Silicon-rich shell, it eventually merges with its partner star.

Imagine two dancers spinning together until they crash. During this crash:

The outer layer of the Silicon-rich star gets ripped off (tidally stripped) and forms a cloud around the system.
The collision triggers a massive explosion.
The explosion hits that pre-made cloud of Silicon and Sulfur, creating the unique light and sound we see as SN 2021yfj.

The Bottom Line

The paper concludes that this "Double White Dwarf Merger" scenario is a viable suspect. It explains how a star could naturally accumulate a cloud of Silicon and Sulfur before exploding. While there are still some uncertainties (like exactly how much material gets ripped off during the crash), the "cooking recipe" of off-center carbon burning successfully produces the ingredients needed for this mysterious type of supernova.

In short: By simulating a star eating helium at a realistic, changing pace, the authors proved that stars can naturally cook up the specific "Silicon-Sulfur" soup found in SN 2021yfj, supporting the theory that these explosions come from colliding dead stars.

Technical Summary: Elemental Abundances from Off-center Carbon Burning in Accreting CO White Dwarfs

Problem Statement
The supernova SN 2021yfj is a recently discovered interacting event characterized by narrow emission lines of silicon (Si), sulfur (S), and argon (Ar). These features indicate the presence of circumstellar material (CSM) enriched with intermediate-mass elements surrounding the progenitor. While the "double white dwarf (WD) merger" scenario has been proposed as a progenitor channel—specifically involving the tidal stripping of a Si-rich WD formed from a CO WD accreting helium from a companion—the viability of this channel remains uncertain. Previous models assumed a constant mass-accretion rate, which may introduce significant uncertainties in the predicted nucleosynthetic yields. Specifically, it is unclear whether off-center carbon burning in an accreting CO WD, under realistic time-dependent accretion histories, can produce Si, S, and Ar in quantities comparable to those inferred for the CSM of SN 2021yfj.

Methodology
The authors utilized the stellar evolution code MESA (version 10398) to simulate the evolution of a $1.1 M_\odot$ CO WD accreting helium-rich material. Unlike previous studies that adopted constant accretion rates, this work employed time-dependent mass-accretion rates derived from binary evolution calculations of a $1.1 M_\odot$ CO WD + $2.3 M_\odot$ He star system (based on Moriya et al. 2026).

Key methodological components include:

Accretion Physics: The simulations incorporated the transition between unstable He-shell flashes (driving super-Eddington winds) and stable He burning (regulated by optically thick winds) when the accretion rate exceeds the critical limit ( $\dot{M}_{cr}$ ).
Initial Conditions: A fiducial model was constructed with an initial composition of 40% $^{12}$ C and 60% $^{16}$ O. To investigate the sensitivity of nucleosynthesis to initial conditions, two additional models were run with initial carbon fractions of 30% and 50%.
Nuclear Networks: The primary simulations used the "approx21.net" network. To assess the impact of nuclear reaction choices, the fiducial model was recalculated using a more extensive 57-isotope network.
Termination Criteria: Calculations were terminated when the off-center carbon flame propagated inward to a mass coordinate of $M_r \approx 0.3 M_\odot$ , representing the outer layers expected to be stripped during a merger.

Key Results

Evolutionary Track: The simulations confirmed that off-center carbon ignition occurs when the WD mass reaches $\sim 1.246 M_\odot$ . The carbon flame propagates inward, reaching temperatures sufficient ( $\log(T_{max}/K) > 9.2$ ) to synthesize intermediate-mass elements.
Nucleosynthetic Yields: The off-center burning produced significant amounts of Ne, Mg, Si, S, and Ar. The study found that the resulting elemental abundances are strongly dependent on the initial carbon abundance of the WD:
- Higher initial carbon fractions lead to more efficient carbon burning, driving the production of heavier intermediate-mass elements (Si, S, Ar) while reducing O and Mg.
- Models with higher initial carbon (50% C) produced S abundances closer to those inferred for SN 2021yfj, though they tended to overproduce Si relative to observations.
Comparison with SN 2021yfj: The average abundances in the outer $0.3 M_\odot$ $0.3 M_{⊙}$ of the simulated WDs were compared with the CSM composition inferred by Schulze et al. (2025).
- The models generally reproduced the presence of Si, S, and Ar.
- Discrepancies exist: the models tend to overproduce Mg and underproduce Ca (though Ca features are not prominent in the observed spectra) and Ar.
- The predicted Si and S abundances fall within a broad range consistent with the observational constraints, provided the initial C/O ratio and accretion history are tuned.
Robustness: The choice of nuclear reaction network (approx21.net vs. 57-isotope) had a negligible effect on the main conclusions, with only minor variations in abundance patterns (e.g., a slight increase in Ca in the 57-isotope run, but still below 0.1% mass fraction).

Significance and Claims
The paper claims that off-center carbon burning in accreting CO WDs, when modeled with realistic time-dependent accretion rates, can naturally produce the key intermediate-mass elements (Si, S, Ar) required to explain the CSM of SN 2021yfj.

The authors conclude that the double WD merger scenario remains a viable progenitor channel for SN 2021yfj-like events. While the models do not perfectly reproduce the observed abundance pattern (specifically regarding Mg overproduction and Ar underproduction), the results demonstrate that the necessary chemical enrichment is achievable within this framework. The study suggests that variations in the initial C/O ratio and accretion history could reconcile the remaining discrepancies. However, the authors explicitly note that significant uncertainties remain regarding the merger dynamics (specifically the ability to eject $\sim 0.3 M_\odot$ of material) and the explosion mechanism itself, which require further multidimensional investigation.