Artistic illustration of a stripped-envelope supernova glowing in infrared, with carbon monoxide molecules and dust in its expanding ejecta.

A Stripped Supernova Made Carbon Monoxide—But Did It Forge Heavy Elements?

JWST's first near- plus mid-infrared (NIR+MIR) look at a broad-lined Type Ic supernova finds the earliest molecular detection in a core-collapse explosion and an infrared continuum consistent with heated dust, but no clear r-process signature.

Jun 26, 2026

On 22 November 2024, the Asteroid Terrestrial-Impact Last Alert System spotted a new point of light in the galaxy NGC 681. The explosion, named SN 2024abup, lies about 23.3 megaparsecs away—roughly 76 million light-years. Astronomers soon classified it as a broad-lined Type Ic supernova: a massive star that had already lost its hydrogen and helium layers before collapsing.

These explosions are rare, making up only about 1 percent of core-collapse supernovae. Their ejecta race outward at tens of thousands of kilometers per second, far faster than ordinary supernovae. That speed and stripped-down structure have made them suspects in two cosmic mysteries: the production of heavy elements through the rapid neutron-capture process, or r-process, and the creation of dust in the early universe.

A rare explosion stripped to its core

For r-process elements, the confirmed factory is merging neutron stars. But their merger rate may be too low to explain the abundances seen in ancient stars. Other sites may be needed. Broad-lined Type Ic supernovae, also called SNe Ic-bl, are plausible candidates because their cores may collapse into a black hole with a surrounding accretion disk, or into a rapidly spinning, strongly magnetized neutron star. Both settings could fling out neutron-rich matter where heavy elements form.

Dust is the other puzzle. Some young galaxies at high redshift already contain dust, but they are too young for low-mass aging stars to have produced it. Supernovae could provide a faster source. Previous searches had found carbon monoxide, a dust precursor, in some Type Ic supernovae, but never in the broad-lined variety.

JWST opened a 1-to-14-micron window

At 41 days after the supernova's visible-light peak, and about 54 days after explosion, the team pointed JWST's near-infrared NIRSpec and mid-infrared MIRI instruments at SN 2024abup. Together they covered the 1-to-14-micron range, infrared wavelengths that carry information about molecules, dust, and cooler ejecta. Supporting data included an optical spectrograph from Las Cumbres Observatory, Swift's ultraviolet telescope, and the Very Large Array radio observatory.

The resulting spectrum is the first JWST near- plus mid-infrared (NIR+MIR) observation of a broad-lined Type Ic supernova. It also combines optical and infrared spectra from nearly the same time with radio observations at later epochs, making a rare multi-wavelength (panchromatic) view of the same event.

A molecule appears at 4.6 microns

The spectrum showed a broad feature around 4.6 microns, the fingerprint of carbon monoxide. This is the first reported detection of CO in a broad-lined Type Ic supernova and the earliest molecular detection in any core-collapse supernova so far.

The team estimates roughly 2 times 10^-5 solar masses of CO. That is early: some models had predicted molecules like CO would form between 100 and 600 days after explosion. CO matters because it radiates heat away, cooling the expanding ejecta so that dust grains can condense. Finding it at this phase suggests that the conditions for dust formation were already in place.

Dust may already be glowing

Beyond 1.5 microns, the spectrum also showed more continuum emission than the model predicted. A qualitative fit suggests heated dust could explain the excess: silicate grains at about 700 kelvin and 0.05 solar masses, plus graphite grains at about 300 kelvin and 0.009 solar masses.

But the source of that dust is not settled. SN 2024abup sits in a thick dust lane in NGC 681, so some of the glow could come from preexisting host-galaxy dust heated by the supernova. Some could be newly formed dust in the ejecta, or the signal could be a mix. The team tried to subtract the host galaxy carefully, but residual contamination remains possible.

Even with that caveat, the result points to a broader idea: broad-lined Type Ic supernovae could help supply dust to young galaxies. They explode quickly and can enrich their surroundings before older stars have had time to evolve.

The heavy-element signal is still buried

The r-process search proved less conclusive. The team compared the JWST spectrum with theoretical predictions for infrared lines from heavy elements. Some predicted lines from rubidium, palladium, bromine, selenium, and tin fell near observed features.

But a spectral synthesis model called SUMO, which contained no r-process elements, could reproduce most of those features using lighter elements such as carbon, oxygen, and magnesium. A bump near 2.1 microns—seen as a possible telltale of tellurium in the neutron-star merger AT 2017gfo—is better explained in SN 2024abup by magnesium. The team therefore reports no compelling evidence for an r-process signature.

The absence is not proof. The spectrum was taken while the ejecta were not fully transparent, and many broad lines overlap. Under simplifying assumptions, the observations are sensitive to heavy-element masses near 10^-4 solar masses. For tin, the limit is about 5 times 10^-5 solar masses if the atoms were mostly neutral and the line optically thin. For other elements, optical-depth uncertainties weaken the constraints. Later nebular-phase infrared spectra, when the ejecta are more transparent, would give a cleaner test.

A radio silence with meaning

The Very Large Array observed SN 2024abup at 84 and 211 days after explosion and detected nothing. Standard models for a relativistic jet viewed off-axis would have produced visible radio emission. The non-detections suggest that any jet was weaker than typical gamma-ray-burst jets, or that it plowed into a low-density environment. That fits a progenitor that had little pre-explosion mass loss.

One spectrum, two answers

SN 2024abup is a single event, and one early infrared spectrum cannot settle the r-process question for its entire class. The work, reported in a preprint and not yet peer-reviewed, also relies on models that are not fine-tuned to this supernova. But the observation does show how much JWST can extract from one stripped, fast-moving explosion.

It found a molecule that had never been seen in this kind of supernova, and it found infrared light consistent with dust grains already warming the ejecta. It did not find heavy r-process elements—at least not yet. The next step is to watch similar explosions later, when the ejecta thin out and their spectral lines separate. For now, SN 2024abup leaves behind a molecular whisper, a possible dust glow, and a heavy-element mystery.

JWST observations of SN 2024abup: First Detection of CO in a broad-lined Type Ic Supernova and Constraints on r-process NucleosynthesisManisha Shrestha, Lindsey A. Kwok, David J. Sand, Stan Bartmentloo, Collin Christy, Anders Jerkstrand, K. Azalee Bostroem, Jennifer E. Andrews, Kate D. Alexander, Yize Dong, Carl E. Fields, Emily Hoang, Griffin Hosseinzadeh, Brian Hsu, Daryl Janzen, Saurabh W. Jha, Joel Johansson, Jeniveve Pearson, M. J. Lundquist, Darshana Mehta, Aidan Martas, Maryam Modjaz, Bernhard Müller, Conor L. Ransome, Aravind P. Ravi, Mathieu Renzo, Nicolás Meza Retamal, Bhagya Subrayan, Nathan Smith, Stefano Valenti, Sergiy Vasylyev, Giacomo Ricigliano, Peter J. Brown, Moira Andrews, Joseph Farah, D. Andrew Howell, Curtis McCully, Megan Newsome, Kathryn Wynn, Ryan Chornock, Natalie LeBaro, Raffaella Margutti, Melissa Shahbandeh, Chris Ashall, Peter Hoeflichhttps://arxiv.org/abs/2606.28561v1