Webb Found an Object With a Planet’s Mass That Was Born Like a Star

Subtitle: In the young cluster IC 348, JWST confirmed brown dwarfs down to roughly twice Jupiter’s mass — including one with a warm disk of planet-forming material — blurring the line between planet and star.

Astronomy’s simplest question is also one of its messiest: where does a planet end and a star begin?

Mass alone used to feel like enough. Stars fuse hydrogen. Brown dwarfs sit below that line and never run a sustained hydrogen furnace. Giant planets assemble later, inside disks around brighter hosts. Then the James Webb Space Telescope started finding free-floating objects that weigh like planets, sit in star-forming nurseries like newborn stars, and — in at least one dramatic case — still wear a disk of gas, dust, and ice that looks ready to build something smaller.

On September 20, 2026, SpaceDaily amplified that puzzle with fresh public framing of a deeper Webb survey of IC 348, a star-forming region about 1,000 light-years away in Perseus. The peer-reviewed core is a June 2025 Astrophysical Journal Letters study by Kevin L. Luhman (Penn State) and Catarina Alves de Oliveira (ESA). NASA’s Webb mission team had already released one of the telescope’s largest IC 348 panoramas on September 15, 2026 — a glowing nursery of protostars, jets, and the faintest confirmed members of the cluster’s substellar population.

The headline object in that story is catalogued as LRL 2296: roughly two Jupiter masses, free-floating, spectroscopically confirmed as a young cluster member, and red enough at wavelengths longer than 3.5 micrometers to betray excess heat from a circumstellar disk. Its mass says planet. Its birthplace and disk say scaled-down star formation.

A nursery where labels break

IC 348 is only a few million years old. Newborn brown dwarfs still glow brightly in the infrared from residual formation heat, which is why a nearby cluster is the right place to hunt the bottom of the initial mass function — the census of how many objects of each mass star formation actually produces.

In 2022, the same team used Webb to push the spectroscopically confirmed floor in IC 348 to about 3–4 Jupiter masses. The deeper program went further. Webb’s NIRCam imaged a wider field in 2024. The team selected 39 brown-dwarf candidates from color and brightness, then used NIRSpec in 2025 to take spectra of 15 of them. Spectroscopy is the filter that separates real newborns from impostors: of those 15, nine are classified as substellar members of the cluster; four are unrelated background T dwarfs; two are active galaxies.

The faintest new members land near ~2 Jupiter masses — about 0.19% of the Sun’s mass, in NASA’s framing. That is far below the hydrogen-burning limit (~75–80 Jupiter masses) and below the rough deuterium-burning marker often used near ~13 Jupiter masses. By nuclear bookkeeping alone, these objects live in planetary-mass territory. By formation bookkeeping, they look like the extreme low-mass tail of the same cloud-collapse process that makes stars and ordinary brown dwarfs.

The disk that makes the story shareable

Two of the new members — LRL 2296 (~2 M_Jup) and LRL 11040 (~10 M_Jup) — show large infrared excesses from circumstellar disks. That makes the fainter one the least massive known brown dwarf with disk evidence. Both also show absorption features that may come from circumstellar ices: water, carbon dioxide, carbon monoxide, and OCN⁻.

Webb did not resolve rings or baby planets. The disk is inferred from colors and spectra: the object is brighter at long infrared wavelengths than atmosphere-only models allow. “Planet-forming material” means a reservoir of gas, dust, and ice is present — not that planet formation has already been caught in the act. Longer-wavelength observations would be needed for a fuller disk census; near-infrared excesses only catch part of the story and therefore set a lower limit on how common disks are among these lightweights.

Still, the implication is delicious. Disks are the natural leftovers of collapse plus angular momentum. If an independently formed, planet-mass body can keep one, then the machinery that builds moons or miniature planetary systems does not require a Sun-like central star. Vocabulary starts arguing with itself: companions born around a planet-mass primary might be called moons; companions around an independently formed brown dwarf might be called planets. The physics does not wait for the dictionary.

Formation, not just weight

A disk favors star-like birth, but it is not a unique birth certificate. A giant planet ejected from a stellar system could, in principle, drag some circumplanetary material along. A separate 2026 analysis of the spatial distribution of ultralow-mass, hydrocarbon-bearing objects in IC 348 found their positions and local stellar densities statistically indistinguishable from stars and heavier brown dwarfs — unlike N-body simulations in which liberated planets tend to scatter into wider, lower-density neighborhoods after a few million years. That tilts the ledger toward cloud fragmentation, with the usual age caveat: if the cluster is closer to five million years old, some simulated ejected populations have more time to relax and blur the contrast.

Mass estimates themselves are model-dependent. LRL 2296 has no dynamical orbit that weighs it directly. Researchers compare luminosity and assumed cluster age with evolutionary cooling tracks — tracks that are least secure below a few percent of a solar mass. The discovery paper notes that even a large systematic error would still leave the object at only a few Jupiter masses. The qualitative punchline survives: star formation can isolate astonishingly little material.

A new letter in the spectral alphabet

While chasing mass, the team also chased chemistry. Eight of the nine new brown dwarfs — plus one previously known member reobserved with NIRSpec — show an unidentified aliphatic hydrocarbon absorption near 3.4 μm, stronger in the faintest objects. Atmospheric models did not predict the feature; it had not been seen in atmospheres outside the Solar System before the IC 348 campaign. The authors propose a new spectral class “H” (for hydrocarbon), defined by that 3.4 μm fundamental band — an extension of the L/T/Y sequence for the coolest newborn brown dwarfs.

That is not a side trivia item. If the coolest atmospheres carry chemistry the models miss, the same models that convert brightness into mass need more work. Classification and weighing are entangled.

What this is — and what it is not

It is not a claim that LRL 2296 is “definitely a star,” that planets no longer exist as a category, or that Webb photographed planets forming in its disk. Masses are evolutionary-model estimates. The disk is an infrared excess, not a resolved architecture. Formation preference rests on cluster membership, spatial statistics, and disk presence — strong clues, not a reconstructed birth movie.

It is a peer-reviewed extension of the spectroscopically confirmed brown-dwarf population in IC 348 down to ~2 Jupiter masses, with at least one disk-bearing object at that floor, a proposed “H” spectral class, and a public science wave (NASA panorama September 15; SpaceDaily September 20, 2026) that makes the classification crisis impossible to ignore. Primary paper: Luhman & Alves de Oliveira, ApJL 986, L14 (2025), DOI 10.3847/2041-8213/addc55; arXiv 2506.08969.

The boundary that refuses to sit still

Sustained hydrogen fusion still draws a hard physical line for “star.” The line between giant planet and brown dwarf is softer: deuterium burning is a convenient mass marker, not a formation verdict. A two-Jupiter-mass body embedded in a young cluster, sharing the spatial habits of stars, wearing a disk, and glowing with hydrocarbon fingerprints looks like the extreme low-mass product of star formation. An object of the same mass orbiting a mature star would be called a planet without an argument.

Webb has not erased the categories. It has shown where their definitions disagree — and handed Sunday-night science a shareable image of the disagreement: a world-weight object, born like a star, possibly assembling a system of its own in the dark of Perseus.

Sources

Luhman, K. L., & Alves de Oliveira, C. (2025). A new spectral class of brown dwarfs at the bottom of the IMF in IC 348. The Astrophysical Journal Letters, 986, L14. https://doi.org/10.3847/2041-8213/addc55
Same title (preprint). arXiv:2506.08969. https://arxiv.org/abs/2506.08969
NASA Webb Mission Team. (2026, September 15). NASA’s Webb reveals dynamic panorama of star formation. https://science.nasa.gov/missions/webb/nasas-webb-reveals-dynamic-panorama-of-star-formation/
Deep Field / SpaceDaily. (2026, September 20). Where does a planet end and a star begin? Webb found a free-floating brown dwarf only about twice Jupiter’s mass, surrounded by a disk of planet-forming material. https://spacedaily.com/t-webb-ic348-two-jupiter-mass-brown-dwarf-disk/