SPACE/COSMOS
Mysterious ‘red dots’ in early universe may be ‘black hole star’ atmospheres
The objects that astronomers at Penn State dubbed ‘universe breakers’ could be an exotic black hole atmosphere, representing a missing link in the rapid growth of supermassive black holes
Penn State
image:
Artist’s impression of a black hole star (not to scale). Mysterious tiny pinpoints of light discovered at the dawn of the universe may be giant spheres of hot gas that are so dense they look like the atmospheres of typical nuclear fusion-powered stars; however, instead of fusion, they are powered by supermassive black holes in their center that rapidly pull in matter, converting it into energy and giving off light.
view moreCredit: T. Müller/A. de Graaff/Max Planck Institute for Astronomy
UNIVERSITY PARK, Pa. — Tiny red objects spotted by NASA's James Webb Space Telescope (JWST) are offering scientists new insights into the origins of galaxies in the universe — and may represent an entirely new class of celestial object: a black hole swallowing massive amounts of matter and spitting out light.
Using the first datasets released by the telescope in 2022, an international team of scientists including Penn State researchers discovered mysterious “little red dots.” The researchers suggested the objects may be galaxies that were as mature as our current Milky Way, which is roughly 13.6 billion years old, just 500 to 700 million years after the Big Bang.
Informally dubbed “universe breakers” by the team, the objects were originally thought to be galaxies far older than anyone expected in the infant universe — calling into question what scientists previously understood about galaxy formation.
Now, in a paper published today (Sept. 12) in the journal Astronomy & Astrophysics, the international team of astronomers and physicists, including those at Penn State, suggest that the dots may not be galaxies but an entirely new type of object: a black hole star.
They said their analysis indicates that the tiny pinpoints of light may be giant spheres of hot gas that are so dense they look like the atmospheres of typical nuclear fusion-powered stars; however, instead of fusion, they are powered by supermassive black holes in their center that rapidly pull in matter, converting it into energy and giving off light.
“Basically, we looked at enough red dots until we saw one that had so much atmosphere that it couldn't be explained as typical stars we’d expect from a galaxy,” said Joel Leja, the Dr. Keiko Miwa Ross Mid-Career Associate Professor of Astrophysics at Penn State and co-author on the paper. “It’s an elegant answer really, because we thought it was a tiny galaxy full of many separate cold stars, but it’s actually, effectively, one gigantic, very cold star.”
Cold stars emit little light due to their low temperatures compared to normal stars, Leja explained. Most stars in the universe are low-mass, colder stars, but they are typically harder to see as they are washed out by rarer, more luminous massive stars. Astronomers identify cold stars by their glow, which is primarily in the red optical or near-infrared spectrum, wavelengths of light that are no longer visible. While the gas around supermassive black holes is typically very hot, millions of degrees Celsius, the light from these “red dot” black holes was instead dominated by very cold gas, the researchers said, similar to the atmospheres of low-mass, cold stars, based on the wavelengths of light they were giving off.
The most powerful telescope in space, JWST was designed to see the genesis of the cosmos with infrared-sensing instruments capable of detecting light that was emitted by the most ancient stars and galaxies. Essentially, the telescope allows scientists to see back in time roughly 13.5 billion years, near the beginning of the universe as we know it, Leja explained.
From the moment the telescope turned on, researchers around the world began to spot “little red dots,” objects that appeared far more massive than galaxy models predicted. At first, Leja said, he and his colleagues thought the objects were mature galaxies, which tend to get redder as the stars within them age. But the objects were too bright to be explained — the stars would need to be packed in the galaxies with impossible density.
“The night sky of such a galaxy would be dazzlingly bright,” said Bingjie Wang, now a NASA Hubble Fellow at Princeton University who worked on the paper as a postdoctoral researcher at Penn State. “If this interpretation holds, it implies that stars formed through extraordinary processes that have never been observed before.”
To better understand the mystery, the researchers needed spectra, a type of data that could provide information about how much light the objects emitted at different wavelengths. Between January and December 2024, the astronomers used nearly 60 hours of Webb time to obtain spectra from a total of 4,500 distant galaxies. It is one of the largest spectroscopic datasets yet obtained with the telescope.
In July 2024, the team spotted an object with a spectrum that indicated a huge amount of mass, making it the most extreme case of such an early and large object. The astronomers nicknamed the object in question “The Cliff,” flagging it as the most promising test case to investigate just what those “little red dots” were.
“The extreme properties of The Cliff forced us to go back to the drawing board, and come up with entirely new models,” said Anna de Graaff, a researcher for the Max Planck Institute for Astronomy and corresponding author on the paper, in a Max Planck Institute press release.
The object was so distant that its light took roughly 11.9 billion years to reach Earth. The spectra analysis of that light indicated it was actually a supermassive black hole, pulling in its surroundings at such a rate that it cocooned itself in a fiery ball of hydrogen gas. The light that Leja and his colleagues spotted was coming not from thick clusters of stars, but from one giant object.
Black holes are at the center of most galaxies, Leja explained. In some cases, those black holes are millions or even billions of times more massive than our solar system’s sun, pulling in nearby matter with such strength that it converts to energy and shines.
“No one's ever really known why or where these gigantic black holes at the center of galaxies come from,” said Leja, who is also affiliated with Penn State's Institute for Computational and Data Sciences. “These black hole stars might be the first phase of formation for the black holes that we see in galaxies today — supermassive black holes in their little infancy stage.”
He added that JWST has already found signs of high-mass black holes in the early universe. These new black hole star objects, which are essentially turbocharged mass-builders, could help explain the early evolution of the universe — and may be a welcome addition to current models. The team is planning future work to test this hypothesis by examining the density of gas and strength of these early black hole stars, Leja said.
Of course, the mysterious “little red dots” are great distance away in both time and space — and their small size makes it especially challenging to get a clear picture.
“This is the best idea we have and really the first one that fits nearly all of the data, so now we need to flesh it out more,” Leja said. “It's okay to be wrong. The universe is much weirder than we can imagine and all we can do is follow its clues. There are still big surprises out there for us.”
A full list of authors is available in the paper. The Penn State aspects of this work were funded by NASA.
Journal
Astronomy and Astrophysics
Method of Research
Imaging analysis
Subject of Research
Not applicable
Article Title
A remarkable ruby: Absorption in dense gas, rather than evolved stars, drives the extreme Balmer break of a little red dot at z = 3.5
Article Publication Date
12-Sep-2025
“Black Hole Stars” could solve JWST riddle of overly massive early galaxies
Max Planck Institute for Astronomy
image:
Artist’s impression of a black hole star (not to scale). The cut-out reveals the central black hole with its surrounding accretion disk. What makes this a black hole star is the surrounding envelope of turbulent gas. This configuration can explain what astronomers observe in the object they are calling “The Cliff.”
view moreCredit: MPIA/HdA/T. Müller/A. de Graaff
In the summer of 2022, less than a full month after the James Webb Space Telescope (JWST) had begun to produce its first scientific images, astronomers noticed something unexpected: little red dots. In pictures taken at JWST’s unprecedented sensitivity, these extremely compact, very red celestial objects showed very clearly on the sky and there appeared to be a considerable number of them. JWST had apparently discovered a whole new population of astronomical objects, which had eluded the Hubble Space Telescope. That latter part is unsurprising. “Very red” is astronomy lingo for objects that emit light predominantly at longer wavelengths. The little red dots emit light predominantly at wavelengths beyond a 10 millionth of a meter, in the mid-infrared. Hubble cannot observe at wavelengths this long. JWST, on the other hand, is designed to cover this range.
Additional data showed that these objects were far away indeed. Even the closest examples were so far away that their light had taken 12 billion years to reach us. Astronomers always look into the past, and we see an object whose light takes 12 billion years to reach us as it was those 12 billion years ago, a mere 1.8 billion years after the Big Bang.
Unexplainable young, massive galaxies?
This is where things get dicey. In order to interpret astronomical observations, you need a model of the object in question. When astronomers point to their data and say, “This is a star,” the statement comes with a lot of baggage. It is trustworthy only because astronomers have robust physical models of what a star is – in short, a giant plasma ball held together by its own gravity, producing energy by nuclear fusion in its center. You also need a good understanding of how stars look both in images and in the rainbow-like decomposition of light known as a spectrum. In turn, if you see an object with the right kind of appearance and the right kind of spectrum, you can confidently state that it is a star.
The little red dots did not seem to fit into any of the usual slots, so astronomers set out to look beyond the standard objects. One of the first interpretations offered was a bombshell in and of itself: In this interpretation, little red dots were galaxies that were extremely rich in stars, their light reddened by huge amounts of surrounding dust. Within our own cosmic neighbourhood, if you put our solar system in a cube one light-year a side, that cube would only contain a single star: our Sun. In the star-rich galaxies postulated to explain little red dots, a cube that size would contain several hundred thousand stars.
In our home galaxy, the Milky Way, the only region that dense in stars is the central nucleus, but that contains only about one thousandth of the stars needed in those little-red-dot models. The sheer number of stars involved, as high as hundreds of billions of solar masses’ worth less than a billion years after the Big Bang, raised major questions about astronomers’ basic understanding of galaxy evolution: Could we even explain how these galaxies produced so many stars, so quickly? Co-author Bingjie Wang (Penn State University) explains: “The night sky of such a galaxy would be dazzlingly bright. If this interpretation holds, it implies that stars formed through extraordinary processes which have never been observed before.“
Galaxies vs. active galactic nuclei
The interpretation itself remained controversial. The community split into two camps: One group that favored the many-stars-plus-dust interpretation, and another that interpreted little red dots as active galactic nuclei, but also obscured by copious dust. Active galactic nuclei are what we see when a steady stream of matter falls onto a galaxy’s central black hole, forming an exceedingly hot, so-called accretion disk around the central object. But this second interpretation came with its own set of limitations. There are marked differences between the spectra of little red dots and those of the dust-reddened active galactic nuclei astronomers had previously observed. In addition, this interpretation would require extremely large masses for the supermassive black holes at the center of those objects – and surprisingly many of those, given the large number of little red dots that had been found.
There was a consensus, too: that in order to resolve the puzzle, astronomers would need more and different observational data. The original JWST observations had provided images. For testing physical interpretations, astronomers need spectra: detailed information about how much light an object emits at different wavelengths. For the top telescopes, there is considerable competition for observing time. Once it became clear just how interesting little red dots were, numerous astronomers worldwide began to apply for time to observe them more closely. One such application was the RUBIES program formulated by Anna de Graaff at the Max Planck Institute for Astronomy in Heidelberg and an international team of colleagues, where the acronym stands for “Red Unknowns: Bright Infrared Extragalactic Survey.”
The distant treasures of RUBIES
The RUBIES application was successful, and between January and December 2024, the astronomers used nearly 60 hours of JWST time to obtain spectra from a total of 4500 distant galaxies, one of the largest spectroscopic data sets obtained with JWST to date. As Raphael Hviding (MPIA) says, “In that data set, we found 35 little red dots. Most of them had already been found using publicly available JWST images. But the ones that were new turned out to be the most extreme and fascinating object.” Most interesting of all was the spectrum for an object the astronomers found in July 2024. The astronomers dubbed the object in question “The Cliff,” and it seemed to be an extreme version of the population of little red dots – and by that very fact a promising test case for interpretations of just what little red dots were. The Cliff is so distant from us that its light took 11.9 billion years to reach us (redshift z=3.55).
“The Cliff” gets its name from the most prominent feature of its spectrum: a steep rise in what would be the ultraviolet region, at wavelengths just a little shorter than that of violet visible light. “Would” because our universe is expanding: A direct consequence is that, for an object as distant as The Cliff, that wavelength is stretched to almost five times its original value, landing squarely in the near-infrared (“cosmological redshift”). A prominent rise of this kind, at these wavelengths, is known as a “Balmer break.” Balmer breaks can be found in the spectra of ordinary galaxies, where they are usually seen in galaxies that form little to no new stars at the time. But in those cases, the rise is much less steep than The Cliff.
A curious similarity to single stars
With this unmissable, unusual feature, The Cliff looked like it did not fit any of the interpretations that had been proposed for little red dots. But De Graaff and her colleagues wanted to make sure. They constructed diverse variations of all the models that tried to cast little red dots either as massive star-forming galaxies or as dust-shrouded active galactic nuclei, attempted to reproduce the spectrum of The Cliff with each one, and failed every single time.
Anna de Graaff says: „The extreme properties of The Cliff forced us to go back to the drawing board, and come up with entirely new models.“ By that time, the idea that Balmer-break features in a spectrum might be due to something other than stars had entered the discussion (in the shape of a September 2024 article by two researchers based in China and the UK). De Graaff and her colleagues had started to wonder about something very similar themselves: Balmer breaks can be found both in the spectra of single, very hot, young stars and in the spectra of galaxies containing a sufficient number of such very hot, young stars. Weirdly, The Cliff looked more like the spectrum of a single star than that of a hole galaxy.
Enter black hole stars
On this basis, de Graaff and her colleagues developed a model some of them have taken to calling a “black hole star,” written as BH*: An active galactic nucleus, that is, a supermassive black hole with an accretion disk, but surrounded and reddened not by dust, but by virtue of being embedded in a thick envelope of hydrogen gas. The BH* is not a star in the strict sense, since there is no nuclear fusion reactor in its center. In addition, the gas in the envelope is swirling much more violently (there is much stronger turbulence) than in any ordinary stellar atmosphere. But the basic physics is similar: The active galactic nucleus heats the surrounding gas envelope, just like the nuclear-fusion-driven center of a star heats the star’s outer layers, so the external appearance has marked similarities.
The models formulated by de Graaff and colleagues at this point are proofs-of-concept – pioneering work, but not by any measure a perfect fit. Still, these black hole star models describe the data much better than any other type of model. In particular, the shape of the name-giving cliff in the spectrum is nicely explained by assuming a turbulent, dense, spherical gas envelope around an AGN. From that perspective, The Cliff would be an extreme example where the central black hole star dominates the object’s brightness. For the other little red dots, their light would be a more even mixture of the central black hole star with the light from stars and gas in the surrounding parts of the galaxy.
A new mechanism for rapid early galaxy formation?
If a black hole star is indeed the solution, it might have another potential advantage. Systems of this kind had previously been studied in a purely theoretical setting, with much lighter intermediate-mass black holes. There, the setup with central black hole and surrounding gas envelope was seen as a way for the mass of a very early galaxies’ central black holes growing particularly quickly. Given that JWST has found solid evidence for high-mass black holes in the early universe, a configuration that could explain ultra-fast mass growth of black holes would be a welcome addition to current galaxy evolution models. Whether the supermassive black hole stars can do the same is still undetermined, but it would be an intriguing expansion of their role if they did!
As promising as this sounds, caveats are in order. The new result is brand-new. Reporting on it conforms with accepted practice of covering scientific results once they are published in, or at least accepted by, a peer-reviewed journal. But in order to know whether this becomes a trusted part of astronomy’s view of the universe, we will need to wait at least a few more years.
Open questions
The present result does represent a major step forward: the first model that can explain the unusual shape of The Cliff, the extreme object’s Balmer break. Like any significant step forward, it leads to new, open research questions: How could such a black hole star have formed? How can the unusual gas envelope be sustained over a longer time? (Since the black hole gobbles up surrounding gas, there needs to be a mechanism for “refueling” the envelope.) How do the other features of the spectrum of The Cliff come about?
Answering those questions requires contributions from astrophysical modeling, but it is also set to benefit from further in-depth observation. In fact, de Graaff and her team already have the approval of JWST follow-up observations for little red dots of particular interest, such as The Cliff, scheduled for next year.
These future observations will shed light on whether black hole stars are indeed the explanation for how today’s galaxies came to be what they are. At this point in time, that outcome is an intriguing possibility, but far from certain.
Background information
The results described here have been accepted for publication as A. de Graaff et al., “A remarkable Ruby: Absorption in dense gas, rather than evolved stars, drives the extreme Balmer break of a Little Red Dot at z = 3.5” in the journal Astronomy & Astrophysics. The paper led by Raphael Hviding that presents the full sample of Little Red Dots in the RUBIES data set has been accepted for publication in the same journal,
The MPIA researchers involved are Anna de Graaff, Hans-Walter Rix and Raphael E. Hviding, in collaboration with Gabe Brammer (Cosmic Dawn Center), Jenny Greene (Princeton University), Ivo Labbe (Swinburne University), Rohan Naidu (MIT), Bingjie Wang (Penn State University and Princeton University), and others.
Journal
Astronomy and Astrophysics
Method of Research
Observational study
Subject of Research
Not applicable
Article Title
A remarkable ruby: Absorption in dense gas, rather than evolved stars, drives the extreme Balmer break of a little red dot at z = 3.5
Article Publication Date
10-Sep-2025
