Science Friday - Have Astrophysicists Spotted Evidence For ‘Dark Stars’?

Episode Date: October 20, 2025

Astrophysicists may have spotted evidence for “dark stars,” an unusual type of star that could possibly have existed in the earliest days of the universe, in data from the James Webb Space Telescope. Instead of being powered by nuclear fusion as current stars are, the controversial theory says that these ancient dark stars would have formed by mixing a huge cloud of hydrogen and helium with a type of self-annihilating dark matter. Dark stars would not have been dark—researchers believe that if they existed, they would actually have been bigger and brighter than current stars.Astrophysicists Katherine Freese, who first proposed the idea of dark stars in 2007, and Cosmin Ilie, who detected the possible signs of the dark stars, join Host Ira Flatow to discuss the theory.  Guests:Dr. Katherine Freese is a theoretical astrophysicist and a professor of physics at the University of Texas at Austin.Dr. Cosmin Ilie is an assistant professor of physics and astronomy at Colgate University.Transcripts for each episode are available within 1-3 days at sciencefriday.com. Subscribe to this podcast. Plus, to stay updated on all things science, sign up for Science Friday's newsletters.

Hi, I'm Irafledo, and you're listening to Science Friday. Today on the podcast, one theory about the early days of our universe and how stars then might not be like stars we have now. This dark matter annihilation takes place throughout the star, which, by the way, is really weird-looking. Its radius is 10 times the distance between the Earth and the Sun. These are cool, big, puffy beasts. Scientists think they may have discovered a new kind of star. They named it a dark star, but it actually shines very brightly. I know it sounds weird.

Considering that the star is believed to be powered by dark matter, not nuclear fusion, weirdness is par for the course in cosmology, is it not? It goes without saying this idea is controversial, but a team of astrophysicists say they have spotted evidence for the existence of these dark stars in data collected by the James Webb Space Telescope. The study was published in the proceedings of the National Academy of Sciences. Joining me now are two of the authors of that research. Dr. Catherine Fries is a theoretical astrophysicist and professor of physics

University of Texas at Austin. She first proposed the existence of dark stars in 2007, and Dr. Cosmin Ilya, an assistant professor of physics and astronomy at Colgate University. Welcome to Science Friday. Thank you. Thank you. It's great to be here. Oh, that's nice to have you. Dr. Freeze, what exactly is a dark star?

A dark star would be the very first kind of star that formed in the history of the universe when it was about 200 million years old. They're made of hydrogen and helium from the Big Bang, almost entirely of ordinary stuff, but they're powered by dark matter. The way the first stars form is inside proto-galaxies that weigh about a million times as much as the sun. and smack in the middle of these objects, there is a lot of dark matter. And it turns out that if the dark matter is made of particles that annihilate among themselves, then when two particles annihilate, they turn into something else. And the stuff that comes out interacts with the hydrogen and gets stuck inside this collapsing cloud and turns it instead into a star.

So it's shining brightly then? It starts out small, weighing about the same as the sun, but it keeps growing and growing and growing until it gets a million times as massive as the sun and a billion times as bright. And if you're a billion times as bright of the sun, yes, I would say it's shining brightly. Dr. Ilya, you say you have four candidates. What do they look like? What's the signature for them? Okay. So physically, if you look at the image, they look like red dots. Okay. They're part of this plethora of objects called red dots. Some of them are called blue monsters. They're really compact objects, very compact, meaning a few hundred parsecs or less,

which emit insane amounts of light. So if they're really galaxies, they have to have packed those stars incredibly efficiently. Like in a small amount of space, you have to have many, many stars. For this reason, our interpretation is, well, if it's actually a star, each could have the same signatures as a galaxy, and that's what we're showing both with spectroscopy in this latest study and with photometry in the 2023 study. The predictions that we made back in 2012 were how much light would come out of these things at different wavelengths. It's called a spectrum. And then we compared that to the JWST filters. And we said, yes, these things would stick out. We would be able to see them in the JWST.

And so the spectrum, how much light at different wavelengths was a specific prediction and it matches. So that is really exciting. But the third property is the one that I'm going to let Cosmine describe, which is a helium line in this spectrum. Yeah, definitely. So there's a telltale signature in the spectra of dark stars. An absorption feature, meaning, you know, if you take the spectrum, there's a deep in

in the spectrum. That means there's a lack of light compared to the light that comes in the, you know, narrowing wavelengths, a lack of an absorbent feature. But a specific wavelength, that line is expected to be there for dark stars. Why? Well, because in their atmospheres, they do have sufficient helium, but more than that, the temperatures of those objects is in such a way that that helium is singly ionized. That means it lost an electron out of the two it has. And more than that, that electron that still remains there is in a specific excited state.

So it takes a combination of factors for an object to actually have this feature as an observant feature. This helium-2-1640 line is a smoking gun for dark stars. If it's in there, then it is not an early galaxy. It is a dark star. And as we keep scanning the JWC, Look, we don't even have to do an observing survey because the data are coming in. And we can watch as it comes in and say, aha, there's a candidate.

And then we look at the details and doesn't have that dip right there or not. When you say it's powered by dark matter, Dr. Freese, what does that mean? That means the, if the dark matter is of the type that self-annualates, two dark matter particles, hit each other and turn into something else, the things that it turns into, which would be particles of light or electrons, hit the hydrogen inside the collapsing hydrogen cloud, and they get captured.

So it's kind of a heat dump into the dark star. The energy that used to be in the mass of the dark matter particles instead becomes energy that powers the dark star. And this is completely different from fusion, I mean, fusion takes place in the hot cores of stars. This dark matter annihilation takes place throughout the star, which, by the way, is really weird looking. Its radius is 10 times the distance between the earth and the sun. These are cool, big, puffy beasts.

They're very unusual. Because they're so cool, is possible for them to keep accreting matter, to keep having more mass flow onto them and getting bigger and bigger and bigger. So this dark matter power is the key to why they're. they can grow so large. If you don't know what dark matter is made of, how do you base a theory around the behavior of it? The key is there's, you're dumping heat into these early clouds

and turning them into stars. That for us is the key. We don't really care what type of dark matter will do that. We don't care the origin of the heat source. So we used as an example, annihilating dark matter, that works. We've also since then looked at other types of dark matter, self-interacting dark matter also works.

We want to generalize what we have to all kinds of different, all different kinds of dark matter. We need the heat to get into the star. Well, if these are dark stars and they exist, does it help you explain, Dr. Freeze, anything else about the universe? It's an interesting concept to imagine that maybe the way the nature of the dark matter will be discovered

by understanding the dark stars, by discovering a dark star, and then it's clear it has to be powered by dark matter, and then you can study how many dark stars there are of what mass and so forth. And based on that, you could even figure out what the mass of the dark matter particle is, or the interaction strength of the dark matter particles. So discovering these stars is a probe of the dark matter, which is 85% of the total mass content of the universe. So that's an exciting prospect.

I also want to say that make a connection to the connection between different types of physics. Back in the day, people were trying to understand what powers the sun. If the sun were just collapsing, and that's the only place that got any energy, it would have died in a million years. Well, that's wrong. The sun has already lived five billion. Along comes nuclear physics and says, well, we can give you fusion. Put that in the center of the star. Now you have a way to keep it alive.

10 billion years. And we're saying, well, okay, what about particle physics? Let's put that inside the star. What does that do? And that can power a different kind of star, a dark star, for who knows how long, millions to billions of years. So this connection between different branches of physics is pretty interesting. I want to go back to the data, to the web, and the fact that the web, in its few years of existence already is placing enormous stresses on previous theories of the formations of the first stars and galaxies. There's too many, too massive, too compact, very early galaxies, such as those blue monsters to be explained with regular astrophysics. So now we have a choice either to modify the astrophysics in ways that are strange, let's say,

or to introduce new flavors of physics, new particles in the mix, such as dark matter, and see what this dark matter does for the first stars. And if eventually, if dark stars do exist, they actually could solve some of those puzzles posed by the web data. The other causal that we haven't mentioned, which is huge, cosmin, is that once the dark stars die, because there's no more dark matter fuel, then they collapse to black holes. There's the big black hole problem of the early universe. There are these supermassive black holes that are also seen in the James Webb data and other data that are just enormous. So you've got a billion solar mass, supermass of black holes,

they're weighing a billion times as much as the sun, already very early in the history of the universe. How do you make those? We really help with that problem because supermas of dark stars that collapse to, let's say, a black hole that weighs a million times as much as the sun, that's a really good seed to merge some of those together or accrete mass onto them. And you can end up explaining where these early giant supermassive black holes came from. This is a huge area of research and people are struggling to try to explain where these things came from. Yeah. Dr. Ilya, what would you need to confirm this or firm up your data? Basically, as we mentioned before, the smoking gun signature is this 1640 absorption feature, which in one of the four objects

we analyzed in the study we're discussing today, we found albeit, Not conclusive at what a stormer is called signal-to-noise ratio. For us, it's a level 2 to 3, so that's not conclusive. If we scan more objects and we analyze more data, hopefully one of those would be irrefutable, meaning at a level of 5 or higher. And with that, then would, in my mind, have a confirmation of the existence of those objects.

Well, do you think that your research, this paper, is going to convince others in the field of this idea? you know, it's hard to determine what other people are going to do. So I never know how to answer a question. I mean, if you have the right idea and you have something new and intelligent to say, people, they pay attention. And if you have the answer, they're going to get it, you know. As theorists, we invent a lot of stuff.

We have ideas, right? Guess what? Within five minutes, they die because some observation breaks them. This is the other way around every few years. the new problem comes along that we're automatically solving. And we're like, how is this possible? There must be something to this. So for us, these are exciting times.

Well, I could go on talking about this all day because, as I say, it's one of our favorite subjects on Science Friday. But we have run out of time. So I'd like to thank both of you, Dr. Catherine Fries, Professor of Physics at the University of Texas at Austin, Dr. Cosmine Ilié, an assistant professor of physics and astronomy

at Colgate University. Thank you. This has been fun. Thank you. It was a real pleasure. Hey, thanks for listening. And if you have a comment or question or a story idea, our listener line, it's always open. Call 8774 SciFRI, 877, the number four, side fry. This episode was produced by Charles Berkwist. I'm Ira Flato. We'll see you soon.