Can Ancient Black Holes Tell Us What Dark Matter Is?

Somewhere in the first billion years after the Big Bang, black holes already weighing millions of Suns were pulling in gas and shining. The James Webb Space Telescope has found them. How did they grow so large so quickly? A preprint by John Ellis, Malcolm Fairbairn, Juan Urrutia, and Ville Vaskonen argues that part of the answer may lie in what dark matter is—and that these ancient black holes can act as dark matter detectors.

The high-redshift black holes JWST found

The puzzle is not just about black holes. In the standard cosmological model, ΛCDM, dark matter is cold and slow. It clumps into halos, from tiny ones to giant galaxy clusters. The smallest halos are the first cradles for stars and black holes. If dark matter is not cold—if it is fuzzy or warm—those small halos are suppressed. That suppression could make it harder to build the supermassive black holes JWST sees at high redshift.

Fuzzy dark matter is a hypothetical ultralight particle, so light that its quantum wavelength stretches across small cosmic scales. Warm dark matter is made of particles moving fast enough to smooth out the smallest structures. Both would suppress the formation of some of the low-mass halos that cold dark matter predicts.

Why the smallest dark matter halos matter

That matters because stars live fast and die young. The bright ultraviolet light JWST uses to map early galaxies comes mostly from short-lived massive stars, so it gives a snapshot. Supermassive black holes are different. They can live practically forever, preserving clues about the halos where they were seeded and the mergers that fed them. As the authors put it, black holes may be better probes of dark matter than stars.

A model that follows black holes from redshift 20 to today

The team built a semi-analytical model, not a full computer simulation. It tracks the average stellar mass and central black hole mass in dark matter halos as they grow through mergers and star formation, with black holes also growing by accretion, from about redshift 20 to today. Star formation is calibrated to Hubble and JWST ultraviolet luminosity data. Black hole growth includes gas accretion, feedback from supernovae and active galactic nuclei, and the probability that black holes merge when their host halos merge.

Then they compared the model's predictions for the relation between galaxy stellar mass and black hole mass with observations at several cosmic epochs. Those observations include JWST measurements of high-redshift black holes and their host galaxies, plus earlier data closer to home. They used Markov-chain Monte Carlo fitting—a statistical method that explores which parameter combinations fit the data—to find the allowed ranges and test different dark matter models.

The limits: fuzzy dark matter below 2.0 × 10⁻²⁰ eV, warm dark matter below 7.2 keV

The headline result is a pair of lower limits. In fuzzy dark matter, particle masses below 2.0 × 10⁻²⁰ eV are disfavored at 95% confidence. In warm dark matter, particle masses below 7.2 keV are disfavored at the same confidence level. 'Disfavored' is not 'ruled out'; it means the data prefer heavier, less fuzzy particles if the rest of the model is right. For fuzzy dark matter, the bound is about as strong as existing limits from the Lyman-alpha forest, a web of hydrogen absorption lines that traces small-scale structure. For warm dark matter, it is slightly stronger than a combined limit from strong gravitational lensing, the Lyman-alpha forest, and Milky Way satellites.

Cold dark matter may not need a rescue

A second result is that the JWST black holes may be less shocking than they first appeared. In cold dark matter, the model can reproduce the observed stellar mass–black hole mass relation across redshift. The best-fit parameters fall within ranges expected from simulations and physical arguments. Light seeds—black holes of roughly 100 to 1,000 solar masses from early stars—need to form in very light halos and merge efficiently. Heavy seeds—about 100,000 solar masses—prefer a lower merger efficiency. The fit also favors accretion at or above the Eddington limit, the maximum rate at which radiation pressure normally allows a black hole to pull in gas, though it does not require it.

The bound hinges on how black holes were seeded

The dark matter bounds depend on which seeding scenario is true. If light seeds are the correct path, the constraints on fuzzy and warm dark matter become much tighter. That is because light seeds rely more on the mergers of small halos, precisely the structures that fuzzy and warm dark matter suppress. Heavy seeds are less sensitive than light seeds.

What could weaken the case

There are important caveats. The work is a preprint and has not been peer-reviewed. It uses a semi-analytical model, so it simplifies complex processes such as star formation, feedback, black hole accretion, and mergers. For the fuzzy and warm dark matter cases, the authors changed only how dark matter affects structure growth, not the black hole accretion model. The black hole mass measurements themselves carry uncertainties, and debates continue about whether some high-redshift black holes are accreting so rapidly that their masses are overestimated. The bounds also rely on the model's predicted stellar mass–black hole mass relation and on comparing it with observations across cosmic time.

Gravitational waves as the next referee

Still, the paper points to a striking possibility: the longest-lived objects in the universe may be the best recorders of dark matter's smallest effects. Future gravitational-wave detectors such as LISA and AION could help test the seeding scenarios; the authors plan to explore this connection in future work. If they do, the question of how the first supermassive black holes grew could also become a sharper probe of what dark matter really is.