Overmassive Early Black Holes Were Born Heavy, Simulations Suggest

JWST has found black holes that seem to have skipped ahead in class. At redshifts greater than 4–6, their estimated masses sit far above the relationship between central black holes and their host galaxies seen nearby. Some appear as little red dots: compact, reddish sources with broad emission lines. How did they get so heavy so early?

A new preprint posted to arXiv offers a possible answer. Sunmyon Chon and colleagues ran cosmological radiation-hydrodynamic simulations that follow the birth, growth, and observable glow of black holes in the early universe. Their result is a physical scenario, not a direct observation: in crowded proto-cluster regions, black hole seeds can be born heavy and then feed at extraordinary rates.

Born Heavy in a Crowded Neighborhood

Supermassive black holes are known to exist less than a billion years after the Big Bang. Standard scenarios often start small. In this work's comparison, a black hole seeded by a Population III remnant had about 800 solar masses, while canonical direct-collapse models predict seeds around 100,000 solar masses. Even those canonical seeds have trouble growing fast enough.

JWST sharpened the puzzle. It found overmassive black holes at redshifts greater than 4–6: their masses are unusually large for their host galaxies, compared with the local black-hole-to-star relation. Little red dots are especially intriguing. They are compact, red, and broad-lined, suggesting dense gas moving rapidly around a growing black hole.

The simulations used the moving-mesh code AREPO, which models gas, gravity, radiation, and chemistry in a cosmological setting. The team focused on one halo identified as a promising heavy-seed site. It sits about 10 kiloparsecs from a luminous neighbor, close enough to be bathed in far-ultraviolet radiation.

That radiation suppresses molecular hydrogen, the coolant that lets primordial gas clump into ordinary stars. Without efficient cooling, the halo accumulates a large gas reservoir. For this target, collapse began at redshift 14. Ionized gas stayed gravitationally bound, feeding central protostars that grew to between 500,000 and 900,000 solar masses, roughly ten times the canonical direct-collapse seed. Multiple seed-forming sites appeared nearby.

A Short, Bright, Super-Eddington Infancy

Once a massive star collapses into a black hole, growth accelerates. The new black hole is surrounded by a dense, optically thick envelope. Accretion becomes super-Eddington: gas falls in several to a few tens of times faster than the Eddington limit, the rate at which radiation pressure would normally balance gravity. Radiation is trapped and dragged inward, so it cannot immediately halt the inflow. This phase lasts less than a million years but pushes the black hole to several million solar masses.

As the gas supply fades, growth settles. By redshift 10, the simulated black holes grow to at least about 10 million solar masses. By redshift 8, they have settled into the central kiloparsec of a galaxy with about a billion solar masses in stars, giving a black-hole-to-stellar-mass ratio near 1 percent, about an order of magnitude above the local relation.

A higher-resolution zoom-in followed the gas disk around one seed down to scales of 500 astronomical units. The disk reached hydrogen densities above 10^10 per cubic centimeter. Such dense gas produces strong H-alpha emission, a hydrogen line. Electron scattering broadens it to widths above 1000 kilometers per second. The modeled H-alpha luminosity, about 1.5 times 10^43 ergs per second within 10^4 astronomical units, is comparable to values inferred for little red dots. In this picture, a little red dot is a short-lived, enshrouded phase of heavy-seed formation and early growth.

The comparison with light seeds is stark. A black hole seeded by a Population III remnant of about 800 solar masses barely grew. Radiation from its parent star and the shallow gravitational potential of its host prevented gas from staying bound and feeding it. Its final mass remained orders of magnitude smaller.

From Red Cocoon to Overmassive Quasar—and What Remains Unproven

If the scenario holds, little red dots are not a separate mystery. They are an early chapter in the same story that leads to overmassive quasars detected by JWST. The simulations suggest the enshrouded phase can end within about 0.1 to 1 million years, as the disk thins and the black hole becomes a less obscured, more AGN-like source.

The simulation yields a lower limit of about 10^-4 massive-black-hole-hosting galaxies per cubic megaparsec, consistent with the abundance needed to explain observed little red dots. Later mergers of these heavy seeds, the authors note, could produce gravitational-wave signals in the millihertz band for the future Laser Interferometer Space Antenna. Those are forecasts, not detections.

This is a model, not a direct observation of a seed forming. The main run follows one representative halo, and the radiation field is partly based on a semi-analytic model of nearby galaxy formation. The team tested weaker far-ultraviolet intensities, reduced by factors of two, five, and ten; final masses changed by less than a factor of three, and even the weakest case reached about 19 million solar masses by redshift 7. A higher-resolution run produced similar growth.

Other caveats remain. The authors do not resolve the full galactic gas inflows needed to sustain long-term Eddington accretion, so they cannot follow the black holes to the billion-solar-mass quasars seen later. One early disk has a Thomson optical depth of 10.2, higher than the rough range of 0.1 to 1 inferred from some observations, though the effective depth for H-alpha photons is smaller. The simulations omit jets, and the LISA rates remain theoretical.

What makes the result striking is how naturally the pieces fit in the simulation. A crowded early environment suppresses ordinary star formation, allowing a heavy seed to form. That seed spends a brief, bright infancy wrapped in dense gas, looking like a little red dot. It then emerges as an overmassive black hole, already too large for its galaxy. If future observations and gravitational-wave searches match these predictions, the first giant black holes may have been born heavy, not just grown fast.