
When the universe was less than 700 million years old, the supermassive black hole powering a quasar called J0313-1806 had already grown to about 3 billion times the mass of the sun. It is one of the most extreme examples of a puzzle that has occupied astrophysicists for two decades: How did supermassive black holes grow so big so quickly?
At its core, the puzzle is a speed limit. A black hole grows by swallowing, or accreting, gas, but the infalling matter heats up and shines, and the radiation pushes on the gas still arriving. At the so-called Eddington limit, radiation pressure balances gravity and infall cannot speed up further. Because this limit scales with the black hole’s mass, growth is exponential. If, as usually assumed, about 10% of the infalling mass is converted to radiation, the mass grows by a factor of e (about 2.7) every 45 million years.
A stellar-mass black hole of about a hundred solar masses would need about 17 such steps, roughly 770 million years, to reach 3 billion solar masses. When you account for the time it took for the stellar black holes to form, the universe had less than 600 million years available.
There are two popular ways out. One is to start with a “heavy seed” of at least 10,000 solar masses, formed by the direct collapse of enormous gas clouds. This would skip the slowest part of the climb. The other is to break the Eddington limit: If the energy released by infalling gas is removed in ways other than radiation, the black hole can grow far faster. Both options have caveats: Direct collapse requires a rare combination of conditions, and it is uncertain whether real accretion flows can stay super-Eddington for hundreds of millions of years.
In a new paper with my colleague Matas Tartėnas, published in Astronomy & Astrophysics, we ask whether a third, older and simpler idea survives confrontation with the latest data. It is called chaotic accretion, and it needs neither a heavy seed nor a broken speed limit.
Why a spinning-down black hole grows faster
The “chaos” in our model comes from how gas reaches the black hole. We consider a slowly rotating dark matter halo, so that gas falls in from essentially random directions. Each stream forms a small disk, tilted randomly relative to the black hole’s spin, feeds it briefly and is quickly followed by another, tilted differently. Some episodes spin the black hole up; others, where the disk counter-rotates, spin it down. Spin-down is rarer but more efficient per unit of mass, so over time the spin stays low.
That matters because a slowly spinning black hole converts less of the infalling mass into light: It radiates away only about 6 to 7.5% of the mass it swallows, rather than 10%. This may seem like a small difference, but it shortens the growth timescale from 45 to roughly 27–32 million years, and the time required to grow J0313-1806 from 770 to 550 million years.
In this scenario, the gas supply on large scales can exceed the rate necessary for the Eddington limit. What matters, though, is how much material reaches the event horizon, and that stays near the Eddington rate, while most of the rest is removed along the way. The large gas reservoir, however, ensures that the black hole may be fed nearly continuously.
What the data say
We compared this prediction with mass estimates for all known supermassive black holes at redshifts above seven. The most reliable (virial) estimates, out to redshift 10.6 and including J0313-1806, can all be explained by a seed of about 100 solar masses or smaller, formed around 100 million years after the Big Bang, growing on a timescale of about 32 million years. This requires almost continuous feeding at least 85% of the time at the Eddington rate, or all the time at 85% of it.
This shows that the scenario works, not that the alternatives fail. Our scenario cannot explain several existing mass estimates, but those are made using methods that rely on more assumptions about the accretion flow. For example, for the galaxy GN-z11, the virial estimate is about seven times lower than the less robust one. Several more distant black holes lack robust masses and would make excellent near-future observation targets.
How to tell the scenarios apart
We also propose ways to distinguish the three explanations. Chaotic accretion predicts slow spins for the most massive early black holes, while sustained super-Eddington growth should leave them spinning rapidly. Spin can be measured from the shape of X-ray emission lines, but current measurements are at lower redshifts and ambiguous. ESA’s planned NewAthena observatory is best suited to the task, but it won’t launch until at least 2037.
In our scenario, the maximum black hole mass should rise very rapidly until redshift 7.5 and then nearly plateau. Current data hint at this but cover too little volume to be conclusive; wide-area surveys such as Euclid and Roman should find enough black holes to test it. In particular, a robustly measured black hole of more than a billion solar masses at redshift 9.1 or higher would falsify our model.
Finally, once a rapidly growing black hole reaches 100,000 to 10 million solar masses, depending on its halo, its accretion energy can drive large-scale outflows that clear the surrounding gas and make it visible, possibly as a “little red dot,” a compact and bright source of the kind discovered by the James Webb Space Telescope. The inferred little red dot masses fall in this range, although what powers them is still debated.
For now, our conclusion is modest. Ordinary stellar-mass seeds growing near the Eddington limit with low spin can explain the earliest known supermassive black holes, without invoking massive seeds or super-Eddington growth. Whether that is what actually happened is a question the next generation of telescopes can answer.
This story is part of Science X Dialog, where researchers can report findings from their published research articles. Visit this page for information about Science X Dialog and how to participate.
Publication details
K. Zubovas et al., Chaotic accretion can (still) explain early supermassive black hole growth, Astronomy & Astrophysics (2026). DOI: 10.1051/0004-6361/202558523
Key concepts
Dr Kastytis Zubovas is a theoretical astrophysicist at the Center for Physical Sciences and Technology in Vilnius, Lithuania. He studies how supermassive black holes affect their host galaxies, primarily through the large-scale outflows they launch during active phases.
Dr Matas Tartėnas is also a theoretical astrophysicist at the Center for Physical Sciences and Technology. He uses numerical hydrodynamical simulations to study supermassive black hole feeding and feedback.
Citation:
Early black holes may have grown through productive chaos (2026, November 16)
retrieved 16 November 2026
from https://phys.org/news/2026-11-early-black-holes-grown-productive.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.

