At the heart of the Milky Way, just 27,000 light-years from Earth, there is a supermassive black hole with a mass of more than 4 million Suns. Nearly all galaxies contain a supermassive black hole, and many of them are much more massive. The black hole in the elliptical galaxy M87 has a mass of 6.5 billion Suns. The largest black holes are more than 40 billion solar masses. We know these monsters lurk in the cosmos, but how did they form?
One idea is that supermassive black holes form over time through mergers. Because of dark matter and dark energy, galaxies formed in clusters separated by voids. Over time the voids grow larger while the galaxies cluster together and eventually merge. The black holes within those galaxies also merge to form the supermassive objects we see today.
Of course, that takes time. If that model is correct, the most distant galaxies should have smaller, million-solar-mass black holes, and we should only see the billion-solar-mass giants in the nearby Universe. But observations from the James Webb Space Telescope found that the supermassive black holes in many of the most distant galaxies are huge. Black holes with a mass of more than a billion Suns already existed when the Universe was only half a billion years old. Those young giants are too massive to be explained by mergers, and they defy conventional explanations.
You might wonder why. After all, the early Universe was incredibly dense. With plenty of matter around for black holes to breakfast upon, why couldn't they fatten up fast? The reason is something known as the Eddington Limit. As matter is pulled toward a black hole, it becomes a super-hot, high-pressure plasma. This pushes more distant matter away from the black hole, slowing down the rate of growth. The Eddington Limit is the fastest rate a black hole can grow. This rate isn't fast enough to account for all the giant black holes we see in the early cosmos.
But the earliest period of the Universe is very different from the Universe today. What if the Eddington Limit didn't apply back then? This is the question examined in a recent paper on the arXiv. The authors created sophisticated hydrodynamic models to look at the formation of black holes during the cosmic dark age. The period after electrons and nuclei cooled to form atoms, but before reionization, when the first stars formed and reignited the cosmos with light. We know that this period is when galaxies started to form, so it's reasonable to presume supermassive black holes also formed during this time.
Based on their simulations, the authors found that there is a super-Eddington period. There are regions dense enough that superhot material close to a black hole can't clear the region. This allowed early black holes to grow at a rate faster than possible today, but only up to about 10,000 solar masses. According to the simulations, after that the Eddington feedback loop kicks in and the growth rate is limited again. The team also found that this super-Eddington growth doesn't help much in the long run. Eventually, even black holes that always grow at a sub-Eddington pace will achieve the same mass. Olympic sprinter Usain Bolt may be the world's fastest human, but marathoner Eliud Kipchoge will pass him in a longer run.
This study strongly suggests that super-Eddington growth can't explain all the billion-solar-mass black holes we see in the early Universe. Since galactic mergers also can't account for them, this work points toward another solution: seed mass black holes that formed very early, perhaps even during the inflationary period soon after the Big Bang.
Reference: Wu, Ziyong, Renyue Cen, and Romain Teyssier. "How Fast Could Supermassive Black Holes Grow At the Epoch of Reionization?" *arXiv preprint* arXiv:2510.16532 (2025).