As far as we can tell, nearly every galaxy out there has a supermassive black hole at its core. And when these black holes are actively ingesting matter, they create quasars, the brightest objects we've ever detected. Quasars appear to be present in some of the earliest galaxies we can detect, from when the Universe was only six percent of its current age.
That's a bit of a problem. The radiation a black hole emits while swallowing matter places a speed limit on the amount of matter it can ingest. Currently, we simply don't know how black holes got big enough to power a quasar less than a billion years after the birth of the Universe. But a paper from last week's edition of Science suggests that the stars present at the galaxy's core might cause gravitational instabilities that let the black hole overcome the speed limit on its growth.
Black holes are famous for having a point of no return, a distance where even photons cannot escape their gravitational draw. But beyond that point, the infalling matter can form what's called an accretion disk, where its interactions with the intense magnetic and gravitational fields send copious amounts of matter and energy flowing away from the black hole.
The photons produced by the infalling matter are what set the speed limit on the black hole's dining. At a point called the Eddington limit, the force exerted by the radiation pressure from these photons becomes greater than the draw of gravity. This creates a balance: too much matter and the black hole cuts off its own food supply until the lack of infalling matter causes the radiation pressure to subside.
The Eddington limit causes an astrophysical conundrum. It tells us that black holes should take a while to reach supermassive levels sufficient to power quasars. Yet observations tell us those quasars were there. So something unusual must be going on.
Researchers have considered a number of possible ways in which the earliest black holes could grow quickly. Some of the things they've considered include the direct collapse of massive clouds of gas without forming stars first or the merger of multiple massive stars to create a single, unusually massive black hole. But the new paper suggests that the Eddington limit could be sidestepped if you manage to get rid of the accretion disk.
The authors note that models of the early collapse of galaxies suggest that the process involves the organized flow of large amounts of cold gas into the galactic core. So, they produced a high-resolution model of what happens to that gas when it gets there. Their model included a large population of existing, massive stars in the core, along with a single, low-mass black hole. They then tracked the dynamics of the whole thing, seeing how the flow of cold gas interacted with the stars and black hole in the galactic core.
The modeling involves a lot of math, but some fascinating facts. For example, radiation pressure from close to the black hole is lower than you might expect from simple calculations. That's because the matter is dense enough that some of the photons are absorbed and re-emitted multiple times. While that's going on, some of the matter gets swept past the event horizon, taking the photons with it.
What they find is that, when an accretion disk forms, the Eddington limit kicks in. But the black hole's gravitational interactions with the massive stars at the galaxy's core regularly disrupt the accretion disk—in fact, until it grows above a certain mass, the black hole ping-pongs around the core of the galaxy due to these gravitational interactions. With the accretion disk disrupted, the inflowing gas can head directly into the black hole without any heed to the Eddington limit.
This process doesn't get the black hole up to the full size needed to power a quasar, but it does allow it to get very large very quickly. From there, things get a bit complex. The authors admit that "It is much more difficult to self-consistently predict the subsequent joint evolution of the BH [black hole] and cluster" at that point. But they say that you'd only need this to work in less than five percent of the initial star-forming regions of the Universe in order to produce the number of quasars we're seeing in the early Universe. While the details still need to be sorted out, it seems like a promising start.