How Black Holes Kill Galaxies (Patreon)

Black holes are really only dangerous if you get too close. Ha, who am I kidding. It turns out they may be responsible for ending star formation across the entire universe.

When we first realized that black holes could have masses of millions or even billions of times that of the sun, it came as a bit of a shock. They were discovered as the driving force behind quasars, where matter is heated to extreme incandescence before its plunge into vast black holes. But if that weren’t enough, we soon realized that every single decent-sized galaxy contains such a supermassive black hole.

 By the beginning of the 21st century it became clear that black holes and the galaxies that contain them are very closely connected. The bigger the galaxy, the bigger its supermassive black hole. That might not sound surprising. What was weird was how closely they were connected. There’s a tight correlation between the mass the central black hole and the mass of the stars in the galactic bulge – that’s the central ball-like part of a spiral galaxy, or the entirety of an elliptical galaxy, and every bulge contains a supermassive black hole around one-one-thousandth its mass. And there’s an even tighter relationship between the black hole mass and the speed that stars are moving in their random orbits within the galactic bulge – the so-called stellar velocity dispersion – which itself depends of the total mass of the galaxy, including dark matter.

So why shouldn’t a galaxy and its black hole be closely connected? A couple of things you need to know. Even though supermassive black holes are big, they’re peanuts compared to the galaxy they live in. Their gravitational influence should only extend to the stars right in the very center of the galaxy. They definitely aren’t directly responsible for speed of stellar orbits outside their local region. Sure, galaxies and their black holes probably grew together – but galaxy formation is an extremely messy process, so you’d expect a huge amount of variation in black hole versus galaxy mass. Because of all this, the tightness black-hole-galaxy connection has been a decades-long conundrum that still isn’t solved. But one of the leading ideas is that central black holes somehow extend their nefarious influence throughout the entire galaxy, disrupting its ability to grow.

To understand this, first we need to understand how galaxies form. In short – small galaxies collapse from the gas of the Big Bang, and they smash together to make bigger galaxies. This is “bottom-up” hierarchical galaxy formation, as opposed to “top-down” which would have the large galaxies collapsing directly from that gas early on. Based on our understanding of the physics of our universe – especially the nature of dark matter and dark energy – we expect bottom-up galaxy formation. This is what we see in our simulations of the universe and also mostly agrees with what we see when we look out into the universe.

Galaxies first grew in the clusters, in the densest parts of the universe – places where enormous wells of dark matter pulled in great rivers of material from all around. Gas poured into the clusters from the outside universe igniting bouts extreme of star formation called starbursts. 

As these galaxies build, so did their black holes. They will have started as already-massive seed black holes formed by the very first generations of stars. They would fall to the centers of their local wispy proto-galaxy, and when wisps merged their central black holes would also merge, all the while gorging on the rich gas supply in the early universe. So black holes grew as galaxies grew. Why is it so surprising that they appear to be so closely correlated in the modern universe?

Two things In order for the black hole to grow in lock-step with the galaxy you need to get a consistent proportion of new material down to the center of the galaxy. That’s challenging given how messy and varied the whole galaxy assembly process seems to be. Second, observations seem to indicate that early black holes actually grew faster than their surrounding galaxies. The galaxy-black hole mass relationships seem to evolve through the history of the universe. Many large supermassive black holes were in place early, leaving their surrounding galaxies to catch up, like a puppy growing into its giant feet. In fact we’ve seen quasars shining out from less that a billion years after the big bang with the mass of 10 billion Suns – easily as large as the largest in the modern universe,

So if black holes and galaxies are NOT growing in lock-step with each other from the same material, and they can’t directly influence each other with gravity. How do they know what size to be? How does a galaxy know to stop growing when it’s 1000 times larger than its central black hole?

Perhaps the best contender is that black holes kill galaxies. And by “kill” I mean make them dead. Which I guess is the usual sense of the word. But in astronomy, a “dead” galaxy refers to its current star formation activity. In particular, the largest galaxies in the universe are giant ellipticals – and we say they are “red and dead” – not because they contain lots of old-west themed planets – that’s not yet known. Rather, it’s because they appear to have almost no active star formation, and haven’t for a long time. The short-lived hot, massive stars that give spiral galaxies like the Milky Way their blue-white sheen have long-since died out in giant ellipticals, leaving them tinged with the orange-red colours of longer-lived cooler stars.

But from what we can tell, those giant galaxies should have kept forming stars. Giant reservoirs of gas still flowed into their clusters from the outside universe. In our simulations of the universe, cluster ellipticals end up much bigger and bluer due to billions more years of ongoing star formations. A likely culprit is the black hole – it killed star formation in those giant galaxies, fixing the maximum size that they could grow. The process is called quenching, and there are a couple of different ways it can happen via black hole – but all involve a quasar switching on and blasting the crap out of the galaxy’s gas.

Rewinding a bit to the whole galaxy formation thing – when galaxies collide, or even get stirred up a bit by eating a smaller galaxy, gas tends to get driven down into the center where it’s gobbled up by the black hole. Well, some of it is. Close to the black hole the gas forms a whirlpool-like accretion disk, heated by the energy of its long fall. It gets so hot that around 10% of the mass of the infalling gas is converted to energy in the form of light. And yet more of the gas is blasted outwards by that same light into colossal winds that rip through the surrounding galaxy, or is channeled into jets by powerful magnetic fields.

A quasar’s intense brightness and these powerful winds and jets can dump an enormous amount of energy into the gas throughout the surrounding galaxy. That can do two things: it can drive gas out of the galaxy and prevent new gas arriving, or it can heat the gas. Either case is bad for star formation. In the latter case it’s because hot gas can’t form stars. Gas has to cool down before the force of gravity can cause it to collapse into stars. Too hot and it stays puffed up. As well as killing star formation, quasar activity limits the black hole’s own growth. It grows while eating the new supply of gas, but then its energy output stops new gas falling into the galactic center.

In the end you have this balancing act – a feedback process. The more massive the galaxy the more stuff it can pull in to feed itself and its black hole, but the more massive the black hole the better it is at shutting down both star formation and its own feeding. The end result is that black hole mass and galaxy size are closely linked. And in the largest galaxies – our red-dead ellipticals, star formation is entirely shut down.

Whether or not black holes kill star formation – or at least are the main culprit – is not established. But we DO know that star formation died. Or is dying. Not just in the largest galaxies – but across the universe. After the first stars formed around 150 million years after the Big Bang, the rate of star formation across the universe slowly rose. It became concentrated in the most dense regions in the universe where  giant ellipticals assembled. The rising storm of star formation reached a crescendo 4 or 5 billion years ago when finally the largest galaxies were in place. Then the work of galaxy formation shifted towards the outer regions, further from the giant clusters. Then around 2 billion years ago the cosmic star formation rate plummeted, and continues to do so. This is mainly due to star formation shutting down in the densest parts of the universe – which, as I mentioned, is strange given they should still have plenty of gas.

There are other explanations besides quasars and black holes for this shutdown. Star formation has to be a somewhat self-limiting process because intense starbursts lead to intense supernova activity, which can also heat or expel gas and limit further formation. For the largest galaxies, the shock experienced by gas falling in from the outside universe can be enough to stop stars forming. All of these probably play some role in killing star formation – they are accomplices to the crime, but many astronomers think supermassive black holes are the main suspect.

Here’s one final clue: compare the cosmic history of black hole growth to that of star formation. They grow together, with black hole growth peaking around 3 or 4 billion years ago, right before the precipitous drop in star formation. Now some very massive black holes existed earlier than this – somewhat perplexingly. But this peak in black hole growth represents the peak of the quasar epoch – when the most outrageously luminous quasars shone from the most ridiculously massive galaxies. Surely enough to shut down those galaxies’ further growth.

In the modern universe, giant dead galaxies harbor fossil quasars – supermassive black holes whose close connection to their surrounding galaxy is a clue. Incriminating evidence that suggests a long and violent struggle – a feedback cycle of growth and quenching that neither quasar nor star-forming galaxy would survive. Perhaps for the best; for the comfort of life-bearing worlds, best to leave raging starbursts and fiery quasars to an earlier epoch of spacetime.