This page is a description of Intermediate-Mass Black Holes (IMBHs) and my studies of them for the general public. I will not go into the details of what makes a black hole a black hole since there are many other web pages that already do a very good job of that. This page is about IMBHs. It is about the reasons we think they do or do not exist and the reasons they are interesting. I know that I need some pictures on this page. I’m working on it! If you like this page or if you have more questions about what is in it, please email me at .

We begin with the observations that first inspired the idea of IMBHs: ultraluminous X-ray sources (ULXs). Let us say the following about ULXs: first, they are X-ray point sources, meaning that they emit X-rays and appear to come from a single point as opposed to spread out over a noticeable region; second, they appear to have an X-ray luminosity greater than 10³⁹ erg/s, meaning its brightness in X-rays alone is 500,000 times brighter than the Sun’s brightness in all wavelengths^*; third, they are non-nuclear, meaning they are found away from the nucleus of their host galaxy; and lastly, they show variability on a scale of weeks or a few months, meaning the ULX’s brightness goes up and down noticeably over the course of weeks or months.

The fact that ULXs are point sources and (more importantly) the fact that they are variable means that they are small. The fact that they are emitting X-rays also means that they are hot. Astronomers have seen objects similar to these before, the most important example of which are the X-ray binaries. In an X-ray binary, what you see is gas that is falling onto a compact object, like an ordinary black hole. The gas comes from a companion to the black hole, and hence the binary part of the name. As the gas falls closer and closer to the black hole, it starts spiraling around the black hole in an accretion disk. The gas flows around and around the black hole and, through friction and compression, heats itself up enough to emit X-rays. Another object that operates in the same way is an active galactic nucleus (AGN), which is a supermassive black hole at the center of a galaxy and is actively accreting gas.

We know that ULXs are not AGNs and are not powered by supermassive black holes (black holes with masses one million to one billion times the mass of the sun) because they are non-nuclear. A black hole that big would quickly sink to the center of its host galaxy. So it must be smaller than a supermassive black hole. We know that ULXs are not ordinary X-ray binaries because they are so luminous in X-rays. If something that is held together by gravity (like a star or an accretion disk) is radiating equally in all directions, it can blow itself apart due to the force that the light imparts on the matter. ULXs appear to be too luminous to be held together by the gravity of an ordinary black hole, which has a mass less than twenty times the mass of the Sun. Thus, in order to explain ULXs, we need a black hole that is larger than an ordinary (stellar-mass) black hole and smaller than a supermassive black hole: we need an intermediate-mass black hole.

* Different scientists have slightly different criteria for how luminous an X-ray object is before it is considered ultra-luminous, but the differences aren’t really important for our discussion. [Back]

There is also evidence that there are black holes at the centers of objects called globular clusters in our Galaxy. This evidence is more circumstantial, but it is interesting because it is a more direct measure of the mass of the suspected IMBH, and its mass is what makes it so special. A globular cluster is a gravitationally bound ball of about a hundred thousand to a million stars. There are about 150 of them in our Galaxy, and they are found in the outskirts of our Galaxy and other galaxies. Optical observations of the centers of two of these clusters reveal that the stars’ motions may be influenced by an unseen mass. The exact value of this mass is the critical thing to know. It is not certain, but the best guesses are 3,000 and 10,000 times the mass of Sun. In both cases, it is impossible to rule out the possibility that there is no IMBH, but the current best guess is that these two particular globular clusters harbor IMBHs.

Not all scientists agree that IMBHs exist. The skeptics say that the dark masses at the centers of globular clusters are consistent with no IMBH, which is true. They also say that observations of ULXs are deceptive. Although we can measure how much energy is radiating in our direction from the ULXs, we cannot tell how much radiation is being radiated in every direction. It is possible that we are seeing an intense beam of radiation that is pointing in our direction. We say that this object is beamed. When we calculate how much energy is radiated in all directions, we overestimate it because we assume that the same amount of energy is being radiated in all directions. So there is no need to think that the black hole must be more massive to keep from blowing away the accretion disk. It could be an ordinary stellar-mass black hole that happens to have its radiation beamed towards us. We see similar behavior in supermassive black holes in other galaxies so it is not extraordinary to expect it from stellar-mass black holes.

On the other hand, there is increasing evidence that some of the ULXs cannot be beamed. Most notably, there is a source that is surrounded by gas that is lit up by radiation so much that the source must be radiating equally in all directions. Although nobody thinks that all ULXs are unbeamed sources, if some of them are, then that means that IMBHs can exist.

Incidentally, some ULXs are neither IMBHs nor beamed stellar-mass black holes. Some ULXs are going to be background AGNs and some are supernova remnants, but these are really just misclassified objects masquerading as ULXs.

So what do astrophysicist want to learn about IMBHs? The first and most pressing issue is how did they form? They cannot form from the same mechanism as stellar mass black holes, which form from the core collapse supernova of ordinary, big stars, or supermassive black holes, which are much more massive. There are three main ideas.

The first is that they come from the core collapse supernova of the first generation of stars (called Population III stars), which were much larger than stars formed around the time the Sun formed. It is unclear if this method can form black holes as big as the largest IMBH candidates, and you wouldn’t expect to find these black holes in some of the places that they are found.

The second method happens in young stellar clusters. Stellar clusters are balls of stars held together by gravity. In these clusters, the heaviest stars sink to the center and can actually collide with one another and become one massive star-like thing. These collisions happen over and over again until you end up with something with a few hundred to a couple thousand times the mass of the sun. This thing then turns into an IMBH. This method is nice because it takes place near where we find IMBHs. The questions surrounding this method have to do with the details of going from that big "star-like thing" to a black hole. It isn’t obvious how this complicated thing evolves into a black hole or if it keeps all of its mass as it does so.

The third method happens in globular clusters, which are large, old stellar clusters. In this method, all of the stellar-mass black holes sink to the center of cluster where they interact and merge with each other. This process is much slower than the process of colliding stars. If you have enough repeated mergers, you end up with an IMBH. The biggest question remaining about this method is whether this method is efficient enough.. There are definitely enough black holes in a globular cluster to make up the IMBH; in fact, there are ten times the minimum number required. During the process of merging black holes, however, many are ejected from the globular cluster. The question is whether the growing black hole will eject all of the black holes from the globular cluster before it has grown.

This last method of formation and the questions surrounding it are what my colleagues and I study. We answer these questions with numerical simulations of the interactions of black holes in globular clusters. We can do this because we understand gravity very well, and we have a very good idea of the conditions in a globular cluster. (Our understanding of gravity is more than just a good idea; it’s the Law of Gravity.) I put all of this together into a computer and wait for the answers to churn out! To see a report on some of the early stuff my collaborators and I did, look at this Science article by Govert Schilling.