Mark Dodici

If a stellar binary — a pair of stars, bound together in orbit — comes too close to a massive black hole, the stars may be torn from each other by the gravity of that black hole. This tragic process is known as the Hills mechanism, named after its discoverer, Jack Hills. The Hills mechanism has two interesting byproducts: one star is left behind on a very close orbit around the massive black hole, while the other star is flung far afield, at speeds as great as 4,000 kilometers per second!

We see both sorts of stars in our Milky Way Galaxy. Some stars orbit very close to Sagittarius A*, the massive black hole in our Galactic Center, and they might have been placed there by such disruptions. Other stars — the fastest ones we know of — move at hundreds to thousands of kilometers per second through the outskirts of the Galaxy.

But what actually causes a binary to come "too close" to a massive black hole? In this work, I point out that massive disks of gas, which occasionally arise around massive black holes, can push binaries living there onto extreme trajectories — bringing them too close, indeed, to said black holes. While this process is far from the only one causing Hills-mechanism disruptions, it is interesting in how temporary it is; these disks only live for a few million years, so they drive binary disruptions in "bursts" of time comparable to or shorter than that window.

This process might have happened recently in the Milky Way! There's a collection of stars around Sagittarius A* that are thought to have been born about 5 million years ago in a massive, gaseous disk. In this work, I find that the disk that formed these stars might have caused ~100 binary disruptions. That means ~100 very tightly bound stars around Sagittarius A*, and ~100 very fast stars flung away from it about 5 million years ago. I would argue it's no coincidence that S5-HVS1 — the fastest known star and the only fast star whose trajectory we can confidently trace back to Galactic Center — left that region 4.8 million years ago...

Stellar binaries orbiting supermassive black holes are likely involved in a range of (often observable) astronomical processes. These processes depend on the typical properties of this binary population. This population should be sculpted by a range of dynamical effects; each binary will be incessantly perturbed both by the supermassive black hole and the other stars in the dense cluster around it.

In this work, we show that these effects will cause a surprisingly large fraction of the binaries in these environments to take on highly elliptical orbits, to the point that their closest approach during an orbit can be as small as a few stellar radii. At such small separations, in an extremely elliptical orbit, the two stars raise significant dynamical tides on each other. These tides rapidly shrink the binary orbit, leaving the pair separated by just a few times their own radii. We find that this "diffusive tidal shrinking" should act on roughly half of the binaries in the central few parsecs of the Milky Way, for example, and suggest that it should sculpt the binary population in the central part of any galaxy with a supermassive black hole.

Start with two objects in space (stars, asteroids, black holes, whatever you like!) that are unbound from each other. If you throw them at each other, they will interact gravitationally. However, it's impossible for them to form a permanent binary pairing unless you have some way to get rid of energy.

For the last few years, researchers have studied this process for stellar-mass black holes in the gaseous disks that sometimes crop up around supermassive black holes at the centers of galaxies. There should be a lot of solo stellar-mass black holes in these regions, and there's a lot of background gas in the disk, so perhaps you can form a lot of binary pairs with these stellar-mass black holes. If you form a lot of binaries, the conditions are also amenable for the black holes to spiral into each other and merge, inducing gravitational waves detectable by LIGO/Virgo.

But how easy is it to actually form a binary in this way? In this work, we take a simple approach to the problem of gas-driven pairings, and we introduce the idea that stars in the centers of galaxies can lead to binary formation in much the same way. We run over 100 million simulations of objects approaching each other under dissipative forces, covering a much wider range of this problem's vast parameter space than past works. We find simple relationships between the properties of the gas disk or surrounding stars and the efficacy of binary formation, and we compile recent results from other researchers to show that they've been finding the same relationships. We also point out that our understanding of these gas disks is really very uncertain, and that different models give hugely different rates of binary formation, which present problems for constraining this channel of LIGO/Virgo source formation.

Conveniently, everything about this work is scale-free, so the results can be applied to completely different scenarios – like the formation of binary asteroids in the gas disks of a nascent planetary system. Let me know if you have a good use-case!