Title: Unveiling the Central Engine of Core-collapse Supernovae in the Local Universe:
Authors: Maurice H. P. M. van Putten, Maryam Aghaei Abchouyeh , and Massimo Della Valle
First Author’s Institution: Sejong University, Seoul, Republic of Korea
Status: Published in The Astrophysical Journal Letters [open access]
The violent delights of massive stars have violent ends. The largest stars in the universe are believed to die in spectacular core-collapse supernovae, where the rapid burning of the last vestiges of a star’s nuclear fuel cause the stalemate of pressure and gravity to suddenly become one sided; the pressure is routed by the force of gravity and the star implodes.
This implosion leaves behind a smoldering corpse, either a neutron star or a black hole. If the neutron star or black hole is spinning very rapidly, we may say that the supernova has a powerful central engine because the rotational energy of the spinning object could release powerful eruptions of energy such as gamma ray bursts. Identifying the type of central engine left over from a supernova based on the electromagnetic or neutrino signal is pretty tough, meaning it’s difficult to confirm our hypotheses about whether these core-collapse supernovae should end up as neutron stars or black holes. However, we live in the 21st century, baby, and gravitational waves are opening up new insights into the universe. Could the identity of the corpses of these supernovae be determined by looking at their gravitational wave signal? The author’s of today’s paper strive to find out!
Black Hole Central Engines
Rapidly spinning black holes are excellent objects for multi-messenger observations (meaning we see them through multiple different types of observations, for example light and gravitational waves) thanks to their extremely energetic nature. Firstly, a black hole is much more efficient at storing energy in the form of angular momentum than a neutron star. If you spun a neutron star fast enough, it would eventually break apart, while a black hole would not. For comparison, a neutron star can only store about one-tenth the energy a black hole can in this way.
Secondly, the spin of a black hole changes the geometry of the local spacetime. All black holes have an innermost stable circular orbit (or ISCO): a radius below which circular orbits are not allowed; anything that is trying to orbit below the ISCO will spiral into the event horizon of the black hole. However, spinning a black hole shrinks its ISCO, and very rapidly spinning ones can have the ISCO almost butt up against its event horizon. Since the ISCO defines a rough inner edge of the accretion disk of orbiting matter, a rapidly spinning black hole can more directly interact with its accretion disk. This allows for some interesting possibilities for multi-messenger signals as the black hole directly interacts with the spin of the matter around it. This would again allow for a black hole-driven signals to be much stronger than a neutron star’s.
To determine what kind of signal a rapidly rotating black hole leftover from a core-collapse supernova produces, the authors consider the “energy budget” of the system. Combining several recent works, the authors conclude that, of the possible avenues by which energy could leave the system, the least energy would likely be carried away by the black hole forming a jet of particles or by its magnetic winds. (See Figure 2.) Slightly more energy would be expected from neutrinos being shot out from processes occurring in the accretion disk. The greatest amount of energy that would be expected to escape in this scenario would be in the form of gravitational waves! This would be second only to the energy that is dispersed by crossing the event horizon of the black hole, never to be seen again. Bummer.
So what kind of gravitational waves might we expect to see? After the core collapse, we expect the central engine to be surrounded by a disk or torus (donut shape) of matter from the death of the star. If this torus is spinning slower than the black hole itself, we’d expect the black hole to lose momentum as the surrounding torus tugs on it. This energy is released as gravitational waves as the mass of the black hole and torus is quickly yanked around. Since the system is losing speed, the authors predict it should release a “descending chirp” of gravitational waves. If you’ve read other Astrobites about gravitational waves, you may have heard about an ascending chirp – when two compact objects spiral in during a merger, they begin to spin around each other faster and faster, resulting in a gravitational wave signal that sweeps up in frequency. Since this system slows down instead, we’d expect the exact opposite. In fact, this descending chirp was observed in the spindown phase of GW170817, the first gravitational wave event also to be observed alongside a light signal, see Figure 3.
From analysis of that event in a previous paper, the authors can predict how much energy should be released in the form of gravitational waves for a similar event. They predict that for every extra 2.8 solar masses of a black hole, you’d expect an extra 6 × 1045 Joules of gravitational wave energy (the way this is denoted in the paper is 3.5% of a solar mass directly converted to energy via E = mc2). GW170817 had a chirp that swept from a frequency of 700 Hz (about the F5 note on a piano) to about 200 Hz (the G2 note, the lowest note on a violin); the authors predict this frequency range will linearly scale with the size of the system. This descending chirp would allow us to distinguish a gravitational wave signal from a rapidly spinning object from that of a merger.
Tell-tale signs of a black hole
As an extra way to possibly distinguish a black hole and neutron star, the gravitational wave signal may show signs of Lense-Thirring precession.
In physics, precession is when an object’s rotational axis changes in time. The horrendously overused but perfect example is when you spin a top and its handle slowly draws out a circle as the top spins. In our case, as the black hole spins, it is spinning so fast and is so heavy that it pulls space time with it a bit, in a process called frame-dragging. If the black hole’s spin is misaligned from the orbital plane of the accretion disk, this effect may cause the entire system to precess, see Figure 4. This precession would give an extra component to the gravitational wave signal; if our main signal is at some frequency, we might expect to see Lense-Thirring “side lobes” at the original frequency plus or minus the Lense-Thirring precession frequency. Going back to the spinning top metaphor, this means we would see a main gravitational wave signal at the frequency of the top’s spin, and an extra signal that is shifted by the frequency at which the axis of the top precesses in a little circle. All of this is to say, if the central engine is a black hole, we expect two extra gravitational wave frequencies on opposite sides of the main frequency, thanks to the Lense-Thirring effect. A neutron star is not dense enough to cause the frame dragging necessary to produce this effect, so we would not expect this Lense-Thirring signal for a neutron star.
The Hunt is on!
So we know what pattern we expect to see from this kind of event and roughly how strong of a signal we’d see; from this information we can predict how likely we are to observe this kind of event! How large would we expect the black holes to be that form from core-collapse supernovae? The authors argue that some fraction of the Helium core that the star had before going supernova should collapse into a black hole, setting a rough lower limit of 5 solar masses for the newborn black hole. Using the scaling arguments made before, we’d expect 7% of a solar mass’s worth of energy to be released as gravitational waves over a frequency chirp of about 100-380 Hz. Thankfully, this just so happens to align nicely with Ligo-Virgo-Kagra’s (LVK) most sensitive region of 100-250 Hz. Plus, given that GW170817 released about 2.5% of a solar mass worth of gravitational wave energy, this signal size is certainly observable! The authors estimate a signal of this kind would be visible out to a distance of 160 Megaparsec or 522 million light years. Combining this information with the increased sensitivity of LVK, now in its fourth observing run, and with the expected frequency of core collapse supernovae, we might expect to see a few of these events per year. The authors, therefore, propose three possible outcomes from searching for these signals.
- If we see a core-collapse supernova’s electromagnetic signal and check for a gravitational wave signal but find nothing in the frequency bands we mentioned above, this suggests that the central engine is a neutron star, and whatever gravitational waves released, if any, were too weak to be detected.
- We see a gravitational wave signal, but it has much less energy than predicted above for a black hole – a few percent of a solar mass’s worth of energy. This would again point to a less energetic object, such as a slowly rotating neutron star.
- If we see a gravitational wave signal in the right energy and frequency ranges, we could claim that the central engine is a black hole! This result would be especially solid with a secondary piece of evidence, such as the ejecta from the supernova having high kinetic energy, or by seeing the Lense-Thirring signal sidelobes.
This method gets much more reliable when following up on a supernova spotted first as a light signal rather than a blind search for gravitational waves. The authors estimate in order to have a few hundred supernovae that can be analyzed by LVK using this method, they would need a survey of a few thousand core-collapse supernovae. With excellent automated transient searches such as the Zwicky Transient Facility already up and running and new proposed missions such as the THESEUS satellite in the pipeline, these surveys could help us spot the gravitational wave signal counterpart. With enough data points, we might be able to use gravitational waves to settle the question: neutron star or black hole?
Edited by Bill Smith
Featured image: Wikimedia Commons, Hubble Space Telescope
