Authors: Luca Boccioli, Giacomo Fragione
First Author’s Institution: University of California, Berkeley
Status: Accepted to Physical Review D [closed access]
It’s surprising how little we know about how stars explode.
At the end of a massive star’s life, its iron core collapses, no longer able to sustain its own weight. Material from the outer layers falls down onto the collapsing core, then bounces outward in a violent explosion, leaving a neutron star (NS) made from the core of the former star.
Or, it doesn’t. Some stars, rather than exploding and leaving a neutron star remnant, simply continue to collapse, leaving a black hole (BH). While a lot of work has gone into determining the mechanisms behind supernovae, it’s still unclear exactly what triggers an explosion, especially in the so-called “low-mass gap” where we have trouble finding both neutron stars and black holes. Whether that gap corresponds to a real physical effect or is simply an effect of our detection methods is also unknown. Today’s authors, therefore, create a new model to describe a star’s explodability! (Yes, that’s the technical term.)
Islands of explodability
Astronomers used to believe that a star’s explodability was determined simply by its mass – stars more than around 20 times the mass of the Sun would collapse into a black hole, while stars less than 20 solar masses but more than around 8 would explode in a supernova. It turns out, though, that the pattern of which initial masses will explode is much more complex, containing several different mass regions at which the star is predicted to explode. As can be seen in Figure 1, different authors find different locations for these “islands of explodability”, since they’re highly dependent on the stellar evolution models used to compute them. It’s therefore difficult to determine the outcome for any given initial mass and metallicity. However, the authors of today’s paper define an explodability criterion that depends instead on the star’s compactness, or how densely material is packed into its center, as well as on the location of the star’s silicon-oxygen interface.
Massive stars near the end of their lives are made up of onion-like layers of different elements, like the model shown in Figure 2. The hydrogen that the star was originally mostly composed of makes up the outer layer, followed by layers made of the products of fusing heavier and heavier elements into each other. The two layers just above the star’s inert iron core are made of oxygen and silicon, and the location of the boundary between these two layers is very important in supernova models. When the outgoing shock wave reaches this interface, it encounters an abrupt drop in the star’s density, and with less material to travel through, the shock wave can suddenly travel much more rapidly. So, if a star’s core is less dense, or if the shock wave reaches a sharp enough drop in density at the Si/O interface early enough in its lifetime, it will explode!
The low-mass gap
The authors defined their explodability criterion by comparing their results to those of supernova simulations, but how does it do when predicting observations? We can test this by looking at the predicted masses of the two possible remnants, neutron stars and black holes, compared to what we see in the data.
In Figure 3, the authors do just that, comparing the observed masses of neutron stars to the predicted NS masses for stars that explode according to their criterion. They test four different methods of predicting the NS mass, and find that a piecewise function based on the star’s compactness before core collapse (yellow line) fits the data best. All four methods, though, fit the data best when a bimodal distribution is assumed – i. e. the number of neutron stars peaks at two different masses. We also see this in the observed mass distribution that we measure from millisecond pulsars, which peaks in the same two places as the authors’ predicted bimodal distributions.
Any stars that do not explode and leave a neutron star, meanwhile, will collapse into a black hole. The authors determine the mass distribution of these black holes based on their explodability criterion, assuming that some fraction fej of the star’s outer hydrogen is ejected and the rest of its material falls onto the newly born BH. They allow this fraction to vary, and construct mass distributions based on both a truncated power law and a Gaussian distribution. Then, they can compare the results to the distribution of BH masses measured by the LIGO, Virgo, and KABRA collaborations from gravitational waves, which they do in Figure 4.
A truncated power law with over 80% of the star’s hydrogen being ejected fits the observed distribution best, and the authors predict a significant population of low-mass black holes in the gap predicted by other studies. Although we don’t have many data points to go on yet and the uncertainties are large, gravitational wave observations, in particular the recent LIGO detection of a probable black hole 3.6 times the mass of the Sun, seem to back up this prediction.
Supernova science is not a trivial task. There are a lot of moving parts, and the field contains many questions that remain open. Today’s authors present a viable predictor for the explodability of massive stars, and use it to make predictions that seem to fit our observations. But who knows? As we get more data, we might find something that surprises us!
Astrobite edited by: Kylee Carden
Featured image credit: NASA/Swift/Skyworks Digital/Dana Berry
Discover more from astrobites
Subscribe to get the latest posts to your email.
