Hacker News new | past | comments | ask | show | jobs | submit login
The possible disappearance of a massive star in the galaxy PHL 293B (oup.com)
104 points by karxxm 5 days ago | hide | past | favorite | 51 comments

To provide a little bit of context for this work, one of the open questions in astronomy is how core collapse supernovae work. Computational scientists have struggled for a long time to get simulations of core collapse to explode. Typically to get an explosion they have to artificially insert a "piston" somewhere in the system that kicks the system outwards during the collapse and triggers the shock wave.

Over maybe the last decade or so there's been some thought that if our simulations have trouble getting supernovae to explode, then perhaps nature has trouble getting supernovae to explode, too. Maybe a lot of massive stars just collapse directly into a black hole with no supernova. If this were the case then it would explain a discrepancy between the observed supernova rate and the total number of black holes; there seems to be about twice as many black holes as there are supernovae.

When I was in grad school one of the ongoing projects in my department was the so-called "Survey about Nothing" in which they took deep images of nearby galaxies and waited to see if any bright stars just... disappeared. (They did find one candidate: https://arxiv.org/abs/1411.1761)

The authors in this paper find another star that just disappeared. The trouble with all this is that if a star disappears it's unclear what exactly happened. Maybe it collapsed into a black hole, or maybe it just temporarily shrouded itself in dust and will be back in a few decades. The new thing about this star is that they have detailed spectra of the star prior to its disappearance, which helps tremendously with modeling the star. The hope is that with more detailed measurements they might be able to place limits on what is currently there and thereby figure out if a direct collapse happened or not.

Could you point to any laymanish articles or media that talks about the non-supernova fate? My knowledge about supernovae, black holes and neutron stars all come from youtube channels ( eg. https://youtu.be/RZkR9zdUv-E ) and they make it sound like inevitable ends for massive stars, because once the neutron star forms the sudden vacuum created by the compacted neutrons makes all the matter around it collapse on it and rebound. Why is that not plausible ?

It’s totally plausible.

But if the core of your star is sufficiently massive then the initial collapse takes it over the Chandrasekhar limit & it forms a black hole immediately. Then the rest of the stellar material just falls into it.

The fine details vary depending on mass, composition & spin of the rest of the material: if the star is spinning fast enough an accretion disk might form and drive jets which is one possible method of generating a GRB.

>> Then the rest of the stellar material just falls into it.

Spherical chickens in a vacuum. This would be true of a non-spinning black hole. In reality they all spin. The star was spinning before collapse. So a tiny amount quietly falls in, but the bulk of the outer shell quickly accelerates, energizes, jets form and generally the massive releases of energy blasts material away.

This is definitely outside my area of competence, but I believe that jets that manage to escape the stellar volume only form under a very specific set of circumstances: https://en.wikipedia.org/wiki/Gamma-ray_burst_progenitors#Co...

I believe current models suggest that a low-metallicity stars (which are the ones who’s core can collage straight to a black hole, with no intermediate neutron star) will typically collapse with no supernova or other visible trace. High mass (> 90 SM) might lead to a GRB, but not a supernovae.

You can follow the references at https://en.wikipedia.org/wiki/Failed_supernova & the supernova wikipedia page for the gory details.

Having part of the star form accretion disc and then slowly radiate the energy is entirely different from the star converting significant part of its mass into energy and radiating it in couple of hours.

Also jets are highly directional and you have good chance to be away from the cone.

I am not a physicist but I suspect that extremely active star turning into black hole might be much less luminous than the star itself. Accretion disks can be very luminous but this is typically for very large holes in the centers of galaxies. Because we observe these huge stars only because they are so highly active, the act of converting it to much less luminous system might be undistinguishable from the star disappearing.

Thanks, very helpful.

I wonder if radio astronomy could put some limits on the obscured by dust scenario. I suppose it's unlikely for the star to be bright enough in such bands, would raise more questions than answers. But maybe astronomy is just one unlikely event after another.

Is there a project that shows which portions of the sky are currently viewed by which telescopes?

Also, how can researchers request/rent usage of telescopes (especially satellite-based ones)?

I have experience only with the Very Large Telescope run by ESO in Chile, maybe what I write is only valid there.

There is usually no real schedule of targets for ground-based telescopes. There are two ways chosen at the telescope: either the astronomer who wrote the application is sitting there and decides what to do, or the staff goes through a list of approved programs and looks for objects that can be observed. This depends on the constraints described in the program (usually atmospheric conditions and height of object above horizon). Which target is chosen next is usually decided during the observation of the last target, there is no real schedule.

This is of course different for robotic telescopes (which is absolutely not the standard), like the Hubble Space Telescope but also ground-based robotic telescopes. But I'm not aware of a live feed of the pointing coordinates for them.

What one could do is regularly query the ESO archive[1] as finished observations appear there immediately (I think) and contain coordinates (just enter night: "2020 01 01" and maybe chose type: object).

In case of the Hubble Space Telescope and also ESO's telescopes, you write a proposal containing the science case, the requested time and instruments, and related previous experience, submit it before a deadline (twice each year for ESO) and hope for the best. The acceptance rate for the HST is currently ~20% [2]. It's a bit better for ESO telescopes. If you are successful you do not have to pay anything. ESO even pays your flight and hotel next to the telescopes in the middle of a desert [3].

The situation is totally different for American telescopes (as far as I know), where you either belong to an institution that has telescope time or not.

[1] http://archive.eso.org/eso/eso_archive_main.html [2] https://www.stsci.edu/contents/newsletters/2019-volume-36-is... [3] https://en.wikipedia.org/wiki/ESO_Hotel

Thanks for the great explanation.

Are we tracking the spectra of everything visible in the sky? How much data is that?

In short: No, we cannot do that at the moment.

We have photometric all-sky surveys that can map the entire sky (visible from the telescope location) during a night up to a certain brightness. But those only take images of the sky, not spectra (Zwicky Transient Facility and the planned Vera C. Rubin Observatory).

What we also have are integral-field spectrographs which can take 2D images with a twist: there is one image for every ~0.1nm from 480nm to 950nm. You take one exposure with the instrument and you get a stack of thousands of images. If you go through the stack at a fixed spatial position you get the spectrum. The problem is that the integral-field spectrograph with the largest field-of-view is already huge (it is called MUSE at is located at the Very Large Telescope). And its field-of-view is "only" 1 arcmin^2 (1 deg = 60 arcmin), which is by far too small for large surveys. If you wanted to image the whole sky each night with MUSE clones, you would need several millions of them. A single MUSE exposure is about ~5 GB in the end but there are intermediate data products which are about 10 GB, if I remember correctly.

here it says that such an event should create detectable gravitational waves https://aasnova.org/2019/07/05/can-we-detect-gravitational-w... The article mentions that we might get good enough detectors in some ten years already.

Very interesting. Do you know the theoretical probability that at least one observable star will collapse/disappear during any given earth year?

Great comment, thank you.

It is very informative but I especially admire your ability to communicate so clearly about a complex topic. I wish I could do that.

IANAP and this could be a bit of a stretch, but could the disappearance of the star been caused by the completion of a Dyson sphere around it?

It would be; first, spheres are unrealistic because of the stresses involved, so you'd be looking at a dyson swarm instead, and second, (and this is with sci-fi glasses on), even advanced civilizations would need many years to build one, so there would be a gradual fading instead.

It could be the result of the closing of the dyson sphere/swarm. If the dyson sphere/swarm is built like giant Venetian blinds, it could work.

Easier: if they build the frame for the sphere first, then added (gigantic) panels one at a time, a part filled with panels might have moved into our line of sight.

Or maybe they built them all in a higher orbit spaced apart, then simultaneously decelerated them all so they fell as one into a Dyson sphere! Have to have some kind of process, why not that one?

Self-propelled modular elements that are built separately, and then placed and connected to others.

A Dyson sphere/swarm has huge emissions on the infrared band that would be easy to detect on the tens of millions of light years distance.

Edit: AFAIK, this group didn't look for that. But the fact that nobody ever found one is telling.

If you are into these kinds of mysteries, I highly recommend reading Pandora's Star by Peter F Hamilton. (even more so if you want to spend a few days reading this epic space opera)

Warning: he writes huge tomes. They're far too long. The first I read was fun, the second a slog. There won't be a third.

He has a whole follow-up trilogy from Pandora's Star. They are every bit as massive and pondorous. I personally liked them just enough to finish them, but I won't re-read them.

However, I absolutely love the Night's Dawn Trilogy and re-read it once every few years. I think I read it at an impressionable age and it really stuck.

One could also consider reading Charles Stross' "Accelerando" which has plenty of stuff about dyson swarms.

I only read Pandora's Star so far, but I really liked the fact that it was a rather optimistic vision of the future. Humanity had colonies on dozens or hundreds of planets, and due to life extension people were basically immortal and could even choose the shape of their body.

Definitely read the sequel Judas Unchained. Those books are probably my favourite novels.

For more Peter F Hamilton optimistic future of humanity, see the Night's Dawn trilogy.

Thank you! Any other optimistic and believable sci-fi you would recommend?

I'd imagine that such a completion would take long enough that we'd see it dimming (maybe irregularly) first. Then again I'd have thought so would a star collapsing...

If anything, the star certainly looks like a perfect candidate to go supernova…

So today I learned that we have the ability to notice when a single star (possibly) disappears from a galaxy 75 million light years away. A small thing, but also a rather wonderful example of what we're capable of.

It was a particularly massive and bright star, though.

Still a great achievement!

This sounds like the plot of Pandoras Star by Peter F Hamilton

At the farthest edge of the Commonwealth, astronomer Dudley Bose observes the impossible: Over one thousand light-years away, a star... vanishes. It does not go supernova. It does not collapse into a black hole. It simply disappears.

Lets hope its not the [[SPOILER]]

Or Vernor Vinge's A Deepness in The Sky, which is also excellent.

Sounds like "Pandora's Star" by Peter F. Hamilton.

Maybe we should check into the financial records of those funding the observation.

A Dyson sphere!

So they hypothesize it either collapsed to a black hole without supernova or stopped fusion. They don't mention it might just might have been swallowed by black hole.

Wouldn't that be a slow process?

That is my impression also. Stars/objects being swallowed by black holes tend to emit a large amount of radiation as it happens, over longer periods. Someone correct me if I'm wrong.

Maybe stupid question - what would happen to the star, if it actually 'collided' with black hole - ie black hole flew right through its core, not just casually going nearby, trapping each other and black hole slowly sucking it dry?

I would expect the thing to be over rather swiftly, albeit probably with some accretion disc still generating some radiation.

A head-on collision with a black hole will likely produce a very noticeable explosion.

Black holes don't just vacuum up all the matter around them silently. Any matter falling into a black hole, not just the accretion disk, emits radiation. This radiation causes surrounding matter to expand outward, counteracting the gravity of the black hole. A black hole trying to eat an entire star at once will quickly exceed its Eddington limit [1] and most of the remaining matter will be blown away.

[1] https://en.wikipedia.org/wiki/Eddington_luminosity

Before the actual head-on collision the star would be spaghettified(that's actual term, see below) by the black hole's gravity and that... is not a silent process, with exception of supermassive black holes.


Slow and exothermic.

A star would be unlikely to directly enter an event horizon but would instead orbit the black hole (with resulting doppler-shift accelerations detectable on Earth via spectral analysis), and strong emissions from IR to x-ray frequencies as matter was ripped from the star by tidal forces and drawn into the event horizon.


More validation of the dark forest theory?

Probably not. If it got sniped by some other civilization, it would probably produce a burst of radiation on many wavelengths.

Sometimes astrologists make me smile:

> important sources of ionizing photons

Nothing better than a morning with cereal, milk and ionized photons.

Wikipedia says that LVB often form clouds around them from their violent outbursts that are sometimes miss-classified as supernovae. We will probably have to wait if it pops up again.

Sometimes astrologists make me smile

Nothing makes astronomers smile more than being called astrologists.

We smile on the outside yet cry on the inside.

Oops... well, they share the same hobbies and with such colleagues they never have to buy any fortune cookies.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact