Black hole mass coupled to expansion -- astrophysical source of dark energy?

[Bandersnatch]

TL;DR Summary

A recent article proposes observed black hole growth in ellipticals being coupled to expansion, and sourcing dark energy.

Observational evidence for cosmological coupling of black holes and its implications for an astrophysical source of dark energy​

Observations have found black holes spanning ten orders of magnitude in mass across most of cosmic history. The Kerr black hole solution is however provisional as its behavior at infinity is incompatible with an expanding universe. Black hole models with realistic behavior at infinity predict that the gravitating mass of a black hole can increase with the expansion of the universe independently of accretion or mergers, in a manner that depends on the black hole's interior solution. We test this prediction by considering the growth of supermassive black holes in elliptical galaxies over 0<z≲2.5. We find evidence for cosmologically coupled mass growth among these black holes, with zero cosmological coupling excluded at 99.98% confidence. The redshift dependence of the mass growth implies that, at z≲7, black holes contribute an effectively constant cosmological energy density to Friedmann's equations. The continuity equation then requires that black holes contribute cosmologically as vacuum energy. We further show that black hole production from the cosmic star formation history gives the value of ΩΛ measured by Planck while being consistent with constraints from massive compact halo objects. We thus propose that stellar remnant black holes are the astrophysical origin of dark energy, explaining the onset of accelerating expansion at z∼0.7.

Comments?

 

[PeterDonis]

It's an interesting hypothesis, but it's going to take time for the claimed observational evidence to be independently evaluated. There are a lot of uncertainties in these observations so I would say it's premature to put too much weight on any particular hypothesis of this sort.

 

[PAllen]

I'll just make the comment that the first two sentences of the abstract:

"Observations have found black holes spanning ten orders of magnitude in mass across most of cosmic history. The Kerr black hole solution is however provisional as its behavior at infinity is incompatible with an expanding universe."

are clearly true, so research in this area is required.

 

[.Scott]

It seems we didn't have to wait long for an additional related observation:

Astrophysical Journal: re Black Hole Growth rates

Also:
Astrophysical Journal Letters: re Black Holes and Dark Energy

A Life Science article suggests that these observations suggest an unexplained factor in BH growth rates, alternate internal BH structure; and a cosmic source for Dark Energy.

 

[PeterDonis]

It seems we didn't have to wait long for an additional related observation:

Astrological Journal: re Black Hole Growth rates

I think you mean Astrophysical Journal?

 

[srce]

Would someone be able to try to provide a bit of a layman's explanation for this?

I was under the impression that dark energy was meant to be evenly distributed. How does this fit in with BHs being a source of dark energy?

 

[Ibix]

How does this fit in with BHs being a source of dark energy?

Other way around, I think. Edit: but see my next post. The textbook models of black holes don't match well to the real universe because they are "asymptotically flat", meaning that the models look more and more like flat spacetime as you get far from the hole. But far from a hole in the real world, spacetime is not flat (even "flat" cosmological solutions are only spatially flat - they're still curved spacetimes). This appears to be fine for modelling short-timescale interactions like black hole mergers, but if you want to talk about the long term behaviour of black holes you need to match up the black hole model's behaviour at infinity to the real world behaviour.

Apparently there's been some recent work that lets people take a crack at doing this (we knew the problem before, but not how to address it). From a skim of the papers cited above, I think they suggest that the masses of black holes evolve in a way that depends on the dark energy content of the universe. Since the mass of a black hole is quite a tricky thing to define precisely that's not at all unbelievable. But, as usual with cutting edge science, this should go firmly in the "maybe" file until the community has had a chance to rip it apart and see how (if) it works.

 

[srce]

Other way around, I think.

From the abstract: "We thus propose that stellar remnant black holes are the astrophysical origin of dark energy" - what does this mean then?

There's a podcast where Pearson says BH's gain mass due to the expansion of space (and uses the analogly of a rubber band being stretched), which seems fair enough. But struggling with the sentence above (which is what is widely being reported in the press)

 

[Ibix]

From the abstract: "We thus propose that stellar remnant black holes are the astrophysical origin of dark energy" - what does this mean then?

That I needed to read more closely. Reading further, I think they are saying that the way they acquire mass means that a large and reasonably uniform population of their $k=3$ black holes acts like a dark energy term in FLRW solutions. But there's a lot I don't understand about the black hole species they're talking about.

 

[PeterDonis]

I was under the impression that dark energy was meant to be evenly distributed.

That's the assumption in our current models, yes. But...

How does this fit in with BHs being a source of dark energy?

The same way our current model deals with ordinary matter: it has an ordinary matter density that is the same everywhere (at a given FRW coordinate time), even though the actual ordinary matter in our universe is clumped. The density in the model is the average density.

 

[PeterDonis]

From a skim of the papers cited above, I think they suggest that the masses of black holes evolve in a way that depends on the dark energy content of the universe.

As you note in your next post, that's not correct.

Reading further, I think they are saying that the way they acquire mass means that a large and reasonably uniform population of their $k = 3$ black holes acts like a dark energy term in FLRW solutions.

More precisely, these "black holes" contain dark energy in their interiors, and a suitable population of them will have an average dark energy density that matches up with what is in our current cosmological models based on the observed accelerated expansion of the universe.

But there's a lot I don't understand about the black hole species they're talking about.

The main point to understand is that they aren't true black holes: they don't have a singularity and they don't have a true event horizon. A reasonable heuristic model is the Bardeen "black hole" and related models that were discussed in a fairly recent thread; see in particular the paper I discuss in this post:

https://www.physicsforums.com/threa...t-collapse-wave-function.1048041/post-6832588

From the outside this type of object looks like a black hole, but its horizon is only an apparent horizon, not an event horizon, and inside it there is dark energy that violates the energy conditions and allows the interior to remain stable (or more precisely meta-stable--it will eventually evaporate) over very long time scales (roughly the same as the Hawking evaporation time scale). I don't know that the paper referenced in the OP of this thread is using the same model, but it should be the same general type of model.

 

[srce]

The same way our current model deals with ordinary matter: it has an ordinary matter density that is the same everywhere (at a given FRW coordinate time), even though the actual ordinary matter in our universe is clumped. The density in the model is the average density.

Ok, but if the dark energy is inside the BHs, how is it "getting out" to drive expansion that is occurring far away from the BHs?

 

[phyzguy]

The main point to understand is that they aren't true black holes: they don't have a singularity and they don't have a true event horizon. A reasonable heuristic model is the Bardeen "black hole" and related models that were discussed in a fairly recent thread; see in particular the paper I discuss in this post:

From the outside this type of object looks like a black hole, but its horizon is only an apparent horizon, not an event horizon, and inside it there is dark energy that violates the energy conditions and allows the interior to remain stable (or more precisely meta-stable--it will eventually evaporate) over very long time scales (roughly the same as the Hawking evaporation time scale). I don't know that the paper referenced in the OP of this thread is using the same model, but it should be the same general type of model.

Thanks for this explanation. If this model is correct, does that mean that there are two types of black holes? (1) "normal" black holes formed from the collapse of ordinary matter, and (2)these more exotic black-hole-like objects that are filled with dark energy?

 

[PeterDonis]

if the dark energy is inside the BHs, how is it "getting out" to drive expansion that is occurring far away from the BHs?

It doesn't have to get out. An average dark energy density can drive an average accelerated expansion. Just as an average normal matter density drives an average decelerated expansion. The cosmological models based on FRW spacetime are large scale averages; the fact that up to now the dark energy density has been assumed to be the same everywhere (because in the absence of any evidence to the contrary that was the simplest model--basically a cosmological constant) does not mean the model stops working if it turns out the dark energy density is just a large scale average the way we already know the ordinary matter density is.

 

[PeterDonis]

If this model is correct, does that mean that there are two types of black holes? (1) "normal" black holes formed from the collapse of ordinary matter, and (2)these more exotic black-hole-like objects that are filled with dark energy?

No. It means that the compact objects we now call "black holes", formed from the collapse of ordinary matter, do not form event horizons and singularities; instead, the collapse process ends up producing dark energy in the deep interior of these objects, so they are not true black holes but these more exotic objects instead. In other words, in our actual universe, if this model is correct, there are no actual black holes, with event horizons and singularities, anywhere.

 

[Ken G]

This strikes me as a potential discovery of seismic proportions. Not only does it change our understanding of what black holes are, in the way being explained so elegantly by PeterDonis, it also explains the largest current puzzle in all of astronomy-- what is the dominant source of the universal dynamics on the largest scales (the acceleration of the expansion). It should be pointed out that it would also resolve a third sticky problem: how supermassive black holes managed to get so massive, when their accretion rate should have been limited by what is called "the Eddington limit."

This is the limit that says as a spherically symmetric object accretes mass, it should give off some of that energy as radiation, which carries momentum and interacts with gas not yet accreted, pushing it away and limiting accretion to the "Eddington rate". Calculations of this maximum rate make it very difficult for black hole masses to rise as fast as they have. Up until now, it was assumed the problem was with the assumptions that go into the Eddington rate calculation, such as spherical symmetry. But with this new suggestion, we have a very simple escape from the Eddington rate limitation-- this new approach to dark energy gives us a free "get out of jail" card.

I agree that more corroboration of both the observations and the theory is needed, but it certainly has me excited for the prospects. One thing that does bother me, though, is that this overall scenario must still be missing something important that remains to be understood. They seem to be essentially saying that as black holes form, they generate dark energy, which explains the acceleration of the expansion, which violates the asymptotically flat boundary condition, which explains why these are not normal black holes. So they are not normal black holes because by being not normal, they are able to be not normal! I don't see the reason in all this that they are not just normal, and there is not acceleration. We know that isn't what happens, but I'm missing the "why can't these just be normal black holes with no universal acceleration?" In other words, it does not sound like this resolution to dark energy pulls dark energy into what we consider to be the current body of theoretical physics, there has to be some new element of theoretical physics that must be added, to explain why an equally possible outcome is that black holes would not generate dark energy, and would exist in an asymptotically flat non-accelerating universe. Hence, we have made key progress, if this is true, but we still have our initial question: what is dark energy?

 

[mfb]

There seems to be some concern about their galaxy selection having potential biases, which can influence the measured growth rate of the black holes. On the other hand, it's well-known that early universe black holes grew faster than we expect, so it's still evidence for this model unless someone can explain it in a better way.

 

[Sagittarius A-Star]

there has to be some new element of theoretical physics that must be added, to explain why an equally possible outcome is that black holes would not generate dark energy, and would exist in an asymptotically flat non-accelerating universe.

In the following paper from 2005 they suggest, that the picture of gravitational collapse provided by
classical general relativity cannot be physically correct:

The picture of gravitational collapse provided by classical general relativity cannot be physically correct because it conflicts with ordinary quantum mechanics. For example, an event horizon makes it impossible to everywhere synchronize atomic clocks. As an alternative it has been proposed that the vacuum state has off-diagonal order, and that space-time undergoes a continuous phase transition near to where general relativity predicts there should be an event horizon. For example, it is expected that gravitational collapse of objects with masses greater than a few solar masses should lead to the formation of a compact object whose surface corresponds to a quantum critical surface for space-time, and whose interior differs from ordinary space-time only in having a much larger vacuum energy [1]. I call such an object a “dark energy star“.

Source:
https://arxiv.org/abs/astro-ph/0503200

 

[Ken G]

It does seem clear that our previous way of thinking about black holes always created contradictions with quantum mechanics (and Hawking spent a lot of time trying to resolve that, too bad he did not live long enough to see this new possibility). The new way doesn't lead to singularities, so perhaps it resolves those contradictions as well, I'm not sure of that.

By the way, the proposers do claim that their black hole picture does not require any new physics at all, it's pure GR just with a different boundary condition, or maybe some other free assumptions that are allowed in GR theory that just weren't being used before. If that is true, I'd say it's a pretty powerful persuasion. Choosing boundary conditions has always been a kind of black art in physics, the part of the theory that is not really in the theory. So if they can convince me that is all they had to do, then I'm on board. If it also resolves contradictions with quantum mechanics, allows the Eddington accretion limit to be avoided, and explain the acceleration of the universe, all in one go, then count me in all the more. But I don't know the GR, to convince me that all this emerges directly from existing equations with no ad hoc "reverse engineering" to make it work, other than choosing a boundary condition that we already know is the one we see. And in particular, I would need to understand why the "asymptotic" influence is not of inverse-square nature, but instead requires extremely large distances to make its presence felt.

Put differently, they are claiming their black hole model explains the acceleration. What if taking an accelerating boundary only explains the increase of black hole masses, without explaining the acceleration in the first place? Then we would still get the answer to the Eddington accretion problem, and perhaps the resolution of the singularity issues, but we still wouldn't know what dark energy is or why the universe is accelerating.

 

[Ken G]

I think I can boil down what I would need to be convinced into getting answers to two questions, that people here with good GR understanding might be able to tackle:
1) What does the empirical k=3 result mean in terms of the GR solution, i.e., is this somehow inevitable for the distant boundary condition of an expanding universe, or is it a parameter of the acceleration that is simply empirically inserted?
2) Why does it make sense that the internal mass of the black hole should be affected by the behavior of spacetime vastly far away, and not the spacetime in the actual galaxy where the black hole resides? (In other words, usually in physics when we assert a boundary condition, we don't look so far away, we go out only as far as we need to in order to convince ourselves that everything even farther than that is somehow irrelevant.)

 

[PeterDonis]

it would also resolve a third sticky problem: how supermassive black holes managed to get so massive

I don't see how this dark energy hypothesis resolves that. The formation of dark energy in the deep interior of these objects does not change their externally measured mass. It only changes the equation of state in the interior. So it doesn't cause these objects to get massive any faster than previous models.

 

[PeterDonis]

They seem to be essentially saying that as black holes form, they generate dark energy

More precisely, the collapsing matter, in the deep interior of these objects, undergoes a sort of phase transition that changes its equation of state from that of normal matter to that of dark energy.

which explains the acceleration of the expansion

Yes.

which violates the asymptotically flat boundary condition

No. The asymptotically flat boundary condition has always been violated for the universe as a whole; the universe as a whole is not and never has been asymptotically flat. That's why asymptotically flat models of anything (including ordinary stars and planets) must be approximations to what is going on in our actual universe. That's always been the case.

As far as I can tell, these "dark energy objects" could just as easily form with a boundary condition that was asymptotically flat; there's nothing about them that precludes an asymptotically flat boundary condition. The comments in the paper that appear to imply that these models somehow fix a "problem" with ordinary black hole models due to asymptotic flatness are one of the parts of the paper that I'm skeptical about. There are already well known models, such as the Bardeen black holes and related solutions that I mentioned earlier, that are asymptotically flat but have dark energy in their deep interiors.

The usual way that models that, in the idealized case, are asymptotically flat are used in the actual universe is to, heuristically, cut a "hole" in the global model (such as FRW spacetime), and then stitch a finite portion of the "black hole" model (or whatever we are modeling, such as the spacetime around an ordinary star or planet) into the hole. This could be done just as easily with a Bardeen black hole or some similar model with a dark energy interior.

If I try to distill out what seems to me to be the key new claim of the paper, it would be that, if the global model is an expanding FRW universe, the "stitching" process I just described above has to take into account the global expansion, and doing that has an effect on the "black hole" model that you are stitching in. This is what I think the "coupling" in the paper is describing.

 

[Ken G]

And just how that emerges becomes the key issue. They claim they are not using anything but standard GR, just doing it right, in some sense. But I did think they were saying that the measured black hole mass, and its influence on its accretion disk and so on, does increase due to their dark energy solution. They seem to be claiming that the reason the ratio of SMBH mass to the stellar mass of the host galaxy rises with time does not require accretion, it is due to this cosmological coupling. That sounds like a kind of "action at a distance" that we are not so used to having-- whereby the distant universe affects the black hole more than its own surrounding "stitched patch", as you say. The usual expectation was that the stitching could be done without affecting the black hole (and especially its measured mass), but somehow that's what they are saying isn't right. That's the part that I think needs to be explained better, so it doesn't seem so "magical."

Also, what is not clear is if the accelerated expansion is causing dark energy to be created in the black hole, as if the distant spacetime was "coupling" to the black hole mass in a way where the causal arrow points from the distant behavior to the black hole response, rather than the other way around as they claim. One cannot claim to explain why the universe is accelerating if it is the acceleration that causes the black hole behavior, one has to show that the black hole behavior causes the expansion or the question of acceleration remains unanswered (though it would still be nice to remove the singularity, and the Eddington accretion problems). In other words, the question "why this kind of black hole and not the Bardeen type, if both are valid GR solutions" remains unanswered, so the cause of the acceleration remains unanswered.

 

[srce]

It doesn't have to get out. An average dark energy density can drive an average accelerated expansion. Just as an average normal matter density drives an average decelerated expansion. The cosmological models based on FRW spacetime are large scale averages; the fact that up to now the dark energy density has been assumed to be the same everywhere (because in the absence of any evidence to the contrary that was the simplest model--basically a cosmological constant) does not mean the model stops working if it turns out the dark energy density is just a large scale average the way we already know the ordinary matter density is.

Ok, but would that mean there's a different source of dark energy that is driving the observed expansion outside of BHs (i.e. the expansion between galaxies)?

 

[Ibix]

Ok, but would that mean there's a different source of dark energy that is driving the observed expansion outside of BHs (i.e. the expansion between galaxies)?

As far as I understand it in this model the black holes are the dark energy. There doesn't need to be any elsewhere, at least in principle.

 

[PeterDonis]

would that mean there's a different source of dark energy that is driving the observed expansion outside of BHs (i.e. the expansion between galaxies)?

Not in the model proposed in the paper under discussion, no. The dark energy in the interiors of the objects described in the paper can drive accelerated expansion just fine.

 

[artis]

Ok , I got bit confused reading all this.
So the claim is that certain types of black holes by whatever unknown mechanism radiate "dark energy" which itself is also unknown to us apart from the fact that it apparently exists, so far so good?

But given every black hole in existence was created out of ordinary matter collapsing that would mean they also "radiate" ordinary gravity.

So these special types of BH radiate both gravity and dark energy?
So in terms of gravity they add to the total matter made gravity constant within the universe but in terms of dark energy they add to the total acceleration potential observed within the universe?

How about the "action at a distance" that @Ken G just mentioned above, is the BH radiated dark energy "fall off" rate from source different than that of gravity or EM fields?

I guess the question that can't be answered but which is most interesting is how does ordinary matter start to radiate energy (under special circumstances) which we cannot directly observe therefore also can't understand and yet we see it's influence through it;'s effects on ordinary matter/ spacetime.

 

[Ibix]

Ok , I got bit confused reading all this.
So the claim is that certain types of black holes by whatever unknown mechanism radiate "dark energy" which itself is also unknown to us apart from the fact that it apparently exists, so far so good?

No. The model is that ordinary stuff falling into black holes becomes what we have been calling dark energy inside the hole. Nothing is being radiated from them, and there is no dark energy outside black holes.

 

[martinbn]

Where can we see a bit more about these black holes? The papers don't give much detail and there are too many references in them to check everything.

 

[Ken G]

It's possible that one way to look at this is that dark energy is the essentially the resolution of the longstanding debate about how GR can be made consistent with quantum mechanics. If quantum mechanics requires that black holes must not contain true singularities, then it forces them to find solutions to GR other than the Schwarzschild or Kerr metrics, so it finds solutions that come with long-range and highly non-Newtonian effects that lead to accelerated expansion. Those solutions allow black holes to "hit outside their weight class", in the sense of having much greater large-scale effects than any of the solutions we expected to be right up until now.

In other words, if this new picture is correct, we can say that our two deepest theories of physics could only both be right in a universe that is accelerating. We already had evidence that both theories are very good, but we certainly did not know that one or the other did not break down inside black holes. If it turns out that neither does, it is a remarkable example of human physics theory also hitting outside its weight class. I don't think anyone expected both theories to be that right!

It should also be noted that in this new theory, the black holes responsible for the acceleration are not the supermassive black holes, they are the vastly more numerous stellar-mass black holes (whose total mass is thought to significantly exceed that of the supermassive ones). You could say that during the age of the quasars, the universe tried to create supermassive black holes whose combined effects were strong enough to cause accelerated expansion, but it fell short of the mark because the supermassive black holes ceased having sufficient accretion rates. So the universe tried again, this time through the more continuous effort of star formation, and it might just have succeeded that way. What's more, had we not lived during this phase of extended star formation, we would never have had any evidence that this was the case (though of course, there also would not have been sufficient metallicity to support the chemistry we need to be here).

 

[ShayanJ]

This is very interesting, but unfortunately I'm a bit rusty on the subject so...

We thus propose that stellar remnant black holes are the astrophysical origin of dark energy, explaining the onset of accelerating expansion at z∼0.7

Is this after inflation or the same as inflation?
Sorry if my question doesn't make sense, but it seems like that, if this is true, we can establish this timeline:

Big Bang => Inflation => temperature goes down as the universe expands => pockets of matter start to form as temperature goes down => some of those pockets of matter form black holes and generate dark energy inside them => dark energy accelerates the expansion further => ...

Does this make sense? Or am I just confused about something?
Thanks

 

[phinds]

Nothing is being radiated from them, and there is no dark energy outside black holes.

This whole discussion is way over my head but that statement in particular seems very strange. If there is no dark energy outside the BH and nothing is being radiated out, why is the universe undergoing an accelerated expansion?

 

[martinbn]

This whole discussion is way over my head but that statement in particular seems very strange. If there is no dark energy outside the BH and nothing is being radiated out, why is the universe undergoing an accelerated expansion?

I think (not sure) the black holes contribute a term to the Friedmann equations that is like a cosmological constant.

 

[ShayanJ]

This whole discussion is way over my head but that statement in particular seems very strange. If there is no dark energy outside the BH and nothing is being radiated out, why is the universe undergoing an accelerated expansion?

Well, black holes were never like the stars they were created from, even before this. When you're talking about a star, you can always separate the matter from the spacetime. But when a blackhole is formed, there is no matter anymore, you just have a little region in spacetime that is behaving in a weird way. So I think this paper would imply that dark energy is only concentrated in those weird regions of spacetime.

 

[.Scott]

Nothing is being radiated from them, and there is no dark energy outside black holes.

Let me see if I have a workable picture.
Would it be fair to say that the dark energy has an effect on the geometry of the space around it?

Let's see if I be more specific in my question: If there was only one supercluster of galaxies in the universe that had stellar-mass black holes, would the result be that the accelerating expansion would be "centered" or "concentrated" about that supercluster?