Can black holes merge to create larger event horizons?

[mfb]

I'm not sure if I understand that. Inside the event horizon, tidal forces grow without bound for any black hole, don't they? So are you talking about the tidal forces for something hovering outside the event horizon?

I don't consider "inside" a region where current physics works. Sure, if you just ignore quantum effects at all, you can use GR to describe that region. But is that how real black holes look like? Where "look" is a bit ironic because we can never "see" it...

 

[Soule]

A black hole is not really formed starting from a microscopic point, it's being formed as the entirety of the stellar core's mass is being compressed catastrophically (when the neutron degeneracy pressure is overwhelmed) into the Schwarzschild radius.

Oh ok. That happens the instant the core reaches its Scwarzschild radius. I should have figured that out as that is the one thing about black holes I can calculate XD.

 

[PeterDonis]

A black hole is not really formed starting from a microscopic point, it's being formed as the entirety of the stellar core's mass is being compressed catastrophically (when the neutron degeneracy pressure is overwhelmed) into the Schwarzschild radius.

Actually, if "black hole" means "event horizon", then the horizon *does* start as a point and grows in radius. This happens inside the collapsing matter; the event horizon reaches the Schwarzschild radius corresponding to the total mass of the collapsing matter at the same instant as the outer surface of the matter reaches that radius as it collapses. After that instant, the horizon stays at the same radius forever (unless more matter falls in).

The tidal forces at the horizon as it grows, however, are never larger than the tidal forces at the horizon after the matter has collapsed through it. So if the black hole's tidal gravity at the horizon isn't enough to separate quarks, the tidal gravity while it's forming won't be either.

 

[PeterDonis]

I don't consider "inside" a region where current physics works. Sure, if you just ignore quantum effects at all, you can use GR to describe that region. But is that how real black holes look like? Where "look" is a bit ironic because we can never "see" it...

I don't think the fact that we can't directly "see" inside the hole's horizon means that "current physics doesn't work" inside the horizon. For a black hole of stellar mass or larger, the curvature at the horizon is much smaller than any sort of curvature threshold where quantum effects should become important, at least according to our best current understanding. So Occam's razor says that physics inside the horizon, at least down to the point where the curvature *does* get large enough for quantum effects to be important, should work the same as physics outside the horizon--meaning that we should be able to use GR to describe it.

Yes, various speculations about how to resolve the black hole information paradox involve assuming that the above is not the case, that there are quantum effects that *are* important even at the horizon of a black hole of this size. But those are speculations. Unless and until we get some actual experimental evidence to back up those speculations, I don't think it's fair to just say that "current physics doesn't work" inside the horizon. Using GR to describe physics inside the horizon, for cases where the curvature there is small enough, is just extrapolating a well-confirmed theory into a regime in which it should still be valid, given our best current knowledge, even though we can't experimentally explore that regime directly. We do that in physics all the time; in fact, if we couldn't do that, physics would be useless, because it could only tell us about experiments that we've already done and regimes that we've already explored.

 

[Matterwave]

Actually, if "black hole" means "event horizon", then the horizon *does* start as a point and grows in radius. This happens inside the collapsing matter; the event horizon reaches the Schwarzschild radius corresponding to the total mass of the collapsing matter at the same instant as the outer surface of the matter reaches that radius as it collapses. After that instant, the horizon stays at the same radius forever (unless more matter falls in).

The tidal forces at the horizon as it grows, however, are never larger than the tidal forces at the horizon after the matter has collapsed through it. So if the black hole's tidal gravity at the horizon isn't enough to separate quarks, the tidal gravity while it's forming won't be either.

From a purely GR standpoint (with all the assumptions of spherical symmetry, and isotropy, etc.) you might be correct. But from an astrophysical standpoint, I don't think you can really say for sure anyone point where an event horizon starts to grow. The collapse of a stellar core is a catastrophic process, occurring in time scales of micro-seconds to milliseconds. Additionally, there is no guarantee that all of the assumptions of isotropy and homogeneity, for example, are met.

I felt it prudent, to inform the OP that the stellar collapse process is a catastrophic one, in order to distance him from the idea that stellar collapse happens starting from one tiny point and slowly growing outwards. But if there is a problem with this idea, then be sure to correct me. :)

 

[PeterDonis]

From a purely GR standpoint (with all the assumptions of spherical symmetry, and isotropy, etc.) you might be correct. But from an astrophysical standpoint, I don't think you can really say for sure anyone point where an event horizon starts to grow.

Dropping the assumption of spherical symmetry certainly makes things more complicated; AFAIK there are no analytical solutions for the general case, only numerical simulations. But AFAIK that does not change the qualitative features that I described. Bear in mind that the event horizon is the boundary of the region of spacetime that can't send light signals to future null infinity; qualitatively, such a region *has* to start with a single point (more precisely, there has to be some earliest spacelike hypersurface that the EH intersects, and it must intersect that hypersurface at a single point), even if the details of the process are not spherically symmetric.

I felt it prudent, to inform the OP that the stellar collapse process is a catastrophic one, in order to distance him from the idea that stellar collapse happens starting from one tiny point and slowly growing outwards. But if there is a problem with this idea, then be sure to correct me. :)

Bear in mind that I was only talking about the event horizon, not about the entire process of collapse. I agree that the collapse process does not start at a single point and grow outwards. Only the EH does.

 

[Matterwave]

Dropping the assumption of spherical symmetry certainly makes things more complicated; AFAIK there are no analytical solutions for the general case, only numerical simulations. But AFAIK that does not change the qualitative features that I described. Bear in mind that the event horizon is the boundary of the region of spacetime that can't send light signals to future null infinity; qualitatively, such a region *has* to start with a single point (more precisely, there has to be some earliest spacelike hypersurface that the EH intersects, and it must intersect that hypersurface at a single point), even if the details of the process are not spherically symmetric.

But if the process is not necessarily spherically symmetric, which point would the universe know to choose to begin creating an EH? Wouldn't it be plausible that at many different places in the core, you have densities high enough to create multiple EHs and then they merge to form a large black hole?

But anyways, my point was only that the supernova process is a very chaotic one. And one should not so quickly jump to conclusions based on very nice initial conditions.

 

[Soule]

But if the process is not necessarily spherically symmetric, which point would the universe know to choose to begin creating an EH? Wouldn't it be plausible that at many different places in the core, you have densities high enough to create multiple EHs and then they merge to form a large black hole?

I was thinking that too >.> lol.

 

[PeterDonis]

But if the process is not necessarily spherically symmetric, which point would the universe know to choose to begin creating an EH?

It's not a question of "choosing to create an EH". The EH is globally defined; it's the boundary of the region that can't send light signals to future null infinity. That boundary must be a null surface, i.e., a surface generated by light rays. And for a single collapsing object (this is to distinguish from the case of multiple black holes merging--see below), the null surface forming the boundary must intersect any spacelike hypersurface in either a point, or a 2-surface; and given a slicing of spacetime into spacelike hypersurfaces, the hypersurface which the EH boundary (of the single collapsing object) intersects in a point must be to the past of any hypersurfaces in the same slicing that the EH intersects in a 2-surface.

Wouldn't it be plausible that at many different places in the core, you have densities high enough to create multiple EHs and then they merge to form a large black hole?

Yes, that's true, and I should have clarified that my statement was talking about each black hole individually, not about the set of all black holes in the universe. In other words, each black hole individually starts out with its own "section" of event horizon that works as I described above; later on, the "sections" of EH associated with different black holes can merge, but I wasn't intending to describe the merge process (it must be to the future of each of the individual sections anyway, so it doesn't invalidate what I said above).

my point was only that the supernova process is a very chaotic one. And one should not so quickly jump to conclusions based on very nice initial conditions.

Nothing that I stated depends on any symmetry in the initial conditions. It's purely a consequence of global geometric facts about *any* null surface and *any* set of spacelike hypersurfaces. It's easier to visualize for a highly symmetric collapse, but that's all.