Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Backreaction from Hawking radiation may prevent formation of event horizon?

  1. Sep 24, 2014 #1

    bcrowell

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I'm trying to understand the ideas in this paper at a nontechnical level:

    Laura Mersini-Houghton, "Backreaction of Hawking Radiation on a Gravitationally Collapsing Star I: Black Holes?," http://arxiv.org/abs/1406.1525

    She says:

    This work investigates the backreaction of Hawking radiation on the interior of a gravitationally collapsing star, in a Hartle-Hawking initial vacuum. It shows that due to the negative energy Hawking radiation in the interior, the collapse of the star stops at a finite radius, before the singularity and the event horizon of a black hole have a chance to form.
    I'm just baffled by this claim. For an astrophysical black hole, the Hawking radiation is the most incredibly tenuous thing you could possibly imagine -- utterly undetectable by any foreseeable technology. How, then, can it cause a back-reaction so vigorous as to prevent the formation of an event horizon?

    In general, I can't overcome a feeling of disbelief in semiclassical gravity. It seems like they get all kinds of gee-whiz results that are just clearly not reasonable, and that should be interpreted simply as a sign that semiclassical gravity doesn't work. In particular, semiclassical gravity seems to predict spectacular stuff happening near the event horizons of black holes. They even get divergences that need to be renormalized away. But for a black hole of mass M, I can make the curvature of spacetime at the horizon as small as desired by making M big enough. Therefore I have a hard time believing any prediction that something special happens at the event horizon, if the prediction is independent of M.
     
  2. jcsd
  3. Sep 24, 2014 #2

    Haelfix

    User Avatar
    Science Advisor

    Suffice it to say, that you are not the only one who is skeptical about the idea that blackholes don't form. I still find Bill Unruh's comments on this the most emminently sensible (first five minutes of below):

    http://online.kitp.ucsb.edu/online/fuzzorfire-m13/unruh/

    As for the loss of coherence at the horizons of blackholes, well that's another problem and that does seem to require something drastic happening.
     
  4. Sep 24, 2014 #3

    bcrowell

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    More comments by Unruh:

    “The [paper] is nonsense,” Unruh said in an email to IFLS. “Attempts like this to show that black holes never form have a very long history, and this is only the latest. They all misunderstand Hawking radiation, and assume that matter behaves in ways that are completely implausible.”

    According to Unruh, black holes don’t emit enough Hawking radiation to shrink the mass of the black hole down to where Mersini-Houghton claims in a timely manner. Instead, “it would take 10^53 (1 followed by 53 zeros) times the age of the universe to evaporate,” he explains.

    “The standard behaviour by such people [who don’t understand Hawking radiation] is to project that outgoing energy back closer and closer to the horizon of the black hole, where its energy density gets larger and larger,” he continued. “Unfortunately explicit calculations of the energy density near the horizon show it is really, really small instead of being large-- Those calculations were already done in the 1970s. To call bad speculation "has been proven mathematically" is, shall we say, and overstatement.”
    From http://www.iflscience.com/physics/p...roven-mathematically-black-holes-do-not-exist
     
  5. Sep 24, 2014 #4

    MTd2

    User Avatar
    Gold Member

    This is the paper that got attention from the press and not the one from june, but this one:
    http://arxiv.org/pdf/1409.1837v1.pdf
    I am skeptical of the formation of black holes and it is indeed because of the reason Uruh said concerning "the standard behavior of such people [who don’t understand Hawking radiation]". There are sorts of assumptions that can be done, with a variety of different "understandings" of the origin of Hawking radiation, with a great variety of final states (fuzzyballis a famous one). So, explicit calculations "already done in the 1970s" is a also claim that cannot be said conclusive at all. There are many approximation of theories of semi classical quantum gravity. For example, this latter paper tackles issue of firewalls.
     
  6. Sep 24, 2014 #5

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    Here is a relatively well known paper explicitly examining backreaction from Hawking radiation, and coming to conclusions supporting Unruh's point of view. I find it rather interesting that the two papers cited in this thread completely ignore this work. It would seem they would need explain the discrepancy.

    http://arxiv.org/abs/0906.1768
     
  7. Sep 25, 2014 #6

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    The calculation scheme in the new papers seems much more ambitious than the Padmanabhon paper of 2009. The latter used shells and 1+1 computation for simplicity. The new papers use a 'realistic' ball model, and considers the full 4-D problem.
     
  8. Sep 26, 2014 #7

    The argument arises from the claim that the rate of change of the gravitational field is so violent that when the star collapses that the Hawking radiation is much stronger than what we expect from an existing black hole. Typically, when the detection of Hawking radiation has been considererd previously, it's been from existing black holes. If she's right, presumably it should be much more feasible to detect the Hawking radiation from a collapsing star, though we might have to wait a long time for a detectable event.
     
    Last edited: Sep 26, 2014
  9. Sep 26, 2014 #8

    PAllen

    User Avatar
    Science Advisor
    Gold Member

  10. Sep 26, 2014 #9

    PAllen

    User Avatar
    Science Advisor
    Gold Member

  11. Sep 26, 2014 #10

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    Ok, now for my own first substantive thought on these papers:

    On the plus side, I really like that they don't use shells and that they use 4-d. On the negative side (pun intended) I am increasingly concerned about use of a negative energy SET component. Precisely because they model all Hawking radiation as being generated, locally, BEFORE formation of the event horizon, negative energy should be unnecessary. My understanding of why Hawking radiation violates energy conditions is that you have an outflow of positive energy across an event horizon, which is a spacelike energy flow. If, instead, you have delayed reception at infinity of radiation that need not violate causality, why use negative energy at all??!!

    A caveat to my concern is that their use of negative energy might just be a bookeeping trick, and the T (total) as they define it maybe never violates DEC.
     
  12. Sep 29, 2014 #11

    bcrowell

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I don't see how it can be true that the rate of change of the gravitational field is violent for all black holes, at the location where the event horizon would be expected to form. First off, we would have to agree on some measure of gravitational field. Presumably that would have to be something coordinate-independent, so it can't be the acceleration that we refer to as a gravitational field in Newtonian mechanics (or, in GR terms, the proper acceleration of a stationary observer). I guess the measure of gravitational field would have to be some measure of curvature. But the curvature can be made as small as desired by making the mass of the black hole big enough. I think the curvature's time rate of change can also be made arbitrarily small, since the only thing that sets a time-scale for the collapse is the black hole's mass (this is the only parameter available). In geometrized units, a larger mass is equivalent to a longer time.

    This quote from Sabine Hossenfelder's blog at http://backreaction.blogspot.com/2014/09/black-holes-declared-non-existent-again.html seems to be getting at the issue I'm talking about:

     
    Last edited: Sep 29, 2014
  13. Sep 29, 2014 #12

    PAllen

    User Avatar
    Science Advisor
    Gold Member

    This paper by George Ellis has some discussion of the first Mersini-Houghton paper. It seems to me the discussion carries over to the newer paper.

    http://arxiv.org/abs/1407.3577
     
    Last edited: Sep 30, 2014
  14. Sep 30, 2014 #13

    aleazk

    User Avatar
    Gold Member

    Thanks guys for all these links!

    I read the 2009 Padmanabhan paper first. Although my knowledge of QFT in curved spacetime is based only on a first and quick reading of Wald's homonymous book, I can at least follow the thing in its basic concepts. I found the Padmanabhan very clear and explicit in, precisely, the difficult and tricky/non-trivial considerations.

    Then I read the M-H paper and I found it confusing and unclear in some parts. Basically, I have the same concern as the one expressed by Sabine Hossenfelder in her blog, i.e., they seem to use the stress-energy from the static situation rather than the correct expression. She (Hossenfelder) also describes all the situation in a very clear way, on the same lines as Padmanabhan.

    Anyway, was fun. It helped me for having more practice in this topic, which I'm just starting to study.
     
  15. Oct 11, 2014 #14
    The way I understand it is, from the frame of a distant observer, it's not the changing curvature, where GR predicts the event horizon would form, rather the changing curvature closer to where it predicts the singlarity would form, before the horizon has formed that is expected to give rise the large amounts of radiation.

    The most interesting thing about this for me, is that the strongest curvature is not necessairily where the largest density of matter is. I think this is what Unruh is referring to when he vaguely suggests that these types of papers have matter behaving in unplausible ways.
     
  16. Oct 11, 2014 #15
    The paper has cranky title and doesn't deserve my reading :s
     
  17. Oct 17, 2014 #16
    I'm new here but here I go...

    Ok folks, listen. We all know that black holes are real. We know that there is one at the center of every galaxy including our own. We know that when an object becomes smaller than its Schwarzschild radius, a black hole is formed. We also all know there is no way to know what is going on inside of the event horizon.

    And I think that most of us suspect there is no such thing as a singularity -- it just means "we don't know."

    Anyway, most smart people in physics these days are starting to believe that if we passed through an event horizon, we would see nothing special.
     
  18. Nov 7, 2014 #17

    jambaugh

    User Avatar
    Science Advisor
    Gold Member

    True. In fact we pass through event horizons continuously (the world-sheets corresponding to t=const are event horizons as are the forward and backward light-cones of an event point). The form of a Schwarzchild BH is that there is a spherically symmetric stationary event horizon, what we typically refer to as "the event horizon".

    However we can pretty well describe the singularity interior to the classical black hole. Extending beyond that horizon the interior solution remains spherically symmetric in the coordinates but the "r" coordinate ceases to be spatial, rather a time-like coordinate. It is the "time until singularity". The interior shape might better be described as a collapsing cylindrical space approaching zero radius as time r goes to 0. The singularity is not in space but in time.

    My reflexive objection to the claims in the paper (I haven't read) are that the "formation" of an event horizon is semantically meaningless. Event horizons do not "form" and they are not physical objects. One would rather observe over time that the event horizons corresponding to future light cones of events at the center will bend more and more into "light cylinders". This phenomenon is macroscopic and I don't see how it could be disrupted by some esoteric quantum effect at the center of the collapsing star. I don't see a quantum "out" for massive star collapse into a BH.
     
  19. Nov 17, 2014 #18
    Once there is outgoing thermal radiation, it takes away energy. This energy has to come from somewhere. The obvious candidate is the black hole itself. And the way it comes from the black hole has to be causal, not? Thus, we wait (conceptually - it does not matter at all how long it takes, no number of zeros matters here) until some amount of energy has been radiated away, and, using Einstein causality, look where it possibly comes from. And this means it all comes from the collapsing body before horizon formation. I see no way in which one can prevent the conclusion that there has to be a firewall, except that the basic claim of Hawking radiation having a fixed temperature is wrong.

    The computations which claim that there is nothing special near the horizon seem more plausible to me, thus, it seems to me that the basic Hawking radiation formula has to be wrong.

    Of course, there is the "nothing special" argument - but it depends on the equivalence principle, and that it holds even for extremal, essentially infinite, time dilations.

    So I see here a contradiction between the EEP, which sees nothing special near the horizon, energy conservation.

    Such a contradiction should not be something unexpected, given that we have no true local energy conservation in GR, we have only pseudotensors, in other words, a form of energy conservation which is not compatible with the EEP.
     
  20. Nov 17, 2014 #19
    To quote this paper:
    But, sorry, $L_H$ is the luminosity of the BH, the Hawking radiation, far away but constant in far away (Schwarzschild) time. And the very point of all the ideas about firewalls is that all this radiation, however small it is, raises to large values near the horizon at the moment immediately before horizon formation. The radiation near the horizon formation should be much greater to obtain a however small but constant in time luminosity far away.

    So, I would (possibly naively) interpret this passage as "now we use the smallness of Hawking radiation far away to make the assumtion that there is no firewall.". With the consequence:

    Sorry, no. The delay in horizon formation is governed by the integral of $L_H(t)$ in external time $t$, which defines the energy which is radiated away. And this integral becomes infinite if $L_H(t)$ is however small but constant, and it comes from the collapsing surface before horizon formation, simply because we can always trace back all the outgoing radiation to that surface, and it reaches this surface by definition of the horizon before horizon formation (because it is outgoing radiation).
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Backreaction from Hawking radiation may prevent formation of event horizon?
  1. Hawking Radiation (Replies: 6)

Loading...