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

What are firewalls and why do they think they exist?

  1. Aug 22, 2014 #1
    I watched a lecture by Polchinski on youtube where he discussed firewalls that exist around blackholes. He said explains that if information is not lost in a black hole then entanglements must be broken. Just as energy is released when chemical bonds are broken, energy is released when entanglements are broken, resulting in a firewall which vaporizes anything getting close to the black hole.

    So a few questions:
    1) Why must be entanglements be broken for information to be conserved? Can anyone give me a basic example to illustrate what i means for an entanglement to be broken?
    2) Why would the existence of firewalls mean that Einstein's theory of relativity could be incorrect?
    3) Did I make any mistakes in my attempt at paraphrasing Polchinski?
  2. jcsd
  3. Aug 23, 2014 #2


    User Avatar

    Staff: Mentor

    There's a wikipedia article at http://en.m.wikipedia.org/wiki/Firewall_(physics [Broken])

    The very first sentence says that they are "hypothetical" - it's by not means agreed that they exist.
    Last edited by a moderator: May 6, 2017
  4. Aug 23, 2014 #3


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Salamon, there have been a number of papers this year suggesting different ways to resolve the information-loss problem and avoid having to postulate a firewall.

    Getting rid of firewall and information-loss seems to be something that attracts researchers. These problems represent what are seen as flaws in the theory needing to be fixed by improving our model of BH.

    There are too many different ideas for how to improve the BH model or otherwise avoid the ugly conclusions. I can't give you a list.

    The one I find most interesting is the Planck Star proposal by Carlo Rovelli and collaborators.

    He will be giving an online talk about this BH model on 14 October which you can listen to if you want. He has also been invited to speak about it at a couple of conferences in September, but I'm not sure those talks will be put on line. And there are several papers available. Let me know if you want links.

    The idea uses gravitational time dilation---the fact that from the standpoint of an outside observer the intense gravity in a deep gravity well makes processes slow down. any kind of clock slows way down in the heart of a BH.
    And according to some proposed quantum gravity (QG) theories quantum effects at extreme density cause the singularity to be replaced by a bounce (quantum nature resists being pinned down :smile:).

    Because of extreme time dilation this bounce is delayed by billions of years (as seen by outsiders). When it eventually breaks out of the time-slowed core region it results in a type of brief intense flash called a gamma ray burst (GRB).

    So there is something for astronomers to look for. This theory, or BH model, predicts something. It also gets rid of the need for firewalls and resolves the paradoxes that plague the earlier Hawking BH evaporation model. They don't evaporate. they explode.

    The explosion releases the information which was previously thought to be lost. If you google "planck star" you'll find various news articles and technical papers about it. Ask if you want links. We also have one or two threads about this idea here at the forum.
  5. Aug 24, 2014 #4
    Thanks a lot Marcus for your response.

    I get what you wrote about time dilation but wouldn't the singularity or "bounce" be hidden from the outside observer since it lies beyond the event horizon?

    Also, is Rovelli saying that Hawking radiation doesn't exist at all?
  6. Aug 24, 2014 #5


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    According to their analysis, the "event horizon" is temporary. A temporary horizon is called a "trapping horizon". It's a feature of the geometry where the lightcones tilt inwards and it can be undone depending on what happens inside. The bounce is an upheaval of geometry which can tilt the lightcones back outwards, when it finally reaches the horizon. So it bursts through the trapping horizon and the BH ends in a final explosion.

    google "Planck star" and take a look at the non-technical parts of the article by Rovelli and Vidotto. they have some diagrams.

    They certainly don't say "Hawking radiation doesn't exist" :biggrin: In fact, as they point out, the Planck star model BH looks just like a conventional Hawking BH for roughly the first half of its life.
    The trapping horizon works effectively like the conventional event horizon (same role in Hawking radiation) the main difference being that in the very long term it is not permanent, because the bounce eventually destroys it.

    You are better off, I think, looking at the papers directly rather than getting me to paraphrase. Please at least give it a try. See if you get the Rovelli Vidotto paper by googling "planck star". If that does not work, tell me and I will get the link to that and to some other papers.

    On a totally different tangent. Remember that this is only one! There are a bunch of different ways people are working on that get rid of the "firewall". It may be a phony problem, a false alarm. Today this paper appeared. I'm not evaluating it or saying its right. It's just symptomatic. A lot of researchers are suspicious of "firewall" and "BH information loss problem" and are trying various ways to get rid of them. Also various proposed solutions get rid of the "singularity". Here's another straw in the wind:
    Here is one of the authors, Andreas Albrecht, profile:
    Mind you I'm not recommending this Albrecht paper or saying it's right, but I see the guy has a very respectable track record. Has co-authored with outstanding people, has an unusually high average number of citations per paper (his papers tend to be cited a lot by other researchers).
    Last edited: Aug 24, 2014
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook