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

Rotating collapse

  1. Nov 21, 2006 #1


    User Avatar
    Staff Emeritus
    Science Advisor

    This may be a repeat, but I don't think the earlier shorter post survived the recent PF upgrade, at least I couldn't find it. This is a better post anyway :-).

    I've been trying to understand the Poisson-Israel model of a charged black hole recently. (The model is also expected to provide insights into rotating black holes as well as charged ones, as I understand it).


    (the full text unfortunately requires one to visit a library with access)

    I'd like to be able to answer simple questions such as "what is the fate of an observer who falls into such a hole?

    Does he get spaghettified? http://en.wikipedia.org/wiki/Spaghettification

    Does he get fried by infinite radiation, crushed by infinite pressures, or see the fate of the universe?

    Some specificity may be needed to answer these questions, but at the moment I'm leaving the exact details of said black hole rather vague. Ultimately, I'm trying to generally understand the fate of someone falling into an actual rotating black hole like the one in the center of our galaxy, but I'm more than willing to consider simpler cases first.

    I'd also like to better understand what the authors mean by "mass inflation".

    So far I'm having a hard time even figuring out the metric. Using Schwarzschild coordinates (r,t) I gather that

    [tex]g^{rr} = 1 - 2m(r,t) / r + e^2/r^2[/tex]

    m(r,t) appears to be the much-talked about "mass function", and it appears to become infinite (?!) near the inner horizon.

    Because r is timelike inside the event horizon, this should be the most interesting metric coefficient, but I'm not sure how one goes about finding the other metric coefficients (other than by redoing the calculations).

    Other related questions (this is probably already too long) - how does one justify the model of "backscattering" used to determine how some small component of infalling radiation gets "backscattered" into outwards going radiation?
  2. jcsd
  3. Nov 22, 2006 #2

    Chris Hillman

    User Avatar
    Science Advisor

    Mass inflation

    Hi, pervect,

    In addition to PF instability, I have been troubled by instability in the electric power grid, so I was forced to enter standby mode for a day or so. I did see your question about CTCs in Kerr, but had to log off just before answering. I'll try to find that again and answer at greater length, but the short answer is that the CTC's in the Kerr vacuum occur in the deep interior (inside the blocks of the Carter-Penrose diagrams which include the timelike curvature singularities). There is an ongoing debate in CQG between Bonnor and some others over whether or not CTCs occur in vacuum solutions in places where their effects could in principle be observed by distant physicists: my short answer is that while there is no doubt that many simple exact solutions suggest the disturbing answer "yes", in my view and that of the physicists arguing with Bonnor, these solutions, which exhibit "massless struts" and so on, are artifacts of inappropriate boundary conditions and thus cannot be regarded as physically acceptable.

    Yes, based upon a mathematical analogy.

    Yes, but there used to be a website with an html version. I don't have that paper in front of me but I'll refer to Israel's article in the book Black Holes and Relativistic Stars, edited by Wald (aka the Chandrasekhar memorial volume). Another reference: the book Black Hole Physics by Frolov and Novikov extensively discussed mass inflation.

    Those with access to none of these can refer to the crude ASCII sketch of the appropriate "Carter-Penrose conformal block diagram" offered toward the bottom of this webpage: http://www.math.ucr.edu/home/baez/RelWWW/history.html [Broken]

    It depends on how he falls in. If he avoids the spacelike (not timelike!) curvature singularity, he must encounter a Cauchy horizon, and gtr can't predict what happens after that. Interestingly enough, as with some exact gravitational plane wave solutions discussed in an important paper by Ellis and Schmidt, such an observer might observe a divergence of tidal forces, but these might occur so quickly that his body, modeled an elastic solid, might not have time to distort enough to harm him before he passes through the danger region. This is the same place where one might worry about him being fried by EM radiation from starlight falling in from the external universe, etc.

    That's the question asked by Penrose. This is a topic of current research interest in classical gravitation.

    Not if he avoids the spacelike curvature singularity (which is "crushing" in terms of volume as well as "sphaghetifying" in terms of shape).

    Well, the diagram I offered doesn't attempt to account for Lambda if that's what you mean. If you sketch the analogous picture starting from a Schwarzchild-de Sitter model, you can see the answer you probably want should be "no, he can't see any events lying above one of the cosmological horizons, hence he can't see the fate of most stuff in the universe".

    Well, this a topic of current research interest. As I mentioned in some other threads, Chandrasekhar and his students Valeria Ferrari and Basilis Xanthopolous (who was tragically murdered by a disgruntled student in Greece, thus cutting short a brilliant career) discovered a beautiful duality between certain colliding plane wave solutions and models of black hole interiors. Strictly speaking, this yields a local isometry of part of the black hole "interior region" (think of roughly [tex]m < r < 2 m[/tex] with the "interaction zone region" in a CPW model, but this suffices to study the question of whether the Cauchy horizon corresponds to a survivable geometric singularity or not. Several researchers have in fact constructed CPW models to model the effect of infalling incoherent radiation and so on. Some of these models suggest that the answer is "yes"; others, "no".

    Another major goal is to try to understand how perturbations (due to infalling matter and radiation) can (probably) change the avoidable timelike singularities of the Kerr deep interior into a strong spacelike scalar singularity plus something at this Cauchy horizon.

    I'd also like to better understand what the authors mean by "mass inflation".

    In the perturbed Reissner-Nordstrom model of an interior, the mass function is defined via the gradient of the "radial coordinate" (which we can treat as a monotonic scalar field):
    [tex] | \nabla r |^2 = 1 - 2 m(t,r)/r + e^2/r^2 [/tex]
    The field equations reduce to two coupled equations and lead to a nonlinear wave equation for m. For a null dust influx (e.g. incoherent EM radiation), Poisson and Israel concluded that m(t,r) diverges near the Cauchy horizon, which is known as "mass inflation". By a mathematical analogy of charge with angular momentum (compare Carter-Penrose diagrams for Kerr and RN with Schwarzschild), one expects something similar to happen with perturbations of Kerr due to influx of null dust.

    The very extensive theory of perturbations of black holes (another major achievement largely due to Chandrasekhar, building upon work by Wheeler, Teukolsky, and others).

    Chris Hillman
    Last edited by a moderator: May 2, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook