16 pages

Please download to get full document.

View again

of 16
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
    a  r   X   i  v  :  g  r  -  q  c   /   9   3   0   7   0   3   8  v   1   2   9   J  u   l   1   9   9   3 Black Hole Entropy is Noether Charge Robert M. Wald University of ChicagoEnrico Fermi Institute and Department of Physics 5640 S. Ellis Avenue Chicago, Illinois 60637-1433  Abstract We consider a general, classical theory of gravity in  n  dimensions, arising froma diffeomorphism invariant Lagrangian. In any such theory, to each vector field, ξ  a , on spacetime one can associate a local symmetry and, hence, a Noether current( n − 1)-form,  j , and (for solutions to the field equations) a Noether charge ( n − 2)-form,  Q , both of which are locally constructed from  ξ  a and the the fields appearingin the Lagrangian. Assuming only that the theory admits stationary black holesolutions with a bifurcate Killing horizon (with bifurcation surface Σ), and thatthe canonical mass and angular momentum of solutions are well defined at infinity,we show that the first law of black hole mechanics always holds for perturbations tonearby stationary black hole solutions. The quantity playing the role of black holeentropy in this formula is simply 2 π  times the integral over Σ of the Noether charge( n − 2)-form associated with the horizon Killing field (i.e., the Killing field whichvanishes on Σ), normalized so as to have unit surface gravity. Furthermore, weshow that this black hole entropy always is given by a local geometrical expressionon the horizon of the black hole. We thereby obtain a natural candidate for theentropy of a dynamical black hole in a general theory of gravity. Our results showthat the validity of the “second law” of black hole mechanics in dynamical evolutionfrom an initially stationary black hole to a final stationary state is equivalent to 1  the positivity of a total Noether flux, and thus may be intimately related to thepositive energy properties of the theory. The relationship between the derivationof our formula for black hole entropy and the derivation via “Euclidean methods”also is explained. PACS #:  04.20.-q, 97.60.Lf  2  One of the most remarkable developments in the theory of black holes in classicalgeneral relativity was the discovery of a close mathematical analogy between certainlaws of “black hole mechanics” and the ordinary laws of thermodynamics. When theeffects of quantum particle creation by black holes [1] were taken into account, thisanalogy was seen to be of a physical nature, and it has given rise to some deep insightsinto phenomena which may be expected to occur in a quantum theory of gravity.The srcinal derivation of the laws of black hole mechanics in classical general relativ-ity [2] used many detailed properties of the Einstein field equations, and, thus, appearedto be very special to general relativity. However, recently it has become clear that atleast some of the laws of classical black hole mechanics hold in a much more generalcontext. In particular, it has been shown that a version of the first law of black holemechanics holds in any theory of gravity derivable from a Hamiltonian [3]. (For the casesof (1 + 1)-dimensional theories of gravity [4] and Lovelock gravity [5], the explicit forms of this law have been given.) Furthermore, analogs of all of the classical laws of blackhole mechanics have been shown to hold in (1 + 1)-dimensional theories [4].However, despite the very general nature of the Hamiltonian derivation [3] of the firstlaw of black hole mechanics, there remains one unsatisfactory aspect of the status of the first law in a general theory of gravity [6]: Although the derivation shows that fora perturbation of a stationary black hole, a surface integral at the black hole horizon(involving the unperturbed metric and its variation) is equal to terms involving thevariation of mass and angular momentum (and possibly other asymptotic quantities)at infinity, the derivation does not show that this surface term at the horizon can beexpressed as  κ/ 2 π  (where  κ  denotes the unperturbed surface gravity) times the variationof a surface integral of the form  S   =    Σ F  , where  F   is locally constructed out of themetric and other dynamical fields appearing in the theory. It is necessary that thehorizon surface term be expressible in this form in order to be able to identify a local,geometrical quantity,  S  , as playing the role of the entropy of the black hole.The main purpose of this paper is to remedy this deficiency by showing that ina general theory of gravity derivable from a Lagrangian, the form of the first law of black hole mechanics for perturbations to nearby stationary black holes is such that the3  surface term at the horizon always takes the form  κ 2 π δS  , where  S   is a local geometricalquantity, and is equal to 2 π  times the Noether charge at the horizon of the horizon Killingfield (normalized so as to have unit surface gravity). The local, geometrical characterof   S   suggests a possible generalization of the definition of entropy to dynamical blackholes. The relationship between black hole entropy and Noether charge also suggests thepossibility of a general relationship between the validity of the second law of black holemechanics (i.e., increase of black hole entropy) and positive energy properties of a theory.An additional byproduct of our analysis is that it will enable us to make contact withthe “Euclidean derivation” of formulas for black hole entropy, thereby demonstratingequivalence of that approach with other approaches – a fact that is not at all easy to seeby a direct comparison of, say, references [3] and [7]. Our considerations in this paper will be limited to a general analysis of all the above issues; applications to particulartheories will be given elsewhere [8]Before presenting our new derivation of the first law, we comment upon the status of other “preliminary laws” of black hole mechanics in a general theory of gravity. We con-sider theories defined on an  n -dimensional manifold  M   with dynamical fields consistingof a (Lorentzian) spacetime metric,  g ab , and possibly other matter fields, such that theequations of motion of the metric and other fields are derivable from a diffeomorphism in-variant Lagrangian. (Our precise assumptions concerning the Lagrangian will be spelledout in more detail below.) We assume that a suitable notion of “asymptotic flatness” isdefined in the theory. The  black hole   region of an asymptotically flat spacetime then isdefined to be the complement of the past of the asymptotic region. In order to beginconsideration of the classical laws of black hole mechanics, it is necessary that the eventhorizon of a stationary black hole be a Killing horizon, i.e., a null surface to which aKilling vector field is normal. This property is known to be true in general relativityby a nontrivial argument using the null initial value formulation [9], so it is not obviousthat it would hold in more general theories of gravity. Nevertheless, this property auto-matically holds for all static black holes (since the static Killing field must be normal tothe event horizon of the black hole), and, hence, it automatically holds for sphericallysymmetric black holes (in the  O ( n − 1) sense) and, thus, in particular, for all black holes4
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!