Добавил:
Upload Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:
Diss / 6 - 753p.pdf
Скачиваний:
48
Добавлен:
27.03.2016
Размер:
15.34 Mб
Скачать

Black Hole Entropy Function and Duality

Gabriel Lopes Cardoso

Abstract The macroscopic entropy and the attractor equations for extremal black hole solutions follow from a variational principle based on an entropy function. We review this variational principle for static extremal black holes in four space-time dimensions and we apply it to N = 2 supergravity theories with higher-curvature interactions.

1 Introduction

String theory provides a microscopic derivation of the macroscopic BekensteinHawking entropy of certain supersymmetric black holes [21]. An important feature of supersymmetric black holes is that they are charged and are supported by scalar fields. In the black hole background these scalar fields vary radially as one moves from spatial infinity to the horizon of the black hole, and they get attracted to specific values at the horizon which are determined by the black hole charges. This is the so-called attractor mechanism, which was first studied in [10, 20, 9, 8]. As a result, the macroscopic entropy is entirely determined in terms of the black hole charges and can be compared with the microscopic entropy based on state counting. The attractor mechanism is, however, not just a feature of supersymmetric black holes, but is also present for extremal non-supersymmetric black holes [11, 12, 18, 13]. The attractor behaviour is encoded in a set of attractor equations, which can be obtained by extremizing a so-called entropy function [18]. In addition, the value of this function at the extremum yields the macroscopic entropy of the extremal black hole.

We review the entropy function for static extremal black hole solutions in four space-time dimensions using the approach of [5] based on electric/magnetic duality

Gabriel Lopes Cardoso

Arnold Sommerfeld Center for Theoretical Physics, Department für Physik, Ludwig-Maximilians- Universität München, Munich, Germany, e-mail: gcardoso@theorie.physik.uni-muenchen.de

V. Sidoraviciusˇ (ed.), New Trends in Mathematical Physics,

127

© Springer Science + Business Media B.V. 2009

 

128

Gabriel Lopes Cardoso

covariance. Then, following [17], we specialize to the case of N = 2 supergravity theories with higher-curvature interactions. We also review the construction of a duality invariant OSV integral formula for supersymmetric black holes [4]. We refer to [5, 4] for a detailed discussion of these topics.

2 Entropy Function and Electric/Magnetic Duality Covariance

Let us consider static extremal black hole solutions to the equations of motion of a general system of Abelian vector gauge fields, scalar and matter fields coupled to gravity in four space-time dimensions. Following [18], we take the near-horizon geometry of such a black hole to be of the form AdS2 × S2. Thus, we consider near-horizon solutions with spherical symmetry, which may be written as

ds(24) = gμν dxμdxν = v1

r2 dt 2 +

dr2

+ v2

dθ 2 + sin2 θ dϕ2 ,

 

r2

(1)

FrIt = eI , FθIϕ =

pI

 

sin θ .

 

 

 

 

4π

 

 

 

 

Here the FμνI denote the field strengths associated with a number of Abelian gauge fields. The θ -dependence of FθIϕ is fixed by rotational invariance and the pI denote the magnetic charges. The fields eI are dual to the electric charges. In addition to the constant fields eI , v1 and v2 there may be a number of other fields which for the moment we denote collectively by uα .

As is well known theories based on Abelian vector fields are subject to electric/magnetic duality, because their equations of motion expressed in terms of the dual field strengths,1

 

=

 

| |

 

∂FρσI

(2)

GμνI

 

 

g

εμνρσ

∂L ,

take the same form as the Bianchi identities for the field strengths FμνI . Adopting the conventions where xμ = (t, r, θ , ϕ) and εt rθ ϕ = 1, and the signature of the space-time metric equals (, +, +, +), it follows that, in the background (1),

G

v

v

 

sin θ

∂L

v

v

 

sin θ

∂L

,

θ ϕI = − 1

 

2

 

∂FrIt

= −

1

 

2

 

 

 

∂eI

(3)

 

 

 

 

 

∂L

 

 

 

 

 

 

 

∂L

 

G

v

v

 

sin θ

= −

4π v

v

 

.

 

 

rt I = − 1

 

2

 

∂FθIϕ

 

 

 

1

 

2 ∂pI

 

1 Here and henceforth we assume that the Lagrangian depends on the Abelian field strengths but not on their space-time derivatives. We also assume that the gauge fields appear exclusively through their field strengths.

Black Hole Entropy Function and Duality

129

These two tensors can be written as qI sin θ /(4π ) and fI . The quantities qI are conjugate to pI and eI , respectively, and can be written as

∂L qI (e, p, v, u) = −4π v1v2 ∂eI ,

∂L fI (e, p, v, u) = −4π v1v2 ∂pI .

and fI

(4)

The qI are constant by virtue of the equations of motion and correspond to the electric charges. Electric/magnetic duality transformations are induced by rotating the tensors FμνI and Gμν I by a constant transformation, so that the new linear combina-

tions are all subject to Bianchi identities. This leads to new quantities (p˜I , q˜I ) and

(e˜I , f˜I ).

Next, we define the reduced Lagrangian by the integral of the full Lagrangian over the horizon two-sphere S2,

F (e, p, v, u) = dθ dϕ |g|L. (5)

We note that the definition of the conjugate quantities qI and fI takes the form,

q

I = −

∂F

, f

I = −

∂F

.

(6)

 

∂eI

 

∂pI

 

It is known that a Lagrangian does not transform as a function under electric/magnetic dualities. Instead we have [6],

˜

 

 

1

 

I

 

˜

 

 

 

 

1

 

 

 

I

 

 

 

 

(e, p, v, u)

+

 

eI q

 

+

fI pI

] = F

(e, p, v, u)

+

 

[

eI qI

+

f

pI

]

(7)

F ˜ ˜

2

[ ˜ ˜

 

 

˜

 

2

 

 

 

 

so that the linear combination F (e, p, v, u) + 12 [eI qI + fI pI ] transforms as a function. It is easy to see that the combination eI qI fI pI transforms as a function as well, so that we may construct a modification of (5) that no longer involves the fI and that transforms as a function under electric/magnetic duality,

E (q, p, v, u) = −F (e, p, v, u) eI qI ,

(8)

which takes the form of a Legendre transform in view of the first equation (6). In this way we obtain a function of electric and magnetic charges. Therefore it trans-

E ˜ ˜

= E

(q, p, v, u).

forms under electric/magnetic duality according to ˜(q, p, v, u)

 

Furthermore the field equations imply that the qI are constant and that the action,dt dr E , is stationary under variations of the fields v and u, while keeping the pI and qI fixed. This is to be expected as E is in fact the analogue of the Hamiltonian density associated with the reduced Lagrangian density (5), at least as far as the vector fields are concerned. The constant values of the fields v1,2 and uα are thus determined by demanding E to be stationary under variations of v and u,

130

 

 

 

 

 

 

Gabriel Lopes Cardoso

 

∂E

=

 

∂E

=

0.

(9)

 

∂v

∂u

 

 

 

 

The function 2π E (q, p, v, u) coincides with the entropy function proposed by Sen [18]. Equation (9) is the so-called attractor equation and the macroscopic entropy is directly proportional to the value of E at the stationary point,

Smacro(p, q) E

attractor.

(10)

 

 

 

The above derivation of the entropy function applies to any gauge and general coordinate invariant Lagrangian, and, in particular, also to Lagrangians containing higher-derivative interactions. The entropy computed by (10) is Wald’s entropy [22, 14, 15] which, in the absence of higher-derivative interactions, reduces to the area law of Bekenstein and Hawking.

In the absence of higher-derivative terms, the reduced Lagrangian F is at most quadratic in eI and pI and the Legendre transform (8) can easily be carried out. For instance, consider the following Lagrangian in four space-time dimensions (we only concentrate on terms quadratic in the field strengths),

L0 = −

1

NI J Fμν+I F

+μνJ N¯ I J FμνI F μνJ ,

(11)

4

where Fμν±I denote the (anti)-selfdual field strengths. In the context of this paper the tensors Fr±t I = ±iFθ±ϕI = 12 (FrIt ± iFθIϕ ) are relevant, where underlined indices refer

to the tangent space. It is straightforward to evaluate the entropy function (8) in this case,

 

v1

(qI

N

I K pK ) (Im

 

)1

]

I J (qJ

¯ J LpL),

(12)

E = − 8π v2

N

 

[

 

 

N

 

which is indeed compatible with electric/magnetic duality. Upon decomposing into real matrices, iNI J = μI J iνI J , this result coincides with the corresponding terms in the so-called black hole potential

VBH

=

1

(p, q)T

p ,

 

μ + νμ1ν

νμ1 , (13)

2

 

 

M q

M =

μ1ν

μ1

discussed in [11], and more recently in [13]. Namely, setting v1 = v2 (which enforces the vanishing of the curvature scalar) we obtain E = (4π )1 VBH.

3 Application to N = 2 Supergravity

We now give the entropy function for N = 2 supergravity coupled to n Abelian N = 2 vector multiplets, first at the two-derivative level and then in the presence of higher-curvature interactions proportional to the square of the Weyl tensor. Here we

Black Hole Entropy Function and Duality

131

follow the conventions of [3], where the charges and the Lagrangian have different normalizations than in the previous section.

The N = 2 vector multiplets contain complex physical scalar fields which we denote by XI , I = 0, . . . , n. At the two-derivative-level, the action for the vector multiplets is encoded in a holomorphic function F (X). The coupling to supergravity requires this function to be homogeneous of second degree, i.e. F (λX) = λ2 F (X). The gauge coupling functions NI J in (11) are given in terms of derivatives of F (X),

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F

 

 

 

 

2i

ImFI K ImFJ LXK XL

,

 

 

 

 

 

 

 

 

(14)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NI J = ¯I J +

 

 

 

 

 

 

ImFMN XM XN

 

 

 

 

 

 

 

 

 

 

 

 

 

where FI = ∂F (X)/∂XI

and FI J

 

= 2F (X)/∂XI ∂XJ .

 

 

 

= v2, the resulting

Imposing the vanishing of the Ricci scalar, i.e. setting v1

entropy function (12) can be brought into the equivalent form [5],

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E =

1

Σ

+

1

 

N I J (

 

 

 

 

 

 

 

F

I K P

K )(

QJ

 

 

 

F

 

L),

 

 

 

 

 

(15)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

QI

 

 

 

 

 

 

 

¯J LP

 

 

 

 

 

 

 

 

where

 

 

 

 

 

= −

 

¯

 

 

 

I

 

 

 

¯I

 

 

 

I

 

 

 

 

 

+

 

¯

 

 

 

+

 

 

 

 

I

+

¯I

 

 

 

 

 

 

 

 

Σ

 

 

 

 

i

 

Y I F

 

 

 

 

 

Y I F

 

 

 

 

 

 

q

(Y I

 

 

Y I )

 

 

 

pI (F

 

 

F

 

),

 

 

(16)

and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

P

I

pI

+

i(Y I

 

 

Y I ),

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(17)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

 

 

 

 

 

 

 

¯

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

QI

q

+

i(F

I

 

¯I

).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Here the Y I

 

are related to the XI by a uniform rescaling and FI

denotes the deriva-

tive of F (Y ) with respect to

Y

I

. Also NI J

 

 

 

 

i

(F

 

 

 

 

 

 

F

I J

)

 

 

 

F

I J

 

2F (Y )/

 

 

=

 

¯I J

 

 

 

 

, where

 

=

 

 

∂Y I ∂Y J .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Varying the entropy function (15) with respect to the scalar fields Y I

yields the

attractor equations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

QI

 

 

I J P

 

 

 

 

i

QK

 

¯KM P

 

 

 

 

 

 

 

 

 

 

 

P I Q

 

 

 

 

QL

¯LN P

 

 

 

 

 

2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F

 

 

 

J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F

 

 

 

 

 

 

M

 

 

N KP F

 

 

 

 

 

 

N QL

 

 

 

 

 

F

 

N

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

= 0,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(18)

where F

P I Q

=

3F (Y )/∂Y P ∂Y I ∂Y Q

. The attractor

equations determine the hori-

 

 

 

I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

, qI ). Because the function

zon value of the Y

 

in terms of the black hole charges (p

 

F (Y )

is homogeneous of second degree,

we have F

I J K

Y K

=

0. Using this relation

 

 

 

K

 

 

 

J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

one deduces from (18) that (QJ FJ K P

 

 

)Y

 

= 0, which is equivalent to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

¯

 

 

 

I

 

 

 

 

 

¯I

 

 

 

=

 

 

 

 

I

 

 

I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i(Y I F

 

 

 

Y I F

)

 

 

 

 

pI F

 

 

 

 

 

q

Y I .

 

 

 

 

 

 

 

 

(19)

Therefore, at the attractor point, we have

 

 

 

 

 

 

 

¯I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=

 

 

 

¯

 

 

 

 

I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Σ

 

 

 

i(Y I F

 

 

 

 

Y I F

 

).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(20)

132

Gabriel Lopes Cardoso

It is possible to incorporate higher-curvature interactions involving the square of the Weyl tensor by including the Weyl multiplet into the function F , so that now F = F (Y, Υ ) subject to

F (λY, λ2Υ ) = λ2F (Y, Υ ).

(21)

Here Υ denotes the rescaled square of the auxiliary field Tab of the Weyl multiplet. The associated entropy function is then given by [5]

 

Y , Υ, U )

 

 

1

U Σ

 

1

U N I J (

 

 

 

 

 

F

 

 

 

K )(

 

 

F

 

 

L)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=

 

2

+ 2

 

 

 

 

 

I K P

 

 

 

 

E (Y, ¯

 

 

 

 

 

 

 

 

 

 

 

 

QI

 

 

 

 

 

QJ ¯J LP

 

 

 

 

 

 

 

 

 

 

4i

 

 

¯

 

 

 

I

 

 

 

 

 

¯I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(Y I F

 

Y I F

 

)(U

1)

 

 

 

 

 

 

 

 

 

 

 

 

Υ

 

¯Υ

 

 

 

 

 

 

 

 

 

 

+

 

2

 

 

 

 

 

 

 

 

 

 

 

Υ

 

 

2

 

 

 

+ 32

 

 

 

 

 

 

 

 

 

 

 

 

 

i(F

 

 

 

F

 

)

 

 

 

 

U Υ

 

 

 

 

(U

 

 

U

1

 

 

)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8(1 + U )

 

 

Υ

,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(22)

where now

 

= F

 

 

¯

 

 

 

 

¯

 

 

 

 

 

 

 

 

+

¯

 

 

+

 

 

 

 

 

 

+

¯I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I (Y

 

 

 

 

 

 

 

 

 

I

 

 

 

 

 

Σ

 

(Y, Y , Υ, Υ )

 

 

q

 

 

 

 

I

 

 

Y I )

 

pI (F

 

 

F

).

 

 

(23)

The quantity F (Y, ¯

 

¯

 

is defined by

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Y , Υ, Υ )

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

¯I

 

 

 

 

 

 

 

 

 

¯

 

 

 

 

F

 

¯

¯

 

= −

 

¯

 

 

I

 

 

 

 

 

 

 

 

Υ

¯Υ

 

 

 

(Y, Y , Υ, Υ )

 

 

 

 

 

i

Y I F

 

 

 

 

Y I F

 

 

 

2i

 

Υ F

 

 

Υ F

 

,

(24)

where FΥ = ∂F /∂Υ . The entropy function (22) depends on the variables U = v1/v2, Υ and Y I . Their attractor values are determined by requiring E to be stationary. We refer to [5] for the detailed form of the associated attractor equations.

With the normalizations used in this section, the entropy (10) reads

Smacro(p, q) = 2π E

attractor.

(25)

 

 

 

Supersymmetric black holes are the subset of extremal black holes satisfying [3]

QI = P J = 0, Υ = −64, U = 1.

(26)

The conditions QI = P J = 0 can also be obtained from a variational principle based on Σ [1, 4]. The entropy of supersymmetric black holes reads [2]

Smacro = π Σ |attractor.

(27)

It can be written as a Legendre transform [16],

 

Smacro = π FE (p, φ) qI φI ,

(28)

where

 

FE (p, φ) = 4Im F (Y, Υ = −64),

(29)

Соседние файлы в папке Diss