IP/BBSR/2004-06-17

hep-th/0406169

Born-Infeld black holes in the presence of a

cosmological constant
arXiv:hep-th/0406169v1 21 Jun 2004

Tanay Kr. Dey

Institute of Physics, Bhubaneswar-751005, India

tanay@[Link]

Abstract

We construct asymptotically anti-deSitter (and deSitter) black hole solutions of Einstein-

Born-Infeld theory in arbitrary dimension. We critically analyse their geometries and
discuss their thermodynamic properties.
1 Introduction

By now, there are substantial evidences which suggests that the type IIB string theory on AdS5 ×
S5 is dual to four dimensional N = 4 super Yang-Mills theory [1]. At high temperature, the
thermal state of the string theory is described by an asymptotically AdS black hole. Therefore,
qualitatively the properties of thermal super Yang-Mills can be understood by studying the black
hole geometry. However, due to the fact that the bulk supergravity describes gauge theory at
strong coupling, the quantitative understanding becomes difficult. Nevertheless, many attempts
were made along this direction (see for example [2], [3], [4], [5], [6]). By going higher in the bulk
coupling, we make the dual boundary theory weaker. For example, one may wonder if we can
atleast qualitatively understand the behaviour of the boundary Yang-Mills theory as we perturb
the black hole geometry by turning on the higher derivative curvature terms in the bulk action.
In fact, such an analysis was carried out in [7] by constructing bulk anti-deSitter black holes in
the presence of certain higher curvature terms in the supergravity action. Furthermore, black
holes of higher curvature gravity were constructed where the electromagnetic coupling was turned
on [8] ∗ . In general, beside the curvature terms, one would also expect higher derivative gauge
field contributions to the supergravity actions. How does the boundary theory respond when we
incorporate such corrections? As a first step to analyse such an issue, in this paper, we study
the effect of adding higher derivative gauge field terms on the bulk adS (dS) black hole geometry.
This is done by explicitly constructing black hole solutions of the supergravity action coupled to
a Born-Infeld gauge field in arbitrary dimension in the presence of a cosmological constant. This
action not only incorporates the higher order gauge field corrections to the Einstein-Maxwell
gravity in the presence of a cosmological constant, but also allows us to find exact black hole
solutions. Note that in general the gravity action may have both higher order curvature terms
as well as the higher derivative terms due to gauge fields. We have not analysed here the black
holes in these.

In the remaining part of the paper, we first construct black holes in (n + 1) dimensional Einstein-
Born-Infeld theory in the presence of a negative or positive cosmological constant† . Next, we
study the thermodynamics of these holes. Here, in particular, we check that these black holes
follow first law of thermodynamics. Then, by calculating specific heat at fixed charge, we show
that for a certain range of parameters these black holes are stable. Expression of free energy at
fixed charge is also calculated for these holes. Furthermore, by expanding our solutions around
Reissner-Nordstrom anti deSitter (RNadS) black holes [9], we find out the effect of higher order
gauge field corrections to the geometry.
∗
Under IIB string compactification, the electromagnetic field appear when we take the compact space to be a
spinning sphere.
†
In three and four dimensions the solutions were constructed in [10] and [11] respectively. However, their
thermodynamical behaviour was not analysed.

2
2 AdS and dS Black hole solutions of Einstein-Born-Infeld theory

In this section, we will explicitly construct the black hole solutions of Einstein-Born-Infeld action
in (n + 1) dimension in the presence of a negative (or positive) cosmological constant Λ. The
action has the form
√ h (R − 2Λ)
Z i
S = dn+1 x −g + L(F ) , (1)
16πG
where L(F ) is given by s
2
 F µν Fµν 
L(F ) = 4β 1 − 1 + . (2)
2β 2
The constant β is the Born-Infeld parameter and has the dimension of mass. In the limit β → ∞
L(F ) reduces to the standard Maxwell form

L(F ) = −F µν Fµν + O(F 4 ). (3)

The black holes in this limit were constructed in [9] and their thermodynamics and phase struc-
tures were studied in [13], [9]. For simplicity, in this paper, we will work with the convention that
16πG = 1, where G is the Newton’s constant.

By varying the action with respect to the gauge field Aµ and the metric gµν , we get the corre-
sponding equations of motion. These are respectively
 F µν 
▽µ q = 0, (4)
F2
1 + 2β 2

and
2 1 2 gµν F 2 2Fαµ F α ν
Rµν + gµν Λ = gµν L + q − q . (5)
n−1 n−1 n − 1 1 + F 22 1 + F2
2
2β 2β

In order to solve the equations of motion, we use the metric ansatz

dr 2
ds2 = −V (r)dt2 + + r 2 dΩ2 n−1 , (6)
V (r)

where, dΩ2 n−1 denotes the metric of an unit (n − 1) sphere. V (r) is an unknown function of r
which we will determine shortly. First of all, a class of solution of equation (4) can immediately
be written down where all the components of F µν are zero except F rt . It is given by
p
rt (n − 1)(n − 2)βq
F = p 2 2n−2 . (7)
2β r + (n − 1)(n − 2)q 2
Here q is an integration constant and is related to the electromagnetic charge. This can be
q
concluded from the behaviour of F rt in the large β limit as F rt ∼ rn−1 . We notice that the
electric field is finite at r = 0. This is expected in Born-Infeld theories. Now, parametrising
Λ = − n(n−1)
2l2 , equation (6) can easily be solved as

h 4β 2
√
m 1i 2 2 2β
Z q
V (r) = 1 − n−2 + + r − 2β 2 r 2n−2 + (n − 1)(n − 2)q 2 dr. (8)
r n(n − 1) l2 (n − 1)r n−2

3
The integral can be done in terms of hypergeometric function and can be written in a compact
form. The result is
h 4β 2
√
m 1i 2 2 2β q
V (r) = 1 − n−2 + + 2 r − 2β 2 r 2n−2 + (n − 1)(n − 2)q 2
r n(n − 1) l n(n − 1)r n−3
2(n − 1)q 2 h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i
+ F
2 1 , , , − . (9)
nr 2n−4 2n − 2 2 2n − 2 2β 2 r 2n−2
In the above expression, m appears as an integration constant and is related to the ADM mass
of the configuration. It can be checked that for n = 3, it reduces to the solution of [11]. Also,
in the l → ∞ limit (that is for Λ = 0), this solution reproduces correctly the asymptotically flat
Born-Infeld black hole (see for example [12]).

Using the fact that 2 F1 (a, b, c, z) has a convergent series expansion for |z| < 1, we can find the
behaviour of the metric for large r. This is given by

m q2 r 2 (n − 1)(n − 2)2 q 4
V (r) = 1 − + + − 2 , (10)
r n−2 r 2n−4 l2 8β (3n − 4)r 4n−6

Note that in the β → ∞, it has the form of Reissner-Nordstrom adS black hole [9]. The last term
in the right hand side of the above expression is the leading Born-Infeld correction to the RNadS
black hole in the large β limit. From the asymptotic behaviour, we see that m is related to the
mass of the configuration. In particular, in our convention, the ADM mass M is

M = (n − 1)ωn−1 m, (11)

where ωn−1 is the volume of the unit (n − 1) sphere. More interesting is the behaviour of V (r)
close to the origin where

m − A h 2cβ i q h 4β 2 1 i 2 h 2cβ i β 2 r n+1

V (r) = 1− − −B(2n−1)q + + r − +B , (12)
r n−2 n r n−3 n(n − 1) l2 n (n − 1)(n − 2)

where n−2
2(n − 1)q 2 n 2β 2 o 2n−2 h 3n − 4 i h 1 i
A= √ Γ Γ , (13)
n π (n − 1)(n − 2)q 2 2n − 2 2n − 2
and h i h i
s 3n−4 −1
2(n − 2) 4β Γ 2n−2 Γ 2n−2
c= and B= h i h i.
(n − 1) cn(2n − 1)q Γ n−2
Γ 2n−3
2n−2 2n−2

From the above expression, we see that for generic values of n ≥ 3, the metric has a curvature
singularity at r = 0. However, for n = 3 and for m = A, the metric is regular at r = 0. We
now proceed to analyse if this singularity is hidden behind a horizon. The horizons correspond
to the locations where V (r) = 0. Though we are unable to solve this equation analytically, we
first plot, in figure 1, the function V (r) for some different values of m and for n = 4. In this
figure, the other parameters such as l, β are kept fixed. First of all, let us note that there can be
one or two horizons depending on the value of m. Furthermore, for certain choices of m there
can be no horizon, leading to a naked singularity at the origin. To have further understanding

4
 VHrL

30

20

10
r
1 2 3
-10

-20

Figure 1: The metric function V (r) as a function of r for n = 4, β = 1, q = 1, l = 1. The dashed line,
solid line, dotted line and dash-dotted line are for m = 1.5, 2.062, 2.6, 3 respectively.

on the nature of the horizons, we plot in figure 2, the mass as a function of the horizon radius.
The mass parameter of the hole can be expressed in terms of horizon radius (r+ ) as
h 4β 2
√
n−2 1i n 2 2βr+ q 2 2n−2
m = r+ + + r − 2β r+ + (n − 1)(n − 2)q 2
n(n − 1) l2 + n(n − 1)
2(n − 1)q 2 h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i
+ n−2 F
2 1 , , , − 2n−2 . (14)
nr+ 2n − 2 2 2n − 2 2β 2 r+
In figure 2, m(r+ ) is shown for different values of β. For a given β, the number of horizons depend
clearly upon the choice of m. If we focus our attention to the solid line (β = 5), we see that upto
certain m, there are two horizons. As we decrease the mass further, the two horizons meet. We
mHr+ L

12
10
8
6
4
2
r+
0.25 0.5 0.75 1 1.25 1.5 1.75

Figure 2: The mass m as a function of r+ for adS Born-Infeld holes for n = 4, l = 1, q = 1, β = 16 (dotted
line) ,5 (solid line) ,0.1 (dashed line).

then call the black hole extremal. It is easy to find out from V (r) that this happens when r+
satisfies the following condition:
h 4β 2
√
n−3 n i n−1 2 2β q 2 2n−2
(n − 2)r+ + + r − 2β r+ + (n − 1)(n − 2)q 2 = 0. (15)
n − 1 l2 + n−1
For β = 0.1 there is only a single horizon, as equation (15) does not have a real solution for r+ and
hence extremality condition can not be satisfied. As we will see later that when this condition is

5
satisfied, the temperature of the black hole vanishes. We also notice that the behaviour of V (r)
crucially depends on the ratio m/A, where A is defined in (13). If m/A ≥ 1, V (r) behaves like
that of a Schwarzschild black hole close to the origin. On the other hand, for m/A < 1, behaviour
of V (r) is more like the Reisssner-Nordstrom one.

From equation (7), we can also calculate the gauge field associated with this configuration. It is
given by

1 q h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i
At = 2 F1 , , , − − Φ, (16)
c r n−2 2n − 2 2 2n − 2 2β 2 r 2n−2
where q is related to the black hole charge Q via
q
Q = 2 2(n − 1)(n − 2)ωn−1 q .

In equation (16), Φ is the gauge potential. We will choose Φ in such a way that At is zero at the
horizon. This gives

1 q h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i
Φ= 2 F1 , , , − . (17)
c r+ n−2 2n − 2 2 2n − 2 2β 2 r+ 2n−2
Behaviour of Φ as a function of the horizon size for n = 4, q = 4, l = β = 1 is shown in figure 3.
Notice that Φ is finite even when r+ = 0.
FHr+ L

2.5

1.5

0.5

r+
2 4 6 8 10
Figure 3: The potential Φ as a function of r+ for n = 4, l = 1, β = 1, q = 4. Note that Φ is finite for
r+ = 0.

We would now like to make some brief comments on the Born-Infeld deSitter black holes. These
are the solutions in the presence of a positive cosmological constant. This can be found from the
earlier expression of V (r) by replacing l2 by −l2 . More explicitly,
h 4β 2
√
m 1i 2 2 2β q
V (r) = 1 − n−2 + − 2 r − 2β 2 r 2n−2 + (n − 1)(n − 2)q 2
r n(n − 1) l n(n − 1)r n−3
2(n − 1)q 2 h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i
+ 2 F1 , , , − . (18)
nr 2n−4 2n − 2 2 2n − 2 2β 2 r 2n−2

6
This metric, in the asymptotic region, goes to the Reisnner-Nordstrom deSitter black holes with
a β dependent Born-Infeld correction. While near r = 0, V (r) behaves similar to equation (12)
with l2 → −l2 . Therefore, the singularity structure of this solution is the same as the previous
one. Now, turning our attention to the nature of the horizon, we find that it is best described in
terms of a plot of m as a function of r+ . This is shown in figure 4, where we have given a plot of
m(r+ ) for different l, keeping the other parameters fixed. For small l, m monotonically decreases
with r+ . So, the configuration has only one horizon (inner or cosmological depending on the
mass). However, for large l, the solution has three horizons; out of them the largest one is the
cosmological horizon and the other two are the black hole inner horizon and the event horizon.
For fixed l, if we decrease m, these two horizons of the black hole come closer and they meet
at a certain value of m. For even lower values of m, only the cosmological horizon exists. Now
for even larger value of l (shown in dashed line in the figure), there can be only two horizons.
Out of that, the larger one is the cosmological horizon. As we now increase m, the event and
the cosmological horizons meet. Beyond that value of m we get naked singularity (all horizons
vanish) at the origin.
mHr+ L
4

r+
1 2 3 4 5
-1

Figure 4: Mass m as a function of r+ for Born-Infeld deSitter holes for n = 4, β=1, q = 1, l = 2 (dotted
line), 3 (solid line) , 4 (dashed line).

3 Thermodynamics

We now would like to study the thermodynamical properties of the black holes we have just
found.

The Hawking temperature of the hole can be calculated using the relation
κ
T = , (19)
2π
where κ is the surface gravity and is given by
1 dgtt
κ=− | . (20)
2 dr r=r+

7
κ can be calculated explicitly from the metric functions. The temperature is then,
√
1 h n − 2 n 4β 2 no 2 2β q i
2 r 2n−2 + (n − 1)(n − 2)q 2 .
T = + + 2 r+ − 2β + (21)
4π r+ n−1 l (n − 1)r+ n−2

We note that when r+ is such that the right hand side of the above equation is zero, the temper-
ature of the black hole is zero. We notice that this gives the same constraint on r+ that we have
already encountered while discussing extremal black hole (see eq (15)). From here we conclude,
in the extremal limit, the temperature of the black hole is zero.

Using the standard formula for entropy

Z  ∂M 
S= T −1 dr+ , (22)
∂r+ Q

we get
S = 4πωn−1 r+ n−1 (23)
as the entropy of the black hole. It is indeed proportional to the area of the horizon. Now
that we have all the relevant thermodynamic quantities, we can easily verify that the first law of
thermodynamics. We find that
dM = T dS + ΦdQ, (24)
is satisfied. To find the stability of the black hole, it is important to find the specific heat of the
hole. This can be easily evaluated using
 ∂M   ∂M   ∂T 
CQ = = / , (25)
∂T Q ∂r+ Q ∂r+ Q

where M and T are given in eqs. (11) and (21). In figure 5, we have plotted CQ as a
CQ
0.3
0.2
0.1
r+
0.1 0.2 0.3 0.4 0.5
-0.1
-0.2
-0.3

Figure 5: Specific heat at fixed charge as a function of r+ where n = 4, l = .2, β = 1 and q = .5, 1, 2, 3, 4
from left to right respectively.

function of horizon radius r+ . We see from the figure that the specific heat is negative if r+ is less
than certain value; making the hole unstable. We therefore conclude that the large Born-Infeld
black holes with adS asymptotics are stable against fluctuations. We can now write down the

8
 FHr+ L

r+

Figure 6: Free energy at fixed charge for Born-Infeld (solid line) and RNadS (dashed line) black holes for
n = 4, l = 1,β = 10 and q = 10.

thermodynamics canonical potential at fixed charge Q, F = E − T S, where E = M − Me and Me

is the ADM mass of the extremal black hole that follows from the condition (15). We get
n 4β 2
√
h
n−2 1 o n 2 2βr+ q 2 2n−2
F = ωn−1 r+ − + r+ + 2β r+ + (n − 1)(n − 2)q 2
n(n − 1) l2 n(n − 1)
2(n − 1)2 q 2 h n − 2 1 3n − 4 (n − 1)(n − 2)q 2 i i
+ n−2 2 F1 , , , − − Me . (26)
nr+ 2n − 2 2 2n − 2 2β 2 r+ 2n−2

In figure 6, we have shown F as a function of r+ for Born-Infeld black holes and for the standard
RNadS black holes. This figure also shows that the larger black holes are more stable than the
smaller ones. We also see that for large r+ , F behaves similar to that of RNadS black holes.
However, for small holes, their behaviours are distinctly different.

4 Discussion

In this letter, we have constructed charged adS and dS black holes of Einstein-Born-Infeld actions.
We believe that this is only the first step before we analyse, following the adS/CFT correspon-
dence, the behaviour of the boundary theory caused by higher order gauge field perturbations
in the bulk. Beside the motivation coming from adS/CFT side, Born-Infeld lagrangian appears
very frequently in string theory. So we expect that it is worthwhile to know various properties of
black hole solutions in this theory.

Though, in this paper we have constructed the Born-Infeld black holes in the presence of a
cosmological constant and discussed their thermodynamical properties, many issues however still
remain to be investigated. We know that Reissner-Nordstrom adS black holes undergo Hawking-
Page phase transition. This transition gets modified as we include Born-Infeld corrections into
account. We hope to carry out a detail study on this issue in the future. Furthermore, in the
context of brane world cosmology, it was found that a brane moving in a Reisssner-Nordstrom adS
background generates non-singular cosmology [14]. However, as shown in [15], the brane always
crosses the inner horizon of the bulk geometry, creating instability. It would be interesting to

9
study cosmology on the brane when it is moving in the charged black hole backgrounds that we
have constructed. Note that since these charged holes does not have inner horizon for certain
range of parameters, we may generate non-singular cosmology without creating the instabilities
that we have just mentioned.

Acknowledgements:

I would like to thank S. Mukherji for suggesting me this problem, for fruitful discussions at every
step of the work and for help in structuring the manuscript. I would also like to thank A. Biswas
for several useful conversations.

References
[1] O. Aharony, S. Gubser, J. Maldacena, [Link] and Y. Oz, Phys. Rept. 323 (2000) 183,
hep-th/9905111.
[2] S. Gubser, I. Klebanov and A. Tseytlin, Nucl. Phys. B534 (1998) 202, hep-th/9805156.
[3] M. Li, JHEP 9903 (1999) 004, hep-th/9807196.
[4] K. Landsteiner, Mod. Phys. Lett. A14 (1999) 379, hep-th/9901143.
[5] M. Vazquez-Mozo, Phys. Rev. D60 (1999) 106010, hep-th/9905030.
[6] C. Burgess, N. Constable, R. Myers, JHEP 9908 (1999) 017, hep-th/9907188.
[7] M. Cvetic, S. Nojiri, S. Odintsov, Nucl. Phys. B628 (2002) 295, hep-th/0112045; S. Nojiri
and S. Odintsov, Phys. Rev. D66 (2002) 044012, hep-th/0204112.
[8] M. Banados, Phys. Lett. B579 (2004) 13, hep-th/0310160.
[9] A. Chamblin, R. Emparan, C. Johnson and R. Myers, Phys. Rev. D60 (1999) 064018,
hep-th/9902170.
[10] M. Cataldo and A. Garcia, Phys. Lett. B456 (1999) 28, hep-th/9903257.
[11] S. Fernando and D. Kung, Gen. Rel. Grav. 35 (2003) 129, hep-th/0306120.
[12] D. Rasheed, ”Non-linear electrodynamics: zeroth and first laws of black hole thermodynam-
ics”, hep-th/9702087.
[13] A. Chamblin, R. Emparan, C. Johnson and R. Myers, Phys. Rev. D60 (1999) 104026,
hep-th/9904197.
[14] A.K. Biswas and S. Mukherji, JHEP 0103 (2001) 046, hep-th/0102138; S. Mukherji and M.
Peloso, Phys. Lett. B 547 (2002) 297, hep-th/0205180; A. Biswas, S. Mukherji and S.S. Pal,
Int. Jour. Mod. Phys. A19 (2004) 557, hep-th/0301144.
[15] J.L. Hovdebo and R.C. Myers, JCAP 0311 (2003) 012, hepth/0308088.

10