Black Hole Thermodynamics

Narit Pidokrajt

Masters Thesis in
Theoretical Physics
Stockholm University
Department of Physics
2003

Abstract
Thermodynamics of various black holes is studied by means of thermodynamic
Riemannian curvatures . The thermodynamic Riemannian curvature is a scalar
curvature of the Ruppeiner metric, which is defined as a metric tensor on the thermodynamic parameter space, mathematically it is the matrix of second derivatives
(Hessian matrix) of the entropy with respect to the energy (mass of the black hole)
and other extensive thermodynamic parameters such as the black holes spin and
the electric charge. In ordinary thermodynamics the Ruppeiner metric is a flat metric if the underlying statistical mechanical system is non-interacting such as that of
the ideal gas. Furthermore, the curvature scalar of the Ruppeiner metric diverges
at the critical point. In this thesis, various black hole families are investigated,
i.e., the BTZ, the Reissner-Nordstrom (RN), the Reissner-Nordstrom-anti-de Sitter
(RNadS), the Kerr and the Kerr-Newman (KN) black holes respectively and it is
found that the BTZ and the RN black holes both have vanishing curvatures. The
RNadS black hole has a curved geometry and gives interpretable resultsnamely
that its curvature diverges in the extremal limit and its metric changes signature
where the thermodynamical stability properties of the black hole change. The Kerr
black hole has a non-zero curvature scalar which diverges at the extremal limit,
whereas the Ruppeiner metric for KN black hole has a non-trivial curvature and
does not give rise any further conclusive remarks.

Contents
Preface

Acknowledgments

viii

1 Introduction

2 Black Holes and General Relativity

2.1 Black hole metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2 The BTZ Black Hole: The (2+1)-dimensional black hole . . . . . . .
2.3 Black holes and Thermodynamics . . . . . . . . . . . . . . . . . . . .

3
5
8
9

3 Thermodynamic Fluctuation Theory

3.1 The role of entropy . . . . . . . . . .
3.2 Fluctuations . . . . . . . . . . . . . .
3.3 Thermodynamics and Geometry . .
3.3.1 Ruppeiner metrics . . . . . .
3.3.2 Weinhold metrics . . . . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

14
14
15
17
18
19

4 Results
4.1 BTZ black hole . . . . . . . . .
4.2 Reissner-Nordstrom black hole
4.3 Kerr black hole . . . . . . . . .
4.4 Kerr-Newman black hole . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

20
21
22
25
27

.
.
.
.

.
.
.
.

.
.
.
.

5 Conclusions and Speculations

30

Appendix A

Thermodynamics in a nutshell

32

Appendix B

Curvature scalar calculation

34

Bibliography

36

Index

37

List of Figures
2.1

2.2

2.3

2.4

2.5

4.1

4.2

4.3

Light cones in Minkowski spacetime. Any point in the future light

cone r = t can be reached by a particle or signal with speed less than
c. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A trajectory of a particle or signal approaching the Schwarzschild
black hole as represented in the Minkowski spacetime. The surface
r = 2M is the event horizon, nothing inside surface is able to escape.
This trajectory is given by solving the Schwarzschild metric ds2 = 0
(null geodesic) with d2 = 0. . . . . . . . . . . . . . . . . . . . . . .
The Schwarzschild black hole in the Kruskal-Szekeres coordinates
have no coordinate singularity hence represents the real spacetime
where r = 0 is the singularity which is the broken thick line in the
figure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A rough sketch of the Kerr black hole which is surrounded by an ergosphere. The ergosphere is a region inside which nothing can remain
stationary. The angular momentum of the Kerr black hole is denoted
by J. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The 2+1 dimensional BTZ black hole. This black hole can be visualized as a circular disc with spin J. Its event horizon depends on its
mass. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Heat capacity of the BTZ black hole as a function of S and J. The
C(S, J) is a truly positive function for any values of S and J. It is
notable that at the extremal limit J = 2S 2 , which is the same as
J = M the temperature vanisheshence there is no heat capacity. .
The Ruppeiner metric for the RN black hole as seen in Minskowski
space when Rindler coordinates are used. The shaded area represents
the Lorentzian metric, i.e., the metric of ( +) signature. . . . . . .
Heat capacity of the RNadS black hole as a function of entropy S
and electric charge Q for 2 different values of l, i.e., we have set l = 1
in Fig.(A) and l = 0.01 in Fig.(B). The heat capacity of the RNadS
black hole vanishes at the extremal limit, whereas in the case of a
small value of l (large ) as in Fig. (B) the heat capacity becomes
linear in the entropy. . . . . . . . . . . . . . . . . . . . . . . . . . . .

ii

21

23

24

4.4

4.5

4.6

4.7

The RNadS black holes Ruppeiner geometrythe shaded areas represent the Lorentian metrics while the strips are the Riemannian
metric which is a positive definite metric with the (+ +) signature.
Fig.(A) is when l = 1 which is the when the cosmological constant
= 1, Fig.(B) represents the geometry when the value of l is large.
In Fig.(C) we have l which means that the cosmological constant is vanishing thus it reduces to the case of RN black hole where
its Ruppeiner metric is a Lorentzian metric. . . . . . . . . . . . . .
The Riemannian curvatures for the RNadS black hole as a function
of Q, S with 2 different values of l. In Fig.(A) we set l = 1 whereas
Fig.(B) represents the curvature where l = 0.001. There are discontinuities in both cases along the extremal limits. . . . . . . . . . . .
The curvature scalar for the Kerr black hole as a function of M and
J. It is noticeable that a divergence of the curvature occurs along
J = M 2 which is an extremal limit of this black hole . . . . . . . . .
Heat capacity for the KN black hole for constant J and Q, we here
set J = Q = 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A.1 Phase transitions in T S diagram. On the left is the phase transition

of first order; it is obvious that there is a jump in entropy hence the
divergence of the heat capacity. The phase transition of second order
is shown on the right in the figure, there is a finite jump in entropy
in this case. We assume the constant pressure for both cases. . . . .

26

26

28
28

33

List of Tables
2.1
2.2
5.1

Analogy between thermodynamic parameters and black holes parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Analogy between the laws of thermodynamic and the laws of blackhole mechanics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12

Geometries of the Ruppeiner and Weinhold metrics for various black

holes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31

iv

10

Preface
It is no exaggeration to say that there is no other wayunless one wishes something
non-scientificto learn about black holes than using our existing and yet-to-bediscovered physics knowledge. By our existing and yet-to-be discovered physics
knowledge I mean certain knowledge of physics that is capable of telling us certain
truths about black holes in a systematically reasonable way, and we do believe that
physics is doing a good job. As the author of this thesis, I believe that black holes
do exist in the Universe and the reason for that is because there have been certain
verifications that followed feasible logics both theoretically and experimentally. I do
also believe that there are very many things that we do not know about them owing
to the fact that the nature of the black holes are not of the common sense that exists
in our mind. As a matter of fact no one has ever seen the black hole with his/her
naked eyes and will never! Therefore the best tools that we do have for the time
being to study these strange objects (entities) seem to be:
Physics together with mathematics: So far they both have been serving as
theoretical tools to get insights into the nature of black holes to some extent.
Our imaginations and our dreams: Without dreams humans would not have
been flyingthe great work by great physicists have always been inspired by
their imaginations and dreams. These will be some of the key factors if one
aims to study the black holes seriously.
Modern astronomy and space technologies: With the Hubble space telescope
being in orbit we are in hope that it will reveal to us more and better pictures
of the Universe, including some indirect measurement of what are believed to
be black holes. Recall that this was not possible some 15 years ago, therefore
it can be stressed that there is still so much one can learn in the future owing
to the speedy advancement of the modern astronomy and space technologies.
Nevertheless, progress in the physics of black holeseither small steps or giants
leapsis yet to come and one cannot say anything more congretely than this. The
mysteries of the black holes may prevail for decades but these queer objects will still
be of interest to the world of physics inarguably. For the sake of history, I would
like to present the timeline for the physics of black holes (black hole physics) and
I hope that this will be useful for some of the readers who may not know how the
concept (knowledge/physics) of black holes began and how it has developing. This
timeline chart was done by Niel Brandtinterestingly under the condition that the
Copyright Notice is posted as it is on his website I am permitted to copy and
paste it here. Thankfully, this text may also be modified up to ones need and my
modification will be in slanted letters.
v

Copyright Niel Brandt 1994.

Copyright Notice (see link below)
(http://www.gsu.edu/other/timeline.html)
1784 : John Michell discusses classical bodies which have escape velocities greater
than the speed of light
1795 : Pierre Laplace discusses classical bodies which have escape velocities greater
than the speed of light
1916 : Karl Schwarzschild solves the Einstein vacuum field equations for uncharged
spherically symmetric systems
1918 : H. Reissner and G. Nordstrom solve the Einstein-Maxwell field equations for
charged spherically symmetric systems
1923 : George Birkhoff proves that the Schwarzschild spacetime geometry is the
unique spherically symmetric solution of the Einstein vacuum field equations
1939 : Robert Oppenheimer and Hartland Snyder calculate the collapse of a pressurefree homogeneous fluid sphere and find that it cuts itself off from communication
with the rest of the universe
1963 : Roy Kerr solves the Einstein vacuum field equations for uncharged rotating
systems
1964 : Roger Penrose proves that an imploding star will necessarily produce a
singularity once it has formed an apparent horizon (trapped surface)
1965 : Ezra Newman, E. Couch, K. Chinnapared, A. Exton, A. Prakash, and Robert
Torrence solve the Einstein-Maxwell field equations for charged rotating systems
1968 : Brandon Carter uses Hamilton-Jacobi theory to derive first-order equations
of motion for a charged particle moving in the external fields of a Kerr-Newman
black hole
1969 : Roger Penrose discusses the Penrose process for the extraction of the spin
energy from a Kerr black hole
1969 : Roger Penrose proposes the cosmic censorship hypothesis
1971 : Identification of Cygnus X-1/HDE 226868 as a binary black hole candidate
system
1972 : Stephen Hawking proves that the area of a classical black holes event horizon
cannot decrease
1972 : James Bardeen, Brandon Carter, and Stephen Hawking propose four laws of
black hole mechanics in analogy with the laws of thermodynamics
1972 : Jacob Bekenstein suggests that black holes have an entropy proportional to
their surface area due to information loss effects
1974 : Stephen Hawking applies quantum field theory to black hole spacetimes and
shows that black holes will radiate particles with a blackbody spectrum which can
cause black hole evaporation
1989 : Identification of GS2023+338/V404 Cygni as a binary black hole candidate
system
1992 : The 2+1 dimensional black hole was discovered by Banados, Teitelboim and
Zanelliknown as the BTZ black hole. It is merely a theoretical playground for
physicists but proved to be valuable as a platform for string theory and suchs.
1994 : Hubble Space Telescope (HST) confirms existence of massive black hole
at heart of active galaxy M87, located 50 million light-years from Earth in the
constellation Virgo.

What this thesis is concerned with is the black hole thermodynamics as it is

entitledthe black holes are studied thermodynamically by means of thermodynamic (classical) fluctuation theory based on the concept of the Riemannian geometry. Contained in this volume are five chapters, two appendices and an index. The
first two chapters are basically introducing messages especially for those who might
not be familiar with the subject of general relativity. The third chapter is basically
the main stream of this work, which is an application of the the thermodynamic
fluctuation theory to the black holes. Results will be given in chapter four followed
by conclusions and speculations of the results. The two appendices are brief information on ordinary thermodynamics and curvature scalars respectively. As a matter
of fact, I am fully aware that writing a thesis is not a so easy task that can be accomplished in a few months time, but with the limited amount of time of roughly
5 months I am pleased enough to present this work as it is in your hand.
It would be an out-of-date fashion to have this thesis published only on papers,
therefore this thesis is also made available on the world-wide web in two formats as
follows:
www.physto.se/narit/bh.pdf (Acrobat reader program is required)
or
www.physto.se/narit/bh.ps (In Postscript format, Ghostview program may
be needed)
I openly welcome comments or suggestions in any aspect that the reader may
have pertaining to this work. I can be reached easily via email at narit@physto.se
and finallyenjoy the reading!

Narit Pidokrajt
Stockholm, April 2003

Acknowledgments
There would not be this thesis had I not met Professor Ingemar Bengtssonmy
supervisorwho has been a significantly constant source of encouragement and
knowledge for me since the first day I arrived at Fysikum in Stockholm. Each
discussion with him has always been fruitful and essential for me to achieve the goal
of this project. I would also like to thank him a lot for being critical in proofreading this thesis. I am grateful to Dr. Jan
Aman for his collaboration and for many
interesting conversations. Many thanks go to
Asa Ericsson for exchanging thoughts
with me on various things both inside and outside physics. I wish to thank Lars
Samuelsson for introducing me to the GRtensor program which makes life a bit
easier. (Thanks go to those in Canada who created it as well!) Dr. Sayed Fawad
Hassan is acknowledged for teaching me how to create transparencies with LATEX. I
am thankful to Moundheur Zarroug and Professor Alexander Zheltukhin (my officemates since the end of February 2003) for widening my views on France and Ukraine
plus some stories about the Russian physicists that I have never heard of from anyone else. Every member of the Falt-och-partikel (FOP) Group is thankworthy for
making this department a very attractive and stimulating work place.
I am indebted to Isabella Malmnas for helping me to get a little home in Katrineholm
as it is always hard to get a place to live in Stockholm, and for making my life besides
physics an emotional and delightful life. Lastly and unforgettably I wish to thank
my family for every support that they have been providing to me although they are
so far way.

viii

Chapter 1

Introduction
Time and space are modes in which we think and not conditions in which we
live in.
A. Einstein

Black holes are some of the most exotic1 entities encountered in physics of the
present time2 . The nature of spacetime within a black hole is enough to make the
science of black holes seem more like a science fiction. Even more astounding are
the connections of black hole physics with thermodynamics. Classically we would
expect the black hole to be a perfectly dead star, namely it should have an absolute zero as a physical temperature. But it was not so since Hawking has found a
startling discovery that the black hole radiates thermally [15, 16] whereas Bekenstein suggested that there is an entropy associated with the black hole [5], i.e., the
black hole entropy. However, that the black hole has an entropy first arose from
the realization that its event horizon surface area exhibits remarkable tendency to
increase when undergoing any transformation as noticed by Floyd and Penrose [27]
and later supported by Christodoulou [10]. Hawking [14] was the first to give a general proof that the surface area of the black hole cannot decrease in any process and
additionally he showed that when two black holes coalesce, the area of the resulting
black hole cannot be smaller than the sum of the initial areas. Easily speaking, it is
clear that changes in black hole generally occur in the direction of increasing area.
This reminds us of the second law of ordinary thermodynamics which states that
changes of a closed thermodynamic system take place in the direction of increasing
entropy. Historically, physicists were not convinced about the validity the black hole
thermodynamics before Hawking radiation was discovered.
Despite the written-down laws of black hole thermodynamics we have never
been able to do any real experiment to verify them, all we can do is the gedanken
experiment 3 . It is thus legitimate to say we kind of perform gedanken experiments
on the black holesthermodynamicallyby means of thermodynamic fluctuation
theory using the language of Riemannian geometry in this thesis work. By the
1

although one may take their existence for granted!

Our conclusions concerning black hole existence are based on observations of matter moving
in the vicinity of the black holes. The evidence may well improve when gravitational waves are
detected.
3
which means thought experiment. It is simply an experiment that can be done in principle,
and which is useful to think about in order to clarify ones ideas.
2

CHAPTER 1. INTRODUCTION
virtue of the thermodynamic Riemannian curvature (i.e., the curvature scalar 4 )
of the Ruppeiner metric of the black hole under consideration (to be discussed in
Sect. 3.3.1)we hopefully will obtain some new information on the black holes. We
investigate also the Weinhold metric for each black hole (Sect. 3.3.2), which is the
metric conformally related to the Ruppeiner metric, and it turns that it gives us a
helping hand when the Ruppeiner metric is too complicated to deal with.
Throughout this report, we will set G = ~ = c = kB = 1 unless otherwise stated.
These units are sometime referred to as geometrized units.

The curvature scalar is independent of the coordinate system in which it is calculated. This
point is essential when we want to endow the curvature with an intrinsic physical meaning.

Chapter 2

Black Holes and General

Relativity
The black holes of nature are the most perfect macroscopic objects there
are in the universe: the only elements in their construction are our concepts of
space and time.
S. Chandrashekar

Black holes1 were at first only a speculation as a result of calculation by Laplace

in 17952 when he discussed the classical body with escape velocity greater than
that of light, but his idea did not attract much attention. Later in 1916 Karl
Schwarzschild was able solve the Einstein field equation in the vacuum for uncharged
spherical systems and his solution is known as the Schwarzschild solution, which
implies the simplest black hole type namely the Schwarzschild black hole that is
only determined by a single parameter, namely its mass M .
The physics of black holes is largely based on Einsteins general theory of relativity (GR), which is a theory of gravitation. Relativity is a geometrical theory because
the mathematical study of spacetime, whether curved or flat, is geometry 3 . There
is no doubt that GR (including SR as well since SR is just a special solution of GR)
is one of the most successful physical theories of mankind (it of course cannot tell
everything about the universe4 , ) owing to its predictions that have been confirmed
by a number of experimental tests.5
In GR physics is expressed in terms of tensors. Each kind of spacetime is assigned
with the proper time and space interval which are described by the line element or
metric which is an invariant interval. For example, in SR there is a Minkowski line
element which takes the form
1

Alternative phrases for the black holes used in the pre-1969 literature were frozen stars and
collapsed stars see [23].
2
At that time the speed of light could not be precisely measured and there was no concept of
the speed of light as the ultimate speed in the Universe. Einsteins special theory of relativity (SR)
was not publicly known until 1905.
3
It can simply be said that in SR we deal with the Euclidean geometry, which is a geometry of
flat spacetime. Whereas in GR we deal with the non-Euclidean geometry which is the Riemannian
spacetime.
4
There have been attempts to create a Theory Of Everything (TOE), but it is still far from an
ultimate success due to the lack of knowledge of gravity at the quantum level.
5
See [39] for example.

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Figure 2.1: Light cones in Minkowski spacetime. Any point in the future light cone
r = t can be reached by a particle or signal with speed less than c.

ds2 = dt2 + dx2 + dy 2 + dz 2

(2.1)

ds2 = dx dx

(2.2)

or in a tensorial form
where

1
0
=
0
0

0
1
0
0

0
0
1
0

0
0
= diag(1, 1, 1, 1)
0
1

(2.3)

and is called the Minkowski metric tensor or just the Minkowski metric for
this particular Cartesian coordinates. This metric can turn out to be a positive,
negative and zero quantityit corresponds to a space-like, time-like and light-like
metric interval repectively. If we use a general coordinate system6 ,
ds2 =

3
X

g dx dx = g dx dx

(2.4)

,=0

The summation sign is suppressed when one uses the Einstein summation convention. For example, in spherical polar coordinates we have x = (t, r, , ) thus the
metric reads
ds2 = dt2 + dr2 + r2 d2 + r2 sin2 d2
(2.5)
or g = diag(1, 1, r 2 , r2 sin2 ) which is a flat metric7 since it can be rewritten
using flat coordinates.
6
It is customary to use Greek letters to denote spacetime coordinates, whereas the Latin letters
are used for space or other coordinates.
7
We can also check whether the metric is flat by considering the Riemann tensor which is a
curvature tensor whose all components become zero if and only if the metric is flat.

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

When we deal with the curved spacetime due to the presence of matter, the
Riemann tensor plays a major role as seen in GR. One very important equation in
this subject is the Einstein field equation, a tensorial equation which takes the form
G = 8T

(2.6)

G is an Einstein tensor which is symmetric and vanishes when spacetime is flat.

T is the so-called energy-momentum tensor which can be thought of as a source
for the gravitational field. It is a divergenceless tensor due to the conservation of
energy, namely T = 0. The proportionality constant is 8 since we use the
natural units, otherwise it would be 8G
. Mathematically, the Einstein tensor is
c4
given by
1
G = R g R
(2.7)
2
where R is called Ricci tensor which is a contraction of the Riemann tensor (R =

R
), and R is a curvature scalar obtained from the Ricci tensor, hence called
Ricci scalar. The full form of the Einstein equation has an extra term owing to the
Cosmological constant () which has been found recently to be an extremely tiny
number but non-zero. It reads
G + g = 8T

(2.8)

The significance of the cosmological constant is involved mostly in the context of

cosmology in which one studies the fate of the universe. The cosmological constant
will appear again when we discuss the BTZ black hole in Section 2.2.

2.1

Black hole metrics

The geometry of a spherically symmetric vacuum, i.e. vacuum spacetime outside

the spherical black hole is the Schwarzschild geometry describable in terms of the
Schwarzschild metric 8

2M 1 2
2M
2
2
dt + 1
ds = 1
dr + r2 d2
(2.9)
r
r
where d2 = d 2 +sin2 d2 . The Schwarzschild metric is indeed a gravitational field
and seems to have a singularity at the surface where r = 2M due to its coordinates,
in which space and time change their meanings. To get an idea of the magnitude
of the Schwarzschild radius we note that for the Earth it is 0.1 cm and for the
star of the size of the Sun, it is about 3 kilometers. The singularity at r = 2M
is however removable by chosing coordinates cleverly, an alternative to that is the
Kruskal-Szekeres coordinates found in 1960, which represents the spacetime more
properly. The surface of the black hole is entitled an event horizon for the fact
that nothing can be seen beyond it. Only region on and outside the the black holes
surface, r 2M is observationally relevant. The Schwarzschild metric is necessarily
8

This vacuum solution is always static in the exterior region of the black hole, c.f. Birkhoffs
theorem.

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Figure 2.2: A trajectory of a particle or signal approaching the Schwarzschild black

hole as represented in the Minkowski spacetime. The surface r = 2M is the event
horizon, nothing inside surface is able to escape. This trajectory is given by solving
the Schwarzschild metric ds2 = 0 (null geodesic) with d2 = 0.
asymptotically flat 9 , that is, for large r,

2M
2M
ds2 1
dt2 + 1 +
dr2 + r2 d2
r
r

(2.10)

and in addition to the Eq. (2.10) one can show that the Newtonian gravity is merely
a limiting case of GR.
So far, we have seen that the Schwarzschild black hole depends only on the
mass. This simple black hole is, astronomically, a collapse of an uncharged and nonrotating star with spherical symmetry. The gravitational collapse of a non-spherical
star with non-zero net charge produces a somewhat different black hole which can
be characterized by the mass M , intrinsic angular momentum or spin J and electric
charge Q. It is found that the structure of a black hole is determined uniquely by
just three parameters, i.e. M , J and Q once it is in the final state10 . In this way it
can be said that black holes have no hair.11 The black hole in the final state with
only M and Q, has a gravitational field given by the Reissner-Nordstr
om metric
which takes the form

Q2
2M
Q2
2M
2
2
+ 2 dt + 1
+ 2
ds = 1
dr2 + r2 d2
(2.11)
r
r
r
r
where M and Q are the total mass and charge respectively as measured by a distant
observer. For the rotating uncharged black hole, (namely the black hole is characterized only by M and J) its geometry is given by the Kerr metric (usually represented
9
This means that its metric approaches the flat metric far away from the black hole, namely
gab ab as r .
10
Final state here stands for stationary state in which everything is settled.
11
Due to the fact that all traces of the matter that formed a black hole disappear except for
M, J and Q thereby black holes look the same, as first introduced by J. Wheeler in 1960s.

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Figure 2.3: The Schwarzschild black hole in the Kruskal-Szekeres coordinates have
no coordinate singularity hence represents the real spacetime where r = 0 is the
singularity which is the broken thick line in the figure.
in the Boyer-Lindquist coordinates) using12 a = J/M . The significance of the value
of a plays a role when the extremal case is considered, i.e.
a
M
a
M

= 0 There is no spin, hence reduced to the Schwarzchild case.

= 1 Extremal Kerr black hole is reached.

The Kerr metric takes the form:

2M r sin2
a2 sin2 2
dt

2a
dt d
2
2
2 2
(r2 + a2 )2 a2 sin2
2
2
sin

d
+
dr + 2 d2
+
2

ds2 =

(2.12)

where
= r2 2M r + a2

(2.13)

= r + a cos

(2.14)

Its event horizons are (assuming that

a2

M 2)

<
p
r = M M 2 a 2

(2.15)

The Kerr metric or Kerr solution is stationary and axially symmetric, and has double
surfaces, i.e., outer and inner surfaces. In between the event horizon and the static
limit lies the so-called Ergosphere 13 inside which nothing can remain stationary.
For the charged rotating black hole (named Kerr-Newman black hole), its geometry is given by the Kerr-Newman metric which is in the same form as Eq. (2.12)
but with
= r2 2M r + a2 + Q2
(2.16)
The event horizons of the Kerr-Newman black hole are
p
r = M M 2 Q 2 a 2
12

13

J/M is refered to as angular momentum per unit mass.

Ergo means energy.

(2.17)

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Figure 2.4: A rough sketch of the Kerr black hole which is surrounded by an ergosphere. The ergosphere is a region inside which nothing can remain stationary. The
angular momentum of the Kerr black hole is denoted by J.
for a2 < M 2 + Q2 . The extremal Kerr-Newman black hole is reached when a2 =
M 2 + Q2 .
It is notable that one would hardly observe charged black holes in nature because
the black hole is already discharged when it is in stationary state. The discharging
process takes place very rapidly. The time it takes to become stationary is called the
characteristic time which is approximately 105 MM second. Therefore, it is obvious
that the black hole of 100000 M only requires approximately 1 second to reach its
stationary state.

2.2

The BTZ Black Hole: The (2+1)-dimensional black

hole

Banados, Teitelboim and Zanelli discovered, in 1992, the black hole solution of
Einsteins equation with a negative cosmological constant, in 2+1 dimensions and
without couplings to matter [2, 3]. This discovery was rather surprising as there
was no speculation that there would exist a black hole solution in 2+1 dimensions
at that time. The BTZ black hole is known as a simple toy model for a number of
studies including the string and supergravity theory. The cosmological constant,
is written as l2 in the BTZ black holes metric which reads:
ds2 = N 2 (r)dt2 + N 2 (r)dr 2 + r2 (N dt + d)2
where
N 2 (r) = M +

J2
J
r2
+
, N = 2
2
2
l
4r
2r

(2.18)

(2.19)

with < t < , 0 < r < and 0 2. N 2 (r) and N are the squared
lapse and angular shift respectively.
The event horizons can be obtained from N 2 (r) = 0 and takes the form:
r
M
r = l
(1 )
(2.20)
2
8

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Figure 2.5: The 2+1 dimensional BTZ black hole. This black hole can be visualized
as a circular disc with spin J. Its event horizon depends on its mass.
where
=

with imposed conditions that

J
Ml

M > 0 and |J| M l

(2.21)

(2.22)

In the extremal case J = |M l|, the two event horizons coincide. Note that l is the
radius of curvature which provides the length scale in order to have dimensionless
mass. If one lets l grows very large the black hole exterior is pushed away to infinity
and one will be left with the inside [2]. The BTZ black hole is similar to its 3+1
counterpart,
the Kerr solution. The BTZ black hole has an ergosphere, namely

rerg = l M and an upper bound in angular momentum for any given mass. The
spacetime geometry of the black hole is one of constant negative curvatures, so it is
locally that of anti-de Sitter (adS) space. The BTZ black hole can only differ from
the adS in its global properties [3]. The BTZ black hole is however different from its
counterpart by the fact that it has a positive heat capacity instead of the negative
heat capicity. Therefore its thermodynamic properties are well-defined and simple.
If we, for the sake of simplicity, set = 1 (hence l = 1) for the spinless BTZ
black hole, its metric is then reduced to

1 2
ds2 = (M + r 2 )dt2 + M + r 2
dr + r2 d2
(2.23)

The metric above is singular when r = M , which is an event horizon of the spinless
BTZ black hole.

2.3

Black holes and Thermodynamics

Black holes are found to be intimately related to ordinary thermodynamics in the

sense that their mechanical laws are seemingly identical to the laws of the thermody9

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Thermodynamic system
Temperature, T
Energy, E
Entropy, S

Black hole
Surface gravity,
Black holes mass, M
Area of event horizon, A

Table 2.1: Analogy between thermodynamic parameters and black holes parameters.
namics. In 1971 Stephen Hawking [14] stated that the area, A of the event horizon
of a black hole can never decrease (but can remain constant) in any process:
A 0

(2.24)

The area of the event horizon increases when (1) mass increases and (2) spin decreases. It was later noted by Bekenstein [5] that this result is analogous to the
statement of the ordinary second law of thermodynamics, namely that the total
entropy, S of a closed system never decreases in any process:
S 0

(2.25)

With these arguments it is legitimate to establish the laws of black hole mechanics
in parallel to the laws of ordinary thermodynamics by using parameters of the black
hole (see Table 2.1) as follows:

Zeroth law: The event horizon is described by a quantity, , the surface

gravity which is constant over the event horizon. The surface gravity is related to
the physical temperature of the black hole14 (Hawking temperature) by
TH =

(2.26)

For the special case of the Schwarzschild black hole, where = 1/(4GM ), the
Hawking temperature becomes:
TH =

M
~
6.2 108
K
8GkB M
M

(2.27)

So this is utterly negligible for solar-mass black holes the black hole absorbs much
more from the microwave background radiation than it radiates itself. In the case of
the rotating Kerr black hole, the Hawking temperature is reduced by the rotation,
explicitly
TH =

1
~
~
M
~
=2 1+
<
2
2
2kB
8M
k
8M
kB
M a
B

(2.28)

where a = J/M . For the charged non-rotating Reissner-Nordstrom black hole,

one has

~
Q4
~
~
TH =
= 1 4
<
(2.29)
2kB
8M kB
r+ 8M kB
14

TH =

~
2kB

when natural units are not used.

10

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Thus, electric charge also reduces the Hawking temperature. As a conclusive remark
one can safely say that the Hawking radiation plays no role in the case of large-sized
black holes. The only type of black hole where one can hope to observe such the
radiation is the so-called mini black hole, which is believed to have existed in the
primordial stage of the Universe.

First law: This law deals with the mass (energy) change, dM when a black
hole switches from one stationary state to another.

dM =
dA + work terms
(2.30)
8
or
dM = TH dSbh + work terms
(2.31)
It is readily seen that the above equations are analogous to the first law of thermodynamics, i.e.
dE = T dS + work terms
(2.32)
And the entropy of the black hole is thus represented by a quarter of the area of the
event horizon, that is
A
(2.33)
Sbh =
4
The factor 41 was indeed found by Hawking [16] based on the application of the
quantum field theory to the black holes which shows that they will absorb and emit
particles as if they were thermal bodies with the Hawking temperature, T H . Sbh is
sometimes called Bekenstein-Hawking entropy in order to honor their discoveries.
The work terms are given differently depending on the type of the black holes.
For the Kerr-Newman black hole family, the first law would be

dA + dJ + dQ
(2.34)
dM =
8
where is the angular velocity of the hole and is the electric potential which are
defined by
M
=
(2.35)
J
M
=
(2.36)
Q

Second law:

In any classical process, the area of the event horizon does not

decrease
dA 1

(2.37)

nor does the black holes entropy, Sbh . The second law of black hole mechanics
can, however, be violated if the quantum effect is taken into account, namely that
the area of the event horizon can be reduced via Hawking radiation. Note that
the proof of this depends on the Cosmic censorship conjecture 15 . It is essential
15
The cosmic censorship conjecture was made by the British mathematician Roger Penrose, which
states that in the universe black holes enshroud singularities so that no information about singularities can reach an outside observers [30]. This issue was seriously investigated by Christodoulou
and Hawking. And this is indeed the number-one open question in classical GR.

11

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Law
Zeroth law
First law
Second law
Third law

Thermodynamic system
T constant on a body
in thermal equilibrium
dE = T dS p dV + dN
dS 0
T = 0 cannot be reached

Black hole
constant over a
black holes event horizon

dA + dJ + dQ
dM = 8
dA 0
= 0 cannot be reached

Table 2.2: Analogy between the laws of thermodynamic and the laws of black-hole
mechanics.
that the black hole radiation is thermal in nature, therefore generates a rise in
entropy in the surrounding region. The generalized entropy, S 0 was introduced by
Bekenstein [5, 6, 36] to account for this sort of entropy. It is defined as the sum of
the black holes entropy, Sbh and the entropy of the surrouding matter, Sm
S 0 = Sbh + Sm

(2.38)

This statement is known as the Generalized Second Law (GSL):

S 0 0

(2.39)

The ordinary second law seems to fail when the matter is dropped into a black
hole because according to classical GR, the matter will disappear into a spacetime
singularity, in this manner the total entropy of the universe decreases as there is no
compensation for the lost entropy. The virtue of the GSL keeps the law of entropy
valid as the total entropy of the universe still increases when that matter falls into
the black hole.
It is natural to question the magnitude of the entropy of the black hole. When
expressed dimensionfully, the Bekenstein-Hawking entropy reads:
Sbh =

kB A
4G~

(2.40)

for the Schwarzschild black hole, this yields [11]

Sbh =

kB R02
G~

(2.41)

Numerically, the entropy of the sun is S 1057 kB whereas a solar-mass black hole
has an entropy of about 1077 kB which is 20 orders of magnitude larger!

Third law: The limit = 0 cannot be reached within a finite time, in other
words it is not possible how many processes we do, we will never reach reach the
limit = 0. However, the extremal black holes, for example the Kerr black hole in
which a/M = 1, do have = 0 thus zero temperature (absolute zero) but non-zero
entropy. To actually reduce the surface gravity to zero is merely an idealized case
because it is forbidden by the Cosmic censorship conjecture.
As far as physical theories are concerned, the laws of black hole mechanics have
only been phenomenological thermodynamical laws [11], thereby a number of open
questions can be raised, such as [17, 20, 37]:
12

CHAPTER 2. BLACK HOLES AND GENERAL RELATIVITY

Is Sbh real or subjective?

Where does it appearon or near the horizon or deep in the hole?
At what stage in the black holes evolution is it createdimmediately upon
formation by gravitation collapse, or only gradually over the long course of
evolution?
What is the dynamical mechanism that makes Sbh a universal function, independent of the holes past history or detailed internal condition?
When the quantum effects are taken into account, one can ask:
Can Sbh be derived from quantum mechanical considerations?
Due to the effect of Hawking radiation (black hole evaporation ), what happens
to Sbh after the black hole has evaporated? Will all the information disappear
after the evaporation?
These questions are somewhat embarrassing, because we do not know with our
present knowledge how to answer them precisely. Nevertheless, it is hoped that
success in modern theory of gravity, e.g., the quantum gravity or the string theory
would be the key to answerif not allsome of these open questions.

13

Chapter 3

Thermodynamic Fluctuation
Theory
The idea that time may vary from place to place is a difficult one, but it is the
idea Einstein used, and it is correctbelieve it or not.
-R. Feynman

Thermodynamics (see [9] for example) is an old subject in physics and it is

applicable to a wide variety class of systems. The ordinary laws of thermodynamics
are not fundamental in their own right, but are laws that arise from the microscopic
properties of the system. The utility of the thermodynamic laws is based on the fact
that they have a universal validity, at least for most systems.

3.1

The role of entropy

Entropy is, in a sense, a measure of the disorder of a system. This quantity was
first introduced by R. Clausius in 1850 as the amount of heat reversibly exchanged
at a temperature T . Entropy undoubtedly plays a major role in thermodynamics
and statistical mechanics. It is also the most characteristic extensive parameter (see
Appendix A) in thermodynamics, namely when it is expressed in terms of other
extensive parametersit basically tells us the physics that underlines the system.
Entropy is a part of the first and the second law of thermodynamics directlyit
enters the first law to complete the differential representation of the internal energy,
namely
dU = T dS p dV + dN
(3.1)
We can also expressed the entropy as a thermodynamic potential as
dS =

1
p

dU + dV dN
T
T
T

(3.2)

which is just the differential form of the entropy. In this view, if the dependence of
the entropy S(U, V, N ) on the variables U, V, N is known, then complete knowledge
of all the thermodynamic parameters is obtained. Furthermore, the entropy tells us
that for isolated systems (where dQreversible = 0) in equilibrium
dS = 0 S = Smaximum
14

(3.3)

CHAPTER 3. THERMODYNAMIC FLUCTUATION THEORY

and for irreversible processes

dS > 0

(3.4)

so in words it says that the state of equilibrium is defined as the state of maximum
entropy. Now as an example we have the entropy of the ideal gas as [13]
(
" #)
T 5/2 p0
(3.5)
S(N, T, p) = N kB s0 (T0 , p0 ) + ln
T0
p
where s0 (T0 , p0 ) is an arbitrary dimensionless function of the state (T0 , p0 ), which
is the result of the integration since Eq. (3.34) is derived from the second law of
thermodynamics when we use P V = N kB T and U = 23 N kB T . It is obvious that
Eq. (3.34) gives us a full information about the ideal gas. The entropy of the ideal
gas can also be expressed in terms of the other extensive parameters, i.e. S(N, V, U )
and it reads [7]
(
" #)
N0 5/2 U 3/2 V
S(N, V, U ) = N kB s0 (N0 , V0 , U0 ) + ln
(3.6)
N
U0
V0
From this equation, all equations of state of the ideal gas can be obtained by partial
differentiation. Therefore by differentiating Eq. (3.6) with respect to the internal
energy yields

S
1
3
1
3
= = N k U = N kT
(3.7)
U N,V
T
2
U
2
whereas differentiating it with respect to the volume we get the equation of state:

p
1
S
= = N k pV = N kT
(3.8)
V N,U
T
V
From statistical mechanics the entropy of the system is given by the natural
logarithm of the number of microscopic states1 which reads (with the Boltzmanns
constant being unity):
S = ln
(3.9)
The microstate is a function of the macrostate, i.e., (U, V, N ) hence the entropy
is a function of these variables. Eq. (3.9) is very important for it provides the basic
connection between macroscopic thermodynamics (entropy) and statistical microscopic physics (number of microscopic states). Notice that S = 0 when = 1, thus
there is only one exact microstate, hence no disorderand no entropy is created.

3.2

Fluctuations

Physical quantities which describe a macroscopic body in equilibrium are, almost

always, close to their mean values. However there are certain small deviations from
the mean values, which is the natural behavior of the system. These deviations
are known as thermodynamic fluctuations. The problem that arises is to find the
probability distribution of these deviations.
1

A microstate is the specification of the quantum numbers of all the atomic constitutents of the
subsystem. A microstate generally determines uniquely a macrostate, but not conversely.

15

CHAPTER 3. THERMODYNAMIC FLUCTUATION THEORY

When Eq. (3.9) is inverted2 one gets
= eS

(3.10)

which is the starting point of thermodynamic fluctuation theory indeed as it was first
done by Einstein. We, however, preferably denote P as the probability distribution,
i.e. P eS , we can Taylor expand the entropy about the fluctuation quantity x
only up to the second order as
1
S(x) = S(0) x2
2
where =
S
x

2S
x2

(3.11)

and S(0) = 0 since the entropy S has a maximum for x = 0 as

= 0. Substituting Eq. (3.11) into Eq. (3.10) yields:

1

P (x) = A e 2 x

(3.12)

In differential form it reads:

P (x) dx = A e 2 x dx

(3.13)

R
The constant A is given by the normalization condition that P (x) dx = 1. The
integration
limit is over all space, i.e., from to . This constant is found to be
p
/2 by Gaussian integration formula. Thus the probability distribution of the
various values of the fluctuation reads:
r
1 x2
P (x) =
e 2
(3.14)
2
This probability distribution is catagolized as a Gaussian distribution . It reaches
a maximum when x = 0 and decreases rapidly and symmetrically as |x| increases.
The mean squared fluctuation is defined as
Z
2
1
x = dx P (x)x2 =
(3.15)

We can then write the Gaussian distribution as

1
x2
p
P (x)dx =
dx
(3.16)
exp
2 hx2 i
2 hx2 i

It is then readily seen that, the smaller the x2 , the sharper the maximum of
P (x), which is the characteristics of the Gaussian distribution. Let us now consider
the Gaussian distribution for more than one variable, namely we will determine
simultaneous deviation of several thermodynamic quantities from their mean values.
We define the entropy S(x1 , . . . , xn ) as a function of the quantities of a simultaneous
deviation, and Taylor expand S in the same manner as done in Eq. (3.11) to the
second-order, thus
n
1 X
S S0 =
ij xi xj
(3.17)
2
i,j=1

This formula was first applied to the study of fluctuations by Einstein in 1907.

16

CHAPTER 3. THERMODYNAMIC FLUCTUATION THEORY

note that ij = ji . For simplicity we will omit the summation sign, we thus write
1
S S0 = ij xi xj
2

(3.18)

and the probability takes the form

1
P = A exp ij xi xj
2

(3.19)

A is the normalization constant which is determined from the normalization condition namely,
Z
Z
Z
(3.20)
dx1 dx2 . . . dxn P (x1 , . . . , xn ) = 1

After some algebraic manipulation (see [21], p. 335-338) we get

A=
(2)n/2

(3.21)

Therefore the desired form of the Gaussian distribution for more than one variable
reads

exp ij xi xj
P =
(3.22)
2
(2)n/2
where
ij =

3.3

2S
, = |ij |
xi xj

(3.23)

Thermodynamics and Geometry

Ruppeiner [28] in 1979 proposed a new way to study thermodynamics by using a

Riemannian geometrical model, in other words, he claims that thermodynamic systems can be represented by Riemannian geometry and certain statistical properties
can be derived from the model. This geometrical model is based on the inclusion of
the theory of fluctuations into the axioms of equilibrium thermodyanmics, namely
there exist equilibrium states which can be represented by points in two-dimensional
surface (manifold) and the distance between these equilibrium states is related to
the fluctuation between them. This concept is associated to probabilities, i.e., the
less probable a fluctuation between states, the further apart they are. This can be
recognized if one considers ij in the distance formula (line element) between the
two equilibrium states
X
ds2 =
ij dxi dxj = ij dxi dxj
(3.24)
ij

where the matrix of coefficients ij is the metric tensor, which is symmetric. We call
a manifold with a rule for distance in the form of Eq. (3.24) a Reimannian manifold.
If we now define
S
= ij xj
(3.25)
Xi =
xi
we will find that [21]

hxi xj i = ij = X i X j
(3.26)
17

CHAPTER 3. THERMODYNAMIC FLUCTUATION THEORY

which is the second moment for the fluctuations or the pair correlation function.
We refer the variables X i as thermodynamically conjugate to xi and we will assign
Sij to be the metric tensor intead of ij . The probability distribution can then be
rewritten as

S
1
i j
P (X) =
exp Sij X X
(3.27)
2
(2)n/2

where

2S
, S = |Sij |
(3.28)
X i X j
It is not entirely clear from the outset how this can be used to describe thermodynamic systems, but if we limit ourselves to the coordinates which are extensive
parameters in ordinary thermodynamics, we will be able to obtain some thermodynamic properties according to Ruppeiner [28]. The following sections will discuss
this:
Sij =

3.3.1

Ruppeiner metrics

The Ruppeiner metric is defined as the second derivatives of the entropy of the system with respect to the internal energy and other extensive charges (thermodynamic
parameters) , namely Qi = (M, N a ), it reads:
Sij =

2S
Qi Qj

(3.29)

N a is a conserved quantity and a = 1, . . . , n. An interesting fact about this metric

is that its inverse form is
2
S ij =
(3.30)
i j

where i = T1 , Ta which are the conjugate variables. is the Legendre transformation of the entropy S, i.e.
=

M
a
G
= S +
Na
T
T
T

(3.31)

where G is the Gibbs free energy defined as:

G = M T S a N a

(3.32)

The sum of the entropy and its Legendre transformation is equal to the product of
the extensive charges and the conjugate variables, namely
S + = i Qi

(3.33)

The Ruppeiner metric is physically meaningful in the equilibrium thermodynamic fluctuation theory, i.e., the thermodynamic Riemannian curvature, in short,
thermodynamic curvature for any given thermodynamic state tells us about the
underlying statistical structure. Calculations of the Ruppeiner metrics as well as
their Riemannian curvatures for different thermodynamic systems have been carried
out. As an example, the ideal gas at fixed volume has the Ruppeiner metric in the
following form [29]:

1 V
CV
2
dT
+
d2
(3.34)
ds2 =
T2
T 2 K T
18

CHAPTER 3. THERMODYNAMIC FLUCTUATION THEORY

where CV is the heat capacity at constant volume and KT is the isothermal compressibility of the system. Notice that we are in the coordinates (T, ) where ( N/V ),
is the density. To figure out whether this system is interacting or not, we can do
so either by coordinate transformation (i.e., rewriting it in the form of Euclidean
metric, which assures the flat space, i.e. the zero curvature) or computing the scalar
curvature of the metric in Eq. (3.34). It is not always possible to do the former
method if the given metric is very complicated, the formula for the latter is given in
Appendix B (Eq. B.9). It turns outfor the ideal gasthat it is a non-interacting
system, which we of course know from our exisiting knowledge. The curvature scalar
for the ideal gas with the particle number rather than the volume as the fixed scale
is also zero as found by Nulton and Salamon [26].

3.3.2

Weinhold metrics

Weinhold [38] was the first to introduce the geometrical concept into thermodynamics. In his approach the Riemannian metric is considered in the energy representation
where extensive parameters of the subsystem is given by
N i = (S, N a )

(3.35)

The Weinhold metric Wij is defined as the second derivatives of the internal energy
(in this case, mass M ) with respect to the entropy and N i as given in Eq. (3.35),
explicitly
2M
Wij =
(3.36)
N i N j
The contravariant form of the Weinhold metric is given by
W ij =

2G
i j

(3.37)

Where
G = M T S a N a

(3.38)

with i = (T, a ) and a= 1, . . . , n. Note that, in general, for black holes these
metrics will not be positive definite due to the fact that the ordinary black holes
have negative heat capacities [12]. The negative heat capacity is a property of the
isolated self-gravitating systems such as stars and astronomical systems [22]. Stars
and black holes display the same phenomenon in this aspect, their temperatures
increase as they lose energy, which is in accordance with the virial theorem.
Remarkably, the Weinhold and the Ruppeiner metrics are conformally related
via temperature T as the conformal factor, mathematically
Wij dN i dN j = T Sij dQi dQj

(3.39)

The conformal relation will be of great importance, when the Ruppeiner metric
seems to be very difficult to handle, by this we mean that instead of computing
the Ruppeiner metric, we will work out the Weinhold metric and turn it into the
Ruppeiner metric via T , which is defined as T = M
S . It is worth noting that the
Ruppeiner metric obtained by using the conformal relation will be in the Weinhold
coordinates, but its geometry is the Ruppeiner geometry.
19

Chapter 4

Results
The opposite of a correct statement is a false statement. But the opposite of a
profound truth may well be another profound truth.
N. Bohr

Having introduced the Ruppeiner and the Weinhold metrics in Chapter 3 we

can now compute them for 4 different families of black holes. In order to do so, we
need an expression for the entropy (for the Ruppeiner metrics) and the mass (for
the Weinhold metrics). We will also use the mass formula to calculate the intensive
parameters such as temeparature T , angular velocity and the electric charge
as well as the heat capacity C for each black hole. Some thermodynamic curvature
scalars obtained from the Ruppeiner metrics will be plotted and discussed.
Smarr [32] in 1973 found the mass formula which contains all the information
about the thermodynamic state of the Kerr-Newman black hole as
s

1 A
4
Q4
Q2
M=
+
J2 +
+
(4.1)
4 4
A
4
2
where J and Q are the black holes spin and the electric charge respectively. A is
the surface (event horizon) area of the black hole which is expressible in terms of
the holes entropy as
1
S = kB A
(4.2)
4
by inserting A = k4SB in Eq. (4.1) with kB = 1/; it is found that the algebra becomes
nicely simplifiedthis simplicity turns out to be extremely useful when we proceed
to compute the Ruppeiner and the Weinhold metrics of these black holes as it reduces
the degree of complication and tediousness. The simplified Smarrs mass formula
takes the form:
s

1
S
Q2
Q2
2
+
(4.3)
J +
+
M=
4
S
4
2
We will use this relation for obtaining certain thermodynamic properties of the
Reissner-Nordstrom, the Kerr and the Kerr-Newman black holes. We will start
with a theoretical BTZ black hole.

20

CHAPTER 4. RESULTS

Figure 4.1: Heat capacity of the BTZ black hole as a function of S and J. The
C(S, J) is a truly positive function for any values of S and J. It is notable that
at the extremal limit J = 2S 2 , which is the same as J = M the temperature
vanisheshence there is no heat capacity.

4.1

BTZ black hole

For the sake of simplicity we will use l = 1 (that is we use l = 1 in Eq. (2.20)).
Therefore we have the event horizons of the BTZ black hole in the form:
r
M
(1 )
(4.4)
r =
2
where
=

J
M

(4.5)

The entropy of the BTZ black hole (note that we use kB = 2/ for this particular
case) is
r
A
M
L
2r+
S=
=
=
=
(1 + )
(4.6)
2
2
2
2
where L is the length of the horizon r+ . By solving Eq. (4.6) for M we have the
mass in terms of the entropy for the BTZ black hole as
M = S2 +

J2
4S 2

(4.7)

The temperature of the BTZ black hole is given by

T =

M
J2
= 2S
S
2S 3

(4.8)

M
J
=
J
2S 2

(4.9)

and its angular velocity is

=

21

CHAPTER 4. RESULTS

Its heat capacity takes the form:

C =

T
M
S
T
=T
= T = 2
M
T
T
S
S 2

C =

S(4S 4 J 2 )
4S 4 + 3J 2

(4.10)

It is readily seen that the heat capactity of the BTZ hole is positive. We have found
that for this black hole it is more convenient to evaluate the Weinhold metric and use
the conformal relation to obtain the Ruppeiner metric. The Weinhold line element
reads

3J 2
2J
1
2
ds2W = 2 +
dS

dSdJ
+
dJ 2
(4.11)
2S 4
S3
2S 2
The Weinhold metric turns out to be a non-flat metric but we are not interested in
its curvature scalar, so we proceed to calculate the Ruppeiner metric for the BTZ
black hole by using ds2R = T1 ds2W which takes the form:

2J
S
4S 4 + 3J 2
2
2
dS
dSdJ +
dJ 2
(4.12)
dsR =
(4S 4 J 2 )S
4S 4 J 2
4S 4 J 2

and it is diagonalizable as
ds2R

S
1
2
dS +
du2
=
S
1 u2

(4.13)

with

J
, 1 u 1
(4.14)
2S 2
This metric yields a flat Riemannian curvature, i.e., the space of thermodynamic
state is a flat space.
u=

4.2

Reissner-Nordstr
om black hole

From the mass formula in Eq. (4.3), the Reissner-Nordstrom black hole has
r
1 Q2 Q2
S
+
+
(4.15)
M=
4
S 4
2
fortunately it can be simplified as

S
Q2
M=
1+
2
S

(4.16)

The temperature and the heat capacity for the Reissner-Nordstrom black hole are
as follow:

M
Q2
1
T =
1
(4.17)
=
S
S
4 S
and

2
1 QS

(4.18)
C = 2S
2
1 3Q
S
22

CHAPTER 4. RESULTS

Figure 4.2: The Ruppeiner metric for the RN black hole as seen in Minskowski
space when Rindler coordinates are used. The shaded area represents the Lorentzian
metric, i.e., the metric of ( +) signature.
and the electric potential takes a simple form:
=

Q
M
=
Q
S

(4.19)

It is obvious that the heat capacity for the Reissner-Nordstrom hole is a negative
quantity. Notice
that there is an absolute-zero temperature associated with the hole

when Q = S, which is the extremal limit. We next calculate the Weinhold metric
and obtain

1
3Q2
2
2
2
dsW =
1
dS 8QdSdQ + 8SdQ
(4.20)
3
S
8S 2
We can diagonalize this metric by using
Q
= , 1 1
S
which yields
ds2W =

(1 2 )dS 2 + 8S 2 d 2

(4.21)

(4.22)
8S
We can turn the Weinhold metric above into the the Ruppeiner metric in the same
coordinates via the conformal relation as

1
4S
ds2R =
dS 2 +
d 2
(4.23)
2S
1 2
3
2

which has the same properties as the Ruppeiner metric obtained directly from the
entropy representation where
r
Q2
S(M, Q) = 2M 2 Q2 + 2M 2 1 2
(4.24)
M
S(M, Q) is just an inversion of the mass formula in Eq. (4.16). By computing its
curvature scalar, it is found that the Ruppeiner metric for the RN black hole is a
23

CHAPTER 4. RESULTS

Fig.(A)

Fig.(B)

Figure 4.3: Heat capacity of the RNadS black hole as a function of entropy S and
electric charge Q for 2 different values of l, i.e., we have set l = 1 in Fig.(A) and
l = 0.01 in Fig.(B). The heat capacity of the RNadS black hole vanishes at the
extremal limit, whereas in the case of a small value of l (large ) as in Fig. (B) the
heat capacity becomes linear in the entropy.
flat metric, which is the same as that of the BTZ black hole. We now introduce new
coordinates

(4.25)
= 2S and sin =
2
The line element in Eq. (4.23) then reads
ds2 = d 2 + 2 d 2

(4.26)

This is a timelike wedge in a Minkowski space as seen in Rindler coordinates, see

Fig. (4.2).

For the sake of completeness we consider also the Reissner-Nordstrom black hole with
a negative cosmological constant (the Reissner-Nordstrom-anti-de Sitter or RNadS
for short) and we use the mass formula [8]:

Q2 J 2
A
A2
A
2
M2 =
+ (4J 2 + Q4 ) +
+ 2 +
+
Q
+
(4.27)
16
4
2
l
8l2
4 32 2 l2
A
with J = 0 and kB = 1/ (hence S = 4
) it simply becomes

S
Q2
S
1+ 2 +
M=
2
l
S

The temperature associated with the RNadS black hole is given by

1
3S
Q2
T =
1+ 2
l
S
4 S

(4.28)

(4.29)

which vanishes in the extremal limit, namely at

3S
Q2
=1+ 2
S
l
24

(4.30)

CHAPTER 4. RESULTS

as well as the heat capacity for the RNadS black hole, which takes the form
"
#

2
S 3S
2 + 1 Q
l

C = 2S
(4.31)
S 3S
1 + 3Q2
l2

The heat capacity of the RNadS black hole becomes linear in S in the limit of l .
We display heat capacities of this black hole for 2 different values of l in Fig. (4.3)
and its electric charge is given by
=

Q
M
=
Q
S

(4.32)

We are now in the position to compute the Ruppeiner metric for the RNadS black
hole; we will do so via the conformal relation as it is obviously simpler, in this case,
to calculate the Weinhold metric directly from the mass formula in Eq. (4.28), we
thus have

1
3S
3Q2
2
2
ds2W =

dS

8QdSdQ
+
8SdQ
(4.33)
3
l2
S
8S 2
this metric can be diagonalized using the same coordinates as in Eq. (4.25) which
yields the diagonal metric in the form

1
3 2
2
2
2
2
2
1 2 d + 2 d
(4.34)
ds =
2
2l
2
1 + 3

2
2l

This metric has a non-trivial geometry, namely there is a change of the signature of
the metric at
Q2
3S
2 =
=1 2
(4.35)
S
l
This corresponds to the stability properties of the thermodynamic system which
change for sufficiently large black holes [1]. We show this in the Fig. (4.4). From
the Ruppeiner metric we obtain the Riemannian curvature scalar in the form

Q2
Q2
3S
S
+
1

S
S
9
l2
l2
R= 2
(4.36)

2
l
Q2
Q2
3S
1 3S

1
+

S
S
l2
l2

which has a divergence in the extremal limit and along the curve where the metric
changes its signature. The 3D plot of this is in Fig. (4.5). This RNadS black hole
comes the ordinary RN black hole in the limit of l .

4.3

Kerr black hole

The mass formula for the Kerr black hole is obtained by setting Q = 0 in Eq. (4.3)
which reads
r
S J2
+
(4.37)
M=
4
S
25

CHAPTER 4. RESULTS

Fig.(A)

Fig.(B)

Fig.(C)

Figure 4.4: The RNadS black holes Ruppeiner geometrythe shaded areas represent the Lorentian metrics while the strips are the Riemannian metric which is a
positive definite metric with the (+ +) signature. Fig.(A) is when l = 1 which is
the when the cosmological constant = 1, Fig.(B) represents the geometry when
the value of l is large. In Fig.(C) we have l which means that the cosmological constant is vanishing thus it reduces to the case of RN black hole where its
Ruppeiner metric is a Lorentzian metric.

Fig.(A)

Fig.(B)

Figure 4.5: The Riemannian curvatures for the RNadS black hole as a function of
Q, S with 2 different values of l. In Fig.(A) we set l = 1 whereas Fig.(B) represents
the curvature where l = 0.001. There are discontinuities in both cases along the
extremal limits.

26

CHAPTER 4. RESULTS

The temperature temperature takes the form:

2

1 4J2
M
T =
= q S
2
S
4 S + 4JS

(4.38)

and it vanishes in the extremal limit, namely S = 2J or in terms of J and M it is

J = M 2 . The angular velocity of the hole is given by
=

2J
M
= q
J
S S+

4J 2
S

On inversion of Eq. (4.37) we obtain the entropy in the form

r
J2
2
2
S = 2M + 2M
1 4
M

(4.39)

(4.40)

hence we obtain the Ruppeiner metric in the following form

( "
#
)
3

4J
1
3J 2
J2 2
2
2
2
2
+ 1 4 dM
dM dJ +
dJ
dsR =
3 2 1 M 4
M
M3
M2
J2 2
1 M
4
(4.41)
by the virtue of the coordinate transformation it is reduced to the diagonal form as
!
!

2M 2
2
2
2
d 2
(4.42)
dM +
dsR = 2 1 + p
3
2
2
2
1
(1 )

where

J
(4.43)
M2
It turns out that the Ruppeiner geometry of this black hole is curvedit scalar
curvature is given by
q
=

1
R=
4M 2

J
1 M
4 2
q
J2
1 M
4

(4.44)

The above curvature scalar has a divergence in the extremal limit of the Kerr black
hole, namely at J = M 2 which is where the extremal limit of the black hole.

4.4

Kerr-Newman black hole

The Kerr-Newman black hole has the mass formula given by Eq. (4.3), namely
s

1
Q2
Q2
S
2
+
(4.45)
J +
+
M (S, J, Q) =
4
S
4
2
Thermodynamics of this black hole can be described by the variation in mass M as
dM = T dS + dJ + dQ
27

(4.46)

CHAPTER 4. RESULTS

Figure 4.6: The curvature scalar for the Kerr black hole as a function of M and J.
It is noticeable that a divergence of the curvature occurs along J = M 2 which is an
extremal limit of this black hole .

Figure 4.7: Heat capacity for the KN black hole for constant J and Q, we here set
J = Q = 1.

28

CHAPTER 4. RESULTS

which is indeed the second law of thermodynamics where the intensive parameters
T, and are given by
T =

S 2 + 4J 2 + Q4
1
1 (S 2 4J 2 + Q4 )
p
=
4 S S(4J 2 + S 2 + 2SQ2 + Q4 )
4
2M S 2

(4.47)

with the entropy S being just the inversion of Eq. (4.46) which takes the form:
r
J2
Q2
2
2
2
1 2 4
S = 2M Q + 2M
(4.48)
M
M
The other intensive parameters are its angular velocity
= p

2J
S(4J 2

S2

2SQ2

Q4 )

J
MS

(4.49)

It is readily seen that the larger the mass and entropy of the black hole, the smaller
the angular velocity, and as expected the angular velocity is proportional to the
angular momentum of the black hole. The electric potential takes the following
form:
Q(S + Q2 )
Q(S + Q2 )
= p
=
(4.50)
2M S
S(4J 2 + S 2 + 2SQ2 + Q4 )

This potential is reduced by mass and entropy of the black hole. As a matter of
fact, the KN black hole has a fairly complicated form of heat capacity, therefore we
will only show it graphically. The heat capacity is a function of S, J and Qfor
simplicity we will consider it for constant angular momentum and charge. Thus by
setting J = Q = 1 we have found that the heat capacity becomes linear in entropy,
see plot in Fig. (4.7).
The Ruppeiner metric of the Kerr-Newman black hole is very complicated and
too tedious to work out by hand, therefore we utilize a computer program (GRtensor)
to ease the calculation. We will not present it here since it does not give any
significance to the rest of this work. However we have found that the KN black
hole has curved geometries of both the Ruppeiner and the Weindhold metricsthe
Ruppeiner is found to be non-conformally flat [1].

29

Chapter 5

Conclusions and Speculations

Somewhere, there is something incredible waiting to be known.
C. Sagan

In this thesis work, thermodynamic Riemannian geometries of various black hole

families are studied, namely the Ruppeiner and Weinhold geometries. The Ruppeiner geometry is constructed on the basis of the fluctuation theory in equilibrium thermodynamicsand it is conformally related to the Weinhold geometry via
temperature in the mathematical expression. It is, however, only the Ruppeiner
geometry that tells us the statistical interaction underlying the system under consideration. Although the black hole is believed to be a thermodynamic system, its
statistical mechanical structure is lacking1 , partly due to own limited knowledge
of gravity at the quantum level. The Ruppeiner geometry for the BTZ and RN
families is flat (zero curvature), as expected because of the simpler structure they
possess The thermodynamic scalar curvature diverges in the extremal limit in the
Kerr, RNadS and KN families that possess a curved geometry. Most interestingly
perhaps, is the RNadS black hole family whose curvature is singular along the line
where the stability properties change, and it coincides with the RN family in the
limit of vanishing cosmological constant. The Weinhold geometry gives rise to no
physical meaningit is curved for the BTZ, RN and RNadS families whereas the
Kerr family has a flat Weinhold geometry. For the sake of tidiness, the obtained
results are tabulated in the following page. We believe that these results are sensible
and simpler than what one might have expected, although no statistical physics of
black holes can be derived from these findings at this point in time2 . Finally, it
is hoped that these results will find an interesting pattern in the future when the
quantum theory of black holes is more concrete.

There have been some results from string theory concerning the counting of the black hole
microstates, for example, see [18, 33].
2
As an encouraging remark, one may try to apply this calculation to black holes in string theory.

30

Black hole family

BTZ
Reissner-Nordstrom
Reissner-Nordstrom-anti-de-Sitter
Kerr
Kerr-Newman

Ruppeiner metric
Flat
Flat
Curved
Curved
Curved

Weinhold metric
Curved
Curved
Curved
Flat
Curved

Table 5.1: Geometries of the Ruppeiner and Weinhold metrics for various black
holes.

31

Appendix A: Thermodynamics in a nutshell

The succint description of thermodynamics as given by W. Greiner [13] is:
The task of thermodynamics is to define appropriate physical quantities (the state quantities), which characterize macroscopic properties of
matter, the so-called macrostate, in a way which is as unambiguous as
possible, and to relate these quantities by means of universally valid
equations (the equations of state and the laws of thermodynamics).
We will give here some terminologies used in thermodynamics, important equations
and descriptions of certain thermodynamic systems [9, 13, 40]:
Temperature: In thermodynamics the temperature (or thermodynamic temperature) is defined as a measure of the average kinetic energy of the particles
in a system. Adding heat to a system causes its temperature to rise. While
there is no maximum theoretically reachable temperature, there is a minimum
temperature, known as absolute zero, at which all molecular motion stops.
Extensive parameters or additive state quantities are the quantities that are
proportional to the amount of matter in a system, e.g. to the particle number
or mass. Obvious examples are the volume and the energy.
Intensive parameters or intensive state quantities are the quantities that are
independent of the amount of matter and are not additive. Examples are:
refractive index, density, temperature, pressure, chemical potential, etc. These
quantities can be locally defined, i.e. they may vary spatially.
Equations of state: We call certain relationships expressing intensive parameters in terms of the independent extensive parameters the equations of state.
For example
T

= T (S, V, N1 , . . . , Nr )

(A.1)

= P (S, V, N1 , . . . , Nr )

(A.2)

= (S, V, N1 , . . . , Nr )

(A.3)

Heat capacities: The heat capacity C of a substance is the amount of heat

required to change its temperature by one degree, and it has a unit of energy
per degree. The heat capacity is an extensive variable since a large quantity
of matter will have a proportionally large heat capacity.
The heat capacity at constant pressure is then defined by

dQ
S
CP =
=T
dT P
T P
while at the constant volume it reads:

dQ
S
E
CV =
=T
=T
dT V
T V
T V
32

(A.4)

(A.5)

transition temp

critical temp

Figure A.1: Phase transitions in T S diagram. On the left is the phase transition
of first order; it is obvious that there is a jump in entropy hence the divergence of
the heat capacity. The phase transition of second order is shown on the right in
the figure, there is a finite jump in entropy in this case. We assume the constant
pressure for both cases.
Reversible processes: These are processes which proceed over equilibrium states,
basically they are nonexistent.
Irreversible processes: In daily life we usually encounter such processes, thermodynamically speaking they are the processes which proceed by themselves
until their equilibrium states are reached.
Phase transitions: Phase transition is, easily speaking, an abrupt change of
the thermodynamic properties of the system. The most common example is
waterit is liquid at room temperature and atmospheric pressure, below the
freezing point it becomes solid and when heated above the boiling point it
vaporizes. At high pressures water can undergo several addition phase transitions from one solid to another, which makes the ices differ in their crystal
structures. We speak of the phase transitions as the change of the entropy
of the system (which differs in the phases). There are two kinds of phase
transitions, i.e.,
First-order phase transition: It is described by an additional heat supply
(or release) during the phase transition. The characterizations of the first-order
phase transitions are (i) a jump in entropy and (ii) the heat capacity tends to
infinity
Second-order phase transition: There is no jump in entropy, S = 0
S
changes discontinuously at the transition point. We can characterize
but T
the second-order phase transition by (i) there is a continous break point in
entropy and (ii) there is a finite jump of heat capacity
Gibbs free energy: The Gibbs free energy is defined as
G = U T S + work term

(A.6)

The work term can be P dV or dN for instance. Practically, G is an indicator of spontaneity of chemical reactions when its change is considered. The
33

reaction is at equilibrium if there is no change in G. If the change is positive

then the reaction is spontaneous, and vice versa.
In thermodynamics we have 3 types of systems, namely
Isolated systems: These do not interact in any way with the surroundings.
The container has to be impermeable to any form of matter or energy. In this
case, the total enery E (mechanical, electrical,. . .) is a conserved quantity of
the system and can be used to characterize the macrostate.
Closed systems: For the closed system, only the exchange of energy with the
surroundings is allowed and the energy is therefore not a conserved quantity.
The actual energy of the system will fluctuate with the surroudings. Nevertheless, if the closed system is in equilibrium with the surroundings the energy
will assume an average value with is related to the temperature of the system.
Open systems: These systems are the most liberal, namely that the exchange
of matter and energy with their surroundings are allowed.
Ensembles: In thermodynamics we have 3 types of ensembles which describe
the condition for the system:
Microcanonical ensemble: When a system is isolated, it is said to have
no interaction with its environment, therefore energy and partical number are
conserved quantities. Nevertheless, this cannot be realized entirely because
any wall in reality is heat-conducting.
Canonical ensemble: This is the case for a closed systemso there are
fluctuations due to the exchange of energy. We only have the particle number
as a conserved quantity.
Macrocanonical ensemble: It is more often called Grand canonical ensemble In the macrocanonical ensemble we consider the open system in equilibrium with its environment, in which energy and matter of the system can
be exchanged with surroundings. In this respect, certain mean values of enery
and particle number are established.

Appendix B: Curvature scalar calculation

We often deal with curvature scalars in GR, especially for this work we heavily
rely on them since the the curvature scalar of the Ruppeiner metric tells us what
statistical model lies beneath the system of interest. To compute them is not always
a simple taskit depends mainly on the degree of complication of the algebra and
the dimension of the metric tensor. In this thesis we deal with 2 and 3 dimensional
metrics and it has proven to be useful to have a formula for obtaining the curvature
scalar of the given metric. It is usual that we do have the metric tensor in a diagonal
form, therefore we will present here a procedure that helps us reduce such a metric

34

to a simpler form. The line element in 2 dimensions is usually given as

ds2 = A(u, v) du2 + 2B(u, v) du dv + C(u, v) dv 2
or
gij =

A(u, v) B(u, v)
B(u, v) C(u, v)

(B.1)

(B.2)

We wish to diagonalize it so that the cross term disappears; let be a variable such
that it satisfies:

B(u, v)2
1
2
2
(A(u, v) du + B(u, v) dv) + C(u, v)
dv 2 (B.3)
ds =
A(u, v) 2
A(u, v)
Notice that there is no change in quantity in Eq. B.3 from Eq. (B.1), only symbolically different. The diagonalized metric then reads:
ds2? = L(u, v) d 2 + M (u, v) dv 2

(B.4)

where
1
A(u, v) 2

B(u, v)2
M (u, v) =
C(u, v)
and
A(u, v)
d = A(u, v) du + B(u, v) dv
L(u, v) =

The desired diagonal metric is thus:

L(u, v)
?
gij =

M (u, v)

(B.5)
(B.6)
(B.7)

(B.8)

and we hereafter work with and v instead of u and v. So it should be obvious that
the task is indeed to find such that the above equations work. However finding
the is not always trivial. It is a matter of trial and errorsome experienced
relativists are able to foresee if it is possible to diagonalize the metric at all. Once
the diagonalized metric is at hand, one can proceed to compute the curvature scalar
of the metric using the following formula:

1 M (u, v)

1 L(u, v)
R=
+
(B.9)

g
g

v
g v
After obtaining the curvature scalar from the above formula; we can then substitute
, which is
Z
Z
=

A(u, v) du +

B(u, v) dv

(B.10)

so that we get the curvature scalar as a function of u and v. It is worth noting that if
is not properly chosen (found), the integration for will not be easily to evaluate.

35

Bibliography
man, J., Bengtsson, I and Pidokrajt, N., Geometry of black hole ther[1] A
modyanmics, General Relativity and Gravitation 3510, p. 1733-1743 (2003).
(gr-qc/0304015)
[2] Banados, M. et.al., Phys. Rev. Lett. 69, 1849 (1992).
[3] Banados, M. et.al., Phys. Rev. D48, 1506 (1993).
[4] Bardeen, J. M., Carter, N. and Hawking, S., Commun. Math. Phys. 31, 161
(1973).
[5] Bekenstein, J. D., Phys. Rev. D7, 949, 2333 (1973).
[6] Bekenstein, J. D., Phys. Rev. D9, 12 (1974).
[7] Benenson W, et. al., Handbook of Physics, Springer (New York 2002).
[8] Caldarelli, M. M., hep-th/9908022.
[9] Callen H. B., Thermodynamics and introduction to thermostatics, 2nd edition,
John Wiley & Son (Singapore 1985).
[10] Christodoulou, D., Phys. Rev. Lett. 25, 1596 (1971).
[11] Falcke, H. and Hehl, F. W., The galactic black hole, IoP (United Kingdom
2003).
[12] Ferrara, S. et. al., hep-tp/9702103.
[13] Greiner, W. et. al., Thermodynamics and Statistical Mechanics, Springer-Verlag
(New York 1995).
[14] Hawking, S. W., Phys. Rev. Lett. 26, 1344 (1971).
[15] Hawking, S. W., Nature (London) 248, 30 (1974).
[16] Hawking, S. W., Commun. Math. Phys. 43, 199 (1975).
[17] Hehl, F. W. et. al., Black holes: Theory and Observation, Springer (Germany
1998).
[18] Horowitz, G., Quantum States of Black Holes, gr-qc/9704072.
[19] Jacobson, T., Introductory Lectures on Black Hole Thermodynamics, University
of Utrecht.
36

[20] Kiefer, C., gr-qc/0110070.

[21] Landau, L. D., Lifshitz, E. M., Statistical Physics, Pergamon Press (Hungary
1980).
[22] Lynden-Bell, D. and Lynden-Bell, R. M., Mon. Not. R. astr. Soc.181, 405-409
(1977).
[23] Misner, C. W., Thorne, K. S. and Wheeler, J. A., Gravitation, W.H. Freeman
and Company (San Francisco 1973).
[24] Nieuwenhuizen, Th. M., cond-mat/0103063.
[25] Novikov, I. D. and Frolov, V. P., Physics of Black holes, Kluwer Academics
Publishing (The Netherlands 1989).
[26] Nulton, J. D. and Salamon, P., Phys. Rev. A31, 2520 (1985).
[27] Penrose, R. and Floyd R. M., Nature 229, 177 (1971).
[28] Ruppeiner, G., Phys. Rev. A20, 1608 (1979).
[29] Ruppeiner, G., Rev. Mod. Phys. 67, 605 (1995).
[30] Savani, J., Princeton Weekly Bulletin, April 14, 1997.
[31] Schutz, B. F., A first course in general relativity, Cambride University Press
(Cambridge 2001).
[32] Smarr, L., Phys. Rev. Lett. 30, 71 (1973).
[33] Strominger, A. and Vafa, C., Microscopic Origin of the Bekenstein-Hawking
Entropy, gr-qc/9601029
[34] Wald, R. M., General Relativity, The University of Chicago Press (Chicago
1984).
[35] Wald, R. M., Black holes and relativistic stars, The University of Chicago Press
(Chicago 1998).
[36] Wald, R. M., The Thermodynamics of Black Holes, Lecture Note, Enrico Fermi
Institute and Department of Physics, University of Chicago.
[37] Wald, R. M., http://www.livingreviews.org/Articles/Volume4/2001-6wald/,
cited on 19 February 2003.
[38] Weinhold, F., J. Chem. Phys. 63, 2479 (1975).
[39] Will, C. M., http://wugrav.wustl.edu/People/CLIFF/exptgravity.html,
cited on 17 February 2003.
[40] http://scienceworld.wolfram.com/physics/HeatCapacity.html , cited on
18 March 2003

37

Index
macrocanonical, 34
microcanonical, 34
entropy
Bekenstein-Hawking, 12
generalized, 12
equations of state, 32
ergosphere, 8
Euclidean geometry, 3
event horizon, 5, 9
extensive parameters, 18, 32

absolute zero, 12, 32

adS, 9
anti-de Sitter space, 9
asymptotically flat, 6
black hole
2+1 dimensional, 8
BTZ, 8
entropy, 1
evaporation, 13
extremal BTZ, 9
extremal Kerr-Newman, 8
Kerr, 6
Kerr-Newman, 8
mechanics, 10
mini, 11
no hair, 6
parameters, 10
physical temperature, 10
radiation, 12
Schwarzschild, 6
spinless BTZ, 9
thermodynamics, 9

frozen star, 3
Gaussian distribution, 16
gedanken experiment, 1
Generalized Second Law, 12
geometrized units, 2
Gibbs free energy, 18, 34
GSL, 12
Hawking
area theorem, 10
radiation, 1, 11, 13
temperature, 10
Hawking temperature, 11
heat capacity, 32

characteristic time, 8
closed system, 34
collapsed star, 3
conformal factor, 19
conformal relation, 19
Cosmic censorship conjecture, 11
cosmological constant, 5
curvature scalar, 2, 5, 34

intensive parameters, 32
invariant interval, 3
irreversible process, 33
isolated system, 34
Kruskal-Szekeres coordinates, 5

Einstein
field equation, 5
general relativity, 3
special theory of relativity, 3
summation convention, 4
ensemble, 34
canonical, 34
grand canonical, 34

Legendre transformation, 18
line element
Minkowski, 3
mean square fluctuation, 16
metric
Kerr, 7
Kerr-Newman, 7
38

Minkowski, 4
Reissner-Nordstrom, 6
Ruppeiner, 18
Schwarzschild, 5
Weindhold, 19
metric interval
light-like, 4
space-like, 4
time-like, 4
negative heat capacity, 9, 19
open questions in black hole thermodynamics, 12
open system, 34
Penrose, Roger, 11
phase transition, 33
positive heat capacity, 9
probability distribution, 15
reversible process, 33
Schwarzschild radius, 5
singularity, 5, 11, 12
Smarr
mass formula, 20
stationary state, 8
surface gravity, 10, 12
tensor
Einstein, 5
energy-momentum, 5
Ricci, 5
Riemann, 5
thermodynamic fluctuation, 15
thermodynamic Riemannian curvature,
2, 18
thermodynamics
ordinary laws, 12

39