Symmetries in General Relativity

Contents

1. Topics 2199

2. Participants 2201
2.1. ICRANet participants . . . . . . . . . . . . . . . . . . . . . . . . 2201
2.2. Past collaborators . . . . . . . . . . . . . . . . . . . . . . . . . . 2201
2.3. Ongoing collaborations . . . . . . . . . . . . . . . . . . . . . . . 2202
2.4. Students . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2202

3. Brief description 2203

3.1. Spacetime splitting techniques in
General Relativity . . . . . . . . . . . . . . . . . . . . . . . . . . 2203
3.1.1. “1+3” splitting of the spacetime . . . . . . . . . . . . . . 2203
3.1.2. Measurement process in general relativity . . . . . . . 2203
3.2. Motion of particles and extended bodies in general relativity . 2204
3.2.1. Test particles . . . . . . . . . . . . . . . . . . . . . . . . 2205
3.2.2. Spinning test particles . . . . . . . . . . . . . . . . . . . 2206
3.2.3. Particles with both dipolar and quadrupolar structure
(Dixon’s model) . . . . . . . . . . . . . . . . . . . . . . . 2206
3.2.4. Exact solutions representing extended bodies with quadrupo-
lar structure . . . . . . . . . . . . . . . . . . . . . . . . . 2206
3.3. Perturbations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2207
3.3.1. Curvature and metric perturbations in algebraically spe-
cial spacetimes . . . . . . . . . . . . . . . . . . . . . . . 2207
3.3.2. Curvature perturbations in type D spacetimes . . . . . 2208
3.3.3. Metric perturbations in a Reissner-Nordström spacetime 2208
3.3.4. Curvature and metric perturbations in de Sitter spacetime2208
3.3.5. Curvature perturbations due to spinning bodies on a
Kerr background . . . . . . . . . . . . . . . . . . . . . . 2209
3.3.6. Metric perturbations due to spinning bodies on a Schwarzschild
background . . . . . . . . . . . . . . . . . . . . . . . . . 2209
3.4. Cosmology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2209
3.4.1. Mixmaster universe and the spectral index . . . . . . . 2209
3.4.2. Wave equations in de Sitter spacetime . . . . . . . . . . 2210
3.5. Exact solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2210
3.5.1. Kerr-Schild ansatz revisited . . . . . . . . . . . . . . . . 2210
3.5.2. Rational metrics . . . . . . . . . . . . . . . . . . . . . . . 2210

2195
Contents

4. Publications (2009 – 2013) 2213

A. Spacetime splitting techniques in general relativity 2235

A.1. Observer-orthogonal splitting . . . . . . . . . . . . . . . . . . . 2237
A.1.1. The measurement process . . . . . . . . . . . . . . . . . 2238
A.2. Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2238
A.3. Comparing measurements by two observers in relative motion 2242
A.3.1. Maps between the LRSs of different observers . . . . . 2242
A.4. Comparing measurements by three or more observers in rela-
tive motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2243
A.5. Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2244
A.6. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2245
A.6.1. Test-particle motion . . . . . . . . . . . . . . . . . . . . 2245
A.6.2. Maxwell’s equations . . . . . . . . . . . . . . . . . . . . 2246

B. Motion of particles and extended bodies in General Relativity 2249

B.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2249
B.2. The Mathisson-Papapetrou model . . . . . . . . . . . . . . . . 2250
B.3. The Dixon-Souriau model . . . . . . . . . . . . . . . . . . . . . 2251
B.4. Particles with quadrupole structure . . . . . . . . . . . . . . . . 2252
B.5. Null multipole reduction world line: the massless case . . . . 2253
B.6. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2255
B.6.1. The special case of constant frame components of the
spin tensor . . . . . . . . . . . . . . . . . . . . . . . . . . 2255
B.6.2. Spin precession in Schwarzschild and Kerr spacetimes 2257
B.6.3. Massless spinning test particles in vacuum algebraically
special spacetimes . . . . . . . . . . . . . . . . . . . . . 2257
B.6.4. Quadrupole effects in black hole spacetimes . . . . . . 2257
B.6.5. Quadrupole effects in gravitational wave spacetimes . 2258
B.6.6. Quadrupolar particles and the equivalence principle . 2259
B.6.7. Poynting-Robertson-like effects . . . . . . . . . . . . . . 2259

C. Metric and curvature perturbations in black hole spacetimes 2261

C.1. Perturbations of charged and rotating Black hole . . . . . . . . 2261
C.2. Perturbations of a Reissner-Nordström black hole by a massive
point charge at rest and the “electric Meissner effect” . . . . . 2270
C.3. Perturbations of a de Sitter spacetime . . . . . . . . . . . . . . 2282
C.3.1. Electric multipoles . . . . . . . . . . . . . . . . . . . . . 2282
C.3.2. Magnetic multipoles . . . . . . . . . . . . . . . . . . . . 2284

D. Cosmology 2287

2196
 Contents

E. Exact solutions 2289

E.1. Kerr-Schild metrics and the Kerr-Schild ansatz revisited . . . . 2289
E.1.1. Modified ansatz . . . . . . . . . . . . . . . . . . . . . . . 2290
E.2. Rational metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 2293

Bibliography 2295

2197
1. Topics
• Spacetime splitting techniques in General Relativity.
1. “1+3” splitting of the spacetime
2. Measurement process in General Relativity

• Motion of particles and extended bodies in General Relativity.

1. Test particles
2. Spinning test particles
3. Particles with both dipolar and quadrupolar structure (Dixon’s model)

• Perturbations
1. Curvature and metric perturbations in algebraically special space-
times
2. Curvature perturbations in type D spacetimes
3. Metric perturbations in a Reissner-Nordström spacetime
4. Curvature and metric perturbations in de Sitter spacetime
5. Curvature perturbations due to spinning bodies on a Kerr back-
ground

• Cosmology
1. Mixmaster universe and the spectral index
2. Wave equations in de Sitter spacetime

• Exact solutions
1. Kerr-Schild ansatz revisited
2. Rational metrics

2199
2. Participants
2.1. ICRANet participants
• Bini Donato (ICRANet and Istituto per le Applicazioni del Calcolo “M.
Picone,” CNR, Italy)

• Cherubini Christian (ICRANet and University Campus Biomedico, Rome,

Italy)

• Geralico Andrea (ICRANet and University of Rome “La Sapienza,” Italy)

• Jantzen T. Robert (ICRANet and Villanova University, USA)

• Kerr P. Roy (ICRANet)

• Wen-Biao Han (ICRANet)

• Ruffini J. Remo (ICRANet and University of Rome “La Sapienza,” Italy)

2.2. Past collaborators

• Carini Paolo (high school science teacher in San Francisco)

• Cruciani Gianluca (ICRANet and high school teacher in Perugia, Italy)

• Ehlers Jurgen (Max Planck Institut für Gravitationsphysik, Golm, Ger-

many, † 2008)

• Merloni Andrea (post-doc at the Max Planck Institute for Extraterres-

trial Physics, Germany)

• Miniutti Giovanni (researcher at the Institute of Astronomy, Cambridge

UK)

• Spoliti Giuseppe (commercial director at RHEA Systems, S.A.)

• Boshkayev Kuantay (Irap PhD student)

2201
2. Participants

2.3. Ongoing collaborations

• De Felice Fernando (University of Padova, Italy)

• Fortini Pierluigi (University of Ferrara, Italy)

• Lusanna Luca (INFN, Florence, Italy)

• Ortolan Antonello (INFN, Padova, Italy)

• Stella Luigi (Astronomic Observatory of Rome, Italy)

• Esposito Giampiero (INFN, Napoli, Italy)

• Damour T. (IHES, Paris)

2.4. Students
• Gregoris Daniele (Irap PhD student)

• Haney Maria (Irap PhD student)

2202
3. Brief description
3.1. Spacetime splitting techniques in
General Relativity
Spacetime splitting techniques play a central role and have fundamental in-
terest in general relativity in view of extracting from the unified notion of
spacetime the separate classical notions of space and time, at the founda-
tion of all of our experience and intuition. Studying all the existing differ-
ent approaches scattered in the literature has allowed the creation a unique
framework encompassing all of them [1] and a more clear geometrical inter-
pretation of the underlying “measurement process” for tensors and tensorial
equations. “Gravitoelectromagnetism” is a convenient name for this frame-
work because it helps explain the close relation between gravity and electro-
magnetism represented by the Coriolis and centrifugal forces on one side and
the Lorentz force on the other side.

3.1.1. “1+3” splitting of the spacetime

During the last century, the various relativistic schools: Zelmanov, Landau,
Lifshitz and the Russian school, Lichnerowicz in France, the British school,
the Italian school (Cattaneo and Ferrarese), scattered Europeans (Ehlers and
Trautman, for example) and the Americans (Wheeler, Misner, etc.), developed
a number of different independent approaches to spacetime splitting almost
without reference to each other.
R. Ruffini [2], a former student of Cattaneo and a collaborator of Wheeler,
looking for a better understanding of black holes and their electromagnetic
properties, stimulated Jantzen, Carini and Bini to approach the problem and
to make an effort to clarify the interrelationships between these various ap-
proaches as well as to shed some light on the then confusing works of Abra-
mowicz and others on relativistic centrifugal and Coriolis forces. By putting
them all in a common framework and clarifying the related geometrical as-
pects, some order was brought to the field [1, 3, 4].

3.1.2. Measurement process in general relativity

The investigations on the underlying geometrical structure of any spacetime
splitting approach show that it is not relevant to ask which of these various

2203
3. Brief description

splitting formalisms is the “best” or “correct” one, but to instead ask what
exactly each one of them “measures” and which is specially suited to a par-
ticular application.
For instance, in certain situations a given approach can be more suitable
than another to provide intuition about or simplify the presentation of the
invariant spacetime geometry, even if all of them may always be used. These
ideas were then used to try to understand better the geometry of circular
orbits in stationary spacetimes and their physical properties where the con-
nection between general relativity and its Newtonian progenitor are most
natural.
The list of problems approached and results obtained together can be found
in Appendix A.

3.2. Motion of particles and extended bodies in

general relativity
The features of test particle motion along a given orbit strongly depend on the
nature of the background spacetime as well as on the model adopted for the
description of the intrinsic properties of the particle itself (e.g., its charge or
spin). As a basic assumption, the dimensions of the test particle are supposed
to be very small compared with the characteristic length of the background
field in such a way that the background metric is not modified by the presence
of the particle (i.e., the back reaction is neglected), and that the gravitational
radiation emitted by the particle in its motion is negligible. The particle can
in turn be thought as a small extended body described by its own energy-
momentum tensor, whose motion in a given background may be studied by
treating the body via a multipole expansion. Thus, a single-pole particle is a
test particle without any internal structure; a pole-dipole particle instead is a
test particle whose internal structure is expressed by its spin, and so on. The
equations of motion are then obtained by applying Einsteins field equations
together with conservation of the energy-momentum tensor describing the
body. For a single-pole particle this leads to a free particle moving along
the geodesics associated with the given background geometry. The motion
of a pole-dipole particle is instead described by the Mathisson-Papapetrou-
Dixon equations which couple background curvature and the spin tensor of
the field. The motion of particles with an additional quadrupolar structure
has been developed mostly by Dixon; because of its complexity, there are
very few applications in the literature. Finally, the discussion of the case in
which the test particle also has charge in addition to spin or mass quadrupole
moment is due to Dixon and Souriau and this situation has been very poorly
studied as well.
A complete list of the original results obtained and a deeper introduction

2204
 3. Brief description

to the models can be found in Appendix B.

3.2.1. Test particles

Since the 1990s we have been investigating the geometrical as well as physical
properties of circular orbits in black hole spacetimes, selecting a number of
special orbits for various reasons. These were already reviewed in a previous
ICRANet report on activities. A recent work has instead been to consider
a given gravitational background a (weak) radiation field superposed on it
and a test particle interacting with both fields. Interesting effects arise like
the Poynting-Robertson effect which have been considered in the framework
of the full general relativistic theory for the first time.
Poynting-Robertson effect can be briefly described as follows.
For a small body orbiting a star, the radiation pressure of the light emitted
by the star in addition to the direct effect of the outward radial force exerts
a drag force on the body’s motion which causes it to fall into the star un-
less the body is so small that the radiation pressure pushes it away from the
star. Called the Poynting-Robertson effect, it was first investigated by J.H.
Poynting in 1903 using Newtonian gravity and then later calculated in lin-
earized general relativity by H.P. Robertson in 1937. These calculations were
revisited by Wyatt and Whipple in 1950 for applications to meteor orbits,
making more explicit Robertson’s calculations for slowly evolving elliptical
orbits and slightly extending them.
The drag force is easily naively understood as an aberration effect: if the
body is in a circular orbit, for example, the radiation pressure is radially out-
ward from the star, but in the rest frame of the body, the radiation appears
to be coming from a direction slightly towards its own direction of motion,
and hence a backwards component of force is exerted on the body which acts
as a drag force. If the drag force dominates the outward radial force, the
body falls into the star. For the case in which a body is momentarily at rest,
a critical luminosity similar to the Eddington limit for a star exists at which
the inward gravitational force balances the outward radiation force, a critical
value separating radial infall from radial escape. Similarly for a body initially
in a circular orbit, there are two kinds of solutions: those in which the body
spirals inward or spirals outward, depending on the strength of the radiation
pressure.
We have considered [156, 162, 159] this problem in the context of a test
body in orbit in a spherically symmetric Schwarzschild spacetime without
the restriction of slow motion, and then in the larger context of an axially
symmetric Kerr spacetime while developing the equations for a more general
stationary axially symmetric spacetime. The finite size of the radiating body
is ignored.
We have also developed applications to cylindrically symmetric Weyl class

2205
3. Brief description

spacetimes (exhibiting a typical naked singularity structure) as well as to the

Vaidya radiating spacetime where the photon field is not a test field but the
source of the spacetime itself.

3.2.2. Spinning test particles

During the last five years we have investigated the motion of spinning test
particles along special orbits in various spacetimes of astrophysical interest:
black hole spacetimes as well as more “exotic” background fields represent-
ing naked singularities or the superposition of two or more axially symmetric
bodies kept apart in a stable configuration by gravitationally inert singular
structures.
In particular, we have focused on the so called “clock effect,” defined by
the difference in the arrival times between two massive particles (as well as
photons) orbiting around a gravitating source in opposite directions after one
complete loop with respect to a given observer [5, 6, 7].
We have also analyzed the motion of massless spinning test particles, ac-
cording to an extended version of the Mathisson-Papapetrou model in a gen-
eral vacuum algebraically special spacetime using the Newman-Penrose for-
malism in the special case in which the multipole reduction world line is
aligned with a principal null direction of the spacetime. Recent applications
concern instead the study of the Poynting-Robertson effect for spinning par-
ticles.

3.2.3. Particles with both dipolar and quadrupolar structure

(Dixon’s model)
We have studied the motion of particles with both dipolar and quadrupolar
structure in several different gravitational backgrounds (including Schwarz-
schild, Kerr, weak and strong gravitational waves, etc.) following Dixon’s
model and within certain restrictions (constant frame components for the
spin and the quadrupole tensor, center of mass moving along a circular orbit,
etc.).
We have found a number of interesting situations in which deviations from
geodesic motion due to the internal structure of the particle can give rise to
measurable effects.

3.2.4. Exact solutions representing extended bodies with

quadrupolar structure
We have investigated geometrical as well as physical properties of exact so-
lutions of Einstein’s field equations representing extended bodies with struc-

2206
 3. Brief description

ture up to the quadrupole mass moment, generalizing so the familiar black

hole spacetimes of Schwarzschild and Kerr.
Recent results involve the use of the equivalence principle to compare geo-
desic motion in these spacetimes with nongeodesic motion of structured par-
ticles in Schwarzschild and Kerr spacetimes, allowing an interesting analysis
which strongly support Dixon’s model.

3.3. Perturbations
A discussion of curvature perturbations of black holes needs many different
approaches and mathematical tools. For example, the Newman-Penrose for-
malism in the tetradic and spinor version, the Cahen-Debever-Defrise self-
dual theory, the properties of the spin-weighted angular harmonics, with
particular attention to the related differential geometry and the group the-
ory, some tools of complex analysis, etc. Furthermore, even using any of the
above mentioned approaches, this remains a difficult problem to handle. It is
not by chance, for instance, that the gravitational and electromagnetic pertur-
bations of the Kerr-Newman rotating and charged black hole still represent
an open problem in general relativity.
During the last years, however, modern computers and software have rea-
ched an exceptional computational level and one may re-visit some of these
still open problems, where technical difficulties stopped the research in the
past. Details can be found in Appendix C.

3.3.1. Curvature and metric perturbations in algebraically

special spacetimes

Most of the work done when studying perturbations in General Relativity

concerns curvature perturbations from one side or metric perturbations from
the other side. In the first case, one can easily deal with gauge invariant
quantities but the problem of finding frame-independent objects arises. Fur-
thermore, the reconstruction of the metric once the curvature perturbations
are known is a very difficult task. In the second case, instead, in order to start
working with an explicit metric since the beginning, the choice of a gauge
condition is necessary. Gauge independent quantities should therefore be
determined properly.
There exist very few examples of works considering both cases of curvature
and metric perturbations on the same level so that we have been motivated
to start working in this direction.

2207
3. Brief description

3.3.2. Curvature perturbations in type D spacetimes

In the Kerr spacetime Teukolsky [8] has given a single “master equation”
to deal with curvature perturbations by a field of any spin (“spin-weight,”
properly speaking). Then the problem of extending the results of Teukolsky
to other spacetimes is raised.
Actually, a very important result that we have obtained framed the Teukol-
sky equation in the form of a linearized de-Rham Laplacian equation for
the perturbing field [9, 10]. In addition, in all the cases (type D spacetime:
Taub-NUT, type D-Kasner, etc) in which an equation similar to the Teukolsky
equation can be written down, one can study the various couplings between
the spin of the perturbing field and the background parameters, i.e., spin-
rotation, spin-acceleration couplings, etc., which can also be relevant in dif-
ferent contexts and from other points of view. We have obtained important
results considering explicit applications to the Taub-NUT, Kerr-Taub-NUT, C-
metric, spinning C-metric, Kasner and de Sitter spacetimes. For example in
the Taub-NUT spacetime we have shown that the perturbing field acquires
an effective spin which is simply related to the gravitomagnetic monopole
parameter ` of the background [11]; in the C-metric case (uniformly accel-
erated black hole spacetime) we have been able to introduce a gravitational
analog of the Stark effect, etc.

3.3.3. Metric perturbations in a Reissner-Nordström

spacetime
We have recently solved the multiyear problem of a two-body system con-
sisting of a ReissnerNordström black hole and a charged massive particle
at rest at the first perturbative order. The expressions for the metric and the
electromagnetic field, including the effects of the electromagnetically induced
gravitational perturbation and of the gravitationally induced electromagnetic
perturbation, have been presented in closed analytic formulas; the details are
indicated in Appendix C. Motivated by our works an exact solution has then
been found by Belinski and AleKseev [149] of which our solution is a lin-
earization with respect to certain parameters.

3.3.4. Curvature and metric perturbations in de Sitter

spacetime
de Sitter spacetime is a conformally flat spacetime and therefore has the Petrov
type O. Due to the high number of symmetries as well as the particular sim-
plicity of the associated metric it results as the best arena to perform curva-
ture and metric perturbations simultaneously. In fact, taking advantage from
the possibility of writing the metric in a spherically symmetric form, one can

2208
 3. Brief description

apply the well known approaches of Regge-Wheeler-Zerilli to discuss metric

perturbations and translate then the results in terms of curvature. Closed an-
alytic expressions involving Heun hypergeometric functions can be obtained;
the details are scketched in appendix C.

3.3.5. Curvature perturbations due to spinning bodies on a

Kerr background
A new scheme for computing dynamical evolutions and gravitational ra-
diations for intermediate-mass-ratio inspirals (IMRIs) based on an effective
one-body (EOB) dynamics plus Teukolsky perturbation theory has been re-
cently derived by Wen-Biao Han and collaborators [169]. This research line
is very promising in view of many possible applications ranging from the
Post-Newtonian physics of binary systems to numerical relativity.

3.3.6. Metric perturbations due to spinning bodies on a

Schwarzschild background
The full reconstruction of the perturbed metric by a pinning particle mov-
ing on a Schwarzschild background is possibile following the original Zerilli
and Ruffini approach, at least perturbatively at various Post-Newtonian or-
ders. This research project is expected to add corrections due to spin to the
relativistic two body problem within the effective one-body formalism intro-
duced by Damour.

3.4. Cosmology
3.4.1. Mixmaster universe and the spectral index
We have recently revisited the Mixmaster dynamics in a new light, reveal-
ing a series of transitions in the complex scale invariant scalar invariant of
the Weyl curvature tensor best represented by the speciality index S, which
gives a 4-dimensional measure of the evolution of the spacetime independent
of all the 3-dimensional gauge-dependent variables except the time used to
parametrize it.
Its graph versus time with typical spikes in its real and imaginary parts cor-
responding to curvature wall collisions serves as a sort of electrocardiogram
of the Mixmaster universe, with each such spike pair arising from a single
circuit or “pulse” around the origin in the complex plane. These pulses in the
speciality index seem to invariantly characterize some of the so called spike
solutions in inhomogeneous cosmology and should play an important role

2209
3. Brief description

in the current investigations of inhomogeneous Mixmaster dynamics. This

interesting work will be certainly continue over the next few years.

3.4.2. Wave equations in de Sitter spacetime

Wave propagation on a de Sitter background spacetime can be considered
for both the electromagnetic and the gravitational case under the preliminar
choice of a gauge conditions. Usually, even in the recent literature, the dis-
cussion is limited to special gauge conditions only, like the harmonic one. Re-
cently, some interest has been raised instead for the development of a system-
atic study in terms of the de Donder gauge since this is close to the Lorentz
gauge of the electromagnetic case. Due to the particular symmetries of the de
Sitter spacetime we expect to be able to reconstruct the perturbed metric by
the wave propagation, at least in PN scheme.

3.5. Exact solutions

3.5.1. Kerr-Schild ansatz revisited
We have recently presented an alternative derivation of Kerr solution by
treating Kerr-Schild metrics as exact linear perturbations of Minkowski space-
time. Actually, they have been introduced as a linear superposition of the
flat spacetime metric and a squared null vector field k multiplied by a scalar
function H.
In the case of Kerr solution the vector k is geodesic and shearfree and it
is independent of the mass parameter M, which enters instead the definition
of H linearly. Without introducing any assumption on the null congruence
k and due to this linearity property, it is possible to solve the field equations
order by order in powers of H in complete generality. The Ricci tensor turns
out to consist of three different contributions: third order equations all im-
ply that k must be geodesic; the latter, in turn, must be also shearfree as a
consequence of first order equations, whereas the solution for H comes from
second order equations. This work is discussed in Appendix E.

3.5.2. Rational metrics

We have reconsidered the general form for stationary and axisymmetric met-
rics which are solutions of the vacuum Einstein’s equations. Using symmetry
adapted coordinates and decomposing properly the 4-dimensional metric as
a sum of 2 2-metrics, we have shown how certain solutions obtained long ago
by using generating techniques and other involved procedures may have a
simpler form with metric coefficients being rational (polynomial) functions.

2210
 3. Brief description

This research line is just started, following an idea due to Prof. Roy P. Kerr
(see the section “Kerr-Newman solution” of the present report)

2211
4. Publications (2009 – 2013)
Refereed journals

1. Bini D., Cherubini C., Filippi S.

On vortices heating biological excitable media,
Chaos, Solitons and Fractals vol. 42, 20572066, 2009
Abstract
An extension of the Hodgkin-Huxley model for the propagation of nerve
signal which takes into account dynamical heat transfer in biological
tissue is derived and ne tuned with existing experimental data. The
medium is heated due to the Joule’s eect associated with action poten-
tial propagation, leading to characteristic thermal patterns. The intro-
duction of heat transfer—necessary on physical grounds—provides a
novel way to directly observe movement of the tip of spiral waves in
numerical simulations and possibly in experiments regarding biologi-
cal excitable media.

2. Bini D., Jantzen R. T., Stella L.

The general relativistic Poynting-Robertson effect
Classical and Quantum Gravity, vol. 26, 055009, 2009.
Abstract
The general relativistic version is developed for Robertsons discussion
of the Poynting-Robertson effect that he based on special relativity and
Newtonian gravity for point radiation sources like stars. The general
relativistic model uses a test radiation field of photons in outward ra-
dial motion with zero angular momentum in the equatorial plane of the
exterior Schwarzschild or Kerr spacetime.

3. Bini D., Cherubini C., Geralico A., Jantzen R. T.

Electrocardiogram of the Mixmaster Universe
Classical and Quantum Gravity, vol. 26, 025012, 2009.
Abstract
The Mixmaster dynamics is revisited in a new light as revealing a se-
ries of transitions in the complex scale invariant scalar invariant of the
Weyl curvature tensor best represented by the speciality index S, which
gives a 4-dimensional measure of the evolution of the spacetime inde-
pendent of all the 3-dimensional gauge-dependent variables except the
time used to parametrize it. Its graph versus time with typical spikes in

2213
4. Publications (2009 – 2013)

its real and imaginary parts corresponding to curvature wall collisions

serves as a sort of electrocardiogram of the Mixmaster universe, with
each such spike pair arising from a single circuit or pulse around the
origin in the complex plane. These pulses in the speciality index seem
to invariantly characterize some of the so called spike solutions in inho-
mogeneous cosmology and should play an important role in the current
investigations of inhomogeneous Mixmaster dynamics.

4. Bini D., Cherubini, C., Geralico, A., Ortolan, A.

Dixon’s extended bodies and weak gravitational waves,
General Relativity and Gravitation, vol. 41, 105, 2009. Abstract
General relativity considers Dixons theory as the standard theory to
deal with the motion of extended bodies in a given gravitational back-
ground. We discuss here the features of the reaction of an extended
body to the passage of a weak gravitational wave. We find that the
body acquires a dipolar moment induced by its quadrupole structure.
Furthermore, we derive the world function for the weak field limit of
a gravitational wave background and use it to estimate the deviation
between geodesics and the world lines of structured bodies. Measuring
such deviations, due to the existence of cumulative effects, should be
favorite with respect to measuring the amplitude of the gravitational
wave itself.

5. Bini D., Esposito G., Montaquila R.V.,

The vector wave equation in de Sitter space-time
General Relativity and Gravitation,, vol. 42, 51-61, 2009.
Abstract
The vector wave equation, supplemented by the Lorenz gauge condi-
tion, is decoupled and solved exactly in de Sitter space-time studied
in static spherical coordinates. One component of the vector field is
expressed, in its radial part, through the solution of a fourth-order or-
dinary differential equation obeying given initial conditions. The other
components of the vector field are then found by acting with lower-
order differential operators on the solution of the fourth-order equa-
tion (while the transverse part is decoupled and solved exactly from
the beginning). The whole four-vector potential is eventually expressed
through hypergeometric functions and spherical harmonics. Its radial
part is plotted for given choices of initial conditions. This is an impor-
tant step towards solving exactly the tensor wave equation in de Sitter
space-time, which has important applications to the theory of gravita-
tional waves about curved backgrounds.

6. Bini D., Geralico A., Luongo O. and Quevedo H.,

Generalized Kerr spacetime with an arbitrary mass quadrupole moment: geo-
metric properties versus particle motion

2214
 4. Publications (2009 – 2013)

Classical and Quantum Gravity, vol. 26, 225006 (23pp), 2009.

Abstract
An exact solution of Einstein’s field equations in empty space first found
in 1985 by Quevedo and Mashhoon is analyzed in detail. This solu-
tion generalizes Kerr spacetime to include the case of matter with an
arbitrary mass quadrupole moment and is specified by three param-
eters, the mass M, the angular momentum per unit mass a and the
quadrupole parameter q. It reduces to the Kerr spacetime in the limit-
ing case q = 0 and to the ErezRosen spacetime when the specific angular
momentum a vanishes. The geometrical properties of such a solution
are investigated. Causality violations, directional singularities and re-
pulsive effects occur in the region close to the source. Geodesic motion
and accelerated motion are studied on the equatorial plane which, due
to the reflection symmetry property of the solution, also turns out to be
a geodesic plane.

7. Bini D., Cherubini C., Filippi S., Geralico A.

Extended bodies with quadrupole moment interacting with gravitational monopoles:
reciprocity relations
General Relativity and Gravitation, vol. 41, 2781, 2009.
Abstract
An exact solution of Einstein’s equations representing the static gravi-
tational field of a quasi-spherical source endowed with both mass and
mass quadrupole moment is considered. It belongs to the Weyl class of
solutions and reduces to the Schwarzschild solution when the quadrupole
moment vanishes. The geometric properties of timelike circular or-
bits (including geodesics) in this spacetime are investigated. Moreover,
a comparison between geodesic motion in the spacetime of a quasi-
spherical source and non-geodesic motion of an extended body also en-
dowed with both mass and mass quadrupole moment as described by
Dixon’s model in the gravitational field of a Schwarzschild black hole
is discussed. Certain “reciprocity relations” between the source and the
particle parameters are obtained, providing a further argument in fa-
vor of the acceptability of Dixon’s model for extended bodies in general
relativity.

8. Bini D. , C. Cherubini, S. Filippi, A. Gizzi and P. E. Ricci

On the universality of spiral waves
Communications in Computational Physics (CiCP), vol. 8, pp. 610-622,
2010.
Abstract
Spiral waves appear in many different contexts: excitable biological tis-
sues, fungi and amoebae colonies, chemical reactions, growing crystals,
fluids and gas eddies as well as in galaxies. While the existing theories

2215
4. Publications (2009 – 2013)

explain the presence of spirals in terms of nonlinear parabolic equa-

tions, in this paper it is shown that self-sustained spiral wave regime
is already present in the linear heat operator, in terms of integer Bessel
functions of complex argument. Such solutions, even if commonly not
discussed in the literature because diverging at spatial infinity, play a
central role in the understanding of the universality of spiral process.
As an example we have studied how in nonlinear reaction-diffusion
models the linear part of the equations determines the wave front ap-
pearance while nonlinearities are mandatory to cancel out the blowup
of solutions. The spiral wave pattern still requires however at least two
cross diffusing species to be physically realized.

9. Bini D., Geralico A., Kerr R.P.

The Kerr-Schild ansatz revised
International Journal of Geometric Methods in Modern Physics (IJG-
MMP) vol.7, 693-703, 2010.
Abstract
KerrSchild metrics have been introduced as a linear superposition of
the flat spacetime metric and a squared null-vector field, say k, mul-
tiplied by some scalar function, say H. The basic assumption which
led to Kerr solution was that k be both geodesic and shearfree. This
condition is relaxed here and KerrSchild Ansatz is revised by treating
KerrSchild metrics as exact linear perturbations of Minkowski space-
time. The scalar function H is taken as the perturbing function, so that
Einsteins field equations are solved order-by-order in powers of H. It
turns out that the congruence must be geodesic and shearfree as a con-
sequence of third- and second-order equations, leading to an alternative
derivation of Kerr solution.

10. Bini D. , Geralico A.

Spinning bodies and the Poynting-Robertson effect in the Schwarzschild space-
time
Classical and Quantum Gravity, vol. 27, 185014, 2010.
Abstract
A spinning particle in the Schwarzschild spacetime deviates from geodesic
behavior because of its spin. A spinless particle also deviates from
geodesic behavior when a test radiation field is superimposed on the
Schwarzschild background: in fact the interaction with the radiation
field, i.e., the absorption and re-emission of radiation, leads to a friction-
like drag force responsible for the well known effect which exists al-
ready in Newtonian gravity, the Poynting-Robertson effect. Here the
Poynting-Robertson effect is extended to the case of spinning particles
by modifying the Mathisson-Papapetrou model describing the motion
of spinning test particles to account for the contribution of the radia-

2216
 4. Publications (2009 – 2013)

tion force. The resulting equations are numerically integrated and some
typical orbits are shown in comparison with the spinless case. Further-
more, the interplay between spin and radiation forces is discussed by
analyzing the deviation from circular geodesic motion on the equato-
rial plane when the contribution due to the radiation can also be treated
as a small perturbation. Finally the estimate of the amount of radial
variation from the geodesic radius is shown to be measurable in princi-
ple.

11. Bini D. , C. Cherubini, S. Filippi, Geralico A.

Effective geometry of the n = 1 uniformly rotating self-gravitating polytrope
Physical Review D, vol. 82, 044005 2010.
Abstract
The “effective geometry” formalism is used to study the perturbations
of a perfect barotropic Newtonian self-gravitating rotating and com-
pressible fluid coupled with gravitational backreaction. The case of a
uniformly rotating polytrope with index n = 1 is investigated, due to
its analytical tractability. Special attention is devoted to the geometri-
cal properties of the underlying background acoustic metric, focusing
in particular on null geodesics as well as on the analog light cone struc-
ture.

12. Bini D. , Geralico A., Jantzen R.T.

Fermi coordinates in Schwarzschild spacetime: closed form expressions
General Relativity and Gravitation, vol. 43, 18371853, 2011.
Abstract
Fermi coordinates are constructed as exact functions of the Schwar-
zschild coordinates around the world line of a static observer in the
equatorial plane of the Schwarzschild spacetime modulo a single im-
pact parameter determined implicitly as a function of the latter coordi-
nates. This illustrates the difficulty of constructing explicit exact Fermi
coordinates even along simple world lines in highly symmetric space-
times.

13. Bini D., Geralico A., Jantzen R. T.

Spin-geodesic deviations in the Schwarzschild spacetime
General Relativity and Gravitation, vol. 43, 959-975, 2011.
Abstract
The deviation of the path of a spinning particle from a circular geodesic
in the Schwarzschild spacetime is studied by an extension of the idea
of geodesic deviation. Within the Mathisson-Papapetrou-Dixon model
and assuming the spin parameter to be sufficiently small so that it makes
sense to linearize the equations of motion in the spin variables as well
as in the geodesic deviation, the spin-curvature force adds an additional
driving term to the second order system of linear ordinary differential

2217
4. Publications (2009 – 2013)

equations satisfied by nearby geodesics. Choosing initial conditions for

geodesic motion leads to solutions for which the deviations are entirely
due to the spin-curvature force, and one finds that the spinning par-
ticle position for a given fixed total spin oscillates roughly within an
ellipse in the plane perpendicular to the motion, while the azimuthal
motion undergoes similar oscillations plus an additional secular drift
which varies with spin orientation.

14. Gizzi A., Bernaschi M., Bini D., Cherubini C., Filippi S., Melchionna S.,
Succi S.
Three-band decomposition analysis of wall shear stress in pulsatile flows
Physical Review E 83, 031902(10), 2011.
Abstract
Space-time patterns of Wall Shear Stress (WSS) resulting from the nu-
merical simulation of pulsating hemodynamic flows in semi-coronal
domains are analyzed, both in the case of regular semi-coronal domains
and semi-coronal domains with bumpy insertions, mimicking aneurysm-
like geometries. A new family of cardiovascular risk indicators, which
we name Three-Band Diagrams (TBD), are introduced, as a sensible
generalization of the two standard indicators, i.e. the time-averaged
WSS and the OSI (Oscillatory Shear Index). TBD’s provide a handy
access to additional information contained in the dynamic structure
of the WSS signal as a function of the physiological WSS risk thresh-
old, thereby allowing a quick visual assessment of the risk sensitiv-
ity to individual fluctuations of the physiological risk thresholds. Due
to its generality, TBD analysis is expected to prove useful for a wide
host of applications in science, engineering and medicine, where risk-
assessment plays a central role.

15. Bini D., Geralico A., Jantzen R. T., Semeřák O. and Stella L.
The general relativistic Poynting-Robertson effect II: A photon flux with nonzero
angular momentum
Classical and Quantum Gravity, vol. 28 035008 (21pp), 2011.
Abstract
We study the motion of a test particle in a stationary, axially and re-
flection symmetric spacetime of a central compact object, as affected by
interaction with a test radiation field of the same symmetries. Con-
sidering the radiation flux with fixed but arbitrary (non-zero) angu-
lar momentum, we extend previous results limited to an equatorial
motion within a zero-angular-momentum photon flux in the Kerr and
Schwarzschild backgrounds. While a unique equilibrium circular or-
bit exists if the photon flux has zero angular momentum, multiple such
orbits appear if the photon angular momentum is sufficiently high.

16. Bini D., Cherubini C., Filippi S.

2218
 4. Publications (2009 – 2013)

Effective geometry of a white dwarf

Physical Review D, vol. 83, 064039 (15pp), 2011.
Abstract
The “effective geometry” formalism is used to study the perturbations
of a white dwarf described as a self-gravitating fermion gas with a com-
pletely degenerate relativistic equation of state of barotropic type. The
quantum nature of the system causes an absence of homological prop-
erties manifested instead by batropic stars and requires a parametric
study of the solutions both at numerical and analytical level. We have
explicitly derived a compact analytical parametric approximate solu-
tion of Padé type which gives density curves and stellar radii in good
accordance with already existing numerical results. After validation of
this new type of approximate solutions, we use them to construct the
effective acoustic metric governing perturbations of any type follow-
ing Chebsch’s formalism. Even in this quantum and relativistic case
the stellar surface exhibits a curvature singularity due to the vanishing
of density, as already evidenced in past studies on non relativistic and
non quantum self-gravitating polytropic star. The equations of the the-
ory are finally numerically integrated, in the simpler case of irrotational
spherical pulsating perturbations including the effect of back-reaction,
in order to have a dynamical picture of the process occurring in the
acoustic metric.

17. Bini D. , de Felice F., Geralico A.

Accelerated orbits in black hole fields: the static case
Classical and Quantum Gravity, vol. 28 225012, 2011.
Abstract
We study non-geodesic orbits of test particles endowed with a structure,
assuming the Schwarzschild spacetime as background. We develop a
formalism which allows one to recognize the geometrical characteriza-
tion of those orbits in terms of their Frenet-Serret parameters and apply
it to explicit cases as those of spatially circular orbits which witness the
equilibrium under conflicting types of interactions. In our general anal-
ysis we solve the equations of motion offering a detailed picture of the
dynamics having in mind a check with a possible astronomical set up.
We focus on certain ambiguities which plague the interpretation of the
measurements preventing one from identifying the particular structure
carried by the particle.

18. Bini D., Geralico A., Jantzen R. T. and Semeřák O.

Poynting-Robertson effect in the Vaidya spacetime
Classical and Quantum Gravity, vol. 28, 245019, 2011.
Abstract
Motion of massive test particles in the nonvacuum spherically symmet-

2219
4. Publications (2009 – 2013)

ric radiating Vaidya spacetime is investigated, allowing for physical

interaction of the particles with the radiation field in terms of which
the source energy-momentum tensor is interpreted. This “Poynting-
Robertson-like effect” is modeled by the usual effective term describing
a Thomson-type radiation drag force. The equations of motion are stud-
ied for simple types of motion including free motion (without interac-
tion), purely radial and purely azimuthal (circular) motion, and for the
particular case of “static” equilibrium; appropriate solutions are given
where possible. The results—mainly those on the possible existence of
equilibrium positions—are compared with their counterparts obtained
previously for a test spherically symmetric radiation field in a vacuum
Schwarzschild background.

19. Bini D. and Geralico A.

Spin-geodesic deviations in the Kerr spacetime
Physical Review D, vol. 84, 104012, 2011.
Abstract
The dynamics of extended spinning bodies in the Kerr spacetime is in-
vestigated by assuming that the actual motion slightly deviates from the
geodesic path due to the spin-curvature force, in order to focus on how
the presence of the spin changes that geodesic motion. As usual, the
spin parameter is assumed to be very small in order to neglect the back
reaction on the spacetime geometry. This approach naturally leads to
solve the Mathisson-Papapetrou-Dixon equations linearized in the spin
variables as well as in the deviation vector, with the same initial condi-
tions as for geodesic motion. General deviations from generic geodesic
motion are studied, generalizing previous results limited to the very
special case of an equatorial circular geodesic as the reference path.

20. Bini D., Fortini P., Haney M., Ortolan A.

Electromagnetic waves in gravitational wave spacetimes
Classical and Quantum Gravity, vol. 28, 235007, 2011.
Abstract
We have considered the propagation of electromagnetic waves in a space-
time representing an exact gravitational plane wave and calculated the
induced changes on the four potential field Aµ of a plane electromag-
netic wave. By choosing a suitable photon round-trip in a Michelson
interferometer, we have been able to identify the physical effects of the
exact gravitational wave on the electromagnetic field, i.e. phase shift,
change of the polarization vector, angular deflection and delay. These
results have been exploited to study the response of an interferomet-
ric gravitational wave detector beyond the linear approximation of the
general theory of relativity.

21. Bini D., Gregoris D. and Succi S.

2220
 4. Publications (2009 – 2013)

Kinetic theory in a curved spacetime: applications to the Poynting-Robertson

effect
submitted, 2011.
Abstract
We discuss the statistical description of a massive and a photon gas in
an arbitrary spacetime. The motion of a massive test particle inside
a (test) photon gas is then studied near a Schwarzschild black hole,
leading to a novel description of radiation scattering in terms of the so
called Poynting-Robertson effect. Based on the new statistical descrip-
tion, and at variance with previous results, it is found that a particle
moving in the gravitational background of a Schwarzschild black-hole,
always ends up within the black-hole horizon.

22. Bini D. , Geralico A., Jantzen R. T.

Separable geodesic action slicing in stationary spacetimes
General Relativity and Gravitation, vol. 44, 603-621, 2012.
Abstract
A simple observation about the action for geodesics in a stationary space-
time with separable geodesic equations leads to a natural class of slic-
ings of that spacetime whose orthogonal geodesic trajectories represent
the world lines of freely falling fiducial observers. The time coordi-
nate function can then be taken to be the observer proper time, lead-
ing to a unit lapse function, although the time coordinate lines still fol-
low Killing trajectories to retain the explicitly stationary nature of the
coordinate grid. This explains some of the properties of the original
Painlevé-Gullstrand coordinates on the Schwarzschild spacetime and
their generalization to the Kerr-Newman family of spacetimes, repro-
ducible also locally for the Gödel spacetime. For the static spherically
symmetric case the slicing can be chosen to be intrinsically flat with
spherically symmetric geodesic observers, leaving all the gravitational
field information in the shift vector field.

23. Bini D. , Esposito G., Geralico A.

de Sitter spacetime: effects of metric perturbations on geodesic motion
General Relativity and Gravitation, vol. 44, 467-490 2012.
Abstract
Gravitational perturbations of the de Sitter spacetime are investigated
using the Regge–Wheeler formalism. The set of perturbation equations
is reduced to a single second order differential equation of the Heun-
type for both electric and magnetic multipoles. The solution so ob-
tained is used to study the deviation from an initially radial geodesic
due to the perturbation. The spectral properties of the perturbed metric
are also analyzed. Finally, gauge- and tetrad-invariant first-order mass-
less perturbations of any spin are explored following the approach of

2221
4. Publications (2009 – 2013)

Teukolsky. The existence of closed-form, i.e. Liouvillian, solutions to

the radial part of the Teukolsky master equation is discussed.

24. Bini D. and Geralico A.

Scattering by an electromagnetic radiation field
Physical Review D, vol. 85, 044001, 2012.
Abstract
Motion of test particles in the gravitational field associated with an elec-
tromagnetic plane wave is investigated. The interaction with the radi-
ation field is modeled by a force term à la Poynting-Robertson entering
the equations of motion given by the 4-momentum density of radia-
tion observed in the particle’s rest frame with a multiplicative constant
factor expressing the strength of the interaction itself. Explicit analyt-
ical solutions are obtained. Scattering of fields by the electromagnetic
wave, i.e., scalar (spin 0), massless spin 12 and electromagnetic (spin 1)
fields, is studied too.

25. Bini D., Falanga M., Geralico A. and Stella L.

The signal from an emitting source moving in a Schwarzschild spacetime un-
der the influence of a radiation field
Classical and Quantum Gravity, vol. 29, 065014, 2012.
Abstract
The motion of matter immersed in a radiation field is affected by a radi-
ation drag, as a result of scattering or absorption and re-emission. The
resulting friction-like drag, also known as the PoyntingRobertson effect,
has been recently studied in the general relativistic background of the
Schwarzschild and Kerr metric, under the assumption that all photons
in the radiation field possess the same angular momentum. We calcu-
late here the signal produced by an emitting point-like specific source
moving in a Schwarzschild spacetime under the influence of such a ra-
diation field. We derive the flux, redshift factor and solid angle of the
hot spot as a function of (coordinate) time, as well as the time-integrated
image of the hot spot as seen by an observer at infinity. The results are
then compared with those for a spot moving on a circular geodesic in a
Schwarzschild metric.

26. Bini D., Gregoris D. and Succi S.

Radiation pressure vs friction effects in the description of the Poynting-Robertson
scattering process
Europhysics Letters, vol. 97, 40007, 2012.
Abstract
The motion of a massive test particle inside a thermal (test) photon gas
is studied near a Schwarzschild black hole, leading to a novel descrip-
tion of the effect of radiation scattering on the particle trajectory, respon-
sible for half of the Poynting-Robertson effect: the azimuthal radiation

2222
 4. Publications (2009 – 2013)

drag. Lacking the outward directed radiation pressure of the latter ef-
fect, gravitationally bound orbits always decay, leading to capture by
the black hole or the central object generating the exterior Schwarzschild
field in which this discussion takes place.

27. Bini D., Geralico A. and Succi S.

Particle scattering by a test fluid on a Schwarzschild spacetime: the equation
of state matters
The European Physical Journal C, vol. 72, Issue 3, 1913, 2012.
Abstract
The motion of a massive test particle in a Schwarzschild spacetime sur-
rounded by a perfect fluid with equation of state p0 = wρ0 is investi-
gated. Deviations from geodesic motion are analyzed as a function of
the parameter w, ranging from w = 1 which corresponds to the case of
massive free scalar fields, down into the so-called “phantom” energy,
with w < −1. It is found that the interaction with the fluid leads to
capture (escape) of the particle trajectory in the case 1 + w > 0 (< 0),
respectively. Based on this result, it is argued that inspection of the tra-
jectories of test particles in the vicinity of a Schwarzschild black hole
with matter around may offer a new means of gaining insights into the
nature of cosmic matter.

28. Bini D. and Geralico A.

Observer-dependent tidal indicators in the Kerr spacetime
Class. Quantum Grav., vol. 29, 055005, 2012.
Abstract
The observer-dependent tidal effects associated with the electric and
magnetic parts of the Riemann tensor with respect to an arbitrary fam-
ily of observers are discussed in a general spacetime in terms of certain
tidal indicators. The features of such indicators are then explored by
specializing our considerations to the family of stationary circularly ro-
tating observers in the equatorial plane of the Kerr spacetime. There ex-
ist a number of observer families which are special for several reasons
and for each of them such indicators are evaluated. The transforma-
tion laws of tidal indicators when passing from one observer to another
are also discussed, clarifying the interplay among them. Our analysis
shows that no equatorial plane circularly rotating observer in the Kerr
spacetime can ever measure a vanishing tidal electric indicator, whereas
the family of Carters observers measures zero tidal magnetic indicator.

29. Bini D., Damour T. and Faye G.

Effective action approach to higher-order relativistic tidal interactions in bi-
nary systems and their effective one body description
Physical Review D, 85, 124034, 2012.
Abstract

2223
4. Publications (2009 – 2013)

The gravitational-wave signal from inspiralling neutron-starneutron-

star (or black-holeneutron-star) binaries will be influenced by tidal cou-
pling in the system. An important science goal in the gravitational-
wave detection of these systems is to obtain information about the equa-
tion of state of neutron star matter via the measurement of the tidal
polarizability parameters of neutron stars. To extract this piece of in-
formation will require accurate analytical descriptions both of the mo-
tion and the radiation of tidally interacting binaries. We improve the
analytical description of the late inspiral dynamics by computing the
next-to-next-to-leadingorder relativistic correction to the tidal interac-
tion energy. Our calculation is based on an effective-action approach to
tidal interactions and on its transcription within the effective-one-body
formalism. We find that second-order relativistic effects (quadratic in
the relativistic gravitational potential u = G (m1 + m2 )/(c2 r )) signif-
icantly increase the effective tidal polarizability of neutron stars by a
distance-dependent amplification factor of the form 1 + a1 u + a2 u2 +
where, say, for an equal-mass binary, a1 = 5/4 = 1.25 (as previously
known) and a2 = 85/14 =≈ 6.07143 (as determined here for the first
time). We argue that higher-order relativistic effects will lead to further
amplification, and we suggest a Pad-type way of resumming them. We
recommend testing our results by comparing resolution-extrapolated
numerical simulations of inspiralling binary neutron stars to their ef-
fective one-body description.

30. Bini D., Gregoris D., Rosquist K. and Succi S.

Particle motion in a photon gas: friction matters
Gen. Rel. Grav., vol. 44, 2669-2680, 2012.
Abstract
The motion of a particle in the Tolman metric generated by a photon gas
source is discussed. Both the case of geodesic motion and motion with
nonzero friction, due to photon scattering effects, are analyzed. In the
Minkowski limit, the particle moves along a straight line segment with
a decelerated motion, reaching the endpoint at zero speed. The curved
case shows a qualitatively different behavior; the geodesic motion con-
sists of periodic orbits, confined within a specific radial interval. Under
the effect of frictional drag, this radial interval closes up in time and in
all our numerical simulations the particle ends up in the singularity at
the center.

31. Bini D., Chicone C. and Mashhoon B.

Spacetime Splitting, Admissible Coordinates and Causality
Physical Review D, vol. 85, 104020, 2012.
arXiv:1203.3454
Abstract

2224
 4. Publications (2009 – 2013)

To confront relativity theory with observation, it is necessary to split

spacetime into its temporal and spatial components. The (1+3) timelike
threading approach involves restrictions on the gravitational potentials
(gµν ), while the (3+1) spacelike slicing approach involves restrictions
on (gµν ). These latter coordinate conditions protect chronology within
any such coordinate patch. While the threading coordinate conditions
can be naturally integrated into the structure of Lorentzian geometry
and constitute the standard coordinate conditions in general relativity,
this circumstance does not extend to the slicing coordinate conditions.
We explore the influence of chronology violation on wave motion. In
particular, we consider the propagation of radiation parallel to the rota-
tion axis of stationary Gödel-type universes characterized by parame-
ters η > 0 and λ > 0 such that for η < 1 (η > 1) chronology is protected
(violated). We show that in the WKB approximation such waves can
freely propagate only when chronology is protected.

32. Bini D., Boshkayev K.,Ruffini R. and Siutsou I.

Equatorial Circular Geodesics in the Hartle-Thorne Spacetime
Proceedings of the 12th Italian-Korean meeting July 4-8, 2011. Pescara
(Italy). To appear on il Nuovo Cimento, 2012
Abstract
We investigate the influence of the quadrupole moment of a rotating
source on the motion of a test particle in the strong field regime. For
this purpose the Hartle-Thorne metric, that is an approximate solution
of vacuum Einstein field equations that describes the exterior of any
slowly rotating, stationary and axially symmetric body, is used. The
metric is given with accuracy up to the second order terms in the body’s
angular momentum, and first order terms in its quadrupole moment.
We give, with the same accuracy, analytic equations for equatorial cir-
cular geodesics in the Hartle-Thorne spacetime and integrate them nu-
merically.

33. Bini D., Kuantay B. and Geralico A.

Tidal indicators in the spacetime of a rotating deformed mass
Class. Quantum Grav., vol. 29, 145003, 2012.
Abstract
Tidal indicators are commonly associated with the electric and magnetic
parts of the Riemann tensor (and its covariant derivatives) with respect
to a given family of observers in a given spacetime. Recently, observer-
dependent tidal effects have been extensively investigated with respect
to a variety of special observers in the equatorial plane of the Kerr
spacetime. This analysis is extended here by considering a more gen-
eral background solution to include the case of matter which is also
endowed with an arbitrary mass quadrupole moment. Relations with

2225
4. Publications (2009 – 2013)

curvature invariants and BelRobinson tensor, i.e. observer-dependent

super-energy density and super-Poynting vector, are also investigated.

34. Bini D., Geralico A., Haney M. and Jantzen R.T.

Scattering of particles by radiation fields: a comparative analysis
Phys. Rev. D., vol 86, 064016, 2012.
Abstract
The features of the scattering of massive neutral particles propagating
in the field of a gravitational plane wave are compared with those char-
acterizing their interaction with an electromagnetic radiation field. The
motion is geodesic in the former case, whereas in the case of an elec-
tromagnetic pulse it is accelerated by the radiation field filling the as-
sociated spacetime region. The interaction with the radiation field is
modeled by a force term entering the equations of motion proportional
to the 4-momentum density of radiation observed in the particles rest
frame. The corresponding classical scattering cross sections are evalu-
ated too.

35. Bini D., Geralico A.

Equilibrium Orbits of Particles Undergoing Poynting-Robertson Effect in Schwarzschild
Spacetime
Proceedings of the 2nd GalileoXuGuangqi Meeting
International Journal of Modern Physics: Conference Series, vol. 12, is-
sue 01, p. 247, 2012.
Abstract
Equilibrium orbits of particles moving on the equatorial plane of a Schwarzschild
spacetime are investigated when a test radiation field is superposed to
the background gravitational field. The radiation flux is endowed with
a fixed but arbitrary (non-zero) angular momentum. It is found that
multiple equilibrium circular orbit exist provided that the photon an-
gular momentum is sufficiently high. The stability of such orbits is also
analyzed.

36. Bini D. , Damour T.

Gravitational radiation reaction along general orbits in the effective one-body
formalism
Phys. Rev. D, vol. 86, 124012, 2012.
Abstract
We derive the gravitational radiation-reaction force modifying the Ef-
fective One Body (EOB) description of the conservative dynamics of
binary systems. Our result is applicable to general orbits (elliptic or hy-
perbolic) and keeps terms of fractional second post-Newtonian order
(but does not include tail effects). Our derivation of radiation-reaction
is based on a new way of requiring energy and angular momentum
balance. We give several applications of our results, notably the value

2226
 4. Publications (2009 – 2013)

of the (minimal) “Schott” contribution to the energy, the radial compo-

nent of the radiation-reaction force, and the radiative contribution to
the angle of scattering during hyperbolic encounters. We present also
new results about the conservative relativistic dynamics of hyperbolic
motions.

37. Bini D., Geralico A.

Slicing black hole spacetimes
Proceedings of the 13th Marcel Grossmann Meeting, July 2-8, 2012, Stock-
holm (Sweden).
Abstract
Some properties of both the intrinsic and extrinsic geometry of special
slicings in black hole spacetimes are discussed. A special role is played
by slicings associated with the so called Painlevé-Gullstrand observers.

38. Bini D., Cherubini C. Filippi S.

On the analog gravity formalism applied to white dwarfs
Proceedings of the 13th Marcel Grossmann Meeting, July 2-8, 2012, Stock-
holm (Sweden).
abstract
The “effective geometry”formalism is here adopted to analyze the per-
turbations of a non rotating white dwarf. After the derivation of a
compact analytical parametric approximate white dwarf solution via
the Padé formalism, we use it to construct the effective acoustic met-
ric governing general fluid perturbations. The equations of the theory,
numerically integrated in the case of irrotational, spherical, pulsating
problem, show the analog space-time structure of the acoustic metric.
In particular it appears that the stellar surface exhibits a curvature sin-
gularity associated to the vanishing of density.

39. Bini D.
Observers, observables and measurements in general relativity
Proceedings of the meeting “Relativity and Gravitation 100 Years after
Einstein in Prague” June 2529, 2012, Prague (Czech Republic).
Abstract
To perform any physical measurement it is necessary to identify in a
non ambiguous way both the observer and the observable. A given
observable can be then the target of different observers: a suitable al-
gorithm to compare among their measurements should necessarily be
developed, either formally or operationally. This is the task of what we
call “theory of measurement,” which we discuss here in the framework
of general relativity.

40. Bini D., Geralico A.

2227
4. Publications (2009 – 2013)

On the occurrence of Closed Timelike Curves and the observer’s point of view
Proceedings of the meeting “The Time-Machine Factory”, October 14-
19, 2012, Torino (Italy) [Invited speaker].
Abstract
The existence of Closed Timelike Curves (CTCs) in a generic spacetime
is often associated with a non-physical choice of coordinates and can
be cured by limiting the admissibility of such coordinates. Lichnerow-
icz conditions, for instance, represent a criterion for admissibility. The
result, however, is a very restrictive limitation which may imply “re-
moval” of important regions (with respect to the peculiarity of phenom-
ena which may happen there) of the spacetime manifold. We consider
here the point of view of a family of observers (Fundamental Slicing
Observers, FSO) having their world lines orthogonal to the surfaces of
constant coordinate time. We say that the time coordinate has not a
global character if the associated FSO change their causality condition
in the domain of validity of the coordinates themselves. Furthermore,
in those regions where FSO have no more timelike world lines CTCs
are present and one may think of special devices or investigation tools
apt to operationally discover them. We will analyze in detail special
features involving (scalar) waves or photons.

41. Donato Bini, Mariateresa Crosta, Fernando de Felice, Andrea Geralico,

Alberto Vecchiato
The Erez-Rosen metric and the role of the quadrupole on light propagation
Class. Quantum Grav., vol. 30 045009(22pp), 2013
Abstract
The gravitational field of a static body with quadrupole moment is de-
scribed by an exact solution found by Erez and Rosen. Here we inves-
tigate the role of the quadrupole in the motion, deflection and lensing
of a light ray in the above metric. The standard lensing observables like
image positions and magnification have been explicitly obtained in the
weak field and small quadrupole limit. In this limit the spacetime met-
ric appears as the natural generalization to quadrupole corrections of
the metric form adopted also in current astrometric models. Hence, the
corresponding analytical solution of the inverse ray tracing problem as
well as the consistency with other approaches are also discussed.

42. Bini D., Gregoris D., Rosquist K. and Succi S.

Effects of friction forces on the motion of objects in smoothly matched inte-
rior/exterior spacetimes
Class. Quantum Grav., vol. 30, 025009, 2013.
Abstract
We investigate the non-geodesic motion of a test particle inside a gas
in equilibrium, as described by the interior Schwarzschild solution, in

2228
 4. Publications (2009 – 2013)

the presence of a Poynting-Robertson-like dissipative force. The ac-

celerated motion in the interior spacetime is (regularly) connected to
the geodesic one in the corresponding external Schwarzschild back-
ground, mimicking physical conditions often occurring in spacetime re-
gions close to stellar or compact objects. A similar analysis is repeated
for the case in which the interior Schwarzschild solution is given by a
Pant-Sah perfect-fluid solution. This latter case, supporting a polytropic
equation of state for the interior fluid source, appears to be more appro-
priate for modeling situations of either stellar or cosmological interest.

43. Bini D. , Geralico A.

On the occurrence of Closed Timelike Curves and the observer’s point of view
Proceedings of the meeting “The Time-Machine Factory”, October 14-
19, 2012, Torino (Italy) [Invited speaker].
EPJ Web of Conferences 58, 01002, (2013)
DOI: 10.1051/epjconf/20135801002
Published by EDP Sciences, 2013.
Abstract

The existence of Closed Timelike Curves (CTCs) in a generic spacetime

is often associated with a non-physical choice of coordinates and can
be cured by limiting the admissibility of such coordinates. Lichnerow-
icz conditions, for instance, represent a criterion for admissibility. The
result, however, is a very restrictive limitation which may imply re-
moval” of important regions (with respect to the peculiarity of phenom-
ena which may happen there) of the spacetime manifold. We consider
here the point of view of a family of observers (Fundamental Slicing
Observers, FSO) having their world lines orthogonal to the surfaces of
constant coordinate time. We say that the time coordinate has not a
global character if the associated FSO change their causality condition
in the domain of validity of the coordinates themselves. Furthermore,
in those regions where FSO have no more timelike world lines, CTCs
are present and one may think of special devices or investigation tools
apt to operationally detect them. We will discuss in detail theoretical
approaches involving (scalar) waves or photons.

44. Bini D. , Geralico A.

Dynamics of quadrupolar bodies in a Schwarzschild spacetime
Phys. Rev. D, vol. 87, 024028, 2013.
Abstract
The dynamics of extended bodies endowed with multipolar structure
up to the mass quadrupole moment is investigated in the Schwarzschild
background according to Dixons model, extending previous works. The
whole set of evolution equations is numerically integrated under the

2229
4. Publications (2009 – 2013)

simplifying assumptions of constant frame components of the quadrupole

tensor and that the motion of the center of mass be confined on the
equatorial plane, the spin vector being orthogonal to it. The equations
of motion are also solved analytically in the limit of small values of
the characteristic length scales associated with the spin and quadrupole
with respect to the background curvature characteristic length. The re-
sults are qualitatively and quantitatively different from previous anal-
yses involving only spin structures. In particular, the presence of the
quadrupole turns out to be responsible for the onset of a nonzero spin
angular momentum, even if initially absent.

45. Bini D. , Geralico A., Gregoris D., Succi S.

Friction forces in cosmological models
EPJC, vol. 73, 2334, 2013.
Abstract
We investigate the dynamics of test particles undergoing friction forces
in a Friedmann-Robertson-Walker (FRW) spacetime. The interaction
with the background fluid is modeled by introducing a Poynting-Robertson-
like friction force in the equations of motion, leading to measurable (at
least in principle) deviations of the particle trajectories from geodesic
motion. The effect on the peculiar velocities of the particles is investi-
gated for various equations of state of the background fluid and dif-
ferent standard cosmological models. The friction force is found to
have major effects on particle motion in closed FRW universes, where it
turns the time-asymptotic value (approaching the recollapse) of the pe-
culiar particle velocity from ultra-relativistic (close to light speed) to a
co-moving one, i.e., zero peculiar speed. On the other hand, for open or
flat universes the effect of the friction is not so significant, because the
time-asymptotic peculiar particle speed is largely non-relativistic also
in the geodesic case.

46. Bini D., Fortini P., Geralico A., Haney M. and Ortolan A.
Light scattering by radiation fields: the optical medium analogy
EPL, vol 102, 20006 (2013)
Abstract
The optical medium analogy of a radiation field generated by either an
exact gravitational plane wave or an exact electromagnetic wave in the
framework of general relativity is developed. The equivalent medium
of the associated background field is inhomogeneous and anisotropic
in the former case, whereas it is inhomogeneous but isotropic in the lat-
ter. The features of light scattering are investigated by assuming the in-
teraction region to be sandwiched between two flat spacetime regions,
where light rays propagate along straight lines. Standard tools of ordi-
nary wave optics are used to study the deflection of photon paths due

2230
 4. Publications (2009 – 2013)

to the interaction with the radiation fields, allowing for a comparison

between the optical properties of the equivalent media associated with
the different background fields.

47. Bini D. , Esposito G., Kiefer C., Krämer M. and Pessina F.,
On the Modification of the Cosmic Microwave Background Anisotropy Spec-
trum from Canonical Quantum Gravity
Phys. Rev. D, vol. 87, 104008 (2013)
Abstract
We evaluate the modifications to the CMB anisotropy spectrum that re-
sult from a semiclassical expansion of the Wheeler–DeWitt equation.
Recently, such an investigation in the case of a real scalar field coupled
to gravity, has led to the prediction that the power at large scales is sup-
pressed. We make here a more general analysis and show that there
is an ambiguity in the choice of solution to the equations describing
the quantum gravitational effects. Whereas one of the two solutions
describes a suppression of power, the other one describes an enhance-
ment. We investigate possible criteria for an appropriate choice of so-
lution. The absolute value of the correction term is in both cases of the
same order and currently not observable. We also obtain detailed for-
mulae for arbitrary values of a complex parameter occurring in the gen-
eral solution of the nonlinear equations of the model. We finally discuss
the modification of the spectral index connected with the power spec-
trum and comment on the possibility of a quantum-gravity induced
unitarity violation.

48. Bini D. , de Felice F., Geralico A.

On the spacetime acting as an optical medium: the observer-dependent ap-
proach
IJGMMP., (2013) to appear
Abstract
A local observer-dependent approach to the optical medium analogy of
a general gravitational background is developed. The optical proper-
ties of the equivalent media associated with several stationary axisym-
metric spacetimes are investigated with respect to different families of
observers playing a special role as well as selected null orbits. Explicit
examples among the class of both exact and approximate solutions of
Einstein’s field equations describing the gravitational field of sources
with a multipolar structure up to the quadrupole are analyzed. The
analytic contribution of the individual polarities to the refraction index
associated with the corresponding effective media is deduced as a func-
tion of the spacetime parameters.

49. Bini D., Damour T.

Analytic determination of the two-body gravitational interaction potential at

2231
4. Publications (2009 – 2013)

the 4th post-Newtonian approximation

Phys. Rev. D. Rapid Communications, vol. 87, 121501, (2013)
Abstract
We complete the analytical determination, at the 4th post-Newtonian
approximation, of the main radial potential describing (within the ef-
fective one-body formalism) the gravitational interaction of two bod-
ies. The (non logarithmic) coefficient a5 (ν) measuring this 4th post-
Newtonian interaction potential is found to be linear in the symmet-
ric mass ratio ν. Its ν-independent part a5 (0) is obtained by an ana-
lytical gravitational self-force calculation that unambiguously resolves
the formal infrared divergencies which currently impede its direct post-
Newtonian calculation. Its ν-linear part a5 (ν) − a5 (0) is deduced from
recent results of Jaranowski and Schaefer, and is found to be signifi-
cantly negative.

50. Bini D. , Geralico A., Gregoris D, Succi S.

Dark energy from cosmological fluids with asymptotic-free non-ideal equation
of state
Phys. Rev. D., vol. 88, 063007, (2013)
Abstract
We consider a Friedmann-Robertson-Walker universe with a fluid source
obeying a non-ideal equation of state with “asymptotic freedom,” namely
ideal gas behavior (pressure changes directly proportional to density
changes) both at low and high density regimes, following a fluid dy-
namical model due to Shan and Chen. It is shown that, starting from
an ordinary energy density component, such fluids naturally evolve
towards a universe with a substantial dark energy component at the
present time, with no need of invoking any cosmological constant. More-
over, we introduce a quantitative indicator of darkness abundance, which
provides a consistent picture of the actual matter-energy content of the
universe.

51. Bini D., Geralico A., Haney M.

Refraction index analysis of light propagation in a colliding gravitational wave
spacetime
Gen. Rel. Grav., (2013), submitted
Abstract
The optical medium analogy of a given spacetime was developed decades
ago and has since then been widely applied to different gravitational
contexts. Here we consider applications to light propagation in a col-
liding gravitational wave spacetime, generalizing previous results con-
cerning single gravitational pulses. In view of complexity of the non-
linear interaction of two gravitational waves in the framework of gen-
eral relativity, typically leading to the formation of either horizons or

2232
 4. Publications (2009 – 2013)

singularities, the optical medium analogy proves helpful. It allows us

to simply capture some interesting effects we discuss here in terms of
wave propagation, interferometric fringe motions, time delays, etc.

Books and book chapters

1. (Book) G. Ferrarese, Bini D.

Introduction to relativistic continuum mechanics,
Lecture Notes in Physics 727, Ed. Springer, 2007.

2. (Book) De Felice F., Bini D.

Classical Measurements in Curved Space-Times
Series: Cambridge Monographs on Mathematical Physics, Cambridge,
UK, 2010
Brief description

The theory of relativity describes the laws of physics in a given space-

time. However, a physical theory must provide observational predic-
tions expressed in terms of measurements, which are the outcome of
practical experiments and observations. Ideal for readers with a math-
ematical background and a basic knowledge of relativity, this book will
help readers understand the physics behind the mathematical formal-
ism of the theory of relativity. It explores the informative power of
the theory of relativity, and highlights its uses in space physics, astro-
physics and cosmology. Readers are given the tools to pick out from
the mathematical formalism those quantities that have physical mean-
ing and which can therefore be the result of a measurement. The book
considers the complications that arise through the interpretation of a
measurement, which is dependent on the observer who performs it.
Specific examples of this are given to highlight the awkwardness of the
problem.
Provides a large sample of observers and reference frames in space-
times that can be applied to space physics, astrophysics and cosmology.
Tackles the problems encountered in interpreting measurements, giv-
ing specific examples. Features advice to help readers understand the
logic of a given theory and its limitations.
Contents
1. Introduction; 2. The theory of relativity: a mathematical overview; 3.
Space-time splitting; 4. Special frames; 5. The world function; 6. Local
measurements; 7. Non-local measurements; 8. Observers in physical
relevant space-times; 9. Measurements in physically relevant space-
times; 10. Measurements of spinning bodies.

2233
4. Publications (2009 – 2013)

APPENDICES

2234
A. Spacetime splitting techniques
in general relativity
The concept of a “gravitational force” modeled after the electromagnetic Lo-
rentz force was born in the Newtonian context of centrifugal and Coriolis
“fictitious” forces introduced by a rigidly rotating coordinate system in a flat
Euclidean space. Bringing this idea first into linearized general relativity and
then into its fully nonlinear form, it has found a number of closely related but
distinct generalizations. Regardless of the details, this analogy between grav-
itation and electromagnetism has proven useful in interpreting the results of
spacetime geometry in terms we can relate to, and has been illustrated in
many research articles and textbooks over the past half century.
ICRANet has itself devoted a workshop and its proceedings to aspects of
this topic in 2003 [2]. In the lengthy introduction to these proceedings, R.
Ruffini has discussed a number of related topics, like “the gravitational ana-
logue of the Coulomb-like interactions, of Hertz-like wave solutions, of the
Oersted-Ampére-like magnetic interaction, etc.,” supporting the thesis that
treating gravitation in analogy with electromagnetism may help to better un-
derstand the main features of certain gravitational phenomena, at least when
the gravitational field may be considered appropriately described by its lin-
earized approximation [12, 13, 14, 15, 16, 17, 18]. A particularly long bib-
liography surveying most of the relevant literature through 2001 had been
published earlier in the Proceedings of one of the annual Spanish Relativity
Meetings [19].
In the 1990s, working in fully nonlinear general relativity, all of the various
notions of “noninertial forces” (centrifugal and Coriolis forces) were put into
a single framework by means of a unifying formalism dubbed “gravitoelec-
tromagnetism” [1, 3, 4] which is a convenient framework to deal with these
and curvature forces and related questions of their effect on test bodies mov-
ing in the gravitational field. More precisely, such a language is based on the
splitting of spacetime into “space plus time,” accomplished locally by means
of an observer congruence, namely a congruence of timelike worldlines with
(future-pointing) unit tangent vector field u which may be interpreted as the
4-velocity field of a family of test observers filling some region of spacetime.
The orthogonal decomposition of each tangent space into a local time direc-
tion along u and the orthogonal local rest space (LRS) is used to decompose
all spacetime tensors and tensor equations into a “space plus time” represen-
tation; the latter representation is somehow equivalent to a geometrical “mea-

2235
A. Spacetime splitting techniques in general relativity

surement” process. This leads to a family of “spatial” spacetime tensor fields

which represent each spacetime field and a family of spatial equations which
represent each spacetime equation. Dealing with spacetime splitting tech-
niques as well as 3-dimensional-like quantities clearly permits a better inter-
face of our intuition and experience with the 4-dimensional geometry in cer-
tain gravitational problems. It can be particularly useful in spacetimes which
have a geometrically defined timelike congruence, either explicitly given or
defined implicitly as the congruence of orthogonal trajectories to a slicing or
foliation of spacetime by a family of privileged spacelike hypersurfaces.
For example, splitting techniques are useful in the following spacetimes:
1. Stationary spacetimes, having a preferred congruence of Killing trajec-
tories associated with the stationary symmetry, which is timelike on a
certain region of spacetime (usually an open region, the boundary of
which corresponding to the case in which the Killing vector becomes
null so that in the exterior region Killing trajectories are spacelike).
2. Stationary axially symmetric spacetimes having in addition a preferred
slicing whose orthogonal trajectories coincide with the worldlines of
locally nonrotating test observers.
3. Cosmological spacetimes with a spatial homogeneity subgroup, which
have a preferred spacelike slicing by the orbits of this subgroup.
From the various schools of relativity that blossomed during the second
half of the last century a number of different approaches to spacetime split-
ting were developed without reference to each other. During the 1950s efforts
were initiated to better understand general relativity and the mathematical
tools needed to flush out its consequences. Lifshitz and the Russian school,
Lichnerowicz in France, the British school, scattered Europeans (Ehlers and
Trautman, for example) and the Americans best represented by Wheeler ini-
tiated this wave of relativity which blossomed in the 1960s. The textbook of
Landau and Lifshitz and articles of Zelmanov [20, 21, 22, 23] presented the
“threading point of view” of the Russian school and of Moller [21] which in-
fluenced Cattaneo in Rome and his successor Ferrarese [24, 25, 26, 27, 28],
while a variation of this approach not relying on a complementary family
of hypersurfaces (the “congruence point of view) began from work initially
codified by Ehlers [17] and then taken up by Ellis [29, 30] in analyzing cos-
mological issues.
However, issues of quantum gravity lead to the higher profile of the “slic-
ing point of view” in the 1960s initiated earlier by Lichnerowicz and devel-
oped by Arnowitt, Deser and Misner and later promoted by the influential
textbook “Gravitation” by Misner, Thorne and Wheeler [31, 32, 33, 34] repre-
sents a splitting technique which is complementary to the threading point of
view and its congruence variation, and proved quite useful in illuminating
properties of black hole spacetimes.

2236
 A. Spacetime splitting techniques in general relativity

R. Ruffini, a former student of Cattaneo and a collaborator of Wheeler,

in his quest to better understand electromagnetic properties of black holes,
awakened the curiosity of Jantzen and Carini at the end of the 1980s, later
joined by Bini, who together made an effort to clarify the interrelationships
between these various approaches as well as shed some light on the then
confusing work of Abramowicz and others on relativistic centrifugal and
Coriolis forces. By putting them all in a common framework, and describ-
ing what each measured in geometrical terms, and how each related to the
others, some order was brought to the field [1, 3, 4].
The ICRANet people working on this subject have applied the main ideas
underlying spacetime splitting techniques to concrete problems arising when
studying test particle motion in black hole spacetimes. Among the various re-
sults obtained it is worth mentioning the relativistic and geometrically correct
definition of inertial forces in general relativity [35, 36, 37, 38, 39], the defi-
nition of special world line congruences, relevant for the description of the
motion of test particles along circular orbits in the Kerr spacetime (geodesic
meeting point observers, extremely accelerated observers, etc.), the specifi-
cation of all the geometrical properties concerning observer-adapted frames
to the above mentioned special world line congruences [40, 41], the charac-
terization of certain relevant tensors in black hole spacetimes (Simon tensor,
Killing-Yano tensor) in terms of gravitoelectromagnetism [42, 43], etc. This
research line is still ongoing and productive.
Over a period of several decades Jantzen, Bini and a number of students at
the University of Rome “La Sapienza” under the umbrella of the Rome ICRA
group have been working on this problem under the supervision of Ruffini.
The collaborators involved have been already listed and the most relevant pa-
pers produced are indicated in the references below [44]–[83]. In the present
year 2010 a book by F. de Felice and D. Bini, including a detailed discussion
of this and related topics, has been published by Cambridge University Press
[157].
Let us now describe some fundamental notions of gravitoelectromagnetism.

A.1. Observer-orthogonal splitting

Let (4)g (signature -+++ and components (4)gαβ , α, β, . . . = 0, 1, 2, 3) be the
spacetime metric, (4)∇ its associated covariant derivative operator, and (4)η
the unit volume 4-form which orients spacetime ((4)η0123 = (4)g1/2 in an ori-
ented frame, where (4)g ≡ | det((4)gαβ )|). Assume the spacetime is also time
oriented and let u be a future-pointing unit timelike vector field (uα uα = −1)
representing the 4-velocity field of a family of test observers filling the space-
time (or some open submanifold of it).
If S is an arbitrary tensor field, let S[ and S] denote its totally covariant

2237
A. Spacetime splitting techniques in general relativity

and totally contravariant forms with respect to the metric index-shifting op-
erations. It is also convenient to introduce the right contraction notation
[S X ]α = Sα β X β for the contraction of a vector field and the covariant in-
dex of a (11)-tensor field (left contraction notation being analogous).

A.1.1. The measurement process

The observer-orthogonal decomposition of the tangent space, and in turn of
the algebra of spacetime tensor fields, is accomplished by the temporal pro-
jection operator T (u) along u and the spatial projection operator P(u) onto
LRSu , which may be identified with mixed second rank tensors acting by
contraction
δα β = T (u)α β + P(u)α β ,
T (u)α β = −uα u β , (A.1.1)
α α α
P(u) β =δ β + u uβ .
These satisfy the usual orthogonal projection relations P(u)2 = P(u), T (u)2 =
T (u), and T (u) P(u) = P(u) T (u) = 0. Let

[ P(u)S]α...β... = P(u)α γ · · · P(u)δ β · · · Sγ...

δ... (A.1.2)

denote the spatial projection of a tensor S on all indices.

The measurement of S by the observer congruence is the family of spatial
tensor fields which result from the spatial projection of all possible contrac-
tions of S by any number of factors of u. For example, if S is a (11)-tensor, then
its measurement

Sα β ↔(uδ uγ Sγ δ , P(u)α γ uδ Sγ δ , P(u)δ α uγ Sγ δ , P(u)α γ P(u)δ β Sγ δ )

| {z } | {z } | {z } | {z } (A.1.3)
scalar vector vector tensor

results in a scalar field, a spatial vector field, a spatial 1-form and a spatial (11)-
tensor field. It is exactly this family of fields which occur in the orthogonal
“decomposition of S” with respect to the observer congruence

Sα β = [ T (u)α γ + P(u)α γ ][ T (u)δ β + P(u)δ β ]Sγ δ

(A.1.4)
= [uδ uγ Sγ δ ]uα u β + · · · + [ P(u)S]α β .

A.2. Examples
1. Measurement of the spacetime metric and volume 4-form

• spatial metric [ P(u)(4)g]αβ = P(u)αβ

2238
 A. Spacetime splitting techniques in general relativity

• spatial unit volume 3-form η (u)αβγ = uδ (4)ηδαβγ ;

In a compact notation: η (u) = [ P(u) u (4)η ]
2. Measurement of the Lie, exterior and covariant derivative
• spatial Lie derivative £(u) X = P(u)£X
• the spatial exterior derivative d(u) = P(u)d
• the spatial covariant derivative ∇(u) = P(u)(4)∇
• the spatial Fermi-Walker derivative (or Fermi-Walker temporal deriva-
tive) ∇(fw) (u) = P(u)(4)∇u (when acting on spatial fields)
• the Lie temporal derivative ∇(lie) (u) = P(u)£u = £(u)u
Note that spatial differential operators do not obey the usual product
rules for nonspatial fields since undifferentiated factors of u are killed
by the spatial projection.
3. Notation for 3-dimensional operations
It is convenient to introduce 3-dimensional vector notation for the spa-
tial inner product and spatial cross product of two spatial vector fields
X and Y. The inner product is just

X ·u Y = P(u)αβ X α Y β (A.2.1)

while the cross product is

[ X ×u Y ]α = η (u)α βγ X β Y γ . (A.2.2)

With the “vector derivative operator” ∇(u)α one can introduce spatial
gradient, curl and divergence operators for functions f and spatial vec-
tor fields X by

gradu f = ∇(u) f = [d(u) f ]] ,

curlu X = ∇(u) ×u X = [∗(u)d(u) X [ ]] , (A.2.3)
divu X = ∇(u) ·u X = ∗(u)[d(u)∗(u) X [ ] ,

where ∗(u) is the spatial duality operation for antisymmetric tensor fields
associated with the spatial volume form η (u) in the usual way. These
definitions enable one to mimic all the usual formulas of 3-dimensional
vector analysis. For example, the spatial exterior derivative formula for
the curl has the index form

[curlu X ]α = η (u)αβγ (4)∇ β Xγ (A.2.4)

which also defines a useful operator for nonspatial vector fields X.

2239
A. Spacetime splitting techniques in general relativity

4. Measurement of the covariant derivative of the observer four velocity

Measurement of the covariant derivative [(4)∇u]α β = uα ;β leads to two

spatial fields, the acceleration vector field a(u) and the kinematical mixed
tensor field k(u)
uα ;β = − a(u)α u β − k (u)α β ,
a(u) = ∇(fw) (u)u , (A.2.5)
k (u) = −∇(u)u .
The kinematical tensor field may be decomposed into its antisymmetric
and symmetric parts:

k(u) = ω (u) − θ (u) , (A.2.6)

with

[ω (u)[ ]αβ = P(u)σα P(u)δβ u[δ;σ]

= 21 [d(u)u[ ]αβ ,
(A.2.7)
[θ (u)[ ]αβ = P(u)σα P(u)δβ u(δ;σ)
= 21 [∇(lie) (u) P(u)[ ]αβ = 21 £(u)u (4)gαβ ,

defining the mixed rotation or vorticity tensor field ω (u) (whose sign
depends on convention) and the mixed expansion tensor field θ (u), the
latter of which may itself be decomposed into its tracefree and pure
trace parts
1
θ (u) = σ(u) + Θ(u) P(u) , (A.2.8)
3
where the mixed shear tensor field σ (u) is tracefree (σ(u)α α = 0) and
the expansion scalar is

Θ(u) = uα ;α = ∗(u)[∇(lie) (u)η (u)] . (A.2.9)

1
Define also the rotation or vorticity vector field ω (u) = 2 curlu u as the
spatial dual of the spatial rotation tensor field
1 (4) αβγδ
ω (u)α = 12 η (u)αβγ ω (u) βγ = 2 η u β uγ;δ . (A.2.10)

5. Lie, Fermi-Walker and co-Fermi-Walker derivatives

The kinematical tensor describes the difference between the Lie and
Fermi-Walker temporal derivative operators when acting on spatial ten-

2240
 A. Spacetime splitting techniques in general relativity

sor fields. For example, for a spatial vector field X

∇(fw) (u) X α = ∇(lie) (u) X α − k(u)α β X β

(A.2.11)
= ∇(lie) (u) X α − ω (u)α β X β + θ (u)α β X β ,

where

ω (u)α β X β = −η (u)α βγ ω (u) β X γ = −[ω (u) ×u X ]α . (A.2.12)

The kinematical quantities associated with u may be used to introduce

two spacetime temporal derivatives, the Fermi-Walker derivative and
the co-rotating Fermi-Walker derivative along u

(4)
∇(fw) (u) X α = (4)∇u X α + [ a(u) ∧ u]αβ Xβ ,
(A.2.13)
(4)
∇(cfw) (u) X α = (4)∇(fw) (u) X α + ω (u)α β X β .
These may be extended to arbitrary tensor fields in the usual way (so
that they commute with contraction and tensor products) and they both
commute with index shifting with respect to the metric and with duality
operations on antisymmetric tensor fields since both (4)g and (4)η have
zero derivative with respect to both operators (as does u itself). For an
arbitrary vector field X the following relations hold

£u X α = (4)∇(fw) (u) X α + [ω (u)α β − θ (u)α β + uα a(u) β ] X β

(A.2.14)
= (4)∇(cfw) (u) X α + [−θ (u)α β + uα a(u) β ] X β .

A spatial co-rotating Fermi-Walker derivative ∇(cfw) (u) (“co-rotating

Fermi-Walker temporal derivative”) may be defined in a way analogous
to the ordinary one, such that the three temporal derivatives have the
following relation when acting on a spatial vector field X

∇(cfw) (u) X α = ∇(fw) (u) X α + ω (u)α β X β

(A.2.15)
= ∇(lie) (u) X α + θ (u)α β X β ,

while ∇(cfw) (u)[ f u] = f a(u) determines its action on nonspatial fields.

It has been introduced an index notation to handle these three operators
simultaneously

{∇(tem) (u)}tem=fw,cfw,lie = {∇(fw) (u), ∇(cfw) (u), ∇(lie) (u)} . (A.2.16)

2241
A. Spacetime splitting techniques in general relativity

A.3. Comparing measurements by two observers in

relative motion
Suppose U is another unit timelike vector field representing a different family
of test observers. One can then consider relating the “observations” of each
to the other. Their relative velocities are defined by

U = γ(U, u)[u + ν(U, u)] ,

(A.3.1)
u = γ(u, U )[U + ν(u, U )] ,

where the relative velocity ν(U, u) of U with respect to u is spatial with re-
spect to u and vice versa, both of which have the same magnitude ||ν(U, u)|| =
[ν(U, u)α ν(U, u)α ]1/2 , while the common gamma factor is related to that mag-
nitude by

γ(U, u) = γ(u, U ) = [1 − ||ν(U, u)||2 ]−1/2 = −Uα uα . (A.3.2)

Let ν̂(U, u) be the unit vector giving the direction of the relative velocity
ν(U, u). In addition to the natural parametrization of the worldlines of U
by the proper time τU , one may introduce two new parametrizations: by a
(Cattaneo) relative standard time τ(U,u)

dτ(U,u) /dτU = γ(U, u) , (A.3.3)

which corresponds to the sequence of proper times of the family of observers
from the u congruence which cross paths with a given worldline of the U
congruence, and by a relative standard lenght `(U,u)

d`(U,u) /dτU = γ(U, u)||ν(U, u)|| = ||ν(U, u)||dτ(U,u) /dτU , (A.3.4)

which corresponds to the spatial arc lenght along U as observed by u.

Eqs. (A.3.1) describe a unique active “relative observer boost” B(U, u) in
the “relative observer plane” spanned by u and U such that

B(U, u)u = U , B(U, u)ν(U, u) = −ν(u, U ) (A.3.5)

and which acts as the identity on the common subspace of the local rest
spaces LRSu ∩ LRSU orthogonal to the direction of motion.

A.3.1. Maps between the LRSs of different observers

The projection P(U ) restricts to an invertible map when combined with P(u)
as follows
P(U, u) = P(U ) ◦ P(u) : LRSu → LRSU (A.3.6)

2242
 A. Spacetime splitting techniques in general relativity

with inverse P(U, u)−1 : LRSU → LRSu and vice versa, and these maps also
act as the identity on the common subspace of the local rest spaces.
Similarly the boost B(U, u) restricts to an invertible map

B(lrs) (U, u) ≡ P(U ) ◦ B(U, u) ◦ P(u) (A.3.7)

between the local rest spaces which also acts as the identity on their common
subspace. The boosts and projections between the local rest spaces differ only
by a gamma factor along the direction of motion.
An expression for the inverse projection
If Y ∈ LRSu , then the orthogonality condition 0 = uα Y α implies that Y has
the form
Y = [ν(u, U ) ·U P(U, u)Y ]U + P(U, u)Y . (A.3.8)
If X = P(U, u)Y ∈ LRSU is the field seen by U, then Y = P(U, u)−1 X and

P(U, u)−1 X = [ν(u, U ) ·U X ]U + X = [ P(U ) + U ⊗ ν(u, U )[ ] X , (A.3.9)

which gives a useful expression for the inverse projection.

This map appears in the transformation law for the electric and magnetic
fields:
E(u) = γP(U, u)−1 [ E(U ) + ν(u, U ) ×U B(U )] ,
(A.3.10)
B(u) = γP(U, u)−1 [ B(U ) − ν(u, U ) ×U E(U )] .

A.4. Comparing measurements by three or more

observers in relative motion

A typical situation is that of a fluid/particle whis is observed by two diferrent

families of observers. In this case one deal with three timelike congruences
(or two congruences and a single line): the rest frame of the fluid U and the
two observer families u e u0 .
All the previous formalism can be easily generalized. One has

U = γ(U, u)[u + ν(U, u)] ,

U = γ(U, u0 )[u0 + ν(U, u0 )] ,
(A.4.1)
u0 = γ(u0 , u)[u + ν(u0 , u)] ,
u = γ(u, u0 )[u0 + ν(u, u0 )] .
and mixed projectors involving the various four-velocities can be introduced.
They are summarized in the following table:

2243
A. Spacetime splitting techniques in general relativity

PROJECTORS
P(u, U, u) P(u) + γ(U, u)2 ν(U, u) ⊗ ν(U, u)
P(u, U, u)−1 P(u) − ν(U, u) ⊗ ν(U, u)
P(u, U, u )0 P(u, u0 ) + γ(U, u)γ(U, u0 )ν(U, u) ⊗ ν(U, u0 )
P(u, U, u0 )−1 P(u0 , u) + γ(u, u0 )[(ν(u, u0 ) − ν(U, u0 )) ⊗ ν(U, u)
+ ν(U, u0 ) ⊗ ν(u0 , u)]
P(U, u)−1 P(U, u0 ) P(u, u0 ) + γ(u, u0 )ν(U, u) ⊗ ν(u, u0 )
P(u0 , u) P(U, u)−1 P(U, u0 ) P(u0 ) + δ(U, u, u0 )ν(U, u0 ) ⊗ ν(u, u0 )
P(u0 , u) P(u0 , U, u)−1 P(u0 ) + δ(U, u, u0 )ν(U, u0 ) ⊗ [ν(u, u0 ) − ν(U, u0 )]

where
γ(U, u0 )γ(u0 , u)
δ(U, u, u0 ) = , δ(U, u, u0 )−1 = δ(u, U, u0 ) , (A.4.2)
γ(U, u)

and
P(u, U, u0 ) = P(u, U ) P(U, u0 )

A.5. Derivatives

Suppose one uses the suggestive notation

(4)
D (U )/dτU = (4)∇U (A.5.1)

for the “total covariant derivative” along U. Its spatial projection with respect
to u and rescaling corresponding to the reparametrization of Eq. (A.3.4) is
then given by the “Fermi-Walker total spatial covariant derivative,” defined
by

D(fw,U,u) /dτ(U,u) = γ−1 D(fw,U,u) /dτU = γ−1 P(u)(4)D (U )/dτU

(A.5.2)
= ∇(fw) (u) + ∇(u)ν(U, u) .

Extend this to two other similar derivative operators (the co-rotating Fermi-
Walker and the Lie total spatial covariant derivatives) by

D(tem,U,u) /dτ(U,u) = ∇(tem) (u) + ∇(u)ν(U, u) , tem=fw,cfw,lie , (A.5.3)

2244
 A. Spacetime splitting techniques in general relativity

which are then related to each other in the same way as the corresponding
temporal derivative operators

D(cfw,U,u) X α /dτ(U,u) = D(fw,U,u) X α /dτ(U,u) + ω (u)α β X β

(A.5.4)
= D(lie,U,u) X α /dτ(U,u) + θ (u)α β X β

when acting on a spatial vector field X. All of these derivative operators

reduce to the ordinary parameter derivative D/dτ(U,u) ≡ d/dτ(U,u) when
acting on a function and extend in an obvious way to all tensor fields.
Introduce the ordinary and co-rotating Fermi-Walker and the Lie “relative
accelerations” of U with respect to u by

a(tem) (U, u) = D(tem) (U, u)ν(U, u)/dτ(U,u) , tem=fw,cfw,lie . (A.5.5)

These are related to each other in the same way as the corresponding deriva-
tive operators in Eq. (A.2.15).
The total spatial covariant derivative operators restrict in a natural way to a
single timelike worldline with 4-velocity U, where the D/dτ notation is most
appropriate; (4)D (U )/dτU is often called the absolute or intrinsic derivative
along the worldline of U (associated with an induced connection along such
a worldline).

A.6. Applications

A.6.1. Test-particle motion

Let’s consider the motion of a unit mass test-particle with four velocity U,
accelerated by an external force f (U ): a(U ) = f (U ). A generic observer u can
measure the particle four velocity U, obtaining its relative energy E(U, u) =
γ(U, u) and momentum p(U, u) = γ(U, u)ν(U, u),

U = E(U, u)[u + p(U, u)] = γ(U, u)[u + ν(U, u)]. (A.6.1)

Splitting the acceleration equation gives the evolution (along U) of the rela-
tive energy and momentum of the particle

dE(U, u) (G)
= [ F(tem,U,u) + F (U, u)] · ν(U, u)
dτ(U,u)
+ e(tem) γ(U, u)ν(U, u) · (θ (u) ν(U, u)) (A.6.2)
D(tem) p(U, u) (G)
= F(tem,U,u) + F (U, u) ,
dτ(U,u)

2245
A. Spacetime splitting techniques in general relativity

where tem=fw,cfw,lie,lie[ refers to the various possible (i.e. geometrically

meaningful) transport of vectors along U, e(tem) = (0, 0, −1, 1) respectively
and
dτ(U,u) = γ(U, u)dτU
(G)
F(tem,U,u) = γ(U, u)[ g(u) + H(tem,u) ν(U, u)]
F (U, u) = γ(U, u)−1 P(u, U ) f (U )
with
H(fw,u) = ω (u) − θ (u) H(cfw,u) = 2ω (u) − θ (u)
(A.6.3)
H(lie,u) = 2ω (u) − 2θ (u) H(lie[ ,u) = 2ω (u) .

The gravitoelectric vector field g(u) = − a(u) = −∇u u and the gravito-
magnetic vector field H (u) = 2[∗(u)ω (u)[ ]] of the observer u (sign-reversed
acceleration and twice the vorticity vector field) are defined by the exterior
derivative of u
du[ = [u ∧ g(u) + ∗(u) H (u)][ . (A.6.4)
and will be essential in showing the analogy between the gravitational force
(G)
F(tem,U,u) and the Lorentz force. The expansion scalar Θ(u) = Tr θ (u) ap-
pears in an additional term in the covariant derivative of u as the trace of
the (mixed) expansion tensor θ (u), of which the shear tensor σ(u) = θ (u) −
3 Θ ( u ) P ( u ) is its tracefree part
1

∇u = − a(u) ⊗ u[ + θ (u) − ω (u) . (A.6.5)

The term D(tem) p(U, u)/dτ(U,u) contains itself the “spatial geometry” con-
tribution which must be added to the gravitational and the external force to
reconstruct the spacetime point of view. Actually, this term comes out nat-
urally and is significant all along the line of the particle: the single terms
∇(fw,u) and ∇(u)ν(U,u) , in which it can be further decomposed, are not indi-
vidually meaningful unless one defines some extension for the spatial mo-
mentum p(U, u) off the line of the particle, which of course is unnecessary at
all.
From this spatial geometry contribution a general relativistic version of in-
ertial forces can be further extracted.

A.6.2. Maxwell’s equations

Maxwell’s equations can be expressed covariantly in many ways. For in-
stance, in differential form language one has

dF = 0 , d∗F = −4π ∗J [ , (A.6.6)

2246
 A. Spacetime splitting techniques in general relativity

where F is the Faraday electromagnetic 2-form and J is the current vector

field, obeying the conservation law

δJ [ = ∗d∗J [ = 0 . (A.6.7)

The splitting of the electromagnetic 2-form F by any observer family (with

unit 4-velocity vector field u) gives the associated electric and magnetic vector
fields E(u) and B(u) as measured by those observers through the Lorentz
force law on a test charge, and the relative charge and current density ρ(u)
and J (u). The “relative observer decomposition” of F and its dual 2-form ∗F
is

F = [u ∧ E(u) + ∗(u) B(u)][ ,

∗
F = [−u ∧ B(u) + ∗(u) E(u)][ ,

while J has the representation

J = ρ(u)u + J (u) . (A.6.8)

If U is the 4-velocity of any test particle with charge q and nonzero rest
mass m, it has the orthogonal decomposition

U = γ(U, u)[u + ν(U, u)] . (A.6.9)

Its absolute derivative with respect to a proper time parametrization of its

world line is its 4-acceleration a(U ) = DU/dτU . The Lorentz force law then
takes the form

ma(U ) = qγ(U, u)[ E(u) + ν(U, u) ×u B(u)] . (A.6.10)

The relative observer formulation of Maxwell’s equations is well known.

Projection of the differential form equations (A.6.6) along and orthogonal to
u gives the spatial scalar (divergence) and spatial vector (curl) equations:

~ (u) ·u E(u) = 0 ,
divu B(u) + H
curlu E(u) − ~g(u) ×u E(u) + [£(u)u + Θ(u)] B(u) = 0 ,
(A.6.11)
~ (u) ·u B(u) = 4πρ(u) ,
divu E(u) − H
curlu B(u) − ~g(u) ×u B(u) − [£(u)u + Θ(u)] E(u) = 4π J (u) ,

This representation of Maxwell’s equations differs from the Ellis represen-

tation only in the use of the spatially projected Lie derivative rather than the
spatially projected covariant derivative along u (spatial Fermi-Walker deriva-
tive). These two derivative operators are related by the following identity for

2247
A. Spacetime splitting techniques in general relativity

a spatial vector field X (orthogonal to u)

[£(u)u + Θ(u)] X = [∇(u)u + {−σ(u) + ω (u)} ] X . (A.6.12)

It is clear, at this point, that for any spacetime tensor equation the “1+3”
associated version allows one to read it in a Newtonian form and to interpret
it quasi-classically.
For instance one can consider motion of test fields in a given gravitational
background (i.e. neglecting backreaction) as described by spacetime equa-
tions and look at their “1+3” counterpart. Over the last ten years, in a similar
way in which we have discussed the splitting of Maxwell’s equations in in-
tegral formulation, we have studied scalar field, spinorial field (Dirac fields),
fluid motions, etc.

2248
B. Motion of particles and
extended bodies in General
Relativity

B.1. Introduction

The motion of an extended body in a given background may be studied by

treating the body via a multipole expansion. The starting point of this method
is the covariant conservation law

∇µ T µ ν = 0 , (B.1.1)

where T µν is the energy-momentum tensor describing the body. The body

sweeps out a narrow tube in spacetime as it moves. Let L be a line inside
this tube representing the motion of the body. Denote the coordinates of the
points of this line by X α , and define the displacement δx α = X α − x α , where
x α are the coordinates of the points of the body. Let us consider now the
quantities
Z Z Z
µν λ µν
T dV , δx T dV , δx λ δx ρ T µν dV , ... (B.1.2)

where the integrations are carried out on the 3-dimensional hypersurfaces

of fixed time t = X 0 = const, the tensor T µν being different from zero only
inside the world tube: these are the successive terms of the multipole expan-
sion. A single-pole particle is defined as a particle that has nonvanishing at
least some of the integrals in the first (monopole) term, assuming that all the
integrals containing δx µ vanish. A pole-dipole particle, instead, is defined as
a particle for which all the integrals with more than one factor of δx µ (dipole
term) vanish. Higher order approximations may be defined in a similar way.
Thus, a single-pole particle is a test particle without any internal structure.
A pole-dipole particle, instead, is a test particle whose internal structure is
expressed by its spin, an antisymmetric second-rank tensor defined by
Z h i
µ 0ν ν 0µ
S µν
≡ δx T − δx T dV . (B.1.3)

2249
B. Motion of particles and extended bodies in General Relativity

The equations of motion are, then, obtained by applying the Einstein’s field
equations together with conservation of the energy-momentum tensor (B.1.1)
describing the body. For a single-pole particle this leads to a free particle
moving along the geodesics associated with the given background field. For
the motion of a pole-dipole particle, instead, the corresponding set of equa-
tions was derived by Papapetrou [84] by using the above procedure. Obvi-
ously, the model is worked out under the assumption that the dimensions of
the test particle are very small compared with the characteristic length of the
basic field (i.e., with backreaction neglected), and that the gravitational radia-
tion emitted by the particle in its motion is negligible. As a final remark, note
that this model can be extended to charged bodies by considering in addition
the conservation law of the current density.

B.2. The Mathisson-Papapetrou model

The equations of motion for a spinning (or pole-dipole) test particle in a given
gravitational background were deduced by Mathisson and Papapetrou [84,
85] and read
DPµ 1
= − Rµ ναβ U ν Sαβ ≡ F (spin) µ , (B.2.1)
dτU 2
DSµν
= Pµ U ν − Pν U µ , (B.2.2)
dτU

where Pµ is the total four-momentum of the particle, and Sµν is a (antisym-

metric) spin tensor; U is the timelike unit tangent vector of the “center of mass
line” used to make the multipole reduction. Equations (B.2.1) and (B.2.2) de-
fine the evolution of P and S only along the world line of U, so a correct
interpretation of U is that of being tangent to the true world-line of the spin-
ning particle. The 4-momentum P and the spin tensor S are then defined as
vector fields along the trajectory of U. By contracting both sides of Eq. (B.2.2)
with Uν , one obtains the following expression for the total 4-momentum

DSµν µ
Pµ = −(U · P)U µ − Uν ≡ mU µ + Ps , (B.2.3)
dτU

where m = −U · P reduces to the ordinary mass in the case in which the

particle is not spinning, and Ps is a 4-vector orthogonal to U.
The test character of the particle under consideration refers to its mass as
well as to its spin, since both quantities should not be large enough to con-
tribute to the background metric. In what follows, with the magnitude of the
spin of the particle, with the mass and with a natural lengthscale associated
with the gravitational background we will construct a dimensionless param-
eter as a smallness indicator, which we retain to the first order only so that

2250
 B. Motion of particles and extended bodies in General Relativity

the test character of the particle be fully satisfied. Moreover, in order to have
a closed set of equations Eqs. (B.2.1) and (B.2.2) must be completed with sup-
plementary conditions (SC), whose standard choices in the literature are the
following

1. Corinaldesi-Papapetrou [86] conditions (CP): Sµν (e0 )ν = 0, where e0 is

the coordinate timelike direction given by the background;

2. Pirani [87] conditions (P): Sµν Uν = 0;

3. Tulczyjew [88] conditions (T): Sµν Pν = 0;

all of these are algebraic conditions.

Detailed studies concerning spinning test particles in General Relativity
are due to Dixon [89, 90, 91, 92, 93], Taub [94], Mashhoon [95, 96] and Ehlers
and Rudolph [97]. The Mathisson-Papapetrou model does not give a priori
restrictions on the causal character of U and P and there is no agreement in
the literature on how this point should be considered. For instance, Tod, de
Felice and Calvani [98] consider P timelike, assuming that it represents the to-
tal energy momentum content of the particle, while they do not impose any
causality condition on the world line U, which plays the role of a mere math-
ematical “tool” to perform the multipole reduction. Differently, according to
Mashhoon [96], P can be considered analogously to the canonical momentum
of the particle: hence, there should be not any meaning for its causality char-
acter, while the world line U has to be timelike (or eventually null) because
it represents the center of mass line of the particle. This uncertainty in the
model itself then reflects in the need for a supplementary condition, whose
choice among the three mentioned above is arbitrary, making the general rel-
ativistic description of a spinning test particle somehow unsatisfactory. When
both U and P are timelike vectors as e0 , all of them can be taken as the 4-
velocity field of a preferred observer family, and all the SC above state that
for the corresponding observer the spin tensor is purely spatial. In a sense,
only P and T supplementary conditions give “intrinsic” relations between the
various unknown of the model and they should be somehow more physical
conditions. In fact, the CP conditions are “coordinate dependent,” being e0
the coordinate timelike vector. It is worth to mention that grounded on phys-
ical reasons, Dixon has shown that the T conditions should be preferred with
respect to the others.

B.3. The Dixon-Souriau model

The equations of motion for a charged spinning test particle in a given grav-
itational as well as electromagnetic background were deduced by Dixon-

2251
B. Motion of particles and extended bodies in General Relativity

Souriau [99, 100, 101, 102]. They have the form

DPµ 1 λ
= − Rµ ναβ U ν Sαβ + qF µ ν U ν − Sρσ ∇µ Fρσ ≡ F (tot) µ , (B.3.1)
dτU 2 2
DS µν
= Pµ U ν − Pν U µ + λ[Sµρ Fρ ν − Sνρ Fρ µ ] , (B.3.2)
dτU

where F µν is the electromagnetic field, Pµ is the total 4-momentum of the par-

ticle, and Sµν is the spin tensor (antisymmetric); U is the timelike unit tangent
vector of the “center of mass line” used to make the multipole reduction. As
it has been shown by Souriau, the quantity λ is an arbitrary electromagnetic
coupling scalar constant. We note that the special choice λ = −q/m (see
[46]) in flat spacetime corresponds to the Bargman-Michel-Telegdi [103] spin
precession law.

B.4. Particles with quadrupole structure

The equations of motion for an extended body in a given gravitational back-
ground were deduced by Dixon [89, 90, 91, 92, 93] in multipole approxima-
tion to any order. In the quadrupole approximation they read

DPµ 1 1
= − Rµ ναβ U ν Sαβ − J αβγδ Rαβγδ ; µ ≡ F (spin) µ + F (quad) µ(B.4.1)
dτU 2 6
DS µν 4
= 2P[µ U ν] − J αβγ[µ Rν] αβγ , (B.4.2)
dτU 3
µ
where Pµ = mU p (with U p · U p = −1) is the total four-momentum of the par-
ticle, and Sµν is a (antisymmetric) spin tensor; U is the timelike unit tangent
vector of the “center of mass line” CU used to make the multipole reduc-
tion, parametrized by the proper time τU . The tensor J αβγδ is the quadrupole
moment of the stress-energy tensor of the body, and has the same algebraic
symmetries as the Riemann tensor. Using standard spacetime splitting tech-
niques it can be reduced to the following form

J αβγδ = Παβγδ − ū[α π β]γδ − ū[γ π δ]αβ − 3ū[α Q β][γ ūδ] , (B.4.3)

where Qαβ = Q(αβ) represents the quadrupole moment of the mass distribu-
tion as measured by an observer with 4-velocity ū. Similarly π αβγ = π α[ βγ]
(with the additional property π [αβγ] = 0) and Παβγδ = Π[αβ][γδ] are essen-
tially the body’s momentum and stress quadrupoles. Moreover the various
fields Qαβ , π αβγ and Παβγδ are all spatial (i.e. give zero after any contraction
by ū). The number of independent components of J αβγδ is 20: 6 in Qαβ , 6 in
Παβγδ and 8 in π αβγ . When the observer ū = U p , i.e. in the frame associated

2252
 B. Motion of particles and extended bodies in General Relativity

with the momentum of the particle, the tensors Qαβ , π αβγ and Παβγδ have an
intrinsic meaning.
There are no evolution equations for the quadrupole as well as higher mul-
tipoles as a consequence of the Dixon’s construction, so their evolution is
completely free, depending only on the considered body. Therefore the sys-
tem of equations is not self-consistent, and one must assume that all unspec-
ified quantities are known as intrinsic properties of the matter under consid-
eration.
In order the model to be mathematically correct the following additional
condition should be imposed to the spin tensor:

Sµν U pν = 0. (B.4.4)

Such supplementary conditions (or Tulczyjew-Dixon conditions [88, 89]) are

necessary to ensure the correct definition of the various multipolar terms.
Dixon’s model for structured particles originated to complete and give
a rigorous mathematical support to the previously introduced Mathisson-
Papapetrou model [84, 85, 86, 87], i.e. a multipole approximation to any order
which includes evolutional equations along the “center line” for all the var-
ious structural quantities. The models are then different and a comparison
between the two is possible at the dipolar order but not once the involved
order is the quadrupole.
Here we limit our considerations to Dixon’s model under the further sim-
plifying assumption[94, 97] that the only contribution to the complete quadrupole
moment J αβγδ stems from the mass quadrupole moment Qαβ , so that π αβγ =
0 = Παβγδ and
[α δ]
J αβγδ = −3U p Q β][γ U p , Qαβ U p β = 0 ; (B.4.5)

The assumption that the particle under consideration is a test particle means
that its mass, its spin as well as its quadrupole moments must all be small
enough not to contribute significantly to the background metric. Otherwise,
backreaction must be taken into account.

B.5. Null multipole reduction world line: the

massless case
The extension of the Mathisson-Papapetrou model to the case of a null multi-
pole reduction world line l has been considered by Mashhoon [96]: the model
equations have exactly the same form as (B.2.1) and (B.2.2), with U (timelike)
replaced by l (null) for what concerns the multipole reduction world line and
τU (proper time parametrization of the U line) replaced by λ (affine parame-

2253
B. Motion of particles and extended bodies in General Relativity

ter along the l line):

DPα 1
= − Rα βρσ l β Sρσ ≡ F (spin) α , (B.5.1)
dλ 2
DS αβ
= [ P ∧ l ]αβ . (B.5.2)
dλ
Equations (B.5.1) and (B.5.2) should be then solved assuming some SC. Let
us limit ourselves to the case of “intrinsic” SC, i.e. Pirani and Tulczyjew, with
Pirani’s conditions now naturally generalized as Sαβ l β = 0. Furthermore,
we require P · l = 0: in fact, we are interested to the massless limit of the
Mathisson-Papapetrou equations, and as the mass of the particle is defined
by m = − P · U the massless limit implies − P · l = 0.

By denoting with {l = e1 , n = e2 , m = e3 , m̄ = e4 } a complex null frame

along the center line l, such that l · l = n · n = m · m = 0, l · n = 1, l · m =
l · m = 0 and m · m̄ = −1, it is possible to parametrize the path so that

Dl µ
= b̄mµ + bm̄µ ,
dλ
Dnµ
= āmµ + am̄µ ,
dλ
Dmµ
= al µ + bnµ + icmµ , (B.5.3)
dλ
where a, b, c are functions of λ and c is real. The metric signature is assumed
now + − − − in order to follow standard notation of Newman-Penrose for-
malism, and the bar over a quantity denotes complex conjugation. Equa-
tions (B.5.3) are the analogous of the FS relations for null lines so that, repeat-
ing exactly the above procedure, one gets the final set of equations. Since
for a massless spinning test particle we have m = − P · l = 0, the total 4-
momentum P has the following decomposition:

Pµ = −[ Bl µ + Amµ + Ām̄µ ] . (B.5.4)

Following Mashhoon [96], Tulczyjew’s conditions Sαβ Pβ = 0 are in general

inconsistent in the presence of a gravitational background if in addition one
has P lightlike: P · P = 0. Thus, even if these inconsistencies concern only
the case of null P, we are clearly forced to consider Pirani’s SC as the only
physically meaningful supplementary conditions. Using the P supplemen-
tary conditions (implying b = 0), Mashhoon has shown that l is necessarily
geodesics: Dl µ /dλ = 0 and

Sµν = f (λ)[l ∧ m]µν + f¯(λ)[l ∧ m̄]µν + ig(λ)[m ∧ m̄]µν , (B.5.5)

2254
 B. Motion of particles and extended bodies in General Relativity

with B real and

df
A= + ic f − ig ā , Pµ Pµ = −2| A|2 . (B.5.6)
dλ
so that P is in general spacelike or eventually null. Furthermore, he has
shown that the spin vector defined by

1 µναβ
Sµ = η lν Sαβ (B.5.7)
2
is constant along l and either parallel or antiparallel to l.
Finally, the generalized momentum of the particle should be determined by
solving equations (B.5.1) and (B.5.2) supplemented by Sαβ l β = 0. The other
components of the spin tensor not summarized by the spin vector should be
determined too. By assuming a = 0 (n parallel propagated along l) without
any loss of the physical content of the solution, Mashhoon has obtained for f
and B the following differential equations:
 2
d
+ ic f = f R1413 + f¯R1414 + igR1434 ,
dλ
dB
− = f R1213 + f¯R1214 + igR1234 , (B.5.8)
dλ
which determine the total 4-momentum and the spin tensor along the path
once they have been specified initially.

B.6. Applications
B.6.1. The special case of constant frame components of the
spin tensor
Due to the mathematical complexity in treating the general case of non-con-
stant frame components of the spin tensor, we have considered first the sim-
plest case of massive spinning test particles moving uniformly along circular
orbits with constant frame components of the spin tensor with respect to a
naturally geometrically defined frame adapted to the stationary observers in
the Schwarzschild spacetime [104] as well as in other spacetimes of astro-
physical interest: Reissner-Nordström spacetime [105], Kerr spacetime [5],
superposed static Weyl field [106], vacuum C metric [107]. A static spin vec-
tor is a very strong restriction on the solutions of the Mathisson-Papapetrou
equations of motion. However, this assumption not only greatly simplifies
the calculation, but seems to be not so restrictive, since, as previously demon-
strated at least in the Schwarzschild case, the spin tensor components still re-

2255
B. Motion of particles and extended bodies in General Relativity

main constant under the CP an T choices of supplementary conditions, start-

ing from the more general non-constant case.
We have confined our attention to spatially circular equatorial orbits in
Schwarzschild, Reissner-Nordström and Kerr spacetimes, and searched for
observable effects which could eventually discriminate among the standard
supplementary conditions. We have found that if the world line chosen for
the multipole reduction and whose unit tangent we denote as U is a circular
orbit, then also the generalized momentum P of the spinning test particle is
tangent to a circular orbit even though P and U are not parallel 4-vectors.
These orbits are shown to exist because the spin induced tidal forces provide
the required acceleration no matter what supplementary condition we select.
Of course, in the limit of a small spin the particle’s orbit is close of being a
circular geodesic and the (small) deviation of the angular velocities from the
geodesic values can be of an arbitrary sign, corresponding to the possible
spin-up and spin-down alignment to the z-axis. When two massive particles
(as well as photons) orbit around a gravitating source in opposite directions,
they make one loop with respect to a given static observer with different ar-
rival times. This difference is termed clock effect (see [50, 108, 109, 110, 111]
and references therein). Hereafter we shall refer the co/counter-rotation as
with respect to a fixed sense of variation of the azimuthal angular coordinate.
In the case of a static observer and of timelike spatially circular geodesics the
coordinate time delay is given by
 
1 1
∆t(+,−) = 2π + , (B.6.1)
ζ+ ζ−

where ζ ± denote angular velocities of two opposite rotating geodesics. In the

case of spinless neutral particles in geodesic motion on the equatorial plane of
both Schwarzschild and Reissner-Nordström spacetimes one has ζ + = −ζ − ,
and so the clock effect vanishes; in the Kerr case, instead, the clock effect reads
∆t(+,−) = 4πa, where a is the angular momentum per unit mass of the Kerr
black hole. These results are well known in the literature. We have then ex-
tended the notion of clock effect to non geodesic circular trajectories consid-
ering co/counter-rotating spinning-up/spinning-down particles. In this case
we have found that the time delay is nonzero for oppositely orbiting both
spin-up or spin-down particles even in both Schwarzschild and Reissner-
Nordström cases, and can be measured. In addition, we have found that
a nonzero gravitomagnetic clock effect appears in the Reissner-Nordström
spacetime for spinless (oppositely) charged particles as well.
An analogous effect is found in the case of superposed Weyl fields corre-
sponding to Chazy-Curzon particles and Schwarzschild black holes when the
circular motion of spinning test particles is considered on particular symme-
try hyperplanes, where the orbits are close to a geodesic for small values of
the spin. In the case of the C metric, instead, we have found that the orbital

2256
 B. Motion of particles and extended bodies in General Relativity

frequency is in general spin-dependent, but there is no clock effect, in contrast

to the limiting Schwarzschild case.

B.6.2. Spin precession in Schwarzschild and Kerr spacetimes

We have then studied the behaviour of spinning test particles moving along
equatorial circular orbits in the Schwarzschild [6] as well as Kerr [7] space-
times within the framework of the Mathisson-Papapetrou approach supple-
mented by standard conditions, in the general case in which the components
of the spin tensor are not constant along the orbit. We have found that preces-
sion effects occur only if the Pirani’s supplementary conditions are imposed,
where one finds a Fermi-Walker transported spin vector along an accelerated
center of mass world line. The remaining two supplementary conditions ap-
parently force the test particle center of mass world line to deviate from a
circular orbit because of the feedback of the precessing spin vector; in addi-
tion, under these choices of supplementary conditions the spin tensor com-
ponents still remain constant. In reaching these conclusions, we only consid-
ered solutions for which both U and P are timelike vectors, in order to have
a meaningful interpretation describing a spinning test particle with nonzero
rest mass.

B.6.3. Massless spinning test particles in vacuum

algebraically special spacetimes
As a final application, we have derived the equations of motion for massless
spinning test particles in general vacuum algebraically special spacetimes,
using the Newman-Penrose formalism, in the special case in which the mul-
tipole reduction world line is aligned with a principal null direction of the
spacetime [112]. This situation gives very simple equations and their com-
plete integration is straightforward. Explicit solutions corresponding to some
familiar Petrov type D and type N spacetimes (including Schwarzschild, Taub-
NUT, Kerr, C metric, Kasner, single exact gravitational wave) are derived and
discussed. Furthermore, we have investigated the motion along (null) circu-
lar orbits, providing explicit solutions in black hole spacetimes.

B.6.4. Quadrupole effects in black hole spacetimes

We have studied the motion of quadrupolar particles on a Schwarzschild as
well as Kerr backgrounds [113, 114] following Dixon’s model. In the sim-
plified situation of constant frame components (with respect to a natural or-
thonormal frame) of both the spin and the quadrupole tensor of the parti-
cle we have found the kinematical conditions to be imposed to the particle’s
structure in order the orbit of the particle itself be circular and confined on the

2257
B. Motion of particles and extended bodies in General Relativity

equatorial plane. Co-rotating and counter-rotating particles result to have a

non-symmetric speed in spite of the spherical symmetry of the background,
due to their internal structure. This fact has been anticipated when studying
spinning particles only, i.e. with vanishing quadrupole moments. We show
modifications due to the quadrupole which could be eventually observed in
experiments. Such experiment, however, cannot concern standard clock ef-
fects, because in this case we have shown that there are no contributions aris-
ing from the quadrupolar structure of the body. In contrast, the effect of the
quadrupole terms could be important when considering the period of revo-
lution of an extended body around the central source: measuring the period
will provide an estimate of the quantities determining the quadrupolar struc-
ture of the body, if its spin is known.
It would be of great interest to extend this analysis to systems with varying
quadrupolar structure and emitting gravitational waves without perturbing
significantly the background spacetime.

B.6.5. Quadrupole effects in gravitational wave spacetimes

We have studied how a small extended body at rest interacts with an incom-
ing single plane gravitational wave. The body is spinning and also endowed
with a quadrupolar structure, so that due to the latter property it can be thus
considered as a good model for a gravitational wave antenna.
We have first discussed the motion of such an extended body by assuming
that it can be described according to Dixon’s model and that the gravitational
field of the wave is weak, i.e. the “reaction” (induced motion) of a “gravi-
tational wave antenna” (the extended body) to the passage of the wave, and
then the case of an exact plane gravitational wave. We have found that in gen-
eral, even if initially absent, the body acquires a dipolar moment induced by
the quadrupole tensor, a property never pointed out before in the literature.
Special situations may occur in which certain spin components change
their magnitude leading to effects (e.g. spin-flip) which can be eventually ob-
served. This interesting feature recalls the phenomenon of glitches observed
in pulsars: a sudden increase in the rotation frequency, often accompanied by
an increase in slow-down rate. The physical mechanism triggering glitches
is not well understood yet, even if these are commonly thought to be caused
by internal processes. If one models a pulsar by a Dixon’s extended body,
then the present analysis shows that a sort of glitch can be generated by the
passage of a strong gravitational wave, due to the pulsar quadrupole struc-
ture. In fact, we have found that the profile of a polarization function can be
suitably selected in order to fit observed glitches and in particular to describe
the post-glitch behavior.

2258
 B. Motion of particles and extended bodies in General Relativity

B.6.6. Quadrupolar particles and the equivalence principle

We have compared the two “reciprocal ” situations of motion of an extended
body endowed with structure up to the mass quadrupole moment in a Shwarz-
schild background spacetime (as described by Dixon’s model) with that of a
test particle in geodesic motion in the background of an exact solution of
Einstein’s field equations describing a source with quadrupolar structure (for
a more detailed study of this kind of solutions generalizing Schwarzschild,
Kerr and Kerr-Newman spacetimes see also the section “Generalizations of
the Kerr-Newman solution,” included in the present report). Under certain
conditions the two situations give perfect corresponding results a fact which
has been interpreted as an argument in favour of the validity of Dixon’s
model.

B.6.7. Poynting-Robertson-like effects

Test particle motion in realistic gravitational fields is of obvious astrophysical
importance and at the same time it provides reliable evidence of the proper-
ties of those gravitational fields. However, in many actual astrophysical sys-
tems the particles are not moving freely but are influenced by ambient matter,
electromagnetic fields and radiation. In typical situations, these “physical”
effects are probably even more important than fine details of the spacetime
geometry alone. The most remarkable conditions, from the point of view of
general relativity as well as astrophysics, appear near very compact objects
where both the pure gravitational and other “physical” effects typically be-
come extraordinarily strong.
In a series of papers, recently, we have focused on the motion of test par-
ticles in a spherically symmetric gravitational field, under the action of a
Thomson-type interaction with radiation emitted or accreted by a compact
center. This kind of problem was first investigated by Poynting using New-
tonian gravity and then in the framework of linearized general relativity by
Robertson (see [165] and the references therein). It involves competition be-
tween gravity and radiation drag, which may lead to interesting types of mo-
tion which do not occur in strictly vacuum circumstances. In particular, there
arises the question of whether equilibrium behavior like circular orbit motion
or even “staying at rest” are possible in some cases. Theoretical aspects of the
Poynting-Robertson effect as well as its astrophysical relevance in specific
situations have been studied by many authors since the original pioneering
work. We first considered this effect on test particles orbiting in the equa-
torial plane of a Schwarzschild or Kerr black hole, assuming that the source
of radiation is located symmetrically not far from the horizon (in the case of
outgoing flux). Successively, we have generalized these results by includ-
ing in our discussion other relevant spacetimes, e.g. Vaidya, or considering
spinning particles undergoing Poynting-Robertson-like effect.

2259
C. Metric and curvature
perturbations in black hole
spacetimes

C.1. Perturbations of charged and rotating Black

hole

The gravitational and electromagnetic perturbations of the Kerr-Newman

metric represent still an open problem in General Relativity whose solution
could have an enormous importance for the astrophysics of charged and ro-
tating collapsed objects. A complete discussion about this problems needs
a plenty of different mathematical tools: the Newman-Penrose formalism in
the tetradic and spinor version, the Cahen-Debever-Defrise self dual theory,
the properties of the spin-weighted angular harmonics, with particular atten-
tion to the related differential geometry and the group theory, some tools of
complex analysis, etc, but in any case it is difficult to handle with the pertur-
bative equations. Fortunately, during the last years, the modern computers
and software have reached an optimal computational level which allows now
to approach this problem from a completely new point of view.
The Kerr-Newman solution in Boyer-Lindquist coordinates is represented
by the metric:

2a sin2 θ Σ
 
2 V
ds = 1− dt2 + Vdtdφ − dr2
Σ Σ ∆
" #
a2 sin2 θ
−Σdθ 2 − r2 + a2 + V sin2 θdφ2 (C.1.1)
Σ

where as usual:

V ≡ 2Mr − Q2 (C.1.2)
∆ ≡ r2 − 2Mr + a2 + Q2
Σ ≡ r2 + a2 cos2 θ

2261
C. Metric and curvature perturbations in black hole spacetimes

and by the vector potential:

Qr
A[ = Aµ dx µ = (dt − a sin2 θ dφ) . (C.1.3)
Σ
To investigate the geometrical features of this metric it is convenient to in-
troduce a symmetry-adapted tetrad. For any type D metric, and in par-
ticular for the Kerr-Newman solution, the best choice is a null tetrad with
two “legs” aligned along the two repeated principal null directions of the
Weyl tensor. The standard theory for analyzing different spin massless wave
fields in a given background is represented by the spinorial tetradic formal-
ism of Newman-Penrose (hereafter N-P)[115]. Here we follow the standard
approach, pointing out that a more advanced reformulation of this formal-
ism, called “GHP” [116] exists, allowing a more geometric comprehension of
the theory. In the N-P formalism, this solution is represented by the follow-
ing quantities [117] (in this section we use an A label over all quantities for a
reason which will be clear later). The Kinnersley tetrad [118]:

1 2
(l µ ) A = [r + a2 , ∆, 0, a]
∆
1 2
(nµ ) A = [r + a2 , −∆, 0, a] (C.1.4)
2Σ
1 i
(mµ ) A =√ [ia sin θ, 0, 1, ],
2(r + ia cos θ ) sin θ

with the 4th leg represented by the conjugate (m∗µ ) A , gives the metric tensor
of Kerr-Newman spacetime the form:
 
0 1 0 0
 1 0 0 0 
η(a)(b) = 
 0 0 0 −1  .
 (C.1.5)
0 0 −1 0

The Weyl tensor is represented by:

Ψ0A = Ψ1A = Ψ3A = Ψ4A = 0

(C.1.6)
Ψ2A = Mρ3 + Q2 ρ ∗ ρ3

and the electromagnetic field is given by:

Q
φ0A = φ2A = 0 , φ1A = . (C.1.7)
2(r − ia cos θ )2

2262
 C. Metric and curvature perturbations in black hole spacetimes

For the Ricci tensor and the curvature scalar we have:

A ∗A
Λ A = 0 , Φnm
A
= 2φm φn (m, n = 0, 1, 2) (C.1.8)

so in Kerr-Newman, the only quantity different from zero is:

A Q2
Φ11 = . (C.1.9)
2Σ2
The spin coefficients, which are linear combination of the Ricci rotation coef-
ficients, are given by:

κ A = σA = λA = νA = eA = 0 ,
−1 −iaρ A ρ∗ A sin θ
ρA = , τA = √ ,
(r − ia cos θ ) 2
−ρ∗ A cot θ ia(ρ A )2 sin θ
βA = √ , πA = √ , (C.1.10)
2 2 2
( ρ A )2 ρ ∗ A ∆ ρ A ρ ∗ A (r − M )
µA = , γA = µA + ,
2 2
α = π − β∗ A .
A A

The directional derivatives are expressed by:

D = l µ ∂µ , ∆ = nµ ∂µ , δ = mµ ∂µ , δ∗ = m∗µ ∂µ . (C.1.11)

Unfortunately in the literature the same letter for (C.1.2) and for the direc-
tional derivative along n it is used. However the meaning of ∆ will always
be clear from the context. The study of perturbations in the N-P formalism is
achieved splitting all the relevant quantities in the form l = l A + l B , Ψ4 =
Ψ4A + Ψ4B , σ = σ A + σ B , D = D A + D B , etc., where the A terms are the
background and the B’s are small perturbations. The full set of perturba-
tive equations is obtained inserting these quantities in the basic equations
of the theory (Ricci and Bianchi identities, Maxwell, Dirac, Rarita-Schwinger
equations etc.) and keeping only first order terms. After certain standard
algebraic manipulations one usually obtains coupled linear PDE’s involving
curvature quantities. In the following, we will omit the A superscript for the
background quantities. Comparing with the standard Regge-Wheeler-Zerilli
[119, 120] approach which gives the equation for the metric, here one gets
the equations for Weyl tensor components. This theory is known as curvature
perturbations. In the case of Einstein-Maxwell perturbed metrics, one gets as
in R-W-Z the well known phenomenon of the “gravitationally induced elec-
tromagnetic radiation and vice versa” [121], which couples gravitational and
electromagnetic fields. In the first formulation, one gets a coupled system for
B and g B quantities. In the N-P approach one has the coupling between
Fµν µν

2263
C. Metric and curvature perturbations in black hole spacetimes

perturbed Weyl and Maxwell tensor components, although it’s possible to

recover the metric perturbations using the curvature one [122]. A discussion
about the connections between these two approaches can be found in [123].
To make a long story short, taking in account the two Killing vectors of this
spacetime, one can write the unknown functions in the form:

F (t, r, θ, φ) = e−iωt eimφ f (r, θ ) . (C.1.12)

In the easier cases of Kerr, Reissner-Nordstrom and Schwarzchild, writing

f (r, θ ) = R(r )Y (θ ) one gets separability of the problem. For instance, the
Reissner-Nordström case [124] is separable using the spin-weighted spherical
harmonics:
  2
m + s2 + 2ms cos θ
  
1 d d m m
sin θ − s Y l ( θ ) = − l ( l + 1) s Y l ( θ )
sin θ dθ dθ sin2 θ
(C.1.13)
and their related laddering operators:
 
d m cos θ
q
m
− −s s Y l ( θ ) = − (l − s)(l + s + 1)s+1 Y m l (θ ) (C.1.14)
dθ sin θ sin θ
 
d m cos θ
q
m m
+ +s s Y l ( θ ) = + ( l + s )( l − s + 1)s−1 Y l ( θ ) . (C.1.15)
dθ sin θ sin θ
The unknown functions can be cast in the form:
(2)
Ψ0B = e−iωt eimφ 2 Y m l (θ ) Rl (r )
(1)
χ1B = e−iωt eimφ 1 Y m l (θ ) Rl (r ) (C.1.16)
∆ (−1)
B
χ− 1 = e−iωt eimφ −1 Y m l (θ ) 2 Rl (r )
2r
∆2 (−2)
Ψ4B = e−iωt eimφ −2 Y m l (θ ) 4 Rl (r )
4r
where ∆ = r2 − 2Mr + Q2 , and after manipulations, one gets two sets of

2264
 C. Metric and curvature perturbations in black hole spacetimes

coupled ODE’s. The first set is:

 4 r (r − M )

Q2

d2
2r
−ω + 4iωr −2 + + − ∆
∆ ∆ 3Mr − 4Q2 dr2
4Q2 ∆ 2Q2
 
d
− 6(r − M ) − − 4 −
r (3Mr − 4Q2 ) dr r2
4Q2 (r2 + 2Mr − 3Q2 ) 3Mr − 4Q2

(2)
+ 2 2
+ 2
(l − 1)(l + 2) Rl (C.1.17)
r (3Mr − 4Q ) 3Mr − 2Q
√ p 3
2 2Q (l − 1)(l + 2)r r2

d 4
= − iω + +
3Mr − 2Q2 ∆ dr r
4Q 2 
(1)
− 2
Rl
r (3Mr − 4Q )

 4 r (r − M )

Q2

d2
2r
−ω + 2iωr −2 + − − ∆
∆ ∆ 3Mr − 2Q2 dr2
2Q2 ∆ 18r2 − 24Mr + 2Q2
 
6∆ d
− + 4(r − M ) − −
r r (3Mr − 2Q2 ) dr r2
12Q2 ∆ 3Mr − 2Q2

(1)
+ 2 2
+ 2
(l − 1)(l + 2) Rl (C.1.18)
r (3Mr − 2Q ) 3Mr − 4Q
√
− 2Q2 (l − 1)(l + 2)∆
p
r2 2 4(r − M )

d
= iω + − +
r3 (3Mr − 4Q2 ) ∆ dr r ∆
2Q2

(2)
− 2
Rl .
r (3Mr − 2Q )
(−1) ∗ (−2) ∗ B e Ψ B ), satisfy the same equations
The quantities ( Rl ) e ( Rl ) (from χ− 1 4
(1) (2)
of Rl and Rl . At this point decoupling this system of ordinary differential
equations is straightforward.

Similarly, the Kerr case is separable using but the so-called spin-weighted
spheroidal harmonics [8, 125]:

(H0 + H1 )Θ(θ ) = − EΘ(θ ) (C.1.19)

where:
m2 + s2 + 2ms cos θ
    
1 d d
H0 = sin θ − (C.1.20)
sin θ dθ dθ sin2 θ

H1 = a2 ω 2 cos2 θ − 2aωs cos θ (C.1.21)

2265
C. Metric and curvature perturbations in black hole spacetimes

and E is the eigenvalue. We have factorized the spherical and the spheroidal
parts to give the problem the form of a typical Quantum Mechanics exercise.
In fact depending if the H1 term is small or not, the way to approach the prob-
lem is very different. Unfortunately, in this case the laddering operators are
not know [126] and this does not allow the same strategy used in the case of
the Reissner-Nordström spacetime. In the case of the Kerr spacetime instead,
this is not a problem because laddering operators are unnecessary to solve
completely the problem. In the case of the Kerr-Newman spacetime this cre-
ates a “formal” problem. In fact the presence of the charge Q generates “ugly”
terms which don’t allow the separation of variables in all known coordinates.
A hypothetical separation of variables in these coordinates would have been
stopped by the explicit absence of laddering operators. During the last 25
years there have been various attempts to solve this problem. One idea, pro-
posed in Chandrasekhar’s monography [126], is to decouple the PDE’s be-
fore the separation of variables, obtaining 4th order or higher linear PDE’s.
This task could be accomplished only using a super-computer, because of the
4th order derivatives. Another formulation was developed using de Cahen-
Debever-Defrise formalism, but a part some elegant conservative equations,
the problem has not been solved [127, 128]. In conclusion the problem re-
mains still open. A new approach has been developed [9, 10] for vacuum
spacetimes which gives directly the full set of perturbative equations. The
direct extension of this work to the case of Einstein-Maxwell or more compli-
cate spacetimes can put in a new light this difficult problem.

After this short historical overview we can discuss the results obtained by
ICRANet researchers in this field. In [129], due to Cherubini and Ruffini,
gravitational and electromagnetic perturbations to the Kerr-Newman space-
time using Maple tensor package are shown; a detailed analysis for slightly
charged, rotating and oblate black hole is presented too. Subsequent to this
article there have been various studies regarding the Teukolsky Master Equa-
tions (TMEs) in General Relativity. To this aim, a new form is found for the
Teukolsky Master Equation in Kerr and interpreted in terms of de Rham-
Lichenrowicz laplacians. The exact form of these generalized wave equations
in any vacuum spacetime is given for the Riemann and Maxwell tensors, and
the equations are linearized at any order, obtaining a hierarchy. It is shown
that the TME for any Petrov type D spacetime is nothing more than a com-
ponent of this laplacian linearized and that the TME cannot be derived by
variational principles [9, 10]. More in detail, the Teukolsky Master Equation
in the Kerr case, can be cast in a more compact form (Bini-Cherubini-Jantzen-

2266
 C. Metric and curvature perturbations in black hole spacetimes

Ruffini form) by introducing a “connection vector” whose components are:

1 M (r 2 − a2 )
 
Γt = − − (r + ia cos θ )
Σ ∆
1
Γr = − (r − M )
Σ
Γθ = 0 
1 a (r − M )

cos θ
Γφ = − +i 2 . (C.1.22)
Σ ∆ sin θ
It’s easy to prove that:

1 1
∇µ Γµ = − , Γµ Γµ = cot2 θ + 4ψ2A (C.1.23)
Σ Σ
and consequently the Teukolsky Master Equation assumes the form:

[(∇µ + sΓµ )(∇µ + sΓµ ) − 4s2 ψ2A ]ψ(s) = 4πT (C.1.24)

where ψ2A is the only non vanishing NP component of the Weyl tensor in
the Kerr background in the Kinnersley tetrad (C.1.5) (with Q = 0). Equa-
tion (C.1.24) gives a common structure for these massless fields in the Kerr
background varying the “s” index. In fact, the first part in the lhs represents
(formally) a D’Alembertian, corrected by taking into account the spin-weight,
and the second part is a curvature (Weyl) term linked to the “s” index too.
This particular form of the Teukolsky Master Equation forces us to extend
this analysis in the next sections because it suggests a connection between
the perturbation theory and a sort of generalized wave equations which dif-
fer from the standard ones by curvature terms. In fact generalized wave op-
erators are know in the mathematical literature as De Rham-Lichnerowicz
Laplacians and the curvature terms which make them different from the or-
dinary ones are given by the Weitzenböck formulas. Mostly known examples
in electromagnetism are

• the wave equation for the vector potential Aµ :

∇α ∇α Aµ − Rµ λ Aλ = −4π Jµ , ∇α Aα = 0 (C.1.25)

• the wave equation for the Maxwell tensor :

∇µ ∇µ Fνλ + Rρµνλ F ρµ − Rρ λ Fνρ + Rρ ν Fλρ = −8π ∇[µ Jν] (C.1.26)

while for the gravitational case one has

2267
C. Metric and curvature perturbations in black hole spacetimes

• the wave equation for the metric perturbations:

∇α ∇α h̄µν + 2Rαµβν h̄αβ − 2Rα(µ h̄ν) α = 0,

1
∇α h̄µ α = 0, h̄µν = hµν − gµν hα α (C.1.27)
2
• the wave equation for the Riemann Tensor

Rαβ γδ;e e = 4R[α [γ;δ] β] − 2R[α e R β]e γδ − 2Rαµβν Rµνγδ

−4R[α µν[γ R β]µν δ] . (C.1.28)

These equations are “non minimal,” in the sense that they cannot be recov-
ered by a minimal substitution from their flat space counterparts. A similar
situation holds in the standard Quantum Field Theory for the electromagnetic
Dirac equation. In fact, applying for instance to the Dirac equation an “ad
hoc” first order differential operator one gets the second order Dirac equa-
tion

(i/∂ − e/
A + m)(i/∂ − e/
A − m)ψ =
h e µν 2
i
(i∂µ − eAµ )(i∂ − eA ) − σ Fµν − m ψ = 0,
µ µ
(C.1.29)
2
where the notation is obvious. It is easy to recognize in equation (C.1.29) a
generalized Laplacian and a curvature (Maxwell) term applied to the spinor.
Moreover this equation is “non minimal”, in the sense that the curvature
(Maxwell) term cannot be recovered by electromagnetic minimal substitu-
tion in the standard Klein-Gordon equation for the spinor components. The
analogous second order Dirac equation in presence of a gravitational field
also has a non minimal curvature term and reduces to the form:
1
(∇α ∇α + m2 + R)ψ = 0 . (C.1.30)
4
The general TME formalism is applied to other exact solutions of the vacuum
Einstein field equations of Petrov type D. A new analysis of the Kerr-Taub-
NUT black hole is given, focussing on Mashhoon spin-coupling and superra-
diance [130, 59].
More in detail, in [130] Bini, Cherubini and Jantzen studied a single mas-
ter equation describing spin s = 0 − 2 test field gauge and tetrad-invariant
perturbations of the Taub-NUT spacetime. This solution of vacuum Ein-
stein field equations describes a black hole with mass M and gravitomagnetic
monopole moment `. This equation can be separated into its radial and an-
gular parts. The behaviour of the radial functions at infinity and near the

2268
 C. Metric and curvature perturbations in black hole spacetimes

horizon is studied. The angular equation, solved in terms of hypergeomet-

ric functions, can be related both to spherical harmonics of suitable weight,
resulting from the coupling of the spin-weight of the field and the gravito-
magnetic monopole moment of the spacetime, and to the total angular mo-
mentum operator associated with the spacetime’s rotational symmetry. The
results are compared with the Teukolsky master equation for the Kerr space-
time.
In [59] instead Bini, Cherubini, Jantzen and Mashhoon have studied a sin-
gle master equation describing spin s ≤ 2 test fields that are gauge- and
tetrad-invariant perturbations of the Kerr-Taub-NUT (Newman - Unti - Tam-
burino) spacetime representing a source with a mass M, gravitomagnetic
monopole moment −`, and gravitomagnetic dipole moment (angular mo-
mentum) per unit mass a. This equation can be separated into its radial and
angular parts. The behavior of the radial functions at infinity and near the
horizon is studied and used to examine the influence of l on the phenomenon
of superradiance, while the angular equation leads to spin-weighted spheroidal
harmonic solutions generalizing those of the Kerr spacetime. Finally, the
coupling between the spin of the perturbing field and the gravitomagnetic
monopole moment is discussed.
In [69] instead Bini and Cherubini investigate the algebraically special fre-
quencies of Taub-NUT black holes in detail in comparison with known results
concerning the Schwarzschild case. The periodicity of the time coordinate, re-
quired for regularity of the solution, prevents algebraically special frequen-
cies to be physically acceptable. In the more involved Kerr-Taub-NUT case,
the relevant equations governing the problem are obtained. The formalism is
applied to the C-metric, and physical speculations are presented concerning
the spin-acceleration coupling.
In [70] Bini, Cherubini and Mashhoon study the vacuum C metric and its
physical interpretation in terms of the exterior spacetime of a uniformly accel-
erating spherically-symmetric gravitational source. Wave phenomena on the
linearized C metric background are investigated. It is shown that the scalar
perturbations of the linearized C metric correspond to the gravitational Stark
effect. This effect is studied in connection with the Pioneer anomaly.
In [71] instead Bini, Cherubini and Mashhoon analysed the massless field
perturbations of the accelerating Minkowski and Schwarzschild spacetimes.
The results are extended to the propagation of the Proca field in Rindler
spacetime. They examine critically the possibility of existence of a general
spinacceleration coupling in complete analogy with the well-known spinro-
tation coupling. They argue that such a direct coupling between spin and
linear acceleration does not exist.
In [72] Cherubini, Bini, Bruni and Perjes consider vacuum Kasner space-
times, focusing on those that can be parametrized as linear perturbations of
the special Petrov type D case. In particular they analyze in detail the per-
turbations which map the Petrov type D Kasner spacetime into another Kas-

2269
C. Metric and curvature perturbations in black hole spacetimes

ner spacetime of Petrov type I. For these ’quasi-D’ Kasner models they first
investigate the modification to some curvature invariants and the principal
null directions, both related to the Petrov classification of the spacetime. This
simple Kasner example allows one to clarify the fact that perturbed space-
times do not retain in general the speciality character of the background. In
fact, there are four distinct principal null directions, although they are not
necessarily first-order perturbations of the background principal null direc-
tions. Then in the Kasner type D background they derive a Teukolsky master
equation, a classical tool for studying black-hole perturbations of any spin.
This further step allows one to control totally general cosmologies around
such a background as well as to show, from a completely new point of view,
the well-known absence of gravitational waves in Kasner spacetimes.

C.2. Perturbations of a Reissner-Nordström black

hole by a massive point charge at rest and
the “electric Meissner effect”
The problem of the effect of gravity on the electromagnetic field of a charged
particle leading to the consideration of the Einstein-Maxwell equations has
been one of the most extensively treated in the literature, resulting in exact
solutions (see [131] and references therein) as well as in a variety of approxi-
mation methods [132]-[140].
The issue of the interaction of a massive charged particle of mass m and
charge q with a Reissner-Nordström black hole with mass M and charge Q
has been addressed by the ICRANet collaboration: Bini, Geralico and Ruffini
[141, 142, 143]. We have solved this problem by the first order perturbation
approach formulated by Zerilli [120] using the tensor harmonic expansion of
the Einstein-Maxwell system of equations.
The results discussed in [141, 142] gave answer to a problem whose inves-
tigation started long ago by Hanni and Ruffini [137]. They obtained the so-
lution for a charged particle at rest in the field of a Schwarzschild black hole
in the case of test field approximation, i.e. under the conditions q/m  1,
m ≈ 0 and q  M, q  Q, by using the vector harmonic expansion of the
electromagnetic field in curved space. The conditions above imply the so-
lution of the Maxwell equations only in a fixed Schwarzschild metric, since
the perturbation to the background geometry given by the electromagnetic
stress-energy tensor is second order in the particle’s charge and the effect of
the particle’s mass is there neglected. As a result, no constraint on the posi-
tion of the test particle follows from the Einstein equations and the Bianchi
identities: the position of the particle is totally arbitrary.
This same test field approximation has been applied to the case of a Reissner-
Nordström black hole by Leaute and Linet [140]. In analogy with the Schwarz-

2270
 C. Metric and curvature perturbations in black hole spacetimes

schild case, they used the vector harmonic expansion of the electromagnetic
field holding the background geometry fixed. However, this “test field ap-
proximation” is not valid in the present context. In fact, in addition to ne-
glecting the effect of the particle mass on the background geometry, this
treatment also neglects the electromagnetically induced gravitational pertur-
bation terms linear in the charge of the particle which would contribute to
modifying the metric as well.

The correct way to attack the problem is thus to solve the linearized Einstein-
Maxwell equations following Zerilli’s first order tensor harmonic analysis
[120]. In fact the source terms of the Einstein equations comprise the energy-
momentum tensor associated with the particle’s mass, the electromagnetic
energy-momentum tensor associated with the background field as well as
additional interaction terms, to first order in m and q, proportional to the
product of the square of the charge of the background geometry and the par-
ticle’s mass (∼ Q2 m) and to the product of the charges of both the particle and
the black hole (∼ qQ). These terms give origin to the so called “electromag-
netically induced gravitational perturbation” [144]. On the other hand, the
source terms of the Maxwell equations contain the electromagnetic current
associated with the particle’s charge as well as interaction terms proportional
to the product of the black hole’s charge and the particle’s mass (∼ Qm),
giving origin to the “gravitationally induced electromagnetic perturbation”
[145].

This has been explicitly done in [141, 142]. Let us briefly summarize the re-
sults and the properties of the solution derived there. In standard Schwarzschild-
like coordinates the Reissner-Nordström black hole metric is

ds2 = − f (r )dt2 + f (r )−1 dr2 + r2 (dθ 2 + sin2 θdφ2 ) ,

2M Q2
f (r ) = 1 − + 2 , (C.2.1)
r r
with associated electromagnetic field

Q
F=− dt ∧ dr . (C.2.2)
r2

The horizons are located at r± = M ± M2 − Q2 ≡ M ± Γ; we consider

p

the case | Q| ≤ M and the region r > r+ outside the outer horizon, with an
extremely charged hole corresponding to | Q| = M (which implies Γ = 0)
where the two horizons coalesce.

The only nonvanishing components of the stress-energy tensor and of the

2271
C. Metric and curvature perturbations in black hole spacetimes

current density are given by

part m
T00 = 2
f (b)3/2 δ (r − b) δ (cos θ − 1)
2πb
0 q
Jpart = δ (r − b) δ (cos θ − 1) , (C.2.3)
2πb2
which enter the system of combined Einstein-Maxwell equations
 
part em
G̃µν = 8π Tµν + T̃µν ,
µ ∗ αβ
F̃ µν ; ν = 4π Jpart , F̃ ;β =0. (C.2.4)

The quantities denoted by the tilde refer to the total electromagnetic and
gravitational fields, to first order of the perturbation:

g̃µν = gµν + hµν , F̃µν = Fµν + f µν ,

 
em 1 1
T̃µν = g̃ρσ F̃ρµ F̃σν − g̃µν F̃ρσ F̃ ρσ ,
4π 4
1
G̃µν = R̃µν − g̃µν R̃ ; (C.2.5)
2
note that the covariant derivative operation makes use of the perturbed met-
ric g̃µν as well. The corresponding quantities without the tilde refer to the
background Reissner-Nordström geometry (C.2.1) and electromagnetic field
(C.2.2). Following Zerilli’s [120] procedure we expand the fields hµν and f µν
as well as the source terms (C.2.3) in tensor harmonics, imposing then the
Regge-Wheeler gauge [119] to simplify the description of the perturbation.
The perturbation equations are then obtained from the system (C.2.4), keep-
ing terms to first order in the mass m of the particle and its charge q which are
assumed sufficiently small with respect to the black hole mass and charge.
The axial symmetry of the problem about the z axis (θ = 0) allows to put
the azimuthal parameter equal to zero in the expansion, leading to a great
simplification. Furthermore, it is sufficient to consider only electric-parity
perturbations, since there are no magnetic sources [144, 145, 120].
The geometrical perturbations hµν for the electric multipoles in the Regge-
Wheeler gauge are given by
 
eν H0 Yl0 H1 Yl0 0 0
 H1 Yl0 e−ν H2 Yl0
 
0 0 
||hµν || =   , (C.2.6)
 
0 0 r 2 KY 0
l0
 
 
2 2
0 0 0 r sin θKYl0

where Yl0 are normalized spherical harmonics with azimuthal index equal to

2272
 C. Metric and curvature perturbations in black hole spacetimes

zero and eν = f (r ) is Zerilli’s notation. The electromagnetic field harmonics

f µν for the electric multipoles are given by
 
∂Y
0 f˜01 Yl0 f˜02 l0 0

 ∂θ 

∂Y
f˜12 l0
 
|| f µν || =  antisym 0 0 
 , (C.2.7)

 ∂θ 
 antisym antisym 0 0 
 
antisym antisym antisym 0

where f˜µν denotes the θ-independent part of f µν , and the symbol “antisym”
indicates components obtainable by antisymmetry. The expansion of the
source terms (C.2.3) gives the relations

∑ A00Yl0 = 16πT00 ∑ vYl0 = Jpart

part 0
, , (C.2.8)
l l

with
√ √
√ m 2l + 1 3/2 1 q 2l + 1
A00 =8 π f ( b ) δ (r − b ) , v = √ δ (r − b) .(C.2.9)
b2 2 π b2

The Einstein-Maxwell field equations (C.2.4) give rise to the following system
of radial equations for values l ≥ 2 of the multipoles (note that the cases
l = 0, 1 must be treated separately)

ν0
     
2ν 00 2 0 0 6 0 1 2λeν
0 = e 2K − W + ν + K −4 2 + W − 2 (W + K )
r r r r r
2
Q e W ν ν
Qe f ˜
−2 4
− 4 2 01 + A00 ,
r r
2λe−ν Q2 e − ν W Qe−ν f˜01
 
2 0 2
0 = W − ν0 + K0 − ( W − K ) − 2 + 4 ,
r r r2 r4 r2
   
00 0 2 0 00 0 1
0 = K + ν + K −W −2 ν + W0
r r
2ν0 Q2 e−ν K 4Qe−ν ˜
 
00 02
+ ν +ν + (K − W ) − 2 + f 01 ,
r r4 r2
Qe−ν f˜02
0 = −W 0 + K 0 − ν 0 W + 4 ,
r2
2 l (l + 1) e−ν f˜02 Q 0
0 = f˜01 0 + f˜01 − − 2 K + 4πv ,
r r2 r
˜ ˜ 0
0 = f 01 − f 02 , (C.2.10)

since H0 = H2 ≡ W, H1 ≡ 0 and f˜12 ≡ 0, where λ = 1

2 (l − 1) (l + 2) and a

2273
C. Metric and curvature perturbations in black hole spacetimes

prime denotes differentiation with respect to r.

We have a system of 6 coupled ordinary differential equations for 4 un-

known functions: K, W, f˜01 and f˜02 . The compatibility of the system requires
that these equations are not independent. Two equations can indeed be elim-
inated provided that the following stability condition holds

b f (b)1/2
m = qQ , (C.2.11)
Mb − Q2

involving the black hole and particle parameters as well as their separation
distance b. If the black hole is extreme (i.e. Q/M = 1), then the particle
must also have the same ratio q/m = 1, and equilibrium exists indepen-
dent of the separation. In the general non-extreme case Q/M < 1 there is
instead only one position of the particle which corresponds to equilibrium,
for given values of the charge-to-mass ratios of the bodies. In this case the
particle charge-to-mass ratio must satisfy the condition q/m > 1. Note that
quite surprisingly Eq. (C.2.11) coincides with the equilibrium condition for
a charged test particle in the field of a Reissner-Nordström black hole which
has been discussed by Bonnor [146] in the simplified approach of test field
approximation, neglecting all the feedback terms.

We then succeed in the exact reconstruction of both the perturbed grav-

itational and electromagnetic fields by summing all multipoles [142]. The
perturbed metric is given by

ds̃2 = −[1 − H̄ − k(r )] f (r )dt2 + [1 + H̄ + k (r )] f (r )−1 dr2

+(1 + H̄)r2 (dθ 2 + sin2 θdφ2 ) ,
H̄0 Q2
k (r ) = 2 , H̄0 = −2qΓ2 /[ Q(Mb − Q2 )] , (C.2.12)
r f (r )

where
m (r − M)(b − M) − Γ2 cos θ
H̄ = 2 f (b)−1/2 ,
br D̄
D̄ = [(r − M)2 + (b − M)2 − 2(r − M)(b − M) cos θ
−Γ2 sin2 θ ]1/2 . (C.2.13)

It can be shown that this perturbed metric is spatially conformally flat; more-
over, the solution remains valid as long as the condition |H̄|  1 is satisfied.
The total electromagnetic field to first order of the perturbation turns out to
be  
Q
F̃ = − 2 + Er dt ∧ dr − Eθ dt ∧ dθ , (C.2.14)
r

2274
 C. Metric and curvature perturbations in black hole spacetimes

with
q Mr − Q2 1

Er = 3 M(b − M) + Γ2 cos θ
r Mb − Q2 D̄
Q2 [(r − M)(b − M) − Γ2 cos θ ]

+
Mr − Q2
r [(r − M)(b − M) − Γ2 cos θ ]

− [(r − M) − (b − M) cos θ ] ,
D̄2
Mr − Q2 b2 f (b) f (r )
Eθ = q sin θ . (C.2.15)
Mb − Q2 D̄3
Note that in the extreme case Q/M = q/m = 1 this solution reduces to the
linearized form of the well known exact solution by Majumdar and Papa-
petrou [147, 148] for two extreme Reissner-Nordström black holes. Further-
more, this solution satisfies Gauss’ theorem
Z
∗
Φ= F̃ ∧ dS = 4π [ Q + qϑ (r − b)] , (C.2.16)
S

where Φ is the flux of the electric field obtained by integrating the dual of
the electromagnetic form (C.2.14) over a spherical 2-surface S centered at the
origin where the black hole charge Q is placed and with variable radius (r
greater or lesser than b), the function ϑ ( x ) denoting the step function.
Recently an important progress has been achieved by Belinski and Alek-
seev [149]. They have obtained an exact two-body solution of the Einstein-
Maxwell equations in explicit analytic form for the system consisting of a
Reissner-Nordström black hole and a naked singularity, by using the mon-
odromy transform approach [150]. They have shown that an equilibrium
without intervening struts or tensions is possible for such a system at se-
lected values of the separating distance between the sources. Furthermore,
their equilibrium condition exactly reduces to our equation (C.2.11) once lin-
earized with respect to the mass and charge of the naked singularity. We have
indeed been able to show explicitly the coincidence between the linearized
form of their exact solution and our perturbative solution.
We have then analyzed in [143] the properties of the perturbed electric field
with special attention to the construction of the lines of force of the electric
field. The two cases have been considered of the sole particle, with the sub-
traction of the dominant contribution of the black hole, as well as of the total
field due to the black hole and the particle. As the black hole becomes ex-
treme an effect similar to the ordinary Meissner effect for magnetic fields in
the presence of superconductors arises: the electric field lines of the point
charge are expelled outside the outer horizon. Note that this effective “elec-
tric Meissner effect” has no classical analogue, as far as we know, and is a
pure general relativistic effect. Let us discuss this issue more in detail.

2275
C. Metric and curvature perturbations in black hole spacetimes

The electric field lines are defined as the solution of the differential equa-
tion
dx α
= E (U ) α , (C.2.17)
dλ
where λ is an affine parameter for the lines and E(U )α are the coordinate
components of the electric field

E (U ) α = F α β U β (C.2.18)

as measured by an observer with four-velocity U. The shape of the lines thus

depends on the choice of the observer and that of the coordinates which are
used to draw the curves. We refer to the static observers with respect to the
metric (C.2.12), whose four-velocity is given by

H̄ + k (r )
 
1 − 1/2
U= √ ∂ t = f (r ) 1+ ∂t , (C.2.19)
− g̃tt 2

to first order of the perturbation. Eq. (C.2.17) thus becomes

dr dθ
= E (U ) r , = E (U ) θ , (C.2.20)
dλ dλ
leading to the equation

− E(U )r dθ + E(U )θ dr = 0 , (C.2.21)

after eliminating the parameter λ.

For a static spacetime and using a static family of observers the electric lines
of force coincide with the constant flux lines [151]. The flux across a generic
2-surface S is given by
Z 
∗
Φ= F̃rφ dr dφ + ∗ F̃θφ dθ dφ ,

(C.2.22)
S

since the only nonvanishing components of ∗ F̃ are

 
∗ Q (0) (1)
F̃θφ = −r sin θ −(1 + H̄) 2 + f tr ≡ ∗ F̃θφ + ∗ F̃θφ ,
2
r
∗ (1)
F̃rφ = f (r )−1 sin θ f tθ ≡ ∗ F̃rφ , (C.2.23)

where the superscripts (0), (1) refer to the zeroth order and first order terms
respectively. Therefore, as the electromagnetic field components do not de-
pend explicitly on φ, if S is a generic revolution surface around the symmetry

2276
 C. Metric and curvature perturbations in black hole spacetimes

z-axis we can write

Z 
∗
Φ = 2π F̃rφ dr + ∗ F̃θφ dθ ,

(C.2.24)
S

so that the elementary flux across an infinitesimal surface (closed, limited

by the two spherical caps: φ ∈ [0, 2π ], θ = θ0 and r = r0 and φ ∈ [0, 2π ],
θ = θ0 + dθ and r = r0 + dr) of this kind is

dΦ = 2π [∗ F̃rφ dr + ∗ F̃θφ dθ ] . (C.2.25)

The lines of constant electric flux (dΦ = 0) are then defined as those curves
solutions of the equation

0 = ∗ F̃rφ dr + ∗ F̃θφ dθ , (C.2.26)

which coincides with Eq. (C.2.21), noting that

√ √
∗ − g̃ r ∗ − g̃
F̃θφ = − E (U ) , F̃rφ = E (U ) θ . (C.2.27)
U0 U0

We are now ready to draw the electric lines of force by numerically in-
tegrating Eq. (C.2.21). Note that if the total electric field is considered the
contribution of the black hole always dominates (see Fig. C.1).
We are mainly interested in studying the “effective field” representing the
net effect of the perturbation induced by the massive charged particle on the
background electric field. The most natural way to separate the two contri-
butions is to directly use the elementary flux equation (C.2.25). By requiring
that the integration over a spherical 2-surface S centered at the origin gives
the first order contribution Φ(1) = 4πqϑ (r − b) only to the total electric flux
(C.2.16), i.e. the charge of the particle only, we get
(1) (1)
dΦ(1) = 2π [∗ F̃rφ dr + ∗ F̃θφ dθ ] . (C.2.28)

The “effective field” lines corresponding to the perturbation with the contri-
bution of the black hole electric field being subtracted are thus defined as the
lines of constant flux dΦ(1) = 0, namely
(1) (1)
0 = ∗ F̃rφ dr + ∗ F̃θφ dθ , (C.2.29)

according to Eq. (C.2.26), which reduces to the previous one when only the
contribution of those terms which are first order is taken into account.
The behavior of the lines of force of the effective electric field of the particle
alone is shown in Fig. C.2.
Following Hanni and Ruffini [137] we now compute the induced charge on

2277
C. Metric and curvature perturbations in black hole spacetimes

Figure C.1.: Lines of force of the total electric field of the black hole and parti-
cle in the X − Z plane (X = r sin θ, Z = r cos θ are Cartesian-like coordinates)
for charges of the same sign q/Q = 0.1 and fixed parameter values b/M = 3
and Q/M = 0.8.

2278
 C. Metric and curvature perturbations in black hole spacetimes

Figure C.2.: Lines of force of the effective electric field of the sole particle in
the non-extreme case for the same choice of parameters as in Fig. C.1. As
explained in the text this “effective field” is obtained by subtracting the dom-
inant contribution of the black hole own electric field to the total perturbed
field, thus representing the net effect of the perturbation induced by the mas-
sive charged particle on the background field.

2279
C. Metric and curvature perturbations in black hole spacetimes

the surface of the black hole horizon. Some lines of force intersect the horizon.
If the particle is positively charged, at angles smaller than a certain critical
angle the induced charge is negative and the lines of force cross the horizon.
At angles greater than the critical angle the induced charge is positive and
the lines of force extend out of the horizon. At the critical angle the induced
charge density vanishes and the lines of force of the electric field are tangent
to the horizon. The total electric flux through the horizon and thus the total
induced charge are zero.
The induced charge density on the horizon σ H (θ ) is defined in such a way
that the amount of induced charge on an infinitesimal portion of the horizon
sphere r = r+ between θ = θ0 and θ = θ0 + dθ equals 1/(4π ) times the
elementary flux across the same surface

1 1 (1) 
dΦ|r+ = 2π ∗ F̃θφ r dθ = 2πr+
2 H
σ (θ ) sin θdθ , (C.2.30)
4π 4π +

implying
∗ F̃ (1) 

θφ r+
2
= 4πσ H (θ ) . (C.2.31)
r+ sin θ
This can be identified with the surface version of the Gauss’ law. The cor-
responding expression for the critical angle θ(crit) comes from the condition
σ H (θ(crit) ) = 0. Hence it results

q Γ2
σ H (θ ) =
4πr+ Mb − Q2
Γ(1 + cos2 θ ) − 2(b − M) cos θ
× , (C.2.32)
[b − M − Γ cos θ ]2
" #
b − M − ( b − M)2 − Γ2
p
θ(crit) = arccos . (C.2.33)
Γ

Assuming then the black hole and particle both have positive charge, one can
evaluate the total amount of negative charge induced on the horizon by the
particle
Z Z θ
(−) (crit)
H
Qind = σ (θ )dΣ = 2
2πr+ σ H (θ ) sin θdθ
Σ 0
Γr+
= −q cos θ(crit) , (C.2.34)
Mb − Q2
√
where dΣ = gθθ gφφ dθ dφ and Σ is the spherical cap 0 ≤ θ ≤ θ(crit) .
Let us study what happens as the black hole approaches the extreme con-
dition | Q| = M (implying Γ = 0). Eq. (C.2.32) shows that the induced charge

2280
 C. Metric and curvature perturbations in black hole spacetimes

Figure C.3.: Lines of force of the effective electric field of the particle alone
in the extreme case Q/M = 1 for the same choice of the distance parameter
as in Fig. C.1. No lines of force intersect the black hole horizon in this case,
leading to the the “electric Meissner effect.”

density on the horizon degenerates to zero for every value of the angle θ;
the critical angle (C.2.33) approaches the value π/2 and the amount of neg-
ative charge (C.2.34) induced on the horizon vanishes identically. Therefore
no lines of force cross the horizon, remaining tangent to it for every value
of the polar angle, since every angle becomes critical: as the black hole ap-
proaches the extreme condition the electric field lines are thus pulled off the
outer horizon and never intersect it when the black hole becomes extreme.
The situation is summarized in Fig. C.3 showing the behavior of the lines of
force of the effective electric field of the particle alone in the extreme case.
The “electric Meissner effect” above described is suitable to a suggestive
interpretation in terms of the nature of the Reissner-Nordström solution. As
soon as the black hole is not extreme the point particle induces charge on
the horizon, and accordingly the electric field lines terminate on it; when the
black hole becomes extreme no further charge induction is possible (unless
one turns the black hole into a naked singularity), and coherently the electric
field lines no more cross the horizon. In a sense the black hole rejects to turn
itself into a naked singularity and this might be thought of as an argument in

2281
C. Metric and curvature perturbations in black hole spacetimes

favor of the Cosmic Censorship conjecture.

C.3. Perturbations of a de Sitter spacetime

Let us consider the perturbations

g̃µν = gµν + hµν (C.3.1)

of the de Sitter spacetime,

gµν dx µ dx ν = − N 2 dt2 + N −2 dr2 + r2 (dθ 2 + sin2 θdθ 2 ) (C.3.2)

where N 2 = (1 − H 2 r2 ), H being the Hubble constant. Since the background

metric is static and spherically symmetric, we decompose the metric pertur-
bation hµν in tensor harmonics and Fourier transform it with respect to time,
as customary. We will use the Regge-Wheeler [119] gauge to simplify the
form of the perturbation.

C.3.1. Electric multipoles

The metric perturbation hµν , for the electric multipoles, is given by

N 2 H0
 
H1 0 0
1
 
 sym H2 0 0
 
 −iωt
||hµν || = 
 N2 e
 Yl0 , (C.3.3)
 sym
 sym r2 K 0 

sym sym sym r2 sin2 θK

where the symbol “sym” indicates that the missing components of hµν should
be found from the symmetry hµν = hνµ and the functions Yl0 are normalized
spherical harmonics with azimuthal index m = 0, defined by
r
1 2l + 1
Yl0 = Pl (cos θ ) . (C.3.4)
2 π

For l ≥ 2, the system of radial equations we have to solve is the following:

iω 2rH 2
0 = H1 0 + ( W + K ) − H1 ,
N2 N2
W iL K
0 = K0 − − 2
H1 + , (C.3.5)
r 2ωr rN 2

2282
 C. Metric and curvature perturbations in black hole spacetimes

where L = l (l + 1), since

H0 = H2 ≡ W , (C.3.6)

together with the algebraic relation

L−2
 
i 2 2L
0= W − ( H 2 L + 2ω 2 ) H1 + 2
1+r ω −N2
K. (C.3.7)
r ω rN 2 2

Let us introduce the dimensionless variables x = Hr, Ω = ω/H and de-

note H1 = i H̃1 . Solving for W in the constraint equation (C.3.7) and substi-
tuting into the system (C.3.5) we get

(2Ω2 + 3L − 4) x 2Ω[(Ω2 + L − 1) x2 + 2 − L]
H̃1 0 = H̃ + K,
(1 − x2 )( L − 2) 1 (1 − x 2 )2 ( L − 2)
[2x2 ( L + 2Ω2 ) + L( L − 2)] x ( L + 2Ω2 )
K0 = − H̃1 − K , (C.3.8)
2( L − 2)Ωx2 (1 − x2 )( L − 2)
where now a prime denotes differentiation with respect to x, which can be
solved in terms of Heun’s functions.

A real solution for the metric can be obtained by considering W and K and
H̃1 as real. In this case the nonvanishing metric components are

gtt = − N 2 (1 − WYl0 cos ωt)

gtr = H̃1 Yl0 sin ωt
grr = N −2 (1 + WYl0 cos ωt)
gθθ = r2 (1 + KYl0 cos ωt)
gφφ = r2 sin2 θ (1 + KYl0 cos ωt) , (C.3.9)

so that
 2
2 2 −2
ds = − N (1 − WYl0 cos ωt) dt + H̃1 N Yl0 sin ωt dr
+ N −2 (1 + WYl0 cos ωt)dr2
+r2 (1 + KYl0 cos ωt)(dθ 2 + sin2 θdφ2 ) , (C.3.10)

at first order in the perturbation quantities. A natural orthonormal frame

2283
C. Metric and curvature perturbations in black hole spacetimes

associated with this form of the metric is then

 
1 −2

Ω = N 1 − WYl0 cos ωt
0̂
dt − H̃1 N Yl0 sin ωt dr
2
 
1
= N 1 − WYl0 cos ωt dt − H̃1 N −1 Yl0 sin ωt dr
2
 
r̂ −1 1
Ω = N 1 + WYl0 cos ωt dr
2
 
1
Ω = r 1 + KYl0 cos ωt dθ
θ̂
2
 
1
Ω = r sin θ 1 + KYl0 cos ωt dφ .
φ̂
(C.3.11)
2

One can then introduce a NP in a standard way

√ √ √
l = (Ω0̂ + Ωr̂ )/ 2 , n = (Ω0̂ − Ωr̂ )/ 2 , m = (Ωθ̂ + iΩφ̂ )/ 2 . (C.3.12)

C.3.2. Magnetic multipoles

The metric perturbation hµν , for the magnetic multipoles, is given by

0 0 0 h0
 

 sym 0 0 h1  −iωt
 
dY
||hµν || =  e sin θ l0 . (C.3.13)
 sym sym
 0 0  dθ
sym sym sym 0

For l ≥ 2, the system of radial equations we have to solve is the following:

N2 L − 2
 
2h0 0
0 = h0 − + ih1 ω − 2 ,
r r ω
2rH 2 iω
0 = h1 0 − h 1 − h0 , (C.3.14)
N2 N4
where L = l (l + 1).
By introducing the dimensionless variables x = Hr, Ω = ω/H as above
and setting h0 = i h̃0 , the system (C.3.14) becomes

2 ( Ω2 + L − 2) x 2 + 2 − L
h̃0 0 = h̃0 − h1 ,
x Ωx2
Ω 2x
h1 0 = 2 2
h̃0 + h1 , (C.3.15)
(1 − x ) 1 − x2

2284
 C. Metric and curvature perturbations in black hole spacetimes

where now a prime denotes differentiation with respect to x, which can be

solved in terms of Heun’s functions.
A real solution for the metric can be obtained by considering h̃0 and h1 as
real. In this case the nonvanishing metric components are

dYl0 dY
gtt = − N 2 , gtφ = h̃0 sin ωt sin θ , grφ = h1 cos ωt sin θ l0 ,
dθ dθ
grr = N −2 , gθθ = r2 , gφφ 2 2
= r sin θ , (C.3.16)

so that
dYl0
ds2 = ds2(dS) + 2 sin θ


h̃0 sin ωtdφ + 2h1 cos ωtdr dt , (C.3.17)
dθ
at first order in the perturbation quantities. A natural orthonormal frame
associated with this form of the metric is then
sin θ dYl0
Ω0̂ = Ndt + (h̃0 sin ωtdφ + h1 cos ωtdr )
N dθ
Ωr̂ = N −1 dr , Ωθ̂ = rdθ , Ωφ̂ = r sin θdφ . (C.3.18)

One can then introduce a NP in a standard way, as in Eq. (C.3.12).

Writing closed form expressions for the electric and magnetic perturbations
allows for the full reconstruction of the metric, a fact of fundamental impor-
tance to proceed with the comparison with curvature perturbations as well
as to perform a systematic analysis of gauge conditions.

2285
D. Cosmology
The Bianchi type IX spatially homogeneous vacuum spacetime also known
as the Mixmaster universe has served as a theoretical playground for many
ideas in general relativity, one of which is the question of the nature of the
chaotic behavior exhibited in some solutions of the vacuum Einstein equa-
tions and another is the question of whether or not one can interpret the
spacetime as a closed gravitational wave. In particular, to describe the mathe-
matical approach to an initial cosmological singularity, the exact Bianchi type
IX dynamics leads to the BLK approximation involving the discrete BLK map
which acts as the transition between phases of approximately Bianchi type I
evolution. The parameters of this map are not so easily extracted from the
numerical evolution of the metric variables. However, recently it has been
realized that these parameters are directly related to transitions in the scale-
free part of the Weyl tensor. In fact this leads to a whole new interpretation
of what the BLK dynamics represents.
For a given foliation of any spacetime, one can always introduce the scale
free part of the extrinsic curvature when its trace is nonzero by dividing by
that trace. In the expansion-normalized approach to spatially homogeneous
dynamics, this corresponds to the expansion-normalized gravitational veloc-
ity variables. This scale free extrinsic curvature tensor can be characterized
by its eigenvalues, whose sum is 1 by definition: these define three functions
of the time parametrizing the foliation which generalize the Kasner indices of
Bianchi type I vacuum spacetimes. A phase of velocity-dominated evolution
is loosely defined as an interval of time during which the spatial curvature
terms in the spacetime curvature are negligible compared to the extrinsic cur-
vature terms. Under these conditions the vacuum Einstein equations can be
approximated by ordinary differential equations in the time. These lead to a
simple scaling of the eigenvectors of the extrinsic curvature during which the
generalized Kasner indices remain approximately constant and simulate the
Bianchi type I Kasner evolution.
The Weyl tensor can be also be repackaged as a second rank but complex
spatial tensor with respect to the foliation and its scale free part is deter-
mined by a single complex scalar function of its eigenvalues, a number of
particular representations for which are useful. In particular the so called
speciality index is the natural choice for this variable which is independent
of the permutations of the spatial axes used to order the eigenvalues, and so
is a natural 4-dimensional tracker of the evolving gravitational field quoti-
enting out all 3-dimensional gauge-dependent quantities. During a phase of

2287
D. Cosmology

velocity-dominated (“Kasner”) evolution, the Weyl tensor is approximately

determined by the extrinsic curvature alone, and hence the scalefree invari-
ant part of the Weyl tensor is locked to the generalized Kasner indices exactly
as in a Kasner spacetime. Of course during transitions between velocity-
dominated evolution where the spatial curvature terms are important, the
generalized Kasner indices and the Weyl tensor are uncoupled in their evolu-
tion, but the transition between one set of generalized Kasner indices and the
next is locked to a transition in the scalefree Weyl tensor. This idealized map-
ping, approximated by the BKL map between Kasner triplets, can be rein-
terpreted as a continuous transition in the Weyl tensor whose scale invariant
part can be followed through the transition directly. For spatially homoge-
neous vacuum spacetimes, the BLK transition is a consequence of a Bianchi
type II phase of the dynamics which can be interpreted as a single bounce
with a curvature wall in the Hamiltonian approach to the problem. One can
in fact follow this transition in the Weyl tensor directly with an additional
first order differential equation which is easily extracted from the Newman-
Penrose equations expressed in a frame adapted both to the foliation and the
Petrov type of the Weyl tensor.
This type of Weyl transition in the Mixmaster dynamics can be followed ap-
proximately using the Bianchi type II approximation to a curvature bounce,
leading to a temporal spike in the real and imaginary parts of the special-
ity index which represents a circuit in the complex plane between the two
real asymptotic Kasner points (a “pulse”). The graph of the speciality index
versus time thus serves as a sort of electrocardiogram of the “heart” of the
Mixmaster dynamics, stripping away all the gauge and frame dependent de-
tails of its evolution except for the choice of time parametrization, which is a
recent nice result of our investigation.

2288
E. Exact solutions
In this appendix we briefly review a recent work on the Kerr-Schild ansatz
used to search for new exact solutions of Einstein’s field equations. More-
over, we introduce a more recent work on exact solution with stationarity and
axial symmetry and with metric functions as rational functions of the non ig-
norable coordinates. A more detailed discussion of both these works can be
found in the section “The Kerr-Newman solution” of the present report.

E.1. Kerr-Schild metrics and the Kerr-Schild

ansatz revisited
Kerr-Schild metrics have the form [152, 153]

ds2 = gαβ dx α dx β ≡ (ηαβ − 2Hk α k β )dx α dx β , (E.1.1)

where ηαβ is the metric for Minkowski space and k α is a null vector

ηαβ kα k β = gαβ kα k β = 0, kα = η αβ k β = gαβ k β . (E.1.2)

The inverse metric is also linear in H

gαβ = η αβ + 2Hkα k β , (E.1.3)

and so the determinant of the metric is independent of H

β
(ηαγ − 2Hk α k γ )(η γβ + 2Hkγ k β ) = δα −→ | gαβ | = |ηαβ | .

Within this class of general metrics the Kerr solution was obtained in 1963
by a systematic study of algebraically special vacuum solutions [154]. If
( x0 = t, x1 = x, x2 = y, x3 = z) are the standard Cartesian coordinates for
Minkowski spacetime with ηαβ = diag[−1, 1, 1, 1], then for Kerr metric we
have
(rx + ay)dx + (ry − ax )dy z
− k α dx α = dt + + dz , (E.1.4)
r 2 + a2 r
where r and H are defined implicitly by

x 2 + y2 z2 Mr3
+ 2 =1, H=− . (E.1.5)
r 2 + a2 r r 4 + a2 z2

2289
E. Exact solutions

Kerr solution is asymptotically flat and the constants M and a are the total
mass and specific angular momentum for a localized source. They both have
the dimension of a length in geometrized units. The vector k is geodesic and
shearfree, implying that Kerr metric is algebraically special according to the
Goldberg-Sachs theorem [155]. Moreover, k is independent of M and hence a
function of a alone. Note that the mass parameter M appears linearly in the
metric, i.e. in H.
We consider here Kerr-Schild metrics (E.1.1) as exact linear perturbations of
Minkowski space and solve Einstein’s field equations order by order in pow-
ers of H. The results of this analysis will be that k must be geodesic and
shearfree as a consequence of third and second order equations, leading to
an alternative derivation of Kerr solution.

E.1.1. Modified ansatz

Let e be an arbitrary constant parameter, eventually to be set equal to 1, so

that the Kerr-Schild metric (E.1.1) reads

gαβ = ηαβ − 2eHk α k β , (E.1.6)

with inverse
gαβ = η αβ + 2eHkα k β , (E.1.7)
and suppose that coordinates are chosen so that the components ηαβ are con-
stants, but not necessarily of the form ηαβ = diag[−1, 1, 1, 1]. The connection
is then quadratic in e
Γγ αβ = e Γ1 γ αβ + e2 Γ2 γ αβ ,
where

Γ γ αβ = −( Hk α kγ ),β − ( Hk β kγ ),α + ( Hk α k β ),λ η λγ , (E.1.8)

1
Γ γ αβ = 2H [ H (k̇ α k β + k̇ β k α ) + Ḣk α k β ]kγ ≡ 2Hkγ ( Hk α k β )˙ , (E.1.9)
2

where a “dot” denotes differntiation in the k direction, i.e. f˙ = k( f ) = f ,α kα .

Note that only the indices of k can be raised and lowered with the Minkowski
metric. Hereafter we will use an “index” 0 to denote contraction with k, i.e.

Γ0 αβ = Γγ αβ k γ = e( Hk α k β )˙ , (E.1.10)
Γγ α0 = Γγ αβ k β = −e( Hk α kγ )˙ , (E.1.11)
Γγ 00 = Γγ αβ kα k β = 0 , (E.1.12)
Γ0 α0 = Γγ αβ k β k γ = 0 . (E.1.13)

2290
 E. Exact solutions

The determinant of the full metric is independent of e

| gαβ | = |ηαβ − 2eHk α k β | = |ηαβ | = const. −→ Γ β αβ = 0 ,

and the contracted Riemann tensor therefore reduces to

Rαβ = Rγ αγβ = Γγ αβ,γ − Γγ αδ Γδ βγ . (E.1.14)

The simplest component is

Rαβ kα k β = Γγ αβ,γ kα k β − Γγ δ0 Γδ γ0 = Γγ 00,γ − 2Γγ α0 kα ,γ (E.1.15)

= 2eH ||k̇||2 . (E.1.16)

If the L.H.S. is zero then ||k̇|| = 0 and so k̇ is a null-vector orthogonal to

another null-vector, k. Hence k̇ must be parallel to k and therefore k is a
geodesic vector.

The Ricci tensor expanded as series in e is given by

Rαβ = e R1 αβ + e2 R2 αβ + e3 R3 αβ + e4 R4 αβ . (E.1.17)

The vacuum Einstein’s equations Rαβ = 0 imply that contributions of all or-
ders must vanish. Let us evaluate all such components.

The highest components of the expansion for the Ricci tensor are

R αβ = − Γ ρ ασ Γ σ βρ = 0 , (E.1.18)
4 2 2

R αβ = − Γ ρ ασ Γ σ βρ − Γ ρ ασ Γ σ βρ = 4H 3 ||k̇||2 k α k β . (E.1.19)
3 1 2 2 1

The next component of Rαβ is

R αβ = Γ ρ αβ,ρ − Γ ρ ασ Γ σ βρ (E.1.20)
2 2 1 1

= 2H ( Hk α k β ) + kσ ,σ ( Hk α k β )˙ − H k̇ α k̇ β
¨
 
(E.1.21)
− H 2 Φk α k β − 2Hk (α ψβ) , (E.1.22)
(E.1.23)

where
Φ = 4η γλ η δµ k [λ,δ] k [µ,γ] , ψα = 2k̇γ ( Hk α ),γ . (E.1.24)
Finally, the first component of the Ricci tensor expansion is

R αβ = Γ γ αβ,γ (E.1.25)
1 1
= Ak α k β + 2k (α Bβ) + Xαβ , (E.1.26)

2291
E. Exact solutions

where

A = η λγ H,λγ ,
1
Bβ = −( Hkγ ),γβ + η λγ ( H 2 k β,γ ),λ ,
h H i
Xαβ = −2H (k (α,β) k ),γ + k (α,|γ| kγ ,β) − η λγ k α,γ k β,λ
γ
h i
−2k H,(α k β),γ + H,γ k (α,β)
γ
h i
= −2H k̇ (α,β) + kγ ,γ k (α,β) − η λγ k α,γ k β,λ
−2 Ḣk (α,β) − 2H,(α k̇ β) . (E.1.27)

Therefore, the Ricci tensor turns out to consist of three different contribu-
tions. In paper ([158]) we have shown that third order equations all imply
that k must be geodesic; it must be also shearfree as a consequence of first or-
der equations, whereas the solution for H comes from second order equations
too.

The same treatment can be generalized to include the electromagnetic field,

i.e. to the case of Kerr-Newman. In fact, even in the charged Kerr solution
the congruence of k-lines depend only on the rotation parameter a and not
on the mass M or charge Q. Furthermore, the electromagnetic field is linear
in Q and the metric is linear in M and Q2 , since the function H is obtained
simply by replacing M → M − Q2 /(2r ).

A wider class of solutions could be obtained by modifying the original “lin-

ear” Kerr-Schild ansatz by a “quadratic” one. Suppose that

gαβ = ηαβ − 2λk (α p β) + µk α k β , ||k|| = 0 , k·p = 0, p·p = 1,

where the vector p has to be spacelike. Its inverse is similar,

gαβ = η αβ + 2λk(α p β) + µ0 kα k β , µ + µ 0 = λ2 .

The determinant of this metric is independent of λ and µ, and so the compo-

nents of the Einstein/Riemann tensors are also polynomials in these param-
eters. It is possible that this ansatz can include a black hole interior. Presum-
ably the null vector k would have to be the same as for Kerr(-Newman). This
investigation is left for future works.

2292
 E. Exact solutions

E.2. Rational metrics

We have considered a general metric of the form

∆ dx2 dy2 dΣ2 (dx3 , dx4 )

 
2
ds = + + , (E.2.1)
H V W ∆

where
dΣ2 = − A(dx3 )2 + 2Cdx3 dx4 + B(dx4 )2 . (E.2.2)
The field equations use the following quantities

Q ab,x Q̃ ab ,x ∆2,x ∆,x V,x 2

 
ab
gab,x g ,x = − 2 − + ,
∆2 VW ∆ ∆ V
Q ab,y Q̃ ab ,y ∆2,y ∆,y W,y 2
 
gab,y g ab ,y = − 2 − + ,
∆2 VW ∆ ∆ W
Q ab,x Q̃ ab ,y ∆,y ∆,x ∆,x V,x ∆,y W,y
  
gab,x g ab ,y = − − + + (. E.2.3)
∆2 VW ∆2 ∆ V ∆ W

The functions A, B, C instead have to satisfy second order differential equa-

tions:
AC,x − CA,x AC,y − CA,y
   
+ =0 (E.2.4)
W∆2 ,x V∆2 ,y

with the integrability conditions (or equivalently the equations for the deriva-
tives of H)

A,x B,y + B,x A,y + 2C,x C,y 2∆,x ∆,y V̇ Ẇ V̇ H,y Ẇ H,x
− − − − = 0,
VW∆2 ∆2 VW V H W H
2 2
A,y B,y + C,y V∆2,x − W∆2,y
A,x B,x + C,x V̇ H,x Ẇ H,y
− − + Ẅ − V̈ − + = 0.
W∆ 2 V∆ 2 ∆ 2 H H
This study is still preliminar (see also the section on “The Kerr-Newman so-
lution,” included in this report) but it seems promising in view of obtaining
new exact solutions of Einstein’s field equations for the case of stationary and
axisymmetric spacetimes.

2293
Bibliography
[1] Jantzen R.T., Carini P., Bini D., Annals of Physics 215, (1992), 1–50.

[2] R. Ruffini and C. Sigismondi Eds., Nonlinear gravitodynamics. The

Lense-Thirring effect. A documentary introduction to current re-
search,” World Scientific, Singapore (2003).

[3] D. Bini, P. Carini, R.T. Jantzen, Int. J. Mod. Phys. D 6, (1997), 1–38.

[4] D. Bini, P. Carini, R.T. Jantzen, Int. J. Mod. Phys. D 6, (1997), 143–198.

[5] Bini D., de Felice F. and Geralico A., Class. Quantum Grav. 21, (2004),
5441.

[6] Bini D., de Felice F., Geralico A., Jantzen R. T., Class. Quantum Grav. 22,
(2005), 2947.

[7] Bini D., de Felice F., Geralico A., Jantzen R. T., Class. Quantum Grav. 23,
(2006), 3287.

[8] Teukolsky S.A., Astrophys. J. , 185, (1973), 635 .

[9] Bini D., Cherubini C., Jantzen R.T. and Ruffini R., Progr. Theor. Phys. 107,
(2002), 967.

[10] Bini D., Cherubini C., Jantzen R.T. and Ruffini R. Int. J. Mod. Phys. D 12,
(2003), 1363.

[11] Bini D., Cherubini C, Jantzen R.T., Class. Quantum Grav. 19, (2002), 5265-
5272.

[12] V.B. Braginsky, C.M. Caves, and K.S. Thorne, Phys. Rev. 15 (1977), 2047.

[13] V.B. Braginsky, A.G. Polnarev, and K.S. Thorne, Phys. Rev. Lett. 53
(1984), 863.

[14] K.S. Thorne, R.H. Price, D.A. Macdonald, editors, “Black Holes: The
Membrane Paradigm,” Yale University Press, New Haven, 1986.

[15] K.S. Thorne, in “Quantum Optics, Experimental Gravitation, and Mea-

surement Theory,” (P. Meystre and M.O. Scully, Eds.), Plenum Press,
New York and London, 1981, and in “Near Zero: New Frontiers

2295
Bibliography

of Physics,” (J.D. Fairbank, B.S. Deaver, Jr., C.W. Everitt, and P.F.
Michelson, Eds.), Freeman, New York, 1988.

[16] R.L. Forward, Proceedings of the IRE 49 (1961), 892.

[17] J. Ehlers, Akad. Wiss. Mainz Abh., Math.-Nat. Kl., Nr. 11 (1961), 793.

[18] S.W. Hawking and G.F.R. Ellis, “The Large Scale Structure of Space-
Time,” Cambridge University Press, Cambridge, 1973.

[19] D. Bini and R.T. Jantzen, Proceedings of the XXIII Spanish Rel-
ativity Meeting on Reference Frames and Gravitomagnetism
(http://hades.eis.uva.es/EREs2000), (J.-F. Pascual-Snchez, L.
Flora, A. San Miguel, and F. Vicente, Eds.), World Scientific,
Singapore, 2001.

[20] L.D. Landau and E.M. Lifshitz, “The Classical Theory of Fields,” Per-
magon Press, New York, 1975.

[21] C. Møller, “The Theory of Relativity,” Oxford University Press, Oxford,

First Edition: 1952, Second Edition: 1972.

[22] A.L. Zel’manov, Dokl. Akad. Nauk. USSR 107 (1956), 805 [Eng. trans. in
Sov. Phys. Doklady 1 (1956), 227].

[23] A.L. Zel’manov, Sov. Phys. Dokl. 18 (1973), 231.

[24] C. Cattaneo, Nuovo Cim. 10 (1958), 318; 11 (1958), 733.

[25] C. Cattaneo, Ann. Mat. Pura ed Appl. 48 (1959), 361.

[26] C. Cattaneo, “Calcolo Differenziale Assoluto su una Varietá Riemanni-

ana,” Libreria E.V. Veschi, Rome, (1961).

[27] G. Ferrarese, Rend. di Mat. 22 (1963), 147; Rend. di Mat. 24 (1965), 57.

[28] E. Massa, Gen. Relativ. Grav. 5 (1974), 555; 5 (1974), 573; 5 (1974), 715.

[29] G.F.R. Ellis, in “General Relativity and Cosmology: Proceedings of

Course 47 of the International School of Physics ‘Enrico Fermi’,” (R.
Sachs, Ed.) Academic Press, New York, 1971.

[30] G.F.R. Ellis, in “Càrgese Lectures in Physics,” Vol. 6, (E. Schatzman,

Ed.), Gordon and Breach, New York, 1973.

[31] A. Lichnerowicz, J. Math. Pures Appl. 23 (1944), 37.

[32] Y. Choquet-Bruhat, J. Rat. Mech. Anal. 5 (1956), 951.

2296
 Bibliography

[33] R. Arnowit, S. Deser and C.W. Misner, in “Gravitation: An Introduction

to Current Research,” (L. Witten, Ed.), Wiley, New York, 1962.

[34] C.W. Misner, K.S. Thorne and J.A. Wheeler, “Gravitation,” Freeman,
San Francisco, 1973.

[35] M.A. Abramowicz, B. Carter, and J.P. Lasota, Gen. Relativ. Grav. 20
(1988), 1173.

[36] M.A. Abramowicz, Mon. Not. R. Ast. Soc. 245 (1990), 733.

[37] D. Bini, P. Carini, R.T. Jantzen, Wilkins D., Phys. Rev. D 49, (1994), 2820–
2827.

[38] D. Bini, P. Carini, R.T. Jantzen, Class. Quantum Grav. 12, (1995), 2549–
2563.

[39] D. Bini, R.T. Jantzen, A. Merloni, Il Nuovo Cimento B, 113, (1998), 611–
620.

[40] D. Bini, R.T. Jantzen, A. Merloni, Class. Quantum Grav. 16, (1999), 1333–
1348.

[41] D. Bini, F. de Felice , R.T. Jantzen, Class. Quantum Grav. 16, (1999), 2105–
2124.

[42] D. Bini, R.T. Jantzen, G. Miniutti, Class. Quantum Grav. 18, (2001), 1–13.

[43] D. Bini, C. Cherubini, R.T. Jantzen, G. Miniutti, Class. Quantum Grav.

21, (2004), 1987–1998.

[44] Bini D., Ruffini R., Spoliti G., Il Nuovo Cimento B, 112, (1997), 173-180.

[45] Merloni A., Stella L., Vietri M., Bini D., MNRAS, 304, (1999), 155-159.

[46] Bini D., Gemelli G., Ruffini R., Phys. Rev. D, 61, (2000), 064013.

[47] Bini D., Jantzen R.T., Class. Quantum Grav. 17, (2000), 1-11.

[48] Bini D., de Felice F., Class. Quantum Grav. 17, (2000), 4627-4635.

[49] Bini D., Germani C., Jantzen R.T., Int. J. Mod. Phys. D 10, (2001), 633-647.

[50] Bini D., Jantzen R.T., Mashhoon B., Class. Quantum Grav. 18, (2001), 653-
670.

[51] Sorge F., Bini D., de Felice F., Class. Quantum Grav. 18, (2001), 2945-2958.

[52] Gemelli G., Bini D., de Felice F., Int. J. Mod. Phys. D 11, (2002), 223.

2297
Bibliography

[53] Bini D., Jantzen R.T., Mashhoon B., Class. Quantum Grav. 19, (2002), 17-
37.

[54] Cherubini C., Bini D., Capozziello S., Ruffini R., Int. J. Mod. Phys. D 11,
(2002), 827-841.

[55] Bini D., Jantzen R.T., Miniutti G., Int. J. Mod. Phys. D 11, (2002), 1439-
1450.

[56] Bini D., Cherubini C., Jantzen R.T., Class. Quantum Grav. 19, (2002),
5481-5488.

[57] Bini D., Cherubini C, Jantzen R.T., Mashhoon B., Class. Quantum Grav.
20, (2003), 457-468.

[58] Bini D, Cruciani G., Lunari A., Class. Quantum Grav. 20, (2003), 341-350.

[59] Bini D., Cherubini C., Jantzen R.T., Mashhoon B.,

Phys. Rev. D, 67, (2003), 084013.

[60] Bini D., Jantzen R.T., Special Observers in the Kerr spacetime, In “Ad-
vances in General Relativity and Cosmology”, Ed. G. Ferrarese,
Pitagora, pag. 287-296, 2003.

[61] Bini D., Jantzen R.T., Gravitomagnetic clock effects in black hole space-
time, In “General Relativity, Cosmology and Gravitational Lens-
ing”, Napoli Series on Physics and Astrophysics, pp. 17-28 Eds. G.
Marmo, C. Rubano, P. Scudellaro, Bibliopolis, 2002.

[62] Bini D., Crosta M. T., de Felice F., Class. Quantum Grav. 20, (2003), 4695-
4706.

[63] Bini D., de Felice F., Class. Quantum Grav. 20, (2003), 2251-2259.

[64] Bini D., Cherubini C., De Mattia M., Jantzen R.T., Gen. Rel. Grav. 35,
(2003), 2249-2260.

[65] Bini D., Cherubini C., Cruciani G., Lunari A., Int. J. Mod. Phys. D 12,
(2003), 1983-2000.

[66] Bini D., Il Nuovo Cimento B, 118, (2003), 1055-1061.

[67] Bini D., Jantzen R.T., Inertial forces: the special relativistic assessment,
in “Relativity in Rotating Frames (Relativistic physics in rotating
reference frames) Ed. by G. Rizzi and M.L. Ruggiero Kluwer Aca-
demic Press, Fundamental Theories of Physics, vol. 135, 221-239,
2004.

[68] Bini D, Geralico A., Int. J. Mod. Phys. D 13, (2004), 983-1003.

2298
 Bibliography

[69] Bini D., Cherubini C., Class. Quantum Grav. 21, (2004), 4523-4529.

[70] Bini D., Cherubini C., Mashhoon B., Phys. Rev. D, 70, (2004), 044020-7.

[71] Bini D., Cherubini C., Mashhoon B., Class. Quantum Grav. 21, (2004),
3893-3908.

[72] Cherubini C., Bini D., Bruni M., Perjes Z., Class. Quantum Grav. 21,
(2004), 4833-4843.

[73] Bini D., Cherubini C., Cruciani G., Jantzen R.T., Int. J. Mod. Phys. D 13,
(2004), 1771-1803.

[74] Bini D., Jantzen R.T., Il Nuovo Cimento B 119, (2004), 863-873.

[75] Bini D., Il Nuovo Cimento B 120, (2005), 1021-1025..

[76] Cherubini C., Bini D., Bruni M., Perjes Z., Class. Quantum Grav. 22,
(2005), 1763-1768.

[77] Bini D., Lusanna L., Mashhoon B., Int. J. Mod. Phys. D 14, (2005), 1413-
1429.

[78] Bini D., De Paolis F., Geralico A., Ingrosso G., Nucita A. Gen. Rel. Grav.
37, (2005), 1263-1276.

[79] Bini D., Geralico A., Jantzen R. T., Class. Quantum Grav. 22, (2005), 4729-
4742.

[80] Bini D., Cherubini C., Filippi S., Geralico A., Class. Quantum Grav. 22,
(2005), 5157-5168.

[81] Bini D., Geralico A., Jantzen R.T. Class. Quantum Grav. 23, (2006), 3963-
3981.

[82] Bini D., de Felice F., Geralico A. Class. Quantum Grav. 23, (2006), 7603-
7626.

[83] Bini D., de Felice F., Geralico A. Phys. Rev. D, 76, (2007), 047502.

[84] Papapetrou A., Proc. Roy. Soc. London 209, (1951), 248.

[85] Mathisson M., Acta Phys. Polonica 6, (1937), 167 .

[86] Corinaldesi E. and Papapetrou A., Proc. Roy. Soc. London 209, (1951),
259.

[87] Pirani F., Acta Phys. Polon. 15, (1956), 389.

[88] Tulczyjew W., Acta Phys. Polon. 18, (1959), 393.

2299
Bibliography

[89] Dixon W. G., Il Nuovo Cimento 34, (1964), 318.

[90] Dixon W. G., Proc. Roy. Soc. London A 314, (1970), 499.

[91] Dixon W. G., Proc. Roy. Soc. London A 319, (1970), 509.

[92] Dixon W. G., Gen. Rel. Grav. 4, (1973), 199.

[93] Dixon W. G., Philos. Trans. R. Soc. London. Ser. A 277, (1974), 59.

[94] Taub A. H., J. Math. Phys. 5, (1964), 112.

[95] Mashhoon B., J. Math. Phys. 12, (1971), 1075.

[96] Mashhoon B., Ann. of Phys. 89, (1975), 254.

[97] Ehlers J. and Rudolph E., Gen. Rel. Grav. 8, (1977), 197.

[98] Tod K. P., de Felice F. and Calvani M., Il Nuovo Cimento B 34, (1976), 365.

[99] Souriau J. M., C.R. Acad. Sc. Paris 271, (1970), 751.

[100] Souriau J. M., C.R. Acad. Sc. Paris 271, (1970), 1086.

[101] Souriau J. M., C.R. Acad. Sc. Paris 274, (1972), 1082.

[102] Souriau J. M., Ann. Inst. H. Poincaré 20, (1974), 22.

[103] Bargman V., Michel L. and Telegdi V., Phys. Rev. Lett. 2, (1959), 435.

[104] Bini D., de Felice F. and Geralico A., Class. Quantum Grav. 21, (2004),
5427.

[105] Bini D., de Felice F., Geralico A., Int. Jour. Mod. Phys. D 14, (2005), 1793.

[106] Bini D., de Felice F., Geralico A., Lunari A., J. Phys. A: Math. Gen. 38,
(2005), 1163.

[107] Bini D., Cherubini C., Geralico A. and Mashhoon B., Class. Quantum
Grav. 22, (2005), 709.

[108] Mashhoon B., Granwald F., Lichtnegger H. I. M., Lectures Notes in

Physics 562, (2001), 83.

[109] Cohen J. M. and Mashhoon B., Phys. Lett. A 181, (1993), 353.

[110] Ciufolini I. and Ricci F., Class. Quant. Grav. 19, (2002), 3863.

[111] Ciufolini I. and Ricci F., Class. Quant. Grav. 19, (2002), 3875.

2300
 Bibliography

[112] Bini D., Cherubini C., Geralico A., Jantzen R. T., Int. Jour. Mod. Phys. D
15, (2006), 737.
[113] Bini D., Fortini P., Geralico A., Ortolan A., Class. and Quantum Grav. 25,
(2008), 035005.
[114] Bini D., Fortini P., Geralico A., Ortolan A., Class. and Quantum Grav. 25,
(2008), 125007.
[115] Newman E. Penrose R. J. Math. Phys. 3, (1962), 566.
[116] Geroch R., Held A. and Penrose R. J. Math. Phys. 14 (1973), 874.
[117] Bose K. S. J. Math. Phys. 16, (1975), 772.
[118] Kinnersley W. J. Math. Phys. 10, (1969), 1195.
[119] Regge T. Wheeler J. A Phys. Rev. 108, (1957), 1063.
[120] Zerilli F. J. Phys. Rev. D 9, (1974), 860.
[121] Ruffini R. in Physics and Astrophysics of Neutron Stars and Black
Holes North-Holland, Amsterdam edited by Giacconi R. and
Ruffini R. (1978).
[122] Chrzanowsky P. L., Phys. Rev. D 13, (1976), 806.
[123] Lee C. H., Report IC/77/121 of the International Center for Theoretical
Physics, Miramare, Trieste , (1977).
[124] Lee C. H., J. Math. Phys 17, (1976), 1226.
[125] Press W. H. Teukolsky S., Astrophys. J. 185, (1973), 649.
[126] Chandrasekhar S., The Mathematical Theory of Black Holes Clarendon
Press, Oxford, (1983).
[127] Fackerell E. D., Proceedings of the Second Marcel Grossman Meeting
on General Relativity (1982), Ed. Ruffini R., North-Holland, Ams-
terdam.
[128] Bellezza V. Ferrari V., J. Math. Phys. 25, (1984), 1985.
[129] Cherubini C. and Ruffini R., Il Nuovo Cimento B 115, (2000), 699.
[130] Bini D., Cherubini C. and Jantzen R.T., Class. Quant. Grav. 19, (2002),
5265.
[131] D. Kramer, H. Stephani, E. Hertl, and M. McCallum, Exact solutions
of Einstein’s field equations Cambridge University Press, Cambridge,
(1979).

2301
Bibliography

[132] E. Fermi, Il Nuovo Cimento 22, (1921), 176.

[133] E. T. Whittaker, Proc. Roy. Soc. A116, (1927), 720.

[134] E. T. Copson, Proc. Roy. Soc. A118, (1928), 184.

[135] R. Hanni, Junior Paper submitted to the Physics Department of Prince-

ton University, (1970), unpublished.

[136] J. Cohen and R. Wald, J. Math. Phys. 12, (1971), 1845.

[137] R. Hanni and R. Ruffini, Phys. Rev. D 8, (1973), 3259.

[138] J. Bičák and L. Dvořák, Gen. Relativ. Grav. 7, (1976), 959.

[139] B. Linet, J. Phys. A: Math. Gen. 9, (1976), 1081.

[140] B. Leaute and B. Linet, Phys. Lett. 58A, (1976), 5.

[141] D. Bini, A. Geralico, and R. Ruffini, Phys. Lett. A360, (2007), 515.

[142] D. Bini, A. Geralico, and R. Ruffini, Phys. Rev. D 75, (2007), 044012.

[143] D. Bini, A. Geralico, and R. Ruffini, Phys. Rev. D 77, (2008), 064020.

[144] M. Johnston, R. Ruffini, and F. J. Zerilli, Phys. Lett. B49, (1974), 185.

[145] M. Johnston, R. Ruffini, and F. J. Zerilli, Phys. Rev. Lett. 31, (1973), 1317.

[146] W. B. Bonnor, Class. Quant. Grav. 10, (1993), 2077.

[147] S. M. Majumdar, Phys. Rev. 72, (1947), 390.

[148] A. Papapetrou, Proc. R. Irish Acad. 51, (1947), 191.

[149] G. A. Alekseev and V. A. Belinski, Phys. Rev. D 76, (2007), 021501(R).

[150] G. A. Alekseev, Sov. Phys. Dokl. 30, (1985), 565; Proc. Steklov Inst. Math.
3, (1988), 215.

[151] D. Bini, C. Germani, and R. T. Jantzen, Int. J. Mod. Phys. D 10, (2001),
633.

[152] R. P. Kerr and A. Schild, “A new class of vacuum solutions of the

Einstein field equations,” in Atti del convegno sulla relatività gen-
erale; problemi dell’energia e onde gravitationali, G. Barbera, Ed., p. 173
(Firenze, 1965).

[153] R. P. Kerr and A. Schild, Proc. Symp. Appl. Math. 17, 199 (1965).

[154] R. P. Kerr, Phys. Rev. Lett. 11, 237 (1963).

2302
 Bibliography

[155] Goldberg J.N. and Sachs R.K., Acta Phys. Polon. 22 Suppl., 13 (1962).

[156] Bini D., Jantzen R. T., Stella L., Classical and Quantum Gravity, 26,
055009 (2009).

[157] De Felice F., Bini D., Classical Measurements in Curved Space-Times

Series: Cambridge Monographs on Mathematical Physics, (Cam-
bridge Univ. Press, Cambridge, 2010).

[158] Bini D., Geralico A., Kerr R.P., International Journal of Geometric Meth-
ods in Modern Physics (IJGMMP) 7, 693-703 (2010).

[159] Bini D., Geralico A., Classical and Quantum Gravity, 27, 185014 (2010).

[160] Bini D. , Geralico A., Jantzen R.T., General Relativity and Gravitation,
vol. 43, 18371853, 2011.

[161] Bini D., Geralico A., Jantzen R. T., General Relativity and Gravitation,
vol. 43, 959-975, 2011.

[162] Bini D., Geralico A., Jantzen R. T., Semeřák O. and Stella L., Classical
and Quantum Gravity, vol. 28 035008 (21pp), 2011.

[163] Bini D. , de Felice F., Geralico A., Classical and Quantum Gravity, vol.
28 225012, 2011.

[164] Bini D., Esposito G., Geralico A., General Relativity and Gravitation, to
appear, 2011.

[165] Bini D., Geralico A., Jantzen R. T. and Semeřák O. Classical and Quan-
tum Gravity, to appear, 2011.

[166] Bini D. and Geralico A. Physical Review D, to appear, 2011.

[167] Bini D., Geralico A., Jantzen R. T. General Relativity and Gravitation,
to appear, 2011.

[168] Bini D., Fortini P., Haney M., Ortolan A. Classical and Quantum Grav-
ity, vol. 28, 235007, 2011.

[169] Wen-Biao Han and Zhoujian Cao, Physical Review D, vol. 84, 044014,
2011.

2303