Black Holes: from Stars to Galaxies – across the Range of Masses

Proceedings IAU Symposium No. 238, 2006 c 2006 International Astronomical Union

V. Karas & G. Matt DOI: 00.0000/X000000000000000X

Black holes: from stars to galaxies

I. Félix Mirabel †
European Southern Observatory, Alonso de Cordova 3107, Santiago, Chile
email: [email protected]

Abstract. While until recently they were often considered as exotic objects of dubious existence,
in the last decades there have been overwhelming observational evidences for the presence of
stellar mass black holes in binary systems, supermassive black holes at the centers of galaxies,
and possibly, intermediate-mass black holes observed as ultraluminous X-ray sources in nearby
galaxies. Black holes are now widely accepted as real physical entities that play an important
arXiv:astro-ph/0612188 v1 7 Dec 2006

role in several areas of modern astrophysics.

Here I review the concluding remarks of the IAU Sympposium No 238 on Black Holes, with
particular emphasis on the topical questions in this area of research.
Keywords. Black hole physics – galaxies: nuclei – galaxies: jets – stars: general

1. The aim of IAU Symposium No 238

The interaction of black holes with their surroundings produces analogous phenom-
ena in AGN and stellar black hole binaries. The scales of length and time of the phe-
nomena are proportional to the mass of the black hole, and the whole phenomenological
diversity that takes place around black holes can be described by the same physical
concepts, however, the observed phenomena exhibit enormous complexity (see figure 1).
Quasars and microquasars can eject matter several times, whereas collapsars form jets
only once. When the jets point to the Earth these objects appear as microblazars, blazars
and gamma-ray bursts, respectively. Synergy between research on stellar mass and su-
permassive black holes has become essential for our understanding of the underlying
physics.
With this in mind, the aim of IAU Symposium No 238 has been to discuss the
relations between different types of astrophysical black holes in a broader evolutionary
context, bringing together astronomers working on AGN with those working on compact
stellar binaries. Several groups that actively search for the elusive intermediate-mass
black holes have had an active participation.

2. Stellar black holes

Current physics suggests that compact objects in stellar binaries with mass functions
larger than 4 solar masses must be black holes. There are about 20 known objects with
such mass functions. They are believed to be the tip of an iceberg, since it is estimated
that in the Milky Way alone there should be at least 1000 dormant black hole X-ray
transients, while the total number of stellar-mass black holes (isolated and in binaries)
could be as large as 100 million. The number statistics of known stellar black holes is
still very small and at present remain open the following questions:
Is the gap in the black hole mass function between 2.2 and 4.0 solar masses real? If
so, which is the physical reason? Why in the Milky Way it has not been found a stellar
† On leave from CEA, France.
1
2 I. F. Mirabel

Figure 1. The same physical mechanism can be responsible for the three different types of
objects: microquasars, quasars/AGN and massive stars that collapse (“collapsars”) to form a
black hole producing Gamma-Ray-Bursts. Each one of these objects contains a black hole, an
accretion disk and relativistic particles jets. Credit: Mirabel & Rodrı́guez (2002).

black hole with mass larger than 14 solar masses? Is this due to poor statistics or to large
mass losses in stellar winds by the metal rich progenitors in the Milky Way?
Kinematic studies of black hole binaries suggest that some stellar black holes form
associated to very energetic supernova explosions (see figure 2). In some cases, this kine-
matic evidence is reinforced by the chemical composition of the donor star, when it
contains elements produced in supernova explosions. However, kinematic studies suggest
that black holes may also form by direct collapse, namely, in the dark (figure 3). There-
fore, it is an open question when stellar black holes form in energetic supernovae and
when by direct core collapse; more specifically, whether the presence of an energetic ex-
plosion depends or not on the mass of the collapsing stellar core. In fact, the kinematics
of the microquasars Cygnus X-1 and GRS 1915+105, which contain black holes of ∼ 10
and ∼ 14 solar masses, respectivelly, did not receive strong kicks from natal energetic
supernova explosions.
This question on the explosive or implosive black hole formation can be approached
by observations of nearby gamma-ray bursts of long duration, which are believed to take
place when stellar black holes are form by core collapse of massive stars. Therefore, the
question on whether all collapsar GRBs are associated with supernovae of type Ia/b,
or some are not, is of topical interest for the understanding of the very last phases of
massive stellar evolution and black hole formation.
 Black holes: from stars to galaxies 3

Figure 2. Galactic trajectory of a black hole wandering in the Galactic halo. This black hole
may have been shoot out from the Galactic plane by an energetic, natal supernova explosion.
Credit: Mirabel et al (2001).

Figure 3. Kinematic evidence for the direct for-

mation of the black hole in Cygnus X-1. Optical
image of the sky around the black hole binary
Cyg X-1 and the parent association of massive
stars Cyg OB3. The red arrow shows the motion
in the sky of the radiocounterpart of Cyg X-1
for the past 0.5 millon years. The yellow arrow
shows the average Hipparcos motion of the mas-
sive stars of Cyg OB3 (circled in yellow) for the
past 0.5 millon years. Cyg X-1 moves along with
the parent association of massive stars indicat-
ing that the compact object did not received an
energetic kick from a natal supernova. Credit:
Mirabel & Rodrı́gues (2003).

3. The supermassive black hole at the Galactic Center

Dynamics is the most direct method to determine the mass of astrophysical compact
objects, and therefore, the best evidence for the existence of a black hole. The first
unambigous dynamic evidence for a supermassive black hole was found following the
motion of water masers around the center of the galaxy NGC 4258.
More robust evidence has been obtained by the motion of stars around the dormant
black hole of 3-4 million solar masses at the center of our Galaxy. These stars seem to be
distributed in two randomly inclined disks of 0.04 pc and 0.5 pc radii. The unexpected
discovery of a compact cluster of massive stars in the central parsec of our Galaxy (see
figure 4) poses new questions and may open new horizons for our understanding of
massive black hole formation and its relation with massive star formation. This central
cluster exhibits a flat mass function and it is only 6 million years old. At present, it is not
clear how it got there. Perhaps it was formed in situ. How this could take place under
the strong tidal forces from the central black hole remains a mystery.

4. Supermassive black holes in external galaxies

Supermassive black holes are ubiquitous at the centers of galaxies. Their mass is
correlated with the mass of the host galaxy, and in particular with that of the stellar
bulge. This indicates that massive black hole and host galaxy formation are tided up.
The most massive black holes are found at z > 6.0 implying that they assembled very
early, in less than one billon years after the Big Bang. On the other hand, supermassive
black holes of lower mass have formed more slowly by merging at z 6 3. The peak
4 I. F. Mirabel

Figure 4. Left: The K-band near-infrared adaptive optics image of the central 20 arcsec of the
Milky Way. Right: HKL colour composite image. Credit: Reinhard Genzel et al (2003).

accretion rate into black holes occurred rather late at z ∼ 0.7. Interestingly, this redshift
is similar to the redshift of the peak density of luminous infrared galaxies at z ∼ 0.8,
which suggests the association of maximum rate of accretion into black holes with dust
enshrouded starbursts in luminous infrared galaxies, which are known to be mergers of
gas-rich galaxies.
Although several models have been proposed, there is no general concensus on how
supermassive black holes were form. More specifically, it is not known how could the most
massive black holes in the universe form so rapidly with no feedback. Did supermassive
black holes form before, after or coevally with the bulges?
The kick velocity due to gravitational recoil in merging supermassive black holes
may displace the merged black hole from the dynamic center of the host galaxies. For
dwarf galaxies the estimated kick velocity is larger than the typical escape velocities
of 10-20 km/s, which may result in the ejection to the intergalactic medium of naked,
massive black holes. How could we identify these runway black holes?

5. Black holes of intermediate mass

The existence of black holes of intermediate mass remains an open question. From
the spectral properties of ultraluminous X-ray sources in nearby galaxies it has been
suggested that some of these sources contain black holes of hundreds and perhaps thou-
sands of solar masses. However, no dynamic evidence from the motion of satellite objects
around such objects has been found. Furthermore, the luminosity function of black hole
binaries as well as the properties of the most luminous x-ray black hole binaries in our
Galaxy indicate that most of the super Eddington sources observed in nearby galax-
ies are a natural extension of the luminosity function of X-ray binaries, most of them
being black holes with masses smaller than 100 solar. It has been argued that the hyper-
accreting Galactic sources SS 433 and GRS 1915+105 seen from a different angle would
be classified as ultraluminous X-ray sources.
If some ultraluminous x-ray sources are black holes of intermediate mass, one may
ask why none has been identified in our Galaxy and/or the Magellanic Clouds, where the
 Black holes: from stars to galaxies 5
mass functions of compact objects can be determined. Black holes of intermediate mass
may exist but are difficult to find.

6. Correlations among black holes of all masses

a) The fundamental plane: At low accretion rates, correlations between radio and
x-ray luminosities are found for black holes across the whole range of masses. At low
rates of accretion most the radiation power is dominated by synchrotron, compact, flat
spectrum jets, from the x-rays to radio waves. For black holes in the jet dominated state,
universal correlations between the x-ray luminosity, radio luminosity, and black hole mass
have been found. Using this black hole fundamental plane it is expected that the masses
of black holes could be determined from the x-ray and radio luminosities.
b) Accretion–jet coupling: Because of their proximity and rapid variability, micro-
quasars have become the most adequate objects to study the connection between in-
stabilities in the accretion disks and the genesis of relativistic jets. Figure 5 shows this
connection as observed in an interval of time of a few tens of minutes. This sequence has
also been observed in quasars, but on scales of a few years insted of tens of minutes.

Figure 5. Real
time observations
in a microquasar
of the connection
between instabili-
ties in the accretion
disk and the genesis
of jets in time
scales of tens of
minutes. Analogous
disk–jet coupling
has been observed
in quasars, but on
time scales of years.
Credit: Mirabel
et al (1998).

c) Massive outflows: Across all mass scales, black holes in high luminous states
exhibit X-ray absorption lines that reveal sub-relativistic winds with mass outflows as
large as 0.3 times the accretion mass. Super-Eddington accretion makes massive outflows
inevitable because all the accreting energy cannot be radiated.
In the case of black holes binaries the joint action of jets and massive outflows
heat and blow away the interstellar medium, as does the black hole binary SS433 in
the host nebula W50. X-ray observations with Chandra and XMM-Newton show that
AGN produce analogous impact on the intergalactic medium, which has solved the long
standing cooling flow paradox.
In this conference it was proposed that massive outflows may result from the inter-
action of relativistic jets with warped accretion disks, in both stellar and supermassive
6 I. F. Mirabel
accreting black holes. Furthermore, it was proposed that bended disks may explain sev-
eral intriguing observational results, such as the large numbers of obscured AGN, as well
as the re-direction of precessing jets, as in SS 433. The following questions remain open:
Are the relativistic jets in AGN and microquasars purely leptonic or hadronic? Are
the hadrons in the jets of SS433 a result of entrainment by the relativistic jets in the
massive winds? Is the nebula W50 that hosts SS433 the remnant of the natal supernova
of the compact object, or a blown away cavity by the super-winds, or both?
d) Quasi-periodic oscillations: Sgr A* accretes at low mass rates, and quasi-periodic,
polarized flares on scales of tens of minutes have now been observed at X-rays, infrared,
submillimeter and radio waves. At longer waves the flares are polarized up to 10%.
From the time lag at longer wavelengths it has been proposed that these flares could
be synchrotron self Compton emission from adiabatically expanding clouds. It is unclear
whether these expanding plasma clouds in Sgr A* are rotating blobs in the accretion
disk, or whether they are collimated expanding jets as observed in microquasars.
e) Spin: Black holes are the simplest objects in the universe. They are defined by
only three parameters: the mass, the spin and the electric charge. Because much of the
radiation emerges within 6 gravitational radii they provide a unique opportunity to probe
gravity in the strong-field regime. The radius of the ultimate stable orbit depends from
the spin, and knowing the mass, distance and inclination, it can be derived the spin by
measuring the X-ray flux and the temperature of the accretion disk when the accretion
disk is in the thermally dominated state and at luminosities 6 0.3 Eddington.
Another way to measure the spin is with skewed fluorescence iron lines. Using both
methods it has been claimed that some microquasars and AGN host extreme Kerr black
holes. The following questions remain unanswered: Would magnetic fields change com-
pletely the physical conditions in the inner parts of accretion disks? Is there a general
correlation between spin and jet power?

7. Historical and epistemological analogy between stellar

astrophysics and black hole astrophysics
At present, black hole astrophysics is in an analogous situation as was stellar as-
trophysics in the first decades of the XX century. At that time, well before reaching the
physical understanding of the interior of stars and the way by which they produce and
radiate their energy, empirical correlations such as the HR diagram were found and used
to derive fundamental properties of the stars, such as their mass. Analogous approaches
are taking place in black hole astrophysics. Using correlations among observables such
as the radiated fluxes in x-rays and radio waves, quasi-periodic oscillations, flickering
frequencies, etc, fundamental parameters that describe astrophysical black holes such as
the mass and spin of the black holes are being derived.

References
Genzel R., Schödel R., Ott T. et al 2003, ApJ, 594, 812
Mirabel I. F., Dhawan V., Chaty S. et al 1998, A&A, 330, L9
Mirabel I. F., Dhawan V., Mignani R. P., Rodrigues I., Guglielmetti F. 2001, Nature, 413, 139
Mirabel I. F., Rodrı́guez L. F. 2002, Sky & Telescope, 103 (May), 32
Mirabel I. F., Rodrı́guez L. F. 2003, Science, 300. 1119
 Black holes: from stars to galaxies 7
Phil Charles: Could the apparent “dark jet” paradox of SS433 be answered by the
high inclination which causes obscuration of most of the flux?

Felix Mirabel: Yes, the accretion disc produces high opacity to the X-rays and, as
shown in the talk by Andrew King, the disc is probably highly warped. The source could
be very bright if seen from a different direction.

Virginia Trimble: Suppose there have never been those jets in SS433, would there
still be a supernova remnant there, or is the host nebula blown mainly by the activity of
central source?

Felix Mirabel: This is an open question. Clearly, the lateral extentions seen in the
nebula W 50 that hosts SS433, have been blown away by the jets.

Gregory Beskin: Is it possible to create jets without a black hole in the centre of the
disc, for example by a neutron star?

Felix Mirabel: Yes indeed, there are jets in binaries with confirmed neutron stars.
Clear cases are Scorpius X-1 and Circinus X-1.

Gloria Dubner: Comment on the previous question by V. Trimble: There is a way

to probe if the whole bubble was created by a compact object. I believe that the whole
nebula had been created by a supernova remnant and its shape was then distorted by the
action of SS433’s jets. The way to address this issue is by searching for spectral changes
in the radio emission – the jets have different spectrum than the rest of the bubble.