Active Galactic Nuclei (AGN)
A little history - First Detections of Seyfert Galaxies
1908 – Fath & Slipher detect strong emission lines similar to planetary
nebulae with line-width of several 1000 km/s in NGC1068.
1913 – Detection of an optical
jet in M87 by Curtis
jet
AGN
A little history - First Detections of Optical Jets
A little history - emission of the Galactic center
Karl Jansky
Cygnus A (6cm Carilli NRAO/AUI)
Jets can cover several hundred kiloparsecs to a couple of
megaparsecs (remember the Milky Way has a diameter of several
10s of kiloparsecs).
1953 - discovery of the large linear radio structure in Cygnus A
centered around the central brightest galaxy of a cluster
Core
Lobes
A little history - the puzzle of Cygnus A
First Radio Surveys
Early radio surveys played a crucial role in discovering quasars
• 3C and 3CRThird Cambridge Catalog (Edge et al. 1959) at 159 Mhz (>9Jy).
Basis for extragalactic radio astronomy, cosmology and discovery of Quasars
• PKS Parkes (Australia, Ekers 1959) survey of southern sky at 408 Mhz (>4Jy)
and 1410MHz (>1Jy).
• 4C 4th Cambridge survey (today 8C). Deeper/smaller
• AO Aricibo Occultation Survey (Hazard et al. 1967). Occultation by moon
(high positional accuracy)
Discovery of Quasars
3C273
The 273rd radio source
in the Cambridge Catalog
Compact radio source looks
like a star except for that
wisp of light!
Maarten Schmidt measured a
redshift of z=0.158
Early attempts to find galaxies
associated with quasars failed
Broad emission lines at “strange” positions
Discovery of Quasars
Maarten Schmidt
Discovery of Quasars
Discovery of Quasars
• 1964 – Schmidt studied sufficient quasars to find:
• Star-like, associated with radio sources
• Time-variable in continuum flux
• Large UV fluxes
• Broad emission lines
• Large redshifts
Not all quasars have all these properties
Discovery of Quasars
Why is this object at redshift of 0.158 so special?
z = Δλ/λ0, then it follows that d = cz/H0 = ~470 h0
-1 Mpc
m – M = 5 log(d/Mpc) + 25
For B =13.1th magitude => MB=-23.3 + 5 log(h0
-1)
(The Milk Way is ~ -19.7 =>
3C273 is 2.5123.6 ~ 30 times brighter)
QuasarVariability
• Quasars are variable in every
waveband and emission lines
•Variability time-scale can be days
to months
• Hence size of emission regions is
light-days to light-months (an
object cannot vary in brightness
faster than it takes light to cross
the object).
• a variability of an hour implies a
size of l<cΔt ~ 7.2 AU (less than
Saturn’s distance from the Sun)
Twinkle, twinkle, quasi-star,
Biggest puzzle from afar.
How unlike the other ones,
Brighter than a trillion Suns.
Twinkle, twinkle, quasi-star,
How I wonder what you are!
George Gamow
A more detailed view of the jet of the nearest quasar
Quasar/QSR = Quasi Stellar Radio-source,
QSO = Quasi-Stellar Object
•Very luminous compact (1045–1049 erg/s) centers of galaxies,
which outshine their host (typical field galaxy 1044erg/s)
• Highly variable across the electromagnetic spectrum
• Optical spectra similar to much fainter Seyfert 1 nuclei, with the
exception that the narrow lines are generally weaker.
Two varieties:
• Radio-loud QSOs (Quasars or RL QSOs)
• Radio-quiet QSOs (or RQ QSOs)
Transitions at P5GHz≈1024.7 W Hz−1 sr–1 / RL QSOs are 5−10% of
the total of QSOs.
AGN taxonomy: Quasars & QSOs
There is a large gap in radio power between RL and RQ varieties
of QSOs (Kellerman et al. 1989, Miller et al. 1990)
(Miller et al. 1990)
P5GHz≈1024.7 W Hz−1 .
Quasars & QSOs
Falcke, Sherwood, Patnaik (1996)
R=radio/optical flux
Radio loud
Radio quiet
Some examples of QSOs
QSOs often outshine their
host galaxies which can be
difficult to detect!
Quasars host-galaxies
often show interactions
Redshift Distribution of Quasars
SDSS
•The quasar redshift distribution
seems to peak around z~2.
•This is not only a selection
effect, but seems real, even after
bias corrections.
•This could be related to the
formation of galaxies and LSS and
the star-formation history.
Broad band spectral energy distribution of 3C273
Contribution
of the galaxy
Contribution
of the jet
•The flux of 3C273 per logarithmic
interval is constant over more than 10
orders of magnitude in frequency!
Big blue bump
Quasar Composite Optical/UV Spectrum
Broad/narrow emission lines
UV excess
QUASARS
• Quasars are very luminous compact (1045–1049 erg/s) centers of
galaxies, which outshine their host (typical
fi
eld galaxy 1044erg/s)
• Highly variable across the electromagnetic spectrum
• Out of 104 known quasars only about 10% are radio loud
BAL QSOs = Broad Absorption Line QSOs
Otherwise normal QSOs that show deep blue-shifted absorption
lines corresponding to resonance lines of C IV, Si IV, NV.
All of them are at z ≥ 1.5 because the
phenomenon is observed in the restframe
UV.At these redshifts, they are
about 10% of the observed population.
BAL QSOs tend to be more polarized
than non-BAL QSOs.
BAL QSO
BL Lac: Is the prototype of its class, an object, stellar in appearance, with
very weak emission lines and variable, intense and highly polarized
continuum.The weak lines often just appear in the most quiescent
stages.
Blazars: Encompass BL Lacs and optically violent-variable (OVV) QSOs.
These are believed to be objects with a strong relativistically beamed jet
in the line of sight.They exhibit the most rapid and largest amplitude
variability.
BL Lac objects and blazars
(Vermeulen et al. 1994)
AGN taxonomy: BL Lac
“Re-discovery” of Seyfert Galaxies
1943 – Seyfert finds multiple galaxies similar to NGC1068
(Hence since then they are called by his name)
1959 – Woltjer draws several important conclusions on
Seyfert galaxies:
* Nuclei are unresolved (<100pc)
* Nuclear emission last for >108 years
(1/100th spirals is a Seyfert and the Universe is 1010 yrs)
* Nuclear mass is very high if emission-line broadening
is caused by bound material (M~v2r/G~109±1 Msun)
Seyfert 1 spectrum
Classification based on the width of the optical emission lines
• Sy 1: broad permitted emission lines (Hα, He II, ... ), of
FWHM ≤ 104 km s−1 that originate in a high-density
medium (ne ≥ 109 cm−3)
•Sy 2: narrow forbidden emission lines ([OIII], [N II], …) that
originate in a low-density medium (ne ≈ 103−106 cm−3)
of FWHM ≤ few x 100 km s−1
• Sy1.x (1.9, 1.8, ...): increase with the width Hα and Hβ lines.
Classification of Seyfert galaxies
The classification for a single
object can change with time,
due to AGN variability!
AGN taxonomy: Seyfert galaxies
Radio-galaxies
• FRI • FRII
NRAO/AUI
NRAO/AUI
Classes of Radio-Galaxies
Classes of Radio-Galaxies
Large radio-galaxies with lobes can be
divided in two types Fanaroff-Riley (1974):
•FR-I
-Weaker radio sources that are bright in the center
and fainter toward the edges (limb-darkened)
-smooth, continuous turbulent double-sided jets
-steepest spectrum emission in outer region
•FR-II
-Radio structure with a faint core and bright endpoints
(limb-brightened)
-steepest spectrum emission in inner region
-invariably have high luminosities, P>1042 erg/s
• Morphological classi
fi
cation
- Lobe dominated
✴ FR I
✴ FR II
Jet-power - black hole spin(?), ISM/IGM/ICM density?
- Core Dominated
Doppler boosting and/or age
- Peculiar
✴ Tailed radio sources
Interaction with ambient cluster medium
• Optical spectra
- Narrow-line radio galaxies
- Broad-line radio galaxies (Seyfert I and QSO)
- Feature less (link to blazars)
Observed properties dependent on viewing direction due to a torus of dust
Radio classification
AGN-1: HR-2007
p. 41
3. Tailed radio sources: Morphology due to movement galaxy through (cluster) gas
VLA observations of NGC 1265:
O’dea and Owen Astrophysical Journal, vol. 301, Feb. 15, 1986, p. 841-859.
• Morphological classi
fi
cation
- Lobe dominated
✴ FR I
✴ FR II
Jet-power - black hole spin(?)
- Core Dominated
Doppler boosting and/or age
- Peculiar
✴ Tailed radio sources
Interaction with ambient cluster medium
• Optical spectra
- Narrow-line radio galaxies
- Broad-line radio galaxies (Seyfert I and QSO)
- Feature less (link to blazars)
Observed properties dependent on viewing direction due to a torus of dust
Radio classification
Some signs of AGN Activity
•Compact (~3pc) luminous centers
•Spectra with strong emission lines
•Strong Non-Thermal Emission
•Strong UV emission from a compact region in the center
•HighVariability on time-scales of days to months
•Compact Radio Core
•Extended radio structures (jets, lobes, hot spots)
•Strongly Doppler-broadened emission lines
•X-ray, γ-ray andTeV-emission
(Not all AGN show each of these, but often several of them)
• AGN show emission not easily attributable to stars
• AGN occur both in spirals and E/S0's (Seyferts/Quasars, distinguished
mostly in the amount of energy emitted)
• AGN emit energy comparable or larger than all the stars in the hostgalaxy,
over a wide range of frequencies (including sometimes the
radio).
• AGN show strong broad emission lines. Combined with the small
emission region this indicates a high central concentration of mass.
• AGN are often highly variable (supporting the small region from which
the emission emanates).
• AGN can show linear structures (jet/lobes/hotspots) in the radio of
order ~Mpc
Summary Quasars, radio and Seyfert galaxies
Summary Quasars, radio and Seyfert galaxies
What powers AGN?
To produce L=1045 erg/s by O stars, 8x105 O stars would be
required. Assuming an emitting region of l=10-2 pc-3, their
density would have to be 1.6x1012 pc-3 (globular clusters 106
pc-3) and an average separation would have to be 3800 RSun.
To produce this luminosity by supernovae would require 104
supernovae shining at their peak luminosity!
Theoretical arguments for SMBHs in AGN:
• Radiation pressure: Lower Limit on M•
• Radiation Efficiency of Accretion on BHs
Observational evidence for SMBH in Galaxies/AGN hosts:
• High central stellar velocity dispersions
• Megamaser disks
• RadialVelocities from Ionized Gas
• Radial velocities of stars within the spheres of influence of BHs
• Broad Iron (Fe) Kα lines (relativistic accretion disk)
• Sgr A* in the Galactic Center
Arguments in Favour of SMBHs as
the Engines of AGN
Theoretical arguments for SMBHs in AGN:
• Radiation pressure: Lower Limit on M•
• Radiation Efficiency of Accretion on BHs
Observational evidence for SMBH in Galaxies/AGN hosts:
• High central stellar velocity dispersions
• Megamaser disks
• RadialVelocities from Ionized Gas
• Broad Iron (Fe) Kα lines (relativistic accretion disk)
• Sgr A* in the Galactic Center
Arguments in Favour of SMBHs as
the Engines of AGN
Radiation Pressure: BH mass limits
(Long-term) stability of the AGN gas requires that
the gravitational force exceeds or equals the radiation
pressure from the AGN:
Fgrav > Frad
r
cr
L
F e
ˆ
4 2rad
π
σ=
r
r
mmGM
F
ep
ˆ
)(
2grav
+
−=
•
Radiation Force on an electron
Gravitational Force on electron
plus proton pair (medium must
be neutral)
Radiation Pressure: BH mass limits
13814
serg)/(1026.1serg1031.6
4 −
•
−
•• ×≈×≈≤ sun
e
p
MMMM
Gcm
L
σ
π
This is known as the Eddington limit, which can be used to establish a
minimum for the mass of the BH:
For typical Seyfert galaxies L ≈ 1044 erg s−1 , so MSy ≈ 8 x 105 Msun
QSOs L ≈ 1046 erg s−1 , so MQSO ≈ 8 x 107 Msun
The Eddington luminosity is the maximum luminosity emitted by a body
of mass M● that is powered by spherical accretion.
MLM 44
5
E 108×= sun
Eddington Limit:
Radiation Pressure: BH mass limits
• Hence, the luminosity of an AGN sets a limit on its mass, independent
from size/distance (both radiation pressure and gravity decrease as 1/r2)
• This does NOT imply a SMBH, but combined with an upper limits on
the volume (e.g. from variability) it can limit alternatives (clusters of
compact objects)
• With the Eddington mass >108 Msun and the size constraints <1pc
from variability one can derive a robust lower limit for the central mass
density ρ >108 MSun pc-3 (for comparison, the central star cluster in our
Galaxy ~4x106 MSun pc-3)
WHY BLACK HOLE?
• With the Eddington mass >108 Msun and the size constraints <1pc from variability one
can derive a robust lower limit for the central mass density
ρ >108 Msun pc-3
• For comparison remember that
• in our vicinity there are only a few stars within a parsec distance.
• the central star cluster in our Galaxy has “only” ~4x106 Msun pc-3
• It was then suggested that the activity in the active nuclei was produced by a accreting
black holes.
• NB:The term ``black hole'' was invented by John Wheeler in 1967 well after the
concept was invented.
WHAT IS A BLACK HOLE
• A black hole is a concentration of mass so large, that even light
cannot escape its gravitational attraction
• A black hole has only two parameters (we ignore charge):
• the mass Mbh and
• the spin 0≤a≤1
• A non-rotating black hole (a=0) is called a Schwarzschild hole
• A rotating black hole (0<a ≤1) is called a Kerr hole
SCHWARZSCHILD RADIUS ET AL.
•Equating kinetic and potential energy in a gravitating system yields:
•This is called the Schwarzschild radius and de
fi
nes the event horizon in
the Schwarzschild metric (non-rotating black hole).
•For the mass of the Earth RS=1 cm.
•For a quasar with M=100 Million Msun we have RS= 2 AU.
•For maximally rotating black hole (a=1) the event horizon is Rg=0.5 RS
2
2
2
2
1
c
GM
R
R
mGM
mv S
cv
S
•
=
•
=⇒=
Theoretical arguments for SMBHs in AGN:
• Radiation pressure: Lower Limit on M•
• Radiation Efficiency of Accretion on BHs
Observational evidence for SMBH in Galaxies/AGN hosts:
• High central stellar velocity dispersions
• Megamaser disks
• RadialVelocities from Ionized Gas
• Broad Iron (Fe) Kα lines (relativistic accretion disk)
• Sgr A* in the Galactic Center
Arguments in Favour of SMBHs as
the Engines of AGN
BLACK HOLES ASTHE BRIGHTEST
REGIONS INTHE UNIVERSE
•Consider a particle falling from in
fi
nity onto an
object of mass M. When it is stopped at radius r, its
energy is radiated away. When mass dm is accreted
in time dt, the mass accretion generates a power:
where we call M-dot the mass accretion rate.
L U G
M
r
dm
dt
GM
M
r
L
G M M
1
c
2
ri n
GM
Mc
2
BLACK HOLES ASTHE BRIGHTEST
REGIONS INTHE UNIVERSE
• The characteristic scale of the emitting region is a few gravitational radii,
r ~ rin Rg (Rg = GM/c2)
•where η=1/rin
•Therefore, for energy dissipation near the black hole with, e.g., rin=10
(means 10Rg) we will have η=0.1 and hence a 10% efficiency in
converting rest mass into energy
INNER DISK RADII
• The top line gives the radius of maximal energy dissipation
• The bottom line gives the location of the marginally stable radius,
i.e. the inner disk radius
• Values plotted as function of angular momentum a
BLACK HOLES ASTHE BRIGHTEST
REGIONS INTHE UNIVERSE
• The efficiency η will depend on the spin (a) of the black hole:
- for a=0 (Schwarzschild) we have η =6% and for a=1 (extreme Kerr)
we have η =40%!
- Note that for nuclear fusion we only have η =0.7%.
• For LQSO=1046 erg/sec and η =10% we have Mdot = 2MSun yr-1
• The Eddington accretion rate also depends on the type of accretion:
- Spherical accretion: Eddington limit is strictly valid only for this type
- ADAF (Advection Dominated Accretion Flow): Quasi-spherical
accretion where energy is not radiated away, but carried into the
black hole (η <<0.1)
- Disk accretion: much of the radiation escapes along rotation axis.
However, strong radiation can induce a disk-wind which becomes
significant near the Eddington limit.
• =>At least for very luminous AGN, the Eddington limit is robust.
Theoretical arguments for SMBHs in AGN:
• Radiation pressure: Lower Limit on M•
• Radiation Efficiency of Accretion on BHs
Observational evidence for SMBH in Galaxies/AGN hosts:
• High central stellar velocity dispersions
• Megamaser disks
• RadialVelocities from Ionized Gas
• Broad Iron (Fe) Kα lines (relativistic accretion disk)
• Sgr A* in the Galactic Center
Arguments in Favour of SMBHs as
the Engines of AGN
M31 – Andromeda:
Stellar Kinematics
• Velocity dispersion increases to 250 km/s
toward center
• Radial velocities increase to 200 km/s
before passing through center
• Kormendy (1988) derived a mass of
about 107 Msun
Sphere of in
fl
uence of a black hole
(gravitational potential of SMBH larger
than that of the surrounding stars):
M87 (Massive Elliptical):
Gas Kinematics
• Radial velocity measurements
using spectroscopy of emission
lines of ionized gas
• Ford et al. conclude a mass of
2.4 x 109 Msun within the inner
18 parsecs of the nucleus
20 cm
1 cm
H2O megamaser @ 22 GHz detected in
NGC 4258 in a warped annulus of 0.14 −
0.28 pc and less than 1015 cm of thickness,
with a beaming angle of 11° (Miyoshi et al.
1995, Maloney 2002).
Combining the Doppler velocities (±900km
s−1) and the time to transverse the angular
distance (0.14 pc) gives the mass of the
nucleus 3.9 x 107 Msun within r ≤ 0.012 pc
NGC 4258:
Megamasers
NGC 4258: Megamasers
X-ray measurements of the innermost discs of AGN
WHY DO WE USE X-RAYS
Rapid variability of AGN reveals that the X-ray photons
come from the innermost accretion disk!
MCG-6-30-15
RELATIVISTICALLY BROADENED AND
SKEWED EMISSION LINES
RELATIVISTICALLY BROADENED AND
SKEWED EMISSION LINES
RELATIVISTIC LINE BROADENING IN
MCG-6-30-15
• emission very close to black
hole ISCO<2Rg
• accretion disk oriented at 30o
from our l.o.s.
• rotates at 0.989 of maximal
Kerr spin
Spin of a Black Hole
Measure radius of the innermost stable circular orbit
RELATIVISTIC BROAD FE LINE
•Mrk 335
Relativistic Broaden lines
NGC 1365 - not gas obscuration
The Galactic Center
Sagittarius A*
Sagittarius A*
Event HorizontTelescope
General Summary
• A massive object is required to avoid highly ionized gas being
blown away by radiation pressure.
• The accretion efficiency of SMBH can be 0.06-0.42, avoiding the
problem with the “low” nuclear burning efficiency (~0.007) of
stars (if they were the cause of AGN)
• Evidence for massive objects (SMBH) come from:
- Stellar/gas kinematics: Increasing to very small radii
- Mega-masers: Keplerian velocity of gas disks
- Broadened Fe lines: Relativistic accretion disks
-Sgr A*:Individial stellar orbits around Galactic center