Download Introduction to Active Galactic Nuclei

Lecture: April 3, 2003
• Summary
of Emission Lines from Ionized Gas
• Introduction
to AGN
• Historical Perspective
• AGN Taxonomy
• Standard Model - Black Hole Paradigm
Useful References
Gaseous Nebulae:
•
•
•
Shu (pg 216)
Online notes
Osterbrock
AGN:
•
•
•
•
C&O Ch 26
Introduction to Active Galactic Nuclei by Peterson(first chapter available free online;
see reading references on class webpage)
Active Galactic Nuclei by Krolik (more advanced graduate-level textook)
Quasars and Active Galactic Nuclei - An Introduction , by A.K.Kembhavi and
J.V.Narlikar (slightly more advanced graduate-level textbook)
Summary of Emission Lines from Ionized Gas
Forbidden Lines from Ionized Gas: “2-Level Atom”
2
E21
1
Upward transitions can occur
through:
Downward transitions can occur
through:
• photoionization from a lower
level
• Spontaneous decay (rate
determined by A21)
• collisional excitation (inelastic
collisions with electrons)
• collisional de-excitation (elastic
collisions with electrons)
In equilibrium, upward transition rate from level 1
balances downard transition rate to level 1
This means:
Einstein A-coefficient
n e n 1 q12 (T)  n 2A21  n e n 2q 21(T)
[1]
Collisional de-excitation coefficient
Increases with ne
Increases with n1
Collisional excitation coefficient
The collisional coefficients are related through the Boltzmann factor
g 2 - EkT21
q12 (T)  q 21 (T) e
g1
Where g2, g1, are the statistical weights of the two levels (=2J+1)
[2]
The collisional de-excitation coefficient is given approximately by:
q 21  v 21
Where the thermal speed of the electrons is given by:
1
3
kT
2
m e v  kT ; v 
2
2
me
And an elastic cross section given approximately by
2
h
 12 
me kT
q 21 
h2
3
2
m e kT 
1
2
A more detailed calculation yields:
8.63 10-6 12
q 21 
g2 T
Where 12 comes from quantum
mechanical calculations
Using Equations [1] and [2], we get that the level populations are related by:
n1n e q 21 (T) g 2 - EkT21
n2 
e
A 21  n e q 21 (T) g1
[3]
The limiting cases are:
n eq 21(T)  A21
g 2 - EkT2 1
n 2  n1 e
g1
Boltzmann distr.
A 21
n e 
 n critical
q 21 (T)
This defines the critical density for a transition; at densities above the critical
density, transitions from the upper to lower level are more likely to occur
through collisions rather than spontaneous decay.
How do we use this information?
The luminosity of a line (per unit volume) caused by a downward transition
from 2 -> 1 is:
L21  n 2 A 21hν
Using equation [3], this becomes:
n1n e q 21 (T)A 21h g 2 - EkT21
L 21 
e
A 21  n e q 21 (T) g1
At low densities:
g 2 - EkT21
L 21  n1n e q 21 (T)h 
e
 n2
g1
At high densities:
g 2 - EkT21
L 21  n1A 21h
e n
g1
The transition between these two regimes occurs at the critical density:
A 21
n e 
 n critical
q 21 (T)
This is a useful diagnostic - if we don't see certain lines of an ion, it may be
because the density is too high for their radiation to occur. This happens
when the mean time between collisions is comparable to or less than the
radiative lifetime A21-1. (This is why we don’t see forbidden lines on Earth)
For permitted transitions (electric dipole radiation), the transitions rates
occur at A ~ 108 sec-1 and corresponding critical densities of 1015 cm-3.
For forbidden lines, the A coefficients are much smaller so the critical
densities are much lower - comparable to densities in typical nebulae.
Critical densities for some common transitions:
Identification Wavelength (Å) log Ncr (cm -3) Identification Wavelength (Å) log Ncr (cm -3)
[C III
1909
9 [Fe VII]
5721.1
7.6
[O II]
3726.1
3.5 [N II]
5754.6
7.5
[O II]
3728.8
2.8 [Fe VII]
6086.9
7.6
[Fe VII]
3760.3
7.6 [O I]
6300.3
6.3
[Ne III]
3868.8
7 [S III]
6312.1
7.2
[Ne III]
3967.5
7 [Fe X]
6374.6
9.7
[S II]
4068.6
6.4 [N II]
6583.4
4.9
[O III]
4363.2
7.5 [S II]
6716.4
3.2
[Ar IV]
4711.3
4.4 [S II]
6730.8
3.6
[Ar IV]
4740
5.6 [Ar III]
7135.4
6.7
[O III]
5006.9
5.8 [O II]
7319.9
6.8
Appenzeller and Östreicher 1988 (AJ 95, 45) and Table 3.11 of Osterbrock
Many of these lines are seen in gaseous nebulae,
and the strength of the lines depends on the
density of the gas
So, summarizing, because these lines are excited primarily by collisions
with electrons, the observed line strengths can be used to measure the
density, temperature, and then composition of the gas!!
To measure density:
Use line ratios from lines with different critical densities (but are close
together in energy to eliminate temperature dependence ) one in the lowdensity limit (L a n2) and one in the high-density limit (L a n).
In this example, the [SII] at 6716Å
comes from a 4S3/2-2D5/2 transition
and has a critical density of 1.5 x
103 cm-3. The 6731Å line comes
from the 4S3/2-2D3/2 transition with
critical density 3.9 x 103 cm-3. Note
that the line ratio is sensitivite to
density between about 100 to
10,000 cm-3.
To measure temperature:
Again use single ion (independent of abundance and ionization) but pick
lines that have large differences in excitation potential.
The most-used example is [O III] with lines at 4363 and 4959+5007 Å.
Once these line ratios are used to constrain density and temperature,
abundances can be derived!
The Nature of the Ionizing Source:
Since the ions observed are ionized by photons, and the ionization
potential of metals varies, line ratios of lines from various ions can be
used to infer the spectral shape of the ionizing radiation field. Here
are the ionization potentials (eV) for a few common ions:
Element
H
He
C
N
O
Ne
I
II
13.6
24.6
11.3
14.5
13.6
21.6
III
54.4
24.4
29.6
35.1
41
IV
47.9
47.4
54.9
63.5
64.5
77.7
77.4
97.1
Since even the most massive (hottest) stars produce very few
photons beyond ~ 50 eV, the presence of lines from highly ionized
species indicated the presence of some source of hard photons other
than stars. What could this be?
Next topic:AGN
Introduction to Active Galactic Nulcei (AGN)
• Historical Background
• Taxonomy - Classes of AGN
• Brief overview of continuum and spectral characteristics
Denition
What are Active Galactic Nuclei?
Active Galactic Nuclei (AGN) are nuclei of galaxies which show energetic
phenomena that cannot be clearly and directly attributed to stars.
Signs of Activity:
•Luminous UV emission from a compact region in the center of a galaxy
•Strongly Doppler-broadened emission lines
•High variability
•Strong non-thermal emission ; polarized emission
•Compact radio core
•Extended, linear radio structures (jets+hotspots)
•X-ray, g-ray, and TeV-emission
In some luminous AGN (quasars) the radiation from a region comparable to the solar system can
be several hundred times brighter than the whole galaxy.
Historical Perspective
First detections
• NGC 1068 (E.A. Fath 1908, Lick Obs.) and V.M. Slipher (Lowell Obs.) - strong
emission lines similar to planetary nebulae, line widths of several hundred
km/sec
• Detection of an optical jet in M87 (Curtis 1913)
• Same period: Einstein develops GR, Schwarzschild metric (1916), but no
connection seen yet
• Hubble (1926) - Nebulae are extragalactic (galaxies)
• Carl Seyfert (1943) - found several galaxies similar to NGC1068 (henceforth
named Seyfert galaxies), i.e. galxies with a bright nucleus and strong emission
lines (NGC1275, NGC3516, NGC 4051, NGC 4151, NGC7469)
• Detection of radio emission from NGC1068 & NGC1275 (1955)
• Woltjer (1959) - Timescale for Seyfert-activity 108 years (1% of Galaxies are
luminous Seyferts), Mass of nucleus is very high 108-10 M(velocity dispersion/line
width several thousandkm/sec at base in unresolved nucleus)
Historical Perspective
Radio Surveys & Quasars
Early radio surveys were very important for the discovery of quasars. For
example:
• 3C and 3CR: The third Cambridge (3C) catalog (Edge et
al. 1959) at 158 MHz. Revision: 3CR catalog (Bennett 1961)
at 178 MHz down to limiting flux density of 9 Jy (1 Jy = 10-26 W m-2 Hz-1) Discovery of first quasars (e.g. 3C273,279)
•PKS: Parkes (Australia, Ekers 1969) survey of southern sky at
408 MHz (> 4 Jy) and 1410 MHz (> 1 Jy).
Sources found:
•Normal Galaxies
•Stars with weird broad emission lines!?
Detection of Quasars
Radio ID of quasars by Hazard et al. (1963), Maarten Schmidt (1963):
The lines in 3C273 are highly redshifted (z=0.158) emission lines !
first detection of quasars (3C 48, 3C273), from Hubbles law follows
(h0 = H0/(100 km s-1 Mpc-1)
(100 times larger than normal galaxy)
quasar = quasi-stellar radio source, QSO = quasi-stellar object
(no longer distinguished)
Quasars are much bluer than stars
AGN were quickly seen to show emission in all astrophysically
relevant wavelengths regimes
AGN Taxonomy
•
•
•
•
•
•
Seyfert galaxies 1 and 2
Quasars (QSOs and QSRs)
Radio Galaxies
LINERs
Blazars
Related phenomena
Seyferts
• Lower-luminosity AGN
MB > -21.5 + 5log(H0/100) (Schmidt & Green 1983)
• Quasar-like nucleus; host galaxy clearly seen
• Seyferts occur mainly in spirals
• Nucleus has strong, high-ionization emission
lines in the optical
Optical spectra of Seyferts
• Broad lines:
FWHM~500-10,000 km/s
permitted lines;
high-density gas (ne> 109 cm-3)
~ pc distance from center
• Narrow lines:
FWHM ~ hundreds km/s
forbidden lines
low-density gas (ne~ 103-6 cm-3)
~ 50-100 pc distance
• Absorption features due to stars in the host galaxy
BLR/NLR Diagnostic
• Dynamics gives clouds location
• Types of lines observed/not observed give info on
temperature, density
Example:
[OIII] λ5007 not observed from BLR
critical de-excitation density ne~106cm-3
lower limit
[CIII]λ1909 is observed from BLR
critical de-excitation density ne~1010cm-3
upper limit
Seyferts 1 and 2
• Seyferts 1: both broad and narrow lines
Balmer lines, [OIII] λ4959,5007
• Seyferts 2: only narrow lines
[OIII] λ4959,5007
• Intermediate types: 1.1-1.9
decreasing intensity of the broad
line component
Blue Continuum
Reddened continuum
Seyfert 2
LINER
Quasars
• Most luminous AGN
MB< -21.5 + 5log(H0/100)
• Unresolved (<7”) on Palomar Plates
• Weak “fuzz” on deep observations (HST)
• Similar nuclear spectra to Seyferts, but
weaker abs lines and narrow/broad ratios
QSO=optically bright (most)
QSR=radio bright (5-10%)
Radio Galaxies
• Occur in giant ellipticals
• Bright radio emission with extended features
(jets, lobes) and compact core
• Broad-Line Radio Galaxies (BLRGs)
Narrow-Line Radio Galaxies (NLRGs)
QSRs
Cygnus A in the radio
LINERs
• Low Ionization Nuclear Emission Lines
• Very faint, and numerous; may be ~50% of
local extragalactic population
• Similar to Sy2 but stronger low-ionization
lines
• What are LINERs?
Ratios of line fluxes depends on
shape of ionizing continuum
Starlight vs AGN light
• Ionization potential for O
to OIII:
54.93 eV
• AGN produce many
photons at these energies
Are LINERs powered
by faint AGN?
LINER diagnostic
Ho et al. 1989
x=HII regions
LINERs
Seyferts
Blazars
• High luminosity, non-thermal continuum from
radio to gamma-rays
• Flat or inverted radio spectrum
• Rapid variability T day- hours
• Large Polarizations ~ 10%
BL Lacs: weak emission lines, nearby
OVVs/FSRQs: strong emission lines, distant
Comparison of Optical Spectra
Related Phenomena
• Starbursts: Intense starformation, often nuclear
bright in optical, X-rays, radio
• Markarian galaxies: from Markarian survey at
Byurakan Observatory, Armenia
UV-excess galaxies
11% Seyferts, 2% QSOs and BL Lacs
• ULIRGs: from IRAS survey in the 80s at λ>10μ
L(8-1000μ) > 1012Lsolar
due to dust heated by AGN/Starburst
What powers AGN?
• It is very difficult to explain the observed
set of characteristics of AGN without
invoking a supermassive central black
hole.
AGN: What are they good for?
• Study black hole physics
• Study high energy physics
• Background sources on cosmological scales:
-Ly-a forrest in optical spectra (absorbing gas in cosmic walls?)
-Are gravitationally lensed by clusters, get Hubble constant from time
delay
-Background radiation to detect absorption lines in host galaxies at
large z
-Produce cosmic background radiation, e.g. at X-ray wavelengths
• Find galaxies at highest redshifts (z > 5.8!)
• Constrain cosmology
• History of our Galaxy; impact on galaxy evolution
The Black Hole Paradigm
Basic question early on: what powers quasars, Seyferts,
and radio galaxies?
Remember: the characteristic signatures of quasars are
• high luminosity (L > 1044 erg/sec, i.e. 1010.5-14.5 L)
• high compactness (variability on time scales of years, compact
radio cores < 1pc)
Such a high luminosity will produce an enormous radiation pressure.
Hence, for material to be gravitational bound to the center of the galaxy,
we can calculate a minimum central mass - the Eddington mass independent of a particular model.
Eddington Limit
•The momentum of photons is l = E/c (one photon: E = h)
• the force F of photons is F = dl/dt = E/c
• the pressure is force per area P = F/A
Hence, the radiation pressure at distance r of an isotropically radiating point source of luminosity L is (L: luminosity, F: flux density)
The force exerted on a single electron is obtained by multiplying
with the electron scattering cross-section (dimension cm2):
Here e is the Thomson cross-section (from classical electron radius)
The inward directed gravitational force of a central mass M is given by
NB: The radiation pressure is mainly acting on the electrons while gravitation mainly acts on protons
(since they have bigger mass). Coulomb forces will keep them together!
In order for matter to be gravitationally bound (to fall inward), we need
which is independent of the radius.
This condition is called the Eddington Limit and states that
fora given luminosity a certain minimum central mass is
required (Eddington mass) or that for a given central mass
the luminosity cannot exceed the Eddington luminosity.
Schwarzchild Radius
Equating kinetic and potential energy in a gravitating system yields:
1
2
2
mv esc

GmM
r
setting v = c-the speed of photons-and canceling m we get:
RS 
2GM
c2
This is called the Schwarzschild radius and defines the event
horizon in the Schwarzschild metric (non-rotating black hole).
•For the mass of the earth we have RS = 1cm
•For a quasar with M = 108 M we have RS = 3 x 1013 cm = 2 AU.
When mass falls onto a black hole, potential energy is converted into
kinetic energy. This energy is either advected into and beyond the
event horizon or released before.
The potential energy of a mass element dm in a gravitational field is
U
GMdm
r
The available energy (luminosity) then is
GM dm GMM

L U 

r dt
r
where we call M the mass accretion rate
The characteristic scale of the emitting region will be a few gravitational radii,
i.e. r ~ rinRs
Efficiency is just a function of how compact the object is
The compactness of the accretion disk will depend on
the spin (a) of the black hole: for a = 0 (Schwarzschild)
we have = 6% and for a = 1 (extreme Kerr) we have =
40%! Note that for nuclear fusion we only have = 0.007.
For a quasar with L=1046 ergs/s and =0.1, we have M =
2 M yr-1.
Accretion is the most efficient energy-generation process
currently known
Where does the luminosity come from?
Accretion Disks
Luminosity of AGN derives from gravitational potential energy of gas
spiraling inward through an accretion disk. Mass streams are in orbit
around the BH; viscosity, an internal force that converts the directed KE
of the bulk mass motion into random thermal motion, causes orbiting gas
to lose angular momentum and fall inwards. Energy is dissipated in the
disk and is radiated away.
Angular momentum transport
If the disk is thin (this means that the orbital velocity is much greater than the
sound speed), then orbital velocity of the gas is Keplerian:
Specific angular momentum vfR is:
i.e. increasing outwards. Gas at large R has too much angular momentum to
be accreted by the black hole.
To flow inwards, gas must lose angular momentum, either:
• By redistributing the angular momentum within the disk (gas at small R loses
angular momentum to gas further out and flows inward)
• By loss of angular momentum from the entire system.e.g. a wind from the disk
could take away angular momentum allowing inflow
Gas surface density
Redistribution of angular momentum within a thin disk is a diffusive process - a narrow
ring of gas spreads out under the action of the disk viscosity:
Radius, R
With increasing time:
• Mass all flows inward to small R and is accreted
• Angular momentum is carried out to very large R by a vanishingly small
fraction of the mass
Radiation from thin disk accretion
Consider gas flowing inward through a thin disk. Easy to
estimate the radial distribution of temperature.
The total energy felt by a mass m of orbiting gas is:
E
R
R-dR
GMm
2r
Conservation of energy required that energy dE radiated in
time t be equal to energy difference between the two
boundaries in picture above:
t
dE
d GMm
MM
dE 
dr  (
)dr  G 2 dr
dr
dr
2r
2r
If the luminosity of the ring is dLring, then:
Using Stefan-Boltzmann law with A=2(2prdr):
Solving for T:
1
4
3
   R 4
 MM

T   G

3  
8
p
R

 r
t
MM
dL ring t  dE  G 2 dr
2r
dL ring

MM
 4prT dr  G 2 dr
2r
4
Correct dependence on mass, accretion rate, and radius, but wrong prefactor. Need
to account for:
• Radial energy flux through the disk (transport of angular momentum also means
transport of energy)
• Boundary conditions at the inner edge of the disk
Correcting for this, radial distribution of temperature is:
…where Rin is the radius of the disk inner edge. For large radii R >> Rin, we
can simplify the expression to:
with Rs the Schwarzschild radius as before.
For a hole accreting at the Eddington limit:
• Accretion rate scales linearly with mass
• Schwarzschild radius also increases linearly with mass
Temperature at fixed number of Rs decreases as M-1/4
Disks around more massive black holes are cooler.
For a supermassive black hole, rewrite the temperature as:
Accretion rate at the Eddington limiting luminosity (assuming =0.1)
A thermal spectrum at temperature T peaks at a frequency:
An inner disk temperature of ~105 K corresponds to strong emission at
frequencies of ~1016 Hz. Wavelength ~50 nm.
Expect disk emission in AGN accreting at close to the Eddington limit to
be strong in the ultraviolet –
origin of the broad peak in quasar SEDs in the blue and UV.
Disk has annuli at many different temperatures – spectrum is weighted sum of
many blackbody spectra.
Consistent with the broad spectral energy distribution of
AGN in the optical and UV regions of the spectrum.