HIGH ENERGY ASTROPHYSICS Gammaray Bursts GRBs

1
HIGH ENERGY ASTROPHYSICS Gamma-ray Bursts (GRBs)

• Historical background
• Main observational properties
• spatial distribution
• temporal
• spectral
• afterglows optical counterparts
• Some basic constraints
• Scenario fireball model
• internal and external shocks
• inner engine
• unsolved problems

2
Historical background
In October of 1963 the US Air Force launched the
first in a series of satellites inspired by a
recently signed nuclear test ban treaty.
Signatories of this treaty agreed not to test
nuclear devices in the atmosphere or in space.
Vela satellites (from the Spanish verb velar,
to watch) were launched and operated in pairs
with two identical satellites on opposite sides
of a circular orbit 250,000 kilometers in
diameter (about a 4 day orbit) so that no part of
the earth was shielded from direct observation.
With the timing accuracy of the later Vela
satellites (1965) Klebesadel and colleagues at
LANL were able to deduce the directions to the
events with sufficient accuracy to rule out the
sun and earth as sources. They concluded that the
gamma-ray events were "of cosmic origin". In
1973, this discovery was announced in Ap.J.
letters by Klebesadel, Strong, and Olsen. Their
paper discusses 16 cosmic gamma-ray bursts
observed by Vela 5a,b and Vela 6a,b between July
1969, and July 1972.
3
Historical background
4
Historical background
5
(No Transcript)
6
(No Transcript)
7
Observationsspatial distribution
Observed burst rate with BATSE 1 burst/day gt
one event/106yr/galaxy
Locations of a total of 2704 GRBs recorded with
BATSE onboard CGRO during 9 years of operation
burst spatial distribution is isotropic
8
Observations spatial distribution
BUT the spatial distribution is inhomogeneous
Intensity distribution log N log P P peak
flux N(P) number of bursts brighter than P For
an homogeneous distribution V?? r3 Flux ? r-2
gt N ? V ? F-3/2 gt log N - 3/2 log F
ct. expected slope of log N log P curve - 3/2
9
- 3/2 slope
Peak flux in phot/cm2/s
10
Observations spatial distribution
BUT the spatial distribution is
inhomogeneous Intensity distribution log N
log P P peak flux N(P) number of bursts
brighter than P For an homogeneous distribution
V?? r3 Flux ? r-2 gt N ? V ? F-3/2 gt log N -
3/2 log F ct. expected slope of log N log P
curve - 3/2 Lack of faint sources (Plt10) is
observed inhomogeneous distribution gt
cosmological effects if sources of weak bursts
are located at large redshifts
11
Observations spatial distribution
Another proof of the inhomogeneity of the spatial
distribution V/Vmax test quantitative
evaluation of the uniformity of the radial
distribution of a sample. For each object, its
radial location within the volume available to it
as determined by the sample limits is
computed. Uniform space distribution gt uniform
distribution of V/Vmax between 0 and 1 gt ltV/
Vmax gt 0.5 ltV/ Vmax gt smaller than 0.5 for
GRBs (0.3-0.4)
12
Observations temporal properties

• Great diversity of time profiles
• Variability down to ms scale
• Burst composed of individual pulses
• Typical individual pulse FRED (fast rise
  exponential decay duration 13)

13
(No Transcript)
14
Pulse width average 1s
Pulse separation average 1.3s
15
Observations spectral properties
Uniform spectra Non thermal, well fitted with
power laws Peak at some hundred keV No lines GeV
photons (EGRET) in some cases TeV photons
(Milagrito) possibly in one case
16
A sample of BATSE spectra uniformity non
thermal broken power-laws Kouveliotou 1994
17
Non thermal spectra, well fitted using two power
laws with a smooth transition at the peak energy
Ep E? exp(-E/E0) Elt Eb N(E) A E?
Eb?-? exp(-Eb/E0) E ? Eb with break energy Eb
(?-?) E0 and E0 Ep / (2 ?)
Band et al. 1993
18

• Eb 100-400 keV
• Average Eb 250 keV
• gt not many soft GRBs
• not real related to lower BATSE sensitivity at
  low E
• BUT X-ray flashes, discovered in 2001
• strong non-thermal emision (2-20 keV)
• weak emission in the GRB band 50-300 keV
• short durations lt some 1000 s
• 17 detected by Beppo-SAX WFC in 5 years

Preece et al. 2000
19
Preece et al. 2000
Average ? -1 Average ? -2.25
20
Observations temporal properties populations
?t 10-3-103s typical duration ? 10 s Bimodal
distribution T90lt 2s T90gt 2s N(short) 1/3
N(long) (but BATSE was less sensitive to short
bursts short burts were detected to smaller d)
Durations of the GRBs recorded with CGRO/BATSE.
T90 time needed to accumulate from 5 to 95 of
the counts in the 50-300 keV band.
21
Observations some correlations

• Hardness ratio duration correlation
• Hardness ratio - intensity correlation pulses
  become softer during the pulse decay
• Pulses are narrower at higher E

22
Piran 2004 (review)
Short bursts harder than long bursts
Hardness ratio (HR) versus duration of the GRBs
recorded with CGRO/BATSE. HRratio of fluence
between channels 3 (100-300 keV) and 2 (50-100
keV)
23
Previous results with a smaller sample (1994)
24
Observations some correlations between temporal
and specral properties

• Hardness ratio duration correlation
• Hardness ratio - intensity correlation pulses
  become softer during the pulse decay
• Pulses peak earlier and are narrower at higher E
  bands

25
Relationship between temporal and spectral
structure narrowing with E W ? E-0.4 Fenimore
1995
26
Observations afterglows
No known counterparts of GRBs at lower energies
were known before 1997 no idea of distance,
energy budget not known, Both a spherical
source distribution limited in spatial extent and
a cosmological population could be adopted (but
cosmological preferred) February 28, 1997
Italian-Dutch satellite BeppoSAX detected the
X-ray afterglow from GRB970228
27
Observations counterparts X-ray afterglows
28
Observations optical counterparts
29
Observations optical counterparts
30
Observations X-ray, optical and radio
counterparts
44 (end of 2003) GRBs with optical afterglows, 33
with known redshifts z 0.1-4.5 typical z ?
1 Before 2000 only 12 with optical afterglows
were known
31
X-ray afterglow

• First (hours) and strongest afterglow signal (in
  around 90 of GRBs)
• Energy emitted a few of GRB energy
• Lines seen in some cases, but not clearly
  confirmed

32
Optical afterglow

• Seen in 50 of (well localized) GRBs
• Fast fading in general (several weeks)
• Dark GRBs high absorption? very large z?
  intrinsically very faint?

33
Radio afterglow

• 80 of GRBs with optical afterglow have a radio
  aterglow
• Size of the emitting region deduced from radio
  emission fluctuations 1017cm at 4 weeks after
  outburst gt proof of relativistic expansion
• Longer timescale direct estimate of the total
  energy in the ejecta

34
Some basic constraints (from main observational
properties)

• Energy released in gamma-rays E? 1051 1054
  (??/4?) erg
• (beaming accounted gt effective E? 5 1050 1051
  erg collimation 10lt?lt200 )
• Short time scale (ms) variability gt small size
  (ltc.?t), compact source (of stellar M)
• Super-Eddington L no static envelope gt wind
• compactness problem number of photons with
  Egt500keV size of the source imply that the
  source is extremely optically thick to pair
  creation (??? ? ee-)
• solution relativistic wind (Lorentz factor
  ?gt100) relativistic fireball model
• Frequency of the events 10-6 per year (per
  galaxy)
• Non thermal spectrum observed emission should
  emerge from an optically thin region

35
Fireball model sources

• Merging of compact stars (10-5/yr)
• Neutron star Neutron star
• Black Hole Neutron star
• (Collapse of a White Dwarf to a Neutron Star -
  accretion induced collapse does not work because
  of baryon pollution - see later)
• Collapse of a massive star (10-3/yr) failed
  supernova, collapsar, hypernova ...
• afterglow seen well inside host galaxies (in
  coinidence with SNe in some cases) . Could
  explain some long bursts but not the short ones
• FINAL CONFIGURATION BH thick disk

36
Fireball model formation of the relativistic wind

• Available energy (BHdisk)
• mass accretion gravitational energy of the disk
• black hole angular momentum extraction by a
  magnetic field (Blandford-Znajek process)
• 1053-1054 erg (OK)
• Energy extraction ? ? annihilation ? ee- ? ??
• magnetic processes
• Problem of baryonic pollution E degraded from ?
  to UV...
• unless ? E/Mc2 gt 100 (E,M wind mass and
  energy)
• solved if injection of E along the system axis
  (baryon free region because of centrifugal
  forces)

37
GRBs as collimated jets from neutron star/black
hole mergersMochkovitch, Hernanz, Isern,
Martin, Nature, 1993
38
GRBs from relativistic beams in neutron star
mergersMochkovitch, Hernanz, Isern, Loiseau,
AA, 1995
39
Fireball model, with relativistic
internal-external shocks

• Three main stages
• Inner engine (not observed directly, hidden)
  produces a relativistic energy flow. Observed
  rapid fluctuations and huge E released gt compact
  source
• Energy transferred relativistically to optically
  thin regions (distances 1013 cm)
• Relativistic ejecta is slowed down internal
  shocks convert KE to internal E of accelerated
  particles which emit the observed ? rays
• Afterglow External shocks due to interaction of
  relativistic matter with surrounding matter
  ISM, circumstellar wind

40
Internal shocks irregular flow, where faster
shells catch up and collide with slower shells
kinetic E ? internal E Highly variable temporal
structure is well reproduced Shocks take place
at 1013 cm Duration of the GRB t during which
inner engine is active
41
(If) GRB arise from internal shocks and aterglow
from external shocks no direct scaling between
GRB and its afterglow gt confirmed by observations
42
Emission mechanism (GRB and afterglow)
synchrotron
X ray to radio spectrum of GRB 970508
?m sync. freq. of an e with Emin (low-E
cutoff) ?c sync. freq. of an e that cools
during the local hydrodynamic time scale. For
slow cooling ?cgt ?m
43
Sari et al, 1998
44

• OPEN QUESTIONS
• Energetics is there an energy crisis? Whats
  the energy distribution between the GRB and its
  afterglow?
• Unknown micropysics how does the inner engine
  accelerate the ejecta to relativistic velocities?
• Observations can help to understand
• the sources GRB-SN association, GRB and star
  forming regions? Detection of gravitational waves
  coincident with GRBs would demonstrate the
  connection with NS-NS mergers, for instance
• physical processes detailed (multi-?)
  information about the promt emission and early
  afterglow (HETE-II, SWIFT)

45
REFERENCES

• Piran, T., 1999, Gamma-ray bursts and the
  fireball model, Physics Reports, 314, 575-667
• Piran, T., 2000, Gamma-ray bursts a puzzle
  being resolved, Physics Reports, 333-334, 529-553
• Piran, T., 2004, The physics of gamma-ray bursts,
  Reviews of Modern Physics (astro-ph/0405503)
• Fishman, G.J, Meegan, C.A., 1995, Gamma-ray
  bursts, Annual Review of Astronomy
  Astrophysics, 33, 415-458
• Kouveliotou, C. et al., 1993, Identification of
  two classes of gamma-ray bursts, Astrophys. J.,
  413, L101-L104
• Sari et al., Spectra and light curves of
  gamma-ray burst afterglows, Astrophys. J., 497,
  L17-L20
• Preece, R.D. et al., 2000, The BATSE gamma-ray
  burst spectral catalog. I., Astrophys. J.
  Supplement, 126, 19-36