High energy emission in Gamma Ray Bursts Gabriele
High energy emission in Gamma Ray Bursts Gabriele Ghisellini INAF – Osservatorio Astronomico di Brera
“Pillars” of knowledge Criterion: the most important and not controversial facts constructing the basics of our understanding
1 st Pillar: GRBs are cosmological 970228 Costa+ 2007 970508; z=0. 835 Metzeger+ 2007 (therefore large energetics, but how large? Depends on collimation…). Thanks to Beppo. SAX and its team, led by Luigi Piro, and to Paczynski)
Attention: not bolometric for Swift
2 nd Pillar: GRBs have large G (From Ge. V; msec variability; radio scintillation; theory) 090510 970508 Frail+ 1997: G~4 two weeks after Abdo+ 2009; Ghirlanda+ 2010; GG+2010; Ackermann+ 2010: G>1000
3 rd Pillar: Prompt+Afterglow (but X-rays may be late prompt). Energy is NOT released ENTIRELY during the prompt. Prompt SAX X-ray afterglow light curve After Swift Willingale et al. 2007 Piro astro-ph/0001436 Before Swift
4 th Pillar: Long & Short But there are exceptions + extended emission SHORT LONG Short – Hard Long - Soft
5 th Pillar: Same Dt of spikes during the prompt Spikes have same duration A process that repeats itself
6 th Pillar: Supernova connection i. e. progenitors. But there are exceptions. Evidence can be gathered only from nearby, under-luminous GRBs. No SN 060614 060218 Campana+ 2006 Della Valle+ 2006 Woosley Bloom 2006
7 th Pillar: Phenomenology of the prompt & “afterglow” Diversity, but some common behavior exists. 2 examples: Epeak ke. V ? flares flat p ee st 10 p 100 rt o Sh ng Lo stee 1000 The early X-ray afterglow is “typical” Log X-ray flux The total energy of the prompt correlates with peak of the spectrum Eiso erg Log time
Ideas (and enigmas)
Central Engine Black hole or magnetar, or more exotic? (quark star? ) Magnetars: Giant flares to explain SGRBs + some short (but numbers are not ok) During the magnetar phase: flat X-ray plateaux Magnetar BH transition (re-edition of Supra. Nova). GRBs from quark stars: one-way membrane for baryons, only e+-, photons, B-fields escape… Paczynski & Haensel 2005 MNRAS 362, L 4
Magnetic or matter dominated? Internal pressure: Random bulk random Disorder disorder “Heavy FB” optical flash R~109 cm Annihilation Blandford: bulk random order disorder Light “FB” no opt. flash, no inertia, very large G=? G~100 G Dissipation at large R. Variability through minijets or small scale instabilities? (Lyutikov)
In any case: L ~ B 02 R 02 c/8 p ~ 1051 B 152 R 62 erg/s B 0 ~1015 R~106 cm G G=? ~Everybody: At the start: B 0~1015 G for BZ Conversion of Poynting to kinetic Cyclo n >mec 2 Smaller scattering cross section Different Eg different B 0? Is the funnel useful to collimate? No, it is a myth, short can do without, as well as blazars
Internal shocks: collisions within the flow. Dissipate RELATIVE kinetic energy Lazzati+ 1999 Efficiency is small. 5% G 2 / G 1 Willingale+ 2007 Log Eafterglow Big prompt/afterglow ratio Even bigger if X-rays are late prompt. Ge. V relax, but not enough. Eaft ~ Eprompt/10 Log Eprompt
Internal shocks: collisions within the flow. Dissipate RELATIVE kinetic energy Efficiency is small. Big prompt/afterglow ratio Even bigger if X-rays are late prompt. Ge. V relax, but not enough. Deep impacts? Lazzati+ 2009
What makes the light we see? For the prompt: we don’t know. Must be efficient: short cooling time. If synchro, or IC: F(E) = k E-1/2. SSC even steeper: k. E-3/4
Line of death for non cooling e- Line of death for cooling e- Kaneko+ 2006 Nava Ph. D thesis 2009
“Afterglows”: X-rays and the optical have often different behaviors. ? k c o h s l opt a n ical r e t x e. e. i ? w o l g r e t f X-ray a ” l a e r “ s i h t TA s I
2 components? Late prompt+forward shock light curves resemble t-5/3, like rate of fallback material alate ~5/3 prompt
Log n. Fn Spectral-energy correlations a Epeak b GBM Log n
Amati, Ghirlanda, Firmani, Yonetoku… Under attack from the start (selection effects). Fiery replies. 97 GRBs ti” a m “A Ep-Eiso 0. 5 Ghirlanda 2009
Amati, Ghirlanda, Firmani, Yonetoku… Under attack from the start (selection effects). Fiery replies. q 2 jet 97 GRBs ti” a m “A Ep-Eiso 0. 5 Ghirlanda 2009
Amati, Ghirlanda, Firmani, Yonetoku… Under attack from the start (selection effects). Fiery replies. 29 GRBs q 2 jet “G hi rl a nd a” Ep-Eγ 1. 03 97 GRBs ti” a m “A Ep-Eiso 0. 5 Ghirlanda 2009
Yet we see the “Epeak-L” correlation in single GRBs Epeak [ke. V] Ghirlanda+ 2009 Rate ! !. s t c e f f e n o i t c e l e s o /2 t 1 e L k = u E pd eak t o n s i FERMI-GBM Th Luminosity [erg/s]
High energy
EGRET: 100 Me. V-10 Ge. V 18 Ge. V Hurley et al. 1994
GG+ 2010 Fermi: 100 Me. V - 100 Ge. V
short
Log n. Fn a b GBM G LAT Log n
Log n. Fn b vs G a b LAT GBM Log n a vs G G G
The 4 brightest LAT GRBs t-10/7 This is puzzling Spectrum and decay: afterglow = forward shock in the circumburst medium
Adiabatic fireballs: Lbolom = a -1 t Radiative fireballs: Lbolom = b t-10/7
The 4 brightest LAT GRBs e! tiv d ia Ra t-10/7
The 4 brightest LAT GRBs e? tiv dia Ra t-10/7
e
e
e ee+
e p ee+
Time
T A -L i rm e F n n n n GRB 090510 Short Very hard z=0. 903 Detected by the LAT up to 31 Ge. V!! Well defined timing Delay: ~Ge. V arrive after ~Me. V (fraction of seconds) Quantum Gravity? Violation of Lorentz invariance?
precursor 8 -260 ke. V 0. 26 -5 Me. V Due to Lorentz invariance. LAT all violation? >100 Me. V 0. 6 s 0. 5 s 31 Ge. V >1 Ge. V Time since trigger (precursor) Abdo et al 2009 Delay between GBM and LAT
0. 1 Ge. V Time resolved 2 30 Ge. V Different component 3 0. 5 -1 s 3 If LAT and GBM radiation are cospatial: 4 G>1000 to 1 avoid photon-photon absorption If G>1000: deceleration of the fireball occurs early afterglow! If G>1000: large electron energies synchrotron afterglow! Energy [ke. V] Abdo et al 2009 n. F(n) [erg/cm 2/s] Average
T A -L i rm e F t 2 Ghirlanda+ 2010 t-1. 5
>1 Ge. V T-T* [s] Ghirlanda+ 2010 0. 1 -1 Ge. V
Strong limit to quantum gravity MQG > 4. 7 MPlanck T-T* [s] Ghirlanda+ 2010 ~Me. V and ~Ge. V emission are NOT cospatial. But the ~Ge. V emission is… No measurable delay in arrival time of high energy photons: tdelay<0. 2 s
Conclusions “Paradigm”: internal+external shocks, synchrotron for both: it does not work Fermi/LAT detection large G Early high energy (and powerful) afterglow Decay suggests radiative afterglows GRB 090510: Violation of the Lorentz invariance? No (not yet)
Pillars Central engine Progenitor Efficiency Rad. process Precursors “Afterglows” Ambient Correlations Ge. V Hosts Shorts High-z New pillar 4 th Pillar: Long & Short (8) Nava+ 2010 Similar spectra, especially for the first second of long Fluence Peak Flux
Energetics Amati corr. Luminosities Yonetoku corr. LONG GRBs Ghirlanda et al. 2009 A 2: Short vs Long: < Energetics ; = Luminosities
FERMI GRBs & TIME INTEGRATED correlations
Pillars Central engine Progenitor Efficiency Rad. process Precursors “Afterglows” Ambient Correlations Ge. V Hosts Shorts High-z New pillar For the prompt: we don’t know. Must be efficient: short cooling time. If synchro, or IC F(E) = k E-1/2. SSC even steeper: k. E-3/4 1057 photons: large entropy (# of photons per particle), t. T>1 For the afterglow: when it is forward shock it is synchrotron, but when it is late prompt… we don’t know.
Pillars Central engine Progenitor Efficiency Rad. process Precursors “Afterglows” Ambient Correlations Ge. V Hosts Shorts High-z New pillar Isotropic or collimated? Attention: not bolometric for Swift
Pillars Central engine Progenitor Efficiency Rad. process Precursors “Afterglows” Ambient Correlations Ge. V Hosts Shorts High-z New pillar Isotropic or collimated? Strongest argument: Ghirlanda relation Nava+ 2006; Ghirlanda+ 2007 G<100 “G hir lan da ” 1 - cos qjet ” m A “ i at
Pillars Central engine Progenitor Correlations Ge. V Efficiency Rad. process For long GRBs: Wolf-Rayet? Isolated or binary? (to give angular “Afterglows” momentum). What triggers the SN, if a BH forms? The jet? In all SN Ic? For short: merging NS-NS?
Pillars Central engine Progenitor Efficiency Rad. process Precursors “Afterglows” Ambient Correlations Ge. V Hosts Shorts High-z New pillar Isotropic or collimated? But this? No jet breaks G<100
pe ak E tru 0 e. 7 E cr 2=1. 27 Ghirlanda, Ghisellini & Lazzati 2004 Epeak(1+z) Peak energy vs. True energy
Nava et al. 2006 Homogeneous density
7 “L or en N tz g~ in co v ar ns ia t~ nt 10 ” 57 Nava et al. 2006 Wind-like density
- Slides: 59