AFTERGLOW PHYSICS Alin Panaitescu Los Alamos National Laboratory - PowerPoint PPT Presentation

AFTERGLOW PHYSICS Alin Panaitescu Los Alamos National Laboratory

Relativistic blast-wave model for GRB afterglows Reverse shock into ejecta Rel. outflow Γ 0 ~ 100-500 ( Sari&Piran 99, Meszaros&Rees 99, Kobayashi 00) Interaction with CBM Forward shock into ambient medium ( Rhoads&Paczynski 93, Meszaros & Rees 97) RS FS

Afterglow radiation mechanism : synchrotron (inverse-Compton much less) Even if E B ~ E γ initially, B is too weak at 10 17 cm Origin of magnetic field and relativistic electrons: 1. magnetic dissipation in Poynting outflow (Rees & Meszaros 97, Lyutikov & Blandford 2000) 2. dissipation of ejecta energy by interaction with CBM (Rees & Meszaros 1994, 1997) + plasma instability (e.g. Weibel – Medvedev 2001 ) or Fermi acceleration (electrons) Afterglow light-curve depends on 1. dynamics of shocked gas (N e , γ e for FS ; B, Lorentz boost) 2. distribution with energy of radiating electrons (sets synchrotron spectrum) & 3. distribution of incoming ejecta (sets N e , γ e for RS ) { R -1/2 , wind CBM n α R − 2 + F ν α ν −β F ν α t − α ( β ; s,e,...) Γα { { R -3/2 , homogeneous CBM Power-law spectrum Power-law light-curve

Power-law decay indices F ν α t −α for RS and FS light-curves

Multi-wavelength afterglow observations VLA OVRO ATCA RADIO ls - VLT OPTICAL HST MDM ls -l X-RAY BSAX SWIFT CXO

Parameters of forward-shock emission up to 4 constraints ( ν a , ν p , ν c ,F p ) 4 parameters blast-wave kinetic energy E~10 53 erg/sr medium density n~0.1-1 cm -3 micro parameters ε B ~ 10 -3 & ε e ~ 10 -2 970805- Wijers & Galama 98 030329

Collimated outflow O B θ S 1/ Γ if Γ > θ -1 (spherical) → α = 1.5 β +(-.5,0,.5) if Γ < θ −1 → α = 2 β , 2 β +1 emitting surface Flux dimming is faster after Γ = θ −1 because Kulkarni et al 99: GRB 90123 1. lack of emitting fluid at angle > θ ∆α coll = 1 / 2(wind), 3/4 (homogeneous) 2. jet lateral spreading : faster deceleration Γ α t -1/4(-3/8) → Γ α t -1/2 , ∆α spread < 1/2

Optical light-curve breaks in pre-Swift afterglows Zeh, Klose & Kann 06

Jet dynamics - Γ ( r ), θ ( r ) - and emission - F ν (t) - calculated numerically + data fit determine jet parameters and medium comparable fits – Jet model homog. better fit than wind – Jet model homog. better fit than wind – SO model

Best fit parameters for uniform jets from fits to multiwavelength afterglow data Results : - high GRB efficiency (10-80%) - narrow jets (2-3 deg) - initial jet kinetic energy comparable with that of SNe - wind density parameter consistent with Galactic WRs - non-universal microphysical parameters

Numerical modeling of broadband emission of 10 GRB afterglows: Jet/Struct.outflow model – uniform CBM fits better than wind: 7-1(2)/6-1(3) → why is ambient medium homogeneous if progenitor is Wolf-Rayet star ? Chevalier, Li & Fransson 04 1. termination shock of free WR wind with radius smaller than R aglow = 0.4 (E 53 t day /n o ) 1/4 pc 2. peculiar motion (~v shock ~50 km/s) of WR star → smaller R shock 3. faster & tenuous wind ( expelled in the last < 1000 yrs before core-collapse ) interacting with WR wind R shock

Swift X-ray afterglows: three phases

Jet-breaks in X-ray light-curves (Swift) 1/3 of Swift X-ray afterglows display breaks 0.5 < α x < 1.5 to 1.5 < α x < 2.5 at 0.5-10 d 1/3 may also have a break at 1-10 d 1/3 do not have a break until > 10 d while ~ 75% of pre-Swift optical afterglows display a break at 0.3-3 d Reason : Swift “sees” dimmer afterglows from wider jets, whose lc breaks occur later

red = light-curves with breaks purple = lcs without breaks until ~10 d Flux & jet-break time dep on θ j if jet energy were universal F ν α dE/d Ω α θ j -2 t break α (dE/d Ω)θ j 4 α θ j 2 F ν α 1/ t break means that afterglows with earlier jet-breaks are brighter b

Jet-breaks = achromatic (late and followed by steep decay)

X-ray plateaus ( 100-600s → 1-10 ks) plateau no spectral evolution at plateau end β x1 = β x2 → no spectral break crosses X-ray Plateaus require a departure from assumptions of standard blast-wave model: 1. constant kinetic energy (but variable before deceleration or if ejecta are anisotropic) 2. constant micro-physical parameters (contrived)

Plateaus from increasing average dE/d Ω over visible 1/ Γ area 1. energy injection (Nousek et al 06, Zhang et al 06, AP et al 06)

Plateaus from increasing average dE/d Ω over visible 1/ Γ area 2. structured outflow (e.g. Eichler & Granot 06)

d(dE/d Ω )/ dt > 0 model decoupled optical & X-ray light-curves cannot be explained by energy injection alone because EI alters dynamics of forward-shock, hence resulting light-curve features should be achromatic

Possible reasons for X-ray and optical decoupled light-curves 1. X-ray = reprocessed synchrotron forward-shock emission scattered in another part of rel. outflow = bulk-scattering (to be continued) 2. X-ray (& optical ?) emission is (are) from 2a. a long-lived reverse shock (Uhm & Beloborodov 07 ) 2b. long-lived internal shocks (Ghisellini et al 07) Note : All require long-lived engine, producing rel. outflow for t source ~ t aglow >10 6 s

Bulk-scattering - relativistic effects rel. boost of specific flux and photon energy by γ & Γ/γ

Unifying forward-shock model for X-ray plateaus Plateaus require existence of an outflow behind forward-shock (FS) either for energy injection or for scattering Scattering negligible rel to FS Scattering dominant rel to FS achromatic light-curve breaks chromatic light-curve breaks O=Synchrotron, X=Sy O=Sy, X=inverse-Compton O&X decays well correlated O&X decays correlated O&X decays decoupled

Early optical emission from reverse shock (RS) AP & Meszaros (1998) first optical flash (1999) excess emission from RS during early afterglow sy-FS Ejecta energized by RS, followed by adiabatic cooling: F ν α ν − β t − α ( β ) with α = (0.67/0.80) + (1.19/1.47) β α = 2.5 for β =1.5 (homogeneous CBM) or β =1.2 (wind) (but optical β at 100-1000s not known for 990123 & 021211)

Measurements of optical spectral slope β during early afterglow enable test of RS expectation : α = 3/4 + 4 β /3 061126: Perley et al 07 080319B: Wozniak et al 08, Bloom et al 08 α 1 =1.96 β 1 =0.89 ? α 2 =0.76 α 2 =1.23(.02) β 2 =0.50? β 2 =0.0+0.2*log(t/1ks) α 1 =2.53 β 1 =0.63 hardening hardening softening α 1 & β 1 consistent with RS model early decay too fast for RS model Note : fast-decaying early emission is softer than slower-decaying late emission, indicating different origins ( early =RS, late =FS)

CONCLUSIONS 1. Confirmed predictions of relativistic blast-wave model (RS or FS) a . power-law afterglow spectra F ν α ν − β ( seen in optical and X-ray) b . power-law flux decay F ν α t − α ( β ) (seen in radio, optical and X-ray) c . optical flashes from RS (very rare, RS emission present until 1ks, afterwards FS) d . light-curve jet breaks ( achromatic breaks, very few cases) 2. Early ( <10 ks ) X-ray LC breaks at end of plateau due to a. achromatic breaks : long-lived injection of energy into FS b. chromatic X-ray LC breaks : “external” scattering in outflow inner to FS or “central engine” (e.g. internal shocks) emission dominant over FS

X-ray plateaus and chromatic lc breaks from reverse-shock Uhm & Beloborodov 07 ( ν c between O & X) Genet et al 07 ( ν p between O & X) resulting lcs distribution of ejecta mass with LF O X X O in contradiction with hardening of optical emission of 061126 & 080319B at B/C (suggesting different emission mechanisms at A-B and B-D)