X-ray Spectra from Magnetar Candidates

Transcription

1 X-ray Spectra from Magnetar Candidates A Twist in the Field R Turolla Department of Physics University of Padova, Italy With L Nobili, S Zane, N. Sartore GL Israel, N Rea

2 SGRs and AXPs X-ray Spectra SGR (Mereghetti et al 2005) AXP 1RXS J Mereghetti et al 2006 (Rea et al 2005) What is the physical interpretation? kev emission usually modelled by an absorbed blackbody (kt ~ 0.5 kev) plus power-law (! ~ 1 4)

3 Twisted Magnetospheres I! The magnetic field inside a magnetar is wound up! The presence of a toroidal component induces a rotation of the surface layers! The crust tensile strength resists! A gradual (quasi-plastic?) deformation of the crust! The external field twists up (Thompson, Lyutikov & Kulkarni 2002)

4 Twisted Magnetospheres - II! TLK02 investigated force-free, axisymmetric magnetic equilibria!! A sequence of models labeled by p or the twist angle

5 Twisted Magnetospheres - III! Twisted magnetospheres are threaded by currents! Charged particles provide large optical depth to resonant cyclotron scattering! Because and, a power- law tail expected instead of an absorption line!, and! Both angle and increase with the twist

6 A Monte Carlo Approach! Follow individually a large sample of photons, treating probabilistically their Preliminary interactions investigation with charged (1D) particles by Lyutikov & Basic Gavriil ingredients: (2006)! Can handle very general (3D) geometries More " Space detailed and energy modeling distribution by Fernandez of the &! Thompson Quite scattering easy (2007) to particles code, fast! New, Ideal " Same up-to-dated for for purely the seed scattering code (primary) (Nobili, media Turolla, photons " Scattering cross sections! Zane Monte 2007) Carlo techniques work well when scat! 1 N scat

7 Magnetospheric Currents! Charges move along the field lines! Spatial distribution Electron contribution only 1D relativistic Mawellian at T e centred at v bulk! Particle motion characterized by a bulk velocity, v bulk, and by a velocity spread "v (Beloborodov & Thompson 2007)! There may be e ± in addition to e-p,, but no detailed model as yet

8 Surface Emission The star surface is divided into patches by a cos # $ grid Each patch has its own temperature to reproduce different thermal maps Blackbody (isotropic) emission

9 Photons in a Magnetized Medium! Magnetized plasma is anisotropic and birefringent, radiative processes sensitive to polarization state! Two normal, elliptically polarized modes in the magnetized vacuum+cold plasma! At the modes are almost linearly polarized The extraordinary (X) and ordinary (O) modes

10 Scattering Cross Sections - I! QED cross section for magnetic Compton scattering available (e.g. Herold Herold 1979; Harding & Daugherty 1991; Araya & Harding 1999; Gonthier et al 2000)! Because of charge motion resonance at Completely differential cross sections at resonance (ERF) d% d' ' d% d' ' r 0 3$ r0c = # (" &" c )cos 8 3$ r0c = # (" &" c ) 8 O& O O& X! Non-relativistic (Thompson) cross section = X & X e 2 / mc 2," c! cos velocity before and after scattering 2 2! ' d% d' ' d% d' ' X & O = eb / mc,!,! ' angles between photon direction and particle 3$ r0c = # (" &" c )cos 8 3$ r0c = # (" &" c )cos 8 2 2!! '

11 Scattering Cross Sections - II! Through repeated scatterings photons may gain enough energy to Violate the condition % << m e c 2 /& Scatter in regions where B ~ B QED! Hard tails produced by up-scattering onto high- energy (non-thermal) electrons (Baring & Harding 2007)? Complete treatment of magnetic Compton scattering highly desirable

12 Scattering Cross Sections - III

13 Nuts and Bolts Generate uniform deviate R, scatter occurs when! = -ln R Generate second deviate R 1 to decide if polarization switching Generate third deviate R 2 to pick up electron velocity (if v 1,2 >0) Generate to further deviates R 3 and R 4 to decide photon direction after scattering

14 Model Spectra Model parameters: "' N-S, B pole, T e, v bulk Surface emission geometry, geometrical angles ((, )) Model archives (fixed B pole ) " Emission from entire star surface at T & " Thermal energy a given 10fraction 14 G of bulk kinetic energy " Geometry included at the post-processing level (a MC run contains all viewing/inclination angles and spin phases) " Total of ~ 500 runs, ~ models G twist increases hardness increases

15 Phase-averaged spectra (B pole = G)

16 Spectral Fitting Model archive with B pole = G completed and implemented in XSPEC (with N.Rea)) Applications to AXPs under way Onset of hard tail

17 Conclusions & Future Developments! Twisted magnetosphere model, within magnetar scenario, in general agreement with observations! Resonant scattering of thermal, surface photons produces spectra with correct properties! Many issues need to be investigated further Twist of more general external fields Detailed models for magnetospheric currents More accurate treatment of cross section including QED effects and electron recoil (in progress) kev tails: up-scattering by (ultra)relativistic( (e ± ) particles? fit of model spectra to observations (in progress)

18 Model Spectra - II Line of sight effects Emission from a single patch at the equator LOS LOS at at the opposite same longitudeof the patch

19 Post-Flare Evolution! After the GF SGR persistent X-ray emission is softer and spin- down rate smaller! Evidence for an untwisting of the magnetosphere

20 Part I: Observational Facts (mainly) Part II: Theoretical Implications (and Speculations ) Soft Gamma Repeaters are ULTRA-MAGNETIZED NEUTRON STARS, i.e. MAGNETARS

21 SGRs and AXPs X-ray Spectra - II! kt BB ~ 0.5 kev,, does not change much in different sources! Photon index *! 1 4, AXPs tend to be softer! SGRs and AXPs persistent emission is variable (months/years)! Variability mostly associated with the non- thermal component

22 Hard X-ray Emission INTEGRAL revealed substantial emission in the kev band from SGRs and APXs Mereghetti et al 2006 Hard power law tails with *! 1-3, hardening wrt soft X-ray emission required in AXPs Hard emission pulsed

23 No Generate a uniform deviate 0<R<1 No Select seed photon (energy and direction) Select particle from distribution Transform photon Advance energy photon, and direction to ERF Escape Compute? photon compute energy after depthscattering Compute new photon direction Yes R? Transform back to LAB Compute scattering Yes Store data