Pulsar Winds. John Kirk. Max-Planck-Institut für Kernphysik Heidelberg, Germany. < > p.1/18

Transcription

1 Pulsar Winds John Kirk Max-Planck-Institut für Kernphysik Heidelberg, Germany < > p.1/18

2 About 50 years after... The Crab Nebula Central star is source of particles and magnetic field (Piddington 1957) and waves (Rees & Gunn 1974). Few particles: magnetic dipole radiation? Damping propagation only for ω pe < Ω. For Crab: r > 10 8 r L (e.g., Melatos & Melrose 1996) Many particles, MHD wind + shock < > p.2/18

3 Topics The wind nebula connection MHD simulation Acceleration of the wind Dissipation in shocks/current sheets Observation of the wind Optical pulse shapes and polarisation TeV emission binary system PSR B Acceleration of particles Two mechanisms? < > p.3/18

4 Wind Nebula Connection 2D relativistic MHD by at least three groups Komissarov & Lyubarsky 2003; Khangoulian & Bogovalov 2003; Del Zanna et al 2004 Key ingredients: relativistic, anisotropic wind (power sin 2 θ) low magnetisation σ (at least near equator) < > p.4/18

5 Implications for the wind Radial flow, toroidal magnetic field Jet formed downstream of termination shock Low value of σ 0.03 No constraints on µ parameter (= L/Ṁc2 ) from the dynamics Problems with the inner ring (knots, front/back brightness ratio) may reflect kinetic effects, such as proposed for the wisps (Gallant & Arons 1994; Spitkovsky & Arons 2000) < > p.5/18

6 Axisymmetric winds Exact solution for force-free, split monopole (Michel 1973): no collimation, B φ sin θ/r (no closed field lines) Super-(magneto)sonic flow: Γ constant (Bogovalov 1997) σ = B2 /8π Γnmc 2 = constant the σ problem < > p.6/18

7 Possible solutions to the σ problem Accelerate the wind: Collimation? Not for monopole-like flows (e.g., Bogovalov & Tsinganos 1999) but in principle possible (Vlahakis 2004) Dissipation? Oblique rotator (Coroniti 1990) and damping of wave component how fast? Problem not really a problem: σ still high after the shock (Begelman 1998)? Difficult to recover nice pictures... the (striped) field dissipates in the termination shock (Lyubarsky 2003) Transition must remain thin < > p.7/18

8 Acceleration of the wind Dissipation forced by charge starvation (B 1/r, n 1/r 2 ) Entropy wave or FMS wave (small wavelength approx. r r L ) Lyubarsky & Kirk 2001; Lyubarsky 2003; < > p.8/18

9 Acceleration of the wind Dissipation forced by charge starvation (B 1/r, n 1/r 2 ) Entropy wave or FMS wave (small wavelength approx. r r L ) Lyubarsky & Kirk 2001; Lyubarsky 2003; Dissipation accel. for Γ > σ < > p.8/18

10 Current sheets Magnetic pressure balanced by hot plasma in sheet. Key question: What controls the dissipation rate? < > p.9/18

11 Modelling the dissipation Short wavelength approximation (Kirk & Skjæraasen 2003) Slow dissipation Tearing-mode Fast Coroniti (1980); Michel (1994); Lyubarsky & Kirk (2001) Lyubarsky (1996) Drenkhahn & Spruit (2002) Γ r 1/2 Γ r 5/12 Γ r 1/3 r max = r ˆL 1/2 r max = µ 4/5 L r ˆL3/10 r max 4/5 = µ ˆL3/10 L r L ˆL = L(π 2 e 2 /m 2 c 5 ), (= for Crab) No consistent conversion mechanism for µ > 10 ˆL 1/4 < > p.10/18

12 Observation of the wind? Gamma-rays from unshocked wind Targets from companion star: swamped by emission from shocked wind (Ball & Kirk 2000) Targets from neutron star surface: acceleration not permitted for r < 5r L in the Crab (Aharonian & Bogovalov 2000, 2003) < > p.11/18

13 Observation of the wind? Gamma-rays from unshocked wind Targets from companion star: swamped by emission from shocked wind (Ball & Kirk 2000) Targets from neutron star surface: acceleration not permitted for r < 5r L in the Crab (Aharonian & Bogovalov 2000, 2003) Optical pulses from unshocked wind if r emission < Γ 2 r L (Pétri) < > p.11/18

14 Observation of the wind? Gamma-rays from unshocked wind Targets from companion star: swamped by emission from shocked wind (Ball & Kirk 2000) Targets from neutron star surface: acceleration not permitted for r < 5r L in the Crab (Aharonian & Bogovalov 2000, 2003) Optical pulses from unshocked wind if r emission < Γ 2 r L (Pétri) TeV emission from shocked wind in PSR B Hadronic emission (Kawachi et al 2004) Inverse Compton model < > p.11/18

15 But L s.d. L Be wind < > p.12/18 Unique pulsar/be star binary X-rays from interacting winds Tavani & Arons (1997) Be star wind: M 10 8 M yr 1 v wind 10 3 km s 1 Pulsar wind: L s.d erg s 1 Momentum balance: r Be r p = L s.d. Ṁv wind c 0.7 periastron epoch τ τ+15 days 03/2004 H.E.S.S. H.E.S.S. 02/2004 SS 2883 H.E.S.S. wind interface stellar disc? 04/2004 pulsar orbit period = 3.4 years eccentricity = 0.87 PSR B τ 18 days wind flow line H.E.S.S. 05/2004 H.E.S.S. 06/2004 Flux (>380 GeV) cm 2 s 1 Based on cartoon by O. de Jager

16 PSR B : predicted spectrum Kirk, Ball, Skjæraasen (1999) < > p.13/18

17 PSR B : predicted spectrum Kirk, Ball, Skjæraasen (1999) B from P, P and r p r Be B 2 /8π 0.1U rad < > p.13/18

18 PSR B : predicted spectrum Kirk, Ball, Skjæraasen (1999) B from P, P and r p r Be B 2 /8π 0.1U rad Normalisation from ASCA/OSSE points < > p.13/18

19 PSR B : predicted spectrum Kirk, Ball, Skjæraasen (1999) B from P, P and r p r Be B 2 /8π 0.1U rad Normalisation from ASCA/OSSE points Anistropic IC targets: δ-function approx. < > p.13/18

20 PSR B : predicted spectrum Kirk, Ball, Skjæraasen (1999) Observations with H.E.S.S. telescopes Schlenker et al (2005), Aharonian et al astro-ph/ B from P, P and r p r Be B 2 /8π 0.1U rad Normalisation from ASCA/OSSE points Anistropic IC targets: δ-function approx. < > p.13/18

21 Degeneracy TeV photon index Electron injection rate d ln N d ln ɛ d ln Q d ln γ 2.7 = q < > p.14/18

22 Degeneracy TeV photon index Electron injection rate d ln N d ln ɛ d ln Q d ln γ 2.7 = q Either q 1.4 (synchrotron/ic cooling important) < > p.14/18

23 Degeneracy TeV photon index Electron injection rate d ln N d ln ɛ d ln Q d ln γ 2.7 = q Either q 1.4 (synchrotron/ic cooling important) Or q 2.4 (adiabatic losses dominate) < > p.14/18

24 Degeneracy TeV photon index d ln N d ln ɛ 2.7 Electron injection rate d ln Q d ln γ = q Either q 1.4 (synchrotron/ic cooling important) Or q 2.4 (adiabatic losses dominate) Connect with acceleration theory? < > p.14/18

25 Crab Nebula ν 1.1 Atoyan (1999) < > p.15/18

26 Crab Nebula ν 1.1 ν 0.3 Atoyan (1999) < > p.15/18

27 log[νfν] Two mechanisms? log(ν) Q(γ) γ q < > p.16/18

28 Two mechanisms? log[νfν] Fermi I relativistic shock q 2.2 log(ν) Q(γ) γ q < > p.16/18

29 Two mechanisms? log[νfν] q 1.4 Fermi I relativistic shock q 2.2 log(ν) Q(γ) γ q < > p.16/18

30 Two mechanisms? log[νfν] q 1.4 Fermi I relativistic shock q 2.2 Shock width gyro radius ν crit log(ν) Q(γ) γ q < > p.16/18

31 Two mechanisms? log[νfν] q 1.4 Fermi I relativistic shock q 2.2 Shock width gyro radius cooling ν crit log(ν) Q(γ) γ q < > p.16/18

32 Two mechanisms? log[νfν] q 1.4 maser? cooling Fermi I relativistic shock q 2.2 Shock width gyro radius resonant cyclotron absorption (p e + )? Hoshino et al 1992 Amato 2004 ν crit log(ν) Q(γ) γ q < > p.16/18

33 Two mechanisms? log[νfν] maser? reconnection? q 1.4 cooling Fermi I relativistic shock q 2.2 Shock width gyro radius resonant cyclotron absorption (p e + )? Hoshino et al 1992 Amato 2004 ν crit log(ν) Q(γ) γ q < > p.16/18

34 PSR : modelling 1. Radiative losses at periastron Crab-type injection dn/dγ γ 1.6 at γ < Γ peak Adiabatic loss timescale > 15D/c 10% efficiency INTEGRAL: Shaw et al 2004 H.E.S.S Γ wind < > p.17/18

35 PSR : modelling 1. Radiative losses at periastron Crab-type injection dn/dγ γ 1.6 at γ < Γ peak Adiabatic loss timescale > 15D/c 10% efficiency INTEGRAL: Shaw et al 2004 H.E.S.S Γ wind < > p.17/18

36 PSR : modelling 2. Adiabatic losses at periastron Crab-type injection dn/dγ γ 2.2 at γ > Γ peak INTEGRAL: Shaw et al 2004 H.E.S.S Adiabatic loss timescale < 0.5D/c high efficiency ( 100%) Γ wind < > p.17/18

37 PSR : modelling 2. Adiabatic losses at periastron Crab-type injection dn/dγ γ 2.2 at γ > Γ peak INTEGRAL: Shaw et al 2004 H.E.S.S Adiabatic loss timescale < 0.5D/c high efficiency ( 100%) Γ wind < > p.17/18

38 Knowledge, Ignorance, Striving < > p.18/18

39 Knowledge, Ignorance, Striving Acceleration of the wind < > p.18/18

40 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab < > p.18/18

41 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? < > p.18/18

42 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? Optical polarisation studies, more observations < > p.18/18

43 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? Optical polarisation studies, more observations Acceleration of the electrons < > p.18/18

44 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? Optical polarisation studies, more observations Acceleration of the electrons 10 TeV electrons already at r = 10 3 r L < > p.18/18

45 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? Optical polarisation studies, more observations Acceleration of the electrons 10 TeV electrons already at r = 10 3 r L How? < > p.18/18

46 Knowledge, Ignorance, Striving Acceleration of the wind Low σ wind in Crab How, where? Optical polarisation studies, more observations Acceleration of the electrons 10 TeV electrons already at r = 10 3 r L How? more theory < > p.18/18