Cosmic Ray acceleration in clusters of galaxies. Gianfranco Brunetti. Istituto di Radioastronomia INAF, Bologna, ITALY

Transcription

1 Cosmic Ray acceleration in clusters of galaxies Gianfranco Brunetti Istituto di Radioastronomia INAF, Bologna, ITALY

2 Outline NT components (CRe,CRp,B) in galaxy clusters : observations Physics and dynamics of CR in galaxy clusters Present constraints on CR protons **new** Physics of NT Mpc scale diffuse radio emission from galaxy clusters: Relics & Halos Shocks and turbulent acceleration in clusters **Future** at low radio frequencies & gamma rays

3 Clusters of galaxies: the largest gravitational structures in the Universe (M M sun, R V 2-3 Mpc) Galaxy cluster mass: Coma Cluster Barions 10% of stars in galaxies 15-20% of hot diffuse gas Dark Matter 70%

4 Clusters of galaxies: the largest gravitational structures in the Universe (M M sun, R V 2-3 Mpc) n 10-3 cm -3 Galaxy cluster mass: galaxies Coma Cluster T K Barions 10% of stars in galaxies 15-20% of hot diffuse gas Dark Matter 70%

5 Coma Cluster 1 Mpc Feretti +al.1998

6 Coma Cluster Bridge Radio Halo Radio Relic 1 Mpc Feretti +al.1998

7 Coma Cluster: high energy NT Wik et al Radio Halo Radio Relic Fusco-Femiano et al.99 Rephaeli et al.99 Radio Relic Ajello et al. 2008; Fujita et al. 2009; Wik et al 2009; Eckert et al. 2008; Perez-Torres et al. 2009; Colafrancesco & Marchegiani 2009; Nevalainen et al 2009, Murgia et al 10, Wik et al 11

8 Inverse Compton Emission from GC?? Syn B IC ph (CMB) L rad -> (U e, U B ) -> K e B 2 L HE -> (U e, U ph ) -> K e U ph L rad /L HE U B /U ph _ -> B B > μg

9 Govoni et al 05 Bonafede et al 2010 B in Galaxy Clusters (also Bruggen,Dolag,Neronov lectures) B few μg Λc few-50 kpc RM probe turbulent motions in the IGM Clarke et al 2001 Murgia et al 2004

10 Injection & Dynamics of CR in GC Cosmological Shocks (e.g. Sarazin 1999, Miniati et al. 2001, Blasi 2001, Gabici & Blasi 2003, Ryu et al. 2003, Pfrommer et al. 2006, 2008, Hoeft & Bruggen 2007, Skillman et al. 2008, Vazza, Brunetti, Gheller 2009, 2010, etc..) AGN, Galactic Winds (e.g. Ensslin et al. 1998; Voelk & Atoyan 1999) Reconnection (turbulent.. Lazarian & Vishniac 99) (e.g. Brunetti & Lazarian 2011, DeGouveia dal Pino et al 2011,.. )

11 Physics of CR Leptons (de/dt) / m e c 2 = b = rate of energy losses in units of m e c 2 Photon Collisions Particle Collisions

12 Physics of CR Leptons (de/dt) ~ E / Time ~ m e c 2 b The life-time of electrons depends on quantities that can be measured

13 Physics of CR Hadrons CR protons more energetics than thermal electrons: Coulomb scattering X m = like leptons Collisions between CR & thermal protons ~30 Gyrs!

14 Das Bild kann zurzeit nicht angezeigt werden. Physics of Cosmic Rays D(GeV) cm 2 /s << cm 2 /s (Schlickeiser +al 1987, Blasi+Colafrancesco 1999, GB +al 2011 ) Diffusion time CR are confined in GC Voelk et al 1996; Berezinsky, Blasi, Ptuskin 1997; CRe CRp CR protons are long living and accumulated Voelk et al 1996; Berezinsky, Blasi, Ptuskin 1997; Ensslin et al 1998; Blasi, Gabici, Brunetti 07 CR electrons are short living particles and accumulated at γ Sarazin 1999; Petrosian 2001;

15 Acceleration & transport of CR : simulations Pfrommer et al. 2007, 08 : simulations of CR+IGM +CR transport/advection - Acceleration efficiency is the free parameter - Diffusion time >> advection time

16 Radiation from Cosmic Rays in GC B Syn Miniati 2003 CRe ph IC FERMI CRp

17 Radiation from Cosmic Rays in GC B Syn Wolfe +al 2008 CRe ph IC p-p p-γ CRp

18 Limits from gamma rays EGRET L γ,π ~ f(δ) <E CR > <E th /T> V γ Reimer +al. 2003; Pfrommer & Ensslin 2004 H.E.S.S. A 85 : Ecr/Eth < 6-15% (hard spectra) Coma : Ecr/Eth < 12% VERITAS (Perkins +al. 2008) Coma : Ecr/Eth < 5-10% (hard spectra) Aharonian + al MAGIC (Aleksic +al. 2010) Perseus : Ecr/Eth < 4% (hard spectra)

19 Gamma rays : energy content of CRp Ackermann et al 2010

20 CRp: limits from Radio Reimer et al 04, Brunetti et al. 07,08 L γ,π ~ f γ (δ) <E CR > <E th /T> V γ L R ~ f R (δ)<e CR > <E th /T> <B δ/2+1 /(B 2 +B cmb2 )>V R δ=2.1 δ=2.9 Assuming that secondary particles are injected in the IGM, their synchrotron emission should be smaller than upper limits to the diffuse radio emission. limits on : (B, E CRp ), δ N(p)=K p -δ

21 CRp: limits from Radio Reimer et al 04, Brunetti et al. 07,08 L γ,π ~ f γ (δ) <E CR > <E th /T> V γ L R ~ f R (δ)<e CR > <E th /T> <B δ/2+1 /(B 2 +B cmb2 )>V R δ=2.1 δ=2.9 Assuming that secondary particles are injected in the IGM, their synchrotron emission should be smaller than upper limits to the diffuse radio emission. limits on : (B, E CRp ), δ N(p)=K p -δ

22 Energy content of CRp RM EGRET FERMI Reimer et al. (2003) Reimer et al. (2004) Pfrommer & Ensslin (2004) Perkins et al. (2006) Brunetti et al. (2007) Brunetti et al. (2008) Perkins et al. (2008) Aharonian et al. (2008 a,b) Aleksic et al. (2009) Ackermann et al (2010) ASKAP/EMU < > Gamma + Radio observations independently suggest that non-thermal components are dynamically NOT important (% level) Additional limits from cluster dynamics (e.g. Churazov et al. 2008; Lagana et al 2009) constrain E CR +E B +E turb below 10% ( < 30% ) Ethermal.

23 Non-thermal emission from GC Both Halos & Relics have steep spectrum, F(ν)=F o ν -, with 1.3 Abell 754 Henry et al Abell 2163 Feretti et al Abell 3376 Bagchi et al Radio Relics Radio Halos

24 Non-thermal emission from GC Both Halos & Relics have steep spectrum, F(ν)=F o ν -, with 1.3 Unpolarised, follow the X-ray brightness (originate from cluster central regions) Polarised, no correlation with X-ray brightness (form in cluster outskirts) Abell 754 Henry et al Abell 2163 Feretti et al Abell 3376 Bagchi et al Radio Relics Radio Halos

25 Connecting LSS formation and merger history of IGM NT-physics B? SHOCKS accelerate e ±, p cr clusters increase their mass via merger with smaller subclusters e ±, p TURBULENCE reaccelerates fossil e ± and secondaries e ± on Mpc scales (eg., Brunetti et al. 2001, 2004, 2009; Petrosian 2001; Miniati et al. 2001; Fujita et al. 2003; Ryu et al. 2003; Gabici & Blasi 2003; Berrington & Dermer 2003; Pfrommer & Ensslin 2004; Brunetti & Blasi 2005; Cassano & Brunetti 2005; Cassano et al. 2006; Brunetti & Lazarian 2007; Hoeft & Bruggen 2007; Pfrommer et al. 2008; Petrosian & Bykov 2008, ) p cr p th π 0 γ rays π ± e ±?

26 Connecting LSS formation and merger history of IGM NT-physics B Shocks: acceleration or reacceleration? e ±, p p cr p th SHOCKS accelerate e ±, p cr π 0 γ rays? clusters increase their mass via merger with smaller subclusters TURBULENCE reaccelerates fossil e ± and secondaries e ± on Mpc scales (eg., Brunetti et al. 2001, 2004, 2009; Petrosian 2001; Miniati et al. 2001; Fujita et al. 2003; Ryu et al. 2003; Gabici & Blasi 2003; Berrington & Dermer 2003; Pfrommer & Ensslin 2004; Brunetti & Blasi 2005; Cassano & Brunetti 2005; Cassano et al. 2006; Brunetti & Lazarian 2007; Hoeft & Bruggen 2007; Pfrommer et al. 2008; Petrosian & Bykov 2008, ) π ± e ±?

27 Connecting LSS formation and merger history of IGM NT-physics B Shocks: acceleration or reacceleration? e ±, p p cr p th SHOCKS accelerate e ±, p cr π 0 γ rays? clusters increase their mass via merger with smaller subclusters TURBULENCE reaccelerates fossil e ± and secondaries e ± on Mpc scales Turbulence? (eg., Brunetti et al. 2001, 2004, 2009; Petrosian 2001; Miniati et al. 2001; Fujita et al. 2003; Ryu et al. 2003; Gabici & Blasi 2003; Berrington & Dermer 2003; Pfrommer & Ensslin 2004; Brunetti & Blasi 2005; Cassano & Brunetti 2005; Cassano et al. 2006; Brunetti & Lazarian 2007; Hoeft & Bruggen 2007; Pfrommer et al. 2008; Petrosian & Bykov 2008, ) π ± e ±?

28 Connecting LSS formation and merger history of IGM NT-physics B Shocks: acceleration or reacceleration? e ±, p p cr p th SHOCKS accelerate e ±, p cr π 0 γ rays? clusters increase their mass via merger with smaller subclusters TURBULENCE reaccelerates fossil e ± and secondaries e ± on Mpc scales Turbulence? (eg., Brunetti et al. 2001, 2004, 2009; Petrosian 2001; Miniati et al. 2001; Fujita et al. 2003; Ryu et al. 2003; Gabici & Blasi 2003; Berrington & Dermer 2003; Pfrommer & Ensslin 2004; Brunetti & Blasi 2005; Cassano & Brunetti 2005; Cassano et al. 2006; Brunetti & Lazarian 2007; Hoeft & Bruggen 2007; Pfrommer et al. 2008; Petrosian & Bykov 2008, ) π ± e ± Relevance of secondary particles??

29 Vazza et al 2009 Cosmological Shocks Natural consequence of the hierarchical process of LSS formation

30 Shocks in Galaxy Clusters Miniati et al. 2001; Ryu et al. 2003; Pfrommer et al. 2006,08; Hoeft & Bruggen 2007; Skillman et al. 2008,11 Vazza et al. 2009, 11 Vazza, Brunetti, Gheller 2009

31 Shocks in Galaxy Clusters Shocks are responsible for the heating of the IGM Markevitch et al 05

32 Shock Relic connection Bagchi+al. 2002, Science van Weeren+al. 2010, Science Abell 3667

33 Shock Acceleration: Radio Relics (Ensslin et al 1998, Roettiger et al. 1999, Sarazin 1999, ) Abell δ l R V d τe(ν) 100 kpc time/distance Log N(γ) Log(γ)

34 Width of Relics & constraints on B Markevitch et al 2005 γ c syn (ν/b) 1/2 Width of Relic V d c s M(M 2 +3)/4M 2

35 Van Weeren et al., 10 From Heft, conference in Bangalore Diffuse Relativistic Plasmas: RRI

36 Mach numbers in galaxy clusters Vazza, Brunetti, Gheller 2009 Semi-analytics : Gabici & Blasi 2003 Berrington & Dermer 2003 some agreement Pfrommer et al. 2008

37 Uncertainties in CR acceleration Vazza, Brunetti, Gheller 2009 Kang & Jones 2007 Pfrommer et al. 2006, 08 SN

38 CR acceleration or REacceleration? Kang & Jones 2007 Merger shocks have M= Reacceleration of pre-existing relativistic electrons at these shocks is efficient (eg. Kang & Ryu 11) Galaxy clusters are unique labs Kang & Jones 2002 to study particle acceleration at weak & LS shocks

39 Radio Halos as labs for CR acceleration in GC Abell 2163 Feretti et al L 2 ~ D x t 1 Mpc diffusion (Mpc) T diff (~10 10 yr) >> T cool (~10 8 yr) (eg. Jaffe 1977) in situ (re)acceleration or injection of CR electrons electrons

40 Statistics of Radio Halos Cassano et al 08 NVSS GMRT 1/10 1/3? Cassano et al.2008 NVSS data (from Giovannini et al. 1999) and deep GMRT observations. 0.41±0.11 for L x > erg/s 0.08±0.04 for L x < erg/s 1/3 1/10 (Venturi et al. 2007, 2008; Cassano et al.2008)

41 Cluster mergers - radio halos connection Venturi et al 2007,08 Brunetti et al 2007,09 Cassano et al 2010 ApJ 721 L82 The radio bimodality has a correspondence in terms of dynamical segregation

42 Das Bild kann zurzeit nicht angezeigt werden. Origin of Radio Halos: turbulence? REacceleration models (eg Brunetti et al 01, Petrosian 01, ) A mechanisms distributed on Mpc scales channels a fraction of the gravitational energy dissipated during mergers into high energy particles Turbulence?? Brunetti et al 2007 Venturi et al 2007,08 protons electrons

43 Radio Halos : are they generated by inefficient mechanism of CRe acceleration? (ν) Schlickeiser +al (Coma) Thierbach et al. 2003

44 Radio Halos : are they generated by inefficient mechanism of CRe acceleration? (ν) Schlickeiser +al (Coma) Evidence of break in the spectrum of the emitting electrons at energies of few GeV τ acc τ loss Re-acceleration Thierbach et al acceleration losses

45 Radio Halos : are they generated by inefficient mechanism of CRe acceleration? (ν) Schlickeiser +al (Coma) Acceleration time-scale 10 8 years Re-acceleration Thierbach et al acceleration losses

46 Radio Halos : are they generated by inefficient mechanism of CRe acceleration? (ν) Schlickeiser +al (Coma) Acceleration time-scale 10 8 years eg., classical Fermi II Thierbach et al > 10 7 yrs

47 Second order Fermi Mechanisms (Fermi 1949) u u + - λ Frequency of collisions: Energy gain per collisions:

48 transport diffusion

49 Stochastic particle acceleration due to particle-mode coupling (Book reviews : Melrose 1980, Berezinskii et al 1990, Schlickeiser 2002) Acceleration is sensitive to our model of turbulence Stochastic acceleration of fast particles diffusing in turbulence (Ptuskin 1988)

50 Stochastic particle acceleration due to particle-mode coupling (Book reviews : Melrose 1980, Berezinskii et al 1990, Schlickeiser 2002) Gyroresonance Acceleration is sensitive to our model of turbulence Stochastic acceleration of fast particles diffusing in turbulence (Ptuskin 1988) Transit Time Damping (TTD) ω-k // v // =0 Interaction btw magnetic moment of particle and parallel gradient of B Suitable for IGM! Isotropic fast modes (Cassano & Brunetti 05, Yan et al 10, Brunetti & Lazarian 07, 11)

51 Das Bild kann zurzeit nicht angezeigt werden. Waves spectra Primary electrons Proton spectra Secondary electrons

52 Das Bild kann zurzeit nicht angezeigt werden. Spectra of radio halos & turbulence Steepening frequency Χ 1/τ acc less efficient Mergers between M<10 15 M sun Big jumps = major mergers Small jumps = minor mergers Mergers between M>10 15 M sun more efficient GHz

53 Observed spectra of radio halos & turbulence Cassano, GB, Setti (2006) Steepening frequency Χ 1/τ acc less efficient Radio Halos with very steep spectrum in the classical radio band must exist GHz

54 Radio Power Turbulent acceleration? GMRT 240 MHz Brunetti +al 2008, Nature 455, 944 =1.9 N(E)=k E -4.8 =1.5 E e =3-5 GeV Acceleration time-scale 2.5 x 10 8 years Frequency Dallacasa et al 2009

55 LOFAR Cassano et al 2010

56 Gamma rays : energy content of CRp & origin of Radio Halos Jeltema & Profumo 2010 RM L γ,π ~ f γ (δ) <E CR > <E th /T> V γ < > L R ~ f R (δ)<e CR > <E th /T> <B δ/2+1 /(B 2 +B cmb2 )>V R L R / L γ,π -> <B δ/2+1 /(B 2 +B cmb2 )> Also Donnert et al 10, Brunetti et al

57 Bonafede et al 2010 Turbulent models and Coma SED E tur 10 % E th E CR = 3 % E -1 ~100 kpc (flat profile) Suzaku Brunetti & Lazarian 11 NHXM, ASTRO-H turbulence EGRET FERMI CTA secondaries

58 UHEp-γ Inoue et al 05 Additional processes Suzaku NHXM, ASTRO-H EGRET FERMI Pinzke & Pfrommer 10 CTA If B is smaller than that estimated from RM (reasonable?) more gamma rays

59