1. Dimitrios Giannios Purdue Workshop, May 12th 2014 Sironi L. and Giannios D. 2014, ApJ in press, arXiv:1312.4538 Is the IGM heated by TeV blazars?

2. TeV blazars Cerenkov Telescopes: Blazars dominate the extragalactic TeV sky (Ghisellini et al 11) (credit: TEVCat) The blazar sequence: a continuous sequence LBL - IBL - HBL TeV blazars are dim

3. TeV photons are absorbed in the IGM TeV photons from blazars pair-produce in the IGM by interacting with ~ eV EBL photons. mean free path is ~100 Mpc The beam of electron-positron pairs has: Lorentz factor     and density ratio     (wrt the IGM plasma) These pairs should IC scatter off the CMB, producing ~ GeV photons. mean free path is ~ 100 kpc (IC cooling length)

4. No excess GeV emission from blazars Every TeV blazar should have a GeV halo of reprocessed light. However, not seen! (Neronov & Vovk 10)

5. IGM fields or plasma instabilities? 2) The pair energy is deposited into the IGM by plasma instabilities (Broderick, Chang, Pfrommer 12, 13) 1) IGM magnetic fields deflect the streaming pairs (Neronov & Vovk 10, Tavecchio et al. 11) Every TeV blazar should have a GeV halo of reprocessed light. However, not seen! Two possibilities: (Tavecchio et al. 11) Fermi upper limits reprocessed GeV emission from pairs deflected by IGM fields intrinsic TeV spectrum absorbed TeV spectrum

6. Plasma instabilities in the IGM Interpenetrating beams of charged particles are unstable (beam-plasma instabilities) Blazar-induced relativistic pairs IGM plasma Two-stream (bump on tail) instability energy from waves to particles: → damping energy from particles to waves: → instability microscopic scales! Oblique instability beam (Sironi & Giannios 14)

7. Beam-plasma linear evolution Linear analysis: the oblique instability grows 10-100 times faster than the IC cooling time. Is the instability growing at the linear rate until it deposits all the beam energy into the IGM? The non-linear evolution of the beam-plasma system requires PIC simulations... (Broderick et al. 12) IF the instability grows until all the beam energy is deposited into the IGM: No reprocessed blazar GeV emission IGM field estimates are invalid IGM heating from blazars will have cosmological implications (Chang et al. 12)

8. The PIC method Particle-in-Cell (PIC) method: 1. Particle currents deposited on a grid Electromagnetic fields solved on the grid via Maxwell’s equations Lorentz force interpolated to particle locations No approximations, plasma physics at a fundamental level Tiny length and time scales need to be resolved  huge simulations, limited time coverage Relativistic 3D e.m. PIC code TRISTAN-MP (Buneman ‘93, Spitkovsky ‘05) Yee mesh

9. Cold beam: non-linear evolution The oblique instability grows fast, but it is quenched by self-heating of the beam heating fraction Exponential phase Blazar-induced beams: Lorentz factor     and density ratio     COLD beam with  and   In the end, the beam longitudinal dispersion ~0.2 , and the plasma heating fraction ~10% Relaxation phase heating fraction B energy E energy

10. 10% in heat, 90% in GeV emission Blazar-induced beams: Lorentz factor     and density ratio     Numerically tractable: Lorentz factor     and density ratio     COLD beams: Regardless of the beam  or , the beam longitudinal dispersion reaches ~0.2 , and the IGM heating fraction ~10%. Only 10% of the beam energy is deposited into the IGM, 90% is still available to power the reprocessed GeV emission. IGM heating fraction (LS & Giannios 14)

11. Blazar beams are not cold The heating fraction can be ≪ 10%:  if the initial longitudinal beam dispersion is already > 0.2   IGM heating fraction (Sironi & Giannios 14) (Miniati et al 13) Blazar beams are born warm: the pair production cross section peaks at ~ few m e c 2. the TeV blazar spectrum and the EBL spectrum are broad. distance

12. Is the IGM heated by TeV blazars? Not much.

13. At the end of the relaxation phase, the beam-plasma system is still highly anisotropic, so still unstable (to the Weibel instability). Blazar-induced pair beams might be a potential mechanism for generating small-scale (~ c/ω p ~ 10 8 cm) magnetic fields in cosmic voids? Beam-aligned electric fieldMagnetic energy beam z [c/   ] y [c/   ] x [c/   ] z [c/   ] y [c/   ] x [c/   ] Long term beam-plasma evolution (Sironi & Giannios, in prep.)

14. TeV photons from blazars will pair-produce in the IGM. The resulting electron-positron beam is unstable to the excitation of plasma instabilities. Electrostatic plasma instabilities deposit ≪ 10% of the beam energy into the IGM. Most of the beam energy will result in GeV emission by IC scattering off the CMB. After the saturation of electrostatic plasma instabilities, the beam is still anisotropic, and it can generate magnetic fields from scratch via the Weibel instability. Summary

16. 10% in heat: a generous upper limit The heating fraction can be ≪ 10%:  if the initial longitudinal beam dispersion is already > 0.2    if pre-existing magnetic fields are dispersing the beam sideways. (Miniati et al 13) suppressed in the presence of density inhomogeneities in the IGM. → suppression

17. Beam distribution function

18. The complete evolution

21. Dependence on the beam properties

22. Dependence on the beam temperature