Shining in the dark: the spectral evolution of the first black holes

Massive black hole (MBH) seeds at redshift z 10 are now thought to be key ingredients to explain the presence of the supermassive (10 9-10 M ) black holes in place <1 Gyr after the big bang. Once formed, massive seeds grow and emit copious amounts of radiation by accreting the left-over halo gas; their spectrum can then provide crucial information on their evolution. By combining radiation-hydrodynamic and spectral synthesis codes, we simulate the timeevolving spectrum emerging from the host halo of a MBH seed with initial mass 10 5 M , assuming both standard Eddington-limited accretion, or slim accretion discs, appropriate for super-Eddington flows. The emission occurs predominantly in the observed infrared-submm (1-1000 m) and X-ray (0.1-100 keV) bands. Such signal should be easily detectable by JWSTaround 1 m up to z 25, and by ATHENA (between 0.1 and 10 keV, up to z 15). Ultra-deep X-ray surveys like the Chandra Deep Field South could have already detected these systems up to z 15. Based on this, we provide an upper limit for the z 6 MBH mass density of 2.5 10 2 M Mpc -3 assuming standard Eddington-limited accretion. If accretion occurs in the slim disc mode the limits are much weaker, 7.6 10 3 M Mpc -3 in the most constraining case.