The WISSH quasars project I. Powerful ionised outflows in hyper-luminous quasars

Models and observations suggest that both the power and effects of AGN feedback should be maximised in hyper-luminous (L-Bol > 10(47) erg s(-1)) quasars, i.e. objects at the brightest end of the AGN luminosity function. In this paper, we present the first results of a multiwavelength observing programme, focusing on a sample of WISE/SDSS selected hyper-luminous (WISSH) broad-line quasars at z approximate to 1.5-5. The WISSH quasars project has been designed to reveal the most energetic AGN-driven outflows, estimate their occurrence at the peak of quasar activity, and extend the study of correlations between outflows and nuclear properties up to poorly investigated, extreme AGN luminosities, i.e. L-Bol similar to 10(47) - 10(48) erg s(-1). We present near-infrared, long-slit LBT/LUCI1 spectroscopy of five WISSH quasars at z approximate to 2.3 - 3.5, showing prominent [OIII] emission lines with broad (FWHM similar to 1200-2200 km s(-1)) and skewed profiles. The luminosities of these broad [OIII] wings are the highest measured so far, with L-[OIII](broad) greater than or similar to 5 x 10(44) erg s(-1), and reveal the presence of powerful ionised outflows with associated mass outflow rates. (M) over dot greater than or similar to 1700 M-circle dot yr(-1) and kinetic powers. (E)over dot(kin) greater than or similar to 10(45) erg s(-1). Although these estimates are affected by large uncertainties because of the use of [OIII] as a tracer of ionised outflows and the very basic outflow model adopted here, these results suggest that in our hyper-luminous targets the AGN is highly efficient at pushing large amounts of ionised gas outwards. Furthermore, the mechanical outflow luminosities measured for WISSH quasars correspond to higher percentages (similar to 1-3%) of L-Bol than those derived for AGN with lower L-Bol. Our targets host very massive (M-BH greater than or similar to 2 x 10(9) M-circle dot) black holes that are still accreting at a high rate (i.e. a factor of similar to 0.4-3 of the Eddington limit). These findings clearly demonstrate that WISSH quasars off er the opportunity to probe the extreme end of both luminosity and supermassive black holes (SMBH) mass functions and revealing powerful ionised outflows that are able to affect the evolution of their host galaxies.