The most luminous blue quasars at 3.0 < z < 3.3. I. A tale of two X-ray populations

The most luminous blue quasars at 3.0 < z < 3.3. I. A tale of two X-ray populationsNardini, E.Lusso, E.Risaliti, G.Bisogni, S.Civano, FrancescaElvis, MartinFabbiano, GiuseppinaGilli, R.Marconi, A.Salvestrini, F.Vignali, C.DOI: info:10.1051/0004-6361/201936911v. 632A109
Nardini, E., Lusso, E., Risaliti, G., Bisogni, S., Civano, Francesca, Elvis, Martin, Fabbiano, Giuseppina, Gilli, R., Marconi, A., Salvestrini, F., and Vignali, C. 2019. "The most luminous blue quasars at 3.0 < z < 3.3. I. A tale of two X-ray populations." Astronomy and Astrophysics 632:A109. https://doi.org/10.1051/0004-6361/201936911
ID: 154538
Type: article
Authors: Nardini, E.; Lusso, E.; Risaliti, G.; Bisogni, S.; Civano, Francesca; Elvis, Martin; Fabbiano, Giuseppina; Gilli, R.; Marconi, A.; Salvestrini, F.; Vignali, C.
Abstract: We present the X-ray analysis of 30 luminous quasars at z ≃ 3.0 - 3.3 with pointed XMM-Newton observations (28-48 ks) originally obtained by our group to test the suitability of active galactic nuclei as standard candles for cosmological studies. The sample was selected in the optical from the Sloan Digital Sky Survey Data Release 7 to be representative of the most luminous, intrinsically blue quasar population, and by construction boasts a high degree of homogeneity in terms of optical and UV properties. In the X-rays, only four sources are too faint for a detailed spectral analysis, one of which is formally undetected. Neglecting one more object later found to be radio-loud, the other 25 quasars are, as a whole, the most X-ray luminous ever observed, with rest-frame 2-10 keV luminosities of 0.5 - 7 × 1045 erg s-1. The continuum photon index distribution, centred at Γ ̃ 1.85, is in excellent agreement with those in place at lower redshift, luminosity, and black-hole mass, confirming the universal nature of the X-ray emission mechanism in quasars. Even so, when compared against the well-known LX-LUV correlation, our quasars show an unexpectedly varied behaviour, splitting into two distinct subsets. About two-thirds of the sources are clustered around the relation with a minimal scatter of 0.1 dex, while the remaining one-third appear to be X-ray underluminous by factors of > 3-10. Such a large incidence (≈25%) of X-ray weakness has never been reported in radio-quiet, non- broad absorption line (BAL) quasar samples. Several factors could contribute to enhancing the X-ray weakness fraction among our z ≃ 3 blue quasars, including variability, mild X-ray obscuration, contamination from weak-line quasars, and missed BALs. However, the X-ray weak objects also have, on average, flatter spectra, with no clear evidence of absorption. Indeed, column densities in excess of a few ×1022 cm-2 can be ruled out for most of the sample. We suggest that, at least in some of our X-ray weak quasars, the corona might experience a radiatively inefficient phase due to the presence of a powerful accretion-disc wind, which substantially reduces the accretion rate through the inner disc and therefore also the availability of seed photons for Compton up-scattering. The origin of the deviations from the LX-LUV relation will be further investigated in a series of future studies.