Super-Eddington accretion in high-redshift black holes and the emergence of jetted AGN

Abstract: In this work we study the co-evolution of central black holes (BHs) and host galaxies by utilizing an advanced iteration of the DELPHI semi-analytical model of galaxy formation and evolution. Based on dark matter halo merger trees spanning the redshift range from $z=20$ to $z=4$, it now incorporates essential components such as gas heating and cooling, cold and hot BH accretion, jet and radiative AGN feedback. We show how different BH growth models impact quasar and galaxy observables at $z \geq 5$, providing predictions that will help discriminate between super-Eddington and Eddington-limited accretion models: despite being both consistent with observed properties of SMBHs and their host galaxies at $z \sim 5-7$, they become very clearly distinguishable at higher redshift and in the intermediate mass regime. We find that the super-Eddington model, unlike the Eddington-limited scenario, predicts a gap in the BH mass function corresponding to the intermediate-mass range $10^4 \mathrm{M_\odot} < M_\mathrm{bh} < 10^6 \mathrm{M_\odot}$. Additionally, it predicts black holes up to two orders of magnitude more massive for the same stellar mass at $z=9$. The resulting velocity dispersion - BH mass relation at $z \geq 5$ is consistent with local measurements, suggesting that its slope and normalisation are independent of redshift. Depending on the Eddington ratio, we also model the emergence of AGN jets, predicting their duty cycle across as a function of BH mass and their potential impact on the observed number density distribution of high-redshift AGN in the hard X-ray band.