The impact of active galactic nuclei on cooling flows

Abstract

This thesis explores the role of active galactic nuclei (AGN) in the heating of the intracluster medium (ICM). In the centre of many clusters the radiative cooling tune of the ICM is much shorter than the Hubble time. Unless cooling is balanced by some form of heating, gas will flow into the cluster centre at rates up to ~ 1000 M© yr(^-1). Recent Chandra and XMM-Newton X-ray observations present almost no evidence that this is happening incluster cores. Moreover, they show that the ICM has a rather complex structure. Some of the features in the X-ray images can be explained as the interaction of the central AGN with the ICM. The most prominent feature are bubbles of hot and underdense gas inflated by jets coming from massive black holes residing in the centre of giant elliptical galaxies. These bubbles are thought to rise buoyantly through the ICM and heat the gas by depositing their energy. I start by introducing the cooling-flow problem and by summarising the current understanding of the ICM heating processes. I then introduce the adaptive mesh refinement (AMR) code FLASH that has been used for the simulations in this thesis and the development of new routines and modules. A model of AGN heating is then applied to model clusters to investigate three issues: (1) the quenching of the cooling-flow by injection of bubbles of energy; (2) the determination of the AGN duty cycle by using measurements of sound wave positions; (3) the presence of a mass threshold below which the heating process is no longer effective. I show that cooling can be effectively balanced by AGN heating in a cluster of mass 3 x lO(^14) Mo. Then, I argue that by using measurements of sound wave positions it is possible to determine the duty cycle of the AGN with good accuracy. Finally, I show that there is a threshold mass for which the heating process is ineffective. In the light of this, I discuss the importance of the process in shaping the luminosity function of galaxies. I also apply the heating model to a cluster that has formed in a cosmological environment and discuss how to improve the code performance.