Towards a Holistic View of the Heating and Cooling of the Intracluster Medium

TL;DR

This paper presents a comprehensive model of intracluster medium heating and cooling, integrating gravitational shock heating, radiative cooling, preheating, and AGN feedback to explain observed entropy profiles and cluster dichotomy.

Contribution

It introduces a unified model incorporating preheating and AGN feedback to explain entropy profiles and the cool core/non-cool core dichotomy in galaxy clusters.

Findings

Gravitational heating explains large-radius profiles.

Preheating is necessary for NCC clusters.

AGN feedback stabilizes CC clusters against overcooling.

Abstract

(Abridged) X-ray clusters are conventionally divided into two classes: "cool core" (CC) clusters and "non-cool core" (NCC) clusters. Yet relatively little attention has been given to the origins of this dichotomy and, in particular, to the energetics and thermal histories of the two classes. We develop a model for the entropy profiles of clusters starting from the configuration established by gravitational shock heating and radiative cooling. At large radii, gravitational heating accounts for the observed profiles and their scalings well. However, at small and intermediate radii, radiative cooling and gravitational heating cannot be combined to explain the observed profiles of either type of cluster. The inferred entropy profiles of NCC clusters require that material is preheated prior to cluster collapse in order to explain the absence of low entropy (cool) material in these systems. We show that a similar modification is also required in CC clusters in order to match their properties at intermediate radii. In CC clusters, this modification is unstable, and an additional process is required to prevent cooling below a temperature of a few keV. We show that this can be achieved by adding a self-consistent AGN feedback loop in which the lowest-entropy, most rapidly cooling material is heated so that it rises buoyantly to mix with material at larger radii. The resulting model does not require fine tuning and is in excellent agreement with a wide variety of observational data. Some of the other implications of this model are briefly discussed.