2010MNRAS.405.1544D

Member galaxies within galaxy clusters nowadays can be routinely identified in cosmological, hydrodynamical simulations using methods based on identifying self bound, locally over dense substructures. However, distinguishing the central galaxy from the stellar diffuse component within clusters is notoriously difficult, and in the centre it is not even clear if two distinct stellar populations exist. Here, after subtracting all member galaxies, we use the velocity distribution of the remaining stars and detect two dynamically, well-distinct stellar components within simulated galaxy clusters. These differences in the dynamics can be used to apply an unbinding procedure which leads to a spatial separation of the two components into a cD and a diffuse stellar component (DSC). Applying our new algorithm to a cosmological, hydrodynamical simulation we find that - in line with previous studies - these two components have clearly distinguished spatial and velocity distributions as well as different star formation histories. We show that the DSC fraction - which can broadly be associated with the observed intracluster light - does not depend on the virial mass of the galaxy cluster and is much more sensitive to the formation history of the cluster. We conclude that the separation of the cD and the DSC in simulations, based on our dynamical criteria, is more physically motivated than current methods which depend on implicit assumptions on a length-scale associated with the cD galaxy and therefore represent a step forward in understanding the different stellar components within galaxy clusters. Our results also show the importance of analysing the dynamics of the DSC to characterize its properties and understand its origin.