TIDAL DISRUPTION, GLOBAL MASS FUNCTION, AND STRUCTURAL PARAMETER EVOLUTION IN STAR CLUSTERS

We present a unified picture for the evolution of star clusters on the two-body relaxation timescale. We use direct N-body simulations of star clusters in a galactic tidal field starting from different multimass King models, up to 10% of primordial binaries and up to Ntot = 65, 536 particles. An additional run also includes a central Intermediate Mass Black Hole. We find that for the broad range of initial conditions we have studied the stellar mass function of these systems presents a universal evolution, which depends only on the fractional mass loss. The structure of the system, as measured by the core to half-mass radius ratio, also evolves toward a universal state, which is set by the efficiency of heating on the visible population of stars induced by dynamical interactions in the core of the system. Interactions with dark remnants (white dwarfs, neutron stars, and stellar mass black holes) are dominant over the heating induced by a moderate population of primordial binaries (3%-5%), especially under the assumption that most of the neutron stars and black holes are retained in the system. All our models without primordial binaries undergo a deep gravothermal collapse in the radial mass profile. However, their projected light distribution can be well fitted by medium concentration King models (with parameter W0 ∼ 8), even though there tends to be an excess over the best fit for the innermost points of the surface brightness. This excess is consistent with a shallow cusp in the surface brightness (μ ∼ R -ν with ν ∼ 0.4-0.7), like it has been observed for many globular clusters from high-resolution Hubble Space Telescope imaging. Generally, fitting a King profile to derive the structural parameters yields to larger fluctuations in the core size than defining the core as the radius where the surface brightness is one half of its central value. Classification of core-collapsed globular clusters based on their surface brightness profile may thus fail in systems that appear to have already bounced back to lower concentrations, particularly if the angular resolution of the observations is limited and the core is not well resolved.