Phase-space mixing of multiple stellar populations in globular clusters

¹ Department of Astrophysics, University of Vienna, Türkenschanzstrasse 17, 1180 Vienna, Austria
² Department of Astronomy, Indiana University, Bloomington, Swain West, 727 E. 3rd Street, IN 47405, USA
³ INAF – Astrophysics and Space Science Observatory Bologna, Via Gobetti 93/3, 40129 Bologna, Italy

^★ Corresponding author: francisco.aros@univie.ac.at

Received: 25 November 2024
Accepted: 1 May 2025

Abstract

Context. Globular clusters (GCs) host multiple populations characterised by abundance variations in a number of light elements. In many cases, these populations also exhibit spatial and/or kinematic differences, which vary in strength from cluster to cluster and tend to decrease with the clusters’ dynamical ages.

Aims. In this work, we aim to study the dynamical mixing of multiple populations and establish a link between the more theoretical aspects of the mixing process and various observational parameters that quantify differences between the populations’ spatial concentration and velocity anisotropy.

Methods. We follow the dynamical mixing of multiple populations in a set of numerical simulations through their distribution in the energy and angular momentum phase space and quantify the evolution of their degree of dynamical mixing.

Results. We present the degree of dynamical mixing traced by the intrinsic differences in the phase-space distribution of the populations. We compare the differences in phase space with three observable quantities that describe the degree of mixing in the structural and kinematic differences of the populations: A⁺, commonly used in the literature for spatial differences; and we introduce two new parameters, ΔA_β which traces the difference in velocity anisotropy, and σ_Lz, which traces the angular momentum distribution of stars.

Conclusions. Our study provides new insights into the dynamics of phase-space mixing of multiple populations in GCs. We show that differences between the first (1P) and second (2P) populations observed in old clusters contain key information on the cluster’s dynamics, as well as the 1P and the 2P spatial and kinematic properties set by the formation processes. However, caution is necessary in using the strength of the present-day differences to quantitatively constrain those imprinted at the time of formation.