What cluster gas expulsion can tell us about star formation, cluster environment and galaxy evolution

Violent relaxation – the protocluster dynamical response to the expulsion of its leftover star forming gas – is a short albeit crucial episode in the evolution of star clusters and star cluster systems. In this contribution, I survey how it influences the cluster age distribution, the cluster mass function and the ratio between the cluster mass and the stellar mass. I highlight the promising potential that the study of this phase holds in terms of deciphering star cluster formation and galaxy evolution, and (some of) the issues which are to be dealt with before achieving this goal.

💡 Research Summary

The paper focuses on the dynamical response of young star clusters to the rapid removal of residual star‑forming gas, a phase commonly referred to as violent relaxation. This brief but pivotal episode reshapes the structural and kinematic properties of clusters and leaves observable imprints on the statistical properties of whole cluster populations. The author first outlines the physical picture: when feedback processes (stellar winds, radiation pressure, supernovae) expel the remaining gas on a timescale of a few Myr, the gravitational potential of the proto‑cluster shallowens abruptly. Stars that were previously in virial equilibrium find themselves on super‑virial orbits; many become unbound and drift into the field, while the bound remnant settles into a new equilibrium after a short relaxation period.

A central claim is that violent relaxation directly sculpts three key observables: the cluster age distribution, the cluster mass function, and the ratio of total cluster mass to the total stellar mass of the host galaxy (the cluster‑to‑field mass ratio, CFM). Regarding the age distribution, the author demonstrates that the rapid loss of gas produces an “infant mortality” effect: the number of observable clusters drops sharply after a few Myr, even if the underlying formation rate remains constant. By incorporating a gas‑expulsion efficiency parameter (ε_gas) and an expulsion timescale (τ_gas) into simple analytic models, the predicted age distribution reproduces the steep decline seen in nearby galaxies such as the LMC, M33, and the Antennae.

The impact on the cluster mass function (CMF) is treated next. Assuming that low‑mass clusters retain a larger fraction of gas relative to their stellar mass, they suffer a higher bound‑fraction loss than massive clusters. Consequently, an initial power‑law CMF (dN/dM ∝ M^−β with β≈2) is transformed into a broken or flattened distribution at the low‑mass end. The author compares the model‑predicted CMF with observed CMFs in several environments and finds good agreement, especially the emergence of a characteristic turnover around 10^4 M_⊙ that is often attributed to dynamical evolution but here is shown to arise already during violent relaxation.

The third observable, CFM, links cluster evolution to galaxy‑scale processes. Stars that escape during gas expulsion populate the galactic disk, halo, or even intergalactic space, thereby reducing the fraction of stellar mass locked in bound clusters. The paper presents a semi‑analytic framework that couples the gas‑expulsion outcome to global galaxy properties such as the rotation curve, gas inflow rate, and the frequency of external perturbations (e.g., minor mergers). The model reproduces the observed trend that dwarf irregulars and starbursting systems have higher CFM values than massive, quiescent spirals, suggesting that the efficiency of gas removal and the depth of the galactic potential well jointly regulate how much stellar mass remains in clusters.

After establishing these connections, the author discusses the current limitations. First, the physical parameters governing gas expulsion (e.g., anisotropy, speed, residual gas fraction) are poorly constrained observationally. Second, the initial conditions of proto‑clusters—density profile, substructure, and primordial binary fraction—remain uncertain, introducing degeneracies in model predictions. Third, most existing simulations assume spherical symmetry and instantaneous gas loss, whereas real systems likely experience asymmetric, multi‑phase outflows. Overcoming these issues will require high‑resolution observations (JWST, ALMA, ELT) capable of probing embedded clusters, combined with state‑of‑the‑art N‑body‑hydrodynamic simulations that resolve both stellar dynamics and gas physics.

In conclusion, the study argues that violent relaxation is not merely a nuisance to be ignored but a powerful diagnostic of star formation physics, cluster survival, and galaxy evolution. By decoding the signatures left on the age distribution, mass function, and CFM, astronomers can infer the efficiency of feedback, the typical gas‑expulsion timescales, and the interplay between clusters and their host galaxies. The paper thus positions the early, gas‑driven phase of cluster evolution as a promising avenue for linking small‑scale star‑formation processes to the large‑scale assembly history of galaxies.