The inefficiency of satellite accretion in forming extended star clusters

TL;DR

This paper investigates whether the accretion process can explain the large sizes of extended star clusters, concluding that their extended structure is likely inherent from formation rather than due to tidal effects during accretion.

Contribution

The study uses N-body simulations to demonstrate that tidal effects during accretion are insufficient to produce the observed extended sizes of star clusters, challenging the accreted origin hypothesis.

Findings

Tidal effects during accretion cause some expansion but are insufficient to match observed sizes.

Clusters formed in extreme environments are more compact than observed extended clusters.

Extended globular clusters are likely formed already extended, not solely through accretion processes.

Abstract

The distinction between globular clusters and dwarf galaxies has been progressively blurred by the recent discoveries of several extended star clusters, with size (20-30 pc) and luminosity (-6 < Mv < -2) comparable to the one of faint dwarf spheroidals. In order to explain their sparse structure, it has been suggested that they formed as star clusters in dwarf galaxy satellites that later accreted onto the Milky Way. If these clusters form in the centre of dwarf galaxies, they evolve in a tidally-compressive environment where the contribution of the tides to the virial balance can become significant, and lead to a super-virial state and subsequent expansion of the cluster, once removed. Using N-body simulations, we show that a cluster formed in such an extreme environment undergoes a sizable expansion, during the drastic variation of the external tidal field due to the accretion process. However, we show that the expansion due to the removal of the compressive tides is not enough to explain the observed extended structure, since the stellar systems resulting from this process are always more compact than the corresponding clusters that expand in isolation due to two-body relaxation. We conclude that an accreted origin of extended globular clusters is unlikely to explain their large spatial extent, and rather favor the hypothesis that such clusters are already extended at the stage of their formation.