# Globular cluster formation with multiple stellar populations: self-enrichment in fractal massive molecular clouds

## TL;DR

This study uses hydrodynamical simulations to explore globular cluster formation from fractal molecular clouds, revealing how multiple stellar populations and chemical abundance spreads can originate through hierarchical merging and feedback effects.

## Contribution

It introduces a new simulation approach to model GC formation from fractal MCs, highlighting the roles of feedback and hierarchical merging in creating multiple stellar populations.

## Key findings

- GCs form from merging hierarchical star cluster complexes.
- Feedback from SNe and AGB stars influences star formation efficiency.
- Simulated GCs show radial helium abundance gradients.

## Abstract

Internal chemical abundance spreads are one of fundamental properties of globular clusters (GCs) in the Galaxy. In order to understand the origin of such abundance spreads, we numerically investigate GC formation from massive molecular clouds (MCs) with fractal structures using our new hydrodynamical simulations with star formation and feedback effects of supernovae (SNe) and asymptotic giant branch (AGB) stars. We particularly investigate star formation from gas chemically contaminated by SNe and AGB stars within MCs with different initial conditions and environments. The principal results are as follows. GCs with multiple generation of stars can be formed from merging of hierarchical star cluster complexes that are developed from high-density regions of fractal MCs. Feedback effects of SNe and AGB stars can control the formation efficiencies of stars formed from original gas of MCs and from gas ejected from AGB stars. The simulated GCs have radial gradients of helium abundances within the central 3 pc. The original MC masses need to be as large as 10^7 Msun for a canonical initial stellar mass function (IMF) so that the final masses of stars formed from AGB ejecta can be 10^5 Msun. Since star formation from AGB ejecta is rather prolonged (10^8 yr), their formation can be strongly suppressed by SNe of the stars themselves. This result implies that the so-called mass budget problem is much more severe than ever thought in the self-enrichment scenario of GC formation. and thus that IMF for the second generation of stars should be `top-light'.