Preparing the next gravitational million-body simulations: evolution of single and binary stars in nbody6++gpu, mocca, and mcluster. Issue 3 (27th December 2021)

• Record Type: Journal Article
• Authors: Kamlah, A W H; Leveque, A; Spurzem, R; Arca Sedda, M; Askar, A; Banerjee, S; Berczik, P; Giersz, M; Hurley, J; Belloni, D; Kühmichel, L; Wang, L
• Is Part Of: Monthly notices of the Royal Astronomical Society. Volume 511:Issue 3(2022)
• Page Start: 4060
• Page End: 4089
• Publication Date: 2021-12-27
• Subjects: methods: numerical -- software: development -- software: documentation -- binaries: general -- stars: general -- globular clusters: general; Astronomy -- Periodicals; Periodicals; 520.5
• DOI: 10.1093/mnras/stab3748
• Languages: English
• ISSNs: 0035-8711

Abstract

ABSTRACT: We present the implementation of updated stellar evolution recipes in the codes nbody6++gpu, mocca, and mcluster . We test them through numerical simulations of star clusters containing 1.1 × 10 5 stars (with 2.0 × 10 4 in primordial hard binaries) performing high-resolution direct N -body (nbody6++gpu ) and Monte Carlo (mocca ) simulations to an age of 10 Gyr. We compare models implementing either delayed or core-collapse supernovae mechanisms, a different mass ratio distribution for binaries, and white dwarf (WD) natal kicks enabled/disabled. Compared to nbody6++gpu, the mocca models appear to be denser, with a larger scatter in the remnant masses, and a lower binary fraction on average. The mocca models produce more black holes (BHs) and helium WDs, while nbody6++gpu models are characterized by a much larger amount of WD–WD binaries. The remnant kick velocity and escape speed distributions are similar for the BHs and neutron stars (NSs), and some NSs formed via electron-capture supernovae, accretion-induced collapse, or merger-induced collapse escape the cluster in all simulations. The escape speed distributions for the WDs, on the other hand, are very dissimilar. We categorize the stellar evolution recipes available in nbody6++gpu into four levels: the one implemented in previous nbody6++gpu and mocca versions (level A ), state-of-the-art prescriptions (level B ), some in a testing phase (level C ), and those that will be added in future versions of our codes.