Dynamically formed black hole binaries: In-cluster versus ejected mergers

Abstract

The growing number of black hole binary (BHB) mergers detected by the Laser Interferometer Gravitational-Wave Observatory have the potential to enable an unprecedented characterisation of the physical processes and astrophysical conditions that govern the formation of compact binaries. In this paper, we focus on investigating the dynamical formation of BHBs in dense star clusters through a state-of-the-art set of 58 direct N-body simulations with N $\leqslant200\,000$ particles which include stellar evolution, gravitational braking, orbital decay through gravitational radiation, and galactic tidal interactions. The simulations encompass a range of initial conditions representing typical young globular clusters, including the presence of primordial binaries. The systems are simulated for $\sim 12$ Gyr. The dataset yields 117 BHB gravitational wave (GW) events, with 97 binaries merging within their host cluster and 20 merging after having been ejected. Only 8% of all ejected BHBs merge within the age of the Universe. Systems in this merging subset tend to have smaller separations and larger eccentricities, as this combination of parameters results in greater emission of gravitational radiation. We confirm known trends from Monte Carlo simulations, such as the anti-correlation between the mass of the binary and age of the cluster. In addition, we highlight for the first time a difference at low values of the mass ratio distribution between in-cluster and ejected mergers. However, the results depend on assumptions on the strength of GW recoils, thus in-cluster mergers cannot be ruled out at a significant level of confidence. A more substantial catalogue of BHB mergers and a more extensive library of N-body simulations are needed to constrain the origin of the observed events.

Stellar-mass black holes in young massive and open stellar clusters – IV. Updated stellar-evolutionary and black hole spin models and comparisons with the LIGO-Virgo O1/O2 merger-event data

“…As discussed in Paper II (see also the references therein), the moderate density and velocity dispersion in the model clusters make them efficient factories of dynamically assembling PN subsystems, particularly, those comprising BHs. As also discussed in Paper II (see also Anagnostou et al 2020), the vast majority of the GR mergers from these computed model clusters are in-cluster BBH mergers. As also demonstrated therein (see also Banerjee 2020a), the final inspiralling phases of such merging BBHs sweep through the LISA and deci-Hertz GW frequency bands before merging in the LVK band.…”

Dynamical double black holes and their host cluster properties

“…We find that the ejected systems that merge (basically, the ejected systems of M-07) are biased towards higher eccentricity, unlike the ejected systems of all other models which roughly follow the thermal distribution. Anagnostou et al (2020) also finds ejected mergers having steeper eccentricity, as expected from Peters (1964) calculations showing that higher ellipticity aids in more efficient emission of gravitational waves (also discussed in section 5.5). It can be argued that the smaller clusters of our datasetas well as Banerjee (2021b)-do not allow enough interactions to eject hard (smaller semi-major axis), eccentric binaries and instead eject softer binaries more readily than they should be compared to more massive clusters.…”

“…We observe a higher fraction of in-situ DBH mergers than ex-situ mergers. This trend is noted in NBODY cluster simulations as shown by Anagnostou et al (2020) using models from de Vita et al (2019), as well as by Banerjee (2021b), unlike Monte-Carlo simulations (e.g., Rodriguez et al 2016a). We find that the ejected systems that merge (basically, the ejected systems of M-07) are biased towards higher eccentricity, unlike the ejected systems of all other models which roughly follow the thermal distribution.…”

“…The escaped binary BHs with significant eccentricity (a signature of the dynamical environment) at the moment of leaving the cluster can also merge within a Hubble time. Such out-of cluster mergers can still be considered of dynamical origin (Anagnostou et al 2020). It has also been noted that clusters with lower metallicity not only form more massive BHs (Mapelli 2016;Banerjee 2017), but enhanced dynamical interactions further increase the rate of BH mergers hence creating even more massive BHs than through isolated binary evolution (Rodriguez et al 2015;Di Carlo et al 2020a).…”

Evidence for Hierarchical Black Hole Mergers in the Second LIGO–Virgo Gravitational Wave Catalog

“…Black holes of ∼45-135M e are not typically expected to form via standard stellar evolution as the pair-instability process either limits the maximum mass of the progenitor star's core or completely disrupts the star entirely (Fryer et al 2001;Heger & Woosley 2002;Belczynski et al 2016;Spera & Mapelli 2017;Farmer et al 2019;Stevenson et al 2019;Farmer et al 2020;Woosley & Heger 2021). Potential (non-mutually exclusive) astrophysical formation mechanisms for black holes in this mass gap include hierarchical mergers, where the remnant of a previous merger becomes part of a new binary (Miller & Hamilton 2002;Antonini & Rasio 2016;Gerosa & Berti 2017;Rodriguez et al 2019;Yang et al 2019;Anagnostou et al 2020;Fragione & Silk 2020;Mapelli et al 2020;Fragione et al 2020a;Banerjee 2021); stellar mergers, which may result in a larger hydrogen envelope around a core below the pairinstability threshold (Spera et al 2018;Kremer et al 2020;Di Carlo et al 2020a;González et al 2021); formation of black holes from Population III stars that are able to retain their hydrogen envelopes (Farrell et al 2021; Kinugawa et al 2021;Vink et al 2021), formation via stellar triples in the field (Vigna-Gómez et al 2021); growth via accretion in an active galactic nucleus (AGN) disk (McKernan et al 2012;Michaely & Perets 2020;Secunda et al 2020;Tagawa et al 2020), or growth via rapid gas accretion in dense primordial clusters (Roupas & Kazanas 2019).…”