Escape from the Bermuda cluster: Orphanization by multiple stellar ejections

Abstract

Context. Dynamical interactions in young stellar clusters can eject massive stars early in their lives and significantly alter their mass functions. If all of the most massive stars are lost, we are left with an orphan cluster. Aims. We study the Bermuda cluster (Villafranca O-014 NW), the most significant young stellar group in the North America and Pelican nebulae, and the massive stars that may have been ejected from it to test if it has been orphaned. Methods. We use Gaia EDR3 parallaxes and proper motions to search for walkaway and runaway stars in the vicinity of the North America and Pelican nebulae. The candidates are analyzed with a combination of spectroscopy and photometry to assess their nature and their trajectories are traced back in time to determine at what time they left the Bermuda cluster. Results. We detect three ejection events, dubbed the Bajamar, Toronto, and HD 201 795 events, which expelled (a minimum of) five, two, and two systems, respectively, or six, three, and three stars if we count the individual components in spectroscopic and eclipsing binaries. The events took place 1.611 ± 0.011 Ma, 1.496 ± 0.044 Ma, and 1.905 ± 0.037 Ma ago, respectively, but our analysis is marginally consistent with the first two being simultaneous. We detect bow shocks in WISE images associated with four of the ejected systems; their orientation agrees with that of their relative proper motions with respect to the cluster. Combining the three events, the Bermuda cluster has lost over 200 M⊙, including its three most massive stars, so it can be rightfully considered an orphan cluster. One consequence is that the present-day mass function of the cluster has been radically altered from its top-heavy initial value to one compatible with a Kroupa-like function. Another is that the cluster is currently expanding with a dynamical timescale consistent with the cause being the ejection events. A scenario in which the Bermuda cluster was formed in a conveyor-belt fashion over several hundreds of ka or even 1 Ma is consistent with all the observables.

Kinematics of the Central Stars Powering Bowshock Nebulae and the Large Multiplicity Fraction of Runaway OB Stars

“…If SBN central stars are representative of the population of field OB stars, as is suggested by the consonance between the samples' multiplicity fractions and runaway fractions, this suggests that the DES contributes more heavily to the field OB population than BSS. The significant numbers of multiple systems ejected from young stellar clusters (Maíz Apellániz et al 2022) adds evidence to this scenario. Alternatively, given the high fraction of SBN central stars in isolated environments with low peculiar velocities ( 14 km s −1 ), this may provide evidence that a fraction of massive stars may form in relative isolation.…”

Kinematics of the Central Stars Powering Bowshock Nebulae and the Large Multiplicity Fraction of Runaway OB Stars

“…If SBN central stars are representative of the population of field OB stars, as is suggested by the consonance between the samples' multiplicity fractions and runaway fractions, this suggests that the DES contributes more heavily to the field OB population than BSS. The significant numbers of multiple systems ejected from young stellar clusters (Maíz Apellániz et al 2022) add evidence to this scenario. Alternatively, given the high fraction of SBN central stars in isolated environments with low peculiar velocities (;14 km s −1 ), this may provide evidence that a fraction of massive stars may form in relative isolation.…”

Gaia-ESO Survey: massive stars in the Carina Nebula. A new census of OB stars

“…Given that the periastron took place at a 3-D separation of just 18.6±1.0 AU (when the system was first spatially resolved in 1996 it was ∼375 AU), an order of magnitude (or even less) smaller than the expected semi-major axis of the inner orbit, and the high eccentricity, it is possible that the system has transitioned from an elliptic orbit to a hyperbolic trajectory and a possible ejection from Villafranca O-002 (Trumpler 14). If that had happened, this could be another example of an orphan cluster where the most (in this case, two) massive stars of a cluster are ejected through a dynamical interaction (Maíz Apellániz et al 2022b). Further observations are needed, especially with HST/STIS later in this decade (if it is still operational) when Aa and Ab are expected to reach plane-of-the-sky separations of ∼40 mas.…”