Pith. sign in

REVIEW 4 major objections 4 minor 43 references

The origin of the most recently ejected OB runaway star from the R136 cluster

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper reconstructs the ejection of the runaway binary Mel 34 from R136 as a five-body triple-binary encounter about 52,000 years ago.

desk verdict A novel and well-documented five-body ejection scenario that deserves serious reading, but the headline 'prediction' is already measured and the claimed reproduction is a tuned scatter, not a quantitative match. read the letter →

arxiv 2507.11795 v1 pith:K6HXLY7V submitted 2025-07-15 astro-ph.SR astro-ph.GAnlin.CD

Simon Portegies Zwart , Michel Stoop , Lex Kaper , Alex de Kooter , Steven Rieder , Tomer Shenar This is my paper
classification astro-ph.SRastro-ph.GAnlin.CD
keywords R136runawaystarsdynamicalejectiontriple-binaryencounterMelnick3439VFTS590massivebinaries
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper reconstructs the most recent ejection of a massive runaway binary, Mel 34, from the young cluster R136. By tracing Mel 34 and the star VFTS 590 backward to a common origin about 52,000 years ago, it concludes that a single strong encounter between a hierarchical triple, ((Mel 39A, Mel 39B), VFTS 590), and the binary (Mel 34A, Mel 34B) simultaneously launched all three objects. The reconstruction is deterministic Newtonian scattering, so it also predicts that Mel 39's unseen companion has about 80 solar masses and that all five stars will explode as supernovae within roughly 5 million years. If this is right, it turns a single runaway into a complete, testable forensic record of a dynamical event in a young massive cluster.

What carries the argument

The central machinery is a five-body scattering reconstruction, summarized by the interaction equation ((Mel39A, Mel39B), VFTS590) + (Mel34A, Mel34B) → (Mel34A, Mel34B), (Mel39A, Mel39B), VFTS590. The argument uses conservation of energy, linear momentum, and angular momentum to restrict the pre-encounter configuration, then samples the remaining parameter space with more than 1.5 million scattering experiments to find the configuration whose post-encounter orbital separations, eccentricities, and runaway velocities match the observed ones. The binary binding energies, expressed in units of the cluster's kinetic temperature kT, provide the energy budget that the encounter must supply.

What would settle it

A decisive check would be to measure the companion of Mel39 spectroscopically: if its mass is not close to 80 solar masses or its orbit is not near 2.48 au in the predicted plane, the proposed encounter fails. Independently, any detected deflection in the backward trajectories of Mel34 or VFTS590 after the supposed ejection, such as another encounter or significant mass loss, would erase the roughly 52,000-year timing coincidence.

Watch

Extended reading notes

Core claim

The paper claims that the runaway binary Mel 34, consisting of a 139 and a 127 solar-mass star in a 154.55-day eccentric orbit, was ejected from R136 about 52 ± 8 kyr ago not by a binary-single or binary-binary encounter but by a five-body interaction of a triple and a binary. In the most probable reconstruction, the triple ((Mel39A, Mel39B), VFTS590) encountered the binary (Mel34A, Mel34B); the encounter preserved both binaries, ejected Mel34 at about 46 km/s, VFTS590 at about 83 km/s, and sent Mel39 out at about 64 km/s with an approximately 80 solar-mass companion in a 2.48 au orbit. The same event is argued to account for the observed orbital separation and eccentricity of Mel34 and for the near-opposite directions of the escaping binaries. This five-body channel is unexpected because triple interactions were not previously thought to produce runaways. The paper further predicts that the five participating stars will explode as supernovae within about 5 Myr at distances of 180–332 pc from R136, producing black-hole binaries that will not merge within a Hubble time.

Load-bearing premise

The reconstruction assumes that the backward trajectories of Mel34 and VFTS590 have been unperturbed since ejection, with no later close encounters, no mass loss, and a static cluster potential, so the common-ejection event dissolves if either star was deflected.

Editorial extensions

If this is right

  • Mel 39 should reveal an approximately 80 solar-mass companion in a roughly 2.48 au orbit, with an orbital plane close to that of Mel 34.
  • All five participating stars will explode as supernovae within about 5 Myr, at distances of roughly 180 to 332 pc from R136.
  • The resulting black-hole binaries will survive as binaries but will not merge within a Hubble time.
  • The rate of binary-preserving encounters that eject VFTS590 and the two binaries is about 27 per Myr, close to the observed runaway production rate of about 24 per Myr.
  • About 52 additional lower-mass runaways below the current detection limit should have been ejected from R136 in the same recent wave.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the reconstruction is right, the same trajectory-rewinding method can be applied to other R136 runaways; each recovered common ejection event would directly measure the cluster's recent encounter rate and the properties of its hard binaries.
  • The paper leaves open the identity of the roughly 90 solar-mass star that a four-body alternative would require; a targeted deep search along the predicted momentum-conserving trajectory about 8.5 pc from R136 would test whether such an object is simply obscured or genuinely absent.
  • The binary-evolution endpoint assumes no significant natal kicks in the supernovae; if kicks are important, the two black-hole binaries could be disrupted, changing the predicted non-merging outcome.
  • A successful search for the predicted 80 solar-mass companion of Mel 39 would turn this single reconstructed encounter into a template for identifying the dynamical ejection events that produce runaway binaries in young clusters.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper reconstructs the dynamical history of the recently ejected OB runaway binary Mel 34 from the young massive cluster R136. The authors propose that Mel 34 was ejected about 52,000 years ago in a five-body encounter between the hierarchical triple ((Mel39A, Mel39B), VFTS590) and the binary (Mel34A, Mel34B), with the most probable interaction given by Eq. (2). They report that this single encounter simultaneously produces the runaway binary Mel34 at ~46 km/s, the runaway star VFTS590 at ~83 km/s, and the escaping binary Mel39, and they predict that Mel39's unseen companion has a mass of about 80 solar masses and orbits nearly coplanarly with Mel34. The paper is supported by extensive 3-, 4-, and 5-body scattering simulations, N-body cluster models, and binary evolution calculations, with code and data made publicly available.

Significance. If validated, the paper would provide a rare, deterministic reconstruction of a specific five-body encounter that ejected multiple massive objects from a cluster, offering a new dynamical channel for producing runaway binaries and isolated runaways simultaneously. The scattering calculations are extensive, with over two million experiments, and the appendices document the assumptions in unusual detail. The public availability of the code and data is a strength. However, the central claims currently rest on a statistical comparison that is not quantified as a joint likelihood, and the abstract presents as predictions several quantities that are either adopted as inputs or are consistent with a very broad distribution. The significance of the paper therefore depends on whether these overclaims can be corrected and whether a proper statistical test supports the scenario.

major comments (4)
  1. [Section 6, Table 2, Table 1] The claim that scenario G 'reproduces statistically the observed orbital separations of Mel 34 and Mel 39, the velocities of both binaries, and the observed velocity of VFTS 590' is not supported by a joint statistical test. The reported match fractions, such as fesc,VFTS590 ≈ 0.44 in Table 1, are marginal rates for escaping with velocity above a threshold (≥27.6 km/s), not probabilities for the full observed final state. Table 2 shows mean simulated final semi-major axes and velocities that differ from the observed values: for Mel34, a_fin = 6.03 ± 9.14 au versus the observed 3.63 au; for Mel39, 1.39 ± 0.57 au versus 2.48 au; and for Mel34, the mean runaway velocity is 70.6 ± 8.6 km/s versus the observed projected 46.3 ± 6.4 km/s (even after a projection correction to ~55 km/s, this is still a 1–2 sigma offset). Without computing the joint probability of satisfying all observed constraints simultaneously, the identification of Eq. (2) as the actual encounter is not demonstrated.
  2. [Abstract, Section 4, Section 5, Appendix E] The abstract and Section 6 present the '80 M⊙ companion' in Mel 39 as a prediction of the scattering calculations, but the mass is adopted as a fixed input. Appendix E states 'we assume that VFTS 590 has a mass of 50 M⊙, Mel39B is 80 M⊙', and Section 4 explicitly notes that this mass is 'consistent with the spectroscopically determined mass of 80 ± 11 M⊙, derived by [20]' (Pollock et al. 2025). The companion mass is therefore an observational constraint adopted into the model, not an output of the simulation. The abstract and conclusions should be reworded to avoid claiming a prediction of a quantity that is already measured and used as an input.
  3. [Abstract, Section 6] The abstract's claim that the orbital planes of Mel 34 and Mel 39 are 'within ∼1°' is contradicted by the distribution reported in Section 6: δi = (−1.1 ± 56)°. A mean near zero with a 56° standard deviation does not constrain the relative inclination to one degree. This claim appears to be based on the mean of a wide distribution, and it should be corrected or removed, because it is presented as a major successful prediction of the model.
  4. [Section 5, 'optimal parameters'] The initial conditions a_in = 4 au, a_p = 6 au, and v_enc = 7 km/s are selected to reproduce the observed final state, but the paper provides no quantitative goodness-of-fit measure for this choice. The text states that these parameters 'reproduce the current observed orbital separations and runaway velocities', yet Table 2 shows that the simulated means differ from the observed values, and the reported match fractions are only marginal. The authors should provide a likelihood or other statistical measure that compares the full observed state against the simulations, rather than relying on qualitative agreement or single-constraint thresholds.
minor comments (4)
  1. [Throughout] Several typos and grammatical errors appear in the manuscript, including 'let to' for 'led to' (abstract, Section 1), 'interation' (Section 4), 'refered' (Section 3), 'insufficienly' (Section 4), 'perfomed' (Appendix E), 'planet' for 'panel' (Fig. F.III caption), 'consisent' (Section 6), and 'avarious' (Appendix E). A careful proofread is needed.
  2. [Table E.IV] The column header for frun lists units of '[au 2]', which appears to be a typo; frun is a fraction and should be dimensionless, while the cross-section columns already have units of 10^6 au^2.
  3. [Figure 2 caption] The caption describes the interaction as 'for scenario B', but the surrounding text and Table 1 identify scenario G as the favored configuration. Please clarify whether the figure shows scenario B or G, and ensure the labels and the scenario designation are consistent.
  4. [Appendix D] The sentence 'The probability, however, that this interaction leads to a collision between two (or even three) stars is more than 60 times larger than the clean exchange and ionization interaction in which VFTS 590 and Mel 34AB would be ejected' would be clearer if it explicitly stated that the 60-fold factor compares collision outcomes to the desired exchange outcome, rather than to an ionization interaction.

Circularity Check

2 steps flagged · score 6.0 of 10

The abstract's 'prediction' of Mel 39's 80 M_sun companion is an adopted measured value, and the scattering initial conditions are tuned to reproduce the observed final orbit, making the claimed reproduction partly by construction.

  1. fitted input called prediction [Abstract; Section 4; Section 5; Appendix E.4]
    "We then predict that Mel 39 is a binary star with an 80 M⊙ companion star that orbits within ∼ 1◦ in the same plane as Mel 34, and escapes the cluster with a velocity of ∼ 64 km/s. ... We reduce parameter space by adopting the masses of VFTS 590 (50 M⊙) and Mel 39B (80 M⊙) unique, and assuming 130 M⊙ for Mel 34A, Mel 34B, and Mel 39A."

    The 80 M⊙ companion mass is not derived from the encounter model: Section 4 already states it is 'consistent with the spectroscopically determined mass of 80 ± 11 M⊙, derived by [20]', and Section 5 fixes Mel39B = 80 M⊙ as an adopted input for all 5-body scattering. The abstract's 'prediction' of an 80 M⊙ companion is therefore the same measured/adopted value presented as an output; the orbital-plane and escape-velocity claims ride on that input, and the 80 M⊙ mass itself is forced by construction.

  2. fitted input called prediction [Section 5 (Reconstructing the interaction) and Section 6 (Results)]
    "The optimal parameters are determined by running 5-body simulations to reproduce the post-encounter binaries, Mel 34 and Mel 39, with their appropriate orbital separations of 3.63 au and 2.48 au, respectively and runaway speeds (see appendix E). ... The cross section is largest for orbital separations of the inner target and projectile binaries of 4 au and 6 au, respectively, and these reproduce the current observed orbital separations and runaway velocities."

    The initial separations (4 au and 6 au) and encounter velocity (7 km/s) are selected precisely because they reproduce the observed final separations (3.63 au and 2.48 au) and runaway velocities. Section 6 then reports 'The interaction reproduces statistically the observed orbital separations ... and the observed velocity of VFTS 590' as support for the scenario. This is a fit to the validation data, not an independent prediction: the observed values used as success criteria are the same values that determined the input parameters. The paper does not quote a joint likelihood, and Table 2's means show the match is marginal, so the claimed reproduction is partly by construction and partly unquantified.

full rationale

The paper's astrometric reconstruction of Mel 34's and VFTS 590's trajectories is independent data processing (Gaia astrometry/Stoop et al. 2024), and the cross-section computations among 8 scenarios are a genuine dynamical calculation, so the work is not wholly circular. The load-bearing circularity is concentrated in two places. First, the abstract's 'prediction' that Mel 39 has an 80 M_sun companion is exactly the spectroscopically measured value adopted as a fixed input in Section 5 and Appendix E; the companion mass is therefore not an output of the encounter model. Second, the initial orbital separations (4 au, 6 au) and encounter velocity (7 km/s) are chosen by running simulations to 'reproduce' the observed final separations and velocities, and the same reproduction is then cited in Section 6 as evidence for scenario G; this is fitting to the validation data, and the reported simulated means in Table 2 actually deviate from the observed values. The 'within ~1 degree' coplanarity statement is drawn from δi = (−1.1 ± 56)°, so it is a statistical overreach rather than a circular step. The self-citations to earlier work on runaways are contextual and not load-bearing. On balance, one central 'prediction' reduces to an input and the quantitative validation partly reduces by construction, so a score of 6 is appropriate.

Assumptions & free parameters 5 free parameters · 6 assumptions · 1 invented entities

The central scenario rests on the cluster model, the astrometric backtracking, the ballistic assumption, the triple-plus-binary initial configuration, and point-mass scattering. The main free parameters are the three initial orbital scales and the encounter velocity, all tuned to reproduce the observed final state; the 80 solar mass companion mass is an adopted observational input. No new physics entities are required by the adopted scenario.

free parameters (5)
  • Initial inner binary separation of the target triple (a_in) = 4 au
    Section 5 and Table 2: chosen because scattering with a 4 au inner binary and a 6 au projectile binary reproduces the observed post-encounter orbital separations and runaway velocities; varied between 2 and 10 au.
  • Initial projectile binary separation (a_p) = 6 au
    Same set as a_in; varied between 2 and 10 au. The observed Mel34 separation of 3.63 au and Mel39 separation of 2.48 au emerge from this choice.
  • Initial outer tertiary orbit (a_out) = 40 to 200 au
    Table E.III: the tertiary orbit is ill-constrained by energetics and has a minor effect on the cross-section; a range is adopted.
  • Relative encounter velocity (v_enc) = 7 km/s
    Appendix E and Table E.II: chosen for consistency with the cluster core velocity dispersion and to give a rate near 34 Myr^-1, close to the observed 24 Myr^-1.
  • Simulation mass assignment = VFTS590 = 50, Mel39B = 80, others = 130 Msun
    Section 5 and Appendix E: adopted to reduce the 5! assignment space. The 80 Msun value is from the cited Pollock et al. (2025) measurement; the others are rounded approximations.
assumptions (6)
  • standard math Newtonian point-mass gravitational dynamics govern the few-body scattering and the cluster evolution.
    Used throughout Sections 3 to 5 and Appendix E for the scattering simulations; no relativistic or hydrodynamical effects are included.
  • domain assumption The cluster R136 can be modeled as a King profile with W0 about 9.8, core radius about 0.1 pc, half-mass radius 2.9 pc, and core velocity dispersion about 7 km/s.
    Appendix A; used to set encounter rates, the velocity dispersion, and the 7 km/s encounter velocity.
  • domain assumption The Gaia-based trajectories of Mel34 and VFTS590 correctly identify a common ejection epoch and location near the cluster center.
    Section 2; the authors rely on Stoop et al. (2024) for the backtracking; the common event is inferred from a timing match within 1.1 standard deviations.
  • domain assumption The ejected stars have moved ballistically since the encounter, with no strong encounters, no mass loss, and a static potential.
    Section 2 and Appendix F; if deflected, the reconstructed encounter site and timing would be wrong.
  • domain assumption The pre-encounter configuration was a hierarchical triple plus a binary, with VFTS590 as the tertiary, rather than a single star or a four-body encounter.
    Sections 4 and 5; alternative three- and four-body encounters are rejected on energy, angular momentum, and cross-section grounds, and the five-body simulations are built on this premise.
  • domain assumption No stellar collisions occur during the favored encounter; all stars are treated as point masses and survive.
    Section 3 and Appendix D; the authors argue collisions are unlikely in the favored five-body scenario, but the scattering calculations are point-mass.
invented entities (1)
  • Approximately 90 solar mass mystery runaway star
    purpose: Would conserve linear and angular momentum in a four-body binary-binary encounter that ejects both Mel34 and VFTS590.
    Introduced in Section 4 and Appendix D as a possible companion to VFTS590 before the encounter. The paper finds no observed candidate along its predicted trajectory and ultimately favors the five-body scenario, so this entity is not part of the adopted solution.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The origin of the most recently ejected OB runaway star from the R136 cluster." pith.science (2026). https://pith.science/paper/K6HXLY7V

@misc{pith2026250711795,
  author       = {Pith},
  title        = {Pith review of: The origin of the most recently ejected OB runaway star from the R136 cluster},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K6HXLY7V}},
  note         = {Machine review of arXiv:2507.11795}
}
abstract

The $\sim 60\,000$ solar-mass (\MSun) star-cluster R136 (NGC~2070) in the Tarantula Nebula in the Large Magellanic Cloud is the host of at least 55 massive stars ($M \apgt 10$\,\MSun) which move away from the cluster at projected velocities $\gteq 27.5$\,km/s \cite{2024Natur.634..809S}. The origin of the high velocities of such runaway stars have been debated since the 1960s, resulting either from dynamical ejections \citep{1961BAN....15..265B,1961BAN....15..291B} or from supernova explosions \citep{1983ApJ...267..322H}. Due to the Gaia satellite's outstanding precision, we can now retrace the most recently ejected binary star, Mel 34, back to the center of R136 and reconstruct the events that 52\,000 years ago let to its removal from R136, i.e., we establish its dynamical interaction and ejection history. We find that this ejection requires the participation of 5 stars in a strong interaction between a triple composed of the tight massive binary Mel~39 orbited by the star VFTS~590, and the binary star Mel~34. The participation of 5 stars is unexpected because runaway stars were not expected to result from triple interactions \cite{2011Sci...334.1380F}. The deterministic nature of the Newtonian dynamics in the scattering enables us to reconstruct the encounter that ejected Mel~34. We then predict that Mel~39 is a binary star with an 80\,\Msun\, companion star that orbits within $\sim 1^\circ$ in the same plane as Mel~34, and escapes the cluster with a velocity of $\sim 64$\,km/s. The five stars will undergo supernova explosions in the coming 5\,Myr at a distance of $\sim 180$\,pc to $\sim 332$\,pc from their birth location (R\,136). The resulting black hole binaries, however, are not expected to merge within a Hubble time.

Figures

Figures reproduced from arXiv: 2507.11795 by the authors.

Figure 1
Figure 1. FIG. 1. Reconstruction of the historic trajectory for the massive binary Mel 34 (red), VFTS 590 (black) and the mystery runaway star that [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Interaction between the triple ((Mel 34A, Mel 39B), VFTS 590) and the binary (Mel39A, Mel34B) for scenario B in a strong [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Orbital separation vs eccentricity for Mel34 (gray shades, see color-bar to the right) and Mel39 (contours, overplotted in the colorbar). [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

43 extracted references · 35 canonical work pages

  1. [20]

    Polak, M.-M

    B. Polak, M.-M. Mac Low, R. S. Klessen, S. Portegies Zwart, E. P. Andersson, S. M. Appel, C. Cournoyer-Cloutier, S. C. O. Glover, S. L. W. McMillan, Massive star cluster formation: II. Runaway stars as fossils of subcluster mergers, A&A 690 (2024) A207. arXiv: 2405.12286, doi:10.1051/0004-6361/202450774. 15

    arXiv 2024

  2. [1]

    It is host of the largest known population of stars more massive than 100 M⊙ and several even in excess of 150M⊙ [7, 8]

    INTRODUCTION The young ( < ∼ 1 Myr) and massive (∼ 60 000M⊙) star cluster NGC2070 (also known as R136) in the Large Magelanic cloud is one of the most dynamically active regions in the local group [6]. It is host of the largest known population of stars more massive than 100 M⊙ and several even in excess of 150M⊙ [7, 8]. With a core-density of n ≃ 4.2 × 1...

    work page 2024

  3. [2]

    THE LAST MASSIVE RUNA W A Y STAR MEL 34 The most recently ejected object is the binary Mel 34AB composed of two stars of 139+21 −18 M⊙ and 127 ± 17 M⊙ in an eccentric (e = 0.68 ± 0.02) 154.55 ± 0.05 day (a ≃ 3.63 au) orbit. Mel 34 was ejected 52 ± 8 kyr ago with a projected velocity arXiv:2507.11795v1 [astro-ph.SR] 15 Jul 2025 2 8 6 4 2 0 2 4 6 8 x [pc] 2...

    work page Pith review arXiv 2025

  4. [3]

    THE BALLISTICS OF THE LAST RUNA W A YS Ejecting a massive binary with such a high velocity requires an unusually strong encounter with other stars. The binary itself can deliver enough energy for such an ejection if the encountering star is of comparable mass, but can we identify this other star? Mel 34 may have formed as a wide binary system, over time d...

    work page

  5. [4]

    THE OTHER STARS PARTICIPA TING IN THE ENCOUNTER Angular momentum can only be conserved if at least one other star participated in the encounter, which would also have been ejected from the cluster (fig. 1). We performed 4-body (binary-binary, and triple-single) scattering experiments to determine the most favorable mass of the unknown runaway (see appendi...

    work page

  6. [5]

    We can limit the available parameter space enormously by requirements that energy, and momentum (linear and angular) are conserved througout the interaction

    RECONSTRUCTING THE INTERACTION Exploring the entire parameter space of a 5-body encounter is hindered by its large volume. We can limit the available parameter space enormously by requirements that energy, and momentum (linear and angular) are conserved througout the interaction. The two binaries, currently with binding energies of ∼ 4300 kT and ∼ 4000 kT...

    work page

  7. [6]

    The triple was composed of ((Mel 39A, Mel 39B), VFTS 590) with orbital separations of 4.0 ± 0.5 au for the inner target orbit, and an outer orbit between 40 au and 200 au

    RESULTS The last energetic encounter in R 136 occurred ∼ 52 000 years ago between a hierarchical triple and a binary. The triple was composed of ((Mel 39A, Mel 39B), VFTS 590) with orbital separations of 4.0 ± 0.5 au for the inner target orbit, and an outer orbit between 40 au and 200 au. Before the encounter, the binary (Mel 34A, Mel 34B) had an orbital ...

    work page

  8. [8]

    S. A. Brands, A. de Koter, J. M. Bestenlehner, P. A. Crowther, J. O. Sundqvist, J. Puls, S. M. Caballero-Nieves, M. Abdul-Masih, F. A. Driessen, M. Garc´ıa, S. Geen, G. Gr¨afener, C. Hawcroft, L. Kaper, Z. Keszthelyi, N. Langer, H. Sana, F. R. N. Schneider, T. Shenar, J. S. Vink, The R136 star cluster dissected with Hubble Space Telescope/STIS. III. The m...

    work page Pith review arXiv 2022

Show all 43 references
  1. [9]

    Stoop, A

    M. Stoop, A. de Koter, L. Kaper, S. Brands, S. Portegies Zwart, H. Sana, F. Stoppa, M. Gieles, L. Mahy, T. Shenar, D. Guo, G. Nelemans, S. Rieder, Two waves of massive stars running away from the young cluster R136, Nat 634 (8035) (2024) 809–812. arXiv:2410. 06255, doi:10.1038...

    2024 doi

  2. [10]

    Blaauw, On the origin of the O- and B-type stars with high velocities (the ”run-away” stars), and some related problems, Bul

    A. Blaauw, On the origin of the O- and B-type stars with high velocities (the ”run-away” stars), and some related problems, Bul. Astron. Inst. Neth. 15 (1961) 265–+

    1961

  3. [11]

    Boersma, Mathematical theory of the two-body problem with one of the masses decreasing with time, Bul

    J. Boersma, Mathematical theory of the two-body problem with one of the masses decreasing with time, Bul. Astron. Inst. Neth. 15 (1961) 291–301

    1961

  4. [12]

    J. G. Hills, The effects of sudden mass loss and a random kick velocity produced in a supernova explosion on the dynamics of a binary star of arbitrary orbital eccentricity - Applications to X-ray binaries and to the binary pulsars, ApJ 267 (1983) 322–333. doi: 10.1086/160871....

    1983 doi

  5. [13]

    M. S. Fujii, S. Portegies Zwart, The Origin of OB Runaway Stars, Science 334 (2011) 1380–. arXiv:1111.3644, doi:10.1126/ science.1211927

    2011 arXiv

  6. [14]

    S. F. Portegies Zwart, S. L. W. McMillan, M. Gieles, Young Massive Star Clusters, ARA&A 48 (2010) 431–493.arXiv:1002.1961, doi:10.1146/annurev-astro-081309-130834

    2010 arXiv

  7. [15]

    P. A. Crowther, O. Schnurr, R. Hirschi, N. Yusof, R. J. Parker, S. P. Goodwin, H. A. Kassim, The R136 star cluster hosts several stars whose individual masses greatly exceed the accepted 150M solar stellar mass limit, MNRAS 408 (2) (2010) 731–751. arXiv: 1007.3284, doi:10.1111...

    2010 arXiv

  8. [16]

    Heggie, P

    D. Heggie, P. Hut, The Gravitational Million-Body Problem: A Multidisciplinary Approach to Star Cluster Dynamics, The Gravitational Million-Body Problem: A Multidisciplinary Approach to Star Cluster Dynamics, by Douglas Heggie and Piet Hut. Cambridge University Press, 2003, 37...

    2003

  9. [17]

    S. F. Portegies Zwart, E. P. J. van den Heuvel, A runaway collision in a young star cluster as the origin of the brightest supernova, Nat 450 (2007) 388–389. arXiv:arXiv:0711.2293, doi:10.1038/nature06276

    2007 arXiv

  10. [18]

    Blaauw, W

    A. Blaauw, W. W. Morgan, The space motions of ae aurigae and &mu; columbae with respect to the orion nebula., ApJ 119 (1954) 625

    1954

  11. [19]

    Poveda, J

    A. Poveda, J. Ruiz, C. Allen, Bolet ´ın de los Observatorios de Tonantzintla y Tacubaya

  12. [21]

    K. A. Tehrani, P. A. Crowther, J. M. Bestenlehner, S. P. Littlefair, A. M. T. Pollock, R. J. Parker, O. Schnurr, Weighing Melnick 34: the most massive binary system known, MNRAS 484 (2) (2019) 2692–2710. arXiv:1901.04769, doi:10.1093/mnras/stz147

    2019 arXiv

  13. [22]

    S. F. Portegies Zwart, S. L. W. McMillan, Gravitational thermodynamics and black-hole mergers, Int, J, of Mod. Phys, A 15 (2000) 4871–4875. doi:10.1142/S0217751X00002135. URL http://journals.wspc.com.sg/139/15/1530/S0217751X00002135.html

    2000 doi

  14. [23]

    D. C. Heggie, Binary evolution in stellar dynamics, MNRAS 173 (1975) 729–787

    1975

  15. [24]

    B. P. Abbott, et al., Kagra Collaboration, VIRGO Collaboration, Prospects for observing and localizing gravitational-wave transients with Advanced LIGO, Advanced Virgo and KAGRA, Living Reviews in Relativity 21 (1) (2018) 3. arXiv:1304.0670, doi: 10.1007/s41114-018-0012-9

    2018 arXiv

  16. [25]

    P. Hut, S. McMillan, R. W. Romani, The evolution of a primordial binary population in a globular cluster, ApJ 389 (1992) 527–545

    1992

  17. [26]

    S. F. Portegies Zwart, S. L. W. McMillan, The Runaway Growth of Intermediate-Mass Black Holes in Dense Star Clusters, ApJ 576 (2002) 899–907. arXiv:astro-ph/0201055, doi:10.1086/341798

    2002 arXiv

  18. [27]

    Most simulated clusters virialize within ∼ 0.2 Myr, at which point, the core has shrunk to ∼ 0.10 pc [28]

    sphere with a half-mass radius of 2.9 pc. Most simulated clusters virialize within ∼ 0.2 Myr, at which point, the core has shrunk to ∼ 0.10 pc [28]. By this time, the mean mass within 0.06 pc is about 11 M⊙ (with an average of ∼ 3.6 M⊙ in the core). At this time the density in...

    2011

  19. [28]

    J. M. Bestenlehner, P. A. Crowther, S. M. Caballero-Nieves, F. R. N. Schneider, S. Sim´on-D´ıaz, S. A. Brands, A. de Koter, G. Gr¨afener, A. Herrero, N. Langer, D. J. Lennon, J. Ma ´ız Apell ´aniz, J. Puls, J. S. Vink, The R136 star cluster dissected with Hubble Space Tele- sc...

    2020 arXiv

  20. [29]

    A. M. T. Pollock, P. A. Crowther, J. M. Bestenlehner, P. S. Broos, L. K. Townsley, Melnick 39 is a very massive intermediate-period colliding-wind binary, MNRAS 539 (2) (2025) 1291–1298. arXiv:2503.17150, doi:10.1093/mnras/staf501

    2025 arXiv

  21. [30]

    R. A. Mardling, S. J. Aarseth, Tidal interactions in star cluster simulations, MNRAS 321 (2001) 398–420

    2001

  22. [31]

    Zahn, Tidal friction in close binary stars, A&A 57 (1977) 383–394

    J.-P. Zahn, Tidal friction in close binary stars, A&A 57 (1977) 383–394

    1977

  23. [32]

    Massey, L

    P. Massey, L. R. Penny, J. Vukovich, Orbits of Four Very Massive Binaries in the R136 Cluster, ApJ 565 (2002) 982–993. arXiv: arXiv:astro-ph/0110088, doi:10.1086/324783

    2002 arXiv

  24. [33]

    Portegies Zwart, S

    S. Portegies Zwart, S. McMillan, Astrophysical Recipes, 2514-3433, IOP Publishing, 2018. doi:10.1088/978-0-7503-1320-9 . URL http://dx.doi.org/10.1088/978-0-7503-1320-9

    2018 doi

  25. [34]

    E. E. Salpeter, The luminosity function and stellar evolution., ApJ 121 (1955) 161

    1955

  26. [35]

    H. C. Plummer, On the problem of distribution in globular star clusters, MNRAS 71 (1911) 460–470

    1911

  27. [36]

    D. P. Caputo, N. de Vries, S. Portegies Zwart, On the effects of subvirial initial conditions and the birth temperature of R136, MNRAS 445 (1) (2014) 674–685. arXiv:1409.4765, doi:10.1093/mnras/stu1769

    2014 arXiv

  28. [37]

    I. R. King, The structure of star clusters. iii. some simple dvriamical models, AJ 71 (1966) 64–75

    1966

  29. [38]

    Makino, P

    J. Makino, P. Hut, On core collapse, ApJ 383 (1991) 181–191. doi:10.1086/170774

    1991 doi

  30. [39]

    V . V . Gvaramadze, A. Gualandris, S. Portegies Zwart, On the origin of high-velocity runaway stars, MNRAS 396 (2009) 570–578. arXiv:0903.0738, doi:10.1111/j.1365-2966.2009.14809.x

    2009 arXiv

  31. [40]

    S. F. Portegies Zwart, P. Hut, S. L. W. McMillan, J. Makino, Star cluster ecology - V . Dissection of an open star cluster: spectroscopy, MNRAS 351 (2004) 473–486. arXiv:arXiv:astro-ph/0301041, doi:10.1111/j.1365-2966.2004.07709.x

    2004 arXiv

  32. [41]

    Gualandris, S

    A. Gualandris, S. Portegies Zwart, P. P. Eggleton, N-body simulations of stars escaping from the Orion nebula, MNRAS 350 (2004) 615–626. arXiv:arXiv:astro-ph/0401451, doi:10.1111/j.1365-2966.2004.07673.x

    2004 arXiv

  33. [42]

    D. C. Heggie, R. D. Mathieu, Standardised Units and Time Scales, in: P. Hut, S. L. W. McMillan (Eds.), The Use of Supercomputers in Stellar Dynamics, V ol. 267 of Lecture Notes in Physics, Berlin Springer Verlag, 1986, p. 233.doi:10.1007/BFb0116419

    1986 doi

  34. [43]

    S. F. Portegies Zwart, F. Verbunt, Population synthesis of high-mass binaries., A&A 309 (1996) 179–196

    1996

  35. [44]

    Portegies Zwart, The ecological impact of high-performance computing in astrophysics, Nature Astronomy 4 (2020) 819–822

    S. Portegies Zwart, The ecological impact of high-performance computing in astrophysics, Nature Astronomy 4 (2020) 819–822. arXiv:2009.11295, doi:10.1038/s41550-020-1208-y

    2020 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.