Richard J. Parker

The R136 star cluster hosts several stars whose individual masses greatly exceed the accepted 150 M⊙ stellar mass limit

Spectroscopic analyses of hydrogen-rich WN5‐6 stars within the young star clusters NGC 3603 and R136 are presented, using archival Hubble Space Telescope and Very Large Telescope spectroscopy, and high spatial resolution near-IR photometry, including MultiConjugate Adaptive Optics Demonstrator (MAD) imaging of R136. We derive high stellar temperatures for the WN stars in NGC 3603 (T∗ ∼ 42±2 kK) and R136 (T∗ ∼ 53± 3 kK) plus clumping-corrected mass-loss rates of 2 ‐ 5 ×10 −5 M⊙ yr −1 which closely agree with theoretical predictions from Vink et al. These stars make a disproportionate contribution to the global ionizing and mechanical wind power budget of their host clusters. Indeed, R136a1 alone supplies ∼7% of the ionizing flux of the entire 30 Doradus region. Compar isons with stellar models calculated for the main-sequence evolution of 85 ‐ 500 M⊙ accounting for rotation suggest ages of ∼1.5 Myr and initial masses in the range 105 ‐ 170 M⊙ for three systems in NGC 3603, plus 165 ‐ 320 M⊙ for four stars in R136. Our high stellar masses are supported by consistent spectroscopic and dynamical mass determinations for the components of NGC 3603 A1. We consider the predicted X-ray luminosity of the R136 stars if they were close, colliding wind binaries. R136c is consistent with a colliding wind binary system. However, short period, colliding wind systems are excluded for R136a WN stars if mass ratios are of order unity. Widely separated systems would have been expected to harden owing to early dynamical encounters with other massive stars within such a high density environment. From simulated star clusters, whose constituents are randomly sampled from the Kroupa initial mass function, both NGC 3603 and R136 are consistent with an tentative upper mass limit of ∼300 M⊙. The Arches cluster is either too old to be used to diagnose the upper mass limit, exhibits a deficiency of very massive stars, or mo re likely stellar masses have been underestimated ‐ initial masses for the most luminous stars in the Arches cluster approach 200 M⊙ according to contemporary stellar and photometric results. The potential for stars greatly exceeding 150 M⊙ within metal-poor galaxies suggests that such pair-instab ility supernovae could occur within the local universe, as has been claimed for SN 2007bi.

Dynamical Mass Segregation on a Very Short Timescale

We discuss the observations and theory of star cluster formation to argue that clusters form dynamically cool (subvirial) and with substructure. We then perform an ensemble of simulations of cool, clumpy (fractal) clusters and show that they often dynamically mass segregate on timescales far shorter than expected from simple models. The mass segregation comes about through the production of a short-lived, but very dense core. This shows that in clusters like the Orion Nebula Cluster the stars ≥ 4 M ☉ can dynamically mass segregate within the current age of the cluster. Therefore, the observed mass segregation in apparently dynamically young clusters need not be primordial, but could be the result of rapid and violent early dynamical evolution.

The early dynamical evolution of cool, clumpy star clusters

Observations and theory both suggest that star clusters form sub-virial (cool) with highly sub-structured distributions. We perform a large ensemble of N-body simulations of moderate-sized (N = 1000) cool, fractal clusters to investigate their early dynamical evolution. We find that cool, clumpy clusters dynamically mass segregate on a short timescale, that Trapezium-like massive higher-order multiples are commonly formed, and that massive stars are often ejected from clusters with velocities > 10 km s−1 (c.f. the average escape velocity of 2.5 km s−1 ). The properties of clusters also change rapidly on very short timescales. Young clusters may also undergo core collapse events, in which a dense core containing massive stars is hardened due to energy losses to a halo of lower-mass stars. Such events can blow young clusters apart with no need for gas expulsion. The warmer and less substructured a cluster is initially, the less extreme its evolution.

Do binaries in clusters form in the same way as in the field

We examine the dynamical destruction of binary systems in star clusters of different densities. We find that at high densities (10 4 -10 5 M ⊙ pc -3 ) almost all binaries with separations > 10 3 au are destroyed after a few crossing times. At low densities [O(10 2 )M ⊙ )pc -3 ], many binaries with separations > 1 0 3 au are destroyed, and no binaries with separations > 10 4 au survive after a few crossing times. Therefore, the binary separations in clusters can be used as a tracer of the dynamical age and past density of a cluster. We argue that the central region of the Orion nebula cluster was ∼100 times denser in the past with a half-mass radius of only 0.1-0.2 pc as (i) it is expanding, (ii) it has very few binaries with separations > 1 0 3 au and (iii) it is well mixed and therefore dynamically old. We also examine the origin of the field binary population. Binaries with separations 10 4 au can survive in any cluster and so must be produced by isolated star formation, but only if all isolated star formation produces extremely wide binaries.

Constraints on Massive Star Formation: Cygnus OB2 was always an Association

We examine substructure and mass segregation in the massive OB association Cygnus OB2 to better understand its initial conditions. Using a well understood Chandra X-ray selected sample of young stars we find that Cyg OB2 exhibits considerable physical substructure and has no evidence for mass segregation, both indications that the association is not dynamically evolved. Combined with previous kinematical studies we conclude that Cyg OB2 is dynamically very young, and what we observe now is very close to its initial conditions: Cyg OB2 formed as a highly substructured, unbound association with a low volume density (< 100 stars pc −3 ). This is inconsistent with the idea that all stars form in dense, compact clusters. The massive stars in Cyg OB2 show no evidence for having formed particularly close to one another, nor in regions of higher than average density. Since Cyg OB2 contains stars as massive as �100 M⊙ this result suggests that very massive stars can be born in relatively low-density environments. This would imply that massive stars in Cyg OB2 did not form by competitive accretion, or by mergers.

Characterizing the dynamical state of star clusters from snapshots of their spatial distributions

We determine the distribution of stellar surface densities, Σ, from models of static and dynamically evolving star clusters with different morphologies, including both radially smooth and substructured clusters. We find that the Σ distribution is degenerate, in the sense that many different cluster morphologies (smooth or substructured) produce similar cumulative distributions. However, when used in tandem with a measure of structure, such as the Q-parameter, the current spatial and dynamical state of a star cluster can be inferred. The effect of cluster dynamics on the Σ distribution and the Q-parameter is investigated using N-body simulations and we find that, depending on the assumed initial conditions, the Σ distribution can rapidly evolve from high to low densities in less than 5 Myr. This suggests that the Σ distribution can only be used to assess the current density of a star-forming region, and provides little information on its initial density. However, if the Σ distribution is used together with the Q-parameter, then information on the amount of substructure can be used as a proxy to infer the amount of dynamical evolution that has taken place. Substructure is erased quickly through dynamics, which can disrupt binary star systems and planets, as well as facilitate dynamical mass segregation. Therefore, dynamical processing in young star-forming regions could still be significant, even without currently observed high densities.

The VLT-FLAMES Tarantula Survey - IV. Candidates for isolated high-mass star formation in 30 Doradus

Whether massive stars (≳30 M⊙) can occasionally form in relative isolation (e.g. in clusters with M < 100 M⊙) or if they require a large cluster of lower-mass stars around them is a key test in the differentiation of star-formation theories as well as how the initial mass function of stars is sampled. Previous attempts to find O-type stars that formed in isolation were hindered by the possibility that such stars are merely runaways from clusters, i.e., their current isolation does not reflect their birth conditions. We introduce a new method to find O-type stars that are not affected by such a degeneracy. Using the VLT-FLAMES Tarantula Survey and additional high resolution imaging we have identified stars that satisfy the following constraints: 1) they are O-type stars that are not detected to be part of a binary system based on radial velocity (RV) time series analysis; 2) they are designated spectral type O7 or earlier; 3) their velocities are within 1σ of the mean of OB-type stars in the 30 Doradus region, i.e. they are not runaways along our line-of-sight; 4) the projected surface density of stars does not increase within 3 pc towards the O-star (no evidence for clusters); 5) their sight lines are associated with gaseous and/or dusty filaments in the interstellar medium (ISM); and 6) if a second candidate is found in the direction of the same filament with which the target is associated, both are required to have similar velocities. With these criteria, we have identified 15 stars in the 30 Doradus region, which are strong candidates for being high-mass stars that have formed in isolation. Additionally, we employed extensive Monte Carlo stellar cluster simulations to confirm that our results rule out the presence of clusters around the candidates. Eleven of these are classified as Vz stars, possibly associated with the zero-age main sequence. We include a newly discovered Wolf-Rayet star as a candidate, although it does not meet all of the above criteria.

The evolution of binary populations in cool, clumpy star clusters

Observations and theory suggest that star clusters can form in a subvirial (cool) state and are highly substructured. Such initial conditions have been proposed to explain the level of mass segregation in clusters through dynamics, and have also been successful in explaining the origin of Trapezium-like systems. In this paper, we investigate, using N-body simulations, whether such a dynamical scenario is consistent with the observed binary properties in the Orion Nebula Cluster (ONC). We find that several different primordial binary populations are consistent with the overall fraction and separation distribution of visual binaries in the ONC (in the range 67–670 au), and that these binary systems are heavily processed. The substructured, cool-collapse scenario requires a primordial binary fraction approaching 100 per cent. We find that the most important factor in processing the primordial binaries is the initial level of substructure; a highly substructured cluster processes up to 20 per cent more systems than a less substructured cluster because of localized pockets of high stellar density in the substructure. Binaries are processed in the substructure before the cluster reaches its densest phase, suggesting that even clusters remaining in virial equilibrium or undergoing supervirial expansion would dynamically alter their primordial binary population. Therefore, even some expanding associations may not preserve their primordial binary population.

Testing the universality of star formation – II. Comparing separation distributions of nearby star-forming regions and the field

We have measured the multiplicity fractions and separation distributions of seven young star-forming regions using a uniform sample of young binaries. Both the multiplicity fractions and separation distributions are similar in the different regions. A tentative decline in the multiplicity fraction with increasing stellar density is apparent, even for binary systems with separations too close (19–100 au) to have been dynamically processed. The separation distributions in the different regions are statistically indistinguishable over most separation ranges, and the regions with higher densities do not exhibit a lower proportion of wide (300–620 au) relative to close (62–300 au) binaries as might be expected from the preferential destruction of wider pairs. Only the closest (19–100 au) separation range, which would be unaffected by dynamical processing, shows a possible difference in separation distributions between different regions. The combined set of young binaries, however, shows a distinct difference when compared to field binaries, with a significant excess of close (19–100 au) systems among the younger binaries. Based on both the similarities and differences between individual regions, and between all seven young regions and the field, especially over separation ranges too close to be modified by dynamical processing, we conclude that multiple star formation is not universal and, by extension, the star formation process is not universal.

On the mass segregation of stars and brown dwarfs in Taurus

ABSTRACT We use the new minimum spanning tree (MST) method to look for mass segregation inthe Taurus association. The method computes the ratio of MST lengths of any chosensubset of objects, including the most massive stars and brown dwarfs, to the MSTlengths of random sets of stars and brown dwarfs in the cluster. This mass segregationratio (Λ MSR ) enables a quantitative measure of the spatial distribution of high-massand low-mass stars, and brown dwarfs to be made in Taurus.We find that the most massive stars in Taurus are inverselymasssegregated, withΛ MSR = 0.70 ±0.10 (Λ MSR = 1 corresponds to no mass segregation), which differsfrom the strong mass segregation signatures found in more dense and massive clusterssuch as Orion. The brown dwarfs in Taurus are not mass segregated, although wefind evidence that some low-mass stars are, with an Λ MSR = 1.25±0.15. Finally, wecompare our results to previous measures of the spatial distribution of stars and browndwarfs in Taurus, and briefly discuss their implications.Key words: methods: data analysis – star clusters: individual: Taurus – stars: lowmass, brown dwarfs