Residual Gas Expulsion Research Articles

Young, wild, and free: The early expansion of star clusters

Early expansion plays a fundamental role in the dynamical evolution of young star clusters. However, until very recently most of our understanding of cluster expansion was based only on indirect evidence or on statistically limited samples of clusters. Here we present a comprehensive kinematic analysis of virtually all known young Galactic clusters (t < 300 Myr) based on the improved astrometric quality of the Gaia DR3 data. Such a large sample provides an unprecedented opportunity to robustly constrain the fraction of clusters and the timescale during which expansion has a prominent impact on the overall kinematics. We find that a remarkable fraction (up to 80%) of clusters younger than ∼30 Myr is currently experiencing significant expansion, whereas older systems are mostly compatible with equilibrium configurations. We observe a trend in which the expansion speed increases with the cluster-centric distance, suggesting that clusters undergoing expansion will likely lose a fraction of their present-day mass. Also, most young expanding clusters are extended, possibly due to the expansion itself. A comparison with a set of N-body simulations of young star clusters shows that the observed expansion pattern is in general qualitative agreement with that found for systems undergoing violent relaxation and evolving toward a final virial equilibrium state. However, we also note that additional processes likely associated with residual gas expulsion and mass loss due to stellar evolution likely also play a key role in driving the observed expansion.

A stellar relic filament in the Orion star-forming region

ABSTRACT We report the discovery of the oldest stellar substructure in the Orion star-forming region (OSFR), the Orion relic filament. The relic filament is physically associated with the OSFR, as demonstrated by Gaia DR2 photometry and astrometry, as well as by targeted radial velocity follow-up observations of a bright subsample of proper motion-selected candidate members. Gaia DR2 parallaxes place the Orion relic filament in the more distant part of the OSFR, ∼430 pc from the Sun. Given its age, velocity dispersion, spatial extent, and shape, it is not possible for the Orion relic filament to have formed as a single stellar cluster, even taking into account residual gas expulsion. The relic filament is also too young to be a tidal stream, because Galactic tides act on much longer time-scales, of the order of 100 Myr. It therefore appears likely that the structure formed from a molecular cloud filament similar to Orion A in the OSFR and retained its morphology despite decoupling from its natal gas. Hence, the Orion relic filament bears witness to the short-lived evolutionary phase between gas removal and dispersion owing to shears and tides, and provides crucial new insights into how stars are formed in molecular clouds.

The Gaia-ESO Survey: asymmetric expansion of the Lagoon Nebula cluster NGC 6530 from GES and Gaia DR2

The combination of precise radial velocities from multi-object spectroscopy and highly accurate proper motions from Gaia DR2 opens up the possibility for detailed 3D kinematic studies of young star forming regions and clusters. Here, we perform such an analysis by combining Gaia-ESO Survey spectroscopy with Gaia astrometry for ~900 members of the Lagoon Nebula cluster, NGC 6530. We measure the 3D velocity dispersion of the region to be $5.35^{+0.39}_{-0.34}$~km~s$^{-1}$, which is large enough to suggest the region is gravitationally unbound. The velocity ellipsoid is anisotropic, implying that the region is not sufficiently dynamically evolved to achieve isotropy, though the central part of NGC 6530 does exhibit velocity isotropy that suggests sufficient mixing has occurred in this denser part. We find strong evidence that the stellar population is expanding, though this is preferentially occurring in the declination direction and there is very little evidence for expansion in the right ascension direction. This argues against a simple radial expansion pattern, as predicted by models of residual gas expulsion. We discuss these findings in the context of cluster formation, evolution and disruption theories.

TheGaia-ESO Survey: a kinematical and dynamical study of four young open clusters

Context.The origin and dynamical evolution of star clusters is an important topic in stellar astrophysics. Several models have been proposed in order to understand the formation of bound and unbound clusters and their evolution, and they can be tested by examining the kinematical and dynamical properties of clusters over a wide range of ages and masses.Aims.We use theGaia-ESO Survey products to study four open clusters (IC 2602, IC 2391, IC 4665, and NGC 2547) that lie in the age range between 20 and 50 Myr.Methods.We employ the gravity indexγand the equivalent width of the lithium line at 6708 Å together with effective temperatureTeffand the metallicity of the stars in order to discard observed contaminant stars. Then we derive the cluster radial velocity dispersionsσc, the total cluster massMtot, and the half mass radiusrhm. Using theGaia-DR1 TGAS catalogue, we independently derive the intrinsic velocity dispersion of the clusters from the astrometric parameters of cluster members.Results.The intrinsic radial velocity dispersions derived by the spectroscopic data are higher than those derived from the TGAS data, possibly due to the different masses of the considered stars. UsingMtotandrhmwe derive the virial velocity dispersionσvirand we find that three out of four clusters are supervirial. This result is in agreement with the hypothesis that these clusters are dispersing, as predicted by the “residual gas expulsion” scenario. However, recent simulations show that the virial ratio of young star clustersmay be overestimated if it is determined using the global velocity dispersion, since the clusters are not fully relaxed.

The kinematics of the Scorpius-Centaurus OB association from Gaia DR1

We present a kinematic study of the Scorpius-Centaurus (Sco-Cen) OB association (Sco OB2) using Gaia DR1 parallaxes and proper motions. Our goal is to test the classical theory that OB associations are the expanded remnants of dense and compact star clusters disrupted by processes such as residual gas expulsion. Gaia astrometry is available for 258 out of 433 members of the association, with revised Hipparcos astrometry used for the remainder. We use these data to confirm that the three subgroups of Sco-Cen are gravitationally unbound and have non-isotropic velocity dispersions, suggesting that they have not had time to dynamically relax. We also explore the internal kinematics of the subgroups to search for evidence of expansion. We test Blaauw's classical linear model of expansion, search for velocity trends along the Galactic axes, compare the expanding and non-expanding convergence points, perform traceback analysis assuming both linear trajectories and using an epicycle approximation, and assess the evidence for expansion in proper motions corrected for virtual expansion/contraction. None of these methods provide coherent evidence for expansion of the subgroups, with no evidence to suggest that the subgroups had a more compact configuration in the past. We find evidence for kinematic substructure within the subgroups that supports the view that they were not formed by the disruption of individual star clusters. We conclude that Sco-Cen was likely to have been born highly substructured, with multiple small-scale star formation events contributing to the overall OB association, and not as single, monolithic burst of clustered star formation.

How can young massive clusters reach their present-day sizes?

The classic question that how young massive star clusters attain their shapes and sizes, as we find them today, remains to be a challenge. Both observational and computational studies of star-forming massive molecular gas clouds infer that massive cluster formation is primarily triggered along the small-scale ($\lesssim0.3$ pc) filamentary substructures within the clouds. The present study is intended to investigate the possible ways in which a filament-like-compact, massive star cluster (effective radius 0.1-0.3 pc) can expand $\gtrsim10$ times, still remaining massive enough ($\gtrsim10^4 M_\odot$), to become a young massive star cluster, as we observe today. To that end, model massive clusters (of initially $10^4 M_\odot-10^5 M_\odot$) are evolved using Sverre Aarseth's state-of-the-art N-body code NBODY7. All the computed clusters expand with time, whose sizes (effective radii) are compared with those observed for young massive clusters, of age $\lesssim100$ Myr, in the Milky Way and other nearby galaxies. It is found that beginning from the above compact sizes, a star cluster cannot expand by its own, i.e., due to two-body relaxation, stellar-evolutionary mass loss, dynamical heating by primordial binaries and stellar-mass black holes, up to the observed sizes of young massive clusters; they always remain much more compact compared to the observed ones. This calls for additional mechanisms that can boost the expansion of a massive cluster after its assembly. Using further N-body calculations, it is shown that a substantial residual gas expulsion, with $\approx30$% star formation efficiency, can indeed swell the newborn embedded cluster adequately. The limitations of the present calculations and their consequences are discussed.

Cygnus OB2 DANCe: A high-precision proper motion study of the Cygnus OB2 association

We present a high-precision proper motion study of 873 X-ray and spectroscopically selected stars in the massive OB association Cygnus OB2 as part of the DANCe project. These were calculated from images spanning a 15 year baseline and have typical precisions < 1 mas/yr. We calculate the velocity dispersion in the two axes to be $\sigma_\alpha(c) = 13.0^{+0.8}_{-0.7}$ and $\sigma_\delta(c) = 9.1^{+0.5}_{-0.5}$ km/s, using a 2-component, 2-dimensional model that takes into account the uncertainties on the measurements. This gives a 3-dimensional velocity dispersion of $\sigma_{3D} = 17.8 \pm 0.6$ km/s implying a virial mass significantly larger than the observed stellar mass, confirming that the association is gravitationally unbound. The association appears to be dynamically unevolved, as evidenced by considerable kinematic substructure, non-isotropic velocity dispersions and a lack of energy equipartition. The proper motions show no evidence for a global expansion pattern, with approximately the same amount of kinetic energy in expansion as there is in contraction, which argues against the association being an expanded star cluster disrupted by process such as residual gas expulsion or tidal heating. The kinematic substructures, which appear to be close to virial equilibrium and have typical masses of 40-400 M$_\odot$, also do not appear to have been affected by the expulsion of the residual gas. We conclude that Cyg OB2 was most likely born highly substructured and globally unbound, with the individual subgroups born in (or close to) virial equilibrium, and that the OB association has not experienced significant dynamical evolution since then.

Possible smoking-gun evidence for initial mass segregation in re-virialized post-gas expulsion globular clusters

We perform a series of direct $N$-body calculations to investigate the effect of residual gas expulsion from the gas-embedded progenitors of present-day globular clusters (GCs) on the stellar mass function (MF). Our models start either tidally filling or underfilling, and either with or without primordial mass segregation. We cover 100 Myr of the evolution of modeled clusters and show that the expulsion of residual gas from initially mass-segregated clusters leads to a significantly shallower slope of the stellar MF in the low- ($m\leq 0.50 M_\odot$) and intermediate-mass ($\simeq 0.50-0.85 M_\odot$) regime. Therefore, the imprint of residual gas expulsion and primordial mass segregation might be visible in the present-day MF. We find that the strength of the external tidal field, as an essential parameter, influences the degree of flattening, such that a primordially mass-segregated tidally-filling cluster with $r_h/r_t\geq 0.1$ shows a strongly depleted MF in the intermediate stellar mass range. Therefore, the shape of the present-day stellar MF in this mass range probes the birth place of clusters in the Galactic environment. We furthermore find that this flattening agrees with the observed correlation between the concentration of a cluster and its MF slope, as found by de Marchi et al.. We show that if the expansion through the residual gas expulsion in primordial mass segregated clusters is the reason for this correlation then GCs most probably formed in strongly fluctuating local tidal fields in the early proto-Milky Way potential, supporting the recent conclusion by Marks \& Kroupa.

Initial conditions of formation of starburst clusters: constraints from stellar dynamics

AbstractHow starburst clusters form out of molecular clouds is still an open question. In this article, I highlight some of the key constraints in this regard, that one can get from the dynamical evolutionary properties of dense stellar systems. I particularly focus on secular expansion of massive star clusters and hierarchical merging of sub-clusters, and discuss their implicationsvis-á-visthe observed properties of young massive clusters. The analysis suggests that residual gas expulsion is necessary for shaping these clusters as we see them today, irrespective of their monolithic or hierarchical mode of formation.

The formation of NGC 3603 young starburst cluster: ‘prompt’ hierarchical assembly or monolithic starburst?

The formation of very young massive clusters or "starburst" clusters is currently one of the most widely debated topic in astronomy. The classical notion dictates that a star cluster is formed in-situ in a dense molecular gas clump followed by a substantial residual gas expulsion. On the other hand, based on the observed morphologies of many young stellar associations, a hierarchical formation scenario is alternatively suggested. A very young (age $\approx$ 1 Myr), massive ($>10^4M_\odot$) star cluster like the Galactic NGC 3603 young cluster (HD 97950) is an appropriate testbed for distinguishing between such "monolithic" and "hierarchical" formation scenarios. A recent study by Banerjee and Kroupa (2014) demonstrates that the monolithic scenario remarkably reproduces the HD 97950 cluster. In the present work, we explore the possibility of the formation of the above cluster via hierarchical assembly of subclusters. These subclusters are initially distributed over a wide range of spatial volumes and have various modes of sub-clustering. Unlike the monolithic model that reproduces HD 97950 very well, the same is found to be prohibitive with hierarchical assembly alone. Only those systems which assemble promptly into a single cluster (in $\lesssim1$ Myr) from a close separation (all within $\lesssim2$ pc) could match the observed density profile of HD 97950 after $\approx70$% gas dispersal. These results therefore suggest that NGC 3603 young cluster has formed essentially monolithically, i.e., either in-situ or via a prompt assembly, followed by a substantial residual gas expulsion. Both scenarios are consistent with the inferred young age and the small age spread of this cluster. Future observations of molecular cloud filaments with ALMA and proper motion measurements of young clusters with Gaia will provide more direct tests of such birth environments. [Abridged]

A PERFECT STARBURST CLUSTER MADE IN ONE GO: THE NGC 3603 YOUNG CLUSTER

Understanding how distinct, near-spherical gas-free clusters of very young, massive stars shape out of vast, complex clouds of molecular hydrogen is one of the biggest challenges in astrophysics. A popular thought dictates that a single gas cloud fragments into many new-born stars which, in turn, energize and rapidly expel the residual gas to form a gas-free cluster. This study demonstrates that the above classical paradigm remarkably reproduces the well-observed central, young cluster (HD 97950) of the Galactic NGC 3603 star-forming region, in particular, its shape, internal motion and the mass distribution of stars, naturally and consistently follow from a single model calculation. Remarkably, the same parameters (star formation efficiency, gas expulsion time scale and delay) reproduce HD 97950 as were found to reproduce the Orion Nebula Cluster, Pleiades and R136. The present results thereby provide intriguing evidences of formation of star clusters through single-starburst events followed by significant residual gas expulsion.

The star cluster formation history of the LMC

The Large Magellanic Cloud is one of the nearest galaxies to us and is one of only few galaxies where the star formation history can be determined from studying resolved stellar populations. We have compiled a new catalogue of ages, luminosities and masses of LMC star clusters and used it to determine the age distribution and dissolution rate of LMC star clusters. We find that the frequency of massive clusters with masses M>5000 Msun is almost constant between 10 and 200 Myr, showing that the influence of residual gas expulsion is limited to the first 10 Myr of cluster evolution or clusters less massive than 5000 Msun. Comparing the cluster frequency in that interval with the absolute star formation rate, we find that about 15% of all stars in the LMC were formed in long-lived star clusters that survive for more than 10 Myr. We also find that the mass function of LMC clusters younger than 1 Gyr can be fitted by a power-law mass function with slope \alpha=-2.3, while older clusters follow a significantly shallower slope and interpret this is a sign of the ongoing dissolution of low-mass clusters. Our data shows that for ages older than 200 Myr, about 90% of all clusters are lost per dex of lifetime. The implied cluster dissolution rate is significantly faster than that based on analytic estimates and N-body simulations. Our cluster age data finally shows evidence for a burst in cluster formation about 1 Gyr ago, but little evidence for bursts at other ages.

DID THE INFANT R136 AND NGC 3603 CLUSTERS UNDERGO RESIDUAL GAS EXPULSION?

Based on kinematic data observed for very young, massive clusters that appear to be in dynamical equilibrium, it has recently been argued that such young systems set examples where the early residual gas-expulsion did not happen or had no dynamical effect. The intriguing scenario of a star cluster forming through a single starburst has thereby been challenged. Choosing the case of the R136 cluster of the Large Magellanic Cloud, the most cited one in this context, we perform direct N-body computations that mimic the early evolution of this cluster including the gas-removal phase (on a thermal timescale). Our calculations show that under plausible initial conditions as consistent from observational data, a large fraction (> 60%) of a gas-expelled, expanding R136-like cluster is bound to regain dynamical equilibrium by its current age. Therefore, the recent measurements of velocity dispersion in the inner regions of R136, that indicate that the cluster is in dynamical equilibrium, are consistent with an earlier substantial gas expulsion of R136 followed by a rapid re-virialization (in approx. 1 Myr). Additionally, we find that the less massive Galactic NGC 3603 Young Cluster (NYC), with a substantially longer re-virialization time, is likely to be found deviated from dynamical equilibrium at its present age (approx. 1 Myr). The recently obtained stellar proper motions in the central part of the NYC indeed suggest this and are consistent with the computed models. This work significantly extends previous models of the Orion Nebula Cluster which already demonstrated that the re-virialization time of young post-gas-expulsion clusters decreases with increasing pre-expulsion density.

Local-density-driven clustered star formation

A positive power-law trend between the local surface densities of molecular gas, $\Sigma_{gas}$, and young stellar objects, $\Sigma_{\star}$, in molecular clouds of the Solar Neighbourhood has been identified by Gutermuth et al. How it relates to the properties of embedded clusters, in particular to the recently established radius-density relation, has so far not been investigated. In this paper, we model the development of the stellar component of molecular clumps as a function of time and initial local volume density so as to provide a coherent framework able to explain both the molecular-cloud and embedded-cluster relations quoted above. To do so, we associate the observed volume density gradient of molecular clumps to a density-dependent free-fall time. The molecular clump star formation history is obtained by applying a constant SFE per free-fall time, $\eff$. For volume density profiles typical of observed molecular clumps (i.e. power-law slope $\simeq -1.7$), our model gives a star-gas surface-density relation $\Sigma_{\star} \propto \Sigma_{gas}^2$, in very good agreement with the Gutermuth et al relation. Taking the case of a molecular clump of mass $M_0 \simeq 10^4 Msun$ and radius $R \simeq 6 pc$ experiencing star formation during 2 Myr, we derive what SFE per free-fall time matches best the normalizations of the observed and predicted ($\Sigma_{\star}$, $\Sigma_{gas}$) relations. We find $\eff \simeq 0.1$. We show that the observed growth of embedded clusters, embodied by their radius-density relation, corresponds to a surface density threshold being applied to developing star-forming regions. The consequences of our model in terms of cluster survivability after residual star-forming gas expulsion are that due to the locally high SFE in the inner part of star-forming regions, global SFE as low as 10% can lead to the formation of bound gas-free star clusters.

The evolution of the surface brightness of a star cluster as a result of residual star-forming gas expulsion

Direct N-body calculations are presented of the early evolution of exposed clusters to quantify the influence of gas expulsion on the time-varying surface brightness. By assuming that the embedded OB stars drive out most of the gas after a given time delay, the change of the surface brightness of expanding star clusters is studied. The influence of stellar dynamics and stellar evolution is discussed. The growth of the core radii of such models shows a remarkable core re-virialisation. The decrease of the surface mass density during gas expulsion is large and is only truncated by this re-virialisation process. However, the surface brightness within a certain radius does not increase noticeably. Thus, an embedded star cluster cannot reappear in observational surveys after re-virialisation. This finding has a bearing on the observed infant mortality fraction.

Evidence for top-heavy stellar initial mass functions with increasing density and decreasing metallicity

Residual-gas expulsion after cluster formation has recently been shown to leave an imprint in the low-mass present-day stellar mass function (PDMF) which allowed the estimation of birth conditions of some Galactic globular clusters (GCs) such as mass, radius and star formation efficiency. We show that in order to explain their characteristics (masses, radii, metallicity, PDMF) their stellar initial mass function (IMF) must have been top-heavy. It is found that the IMF is required to become more top-heavy the lower the cluster metallicity and the larger the pre-GC cloud-core density are. The deduced trends are in qualitative agreement with theoretical expectation. The results are consistent with estimates of the shape of the high-mass end of the IMF in the Arches cluster, Westerlund 1, R136 and NGC 3603, as well as with the IMF independently constrained for ultra-compact dwarf galaxies (UCDs). The latter suggests that GCs and UCDs might have formed along the same channel or that UCDs formed via mergers of GCs. A fundamental plane is found which describes the variation of the IMF with density and metallicity of the pre-GC cloud-cores simultaneously. The implications for the evolution of galaxies and chemical enrichment over cosmological times are expected to be major.

From the molecular-cloud- to the embedded-cluster-mass function with a density threshold for star formation

The mass function of molecular clouds and clumps is shallower than the mass function of young star clusters, gas-embedded and gas-free alike, as their respective mass function indices are $\beta_0 \simeq 1.7$ and $\beta_\star \simeq 2$. We demonstrate that such a difference can arise from different mass-radius relations for the embedded-clusters and the molecular clouds (clumps) hosting them. In particular, the formation of star clusters with a constant mean {\it volume} density in the central regions of molecular clouds of constant mean {\it surface} density steepens the mass function from clouds to embedded-clusters. This model is observationally supported since the mean surface density of molecular clouds is approximately constant, while there is a growing body of evidence, in both Galactic and extragalactic environments, that efficient star-formation requires a hydrogen molecule number density threshold of $n_{th} \simeq 10^{4-5}\,cm^{-3}$. In the framework of the same model, the radius distribution steepens from clouds (clumps) to embedded-clusters, which contributes to explaining observed cluster radius distributions. [Abridged]

Initial conditions for globular clusters and assembly of the old globular cluster population of the Milky Way

Comparing N-body calculations that include primordial residual-gas expulsion with the observed properties of 20 Galactic globular clusters (GCs) for which the stellar mass function (MF) has been measured, we constrain the time-scale over which the gas of their embedded counterparts must have been removed, the star formation efficiency the progenitor cloud must have had and the strength of the tidal-field the clusters must have formed in. The three parameters determine the expansion and mass-loss during residual-gas expulsion. After applying corrections for stellar and dynamical evolution we find birth cluster masses, sizes and densities for the GC sample and the same quantities for the progenitor gas clouds. The pre-cluster cloud core masses were between 10^5-10^7 Msun and half-mass radii were typically below 1 pc and reach down to 0.2 pc. We show that the low-mass present day MF slope, initial half-mass radius and initial density of clusters correlates with cluster metallicity, unmasking metallicity as an important parameter driving cluster formation and the gas expulsion process. This work predicts that PD low-concentration clusters should have a higher binary fraction than PD high-concentration clusters. Since the oldest GCs are early residuals from the formation of the Milky Way (MW) and the derived initial conditions probe the environment in which the clusters formed, we use the results as a new tool to study the formation of the inner GC system of the Galaxy. We achieve time-resolved insight into the evolution of the pre-MW gas cloud on short time-scales (a few hundred Myr) via cluster metallicities. The results are shown to be consistent with a contracting and self-gravitating cloud in which fluctuations in the pre-MW potential grow with time. An initially relatively smooth tidal-field evolved into a grainy potential within a dynamical time-scale of the collapsing cloud.

Evolution of two stellar populations in globular clusters

We investigate the early evolution of two distinct populations of low-mass stars in globular clusters under the influence of primordial gas expulsion driven by supernovae to study if this process can increase the fraction of second generation stars at the level required by observations. We analyse N-body models that take into account the effect of primordial gas expulsion. We divide the stars into two populations which mimic the chemical and dynamical properties of stars in globular clusters so that second generation stars start with a more centrally concentrated distribution. The main effect of gas expulsion is to eject preferentially first generation stars while second generation stars remain bound to the cluster. In the most favourable cases second generation stars can account for 60% of the bound stars we see today. We also find that at the end of the gas expulsion phase, the radial distribution of the two populations is still different, so that long-term evolution will further increase the fraction of second generation stars. The large fraction of chemically anomalous stars is readily explainable as a second generation of stars formed out of the slow winds of rapidly rotating massive stars if globular clusters suffer explosive residual gas expulsion for a star formation efficiency of about 0.33.

Dynamical mixing of two stellar populations in globular clusters

AbstractStars in globular clusters (GCs) exhibit a peculiar chemical pattern with strong abundance variations in light elements along with a constant abundance in heavy elements. These abundance anomalies can be explained by a primordial pollution due to a first generation of fast rotating massive stars which released slow winds into the ISM from which a second generation of chemically anomalous stars can be formed. In particular the observed ratio of anomalous and standard stars in clusters can be used to constrain the dynamical evolution of GCs as around 95% of the standard stars need to be lost by the clusters. We show that both residual gas expulsion during the cluster formation and long term evolution are needed to achieve this ratio. (© 2008 WILEY‐VCH Verlag GmbH &amp; Co. KGaA, Weinheim)