The turbulent youth of globular clusters
April 11, 2013
The massive stellar clusters accompanying our galaxies as well as other galaxies have passed through a more complex evolution than previously thought. New observations have found evidence for several generations of stars, which can now be explained by a research team from MPE, the observatory of the University of Geneva, and the French science organisation CNRS. In their scenario, some of the first generation stars are much more massive than the left-over stars seen today. Their fast evolution up to violent supernova explosions will have substantial influence on the formation of the following stellar generations.
Globular clusters are real jewels in the sky. They consist of hundreds of thousands of stars, and belong to the oldest objects of the Universe. They probably formed at the same time as their host galaxy. Today, they can be observed in any type of galaxy; the Milky Way hosts almost 200, the Andromeda galaxy some 500, and 15,000 have been counted in the giant elliptical galaxy M87, which resides at the centre of the Virgo cluster of galaxies.
Recent observations with ESO’s Very Large Telescope and the Hubble Space Telescope have revealed the presence of multiple stellar generations, showing a large spread in chemical composition and in particular strong anomalies in sodium and oxygen. Up to now these objects were thought to consist of only one generation of stars with homogeneous chemical composition. The discovery of different generations of stars thus challenges one of the oldest paradigms of astrophysics and constitutes a real puzzle. A team of researchers from MPE, the observatory of the University of Geneva, and the French science organisation CNRS have now proposed a solution.
The group developed an innovative scenario, which shows the impact that a first generation of massive stars can have on their environment just after the formation of a globular cluster. Born in the most central regions of the cluster, these massive stars, which have long since disappeared, should in fact have spun close to the critical disruption velocity and thus lost a great deal of the heavier elements produced from hydrogen burning in their centre. The ejecta should have mixed with interstellar gas in a disc surrounding the star to give birth to stars of succeeding generations with the masses and the chemical compositions that are observed today. This new mode of star-formation in very dense environments that was now proposed by the team shows clear parallels to planet formation in circumstellar discs.
The scenario implies that the initial mass of globular clusters has been 20 to 30 times higher than today. For very massive clusters, like NGC 2808 (image) this means several million solar masses.
"Most of the massive first generation stars disappear some 40 million years after their formation into the galactic halo", explains Martin Krause from MPE. "This happens as the massive stars of the first generation become black holes and neutron stars, releasing huge amounts of energy. The leftover interstellar gas is expulsed from the cluster, which would considerably reduce the gravitational attraction of the globular cluster and favour the escape of some part of the stars."
The lost stars would constitute a large fraction of the Galactic halo and should be identifiable by ESA’s space mission GAIA which is to map the Milky Way and expected to be launched by the end of 2013.
Schematic illustration of how the second generation stars form in a disk. Massive stars enrich the surrounding interstellar gas with heavy elements produced in their interior by hydrogen burning. A new generation of stars is then formed in an accretion disk, much like planets form in dusty disks around stars.[less]
Schematic illustration of how the second generation stars form in a disk. Massive stars enrich the surrounding interstellar gas with heavy elements produced in their interior by hydrogen burning. A new generation of stars is then formed in an accretion disk, much like planets form in dusty disks around stars.
Next, the MPE scientists and their collaborators aim to confirm certain key-hypotheses in their scenario of the chemical and dynamical evolution of massive clusters. Using multi-dimensional hydrodynamics simulations they want to model in particular the interactions between the massive-star ejecta and the interstellar matter in extreme environments. This should help to better understand how star formation is induced - one of the hottest questions of modern astrophysics. Their work should also permit a redetermination of the age of globular clusters, an independent constraint on the age of the Universe.