Work Package 2
From CONSTELLATION
| Table of contents |
The birth and influence of massive stars
Massive stars in excess of 20 solar masses are relatively rare and yet a vital outcome of the star formation process: they dominate the energy feedback in the interstellar medium, produce all elements heavier than carbon, and are responsible for virtually all the visible light seen from the distant universe. However, despite progress in understanding the formation of low-mass stars like the Sun, the birth of massive stars remains deeply enigmatic. The fundamental problem is that the radiation pressure from a highly-luminous mas- sive protostar should rapidly build up and stop accretion completely, before the star can achieve the sorts of high masses actually observed (Yorke & Krugel 1977; Wolfire & Casinelli 1987). Several theories have been developed to resolve this quandary. One posits that the standard model of low-mass star formation can indeed be scaled-up and that sufficiently high accretion rates can overcome the radiation pressure (McKee & Tan 2003; Palouˇs et al. 2004). An alternative scenario suggests that massive stars form in a densely clustered environment, where gas is funnelled by the gravitational potential of the cluster to its centre and then accreted by the developing massive star (Bonnell et al. 2001; Bonnell, Vine, & Bate 2004). In both models, the radiation pressure could be circumvented through disk accretion, radiation beaming (Yorke & Sonnhalter 2002), or Rayleigh-Taylor instabilities in the infalling gas (Krumholz et al. 2005). A third option involves mergers of single or binary low-mass stars in the core of a dense cluster, thus avoiding the radiation pressure entirely (Bonnell, Bate, & Zinnecker 1998; Stahler, Palla, & Ho2000; Bonnell & Bate 2005; Bally & Zinnecker 2005). Young high-mass stars are hard to study because they are relatively rare, form and evolve rapidly, and are heavily enshrouded in gas and dust at birth. As a consequence, thermal-IR, millimetre, and radio observations are required to survey for sites of massive star formation and then study them in detail. Newly or soon to be available space IR telescopes, millimetre imaging systems, and radio interferometers now enable such surveys and follow-up studies, and CONSTELLATION will enjoy excellent access to proprietary and publicly-available surveys, making it possible to build a detailed picture of massive star- formation in the Milky Way. Advances in computational techniques also ensure that such observations can be compared to state-of-the-art models of the massive star formation process. Key questions include:
— Are massive stars born via accretion like low-mass stars or via a separate process, e.g., coalescence?
— Do massive stars always form in stellar clusters and associations or can they form in isolation?
— Why are massive stars relatively rare and is there an upper limit to their masses?
— Does feedback from massive stars trigger new star formation or halt the process completely?
Task 2A—The initial conditions for massive star formation
A key task is to study the initial conditions in the recently-discovered IR dark clouds, extremely dense, massive molecular cloud cores which are likely to form massive stars (Menten, Pillai, & Wyrowski 2005). For high-mass stars to form like low-mass stars, these high-mass cores should be near virial equilibrium and resistant to fragmentation. In contrast, cluster-mode massive star formation models predict a highly unstable configuration with significant substructure to seed fragmentation and a large extended gas component allowing accretion to form the massive stars (Bonnell et al. 2004). Finally, stellar coalescence requires extreme densities of intermediate-mass stars in the centres of the very young star clusters which can be measured via radio in- terferometry of masers and thermal-IR interferometry of the stars themselves. Thus a detailed knowledge of the structure and kinematics of these massive molecular cores using far-IR (e.g., Herschel) to radio continuum and line observations (e.g., ATCA, VLA) is essential for comparison with numerical models, in order to distinguish between the various formation mechanisms.
Task 2B—The influence of environment on massive star formation
Once formed, direct measurements of massive stars and their environment can also provide important insights into their formation mechanisms. Searches for clustering of low-mass stars around them will be used to establish whether massive star formation requires the presence of a stellar cluster to provide the mass reservoir (Bonnell et al. 2004), or if some can truly form in isolation (de Wit et al. 2005). Millimetre interferometry will be employed to probe the circumstellar gas around young massive stars: well-ordered disks and colli- mated outflows would argue for a scaled-up version of low-mass star formation, while disordered disks would argue in favour of mergers (Bally & Zinnecker 2005). Finally, massive stars are even more likely to be members of binary and multiple systems than their low-mass counterparts (Preibisch et al. 1999). Measurements of their properties (mass ratios, hierarchical distribution) using adaptive optics and inter- ferometry at near-IR wavelengths can also be used to provide important tests of new, high-resolution numerical models of massive star formation.
Task 2C—Feedback from massive stars
Mechanical and radiative feedback driven by winds and ionisation from massive stars are vitally important, not only affecting massive star formation itself (Yorke & Sonnhalter 2002), but also affecting the surrounding environment, unbinding stellar clusters, suppressing and/or triggering new episodes of star formation (Hennebelle et al. 2003; Kessel-Deynet & Burkert 2003), and evaporating the circumstellar disks of neighbouring stars. Radio and X-ray observations of the ambi- ent gas will be used to trace the feedback from massive stars and its role in cloud dispersal. Conversely, likely locations of shock-triggered star formation near H II regions (e.g., Deharveng et al. 2003) will be identified by optical and radio observations, followed with wide-field IR imaging and sub-millimetre mapping to search for young stars and protostars. These observations will be compared to predictions of numerical simulations in which it is now possible to include feedback (e.g., Dale et al. 2005). The balance between the destructive effects of photoionisation and stellar winds, and their potentially constructive role in triggering successive new generations of stars will be examined.