Work Package 3

The physics of the low-mass end of the IMF

[WP3 internal pages and updates]

For all their importance, massive stars are vastly outnumbered by the low-mass stars and brown dwarfs (BDs) below the stellar/sub-stellar boundary at 0.075 solar masses, which populate the lower end of the IMF: indeed, current estimates are that there are probably roughly equal numbers of stars and BDs in the Milky Way (Chabrier 2003). In the past decade, many old field BDs have been discovered in wide-field optical and IR surveys, resulting in the definition of two new spectral types, the L and T dwarfs (e.g., Martın et al. 1999). BDs have also been detected in young clusters and there appear to be substantial populations of BDs down to at least 0.030 solar masses (e.g., Moraux et al. 2003). In the youngest clusters (∼ 1 Myr), the IMF extends as low as 5–15 MJup , although there is considerable ambiguity in its form at these masses (e.g., McCaughrean et al. 2002; Lucas, Roche, & Tamura 2005). These very lowest-mass objects overlap in mass with giant planets and thus likely share their physical characteristics (e.g., Allard et al. 2001). Crucially, these isolated BDs can be used as proxies for close-in planets orbiting bright stars, as they are considerably easier to study. Finally, the IMF must end at some point: recent theoretical studies confirm long-standing thoughts that this limit should lie at around 5 MJup (Boyd & Whitworth 2005), and observational confirmation or denial of these predictions would provide valuable insight into the physics of BD formation and the distinction between star- and planet-formation processes. Despite their ubiquity, it is not yet known whether BDs form via fragmentation in molecular cores like stars (Padoan & Nordlund 2002), via agglomeration in circumstellar disks like planets, or via yet another mechanism, such as early ejection from a multiple star system via dynamical encounters (Reipurth & Clarke 2001; Bate et al. 2002; Umbreit et al. 2005) or photoevaporation of cores by ionising radiation from massive stars (Whitworth & Zinnecker 2003). Using new wide-field IR cameras, CONSTELLATION will carry out the largest and deepest near-IR surveys to date searching for BDs and planetary-mass objects in star-forming regions and young open clusters within 500 pc, followed by multi-object spectroscopy to confirm membership. Selected clusters will also be surveyed at X-ray wavelengths, which provide complementary method of detecting very low-mass objects in star-forming regions (e.g., Mokler & Stelzer 2002). By studying their individual and group characteristics (e.g., mass function, kinematics, location relative to massive stars, circumstellar disk properties), it will be possible to shed light on the origin of these very low-mass objects:

— Are low-mass stars and BDs simply the low-mass limit of star formation or do they form differently?

— How do the lowest-mass brown dwarfs relate to planets; are they fundamentally different objects?

— Is the low-mass IMF universal or does it depend on environment?

— Is there continuity between the mass function of the lowest-mass BDs and that of planets?

Task 3A—The observational properties of young brown dwarfs

The various models of low-mass star and BD formation outlined above make specific and distinct predictions regarding the observational properties of BDs. For example, numerical simulations currently favour the ejection hypothesis and predict that low-mass stars and BDs will have a far lower multiplicity than stars ( ∼ 10%, compared to >50%) and relatively small (frequently <10 AU) circumstellar disks due to their ejection at ∼ 1–2 km s^-1 from small-N protostellar systems. Improved calculations will also provide a better statistical prediction for the proper motions and spatial distributions arising from the ejection scenario. Explicit tests of the ejection hypothesis will be made by determining the spatial distribution of BDs in star-forming regions and by searching for resolved binary BD systems via high angular resolution space imaging, adaptive optics, and interferometry, and spectroscopic binaries using high spectral resolution spectrographs. The identification of short-period binary BDs will also allow dynamical masses to be compared to theoretical models (Close et al. 2005; Bouy et al. 2005). Finally, the sizes and masses of disks around BDs in star-forming regions will be derived from space IR and ground-based millimetre observations.

Task 3B: What is the form of the low-mass end of the IMF?

Recent surveys of low-mass stars and BDs in Orion and in Taurus suggest that the low-mass IMF may not be universal (Briceno et al. 2002; Luhman et al. 2003), but this result has recently been challenged by Guieu et al. (2005). If the apparent discrepancies in the IMFs are real, then there are two possibilities for the differences: that the outcome of star formation in different environments is different (e.g., Goodwin et al. 2003; Bate & Bonnell 2005), or that dynamical evolution in diffuse environments causes the depletion of the low-mass end of the IMF (e.g., Kroupa & Bouvier 2003). To settle this issue, combined wide-field, deep, and pencil beam IR surveys will be carried out in Orion, Ophiuchus, IC 348, NGC 2264, IC 4665, and the Pleiades, to obtain a complete census of their BD populations and examine the universality of the low-mass IMF down to masses of ≤0.013 solar masses.