One of the outstanding goals of modern astronomy is to understand how stars and planets form and evolve. There is a generally accepted theory on how a star forms through the collapse of a molecular cloud core and contracts to the main-sequence1,2
. This process produces a distribution of stellar masses that is empirically described as a power-law at high masses (M>1MSun
) and a lognormal distribution at lower masses (or a combination of power laws) called the Initial Mass Function (IMF)3,4
. This distribution provides unique quantitative evidence to the underlying physics in the star formation process. Understanding the origin of this stellar mass distribution is crucial at different levels, since it includes the study of the basic physical laws that govern the observable universe, and the formation and evolution of galaxies, strongly dependent on its stellar content.
Understanding the IMF and its origin in our Galaxy, is paramount to address questions on the IMF of other galaxies or even of the stellar mass distribution in the early universe. Moreover, it is fundamental to understand why most stars have masses similar to our Sun (the IMF has a peak at half of a Solar mass), and therefore gather all the conditions to produce planets within an habitable zone, which ultimately can harbour life. At the lower mass end, the discovery of the first brown dwarf (BD) and extrasolar planets in 19955,6,7,8
imply that one of the key elements in understanding the IMF is the study of the formation of these objects and the relation between BDs and extrasolar planets.
Brown dwarfs are objects that have masses below 75MJupiter
, and according to stellar interior models are not massive enough to ignite hydrogen fusion9
, the source of energy used by stars to establish the equilibrium with gravitational energy, allowing them to have a complete life cycle. In broad terms, these objects are thought to form in the same way as Hydrogen burning stars, on a dynamical timescale in a gravity dominated process, constituted mainly of light elements. This is intrinsically different from the formation process of a planet around a star, which is thought to occur on a much longer timescale, involving the formation of a rocky core, and to have a deficit of light elements in its constitution10
. There is however an overlap in terms of masses, with the least massive BDs coinciding with the heaviest planets, and any successful star formation theory must be able to explain both domains. There is not yet a consensus regarding their formation mechanism and several scenarios are currently proposed. These defend either a star-like formation, i.e. fragmentation of molecular clouds into low mass cores driven by turbulence11,12
; or distinct formation processes such as gravitational instabilities in discs14
, premature ejection from pre-stellar cores15
, or photo-erosion of cores16. These mechanisms are not mutually exclusive and could be all at play simultaneously within a cluster.
The existing observational properties of BDs in young clusters show a global scaling down trend from those of stars, arguing in favour of a common formation scenario17
. However, up to date, most studies in young star forming regions suffer from incompleteness both at lower masses (below ~40MJup) and in the spatial content, frequently focusing on the inner regions of clusters. It is therefore unknown if, as one moves to lower masses, other formation mechanisms take over18
. At the same time, the increasing capacity of numerical simulations to reproduce the collapse of entire molecular clouds is such that the comparison between simulations and observations starts to be hampered by our small-number statistics empirical knowledge of the low-mass end of the brown dwarf regime. Extensive studies of the lowest mass populations in different environments and stages of evolution are therefore required to address these pressing questions. For that end, we have carried out a large photometric survey at optical and near-IR wavelengths of various molecular clouds using MegaCam and WIRCam/CFHT. We adopted a twofold approach to this study, using both near-IR photometry to find all candidate substellar members and extensive spectroscopic follow-up to derive a spectroscopic IMF, and deep methane imaging to search for the lowest mass members.
The Rho Ophiuchi Molecular Cloud
We selected from the WIRCam/CFHT JHKs catalogue of 57000 sources, 110 candidate members of Rho Oph. By carrying out various spectroscopic follow-up campaigns, we have confirmed 19 new brown dwarfs as member of the cluster, more than doubling the know substellar population. By compiling a census of the spectroscopically confirmed population from the literature, totalling 250 members, we find that the distribution of spectral types peaks at M5, similar to what is seen in IC 348 and Cha I. The mass function peaks at 0.12 MSun
, which is consistent with what is found in other clusters and the field. Based on these results, we conclude that there is no evidence for variations in the mass function of ρ Oph when compared to other nearby young clusters. Read more about the results in the original articles (Alves de Oliveira et al. 2010
, Alves de Oliveira et al. 2012
The IC 348 cluster
We have observed the IC 348 cluster with MegaCam and WIRCam at the CFHT, and uncovered several candidate brown dwarfs. We are finalizing the analysis of two large spectroscopic follow-up campaigns, where we confirm the membership and youth of new isolated, planetary-mass brown dwarfs. Keep an eye out for any updates, we are preparing an article on this very interesting results!
Other projects within the WIRCam survey
Check other projects of this survey; I will update this list whenever new results are published:
Searching for planetary-mass T-dwarfs in the core of Serpens (submitted)
L. Spezzi1, C. Alves de Oliveira, E. Moraux, J. Bouvier, E. Winston, and P. Hudelot
Young T-dwarf candidates in IC 348 (A&A 508, 823-831)
A. S. M. Burgess, E. Moraux, J. Bouvier, C. Marmo, L. Albert, and H. Bouy 2009.
References: 1. Shu et al. 1987, 2. Larson 1973, 3. Kroupa 2001; 4. Chabrier 2003; 5. Rebolo et al. 1995; 6. Nakajima et al. 1995; 7. Oppenheimer et al. 1995; 8. Mayor et al. 1995; 9. Chabrier et al. 1997; 10. Whitworth et al. 2007; 11. Padoan & Nordlund 2002; 12. Hennebelle & Chabrier 2008; 13. Bonnel et al. 2008; 14. Stamatellos & Whitworth 2009; 15. Reipurth & Clarke 2001; 16. Kroupa & Bouvier 2003; 17. Luhman et al. 2007; 18. Whitworth et al. 2010