Brown Dwarf Formation from Disk Fragmentation and Ejection

Transcription

1 Brown Dwarf Formation from Disk Fragmentation and Ejection Shantanu Basu Western University, London, Ontario, Canada Collaborator: Eduard Vorobyov (University of Vienna) 50 years of Brown Dwarfs Ringberg Castle, Germany Tuesday, October 23, 2012

2 Origin of Low End of Stellar Mass Function? IMF = initial mass function of stars, i.e., stellar mass distribution, average mass ~ 0.25 M sun Brown dwarfs Lowest mass clumps not well sampled, but also may not be gravitationally bound. Where do lowest mass stars come from? Data: Motte et al. (2001), Johnstone et al. (2001), and Nutter & Ward- Thompson (2007) Figure: André, Basu, & Inutsuka (2009)

3 Gravitational Collapse Jeans mass in present day star-forming molecular cloud: M J 3 6 1/2 3/ /2 cs T 10 cm G 10 K n M Can direct gravitational collapse from interstellar clouds explain low mass stars, brown dwarfs, planets? Disk fragmentation provides an alternate path due to high density and lifetime of many orbit times. Ejections can be invoked to explain free-floaters.

4 A Global Model, Nonaxisymmetric Model for Disk Formation/Evolution Thin disk approximation. Logarithmically spaced grid in r-direction, uniform in f direction Simulations require high resolution in the inner regions, while a lower resolution may be sufficient in the outer regions Models run with grids. Span large dynamic range in space (outer boundary at ~10,000 AU, but innermost grid resolution ~ 0.1 AU) and time (can follow evolution for several Myr after disk formation). Central sink cell with unresolved physics, size 6 AU.

5 Global Modeling, Thin-Disk Approx. Central star (polytrope) Inner inflow boundary (sink cell ~ 5 AU) Vertical motions are neglected, local vertical hydrostatic equilibrium is assumed.

6 Basic Equations, Thin-Disk Approximation u 0, t u uu P t 4 4 P C T T u u irr 2 t 1 1 u ur r u ; r r, r T T F, T 4 4 irr bg irr bg / 2, st background temperature, F =A cos - stellar irradiation flux irr 2 irr L 4 r midplane optical depth opacity from Bell & Lin (1990); C -1 = tan /3

7 column density angular velocity Initial Conditions of Prestellar Core radius radius Overall qualitative character of disk evolution is independent of the initial profiles of these quantities. 2, 1 r asymptotically, with flat central region a s vr 0, T const. 10 K c G 0

8 Disk Evolutionary Images Clump formation (gravitational instability) and migration Long-term evolution modeling is essential to determine the fate of the clumps. Based on Vorobyov & Basu (2006, ApJ, 650, 956)

9 Radial distance (AU) Radial distance (AU) Key Results for Early Accretion Phase Mass accretion rate (M yr -1 ) 1e-3 1e-4 1e-5 1e-6 1e-7 1e-8 1e-9 smooth mode burst mode 1e Time (Myr) residual disk accretion FU Ori eruptions flickering Bursts of accretion occur during the early accretion phase, as clumps are formed and driven inward. This is followed by a more quiescent phase that is still characterized by flickering accretion Vorobyov & Basu (2006, ApJ, 650, 956) Just before a burst Radial distance (AU) Nonlinear instability clumps efficient angular momentum transport Quiescent period Radial distance (AU)

10 3D non-ideal MHD simulations explain outflows + confirm aspects of disk fragmentation/episodic accretion 3D nested grid simulations Machida, Inutsuka, and Matsumoto (2011) Inutsuka, Machida, and Matsumoto (2010)

11 Migrating Embryo (ME) Scenario Clumps form in outer disk, radius ~ AU, migrate inward (migration time ~ yr). Many possibilities below Clump is accreted to center, mostly intact mass accretion and luminosity burst, Vorobyov & Basu (2006, 2010) Clump opens up a gap in disk and settles into a stable orbit form companion star, BD, giant planet VB (2010, ApJ, 714, L133) Multiple clumps can lead to ejection of a clump free floating BDs, very low mass stars Basu & Vorobyov (2012, ApJ, 750, 30) Clump is tidally dispersed in inner disk, but not before achieving T ~ 1000 K in core release thermally processed solids to inner disk Boley et al. (2010), Vorobyov (2011, ApJ, 728, L45) Clump migration and partial dispersal inner gas giant Clump gas dispersed but sedimentation occurs beforehand solid core remains terrestrial planet formation (Nayakshin 2010)

12 Radial distance (AU) Multiple Fragments in Massive Disk Ejection of Low Mass Fragment No sink cells employed to follow clumps. M 0.95M sun Ejection correlated with higher mass and angular momentum in initial state. Basu & Vorobyov (2012, ApJ, 750, 30)

13 Ejections occur in many models Lowest mass clumps more likely to be sheared by tidal effects arising from ejection. Ejected clumps span the substellar to low mass star regime, and have ejection speeds 0.8 +/ km/s. Basu & Vorobyov (2012, ApJ, 750, 30) Some models exhibit multiple ejections. Occurrence and number of ejections correlate with amount of initial model angular momentum and mass.

14 Pre-BD ejection, not BD ejection Hybrid model of clump ejection rather than finished BD ejection, overcomes some principal objections to ejection model: Disks expected to accompany BDs in this scenario Ejection speeds ~ 1 km/s comparable to velocity dispersion of stars/bds in clusters (Joergens 2006) BDs expected to be co-located in YSO clusters Expected number? No firm prediction but ejection model may lead to greater numbers than a model of only direct collapse from CMF. Testable?

15 Object definition from formation mechanism?! Image: Basu (2012, Science, 337, 43)

16 Summary Disk evolution calculated from self-consistent collapse of dense core yields a Migrating Embryo model of episodic clump formation, inward migration, dissolution, or ejection Ejection leads naturally to ejected low mass clumps that straddle the substellar mass limit. Can expect the formation of isolated brown dwarfs and very low mass stars that have their own disks Ejection of proto-brown-dwarfs avoids the problem of fast ejection associated with traditional ejection models Core accretion, disk instability, and direct collapse planet, brown dwarf, star. However, not a one-to-one mapping. New classification can be based on the formation mechanisms, at least core accretion vs. gravitational instability