Title: Exploring brown dwarf disks: A 1.3 mm survey in Taurus

Abstract: We have carried out sensitive 1.3 mm observations of 20 young brown dwarfs in the Taurus star-forming region, representing the largest sample of young substellar objects targeted in a deep millimeter continuum survey to date. Under standard assumptions, the masses of brown dwarf disks range from ~<0.4 to several Jupiter masses. Their relative disk masses are comparable to those derived for coeval low-mass stars: most of them are in the ~<1-5% range, and there is no clear change of relative disk mass with object mass from 0.015 to 3 solar masses. Specifically, we do not find evidence for disk truncation, as would be expected in the ejection scenario for brown dwarf origin, although the signature of ejection may be hidden in our non-detections. We use the derived mm fluxes, complemented by mid-infrared data from the Spitzer Space Telescope and ground-based near-infrared images, to construct spectral energy distributions (SEDs) for six of our sources, and model those SEDs with a Monte Carlo radiative transfer code. While the model fits are by no means unique, they allow us to investigate disk properties such as the degree of flaring and minimum radii. In several cases, we find that the SEDs in the mid-infrared exhibit lower flux levels than predicted by hydrostatic models, implying dust settling to the disk midplane. What's more, at least 25% of our targets are likely to have disks with radii >10 AU; models with smaller disks cannot reproduce the mm fluxes even if they are very massive. This finding is in contrast to the results of some simulations of the ejection scenario for brown dwarf formation that suggest only ~5% of ejected objects would harbor disks larger than 10 AU. Our findings imply that ejection is probably not the dominant formation process, but may still be relevant for some brown dwarfs.