What are the initial conditions in young disks, and what is the likelihood that they are, in fact, proto-planetary? The raw material of planetary embryos, Earth-like rocks, and Jupiter-like gas giants is indeed abundant, if not ubiquitous, in young disks. But whether any individual disk will form planets is, of course, unknowable. What we can say is that many of the disks we observe are at least capable of forming planetary systems similar to our own, as evidenced from measured disk sizes, masses, and composition/chemistry. However, as detailed below, the mean disk properties are not yet known due to sensitivity limitations and therefore comparisons to our own proto-solar system based on existing data may be biased.
Disks around young stars were spatially resolved for the first time at millimeter wavelengths (e.g., Sargent & Beckwith 1987) which measure cold dust and gas in the outer disk regions. Unequal axial ratios, combined with implied dust masses large enough that the central stars should not be optically visible if the dust geometry is spherically symmetric, stood as the strongest evidence for close to a decade of disks surrounding young stars. Further, kinematic models of spatially resolved CO emission demonstrated consistency with Keplerian rotation (e.g., Koerner et al. 1993; Mannings et al. 1997; Simon et al. 2000; Qi et al. 2003).
Continued interferometric work (e.g., Lay et al. 1994; Dutrey et al. 1996; Duvert et al. 2000; Kitamura et al. 2002; Qi et al. 2003, Semenov et al. 2005), suggested that disk diameters—in instances where spatially resolved, as opposed to point-like, images are obtained—range from —70-700 AU and are even as large as —2000 AU in some cases. These disk-size estimates are consistent with those inferred from optical/near-infrared scattered light or silhouette images (e.g., McCaughrean & O'Dell 1996; Padgett et al. 1999; Bally et al. 2000), and in the typical case are comparable to, or larger than, the orbit of the outermost gas giant in our Solar System, Neptune. Surface density profiles, e.g., simple power-laws with £(r) x r-p or viscous disk "similarity solutions" with
E(r) x r-pe-r(2-p' , have suggested a wide range in the value of p (0-1.5 for the power-law case).
Disk masses are derived from optically thin millimeter flux and an adopted opacity-wavelength relationship which leads to uncertainties of factors of 5-10 in disk masses. Under common assumptions, the calculated dust masses range from 10-45 to 10-3 Mq (e.g., Beckwith et al. 1990). Making the further assumption that the dust:gas ratio by mass is unaltered from the canonical interstellar value of 1:100, total disk masses average around 0.02 Mq, or about the Minimum Mass Solar Nebula (Kusaka et al. 1970; Weidenschilling 1977), the reconstitution of present-day solar system mass and composition to solar consistency. It should be stressed that detection at all of millimeter flux is made amidst an increasing number of upper limits measured for stars with other indicators of disks at shorter wavelengths, and so the true "mean mass" is even lower than that quoted above.
The composition of both young primordial and older debris disks has been shown to resemble that of solar system comets. Ground-based 10 and 20 ¡m work on samples of brighter sources (e.g., Hanner et al. 1995, 1998; Sitko et al. 1999; van Boekel et al. 2003; Kessler-Silacci et al. 2005) and especially ISO 2-30 ¡m spectroscopy (e.g., Meeus et al. 2001; Bouwmann et al. 2001) have revealed an impressive suite of solid state (and PAH) dust features. Mineralogical details of the dust are modeled on a case-by-case basis due to cosmic variance, but the mean composition appears to be ~70-80% amorphous magnesium-rich olivines, ~1-10% crystalline forsterite, ~10-15% carbons, ^3-5% irons, and other trace components such as silicas. In particular, crystallinity is advocated in ~10% of sources.
In summary, the observed sizes, masses, and chemical composition of young disks are all consistent with solar nebula estimates. This is a weak statement, however, since the mean disk properties are biased at present by detection limits and selection effects.
Was this article helpful?