Interferometric Spectro-imaging of Molecular Gas inProto-Planetary DisksAnne Dutrey and St´ephane GuilloteauL3AB, Observatoire de BordeauxPaul HoAcademia Sinica Institute of Astronomy and Astrophysics and Smithsonian Astrophysical ObservatoryProto-planetary disks are found to orbit around low and intermediate mass stars. Currenttheories predict that these disks are the likely sites for planet formation. In this review, wesummarize the improvement in our knowledge of their observed molecular properties sincePPIV. This is timely since a new facility, the Submillimeter-Array (SMA), has recently begunoperation and has opened the submillimeter atmospheric windows to interferometry, allowingstudies of warmer gas and dust in disks at subarcsecond resolution. Using results from theIRAM array and the SMA, we focus on two complementary main topics: 1) the determinationof the physical structure of the disks from multi-transition CO isotopic analysis of high angularresolution millimeter interferometric data and 2) the observations of molecules other thanCO (and isotopes), which enable investigations of the chemistry in proto-planetary disks. Inparticular, we emphasize how to handle the available data to provide relevant constraints on thethermal, physical and chemical structure of the disks as a function of radius, within the currentlimitations in sensitivity and angular resolution of the existing arrays. These results suggestthe importance of photo-dissociation effects and X-ray heating. They also reveal unexpectedresults, such as the discovery of non-Keplerian rotation in the AB Aur disk. We also discusshow to extrapolate these results in the context of the tremendous capabilities of the ALMAproject currently under construction in Chile.1. INTRODUCTIONIn the last fifteen years, observations from the opticalup to the millimeter wavelength domains have revealed thatlow and intermediate mass stars of ages around a millionyears, such as TTauri and Herbig Ae stars, are surroundedby circumstellar disks. These disks are often called ”proto-planetary” because they still contain enough original mate-rial (from the parent cloud) to form giant planets. Indeed,although dust is the easiest tracer of proto-planetary disks,molecules represent more than 70 % of the total mass indisks. H2, which does not deplete on dust grains, is byfar the most abundant molecule in disks but it remains dif-ficult to observe because of the lack of a dipole moment.Even if a few direct H2detections are now possible, the ob-served H2lines mainly trace the warm gas located in theinner disks (R< 10-20 AU), while, in many cases, disksare known to extend out to several 100 AU. These outer re-gions, which may even contain most of the mass, are coldand can only be characterized by molecules having rota-tional lines at low energy levels and detectable with largemillimeter and sub-millimeter interferometers. After H2,CO is the most abundant molecule (even if it can depleteon dust grains) and its lowest rotational lines are easily ex-cited by collisions with H2in disks. Only heterodyne ar-rays give the sensitivity, resolving power and spectral res-olution needed to map cold molecular disks. In the lastten years, millimeter spectroscopic studies of disks haveshown that they are in Keplerian rotation. More recently,the SMA (Ho et al., 2004) has opened a new the era of sub-millimeter interferometry, while the IRAM Plateau de Bureinterferomter (PdBI) routinely provides images of 0.6” res-olution at 1.3mm. Since PPIV, observations of moleculardisks have considerably improved, and mm/submm arrayshave provided many new direct constraints on physical andchemical structure of disks which could not be addressedby disk continuum observations.In this review we focus on the recent molecular resultsobtained with the SMA and the IRAM array. We summarizein Section 2 the sensitivity which can be achieved by ob-serving CO transitions in disks with the PdBI and the SMA.We also calculate the brightness temperature for transitionsJ=1-0 up to J=3-2 and describe new methods of analysis forinterferometric data. In Section 3, we present the CO diskproperties as inferred from SMA and IRAM array observa-tions. Then Section 4 is dedicated to new results for somespecific disks because they provide new quantitative infor-mation on the disk physics since PPIV. Particular attentionis given to the observation of molecular chemistry in Sec-tion 5. Then we conclude by presenting the sensitivity ofALMA. This paper focusses on line data and outer disksonly. In these proceedings, more information on continuumemission can be found in the chapters by Testi et al., and onchemistry by Bergin et al.12. MOLECULAR LINE FORMATION AND DISKMODELSThe line formation process in proto-planetary disks iscomplicated because of the combination of strong velocity,temperature and density gradients. The regular pattern ofrotational motions provides a direct link between the pro-jected velocity and position in the disk, a property whichprovides some effective super-resolution, under the guid-ance of an applicable model. In addition, proto-planetarydisks have two specific properties which make the deriva-tion of physical parameters from observations particularlyrobust and relatively simple:1. Power laws are good approximations for the radialdependence of many physical quantities2. As a result, molecular column densities can be de-rived from the observation of a single (partially opti-cally thin) transition.However, radiative transfer models are required to estimatethe line brightness distribution as a function of projectedvelocity. This implies the use of dedicated codes, whoseprecision should be matched to the sensitivity of the obser-vations to be analyzed. Such detailed models have becomenecessary only because the observations now provide suffi-cient sensitivity. In this section, we describe the availabletools and their use.2.1. Disk DescriptionDisks models (such as those described in Dutrey etal., 1994) are usually based on the description of Pringle(1981): a geometrically thin disk in hydrostatic equilibriumwith sharp inner and outer edges. Temperature, velocity andsurface density assume power law radial dependencies:For the kinetic temperature: T (r) = To(r/ro)−qthe surface density: Σ(r) = Σo(r/ro)−pand the velocity: V (r) = Vo(r/ro)−v, with v = 0.5 forKeplerian disks.We start by the simplest approach which assumes powerlaw dependencies versus radius. This is of course
View Full Document
Unlocking...