We present a spectro-astrometric survey of molecular gas in the inner regions of 16 protoplanetary disks using CRIRES, the high resolution infrared imaging spectrometer on the Very Large Telescope. Spectro-astrometry with CRIRES measures the spatial
extent of line emission to sub-milliarcsecond precision, or <0.2 AU at the distance of the observed targets. The sample consists of gas-rich disks surrounding stars with spectral types ranging from K to A. The properties of the spectro-astrometric signals divide the sources into two distinct phenomenological classes: one that shows clear Keplerian astrometric spectra, and one in which the astrometric signatures are dominated by gas with strong non-Keplerian (radial) motions. Similarly to the near-infrared continuum emission, as determined by interferometry, we find that the size of the CO line emitting region in the Keplerian sources obeys a size-luminosity relation as $R_CO L_*^0.5. The non-Keplerian spectro-astrometric signatures are likely indicative of the presence of wide-angle disk winds. The central feature of the winds is a strong sub-Keplerian velocity field due to conservation of angular momentum as the wind pressure drives the gas outwards. We construct a parametrized 2-dimensional disk+wind model that reproduces the observed characteristics the observed CO spectra and astrometry. The modeled winds indicate mass-loss rates of >10^-10 to 10^-8 Msol/yr. We suggest a unifying model in which all disks have slow molecular winds, but where the magnitude of the mass-loss rate determines the degree to which the mid-infrared molecular lines are dominated by the wind relative to the Keplerian disk surface.
We present ground-based high resolution N-band spectra (Delta v = 15 km/s) of pure rotational lines of water vapor in two protoplanetary disks surrounding the pre-main sequence stars AS 205N and RNO 90, selected based on detections of rotational wate
r lines by the Spitzer IRS. Using VISIR on the Very Large Telescope, we spectrally resolve individual lines and show that they have widths of 30-60 km/s, consistent with an origin in Keplerian disks at radii of ~1 AU. The water lines have similar widths to those of the CO at 4.67 micron, indicating that the mid-infrared water lines trace similar radii. The rotational temperatures of the water are 540 and 600K in the two disks, respectively. However, the lines ratios show evidence of non-LTE excitation, with low-excitation line fluxes being over-predicted by 2-dimensional disk LTE models. Due to the limited number of observed lines and the non-LTE line ratios, an accurate measure of the water ortho/para ratio is not available, but a best estimate for AS 205N is ortho/para = 4.5 +/- 1.0, apparently ruling out a low-temperature origin of the water. The spectra demonstrate that high resolution spectroscopy of rotational water lines is feasible from the ground, and further that ground-based high resolution spectroscopy is likely to significantly improve our understanding of the inner disk chemistry recently revealed by recent Spitzer observations.
We present a Spitzer InfraRed Spectrometer search for 10-36 micron molecular emission from a large sample of protoplanetary disks, including lines from H2O, OH, C2H2, HCN and CO2. This paper describes the sample and data processing and derives the de
tection rate of mid-infrared molecular emission as a function of stellar mass. The sample covers a range of spectral type from early M to A, and is supplemented by archival spectra of disks around A and B stars. It is drawn from a variety of nearby star forming regions, including Ophiuchus, Lupus and Chamaeleon. In total, we identify 22 T Tauri stars with strong mid-infrared H2O emission. Integrated water line luminosities, where water vapor is detected, range from 5x10^-4 to 9x10^-3 Lsun, likely making water the dominant line coolant of inner disk surfaces in classical T Tauri stars. None of the 5 transitional disks in the sample show detectable gaseous molecular emission with Spitzer upper limits at the 1% level in terms of line-to-continuum ratios (apart from H2). We find a strong dependence on detection rate with spectral type; no disks around our sample of 25 A and B stars were found to exhibit water emission, down to 1-2% line-to-continuum ratios, in the mid-infrared, while almost 2/3 of the disks around K stars show sufficiently intense water emission to be detected by Spitzer. Some Herbig Ae/Be stars show tentative H2O/OH emission features beyond 20 micron at the 1-2 level, however, and one of them shows CO2 in emission. We argue that the observed differences between T Tauri disks and Herbig Ae/Be disks is due to a difference in excitation and/or chemistry depending on spectral type and suggest that photochemistry may be playing an important role in the observable characteristics of mid-infrared molecular line emission from protoplanetary disks.
The material that formed the present-day Solar System originated in feeding zones in the inner Solar Nebula located at distances within ~20 AU from the Sun, known as the planet-forming zone. Meteoritic and cometary material contain abundant evidence
for the presence of a rich and active chemistry in the planet-forming zone during the gas-rich phase of Solar System formation. It is a natural conjecture that analogs can be found amoung the zoo of protoplanetary disks around nearby young stars. The study of the chemistry and dynamics of planet formation requires: 1) tracers of dense gas at 100-1000 K and 2) imaging capabilities of such tracers with 5-100 (0.5-20 AU) milli-arcsec resolution, corresponding to the planet-forming zone at the distance of the closest star-forming regions. Recognizing that the rich infrared (2-200 micron) molecular spectrum recently discovered to be common in protoplanetary disks represents such a tracer, we present a new general raytracing code, RADLite, that is optimized for producing infrared line spectra and images from axisymmetric structures. RADLite can consistently deal with a wide range of velocity gradients, such as those typical for the inner regions of protoplanetary disks. The code is intended as a backend for chemical and excitation codes, and can rapidly produce spectra of thousands of lines for grids of models for comparison with observations. Such radiative transfer tools will be crucial for constraining both the structure and chemistry of planet-forming regions, including data from current infrared imaging spectrometers and extending to the Atacama Large Millimeter Array and the next generation of Extremely Large Telescopes, the James Webb Space Telescope and beyond.
