Scanning disk rings and winds in CO at 0.01-10 au: a high-resolution $M$-band spectroscopy survey with IRTF-iSHELL

Abstract

We present an overview and first results from a M-band spectroscopic survey of planet-forming disks performed with iSHELL on IRTF, using two slits that provide resolving power R ≈ 60,000-92,000 (5-3.3 km/s). iSHELL provides a nearly complete coverage at 4.52-5.24 µm in one shot, covering > 50 lines from the R and P branches of 12 CO and 13 CO for each of multiple vibrational levels, and providing unprecedented information on the excitation of multiple emission and absorption components. Some of the most notable new findings of this survey are: 1) the detection of two CO Keplerian rings at < 2 au (in HD 259431), 2) the detection of H 2 O ro-vibrational lines at 5 µm (in AS 205 N), and 3) the common kinematic variability of CO lines over timescales of 1-14 years. By homogeneously analyzing this survey together with a previous VLT-CRIRES survey of cooler stars, we discuss a unified view of CO spectra where emission and absorption components scan the disk surface across radii from a dust-free region within dust sublimation out to ≈ 10 au. We classify two fundamental types of CO line shapes interpreted as emission from Keplerian rings (double-peak lines) and a disk surface plus a low-velocity part of a wind (triangular lines), where CO excitation reflects different emitting regions (and their gas-to-dust ratio) rather than just the irradiation spectrum. A disk+wind interpretation for the triangular lines naturally explains several properties observed in CO spectra, including the line blue-shifts, line shapes that turn into narrow absorption at high inclinations, and the frequency of disk winds as a function of stellar type.

The Role of Disk Winds in the Evolution and Dispersal of Protoplanetary Disks

“…Out of their 36 T Tauri stars in Taurus they detect the [O I] λ6300 line in all but one source and the H 2 ν =1-0 S(1) at 2.12 µm in half of their sample. H 2 profiles are found to be less complex than the [O I] λ6300 profiles, and mostly single-peaked with only modest blueshifts (< 20 km/s), see also Bary et al (2003). Although the H 2 and [O I] LVC-NC luminosities are not correlated, for sources where both lines are detected, their peak velocities are quite similar and their FWHMs correlate, pointing to a common origin.…”

“…This possibility has been invoked in the literature as the source of some slow ≤ 500 au scale resolved H 2 and CO flows observed in Class 0/I/II sources (e.g., Pyo et al 2003;Takami et al 2006;Beck et al 2008;Zapata et al 2015;Lee et al 2018a;Bjerkeli et al 2019). However, on such small scales the ambient environment will no longer be a static self-gravitating 1/r 2 core, as for large scale entrained flows (Shang et al 2007, 2020, andreferences therein), but rather an infalling rotating envelope, where centrifugal de-flection and gravity of the central star become dominant.…”

“…Another example is HL Tau with a slow CO flow seen with ALMA (Garufi et al 2020, Bacciotti, in prep. ) surrounding both the high velocity jet seen in forbidden lines and a wide angle H 2 cone (Takami et al 2007;Beck et al 2008). The more massive (∼ 1.9 M ) star HD 163296 also has an optical jet surrounded by both a very slow molecular outflow (∼30 m/s) thought to originate beyond ∼300 au (Teague et al 2019) and a much larger and faster (20 km/s) molecular flow extending vertically out to ∼1000 au (Klaassen et al 2013;Booth et al 2021).…”

“…At high inclinations CO profiles not only broaden but, unlike forbidden lines, begin to show central absorptions, also with small blueshifts, attributed to the line of sight grazing a low velocity wind over the disk surface. In all, the majority of T Tauri CO profiles are interpreted as arising in a low velocity disk wind, although blueshifts above 5 km/s are limited to a handful of sources (Bast et al 2011;Pontoppidan et al 2011;Brown et al 2013;Banzatti et al 2022) The nested velocity structure that is well documented in resolved outflows appears to persist when the spatial scale of the outflows has diminished and can only be traced spectroscopically. This conclusion is based on attributing forbidden line HVC to a fast axial jet and the forbidden line widths of the (faster) broad and (slower) narrow components of the LVC to the radii at which the disk winds arise.…”

An Unusual Reservoir of Water Emission in the VV CrA A Protoplanetary Disk

“…Bast et al 2011). Lorentzian-like M-band CO rovibrational emission has previously been attributed to the presence of low-velocity disk winds (Pontoppidan et al 2011;Banzatti et al 2022). A contribution from a CPD would also cause deviations from the simple disk velocity trend, but we cannot say definitively how the L-band emission line profile compares to other wavelengths.…”