Gallery of ALMA continuum images of disks with inner cavities (credits Nienke van der Marel).
My research focuses on the density structure of gas and dust in circumstellar disks through ALMA observations. I work mostly on large samples and aim to connect my work with exoplanet properties and planet formation models. A key aspect of my work are dust traps: concentrations of millimeter-dust pebbles seen as rings or asymmetries in disks which prevent the dust from drifting inwards. Dust traps can be used to indicate the presence of giant planets in disks, but also regulate the dust disk evolution, enhance planetesimal formation and affect the icy chemistry in disks, and are thus connected with every aspect of planet formation. I analyze observational data through e.g. visibility modeling, radiative transfer and physical-chemical modeling and connect the results with theoretical work on dust evolution, planet-disk interaction and hydrodynamical models, and with exoplanet populations, exoplanets revealed through direct imaging and studies of exoplanet atmospheres.
Recent paper highlights (papers led by myself and my students)
Hypothesis of the C/O distribution in disks as the result of dust traps
We reanalyze the C2H detections and non-detections in 26 disks from the literature. We find a likely correlation between the C2H peak brightness and the 'icy dust mass' (the dust mass outside the CO snowline). This is evidence that the C/O ratio (as traced by C2H) is regulated by dust transport and the location of the CO snowline. This is quite different from the classical view of the C/O ratio as a radial step function at snowline locations, and has important consequences for the interpretation of exoplanet atmospheres, as those planets create pressure bumps outside their orbit. If this complex interplay between pressure bumps and snowline locations is indeed the main explanation for the observed bimodal distribution in the C2H, this might also help to understand and quantify CO depletion in disks, used to measure gas masses!
The cartoon shows the 4 possible scenarios of dust ring vs CO snowline and the resulting C/O. More data is obviously needed, as C2H studies are often biased towards the brightest mm-disks, which are mostly ring disks! But also chemical models really should start to take dust transport into account to quantify these effects.
A YouTube presentation is following soon...
Michel et al. 2021, ApJ, in press - "Bridging the gap between protoplanetary and debris disks: separate evolution of millimeter and micrometer-sized dust"
Disk processes connecting Class II, III and debris disks
Based on the analysis of Class II and III disks in nearby star-forming regions, we propose that structured Class II disks (ring and transition disks) are the most likely progenitors to the known debris disks. In structured disks, dust traps allow for the formation of planetesimal belts at large radii, such as those later observed in debris disks. However, in disks without dust traps, radial drift rapidly removes most of the mm dust.
During the disk evolution, we suggest a separation in the mm and micron dust evolution taking place by different mechanisms. This provides evidence for slow disk dissipation in line with low alpha values and dust evolution models.
In addition, using Gaia-based membership of 11 low-UV nearby regions, we find evidence for long infrared protoplanetary disk timescales on the order of ~8 Myr. This means that the closest and well-observed/studied disks survive 2-3 times longer than previously thought in the absence of strong external photoevaporation: this is an important update of the Mamajek 2009 disk lifetime prediction of 2-3 Myr.
A 10 minute YouTube presentation by Arnaud Michel given at EAS on this work can be found here.
ALMA detections of 13CO, methanol, formaldehyde, sulphur monoxide and sulphur dioxide in IRS48
We identified 7 H2CO, 6 CH3OH, 2 SO2 and 34SO2 lines in an ALMA dataset of the IRS48 dust trap, and all line emission is cospatial with the asymmetric dust trap. SO2 has never been detected in a disk before, and this is only the second detection of CH3OH in a warm Herbig disk, following HD100546.
H2CO and CH3OH are products of hydrogenation of CO-ice on dust grains, and these icy grains may be inherited from the colder dark cloud phase and transported to the dust trap. Analysis of the rotational diagrams and RADEX indicate that the molecules are warm (>100 K!), indicating efficient vertical mixing.
Second result is the SO2 detection, and we recovered a detection of SO in a previous dataset, plus a non-detection of CS. We derive CS/SO < 0.01 so C/O < 1 at the dust trap location! In most disks observed so far C/O is > 1, so pebble drift and trapping may play a crucial role in regulating the C/O of the gas at a pressure bump location, which might even accrete onto the planet inside the gap.
IRS 48 is the first protoplanetary disk with a clear link between the morphology of COM and sulphur line emission and dust. Dust traps must be taken into account in chemical disk modeling, and we should spatially resolve molecules in disks for comparison with the dust substructure.
A 20 minute YouTube presentation at Astrochemistry Discussions on this work can be found here.
Proposed connection between disk demographics and exoplanet populations
In a large disk survey, combining the data from more than 700 continuum disks imaged with ALMA, we identified two important trends: first, the gapped disks (transition and ring disks) maintain their dust mass with age, whereas the bulk of the compact disks decrease in dust mass, which is evidence for two evolutionary pathways, for disks with pressure bumps and disks dominated by radial drift.
Second, the gapped disks show a clear stellar mass dependence, which is also seen in giant exoplanet occurrence rates: we can match the two when transition disks are caused by ~>1 MJup planets and ring disks by 0.2-1 MJup planets. These exoplanets are typically located at 2-3 au, so they must migrate inwards from the gap location. On the other hand, the compact disks matches the anti-correlation of the occurrence of close-in super-Earths, suggesting that the high pebble flux in drift-dominated disks creates the perfect conditions for the formation of these planets.
A 20 minute YouTube presentation on this work can be found here. This work was accompanied by a NRAO press release.
Norfolk et al. 2021, MNRAS - "Dust Traps and the Formation of Cavities in Transition Discs: A millimetre to sub-millimetre comparison survey"
Comparison between ATCA (cm) and ALMA (mm) radial profiles of a selection of the transition disks in the sample
We present the results from the Discs Down Under program, an 8.8-mm continuum Australia Telescope Compact Array (ATCA) survey targeting 15 transition discs with large (≳20 au) cavities and compare the resulting dust emission to Atacama Large millimetre/sub-millimetre Array (ALMA) observations. The ATCA observations resolve the inner cavity for 8 of the 14 detected discs. We fit the visibilities and reconstruct 1D radial brightness models for 10 sources with a S/N > 5sigma. We find that, for sources with a resolved cavity in both wavebands, the 8.8 mm and sub-mm brightness distributions peak at the same radius from the star. We suggest that a similar cavity size for 8.8 mm and sub-mm dust grains is due to a dust trap induced by the presence of a companion.
A 12 minute YouTube presentation on this work by Brodie Norfolk at the "Five years after HL Tau" conference can be found here.
Sample of asymmetric and axisymmetric disks in this study.
Protoplanetary disks with large inner dust cavities are thought to host massive planetary or substellar companions. These disks show asymmetries and rings in the millimeter continuum caused by dust trapping in pressure bumps and potentially vortices or horseshoes. The origin of the asymmetries and their diversity remains unclear. We present a comprehensive study of 16 disks for which the gas surface density profile has been constrained by CO isotopologue data. First, we compare the azimuthal extents of the dust continuum profiles with the local gas surface density in each disk and find that the asymmetries correspond to higher Stokes numbers or low gas surface density. We discuss which asymmetric structures can be explained by a horseshoe, a vortex, or spiral density waves. Second, we reassess the gas gap radii from the 13CO maps, which are about a factor of 2 smaller than the dust ring radii, suggesting that the companions in these disks are in the brown dwarf (15-50 MJup) or super-Jovian (3-15 MJup) mass regime on eccentric orbits. This is consistent with the estimates from contrast curves on companion mass limits. These curves rule out (sub)stellar companions (q > 0.05) for the majority of the sample at the gap location, but it remains possible at even smaller radii. Third, we find that spiral arms in scattered-light images are primarily detected around high-luminosity stars with disks with wide gaps, which can be understood by the dependence of the spiral arm pitch angle on disk temperature and companion mass.
A 1 hour YouTube presentation on this work at the Arizona Origins seminar series can be found here.
Francis & van der Marel 2020, ApJ - "Dust-depleted Inner Disks in a Large Sample of Transition Disks through Long-baseline ALMA Observations"
Sample of transition disks with inner dust disks
Transition disks with large inner dust cavities are thought to host massive companions. However, the disk structure inside the companion orbit and how material flows toward an actively accreting star remain unclear. We present a high-resolution continuum study of inner disks in the cavities of 38 transition disks. Measurements of the dust mass from archival ALMA observations are combined with stellar properties and spectral energy distributions to assemble a detailed picture of the inner disk. An inner dust disk is detected in 18 of 38 disks in our sample. Of the 14 resolved disks, 8 are significantly misaligned with the outer disk. The near- infrared excess is uncorrelated with the mm-dust mass of the inner disk. The size-luminosity correlation known for protoplanetary disks is recovered for the inner disks as well, consistent with radial drift. The inner disks are depleted in dust relative to the outer disk, and their dust mass is uncorrelated with the accretion rates. This is interpreted as the result of radial drift and trapping by planets in a low alpha disk, or a failure of the alpha-disk model to describe angular momentum transport and accretion. The only disk in our sample with confirmed planets in the gap, PDS 70, has an inner disk with a significantly larger radius and lower inferred gas-to-dust ratio than other disks in the sample. We hypothesize that these inner disk properties and the detection of planets are due to the gap having only
been opened recently by young, actively accreting planets.
Bi et al. 2020, ApJL - "GW Ori: Interactions between a Triple-star System and Its Circumtriple Disk in Action"
The triple ring disk system around the triple star GW Ori
GW Ori is a hierarchical triple system with a rare circumtriple disk. We present Atacama Large Millimeter/submillimeter Array (ALMA) observations of 1.3 mm dust continuum and 12CO J = 2 - 1 molecular gas emission of the disk. For the first time, we identify three dust rings in the GW Ori disk at ∼46, 188, and 338 au, with estimated dust mass of 74, 168, and 245 Earth masses, respectively. To our knowledge, its outermost ring is the largest dust ring ever found in protoplanetary disks. We use visibility modeling of dust continuum to show that the disk has misaligned parts, and the innermost dust ring is eccentric. The disk misalignment is also suggested by the CO kinematics. We interpret these substructures as evidence of ongoing dynamical interactions between the triple stars and the circumtriple disk.
This work was accompanied by an NRAO press release and highlighted by Nature Astronomy (Paul Woods).
Interested?
For any inquiries on my on-going work, student projects and collaborations, please contact me.
