A dusty origin for the correlation between protoplanetary disc accretion rates and dust masses

Abstract

Recent observations have uncovered a correlation between the accretion rates (measured from the UV continuum excess) of protoplanetary discs and their masses inferred from observations of the sub-mm continuum. While viscous evolution models predict such a correlation, the predicted values are in tension with data obtained from the Lupus and Upper Scorpius star forming regions; for example, they underpredict the scatter in accretion rates, particularly in older regions. Here we argue that since the sub-mm observations trace the discs dust, by explicitly modelling the dust grain growth, evolution, and emission, we can better understand the correlation. We show that for turbulent viscosities with α ≲ 10−3, the depletion of dust from the disc due to radial drift means we can reproduce the range of masses and accretion rates seen in the Lupus and Upper Sco datasets. One consequence of this model is that the upper locus of accretion rates at a given dust mass does not evolve with the age of the region. Moreover, we find that internal photoevaporation is necessary to produce the lowest accretion rates observed. In order to replicate the correct dust masses at the time of disc dispersal, we favour relatively low photoevaporation rates ≲ 10−9 M⊙ yr−1 for most sources but cannot discriminate between EUV or X-ray driven winds. A limited number of sources, particularly in Lupus, are shown to have higher masses than predicted by our models which may be evidence for variations in the properties of the dust or dust trapping induced in substructures.

PENELLOPE: The ESO data legacy program to complement the Hubble UV Legacy Library of Young Stars (ULLYSES)

“…Recently, the connection of measurements ofṀ acc from spectroscopy and of disk masses with millimeter interferometry, mainly with the Atacama Large Millimeter/submillimeter Array (ALMA), has also revealed the presence of a correlation between these two quantities (e.g., Manara et al 2016b;Mulders et al 2017), which is possibly in line with expectations drawn from viscous evolution (e.g., Jones et al 2012;Lodato et al 2017;Rosotti et al 2017), but this also reveals tensions with respect to this simplified description of disk evolution (e.g., Mulders et al 2017;Manara et al 2020). These tensions may be relieved by introducing complications to a simplistic model, including external or internal photoevaporation (e.g., Rosotti et al 2017;Sellek et al 2020a;Somigliana et al 2020), or dust evolution (Sellek et al 2020b).…”

On the secular evolution of the ratio between gas and dust radii in protoplanetary discs

“…Because only the grains larger than the wavelength have significant opacity to be observed, the physics of grain growth and drift is fundamental in setting the observed disc size. Moreover, Sellek et al (2020) show that smooth models including radial drift can also explain the distribution of observations in the accretion rate -disc mass plane, even in the old Upper Sco region.…”

“…Indeed, in this case, the disc radius after 3 Myr is very similar to the initial value, implying a strong dependence on the initial conditions of protoplanetary discs and no disc evolution: initially large discs remains large while initially compact discs are observed with small dust radii. Moreover, Sellek et al 2020 found that the resulting dust mass of the discs are proportional to 𝑓 grow , so that a large reduction in growth efficiency also corresponds to a correspondingly large increase in dust mass: that would help to explain the observed large masses of a few discs in Lupus, but the bulk of the population would be too massive compared to observations.…”

“…where Ω is the Keplerian angular frequency and the growing factor 𝑓 grow is an extra factor (considered to be equal to 1 in the "standard" two population model and in this work) used e.g in Booth & Owen 2020 to measure the fraction of collisions that lead to growth: 𝑓 grow = 1 means that all the collisions result in dust growth while 𝑓 grow = 100 means that only 1% of the collisions result in growth, leading to longer lifetimes for dust in discs due to a less efficient radial drift (for an extended discussion see Booth & Owen 2020or Sellek et al 2020).…”

“…While internal photo-evaporation mostly affects the inner part of the disc, external photo-evaporation preferentially removes mass from the outer part of discs, potentially impacting disc radii. Indeed, Sellek et al 2020 found that external photo-evaporation can reduce the amount of dust in the outer part of the disc, increasing the efficiency of radial drift, thus reducing 𝑅 dust . At the same time, also 𝑅 CO is expected to decrease if the accretion rate is smaller than the photo-evaporative mass loss rate.…”