Most of the past and ongoing exoplanet surveys have focused on stars as massive as or less massive than the Sun, and about 90% of the ∼4500 known exoplanets lie closer to their stars than the Earth is to the Sun. What lurks in the outskirts of high-mass stars systems, waiting to be discovered? This is the scientific question that the B-star Exoplanet Abundance Study (BEAST) is trying to answer.
With just one third of the targets observed twice, BEAST has already started to give exciting results. We have recently presented the discovery of a giant planet with 10.9±1.6 M_J around the binary b Centauri (Janson et al., Nature 600, 231), and one (possibly two) companion(s) around the massive star μ2 Scorpii, the first planet-hosting star that will end its life as a supernova (Squicciarini et al. 2022). How can the existence of these planets be reconciled with the current models of planetary formation?
What is the frequency of giant planets and brown dwarfs around B stars? What is the formation model that best explains the observations? Every new discovery is a piece of a huge jigsaw puzzle, whose complete picture we are trying to grasp. But to have each piece in the right spot, precise estimate of both stellar properties (mass, age) and companion properties (mass, orbital parameters) are required. To this end, I have combined kinematic information from Gaia and a new age determination tool, MADYS.
Precise system ages are mandatory both for a meaningful comparison of direct imaging observations with model predictions and for demographic purposes. However, due to a combination of factors (distance uncertainty, rapid rotation, magnetic fields, unresolved multiplicity, luminosity almost independent of age after 2-3 Myr), direct age determinations for B stars are extremely difficult. Tackling this problem has been a major part of my involvement within BEAST.
Both random and systematic uncertainties prevent a precise age estimate for main-sequence B stars.
Using Gaia parallaxes and proper motions, I identified comoving stars (CMS), i.e. stars with similar 2D-projected space motions, to most BEAST targets. The identified groups can be seen as co-moving streams sprouting from a common extended star formation event, giving the opportunity to indirectly constrain the age of the target star. Individual isochronal ages for each CMS are computed using MADYS.
The renewed interest in young clusters and associations thanks to Gaia, on the one hand, and the flourishing of planet-hunting direct imaging surveys, on the other hand, highlight the need for a uniform and consistent derivation of stellar parameters (age, mass) for large groups of stars. To this end, I have developed MADYS, a flexible Python tool for age and mass determination of young stellar and substellar objects based on the comparison between photometric data and theoretical isochrone grids.
A mock CMD from MADYS.
Given a list of stars, MADYS automatically retrieves and cross-matches photometry from different catalogues, corrects for interstellar extinction, and employs reliable photometric data to rapidly derive ages and masses of individual stars. Harmonizing the heterogeneity of publicly-available isochrone grids, MADYS allows to select one among 17 models, many of which offering the possibility to customize metallicity, rotational velocity and other astrophysical parameters, for a total of ~120 different isochrone grids.
The versatility of MADYS is one among its chief strengths: from indirect age determinations (Janson+21) to direct mass estimates (Bonavita+21), from the computation of detection map for direct imaging surveys to a thorough age comparison of different populations (Squicciarini+21), and even to the selection of the best targets for future surveys (Ray et al. in prep), MADYS has the potential to become highly beneficial to the whole community.
Members of stellar associations can be distinguished from field interlopers because of a common average proper motion. This common motion, gradually eroded by the galactic tidal field over time, is in turn reminiscent of the initial kinematic structure. Combining astrometry and kinematics from Gaia EDR3, and radial velocities from GALAH and APOGEE, I traced back the present positions of stars belonging to Upper Scorpius (USCO), a subgroup of Scorpius–Centaurus, the nearest OB association to the Sun.
One half of USCO (the clustered population) appears composed of many smaller entities, which were in a more compact configuration in the past. Thanks to MADYS, I confirmed an age spread between this younger clustered population and an older diffuse population. Star formation in USCO appears to have been highly substructured and long-lasting (>10 Myr); the initial velocity structure is being slowly eroded, as testified by the presence of the diffuse component.
