SaNDi-SHoP: Searching for Satellites'N'Disks with a Star-Hopping Program I. Analysis of the close surroundings of DI companions

Abstract

We aim to search for satellites and circumplanetary or circumsubstellar disks around directly imaged substellar companions, exploring their immediate environment to constrain the conditions for satellites and disk formation. We conducted a dedicated survey of twelve planets and brown dwarfs with VLT/SPHERE using a novel application of the star hopping technique. By building libraries of contemporaneous point spread function (PSF) references from nearby stars, we applied a frame-by-frame subtraction of each companion's flux using the Negative Fake Companion method (NEGFC). This approach mitigates temporal PSF variability and enhances sensitivity to faint circumplanetary features. We derived contrast curves, translated them into mass detection limits using evolutionary models, and constrained the dynamically stable regions through estimates of Hill radii from orbital fits. Our analysis yields stringent limits on the presence of massive satellites, generally excluding companions more massive than a few Jupiter masses at separations beyond 1-5 au, depending on each system's Hill radius. In most cases, no convincing point-like or extended residuals were found. However, we identify promising signals for three systems: extended residuals consistent with a circumplanetary disk around CT Cha b, tentative repeated residuals near TYC 8047-232-1 B that may trace a bound satellite companion of 3-6 MJup, and marginal residual signals at the location of the previously reported candidate around DH Tau b, whose interpretation, however, remains uncertain due to possible contamination by instrumental effects. These results confirm the power of star hopping in reducing PSF-related artifacts and provide some of the most stringent constraints to date on the mass and location of potential satellites and disks around directly imaged substellar companions.

Figures (13)

• Figure 1: Final ADI-processed images of each system and the residual outcome of the PSF subtraction. For each target, the large panel shows the full field of view, while the inset presents a $30\times30$ pixel zoom on the subtracted companion and its residuals. Positive residuals are shown in red while negative ones are in blue. Detected background sources are marked in green. Pink shading denotes the range of possible Hill radii, with dashed lines for the extrema and a solid circle for half of the mean Hill radius.
• Figure 2: Left: 5$\sigma$ contrast curves in H2-band expressed in arcseconds versus contrast relative to the host star, truncated at half mean (solid curves) and half maximum (dashed curves) Hill radii for dynamical stability. Right: corresponding detection limits converted into mass using ATMO evolutionary models.
• Figure 3: H2-H3 color magnitude diagram for the objects in our sample, adapted from Bernardi et al submitted. The candidates around DH Tau b and TYC 8047-232-1 B are indicated with squares and triangles, respectively, both for datasets presented in this paper (green) and archival epochs (red).
• Figure 4: Residuals around TYC 8047-232-1 B after PSF subtraction for 2023-10-11 (top row) and 2024-08-27 (bottom raw) epochs and for H2 (left side) and H3 (right side) filters. Each set of three panels shows, from left to right, the substellar companion, the model PSF, and the residuals, highlighting a recurrent excess feature with a white circle.
• Figure 5: Archival SPHERE/IRDIS observations of TYC 8047-232-1 B from 2015-09-25 (top row) and 2016-01-17 (bottom row) in H2 filter. In each row, from left to right, the substellar companion, the model PSF, and the residuals are displayed. The recurring excess is highlighted with a white circle.