Circumplanetary Disk Candidate in the Disk of HD 163296 Traced by Localized Emission from Simple Organics

TL;DR

This work presents the first evidence of simple organic molecules tracing a candidate circumplanetary disc around a forming planet in HD 163296, detected via localized HCN and C2H emission co-located with a known CO kinematic perturbation (P94). By folding spectra from the disc halves and modeling the CPD with Keplerian, subpixel-accurate channel maps across multiple temperatures, the authors derive an upper bound on the planet mass, find a compact CPD size, and show that CPD temperatures must exceed ~$150\ \mathrm{K}$ to reproduce the observations. The results imply significant planetary heating and UV irradiation in the CPD, driving desorption and high-temperature chemistry that enhances organic abundances, and they support the use of chemical anomalies as a diagnostic for embedded planets when infrared signatures are obscured. Collectively, the findings advance our understanding of CPD chemistry, the thermal state of young planets, and satellite formation conditions in young planetary systems, while illustrating a powerful observational pathway for detecting and characterizing deeply embedded planets.

Abstract

Atacama Large Millimeter/submillimeter Array observations suggest that the disc of HD 163296 is being actively shaped by embedded, yet unseen protoplanets, as indicated by numerous gas and dust substructures consistent with planet-disc interaction models. We report the first detection of simple organic molecules, HCN and C2H, tracing a candidate circumplanetary disc (CPD) in the HD 163296 system, located at an orbital radius of $R=88\pm7$ au and azimuth $φ=46\pm3^\circ$ (or $R=0.75''$, $\rm{PA}=350^\circ$ in projected sky coordinates), and originating near the midplane of the circumstellar disc. The signature is localised but spectrally resolved, and it overlaps with a previously reported planet candidate, P94, identified through kinematic perturbations traced by CO lines. We propose a scenario in which the observed chemical anomalies arise from increased heating driven by the forming planet and ongoing accretion through its CPD, facilitating the thermal desorption of species that would otherwise remain frozen out in the disc midplane, and potentially triggering the activation barriers of chemical reactions that lead to enhanced molecular production. Based on a first-order dynamical analysis of the HCN spectrum from the CPD--isolated with a 7$σ$ significance--we infer an upper limit on the planet mass of 1.8 $M_{\rm Jup}$, consistent with predictions from CO kinematics and constraints from direct imaging studies. By comparing the CPD sizes derived from our models with theoretical expectations where the CPD radius corresponds to roughly one-third of the planet's Hill radius, we favor CPD gas temperatures $T > 150$ K, planet masses $M_{\rm p} < 1.0$ $M_{\rm Jup}$, and CPD radii $R_{\rm CPD} < 2$ au.

Figures (11)

• Figure 1: HCN and millimeter dust continuum emission from the disc of HD 163296 as observed with ALMA. The left panel shows the absolute peak intensity of HCN $J=3\rightarrow2$ line data from the MAPS project oberg+2021, computed using the Rayleigh-Jeans approximation, while the middle panel provides a zoomed-in view of the central region. For the same section, the right panel displays the millimeter dust continuum emission from DSHARP andrews+2018isella+2018, with dashed and solid lines marking the radial locations of selected continuum gaps (DXX) and rings (BXX), respectively. Overlaid is the localised $^{12}$CO velocity perturbation identified by izquierdo+2022, along with contours of HCN emission at $T=10$ K for reference. The circle labeled P94 marks the location of the planet candidate associated with this velocity feature, which we propose is also responsible for the localised intensity signals observed in the channel maps of HCN and C$_{\rm 2}$H lines (see Sect. \ref{['subsec:signals']}).
• Figure 2: Selected channel maps of HCN line intensity from the HD 163296 disc (top row), compared with those from the best-fit discminer model (middle row), computed using the Rayleigh-Jeans approximation. Contours in the top row enclose emission above five times the rms noise level of 0.33 K. Also shown are intensity residuals for each velocity channel (bottom row), with overlaid isovelocity contours from the model upper and lower surfaces as solid and dashed lines, respectively. For reference, the best-fit systemic velocity is $\upsilon_{\rm sys}=5.79$ km s$^{-1}$. The synthesized beam of the observations is shown in the bottom left corner of the panels in the top row. Residuals with amplitudes lower than four times the rms noise have been masked.
• Figure 3: Elevation (top and middle) and peak intensity (bottom) profiles for the front side of the disc of HD 163296, probed by HCN $J=3\rightarrow2$ emission, shown for both the best-fit model (blue curves) and the datacube (orange points). The DXX dashed lines mark the locations of the Band 6 dust gaps, while K94 and K260 indicate the orbital separations of previously reported kinematic planet candidates. The surface points in the top panel, extracted with disksurf, reveal significant vertical substructure, including two prominent 30 au wide dips at 69 and 135 au, and a tentative dip at 271 au, labeled as HXX in the middle panel. An optically thin window, highlighted with gray shading and centred on the H69 dip, provides a view of the disc midplane near D85, where a localised signal associated with the P94 planet arises (see Sect. \ref{['subsec:signals']}). The gray points in the bottom panel are peak intensity values mapped into the disc frame using the model's best-fit geometrical parameters.
• Figure 4: Cartoon illustrating the radial modulations in the disc surface traced by HCN emission and the relative location of the CPD candidate, whose emission is consistent with an origin near the midplane. This region is partially accessible thanks to a prominent HCN gap near the CPD's orbital separation and the moderate inclination of the disc.
• Figure 5: Illustration of the folding procedure used to isolate the HCN $J=3\rightarrow2$ spectra from the localised signal around the kinematic planet P94, possibly tracing circumplanetary disc (CPD) emission. The blueshifted side of the circumstellar disc (CSD) is spectrally mirrored and subtracted from the redshifted side, where the planet resides, retaining the excess intensity contribution from the candidate CPD. The bottom panel shows the resulting folded spectra extracted from a $7\times7$ pixel box ($\sim\!2$ beams per side) centred on the proposed CPD location at $R = 0\hbox{$.\!\!^{\prime\prime}$}{75}$ and $\mathrm{PA} = 350^\circ$. Gaussian fits to the brightest 20% of the spectra are overlaid as orange lines. The global peak profile has an amplitude of 5.9 mJy beam$^{-1}$ and a centroid velocity of 2.29 km s$^{-1}$ marked by the vertical dashed line, measured with respect to the inferred systemic velocity of the CSD.