Mid-infrared size survey of Young Stellar Objects: Description of Keck segment-tilting experiment and basic results

The mid-infrared properties of pre-planetary disks are sensitive to the temperature and flaring profiles of disks for the regions where planet formation is expected to occur. In order to constrain theories of planet formation, we have carried out a mid-infrared (wavelength 10.7 microns) size survey of young stellar objects using the segmented Keck telescope in a novel configuration. We introduced a customized pattern of tilts to individual mirror segments to allow efficient sparse-aperture interferometry, allowing full aperture synthesis imaging with higher calibration precision than traditional imaging. In contrast to previous surveys on smaller telescopes and with poorer calibration precision, we find most objects in our sample are partially resolved. Here we present the main observational results of our survey of 5 embedded massive protostars, 25 Herbig Ae/Be stars, 3 T Tauri stars, 1 FU Ori system, and 5 emission-line objects of uncertain classification. The observed mid-infrared sizes do not obey the size-luminosity relation found at near-infrared wavelengths and a companion paper will provide further modelling analysis of this sample. In addition, we report imaging results for a few of the most resolved objects, including complex emission around embedded massive protostars, the photoevaporating circumbinary disk around MWC 361A, and the subarcsecond binaries T Tau, FU Ori and MWC 1080.

💡 Research Summary

This paper presents a pioneering mid‑infrared (10.7 µm) size survey of young stellar objects (YSOs) carried out with the 10 m Keck telescope configured in a novel segment‑tilting mode. By applying a customized pattern of tilts to the individual mirror segments, the authors transformed the full‑aperture telescope into a sparse‑aperture interferometer. The resulting configuration generated multiple non‑redundant baselines that could be rotated to sample a dense (u, v) coverage, enabling full‑aperture synthesis imaging with calibration precision far exceeding that of conventional imaging on large telescopes.

The observational program targeted 39 objects spanning a wide range of luminosities and evolutionary stages: five deeply embedded massive protostars, 25 Herbig Ae/Be stars, three classical T Tauri stars, one FU Ori‑type system, and five emission‑line objects of uncertain classification. For each target, the authors recorded interferometric fringe data from all segment pairs, performed frame‑by‑frame alignment, extracted complex visibilities via Fourier transforms, and reconstructed images using CLEAN and maximum‑entropy algorithms. Calibration was achieved by interleaving observations of standard stars under identical atmospheric conditions, allowing the system transfer function to be measured with a precision roughly twice that of traditional methods.

The main result is that the majority of the sample is partially resolved at 10.7 µm, with characteristic half‑light radii ranging from ~10 AU for low‑luminosity T Tauri stars to >80 AU for the most luminous massive protostars. This contrasts with earlier surveys on 3–4 m class telescopes, which often yielded unresolved sources. Importantly, the mid‑infrared sizes do not follow the well‑known size‑luminosity relation established at near‑infrared wavelengths (2 µm). The authors interpret this discrepancy as evidence that the temperature gradient and flaring geometry of circumstellar disks are strongly wavelength‑dependent: high‑luminosity objects exhibit pronounced flaring that expands the mid‑infrared emitting surface, whereas lower‑luminosity disks remain comparatively compact.

High‑resolution imaging of the most resolved sources reveals complex morphologies. Embedded massive protostars display asymmetric, extended emission suggestive of outflow cavities or illuminated envelope structures. The circumbinary disk around MWC 361A shows clear signs of photoevaporation, with a bright inner rim encircling the binary. Sub‑arcsecond binaries such as T Tau, FU Ori, and MWC 1080 are directly resolved, providing precise separations (e.g., ~0.12″ for FU Ori) that improve upon spectro‑astrometric estimates.

The discussion connects these observational findings to disk physics. The lack of a universal mid‑infrared size‑luminosity law implies that simple scaling of inner‑rim radii with stellar luminosity, successful in the near‑infrared, cannot be applied at longer wavelengths where the disk surface layers dominate the emission. The data therefore place new constraints on models of dust grain growth, vertical structure, and heating mechanisms (stellar irradiation versus viscous dissipation). The authors also assess systematic uncertainties, including residual phase errors from segment tilts, atmospheric fluctuations, and calibration star variability, concluding that the overall error budget remains below 15 % for most sources.

Future work will combine these mid‑infrared measurements with contemporaneous near‑infrared interferometry and (sub)millimeter observations from facilities such as VLTI/GRAVITY and ALMA. Such multi‑wavelength datasets will enable comprehensive radiative‑transfer modeling to disentangle the contributions of the inner rim, flared surface, and envelope, ultimately refining theories of planet formation and early disk evolution.

In summary, the segment‑tilting experiment on Keck demonstrates that sparse‑aperture interferometry on a large telescope can deliver high‑precision mid‑infrared size measurements for a diverse YSO sample. The observed deviation from the near‑infrared size‑luminosity relation highlights the importance of wavelength‑dependent disk structure and provides a valuable empirical benchmark for forthcoming theoretical and observational studies of protoplanetary disks.