First millimeter detection of the disk around a young, isolated, planetary-mass object

TL;DR

This study reports the first millimeter detection of a disk around a young, isolated planetary-mass object, providing insights into its properties and formation mechanisms by comparing it with higher-mass objects.

Contribution

It presents the first millimeter continuum detection of a disk around a free-floating planetary-mass object, expanding understanding of disk properties at planetary masses.

Findings

Detected the disk at 233 GHz with ALMA.

Estimated disk mass between 0.07 and 0.63 Earth masses.

Disk properties follow a scaling relation with the central object mass.

Abstract

OTS44 is one of only four free-floating planets known to have a disk. We have previously shown that it is the coolest and least massive known free-floating planet ($\sim$12 M$_{\rm Jup}$) with a substantial disk that is actively accreting. We have obtained Band 6 (233 GHz) ALMA continuum data of this very young disk-bearing object. The data shows a clear unresolved detection of the source. We obtained disk-mass estimates via empirical correlations derived for young, higher-mass, central (substellar) objects. The range of values obtained are between 0.07 and 0.63 M$_{\oplus}$ (dust masses). We compare the properties of this unique disk with those recently reported around higher-mass (brown dwarfs) young objects in order to infer constraints on its mechanism of formation. While extreme assumptions on dust temperature yield disk-mass values that could slightly diverge from the general trends found for more massive brown dwarfs, a range of sensible values provide disk masses compatible with a unique scaling relation between $M_{\rm dust}$ and $M_{*}$ through the substellar domain down to planetary masses.

Equations

$$M_{dust}=\frac{F_{\nu }\times d^2}{{\kappa }_{\nu }\times B_{\nu }(T_{dust})}$$

Full text

Brown dwarfs — stars: pre-main sequence — stars: low-mass — stars: formation

††software: astropy (Astropy Collaboration et al., 2013)

1 Introduction

Key questions are still unanswered about how low-mass stars come to emerge and gain their masses from their natal molecular clouds, but when we move toward the substellar domain (with masses below the hydrogen-burning mass limit, $\leq$ 0.072$M_{\odot}$), these questions become even more fundamental.

Several scenarios have been proposed to prevent a substellar core in a dense environment from accreting to stellar mass (e.g., dynamical interactions, Reipurth & Clarke 2001; Umbreit et al. 2005, disk fragmentation, Goodwin & Whitworth 2007; Stamatellos et al. 2007, or photoevaporation, Whitworth & Zinnecker 2004). Alternatively, brown dwarfs and free-floating planets (with masses below the deuterium-burning mass limit, $\leq$ 13$M_{Jup}$) could form in an isolated mode by direct collapse. This could be possible either introducing turbulence so that the Jeans mass decreases in the first place (Padoan & Nordlund, 2002; Hennebelle & Chabrier, 2008) or from a filament collapse (e.g., Inutsuka & Miyama 1992) forming low-mass cores that experience high self-erosion in outflows (Machida et al., 2009).

In order to test these models, observational constraints need to be placed on the main features of star formation (disk properties / morphology, accretion, outflows, etc.) toward the lowest possible masses, so that conclusions can be drawn from the behaviors of those properties with the mass of the central object.

Young brown dwarfs were shown to have substantial circumstellar material based on near-infrared (IR; e.g. Oasa et al. 1999; Muench et al. 2001), mid-IR (ISO & $Spitzer$, e.g., Natta & Testi 2001; Natta et al. 2002; Apai et al. 2005; Barrado y Navascués et al. 2007; Luhman et al. 2008; Bayo et al. 2012), far-IR ($Herschel$, e.g., Harvey et al. 2012a, b; Alves de Oliveira et al. 2013; Joergens et al. 2013; Liu et al. 2015) and single-dish millimeter continuum photometry (e.g., Klein et al. 2003; Scholz et al. 2006; Mohanty et al. 2013). Only a few years ago, millimeter interferometers became more sensitive, and the disk around 2MJ0444 (M7.25, $\sim$0.05$M_{\odot}$) was the first of its kind to be spatially resolved at 1.3mm (CARMA observations by Ricci et al. 2013) with an estimated disk radius of 15–30 au. Even more recently, ALMA allowed several groups to perform small surveys in the substellar domain including intermediate-to-high-mass brown dwarfs (Ricci et al., 2014; Daemgen et al., 2016; Pascucci et al., 2016; Testi et al., 2016; van der Plas et al., 2016). Measurements of low spectral indices of a handful objects showed that dust grains have grown to millimeter sizes even in these very low-mass environments (e.g. Ricci et al. 2013), challenging models of planetesimal formation, which predict that dust growth is more limited by radial drift than in disks around stellar objects (Pinilla et al., 2013). Hints of disk truncation in the substellar domain were presented (Testi et al., 2016), as well as different estimates of the temperature of the dust in substellar disks and an evolution (Andrews et al., 2013; van der Plas et al., 2016), within the first 10 Myr, of the $M_{\rm dust}$–$M_{*}$ relationship (pointing again toward grain growth, drift, and fragmentation).

Moving down in mass, the lowest-mass isolated objects found to harbor a disk are, to the best of our knowledge, Proplyd 133-353 ($\leq 13\rm{M_{\rm Jup}}$, Fang et al. 2016), Cha 1109-7734 ($\sim 8\rm{M_{\rm Jup}}$, Luhman et al. 2005a), J02265658-5327032 ($\sim 13\rm{M_{\rm Jup}}$; Boucher et al. 2016), and OTS44 ($\sim 12\rm{M_{\rm Jup}}$; Joergens et al. 2013), with central object masses well below what has been studied with ALMA until now.

In this Letter we present the first millimeter detection of one of these four extremely low-mass objects, OTS44. In Section 2, we describe in more detail the target; in Section 3, we describe our observations; in Section 4, we present our disk estimate and the comparison with the literature; and in Section 5, we present our conclusions.

2 OTS44

OTS44 is the object with the latest spectral type in the Chamaeleon I (Cha I) star forming region (M9.5) with a mass below or close to the planetary border (6-17$\rm{M_{\rm Jup}}$, Luhman et al. 2005b; Bonnefoy et al. 2014). First evidence for a disk around OTS44 came from mid- and far-IR excess emission detected with $Spitzer$ and $Herschel$ (Luhman et al., 2005b; Harvey et al., 2012a, b). In addition, we observed OTS44 with VLT/SINFONI and detected strong, broad, and variable Pa $\beta$ emission, which is evidence for active disk accretion in the planetary regime with a relatively high mass-accretion rate (8$\times$10*-12* M*⊙* yr*-1*; Joergens et al. 2013).

We recently determined the properties of the disk of OTS44 (Joergens et al., 2013; Liu et al., 2015) through radiative transfer modeling of its spectral energy distribution (SED) from the optical to the far-IR applying the radiative transfer code MC3D (Wolf, 2003) and a Bayesian analysis. The disk model that fitted the mid- and far-IR data best was that of a highly flared disk with a dust mass of 0.17$\rm{M_{\oplus}}$. However, our far-IR $Herschel$ measurements (a detection at 70$\mu$m and an upper limit at 160$\mu$m) are insensitive to millimeter-sized grains, which prevented us from concluding about the presence of large grains (a maximum grain size of 100$\mu$m was assumed in Liu et al. 2015), with this potentially leading to an underestimation of the disk dust mass. In this Letter we report the first ALMA detection of the disk of a planetary-mass object, providing a more robust estimate of its mass and supporting the idea that the value obtained in Liu et al. (2015) was indeed an underestimation.

3 ALMA data

ALMA Cycle 3 Band 6 continuum data were obtained as part of the program 2015.1.00243.S. Four spectral windows (centered at 224, 226, 240, and 242 GHz and each one with $\sim$1.9 GHz bandwidth) were defined to be collapsed in a single “broadband” continuum image.

The data processing was performed with CASA (McMullin et al., 2007), following the standard steps starting from the measurement sets: visual inspection of the performance of antennas and scans, flagging corrupted or useless data (using solutions derived from the water vapor radiometer), correcting in bandpass, flux and phase, further flagging (shadowing, spectral window edges anomalies, etc.), deriving the bandpass solution per spectral window, and creating the cube from the calibrated data (the derived flux uncertainty is $\sim$8%, but, based on the ALMA documentation, we assumed a more conservative 10% value from now on).

In addition to these general steps, we tried to apply self-calibration to improve the extended flux recovery but the signal-to-noise ratio (S/N) of the data was not good enough. Finally, we tried to bin our data in two spectral windows as separated as possible to obtain a spectral index, but once again, the S/N of the data was not good enough for this purpose.

The final CLEAN, primary beam corrected image (natural weighting) has a beam of 1.6” $\times$ 1.6”, a RMS of 9.8$\mu$Jy/beam, a central reference frequency of 233 GHz, and a frequency range covered (not continuously) of 20 GHz. We detected OTS44 as a point-like source with a peak flux value 0.101$\pm$0.01mJy (see Fig. 1).

4 Results and discussion: dust disk mass

The most direct quantity that we can derive from our data is the dust mass of the disk. In this section, we provide such estimates and compare them with the available literature.

4.1 Disk mass via analytical prescription

In order to estimate the disk dust mass from millimeter data, we assume that the emission is optically thin and isothermal at temperature Tdust, and therefore

[TABLE]

where $F_{\nu}$ is the flux density, $d$ is the distance to Cha I (160 pc is assumed; Whittet et al. 1997), $\kappa_{\nu}$ is the mass absorption coefficient, and $B_{\nu}$ is the Planck function of temperature $T_{\rm dust}$ at the observed frequency.

We have adopted a $\kappa_{\nu}$ value of 2.3 cm2 g*-1* at 230 GHz with a frequency dependence of $\nu^{0.4}$, the same as in Andrews et al. (2013) for Taurus, and more recently by Pascucci et al. (2016) for Cha I. For consistency, we have adapted all Mdust values from the literature to be compatible with this assumption.

The remaining strong assumption rests on the choice of Tdust. A typical value used for Tdust is 20 K, but Andrews et al. (2013) showed that this temperature scales significantly with the luminosity of the central object, proposing the relation: $T_{\rm dust}=25\times(L_{*}/L_{\odot})^{1/4}K$. However, this empirical relation yields a temperature of 5.5 K for OTS44 (assuming a luminosity of 0.0024 $L_{\odot}$; Joergens et al. 2013), which is unrealistic given the higher temperatures reported merely by heating due to the interstellar radiation field (IRF; Draine 2011). Very recently, van der Plas et al. (2016) revised this dependence as $T_{\rm dust}=22\times(L_{*}/L_{\odot})^{0.16}K$, which translates to a dust temperature of 8.4 K for OTS44 (with the same $L_{\odot}$ as before).

For completeness we have considered four different temperatures: the “classical” 20 K, the 8.4 K obtained assuming the relationship from van der Plas et al. (2016), that of IRF (7.5 K, Draine 2011), and the 5.5 K derived following the relationship from Andrews et al. (2013). The corresponding estimates for the dust mass of the disk of OTS44 are: 0.07, 0.27, 0.33, and 0.63 M*⊕*, respectively.

In Fig. 2, we show how our four estimates of the dust mass in the disk around OTS44 compare to literature values for stellar and substellar objects in the literature from Pascucci et al. (2016); van der Plas et al. (2016); Daemgen et al. (2016); Testi et al. (2016). We note that we are only comparing values derived from interferometric measurements since those derived from single-dish observations tend to be higher pointing toward a contamination of the measurement by the molecular cloud.

The estimates for the different samples shown in Fig. 2, are consistent with each other since we recalculated the dust masses using the same absorption coefficient (to rescale the values in Testi et al. 2016). In the case where Tdust scales with the bolometric luminosity of the central object, the same caution was taken in the use of consistent opacities, and we adopted the stellar parameters from Manara et al. (2014, 2016) and Manara et al. (2017), for the Pascucci et al. (2016) sample (Cha I objects, as OTS44), where those values are generally in agreement with Luhman (2004) or more recently Bayo et al. (2017).

The conclusion from these comparisons is that a “very high” Tdust value of 20 K translates in an extremely low-mass disk that would fall in the lower envelope of the $M_{*}-M_{disk}$ relationship drawn by the literature data. However, 20 K is probably an unrealistic value for the dust temperature unless disks around these very low-mass central objects are much smaller and flared than those around low-mass stars, a trend challenged, for example, by Liu et al. (2015), but worthy of further scrutiny. On the other hand, values between 5.5 K and 8.4 K, yield disk masses compatible with the dispersion observed in the literature data (although 5.5 K is probably also unrealistic due to the fact that including an IRF already brings this value to 7.5 K), pushing the $M_{*}-M_{disk}$ correlation into, or at the border of, the planetary-mass domain.

5 Conclusions

We have presented the first millimeter detection of the dusty disk around an isolated planetary-mass object. Taking into account the strong assumptions to derive its dust mass, and following different approaches to do so, the values we obtain are consistent with the log-log linear relation between $M_{*}$ and Mdust, holding even at the planetary-mass domain.

However, these mass estimates are severely limited by assumptions on poorly constrained parameters such as the dust properties in disks around these extremely low-mass objects. In addition, crucial aspects such as grain growth cannot be probed with one-band millimeter observations. To tackle these questions, the ideal complementary data would include $\sim$200$\mu$m (unfortunately, most likely beyond the limits of ALMA Band 10, $\sim$869 GHz, capabilities) observations and extremely sensitive ALMA Band 3 observations.

This Letter makes use of the following ALMA data: ADS/JAO.ALMA#2015.1.00243.S. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), NSC and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO and NAOJ. A. Bayo acknowledges financial support from the Proyecto Fondecyt Iniciación 11140572, and I. de Gregorio for very useful discussions. A.N. acknowledges support from Science Foundation Ireland (grant 13/ERC/12907). Y.L. acknowledges support by NSFC grant 11503087, by the Natural Science Foundation of Jiangsu Province of China (grant No. BK20141046), and by the German Academic Exchange Service and the China Scholarship Council. J.O. acknowledges the UV support to the MPIA-UV Max Planck Tandem Group. H.B. acknowledges support from the European Science Council under the Horizon 2020 framework program via the ERC Consolidator grant CSF-648505. P.P. acknowledges support by NASA through Hubble Fellowship grant HST-HF2-51380.001-A awarded by the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., for NASA, under contract NAS 5- 26555. This research made use of: APLpy, an open-source plotting package for Python hosted at http://aplpy.github.com; Astropy, a community-developed core Python package for Astronomy (Astropy Collaboration et al., 2013); NASA’s Astrophysics Data System; the SIMBAD database, operated at CDS, Strasbourg, France; and the VizieR catalog access tool, CDS, Strasbourg, France (Ochsenbein et al., 2000). ALMA