BANYAN. VI. Discovery of a Companion at the Brown Dwarf
Total Page:16
File Type:pdf, Size:1020Kb
BANYAN. VI. Discovery of a companion at the brown dwarf/planet-mass limit to a Tucana-Horologium M dwarf Etienne´ Artigau1, Jonathan Gagn´e 1, Jacqueline Faherty2, Lison Malo3 , Marie-Eve Naud1, Ren´eDoyon1, David Lafreni`ere1, Yuri Beletsky4 Send correspondence to [email protected] ABSTRACT We report the discovery of a substellar companion to 2MASS J02192210–3925225, a young M6 γ candidate member of the Tucana-Horologium association (30 − 40 Myr). This L4 γ com- panion has been discovered with seeing-limited direct imaging observations; at a 4′′ separation (160 AU) and a modest contrast ratio, it joins the very short list of young low-mass companions amenable to study without the aid of adaptive optics, enabling its characterization with a much wider suite of instruments than is possible for companions uncovered by high-contrast imaging surveys. With a model-dependent mass of 12–15MJup, it straddles the boundary between the planet and brown dwarf mass regimes. We present near-infrared spectroscopy of this companion and compare it to various similar objects uncovered in the last few years. The J0219–3925 system falls in a sparsely populated part of the host mass versus mass ratio diagram for binaries; the dearth of known similar companions may be due to observational biases in previous low-mass companion searches. Subject headings: 1. Introduction Psc b and Ross 458(AB)c, respectively 2000 and 1200 AU from their hosts (Goldman et al. 2010; The spectacular discoveries brought on by high- Burningham et al. 2011; Naud et al. 2014), and contrast imaging of exoplanets around nearby the presence of companions on such wide orbits stars (e.g., HR8799’s system, Marois et al. 2008, presents a significant challenge to planetary for- 2010; β Pictoris b, Lagrange et al. 2010; Foma- mation models. lhaut’s companion, Kalas et al. 2008 ) eclipse a These companions are on orbits too large fact that is often overlooked: most of the region for in situ formation by either core accretion around a star where a planet would be gravita- (Pollack et al. 1996; Alibert et al. 2005) or grav- arXiv:1505.01747v1 [astro-ph.SR] 7 May 2015 tionally bound for Gyrs is readily accessible to itational instability within protoplanetary disks seeing-limited observations for nearby stars. A (Cameron 1978; Boss 1997). Vorobyov (2013) es- handful of planetary-mass objects have indeed tablished that wide companions may form in wide been uncovered in wide-field surveys, such as GU orbits within gravitationally unstable protoplan- etary disks, but only around massive (> 0.7 M⊙) 1Institut de Recherche sur les Exoplan`etes (IREx), D´epartement de Physique, Universit´ede Montr´eal, C.P. hosts with massive (> 0.2 M⊙) protoplanetary 6128, Succ. Centre-Ville, Montr´eal, QC, H3C 3J7, Canada disks, and cannot account for the discovery of 2Department of Terrestrial Magnetism, Carnegie Insti- such companions around K stars or later-type tution of Washington, Washington, DC 20015, USA hosts. Turbulent fragmentation of a pre-stellar 3 Canada-France-Hawaii Telescope Corporation, 65-1238 core (Padoan & Nordlund 2002) is a viable alter- Mamalahoa Highway, Kamuela, HI 96743, USA native to explain the existence of such systems. 4Las Campanas Observatory, Carnegie Institution of Washington, Colina el Pino, Casilla 601, La Serena, Chile Outward migration through planet-planet inter- action is also plausible, but remains to be tested 1 as a credible mechanism at such separations and metric follow-ups of the companion and its host mass ratios (Veras et al. 2009). Statistical anal- are described in section 3. Results are described yses of the mass, age and separation of planets in section 4 and discussed in section 5. uncovered through high-contrast imaging cam- paigns suggest that these companions constitute 2. The survey a low-mass tail of the brown dwarf distribution and are therefore the results of disk or cloud frag- J0219–3925 has been observed as part of a sur- vey of 300 stars conducted at the CTIO-1.5m mentation. The dearth of low-mass (< 5 MJ ) companions found by direct imaging surveys with telescope with the SIMON. This survey was un- adaptive optics (AO) can be explained in that dertaken following the discovery of the planetary- context (Brandt et al. 2014). This analysis mostly mass companion around the M3 AB Doradus contrains the occurence of companions out to 100- member GU Psc (Naud et al. 2014), in an attempt 200 AU; whether this paucity of companions holds to identify additional comparable planetary-mass for more distant brown dwarfs and planetary-mass companions and assess their overall frequency. companions remains to be determined in a statis- The target sample consists of both confirmed and tical framework. Overall, much work needs to be strong members of young moving groups within done to understand the origin and demographics 70pc (< 120 Myr; β Pictoris, AB Doradus, THA, of these distant companions, both theoretically Columba, Carina and Argus), and it was as- and observationally. sembled from objects in Malo et al. (2013) and Gagn´eet al. (2015). For each star, the Bayesian These distant companions have distinct ad- Analysis for Nearby Young AssociatioNs II tool vantages compared to either isolated planetary- (BANYAN II; Gagn´eet al. 2014c) provided a sta- mass objects or planets detected through AO sur- tistical distance estimate, allowing us to derive a veys. The presence of a host star allows a cross- projected distance for putative companions. calibration of the properties of the planet (paral- lax, mass, age, metallicity, membership to a young By combining SIMON J-band imaging with group), and the projected separation to that host WISE photometry (Wright et al. 2010), we iden- is sufficient to allow direct study through means tified objects that had a similar position to brown not compatible with extreme-AO, such as accurate dwarfs or known very-low gravity L dwarfs in an − spectro-photometry, high-resolution spectroscopy, MW 1 vs J W 2 diagram, but that did not have optical imaging, etc. Indeed, the detailed analysis optical digitized sky survey (DSS) counterparts. performed with intermediate-resolution spectro- In a way similar to the survey that allowed photometry on GU Psc b and Ross 458(AB)c will the discovery of GU Psc b, candidates were then be impossible to obtain in the near future for ex- followed-up with Gemini-South using deep i and z oplanets uncovered by AO surveys. band imaging to confirm the very red i−z color ex- As distant companions of nearby young stars pected for an L or T dwarf. The selection methods provide important benchmarks to understand self- will be detailed in a future paper (Artigau et al. luminous gas giants, we undertook various seeing- in preparation), as refinement of selection methods limited observations using the SIMON near- and follow-up of candidates are still ongoing. infrared spectro-imager (Doyon et al. 2000) at As the SIMON observations have significantly the CTIO-1.5m telescope and GMOS-S at Gem- better resolution than 2MASS (∼1′′ versus ∼2′′ ini South in order to identify such new objects full-width at half maximum) and are significantly through their distinctive far-red and near-infrared deeper (10 σ at J ∼ 18), they provide an oppor- colors. We report here a first discovery from tunity to identify relatively tight (2 − 6′′) over- this survey; a co-moving companion to 2MASS looked companions. We performed a radial profile J02192210–3925225 (J0219–3925), an M6 γ candi- subtraction on all primaries and visually inspected date member of the Tucana-Horologium associa- the residuals. Through this process, a single can- tion (THA) that has a low-gravity L4 γ compan- didate was identified, at an angular distance of ion. about 4′′ from the M6 star J0219–3925. No other The survey that led to this discovery is de- star within our sample presented a similar candi- − ′′ scribed in section 2. The discovery and photo- date companions within in a 2 6 annulus down 2 to a contrast of ∆J < 5. 3. Observation and reduction 3.1. Imaging The J-band discovery dataset of J0219–3925 g, GMOS i, GMOS was obtained with SIMON on 2013 November 5 6 6 at the CTIO, and follow-up H and Ks-band imag- 4 4 ing were obtained on 2014 February 2, with the 2 2 same instrument and telescope. For all 3 imaging 0 0 −2 −2 sequences, we employed a 4-point dither pattern DEC offset (") DEC offset (") ′ ′ along the corners of a 2 × 2 square. At each −4 −4 dither position, 15 images (J) and 5 images (H −6 −6 6 4 2 0 −2 −4 −6 6 4 2 0 −2 −4 −6 and Ks) were taken. All sequences used a 30s RA offset (") RA offset (") per-frame exposure time. The images were sky- J, SIMON J, F2 subtracted using a sky frame constructed from the 6 6 median combination of all science images taken 4 4 on that night within that band. Flat-fielding was 2 2 performed with a flat constructed from images of 0 0 −2 −2 a flat screen. The astrometric solution was per- DEC offset (") DEC offset (") formed using a cross-match of the 2MASS cata- −4 −4 −6 −6 log with field stars. All images were registered 6 4 2 0 −2 −4 −6 6 4 2 0 −2 −4 −6 and median-combined to produce the final science RA offset (") RA offset (") frame. K, 2MASS Gemini giJ 10 The contrast ratio between the two components 6 4 was determined by performing a PSF fitting of 5 2 the two J0219–3925 components using two isolated ′ 0 0 bright and nearby field stars (< 2.5, J > 13.5) as −2 DEC offset (") DEC offset (") −5 input PSFs.