Direct Detection of Giant Exoplanets
ING Banner
Home > Public Information > ING Newsletter > No. 9, March 2005 > Direct Detection of Giant Exoplanets

ING Newsletter No. 9, March 2005

Previous: Wide Field Survey: Final Data Products | Up: Table of Contents | Next: LIRIS Observations of SN 2004ao
Reference: ING Newsl., No. 9, page 11-14.
Article mirrored at: La Palma server | Cambridge server
Other available formats: PDF

Direct Detection of Giant Exoplanets

J. A. Caballero* (IAC), V. J. S. Béjar (Proyecto Gran Telescopio Canarias, IAC)

Since the discovery in 1995 of the first extrasolar planet candidate around a solar type star using the radial velocity method (Mayor & Queloz, 1995), to date (beginning of 2005), 135 candidate planets around main sequence stars have been discovered by the transit and the radial velocity (RV) methods. Their minimum masses are in the range 0.045 to 13 MJup. The proximity of these planets to their host stars has prevented direct imaging and spectroscopy, making a precise characterisation of their physical structure and chemical composition difficult.

The least massive objects imaged and spectroscopically confirmed outside the Solar System are the so called isolated planetary-mass objects (IPMOs) discovered in the s Orionis cluster (age ~3Myr, distance ~350pc), with masses in the range 3–13MJup  (Zapatero Osorio et al., 2000, 2002; Béjar et al., 2001). Very recently, Chauvin et al. (2004) have announced the discovery of a ~ 5MJup object at a projected separation of 55AU of a brown dwarf of the TW Hydrae association (age ~ 8 Myr, distance ~70pc). This object awaits confirmation by proper motion studies and high signal to noise spectroscopy. Slightly more massive is G196-3B, a substellar companion of a young nearby M dwarf (Rebolo et al., 1998). Its mass could be significantly lower than 25MJup if the age of the system is confirmed to be much less than 100Myr (McGovern et al., 2004).

The JOVIAN Project

The aim of the JOVIAN project (Jupiter-like Objects in the Visible and in the Infrared: their Astrophysical Nature; P. I.: R. Rebolo) is to achieve the direct detection and characterisation of objects down to the mass of Jupiter, and help, through selected observations, to shed light on the formation of massive planets.

Logo of the JOVIAN Project. [ JPEG | TIFF ].

In the last six years we have followed two major strategies for direct detection of such objects: wide field imaging searches for 1 to ~13MJup objects in several very young open clusters; and high spatial-resolution imaging, with  Adaptive Optics (AO) or the Hubble Space Telescope (HST), of young nearby late-type stars in the solar neighbourhood (age 600–30Myr, distance <50pc). In both strategies, youth is a key parameter given the large overluminosity of ultra-low mass objects during the contraction phase. We summarise below some of the results achieved in the ongoing JOVIAN project.

Figure 1
Figure 1 (left). JHKs composite image of the G196-3AB system. The images were taken with the NAOMI+INGRID Adaptive Optics system at the William Herschel Telescope. North is up and East is left. Size is 41 arcsec´41 arcsec. G196-3A, an extremely young nearby M star, is in the centre of the field of view. G196-3B, 13.5 arcsec to the SWS, is a late L spectral type brown dwarf that shares common proper motion with the A component. Estimated mass of G196-3B is 25MJup or less. These images are sensitive to discover superjupiters at separations >50 AU to the central star.

Figure 2 (right). Same as previous figure, but for the V639 and V647 Herculis system. The double object at 6.5 arcsec SSE of the brightest star is a blue background object of unknown nature. It does not belong to the stellar system. The FWHM of the JHKs images is better than 0.2 arcsec. [ JPEG | TIFF ].

IPMOs in the σ Orionis Cluster

The σ Orionis cluster has revealed as a paradigmatic place for understanding the formation of stellar and substellar objects. Following our first discoveries of massive brown dwarfs (50–30MJup) in the120Myr-old Pleiades cluster (Rebolo et al., 1995, 1996), we decided to investigate in a much younger cluster the formation of less massive objects down to the deuterium burning limit. The region around the multiple stellar system σ Orionis was selected because of its proximity, youth and low extinction. We have conducted RIZ surveys with the IAC-80 telescope (Observatorio del Teide) and the Wide Field Camera at the Isaac Newton Telescope (Observatorio del Roque de los Muchachos). From these surveys we covered the whole brown dwarf domain. Most of the one hundred brown dwarf candidates discovered were confirmed as bona fide cluster members, by follow-up near-infrared photometry and optical spectroscopy. Many of them have been confirmed using the ISIS spectrograph at the William Herschel Telescope or LRIS at Keck Observatory.

Figure 3
Figure 3. The σ Orionis region and the WFC surveys. The bright star to the up left corner (North East) is Alnitak, one of the stars of the Orion Belt. Blue background: I-band image from Digitised Sky Survey. Coloured intermediate level: WFC survey in VRI-bands (note the emission in R band (or Hα) of the nebula associated to Alnitak and the Horsehead Nebula). Grey foreground: very deep I-band WFC survey (Ilim about 24.5).  [ JPEG | TIFF ].

From these studies we have characterised the complete spectral sequence of the cluster in the brown dwarf domain from spectral type M6 to L1.5 (roughly 75 MJup to 13 MJup). We have found that the substellar mass spectrum increases toward lower masses and can be represented by a power-law, dN/dM ~M–0.8±0.4 (Béjar et al., 2001). Our results indicate that brown dwarfs are very common in the cluster and suggest that a similar behaviour of the mass spectrum is possible at lower masses.

In order to detect fainter and less massive objects, we have performed and planned to conduct deeper surveys in the optical (I band, using the Wide Field Camera) and in the near-infrared (JHKs bands,with ISAAC/VLT, INGRID– LIRIS/WHT, Omega-2000/3.5m Calar Alto). From the new processed data we have identified about 15 new cluster member candidates with masses in the planetary domain. Our faintest candidate, S Ori 70, resulted from a JH-band mini-survey performed with INGRID at the WHT. Near-infrared low-resolution spectroscopy obtained at the Keck Observatory led us to derive a T6 spectral type and a mass in the range 2 to 8MJup.
Figure 4
Figure 4. Pictorical view of several isolated planetary-mass objects (IPMOs) in the σ Orionis cluster (Zapatero Osorio et al., 2000).
[ JPEG | TIFF ].

Figure 5
Figure 5. Near-infrared image of S Ori 70 (marked with two lines), overimposed onto a Digitised Sky Survey image centred in the multiple stellar system s Orionis, that gives the name to the cluster. The mass of S Ori 70 is calculated to be in the range 2 to 8 Jupiter masses. [ JPEG | TIFF ].

Substellar Companions of Stars

In order to detect faint cool companions of young nearby stars, we have used the NICMOS instrument with the coronograph at the HST and AO systems attached at 4 m-class telescopes: Alfa+Omega-Cass at the 3.5-m Calar Alto, AdOpt@TNG+NICS at Telescopio Nazionale Galileo and, especially, NAOMI+INGRID at the WHT. The data taken by our group allow to resolve faint objects down to separations of  ~1 arcsec of relatively bright stars. This separation in a stellar system at 10pc corresponds to ~10AU. The sensitivity to planetary-mass companions improves when the spatial resolution is higher (i.e. nearby stars) and the contrast is lower (i.e. primaries are low-mass stars and planets are intrinsically brighter due to youth).

We are studying more than fifty stellar systems closer than 50pc, with spectral types later than solar and with features indicative of youth (high lithium abundance, X-ray and/or UV emission, membership to young proper motion associations, etc.). The ages of the stellar systems range between 30 and 600Myr. Forty of them have been completely analysed, comparing first and second astrometric epochs and performing photometry when possible. Although the data would allow us to discover objects with masses down to 3–10MJup in several of the systems, we have not detected any previously unknown substellar companion at distances between ~30 and ~1000 AU of the primaries. We have only detected two, possibly three, stellar companions in very close orbits and a previously known L-type dwarf secondary. From our study, the frequency of substellar companions at intermediate and large separations of the primary stars is <4%.

This apparently disappointing result is of great interest, since together with work performed by other authors, allows to conclude that only ~1% of the solar-like stars have massive planetary companions and brown dwarfs at intermediate and large distances (e.g. McCarthy & Zuckerman, 2004). This figure must be compared with the 7.3±1.5% of the solar-like stars that have exoplanet candidates discovered at small separations with the RV method.

Future Prospects

The ultimate goal of the JOVIAN project is to set observational constraints on the scenarios of formation of giant planets with masses from 1 to ~13MJup (jupiters and superjupiters). These objects appear to be quite abundant, as they exist at close distances of relatively old solar-like stars, but also free floating in very young open clusters. Is the lack of massive giant planets at intermediate and large distances related to the existence of IPMOs? Is there any scenario that could explain qualitatively and quantitatively the observational features? Are IPMOs the result of direct collapse and fragmentation of clouds? These questions will also be addressed by the JOVIAN project using the first light instruments of the Gran Telescopio Canarias.

Members of the JOVIAN project at Instituto de Astrofísica de Canarias: R. Rebolo, E. Martín, V. J. S. Béjar, J. A. Caballero, G. Bihain and J. Licandro (also at ING); LAEFF/INTA: M. R. Zapatero Osorio and D. Barrado y Navascués; Universidad Politécnica de Cartagena: A. Díaz, A. Pérez and I. Villo; Max-Planck-Institut für Astronomie: C. Bailer-Jones and R. Mundt; Thüringer Landessternwarte Tautenburg: J. Eisloeffel.

More information on JOVIAN can be found at ¤


*: Email contact: José A. Caballero (

Top | Back

Contact:  (Public Relations Officer)
Last modified: 21 December 2010