Brown dwarfs, extrasolar planets and protoplanetary disks with CanariCam

Brown dwarfs
Brown dwarfs are of particular interest, because they represent an extension of the stellar initial mass function to masses well below that of the least-massive star. Thus, determination of how many are formed as a function of mass and in different types of star forming environments is fundamental to our understanding of the process of star formation itself. In addition, they may account for some of the "missing mass" which forms a dynamically significant but as yet unseen halo for our Milky Way Galaxy - and other galaxies as well. CanariCam will permit the search for different types of brown dwarfs to determine whether they are important in the gravitational dynamics of the Milky Way. The wide wavelength coverage will permit both brown dwarfs and low mass stars. Young brown dwarfs of relatively high mass will have their blackbody peak well into the near-infrared range, whilst older and lower mass objects will peak in the mid-infrared. CanariCam can provide constraints on the properties of dark, massive objects in the halo of the Galaxy, to search young stellar clusters for recently formed brown dwarfs, and to look for "super-planets" - with masses between 0.1% and 1% that of the Sun - orbiting nearby stars. Given the high sensitivity of CanariCam to a range of different types of brown dwarf the instrument, combined with the GTC, would have major advantages over other telescopes and their infrared instrumentation.

Detection of brown dwarfs

A 0.01 Msun brown dwarf, at 5 pc distance would have a flux of ~ 0.1 mJy between 10 and 25 microns, comparatively "easy" to detect with the Gran Telescopio Canarias.
A field brown dwarf of age 10¹⁰ years, mass 0.03 Msun, at 10 pc distance, would have a flux around 3 mJy at 10 microns.
A 0.01Msun brown dwarf, located in the Ophiuchus star-forming region (at 170 pc and with an age of 10⁶ years), would have its maximum flux around 5 microns, but would still have a flux of ~1 mJy at 10 microns.

Spectral identification of brown dwarfs

In the mid-infrared, brown dwarfs also have very distinct spectroscopic signatures.

Component	Range
H₂O	11-15 microns
CH₄	8 microns
NH₃	8.5-13 microns

Detecting the mid-infrared spectrum of brown dwarfs is a challenge even with an instrument as capable as CanariCam on the 10-m telescope. The maximum fluxes expected in the continuum of the relatively bright brown dwarf GL 299B are ~3 mJy and ~1 mJy in the minimum of the lines (R=50). To obtain s/n~ 5 in the minimum of the lines, a 25000-sec exposure is needed, despite degrading the low resolution of the spectrograph by a factor of 2. Although this is a long exposure time (7 hours), it is feasible with CanariCam on the 10-m telescope.

Extrasolar planets

The direct observation of extrasolar planets is not currently possible, except, perhaps, for super-Jupiter sized objects very distant from their parent star. However, a 10-m telescope equipped with diffraction-limited mid-infrared instrumentation would make direct observation of giant planets at solar system distances, around nearby stars, possible for the first time. In the visible, the difference in magnitude between planets and their parent stars is around 20 magnitudes, or a factor of 10⁸ in luminosity, but in the mid-infrared the contrast increases substantially. A solar-type star is approximately a 6000K blackbody with a spectrum peaking in the visible. On the other hand, a giant planet at 5 -10 AU from a solar-type star will be close to its maximum blackbody emission at 25 microns. Even more distant planets will still be close to the maximum of the blackbody curve and maximum contrast and, in addition, we increase the angular separation between the planet and the star.

The coronagraphic may provide the opportunity to detect extrasolar planets around many nearby stars. In fact, plans already exist for a search of this type with the Keck, although this requires a preparatory study of several years duration and so the GTC will still be highly competitive in this field at the time of first light. This will not be the case if the installation of CanariCam on the telescope were to be seriously delayed. An even more ambitious project for the future is the spectral observation of any detected planets. Theoretically the planet can be classified by means of its spectrum as key components in its atmosphere such as methane; ozone and water vapour have lines in the thermal infrared.

Proto-planetary disks

The discovery of disks of particulate matter orbiting solar type stars was one of IRAS's major discoveries. Modelling of a variety of IRAS and groundbased data suggests that these disks contain particles ranging in size from a few microns to as large as asteroids and comets, with some arguments suggestive of still larger, planet-sized objects. Further exploration of this phenomenon will help us to understand the frequency and properties of planetary systems around nearby stars. CanariCam has a unique role in this field because its wide wavelength coverage allows the disks to be detected at a wide range of blackbody temperatures and thus distances from the parent star. Detailed imaging of the nearby disks at a range of wavelengths can constrain models for the grain properties and distribution while searching for the central voids inferred from the IRAS data. In addition, CanariCam can compare low resolution spectra of the circumstellar material with spectra of emission from comets in our own solar system in order to compare the physical conditions in the primitive solar system to those in the debris disks.

The study of the infrared excesses of these stars can provide valuable information about the dynamics of the star's planetary system. Dust at 80 AU (the typical distance of the inner Kuiper Belt in the Solar System) from Beta Pictoris is at a temperature of 160 K. Based on the blackbody temperature expected at this distance (50 K), we can estimate that the grains are typically ~0.1- 0.2 microns in diameter. Dust of this size should be removed rapidly from the system (in 3x10⁵ years) by Poynting-Robertson drag, and so the disk particles must be constantly replenished, probably due to cometary activity. Strong support for this cometary renovation hypothesis comes from the fact that the silicate feature in the Beta Pictoris spectrum does not resemble other interstellar and circumstellar silicate features. Instead, it shows a strong resemblance to the cometary silicate line. This even allows the possibility of carrying out mineralogy - identifying the mineralogical composition of the silicate grains in protoplanetary disks.

In the same way, dust falling towards the star because of the Poynting-Robertson effect should fill the disk down to the star. In contrast, what is seen is a dust-free central zone, which suggests that there are several planets in the system, which clear the dust from this region. In fact, there is evidence of a central warp to the disk, which is may be due to gravitational perturbations of at least one giant planet. The angular resolution of CanariCam would be 2.5AU (at 8 microns) at a distance of 16 pc from the Sun (the distance of Beta Pictoris).

Spectroscopy at both 10 and 20 microns is useful for maximum scientific return, which is underscored by looking at the ISO-SWS/LWS spectrum of the Herbig Ae/Be star HD 100546 (Malfait et al. 1998, A&A, 332, L25). The HD 100546 spectrum, made with a spectral resolution of about 1000, is remarkably similar to that of Comet Hale-Bopp, a situation reminiscent of the similarity of the Beta Pic and Comet Halley 10 microns silicate features mentioned above. The 8-30 micron spectrum of HD 100546 is well fit by a superposition of spectra of three primary constituents: (1) crystalline forsterite (Mg₂SiO₄), to account for the relatively narrow emission bands; (2) amorphous olivines, to account for the broad 10 and 18 microns features; and (3) FeO, to account for a broad feature near 23 microns. It is obvious that these constituents are powerfully diagnosed with both 10 and 20 micron spectroscopy. As we have shown previously for Beta Pic (Knacke et al.), a spectral resolution R ~ 100, compatible with the CanariCam LoRes modes, is sufficient to resolve the narrow crystalline features. CanariCam's high-resolution mode, which provides R » 1400 at 10 microns, can be used for much more refined mineralogy of the features found at the lower resolution.

The continuum sensitivities with the best spectral resolutions in the LoRes-10 (R ~ 100) and HiRes-10 (R ~ 1000) gratings are about 0.1 mJy and 1 mJy for 1 h of chopped integration on a point source. It is this continuum sensitivity (rather than the line sensitivity) that applies to observations of the broad silicate feature. The silicate emission feature in Beta Pic (at 16 pc) has a strength above the underlying continuum of about 1 Jy at 10 microns. Thus, we can do HiRes-10 spectroscopy with a signal-to-noise ratio of 10 in 1 hour of chopped integration on a Beta -Pic-like disk that is 160 pc away from the Sun, and LoRes-10 spectroscopy with that signal-to-noise ratio on a Beta -Pic-like disk that is 500 pc away from the Sun. This implies that a productive search for the silicate feature on nearly all of the IRAS-selected Vega-type disks can be carried out even though many are much fainter than Beta Pic.

Back to the CanariCam Science Page.