The identification of Gl 229B marks a pivotal point in brown dwarf research. Like GD 165B, this low-luminosity companion was discovered through direct imaging, in this case using a coronagraph to occult most of the radiation from the primary star. The characteristics of Gl 229B place it unequivocally below the hydrogen-burning limit and render it the prototype for spectral class T.
Gl 229 is an early-type M-dwarf, Mv = 9.4, lying at a distance of only 5.7 parsecs (see Appendix). As one of the nearest M dwarfs, the star was included in Henry's near-infrared speckle survey of stars within 8 parsecs of the Sun [H6]. Those observations showed no evidence for any companions within 20 AU (±2 arcseconds). However, Gl 229 was also observed by T. Nakajima and collaborators, using the Johns Hopkins coronagraph on the Palomar 60-inch [N1]. This system uses an apodising mask and occulting disk to obscure the central 3 arcseconds and cut down scattered light, while tip-tilt correction (using the reflected image of the primary) provides image sharpening over the full 60 x 60 arcsecond field of view.
I-band images of Gl 229, obtained in October, 1994 revealed a potential faint companion (Figure 6.5), 13 magnitudes fainter (MI = 20.3) and at 7.8 arcseconds separation [N2]. Subsequent measurements in October, 1995 confirmed that the companion shares the proper motion of the primary, while K-band observations on the Palomar Hale 200-inch showed that the fainter star has extremely red optical/infrared colours, (I-K) = 6.2 mag. [M9]. Integrating the broadband photometry gives a luminosity L ~ 6 x 10 —6 L0, placing the object firmly in the brown dwarf regime.
Initially, the most surprising aspect of these observations was the infrared colours, (J-H) ~ 0.0, (J-K) ~ —0.1, strikingly different from late-type M dwarfs and GD 165B. Low-resolution spectroscopy [O2] supplied the answer to this puzzle, showing that the 1-2.5-^m spectrum exhibits strong molecular bands that cut into the longer wavelength regions of both the H and K atmospheric windows (Figure 6.6). As described in Section 2.2.4, those bands are due to methane, and Gl 229B is the prototype T dwarf. The prediction that cool brown dwarfs would have methane absorption dates back to the early 1960s ([T11]), but most observational astronomers had either forgotten, or never encountered, that prediction (although the spectra of Jupiter and Saturn should have given a clue).
Detailed comparison of the spectral energy distribution of Gl 229B with theoretical models leads to a temperature estimate of 960 ± 70 K [M3]; this is far too cool for a hydrogen-burning star and, like the luminosity, clearly identifies Gl 229B as a brown dwarf. As with GD 165B, the mass depends on the age, which is best estimated using the observed characteristics of the primary. Gl 229A is inactive chromospherically (no Ha emission) and coronally (weak X-ray flux), and is
Was this article helpful?