Abstract

The recent discoveries of extra-solar planets have generated great excitement. However, all of the detections have been indirect, and are resulted from only one technique, namely high precision Doppler shift measurements. The Palomar Testbed Interferometer is capable of directly imaging these stars. Our latest results for the well-known spectroscopic binary $\iota$~Peg demonstrate a resolving power of 2 mas, and a measurement accuracy of 0.2 mas. The geometric and physical parameters of the iota Peg system provide important clues for 51 Peg. Our results from PTI clearly indicate that 51 Peg has been resolved.

Discoveries of extra-solar planets during the past few years have resulted in great interest in these solar type stars. In particular, 51~Peg, around which the first extra-solar planet was discovered, has become the prototype of the so-called ``Hot Jupiters''. This planet was discovered by Mayor, M. & Queloz, D. (1995), and confirmed immediately by Marcy, J. et al (1997). However, it is important to keep in mind that the high precision radial velocity measurements are only indirect evidence for the existence of planets, and that there are several possible alternative interpretations for small changes of radial velocity. For example, a scenario in which these changes are produced by stellar pulsations has been advocated by Gray, D. F. (1997).

For the planet around 51~Peg, the estimated minimum mass m.sini is 0.45 M_J, where M_J is the mass of Jupiter. The minimum separation between the star and the planet is 0.051 AU. At this small separation, the temperature of the planet is expected to be at least 1300^oK, making it very difficult for a Jupiter-mass planet to form and to survive. However, this separation is within the resolving capability of the Palomar Testbed Interferometer (PTI).

Our latest PTI results on the spectroscopic binary iota Peg (Pan et al 1996) have demonstrated an angular resolution of 2 mas and a measurement accuracy of 0.2 mas. The orbital period derived from the PTI data is consistent with the spectroscopic results to 2 10^-5. Another common parameter between interferometry and spectroscopy, i.e., the orbital eccentricity e, agrees to within 1 %. Good agreement between spectroscopy and interferometry indicate the results are very reliable. Since the mass ratio is accurately determined by spectroscopy, the masses of both components are determined to within ~ 1 %, and are found to be consistent with the spectral types. It is interesting to note the similarities between iota Peg and 51 Peg. Both of the primaries are solar-type stars with companions in circular orbits with short periods (10.2 days for iota Peg). The separation of iota Peg is only 0.12 AU, and the minimum separation of 51 Peg is 0.05 AU, which is consistent with its period of 4.23 days.

Preliminary results from PTI indicate that 51 Peg has been resolved. There are two possible scenarios: either the disk of the primary is resolved, or the system of 51 Peg is resolved. Originally, we believed that the stellar disk of 51 Peg was resolved. This seemed plausible because the spectral classification of 51 Peg was uncertain, with proposed values of G2.5 IV, G5 V , and others in the literature. Since then, however, a consensus has emerged that 51 Peg is a G2 - G3 main-sequence star, and is definitely not a giant or sub-giant. In addition, the distance of 51 Peg has been determined accurately by Hipparcos as 15.36 ± 0.01 pc. Thus the estimated stellar angular diameter of 51 Peg is 0.6 ± 0.1 mas if it is a main-sequence star, significantly below the resolution limit of PTI. We now have had one years data, and hope to finish our data analysis soon and to confirm our conclusions in the near future.

References

Mayor, M. and Queloz 1995, Nature, 378, 355
Gray D. F. 1997, Nature, 385, 795
Marcy, J. W., Cutler, R. P., Williams, E., Bildsten, L., Graham, J. R., Ghez, A. M., and Jernigan, J. G. 1997, ApJ, (accepted)
Pan, X. P., Kulkarni, S., Colavita, M. M. and Shao, M. 1996, BAAS, 28, No. 4, 1312