A most intriguing challenge is to understand how Asymptotic Giant Branch (AGB) stars transform their surrounding mass-loss shells in a couple of thousand years into the variety of shapes and sizes observed in Planetary Nebulae (PNe). There are a number of theories currently being investigated. In the generalized interacting stellar wind model, a variety of axisymmetric PN shapes are obtained by the interaction of a very fast central star wind with the progenitor AGB circumstellar envelope (Kwok 1982), when the latter is denser near the equator than the poles (Frank et al. 1993). Sahai & Trauger (1998) proposed that the primary agent for shaping PNe is not the density contrast, but a high speed collimated outflow of a few 100 km s-1. No consensus about the dominant physical process responsible for the shaping of PNe has emerged so far, but there is agreement that they occur during the early AGB-to-PN transition stage. However, details of the rate of evolution, the strength of the stellar wind, and the impact of ionization across the transition phase are very poorly known. In order to test model predictions it is essential to study nebulae from the early post-AGB phase through the very early PN phases. However, post-AGB objects are difficult to find, because this phase is very short and the star is usually obscured by a thick circumstellar dust shell.
One way of identifying new post-AGB stars is through their dust emission. In our search for new obscured PNe and post-AGB stars, we selected candidates from the IRAS Point Source Catalogue based on infrared colors typical of PNe. When we observed these candidates in the radio continuum at 6 cm, on average 20% of the objects were detected (Van de Steene & Pottasch 1993, 1995). Subsequent optical spectroscopy showed that the PN candidates detected in the radio have emission line spectra typical of PNe (Van de Steene et al. 1996a, 1996b). However, the question remained: what is the evolutionary status of those IRAS sources with colors typical of PNe, which had no detectable ionization in the radio (i.e. fluxes below 3 mJy)? It is possible that a few large, low surface-brightness PNe have escaped detection. Others of the remaining non-identified PN candidates could be very young and small PNe. Their radio flux would have been below our detection threshold and they wouldn't be identified in H due to extinction. However, the true evolutionary status of most objects with IRAS colors typical of PNe has remained unknown.
We have calculated evolutionary tracks for post-AGB stars in the IRAS color-color diagram (van Hoof et al. 1997). From this study, it became clear that post-AGB objects and PNe could be located in the same region in the IRAS color-color diagram. No other types of objects seem to have typically these particular IRAS colors. We therefore adopted the working hypothesis that the non-detected PN candidates are AGB-to-PN transition objects. The goal of this project is to determine the evolutionary status of these post-AGB candidates, their physical properties, and to relate them to their morphology.
To ensure the identification of the correct counterpart of the IRAS source and obtain accurate positions for follow-up observations, we first imaged the post-AGB candidates in the N -band. In order to investigate whether the sources have some ionization, we obtained Br spectra. We reobserved the sources showing Br emission at 3 cm and 6 cm. To confirm the identification, obtain finding charts, and photometry, we took high resolution near-infrared (J H K L ) images.
The next section describes the observations. In Sect. 3 we describe the data reduction and analysis, in Sect. 4 we discuss the individual objects and in Sect. 5 we discuss the general results. We give a summary in the last section.
© European Southern Observatory (ESO) 2000
Online publication: October 30, 2000