2009ApJ...703..585C

We have observed a small sample of hot post-asymptotic giant branch (AGB) stars with the Infrared Array Camera (IRAC) and the Infrared Spectrograph (IRS) on board the Spitzer Space Telescope. The stars were selected from the literature on the basis of their far-infrared (IR) excess (i.e., post-AGB candidates) and B spectral type (i.e., close to the ionization of the envelope). The combination of our IRAC observations with Two Micron All Sky Survey and IRAS catalog data, along with previous radio observations in the cm range (where available) allowed us to model the spectral energy distributions of our targets and find that in almost all of them at least two shells of dust at different temperatures must be present, the hot dust component ranging up to 10³ K. In several targets, grains larger than 1 µm are needed to match the far-IR data points. In particular, in IRAS 17423-1755 grains up to 100 µm must be introduced to match the emission in the millimeter range. We obtained IRS spectra to identify the chemistry of the envelopes and found that more than one-third of the sources in our sample have mixed chemistry, showing both mid-IR bands attributed to polycyclic aromatic hydrocarbons (PAHs) and silicate features. The analysis of the PAH features indicates that these molecules are located in the outflows, far away from the central stars. We consider the larger than expected percentage of mixed-chemistry targets as a selection bias toward stars with a disk or torus around them. Our results strengthen the current picture of mixed chemistry being due to the spatial segregation of different dust populations in the envelopes.