The mineralogy, geometry and mass-loss history of IRAS 16342-3814

Dijkstra C., Waters L.B.F.M., Kemper F., Min M., Matsuura M., Zijlstra A.A., de Koter A., Hony S., Dominik C., 2003, A&A, 399, 1037

Abstract

We present the 2-200 micron Infrared Space Observatory (ISO) spectrum and 3.8-20 micron ISAAC and TIMMI2 images of the extreme OH/IR star IRAS 16342-3814. Amorphous silicate absorption features are seen at 10 and 20 micron, together with crystalline silicate absorption features up to almost 45 micron. No other OH/IR star is known to have crystalline silicate features in absorption up to these wavelengths. This suggests that IRAS 16342-3814 must have, or recently had, an extremely high mass-loss rate. Indeed, preliminary radiative transfer calculations suggest that the mass-loss rate may be as large as 10-3 Msun yr-1. The 3.8 micron ISAAC image shows a bipolar reflection nebula with a dark equatorial waist or torus, similar to that seen in optical images taken with the Hubble Space Telescope (HST). The position angle of the nebula decreases significantly with increasing wavelength, suggesting that the dominant source of emission changes from scattering to thermal emission. Still, even up to 20 micron the nebula is oriented approximately along the major axis of the nebula seen in the HST and ISAAC images, suggesting that the torus must be very cold, in agreement with the very red ISO spectrum. The 20 micron image shows a roughly spherically symmetric extended halo, approximately 6 arcsec in diameter, which is probably due to a previous phase of mass-loss on the AGB, suggesting a transition from a (more) spherically symmetric to a (more) axial symmetric form of mass-loss at the end of the AGB. Using a simple model, we estimate the maximum dust particle sizes in the torus and in the reflection nebula to be 1.3 and 0.09 micron respectively. The size of the particles in the torus is large compared to typical ISM values, but in agreement with high mass-loss rate objects like AFGL 4106 and HD161796. We discuss the possible reason for the difference in particle size between the torus and the reflection nebula.

paper (pdf 500 kb)


Last update: June 10, 2003