The Dust Ring of Luminous Blue Variable Candidate HD 168625: Infrared Observations and Model Calculations
We present a 2.218 lm image from the Hubble Space Telescope/Near Infrared Camera and Multi-Object Spectrometer (NICMOS) and a 55 lm image from ISOPHOT of the dust ring surrounding the luminous blue variable (LBV) candidate HD 168625, together with new temperature and optical depth maps derived from mid-IR images. The shell is detached from the star in the near-IR, and substructure in the overall toroidal shell is visible. The far-IR image constrains the extent of the dust shell to 2500 in diameter, providing an upper radius limit for modeling. The temperature maps and the NICMOS image show evidence for very small transiently heated dust grains in the shell. The opacity maps show higher optical depth in the limbs, consistent with interpretation of the dust shell as an equatorially enhanced torus inclined 60 with respect to the observer. An overall trend in the dust emission location with wavelength is observed and interpreted as a variation with respect to location in the nebula of either the dust grain size distribution or gas-to-dust mass ratio. Radiative transfer calculations using 2-Dust indicate that a mass-loss event occurred 5700 yr ago with a rate of ð1:9 0:1Þ 10 4 M yr 1, creating a dust torus that currently has a V 0:22 in the equatorial plane and a dustmass of ð2:5 0:1Þ 10 3 M .Using published values for the gas mass, we find a gas-to-dust mass ratio of 840, which is 4 times higher than current estimates for the interstellar medium. In addition to a high equator-to-pole density ratio ( 31) torus, an ellipticalmidshell is needed to reproduce the appearance and spectral energy distribution of the dust. Therefore, HD 168625 is an excellent example of proposed models of LBV nebulae in which a stellar wind interacts with a preexisting density contrast and creates a blowout in the polar direction perpendicular to the equatorial ring. The circumstellar shell is much lower in mass than that of LBV Carinae, suggesting thatHD168625 had a lowermass progenitor.
Timothy B. O'Hara et al. 2003 ApJ 598 1255