Planetary nebulae (PNe) are one of the last evolutionary stages of the majoirty of low- and intermediate mass stars (1-8 Msun). During the previous asymptotic giant branch (AGB) evolutionary phase the star ejects a large fraction of its mass during brief phases of high mass-loss. This exposes the central core of the star that radiates strongly in the UV and ionizes the surrounding nebula making it visible for study. Despite supposedly spherically symmetric mass-loss while on the AGB, many PNe show point- or axi-symmetric shapes and complex morphologies that have been interpreted as evidence for the influence of a binary companion. This and other outstanding issues in PNe research such as their small-scale structures, and abundance discrepancies suggest that our understanding of the late stages of stellar evolution is far from complete.
NGC 6302 is one of the most complex and extreme examples of a planetary nebula. The nebula is broadly bipolar, with a second narrower pair of lobes visible further out. It has a highly pinched waist that is typical of 'butterfly' bipolar PNe, as is a highly complex structure with many clumps and knots. The bipolar structure has been attributed to the presence of a very dense circumstellar disc that is seen almost edge-on and almost completely obscures the central star, which has only recently been detected.
I produced a 3D photoionization model of NGC 6302 using the Mocassin photoionization and radiative transfer code. The model is composed of an extremely dense geometrically thing circumstellar disc and a large pair of diffuse bipolar lobes, a combination which was necessary to reproduce the observed emission-line spectrum. The masses of these components, 2.2 Msun and 2.5 Msun, respectively, give a total nebular mass of 4.7 Msun, of which 1.8 Msun (39 per cent) is ionized. Discrepancies between our model fit and the observations are attributed to complex density inhomogeneities in the nebula. The potential to resolve such discrepancies with more complex models is confirmed by exploring a range of models introducing small-scale structures.
Compared to solar abundances helium is enhanced by 50 per cent, carbon is slightly subsolar, oxygen is solar, and nitrogen is enhanced by a factor of 6. These all imply a significant third dredge-up coupled with hot-bottom burning CN-cycle conversion of dredged-up carbon to nitrogen. Aluminium is also depleted by a factor of 100, consistent with depletion by dust grains.
The central star of NGC 6302 is partly obscured by the opaque circumstellar disc, which is seen almost edge-on, and as such its properties are not well constrained. However, emission from a number of high-ionization 'coronal' lines provides a strong constraint on the form of the high-energy ionizing flux. We model emission from the central star using a series of stellar model atmospheres, the properties of which are constrained from fits to the high-ionization nebular emission lines. Using a solar abundance stellar atmosphere we are unable to fit all of the observed line fluxes, but a substantially better fit was obtained using a 220 000 K hydrogen-deficient stellar atmosphere with log g = 7.0 and L = 14,300 Lsun.
The H-deficient nature of the central star atmosphere suggests that it has undergone some sort of late thermal pulse, and fits to evolutionary tracks imply a central star mass of 0.73-0.82 Msun. Time-scales for these evolutionary tracks suggest the object left the top of the AGB approximately 2100 years ago, in good agreement with sutides of the recent mass-loss event that formed one pair of the bipolar lobes. Based on the modelled nebular mass and central star mass we estimate the initial mass of the central star to be 5.5 Msun, in approximate agreement with that derived from evolutionary tracks.
You can read more about this work in Wright et al. 2011a.