There is often dust associated with ionized regions. Compact HII regions usually have associated dust from a molecular cloud in which the massive star is being "born". Even for more evolved O-type stars and early B-type stars there is often a molecular cloud nearby whose dust is heated by the radiation from the massive star or stars. The case of the Trapezium cluster in the Orion Nebula is typical in this regard. For such stars the dust is not produced by the star, and a variety of dust features are observed. The planetary nebulae also usually have associated dust from the previous stage of evolution where they were either carbon stars or M-type giants.
The figure above shows a few example spectra of this type. The Orion nebula spectrum at upper left shows the cold dust in the molecular cloud along with the nebular emission lines from the ionized region. At short wavelengths weak UIRs are observed, apparently from the photo-dissociation region just outside the main ionized nebula. Such spectra of HII region sources can show a variety of features: amorphous and crystalline silicate features along with the UIRs in many cases. The amorphous silicate features can be in either emission or in absorption.
Planetary nebula spectra can be emission-line dominated, as with the spectrum of NGC 2440 at lower left, or have significant dust emission as in the case of the spectrum of IC 418 at upper right. Depending on whether this dust is carbon-based or oxygen-based one either typically sees the aromatic hydrocarbon/UIR features plus the 30 micron feature or silicate emission features. The feature near 11.5 microns in the IC 418 spectrum may be due to SiC dust. This is only rarely seen in the planetary nebulae and never in galactic objects in the transition phase between the red giant phase and the planetary nebula phase. The spectrum of IC 418 may have a very weak feature at about 20 microns. There is a small inflection in the continuum at this wavelength. If it is a feature, then it is the same feature as seen in the spectrum shown in the lower right panel. However its very, very weak if present. Most people don't believe that there is a feature there.
One oddity is that a strong SiC feature is seen in some cases in these post-AGB type objects in the Large or Small Magellanic Clouds, but never in the Galaxy. This is not explained. Something is different between the two satellite galaxies and our own galaxy.
An example spectrum of a star we think is in this transition phase (see the proto-planetary nebula page) is given in the lower right panel. This object is newly discovered in the Large Magellanic Cloud but I selected it because it shows all the features seen in carbon-rich proto-planetary nebula spectra. Object IRAS 18095+2704 shown in the previous page is thought to be another such PPN object with oxygen-rich dust. The carbon-rich PPNs show the UIR features, although with somewhat usual properties, the 30 micron feature, and an unidentified transient dust feature called the "21 micron feature". That feature is very odd because it seems to be transient, and is only strong in the AGB to PN transition phase. There are a few possible cases of the feature being seen very weakly in a bare handful of either extreme carbon stars similar to IRAS 21489+5301 or in carbon-rich PNs like IC 418. In these very few cases the feature is so small as to be extremely debatable. Thus the 21 micron feature seems to appear and then disappear over a period of a few hundred years during the PPN evolution, which is very odd behavior. There is no generally accepted carrier identified for the feature. At one time TiC appeared to be a good candidate for the dust grain that produces this feature, but that seems to not work because there should be other features as well that are not observed.
Note that this "21 micron feature" peaks at 20.1 microns, and is not the same as the Cas A "21 micron feature" shown on the previous page. You can see that the features are different in the plot below where I show the Cas A spectrum and the spectrum of one of the post-AGB sources similar to J052520-705007.5. The feature in the spectrum of IRAS 23304+6147 is clearly not the same as the one in the Cas A spectrum. Lamentably, though, they are both called "the 21 micron feature".
up to the top page
"bare" stellar spectra in the mid-infrared
the most common features: silicates and SiC
less common features: extreme carbon stars, the AlO/silicate complex, the "unidentified infrared" features, unusual silicate features
unusual features: mixed chemistry sources, Wolf-Rayet stars, ice bands, featureless carbon-star spectra