Another candidate for the craziest paper title
The Wolf-Rayet star WR124 as seen by the HST CREDIT: Yves Grosdidier, Anthony Moffat, Gilles Joncas, Agnes Acker, STScI, and NASA
On astro-ph today there is an article titled Pixie Dust: The Silicate Features in the Diffuse Interstellar Medium. Sounds very odd but, honestly, it describes serious research. The two researchers who wrote the paper (J. Chiar and A. Tielens) have looked at examples of a type of star known as Wolf-Rayet stars and used them to investigate the interstellar medium (ISM).
The image on the left shows an example of a Wolf-Rayet, or WR, star known as WR124 seen by the Hubble Space Telescope. This one is about 15,000 light years away in the constellation of Sagitta.
There are actually three types of WR star, classified by their spectra. The spectra of WN stars are dominated by nitrogen, those of WC are dominated by carbon, and WO are oxygen-dominant. There is an interesting list of lines seen in the different spectral types along with example spectra for each class at the CfA.
The researchers who wrote this paper looked at four WC stars in our Galaxy in the infra-red part of the electromagnetic spectrum using the Infrared Space Observatory (ISO). These particular stars are heavily affected by extinction: there is a lot of stuff (gas and dust) in the interstellar medium between them and us which absorbs light coming from objects in the distance. This extinction is a common problem in astronomy (it is why there appear to be dark bands in the Milky Way) and, if not corrected for accurately, can cause false conclusions. The problem is that the required corrections are not always known that accurately. This is where the "pixie dust" referred to in the title comes from. The corrections applied seem to work, even though they are not perfectly constrained or always well understood.
By carefully analysing the spectra seen when looking at these four stars the researchers have been able to determine some of the chemistry and mineralogy of the ISM and so better describe this "pixie dust". It may sound strange to observe a star when tryig to investigate the dust, but it works. You can think of it like using a lighthouse beam to probe fog, if you know the properties of the lamp then the amount of light you can see tells you how thick the fog is. Similarly, astronomers have models of these kinds of stars which tell them what the spectra should look like. This ideal spectrum can be subtracted from the actual data to leave just the spectrum due to the extinction.
The astronomers who carried out this experiment found that the absorption can be described quite well using a model made up of grains of silicates such as pyroxene and olivine. They have produced extinction curves so that other researchers can make use of them when trying to correct for local dust or dust in the Galactic centre. So the pixie dust is not be so mysterious after all.