Thesis Topic: Silicates in starburst galaxies


Thesis Supervisor: Francisca Kemper


Among the most dusty objects in the universe are the (ultra)luminous infrared galaxies, a class of objects dominated by their infrared emission, due to significant reprocessing of starlight by dust grains, which can be seen up to high redshifts in the far-infrared and submillimeter. The members of this class commonly have AGN or show starburst activity, and often at the same time. In comparison to the Milky Way and other quiescent galaxies, active and starbursting galaxies exhibit a more intense radiation field, which heats the grains; dust grains are more exposed to destructive interstellar shocks; and grains also see a higher cosmic ray fluence, which may damage the lattice structure of crystalline minerals. We know that this has its ramifications on the dust properties, as the mid-infrared spectrum shows non-standard silicate dust (Sturm et al. 2005), and a particularly strong crystalline silicate signature in some galaxies (Spoon et al. 2006), while the ISM in the Milky Way contains little or no crystalline silicates, and shows a markedly different amorphous silicate profile (Kemper et al. 2004). Exactly how the physical conditions give rise to such differences in dust mineralogy compared between galaxies is unclear, and, indeed, the observed mineralogy (more crystallinity of silicates in active and starbursting galaxies, rather than less, as one would expect due to the increased cosmic-ray flux) seems contradictory. 

Mid-infrared spectroscopy provides an excellent tool to study the mineralogy of interstellar and circumstellar dust, and it is with such observations that the dust composition of the main dust producers in galaxies, evolved stars, has been analysed. Focusing on the lattice structure of silicates, it is known that only oxygen-rich AGB stars with the highest mass loss rates actually show crystalline silicates in their spectra (Sylvester et al. 1999), although this might be due to an observational effect, and that a crystallinity of 10% over all mass-loss rates is consistent with the observations (Kemper et al. 2001). These findings should be contrasted with the silicate composition in the diffuse interstellar medium, where I found a degree of crystallinity < 2% (Kemper et al. 2004). Clearly, amorphization in the ISM is ongoing, and an amorphization time scale in the ISM of our Milky Way of at most 90 million years, presumably due to ion bombardment, can be estimated. 

One of the most interesting Spitzer results revealed the presence of crystalline silicates in absorption in a sample of 12 out of 77 starburst galaxies (Spoon et al. 2006), with crystallinities in the range from 6-13 %. The high levels of crystallinity can be understood as being due to freshly produced dust that has not yet been amorphized by cosmic ray hits, which may be possible under certain conditions (Kemper et al. 2011). Crystalline silicates in the ISM of external studies were later also reported by others (Willett et al. 2011; Aller et al. 2012; Stierwalt et al. 2014), but these studies did not measure the crystalline fraction of the silicates. 

With the advent of the James Webb Space Telescope (JWST), which will be launched later in 2021, a deluge of spectroscopic data of external galaxies is to be expected. JWST's MIRI instrument has a wavelength coverage of 5-28 μm, which, for modestly redshifted galaxies encompasses the prominent 10 μm Si-O stretching mode in silicates, and will therefore be a great tool to study the mineralogy of nearby and modestly red-shifted galaxies. Simultaneously, the Spitzer archive remains largely unexplored on this particular topic. 

This thesis project will focus on further developing the dust evolution model for silicates in starforming galaxies (Kemper et al. 2011), and comparing the predictions of this model with observational results that can be obtained from newly observed JWST spectroscopy and existing archival Spitzer/IRS spectroscopy.