An infrared inventory of the mineralogy of dust in active galaxies
Thesis Supervisor at ESO (as of 01.11.2018): Ciska Kemper
This project aims to do an extensive investigation of the dust mineralogy in active galaxies. Mid-infrared spectroscopy, between ~5
and ~100 μm shows the resonances due to various astrophysically relevant materials, with the highest concentration of features between
8 and 45 μm. In the interstellar medium of the Milky Way can be well-fitted with the observationally derived "astronomical silicate"
optical properties (Draine & Lee 1984, Draine & Li 2007), however, it is becoming apparent that the silicates seen in the tori around active galactic nuclei cannot be fitted with these optical properties, and furthermore that the silicates in starburst galaxies can be considerably crystalline, thus demonstrating that dust properties are not universal.
A number of studies have attempted to resolve the mineralogy of AGN environments using laboratory optical properties (Jaffe et al. 2004; Markwick-Kemper et al. 2007; Nikutta et al. 2009; Köhler & Li 2010; Smith et al. 2010; Dopita et al. 2011; Xie et al. 2014; Srinivasan et al. 2017), as well as establishing the presence of crystalline silicates in the global ISM for a subset of galaxies (Spoon et al. 2006; Willett et al. 2011; Aller et al. 2012; Stierwalt et al. 2014). However, the information about the dust mineralogy derived from such fits is still anecdotal. Most studies have targeted small samples or even a single object, with virtually every study approaching their sample with a different scientific question, a different physical model, and a different set of minerals and corresponding optical constants to perform the fit. Due to the highly non-linear nature of dust radiative transfer, it is difficult to establish the goodness-of-fit and derive the confidence limits on the derived physical parameters and mineralogical properties. Thus, a systematic comparison between dust mineralogy and physical parameters in and evolutionary status of the galaxies is lacking.
We have developed a python tool called AMPERE which, through Bayesian inference, will fit spectral and photometric data to dust radiative transfer models for galaxies, such as SKIRT. AMPERE allows for a systematic automated analysis of large sets of spectroscopic data, overcoming the issues raised above. A wealth of data is available to apply AMPERE to, and we will initially work with archival data to execute this project. In particular, the Spitzer archive contains hundreds of 5–38 μm spectra of external galaxies suitable for this analysis. The CASSIS database (Lebouteiller et al. 2011; Lebouteiller et al. 2015) contains the optimally reduced spectra of all the objects targeted with Spitzer. Specifically, it contains high-and low-resolution spectra of 14,405 distinct targets, of which in excess of 5,000 objects are (lines of sight in) external galaxies, at all redshifts, of which about half are AGN, quasars, ULIRGS and starburst galaxies.
In this project, we will carry out a systematic and statistical survey on this Spitzer data set, and put constraints on the mineralogy of
a large number of objects. We will correlate the derived mineralogy to a variety of galaxy properties with the goal of understanding dust formation and deriving the physical properties in the circumnuclear tori and interstellar medium of these galaxies in mind.
In addition, in this project we aim to obtain data with the JWST, by responding to the call for proposals, but we are also in collaboration with relevant JWST teams, in particular the GOODS-JWST team. JWST is expected to launch in 2021,
and the MIRI instrument can be used to resolve the mineralogy.