Dust grains play a major role in the physics of the interstellar medium. They absorb and reemit in the infrared most of the radiated stellar power. Moreover, they are responsible for the gas heating in photodissociation regions (PDR) and serve as catalysts of numerous chemical reactions. Their properties (chemical composition, size distribution, etc.) are however currently poorly known. These uncertainties put caution on numerous aspects of our knowledge of the interstellar medium: mass estimates, PDR models, unreddening, etc. Refining our comprehension of dust is crucial to understand the life cycle of interstellar matter and its effect on galaxy evolution.
One of the approaches, to tackle these open questions, consists in studying the way the observed grain properties vary with the physical conditions they experience. The PhD thesis we propose is aimed at focussing on the properties of the smallest grains (with a radius smaller than ?10 nm) and of polycyclic aromatic hydrocarbons (PAH). These interstellar medium components radiates out of equilibrium in the mid-infrared (?5-40 µm), and are the carriers of numerous resonance bands. This study will focus on several nearby galaxies, including the Magellanic clouds. The interest of studying nearby galaxies rather than the interstellar medium of our own galaxy resides in the diversity of the physical conditions of the environments we can access (metallicity, stellar radiation field intensity, etc.).
Numerous studies have already been published on this subject. However, most of them were superficial. There remains many aspects to study: identifying and physically modeling several bands of solids in star forming regions, and the correlation of the properties of the main PAH bands with the physical conditions diagnosed thanks to the new Herschel data.
The thesis will have several aspects. First, the analysis of mid-infrared spectra, obtained with the satellite Spitzer. Most of these spectra are already reduced. Most of this first step will consist in critically selecting the spectra to study, and homogenizing the data. Then, quantifying the physical components, which is not trivial, will be performed in a sophisticated manner. We propose that the student will develop a hierarchical bayesian model for spectral decomposition, which will allow him a precise quantification of the uncertainties and of the correlations between physical parameters. This new tools and its meticulous application to the data is the guaranty of a precise and original interpretation of the physical processes in the studied regions.
This thematics is particularly relevant for planning the scientific objectives of the James Webb Space Telescope (JWST), which should be launched in 2018.