An sizable fraction of present-day astrophysics is based on physical models of stellar atmospheres, which allow us to interpret observations of planets, stars, nebulae and galaxies. 

Traditionally, these models adopt a series of simplifying assumptions, including Local Thermodynamical Equilibrium (LTE), implying that matter and radiation are in thermal equilibrium in each and every layer of the atmosphere. Recently, progress in our understanding of processes for the exchange of energy through collisions between hydrogen atoms and metal ions have opened the possibility to relax the LTE hypothesis and make more accurate computations.  These detailed calculations are a must to break the precision barrier that limits the analysis of stellar spectra (and other sources), currently at about 20%, reducing the uncertainties to 1-2%. The goal of this project is to build iron models including the relevant physical processes, in particular inelastic colllisions with hydrogen atoms, and apply them to interpret astronomical observations.