To decrease vehicle masses (reduction of CO2 emission), steels with improved formability are developed using alloying elements such as aluminium, silicon or manganese. During the recrystallization annealing of steel strips at ~ 1070 K under reductive atmosphere (N2 + H2), those elements are prone enough to oxidation leading to oxygen-induced segregation at the steel surface. A strong industrial concern is that the bad wetting of these oxides by zinc degrades the efficiency of galvanization, which is the historical method to prevent corrosion whose principle is that zinc being less noble to iron, it sacrificially corrodes to protect the substrate steel. Therefore, galvanization faces a new paradigm since it switches from what is close to a reactive interface with bare iron to a high-energy interface with wide bandgap oxides. To improve the coatability of steel by zinc, a strategy consists in trying to control upstream the surface oxidation of the steel. A detailed analysis of alloyed steel and environment being far out of grasp, the present SURFOX project, focuses on the oxidation of Fe-Al alloy, a common alternative to Al-alloyed steel. It involves two partners, University Pierre and Marie Curie and ArcelorMittal Maizières Research represented by two groups, “Low-dimensional oxides” (Paris group) and “Physical Chemistry and Surfaces” (Maizières group). Three contributions are foreseen, experiments on single crystals and numerical simulation from Paris group, experiments on polycrystalline FeAl plates from Maizières group. It has been chosen to study Fe-15at.%Al samples since at annealing temperature (around 1 070 K) the alloy matrix corresponds to the ferrite phase as in the case of industrial Al-alloyed grade. There is some suggestion that local arrangements in supported aluminium oxide layers mildly depend on the nature and orientation of the support and of the Al-content of the alloy, although long-range structures and oxidation kinetics are different. This suggests to define common fingerprints to all orientations prior to examining on this basis the parameters (orientation, environment) which affect the oxidation of FeAl samples. • Toward common fingerprints of all Fe-15at.%Al oxidized surfaces - By analysis relying on surface science methods in ultra-high vacuum conditions, including electron and ion spectroscopies, electron diffraction and near-field microscopy, the objective of Paris experimental group is (i) to determine the structures of the Al oxide layers formed at the Fe-15at.%Al(110), (100) and (111) surfaces, considered herein as reference data, and (ii) to explore the exciting issue of the representation of all structures by similar short-range arrangements. These will be defined with the support of numerical simulations. The structure of the oxide layers formed on Fe-15at.%Al polycrystalline plates, second series of reference data as studied by surface and material science methods (Maizières group) including electron spectroscopies and microscopies, will then be compared to those fingerprints in order to achieve an unified picture. • Orientation and environment-dependent oxidation – On the basis of this picture and by combining numerical simulation (Paris group) and experiments in various atmospheres (O2, H2O, H2O/H2, Maizières and Paris group), it is foreseen to examine the rationale of (i) the way the oxidation during recrystallization annealing depends on the dew point of the atmosphere in which the process is operated, in particular to lead to external and internal oxidation, and of (ii) the orientation-dependent oxidation anisotropy of the ferrite grains (kinetics, stability) which is of paramount importance in the industrial context. The challenging issue is to set up a unique structural model for oxide adlayers formed at the surface of FeAl alloys and to use it to get insights into the behaviour of these oxide layers as a function of both environment and crystallographic orientations