To achieve an understanding of alloy oxidation at room temperature, AES sputter depth profiling was used to disclose the elemental distribution, the existence of a sequence of different oxide layers and their respective thickness. A polycrystalline, homogenized, and polished sample of Ni21 at % Cr-12 at % Fe was oxidized in air at room temperature and atmospheric pressure for 20 min after sputter cleaning. Additionally, a polycrystalline sample of Fe18 a t% Cr was passivated for 24 h in deaerated 0.5 tool/1 H2SO 4 at 1 VN~E. Then, photoelectron spectra (2p of the alloy elements and Ols) were taken using an Mg K~ X-ray source to characterize the chemical state of the oxides and to determine the oxide thickness. The depth profile analysis was performed with a rastered Ar + ion beam of 3 keV with 34 ~ incidence angle to the surface normal and an Ar pressure of 4104 Pa in the analysis chamber. After each sputtering cycle the Auger peaks of the elements Ni (848 eV), Cr (529 eV), Fe (650 eV) and O (510 eV) were recorded in the derivative mode with 2 Vpp modulation voltage using a double pass cylindrical mirror analyser with 0.6% energy resolution. Additionally, the M23VV transitions of the alloy elements were measured in the low energy region between 20 eV and 70 eV. For the quantitative evaluation of the composition in the oxide layers and the oxide thicknesses the sequential layer sputtering model (SLS) [ / 3] was applied to the measured intensity/sputtering time profiles. This model assumes a monoatomic layer structure and a Poisson distribution as resolution function in combination with the exponential escape depth correction and is appropriate for very thin layers if the information depth of the electrons used for the analysis is comparable or larger than the thickness of the film [1]. It is suitable for an automated data evaluation by computers. The sputtering time scale is first converted into a depth scale by calibration with the total oxide layer thickness obtained from XPS measurements [4, 5]. Assuming constant oxide density and sputtering rate and neglecting a possible preferential sputtering, an intensity corresponding to each removed monolayer equivalent is obtained by a spline interpolation to the measured intensities. The functional dependence of these elemental intensities on the unknown concentrations in each atomic layer is given by Eq. (13) in ref. [6] which can be modified into a linear equation system