Operando study of Ni-22Cr oxidation dynamics with nanoscale resolution in an X-ray photoemission electron microscope

Understanding the initial oxidation pathways on binary Ni-Cr surfaces is import for alloy development and rational design. The early-stage oxidation of Ni-22wt% Cr is studied in-situ and in-operando with synchrotron-based x-ray photoelectron microscopy (XPEEM) and x-ray photoelectron spectroscopy (XPS) at 500 °C and oxygen exposure from 0 to 65 Langmuir. Preferential Cr oxidation is accompanied by Cr depletion in the alloy. Analysis of XPEEM timeseries and hyperspectral datasets presented the combined challenge of large datasets and varying image background. Hyperspectral data was approached using a combination of data dimensionality reduction techniques including principal component analysis (PCA), non-negative matrix approximation (NNMA), cosine similarity (CS), and tiling. These methods deliver spatially resolved chemical identification and extract the evolution of oxide island distribution over time. XPEEM timeseries thresholding shows that chromia islands are embedded in a thinner layer of oxidic Cr components, and larger islands emerge at ∼30 L. Their growth rate and size distribution differs markedly for (104) and (212) surfaces studied here. The layer-plus-island growth mode is reminiscent of a Stranski-Krastanov type growth and increases the oxide layer heterogeneity. This work demonstrates a local-background framework for island identification and the utility of PCA, NNMA, and cosine similarity for analysis of XPEEM datasets.