Figure 3
Summary of hyperspectral acquisition strategy and subsequent post-treatment analysis for a Cu K-edge dataset recorded for a sample prepared from the juxtaposition of a pellet of CuS (on the left) and a pellet of Cu2O (on the right). (a) At the same position of the camera sensor, hyperspectral cubes of a reference Cu metallic foil, of the flat-field image for I0 and of the sample of interest are recorded successively. (b) Using the Beer–Lambert law, the spectra of each pixel of the reference image (blue cross) and of the sample image (green and red crosses) can be calculated. (c) After pixel binning, the first derivatives of the spectra of the Cu reference foil hyperspectral cube are calculated and realigned to the first derivative (black curve) of the quick-EXAFS spectrum of the reference foil absolute energy calibrated. An energy-shift map is therefore obtained consisting at each pixel of the ΔE offset value used for absolute energy scale alignment. (d) The energy shift map is used for energy alignment at each pixel of the spectra of the sample hyperspectral cubes, performed after pixel binning similar to the one used for the reference and before normalization. (e) Outcomes of the sequence of Jupyter notebooks visualized along vertical lines of pixels in the CuS and Cu2O areas located in the edge jump map (white lines) for which values are related to the color scale displayed at the top of the map. |