Fig. 2: Data mining of diffraction patterns (DPs) for CSRO detection in CrCoNi. | Nature Communications

Fig. 2: Data mining of diffraction patterns (DPs) for CSRO detection in CrCoNi.

From: Data-driven electron-diffraction approach reveals local short-range ordering in CrCoNi with ordering effects

Fig. 2

Four steps (a–d) are involved. a Data acquisition, energy-filtered diffraction patterns (DPs) in a scan are recorded using an ~1 nm sized electron probe, with a pixelated detector. b The DPs are transformed into difference cepstra, which are then used to form cepstral annular dark-field (ADF) images for nanocluster (NC) detection. c Bragg reflections are removed, using the log method, and diffuse DPs belonging to NCs are averaged and grouped together. d Diffuse DP templates are identified and used to detect the CSRO-strengthened NCs by the correlation method. The diffuse DPs from the same CSRO-strengthened NCs are then averaged, and the process is iterated to yield the refined templates. N is the number of collected DPs (104), M is the number of distinct cluster diffuse DPs (101), and n (n2 = N) is the number of scan points. ADF stands for annular dark field imaging, using the cepstral signals here.

Back to article page