Data Mining for collecting previously unavailable information
We developed a novel data mining technique for collecting previously unavailable information about the energy landscape and friction during the glide of dislocations in a High-Entropy Alloy (HEA) as part of a collaboration between an in-situ TEM group and a materials data science group. More specifically, we use dislocations as probes to systematically scan the "energy landscape" of a CoCrFeMnNi alloy. Our scanning method uses a unique data-mining strategy that is specifically designed for in-situ TEM observations and can perform spatio-temporal coarse-graining of dislocation line properties. It provides the opportunity to perform ensemble averages of many time snapshots during dislocation motion. With this method, we can quantitatively investigate the influence of pinning points on dislocation glide behavior and find that (i) the pinning point strength evolves when dislocations cross, and (ii) the pinning point location is shifted in a direction close to the Burgers vector direction.