Nanoscale light element identification using machine learning aided STEM-EDS

Abstract

Light element identifcation is necessary in materials research to obtain detailed insight into various material properties. However, reported techniques, such as scanning transmission electron microscopy (STEM)-energy dispersive X-ray spectroscopy (EDS) have inadequate detection limits, which impairs identifcation. In this study, we achieved light element identifcation with nanoscale spatial resolution in a multi-component metal alloy through unsupervised machine learning algorithms of singular value decomposition (SVD) and independent component analysis (ICA). Improvement of the signal-to-noise ratio (SNR) in the STEM-EDS spectrum images was achieved by combining SVD and ICA, leading to the identifcation of a nanoscale N-depleted region that was not observed in as-measured STEM-EDS. Additionally, the formation of the nanoscale N-depleted region was validated using STEM&#8211

electron energy loss spectroscopy and multicomponent difusional transformation simulation. The enhancement of SNR in STEM-EDS spectrum images by machine learning algorithms can provide an efcient, economical chemical analysis method to identify light elements at the nanoscale.