Curved Structure of Si by Improving Etching Direction Controllability in Magnetically Guided Metal-Assisted Chemical Etching.

Metal-assisted chemical etching (MACE) is widely used to fabricate micro-/nano-structured Si owing to its simplicity and cost-effectiveness. The technique of magnetically guided MACE, involving MACE with a tri-layer metal catalyst, was developed to improve etching speed as well as to adjust the etching direction using an external magnetic field. However, the controllability of the etching direction diminishes with an increase in the etching dimension, owing to the corrosion of Fe due to the etching solution; this impedes the wider application of this approach for the fabrication of complex micro Si structures. In this study, we modified a tri-layer metal catalyst (Au/Fe/Au), wherein the Fe layer was encapsulated to improve direction controllability; this improved controllability was achieved by protecting Fe against the corrosion caused by the etching solution. We demonstrated curved Si microgroove arrays via magnetically guided MACE with Fe encapsulated in the tri-layer catalyst. Furthermore, the curvature in the curved Si microarrays could be modulated via an external magnetic field, indicating that direction controllability could be maintained even for the magnetically guided MACE of bulk Si. The proposed fabrication method developed for producing curved Si microgroove arrays can be applied to electronic devices and micro-electromechanical systems.

Nanoscale light element identification using machine learning aided STEM-EDS.

Light element identification 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 identification. In this study, we achieved light element identification 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 identification 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-electron energy loss spectroscopy and multicomponent diffusional transformation simulation. The enhancement of SNR in STEM-EDS spectrum images by machine learning algorithms can provide an efficient, economical chemical analysis method to identify light elements at the nanoscale.