A Novel Method for Hydrogen Permeation Detection Using Reflective Microscopy and Machine Learning

Localized detection of hydrogen permeation in steel membranes is crucial for practical applications but remains challenging. We present a reflective microscopy (RM) approach combined with machine learning (ML)-driven image analysis to address this issue. Hydrogen permeation in press-hardened steel alters the Fe(II)/Fe(III) ratio in the surface layer, affecting reflected light intensity due to changes in the refractive index. Although initially near the optical detection limit, this subtle contrast was enhanced using time-series analysis with UMAP and K-means clustering. The detected regions corresponded to hydrogen-affected zones confirmed by Scanning Kelvin Probe measurements. This work establishes a novel RM-ML framework for detecting and quantifying hydrogen permeation, offering a promising and accessible alternative to conventional methods for localized analysis of metallic membranes.