Calculation of the volume of nano- and microstructures formed on the surface of PbTe during ion-plasma treatment using machine learning

The formation of porous Pb nano- and microstructures on the surface of PbTe films during low-energy ion-plasma treatment has been studied using scanning electron microscopy (SEM) combined with machine learning-based image analysis. PbTe epitaxial films with (111) crystallographic orientation were exposed to argon plasma at an ion energy of approximately 25 eV for varying durations (60−240 s). SEM imaging at a tilt angle of 70° enabled three-dimensional size estimation of the formed structures, which, together with automated image processing using the DLgram⁰¹ deep learning service, allowed for precise calculation of particle number, area, height, and volume. In this paper, a comparative analysis of the parameters of the Pb structure on the surface of lead telluride films with the orientation (111)
and PbTe single crystals with the orientation (100) is carried out. This study demonstrates the effectiveness of machine learning for quantitative analysis of surface nanostructures after lowenergy argon plasma treatment of PbTe surface.