Pore Network 3D Reconstruction for Clays Using FIB-SEM Images

Abstract The aim of this study is to propose a new approach to explore the 3D evolution of pore properties in a saturated and remoulded clay in relation to mechanical loading. Oedometric loading until the desired stress level was applied to a clay sample. A post-mortem observation using scanning electron microscopy (SEM) coupled with focused ion beam (FIB) was used to obtain the 3D microstructure of a micro-volume of the sample. A novel processing method, incorporating machine learning for pore segmentation was performed to reconstitute the micro-volume. Pore network properties were studied through different parameters like orientation and morphology. The results obtained and the method developed in this study highlight the significant potential of 3D FIB-SEM observations in characterizing the microstructures of saturated clays.