Characterization of the pore structure in Chinese anthracite coal using FIB-SEM tomography and deep learning-based segmentation

Abstract

Coalbed methane (CBM) is an unconventional natural gas that possesses significant impacts on energy supply, mining safety, and environmental conservation. CBM is primarily stored within the pores of coal, highlighting the significance of pore structures for both methane storage and migration. However, comprehensively understanding the intricate pore structures in coal pose challenges. This study employed focused ion beam-scanning electron microscopy (FIB-SEM) tomography and deep learning-based segmentation to characterize the pore structures within a Chinese anthracite sample. The obtained pore structures exhibited a considerable degree of disconnection, comprising numerous separate pore components. Isolated pores prevailed in number, while connected pores dominated in surface area and pore volume. Mesopores (100–1000 nm) contributed the most to pore number, surface area, and pore volume. Pore size distribution analysis revealed distinct patterns among different pore structure properties, with pore number exhibiting an intensive distribution while surface area and pore volume displaying dispersed distributions. Pore structure connectivities exhibited a hierarchical nature and held distinct meanings at the levels of pore, pore component, and pore network. The pore structure characteristics observed in this study have implications for primary CBM recovery, emphasizing the necessity to improve connectivity between pore components and fractures to enhance production rates and recoverability.