Extraction of physically-realistic pore network properties from three-dimensional synchrotron microtomography images of unconsolidated porous media

Abstract

Algorithms were implemented to obtain high resolution three-dimensional images using synchrotron microtomography. Morphological algorithms were developed to extract physically-realistic pore-network structure from unconsolidated porous media systems imaged using synchrotron microtomography. The structure can be used to correlate pore-scale phenomena with the pore structure and can also be incorporated into a pore-network model to verify existing models, understand, or predict transport and flow processes and phenomena in complex porous media systems. The algorithms are based on the three-dimensional skeletonization of the pore space in the form of nodes connected to paths. Dilation algorithms were developed to generate inscribed spheres on the nodes and paths of the medial axis to represent pore-bodies and pore-throats of the network, respectively. Pore-network structure is captured by three-dimensional spatial distribution of pore-bodies and pore-throats, pore-body and pore-throat size distributions, and the connectivity. Theoretical packings were used to verify the algorithms. Systems of glass bead and natural sand were used in this study to investigate the applicability of the algorithms. Additionally, porosity, specific surface area, and representative elementary volume (REV) analysis of porosity were calculated. The impact of resolution was investigated using perfect glass bead and natural sand systems. Finally, semivariograms and integral scale concepts were used as a tool to investigate the spatial correlation of the network. Results showed that microtomography is an effective tool to provide quantitative analysis of three-dimensional systems. The quality of the datasets depends on photon energy, photon flux, size and type of the sample, and the number of projections. The resolution has a significant impact on the construction of the medial axis and extraction of pore network parameters. This impact varies in its significance based on the system and the properties being calculated. Results highlighted the difficulty of creating a unique network from a complex, continuum pore space. Results showed that the algorithms developed are general in use and can be applied to any three-dimensional unconsolidated porous media system. Spatial correlation results showed that systems have different correlation behavior; therefore, it might be not correct if a correlation model is assigned a priori into a pore-network model.