Please use this identifier to cite or link to this item:
|Title:||PERMEABILITY AND PORE STRUCTURE OF GRANULAR MATERIALS|
|Abstract:||The influence of the microstructure on the permeability of porous media is very important for various engineering problems in different engineering disciplines. The challenges for research in this area include the characterization and the quantification of microstructure as well as proper involvement of such microstructure measures in mathematical models with clear physical meaning. This study focuses on the reconstruction of pore structure of porous materials, the quantitative description of pore structure using second-rank fabric tensors, as well as the determination of permeability as a function of microstructural measures via different approaches, including laboratory tests, theoretical analysis and numerical simulations. This research is centred on the quantification of pore structure of porous materials. A procedure is established to reconstruct the microstructure of a porous material using 3D CT-scan images. A new algorithm is developed to quantify the internal structure based on the 3D CT-scan images. More specifically, two fabric descriptors, Mean Intercept Length and Areal Pore Size, are employed to construct fabric tensors as measures for the internal structure. The developed procedure and algorithm are verified for difference scenarios and then used to determine the internal structures of different materials. An analytical model is developed to describe flow in porous medium with internal structures using the technique of homogenization. With the fluid-solid interaction taking into account on the particle level, the permeability tensor of a porous medium is derived from the macroscopic momentum balance equation of the fluid. The microstructure of the porous media is taken into account using the fabric tensors (based on both the Mean Intercept Length and Areal Pore Size) obtained from 3D microstructure reconstructions. When interpreted in the classical meaning, the two tensors describe the directional dependency of the tortuosity and hydraulic diameter, respectively. The well-known Kozeny-Carmman model is recovered for isotropic materials. In addition to the directional variation of mean pore sizes, the variation of pore size distribution in different directions also affect the permeability of porous materials. Pipe Network Models are used to explore the influence of pore size variation on the permeability of granular materials. A new algorithm is developed to compute the pore size distribution and its variation in different directions in porous media using the CT-scan images. When taking into account the directional variation of pore size distributions, Pipe Network Models can be used to determine the principal permeability of an anisotropic porous material. The success of this approach highly depends on reliable characterization of pore structures To verify the theoretical and numerical approaches developed in this research, laboratory tests are carried out to obtain the permeability of different materials. To determine the anisotropy of permeability of granular materials with internal structure, a new permeameter is developed to allow the hydraulic conductivity of cohesionless materials to be measured when flow takes place in any direction relative to bedding plane of the specimens. Materials with different type of particle shape and particle size distributions are examined using this permeameter. Good agreements are obtained from comparisons of laboratory test results with theoretical models and numerical simulations.|
|Appears in Collections:||Open Access Dissertations and Theses|
Files in This Item:
|Dissertation final.pdf||4.81 MB||Adobe PDF||View/Open|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.