micro-XRCT datasets of stochastically reconstructed 3D porous media micromodels manufactured by additive manufacturing

This dataset contains micro X-ray Computed Tomography (micro-XRCT) scan data sets (projection, reconstructed, and binarized images) of 3D porous media micromodels manufactured by additive manufacturing using the Material Jetting (MJ) method. The micromodel geometries were designed using the stochastic model proposed by Quiblie (1984), Adler et al. (1990), and Hyman et al. (2014). For this study, four samples were fabricated, all possessing the same porosity of 0.45, but with varying correlation lengths (id: 15, 25, 35, and 45) that define the pore size distribution. The four cylindrical samples have a length of 50 mm and a diameter of 16 mm. For more details, see the related publication Lee et al. (2023) The samples were completely scanned. Due to the limitation of the field of view at the required resolution, first the bottom part followed by the top part of the respective sample was scanned. Reconstruction was carried out separately for the top and bottom scans. Merging of the bottom and top parts was performed based on the reconstructed images. During the merging process, duplicated slices were naturally eliminated. The grayscale images obtained after the reconstruction and merging processes underwent segmentation, distinguishing between solid phase and pore space regions based on intensity values. Subsequently, the misclassified voxels resulting from the inherent noise in the micro-XRCT data were adjusted accordingly by assessing the connectivity between pixels (isolated pixels were reclassified to the neighboring class). Simulations using the lattice Boltzmann method to determine the permeability of the scanned mircomodels can be found in the related dataset Lee et al. (2023).