Pore‐Scale 3D Morphological Modeling and Physical Characterization of Hydrate‐Bearing Sediment Based on Computed Tomography

Abstract

Natural gas hydrate is considered to be a promising future energy resource; therefore, obtaining its physical properties is crucial for evaluating gas production efficiency and developing reasonable exploitation strategies. In this study, we present a novel pore-scale 3D morphological modeling algorithm considering various saturation and occurrences (cementing and pore-filling) of hydrate in sediments. To evaluate the performance of the presented algorithm, 14 hydrate-bearing sediment models were constructed based on X-ray computed tomographic images of a consolidated sandy specimen under an effective stress of 3 MPa. Morphologically, the new algorithm generated pore-scale hydrate occurrences coincide with published morphological behaviors of hydrate observed via computed tomography. The tortuosity and fractal dimension of pore spaces of these models were then characterized. The pore networks are also extracted, based on which the evolution of the pore characteristics including the distributions of pore radius, throat radius, throat length, and the coordination number were investigated. Using these generated models, simulations of permeability, thermal conductivity, and electrical conductivity were conducted to evaluate the influences of hydrate saturation and occurrence. These results are validated against published data, demonstrating that this algorithm could be an effective way to construct digital hydrate-bearing sediment models using a single set of computed tomographic images of a hydrate-free specimen. This new method can also be significant for the physical evaluation of natural cores in which hydrate has dissociated during core recovering and transferring.

Mechanical behaviors of hydrate-bearing sediment with different cementation spatial distributions at microscales

“…In addition, it has to be noted that although the median radii of the pore space in the two specimens are around 100 μm, the hydrate particle could be even larger than 10 mm 3 . This is because the hydrate particle is not an isolated small particle only existing in a single pore as the idealized conceptual model ( Clennell et al., 1999 ; Wu et al., 2020c ), but a cluster existing in multiple pores and throats at the same time.…”

Influence of Particle Size Distribution on the Physical Characteristics of Pore-Filling Hydrate-Bearing Sediment

“…A novel pore-scale 3D morphological modeling algorithm for hydrate generation proposed by Wu et al [24] was used in this study. As illustrated in Figure 2(b), the spe-cific steps are as follows: Firstly, conducting CT scanning for the host sediment.…”

“…Finally, assembling the sand phase, hydrate phase, and corresponding pore phase for the image stacks of the pore-filling HBS. The more detailed introduction could be found in the morphological algorithm proposed by Wu et al [24]. In addition, the bulk variation since the hydrate generation is not considered, this is because Lee et al [27] found that during hydrate generation, the HBS's volumetric strain is only around 0.1% which could be insignificant.…”

“…After 360°rotations of a specimen, a series of two-dimensional slice images were collected and a median filter was utilized to process these two-dimensional images. Then, we utilized the inspeXio (Shimadzu Co., Ltd, Japan) to reconstruct these two-dimensional CT images into a three-dimensional model and chose the middle area as the representative elemental volume (REV) according to the operation of Wu et al [24]. The REV is a cube consisting of 500 × 500 × 500 voxels, and the voxel size is 21 μm/voxel.…”

“…This part simulated the formation of hydrates under the gas saturation condition. And following Wu et al [24], a partial differential equation based on Fourier's law considering multiple phases and multiscale homogenization theory was employed as follows: 8 Geofluids…”