In recent years, advancements in microcomputed tomography (microCT) imaging technology and image processing have significantly enhanced our understanding of the internal structure of rock cores and the distribution of fluids during multiphase flow. Herein, we use microCT imaging to explore the impact of wettability on fluid flow in a sandstone rock at the pore scale in combination with relative permeability measurements. Steady- state experiments were conducted by the coinjection of decane and water into a sandstone core under different fractional flows (F-w = 0, 0.25, 0.5, 0.75, and 1) and wetting conditions. At each stabilized fractional flow, the system was imaged with microCT at a resolution of 6.7 & micro;m under dynamic flow conditions. The sandstone core is initially tested in a clean state (i.e., water- wet condition), followed by aging in crude oil for 2 weeks at 90 degrees C to create an aged state (i.e., oil- wet/mixed- wet condition). For the aged- state core, it was observed that the oil/water interface was dynamically changing under so- called steady- state conditions while the clean- state core provided less dynamic changes. Consequently, the aged- state core was also imaged under static conditions to capture fluid/fluid interfaces and common lines. Based on the scanned images, parameters such as contact angle, curvature, and pore occupancy were calculated, comparing differences between the clean- state and aged- state conditions. Furthermore, relative permeabilities were measured to analyze the flow characteristics at the continuum scale, providing a comprehensive understanding of pore- scale mechanisms linked to relative permeability behavior.

Cyclic drainage and imbibition processes in porous media are characterized by the saturation state, flow pathway, and hysteresis. These processes play a crucial role in applications such as geological storage of carbon dioxide and hydrogen. While classical empirical hysteresis models have been shown to adequately describe bounding behaviour (primary drainage and imbibition), their accuracy in capturing the scanning and intermediate saturation regions (where most field processes occur) is not well understood. In this work, we investigate the impact of hysteresis and trapping at intermediate saturations during multiple cycles of drainage and imbibition in a water-wet and mixed-wet Bentheimer sandstone. Using fast 4D micro-CT imaging, we characterise the dynamic evolution of topological (Euler characteristics), geometric (interfacial area) and petrophysical (relative permeability) measurements. For both wetting conditions, across all measures, we find that hysteresis is present in the primary bounding cycle, but subsequent cycles are fully reversible along the same hysteretic pathway. Yet, we see clear differences in the trapping extent in the water-wet and mixed-wet samples, indicating that if the primary hysteresis behaviour is well characterised, subsequent cyclic behaviour is predictable. Finally, we show that the relative permeability hysteresis extent is well predicted by changes in the specific interfacial area, highlighting the key role of geometric hysteresis in multiphase flow.

In coal seam gas reservoirs, reducing capillary pressure and enhancing gas relative permeability play important roles in optimising gas production. This study expands on our earlier work (Xi et al. 2025), where we identified the potential of an enzyme-based surfactant for coal seam gas (CSG) recovery, but that study was limited to relative permeability assessments using helium. Here, we present a comprehensive evaluation of its mechanism and effectiveness in enhancing CSG recovery. Captive bubble tests first reveal an apparent critical micelle concentration (CMC) of 1.2 wt.%, achieving capillary pressure reductions of 22% for the CH4-water system, with minimal temperature sensitivity (±10% at CMC) across reservoir-relevant conditions (18–50 °C). In addition, this enzyme-based surfactant decreases CH4-water IFT by 28.5% (67.0→47.9 mN/m), alongside a contact angle reduction of up to 20%. Compared with synthetic surfactants, the enzyme-based surfactant emerges as a natural and degradable alternative, surpassing cationic and non-ionic surfactants in capillary pressure reduction. Finally, its influence on coal relative permeability was investigated. Micro-CT images of the coal sample highlight an 82% decline in trapped gas saturation (17%→3%) during imbibition and a 29% reduction in residual water saturation (73%→52%) during drainage. These enhancements lead to better fluid flow, with the relative permeability of methane increasing by 65% (0.4→ 0.6) and the cumulative water production increasing by 78% (0.3→0.5 PV). These results indicate that the observed improvement in relative permeability is mainly driven by capillary-pressure reduction: although the tested enzyme-based surfactant treatment makes the coal surface more hydrophilic, the simultaneous decrease in methane–water IFT produces an overall reduction in capillary pressure. The comparative performance further confirms that natural surfactants can be effective alternatives to conventional synthetic agents for enhancing CSG production while better accounting for potential environmental impacts. Moreover, the integrated workflow, combining gas-specific captive-bubble measurements with core flooding under reservoir conditions and micro-CT visualisation, provides a general framework for evaluating surfactants under conditions that are more representative than standard ambient, air-based tests, while directly linking capillary-pressure reduction to shifts in gas–water relative permeability.

Multiphase flow in porous media is an extreme case in colloid and interface science. The large surface area amplifies solid-fluid interactions and the complex pore space causes a wide range of flow regimes with rich spatio-temporal dynamics posing a major challenge for deriving transport equations. Historically, macroscopic two-phase flow is described through phenomenological extensions of Darcy's law, which - besides many other shortcomings and inconsistencies - covers strictly only the connected pathway flow regime at very low flow rates while regimes with moving interfaces and associated topological changes are entirely implicit. Developing a description for all flow regimes by upscaling from pore to Darcy scale represents a long-standing challenge. Over the past decades, the field advanced by introducing thermodynamic approaches, geometric state variables for capillarity and capturing non-equilibrium effects. Experimental insights, enabled by advances in pore-scale imaging and modeling, has motivated several novel recent approaches which inherently include fluctuations and intermittency, and thereby avoid previous limiting assumptions. They cover the physics of the three dominant flow regimes: (I) the capillary-dominated regime, consisting of connected pathway flow with capillary fluctuations is covered by the space-time averaging approach and by the extended nonequilibrium thermodynamic theory (NET), resulting in linear laws; (II) the nonlinear flow regime, where capillary states become increasingly accessible by viscous mobilization leading to ganglion dynamics and intermittency, is described by the statistical thermodynamics approach; (III) the viscous limit consisting of drop-traffic, is described by the NET approach, which utilizes the fluctuation-dissipation theorem and Onsager reciprocal relationships leading again to a linear law, or the statistical thermodynamics approach. Most applications reside in regime I which is the most complex and least intuitive because it is a "frozen state". A better starting point is regime III which is from the perspective of dynamics, and then approaching successively regime II and I. We conclude with open questions and invite to contribute steering the theoretical advances towards application. The most immediate is using the co-moving velocity, which utilizes inherent symmetries in the 2-phase Darcy equations, to constrain the functional form of relative permeability and thereby simplify measurement protocols. The choice of state variables and the statistical thermodynamics approach that establishes relationships between them can be used to replace empirical hysteresis models. Grounding transport laws in thermodynamic concepts opens new possibilities for describing coupled transport phenomena in many relevant applications.

The role of phase topology in hysteresis during fluid injection and withdrawal in porous media is not fully understood. We address this by providing experimental and theoretical evidence on three key findings. (1) The topological evolution of the nonwetting fluid is distinct from the capillary pressure and the specific interfacial area, as shown by experiments and a generalized model. (2) Saturation paths with identical capillary pressure and interfacial area show different topologies, revealing insights into energy dissipation and phase connectivity. (3) The topological evolution of the nonwetting phase follows predictable trajectories captured by a piecewise non-linear model. These findings offer practical implications for optimizing subsurface hydrogen and carbon dioxide storage systems and provide a novel approach to study complex systems with topological singularities.

This paper introduces a workflow for creating and analysing an engineered dual-porosity system, similar to those in copper heap leaching, by synthesising micro-porous chalcopyrite-glass beads. These beads are designed to match the size and shape of single-porosity glass beads. X-ray micro-CT imaging is utilised alongside mass balance measurements on comparative irrigation experiments in the dual and single-porosity systems to investigate the complex fluid dynamics governing the interaction between fluid flow in the micro-porous (intraparticle) and macroscopic (interparticle) domains. Results demonstrate that the dual porosity system, with 20% micro-porosity and 46% macro-porosity, retains over twice the liquid volume compared to the single porosity system, with the same macro-porosity and tortuosity. An increased macroscopic flow connectivity and liquid content is observed in the dual porosity system, due to lateral flow within the micro-pores that enhance surface area and connectivity at the bead contact points. A considerable amount of liquid is retained in the micro-pores through capillary forces, which impacts leaching performance in a large scale system. Overall, this imaging-based methodology and workflow provides a robust framework for designing and analysing engineered dual-porosity systems found in geosciences, chemical engineering, and hydrometallurgy, enabling improved prediction and optimisation of reactive transport and resource recovery processes in complex porous media.

Three-dimensional porous media images are crucial in subsurface characterization due to their detailed microstructural descriptions. However, imaging techniques often yield a limited number of small-sized images that are often incomplete due to challenges in sample handling and acquisition. This in turn restricts researchers' access to comprehensive data, diminishing the effectiveness of characterization studies. Therefore, generating diverse large-scale representations from such limited information is essential to enhancing porous material characterization. This paper proposes a straightforward framework for constructing 3D images of any size, independent of the training image sizes. The approach leverages the inpainting/outpainting capabilities of generative adversarial networks (GANs). Large images are generated by tiling 643 central parts of inpainted/outpainted 963 images. The method is validated visually and statistically across image sizes of 1283, 2563, and 5123. Visual comparisons confirm its ability to replicate real data features, while statistical analyses show that generated image distributions reasonably match real ones in shape and range. The mean and median error percentages of these distributions are compared for permeability, porosity, specific surface area, and Euler number metrics. The highest errors are observed in the permeability, at similar to 11% for the mean and similar to 16% for the median. Moreover, limitations in constructing images of non-overlapping spheres are highlighted using the sphericity metric. Overall, this approach effectively generates porous materials of any dimension by extrapolating extremely limited data, with notable performance such as producing a 20483 image in similar to 40 min. This study offers potential applications in carbon dioxide sequestration, hydrogen storage, water resources studies and hydrocarbon recovery.

Low-cost proton exchange membrane fuel cells are critical to the hydrogen economy. Despite their potential for major cost reductions, most low-platinum oxygen reduction reaction catalysts perform poorly in commercial fuel cells, operating with H-2-Air, ambient pressure, and high current densities (> 1 A cm(-2)) due to severe gas diffusion limitations. Herein, we engineer a serpentine flow field with 100 & micro;m lateral bypasses and 100 & micro;m micro-ribs, or so-called lateral bypass flow field, to improve gas diffusion to the active sites for low-platinum fuel cells. The rationale behind this concept is to remove water saturating within the 100 & micro;m pores of the gas diffusion electrode at the interface with the flow field ribs. Using cathode loadings of 0.1 mgPtCo cm(-2), 0.1 mgPt cm(-2), and 0.2 mgPt cm(-2), the lateral bypass fuel cell achieves 0.63, 0.76, and 1.1 W cm(-2) in H2-Air at ambient pressure, which is up to 75% higher than the conventional serpentine fuel cell. Advanced characterization and simulations are conducted to further elucidate the underlying mechanisms. Operando electrochemical impedance spectroscopy and operando neutron imaging, coupled with advanced two-phase flow simulations using computational fluid dynamics with the Volume of Fluid and the Lattice Boltzmann Methods, and a zero-dimensional analytical model, reveal that the lateral bypasses not only suppress water accumulation beneath the flow field ribs but also enhance water transport across adjacent channels and increase the oxygen content in the catalyst layer. This design also significantly improves the performance of platinum-free catalysts and paves the way for highperformance flood-free fuel cells.

We present a comprehensive dataset of two-phase flow Lattice-Boltzmann simulations, generated using over 100 million GPU hours, covering a wide range of wetting conditions, capillary numbers, and porous geometries. While multiphase flow has traditionally been studied through laboratory experiments, the growing power of computational simulations provides a scalable and efficient alternative. Our simulations, validated against synchrotron beamline experiments, reveal key insights into the effects of wettability, ganglion dynamics, and flow behaviors that can be used to either substantiate current upscaling theories or develop new approaches. The dataset includes 50 relative permeability curves and over 25,000 distinct fluid configurations. Acquiring equivalent data through experiments would be impractical using current techniques, and the computational resources required far exceed those typically available without direct access to high-performance facilities. This open-access dataset enables broad collaboration within the porous media research community and offers a valuable foundation for future studies on pore-scale transport, relative permeability prediction, and data-driven modeling approaches.

Underground hydrogen storage (UHS) in depleted hydrocarbon reservoirs can play a critical role for managing renewable energy intermittency, yet the viability of depleted coalbed methane (CBM) reservoirs remains poorly quantified. This study presents a comprehensive field-scale assessment of hydrogen storage performance across different CBM depletion states using a dual-permeability numerical model in CMG-GEM. The model incorporates viscous flow in fractures, gas diffusion in the coal matrix, and competitive H2-CH4 adsorption dynamics. Storage performance was systematically compared between highly depleted (0.5 MPa) and partially depleted (1.38 MPa) CBM reservoirs and benchmarked against an equivalent sandstone reservoir. This study simulates seasonal-scale cycling operations without dedicated cushion gas, relying on residual CH4; all conclusions regarding site suitability are conditional on this operational assumption. Results demonstrate that initial reservoir pressure fundamentally controls the dominant storage mechanism and operational viability. Highly depleted CBM reservoirs exhibit 72% higher maximum storage capacity than sandstone due to extensive H2 adsorption on coal matrix surfaces. However, under the 7-day production window simulated, adsorptive retention limits extraction purity to 22 mol% H2, rendering this configuration unsuitable for short-term seasonal cycling under the operational conditions examined. Partially depleted CBM reservoirs demonstrate free-gas-dominated storage with performance comparable to conventional sandstone and achieve significantly higher extraction purity (approximate to 98%), making them suitable for seasonal storage requiring high extraction purity. Sensitivity analysis reveals a direct correlation between matrix porosity and storage capacity, and an inverse correlation between residual CH4 content and net H2 storage. This work establishes that depleted CBM reservoirs are technically feasible for UHS, but site selection requires careful alignment with depletion state to optimize the critical trade-off between storage capacity and extraction purity for intended operational cycles. These findings provide quantitative screening criteria for CBM site selection in emerging hydrogen economy projects, with partially depleted sites recommended for seasonal storage and highly depleted sites suitable primarily for long-term strategic reserves.

We introduce a nonequilibrium thermodynamic (NET) framework for immiscible two-phase flow in porous media, in which the total flux of both phases is driven by the gradient of an effective pressure. We show that the classical two-phase Darcy formulation emerges as a projected limit of the full linear flux-force relations under steady-state saturation constraints, with cross-coupling terms arising naturally. Large-scale lattice Boltzmann simulations reveal that the full Onsager transport matrix remains symmetric at the fundamental level, while its reduced Darcy representation can appear asymmetric because the standard force-flux pairs are not independent. Projecting onto the physically admissible subspace restores reciprocity and recovers the total phase mobility from the main- and cross-coupling coefficients. These results resolve the long-standing debate over apparent violations of Onsager symmetry and establish Darcy's law as an emergent thermodynamic limit for multiphase porous-media flow.

Generating micro-CT images of porous media is essential for rock characterization. However, such data are typically sparse and limited to small scanned intervals, leaving large sections without this valuable information. This paper presents a deep learning framework for generating large missing intervals of micro-CT data conditioned on petrophysical information. The workflow integrates the strengths of diffusion models for conditional image synthesis with generative adversarial networks for spatially consistent stitching of consecutive sub-images.The study investigates the performance of multiple conditional generative models and conditioning mechanisms to create micro-CT images that adhere closely to the petrophysical constraints. Additionally, several inpainting variants are evaluated to ensure high-quality stitching across sub-image boundaries. Model performance is assessed using four metrics including porosity, surface area, Euler number, and permeability. Furthermore, the large-scale micro-CT images are compared against real data using mean square differences between consecutive slices and their adherence to the conditional information.Overall, this study demonstrates the advantages of integrating diffusion and adversarial generative models to produce high-quality, conditionally consistent micro-CT images over large intervals from limited training data. The outcomes of this work have applications across multiple fields, including hydrogeology, geoscience, petroleum engineering, as well as hydrogen and carbon storage studies.

The internal micro/nanostructure of materials is crucial for both performance and optimizing the manufacturing process, and it is commonly characterized using 2-dimensional (2D) imaging techniques in practical materials applications, such as scanning electron microscopy (SEM). However, analyzing 3-dimensional (3D) materials with 2D techniques presents many challenges, including sampling errors and misrepresentation of structural features, leading to inaccurate estimates of key physical properties and unreliable predictions of material performance. Generative deep-learning methods have been developed to reconstruct 3D domains from 2D high-resolution (HR) images, enabling the analysis and characterization of structures within these domains. However, generative models are computationally intensive and lack scalability when dealing with the 3D domain, resulting in the generation of 3D images with limited resolution and field of view (FOV), far below the usable representative field for real-life SEM images. We introduce a new generative pipeline, Surface-to-Volume Generative Adversarial Network (SurVol), featuring a modified 2D-3D generator with a dual super-resolution network for super-large HR domain generation. We present engineering applications of SurVol across four real-life materials. Our results demonstrate that the generated domains accurately preserve structural information and physical parameters related to macro-micro pores, solid morphology, effective diffusivity, and permeability. The ability of SurVol to generate super-large HR 3D domains with large FOV enables advanced material characterization, surpassing the hardware limitations of 2D imaging and resolving the software applicability challenge of deep learning for material engineering.

The composition, microstructure, and physical behavior of rock at the core‐to‐pore scale are commonly visualized and characterized using 3D X‐ray microcomputed tomography (micro‐CT). However, a key geophysical problem in micro‐CT rock characterization is how to efficiently handle and process large numbers of high‐resolution 3D rock images, and how to preserve multiscale features for characterization at coarse resolution. Typically, a single high‐resolution 3D micro‐CT image can be extremely large with more than voxels, and in the case of dynamic scans, hundreds of such large images will be generated, posing challenges in storage, downstream analysis, modeling, and their application to deep learning. Existing image rescaling methods face challenges such as information loss during down‐sampling and limited feature recovery during up‐sampling. Herein, we present TopoRSNet, an image rescaling network designed to adjust micro‐CT images to an optimal size and resolution while preserving global and local representative features. Feature preservation is ensured by introducing three feature‐based loss functions: adversarial loss, a new feature consistency loss, and a novel persistent homology loss (PH loss). Combined with a pixel consistency loss, we assure the preservation of both pixels and features during rescaling. The efficacy of TopoRSNet is validated on two common geological rocks, and the results are compared to other rescaling methods. This method enables scalable analysis of multiscale rock features, allowing for broader integration of 3D imaging in geoscientific modeling, simulation, and machine learning workflows.

Accurate prediction of relative permeability is essential for continuum-scale simulations of multiphase flow in porous media. This study presents a workflow that couples a thermodynamic model with a data-driven approach to estimate relative permeability directly from continuum-scale wettability. By leveraging a pre-trained neural network, the method bypasses the need for repetitive pore-scale simulations and rapidly predicts permeability based on fluid configurations. Validation using CT images of a glass bead pack confirms the accuracy and physical consistency of the predictions. Integration with the MATLAB Reservoir Simulation Toolbox framework demonstrates the workflow's scalability and ease of application. This approach offers an innovative and efficient solution for modeling complex multiphase flow, advancing the computational tools available for large-scale porous media simulations.

This study hypothesizes that a pore morphology method (PMM) can be used to accurately determine representative contact angles by effectively capturing fluid morphologies within porous media, thereby overcoming the challenges of accurate wettability characterization for porous materials. We introduce a methodology for the estimation of the wettability, along with measurements of capillary pressure and relative permeability, using a PMM. This approach employs morphological operations to model quasistatic drainage under different surface wetting conditions. To assess PMM, fluid morphologies resulting from the simulation were compared with experimentally derived geometric and thermodynamic contact angles, along with surface area, and Euler characteristic measurements. Based on fluid configurations under different wettability conditions, we find that PMM effectively captures realistic fluid morphologies. At lower capillary pressures, PMM exhibits superior adaptability to a wide range of wetting behaviors. However, at higher capillary pressures, PMM does not reflect the true morphologies of the fluid due to the interfaces that exist in the pendular state. The influence of these effects at higher capillary pressures introduces an inaccuracy in the simulated relative permeability of the wetting phase, though they do not affect the relative permeability of the nonwetting phase. Overall, these findings can significantly enhance the accuracy of wettability characterization in porous media, thereby advancing our understanding and prediction of fluid behavior in surface-based research of porous materials.

Multiphase flow in porous media is fundamental to various geological processes, including carbon capture, geothermal energy production, and enhanced oil recovery. However, the role of fluid properties and surface wettability in determining displacement patterns during flow remains not fully understood. This study addresses this gap by examining the effects of fluid viscosity and wettability on two-phase flow through porous media using a combination of microfluidic experiments and high-resolution numerical simulations. Our findings indicate that viscosity and wettability significantly influence the morphology of fluid displacement, with lower viscosity ratios leading to viscous finger-like invasion patterns, while higher viscosity ratios result in more compact displacement fronts. A significant increase in interface area generation is identified during the transition from compact displacement to viscous flow. This aligns with the energy balance analysis, which reveals that a greater portion of the injected fluid energy is expended on creating new interfaces. Wettability also plays a critical role in displacement patterns, especially under intermediate conditions, causing more interfacial dynamics than water-wet and oil-wet conditions. These insights advance our understanding of pore-scale mechanisms and contribute to more accurate multiphase flow models, ultimately informing applications in resource extraction and underground fluid management.

We introduce a novel experimental approach for measuring Onsager coefficients in steady-state multiphase flow through porous media, leveraging the fluctuation-dissipation theorem to analyse saturation fluctuations. This method provides a new tool for probing transport properties in porous media, which could aid in the characterisation of key macroscopic coefficients such as relative permeability. The experimental set-up consists of a steady-state flow system in which two incompressible fluids are simultaneously injected into a modified Hele-Shaw cell, allowing direct visualisation of the dynamics through optical imaging. By computing the temporal correlations of saturation fluctuations, we extract Onsager coefficients that govern the coupling between phase fluxes. Additionally, we have performed a statistical analysis of the fluctuations in the derivative of saturation under different flow conditions. This analysis reveals that while the fluctuations follow Gaussian statistics up to 2-3 standard deviations, they exhibit heavy tails beyond this range. This work provides an experimental foundation for recent theoretical developments in the extention of non-equilibrium thermodynamics to multiphase porous media flows. By linking microscopic fluctuations to macroscopic transport behaviour, our approach offers a new perspective that may complement existing techniques in the study of multiphase flow, making it relevant to both statistical physics and the broader fluid mechanics community.

Relative permeability plays an important role in the upscaling of multiphase flow in porous media from the pore scale to the Darcy scale. The entire concept of relative permeability is contingent on the existence of a representative elementary volume (REV). As we move to smaller samples to measure relative permeability, such as with digital core analysis, the concept of a classical REV has become increasingly unlikely when using the conventional approach to defining a representative volume. The “‘conventional”’ understanding of an REV is that a large enough volume must be considered such that spatial variability averages out. In digital rock methods, such as pore-scale simulations based on micro-computed tomography (CT) images, the domain size is typically 2 to 4 mm. This is approximately the length scale of a single-phase flow REV using the classic REV approach. However, the single-phase perspective does not consider the complex dynamics and fluctuations often observed in multiphase flow systems, even at centimeter-scale experiments and/or simulations. A fundamental question is, therefore, whether the domain size commonly used in digital rock simulations can provide a consistent energy budget such that the concept of relative permeability exists. Based on first principles, relative permeability accounts for the rate of energy dissipated in a stationary process. If the dynamics are fluctuating, the energy dissipated can vary but will average out over a long enough timescale. The key to determining the validity of the relative permeability is the timescale of the measurement, not the spatial scale. The conventional REV theory assumes that spatial, temporal, and ensemble averages are equivalent in an ergodic system, but it does not provide a way to test this assumption. We provide a formal way to identify the timescale where the relative permeability accurately captures energy dissipation as a way to validate relative permeability measurements and quantitatively assess their accuracy. This result will be tested for a practical SCAL test, determining how long a flow experiment needs to be run to accurately characterize the rate of energy dissipation by the flow. The outcome will be a best practice guide for the determination of relative permeability from core-scale experiments and/or digital core simulations that ensure the energy budget is fully accounted for in the relative permeability coefficient.

Digital imaging and modeling are essential tools for characterizing rock structures and understanding fluid flow behavior. These efforts often rely on X-ray micro-computed tomography (micro-CT), which faces an inherent trade-off between resolution and field-of-view (FOV). Deep learning super-resolution (SR) methods have been developed to overcome this limitation, but their application to carbonate rocks is challenged by complex micro-nanometer features. Due to the resolution limits, micro-CT fails to capture sub-micrometer features such as micropores in carbonates, and using such data as high-resolution (HR) training images limits the SR model's ability to accurately reconstruct the micropore structures. We introduce a cascading SR pipeline designed to address these challenges and reveal sub-micrometer features in carbonate rocks. The approach integrates multi-stage 2D SR networks to progressively enhance low-resolution (LR) images toward the HR domain, followed by a third-plane SR network for 3D reconstruction. We evaluate this method on a three-stage SR task: starting from a 3 mu ${\upmu }$m resolution micro-CT image, super-resolving to an intermediate 1 mu ${\upmu }$m resolution, and ultimately reaching 0.1 mu ${\upmu }$m resolution based on scanning electron microscopy (SEM), achieving a 30x ${\times} $ scale factor. Validation with unseen SEM demonstrates that the reconstructed domains retain essential structural and physical properties. This approach provides a practical solution to current imaging limitations and enables the integration of multi-resolution modalities for improved rock characterization.