Abstract
This study develops a Lattice Boltzmann Method (LBM) framework to investigate isothermal two-phase fluid flow within porous media. Utilizing the Shan-Chen multiphase model, the implementation allows for the natural coexistence and separation of two fluid phases. The model is validated against theoretical benchmarks, including phase separation, surface tension (Laplace’s Law), and droplet dynamics. Furthermore, the research explores the critical role of wettability on fluid intrusion and trapping phenomena in complex geometries, demonstrating the model's capability to reproduce complex flooding behaviors under varying surface conditions.
Introduction
Simulating fluid transport through porous materials—such as soil, industrial filters, or hydrocarbon reservoirs—presents a significant challenge for traditional Computational Fluid Dynamics (CFD). Conventional methods typically solve macroscopic Navier-Stokes equations, which often struggle to accurately capture the intricate, shifting interfaces between fluid phases and the highly irregular boundaries of pore networks.
The Lattice Boltzmann Method (LBM) offers a "bottom-up" alternative. By tracking the distribution of particle populations on a discrete lattice rather than solving continuous equations, LBM naturally handles complex geometries and interfacial physics. This paper implements a D2Q9 lattice-BGK model to bridge the gap between microscopic interactions and macroscopic flow behavior in porous media.
Methodology: The Shan-Chen Model
The core of the simulation relies on the Shan-Chen Multiphase Model. This approach introduces a short-range attractive force between fluid particles, enabling a single substance to spontaneously separate into distinct liquid and vapor phases based on local density fluctuations.
To model the interaction between fluids and solid surfaces, the Martys Adhesive Force is integrated. By adjusting the interaction strength ($G_{ads}$), the model can simulate a full spectrum of wettability, from non-wetting to perfectly wetting surfaces. Boundary conditions at the solid-fluid interface are handled via the Bounce-Back method, ensuring no-slip conditions with high computational efficiency.
Validation and Emergent Physics
A key advantage of the LBM is that complex macroscopic physical laws emerge from simple local rules. The implementation was validated using several idealized scenarios:
- Phase Separation: Starting from a nearly uniform density, the system naturally evolves into stable droplets and vapor regions.
- Laplace’s Law: The relationship between the pressure difference across a droplet interface ($\Delta P$) and its radius ($r$) was tested. The simulation results showed excellent agreement with the theoretical Laplace equation.
Insights into Porous Media Dynamics
The study applied the validated model to a synthetic "chamber and throat" pore network. This geometry mimics real-world porous structures where large pores are connected by narrow capillaries.
The Impact of Wettability
Wettability was found to be the primary driver of flow patterns and trapping mechanisms. In scenarios with "strong wettability," the fluid forms a thin film along the pore walls. This leads to a phenomenon known as "snap-off," where the wetting fluid pinches off and traps the non-wetting phase (e.g., gas or oil) in the center of the pores. This significantly reduces the "sweeping efficiency" of the fluid displacement process.
Path Selection
When droplets encounter obstacles or branching paths, they naturally gravitate toward wider openings. This behavior occurs because wider paths offer lower capillary resistance—a physical reality that the LBM captures intrinsically without the need for explicit path-finding algorithms.
Conclusion
The Lattice Boltzmann implementation successfully models the complex interplay of interfacial tension, wettability, and geometry in two-phase flow. By accurately reproducing theoretical benchmarks like Laplace’s Law and capturing emergent behaviors like snap-off in porous networks, this framework provides a robust tool for studying unstable and intermittent flows that are difficult to resolve with standard continuum models. Future work will focus on extending these simulations to three-dimensional domains and exploring the effects of varying viscosity ratios between phases.
Summary updated