Linear Stability Analysis of Two-phase, Two-Component Flow in Porous Media

This study extends linear stability analysis to partially miscible two-phase, two-component flow in porous media by deriving jump conditions for discontinuous eigenfunction derivatives and demonstrating that interphase mass transfer predominantly stabilizes viscous fingering instabilities by reducing viscosity contrast and altering shock properties, while revealing complex interactions between capillary forces and mechanical dispersion.

The Gist

The Big Picture: The "Finger" Problem

Imagine you are trying to push a thick, sticky substance (like honey) out of a sponge using a thinner, runnier substance (like water). In a perfect world, the water would push the honey out in a neat, straight line, like a piston.

However, in reality, the water doesn't push evenly. Because the water is thinner, it finds the path of least resistance and shoots through the honey in thin, branching streams. These streams look like fingers reaching out. This is called viscous fingering.

In the real world, this is a problem for things like:

• Oil Recovery: Trying to get oil out of rock.
• Carbon Storage: Trying to bury CO2 underground safely.
• Cleaning Groundwater: Trying to flush out pollutants.

When these "fingers" form, they bypass the target fluid, leaving it behind and making the process inefficient.

What This Paper Studied

Most previous studies looked at two extreme scenarios:

1. Completely Immiscible: The two fluids hate each other and never mix (like oil and water).
2. Completely Miscible: The two fluids mix perfectly like sugar in tea.

This paper looked at the middle ground: Partially Miscible Flow. This is when the fluids do mix a little bit. Imagine pouring a little bit of alcohol into water; they mix, but not instantly or perfectly everywhere. The paper specifically studied what happens when a gas (like CO2) is injected to push out a liquid (like oil), and a small amount of the gas dissolves into the liquid as they meet.

The Main Discovery: Mixing is a Stabilizer

The researchers found that this "partial mixing" acts like a stabilizer.

• The Analogy: Think of the "fingers" as a race car trying to speed past a slower car. If the race car (the gas) is super thin and the other car (the oil) is super thick, the race car shoots past easily, creating chaos (fingers).
• The Effect of Mixing: When the gas mixes slightly with the oil, it changes the oil's properties. It makes the oil less thick (less viscous) right at the boundary where they meet.
• The Result: Because the oil isn't as thick anymore, the gas can't shoot through it as easily. The "fingers" become shorter and less chaotic. The paper concludes that mass transfer (mixing) generally calms down the instability.

The "Jump" in the Math

The math behind this was tricky. Usually, when fluids mix, the change is smooth. But in this specific scenario, the researchers found a "cliff" in the math.

• The Analogy: Imagine driving a car. In the "two-phase" zone (where gas and liquid mix), the road is smooth. But the moment the gas finishes dissolving and you enter the "pure liquid" zone, the road suddenly changes texture.
• The Challenge: The mathematical equations describing the flow have a sudden "jump" or discontinuity at this transition point. The researchers had to invent a special set of rules (called "jump conditions") to connect the math on one side of the cliff to the other, allowing them to solve the puzzle.

Surprising Findings: The "Goldilocks" of Dispersion

The paper also looked at dispersion, which is like the "smearing" effect of the fluids as they move through the tiny holes in the rock.

• Expectation: You might think that more smearing (dispersion) always makes the flow more stable and less chaotic.
• Reality: The researchers found a "Goldilocks" zone.
  • If there is too little smearing, the flow is unstable.
  • If there is too much smearing, the flow becomes stable.
  • But: There is a specific, "just right" amount of smearing where the instability actually gets worse than it would be with very little or very much smearing. It's as if the forces of the rock (capillary forces) and the movement of the fluid (mechanical dispersion) conspire together to create the worst possible "fingers" at a specific setting.

Gravity's Role

The paper also checked what happens if you push the fluids up (against gravity) or down (with gravity).

• Pushing Up: Usually, pushing a light fluid (gas) up through a heavy fluid (liquid) is very unstable because the heavy fluid wants to fall back down. However, the paper found that the mixing effect helps fight this. The mixing changes the density and viscosity in a way that dampens the instability caused by gravity.
• Pushing Down: Both gravity and mixing work together to keep the flow stable.

Summary

This paper built a new mathematical model to understand how fluids behave when they are pushed through rock and mix just a little bit. They discovered that:

1. Mixing helps: Even a little bit of mixing between the gas and liquid makes the flow more stable and reduces the chaotic "fingers."
2. The math is bumpy: The transition between mixed and pure fluid creates a mathematical "cliff" that required special rules to solve.
3. Smearing isn't always good: There is a specific amount of fluid smearing that makes the instability worse, which is a surprising and complex interaction between the rock and the fluid.

The authors did not apply these findings to specific new technologies or clinical uses; they focused strictly on understanding the physics and math of this specific type of fluid flow.

Technical Summary

Technical Summary: Linear Stability Analysis of Two-phase, Two-Component Flow in Porous Media

Problem Statement
Viscous fingering instabilities during fluid displacement in porous media significantly compromise the efficiency of applications such as enhanced oil recovery, CO2 sequestration, and groundwater remediation. While extensive literature exists on linear stability analysis for fully immiscible and fully miscible displacements, the intermediate case of partially miscible flow—where limited mass transfer occurs between phases—remains largely unexplored. Specifically, there is a lack of analytical frameworks that account for the complex interplay of gravity, capillary forces, mechanical dispersion, and interphase mass transfer in a two-phase, two-component system. This study addresses the prediction of unstable perturbation modes in such partially miscible systems, focusing on the scenario where a gaseous fluid displaces a liquid.

Methodology
The authors develop a mathematical model for two-dimensional flow in an isotropic porous medium, assuming thermodynamic equilibrium for a binary mixture (components "a" and "b"). The methodology proceeds through the following steps:

1. Governing Equations: The model derives mass balance equations for a two-phase, two-component system, incorporating Darcy's law, fractional flow, capillary pressure, and mechanical dispersion. A key simplifying assumption is that the liquid phase density is constant in the single-phase region, while viscosity varies with composition.
2. Base State Formulation: The base state is defined as a one-dimensional, travelling-wave solution (shock front) derived from the Buckley-Leverett approach, adapted for mass transfer. The shock configuration is calculated using the method of characteristics, accounting for the discontinuity in flow functions at the transition from the two-phase region to the pure-liquid region.
3. Linear Stability Analysis: Small-amplitude, wave-like perturbations are superimposed on the base state. The governing equations are linearized to form an eigenvalue problem where the eigenvalue represents the perturbation growth rate ($\sigma$) and the eigenfunctions represent the disturbance profiles.
4. Numerical Solution: A critical challenge identified is the discontinuity in the derivatives of the eigenfunctions at the transition point between two-phase and single-phase flow. To resolve this, the authors derive specific jump conditions for the derivatives at this interface. The eigenvalue problem is then solved numerically using the Matched Initial Value Problem (MIVP) method. This involves integrating the linearized ordinary differential equations from far-field boundaries toward the transition point and matching the solutions using the derived jump conditions.

Key Contributions

• Novel Framework: This work presents, to the authors' knowledge, the first linear stability analysis of a genuine two-phase, two-component system that explicitly accounts for interphase mass transfer.
• Jump Conditions: The derivation of jump conditions for the derivatives of eigenfunctions at the phase-transition discontinuity allows for the accurate solution of the eigenvalue problem in systems where flow functions are discontinuous.
• Comprehensive Parameter Study: The study systematically investigates the effects of initial composition, dispersivities (longitudinal and transverse), miscibility levels (via binary interaction coefficients), viscosity ratios, and gravity on the stability of the displacement front.

Results

• Stabilizing Effect of Mass Transfer: The primary finding is that mass transfer has a predominantly stabilizing effect on the displacement. It reduces the maximum growth rate ($\sigma_{max}$) by lowering the equilibrium viscosity ratio ($M_e$) and altering shock properties (reducing shock velocity $v_s$ and increasing gas saturation $S_{gs}$). This stabilization is particularly pronounced for high viscosity contrasts.
• Gravity Interactions: In upward displacements, mass transfer dampens gravity-induced instability. The density contrast that drives gravitational instability also enhances mass transfer, which counteracts the destabilizing effect. In downward displacements, both gravity and mass transfer act as stabilizing mechanisms.
• Dispersion Non-linearity: The study identifies a non-linear relationship between the dimensionless longitudinal dispersion coefficient ($D^*_{\ell,xx}$) and instability. There exists a critical "most dangerous" value of $D^*_{\ell,xx}$ where both the growth rate and the cutoff wavenumber ($n_{cut}$) are maximized. This suggests a complex interaction where intermediate levels of mechanical dispersion, relative to capillary forces, can actually increase instability compared to both very low and very high dispersion regimes.
• Cutoff Wavenumber: While mass transfer significantly reduces the growth rate, its effect on the cutoff wavenumber is less pronounced and more complex, sometimes resulting in higher $n_{cut}$ values for partially miscible cases compared to immiscible ones under specific conditions.

Significance and Claims
The paper claims to bridge the gap between well-studied limiting cases of immiscible and miscible displacement. By demonstrating that mass transfer acts as a stabilizing mechanism, the results suggest that partially miscible conditions may offer more robust displacement fronts than purely immiscible scenarios, particularly in high-viscosity-contrast environments. The identification of the critical dispersion coefficient value highlights a previously unreported complex interaction between capillary forces and mechanical dispersion in the context of viscous fingering. The authors note that while their model simplifies the physics of multi-component systems, it provides a necessary framework for understanding the fundamental stability mechanisms in partially miscible flows. The study acknowledges limitations, such as the assumption of constant liquid density in the single-phase region and fixed relative permeability curves, suggesting these as areas for future refinement.