Flow in Disordered Media

Anders Malthe‐Sørenssen, Anders Malthe-SorenssenLecture Notes in Physics 2024 · DOI · OpenAlex

Abstract

In this paper, we introduce the basic concepts of disordered media. We introduce properties of flow of current or fluids, and then address flow in a percolating system close to $p_c$. We study the behavior numerically, develop a scaling theory, and find properties using the renormalization group approach. Our initial studies focus on the binary porous medium of the percolation system. However, we also extend our results to more general random media, demonstrating how this can be achieved.

Introduction

In traditional materials science, properties like electrical conductivity or fluid permeability are often treated as material constants. However, in disordered systems—such as geological formations, porous membranes, or composite materials—these properties become highly sensitive to the system's connectivity.

The core challenge is "up-scaling": how do we predict the flow behavior of a large-scale system when we can only practically model it at a microscopic level? Near the percolation threshold ($p_c$), where a system just barely becomes connected, the geometry is fractal. This work addresses how this fractal geometry leads to anomalous, size-dependent transport properties that defy classical intuition.

Methodology

To bridge the gap between microscopic disorder and macroscopic flow, a three-pronged approach is employed:

  1. Numerical Simulation: Using Python (specifically NumPy and SciPy), disordered media are modeled as Random Resistor Networks. By applying Kirchhoff’s rules and solving large systems of linear equations, local potentials and total currents are calculated across $L \times L$ lattices.
  2. Scaling Theory: A scaling ansatz is developed to describe how conductance ($G$) varies with system size ($L$) and the correlation length ($\xi$). This includes distinguishing between the total spanning cluster and the "backbone"—the subset of the cluster that actually carries current.
  3. Renormalization Group (RG) Calculation: To find analytical estimates for scaling exponents, an "H-cell" renormalization scheme is used on a $2 \times 2$ bond lattice. This allows for the calculation of the resistance exponent ($\tilde{\zeta}_R$) by observing how average resistance changes as the system is viewed at increasingly larger scales.
Numerical simulation of the potential distribution across a large percolating lattice.

The study utilizes synthetically generated percolation lattices, typically $L \times L$ grids (with $L$ up to 400) where sites or bonds are occupied with a probability $p$. These synthetic benchmarks are standard in statistical physics for testing universality and scaling laws.

Key Insights and Theoretical Framework

The Backbone vs. The Spanning Cluster

A crucial insight is the realization that most of a connected cluster is "dead weight" for flow. "Dangling ends" contribute to the mass of the cluster but carry zero current. Only the backbone—the part of the cluster with at least two paths to the boundaries—matters for conductivity.

Visualization of the backbone (green/yellow) and dangling ends (blue) within a spanning cluster.

Conductivity vs. Density

Because the backbone is much sparser than the total spanning cluster, conductivity ($g$) increases much more slowly than the cluster density ($P$) as the system moves above the percolation threshold. This explains why the conductivity exponent $\mu$ is significantly larger than the mass exponent $\beta$.

Multifractality and Universality

The distribution of currents within the backbone is not a simple fractal. Different "moments" of the current distribution scale with different exponents, a phenomenon known as multifractality. This means a single fractal dimension is insufficient to describe the internal physics of flow. Furthermore, at low velocities, the mathematics for electrical current (Ohm’s Law) and fluid flow (Darcy’s Law) is identical, allowing resistor network theory to solve complex hydraulic problems.

Key Results

  • Scaling Exponents: For 2D systems, the conductivity exponent $\mu$ is found to be $\approx 1.3$. This is contrasted with the cluster density exponent $\beta \approx 0.14$, quantitatively proving that flow properties emerge much more "reluctantly" than physical connectivity.
  • RG Accuracy: The renormalization group calculation yields a resistance exponent $\tilde{\zeta}_R \approx 0.939$, consistent with the theoretical bounds established by the "blob model" of the backbone.
  • Data Collapse: The research successfully demonstrates finite-size scaling collapse. By plotting $L^{\mu/\nu}g(p,L)$ against $L^{1/\nu}(p-p_c)$, data from different system sizes fall onto a single curve, validating the scaling theory and the universality of the results.

Limitations and Future Work

While providing a robust framework, the current model has limitations. It assumes uncorrelated porosity, whereas real-world geological systems often have correlated porosity due to sedimentary processes. Most of the focus is on binary conductance (on/off systems); future work could explore continuous distributions of material properties. Additionally, the study is limited to incompressible, single-phase flow. Extending this to multiphase flow (e.g., oil and water) would require complex invasion-percolation models to account for capillary forces.

Summary updated

Abstract

Abstract In this chapter, we introduce the basic concepts of disordered media. We introduce properties of flow of current or fluids, and then address flow in a percolating system close to $$p:c$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mrow> <mml:mi>p</mml:mi> </mml:mrow> <mml:mrow> <mml:mi>c</mml:mi> </mml:mrow> </mml:msub> </mml:math> . We will study the behavior numerically, develop a scaling theory, and find properties using the renormalization group approach. Our initial studies will be on the binary porous medium of the percolation system. However, we can also extend our results to more general random media, and we demonstrate how this can be done towards the end of the chapter.
BibTeX
@inproceedings{malthesrenssen2024flow,
  author = {Anders Malthe‐Sørenssen and Anders Malthe-Sorenssen},
  title = {{Flow in Disordered Media}},
  booktitle = {Lecture Notes in Physics},
  year = {2024},
  doi = {10.1007/978-3-031-59900-2\_9},
}
Percolation Theory Using PythonLecture Notes in Physics · 2024
Subset GeometryLecture Notes in Physics · 2024
Dynamic Processes in Disordered MediaLecture Notes in Physics · 2024
Introduction to PercolationLecture Notes in Physics · 2024
Diffusion in Disordered MediaLecture Notes in Physics · 2024
Geometry of ClustersLecture Notes in Physics · 2024
Figure