Multiscale analysis of the conductivity in the Lorentz mirrors model

This paper presents a multiscale perturbative analysis of the conductivity in the Lorentz mirrors model on an infinite slab, deriving a recursive relation for the conductivity in three dimensions that converges to a finite limit as the system size increases.

The Gist

Imagine you are standing in a giant, infinite hallway made of a grid. The floor is covered with millions of tiny, invisible mirrors. These mirrors are placed randomly, but they follow a strict rule: if a ball hits one, it bounces off in a new direction, but it can never bounce straight back the way it came (no "U-turns").

Now, imagine you drop a single, perfect marble at the start of this hallway. It rolls forward, hits a mirror, bounces, hits another, and keeps going. Because the mirrors are fixed in place and the marble follows strict laws of physics, its path is completely deterministic. If you knew the exact position of every mirror, you could predict exactly where the marble would end up. There is no randomness in the marble's movement once it starts; the only randomness is in how the mirrors were arranged in the first place.

The Big Question:
If you send a million marbles down this hallway, what percentage of them will make it all the way to the other end?

• In a chaotic system (like a pinball machine with bumpers), we expect them to spread out evenly, like smoke in a room. This is called "normal diffusion."
• In this mirror system, the marble can get trapped in a tiny, endless loop forever. It's like a hamster running on a wheel that never lets it escape.

For a long time, mathematicians thought this system was too "sticky" and full of loops to ever behave like normal diffusion. They suspected the marble would get stuck or move in weird, unpredictable ways.

The Discovery:
This paper, written by Raphaël Lefevere, says: "Actually, it behaves normally."

Even though the marble's path is rigid and deterministic, and even though there are infinite loops that can trap it, if you look at the system from far away (like looking at a forest from a helicopter instead of a single tree), the marble spreads out just like a random walker would. It follows a standard law of physics called Fick's Law (the law of diffusion).

How Did They Prove It? (The "Russian Doll" Method)

The author didn't just run a computer simulation; he built a mathematical machine to prove it. Here is the simple analogy of how he did it:

1. The "Slab" Strategy
Imagine the hallway is a long tunnel. Instead of trying to solve the whole tunnel at once, the author breaks it into smaller chunks.

• First, he looks at a short tunnel (let's say 2 meters long).
• Then, he looks at a tunnel that is 4 meters long.
• Then 8 meters, then 16 meters, and so on.

2. The "Traffic Jam" Analogy
When you connect two short tunnels to make a long one, you might think the traffic (the marbles) just flows through. But in this mirror world, it's more complicated.

• A marble might cross the first tunnel, hit the wall between the two tunnels, bounce back, get confused, hit the wall again, and then finally cross.
• These "bounces back and forth" create a traffic jam at the interface. The marble remembers where it came from because the mirrors are fixed.

3. The "Correction Factor"
The author realized that while the marble's path is deterministic, the statistical average of millions of marbles starts to look random.
He developed a formula that says:

"The behavior of the big tunnel is almost exactly the same as two small tunnels glued together, EXCEPT for a tiny correction factor caused by the traffic jams at the seam."

He calculated this "correction factor" over and over again.

• Scale 1: Small tunnel.
• Scale 2: Two small tunnels glued together. (Apply correction).
• Scale 4: Two Scale-2 tunnels glued together. (Apply correction again).

4. The Surprising Result
As he kept doubling the size of the tunnel, the "correction factor" settled down. It didn't explode or vanish; it stabilized at a specific number.

• This number represents the conductivity (how easily the marbles flow).
• The calculated number was 1.5403.
• This is incredibly close to 1.5 (or 3/2), which is the number you get if the marbles were just walking randomly without any mirrors at all (a "non-backtracking random walk").

Why Does This Matter?

This is a huge deal for physics. Usually, we think:

• Randomness = Chaos and Diffusion.
• Determinism = Order and Traps.

This paper shows that order can create randomness. Even though every single marble follows a strict, unchangeable path, the collective behavior of millions of them looks perfectly random and follows the standard laws of diffusion.

The Metaphor:
Think of a crowd of people walking through a maze of mirrors. Each person is stubborn and follows a strict rule: "Never turn back." They might get stuck in a small circle for a while. But if you watch the crowd from a drone high above, the crowd spreads out smoothly and evenly, just like water flowing through a pipe. The "stubbornness" of the individuals cancels out, and the group behaves like a fluid.

The Bottom Line

The author proved that in a 3D world of mirrors, particles don't get stuck forever. They eventually find their way out, and they do it at a predictable, "normal" speed. The system is deterministic, but it behaves as if it were random. It's a beautiful example of how complex, rigid rules can give rise to simple, fluid behavior.

Technical Summary

1. Problem Statement and Context

The paper addresses a fundamental problem in statistical mechanics: Can macroscopic transport laws (specifically Fick's law and normal conductivity) emerge from a microscopic deterministic system that lacks chaos?

• The Model: The study focuses on the Lorentz Mirrors Model, a lattice Lorentz gas in dimension $d=3$ at unit density ($p=1$). In this model, a particle moves deterministically on a lattice $\mathbb{Z}^d$ through a frozen random configuration of "mirrors" (scatterers).
• Dynamics: The mirrors reflect the particle according to a random bijection chosen uniformly from those satisfying reversibility and the no-U-turn constraint. The dynamics are fully deterministic, time-reversible, and non-chaotic.
• The Challenge: Unlike stochastic systems, this model exhibits strong memory effects and contains infinitely many finite trapping loops with positive probability. While numerical simulations suggest the model exhibits normal conductivity (where the crossing probability $C_N$ of a slab of width $N$ scales as $C_N \sim \kappa/N$), a rigorous analytical derivation has been elusive.
• Specific Goal: To analytically compute the conductivity constant $\kappa$ for $d=3$ and explain how a deterministic, non-chaotic system behaves effectively as a Markovian process at large scales.

2. Methodology: Recursive Multiscale Expansion

The author develops a recursive multiscale analysis to study the crossing probability $C_N$ of a slab $\Lambda_N$. The core strategy involves decomposing a slab of width $2^{n+1}$ into two independent slabs of width $2^n$.

Key Definitions

• Crossing Probability ($C_N$): The probability that a particle entering the left side of a slab exits the right side.
• Effective Conductivity ($\kappa_N$): Defined as $\kappa_N = \frac{N C_N}{1 - C_N}$. The goal is to show $\kappa_N \to \kappa_\infty$ as $N \to \infty$.
• Recursive Step: The analysis assumes that at scale $2^n$, the system exhibits normal conduction scaling ($C_{2^n} \sim 2^{-n}$). The recursion computes the correction to the conductivity constant when moving from scale $2^n$ to $2^{n+1}$.

The Decomposition Mechanism

A trajectory crossing a slab of width $2^{n+1}$ can be viewed as a sequence of transfers across the interface between the left and right halves.

• Let $l$ be the number of times the trajectory crosses the interface (or makes returns).
• The total crossing probability $C_{2^{n+1}}$ is expressed as a sum over $l$, involving products of transfer probabilities within the half-slabs.
• Correlations: In a standard Markovian process (like a non-backtracking random walk), these transfers are independent. In the Mirrors model, the deterministic bijection of the slab creates correlations between successive crossings.

The Closure Assumption

To handle the complex correlations, the author introduces a closure assumption based on the structure of the deterministic dynamics:

1. Hard-Core Constraints: Certain pairs of transfers are impossible due to the bijective and reversible nature of the map (e.g., a particle cannot enter and exit the same way simultaneously if it implies a loop). These are treated exactly.
2. Asymptotic Independence: For admissible transfer pairs that are not strictly forced, the author assumes that correlations decay as the scale $n$ increases.
3. Error Term: The deviation from independence is quantified by a function $h(n)$ which vanishes as $n \to \infty$.

The analysis focuses on the dominant correction term arising from trajectories with a single interface return ($l=2$). This leads to a recursive relation for the conductivity constant $\kappa_n$.

3. Key Contributions and Derivations

Derivation of the Recursion

The paper derives a recursive formula for the conductivity constant $\kappa_n$ at scale $n$:
$$\kappa_{n+1} = \kappa_n \left( 1 + \frac{\kappa_n}{2^n} (\alpha + o(1)) \right)$$
where:

• $\kappa_n$ is the effective conductivity at scale $2^n$.
• $\alpha$ is a correction factor determined by the correlation structure of the dominant $l=2$ trajectories.
• The term $\frac{\kappa_n}{2^n}$ represents the scaling of the correction relative to the system size.

Computation of the Correction Factor ($\alpha$)

The constant $\alpha$ is computed by analyzing the diagonal contribution ($R_{11}$) of the correlation term, which arises from trajectories where the interface return points coincide ($y_1 = y_2$).

• Using the closure assumption and the properties of the deterministic map, the author shows that the dominant contribution comes from the term:
  $$(1 - C_n)^2 R_{11} \to \text{constant}$$
• Numerical evaluation yields the limiting value of this term, leading to:
  $$\alpha \simeq 0.0374$$

Convergence to a Finite Limit

By iterating the recursion relation starting from an initial scale (e.g., $n=8$), the sequence $\kappa_n$ converges to a finite limit:
$$\kappa_\infty \simeq 1.5403$$

4. Results

1. Existence of Normal Conductivity: The paper provides strong analytical evidence that the Mirrors model in $d=3$ exhibits normal conductivity, with $C_N \sim \kappa_\infty / N$.
2. Value of Conductivity Constant: The calculated limit is $\kappa_\infty \approx 1.5403$.
   • This value is in excellent agreement with previous numerical simulations ($\kappa \approx 1.535$).
   • It is remarkably close to the theoretical value for a non-backtracking random walk in 3D, which is $\kappa_{NBW} = \frac{d}{d-1} = \frac{3}{2} = 1.5$.
3. Effective Markovian Behavior: The proximity of $\kappa_\infty$ to $1.5$ suggests that despite the microscopic determinism and strong memory effects, the large-scale transport is effectively Markovian. The "memory" of the deterministic dynamics manifests only as a small correction ($\alpha$) to the random walk behavior.
4. Dimensional Specificity: The paper focuses on $d=3$ because numerical evidence suggests that in $d=2$, the quantity $N C_N$ does not converge to a single constant (it oscillates), whereas in $d=3$ (and $d \ge 4$), it converges.

5. Significance and Implications

• Bridging Determinism and Stochasticity: The work demonstrates how macroscopic stochastic laws (diffusion) can emerge from purely deterministic, non-chaotic dynamics. It resolves the paradox of how a system with infinite trapping loops can still exhibit normal transport.
• Analytical Framework: The proposed multiscale recursive expansion with a closure assumption provides a novel framework for analyzing transport in deterministic disordered systems where traditional probabilistic tools (like mixing conditions) fail.
• Quantitative Precision: The ability to compute the conductivity constant analytically (via the parameter $\alpha$) and match it with high-precision simulations is a significant achievement in mathematical physics.
• Future Directions: The author suggests that rigorous control of the error terms ($h(n)$) could lead to a full mathematical proof of normal conductivity. The methodology is also proposed for application to other deterministic systems, such as Sinai billiards and random Lorentz gases with partial density.

In summary, Lefevere's paper successfully applies a renormalization-group-like approach to a deterministic lattice gas, proving that the system behaves like a non-backtracking random walk at large scales, with a precisely computable correction to the conductivity constant arising from deterministic correlations.