Boundary-induced classical Generalized Gibbs Ensemble with angular momentum

This study demonstrates that the boundary shape of a confined classical hard disk system dictates its thermalization outcome, where circular boundaries induce a non-ergodic Generalized Gibbs Ensemble characterized by conserved angular momentum and near-boundary condensation, in contrast to the standard Gibbs Ensemble observed with square boundaries.

The Gist

Imagine you have a giant, empty room filled with thousands of tiny, bouncy ping-pong balls. You shake the room, and the balls bounce around, hitting each other and the walls. In the world of physics, we usually expect that after enough time, these balls will settle into a predictable, calm pattern called the Gibbs Ensemble. Think of this as the "standard party" where everyone is moving randomly, and the energy is spread out evenly across the whole room.

This paper asks a simple but profound question: What happens if the shape of the room changes?

The authors discovered that if you change the room from a square to a circle, the balls don't just settle down normally. Instead, they get stuck in a weird, swirling dance that breaks all the usual rules of physics.

Here is the breakdown of their discovery using everyday analogies:

1. The Square Room vs. The Circular Room

• The Square Room (The Normal Party): If the walls are square, the balls bounce off at random angles. Over time, they forget where they started. They spread out evenly, and their speeds follow a standard bell-curve distribution. This is the "Gibbs" state, which is what almost all physics textbooks teach us to expect.
• The Circular Room (The Swirling Dance): If the walls are a perfect circle, something magical and strange happens. Because the wall is curved, every time a ball hits it, it bounces off in a way that preserves its spin (angular momentum).
  • The Analogy: Imagine a marble rolling inside a smooth, round bowl. If you give it a spin, it keeps spinning around the edge forever. It doesn't wander into the middle. In a square bowl, the marble would eventually hit a corner and lose that specific spin pattern. In a circle, the spin is "locked in."

2. The "Condensation" Phenomenon

Because the balls in the circular room can't lose their spin, they don't spread out evenly. Instead, they condense near the walls.

• The Analogy: Imagine a crowd of people in a circular arena. If everyone is forced to run in a circle without stopping, they will all end up hugging the outer wall, leaving the center of the arena completely empty.
• The paper shows that in this circular room, the balls pile up against the boundary, creating a high-density ring, while the middle becomes a ghost town. This is a "condensation" caused purely by the shape of the container.

3. The "Generalized Gibbs Ensemble" (GGE)

In standard physics, we assume that Energy is the only thing that matters when things settle down. This paper introduces a new rule: Angular Momentum (spin) is also a conserved quantity that matters just as much.

• The Analogy: Think of a library.
  • Standard Physics (Gibbs): You only care about how many books are in the library (Total Energy). You don't care who borrowed them.
  • This Paper (GGE): You realize that the library also has a rule that "books must stay on the shelf they were taken from" (Conserved Angular Momentum). Because of this extra rule, the books don't get shuffled randomly; they stay in specific patterns.
• The authors call this new state the Generalized Gibbs Ensemble. It's a "super-ordered" state that only appears when you have both a circular shape and spinning particles.

4. Why This Breaks the Rules (Ergodicity)

There is a famous rule in physics called Ergodicity. It basically says: "If you wait long enough, a system will visit every possible state."

• The Analogy: If you drop a drop of ink in a glass of water, it will eventually spread to every corner.
• The Breakdown: In this circular room, the ink never spreads to the center. The system gets "stuck" in a specific state based on how it started. If you start the balls spinning clockwise, they stay spinning clockwise forever. They never "forget" their initial spin. This is a hard break in the rules of how we think thermalization works.

5. The "Magnetism" Surprise

The paper ends with a mind-bending twist involving the Bohr-van Leeuwen Theorem.

• The Old Rule: This famous theorem says that in a purely classical world (no quantum mechanics), you cannot create magnetism just by moving charged particles around. Thermal motion should cancel out any magnetic effects.
• The New Discovery: Because the circular boundary forces the particles to keep their spin (breaking time-reversal symmetry), the authors show that you could theoretically generate a magnetic field just by spinning these particles in a circle.
• The Analogy: It's like saying, "We thought a spinning fan couldn't create a magnetic field, but if you put the fan inside a perfect, frictionless circular tube, suddenly it does."

Summary

The authors used computer simulations to show that geometry matters.

1. Square walls = Normal physics (Gibbs Ensemble).
2. Circular walls = Weird physics (Generalized Gibbs Ensemble).

In the circular case, the particles refuse to settle down evenly. They cling to the walls, spin forever, and break the standard laws of thermal equilibrium. This suggests that if we want to simulate real-world systems (like granular materials or even certain quantum systems) accurately, we can't just use standard math; we have to account for the shape of the container and the "spin" of the particles.

The Takeaway: The shape of the box isn't just a container; it's an active participant that can change the fundamental laws of how the system behaves.

Technical Summary

1. Problem Statement

The paper addresses a fundamental question in statistical mechanics: how do confined systems of classical particles thermalize, and under what conditions do they deviate from the standard Gibbs ensemble?

• Context: Traditionally, the Gibbs distribution is derived under the assumption of the ergodic hypothesis, where energy is the sole conserved quantity.
• The Gap: While Generalized Gibbs Ensembles (GGE) are well-studied in quantum integrable systems, their emergence in simple classical systems is often overlooked. Specifically, the role of boundary geometry in inducing extra conservation laws (beyond energy) has been neglected.
• Hypothesis: The authors propose that for a system of hard disks confined within a circular boundary, the conservation of total angular momentum ($L$) alongside energy ($E$) prevents the system from reaching a standard Gibbs equilibrium. Instead, the system converges to a GGE characterized by a non-zero angular momentum, leading to a breakdown of ergodicity and time-reversal invariance.

2. Methodology

The authors employ a combination of analytical derivations based on the Maximum Entropy (MaxEnt) principle and numerical simulations using two distinct methods.

A. Analytical Framework (MaxEnt Principle)

• Constraints: The system is constrained by the conservation of total energy ($E$) and total angular momentum ($L_z$).
• Derivation: Using the MaxEnt principle, the authors maximize the entropy $S = -\int P \ln P$ subject to constraints on $E$ and $L_z$. This introduces two Lagrange multipliers: $\beta_T$ (inverse temperature) and $\beta_L$ (conjugate to angular momentum).
• Resulting Distribution: The derived probability distribution for a single particle is:
  $$P(x, p) \propto \exp\left(-\beta_T \frac{p^2}{2m} - \beta_L (p \wedge x)_z\right)$$
  This distribution is not time-reversal invariant because the angular momentum term changes sign under time reversal ($T: p \to -p$).

B. Numerical Simulations

The authors simulate a 2D system of $N$ hard disks (radius $r$, mass $m$) at low packing fractions ($\eta \approx 0.1$) to ensure the gas regime.

1. Event-Driven Molecular Dynamics (EDMD): Particles move in straight lines between instantaneous elastic collisions. This method is exact for hard disks.
2. Time-Driven Molecular Dynamics (TDMD): A standard integration scheme with small time steps ($\delta t$), allowing for slight overlaps to approximate hard-core potentials.
3. Boundary Conditions: Simulations were run with two distinct geometries:
   • Square Boundary: Standard reflecting walls.
   • Circular Boundary: Perfectly reflecting walls where the normal vector aligns with the radius.
4. Order Parameter ($\Xi$): A dimensionless parameter is introduced to quantify the degree of angular momentum conservation relative to total energy:
   $$\Xi = \frac{\|\sum x_i \wedge p_i\|^2}{2R^2 (\sum m_i)(\sum p_j^2 / 2m_j)}$$
   • $\Xi = 0$: Zero angular momentum (Gibbs regime).
   • $\Xi \to 1$: Maximized angular momentum (GGE regime).

3. Key Contributions

A. Discovery of Boundary-Induced GGE

The paper demonstrates that the geometry of the boundary dictates the asymptotic ensemble:

• Square Boundaries: Angular momentum is not conserved due to the lack of rotational symmetry in the boundary reflections. The system thermalizes to the standard Gibbs Ensemble regardless of initial angular momentum.
• Circular Boundaries: The geometry preserves the total angular momentum. The system fails to thermalize to the Gibbs distribution and instead converges to a Generalized Gibbs Ensemble (GGE).

B. Condensation Phenomenon

In the circular boundary case with high $\Xi$ (conserved angular momentum), the system exhibits a "condensation" near the boundary:

• Spatial Distribution: Particles accumulate near the wall ($r \to R$). In the limit $\Xi \to 1$, the radial distribution approaches a Dirac delta function at the boundary.
• Momentum Distribution: The momentum distribution shifts away from zero, peaking at higher momenta, and the velocity distribution becomes non-Gaussian.

C. Violation of Time-Reversal Invariance and Ergodicity

• The derived GGE distribution is explicitly not time-reversal invariant due to the linear dependence on angular momentum in the exponent.
• The system exhibits ergodicity breaking: the asymptotic state depends on the initial conditions (specifically the initial angular momentum), which is trapped by the boundary geometry.

D. Modification of Monte Carlo Algorithms

The authors show that standard Metropolis-Hastings algorithms (which assume Gibbs distributions) fail to reproduce the correct equilibrium state for circular boundaries. They propose a Modified Metropolis Algorithm (MMA) that includes angular momentum in the acceptance probability:
$$P_{acc} = \min\left(e^{-\beta_T \Delta E - \beta_L \Delta L}, 1\right)$$
This modification successfully recovers the GGE distribution predicted by the analytical theory.

E. Implications for the Bohr-van Leeuwen Theorem

The paper discusses a paradoxical violation of the Bohr-van Leeuwen theorem, which states that classical statistical mechanics cannot explain magnetism (average magnetization is zero).

• Since the GGE distribution is not time-reversal invariant, the average magnetic moment $\langle M \rangle$ is non-zero in the presence of conserved angular momentum.
• This suggests that classical systems with specific symmetries and boundary constraints can exhibit intrinsic magnetic properties, challenging the traditional interpretation of the theorem.

4. Key Results

• Analytical Distributions:
  • Radial distribution: $P(r) \propto r e^{\gamma r^2}$ (where $\gamma$ depends on $\Xi$). As $\Xi \to 1$, $P(r) \to \delta(1-r)$.
  • Momentum distribution: Deviates from Maxwell-Boltzmann, showing a peak at non-zero momentum.
• Simulation Verification:
  • Square Boundary: Both EDMD and TDMD confirm uniform spatial distribution and Maxwell-Boltzmann velocity distribution, even with initial angular momentum ($\Xi \approx 0.6$).
  • Circular Boundary: Both methods confirm the condensation phenomenon. Particles cluster at the boundary, and velocity distributions are non-centered.
  • Eccentricity Test: When the circular boundary is perturbed into an ellipse (eccentricity $\epsilon > 0$), the order parameter $\Xi$ decays exponentially over time, and the system eventually returns to the Gibbs ensemble. This confirms that the GGE state is fragile and relies on perfect rotational symmetry.
• Equation of State: The presence of angular momentum modifies the ideal gas law. The pressure $P$ increases with angular momentum:
  $$PV = N k_B T \left(1 + \frac{V m \beta_L^2}{4\pi \beta_T}\right) + O(\beta_L^4)$$

5. Significance and Conclusion

• Fundamental Physics: The work challenges the universality of the Gibbs ensemble in classical mechanics, showing that boundary conditions can act as a "control knob" to select between different statistical ensembles (Gibbs vs. GGE).
• Granular Matter & Simulations: It highlights that standard simulation methods (like standard Metropolis MC) may yield incorrect results for systems with specific symmetries (e.g., circular granular containers) if conservation laws other than energy are ignored.
• Ergodicity: It provides a clear, simple example of "hard" ergodicity breaking in a classical many-body system driven solely by geometry, without the need for integrability in the bulk interactions.
• Future Directions: The authors suggest extending these results to higher dimensions (spheres, cylinders) and investigating the impact on phase transitions (e.g., liquid-hexatic transitions) in hard disk systems, where the GGE might alter critical densities.

In summary, the paper establishes that boundary shape is a critical determinant of thermalization, capable of inducing a Generalized Gibbs Ensemble with conserved angular momentum, leading to unique physical phenomena like boundary condensation and non-zero classical magnetization.