Quasiparticle dynamics and hydrodynamics of 1d hard rod gas on diffusion scale

This paper derives analytical expressions for the mean, variance, and autocorrelation of a quasiparticle in a one-dimensional hard rod gas, revealing that long-range initial correlations induce a diffusive-scale correction to the standard Euler generalized hydrodynamic equations that depends on the specific form of those correlations.

The Gist

The Big Picture: A Crowd of Bouncing Billiards

Imagine a long, narrow hallway filled with thousands of identical billiard balls (these are the "hard rods"). They are all rolling around, bouncing off the walls and, more importantly, bouncing off each other.

In physics, we usually try to predict how this crowd moves in two ways:

1. The "Euler" View (The Big Picture): We ignore the individual bounces and just look at the flow, like watching a river. This is called Generalized Hydrodynamics (GHD). It works great for predicting where the "center" of the crowd will be after a long time.
2. The "Diffusion" View (The Wobble): If you watch a single ball closely, you'll see it doesn't just move in a straight line. It gets bumped, jiggles, and wanders off its path. This wandering is called diffusion.

The Problem: For a long time, scientists thought they understood how to describe this "wandering" (diffusion) for these billiard balls. They had a standard formula (like the Navier-Stokes equations) that worked for normal fluids (like water). But for these special "hard rod" systems, that formula was failing.

The Discovery: The author of this paper, Anupam Kundu, found out why the standard formula fails and derived a new, corrected version. The key culprit? Long-Range Correlations.

---

The Core Concept: The "Whispering Crowd"

To understand the paper, you need to understand two different ways the billiard balls could start their journey:

Scenario A: The "Silent Room" (Standard Initial State)

Imagine the balls are placed in the hallway completely randomly, with no pattern. They don't know anything about their neighbors.

• What happens: As they move, they bounce. If you look at the crowd later, the only thing that matters is what's happening right next to a specific ball.
• The Result: The standard physics formulas work fine here. The "wandering" is predictable.

Scenario B: The "Whispering Room" (The New Initial State)

Now, imagine the balls are placed in a specific pattern where they "know" about each other across the room. Maybe the balls on the left are all moving slightly faster because the balls on the right told them to (metaphorically). This is a Long-Range (LR) Correlation.

• What happens: Even though the balls are far apart, their movements are linked. When a ball on the left bounces, it sends a "ripple" of information that affects a ball on the right much later.
• The Result: The standard formulas break! The "wandering" (diffusion) is different because the balls are influenced by the whole crowd, not just their immediate neighbors.

The "Quasiparticle" Detective

The author studies a single "tagged" ball (a quasiparticle).

• The Tag: Imagine you put a bright red sticker on one specific billiard ball.
• The Trick: In a hard rod gas, when two balls collide, they swap velocities. So, the "red sticker" (the identity of the quasiparticle) jumps from one ball to another!
• The Journey: The red sticker doesn't move in a straight line. It hops from ball to ball. Every time it hops, it gets pushed forward or backward by the length of the ball it just hit.

The author asked: "If I tag a ball, where will that tag be after a long time? How much will it wiggle?"

The Three Main Findings

1. The "Ghost" Correction:
   The author calculated the average position and the "wobble" (variance) of that red tag. He found that for the "Whispering Room" (Scenario B), there is an extra "ghost" force pushing the tag. This force comes entirely from those long-distance connections (correlations) between the balls.

   • Analogy: It's like walking through a crowd where people are holding hands across the room. If someone far away sneezes, the whole chain of hands shifts, and you get nudged, even though you didn't see the sneeze.

2. The New Equation:
   Because of this extra "nudge," the standard equation for how the crowd spreads out (the Diffusion equation) needs a new term.

   • The old equation said: "Spread depends on local density."
   • The new equation says: "Spread depends on local density PLUS how the density is changing across the whole room."
   • This new term is different depending on how the crowd was arranged at the start.

3. Two Different Rules for Two Different Starts:
   The paper proves that if you start with a "factorized" crowd (random, no long-range links), you get one type of diffusion. But if you start with a "correlated" crowd (links exist from the start), you get a different type of diffusion.

   • Metaphor: It's like driving a car. If you start on a smooth, empty highway, your fuel consumption follows one rule. If you start on a highway where everyone is already bumper-to-bumper and reacting to each other's brake lights from miles back, your fuel consumption follows a completely different rule.

Why Does This Matter?

This isn't just about billiard balls. This research helps us understand:

• Quantum Computers: Many quantum systems behave like these hard rods. If we want to build better quantum computers, we need to know exactly how information (heat, energy) spreads through them.
• New Materials: Understanding how particles move in "integrable" systems (systems with perfect order) helps us design materials with unique properties, like super-efficient energy transport.

Summary in One Sentence

The author discovered that when a crowd of particles starts with "long-distance friendships" (correlations), they don't just wander randomly like normal; they move in a coordinated, complex way that requires a brand-new mathematical rule to describe their spread.

Technical Summary

1. Problem Statement

The paper addresses the derivation of Generalized Hydrodynamics (GHD) for a one-dimensional gas of hard rods on the diffusive scale. While the Euler-scale GHD (ballistic transport) is well-established for integrable systems, the derivation of diffusive corrections (Navier-Stokes terms) is more complex.

• The Core Issue: Standard hydrodynamic derivations often rely on the Local Equilibrium (LE) assumption, which posits that correlations decay rapidly. However, in integrable systems like the hard-rod gas, Long-Range (LR) correlations emerge dynamically on the Euler space-time scale due to the coherent transport of initial fluctuations. These LR correlations invalidate the standard LE approximation, leading to diffusive corrections that differ from the standard Navier-Stokes form.
• The Gap: Previous microscopic derivations (e.g., by Kundu, Dhar, et al.) established these corrections for an initial state where the distribution is factorized in hard-rod coordinates (ICfhr). However, a rigorous derivation was missing for the initial state factorized in point-particle coordinates (ICfhp). In the ICfhp state, the hard rods possess inherent long-range correlations from $t=0$, unlike the ICfhr state where they are initially uncorrelated (except for the hard-core exclusion).

2. Methodology

The author employs a microscopic approach to derive the diffusive-scale hydrodynamic equations by analyzing the stochastic dynamics of quasiparticles.

• System Definition: A 1D gas of $N$ hard rods (mass 1, length $a$) moving on an infinite line. The system is mapped to a non-interacting hard-point gas (HPG) via a coordinate transformation ($x_i = X_i - (i-1)a$) to simplify dynamics, then mapped back to hard-rod coordinates.
• Initial Conditions:
  1. ICfhr: Factorized in hard-rod coordinates (Eq. 3).
  2. ICfhp: Factorized in point-particle coordinates (Eq. 5). The paper focuses primarily on deriving results for ICfhp.
• Quasiparticle Dynamics: A quasiparticle is defined as a tagged rod. Due to collisions, the tag jumps between rods, and the quasiparticle undergoes stochastic displacements. The author calculates the mean, variance, and autocorrelation of the quasiparticle position $X_q(t)$.
• Correlation Analysis:
  • The phase-space density correlations are decomposed into a singular part (Generalized Gibbs Ensemble, GGE) and a non-singular Long-Range (LR) part.
  • The author explicitly computes the LR correlations for the ICfhp initial state using "height field" variables ($\phi$ and $\varphi$) and their linearized fluctuations.
  • Key insight: The LR correlation function exhibits a discontinuity (jump) at equal positions ($X=Y$), which is crucial for the diffusive correction.
• Derivation of Hydrodynamics:
  • The evolution of the mean phase-space density $\bar{f}(X, v, t)$ is derived by expanding the displacement of a quasiparticle over an infinitesimal time $dt$.
  • The mean displacement includes a correction term $j_d$ arising from the LR correlations.
  • The variance of the displacement yields the diffusion coefficient.

3. Key Contributions

1. Explicit Derivation of LR Correlations for ICfhp: The paper provides the first analytical derivation of phase-space density correlations for the hard-rod gas starting from an initial state factorized in point-particle coordinates. It confirms that this state also possesses LR correlations with a specific jump structure at $X=Y$.
2. Unified Quasiparticle Dynamics: It demonstrates that the expressions for the mean, variance, and autocorrelation of a quasiparticle are universal for both ICfhr and ICfhp initial states, provided the specific structure of the LR correlations (singular + non-singular with a jump) is accounted for.
3. Diffusive Correction to GHD: The paper derives the specific form of the diffusive correction term for the ICfhp initial state. It proves that this correction is different from the standard Navier-Stokes term derived under LE assumptions and also different from the correction derived for the ICfhr state.
4. Cancellation Mechanism: It rigorously proves that for the ICfhp state, the "antisymmetric" part of the LR correction cancels exactly with the GGE contribution, leaving only a "symmetric" LR term to modify the hydrodynamic equation.

4. Key Results

A. Quasiparticle Statistics

For a quasiparticle starting at $X_q$ with velocity $v_q$, the mean position $\langle X_q(t) \rangle$ and variance are given by:

• Mean: $\langle X_q(t) \rangle = X^{eu}_t + \frac{1}{\ell} \int_0^t j_d(X^{eu}_s, v_q, s) ds + O(1/\ell^2)$
  • Where $X^{eu}_t$ is the Euler trajectory.
  • $j_d = j^{gge}_d + j^{lr}_d$ is the drift correction.
• Variance: $\langle (\Delta X_q(t))^2 \rangle = \frac{a^2}{\ell} \int_0^t D_s(X^{eu}_s, v_q) ds + O(1/\ell^2)$
  • $D_s$ is the diffusion coefficient derived from GGE correlations.
  • Crucially, the LR correlations do not contribute to the variance at leading order (Eq. 51d), only to the mean drift.

B. Hydrodynamic Equation on Diffusive Scale

The evolution of the mean phase-space density $\bar{f}(X, v, t)$ is governed by:
$$\partial_t \bar{f} + \partial_X (v^{eff} \bar{f}) = -\frac{1}{\ell} \partial_X \left[ j^{lr, sym}_d(X, v, t) \bar{f}(X, v, t) \right]$$

• $v^{eff}$: The effective velocity from Euler GHD.
• $j^{lr, sym}_d$: The symmetric part of the LR-induced drift.
  • For ICfhp, this term is explicitly derived (Eq. 81b/81c) and depends on the spatial gradients of the density $\partial_X \bar{f}$ and the integrated density $\bar{F}$.
  • This term is non-zero and distinct from the standard Navier-Stokes diffusion term.
  • For ICfhr, the form is different, confirming that the diffusive correction is sensitive to the specific nature of the initial long-range correlations.

C. Homogeneous Limit

For a homogeneous initial state, the LR drift terms vanish ($j^{lr}_d = 0$), and the system recovers the standard Navier-Stokes equation derived in previous works (e.g., [33, 42]), consistent with the fact that LR correlations are absent in homogeneous equilibrium.

5. Significance and Implications

• Beyond Local Equilibrium: The work reinforces the concept that integrable systems cannot be described by standard hydrodynamics based solely on local equilibrium. The history of the system (initial correlations) imprints itself on the diffusive scale dynamics.
• Universality vs. Specificity: While the structure of the quasiparticle dynamics (mean/variance formulas) is universal across different initial ensembles, the specific form of the hydrodynamic correction depends on the microscopic details of the initial state's correlations.
• Fluctuating Hydrodynamics: The results provide the necessary ingredients for constructing a fluctuating hydrodynamic theory for hard rods at the mesoscopic scale, bridging the gap between microscopic integrable dynamics and macroscopic diffusive behavior.
• Theoretical Rigor: By providing a microscopic derivation for the ICfhp case, the paper closes a gap in the literature, showing that even when initial correlations are "built-in" (as in point-particle factorization), the resulting hydrodynamics is modified in a calculable way, distinct from the "dynamically generated" correlations of the ICfhr case.

In summary, this paper establishes that diffusive corrections in integrable systems are not universal constants but are dynamically determined by the specific long-range correlation structure of the initial state, necessitating a revision of standard hydrodynamic equations for non-equilibrium integrable gases.