Brownian motion: non-equilibrium states from equilibrium trajectories -- recovering hydrodynamic regimes from prepared displacement measurements

This paper demonstrates that analyzing the second moments of a single equilibrium Brownian trajectory allows for the recovery of non-equilibrium hydrodynamic regimes, revealing that short-time displacement statistics are governed by correlated thermal-hydrodynamic forces and follow a $t^4$ scaling at very short times, superseding the previously established $t^{5/2}$ law.

The Gist

Imagine you are watching a single, tiny speck of dust floating in a glass of water. It jitters and jiggles randomly, a dance caused by invisible water molecules bumping into it. This is Brownian motion.

For a long time, scientists have studied this dance by watching the particle settle into a calm, steady rhythm (equilibrium). But this paper proposes a clever trick: you can learn about the chaotic, "out-of-control" moments of the particle just by carefully picking apart its calm, steady dance.

Here is the breakdown of the paper's ideas using simple analogies:

1. The "Movie Reel" Trick (Equilibrium vs. Non-Equilibrium)

Think of the particle's steady, random movement as a long movie reel of a busy city street.

• The Old Way: Scientists usually just watched the whole movie to see the average traffic flow.
• The New Way: The authors say, "Wait! If we pause the movie at a specific moment and say, 'Let's pretend this exact moment is the very beginning of a new story,' we can see what happens next."

By taking a snapshot of the particle when it happens to be at a specific spot with zero speed (a "Z-preparation") and watching how it moves from there, they can uncover hidden details about the water's behavior that are usually invisible. It's like realizing that every calm moment in a storm contains the blueprint for the next gust of wind.

2. The "Speed Limit" of the Water

The paper focuses on how fast the particle moves in the very first split-second after that "pause."

• The Old Belief: Scientists thought that in liquids, the particle's movement followed a specific rule (a $t^{5/2}$ law) caused by the water's "inertia" (its resistance to changing motion, like a heavy truck taking time to stop). This is similar to the Basset force, a drag effect that lingers.
• The New Discovery: The authors found that if you look really closely at the very beginning, before the water's "heavy truck" inertia kicks in, the movement follows a different, faster rule (a $t^4$ law).

The Analogy: Imagine pushing a heavy shopping cart.

• The $t^4$ Law: This is the split second before the wheels even start to roll, when you are just applying force. The movement is smooth and predictable because the force you apply is "correlated" (it doesn't jump around wildly).
• The $t^{5/2}$ Law: This is the moment the wheels start to spin and the cart's weight (inertia) fights back. This happens slightly later.

The paper argues that for a tiny fraction of a second, the "smooth push" ($t^4$) dominates before the "heavy inertia" ($t^{5/2}$) takes over.

3. The "Roughness" of the Dance

The paper connects how the particle moves to how "rough" or "smooth" its path is.

• Imagine drawing the path of the particle on a piece of paper.
• If the path is very jagged and fractal (like a lightning bolt), it means the particle is changing direction wildly.
• If the path is smoother, it means the particle's speed changes more gently.

The authors show that by measuring how the particle's position changes in those first few moments, you can calculate the "roughness" of its speed.

• If the movement follows the $t^4$ rule, the speed is very smooth (like a car on a highway).
• If it follows the $t^{5/2}$ rule, the speed is a bit rougher (like a car hitting bumps).

4. Why This Matters (Without the Hype)

The paper doesn't claim this will cure diseases or build new engines. Instead, it offers a new microscope for fluid dynamics.

By using this "pause and restart" method on a single particle, scientists can now:

1. Distinguish between different types of fluids: Is the liquid acting like a simple water (Newtonian) or a thick, gooey fluid (viscoelastic)? The "early seconds" of the particle's dance tell the story.
2. Check the math: It confirms that the "heavy inertia" effects (Basset force) are real, but it also shows that there is an even earlier, smoother phase of motion that was previously missed because it happens so fast.

Summary

The paper is like finding a secret code in a calm river. By stopping the river at a specific point and watching how a leaf moves immediately after, you can learn about the water's hidden properties (like its thickness and how it resists motion) that you couldn't see just by watching the river flow calmly. It reveals that the very first instant of movement is smoother and more predictable than we thought, before the water's "weight" starts to drag on the particle.

Technical Summary

Technical Summary: Brownian Motion: Non-equilibrium States from Equilibrium Trajectories

Problem Statement
The paper addresses the challenge of characterizing the fine structure of fluid-particle interactions at microscales, specifically within the context of Brownian motion in liquid media. While recent high-resolution experiments have confirmed hydrodynamic theories regarding inertial effects (such as the Basset force) at short time scales, fundamental questions remain regarding the regularity of particle velocity realizations and the precise scaling of mean square displacement (MSD) in the very short-time limit. Traditional analyses often focus on equilibrium second-order moments. However, the authors posit that equilibrium trajectories contain embedded information about non-equilibrium states that, if properly extracted, can reveal distinct hydrodynamic regimes and the regularity properties (Hölder continuity) of particle velocity that are otherwise obscured.

Methodology
The study employs a theoretical framework grounded in the Chapman-Kolmogorov equation for Markovian dynamics, extended to non-Markovian cases via Markovian embedding. The core methodology involves:

1. Subsampling of Equilibrium Trajectories: The authors propose that any equilibrium trajectory can be viewed as a superposition of uncountable non-equilibrium states. By "subsampling" an equilibrium trajectory—specifically selecting segments where the particle position and velocity are initially zero (a "Z-preparation")—one can reconstruct the propagator $p(x, t | x_0, 0)$ and analyze the subsequent non-equilibrium relaxation.
2. Generalized Langevin Equation (GLE) and Modal Expansion: The hydrodynamics of the solvent are modeled using a GLE with memory kernels representing dissipative ($h(t)$) and fluid-inertial ($k(t)$) effects. To handle singularities (like the $1/\sqrt{t}$ Basset kernel) and ensure physical consistency (finite propagation speeds), the authors utilize a modal representation (Prony series expansion) of these kernels. This transforms the non-Markovian GLE into a system of coupled Markovian Langevin equations involving auxiliary "hidden" variables (memory modes).
3. Moment Graph Analysis: A topological approach is used to analyze the hierarchy of second-order moments. By constructing an oriented graph where nodes represent moments and edges represent integration steps from constant forcings, the authors derive the initial time-scaling exponents of the MSD ($m_{xx}(t) = \langle x^2(t) \rangle$) based on the number of integration steps required to reach the position moment.
4. Correlated Fluctuational Patterns: The study extends the standard fluctuation-dissipation theory by introducing "correlated fluctuational patterns," where stochastic forcings possess finite correlation times rather than being purely $\delta$-correlated, to account for the finite propagation velocity of hydrodynamic fluctuations.

Key Contributions and Results

• Recovery of Hydrodynamic Regimes: The analysis demonstrates that the initial scaling of the MSD, $m_{xx}(t) \sim t^\phi$, in Z-prepared conditions serves as a signature of the underlying hydrodynamic regime.

  • Einstein-Langevin (Stokesian) & Finite Inertial Kernels: When fluid-inertial effects are negligible or the inertial kernel is regular (finite at $t=0$), the initial scaling is $\phi = 3$ ($m_{xx}(t) \sim t^3$).
  • Singular Basset Kernel: In the presence of a strictly singular Basset kernel ($k(t) \sim t^{-1/2}$), the scaling transitions to $\phi = 5/2$ ($m_{xx}(t) \sim t^{5/2}$), confirming previous findings by Boynewicz et al. (2026).
  • Correlated Fluctuations: When the stochastic forcings are modeled with finite correlation times (correlated fluctuational patterns), the initial scaling is found to be $\phi = 4$ ($m_{xx}(t) \sim t^4$). This scaling supersedes the $t^{5/2}$ law at very short times if the regularity of the velocity is considered.

• Connection to Velocity Regularity: A central result is the derivation of a direct relationship between the MSD scaling exponent $\phi$ and the Hölder exponent $H_v$ of the particle velocity realizations:
  $$H_v = \frac{\phi - 2}{2}$$
  This implies:

  • For $\phi = 3$ (Stokesian/Finite Inertia), $H_v = 1/2$, corresponding to a fractal dimension $d_v = 3/2$.
  • For $\phi = 5/2$ (Singular Basset), $H_v = 1/4$, corresponding to $d_v = 7/4$.
  • For $\phi = 4$ (Correlated/Viscoelastic), $H_v = 1$, corresponding to a Lipschitz continuous velocity ($d_v = 1$).

• Resolution of Singularities: The paper clarifies that the $t^{-3/2}$ singularity in the global memory kernel derived from the Basset force is a mathematical artifact of assuming infinite propagation velocity. When viscoelastic effects (finite relaxation times) or finite correlation times are included, the kernel becomes regular at $t=0$, leading to the $t^4$ scaling initially, followed by a crossover to the $t^{5/2}$ regime at intermediate times.

• Anomalous Diffusion: The framework is extended to anomalous diffusion scenarios, including power-law memory kernels and diffusion in fractal geometries (e.g., Sierpinski carpets), showing that short-time dynamics remain governed by the regularity of the velocity and the initial conditions, distinct from long-term anomalous scaling.

Significance and Claims
The authors claim that this work provides a theoretical foundation for extracting non-equilibrium dynamics from equilibrium data, a capability rooted in the Chapman-Kolmogorov equation. The significance lies in:

1. Experimental Accessibility: The proposed "Z-preparation" (selecting trajectory segments with zero initial position and velocity) is experimentally feasible using high-resolution Brownian motion data, offering a new method to probe fluid properties without external perturbations.
2. Universality Classes: The study defines distinct universality classes for Brownian motion based on the scaling exponent $\phi$, linking macroscopic transport properties to the microscopic regularity of velocity fluctuations.
3. Theoretical Consistency: By incorporating correlated fluctuational patterns and finite propagation speeds, the paper resolves inconsistencies in the standard GLE formulation regarding the long-term behavior of singular kernels and the physical realizability of the fluctuation-dissipation relations.
4. Validation: The results are presented as consistent with recent experimental observations of the $t^{5/2}$ law in liquids and the fractal dimension of velocity in liquids, while predicting a $t^4$ regime that may be observable at time scales currently beyond experimental resolution or in systems with significant viscoelasticity.

The paper concludes that the regularity of Brownian motion is not merely a mathematical curiosity but a physical quantity comparable in relevance to standard moments, connecting equilibrium properties (velocity autocorrelation, fractal dimension) directly to non-equilibrium scaling laws derived from subsampled trajectories.