The stochastic discrete nonlinear Schrödinger equation: microscopic derivation and finite-temperature phase transition

This paper presents a first-principles derivation of a stochastic one-dimensional discrete nonlinear Schrödinger equation that satisfies statistical mechanics principles, revealing a finite-temperature phase transition and coarsening dynamics reminiscent of stochastic resonance, thereby offering a pathway to experimentally realize negative-temperature phase transitions in systems coupled to positive-temperature heat baths.

The Gist

Imagine a long line of people holding hands, each holding a small, glowing lantern. In the world of physics, this line represents a Discrete Nonlinear Schrödinger Equation (DNSE). Usually, these lanterns are perfectly synchronized, passing energy back and forth in a predictable, mathematical dance. This is the "deterministic" version: if you know the starting position, you know exactly where the light will be a second later.

But in real life, nothing is perfect. There is always wind, bumps, and random jolts. This paper asks: What happens if we add a little bit of "wind" (random noise) to this line of lanterns?

The authors, Mahdieh Ebrahimi, Barbara Drossel, and Wolfram Just, built a new model to study this. They didn't just guess; they derived their model from the ground up using the laws of thermodynamics (the rules of heat and energy). Here is the story of what they found, explained simply.

1. The Setup: A Line of Lanterns in a Storm

Think of the system as a chain of $L$ lanterns.

• The Rules: The lanterns are connected. If one gets bright, it can make its neighbors bright.
• The "Wind" (Noise): The authors added a "heat bath" to the system. Imagine the whole line is standing in a gentle, random breeze. Sometimes the wind pushes a lantern up; sometimes it pushes it down. This represents the stochastic (random) part of their equation.
• The Constraint: No matter how much the wind blows, the total amount of light in the system stays exactly the same. You can't create new light or destroy it; you can only move it around.

2. The Big Discovery: A "Freezing" Phase Transition

In physics, a phase transition is like water turning into ice. It's a sudden change in how a system behaves.

The researchers found that this line of lanterns has two distinct "moods" depending on the temperature (how hot or cold the "wind" is):

• The Hot/Chaotic Mood (High Temperature): When the "wind" is strong (high temperature), the lanterns are jittery and random. The light is spread out evenly across the whole line. Everyone is dancing chaotically. This is the disordered phase.
• The Cold/Focused Mood (Low Temperature): When the "wind" is gentle (low temperature), something magical happens. The light suddenly stops spreading out. Instead, it collapses into one single, super-bright lantern (or a very small group of them). The rest of the line goes almost dark. This is the localised phase (or a "breather").

The Analogy: Imagine a crowded room where everyone is chatting randomly (Hot). Suddenly, the room gets quiet, and all the attention focuses on one person telling a story, while everyone else stops talking (Cold). The energy has "condensed" into one spot.

3. The "Negative Temperature" Twist

Here is where it gets weird. In this specific model, the "Cold" state actually corresponds to what physicists call Negative Temperature.

• Normal Physics: Usually, "hot" means high energy and "cold" means low energy.
• This Model: Because the total light is fixed, there is a maximum amount of energy the system can hold. Once you hit that ceiling, adding more "heat" actually makes the system behave like it's "colder" in a mathematical sense.
• The Paper's Trick: The authors showed that by flipping a switch (changing a parameter called $\alpha$), they could make this "Negative Temperature" phenomenon happen at positive temperatures. It's like finding a way to make ice melt when you turn up the heat, but in a controlled, mathematical way.

4. The "Goldilocks" Noise Effect (Stochastic Resonance)

One of the most surprising findings was about the strength of the noise (the wind).

Usually, you'd think:

• No wind = The system gets stuck.
• Too much wind = The system gets too chaotic to organize.
• Just the right amount of wind = The system organizes itself fastest.

The researchers found that if they started with the light spread out and wanted it to collapse into a single bright spot, there was a "Goldilocks" level of noise.

• If the wind was too weak, the lanterns moved too slowly to find each other.
• If the wind was too strong, it kept knocking the light apart.
• At a medium, optimal wind speed, the lanterns found each other and collapsed into a bright spot faster than at any other noise level.

This is called Stochastic Resonance. It's like trying to push a heavy swing. If you push too hard or too soft, it doesn't go high. But if you push with just the right rhythm and force, it soars.

5. Why Does This Matter?

You might ask, "Who cares about a line of mathematical lanterns?"

1. Real-World Applications: This model describes real things like:
   • Optical fibers: Where light pulses travel through glass.
   • Bose-Einstein Condensates: Super-cold clouds of atoms.
   • DNA: How energy moves through the double helix.
2. Experimental Proof: For a long time, scientists could only see these "phase transitions" in computer simulations or in systems that were perfectly isolated (which is hard to do in real life). This paper provides a blueprint for how to see these transitions in a real lab, even if the system is connected to a warm environment (a "heat bath").
3. Understanding Order from Chaos: It shows how randomness (noise) and order (structure) can work together to create stable, focused patterns.

Summary

The paper tells the story of a line of energy that can either be a chaotic mess or a focused beam. By adding a specific type of "wind" (noise) and controlling the temperature, the authors showed that the system can suddenly snap from chaos to order. They also discovered that a little bit of chaos is actually necessary to help the system organize itself quickly—a bit like how a little bit of stress can sometimes help a team focus and get things done faster.

They proved this using a mix of heavy math (to ensure the laws of physics weren't broken) and computer simulations (to watch the lanterns dance), providing a clear map for how to find these transitions in the real world.

Technical Summary

1. Problem Statement

The paper addresses the dynamics of the one-dimensional Discrete Nonlinear Schrödinger Equation (DNSE), a Hamiltonian system known for exhibiting localized modes (breathers) and negative temperature states. While the thermodynamics of the isolated DNSE (microcanonical ensemble) are well-studied, the behavior of the system when coupled to a heat bath (canonical ensemble) remains less understood, particularly regarding:

• How to rigorously derive a stochastic version of the DNSE that satisfies statistical mechanics principles (detailed balance, H-theorem).
• Whether a phase transition between disordered and localized phases occurs at finite positive temperatures when the system is coupled to a thermal bath.
• How the system relaxes to equilibrium and how noise strength influences this non-equilibrium coarsening dynamics.

2. Methodology

A. Microscopic Derivation of the Stochastic Model

The authors derive a Stochastic Discrete Nonlinear Schrödinger Equation (SDNSE) from first principles by coupling the DNSE Hamiltonian ($H_S$) to a microscopic heat bath of harmonic oscillators.

• Derivation: Using projection operator techniques and eliminating bath degrees of freedom, they obtain a Langevin-type equation.
• The Equation: The resulting stochastic differential equation (SDE) is:
  $$\dot{c}_k(t) = i\frac{\partial H_S}{\partial \bar{c}_k} - \frac{\beta \sigma^2}{2} c_k(t) \left( \bar{c}_k \frac{\partial H_S}{\partial \bar{c}_k} - c_k \frac{\partial H_S}{\partial c_k} \right) + i\sigma c_k(t) \xi_k(t)$$
  where $\xi_k(t)$ is Gaussian white noise, $\sigma$ is the noise strength, and $\beta$ is the inverse temperature.
• Constraints: The model strictly preserves the normalization $N = \sum |c_k|^2 = 1$, a crucial constant of motion for the DNSE.
• Properties: The derivation ensures the system obeys detailed balance and an H-theorem, guaranteeing relaxation to a constrained canonical equilibrium distribution $\rho_\beta \propto e^{-\beta H_S} \delta(N-1)$, valid for both positive and negative $\beta$.

B. Numerical Simulations

• Integration Scheme: The authors employ a specialized numerical integration scheme (Appendix D) that preserves the normalization $N=1$ and reduces to a symplectic scheme in the limit of zero noise. This avoids numerical artifacts common in standard SDE solvers.
• Parameters: Simulations were performed for system sizes $L=32, 64, 128, 1024$, varying the nonlinearity strength $\alpha$, inverse temperature $\beta$, and noise strength $\sigma$.
• Observables: Key metrics included mean energy $E(\beta)$, energy variance (susceptibility), inverse participation ratio (localization measure), and autocorrelation functions.

C. Analytical Approach

• Mean-Field Theory: A simplified mean-field model was developed to predict the phase transition. It assumes a symmetry-broken state where one site holds a dominant weight ($a$) while others share the remainder ($b$).
• Linear Limit: Exact analytical results for the caloric equation of state were derived for the linear case ($\alpha=0$) to serve as a baseline.

3. Key Contributions and Results

A. Finite-Temperature Phase Transition

The study confirms a first-order phase transition in the canonical ensemble at positive temperatures (for attractive nonlinearity $\alpha > 0$).

• Phases:
  • High-Temperature (Disordered) Phase: Characterized by random fluctuations and delocalized energy.
  • Low-Temperature (Localized) Phase: Characterized by the formation of discrete breathers (energy concentrated on a single site).
• Control Parameter: The transition is governed by a rescaled inverse temperature variable:
  $$\bar{\beta} = \frac{\alpha \beta}{2L}$$
  The critical point occurs at $\bar{\beta}_c \approx 2.45$.
• Symmetry: The system exhibits a symmetry $(\alpha, \beta) \to (-\alpha, -\beta)$. This implies that the phase transition observed at positive $\beta$ with attractive nonlinearity is mathematically equivalent to the negative-temperature transitions reported in previous microcanonical studies of repulsive nonlinearity.

B. Coarsening Dynamics and Stochastic Resonance

A significant finding concerns the transient dynamics of the system approaching equilibrium:

• Non-Monotonic Noise Dependence: The time required for the system to form localized structures (from random initial conditions) or decay them (from localized initial conditions) is non-monotonic with respect to the noise strength $\sigma$.
• Optimal Noise: There exists an optimal noise strength that minimizes the relaxation time. This phenomenon is reminiscent of stochastic resonance, where noise aids the system in overcoming energy barriers or navigating the phase space more efficiently than pure deterministic dynamics or high-noise diffusion.

C. Analytical Validation

The mean-field theory successfully predicts:

• The critical value $\bar{\beta}_c \approx 2.46$.
• The nature of the transition (first-order jump in the order parameter).
• The critical energy line $E_c(\alpha)$ in the phase diagram, which aligns well with numerical simulation data and extends previous microcanonical results to the canonical ensemble.

4. Significance and Implications

• Bridging Ensembles: The paper successfully translates the concept of negative-temperature phase transitions (traditionally studied in isolated systems) into a finite-temperature canonical framework. This demonstrates that such transitions are accessible in systems coupled to positive-temperature heat baths, provided the nonlinearity sign is adjusted.
• Experimental Relevance: By showing that the phase transition occurs at finite, positive temperatures in a dissipative setup, the authors provide a theoretical pathway for observing these phenomena in real-world experiments (e.g., optical waveguide arrays or Bose-Einstein condensates), which are inherently coupled to thermal environments.
• Non-Equilibrium Physics: The discovery of the non-monotonic dependence of coarsening dynamics on noise strength highlights complex interactions between deterministic nonlinearities and stochastic forces, suggesting new avenues for controlling pattern formation and soliton dynamics via noise tuning.
• Theoretical Rigor: The derivation from a microscopic Hamiltonian ensures the stochastic model is physically consistent, satisfying fundamental statistical mechanics requirements (detailed balance, H-theorem) often overlooked in phenomenological stochastic models.

In summary, this work provides a rigorous microscopic foundation for the stochastic DNSE, uncovers a finite-temperature phase transition linked to negative-temperature physics, and reveals unexpected noise-induced acceleration in the formation of localized structures.