🔬 mesoscale physics

Temporal hopping dynamics in exciton-polariton condensation

This study reveals that exciton-polariton condensation near the threshold is not a static phase transition but a dynamical process characterized by stochastic hopping between condensed and non-condensed states, where coherence progressively builds up despite strong temporal fluctuations.

Original authors: Elena Rozas, Wojciech Bukalski, Yannik Brune, Adbhut Gupta, Kirk Baldwin, Loren N. Pfeiffer, Hassan Alnatah, Jonathan Beaumariage, David W. Snoke, Paolo Comaron, Marzena H. Szymanska, Marc Aßmann

Published 2026-04-29

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Elena Rozas, Wojciech Bukalski, Yannik Brune, Adbhut Gupta, Kirk Baldwin, Loren N. Pfeiffer, Hassan Alnatah, Jonathan Beaumariage, David W. Snoke, Paolo Comaron, Marzena H. Szymanska, Marc Aßmann

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a crowded dance floor where people (particles of light and matter called polaritons) are trying to decide whether to dance together in perfect unison or just wander around randomly. Usually, scientists think of this decision as a simple "on/off" switch: once you turn up the music (pump power) enough, everyone suddenly starts dancing in sync, forming a condensate.

This paper, however, reveals that the moment right before everyone starts dancing is much more chaotic and interesting than we thought. Instead of a smooth transition, the system goes through a phase of "temporal hopping."

Here is a simple breakdown of what the researchers found:

1. The Setup: A Special Dance Floor

The scientists used a tiny, high-tech "microcavity" (a sandwich of mirrors) to trap these particles. To keep them organized, they used a laser to create a ring-shaped trap (like a hula hoop on the floor). This forces the particles to gather in the center of the ring, away from the messy edges.

2. The "Hopping" Phenomenon

When they slowly turned up the laser power to the exact point where the particles should start dancing in sync (the threshold), something strange happened. The particles didn't just settle down immediately. Instead, they kept hopping back and forth between two states:

State A (The Party): All the particles are in the center, dancing in perfect rhythm (a condensate).
State B (The Crowd): The particles scatter, stop dancing together, and wander around the ring (non-condensed).

This wasn't a steady state. It was an intermittent flicker. The system would form a perfect condensate for a split second, then lose it, then form it again, over and over. It's like a lightbulb that is flickering rapidly between "on" and "off" right before it finally stays lit.

3. Measuring the Flicker

To see this happening, the team used a special camera technique called homodyne detection. Think of this as a super-sensitive microphone that listens to the "beat" of the light.

They measured how many photons (light particles) were in the trap at any given millisecond.
They also measured a value called $g^{(2)}(0)$ . In simple terms, this number tells you how "organized" the light is.
- A high number means the light is chaotic (like a crowd shouting).
- A number close to 1 means the light is perfectly organized (like a choir singing in unison).

4. The Big Surprise: Order in Chaos

The most exciting discovery was what happened to the "organization" number ( $g^{(2)}(0)$ ) while the system was still hopping.

Even though the condensate was appearing and disappearing (hopping), the "organization" number was slowly getting lower, moving toward 1.
The Analogy: Imagine a group of people trying to start a synchronized wave in a stadium. At first, they are just standing around. Then, for a few seconds, a small group does the wave, then they stop, then another group tries. Even though the wave keeps stopping and starting, the quality of the wave gets better and better each time it happens.
This proved that coherence (the ability to dance in sync) can build up even while the system is unstable. You don't need a perfectly stable party to start having a synchronized dance; you can build the rhythm even while the music is skipping.

5. Computer Simulations

The researchers built a computer model to see if this was real or just a glitch in their equipment. They simulated the particles with all the natural "noise" and randomness included. The computer model showed the exact same hopping behavior.

This confirmed that the hopping isn't caused by a broken machine or outside interference.
It is an intrinsic feature of the physics itself. The system naturally wants to hop between states right at the edge of condensation because of the delicate balance between energy coming in and energy leaking out.

Summary

In the past, scientists thought that once a polariton condensate formed, it was a stable, solid thing. This paper shows that right at the edge of formation, the system is actually a jittery, hopping mess.

However, even in this messy, flickering state, the particles are slowly learning to coordinate. They are building a "perfect dance" one hop at a time, eventually settling into a stable, synchronized state once the power is turned up just a little bit higher. This gives us a new, dynamic way to understand how order emerges from chaos in the quantum world.

1. Problem Statement

Macroscopic coherence in driven-dissipative systems, such as exciton-polariton condensates, is traditionally characterized by time-averaged observables (e.g., spectral linewidth narrowing and intensity saturation). These static measurements implicitly assume that once a condensate forms, it exists in a dynamically stable phase. However, near the condensation threshold ( $P_{th}$ ), driven-dissipative systems are highly susceptible to fluctuations and instabilities due to the fragile balance between gain and loss.

The core problem addressed is the lack of understanding regarding the temporal dynamics of condensate formation near the threshold. Specifically, it is unknown whether the transition to a coherent state is a smooth, static process or if it involves transient, stochastic instabilities where the condensate intermittently forms and disappears. Existing time-integrated measurements often mask these rapid fluctuations, leading to an incomplete picture of non-equilibrium phase transitions in open quantum systems.

2. Methodology

The authors employed a combined experimental and theoretical approach to probe the system on timescales faster than the intrinsic dynamics of the condensate.

Experimental Setup:

Sample: A high-quality factor ( $Q \approx 300,000$ ) GaAs microcavity containing 12 quantum wells, yielding a polariton lifetime of $\sim 300$ ps.
Confinement: An annular (ring-shaped) optical trap was created using a spatial light modulator (SLM) and a non-resonant continuous-wave (CW) laser. This geometry decouples the condensate from the incoherent exciton reservoir, minimizing dephasing and allowing polaritons to relax to the trap center.
Detection: The team utilized time-resolved homodyne detection. By interfering the emitted signal with a stable resonant local oscillator (LO) on a balanced detector, they measured a single field quadrature.
Data Acquisition: They collected 30,000 quadrature samples per window (approx. 40 $\mu$ s) to reconstruct the photon number distribution and the second-order correlation function, $g^{(2)}(0)$ , with high temporal resolution.
Metrics:
- Photon Number: Tracked to observe intensity fluctuations.
- $g^{(2)}(0)$ : Used to quantify the degree of coherence (where $g^{(2)}(0)=1$ indicates a coherent state and $g^{(2)}(0)=2$ indicates a thermal state).
- Image Difference ( $I_d$ ): A metric defined as the normalized root-mean-square difference between consecutive real-space images to quantify temporal instability.

Theoretical Framework:

Model: The system was modeled using the Truncated Wigner Approximation (TWA).
Equations: A stochastic partial differential equation (SPDE) for the polariton field $\psi(r,t)$ coupled to a rate equation for the exciton reservoir $n_R(r)$ .
Noise: The model includes intrinsic quantum noise via a complex Gaussian stochastic term ( $dW_c$ ) representing coupling to the environment and reservoir interactions.
Simulation: Numerical simulations were performed on a discretized grid to reproduce the experimental conditions, tracking the evolution of the field and calculating $g^{(2)}(0)$ using a time-averaging strategy to mimic experimental shot-to-shot analysis.

3. Key Contributions

Discovery of Temporal Hopping: The paper identifies a distinct dynamical regime near the condensation threshold characterized by stochastic hopping. The system does not transition smoothly to a stable condensate; instead, it undergoes irregular, non-periodic switching between low-occupation (non-condensed) and high-occupation (condensed) states.
Decoupling Coherence from Stability: The authors demonstrate that macroscopic coherence (indicated by $g^{(2)}(0) \to 1$ ) can build up progressively even while the system remains temporally unstable. This challenges the conventional view that a stable stationary state is a prerequisite for coherence.
Intrinsic vs. Extrinsic Fluctuations: Through TWA simulations, the study proves that these hopping dynamics are intrinsic to the driven-dissipative nature of polariton condensation (arising from noise and reservoir interactions) rather than being caused by extrinsic experimental noise.
Methodological Advance: The combination of ring-trapping with high-speed homodyne detection provides a new window into the transient dynamics of open quantum systems, moving beyond static threshold criteria.

4. Key Results

Instability Window: Both experiments and simulations reveal an instability window extending from approximately $0.95 P_{th}$ to $1.10 P_{th}$ (experiment) and $0.95 P_{th}$ to $1.20 P_{th}$ (simulation). Within this range, the "Image Difference" metric ( $I_d$ ) is high, indicating strong temporal fluctuations.
Photon Number Dynamics:
- Below Threshold ( $< P_{th}$ ): No condensate emission is detected (signal indistinguishable from vacuum noise due to blueshift).
- Near Threshold ( $1.00 - 1.10 P_{th}$ ): The photon number exhibits irregular, non-periodic fluctuations. The condensate forms for short durations (e.g., 10 ms) and then disappears, reflecting intermittent switching.
- Above Threshold ( $> 1.10 P_{th}$ ): The hopping dynamics vanish, and the photon number stabilizes, indicating a permanent condensate phase.
Evolution of Coherence ( $g^{(2)}(0)$ ):
- In the hopping regime, $g^{(2)}(0)$ shows a gradual reduction from thermal values ( $\approx 2$ ) toward unity ($1.0$), despite the strong intensity fluctuations.
- At $1.06 P_{th}$ , $g^{(2)}(0)$ drops to $\approx 1.19$ , signaling the progressive build-up of coherence even as the system hops between states.
- At high powers ( $1.35 P_{th}$ ), $g^{(2)}(0)$ saturates at $\approx 1.06$ , confirming a highly coherent, stable state.
Simulation Validation: The TWA simulations successfully reproduced the experimental hopping dynamics and the specific evolution of $g^{(2)}(0)$ , confirming that the observed phenomena are governed by the fundamental stochastic equations of the driven-dissipative field.

5. Significance

Redefining Phase Transitions: This work shifts the perspective of polariton condensation from a static phase transition to a dynamical process. It highlights that near the threshold, the system exists in a metastable, fluctuating state where coherence is established intermittently.
Fundamental Physics: It provides a unified picture linking photon statistics, coherence, and stability in non-equilibrium systems, demonstrating that macroscopic coherence can emerge from strong temporal fluctuations.
Technological Implications: Understanding these intrinsic noise-driven dynamics is crucial for the development of quantum technologies based on polaritons (e.g., polariton lasers, simulators, and logic gates). It suggests that noise and nonlinearity can be harnessed to engineer novel dynamical states rather than merely being viewed as detrimental factors.
General Applicability: The findings offer a framework for studying non-equilibrium phase transitions in other open quantum systems where time-resolved access to fluctuations is possible.