⚛️ quantum physics

Permanent Thermodynamic Mpemba Inversion from Quantum-Jump Feedback

This paper demonstrates that a quantum-jump feedback protocol can permanently reverse the nonequilibrium free-energy ordering of two initial states by exploiting collective channel compensation to eliminate the slow-sector contribution of the higher-energy state, thereby achieving a permanent thermodynamic Mpemba inversion while preserving Gibbs equilibrium.

Original authors: Jincheng Lu, Chen Wang

Published 2026-10-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Jincheng Lu, Chen Wang

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

In the quiet world of thermodynamics, there is a rule that seems unbreakable: hot things cool down, and cold things warm up, until they both reach the same temperature as their surroundings. This journey toward balance is called relaxation, and for over a century, scientists have understood it as a slow, steady march where the starting point determines the pace. If you begin with a system that is very far from equilibrium, it generally takes longer to settle than one that starts closer to the finish line. This intuition is so deeply rooted that it feels like a law of nature, much like gravity pulling an apple to the ground. Yet, nature occasionally offers a surprise. In the 1960s, a student named Erasto Mpemba noticed that under certain conditions, hot water could freeze faster than cold water, a counterintuitive phenomenon that has since been observed in many different systems, from simple fluids to complex quantum materials. This "Mpemba effect" suggests that the history of a system matters less than how its internal parts are arranged, allowing a distant state to overtake a closer one on the road to equilibrium.

For a long time, researchers believed this effect was a fleeting trick of the initial conditions, a temporary crossover that eventually gave way to the standard order of things. However, a new study by Jincheng Lu and Chen Wang challenges this view by showing that the order of relaxation can be permanently flipped. They demonstrate that by using a specific type of quantum control, it is possible to take a system that starts with more energy and disorder than another, and force it to stay ahead of the slower system forever. This is not a temporary race where the lead changes hands; it is a fundamental rewriting of the rules of thermal relaxation, ensuring that the initially "hotter" state remains thermodynamically superior after a specific, finite time has passed.

The researchers achieved this by working with a tiny quantum system, specifically a three-level atom known as a qutrit, which acts as a miniature laboratory for testing these ideas. In the natural world, such a system would interact with its environment, absorbing and emitting energy in random bursts called quantum jumps. These jumps are the mechanism by which the system loses its excess energy and settles into a calm, stable state. The team introduced a feedback loop that watches for these jumps in real time. Whenever a jump is detected, the system receives an immediate, precise nudge—a pulse of energy designed to steer its path. The crucial innovation lies in how these nudges are applied. Instead of trying to fix the system after every single jump, the researchers designed the feedback so that the corrections from different types of jumps cancel each other out when the system is perfectly at rest. This ensures that the stable, balanced state remains untouched. However, when the system is out of balance, the jumps do not happen at the same rate, and the cancellation is incomplete. This leaves a net effect that actively reshapes how the system relaxes.

By carefully tuning the timing and strength of these feedback pulses, the team found they could eliminate the slowest, most stubborn part of the relaxation process for one specific starting state, while leaving that slow part intact for the other. Imagine two runners starting a race where one is heavily burdened by a heavy backpack that slows them down. The feedback acts like a guide who, upon seeing the burdened runner stumble, instantly removes the backpack for them, while the other runner keeps theirs. In this quantum scenario, the "backpack" is a specific pattern of internal motion that takes a long time to fade away. The feedback protocol successfully strips this pattern away from the initially more energetic state, allowing it to sprint toward equilibrium using only the fast, efficient pathways. The other state, lacking this intervention, is forced to drag its slow, heavy pattern along, taking much longer to finish.

The result is a permanent inversion of the thermodynamic order. The state that started with more free energy, which would normally take the longest to cool down, ends up reaching the stable state faster and stays there. The researchers proved that this crossover is not a temporary glitch but a permanent feature of the system's future. They calculated that the ratio of the remaining energy in the two states would eventually vanish, meaning the initially hotter state would remain strictly "cooler" in a thermodynamic sense for all time after a specific crossing point. This effect holds true even when the monitoring equipment is not perfect. The study shows that as long as the detection of jumps is efficient enough—specifically, if the system detects at least about 72 percent of the jumps—the feedback can be adjusted to maintain this permanent advantage through exact cancellation. Below this threshold, the exact elimination of the slow pattern is no longer possible, yet the permanent inversion can still survive if the remaining slow patterns happen to favor the initially hotter state.

This work does more than just demonstrate a faster way to cool a quantum system; it reveals a new way to control the very nature of time and order in the microscopic world. By using the information gained from watching a system change, researchers can now dictate not just how fast it relaxes, but the fundamental hierarchy of its relaxation. The study confirms that with the right feedback, the initial conditions of a system do not have to dictate its final fate. Instead, a common set of rules can be applied to different starting points to ensure that the one that starts furthest from the goal is the one that arrives first, and stays there. This discovery opens the door to designing quantum devices that can be reset or stabilized with unprecedented speed and precision, turning the counterintuitive Mpemba effect from a curious anomaly into a reliable tool for engineering.

Technical Summary: Permanent Thermodynamic Mpemba Inversion from Quantum-Jump Feedback

Problem Statement
The Mpemba effect describes a phenomenon where a system initially further from thermal equilibrium can relax to equilibrium faster than a system initially closer to it. While this effect has been observed in classical and quantum systems through initial-state preparation or Liouvillian engineering, a fundamental challenge remains: can a common feedback rule reverse the thermal relaxation order of two fixed quantum states while strictly preserving their Gibbs equilibrium?

Standard approaches face limitations. Unitary preparation suppresses slow modes for a specific initial state but does not apply a common rule to different preparations. Liouvillian engineering often alters the stationary state or requires distinct generators for different trajectories. Furthermore, in thermal relaxation, a state may retain coherence and excess free energy even after populations reach Gibbs values. A control protocol must therefore manage both population imbalance and coherence to reverse the ordering of nonequilibrium free energy ( $\Delta F_\beta$ ) without shifting the thermal fixed point.

Methodology
The authors propose a quantum-jump feedback protocol that utilizes collective compensation between monitored channels. The system evolves under a Davies generator $L_0$ coupled to a thermal reservoir. Upon detection of a jump in channel $k$ , a completely positive and trace-preserving operation $F_k(\theta)$ is immediately applied.

The core mechanism relies on the following principles:

Gibbs Preservation: The feedback parameters $\theta$ are constrained to a set $G_\beta$ such that the collective feedback corrections cancel out at the Gibbs state $\tau_\beta$ ( $\sum_k [F_k(\theta) - I] J_k[\tau_\beta] = 0$ ). This ensures $\tau_\beta$ remains the unique stationary state, even though individual channels may produce non-zero corrections away from equilibrium.
Selective Slow-Sector Elimination: The protocol aims to eliminate the projection of the initially higher-free-energy state ( $\rho_h$ ) onto the complete slow sector of the controlled generator $L_{fb}(\theta)$ , while retaining a non-zero projection for the comparison state ( $\rho_c$ ).
Spectral Control: By adjusting feedback parameters (specifically mixing angles and phases in a qutrit realization), the authors manipulate the overlap between initial states and the eigenmodes of the generator. This allows the "slow" decay modes (those with the smallest magnitude real eigenvalues) to be suppressed for $\rho_h$ but not for $\rho_c$ .

Key Contributions and Results

Theoretical Proof of Permanent Inversion: The paper proves that if the slow-sector contribution of $\rho_h$ is completely eliminated ( $P_s[\rho_h - \tau_\beta] = 0$ ) while $\rho_c$ retains a slow component ( $P_s[\rho_c - \tau_\beta] \neq 0$ ), the free-energy ratio $\Delta F_\beta(\rho_h(t)) / \Delta F_\beta(\rho_c(t))$ vanishes asymptotically. This guarantees a permanent thermodynamic Mpemba inversion: after a finite time $t_{per}$ , the initially "hotter" (higher free energy) state remains permanently below the "colder" state in terms of free energy, despite both approaching the same equilibrium.
Qutrit Realization: The mechanism is explicitly realized using a thermal qutrit (three-level ladder system) with two monitored emission channels.
- The system utilizes destructive interference between initial coherence and coherence generated by population imbalance.
- By tuning feedback angles, the authors demonstrate that the coherence sector can be made the slowest mode.
- At a specific operating point, the slow-mode amplitude for $\rho_h$ is exactly cancelled, while $\rho_c$ (which has thermal populations and no population-induced coherence) remains unaffected by the feedback's population dynamics, preserving its slow decay.
Detection Efficiency Thresholds: The study analyzes the impact of finite detection efficiency ( $\eta$ $η$ ). It identifies two distinct thresholds:
1. Spectral Selection Threshold ( $\eta_{sp}$ ): The minimum efficiency required to make the coherence sector the slowest mode.
2. Exact Cancellation Threshold ( $\eta_c$ ): The minimum efficiency required to achieve exact selective elimination of the slow sector for $\rho_h$ .
- The paper shows that while exact cancellation requires $\eta \ge \eta_c$ (approx. 0.72 in the model), permanent inversion can persist below this threshold if the residual slow amplitude still favors $\rho_h$ , though the asymptotic decay rates become identical.
Robustness and Retuning: The authors demonstrate that feedback parameters can be "retuned" to compensate for deviations in detection efficiency, restoring the exact cancellation condition.

Significance and Claims
The paper claims to establish a sufficient condition for permanent thermodynamic Mpemba inversion under a common controlled dynamics. Unlike previous demonstrations where the inversion might be transient or dependent on specific initial preparations, this protocol ensures that the ordering is reversed and maintained indefinitely as $t \to \infty$ .

Key distinctions highlighted by the authors include:

Collective Compensation: The feedback preserves the Gibbs state not by leaving individual channels inert, but through a collective cancellation of corrections. This allows the feedback to remain dynamically active during relaxation.
Thermodynamic Measure: The inversion is defined via the nonequilibrium free energy (incorporating both entropy and energy), providing a rigorous thermodynamic measure rather than just a population-based metric.
Experimental Feasibility: The authors identify superconducting circuits as a viable platform for implementation, citing existing capabilities in qutrit control, real-time feedback, and itinerant microwave-photon detection. They note that the required reservoir engineering and detection efficiencies are within the realm of current experimental parameters, provided the specific balance of rates and feedback latency can be managed.

The work concludes that collective Gibbs-preserving feedback offers a robust method for controlling the asymptotic ordering of thermal relaxation, enabling the acceleration of a specific state relative to another without altering the final equilibrium state.

More like this