⚛️ quantum physics

All steerable quantum correlations can provide thermodynamic advantages in cooling

This paper demonstrates that all steerable quantum correlations provide a provable thermodynamic advantage in cooling tasks compared to unsteerable classical correlations, with the maximum advantage quantified by the steerability robustness, thereby establishing thermodynamic performance as a witness for steerability.

Original authors: Tanmoy Biswas, Chandan Datta, Luis Pedro Garcia-Pintos

Published 2026-09-16

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Tanmoy Biswas, Chandan Datta, Luis Pedro Garcia-Pintos

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

Computing, whether it runs on silicon chips in a laptop or on delicate quantum states in a laboratory, is fundamentally a process of managing heat. Every time information is processed or a calculation is completed, energy is converted into waste heat. In the quantum world, this heat is particularly dangerous because even the slightest thermal noise can destroy the fragile superpositions and entanglements that make quantum computers powerful. To keep these machines running, scientists must constantly cool them down, often to temperatures near absolute zero. The challenge is not just removing this heat, but doing so efficiently. If the cooling process itself requires too much energy or creates too much disorder, the entire system becomes impractical. This is where the nature of the connection between different parts of a quantum system becomes critical. Researchers have long known that quantum particles can be linked in ways that have no classical equivalent, allowing them to influence one another instantly across distances. One specific type of this connection, known as steering, allows one observer to effectively "steer" the state of a distant particle by choosing how to measure their own. While this phenomenon has been studied for its potential in secure communication and random number generation, its role in thermodynamics has remained largely unexplored.

A team of researchers has now demonstrated that this specific quantum connection, steering, can be used as a fuel to cool a system more effectively than any classical link ever could. In their work, they designed a theoretical cooling task that mimics the operation of a microscopic refrigerator. Imagine two scientists, Alice and Bob, who are far apart from each other. They share a pair of quantum particles that are linked together. Alice performs a measurement on her particle, and the result of that measurement instantly determines the state of Bob's particle, even though she has not touched it. Bob then uses this newly prepared state of his particle to interact with a heat bath, a reservoir of thermal energy. By carefully timing how he changes the energy levels of his particle and allowing it to settle, Bob can extract heat from the bath. The researchers found that when Alice and Bob share a quantum link that possesses the property of steering, Bob can pull significantly more heat out of the bath than if they were sharing a link that could be explained by classical physics alone.

The study proves that this advantage is not a minor fluctuation but a fundamental feature of the quantum world. For every scenario where the quantum link allows for steering, there exists a specific cooling protocol where the quantum method outperforms the best possible classical method. The researchers quantified this advantage by comparing the amount of heat removed in the quantum scenario against the amount removed in the classical scenario. They showed that the ratio of these two amounts is always greater than one whenever steering is present. In fact, the size of this advantage is directly tied to a mathematical measure of how strong the steering is. The stronger the steering, the more heat can be removed. This relationship is so precise that the amount of extra cooling achieved can serve as a witness, a clear signal that the system is indeed using quantum steering. If the cooling advantage is greater than a certain threshold, it is mathematically impossible for the system to be operating with classical correlations alone.

The researchers also explored how this advantage changes as the size of the quantum system grows. They constructed a specific example using particles that exist in a high-dimensional space, rather than just the simple two-state systems often used in basic experiments. They found that as the dimension of the system increases, the cooling advantage grows as well. In systems with a large number of possible states, the quantum method can remove a vastly larger amount of heat compared to the classical limit. This suggests that as quantum technologies scale up to handle more complex information, the benefits of using these steering correlations for cooling will become even more pronounced. The work establishes that steering is not just a curiosity of quantum mechanics but a genuine thermodynamic resource. It provides a new way to think about cooling, suggesting that the very act of measuring one part of a quantum system can be harnessed to drive a refrigerator, offering a path toward more efficient thermal management in future quantum devices.

The protocol proposed by the team relies on standard operations that are already within the reach of current experimental technology. It involves rapidly changing the energy settings of a particle, a process known as a quench, and then allowing the particle to interact with a heat bath until it reaches a stable temperature. These steps are similar to those used in other microscopic engines and refrigerators. Because the method does not require exotic new equipment, but rather the clever use of existing quantum links, it could be tested in laboratories using superconducting circuits, trapped ions, or even the nuclear spins of defects in diamond crystals. The findings suggest that the future of quantum computing may depend not just on building better qubits, but on understanding how to best exploit the unique ways in which those qubits can be connected to one another to manage the heat they generate. By proving that steering offers a provable thermodynamic edge, the researchers have opened a new chapter in the study of quantum thermodynamics, linking the abstract concept of non-local influence directly to the practical task of cooling.

Problem Statement
The removal of heat generated during computation is a fundamental challenge for both classical and quantum information processing. In quantum systems, this is particularly critical due to the need to reset qubits to pure states before computation and the degradation of superposition and entanglement by thermal noise. While previous studies have explored thermodynamic signatures of quantum correlations in work extraction and anomalous heat flow, there remains a need to identify specific operational tasks where quantum correlations provide a provable advantage over classical correlations in the context of cooling. Specifically, the paper addresses whether "steerability"—a specific type of quantum correlation—can serve as a genuine resource to enhance heat withdrawal from a thermal bath compared to scenarios restricted to unsteerable (classical) correlations.

Methodology
The authors propose a cooling protocol inspired by the microscopic Otto engine and refrigerator, but adapted to utilize shared quantum correlations as fuel. The setup involves two spatially separated parties, Alice and Bob:

Resource Preparation: Alice and Bob share a bipartite quantum state $\hat{\rho}_{AB}$ . Alice performs a measurement $M_x$ (chosen from a set of inputs $x$ ) on her subsystem and communicates the outcome $(a, x)$ to Bob via a classical channel. This process generates an "assemblage" of conditional states $\{\hat{\rho}_{a|x}\}$ on Bob's side.
Cooling Cycle: Upon receiving $(a, x)$ $(a, x)$ , Bob executes a three-step process:
- Quench: He changes his system's Hamiltonian from $H=0$ to a specific Hamiltonian $H_{a|x}$ .
- Thermalization: He allows his system to irreversibly thermalize with a heat bath at inverse temperature $\beta$ , reaching a thermal state $\hat{\gamma}_{a|x}$ . The heat exchanged during this step is $Q_c = \text{Tr}(H_{a|x}\hat{\gamma}_{a|x}) - \text{Tr}(H_{a|x}\hat{\rho}_{a|x})$ .
- Restoration: He quenches the Hamiltonian back to $H=0$ without interacting with the bath.
Comparison Framework: The authors compare the average heat withdrawn ( $Q_c$ ) when Bob uses assemblages derived from steerable quantum states versus those derived from unsteerable (classical) states. To ensure a fair comparison, they impose that both scenarios yield the same probability distribution $p(a|x)$ for the outcomes.
Quantification: The advantage is quantified by the ratio $\xi_{max}$ of the heat withdrawn using a steerable assemblage to the optimal heat withdrawn using any compatible unsteerable assemblage. This is related to the steerability robustness ( $R$ ), a geometric measure defined as the minimal noise required to make a steerable assemblage unsteerable.

Key Contributions and Results

Provable Thermodynamic Advantage: The central result (Theorem 1) establishes that for any steerable assemblage, there exists a choice of Hamiltonians such that the heat withdrawn exceeds that achievable by any unsteerable assemblage. Specifically, the ratio of quantum to classical heat withdrawal is strictly greater than 1 and is lower-bounded by the steerability robustness: $\xi_{max} > 1 + R$ .
Thermodynamic Witness of Steerability: The authors demonstrate that the quantity $\xi_{max}$ itself serves as a necessary and sufficient witness for steerability. If $\xi_{max} > 1$ , the assemblage is steerable; if $\xi_{max} \le 1$ , it is unsteerable. This provides the first operational interpretation of steerability robustness within a thermodynamic setting.
Dimensional Scaling: By analyzing assemblages generated from Mutually Unbiased Bases (MUBs) on isotropic entangled states, the authors show that the quantum advantage scales with the Hilbert space dimension $d$ . For maximally entangled states ( $\eta=1$ ), the advantage scales as $O(\sqrt{d})$ , representing an unbounded quantum advantage as the system dimension increases.
Temperature Dependence: The derived bounds indicate that the quantum advantage is more prominent at lower bath temperatures.

Significance and Claims
The paper claims to establish a direct link between the geometric measure of steerability (robustness) and a thermodynamic performance metric (heat extraction). The significance lies in:

Resource Identification: Identifying steerability as a distinct resource that enables superior cooling performance compared to classical correlations, distinct from entanglement or Bell nonlocality in this specific context.
Operational Interpretation: Providing a concrete operational task (cooling) that quantifies steerability robustness, moving beyond abstract geometric definitions.
Experimental Feasibility: The protocol relies only on standard operations—quenching and thermalization—which are accessible in current quantum platforms such as superconducting qubits, trapped ions, and Nitrogen-Vacancy (NV) centers.
Scalability: The demonstration that the advantage grows with system dimension suggests potential utility for scalable quantum technologies where efficient cooling is a bottleneck.

The authors explicitly state that they do not account for the cost of preparing the initial shared state $\hat{\rho}_{AB}$ , following the standard approach in quantum advantage studies where the focus is on the enhancement provided by the quantum resource itself. They also clarify that the protocol is not a closed thermodynamic cycle in the traditional sense, as the "fuel" (the shared state) is not returned to its initial state, but rather the protocol is restarted by Alice's measurements.

More like this