A new theoretical cooling task has been developed to exploit nonlocal correlations to draw heat from a bath, revealing an advantage of quantum over classical resources.
For all their usefulness, quantum systems are demanding. Prized features like superposition and entanglement are easily washed away by interactions with the environment, so one has to work at low temperatures to preserve them. Quantum computation produces extra heat, which must be quickly ferried away. How can we make this cooling process more efficient? According to a recent paper in Physical Review Letters, the solution may lie in quantum physics itself.
An international team of researchers, led by Tanmoy Biswas at Germany’s University of Cologne, has devised a theoretical cooling task in which an experimenter sacrifices an assemblage – a set of labelled quantum states – to withdraw heat from a bath. Although the states must have finite dimension, their form is otherwise unrestricted. They might, for instance, be nuclear spins in nitrogen–vacancy centres in diamond.
Every state in the ensemble is distinct, so each can be given a distinctive label. According to the label of a chosen state, the experimenter drives the energy of the system to a specified level. The quantum state, seeking thermal equilibrium, is obliged to soak up heat from the bath. The state is then discarded (it is not used again throughout the cooling cycle), and the cooling cycle begins afresh with a new state. Not all assemblages are up to the task, however. Those that cool effectively share a common feature: they can be used to demonstrate a form of nonlocality known as quantum steering.
Steering from afar
To understand this suppose an entangled state is shared by two distant parties. Quantum steering occurs when the outcome of one party’s measurement on their part of the state influences the post-measurement state held by the other. Originally suggested by Erwin Schrödinger in 1935, this nonlocal phenomenon is fundamentally nonclassical. Indeed, to demonstrate it we need shared entanglement and incompatible measurements, which are two quintessentially quantum resources.
In the team’s setup, the distant party, Alice, randomly chooses a measurement to make on her part of the entangled state – she might roll some dice first to decide which one. Unbeknownst to the experimenter, Bob, his system assumes a particular form as a result. Alice then acts as a virtual adviser: she contacts Bob, perhaps by phone, to reveal her measurement choice and outcome. Taken together, these specify Bob’s post-measurement state and tell him what to do. Somewhat counterintuitively, although Alice has no access to the bath, she still helps prepare the assemblage used to cool it.
As the team proves, assemblages that demonstrate quantum steering provide a strict advantage in the cooling task over those that do not, which exhibit only classical correlations.
“Our goal,” says Biswas, “was not simply to demonstrate an advantage for a particular quantum state, but to establish a universal connection between quantum steering and thermodynamic performance.”
From steering to thermodynamics
Beyond its foundational appeal, quantum steering has many applications: it can be used to power heat engines, certify entanglement, and even generate randomness. By contrast, the cooling task considered here is not cryptographic (neither Alice nor Bob care about the secrecy of their protocols), and entangled states participate only indirectly.
Quantum-scale thermodynamics offers a tighter definition of entropy
The researchers now hope their results will find use in current experimental platforms for quantum computation, such as trapped ions and superconducting qubits.
While the advantage of using quantum resources has been established, the question of which specific operations and assemblages are optimal for this task remains open. In considering quantum correlations together with thermodynamics, however, this task brings us a step closer to reconciling these strands of physics. According to Biswas, the most interesting insight from their work is “that every steerable quantum correlation, no matter how weak, can provide a genuine thermodynamic advantage in cooling. This establishes a direct connection between a fundamental feature of quantum mechanics and a practical thermodynamic task.”
