Running hot and cold: A Quantum Engine

Image courtesy: Magda Ehlers

This post is about the recent demonstration of a quantum heat engine, using a superconductor.

Heat Engines

The job of a heat engine is to change heat into useful mechanical work or motion. Now, in order to flow, heat requires the existence of a temperature difference. So typically the engine takes some heat from a hot place (at temperature Th), turns some of it into work, and then releases some heat into a cold place (at temperature Tc). The efficiency of the heat engine is given by 1-(Tc/Th). This formula, called the Carnot efficiency, was arrived at by using the laws of classical (i.e. non-quantum) physics.

Due to the second law of thermodynamics, which forbids the existence of absolute zero (i.e. Tc cannot be zero), no heat engine can be 100% efficient - therefore some heat is always wasted in the production of work.

In our daily lives heat engines are almost indispensable - they are used in cars, trucks, planes and thermal power stations, for example.

Enter the Quantum

The so-called first quantum revolution took place in the early 1900s, when the subject was initially discovered and subsequently fleshed out. A second quantum revolution is now underway, in which quantum applications to our lives are being realized and explored. The best known example of this is probably the quantum computer, which is expected to outperform existing classical computers.

Another such application being explored is a quantum heat engine. Practically, we are now able to manipulate small objects (like atoms) and very cold objects (like superconductors). At these sizes and temperatures, quantum physics becomes essential for describing the behavior of such systems. The question arises, if we want to make useful nanodevices using these platforms, as to how the laws of thermodynamics change when the engines follow the laws of quantum physics. Coming at this from an opposite direction, machines are becoming more and more miniaturized and devices are now so small that standard (classical) rules of heat and energy no longer apply accurately. So we need to figure out how thermodynamics works at these small dimensions.

Remarkably, it has been shown that using quantum sources of heat may allow to improve upon Carnot's formula and construct more efficient heat engines than are permitted classically. There is also interest in how fundamentally quantum capabilities like superposition and entanglement can improve the working of heat engines.

Considerations of this sort have led to a new area of physics, called quantum thermodynamics. Quantum thermodynamics is intimately inked to the operation of a quantum computer. It provides the physical rules for managing energy, suppressing thermal noise, and cooling individual qubits, which are essential processes for building stable, scalable quantum hardware.

Experimentally, quantum heat engines have been constructed using nanoparticles like atoms, ions, nuclei and well-isolated defects in materials. These platforms, however, are not easy to scale and the demonstrations were mostly proof of principle.

Superconductors

Some materials lose their resistance to current flow at low temperatures. These are called superconductors and their behavior can only be explained by quantum physics. Circuits can be made with these materials (Aluminum and Niobium are popular choices) to implement quantum logic. They can be scaled readily, so in fact superconducting circuits are a leading platform for the realization of a quantum computer.

Heat Engine in a Superconducting Circuit

The paper referred to at the beginning of this post realizes a cyclic quantum heat engine in a superconducting circuit [1]. Though this engine produces a very small amount of work (23 orders of magnitude smaller than required to push a car!), it could be scaled to affect the working of a quantum computer and to make nanodevices.

[1] Initial demonstration of a quantum heat engine based on dissipation-engineered superconducting circuits, T. Uusnäkki, T. Mörsted, W. Teixeira, M. Rasola and M. Möttönen, Nature Communications 17, 6054 (2026).