In theory, quantum physics could harvest energy without entropy

A team at City University of Hong Kong set out to prove that a certain kind of energy transfer could not work. It ended up finding a quantum way to make it work. Led by Oscar Dahlsten, associate professor in the Department of Physics, the group reports the result in Physical Review Letters, where the paper is titled "Nonunique Decompositions of Mixed States and Deterministic Energy Transfers".

Energy from natural sources often arrives with entropy, the physicists' word for randomness. That randomness is the problem: it makes the output hard to predict, and hard to turn into the steady power needed to charge a battery or run an electronic device. The study argues that a particular feature of quantum theory, the fact that a mixed state can be decomposed in more than one way, allows deterministic energy harvesting, or DEH. In this scheme, energy passes from source to harvester with no matching transfer of entropy.

The mechanism works like this. A quantum harvester can be charged along several paths, each passing through a different intermediate superposition state, a combination of states that quantum theory allows at once. As long as the source drives the harvester into one of those transitions, the harvester reaches its charged state with certainty. Different source states could therefore, the authors argue, deliver energy in a predictable way.

"We initially set out to show that such a protocol might be impossible," Dahlsten said. "To our surprise, our efforts led us to a quantum solution that could achieve exactly the deterministic energy transfer we were looking for."

The result is a theoretical foundation, not a device. The preprint appeared on arXiv in 2025 and the journal version in 2026, and the study describes no prototype and no measurement. The team gives no date for a working component, and the step from equations to hardware is not described. The authors present the framework as a route to converting ambient energy into usable power with minimal loss.

For the devices the authors name, the framework is aimed at internet of things sensors, wearables, smart-city sensors and 6G devices that could draw power from ambient sources. The framework also points to a possible way to cut the fundamental losses of conventional AC-to-DC rectifiers, the circuits that turn the alternating current of the grid or of a radio signal into the direct current electronics use. Whether that gain holds outside a paper is the next test, and no timeline has been set for it.