Quantum error correction (QEC) does not have a single, universal energy cost per logical qubit. The expense depends on the error-correcting code, the number of physical qubits required, how often syndromes are measured, and the classical processing and cryogenic infrastructure that supports the quantum hardware. Fundamental thermodynamic limits set a tiny baseline, while practical systems incur costs many orders of magnitude higher.
Sources of energy cost
The unavoidable theoretical lower bound for any irreversible bit operation is given by the Landauer limit, formulated by Rolf Landauer IBM. At room temperature this limit is roughly 3 × 10^-21 joules per bit erased; at cryogenic temperatures used for many qubit technologies the bound falls proportionally with temperature, for example to around 4 × 10^-23 joules per bit at 4 kelvin. These values are minimum thermodynamic costs and do not capture real hardware overhead.
Practical QEC energy consumption comes from three major components: repeated syndrome measurements performed on physical qubits, classical decoding and control electronics that compute corrections, and maintaining the cryogenic environment. Theoretical and engineering work on QEC overhead by John Preskill Caltech and on the surface code by Austin Fowler University of Melbourne explain that the multiplicative factor of physical qubits per logical qubit and the required syndrome measurement rate directly scale the classical and refrigeration load.
Estimates and bounds
Error-correction schemes such as the surface code typically demand hundreds to thousands of physical qubits for one robust logical qubit under realistic noise rates, as discussed by Austin Fowler University of Melbourne and reviewed by John Preskill Caltech. Multiplying syndrome operations by classical decoding steps means the energy per logical qubit is the product of many small operations plus the continuous power for cryogenics and room-temperature control equipment. Industry and academic reports indicate that these supporting systems, not the quantum bits themselves, often dominate total power use in current prototypes, so practical energy per logical qubit is many orders of magnitude above fundamental limits.
The consequences are significant: scaling QEC affects the environmental footprint, capital and operational costs, and geographic deployment choices according to local electricity carbon intensity. For equitable access and sustainable deployment, researchers and engineers must optimize codes, low-power classical decoding, and cryogenic efficiency alongside improving qubit fidelity.