The 'lesser demon' engine of Seah, Nimmrichter, and Scarani [Phys. Rev. Lett. 124, 100603 (2020)] extracts work from a thermally driven qubit using a demon that can only interrogate a macroscopic pointer - a damped harmonic oscillator displaced to positions depending on the qubit state - so that measurement, feedback, and erasure costs all appear as explicit physical energy flows. The published theory rests on Born-Markov secular master equations and, for the passively driven variant, on weak driving, and the authors themselves acknowledge that the theory becomes unreliable at strong coherent driving, that the accounting of measurement-induced energetic costs and pointer erasure remains contentious, and that implementation is only sketched. This problem asks for the completion of that theory along three linked strands: (1) the steady-state power, efficiency, and energy flows of the qubit-pointer engine at strong interrogation, obtained with dynamics that remain valid there; (2) a principled assignment of measurement-backaction and pointer-reset costs to heat or work, resolving the contentious interpretation of these costs; and (3) a concrete realization path in a physical platform.

Background

Maxwell's demon is a hypothetical agent that measures a system and uses the acquired information to extract work, apparently in violation of the second law of thermodynamics; the paradox is resolved because erasing the demon's memory of one bit costs at least kBTln⁡2k_B T\ln 2 of heat (Landauer's principle). In quantum thermodynamics the measurement step itself has an energetic signature: measuring an observable that does not commute with the system Hamiltonian changes the system's mean energy, a contribution often called 'quantum heat', and this backaction can be used as a non-thermal resource to drive engines and refrigerators. A standing difficulty for such measurement-based machines is the bookkeeping: how the energy flows associated with the measurement apparatus should be split into heat and work, and how the genuine cost of resetting a realistic, macroscopic pointer compares with the ideal Landauer cost, remain contested topics across the trajectory-based, autonomous, and operational formulations of quantum thermodynamics.

Seah, Nimmrichter, and Scarani addressed these questions with a self-contained spin-boson engine (a qubit linearly coupled to a harmonic-oscillator mode and to thermal baths, with no external apparatus beyond the pointer itself). The working medium is a qubit of transition frequency Ω\Omega; a harmonic oscillator of frequency ω\omega acts as the pointer and couples to the qubit through the Hamiltonian

H^=ℏΩ2σ^z+ℏω(a^†a^+12)+ℏωx0σ^za^+a^†2,\hat{H}=\frac{\hbar\Omega}{2}\hat{\sigma}_z+\hbar\omega\left(\hat{a}^\dagger\hat{a}+\frac{1}{2}\right)+\hbar\omega x_0\hat{\sigma}_z\frac{\hat{a}+\hat{a}^\dagger}{\sqrt{2}},

so that the pointer's equilibrium position is displaced to ±x0\pm x_0 depending on the qubit state. A hot bath (rate κh\kappa_h, temperature ThT_h) randomly excites the qubit; a cold bath (rate κc\kappa_c, TcT_c) continuously thermalizes the pointer, thereby erasing the information stored in it. In the intended hierarchy Ω≫ω≫κc≫κh\Omega\gg\omega\gg\kappa_c\gg\kappa_h with the displacement x0x_0 exceeding the pointer's thermal width xthx_{\rm th} (ultrastrong qubit-oscillator coupling), the two displaced pointer states are spatially distinguishable, so reading the pointer position amounts to a projective measurement of the qubit. Two demon implementations were analyzed: an 'active' demon performing Poisson-distributed projective pointer measurements at rate γ\gamma followed by conditional spin flips, and a 'passive' demon consisting of a position-dependent coherent drive V^(t)=ℏζf(x^)e−i(Ω−Δ)t∣e⟩⟨g∣+h.c.\hat{V}(t)=\hbar\zeta f(\hat{x})e^{-i(\Omega-\Delta)t}|e\rangle\langle g|+\mathrm{h.c.} of strength ζ\zeta and detuning Δ\Delta. Key results: the engine can attain the benchmark power γWmax⁡\gamma W_{\max} (with Wmax⁡=ℏ(Ω−2ωx02)p∞W_{\max}=\hbar(\Omega-2\omega x_0^2)p_\infty per event, where p∞p_\infty is the steady-state excitation probability of the qubit) simultaneously with a benchmark efficiency that at vanishing κh\kappa_h approaches η≈1−4x02ω/Ω\eta\approx 1-4x_0^2\omega/\Omega, below the Otto efficiency 1−ω/Ω1-\omega/\Omega; it operates beyond the Otto window (the range nˉh>nˉc\bar{n}_h>\bar{n}_c of bath occupations in which a finite-dimensional engine with a qubit pointer functions) as long as x0≳xthx_0\gtrsim x_{\rm th}; at measurement rates γ>κc\gamma>\kappa_c the quantum Zeno effect (freezing of the dynamics by sufficiently frequent measurements) stops the engine; and the pointer-reset energy 2ℏωx02≥2ℏωxth2>4kBTc2\hbar\omega x_0^2\ge 2\hbar\omega x_{\rm th}^2>4k_BT_c always exceeds the ideal Landauer cost kBTcln⁡2k_BT_c\ln 2, so second-law compliance is maintained with a macroscopic pointer.

The authors state the limitations of this analysis explicitly. The derivations use weak-coupling master equations in the secular approximation (dropping non-resonant terms), and the passive scheme is treated only in the weak-driving limit ζ≪ω,Ω\zeta\ll\omega,\Omega; the caption of their Fig. 4 warns that the master-equation model 'may no longer be reliable for ζ∼ω\zeta\sim\omega'. The general treatment of measurement-induced energetic costs and of pointer erasure is described as remaining contentious. The efficiency can be defined either as η=W˙/(Q˙h+Q˙ba)\eta=\dot{W}/(\dot{Q}_h+\dot{Q}_{\rm ba}), counting the measurement-backaction flow Q˙ba\dot{Q}_{\rm ba} as an additional heat input, or as η=(W˙−Q˙ba)/Q˙h\eta=(\dot{W}-\dot{Q}_{\rm ba})/\dot{Q}_h, counting it as a work expense; the definitions coalesce in the regime considered because Q˙ba∝ω\dot{Q}_{\rm ba}\propto\omega whereas W˙,Q˙h∝Ω≫ω\dot{W},\dot{Q}_h\propto\Omega\gg\omega. Finally, experimental platforms (molecular vibrations, hybrid optomechanical systems, trapped ions) are only sketched, since realizing Ω≫ω≫κc≫κh\Omega\gg\omega\gg\kappa_c\gg\kappa_h together with x0>1x_0>1 and fast pointer relaxation is demanding.

Problem Statement

Complete the thermodynamic theory of the pointer-measurement-driven quantum engine of Seah, Nimmrichter, and Scarani, so that the engine's operation is described by dynamics valid where it actually has to work and evaluated with a cost bookkeeping that does not depend on interpretation. The common objective uniting the requested results is a self-contained account of a measurement-driven engine whose measurement apparatus is an explicit physical degree of freedom, with all measurement, feedback, and erasure costs appearing as physical energy flows. Concretely, an answer must resolve three linked strands:

  1. Strong-interrogation dynamics. Extend the steady-state analysis of the qubit-pointer engine beyond the weak-driving and secular approximations - to coherent drive strengths ζ\zeta comparable to the pointer frequency ω\omega for the passive demon (where the source's master equation is stated to be unreliable), and correspondingly to measurement rates γ\gamma approaching and exceeding the pointer damping κc\kappa_c for the active demon, where the coarse-grained description is untested and quantum-Zeno freezing sets in. Using dynamics that remain valid there, determine the steady-state power, efficiency, and energy flows W˙\dot{W}, Q˙h\dot{Q}_h, Q˙c\dot{Q}_c, Q˙ba\dot{Q}_{\rm ba}, and establish whether the engine's signature features survive: simultaneous attainment of power and efficiency benchmarks, operation beyond the Otto window, sub-Carnot compliance, and the fate of the maximum efficiency η≈1−4x02ω/Ω\eta\approx 1-4x_0^2\omega/\Omega.

  2. Cost accounting and efficiency definitions. Provide a principled assignment of the measurement-backaction flow Q˙ba\dot{Q}_{\rm ba} and the pointer erasure/reset cost (bounded below by 2ℏωx02>4kBTc2\hbar\omega x_0^2 > 4k_BT_c in this model, above the Landauer cost kBTcln⁡2k_BT_c\ln 2) to heat or work, resolving the contention noted in the source over the proper treatment of the erasure cost of pointers and the interpretation of incoherent measurement schemes as a form of heat and work exchange, and address whether the two efficiency definitions that include or exclude Q˙ba\dot{Q}_{\rm ba} continue to coalesce beyond the regime of operation considered in the source.

  3. Physical realization. Identify a concrete platform (molecular, optomechanical, trapped-ion, or other) in which the required hierarchy Ω≫ω≫κc≫κh\Omega\gg\omega\gg\kappa_c\gg\kappa_h, pointer displacement x0x_0 at least of order the thermal width xthx_{\rm th}, and fast pointer relaxation can be realized, turning the source's qualitative sketch into a concrete realization proposal for the engine.

The three strands jointly support one objective: a measurement-driven engine theory whose apparatus costs are physical rather than phenomenal, whose dynamical regime is the one in which it is claimed to operate, and whose performance claims are grounded in an implementable platform.

References

  1. S. Seah, S. Nimmrichter, and V. Scarani, 'Maxwell's lesser demon: A quantum engine driven by pointer measurements', Physical Review Letters 124, 100603 (2020). DOI: 10.1103/PhysRevLett.124.100603; https://arxiv.org/abs/1908.10102
  2. C. Elouard, D. A. Herrera-Martí, M. Clusel, and A. Auffèves, 'The role of quantum measurement in stochastic thermodynamics', npj Quantum Information 3, 9 (2017). DOI: 10.1038/s41534-017-0008-4; https://doi.org/10.1038/s41534-017-0008-4
  3. C. Elouard, D. Herrera-Martí, B. Huard, and A. Auffèves, 'Extracting work from quantum measurement in Maxwell's demon engines', Physical Review Letters 118, 260603 (2017). DOI: 10.1103/PhysRevLett.118.260603; https://doi.org/10.1103/PhysRevLett.118.260603
  4. R. Landauer, 'Irreversibility and heat generation in the computing process', IBM Journal of Research and Development 5, 183 (1961). DOI: 10.1147/rd.53.0183; https://doi.org/10.1147/rd.53.0183
  5. R. Alicki and M. Horodecki, 'Information-thermodynamics link revisited', Journal of Physics A: Mathematical and Theoretical 52, 204001 (2019). DOI: 10.1088/1751-8121/ab076f; https://doi.org/10.1088/1751-8121/ab076f
  6. P. Strasberg, 'Operational approach to quantum stochastic thermodynamics', Physical Review E 100, 022127 (2019). DOI: 10.1103/PhysRevE.100.022127; https://doi.org/10.1103/PhysRevE.100.022127
  7. D. von Lindenfels, O. Gräb, C. T. Schmiegelow, V. Kaushal, J. Schulz, M. T. Mitchison, J. Goold, F. Schmidt-Kaler, and U. G. Poschinger, 'Spin heat engine coupled to a harmonic-oscillator flywheel', Physical Review Letters 123, 080602 (2019). DOI: 10.1103/PhysRevLett.123.080602; https://doi.org/10.1103/PhysRevLett.123.080602
  8. S. K. Manikandan, C. Elouard, K. W. Murch, A. Auffèves, and A. N. Jordan, 'Efficiently fueling a quantum engine with incompatible measurements', Physical Review E 105, 044137 (2022). DOI: 10.1103/PhysRevE.105.044137; https://doi.org/10.1103/PhysRevE.105.044137
  9. J. Stevens, D. Szombati, M. Maffei, C. Elouard, R. Assouly, N. Cottet, R. Dassonneville, Q. Ficheux, S. Zeppetzauer, A. Bienfait, A. N. Jordan, A. Auffèves, and B. Huard, 'Energetics of a single qubit gate', Physical Review Letters 129, 110601 (2022). DOI: 10.1103/PhysRevLett.129.110601; https://arxiv.org/abs/2109.09648
  10. V. F. Lisboa, P. R. Dieguez, J. R. Guimarães, J. F. G. Santos, and R. M. Serra, 'Experimental investigation of a quantum heat engine powered by generalized measurements', Physical Review A 106, 022436 (2022). DOI: 10.1103/PhysRevA.106.022436; https://arxiv.org/abs/2204.01041
  11. K. Wang, R. Xia, L. Bresque, and P. Xue, 'Experimental demonstration of a quantum engine driven by entanglement and local measurements', Physical Review Research 4, L032042 (2022). DOI: 10.1103/PhysRevResearch.4.L032042; https://doi.org/10.1103/PhysRevResearch.4.L032042
  12. B. Bhandari, R. Czupryniak, P. A. Erdman, and A. N. Jordan, 'Measurement-based quantum thermal machines with feedback control', Entropy 25, 204 (2023). DOI: 10.3390/e25020204; https://arxiv.org/abs/2212.01502
  13. A. Panda, F. C. Binder, and S. Vinjanampathy, 'Nonideal measurement heat engines', Physical Review A 108, 062214 (2023). DOI: 10.1103/PhysRevA.108.062214; https://arxiv.org/abs/2308.02381
  14. G. Perna and E. Calzetta, 'Limits on quantum measurement engines', Physical Review E 109, 044102 (2024). DOI: 10.1103/PhysRevE.109.044102; https://arxiv.org/abs/2312.08148
  15. L. M. Cangemi, C. Bhadra, and A. Levy, 'Quantum engines and refrigerators', Physics Reports 1087, 1-71 (2024). DOI: 10.1016/j.physrep.2024.07.001; https://arxiv.org/abs/2302.00726
  16. G. Schaller, 'How small can Maxwell's demon be? - Lessons from autonomous electronic feedback models', in 'Crossroad of Maxwell Demon' (Advances in Atom and Single Molecule Machines, Springer, 2024). DOI: 10.1007/978-3-031-57904-2_3; https://doi.org/10.1007/978-3-031-57904-2_3
  17. A. de Oliveira Junior, J. B. Brask, and R. Chaves, 'A friendly guide to exorcising Maxwell's demon', PRX Quantum 6, 030201 (2025). DOI: 10.1103/phkv-wrsd; https://arxiv.org/abs/2503.07740
  18. R. E. Kastner, 'Maxwell's demon is foiled by the entropy cost of measurement, not erasure', Foundations 5(2), 16 (2025). DOI: 10.3390/foundations5020016; https://arxiv.org/abs/2503.18186
  19. J. Yi and Y. W. Kim, 'Finite-measurement-duration limits on measurement-driven quantum engines', Physical Review A 113, 042220 (2026). DOI: 10.1103/1t8q-rlx5; https://doi.org/10.1103/1t8q-rlx5