Complete positivity versus Redfield accuracy: the existence of thermodynamically consistent Lindblad master equations for systems weakly coupled to multiple baths

For a finite-dimensional quantum system weakly coupled (coupling $\epsilon$) to several macroscopic thermal baths, the microscopically derived Born–Markov or Redfield master equation reproduces the correct leading-order ($O(\epsilon^2)$) non-equilibrium steady-state (NESS) populations and coherences, thermalizes when all baths coincide, and preserves local conservation laws, but generically violates complete positivity; Tupkary, Dhar, Kulkarni and Purkayastha (Phys. Rev. A 105, 032208 (2022)) showed that every known approximation of it to Lindblad (completely positive) form restores complete positivity at the price of violating one or more of these requirements. The open problem is whether any completely positive Markovian description exists that simultaneously satisfies all the requirements for arbitrary time-independent system Hamiltonians, or whether the trade-off can be proven fundamental — including whether the steady-state positivity violations of the Redfield equation can be cured by next-order ($O(\epsilon^4)$) corrections inferred via perturbative steady-state reconstruction without deriving the full fourth-order time-convolutionless generator. Later work turned the trade-off into rigorous instance-level semidefinite-programming no-go results and produced partial escapes (memory-undressing within the Universal Lindblad Equation, exact locality–thermodynamic-consistency coexistence for quadratic systems, non-secular generators satisfying KMS (Kubo–Martin–Schwinger) detailed balance for equilibrium thermalization), but the general question at non-equilibrium steady states remains open, as explicitly stated by the source authors in 2026.

Background

A Markovian quantum master equation (QME) describes the density matrix of an open quantum system using only system-level variables. For a system coupled weakly (coupling strength $\epsilon$) to thermal baths, the standard microscopic derivation combines the Born approximation (bath stays equilibrium and uncorrelated) with a Markov approximation, giving the Redfield equation (RE), which is accurate to order $\epsilon^2$ but is not of Lindblad form. Complete positivity (CP) means the evolution maps every density matrix, even of an extended system, to a positive semidefinite operator; Gorini–Kossakowski–Sudarshan and Lindblad (GKLS) proved that CP plus trace preservation forces a specific Lindblad form with non-negative rates. The RE generically violates CP: for some initial states and times it produces unphysical density matrices with negative eigenvalues. The usual cure is the secular (rotating-wave) approximation, which discards rapidly oscillating terms in the system energy eigenbasis and yields the eigenbasis (global) Lindblad equation (ELE); alternatively one derives dissipators acting locally at the bath-attached sites (local Lindblad equation, LLE). Both cures have long been known to distort the steady state in multi-bath, interacting systems. The Universal Lindblad Equation (ULE) of Nathan and Rudner (Phys. Rev. B 102, 115109 (2020)) is a more refined construction whose accuracy was argued to match the RE while remaining of Lindblad form. A non-equilibrium steady state (NESS) is the long-time state reached when baths are held at different temperatures or chemical potentials; the Lamb shift is the bath-induced Hamiltonian correction of order $\epsilon^2$. Local conservation laws require continuity equations: the current entering the system through a bath-contact site must equal the current leaving the next site, so that energy or particle flows through the bulk without spurious sources or sinks — a condition on coherences in the energy eigenbasis, not only on populations. Tupkary, Dhar, Kulkarni and Purkayastha (Phys. Rev. A 105, 032208 (2022)) laid down the fundamental requirements any further approximation of the RE should satisfy at weak coupling for time-independent Hamiltonians: (C1) correct $O(\epsilon^2)$ populations and coherences in the system energy eigenbasis simultaneously, (C2) thermalization to the Gibbs state to the appropriate order when all baths have the same temperature and chemical potential, (C3) local conservation laws at NESS. They showed the RE violates only CP, while, to their knowledge, every existing weak-coupling Lindblad description violates at least one of C1–C3; for the ULE specifically, populations are accurate but coherences are not, so local conservation laws fail at NESS. They concluded that there seems to be no consistent way to cure the RE's positivity violation without going to higher order in $\epsilon$, and pointed to two escape routes outside their assumptions: steady-state reconstruction techniques that infer next-order ($O(\epsilon^4)$) corrections from the RE without deriving the full fourth-order time-convolutionless (TCL) generator (Thingna–Wang–Hänggi and Xu–Thingna–Wang), and descriptions with dissipators acting on bath degrees of freedom or derived under particular driving (collision models), which can be valid even beyond weak coupling.

Problem Statement

For a finite-dimensional quantum system with an arbitrary time-independent Hamiltonian, weakly coupled (coupling $\epsilon$) to multiple macroscopic thermal baths that may be at different temperatures and chemical potentials, does there exist a completely positive, trace-preserving Markovian master equation in Lindblad form whose non-equilibrium steady state simultaneously (i) reproduces the exact Redfield-order $O(\epsilon^2)$ populations and coherences in the system energy eigenbasis, (ii) reduces to the thermal (Gibbs) state to the appropriate order in $\epsilon$ when all baths are at equal temperature and chemical potential, and (iii) satisfies the local continuity equations (local conservation laws) for the conserved quantities exchanged with the baths? Equivalently, if no such description exists, can a general no-go theorem be proven for the class of weak-coupling Markovian Lindblad descriptions? The problem includes as a specific sub-question whether the positivity violations of the Redfield steady state can be cured by incorporating or inferring next-order $O(\epsilon^4)$ corrections to the steady-state populations and coherences via the perturbative reconstruction techniques of Thingna–Wang–Hänggi and Xu–Thingna–Wang, without a full derivation of the fourth-order time-convolutionless generator. This sub-question is a required part of the problem that is independent of the general existence-or-no-go question, so that resolving either branch does not by itself settle the other. Escape routes outside the weak-coupling system-side Lindblad class — dissipators acting on bath degrees of freedom or Lindblad equations derived under particular external driving such as collision models — are excluded from the core existence-or-no-go question, which is the primary required deliverable and whose general answer need not invoke them. As part of the sought answer rather than a restriction of the question, a complete answer also states the general extent to which special system classes evade the trade-off, with quadratic Hamiltonians and equilibrium all-sites-attached geometries as illustrations rather than an exhaustive list of instances to cover.

References

1. Devashish Tupkary, Abhishek Dhar, Manas Kulkarni, Archak Purkayastha, "Fundamental limitations in Lindblad descriptions of systems weakly coupled to baths", Phys. Rev. A 105, 032208 (2022). DOI: 10.1103/PhysRevA.105.032208; arXiv:2105.12091, https://arxiv.org/abs/2105.12091
2. Devashish Tupkary, Abhishek Dhar, Manas Kulkarni, Archak Purkayastha, "Searching for Lindbladians obeying local conservation laws and showing thermalization", Phys. Rev. A 107, 062216 (2023). DOI: 10.1103/PhysRevA.107.062216; arXiv:2301.02146, https://arxiv.org/abs/2301.02146
3. Frederik Nathan, Mark S. Rudner, "Universal Lindblad equation for open quantum systems", Phys. Rev. B 102, 115109 (2020). DOI: 10.1103/PhysRevB.102.115109; arXiv:2004.01469, https://arxiv.org/abs/2004.01469
4. Frederik Nathan, Mark S. Rudner, "Quantifying the accuracy of steady states obtained from the Universal Lindblad Equation", Phys. Rev. B 109, 205140 (2024). DOI: 10.1103/PhysRevB.109.205140; arXiv:2206.02917, https://arxiv.org/abs/2206.02917
5. Soumyadeep Sarma, Manas Kulkarni, Archak Purkayastha, Devashish Tupkary, "Semidefinite programming for understanding the limitations of Lindblad equations", Phys. Rev. A 114, 012423 (2026). DOI: 10.1103/jyck-mvlg; arXiv:2602.01794, https://arxiv.org/abs/2602.01794
6. Juzar Thingna, Jian-Sheng Wang, Peter Hänggi, "Generalized Gibbs state with modified Redfield solution: Exact agreement up to second order", J. Chem. Phys. 136, 194110 (2012). DOI: 10.1063/1.4718706, https://doi.org/10.1063/1.4718706
7. Juzar Thingna, Jian-Sheng Wang, Peter Hänggi, "Reduced density matrix for nonequilibrium steady states: A modified Redfield solution approach", Phys. Rev. E 88, 052127 (2013). DOI: 10.1103/PhysRevE.88.052127; arXiv:1307.2512, https://arxiv.org/abs/1307.2512
8. Xiansong Xu, Juzar Thingna, Jian-Sheng Wang, "Finite coupling effects in double quantum dots near equilibrium", Phys. Rev. B 95, 035428 (2017). DOI: 10.1103/PhysRevB.95.035428, https://doi.org/10.1103/PhysRevB.95.035428
9. Tobias Becker, Alexander Schnell, Juzar Thingna, "Canonically consistent quantum master equation", Phys. Rev. Lett. 129, 200403 (2022). DOI: 10.1103/PhysRevLett.129.200403; arXiv:2205.12848, https://arxiv.org/abs/2205.12848
10. Juzar Thingna, Xiansong Xu, Daniel Manzano, "Polaron transformed canonically consistent quantum master equation", arXiv:2604.02731 (2026), https://arxiv.org/abs/2604.02731
11. Diego Fernández de la Pradilla, Esteban Moreno, Johannes Feist, "Recovering an accurate Lindblad equation from the Bloch-Redfield equation for general open quantum systems", Phys. Rev. A 109, 062225 (2024). DOI: 10.1103/PhysRevA.109.062225; arXiv:2402.06354, https://arxiv.org/abs/2402.06354
12. Antonio D'Abbruzzo, Donato Farina, Vittorio Giovannetti, "Recovering complete positivity of non-Markovian quantum dynamics with Choi-proximity regularization", Phys. Rev. X 14, 031010 (2024). DOI: 10.1103/PhysRevX.14.031010; arXiv:2309.16320, https://arxiv.org/abs/2309.16320
13. Koki Shiraishi, Masaya Nakagawa, Takashi Mori, "Davies equation without the secular approximation: Reconciling locality with quantum thermodynamics for open quadratic systems", Phys. Rev. A 113, 052212 (2026). DOI: 10.1103/m1zr-7rmv; arXiv:2507.10080, https://arxiv.org/abs/2507.10080
14. Matteo Scandi, Álvaro M. Alhambra, "Thermalization in open many-body systems and KMS detailed balance", Phys. Rev. X 16, 011040 (2026). DOI: 10.1103/sfp3-3sqf; arXiv:2505.20064, https://arxiv.org/abs/2505.20064
15. Camilo Santiago Tello Breuer, Tobias Becker, André Eckardt, "Benchmarking quantum master equations beyond ultraweak coupling", Phys. Rev. B 110, 064319 (2024). DOI: 10.1103/PhysRevB.110.064319; arXiv:2403.08320, https://arxiv.org/abs/2403.08320
16. Elyana Crowder, Lance Lampert, Grihith Manchanda, Brian Shoffeitt, Srikar Gadamsetty, Yiting Pei, Shantanu Chaudhary, Dragomir Davidović, "Invalidation of the Bloch-Redfield equation in the sub-Ohmic regime via a practical time-convolutionless fourth-order master equation", Phys. Rev. A 109, 052205 (2024). DOI: 10.1103/PhysRevA.109.052205; arXiv:2310.15089, https://arxiv.org/abs/2310.15089
17. Teruhiro Ikeuchi, Takashi Mori, "Error bounds on the universal Lindblad equation in the thermodynamic limit", Phys. Rev. B 112, 094309 (2025). DOI: 10.1103/bn2y-tw49; arXiv:2503.14916, https://arxiv.org/abs/2503.14916
18. Patrick P. Potts, Alex Arash Sand Kalaee, Andreas Wacker, "A thermodynamically consistent Markovian master equation beyond the secular approximation", New J. Phys. 23, 123013 (2021). DOI: 10.1088/1367-2630/ac3b2f, https://doi.org/10.1088/1367-2630/ac3b2f
19. Marco Cattaneo, Gabriele De Chiara, Sabrina Maniscalco, Roberta Zambrini, Gian Luca Giorgi, "Collision Models Can Efficiently Simulate Any Multipartite Markovian Quantum Dynamics", Phys. Rev. Lett. 126, 130403 (2021). DOI: 10.1103/PhysRevLett.126.130403, https://doi.org/10.1103/PhysRevLett.126.130403
20. Marlon Brenes, Juan José Mendoza-Arenas, Archak Purkayastha, Mark T. Mitchison, Stephen R. Clark, John Goold, "Tensor-Network Method to Simulate Strongly Interacting Quantum Thermal Machines", Phys. Rev. X 10, 031040 (2020). DOI: 10.1103/PhysRevX.10.031040, https://doi.org/10.1103/PhysRevX.10.031040