Pith. sign in

REVIEW 3 major objections 5 minor 46 references

Charging and Discharging a Hubbard-Holstein Quantum Battery: Specific Mechanisms and General Insights

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A Hubbard–Holstein dimer is a correlation-driven quantum battery whose ergotropy survives dissipation, but discharging efficiently requires a spectrally matched load.

desk verdict Worth a careful referee on the strength of the discharge analysis; the dissipative robustness claim needs an exact benchmark before it can be trusted. read the letter →

arxiv 2607.29339 v1 pith:B7BSGW42 submitted 2026-07-31 quant-ph cond-mat.mes-hall

Emil Östberg , Arvid Steen , Najmeh Abiri , Irene D'Amico , Claudio Verdozzi This is my paper · ORCID
classification quant-phcond-mat.mes-hall
keywords quantumbatteryHubbard–Holsteindimerergotropyelectron–phononcouplingspectralmatchingdissipativedynamicsworkextractionfinite-bandwidthbath
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that a minimal two-site model—two electrons on a dimer coupled to a single phonon mode—can serve as a quantum battery whose stored work (ergotropy) survives interaction with a dissipative environment, provided the bath has finite bandwidth. It shows that the textbook optimal discharge protocol can extract all stored energy only at the price of an unrealistically structured, fine-tuned coupling. Instead, a practical discharge is achieved by coupling the charged battery to a finite chain whose excitation spectrum overlaps the battery's electron-removal spectrum. The general conclusion is that quantum batteries are not plug-and-play: the microscopic storage mechanism constrains which loads can efficiently draw the energy, so battery and load must be designed together.

What carries the argument

The central objects are the Hubbard–Holstein dimer Hamiltonian, a two-site two-electron model with a single local phonon mode coupled to one site, which provides the storage mechanism; the ergotropy W(ρ,H), the maximum work extractable by unitary operations, which quantifies battery performance; the charged-state electron-removal spectral function of the battery and the electron-addition spectral function of the load, whose overlap is the spectral-matching criterion; and the Hermitian generator of the perfect-discharge unitary, whose dense structure is the obstacle to optimal protocols. The phonon mode does double duty: it stores energy during charging and, by generating a broad vibronic spe

What would settle it

Compare the ergotropy at t=100 for the same charging and discharging protocol using an exact quantum simulation of the finite-bandwidth bath (for instance, a numerically converged tensor-network or hierarchy calculation) instead of the mean-field equations; if it decays substantially below the reported value, the robustness claim fails. Alternatively, scan the load's on-site energies through a range where the addition spectrum has zero overlap with the charged battery's removal spectrum while all other parameters stay fixed: if significant work is still extracted, spectral matching is not the

Watch

Extended reading notes

Core claim

Energy storage in this model is carried jointly by electronic correlations and the phonon mode, and it is robust: after the charging field is off, the ergotropy stays nearly flat even with the oscillator bath, because the charged system's spectral weight shifts away from the bath's frequency window. The exact inverse-construction protocol for complete discharge, by contrast, needs a dense generator whose off-diagonal elements correspond to high powers of boson operators—physically implausible. The practical alternative is a small chain load optimised so that its electron-addition spectrum overlaps the charged battery's electron-removal spectrum; with that overlap, substantial work is extract

Load-bearing premise

The load-bearing assumption is that the mixed quantum-classical mean-field treatment of the oscillator bath—which the paper itself notes does not satisfy detailed balance and typically overestimates quantum coherence—captures how the stored ergotropy relaxes in time; if an exact treatment shows significantly faster relaxation, the robustness claim weakens.

Editorial extensions

If this is right

  • If correct, a charged quantum battery cannot be treated as a generic energy reservoir: the same battery may be efficiently drained by one load and nearly inert for another, so battery and load must be co-designed.
  • The phonon mode's broad vibronic spectrum is what makes small finite loads viable; systems without such broadening will need a denser load spectrum or additional engineered channels.
  • Finite-bandwidth environments matter: a spectrally shifted charged state can suppress relaxation, so broadband Markovian dissipators would misestimate the retention time.
  • The matching principle gives machine-learning search a target objective: optimise load topology and parameters against the charged battery's spectral function rather than against an arbitrary discharge curve.
  • Optimal universal discharge, though exact in principle, is a poor design target for systems with bosonic modes, because the inverse-constructed coupling is essentially inaccessible.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same spectral-matching logic suggests a practical characterisation protocol: measure the charged battery's removal spectrum before choosing a load; if the charged spectrum is sparse, larger or engineered loads will be needed.
  • A testable extension is to vary the phonon frequency: lower frequencies should make discharge easier by placing more vibronic sidebands in the matching window, while very high frequencies should suppress extracted work.
  • The robustness claim could be probed directly by exact quantum bath methods for this small model; if the true dynamics relax ergotropy faster than the mean-field result, the finite-bandwidth spectral-shift explanation would still survive, but the quantitative retention would need revision.
  • The principle may carry over to other boson-mediated energy-transfer devices, such as emitters coupled to waveguides, where the same spectral-overlap condition should govern efficiency.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes a Hubbard-Holstein dimer (one phonon mode coupled to one site of a two-site Hubbard model) as a quantum battery. The battery is charged by a time-dependent parity-odd onsite potential, then allowed to interact with a Caldeira-Leggett bath of harmonic oscillators, and finally discharged either by an ideal unitary protocol (Allahverdyan–Balian–Nieuwenhuizen) or by coupling to a small non-interacting chain (the load). The main claims are: (i) the charged state retains ergotropy robustly under dissipation; (ii) ideal full discharge requires an unrealistically structured driving Hamiltonian, as demonstrated by pruning the generator; (iii) an optimized load can extract substantial energy, and this is attributed to spectral matching between the charged battery's electron-removal spectrum and the load's electron-addition density of states. The paper concludes with a general design principle that quantum battery discharge is not universal and requires mutual spectral compatibility, which the phononic vibronic spectrum helps to achieve. The numerics use a Suzuki–Trotter propagator for the electron-phonon system and a mixed quantum-classical Ehrenfest treatment for the bath.

Significance. If the claims are validated, this work would provide a concrete, minimal model of a correlation-driven quantum battery with phononic storage, and would highlight an underappreciated practical limitation of ideal work-extraction protocols. The paper is methodically careful in several respects: the optimal-discharge control test (pruning matrix elements of Λ and observing degraded extraction) is a good diagnostic; the optimization of the load parameters using analytical gradients is clearly presented; and the spectral-function analysis is a standard and informative tool. The paper also explicitly acknowledges the qualitative status of its dissipative spectral argument and defers a rigorous treatment to future work. However, the central dissipation-robustness claim rests entirely on the Ehrenfest approximation, whose limitations are admitted, and the spectral-matching design principle is derived from a single fitted example without a sufficiency test. The work is therefore promising but not yet conclusive; the major claims require additional support before they can be accepted as established.

major comments (3)
  1. [§3, Eq. (3), Fig. 2, App. E] The robustness of ergotropy against dissipation is the headline claim, but it is computed entirely within the Ehrenfest mean-field approximation of Eqs. (3a)–(3b). The paper itself states that this approximation 'does not, in general, satisfy detailed balance' and 'typically overestimates quantum coherence.' Ergotropy (Eq. (4)) is a functional of the full density-matrix spectrum, so artificially preserved coherences can directly inflate W. The near-constant plateau in Fig. 2 for 50≤t≤100 may therefore be an artifact of the mean-field treatment. No benchmark against an independent method—such as hierarchical equations of motion, a non-Markovian master equation, or even a stochastic surface-hopping approach—is provided. Appendix E explicitly concedes that a 'conclusive assessment' and a 'rigorous determination' are deferred to future work. Given that the abstract and conclusions advance di
  2. [§5, Fig. 3(c), Eq. (6), paragraph beginning 'Overall, the results of Fig. 3 suggest...'] The general design principle—that efficient work extraction requires spectral matching between battery and load—is derived from a post-hoc interpretation of a single optimized four-site chain. The load parameters are optimized to maximize ergotropy extraction, and then the spectral overlap is shown to be strong. This is not circular, but it does not establish that spectral overlap is sufficient (or even necessary) for extraction. The paper does not test whether a load engineered to have large spectral overlap but with different parameters (e.g., different matrix elements) extracts work, nor does it quantify the relationship between overlap and extracted W. Moreover, the specific claim that the phonon-induced vibronic spectrum 'relaxes' the matching requirements is not supported by a comparison: no benchmark with g=0 (or a pure Hubbard dimer) is shown to demonstrate that phonons make matc
  3. [§3, paragraph 'This behavior is consistent with the phonon mode acting as an energy reservoir...'] The paper attributes the robustness of W to the phonon mode storing energy and releasing it slowly to the bath. However, the supporting evidence is only that ⟨n_ph⟩ grows during charging and remains nearly constant during dissipation. This is an indirect correlation and does not establish that the phonons are causally responsible for the ergotropy plateau. The paper does not compute a partition of ergotropy or energy between electronic and phononic subsystems, nor does it show that removing or heavily detuning the phonon mode (g→0 or ω large) destroys the plateau. Without such a control, the storage mechanism remains plausible but unproven. The claim in the abstract that ergotropy is 'correlation-driven' and 'robustly stored' would be better supported by such a comparison.
minor comments (5)
  1. [§2, Eq. (1) and App. C] The notation '4N_ph' in the text (dimension of the density matrix) should be typeset as '4 N_ph' or '4N_ph' with proper math mode. Also, in Eq. (7) of App. C, the battery–load coupling appears as V_i(t) but the text describes coupling only after τ; please clarify whether Vi(t) is time-dependent and how the switching is implemented.
  2. [§5, Fig. 3(c)] The spectral functions are broadened with γ=0.05 (App. C) but the energy axis and the scaling of the load spectrum (divided by 8) are not explained in the caption. Please add a short description of how the spectra are normalized and what the factor of 8 represents.
  3. [References] Several references have malformed DOIs or URLs that appear to be placeholders, e.g. Ref. [10] (doi.org/10.1103/ndlt-qszr) and Ref. [13] (doi.org/10.1103/kzvn-dj7v). These will not resolve for the reader. Please correct these to the actual DOIs or remove the broken hyperlinks.
  4. [Throughout] There are several typos and formatting issues: 'AHubbard–Holsteindimerfunctionsas...' in the abstract, 'Hubbard-HolsteinV(t)' missing a space in §4, and the duplicate affiliation line for E. Östberg in the author list. These should be corrected in a final revision.
  5. [App. E] The statement that the data are not publicly available 'due to privacy or ethical restrictions' is unusual for a purely theoretical study. Please clarify what privacy or ethical considerations apply, or consider making the datasets (or the code) available, which would strengthen reproducibility.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central results are direct simulations or follow an external theorem; self-citations are minor and the spectral-matching interpretation is post-hoc, not a fitted prediction.

full rationale

The paper's derivation chain is essentially self-contained. The battery dynamics in Figs. 1-2 come from solving the stated equations of motion (3a)-(3b), with ergotropy evaluated from the definition (4); no parameter in that definition is fitted to the W values that are then reported. The full-discharge protocol in Fig. 2c directly implements the external unitary construction of Ref. [19], U(τ)=Σ_k|ε_k><r_k|, with V(t) generated by the inverse-construction procedure in App. B, so the near-complete extraction is a direct implementation of a known theorem rather than a circular prediction. The non-optimal discharge in Sec. 5 optimizes load parameters by BFGS (App. C) to maximize extraction, and the spectral-function overlap in Fig. 3c is an interpretation of the optimized solution, not an independent predictive constraint; this is post-hoc explanation, not a fitted input renamed as a prediction. The self-citations [29,32] are acknowledged student-paper precursors for preliminary dynamics and for a numerical propagator technique, and they do not carry the abstract's robustness or design-principle claims. The passages noting that Ehrenfest dynamics 'does not, in general, satisfy detailed balance' and 'typically overestimates quantum coherence' (Sec. 2), and App. E's statement that the spectral interpretation 'remains qualitative' with rigorous analysis deferred, are accuracy/scope limitations rather than circularity. No equation in the paper reduces to its own inputs by construction.

Assumptions & free parameters 9 free parameters · 8 assumptions · 0 invented entities

The model parameters are hand-chosen for a single representative regime, not fitted to experimental data; the load-chain parameters are genuinely fitted to maximize discharge. The chief unvalidated assumption is Ehrenfest dynamics for dissipation and the post-hoc spectral-matching interpretation. No new entities are postulated.

free parameters (9)
  • Hubbard interaction U = 2 (in units of J=1)
    Electron-electron repulsion chosen for the representative regime; no scan over U is reported.
  • Phonon frequency ω = 0.1
    Slow phonon regime; central to the battery storage mechanism.
  • Electron-phonon coupling g = 0.2
    Coupling strength; chosen to give manageable dynamics.
  • Inverse temperature β = 2
    Initial thermal state of the dimer.
  • Charging field parameters ξ and Tmax = ξ=10, Tmax=50
    Drive amplitude/frequency and duration; chosen to produce charging.
  • Bath parameters (A, a, Δ, Nν) = A=9e-4, a=0.2, Δ=0.001, Nν=1000
    Sub-Ohmic finite-bandwidth bath; parameters chosen for finite bandwidth and weak dissipation.
  • Phonon Hilbert-space truncation Nph = 50
    Numerical truncation; no convergence test shown.
  • Optimized load parameters (ε_i, V_i, V_ij) = ε=(2.5151,2.4936,2.4840,2.4334), V1=0.6872, V2=-0.5746, V12=-0.8814, V23=-0.8969, V34=-1.1857
    BFGS-fitted to maximize ergotropy extraction at target time; these are genuine fitting parameters central to the discharge demonstration.
  • Discharge optimization times = T=10, td=110, final evolution 150
    Time horizons for optimization and observation; chosen to show extraction before finite-size revivals.
assumptions (8)
  • domain assumption Ehrenfest mean-field decoupling of the Caldeira-Leggett bath (Eq. 3a-3b) correctly captures the dissipative dynamics qualitatively
    The paper states this approximation does not satisfy detailed balance and overestimates coherence, yet uses it for all dissipation-robustness results.
  • domain assumption A harmonic-oscillator bath with finite bandwidth (ω∈[0,1]) and no counterterm is an adequate environment model
    Used in Section 2 and Appendix E; the robustness conclusion depends on the finite bandwidth.
  • domain assumption Initial state is thermal (β=2) and decoupled from the bath at t=0
    Sets the initial condition for all simulations.
  • domain assumption Truncating phonons at Nph=50 is sufficient
    No convergence test against larger Nph is provided.
  • standard math The Allahverdyan-Balian-Nieuwenhuizen construction (Ref. [19]) yields the optimal full-discharge unitary
    External theorem used to derive Λ and V(t); not re-derived here.
  • ad hoc to paper Electron-removal spectral function of battery and electron-addition LDOS of load determine work extraction (Bardeen analogy)
    Proposed in Section 5 as the mechanism; matrix elements are not computed, so this is a heuristic explanatory assumption.
  • standard math Principal branch of matrix logarithm gives a valid Hermitian generator Λ
    Used in Appendix B; branch choice is standard but could affect Λ.
  • standard math Suzuki-Trotter decomposition with O(Δ^3) error is accurate
    Standard numerical integration assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Charging and Discharging a Hubbard-Holstein Quantum Battery: Specific Mechanisms and General Insights." pith.science (2026). https://pith.science/paper/B7BSGW42

@misc{pith2026260729339,
  author       = {Pith},
  title        = {Pith review of: Charging and Discharging a Hubbard-Holstein Quantum Battery: Specific Mechanisms and General Insights},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B7BSGW42}},
  note         = {Machine review of arXiv:2607.29339}
}
read the original abstract

A Hubbard-Holstein dimer functions as a correlation-driven quantum battery, with ergotropy robustly stored, under some conditions, even in the presence of dissipation. We find that, although optimal work extraction can in principle recover all the stored energy, it requires unrealistically fine-tuned couplings. By contrast, a physically realizable protocol based on spectral matching between the battery and the load achieves substantial, albeit suboptimal, energy extraction. Our results identify a mechanism for quantum energy storage, provide a realistic route to work extraction that is amenable to machine-learning-based theoretical exploration, and suggest that quantum batteries may not be universally deployable: the microscopic mechanism responsible for storing energy can constrain the classes of systems able to efficiently extract it.

Figures

Figures reproduced from arXiv: 2607.29339 by the authors.

Figure 1
Figure 1. Battery schematic and time evolution. (a) Isolated battery (Hubbard-Holstein dimer) and external-field parameters. (b) Battery coupled to a dissipative oscillator bath. (c) Battery–load setup. (d) Time evolution from an initial thermal state (β = 2, U = 2, ω = 0.1, g = 0.2). Shown are the phonon displacement ⟨ˆb † + ˆb⟩ and site occupations ⟨nˆL⟩ and ⟨nˆR⟩. Vertical dashed lines separate the charging, dissipation, a… view at source ↗
Figure 2
Figure 2. Battery performance. (a)–(c) Ergotropy W (red curve) and average phonon occupation ⟨ˆb †ˆb⟩/4 (green curve) during charging, dissipation, and discharging. Parameters are as in [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. (a) Ergotropy W during load optimization for four battery–load couplings: optimized load coupled to both battery sites, to the left site, or to the right site, and a non-optimized shifted load coupled to both sites. (b) Optimized case: battery-site densities (thick black curves), load-site densities (thin curves), and phonon displacement/occupation (blue curves). Electronic and phononic quantities use the left and r… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Site labels for the battery+chain system Here εi denotes the on-site energies of the external chain and Vij its nearest-neighbor hopping amplitudes. We consider two possible couplings Vi between the battery sites and the first site of the external chain ( [PITH_FULL_I…
Figure 5
Figure 5. Figure 5: Comparison between the spectral functions for the thermal state considered in the main text and the spectral functions obtained from (8) The initial distributions at t = 0 are quite different ( [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 15 canonical work pages

  1. [1]

    Andolina G M, Keck M, Mari A, Giovannetti V and Polini M 2019Phys. Rev. B99(20) 205437 URLhttps://link.aps.org/doi/10.1103/PhysRevB.99.205437

    work page doi:10.1103/physrevb.99.205437

  2. [2]

    Fox E J, Herrera M, Schmidt-Kaler F and D’Amico I 2024Entropy26ISSN 1099-4300 URL https://www.mdpi.com/1099-4300/26/11/952

  3. [3]

    Shaghaghi V, Singh V, Benenti G and Rosa D 2022Quantum Science and Technology704LT01 URLhttps://doi.org/10.1088/2058-9565/ac8829

    doi:10.1088/2058-9565/ac8829

  4. [4]

    Campbell S, D’Amico I, Ciampini M A, Anders J, Ares N, Artini S, Auffèves A, Bassman Oftelie L, Bettmann L P, Bonança M V S, Busch T, Campisi M, Cavalcante M F, Correa L A, Cuestas E, Dag C B, Dago S, Deffner S, Del Campo A, Deutschmann-Olek A, Donadi S, Doucet E, Elouard C, Ensslin K, Erker P, Fabbri N, Fedele F, Fiusa G, Fogarty T, Folk J, Guarnieri G, ...

  5. [6]

    Herrera M, Zawadzki K and D’Amico I 2018The European Physical Journal B91248 ISSN 1434-6036 URLhttps://doi.org/10.1140/epjb/e2018-90186-5

    work page doi:10.1140/epjb/e2018-90186-5

  6. [7]

    Campaioli F, Gherardini S, Quach J Q, Polini M and Andolina G M 2024Rev. Mod. Phys. 96(3) 031001 URLhttps://link.aps.org/doi/10.1103/RevModPhys.96.031001

    doi:10.1103/revmodphys.96.031001

  7. [8]

    org/10.1007/s44464-026-00012-0 11 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al

    Guha Majumdar M 2026Discover Quantum Science210 ISSN 3059-4529 URLhttps://doi. org/10.1007/s44464-026-00012-0 11 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al

    work page doi:10.1007/s44464-026-00012-0

  8. [9]

    Hovhannisyan K V, Perarnau-Llobet M, Huber M and Acín A 2013Phys. Rev. Lett.111(24) 240401 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.111.240401

    doi:10.1103/physrevlett.111.240401

Show all 46 references
  1. [10]

    1103/ndlt-qszr

    Khan N A and He D 2026PRX Energy5(2) 023003 URLhttps://link.aps.org/doi/10. 1103/ndlt-qszr

  2. [11]

    Quach J Q, Cerullo G and Virgili T 2023Joule72195–2200 ISSN 2542-4785 URLhttps: //doi.org/10.1016/j.joule.2023.09.003

    2023 doi

  3. [12]

    Downing C A and Ukhtary M S 2023Communications Physics6322 ISSN 2399-3650 URL https://doi.org/10.1038/s42005-023-01439-y

    doi

  4. [13]

    Andolina G M, Stanzione V, Giovannetti V and Polini M 2025Phys. Rev. Lett.134(24) 240403 URLhttps://link.aps.org/doi/10.1103/kzvn-dj7v

    doi

  5. [14]

    Campaioli F, Pollock F A and Vinjanampathy S 2018Quantum Batteries(Cham: Springer In- ternational Publishing) pp 207–225 ISBN 978-3-319-99046-0 URLhttps://doi.org/10.1007/ 978-3-319-99046-0_8

  6. [15]

    Bhattacharjee S and Dutta A 2021The European Physical Journal B94239 ISSN 1434-6036 URLhttps://doi.org/10.1140/epjb/s10051-021-00235-3

    doi

  7. [16]

    Ferraro D, Campisi M, Andolina G M, Pellegrini V and Polini M 2018Phys. Rev. Lett.120(11) 117702 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.120.117702

    doi

  8. [17]

    Catalano A, Giampaolo S, Morsch O, Giovannetti V and Franchini F 2024PRX Quantum5(3) 030319 URLhttps://link.aps.org/doi/10.1103/PRXQuantum.5.030319

    doi

  9. [18]

    Ahmadi B, Mazurek P, Horodecki P and Barzanjeh S 2024Phys. Rev. Lett.132(21) 210402 URLhttps://link.aps.org/doi/10.1103/PhysRevLett.132.210402

    doi

  10. [19]

    Allahverdyan A E, Balian R and Nieuwenhuizen T M 2004Europhysics Letters67565 URL https://doi.org/10.1209/epl/i2004-10101-2

    doi

  11. [20]

    Joshi J and Mahesh T S 2026Phys. Rev. A113(3) 032407 URLhttps://link.aps.org/doi/ 10.1103/7pjs-146q

    doi

  12. [21]

    Hubbard J 1963Proceedings of the Royal Society of London. A. Mathematical and Physical Sci- ences276238–257 ISSN 0080-4630 (Preprinthttps://royalsocietypublishing.org/rspa/ article-pdf/276/1365/238/54456/rspa.1963.0204.pdf) URLhttps://doi.org/10.1098/ rspa.1963.0204

    1963

  13. [23]

    Wellein G, Röder H and Fehske H 1996Phys. Rev. B53(15) 9666–9675 URLhttps://link. aps.org/doi/10.1103/PhysRevB.53.9666

    doi

  14. [24]

    Jeon G S, Park T H, Han J H, Lee H C and Choi H Y 2004Phys. Rev. B70(12) 125114 URL https://link.aps.org/doi/10.1103/PhysRevB.70.125114

    doi

  15. [25]

    Macridin A, Sawatzky G A and Jarrell M 2004Phys. Rev. B69(24) 245111 URLhttps: //link.aps.org/doi/10.1103/PhysRevB.69.245111

    doi

  16. [26]

    Berciu M 2007Phys. Rev. B75(8) 081101(R) URLhttps://link.aps.org/doi/10.1103/ PhysRevB.75.081101

  17. [27]

    Zhang Y Y, Liu T, Chen Q H, Wang X and Wang K L 2009Journal of Physics: Condensed Matter21415601 URLhttps://doi.org/10.1088/0953-8984/21/41/415601

    doi

  18. [28]

    Boström E V n, Helmer P, Werner P and Verdozzi C 2019Phys. Rev. Res.1(1) 013017 URL https://link.aps.org/doi/10.1103/PhysRevResearch.1.013017

    doi

  19. [29]

    Östberg E 2021 Quantum thermodynamics explorations with a Hubbard-Holstein dimer Student Paper, Lund University URLhttps://lup.lub.lu.se/student-papers/record/9055348

    2021

  20. [30]

    Bätge J, Levy A, Dou W and Thoss M 2022Phys. Rev. B106(7) 075419 URLhttps://link. aps.org/doi/10.1103/PhysRevB.106.075419 12 IOP PublishingJournalvv(yyyy) aaaaaa Authoret al

    doi

  21. [31]

    Zhou J, Li A and Galperin M 2024Phys. Rev. B109(8) 085408 URLhttps://link.aps.org/ doi/10.1103/PhysRevB.109.085408

    doi

  22. [32]

    Student Paper, Lund University URLhttps://lup.lub

    Steen A 2026 On the ergotropy, practical-viability, and dissipation characterization of an electron-phonon quantum battery. Student Paper, Lund University URLhttps://lup.lub. lu.se/student-papers/record/9220773

    2026

  23. [33]

    Manzano D 2020AIP Advances10025106 ISSN 2158-3226 URLhttps://doi.org/10.1063/ 1.5115323

  24. [34]

    Ahmadi B, Mazurek P, Barzanjeh S and Horodecki P 2025Phys. Rev. Appl.23(2) 024010 URL https://link.aps.org/doi/10.1103/PhysRevApplied.23.024010

    doi

  25. [35]

    Gopalakrishna M, Viñas Boström E and Verdozzi C 2023SciPost Phys.15138 URLhttps: //scipost.org/10.21468/SciPostPhys.15.4.138

    doi

  26. [36]

    Grabert H and Thorwart M 2018Phys. Rev. E98(1) 012122 URLhttps://link.aps.org/ doi/10.1103/PhysRevE.98.012122

    doi

  27. [37]

    Caldeira A and Leggett A 1983Annals of Physics149374–456 ISSN 0003-4916 URLhttps: //www.sciencedirect.com/science/article/pii/0003491683902026

  28. [38]

    Horsfield A P, Bowler D R, Fisher A J, Todorov T N and Montgomery M J 2004Journal of Physics: Condensed Matter163609 URLhttps://dx.doi.org/10.1088/0953-8984/16/21/ 010

    doi

  29. [39]

    aps.org/doi/10.1103/PhysRevLett.119.227203

    Stahl C and Potthoff M 2017Physical Review Letters119(22) 227203 URLhttps://link. aps.org/doi/10.1103/PhysRevLett.119.227203

    doi

  30. [40]

    org/doi/10.1103/PhysRevLett.128.197202

    Bai H, Han L, Feng X Y, Zhou Y J, Su R X, Wang Q, Liao L Y, Zhu W X, Chen X Z, Pan F, Fan X L and Song C 2022Physical Review Letters128(19) 197202 URLhttps://link.aps. org/doi/10.1103/PhysRevLett.128.197202

    doi

  31. [41]

    Gay-Balmaz F and Tronci C 2023Journal of Physics A: Mathematical and Theoretical56 144002 URLhttps://dx.doi.org/10.1088/1751-8121/acc145

    doi

  32. [42]

    1021/ct800518j

    Andrade X, Castro A, Zueco D, Alonso J L, Echenique P, Falceto F and Rubio A 2009Journal of Chemical Theory and Computation5728–742 ISSN 1549-9618 URLhttps://doi.org/10. 1021/ct800518j

  33. [43]

    Pusz W and Woronowicz S L 1978Communications in Mathematical Physics58273–290 URL https://doi.org/10.1007/BF01614224

    doi

  34. [44]

    Lenard A 1978Journal of Statistical Physics19575–586 URLhttps://doi.org/10.1007/ BF01011769

  35. [45]

    Nocedal J and Wright S J 2006Numerical Optimization2nd ed (New York: Springer Series in Operations Research and Financial Engineering)

  36. [46]

    Bardeen J 1961Phys. Rev. Lett.6(2) 57–59 URLhttps://link.aps.org/doi/10.1103/ PhysRevLett.6.57

  37. [47]

    org/10.1007/11526216_2

    Hatano N and Suzuki M 2005Finding Exponential Product Formulas of Higher Orders(Berlin, Heidelberg: Springer Berlin Heidelberg) pp 37–68 ISBN 978-3-540-31515-5 URLhttps://doi. org/10.1007/11526216_2

    doi

  38. [48]

    Wilcox R M 1967Journal of Mathematical Physics8962–982 ISSN 0022-2488 13

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.