Pith. sign in

REVIEW 3 major objections 4 minor 106 references

Information thermodynamics for Markov jump processes coupled to underdamped diffusion: Application to nanoelectromechanics

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Mixed Markov-jump and underdamped-diffusion systems satisfy integral fluctuation theorems for each subsystem's partial entropy production.

desk verdict Solid extension of information thermodynamics to mixed jump–underdamped systems, with honest limitations; worth refereeing despite a narrower scope than the abstract suggests. read the letter →

arxiv 2412.03226 v1 pith:QBJUCZ3Y submitted 2024-12-04 cond-mat.stat-mech

Ashwin Gopal , Nahuel Freitas , Massimiliano Esposito This is my paper · ORCID
classification cond-mat.stat-mech MSC 82C3182C7060J75 PACS 05.40.-a05.70.Ln85.85.+j
keywords informationthermodynamicsMarkovjumpprocessunderdampeddiffusionpartialentropyproductionintegratedfluctuationtheoremlocalsecondlawelectronshuttlenanoelectromechanicalsystems
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 extends information thermodynamics to autonomous systems in which one part jumps between discrete states while the other moves through an underdamped, inertial continuum. The authors establish local second laws for each part, with the missing entropy accounted for by information flows between the parts, and prove integrated fluctuation theorems for the partial entropy production of each subsystem. They then analyze two limit regimes—fast jumps with slow inertial motion, and large oscillator mass—and apply the framework to the single-electron shuttle. The shuttle analysis concludes that self-oscillations are sustained mainly by energy transfer from the dot to the oscillator, with information flows saturating and becoming relatively negligible at large voltage.

What carries the argument

The load-bearing object is the decomposition of the total entropy production into partial contributions $\delta\sigma_X$ and $\delta\sigma_Y$, assigned to the two subsystems' own dynamics, together with the information flows $\dot I_X$ and $\dot I_Y$ that measure each subsystem's rate of change of mutual information. The proof of the integrated fluctuation theorems constructs modified time-reversed dynamics: for the jump part, rates are reweighted by the instantaneous conditional distribution, and for the underdamped part, the drift is reversed and reweighted, so that the log-ratio of forward to modified-reverse path probabilities equals the partial entropy production. A second mechanism is the systematic $\epsilon$-expansion of the joint master equation around the adiabatic limit, which yields a closed Fokker-Planck equation for the slow underdamped marginal with effective position-dependent damping $\gamma_{\rm eff}(x)$ and diffusion $D_{\rm eff}(x)$, capturing the first nonzero energy and information flows. A third is the large-mass expansion, where the conditional distribution collapses to the dot marginal and the closed dynamics reduces to the mean-field shuttle equations.

What would settle it

Simulate a two-state dot coupled to an underdamped oscillator with jump rates that include an explicit velocity dependence, for example $\lambda(x,v)=\lambda_0\exp(\pm x/\lambda + a v)$, and check two predictions: the integrated fluctuation theorems either fail or change their information-flow term, and the adiabatic energy and information flows no longer vanish at leading order. Alternatively, for the shuttle at fixed coupling, measure the steady-state ratio $T\dot I/\dot E$ as a function of voltage; the paper predicts it vanishes in the self-oscillating regime, so a controlled experiment or numerical simulation showing that it remains order-one would falsify the energy-dominance claim.

Watch

Extended reading notes

Core claim

The central claim is that the bipartite information-thermodynamics formalism extends to a subsystem described by Markov jump dynamics coupled to a subsystem described by underdamped diffusion, in which position-dependent jump rates satisfy local detailed balance. Concretely, the paper derives the trajectory-level partial entropy productions $\sigma_X$ and $\sigma_Y$ and shows they obey the integrated fluctuation theorems $\langle e^{-\sigma_Y/k_B}\rangle=\langle e^{-[\Sigma_Y-\Delta I_Y]/k_B}\rangle=1$ and $\langle e^{-\sigma_X/k_B}\rangle=1$, so each subsystem's average partial entropy production is nonnegative once the information flow into or out of it is included. The paper further shows that in the adiabatic limit, where the jump dynamics is infinitely fast, a purely conservative underdamped subsystem remains at equilibrium and all steady-state energy and information flows vanish; nonvanishing flows first appear at next order in the timescale separation, through effective damping and diffusion terms that break detailed balance. In the large-mass limit, the underdamped subsystem becomes deterministic and the coupled dynamics reduces to a mean-field approximation, where energy flows survive but information flows vanish. Applied to the single-electron shuttle, these results yield a measurement-feedback picture in which the oscillator acts as a sensor of the dot state, and a thermodynamic efficiency $\eta=\dot E/T\dot\sigma_{\rm tot}$ for converting electrical work into mechanical oscillations.

Load-bearing premise

The argument assumes the jump rates depend only on the underdamped particle's position, never on its velocity; if real rates carry velocity dependence, the adiabatic limit can drive the slow subsystem out of equilibrium and the derived flow hierarchy stops holding.

Editorial extensions

If this is right

  • The local second laws become exact once information flow is included, so apparent violations of the second law in the marginalized dynamics are resolved and each subsystem's entropy production is nonnegative.
  • In the fast-jump, slow-oscillator limit, any steady-state energy and information flow is at least of order $\epsilon^2$ for conservative interactions, so the slow inertial subsystem can be driven out of equilibrium only by going beyond the adiabatic approximation.
  • In the large-mass limit, the mean-field dynamics captures the limit-cycle attractor and the leading energy flow, but predicts zero information flow, which only appears through the higher-order diffusion term.
  • For the single-electron shuttle, the oscillator acts as a sensor that acquires mutual information from the dot, while the self-oscillating regime is maintained predominantly by energy flow, with the ratio $T\dot I/\dot E$ tending to zero at large voltage.
  • The framework provides a path to information-thermodynamic analyses of hybrid CMOS and mechanical clock circuits, where the derived fluctuation theorems and local second laws can be used to bound performance.

Reading between the lines

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

  • A velocity-dependent jump rate would likely break the adiabatic flow-hierarchy result; testing this extension directly would show whether the position-only assumption is essential or merely convenient.
  • The integrated fluctuation theorems for partial entropy production could be combined with standard uncertainty-relation arguments to derive thermodynamic uncertainty relations for the shuttle's current or oscillation amplitude, a step the paper does not take.
  • In non-isothermal conditions, the information-to-energy ratio should grow because the cost and gain scale with different temperatures, so a temperature-biased shuttle may act as a genuine information engine.
  • The saturation of the information flow at large voltage suggests a calibrated nanomechanical shuttle could serve as a direct probe of information-thermodynamic bounds, comparing the measured $T\dot I$ against the local second-law inequality.
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 / 4 minor

Summary. This paper develops a stochastic-thermodynamics framework for autonomous systems in which an underdamped diffusive subsystem X is coupled to a Markov jump subsystem Y. The authors derive trajectory-level energy and entropy balances, identify internal energy and information flows, establish local second laws for each subsystem, and prove integrated fluctuation theorems for the partial entropy productions (Eq. (39)). They then analyze two asymptotic regimes: a timescale-separated (adiabatic) limit in which the leading-order energy and information flows vanish and the first non-zero contributions are computed at O(epsilon^2), and a large-mass limit in which the dynamics reduces to a deterministic mean-field description. The framework is applied to a single-electron shuttle, comparing stochastic simulations with the timescale-separated and mean-field predictions, and concluding that in the self-oscillating regime energy flows dominate information flows and that the shuttle acts as a sensor rather than an information engine.

Significance. If the results hold, the paper substantially extends bipartite information thermodynamics to a previously untreated class of mixed Markov-jump/underdamped-diffusion systems. The IFT proof in Appendix C constructs modified forward/backward dynamics for each subsystem and is a nontrivial contribution. The adiabatic and large-mass perturbation expansions are derived systematically from the master equation rather than assumed ad hoc, and the electron-shuttle application yields concrete, falsifiable predictions (Eqs. (101)-(102), Fig. 5) that are compared with direct stochastic simulations of the same model. The authors are also explicit about the main modeling restriction (position-dependent jump rates) in Sec. VII, which is a strength of the presentation. However, the advertised generality of the framework is wider than what is actually proven, and the quantitative shuttle analysis relies in part on uncontrolled higher-order terms, so the central claims need to be carefully delimited.

major comments (3)
  1. [Sec. II A, Eq. (5); Sec. VII] The load-bearing assumption that the jump rates lambda_{y',y}^rho(x) depend only on the position of the underdamped particle is not generic for MJ-UD systems, and the paper's central theoretical claims are proven only under this restriction. For velocity-dependent rates, the parity-consistent local detailed balance condition would acquire odd-parity corrections, and the modified-process constructions used in Appendix C (Eqs. (C11) and (C19)) would need to be re-derived. The authors themselves concede in Sec. VII that 'further analysis is required to verify the validity of the generalized integrated fluctuation theorem (IFT) under these conditions.' Since the abstract and introduction present the result as a general extension of information thermodynamics to mixed MJ-UD dynamics, the claims should either be extended to velocity-dependent rates or the abstract should explicitly state the position-only restriction. This is a genuine limitation, not merely a presentational one, because the adiabatic vanishing-flow results of Eqs. (55)-(56) and the subsequent O(epsilon^2) formulas depend directly on the velocity-independence of the fast conditional distribution.
  2. [Sec. VI B 1 and Appendix D] The quantitative shuttle results, including the advertised conclusion that energy flows dominate information flows in the self-oscillating regime, are obtained by solving the full timescale-separated Fokker-Planck equation (Eq. (82)) and using the full steady-state solution P_TS(E) from Eq. (D8), which contains uncontrolled higher-order terms in epsilon. Appendix D (Fig. D.2) shows that the exact O(epsilon^2) expressions for the flows, Eqs. (D13), agree with stochastic simulations only for small voltages, and the range of validity shrinks with increasing electromechanical coupling. The main-text comparisons in Figs. 5 and 8 therefore rely on a resummation that is not controlled, even though it is described as 'retaining these higher order terms.' The paper should either justify this resummation or present the controlled O(epsilon^2) results as the primary quantitative evidence, with the full-solution results clearly labeled as an uncontrolled but empirically better approximation.
  3. [Sec. V and Sec. VI A 1, Eq. (76), Eq. (90)] The large-mass mean-field dynamics is derived to O(m^{-1}) using a delta-function ansatz for the marginal distribution, but the exact heat flow Q_X in Eq. (90) contains a diffusive contribution -gamma/(m beta) that is of order m^{-1} and is omitted in the mean-field thermodynamic expressions (e.g., Eq. (74) and Eq. (76)). The paper acknowledges this in Sec. V, yet Fig. 6(a) compares the mean-field energy flow with stochastic simulations that include the full 1/m correction. The order of accuracy of each thermodynamic quantity computed from the mean-field dynamics should be stated explicitly, and the comparison should be adjusted or the missing 1/m term included, otherwise the apparent quantitative agreement in Fig. 6(a) mixes different orders in the expansion.
minor comments (4)
  1. [Sec. III A, Eq. (12)] The notation d_tE_X and d_tE_Y in Eq. (12) is not defined before first use; the reader must infer that these are the rates of change of the bare energies E_X and E_Y. Please define these quantities explicitly.
  2. [Sec. IV B, Eq. (62)] Equation (62) contains a misplaced parenthesis: the term D_eff(x) partial_v^2 is written inside the drift bracket with an extra closing parenthesis. This makes the Fokker-Planck structure harder to read and should be corrected.
  3. [Sec. III D, Eq. (39)] The notation Delta-hat-I_X(Y)(t) in Eq. (39) is ambiguous because the subscript X(Y) appears both as a label and as a placeholder. Writing the two identities with explicit subscripts (e.g., for σ_X and σ_Y separately) would improve clarity.
  4. [Sec. VI A 1, Eq. (81)] The mean-field equations (Eq. (81)) are written as closed equations for (x,v,p_1), but the derivation in Sec. V shows that they hold only up to O(m^{-1}) and that the diffusive term of order m^{-2} is neglected. The text around Eq. (81) should remind the reader of this limitation, since the subsequent comparison in Fig. 6 uses finite masses.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the IFTs are proved by explicit modified-dynamics constructions and the shuttle analysis is benchmarked against stochastic simulations, not fitted.

full rationale

The central claims are derived rather than assumed. Equation (39) is an integral fluctuation theorem whose nontrivial content (the =1 equalities) is proved in Appendix C by constructing modified underdamped and jump dynamics (Eqs. (C11) and (C19)) and verifying that the log-ratio of forward to modified-reverse path probabilities equals the corresponding partial entropy production; the first equality in Eq. (39) is a definitional identity following from s = s_marg_X + s_marg_Y - i and the definitions of Sigma and I, but the fluctuation theorem itself is independent. The adiabatic vanishing of energy and information flows, Eqs. (55)-(56), follows from Eq. (49), where the zeroth-order conditional distribution is velocity-independent because the jump rates depend only on x; this is a clearly stated modeling assumption in Sec. II A, not a fitted result renamed as a prediction. The beyond-adiabatic Fokker-Planck equation and the O(epsilon^2) flow formulas are obtained by systematic perturbation theory, and the electron-shuttle predictions are compared with stochastic simulations of the full master equation (Eq. (79)) using the same parameters, so the comparisons are same-model benchmarks rather than fits. Self-citations such as [12], [17], and [60] are contextual, motivational, or references to prior examples and do not carry the load-bearing derivations, which are self-contained in this paper. The Sec. VII admission that velocity-dependent jump rates would require further analysis of the local detailed balance condition and of the generalized IFT is an honest limitation on scope, not evidence that a target result was assumed into the derivation.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The central results assume a specific stochastic model: position-dependent jump rates, local detailed balance, white-noise friction, timescale separation or large-mass scaling, and a single-electron shuttle idealization. No fitted constants are introduced; the parameters are physical model inputs. The most fragile assumption is the position-only dependence of the jump rates, which the paper itself flags in the conclusion.

assumptions (6)
  • domain assumption Local detailed balance for the Markov jump rates, Eq. (5), relating rate ratios to free energy differences and reservoir work.
    This is the standard thermodynamic consistency condition for jump processes and is stated as a requirement in Sec. II A.
  • domain assumption Jump rates depend only on position x, not on velocity v.
    Stated explicitly in Sec. II A and used throughout Sec. IV to obtain the adiabatic vanishing of flows and the higher-order corrections.
  • standard math Underdamped Langevin dynamics with linear friction, Gaussian white noise, and Stratonovich convention for heat.
    This is the standard model for the continuous subsystem, introduced in Eq. (3) and used in the first-law and entropy-production derivations.
  • domain assumption Timescale separation epsilon << 1 and the assumption that the conditional distribution has relaxed to steady state for the instantaneous marginal density.
    Used in Sec. IV B to derive the closed Fokker-Planck equation, Eq. (62), and the flow formulas, Eqs. (65) and (66).
  • domain assumption Large-mass scaling: the conservative potential scales with mass and the interaction potential contribution vanishes in the m to infinity limit.
    Stated in Sec. V and used to derive the mean-field dynamics, Eq. (72), and its thermodynamic estimates.
  • domain assumption For the electron shuttle, ultra-strong Coulomb blockade so the dot hosts at most one electron, with Fermi-function tunneling rates.
    Stated in Sec. VI A as the model of the NEMS example; this constrains the numerical application.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Information thermodynamics for Markov jump processes coupled to underdamped diffusion: Application to nanoelectromechanics." pith.science (2026). https://pith.science/paper/QBJUCZ3Y

@misc{pith2026241203226,
  author       = {Pith},
  title        = {Pith review of: Information thermodynamics for Markov jump processes coupled to underdamped diffusion: Application to nanoelectromechanics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QBJUCZ3Y}},
  note         = {Machine review of arXiv:2412.03226}
}
read the original abstract

We extend the principles of information thermodynamics to study energy and information exchanges between coupled systems composed of one part undergoing a Markov jump process and another underdamped diffusion. We derive integral fluctuation theorems for the partial entropy production of each subsystem and analyze two distinct regimes. First, when the inertial dynamics is slow compared to the discrete-state transitions, we show that the steady-state energy and information flows vanish at the leading order in an adiabatic approximation, if the underdamped subsystem is governed purely by conservative forces. To capture the non-zero contributions, we consistently derive dynamical equations valid to higher order. Second, in the limit of infinite mass, the underdamped dynamics becomes a deterministic Hamiltonian dynamics driving the jump processes, we capture the next-order correction beyond this limit. We apply our framework to study self-oscillations in the single-electron shuttle - a nanoelectromechanical system (NEMS) - from a measurement-feedback perspective. We find that energy flows dominate over information flows in the self-oscillating regime, and study the efficiency with which this NEMS converts electrical work into mechanical oscillations.

Figures

Figures reproduced from arXiv: 2412.03226 by the authors.

Figure 1
Figure 1. FIG. 1. (a): Schematic diagram of the composite system and its associated reservoirs. The system is composed of subsystem [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. (a): Schematic diagram of the electron shuttle, with an oscillating dot, which experiences a electrostatic force [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. (a): On the left panel, we plot the ensemble averaged trajectory, calculated over 10 [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: FIG. 4. (a): steady-state distribution from the stochastic simulations (left) and the timescale separated dynamics using Eq. (82) (right) (b): [PITH_FULL_IMAGE:figures/full_fig_p017_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. (a): Energetic flow due to the dot dynamics [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. We plot the internal flows as a function of the oscillator mass [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Steady-state thermodynamic flows in the electron shuttle. Arrows indicate the positive flows. [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. (a): The total entropy production [PITH_FULL_IMAGE:figures/full_fig_p023_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

106 extracted references · 74 canonical work pages

  1. [1]

    (79) is equivalent to a meanfield (MF) approximation in the large mass limit, by keeping terms up toO(𝑚−1) in the full master equation

    Large mass limit : Meanfield dynamics In Sec.V, we established that the stochastic dynamics described by Eq. (79) is equivalent to a meanfield (MF) approximation in the large mass limit, by keeping terms up toO(𝑚−1) in the full master equation. The MF approximation effectively neglects fluctuations of the oscillator in the jump dynamics, simplifying the t...

  2. [2]

    when 𝜏𝕐 = 𝛤−1 𝑏 ≪ 𝜏𝕏 =(2𝜋)/𝜔, the timescale separated (TS) Fokker-Planck equation (Eq

    Fast dot dynamics limit: Timescale separated dynamics In the limit of fast tunneling compared to the oscillation time period, i.e. when 𝜏𝕐 = 𝛤−1 𝑏 ≪ 𝜏𝕏 =(2𝜋)/𝜔, the timescale separated (TS) Fokker-Planck equation (Eq. (62)) for the marginal density of the oscillator in the dimensional form is given as: 𝑑𝑡𝑃 𝕏 𝑡 = h −𝑣𝜕𝑥− 𝜕𝑣[𝑓 eff 𝕏 (𝑥)+ 𝑔eff(𝑥)− 𝛾 eff(𝑥)𝑣]...

  3. [3]

    IV B, the dynamics and thermodynamics can be computed using the TS approximation

    Energy and information flows : Comparison with TS and MF dynamics As shown in Sec. IV B, the dynamics and thermodynamics can be computed using the TS approximation. The steady-state distribution 𝑃𝑠𝑠(Γ,𝑦) is accurate up to O(𝜖) and the thermodynamic quantities due to the slow oscillator 𝕏 dynamics are valid up to the orderO(𝜖2). The energy flows ¤E (Eq. (9...

  4. [4]

    (94), the inputted power at the dot is equal to the entropy production rate multiplied by the temperature 𝑇¤𝜎tot, i.e

    Thermal Engine: From electric to mechanical As shown in Eq. (94), the inputted power at the dot is equal to the entropy production rate multiplied by the temperature 𝑇¤𝜎tot, i.e. ¤𝑊𝕐 =𝑞𝑒𝑉⟨𝐽 1,0 𝐿 (Γ)⟩𝑠𝑠 =𝑇¤𝜎tot. From the TS dynamics, we can also compute the total entropy production rate (or input power) in the steady-state as: 𝑇¤𝜎tot = 𝑞𝑒𝑉⟨𝐽 1,0 𝐿 (Γ)⟩TS+...

  5. [5]

    Maruyama, F

    K. Maruyama, F. Nori, and V. Vedral, Reviews of Modern Physics 81, 1 (2009)

    2009

  6. [6]

    H. S. Leff and A. F. Rex, Maxwell’s demon: entropy, information, computing (Princeton University Press, 1990)

    1990

  7. [7]

    Leff and A

    H. Leff and A. F. Rex, Maxwell’s Demon 2 Entropy, Classical and Quantum Information, Computing (CRC Press, 2002)

    2002

  8. [8]

    Szilard, Zeitschrift f ¨ur Physik 53, 840 (1929)

    L. Szilard, Zeitschrift f ¨ur Physik 53, 840 (1929)

    1929

Show all 106 references
  1. [9]

    Landauer, IBM journal of research and development 5, 183 (1961)

    R. Landauer, IBM journal of research and development 5, 183 (1961)

    1961

  2. [10]

    C. H. Bennett, International Journal of Theoretical Physics 21, 905 (1982). 25

    1982

  3. [11]

    Sagawa, Progress of theoretical physics 127, 1 (2012)

    T. Sagawa, Progress of theoretical physics 127, 1 (2012)

    2012

  4. [12]

    J. M. Parrondo, J. M. Horowitz, and T. Sagawa, Nature physics 11, 131 (2015)

    2015

  5. [13]

    Ehrich and D

    J. Ehrich and D. A. Sivak, Frontiers in Physics 11, 1108357 (2023)

    2023

  6. [14]

    J. M. R. Parrondo, J. Tabanera-Bravo, F. Fedele, and N. Ares, inCrossroad of Maxwell Demon, edited by X. Bouju and C. Joachim (Springer Nature Switzerland, Cham, 2024) pp. 1–31

    2024

  7. [15]

    Sagawa and M

    T. Sagawa and M. Ueda, Phys. Rev. Lett. 102, 250602 (2009)

    2009

  8. [16]

    J. M. Horowitz and M. Esposito, Physical Review X 4, 031015 (2014)

    2014

  9. [17]

    A. C. Barato, D. Hartich, and U. Seifert, New Journal of Physics 16, 103024 (2014)

    2014

  10. [18]

    A. E. Allahverdyan, D. Janzing, and G. Mahler, Journal of Statistical Mechanics: Theory and Experiment 2009, P09011 (2009)

    2009

  11. [19]

    J. M. Horowitz, Journal of Statistical Mechanics: Theory and Experiment 2015, P03006 (2015)

    2015

  12. [20]

    J. M. Horowitz and H. Sandberg, New Journal of Physics 16, 125007 (2014)

    2014

  13. [21]

    Herpich, K

    T. Herpich, K. Shayanfard, and M. Esposito, Physical Review E 101, 022116 (2020)

    2020

  14. [22]

    Strasberg, G

    P. Strasberg, G. Schaller, T. Brandes, and M. Esposito, Physical Review X 7, 021003 (2017)

    2017

  15. [23]

    Ptaszy ´nski and M

    K. Ptaszy ´nski and M. Esposito, Physical review letters 122, 150603 (2019)

    2019

  16. [24]

    Sagawa and M

    T. Sagawa and M. Ueda, Physical review letters 104, 090602 (2010)

    2010

  17. [25]

    M. L. Rosinberg and J. M. Horowitz, Europhysics Letters 116, 10007 (2016)

    2016

  18. [26]

    Shiraishi and T

    N. Shiraishi and T. Sagawa, Physical Review E 91, 012130 (2015)

    2015

  19. [27]

    Prech and P

    K. Prech and P. P. Potts, Physical Review Letters 133, 140401 (2024)

    2024

  20. [28]

    M. P. Leighton and D. A. Sivak, arXiv preprint arXiv:2406.10355 (2024)

    2024 arXiv

  21. [29]

    S ´anchez, P

    R. S ´anchez, P. Samuelsson, and P. P. Potts, Physical Review Research 1, 033066 (2019)

    2019

  22. [30]

    Amano, M

    S. Amano, M. Esposito, E. Kreidt, D. A. Leigh, E. Penocchio, and B. M. Roberts, Nature Chemistry 14, 530 (2022)

    2022

  23. [31]

    T. K. Saha, J. N. Lucero, J. Ehrich, D. A. Sivak, and J. Bechhoefer, Proceedings of the National Academy of Sciences 118, e2023356118 (2021)

    2021

  24. [32]

    Serreli, C.-F

    V. Serreli, C.-F. Lee, E. R. Kay, and D. A. Leigh, Nature 445, 523 (2007)

    2007

  25. [33]

    M. G. Raizen, Science 324, 1403 (2009)

    2009

  26. [34]

    Toyabe, T

    S. Toyabe, T. Sagawa, M. Ueda, E. Muneyuki, and M. Sano, Nature physics 6, 988 (2010)

    2010

  27. [35]

    J. V. Koski, V. F. Maisi, J. P. Pekola, and D. V. Averin, Proceedings of the National Academy of Sciences111, 13786 (2014)

    2014

  28. [36]

    J. V. Koski, A. Kutvonen, I. M. Khaymovich, T. Ala-Nissila, and J. P. Pekola, Physical review letters 115, 260602 (2015)

    2015

  29. [37]

    P. A. Camati, J. P. Peterson, T. B. Batalhao, K. Micadei, A. M. Souza, R. S. Sarthour, I. S. Oliveira, and R. M. Serra, Physical review letters 117, 240502 (2016)

    2016

  30. [38]

    Chida, S

    K. Chida, S. Desai, K. Nishiguchi, and A. Fujiwara, Nature communications 8, 15301 (2017)

    2017

  31. [39]

    Masuyama, K

    Y. Masuyama, K. Funo, Y. Murashita, A. Noguchi, S. Kono, Y. Tabuchi, R. Yamazaki, M. Ueda, and Y. Nakamura, Nature communications 9, 1291 (2018)

    2018

  32. [40]

    Ribezzi-Crivellari and F

    M. Ribezzi-Crivellari and F. Ritort, Nature Physics 15, 660 (2019)

    2019

  33. [41]

    Ciliberto, Physical Review X 7, 021051 (2017)

    S. Ciliberto, Physical Review X 7, 021051 (2017)

    2017

  34. [42]

    Freitas and M

    N. Freitas and M. Esposito, Physical Review Letters 129, 120602 (2022)

    2022

  35. [43]

    Freitas and M

    N. Freitas and M. Esposito, Physical Review E 107, 014136 (2023)

    2023

  36. [44]

    Penocchio, F

    E. Penocchio, F. Avanzini, and M. Esposito, The Journal of Chemical Physics 157 (2022)

    2022

  37. [45]

    Bilancioni, M

    M. Bilancioni, M. Esposito, and N. Freitas, The Journal of Chemical Physics 159 (2023)

    2023

  38. [46]

    Ptaszynski and M

    K. Ptaszynski and M. Esposito, arXiv preprint arXiv:2410.13375 (2024)

    2024 arXiv

  39. [47]

    Lassagne, Y

    B. Lassagne, Y. Tarakanov, J. Kinaret, D. Garcia-Sanchez, and A. Bachtold, Science 325, 1107 (2009)

    2009

  40. [48]

    G. A. Steele, A. K. H ¨uttel, B. Witkamp, M. Poot, H. B. Meerwaldt, L. P. Kouwenhoven, and H. S. van der Zant, Science 325, 1103 (2009)

    2009

  41. [49]

    Okazaki, I

    Y. Okazaki, I. Mahboob, K. Onomitsu, S. Sasaki, and H. Yamaguchi, Nature Communications 7, 11132 (2016)

    2016

  42. [50]

    Ekinci and M

    K. Ekinci and M. Roukes, Review of scientific instruments 76 (2005)

    2005

  43. [51]

    Vigneau, J

    F. Vigneau, J. Monsel, J. Tabanera, K. Aggarwal, L. Bresque, F. Fedele, F. Cerisola, G. Briggs, J. Anders, J. M. Parrondo, et al., Physical Review Research 4, 043168 (2022)

    2022

  44. [52]

    Y. Wen, N. Ares, F. Schupp, T. Pei, G. Briggs, and E. Laird, Nature physics 16, 75 (2020)

    2020

  45. [53]

    LaHaye, J

    M. LaHaye, J. Suh, P. Echternach, K. C. Schwab, and M. L. Roukes, Nature 459, 960 (2009)

    2009

  46. [54]

    A. Reed, K. Mayer, J. Teufel, L. Burkhart, W. Pfaff, M. Reagor, L. Sletten, X. Ma, R. Schoelkopf, E. Knill, et al., Nature Physics 13, 1163 (2017)

    2017

  47. [55]

    R. G. Knobel and A. N. Cleland, Nature 424, 291 (2003)

    2003

  48. [56]

    Moser, J

    J. Moser, J. G ¨uttinger, A. Eichler, M. J. Esplandiu, D. Liu, M. Dykman, and A. Bachtold, Nature nanotechnology 8, 493 (2013)

    2013

  49. [57]

    Tonekaboni, B

    B. Tonekaboni, B. W. Lovett, and T. M. Stace, arXiv preprint arXiv:1809.04251 (2018)

    2018 arXiv

  50. [58]

    Strasberg, C

    P. Strasberg, C. W. W ¨achtler, and G. Schaller, Physical Review Letters 126, 180605 (2021)

    2021

  51. [59]

    Vittoz, Low-power crystal and MEMS oscillators: the experience of watch developments (Springer Science & Business Media, 2010)

    E. Vittoz, Low-power crystal and MEMS oscillators: the experience of watch developments (Springer Science & Business Media, 2010)

    2010

  52. [60]

    Thommen, in Proceedings of the 25th European Solid-State Circuits Conference (IEEE, 1999) pp

    W. Thommen, in Proceedings of the 25th European Solid-State Circuits Conference (IEEE, 1999) pp. 146–149

    1999

  53. [61]

    Siniscalchi, F

    M. Siniscalchi, F. Silveira, and C. Galup-Montoro, IEEE Transactions on Circuits and Systems I: Regular Papers 67, 1846 (2020)

    2020

  54. [62]

    Freitas, J.-C

    N. Freitas, J.-C. Delvenne, and M. Esposito, Physical Review X 10, 031005 (2020)

    2020

  55. [63]

    Freitas, J.-C

    N. Freitas, J.-C. Delvenne, and M. Esposito, Physical Review X 11, 031064 (2021)

    2021

  56. [64]

    Gopal, M

    A. Gopal, M. Esposito, and N. Freitas, Physical Review B 109, 085421 (2024)

    2024

  57. [65]

    C. W. W ¨achtler, P. Strasberg, S. H. Klapp, G. Schaller, and C. Jarzynski, New journal of physics 21, 073009 (2019)

    2019

  58. [66]

    W ¨achtler, P

    C. W ¨achtler, P. Strasberg, and G. Schaller, Physical Review Applied 12, 024001 (2019)

    2019

  59. [67]

    Esposito, Physical Review E—Statistical, Nonlinear, and Soft Matter Physics 85, 041125 (2012)

    M. Esposito, Physical Review E—Statistical, Nonlinear, and Soft Matter Physics 85, 041125 (2012)

    2012

  60. [68]

    Gorelik, A

    L. Gorelik, A. Isacsson, M. Voinova, B. Kasemo, R. Shekhter, and M. Jonson, Physical review letters 80, 4526 (1998). 26

    1998

  61. [69]

    Rao and M

    R. Rao and M. Esposito, Physical Review X 6, 041064 (2016)

    2016

  62. [70]

    Pietzonka, Physical Review Letters 128, 130606 (2022)

    P. Pietzonka, Physical Review Letters 128, 130606 (2022)

    2022

  63. [71]

    Rao and M

    R. Rao and M. Esposito, The Journal of chemical physics 149 (2018)

    2018

  64. [72]

    Tuominen, R

    M. Tuominen, R. Krotkov, and M. Breuer, Physical review letters 83, 3025 (1999)

    1999

  65. [73]

    D. V. Scheible and R. H. Blick, Applied physics letters 84, 4632 (2004)

    2004

  66. [74]

    A. Erbe, C. Weiss, W. Zwerger, and R. Blick, Physical Review Letters 87, 096106 (2001)

    2001

  67. [75]

    A. V. Moskalenko, S. N. Gordeev, O. F. Koentjoro, P. R. Raithby, R. W. French, F. Marken, and S. E. Savel’ev, Physical Review B—Condensed Matter and Materials Physics 79, 241403 (2009)

    2009

  68. [76]

    D. R. K ¨onig and E. M. Weig, Applied Physics Letters 101 (2012)

    2012

  69. [77]

    Isacsson, L

    A. Isacsson, L. Gorelik, M. Voinova, B. Kasemo, R. Shekhter, and M. Jonson, Physica B: Condensed Matter 255, 150 (1998)

    1998

  70. [78]

    A. A. Clerk and S. Bennett, New Journal of Physics 7, 238 (2005)

    2005

  71. [79]

    Weiss and W

    C. Weiss and W. Zwerger, Europhysics Letters 47, 97 (1999)

    1999

  72. [80]

    Boese and H

    D. Boese and H. Schoeller, Europhysics Letters 54, 668 (2001)

    2001

  73. [81]

    Braig and K

    S. Braig and K. Flensberg, Physical Review B 68, 205324 (2003)

    2003

  74. [82]

    Sapmaz, Y

    S. Sapmaz, Y. M. Blanter, L. Gurevich, and H. Van der Zant, Physical Review B 67, 235414 (2003)

    2003

  75. [83]

    Novotn `y, A

    T. Novotn `y, A. Donarini, and A.-P. Jauho, Physical review letters 90, 256801 (2003)

    2003

  76. [84]

    Pistolesi, Physical Review B—Condensed Matter and Materials Physics 69, 245409 (2004)

    F. Pistolesi, Physical Review B—Condensed Matter and Materials Physics 69, 245409 (2004)

    2004

  77. [85]

    Novotn `y, A

    T. Novotn `y, A. Donarini, C. Flindt, and A.-P. Jauho, Physical review letters 92, 248302 (2004)

    2004

  78. [86]

    Pistolesi and R

    F. Pistolesi and R. Fazio, arXiv preprint cond-mat/0408056 (2004)

    2004 arXiv

  79. [87]

    Fedorets, L

    D. Fedorets, L. Y. Gorelik, R. I. Shekhter, and M. Jonson, Physical review letters 92, 166801 (2004)

    2004

  80. [88]

    Bennett and A

    S. Bennett and A. Clerk, Physical Review B—Condensed Matter and Materials Physics 74, 201301 (2006)

    2006

  81. [89]

    D. W. Utami, H.-S. Goan, C. Holmes, and G. Milburn, Physical Review B—Condensed Matter and Materials Physics 74, 014303 (2006)

    2006

  82. [90]

    Y. M. Blanter, O. Usmani, Nazarov, and Y. V., Phys. Rev. Lett.93, 136802 (2004)

    2004

  83. [91]

    R. J. Matthys, New York (1983)

    1983

  84. [92]

    Culhane, M

    O. Culhane, M. J. Kewming, A. Silva, J. Goold, and M. T. Mitchison, New Journal of Physics 26, 023047 (2024)

    2024

  85. [93]

    C. W. W ¨achtler and J. Cerrillo, Physical Review Applied 18, 014011 (2022)

    2022

  86. [94]

    R. E. Spinney and I. J. Ford, Physical review letters 108, 170603 (2012)

    2012

  87. [95]

    H. K. Lee, C. Kwon, and H. Park, Physical Review Letters 110, 050602 (2013)

    2013

  88. [96]

    Grelier, D

    M. Grelier, D. A. Sivak, and J. Ehrich, Physical Review E 109, 034115 (2024)

    2024

  89. [97]

    Celani, S

    A. Celani, S. Bo, R. Eichhorn, and E. Aurell, Physical review letters 109, 260603 (2012)

    2012

  90. [98]

    Murashita and M

    Y. Murashita and M. Esposito, Physical Review E 94, 062148 (2016)

    2016

  91. [99]

    K. H. Kim and H. Qian, Phys. Rev. Lett. 93, 120602 (2004)

    2004

  92. [100]

    K. H. Kim and H. Qian, Phys. Rev. E 75, 022102 (2007)

    2007

  93. [101]

    Munakata and M

    T. Munakata and M. Rosinberg, Journal of Statistical Mechanics: Theory and Experiment 2012, P05010 (2012). Appendix A: Second Law : Average Description At the average description, the nonequilibrium system entropy at a given time𝑡 of the composite system is given as, 𝑆(𝑡) = ∑︁...

    2012

  94. [102]

    This situation arises when one can only access the position and velocity of the subsystem 𝕏, whereas changes in the discrete states are inaccessible

    Marginal UD dynamics We will first analyze the case with only continuous degrees of freedom Γ, while integrating out the discrete electronic state𝑦. This situation arises when one can only access the position and velocity of the subsystem 𝕏, whereas changes in the discrete sta...

  95. [103]

    Here, we will be left with the marginal distribution𝑃 𝕐 𝑡(𝑦) = ∫ 𝑑Γ𝑃𝑡(Γ,𝑦), after integrating out continuous degrees of freedom

    Marginal MJ dynamics Now we will shift our focus to the complementary case where we can only keep track of the electronic degrees of freedom𝑦. Here, we will be left with the marginal distribution𝑃 𝕐 𝑡(𝑦) = ∫ 𝑑Γ𝑃𝑡(Γ,𝑦), after integrating out continuous degrees of freedom. Integ...

  96. [104]

    (C5) where the ensemble average is over all the trajectories up to time t

    We will demonstrate below the existence of a generalized integrated fluctuation theorem (IFT) for the partial entropy production, ˆ𝜎𝑒(𝑚)(𝑡), given as: ⟨𝑒− ˆ𝜎𝕐(𝑡)/𝑘𝐵⟩ = 1 and ⟨𝑒− ˆ𝜎𝕏(𝑡)/𝑘𝐵⟩ = 1. (C5) where the ensemble average is over all the trajectories up to time t. The IFT ...

  97. [105]

    Electronic partial entropy production For the electronic partial entropy production, this requirement can be further simplified to ˆ𝜎𝕐(𝑡) =𝑘𝐵 log P[𝑦, Γ] P∗ 1[𝑦†, Γ†] (C7) =− ∑︁ 𝜌 ˆ𝑄𝜌(𝑡) 𝑇𝕐 + log P𝕏[𝑦, Γ|𝑦0, Γ0]𝑃0(𝑦0.Γ0) P𝕏∗[𝑦†, Γ†|𝑦𝑡, Γ𝑡]𝑃𝑡(𝑦𝑡, Γ𝑡) (C8) =− ∑︁ 𝜌 ˆ𝑄𝜌(𝑡) 𝑇𝕐 +𝛥ˆ𝑠...

  98. [106]

    Partial entropy production : UD dynamics For the partial entropy production due to UD dynamics, the requirement of Eq. (??) implies that the above equation can be further simplified as, ˆ𝜎𝕏(𝑡) =𝑘𝐵 log P[𝑦, Γ] P∗ 2[𝑦†, Γ†] (C15) =− ˆ𝑄𝕏(𝑡) 𝑇𝕏 +𝑘𝐵 log P𝕐[𝑦, Γ|𝑦0, Γ0]𝑃0(𝑦0.Γ0) P∗ ...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.