Pith. sign in

REVIEW 3 major objections 5 minor 14 references

Chaos in Nonequilibrium Two-Temperature $(T_x, T_y)$ Nos\'e-Hoover Cell Models

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A two-thermostat particle cell is chaotic at every studied anisotropy, and its dissipation grows as $\delta^{2.44}$ rather than the linear-response $\delta^2$ law.

desk verdict A careful, honest numerical study of a known toy model; the new scaling numbers are plausible and internally checked, though the fits rest on few points. read the letter →

arxiv 2507.10863 v7 pith:IJ32EW4Y submitted 2025-07-14 cond-mat.stat-mech nlin.CD

Hesam Arabzadeh , Carol Griswold Hoover , William Graham Hoover , Brad Lee Holian This is my paper · ORCID
classification cond-mat.stat-mechnlin.CD PACS 05.45.-a05.70.Ln
keywords Nosé–Hooverthermostattwo-temperatureLyapunovspectrumphase-spacecontractionentropyproductionKaplan–Yorkedimensionanisotropytimereversibility
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

Using a single particle in a periodic square cell with four soft repulsive corners, the paper couples the $x$ and $y$ motions to two deterministic thermostats held at different temperatures. From very long simulations ($10^{11}$ steps) it obtains the full six-dimensional Lyapunov spectrum and shows the system is chaotic for every thermostat pair studied. The central quantitative claim is that the total phase-space contraction rate $\Lambda$, the Lyapunov sum and a proxy for entropy production, grows with the temperature anisotropy $\delta = 0.5 - T_y$ faster than the quadratic law expected from linear-response theory. Two fits are offered, $\Lambda = -0.95\,\delta^{2.44}$ and $\Lambda = -0.45\,\delta^2 - 1.10\,\delta^4$. The paper also reports an empirical linear link between dissipation and attractor-dimension loss, $\Lambda \approx (D_{\rm KY} - 6)/3$, and shows the flow remains time-reversible despite being dissipative.

What carries the argument

The central object is the two-temperature extension of Nosé–Hoover dynamics: two extra variables $\zeta_x$, $\zeta_y$ feed back onto $p_x$, $p_y$ so that $\langle p_x^2\rangle$ relaxes to $T_x$ and $\langle p_y^2\rangle$ to $T_y$, while the particle is confined by a smooth four-corner repulsive potential in a periodic $2\times2$ cell. The identity that carries the quantitative argument is the phase-space divergence $\Lambda = \nabla_\Gamma \cdot \dot{\Gamma} = -\zeta_x - \zeta_y$, derived two ways in the appendix and used to validate the Lyapunov-sum measurement of $\Lambda$. The Lyapunov spectrum itself, computed by orthonormalizing tangent vectors over $10^{11}$ steps to an uncertainty below 0.001, is what makes the nonlinear $\Lambda(\delta)$ scaling and the dimension–dissipation relation numerically credible.

What would settle it

Run the same cell to much smaller $T_y$ (for example 0.02, 0.01, and 0.005) with a regularized thermostat that avoids the $T_y=0$ singularity, and check whether the measured $\Lambda$ and $D_{\rm KY}$ meet the linear extrapolation; if they do not, or if the residuals $3\Lambda/(D_{\rm KY}-6) - 1$ exceed one standard deviation, the empirical dimension–dissipation relation and the extrapolated endpoint are falsified. The scaling-law claim itself would be falsified if the new points pull the fitted exponent from 2.44 back toward 2.

Watch

Extended reading notes

Core claim

The central discovery is that in this minimal two-temperature thermostat model the dissipation is nonlinear in the driving. With $T_x$ fixed at 0.5 and $T_y$ decreasing from 0.5 to 0.05, the measured contraction rate rises from 0 to $-0.139$, and its $\delta$-dependence is described both by $\Lambda = -0.95\,\delta^{2.44}$ and by $\Lambda = -0.45\,\delta^2 - 1.10\,\delta^4$. The power law fits the near-equilibrium points marginally better; the quadratic-plus-quartic form better captures the strongly driven regime while remaining consistent with linear-response theory as $\delta \to 0$. Over the same scan, the approximate Kaplan–Yorke dimension falls from 6.00 to 5.49, and the empirical relation $\Lambda \approx (D_{\rm KY} - 6)/3$ assigns each unit of dissipation to the three $y$-directed phase-space variables ($y, p_y, \zeta_y$). Time reversal of momenta and thermostat variables retraces the trajectory, so microscopic reversibility and macroscopic entropy production coexist here.

Load-bearing premise

The load-bearing numerical step is the $T_y=0$ limit, obtained by linearly extrapolating both $\Lambda$ and $D_{\rm KY}$ from finite-$\delta$ data even though the $y$-thermostat equation is singular at $T_y=0$; the empirical $\Lambda \approx (D_{\rm KY} - 6)/3$ relation leans on that extrapolated endpoint.

Editorial extensions

If this is right

  • If the $\delta^{2.44}$ scaling holds, linear-response theory's $\delta^2$ law for entropy production is not the whole story even at moderate anisotropy in a single-particle deterministic model.
  • The relation $\Lambda \approx (D_{\rm KY} - 6)/3$, if it survives closer to $T_y=0$, gives a practical bridge: attractor-dimension measurements can estimate entropy production in anisotropic thermostatted systems.
  • Chaos for all studied thermostat pairs means deterministic thermostats can sustain fractal nonequilibrium attractors without any stochastic noise, supporting their continued use in nonequilibrium molecular dynamics.
  • Because the quadratic-plus-quartic fit has a negative quartic term, it predicts that strong driving suppresses dissipation relative to a pure power law, a signature that could be sought in other driven thermostat models.

Reading between the lines

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

  • An extension the authors leave implicit: a three-dimensional two-temperature version with separate $x$, $y$, $z$ thermostats would directly test whether the $\delta^{2.44}$ exponent survives and whether the dimension–dissipation prefactor tracks the number of phase-space variables controlled by the colder thermostat.
  • My reading of the factor $1/3$ is that it counts the three $y$-directed phase-space variables; swapping the roles of $T_x$ and $T_y$ (or varying both) would test whether the coefficient always equals one over the number of colder-thermostat variables.
  • The strongly driven run shows a sixth momentum moment $R_6 \approx 160$, far above the Gaussian value of 15, hinting that rare large excursions in $p_y$ may contribute disproportionately to the dissipation; the paper reports this statistic but does not connect it to the scaling exponent.
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 manuscript studies a single particle confined in a periodic 2x2 cell with smooth repulsive corners and coupled to two Nosé-Hoover thermostats at temperatures Tx and Ty. The authors integrate the six-dimensional equations of motion, compute the full Lyapunov spectrum, and report that the total phase-space contraction rate Λ becomes increasingly negative with the anisotropy δ = 0.5 − Ty, following approximately Λ = −0.95 δ^2.44 or, alternatively, a quadratic-plus-quartic form Λ = −0.45 δ^2 − 1.10 δ^4. They also report an empirical linear relation between Λ and the Kaplan-Yorke dimension deficit, non-Gaussian momentum statistics under strong driving, and the strict time-reversibility of the dissipative flow.

Significance. If the reported scaling is robust, the paper provides a minimal deterministic model in which a complete Lyapunov spectrum can be computed with high precision and used to test linear-response predictions for entropy production beyond equilibrium. The paper's strengths are its clearly stated equations, the internal cross-check between the Lyapunov-sum route and the phase-space divergence route (Tables I and II), and the explicit labeling of its scaling laws as empirical fits. The main limitations are the absence of statistical uncertainties on the fit parameters, the Ty=0 extrapolation used for one auxiliary relation, and the lack of released code and data. On balance, the central chaos claim appears supported, but the quantitative scaling claim needs additional statistical grounding.

major comments (3)
  1. [Section III.C, Eqs. (16) and (19)] The central quantitative claim is that Λ deviates from the quadratic δ^2 scaling, yet the supporting fits are presented without confidence intervals on the parameters or any goodness-of-fit measures. With only ten data points (δ = 0.05, 0.10, ..., 0.45), the exponent 2.44 and the quartic coefficient 1.10 could be sensitive to the last few points and to the chosen fitting form. Please report parameter uncertainties (for example, via bootstrap or covariance analysis), residual plots, and a model comparison such as AIC or reduced chi-square, and state explicitly whether the deviation from the pure quadratic baseline is statistically significant.
  2. [Section III.C, Eq. (20) and Fig. 6] The empirical linear relation Λ ≈ (D_KY − 6)/3 is supported by residual analysis, but its strongest point (δ = 0.5) is obtained by linear extrapolation of both Λ and D_KY to Ty = 0. The equation of motion for ζ_y is singular at Ty = 0, as the paper itself acknowledges via the 'Toda demon' remark, so the extrapolation is not physically justified. Because this point is the most extreme and therefore the most influential for the fitted line in Fig. 6, the relation should be presented as tentative, or the analysis should be repeated with a regularized low-temperature thermostat or with data restricted to Ty > 0.
  3. [Section III.A and Section II.B.2] The main text states that each Lyapunov spectrum contains one to three positive exponents and identifies chaos, but the actual six exponents are only referenced as Supplementary Table S1 and are not included in the manuscript as provided. Since the chaos claim is central, please include the full spectrum (or at least the positive exponents and their uncertainties) in the main text or a data availability appendix, and provide the code and input parameters needed to reproduce Tables I and II.
minor comments (5)
  1. [Appendix, Eq. (A.14)] In the derivation of ⟨ζx⟩, the second term should be ⟨Ty⟩⟨ζy⟩, not ⟨Tx⟩⟨ζy⟩; as written, the equation is inconsistent with the following ratio ⟨ζx⟩/⟨ζy⟩ = −Ty/Tx.
  2. [Section II.B.1] The statement that Lyapunov exponents converge with 'an uncertainty of 0.001' needs a description of how the uncertainty was estimated (for example, block averaging or time averages).
  3. [Table I] The values are rounded to three decimal places; for δ = 0.05, Table I gives Λ = −0.001 while Table II gives Λ = −0.0008. Please state the rounding convention and give the raw values or uncertainties.
  4. [Section III.D] The 'complete set of values and uncertainties' for R4 and R6 is only in Supplementary Table S2; please include the table or a data availability statement so the non-Gaussian claim can be checked.
  5. [Section III.C and Fig. 5] The caption refers to 'error bars' on the data, but no error-bar magnitude is defined; please state the one-sigma uncertainty on Λ and how it was obtained.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the nonlinear scaling laws and the dimension-dissipation relation are explicitly empirical fits to simulation data, not derived predictions, and the Lyapunov contraction is independently cross-checked.

full rationale

The paper's central claims are empirical fits, not hidden predictions. Section III.C explicitly says 'we fit the total phase-space contraction rate Λ to two different models' and calls the resulting power-law 'the empirical relation' (Eq. 16). The quadratic-plus-quartic form is likewise presented as a fit, with the paper stating 'we present both, leaving the precise functional form open to interpretation.' Reporting a fitted scaling is not circular merely because the fitted curve passes through the data used for the fit. The claim that this deviates from linear-response theory is a comparison against an external expectation, not a restatement of an input. The relation Λ ≈ (D_KY − 6)/3 is also explicitly labeled 'empirical scaling relation' and is derived from the same Lyapunov spectra; although it is not an independent prediction, the paper does not claim it is one. The Appendix provides an independent validation of Λ via the phase-space divergence identity Λ = −ζx − ζy, checked numerically in Table II, so the Lyapunov sum is not an unverified input. Citations to prior work by the same authors ([6], [9], [12], [14]) are used for model setup and standard identities, not to force a conclusion. The Ty = 0 extrapolation caveat in Section III.C and Fig. 6 is a robustness limitation, not a circular step. No load-bearing step reduces, by the paper's own equations or by self-citation, to its own inputs.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The scaling laws are determined by six fit coefficients (two for the power law, two for the quartic, plus empirical relation coefficients) and by several hand-chosen model parameters (τ=1, potential). The paper relies on standard identities (K-Y formula, NESS averages) and on one ad hoc extrapolation to Ty=0. No new physical entities are introduced.

free parameters (6)
  • Power-law prefactor = -0.95
    Least-squares fit to simulated Λ(δ) data (Eq. 16).
  • Power-law exponent = 2.44
    Least-squares fit to simulated Λ(δ) data (Eq. 16).
  • Quadratic coefficient = -0.45
    Fit coefficient in quadratic-plus-quartic model (Eq. 19).
  • Quartic coefficient = -1.10
    Fit coefficient in quadratic-plus-quartic model (Eq. 19).
  • Thermostat relaxation times τx, τy = 1 (both)
    Set to unity by hand (Section II.A.1); not fitted but chosen.
  • Corner potential parameters = corners at (±1,±1), exponent 4, cutoff 1
    Chosen from previous work [9]; these shape the dynamics and influence all results.
assumptions (5)
  • domain assumption Identification of the negative phase-space contraction rate with entropy production, Ṡ_prod = -kΛ (Eq. 15).
    This standard relation from Posch and Hoover (1989) is invoked to interpret Λ as entropy production; it is not derived in this paper.
  • standard math The Kaplan-Yorke formula D_KY = j + (∑_{i=1}^j λ_i)/|λ_{j+1}| estimates the attractor dimension.
    Used in Section III.A to compute D_KY; standard.
  • domain assumption Near equilibrium, linear response theory gives entropy production ∝ δ^2 (Eq. 18).
    Used to motivate the quadratic-plus-quartic fit as consistent with linear response near equilibrium.
  • domain assumption Long-time averages of ζ_x and ζ_y vanish in the steady state, and the system reaches a unique NESS (Appendix A).
    Required for the identity ⟨Ṡ_prod⟩ = ⟨ζ_x + ζ_y⟩ and the energy balance derivation.
  • ad hoc to paper Λ and D_KY can be linearly extrapolated to Ty = 0.
    Used to add the final point in Fig. 6 despite the singularity in the thermostat equation at Ty = 0.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Chaos in Nonequilibrium Two-Temperature $(T_x, T_y)$ Nos\'e-Hoover Cell Models." pith.science (2026). https://pith.science/paper/IJ32EW4Y

@misc{pith2026250710863,
  author       = {Pith},
  title        = {Pith review of: Chaos in Nonequilibrium Two-Temperature $(T_x, T_y)$ Nos\'e-Hoover Cell Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJ32EW4Y}},
  note         = {Machine review of arXiv:2507.10863}
}
abstract

We revisit a two-temperature Nos\'e-Hoover wanderer particle embedded in a two-dimensional periodic 2x2 cell with four smooth repulsive corners at $(x,y) = (\pm 1, \pm 1)$ to explore chaos with anisotropic thermostatting. The model employs separate thermostats in the x and y directions, enabling controlled deviations from equilibrium. By integrating the full six-dimensional equations of motion and computing the complete Lyapunov spectrum, we confirm chaos and quantify phase-space contraction with high numerical precision. The total contraction rate, interpreted as entropy production, increases nonlinearly with the thermostat anisotropy, deviating from the quadratic dependence expected from linear-response theory, $\Lambda\propto -\delta^{2}$. We compare two fits for $\Lambda$ as a function of $\delta = 0.5 -T_y$: 1) a power law, $\Lambda\propto -\delta^{2.44}$, 2) a quadratic-plus-quartic expansion. While the former captures low-driving behavior slightly better, the latter more accurately describes the strongly driven regime and remains consistent with linear response theory near equilibrium. An empirical linear relation between dissipation and phase-space dimensionality loss is also identified, $\Lambda\propto (D_{KY}-6) / 3$, where $D_{KY}$ is the approximate Kaplan-Yorke dimension. Our results demonstrate that nonlinear dissipation scaling emerges naturally even in minimal driven systems. Momentum statistics show significant non-Gaussian behavior under strong driving. Despite its dissipative nature, the model remains strictly time-reversible, offering a pedagogically rich example of microscopic reversibility coexisting with macroscopic entropy production.

Figures

Figures reproduced from arXiv: 2507.10863 by the authors.

Figure 1
Figure 1. FIG. 1: Contour plot of the corner potential [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: Particle trajectories under Hamiltonian mechanics, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3: Particle trajectories under two-thermostat Nos´e– [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: FIG. 5: Total phase-space contraction rate, Λ, as a function [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 4
Figure 4. Figure 4: FIG. 4: Poincar´e sections showing the position angle ver [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 6
Figure 6. Figure 6: FIG. 6: Residual deviation from the phase space [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 13 canonical work pages

  1. [1]

    W. G. Hoover, Physical review A 31, 1695 (1985)

    work page 1985

  2. [2]

    Nos´ e, Molecular Physics52, 255 (1984)

    S. Nos´ e, Molecular Physics52, 255 (1984)

    work page 1984

  3. [3]

    W. G. Hoover and C. Hoover, Computational Methods in Science & Technology 30 (2024)

    work page 2024

  4. [4]

    D. J. Evans and G. Morriss, Statistical Mechanics of Nonequilibrium Liquids (Cambridge University Press, 2008)

    work page 2008

  5. [5]

    G. J. Martyna, M. L. Klein, and M. Tuckerman, The Journal of chemical physics 97, 2635 (1992)

    work page 1992

  6. [6]

    W. G. Hoover, E. Craig, H. A. Posch, B. L. Holian, and C. G. Hoover, Chaos An Interdisciplinary Journal of Non- linear Science 1, 343 (1991)

    work page 1991

  7. [7]

    T. M. Squires and S. R. Quake, Reviews of modern physics 77, 977 (2005)

    work page 2005

  8. [8]

    D. G. Cahill, P. V. Braun, G. Chen, D. R. Clarke, S. Fan, K. E. Goodson, P. Keblinski, W. P. King, G. D. Mahan, A. Majumdar, et al., Applied physics reviews 1 (2014)

    work page 2014

Show all 14 references
  1. [9]

    Hoover and C

    W. Hoover and C. Hoover, Computational Methods in Science and Technology 21 (2015)

    2015

  2. [10]

    D. J. Evans, E. G. D. Cohen, and G. P. Morriss, Physical review letters 71, 2401 (1993)

    1993

  3. [11]

    Benettin, L

    G. Benettin, L. Galgani, and J.-M. Strelcyn, Physical Review A 14, 2338 (1976)

    1976

  4. [12]

    H. A. Posch and W. G. Hoover, Physical Review A 39, 2175 (1989)

    1989

  5. [13]

    B´ alint, H

    P. B´ alint, H. Bruin, and D. Terhesiu, Probability Theory and Related Fields 186, 159 (2023)

    2023

  6. [14]

    B. L. Holian, A. F. Voter, and R. Ravelo, Physical Review E 52, 2338 (1995)

    1995

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.