Pith. sign in

REVIEW 4 major objections 5 minor 28 references

The paper derives the second law of thermodynamics from a single macrostate probability condition and argues that locally nonchaotic systems—exemplified by Knudsen-gas cells—can violate it.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 00:31 UTC pith:3X7EQ7A6

load-bearing objection The macrostate-level framing is genuinely new, but the derivation rests on unproven smoothness and a near-tautological definition of full chaoticity, while the Knudsen boundary claim collapses once you use the flux-weighted wall emission condition. the 4 major comments →

arxiv 2607.28765 v1 pith:3X7EQ7A6 submitted 2026-07-30 cond-mat.stat-mech physics.class-ph

Deriving the second law of thermodynamics and exploring its boundaries

Yu Qiao This is my paper
classification cond-mat.stat-mech physics.class-ph
keywords second law of thermodynamicsBoltzmann entropymacrostate probabilityphase-space continuity equationKnudsen gasintrinsic nonequilibriumthermodynamic limit
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper aims to prove that the second law of thermodynamics (entropy never decreases in an isolated system) follows from one statistical condition: the probability f of a macrostate increases monotonically with the number Ω of microstates available to it. From that condition and a continuity equation for f in phase space, the paper derives Ṡ ≥ 0 for fully chaotic systems without needing the equations of motion. It then claims that in locally nonchaotic systems, such as a Knudsen gas with mean free path larger than the container, ∂f/∂Ω can be non-positive, so entropy can decrease spontaneously, enabling cold-to-hot heat transfer and work extraction from a single thermal reservoir. If correct, this would give a general proof of the second law and define a concrete boundary past which ordinary thermodynamics ceases to apply.

Core claim

The paper's central claim is that the monotonicity ∂f/∂Ω > 0 — which includes Boltzmann's equal a priori probabilities as a special case — is the only condition needed to derive the second law. Using the macrostate-level continuity equation ∂f/∂t = −Ω̇ ∂f/∂Ω − f ∂Ω̇/∂Ω in a phase space organized as a funnel by increasing Ω, it shows by a slice-by-slice argument that Ω̇ ≥ 0 for every macrostate, hence Ṡ ≥ 0. For fully chaotic systems, microstate probability ρ depends only on energy, guaranteeing ∂f/∂Ω > 0; the derivation therefore avoids molecular-chaos assumptions and applies equally to classical and quantum systems. The paper then argues that locally nonchaotic systems can have ∂f/∂Ω ≤ 0,

What carries the argument

The argument rests on a macrostate-level continuity equation (Eq. 1) that governs the flow of macrostate probability f along the Ω-axis of phase space, combined with the positive-monotonicity condition ∂f/∂Ω > 0. Slice-by-slice induction from the bottom of the funnel-shaped phase space converts these into Ω̇ ≥ 0. The boundary exploration uses a Knudsen-gas model whose microstate probability is non-Boltzmann, ρ ∝ e^{-βε}/κ_p (inversely proportional to particle speed), where κ_p is the product of the particles' momentum magnitudes; this weighting is what produces the lower kinetic temperature and the predicted violation.

Load-bearing premise

The paper's boundary claim rests on the assumption that a diffuse thermal wall at temperature T drives a Knudsen-gas interior into the steady state ρ ∝ e^{-βε}/κ_p (a 1/v probability weighting) with kinetic temperature T_k = 2T/3 < T; if the wall instead maintains the standard Maxwell–Boltzmann interior distribution at T, the predicted entropy decrease and second-law violation collapse.

What would settle it

Run a direct Monte Carlo simulation of a Knudsen gas cell with diffuse walls whose emission law respects detailed balance (emitted flux proportional to v times the Boltzmann distribution at T). If the steady-state velocity distribution inside is Maxwellian with kinetic temperature T, then the paper's T_k = 2T/3 steady state does not exist and the claimed boundary is an artifact. Equivalently, measure the speed distribution in a small dilute-gas cell held at uniform wall temperature T: observing T_k = T would disprove the counterexample.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Fully chaotic isolated systems—classical or quantum, far or near equilibrium—obey Ṡ ≥ 0 without assuming molecular chaos or a specific dynamics.
  • Entropy increase appears as a statistical consequence of phase-space structure, not as a dynamical law, so it remains compatible with micro-reversibility and Poincaré recurrence.
  • The second law holds only when macrostate probability rises with microstate number; systems outside that regime (locally nonchaotic Knudsen-like gases) can spontaneously decrease entropy.
  • The thermodynamic limit is not the boundaries of the second law; the actual boundary is the sign of ∂f/∂Ω for the macrostates present.
  • In the paper's counterexample, cyclic insertion and removal of frictionless walls in a Knudsen gas extracts work from a single reservoir, violating the Kelvin–Planck statement.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the Knudsen-gas boundary condition is correct, analogous 1/v weighting should appear in other rarefied or confined systems (nanopores, strong rarefaction), giving a simple experimental check: measure the kinetic temperature in a closed dilute-gas cell and see whether it reads 2T/3 rather than T.
  • The criterion ∂f/∂Ω > 0 may serve as a necessary condition for applying thermodynamics; it could be used to audit proposed engines or refrigerators that claim to violate the second law—if they rely on nonchaotic dilute gases, they are consistent with the paper's boundary, if they rely on chaotic systems they are not.
  • One could extend the derivation by replacing Ω with any coarse-grained measure of accessible volume; this would generalize the proof to other entropy definitions (e.g., Gibbs entropy) and possibly to information-theoretic settings.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes a proof of the second law of thermodynamics from a macrostate-level continuity equation. It defines f as the macrostate probability and Ω as the number of microstates, asserts Eq. (1), and claims that if ∂f/∂Ω > 0, a slice-by-slice induction yields Ω̇ ≥ 0 and hence Ṡ ≥ 0. It presents ℬ (equal a priori probabilities) as a sufficient condition and claims that full chaoticity implies ℬ. The second half introduces Knudsen-gas models said to exhibit intrinsic nonequilibrium with ρ ∝ e^{-βε}/κ_p, T_k = 2T/3, and ∂f/∂Ω ≤ 0, implying entropy decrease and violations of the Kelvin and Clausius statements. Appendices contain further models and discussion.

Significance. If the proof were valid, this would be a significant contribution: a derivation of the second law independent of microscopic dynamics and a concrete boundary condition for its validity. The paper's honesty in stating that Eq. (1) alone has no arrow and in identifying ℬ as an assumption is a strength. However, the central proof is not rigorous, and the main physical example is based on a boundary condition that standard kinetic theory shows is not a thermal wall. The manuscript does not provide machine-checked proofs, reproducible data for the simulations/experiments, or a derivation of Eq. (10) from Hamiltonian dynamics. The potential significance is therefore not realized as written.

major comments (4)
  1. [§2.4–2.5, especially Eqs. (1), (4), (5)] The derivation of Ω̇ ≥ 0 is not a proof. The continuity equation (1) assumes f is a smooth density on an Ω-axis and Ω̇ is a well-defined smooth velocity; but Ω is a macrostate sum over discrete microstates, and the paper's finite layer thickness δΩ is never reconciled with the derivatives. More importantly, the slice-by-slice argument uses inconsistent hypotheses: Eq. (3) (f = ρΩ) assumes a global constant ρ, while the 'initial condition f = 1 at L2, f = 0 elsewhere' introduced at L2 violates that assumption. The inference 'if ∂Ω̇/∂Ω > 0 at L2, then at nearby L1 ∂Ω̇/∂Ω ≥ 0, and since Ω̇ ≥ 0 at L1, Ω̇ ≥ 0 at L2' does not follow; a derivative sign at a lower point does not control the value at an upper point. The closure of subspace A1A2B1B2 is asserted, not established. Thus Inequality (5) — and with it the central claim Inequality (6) — is unsupported.
  2. [§1 and §2.7] The claimed justification of ℬ via 'full chaoticity' is circular. Full chaoticity is defined on p. 3 as the condition that ρ 'does not explicitly depend on the EOM but is constrained only by normalization and energy conservation,' which is essentially ℬ itself. Section 2.7 then concludes that under this condition ℬ holds, and with ℬ the second law follows. The logic is therefore an assumption, not a derivation from chaos. The statement on p. 15 that equations (8)–(9) 'alone are sufficient to obtain these equilibrium distributions' because 'entropy maximization (Inequality 6) is itself a consequence of ℬ' confirms that the second law is being put in as input rather than derived.
  3. [§4.1, Eq. (10); Appendix 1.4] The Knudsen-gas steady state is physically incorrect for a thermal wall. At a diffuse wall at temperature T, emitted particles have a flux distribution ∝ v_n f_MB(v); free streaming then gives the interior one-particle distribution f_MB(v), hence T_k = T. The 1/|p| weighting in Eq. (10) is obtained only by sampling reflected speeds from the unweighted Maxwell–Boltzmann speed distribution with a random direction, which violates detailed balance at the wall. Appendix 1.4's assertion that 'no reasonable boundary condition can keep the Knudsen-gas cells in equilibrium' is therefore not correct: the standard diffuse boundary condition does. If Eq. (10) fails, the subsequent 'intrinsic nonequilibrium' state, the ∂f/∂Ω ≤ 0 curve in §4.2, and the claimed cold-to-hot transfer and work-from-single-reservoir are artifacts of the chosen boundary rule, not consequences of local nonchaoticity.
  4. [§4.2, Eqs. (11)–(12), Fig. 3(b)] Even granting the boundary rule, the parametric family n_w(K) = ξ_a K^a e^{-β_a K} is not shown to describe macrostates of the Knudsen gas. Equation (11) evaluates ln f by inserting n_w into the equilibrium combinatorial expression for an ideal gas, but f is supposed to be the probability of a macrostate, i.e., a sum over the microstates in that macrostate; imposing a kinetic-energy distribution after wall collisions does not fix the macrostate probability. The additive constants C_1, C̅_1, e_0 are declared 'independent of a' without proof; if they are not, the slope of the ln f versus ln Ω curve in Fig. 3(b) is undetermined. The observation that ∂f/∂Ω ≤ 0 between a = 1/2 and a = 1 is a property of an ad hoc interpolation, not a demonstrated property of any Hamiltonian system.
minor comments (5)
  1. [Typos] Fig. 5 caption: 'Monte Calo' should be 'Monte Carlo'; Appendix 1.3: 'evolute' should be 'evolve'; inconsistent use of Ω_x(y_l) versus Ω.
  2. [§2.2] The funnel construction, the layer thickness δΩ, and the 'bijectively related' rearrangement are not mathematically defined; the reader cannot verify that the phase-space layers have the required continuity or closure properties.
  3. [§5.2] The first additional assumption, 'f has no explicit time dependence,' is in tension with Eq. (1), which contains ∂f/∂t. The paper should clarify the meaning of 'explicit' time dependence.
  4. [§2.7] The derivation for an isolated system uses the isothermal constraint (9) alongside the microcanonical completeness constraint (8); the distinction between microcanonical and canonical ensembles is not explained, and the microcanonical case has ρ = 1/Ω_tot, for which f = Ω/Ω_tot already gives ∂f/∂Ω > 0 trivially.
  5. [Figures 5 and 6] The numerical and experimental evidence for the central boundary claims is taken from refs. [11–14]; no data, simulation parameters, or code are provided in this manuscript, so those results are not independently verifiable from the present text.

Circularity Check

3 steps flagged

Central claims reduce to their inputs: 'full chaoticity' is defined as ℬ (so the second-law derivation restates the postulate), and the claimed second-law boundary (T_k = 2T/3 Knudsen state) is built into a non-thermal wall emission rule inherited from the author's own prior work [12]; an equilibrium diffuse wall gives T_k = T.

specific steps
  1. self definitional [§1 (definition of full chaoticity); §2.5 (Eq. 2, ℬ); §2.7]
    "microscopic details are effectively lost at the macroscopic level, such that the microstate probability (ρ) does not explicitly depend on the EOM but is constrained only by normalization and energy conservation. An example of full chaoticity is ρ ∝ e−βε ... Under this condition, ℬ (Equations 2 and 3) is justified, which ensures ∂f/∂Ω > 0 (Inequality 7). With ℬ and ∂f/∂Ω > 0, the second law of thermodynamics (Inequality 6) can be readily derived."

    ℬ (Eq. 2) is defined as 'the microstate probability ρ depends only on the energy level ε'; full chaoticity is defined (§1) as the condition that ρ is 'constrained only by normalization and energy conservation' — the same condition. Hence §2.7's inference 'full chaoticity ⇒ ℬ' holds by construction, and for fully chaotic systems the second law is the postulate ℬ (or the special case ∂f/∂Ω > 0) restated. The paper concedes the arrow is an input: §2.6 'Without additional assumptions, Equation (1) offers no particularly useful information about the evolution of f or Ω' and §4 intro that Inequality (7) is the 'driving force' rendering Eq. (1) directional.

  2. ansatz smuggled in via citation [§4.1, Eq. (10), Eq. (13); Appendix 1.4]
    "When a gas particle collides with a thermal wall, the reflected direction is random; the reflected speed randomly follows the Maxwell-Boltzmann distribution at T, uncorrelated with the incident velocity ... the probability density of microstate ρ ∝ e−βε/κp is non-Boltzmannian ... As long as the particle-particle collisions are sparse, no reasonable boundary condition can keep the Knudsen-gas cells in equilibrium."

    The 1/|p| weighting of Eq. (10) and the resulting T_k = 2T/3 of Eq. (13) (a = 1/2) follow from sampling reflected speeds from the unbiased Maxwell-Boltzmann speed distribution (∝ v²e^{−βE}) and weighting interior occupation by residence time ∝ 1/v. A thermal wall in equilibrium emits particles with flux ∝ v·f_MB(v); that 1/v residence bias then cancels exactly, and the interior velocity distribution is f_MB at T with T_k = T (detailed balance preserved). Thus the predicted non-Boltzmann steady state is not derived from the thermal reservoir — it is manufactured by the adopted emission rule, which is not an equilibrium wall rule. The asserted generality of Appendix 1.4 ('no reasonable boundary condition...') fails for the standard diffuse wall.

  3. self citation load bearing [Appendix 1 intro; §4.3; Appendix 3; refs [11–14]; Eq. (15)]
    "the analyses and experiments in [12] suggest that repeatedly converting a large ideal-gas body into a cluster of small Knudsen-gas cells ... generates a cyclic heat flux ... incompatible with the conventional heat-engine statement of the second law ... For locally nonchaotic systems, recent research [11-14] has revealed counterexamples that do not satisfy Boltzmann's assumption."

    The 'counterexamples' carrying the paper's boundary claim (§4.3, Appendices 1.1–1.2) are entirely the author's own prior work: refs [11–14] are all by Qiao/Qiao & Shang. The cited model (Eq. 15, ρ ∝ (∏τ_j)e^{−βε} with τ_j ∝ 1/v_j) is the same 1/v ansatz whose input is the non-thermal wall rule flagged above, so the MC simulation and nanopore interpretation cited from [12] do not independently test the rule. The load-bearing evidence for spontaneous entropy decrease, cold-to-hot transfer, and work from a single reservoir is therefore a self-citation chain whose premise is the disputed emission rule.

full rationale

Two central claims are examined. (i) Positive claim: for fully chaotic systems the second law 'follows' from the continuity equation (Eq. 1) under ℬ/∂f/∂Ω > 0. This reduces to a definition: full chaoticity is defined in §1 as ρ being constrained only by normalization and energy conservation, which is precisely ℬ (Eq. 2); §2.7's 'justification' of ℬ is therefore by construction, and the paper concedes that Eq. (1) alone 'offers no particularly useful information' and that Inequality (7) is the 'driving force' (§4 intro). The arrow is an input, not a consequence of chaotic dynamics. (ii) Negative claim: locally nonchaotic Knudsen gases allow ∂f/∂Ω ≤ 0 and spontaneous entropy decrease. This rests on Eq. (10), ρ ∝ e^{−βε}/κ_p, whose 1/v weighting is generated by the adopted wall rule 'reflected speed randomly follows the Maxwell-Boltzmann distribution at T' combined with residence time ∝ 1/v in the interior. An equilibrium thermal wall emits with flux ∝ v·f_MB(v); the residence-time bias then cancels and the interior is Maxwellian at T (T_k = T). Hence the predicted T_k = 2T/3 state and the claimed breakdown of the second law are artifacts of a non-thermal boundary rule, contrary to Appendix 1.4's assertion that 'no reasonable boundary condition can keep the Knudsen-gas cells in equilibrium.' The counterexample evidence is self-cited ([11–14]), with the same 1/v ansatz already present in the cited model (Eq. 15). Independent content does exist: the continuity-equation formalism, the slice-by-slice phase-space argument (if valid), and Appendix 2's moving-frame example satisfying ∂f/∂Ω > 0 without ℬ. But the positive and negative central claims each reduce, partially, to their inputs — definition in the first case, self-cited ansatz in the second — so a score of 6 is appropriate rather than higher. No ad hominem is intended; the issue is structural reduction, not authorial intent.

Axiom & Free-Parameter Ledger

2 free parameters · 8 axioms · 2 invented entities

The derivation of Ṡ ≥ 0 imports, without proof, smoothness of the Ω-flow and closure of the constructed subspace; the statistical content is concentrated in ℬ/∂f/∂Ω > 0, whose 'derivation' from full chaoticity is definitional. The boundary claims import the Knudsen steady-state model from the author's self-cited [12] and assume equilibrium entropy formulas for off-equilibrium states; the wall-reflection rule behind Eq. (10) violates detailed balance, which would remove the effect entirely.

free parameters (2)
  • a (wall-distribution exponent in n_w(K) = ξ_a K^a e^{-β_a K}) = scan over 0 ≤ a ≤ 1; steady state at a = 1/2
    Parametrizes the macrostate family in §4.2; the claimed ∂f/∂Ω ≤ 0 region is read off this scan. Not fitted to data, but varying a changes the wall condition, so the comparison mixes different physical setups.
  • C_1, C̅_1, e_0 (undetermined constants in ln f, ln Ω) = unspecified; declared independent of a
    Constants in Eqs. (11)–(12) that cancel in the f–Ω comparison; their a-independence is asserted, not proven.
axioms (8)
  • domain assumption ∂f/∂Ω > 0 (Inequality 7): macrostate probability increases with microstate count
    The central postulate from which Ṡ ≥ 0 is derived; the paper acknowledges Eq. (1) alone yields no arrow (§2.6). Content is close to the conclusion.
  • domain assumption Boltzmann's equal a priori probabilities ℬ (Eq. 2): ρ = P_E(ε)
    Standard statistical-mechanics postulate, stated as sufficient condition for ∂f/∂Ω > 0 (§2.5).
  • ad hoc to paper 'Full chaoticity' implies ρ independent of EOM (Section 1 definition)
    The definition of full chaoticity includes the conclusion it is used to justify (ℬ) in §2.7 — a definitional slide.
  • standard math Micro-reversibility for superjacent/subjacent accessible microstates
    Used in §2.5 to argue exclusions are conservative; standard Hamiltonian time-reversal.
  • ad hoc to paper Smoothness: ∂Ω̇/∂Ω well-defined and finite, |∂²Ω̇/∂Ω²| finite, layers arbitrarily close
    Required by the slice-by-slice induction (§2.5); Ω is defined on macrostates, so Ω̇ as a smooth field is not guaranteed.
  • ad hoc to paper Closure of phase-space subspace A1A2B1B2
    Asserted in §2.3 without proof; the derivation of Ω̇ ≥ 0 for macrostate Ψ is conducted inside this subspace.
  • domain assumption Knudsen steady state: ρ ∝ e^{-βε}/κ_p (Eq. 10)
    Carried over from self-cited [12]; yields T_k = 2T/3. Conflicts with standard flux balance for diffuse walls (see red flag).
  • domain assumption Gas entropy during the non-equilibrium transition is the equilibrium function of T_k (Eq. 17)
    Appendix 1.1 uses dS_G = dU/T_k to compute the claimed net entropy decrease; requires the off-equilibrium gas to satisfy equilibrium S(U).
invented entities (2)
  • Intrinsic-nonequilibrium steady state independent evidence
    purpose: A steady, unforced out-of-equilibrium state of locally nonchaotic systems; the basis for the claimed second-law boundary (entropy decrease, cold-to-hot transfer).
    Falsifiable handles exist (gas kinetic temperature, heat flux direction; the nanopore defiltration temperature drop), but all current evidence is from the author's own prior papers [11–14]; no independent replication.
  • Kinetic factor κ_p = ∏|p_j| in Eq. (10) no independent evidence
    purpose: Produces the non-Boltzmann microstate weighting ρ ∝ 1/κ_p that generates T_k < T.
    A modeling device imported from [12]; its physical basis (unbiased wall reflection) is the load-bearing questionable assumption.

pith-pipeline@v1.3.0-alltime-deepseek · 21057 in / 46660 out tokens · 469224 ms · 2026-08-03T00:31:51.152635+00:00 · methodology

0 comments
read the original abstract

The second law of thermodynamics still lacks a general proof applicable to both classical and quantum systems across broad ranges of time scales, interactions, and degrees of nonequilibrium. In this paper, we show that when the macrostate-level probability $f$ of an isolated system increases monotonically with the number of possible microstates $\Omega$ (i.e., $\partial f/\partial \Omega > 0$), the second law emerges naturally from the continuity equation of $f$ in phase space; a special case of $\partial f/\partial \Omega > 0$ is Boltzmann's assumption of equal a priori equilibrium probabilities. Based on this finding, the second law can be readily derived for fully chaotic systems. The derivation does not rely on dynamical details, highlighting the statistical nature of entropy increase. In contrast, for a locally nonchaotic system, the positive correlation between $f$ and $\Omega$ may break down (i.e., $\partial f/\partial \Omega \leq 0$). Consequently, the conventional framework of thermodynamics does not apply, and entropy can decrease spontaneously without any energetic penalty.

Figures

Figures reproduced from arXiv: 2607.28765 by Yu Qiao.

Figure 1
Figure 1. Figure 1: (a) Schematic of the funnel-shaped phase space of an isolated system. Each horizontal “layer” (hyperplane) represents the region corresponding to a certain value of , i.e., the subset of the microstates that satisfy x (𝑦𝑙 ) = . The layer size represents the number of microstates in the subset. From bottom to top,  increases. The grey circle in hyperplane Ly indicates a macrostate (). The grey dotted c… view at source ↗
Figure 3
Figure 3. Figure 3: (a) Schematic of a Knudsen gas [12] (see Appendix 1 for more details). The particle mean free path 𝜆F is much larger than the characteristic size of the gas container 𝐷. The system is immersed in a thermal reservoir, using thermal-wall boundary condition. Because particle-particle interactions are sparse, fast particles tend to collide rapidly with the container walls and release heat to the environment, a… view at source ↗
Figure 4
Figure 4. Figure 4: Schematic of a microstate evolution path (the blue curve) in the funnel-shaped phase space. The band of Ly indicates a constant-x hyperplane with a finite “layer” thickness (𝛿) [PITH_FULL_IMAGE:figures/full_fig_p019_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: A Knudsen-gas model [12]. (a) An ideal gas (the undivided setup) is cyclically converted to a Knudsen-gas cell cluster (the cellular setup), by repeatedly inserting and withdrawing a set of frictionless inner cell walls. The container size (𝐿c ) is much larger than the particle mean free path (𝜆F); the cell size (𝐷) is smaller than 𝜆F. In a Knudsen-gas cell, as 𝐷 < 𝜆F, particle-particle collisions are spar… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

28 extracted references

  1. [1]

    The Noether Theorems

    Kosmann-Schwarzbach Y , Schwarzbach BE (2010). The Noether Theorems. Springer

    2010

  2. [2]

    Statistical Physics of Particles

    Kardar M (2007). Statistical Physics of Particles. Cambridge Univ. Press

    2007

  3. [3]

    Cercignani, R

    C. Cercignani, R. Illner, M. Pulvirenti. The Mathematical Theory of Dilute Gases (Springer Sci., 1994)

    1994

  4. [4]

    Ampatzoglou, J

    I. Ampatzoglou, J. K. Miller, N. Pavlović. A rigorous derivation of a Boltzmann system for a mixture of hard-sphere gases. SIAM J. Math. Anal 54, 2320-2372 (2022)

    2022

  5. [5]

    L., Tumulka, R., Zanghì, N

    Goldstein, S., Lebowitz, J. L., Tumulka, R., Zanghì, N. Gibbs and Boltzmann Entropy in Classical and Quantum Mechanics. In: V . Allori (ed), Statistical Mechanics and Scientific Explanation (World Sci. Publ., 2020)

    2020

  6. [6]

    Lebowitz, J. L. Boltzmann’s entropy and time’s arrow, Phys. Today 46, 32-38 (1993)

    1993

  7. [7]

    M. L. Bellac, F. Mortessagne, G. G. Batrouni. Equilibrium and Non-Equilibrium Statistical Thermodynamics (Cambridge University Press, 2009)

    2009

  8. [8]

    Hypocoercive relaxation to equilibrium for some kinetic models

    Pierre Monmarché. Hypocoercive relaxation to equilibrium for some kinetic models. Kinetic and Related Models 7, 341-360 (2014)

    2014

  9. [9]

    E. Dolera. Exponential convergence to equilibrium for solutions of the homogeneous Boltzmann equation for Maxwellian molecules. Mathematics 10, 2347 (2022)

    2022

  10. [10]

    Boltzmann’s Approach to Statistical Mechanics

    Goldstein, S. Boltzmann’s Approach to Statistical Mechanics. In: Bricmont, J., Ghirardi, G., Dürr, D., Petruccione, F., Galavotti, M.C., Zanghi, N. (eds.), Chance in Physics: Foundations and Perspectives (Springer, Berlin, 2001)

    2001

  11. [11]

    Molecular-sized outward-swinging gate: experiment and theoretical analysis of a locally nonchaotic barrier

    Qiao Y , Shang Z, Kou R (2021). Molecular-sized outward-swinging gate: experiment and theoretical analysis of a locally nonchaotic barrier. Phys. Rev. E 104, 064133. https://escholarship.org/uc/item/5hp0w1vv

    2021

  12. [12]

    Shang (2024)

    Qiao Y , Z. Shang (2024). Second law of thermodynamics: spontaneous cold-to-hot heat transfer in a nonchaotic medium. Phys. Rev. E 110, 054113. https://escholarship.org/uc/item/1g6074r3

    2024

  13. [13]

    On the second law of thermodynamics: A global flow spontaneously induced by a locally nonchaotic energy barrier

    Qiao Y , Shang Z (2024). On the second law of thermodynamics: A global flow spontaneously induced by a locally nonchaotic energy barrier. Physica A 647, 129828. https://escholarship.org/uc/item/5632b1ts 33

    2024

  14. [14]

    Searching for quantum non-thermodynamic phenomena, Phys

    Qiao Y (2025). Searching for quantum non-thermodynamic phenomena, Phys. Rev. E 111, https://escholarship.org/uc/item/55c900dz

    2025

  15. [15]

    Van Kampen

    N.G. Van Kampen. Stochastic Processes in Physics and Chemistry (Elsevier, 2007)

    2007

  16. [16]

    Lagrangian and Hamiltonian Mechanics

    Calkin MG (1996). Lagrangian and Hamiltonian Mechanics. World Scientific Publ

    1996

  17. [17]

    R. K. Pathria, P. D. Beale. Statistical Mechanics (Butterworth-Heinemann, 2011)

    2011

  18. [18]

    J. R. Dorfman. An Introduction to Chaos in Nonequilibrium Statistical Mechanics (Cambridge University Press, Cambridge, 1999)

    1999

  19. [19]

    D. P. Feldman (2019). Chaos and Dynamical Systems (Princeton Univ. Press)

    2019

  20. [20]

    A Treatise on Probability

    Keynes JM (1921). A Treatise on Probability. MacMillan, London, UK

    1921

  21. [21]

    Non-Equilibrium Thermodynamics

    de Groot SR, Mazur P (2013). Non-Equilibrium Thermodynamics. Dover Publ

    2013

  22. [22]

    Jarzynski (1997)

    C. Jarzynski (1997). Nonequilibrium equality for free energy differences. Phys. Rev. Lett 78, 2690

    1997

  23. [23]

    D. Wallace. The Emergent Multiverse: Quantum Theory according to the Everett Interpretation (Oxford Univ. Press, 2012)

    2012

  24. [24]

    Cabello, M

    A. Cabello, M. Gu, O. Gühne, J. -Å. Larsson, K. Wiesner. Thermodynamical cost of some interpretations of quantum theory. Phys. Rev. A 94, 052127 (2016)

    2016

  25. [25]

    Deffner, J

    S. Deffner, J. P. Paz. Quantum work and the thermodynamic cost of quantum measurements Phys. Rev. E 94, 010103 (2016)

    2016

  26. [26]

    C. L. Latune, C. Elouard. A thermodynamically consistent approach to the energy costs of quantum measurements. Quantum 9, 1614 (2025)

    2025

  27. [27]

    Theory of Heat (Longmans, Green, and Co., London, 1872, p

    Maxwell JC (1872). Theory of Heat (Longmans, Green, and Co., London, 1872, p. 300)

  28. [28]

    Tolman temperature gradients in a gravitational field

    Santiago J, Visser M (2019). Tolman temperature gradients in a gravitational field. Eur. J. Phys. 40, 025604

    2019