Pith. sign in

REVIEW 4 major objections 6 minor 21 references

The Second Law as a constraint and admitting the approximate nature of constitutive assumptions

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper proposes treating the Second Law of thermodynamics as an equality constraint on processes, introducing excess fields to absorb approximate constitutive assumptions, and converting the resulting constrained problem into an…

desk verdict New variational strategy for treating the Second Law as a constraint, with sound formal algebra, but the central claim about the concave maximizer producing primal solutions is not established due to domain and existence gaps. read the letter →

arxiv 2412.19914 v1 pith:XEFHHCAA submitted 2024-12-27 physics.class-ph

Amit Acharya This is my paper · ORCID
classification physics.class-ph MSC 74A1574B2049S05 PACS 05.70.-a
keywords SecondLawofthermodynamicsClausius-Duheminequalityconstitutiveassumptionsexcessfieldsdualvariationalprincipleconcavemaximizationelastodynamicsentropy
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

The paper argues that the Second Law of thermodynamics need not be used only to restrict constitutive equations, as in the classical Coleman-Noll procedure; it can instead be imposed as an equality constraint on every admissible process, on the same footing as the balance of mass, momentum, and energy. To accommodate the fact that constitutive equations are never known exactly, the stress and free-energy density are augmented by 'excess' fields that are required to be as small as possible. The resulting constrained optimization problem is rewritten as an unconstrained, concave maximization problem in dual variables, and a dual-to-primal (DtP) mapping converts any maximizer back into a solution of the original mechanical PDE system with the Second Law satisfied exactly. If the construction works as claimed, modelers could use approximate material data and still guarantee non-negative entropy production, and the scheme would not fabricate solutions when the primal problem has none.

What carries the argument

The load-bearing object is the dual-to-primal (DtP) mapping $U=U_H(D,\bar U)$, defined by $\partial_U \mathcal L(U_H(D,\bar U), D, \bar U)=0$ for a Lagrangian $\mathcal L$ whose primal-dependent part is a sum of the constraint equations (5) and a quadratic auxiliary potential $H$ that penalizes deviations from base states $\bar U=(\bar v,\bar e,0,\bar d,0,0)$. The associated dual functional $\tilde S[D]=\inf_U \hat S[U,D]$ is concave in $D$ because $\hat S$ is affine in $D$, and the domain conditions (19) enforce positive-semidefiniteness of the Hessian so that the infimum is attained and the DtP mapping is single-valued. The argument then rests on the identity that the Euler-Lagrange equations of $S$ (hence the critical-point condition of $\tilde S$) are precisely the primal system (5) with $U\to U_H$.

What would settle it

Take a set of boundary and initial data for the elastodynamic bar for which the primal system (5)-(6) is known to have no solution; if solving the concave maximization (20) nevertheless yields a convergent maximizing sequence whose DtP image satisfies the PDEs to numerical tolerance, the no-spurious-solutions claim fails. Alternatively, for the softening stress (8), search for a sequence of admissible dual fields satisfying (19) along which $-\tilde S$ stays bounded while the fields grow without bound; such a sequence would disprove the coercivity on which existence of a maximizer rests.

Watch

Extended reading notes

Core claim

For a one-dimensional elastodynamic bar with a constitutively determined stress response (8) that exhibits softening, the author constructs a dual functional $\tilde S[D]$ by taking the infimum over primal fields of a Lagrangian built from the primal PDE system (5) and a quadratic penalty $H$ on deviations from prescribed base states. Maximizing this concave dual functional over dual fields $D=(\lambda,\mu,\beta,\rho,\gamma)$ produces, through the DtP mapping $U=U_H(D,\bar U)$ defined by setting the Lagrangian's primal derivative to zero, a solution of the primal system (5)-(6) in which the Second Law holds as the equality $T e_t - \psi_t - s^2 = 0$ and the excess fields $X, s, g$ are pointwise as small as possible. The Euler-Lagrange equations of the dual functional are exactly the primal equations with $U$ replaced by $U_H$, and since $\hat S$ is affine in $D$, any interior maximizer of $\tilde S$ is a critical point of $S$ and hence yields a primal solution; conversely, if the primal system has no solution, no dual extremal can exist, so the scheme is claimed not to produce spurious solutions.

Load-bearing premise

The construction needs the optimization problem to have a solution: the search space must be closed and the objective must grow without bound as the variables grow, which the paper assumes but does not prove; it also assumes the reference motions chosen are close to a real solution.

Editorial extensions

If this is right

  • The Second Law can be enforced as a pointwise equality with a slack variable $s^2$ for physical dissipation, leaving constitutive functions $\hat T, F$ unrestricted rather than forcing the Coleman-Noll identity $\hat T = F'$.
  • Excess stress and energy corrections become part of the solved fields and are generally nonlocal, in the same way the pressure field is nonlocal for an incompressible material.
  • For each solution of the primal problem, the dual scheme has at least one critical point, namely $D=0$ with base states set to that solution, which serves as an exact consistency check of the formulation.
  • Weak discontinuities in dual fields are admitted and map through the DtP relation to primal fields with discontinuities aligned with characteristics, so shock-type solutions can be represented without added higher-order regularization.
  • Because the dual problem is concave, it has a single maximum, and different primal solutions can be selected by choosing base states $(\bar v,\bar e,\bar d)$, a role the paper likens to a selection criterion for non-unique elastodynamics.

Reading between the lines

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

  • A direct numerical test of the central claim would run the concave maximization (20) on the Needleman viscoplastic example with time-dependent $A(t)$ of both signs; success would show the Second Law can be enforced locally without sacrificing stability, a question Needleman's papers leave open.
  • The unproved coercivity of $-\tilde S$ on the admissible set (19) is the chief mathematical risk; if coercivity fails for the softening law (8), adding a small strictly concave regularization of the dual functional could restore existence but would alter the no-spurious-solutions guarantee.
  • The paper works out the construction in one dimension with six primal and five dual fields; the same ideas in three dimensions would involve a tensor-valued excess stress and a scalar inequality, likely changing the balance between degrees of freedom and constraints.
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

4 major / 6 minor

Summary. The manuscript proposes a variational treatment of the Second Law of thermodynamics in continuum thermomechanics, treating it as a constraint on processes rather than as a restriction on constitutive assumptions. The author augments an approximate constitutive relation for stress and free energy with excess fields X and a, introduces a dissipation slack s and its spatial gradient g, and writes the primal system as (5)-(6). For a 1D elastodynamic bar with the softening stress-strain law (8), the paper constructs a pre-dual functional (7), derives a dual-to-primal mapping (13), defines the dual functional S in (18) and its concave inf-version \tilde S in (19)-(20), and claims that solving the concave maximization problem argmax_D \tilde S[D] produces, through the DtP mapping, a solution of the primal system (5)-(6). The paper also discusses cases where the Second Law is over-constraining or under-constraining, including a viscoplastic example related to Needleman's work and shock solutions with concentrated dissipation.

Significance. If the central claim is made rigorous, the paper would offer a genuinely new computational and conceptual route: a concave variational principle for a non-convex, non-monotone elastodynamic system, with the Second Law enforced as an equality and excess constitutive fields determined by the optimization rather than by additional constitutive postulates. The explicit construction of the DtP mapping, the formal Euler-Lagrange consistency with the primal system, and the consistency check in the Remark after Eq. (18) are valuable and clearly presented. The idea of using prescribed base states as a solution-selection device is original and potentially useful. However, the main claim rests on two unproved premises: existence of a maximizer of \tilde S, and the transfer of the critical-point equivalence to constrained maximizers of the concave program. These gaps are load-bearing rather than cosmetic. The paper also contains no numerical demonstration itself; computational evidence is cited from the author's prior works [9]-[12], which are not reproduced here.

major comments (4)
  1. [Sec. 3, Eq. (19)-(20)] The step from the concave functional \tilde S to the claim that argmax_D \tilde S[D] produces a primal solution is not justified as stated. The text after (19) asserts that a maximizer is also a critical point 'at least formally,' but \tilde L is finite only on the constrained domain (19). The natural consistency point D=0 does not lie in that domain: since c_s<0, the condition c_s-2\rho\ge 0 fails at \rho=0, so \tilde S[0] is -\infty by the 'otherwise' branch of (19). Therefore the Remark after (18), which shows that D=0 is a critical point of S when \bar U solves (5)-(6), does not transfer to the concave program. A boundary maximizer must be characterized by KKT inequality multipliers, and its first-order conditions need not be the primal equations (5)-(6). The central claim needs either a proof that every (or an appropriately selected) maximizer lies in the interior of (19), or a KKT analysis showing that boundary maximizers still map to primal solutions, or a modification of H so that the consistency point lies inside the admissible domain.
  2. [Sec. 3, after Eq. (19)] Existence of a maximizer of \tilde S is asserted to depend only on coercivity of -\tilde S, but no coercivity proof is supplied. The dual functional contains terms with denominators c_s-2\rho and c_e-2E_*\mu_x, whose signs are controlled by the constraints in (19); the boundary of that constraint set can cut off maximizing sequences or fail to be closed under weak limits. No compactness argument is given and no hypotheses on the data and parameters are stated that would guarantee attainment. Because (20) is defined as an argmax, this gap is load-bearing: without existence, the central claim has no definite object to which it refers. The author should either prove existence for the model problem of Sec. 3 under explicit conditions, or state precisely what additional assumptions on c_X,c_s,c_g,c_d,c_e,c_v and the base states are needed.
  3. [Sec. 4, Observation 1] The no-spurious-solutions guarantee is stated for critical points of the dual functional, but the proposed computational scheme is the constrained maximization (20). Since maximizers of \tilde S may lie on the boundary of the domain (19), and since the first-order conditions of a constrained maximum differ from the unconstrained Euler-Lagrange equations, Observation 1 does not cover the actual problem being solved. The claim should be restated and proved for maximizers of (20), or explicitly limited to unconstrained critical points of S.
  4. [Sec. 4, Observation 6 and text after Eq. (20)] The claim that by choosing base states (\bar v,\bar e,\bar d) 'one can home in on different primal solutions in a reliable manner' is presented as an expectation, not as a theorem. Since existence, uniqueness, and stability of the maximizer are not established, it is unclear in what sense the solution depends continuously on the base states or whether a base state close to a desired solution is actually selected. This is a second load-bearing premise for the practical scope of the method. The paper should either prove a selection property under explicit hypotheses, or clearly label this as a conjecture supported only by the cited computational examples.
minor comments (6)
  1. [Abstract] There is a typo in the abstract: 'cons traint' should be 'constraint'.
  2. [Sec. 3, Eq. (7)] The notation for the Lagrangian is inconsistent: the first line writes L(U,D), while later lines use L(U,D,\bar U); the functional \hat S also includes boundary terms that are not shown in the displayed definition of L.
  3. [Sec. 3, Eq. (18)] The notation K|^{-1}_\rho J in the displayed dual Lagrangian is undefined; based on the preceding algebra it appears to denote the Schur complement of K with respect to \rho, but this should be stated explicitly.
  4. [Sec. 3, after Eq. (5)] The sentence 'we will assume T^\sharp(e)=0; a=0' is an assumption that removes the excess free-energy rate a from the analysis, but this is not flagged as a restriction that limits the scope of the subsequent variational principle; the reader should be told explicitly that the method also covers the case a\neq 0.
  5. [Sec. 3, end of Section] The phrase 'a concave maximization problem which has a single maxima' is imprecise: concavity alone gives a convex set of maximizers, not necessarily a single point. The sentence should be revised to say that the problem is concave and hence has no spurious local maxima.
  6. [Sec. 4, Observation 5] In the alternative H appearing after Eq. (22), the term |s|^{1-p} is not differentiable at s=0 when p=0; if first-order optimality conditions are used for the dual scheme, the non-differentiability should be addressed or p should be restricted to (0,1).

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the variational equivalence is by construction, self-citations are non-load-bearing, and the main gaps are existence/coercivity rather than circular reductions.

full rationale

The paper's central construction is a dual variational principle whose Euler-Lagrange equations are, by construction, the primal system (5)-(6) with U replaced by the DtP mapping U_H. This is an explicit equivalence derived in the text, not a fitted prediction: no parameters are calibrated to data, and the D=0 consistency point is explicitly labeled a consistency check rather than a predictive output. The self-citations ([9]-[12]) supply computational and theoretical evidence for the dual scheme in other problems, but the derivation of the dual functional and its consistency with the primal system is carried out directly in this paper, so the self-citations are not load-bearing for the central claim. The more serious limitations are mathematical rather than circular: coercivity of -\tilde S is assumed but never proved, and the domain in (19) may exclude the natural consistency point D=0 because c_s<0, so boundary maximizers of \tilde S need not satisfy the unconstrained primal first-order conditions. These are existence and correctness gaps, not reductions of the claim to its inputs. Accordingly, no specific circular step can be exhibited under the required standard.

Assumptions & free parameters 2 free parameters · 4 assumptions · 4 invented entities

The construction introduces excess fields and a dissipation slack to absorb constitutive error, and several arbitrary constants and base states that shape the solution. None of these have independent physical evidence. The main unproved mathematical inputs are coercivity and the selection power of base states.

free parameters (2)
  • Penalty constants c_X, c_s, c_g, c_d, c_e, c_v = not specified; c_X, c_g, c_d, c_e, c_v > 0 and c_s < 0 O(1)
    Chosen arbitrarily in the auxiliary potential H to define the objective; the solution depends on these weights, and no canonical values or data are given.
  • Base-state fields \bar v, \bar e, \bar d = arbitrary functions of space-time
    Inputs that bias the dual maximizer toward a selected solution; the method's usefulness depends on choosing them well, but no construction rule is given.
assumptions (4)
  • standard math Standard calculus of variations and Lagrange multiplier duality; first-order optimality conditions are sufficient for the concave maximization problem.
    Used throughout Sec. 3 to pass from the pre-dual functional to the dual functional and to claim extremals solve the primal system.
  • ad hoc to paper The dual functional -\tilde S is coercive on a closed convex set of dual fields, so a maximizer exists.
    Invoked after Eq. (20); the paper does not prove coercivity and acknowledges it is the remaining condition.
  • domain assumption The positivity and semi-definiteness restrictions in (19) hold on the relevant domain.
    Needed for \tilde L to be finite and for the DtP mapping U_H to be well-defined; no guarantee for general solutions.
  • ad hoc to paper Base states \bar U can be chosen close enough to a desired solution so that the concave maximizer selects that solution.
    Sec. 4 observation 6 says 'may be expected' that the dual scheme recovers a nearby shock solution; this is not proven.
invented entities (4)
  • Excess stress field X
    purpose: Compensates for error in the assumed stress constitutive equation so balance laws hold exactly.
    Introduced in Eq. (2); determined by the model, with no direct physical measurement.
  • Excess free energy rate field a
    purpose: Compensates for error in the assumed free-energy constitutive law in the Second Law expression.
    Introduced in Eq. (2) and used in the dissipation inequality; no independent physical handle.
  • Dissipation slack s with s^2 physical dissipation
    purpose: Converts the Second Law inequality into an equality and is maximized to select dissipation.
    Introduced in Sec. 2; a mathematical device, not a measured quantity.
  • Spatial gradient of dissipation g = s_x
    purpose: Optionally regularizes spatial gradients of the dissipation; when active, prevents concentration on curves.
    Introduced in system (5) and discussed in Sec. 4 observation 6.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Second Law as a constraint and admitting the approximate nature of constitutive assumptions." pith.science (2026). https://pith.science/paper/XEFHHCAA

@misc{pith2026241219914,
  author       = {Pith},
  title        = {Pith review of: The Second Law as a constraint and admitting the approximate nature of constitutive assumptions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XEFHHCAA}},
  note         = {Machine review of arXiv:2412.19914}
}
read the original abstract

A scheme for treating the Second Law of thermodynamics as a constraint and accounting for the approximate nature of constitutive assumptions in continuum thermomechanics is discussed. An unconstrained, concave, variational principle is designed for solving the resulting mathematical problem. Cases when the Second Law becomes an over-constraint on the mechanical model, as well as when it serves as a necessary constraint, are discussed.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 19 canonical work pages

  1. [9]

    A hi dden convexity of nonlinear elas- ticity

    Siddharth Singh, Janusz Ginster, and Amit Acharya. “A hi dden convexity of nonlinear elas- ticity”. In: Journal of Elasticity 156.3 (2024), pp. 975–1014

    work page 2024

  2. [12]

    Inviscid Burgers as a degenerate elliptic problem

    Uditnarayan Kouskiya and Amit Acharya. “Inviscid Burg ers as a degenerate elliptic prob- lem”. In: published online, Quarterly of Applied Mathematics (https://arxiv.org/abs/ 2401.08814, 2024)

    work page Pith review arXiv 2024

  3. [1]

    The thermodynamics of elastic materials with heat conduction and viscosity

    Bernard D. Coleman and Walter Noll. “The thermodynamics of elastic materials with heat conduction and viscosity”. In: Archive for Rational Mechanics and Analysis 13 (1963), pp. 167– 178

    work page 1963

  4. [2]

    The Classical Fie ld Theories

    Clifford Truesdell and Richard Toupin. “The Classical Fie ld Theories”. In: Principles of clas- sical mechanics and field theory/Prinzipien der Klassische n Mechanik und Feldtheorie; En- cyclopaedia of Physics . Ed. by Flugge S. Vol. III/1. Springer, 1960, pp. 226–858

    work page 1960

  5. [3]

    Berdichevsky

    Victor L. Berdichevsky. Thermodynamics of chaos and order; In Pitman Monographs and Surveys in Pure and Applied Mathematics . Vol. 90. Longman, Harlow, 1997

    work page 1997

  6. [4]

    Microcanonical entropy and mesoscale di slocation mechanics and plasticity

    Amit Acharya. “Microcanonical entropy and mesoscale di slocation mechanics and plasticity”. In: Journal of Elasticity 104 (2011), pp. 23–44

    work page 2011

  7. [5]

    A Perspective on Plasticity, Dissipat ion and the Second Law of Thermo- dynamics

    Alan Needleman. “A Perspective on Plasticity, Dissipat ion and the Second Law of Thermo- dynamics”. In: Journal of Applied Mechanics 91.6 (2024)

    work page 2024

  8. [6]

    Discrete defect plasticity and implic ations for dissipation

    Alan Needleman. “Discrete defect plasticity and implic ations for dissipation”. In: European Journal of Mechanics-A/Solids 100 (2023), p. 105002

    work page 2023

Show all 21 references
  1. [7]

    Plasticity without phenomenology: a first step

    Sabyasachi Chatterjee, Giacomo Po, Xiaohan Zhang, Amit Acharya, and Nasr Ghoniem. “Plasticity without phenomenology: a first step”. In: Journal of the Mechanics and Physics of Solids 143 (2020), p. 104059

    2020

  2. [8]

    A dual variational principle for nonline ar dislocation dynamics

    Amit Acharya. “A dual variational principle for nonline ar dislocation dynamics”. In: Journal of Elasticity 154.1 (2023), pp. 383–395. 13

    2023

  3. [10]

    A hidden convexity in continuum mechani cs, with application to classical, continuous-time, rate-(in) dependent plasticity

    Amit Acharya. “A hidden convexity in continuum mechani cs, with application to classical, continuous-time, rate-(in) dependent plasticity”. In: Mathematics and Mechanics of Solids (2024). doi: https://doi.org/10.1177/10812865241258154

    2024 doi

  4. [11]

    Hidden convex ity in the heat, linear transport, and Euler’s rigid body equations: A computational approach

    Uditnarayan Kouskiya and Amit Acharya. “Hidden convex ity in the heat, linear transport, and Euler’s rigid body equations: A computational approach ”. In: Quarterly of Applied Math- ematics 82 (2024), pp. 673–703

    2024

  5. [13]

    Thermodynami cs with internal state variables

    Bernard D. Coleman and Morton E. Gurtin. “Thermodynami cs with internal state variables”. In: The journal of chemical physics 47.2 (1967), pp. 597–613

    1967

  6. [14]

    The entropy rate admissibil ity criterion for solutions of hyperbolic conservation laws

    Constantine M. Dafermos. “The entropy rate admissibil ity criterion for solutions of hyperbolic conservation laws”. In: Journal of Differential Equations 14.2 (1973), pp. 202–212

    1973

  7. [15]

    Maximal dissipation in equa tions of evolution

    Constantine M. Dafermos. “Maximal dissipation in equa tions of evolution”. In: Journal of Differential Equations 252.1 (2012), pp. 567–587

    2012

  8. [16]

    The Least Action Admissibility Principle

    Heiko Gimperlein, Michael Grinfeld, Robin J Knops, and Marshall Slemrod. “The Least Action Admissibility Principle”. In: arXiv preprint arXiv:2409.07191 (2024)

    2024 arXiv

  9. [17]

    Rohan Abeyaratne and James K. Knowles. Evolution of phase transitions: a continuum the- ory. Cambridge University Press, 2006

    2006

  10. [18]

    Admissibility criteria for propag ating phase boundaries in a van der Waals fluid

    Marshall Slemrod. “Admissibility criteria for propag ating phase boundaries in a van der Waals fluid”. In: Archive for Rational Mechanics and Analysis 81 (1983), pp. 301–315

    1983

  11. [19]

    A limiting “viscosity

    Marshall Slemrod. “A limiting “viscosity” approach to the Riemann problem for materials exhibiting change of phase”. In: Archive for Rational Mechanics and Analysis 105.4 (1989), pp. 327–365

    1989

  12. [20]

    About the “normal growth

    Lev Truskinovsky. “About the “normal growth” approxim ation in the dynamical theory of phase transitions”. In: Continuum mechanics and thermodynamics 6 (1994), pp. 185–208

    1994

  13. [21]

    V ariational Dual Solutions of Chern-Simons Theory

    Amit Acharya, Janusz Ginster, and Ambar N. Sengupta. “V ariational Dual Solutions of Chern-Simons Theory”. In: Arxiv preprint (https://arxiv.org/abs/2411.17635,2024). 14

    2024 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.