{"id":"ce63eb0f-96cb-4516-9040-b55fc619a2ee","arxiv_id":"2501.12532","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A modified pressure-based DG method with correction terms preserves velocity and pressure equilibria and semidiscrete total energy for thermally perfect multicomponent flows.","lead":"This paper builds a discontinuous Galerkin scheme for multicomponent gas flows that keeps pressure and velocity flat across gas interfaces while still conserving total energy in the semidiscrete limit. It targets a known source of spurious pressure oscillations in smooth mixed-gas simulations, and tests the scheme in one, two, and three dimensions.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 17 proves only pure non-uniform or pure constant states; the actual P3 scheme applies the face-based correction at faces between uniform and non-uniform elements, a mixed case the proof does not cover.","rationale":"The reader's weakest_assumption flagged the 'clean split between elementwise-nonuniform and elementwise-constant states' as an assumption, but did not identify the more specific and more damaging issue: at a face shared between a uniform and a non-uniform element, the numerical flux is common to both elements, so the face-based correction cannot be active for only one of them. This makes the case distinction in Theorem 17 non-exhaustive for the actual P3 scheme, which is explicitly intended to handle both element types coexisting in one domain. The numerical experiments in Sec. 4 necessarily contain such mixed faces (the thermal bubble has a non-uniform interior region and a near-constant exterior), so the central claim is not proven for the computed configurations even though the numerical evidence is consistent with the claim. The correction algebra for velocity and pressure preservation is robust in the pure cases, and the energy-conservation arguments for those cases are internally sound; the gap is specifically the missing treatment of mixed-face neighborhoods. Because the paper is otherwise careful and the empirical results are strong, the appropriate outcome is a conditional acceptance: the authors should either supply a proof for the mixed-face case or precisely specify the face-correction switching rule and verify the energy balance for both elements adjacent to such a face. This does not require rejecting the paper, but it does require resolving the ambiguity before the central theorem can be taken as stated.","tokens_in":27658,"tokens_out":19788,"duration_ms":204616,"concrete_test":"Construct a periodic two-element configuration in v0, P0 equilibrium: the left element has a non-uniform smooth concentration/temperature profile (so elementwise correction is active) and the right element is exactly constant (so face correction is active). Run the P3 scheme exactly as specified in Sec. 3.3 and monitor max|P-P0| and total-energy drift. Independently, write out the residual for the non-uniform element at the mixed face with βJbzK included and verify the local conservation statement (3.15) algebraically. If the algorithm cannot be run without specifying β at that face, or the derived balance fails, the mixed-face case in Theorem 17 is unresolved.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Theorem 17 (Sec. 3.3.1) splits the equilibrium analysis into case (a) elementwise-non-uniform and case (b) elementwise-constant, and in each case assumes only one correction mechanism is active: the elementwise correction (3.17) in (a), the face correction (3.19) in (b). The proposed scheme, however, is described as combining the two corrections (Sec. 3.3), and the note after (3.20) explicitly invokes 'an element with a uniform state and an element with a non-uniform state' to justify the modified total-energy flux. At a face shared by a uniform and a non-uniform element, the numerical flux is single-valued, so βJbzK in (3.19) cannot be applied to only one side. If it is applied, the non-uniform element receives a correction not accounted for in case (a); its local energy balance and the velocity/pressure argument in Theorem 17 would need to be re-derived with β present. If it is not applied, the uniform element loses the face correction on which case (b)'s energy conservation relies. The paper gives no precise selection rule for β at such mixed faces and no proof for this configuration. Since the thermal-bubble tests contain exactly these mixed transitions, the central claim that the P3 scheme simultaneously preserves equilibria and conserves energy is not established by the theorem for the configurations computed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper addresses pressure- and velocity-equilibrium preservation in discontinuous Galerkin (DG) discretizations of the compressible, multicomponent Euler equations for smooth, inviscid flows of thermally perfect gas mixtures. The authors first analyze a standard conservative DG formulation and show that it can preserve velocity equilibrium under flux constraints but generally fails to preserve pressure equilibrium. They then introduce a pressure-based DG formulation, replacing the total-energy equation with a pressure-evolution equation, and restore semidiscrete total-energy conservation using the correction terms of Abgrall and co-workers. Because these corrections destroy the equilibrium-preservation property, the authors propose a modified correction term and a face-based correction, and they prove (Theorem 17) that the resulting scheme is velocity-equilibrium-preserving, pressure-equilibrium-preserving, and semidiscretely energy-conservative. Numerical experiments in one, two, and three dimensions, including curved-mesh cases, demonstrate stability, pressure-equilibrium preservation, and energy-conservation behavior for smooth thermal-bubble and Gaussian-wave problems, with an acknowledged robustness limitation for p=1 on coarse meshes.","tokens_in":27911,"tokens_out":13353,"duration_ms":126135,"significance":"If the results hold, the paper makes a valuable contribution to a long-standing problem: constructing fully conservative schemes that exactly preserve pressure equilibrium at fluid interfaces for thermally perfect multicomponent gases, a regime where prior conservative schemes either produced spurious oscillations or required calorically perfect gas assumptions. The paper's strengths include clean algebraic proofs of the equilibrium conditions and of the modified correction-term identities, transparent discussion of the tradeoffs involved (semidiscrete versus fully discrete conservation, exact-integration caveats, and p=1 fragility), and a substantial set of numerical experiments that demonstrate the practical benefit of the proposed P3 scheme. The independent verification on 1D/2D/3D thermal-bubble problems and on curved meshes provides convincing evidence that the proposed corrections are viable in realistic smooth-flow settings. The explicit statement of limitations and the clear distinction between proven and empirically observed behavior are commendable.","major_comments":[{"comment":"The proof of Theorem 17 covers two pure cases: elementwise-nonuniform solutions with only the elementwise correction (3.17), and elementwise-constant solutions with only the face-based correction (3.19). However, the P3 scheme as defined applies the face correction through the numerical flux at every face where it is activated, so at a face shared by a uniform and a non-uniform element the beta JbzK term enters the residual of the non-uniform element, whose elementwise correction is also active. The note after Eq. (3.20) explicitly relies on this mixed-face configuration to maintain local energy conservation, but neither case (a) nor case (b) of the proof establishes energy conservation or equilibrium preservation for it. Since the scheme and theorem are stated for the combined corrections, the central theoretical claim is incomplete for mixed faces. Please either provide a proof for the mixed-face case or restrict the theorem and algorithmic guarantee to configurations in which each face connects elements of the same type, and revise the text accordingly.","section":"Sec. 3.3.1 (Theorem 17) and Sec. 3.3 (Eqs. 3.19–3.20)"},{"comment":"The proposed fix for preserving zero species concentrations (zeroing components of r_k, and similarly for the face correction) is applied after alpha and beta have been determined from the energy-balance conditions (3.10) and (3.19). The paper does not analyze whether this post-hoc zeroing preserves the semidiscrete total-energy identity (3.15); in general, removing a nonzero component of the correction term will break the equality (3.10). Since the introduction advertises that the modifications 'enable exact preservation of pressure equilibrium, velocity equilibrium, and zero species concentrations (while maintaining semidiscrete total-energy conservation)', the compatibility of the zero-species fix with energy conservation should be either proved or explicitly qualified. The numerical test in Sec. 4.2 demonstrates zero-species preservation but does not report the energy error for that configuration.","section":"Sec. 3.3, Remark 19"}],"minor_comments":[{"comment":"The text states that the L2 error is computed at t = 0.2 (one advection period), while the caption of Figure 4.1 states t = 1; please correct the inconsistency.","section":"Sec. 4.1 and Fig. 4.1"},{"comment":"In the temporal-convergence paragraph, the results are said to be presented in Figure 4.2, but the correct reference appears to be Figure 4.7.","section":"Sec. 4.4"},{"comment":"The clause 'is the component corresponding to the ith species concentration,.' contains a stray comma; also, the 'if and only if' statement of Lemma 1 should carry the exact-integration qualification stated in Remark 4 directly in the lemma statement, rather than only in a later remark.","section":"Sec. 2.1.1, Eq. (2.13)"},{"comment":"The notation q wT F (y) y is nonstandard and not introduced; replacing it with the average notation used elsewhere, e.g., {{w^T F}}·n, would improve readability.","section":"Sec. 3.3, Eq. (3.20)"},{"comment":"References [20] and [50] appear to be the same paper (Johnson & Kercher, JCP 423, 2020, 109826); please deduplicate the bibliography.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The mixed-face gap in Theorem 17 is the main reason for major revision. Note, however, that all numerical experiments in the paper concern smooth, elementwise non-uniform flows, so the face-based correction is never actually activated in the computations; the practical claims for the tested configurations are supported. The zero-species issue is also real but is confined to an optional modification presented in Remark 19. If the authors can either prove the mixed-face case or explicitly narrow the theorem's scope to the configurations tested, and clarify the energy-conservation status of the zero-species fix, the paper would be a solid contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Candid take: this is a genuinely useful paper, and the main idea is new. It builds a DG scheme for multicomponent thermally perfect gases that replaces total energy with pressure as a state, then uses Abgrall-style correction terms to restore semidiscrete energy conservation, and modifies the correction vector to z so that velocity and pressure equilibria are preserved. The algebraic proofs of Lemmas 1, 9, and Theorem 17 are clean. The numerical experiments in 1D/2D/3D, including curved meshes, show the proposed P3 scheme stays stable and pressure-stable where total-energy DG variants diverge or oscillate, and the temporal convergence of energy conservation is right. The authors are also honest about the limits: semidiscrete-only conservation, the exact-integration caveat in Remark 4, p=1 coarse-mesh fragility, and no released code/data.\n\nThe soft spot is real. The stress-test note about mixed faces is correct. Theorem 17 proves two separate cases: elementwise-nonuniform with only the elementwise correction, and elementwise-constant with only the face correction. The actual scheme is described as combining the two, and the note after (3.20) explicitly invokes a uniform element adjacent to a non-uniform one. At such a face the numerical flux is single-valued, so beta JzK either enters the non-uniform element's balance (case (a) doesn't include beta) or is suppressed, breaking the uniform element's energy argument. The paper gives no selection rule for beta at mixed faces and no proof for that configuration. Since the thermal-bubble tests contain these mixed transitions, the theorem as written does not establish the central claim for the computed cases. The gap is not fatal—the numerics are strong and the fix is likely straightforward—but it needs to be addressed before the proof can be called complete.\n\nCitation pattern is fine; the authors position against Fujiwara, Abgrall, Gaburro, and their own prior work appropriately. I'd send this to a serious referee. Recommend asking the authors to either prove the mixed-face case with an explicit beta rule or soften the claim to the pure cases and present mixed-face behavior as empirical. For my own work: I'd cite it for the correction-vector modification and the numerical demonstrations.","headline":"New and useful DG scheme with clean algebra, but Theorem 17 doesn't cover the mixed uniform/non-uniform faces the actual scheme and tests use.","tokens_in":28451,"tokens_out":3334,"would_cite":true,"duration_ms":32332,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["65M60","76N15"],"pacs":[],"model":"deepseek-v4-flash","headline":"A discontinuous Galerkin scheme that preserves velocity and pressure equilibria while conserving total energy for smooth multicomponent flows.","keywords":["discontinuous Galerkin method","multicomponent flow","pressure equilibrium preservation","velocity equilibrium preservation","total energy conservation","correction terms","compressible Euler equations","thermally perfect gas"],"falsifier":"Simulate an elementwise-constant, two-species interface with equal pressure and velocity, or a smooth thermal bubble, using the modified scheme with third-order Runge-Kutta and systematically halved time steps; if pressure or velocity errors rise above round-off level, or if the global energy error does not shrink at third order with the time step, the central theorem is contradicted. A shock-tube version of the same setup tests whether the promised preservation extends past the stated smooth, constant-equilibrium assumptions.","tokens_in":27434,"feed_emoji":"💨","tokens_out":9699,"duration_ms":93079,"temperature":0.7,"pith_summary":"This paper attacks a long-standing failure mode of conservative numerical methods for compressible multicomponent flows: at interfaces between gases with different specific heats, the computed pressure develops spurious oscillations and the solver can diverge. The authors prove that a standard discontinuous Galerkin (DG) discretization of the conservative Euler equations preserves a constant velocity under Lax-Friedrichs-type fluxes but cannot preserve a constant pressure. They therefore replace the total-energy conservation equation with a pressure-evolution equation, which preserves pressure equilibrium, and then add correction terms that restore semidiscrete total-energy conservation. A modified choice of auxiliary variables makes these corrections invisible to the velocity and pressure equations in constant-pressure, constant-velocity flows, and face-based corrections cover elementwise-constant states. The result is a DG scheme that, for smooth thermally perfect multicomponent flows initially at uniform pressure and velocity, keeps pressure and velocity flat while conserving total energy, and it stays stable for hundreds of advection periods in one, two, and three dimensions.","feed_headline":"New scheme keeps pressure flat while conserving energy","feed_subtitle":"A pressure-based DG method stops spurious oscillations at smooth multicomponent interfaces without sacrificing energy conservation.","key_machinery":"The load-bearing object is the modified correction term $r_k = \\alpha(\\hat{z}_k - \\bar{z})$, where $\\alpha$ is chosen so that the corrected residual satisfies a discrete total-energy balance, and the auxiliary variable is $z = (v \\sum_i W_i \\partial(\\rho e_t)/\\partial C_i,\\; P,\\; \\partial(\\rho e_t)/\\partial C_1, \\ldots, \\partial(\\rho e_t)/\\partial C_{n_s})^T$, with $v$ the velocity and $W_i$ the molar masses. Replacing the usual auxiliary variable $w$ by $z$ makes the correction drop out of the velocity and pressure equations whenever $P=P_0$ and $v=v_0$, because $\\Delta z_{\\rho v} = v_0 \\sum_i W_i \\Delta(\\partial \\rho e_t/\\partial C_i)$ cancels identically, while $\\Delta z_P = 0$. A face-based correction $\\beta \\llbracket z \\rrbracket$ handles elementwise-constant states, and the nonnegativity of the denominator follows from $\\Delta w^T \\Delta z = \\sum_i [\\Delta(\\partial \\rho e_t/\\partial C_i)]^2 \\ge 0$ in equilibrium.","core_discovery":"The central claim is Theorem 17: in a constant-pressure, constant-velocity flow, the DG scheme (3.7) with the modified correction term (3.17) preserves both velocity and pressure equilibria and, assuming continuity in time, conserves total energy at the semidiscrete level. The key is that the correction term is built from an auxiliary variable z whose momentum components are v Σ_i W_i ∂(ρe_t)/∂C_i and whose pressure component is P; in the equilibrium state the correction's contribution to the velocity equation cancels term-by-term and its pressure component is exactly zero. A face-based correction of the form β ⟦z⟧ extends the argument to elementwise-constant solutions with inter-element jumps, and the numerical total-energy flux is defined so that local conservation holds across mixed element types. The numerical experiments show pressure errors at finite-precision levels for the modified scheme over 100 advection periods in one, two, and three dimensions, with third-order temporal convergence of the total-energy error when paired with SSPRK3 time integration.","pith_inferences":["Away from constant-pressure and constant-velocity equilibria, the modified correction's denominator can become negative, so a general-purpose implementation would likely need to switch back to the original correction term or add stabilization; the paper sketches such a hybrid but does not develop it.","The construction is not obviously restricted to the Euler equations: any system where a secondary conservation law is desired alongside a nonconservative primary formulation could use the same auxiliary-variable switching, though the paper does not explore this.","Because the method stores pressure instead of total energy, temperature is obtained directly from the state, which may simplify future real-fluid and chemically reacting extensions; the paper only demonstrates thermally perfect gas mixtures."],"forward_implications":["Smooth interfacial flows of thermally perfect gas mixtures initialized at uniform pressure and velocity can be advected for many periods with only round-off-level pressure error and no artificial viscosity or limiting.","Total-energy conservation is restored to the pressure-based formulation at the semidiscrete level; with explicit third-order SSPRK time stepping the global energy error converges at third order in the time step.","The method works on unstructured and curved meshes in two and three dimensions, where the standard total-energy DG scheme with either colocated or overintegrated quadrature diverges in the tested cases.","The original correction term causes spurious production and destruction of zero-concentration species; the modified correction with zeroed components prevents this, which matters for reacting-flow extensions.","For polynomial degree p=1 the modified correction has a smaller denominator and can degrade accuracy or cause divergence on coarse meshes; p≥2 cases are largely unaffected."],"supporting_citations":[{"why":"Supplies the elementwise correction term used to restore semidiscrete satisfaction of an auxiliary conservation law, here total energy.","marker":"[1]"},{"why":"Extends the correction-term framework to unsteady DG schemes and overintegration, the starting point for the present modified corrections.","marker":"[2]"},{"why":"Provides the discrete pressure-equilibrium condition and the fully conservative flux analysis that the present work extends to DG for thermally perfect mixtures.","marker":"[24]"},{"why":"Introduces the face-based corrections and the total-energy flux form used here for elementwise-constant solutions with jumps.","marker":"[30]"},{"why":"Derives the pressure-evolution equation with the (ρc²-P)∇·v term that replaces the total-energy equation in the pressure-based formulation.","marker":"[7]"},{"why":"Provides the pressure-evolution equation formulation for arbitrary equations of state used in the governing equations.","marker":"[8]"},{"why":"Documents the spurious pressure oscillations and stability failures of fully conservative DG simulations with different integration strategies, providing the numerical baseline for the test cases.","marker":"[3]"}],"fun_headline_variants":["DG scheme preserves pressure and velocity equilibria","Pressure-based DG conserves energy exactly","Energy-conserving DG keeps pressure flat","Multicomponent flow DG preserves equilibria","Pressure-evolution DG avoids spurious oscillations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The guarantee is proven only for smooth flows sitting exactly at constant pressure and velocity; away from that state, at shocks, under nonuniform equilibria, or with real-fluid equations of state, the modified correction can lose its nonnegative denominator and the theorem's cancellations are not shown.","fun_headline_variants_meta":{"raw":{"variants":["DG scheme preserves pressure and velocity equilibria","Pressure-based DG conserves energy exactly","Energy-conserving DG keeps pressure flat","Multicomponent flow DG preserves equilibria","Pressure-evolution DG avoids spurious oscillations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0002,"raw_usage":{"total_tokens":1420,"prompt_tokens":1034,"completion_tokens":386,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":650,"completion_tokens_details":{"reasoning_tokens":321}},"tokens_in":650,"tokens_out":386,"duration_ms":4298,"temperature":1.0,"reasoning_tokens":321,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T17:06:28.067110+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate an elementwise-constant, two-species interface with equal pressure and velocity, or a smooth thermal bubble, using the modified scheme with third-order Runge-Kutta and systematically halved time steps; if pressure or velocity errors rise above round-off level, or if the global energy error does not shrink at third order with the time step, the central theorem is contradicted. A shock-tube version of the same setup tests whether the promised preservation extends past the stated smooth, constant-equilibrium assumptions.","supporting_citations":[{"cited_title":"Abgrall, A general framework to construct schemes satisfying additional conservation relations","cited_arxiv_id":null,"evidence_quote":"Supplies the elementwise correction term used to restore semidiscrete satisfaction of an auxiliary conservation law, here total energy."},{"cited_title":"Abgrall, P","cited_arxiv_id":null,"evidence_quote":"Extends the correction-term framework to unsteady DG schemes and overintegration, the starting point for the present modified corrections."},{"cited_title":"Fujiwara, Y","cited_arxiv_id":null,"evidence_quote":"Provides the discrete pressure-equilibrium condition and the fully conservative flux analysis that the present work extends to DG for thermally perfect mixtures."},{"cited_title":"Abgrall, S","cited_arxiv_id":null,"evidence_quote":"Introduces the face-based corrections and the total-energy flux form used here for elementwise-constant solutions with jumps."},{"cited_title":"Terashima, M","cited_arxiv_id":null,"evidence_quote":"Derives the pressure-evolution equation with the (ρc²-P)∇·v term that replaces the total-energy equation in the pressure-based formulation."},{"cited_title":"Kawai, H","cited_arxiv_id":null,"evidence_quote":"Provides the pressure-evolution equation formulation for arbitrary equations of state used in the governing equations."},{"cited_title":"A note on reducing spurious pressure oscillations in fully conservative discontinuous Galerkin simulations of multicomponent flows","cited_arxiv_id":"2310.17792","evidence_quote":"Documents the spurious pressure oscillations and stability failures of fully conservative DG simulations with different integration strategies, providing the numerical baseline for the test cases."}],"review_version":1}