{"id":"cb29e6a9-ded9-4cbd-8d07-a956609bdd8e","arxiv_id":"1908.03547","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"An interpolation-free adaptive ALE finite-volume scheme is shown to capture unsteady compression, expansion, and shock waves in NICFD and non-classical regimes on unstructured meshes.","lead":"This paper tests an existing interpolation-free adaptive finite-volume scheme on unsteady flows of dense, non-ideal vapors close to the saturation curve (the NICFD regime). The simulations show the adaptive grid tracks shocks, expansion waves, and reflections in 2D and 3D piston problems without the solution oscillations that remapping usually introduces.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Oscillation-free claim lacks a control: no comparison to interpolation-based or fixed-grid adaptation isolates the method's benefit.","rationale":"I read the paper as a numerical-methods assessment whose central claim is about the interpolation-free adaptive mechanism, not about the fidelity of the Peng-Robinson model. The strongest evidence for that mechanism is the absence of oscillations under adaptation, but the paper never runs a comparison that would let a reader attribute that absence to the interpolation-free construction. The verification against the analytical oblique-shock solution and the 3D symmetry recovery are genuine strengths, and I do not doubt that the method works for the tested cases. The concern is that the headline robustness claim is underdetermined by the evidence: a control experiment is missing. The reader's weakest assumption (polytropic Peng-Robinson EoS fidelity) is also valid and is explicitly flagged by the paper's own v_c mismatch, but it is secondary for the numerical method: the same scheme could be assessed with any EoS, and the non-classical results would change only if the EoS is changed. I therefore disagree with the reader's selection of the single most load-bearing concern, while agreeing that the overall verdict should stay conditional. The recommended check is inexpensive relative to the cost of the published simulations and would decisively separate a property of the method from a property of the test cases.","tokens_in":34115,"tokens_out":9503,"duration_ms":109200,"concrete_test":"Re-run the Section 6.1 TestNI oscillating-piston case (NT = 200) in three configurations: (a) the published interpolation-free ALE scheme, (b) the same Mmg adaptation with the conservative variables linearly interpolated onto the adapted mesh after each adaptation cycle, and (c) a fixed unstructured mesh refined to about h_min throughout. Compare pressure overshoot and undershoot relative to a fine-reference solution, collapse of the scatter to the one-dimensional profile, and total mass and momentum drift over three periods. If configuration (b) shows comparable accuracy and no additional oscillations, and (c) matches (a) within discretization error, then the claimed interpolation-free advantage is not demonstrated and the central assertion should be weakened. If (b) oscillates, loses positivity, or fails near the VLE curve, the paper's mechanism claim is supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing assertion in Section 6.1 is that the method captures all expected NICFD phenomena 'without introducing spurious oscillations due to mesh adaptation.' Every reported run uses the interpolation-free adaptive scheme itself, so the absence of visible oscillations in the pressure scatter plots and profiles cannot be attributed to the absence of interpolation. A standard interpolation-based remap could also be oscillation-free on these fairly smooth piston-generated waves, and the small symmetry disturbances visible in Figures 10 and 11 are noted but not analyzed against any baseline. The interpolation-free ALE machinery is a prior contribution; the new contribution is its NICFD assessment. What would make the central claim load-bearing is a control that isolates the adaptation mechanism: the same solver and Mmg modifications with a solution-interpolation step after remeshing, or a fixed fine mesh at matched resolution. Without such a control, 'robust' is supported only by the absence of failure in a limited set of tests, not by a demonstrated advantage over standard interpolation-based adaptation. The EoS-fidelity issue (v_c^PR/v_c = 1.149, TestNC at Gamma0 = -0.0064) is a real secondary limitation, but it concerns the thermodynamic model rather than the numerical method's claimed mechanism.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper applies an interpolation-free arbitrary Lagrangian-Eulerian (ALE) finite-volume scheme with dynamic unstructured mesh adaptation to the non-ideal compressible fluid dynamics (NICFD) regime. The governing equations are the unsteady Euler equations closed by a polytropic Peng-Robinson equation of state. The method is first exercised on steady oblique and conical shock problems in an unsteady frame, and then on a series of 2D and 3D piston-in-tube problems for the siloxane MD4M, covering dilute, NICFD, and nominally non-classical (Gamma<0) regimes. The central claims are that the scheme detects the expected NICFD phenomena, introduces no spurious oscillations attributable to mesh adaptation, and is robust for unsteady moving-boundary dense-gas flows; the paper also claims to be the first application of unsteady mesh adaptation to NICFD problems.","tokens_in":34332,"tokens_out":4965,"duration_ms":54579,"significance":"If the central claims are substantiated, the paper would provide a useful tool for unsteady dense-gas and organic-vapor flows with moving boundaries, and would extend the rather small body of NICFD numerical validation to adaptive unstructured meshes. The paper has real strengths: the oblique-shock validation in Sec. 5.1 reproduces the analytical post-shock state across three refinement levels and three Courant numbers; the 3D conical-shock test recovers cylindrical symmetry despite an unstructured grid; and the piston tests demonstrate that the adaptive metric tracks traveling waves, including shock formation and wall reflections. The interpolation-free ALE construction, inherited from the authors' earlier work [35-37], is conservative and GCL-compliant, and this prior derivation is cited appropriately. The main limitation is that the paper's differentiating claim, that the absence of solution interpolation prevents spurious oscillations, is not verified by any controlled comparison; and the thermodynamic model's fidelity in the non-classical region is acknowledged but not quantified.","major_comments":[{"comment":"The assertion that the method detects all expected NICFD phenomena \"without introducing spurious oscillations due to mesh adaptation\" is not supported by a comparative experiment. Every piston simulation in the paper uses the interpolation-free adaptive scheme; there is no run with an interpolation-based remap after remeshing and no fixed fine-mesh reference at matched resolution, so the absence of visible oscillations in Figs. 12-13 and 21-22 cannot be attributed to the interpolation-free mechanism. Since the ALE adaptation machinery is taken from Refs. [35,36,37], the paper's incremental contribution is the NICFD assessment; as written, the central robustness claim is an absence-of-failure observation. The small symmetry disturbances in Figs. 10-11 are acknowledged but dismissed without quantitative analysis against any baseline. I request either a control computation that isolates the remap step, or a reformulation of the claim as a limitation.","section":"Sec. 6.1 (concluding paragraph) and Sec. 7"},{"comment":"The non-classical and NICFD results are obtained with the polytropic Peng-Robinson model using the Edmister acentric factor and a constant dilute-gas cv. The paper itself notes that v_c^PR/v_c = 1.149 (Fig. 9 caption), and TestNC starts at Gamma0 = -0.0064, i.e., only marginally inside the Gamma<0 region. The location and width of the Gamma<0 region are therefore model-dependent, so the rarefaction-related behavior reported in Sec. 6.4 could be an artifact of the equation of state rather than a property of MD4M. The authors should either add a sensitivity study (for example, varying the acentric factor or comparing with the multi-parameter EoS of Ref. [99]) or explicitly restrict the conclusions to the PPR model.","section":"Sec. 2.1, Table 3, Fig. 9"},{"comment":"The quantitative validation of the impulsive-piston test is substantially weaker than the oblique-shock validation. The comparison with the analytical post-shock pressure is described only as \"fairly good\", the deviation visibly grows after the second reflection, and no error metric or resolution study is provided for the piston problems beyond the NT sweep reported in Sec. 6.1. Because the paper's overall claim is that the adaptive scheme is robust for unsteady NICFD flows, a quantitative convergence statement (for example, an L1 error versus hmin or NT for the one-dimensional piston configuration) would materially strengthen the argument.","section":"Sec. 6.3, Fig. 19"}],"minor_comments":[{"comment":"The conical-shock validation has no analytical solution and is checked only through steady/unsteady consistency and azimuthal symmetry; this is a clear and honest limitation, but it should be stated explicitly in the main text rather than left implicit in the description of Fig. 7.","section":"Sec. 5.2"},{"comment":"The caption contains a typo, \"Peng-robinson\" should be \"Peng-Robinson\".","section":"Fig. 9 caption"},{"comment":"The phrase \"the previous analysis\" in the concluding paragraph is vague; it would be clearer to refer to the specific figures and quantitative comparisons that support the claim.","section":"Sec. 6.1"}],"recommendation":"major_revision","confidential_remarks":"This is a solid application-oriented paper for a journal like Shock Waves, and the numerical machinery is competently exercised. The main issue is that the paper's central differentiating claim, the absence of interpolation-induced oscillations, is not isolated by any control; this can be fixed by adding a comparison with interpolation-based remap or a matched fixed-mesh run. The EoS-fidelity caveat for the non-classical tests is also worth addressing or tempering. I do not see a fatal flaw, but the load-bearing claim needs additional support before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid assessment paper, not a method-development one. The interpolation-free ALE machinery comes from the authors' own earlier papers; what's new is the extension to the NICFD regime with Peng-Robinson thermodynamics and the piston-test campaign. It deserves a referee.\n\nThe oblique-shock test is the best part: the method reproduces the analytical post-shock state across three minimum edge sizes and Courant numbers up to 4, and the 3D conical-shock test recovers cylindrical symmetry. That is real evidence the adaptive ALE scheme works in the NICFD regime. The piston tests are extensive — open and closed tubes, impulsive start, non-classical states, 3D — and the PPR-vs-PIG comparison in Test D, with pressure deviations around 50%, makes a genuinely useful point about thermodynamic modeling. The paper is also honest about the PR model's known weakness: v_c^PR/v_c = 1.149 is stated in the Figure 9 caption.\n\nThe soft spots are real but not fatal. The Section 6.1 claim that the method works 'without introducing spurious oscillations due to mesh adaptation' lacks a control: every run uses the interpolation-free scheme, so the clean 1D scatter plots don't isolate the mechanism. A head-to-head with an interpolation-based remap, or a fixed fine mesh, would make the advantage claim load-bearing. As written, the evidence supports 'robust in these tests' but not 'better than interpolation-based adaptation.' That is probably a wording problem more than a method problem, but it matters because the abstract leans on it. Second, the non-classical tests sit right at the edge of the PR model's Gamma < 0 region (Gamma0 = -0.0064), and the width of that region depends on an EoS that is 15% off in critical volume. The paper flags the v_c issue but does not say how much the TestNC wave behavior might shift with a more accurate EoS. The piston comparisons against shock relations and the same code's steady solution are useful but not fully independent benchmarks. No code is released, which for a 2019 methods paper is normal but worth noting.\n\nWho this is for: NICFD CFD practitioners, especially people simulating moving-boundary flows in ORC or supercritical-CO2 components, and anyone tracking how adaptive methods behave with real-gas thermodynamics. It deserves a serious referee; the requested revisions are modest — sharpen the interpolation-free claim and add or acknowledge the missing control.","headline":"A solid assessment of the authors' interpolation-free ALE scheme in the NICFD regime, with good shock validation but an oscillation-free claim that lacks a direct control.","tokens_in":34855,"tokens_out":5313,"would_cite":true,"duration_ms":50204,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76M12","76N15","65M50"],"pacs":["47.40.-x","47.11.-j"],"model":"deepseek-v4-flash","headline":"A conservative interpolation-free ALE scheme applies unsteady mesh adaptation to non-ideal compressible-fluid dynamics, reproducing dense-gas and non-classical wave behavior in moving-piston tests.","keywords":["Non-Ideal Compressible Fluid Dynamics","mesh adaptation","Arbitrary Lagrangian-Eulerian","Peng-Robinson equation of state","fundamental derivative of gasdynamics","piston problem","rarefaction shock","siloxane MD4M"],"falsifier":"Repeat the harmonic-piston tests of Section 6.4 using a multiparameter Helmholtz equation of state for MD4M calibrated to the most accurate critical-region data, and compare where rarefaction-shock-like structures form. If the Γ < 0 region from the accurate model lies somewhere other than where the Peng-Robinson model places it—or disappears—then the non-classical results are model-dependent. Alternatively, a direct measurement of the fundamental derivative Γ from sound-speed experiments in near-critical MD4M vapor would settle the same question.","tokens_in":33906,"feed_emoji":"🌊","tokens_out":5384,"duration_ms":53567,"temperature":0.7,"pith_summary":"This paper argues that unsteady mesh adaptation, previously risky in the non-ideal compressible-fluid-dynamics (NICFD) regime because solution interpolation can push thermodynamic states below the vapor-liquid equilibrium curve, can be made safe with an interpolation-free scheme. In the scheme, every local grid change—node insertion, deletion, or edge swap—is treated as a fictitious continuous deformation of the finite volumes inside an Arbitrary Lagrangian-Eulerian (ALE) discretization. Because no solution is ever interpolated onto the new grid, conservation holds by construction and spurious oscillations are avoided. The authors demonstrate the method on piston-in-tube problems filled with the siloxane MD4M, reporting correct compression and expansion waves, shock formation and reflections, and non-classical rarefaction behavior where the fundamental derivative Γ is negative. If the method is right, it supplies a reliable engineering tool for unsteady dense-gas flows with moving boundaries, including the design of organic Rankine cycle and supercritical CO2 components.","feed_headline":"No-interpolation mesh adaptation captures dense-gas shocks and rarefactions","feed_subtitle":"Piston-driven siloxane tests reproduce non-ideal and non-classical wave behavior, with grids that refine and coarsen automatically.","key_machinery":"The load-bearing object is the three-step fictitious deformation that converts a mesh-adaptation operation into an ALE motion. At fictitious time 0 < τ < 0.5 all elements involved in a local grid modification collapse to a point; at τ = 0.5 the connectivity is changed while volumes are null, so no interface sweeps volume and no flux is exchanged; for 0.5 < τ < 1 the surviving elements expand to the final configuration. The volumes swept during collapse and expansion determine the interface velocities through the discrete geometric conservation law, so the conservative properties of the fixed-connectivity scheme carry over to adaptive grids. Around this core, the method uses a node-centered edge-based finite-volume discretization, a Roe-type flux with a non-ideal-gas average, and metric-based adaptation driven by the Hessian of the Mach number.","core_discovery":"The central claim is that a conservative, interpolation-free ALE finite-volume scheme can track unsteady NICFD flow features over dynamically adapted unstructured meshes without contaminating the solution. Connectivity changes are encoded as collapse-and-expansion deformations of the involved volumes, all occurring within a single time step, with the connectivity change occurring at the instant of null volume so no flux is exchanged; the resulting swept volumes feed a geometry-conservation-law-compliant computation of interface velocities. On piston-driven flows of MD4M, the method reproduces the expected qualitative behavior of dense vapors: steepening compressive waves, spreading rarefactions, shock-wall and shock-piston reflections, supercritical excursions, and acoustic-impedance extrema at Γ = 0 in the non-classical region. The authors state that this is the first application of unsteady mesh adaptation to NICFD problems.","pith_inferences":["If the polytropic Peng-Robinson model misplaces the Γ < 0 region—its critical volume is about 15 percent too high—the specific non-classical wave patterns in Section 6.4 may not occur for real MD4M; testing with a multiparameter Helmholtz equation of state would settle this.","The interpolation-free property should matter most for equations of state with a narrow single-phase stability domain; the same adaptation machinery could be combined with tabulated or multiparameter thermodynamics.","The metric built from the Hessian of the Mach number appears preferable to density-based indicators in the NICFD regime because Mach number varies non-monotonically with density; extending this to viscous and turbulent dense-gas flows is a natural next step."],"forward_implications":["Unsteady NICFD simulations with moving boundaries can now follow shocks, rarefactions, and contact discontinuities with meshes that refine and coarsen automatically, at Courant numbers up to about 3 or 4 in the tests.","Large boundary displacements—more than 40 percent of the tube length in the impulsive-start test—can be accommodated without remapping the solution.","The absence of interpolation keeps thermodynamic states away from the two-phase region, making the scheme usable for fluids whose non-ideal region sits close to the vapor-liquid equilibrium curve.","The same solver toolchain can be applied to design dense-gas piston-tube experiments and to improve component simulation in organic Rankine and supercritical CO2 power cycles.","Because the 2D and 3D results agree on one-dimensional flow physics, the method is ready for multidimensional geometries with genuinely three-dimensional wave patterns."],"supporting_citations":[{"why":"Introduces the two-dimensional ALE scheme with connectivity changes whose swept-volume logic the present paper extends to the NICFD regime.","marker":"[35]"},{"why":"Develops the interpolation-free finite-volume treatment over adaptive grids in two dimensions.","marker":"[36]"},{"why":"Extends the interpolation-free ALE scheme to three dimensions and large boundary displacements.","marker":"[37]"},{"why":"Supplies the steady NICFD mesh-adaptation criteria, including Mach-number-based error estimates, that the unsteady method builds on.","marker":"[49]"},{"why":"Provides the Peng-Robinson equation of state used for all non-ideal thermodynamic evaluations.","marker":"[40]"},{"why":"Provides the non-ideal oblique-shock reference case and the phenomenon of Mach-number increase across the shock that the validation reproduces.","marker":"[2]"},{"why":"Performs the local node insertions, deletions, swaps, and regularization that the ALE machinery converts into fictitious deformations.","marker":"[44]"},{"why":"Gives the simplified non-ideal Roe average used to build the numerical fluxes for dense-gas computations.","marker":"[77]"},{"why":"Justifies the Jacobian-form Roe linearization and the choice of intermediate state for non-ideal gases.","marker":"[71]"},{"why":"Provides multiparameter siloxane data used to set critical properties, heat capacity, and related thermodynamic inputs for MD4M.","marker":"[99]"}],"fun_headline_variants":["Interpolation-free ALE captures dense-gas shocks and rarefactions","Adaptive mesh without interpolation resolves dense-gas waves","Interpolation-free mesh adaptation tracks non-ideal gas shocks","Dense-gas shocks and rarefactions via interpolation-free ALE"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the polytropic Peng-Robinson equation of state, with constant dilute-gas heat capacity and a correlation-based acentric factor, represents MD4M well enough that the predicted location and width of the Γ < 0 (non-classical) region match the real fluid; if the model misplaces that region, the reported non-classical wave behavior is an artifact of the thermodynamic model rather than a property of the fluid.","fun_headline_variants_meta":{"raw":{"variants":["Interpolation-free ALE captures dense-gas shocks and rarefactions","Adaptive mesh without interpolation resolves dense-gas waves","Interpolation-free mesh adaptation tracks non-ideal gas shocks","Dense-gas shocks and rarefactions via interpolation-free ALE"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001421,"raw_usage":{"total_tokens":5756,"prompt_tokens":985,"completion_tokens":4771,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":601,"completion_tokens_details":{"reasoning_tokens":4702}},"tokens_in":601,"tokens_out":4771,"duration_ms":31696,"temperature":1.0,"reasoning_tokens":4702,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:09:48.692980+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the harmonic-piston tests of Section 6.4 using a multiparameter Helmholtz equation of state for MD4M calibrated to the most accurate critical-region data, and compare where rarefaction-shock-like structures form. If the Γ < 0 region from the accurate model lies somewhere other than where the Peng-Robinson model places it—or disappears—then the non-classical results are model-dependent. Alternatively, a direct measurement of the fundamental derivative Γ from sound-speed experiments in near-critical MD4M vapor would settle the same question.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the two-dimensional ALE scheme with connectivity changes whose swept-volume logic the present paper extends to the NICFD regime."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the steady NICFD mesh-adaptation criteria, including Mach-number-based error estimates, that the unsteady method builds on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the non-ideal oblique-shock reference case and the phenomenon of Mach-number increase across the shock that the validation reproduces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Justifies the Jacobian-form Roe linearization and the choice of intermediate state for non-ideal gases."}],"review_version":1}