{"id":"a574d7ff-431b-4433-9a3c-82228ee6b172","arxiv_id":"2601.18404","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"An exact solution framework for the flash-evaporation Riemann problem is derived, and Wood's sound-speed model is shown to be thermodynamically inconsistent with an equilibrium mixture entropy.","lead":"This paper defines a flash-evaporation Riemann problem (a high-pressure liquid suddenly expanding into low-pressure gas while vaporizing) and derives an exact mathematical solution under an equilibrium two-phase model, including non-classical wave structures. It also shows that a widely used simpler sound-speed model (Wood's) produces a non-physical entropy decrease, and supplies MATLAB code for CFD benchmark tests.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sign errors in the §3.3 Jacobian for the RSR/RS Newton solver undermine the published exact-solution algorithm; the physical HEM idealization is secondary.","rationale":"The reader identified the HEM/VLE physical idealization as the weakest assumption. That is a legitimate modeling caveat, but the paper explicitly frames HEM/VLE as an idealization and the exact Riemann-solution construction can still be mathematically valid within that model. The more decisive issue is internal: the Jacobian formulas in §3.3 appear to have reversed signs in the D_b terms. Newton methods are only guaranteed to converge when the Jacobian is the derivative of the residual; with the printed signs, the iteration is not Newton's method for F_RSR. The paper's central claim is an exact, stable solution framework for non-classical waves, so a concrete sign error in the core solver directly undermines that claim. The conditional verdict is still appropriate—the MATLAB code may be correct and the paper may contain transcription errors—but the condition should now explicitly require verification and correction of the §3.3 Jacobian, rather than only the physical modeling assumption. This is why I do not accept the reader's weakest-assumption identification.","tokens_in":27728,"tokens_out":39064,"duration_ms":330599,"concrete_test":"Using the paper’s own PR-EoS routines, evaluate the residual F_RSR of Eq. (49) at the converged Case 1 state and compare the analytic Jacobian from Eqs. (59)–(66) with a finite-difference Jacobian computed from F_RSR. If the printed analytic J differs by the sign of the D_{b,Pa}/D_{b,Pb} terms, correct the signs in Eqs. (59) and (62) and re-run the Case 1/2 Newton iterations. Successful convergence and unchanged Figures 9–10 would show the error is typographical; divergence would show the published algorithm is not reproducible as written.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Most load-bearing: §3.3's Jacobian for the RSR/RS Newton solver contains sign errors in the D_b terms. From the Rayleigh relation j^2=(P_a-P_b)/(v_b-v_a), differentiation at fixed P_b gives (∂j^2/∂P_a)_{P_b}=1/D + j^2 D_{b,Pa}/(ρ_b^2 D) − j^2/(ρ_a^2 c_a^2 D), with D=v_b−v_a. Eq. (59), rewritten with v_a−v_b=−D, has the D_{b,Pa} term with the opposite sign. Similarly, (∂j^2/∂P_b)_{P_a}=−1/D + j^2 D_{b,Pb}/(ρ_b^2 D), while Eq. (62) has the opposite sign on the D_{b,Pb} term. Since the 2×2 Newton step (50) inverts this matrix and Eqs. (68)–(69) propagate J11/J12, the published iteration matrix is not the derivative of the residual (49). A concrete check with an ideal-gas Hugoniot (γ=1.4, P_a=10, P_b=5, v_b on the Hugoniot) gives finite-difference ∂j^2/∂P_a ≈ +0.77, whereas Eq. (59) as printed gives ≈ −0.60. This is an internal mathematical inconsistency, independent of the acknowledged HEM/VLE idealization. If the MATLAB code reproduces the figures, it cannot be using the printed signs; if it does use them, the claimed robust convergence to Chapman–Jouguet double-sonic shocks is unsupported.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper formalizes the Flash evaporation Riemann Problem (FeRP) as a Riemann problem for the one-dimensional Euler equations in which the expansion branch crosses the saturation line, closed with the Homogeneous Equilibrium Model and Vapor-Liquid Equilibrium for a two-parameter equation of state (Peng–Robinson). It derives the required thermodynamic derivatives for single-phase and two-phase states, evaluates the Landau fundamental derivative, and identifies three expansion-wave topologies: simple rarefaction (R), rarefaction + upstream sonic expansion shock (RS), and rarefaction + double sonic expansion shock + rarefaction (RSR). Newton solvers are proposed for the intermediate state and for the Chapman–Jouguet-constrained non-classical waves, with a MATLAB implementation. The paper also develops an exact FeRP solution based on Wood's speed of sound and argues that Wood's model implicitly defines a mechanical mixture entropy, producing a \"density lag\" and a non-physical entropy decrease; benchmark n-dodecane injection cases show that Wood's model underestimates intermediate pressure, velocity, and vaporization relative to the complete equilibrium model.","tokens_in":28156,"tokens_out":8645,"duration_ms":84984,"significance":"If the results are correct, this is a valuable contribution: it gives a complete, explicit thermodynamic-derivative apparatus for PR-EoS two-phase Riemann problems, a concrete construction of non-classical composite waves in the FeRP, a benchmark-quality exact solver, and a clear thermodynamic critique of a widely used sound-speed approximation. The provision of open-source MATLAB code and NIST-calibrated thermophysical parameters are significant reproducibility strengths. However, the central exactness claim depends on the correctness of the §3.3 Jacobian and on independent verification, both of which need attention before the benchmark status can be accepted.","major_comments":[{"comment":"The Jacobian entries for the RSR/RS Newton solver contain sign errors in the D_b terms. From j^2=(P_a-P_b)/(v_b-v_a), with D=v_b-v_a and ∂v_b/∂P_a=-D_{b,Pa}/ρ_b^2, one obtains ∂j^2/∂P_a = 1/D + j^2 D_{b,Pa}/(D ρ_b^2) - j^2/(D ρ_a^2 c_a^2). In the notation of Eq. (59), v_a-v_b=-D, so the coefficient of D_{b,Pa} must be negative, not positive as printed. Similarly, ∂j^2/∂P_b = -1/D + j^2 D_{b,Pb}/(D ρ_b^2), so the coefficient in Eq. (62) must also be negative. A finite-difference check on an ideal-gas Hugoniot (γ=1.4, P_a=10, P_b=5) gives ∂j^2/∂P_a ≈ +0.77, whereas the printed Eq. (59) gives approximately -0.60. Since the Newton step (50) and the RS derivative (68) inherit these entries, the published iteration matrix is not the derivative of residuals (49) and (67). The authors must correct the signs, confirm whether the supplied MATLAB code uses the corrected signs, and rerun/report the","section":"§3.3, Eqs. (59) and (62)"},{"comment":"The paper claims an exact solution and proposes it as a CFD benchmark, but the solver is not verified by any independent check. No convergence residuals, no energy-conservation or Rankine–Hugoniot residuals, no grid-convergence study of the self-similar profiles, and no comparison with another exact solver or with a well-resolved numerical method are reported. Given that the FeRP solution involves non-classical composite waves and that the present review found sign errors in the governing Jacobian, an external verification step is load-bearing for the benchmark claim. I ask the authors to add a validation section quantifying solver residuals and comparing at least one case against an independent reference solution.","section":"§4.3 and Appendix H"}],"minor_comments":[{"comment":"The text introduces Shock–Compression Fan–Shock (SCS) waves but then says \"we will not further discuss the cases involving SRS waves\"; the acronym should be SCS, not SRS.","section":"§3.1, final paragraph"},{"comment":"The caption describes the figure as the phase diagram of \"a low heat capacity fluid,\" but panel (a) is n-dodecane, which the text identifies as a high-heat-capacity fluid. Please reword the caption to cover both the retrograde and non-retrograde cases.","section":"Figure 6 caption"},{"comment":"For RSR waves, λ_tail is listed as u_{CJ,1}-c_{CJ,1}, which is the tail of the head rarefaction segment. Because the full RSR wave also contains a tail rarefaction, clarify that Eq. (37) refers to the tail of each individual rarefaction segment, not to the trailing edge of the entire RSR structure.","section":"§2.5, Eq. (37)"},{"comment":"The symbols c_v and c_l are used for the saturated-phase speeds of sound while C_v is used for constant-volume specific heat. This is a potential source of confusion; consider denoting the saturated speeds of sound by c_sat,v and c_sat,l or similar.","section":"Notation, Eq. (13)"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope. The self-citation (Bai et al. 2025) is limited to iterative inversion routines and is not a circularity concern. The sign errors in §3.3 are local and fixable, but they are load-bearing for the exact-solver claim, so they must be addressed together with an independent validation check before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does two genuinely useful things. It formalizes the flash-evaporation Riemann problem under HEM/VLE, and it gives a credible thermodynamic argument that Wood's speed of sound, when used as the Euler eigenvalue, implicitly redefines mixture entropy and produces a non-physical entropy decrease across a complete expansion. That critique is the most citable part of the paper and I think it is essentially right. The solver design — split rarefactions, RS/RSR composite waves with CJ-constrained Newton — is also a real step beyond prior exact Riemann work for real fluids.\n\nBut there is a load-bearing problem. The stress-test note is correct: differentiating the Rayleigh line j²=(P_a−P_b)/(v_b−v_a) gives a D_{b,Pa} term with the opposite sign to what Eq. (59) prints, and similarly for Eq. (62). I checked the differentiation myself; the note's signs are right. As written, the 2×2 Jacobian in §3.3 is not the derivative of the residual (49), so the published iteration matrix cannot be what the code uses if the code converges as claimed. This is an internal inconsistency, independent of the HEM idealization.\n\nOther soft spots are smaller. The authors promise validation but don't deliver it: no convergence residuals, no energy-conservation errors, no comparison against an independent exact or high-resolution numerical solution. The \"arbitrary EoS\" claim is only exercised with the PR EoS, and the free parameters are just the n-dodecane constants and caloric coefficients — that is fine, but the generality is overstated. The HEM/VLE idealization is real and the authors acknowledge it; I don't hold that against them since they frame it as the intended asymptotic limit.\n\nWho is this for? Anyone building exact or approximate Riemann solvers for two-phase real fluids, and CFD practitioners choosing sound-speed models for flashing propulsion flows. The paper deserves a serious referee — it is not a desk reject. But the referee should require the authors to correct the sign errors in §3.3, state whether the MATLAB code uses the printed signs, and add at least one independent validation case. If the code already has the correct signs, the paper is much closer to being acceptable than the printed equations suggest.","headline":"The FeRP idea and the Wood's-model critique are worth engaging, but the printed §3.3 Jacobian has sign errors that make the published exact-solver derivation internally inconsistent as written.","tokens_in":28616,"tokens_out":5561,"would_cite":true,"duration_ms":50639,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Flash evaporation, treated as a Riemann problem with instantaneous vapor-liquid equilibrium, has an exact solution with composite expansion shocks; Wood's sound-speed model systematically mis-predicts it.","keywords":["Flash evaporation","Riemann problem","Vapor-liquid equilibrium","Homogeneous equilibrium model","Landau fundamental derivative","Chapman-Jouguet condition","Wood's speed of sound","Peng-Robinson equation of state"],"falsifier":"A flash-evaporation shock-tube experiment on n-dodecane with a high-pressure liquid state expanding into low-pressure vapor: if the measured wave profile shows a smooth single rarefaction without the split-wave plateau at the saturation point, or without the predicted expansion-shock jump in velocity and pressure, the HEM/VLE assumption or the underlying thermodynamic derivative derivation would be contradicted. Alternatively, a direct measurement of two-phase sound speed that matches Wood's model rather than c_eq would falsify the claim that complete equilibrium is the correct physical limit.","tokens_in":27660,"feed_emoji":"⚡","tokens_out":4727,"duration_ms":52159,"temperature":0.7,"pith_summary":"Flash evaporation — the violently fast liquid-to-vapor transition triggered by depressurization in engines and fuel injectors — has so far lacked a clean theoretical benchmark. This paper argues that, under the Homogeneous Equilibrium and Vapor-Liquid Equilibrium assumptions, flash evaporation can be formulated as a Riemann problem (two uniform states separated by a discontinuity) that admits an exact self-similar solution for arbitrary two-parameter equations of state. The solution reveals that crossing the saturation line splits rarefaction waves and, for high-heat-capacity fuels like n-dodecane, generates composite waves containing expansion shocks that satisfy the Chapman-Jouguet condition. It then shows that the widely used Wood's speed-of-sound model, despite capturing the qualitative trend, redefines the mixture entropy, produces a 'density lag', and significantly underestimates the intermediate pressure, velocity, and vaporization extent — a result with direct consequences for scramjet fuel-spray CFD.","feed_headline":"Flash evaporation solved exactly as a Riemann problem","feed_subtitle":"Full equilibrium predicts expansion-shock waves; the common Wood's model under-predicts pressure and vaporization.","key_machinery":"The central object is the FeRP formulation under HEM+VLE, together with the Landau fundamental derivative (a measure of isentrope convexity, whose sign and delta-function singularities classify wave structures). The key mechanism is the Chapman-Jouguet condition — the requirement that an expansion-shock speed equal the pre- or post-shock acoustic impedance — which pins down the states in the RS and RSR composite waves and is enforced as the outer constraint of a nested Newton iteration, with Rankine-Hugoniot and isentropic relations as inner constraints. For the model comparison, the central quantity is the difference between the complete equilibrium speed of sound c_eq and Wood's speed c_W,","core_discovery":"The paper establishes an exact solution framework for the Flash evaporation Riemann Problem (FeRP) under the Homogeneous Equilibrium Model and Vapor-Liquid Equilibrium assumptions. It shows that when an isentropic expansion crosses the saturation line, the discontinuity in the equilibrium speed of sound splits rarefaction waves and creates a constant-state plateau in the x–t plane; for retrograde crossing (as with n-dodecane expanding into vapor), a negative delta-function singularity in the Landau fundamental derivative gives rise to non-classical RS and RSR composite waves containing expansion shocks that satisfy the Chapman-Jouguet condition. All thermodynamic derivatives are derived anal","pith_inferences":["The strategy of using the Chapman-Jouguet condition as an outer constraint to avoid trivial convergence likely transfers to other non-classical wave problems, such as BZT fluids or two-phase compression waves, beyond flash evaporation.","The 'density lag' argument implies a simple diagnostic for any two-phase CFD scheme using Wood's speed of sound: monitoring entropy across phase boundaries should reveal the predicted non-physical entropy decrease.","The R/RS/RSR classification depends on the sign of the Landau-derivative singularity at the vapor saturation line, which may be sensitive to the equation of state; testing with a multiparameter EoS could identify fluids that switch between classical and non-classical behavior.","For low-heat-capacity fluids such as CO2 and CH4, the framework predicts only split rarefactions without expansion shocks, offering a directly testable distinction that existing shock-tube data could check."],"forward_implications":["Provides an exact benchmark for verifying CFD codes and approximate Riemann solvers for two-phase flashing flows, including scramjet fuel injection.","For n-dodecane, the expansion branch is classified into R, RS, or RSR waves depending on the intermediate pressure relative to characteristic saturation and Chapman-Jouguet pressures.","Wood's model is thermodynamically inconsistent within the HEM framework, producing a density lag, a non-physical entropy decrease, and systematic under-prediction of intermediate pressure, velocity, and vaporization extent.","The choice of sound-speed model changes wave structures and thermodynamic paths, not just numerical values; in the absence of experimental justification, the complete equilibrium model is recommended."],"fun_headline_variants":["Exact solution to flash evaporation Riemann problem reveals expansion shocks","Flash evaporation: Wood's model under-predicts pressure, exact solution shows","Expansion shocks in flash evaporation: exact Riemann solution","Flash evaporation solved exactly: non-classical waves and Wood's model under-prediction"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The solution assumes phase transition starts exactly on the saturation line and proceeds instantaneously to homogeneous equilibrium, neglecting metastable superheated-liquid states, nucleation barriers, and finite relaxation times; if these timescales are not negligible compared with the macroscopic flow, the Euler equations lose self-similarity and the exact FeRP solution is no longer the relevant physical limit.","fun_headline_variants_meta":{"raw":{"variants":["Exact solution to flash evaporation Riemann problem reveals expansion shocks","Flash evaporation: Wood's model under-predicts pressure, exact solution shows","Expansion shocks in flash evaporation: exact Riemann solution","Flash evaporation solved exactly: non-classical waves and Wood's model under-prediction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000519,"raw_usage":{"total_tokens":2357,"prompt_tokens":754,"completion_tokens":1603,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":498,"completion_tokens_details":{"reasoning_tokens":1527}},"tokens_in":498,"tokens_out":1603,"duration_ms":13922,"temperature":1.0,"reasoning_tokens":1527,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T07:57:59.876503+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A flash-evaporation shock-tube experiment on n-dodecane with a high-pressure liquid state expanding into low-pressure vapor: if the measured wave profile shows a smooth single rarefaction without the split-wave plateau at the saturation point, or without the predicted expansion-shock jump in velocity and pressure, the HEM/VLE assumption or the underlying thermodynamic derivative derivation would be contradicted. Alternatively, a direct measurement of two-phase sound speed that matches Wood's model rather than c_eq would falsify the claim that complete equilibrium is the correct physical limit.","supporting_citations":[],"review_version":1}