{"id":"9d32078e-6fe8-428c-981c-b30ce771000d","arxiv_id":"2411.16292","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A temperature-dependent dispersion relation and phase-field simulations show how the Rayleigh-Taylor instability weakens as a binary fluid mixture approaches its upper critical solution temperature.","lead":"This paper extends the classic Rayleigh-Taylor instability to binary fluids whose two components become more miscible as temperature rises toward a critical point. It derives a temperature-dependent growth-rate formula and backs it with phase-field simulations, which could help design temperature-controlled mixing and separation in microfluidics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The empirical exponent a, fixed at 0.5 for every simulation, controls all temperature exponents; the paper's own IBW value a=0.27 shifts the predicted threshold by about 10.7% and changes near-critical capillary forces by roughly 28x, so quantitative temperature predictions are not robust.","rationale":"The reader identified the same weakest assumption: the empirical exponent a=0.5 is fixed without sensitivity analysis. My stress-test confirms this concern and quantifies it: a is not a minor fitting parameter but controls the two temperature exponents in the dispersion relation. Using the experimentally cited value a=0.27 changes the predicted threshold by ~11%, and near the UCST the capillary stabilization term is lower by a factor of ~28, which would substantially alter growth-rate predictions and potentially the zone boundaries and KH-roll suppression. The paper's numerical validation is internally consistent but cannot establish physical validity because the same model with the same a is used in both linear theory and simulation. The derivation itself is algebraically sound: I checked the equilibrium profile, surface-tension integral, and the Fourier relation leading to eq. 4.2, and the sign and factors are correct. Therefore the concern is not a fatal flaw but a missing validation and sensitivity step; the CONDITIONAL verdict remains appropriate.","tokens_in":18996,"tokens_out":21438,"duration_ms":186018,"concrete_test":"Recompute eq. 6.2 for the §6.1.1 test parameters (A=0.1, We=100, k=2π) with a=0.27, the isobutyric acid-water value cited in the paper; the analytic threshold shifts from r_th=0.465 (a=0.5) to r_th=0.515, a ~10.7% change. To settle the dynamical impact, rerun the numerical threshold determination of fig. 9 with a=0.27 and check whether the marginal r shifts correspondingly and whether near-critical growth rates (eq. 6.1) change by more than ~10%.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims—threshold r_th (eq. 6.2), growth-rate scaling (eq. 6.1), and the zone boundaries in fig. 8(b)—all depend on the empirical exponent a through the factors r^{(1-a)/2} and r^{(3-2a)/2}. The manuscript fixes a=0.5 for all simulations with no justification or sensitivity analysis. This value is not representative: eq. 2.8 and the cited May & Maher (1991) data for isobutyric acid-water give a=0.27, i.e., a surface-tension exponent 1.23 instead of 0.5. Recomputing the test case of §6.1.1 (A=0.1, We=100, k=2π) with eq. 6.2 changes r_th from 0.465 to 0.515, an 11% shift; at r=0.01 the capillary term r^{(3-2a)/2} differs by a factor of 28 between a=0.5 and a=0.27. Thus the numerical 2.15% threshold agreement only verifies that the nonlinear solver matches the linear theory for the same arbitrarily chosen a; it provides no evidence that the model predicts real binary-fluid behavior. The qualitative KH-roll suppression near the UCST may also depend on this choice, because it relies on the relative strengths of the same r-dependent buoyancy and capillary terms.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a phase-field (Cahn-Hilliard-Navier-Stokes) model for binary fluids with a temperature-dependent miscibility gap, using a quartic bulk free energy depending on the reduced temperature r=(T_c-T)/T_c and an empirical exponent a. For single-mode Rayleigh-Taylor instability the authors derive an inviscid Boussinesq dispersion relation, eq. (4.3): alpha^2(k)=k A g r^{(1-a)/2} - k^3/(2 We_B) r^{(3-2a)/2}. They use it to classify stable/unstable regions and three growth-rate zones, then solve the nonlinear CH-NS equations with an OpenFOAM-based solver. The simulations reproduce the linear threshold (r=0.475 vs r_th=0.465) in §6.1.1, benchmark the r=1 limit against Hamzehloo et al., compare configuration (2) with Lyubimova et al., and document late-time Kelvin-Helmholtz roll behaviour as a function of r.","tokens_in":19421,"tokens_out":10952,"duration_ms":95987,"significance":"If the temperature scalings are robust, the paper offers a useful analytic and numerical framework for RT instability in partially miscible binary fluids. The dispersion relation transparently reduces to Chandrasekhar's result at r=1, the numerical study is carefully set up with grid-independence checks and two published benchmarks, and the three-zone classification together with the observed suppression of Kelvin-Helmholtz rolls near the UCST are potentially valuable predictions. The main caveat is that all quantitative temperature dependencies are controlled by the empirical exponent a, which is fixed at a=0.5 without sensitivity analysis or experimental calibration for the fluids simulated; the numerical agreement with the linear theory then tests internal consistency rather than external predictive accuracy.","major_comments":[{"comment":"The central quantitative claims - the threshold r_th, the growth-rate scaling, and the zone boundaries in fig. 8(b) - all depend on the empirical exponent a through the factors r^{(1-a)/2} and r^{(3-2a)/2}, yet a is fixed at 0.5 for every simulation with no sensitivity analysis. The only experimentally calibrated value cited in the paper (May & Maher 1991 for isobutyric acid-water) gives a=0.27, i.e. a surface-tension exponent (3-2a)/2 = 1.23 rather than 1.0. For the test case of §6.1.1 (A=0.1, We=100, k=2π), eq. (6.2) changes r_th from 0.465 to 0.515, an 11% shift; the near-critical capillary term r^{(3-2a)/2} changes from r to r^{1.23}. Because the numerical simulations and the linear theory use the same free energy and the same a, the reported 2.15% agreement in §6.1.1 validates the internal consistency of the solver with the linear theory, not the predictive accuracy for a real binary fluid. Please add a sensitivity study over a, calibrate a for the fluids considered, or explicitly restrict the quantitative predictions to a model fluid with a=0.5.","section":"§6.1, eqs. (6.1)-(6.2), figs 7-8"},{"comment":"External validation is limited to the immiscible limit r=1 (against Hamzehloo et al.) and to configuration (2) (against Lyubimova et al.); there is no independent benchmark in the partially miscible regime 0<r<1, where the temperature-dependent terms are active. The claim that the simulations corroborate the linear threshold is therefore not an independent test of the model's temperature dependence. Please add at least one quantitative check in the partial-miscibility regime - for example, a comparison with measured surface tension versus temperature for a specific fluid pair, or with a published miscible-RT experiment - or state clearly that the r-dependent predictions are model predictions to be tested.","section":"§5.3 and §6.1.1"}],"minor_comments":[{"comment":"The phrase 'reformulate the reformulate the governing equations' contains a duplicated word and should be corrected.","section":"Abstract/§1"},{"comment":"The integration constant Z in the derivation of the equilibrium profile appears to be missing factors: from eqs (A 2)-(A 3) one obtains Z = (Lambda/(4 epsilon^2)) r^{2-a}, but the text as printed reads as if Z were simply (1/4) epsilon^2 r^{2-a}.","section":"Appendix A"},{"comment":"The sentence 'The growth behaviour demonstrates week dependence on We' should read 'weak dependence'.","section":"§6.1.2"},{"comment":"Editorial template placeholders such as 'Focus on Fluids articles must not exceed this page length' and 'Rapids articles must not exceed this page length' appear in the text and should be removed before submission.","section":"Manuscript text"},{"comment":"The step from eq. (B 5) to the dispersion relation (B 7) is compressed; since eq. (4.3) is the central analytical result, please expand the Fourier-transform step or give a clear pointer to the supplementary material where it is performed.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The core derivation appears sound and the numerical work is careful, so rejection is not warranted. The main editorial question is whether the quantitative claims can be made without calibrating or varying the empirical exponent a; I would ask for sensitivity analysis or a clear re-scoping of the claims before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis is a solid extension of phase-field RT theory to binary fluids with a temperature-dependent miscibility gap. The genuinely new items are the r-dependent dispersion relation (eqs 4.2/4.3), the threshold formula (4.4/6.2), and the three-zone regime map in fig. 8(b). The derivation follows Celani et al.'s potential-flow approach, reduces cleanly to the Chandrasekhar result at r=1, and the linear theory matches the early-stage simulations, including a threshold bracketed at r=0.475 vs 0.465 from eq. 6.2. The finite-viscosity simulations are careful, and the comparisons against Hamzehloo et al. (r=1) and Lyubimova et al. (non-equilibrium configuration) are appropriate external anchors. The late-time observation that KH-roll formation is suppressed close to the UCST is a nice qualitative result.\n\nThe main soft spot is the empirical exponent a. The model leaves a as a free fluid-specific parameter, and the paper fixes a=0.5 for all simulations without any sensitivity analysis. But the paper itself cites May & Maher (1991) giving a=0.27 for isobutyric acid-water (surface-tension exponent 1.23). Since every temperature exponent in eqs 4.2, 4.3, 6.1 and 6.2 scales with a, this choice matters quantitatively: for the section 6.1.1 test case, a=0.27 shifts the predicted threshold from 0.465 to about 0.515 (an 11% change), and at r=0.01 the capillary term r^{(3-2a)/2} changes by a factor of roughly 28. So the 2.15% threshold agreement is an internal consistency check between the solver and the same free energy, not independent validation of the temperature dependence. The qualitative three-zone picture probably survives for any a in (0,1), but the predicted thresholds, growth-rate scalings, and zone boundaries should not be treated as quantitative for any real fluid yet. No code or data are provided, and the growth-rate comparisons lack error bars.\n\nWho it's for: phase-field and RT specialists, and anyone designing temperature-controlled microfluidic or extraction processes with UCST mixtures. It deserves a serious referee: the derivation is new and the numerics are competent, but the referee should require a sensitivity analysis in a and ideally a demonstration with a real fluid pair (e.g., IBW) where a is known. With that addition, the paper becomes a reliable reference for the field.\n\nRecommendation: send it to review, and expect heavy but doable revision focused on the a-dependence.","headline":"A competent phase-field extension of RT theory to UCST binary fluids with a genuinely new r-dependent dispersion relation, but the quantitative temperature predictions rest on an unexamined empirical exponent a=0.5.","tokens_in":19865,"tokens_out":3126,"would_cite":true,"duration_ms":28470,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["47.20.Ma"],"model":"deepseek-v4-flash","headline":"The paper derives a temperature-dependent dispersion relation for Rayleigh-Taylor instability in binary fluids with a miscibility gap and confirms it with phase-field simulations.","keywords":["Rayleigh-Taylor instability","binary fluids","miscibility gap","phase-field method","Cahn-Hilliard-Navier-Stokes","upper critical solution temperature","dispersion relation","Kelvin-Helmholtz instability"],"falsifier":"Prepare a temperature-controlled cell with a single-mode perturbation between two fluids of known surface-tension scaling (for example isobutyric acid and water, with a=0.27), measure the perturbation amplitude versus time below the UCST, and compare the growth-rate exponent and threshold r_th against eq. (6.1); a mismatch in the temperature dependence of the growth rate would falsify the model.","tokens_in":18776,"feed_emoji":"🌡️","tokens_out":11170,"duration_ms":126523,"temperature":0.7,"pith_summary":"What the paper sets out to establish: for a binary fluid pair whose mutual solubility is controlled by temperature through an upper critical solution temperature, the early growth of a Rayleigh-Taylor perturbation is a predictable function of one dimensionless parameter, $r = (T_c - T)/T_c$, measuring how far the system sits from the critical point. The paper derives the dispersion relation $\\alpha^2(k) = k A g r^{(1-a)/2} - k^3/(2 We_B) r^{(3-2a)/2}$ from potential-flow and Boussinesq assumptions, then shows that coupled Cahn-Hilliard-Navier-Stokes simulations reproduce its threshold and growth-rate predictions at early times. It also classifies fluid pairs into three zones according to how the growth rate responds as the miscibility gap is approached, and reports that late-time Kelvin-Helmholtz secondary rolls appear only when the interface is sharp enough, far enough from the critical point. If correct, the model gives a way to design or suppress interfacial mixing simply by choosing the operating temperature relative to the critical temperature.","feed_headline":"Formula predicts Rayleigh-Taylor growth in partially miscible fluids","feed_subtitle":"Growth-rate curve and simulations show temperature alone can switch binary-interface instability on or off.","key_machinery":"The machinery is a Cahn-Hilliard phase-field model with a temperature-dependent bulk free energy $f_0 = (\\Lambda/\\epsilon^2)(\\tfrac14 |r|^a c^4 - \\tfrac12 r c^2)$, where $r=(T_c-T)/T_c$ is the miscibility parameter. This free energy yields an equilibrium interface profile $c = -r^{(1-a)/2}\\tanh\\big((y-y_0)/(\\sqrt{2}\\,\\epsilon/\\sqrt{r})\\big)$ and a surface tension $\\sigma = \\sigma_0 r^{(3-2a)/2}$, so both interfacial width and capillary force become functions of temperature. Linearizing the inviscid, Boussinesq momentum equation about this profile and Fourier-decomposing the perturbed interface gives the dispersion relation (4.3), the object whose predictions the simulations test.","core_discovery":"The central claim is that the dispersion relation $\\alpha^2(k)=k A g r^{(1-a)/2} - k^3/(2 We_B) r^{(3-2a)/2}$ is the correct linear-theory description of single-mode Rayleigh-Taylor instability in binary fluids with a temperature-dependent miscibility gap. In the immiscible limit $r\\to 1$ it reduces to the classical Rayleigh-Taylor dispersion relation, and temperature enters through powers of $r$: buoyant destabilization scales as $r^{(1-a)/2}$ while capillary stabilization scales as $r^{(3-2a)/2}$. Because $3-2a > 1-a$ for $0<a<3/2$, approaching the critical point weakens the stabilizing term faster than the destabilizing term, producing the threshold $r_{th}$ and the three observed growth-rate zones. The numerical solutions of the coupled Cahn-Hilliard-Navier-Stokes equations corroborate the inviscid linear analysis in the early stage, reproduce gravity-capillary waves and marginal stability, and at late times show Kelvin-Helmholtz rolls for $r=1$ and $r=0.3$ but not for $r=0.01$. The paper further claims that initialization out of thermodynamic equilibrium makes the temperature influence weak except when surface tension is large.","pith_inferences":["One testable extension the paper leaves implicit: the exponent $a$ is fixed at $0.5$ in all simulations, but real fluid pairs differ (isobutyric acid-water has $a=0.27$), so a sensitivity scan over $a$ would show how much the zone boundaries and threshold shift for a given pair.","The same dispersion-relation structure should extend to LCST mixtures by redefining $r$ with the lower critical temperature, since the surface-tension scaling $\\sigma\\sim r^{(3-2a)/2}$ comes from the equilibrium free energy rather than from the RT setup itself.","A direct experimental check would be to measure the early-time growth rate of a single-mode perturbation for a known binary pair at several temperatures below the UCST and compare the temperature exponents of the growth-rate curve with $(1-a)/2$ and $(3-2a)/2$.","The suppression of Kelvin-Helmholtz rolls near the UCST could become a practical control knob in microfluidic extraction or droplet formation, where thermal tuning of the miscibility gap is already used."],"forward_implications":["The predicted threshold $r_{th}$ fixes the temperature at which a given binary-fluid interface turns unstable; systems closer to the critical point than this threshold remain stable to small single-mode perturbations.","The three-zone classification means that for low-Atwood, high-surface-tension pairs, approaching the UCST first destabilizes the interface, while for high-Atwood pairs it monotonically stabilizes it.","Late-time Kelvin-Helmholtz rolls are temperature-dependent, so tuning $T$ relative to $T_c$ can suppress or promote secondary mixing without changing the prescribed density contrast.","For fluids brought into contact out of equilibrium, the model predicts that at low surface tension the interface evolution is nearly independent of whether the system temperature lies above or below the critical point.","Near the UCST the concentration field almost homogenizes, which reduces early growth but delays and strengthens the reacceleration phase of the falling spike."],"supporting_citations":[{"why":"Supplies the phase-field linear-stability method that the paper extends to binary fluids with a miscibility gap.","marker":"Celani et al. (2009)"},{"why":"Provides the classical RT dispersion relation that the new relation must reduce to in the immiscible limit.","marker":"Chandrasekhar (1961)"},{"why":"Provides the direct numerical simulation benchmark for bubble velocity used to validate the solver.","marker":"Hamzehloo et al. (2021)"},{"why":"Earlier phase-field model for binary miscible/immiscible fluids whose free-energy treatment the present model improves.","marker":"Bestehorn et al. (2021)"},{"why":"Extends the earlier model to temperature-field evolution; the present model corrects the Korteweg-stress handling beyond the UCST.","marker":"Borcia et al. (2022)"},{"why":"Gives the diffuse-interface free-energy functional and surface-tension definitions used in the model.","marker":"Yue et al. (2004)"},{"why":"Supplies the Cahn-Hilliard/Navier-Stokes coupling and Korteweg-stress formulation underlying the simulations.","marker":"Jacqmin (1999)"},{"why":"Provides the experimental surface-tension scaling for isobutyric acid-water that fixes the example value a=0.27.","marker":"May & Maher (1991)"},{"why":"Documents the late-time reacceleration and Kelvin-Helmholtz dynamics used to interpret the secondary instability.","marker":"Ramprabhu et al. (2012)"},{"why":"Provides the deceleration-acceleration mechanism and Q-criterion vortex identification used in the viscous results.","marker":"Hu et al. (2019)"}],"fun_headline_variants":["Temperature switch for binary-fluid Rayleigh-Taylor growth","Temperature alone can toggle Rayleigh-Taylor in binary fluids","Miscibility gap sets Rayleigh-Taylor growth-rate zones","Temperature controls Rayleigh-Taylor instability in miscibility-gap fluids"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the empirical exponent a, fixed at a = 0.5 throughout the simulations, correctly represents the fluid pair; every temperature exponent in the dispersion relation, threshold, and zone boundaries depends on a, and the paper gives no experimental validation or sensitivity analysis for this choice.","fun_headline_variants_meta":{"raw":{"variants":["Temperature switch for binary-fluid Rayleigh-Taylor growth","Temperature alone can toggle Rayleigh-Taylor in binary fluids","Miscibility gap sets Rayleigh-Taylor growth-rate zones","Temperature controls Rayleigh-Taylor instability in miscibility-gap fluids"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000727,"raw_usage":{"total_tokens":3296,"prompt_tokens":1023,"completion_tokens":2273,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":639,"completion_tokens_details":{"reasoning_tokens":2208}},"tokens_in":639,"tokens_out":2273,"duration_ms":16951,"temperature":1.0,"reasoning_tokens":2208,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:17:19.057999+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Prepare a temperature-controlled cell with a single-mode perturbation between two fluids of known surface-tension scaling (for example isobutyric acid and water, with a=0.27), measure the perturbation amplitude versus time below the UCST, and compare the growth-rate exponent and threshold r_th against eq. (6.1); a mismatch in the temperature dependence of the growth rate would falsify the model.","supporting_citations":[{"cited_title":", Mazzino, A","cited_arxiv_id":null,"evidence_quote":"Supplies the phase-field linear-stability method that the paper extends to binary fluids with a miscibility gap."},{"cited_title":"1961 Hydrodynamic and Hydromagnetic stability\\/","cited_arxiv_id":null,"evidence_quote":"Provides the classical RT dispersion relation that the new relation must reduce to in the immiscible limit."},{"cited_title":", Bartholomew, P","cited_arxiv_id":null,"evidence_quote":"Provides the direct numerical simulation benchmark for bubble velocity used to validate the solver."},{"cited_title":", Sharma, D","cited_arxiv_id":null,"evidence_quote":"Earlier phase-field model for binary miscible/immiscible fluids whose free-energy treatment the present model improves."},{"cited_title":", Borcia, I.D","cited_arxiv_id":null,"evidence_quote":"Extends the earlier model to temperature-field evolution; the present model corrects the Korteweg-stress handling beyond the UCST."},{"cited_title":", Feng, J.J","cited_arxiv_id":null,"evidence_quote":"Gives the diffuse-interface free-energy functional and surface-tension definitions used in the model."},{"cited_title":"1999 Calculation of two-phase Navier-Stokes flows using Phase-Field modeling","cited_arxiv_id":null,"evidence_quote":"Supplies the Cahn-Hilliard/Navier-Stokes coupling and Korteweg-stress formulation underlying the simulations."},{"cited_title":"& Maher, J.V","cited_arxiv_id":null,"evidence_quote":"Provides the experimental surface-tension scaling for isobutyric acid-water that fixes the example value a=0.27."},{"cited_title":", Dimonte, G","cited_arxiv_id":null,"evidence_quote":"Documents the late-time reacceleration and Kelvin-Helmholtz dynamics used to interpret the secondary instability."}],"review_version":1}