{"id":"7db1fce2-469a-4aa9-9105-7d9149ff3902","arxiv_id":"2411.17968","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A zero-parameter Flory-style entropy of mixing correction to the ideal solution model improves predicted liquidus and eutectic behavior for ten aqueous organic binary systems.","lead":"Researchers added a size-dependent entropy term to the classical ideal solution model and used it to predict freezing curves of water mixed with ten organic molecules. The model needs only known properties of the pure components, and it cuts average liquidus temperature error by 59% compared with the standard ideal model.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The zero-enthalpy, pure-size-entropy interpretation is not independently tested: the size term always lowers the liquidus, so improved liquidus fits alone do not establish entropic dominance. Eqn (4)'s predicted water activities should be checked against measured values.","rationale":"The derivation was checked: differentiating Eqn (2) gives Eqn (4) with the correct Flory-Huggins form, and Eqn (9) follows from equating chemical potentials with a pure solid and using DeltaG_fus = DeltaH_fus(1 - T/Tm). There is no algebraic error in the central construction, and the zero-parameter nature of the model is genuine. The reader's CONDITIONAL verdict is fair. The remaining gap is that the paper's evidence for the physical-attribution claim is indirect. Liquidus temperature is an integrated quantity; an athermal size-entropy term and an enthalpic interaction term can produce similar liquidus depressions, and the one-directional sign of the Flory correction makes improvement over the classic ideal model almost guaranteed for systems whose experimental liquidi lie below Raoult behavior. The sucrose residual, explicitly acknowledged in Fig. 1c, shows that the zero-enthalpy premise is not universal, and the restriction to non-hydrate, non-ionic simple-eutectic systems limits the breadth of the 'aqueous organic solutions' conclusion. A direct check of the model's predicted water activities from Eqn (4) would settle whether the size-entropy terms are thermodynamically real or merely a convenient bias. I therefore agree with the reader's weakest assumption and recommend keeping the CONDITIONAL verdict, with the condition expanded to include this independent activity cross-check.","tokens_in":9331,"tokens_out":16310,"duration_ms":148830,"concrete_test":"Take measured water activities for water-glycerol, water-urea, and water-sucrose (plus glucose, ethylene glycol, and sorbitol if data are available) at 25 C over the composition range up to the eutectic or solubility limit. Compare them with a_w = phi_w exp[(1-phi_w)(1 - v_w/v_solute)] from Eqn (4), using the same pure-component molar volumes as the paper. If the predicted water activity deviates by more than about 0.05 for the strongly hydrogen-bonding solutes while the liquidus still matches, the zero-enthalpy/size-entropy interpretation is not supported, and the reported liquidus improvement should be treated as a biased-baseline artifact rather than evidence of entropic dominance.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The model's practical claim (Eqn (9) is a zero-parameter liquidus predictor) is supported by the internal derivation and the reported error reductions. The load-bearing weakness is the paper's causal interpretation: that 'much of the deviation from ideality ... may be explained by simple entropic size effects' rather than by molecular interactions. This inference is made from liquidus comparisons alone. But the Flory term in Eqn (1) lowers the predicted liquidus for every VR != 1, and the ten simple-eutectic systems studied have experimental liquidi that lie predominantly below the classic ideal prediction. A one-directional offset will mechanically reduce the average error for this dataset whether or not the true non-ideality is entropic. The paper itself concedes in Fig. 1c and the Discussion that sucrose retains deviations attributed to 'additional interaction factors', so the zero-enthalpy premise is already known to fail in at least one of the ten systems. Eqn (2) makes an unambiguous, parameter-free prediction for component activities, e.g. for water from Eqn (4): a_w = phi_w exp[(1-phi_w)(1 - v_w/v_solute)]. These predictions are never compared with experimental water activities. If they fail systematically for hydrogen-bonding solutes (glycerol, sucrose, urea), the liquidus improvement cannot be taken as evidence of entropic dominance; it would be a biased-baseline artifact. This is the central load-bearing concern because it targets the paper's main physical conclusion, not just the empirical correlation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes a size-dependent extension of the ideal solution model for binary aqueous organic solutions. The key idea is to replace the mole-fraction entropy of mixing with a Flory-style volume-fraction entropy (Eqn. 1), so that the chemical potential of each component depends on molar-volume ratios. The authors derive a liquidus-temperature formula (Eqn. 9) whose only inputs are pure-component melting temperature, enthalpy of fusion, and molar volume, and they compare its predictions with literature phase diagrams for ten aqueous organic systems. They report a 59% reduction in liquidus residual temperature, a 45% reduction in eutectic temperature error, and a 43% reduction in eutectic composition error relative to the classic ideal model, and they interpret the improvement as evidence that size-dependent mixing entropy, rather than molecular interactions, dominates deviations from ideality in these solutions.","tokens_in":9570,"tokens_out":17649,"duration_ms":148968,"significance":"If the central claims hold, the model offers a genuinely zero-parameter improvement over the ideal solution model for a practically important class of mixtures, with possible applications in cryopreservation, deep eutectic solvents, and solution design. The derivation is transparent and the inputs are standard pure-component properties, which makes the approach easy to adopt and test. The strength of the paper is its parameter-free predictive structure and the explicit comparison against external experimental phase diagrams. However, the manuscript's physical interpretation is not supported by independent thermodynamic evidence, and the printed central equation contains a sign error that as written would make the model predict liquidus temperatures above the melting point. These issues need to be resolved before the quantitative claims can be accepted.","major_comments":[{"comment":"The sign in Eqn. (9) is inconsistent with Eqn. (8). Solving Eqn. (8) for 1/T_liq gives 1/T_liq = 1/T_m + R[ln(1/φ_A) - (1-φ_A)(1 - 1/V_R,A)] / Δ_fus H_A^0, not the minus sign printed in Eqn. (9). With the printed minus sign, the predicted liquidus lies above the pure melting temperature for ordinary cases with V_R>1; for example, with φ_A=0.5, V_R=4, T_m=273 K, and Δ_fus H=6000 J/mol, Eqn. (9) gives T_liq≈310 K, whereas the physically correct plus sign gives T_liq≈244 K. This contradicts the text's assertion in the same section that 'any V_R,A ≠ 1 will lower the liquidus temperatures.' The results in Figs. 1-3 depend on Eqn. (9), so the equation must be corrected and all figures and error statistics rechecked against the corrected formula.","section":"The Size-Dependent Ideal Solution Model, Eqn. (9)"},{"comment":"The paper's central physical conclusion—that much of the deviation from ideality in aqueous organic solutions is entropic rather than enthalpic—is not established by liquidus comparisons alone. The size term in Eqn. (1) always depresses the predicted liquidus for V_R≠1, and the ten experimental datasets are systems whose liquidi lie predominantly below the classic ideal prediction, so a one-directional baseline shift will reduce the mean error whether or not the true non-ideality is entropic. Eqn. (4) makes a parameter-free prediction for component activities, e.g. water activity a_w = φ_w exp[(1-φ_w)(1 - v_w/v_solute)], which can be compared directly with published water-activity measurements for glycerol, sucrose, urea, and other systems; the manuscript reports no such comparison. Without this independent test, or an explicit disclaimer that the zero-enthalpy assumption is a modeling ansatz, the 'underappreciated dominance of mixing entropy' conclusion is an overreach. The sucrose case in Fig. 1c, where residual deviations are attributed to 'additional interaction factors,' already shows that the zero-enthalpy premise fails in at least one of the ten systems.","section":"Discussion, Eqn. (4)"},{"comment":"The headline 59% reduction in liquidus error is based on an 'aggregate residual' defined as a trapezoidal integral of the difference between experimental and predicted liquidus temperatures. If this is a signed integral, positive and negative deviations cancel, so a poor fit that crosses the data can appear better than a consistently biased fit. The caption does not specify whether absolute values were used, and the text in the same section that 'the residual of the classic ideal model diverges' suggests a signed cumulative quantity. The paper should report a mean absolute deviation or root-mean-square error over the composition range, and if the figures used signed residuals, the three quantitative claims in the abstract need to be recomputed with an absolute metric.","section":"Comparison of model to experimental data, Fig. 2a"},{"comment":"The validation set is limited to ten non-ionic aqueous organic systems exhibiting simple eutectic behavior, and some of the reference data are not fully equilibrium simple-eutectic data: ethylene glycol and D-fructose are compared using metastable simple eutectics, and sorbitol and maltitol have interpolated rather than measured eutectic compositions. Because Eqn. (9) can only generate a simple eutectic between two pure solids, using metastable or interpolated reference points may bias the comparison in the model's favor. The abstract and conclusion generalize to 'aqueous organic solutions' without these restrictions; the claims should be explicitly scoped to non-ionic simple-eutectic systems, or additional systems with compound formation or hydrates should be tested.","section":"Comparison of model to experimental data, first paragraph and Fig. 2"}],"minor_comments":[{"comment":"The word 'entopic' in the abstract should be 'entropic.'","section":"Abstract"},{"comment":"The text says the free energy is 'differentiated with respect to x_A,' but the chemical potential is defined by differentiation with respect to the amount of component A at constant T, P, and n_B; the resulting expression appears correct, but the stated operation is not the standard definition.","section":"Model, Eqn. (3)"},{"comment":"The manuscript asserts that excess volumes in aqueous organic solutions 'generally remain <1%' and refers to Supplementary Note 1, but the relevant evidence is not presented in the main text; a brief sensitivity table would make the claim checkable.","section":"Model, Discussion of excess volume"},{"comment":"For reproducibility, the authors should provide the tabulated pure-component input data, the ten experimental datasets, and the model curves in a repository, since the paper's value proposition depends on the exact numerical comparison.","section":"Data availability"},{"comment":"Reference 16 contains a typo ('thoery' should be 'theory').","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a useful problem and the parameter-free idea is appealing, but the printed Eqn. (9) has a sign error that, unless corrected, makes the model unphysical. In addition, the claim of entropic dominance is not supported by independent activity or calorimetric evidence; if the authors soften that interpretation and present the model as a useful zero-parameter predictor for non-ionic simple-eutectic systems, the paper would be much stronger. I would recommend major revision rather than rejection because the errors appear fixable within the scope of the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The practical part is genuinely useful: a zero-parameter liquidus equation for binary non-ionic aqueous organics, built from Flory-type entropy with volume fractions, using only pure-component melting temperature, enthalpy of fusion, and molar volume. The derivation is clean, the model is validated on ten external datasets, and the reported error reductions over Raoult's law (59% on liquidus residual, 45% on eutectic temperature, 43% on eutectic composition) are real improvements for a class of systems that usually needs solution-specific parameters. The citation pattern is fair—they credit Flory/Huggins, Hildebrand, Hansen, Coutinho, and Martin—and the SI sensitivity checks on molar volume assumptions are the right kind of diligence.\n\nThe soft spot is the physical interpretation. The authors argue that much of the deviation from ideality in these solutions is an entropic size effect rather than molecular interaction. That is plausible but not tested. Because the Flory term always lowers the predicted liquidus whenever the volume ratio differs from 1, and because the chosen experimental liquidi sit mostly below the classic ideal curve, part of the improved fit could be a mechanical offset rather than the correct physics. The paper itself concedes that sucrose retains unexplained deviations near the eutectic. The clean, falsifiable check would be to compare Eqn (4)'s predicted water activity against measured water activities for these systems; the authors don't do it. That gap should be acknowledged and ideally filled before selling the 'entropic dominance' claim.\n\nThe validation set also has a couple of softer spots: two systems use metastable eutectics and two use interpolated eutectic compositions, so the eutectic error statistics are slightly less solid than they look. These are secondary, though. The zero-parameter nature of the model is a genuine strength, and the practical claim—that Eqn (9) is a simple, repeatable, better-than-Raoult predictor for a useful class of aqueous organic solutions—holds up.\n\nWho's this for? Cryopreservation researchers, NADES people, and formulation chemists who need a quick a priori estimate of liquidus and eutectic behavior. It deserves a serious referee, not a desk reject. I'd send it to peer review and ask the referee to require the water-activity comparison and a discussion of the one-directional bias. With that, it's publishable; without it, the predictive model is still worth reporting, but the causal claim should be framed as a hypothesis.","headline":"Useful zero-parameter liquidus model for aqueous organics; the 'entropy dominates' interpretation needs an activity check before it's sold as physics.","tokens_in":10167,"tokens_out":2709,"would_cite":true,"duration_ms":26524,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that a volume-fraction entropy of mixing turns the ideal solution model into a zero-parameter predictor for aqueous organic liquid-solid phase diagrams, cutting average liquidus error by 59% across ten binaries.","keywords":["ideal solution model","Flory entropy of mixing","volume fraction","liquidus temperature prediction","eutectic prediction","aqueous organic solutions","mixing entropy","phase diagram prediction"],"falsifier":"Measure the enthalpy of mixing of aqueous urea or glycerol across the full composition range; if $\\Delta H_{\\mathrm{mix}}$ near the eutectic is comparable to the size-entropy term $RT[\\ln\\phi_A + (1-\\phi_A)(1-1/V_{R,A})]$, then the claim that entropic size effects dominate would be falsified. A second check would be to apply Eqn (9) to a strongly hydrogen-bonded aqueous system with a large molar-volume ratio and see whether systematic failure tracks the neglected excess enthalpy.","tokens_in":9083,"feed_emoji":"❄️","tokens_out":10163,"duration_ms":82326,"temperature":0.7,"pith_summary":"This paper claims that much of the apparent non-ideality of aqueous organic solutions is not caused by molecular interactions but by the size difference between small water molecules and larger organic solutes acting on the entropy of mixing. Replacing the mole-fraction entropy of mixing in the classic ideal solution model with a Flory-style volume-fraction entropy yields a size-dependent chemical potential and a closed-form liquidus equation that require only each pure component's molar volume, enthalpy of fusion, and melting temperature. Tested on ten water-organic binaries, the model lowers average liquidus, eutectic-temperature, and eutectic-composition errors by 59%, 45%, and 43% relative to the classic ideal model. If correct, it provides a zero-parameter way to predict simple-eutectic aqueous organic phase diagrams, which would accelerate screening of cryoprotectants and natural deep eutectic solvents.","feed_headline":"Size effects trim freezing-curve error 59% in aqueous organics","feed_subtitle":"A Flory-style entropy term, using volume fractions, predicts water-organic phase diagrams with no solution data.","key_machinery":"The machinery is Eqn (1), the Flory entropy of mixing $\\Delta S_{\\mathrm{mix}} = -R\\sum_i x_i \\ln\\phi_i$, with volume fractions $\\phi_i = x_i v_i / \\sum_j x_j v_j$. Substituting this into the ideal-solution Gibbs energy and differentiating gives the chemical potential $\\mu_A = G_A + RT[\\ln\\phi_A + (1-\\phi_A)(1-1/V_{R,A})]$, where $V_{R,A}=v_B/v_A$. A Gibbs-Helmholtz integration then yields Eqn (9), the liquidus temperature as a function of volume fraction: $T_{\\mathrm{liq},A} = \\left(1/T_{m,A} - R[\\ln(1/\\phi_A)-(1-\\phi_A)(1-1/V_{R,A})]/\\Delta_{\\mathrm{fus}}H_A^0\\right)^{-1}$. This equation is the zero-parameter predictor; it reduces to the classical ideal liquidus when $V_{R,A}=1$, and it is evaluated for each component with the stable (higher) liquidus retained to construct the phase diagram.","core_discovery":"The central discovery is that incorporating the Flory entropy of mixing, $\\Delta S_{\\mathrm{mix}} = -R\\sum_i x_i \\ln \\phi_i$, where $\\phi_i$ is the volume fraction, into the ideal-solution Gibbs energy produces a chemical potential and a liquidus temperature that depend on the molar-volume ratio of the two components. The resulting liquidus equation, Eqn (9), predicts binary phase diagrams for water with each of ten non-ionic organic solutes using only pure-component melting temperature, enthalpy of fusion, and molar volume, and it outperforms the classical ideal model in every case studied. Because the model deliberately keeps the ideal assumption of zero enthalpy of mixing, the paper reads the systematic improvement as evidence that much of the deviation typically attributed to hydrogen bonding in these solutions is actually an entropic size effect. The paper also notes residual deviations for heavily interacting solutes such as sucrose, which it attributes to additional interaction factors beyond the size-entropy term.","pith_inferences":["A direct test of the zero-enthalpy premise would be to measure excess enthalpies for urea-water or glycerol-water; if these approach the size-entropy term near the eutectic, the entropic explanation would be weakened.","The model's structure suggests that activity coefficients in non-ionic solutions could be decomposed into a volume-fraction reference term and a residual enthalpic term, isolating interactions in a way the mole-fraction basis cannot.","If the entropic dominance is real, cryoprotectant toxicity and permeability screens may correlate more strongly with solute molar volume ratios than with chemical interaction strength, a prediction that existing toxicity datasets could test.","Temperature-dependent molar volumes could extend the model's accuracy near the eutectic, where the constant-enthalpy-of-fusion assumption is known to weaken."],"forward_implications":["For any non-ionic aqueous organic system with a simple eutectic, liquidus curves and eutectic points can be estimated from pure-component properties alone, with no solution-specific measurements.","Across the ten systems tested, the size-dependent model cuts the average liquidus residual by 59%, the eutectic-temperature error by 45%, and the eutectic-composition error by 43% relative to the classical ideal model.","The classic attribution of depressed liquidus curves in aqueous organics to hydrogen bonding is called into question; a substantial part of that depression may be a size-entropy effect.","Because the derivation preserves the ideal model's zero-parameter structure, it can be dropped into derivative formulations, suggesting routes to size-dependent regular-solution and activity models.","The model's accuracy holds across disparate molar-volume ratios, meaning it is equally applicable to large solutes in water and to small solutes in large solvents."],"supporting_citations":[{"why":"Supplies the volume-fraction entropy of mixing used in Eqn (1).","marker":"(13)"},{"why":"Complementary polymer-theory derivation of the same size-dependent mixing entropy.","marker":"(14)"},{"why":"Recent application of size-dependent entropy to aqueous non-ionic solutions; supports the excess-volume assumption.","marker":"(15)"},{"why":"Historical source documenting size effects in non-polymer solutions and the lineage of the chemical potential form.","marker":"(8)"},{"why":"Prior use of similar chemical potential forms in alkane solid-liquid equilibria; supports the constant-enthalpy-of-fusion assumption.","marker":"(17)"},{"why":"Thermodynamic model for wax formation that arrived at a similar size-dependent chemical potential.","marker":"(18)"},{"why":"Recent account attributing aqueous eutectic depressions to hydrogen bonding; provides the counterpoint the paper argues against and the elevated-eutectic caveat.","marker":"(12)"},{"why":"Experimental glycerol-water phase diagram used as a validation dataset.","marker":"(29)"},{"why":"Experimental sucrose-water phase diagram used as the clearest residual-deviation case.","marker":"(33)"},{"why":"Experimental urea-water equilibria used as the small-solute validation dataset.","marker":"(25)"}],"fun_headline_variants":["Size entropy cuts liquidus error 59% in water-organic systems","Volume-fraction entropy predicts phase diagrams with pure data only","Flory entropy term explains organic aqueous deviations","Simple size term halves freezing-point prediction error","Entropic size effects dominate aqueous organic solution ideality"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model assumes the enthalpy of mixing is exactly zero and that all non-ideality is captured by the volume-fraction entropy term; the paper's own sucrose results show this fails for strongly interacting solutes, so the improvement would be at least partly coincidental if enthalpic interactions materially control the liquidus.","fun_headline_variants_meta":{"raw":{"variants":["Size entropy cuts liquidus error 59% in water-organic systems","Volume-fraction entropy predicts phase diagrams with pure data only","Flory entropy term explains organic aqueous deviations","Simple size term halves freezing-point prediction error","Entropic size effects dominate aqueous organic solution ideality"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000652,"raw_usage":{"total_tokens":3030,"prompt_tokens":1030,"completion_tokens":2000,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":646,"completion_tokens_details":{"reasoning_tokens":1923}},"tokens_in":646,"tokens_out":2000,"duration_ms":13631,"temperature":1.0,"reasoning_tokens":1923,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:40:11.494825+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the enthalpy of mixing of aqueous urea or glycerol across the full composition range; if $\\Delta H_{\\mathrm{mix}}$ near the eutectic is comparable to the size-entropy term $RT[\\ln\\phi_A + (1-\\phi_A)(1-1/V_{R,A})]$, then the claim that entropic size effects dominate would be falsified. A second check would be to apply Eqn (9) to a strongly hydrogen-bonded aqueous system with a large molar-volume ratio and see whether systematic failure tracks the neglected excess enthalpy.","supporting_citations":[],"review_version":1}