{"id":"680c7b90-acc3-46c3-aa7a-24d1492589be","arxiv_id":"2505.01055","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Ice grown in DMSO or proline solutions at atmospheric pressure undergoes a kinetic roughening transition at -16.0 +/- 0.2 °C, yielding hexagonal facets without ice-active molecules and showing different growth versus melting kinetics.","lead":"Ice crystals in DMSO and proline solutions switch from round disks to faceted hexagons at about -16 °C, the same temperature seen earlier only under high pressure. This means faceted ice shapes below that temperature can form without antifreeze proteins, so morphology alone is not proof of ice-binding activity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reported TR may be an orientation artifact: the crystal physically rotates in the -15.6 to -18.2 °C window where the sigmoid is fitted, so projected roundness need not track prism-face roughening.","rationale":"The strongest claim survives only if R(T) measures prism-face roughening. The paper's own report of rotation in the transition window makes this condition insecure; the high-pressure -16 °C value gives plausibility but cannot rule out a projection artifact in this particular setup because rotation is observed only in the atmospheric-pressure experiment. I am not rejecting the claim: the visual evidence, the two solutes, and the agreement with Maruyama's high-pressure value are real. But a single orientation-gated reanalysis of Video SI1 would resolve whether the sigmoid is an equilibrium roughening transition or a geometric alignment event. The complementary near-equilibrium issue, with undercooling of about 5 °C at the low-temperature end, reinforces the need for such a check; however, it is secondary to the orientation confound. The reader's conditional verdict is appropriate.","tokens_in":16386,"tokens_out":11158,"duration_ms":126116,"concrete_test":"Using the supplied time-lapse (Video SI1) and the MATLAB contour output, gate the roundness analysis on crystal orientation: estimate the out-of-plane tilt of the basal outline from the projected major/minor axis ratio (or from a z-focal series), retain only frames where the basal normal is within ±5° of the optical axis, and refit Eq. 2 on the gated subset. If the inflection shifts by more than 1 °C or the sigmoidal step disappears, the reported TR is an orientation artifact rather than a roughening transition.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Central claim rests on R(T) in Fig. 4 (and Eq. 2) being a faithful order parameter for prism-face roughening, which requires the crystal's basal-plane orientation relative to the optical axis to stay fixed. That condition is unverified, and the paper itself reports a physical rotation of the crystal between -15.6 and -18.2 °C (Sec. 3.1, Fig. 3b-c), exactly the range where the sigmoid is fitted. A hexagonal plate viewed with any out-of-plane tilt projects as a rounded or elliptical outline; as it rotates toward a basal-on view, the projected perimeter becomes more hexagonal and R decreases from the disk baseline to the hexagon value by projection alone. The fitted inflection at -16.0 ± 0.2 °C could therefore locate the rotation event, not the onset of prism-face step nucleation. The near-equilibrium premise is also strained: using the paper's own area fractions (12%, 21%, 33%) and Semenov melting-point estimates, the undercooling at -18.2 °C is about 5 °C (Tm ≈ -13 °C), far from the near-equilibrium condition the abstract invokes. A morphology change under 5 °C undercooling is a kinetic growth-regime effect unless explicitly separated from the equilibrium roughening transition.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a kinetic roughening transition of ice crystals grown from DMSO–water and proline–water solutions at atmospheric pressure, with a transition temperature TR = -16.0 ± 0.2 °C inferred from a sigmoidal fit to the roundness of single crystals during slow cooling. The authors propose that this transition is independent of solute identity and matches the value previously measured under high pressure, implying that faceted ice morphologies below TR can arise without ice-active molecules. They also report an asymmetry between growth (sigmoidal roundness evolution) and melting (exponential relaxation) and show that antifreeze protein type III promotes faceting above TR. The paper includes cryomicroscopy videos, a MATLAB analysis script, and comparisons to literature benchmarks.","tokens_in":16683,"tokens_out":7630,"duration_ms":76997,"significance":"If the central claim holds, the paper provides a useful benchmark for distinguishing intrinsic kinetic roughening from adsorption-mediated ice shaping, with direct implications for interpreting ice morphology in cryobiology and atmospheric science. The study is well motivated, uses two solutes to probe solute independence, and ships reproducible analysis code and supporting videos. The growth–melt asymmetry, if confirmed, is a novel observation that could stimulate further work on interfacial kinetics. However, the headline TR value rests on a small number of single-crystal fits, and the near-equilibrium premise is not quantitatively established by the paper's own data.","major_comments":[{"comment":"","section":"Section 3.1, Fig. 3 and Fig. 4"},{"comment":"","section":"Section 3.1 and SI 'Estimation of concentration...'"},{"comment":"","section":"Section 3.1, Table 2 vs. SI Fig. SI2"},{"comment":"","section":"Section 3.1, Eq. 2"},{"comment":"","section":"Section 3.2, Figs. 5 and 7"}],"minor_comments":[{"comment":"","section":"Page 20, figure captions"},{"comment":"","section":"Section 3.1, text near Fig. 3"},{"comment":"","section":"Section 2.1"},{"comment":"","section":"Eq. 2 and Table 2"},{"comment":"","section":"SI, MATLAB code and thresholding"},{"comment":"","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses a topic of broad interest and the experimental design is thoughtful, but the headline value of TR = -16.0 ± 0.2 °C is currently supported by only three single-crystal fits, with an unresolved internal inconsistency between the main text and the SI, and with an orientation-rotation artifact that is not ruled out. The near-equilibrium premise is also quantitatively strained by the paper's own undercooling estimates. These issues are fixable within the scope of a revision: the authors could add more crystals, perform an orientation check, reconcile the TR values, and either measure the undercooling or temper the near-equilibrium claim. I recommend major revision rather than rejection, as the underlying observation of a temperature-driven morphological transition near -16 °C appears real and valuable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Core observation is real: ice grown from DMSO or proline solutions switches from rounded disks to hexagonal plates near -16 °C, matching the roughening transition seen under high pressure. That atmospheric confirmation, with two non-adsorbing solutes, is a genuine addition, and the growth-sigmoid vs melt-exponential asymmetry is a nice observation with practical implications for antifreeze-protein assays. The qualitative morphology change is visually solid: at -5 °C you see disks, at -19 °C you see hexagons, for two solutes and several concentrations.\n\nThe soft spots are in the quantitative claim, not the qualitative one. The reported TR = -16.0 ± 0.2 °C comes from fits to single crystals, n=3 total across three conditions. That by itself would be minor if the fits were clean. The bigger issue is that the featured crystal in Figure 3 physically rotates between -15.6 and -18.2 °C — exactly the window where the sigmoid is fitted. Roundness of a projected outline changes with tilt, so unless the crystal's orientation relative to the optical axis is verified, part of the roundness decrease could be projection, not roughening. The paper notes the rotation but doesn't correct for it. That's a real crack in the precision claim.\n\nThe near-equilibrium premise is also strained. Using the paper's own numbers, at -18.2 °C the local DMSO concentration has risen to ~27%, giving Tm ≈ -13 °C, so the undercooling is about 5 °C, not near-equilibrium. A morphological change under that driving force could be a kinetic growth-regime effect rather than a clean roughening transition. This doesn't kill the paper — the faceting at low temperature still implies the prism face is below its roughening temperature — but it does undercut the sharp 0.2 °C error bar and the 'universality' language.\n\nThe data handling is otherwise honest: MATLAB code is supplied, the fitting is simple, the high-pressure benchmarks are properly cited. No sign of fitting to noise or invented entities.\n\nBottom line: this is a useful paper for cryobiologists, ice physicists, and anyone who interprets faceted ice morphology as evidence of ice-binding activity. It should go to peer review, not be desk-rejected. The referees should push on orientation tracking and on measuring (or at least bounding) the undercooling during the growth trace. If the authors can separate projection effects from true roughening, the paper becomes solid.","headline":"Atmospheric confirmation of the -16 °C prism-face roughening transition is real and useful, but the sub-degree TR precision and the growth-melt asymmetry rest on single-crystal fits with unaddressed orientation and undercooling issues.","tokens_in":17178,"tokens_out":4940,"would_cite":true,"duration_ms":52463,"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":"Ice crystals grown near equilibrium in DMSO or proline solutions switch from rounded disks to faceted hexagons at $-16.0 \\pm 0.2\\,^{\\circ}\\mathrm{C}$, a kinetic roughening transition that is independent of the solute.","keywords":["kinetic roughening transition","ice crystal morphology","ice recrystallization","antifreeze proteins","cryomicroscopy","DMSO","proline","growth-melt asymmetry"],"falsifier":"If a single ice crystal in a DMSO or proline droplet kept below $-16\\,^{\\circ}\\mathrm{C}$ with its basal face toward the observer continued to grow as a circular disk, or if direct step-imaging of the prism face showed no change in layer-nucleation behavior at $-16\\,^{\\circ}\\mathrm{C}$, the identification of $T_R$ as a kinetic roughening transition would be contradicted.","tokens_in":16185,"feed_emoji":"❄️","tokens_out":7333,"duration_ms":64760,"temperature":0.7,"pith_summary":"This paper establishes that the shape of ice crystals growing slowly from aqueous solution is governed by a kinetic roughening transition at a sharply defined temperature. At temperatures above $-16.0 \\pm 0.2\\,^{\\circ}\\mathrm{C}$ the prism face of ice grows in a rough, rounded-disk morphology; below that temperature the same face grows by layer propagation and the crystal develops flat hexagonal facets. The transition temperature is the same in dimethyl sulfoxide and proline solutions and matches the value previously measured only under high pressure, showing that it is set by temperature alone rather than by the solute. The authors then show that growth and melting follow different kinetic laws across this transition, and that a low concentration of antifreeze protein type III can induce faceting even above the transition. The practical point is that a faceted, hexagonal ice crystal is not by itself evidence that an ice-binding molecule is present.","feed_headline":"Ice crystals turn hexagonal at -16 °C, no protein needed","feed_subtitle":"A kinetic roughening transition, independent of solute, separates rounded disks from faceted plates in solution-grown ice.","key_machinery":"The central object is the roundness function $R = 4\\pi A/P^2$, computed in real time from microscopy images of single ice crystals, and its dependence on temperature. The paper's mechanism is the kinetic roughening transition: above $T_R$ the prism face can nucleate new molecular layers spontaneously, so the interface is rough and the projected outline is circular; below $T_R$ layer growth requires step nucleation, so the crystal exposes flat prism facets and appears hexagonal when viewed along the basal axis. The measurement is forced into the slow-growth, near-equilibrium regime by growing crystals in microdroplets, where the excluded solute progressively depresses the melting point. A sigmoidal fit (Eq. 2) to the roundness-versus-temperature curve supplies the inflection point identified as $T_R$, and the exponential fits (Eqs. 3) supply the melting relaxation.","core_discovery":"Using cryomicroscopy on single ice crystals in emulsified microdroplets of DMSO- and proline–water solutions cooled at 0.1 °C/min, the authors measured the roundness $R = 4\\pi A/P^2$ of the crystal outline as a function of temperature. A sigmoidal fit to the roundness curve has its inflection point at $T_R = -16.0 \\pm 0.2\\,^{\\circ}\\mathrm{C}$, where the growth morphology switches from rounded disks to hexagonal plates; the 10–90% transition spans $-18.8$ to $-13.2\\,^{\\circ}\\mathrm{C}$. The transition appears at the same temperature in both solutes and with or without a cover glass. During melting the same crystals lose their facets exponentially (time constant $\\tau \\approx 47$ s for isothermal melting, $\\sigma_m \\approx 3.5\\,^{\\circ}\\mathrm{C}$ for gradual warming), so growth and melting are kinetically asymmetric. Adding 0.5 µM antifreeze protein type III produces hexagonally faceted growth even at temperatures above $T_R$, indicating that adsorption raises the effective roughening temperature.","pith_inferences":["If the transition is set purely by temperature, the same $T_R \\approx -16\\,^{\\circ}\\mathrm{C}$ should appear in other non-interacting cryoprotectants such as glycerol, ethylene glycol, or sugars; a survey measuring roundness hysteresis in those systems would test the universality claim.","The crystal rotation observed between $-15.6$ and $-18.2\\,^{\\circ}\\mathrm{C}$ hints that the torque arises from step propagation on the faceting prism face; tracking orientation versus temperature could give a mechanical measure of the step dynamics.","The growth–melt asymmetry may extend beyond ice: any crystal whose rough face crosses a kinetic roughening transition should show sigmoidal faceting on growth but exponential rounding on melting, which could be checked in organic crystals grown from solution.","The 10–90% transition band from $-18.8$ to $-13.2\\,^{\\circ}\\mathrm{C}$ suggests a practical guideline: morphological assays of ice activity conducted above $-13\\,^{\\circ}\\mathrm{C}$ are unlikely to be confounded by intrinsic roughening, while those below $-18\\,^{\\circ}\\mathrm{C}$ almost certainly are."],"forward_implications":["Faceted hexagonal ice at temperatures below about $-13\\,^{\\circ}\\mathrm{C}$ can form in solutions of solutes that do not bind ice, so faceting alone does not prove ice-binding activity.","Recrystallization studies conducted below $-16\\,^{\\circ}\\mathrm{C}$ will see intrinsic faceting during growth and rounding during melting, and should not attribute the hexagonal shapes to ice-active agents.","Growth and melting across the roughening transition follow different kinetic forms: sigmoidal for growth and exponential for melting, which explains the previously reported growth–melt asymmetry in ice crystals.","Antifreeze proteins such as AFPIII can elevate the effective roughening temperature, meaning ice-shaping activity and recrystallization-inhibition activity share a common threshold."],"supporting_citations":[{"why":"First optical study of the roughening transition on the prism plane of ice under pressure; supplies the high-pressure $T_R$ the present work reproduces at atmospheric pressure.","marker":"[10]"},{"why":"Confirms the prism-face roughening temperature for ice grown from the melt under pressure; baseline that the atmospheric-pressure value is compared against.","marker":"[11]"},{"why":"Documents growth–melt asymmetry and twelve-sided snowflake shapes; the melting-rounding behavior the paper compares to its exponential relaxation.","marker":"[12]"},{"why":"Observed faceted hexagonal ice crystals in concentrated fructose solutions at low temperature; motivated the hypothesis that kinetic roughening, not ice-active solutes, causes faceting.","marker":"[6]"},{"why":"Shows antifreeze proteins modify both growth and melting shapes of ice; reference for AFP-induced faceting and for the growth–melt asymmetry.","marker":"[16]"},{"why":"Measured growth–melt asymmetry in ice crystals under antifreeze protein influence; supplies the 'apparent rotation during melting' explanation.","marker":"[37]"},{"why":"Provides the phenomenological melting-point–concentration relation used to estimate how much the melting point drops as the crystal grows inside the droplet.","marker":"[46]"},{"why":"Quantitative classification of ice recrystallization inhibition agents; used to state that 0.5 µM AFPIII corresponds to roughly 50% IRI.","marker":"[59]"},{"why":"Review of the roughening transition that supplies the theoretical framework of step nucleation and spontaneous roughening.","marker":"[9]"}],"fun_headline_variants":["Ice roughening transition at -16 °C flips disks to plates","No proteins required: ice facets at -16 °C transition","Kinetic roughening sets ice crystal shape at -16 °C","Ice growth switches to hexagonal plates below -16 °C","Intrinsic ice transition at -16 °C: disks become facets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the measured roundness of the projected crystal outline tracks the kinetic roughening of the prism face itself, rather than a change in crystal orientation or in the growth regime, so that the inferred $T_R$ is a true surface transition.","fun_headline_variants_meta":{"raw":{"variants":["Ice roughening transition at -16 °C flips disks to plates","No proteins required: ice facets at -16 °C transition","Kinetic roughening sets ice crystal shape at -16 °C","Ice growth switches to hexagonal plates below -16 °C","Intrinsic ice transition at -16 °C: disks become facets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000364,"raw_usage":{"total_tokens":2031,"prompt_tokens":1086,"completion_tokens":945,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":702,"completion_tokens_details":{"reasoning_tokens":852}},"tokens_in":702,"tokens_out":945,"duration_ms":7746,"temperature":1.0,"reasoning_tokens":852,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:28:47.692926+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"If a single ice crystal in a DMSO or proline droplet kept below $-16\\,^{\\circ}\\mathrm{C}$ with its basal face toward the observer continued to grow as a circular disk, or if direct step-imaging of the prism face showed no change in layer-nucleation behavior at $-16\\,^{\\circ}\\mathrm{C}$, the identification of $T_R$ as a kinetic roughening transition would be contradicted.","supporting_citations":[{"cited_title":"& Sawada, T","cited_arxiv_id":null,"evidence_quote":"First optical study of the roughening transition on the prism plane of ice under pressure; supplies the high-pressure $T_R$ the present work reproduces at atmospheric pressure."},{"cited_title":"Roughening transition of prism faces of ice crystals grown from melt under pressure","cited_arxiv_id":null,"evidence_quote":"Confirms the prism-face roughening temperature for ice grown from the melt under pressure; baseline that the atmospheric-pressure value is compared against."},{"cited_title":"& Wettlaufer, J","cited_arxiv_id":null,"evidence_quote":"Documents growth–melt asymmetry and twelve-sided snowflake shapes; the melting-rounding behavior the paper compares to its exponential relaxation."},{"cited_title":"Ripening of faceted ice crystals","cited_arxiv_id":null,"evidence_quote":"Observed faceted hexagonal ice crystals in concentrated fructose solutions at low temperature; motivated the hypothesis that kinetic roughening, not ice-active solutes, causes faceting."},{"cited_title":"S., Davies, P","cited_arxiv_id":null,"evidence_quote":"Shows antifreeze proteins modify both growth and melting shapes of ice; reference for AFP-induced faceting and for the growth–melt asymmetry."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Measured growth–melt asymmetry in ice crystals under antifreeze protein influence; supplies the 'apparent rotation during melting' explanation."},{"cited_title":"P., Mendgaziev, R","cited_arxiv_id":null,"evidence_quote":"Provides the phenomenological melting-point–concentration relation used to estimate how much the melting point drops as the crystal grows inside the droplet."},{"cited_title":"staircased","cited_arxiv_id":null,"evidence_quote":"Quantitative classification of ice recrystallization inhibition agents; used to state that 0.5 µM AFPIII corresponds to roughly 50% IRI."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Review of the roughening transition that supplies the theoretical framework of step nucleation and spontaneous roughening."}],"review_version":1}