{"id":"7e855bab-2f0d-4105-846a-1575a399f318","arxiv_id":"2411.17531","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A monodisperse granular hard-disk system can show coexistence of a hotter solid and a colder liquid when energy is supplied by a thermal bath, an effect explained by the lower collision frequency of the solid.","lead":"Simulations of dissipative hard disks show that a solid phase can be hotter than the coexisting liquid, overturning the common rule that denser granular phases are always colder. The effect arises because the solid phase can have a lower collision frequency, and a kinetic theory explains the temperature ordering for two different driving mechanisms.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The hot-solid criterion for the GLM depends on the Enskog collision-frequency relation holding inside the solid, which the paper itself flags as questionable in Sec. IV.","rationale":"I read the paper as making two claims: a direct simulation observation that in the GLM the coexisting solid has a higher granular temperature than the liquid, and a kinetic-theory explanation/criterion (Λ>1) for this ordering. The simulation observation is a clean, parameter-free event-driven result; I do not see an internal contradiction there, though the absence of error bars means the 0.1–0.3% effect should be checked. The load-bearing theoretical step is the elimination of ϕg+ via G(T) in Eq. (13), which requires Eq. (8) to hold separately in both phases, including the dense/hexatic/crystalline solid. The paper itself acknowledges in Sec. IV that molecular chaos and Gaussian velocity distributions 'can easily break down... particularly in the solid phase,' so this is a live risk, not a manufactured one. The Fig. 2 comparison using measured g+ is good supporting evidence but does not independently verify Eq. (8) in the coexisting solid; a direct measurement of ω versus the Enskog expression, and of the resulting temperature ordering, would settle it. This is the same weakest assumption identified by the reader, so the verdict stays conditional with no change.","tokens_in":19679,"tokens_out":7695,"duration_ms":78867,"concrete_test":"In the GLM coexistence run, measure the actual collision frequency ω directly in bulk slabs of each coexisting phase and compare it, phase by phase, with the Enskog value 8ϕg+√(T/πm)/σ using the locally measured g+ and T. Then recompute the phase-ordering criterion using the measured ω in the temperature balance instead of Eq. (8): if the equality p_s=p_l still yields T_s>T_l for Λ>1, the concern is retired; if using measured ω reverses the ordering or makes it indeterminate, the theoretical criterion in Eq. (21) needs revision. Report block error bars on T_s−T_l and on ω_s−ω_l so the 0.1–0.3% effect is statistically resolvable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The route to the Λ>1 prediction is: Eq. (8) sets ω=8ϕg+√(T/πm)/σ; inserting this into the steady-state balance Eq. (11) gives ϕg+=G(T) in Eq. (13); eliminating ϕg+ from the pressure yields p̃(T) in Eq. (19); mechanical equilibrium then gives T_s>T_l whenever p̃ decreases. Every link in this chain uses molecular chaos and a Gaussian velocity distribution inside each phase, and the solid is precisely where those assumptions are least secure—the authors concede in Sec. IV that they 'can easily break down... particularly in the solid phase.' The comparison in Fig. 2 tests Eq. (11) with measured g+ but does not separately validate Eq. (8) in the coexisting solid, and the predicted phase-temperature difference is only 0.1–0.3%, so a breakdown of Eq. (8) in the solid could change the ordering predicted by the theory. The direct simulation observation of a hotter solid would survive such a failure, but the paper's parameter-free criterion—and the claimed extension of kinetic theory to the solid—would not.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports event-driven molecular dynamics simulations of two driven granular hard-disk models, the Delta+gamma model and the granular Langevin model (GLM), in the liquid-solid coexistence region. For the Delta+gamma model the coexisting solid is colder than the liquid, whereas for the GLM, in which collisions are purely dissipative, the authors observe the opposite: a hotter solid coexisting with a colder liquid. The hotter-solid effect is attributed to a lower collision frequency in the solid relative to the liquid at coexistence. The paper develops a kinetic theory based on molecular chaos, Gaussian velocity distributions, and the Enskog collision-frequency relation, and derives a criterion, Eq. (17), that determines the relative phase temperatures from the function G(T) without fitting parameters. For the GLM the theory predicts a hotter solid whenever the dimensionless parameter Lambda > 1, and a colder solid in the Delta+gamma model. The theoretical temperatures are compared with simulation in Fig. 2 using measured values of g+.","tokens_in":4,"tokens_out":9638,"duration_ms":160896,"significance":"If the result holds, the paper provides a direct counterexample to the common expectation that the denser coexisting phase must be the colder one in dissipative granular systems. The numerical observation is supported by Mayer-Wood loops, local density and temperature profiles, and averages over many realizations, which is a genuine strength. The kinetic-theory route is elegant: no parameters are fitted, and the final criterion is expressed in terms of a single dimensionless parameter. The claim that kinetic theory can be extended to the solid phase, and the associated prediction Lambda > 1 for a hotter solid, are conceptually novel and would be useful for future experiments. The main weakness is that the theoretical criterion relies on the Enskog collision-frequency relation inside the solid, an assumption that the authors themselves flag as fragile; the manuscript needs to either validate this relation in the coexisting solid or clearly restate the theoretical claim as conditional on it.","major_comments":[{"comment":"The parameter-free prediction for the GLM rests on substituting Eq. (8), omega = 8 phi g+ sqrt(T/pi m)/sigma, into the steady-state balance Eq. (11) to obtain G(T) in Eq. (13). If Eq. (8) fails inside the solid, then Eq. (19) for p-tilde and the conclusion T_s > T_l for Lambda > 1 do not follow. Figure 2 tests Eq. (11) combined with Eq. (8) using measured g+ in homogeneous systems, but it does not separately verify Eq. (8) in the coexisting solid slab of Fig. 1(f). Since the observed temperature difference is only 0.1-0.3%, and since Sec. IV concedes that molecular chaos and Gaussian velocity distributions 'can easily break down... particularly in the solid phase,' I ask for a direct test of Eq. (8) inside the coexisting solid, for example by measuring the collision frequency and g+ in the same solid region, or, failing that, for a clear statement that the Lambda > 1 criterion is a theoretical conjecture rather than a fully validated prediction.","section":"Sec. III.B-III.C, Eqs. (8), (13), (19)"},{"comment":"The proof that p-tilde is decreasing for all physical T < T_b when Lambda > 1 relies on the assertion that the root T* of Eq. (E4) is monotonically increasing in Lambda. The manuscript provides only an asymptotic expansion around Lambda = 1, Eq. (E6), and a representative figure. Because the conclusion T_s > T_l for all Lambda > 1 depends on this monotonicity, a short rigorous argument or a numerical plot of T*(Lambda) over the full range Lambda > 1 should be supplied.","section":"Appendix E, Eq. (E3)-Eq. (E7)"}],"minor_comments":[{"comment":"Equation (15) states p = phi p-tilde, while Eq. (16) defines p-tilde = sigma^2 pi p / (4 phi). These two relations are inconsistent by the constant factor 4/(pi sigma^2); the factor cancels in the inequality Eq. (17), but the equations should be corrected for clarity.","section":"Eqs. (15)-(16)"},{"comment":"The error term in the expansion of T* around Lambda = 1 is written as O((Lambda - 2)^3); it should be O((Lambda - 1)^3).","section":"Eq. (E6)"},{"comment":"The temperature difference between the two phases is only about 0.1-0.3%, but the profiles in panels (c) and (f) are shown without error bars. Given that the central qualitative claim rests on the sign of this small difference, statistical uncertainties should be reported, at least for the GLM profiles.","section":"Fig. 1"},{"comment":"The labels 'Energy' in panels (b) and (e) are used interchangeably with granular temperature; please use one consistent term throughout the figures and text.","section":"Fig. 1 caption and Sec. II.B"},{"comment":"The definition of the reference temperature T_o appears only in the caption of Fig. 1; since it is used to normalize the Delta+gamma data, it would be clearer to define it in the main text.","section":"Sec. II.A"}],"recommendation":"major_revision","confidential_remarks":"The direct simulation observation of a hotter solid in the GLM appears solid and would be of interest to the granular and active-matter communities. My main reservation is the strength of the theoretical claim: the Lambda > 1 criterion is presented as a parameter-free prediction, but its derivation uses the Enskog relation deep in the solid phase, and the validation in Fig. 2 does not isolate that assumption in the coexisting solid. This is fixable with additional analysis or a scoped claim, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the central observation is real and worth publishing; the parameter-free criterion is a nice explanation but leans on Enskog assumptions in the solid that the authors themselves concede are delicate.\n\nWhat's actually new: a hotter solid than coexisting liquid in a monodisperse granular hard-disk system. Earlier evidence came from a bidisperse mixture where geometric effects could contaminate the result, so the monodisperse demonstration closes a real gap. The event-driven MD shows a Mayer-Wood loop and local density/temperature profiles that directly support the temperature ordering. The kinetic theory is elegant: they eliminate g+ via the steady-state energy balance and get a parameter-free hot-solid condition Λ > 1. The comparison in Fig. 2 shows the theory tracks the measured temperature well when measured g+ is used.\n\nWhere I'd push back: the effect is 0.1–0.3% and there are no error bars anywhere. The pressure loops and profiles look consistent, but with such a small signal I'd want statistical uncertainties and independent runs. The stress-test concern is fair: the derivation of G(T) and p̃ relies on the Enskog collision-frequency relation holding inside the crystalline/hexatic phase, and that is exactly where molecular chaos and Gaussian velocity assumptions are least secure. The paper acknowledges this in Sec. IV but does not separately validate Eq. (8) in the coexisting solid. That does not undermine the simulation observation, which stands alone; it means the \"parameter-free prediction\" is not as airtight as the text sometimes suggests. Also, no code or data is deposited; \"available upon request\" is weak.\n\nThe citation pattern looks fine, and the paper is honest about the limitations of its model relative to realistic vibrated granular matter.\n\nThis paper is for the non-equilibrium coexistence community, and it deserves a serious referee. I would send it out. The revisions should ask for error estimates, a sensitivity check on solid-phase Enskog assumptions, and ideally deposited data/code.","headline":"A hot solid coexisting with a cold liquid in a monodisperse granular hard-disk model is a real, simulation-supported result; the parameter-free theory is elegant but rests on Enskog assumptions that are least secure in the solid.","tokens_in":20403,"tokens_out":3530,"would_cite":true,"duration_ms":46633,"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":"A denser granular crystal can be hotter than its coexisting liquid in a model of dissipative hard disks driven by a Langevin bath, because the crystal's collision frequency is lower.","keywords":["granular matter","dissipative hard disks","non-equilibrium phase coexistence","granular temperature","kinetic theory","Enskog collision frequency","liquid-solid transition","Langevin bath"],"falsifier":"Run event-driven GLM simulations at coexistence (for instance $\\alpha=0.99$, $T_b/[m(\\sigma\\gamma)^2]=0.125$) and count collisions per particle separately in the solid and liquid domains. If the hotter solid does not show a lower collision frequency than the colder liquid, the mechanism and the $\\phi g^+$ elimination fail; equally, measuring local pressures and temperatures should either satisfy or violate $\\tilde p(T_s)<\\tilde p(T_l)$.","tokens_in":19464,"feed_emoji":"🌡️","tokens_out":10170,"duration_ms":86289,"temperature":0.7,"pith_summary":"This paper establishes that, in a simple two-dimensional model of granular hard disks driven by a Langevin bath, a crystal can coexist with a liquid even though the crystal is denser and all collisions are dissipative. The hot-phase ordering is set by collision frequency rather than density: at coexistence the solid's Enskog collision rate $\\omega=8\\phi g^+\\sqrt{T/\\pi m}/\\sigma$ is lower than the liquid's, so the solid dissipates less and runs hotter. The authors extend kinetic theory, using measured contact values $g^+$, to predict granular temperatures across the coexistence region, including inside the solid phase, and obtain a dimensionless criterion $\\Lambda>1$ for which the solid is necessarily hotter. In the companion $\\Delta+\\gamma$ model, where collisions inject rather than only remove energy, the same reasoning gives a colder solid, matching simulations. This changes the usual expectation that the denser coexisting phase must be the colder one and identifies collisional dynamics as the controlling factor.","feed_headline":"A denser granular crystal can be hotter than its liquid","feed_subtitle":"Coexistence temperatures are set by collision rate, not density; kinetic theory predicts when the crystal is hot.","key_machinery":"The load-bearing object is Enskog's collision frequency $\\omega(T,\\phi g^+)=8\\phi g^+\\sqrt{T/\\pi m}/\\sigma$, combined with the steady-state energy balance $(\\omega/2)[m\\Delta^2+\\alpha\\Delta\\sqrt{\\pi m T}-(1-\\alpha^2)T]-2\\gamma(T-T_b)=0$. Because the temperature equation depends on density only through $\\phi g^+$, the paper isolates $\\phi g^+\\equiv G(T)$, substitutes $G(T)$ into the virial pressure, and writes the pressure as $p=\\phi\\,\\tilde p(T,G(T))$, linear in $\\phi$. Mechanical equilibrium $p_s=p_l$ then yields $\\tilde p(T_s)<\\tilde p(T_l)$ whenever $\\phi_s>\\phi_l$, and the monotonicity of $\\tilde p$ fixes which phase is hotter. For the GLM, $\\tilde p_{\\mathrm{GLM}}(\\tilde T)=T_b\\tilde T[1+\\Lambda(1-\\tilde T)\\tilde T^{-3/2}]$; for $\\Lambda>1$ this function decreases over the physical interval $\\tilde T<1$, yielding $T_s>T_l$. The derivation assumes molecular chaos and Gaussian velocity statistics, which are used to obtain Enskog's $\\omega$ and the collisional averages.","core_discovery":"At the level of the granular Langevin model (GLM), the central claim is that liquid-solid coexistence with $T_s>T_l$ occurs near the equilibrium liquid-hexatic transition even though the only energy input is the bath and collisions purely dissipate energy. The sign of the temperature difference follows from pressure balance: equal hard-disk pressures force $\\phi_s g_s^+<\\phi_l g_l^+$, and because $\\omega\\propto\\phi g^+$, the denser phase has the smaller collision frequency and therefore the smaller dissipation rate. Eliminating $\\phi g^+$ from the steady-state energy balance turns coexistence into the inequality $\\tilde p(T_s)<\\tilde p(T_l)$; for the GLM, $\\tilde p$ is decreasing on the physical temperature interval whenever $\\Lambda>1$, which forces $T_s>T_l$. The same construction makes $\\tilde p$ increasing for the $\\Delta+\\gamma$ model, forcing $T_s<T_l$ throughout its physical parameter range. Event-driven molecular-dynamics simulations with up to about $10^5$ particles confirm both orderings, with phase-temperature differences of order $0.1$--$0.3\\%$.","pith_inferences":["A direct experimental check would be to track per-particle collision rates separately in ordered and disordered regions of a quasi-2D vibrated monolayer; the mechanism predicts that the hotter phase is the one with fewer collisions, not the less dense one.","Because the sign of the temperature difference is controlled by whether $\\tilde p$ is increasing or decreasing, any other dissipative model with the same pressure-versus-temperature structure but a different collision-frequency law could show a different ordering; the paper's construction gives a template for classifying such cases.","For multicomponent or quasicrystalline granular solids, structurally forbidden collisions in the ordered phase suppress the collision frequency further, so the hot-solid effect should be stronger than the monodisperse difference of about 0.1%--0.3%.","The near-equilibrium assumption limits the effect to small temperature differences; as dissipation grows, the first-order coexistence disappears, so an experimental search for a hot solid should stay close to the equilibrium melting point."],"forward_implications":["For the GLM with $\\Lambda>1$, the hotter-solid ordering follows from the theory without an equation of state or a measured $g^+$, so it can be tested purely through macroscopic parameters $\\alpha$, $\\gamma$, and $T_b$.","Kinetic-theory temperatures agree with measured granular temperatures from dilute conditions through the solid phase, so coexistence temperatures can be predicted rather than extracted from histograms.","In the $\\Delta+\\gamma$ model the solid is always colder for physical parameters, showing that whether the dense phase is hotter or colder depends on where collisions inject or remove energy.","In the GLM, decreasing $\\Lambda$ below 1 removes the first-order coexistence: the pressure loop that marks it disappears, while an energy loop survives at the transition."],"supporting_citations":[{"why":"Supplies the equilibrium hard-disk pressure formula $p=4\\phi T(1+2\\phi g^+)/\\sigma^2\\pi$ that links coexistence pressure balance to the contact value $g^+$.","marker":"[96]"},{"why":"Gives the Enskog collision-frequency expression $\\omega=8\\phi g^+\\sqrt{T/\\pi m}/\\sigma$ that the temperature equation and the $\\phi g^+$ elimination rely on.","marker":"[97]"},{"why":"Fixes the equilibrium two-dimensional liquid-hexatic coexistence region used as the near-equilibrium reference for the simulations.","marker":"[87]"},{"why":"Introduces the granular Langevin model and reports a hotter solid in a binary granular mixture, motivating the present monodisperse test.","marker":"[72]"},{"why":"Provides the granular Brownian-motion formulation underlying the GLM's Langevin bath.","marker":"[84]"},{"why":"Supplies the dissipative collision rule with $\\Delta$ and the theoretical reference temperature $T_o$ for the $\\Delta+\\gamma$ model.","marker":"[83]"},{"why":"Is the event-driven molecular-dynamics method used to generate the simulation data.","marker":"[91]"}],"fun_headline_variants":["Hotter crystal than liquid: granular coexistence flip","Granular crystal can be hotter than its coexisting liquid","Dense solid, hot; dilute liquid, cold: granular twist","Collision rate predicts which granular phase runs hotter","Kinetic theory explains hot crystal in granular coexistence"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation assumes that collisions are uncorrelated (molecular chaos) and that particle velocities are nearly Gaussian even inside the crystal, so the Enskog collision-frequency formula and the elimination of $\\phi g^+$ through $G(T)$ remain valid in the solid phase.","fun_headline_variants_meta":{"raw":{"variants":["Hotter crystal than liquid: granular coexistence flip","Granular crystal can be hotter than its coexisting liquid","Dense solid, hot; dilute liquid, cold: granular twist","Collision rate predicts which granular phase runs hotter","Kinetic theory explains hot crystal in granular coexistence"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000223,"raw_usage":{"total_tokens":1490,"prompt_tokens":1014,"completion_tokens":476,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":630,"completion_tokens_details":{"reasoning_tokens":399}},"tokens_in":630,"tokens_out":476,"duration_ms":23526,"temperature":1.0,"reasoning_tokens":399,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:59:43.460501+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run event-driven GLM simulations at coexistence (for instance $\\alpha=0.99$, $T_b/[m(\\sigma\\gamma)^2]=0.125$) and count collisions per particle separately in the solid and liquid domains. If the hotter solid does not show a lower collision frequency than the colder liquid, the mechanism and the $\\phi g^+$ elimination fail; equally, measuring local pressures and temperatures should either satisfy or violate $\\tilde p(T_s)<\\tilde p(T_l)$.","supporting_citations":[{"cited_title":"Interfacial tension effects in finite, periodic, two-dimensional systems","cited_arxiv_id":null,"evidence_quote":"Supplies the equilibrium hard-disk pressure formula $p=4\\phi T(1+2\\phi g^+)/\\sigma^2\\pi$ that links coexistence pressure balance to the contact value $g^+$."},{"cited_title":"Hard-disk equation of state: First-order liquid-hexatic transition in two dimensions with three simulation methods","cited_arxiv_id":null,"evidence_quote":"Gives the Enskog collision-frequency expression $\\omega=8\\phi g^+\\sqrt{T/\\pi m}/\\sigma$ that the temperature equation and the $\\phi g^+$ elimination rely on."},{"cited_title":"Transport and fluctuations in granular fluids: From Boltzmann equation to hydrodynamics, diffusion and motor effects","cited_arxiv_id":null,"evidence_quote":"Fixes the equilibrium two-dimensional liquid-hexatic coexistence region used as the near-equilibrium reference for the simulations."},{"cited_title":"Self-assembly and non-equilibrium phase coexistence in a binary granular mixture","cited_arxiv_id":null,"evidence_quote":"Introduces the granular Langevin model and reports a hotter solid in a binary granular mixture, motivating the present monodisperse test."},{"cited_title":"Hyperuniform states generated by a critical fric- tion field","cited_arxiv_id":null,"evidence_quote":"Provides the granular Brownian-motion formulation underlying the GLM's Langevin bath."},{"cited_title":"Universality class of the motility-induced critical point in large scale off-lattice simulations of active particles","cited_arxiv_id":null,"evidence_quote":"Is the event-driven molecular-dynamics method used to generate the simulation data."}],"review_version":1}