{"id":"19e0b5a5-014b-4728-b677-9b57df3ec328","arxiv_id":"2502.03354","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The paper derives an analytical expression for the random close packing density of polydisperse hard disks in 2D, which grows from 0.886 to about 0.98 with size polydispersity and matches a recent simulation at moderate polydispersity.","lead":"An analytical formula is proposed for the maximum density of randomly packed mixtures of hard disks of different sizes in two dimensions. The formula is useful for predicting how granular and colloidal films pack, and it agrees with a recent state-of-the-art simulation at the one point tested.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central claim relies on unvalidated transfer of equilibrium polydisperse hard-disk EOS to jammed 2D packings; the single swap-MC comparison point cannot support the full phi_RCP(s) curve.","rationale":"After reading the paper in good faith, the central claim is the analytical prediction of phi_RCP(s) from Eq. (13). The derivation is algebraically straightforward and internally consistent: the monodisperse limit of Eq. (12) reduces to the chosen EOS, and Table I shows stable values across three EOS choices. The paper also makes a falsifiable contact with the swap-MC result at s=0.25. However, the single most load-bearing assumption is that the equilibrium Santos polydisperse EOS (Eqs. 10-12) can be used to evaluate the contact value g(sigma;phi) in the jammed state through Eq. (7). This assumption is imported from the 3D hard-sphere analysis of Ref. [10]; no 2D numerical or theoretical evidence is given. In 2D, the proximity of RCP to CP and the importance of hexatic order make this transfer particularly nontrivial. The one comparison point (s=0.25) is not a critical test: the swap-MC packings are not necessarily isostatic, so any theory that predicts a lower density could 'compare well.' Additional concerns, the abstract's 'power-law' mislabel, the ambiguity in the typesetting of Eq. (12), and the choice of EOS, are real but secondary; the first is a typo, the second is resolved by the monodisperse limit, and the third only shifts predictions by small amounts. I therefore agree with the reader's weakest-assumption identification and see no reason to change the CONDITIONAL verdict. The proposed simulation test would directly check whether Eq. (8) plus Eq. (13) holds in 2D, settling the concern.","tokens_in":8547,"tokens_out":20169,"duration_ms":163258,"concrete_test":"Perform event-driven or swap-Monte-Carlo simulations of jammed polydisperse hard disks with a log-normal size distribution at s=0.25 and at least one other s (e.g., 0.1 or 0.5). For each jammed configuration, measure the mean contact number z (excluding rattlers) and the packing fraction phi. Test the prediction z = 4 C0[Z_Santos(phi)-1] from Eq. (8) with C0 fixed by Table I (e.g., C0=0.0118828 for the Henderson EOS). If the measured z deviates from the predicted z by more than about 10% at the same phi, or if the phi at which z=4 differs from phi_RCP(s) by more than the statistical error, then Eq. (13) is falsified and the equilibrium-to-jamming transfer is invalid.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central result, Eq. (13) C0[Z(phi_RCP)-1]=1, is obtained by combining (i) the equilibrium virial relation (7) with (ii) the Santos polydisperse EOS (Eqs. 10-12), which was derived for equilibrium hard-disk fluids. For jammed packings near RCP, the pair distribution function has a singular contact peak (Eq. 3), and the pressure/contact-value relation is not that of an equilibrium fluid. The only numerical support cited is Ref. [10] for 3D hard spheres; no 2D validation is given. The comparison with the swap-Monte-Carlo point phi=0.905 at s=0.25 (Ref. [69]) is a single point and is not a critical test: the authors themselves state those packings are not necessarily isostatic (z may exceed 4), so an underestimating theory could still agree. If the mapping in Eq. (10) fails at high density, the entire phi_RCP(s) curve, including the saturation near 1/(1+C0) about 0.97-0.98, is unsupported. The monodisperse limit (Table I) gives phi_RCP about 0.886, within the broad quoted range 0.81-0.89, but this does not validate the polydisperse extension.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes an analytical route to the random close packing density of polydisperse hard disks. It combines a jammed-state contact relation, z = 4 C0 [Z(phi) - 1], with equilibrium hard-disk equations of state, after calibrating C0 at the monodisperse triangular close-packing point. Polydispersity is introduced through the Santos effective-packing-fraction EOS, Eq. (12). Solving C0[Z(phi_RCP) - 1] = 1 yields phi_RCP as a function of the width of the size distribution; the result is compared with swap-Monte-Carlo data at s = 0.25, giving phi_RCP = 0.8924 versus 0.905.","tokens_in":8871,"tokens_out":12833,"duration_ms":107765,"significance":"If its assumptions held, this would be the first closed-form analytical formula for the 2D polydisperse random close packing fraction, with potential applications to granular monolayers, interfacial assemblies, and composite materials. The derivation is concise, and the s = 0.25 comparison is an external benchmark rather than a fit, because C0 is fixed at the monodisperse close-packing point. However, the two central transfers—equilibrium fluid equation of state to jammed packings, and monodisperse calibration to arbitrary polydispersity—are not validated in 2D, and Eq. (12) as printed does not reduce to the monodisperse equation of state. The paper therefore cannot be accepted without substantial revision.","major_comments":[{"comment":"Equation (12) as written is internally inconsistent: for a monodisperse distribution one has lambda = 1, alpha = 1, and phi_eff = phi, so the equation reduces to Z = (1 + phi) + phi^2 Z_s(phi), not to Z_s(phi). The Santos FMT mapping has the structure Z = 1/(1 - phi) + alpha [ Z_s(phi_eff) - 1/(1 - phi_eff) ], so Eq. (12) appears to contain an erroneous factor phi and to be missing the division by phi_eff (or an equivalent rearrangement). Since Eq. (13) and all polydisperse numerical values, including phi_RCP = 0.8924 and the curves in Fig. 1, are computed from Eq. (12), the polydisperse predictions must be recomputed with the correct EOS before the central claim can be assessed.","section":"Section 'To extend the EOS', Eq. (12)"},{"comment":"The proportionality in Eq. (7) is the pivot of the theory: it takes the equilibrium virial contact value, (Z - 1)/(2 phi), and asserts that the jammed contact value g(sigma; phi) is proportional to it. The paper justifies this by Ref. [10] for 3D hard spheres, but provides no 2D numerical or theoretical check. In a jammed disk packing near RCP, the contact peak is singular and the pressure is not given by the equilibrium virial expression; C0 absorbs the proportionality factor at one density, but the assumed functional form z proportional to Z(phi) - 1 across all densities and polydispersities is unchecked. This makes the entire phi_RCP(s) curve in Fig. 1 conditional on an unvalidated analogy.","section":"Section 'EOS', Eq. (7)"},{"comment":"The only quantitative validation of the polydisperse theory is a single point at s = 0.25, where the paper predicts phi_RCP = 0.8924 against the swap-MC value 0.905. The paper itself states that the packings in Ref. [69] are not necessarily isostatic, i.e. z may exceed 4, so the simulated phi_RCP may be set by a different criterion than z = 4. With one point and an acknowledged mismatch in the defining condition, the agreement cannot distinguish between the proposed mechanism and a systematic underestimate. The authors should compare with additional polydisperse data (e.g. bidisperse disk packings or other distribution widths) or directly measure z(phi) in simulated jammed packings to test Eq. (8).","section":"Section 'Comparison with Ref. [69]'"}],"minor_comments":[{"comment":"The constant g0 in the contact-peak ansatz is said to be determined by a boundary condition, but the boundary condition is never specified; please state it explicitly.","section":"Section 'EOS', Eq. (3)"},{"comment":"The abstract describes the comparison as a 'power-law size distribution' with s = 0.246, while the text and Fig. 1 use a log-normal distribution with sigma approximately 0.2462; please reconcile the nomenclature.","section":"Abstract and text"},{"comment":"For polydisperse disks, the average contact relation should specify how g(r) is averaged over ij pairs and how the single mean diameter sigma follows from that average; currently the reduction from the polydisperse contact condition sigma_ij = (sigma_i + sigma_j)/2 is implicit.","section":"Section 'Polydisperse contacts', Eq. (2)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a short extension of the author's prior 3D framework, and the novelty is the 2D result. The most serious issue is the apparent error in Eq. (12), which by itself invalidates the numerical predictions as printed; this is fixable, but the authors must also confront the lack of direct 2D validation of the equilibrium-to-jammed transfer. If they correct the EOS and add even one or two independent tests, the paper could become suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper gives the first analytical curve for phi_RCP as a function of polydispersity in 2D hard disks. It combines Zaccone's earlier crowding relation, z = 4C0[Z(phi)-1], with Santos's equilibrium polydisperse equation of state, then sets z=4 at isostaticity. The derivation is transparent and the algebra checks out. The scaling constant C0 is pinned at the exact triangular close-packing point rather than fitted to RCP data, which is a plus. The monodisperse limit lands near 0.886, inside the broad 0.81-0.89 range quoted in the literature, though that range is so wide it does not provide a sharp test.\n\nThe real soft spot is the transfer. Equation (8) assumes the jammed contact value is proportional to the equilibrium virial contact value. That proportionality is imported from the author's 3D work and is not independently checked for 2D jammed disks. The Santos EOS used for the polydisperse mixture is an equilibrium high-density mapping, and applying it to a jammed state with a singular contact peak is an extrapolation. The single comparison with swap Monte Carlo at s=0.25 is not a strong test: the authors themselves note those packings are not necessarily isostatic, so the agreement between 0.892 and 0.905 could easily be partly fortuitous. I would want at least one more polydispersity benchmark before believing the full curve, especially the saturation plateau near 0.97-0.98.\n\nThere are also smaller issues. The abstract calls the size distribution a power law, but the text uses a log-normal. Equation (12) as printed is ambiguous; the denominator structure is not clear. These are fixable but suggest the manuscript was not checked carefully enough.\n\nCandidly, the central argument is coherent on its own terms. The paper does not oversell: it states the proportionality is an assumption and flags the non-isostatic nature of the swap packings. So this is a real theoretical proposal, not a hand-wave. I would send it to peer review rather than desk-reject. A good referee should push for either a direct 2D numerical check of Eq. (8) or a much more explicit caveat, and for correction of the abstract and Eq. (12).\n\nWho benefits? People who need a quick analytical estimate for polydisperse monolayers, colloidal films, or granular packing, and who accept the equilibrium-to-jammed transfer as a working approximation. For my own work, I would not cite it as a validated result yet, but I would mention it as a testable prediction. It deserves a serious referee.","headline":"Useful first analytical 2D polydisperse RCP formula, but the central EOS transfer to jammed disks is unvalidated and one swap-MC point is too thin a benchmark.","tokens_in":9371,"tokens_out":3182,"would_cite":false,"duration_ms":32906,"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":"The paper claims that the random close packing density of polydisperse hard disks is obtained by solving $C_0[Z(\\phi_{\\mathrm{RCP}})-1]=1$ with the polydisperse compressibility $Z$ from an effective-density equation of state, yielding…","keywords":["random close packing","polydisperse hard disks","effective packing fraction","isostaticity","equation of state","jamming","size polydispersity","2D disk packings"],"falsifier":"Simulate slow compressions or swap-based packings of log-normal polydisperse hard disks at several reduced standard deviations $s$ and compare the measured maximum disordered packing fraction with Eq. (13); a systematic gap that grows with $s$, or a non-monotone curve, would rule out the assumed mapping.","tokens_in":8322,"feed_emoji":"📐","tokens_out":7813,"duration_ms":71484,"temperature":0.7,"pith_summary":"This paper claims that the densest disordered packing of hard disks with a spread of sizes can be computed analytically, without simulation, by solving a single algebraic equation. The equation, $C_0[Z(\\phi_{\\mathrm{RCP}})-1]=1$, combines the isostatic condition that each disk has on average four contacts with an equilibrium equation of state for polydisperse disks. The solution grows with the reduced standard deviation $s$ of the size distribution and saturates near $\\phi_{\\mathrm{RCP}}\\approx 0.97$--$0.98$ at large $s$; at $s=0.25$ it gives $0.8924$, close to the recent numerical estimate $0.905$. If correct, the result turns a long-standing packing problem into a formula that depends only on the first three moments of the disk-size distribution, with consequences for granular monolayers, particle-stabilized interfaces, and two-dimensional materials.","feed_headline":"One equation predicts mixed-disk packing density","feed_subtitle":"A closed-form formula gives the densest disordered packing of mixed-size disks, matching simulations within 1.4 percent.","key_machinery":"The load-bearing object is the effective-density equation of state for polydisperse hard disks, which maps a mixture at packing fraction $\\phi$ onto an equivalent monodisperse fluid at $\\phi_{\\mathrm{eff}}=\\phi/[\\phi+\\lambda(1-\\phi)]$, with $\\lambda=m_3/m_2^2$ determined by the dimensionless moments of the disk-size distribution. Its compressibility $Z(\\phi)$, meaning the pressure divided by density in thermal units, enters the isostatic condition $C_0[Z(\\phi_{\\mathrm{RCP}})-1]=1$, obtained by writing the mean contact number as $z=4C_0[Z(\\phi)-1]$ and setting $z=4$. The work this machinery does is to reduce the many-species contact problem to a one-species equation of state, so $\\phi_{\\mathrm{RCP}}$ becomes a function of the distribution's first three moments.","core_discovery":"The central claim is that random close packing in two dimensions is the rigidity-onset density, where the mean contact number reaches the isostatic value $z=4$, and that this density is the solution of $C_0[Z(\\phi_{\\mathrm{RCP}})-1]=1$. Here $Z$ is the compressibility of the polydisperse disk fluid obtained from the effective packing fraction mapping $\\phi_{\\mathrm{eff}}=\\phi/[\\phi+\\lambda(1-\\phi)]$ with $\\lambda=m_3/m_2^2$, and $C_0$ is fixed by the monodisperse triangular close packing at $\\phi=\\pi/\\sqrt{12}\\approx 0.9069$. The paper evaluates $C_0$ with three standard disk equations of state to show the prediction is robust, and compares the log-normal result at $s=0.25$ ($\\phi_{\\mathrm{RCP}}=0.8924$) with the recent numerical estimate $0.905$ from irreversible swap Monte Carlo. This is presented as the first analytical solution to the polydisperse random close packing problem for two-dimensional disks.","pith_inferences":["Because the solution depends on the size distribution only through its first three moments, two very different distributions with the same mean, variance, and skewness would be predicted to have identical $\\phi_{\\mathrm{RCP}}$; this moment-collapse is a direct, testable consequence not highlighted in the paper.","The current numerical comparison is at a single value of $s$; a stronger test would be to simulate log-normal disks over a range of $s$ and compare the whole $\\phi_{\\mathrm{RCP}}(s)$ curve with Eq. (13).","If the equilibrium-to-jammed transfer holds, the same effective-density route could likely be applied to models with soft repulsive interactions at jamming, where isostaticity and the contact number are already known to control the mechanical response.","The predicted saturation near 0.97--0.98 suggests an upper bound for isostatic disordered disk packings; a natural extension is to ask whether any protocol that preserves strict disorder can exceed this plateau."],"forward_implications":["The random close packing fraction of any polydisperse disk system can be estimated by solving one algebraic equation from the first three moments of its size distribution, bypassing expensive packing simulations.","At a fixed size spread $s=0.25$, the theory yields $\\phi_{\\mathrm{RCP}}=0.8924$, about 1.4 percent below the $0.905$ value from the latest irreversible swap Monte Carlo simulations.","Increasing polydispersity raises the predicted packing fraction monotonically toward a plateau near $\\phi_{\\mathrm{RCP}}\\approx 0.97$--$0.98$, always below the geometric limit of 1.","The final formula is insensitive to which of the three standard disk equations of state is used at moderate polydispersity, so the prediction is not an artifact of a particular choice.","The same approach previously applied to three-dimensional polydisperse spheres now covers two-dimensional disk monolayers, connecting to interfacial, polymer, and biological packing problems."],"supporting_citations":[{"why":"Introduces the crowding model that relates the contact number to the compressibility and calibrates the proportionality constant through the close-packing reference.","marker":"[9]"},{"why":"Supplies the numerical evidence that a rescaled equilibrium hard-sphere theory describes jammed packings, the justification borrowed for the two-dimensional disk derivation.","marker":"[10]"},{"why":"Establishes the isostatic criterion $z=2d$ used to identify random close packing with the onset of rigidity.","marker":"[32]"},{"why":"Provides the effective packing fraction mapping and the polydisperse compressibility equation that feed Eq. (12) and ultimately Eq. (13).","marker":"[66]"},{"why":"Supplies one of the monodisperse disk equations of state used to compute the benchmark prediction $\\phi_{\\mathrm{RCP}}=0.8924$.","marker":"[64]"},{"why":"Provides the irreversible swap Monte Carlo estimate $\\phi_{\\mathrm{RCP}}=0.905$ to which the analytical result is compared.","marker":"[69]"}],"fun_headline_variants":["Analytic solution solves 2D random packing of mixed disks","Polydisperse disk packing density from a single formula","First analytical theory for mixed-disk random packing","Closed-form equation predicts densest disorder in 2D disks"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The equilibrium pressure-density relation for polydisperse hard disks is assumed to remain valid for jammed, out-of-equilibrium packings near random close packing, a transfer that the paper does not test in two dimensions.","fun_headline_variants_meta":{"raw":{"variants":["Analytic solution solves 2D random packing of mixed disks","Polydisperse disk packing density from a single formula","First analytical theory for mixed-disk random packing","Closed-form equation predicts densest disorder in 2D disks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000692,"raw_usage":{"total_tokens":3153,"prompt_tokens":990,"completion_tokens":2163,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":606,"completion_tokens_details":{"reasoning_tokens":2096}},"tokens_in":606,"tokens_out":2163,"duration_ms":15834,"temperature":1.0,"reasoning_tokens":2096,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T05:00:02.101722+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate slow compressions or swap-based packings of log-normal polydisperse hard disks at several reduced standard deviations $s$ and compare the measured maximum disordered packing fraction with Eq. (13); a systematic gap that grows with $s$, or a non-monotone curve, would rule out the assumed mapping.","supporting_citations":[{"cited_title":"Zaccone, Physical Review Letters 128, 028002 (2022)","cited_arxiv_id":null,"evidence_quote":"Introduces the crowding model that relates the contact number to the compressibility and calibrates the proportionality constant through the close-packing reference."},{"cited_title":"Anzivino, M","cited_arxiv_id":null,"evidence_quote":"Supplies the numerical evidence that a rescaled equilibrium hard-sphere theory describes jammed packings, the justification borrowed for the two-dimensional disk derivation."},{"cited_title":"Zaccone and E","cited_arxiv_id":null,"evidence_quote":"Establishes the isostatic criterion $z=2d$ used to identify random close packing with the onset of rigidity."},{"cited_title":"Santos, S","cited_arxiv_id":null,"evidence_quote":"Provides the effective packing fraction mapping and the polydisperse compressibility equation that feed Eq. (12) and ultimately Eq. (13)."},{"cited_title":"Ghimenti, L","cited_arxiv_id":null,"evidence_quote":"Provides the irreversible swap Monte Carlo estimate $\\phi_{\\mathrm{RCP}}=0.905$ to which the analytical result is compared."}],"review_version":1}