{"id":"5731a115-ddef-4977-8056-d6f4a4ec0a26","arxiv_id":"2509.00216","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Entangled C-shaped granular particles form networks that undergo a standard 3D percolation transition, captured quantitatively by a continuum percolation model of rings.","lead":"Shaking C-shaped steel particles makes them interlock and form clusters that grow as shaking time increases. The entanglement networks behave like a standard percolation process, and a new model of randomly entangled rings reproduces the transition, which may help in designing interlocking metamaterials and robot collectives.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Hopf-link criterion (Eq. 5) is exact for closed rings but is applied to open C-particles; the paper admits this is only approximate for small θ, whereas the CP-model comparisons include θ up to 98°, so the network edges and universality claim rest on unvalidated ground at large θ.","rationale":"The reader's weakest-assumption identification is also the most load-bearing one. The central claim is that entanglement networks of C-particles, with edges defined by Eq. (5), undergo a percolation transition described by a ring continuum-percolation model and the standard 3D percolation universality class. That claim inherits all of its content from the edge definition. If Eq. (5) overcounts mechanical interlocking for open C-particles at larger opening angles, then every derived network property, threshold, and model comparison at those angles is suspect. The manuscript honestly flags this uncertainty, but it still uses the criterion at θ up to 98° in the in-container comparisons that support the CP-model description. The finite-size scaling via ηeff is a real secondary concern because it imports the ring CP model's own calibration and critical parameters, but it does not by itself threaten the small-θ universality evidence, which also rests on cluster-size distributions. An independent test of the linking criterion, as proposed above, would settle whether the large-θ scope of the claim is justified. Because the paper already receives a CONDITIONAL verdict, this stress-test leaves the reader's recommendation unchanged rather than escalating to rejection; the concern is a scope/validation issue, not an evident internal inconsistency.","tokens_in":19144,"tokens_out":6977,"duration_ms":91547,"concrete_test":"In the DEM data, for each pair classified as Hopf-linked by Eq. (5), perform a rigid-body separation check: attempt to move the two C-particles to infinity by continuous rigid-body motions without interpenetration (e.g., by a penetration-depth descent or a collision-free motion planner). Measure the false-positive rate of Eq. (5) relative to true mechanical non-separability for θ = 20°, 46°, 72°, 98°, and 112°. Then rebuild S1(⟨k⟩), χ(⟨k⟩), k1/2(θ), and the finite-size scaling collapse using only mechanically non-separable edges. If the false-positive rate is substantial for θ ≳ 72° or the rebuilt curves stop collapsing onto the ring CP model, the topological-criterion assumption is the limiting step and the large-θ conclusions must be revised; if the false-positive rate is negligible in exactly the θ range used for the universality claim, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Every network statistic in the paper — P(k), S1(⟨k⟩), χ(⟨k⟩), clustering, spectra, ns, and finite-size scaling — is built from edges defined by Eq. (5), a linking condition for closed circles through the C-particle centerlines. The paper itself states that this criterion \"well approximates their mechanical interlocking, especially when θ is small\" (Experiments and Simulation Methods) and reports that lifted clusters disintegrate for θ ≥ 115° (Fig. 1j). Unlike closed rings, open C-particles can have circle extensions that are Hopf-linked while the physical particles remain separable by threading through the gap; the false-positive rate should grow with opening angle, exactly where the manuscript also shows mechanical bonding degrading. The ring CP model uses the same Eq. (5) on closed rings, so agreement between the model and C-particle networks at finite θ may partly reflect shared use of the same proxy rather than an independent physical universality. The large-θ results — in particular the linear rise of k1/2 with θ and the claim that θ mainly changes mechanical bond survival — are most exposed because they cover θ up to 98° (in-container) and 151° (simulation). The universality-class assertion itself is presented for θ = 20°–25°, where the proxy is credible, so the core claim may survive for small θ; but the abstract-level claim that these C-particle networks are captured by a ring CP model needs the linking criterion validated at every θ used, not only where lifting happens to preserve clusters.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies C-shaped granular particles under vertical vibration and constructs 'entanglement networks' by declaring two particles linked when the circles completing their centerlines form a Hopf link (Eq. 5). Cluster statistics are measured in experiments (by lifting) and DEM simulations. The authors observe that the largest-cluster fraction grows with shaking time while the susceptibility peaks, identifying a percolation transition in the mean degree <k>. They introduce a continuum percolation (CP) model of randomly placed infinitely thin rings with the same Hopf-link rule, derive its excluded volume analytically as pi D^3/3, and obtain its threshold and critical exponents from Monte Carlo finite-size scaling. They report parameter-free collapses of in-container C-particle S1(<k>) and chi(<k>) onto the ring CP curves, plus similar clustering coefficients, adjacency spectra, and cluster-size distributions, and an approximate FSS collapse of DEM networks using a ring-CP-calibrated effective density. The paper concludes that the C-particle networks belong to the standard 3D percolation universality class and that opening angle mainly changes mechanical bond survival.","tokens_in":19555,"tokens_out":7638,"duration_ms":93057,"significance":"If the conclusions hold, the paper is a striking demonstration that a purely geometric null model—random rings connected by Hopf links—captures the structure of entangled granular networks, with implications for granular metamaterials, entangled robotic swarms, and network-based analysis of nonconvex granular matter. Strengths include the large experimental dataset, controlled DEM simulations, a clean Monte Carlo realization of the ring CP model, parameter-free in-container comparisons for S1 and chi, and a useful analytic excluded-volume calculation for rings. The claimed universality with 3D percolation is plausible for small opening angles but is not fully established, because the C-particle FSS is calibrated through the ring CP model and the topological edge definition is only validated mechanically for small theta.","major_comments":[{"comment":"The Hopf-link criterion in Eq. (5) is exact for closed rings, but the manuscript applies it to open C-particles and itself states that it 'well approximates their mechanical interlocking, especially when theta is small.' The paper also reports that lifted clusters disintegrate for theta >= 115 deg (Fig. 1j). Nevertheless, Fig. 3a,b and the k1/2-versus-theta inset include theta = 98 deg, and Fig. 2d includes simulation data up to theta = 151 deg. Because the ring CP model uses the same Eq. (5) on closed rings, the collapse of C-particle networks onto the ring CP curve at large theta could reflect the shared topological proxy rather than genuine mechanical interlocking. Please quantify the correspondence between Eq. (5) links and force-bearing interlocking contacts for theta > roughly 70 deg (e.g., by checking which Hopf links survive during lifting or transmit tensile force in DEM), or ex","section":"Entanglement criterion (Methods, Eq. 5); Fig. 1j and Fig. 3a,b"},{"comment":"The FSS analysis for C-particle networks is not an independent test of the universality class. eta_eff is defined by inverting the empirical ring-CP relation eta = p1<k> + p2<k>^p3 (Fig. 4a, Supp. Table 2), and the granular data are then collapsed using the ring-CP threshold eta_c = 2.11 and exponents beta = 0.41, gamma = 1.80, nu-bar = 2.64. This establishes consistency with the ring CP scaling functions, but it cannot by itself show that the C-particle networks belong to the 3D percolation universality class. The reported collapse is also only 'reasonably good' and is imperfect for eta_eff > eta_c. I request an independent analysis: for example, perform FSS directly in <k> without the eta_eff mapping, or fit beta, gamma, and nu-bar to the DEM data and report confidence intervals. If the exponents are fixed a priori, the claim should be weakened to 'consistent with' rather than 'belongs","section":"Finite-size scaling, Eq. (3)-(4), Fig. 4e,f"}],"minor_comments":[{"comment":"The logarithmic growth law <k>(t) = k0 + alpha ln(1 + t/t0) is described in the text as purely empirical. The manuscript should be careful not to imply a mechanistic explanation for logarithmic aging in the abstract or discussion, since no microscopic derivation is offered.","section":"Evolution of the degree distribution, Eq. (2)"},{"comment":"The text states that the experimental procedure was repeated in 578 trials, while the Extended Data Fig. 1 caption reports 543 experimental trials. Please reconcile these numbers and clarify which trials are included in the averages.","section":"Results: experimental trials"},{"comment":"The claim that n_s ~ s^-2.19 is currently based on a visual comparison with a reference line. Please provide a quantitative exponent estimate (e.g., maximum-likelihood fit with uncertainty) for the DEM, experimental, and ring CP cluster-size distributions, particularly near <k> ~ 2.1.","section":"Fig. 3g and Extended Data Fig. 6"},{"comment":"The symbol k is used both for the node degree and for the vector k = n_i x n_j in Eq. (5). Please distinguish the two, for example by using boldface for vectors.","section":"Notation, Eq. (5)"},{"comment":"The granular FSS uses only N = 500, 1400, and 4000 for one opening angle (theta = 20 deg). It would be helpful to state explicitly whether larger N were attempted for the DEM system and to discuss the limited N range as a caveat when interpreting the quality of the collapse.","section":"Finite-size scaling, Fig. 4"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of cond-mat.soft and reports an original and potentially influential result. The main risk is overclaiming universality given the Hopf-link proxy for open C-particles and the ring-CP-calibrated FSS. The requested additional analyses are feasible and would substantially strengthen the manuscript. No concerns about citation patterns or novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The punchline: this is a real, useful paper. It introduces a continuum percolation model of randomly oriented infinitely thin rings, derives the excluded volume analytically (πD^3/3), finds the threshold ηc=2.11, and shows the model falls in the standard 3D percolation universality class. Then it shows that the entanglement networks of C-shaped granular particles, built from a topological linking criterion, match this model surprisingly well—both in simulation and experiment. That is a genuine advance for entangled granular matter.\n\nWhat's new: the ring CP model itself, the analytic excluded volume, the first percolation characterization of granular entanglement networks. The MC simulations are clean and the FSS of the model is convincing. The Poisson degree distribution, the clustering below sphere CP but above ER, the spectral comparisons—all consistent and well presented.\n\nWhere it gets weaker: the linking criterion. Eq. (5) is exact for closed rings, but C-particles are open, and the paper admits the approximation works best at small θ. The in-container S1(⟨k⟩) collapse includes θ up to 98°, where false positives should be more common. The universality claim itself is tested at θ=20–25°, so the core survives, but the abstract's blanket statement that \"these networks\" are captured by the ring model is stronger than what's validated. The FSS test for C-particles uses an effective density ηeff obtained from the ring model's own η(⟨k⟩) calibration, so it's not fully independent—good to see, but it can't stand alone. The experimental cutoff at 3.5 g is a real concern for the cluster size distribution; it doesn't break the S1 and χ curves, but claiming τ=2.19 from those data is risky without knowing the bias. And Eq. (2) is an admitted empirical fit; that's fine, but it means the logarithmic growth is an observation, not a result.\n\nNone of these are fatal. The central claim—percolation of entanglement networks and consistency with standard 3D percolation for small openings—is solid. The large-θ part needs a more careful validation of the edge definition, or a tempering of the language.\n\nFor a referee: this deserves serious engagement. I'd send it out, expecting the authors to either validate Eq. (5) against a more physical interlocking measure for θ ≳ 70°, or restrict their claims to the regime where it's credible. A data release would help. I'd cite this if I worked on entangled matter.","headline":"Serious experimental+simulation study: entangled C-particle networks percolate and match a new ring continuum percolation model; core claim holds at small opening angles, but the Hopf-link edge criterion is unvalidated at large θ and the abstract overreaches.","tokens_in":20033,"tokens_out":3973,"would_cite":true,"duration_ms":43549,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["64.60.ah","45.70.-n"],"model":"deepseek-v4-flash","headline":"Entanglement networks of vibrating C-shaped granular particles undergo a well-defined percolation transition, and a continuum percolation model of infinitely thin rings describes it, including its three-dimensional percolation universality","keywords":["granular matter","entanglement network","percolation transition","continuum percolation","Hopf link","nonconvex particles","discrete element method","finite-size scaling"],"falsifier":"In DEM simulations at θ≈125°, build the network from force-transmitting contacts rather than Hopf links and measure S1(⟨k⟩), χ(⟨k⟩), and the cluster-size exponent; if these no longer collapse onto the ring CP model or follow τ=2.19, the claimed universality is an artifact of the link criterion rather than a property of the granular material.","tokens_in":19000,"feed_emoji":"🔗","tokens_out":7925,"duration_ms":81055,"temperature":0.7,"pith_summary":"The paper tries to establish that the random-looking tangles formed by vibrating C-shaped particles have a sharp, predictable connectivity structure: particles and their geometric links form a network that percolates as links accumulate, and the transition is the same universal geometric percolation found in random geometric graphs. Experiments and simulations show a giant entangled cluster appears once the mean number of links per particle passes a threshold near 2.1, and the order parameter, susceptibility, cluster-size distribution, average cluster length, clustering coefficient, and finite-size scaling all match a continuum percolation model of infinitely thin rings. If correct, the bulk connectivity of entangled granular matter can be predicted from geometry rather than from detailed contact mechanics. The paper also reports that the mean link number grows logarithmically in vibration time, linking entanglement buildup to slow aging in disordered systems.","feed_headline":"Tangled C-shaped grains percolate like random linked rings","feed_subtitle":"Vibrated particles form a giant cluster only past a link threshold near 2.1, matching 3D percolation universality.","key_machinery":"The central object is the entanglement network, whose nodes are particles and whose edges are Hopf links between their centerline circles, detected by the topological criterion of Eq. (5). On top of this sits the ring continuum percolation model—randomly placed, infinitely thin rings, connected when Hopf-linked—whose analytically derived excluded volume v_ex=πD³/3 supplies the reduced density η that makes data collapse possible. The combination turns a mechanically noisy granular system into a parameter-free geometric null model against which measured networks can be compared.","core_discovery":"Entanglement networks of C-shaped granular particles under vibration undergo a percolation transition belonging to the standard three-dimensional percolation universality class (τ≈2.19, β≈0.41, γ≈1.80), with an infinite-size threshold of ⟨k⟩c=2.11 links per particle. The evidence comes from experiments and DEM simulations: S1(⟨k⟩) and χ(⟨k⟩) for in-container networks at opening angles 20°–98° collapse onto Monte Carlo data for randomly placed infinitely thin rings; the same ring model captures average eccentricity, clustering coefficient, adjacency spectra, and cluster size distributions; and finite-size scaling with an effective density reproduces the standard critical exponents. The articl","pith_inferences":["Editorial inference: if the universality holds, mechanical failure of granular tangles—such as the peak in vertical cluster length at the threshold—could be predicted from percolation observables before any force-chain calculation, simplifying the design of load-bearing entangled structures.","Editorial inference: the Hopf-link criterion is defined on closed circles, so a direct test is to build networks from force-transmitting contacts at large opening angles; deviations would show where topological percolation stops being mechanical percolation.","Editorial inference: the ring-model equivalence suggests that other systems held together by topological links—kinetoplast DNA networks, mechanically interlocked polymers, and entangled robot collectives—may share these universal connectivity statistics even when their microscopic dynamics differ."],"forward_implications":["The connectivity of entangled granular assemblies can be described by a geometric null model: once the mean degree is known, quantities such as giant-cluster size, susceptibility, cluster length, and clustering coefficient are determined without mechanical fitting parameters.","The percolation threshold in mean degree is about 2.11, far above the value 1 for Erdős–Rényi random networks, so spatial packing constraints materially delay the onset of a giant entangled cluster.","Opening angle controls whether topological links behave as mechanical bonds: for large angles, lifted clusters disintegrate, the measured threshold shifts upward, and the network statistics deviate from the ring model.","Cluster size distributions and finite-size collapses follow the standard 3D percolation exponents, implying universal critical behavior rather than system-specific connectivity statistics.","The logarithmic growth of mean degree with vibration time connects entanglement buildup to logarithmic aging and memory effects seen in other disordered systems."],"supporting_citations":[{"why":"Supplies the Hopf-link topological entanglement criterion (Eq. 5) used to define edges in both DEM and Monte Carlo networks.","marker":"[19]"},{"why":"Supplies percolation theory definitions, scaling forms, and the standard 3D critical exponents the paper compares against.","marker":"[22]"},{"why":"Provides the universal susceptibility behavior for off-lattice percolation used to identify the transition and universality class.","marker":"[23]"},{"why":"Defines continuum percolation / random geometric graphs, the model family the ring CP model belongs to.","marker":"[26]"},{"why":"Provides the reduced density η=nv_ex formulation that connects mean degree to density in continuum percolation.","marker":"[38]"},{"why":"Supplies the sphere CP model's threshold and random-geometric-graph properties used as comparison points.","marker":"[39]"},{"why":"Supplies the disk CP threshold in 3D (⟨k⟩c=2.27) used to situate the ring threshold of 2.11.","marker":"[40]"},{"why":"Justifies using the S2/S1 intersection to locate the critical point independent of system size.","marker":"[45]"}],"fun_headline_variants":["C-grain entanglement percolates at 2.11 links per particle","Tangled C-grains percolate with universal 3D exponents","Percolation in entangled C-grain networks hits 3D universality","C-grain network percolation matches random ring model"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The load-bearing premise is that the topological criterion for two closed circles being interlocked (a Hopf link, Eq. 5) also represents a mechanical bond between open C-particles; the paper's own data show this approximation degrades at large opening angles, where lifted clusters frequently disintegrate.","fun_headline_variants_meta":{"raw":{"variants":["C-grain entanglement percolates at 2.11 links per particle","Tangled C-grains percolate with universal 3D exponents","Percolation in entangled C-grain networks hits 3D universality","C-grain network percolation matches random ring model"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000679,"raw_usage":{"total_tokens":2897,"prompt_tokens":692,"completion_tokens":2205,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":436,"completion_tokens_details":{"reasoning_tokens":2139}},"tokens_in":436,"tokens_out":2205,"duration_ms":21882,"temperature":1.0,"reasoning_tokens":2139,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T13:49:44.146522+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In DEM simulations at θ≈125°, build the network from force-transmitting contacts rather than Hopf links and measure S1(⟨k⟩), χ(⟨k⟩), and the cluster-size exponent; if these no longer collapse onto the ring CP model or follow τ=2.19, the claimed universality is an artifact of the link criterion rather than a property of the granular material.","supporting_citations":[{"cited_title":"Hoell and H","cited_arxiv_id":null,"evidence_quote":"Supplies the Hopf-link topological entanglement criterion (Eq. 5) used to define edges in both DEM and Monte Carlo networks."},{"cited_title":"Stauffer and A","cited_arxiv_id":null,"evidence_quote":"Supplies percolation theory definitions, scaling forms, and the standard 3D critical exponents the paper compares against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the universal susceptibility behavior for off-lattice percolation used to identify the transition and universality class."},{"cited_title":"Penrose, Random Geometric Graphs(Oxford Univer- sity Press, 2003)","cited_arxiv_id":null,"evidence_quote":"Defines continuum percolation / random geometric graphs, the model family the ring CP model belongs to."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the reduced density η=nv_ex formulation that connects mean degree to density in continuum percolation."},{"cited_title":"Dall and M","cited_arxiv_id":null,"evidence_quote":"Supplies the sphere CP model's threshold and random-geometric-graph properties used as comparison points."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the disk CP threshold in 3D (⟨k⟩c=2.27) used to situate the ring threshold of 2.11."},{"cited_title":"Almeira, O","cited_arxiv_id":null,"evidence_quote":"Justifies using the S2/S1 intersection to locate the critical point independent of system size."}],"review_version":1}