{"id":"36cbfa7a-16fa-49b3-bd63-f09c9f63f63f","arxiv_id":"2411.16293","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Soft granular spheres under circular shear show rate-dependent quaking: smooth flow at high rates, localized clusters at intermediate rates, and system-spanning clusters at low rates.","lead":"This paper builds a circular shear cell that shears soft granular particles continuously, and films the 3D motion of thousands of particles inside the pack. It shows that 'quakes', sudden stress drops, change character with shear rate and that a dimensionless rate from the same group's earlier work can line up results from different geometries.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The S_l collapse rests on a single post-hoc value of V* that drifted by more than an order of magnitude; without a sensitivity analysis the central claim is not quantitatively secured.","rationale":"The reader's weakest assumption identifies exactly the load-bearing concern: the post-hoc measurement of V* and the absence of a sensitivity analysis. Appendix E is explicit about the drift, and the central claim of geometry-independent unification via S_l is quantitatively hostage to this single number. I agree with the conditional verdict: the qualitative observations (clusters growing from sparse to system-spanning to rare) are well supported by the data, but the quantitative collapse claim needs either an error bar on V* or a sensitivity sweep. I do not see a separate, more fundamental flaw: the paper's qualitative picture is internally consistent, the authors acknowledge the one-packing-fraction and tracking-region limitations, and the clustering analysis includes a robustness check for the percentile threshold. Therefore no adjustment to the reader's verdict is needed; the conditional status already reflects the appropriate level of confidence.","tokens_in":13715,"tokens_out":1194,"duration_ms":13646,"concrete_test":"Re-plot Fig. 10(b) with V* = 0.2, 1, and 5 mm/s for the current CSC data while keeping the Ref.[17] curve fixed at its measured V* = 5 mm/s. Quantify the collapse using a standard metric (e.g., the mean squared log-distance between the two LD-count curves in the overlapping S_l range). If the collapse metric worsens by more than, say, a factor of 2 when V* is changed within the documented drift range, the claim that S_l unifies the two geometries is not robust. As a complementary check, repeat the tribology measurement of V* (Appendix E) immediately before and after a shearing run at case III to bracket any concurrent drift.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim is the collapse of LD-count curves from two geometries onto one master curve when plotted against S_l = l*gamma_dot/V* (Fig. 10b). This collapse depends entirely on the value assigned to V*, the material-specific sliding-speed threshold. Appendix E states that V* for the same PDMS particles was ~5 mm/s two years before the experiments, ~1 mm/s after one year of submersion, and 0.2 mm/s immediately after data collection. The value 0.2 mm/s was therefore not measured concurrently with the shearing runs, and the particles' surface state could have changed between the last data run and the post-hoc measurement. Because S_l is inversely proportional to V*, a factor-of-5 drift in V* shifts the present study's curve horizontally by the same factor on the log axis; the claimed 'reasonably good collapse' in Fig. 10(b) could degrade substantially or even vanish if the in-situ V* were, say, 1 mm/s instead of 0.2 mm/s. No sensitivity analysis is provided, and the collapse is judged only by eye. This is not a claim of internal inconsistency, but a concrete uncertainty in the one parameter that is doing all the unificatory work. The authors' own appendix flags the drift, so the concern is grounded in manuscript text, not speculation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports experiments in a circular shear cell on tightly packed PDMS spheres immersed in an index-matched fluid, with 3D particle tracking and simultaneous shear-stress measurement. It identifies a rate-dependent transition in internal motion: smooth, homogeneous flow at high shear rates; intermittent 'quaking' with localized clusters of large displacements at intermediate rates; system-spanning clusters at lower rates; and increasingly rare large events at the lowest rates. The authors then compare large stress-drop (LD) counts with their previous PRL 2021 experiment and claim that the dimensionless shear rate S_l = l*gamma_dot/V* collapses the LD-count curves from the two geometries into one master curve. The paper also documents cluster-size distributions, substructure within large clusters, and several appendices that address imaging, force-signal artifacts, and long-term drift of the tribological threshold V*.","tokens_in":13934,"tokens_out":4976,"duration_ms":49348,"significance":"If the central claims hold, the paper provides valuable three-dimensional, grain-scale evidence that the rate-dependent quaking transition is not specific to one shear geometry and that the previously proposed S_l parameter indeed organizes transitions across experiments. The paper has concrete strengths: the spatial cluster observations (cases I to IV) are directly illustrated and do not rely on S_l; the clustering results are checked against threshold choice in Appendix D; and the tribology input V* is obtained from an independent measurement reported in Appendix E rather than fitted to the LD-count curves. The paper is also candid about experimental limitations, including the drift of V* and the finite observation window. The main weaknesses are quantitative: the collapse in Fig. 10(b) depends on a single post-hoc V* value with no sensitivity analysis, the lowest-rate case rests on a single quaking event, and the variable strain-step convention across runs complicates the displacement-based comparisons.","major_comments":[{"comment":"The central quantitative claim that S_l collapses the large-drop counts rests entirely on assigning V* = 0.2 mm/s, but Appendix E reports that V* for the same PDMS particles was 5 mm/s two years earlier and 1 mm/s after one year of submersion, and that 0.2 mm/s was determined only after all shearing data were collected. Because S_l is inversely proportional to V*, a factor-of-2.5 to 5 drift in the in-situ V* would shift the CSC curve horizontally on the log axis by that factor, and the 'reasonably good collapse' in Fig. 10(b) is judged only visually without error bars. Please provide a sensitivity analysis that recomputes the collapse using V* values covering the full documented drift range (e.g., 0.2, 1, and 5 mm/s), or concurrent V* measurements; without that, the claimed unification across geometries is not quantitatively secured.","section":"Section IV, Fig. 10(b), Appendix E"},{"comment":"The non-monotonic decrease in Σ and in the higher moments <S^n>^{1/n} from case III to case IV is based on a single quaking event: the text states that case IV contains 'only a single quaking event' and that at the second-lowest shear rate a repeated run found one big event, but the plotted case is still n=1. This single-event statistics does not distinguish an intrinsic rareness from sampling noise. Please provide confidence intervals or replicate statistics, or at least report the repeated-run result quantitatively in the main figure, before treating the III-to-IV drop as part of the rate-dependent transition.","section":"Section III.A, Fig. 5(b); Section III.B, Fig. 7(b)"},{"comment":"The fluctuating displacement δr⃗ is computed over a strain step Δγ that is not the same for all runs: case I is sampled at 16Δγ and the case at 2.8×10^{-2} s^{-1} at 4Δγ, while the rest use Δγ = γ0/800 ≈ 3.5×10^{-3}. Since |δr⃗|, the top-5% selection, and hence the cluster statistics all depend on the strain interval over which particle motion is accumulated, the comparison between case I and the other cases mixes the effect of shear rate with the effect of sampling interval. Please show the results for case I and the 2.8×10^{-2} s^{-1} run at the standard Δγ, or demonstrate explicitly that the conclusions are insensitive to the strain interval.","section":"Section III.A and footnote 1"}],"minor_comments":[{"comment":"The last paragraph begins 'system. and the cluster exhibits substructures'; this should be one sentence: 'system, and the cluster exhibits substructures...'.","section":"Abstract"},{"comment":"The phrase 'the father half of the imaging area' should be 'the farther half of the imaging area'.","section":"Section II, paragraph after Fig. 2"},{"comment":"The phrase 'at slow driving ratess' contains a typo; it should be 'at slow driving rates'.","section":"Section III.A"},{"comment":"The caption says the shear-rate values are 'marked at the upper-right corner of every sub-plot', but those values are not legible in the printed figure; please add explicit numeric labels in the caption or in the panels.","section":"Fig. 3 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the journal's scope and the experimental effort is substantial. The central S_l collapse is the load-bearing claim and is currently sensitive to the unmeasured-in-situ V* value; the requested sensitivity analysis should be feasible. The single-event statistics for case IV and the unequal strain steps also need to be addressed quantitatively. With those repairs, the paper could become publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper's real contribution is the 3D spatial picture of rate-dependent quaking clusters in a circular shear cell. That part is new, well-illustrated, and honestly presented. The secondary claim—that the dimensionless shear rate S_l from the group's earlier work collapses the LD-count data from two geometries—is not quantitatively secured, because V* was measured after the runs and drifted by more than an order of magnitude. The S_l collapse should be treated as suggestive, not established.\n\nWhat is genuinely new: the circular shear cell imposes uniform, unlimited shear, and the index-matched tomography gives grain-resolved displacements for ~1200 particles. The cluster analysis (top 5% large displacements, Laguerre neighbors) shows a clear progression: sparse small clusters at high rate, large clusters at intermediate, system-spanning clusters at low, and rare extreme events at the slowest rates. The substructure within a large cluster—particles moving both with and against the base flow in different regions—is an interesting observation that the authors themselves flag as awaiting further study. These spatial results are robust to the threshold choice (Appendix D) and do not depend on V* at all. Credit is due for removing the mechanical background artifact carefully (Appendix C) and for reporting the V* drift honestly in Appendix E.\n\nThe soft spot is the S_l collapse in Fig. 10(b). V* for the same PDMS particles was 5 mm/s two years before the experiments, 1 mm/s after a year, and 0.2 mm/s immediately after data collection. Since S_l is inversely proportional to V*, a factor of 5 drift shifts the present curve horizontally by the same factor on a log axis. The claimed 'reasonably good collapse' with the PRL 2021 curve could substantially degrade if the in-situ V* were closer to 1 mm/s. No sensitivity analysis or error bars are given; the comparison is by eye. The LD count itself also lacks error bars, and case IV rests on a single quaking event. These are addressable concerns, not refutations, but they mean the unificatory claim is not yet demonstrated.\n\nThe paper is honest about its own limitations: one packing fraction, a restricted tracking region, and no particle tracking for the slowest run. These are reasonable and do not undermine the main spatial observations.\n\nWho this is for: granular physics experimentalists working on stick-slip and shear localization. The 3D cluster data will be useful even if the S_l collapse later needs revision. I would accept this for peer review; it deserves a serious referee, with revision requests focusing on uncertainty quantification and a sensitivity analysis for V*.","headline":"Genuinely new 3D cluster picture of rate-dependent quaking, but the S_l collapse is not quantitatively secured because V* drifted by an order of magnitude and was measured post-hoc.","tokens_in":14532,"tokens_out":3747,"would_cite":true,"duration_ms":31886,"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":"This paper claims that rate-dependent quaking in soft granular matter is a generic friction effect, captured by a single dimensionless shear rate that collapses data from two different shear geometries onto one curve.","keywords":["granular matter","stick-slip","quaking","soft particles","tribology","dimensionless shear rate","3D particle tracking","circular shear cell"],"falsifier":"Measure the large-drop count curves with $V_*$ determined in situ during the same runs, or with a fresh batch of particles whose threshold is stable; if the double-cone and circular-cell curves stop collapsing once the threshold is measured concurrently, the $S_\\ell$ unification rests on a post-hoc choice. A sharper check: repeat the circular-cell experiment with newly molded PDMS particles ($V_* \\approx 5$ mm/s) and verify that the quaking peak in the LD count shifts to an $S_\\ell$ value matching the old particles' peak.","tokens_in":13469,"feed_emoji":"🌀","tokens_out":11101,"duration_ms":93957,"temperature":0.7,"pith_summary":"Tightly packed soft particles under slow shear do not flow smoothly; instead they intermittently release stress in sudden drops that the authors call quaking. The paper reports that this rate-dependent quaking, previously seen in a double-cone shear experiment, appears in a completely different circular shear cell, and that the two datasets line up when the shear rate is divided by a material-specific friction threshold speed to form the dimensionless number $S_\\ell$. The new experiment also tracks particles in three dimensions through the bulk, showing that quakes are clusters of unusually large displacements: sparse patches at intermediate rates, system-spanning clusters at lower rates, and rare events at very low rates. The authors conclude that quaking is a generic, tribology-driven transition rather than a quirk of one geometry.","feed_headline":"Quaking in soft grains obeys one friction-scaled shear rate","feed_subtitle":"Intermittent stress drops in soft granular packs line up across two shear cells once speed-dependent friction is included.","key_machinery":"The load-bearing object is the dimensionless shear rate $S_\\ell \\equiv \\ell \\dot{\\gamma}/V_*$, which compares the typical sliding speed between neighboring particles, of order $\\ell \\dot{\\gamma}$, to $V_*$, the material-specific speed at which the friction between PDMS particles drops dramatically — the Stribeck transition from solid-solid contact to mixed lubrication. Equally important are the experimental components: a circular shear cell that imposes a uniform, cyclically rotating shear with strain $\\gamma_0 \\approx 2.8$ per cycle so particles never return to their old positions; refractive-index-matched 3D imaging with particle tracking that gives bulk grain trajectories; and a cluster analysis that keeps the top 5% of fluctuating displacements, connects nearest neighbors through Laguerre tessellation, and counts connected clusters. Together these make quaking visible as a spatial, collective event and tie its rate dependence to the friction threshold.","core_discovery":"The central claim is that intermittent stress drops — quaking — in tightly packed soft granular solids are a rate-dependent, tribology-controlled phenomenon whose onset and decline are organized by one dimensionless number, $S_\\ell \\equiv \\ell \\dot{\\gamma}/V_*$, where $\\ell$ is particle size, $\\dot{\\gamma}$ the shear rate, and $V_*$ the sliding speed at which interparticle friction drops sharply. In the new circular cell, quaking appears only at intermediate $S_\\ell$: high rates give smooth flow, intermediate rates give clusters of large fluctuating displacements, low rates give rare system-spanning events, and the large-drop (LD) count per strain from this geometry collapses onto the earlier double-cone data once plotted against $S_\\ell$ rather than $\\dot{\\gamma}$ alone. Within a system-spanning quake, particles do not all move together; patches move with and against the base flow, revealing substructure in the rearrangement.","pith_inferences":["Beyond the paper: if $V_*$ changes with surface age, the practical implication is that $S_\\ell$ should be treated as time-dependent; embedding a tribology reference measurement in every shearing run would turn the collapse into a continuously testable relation.","Beyond the paper: the substructure observed within large clusters suggests a quake is a percolating sequence of localized rearrangements rather than a single slip plane; one could test this by tracking the time ordering of activated patches within one event.","Beyond the paper: the cluster-size distributions reported here are the raw material for a soft-matter analogue of earthquake frequency-magnitude statistics; the analogue of the b-value should vary systematically with $S_\\ell$, and that variation could be predicted from these data.","Beyond the paper: the same circular shear cell with variable packing fraction, which the authors say is under construction, would show whether $S_\\ell$ remains the control parameter as the system moves away from volume fraction 0.59."],"forward_implications":["The same dimensionless number $S_\\ell$, not the raw shear rate, should organize quaking statistics in any quasi-static granular shear geometry once the material's $V_*$ is known.","Quaking is a bulk collective phenomenon: at intermediate rates the rearranging particles form connected clusters that can span the whole sample, so the transition cannot be reduced to isolated local shear transformation zones.","At very low $S_\\ell$, quakes become rare and widely separated; below the lowest rate tested, quaking is undetectable within a strain window of about 0.7, so the material appears smooth again.","The stress-drop signature and the displacement-cluster signature mark the same transition: the peaks in higher moments of cluster size coincide with the peak in large-drop counts at intermediate rates.","Particle motion in a system-spanning quake is not coherent translation; the cluster contains patches moving along and against the local flow direction, implying internal rearrangement rather than rigid slip."],"supporting_citations":[{"why":"Provides the earlier double-cone experiment whose large-drop counts and rate-dependent stick-slip transition are the main comparison dataset.","marker":"[17]"},{"why":"Introduces the dimensionless shear rate S_l and the numerical demonstration that speed-dependent interparticle friction generates quaking.","marker":"[24]"},{"why":"Supplies the direct tribology measurements between PDMS particles in glycerol from which the threshold speed V* is determined.","marker":"[23]"},{"why":"Establishes the Stribeck-type lubrication background for the sharp drop in PDMS friction beyond a critical sliding speed.","marker":"[22]"},{"why":"Provides the Laguerre tessellation geometry used to define nearest neighbors and thus to build displacement clusters with unequal particle sizes.","marker":"[37]"}],"fun_headline_variants":["Soft grain quaking obeys one friction-scaled shear rate","One rate parameter unifies granular quaking transitions","Friction-scaled shear rate collapses soft-grain quaking","Granular quaking: one number ties rate to internal motion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The collapse onto one curve assumes that the friction threshold speed used in the analysis, 0.2 mm/s, was valid during the shearing runs, even though it was measured only after data collection and had drifted from 5 mm/s to 1 mm/s over the preceding two years.","fun_headline_variants_meta":{"raw":{"variants":["Soft grain quaking obeys one friction-scaled shear rate","One rate parameter unifies granular quaking transitions","Friction-scaled shear rate collapses soft-grain quaking","Granular quaking: one number ties rate to internal motion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000536,"raw_usage":{"total_tokens":2590,"prompt_tokens":976,"completion_tokens":1614,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":592,"completion_tokens_details":{"reasoning_tokens":1544}},"tokens_in":592,"tokens_out":1614,"duration_ms":13746,"temperature":1.0,"reasoning_tokens":1544,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:16:19.616037+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the large-drop count curves with $V_*$ determined in situ during the same runs, or with a fresh batch of particles whose threshold is stable; if the double-cone and circular-cell curves stop collapsing once the threshold is measured concurrently, the $S_\\ell$ unification rests on a post-hoc choice. A sharper check: repeat the circular-cell experiment with newly molded PDMS particles ($V_* \\approx 5$ mm/s) and verify that the quaking peak in the LD count shifts to an $S_\\ell$ value matching the old particles' peak.","supporting_citations":[{"cited_title":"(JC) Tsai, G.-H","cited_arxiv_id":null,"evidence_quote":"Provides the earlier double-cone experiment whose large-drop counts and rate-dependent stick-slip transition are the main comparison dataset."},{"cited_title":"Tsai, W.-C","cited_arxiv_id":null,"evidence_quote":"Introduces the dimensionless shear rate S_l and the numerical demonstration that speed-dependent interparticle friction generates quaking."},{"cited_title":"Tsai and J.-C","cited_arxiv_id":null,"evidence_quote":"Supplies the direct tribology measurements between PDMS particles in glycerol from which the threshold speed V* is determined."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the Stribeck-type lubrication background for the sharp drop in PDMS friction beyond a critical sliding speed."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Laguerre tessellation geometry used to define nearest neighbors and thus to build displacement clusters with unequal particle sizes."}],"review_version":1}