{"id":"4cf461ed-e539-4eaa-a1c8-7cff2e3e00d0","arxiv_id":"2411.13463","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new continuous-rotation shear protocol removes the reversible-irreversible transition in dense suspensions while leaving the rheology qualitatively unchanged.","lead":"This paper introduces a new way to shear dense particle suspensions, called rotary shear, where the flow direction rotates continuously rather than reversing back and forth. The key finding is that particles always wander diffusively under rotary shear, even though the suspension's viscosity curve looks the same as in ordinary oscillatory shear.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'always diffusive' claim for RS rests on 300-strain-unit runs from a single presheared, contact-bearing initial state; the absence of a dynamics-specific initial-condition check leaves open that an isotropic, contact-free start could self-organize into a reversible state at small strain…","rationale":"The paper is a well-executed computational study that introduces a genuinely novel protocol and provides a clean negative result. The rheological findings are carefully validated against experiments (Appendix D), and the RRS control strengthens the mechanistic interpretation. However, the headline dynamical claim is a universal negative ('no self-adsorbing states at any γ0, always diffusive') that is extrapolated from finite simulations with a single preparation. The initial-condition check in Appendix D covers only viscosity, not the MSD/Deff/Z that define the RIT, so the possibility of an isotropic contact-free initial state evolving reversibly is not excluded. This is load-bearing because the paper's own mechanism is contact-induced irreversibility; verifying the same conclusion from a contact-free start is necessary. A longer simulation (1000 strain units) would also test whether the 300-strain-unit data have converged. The concern is addressable and does not undermine the rheology results, so the conditional verdict remains appropriate; no stronger action is warranted.","tokens_in":18967,"tokens_out":10678,"duration_ms":122692,"concrete_test":"Run RS simulations at φ = 0.55, μc = 0.5 and φ = 0.40, μc = 0.0 for γ0 = 0.05, 0.1, 0.5, and 1.0, using two initial conditions: (i) the standard presheared state and (ii) a freshly randomized, isotropic, contact-free configuration (no preshear). Integrate for at least 1000 strain units, recording MSD, Deff, and Z as functions of γt. If the isotropic start at γ0 = 0.05 gives Deff → 0 (or MSD slope < 1 persisting beyond 300 units) and Z → 0, the central 'always diffusive' claim is initial-condition and time dependent. Conversely, if both starts converge to the same finite Deff and Z > 0, the claim is validated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (abstract; §3.2) is that RS suspensions are stroboscopically diffusive at every strain amplitude and never exhibit absorbing states. The evidence is MSD/Deff data from simulations run for at least γt = 300 (§2.1) starting from a presheared, anisotropic, contact-bearing configuration (§2.2). Appendix D (Fig. 15) verifies initial-condition independence only for the relative complex viscosity, not for the dynamical observables (MSD, Deff, Z) that define the RIT. Because the paper's own mechanism is that contacts cause irreversibility, it is circular to start every run with a contact-rich state and then conclude contacts are always present. A random, isotropic, contact-free initial condition at small γ0 might remain essentially collision-free and trace affine orbits that close after each period; the integrated RS strain over a cycle is zero, so absent interactions the stroboscopic positions would be identical. If such a state were dynamically reversible, the abstract's 'at any γ0' would be false, and the conclusion that rheological signatures (minimum viscosity, onset of N2) are not sufficient for RIT would need qualification. The finite 300-strain-unit horizon adds to this: Deff is extracted from a linear fit, and at the smallest γ0 the number of strain cycles is large but the per-cycle displacement is small, so a slow crossover to subdiffusive or absorbing behavior cannot be excluded from the reported data.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper introduces a rotary shear (RS) protocol for dense non-Brownian suspensions, in which two orthogonal oscillatory shears are imposed out of phase so that the shear direction rotates continuously, and compares it with classical oscillatory shear (OS) and a reversible variant (RRS). Using hybrid lubrication/granular dynamics simulations at volume fractions 0.40–0.55 and friction coefficients 0.0–0.5, the authors report that the complex viscosity and normal stress differences are qualitatively similar across protocols, with a non-monotonic viscosity and a minimum at an intermediate strain amplitude, but that the particle dynamics differ fundamentally: OS and RRS exhibit a reversible–irreversible transition, whereas RS is claimed to be stroboscopically diffusive at all strain amplitudes, with no absorbing state. The paper attributes this to the absence of shear reversal in RS, which leaves a persistent, contact-bearing anisotropic microstructure, and validates the OS rheology and diffusion data against Bricker & Butler (2006) and Pine et al. (2005).","tokens_in":19254,"tokens_out":12602,"duration_ms":142658,"significance":"If the central claim holds, the paper is significant because it demonstrates that rheological signatures previously associated with the reversible–irreversible transition—minimum complex viscosity and onset of second normal stress differences—are not sufficient conditions for the transition, thereby decoupling bulk rheology from dynamical phase behavior. The study also introduces a practically realizable protocol and an internal control (RRS) that helps isolate the role of sudden shear reversal. Strengths include validation against two independent datasets, the absence of fitted target quantities, and consistency between the stress decomposition, coordination number, and observed dynamics. The main uncertainties concern the support for the strong 'always diffusive at any γ0' claim.","major_comments":[{"comment":"The claim that suspensions under RS 'do not show self-adsorbing states at any γ0' is established only for the presheared, contact-rich initial condition described in §2.2. Appendix D verifies initial-condition independence only for the relative complex viscosity; the dynamical observables that define the RIT—MSD, Deff, and Z—are never tested from a different initial state. This is load-bearing because the affine RS deformation over a full cycle is the identity: the velocity-gradient tensor L(t) is nilpotent with L(t)L(s)=0, so a collision-free trajectory would close stroboscopically every period. The authors' own mechanism attributes RS irreversibility to persistent contacts, and the simulations begin every run with contacts present; a dynamics-specific test from an initially isotropic, contact-free configuration at small γ0 is therefore required to rule out a reversible branch. Without such a test, the 'at any γ0' assertion is not established, and the conclusion that rheological signatures are not sufficient for RIT may need qualification.","section":"§3.2 and Appendix D (Fig. 15)"},{"comment":"The 'always diffusive' conclusion rests on single 500-particle runs of 300 strain units, with Deff obtained from a linear fit over an unspecified window. At γ0=0.05, 300 strain units corresponds to roughly 955 cycles, but the per-cycle displacement is small and the MSD data are shown over only about 1.5 decades in strain; a slow crossover to subdiffusive or absorbing behavior beyond the simulated window cannot be excluded from the reported data. The authors should provide error bars or a second independent run for representative cases, specify the fitting window used for Deff, and perform a longer-time check (or a finite-time scaling analysis) for at least one small and one intermediate γ0 under RS.","section":"§3.2, Figs. 7–8"}],"minor_comments":[{"comment":"The term 'self-adsorbing states' is nonstandard and likely intended to mean 'absorbing states' or 'self-organized reversible states'; the wording should be corrected for clarity.","section":"Abstract and §3.2"},{"comment":"The mean squared displacement formula is typeset as '[Δr(γt)/d]^2 = 6Deffγt'; it should use an ensemble average notation such as ⟨|Δr(γt)|^2⟩/d^2.","section":"Eq. (2.15)"},{"comment":"The sign convention for the rotation angle θ is inconsistent: §2.1 states θ=ωt, while Appendix C states that θ is negative for the clockwise RS rotation. This could affect the sign of the elastic component η′′ and should be clarified.","section":"§2.1 and Appendix C"},{"comment":"The caption states that the figure is shown for φ=0.40 and μc=0.0, but the figure contains multiple panels for OS and RS at different strain amplitudes; each panel's parameters should be identified.","section":"Fig. 10"},{"comment":"The phrase 'a total strain of ˙γt = 40' uses an odd notation; the accumulated strain should be written as γt=40 to distinguish it from the shear rate.","section":"§2.2"},{"comment":"The sentence 'Our RS and RSS protocols...' contains a typo: 'RSS' should be 'RRS'.","section":"§4 (Conclusion)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope and the topic is timely. The main weakness is the load-bearing support for the null dynamical claim: the initial-condition check covers only rheology, and the finite-time, single-run nature of the simulation data leaves the 'always diffusive at any γ0' conclusion under-supported. I would ask the authors to add a dynamics-specific initial-condition test and to quantify the statistical and finite-time uncertainty of Deff. The RRS control and the validation against Pine et al. and Bricker & Butler are strong points."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"I read this with some care because of the reversible-irreversible transition. The new thing is the rotary shear protocol: two out-of-phase orthogonal oscillatory shears that rotate the shear direction continuously. That sounds like a small variation, but it makes a clean point. In OS you get the usual RIT, while RS preserves the non-monotonic viscosity and normal-stress signatures but the particle dynamics stay diffusive at every amplitude they tested. The RRS protocol, which adds an instantaneous reversal to RS, restores RIT. That control is what makes the paper worth reading: it isolates the role of shear reversal in the absorbing state and gives a concrete counterexample to the idea that a viscosity minimum plus finite N2 implies RIT.\n\nThe simulation work is careful. The complex viscosity is validated against Bricker & Butler, the dynamics against Pine et al. The stress decomposition and coordination-number analysis are consistent. The microstructure story—OS and RRS go from contact-free, isotropic to in-contact, anisotropic at the transition, while RS is always in contact and anisotropic—fits the dynamics.\n\nThe soft spot is the strength of the universal claim. 'Always diffusive' is inferred from 300 strain units, starting from one presheared state, with the initial-condition check done only for the rheology. That is a real gap. The stress-test scenario of a random, contact-free initial state that might stay reversible at small γ0 is not obviously impossible, though I think it is unlikely: at φ≥0.40 even small strains produce collisions in a dense system, and once you have collisions, the RS protocol has nothing that returns particles to their prior positions. Still, the authors should either run that check or soften the wording. The lack of error bars and single runs also bothers me a little, but the trends are clear.\n\nThere is no code or data deposited, which makes verification harder. Minor issue: the Data Availability Statement is just a heading.\n\nThis paper deserves a serious referee. It is not a breakthrough outside the subfield, but it changes how we think about the rheology–dynamics connection and gives a new protocol to play with. I would bring it to the reading group and would cite it if I write anything on RIT in the next year.","headline":"A genuinely new shear protocol with a clean decoupling result; the 'always diffusive' claim is somewhat stronger than the evidence.","tokens_in":19796,"tokens_out":4299,"would_cite":true,"duration_ms":49144,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76T20"],"pacs":["47.57.Qk","83.80.Hj"],"model":"deepseek-v4-flash","headline":"Rotary shear makes dense suspensions always diffusive, and viscosity minima no longer signal a reversible–irreversible transition.","keywords":["dense suspensions","rotary shear","oscillatory shear","reversible-irreversible transition","non-Brownian suspensions","microstructure","normal stress differences","shear reversal"],"falsifier":"Extend the RS simulation at $\\phi=0.55$, $\\mu_c=0.5$, and $\\gamma_0=0.05$ to at least $10^4$ strain units from both presheared and random isotropic initial conditions, and track the stroboscopic MSD and coordination number; the claim that RS is always diffusive collapses if the MSD slope drops below 1 or $Z$ falls to zero at late times.","tokens_in":18772,"feed_emoji":"🔄","tokens_out":8366,"duration_ms":80060,"temperature":0.7,"pith_summary":"This paper introduces rotary shear (RS), a periodic flow in which two orthogonal oscillatory shears are applied out of phase so that the shear direction rotates continuously about the velocity-gradient axis while the shear-rate magnitude stays constant. Using simulations of dense non-Brownian suspensions, the authors try to establish that RS separates rheology from dynamics: although the suspension viscosity shows the same non-monotonic strain-amplitude dependence and the same onset of second normal stress differences as in oscillatory shear (OS), the particles never organize into a reversible absorbing state. At every strain amplitude tested, stroboscopic particle motion is diffusive, because the absence of sudden shear reversal keeps the microstructure in contact and anisotropic at all times. A reversible variant (RRS) that reverses the rotation each half-cycle behaves like OS and recovers the reversible–irreversible transition. The upshot is that viscosity minima and normal-stress onsets are not sufficient evidence for reversible–irreversible transitions; the time-reversibility of the driving protocol is the controlling factor.","feed_headline":"Rotary shear erases the reversible–irreversible transition","feed_subtitle":"Viscosity minima and N2 onset still appear, so rheology alone cannot signal reversible dynamics.","key_machinery":"The carrying object is the rotary-shear rate-of-strain tensor $$$E^{{\\infty}}$_{\\text{RS}} = \\begin{pmatrix} 0 & 0 & \\dot{\\gamma}_{xz}/2 \\\\ 0 & 0 & \\dot{\\gamma}_{yz}/2 \\\\ \\dot{\\gamma}_{zx}/2 & \\dot{\\gamma}_{zy}/2 & 0 \\end{pmatrix}$$ with $\\dot{\\gamma}_{xz} = \\gamma_0\\omega \\sin(\\omega t)$ and $\\dot{\\gamma}_{yz} = \\gamma_0\\omega \\cos(\\omega t)$. This imposes shear of constant magnitude whose flow–vorticity plane rotates around the gradient direction without any sudden reversal, so the protocol is not time-reversible. The control protocol RRS reverses both shears and the rotation direction every half-cycle. The argument is carried by the microstructure diagnostics: coordination number $Z$ and pair distribution $g(h,\\theta)$ show that OS/RRS absorbing states are contact-free and isotropic at low amplitude, while RS is in contact and anisotropic at all amplitudes; persistent contacts produce collisions and hence diffusion.","core_discovery":"The paper establishes that rotary shear, a periodic drive with constant shear-rate magnitude but continuously rotating direction, produces diffusive particle dynamics at every strain amplitude tested, for volume fractions 0.40 to 0.55 and friction coefficients 0, 0.2, and 0.5. Unlike oscillatory shear, no reversible absorbing state forms under RS: the stroboscopic mean squared displacement grows linearly with strain, the effective diffusivity remains finite even at the smallest amplitude, and the coordination number stays positive. The microstructure is always in contact and anisotropic, whereas OS and RRS show a contact-free, isotropic microstructure at low amplitudes and a transition to an in-contact, anisotropic state above the critical amplitude. From this, the paper concludes that rheological markers such as a minimum viscosity and the onset of the second normal stress difference are not sufficient conditions for a reversible–irreversible transition; the absence of shear reversal is what keeps RS dynamics permanently diffusive.","pith_inferences":["Editorial inference: because RS keeps particles diffusive at arbitrarily small strain amplitudes, it offers a clean protocol for measuring shear-induced diffusion in dense suspensions without an absorbing state masking the signal.","Editorial inference: the empirical $\\gamma_{0,c} \\sim \\phi^{-2}$ scaling reported for OS is tied to reversal-based protocols; under RS the transition disappears, so the scaling is not a universal property of periodic shear.","Editorial inference: varying the reversal angle in RRS between 0 and $\\pi$ should interpolate continuously between RS-like and OS-like dynamics, giving a quantitative measure of how much shear reversal is needed to nucleate an absorbing state."],"forward_implications":["Rheological signals alone cannot certify reversible dynamics: a suspension can show a viscosity minimum and the onset of $N_2$ without undergoing a reversible–irreversible transition.","Shear reversal is the controlling ingredient for absorbing states; a periodic drive that merely rotates the shear direction keeps particles diffusive at every amplitude.","The RRS protocol restores OS-like dynamics, showing that the suppression of RIT comes from the absence of sudden reversal, not from rotation itself.","For processing applications, RS provides continuous diffusive transport even at tiny strain amplitudes, which could be used for mixing without large deformations.","The microstructure criterion—contact-free and isotropic for reversible states versus in-contact and anisotropic for diffusive states—holds across all three protocols."],"supporting_citations":[{"why":"Defines the reversible–irreversible transition and the empirical critical-amplitude scaling used as the benchmark for OS dynamics.","marker":"(Pine et al. 2005)"},{"why":"Claims viscosity minimum and N2 onset coincide with RIT in OS; this paper tests whether those markers are sufficient.","marker":"(Ge & Elfring 2022)"},{"why":"Experimental non-monotonic OS viscosity used to validate the simulation rheology.","marker":"(Bricker & Butler 2006)"},{"why":"Random-organization mechanism that produces absorbing states; cited as the reason contacts prevent reversibility.","marker":"(Corte et al. 2008)"},{"why":"Sudden rotation of shear axes lowers viscosity; provides the prior protocol that RS makes continuous.","marker":"(Blanc et al. 2023)"},{"why":"Alternating shear rotations reduce dissipation; RS and RRS extend this to continuous rotation with reversal.","marker":"(Acharya & Trulsson 2024)"},{"why":"Supplies the hybrid lubrication/granular dynamics force model used in the simulations.","marker":"(Cheal & Ness 2018)"},{"why":"Provides the model implementation and lubrication resistance expressions underlying the simulation method.","marker":"(Ge & Brandt 2020)"}],"fun_headline_variants":["Rotary shear kills the reversible-irreversible transition","Rotary shear: no reversible absorbing states at any amplitude","In rotary shear, dynamics stay diffusive without exception","Rotary shear abolishes the transition to reversible motion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that 300 strain units of simulation are enough to reveal the long-time dynamical state, so no reversible absorbing state appears later under RS, and that the presheared starting configuration used for the dynamics is representative.","fun_headline_variants_meta":{"raw":{"variants":["Rotary shear kills the reversible-irreversible transition","Rotary shear: no reversible absorbing states at any amplitude","In rotary shear, dynamics stay diffusive without exception","Rotary shear abolishes the transition to reversible motion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000198,"raw_usage":{"total_tokens":1410,"prompt_tokens":1030,"completion_tokens":380,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":646,"completion_tokens_details":{"reasoning_tokens":316}},"tokens_in":646,"tokens_out":380,"duration_ms":4643,"temperature":1.0,"reasoning_tokens":316,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:22:48.527073+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Extend the RS simulation at $\\phi=0.55$, $\\mu_c=0.5$, and $\\gamma_0=0.05$ to at least $10^4$ strain units from both presheared and random isotropic initial conditions, and track the stroboscopic MSD and coordination number; the claim that RS is always diffusive collapses if the MSD slope drops below 1 or $Z$ falls to zero at late times.","supporting_citations":[{"cited_title":"Nature 438 (7070), 997--1000","cited_arxiv_id":null,"evidence_quote":"Defines the reversible–irreversible transition and the empirical critical-amplitude scaling used as the benchmark for OS dynamics."},{"cited_title":"Physical Review E 106 (5), 054616","cited_arxiv_id":null,"evidence_quote":"Claims viscosity minimum and N2 onset coincide with RIT in OS; this paper tests whether those markers are sufficient."},{"cited_title":"Journal of rheology 50 (5), 711--728","cited_arxiv_id":null,"evidence_quote":"Experimental non-monotonic OS viscosity used to validate the simulation rheology."},{"cited_title":"Nature Physics 4 (5), 420--424","cited_arxiv_id":null,"evidence_quote":"Random-organization mechanism that produces absorbing states; cited as the reason contacts prevent reversibility."},{"cited_title":"Physical Review Research 6 , 033327","cited_arxiv_id":null,"evidence_quote":"Alternating shear rotations reduce dissipation; RS and RRS extend this to continuous rotation with reversal."},{"cited_title":"Journal of rheology 62 (2), 501--512","cited_arxiv_id":null,"evidence_quote":"Supplies the hybrid lubrication/granular dynamics force model used in the simulations."},{"cited_title":"Implementation note on a minimal hybrid lubrication/granular dynamics model for dense suspensions","cited_arxiv_id":"2005.12755","evidence_quote":"Provides the model implementation and lubrication resistance expressions underlying the simulation method."}],"review_version":1}