{"id":"d42cae9d-8b78-489e-a066-9fbf52a1d715","arxiv_id":"2607.13997","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"A lattice with chiral gears driven at its boundaries exhibits an odd shear modulus, enabling non-conservative work and non-Hermitian skin-effect wave localization without electronics or feedback.","lead":"This paper shows in computer simulations that a frame of square cells with toothed chiral gears at its edges can act like a material with 'odd elasticity' — where pulling it one way makes it twist unexpectedly, but twisting it does not pull it back. The design needs no electronics or sensors, only continuously rotating gears, and could lead to mechanical devices that direct vibrations or produce energy from cyclic loading.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"A is measured from monotonic ratcheting steps and the paper's own text admits gear-phase dependence; without reverse-loading/phase tests, A is not established as a material constant, so the non-conservative work and NHSE claims are unsupported.","rationale":"The reader's weakest assumption already flagged that A must be independent of driving phase, frequency, amplitude, friction, clearance, and strain-step size. My concern is the same broad issue but sharpened to a specific and testable defect: the measurement protocol is monotonic and the mechanism is a gear-tooth ratchet, so the response may be plastic/hysteretic rather than an odd elastic modulus. The paper's own sentence in Section B admitting phase-dependence, combined with the absence of any reverse-loading test, makes this the most load-bearing gap. The non-conservative work and NHSE results are purely derived from the fitted A; if A is not a material constant, those predictions do not follow from the simulations. This does not necessarily invalidate the work—additional FE tests could show that the response is reversible and phase-independent—but until they are performed, conditional acceptance is appropriate. The reader's verdict remains CONDITIONAL, so I do not change the final classification, but I recommend making the reverse-loading and phase-sweep tests explicit conditions for acceptance. I disagree slightly with the reader's emphasis: they list many parameters, whereas the decisive issue is path-dependence and the absence of any unloading data.","tokens_in":7235,"tokens_out":4650,"duration_ms":52881,"concrete_test":"In the same FE model used for Fig. 2, apply a full closed strain cycle in normal strain: after reaching N positive increments of Δϵyy = +0.04%, apply the same increments with negative sign (−0.04%) back to zero, continuing the periodic gear driving throughout, and record the time-averaged σyx after each step. If σyx returns to its initial value (or follows the same slope with opposite sign), the constitutive tensor may be valid; if σyx retains a residual offset or evolves nonlinearly on unloading, A is a ratchet/plastic increment rather than an elastic modulus. Separately, repeat the monotonic loading protocol with the gears rotated by half a gear-tooth pitch before strain application; if A changes significantly in magnitude or sign, the paper's own phase-dependence statement applies and Eq. (2) is not a material property.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central quantity is the odd shear modulus A = Δσyx/Δϵyy defined in Eq. (3) and used to construct the asymmetric elasticity tensor in Eq. (2). In Section B, A is obtained by applying only positive increments of normal strain (Fig. 2(b)) and measuring the time-averaged shear stress after each step. The mechanism is explicitly a ratchet: the gear 'locks in a new equilibrium position' by jumping to the next tooth, and the paper states that 'depending on the phase of the gear with respect to the wall tied to the metamaterial, application of normal strain can only result in either no jump (and no odd shear strain), or odd shear occurring in only one direction.' This phase dependence means the measured A may depend on the gear's angular position at the moment the strain step is applied, and no evidence is provided that the response is reversible or even well-defined for negative strain increments. The constitutive tensor in Eq. (2) and the closed-cycle work calculation in Section C assume that time-averaged stresses are a linear, path-independent (up to the odd term) function of total strains. If the shear increment is instead a plastic offset accumulated once per step, then A is not an elastic modulus, and Eq. (6) would predict non-conservative work of the same trivial kind as any hysteretic plastic material, not the odd-elasticity effect claimed. The paper defers robustness to the SI, but the main text already contains a direct admission of phase-dependence, which is exactly the condition needed for A to be protocol-dependent. Without demonstrating that A is reproducible under reversed loading and independent of gear phase, the central claim that this is an odd elastic solid is not supported.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The authors propose a metamaterial made of a square lattice with chiral gears attached at top and bottom, driven by continuous oscillatory gear rotation. Under stepwise normal strain, the gears ratchet to new tooth positions, producing an increment of time-averaged shear stress. From this they extract an odd shear modulus A = Δσyx/Δεyy = 3125 GPa (Eq. 3), giving an asymmetric elasticity tensor (Eq. 2). The paper then argues analytically that this A yields non-conservative work in a closed normal/shear strain cycle (Section C) and, through the resulting non-Hermitian dynamical matrix, the non-Hermitian skin effect (Section D). The central reported result is a single FEM demonstration, with robustness, parameter sensitivity, and mesh-convergence details deferred to an SI that is not included in the manuscript.","tokens_in":7639,"tokens_out":4976,"duration_ms":60675,"significance":"If the measured A were established as a genuine, protocol-independent effective modulus, this would be a notable advance: it would show driven odd elasticity in a passive mechanical structure without electronic feedback, with concrete static and dynamic consequences. The proposed gear mechanism is physically plausible, and Sections C and D correctly work out the consequences of an elasticity tensor of the form of Eq. (2). However, the paper's own text in Section B admits a phase dependence that undermines the material-constant interpretation of A, and the later sections use that same A rather than providing independent validation. The significance therefore depends entirely on whether the missing SI and additional controls establish A as a well-defined material property.","major_comments":[{"comment":"A is measured from a single monotonic sequence of positive normal-strain increments. The main text itself states that 'depending on the phase of the gear with respect to the wall tied to the metamaterial, application of normal strain can only result in either no jump (and no odd shear strain), or odd shear occurring in only one direction.' This directly implies that the measured Δσyx/Δεyy is phase- and history-dependent. No reversed loading, no phase sweep, no frequency/amplitude sweep, and no step-size convergence study are reported in the main text. Until such tests show that A is independent of these control parameters, the time-averaged constitutive tensor in Eqs. (1)-(2) is not established, and all subsequent claims that rely on A are unsupported.","section":"Section B, Eq. (3)"},{"comment":"The non-conservative work and the non-Hermitian skin effect are not independent tests of the mechanism; they are analytic consequences of substituting the fitted A into Eq. (2). For example, Eq. (6) gives a nonzero closed-cycle work for any constitutive relation of the form Eq. (4), including a purely plastic or ratcheting material with an asymmetric incremental response. The paper does not report a direct FEM simulation of the full closed strain cycle on the gear-lattice system, nor a direct dynamic simulation of the finite metamaterial with open boundaries. Without such simulations, the paper has not shown that the metamaterial actually produces non-conservative work or the NHSE; it has only shown that if A is a material constant, these effects follow.","section":"Sections C and D, Eqs. (6) and (8)"},{"comment":"The robustness claims are repeatedly deferred to an SI that is not included in the submitted manuscript. The main text does not provide the mesh density, element type, time-step size, averaging window, or convergence criteria for the ABAQUS simulations, and no uncertainty is attached to the reported A = 3125 GPa. Given that this value is more than an order of magnitude larger than the Young's modulus E = 300 GPa, and that the mechanism is discrete (one tooth jump per strain increment), the step-size dependence and averaging procedure are load-bearing. The authors should either include the SI or summarize the required convergence/sensitivity data in the main text.","section":"General (Section B, Fig. 2)"}],"minor_comments":[{"comment":"The notation uses both σyx/σxy and εyx/εxy as independent components. Please clarify whether εxy and εyx are engineering shear strains or tensor shear strains, and how they are defined in terms of displacement gradients. As written, the repeated G entries in C_e are ambiguous and affect the derivation of Eq. (8).","section":"Equations (1)-(2)"},{"comment":"The text says the time-averaged shear stress 'converges to a constant after a few periods.' Please specify the averaging window, the number of periods, and the convergence tolerance used to define the equilibrium value.","section":"Section B, Fig. 2(c)"},{"comment":"It is unclear whether the strain loop in Fig. 3 is a schematic of Eq. (4) or an actual simulated path. Please state explicitly which, and if it is analytic, note that the FEM model was not used to verify the work integral.","section":"Section C, Fig. 3"},{"comment":"References [13] and [26] appear to refer to the same paper; please check and consolidate. Also, a brief comparison with the driven odd elasticity model of Huang et al. [27] in the main text (not just the introduction) would help the reader understand what is new.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The key issue is whether the missing SI can address the phase-dependence admission in Section B. If the SI includes reverse loading, phase sweeps, frequency/amplitude sweeps, step-size convergence, and direct closed-cycle and open-boundary simulations, the paper could be viable. Otherwise, the current manuscript reports only a single ratcheting measurement, and the derived consequences are circular rather than predictive. The editor should request the SI before further consideration."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Punchline: the chiral-gear ratchet is a genuinely new mechanism, the FEM demonstration is internally consistent, and the paper is honest — but the odd modulus A = 3125 GPa is not yet established as a material constant, and the statics and dynamics claims ride entirely on that. I agree with your conditional verdict, with two shifts in emphasis.\n\nWhat's new and good: the mechanism directly targets the open challenge from Huang et al. [27] — driven odd elasticity without electronics or feedback. Physically it makes sense: each normal-strain step detaches the gears, they re-contact at the next tooth, and the lattice picks up a shear increment. The paper also states in the main text, not in the SI, that depending on gear phase a strain step can produce no jump at all or shear in only one direction. That admission is the key to the soft spot.\n\nThe load-bearing question is whether A is a modulus or a ratchet artifact. A comes from monotonic positive strain steps only; there is no reverse-loading test, no phase sweep, no step-size or frequency dependence shown. If the shear offset is a one-way configurational change, Eq. (6)'s non-conservative work is the ordinary work of a hysteretic plastic material, and the difference between that and odd elasticity is exactly whether reversing the cycle reverses the work sign. The stress-test note is right on the facts; I'd only soften 'unsupported' to 'not established.' The consequences do follow if A is a true effective modulus — the missing piece is evidence that it is.\n\nThe circularity point is real but should be stated precisely. Sections C and D don't validate A; they compute what follows from inserting it into the tensor. That is a standard effective-medium move, not a logical circle. The bigger gap is dynamics: the time-averaged modulus is assumed to govern the instantaneous equations of motion, and no dynamic simulation of the actual metamaterial confirms the skin effect. The NHSE is a prediction of the effective continuum model, not a demonstrated property of the mechanism.\n\nMinor: 'passive' in the title overstates — the components are passive, the system is externally driven. No mesh convergence or UQ in the main text; robustness is deferred to an SI that is not in the version we have.\n\nBottom line: a serious, honest, citable proposal of a new mechanism whose central premise needs one decisive test — reverse loading at controlled gear phase. Send it to peer review; a good referee will ask exactly the right questions.","headline":"A genuinely new chiral-gear ratchet mechanism for driven odd elasticity, plausibly demonstrated in FEM — but A is measured from a single monotonic protocol, the main text admits gear-phase dependence, and no reverse-loading test establishes that A is a true material constant.","tokens_in":8137,"tokens_out":8730,"would_cite":true,"duration_ms":91843,"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":"Periodically driven chiral gears make a square-lattice metamaterial respond as an odd elastic solid, with a one-way shear modulus of 3125 GPa that enables non-conservative work and boundary-localized skin modes.","keywords":["odd elasticity","mechanical metamaterial","chiral gears","non-reciprocity","non-Hermitian skin effect","non-conservative work","passive components","driven solids"],"falsifier":"Measure Δσ_yx/Δϵ_yy under two different gear initial phases or two driving frequencies; if the ratio changes, A is protocol-dependent and the non-conservative-work and skin-effect predictions do not follow. Alternatively, run a closed normal-shear strain cycle and measure the net work; if the net work vanishes once frictional dissipation is subtracted, the ratcheting response is dissipative, not non-conservative elastic work.","tokens_in":7117,"feed_emoji":"⚙️","tokens_out":4838,"duration_ms":43140,"temperature":0.7,"pith_summary":"The paper claims that a square-lattice metamaterial fitted with chirally toothed gears, driven by a periodic rotation at its boundaries, behaves as an odd elastic solid. The key result is an odd shear modulus A = 3125 GPa that connects normal strain to shear stress while the converse connection is absent, making the effective elasticity tensor asymmetric. Because of this asymmetry, a closed cycle of normal and shear strain produces net non-conservative work, and the lattice's equations of motion become non-Hermitian, yielding boundary-localized skin modes with one-way amplification. The structure is entirely passive: no internal energy source, electronic feedback, or robotic control is required, the driving motion itself supplying the energy. If the claim holds, odd elasticity moves from active matter and feedback-controlled devices to ordinary mechanical metamaterials.","feed_headline":"Passive chiral gears create a 3125 GPa one-way shear response","feed_subtitle":"A driven chiral-gear lattice couples tension to shear in one direction, enabling non-conservative work and boundary-localized waves without","key_machinery":"The central object is the chiral gear attached to the lattice boundary, whose asymmetric tooth geometry (contact lengths l1 ≠ l2) and clearance δ produce a one-way mechanical ratchet. During a normal strain step, the gear momentarily separates from the wall and re-engages at the next tooth, locking a new shear displacement into the lattice; repeated strain steps accumulate shear in a single direction regardless of gear phase. The driven periodic rotation provides the energy input, and the time-averaged response over a driving period defines the effective elastic constants. The gear tooth length l sets the shear increment per strain step, giving a design handle on A.","core_discovery":"The central discovery is that a ratcheting gear mechanism converts each increment of normal strain into a locked-in increment of shear strain, so that the time-averaged shear stress grows linearly with normal strain. The authors express the response as C = C_e + C_o, where C_e is the isotropic symmetric part (bulk modulus B = 214.3 GPa, shear modulus G = 115.4 GPa) and C_o contains a single non-zero odd modulus A = Δσ_yx/Δϵ_yy = 3125 GPa. This non-reciprocal term means extension produces shear while shear produces no extension, breaking Maxwell-Betti reciprocity. The mechanism relies on the chiral gear teeth's unequal contact lengths (l1 and l2) to bias the shear direction when the gear dise","pith_inferences":["The main-text evidence for A rests on a single driving protocol; whether A stays constant under changes of driving frequency, amplitude, friction coefficient, or strain-step size is deferred to the SI, so the 'material constant' status of A is the point to scrutinize.","A direct test of reciprocity would be to apply shear strain first and measure normal stress; if any normal stress appears, the tensor is not simply the claimed C_o.","The non-conservative work calculation treats the strain cycle as closed in strain space, but the gear re-engagement involves friction and dissipation; separating genuine odd work from frictional loss would clarify the energetics.","The same chiral-ratchet concept could be ported to other lattices or to three dimensions, potentially producing odd bulk responses or odd terms in other stress components; a topology search over gear tooth shapes could optimize A."],"forward_implications":["Odd elasticity can be realized with passive components, eliminating the need for embedded energy sources, feedback loops, or robotic controllers.","The large odd modulus (A = 3125 GPa, about 27 times the shear modulus) means non-conservative work per strain cycle is substantial and should be measurable.","A normal-strain input is converted into one-way shear output, so the structure acts as a mechanical rectifier or strain-direction valve.","The predicted non-Hermitian skin effect makes the lattice a passive directional amplifier/attenuator for elastic waves at open boundaries.","Because A depends on gear tooth geometry, the odd response can be tuned by design rather than by external control parameters."],"fun_headline_variants":["Passive chiral gears yield a 3125 GPa one-way shear","Nonreciprocal shear from gears: passive odd elasticity","Gears only: passive metamaterial shows odd elasticity","3125 GPa one-way shear via passive chiral gear lattice","Odd elasticity without electronics: chiral gear lattice"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The odd shear modulus A is a genuine time-averaged material constant of the driven metamaterial, independent of the gear's initial phase, the driving frequency and amplitude, the friction coefficient, the clearance, and the size of the normal strain increments.","fun_headline_variants_meta":{"raw":{"variants":["Passive chiral gears yield a 3125 GPa one-way shear","Nonreciprocal shear from gears: passive odd elasticity","Gears only: passive metamaterial shows odd elasticity","3125 GPa one-way shear via passive chiral gear lattice","Odd elasticity without electronics: chiral gear lattice"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000215,"raw_usage":{"total_tokens":1217,"prompt_tokens":644,"completion_tokens":573,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":388,"completion_tokens_details":{"reasoning_tokens":493}},"tokens_in":388,"tokens_out":573,"duration_ms":5313,"temperature":1.0,"reasoning_tokens":493,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T03:03:43.649302+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure Δσ_yx/Δϵ_yy under two different gear initial phases or two driving frequencies; if the ratio changes, A is protocol-dependent and the non-conservative-work and skin-effect predictions do not follow. Alternatively, run a closed normal-shear strain cycle and measure the net work; if the net work vanishes once frictional dissipation is subtracted, the ratcheting response is dissipative, not non-conservative elastic work.","supporting_citations":[],"review_version":1}