{"id":"6bc882fe-a717-4327-a180-a840b03a8ec5","arxiv_id":"1908.05454","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"Using X-ray CT and material-line tracking, the authors report periodic radial shear strain and strain concentration near faults in compressed sandstone.","lead":"This paper uses X-ray CT scans of a small sandstone sample to measure how individual mineral grains moved and deformed under compression, and then uses material lines to map shear strain in the rock. The authors report that shear strain repeats periodically in the radial direction and concentrates near the fracture zone.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The shear-strain periodicity claim rests on the unvalidated equation θ = (ds0 − ds)/ds0; the paper's own statement that the measurement error 'cannot be estimated' (§3.4.3) is the weakest link.","rationale":"The reader's weakest_assumption is precisely the unvalidated mapping from material-line rotation to local shear strain, coupled with the admitted unquantified error. My analysis agrees: this is the load-bearing point because every headline claim—periodicity, local concentration, fracture mechanism—is derived from the shear strain maps in Fig. 16. The rejection verdict is justified by the combination of (a) an unexplained definitional jump from 'local rotation angle' to length-change ratio, (b) an explicit in-text concession that the error cannot be estimated, and (c) absence of any validation, synthetic or otherwise, that the measurement pipeline recovers a known strain field. The strongest_claim's periodicity assertion is therefore unsupported by the evidence presented. I did not find a separate, independent fatal flaw; rather, the single most load-bearing weakness is the unvalidated measurement, which undermines all downstream interpretations. The paper does contain useful descriptive observations of grain movements, but the quantitative and physical conclusions cannot be accepted without addressing this gap. A conditional acceptance would require the proposed calibration experiment; as written, rejection with an invitation to resubmit after validation is the appropriate verdict.","tokens_in":12346,"tokens_out":800,"duration_ms":10113,"concrete_test":"Perform a synthetic-deformation calibration: take the pre-loading CT volume, apply a known rigid-body rotation or a known uniform shear strain field to the gray-scale image, re-segment and track material lines with the same algorithm, and compare the recovered θ with the applied shear. If the recovered field differs by more than the amplitude of the claimed periodic oscillations (approximately 0.3–0.6 in Fig. 16), the periodicity claim is within noise and should be withdrawn or substantially qualified. In addition, compute the radial power spectrum of θ(r) for the four profiles in Fig. 17 and compare it with the spectrum of a random-phase null ensemble; report a p-value or confidence interval for the claimed periodicity.","verdict_should_be":"REJECT","load_bearing_attack":"The central claim is that shear strain is periodic in the radial direction and that positive/negative shear strains concentrate locally. This claim depends entirely on the material-line angle change being a faithful proxy for local shear strain. The paper defines shear strain as θ = (ds0 − ds)/ds0 in Eq. (11), but the derivation is not shown: a material line's length change does not, by itself, equal shear strain in a deformation field, and the text switches between angle and length measures without justification. Moreover, the matching of material lines before and after loading across segmentation is not validated. The authors write in §3.4.3: 'due to the problem of image accuracy and material point recognition rate, a certain error is caused, and the error cannot be estimated.' This explicit admission means the observed 'periodic' variation of θ along a radial profile could be an artifact of tracking noise, segmentation boundaries, or the chosen line-selection criteria (§3.4.1 lists 'material line is periodic' and 'length less than half the spatial wavelength' as selection rules, which could pre-select oscillating lines). The periodicity itself is also only asserted from visual inspection of profiles in Fig. 17; no statistical test, Fourier analysis, or comparison against a null random field is provided. Without bounding the tracking error and without a quantitative periodicity test, the headline observation is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an in-situ X-ray computed tomography study of a 3.8 mm × 7.4 mm sandstone cylinder under uniaxial compression, with scans taken before loading and after unloading at 4.6 μm resolution. It tracks 689 mineral grains, computes grain translations, rotations, principal strains, and stresses using a deformation gradient constructed from grain bounding-box axes, and introduces material lines on CT slices to measure a 'Stokes local rotation' interpreted as shear strain. The headline findings are that shear strain is periodic in the radial direction, that positive and negative shear strains concentrate locally, and that fracture is caused by shear stress concentration near the fracture zone.","tokens_in":12603,"tokens_out":4829,"duration_ms":44191,"significance":"If the methodology were sound, the paper would provide a valuable three-dimensional grain-scale view of sandstone deformation and a new material-line-based approach to measuring local shear strain. The strengths of the work include the in-situ XCT acquisition, the explicit use of a radial distribution function to check grain registration, and the attempt to separate translation, rotation, and strain at the grain scale. However, the central claims are not currently supported: the quantity labeled shear strain in Eq. (11) is a relative length change rather than a shear strain or rotation angle, the measurement error is admitted to be unquantified in §3.4.3, and the stress computation from residual strains via elastic constants is invalid. These are load-bearing issues affecting the main conclusions, not presentation defects.","major_comments":[{"comment":"The definition θ = (ds0 − ds)/ds0 is a relative change in length of a material line, i.e., a normal strain component, not a shear strain or a rotation angle. No derivation is provided that connects the change in length of a material line to the local shear strain, and the text switches between angle-based and length-based measures without justification. Because the paper's central periodicity claim is formulated in terms of this θ, the main quantitative result is not established.","section":"Section 3.4.1, Eq. (11)"},{"comment":"The authors state explicitly that 'the error cannot be estimated' for the material-line measurements. Without an error bound or an independent validation of the material-line tracking (for example, against a synthetic deformation field or an independent measurement technique), the observed periodic oscillations in the shear-strain profiles in Fig. 17 cannot be distinguished from tracking noise, segmentation boundary effects, or artifacts introduced by the line-selection rules in §3.4.1, which include selecting 'periodic' material lines and lines shorter than half the spatial wavelength. This is a load-bearing gap for the headline observation.","section":"Section 3.4.3"},{"comment":"The grain stress tensor is computed by applying isotropic linear elasticity with Lamé constants to strain components measured on grains after unloading. Since the sample has undergone inelastic, plastic, and damage-related deformation, residual strains cannot be converted to stresses using elastic constants. Consequently, the reported stress magnitudes and the conclusion of 'shear stress concentration near the fracture zone' are not supported by the presented analysis.","section":"Section 3.3.1, Eqs. (7)-(8)"},{"comment":"The claim that shear strain is periodic in the radial direction is based on visual inspection of profiles at four angles. No Fourier analysis, autocorrelation, statistical test, or comparison against a null random field is provided, and the 'three regions' are described qualitatively. In addition, the assumptions stated in §3.4.1 that the medium has planar isotropic distribution with S11=S22 and that the mean Green strain satisfies ε11≈ε22≈0 are introduced without justification; these assumptions are load-bearing for the decomposition and are not verified against the experimental data.","section":"Section 3.4.2, Fig. 17"},{"comment":"The statement that the local rotation angle satisfies the Laplace equation ∇²Θ=0 and gives rise to 'circular fatigue striation' is asserted without derivation or supporting evidence, and the 'characteristic deformation scale L0' is never defined operationally or measured. These speculative claims go beyond the data and should be removed or properly derived in a revised manuscript.","section":"Section 3.4.3"}],"minor_comments":[{"comment":"The abstract contains typos such as 'in suit' for 'in situ' and 'principle strains' for 'principal strains'; these should be corrected throughout.","section":"Abstract"},{"comment":"The phrase 'angel' should be 'angle', and Fig. 7 is referenced twice in the text for different purposes; the figure labels and captions should be clarified.","section":"Section 3.2.2"},{"comment":"The caption lists 'L3/L1, L3/L1, L3/L1' for panels (a)-(c); based on the text, these appear to be L3/L1, L3/L2, and L2/L1 respectively, and the caption should be corrected.","section":"Fig. 6 caption"},{"comment":"The equation for the covariance matrix is garbled in the typesetting and should be rewritten with clear notation.","section":"Section 3.2.1, Eq. (3)"},{"comment":"The statement that εxx and εyy fluctuate 'within a range of 0.2-0.5' appears inconsistent with Fig. 10(a), where the axes and values are unclear; please reconcile the text with the figure.","section":"Section 3.3.2"},{"comment":"Reference formatting is inconsistent, with some entries lacking complete bibliographic details and some Chinese-language sources presented without sufficient information; please unify the reference style.","section":"References"}],"recommendation":"reject","confidential_remarks":"For the editor: the manuscript contains an interesting experimental dataset, but the central analysis is not sound. The shear-strain measure is not actually a shear strain, the stress calculation is invalid, and the key error is admitted to be unquantified. I do not see how these issues can be fixed by revision within the scope of the present manuscript; a fundamental reworking of the method and validation against an independent measure would be required before the central claims could be considered."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take on Feng et al. The genuinely interesting piece is the measurement idea: tracking material lines in 3D XCT images to map local rotation (Stokes local rotation / shear strain) inside sandstone after uniaxial compression, and the claim that shear strain is radially periodic. If that observation held up, it would be new and would give the fracture-mechanics subfield a spatial descriptor. The authors also do some careful-looking grain tracking (689 grains, RDF, bounding-box shapes) that could be useful in its own right.\n\nThe problem is that the central measurement is not validated. Eq. (11) defines shear strain as a length change, (ds0 - ds)/ds0, with no derivation explaining why that equals shear strain; Eq. (12) then switches to an angle-based expression. The text never makes clear what the measured quantity actually is. More seriously, the authors state in §3.4.3 that a certain error arises from image accuracy and material point recognition and that 'the error cannot be estimated.' For a method that depends on matching material lines across segmentation and noise, that is a load-bearing gap. The periodicity itself is asserted from visual inspection of radial profiles, with no Fourier analysis, no statistical test, and no comparison to a null random field. Given that the material-line selection criteria include 'material line is periodic' and 'length less than half the spatial wavelength,' there is a real risk that the selection procedure pre-biases toward oscillating lines.\n\nOther weaknesses are real but secondary. The stress computation in §3.3.1 applies elastic Lame constants to residual strains after unloading, which is not valid for inelastic residual deformation. The Laplace equation for the rotation field and the characteristic deformation scale L0 appear without derivation and are essentially speculative.\n\nThe paper is not a lost cause. The measurement concept is a legitimate extension of DIC-style thinking to 3D CT, and the experimental data collection appears careful. But the headline finding is not established as it stands. Rock mechanics readers interested in grain-scale strain measurement might find the approach worth discussing, but I would not cite it yet.\n\nRecommendation: reject in current form, but send to peer review rather than desk reject. It deserves serious referee time because the method is novel and the observation is checkable. Major revisions should include an error bound, validation on a known displacement field, a proper periodicity test, and dropping or deriving the speculative theoretical claims.","headline":"Novel material-line tracking in 3D CT yields an interesting but unvalidated claim of periodic shear strain in sandstone; the admitted inability to estimate tracking error is the load-bearing weakness.","tokens_in":13149,"tokens_out":3136,"would_cite":false,"duration_ms":32456,"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 shear strain inside fractured sandstone is periodic in the radial direction and locally concentrated in positive and negative bands, so fracture is driven by shear stress concentration near the fracture zone.","keywords":["grain deformation","sandstone","Stokes local rotation","material line","shear strain","X-ray computed tomography","fracture zone","uniaxial compression"],"falsifier":"Recompute the same shear-strain field from the same two CT volumes using an independent full-field technique, for example digital volume correlation of the grayscale images or the tracked grain deformation gradients from Section 3.3, and compare the radial profiles at the same four angles. If an independent map does not show the same near-periodic oscillation with the same wavelength and the same sign-concentrated islands, the periodicity is an artifact of material-line tracking. A more direct variant is to re-pick the random material lines several times and check whether the peak positions in the radial profiles stay fixed.","tokens_in":12093,"feed_emoji":"🪨","tokens_out":10042,"duration_ms":96269,"temperature":0.7,"pith_summary":"Using in-situ X-ray computed tomography at 4.6 μm resolution, the authors tracked mineral grains inside a small sandstone cylinder before and after uniaxial compression. They then drew material lines on CT slices and treated each line's angle change as a Stokes local rotation, a proxy for local shear strain. The central finding is that after unloading the shear strain varies periodically in the radial direction around the main fracture, with positive and negative shear strains concentrated in local bands and in round zones at the fracture ends. The paper interprets this as showing that sandstone fracture is a shear-concentration process: crack initiation and propagation follow irreversible local rotation, and fracture occurs when the local rotation angle reaches a critical value. The authors caution that the measurement error from image accuracy and material-point recognition cannot be estimated.","feed_headline":"Fractured sandstone shows periodic shear strain","feed_subtitle":"A CT material-line method links grain rotation to crack growth and points to shear stress concentration as the driver.","key_machinery":"The central object is the material line: a curve drawn on a CT slice that is tied to the same physical points before and after deformation, so that its change in orientation records how the grain cluster rotated. The load-bearing identity is Stokes local rotation, defined by $\\omega = (\\alpha - \\alpha_0)/ds_0$ in Eq. (11), where $\\alpha_0$ and $\\alpha$ are the angles of the material line before and after loading and $|ds_0|=1$ is the unit scale; this angle change is taken to be the local shear strain. Because Green shear strain cannot be obtained for irregular grains of varied size and distribution, the paper assumes planar isotropic distribution and negligible mean Green strain ($\\varepsilon_{11} \\approx \\varepsilon_{22} \\approx 0$), leaving the local rotation angle as the shear measure. The S+R decomposition theorem in Lagrange coordinates supplies the theoretical framing for separating shear deformation into stretch and rotation parts. The material lines are selected under rules derived from sampling considerations, such as using a line shorter than half the spatial wavelength where the field is periodic, and 352 and 364 lines at two scales are used to build the contour maps.","core_discovery":"The paper's central discovery is a near-periodic distribution of residual shear strain inside fractured sandstone, measured not from continuum strain gauges but from the rotation of material lines observed in CT slices before and after loading. On four radial profiles cut at 0°, 45°, 90° and 135°, the shear strain rises and falls periodically with distance from the center, with approximate symmetry about an axis, while the contour maps show alternating positive and negative shear-strain islands whose sizes match grain dimensions. In the fracture zone the shear strain is larger than the macroscopic compression strain, which the authors take as evidence that the basic mechanism of fracture is shear stress concentration near the fracture zone. They further report that grain strain in the fault zone is about 30 times the macroscopic sample strain, and about 5 times in the non-fault zone, indicating that grain-scale deformation is largely plastic. On this picture, cracks initiate and propagate because irreversible local rotation angles accumulate; when the local rotation angle reaches a critical value, a crack opens in the rotation direction.","pith_inferences":["Editorial inference: the near-periodic radial shear strain could be compared with analytic elastic solutions for a mode-II crack in a circular domain; if the period and decay match such a solution, the observed periodicity would be a geometric consequence of the crack-tip stress field rather than a new material property.","Editorial inference: since only two material-line length scales were tested, a decisive check of the periodicity claim would be to measure radial profiles at several more line lengths and test whether the half-period $L_0$ remains constant; the paper does not provide that check.","Editorial inference: the assumption of negligible mean Green strain could be checked with the same dataset by computing the full deformation gradient from the tracked grain axes; if Green strain is not small, the reported shear strain magnitudes would need adjustment.","Editorial inference: if the periodicity is real, it should reappear in other poorly cemented or granular rocks loaded to similar strains, and its wavelength should scale with grain size; that cross-material prediction is not tested in the paper."],"forward_implications":["If the central claim is correct, the measured shear strain field gives a grain-scale failure criterion: cracks initiate and propagate in the direction of rotation when the local rotation angle reaches a critical value, rather than when macroscopic stress alone reaches a threshold.","The characteristic deformation scale $L_0$, identified with half the spatial period of the shear strain oscillation, becomes a measurable microstructural length that could be used to compare fatigue and fracture resistance of different sandstones.","The two-scale material-line measurements imply that strain maps are scale-dependent: small-scale lines resolve complex large local deformation while large-scale lines show a nearly uniform small deformation, so any grain-scale strain measurement must report its line length.","Grain strain in the fracture zone being roughly 30 times the macroscopic strain means that models linking CT microstructure to sandstone failure must include large local plastic deformation, not just linear elastic strain.","The finding that residual shear strain persists after unloading implies the local rotation responsible for cracking is irreversible, so unloading measurements can reveal the damage field responsible for fracture."],"supporting_citations":[{"why":"Supplies the material-line construction and the Stokes local rotation definition that the shear strain measurement is built on.","marker":"34"},{"why":"Provides the S+R decomposition theorem in Lagrange coordinates used to justify separating shear strain into rotation and stretch parts.","marker":"35"},{"why":"Gives the non-local means image filtering method that prepares the CT slices for material-point and grain tracking.","marker":"29"},{"why":"Supplies the watershed segmentation algorithm used to extract mineral grain pixels from the de-noised CT images.","marker":"30"},{"why":"Provides the radial distribution function used to check that grains do not contact or overtake each other, supporting reliable grain matching.","marker":"31"},{"why":"Supplies the equivalent diameter formula used to characterize grain size and to normalize the radial distribution function.","marker":"32"},{"why":"Provides the normal-vector decomposition/eigenvalue method for extracting grain axes and constructing the deformation gradient and principal strains.","marker":"33"}],"fun_headline_variants":["Sandstone cracks show periodic shear strain","Material-line rotation reveals periodic shear in sandstone","CT scans map periodic shear strain in fractured rock","Shear strain oscillation in fractured sandstone","Grain-scale shear periodic in compressed sandstone"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the angle change of a material line between the pre-loading and post-loading CT images directly equals the local shear strain, and that the same material line can be reliably matched across segmentation and noise; the paper itself states in Section 3.4.3 that the error from image accuracy and material-point recognition cannot be estimated.","fun_headline_variants_meta":{"raw":{"variants":["Sandstone cracks show periodic shear strain","Material-line rotation reveals periodic shear in sandstone","CT scans map periodic shear strain in fractured rock","Shear strain oscillation in fractured sandstone","Grain-scale shear periodic in compressed sandstone"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000241,"raw_usage":{"total_tokens":1495,"prompt_tokens":892,"completion_tokens":603,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":508,"completion_tokens_details":{"reasoning_tokens":538}},"tokens_in":508,"tokens_out":603,"duration_ms":6236,"temperature":1.0,"reasoning_tokens":538,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:12:50.458670+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the same shear-strain field from the same two CT volumes using an independent full-field technique, for example digital volume correlation of the grayscale images or the tracked grain deformation gradients from Section 3.3, and compare the radial profiles at the same four angles. If an independent map does not show the same near-periodic oscillation with the same wavelength and the same sign-concentrated islands, the periodicity is an artifact of material-line tracking. A more direct variant is to re-pick the random material lines several times and check whether the peak positions in the radial profiles stay fixed.","supporting_citations":[{"cited_title":"Strain Geometric Field Theory","cited_arxiv_id":null,"evidence_quote":"Supplies the material-line construction and the Stokes local rotation definition that the shear strain measurement is built on."},{"cited_title":"Study on mechanical properties and finite deformation constitutive model of red sandstone subjected to temperature -water-mechanics coupling","cited_arxiv_id":null,"evidence_quote":"Provides the S+R decomposition theorem in Lagrange coordinates used to justify separating shear strain into rotation and stretch parts."},{"cited_title":"Digital Image Processing","cited_arxiv_id":null,"evidence_quote":"Gives the non-local means image filtering method that prepares the CT slices for material-point and grain tracking."},{"cited_title":"Characterization of the crystal structure, kinematics, stresses and rotations in an gular granular quartz during compaction","cited_arxiv_id":null,"evidence_quote":"Supplies the watershed segmentation algorithm used to extract mineral grain pixels from the de-noised CT images."},{"cited_title":"Molecular dynamics simulation: elementary methods","cited_arxiv_id":null,"evidence_quote":"Provides the radial distribution function used to check that grains do not contact or overtake each other, supporting reliable grain matching."},{"cited_title":"Anal ysis of EBSD G rain Size Measurements Using Microstructure Simulations and a Customizable Pattern Matching Library for Grain Perimeter Estimation","cited_arxiv_id":null,"evidence_quote":"Supplies the equivalent diameter formula used to characterize grain size and to normalize the radial distribution function."},{"cited_title":"Effec tive elastic moduli of a heterogeneous oolitic rock containing 3 -D irregularly shaped pores","cited_arxiv_id":null,"evidence_quote":"Provides the normal-vector decomposition/eigenvalue method for extracting grain axes and constructing the deformation gradient and principal strains."}],"review_version":1}