{"id":"01bf6f71-5987-430f-8e1e-f5e2812b00a8","arxiv_id":"2607.17155","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Closed-form Mode-I stress-intensity factors and crack opening displacement for two moving collinear cracks in a pre-stressed strip with a shear-modulus 'sandiness' rescaling and traveling punch loading, derived via Fourier transforms, asymptotic kernels, and the finite Hilbert transform.","lead":"This mathematical paper derives closed-form formulas for the crack-tip stress intensity factors of two moving collinear cracks in a pre-stressed 'dry sandy' strip under a traveling punch load. The value is as an analytical benchmark for dynamic-fracture models in pavements and geotechnical systems, with the caveat that the sandy material is modeled as isotropic elasticity with a rescaled shear modulus.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The stated upper-surface condition u1=0 is never enforced: §3.3 solves a smooth-punch problem (σ12=0 and u2=0 outside), so the SIF/COD may correspond to different boundary conditions than claimed.","rationale":"The reader's weakest assumption was the large-thickness kernel reduction in §4.1. That is a legitimate verification gap, and an independent check of the coth-kernel asymptotics would be useful. However, the more concrete and decisive issue is the boundary-condition mismatch: the solution does not enforce the stated u1=0 condition at the upper surface, so the central claim 'closed-form SIFs for the stated configuration' is not established for that configuration. This is an internal inconsistency, not merely an unproved algebraic step. It supports the same CONDITIONAL verdict, but for a different reason, so the reader's verdict is unchanged while the stated weakest assumption is only partially aligned.","tokens_in":22080,"tokens_out":41200,"duration_ms":382395,"concrete_test":"Evaluate u1(x1,h) from the derived solution: substitute (41) and (49)-(56) at x2=h, with the asymptotic g1,g2 from §4.2, and compute u1(x1,h)=∫0∞ U(ζ,h)sin(ζx1)dζ for a point |x1|<c or |x1|>e. If the result is not identically zero (or at least O(h^-2)), Eq. (19) is violated. Equivalently, check whether ∂2U(ζ,h)=0 follows from (49)-(56); it does not, which proves the solution cannot satisfy u1=0 on the unloaded intervals while maintaining σ12=0 there.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The manuscript states the upper-surface conditions as u1=0 and u2=0 outside the loaded intervals plus σ12=0 everywhere (Eqs. 19-21). But the transform solution in §3.3 enforces only u2=0 there (through the g2 representation in Eq. 51) and σ12=0 everywhere (Eq. 21); it never constrains U(ζ,h), the transform of u1 at the upper surface. Consequently, u1(x1,h), obtained from Eq. (41) with the amplitudes (49)-(56), is generally nonzero on |x1|<c and |x1|>e. Indeed, on those intervals u2≡0, so σ12=S∂2u1; imposing both u1=0 and σ12=0 would require ∂2U(ζ,h)=0 as an extra condition, which is absent from the derivation. The boundary-value problem actually solved is the standard smooth-punch one (u2 prescribed outside, σ12=0 everywhere), not the fully adhesive one described by Eq. (19). This is load-bearing because the integral equations (58)-(59) and all subsequent SIF/COD formulas inherit this boundary-condition choice; if the intended punch is adhesive, the results are for a different configuration.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper treats two steadily moving collinear Mode-I Griffith cracks in an initially stressed dry sandy strip of finite thickness, with concentrated crack-face loads and a moving punch pressure on the outer surfaces. The authors use a moving coordinate frame, Fourier transforms, and a large-thickness asymptotic reduction to convert the mixed boundary-value problem into coupled Cauchy singular integral equations (58)-(59), which they solve analytically with the finite Hilbert transform. The output consists of the crack-density functions (81)-(82), closed-form Mode-I stress intensity factors at the inner and outer crack tips (87)-(88), and a crack-opening-displacement expression (89), all to O(h^{-2}), together with parametric studies of the effects of crack speed, sandiness, initial stress, strip thickness, geometry, and loading position. The derivation is self-contained and the algebra connecting the integral equations to the asymptotic solution is internally consistent; the main weaknesses are in the statement of the boundary conditions and in the unproved large-thickness kernel reduction.","tokens_in":22327,"tokens_out":11464,"duration_ms":139469,"significance":"If the results are correct, this is a useful analytical benchmark for a fairly complicated configuration: two interacting moving cracks in a pre-stressed, sand-modified elastic strip under combined concentrated and distributed moving loads. The paper provides explicit, closed-form expressions rather than purely numerical results, and it contains several limiting-case checks, including a claimed reduction to a previously published monoclinic-strip problem. These are genuine strengths. However, the significance is conditional on resolving two load-bearing issues: the boundary-value problem actually solved differs from the one stated, and the asymptotic kernel formulas that generate the finite-thickness/punch corrections are asserted without derivation or numerical verification.","major_comments":[{"comment":"The stated boundary-value problem and the problem actually solved are not the same. Eq. (19) requires u1(x1,h)=0 for |x1|<c and |x1|>e, and Eq. (23) similarly requires u1(x1,0)=0 outside the cracks. The transform solution never imposes these conditions: the amplitude relations (49)-(54) enforce σ12=0 at x2=0 and x2=h and use the representations (50)-(51) to enforce u2=0 on the relevant intervals, but U(ζ,h) and U(ζ,0) remain unconstrained. Consequently the SIE system (58)-(59) and all subsequent SIF/COD formulas correspond to the smooth-punch/Mode-I problem with u2=0 and σ12=0 prescribed, not to the clamped conditions in (19) and (23). In fact, for a Mode-I crack the correct mid-plane condition is u2=0 outside the cracks, not u1=0; so Eq. (23) is not a Mode-I condition either. The manuscript must either delete the u1=0 conditions from the problem statement or extend the transform solutio","section":"§2.4 and §3.3, Eqs. (19), (23), (49)-(54)"},{"comment":"The large-thickness asymptotic reduction is the only source of the finite-thickness and punch-coupling corrections in the final SIFs, yet it is asserted without derivation. The exact kernel multipliers in (63)-(67) contain F1, F2, F3, F4 built from 1/sinh(r_j ζh), which near ζ=0 behave as 1/(r_j ζh), giving 1/ζ poles. For the leading terms to be O(h^{-2}) and linear in s as claimed in (70)-(72), these poles must cancel after combining Π_j, Λ_j, r_j, and the finite parts must collapse to the single constant P in (73). None of this cancellation is exhibited, and no numerical check of (70)-(72) is reported. Because H1 and H2 in (83)-(84) and the h^{-2} contributions to K_I^(c) and K_I^(e) in (87)-(88) depend directly on these forms, this is a load-bearing gap. Please provide the derivation or an appendix, or verify the asymptotic kernels numerically against the exact integral representation","section":"§4.1, Eqs. (70)-(72)"},{"comment":"The 'sandiness-free' limiting case is mis-specified. The constitutive relation (4) contains the factor 2µ/χ with χ>1, and §2.2 explicitly states that χ=1 recovers the classical isotropic response. Setting χ=0 in Eq. (106), and subsequently in (107)-(111), makes the effective shear modulus singular and is not a well-defined limit. If the intended consistency check is a sandiness-free medium, the parameter should be χ=1 throughout this subsection; otherwise the claimed reduction to the corresponding case in [9] is not established. This also propagates into the formulas for H_1^(b) and H_2^(b) in (114)-(115).","section":"§4.5.3, Eq. (106)"}],"minor_comments":[{"comment":"The second objective states that the solution uses the 'Schmidt method', but the paper actually uses the finite Hilbert transform technique. Please correct the wording.","section":"§1.6"},{"comment":"The sentence defining the coefficients H1 and H2 ('The coefficients H1 and H2 ... are given by') appears twice in succession. Please remove the duplication.","section":"§4.2, after Eq. (82)"},{"comment":"The symbol D_j is used for two different sets of quantities: the effective coefficients in Eq. (31) and the integration constants in Eqs. (81)-(82). This is confusing and should be changed, e.g., to A_j for the integration constants.","section":"§2.3 and §4.2"},{"comment":"The quantity Q_s/Q_c is repeatedly called the 'tangential loading ratio', but Q_s is the normal punch pressure in Eq. (80) and Q_c is the concentrated normal crack-face load. A neutral term such as 'punch-to-concentrated load ratio' would be more accurate.","section":"§5, Table 1"},{"comment":"The constant P is defined in (73) as a combination of the Π_j, Λ_j, r_j, and C, but the text does not explain why the same P appears in all three kernel asymptotics (70)-(72). A sentence indicating the algebraic origin of this common factor would help.","section":"§4.1, Eq. (73)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's main idea is sound, the algebra appears internally consistent, and the closed-form character of the results is valuable. However, the boundary-condition mismatch and the unproved kernel asymptotics are both load-bearing and must be addressed before the results can be trusted. The χ=0 issue in the sandiness-free limit is a likely typo but should also be fixed. I therefore recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know before you read it. First, the problem actually solved does not match the stated boundary conditions. The paper claims u1=0 on the upper surface outside the loaded intervals, but the transform solution never enforces that condition; it imposes only u2=0 there and σ12=0 everywhere. That is the standard smooth-punch condition, not the adhesive punch described in Eq. (19). This is load-bearing because the integral equations and all subsequent SIF/COD formulas inherit that choice. Second, the headline closed forms are not fully closed: the stress intensity factor formulas (87)-(88) contain constants D1..D3 that are never evaluated, so the results are a template rather than explicit numbers.\n\nWhat is genuinely new is the configuration: two moving collinear cracks, initial stress, sandiness, punch loading, and finite strip thickness combined in one analytical treatment. The method is the standard Fourier-transform à Cauchy-singular-integral à finite-Hilbert-transform program, and the broad algebraic structure is internally consistent. I checked the O(h^-2) bookkeeping between Eq. (58) and Eq. (77), and the πB factor in the COD formula comes out correctly. There is real effort here and the derivation mostly hangs together at the level of formal manipulation.\n\nThe soft spots are substantial. The large-thickness kernel reduction in §4.1 is asserted rather than derived; the ζ→0 pole cancellations are not exhibited and the asymptotic forms are never checked numerically. The limiting case in §4.5.3 sets χ=0 even though the constitutive law requires χ>1; the sandiness-free limit should be χ=1. The unevaluated constants are a genuine gap—they make the \"closed-form\" results incomplete. And there is no independent benchmark, either a classical limit, a numerical check, or an experiment. The boundary-condition mismatch is the most serious issue: if the intended punch is adhesive, the results solve a different mechanical problem.\n\nWho gets value from this: specialists in analytical fracture mechanics who want to see how far the standard integral-equation machinery can be pushed in a niche geomechanics setting, not design engineers looking for usable formulas.\n\nRecommendation: send it to peer review rather than desk-rejecting. The flaws are nameable and addressable—derive the kernel asymptotics, evaluate the constants, fix the χ=0 slip, and either correct the boundary conditions or clearly state that a smooth punch is being solved. A competent referee can give the authors a specific revision path.","headline":"The paper offers a new analytical configuration but solves a boundary-value problem different from the one it states, and its 'closed-form' SIFs contain unevaluated constants.","tokens_in":22917,"tokens_out":3678,"would_cite":false,"duration_ms":34190,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["74R10","45E05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper derives closed-form formulas for the Mode-I stress intensity factors and crack opening displacement of two collinear cracks moving in a pre-stressed dry sandy strip under a moving punch, with explicit finite-thickness corrections.","keywords":["moving collinear cracks","dry sandy medium","initial stress","moving punch load","stress intensity factor","crack opening displacement","singular integral equations","finite Hilbert transform"],"falsifier":"Solve the coupled integral equations with the exact kernels numerically for the same parameters as Section 5 and compare the resulting stress intensity factors against the closed-form formulas; if the difference is not O(h⁻²) or does not vanish as h/e increases, the asymptotic reduction is refuted. A cheaper check is to expand the kernel multipliers around ζ = 0 and confirm that the 1/ζ poles cancel exactly, leaving the claimed coefficient P.","tokens_in":21825,"feed_emoji":"🏗️","tokens_out":10805,"duration_ms":91287,"temperature":0.7,"pith_summary":"This paper sets out to prove that the fracture response of two collinear Griffith cracks (idealized line cracks) moving steadily in a pre-stressed dry sandy strip under a moving punch can be written in closed form. The target results are explicit formulas for the crack density functions, the Mode-I stress intensity factors at the inner and outer crack tips, and the crack opening displacement, each carried to the leading correction in the inverse square of the strip thickness. The model simultaneously includes crack speed, initial stress, sandiness, strip thickness, crack geometry, and punch position, so the formulas expose which combinations of parameters drive crack-tip intensification. A careful reader would care because this is the geometry of pavement, railway foundation, and underground excavation failures, and because an analytical result for this parameter combination has not previously been available. The route is steady-state transformation, Fourier transforms, coupled Cauchy-type singular integral equations, a thick-strip kernel reduction, and inversion by the finite Hilbert transform.","feed_headline":"Moving cracks in sandy strips get closed-form stress-intensity factors","feed_subtitle":"Closed-form formulas include crack speed, prestress, sandiness, strip thickness, and punch position in one model.","key_machinery":"The mechanism carrying the argument is the asymptotic kernel reduction of Section 4.1. The exact transform-domain kernel multipliers N₁₁, N₁₂, N₂₁ contain 1/sinh(r_j ζh) factors; for a thick strip the paper reduces them, using the series identities for 1/k² and 1/(2k+1)², to simple linear forms proportional to P s/h². That reduction changes two coupled Cauchy-type singular integral equations into four algebraic conditions, which are inverted in closed form by the finite Hilbert transform, an integral inversion technique for Cauchy principal-value equations. All physical parameters enter through the effective elastic coefficients D₁–D₅, the sandiness-dependent shear coefficient S = µ/χ, the i","core_discovery":"The central claim is that the coupled problem can be solved, not just reduced, after a thick-strip simplification: for large h the regular kernels collapse to linear forms proportional to s/h² with one common material-loading constant P, and the density functions then split into an unbounded-strip part plus an h⁻² correction. Substitution into the near-tip limits gives the Mode-I stress intensity factor at the inner tip, K_I^(c) = (π/(2Q₁)) (1/(√c√(e²−c²))) (D₁ + D₃/h²), and an analogous closed-form expression at the outer tip, with punch loading and crack interaction entering through constants H₁ and H₂. The crack opening displacement follows as an elliptic-integral expression with the same","pith_inferences":["Beyond the paper: the thick-strip kernel reduction could be tested numerically by discretizing the exact integral equations; a mismatch that is not O(h⁻²) would map out where the asymptotic window ends.","Beyond the paper: because the sandiness and initial-stress effects enter only through effective coefficients, the closed-form structure should survive for any isotropic medium with a modified shear modulus; the geometrical form of the formulas is set by the crack configuration, not by the material model.","Beyond the paper: the same reduction-and-Hilbert-transform route could extend to periodic collinear crack arrays, where the two-crack ligament is replaced by a lattice spacing and the Hilbert-transform weights would need only modest changes.","Beyond the paper: the predicted non-monotonic inner-tip response as a function of crack length, with a minimum near c/e ≈ 0.10–0.15, is a specific signature that targeted simulations or experiments on crack interaction could confirm or refute."],"forward_implications":["If the paper is correct, the Mode-I stress intensity factors at both crack tips of a thick sandy strip can be computed from closed-form expressions instead of solving singular integral equations.","The formulas quantify a specific asymmetry: the outer crack tip carries a larger stress intensity factor than the inner tip because the moving concentrated load sits closer to it.","The results show a common dynamic pattern: crack-tip intensity remains nearly flat until the crack speed approaches the effective shear-wave speed, then rises sharply across all material and geometric parameters.","The limiting cases V = 0, zero initial stress, and the zero-sandiness/punch-free configuration reproduce earlier simpler models, providing internal consistency checks.","The finite-thickness correction couples punch loading and inner-tip intensity through the constants H₁ and H₂, so the model captures the interaction between boundary loading and crack interaction at the first nontrivial order."],"fun_headline_variants":["Closed-form SIFs for moving cracks in prestressed dry sand","Collinear crack motion in sand: exact intensification formulas","Thick-strip trick yields analytic SIFs for moving collinear cracks","Moving punch loads on sandy cracks: closed-form stress fields","Exact SIFs for dynamic cracks in sandy strips under moving loads"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that the exact thick-strip kernels reduce to the simple O(h⁻²) linear forms with the single constant P using only two series identities; the paper does not exhibit the cancellation of the kernel poles at ζ = 0 or verify the reduction numerically, so if that reduction is incomplete the finite-thickness corrections in the stress-intensity-factor formulas are wrong.","fun_headline_variants_meta":{"raw":{"variants":["Closed-form SIFs for moving cracks in prestressed dry sand","Collinear crack motion in sand: exact intensification formulas","Thick-strip trick yields analytic SIFs for moving collinear cracks","Moving punch loads on sandy cracks: closed-form stress fields","Exact SIFs for dynamic cracks in sandy strips under moving loads"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001718,"raw_usage":{"total_tokens":6646,"prompt_tokens":769,"completion_tokens":5877,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":513,"completion_tokens_details":{"reasoning_tokens":5787}},"tokens_in":513,"tokens_out":5877,"duration_ms":37379,"temperature":1.0,"reasoning_tokens":5787,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T18:53:20.342130+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Solve the coupled integral equations with the exact kernels numerically for the same parameters as Section 5 and compare the resulting stress intensity factors against the closed-form formulas; if the difference is not O(h⁻²) or does not vanish as h/e increases, the asymptotic reduction is refuted. A cheaper check is to expand the kernel multipliers around ζ = 0 and confirm that the 1/ζ poles cancel exactly, leaving the claimed coefficient P.","supporting_citations":[],"review_version":1}