REVIEW 4 major objections 5 minor 45 references
Why planar cracks fragment into echelon cracks
T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash
Pith's one-line read A strength-constrained energy minimization—not energy competition alone—predicts when and where a planar crack fragments into echelon daughter cracks.
desk verdict Interesting and likely important claim—strength matters for crack path—but the phase-field regularization that produces the echelon pattern is imported from prior work without a proof of fidelity in this geometry; the broad claims exceed the current evidence. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The strength-constrained phase-field formulation: the classical variational fracture functional is minimized subject to the constraint that the crack set lies inside the region where the stress satisfies the Drucker–Prager strength surface F(S)=0. In practice, this is implemented by adding an explicit stress-based driving force ce to the phase-field evolution equation, with coefficients calibrated on uniform stress states. Crack growth then requires both that the strength surface be exceeded (nucleation) and that energy be minimized (propagation).
What would settle it
A tearing experiment on a brittle material with known strength ratio, varying plate thickness H while holding the material length scale fixed: if echelon cracks appear for H/lss_ch below the critical threshold (or fail to appear above it), the predicted scaling is wrong. Alternatively, varying the shear-to-tensile strength ratio by changing the material should move the crack orientation angle between about 0° and 45° as the paper's Fig. 4(b) predicts.
Extended reading notes
Core claim
Echelon crack formation is a nucleation phenomenon, not an instability of a smooth crack front: the parent crack spawns disconnected daughter cracks in regions where the local stress state first violates the material's strength surface. A phase-field model that minimizes elastic plus surface energy only among crack sets confined to these strength-exceeded regions reproduces the observed angled, stepped crack patterns; the purely energetic model instead predicts continued planar growth. The orientation angle of the daughter cracks increases with the shear-to-tensile strength ratio, and fragmentation occurs only when the plate thickness exceeds a characteristic strength-based length scale. Thu
Load-bearing premise
The specific analytical driving force that encodes the strength constraint, calibrated on simple axisymmetric test geometries, is assumed to remain a faithful regularization of the constrained energy minimization in the mixed-mode tearing geometry; if it does not, the predicted echelon fragmentation could be a numerical artifact.
Editorial extensions
If this is right
- Predicts echelon crack formation in both soft and hard brittle materials without invoking disorder, defects, or front perturbations.
- Reconciles energy-based (maximum energy release rate) and stress-based (local mode I) path criteria: which one dominates is controlled by the shear-to-tensile strength ratio.
- Identifies two non-dimensional parameters, σss/σts and H/lss_ch, that govern crack orientation and fragmentation, giving experimentally testable scaling laws.
- Shows the purely variational Griffith model to be fundamentally incomplete for large-crack growth under mixed tension/shear.
- Because the framework uses only measurable elastic, toughness, and strength properties, it offers a direct route to prediction in new materials.
Reading between the lines
- The same strength-constraint mechanism may explain other 'spontaneous fragmentation' fracture patterns, such as crack front segmentation under mixed-mode loading in geological or composite materials.
- If the two-parameter scaling is robust, experiments that vary only the sample thickness should show a sharp transition from planar to echelon cracking at a critical H/lss_ch, which could be tested with a series of hydrogel or polymer plates.
- The authors' claim that the phase-field regularization length ε is freely tunable suggests the model may be used to resolve process-zone effects directly, perhaps linking echelon spacing to microstructure.
- The framework's extension to anisotropic strength surfaces (e.g., Mohr-Coulomb or Hoek-Brown) is straightforward, which may yield orientation preferences in rocks with anisotropic cleavage.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. Using two phase-field formulations, the paper revisits Knauss's mode-III tearing experiment on a notched plate. The classical variational Griffith model (3), minimizing elastic plus fracture energy, predicts continued planar crack growth for both graphite and PDMS. The authors then introduce a strength-constrained model (10)-(13) in which a Drucker-Prager strength surface restricts where crack growth can occur; this model spontaneously produces echelon daughter cracks without stochastic disorder or geometric imperfections. From the simulations, the paper identifies the shear-to-tensile strength ratio sigma_ss/sigma_ts and the thickness-to-material-length ratio H/l_ss^ch as the controlling parameters, and argues that energy-based and stress-based crack-path criteria are reconciled within one framework. The central conclusion is that Griffith energy minimization alone is fundamentally incomplete for predicting large-crack growth and that a strength constraint is necessary.
Significance. If correct, these results would be a substantive advance: they offer a deterministic explanation of a classical crack morphogenesis problem and tie together competing empirical criteria. The paper's strengths include a clean binary comparison between the classical and strength-constrained models on the same geometry, use of independently measurable material parameters (with the PDMS caveat below), and distribution of an open-source FEniCS implementation. However, the main prediction depends on an imported phase-field regularization of the strength constraint whose validity in the mixed-mode tearing state is not established; and the quantitative experimental support is a single historical data point. Thus the work is promising, but at present the force of the central claim exceeds its verification.
major comments (4)
- [§3, Eqs. (10)-(13)] The echelon patterns in Figs. 1(d), 3, and 5 are generated entirely by the added driving force c_e in Eq. (10). Equations (11)-(13) are taken from Refs. [30,40], where the coefficients beta1^eps, beta2^eps, delta_eps were calibrated against homogeneous uniaxial and hydrostatic stress states (poker-chip, indentation). It is not demonstrated that this analytical form enforces the sharp constrained minimization (9) in the strongly inhomogeneous, mixed-mode stress field of the tearing problem. Since the classical model (3) gives planar growth in the same geometry, the strength-surface explanation is currently as sensitive to the regularization as to the physics. The paper's own caveat in the §3 summary ('the phase-field model is not completely connected') underscores the gap. I request an eps-convergence study at fixed h/eps, an independent implementation check, or a direct comparison with t
- [Appendix A, Table 2] For PDMS, the paper states 'We have considered lower strength values than the actual values to make the simulations computationally less expensive.' This is a significant departure from the stated goal of using standard, experimentally measured material parameters. The PDMS results in Figs. 2(d) and 3(b), and the abstract claim of 'general applicability to any brittle material', are therefore not a test of the model for actual PDMS. In addition, the full-geometry comparison with Knauss's experiment (§3, Fig. 1(d)) does not state which material parameters were used; if those are the reduced PDMS values, the approximate 40° result cannot be viewed as a quantitative validation. Please rerun at least one PDMS case at actual strength or reframe the claim as qualitative.
- [§3 and Fig. 5] The text states that 'both the number and orientation of daughter cracks were found to depend primarily on the value of l_ss^ch,' but Fig. 5 plots only the number of daughter cracks. No orientation data are shown for the H/l_ss^ch study, so the second key parameter is not supported by the presented data. Also, the error bars are not defined and no statistical procedure is described. Please either add orientation results or temper the claim.
- [§3, experimental comparison] The sole quantitative experimental benchmark is Knauss's reported inclination of about 45° compared with the predicted approximately 40°. There is no uncertainty on either value, no discussion of how the experimental angle was measured, and no account of the differences between Solithane (experiment) and the constitutive models used here. Given that the Abstract states the model is shown 'through comparison with classical experiments' to explain echelon formation, this comparison is too thin to carry the load. Additional validation against existing hydrogel/glass data, or at least a more careful treatment of this single data point, is needed.
minor comments (5)
- [Eq. (9)] The sharp constrained minimization statement is informal: 'Gamma subset V_F(t)' conflates a 2D crack set with a 3D point set, and no topology/admissibility is specified. A precise formulation would help, especially because the paper argues that (10) is its regularization.
- [§3, angle extraction] Please define how theta is measured from the phase-field contours. Only two crack extensions are used to support the orientation claim, so the measurement procedure matters.
- [§3 and Fig. 5] There is an inconsistency in whether the sweep varied sigma_ss or sigma_ss^2 by factors of 1-5. The text near Eq. (14) says 'sigma_ss^2' and a few lines later says 'sigma_ss was varied.' Please check consistency.
- [Appendix B] The regularization length eps is described as a free parameter, but no convergence study in eps is reported. A sentence or figure showing that the reported morphologies are stable as eps decreases, with h=eps/4, would strengthen the paper.
- [Throughout] Some figure panels are difficult to read in the current typesetting (e.g., Fig. 2(b) axes, Fig. 4(d) labels). A cleaner presentation would improve reproducibility and readability.
Circularity Check
No significant circularity: echelon prediction is not built from fitted inputs, but the central phase-field regularization is inherited from the authors' prior work without proof of fidelity in the tearing geometry.
full rationale
The paper's derivation chain is: (i) show that the classical variational model (3) predicts planar growth; (ii) introduce the strength-constrained minimization (9); (iii) implement it through the phase-field driving force ce in Eqs. (10)-(13); (iv) simulate echelon fragmentation; and (v) map orientation and crack count versus σss/σts and H/lch. The claimed outputs are not equivalent to the inputs by construction: the two 'governing parameters' are independent material/geometric quantities (graphite strengths come from Sato's tube tests; PDMS properties from prior characterization, with lower strengths chosen only to reduce computational cost), and the echelon angle and fragmentation count emerge from the boundary-value problem rather than being fit to Knauss's observations. The main weakness is that the strength-constrained phase-field model itself is taken from the authors' own prior work [30,40], and the paper admits 'the phase-field model is not completely connected' to the sharp constrained minimization (9). No Γ-convergence or numerical equivalence proof is given for the mixed-mode tearing geometry; if Eqs. (11)-(13) do not faithfully regularize (9), the echelon morphology could be an artifact of the imported ansatz. That is a genuine verification gap and a correctness risk, but it is not circularity: the paper does not fit the model to the echelon data it claims to predict, and the cited prior work was validated on independent problems. Under the stated rules, this warrants a low circularity score rather than a charge of circular derivation.
Assumptions & free parameters
free parameters (2)
- PDMS tensile and hydrostatic strengths (reduced) =
σts = 0.1 MPa, σhs = 0.125 MPa
- Phase-field regularization length ε =
ε = l_ss^ch / 3 in final runs (mesh h = ε/4)
assumptions (4)
- domain assumption The Drucker–Prager surface with parameters σts and σss adequately describes the fracture strength of graphite and PDMS under the triaxial stress states near the crack front.
- ad hoc to paper The regularized strength-constrained model (10)–(13), including the analytical driving force ce and coefficients β1ε, β2ε, δε, faithfully represents the sharp constrained minimization (9).
- domain assumption The reduced geometry in Fig 2(a) (and the non-strict mode III loading) yields qualitatively equivalent results to the full Knauss tearing experiment.
- domain assumption The classical variational phase-field model without energy split is the correct baseline for the claim that 'purely energetic minimization is insufficient.'
Cite this review
Pith. "Pith review of Why planar cracks fragment into echelon cracks." pith.science (2026). https://pith.science/paper/TPELMMZW
@misc{pith2026251216053,
author = {Pith},
title = {Pith review of: Why planar cracks fragment into echelon cracks},
year = {2026},
howpublished = {\url{https://pith.science/paper/TPELMMZW}},
note = {Machine review of arXiv:2512.16053}
}
read the original abstract
Predicting the path and shape of growing cracks is fundamental to understanding of fracture. Under out-of-plane shear loading, an initially planar crack may spontaneously fragment into multiple cracks, forming a striking echelon crack pattern. Explaining this crack morphogenesis in brittle materials has been a long-standing open problem essential to developing a complete theory of crack growth. Here, through comparison with classical experiments, we show that a strength-constrained minimization of the sum of elastic and surface energies explains echelon crack formation. Results are presented for both soft and hard materials, confirming the model's general applicability to any brittle material. As a corollary, we show that, contrary to prevailing views, a purely energetic minimization model is insufficient to predict the growth of large cracks. We identify two key non-dimensional parameters governing crack fragmentation and orientation, and demonstrate that these reconcile the various energy-based and stress-based empirical criteria proposed in the literature for crack path.
Figures
Reference graph
Works this paper leans on
-
[1]
A. A. Griffith, VI. the phenomena of rupture and flow in solid s, Philosophical transactions of the royal society of londo n. Series A, containing papers of a mathematical or physical ch aracter 221 (1921) 163–198
1921
-
[2]
Sommer, Formation of fracture ‘lances’ in glass, Engi neering Fracture Mechanics 1 (1969) 539–546
E. Sommer, Formation of fracture ‘lances’ in glass, Engi neering Fracture Mechanics 1 (1969) 539–546
1969
-
[3]
Knauss, An observation of crack propagation in anti-p lane shear, International Journal of Fracture Mechanics 6 ( 1970) 183–187
W. Knauss, An observation of crack propagation in anti-p lane shear, International Journal of Fracture Mechanics 6 ( 1970) 183–187
1970
-
[4]
Ronsin, C
O. Ronsin, C. Caroli, T. Baumberger, Crack front echelon instability in mixed mode fracture of a strongly nonlinear elastic solid, Europhysics Letters 105 (2014) 34001
2014
-
[5]
K. Pham, K. Ravi-Chandar, On the growth of cracks under mi xed-mode I+ III loading, International Journal of Fracture 199 (2016) 105–134
2016
-
[6]
W ang, M
M. W ang, M. Adda-Bedia, J. M. Kolinski, J. Fineberg, How h idden 3D structure within crack fronts reveals energy balance, Journal of the Mechanics and Physics of Solids 161 ( 2022) 104795
2022
-
[7]
M. L. Cooke, D. D. Pollard, Fracture propagation paths un der mixed mode loading within rectangular blocks of polymet hyl methacrylate, Journal of Geophysical Research: Solid Eart h 101 (1996) 3387–3400
1996
-
[8]
E. A. Zimmermann, M. E. Launey, H. D. Barth, R. O. Ritchie, Mixed-mode fracture of human cortical bone, Biomaterials 30 (2009) 5877–5884
2009
Show all 45 references
-
[9]
A. T. Zehnder, N. K. Zella, Spiral to flat fracture transit ion for notched rods under torsional loading, Internationa l Journal of Fracture 195 (2015) 87–92
2015
-
[10]
Barenblatt, G
G. Barenblatt, G. Cherepanov, On brittle cracks under l ongitudinal shear, Journal of Applied Mathematics and Mech anics 25 (1961) 1654–1666
1961
-
[11]
Erdogan, G
F. Erdogan, G. Sih, On the crack extension in plates unde r plane loading and transverse shear, Journal of basic engin eering 85 (1963) 519–525
1963
-
[12]
G. C. Sih, Strain-energy-density factor applied to mix ed mode crack problems, International Journal of fracture 1 0 (1974) 305–321
1974
-
[13]
R. V. Gol’dstein, R. L. Salganik, Brittle fracture of so lids with arbitrary cracks, International Journal of Fract ure 10 (1974) 507–523
1974
-
[14]
Nuismer, An energy release rate criterion for mixed m ode fracture, International Journal of Fracture 11 (1975) 2 45–250
R. Nuismer, An energy release rate criterion for mixed m ode fracture, International Journal of Fracture 11 (1975) 2 45–250
1975
-
[15]
W u, Fracture under combined loads by maximum-ene rgy-release-rate criterion, Journal of Applied Mechanics 45 (1978) 553
C.-H. W u, Fracture under combined loads by maximum-ene rgy-release-rate criterion, Journal of Applied Mechanics 45 (1978) 553
1978
-
[16]
Lazarus, F.-G
V. Lazarus, F.-G. Buchholz, M. Fulland, J. Wiebesiek, C omparison of predictions by mode II or mode III criteria on cr ack front twisting in three or four point bending experiments, I nternational Journal of Fracture 153 (2008) 141–151
2008
-
[17]
Chambolle, G
A. Chambolle, G. A. Francfort, J.-J. Marigo, When and ho w do cracks propagate?, Journal of the Mechanics and Physics of Solids 57 (2009) 1614–1622
2009
-
[18]
K. Pham, K. Ravi-Chandar, The formation and growth of ec helon cracks in brittle materials, International Journal o f Fracture 206 (2017) 229–244
2017
-
[19]
Mittelman, Z
B. Mittelman, Z. Yosibash, Energy release rate cannot p redict crack initiation orientation in domains with a sharp V-notch under mode III loading, Engineering Fracture Mechanics 141 (2015) 230–241
2015
-
[20]
A. J. Pons, A. Karma, Helical crack-front instability i n mixed-mode fracture, Nature 464 (2010) 85–89
2010
-
[21]
Leblond, A
J.-B. Leblond, A. Karma, V. Lazarus, Theoretical analy sis of crack front instability in mode I+ III, Journal of the Mechanics and Physics of Solids 59 (2011) 1872–1887
2011
-
[22]
Kolvin, G
I. Kolvin, G. Cohen, J. Fineberg, Topological defects g overn crack front motion and facet formation on broken surfa ces, Nature materials 17 (2018) 140–144
2018
-
[23]
G. A. Francfort, J.-J. Marigo, Revisiting brittle frac ture as an energy minimization problem, Journal of the Mecha nics and Physics of Solids 46 (1998) 1319–1342
1998
-
[24]
Chambolle, G
A. Chambolle, G. A. Francfort, J.-J. Marigo, Revisitin g energy release rates in brittle fracture, Journal of Nonli near Science 20 (2010) 395–424
2010
-
[25]
Francfort, Variational fracture: twenty years afte r, International Journal of Fracture 237 (2022) 3–13
G. Francfort, Variational fracture: twenty years afte r, International Journal of Fracture 237 (2022) 3–13
2022
-
[26]
Bourdin, G
B. Bourdin, G. A. Francfort, J.-J. Marigo, Numerical ex periments in revisited brittle fracture, Journal of the Mec hanics and Physics of Solids 48 (2000) 797–826
2000
-
[27]
S. Sato, H. Awaji, K. Kawamata, A. Kurumada, T. Oku, Frac ture criteria of reactor graphite under multiaxial stesses , Nuclear Engineering and Design 103 (1987) 291–300
1987
-
[28]
Kumar, B
A. Kumar, B. Bourdin, G. A. Francfort, O. Lopez-Pamies, Revisiting nucleation in the phase-field approach to brittl e fracture, Journal of the Mechanics and Physics of Solids 142 (2020) 104027
2020
-
[29]
Poulain, V
X. Poulain, V. Lefevre, O. Lopez-Pamies, K. Ravi-Chand ar, Damage in elastomers: nucleation and growth of cavities , micro-cracks, and macro-cracks, International Journal of Fracture 205 (2017) 1–21. 12
2017
-
[30]
Kamarei, A
F. Kamarei, A. Kumar, O. Lopez-Pamies, The poker-chip e xperiments of synthetic elastomers explained, Journal of t he Mechanics and Physics of Solids 188 (2024) 105683
2024
-
[31]
Moln´ ar, A
G. Moln´ ar, A. Doitrand, V. Lazarus, Phase-field simula tion and coupled criterion link echelon cracks to internal l ength in antiplane shear, Journal of the Mechanics and Physics of S olids 188 (2024) 105675
2024
-
[32]
Khayaz, A
U. Khayaz, A. Dahal, A. Kumar, A comparison of phase field models of brittle fracture incorporating strength, I: Mixe d- mode loading, Engineering Fracture Mechanics 330 (2025) 11 1679
2025
-
[33]
C. Liu, A. Kumar, Emergence of tension–compression asy mmetry from a complete phase-field approach to brittle fract ure, International Journal of Solids and Structures 309 (2025) 1 13170
2025
-
[34]
Vicentini, C
F. Vicentini, C. Zolesi, P. Carrara, C. Maurini, L. De Lo renzis, On the energy decomposition in variational phase-fi eld models for brittle fracture under multi-axial stress state s, International Journal of Fracture 247 (2024) 291–317
2024
-
[35]
Kumar, G
A. Kumar, G. A. Francfort, O. Lopez-Pamies, Fracture an d healing of elastomers: A phase-transition theory and nume rical implementation, Journal of the Mechanics and Physics of Sol ids 112 (2018) 523–551
2018
-
[36]
Kumar, O
A. Kumar, O. Lopez-Pamies, The poker-chip experiments of Gent and Lindley (1959) explained, Journal of the Mechani cs and Physics of Solids 150 (2021) 104359
1959
-
[37]
Kumar, K
A. Kumar, K. Ravi-Chandar, O. Lopez-Pamies, The revisi ted phase-field approach to brittle fracture: application t o indentation and notch problems, International Journal of F racture 237 (2022) 83–100
2022
-
[38]
Kumar, Y
A. Kumar, Y. Liu, J. E. Dolbow, O. Lopez-Pamies, The stre ngth of the brazilian fracture test, Journal of the Mechanic s and Physics of Solids 182 (2024) 105473
2024
-
[39]
Kamarei, B
F. Kamarei, B. Zeng, J. E. Dolbow, O. Lopez-Pamies, Nine circles of elastic brittle fracture: A series of challenge p roblems to assess fracture models, Computer Methods in Applied Mech anics and Engineering 448 (2026) 118449
2026
-
[40]
Lopez-Pamies, F
O. Lopez-Pamies, F. Kamarei, When and where do large cra cks grow? Griffith energy competition constrained by materia l strength, Extreme Mechanics Letters 81 (2025) 102417
2025
-
[41]
S. Chockalingam, On the construction of explicit analy tical driving forces for crack nucleation in the phase field a pproach to brittle fracture with application to mohr–coulomb and dr ucker–prager strength surfaces, Journal of Applied Mechan ics 92 (2025) 041002
2025
-
[42]
Chockalingam, A
S. Chockalingam, A. B. Tepole, A. Kumar, The phase-field model of fracture incorporating mohr-coulomb, mogi-coulo mb, and hoek-brown strength surfaces, arXiv preprint arXiv:25 11.04627 (2025)
2025
-
[43]
W ard, A
O. W ard, A. Kumar, Data for echelon crack formation, https://github.com/Aditya-Kumar-Lab-GT/Echelon-Crack-Formation,
-
[44]
Santarossa, N
A. Santarossa, N. R. Varela-Rosales, P. Steinmann, M. A . Moreno-Mateos, Configurational forces explain echelon cr acks in soft materials, arXiv preprint arXiv:2507.12247 (2025)
2025 arXiv
-
[45]
Slootman, V
J. Slootman, V. W altz, C. J. Yeh, C. Baumann, R. G¨ ostl, J . Comtet, C. Creton, Quantifying rate-and temperature- dependent molecular damage in elastomer fracture, Physica l Review X 10 (2020) 041045. 13
2020
Reviewed August 3, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.