How Geometry Tames Disorder in Lattice Fracture

Alessandra Lingua; Antoine Sanner; David S. Kammer; Leo de Waal; Marcelo A. Dias; Matthaios Chouzouris

REVIEW 3 major objections 4 minor 47 references

How Geometry Tames Disorder in Lattice Fracture

T0 review · 3 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The slenderness ratio of a triangular beam lattice controls how Weibull-strength disorder is expressed, yielding three fracture regimes in one phase diagram.

desk verdict A clean, largely parameter-free theory for how slenderness ratio controls crack-path disorder in the weak-to-moderate disorder regime; the strong-disorder part of the phase diagram rests on simulation, but the core result stands. read the letter →

arxiv 2602.09737 v1 pith:JCAKXJLV submitted 2026-02-10 cond-mat.mtrl-sci cond-mat.softcond-mat.stat-mechphysics.app-ph

Matthaios Chouzouris , Leo de Waal , Antoine Sanner , Alessandra Lingua , David S. Kammer , Marcelo A. Dias This is my paper

classification cond-mat.mtrl-scicond-mat.softcond-mat.stat-mechphysics.app-ph

keywords beamlatticeWeibulldisorderslendernessratiofracturetoughnesscrackpathdiffusefailuredamageevolutiontriangular

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that in a pre-cracked triangular beam lattice with Weibull-distributed failure stresses, the slenderness ratio (unit-cell size divided by beam thickness) is the geometric control parameter that determines how disorder expresses itself: high slenderness suppresses disorder, intermediate values cause local crack-tip scattering that roughens the crack, and low values allow initially diffuse bulk failure. The authors derive closed-form probabilities for these outcomes from pairwise Weibull failure statistics together with the crack-tip stress ratios, then test them against quasistatic lattice simulations. They also reproduce the disorder-induced increase in apparent fracture energy and find that, for Weibull moduli n at or above about 10, it collapses to G/G_u = 1 + B/n, independent of slenderness, while the absolute fracture energy depends non-monotonically on disorder. Their key negative result is that this toughening does not scale with the number of excess broken bonds or crack tortuosity, challenging the common explanation of disorder toughening by diffuse damage. The paper is explicit that it does not yet model the non-monotonic peak at strong disorder, and its own simulations show the independent-event approximation degrades for n below about 7 to 9, where diffuse damage is significant.

What carries the argument

The central object is the Slenderness Ratio λ ≡ a/t, the ratio of the triangular unit-cell size to the beam thickness, which sets the relative axial and bending stress content of the six crack-tip beams and therefore the stress ratios κ_ij among them. The central identity is the pairwise Weibull failure probability P(s_i > s_j) = κ_ij^n / (1 + κ_ij^n), obtained by absorbing κ_ij into a rescaling of the reference failure stress; this identity makes the statistics functions of λ and n only. The paper extends this pairwise rule to groups using the Weibull stress, the L^n norm of the element stresses, which lets it compare the whole crack-tip set against the bulk and thus predict diffuse failure

What would settle it

The independent-event assumption predicts that the scattering probability at a given λ and n is unaffected by pre-existing diffuse damage; therefore, a lattice in which a few random bulk bonds are removed before loading should show the same P_s as an undamaged one, and the G/G_u master curve should not shift. If either changes, the fixed-κ picture is wrong.

Watch

Extended reading notes

Core claim

The central claim is that the slenderness ratio λ = a/t, the ratio of the triangular unit-cell size to the in-plane beam thickness, acts as a geometric control parameter that selects among three fracture regimes in a disordered beam lattice: disorder suppressed, local crack-tip scattering, and initially diffuse failure. The mechanism is the reordering of the six crack-tip stress ratios with λ, because the Weibull survival function is scale-invariant, the probability that element i fails before element j is exactly the ratio κ_ij^n/(1 + κ_ij^n), where κ_ij is the fixed stress ratio. Summing these over the crack-tip elements and comparing the crack tip to the bulk through the Weibull stress yi

Load-bearing premise

The load-bearing premise is that the crack-tip stress ratios κ_ij and the bulk-to-tip stress ratio that enters the diffuse-failure probability stay fixed at their undamaged-lattice values, so each failure event sees the same stress hierarchy and successive events are statistically independent; the paper's own simulations show this premise degrades once the Weibull modulus falls below about 7 to 9, where diffuse damage alters the stress field.

Editorial extensions

If this is right

Choosing the slenderness ratio at fixed disorder strength lets a designer select the fracture regime: cracks can be forced straight, made locally tortuous, or driven to fail diffusely, independent of the statistical spread of failure strengths.
In the weak-to-moderate disorder regime (n ≳ 10), the disorder-induced toughening relative to the uniform lattice collapses onto a single 1/n curve, so apparent fracture energy enhancement can be predicted without fitting slenderness-dependent parameters.
The amount of excess damage along the crack path is not a proxy for fracture energy: lattices with very different degrees of tortuosity can show the same normalized toughening.
Below a Weibull modulus of about 7 to 9, diffuse damage accumulates before the main crack advances, the independent-event approximation breaks down, and geometric control of the crack path weakens.
Because the uniform reference fracture energy G_u depends on slenderness while the relative enhancement G/G_u does not, geometry and statistical disorder contribute separately to the fracture energy in the collapsed regime.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural extension is to pattern slenderness spatially across a lattice: gradients in λ should steer the crack through different regimes along a single specimen, effectively writing the crack path by design rather than by disorder realization.
The identity P = κ^n/(1 + κ^n) suggests an inverse-problem use: measuring the scattering probability as a function of n at fixed λ would recover the effective crack-tip stress ratio κ_s, making disorder a quantitative probe of crack-tip micromechanics.
The 1/n collapse hints that weak Weibull disorder acts like a spatially fluctuating local toughness field; a testable consequence is that any weak disorder distribution whose failure thresholds have variance proportional to 1/n should produce the same master curve.
The low-n breakdown could itself be modeled by letting κ_ij depend on accumulated diffuse damage, effectively renormalizing the crack-tip stress ratios as the bulk weakens; such an extension would move the phase diagram's strong-disorder boundary from an assumption into a prediction.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Dear colleague,

This one is worth a look. The genuinely new piece: geometry—specifically the slenderness ratio λ—is treated as a control variable for whether Weibull disorder in a triangular lattice expresses itself in the crack path, and the paper derives closed-form probabilities for scattering and diffuse failure from crack-tip stress ratios and Weibull order statistics. The theory lines have no free parameters and match simulations for n ≳ 7–9 across all tested λ. That is a real result, and the code/data are public.

What the paper does well. The three-element discretization is a sensible extension of the random beam model and is experimentally anchored to prior PMMA work. The derivation of Ps (Eq. 18) and Pd(0) (Eq. 14) is clean, and the renormalization via Weibull stress is a nice touch. The reproduction of disorder-induced toughening and the collapse of G/G_u onto 1 + B/n for n ≳ 10 is convincing evidence for the crack-arrest interpretation over simple tortuosity arguments. The authors are also unusually honest: they flag the breakdown at low n as “a notable feature” and speculate about the cause rather than sweeping it under the rug.

Soft spots, in proportion. The central weakness is exactly where the reader placed it: the frozen-stress-ratio and independent-event approximation is not derived, and the paper’s own Fig. 5 shows it under-predicts scattering for n ≲ 7–9, coinciding with where diffuse damage sets in. So regime (iii) and the strong-disorder part of the phase diagram are empirical observations, not predictions of the framework. The non-monotonic toughening peak lives in that same regime and is unexplained, and the 1/n law is fitted in the collapsed regime only (with B a free parameter). Also, the claim that crack-path morphology is system-size-independent is asserted but never tested by varying system size for disordered lattices; only Gu convergence in uniform lattices is tested. Those are addressable rather than fatal, but they should be acknowledged more carefully.

Who it’s for: anyone working on fracture in architected lattices or on disorder-controlled damage. The weak-to-moderate disorder regime is established, and the separation of morphology from energetics is worth arguing with.

I’d send it to a serious referee. My own verdict would be conditional: keep the strong-disorder claims descriptive, test the size-independence claim, and the paper will be solid.

Referee Report

3 major / 4 minor

Summary. The manuscript proposes a statistical framework for fracture of pre-cracked triangular beam lattices with Weibull-distributed beam strengths. The central idea is that the slenderness ratio λ = a/t controls the relative crack-tip stresses, and therefore controls whether disorder is expressed along the main crack ('scattering') or in the bulk ('diffuse failure'). Using the Weibull survival function, the authors derive closed-form probabilities P_s (Eqs. 18–19) and P_d^(0) (Eqs. 14–15), identify three regimes in Fig. 4, and validate against simulations in Fig. 5. They further report that normalized fracture energy G/G_u collapses to 1 + B/n for n ≳ 10 (Eq. 21), while absolute G is non-monotonic in n (Fig. 6b). They argue that disorder-induced toughening is not explained by crack tortuosity or damage count. The paper is clearly written, and the authors explicitly flag the regime where their model breaks down, but the abstract and final remarks present the three-regime picture as more broadly established than the supporting analysis warrants.

Significance. If correct, the paper establishes geometry as a design axis for controlling disorder expression and provides a parameter-free prediction of crack-path statistics at moderate disorder. Its concrete strengths include the overlaid theory lines in Fig. 5, which contain no fitted parameters, and the public availability of code and data. The paper also cleanly separates the geometric contribution G_u from the statistical contribution G/G_u. However, the framework's predictive range is limited: the assumption of independent events with frozen crack-tip stress ratios is explicitly shown to fail for Weibull moduli n ≲ 7–9, exactly where diffuse failure and the non-monotonic toughening maximum occur. The phase diagram and the strong-disorder portion of the central claim therefore rest on simulation data rather than on the statistical model, and the manuscript overstates the degree to which the theory explains them.

major comments (3)

[Sec. V, Eqs. (18)–(19); Fig. 5(a)] The scattering probability is constructed from crack-tip stress ratios κ_i(λ) computed in the undamaged lattice and assumes statistically independent failure events via Eq. (3). The paper's own Fig. 5(a) shows this prediction systematically under-predicts excess broken bonds for Weibull moduli n ≲ 7–9, the range where diffuse damage appears and the stress field seen by the crack changes. Because regime (iii) in Fig. 4 and the peak of G in Fig. 6(b) lie in this range, the phase-diagram boundaries and the strong-disorder portion of the central claim are not predictions of the framework; they are empirical observations. The authors note the breakdown, but the three-regime picture is presented as established in the abstract and final remarks. Please either restrict the theoretical claims to the validated regime or extend the model (e.g., by including the effect of pre-existing diffuse damage
[Sec. VI B, Eq. (21); Fig. 6(b)] The master curve G/G_u = 1 + B/n is obtained by fitting B to data normalized by the simulated G_u; B is not predicted from the model. The explanation via crack-arrest scaling with the standard deviation of Weibull strengths (∼1/n) is an analogy, not a derivation from the lattice model. Moreover, the claim that toughening is decoupled from the amount of damage/tortuosity is not demonstrated by a direct comparison: the paper shows that G/G_u collapses while P_s varies with λ, but it does not plot G/G_u against the measured excess-damage fraction or effective crack path. To make this central interpretative claim quantitative, include such a plot or correlation test, or soften the wording to say the data are not directly correlated rather than 'cannot be connected' to damage.
[Sec. IV C, Eqs. (14)–(15), (17); Fig. 5(b)] The comparison in Fig. 5(b) is less direct than stated. P_d^(0) is exact only for the first failure event (as noted in the text), yet the simulation values are y-intercepts of fits of Eq. (17), which contains free parameters β and γ. The theoretical line for P_d^(0) also depends on a representative bulk stress σ̄ and the effective volume V/V0, whose values are not specified in the main text. Please provide either a direct measurement of the first-failure probability from simulations or a clear statement of how σ̄ and V/V0 are obtained, so the reader can assess whether the agreement in Fig. 5(b) is a parameter-free test.

minor comments (4)

[Abstract and Sec. II] Typographical issues: 'theSlenderness Ratio' in the abstract, and Sec. II's 'mode axial and bending failure stresses' should likely be 'moduli' or 'ultimate axial and bending failure stresses'. Please define λ at first use.
[Eq. (5)] The hazard-rate construction uses W(σ_i; σ_0, n) and S(σ_j; σ_0, n); please state explicitly that S is the survival probability of element j and that the two failure thresholds are assumed independent and identically distributed.
[Appendix D] The nonlinear correction in Appendix D is useful, but it is not incorporated into the main predictions. State explicitly that the main-text results assume constant κ_ij and that the nonlinear correction is a separate, non-validated extension.
[Appendix C] The hierarchical agglomerative clustering cutoff of 1.1a is central to extracting the main-crack cluster, but no sensitivity analysis is given. A brief robustness check (varying the cutoff) would increase confidence in the reported N_f and excess-damage ratios.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Ps and Pd predictions are parameter-free overlays of Weibull pair-comparison statistics with undamaged-lattice crack-tip stress ratios, and the validation is a genuine test of the independence approximation; the 1+B/n toughening law is explicitly a fitted scaling law rather than a prediction.

full rationale

The paper's derivation chain is not circular in any of the enumerated senses. The central probabilities Ps (Eq. 18) and Pd^(0) (Eq. 14) are derived from the standard Weibull pair-comparison result Eq. (8), which is obtained by integrating the joint failure-rate expression Eq. (7). The inputs are the Weibull modulus n and the crack-tip stress ratios κ_i(λ) obtained from an undamaged micromechanical analysis (Sec. V). These stress ratios are computed, not fitted to the failure statistics or to the damage morphology data. The subsequent comparison in Fig. 5 is explicitly described as having 'no free parameters' and being 'overlaid, not fitted to the data,' so the agreement is a real test of the independent-event and linear-scaling approximations. The paper itself flags the breakdown of this agreement for n ≲ 7–9, which is an acknowledged limitation of the model rather than a circular step. The toughening law G/G_u = 1 + B/n is openly presented as a fit ('with the fitted curve plotted as a dashed black line') and is only post hoc connected to the ∼1/n scaling of the Weibull standard deviation; it is not disguised as a parameter-free prediction. The non-monotonic peak in G(n) is also explicitly left without a mechanistic model. The self-citations to [6] and [18] provide the beam-discretization model and material parameters, and [6] is an experimental validation; these are load-bearing modeling choices but not used as an unverified uniqueness claim or an external mathematical theorem. No equation reduces to another by construction, and no fitted parameter is renamed as a prediction. The low-n breakdown is a correctness/scope concern, not circularity.

Assumptions & free parameters 4 free parameters · 7 assumptions · 0 invented entities

The central predictions (P_s, P_d^{(0)}) are genuinely derived from Weibull statistics and geometry-dependent stress ratios, but the stress ratios themselves come from the same model family used for validation, and the empirical curves (1/n master curve, diffuse damage evolution) add fitted parameters. No new physical entities are introduced.

free parameters (4)

B
Fitted constant in the master curve G/G_u = 1 + B/n (Eq. 21) characterizing disorder-induced toughening in the weak-disorder regime.
A and G_∞
Free parameters in the empirical finite-size convergence form G_u = G_∞ A^(1/N_y) (Eq. 20), used to choose simulation system size.
β, γ
Free parameters in the empirical evolution of diffuse failure probability P_d(N_f) = P_d^(0)/(1 + β N_f^(γ/n)) (Eq. 17), used to extract P_d^(0) from data.
HAC distance cutoff = 1.1 a
Termination criterion for hierarchical clustering that defines the main crack cluster; chosen by hand (10% margin above one unit-cell spacing).

assumptions (7)

domain assumption Failure stresses of beams are independent and identically distributed Weibull variates
Standard for brittle fracture (Sec. II, Eq. 1-2); the entire statistical framework relies on it.
domain assumption Stresses scale linearly with a common load factor, so stress ratios κ_ij = σ_i/σ_j are constant
Assumed in Sec. IV A; corrected to first order in Appendix D but corrections are argued negligible for the system.
ad hoc to paper The crack-tip stress ratios σ_k(λ) are taken from the undamaged uniform lattice and remain representative during crack advance
Used in Eqs. 18-19 and Sec. V; this is the independence/self-similarity assumption, validated only at n≳7-9.
domain assumption The bulk stress can be represented by a single value σ̄ (volume-averaged) to compute diffuse failure probability
Assumed in Sec. IV C; footnote 28 notes a position-dependent σ̄(r) gives same qualitative result.
ad hoc to paper Successive failure events are statistically independent, so expected number of broken bonds is N_f = 2N_h(1+P_s)
Eq. 3 in Sec. III; the paper calls it an assumption and tests it, finding breakdown at low n.
domain assumption Total fracture energy per horizontal crack extension is an adequate measure of energy release rate, G = dU/ds
Defined in Sec. VI B; choice is motivated by mode-I loading but is non-trivial for tortuous cracks.
standard math Weibull coefficient of variation ~1/n for n≥4
Used to explain the 1/n master-curve scaling (Sec. VI B); asymptotic result, not exact for small n.

how reviews work

0 comments

Cite this review

Pith. "Pith review of How Geometry Tames Disorder in Lattice Fracture." pith.science (2026). https://pith.science/paper/JCAKXJLV

@misc{pith2026260209737,
  author       = {Pith},
  title        = {Pith review of: How Geometry Tames Disorder in Lattice Fracture},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JCAKXJLV}},
  note         = {Machine review of arXiv:2602.09737}
}

read the original abstract

We investigate the fracture behavior of pre-cracked triangular beam-lattices whose elements have failure stresses drawn from a Weibull distribution. Through a statistical analysis and numerical simulations, we identify and verify the existence of three distinct failure regimes: (i) disorder is effectively suppressed, (ii) disorder manifests locally near the crack tip, modifying the crack morphology, and (iii) disorder manifests globally, leading to initially diffuse failure. Our model naturally reveals the key parameters governing this behavior: the Weibull modulus, quantifying the spread in failure thresholds, and a geometric quantity termed the Slenderness Ratio. We also reproduce the disorder-induced toughening reported in previous experimental and numerical studies, further demonstrating that its manifestation depends non-monotonically on disorder. Crucially, our results indicate that this toughening cannot be simply connected to the amount of damage in the lattice, challenging interpretations that attribute increased fracture energy solely to enhanced crack tortuosity or diffuse failure. Overall, our results establish geometry as a powerful control parameter for regulating how disorder is expressed during fracture in beam-lattices, with broader implications for the disorder-induced toughening in engineered materials.

Figures

Figures reproduced from arXiv: 2602.09737 by the authors.

**Figure 2.** Micromechanics and damage evolution after an anomalous failure event. (a): Crack-tip stress hierarchy [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Renormalization of a set G of m elements. The resulting element can be regarded to be at a Weibull stress ¯σG ≥ max({σ1, σ2, ...σm}), i.e., larger than any of the stresses of its constituents. The resulting element survives iff all of its constituents also survive. For our study, this is important as it means that Eq. (8) is also relevant for describing the failure statistics of two groups (G1, G2) of elements, with… view at source ↗

Figures from the paper (6 more)

**Figure 4.** Figure 4: Fracture regimes resulting from the mechanically informed statistical analysis. At each point in parameter [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

**Figure 5.** Figure 5: Simulation data plotted over theoretical predictions for the probabilities of scattering and diffuse failure. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

**Figure 6.** Figure 6: Energy release rate G in uniform and disordered lattices. (a): Convergence of Gu/G∞ with horizontal system size Nx for different SR in uniform—n → ∞— lattices; the inset shows G∞ vs SR. The smallest Nx for which Gu ≥ 0.95G∞ is used in simulations for subsequent measu…

**Figure 8.** Figure 8: Splitting of elements in a lattice when element [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]

**Figure 9.** Figure 9: Visualization of scattering map and map of row [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]

**Figure 10.** Figure 10: Resulting damage point segregation after per [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

47 extracted references · 1 canonical work pages

[1]

This is exactly proportional to the inverse of a non-dimensionalizedStress Intensity Factor(SIF) YSIF =K SIF/¯σ∝σ 1/¯σ, often called the dimen- sionless or geometrical SIF [29]

The ratio of the (mean) stress in the bulk, to the peak stress on the crack tip - ¯σ/σ 1. This is exactly proportional to the inverse of a non-dimensionalizedStress Intensity Factor(SIF) YSIF =K SIF/¯σ∝σ 1/¯σ, often called the dimen- sionless or geometrical SIF [29]
[2]

This term be- comes larger, the more even the stresses are on the crack-tip, but is still relatively close to 1 for large n(e.g., forn= 4, the largest it can become is 61/4 ∼1.565

The relative stresses of the elements on the crack tip - ( P i κn i )1/n ∈[1,6],∀n≥1. This term be- comes larger, the more even the stresses are on the crack-tip, but is still relatively close to 1 for large n(e.g., forn= 4, the largest it can become is 61/4 ∼1.565. . .). Ignoring the dependence on the latter term, which for our setup (n≥4) and thecrack-t...
[3]

Zheng, H

X. Zheng, H. Lee, T. H. Weisgraber, M. Shusteff, J. DeOtte, E. B. Duoss, J. D. Kuntz, M. M. Biener, Q. Ge, J. A. Jackson, S. O. Kucheyev, N. X. Fang, and C. M. Spadaccini, Ultralight, ultrastiff mechanical meta- 12 materials, Science (American Association for the Ad- vancement of Science)344, 1373 (2014)

2014
[4]

Y. Lyu, X. Song, H. Wang, and J. Jiang, A novel mechan- ical metamaterial with tunable stiffness and individually adjustable poisson’s ratio, Materials today communica- tions40, 110135 (2024)

2024
[5]

Schwaiger, L

R. Schwaiger, L. Meza, and X. Li, The extreme mechanics of micro- and nanoarchitected materials, MRS bulletin 44, 758 (2019)

2019
[6]

Rafsanjani and D

A. Rafsanjani and D. Pasini, Bistable auxetic mechani- cal metamaterials inspired by ancient geometric motifs, Extreme Mechanics Letters9, 291 (2016)

2016
[7]

Zhang, X

H. Zhang, X. Guo, J. Wu, D. Fang, and Y. Zhang, Soft mechanical metamaterials with unusual swelling behav- ior and tunable stress-strain curves, Science advances4, eaar8535 (2018)

2018
[8]

de Waal, M

L. de Waal, M. Chouzouris, and M. A. Dias, Cracking down on fracture to functionalize damage, Physical Re- view Letters135, 148202 (2025)

2025

Show all 47 references

[9]

M. R. Khosravani, D. Anders, M. R. Ayatollahi, and T. Reinicke, Fabrication of mechanical metamaterials by 3d printing: recent advancements and current challenges, Archives of Civil and Mechanical Engineering25, 244 (2025)

2025
[10]

Masuo, Y

H. Masuo, Y. Tanaka, S. Morokoshi, H. Yagura, T. Uchida, Y. Yamamoto, and Y. Murakami, Influence of defects, surface roughness and hip on the fatigue strength of ti-6al-4v manufactured by additive manufacturing, In- ternational Journal of Fatigue117, 163 (2018)

2018
[11]

B. C. d. S. Silva, B. Callegari, L. F. Seixas, M. Kr´ ol, W. Sitek, G. Matula, L. Krzemi´ nski, R. S. Coelho, and G. F. Batalha, Investigation of distortion, porosity and residual stresses in internal channels fabricated in marag- ing 300 steel by laser powder bed fusion, Mate...

2025
[12]

Lertthanasarn, C

J. Lertthanasarn, C. Liu, and M.-S. Pham, Mechani- cal behaviour of additively manufactured ti6al4v meta- crystals containing multi-scale hierarchical lattice struc- tures, arXiv preprint arXiv:2011.14201 (2020)

2011 arXiv
[13]

Shekhawat, S

A. Shekhawat, S. Zapperi, and J. P. Sethna, From dam- age percolation to crack nucleation through finite size criticality, Physical review letters110, 185505 (2013)

2013
[14]

Ziemke, O

P. Ziemke, O. Finney, R. G. Chambers, R. Thiraux, L. Valdevit, and M. R. Begley, The defect sensitivity of brittle truss-based metamaterials, Materials & Design 239, 112776 (2024)

2024
[15]

Fulco, P

S. Fulco, P. K. Purohit, M. K. Budzik, and K. T. Turner, Fracture of disordered and stochastic lattice materials, arXiv preprint arXiv:2508.21187 (2025)

2025 arXiv
[16]

Urabe and S

C. Urabe and S. Takesue, Fracture toughness and max- imum stress in a disordered lattice system, Physical Re- view E—Statistical, Nonlinear, and Soft Matter Physics 82, 016106 (2010)

2010
[17]

Fulco, M

S. Fulco, M. K. Budzik, H. Xiao, D. J. Durian, and K. T. Turner, Disorder enhances the fracture toughness of 2d mechanical metamaterials, PNAS nexus4, pgaf023 (2025)

2025
[18]

C. M. Hartquist, S. Wang, B. Deng, H. K. Beech, S. L. Craig, B. D. Olsen, M. Rubinstein, and X. Zhao, Fracture of polymer-like networks with hybrid bond strengths, Journal of the Mechanics and Physics of Solids195, 105931 (2025)

2025
[19]

M. J. Alava, P. K. Nukala, and S. Zapperi, Role of dis- order in the size scaling of material strength, Physical review letters100, 055502 (2008)

2008
[20]

de Waal, M

L. de Waal, M. Chouzouris, and M. A. Dias, Architect- ing mechanisms of damage in topological metamaterials, Physical Review Research7, 033177 (2025)

2025
[21]

W. Weibull,A statistical theory of the strength of mate- rials, Ingeni¨ ors vetenskapsakademiens handlingar (Gen- eralstabens litografiska anstalts f¨ orlag, Stockholm, 1939) oCLC: 30416455

1939
[22]

Bertalan, A

Z. Bertalan, A. Shekhawat, J. P. Sethna, and S. Zapperi, Fracture strength: stress concentration, extreme value statistics, and the fate of the weibull distribution, Phys- ical Review Applied2, 034008 (2014)

2014
[23]

Sanner, L

A. Sanner, L. Michel, A. Lingua, and D. S. Kammer, Less is more: removing a single bond increases the tough- ness of elastic networks, International Journal of Fracture 249, 1 (2025)

2025
[24]

Charles and F

Y. Charles and F. Hild, On crack arrest in ceramic/metal assemblies, International journal of fracture115, 251 (2002)

2002
[25]

Curtin and K

W. Curtin and K. Futamura, Microcrack toughening?, Acta Metallurgica et Materialia38, 2051 (1990)
[26]

Weihull, A statistical distribution function of wide applicability, J Appl Mech18, 290 (1951)

W. Weihull, A statistical distribution function of wide applicability, J Appl Mech18, 290 (1951)

1951
[27]

S. P. Timoshenko and J. M. Gere,Theory of elastic sta- bility(Courier Corporation, 2012)

2012
[28]

Andrieu, A

A. Andrieu, A. Pineau, J. Besson, D. Ryckelynck, and O. Bouaziz, Beremin model: Methodology and applica- tion to the prediction of the euro toughness data set, Engineering fracture mechanics95, 102 (2012)

2012
[29]

K. G. Wilson, The renormalization group: Critical phe- nomena and the kondo problem, Reviews of modern physics47, 773 (1975)

1975
[30]

We avoid it to de-clutter the notation

A renormalization can also be performed considering a different ¯σat each location (exp −(1/V0) R V (¯σ(r)/σ0)n d3r ), which would just lead to a different definition of ¯σ B, but the same qualitative result. We avoid it to de-clutter the notation
[31]

Andrasic and A

C. Andrasic and A. Parker, Dimensionless stress intensity factors for cracked thick cylinders under polynomial crack face loadings, Engineering Fracture Mechanics19, 187 (1984)

1984
[32]

N. A. Fleck and X. Qiu, The damage tolerance of elastic– brittle, two-dimensional isotropic lattices, Journal of the Mechanics and Physics of Solids55, 562 (2007)

2007
[33]

Berkache, S

K. Berkache, S. Phani, and J.-F. Ganghoffer, Micropo- lar effects on the effective elastic properties and elastic fracture toughness of planar lattices, European Journal of Mechanics-A/Solids93, 104489 (2022)

2022
[34]

Murtagh and P

F. Murtagh and P. Contreras, Algorithms for hierarchical clustering: an overview, Wiley interdisciplinary reviews: data mining and knowledge discovery2, 86 (2012)

2012
[35]

X. Li, L. Men, Y. Yu, Z. Hou, and Z. Wang, Specimen size effect on the fracture energy of architected stretch- able materials, International Journal of Smart and Nano Materials14, 420 (2023)

2023
[36]

Sreejith, K

P. Sreejith, K. Kannan, and K. Rajagopal, A ther- modynamic framework for additive manufacturing, us- ing amorphous polymers, capable of predicting residual stress, warpage and shrinkage, International Journal of Engineering Science159, 103412 (2021)

2021
[37]

Issametova, N

M. Issametova, N. V. Martyushev, A. Zhastalap, L. B. Sabirova, U. Assemgul, A. Tursynbayeva, and G. Abile- zova, Determination of residual stresses in 3d-printed polymer parts, Polymers16, 2067 (2024). 13
[38]

Taherkhani, C

K. Taherkhani, C. Eischer, and E. Toyserkani, An un- supervised machine learning algorithm for in-situ defect- detection in laser powder-bed fusion, Journal of Manu- facturing Processes81, 476 (2022)

2022
[39]

Maleki, S

E. Maleki, S. Bagherifard, M. Bandini, and M. Guagliano, Surface post-treatments for metal additive manufacturing: Progress, challenges, and opportunities, Additive Manufacturing37, 101619 (2021)

2021
[40]

X. Yan, R. Lupoi, H. Wu, W. Ma, M. Liu, G. O’Donnell, and S. Yin, Effect of hot isostatic pressing (hip) treat- ment on the compressive properties of ti6al4v lattice structure fabricated by selective laser melting, Materials Letters255, 126537 (2019)

2019
[41]

Chouzouris, Simlad—simulator for lattice damage, https://github.com/McHouzou/SimLaD.git(2026)

M. Chouzouris, Simlad—simulator for lattice damage, https://github.com/McHouzou/SimLaD.git(2026)

2026
[42]

Chouzouris, L

M. Chouzouris, L. de Waal, A. Sanner, A. Lingua, D. Kammer, and M. A. Dias, How geometry tames dis- order in lattice fracture: Data, 10.5281/zenodo.18482638 (2026)

2026 doi
[43]

Bitzek, P

E. Bitzek, P. Koskinen, F. G¨ ahler, M. Moseler, and P. Gumbsch, Structural relaxation made simple, Phys- ical review letters97, 170201 (2006)

2006
[44]

Gu´ enol´ e, W

J. Gu´ enol´ e, W. G. N¨ ohring, A. Vaid, F. Houll´ e, Z. Xie, A. Prakash, and E. Bitzek, Assessment and optimiza- tion of the fast inertial relaxation engine (fire) for energy minimization in atomistic simulations and its implemen- tation in lammps, Computational Materials Sc...

2020
[45]

S. Luan, E. Chen, and S. Gaitanaros, Energy-based frac- ture mechanics of brittle lattice materials, Journal of the Mechanics and Physics of Solids169, 105093 (2022)

2022
[46]

σ′′ i (0) κ(0) ij −σ ′′ j (0) # = 2 α(1) j 2

A. Lingua, A. Sanner, F. Hild, and D. S. Kammer, Break- ing better: Imperfections increase fracture resistance in architected lattices, arXiv preprint arXiv:2504.08873 (2025). 14 Appendix A: Method We adopt the modeling approach described in [18], which has been experimentally...

2025 arXiv
[47]

Theory We begin by considering the underlying survival distri- bution of the bulk elements, while they are being progres- sively and uniformly stressed until a certain small amount of them has failed. Although there are similar discus- sions in the literature regarding the cra...

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.

[1] [1]

This is exactly proportional to the inverse of a non-dimensionalizedStress Intensity Factor(SIF) YSIF =K SIF/¯σ∝σ 1/¯σ, often called the dimen- sionless or geometrical SIF [29]

The ratio of the (mean) stress in the bulk, to the peak stress on the crack tip - ¯σ/σ 1. This is exactly proportional to the inverse of a non-dimensionalizedStress Intensity Factor(SIF) YSIF =K SIF/¯σ∝σ 1/¯σ, often called the dimen- sionless or geometrical SIF [29]

[2] [2]

This term be- comes larger, the more even the stresses are on the crack-tip, but is still relatively close to 1 for large n(e.g., forn= 4, the largest it can become is 61/4 ∼1.565

The relative stresses of the elements on the crack tip - ( P i κn i )1/n ∈[1,6],∀n≥1. This term be- comes larger, the more even the stresses are on the crack-tip, but is still relatively close to 1 for large n(e.g., forn= 4, the largest it can become is 61/4 ∼1.565. . .). Ignoring the dependence on the latter term, which for our setup (n≥4) and thecrack-t...

[3] [3]

Zheng, H

X. Zheng, H. Lee, T. H. Weisgraber, M. Shusteff, J. DeOtte, E. B. Duoss, J. D. Kuntz, M. M. Biener, Q. Ge, J. A. Jackson, S. O. Kucheyev, N. X. Fang, and C. M. Spadaccini, Ultralight, ultrastiff mechanical meta- 12 materials, Science (American Association for the Ad- vancement of Science)344, 1373 (2014)

2014

[4] [4]

Y. Lyu, X. Song, H. Wang, and J. Jiang, A novel mechan- ical metamaterial with tunable stiffness and individually adjustable poisson’s ratio, Materials today communica- tions40, 110135 (2024)

2024

[5] [5]

Schwaiger, L

R. Schwaiger, L. Meza, and X. Li, The extreme mechanics of micro- and nanoarchitected materials, MRS bulletin 44, 758 (2019)

2019

[6] [6]

Rafsanjani and D

A. Rafsanjani and D. Pasini, Bistable auxetic mechani- cal metamaterials inspired by ancient geometric motifs, Extreme Mechanics Letters9, 291 (2016)

2016

[7] [7]

Zhang, X

H. Zhang, X. Guo, J. Wu, D. Fang, and Y. Zhang, Soft mechanical metamaterials with unusual swelling behav- ior and tunable stress-strain curves, Science advances4, eaar8535 (2018)

2018

[8] [8]

de Waal, M

L. de Waal, M. Chouzouris, and M. A. Dias, Cracking down on fracture to functionalize damage, Physical Re- view Letters135, 148202 (2025)

2025