Iterative convergence in phase-field brittle fracture computations: exact line search is all you need

Francesco Vicentini; Jonas Heinzmann; Laura De Lorenzis; Pietro Carrara

REVIEW 3 major objections 4 minor 78 references

Iterative convergence in phase-field brittle fracture computations: exact line search is all you need

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 ↗ Review JSON

Pith's one-line read Alternate minimization for phase-field brittle fracture can be made globally convergent by equipping each Newton solve with an exact line search that minimizes the energy along the search direction.

desk verdict Solid, honest, and useful paper on exact line search for phase-field fracture; but the printed bisection algorithm has a sign bug in its termination test that breaks the exactness assumption, and the title oversells the unproved cases. read the letter →

arxiv 2511.23064 v2 pith:EO3HA37I submitted 2025-11-28 cs.CE

Jonas Heinzmann , Francesco Vicentini , Pietro Carrara , Laura De Lorenzis This is my paper · ORCID

classification cs.CE MSC 74R1065K0565H10

keywords phase-fieldfracturebrittlealternateminimizationNewton'smethodexactlinesearchbisectionnon-convexiterativeconvergence

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

The paper targets a practical breakdown: in staggered (alternate minimization) solution of variational phase-field brittle fracture, the per-subproblem Newton solvers often fail to converge, stalling the whole computation. It claims the failure is a step-length problem, not a direction problem: full Newton steps overshoot the one-dimensional minimum of the energy along the Newton direction. The proposed fix is an exact line search that locates this minimum by bisection on the directional derivative, which is monotone when the subproblem energy is strictly convex. Under strict convexity, coercivity, and uniform spectral bounds on the Hessian, Newton with this line search is proven to drive the residual norm to zero, so the outer alternate-minimization loop is guaranteed to reach a critical point. The same recipe is shown numerically to resolve all tested benchmark failures, including the star-convex energy decomposition for which the theoretical proof does not apply, and a demanding 3D Brazilian test with contact.

What carries the argument

The load-bearing mechanism is the exact line search implemented by bisection on the directional derivative φ'(λ)=R_w(w+λΔw)·Δw. Along a Newton direction of a strictly convex subproblem energy, φ is strictly convex, so φ' is monotone and has at most one root in (0,1], which is the energy minimizer. Bisection converges linearly with bracket error |λ_l−λ⋆|≤2^{-l}. The global-convergence proof then couples this with the eigenvalue bounds ρ_L≤λ_min(K_ww) and ρ_U≥λ_max(K_ww): an upper bound on the energy decrease per iteration, E(w_{k+1})−E(w_k) ≤ −(1/2)(ρ_L/ρ_U²)||R_w(w_k)||², telescopes into a finite sum of squared residual norms, forcing the residual to zero.

What would settle it

Run the bisection line search on a strictly convex subproblem instance and record the Hessian eigenvalue range along the Newton iterates: a violation of (34), or a Newton direction along which φ' is not monotone (detectable by sampling φ' at 100 points and finding a sign-change pattern inconsistent with a single root), would invalidate the global-convergence guarantee. Concretely, the residual norm failing to decrease to zero on such an instance—or the bisection bracketing losing the root because φ' changes sign more than once—would refute Proposition 2's scope.

Watch

Extended reading notes

Core claim

The central discovery is that Newton's method with an exact line search—minimizing the energy along the Newton update direction—converges globally for the strictly convex displacement and damage subproblems that arise in phase-field brittle fracture. The paper proves that the sequence of residual norms generated by this 'greedy' Newton converges to zero whenever the Hessian eigenvalues stay within a uniform interval [ρ_L, ρ_U]; strict convexity then gives convergence to the unique subproblem minimizer. Consequently, if both subproblems are strictly convex, alternate minimization with exact line search is guaranteed to converge to a critical point of the total energy. For cases outside the pr

Load-bearing premise

The proof rests on assuming that, for every Newton iterate, the Hessian eigenvalues stay inside a fixed interval [ρ_L,ρ_U] and that the energy along the search direction is strictly convex so its derivative is monotone; for the star-convex and reduced-space cases only numerical sampling supports these assumptions.

Editorial extensions

If this is right

With the exact line search active, every benchmark in the obstacle course converges — including the volumetric–deviatoric, spectral, no-tension, Drucker–Prager-like, and star-convex splits — where plain Newton previously stalled in the mechanical subproblem.
When irreversibility is enforced by the reduced-space active-set method, applying the line search to the mechanical subproblem alone is sufficient for convergence in all tested cases; the damage subproblem remains linear and needs no line search.
When irreversibility is enforced by penalization, the damage subproblem becomes strongly nonlinear and the exact line search must also be applied there for several cases; with it, all cases converge.
In the tested cases the bisection line search needs fewer total Newton iterations than backtracking or secant line searches, at the cost of more residual evaluations per iteration, which becomes cheaper relative to the linear solve as problems grow.
The approach computed the 3D Brazilian test with the star-convex model and contact constraints, producing a branched crack pattern within a single load step — a case that previously could not be computed.

Reading between the lines

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

The proof's ingredients — strict convexity, coercivity, uniform Hessian spectrum — are generic, so the same bisection line search should globalize Newton's method for any strictly convex finite-element minimization, not just phase-field fracture; a direct test on a different convex nonlinear PDE problem would verify that.
The paper shows numerical reliability for the star-convex γ⋆>0 case but replaces the proof with a 100-point monotonicity check of φ'; a sharper investigation would seek whether non-convexity can actually occur and, if so, whether the bisection still converges or needs a bracketing fix.
The claim 'exact line search is all you need' implicitly argues that globalization can be handled by the line search alone, without trust regions or inertia corrections; comparing this method head-to-head against such monolithic globalization strategies on the same tests would test that stronger claim.
Because the line search minimizes along each Newton direction, the method's energy-monotonicity may also help with second-order stability checks (assessing whether the found critical point is a local minimizer), a connection the paper notes but does not exploit.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Here's the short version: this is a genuinely useful paper, and it should go to review. The authors don't hide what they can't prove, the benchmarks are extensive, and the code is public. But the pseudocode for their bisection line search (Algorithm 2) omits the absolute value in the termination test, and as written that makes their own convergence proof inapplicable to the method they specify. That needs fixing before publication.

What's actually new: the systematic application of exact bisection line search to the two subproblems of staggered phase-field fracture, including the constraint-aware variant for active-set and penalty formulations, plus the 3D Brazilian test with contact and the star-convex model. The convergence result (Proposition 2) is classical in spirit—greedy Newton is not new, see Shea & Schmidt—but the adaptation to phase-field subproblems and the empirical demonstration of robustness across many energy splits is a real contribution. The paper is honest: Remark 4 and Section 4.2 explicitly say the reduced-space active-set and star-convex gamma*>0 cases are not covered by the proof, and they rely on a 100-point sampling of phi' instead. Good.

The soft spots:

1. Algorithm 2. The exit condition is `phi'(lambda_l)/||Delta w||^2 <= atol`, not `|phi'(lambda_l)|/...`. For any lambda left of the minimizer, phi' is negative and hence trivially <= positive atol. So if the first bisection point 0.5 lies left of the root, the loop exits at 0.5 without bracketing. That means the printed algorithm does not perform exact line search in general, and the key inequality (50) in the proof does not hold for the algorithm as specified. If the PETSc code actually uses abs, then the pseudocode is wrong; if it follows the pseudocode, the shipped solver isn't covered by Proposition 2. Either way it's a concrete fix, but it's load-bearing because the entire theoretical claim rests on exactness.

2. The title. 'Exact line search is all you need' overstates. The proof covers the strictly convex subproblems; the active-set and star-convex cases are empirical. The authors say this themselves, but the title and conclusions lean harder than the results.

3. Minor: the novelty boundary with the authors' own prior work [76] is not clearly drawn. Some readers will want a sentence on what's new relative to that paper.

The math is otherwise fine—Appendix D's proof is consistent under its stated assumptions, and the numerical study goes well beyond what's typical. This deserves a serious referee, but the reviewer should request the Algorithm 2 fix and a title/scope adjustment.

Referee Report

3 major / 4 minor

Summary. The paper addresses iterative non-convergence of Newton solvers inside staggered (alternate minimization) schemes for phase-field brittle fracture. It proposes to globalize each Newton solve by an 'exact' line search that minimizes the restricted energy φ(λ) along the Newton direction, using a bisection algorithm on the directional derivative φ'(λ). Two theoretical results are stated: Proposition 1 gives linear convergence of the bisection root for strictly monotone φ', and Proposition 2 gives convergence of the residual norm to zero for Newton's method with exact line search under strict convexity, coercivity, and uniform spectral bounds on the Hessian. The numerical part shows that the proposed line search resolves the reported convergence failures on an 'obstacle course' of five benchmarks, with several strain-energy decompositions and two irreversibility treatments, and demonstrates the method on a demanding 3D Brazilian test. The paper explicitly acknowledges that the convergence proof does not cover the reduced-space active-set strategy and the star-convex model with γ*>0, for which robustness is established empirically.

Significance. If the correctness gap identified below is closed, this is a practically valuable contribution: the bisection-based exact line search is simple, is contributed to PETSc, and the implementation is publicly available. The proof of Proposition 2 in Appendix D is internally consistent under its stated assumptions, and Proposition 1 is a standard bisection result. The numerical study is unusually broad for this area, covering multiple splits, two irreversibility strategies, and a large 3D problem, and the paper is honest about the cases not covered by the theory. These strengths make the manuscript worth serious consideration. However, the link between the proved statement and the algorithm as printed is broken by a concrete detail in the termination test of Algorithm 2, and this must be resolved before the central claim can be accepted.

major comments (3)

[Section 3.1, Algorithm 2] Algorithm 2 uses the termination test φ'(λ_l)/||Δw||² ≤ atol, not |φ'(λ_l)|/||Δw||² ≤ atol. Since φ'(0)<0 for a Newton direction, φ'(λ_l) is negative for every λ_l to the left of the true root, so any such λ_l automatically satisfies φ'(λ_l)/||Δw||² ≤ atol for positive atol. If the first bisection midpoint lies to the left of the minimizer, the algorithm exits immediately at that midpoint without bracketing the root. The returned step is then not the minimizer of (30), and the key inequality (50) — and with it the residual-norm convergence proof of Proposition 2 — does not apply to the algorithm as written. The pseudocode must be corrected to use |φ'(λ_l)|, or, if the PETSc implementation already uses the absolute value, the pseudocode and the text in Section 3.1 and Appendix F.2 must be corrected to match the implementation. As printed, the global convergence guarantee does not cover th
[Section 3.2 / Appendix D, Eq. (50)] Even after inserting the absolute value, Proposition 2 assumes an exact minimizer of φ over [0,1], whereas Algorithm 2 stops at finite tolerances atol and ltol. With fixed tolerances the returned λ* is only an approximate minimizer. The paper should state explicitly that the theorem is for the idealized exact line search, and clarify the relation to the finite-tolerance implementation (e.g., a limiting argument with tolerances tending to zero, or a separate convergence result for an inexact line search satisfying a sufficient-decrease condition). As written, the proof of (50) and the subsequent residual-norm bound apply only to the idealized algorithm, not directly to the stopping criteria actually used in the numerical experiments.
[Remark 4 / Eq. (34)] Remark 4 states that if the mechanical and damage subproblems are strictly convex, alternate minimization with exact line search is 'guaranteed' to converge to a critical point. However, Proposition 2 requires not only strict convexity but also the uniform spectral bounds (34) on the Hessian for all iterates. Strict convexity alone does not imply a positive uniform lower bound on the smallest eigenvalue over unbounded sets. The remark should explicitly carry over the spectral-bound assumption; otherwise the guarantee is stronger than the proof supports. This is a clarity issue in the main claim, though the proof itself is sound under the stated assumptions.

minor comments (4)

[Remark 4 / References] The citation '[1, Theorem 1]' for convergence of alternate minimization appears to point to the 2000 Bourdin–Francfort–Marigo paper; the relevant convergence theorem is more commonly associated with Bourdin's 2007 paper (reference [27]) or Burke–Ortner–Süli (reference [28]). Please verify and correct the citation.
[Section 3.1, Algorithm 2] The initialization 'λ_0 = λ_left_l = 0, λ_l = λ_right_l = 1' overloads λ_l as both the current iterate and the right bracket. The pseudocode is understandable but could be clearer, e.g. by using λ_mid for the evaluated point.
[Appendix F.2] The description of SNESLineSearchCP uses the same non-absolute termination test φ'(λ_l)/||Δw||² ≤ atol as Algorithm 2. If the shipped PETSc implementation does not take an absolute value, the efficiency comparison in Section 4.3 may be affected by the same early-exit behavior; please clarify.
[Section 4.2] The empirical convexity check by sampling 100 points per Newton iteration is a reasonable sanity check, but the wording 'could not find any evidence of non-convexity up to numerical tolerances' should be kept clearly separate from a proof. The current wording is mostly careful, but a sentence explicitly stating that this check is not a substitute for the missing proof would help avoid overstatement.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the global-convergence proof is self-contained; self-citations concern model validity and are not load-bearing; acknowledged proof gaps are correctness limitations, not circular derivations.

full rationale

The derivation chain is not circular. The central result, Proposition 2 (Section 3.2, proof in Appendix D), assumes strict convexity, coercivity, and uniform Hessian eigenvalue bounds (Eq. 34), plus an exact line search that minimizes the energy along the Newton direction; it then proves via the mean-value expansion (Eqs. 46–53) that the residual norms are square-summable and hence converge to zero. This is a substantive argument: the exactness assumption enters as Eq. (50), but it is not the same as the conclusion ||R_w(w_k)|| -> 0, and no fitted parameter or benchmark outcome is used in the proof. Proposition 1 is a standard bisection-convergence result under monotonicity of phi'. The paper explicitly limits its own guarantee: Remark 4 states that the result does not generally hold for the reduced-space active-set strategy or for the star-convex model with gamma* > 0, and Section 4.2 replaces the missing proof with empirical 100-point sampling of phi' monotonicity. That is an honest scope limitation, not circularity. Self-citations [29,39,66,76] concern the star-convex energy decomposition and earlier numerical studies; they are not used to justify the convergence theorem. The Algorithm 2 termination test `phi'(lambda_l)/||Delta w||^2 <= atol` omitting the absolute value is a genuine correctness/implementation concern (the implemented search may not be exact even in strictly convex subproblems), but it is a gap between the theorem's assumption and the pseudocode, not a reduction of the conclusion to its inputs; therefore it does not constitute circularity.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

No new physical entities are introduced. The theoretical claim borrows convexity and coercivity from the phase-field model and exactness of line search from classical optimization; the only paper-specific inputs are hand-chosen tolerances and an empirical treatment of the uncovered star-convex/active-set regimes.

free parameters (4)

Bisection line search tolerances atol, ltol, lmax = atol=1e-12, ltol=1e-6, lmax=20; relaxed to atol=1e-16, ltol=1e-10 for two cases
Hand-chosen to approximate exact line search; the convergence proof assumes exact minimization, so the practical guarantee depends on these being sufficiently tight, with adequacy demonstrated only empirically.
Newton and alternate-minimization tolerances TOL_NM, TOL_AM = TOL_NM=1e-8 (relaxed to 1e-7 in two cases), TOL_AM=1e-6
Convergence criteria defining 'success' in the numerical tables; not fitted to benchmark outcomes but they affect the reported convergence counts.
Irreversibility penalty tolerance TOL_ir = 1e-4
Sets the penalty parameter through the formula from [6]; the intentionally strict value challenges the damage solver and affects iteration counts.
Star-convex strength parameter gamma* = 1, 5 for benchmarks; 5.5 for Brazilian test
Material-model tuning parameter controlling tensile/compressive strength ratio; chosen to probe different regimes and to produce the Brazilian failure pattern, not fitted to the convergence claim.

assumptions (6)

domain assumption Strict convexity and coercivity of E_n(u,alpha) in each subproblem for 'classical' energy splits (Properties 1-2)
Required for uniqueness of subproblem minima and for Proposition 2. The paper asserts it for volumetric-deviatoric, spectral, no-tension, and DP-like splits; star-convex gamma*>0 is explicitly excluded in Section 4.2.
domain assumption Uniform spectral bounds rho_L <= rho_U on Kww(w) across all Newton iterates
Explicit assumption in Proposition 2 (Eq. 34); no finite-element-level verification is provided. If an iterate leaves the convex/spectral regime, the energy-decrease argument in Appendix D breaks.
domain assumption The line-search subproblem is solved exactly: phi(lambda) is strictly convex and bisection converges to the unique root of phi'
Bisection's convergence relies on monotone increasing phi' (Proposition 1). For the active-set projection, the directional derivative (32) is not proven monotone, so this assumption may fail outside the strictly convex regime.
standard math Convergence of the outer alternate minimization to a critical point when subproblems are solved successfully (Bourdin 2007; Burke-Ortner-Sueli 2010)
External theorem used to chain subproblem convergence into overall convergence in Remark 4; not re-proved.
ad hoc to paper Reduced-space active-set solver converges for the damage subproblem
The paper states that no general proof of convergence is available for this method ([49, p.161]); the overall guarantee therefore inherits an unproved component in all reduced-space computations.
domain assumption First-order stationarity (critical points) is the target; second-order minimality is not checked
Remark 1 restricts the entire study to the first-order KKT problem (17), so 'convergence' means stationary points, not necessarily local minimizers.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Iterative convergence in phase-field brittle fracture computations: exact line search is all you need." pith.science (2026). https://pith.science/paper/EO3HA37I

@misc{pith2026251123064,
  author       = {Pith},
  title        = {Pith review of: Iterative convergence in phase-field brittle fracture computations: exact line search is all you need},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EO3HA37I}},
  note         = {Machine review of arXiv:2511.23064}
}

read the original abstract

Variational phase-field models of brittle fracture pose a local constrained minimization problem of a non-convex energy functional. In the discrete setting, the problem is most often solved by alternate minimization, exploiting the separate convexity of the energy with respect to the two unknowns. This approach is theoretically guaranteed to converge, provided each of the individual subproblems is solved successfully. However, strong non-linearities of the energy functional may lead to failure of iterative convergence within one or both subproblems. We analyze and visualize the energy along Newton directions to illustrate why Newton's method without line search fails. Motivated by this, we propose to employ an exact line search algorithm based on bisection, which (under certain conditions) can guarantee global convergence of Newton's method for each subproblem and consequently the successful determination of critical points of the energy through the alternate minimization scheme. Through several benchmark tests computed with various strain energy decompositions and two strategies for the enforcement of the irreversibility constraint in two and three dimensions, we demonstrate the robustness of the approach and assess its efficiency in comparison with other commonly used line search algorithms. With the outlined approach, we are able to compute the especially demanding Brazilian test featuring contact in 3D with the star-convex model.

Figures

Figures reproduced from arXiv: 2511.23064 by the authors.

**Figure 1.** Oscillation of ∥Ru∥ during Newton iterations for the nucleation test with the volumetric-deviatoric split (a), and behavior of ∥Ru∥2(λ) and ϕ(λ) between the previous solution iterate and the full Newton step for selected iterations (b). In (b), the value of λ ∈ [0, 1] at which the energy is minimized is highlighted with a vertical dashed line. 3. Exact line search with bisection To determine a ’better’ step length, … view at source ↗

**Figure 2.** Iterations of the bisection line search algorithm for the illustrative example of Fig. 1b at Newton iteration [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗

**Figure 3.** Staggered paths for various cases from the obstacle course at selected load steps. For each constraint enforcement [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗

Figures from the paper (9 more)

**Figure 4.** Figure 4: Crack-driving contribution of the strain energy density [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]

**Figure 5.** Figure 5: Performance metrics for the nucleation test with the DP-like split (a), the plate with a hole with the star-convex [PITH_FULL_IMAGE:figures/full_fig_p018_5.png]

**Figure 6.** Figure 6: Brazilian test setup (a), strain states in the volumetric-deviatoric strain space (b), and reaction force (c). [PITH_FULL_IMAGE:figures/full_fig_p019_6.png]

**Figure 7.** Figure 7: Final phase-field state on the back (a), the left (b), the front (c) sides, and from a 3D perspective (d). For clarity, [PITH_FULL_IMAGE:figures/full_fig_p020_7.png]

**Figure 8.** Figure 8: Staggered convergence behavior for the Brazilian test during critical step [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]

**Figure 9.** Figure 9: Unrealistic phase-field solutions obtained for the plate with hole test with the spectral (a) and DP-like (b) splits, the [PITH_FULL_IMAGE:figures/full_fig_p029_9.png]

**Figure 10.** Figure 10: Sampling of the energy and directional derivative along the Newton direction for the mechanical problem, for selected [PITH_FULL_IMAGE:figures/full_fig_p033_10.png]

**Figure 11.** Figure 11: Sampling of the energy and directional derivative along the Newton direction for the damage problem with the [PITH_FULL_IMAGE:figures/full_fig_p034_11.png]

**Figure 12.** Figure 12: (a) Schematic of the backtracking line search, applied to the example of the energy in Fig. 1b at [PITH_FULL_IMAGE:figures/full_fig_p035_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

78 extracted references · 9 canonical work pages

[76]

Heinzmann, F

J. Heinzmann, F. Vicentini, P. Carrara, L. De Lorenzis, Improving convergence of phase-field fracture computations through line-search-enhanced alternate minimization, in: Advances and Challenges in Computational Mechanics, Springer, 2025.doi:10.1007/978-3-031-93213-7_12

work page doi:10.1007/978-3-031-93213-7_12 2025
[1]

Bourdin, G

B. Bourdin, G. Francfort, J.-J. Marigo, Numerical experiments in revisited brittle fracture, Journal of the Mechanics and Physics of Solids 48 (4) (2000) 797–826.doi:10.1016/S0022-5096(99)00028-9

doi:10.1016/s0022-5096(99)00028-9 2000
[2]

A. A. Griffith, The phenomena of rupture and flow in solids, Philosophical Transactions of the Royal Society of London 221 (1921) 163–198.arXiv:91192. URLhttps://www.jstor.org/stable/91192

1921
[3]

Francfort, J.-J

G. Francfort, J.-J. Marigo, Revisiting brittle fracture as an energy minimization problem, Journal of the Mechanics and Physics of Solids 46 (8) (1998) 1319–1342.doi:10.1016/S0022-5096(98)00034-9

doi:10.1016/s0022-5096(98)00034-9 1998
[4]

Ambrosio, V

L. Ambrosio, V. M. Tortorelli, On the approximation of free discontinuity problems, Bollettino della Unione Matematica Italiana 6 (1992) 105–123. 22

1992
[5]

K. Pham, H. Amor, J.-J. Marigo, C. Maurini, Gradient Damage Models and Their Use to Approximate Brittle Fracture, Int. J. Damage Mech. 20 (4) (2011) 618–652.doi:10.1177/1056789510386852

doi:10.1177/1056789510386852 2011
[6]

Gerasimov, L

T. Gerasimov, L. De Lorenzis, On penalization in variational phase-field models of brittle fracture, Computer Methods in Applied Mechanics and Engineering 354 (2019) 990–1026.doi:10.1016/j.cma. 2019.05.038

doi:10.1016/j.cma 2019
[7]

Svolos, J

L. Svolos, J. N. Plohr, G. Manzini, H. M. Mourad, On the convexity of phase-field fracture formulations: Analytical study and comparison of various degradation functions, International Journal of Non-Linear Mechanics 150 (2023).doi:10.1016/j.ijnonlinmec.2023.104359

arXiv 2023

Show all 78 references

[8]

Freddi, L

F. Freddi, L. Mingazzi, Mesh refinement procedures for the phase field approach to brittle fracture, Computer Methods in Applied Mechanics and Engineering 388 (2022).doi:10.1016/j.cma.2021. 114214

2022 doi
[9]

Gupta, U

A. Gupta, U. M. Krishnan, T. K. Mandal, R. Chowdhury, V. P. Nguyen, An adaptive mesh refinement algorithm for phase-field fracture models: Application to brittle, cohesive, and dynamic fracture, Com- puter Methods in Applied Mechanics and Engineering 399 (2022).doi:10.1016/j.c...

2022
[10]

P. K. Kristensen, E. Martínez-Pañeda, Phase field fracture modelling using quasi-newton methods and a new adaptive step scheme, Theoretical and Applied Fracture Mechanics 107 (2020).doi:10.1016/ j.tafmec.2019.102446

2020
[11]

J.-Y. Wu, Y. Huang, V. P. Nguyen, On the bfgs monolithic algorithm for the unified phase field damage theory, Computer Methods in Applied Mechanics and Engineering 360 (2020).doi:10.1016/j.cma. 2019.112704

2020
[12]

Rohracker, P

M. Rohracker, P. Kumar, P. Steinmann, J. Mergheim, Efficient phase-field fracture simulations for fracture analysis in heterogeneous materials, Computational Mechanics (2025).doi:10.1007/ s00466-025-02685-3

2025
[13]

Badri, G

M. Badri, G. Rastiello, E. Foerster, Preconditioning strategies for vectorial finite element linear systems arising from phase-field models for fracture mechanics, Computer Methods in Applied Mechanics and Engineering 373 (2021).doi:10.1016/j.cma.2020.113472

2021
[14]

Kopaničáková, H

A. Kopaničáková, H. Kothari, R. Krause, Nonlinear field-split preconditioners for solving monolithic phase-field models of brittle fracture, Computer Methods in Applied Mechanics and Engineering 403 (2023).doi:10.1016/j.cma.2022.115733

2023
[15]

Jodlbauer, U

D. Jodlbauer, U. Langer, T. Wick, Parallel matrix-free higher-order finite element solvers for phase- field fracture problems, Mathematical and Computational Applications 25 (3) (2020).doi:10.3390/ mca25030040

2020
[16]

Chafia, J

Z. Chafia, J. Yvonnet, J. Bleyer, S. Vincent, S. El Ouafa, Massively parallel phase field fracture simu- lations on supercomputers: towards multi-billion degree-of-freedom computations, Advanced Modeling and Simulation in Engineering Sciences 11 (2024).doi:10.1186/s40323-024-00280-4

2024 doi
[17]

Ziaei-Rad, Y

V. Ziaei-Rad, Y. Shen, Massive parallelization of the phase field formulation for crack propagation with time adaptivity, Computer Methods in Applied Mechanics and Engineering 312 (2016) 224–253, phase Field Approaches to Fracture.doi:10.1016/j.cma.2016.04.013

2016 doi
[18]

Gerasimov, L

T. Gerasimov, L. De Lorenzis, A line search assisted monolithic approach for phase-field computing of brittle fracture, Computer Methods in Applied Mechanics and Engineering 312 (2016) 276–303. doi:10.1016/j.cma.2015.12.017. 23

2016 doi
[19]

Lampron, D

O. Lampron, D. Therriault, M. Lévesque, An efficient and robust monolithic approach to phase-field quasi-static brittle fracture using a modified newton method, Computer Methods in Applied Mechanics and Engineering 386 (2021).doi:10.1016/j.cma.2021.114091

2021
[20]

Kopaničáková, R

A. Kopaničáková, R. Krause, A recursive multilevel trust region method with application to fully mono- lithic phase-field models of brittle fracture, Computer Methods in Applied Mechanics and Engineering 360 (2020).doi:10.1016/j.cma.2019.112720

2020
[21]

Heister, M

T. Heister, M. F. Wheeler, T. Wick, A primal-dual active set method and predictor-corrector mesh adaptivity for computing fracture propagation using a phase-field approach, Computer Methods in Applied Mechanics and Engineering 290 (2015) 466–495.doi:10.1016/j.cma.2015.03.009

2015 doi
[22]

T. Wick, Modified newton methods for solving fully monolithic phase-field quasi-static brittle fracture propagation, Computer Methods in Applied Mechanics and Engineering 325 (2017) 577–611.doi: 10.1016/j.cma.2017.07.026

2017 doi
[23]

Kirkesæther Brun, T

M. Kirkesæther Brun, T. Wick, I. Berre, J. M. Nordbotten, F. A. Radu, An iterative staggered scheme for phase field brittle fracture propagation with stabilizing parameters, Computer Methods in Applied Mechanics and Engineering 361 (2020).doi:10.1016/j.cma.2019.112752

2020
[24]

C. Luo, Fast staggered schemes for the phase-field model of brittle fracture based on the fixed-stress concept, Computer Methods in Applied Mechanics and Engineering 404 (2023).doi:10.1016/j.cma. 2022.115787

2023
[25]

Farrell, C

P. Farrell, C. Maurini, Linear and nonlinear solvers for variational phase-field models of brittle fracture, International Journal for Numerical Methods in Engineering 109 (5) (2017) 648–667.doi:10.1002/ nme.5300

2017
[26]

Storvik, J

E. Storvik, J. W. Both, J. M. Sargado, J. M. Nordbotten, F. A. Radu, An accelerated staggered scheme for variational phase-field models of brittle fracture, Computer Methods in Applied Mechanics and Engineering 381 (2021).doi:10.1016/j.cma.2021.113822

2021
[27]

Bourdin, Numerical implementation of the variational formulation for quasi-static brittle fracture, Interfaces and Free Boundaries (2007) 411–430doi:10.4171/IFB/171

B. Bourdin, Numerical implementation of the variational formulation for quasi-static brittle fracture, Interfaces and Free Boundaries (2007) 411–430doi:10.4171/IFB/171

2007 doi
[28]

Burke, C

S. Burke, C. Ortner, E. Süli, An adaptive finite element approximation of a variational model of brittle fracture, SIAM Journal on Numerical Analysis 48 (3) (2010) 980–1012.doi:10.1137/080741033

2010 doi
[29]

Vicentini, C

F. Vicentini, C. Zolesi, P. Carrara, C. Maurini, L. De Lorenzis, On the energy decomposition in vari- ational phase-field models for brittle fracture under multi-axial stress states, International Journal of Fracture (2024).doi:10.1007/s10704-024-00763-w

2024 doi
[30]

Amor, J.-J

H. Amor, J.-J. Marigo, C. Maurini, Regularized formulation of the variational brittle fracture with unilateral contact: Numerical experiments, Journal of the Mechanics and Physics of Solids 57 (8) (2009) 1209–1229.doi:10.1016/j.jmps.2009.04.011

2009 doi
[31]

Miehe, F

C. Miehe, F. Welschinger, M. Hofacker, Thermodynamically consistent phase-field models of fracture: Variational principles and multi-field fe implementations, International journal for numerical methods in engineering 83 (10) (2010) 1273–1311

2010
[32]

Freddi, G

F. Freddi, G. Royer-Carfagni, Regularized variational theories of fracture: A unified approach, Journal of the Mechanics and Physics of Solids 58 (8) (2010) 1154–1174.doi:10.1016/j.jmps.2010.02.010

2010 doi
[33]

De Lorenzis, C

L. De Lorenzis, C. Maurini, Nucleation under multi-axial loading in variational phase-field mod- els of brittle fracture, International Journal of Fracture 237 (1-2) (2022) 61–81.doi:10.1007/ s10704-021-00555-6. 24

2022
[34]

Nocedal, S

J. Nocedal, S. J. Wright, Numerical Optimization, 2nd Edition, Springer New York, 2006.doi:10. 1007/978-0-387-40065-5

2006
[35]

Balay, S

S. Balay, S. Abhyankar, M. F. Adams, S. Benson, J. Brown, P. Brune, K. Buschelman, E. M. Constanti- nescu, L. Dalcin, A. Dener, V. Eijkhout, J. Faibussowitsch, W. D. Gropp, V. Hapla, T. Isaac, P. Jolivet, D. Karpeev, D. Kaushik, M. G. Knepley, F. Kong, S. Kruger, D. A. May, L....

2024
[36]

Balay, S

S. Balay, S. Abhyankar, M. F. Adams, S. Benson, J. Brown, P. Brune, K. Buschelman, E. Constanti- nescu, L. Dalcin, A. Dener, V. Eijkhout, J. Faibussowitsch, W. D. Gropp, V. Hapla, T. Isaac, P. Jolivet, D. Karpeev, D. Kaushik, M. G. Knepley, F. Kong, S. Kruger, D. A. May, L. C....

2024
[37]

Balay, W

S. Balay, W. D. Gropp, L. C. McInnes, B. F. Smith, Efficient management of parallelism in object oriented numerical software libraries, in: E. Arge, A. M. Bruaset, H. P. Langtangen (Eds.), Modern Software Tools in Scientific Computing, Birkhauser Press, 1997, pp. 163–202

1997
[38]

A. A. L. Baldelli, C. Maurini, Numerical bifurcation and stability analysis of variational gradient- damage models for phase-field fracture, Journal of the Mechanics and Physics of Solids 152 (2021)

2021
[39]

Vicentini, C

F. Vicentini, C. Zolesi, P. Carrara, C. Maurini, L. De Lorenzis, On the energy decomposition in vari- ational phase-field models for brittle fracture under multi-axial stress states, International Journal of Fracture 247 (3) (2024) 291–317

2024
[40]

Lorentz, S

E. Lorentz, S. Cuvilliez, K. Kazymyrenko, Convergence of a gradient damage model toward a cohesive zone model, Comptes Rendus. Mécanique 339 (1) (2011) 20–26

2011
[41]

J.-Y. Wu, V. P. Nguyen, A length scale insensitive phase-field damage model for brittle fracture, Journal of the Mechanics and Physics of Solids 119 (2018) 20–42

2018
[42]

Zolesi, C

C. Zolesi, C. Maurini, Stability and crack nucleation in variational phase-field models of fracture: effects of length-scales and stress multi-axiality, Journal of the Mechanics and Physics of Solids 192 (2024). doi:10.1016/j.jmps.2024.105802

2024
[43]

Marengo, A

A. Marengo, A. Patton, M. Negri, U. Perego, A. Reali, A rigorous and efficient explicit algorithm for irreversibility enforcement in phase-field finite element modeling of brittle crack propagation, Computer Methods in Applied Mechanics and Engineering 387 (2021)

2021
[44]

J. E. J. Dennis, R. B. Schnabel, Numerical Methods for Unconstrained Optimization and Nonlinear Equations, SIAM, 1996.doi:10.1137/1.9781611971200

1996 doi
[45]

Miehe, M

C. Miehe, M. Hofacker, F. Welschinger, A phase field model for rate-independent crack propagation: Robust algorithmic implementation based on operator splits, Computer Methods in Applied Mechanics and Engineering 199 (2010) 2765–2778.doi:10.1016/j.cma.2010.04.011

2010 doi
[46]

Wheeler, T

M. Wheeler, T. Wick, W. Wollner, An augmented-lagrangian method for the phase-field approach for pressurized fractures, Computer Methods in Applied Mechanics and Engineering 271 (2014) 69–85. doi:10.1016/j.cma.2013.12.005

2014 doi
[47]

T. Wick, An error-oriented newton/inexact augmented lagrangian approach for fully monolithic phase- field fracture propagation, SIAM Journal on Scientific Computing 39 (4) (2017) B589–B617.doi: 10.1137/16M1063873. 25

2017 doi
[48]

Wambacq, J

J. Wambacq, J. Ulloa, G. Lombaert, S. François, Interior-point methods for the phase-field approach to brittle and ductile fracture, Computer Methods in Applied Mechanics and Engineering 375 (2021). doi:10.1016/j.cma.2020.113612

2021
[49]

S. J. Benson, T. S. Munson, Flexible complementarity solvers for large-scale applications, Optimization Methods and Software 21 (1) (2006) 155–168.doi:10.1080/10556780500065382

2006 doi
[50]

Wolfe, Convergence conditions for ascent methods, SIAM Review 11 (2) (1969) 226–235.doi: 10.1137/1011036

P. Wolfe, Convergence conditions for ascent methods, SIAM Review 11 (2) (1969) 226–235.doi: 10.1137/1011036

1969 doi
[51]

Armijo, Minimization of functions having lipschitz continuous first partial derivatives, Pacific Journal of Mathematics 16 (1) (1966) 1–3.doi:10.2140/pjm.1966.16.1

L. Armijo, Minimization of functions having lipschitz continuous first partial derivatives, Pacific Journal of Mathematics 16 (1) (1966) 1–3.doi:10.2140/pjm.1966.16.1

1966 doi
[52]

J. J. Moré, D. J. Thuente, Line search algorithms with guaranteed sufficient decrease, ACM Trans. Math. Softw. 20 (3) (1994) 286—-307.doi:10.1145/192115.192132

1994
[53]

J. R. R. A. Martins, A. Ning, Engineering Design Optimization, Cambridge University Press, Cam- bridge, UK, 2022.doi:10.1017/9781108980647. URLhttps://mdobook.github.io

2022 doi
[54]

U. M. Ascher, C. Greif, A First Course in Numerical Methods, Society for Industrial and Applied Mathematics, Philadelphia, PA, 2011.doi:10.1137/9780898719987

2011 doi
[55]

B. Shea, M. Schmidt, Greedy newton: Newton’s method with exact line search, Optimization Letters (2025) 1–19

2025
[56]

R. L. Burden, J. D. Faires, Numerical Analysis, 9th Edition, Cengage Learning, 2010

2010
[57]

I. F. D. Oliveira, R. H. C. Takahashi, An enhancement of the bisection method average performance preserving minmax optimality, ACM Trans. Math. Softw. 47 (1) (2020).doi:10.1145/3423597

2020 doi
[58]

E. M. R. Fairbairn, F.-J. Ulm, A tribute to fernando l. l. b. carneiro (1913–2001) engineer and scientist who invented the brazilian test, Materials and Structures 35 (3) (2002) 195–196.doi: 10.1007/BF02533589

1913 doi
[59]

X. Jin, C. Hou, X. Fan, C. Lu, H. Yang, X. Shu, Z. Wang, Quasi-static and dynamic experimental studies on the tensile strength and failure pattern of concrete and mortar discs, Scientific Reports 7 (15305) (2017).doi:10.1038/s41598-017-15700-2

2017 doi
[60]

Erarslan, D

N. Erarslan, D. J. Williams, Experimental, numerical and analytical studies on tensile strength of rocks, International Journal of Rock Mechanics and Mining Sciences 49 (2012) 21–30.doi:10.1016/ j.ijrmms.2011.11.007

2012
[61]

M. R. Khosravani, M. Silani, K. Weinberg, Fracture studies of ultra-high performance concrete using dynamic brazilian tests, Theoretical and Applied Fracture Mechanics 93 (2018) 302–310.doi:10.1016/ j.tafmec.2017.10.001

2018
[62]

ASTM International, Standard test method for splitting tensile strength of intact rock core specimens with flat loading platens, Tech. Rep. D3927-23, ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States (2023)

2023
[63]

International Society for Rock Mechanics, Suggested methods for determining tensile strength of rock materials, International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts 15 (3) (1978) 99–103.doi:10.1016/0148-9062(78)90003-7

1978 doi
[64]

Kumar, Y

A. Kumar, Y. Liu, J. E. Dolbow, O. Lopez-Pamies, The strength of the brazilian fracture test, Journal of the Mechanics and Physics of Solids 182 (2024).doi:10.1016/j.jmps.2023.105473. 26

2024
[65]

Bilgen, S

C. Bilgen, S. Homberger, K. Weinberg, Phase-field fracture simulations of the brazilian splitting test, International Journal of Fracture 220 (1) (2019) 85–98.doi:10.1007/s10704-019-00401-w

2019 doi
[66]

Vicentini, J

F. Vicentini, J. Heinzmann, P. Carrara, L. De Lorenzis, Variational phase-field modeling of cohesive fracture with flexibly tunable strength surface, Journal of the Mechanics and Physics of Solids 207 (2026).doi:10.1016/j.jmps.2025.106424

2026
[67]

A. R. Ingraffea, F. E. Heuze, Finite element models for rock fracture mechanics, International Jour- nal for Numerical and Analytical Methods in Geomechanics 4 (1) (1980) 25–43.doi:10.1002/nag. 1610040103

1980 doi
[68]

Nguyen, J

T. Nguyen, J. Yvonnet, Q.-Z. Zhu, M. Bornert, C. Chateau, A phase field method to simulate crack nucleation and propagation in strongly heterogeneous materials from direct imaging of their microstruc- ture, Engineering Fracture Mechanics 139 (2015) 18–39.doi:10.1016/j.engfracm...

2015 doi
[69]

I. A. Baratta, J. P. Dean, J. S. Dokken, M. Habera, J. S. Hale, C. N. Richardson, M. E. Rognes, M. W. Scroggs, N. Sime, G. N. Wells, Dolfinx: The next generation fenics problem solving environment (2023). doi:10.5281/zenodo.10447666

2023 doi
[70]

M. W. Scroggs, J. S. Dokken, C. N. Richardson, G. N. Wells, Construction of arbitrary order finite element degree-of-freedom maps on polygonal and polyhedral cell meshes, ACM Trans. Math. Softw. 48 (2) (2022).doi:10.1145/3524456

2022 doi
[71]

M. W. Scroggs, I. A. Baratta, C. N. Richardson, G. N. Wells, Basix: a runtime finite element basis evaluation library, Journal of Open Source Software 7 (73) (2022).doi:10.21105/joss.03982

2022 doi
[72]

M. S. Alnaes, A. Logg, K. B. Olgaard, M. E. Rognes, G. N. Wells, Unified form language: A domain- specific language for weak formulations of partial differential equations, ACM Trans. Math. Softw. 40 (2) (2014).doi:10.1145/2566630

2014 doi
[73]

L. D. Dalcin, R. R. Paz, P. A. Kler, A. Cosimo, Parallel distributed computing using python, Advances in Water Resources 34 (9) (2011) 1124–1139, new Computational Methods and Software Tools.doi: 10.1016/j.advwatres.2011.04.013

2011 doi
[74]

Amestoy, I

P. Amestoy, I. S. Duff, J. Koster, J.-Y. L’Excellent, A fully asynchronous multifrontal solver using distributed dynamic scheduling, SIAM Journal on Matrix Analysis and Applications 23 (1) (2001) 15–41

2001
[75]

Amestoy, A

P. Amestoy, A. Buttari, J.-Y. L’Excellent, T. Mary, Performance and scalability of the block low-rank multifrontal factorization on multicore architectures, ACM Transactions on Mathematical Software 45 (2019) 1–26

2019
[77]

P. R. Brune, M. G. Knepley, B. F. Smith, X. Tu, Composing scalable nonlinear algebraic solvers, SIAM Review 57 (4) (2015) 535–565.doi:10.1137/130936725. Appendix A. Strain energy splits In the following, we report the degradable and residual parts of the strain energy density ...

2015 doi
[78]

suggest that these variants are not necessarily advantageous for the phase-field problem, which is why they are not used in this work. F.2. Secant-based line searches Secondly, we compare the bisection line search to secant-based methods. These look for minima in the objective...

Pith tools

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

[1] [76]

Heinzmann, F

J. Heinzmann, F. Vicentini, P. Carrara, L. De Lorenzis, Improving convergence of phase-field fracture computations through line-search-enhanced alternate minimization, in: Advances and Challenges in Computational Mechanics, Springer, 2025.doi:10.1007/978-3-031-93213-7_12

work page doi:10.1007/978-3-031-93213-7_12 2025

[2] [1]

Bourdin, G

B. Bourdin, G. Francfort, J.-J. Marigo, Numerical experiments in revisited brittle fracture, Journal of the Mechanics and Physics of Solids 48 (4) (2000) 797–826.doi:10.1016/S0022-5096(99)00028-9

doi:10.1016/s0022-5096(99)00028-9 2000

[3] [2]

A. A. Griffith, The phenomena of rupture and flow in solids, Philosophical Transactions of the Royal Society of London 221 (1921) 163–198.arXiv:91192. URLhttps://www.jstor.org/stable/91192

1921

[4] [3]

Francfort, J.-J

G. Francfort, J.-J. Marigo, Revisiting brittle fracture as an energy minimization problem, Journal of the Mechanics and Physics of Solids 46 (8) (1998) 1319–1342.doi:10.1016/S0022-5096(98)00034-9

doi:10.1016/s0022-5096(98)00034-9 1998

[5] [4]

Ambrosio, V

L. Ambrosio, V. M. Tortorelli, On the approximation of free discontinuity problems, Bollettino della Unione Matematica Italiana 6 (1992) 105–123. 22

1992

[6] [5]

K. Pham, H. Amor, J.-J. Marigo, C. Maurini, Gradient Damage Models and Their Use to Approximate Brittle Fracture, Int. J. Damage Mech. 20 (4) (2011) 618–652.doi:10.1177/1056789510386852

doi:10.1177/1056789510386852 2011

[7] [6]

Gerasimov, L

T. Gerasimov, L. De Lorenzis, On penalization in variational phase-field models of brittle fracture, Computer Methods in Applied Mechanics and Engineering 354 (2019) 990–1026.doi:10.1016/j.cma. 2019.05.038

doi:10.1016/j.cma 2019

[8] [7]

Svolos, J

L. Svolos, J. N. Plohr, G. Manzini, H. M. Mourad, On the convexity of phase-field fracture formulations: Analytical study and comparison of various degradation functions, International Journal of Non-Linear Mechanics 150 (2023).doi:10.1016/j.ijnonlinmec.2023.104359

arXiv 2023