Nucleation and propagation of brittle fracture as a constrained energy minimization problem

arXiv:2608.28326 · cond-mat.mtrl-sci, math-ph, math.MP · Submitted 2026-08-26 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Nucleation and propagation of brittle fracture as a constrained energy minimization problem".

Mira: Here's the gist of the paper. Nucleation and propagation of brittle fracture as a constrained energy minimization problem presents a macroscopic continuum theory to describe when, where, and why cracks nucleate and propagate in nominally elastic brittle materials under arbitrary monotonic quasi-static mechanical loads<ref:2608.28326#pg2> The proposed sharp theory posits that cracks nucleate and propagate exclusively in regions where the strength surface of the material is exceeded, with their evolution dictated by the minimization of the sum of the potential — the elastic energy minus the work done by the externally applied forces — and surface energies<ref:2608.28326#pg4> Foundational Definitions and Principles The theory is built upon defining strength as a set of critical stresses at which fracture nucleates under monotonically increasing, spatially uniform, but otherwise arbitrary stress<ref:2608.28326#pg2> This definition establishes the strength surface as a surface in stress space defined by the condition F(S) = 0<ref:2608.28326#pg2> Fracture necessitates three macroscopic properties: elasticity, its strength surface, and its fracture toughness or critical energy release rate<ref:2608.28326#pg2> The governing criterion for fracture is two-fold: cracks can only nucleate or grow where the strength surface F(S) = 0 has been exceeded, and their evolution is governed by a minimization of a functional involving the competition between deformation and surface energies<ref:2608.28326#pg2> The Mathematical Formulation The theory is cast as a time-discretized problem where the pair deformation field-crack set (yk = yk(X), Γk) minimizes E(y, Γ):= P(y, Γ) + Gc H2(Γ)<ref:2608.28326#pg2> The minimization is subject to admissibility constraints that restrict competing add-cracks to lie entirely within the strength violation set VF (t), defined as the set of all material points at which the strength surface F(S) = 0 is exceeded at time t<ref:2608.28326#pg2> The formulation explicitly accounts for body forces and prescribed boundary conditions through terms in the potential energy P(y, Γ)<ref:2608.28326#pg2> Experimental Validation Across Nine Circles The theory is tested against the "Nine Circles of Elastic Brittle Fracture," spanning nine challenge problems on both silicate glass and synthetic rubber<ref:2608.28326#pg2> Two circles, the simple tension test and nucleation dominated by material strength under uniform stress, have been successfully tested<ref:2608.28326#pg2> The results in the simple tension test show that the proposed constrained energy minimization delivers both the when (u/L = 5.7 × 10−4) and the where (A = 0.65 mm and α = 90◦) of crack nucleation<ref:2608.28326#pg2> The theory also predicts finite-size growth in the direction orthogonal to the applied tension under force control, consistent with experimental observations<ref:2608.28326#pg2> Implications for Material Scales and Growth The analysis reveals critical material length scales, such as lc, below which the strength surface F(S) = 0 ceases to be a valid macroscopic material property<ref:2608.28326#pg2> The proposition derived from the theory establishes a minimal length-scale for L 2(omega)/Per(omega) that ensures some kind of nucleation event can take place<ref:2608.28326#pg2> Furthermore, the discrete evolution formulation predicts that crack propagation occurs by finite-size growth in the direction orthogonal to the applied tension rather than a time-continuous smooth growth<ref:2608.28326#pg2> Future Directions and Challenges The authors note that a time-continuous formulation remains an open challenge due to the nature of dissipation carried by the strength eluding current understanding<ref:2608.28326#pg2> Establishing whether phase-field theory provides a mathematically rigorous regularization of the sharp formulation is key for future work<ref:2608.28326#pg2> Developing an efficient and robust numerical implementation of the proposed theory is another crucial priority<ref:2608.28326#pg2> The ability to predict fracture under body forces, such as gravity, should be further validated in future work<ref:2608.28326#pg2> Compliance with Fracture Modes The theory's predictions for propagation modes are consistent with experimental results across various tests<ref:2608.28326#pg2> For instance, in the double cantilever beam test, the theory predicts stable propagation via small jumps rather than time-continuous smooth growth, aligning with reference results<ref:2608.28326#pg2> The energy minimization approach allows for growth by jumps, which is precisely what it predicts for this test<ref:2608.28326#pg2> This contrasts with the classical Griffith criterion which restricts add-cracks to be infinitesimally small<ref:2608.28326#pg2> The results suggest that the energy-minimizing direction coincidentally aligns with the plane orthogonal to the maximum principal stress, explaining observed fracture angles<ref:2608.28326#pg2> Concluding Remarks on Strength Measurement The strength surface is arguably the most challenging intrinsic material property to measure in its entirety through direct experiments<ref:2608.28326#pg2> However, because the strength surface strictly constrains where fracture can occur, non-uniform stress fields can be leveraged for inverse calibration of its parameters<ref:2608.28326#pg2> The theory's imposition of a length-scale acts as a selection criterion for domains for which fracture can take place<ref:2608.28326#pg2> The formulation builds upon the variational framework of Francfort and Marigo by introducing constraints that ensure competing add-cracks lie within the strength violation set<ref:2608.28326#pg2> The formulation reduces to the classical Griffith criterion G = Gc in the limiting case of stable infinitesimal growth<ref:2608.

Title and authors: Kai: So we're looking at this paper titled "Nucleation and propagation of brittle fracture as a constrained energy minimization problem". It sounds like they're trying to get a big picture view on when and where cracks form in materials under normal mechanical loads.

Mira: Exactly. What interests me right away is that they're framing this not just as some curve, but as an energy minimization problem involving material energy, applied forces, and surface energies.

Lev: It sounds like a continuum theory aiming to describe the whole process of crack nucleation and growth in brittle stuff under monotonic loads.

Kai: Right. The authors are trying to establish a sharp theory that says cracks only happen where the material's strength is exceeded, and their movement is driven by minimizing this total potential energy plus surface energies.

The paper's summary: Mira: So, they define strength very precisely first. They state that for elastic brittle materials, strength is the set of all critical stresses at which fracture starts under a specific kind of stress—monotonically increasing and spatially uniform but otherwise arbitrary <ref:2608.28326#pg2>.

Kai: And this creates a surface in stress space called the strength surface, F(S) equals zero, which is basically where the material's failure limit lies.

Lev: The paper then says that for brittle fracture to happen, you need three things: elasticity of the material itself, that strength surface we just talked about, and finally the fracture toughness or critical energy release rate.

Mira: And they set up a mathematical problem where crack evolution is dictated by minimizing this sum of potential energy minus the work done by external forces and those surface energies <ref:2608.28326#pg4>.

Kai: It's basically setting up a constrained optimization problem where the cracks must stay within that region where the strength is already violated.

The paper's improvements: Lev: I see they’re using a time-discretized formulation for this minimization, which means they break the problem down over time, and they have to make sure any new cracks don't jump outside the allowed region defined by that strength violation set VF(t) <ref:2608.28326#pg2>.

Kai: And they’re explicitly including body forces and boundary conditions in their potential energy term P(y, Γ), so it’s not just about the material itself but the whole setup.

Mira: The paper suggests a key improvement is introducing a minimal length scale, some critical length lc, below which the strength surface F(S) equals zero stops being a valid macroscopic property for nucleation events <ref:2608.28326#pg2>.

Lev: So that means they are proposing this minimum length scale as a selection criterion for domains where fracture can actually occur, which is a big constraint on the theory.

Kai: And they also suggest that crack propagation isn't necessarily smooth over time; instead, it predicts finite-size growth in the direction orthogonal to the applied tension <ref:2608.28326#pg2>.

Conclusion: Mira: So, to wrap up, this paper on "Nucleation and propagation of brittle fracture as a constrained energy minimization problem" moves us toward a formulation where cracks are governed by minimizing energy while respecting the material's strength constraints.

Kai: It gives us tools to predict when and where these things happen under arbitrary loads, like the simple tension test, showing nucleation at zero point sixty-five millimeters with an angle of ninety degrees <ref:2608.28326#pg2>.

Lev: And on the practical side, it shows how we can model this evolution using a discrete set formulation instead of just smooth time-continuous growth, which is important for simulations running on real hardware.

Kai: The implication for us is that we need to figure out how to handle those critical length scales and dissipation that are currently tricky in time-continuous formulations.

Mira: It sets up the next step, which is seeing if phase field theory can rigorously regularize this sharp formulation so we have a solid mathematical footing.

Lev: For real hardware, the priority will be developing a numerical implementation that’s efficient and robust enough to handle these discrete evolution schemes without too much overhead.

Oscar Lopez-Pamiesa, Farhad Kamareia, Gilles A. Francfortb, Alessandro Giacominic

Department of Civil and Environmental Engineering, University of Illinois Urbana-Champaign · Flatiron Institute · DICATAM, Sezione di Matematica, Universita degli Studi di Brescia

cond-mat.mtrl-sci, math-ph, math.MP

Submitted: 2026-08-26

Updated: 2026-11-15

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 91/100

The gist: Nucleation and propagation of brittle fracture as a constrained energy minimization problem presents a macroscopic continuum theory to describe when, where…

Key concepts

Strength Surface
This is a mathematical surface in stress space that defines the exact stresses at which a material will begin to fracture under uniform loading. Fracture only occurs when the applied stress exceeds this defined surface.
Energy Minimization Functional
The crack's evolution is governed by minimizing a total energy value. This value balances the stored elastic energy from deformation, the work done by external forces, and the energy required to create new surfaces (surface energies). Cracks evolve toward the lowest possible total energy state.
Strength Violation Set (VF(t))
This set identifies all material points where fracture has already occurred or is currently occurring at a specific time. The theory requires that any potential crack growth must remain confined within this set to ensure physical admissibility.

Terminology

Summary

Nucleation and propagation of brittle fracture as a constrained energy minimization problem presents a macroscopic continuum theory to describe when, where, and why cracks nucleate and propagate in nominally elastic brittle materials under arbitrary monotonic quasi-static mechanical loads<ref:2608. p. 0

Where the paper says it

Nucleation and propagation of brittle fracture as a constrained energy minimization problem Oscar Lopez-Pamiesa , Farhad Kamareia , Gilles A. Francfortb , Alessandro Giacominic aDepartment of Civil and Environmental Engineering, University of Illinois, Urbana–Champaign, IL 61801, USA bFlatiron Institute, 162 Fifth Avenue, New York, NY 10010, USA cDICATAM, Sezione di Matematica, Universit`a degli Studi di Brescia, Brescia, Italy Abstract This paper presents a macroscopic, or continuum, theory aimed at describing when, where, and why cracks nucleate and propagate in nominally elastic brittle materials under monotonic, quasi-static, but otherwise arbitrary mechanical loads.

28326#pg2> The proposed sharp theory posits that cracks nucleate and propagate exclusively in regions where the strength surface of the material is exceeded, with their evolution dictated by the minimization of the sum of the potential — the elastic energy minus the work done by the externally applied forces — and surface energies<ref:2608. p. 0
Where the paper says it

Francfortb , Alessandro Giacominic aDepartment of Civil and Environmental Engineering, University of Illinois, Urbana–Champaign, IL 61801, USA bFlatiron Institute, 162 Fifth Avenue, New York, NY 10010, USA cDICATAM, Sezione di Matematica, Universit`a degli Studi di Brescia, Brescia, Italy Abstract This paper presents a macroscopic, or continuum, theory aimed at describing when, where, and why cracks nucleate and propagate in nominally elastic brittle materials under monotonic, quasi-static, but otherwise arbitrary mechanical loads. Motivated by recent insights, the proposed sharp theory posits that: cracks nucleate and propagate exclusively in regions where the strength surface of the material is exceeded, with their evolution dictated by the minimization of the sum of the potential — the elastic energy minus the work done by the externally applied forces — and surface energies.

28326#pg4>

Foundational Definitions and Principles

The theory is built upon defining strength as a set of critical stresses at which fracture nucleates under monotonically increasing, spatially uniform, but otherwise arbitrary stress<ref:2608.28326#pg2> This definition establishes the strength surface as a surface in stress space defined by the condition F(S) = 0<ref:2608. p. 1

Where the paper says it

1 below, the strength of an elastic brittle material is defined as the set of all critical stresses1 S at which fracture nucleates in a specimen2 under a state of monotonically increasing, spatially uniform, but otherwise arbitrary stress. This collection of critical stress states defines a surface in stress space, F(S) = 0, referred to as the strength surface of the material.

28326#pg2> Fracture necessitates three macroscopic properties: elasticity, its strength surface, and its fracture toughness or critical energy release rate<ref:2608. p. 1
Where the paper says it

This collection of critical stress states defines a surface in stress space, F(S) = 0, referred to as the strength surface of the material. Then, that elastic brittle fracture necessitates three distinct macroscopic properties: (i) the elasticity of the material; (ii) its strength surface; and (iii) its fracture toughness or critical energy release rate.

28326#pg2> The governing criterion for fracture is two-fold: cracks can only nucleate or grow where the strength surface F(S) = 0 has been exceeded, and their evolution is governed by a minimization of a functional involving the competition between deformation and surface energies<ref:2608. p. 1
Where the paper says it

Cracks can nucleate or grow only in regions where the strength surface of the material, F(S) = 0, has been exceeded. Within the regions of strength violation, crack evolution is governed by a minimization of a functional, involving the competition between deformation and surface energies.

28326#pg2>

The Mathematical Formulation

The theory is cast as a time-discretized problem where the pair deformation field-crack set (yk = yk(X), Γk) minimizes E(y, Γ):= P(y, Γ) + Gc H2(Γ)<ref:2608.28326#pg2> The minimization is subject to admissibility constraints that restrict competing add-cracks to lie entirely within the strength violation set VF (t), defined as the set of all material points at which the strength surface F(S) = 0 is exceeded at time t<ref:2608. p. 1

Where the paper says it

Cracks can nucleate or grow only in regions where the strength surface of the material, F(S) = 0, has been exceeded. Within the regions of strength violation, crack evolution is governed by a minimization of a functional, involving the competition between deformation and surface energies.

28326#pg2> The formulation explicitly accounts for body forces and prescribed boundary conditions through terms in the potential energy P(y, Γ)<ref:2608. p. 5
Where the paper says it

A further limitation of the variational approach is its apparent inability to handle non-zero Neumann boundary conditions and/or body forces. An attempt at removing this limitation was recently proposed by Larsen (2021) but it yields a formulation that is inconsistent with experimental observations.

28326#pg2>

Experimental Validation Across Nine Circles

The theory is tested against the Nine Circles of Elastic Brittle Fracture, spanning nine challenge problems on both silicate glass and synthetic rubber<ref:2608.28326#pg2> Two circles, the simple tension test and nucleation dominated by material strength under uniform stress, have been successfully tested<ref:2608. p. 1

Where the paper says it

As a preliminary, Section 2 analyzes three motivating experiments: a silicate glass specimen in simple tension under both displacement and force control, and the self-weight cantilever bending of a mortar beam. In Section 5, we confront the theory with a set of nine tests spanning the complete spectrum of well-settled experimental knowledge on fracture nucleation and propagation — the so-called “Nine Circles of Elastic Brittle Fracture” introduced in (Kamarei et al.

28326#pg2> The results in the simple tension test show that the proposed constrained energy minimization delivers both the when (u/L = 5. p. 7
Where the paper says it

7×10−4 . Crucially, observe that the proposed constrained energy minimization delivers both the when (u/L = 5.

7 × 10−4) and the where (A = 0.65 mm and α = 90◦) of crack nucleation<ref:2608. p. 0
Where the paper says it

giacomini@unibs. it (Alessandro Giacomini) arXiv:2608.

28326#pg2> The theory also predicts finite-size growth in the direction orthogonal to the applied tension under force control, consistent with experimental observations<ref:2608. p. 9
Where the paper says it

7×10−4 does not stay put, instead it propagates by a finite-size growth in the direction orthogonal to the applied tension. Repeating the procedure once again leads to a crack, orthogonal to the direction of applied tension, that completely severs the specimen at u/L = 5.

28326#pg2>

Implications for Material Scales and Growth

The analysis reveals critical material length scales, such as lc, below which the strength surface F(S) = 0 ceases to be a valid macroscopic material property<ref:2608.28326#pg2> The proposition derived from the theory establishes a minimal length-scale for L 2(omega)/Per(omega) that ensures some kind of nucleation event can take place<ref:2608.28326#pg2> Furthermore, the discrete evolution formulation predicts that crack propagation occurs by finite-size growth in the direction orthogonal to the applied tension rather than a time-continuous smooth growth<ref:2608. p. 9

Where the paper says it

7×10−4 does not stay put, instead it propagates by a finite-size growth in the direction orthogonal to the applied tension. Repeating the procedure once again leads to a crack, orthogonal to the direction of applied tension, that completely severs the specimen at u/L = 5.

28326#pg2>

Future Directions and Challenges

The authors note that a time-continuous formulation remains an open challenge due to the nature of dissipation carried by the strength eluding current understanding<ref:2608.28326#pg2> Establishing whether phase-field theory provides a mathematically rigorous regularization of the sharp formulation is key for future work<ref:2608.28326#pg2> Developing an efficient and robust numerical implementation of the proposed theory is another crucial priority<ref:2608.28326#pg2> The ability to predict fracture under body forces, such as gravity, should be further validated in future work<ref:2608. p. 0

Where the paper says it

Recall that an elastic brittle material — the most basic idealization of a solid capable of fracture — is one that, in response to mechanical forces, either deforms elastically or creates new surfaces, the latter being the sole mechanism by which it dissipates energy. Specifically, this first paper addresses these questions for the case of monotonic quasi-static loading, deferring non-monotonic and dynamic loading to future work.

28326#pg2>

Compliance with Fracture Modes

The theory's predictions for propagation modes are consistent with experimental results across various tests<ref:2608.28326#pg2> For instance, in the double cantilever beam test, the theory predicts stable propagation via small jumps rather than time-continuous smooth growth, aligning with reference results<ref:2608.28326#pg2> The energy minimization approach allows for growth by jumps, which is precisely what it predicts for this test<ref:2608. p. 4

Where the paper says it

In a nutshell, the fracture state in a body is viewed as a global minimization problem for the sum of the elastic and surface energies with arbitrary add-cracks as test fields. This variational approach is in line with Griffith’s original idea of energy competition and is amenable to generalizations to nonlinear elastic and anisotropic materials in a 3D setting (Bourdin et al.

28326#pg2> This contrasts with the classical Griffith criterion which restricts add-cracks to be infinitesimally small<ref:2608. p. 3
Where the paper says it

In his own words (Griffith, 1924, page 57): “The condition that the crack may extend, that is, that rupture may occur, is then obtained by expressing the condition that the total energy [the sum of the potential energy of the applied tractions, the strain energy and the surface energy of the crack] is unchanged by small variations in the length of the crack. ” This reveals three significant aspects of his approach: first, the criterion applies strictly to the infinitesimal growth of pre-existing cracks rather than finite-size growth or fracture nucleation; second, it offers no predictive information regarding the resulting crack path; and third, its application amounts to computing variations of the elastic energy minus the work done by the applied forces with respect to infinitesimal add-cracks along a pre-supposed crack path, a quantity that is now classically denoted as G and referred to as the energy release rate.

28326#pg2> The results suggest that the energy-minimizing direction coincidentally aligns with the plane orthogonal to the maximum principal stress, explaining observed fracture angles<ref:2608. p. 8
Where the paper says it

75 mm drop by nucleation drop by propagation Figure 4: Theoretical prediction of the tension test on silicate glass under displacement control. The results show the global stress S = P/(BH) as a function of the applied global strain u/L, as well as a depiction of the fractured state of the specimen following the nucleation and propagation of a crack orthogonal to the direction of applied tension.

28326#pg2>

Concluding Remarks on Strength Measurement

The strength surface is arguably the most challenging intrinsic material property to measure in its entirety through direct experiments<ref:2608.28326#pg2> However, because the strength surface strictly constrains where fracture can occur, non-uniform stress fields can be leveraged for inverse calibration of its parameters<ref:2608. p. 3

Where the paper says it

In his own words (Griffith, 1924, page 57): “The condition that the crack may extend, that is, that rupture may occur, is then obtained by expressing the condition that the total energy [the sum of the potential energy of the applied tractions, the strain energy and the surface energy of the crack] is unchanged by small variations in the length of the crack. ” This reveals three significant aspects of his approach: first, the criterion applies strictly to the infinitesimal growth of pre-existing cracks rather than finite-size growth or fracture nucleation; second, it offers no predictive information regarding the resulting crack path; and third, its application amounts to computing variations of the elastic energy minus the work done by the applied forces with respect to infinitesimal add-cracks along a pre-supposed crack path, a quantity that is now classically denoted as G and referred to as the energy release rate.

28326#pg2> The theory's imposition of a length-scale acts as a selection criterion for domains for which fracture can take place<ref:2608.28326#pg2> The formulation builds upon the variational framework of Francfort and Marigo by introducing constraints that ensure competing add-cracks lie within the strength violation set<ref:2608. p. 4
Where the paper says it

At the end of the 1990s, Francfort and Marigo (1998) introduced a new idea to describe crack path within the setting of linear elastic brittle materials subject to mixed Dirichlet and zero Neumann boundary conditions and in the absence of body forces. In a nutshell, the fracture state in a body is viewed as a global minimization problem for the sum of the elastic and surface energies with arbitrary add-cracks as test fields.

28326#pg2> The formulation reduces to the classical Griffith criterion G = Gc in the limiting case of stable infinitesimal growth<ref:2608. p. 4
Where the paper says it

In addition to describing the crack path, the formulation transcends the limitations of Griffith’s approach in two additional ways. First, the growth of pre-existing cracks need not be infinitesimal, instead, finite-size crack growth is possible.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI system can predict when and where cracks nucleate and propagate in nominally elastic brittle materials under arbitrary monotonic quasi-static loading conditions by minimizing a functional involving potential energy, surface energies, and strength constraints.

  2. The system can perform sharp fracture predictions for isotropic linear elastic materials using a time-discretized formulation, explicitly considering the competition between deformation and surface energies as dictated by the two-fold criterion.

  3. The AI can analyze nine specific test cases spanning experimental knowledge (Nine Circles of Elastic Brittle Fracture) to validate fracture models against both hard silicate glass and soft synthetic rubber materials.

  4. The system can determine critical material length scales, such as the critical material length scale lc, which defines a boundary below which the strength surface is no longer a macroscopic property for nucleation events.

  5. The system can provide a mathematical framework for time-discretized evolution of cracks by defining an admissible configuration set and a minimization scheme over potential energy and surface energies.

  6. The AI can predict crack growth behavior, including finite-size growth via jumps rather than smooth time-continuous processes, as demonstrated in the double cantilever beam test scenario.

Abstract

This paper presents a macroscopic, or continuum, theory aimed at describing when, where, and why cracks nucleate and propagate in nominally elastic brittle materials under monotonic, quasi-static, but otherwise arbitrary mechanical loads. Motivated by recent insights, the proposed sharp theory posits that: cracks nucleate and propagate exclusively in regions where the strength surface of the material is exceeded, with their evolution dictated by the minimization of the sum of the potential --- the elastic energy minus the work done by the externally applied forces --- and surface energies. While the theory applies to materials with any elasticity (linear or nonlinear) and any material symmetry (isotropic or anisotropic), attention is restricted here to the most basic case of isotropic elastic brittle materials. For demonstration purposes, the theory is confronted with a set of nine tests that span the entire range of well-settled experimental knowledge on fracture nucleation and propagation --- the so-called ``Nine Circles of Elastic Brittle Fracture'' --- on both a hard material (a silicate glass) and a soft material (a synthetic rubber).

Related papers