Seminars

Variational phase-field models for brittle fracture are powerful computational tools for studying Griffith-type crack propagation under complex three-dimensional and multiaxial loading scenarios. However, they struggle to accurately capture fracture nucleation, i.e., the onset of cracks in quasi-brittle materials. While effective for tensile-driven fractures (mode I), they fail under multiaxial loading due to the lack of flexibility in prescribing a material-specific strength surface.
Traditional energy decomposition approaches often lead to questionable residual stresses, prompting non-variational modifications that sacrifice the physical, mathematical, and numerical advantages of an energy minimization framework.
This limitation stems from the fact that classical phase-field models merely regularize the sharp Griffith fracture model, which lacks a nucleation concept, unlike sharp cohesive
fracture models.
To overcome this, we propose a variational phase-field model that approximates cohesive fracture allowing the inclusion of an arbitrary strength surface as a material property. Additionally, similar to what was observed in gradient damage models coupled with plasticity, we demonstrate that this formulation enables sharp cohesive fractures, providing an approximation that avoids smearing of the displacement field. The model naturally incorporates a sharp non-interpenetration condition, thus eliminating the need for additional energy decompositions.

Bio: Francesco Vicentini recently completed his PhD in September 2025 at ETH Zurich under the supervision of Prof. Laura De Lorenzis, with a thesis on phase-field modeling of brittle and cohesive fracture. His doctoral research focused on fracture nucleation and fracture in heterogeneous materials from a theoretical and numerical perspective. He is currently a postdoctoral researcher in the laboratory of Prof. Ravi-Chandar at The University of Texas at Austin, with a fellowship granted by the Swiss National Science Foundation.