International Journal of Fracture最新文献

We investigate how the removal of a single bond affects the fracture behavior of triangular spring networks, whereby we systematically vary the position of the removed bond. Our simulations show that removing the bond has two contrasting effects on the fracture energy for initiation of crack propagation and on the fracture energy for failure of the entire network. A single missing bond can either lower or raise the initiation fracture energy, depending on its placement relative to the crack tip. In contrast, the failure fracture energy is always equal to or greater than that of a perfect network. For most initial placements of the missing bond, the crack path remains straight, and the increased failure fracture energy results from arrest at the point of maximum local fracture resistance. When the crack deviates from a straight path, we observe an even higher fracture energy, which we attribute primarily to crack bridging. This additional toughening mechanism becomes active only at low failure strains of the springs; at higher failure strains, the crack path tends to remain straight. Altogether, our results demonstrate that even a single bond removal can significantly enhance toughness, offering fundamental insights into the role of defects in polymer networks and informing the design of tough architected materials.

The reliability and structural integrity of the components get reduced when cracks grow during their service. This becomes even more critical when casting defects, such as macropores, are present in the component. This work investigates the crack trajectory in the presence of selected hole configurations by integrating phase field modelling with photoelasticity, serving as a precursor for modelling macroporosity. By lowering the computational burden typically associated with phase field modelling, photoelasticity has helped to arrive at a simple and consistent PFM approach capturing complex crack trajectories, validated experimentally for a wide configuration of holes with repeatability. Numerically simulated isochromatics are used to corroborate the findings from the phase field simulations and experiments.

The fracture mode of AA 5083 Al–Mg alloy is predominantly intergranular. For this, a three-dimensional Crystal Plasticity Finite Element (CPFE) model is developed to investigate Grain Boundary (GB) fracture behavior under uniaxial tensile loading. Following parameter calibration, the simulated tensile strength demonstrates consistency with experimental data, with a high accuracy of 97%. Furthermore, by integrating CPFE model incorporating the dislocation slips with a GB model, a two-dimensional tri-junction crystal plasticity framework is established, and cohesive elements are implemented within the three grains to simulate GB fracture process under uniaxial tension. This tri-junction model shows a good convergence. The effect of grain orientation, location and GB thickness on GB fracture is investigated. The results reveal that grain orientation and location significantly influences GB fracture behavior. It is found that the Goss-Brass-P orientation combination demonstrates the optimal comprehensive mechanical properties, achieving a strain energy density of 7.09 MJ/m³. Furthermore, there exists a threshold effect about GB thickness on comprehensive property. The optimal performance is obtained when the cohesive that determines GB thickness is set to 2.

We formulate a multiscale model of adhesive layers undergoing impurity-dependent cohesive fracture. The model contemplates three scales: i) at the atomic scale, fracture is controlled by interatomic separation and the thermodynamics of separation depends on temperature and impurity concentration; ii) the mesoscale is characterized by the collective response of a large number of interatomic planes across the adhesive layer, resulting in a thickness-dependence strength; in addition, impurities are uptaken from the environment and diffuse through the adhesive layer; and iii) at the macroscale, we focus on lap joints under the action of static loads and aggressive environments. Within this scenario, we obtain closed form analytical solutions for: the dependence of the adhesive layer strength on thickness; the length of the edge cracks, if any, as a function of time; the lifetime of the joint; and the dependence of the strength of the joint on time of preexposure to the environment. Overall, the theory is found to capture well the experimentally observed trends. Finally, we discuss how the model can be characterized on the basis of atomistic calculations, which opens the way for the systematic exploration of new material specifications.

Once shear fractures begin to propagate in relatively thin weak layers underneath dry snow slabs, avalanche release is imminent. The evolution of the dynamic weak layer fracture likely determines the size or destructive potential once release takes place. In this paper, 54 values of mode II fracture speed collected from field tests are analyzed. In addition, values of speed collected with connection with avalanche release which involves both mode II and mode III shear fracture are analyzed. The Rayleigh wave speed is the normal maximum speed limit for a mode II fracture based on continuum formulations and the shear wave speed is the companion speed limit for mode III fracture. The 54 values of speed data from field tests are shown to have a median value of 12% of the predicted Rayleigh wave speed with a range between 2 and 56%. The data collected during avalanche release do not exceed 59% of the Rayleigh wave speed. The test data were collected from weak layers composed of highly porous crystal forms which are known to exhibit dynamic porosity loss during fracture propagation. Such events imply viscous dissipation and high friction which help to explain the low speeds. For snow slab avalanche release, which occurs for slope angles greater than (25^{0}), field observations show that the mixed mode (II, III) weak layer fracture terminates with mode I fracture through the slab following dynamic propagation. Analysis using the avalanche speed data with respect to terminal mode I fracture suggests that the shear stress drop behind the fracture process zone can contribute to produce the mode I fracture to promote avalanche release but dynamic contraction of the fracture process zone is a minor effect at best.

In this article, an isotropic eikonal gradient-enhanced damage modeling technique for studying time-dependent fracture mechanisms in brittle/quasi-brittle solids is presented. As is consistent with the classical gradient damage theory, the eikonal-type gradient damage modeling framework is mesh-independent and simple to implement, as it does not require any numerical tracking algorithm for discontinuities in the displacement field during crack propagation. With the evolving interaction known as the damage-based transient internal length scale, eikonal damage formulations could overcome the drawbacks of spurious damage diffusion and incorrect damage initiation by the traditional gradient theory. More realistic failure behavior could be predicted by the eikonal-based damage models. In the work, the derived governing equations of the momentum and a smooth Rankine strain-based eikonal gradient-enhanced evolution for the motion of the body and evolution of the nonlocal equivalent strain (time-dependent fracture driving term) are effectively solved using the standard finite element method (FEM). For temporal discretization, a robust explicit solver of the central difference method integrated with the row-sum technique for mass lumping is adopted. The advantages and abilities of the present dynamic model are shown by considering a number of time-dependent crack propagation problems in two-dimensional (2D) and three- dimensional (3D) solids.

The main objectives of this paper are to simulate 3-D fatigue crack propagation using a Generalized Finite Element Method (GFEM) and to validate the results against experimental data. This GFEM adopts a high-order p-hierarchical basis and explicit representations of crack surfaces. Both h-refinement around the fracture fronts and non-uniform p-enrichment of the analysis domain are used to control discretization errors. A systematic validation of this GFEM applied to 3-D fatigue crack propagation has not been reported in the literature. The Displacement Correlation Method (DCM) is used to extract stress intensity factors. The effect of material parameters adopted in the DCM on the crack growth rate and fracture shape is investigated. Three increasingly complex fatigue crack propagation problems are solved. The first involves mixed-mode loading in a modified compact tension specimen. The second one involves the transition from 2-D to 1-D crack surfaces and interactions between the crack front and the corners of the domain boundary. The final problem simulates the growth of a circumferential surface crack in a steel pipe subjected to fatigue bending with overloading, where interactions between the crack and the pipe’s inner surface result in the splitting of the crack front. Another contribution is an algorithm designed to manage cyclic load histories featuring variable loading ranges and ratios between minimum and maximum load magnitudes.

Unfolding failure is a sudden delamination occurring in highly curved fiber-reinforced polymer composite laminates when they are subjected to opening bending moments. Depending on the stacking sequences involved, unfolding failure includes intralaminar damage, interlaminar damage and their complex interactions. The use of these laminates in primary structural components of commercial aircrafts has intensified interest in understanding this failure mechanism. To tackle this, a simple phase field formulation to model the onset and propagation of transverse damage in highly curved laminates is presented. Firstly, the Abaqus implementation of the phase field to model intralaminar damage in composite laminates is established, exploiting the analogy between phase field and the heat transfer model to use Abaqus temperature as the phase field damage parameter. In addition, since residual stresses developed during manufacturing processes affect damage propagation, they are incorporated into the formulation as constant residual strains. Secondly, the proposed implementation is validated by performing tests on benchmark problems. Finally, a preliminary analysis of the unfolding failure test is provided to highlight the importance of including residual stresses in the transverse damage evolution.