Predict Material Failure Research Articles

Towards Prediction of Failure in Ti‐6AI‐4V Aerospace Materials by Surface Observations

Abstract: This paper deals with the development of a methodology for the prediction of material failure in metallic aerospace alloys by evaluating changes in surface characteristics directly prior to unstable fatigue crack propagation. The study is based on in situ nondestructive characterisation of the depression zone ahead of the crack tip of fatigue‐pre‐cracked titanium alloy specimens subjected to static loading. A relationship between the surface characteristics of the deformation zone ahead of the crack and the stress intensity factor of the material was obtained. This relationship was common to a variety of microstructural conditions such as mill‐annealed and β‐annealed microstructures. Based on the analysis, prediction of the impending fracture in cracked samples of the material was enabled. The outcome of this study can be used for optimising the service life of structural components.

Micromechanical analysis on the influence of the Lode parameter on void growth and coalescence

A micromechanical model consisting of a band with a square array of equally sized cells, with a spherical void located in each cell, is developed. The band is allowed a certain inclination and the periodic arrangement of the cells allow the study of a single unit cell for which fully periodic boundary conditions are applied. The model is based on the theoretical framework of plastic localization and is in essence the micromechanical model by Barsoum and Faleskog (Barsoum, I., Faleskog, J., 2007. Rupture mechanisms in combined tension and shear—micromechanics. International Journal of Solids and Structures 44(17), 5481–5498) with the extension accounting for the band orientation. The effect of band inclination is significant on the strain to localization and cannot be disregarded. The macroscopic stress state is characterized by the stress triaxiality and the Lode parameter. The model is used to investigate the influence of the stress state on void growth and coalescence. It is found that the Lode parameter exerts a strong influence on the void shape evolution and void growth rate as well as the localized deformation behavior. At high stress triaxiality level the influence of the Lode parameter is not as marked and the overall ductility is set by the stress triaxiality. For a dominating shear stress state localization into a band cannot be regarded as a void coalescence criterion predicting material failure. A coalescence criterion operative at dominating shear stress state is needed.

An experimentally validated micromechanical model of a rat vertebra under compressive loading

In recent years, finite element analysis (FEA) has been increasingly applied to examine and predict the mechanical behaviour of craniofacial and other bony structures. Traditional methods used to determine material properties and validate finite element models (FEMs) have met with variable success, and can be time-consuming. An implicit assumption underlying many FE studies is that relatively high localized stress/strain magnitudes identified in FEMs are likely to predict material failure. Here we present a new approach that may offer some advantages over previous approaches. Recently developed technology now allows us to both image and conduct mechanical tests on samples in situ using a materials testing stage (MTS) fitted inside the microCT scanner. Thus, micro-finite element models can be created and validated using both quantitative and qualitative means. In this study, a rat vertebra was tested under compressive loading until failure using an MTS. MicroCT imaging of the vertebra before mechanical testing was used to create a high resolution finite element model of the vertebra. Load-displacement data recorded during the test were used to calculate the effective Young's modulus of the bone (found to be 128 MPa). The microCT image of the compressed vertebra was used to assess the predictive qualities of the FE model. The model showed the highest stress concentrations in the areas that failed during the test. Clearly, our analyses do not directly address biomechanics of the craniofacial region; however, the methodology adopted here could easily be applied to examine the properties and behaviour of specific craniofacial structures, or whole craniofacial regions of small vertebrates. Experimentally validated micro-FE analyses are a powerful method in the study of materials with complex microstructures such as bone.

Application of a stress triaxiality dependent fracture criterion in the finite element analysis of unnotched Charpy specimens

Nonlinear dynamic finite element analysis (FEA) is conducted to simulate the fracture of unnotched Charpy specimens of steel under pendulum impact loading by a dedicated, oversized and nonstandard Bulk Fracture Charpy Machine (BFCM). The impact energy needed to fracture an unnotched Charpy specimen in a BFCM test can be two orders of magnitude higher than the typical impact energy of a Charpy V-notch specimen. To predict material failure, a phenomenological, stress triaxiality dependent fracture initiation criterion and a fracture evolution law in the form of strain softening are incorporated in the constitutive relations. The BFCM impact energy results obtained from the FEA simulations compare favorably with the corresponding experimental data. In particular, the FEA predicts accurately the correlations of the BFCM impact energy with such factors as specimen geometry, impactor tup width and material type. The analyses show that a specimen’s progressive deterioration through the thickness dimension displays a range of shear to ductile fracture modes, demonstrating the necessity of applying a stress state dependent fracture initiation criterion. Modeling the strain softening behavior helps to capture the residual load carrying capability of a ductile metal or alloy beyond the onset of damage. The total impact energy can be significantly under predicted if a softening branch is not included in the stress–strain curve. This research supports a study of the puncture failure of railroad tank cars under dynamic impact loading. Applications of the presented fracture model in failure analyses of other structures are further discussed.

Modeling material responses by arbitrary Lagrangian Eulerian formulation and adaptive mesh refinement method

In this paper we report an efficient numerical method combining a staggered arbitrary Lagrangian Eulerian (ALE) formulation with the adaptive mesh refinement (AMR) method for materials modeling including elastic–plastic flows, material failure, and fragmentation predictions. Unlike traditional AMR applied on fixed domains, our investigation focuses on the application to moving and deforming meshes resulting from Lagrangian motion. We give details of this numerical method with a capability to simulate elastic–plastic flows and predict material failure and fragmentation, and our main focus of this paper is to create an efficient method which combines ALE and AMR methods to simulate the dynamics of material responses with deformation and failure mechanisms. The interlevel operators and boundary conditions for these problems in AMR meshes have been investigated, and error indicators to locate material deformation and failure regions are studied. The method has been applied on several test problems, and the solutions of the problems obtained with the ALE–AMR method are reported. Parallel performance and software design for the ALE–AMR method are also discussed.

An Investigation on Low Cycle Lifetime of Al2024 Alloy

One of the important considerations in the design of components is the estimation of cyclic lifetime and analysis of the critical regions of a construction. The local approach of lifetime estimation using continuum damage mechanics (CDM) has shown a great potential in predicting material failure not only for monotonic, but also for fully reversed loadings. In this paper, the CDM model of Desmorat-Lemaitre [1] was investigated regarding the prediction of cyclic lifetime. A series of experiments on tension specimens with different geometries were performed. The latter were used for the determination of model parameters as well as for the validation of the predictive capability of the model.

Generalized Variational Principles for Uncertainty Quantification of Boundary Value Problems of Random Heterogeneous Materials

Asymptotic theories of classical micromechanics are built on a fundamental assumption of large separation of scales. For random heterogeneous materials the scale-decoupling assumption however is inapplicable in many circumstances from conventional failure problems to novel small-scale engineering systems. Development of new theories for scale-coupling mechanics and uncertainty quantification is considered to have significant impacts on diverse disciplines. Scale-coupling effects become crucial when size of boundary value problems (BVPs) dynamic wavelength is comparable to the characteristic length of heterogeneity or when local heterogeneity becomes crucial due to extreme sensitivity of local instabilities. Stochasticity, vanishing in deterministic homogenization, resurfaces amid multiscale interactions. Multiscale stochastic modeling is expected to play an increasingly important role in simulation and prediction of material failure and novel multiscale systems such as microelectronic-mechanical systems and metamaterials. In computational mechanics a prevalent issue is, while a fine mesh is desired to achieve high accuracy, a certain mesh size threshold exists below which deterministic finite elements become questionable. This work starts investigation of the scale-coupling problems by first looking at uncertainty of material responses due to randomness or incomplete information of microstructures. The classical variational principles are generalized from scale-decoupling problems to scale-coupling BVPs, which provides upper and lower variational bounds for probabilistic prediction of material responses. It is expected that the developed generalized variational principles will lead to novel computational methods for uncertainty quantification of random heterogeneous materials.

New Evaluation Method of Material Degradation Considering Synergistic Effects of Radiation Damage

In core structural materials of next generation reactors such as a liquid-metal cooled fast breeding reactor and a supercritical-water cooled thermal or first reactor, materials' degradation behavior by neutron irradiation damage and thermal (cyclic) stress should be considered with fair accuracy in design process (including maintenance and repair plans), because the materials are used under higher temperature gradients and higher neutron flux fields than those in the present light water reactors. In the current experiential design rules, service lives of core structural components were determined by the materials degradation such as the increase of ductile-to-brittle transition temperature after post irradiation examination data. However, other materials degradations such as irradiation-assisted stress corrosion cracking (IASCC), which occurs by the degradation synergistically interacting with radiation hardening, local chemical composition change, swelling and radiation creep, should be considered reasonably in the design process of the next generation reactors, because of the anticipation of the beneficial effects by synergy of radiation damage. The radiation hardening and local chemical composition change at grain boundaries due to radiation-induced segregation increased with increasing dose. Above some threshold dose, swelling increased rapidly with increasing dose. Residual stress due to thermal stress and welding procedure decreased with increasing dose. To predict material failure by IASCC with reasonable accuracy, in this study, each material degradation phenomenon with different dose dependence was modeled with consideration of radiation induced stress relaxation. And then the models were integrated to simulate the failure behavior for the duration of reactor operation period. In this paper, the models obtained by ion-irradiation experiments and compared by data from neutron irradiation experiments were presented, and the concept of our new evaluation method and the programming code for the failure simulation were outlined.

Anisotropy of the fatigue behaviour of cancellous bone

The fatigue behaviour of materials is of particular interest for the failure prediction of materials and structures exposed to cyclic loading. For trabecular bone structures only a few sets of lifetime data have been reported in the literature and structural measures are commonly not considered. The influence of load contributions which are not aligned with the main physiological axis remains unclear. Furthermore site and species dependent relationships are not well described. In this study five different groups of trabecular bone, defined in terms of orientation, species and site were exposed to cyclic compression. In total, 108 fatigue tests were analysed. The lifetimes were found to decrease drastically when off-axis loads were applied. Additionally, species and site strongly affect fatigue lifetimes. Strains at failure were also found to be a function of orientation.

A generalized stress singularity approach for material failure prediction and its application to adhesive joint strength analysis

The concept of stress is very useful to describe the effect of external loads on structures. However, as a basis for the prediction of failure the concept of stress becomes meaningless when the structure encompasses singularities as a result of discrete stiffness steps or geometric anomalies such as cracks. In this article it is argued that the concept of failure stress is incorrect and should be replaced by a generalized concept based on stress intensity factors and singularity orders. It appears that material failure stress is the critical stress intensity factor for a zero-order singularity stress field. By plotting the critical stress intensity factor as a function of singularity order, the strength of a material can be characterized in a general fashion that integrates tensile strength, fracture toughness and critical singularities in adhesive joints. It is also shown that plasticity does not eliminate the stress singularity in an adhesive joint but changes the order of the singularity due to the induced change in interface corner angle between the dissimilar materials in the joint.

Failure prediction for advanced crashworthiness of transportation vehicles

During the past two decades explicit finite element crashworthiness codes have become an indispensable tool for the design of crash and passenger safety systems. These codes have proven remarkably reliable for the prediction of ductile metal structures that deform plastically; however, they are not reliable for joining systems and materials such as high strength steels, plastics and low ductility lightweight materials all of which are liable to fracture during the crash event. In order to improve crash failure prediction of materials and joining systems the CEC has recently funded a 3 year European research project dedicated to this topic. Specifically the project concerned aluminium, magnesium, high strength steels, plastics and two primary joining techniques; namely spotwelds and weldlines. Numerous new developments were undertaken including improved failure laws, adaptive meshing and element splitting to treat crack propagation. In the case of sheet stamping, investigations have also tried to account for process history effects and the metallurgical changes that occur during manufacture. This project has recently finished and this paper presents some of the key research results of the work concerning materials failure modelling.

On damage theory of a cohesive medium

A micro-mechanics damage model is proposed based on homogenization of penny-shaped cohesive micro-cracks (Barenblatt–Dugdale type) in a three dimensional representative volume element. By assuming that macro-hydrostatic stress state has dominant effect on permanent crack opening, a class of pressure sensitive yielding potentials and corresponding damage evolution laws have been derived. The merits of this class of damage models are: (1) Its ability to model and predict material failure and degradation due to cohesive micro-crack growth; (2) its ability to estimate the influence of Poisson's ratio on material's damage. One of the distinguished features of the new damage model is at macro-level the reversible part of effective constitutive relation is characterized as a nonlinear elasticity, whereas the irreversible part of effective constitutive relation is a form of pressure-sensitive plasticity, both of which are significantly different from material's behaviors at micro-level before homogenization.

Comparison of computational predictions of material failure using nonlocal damage models

In computational analysis of damage failure the strain delocalizations are of great importance in predicting assessment of structure integrity. In this paper we are investigating effects of the intrinsic material length on computational prediction of material failure using both cell model, i.e. the conventional micro-mechanical damage model with the constant–sized finite elements for the damage zones, and nonlocal damage model based on the gradient plasticity. The corresponding experiments performed for an engineering steel are taken as reference for verification. The experimental observation has revealed that reducing the specimen size will arise the specific strength of small notched specimen which cannot be predicted using the cell damage model. The nonlocal damage model based on the strain gradient-dependent constitutive plasticity theory reproduces the experimental records. The material length affects evolution of the material porosity and gives an understandable explanation of the size effect.

A three-constituent multicontinuum theory for woven fabric composite materials

Multicontinuum theory (MCT) refers to the use of phase averaged constituent stress/strain fields for predicting failure in composite structural analysis. Given the composite material mechanical properties as well as those of the constituents, well known closed form algebraic expressions exist to decompose the composite stress/strain fields down to the constituent level. Recent research indicates constituent based failure algorithms show a great deal of promise in predicting material failure when coupled to nonlinear finite element codes. A limitation of MCT is that the traditional constituent decomposition is only valid for materials composed of two constituents. In this paper, the MCT decomposition is generalized to handle composite materials composed of three constituents. The application of interest is a woven fabric composite material. The three constituents consist of the warp bundles, fill bundles, and pure matrix pockets. Numerical results are presented for the proposed three-constituent decomposition and are shown to be in good agreement with phase averaged stresses obtained from direct volume averaging of finite element micromechanics models.

Formation of shear bands in plane sheet

The formation of shear bands in plane sheet is studied, both analytically and experimentally, to enhance the fundamental understanding of this phenomenon and to develop a capability for predicting material failure. The evolution of voids is measured and its interaction with the process of shear banding is examined. The evolving dilatancy in plasticity is shown to have a vital role in analysing the shear-band type of bifurcation, and tremendously reduces the theoretical value of critical stresses. The analyses, referring to both localized and diffuse modes of bifurcation, fairly explain the corresponding observations obtained through testing a dual-phase steer sheet and provide a justification of the constitutive model used.