Fatigue crack closure: a review of the physical phenomena

Plasticity‐induced, roughness‐induced and oxide‐induced crack closures are reviewed. Special attention is devoted to the physical origin, the consequences for the experimental determination and the prediction of the effective crack driving force for fatigue crack propagation. Plasticity‐induced crack closure under plane stress and plane strain conditions require, in principle, a different explanation; however, both types are predictable. This is even the case in the transition region from the plane strain to the plane stress state and all types of loading conditions including constant and variable amplitude loading, the short crack case or the transition from small‐scale to large‐scale yielding. In contrast, the prediction of roughness‐induced and oxide‐induced closures is not as straightforward.     

Influence of through-thickness crack shape on plasticity induced crack closure

In most of the previous three‐dimensional (3D) numerical studies of plasticity induced crack closure (PICC), ideal shapes have been assumed for the cracks. The aim of present paper is to study the effect of crack shape on PICC. With this objective a 3D numerical model was developed to predict PICC in middle‐tension (MT) specimens with different thicknesses and crack shapes. The radial size of crack tip elements and the stabilization of closure level were studied to ensure the quality of numerical predictions. Simultaneously, an independent numerical model was developed to predict crack shape evolution, stable crack shapes and corresponding K distributions. Crack closure was found to produce a significant tunnelling effect, with maximum values of ΔK and K_max at the surface. The curved crack presented significant plastic deformation near the free surface which has a high impact on the computation time, compared to the straight crack. The modification of ΔK and K_max with crack shape produced a variation of 38% in opening values at the interior positions, but relatively small variations at the surface. Considering the great influence of crack shape on PICC, it is fundamental to model realistic crack shapes.     

A numerical study of fatigue crack closure induced by plasticity

Crack closure delays the intrinsic mechanisms responsible for crack growth, therefore, it must be considered in fatigue crack growth modelling. The objective of this work is to develop a numerical procedure to predict crack closure induced by plasticity. First the crack closure was experimentally measured on M(T) 6082‐T6 aluminium alloy specimens of 3 mm thickness. A pin microgauge was used with the compliance technique. Then different parameters of the numerical procedure were analysed, namely the finite element mesh and the crack propagation scheme. The size of crack‐tip elements has an important influence and it is recommended to be of the same order of cyclic plastic zone. Crack‐opening levels only 10% lower than experimental results were obtained considering kinematic hardening and two load cycles in each increment.     

The effect of plasticity on dynamic crack growth across an interface

The effect of plasticity on the growth of a crack originating in an elastic solid across an interface and into an elastic-viscoplastic solid is analyzed numerically. The analyses are carried out within a framework where the continuum is characterized by two constitutive relations; one relating the stress and strain in the bulk material and the other relating the traction and separation across a specified set of cohesive surfaces. Crack initiation, crack growth and crack arrest arise naturally as a consequence of the imposed loading, without a priori assumptions concerning criteria for crack growth and for crack path selection. Full transient analyses are carried out. Various values of initial flow strength and cohesive strength of the elastic-viscoplastic solid and of cohesive strength of the interface are considered. With the ratio of cohesive strength of the elastic-viscoplastic solid to cohesive strength of the interface fixed, increasing the ratio of cohesive strength to initial flow strength for the elastic-viscoplastic solid is found to promote crack deflection into the interface.     

Fatigue crack growth and crack closure in an AlMgSi alloy

This study reports an experimental investigation of fatigue crack propagation in AlMgSi1-T6 aluminium alloy using both constant and variable load amplitudes. Crack closure was monitored in all tests by the compliance technique using a pin microgauge. For the constant amplitude tests four different stress ratios were analysed. The crack closure parameter U was calculated and related with Δ K and the stress ratio, R . The threshold of the stress intensity factor range, Δ K _th , was also obtained. Fatigue crack propagation tests with single tensile peak overloads have been performed at constant load amplitude conditions. The observed transient post overload behaviour is discussed in terms of the overload ratio, Δ K baseline level and R . The crack closure parameter U trends are compared with the crack growth transients. Experimental support is given for the hypothesis that crack closure is the main factor determining the transient crack growth behaviour following overloads on AlMgSi1-T6 alloy for plane stress conditions.     

Transient behaviour in the numerical analysis of plasticity induced crack closure

A transient behaviour is observed in the numerical analysis of plasticity induced crack closure at the beginning of crack propagation, as the residual plastic field is being formed. The extent of crack propagation prior to plasticity induced crack closure measurement has a major influence on the accuracy of numerical prediction and on computation time. The objective here is to quantify and understand the minimum propagation, Δa_stb, required to obtain stabilized crack opening values. For plane stress state, Δa_stb was found to increase with ΔK. Under plane strain conditions, a peak of closure exists at the beginning of crack propagation for relatively low ΔK values, which promotes relatively large transient periods. Two driving forces explain the stabilization behaviour, the formation of residual plastic wake and the stabilization of plastic strain, but the second seemed to control the phenomenon. Finally, two strategies are proposed to accelerate convergence. The first, consisting of a progressive increase of maximum load, is relevant in plane strain and 3D studies, in order to eliminate the initial peak. The second strategy consists of an extrapolation model and is very effective for plane stress conditions.     

Fatigue crack propagation prediction by cyclic plasticity damage accumulation models

Different analytical models of damage accumulation by cyclic plasticity have been developed to predict fatigue crack growth from monotonic, cyclic, fracture toughness and crack propagation threshold properties. The models' development logic is condensed as a flowchart, which emphasizes, in a clear and easily comprehensive way, all the required modeling steps. 1020 and API 5L X60 steels and 7075‐T6 aluminum alloy were used in the experimental verification of the models. Samples were extracted from materials of the same heat, in order to have a reliable comparison. The experimental results are better predicted by the models that use the plastic part of Coffin–Manson's equation to calculate the fatigue life of small volume elements ahead of the crack tip, and expressions of the HRR type to represent the elastic–plastic strain amplitude in the cyclic plastic zone.     

Effect of crack closure on non-linear crack tip parameters

Crack closure concept has been widely used to explain different issues of fatigue crack propagation. However, some authors have questioned the relevance of crack closure and have proposed alternative concepts. The main objective here is to check the effectiveness of crack closure concept by linking the contact of crack flanks with non-linear crack tip parameters. Accordingly, 3D-FE numerical models with and without contact were developed for a wide range of loading scenarios and the crack tip parameters usually linked to fatigue crack growth, namely range of cyclic plastic strain, crack tip opening displacement, size of reversed plastic zone and total plastic dissipation per cycle were investigated. It was demonstrated that: (i) LEFM concepts are applicable to the problem under study; (ii) the crack closure phenomenon has a great influence on crack tip parameters decreasing their values; (iii) the ΔK_eff concept is able to explain the variations of crack tip parameters produced by the contact of crack flanks; and (iv) the analysis of remote compliance is the best numerical parameter to quantify the crack opening level. Therefore the crack closure concept seems to be valid. Additionally, the curves of crack tip parameters against stress intensity factor range obtained without contact may be seen as master curves.     

A numerical study of plasticity induced crack closure under plane strain conditions

The level of plasticity induced crack closure (PICC) is greatly affected by stress state. Under plane strain conditions, however, the level and even the existence of PICC still are controversial. The objective here is to study the influence of the main numerical parameters on plane strain PICC, namely the total crack propagation, the number of load cycles between crack increments, the finite element mesh and the parameter used to quantify PICC. The PICC predictions were included in a parallel numerical study of crack propagation, in order to quantify the impact of plane strain values on fatigue life. The results indicate that literature may be overestimating plane strain PICC due to incorrect numerical parameters. The number of load cycles usually considered is unrealistically small, and its increase was found to vanish crack closure, particularly for kinematic hardening. This effect was linked to the ratcheting effect observed at the crack tip. The total crack increment, Δa, must be large enough to obtain stabilized PICC values, but this may imply a huge numerical effort particularly for 3D models. The size of crack tip plastic zone may be overestimated in literature, which means that the meshes used may be too large. Additionally, the crack propagation study showed that the plane strain PICC has usually a dominant effect on fatigue life, and plane stress PICC is only relevant for relatively thin geometries.     

Fatigue crack propagation in Ti–6Al–4V subjected to high strain amplitudes

Fatigue crack propagation in circular Ti–6Al–4V specimens subjected to high strain amplitudes has been investigated. Crack closure was measured with an electrical potential‐drop technique. Closure was shown not to depend on strain ratio but to be a function of the applied strain range. At higher strain ranges, the crack was found to be closed for a smaller part of the load cycle than at lower strain ranges due to blunting of the crack tip. Furthermore, the use of a strain‐intensity approach to predict crack‐propagation rate was investigated, and it was found that for the upper parts of the da/dN curves the effective strain intensity yields good predictions. Also, the effective stress‐intensity factor was found to collapse the da/dN curves for different load ratios.     

Empirical model for plasticity‐induced crack closure based on Kmax and ΔK

The mean stress has a significant effect on crack propagation life and must be included in prediction models. However, there is no consensus in the fatigue community regarding the dominant mechanism explaining the mean stress effect. The concept of crack closure has been widely used and several empirical models can be found in literature. The stress ratio, R, is usually the main parameter of these models, but present numerical results showed a significant influence of K_max. A new empirical model is therefore proposed here, dependent on K_max and ΔK, with four empirical constants. The model also includes the effect of material's yield stress, and two additional parameters were defined to account for stress state and crack closure parameter. A comparison was made with Kujawski's and Glinka's parameters, for a wide range of loading conditions. ΔK_eff lies between Kujawski's and Glinka's parameters, and some agreement is evident, although the concepts are quite different. The crack opening model was applied to literature results and was able to collapse da/dN–ΔK curves for different stress ratios to a single master curve.     

Fatigue crack propagation in cast duplex stainless steels: thermal ageing and microstructural effects

The ferrite phase of cast duplex stainless steels becomes embrittled after thermal ageing, leading to a significant decrease in fracture properties. In the present paper, the influence of ageing and solidification structure on the fatigue crack growth rates (FCGRs) and on the fatigue crack growth mechanisms in a cast duplex stainless steel is studied. FCGRs measured at room temperature increase slightly after ageing at 400 °C, due to ferrite cleavage and to the resulting irregular shape of the crack front. The crack propagates without any preferential path by successive ruptures of ferrite and austenite phases. The macroscopic crack propagation plane depends on the crystallographic orientation of the ferrite grain. Secondary cracks can appear due to the complex solidification structure. This in turn influences the FCGR. The fatigue crack closure level decreases with increasing ageing. This can be explained by a decrease in the kinematic cyclic hardening of these materials.     

Fatigue crack propagation assessment based on residual stresses obtained through cut-compliance technique

Crack growth rate versus crack length curves of heavily overloaded parent material specimens and fatigue crack propagation curves of friction‐stir‐welded aluminium samples are presented. It is shown that in both cases the residual stresses have a strong effect on the crack propagation behaviour under constant and variable amplitude loading. As a simplified engineering approach, it is assumed in this paper, that in both cases residual stresses are the main and only factor influencing crack growth. Therefore fatigue crack propagation predictions are performed by adding the residual stresses to the applied loading and by neglecting the possible effects of overloading and friction stir welding on the parent material properties. For a quantitative assessment of the residual stress effects, the stress intensity factor due to residual stresses K_res is determined directly with the so‐called cut‐compliance method (incremental slitting). These measurements are particularly suited as input parameters for the software packages AFGROW and NASGRO 3.0, which are widely used for fatigue crack growth predictions under constant and variable amplitude loading. The prediction made in terms of crack propagation rates versus crack length and crack length versus cycles generally shows a good agreement with the measured values.     

A parameter for quantitative analysis of plasticity induced crack closure

Numerical models have been successfully developed to predict plasticity induced crack closure (PICC). However, despite the large research effort a full understanding of the links between physical parameters, residual plastic wake and PICC has not been achieved yet. The plastic extension of material behind crack tip, Δy_p, obtained by the integration of vertical plastic deformation perpendicularly to crack flank, is proposed here to quantify the residual plastic field. The values of Δy_p and PICC were obtained numerically in a M(T) specimen using the finite element method. An excellent correlation was found between PICC and Δy_p which indicates that this parameter controls the phenomenon, and can be used to quantify the effect of physical parameters. An empirical model was developed to predict PICC assuming that the residual plastic field is a set of vertical plastic wedges, that the linear superposition principle applies and that the influence of a particular wedge exponentially decreases with distance to crack tip. The model was applied successfully to predict PICC for different residual plastic fields which provided an additional validation of Δy_p as the parameter controlling PICC.     

The effect of overloads on threshold and crack closure

The effect of tensile and compressive overloads on the threshold stress intensity level and crack closure behaviour of one aluminium alloy and three steels has been investigated. A few tensile overloads significantly decreased the crack propagation rate and increased the threshold stress intensity. An initially decreased and then increased opening stress was mostly responsible for the delayed retardation and crack arrest. Intermittant compressive overloads significantly accelerated the crack propagation and decreased the threshold stress intensity which was a function of the frequency of overloading. The opening stress was decreased to below zero after a large compressive peak load, and it took >10⁵ cycles for the opening stress to return to its stable level. During this period an initially high crack propagation rate also gradually decreased to the stable value.     

Characteristics of fatigue crack propagation behaviour as identified by hysteresis loop at the crack tip

Fatigue crack propagation tests have been carried out under various load conditions. Hysteresis loops denoting the relationship between load and strain at the crack tip are obtained by using local compliance measurement. Crack growth acceleration, delayed retardation and non‐propagation phenomena are investigated by considering the variation of hysteresis loop expansion and hysteresis loop tail. Based on the physical meaning of hysteresis loops, two types of crack closure are ascertained and the effect of crack closure on fatigue crack propagation is studied. Results show that change of the effective amplitude of the stress intensity factor at the crack tip is the reason that crack propagation rates vary.     

Numerical simulation of plasticity induced crack closure: Identification and discussion of parameters

Numerical studies play a major role in the understanding and prediction of plasticity induced crack closure (PICC). However, the available numerical models can be considered simplifications of reality as they consider discrete crack propagations, relatively high fatigue crack growth rates (FCGR), sharp cracks, and propagation occurring at well-defined loads. Besides, there are a great number of numerical and physical parameters affecting the predictions of PICC. The aim of this paper is to discuss the numerical study of PICC. The numerical parameters affecting the accuracy of the numerical simulations, and the dependent parameters used to characterise the plastic wake and the closure level, are identified. The influence of the radial size of crack front elements and crack propagation is analysed. An extrapolation model is proposed, with excellent results. An intrinsic uncertainty is associated with the number of load cycles between crack increments and the definition of crack closure level. Finally, the effect of the stress ratio (R) on crack closure level is analysed.     

3D fracture mechanics investigation on surface fatigue crack propagation

Surface fatigue crack propagation is the typical failure mode of engineering structures. In this study, the experiment on surface fatigue crack propagation in 15MnVN steel plate is carried out, and the crack shape and propagation life are obtained. With the concept of ‘equivalent thickness’ brought into the latest three‐dimensional (3D) fracture mechanics theory, one closure model applicable to 3D fatigue crack is put forward. By using the above 3D crack‐closure model, the shape and propagation life of surface fatigue crack in 15MnVN plates are predicted. The simulative results show that the 3D fracture mechanics‐based closure model for 3D fatigue crack is effective.     

Some experimental observations on crack closure and crack-tip plasticity

This study presents a methodology for evaluating crack closure and the effect of crack-tip plasticity on stress intensity. Full-field displacement maps obtained by digital image correlation are used to obtain the mixed-mode, crack-driving force. The methodology allows the quantification of the effect of a range of contact phenomena: effects arising from interlocking, plastic deformation of crack face asperities and wedging generated as a consequence of sliding displacements of fatigue cracks have been identified. By evaluating the effective crack-tip stress intensity factor, crack opening levels can be quantified for both mode I and mode II. Moreover, the approach can take into account plasticity effects local to the crack in determining the stress intensity factor. All the information can be extracted in a non-contacting fashion with equipment that can be easily incorporated into industrial environments.     

A numerical analysis of the mechanisms behind plasticity induced crack closure: Application to variable amplitude loadings

The effect of loading parameters on fatigue crack growth has been explained using the concept of crack closure. Plasticity induced crack closure (PICC) is linked to the crack tip plastic deformation, which becomes residual with crack propagation. The objective here is to identify the main mechanisms behind PICC, and for that different loading cases were considered namely overloads and load blocks. An analytical model was used to isolate the effect of residual plastic deformation on PICC, however significant differences were obtained relatively to finite element results. A second mechanism, which is crack tip blunting, was used to explain the transient behaviour observed after overloads and load blocks. For overloads and low–high load sequences there is a sudden increase of crack tip blunting with load increase which explains the sudden decrease of crack opening level. For high–low load sequences there is a sudden decrease of crack tip blunting which enhances the effect of residual plastic wake. Finally, the partial closure concept was tested looking to non-linear crack tip parameters but no evidences of Donald’s effect were found for the material studied.