Through-thickness Crack Research Articles

Investigation on new type of fracture in Cr-Mo-V steel slab

Abstract The causes of through-thickness cracking in a continuous casting steel slab with low carbon and high alloys were investigated by observation of the fractured surface and microstructures of a fractured slab, and analysis of phase transformation through a continuous cooling transformation (CCT) diagram. Intergranular brittle fracture and ductile fracture were observed on the surface of the slab and inside the slab, respectively. A bainite and martensite microstructure was transformed in the vicinity of the slab surface where intergranular fracture occurred due to excessive water-cooling during continuous casting, and ferrite and pearlite were transformed inside the slab due to a slow cooling rate and relatively high temperature. Immediately after continuous casting, the temperature and the cooling rate for the surface of the slab were 300–450 °C and about 0.1 °C/s, respectively, and thus bainite and martensite on the surface of the slab were tempered simultaneously during transfer to the slab yard. As a result, the tempered bainite and martensite on the surface of the slab and the mixed microstructure of ferrite and pearlite inside the slab were transformed. However, segregation of P and S at grain boundaries was observed in the tempered bainite and martensite at the surface of the slab. The segregation of P and S at grain boundaries could occur by tempering at the temperature between 300–450 °C and the cooling rate as slow as 0.1 °C/s at the surface of the slab, and is a critical cause on ‘temper embrittlement’. Therefore, it is believed that the intergranular fracture at the surface of the slab was due to the temper embrittlement, and ductile fracture occurred inside the slab having a mixed microstructure of ferrite and pearlite, which is ductile.

Transmission of Lamb waves across a partially closed crack: Numerical analysis and experiment

The transmission characteristics of Lamb waves across a partially closed through-thickness crack in a plate are investigated numerically and experimentally. In the numerical analysis, the spectral element method is used to simulate the transmission of the lowest-order symmetric (S0) and antisymmetric (A0) Lamb modes across a crack in a low-frequency range. The analysis is carried out for an open crack with traction-free surfaces as well as for a partially closed crack modeled as a spring-type interface characterized by normal and tangential stiffnesses. The transmission ratios of both modes are obtained from the spectral amplitude of the simulated transmission waveforms for different crack lengths and interfacial stiffnesses. The numerical results show that the transmission ratio of the S0 mode increases monotonically with the interfacial stiffness, but that of the A0 mode depends on the interfacial stiffness in a non-monotonic manner depending on the frequency. The Lamb wave transmission measurements are carried out for aluminum alloy plates with artificial slits or a fatigue crack. The experimental results for the plates with slits show reasonable agreement with the numerical results for open cracks. The measured transmission ratio of the S0 mode is shown to decrease with the tensile load applied to the plate, but that of the A0 mode shows different load dependence for different frequencies. The qualitative features of the experimental results for the fatigue crack are discussed based on the numerical simulation for closed cracks.

Generalized Equation of the Opening of a Through-Thickness Ring Crack in a Clad Thick-Wall Pipeline

Based on numerical-analytical analysis of a straight pipeline section produced of pearlitic steel clad by buttering of austenitic steel, a crack opening was investigated. These steels have different line expansion coefficients. The generalized equation of an opening of a through-thickness ring crack in a thick-wall clad pipeline was obtained taking into account the internal pressure and thermal stresses. The need to take into account the thermal stresses in the evaluation of an opening of a through-thickness ring crack in the clad pipeline was shown. The investigation results can be used effectively when the foundation of the “leak before break” concept in pipeline transport.

Modelling of the ductile-brittle fracture transition in steel structures with large shell elements: A numerical study

In this paper, the strain energy density (SED) criterion is proposed for predicting the ductile-brittle fracture transition (DBFT) in ships and offshore structures. In finite element simulations, these structures are discretized by relatively large shell elements which precludes the modelling of the local stress and strain states in the vicinity of a crack. Critical values of the SED are determined based on local simulations of fracture for a range of temperatures and plane stress states. The local simulations of fracture are based on combined use of the Gurson model for ductile damage and fracture and the Richie-Knott-Rice (RKR) criterion for brittle fracture. After calibrating the Gurson model and the RKR criterion to existing experiments on offshore steel, critical values of the SED are found by analysing a representative plate element with a generic through-thickness crack using a refined solid element mesh. The proposed failure model is evaluated by simulating drop tests on steel-plated structures found in the literature. The present study indicates that the SED criterion is a useful concept for practical design of ships and offshore structures at sub-zero temperatures, but further assessment against experimental data is necessary to fully establish its credibility.

Experimental and numerical studies on the structural response of normal strength concrete slabs subjected to blast loading

To assess the blast performance of concrete structures, a shock tube has been used to simulate blast loading against concrete slabs of two different mixes ordered as C45/55 and C20/25. The C45/55 slabs were tested without reinforcement, whereas the C20/25 slabs were tested both with and without reinforcement to investigate the effect of adding steel bars on the structural response. The performance of the shock tube itself was also evaluated. Compression tests on cubes and cylinders of both concrete recipes were performed to obtain material input data for numerical simulations. Numerous compression tests were also conducted using the C20/25 mix to obtain insights into the statistical variation in the material properties as concrete is inherently heterogeneous at the macroscopic level. In addition, tensile splitting tests were conducted on both materials. The shock tube tests show that the boundary conditions are important and that the reinforcement will direct the crack pattern. Fluid-structure interaction (FSI) effects are not dominant for this particular problem, because the concrete slabs suffer relatively small deformations before through-thickness cracks appear. Once cracks extend through the thickness, FSI effects may become influential, but it appears that when this occurs there is little or no residual capacity left in the slabs. Numerical simulations were used to investigate the influence of various parameters, revealing that the results are highly dependent on the boundary conditions and particularly on the tensile strength of the concrete. It was found necessary to model the entire clamping assembly to recreate the experimental observations.

Diffusion stable aluminide coatings: Behaviour under steam and oxy-fuel fire-side conditions at 650 °C and accelerated diffusion tests at 900 °C

Two diffusion aluminide coatings obtained at 1050 °C were produced to protect ferritic steels both from steam oxidation and fire-side corrosion at 650 °C. One of them was a simple aluminide coating, whereas the other one was enriched with Cr prior to the aluminizing step. As opposed to aluminide coatings obtained at 700 °C, the new “as deposited” coatings do not exhibit through-thickness cracks and the Al content at the surface is lower. The Cr enriched coating has twice as much Cr at the surface than the aluminide coating produced under the same conditions but without Cr enrichment. Cr at the surface may increase the lifetime of this coating as it is known that it decreases the critical amount of Al required to maintain a protective alumina scale. The two aluminide coatings are very stable under atmospheric steam, as well as under an oxy-combustion atmosphere at 650 °C. They show no significant degradation by diffusion, nor evidence of substrate attack despite the development of some cracks, likely formed during the tests cooling cycles. None of these cracks reached the substrate. This behaviour contrasts that of the coating obtained at 700 °C, which exhibits fast diffusion during exposure at 650 °C. Accelerated diffusion experiments were also carried out at 900 °C under steam, comparing the Cr enriched and unenriched aluminide coatings in order to establish if the higher Cr content present in the enriched coating allows maintenance of the protective alumina scale for longer exposures. However, Cr from the substrate diffused outwards equalizing the Cr surface content of both coatings.

Identification of Orientation-Specific Damage Using Guided Waves and Diagnostic Imaging

Guided-waves-based diagnostic imaging techniques have been attracting much attention due to their merits including easily interpretable image, high identification accuracy and suitable for online surveillance. In this study, to envisage the difficulty in detecting orientation-specific damage (crack, notchetc.), a novel guided-waves-based diagnostic imaging technique capable of inspecting complicated engineering structures was developed, in terms of the relationship between damage parameters (location, orientation and severity) and extracted guided waves signal features (time-of-flight, signal correlation and signal energy). Experimental studies were performed to verify the developed diagnostic imaging approach, where a through-thickness crack was successfully identified in a metallic plate and a part of real rail structure respectively.

Study of thermal ageing effects on Rh coating's mechanical performance upon CuCrZr substrate through modeling and experimental methods

Rhodium (Rh) coating on CuCrZr substrate is a promising material option for optical, structural and electrical applications on nuclear fusion reactors. For these applications, Rh coated CuCrZr components subject to long time of thermal ageing due to pre-treatment or normal operation condition. In this paper, both finite element method (FEM) and experimental method were applied to investigate the effects of thermal ageing on mechanical performance of Rh coating after 250 °C, 500 h baking in vacuum. Based on FEM analysis, thermal stresses which concentrate at Rh coating interface is the main source of cracking, and such stresses can be minimized efficiently by introducing a 0.5 μm Au interlayer into the coating layer structure. According to thermal ageing experiments, through-thickness cracking in the Rh coating due to thermal stress releasing and voids generated at the Rh bonding interface caused by Kirkendall effect were the main micro-structure changes in the coating system. The solid-solution hardening caused by significant Cu diffusion into Rh is the dominant factor that affected the Rh coating's hardness. The existing of large amount of cracks in the Rh coating and voids at the Rh coating interface deteriorated the adhesion performance of Rh on CuCrZr substrate by 30%.

Crack Propagation Analysis of Fiber-Reinforced Composite Hollow Transmission Shaft

The main purpose of this research was to investigate crack propagation resistance of shaft. Hollow transmission composite shafts were produced and their crack behavior was investigated. Glass fiber-reinforced polyester polymer composite material is used for the hollow transmission shaft of Hidasie H260 helicopter model of Dejen Aviation Engineering Complex, DAEC. The elastic engineering constants of composite transmission shaft are determined using the Classic Laminate Theory, CLT. Linear Elastic Fracture Mechanics, LEFM was used to analyze its crack propagation by taking a through-thickness central crack. The stress at the crack tip is analyzed and it was found that the glass fiber-reinforced polyester polymer composite transmission shaft has good resistance to crack propagation. FEA method was used to crack propagation analyses. The model of shaft is generated using Finite Element Method. Due to the symmetry of the problem, only a quarter part of it is modeled and analyzed. The crack-tip region was meshed using quarter point (singular) 8-node quadrilateral elements to get accurate results and the analysis used a fit of the nodal displacements in the vicinity of the crack tip. The maximum stress is at the crack tip since there is a stress concentration at sharp edges.

Fracture toughness and damage development in limpet shells

Fracture toughness is an important property for many biological materials, but it can be difficult to obtain accurate and relevant values of toughness in such materials owing to complexities of geometry, material anisotropy, etc. Here we present the results of the first ever attempt to describe and measure cracking and fracture toughness in the shells of limpets. Three different experiments were devised. Firstly, small single-edge-notched bend specimens were machined, enabling us to measure KIC for through-thickness cracks growing in the circumferential direction in the shell walls, giving a value of 0.98 MPa√m. Secondly, radial notches were cut into intact shells which were loaded in compression through the apex. Failure occurred by crack propagation from the notch roots, and finite element analysis (FEA) was used to obtain critical K values. However the analysis gave a surprisingly high toughness value and the results were very sensitive to test variables, especially friction. The experiment demonstrated the remarkable resistance of shells to this kind of damage, but could not be used to measure KIC. Thirdly, impact tests were carried out to create internal damage in the form of delamination cracks. This allowed us to estimate toughness in terms of a crack propagation energy GIC for these cracks of 146 J/m2, equivalent to a KIC of 2.59 MPa√m. Scanning electron microscopy showed that the delamination cracks had much smoother fracture surfaces than those from the through-thickness cracks, however they displayed a regular structure of folds or pleats at the 100 nm scale which may act to hinder crack face movements during shear/compression loading as occurs under impact, which is a common cause of damage for these shells in their natural surroundings.

A novel approach to evaluate mixed-mode SIFs for a through-thickness crack in a welding residual stress field using an effective welding simulation method

This paper accurately evaluates mixed-mode SIFs (MM-SIFs) for cracks in welding residual stress (WRS) fields using a novel approach based on finite element method (FEM). For this reason, the interaction integral method (IIM) available in WARP3D, an open source code, was developed to include normal and shear stress components in MM-SIF solutions. The effectiveness of the developed WARP3D-IIM was verified based on the superposition method and by means of well-established analytical and numerical reference solutions using an inclined through-thickness cracked FE model. After validating the adequacy of the developed WARP3D-IIM, different welding simulations were conducted using JWRIAN (Joining and Welding Research Institute ANalysis). JWRIAN, a sophisticated welding simulation in-house code, was adopted as a welding simulation tool to predict WRS. Different weld line (WL) orientations, as well as different perpendicular distances between WLs, were performed to investigate their influence on the behavior of MM-SIFs. The developed WARP3D-IIM and superposition method were thereafter employed to evaluate MM-SIF solutions for an inclined through-thickness crack in the simulated WRS fields. Different crack angles were also employed to examine the effect of the induced WRS on calculated MM-SIFs. SIF solutions resulting from WRS were validated with analytical reference solutions. Results showed that WRS has a considerable influence on the behavior of MM-SIFs. Moreover, WL orientation as well as the perpendicular distance between WLs revealed a significant influence on the induced WRS and thus MM-SIF solutions. On the other hand, the crack angle showed a remarkable effect on the behavior of MM-SIFs. MM-SIFs calculated at through-thickness crack fronts were influenced by the induced WRS distribution as well as crack angle.

Heat cracks in brake discs for heavy vehicles

Existing models describing the formation of heat cracks in heavy-duty vehicles are not detailed enough to predict the sensitivity of brake discs for heat cracks. In this paper, results from a dynamometer experiment are discussed. They allow a deeper understanding of the processes that result in the failure of brake discs. The experiment described is a standardized heat crack test, which was performed over 300 cycles and ended with a through-thickness crack. During the test, the brake disc was monitored via a thermographic camera, a pyrometer, two sliding thermocouples, a set of capacitive displacement sensors, and an eddy current crack detector. The results show that the current explanation model might not consider several factors that contribute to the crack propagation. Cyclic shear movements at the disc surface are observed, which are likely caused by the cooling channel pins. The variation of the coning of the disc correlates with the maximum disc surface temperature and the crack growth rate. The consequences of the observations are discussed to develop a more detailed model to explain the formation of heat cracks.

Analysis of cracked plate using higher-order shear deformation theory: Asymptotic crack-tip fields and XIGA implementation

In this work, an extended isogeometric analysis (XIGA) is used for the analysis of through-thickness crack in a homogeneous and isotropic plate. In isogeometric analysis (IGA), non-uniform rational B-splines (NURBS) are used as a basis function. The plate kinematics is modelled by Reddy’s higher-order shear deformation theory (HSDT). The C1 continuity requirement of HSDT can be easily fulfilled by the NURBS basis functions. In order to obtain the plate fracture parameters (moment intensity factors), the expressions of crack-tip fields (auxiliary fields) are derived using separation of variables and Eigen-function approach. A new expression for moment intensity factors is developed using auxiliary fields solution (crack-tip fields) and interaction integral approach. Several cracked plate problems are solved by XIGA using HSDT. The results obtained by HSDT based XIGA (HSDT-XIGA) are compared with the FSDT based XIGA (FSDT-XIGA) and literature solutions.

In-situ stress gradient evolution and texture-dependent fracture of brittle ceramic thin films under external load

Brittle ceramic thin films failure problems induced by film stress during service have been recently getting more attention. Thus, it is crucial to know the stress to understand correlated film fracture. In this work, in-situ stress gradient evolution during cracking of TiN films under external load was investigated by grazing-incidence X-ray diffraction using the optimized cos2α sin2ψ method. Counterintuitively, cross-sectional scanning electron microscopy clearly demonstrated that cracks initiated in the middle of the film thickness and propagated to the surface and/or the interface with increased load. Eventually, film through-thickness crack and even delamination occurred. Preferred orientation evolution and texture transition along the film cross-section were characterized by selected area electron diffraction in transmission electron microscopy and the Kikuchi diffraction analysis. Thermodynamics modeling revealed that system energy was the driving force for dominant texture transition. The texture-dependent film fracture toughness accounts for the film cracking behavior. A model of fracture toughness is proposed to consider and evaluate the effect of the preferred orientation. The fracture toughness in the middle of the film, where the texture transition zone is located, is smaller than at the surface or the interface. Consequently, the texture transition zone of the film becomes the crack initiation point, which finally leads to the whole film-substrate system failure. Consequently, the texture transition zone of the film becomes the crack initiation point, which finally leads to the whole film-substrate system failure.

Study on Fatigue Design Method of an Engine Component Produced by Flake Graphite Cast Iron for Different Loading Modes and Mean Stresses

In order to improve reliability of an automotive engine, a fatigue design method was studied based on the fatigue mechanisms in flake graphite cast iron. In the present study, fatigue test specimens were cut from an actual engine component. The material used showed nonlinear behavior in stress–strain relationship from lower stress level during tensile testing. Rotating bending fatigue testing and axial load fatigue testing were performed. Fatigue crack initiation and propagation behavior were observed during the fatigue tests by a replication technique. Fatigue cracking initiated from the tip of flake graphite in early stage of the total fatigue life when the applied stress was higher than the fatigue limit. On the other hand, fatigue cracking was not observed in the specimen tested with applied stress below the fatigue limit. The threshold stress intensity factor obtained with a small surface crack was lower than that obtained with a through-thickness crack. Fatigue limits under various loading conditions such as bending loading and axial loading were predicted well based on fracture mechanics using the $$ \sqrt {\text{area}} $$ parameter. Threshold stress intensity factor for a small surface crack and the maximum graphite size assumed as an initial crack were used for the prediction. Axial load fatigue tests with different stress ratios were performed. The fatigue limit diagram obtained by the fatigue tests did not follow the modified Goodman relation. On the other hand, fatigue limits prediction based on fracture mechanics approach showed good agreement with fatigue limit obtained with fatigue tests in a wider range of mean stresses σmean varying from − 150 to 50 MPa, in negative stress ratio. In case of σmean higher than 50 MPa, in a positive stress ratio, a smaller effect of stress ratio in the fatigue limit diagram was observed. It was considered that this phenomenon could be explained by the lower effect on the crack closure in a small surface crack initiated from the tip of graphite flake and also influence of local plastic behavior of the matrix around the tip of graphite flake. It can be summarized that the fracture mechanics approach is an effective way to predict fatigue limits of an engine component produced by flake cast iron in different loading conditions and in wider range of mean stress conditions.

Cracking in paintings due to relative humidity cycles

A numerical study is performed using the finite element method to consider the effects of low-cycle fatigue, specifically induced through relative humidity cycles on paintings. It has been identified that there are two major crack types in paintings, these being (i) an interfacial crack (delamination) between paint and support and (ii) a through-thickness (channel) crack in the paint layer itself, arresting on the interface. Therefore a 2D plane strain model for each type of crack has been created, which both consist of an alkyd paint modelled using a visco-hyperelastic material model and a primed canvas which is assumed to behave in a linear elastic manner. To account for fatigue damage in both models, cohesive elements located along the interface or through the film thickness respectively, are used and the traction-separation law has been modified to incorporate a fatigue damage parameter. It is possible to expose the models to the same relative humidity cycles, which would typically be seen in museums, enabling the prediction of time to first crack and which crack type is more readily grown in the painting.

Characterization of Membrane Degradation Growth in Fuel Cells Using X-ray Computed Tomography

Perfluorosulfonic acid ionomer membranes are subjected to simultaneous chemical and mechanical degradation under fuel cell operation. Despite the importance of membrane durability, the understanding of its structural degradation and failure modes has been considerably restricted by conventional 2D imaging. In this work, non-invasive micro X-ray computed tomography (XCT) is adopted to visualize the 3D membrane decay at different life stages during combined chemical and mechanical degradation. A detailed survey exhibits damage density of 6 and 10 cracks per mm2 observed at the near-final and final end of life stages respectively. Through-thickness membrane cracks with unbranched I-shaped cracks and Y- and X- shaped cracks with one and two branches respectively are observed. The observed damage development at each life stage is correlated to supplementary diagnostic data including hydrogen leak rate, open circuit voltage, and tensile strength. In particular, large X-shaped cracks formed due to embrittlement from underlying chemical degradation are deemed to have a critical impact on the eventual failure development by facilitating large hydrogen leaks. Overall, the comprehensive 3D perspective enabled by XCT imparts new knowledge pertaining to the degradation process, and could also be extended to other fuel cell failure modes and degradation mechanisms.

Blast loading of concrete pipes using spherical centrically placed C-4 charges

The Norwegian Public Roads Administration has initiated a large research project on a potential ferry-free coastal highway route E39 from Trondheim to Kristiansand. Parts of this route involve crossing somewide and deep fjords, where one of the suggested crossing alternatives is a Submerged Floating Tunnel (SFT)built in concrete. A major concern regarding an SFT is internal blast loading from accidents involving gas orfuel tankers, or from terrorist attacks. To assess the performance of tubular concrete structures exposed to blastloading, 1760 mm long plain concrete pipes with an internal diameter of 200 mm and pipe wall thickness of41 mm were loaded by spherical C-4 charges of various sizes up to 25 g. Pressure sensors in different directionsequidistant from the charge location recorded the blast pressure, while two high-speed cameras captured theevents. Using mass produced commercial off-the-shelf pipes ensures consistent geometry and material proper-ties. A wooden pallet was used as support for the pipes, and an electrically initiated blasting cap was used todetonate the C-4. A charge size of at least 13 g C-4 was needed to breach the pipe wall. Larger charge sizescaused longitudinal cracks and broke the pipe into oblong pieces. The test results were used for comparisonwith simulations using the explicit finite element solver EUROPLEXUS, intended for fast transient dynam-ics involving fluid-structure interaction. Both purely Lagrangian simulations and fully coupled fluid-structureinteraction simulations were carried out. While the former method is more efficient computationally, the lat-ter approach accounts for reflections and the confinement of the pipe. It can be difficult to predict the exactcharge size where through-thickness cracking occurs, but the qualitative results were encouraging for furtherwork. When accurate simulations of the experiments are obtained, the simulations will be extended to a fullcross-section of a submerged floating tunnel.

Effects of compressive residual stress on short fatigue crack growth in a nickel-based superalloy

The effects of compressive residual stress on short fatigue crack growth in Inconel 718 have been investigated. Using two different indentation procedures, controlled plane-strain compressive residual stress fields were applied to short through-thickness cracks, which had been generated by machining away wakes of long cracks that had been grown down to threshold levels at a stress ratio of 0.1. Short fatigue crack growth tests were conducted at stress ratios of 0.1 and 0.7. At a stress ratio of 0.1, the residual stress-free testpieces demonstrated typical short fatigue crack growth behaviour, indicated by a reduction of threshold value and increases in fatigue crack growth rates compared to those of long fatigue cracks. A significant decrease in short crack growth rates was observed within the compressive residual stress region of indented testpieces, together with an increase of threshold values. At a high stress ratio of 0.7, any similar decrease in rate is barely observable. This indicates that the effects of compressive residual stress on short fatigue crack growth are monotonic in sign and could perhaps be simulated by a crack closure approach similar to that applied for long fatigue cracks. However the superposition principle which incorporates the compressive residual stress as a negative stress intensity factor, despite many successful applications to long fatigue cracks, is found to be inappropriate here. Hence one should be cautious when extending such a methodology to short fatigue cracks as it may result in optimistic life estimations.

Effect of thickness on the damage tolerance of glass/epoxy laminates subject to repeated impacts

To evaluate the influence of layer and laminate thickness on the repeated impact response of E-glass/epoxy laminates, two types of cross-ply stacking sequences ([0/90]3n and [03/903]n), with two different thicknesses (4 mm and 8 mm), were analyzed. The investigation, led by running quasi-static perforation tests and repeated impact tests, outlined the better performance of [03/903]n laminates. Experimental results were analyzed through a macroscopic damage variable (damage index DI) and by microscopic observation and digitalized profiles of impacted specimens. The higher damage tolerance of [03/903]n laminates was found to depend on the capability of spreading the damage over a more extended area and of tolerating a higher permanent deflection. For both laminate types, the end of the steady damage accumulation phase was observed to correspond to through-thickness transverse cracking.