La conception de structures complexes requiert une connaissance exhaustive des matériaux utilisés à la fois au niveau macroscopique et au niveau microscopique. Dans le cas du tantale, peu d'études ont été menées pour des sollicitations cycliques et pour l'influence du vieillissement (au sens de la diffusion des atomes interstitiels vers les dislocations) sur le comportement macroscopique et sur les champs locaux de déformation. Afin de mieux comprendre les mécanismes microstructuraux régissant la déformation, cette étude mène de front des essais avec un suivi de déformation à l'échelle locale couplés à une modélisation adaptée, autorisant la prise en compte explicite de la microstructure.Dans un premier volet, une étude macroscopique du vieillissement et du comportement cyclique du tantale est présentée. Plusieurs techniques expérimentales ont été utilisées, mettant en évidence la propagation d'une bande de localisation de la déformation lors de l'entrée en plasticité. En parallèle, un modèle phénoménologique EKMC (de Estrin, Kubin et McCormick) rendant compte du comportement macroscopique (notamment du pic de traction dû au vieillissement statique) a été identifié ; permettant ensuite des investigations numériques sur les manifestations de la localisation de la déformation.Ensuite, l'étude a été poursuivie à l'échelle locale à la fois expérimentalement et numériquement. Pour ce faire, des matrices micrométriques de plots en nickel ont été déposées à la surface d'un échantillon. Les images sucessives autorisent le calcul des cartes de déformation expérimentales lors d'un essai de traction interrompu. La modélisation explicite de la microstructure a été rendue possible par la génération d'agrégats polycristallins spécifiques avec des conditions de surface libre. L'étude de l'influence de la prise en compte du vieillissement sur l'hétérogénéité des champs locaux de déformation a permis de confronter ces derniers avec les champs expérimentaux.Enfin, une étude multi-échelles du comportement en fatigue a été effectuée. Un essai de fatigue interrompu avec observations microscopiques et construction de cartes de déformation a été réalisé. L'hétérogénéité de déformation locale a ainsi été identifiée et quantifiée, jusqu'à l'amorçage de fissures. Une simulation d'une sollicitation équivalente a été réalisée sur un agrégat polycristallin, permettant la comparaison directe des champs locaux de surface. Des investigations approfondies ont été menées sur l'agrégat pour mettre en place un critère d'amorçage basé sur des grandeurs physiques en accord avec l'expérience. / Designing complex structures requires an exhaustive knowledge of the materials used at both macroscopical and microscopical scales. In the case of tantalum, only few studies have been focused on the cyclic behaviour and on static strain aging's (aging in the sense of atoms diffusion to dislocations) influence on macroscopical behaviour and on the local strain fields. In order to extend the comprehension of microstructural mechanisms driving deformation, this study deals with experiments following the local strain fields evolution and with an appropriate modelling, taking into account microstructure explicitly.In a first step, a macroscopical study of aging and cyclic behaviour is presented. Several experimental techniques have been used in order to highlight a strain localisation band propagation associated to the anomalous yield point phenomenon. In parallel, a phenomenological EKMC (from Estrin, Kubin and McCormick) macroscopic model has been adopted, capturing the macroscopic behaviour (especially the anomalous yield point associated to static strain aging) ; and then used for several numerical investigations about the strain localisation occurence.Then, the study has been continued at the microscale in both numerical and experimental matters. Thus, several micrometrical matrixes of nickel dots have been led on the sample's surface. The have then been used for the experimental strain maps computation during an interrupted tensile test. The explicit computation of the microstructure through a specific generation of polycrystalline aggregates using free surface conditions has permitted to study the influence of aging on the local strain fields heterogeneity and to compare them with the experimental ones.Finally, a multi-scale study of the cyclic behaviour has been carried out. An interrupted fatigue test with microscopic observations and strain maps computations has been carried out, permitting the identification and quantification of the local strain heterogeneity up to the crack initiation. A computation of an equivalent loading has been done on a polycrystalline aggregate in order to set up a fatigue criterion based on physical quantities in agreement with experimental datas.

The aim in this dissertation was to improve our understanding of the effectiveness of shot peening in prolonging fatigue life, of 7075-T6 Aluminium Alloy round bar, taking into consideration surface residual stress, microstructural and micro-hardness parameters. Three point bending, high stress, moderately low cycle, fatigue tests were conducted to study the effects of shot peening and associated surface residual compressive stresses on fatigue life. The influence of shot peening on the microstructure was explored, including the application of mechanical small plastic straining and surface skimming, to vary the surface residual compressive stresses and induce strain hardening. Tests were performed to measure residual stress-depth distribution, plastic straining, micro-hardness, and the microstructure analysed on scanning electron microscopy (SEM) fractographs. The Juvinall and Marshek life prediction model was used in conjunction with the Gerber equation for non-zero mean stress applications to generate a proposed life prediction model for this material which is user-friendly. The proposed life prediction model has a linear equation format with the flexibility to conservatively accommodate most of the various types, and combinations, of treatments applied in this research by the use of customised constants. The results show that there was good correlation between actual and predicted fatigue life as well as useful insights into the role of the microstructure in explaining fatigue life behaviour.

Indiana University-Purdue University Indianapolis (IUPUI) / High-temperature alloys have been extensively used in many applications, such as furnace muffles, fuel nozzles, heat-treating fixtures, and fuel nozzles. Due to such conditions, these materials should have resistance to cyclic loading, oxidation, and high heat. Although there are numerous prior experimental and theoretical studies, there is insufficient understanding of application of the unified viscoplasticity theory to finite element software for fatigue life prediction. Therefore, the goal of this research is to develop a procedure to implement unified viscoplasticity theory in finite element (FE) model to model the complex material deformation pertaining to thermomechanical load and implement an incremental damage lifetime rule to predict the thermomechanical fatigue life of high-temperature alloys. The objectives of the thesis are: 1. Develop a simplified integrated approach to model the fatigue creep deformation under the framework of ‘unified viscoplasticity theory’ 2. Implement a physics - based crack growth damage model into the framework 3. Predict the deformation using the unified viscoplastic material model for ferritic cast iron (Fe-3.2C-4.0Si-0.6Mo) SiMo4.06 4. Predict the isothermal low cycle fatigue (LCF) and LCF-Creep life using the damage model In this work, a unified viscoplastic material model is applied in a FE model with a combination of Chaboche non-linear kinematic hardening, Perzyna rate model, and static recovery model to model rate-dependent plasticity, stress relaxation, and creep-fatigue interaction. Also, an incremental damage rule has been successfully implemented in a FE model. The calibrated viscoplastic model is able to correlate deformations pertaining to isothermal LCF, LCF-Creep, and thermal-mechanical fatigue (TMF) experimental deformations. The life predictions from the FE model have been fairly good at room temperature (20°C), 400°C, and 550°C under Isothermal LCF (0.00001/s and 0.003/s) and LCF-Creep tests. The material calibration techniques proposed for calibrating the model parameters resulted in a fairly good correlation of FE model derived hysteresis loops with experimental hysteresis, pertaining to Isothermal LCF (ranging from 0.00001/s to 0.003/s), Isothermal LCF-Creep tests (withhold time) and TMF responses. In summary, the method and models developed in this work are capable of simulating material deformation dependency on temperature, strain rates, hold time, therefore, they are capable of modeling creep-stress relaxation and fatigue interaction in high-temperature alloy design.

<div>High temperature alloys have been extensively used in many applications, such as furnace muffles, fuel nozzles, heat treating fixtures and fuel nozzles. Due to such conditions these materials should have resistance to cyclic loading, oxidation and high heat. Although there are numerous prior experimental and theoretical studies, there is insufficient understanding of application of the unified viscoplasticity theory to finite element software for fatigue life</div><div>prediction.</div><div><br></div><div>Therefore, the goal of this research is to develop a procedure to implement unified viscoplasticity</div><div>theory in finite element (FE) model to model the complex material deformation pertaining to thermomechanical load and implement an incremental damage lifetime rule to</div><div>predict thermomechanical fatigue life of high temperature alloys.</div><div><br></div><div>The objectives of the thesis are:</div><div>1. Develop a simplified integrated approach to model the fatigue creep deformation</div><div>under the framework of ‘unified viscoplasticity theory’</div><div><br></div><div>2. Implement a physics - based crack growth damage model into the framework</div><div><br></div><div>3. Predict the deformation using the unified viscoplastic material model for ferritic</div><div>cast iron (Fe-3.2C-4.0Si-0.6Mo) SiMo4.06</div><div><br></div><div>4. Predict the isothermal low cycle fatigue (LCF) and LCF Creep life using the damage model</div><div><br></div><div>In this work, a unified viscoplastic material model is applied in a FE model with a combination of Chaboche non-linear kinematic hardening, Perzyna rate model and static recovery</div><div>model to model rate dependent plasticity, stress relaxation, and creep-fatigue interaction.</div><div><br></div><div>Also, an incremental damage rule has been successfully implemented in a FE model. The calibrated viscoplastic model is able to correlate deformations pertaining to isothermal LCF, LCF-Creep and thermal-mechanical fatigue (TMF) experimental deformations. The life predictions</div><div>from the FE model have been fairly good at room temperature (20°C), 400°C and 550°C under Isothermal LCF (0.00001/s and 0.003/s) and LCF-Creep tests.</div><div><div><br></div><div>The material calibration techniques proposed for calibrating the model parameters resulted in a fairly good correlation of FE model derived hysteresis loops with experimental hysteresis, pertaining to Isothermal LCF (ranging from 0.00001/s to 0.003/s), Isothermal LCF-Creep tests (with hold time) and TMF responses. In summary, the method and models developed in this work are capable of simulating material deformation dependency on temperature, strain-rates, hold time, therefore, they are capable to modeling creep-stress relaxation and fatigue interaction in high-temperature alloy design.</div></div>

La fiabilité des composants électroniques, surtout dans les technologies de pointe, prend de l’ampleur. Cela est notamment motivé par la réduction de volume demandée dans des structures fortement sollicitées et dont la durée de vie exigée peut atteindre plusieurs dizaines d’années. Nous considérons, ici, un substrat électronique. Il sert de support à un circuit d’électronique de puissance. Dans un tel circuit, les composants, comme les puces, s’échauffent. Cette augmentation de température au niveau de la puce se reporte sur le substrat. Il subira d’autre part des variations de températures environnementales, telles que le cycle jour/nuit dans les satellites ou imposée par l’ambiance comme dans un puits de forage. Le substrat utilisé est composé de plusieurs couches de céramique et de pistes de conduction en métal entre et à travers ses couches. Lorsque ces différents matériaux se dilatent, cela induit des concentrations et gradients de contraintes. Ces variations étant sur des grandes durées, elles peuvent mener le substrat à rupture par fatigue. Dans le cadre de cette étude, l’objectif est de comprendre les mécanismes de défaillances qui peuvent mener à rupture et les moyens de les éviter. Nous cherchons à déterminer des règles de conception simples permettant d’éviter ces défaillances, comme la taille et la distance entre les différentes pistes traversant la céramique. C’est grâce à l’application de la mécanique de la rupture sur le substrat et en fonction de l’utilisation décrite précédemment, que nous évaluerons les défaillances critiques. La modélisation mécanique du substrat dégagera ainsi des règles de dimensionnement permettant d’éviter ces défaillances. / The reliability of electronic compound, especially in advanced technologies, is becoming very important. This is motivated by the volume reduction asked in highly loaded structures. Moreover, its required lifetime can be about decades. Here, we have an electronic substrate. A power electronic circuit lays on it. In such circuit, compounds like chip are heating up. This rising of temperature from the chip propagates on the substrate. Furthermore others environmental amplitudes of temperature will be imposed like the day/night cycle for satellites or the rock temperature in well bores. The substrate is composed of several ceramic layers with metal conducting tracks inside and in between. When theses materials dilate it induces stress concentrations and gradients. Since theses changes are occuring on important durations, they cans lead the substrate to a fatigue failure. The point of this study is to understand the failure mechanisms leading to break and the means to avoid them. Then we seek to determine simple conception rules such as the size and the gap between different tracks going through the ceramic. Thanks to the use of the fracture mechanics on the substrate and according to the previously described use, we will evaluate the critical failures. Thus, the mechanical modelling of the substrate will generate some rules for the dimensions.

This study experimentally quantified the structure-property relations with respect to fatigue of an extruded AZ61 magnesium alloy and captured the behavior with a microstructure-sensitive MultiStage Fatigue Model. Experiments were conducted in the extruded and transverse directions under low and high cycle strain control fatigue conditions. The cyclic behavior of this alloy displayed varying degrees of cyclic hardening depending on the strain amplitude and the specimen orientation. The fracture surfaces of the fatigued specimens were analyzed using a scanning electron microscope in order to quantify structure-property relations with respect to microstructural features. Correlations between particle size, nearest neighbor distance, and grain size as a function of failure cycles were quantified. Finally, a multistage fatigue model based on the structure-property relations quantified in this study was employed to capture the anisotropic fatigue damage of the AZ61 magnesium alloy.

Indiana University-Purdue University Indianapolis (IUPUI) / Cancer treatment-related fatigue (CTRF) is recognized as a prevalent and bothersome symptom for patients with cancer. In a model of the CTRF experience, CTRF representation, or the beliefs, thoughts and emotions surrounding the experience of CTRF, is believed to mediate the relationship between CTRF intensity and CTRF distress. To date, there is no reported measure of CTRF representation. The purpose of this descriptive, cross-sectional study guided by Leventhal’s Common Sense Model of Self-Regulation was to evaluate an instrument designed to measure CTRF representation, the CTRF Representation scale (CTRFRep), based on an existing measure, the Illness Perception Questionnaire (IPQ-R). The study included 47 patients (mean age=57.7 years) receiving radiation therapy for cancer interviewed one month post-treatment. 77% of patients had fatigue during treatment. Three content experts and one theory expert assessed content validity of the CTRFRep. The content experts included three behavioral oncology nurse researchers whose focus is on symptom management and/or fatigue. The theory expert was a nurse researcher who is an expert in the area of self-regulation theory. As tested, the CTRFRep consisted of 105 items in 10 subscales addressing beliefs about the Identity, Timeline (Acute vs. Chronic/Cyclical), Consequences (positive/negative), Cause, Control (Treatment/Personal), Symptom Coherence, and Emotional Representation of CTRF. When evaluating psychometrics, the Identity and Cause subscales are analyzed independent of the other subscales. For the Identity subscale, symptoms most reported as related to CTRF were lack of energy, loss of strength, and feeling blue. For the Cause subscale, the most common beliefs regarding causes of CTRF were cancer treatment(s), having cancer, and stress or worry. Results indicate adequate reliability in six of eight remaining subscales (α>=0.70); the item N in those subscales was reduced from 56 to 34. To address construct validity, logistic regression assessed whether CTRFRep mediated the relationship between CTRF intensity and CTRF distress. After controlling for negative affect, the Identity and Consequences subscales were significant mediators – the Acute vs. Chronic Timeline and Emotional Representation scales were partial mediators – of the relationship between CTRF intensity on CTRF distress. These findings indicate fatigue is a problem for people undergoing treatment for cancer, and the CTRFRep may be a reliable and valid measure of CTRF representation for patients undergoing radiation treatment for cancer. Small sample size prevented successful factor analysis of the CTRFRep. Further research of the CTRFRep is warranted.

Fatigue is a familiar and commonplace occurrence, but attempts to investigate the nature of fatigue have been inconclusive. Following more than a hundred years of extensive research, the construct is still ill-defined. This has resulted in a series of different strands of research, producing results concomitant with each researcher's own idea of what constitutes fatigue. Two central questions remain unresolved: (1) what sort of a construct is fatigue? and (2) should fatigue be conceptualised as a single, one dimensional state, generated by a range of different conditions, or a multidimensional state, incorporating a number of distinct but related states? There is an implicit assumption within the literature (and every-day language) that there is more than one 'type' of fatigue. However, there is currently no theoretical model which outlines the types of fatigue which should be incorporated in a theoretical framework and which explains the relationships between these fatigue types. The work presented in this thesis represents an attempt to address these issues using both psychometric and experimental approaches. Preliminary work investigated the psychometric basis for a unitary or multidimensional construct. This separately addressed the constructs of state and trait fatigue and, on the basis of the findings, state and trait multiple fatigue questionnaires were developed. A series of four experiments were then carried out which manipulated different types of work to facilitate an investigation of the dynamic development of fatigue. The first three experiments focused on the separate effects of mental and physical fatigue, and the final experiment considered the nature of their interaction.Both experimental and psychometric analyses supported the proposition of a multidimensional construct. The evidence in support of a multidimensional construct of trait fatigue was particularly strong. However, while the evidence in support of a multidimensional construct of state fatigue was less convincing, the experimental manipulations of different types of workload did produce states of fatigue that were subjectively different and also different patterns of fatigue after-effects.

Extensive comparative studies have been undertaken on various theoretical approaches used for analysing spiral strands, in order to identify their appropriate ranges of applicability. Based on the findings of this exercise, a theoretical model is proposed for analysing the coupled axial/torsional response of a stranded wire rope with either an independentwire rope or fibre core. A theoretical model is also proposed for the analysis ofaxial/torsionai coupling of locked coil cables, regarding which, there was previously a . paucity of reliable analytical or experimental information, despite their wide range of applications. Axial hysteresis of locked coil cables is estimated, and it is shown that it is significantly higher than the axial hysteresis associated with spiral strands. Simplified routines which are amenable to hand calculations are developed for obtaining the no-slip stiffness coefficients for the coupled axial/torsional stiffness matrix of axially preloaded spiral strands, with the effect of hydrostatic pressure on sheathed strands catered for. Moreover, the concept of orthotropic sheets for calculating the pattern of interwire/interlayer contact forces in spiral strands is used to determine the recovery length of a broken wire in any layer of a multi-layered spiral strand. Wherever possible, the final theoretical results are compared with experimental data from the available literature and very encouraging correlations have been found in all cases.

Fixed jacket offshore structures. which react environmental wave loading. are generally constructed using tubular steel members. When subject to load. severe surface and through-thickness stress gradients occur due to local bending of the tubular wall. The cyclic nature of the environmental wave loading results in high stress concentration at the joints. which can lead to fatigue cracking. British Gas currently operates twelve fixed offshore structures. two of which. the Rough A-Complex structures. are now ageing and showing signs of fatigue cracking. The objective of the work described in this thesis is the development of a fracture mechanics-based methodology for the assessment of fatigue cracking in these structures. The fracture mechanics approach uses the stress intensity factor (SIl) to characterise crack-tip conditions. and provides a means of analysing the behaviour of cracks. The SIF is defined in terms of the crack site stress distribution and the change in structural compliance with crack size. Difficulties in the application of fracture mechanics lie in the derivation of accurate solutions for the SIF. The tubular joints of offshore jacket structures present particular difficulties due to their complex loading and geometry. The current work starts with a review and assessment of tubular joint fracture mechanics models. followed by a numerical study of cracked tubular joints using shell finite element (FE) models incorporating line spring crack representation. Based on the results of this study. a general fracture mechanics model for the prediction of SIF solutions for tubular joints. is derived and assessed. The general fracture mechanics model is incorporated into a crack growth model. which is best implemented using sophisticated commercial software. Crack growth and fatigue life predictions obtained are validated against full scale tubular joint fatigue data. Finally. a complete methodology for the assessment of fatigue cracks in any tubular joint is proposed, and applied to the assessment of fatigue cracking in the Rough A-Complex structures.