No description available.

This thesis explores methods by which carbon fibre reinforced polymers may be fficiently designed with the inclusion of damage tolerance criteria. An efficient method of modelling the compression after impact (CAI) strength of composite materials is selected, and this forms the basis of analysis performed. The CAI model is initially used as the objective in an optimisation routine using a simple genetic algorithm. This indicates features of a damage tolerant composite laminate, namely that plies near the surface are less axially sti® in the loading direction than those nearer the laminate midplane, with a lower Poisson's ratio than the full laminate. This delays sublaminate buckling under laminate uniaxial compression, thus restricting delamination propagation. The designs produced by the optimisation are verified experimentally. In order to improve the computational efficiency of the CAI model a simple surrogate modelling technique for sublaminate buckling is presented. This allows a complete database of results to be produced for a given set of ply angles, in this case standard 0/90/§45± plies. This is used in the full analysis of a collection of layups produced elsewhere to be fully uncoupled, but without the stipulation of midplane symmetry. The surrogate method is shown to reduce computation time by over 99%, and produce results with an average error of less than 0.1% compared to exhaustive analysis. The analysis of the damage tolerance of fully uncoupled laminates shows that the relaxation of midplane symmetry as a design rule gives the designer far more flexibility in layup, and may allow for more damage tolerant laminates to be selected. Finally, the CAI model is incorporated into a stiffened panel design optimisation problem as a constraint. Firstly the panel is optimised using the in¯nite strip analysis tool VICONOPT, with three stiffener geometries. The objective function is minimum mass for a panel subject to compressive and out-of-plane loading, with buckling and strain allowable constraints applied. Damage tolerance constraints are then applied in place of a strain allowable, using a bi-level optimisation approach. This method is shown to allow efficient inclusion of damage tolerance as a constraint in stiffened panel design, although it does not account for interactions in global buckling and local sublaminate buckling which may reduce the strength of the panel. Results indicate that the inclusion of damage tolerance analysis in stiffened panel design shows little benefit for low load panels, but can give significant reductions in mass (up to 30%) for higher load panels.

Barely Visible Impact Damage (BVID) can occur when laminated composite material is subject to impact, i.e. from runway debris or dropped tools, and may result in a significant reduction in the compressive strength of composite structures. A component containing BVID subjected to compression may fail via a number of mechanisms. However, it is assumed that the impact damage problems to be modelled will fail by delamination buckling leading to propagation of damage away from the original site. This precludes problems where the initial mechanism of failure is via kink banding or buckling of the full laminate. An analytical model is presented, for application to various composite structures, which predicts the level of compressive strain below which growth of BVID following local buckling of a delaminated sublaminate will not occur. The model is capable of predicting the critical through-thickness level for delamination, the stability of delamination growth, the sensitivity to experimental error in geometric measurements of the damage area and additionally establishes properties desirable for laminates optimised for damage tolerance. Problems treated with the model are split into two impact categories; ‘face’ (i.e. an out-of-plane skin impact) and ‘free edge’ (i.e. an in-plane stiffener edge impact) and two compressive loading regimes; ‘static’ and ‘fatigue’. Analytical results for static and fatigue compression of face impacted plates show an agreement of threshold strain to within 4% and 17% of experimental values respectively. In particular, for impacts to the skin under a stiffener subject to static loading the model is accurate to within 5%. An optimised laminate stacking sequence has shown an experimental increase of up to 29% in static strength can be achieved in comparison to a baseline configuration. Finally, compression testing has been undertaken on three coupons in order to validate an analysis of static free edge problems. Analytical results are, on average, within 10% of experimental results. An optimised laminate is theoretically predicted to increase static compression after free edge impact strength by at least 35%.

The exploitation of sandwich structures as a means toachieve high specific strength and stiffness is relatively new.Therefore, the knowledge of its damage tolerance is limitedcompared to other structural concepts such as truss bars andmonocoque plate solutions. Several aspects of the damage tolerance of sandwichstructures are investigated. The influence of impact velocityonresidual strength is investigated. Sandwich panels withfaces of glass fiber reinforced vinylester are impacted bothwith very high velocity and quasi static. The residual strengthafter impact is found to be similar for both cases of impactvelocity. Curved sandwich beams subjected to opening bending momentare studied. Faceñcore debonds of varying size areintroduced between the compressively loaded face sheet and thecore. Finite element analysis in combination with a pointstress criterion is utilized to predict the residual strengthof the beams. It is shown that it is possible to predict thefailure load of the beams with face-core debond. Using fractography the governing mode of failure ofcompressively NCF-carbon is characterized. Sandwich panelssubjected to compression after impact are shown to fail byplastic micro buckling. The residual compressive strength after impact of sandwichpanels is investigated. Sandwich panels with face sheets ofnon-crimp fabric (NCF) carbon are subjected to different typesof impact damages. Predictions of residual strength are madeusing the Budiansky, Soutis, Fleck (BSF) model. The residualstrength is tested, and the results are compared topredictions. Predictions and tests correlate well, and indicatethat the residual strength is dependent on damage size and notthe size of the damaged panel. A study of the properties of a selection of fiberreinforcements commonly used in sandwich panels is conducted.The reinforcements are combined with two types of core materialand three types of matrix. Also the influence of laminatethickness is tested. Each combination materials is tested inuni-axial compression, compressive strength after impact andenergy absorption during quasi static indentation. Thespecimens which are tested for residual strength are eithersubjected to quasi-static or dynamic impact of comparableenergy level. Prediction of the residual strength is made andcorrelates reasonably whith the test results. The tests showthat if weight is taken into account the preferred choice offiber reinforcement is carbon.

DNA is susceptible to both exogenous and endogenous damaging agents. Damage is constantly reversed by a wide range of DNA repair pathways. Lesions which escape such repair may cause nucleotide mis-pairing and stalled replication, resulting in mutagenesis and cell death, respectively if left unresolved. Stalled replication is particularly dangerous because replication fork collapse can lead to double-strand breaks (DSBs) and chromosome rearrangement, a hallmark of cancer. DNA damage tolerance (DDT) is defined as a mechanism that allows DNA synthesis to occur in the presence of replication-blocking lesions.<p> DDT, also known as post-replication repair (PRR) in yeast, has been well characterized in the lower eukaryotic model Saccharomyces cerevisiae to consist of error-free and error-prone (mutagenic) pathways. Mono-ubiquitination of proliferating cell nuclear antigen (PCNA) by the Rad6-Rad18 complex promotes mutagenesis by recruiting low fidelity translesion synthesis (TLS) polymerases, while continual Lys63-linked poly-ubiquitination of PCNA by the Mms2-Ubc13-Rad5 complex promotes error-free lesion bypass. Since most of the genes involved in DNA metabolism are conserved within eukaryotes, from yeast to human, I tested the hypothesis that mammalian cells also possess two-pathway DDT in response to DNA damage. Namely, the error-free pathway is dependent on the Ubc13-Mms2 complex, while the error-prone pathway utilizes the TLS polymerases, such as Rev3.<p> By utilizing cultured mammalain cells and producing antibodies against human Ubc13, Mms2 and Rev3, I was able to show that all three proteins associate with PCNA in S-phase cells, and that this association is enhanced following DNA damage. Ubc13-Mms2 association with PCNA was enhanced in response to DSBs. Furthermore, suppression of Ubc13 or Mms2 using interfering RNA technology resulted in increased spontaneous DSBs. In response to UV exposure, Rev3 co-localized with PCNA and two other TLS polymerases, Rev1 and Pol-Ø, at the damage site. UV-induced Rev3 nuclear focus formation was dependent on Rev1 but independent of Pol-£b. Surprisingly, over-expression of Pol-£b was sufficient to induce spontaneous Rev3 nuclear foci. It was further demonstrated that Rev1 and Pol-Ø were independently recruited to the damage site and did not require Rev3. These observations support and extend the polymerase switch model which regulates the activity of the replicative and TLS polymerases. Finally, simultaneous suppression of Rev3 along with Ubc13 or Mms2 resulted in a synergistic sensitivity to UV, whereas simultaneous suppression of Ubc13 and Pol-Ø resulted in an additive effect. These results are consistent with those in yeast cells, implying a comparable mammalian two-pathway DDT model.<p> Additional interesting observations were made. Firstly, Ubc13 interacts with Uev1A, a close homolog of Mms2, which is involved in the NF-£eB signaling pathway independent of DNA damage. Secondly, Rev3 appears to be excluded from the nucleus in a fraction of low passage normal non-S-phase cells, whereas in tumor derived cell lines, Rev3 is consistently enriched in the nucleus independent of cell cycle stage. Finally, Rev3 is elevated during mitosis and associates with condensed chromosomes, suggesting a possible novel role in mitosis. Consistent with this notion, chronic ablation of Rev3 resulted in cell death with inappropriate chromosome segregations. The above preliminary observations require further investigation.

DNA is susceptible to both exogenous and endogenous damaging agents. Damage is constantly reversed by a wide range of DNA repair pathways. Lesions which escape such repair may cause nucleotide mis-pairing and stalled replication, resulting in mutagenesis and cell death, respectively if left unresolved. Stalled replication is particularly dangerous because replication fork collapse can lead to double-strand breaks (DSBs) and chromosome rearrangement, a hallmark of cancer. DNA damage tolerance (DDT) is defined as a mechanism that allows DNA synthesis to occur in the presence of replication-blocking lesions.<p> DDT, also known as post-replication repair (PRR) in yeast, has been well characterized in the lower eukaryotic model Saccharomyces cerevisiae to consist of error-free and error-prone (mutagenic) pathways. Mono-ubiquitination of proliferating cell nuclear antigen (PCNA) by the Rad6-Rad18 complex promotes mutagenesis by recruiting low fidelity translesion synthesis (TLS) polymerases, while continual Lys63-linked poly-ubiquitination of PCNA by the Mms2-Ubc13-Rad5 complex promotes error-free lesion bypass. Since most of the genes involved in DNA metabolism are conserved within eukaryotes, from yeast to human, I tested the hypothesis that mammalian cells also possess two-pathway DDT in response to DNA damage. Namely, the error-free pathway is dependent on the Ubc13-Mms2 complex, while the error-prone pathway utilizes the TLS polymerases, such as Rev3.<p> By utilizing cultured mammalain cells and producing antibodies against human Ubc13, Mms2 and Rev3, I was able to show that all three proteins associate with PCNA in S-phase cells, and that this association is enhanced following DNA damage. Ubc13-Mms2 association with PCNA was enhanced in response to DSBs. Furthermore, suppression of Ubc13 or Mms2 using interfering RNA technology resulted in increased spontaneous DSBs. In response to UV exposure, Rev3 co-localized with PCNA and two other TLS polymerases, Rev1 and Pol-Ø, at the damage site. UV-induced Rev3 nuclear focus formation was dependent on Rev1 but independent of Pol-£b. Surprisingly, over-expression of Pol-£b was sufficient to induce spontaneous Rev3 nuclear foci. It was further demonstrated that Rev1 and Pol-Ø were independently recruited to the damage site and did not require Rev3. These observations support and extend the polymerase switch model which regulates the activity of the replicative and TLS polymerases. Finally, simultaneous suppression of Rev3 along with Ubc13 or Mms2 resulted in a synergistic sensitivity to UV, whereas simultaneous suppression of Ubc13 and Pol-Ø resulted in an additive effect. These results are consistent with those in yeast cells, implying a comparable mammalian two-pathway DDT model.<p> Additional interesting observations were made. Firstly, Ubc13 interacts with Uev1A, a close homolog of Mms2, which is involved in the NF-£eB signaling pathway independent of DNA damage. Secondly, Rev3 appears to be excluded from the nucleus in a fraction of low passage normal non-S-phase cells, whereas in tumor derived cell lines, Rev3 is consistently enriched in the nucleus independent of cell cycle stage. Finally, Rev3 is elevated during mitosis and associates with condensed chromosomes, suggesting a possible novel role in mitosis. Consistent with this notion, chronic ablation of Rev3 resulted in cell death with inappropriate chromosome segregations. The above preliminary observations require further investigation.

<p>The exploitation of sandwich structures as a means toachieve high specific strength and stiffness is relatively new.Therefore, the knowledge of its damage tolerance is limitedcompared to other structural concepts such as truss bars andmonocoque plate solutions.</p><p>Several aspects of the damage tolerance of sandwichstructures are investigated. The influence of impact velocityonresidual strength is investigated. Sandwich panels withfaces of glass fiber reinforced vinylester are impacted bothwith very high velocity and quasi static. The residual strengthafter impact is found to be similar for both cases of impactvelocity.</p><p>Curved sandwich beams subjected to opening bending momentare studied. Faceñcore debonds of varying size areintroduced between the compressively loaded face sheet and thecore. Finite element analysis in combination with a pointstress criterion is utilized to predict the residual strengthof the beams. It is shown that it is possible to predict thefailure load of the beams with face-core debond.</p><p>Using fractography the governing mode of failure ofcompressively NCF-carbon is characterized. Sandwich panelssubjected to compression after impact are shown to fail byplastic micro buckling.</p><p>The residual compressive strength after impact of sandwichpanels is investigated. Sandwich panels with face sheets ofnon-crimp fabric (NCF) carbon are subjected to different typesof impact damages. Predictions of residual strength are madeusing the Budiansky, Soutis, Fleck (BSF) model. The residualstrength is tested, and the results are compared topredictions. Predictions and tests correlate well, and indicatethat the residual strength is dependent on damage size and notthe size of the damaged panel.</p><p>A study of the properties of a selection of fiberreinforcements commonly used in sandwich panels is conducted.The reinforcements are combined with two types of core materialand three types of matrix. Also the influence of laminatethickness is tested. Each combination materials is tested inuni-axial compression, compressive strength after impact andenergy absorption during quasi static indentation. Thespecimens which are tested for residual strength are eithersubjected to quasi-static or dynamic impact of comparableenergy level. Prediction of the residual strength is made andcorrelates reasonably whith the test results. The tests showthat if weight is taken into account the preferred choice offiber reinforcement is carbon.</p>

DNA damage tolerance (DDT) is a process utilized by cells to bypass replication blocking lesions in the DNA, preventing replication fork collapse and maintaining genomic stability and cell viability. In Saccharomyces cerevisiae DDT consists of two branched pathways. One branch allows direct replication past lesions in the DNA utilizing specific error-prone polymerases, a process known as translesion DNA synthesis (TLS). The other branch utilizes homologous recombination and template switch to replicate past damaged DNA in an error-free manner. RAD5 has traditionally been characterized as belonging to the error-free pathway of DNA damage tolerance. The protein is multi-functional, with several specific activities identified and classified to the error-free branch of DDT. However, there is also evidence for additional uncharacterized activities of the protein. The goal of this research was to determine which branches of DNA damage tolerance the uncharacterized activities of Rad5 are involved in. A two-pronged approach was utilized, elucidation of the physical interactions of the protein, and examination of the genetic interactions between RAD5 and other DDT genes. The evidence indicates that Rad5 plays a partial role in TLS and the protein is known to physically interact with Rev1, a member of the TLS pathway. We assumed this physical interaction mediates the TLS activity of Rad5. The yeast two-hybrid assay was utilized to examine the interaction between Rev1 and truncated Rad5 fragments, and the N-terminal 30 amino acids of Rad5 proved sufficient to maintain the interaction. This research sets the stage to identify key residues in Rad5 for the interaction with Rev1, and the creation of a TLS deficient rad5 mutant by targeting those key residues. Genetic interactions between RAD5 and genes required for the initiation of DDT in the cell were examined based on sensitivity to killing by various DNA damaging agents. We determined that the functions of Rad5 rely on PCNA modification, and thus do not function in a cellular process unrelated to Rad5. Potential uncharacterized functions are discussed on the basis of these results and the results of the interaction studies. Future structural and functional studies are proposed to better understand the role of Rad5 in the cell.

Ce travail vise le développement d’une méthode non destructive permettant de contrôler la qualitémécanique des joints collés aéronautiques, en utilisant les ondes de choc générées par impact laser (projetENCOMB). Des chocs ont été réalisés à l’aide de dispositifs tels que des sources laser ou des canons à gaz.Différents diagnostiques ont été utilisés : le VISAR, la VH, la visualisation transverse, la microscopie optiqueet confocale, la radiographie X, le contrôle ultrasons...Des échantillons de résine et des compositesstratifiés carbone/époxy ont d’abord été étudiés. Des chocs laser instrumentés, couplés à des analysespost-mortem, ont permis une meilleure compréhension des phénomènes de choc dans ces matériaux. Lesrésultats obtenus sur les assemblages composites collés montrent que la technique de choc laser permetde discriminer différents degrés d’adhérence. L’utilisation de modèles numériques, développés grâce auxdonnées expérimentales, a permis d’analyser la propagation du choc dans ces assemblages complexes. Cesrésultats ont démontré la nécessité d’optimiser la technique, afin de tester exclusivement l’adhérence dujoint collé, sans endommager les composites de l’assemblage. Plusieurs solutions d’optimisation sontproposées tels que l’utilisation d’une impulsion variable, ou celle de double chocs. Ces solutions ont étévalidées expérimentalement et l’optimisation numérique a fourni les paramètres de choc pour de futurstests. Finalement, ce travail fournit des résultats originaux sur le comportement dynamique de compositesstratifiés et permet de progresser vers l’adaptation du test d’adhérence par choc laser à différentsassemblages composites. / This work aims the development of a non-destructive technique to control the mechanical quality ofaeronautics adhesive bonds (ENCOMB project). Shocks were realized by use of laser sources or gas gun, anddifferent techniques were used to analyse the shocks such as: VISAR, PDV, Shadowgraphy, optical andconfocal microscopy, X-ray radiography, ultrasound testing…Epoxy resins and carbon/epoxy compositelaminates were first investigated. Monitored laser shocks, in addition to post-mortem analyses, enabled tobetter understand the shock phenomenon on these complex materials. The results obtained on bondedcomposite showed that the laser shock technique can be used to discriminate different adherence levels.The use of numerical models, developed thanks to the experimental data, enabled to analyze the shockpropagation in these complex assemblies. They also evidenced the need for optimization, in order to testonly the bond interface and not to break the composite parts on the assembly. Several optimizationsolutions are formulated such as the use of tuneable pulse duration, or double shock configurations. Someof these solutions have been experimentally validated, and the numerical optimization gives the shockparameters to use for the next experimental campaigns. Finally, this work provides original results on thedynamic behaviour of composite materials under laser shock and leads to the adaptation of the laser shockadhesion test to any kind of bonded composite assemblies.

SAFRAN Aircelle produit des panneaux sandwiches constitués de peaux composites collées sur une âme nid d’abeille (« Nida ») en aluminium pour des applications de nacelle de moteur d’avion. Des décollements locaux peuvent exister à l’interface peau/Nida et réduire significativement la capacité structurale de ces sandwiches. L’objectif de cette thèse entre l’ISMANS et SAFRAN Aircelle réside dans la mise en place d’une approche de type « Virtual testing » pour réduire les coûts de validation et de vérification des sandwiches possédant des décollements.L’originalité de ces travaux vient du choix de la représentation des sandwiches à travers deux familles de modèles : les modèles « âme pleine » (modélisation volumique) prévus pour une utilisation en bureau d’études et les modèles « âme creuse » (modélisation physique des clinquants) permettant d’analyser finement les mécanismes de propagation des décollements peau/Nida. Une technique de sous-structuration a été mise en oeuvre avec succès et a permis de réduire significativement les temps de calcul du modèle « âme creuse ».La mécanique de l’endommagement, avec des éléments à zone cohésive, a été choisie pour déterminer numériquement le seuil de propagation des décollements, après une étude comparative avec la mécanique linéaire élastique de la rupture et une méthode de recalage essais. Pour alimenter les lois d’endommagement disponibles dans les codes industriels, une méthode expérimentale a été proposée. Des essais de type DCB sandwiches ont permis de remonter aux propriétés mécaniques de l’interface en mode d’ouverture et en mode mixte, avec un seul montage. Ces essais ont été recalés avec succès, notamment en mode I pur. / SAFRAN Aircelle manufactures sandwich structures made of composite skins bonded to aluminium honeycomb core for aircraft’s engines nacelles applications. Local disbonds may occur at the skins/core interface and lead to significant strength reduction under in-service loadings. The present work was done with ISMANS and SAFRAN Aircelle and deals with the introduction of a “Virtual Testing” approach in order to reduce substantiation and validation cost of sandwiches structures with embedded disbond.In this study we have two kinds of models at two scales to describe the sandwich behaviour: the “homogeneized” model, which provide an industrial tool for design offices and the “detailed” model, which provide a specific tool for accurate analysis of disbond growth initiation at skin/core interface. Due to the numerical cost of the second type of model, the superelement technique has been successfully used which permits to gain computational costs without altering the result quality.Damage mechanics, with cohesive zone elements, have been chosen to numerically determine the disbond growth threshold after a comparative study with linear fracture mechanics and a full experimental method. Inputs data for industrial code using cohesive zone elements have been studied through an experimental investigation. DCB type tests on reinforced sandwiches were perfomed in order to determine the energy release rate at the skin/core interface. Various mode-mixity and pure opening mode are available using the same testing tool. The latter has been successfully modelled.