This research investigates the mechanical properties and performance of structural materials at a small volume scale. In situ observation was made possible through the Small Punch Test (SPT) method as well as tribological testing. Experimentally, aluminum and titanium alloys were examined using those two techniques. Analysis of their behavior in comparison with their published mechanical properties made it possible to establish connections between test parameters and conventional uniaxial tensile test properties. Connections were generated between SPT parameters and tribological performance. This research used experimental approaches to develop an understanding of the material behaviors during small punch testing and apply them to hydrogen embrittlement. The SPT for such alloys were highly repeatable and specimen surface roughness did not have visible impacts on repeatability. Analysis indicated that there is a link between the SPT and conventional mechanical properties. The relationship between the applied force and the slop of the FvE curve is associated with the tensile strength and elastic modulus. It was found that the SPT can be used to qualitatively gage wear resistance. The SPT was used to analyze hydrogen effects, and no significant effects were seen on 3003-H14 and 2618-T61 aluminum alloys; however, effects were seen on a Ti-6Al-4V alloy. It was also found that hydrogen showed no visible effects on friction and wear. The SPT can now be applied more accurately to the testing of aluminum alloys and new doors for the potential of small punch testing in the application of hydrogen embrittlement and surface characterization have been opened. This thesis consists of six chapters. The first chapter serves as an introduction to the background necessary to understand the rational and motivation for the present research. The second chapter will go into detail about the motivation and the objects of the research while the third chapter will explain the experimental procedures that were conducted to fulfill these objectives. The fourth chapter will present the results of these experiments, and they will be discussed in the fifth chapter. Finally, in the sixth chapter, conclusions will be stated and future work will be discussed.

Small punch test (SPT) has been widely used to evaluate in-service materials in nuclear fusion facilities. Early use of SPTs is largely based on empirical relations or curve fitting from experimental data, while recent applications of SPTs take advantage of finite element methods. In this study, an improved inverse finite element analysis procedure is proposed to obtain constitutive relations from load-displacement curves recorded in SPTs. In addition, a parametric study is performed to evaluate the effects of SPT parameters including friction coefficient, punch head diameter, sample thickness, specimen scale and boundary conditions. The proposed inverse finite element (FE) method improves the accuracy of existing inverse FE methods, and the current parametric study provides a basis for the standardization of SPT procedures in the future.

Master's thesis deals with evaluating applicability of correlation relationships between material characteristics determined by Small Punch Test and standard tensile test for material AlSi7Mg0,6 manufactured by casting and technology SLM. Results of Small Punch Tests are correlated with yield strength, tensile strength, elongation, and Young's modulus of elasticity. For each material characteristic various correlation methods are compared, while for each method corresponding coefficients are determined. Consequently, the applicability of individual methods is evaluated by substituting coefficients determined by various studies. Primarily analyzed are correlation methods for which future normalization is expected. The results of master's thesis show that structural inhomogeneity caused by SLM process does not result in high inaccuracies in determining material characteristics. Larger impact on material characteristics has high porosity, which was identified in cast material and led to significant deviations in evaluating tensile strength and elongation.

Die vorliegende Arbeit umfasst die Entwicklung und Implementierung eines nichtlokalen duktilen Schädigungsmodells, die Anwendung von Verfahren zur Bestimmung von Materialparametern aus Versuchsdaten sowie die Berechnung von bruchmechanischen Kennwerten im spröd-duktilen Übergangsbereich durch die numerische Simulation von Bruchmechanikversuchen unter Verwendung der ermittelten Parameter. Das entwickelte nichtlokale duktile Schädigungsmodell basiert auf dem Modell nach Gurson-Tvergaard-Needleman (GTN). Durch die Einführung eines zusätzlichen Längenparameters mittels einer impliziten Gradientenformulierung wird die Netzabhängigkeit des GTN-Modells eliminiert. Zur Lösung des nunmehr gekoppelten Feldproblems wird das nichtlokale Schädigungsmodell in Form eines benutzerdefinierten Elements in ein Finite-Elemente-Programm implementiert. Zur Parameterbestimmung werden Kraft-Verschiebungs-Kurven des Small-Punch-Tests (SPT), einem Kleinstprobenversuch, ausgewertet. Aufgrund des sehr geringen Materialbedarfs für die Herstellung der benötigten SPT-Proben werden Rest- bzw. Bruchstücke von Charpy-Tests weiterverwendet. Es werden zwei ferritische Reaktorstähle im bestrahlten und unbestrahlten Zustand untersucht. Die Versuchsreihen decken den vollständigen Bereich der Zähigkeit beider Stähle ab. Die Bestimmung von Bruchzähigkeiten erfolgt nach dem Konzept des Local Approach allein durch numerische Berechnung des Spannungs- und Verformungszustandes in Bruchmechanikproben. Hierbei werden die zuvor aus dem SPT bestimmten Fließkurven und Schädigungsparameter verwendet. Die berechneten Bruchzähigkeiten werden mit experimentellen Ergebnissen verglichen. / This work comprises the development and implementation of a non-local ductile damage model, the application of methods for the identification of material parameters from experimental data as well as the calculation of fracture mechanics parameters in the brittle-ductile transition zone through numerical simulations of fracture mechanical tests using the identified parameters. The developed non-local ductile damage model is based on the Gurson-Tvergaard-Needleman model (GTN). The pathological mesh sensitivity of the GTN model is eliminated by introducing an additional length parameter by means of an implicit gradient formulation. To solve the coupled field problem, the non-local damage model is implemented in a finite element program in the form of a userdefined element. Force-displacement-curves of the small punch test (SPT), a miniaturised test, are evaluated for the determination of material parameters. Given the modest material requirements for the preparation of the required samples remnants of Charpy-specimens are reused. Two ferritic reactor steels, both irradiated and unirradiated, are examined. The experiments cover the full brittle-ductile transition region of the steels. Following the concept of the Local Approach, fracture toughness values are determined by numerical calculation of the stress and deformation state in fracture mechanics specimen only. Here, the yield curves and damage parameters previously determined from the SPT are used. The calculated fracture toughness values are compared with experimental results.

Die vorliegende Arbeit umfasst die Entwicklung und Implementierung eines nichtlokalen duktilen Schädigungsmodells, die Anwendung von Verfahren zur Bestimmung von Materialparametern aus Versuchsdaten sowie die Berechnung von bruchmechanischen Kennwerten im spröd-duktilen Übergangsbereich durch die numerische Simulation von Bruchmechanikversuchen unter Verwendung der ermittelten Parameter. Das entwickelte nichtlokale duktile Schädigungsmodell basiert auf dem Modell nach Gurson-Tvergaard-Needleman (GTN). Durch die Einführung eines zusätzlichen Längenparameters mittels einer impliziten Gradientenformulierung wird die Netzabhängigkeit des GTN-Modells eliminiert. Zur Lösung des nunmehr gekoppelten Feldproblems wird das nichtlokale Schädigungsmodell in Form eines benutzerdefinierten Elements in ein Finite-Elemente-Programm implementiert. Zur Parameterbestimmung werden Kraft-Verschiebungs-Kurven des Small-Punch-Tests (SPT), einem Kleinstprobenversuch, ausgewertet. Aufgrund des sehr geringen Materialbedarfs für die Herstellung der benötigten SPT-Proben werden Rest- bzw. Bruchstücke von Charpy-Tests weiterverwendet. Es werden zwei ferritische Reaktorstähle im bestrahlten und unbestrahlten Zustand untersucht. Die Versuchsreihen decken den vollständigen Bereich der Zähigkeit beider Stähle ab. Die Bestimmung von Bruchzähigkeiten erfolgt nach dem Konzept des Local Approach allein durch numerische Berechnung des Spannungs- und Verformungszustandes in Bruchmechanikproben. Hierbei werden die zuvor aus dem SPT bestimmten Fließkurven und Schädigungsparameter verwendet. Die berechneten Bruchzähigkeiten werden mit experimentellen Ergebnissen verglichen. / This work comprises the development and implementation of a non-local ductile damage model, the application of methods for the identification of material parameters from experimental data as well as the calculation of fracture mechanics parameters in the brittle-ductile transition zone through numerical simulations of fracture mechanical tests using the identified parameters. The developed non-local ductile damage model is based on the Gurson-Tvergaard-Needleman model (GTN). The pathological mesh sensitivity of the GTN model is eliminated by introducing an additional length parameter by means of an implicit gradient formulation. To solve the coupled field problem, the non-local damage model is implemented in a finite element program in the form of a userdefined element. Force-displacement-curves of the small punch test (SPT), a miniaturised test, are evaluated for the determination of material parameters. Given the modest material requirements for the preparation of the required samples remnants of Charpy-specimens are reused. Two ferritic reactor steels, both irradiated and unirradiated, are examined. The experiments cover the full brittle-ductile transition region of the steels. Following the concept of the Local Approach, fracture toughness values are determined by numerical calculation of the stress and deformation state in fracture mechanics specimen only. Here, the yield curves and damage parameters previously determined from the SPT are used. The calculated fracture toughness values are compared with experimental results.

Diploma thesis is focused on comparison of small punch test results and tensile test results. For experiments were chosen steel P91, 10Ch2MFA, 20CrNi2MoV, 11 416 and 42 2707. These steel are used in nuclear power engineering, or they were developed for this purpose. Basic mechanical characteristics, namely ultimate tensile stress and yield stress, were evaluated from small punch test records of observed materials. This is done via convenient correlation relationships. Obtained values were compared with values obtained by conventional tensile tests. There are mentioned own suggested correlation relationships and comparison of temperature curves gained from small punch test and tensile test in the thesis. Assessment of microstructure of observed materials and analysis of fracture surfaces was provided as well. It was found out, that for assessing basic mechanical characteristics (ultimate tensile stress and yield stress) is the best to suggest own correlation relationships, instead of using universal relationships from literature. The fracture mechanism of all types of assessed steel was ductile in entire temperature range (up to -40 °C).

Der Small-Punch-Test (SPT) ist eine vielversprechende minimalinvasive mechanische Prüfmethode, wenn nur sehr wenig Material für Proben zur Verfügung steht. Die vorliegende Arbeit hat das Ziel, aus Small-Punch-Kleinstproben wahre Materialparameter der elastisch-plastischen Verformung und des spröden Versagens zu bestimmen. Die Kraft-Verschiebungs-Kurve des Versuchs stellt die nichtlineare Materialantwort der inhomogen beanspruchten Probe dar. Das inverse Problem der Identifikation konstitutiver Materialparameter wird numerisch mit Hilfe von Finite-Elemente-Simulationen in Verbindung mit einem Response-Surface-Modell und nichtlinearer Optimierungsverfahren gelöst, indem die Abweichung zwischen gemessener und simulierter Kurve minimiert wird. Mit Hilfe einer eigens entwickelten Kühlapparatur wurden Versuche mit ferritischen Stählen von Raumtemperatur bis hinunter zu -191°C durchgeführt und die temperaturabhängigen Fließkurven identifiziert. Bei tiefen Temperaturen wurden die Weibull-Parameter der zufällig streuenden Sprödbruchfestigkeit bestimmt und die Bruchzähigkeitsverteilung durch Simulation einer CT-Probe vorhergesagt. Für eine Aluminiumoxidkeramik wurden ebenfalls die Weibull-Parameter bestimmt sowie mit Indenterrissen versehene Proben zur Abschätzung der Bruchzähigkeit verwendet. / The small punch test (SPT) is a promising minimally invasive material testing method, especially in cases where only small amounts of material are available. This thesis is aimed at identifying true material parameters of elastic-plastic deformation and brittle fracture. The load-displacement curve of the test represents the non-linear material response of the nonuniformly stressed specimen. The identification of material parameters of constitutive laws is an inverse problem, which is solved numerically. Finite element simulations together with a response surface model and nonlinear optimization techniques are applied to minimize the error between measured and simulated curves. A specially developed cooling apparatus was used to perform tests with ferritic steels from room temperature down to -191°C. The temperature dependent yield stresses and hardening curves were identified. At low temperatures the Weibull parameters of randomly distributed cleavage fracture strength were estimated. The fracture toughness distribution was then predicted by the help of a finite element simulation of a CT specimen. Furthermore the Weibull parameters of an alumina ceramic were determined and its fracture toughness was predicted using specimens prepared with indentation cracks.

Der Small-Punch-Test (SPT) ist eine vielversprechende minimalinvasive mechanische Prüfmethode, wenn nur sehr wenig Material für Proben zur Verfügung steht. Die vorliegende Arbeit hat das Ziel, aus Small-Punch-Kleinstproben wahre Materialparameter der elastisch-plastischen Verformung und des spröden Versagens zu bestimmen. Die Kraft-Verschiebungs-Kurve des Versuchs stellt die nichtlineare Materialantwort der inhomogen beanspruchten Probe dar. Das inverse Problem der Identifikation konstitutiver Materialparameter wird numerisch mit Hilfe von Finite-Elemente-Simulationen in Verbindung mit einem Response-Surface-Modell und nichtlinearer Optimierungsverfahren gelöst, indem die Abweichung zwischen gemessener und simulierter Kurve minimiert wird. Mit Hilfe einer eigens entwickelten Kühlapparatur wurden Versuche mit ferritischen Stählen von Raumtemperatur bis hinunter zu -191°C durchgeführt und die temperaturabhängigen Fließkurven identifiziert. Bei tiefen Temperaturen wurden die Weibull-Parameter der zufällig streuenden Sprödbruchfestigkeit bestimmt und die Bruchzähigkeitsverteilung durch Simulation einer CT-Probe vorhergesagt. Für eine Aluminiumoxidkeramik wurden ebenfalls die Weibull-Parameter bestimmt sowie mit Indenterrissen versehene Proben zur Abschätzung der Bruchzähigkeit verwendet. / The small punch test (SPT) is a promising minimally invasive material testing method, especially in cases where only small amounts of material are available. This thesis is aimed at identifying true material parameters of elastic-plastic deformation and brittle fracture. The load-displacement curve of the test represents the non-linear material response of the nonuniformly stressed specimen. The identification of material parameters of constitutive laws is an inverse problem, which is solved numerically. Finite element simulations together with a response surface model and nonlinear optimization techniques are applied to minimize the error between measured and simulated curves. A specially developed cooling apparatus was used to perform tests with ferritic steels from room temperature down to -191°C. The temperature dependent yield stresses and hardening curves were identified. At low temperatures the Weibull parameters of randomly distributed cleavage fracture strength were estimated. The fracture toughness distribution was then predicted by the help of a finite element simulation of a CT specimen. Furthermore the Weibull parameters of an alumina ceramic were determined and its fracture toughness was predicted using specimens prepared with indentation cracks.

La mise au point de réacteurs nucléaires sous-critiques (Accelerator Driven System - ADS), intrinsèquement sûrs, pour incinérer les déchets nucléaires par transmutation d’actinides de durée de vie longue en radioéléments de plus courte durée de vie, réclament des sources intenses de neutrons. Ces sources sont produites par la technique de spallations « éjections », c’est-à-dire l’interaction d’un faisceau de protons de haute énergie et haute intensité avec une cible contenant des métaux lourds. Compte tenu de la puissance déposée par les protons dans la cible (de l'ordre de quelques MW) seuls des métaux sous forme liquide peuvent à la fois résister aux défauts induits et évacuer la chaleur. Dans ce contexte l'expérience MEGAPIE (MEGawatt PIlot Experiment) fonctionnant avec une fenêtre au niveau de la cible de spallation est pionnière et ouvre la voie vers des cibles de haute puissance. Pour la mise en œuvre de cette cible MEGAPIE, les choix de l’alliage de la cible de spallations, des matériaux de structures de la fenêtre et des parties chaudes sont portés respectivement sur l’eutectique plomb-bismuth, bon rendement neutronique, et l’acier T91 (1%Mo9%Cr en masse), bonne tenue à l’irradiation. Pour les parties froides l’acier 316L a été choisi. Toutefois, malgré sa bonne tenue à l’irradiation, le contact de ces deux alliages (eutectique plomb-bismuth et acier T91) présente quelques inconvénients qui nécessitent d’approfondir la connaissance de leurs propriétés physico-chimiques dans l’environnement MEGAPIE et motive cette présente étude. La contribution de ce travail au développement du réacteur de transmutation a consisté : - d’une part à l'élaboration de revêtements destinés à améliorer les performances de l'acier de structure T91 vis à vis de l'oxydation et de la fissuration par l'eutectique Pb-Bi liquide, - d'autre part à l'étude thermodynamique du système Pb-Bi-(Hg)-Fe-Cr-Al-O pour connaître l’ensemble des équilibres entre phases existant dans les conditions imposées par le fonctionnement du réacteur ADS. Le deuxième volet de ce travail trouve sa justification dans les résultats du premier qui ont mis en évidence le rôle primordial joué par les oxydes mixtes dans l'interaction corrosive entre l'acier et l'eutectique Pb-Bi faiblement oxygéné. / The development of sub-critical nuclear reactors (Accelerator Driven System - ADS), intrinsically safe to incinerate nuclear waste by transmutation of long-lived actinides to radionuclides with a shorter existence, requires intense sources of neutrons. These sources are produced by the "ejected" spallation technique, i.e. the interaction of a high energy, high intensity proton beam with a target containing heavy metals. Considering the power of the proton beam transferred to the target (of a few MW), only metals in the liquid form can be cooled in a closed cooling system and resist to nuclear reactor induced by the proton beam. In this context, the MEGAPIE project (MEGawatt PIlot Experiment), which works with a window at the spallation target, is pioneer and opens the way for high-power targets. The choice of the alloy for the MEGAPIE spallation target is a lead-bismuth eutectic, which is efficient for producing neutrons. For the window beam and hot parts, the best choice at present is the T91 steel (1%Mo9%Cr in mass), due to its good irradiation resistance. For cold parts, the 316L steel was chosen. However, the contact of the two alloys (eutectic lead-bismuth and T91 steel) presents some inconveniences that need to be investigated in relation to their physico-chemical properties in the MEGAPIE environment, which motivates this study. The contribution of this work to this subject consisted of : - In a part, the development of coatings designed to improve the chemical and mechanical performances of the T91 steel in contact of the lead-bismuth eutectic liquid alloy, - In another part, the investigation of the thermodynamic properties of the Pb-Bi-(Hg)-Fe-Cr-Al-O system for all phase equilibria within the conditions imposed by the operation of the ADS reactor. The second part of this work is justified by the results, which, for the first time, highlight the role played by the mixed oxide phases in the corrosive interaction between the T91 steel and Pb-Bi eutectic, in a poorly oxygenated environment.

Coatings are frequently applied on gas turbine components inorder to restrict surface degradation such as corrosion andoxidation of the structural material or to thermally insulatethe structural material against the hot environment, therebyincreasing the efficiency of the turbine. However, in order toobtain accurate lifetime expectancies and performance of thecoatings system it is necessary to have a reliableunderstanding of the mechanical properties and failuremechanisms of the coatings. In this thesis, mechanical and fracture behaviour have beenstudied for a NiAl coating applied by a pack cementationprocess, an air-plasma sprayed NiCoCrAlY bondcoat, a vacuumplasma-sprayed NiCrAlY bondcoat and an air plasma-sprayed ZrO2+ 6-8 % Y2O3topcoat. The mechanical tests were carried out ata temperature interval between room temperature and 860oC.Small punch tests and spherical indentation were the testmethods applied for this purpose, in which existing bending andindentation theory were adopted for interpretation of the testresults. Efforts were made to validate the test methods toensure their relevance for coating property measurements. Itwas found that the combination of these two methods givescapability to predict the temperature dependence of severalrelevant mechanical properties of gas turbine coatings, forexample the hardness, elastic modulus, yield strength, fracturestrength, flow stress-strain behaviour and ductility.Furthermore, the plasma-sprayed coatings were tested in bothas-coated and heat-treated condition, which revealedsignificant difference in properties. Microstructuralexamination of the bondcoats showed that oxidation with loss ofaluminium plays an important role in the coating degradationand for the property changes in the coatings. Keywords:small punch test, miniaturised disc bendingtests, spherical indentation, coatings, NiAl, APS-NiCoCrAlY,VPS-NiCrAlY, mechanical properties