Friction stir welding (FSW) and aluminium alloy 2139-T8 are currently being considered for use in future military vehicles. However, stringent regulations on weld integrity under extreme loading conditions limit the adoption of new technologies. Moreover, current finite element (FE) based methods do not give reliable predictions of strain distribution in welds, which makes it difficult to assess the performance of structures. Therefore, an extensive research program was carried out to develop a generalised finite element (FE) based methodology to predict the response of welded structures under complex loading configurations. The methodology enables the complex distribution of mechanical properties arising from welding, which is linked to microstructural variation, to be incorporated into a macro scale structural model. The method is general, and is applicable for any heat treatable aluminium alloy under a range of joining processes. To achieve this, the microstructure of 2139-T8 alloy was characterised at a range of length scales, with particular emphasis on the size and distribution of strengthening Omega precipitates. 2139-T8 was subjected to bead on plate FSW to enable characterisation of the effects of processing on the local microstructure. In addition, kinetic data for 2139-T8 was generated, allowing a simple softening model to be developed; this allowed the post-weld strength distribution to be predicted. The model was also used to recreate bulk specimens of 2139-T8 with equivalent local weld microstructure, which was verified by transmission electron microscopy. Material with equivalent microstructure was used to estimate the local mechanical property distributions across the weld, including the initial yield stress and plastic response; the mechanical properties of 2139-T8 are known to be representative of 2139-T84. From observations of this combined data, a methodology was developed to enable the estimation of the complex mechanical property distributions arising during welding. Furthermore, an automated computer program was written to implement the property distributions into FE based models. The methodology was verified using data generated for 2139-T8 and was used to simulate the response of FSW 2139-T8 loaded in uniaxial tension. The simulations were verified experimentally using digital image correlation (DIC) and the methodology was shown to demonstrate increased accuracy and reliability over existing FE methods, with respect to strain predictions. In addition, the method eliminates the need to calibrate the structural model to a particular loading configuration. Theoretically, the models are insensitive to loading and this property was tested by extending the model to simulate the strain distribution of large scale welded panels subject to explosive blast loading. The simulations were verified against blast tests where FSW 2139-T84 panels were subjected to blast loading from the detonation of plastic explosive. The results indicate that the modelling methodology developed is capable of producing accurate and reliable predictions of strain distribution in welded structures under complex loading configurations.

The effect of friction stir processing (FSP) on the mechanical wear behavior was investigated for Ti-1Al-8V-5Fe (Ti-185) and stainless steel (Incoloy® A-286) alloys. The Ti-185 and A-286 alloys were tested in different processing conditions, including as rolled (AR), AR+FSP, and AR+FSP+aged. A high frequency reciprocating rig was used to simulate fretting-type wear of these alloys at room temperature. The Vickers micro-hardness and wear rates were calculated and compared for each processing condition. It was determined that along with increasing hardness in the stir zones, FSP resulted in improved wear resistance for both alloys. Specifically, wear rates in the stir zones were reduced to lowest values of 1.6 x 10-5 and 5.8 x 10-7 mm3/N·m for the AR+FSP+aged Ti-185 and A-286 alloys, respectively, despite lower hardness for A-286 alloy. Mechanistic studies were conducted to determine the reason behind these improvements in wear resistance and the effect of FSP on the microstructural evolution during wear. For the Ti-185 alloy, x-ray diffraction revealed that there was a phase transformation from β-Ti (AR+FSP) to α-Ti (AR+FSP+aged). This phase decomposition resulted in the harder and stiffer Ti phase responsible for lowering of wear rate in Ti-185. While x-ray diffraction confirmed the A-286 alloy retains its austenitic structure for all conditions, scanning electron microscopy revealed completely different wear track morphology structures. There was increased coarse abrasion (galling) with the AR+aged A-286 alloy compared to the much finer-scale abrasion with the AR+FSP+aged alloy, which was responsible for smaller and less abrasive wear debris, and hence lower wear rate. Furthermore, cross-sectional focused ion beam microscopy studies inside the stir zone of AR+FSP+aged A-286 alloy determined that a) increased micro-hardness was due to FSP-induced microscopic grain refinement, and b) the corresponding wear rate decrease was due to even finer wear-induced grain refinement. With both effects combined, the level of damage and surface fatigue wear was suppressed resulting in lowering of the wear rate. In contrast, the absence of FSP-induced grain refinement in the AR+aged A-286 alloy resulted in lower hardness and increasing wear rate. In addition, micro-Raman spectroscopy inside the stir wear zone determined that the wear debris contained metal oxides of Fe3O4, Cr2O3, and NiO, but were a consequence and not the cause of low wear. Overall, FSP of titanium and stainless steel alloys resulted in lowering of wear rates suggesting it is a viable surface engineering technique to target and mitigate site-specific wear.

Friction Stir Welding (FSW) is a new solid-state welding process that shows great promise for use in the aerospace and transportation industries. One of the primary benefits of this process is that mechanical properties of the base material are not as severely degraded as they are with conventional fusion welding. However, fatigue crack initiation and growth properties of the resulting weld nugget are not fully understood at this time. The primary goal of this project is to characterize the fatigue crack growth properties of friction stir welds in 6061-T6 aluminum as relates to the microstructural evolution of the weld. This was accomplished by producing friction stir welds and testing fatigue crack growth response in different crack orientations with respect to the weld. In addition, residual stress measurements were conducted for all cases, using both the crack compliance and contour methods. The results from the methods were compared in order to evaluate the accuracy of each method. Being an immature technology, the potential for discovery of new applications for the FSW process exist. With this in mind, novel applications of the FSW process, including the addition of particles during welding were explored. The first step was the investigation of property changes that occur when secondary cast phases are refined using the FSW process. The FSW process successfully refined all secondary phases in A380 and A356, producing an increase in hardness. Next, methods for the creation of particle metal matrix composites using FSW will be investigated. Nano-scale alumina particles were successfully added to the matrix and homogenously distributed. Using multiple weld passes through the composite was found to increase the uniformity of particle distribution. However, the alumina particle composite failed to provide any statistically significant hardness increase over the base material. The FSW process was also evaluated for weldability of traditionally difficult alloy systems. FSW was found to show very good weldability for dissimilar cast and wrought alloys, as well as for high-pressure die castings. Lastly, the feasibility of friction stir welding/processing in repairing crack defects in complex structural members in combination with cold-spray technology was determined. Friction Stir processing was used on a cold spray 6061-T6 block, resulting in significant increases in hardness over the base material, as well as a reduction in porosity. In addition, FSP was shown to eliminate crack-type defects in cold spray materials, a finding that has important applications in part repair. The deliverables of this work include an understanding of the fatigue crack growth response of FSW/FSP 6061-T6, as well as a feasibility study exploring novel uses for the FSW/FSP process. In addition, the deliverables include CNC code, fixtures, procedures, and analytical code for the creation and analysis of FSW/FSP joints. This will be important for the continuation of FSW/FSP work at WPI.

As ligas a base de Níquel são frequentemente utilizadas na indústria do petróleo devido à elevada resistência mecânica e excelente resistência à corrosão. Neste cenário, a utilização da soldagem como etapa na fabricação de dutos rígidos de grande comprimento é inevitável. Assim, chapas da liga Inconel 625® foram unidas através do processo Friction Stir Welding (FSW). A soldagem foi executada com uma ferramenta de nitreto cúbico de boro policristalino e o inerente desta ferramenta foi analisado. Em um primeiro momento, na seleção de parâmetros de processo, diversas condições de processo foram analisadas para escolha da melhor solda. Após esta etapa, novas juntas soldadas foram produzidas com o parâmetro de processo adequado (velocidade de rotação da ferramenta de 200 rpm e velocidade de avanço de 1 mm/s). A caracterização da superfície de topo da solda foi realizada através da microscopia ótica, microscopia eletrônica de varredura, microdureza e ensaios de dobramento. Defeitos superficiais foram inspecionados com líquidos penetrantes. Além disso, os estados de tensões residuais foram obtidos com a difração de raios-X. Ainda, a resistência à corrosão localizada foi estudada com o ensaio de imersão em cloreto férrico e a determinação da temperatura crítica de pite (conhecida como CPT) Por fim, a susceptibilidade à corrosão intergranular foi avaliada com o ensaio Streicher Test. O processo de soldagem promoveu o refino de grão e o aumento de dureza na solda. As micrografias do metal de base (MB) mostraram carbonetos do tipo MC, M6C, M23C6. Por outro lado, a melhor junta soldada apresentou uma microestrutura livre de defeitos, homogênea, onde não foram percebidos carbonetos do tipo M6C, M23C6 nos contornos de grão, foi aprovada nos ensaios de dobramento, líquidos penetrantes e imersão em FeCl3. Entretanto, ocorreu o um desgaste da ferramenta, mas reduzido e aceitável para esta solda. Tensões residuais longitudinais trativas da ordem de 50 MPa foram observadas na zona de mistura. Considerando o valor médio, o MB alcançou uma CPT de 77 °C, enquanto no centro da solda a CPT foi de 86°C. A taxa de corrosão média alcançada após o ensaio de corrosão intergranular foi de 0,4406 mm/ano, valor que sugere uma boa qualidade da solda. / Nickel based alloys are often used in the oil and gas industry due to their great mechanical strenght as well as excelent corrosion resistance. In this scenario, welding as manufacturing step for long lenght rigid tubes is inevitable. Therefore, Inconel 625® sheets were joined by Friction Stir Welding (FSW). The welding was performed with a Polycrystalline Cubic Boron Nitride (pcBN) tool and the unavoidable tool wear was verified. At first, in the process parameter development step, different process conditions were analyzed for choosing the best welded joint. Afterwards, new welded joints were produced with the suitable process parameter (tool rotational speed of 200 rpm and welding speed of 1 mm/s). Top surface characterization of friction-stir-welded Inconel 625 was carried out by optical microscopy, scanning electron microscopy, microhardness and bending tests. Surface defects were verified by dye penetrant inspection (DPI). In addition, residual stress states were obtained by X-ray diffraction. Furthermore, localized corrosion resistance was studied by the immersion test in FeCl3 and the critical pitting temperature (CPT) determination. Finally, the susceptibility to intergranular corrosion was evaluated by Streicher Test. FSW process promoted grain refining as well as increased microhardness in the joint Base material micrographs showed different carbides as MC, M6C, M23C6. On the other hand, the best joint presented a sounds weld without defetcs, homogeneous microstructure, grain boundaries free of M6C, M23C6 carbides, approved in bending tests, DPI and the immersion test. However, a tool wear was also noted, but reduced and acceptable for this weld. Tensile longitudinal residual stresses of 50 MPa were observed in the stir zone. As an avarage, the base material CPT was 77 °C, while the CPT was 86 °C at the weld center. The mean corrosion rate after intergranular corrosion test was 0.4406 mm/year, a value which suggests a good weld quality.

Stationary (static) shoulder friction stir welding (SSFSW) is a variant of conventional friction stir welding (FSW) that was originally invented to improve the quality of welds produced with titanium alloys. Its predominant advantage is a reduction of the severe through thickness temperature gradients seen in conventional FSW, when welding low thermal conductivity alloy. However, SSFSW has rarely been utilised as a method to weld aluminium alloys because it is generally thought that in conventional FSW the rotating shoulder plays an essential role in the heat generation and, due to the high thermal conductivity of aluminium alloys, a rotating shoulder is beneficial for the welding process. In the work presented, the advantages of SSFSW have been examined when welding a typical high strength aluminium alloy AA7050-T7651. The process window for each approach has first been determined, and the optimum welding conditions were systematically evaluated, using power-rotation rate curves. Direct comparison of the two processes was subsequently carried out under these optimum conditions. It has been demonstrated that SSFSW can dramatically improve the quality of a weld's surface finish. Under optimum conditions it has also been shown that SSFSW was able to weld with approximately a 30% lower heat input than FSW and the stationary shoulder led to a narrower heat affected zone (HAZ). As a result, the through thickness properties of SSFSW were much better and more homogeneous than that for FSW, in terms of grain sizes, hardness and cross-weld mechanical properties. Uniaxial tensile tests proved that the average tensile strength of SSFSW samples was around 500 MPa, which was about 100 MPa larger than that of the FSW sample. Also, it was shown that during tensile testing the deformation zones, which correspond to minima in the hardness distribution of SSFSW welds, were about half the size of those found in FSW welds under the same traverse speed. The mechanisms that give rise to these advantages have been investigated systematically, focusing on directly comparing the SSFSW and FSW processes, and are discussed aided by finite element modelling (FEM) of the heat distribution in welds produced by each process and microstructural investigations.

In this project, we have successfully joined MA956 Oxide Dispersion-Strengthened (ODS) steel plates using Friction Stir Welding (FSW). ODS steels are prime candidate materials for the fuel cladding in Generation IV nuclear fission reactors and as first wall components in nuclear fusion reactors. This is due to their exhibiting excellent high temperature strength and creep behaviour, together with enhanced resistance to radiationinduced void swelling. ODS steels are heavily reliant on a fine dispersion of (Y-Al-O) nanooxide particles to provide the aforementioned properties. This, however, makes ODS steels particularly problematic to join. Most joining techniques melt the material along the joint line, but this would severely alter or deplete the nano-oxide dispersion and hence be highly detrimental to the material’s performance in a nuclear environment. FSW is a solid-state joining technique, and therefore can join ODS steel without melting the material. Although FSW can potentially alter the microstructure of the base material and affect the distribution of nano-oxide particles, if a sufficient number of nano-sized particles and a sufficiently homogeneous dispersion remain after the welding process, then a major roadblock for the implementation of ODS steels will have been removed. The research of this thesis focused on the impact of FSW on: i) the microstructure, ii) the mechanical properties, iii) the residual stresses, and iv) the abnormal grain growth behaviour of ODS steels; utilizing a wide array of techniques to assess the micro-to-nano scale structure and the properties of the base material and welds, including optical, scanning and transmission and electron microscopy, X-ray and neutron diffraction, small-angle neutron scattering, tensile testing and micro-hardness measurements. We also produced welds with systematic changes to the tool traverse speed and rotation speed to investigate the impact of changing the welding parameters on the weld microstructure, and therefore optimise the process parameters for enhanced radiation and mechanical performance of the ODS steel welds.

Orientador: Carlos Kiyan / Banca: Messias Borges da Silva / Banca: Luis Rogério de Oliveira Hein / Banca: Olivério Moreira de Macedo Silva / Resumo: O estudo do processamento de compósitos de matriz metálica pelo método de fundição com agitação é de grande interesse em engenharia pelo seu baixo custo e possibilidade de utilizar equipamentos convencionais de fundição. Nesse trabalho é feito o estudo do processo de síntese de compósitos da matriz da liga A356 reforçada com partículas de alumina, focalizando a influência dos fatores de processo na fração de partícula incorporada. Foram sintetizados lingotes de compósitos variando as condições de processo. Foram desenvolvidos métodos específicos de amostragem estatística associados à análise e processamento de imagens e utilizados na obtenção de estimativas confiáveis da fração volumétrica. Para investigar a influência dos fatores e suas interações foi projetado um experimento fatorial com 4 fatores, definidos em ensaios preliminares. Os resultados confirmam achados teóricos e experimentais anteriores e indicam novos caminhos para se obter uma melhor incorporação de partículas, além de um entendimento do mecanismo da molhabilidade da liga no estado semissólido. Compósitos com maior grau de partículas incorporadas foram examinados por microscópio eletrônico de varredura e microscópio óptico, mostrando que a partícula se aloja preferencialmente na região interdendrítica da matriz. / Abstract: The study of metal matrix composite processing through stir casting method is of great interest for engineering because of its low cost and possibility of using conventional casting equipment as well. The synthesis of hypoeutectic A356 matrix alloy reinforced with alumina particle was done in this work focusing the influence of the processing factors on the incorporated particle fraction. Ingots of composites were synthesized by varying the process conditions. Methods of statistic sampling connected to the analysis and processing images were developed and used for reliable estimate of volumetric fraction. To investigate the influence of the factors and their interactions, a factorial experiment was planned with four factors defined in a preliminary test. The results confirm theorical and experimental findings and points to a new way of getting better particle incorporation and an insight of the wettability mechanism. Composites with high level of particle incorporation were checked through electronic and optical microscopes showing that the particle stays mainly in interdendritic regions. / Doutor

Neste trabalho é estudada a influência de Friction Stir Welding (FSW) em um processo subsequente de anodização porosa, avaliando-se o impacto da microestrutura produzida por FSW na espessura, na estrutura e na resistência à corrosão de óxidos anódicos. Um perfil da liga AA7050 foi soldado por FSW a um chapa da liga AA2024, simulando a junta sobreposta entre reforçadores e revestimento na fuselagem de um avião. Como uma primeira aproximação, a superfície da liga AA7050 em contato com a ferramenta de soldagem foi selecionada para a anodização, por ser o lado que sofre maior modificação microestrutural. A solda foi caracterizada por técnicas de microscopia ótica e eletrônica, distinguindo-se três zonas: zona afetada pelo calor (ZAC), zona termomecanicamente afetada (ZTMA) e nugget (onde ocorre recristalização dinâmica). Medições de dureza Vickers revelaram uma menor dureza na fronteira entre ZTMA e ZAC, provavelmente devido à dissolução de precipitados. Amostras da solda foram anodizadas galvanostaticamente em eletrólito sulfúrico, sendo algumas seladas em água deionizada fervente. A camada de óxido produzida sobre a ZTMA é ainda mais espessa do que sobre o material base e apresenta uma interface plana com o substrato, enquanto no nugget a camada de óxido é mais fina e tem interface rugosa. Infere-se que a presença de precipitados endurecedores (como MgZn2) afeta a espessura dos filmes de óxido. Defeitos relacionados à dissolução de precipitados ricos em Cu são encontrados nos filmes de óxido em todas as zonas. A técnica de varredura com eletrodo vibratório (SVET) foi utilizada para testar a resistência à corrosão dos filmes de óxido em eletrólito contendo NaCl, com polarização simultânea acima do potencial de pite. Ocorre corrosão por pite preferencialmente no nugget e em sua fronteitra com a ZTMA. Por fim, transientes de corrente são calculados para cada pite a partir dos mapas de SVET. Conclui-se que as soldas FSW anodizadas apresentam região localizada de menor resistência à corrosão pelo crescimento heterogêneo da camada anódica, mas alta resistência à corrosão é atingida após selagem. / In this work, the influence of Friction Stir Welding (FSW) on a posterior porous anodizing process is studied, with emphasis on how the friction stir weld’s microstructure affects thickness, structure and corrosion resistance of anodic oxide films. An AA7050 profile was friction stir welded to an AA2024 sheet in order to simulate the assembly of airplane stringers and skin. As a first approach, the AA7050 surface facing the welding tool was selected for anodizing, for it is the side that is most modified. The weld was characterized by optical and electron microscopy techniques, and 3 different zones were distinguished: heat affected zone, thermomecanically affected zone (TMAZ), and the nugget (where dynamic recrystallization occurs). Vickers hardness measurements revealed a lower hardness at the frontier between thermomecanically and heat affected zones, possibly due to dissolution of precipitates. Welded samples were galvanostatically anodized in sulfuric electrolyte and some of them were sealed in boiling deionized water. Micrographs showed that on the TMAZ the anodic oxide is thicker than on the base material and the metal/oxide interface is flat. On the nugget zone, however, the anodic oxide is thinner and more defective, and the metal/oxide interface is rougher. It is inferred that the presence of hardening precipitates (such as MgZn2) affects the thickness of oxide films. Defects related to the dissolution of Cu-rich precipitates were found in the oxide layers on all zones and inferences about how they affect anodization kinetics are drawn. The Scanning Vibrating Electrode Technique (SVET) was used to test the corrosion resistance of anodic oxide layers in NaCl electrolyte. The samples were scanned simultaneously to anodic polarization at a potential higher than the AA7050 alloy’s pitting potential. Results showed that pitting happened preferentially on the nugget zone and its frontier with the TMAZ. Finally, pit current transients were calculated from SVET maps. It can be concluded that anodized friction stir welds present a region of diminished corrosion resistance caused by heterogeneous anodic oxide growth, but high corrosion resistance can be achieved after sealing.

The objective of this research was to evaluate the capabilities of ALEGRA, a Sandia National Labs hydrocode, and Forge3, a Transvalor S.A. product, to accurately model the Friction Stir Welding Process. ALEGRA and Forge3 are discussed in light of the inherent challenges of modeling Friction Stir Welding, and a rotational boundary condition is added to ALEGRA. Results are presented from Friction Stir Welding modeling outputs from both ALEGRA and Forge3. ALEGRA is shown to be incapable of modeling the Friction Stir Welding process, in large part due to its focus on shock propagation, which causes extremely small time steps. Forge3 is shown capable of modeling of the FSW plunge process in a transient manner, but overestimates the temperature profiles 90% to 100% in comparison to experimentally measured values. It appears that the adiabatic boundary condition is the source of much of the error. It is recommended that future work focus on improving estimates of the boundary conditions utilized in the Forge3 model.

To evaluate potential uses for friction stir welding in additive manufacturing, two separate parts were fabricated, one of 2195-T84 and the other 2219-T87, utilizing fixed pin techniques and additive lap welds. The parts were cut into samples, artificially aged and subjected to Rockwell hardness (HRB), Vickers hardness, micrographic photography, and metallographic imaging on both pre- and post- heat treatment. Additionally, tensile testing was performed on the heat-treated samples. A comparisons of test results showed a minimal increase in the yield strength of the 2195-T84 samples compared to as-welded tensile results obtained from a previous project. The ultimate tensile strength was reduced by approximately 16%. Further testing will be required to determine the nature of this reduction. No previous results were available for the as-welded 2219-T87, but UTS of the artificially aged samples was approximately 91% that of the parent material.