Material property heterogeneity is present ubiquitously in various natural and man-made materials, such as bones, seashells, rocks, concrete, composites, and functionally graded materials. A fundamental understanding of the structure-property relationships in these material systems is crucial for the development of advanced materials with extreme properties. Well-developed homogenization schemes exist to establish such relationships in elasticity, electrostatics, magnetism, and other time- or history-independent material properties. Nevertheless, one’s understanding of the effective fracture properties of heterogeneous media is remarkably limited. The challenge here is that heterogeneous fracture, as a history-dependent process, involves complex interaction and negotiation of a discontinuity front with local heterogeneities. The determination of effective fracture properties necessitates a critical interrogation of this evolutionary process in detail. In this work, a combined experimental and modeling effort is made to examine and control fracture mechanisms in heterogeneous elastic solids. A two-phase laminated composite, which mimics the key microstructural features of many tough biological materials, is selected as a model material. In the computational part of this work, finite element analysis with cohesive zone modeling is used to model crack propagation and arrest in the laminated direction. A crack-tip-opening controlled algorithm is implemented to overcome the instability problems associated with inherently unstable crack growth. Computational results indicate that the mismatch of elastic modulus is an important factor in determining the fracture behaviors of the heterogeneous model material. Significant enhancement in the material’s effective fracture toughness can be achieved with appropriate modulus mismatch. Systematic parametric studies are also performed to investigate the effects of various material and geometrical parameters, including modulus mismatch ratio, phase volume fractions, T-stress, and cohesive zone size. Concurrently, a novel stereolithography-based additive manufacturing system is developed and used for fabricating heterogeneous test specimens with well-controlled structural and material properties. Fracture testing of each specimen is performed using the tapered double-cantilever beam (TDCB) test method. With optimized material and geometrical parameters, heterogeneous TDCB specimens are found to exhibit higher fracture toughness than their homogenous counterparts, which is in good agreement with the computational predictions. The integrative computational and experimental study presented here provides a fundamental mechanistic understanding of the fracture mechanisms in brittle heterogeneous materials and sheds light on the rational design of ultra-tough materials through patterned heterogeneities.

Topology optimization tools are useful for distributing material in a geometric domain to match targets for mass, displacement, structural stiffness, and other characteristics as closely as possible. Topology optimization tools are especially applicable to additive manufacturing applications, which provide nearly unlimited freedom for customizing the internal and external architecture of a part. Existing topology optimization tools, however, do not take full advantage of the capabilities of additive manufacturing. Prominent tools use micro- or meso-scale voids or artificial materials to parameterize the topology optimization problem, but they use filters, penalization functions, and other schemes to force convergence to regions of fully dense (solid) material and fully void (open) space in the final structure as a means of accommodating conventional manufacturing processes. Since additive manufacturing processes are capable of fabricating intermediate densities (e.g., via porous mesostructures), significant performance advantages could be achieved by preserving and exploiting those features during the topology optimization process. Towards this goal, a topology optimization tool has been created by combining homogenization with parametric smoothing functions. Rectangular mesoscale voids are used to represent material topology. Homogenization is used to analyze its properties. B-spline based parametric smoothing functions are used to control the size of the voids throughout the design domain, thereby smoothing the topology and reducing the number of required design variables relative to homogenization-based approaches. Resulting designs are fabricated with selective laser sintering technology, and their geometric and elastic properties are evaluated experimentally. / text

Additive Manufacturing (AM) processes which include Selective Laser Sintering (SLS) have experienced tremendous growth and development since their introduction over 20 years ago. It becomes highly important at this stage to evaluate the sustainability of the process and refine it to reduce energy and material consumption. In this study, a sustainability analysis was performed on the SLS process with Nylon-12 using the Environmental and Resource Management Data (ERMD) known as Eco-Indicators. The energy perspective alone was considered and a Total Energy Indicator (TEI) value was calculated using various parameters to quantify process sustainability: process productivity, energy consumption rate, etc. Precise thermal control of selective laser sintering (SLS) is desirable for improving geometric accuracy, mechanical properties, and surface finish of parts produced. An experimental setup to monitor the temperature distribution was designed using Resistance Temperature Detectors (RTD) as a part of this study. Discrepancies in temperature profiles were investigated and recommendations were made to improve thermal characteristics of the SLS process. / text

Ink-jet deposition is an emerging technology that provides a more efficient, economic, scalable method of manufacturing than other traditional additive techniques by laying down droplets layer by layer to build up 3-D objects. The focus of this thesis is to investigate the material interface evolution during the droplet deposition process, which holds the key to understanding the material joining process. Droplet deposition is a complicated process and can be broken down into droplet impingement dynamics and droplet hardening. This research focuses on the study of the interface dynamics of droplet impingement. In order to study the interface dynamics, a novel metric is developed to quantify the evolving geometry of the droplet interface in both 2-D and 3-D for single and multiple droplets respectively, by measuring the similarity between the evolving droplet geometry and a desired shape. With the developed shape metric, the underlying physics of the interface evolution for single droplet impingement are examined with simulations using an experimentally validated numerical model. Results show that the Weber number determines the best achievable shape and its timing during the droplet impingement when Ohnesorge number is smaller than 1, while the Reynolds number is the determining factor when Ohnesorge number is larger than 1. A regime map is constructed with the results and an empirical splash criterion to guide the choice of process parameters for given fluid properties in order to achieve the best shape without splash for single droplet impingement. In order to study the interface dynamics for multiple droplet interaction, which is computationally prohibitive for commercial software packages, an efficient numerical model is developed based on the Lattice Boltzmann (LB) method. A new LB formulation equivalent to the phase-field model is developed with consistent boundary conditions through a multiscale analysis. The numerical model is validated by comparing its simulation results with that of commercial software COMSOL and experimental data. Results show our LB model not only has significant improvement of computational speed over COMSOL but is also more accurate. Finally, the developed numerical solver is used to study the interface evolution of multiple droplet interaction with the aid of the 3-D shape metric proposed before. Simulations are performed on a wide range of impingement conditions for two-droplet, a-line-of-droplet, and an-array-of-droplet interactions. The underlying physics of the interface coalescence and breakup coupling with the impingement dynamics are examined. For line-droplet interaction, the strategy for achieving the equilibrium shape in the shortest time is studied. An important issue is discovered for array-droplet interaction, which is the air bubble formation during the droplet interaction. The mechanism for the air bubble formation is investigated and the strategy to avoid this undesirable effect is also suggested. This thesis has largely reduced the gap between basic science of studying droplet impingement dynamics and engineering application in inkjet deposition and provided preliminary insights on the material joining process for additive manufacturing.

Additive manufacturing (AM) offers reduced material waste and energy usage, as well as an increase in precision. Direct metal AM is used not only for prototyping, but to produce final production parts in the aerospace, medical, automotive and other industries. Process mapping is an approach that represents process outcomes in terms of process input variables. Solidification microstructure process maps are developed for single bead and thin wall deposits of Ti-6Al-4V via an electron beam wire feed and electron beam powder bed AM process. Process variable combinations yielding constant beta grain size and morphology are identified. Comparison with the process maps for melt pool geometry shows that by maintaining a constant melt pool cross sectional area, a constant grain size will also be achieved. Additionally, the grain morphology boundaries are similar to curves of constant melt pool aspect ratio. Experimental results are presented to support the numerical predictions and identify a proportional size scaling between beta grain widths and melt pool widths. Results demonstrate that in situ, indirect control of solidification microstructure is possible through direct melt pool dimension control. The ability to control solidification microstructure can greatly accelerate AM process qualification potentially allow for tailored microstructure to the desired application.

An orthosis, or orthotic device is used to straighten or correct the posture of part of the body. A foot orthosis (FO) is the subject of study for this dissertation. A FO is situated between the foot and the midsole of the shoe and replaces the insole. Foot orthoses (FOs) are intended to prevent or aid in the recovery of injury by acting to redistribute pressure experienced by the plantar surface of the foot as well as cause adjustments to the relative positions of the foot's bones during standing and gait. Traditional methods for developing a FO require extensive skilled manual labour and are highly dependent on subjective input. Modern FO development methods have sought to address these issues through the use of computer driven technological advancements. Foot scanners record geometry, computer aided design (CAD) software is used to develop the FO geometry, and automated manufacturing tools are used to either fabricate the FO or fabricate a mould about which the FO can be formed. A variety of modern solutions have successfully automated the process, however, it remains highly subjective. Skilled manual labour has merely been replaced with equally subjective skilled computer labour. In particular, adjustments to the foot are made with basic deformation functions to the static surface foot models generated by modern digitizers. To improve upon this, a model that describes the mechanics and properties of the various tissues of the foot is required. Such a model will also be useful for validating and optimizing FO designs prior to fabrication through simulation of weight-bearing conditions. Given the deformable characteristics of the tissues of the foot, the finite element (FE) modeling method is appropriate. The FE foot model has become a common medical and engineering tool in recent years. Its application, however, has primarily been limited to research as few clinical applications warrant the development cost. High cost stems from the MRI or CT scan and the skilled labour required to assemble the model for FE analysis. Consequently, the FE modeling approach has previously been out of reach for the application of FO development. The solution proposed and implemented was to map a detailed generic FE foot model to an inexpensive surface scan obtained from a modern digitizer. The mapping accurately predicted anatomical geometry and resulted in simulation models that can be used in the FO development process first to carry out postural adjustments prescribed by a practitioner and second in a validation step where a FO design can be tested prior to fabrication. In addition to simulation tools, novel complementary tools were developed for designing and fabricating FOs. The simulation, design, and fabrication tools were incorporated into a novel, seven step FO development process. The proposed process is beneficial to FO development as it reduces the required subjective input from practitioners and lab technicians and allows for the validation of potential FO designs prior to fabrication. Future work is required to improve computational efficiency of the FE foot models and to fully automate the process to make it commercially viable. In addition to FOs, the proposed approach also presents opportunities for improving other orthoses and prostheses for the human body.

Orientador: Ruís Camargo Tokimatsu / Resumo: Esse trabalho tem a finalidade de projetar, com base em levantamento bibliográfico, uma prótese de disco intervertebral de Ti6Al4V pelo processo DMLS. Os sistemas de imagem de tomografias computadorizadas e ressonâncias magnéticas trabalham de maneira semelhante. Imagens em 3D de dados dos pacientes são construídas através da combinação de fatias 2D tomadas a partir de sistemas de sensores e da interpolação entre eles. A distância entre as fatias desses sistemas de imagens é grosseira comparada com a tecnologia da manufatura aditiva, portanto, esses processos combinam as imagens das fatias dentro de um software para criar um modelo 3D sobrepondo camadas, devido a essa sinergia entre os dois tipos de sistema, pesquisadores de todo o mundo tentam melhorar e inventar próteses para substituir o disco intervertebral através da ciência. Os resultados mostraram que a fabricação de uma prótese de disco com junta esférica pelo processo de manufatura aditiva é viável, porém nos segmentos menores, como na cervical ainda são difíceis de produzir devido à altura do disco intervertebral e ao limite da dimensão das camadas da manufatura aditiva, que atualmente estão por volta dos 20μm, por isso requerem tratamentos superficiais após sua fabricação. Espera-se que este trabalho possa ajudar futuros pesquisadores da área de biomecânica da coluna vertebral. / Abstract: This work has the purpose of designing, based on a bibliographical survey, an intervertebral disc prosthesis of Ti6Al4V by the DMLS process. Computed tomography (CT) imaging systems and MRIs work similarly. 3D images of patients data are constructed by combining 2D slices taken from sensor systems and interpolating between them. The distance between the slices of these imaging systems is rough compared to the additive manufacturing technology, so these processes combine the slice images within a software to create a 3D model overlapping layers because of this synergy between the two system types, researchers around the world try to improve and devise prostheses to replace the intervertebral disc through science. The results showed that the manufacture of a disc prosthesis with spherical joint by the additive manufacturing process is feasible, but in the smaller segments, such as in the cervical still difficult to produce due to the height of the intervertebral disc and to the limit of the size of the layers of the manufacture additives, which are currently around 20μm, so they require surface treatments after their manufacture. It is hoped that this work may help future researchers in the area of spinal biomechanics. / Mestre

La prise en compte des impacts environnementaux liés à la Fabrication Additive (FA) est un sujet encore en développement et loin d'être consolidé. Cette thèse tente de répondre à la question suivante: Comment identifier et maîtriser les impacts environnementaux liés à la mise en usage des technologies de Fabrication Additive ?Une approche centrée sur l’analyse de scénarios d’usages a été appliquée. Trois scénarios ont été identifiés et étudiés: 1) La Fabrication Personnelle; 2) l’Utilisation commerciale de la FA et 3) l’Utilisation Industrielle de la FA. L'analyse de la littérature et l’application de la méthode d’Analyse du Cycle de Vie – ACV ont fourni les données nécessaires afin d’identifier les différents hotspots environnementaux dans chaque scénario.Pour le cas de la « Fabrication Personnelle », l'étude souligne l'importance de mettre en évidence les aspects humains. La façon dont les gens utilisent la technologie génère bien souvent plus d’impact que la technologie elle-même. Pour le cas du scénario «Utilisation Commerciale», c’est la phase de production qui génère le plus d'impacts environnementaux, en raison de la consommation d’énergie du processus d'impression. Les résultats du scénario « Utilisation Industriel», avec l’utilisation de la technologie EBM, indiquent que la consommation d’énergie du processus d'impression, la fabrication de poudre, ainsi que les processus de Finition sont les principales sources d’impacts.Les résultats des trois études ont été analysés et rassemblés de manière à concevoir un cadre systémique montrant les contributeurs aux impacts dans le cas de la mise en usage de technologies de FA. Quatre groupes de contributeurs ont ainsi été identifiés en lien avec: le produit, le processus d'impression (aspects techniques), et le niveau d'expérience de l'utilisateur en matière de CAO et d'impression (aspects humains). Pour chaque groupe, plusieurs contributeurs ont été associés (consommation d’énergie, masse du produit, temps d’usage de la CAO,…). A partir de ce cadre, nous avons pu construire un modèle simplifié pour l’ACV adapté au cas de la mise en usage de la FA. Il a donné lieu à l’élaboration d’un démonstrateur pour estimer les impacts environnementaux liés à la mise en usage de la fabrication additive. / The subject of the environmental impacts related to Additive Manufacturing is evolving and far from being consolidated. This thesis tries to answer the following question: How to identify and control the environmental impacts associated to the use of Additive Manufacturing technologies? An approach centred on scenarios was applied. Three specific scenarios were identified and studied: 1) Personal Fabrication; 2) Business use of AM and 3) Industrial use of AM. Literature analysis and the use of the Life Cycle Assessment-LCA approach provided accurate data to identify the different hotspots of environmental impacts in each scenario.In the Personal Fabrication case, the study highlights the importance of placing a greater focus on the Human aspects. The way people use the technology often generates more impacts than the technology itself. In the ‘Business’ scenario the findings show that the energy consumed in Production phase is the most representative phase in terms of environmental impacts. Results from the ‘Industrial’ scenario using EBM technology suggests that the energy consumption of the Printing Process, the powder manufacturing, as well as the Post-treatment process are the main sources of the environmental impacts.The results of the three studies were analysed and gathered in such a way to design a systemic framework of impacts in the case of AM use. Four group of impacts drivers are recognised: The product and the Printing Process (Technical aspects) and the User’s level of experience in CAD and Printing (Human Aspects). For each group, several impact contributors are identified (product height, CAD use time, etc.). From this framework, an adapted LCA model was designed and a software concept was created to estimate the environmental impacts related to use of AM technologies.

Il est maintenant possible de fabriquer des structures treillis métalliques facilement avec la fabrication additive. Les structures en treillis peuvent être utilisées pour produire des pièces de faible masse et de haute résistance. Il n’existe pas de méthode de conception pour les structures treillis. Cette thèse se concentre sur les méthodes de conception des structures treillis et la manipulation dans le CAO et FAO pour faciliter l'intégration des structures treillis dans les produits. La thèse a abordé les questions de recherche suivantes:• Pourquoi les structures treillis sont-elles si peu utilisées dans la conception?• Quelles sont les informations nécessaires pour aider les concepteurs à concevoir des pièces contenant des structures treillis?• Comment les structures treillis peuvent-elles être créées rapidement et facilement dans le CAO?Les principales contributions sont les suivantes:• Une évaluation des outils CAO actuels dans la conception de structures en treillis en termes d'interface homme machine, de formats de fichiers CAO et de FAO pour la fabrication d'additive a été effectuée. Les résultats montrent que les outils de CAO et les formats de fichier CAO actuels ont des performances insuffisantes dans le contexte de la conception pour la fabrication d'additive. Les outils de CAO actuels créent et représentent actuellement des structures en treillis utilisant les surfaces limites des volumes. Cela contribue ainsi à la grande taille des fichiers, à une consommation élevée de mémoire vivre, ainsi des opérations fastidieuses pour les modélisations.• Une nouvelle stratégie de conception de structures treillis. Cette méthode sert de guide aux concepteurs pour l'intégration des structures en treillis dans les pièces fabriquées par fabrication additive en utilisant le matériau équivalent. Les concepteurs auront à leur disposition les informations nécessaires pour choisir les types et la densité des structure treillis à utiliser.• Une méthodologie pour calculer les propriétés matériau équivalent. Ces matériaux équivalents remplacent le besoin de créer des structures treillis dans le CAO et de les calculer par éléments-finis. Cela permettra d'économiser du temps dans la création de modèles CAO 3D et les calculs éléments finis.• Les principales caractéristiques géométriques des structures treillis ont été déterminées. Un modèle squelettique a été présenté pour définir les structures treillis à partir de points, de lignes, de sections et de joints au lieu des surfaces et des volumes. Une méthode est présentée pour visualiser et découper les structures treillis à partir du modèle squelette. / It is now possible to manufacture metallic lattice structures easily with additive manufacturing. Lattice structures can be used to produce high strength low mass parts. However, it does not exist a method to design lattice structures for additive manufacturing. This PhD focuses on lattice structure design methods and manipulation in CAD, CAE and CAM tools to facilitate the wide use of lattice structures in products. The thesis addressed the following research questions:• Why are lattice structures so little used in part designs?• What are the information necessary to help designers to design parts containing lattice structures?• How can lattice structures be created quickly and easily in CAD?The main contributions are:• An evaluation of current CAD tools in terms of human machine interface, CAD file formats, CAE and CAM to design lattice structures was conducted. The results show that current CAD tools and CAD file formats have insufficient performance in the context of design for additive manufacturing. Current CAD tools create and represent lattice structures using surfaces and volumes. This contributes to large file sizes, high RAM consumption, as well as time-consuming creations and operations.• A new lattice structure design strategy. This method serves as a guideline for designers to integrate lattice structures in additive manufactured parts using the concept of equivalent material. Designers will be able to choose lattice structure patterns and densities.• A methodology to create equivalent materials is presented. It is solid and does not contain any struts, thus has few surfaces only. With this equivalent material, it will be easier and quicker to conduct FEA due to the small number of surfaces involved. The characteristics of different lattice structure patterns and densities were determined, which are the relative Young’s modulus and relative strength in function of the relative density. This methodology can be applied to all lattice structures.• The main lattice structure geometrical characteristics were determined. A skeleton model was presented to define lattice structures with points, lines, sections and joints instead of surfaces and volumes. A method is presented to visualise in CAD and slice lattice structures in CAM from the skeleton model.

No description available.