Committee:
     PRESIDENT: CALVO MUÑOZ, JESICA
     SECRETARI: SUÑOL MARTÍNEZ, JUAN JOSÉ
     VOCAL: AGUILAR, CLAUDIO
Thesis abstract: This doctoral research evaluated the feasibility of using various aluminum-based feedstocks in additive manufacturing (AM) to develop cost-effective and environmentally friendly alternatives to traditional metal fabrication. The study systematically examined AA6061 filament, AlSi10Mg granules (commercial and recycled), and AlSi10Mg powder paste across three AM techniques: Fused Deposition Modeling (FDM), screw-based extrusion, and Direct Ink Writing (DIW). The main objectives were to optimize printing, thermal debinding, and sintering parameters for each feedstock and AM technique, and to assess the resulting mechanical properties and microstructures of the fabricated parts.For printing, AA6061 filament processed via FDM achieved optimal results with a 0.8 mm nozzle diameter at 205 °C. AlSi10Mg granules (commercial and recycled) and AlSi10Mg powder paste, used in screw-based extrusion and DIW respectively, performed best with 0.6 mm nozzles and lower temperatures. These optimizations established critical baseline conditions for subsequent processing steps, emphasizing the distinct requirements of each material and technique.Thermal debinding, essential for removing polymeric binders before sintering, was optimized for each feedstock. For AA6061 filament, 550 °C with holding times up to 3 hours was most effective. For commercial AlSi10Mg granules, 350 °C for 3 hours yielded optimal results, a condition that also worked for recycled granules and powder paste. These parameters minimized defects and prepared the parts for successful sintering.Sintering parameters were rigorously optimized to ensure densification and desired mechanical properties. AA6061 filament was best sintered at 635 °C, while commercial AlSi10Mg granules and powder paste achieved optimal results at 600 °C. Recycled AlSi10Mg granules reached peak performance at 620 °C. All sintering was conducted for 3 hours under a nitrogen atmosphere with vacuum and oxygen traps. SEM analysis confirmed increased densification and uniform microstructures under these conditions.A pre-sintering pressing technique was introduced to further enhance densification and reduce porosity. This step significantly improved the relative density of sintered parts by 19.25–45.55%, with pressed samples achieving densities up to 93.65%. Mechanical testing showed that recycled AlSi10Mg granules provided the highest compressive strength (168.34 MPa), followed by commercial granules, AA6061 filament, and powder paste.

Committee:
     PRESIDENT: TYLKOWSKI, BARTOSZ
     SECRETARI: DEL VALLE MENDOZA, LUIS JAVIER
     VOCAL: CABRERA, MIGUEL
Thesis abstract: In-Mold Electronic technology enables the creation of three-dimensional shaped electronic surfaces by combining printed electronics with In-Mold Decoration. This innovative approach enables the integration of electronic circuits into a wide range of complex and diverse applications, going from automotive interiors to consumer electronics. The manufacturing process for IME devices demands that all materials involved, including adhesives, endure the rigorous transformation conditions to ensure functionality and reliability of the final device. Adhesives, while critical for securing surface-mounted components and maintaining circuit integrity, possess properties that can significantly influence film deformation during the thermoforming process. These deformations can have a great impact on the overall performance and durability of the device. This research is dedicated to thoroughly examining the effects of the structural adhesive on film deformation during the thermoforming process, as well as the subsequent impact on device functionality. The study meticulously investigates how varying quantities and disposition placements of structural adhesives affect the film’s stretching behaviour during both thermoforming process and injection molding, comparing printed and hybridized samples. The research findings reveal that hybridized samples, those with structural adhesive, exhibit deformation patterns distinct from printed samples. Specifically, areas around the adhesive show reduced stretching, while regions farther from the adhesive compensate with increased stretching. This compensation often leads to overstretching in certain areas, which, in some cases, results in circuit discontinuities and compromised electrical performance. Moreover, the variation in stretching not only affects the circuit integrity but also alters the final positioning of electronic components. This misalignment can cause components to shift from their intended locations, potentially leading to device malfunctions or failures. To address these challenges, the study recommends adjusting the amount and placement of structural adhesive based on localized deformation patterns. By tailoring the adhesive application, unnecessary stretching can be minimized, ensuring both the mechanical and electrical integrity of the device. Additionally, further analysis during the injection molding process highlighted that careful management of adhesive quantity effectively secures all components in their designated positions without compromising functionality. The insights gained from this research are critical for the design and manufacturing of IME devices. They provide guidelines for optimizing adhesive placement and quantity, ultimately enhancing the reliability, performance, and yield of functional devices in high-volume production.

Committee:
     PRESIDENT: CASELLAS PADRO, DANIEL
     SECRETARI: FARGAS RIBAS, GEMMA
     VOCAL: MOSELEY, STEVEN
Thesis abstract: This PhD thesis explores the microstructure–mechanical property relationships of WC-Co cemented carbides (hardmetals) produced using two sinter-based additive manufacturing (AM) techniques: Binder Jetting Technology (BJT) and Direct Ink Writing (DIW). While AM offers a promising route for fabricating high-performance composites, existing research mainly focuses on optimizing processing parameters, with limited understanding of how AM affects microstructure and mechanical performance. This thesis addresses this gap to support the industrial use of AMed hardmetals.The study systematically examines mechanical behaviour—specifically hardness, sliding contact response, and fracture resistance—of WC-Co composites shaped via BJT and DIW, then consolidated by sintering. Special attention is given to how the layer-wise building strategy and layer orientation influence performance. The work also considers the multiscale nature of these materials, with structural features spanning nanometres to millimetres.A key finding is the confirmation of microstructural homogeneity across all samples, independent of AM route or printing orientation. This isotropy is reflected in uniform hardness, wear resistance, and fracture toughness (KIc). The study validates the Single Edge micro-V-Notch Beam method for KIc measurement, using ultra-short pulse laser ablation for micro-notching. The lack of anisotropy indicates that liquid-phase sintering, common to both AM techniques, effectively homogenizes the microstructure.Hardness testing under different loads produced values comparable to conventionally made WC-Co grades, confirming AM’s viability. A noticeable length-scale effect emerged in microstructures with non-uniform carbide distributions: hardness variation increased under lower loads, highlighting the importance of scale in testing AM parts. High-speed nanoindentation mapping, combined with statistical analysis, enabled phase-level mechanical characterization and demonstrated the technique’s value as a mechanical microscopy tool.Scratch testing showed that AMed hardmetals achieve wear resistance similar to, or greater than, conventionally processed grades. Some BJT samples with bimodal carbide size distributions even outperformed reference grades. This enhanced performance is linked to synergistic wear mechanisms: coarse carbides contribute energy absorption, while finer carbides provide higher strength. Tribo-layer formation also varied between samples, influencing wear response.Unlike hardness and toughness, flexural strength was influenced by the orientation of layer stacking (though not by printing direction). This sensitivity is mainly due to AM-induced defects, such as surface waviness, interlayer misalignment, and cobalt content variation. The findings suggest that strength variability is driven by defects rather than by intrinsic structural anisotropy. Linear Elastic Fracture Mechanics was successfully applied to model fracture behaviour, with subsurface penny-shaped and surface semi-elliptical cracks observed and matched to experimental results.In summary, this work confirms that sinter-based AM techniques like BJT and DIW can produce WC-Co hardmetals with mechanical and microstructural properties comparable to those of conventionally processed materials. The thesis offers a detailed understanding of structure–property relationships and provides guidelines for the industrial application of AMed hardmetals. It also highlights the critical role of microstructural design and defect control in achieving reliable performance in advanced tool materials.

Committee:
     PRESIDENT: ENGIN VRANA, NIHAL
     SECRETARI: GODOY GALLARDO, MARIA
     VOCAL: REVERON CABOTTE, HELEN
Thesis abstract: Tetragonal zirconia polycrystals stabilized with 3 mol% yttria (3Y-TZP) has gained growing interest as an alternative to titanium for dental implants, owing to its excellent biocompatibility, high mechanical strength and corrosion resistance and superior aesthetics. Despite these advantages, the clinical performance of zirconia implants still depends on their ability to promote osseointegration while simultaneously minimizing the risk of bacterial colonization, a competitive process known as "race for the surface". Surface properties of dental implants, such as topography, chemistry, and wettability, critically influence the biological response at the tissue-implant interface. In particular, micro- and nanotopographies directly impact cell-material interaction and can modulate several cellular functions including adhesion, migration, proliferation, and differentiation. Similarly, these topographical features affect bacterial response, either promoting bacterial adhesion or, conversely, reducing colonization through antifouling or bactericidal effects. For this reason, surface modifications have become a widely explored strategy to enhance the biological performance of implants. Nevertheless, the major challenge lies in designing surfaces that simultaneously support osseointegration while also preventing bacterial adhesion.This PhD Thesis addressed this challenge by investigating different surface modification approaches to improve the biological performance of zirconia. The aim was to create topographies that simultaneously improve cell behavior while exhibiting antibacterial properties. In concrete, we developed and characterized a series of micro and nanostructured zirconia surfaces and evaluated their biological performance both in terms of human mesenchymal stem cell (hMSC) response and bacterial adhesion of different strains. Prior to experimental work, a comprehensive bibliographic review on topographical modification strategies for 3Y-TZP was conducted (Chapter I), highlighting existing knowledge gaps and guiding the selection of surface treatments. Following this, different surface modification techniques were employed, including hydrofluoric acid (HF) etching for generating nanotopography (Chapter II) and laser patterning via nanosecond (ns-) and femtosecond (fs-) laser to create defined microstructures (Chapter III). These techniques were also combined to evaluate a potential synergistic effect of hierarchically rough micro- and nanotopographies on the biological response (Annex I). Our findings demonstrate both chemical etching and laser patterning techniques successfully enhanced the biological performance of zirconia by improving the hMSCS behavior and reducing bacterial adhesion. However, their combination did not result in a synergistic improvement. Among all the surfaces, the 3 μm linear pattern (L3) created through fs-laser patterning offered the best balance by simultaneously enhancing hMSC adhesion, migration, and osteogenic differentiation, while significantly reducing the adhesion of Staphylococcus aureus and Pseudomonas aeruginosa bacteria. It also led to the most favorable biological outcome under competitive co-culture conditions. Furthermore, biofunctionalization of this topography using a multifunctional peptide containing both antimicrobial and cell-adhesive sequences showed promising synergistic biological effects (Annex II). Importantly, these improvements were achieved without compromising the mechanical integrity of the L3-patterned surface (Annex III). In conclusion, this PhD Thesis demonstrated that topographical modification of zirconia offers promising strategies for developing zirconia implant with improved biological performance, enhancing both osseointegration and antibacterial properties. Future directions should focus on integrating biochemical cues, in vivo validations, and complete assessment of the mechanical integrity.

Committee:
     PRESIDENT: TROMAS, CHRISTOPHE
     SECRETARI: RAMÍREZ SANDOVAL, GISELLE
     VOCAL: GAILLARD, YVES
Thesis abstract: This doctoral thesis focuses on the micromechanical characterization of multiphase materials through high-speed nanoindentation mapping (HSNM), combined with statistical and machine learning techniques. The aim is to extract and interpret mechanical properties with spatial resolution from large datasets, improving the understanding of microstructure-property relationships in complex systems such as ceramic-metal composites and heterogeneous steels.HSNM enables the acquisition of localized data with micrometric resolution over large areas, but its use presents several challenges: optimizing indentation spacing, interpreting scattered data, limitations of Gaussian distributions in representing micromechanical properties, and difficulties in classifying regions near interfaces. Moreover, there is growing interest in automating interpretation using machine learning.The objectives of this thesis include: (i) assessing industrially relevant materials with HSNM, (ii) applying unsupervised learning to quantify micromechanical transitions, (iii) introducing skewed distributions as alternatives to Gaussian fitting, and (iv) developing supervised models to classify nanoindentation responses based on the full curve shape.Methodologically, the thesis implements Gaussian Mixture Models (GMM) to cluster mechanical properties and identify phases in materials such as WC-Co and superduplex steels. This strategy allows for wide-area surface analysis and the detection of mechanical transitions, such as hardening gradients in advanced high-strength steels (AHSS) and property changes induced by electron beam melting (PBF-EB) in 316L/V4E alloys.To address asymmetric or dispersed data, Skew-normal distribution fitting is introduced, offering a more faithful representation of reality, especially in interface-influenced regions like hardmetals. This approach improves phase classification compared to traditional Gaussian fits.The thesis also develops a supervised model based on convolutional neural networks (CNNs), trained with mechanical response curves transformed into two-dimensional images that preserve their shape. This model enables accurate classification of responses into known phases and provides a continuous confidence score for each classification. This represents a paradigm shift toward similarity-based classification, facilitating the construction of continuous maps capable of realistically capturing micromechanical transitions and interfacial behavior.Overall, this work demonstrates the potential of HSNM combined with statistical and machine learning methods for characterizing complex multiphase materials. The thesis opens new pathways for improving the interpretation of heterogeneous mechanical behavior and integrating it with microstructural data, contributing to the development of more robust and automated methodologies in materials science.

Committee:
     PRESIDENT: MOLINA ALDAREGUIA, JON
     SECRETARI: FARGAS RIBAS, GEMMA
     VOCAL: CARO PRADOS, JAUME
Thesis abstract: WC-Co cemented carbides are widely used in industry, particularly for cutting tools, due to its exceptional combination of mechanical properties. They are composites consisting of tungsten carbide (WC) particles embedded within a cobalt (Co) metallic binder. This microstructural assemblage provides a unique combination of hardness, wear resistance, and fracture toughness. However, these properties can degrade at high temperatures (>500 ºC) due to corrosion sensitivity. To address this limitation, protective coatings, such as titanium aluminium nitride (TiAlN) or titanium silicon nitride (TiSiN) are frequently applied, enhancing performance under high-temperature loading.Grinding is the most conventional post-processing technique used to achieve the final geometry and dimensional tolerances of WC-Co components. However, this method impacts surface integrity, inducing defects like voids, cracks, high roughness (0.2 - 0.6 µm), phase transformations in the metallic Co binder, and compressive residual stresses, extending up to 10-20 µm in depth. Accordingly, removal of the damaged layer though polishing is commonly advised.In recent years, a new dry-electropolishing technology has emerged for metallic alloys, allowing precise defect reduction and surface improvement. Unlike conventional electropolishing, which uses liquid electrolytes, dry-electropolishing uses solid porous polymeric particles (ion exchange resins) filled with an electrolytic medium. This method selectively removes material layer-by-layer from surface irregularities, reducing roughness and improving surface quality.This industrial Ph.D. thesis focus on the development of a novel dry-electrolyte specifically tailored for electropolishing WC-Co composites using DryLyte® Technology. The research explores the underlying polishing mechanisms, optimizes electrical parameters, and identifies key chemical reactions involved in the process. Results demonstrate that this process effectively eliminates surface defects, reduces roughness, and preserves the WC-Co microstructure, avoiding Co leaching and WC embrittlement.A comparative microstructural and mechanical analysis was conducted, evaluating various shaping and finishing methods, including grinding, electrical discharge machining (EDM), conventional polishing, and dry-electropolishing. The findings reveal that, while grinding and EDM are widely used, they generate surface defects such as micro-cracks, porosity, and harmful oxides. Conventional polishing improves surface roughness but removes hundreds of microns of material and fails to preserve compressive residual stresses, thereby diminishing the mechanical performance of the system. Conversely, dry-electropolishing precisely eliminates deformed layers, achieves smooth surfaces between phases, and retains beneficial compressive residual stresses.The thesis further explores the industrial applications of this technology, such as cutting edge preparation for WC-Co end-mill tools. Previous studies have demonstrated that an optimal edge preparation increases tool life. The results of this research confirm that dry-electropolishing can fine-tune the cutting edge radius, refine micro-geometric parameters, reduce roughness, and eliminate manufacturing-induced defects. A direct correlation between polishing time and micro-geometric parameters is established, showing rapid changes within the first three minutes, followed by stabilization.Additionally, the research investigates the stripping of worn TiAlN/TiSiN coating on WC-Co substrates to extend their operational lifespan and reduce costs. This process removes the films precisely, leaving an oxide surface layer without compromising the substrate microstructure. The process demonstrates a linear relationship between polishing time and remaining coating thickness, ensuring a controllable, efficient, and homogeneous method.

Committee:
     PRESIDENT: GINER MARAVILLA, EUGENIO
     SECRETARI: RAMÍREZ SANDOVAL, GISELLE
     VOCAL: MAIMÍ VERT, PERE
Thesis abstract: In this thesis, the mechanical behavior of tungsten carbide-cobalt (WC-Co) hardmetal, a multiphase composite, is thoroughly investigated through computational modeling techniques. First, the thesis focusses on the small-scale, where the constituent’s assemblage and constitution are not only clearly visible and discernible, but also play a major role in the mechanical behavior. Then, stochastic factors that affect small-scale specimens’ strength are accounted for. Finally, the thesis focuses on the large-scale, through the implementation of a model that connects small- and large-scale properties into a single framework.At the first stage of the work, data published on (1) nanoindentation on WC particles and the Co matrix; (2) tensile tests on nanowires (NWs) made of WC-Co hardmetals; and (3) compression tests on micropillars made of WC-Co hardmetals are used for the development of a numerical methodology. To do so, it is developed a novel computational framework that includes two distinct microplane constitutive models developed for the WC and Co phases separately. For the Co matrix, the microplane J2-plasticity model, called MPJ2, is developed, while for the WC particles, a modified version of the microplane model M7 is used, called M7WC. As for simulation meshes, a full realistic 3D representation is used, derived from experimental tomography reconstructions of two WC-Co hardmetal grades. After optimizing the parameters of MPJ2 and the M7WC models with experimental data, the finite element (FE) predictions not only confirm the extensive experimental observations but also provide further insights into the mechanical behavior of these composites.In the second stage of the thesis, significant uncertainties in the mechanical behavior of the WC-Co hardmetals at the small specimen level are addressed. They arise due to factors such as the intrinsic randomness of the microstructure and possible existence of defects. A stochastic finite element method (SFEM) is used in conjunction with the deterministic models, M7WC and MPJ2, by introducing controlled randomness to some of the parameters of these models. The meshes are sampled from an existing tomography of a WC-Co grade using LHS. The results effectively capture the strength distribution of these ceramic-metal composites at small-scale under tension.Finally, in the third part of the thesis, and aiming to demonstrate that at the large-scale the mechanical properties are also dependent on the microstructure, a new constitutive model for the FE modeling of WC-Co hardmetals at the large-scale is introduced, effectively serving as a multiscale approach. Known as the microplane model for hardmetals (MPHM), the model is calibrated using stress-strain test data obtained under uniaxial tension and compression from specimens with varying grain sizes and cobalt contents. Once calibrated, the model with fixed parameters is employed to predict additional experimental data from uniaxial tension, uniaxial compression, and four-point bending tests sourced from literature. The model demonstrates a high level of accuracy in predicting experimental data across a broad range of cobalt weight fractions (3 to 27 wt%) and WC grain sizes (0.35 to 1.85 μm). The model requires only four commonly available material constants as inputs: cobalt content, grain size, uniaxial compressive strength, and uniaxial tensile strength. Put simply, it is developed a model for large-scale behavior of a wide range of WC-Co grades where only easily measurable material properties are necessary as input parameters.