Desarrollo y mejora del proceso de fabricación DIW de piezas de hematita porosa

11 B Equation Chapter 1 Section 1 Master Thesis Industrial Engineering Development and improvement of DIW process for hematite-based porous materials Author: Miguel Paúl Navarrón Tutor: Ranier Enrique Sepúlveda Ferrer Dpto. de Ciencia de Materiales Escuela Técnica Superior de Ingeniería Universidad de Sevilla Sevilla, 2023 Master Thesis Industrial Engineering Development and improvement of DIW process for hematite-based porous materials Author: Miguel Paúl Navarrón Tutor: Ranier Enrique Sepúlveda Ferrer Profesor titular Dpto. de Ciencia de Materiales Escuela Técnica Superior de Ingeniería Universidad de Sevilla Sevilla, 2023 i Abstract A diversity of technologies is the best solution to the multiple challenges humanity faces. Yet energy distribution, a pilar of society, relies primarily on two sources: electricity and fossil fuels. In this context, hydrogen stands out as a promising alternative, showing great potential for integrating renewable energy into the grid or transforming the transport industry. However, these aspirations are distant as long as hydrogen production remains tied to fossil fuels. Electrolysis, splitting water into hydrogen and oxygen using electricity, pushed by the dropping costs of renewable energies, emerges as a key solution. Yet, the significant cost gap, which goes from $6 to $2 per kilogram from electrolysis to fossil fuel-based hydrogen, remains the primary barrier to a broader adoption. Embracing technological diversity, this project explores Steam Iron Reforming (SIM) as a potential bridge between hydrogen derived from fossil fuels and renewable electricity. SIM leverages the redox cycle of iron to produce both syngas (a mix of H2 and CO2 used in industry) and pure hydrogen. The primary SIM approach involves transporting iron powder between reactors, enabling continuous production of both outputs and allowing iron replacement during operation. However, to increase efficiency, these systems need to be scaled up, making them suitable primarily for large-scale industrial usage, that would rely on an unexacting hydrogen distribution infrastructure. An alternative approach, single-reactor SIM systems, can be applied to distributed hydrogen production closer to the consumption. However, their usage of fluidized beds, the injection of pressurized gas through a powder material to make it behave as a liquid, faces its own challenges, main one being particles sintering under repeated cycles. To address this issue and improve the stability of single-reactor systems, bulk porous iron catalysts have emerged as a promising solution. Traditional methods for producing porous metals involve casting metal powders mixed with sacrificial materials, which, when removed, leave behind voids in the metal structure. Still, these processes are difficult to scale and are limited in shape flexibility, main corners on geometry being them only targeting microscopic porosity. This project presents an approach for the production of doped hematite (Fe2O3) porous catalysts for SIP by Direct Ink Writing (DIW). DIW is an additive manufacturing technique that deposits liquid ink, which solidifies, following a programmed path, all resulting in a final shape. The ink, composed of the target powder material and a carrying medium, allows the printing of complex geometries while enabling the removal of the medium to produce a final part with both macroand microscopic porosity. To achieve successful hematite DIW printing, this work investigates three main areas: hematite powder synthesis, ink development, and DIW equipment modification. These areas are then combined to demonstrate the feasibility of the process, leading to the production of final hematite catalysts. Hematite synthesis is conducted via Sol-Gel, a hydrolysis-based process, which enables the creation of hematite from metal salts (such as chlorides and nitrates). This method allows to produce nanoparticles and the doping of hematite with secondary metals like titanium or aluminum, improving the stability of iron during multiple SIP cycles. In this first section this work aims to standardize the procedure while proving the viability of doping the hematite, both with aluminum and titanium. Ink development focuses on colloids, suspensions of non-aye visible particles in a medium, to create a material suitable for printing. Exploring ink rheology, viscosity-displacement correlation which mainly influence printability prompted the use of non-aqueous gels as ink, formed by an organic medium, camphene, and the combination of two additives, stearic acid (SA) and polystyrene (PS). This section details the evolution of ink fabrication and composition throughout the project. As the project developed, continuous improvements to the DIW system were made to address challenges during printing. Using digital modeling, fused filament printing, and basic machining, necessary parts were designed ii and produced. The progression from the initial to the final stages of the system is presented, along with a foundational understanding of the system's operation. The successful production of final hematite samples with intricate geometries demonstrates the viability of this approach. However, parallel routes developed alongside this project highlight the need for further improvements in ink composition, printing conditions, and printing path design to enhance mainly repeatability, especially as scaling up the system demands longer printing periods. iii iv - Index Abstract i Index iv 1 Introduction 1 1.1 Objectives 5 2 Theorical framework 6 2.1 Direct Ink Writing 6 2.1.1 Three axial-extrusion systems 8 2.1.2 Geometrical control 10 2.1.3 DIW design tool 10 2.2 Rheology in DIW 11 2.2.1 Oscillatory test 13 2.2.2 Rotatory test 13 2.3 Ceramic inks 14 2.3.1 Colloids as inks 14 2.3.2 Colloids stability 17 2.3.3 Gels as inks 23 2.4 Hematite nanoparticles synthesis 26 2.4.1 Sol-Gel synthesis 27 2.5 Hematite Additives 29 2.6 Hematite foam 30 3 Experimental procedure 32 3.1 Sol-gel hematite synthesis 32 3.1.1 Precursors and agents 34 3.1.2 Procedure 35 3.1.3 Hematite particles functionalization 40 3.2 Hematite ink development 41 3.2.1 Ink constituents 41 3.2.2 Ink initial fabrication route 42 3.2.3 Ink modification with HDDA 44 3.2.4 Ink final fabrication route 45 3.2.5 Ink final composition 46 3.3 DIW development 47 3.3.1 DIW initial system 47 3.3.2 Initiation in DIW 48 3.3.3 DIW system improvement 49 3.3.4 Equipment current state 52 3.3.5 DIW parameters 53 3.3.6 Green samples 54 3.3.7 Path design 54 3.4 Samples post-process 56 3.4.1 Freeze drying 56 3.4.2 Sintering 57 v 4 Results 59 5 Conclusion 64 5.1 Future steps 64 5.1.1 Non camphene base ink 65 5.1.2 Application of FGF 66 5.1.3 Application of FFF with commercial hematite filaments 67 Bibliography 69 Annex List 78 Annex A 79 Annex B 86 Annex C 89 6 2 THEORICAL FRAMEWORK The fabrication route core of this project, table 1.3, can be englobed in a wider family, powder manufacturing or powder metallurgy (PM), as the base material selected consists in doped hematite nanoparticles (NPs). PM fabrication is a family of process that takes advantage of sintering and compacting of fine powders to avoid the high temperatures required for melting in traditional casting techniques, while allowing near-net-shape production for most metals, metal oxides and their alloys. PM main advantage is the ability to obtain mechanical parts with specific properties and inaccessible materials by traditional ingot metallurgy, exampled in self-lubricating bearings or refractory metals [31], however incomplete compacting generally leads to an intrinsic porosity resulting in lower mechanical performance, particularly considering strength and toughness. This seems to favor the fabrication of porous materials, as aimed in this work, yet lacking interconnectivity and pore size control these techniques require additional steps to obtain functional pieces. The common approach to obtain interconnected and controlled porosity materials relay on a sacrificial phase which is removed during sintering or by an additional chemical or physical processing [32]. The two phases precursor can be achieved by two methods: infiltration, which introduce aimed material in an already porous structure; or phase growth control techniques, like freeze casting, that looks to control the shape formation of the sacrificial phase inside the material. Getting further away from traditional PM, involve mainly in casting, this project looks to take advantage of rapid prototyping techniques. Main PM technique inside this philosophy is select laser sintering (SLS) which allows the progressive direct point melting of fine powder layers following the shape of the selected design [33]; however, this method achieves similar dense parts to traditional PM, and although it could achieve macroporosity by geometry design, it lacks on micro-porosity control advantages. Direct ink writing (DIW) applies the direct extrusion of an ink which its solidification allows the progressive deposition of a 3D layered shape [34]. The DIW ink can be obtained by the formation of a suspension with the desired powder and a liquid base which can act later as sacrificial element to obtain a porous material, this way DIW can be connected to PM porous material fabrication to obtain shape free pieces looking to take advantage of both micro and macro porosity. 2.1 Direct Ink Writing DIW was developed to produce ceramics parts with a complex shape, which are not achievable by traditional casting, due to molds geometry limitations; or by shaping and machining, based on failures lead by ceramics high brittleness and hardness [35]. There are two DIW families, droplet-based writing and filament base writing [36]. Both of this process are based on the extrusion of a feedstock synchronize with the movement of the extrusion equipment to produce a path layered 3D design, with the distinction that droplet-based systems applied a batch outlet; while filament base uses a continuous one, equivalent to traditional additive manufacturing (AM), as illustrated in figure 2.1. 2.1 Figure: Schematic process illustration of filament-based DIW. 7 Robocasting filament DIW, where a three-axis robot controls the movement of the printing head, is the specific process concerning this project. The first characteristic of this fabrication approach is the extrusion method that control the material flow, where there are three commons ways: piston pressing, pneumatic pressing or endless screw driven, as schemed in figure 2.2. 2.2 Figure: Extrusion methods in DIW. The second aspect to account on DIW is the liquid-solid transition method which allows the initial extrusion and later the shape retention once deposited, schemed in figure 2.3. The application of any ink really on flow/pressure relation, therefore a rheological study of the ink is often necessary to fully understand its behavior, and due to its importance, a specific section ahead dives in this topic. Non-Newtonian substances which become liquid under stress (shear-thinning substances) are ideal for this process letting DIW to be performed at room temperature. Other process as temperature solidification, polymerization or solvent evaporation are also applied for this purpose but more complex control or printing equipment with additional heating elements or additional external stimulants are required, very well illustrated in the review by M.A.S.R Saadi on different DIW applications [34]. 2.3 Figure: Liquid-solid transition during deposition process. Third aspect to account in DIW, which is major in any other AM, is the triad: tolerance, accuracy and precision, figure 2.4. For DIW as in fused deposition modelling (FDM) this triad is mainly characterized by the nozzle diameter (D), which influences the layer high (h), the extrusion filament width (w) and the detail size (R), magnitudes represented in figure 2.5. A starting assumption for printing control would be approximate all these parameters to be the same value. Considering the nozzle diameter as main dimension for the printing process is only possible if the printer axes and extrusion movement tolerance are a magnitude under this parameter, allowing to adjust the printing control parameters to the nozzle dimension. 2.4 Figure: Scheme of tolerance, accuracy and precision. 8 2.5 Figure: Main extruded filament dimensions for 3D path printing. Nozzle diameter is not just limited by accuracy needs or printer tolerance, when aimed dimensions are reduced common limit faced in printing is the avoiding nozzle clogging while assuring a high particles-density in the ink [35]. Nozzle clogging takes place under two mechanism: reduction of the nozzle diameter leads directly to an increase of the pressure load required for printing, resulting in clogging if this load is not achievable, as the pressure difference is directly correlated with the shear stress, magnitude responsible of the liquid flow, by the equation 2.1, shear stress for a revolution shape at different lengths [36]; second process that produce nozzle clogging is the solidification of the nozzle tip, this process is common in inks where temperature, evaporation or polymerization is applied as solid-liquid transition paths, an limits nozzle diameter to sizes upper 500 μm [36]. For smaller nozzle size up to 100 μm oil base printing, which provides additional structural support and looks to separate solvent evaporation process to extrusion can be applied for this ink variants [37]. τ𝑟=𝑟 ∙ ∆𝑃 2𝑙 2.1 A second aspect must be considered to avoid clogging. When particles suspensions are applied as ink, particles agglomeration and distribution irregularities provoke flow variations which lead to printing errors and clogging. Different ink preparation aspect must be approached to avoid this: smaller particles size usage, commonly an order of magnitude under the accuracy required; fabrication methods that assure a homogeneous well dispersion of the particles; and the application of dispersants to reduce agglomeration [35], [38]. Colloidal ink preparation will be further investigated in a specific section. 2.1.1 Three axial-extrusion systems Three main aspects for DIW have been introduced, the extrusion methodology, the ink solid-liquid transition, and the dimension characterization of the DIW process. But over DIW an aspect on any robotic systems is the movement it is capable of, or in a technical approach, the degrees of freedom the system handles. Reasoned on price/accuracy, basic AM system that apply a point-to-point printing method as DIW, FDM or SLS, use a 3-axis lineal movement system, as this approach do not add up movement error as other common systems like robot arms, meaning that a good precision can be achieved with a relative cheap and simple technology. Three axis lienal systems (X, Y, Z) in 3D printing, that required achieving complex geometry with low tolerance, relay on step motors to control the three movements. In simple printers, real time position is not followed by a sensor, instead a switch in each axe allows the printer to set a home position (0, 0, 0). This together with digital control of forward and backward movement determinates the printer head, moving part that englobes the nozzle, current position. In DIW or FDM, on top of the three axes, an additional step motor controls the extrusion process, except on pneumatic pressing, but this method is not under current interest for this project. The step motor controls the extrusion distance (E), which in basic systems do not require any type of sensor, meaning that at the start of any printing process the motor current position is set as zero value reference. Basic printing systems, based on explained control method, are very susceptible to miss-location, due to wrong parameters settings, undesired movements or crashing; issues that must be valued when troubleshooting, as they 9 can lead to serious damage of the printer or to print errors. A basic approach to avoid miss-location is to include a home positioning process at the beginning of the code to avoid starting the process from a wrong zero point. For the extrusion control the situation is less desirable as in basic printers control, with no extrusion zero sensor, the real extrusion position is unknown; in filament FDM this leads to not critical failures as the absence of filament is the only common problem, but in piston extrusion DIW this unknowledge can produce the piston to reach its limit and suddenly increase drastically the torque seeing by the motor, consequently damaging it. As base theory, for three axial-extrusion systems, the main parameters to consider are illustrated in figure 2.6. As seen in this figure in extrusion systems the extruded material diameter (DE), the filament diameter in FDM or the deposit diameter in DIW, does not match the nozzle outlet diameter (Dn), which implies that slower feed rates [mm/min] can be applied in diminish of increasing the step motor precision requirement. 2.6 Figure: Main parameters in three axis-extrusion printing systems. The last parameter to consider, shown as well in figure 1.6, is the feed (F) or the speed of the printing process, normally expressed in millimeters per minute. This value, in printing, controls both the movement and the extrusion process, which allows to control the steps required in each motor. In system of equations 2.2, it is expressed how the feed affects the horizontal step motor speed (vx) between two points (A-B), being nx the number of steps required for the movement and fx the step to displacement ratio of the motor. d2= (xB− xA)2+ (yB− yA)2+ (yB− yA)2 2.2 t = d/F nx=(xB− xA)/fx 𝑣x= nx/t To determine the extrusion distance (E) or extrusion speed (ve), for the motor control, the assumption of a square filament deposition base on the nozzle diameter, as in figure 2.5, is usually applied as a good approximation. This assumption together with some equations in 2.2 previous calculations allow this determination as is shown in equations system 2.3. V = w ∙ h ∙ d = Dn2∙ d {𝑤 = ℎ = D𝑛} 2.3 E = 4 ∙ V/ ( D𝐸2 ∙ π ) 𝑛𝐸= 𝐸/𝑓𝐸 𝑣𝐸= 𝑛𝐸/𝑡 10 2.1.2 Geometrical control The base language for any numeric control (NC) computer assisted manufacturing (CAM) system, like 3Dprinters, is known as Geometrical Coding (G-Code). This programming language is structured in lines each one containing a command which can be read looking at the symbol letters used and the number associated, as shown in figure 2.7. 2.7 Figure: G-Code structure for printing equipment. Most 3D printing path are based on simple linear movements between two points, in these cases the G-codes consist basically of G1 and G0 commands, extruding and not extruding linear movements, where the coordinates (X, Y, Z) indicate the final aimed point and E the extrude distance required. G-code is a very direct programming language, which relates directly to the parameters explained in previous sub-section, figure 2.6, and corresponds to the file format readable by 3D printers (nc. or gcode.). Being the final file before the printing process justifies a minimum knowledge on this programming language by itself, but this knowledge is especially useful when analyzing, step by step, the printing process, as this code file gives exactly what parameters are influencing the printing in any point. A last comment on G-Code, needed for its proper understanding, is that this language, as many programming tools, is ‘a-dimensional’; numbers on the code are interpreted by the parameters on the printer settings, for example the step to displacement ratio of the motor (f), therefore to properly read a G-Code is required to know the equipment it is going to be used and its configuration. 2.1.3 DIW design tool DIW is a technique which reproduced a 3D geometry by layering and pathing. This is commonly achieved by going from the initial 3D design, done with CAD software, to the G-Code readable by the printer by a slicing and translating software, as schemed in figure 2.8. This approach useful for commercial and standardized printers leads to good results with a user-friendly interface that can handle complex geometries, however for printing system development this process lacks direct parameter control. A second approach would be to directly design the path by programming the G-Code, but this method is tedious and limited to relatively simple geometries. An intermediate approach is found in a free phyton library, FullControl (FC), developed by Loughborough University PHD, Andy Gleadall [39], found in their website https://fullcontrolgcode.com. 2.8 Figure: Common AM processing steps. FC looks to develop a series of steps, based on points, that together with geometry or control commands, printer parameters and direct G-Code incorporation allows the user to fully control of the final G-Code with an easier approach to direct G coding, also allowing the design of complex control sequences or geometries by taking advantage of phyton programming capabilities. This approach also shows AM potential by targeting path designs which are difficult to achieve in common CAD as illustrated in figure 2.9. 11 2.9 Figure: FullControl designs examples from https://fullcontrolgcode.com. Comparing FC and G-Code base printing variables, as in figure 2.10, it can be seen how FC brings the user easy physical parameters that relate to the printing process directly and are equivalent to the parameters commonly ask in any slicing software. FC design process starts with the definition of the printing parameters; followed by the writing of a list of steps, which will define the movement, therefore the geometry; and lastly the automatic translation to G-code. 2.10 Figure: FullControl and G-Code base printing parameters. 2.2 Rheology in DIW As mentioned, shear thinning is a desired, if not required, aspect for inks in DIW. This property is part of the rheological behavior of the ink, the flow-deformation behavior of the material, which controls its printability: extrudability, self-supporting characteristic, shape fidelity, or as a simplification the ability of an ink to be extruded continuously as a filament be deposited and maintain structural stability [34]. Rheology is mainly studied by three different behavioral graphs illustrated in figure 2.11, the represented curves display shear thinning materials expected results. The initial parameters for rheology are: shear rate [s⁻¹], the rate at which one layer of fluid moves relative to a consecutive layer; shear stress [Pa], tangent force applied to one layer of fluid per unit of area; this two parameters in Newtonian fluids are related by a constant known as shear viscosity, equation 2.4, where τ is the shear stress, η the shear viscosity and γ󰇗 the shear rate. 12 2.11 Figure: Behavioral schemed graphs for shear thinning rheology, based on results on Shahzad A. study on DIW [35]. 𝜏 = γ󰇗 ∙ 𝜂 2.4 With the case shown, figure 2.11, it can be understood how in a shear thinning material, where viscosity has an invert proportionality to the shear rate, the force required to extrude the material is reduced as it starts flowing, allowing it to be properly extruded. Moreover, thinning behavior is dynamic, explaining why continuous pouring, initial over-pouring or sacrificial initial pouring are common practices to achieve proper results, as they lead to printing in a stable behavior stage. The other material property that can be extracted from the rheology behavior graphs, figure 2.11 is the structural stability of the ink. G’, known as the storage modulus, represents the potential energy stored in a material after its deformation, it is usually illustrated as the energy that allows an elastic band to recover its shape after it has been stretched. G’’ named as loss module determinates the energy, dissipated as heat, of a material undergoing a deformation. In the viscoelastic area (G’ > G’’) the deformed energy stored is higher that the energy the viscosity could dissipate which leads the material to retain its shape; in the other case (G’ < G’’) dissipation process is capable to handle the deformation energy achieving a viscous-liquid behavior. In a shear thinning material the viscoelastic or ‘solid’ state is achieved at low shear stress giving them the shape retention under no external force, for example after their deposition in DIW [34]. The only force seeing by the material after deposition is its own weight. If the thinning behavior of the material is too severe this force could lead to the deformation of the initial layers as consecutive layers are built on top which leads to what is colloquially known as elephant foot, as illustrated in figure 2.12. This failure not only leads to geometry inaccuracy on bottoms layers, but it also leads to nozzle high miss position consequent of accumulative layers compaction. 2.12 Figure: Elephant foot failure in printing. To finalize the ink rheology study, it is necessary to contemplate the relation of the viscosity with other aspects like temperature or solvent evaporation, mentioned as solid transition processes. As there are less methodologies to evaluate these relations, together with the high influence of the printing system on these relations, an ambient and temperature control is recommended to be incorporated to any DIW equipment. 13 2.2.1 Oscillatory test To better understand rheology, it is useful to investigate its testing. The base study of rheology is achieved by an oscillatory test, in which a liquid is embedded between two plates that oscillate, lustrated in the scheme in figure 2.13. The difference between the plates movement and the liquid characterized its viscosity [40]. 2.13 Figure: Scheme of a rheology oscillatory test, based on the material found in Anton Paar Gmbh website [40]. When a tangent force (F) is applied between the two faces the displacement seeing by the top face (s) is different from the displacement of the liquid (s’); moreover, when the applied force is oscillatory the liquid follows the oscillation with a phase retard (δ). Quantifying these variables, illustrated in figure 2.13, the parameters that represent the rheological behavior can be determined, system of equations 2.5. γ󰇗 = 𝑠/ℎ { γ󰇗 ′ = 𝑠′/ℎ } 2.5 τ = F/A { A ≡ 𝐿𝑖𝑞𝑢𝑖𝑑 𝑐𝑜𝑛𝑡𝑎𝑐𝑡 𝑎𝑟𝑒𝑎} G′= G∗∙sin (𝛿) { G∗= 𝑠′/𝑠 = γ󰇗′/γ󰇗 } G′′ = G∗∙ cos (𝛿) The application of consecutive oscillatory test with different tangent force applied gives access to the rheological behavior graph {G’,G’’/τ}, figure 2.11, and helps to approximate the minimum shear stress that leads to the transition from liquid to solid, named as critical shear stress (τc). This value allows to dimension the pressure required to extrude the ink in a DIW system, following basic extrusion fluid models. 2.2.2 Rotatory test Although an approximation of the viscosity could be approached with the oscillatory test, a specific test for is measurement is found with a rotary rheometer. The basic type for this equipment, the cone-plate, locates the liquid between a plane plate and a rotatory conic element, as seen in figure 2.14. The objective of the conic shape is to apply a constant shear to the liquid; this allows the determination of the viscosity base on the relationship between the applied torque (T) and reached speed (w) measured, equations in the system 2.6 [41]. γ󰇗 = w/𝛩 {𝑐𝑡𝑒} 2.6 τ = 3T/(2π ∙ 𝑅3) 𝜂 = τ/γ󰇗 = 3T ∙ 𝛩/(2π ∙ w ∙ 𝑅3) 14 2.14 Figure: Rheology rotatory test scheme. Like the oscillatory test, the consecutive testing at different torque or speed allows the determination of viscosity graph, {τ/γ󰇗} and {𝜂/γ󰇗}, and therefore its determinates the shear behavior of the material. When torque is applied as control parameter this experiment fits the liquid behavior on force driven applications [40], like extrusion, assisting on the design of systems as DIW equipment. 2.3 Ceramic inks An ink is a solution that incorporates at least one pigment, and which composition allows the painting process. Following this definition, in DIW inks are understood as the medium that carries the desired material for the final part and allows the printing process by accomplishing the properties that have been discussed in previous sections. In DIW literature there can be found both: inks base on a single material that forms the final part, and has printing properties, like polymers and resins [42], [43], [44]; or inks base on a liquid medium that carries the desired material, and which combination properties or with additives achieves printability, in what is call a green stage [45], [46], [47], [48]. When an oxide, as hematite, is desired as final material, producing a ceramic, three possible approaches can be taken to formulate an ink: colloidal suspensions, hydro-gels or organic-gels; from simpler to more complex systems [49], schemed in figure 2.15. 2.15 Figure: Scheme of ink solutions approximation for oxides particles, based on figure 3 in del-Mazo-Barbara L review in ceramic inks rheology [49] 2.3.1 Colloids as inks The initial and main approach to produce and ink after a powder is to directly try suspending it into a liquid medium, commonly in a first attend, water. To achieve this, friction forces must be capable of stopping the particles precipitation, figure 2.16, to assure the ink stability over time [50]. 15 2.16 Figure: Particle suspended in a liquid medium forces scheme. The suspension of a particle is achieved when its sedimentation speed in the medium (vp), sedimentation speed under a laminar flow, equation 1.10, tends to be zero (vp ≈ 0). There are three ways to enhance the suspension of particles base on the sedimentation speed: applying particles with closer density to the liquid (𝜌𝑃≈ 𝜌𝐿), well exampled in the application of oxides particles over their metal pairs; selecting liquid mediums with higher viscosity (𝜂 ↑), approach that relates to the application of gels over colloids; and lastly, decreasing the particle size (r ↓), which has been implied by referencing colloidal suspensions directly. 𝐹 𝑔=(𝜌𝑃− 𝜌𝐿)∙𝑔∙43 ⁄∙ 𝜋 ∙ 𝑟3 2.7 𝐹 𝑓= 6 ∙ 𝜋 ∙ 𝑟 ∙ 𝜂 ∙ 𝑣𝑃 2.8 𝑣𝑃=29 ⁄∙(𝜌𝑃− 𝜌𝐿)∙ 𝑔 ∙ 𝑟2/𝜂 {𝐹 𝑔= 𝐹 𝑓} 2.9 Colloids, mixtures where suspended elements in a medium are not appreciated with an optic microscope, are characterized by particles dimension in the range of 1 to 100 μm. In this range particles suspension is easily achieved, granting these dispersions a good stability and homogeneous appearance. The use of colloids as inks is backed on the rheological behavior that these complex or structured liquids manifest. Although this behavior, due to its complexity, has not reach standard models that explains and allows its complete understanding [51]; an initial approximation, hard sphere particles model, allows to have an insight on how particles ordering under shear movement influence the viscosity, as schemed in figure 2.17 [52]. 2.17 Figure: Hard sphere particles ordering under shear rate and its influence in viscosity. Based on Daniel T.N. Chen insight on soft materials, figure 3 [52]. 22 Stearic stabilization Medium and particles are often implied by criteria other than stability, as well as other parameters like pH or ions concentration, which cannot always be modified. Under this constrains new ideas had to be explored to assure the colloid state over time. Since surface interactions govern the physical behavior of colloids, surface manipulation through particles coating naturally emerges as a key solution. Steric stabilization involves the coating of the particle with surfactants or polymer, which allows both: the selection of the surface electro-chemical properties, allowing solvation or electrostatic stabilization; and the generation of a physical boundary that prevents agglomeration, equivalent to solvation. Further to this, both mechanisms can be combined in electro-stearic stabilization, figure 2.29. 2.29 Figure: Stearic and electro-stearic surface modification scheme. The role of stearic stabilization can be introduced in the DLVO as a repulsive force, that prevents the contact of particles in close distances, as seeing in figure 2.30. To consider this stabilization method it must be understood that particles surface manipulation has a secondary role on rheological properties by both increasing particle characteristic diameter and its interaction with the medium. Studying the change on the ink behavior is necessary. 2.30 Figure: Stearic force, brush repulsion, scheme inside DLVO theory, based on fig 1.6, Henk N. W. theory on colloids and depletion [67]. Depletion The last idea that comes to mind for the modification of colloids stability, is the addition of third substance to the particles-medium mix, which do not entangle with the particles surface, commonly a non-adsorbing polymer. This addition generates an intermediate body between particles that prevents their agglomeration, while also allowing, secondary modifications, like Van der Walls interaction screening or the modification of the medium electrostatic nature [50]. However, depletion is normally associated to particles agglomeration, as the present of a third compound can lead to the phase separation of the different substances as osmotic pressure rise, this is especially true as particles and depletants concentration rise. This phenomenon is visually explained if it is imagined that the area around the particles has a lower depletant concentration, due to is non-adsorbing nature, this means that the union of two particles leads to a higher free space for the depletants movement, leading to an attractive force between particles, at what is casually know as attraction by repulsion [67], this process is schemed in figure 2.31. 23 2.31 Figure: Scheme of agglomeration by depletion. Under this project interest, the role of depletion is important because it is strongly tied to stearic stabilization. When a polymer is added to the particles-medium mixture it cannot be precisely control whether the polymer is going to attach to the particles or if it is going to knot and act as a depletant, especially as the polymer concentration rises and able particle surface decrease. Both interactions can be included together in the DLVO theory, figure 2.32. Considering this dual role of polymers, these are usually classified as binders of dispersants depending on their common usage, as it will be further discussed in next section. 2.32 Figure: DLVO theory, scheme of depletion and stearic stabilization as a joined phenomenon. 2.3.3 Gels as inks The application of colloids as inks is limited by lack of control over rheological properties as mainly the only control parameter is the powder concentration, or if manageable the change of particles size or surface characteristics. To achieve greater control multi-phase mixtures is the natural next step, slightly introduced in depletion previous subsection. The most common way to produce these samples are the addition of small amounts of binders and dispersants, resulting in what can be referred as gels, considering gels as substances with high viscosity to pseudo-plastic behavior, in this case suitable to rheological printing properties. Not many samples of the usage of aqueous based gels for iron inks can be found, showing the difficulty of usage of water as base medium for the DIW of these materials, table 2.1. 2.1 Table: List of exampled aqueous gels formulas for its application as DIW inks for iron base powders. Powder Concentration State Addon Reference Magnetite 50 wt% Air foam Amphiphile hexylamine [68] Fe and C 20-80 wt% Gel Metakaolin, sodium metasilicate, sodium hydroxide [69] The next step in optimizing the mixture for use as ink involves the direct substitution of water with other organic liquids, as illustrated in table 2.2, which lists various inks for iron-based powders. The choice of alternative mediums can vary depending on the specific application. For metallic iron, the substitution is often aimed at preventing oxidation during prolonged contact with water. However, for hematite or other iron oxides, the change might be driven by the need to increase system viscosity, raise medium density, set different working temperatures, or enable specific treatments during post-processing of green samples. 24 2.2 Table: List of exampled gels, non-aquas medium, formulas for its application as DIW inks. Powder Concentration Medium Addon Reference Fe - Mn 10-70 % Dichloromethane Polylactic acid [71] Fe 5:1 Dichloromethane Polylactic acid [72] Fe2O3 7:3 Dichloromethane (DCM), ethylene glycol butyl ether, and dibutyl phthalate Polylactic-co-glycolic acid copolymer [73] Fe 3:1 Dichloromethane Polylactic acid [74] Fe3O4 10-30% Polydimethylsiloxane - [75] Medium Camphene, the medium of interest of this project, table 1.3, is an organic compound, with the structure shown in figure 2.33, a density of 0.803 (70°C) – 0.850 (5 °C) g/cm3, and a fusion temperature around 44 - 48 ºC. Camphene application come from its usage as medium for porous materials casting techniques, further explain in subsection 2.6, but can already be seen in the application for DIW [76]. This ‘crystal polymer’ has the advantage of allowing its removal, its debinding, by the application of lyophilization, as camphene presents a high vapor pressure in its solid stage, 1.3 kPa (25ºC), meaning that at room temperature camphene complete debinding occurs after 24-48 hours. Together with this camphene allows the dissolution of polymers, as PS, and brings thermal solidification as a possible liquid-solid transition method, without reaching high printing temperatures, commonly around 200ºC for traditional printing polymers like PLA or ABS. As a medium camphene also present a similar viscosity to water, 1.4 mPa·s a 47 °C, and has a low thermal contraction after solidification, 3.1%, which decrease sample damage after printing. 2.33 Figure: Camphene molecule. Dispersants Dispersants are typically surfactants, meaning they are attached to the surface of particles. These substances can be either organic or inorganic, with the most common source being oil-based organic compounds. They are usually classified by their length and polarity into four types: anionic, cationic, non-ionic, and polymeric. The application of dispersants is managed with the premise that the correct amount of dispersant is related to the saturation of the particles surfaces which leads to a point of minimum viscosity, meaning that any excess or default amount is undesired. This common assumption is correlated with a stearic stability. Selected a substance suited to the medium with a strong anchorage to the particle, the main parameter to consider is the molecule length. Larger molecules are related to stable systems as 10 to 20 nm Van der Walls threshold is surpassed, leading to a physical gup between particles that prevents their binding. It must be considered that longer molecules increase the overall space occupied by the suspended particle affecting the maximum powder load that can be added to the gel. A compromise between stability, dispersant length, and power load is then present in the design of gels as inks. Some shorter organic compounds like oleic acid or stearic acid are interesting under organic mediums. Although their length does not lead to a pure stearic stability, these substances present a pseudo-stearic stabilization where their physical barrier is joined to the reduction of Hamaker particle constant, which leads to lower Van der Walls forces. The stearic acid, formula 2.14, is a lineal chain, around 2.44 nm length, with no branches, that presents a carboxylic acid as functional group. The carboxylic head allows the formation of hydrogen bonds to the basic surface of the suspended particle generating a strong connection. The tale of this king molecule presents a good behavior, leading to lower viscosities, in non-polar organic mediums, understood by a better solvation of the tale under these mediums and a greater reduction of Hamaker constant. 25 𝐶17𝐻35 − 𝐶𝑂𝑂𝐻 2.14 Binders The use of binders may seem counterintuitive when considering colloid stability. However, in the case of inks used for solid formation, binders play a crucial role in maintaining structural integrity during the post-processing phase. Binders are typically non-ionic polymers that dissolve in the liquid medium and possess low glass transition or melting points, enabling them to form bridges between particles through hydrogen bonding, which enhances structural stability [50]. Despite these benefits, the addition of binders significantly impacts ink properties, particularly by increasing medium viscosity due to the complex interactions between particles, binder chains, and the medium. Moreover, achieving proper ink homogenization becomes more challenging with the presence of binders. The chemical or thermal decomposition of binders is selected to differ from that of the medium, allowing the binder to serve as what is referred to in sintering literature as a "backbone material." Ideally, before the sintering process, the goal is to remove as much non-sinterable material (i.e., the medium) while preserving the structure of the intended sample. The voids left by debinding enhance particle mobility and the formation of sintering bridges. This process is achievable because the binder remains intact while the medium is eliminated either through chemical means or thermal decomposition [70], as schemed in figure 2.34. 2.34 Figure: Debinding process scheme. The combined addition of dispersants and binders introduces complex interactions that can reduce stability, even when optimal amounts are used in the ink formulation. In such cases, rheology and the actual printing experience often become the primary tools for evaluating the ink’s suitability. These assessments help gauge how well the mixture of four key elements—powder, medium, dispersant, and binder—interacts during the printing process. The selected binder for this work is polystyrene (PS). This thermoplastic, derived from styrene, shown in figure 2.35, presents low glass-transition point around 100ªC, with a commercial density of 1.05 g/cm3 (at 25 °C), and a molecular weight that goes from 17000 to 350000 g/mol, directly correlated to the polymer string length. The application of higher molecular weights brings higher structural properties, but also the raising of viscosity and non-homogeneities. 2.35: Polystyrene organic form. 26 2.4 Hematite nanoparticles synthesis There are three commonly known forms of iron oxide, Wüstite (FeO, iron II), Magnetite (Fe3O4, iron II-III) and Hematite (α-Fe2O3, iron III), inside iron oxide family, table 2.3. Hematite is the most stable polymorph of iron oxides in ambient conditions. It is half-metallic ferromagnetic oxide, with interesting semiconductor and magnetic properties[77], very attractive for nanoparticles formation due to the multiple nanostructures it can adopt, exampled in figure 2.36, together with the ability of forming complex hierarchical structures constructed with nanoscale building blocks [78]. 2.3 Table: Iron oxides family [79]. Fe II Fe II-III Fe III FeO Wüstite Fe3O4 Magnetite α-Fe2O3 Hematite Fe4O5 β-Fe2O3 Akageneite Fe5O7 γ-Fe2O3 Maghemite Fe25O32 ε-Fe2O3 Fe13O19 2.36 Figure: Hematite nanoparticles figures, nanorods, hollow microspheres, nanoparticles and microcubes, left to right from Lian J. work on hematite different shape synthesis [80]. The interest in nanoparticles (NPs) resides in their inner different behavior and properties, commonly optical, magnetic, and chemical properties owing to their small sizes, compared to their macro scale equivalent, differences that are more pronoun with sizes between 1 to 20 nm, and highly decrease over 50 nm [81]. Hematite nanoparticles are in general low processing cost material with a great chemical stability leading them to many potential usages in diverse fields such as catalysis, gas sensors, pigments… etc. [82]. A deep analysis on hematite properties and possible nanostructure can be found in Wan H. review on hematite newly emerging applications [83]. Finally, three main hematite physical properties are gathered in table 2.4. 2.4 Table: Hematite main properties. Density Molecular Weight Fusion Temperature 5.24 g/cm3 159.69 g/mol 1565 °C Hematite nanoparticles can be produced by several methods reported [78], main ones are explored in table 2.5. Solvent base methods are leading hematite NPs production as they avoid grain coarsening and particle sintering at traditional high temperature solid stages methods. Nonetheless, wet methods still require expensive metal compounds such as precursors and commonly strong acids, bases, or organic solvents [84]. For this reason, large-scale production with precise control of the size, morphology and dispersity are considered still underdeveloped [85]. 27 2.5 Table: Hematite nanoparticles synthesis methods. Process Definition Temp Related articles Microemulsion Thermodynamically stable isotropic dispersion of immiscible liquids which forms nanoreactors where a desired reaction can be achieved while controlling crystal nucleation and growth [86]. 20-30°C [87] Thermolysis / Forced hydrolysis Heat evaporation of an ion water solution (Thermolysis), with the addition of concentrated alkalis (Forced hydrolysis), followed by washing steps to remove unreacted species [86]. 90100°C [88], [89] Sol-gel hydrolysis Addition of organic molecules or polymers to colloidal dispersions of metal ions to form the sol, concentration of this sol by removal of the solvent, followed by thermal treatment to obtain the product [86]. 70100°C [90], [91], [91] Precipitation or coprecipitation Addition of alkali to a solution of soluble metal salts to form precipitates which are then aged at certain temperatures to obtain the required product [86]. 50-80°C [92], [92], [93] Polymerization Production of polymer base nanoparticles that act as precursors. Pyrolysis, combustion or thermal decomposition treatment is applied for final metal particles obtention [94]. - [95], [95] Direct oxidation Oxides production by leaving metallic particles in a corrosive ambient. An analysis on iron nano particles oxidation was develop by Kumar N. [96]. * Thermal decomposition Direct conversion of an under-oxide iron form, obtained by a other method, to hematite by high temperature treatment in air or oxygen atmosphere. This method is included as a purification step in other methods as sol-gel or solvothermal method. 500700°C [97], [97] Sono-chemical Enhance chemical reaction of a precursor solution by direct ultrasonic irradiation, followed by precipitation, drying step and heat treatment [98]. 20-80°C [85], [98] Hydrothermal / Solvothermal High temperature and pressure chemical reactions taking place a in a sealed vessel with water (Hydrothermal) or other organic solvents (Solvothermal), flowed by various techniques to control size and morphology of the synthesized materials [86]. 160250°C [80], [82], [99] Electrochemical Application of a control voltages to a metal ion or cathode in a electrolytic bath, commonly aqueous based, which leads to particles deposition in the opposite electrode. 20-60°C [100] Solution combustion Autocatalytic combustion of a metal ions solution, commonly aqueous solution with and additional fuel. For nanoparticles fabrication atomization of the liquid before reaction previous to combustion is required [101]. - [102], [102] Pyrolysis Thermal decomposition in the absence of oxygen of precursors at high temperature. Precursor solutions are atomized into the pyrolysis oven to achieve nano scale particles [81]. 3001100°C [81], [103] Gamma-irradiation techniques Application of gamma-irradiation into a precursor’s solution for decomposition into metal nanoparticles [104]. 20-30°C [104] Mechanochemical Chemical reaction that takes place during grinding of dry solid precursors to form mixture nanocrystals [105]. Additional heat treatment may be necessary to complete the reaction [106]. - [106], [106] * No synthesis paper was found for this method. 2.4.1 Sol-Gel synthesis As explained in table 2.5, Sol-Gel synthesis is a derivation of hydrolysis where solution turns into a gel-foam during evaporation as sample condensation leads to bridging bonds of the reactive, forming a macromolecule. After a heat treatment is followed to eliminate any organic material obtaining the desired powder. Schemed in figure 2.37. This process allows for a better particle morphology and size control over common hydrolysis as metal ions isolation avoid agglomeration [86]. This production method is regarded as a simple process but allows a homogeneous controlled composition of final NPs, although expensive precursors are often required. Sol-gel is a common chemical route for producing metal oxide nanostructures, or with further processing, ceramic materials, in this process metal alkoxides or inorganic metal salts (chloride, nitrate, etc.) are dissolved in water or alcohol together with a jellification agent, to later obtain the gel by stirring and heating. Low temperature required for the process allows for a good control on particles composition and fits this process for large industrial applications [107]. 28 2.37 Figure: Sol-gel stages. This synthesis method started with studies on silica gels and colloidal suspensions in the mid-1800s, by J. Ebelman [108] and T. Graham [109]. A good early review on the subject can be found in Larry L. and Hench L. work [110], where they include the basic steps for this process: mixing, formation of a colloidal suspension by stirring with the required pH to avoid particles precipitation; gelation, colloidal particles and jellification agent link together to become a three-dimensional network molecule, this step can be process while casting or coating to achieve different products; aging, increases the thickness of interparticle necks to obtain required structure resistance and stability; drying or syneresis, liquid extraction from the network, during this step large capillary stresses can develop causing gel crack catastrophically unless surfactant or critical conditions are applied. This set is commonly followed by chemical stabilization and densification process, which are not required for particles obtention, neither is the aging step. The drying step is the heart of the process as leads to different production methods depending on the evaporation rate and applied time. Three common routes with different objectives are normally used: supercritical drying of formed gel, avoiding liquid-gas transition by applying critical conditions, reaches aero-gel formations, highly porous and extremely low density materials; thermal drying, standard evaporation of the liquid solution by heat, looks for gels cracking during moisture extraction to obtain a dry pre-material for further processing; and freeze drying, which apply void and low temperatures, aiming cryo-gels, porous materials with high interconnectivity [107]. For nanoparticles production the second route is applied, followed commonly by washing and heat treatment to obtain final chemical composition. Hematite sol-gel process There are several sol-gel routes for hematite nanoparticles spread in the literature, some literature examples are gathered in table 2.6. Although literature mentions alkoxides as main precursors for sol-gel process, this is due they are proven to be successful resources for metals as Si, Al or Zr, but not for others, mainly for being expensive, sensitive or directly no commercially available products. For iron, ferric nitrates seems one of the most standard precursors although some research can be found on other origins as ferric chloride salts [111]. 2.6 Table: Various literature Sol-gel routes for hematite synthesis. Precursor Jellification Agent Solvent Conditions Post process annealing Reference Ferric nitrate Sodium bicarbonate Water Room temperature, pH 8 200-500°C [112] Ferric nitrate Citric acid Water 70°C 150-530°C [91] Ferric nitrate Sodium hydroxide Water 75°C, pH 11 500°C [93] Byproducts of the steel industry Ammonium hydroxide Ethylene glycol, Citric acid, 140°C, basic pH 300-1200°C [90] Ferric nitrate 1,2-ethanediol 0.2 M Acetic Acid 70°C, dried at 100°C 500-1000°C [113] Iron acetate Oxalic acid Ethanol, Water 65°C 400°C [114] Iron chloride Propylene oxide Propanol, ethanol, water Supercritical point - [111] 29 Selected route for hematite NPs production in this project follows previous team research and knowledge. Cepero Mejias M. master thesis [115], reports the application of iron III nitrate nonahydrate, Fe(NO3)3 + 9H2O, as precursor and citric acid, C6H8O7, as chelating agent in an aqueous solution to obtain 100 nm submicrometric particles. This process is based on Zhang J. work where this technique was successfully applied for hematite nanoparticles production [91]. Citric acid is a largely available and low cost green crosslinker applied widely as gel agent for hydrogel fabrication in pharmaceutical industry [116]. Its potential for sol-gel NPs fabrication, also studied in medical applications [117], derives from its characteristic as a crosslink agent, which its hydrophilic nature prevents the particles agglomeration during evaporation forming co-crystal-layered structures [118]. Link capability of citric acid with metal ions, or their hydroxyl species, comes from the bonding to the carboxylic groups (R−COOH) and the hydroxyl group (R-OH) presence in citric acid, which creates weak soluble bonds boosting the negative particle charge preventing agglomeration [118]. Correlation between citric acid, metal ions, and water determinates the gel formation process and therefore nanoparticles size. Further studies of this process are still required but some research can be found on the effect of citric acid concentration effects [118]. Sol-gel hematite doping One of the main advantages of the sol-gel process is its application for doping materials, as this fabrication system can be applied to multiple metal oxides, doping is achieved simple by choosing equivalent or compatible fabrication routes for the different materials to be added, only requiring a weight calculation base on objective composition ratios. Sol-gel allows small doping quantities as resources solution and particles isolation during gel formation led to a homogeneous distributed composition. Aluminum oxide synthesis by sol-gel is compatible with iron nitrate selected route as aluminum nitrate can be used as precursor for the process following similar steps as can be seen in Cepero’s work [115], or in Khan U. paper on iron and aluminum nitrate co-sol-gel application [119]. Additional papers can be found on aluminum nitrate usage in different sol-gel synthesis [120], [121] In opposition to aluminum for titanium addition a relatively more complex route is required. Titanium sol-gel process commonly are based on alkoxides as, Ti(OC2H5), Ti(OC3H7)4, or Ti(OC4H9) [122] which can be dissolves in an alcohol or an acid base, this means a previous solving step is required for the addition of this precursors into an aqueous solution, as the one used for iron nitrate, exampled in sol-gel routes apply for irondoped titanium dioxide applications [78], [123]. Papers on aqueous route iron-titanium sol-gel can be also found however they apply other precursors such as Iron Chloride [124]. 2.5 Hematite Additives As previously mentioned, the potential of iron as an OC is limited by its low reactivity after consecutive cycles, its deactivation. Three main process relates to this issue, carbon deposition, pore sintering, and iron slow reduction kinetics (Fe2O3 → Fe3O4 → FeO → Fe). This last challenge is particularly significant when attempting to fully reduce hematite to metallic iron, especially in the interior regions of the sample, which is often referred to as deep reduction. The issue arises from the limited oxygen diffusion within the material. Up to 18 metals are found in literature to be available as iron dopants (Li, Na, Mg, Ca, K, Sc, Ti, V, Cr, Mn, Co, Ni, Cu, Zn, Y, Zr, La y Ce). From those two Ti and Al stands as main contenders. Al, as AlO2, applied as structural promoter. Ti, as TiO2, enhances oxygen diffusion. Alumina has emerged as the most promising inert structural promoter for hematite [125], [126]. In these cases, alumina functions as an inert compound with high thermal stability, effectively limiting iron mobility and thereby preventing pore sintering. However, its role as a catalyst for coke formation restricts its application to an optimal range of 2% to 5% by weight [24]. 30 The role of titanium is more ambiguous. While there is extensive literature on hematite doping with titanium for photocatalytic water splitting, studies on hematite OCs doped with TiO2 are scarce [127]. Moreover, existing research often involves high amounts of TiO2, up to 40% by weight. However, this addition has been proved to enhance vacancy formation, leading to higher porosity of doped hematite samples, obtaining an increase on oxygen diffusivity. This dopant is meant to enhance hematite reactivity. In both cases, iron can react with the added compounds to form complex phases, such as FeTiO3 and Fe2TiO with titanium, or FeAl2O4 with aluminum. These phases can lead to lower reactivity (Ro) as different reactions become involved in the reduction-oxidation cycle. Further studies are needed to understand the effects of hematite doping, particularly at low doping agent concentrations of up to 5% by weight. Aswell no literature has been found on the fabrication of hematite nanoparticles doped with titanium and aluminum oxides. 2.6 Hematite foam The interest in porous materials, those with intentionally integrated pores in their structure, is largely driven by their wide-ranging applications across industries, including medical implants, heat exchangers, solid cathodes, gas and liquid transport, mufflers… [128]. In this project, the emphasis is on sponge-like porous materials, where the porosity is not neatly divided into well-defined cells. This structure leads to a significantly larger surface area compared to other geometries, making these materials particularly valuable as potential catalysts, especially in flow-through applications. Multiple fabrication routes exist for creating porous metals, such as traditional powder metallurgy (PM), air injection into molten metals, or direct foaming. However, these methods often struggle with limited control over pore size and morphology. To address this, freeze casting has emerged as a promising solution, originally developed for forming porous ceramics [129]. Freeze casting exploits the freezing and sublimation of the liquid medium in a powder suspension to form a solid where the pore structure mirrors the ice crystal morphology formed during freezing. By selecting the appropriate medium and freezing conditions, pore morphology can be controlled. This is exemplified by unidirectional freezing, where gradual solidification leads to the formation of a sponge-like material with predominantly aligned pores, figure 2.38. 2.38 Figure: Directional freeze casting scheme. Freeze casting enables the fabrication of hematite foams for the production of SIP catalysts, as demonstrated in Lloreda P’s thesis work [50], with an example shown in figure 2.39. While this technique allows control over pore morphology by adjusting the freezing conditions and medium, effective control is primarily achievable in small samples with simple geometries, where the temperature gradient can be accurately predicted. This limitation poses challenges to scaling up the process for larger or more complex structures. 31 2.39 Figure: Alumina doped hematite sponge fabricated by freeze casting [130]. The hematite foam shown in figure 2.29 was developed using a hematite camphene-polystyrene-stearic acid dispersion. The process of obtaining this dispersion for freeze casting relies on previously described theories, such as colloid formation, rheology, debinding, and nanoparticle synthesis. This knowledge can be directly applied to Direct Ink Writing (DIW). DIW offers the advantage of creating geometries where both macro and micro porosity can be utilized, as demonstrated by simple mesh or grid patterns, illustrated in figure 2.40. However, for DIW micro pore morphology is tied to the debinding and sintering process, and therefore this must be tidily followed to understand the results obtain and select a proper method. 2.40 Figure: Scheme comparing traditional casting shape to DIW additive manufacturing grid pattern shape. A prior colleague's work focused on the production of hematite materials using DIW. In this case, Castro F. developed a camphene-based ink composed of 10% hematite powder by volume, 5% polystyrene (PS) by volume, and 3% stearic acid by weight relative to the powder [131]. With this material a base grid geometry was achieved, example in the result in figure 2.41, settling the bases for this project. 2.41 Figure: Hematite-camphene grid developed by DIW. 38 Calcination For the heat treatment the dry gel is placed in a vitrified alumina dish and set inside the calcination oven. This process is performed open in air at atmospheric pressures aiming metals complete oxidation, complete elimination of any organic compound left, and the recrystallization of hematite from amorphous iron oxyhydroxide (FeOOH) undesired sub-product of gel formation [134]. Followed heat treatment applies a 1°C/min warming speed and two plateaus: first one, at 350°C for two hours, aims organics calcination; second and main one, at 620°C for eight hours, looks for sample complete hematite formation. Calcination diagram is introduced for better visual of the process, figure 3.8. Cooling process follows natural cooling of the oven until 150°C, when sample can be taken out for faster cooling. A sample photo after this stage is introduced in figure 3.9. 3.8 Figure: Sol-gel followed calcination diagram. 3.9 Figure: Hematite calcinated gel. Humid ball milling To crack down the obtained calcinated gel, from it agglomerated stage into and homogenous fine powder, it is wet milled in distilled water in a 100 mm diameter (5 mm deep) rotatory mill filled with stainless steel 3mm balls, according to a 1:10 weight sample to balls proportion, and the volume of water matching the volume of balls. A 24 hour milling is considered appropriate to obtain desire homogenization and dispersion. Photography of applied instruments is gathered in figure 3.10. 39 3.10 Figure: 10 cm stainless steel ball milling drum. Once the milling is finished an additional step, balls separation, is required to extract the sample from the mill container. Washing over a 1 litter beaker with distilled water and ethanol is proceeded, a stainless-steel mesh avoids milling balls to fall in the beaker, as shown in figure 3.11. This step is critical to avoid production loss. 3.11 Figure: Set-up for sample recovery by washing after ball milling. Solution precipitation Natural precipitation ff the solution inside a beaker, proceeded for 24 hours wait, allows the separation of the main volume of washing liquids to the particles, figure 3.12. Straight drying without this step leads to nitrates leftovers in the samples, washing content removal allows these impurities, dissolved in the liquid, to go away. Particles loss still suspended after 24 hours on the liquid is minimum and was measured by 2h centrifugation at 4000 rpm and heat evaporation which resulted in less than 0.1 grams of solid residue from the 20g aimed. (a) (b) (c) 3.12 Figure: Solution precipitation. (a) Initial solution, (b) after 24 hours, (c) water separation. 40 Powder drying After the main volume of washing liquid is removed, sample is set to dry at 90°C inside a sand bath for 3 hours. Slow evaporation is necessary as higher temperatures lead to bubbles formation and splashing, which contaminates the equipment and results in the loss of material. Dry particles resulting from this process can be considered the final product. Nonetheless, an additional hand grinding step allows for a better powder look, as agglomerates separate, but does not interfere in the particles properties and application. Additionally, final powder is weighted, labelled and storage. Common weight recovery of the complete process sets around 97%. To achieve the desired quantities an extra 4% weight on initial resources is enough. Photography of the final powder is shown as result for this process, figure 3.13. 3.13 Figure: Hematite doped with aluminum and titanium process with sol-gel technique. 3.1.3 Hematite particles functionalization Most of the improvements developed are directly included in the standard procedure, as result of early or minor changes, an example of those are: the avoiding of the solution splashing during the drying which led to low temperature slow evaporation; the inclusion of the precipitation step to decrease sample contamination with nitrates; or the selection of proper beakers and equipment for the process. However, a major change aimed, not yet included under the standard measured procedure, is the functionalization of the particles. This addition looked to improve the colloidal behavior of the particles reaching a stearic stability. For this purpose, stearic acid was set as dispersant, already introduced in section 2.3.3. The functionalization of the particles was proceeded during the ball milling process, in this stage water was substituted by ethanol with a 3%w/W of stearic acid addon, table 3.4, based on the final hematite weight. After 24 hours milling a washing process was applied to recover the sample, only in this case just ethanol was applied. 3.4 Table: Stearic acid origin. Name Description Formula Molecular Weight [g/mol] Quality [%w/w] Distributor Stearic Acid Soft waxy white solid in dust shape. Natural fattyacid found in many animal and vegetable fats. CH3(CH2) 16CO2H 284.48 98 Alfa Aesar Due to ethanol increase usage and its higher value compared to distilled water a recovery step was added after the precipitation. Rotative vacuum equipment allowed the purification of the separated ethanol and the recovery of the left ethanol mix with the particles, which leads to the dry final product. Photography of the equipment during the process is shown in figure 3.14. 41 3.14 Figure: Left, milled hematite powder in ethanol poured in desired flask. Right, sample under vacuum drying. 3.2 Hematite ink development The ink development process did not achieve full standardization due to the ongoing adjustments required to address challenges in storage and handling during printing. Therefore, this section will first introduce the initial fabrication method, followed by a discussion of the issues encountered, and finally, the proposed current method along with its improvements. 3.2.1 Ink constituents A shared aspect between both the initial and current methods is the use of similar constituents for ink fabrication. As outlined in the introduction table 1.3, and subsection 2.3.3, the ink development for this project is based on a camphene/NPs suspension with an added adhesive agent, PS. The sources for these materials are listed in table 3.5 In addition to the materials mentioned, the NPs powder introduced in the previous section, along with stearic acid, table 3.4, were also used for the ink preparation. 3.5 Table: Ink constituents’ sources. Name Description Formula Molecular Weight [g/mol] Quality [%w/w] Distributor Camphene Colorless, crystalline organic compound with a strong odor, commonly used as a solvent or additive in various industrial processes. It is derived from natural sources like turpentine. C10H16 136.238 98 Alfa Aesar Polystyrene White pellet shaped synthetic thermoplastic polymer made from the monomer styrene. [C8H8]x 192.000 100 SigmaAldrich As the project evolved different compositions were applied, for this reason a spreadsheet for the weight required for each constituent was developed, figure 3.15. 42 3.15 Figure: Spreadsheet for the obtention of the ink constituents weight requirements. 3.2.2 Ink initial fabrication route The initial route for ink fabrication selected ball milling as a mixing process, process schemed in figure 3.16. This method was chosen to achieve a well-homogenized ink and was originally developed by two colleagues, as referenced in subsection 2.6. This proceeding will be briefly introduced to provide context for the subsequent changes made to improve it. 3.16 Figure: Ball milling scheme for ink fabrication based on a container and two rollers. This ball milling procedure was designed to produce small quantities of ink, ranging from 10 to 20 ml, sufficient for printing applications. As no equipment was available for this type of fabrication, the team developed a custom one by modifying an incubator with a pair of rollers, figure 3.17. This system enabled the continuous rotation of glass containers used for the required volumes, figure 3.18, maintaining a controlled temperature throughout the process. To obtain the ink, the selected container is initially filled with 3mm stainless steel balls following a 5:1 ratio to the ink's final weight and the required amount of camphene. The container is then rotated at 60ºC for 30 minutes to warm the balls and melt the camphene. Afterward, the weighted PS is incorporated, and the mixture is rotated for 6 hours to ensure complete dissolution of the PS. Following this, the needed stearic acid is added, and the system is kept rotating for an additional 30 minutes to homogenize the mixture. With the organic base ready, the hematite NPs are incorporated, and the colloid is ball-milled for an extra 12 hours to ensure a homogeneous ink. Desired volume Constituents Final weights Desired composition 43 3.17 Figure: Modified incubator for ink ball milling and rotating container. 3.18 Figure: 20and 30-ml glass bottles. The final ink is a red liquid with low viscosity that solidifies below 60ºC. The typical composition for this preparation method ranges from 5 to 15% volume ratio of powder. Higher particle concentrations lead to a viscous system that stack the milling balls, preventing proper mixing. Additionally, the printing process showed clogging issues with these higher concentrations, indicating improper blending. This was later confirmed as the final fabrication method allowed for higher hematite doses without these symptoms. Before printing, handling the ink begins with preheating the necessary syringe and two stainless steel blunt dispensing needles, a 10-gauge needle for extracting the liquid from the glass container and a 16to 18-gauge needle for printing, see figure 3.19. This preheating is done inside the fabrication incubator for at least 30 minutes. Afterward, the syringe with the 10-gauge needle is used to extract the ink. Since the extraction occurs outside the heated environment, it must be done quickly. As an additional precaution, the filled syringe, now attached to the printing needle, is placed back in the incubator to warm for another 30 minutes. Any leftover ink is kept rotating inside the chamber. A final consideration for handling the ink is the preheating to 60ºC of the printing equipment to prevent the ink from solidifying. Additionally, the syringe should be transported quickly between the incubator and the printer, which should be positioned close to one another to avoid any cooling of the ink during the transfer. 44 3.19 Figure: Stainless steel blunt dispensing needles, 8to 30-gauge. Some inconveniences were encountered with this process. The primary issue was ink loss: the milling balls prevented the complete extraction of the ink, as it tended to adhere to the balls rather than settling at the bottom of the container. This resulted in the syringe drawing air bubbles instead of ink. Secondly, for large production or long-term storage, the continuous rotation caused the ink to darken, likely due to contamination from the milling balls' black oxide blending with the ink. Additionally, the ink's solidification led to irreversible changes, making it more difficult to print with it after. Therefore, storing the ink at room temperature was not suitable, and handling it required the previously mentioned precautions. These inconveniences necessitated both modifications to the ink composition and the development of a new fabrication route. 3.2.3 Ink modification with HDDA The modification of the ink composition looked to allow its usage under room conditions, this meant that no longer the liquid to solid transformation during printing could really on a heat solidification process, and rather, it had to depend on the rheology of the ink as introduced in section 2.2. To achieve this, it was proposed to increase the powder volume ratio enhancing the shear thinning behavior together with the physical behavior of camphene by its modification with HDDA (1,6-Hexanediol diacrylate), table 3.6, which allows to maintain camphene liquid at room temperature. 3.6 Table: HDDA constituent origin. Name Description Formula Molecular Weight [g/mol] Quality [%w/w] Distributor 1,6-Hexanediol diacrylate Plastic smelling translucent liquid. HDD is difunctional acrylate ester monomer used in the manufacture of polymers C12H8O4 226.27 80 SigmaAldrich To assess the influence of HDDA, nine samples with 15% volume of powder and varying concentrations of HDDA and PS were studied using thermogravimetric analysis. This also provided insights into the impact of PS on the solidification process. The concentrations and results of these experiments are presented in table 3.7. 45 3.7 Tab: Main temperatures summary of the thermogravimetric analysis of Hematite-Camphene-HDDA-PS inks. 1st cycle 2nd cycle Sample HDDA [%v/V] PS [%v/V] Melting [ºC] Solidification [ºC] Melting [ºC] Solidification [ºC] 1 5 1 36,3 30,9 35,9 30,6 2 10 1 32,9 26,2 33,8 26,8 3 15 1 18,5 10,6 18,9 11,6 4 5 1,5 37,0 32,7 (29,0) 36,4 31,3 (28,4) 5 10 1,5 35,8 34,3 (27,0) 35,0 31,7 (28,1) 6 15 1,5 24,7 16,3 (14,0) 22,3 15,7 (13,8) 7 5 2 37,0 32,7 (28,8) 35,0 31,7 (28,5) 8 10 2 32,4 28,5 (26,4) 33,1 27,9 (26,9) 9 15 2 24,6 18,5 (17,4) 25,4 18,1 (17,3) The influence of HDDA was evident, as it lowered the natural melting point of camphene from 52°C to approximately 35°C at 5% concentration, and approximately 18°C at 15%. Additionally, the presence of PS increased both the melting and solidification points. Notably, small amounts of PS (over 1.5%) caused a twostage solidification, with PS solidifying at a higher temperature. Based on this analysis, the optimal composition was determined to be 5% HDDA and 1% PS, which lowered the operation temperature to 40°C while avoiding the prolonged solidification process induced by PS. HDDA addition in this quantities did not complete allowed room temperature handling but lowered the operation temperature both allowing an easier temperature control on the printer and a less rough solidification process for the ink. 3.2.4 Ink final fabrication route Considering the aforementioned challenges, a new approach was devised for producing a homogeneous ink. Magnetic stirring replaced ball milling as the blending method, allowing the syringe needle to access more ink due to the smaller volume occupied by the magnet compared to steel balls, and eliminating contamination from the milling balls. To maintain the desired temperature during stirring, a water bath was introduced between the hot plate and the inner beaker containing the ink, ensuring uniform temperature control as depicted in figure 3.20. To prevent camphene sublimation, the inner beaker enclosing the ink was sealed tightly, and periodic weight measurements were conducted to compensate for any lost camphene throughout the fabrication process. 3.20 Figure: Ink fabrication scheme based on a water hot bath set on a magnetic stirrer. Two fabrication routes were developed for this purpose: the first involves higher temperatures and longer processing times, while the second allows decrease both parameters with the addition of acetone as a solvent. Both methods will be briefly introduced below. The first method follows a procedure similar to the ball milling approach. Initially, camphene is added and melted under continuous rotation at 60ºC. The temperature is then increased to 80ºC to aid in dissolving PS in camphene, which is stirred for 4 hours. Subsequently, HDDA is added and stirred while the temperature is gradually lowered to 40ºC. At this temperature, stearic acid is introduced and allowed to mix for 30 minutes. Once the organic medium is prepared, hematite powder is slowly added to prevent clustering. The ink is then stirred for an additional two hours to achieve the desired homogeneity. In the second route for this process, acetone was introduced as a solvent. To start the water bath is set at 50ºC. Then acetone is poured inside the beaker, looking to add the same volume of the aimed ink. PS was introduced 46 and stirred for 3 hours to ensure complete dissolution. The bath temperature was then lowered to 40ºC to incorporate camphene, HDDA, and stearic acid, followed by 30 minutes of mixing. After preparing the organic solution, at room temperature, hematite powder was slowly added. Then the ink was left to dry until the acetone was fully removed, obtaining a solid ink. When remelting for usage compensating for any camphene lost through sublimation can be added. Although this process did not significantly reduce the time required to obtain the ink, as the drying process compensates for any win time, the larger liquid volume facilitated better blending of the powder. Equivalent to the initial route a red ink was obtained, in this scenario the viscosity increased as larger amounts of powder can be incorporated, with successful productions of ink reaching 35% of powder volume. Photography of an ongoing mixing sample is found in figure 3.21. 3.21 Figure: Photography of hematite/camphene ink mixing by new route. 3.2.5 Ink final composition The initial fabrication route and prior experience sets the initial ink composition: 10% v/V hematite, 5% v/V PS (80,000 g/cm3 MW), 85% v/V camphene (referred to as 10Fe85Camp5PS), along with the addition of 3% stearic acid relative to the powder weight. The modifications introduced with the incorporation of HDDA and the improved homogenization by the new fabrication route, produced a less viscous ink during extrusion. A lower viscosity with a softer solidification process, due to the reduction in the printing temperature gradient from 70–25ºC (without HDDA) to 40–25ºC, led to a not printable ink as its lost its structural retention capability. This meant a modification on the composition was necessary to achieve the ink printability again. Helping to obtain higher viscosities, the new fabrication route allowed higher powder concentrations. Therefore, inks moving from 15to 30-% v/V were decided to be studied. Increasing the powder content alone would naturally lead to higher viscosities, potentially counteracting the observed reduction. To further balance this, an increase in the binder percentage was tested. Raising the PS concentration to 10% v/V led to longer solidification periods, in line with the expected trend, even though only up to 2% v/V PS had been previously studied in the presence of HDDA. Additionally, two higher molecular weight PS samples were tested: one with a molecular weight of 192,000 g/mol and another with 350,000 g/mol. Inks containing 15% hematite and 10% PS (192,000 g/mol) were successfully printed. However, as anticipated, the prolonged solidification resulted in the printed part being malleable rather than forming a fragile solid, which adversely affected the lyophilization process (see subsection 3.4.1) and led to bubble formation on the sample, as shown in figure 3.22. Using the 350,000 g/mol PS resulted in inks with poor homogeneity, where visible polymer strings formed, and the increased stickiness compromised printing accuracy by causing the ink to follow the nozzle instead of depositing correctly. 47 3.22 Figure: Samples 15Fe75Camp10PS, bauble formation during lyophilization. As a result of this experiment, 192000 g/mol PS was selected but concentrations would only apply to a 5%v/V of it. Also, different powder concentrations were still suitable, and 15-20-30%v/V of hematite will be shown in the results section. 3.3 DIW development The application and development of the DIW process is one of the cores of this project. This section means to present the starting point, with the system first stage and the initiation into the usage of the system, which will bring the context to understand later modifications. To finalize it will include the system current stage, path design and expected results of the DIW system. 3.3.1 DIW initial system The initial DIW system, developed in previous team projects, is shown in figures 3.23 and 3.24. Built on an Ender-3 Creality printer, the system involved modifications to both the printer head and the deposition bed. The extrusion head consisted of a drilled aluminum body attached to the printer's heating block, with a 2 mm nozzle to fit both 16and 18-gauge needles. The syringe, mounted to the system by two ears, was connected to an 8mm 8T screw (typically found in the vertical axis of 3D printers), which was driven by a Creality 42-40 step motor. The bed was also modified with a cooling system that kept it at a temperature range of 10 to 15ºC. 3.23 Figure: Initial DIW systems, printer extruder head close up. 10 mm 10 mm 54 A future analysis of the printing conditions by consecutive printing for a final geometry and ink formulation for the catalyst could allow to adjust these values to achieve better accuracy, mainly in layer width as mentioned above. This analysis could also ably print at higher speeds. 3.3.6 Green samples Green samples are the result of the DIW printing process. In this step the part still fresh and harden as camphene sublimates. Depending on the composition they can be still malleable when printed and after complete hardening they are still fragile. This means careful handling of green samples is required. Jet, camphene sublimation starts during the printing process and continuous on after it is finish meaning that green part weight defers as time passes, defaulting the traceability of the process. For better handling a weight paper was attached to the magnetic bed by placing magnets in each corner, this allows the sample to be easily recovered just by grabbing of the paper, figure 3.35. This paper also allows a lower first layer over spreading as it absorbs part of the ink. Weighting papers were selected as they have a low and standard weight with a glazed smooth surface. Detachment from the paper is only possible after freeze drying post-processing. 3.35 Figure: Green parts handling with paper bed. 3.3.7 Path design As printing progress multiples path design were tested, at the beginning cylinders with different grid designs were meant, but those geometries result in big inaccuracies due to big changes in direction and small angles found between the grid design and the outer circle wall, leading to holes clogging or other imperfections. This is also a result enhanced in low viscosity inks where capillarity forces move the material out of the nozzles path as it is deposited to areas where material was already placed. An example of inaccuracy is shown in figure 3.36. 3.36 Figure: Shape difference from path design to sample. Path design does not represent layer real width and height programmed. 55 Multiples path design varying grid distance to the circle, grid amplitude or frequency, as example in figure 3.37, were tested but none showed promising results. Trying to avoid capillarity forces, layer complete solidification between layers was set as a priority, for these different bridges design was tried, however result did not accomplish aimed printing quality, example in figure 3.38. 3.37 Figure: Examples of different circle grid layouts. 3.38 Figure: Bridge approach to assure layer solidification. Path design, uncut and cut green sample. Seeing the easy cutting of the green sample, as grid continuous design led to better results a molde cutting techniques was thought to be a viable route, figure 3.39, testing gave proper results, and different samples were printed this way, example in figure 3.40. 3.39 Figure: Molde cutting to obtain samples from a printed grid. 56 3.40 Figure: Sample obtained by mold cutting after DIW. Fitting path design and ink composition modification in the same experiment meant result could not bet completely associated to any decision, so for further testing design was switch to a traditional square grid, leaving design modification to be a future study over a settle formula and fabrication process. To finish this subsection an example of phyton code necessary and the generated GCode by that program are found, respectively, in annex B and C. 3.4 Samples post-process Samples post-processing looks to obtain a final hematite solid part, for these two steps where selected, freeze drying for medium debinding and thermal treatment for final sintering. 3.4.1 Freeze drying Freeze drying or lyophilization is the process of sublimation of compound by reducing both temperature and pressure looking to avoid the transition through liquid, as liquid evaporation tends to elevate pressure forming crack in the sample leading to worst quality and material final mechanical properties, a failure relates to camphene evaporation is found in sample photography in figure 3.36. Camphene which easily sublimates at room temperature allows an easy lyophilization, this treatment is performed during 24h at -30ºC and 0.01 mpa. Lyophilization equipment is found in figure 3.41. The presence of PS in the samples maintains its structure, accomplishing its role as a binder. The debinding process of camphene easy the sintering. 3.41 Figure: Lyophilization equipment: Telstar LyoQuest HT40. 57 Brown samples, samples denomination after debinding, are then photography and weighted in this stage as no shrinkage or aspect change is detected in this step and samples are dry and hardened making them easier to handle. Precaution is still required as samples are still fragile. 3.4.2 Sintering The sintering process looks to reach two aimed temperatures, 300ºC, where polymer calcination happens finishing the debinding process, and 1100ºC, at which hematite sintering process concludes. At both temperatures a precaution time are given to ensure debinding and sintering properly conclude. The thermal treatment follows the chart in figure 3.42. This treatment is proceeded with the equipment shown in figure 3.43. 3.42 Figure: Thermal treatment for hematite brown samples sintering. 3.43 Figure: Muffle furnace. Naberterm L9/14. During this process samples reduce a little due to the polymer gases formation, this gives sintered samples a metallic look, as shown by figures 3.44 and 3.45 comparation. During this step shrinkage is expected, which deeply depends on sample geometry and powder proportion. This shrinkage is meant to be studied next to samples densification over the sintering. 58 3.44 Figure: Extruded hematite filament by DIW system, brown stage. 3.45 Figure: Extruded hematite filament by DIW system, sintered. 5 mm 5 mm 59 4 RESULTS The first result of this work it the successful fabrication of hematite doped both with titanium and aluminum. Through scanning electron microscopy (SEM) and later image analysis it was determinate that particles size range 10-150 µm, showing an irregular shape with rough surface regardless the chemical composition. Powder samples sintered at 1100ºC, following and equivalent treatment as described in the post processing section, were studied to determinate obtained crystallography, which resulted in dominant Fe2O3 hematite structure with no other phase detected in general, meaning solid solutions of the samples were achieved. Low levels of doping (under 5% in volume) avoided the formation of other stoichiometric complex oxides. Further results on the studies of these powders are meant to be presented in an article still in the process of being published. Objectives two to four are condensate in the following result. The continuous improvement of the ink and the system allowed the printing of higher ink composition up to a success sample with a 30%v/V of hematite. This sample was properly sintered and measured showing a 50% volume decrement over the hematite volume proportion which meant inner and outer porosity was expected. The analysis with Arquimides showed an outer porosity around 25%v/V of the volume with an inner porosity of just 0.5%v/V. Therefore, this sample, figures 4.1 and 4.2, accomplish our aim of obtaining a good material with high density which porosity was open incrementing free surface area. Also, it must be considered that this porosity did not account for the macroporosity obtained by the printing grid design. 4.1 Figure: Success hematite sample, 30Fe65Camp5PS, green sample 4.2 Figure: Success hematite sample, 30Fe65Camp5PS, brown and sintered sample. 5 mm 5 mm 60 4.1 Table: Success hematite sample, 30Fe65Camp5PS, weight and geometry measurement follow up. Stage Weight [g] Width [mm] Length [mm] Height [mm] Thickness [mm] Volume* [cm3] Brown 1.39 12.80 12.60 5.00 1.30 0.61 Sintered 1.27 10.40 10.10 4.00 1.00 0.31 Decrement 8.29% 18.75% 19.84% 20.00% 23.08% 49.53% *Obtained by geometrical parameters. 4.2 Table: Success hematite sample, 30Fe65Camp5PS, Arquimides test, four attempts. First Second Third Fourth Unit Average Dry weight 1.27 1.27 1.27 1.27 g 1.27 Saturated weight 1.45 1.35 1.32 1.37 g 1.36 Submerged weight 1.02 1.02 1.02 1.03 g 1.02 Volume 0.43 0.33 0.31 0.34 cm3 0.34 Bulk density 2.94 3.84 4.15 3.74 g/cm3 3.79 Open porosity volume 0.18 0.08 0.05 0.09 cm3 0.09 Open porosity ratio 41,44% 24,34% 16,60% 27,34% 25,84% Theory density 5.16 5.16 5.16 5.16 g/cm3 5.16 Theory volume 0.25 0.25 0.25 0.25 cm3 0.25 Close porosity volume 0,19 0,09 0,06 0,09 cm3 0,09 Close porosity ratio 0,65% 0,43% 0,91% 0,03% 0,54% To back the sample result SEM photography was applied to study different samples cut, figures from 4.3 to 4.5. SEM photography shows how density increases with powder density, giving almost full density when 30%v/V of powder is reached, as expressed by Arquimedes test. When intermediate concentration is applied, 20%v/V, higher density is shown in the sample surface, meaning gasses are trapped during sintering. Inner porosity is homogeneous once lower concentrations are reached, 15%v/V. This meant concentration could be selected depending on the final peace aim. Samples for other concentrations, 15and 20-%v/V did not accomplish a proper printing, and just fragments where recovered and sintered for the SEM study. The testing for further results in this investigation line was stopped as other parallel projects were showing better results, and therefore time and material resources were moved from this line. This discussion will be continued in the conclusion. 61 4.3 Figure: SEM image of 15%v/V hematite sintered sample. Progressive zoom from first to third image. 400 µm 20 µm 100 µm x200 x800 x4000 62 4.4 Figure: SEM image of 20%v/V hematite sintered sample. Progressive zoom from first to third image. 200 µm 100 µm 40 µm x400 x1000 x2000 63 4.5 Figure: SEM image of 30%v/V hematite sintered sample. Progressive zoom from first to third image. 100 µm 50 µm 40 µm x600 x1000 x2000 70 Energy, vol. 19, no. 5, pp. 467–475, 1977, doi: 10.1016/0038-092X(77)90102-5. [18] A. Messerschmitt, “Process of producing hydrogen,” 971206, Sep. 27, 1910 [19] M. Damizia, “Pure hydrogen production by chemical looping technology: use of iron as redox element and bioethanol as renewable reductant,” PhD in Chemical Processes for the Industry and the Environment: XXXV cycle, Sapienza University of Rome, Rome, 2023. [20] J. R. Scheffe and A. Steinfeld, “Oxygen exchange materials for solar thermochemical splitting of H2O and CO2: a review,” Materials Today, vol. 17, no. 7, pp. 341–348, Sep. 2014, doi: 10.1016/j.mattod.2014.04.025. [21] M. Luo et al., “Review of hydrogen production using chemical-looping technology,” Renewable and Sustainable Energy Reviews, vol. 81, pp. 3186–3214, Jan. 2018, doi: 10.1016/j.rser.2017.07.007. [22] Z. Yu et al., “Iron-based oxygen carriers in chemical looping conversions: A review,” Carbon Resources Conversion, vol. 2, no. 1, pp. 23–34, Apr. 2019, doi: 10.1016/j.crcon.2018.11.004. [23] L. Protasova and F. Snijkers, “Recent developments in oxygen carrier materials for hydrogen production via chemical looping processes,” Fuel, vol. 181, pp. 75–93, Oct. 2016, doi: 10.1016/j.fuel.2016.04.110. [24] M. Damizia, M. P. Bracciale, F. Anania, L. Tai, P. De Filippis, and B. de Caprariis, “Efficient utilization of Al2O3 as structural promoter of Fe into 2 and 3 steps chemical looping hydrogen process: Pure H2 production from ethanol,” Int J Hydrogen Energy, Apr. 2023, doi: 10.1016/j.ijhydene.2023.04.067. [25] D. Cebrucean, V. Cebrucean, I. Ionel, and H. Spliethoff, “Performance of two iron-based syngasfueled chemical looping systems for hydrogen and/or electricity generation combined with carbon capture,” Clean Technol Environ Policy, vol. 19, no. 2, pp. 451–470, Mar. 2017, doi: 10.1007/s10098-016-1231-y. [26] S. Chen, Z. Xue, D. Wang, and W. Xiang, “Hydrogen and electricity co-production plant integrating steam-iron process and chemical looping combustion,” Int J Hydrogen Energy, vol. 37, no. 10, pp. 8204–8216, May 2012, doi: 10.1016/j.ijhydene.2012.02.098. [27] F. Li, L. Zeng, and L.-S. Fan, “Techno-Economic Analysis of Coal-Based Hydrogen and Electricity Cogeneration Processes with CO 2 Capture,” Ind Eng Chem Res, vol. 49, no. 21, pp. 11018–11028, Nov. 2010, doi: 10.1021/ie100568z. [28] P. Gupta, L. G. Velazquez-Vargas, and L.-S. Fan, “Syngas Redox (SGR) Process to Produce Hydrogen from Coal Derived Syngas,” Energy & Fuels, vol. 21, no. 5, pp. 2900–2908, Sep. 2007, doi: 10.1021/ef060512k. [29] H. Nazir et al., “Is the H2 economy realizable in the foreseeable future? Part II: H2 storage, transportation, and distribution,” Int J Hydrogen Energy, vol. 45, no. 41, pp. 20693–20708, Aug. 2020, doi: 10.1016/j.ijhydene.2020.05.241. [30] M. El Hage et al., “Bioethanol Production from Woody Biomass: Recent Advances on the Effect of Pretreatments on the Bioconversion Process and Energy Yield Aspects,” Energies (Basel), vol. 16, no. 13, p. 5052, Jun. 2023, doi: 10.3390/en16135052. [31] H. Danninger, R. de Oro Calderon, and C. Gierl‐Mayer, “Powder Metallurgy and Sintered Materials,” in Ullmann’s Encyclopedia of Industrial Chemistry, Wiley, 2017, pp. 1–57. doi: 10.1002/14356007.a22_105.pub2. [32] Peisheng Liu and Guo-Feng Chen, Porous Materials: Processing and Applications. Elsevier, 2014. [33] K. Wang et al., “Advances on direct selective laser printing of ceramics: An overview,” J Alloys Compd, vol. 975, p. 172821, Feb. 2024, doi: 10.1016/j.jallcom.2023.172821. [34] M. A. S. R. Saadi et al., “Direct Ink Writing: A 3D Printing Technology for Diverse Materials,” Advanced Materials, vol. 34, no. 28, Jul. 2022, doi: 10.1002/adma.202108855. 71 [35] A. Shahzad and I. Lazoglu, “Direct ink writing (DIW) of structural and functional ceramics: Recent achievements and future challenges,” Compos B Eng, vol. 225, p. 109249, Nov. 2021, doi: 10.1016/j.compositesb.2021.109249. [36] J. A. Lewis, J. E. Smay, J. Stuecker, and J. Cesarano, “Direct Ink Writing of Three‐Dimensional Ceramic Structures,” Journal of the American Ceramic Society, vol. 89, no. 12, pp. 3599–3609, Dec. 2006, doi: 10.1111/j.1551-2916.2006.01382.x. [37] J. E. Smay, J. Cesarano, and J. A. Lewis, “Colloidal Inks for Directed Assembly of 3-D Periodic Structures,” Langmuir, vol. 18, no. 14, pp. 5429–5437, Jul. 2002, doi: 10.1021/la0257135. [38] J. E. Smay, G. M. Gratson, R. F. Shepherd, J. Cesarano, and J. A. Lewis, “Directed colloidal assembly of 3D periodic structures,” Advanced Materials, vol. 14, no. 18, pp. 1279–1283, Sep. 2002, doi: 10.1002/1521-4095(20020916)14:18<1279::AID-ADMA1279>3.0.CO;2-A. [39] A. Gleadall, “FullControl GCode Designer: Open-source software for unconstrained design in additive manufacturing,” Addit Manuf, vol. 46, p. 102109, Oct. 2021, doi: 10.1016/j.addma.2021.102109. [40] Anton Paar GmbH, “Basics of rheology,” https://wiki.anton-paar.com/en/basics-ofrheology/#rotational-tests. [41] Faith A. Morrison, “Rheometry,” Apr. 27, 2018, Department of Chemical Engineering, Michigan Technological University. [42] S. Guo, F. Gosselin, N. Guerin, A. Lanouette, M. Heuzey, and D. Therriault, “Solvent‐Cast Three‐Dimensional Printing of Multifunctional Microsystems,” Small, vol. 9, no. 24, pp. 4118– 4122, Dec. 2013, doi: 10.1002/smll.201300975. [43] K. B. Manning et al., “Self Assembly–Assisted Additive Manufacturing: Direct Ink Write 3D Printing of Epoxy–Amine Thermosets,” Macromol Mater Eng, vol. 304, no. 3, Mar. 2019, doi: 10.1002/mame.201800511. [44] S. Chandrasekaran, E. B. Duoss, M. A. Worsley, and J. P. Lewicki, “3D printing of high performance cyanate ester thermoset polymers,” J Mater Chem A Mater, vol. 6, no. 3, pp. 853– 858, 2018, doi: 10.1039/C7TA09466C. [45] T. An, K.-T. Hwang, J.-H. Kim, and J. Kim, “Extrusion-based 3D direct ink writing of NiZnferrite structures with viscoelastic ceramic suspension,” Ceram Int, vol. 46, no. 5, pp. 6469–6476, Apr. 2020, doi: 10.1016/j.ceramint.2019.11.127. [46] Z. Ye et al., “Study on 3D-Direct Ink Writing based on adding silica submicron-particles to improve the rheological properties of alumina ceramic ink,” Mater Today Commun, vol. 28, p. 102534, Sep. 2021, doi: 10.1016/j.mtcomm.2021.102534. [47] C. F. Revelo and H. A. Colorado, “3D printing of kaolinite clay ceramics using the Direct Ink Writing (DIW) technique,” Ceram Int, vol. 44, no. 5, pp. 5673–5682, Apr. 2018, doi: 10.1016/j.ceramint.2017.12.219. [48] K. Huang, H. Elsayed, G. Franchin, and P. Colombo, “Embedded direct ink writing of freeform ceramic components,” Appl Mater Today, vol. 23, p. 101005, Jun. 2021, doi: 10.1016/j.apmt.2021.101005. [49] L. del-Mazo-Barbara and M.-P. Ginebra, “Rheological characterisation of ceramic inks for 3D direct ink writing: A review,” J Eur Ceram Soc, vol. 41, no. 16, pp. 18–33, Dec. 2021, doi: 10.1016/j.jeurceramsoc.2021.08.031. [50] P. J. Lloreda, J. Directores, : Ranier, S. Ferrer, and A. P. Escolano, “Fabricación de materiales porosos a partir de polvos nanométricos y submicrométricos mediante la técnica de solidificación direccional,” Tesis Doctoral Ciencia y Tecnología de Nuevos Materiales, Escuela Técnica Superior de Ingeniería Universidad de Sevilla, Sevilla, 2022. [51] J. J. Stickel and R. L. Powell, “Fluid mechanics and rheology of dense suspensions,” Annu Rev Fluid Mech, vol. 37, no. 1, pp. 129–149, Jan. 2005, doi: 72 10.1146/annurev.fluid.36.050802.122132. [52] D. T. N. Chen, Q. Wen, P. A. Janmey, J. C. Crocker, and A. G. Yodh, “Rheology of Soft Materials,” Annu Rev Condens Matter Phys, vol. 1, no. 1, pp. 301–322, Aug. 2010, doi: 10.1146/annurev-conmatphys-070909-104120. [53] Franck A., “Understanding Rheology of Structured Fluids,” 2024. Accessed: Jan. 11, 2024. [Online]. Available: https://www.tainstruments.com/pdf/literature/AAN016_V1_U_StructFluids.pdf [54] S. Mueller, E. W. Llewellin, and H. M. Mader, “The rheology of suspensions of solid particles,” Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 466, no. 2116, pp. 1201–1228, Apr. 2010, doi: 10.1098/rspa.2009.0445. [55] Z. An, Y. Zhang, Q. Li, H. Wang, Z. Guo, and J. Zhu, “Effect of particle shape on the apparent viscosity of liquid–solid suspensions,” Powder Technol, vol. 328, pp. 199–206, Apr. 2018, doi: 10.1016/j.powtec.2017.12.019. [56] S. Mueller, E. W. Llewellin, and H. M. Mader, “The effect of particle shape on suspension viscosity and implications for magmatic flows,” Geophys Res Lett, vol. 38, no. 13, p. n/a-n/a, Jul. 2011, doi: 10.1029/2011GL047167. [57] G. Trefalt and M. Borkovec, “Overview of DLVO Theory,” 2014. [Online]. Available: www.colloid.ch/dlvo [58] B. Derjaguin, “A theory of interaction of particles in presence of electric double-layers and the stability of lyophobe colloids and disperse systems,” Acta Phys. Chim., vol. 10, pp. 333–346, 1939. [59] B. Derjaguin and L. D. Landau, “Theory of the stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solutions of electrolytes,” Acta Phys. Chim., vol. 14, pp. 633–662, 1941. [60] E. J. W. Verwey and J. T. G. Overbeek, Theory of Stability of Lyophobic Colloids. Elsevier Amsterdam, 1948. [61] Frank Babick, Suspensions of Colloidal Particles and Aggregates, vol. 20. Berlin: Springer, 2012. [62] A. Nikolov, J. Lee, and D. Wasan, “DLVO surface forces in liquid films and statistical mechanics of colloidal oscillatory structural forces in dispersion stability,” Adv Colloid Interface Sci, vol. 313, p. 102847, Mar. 2023, doi: 10.1016/j.cis.2023.102847. [63] Gregor Trefalt and Michal Borkovec, “Overview of DLVO theory,” http://www.colloid.ch/index.php?name=dlvo. [64] Ashley N. Mapile, Erik Svensson Grape, and Carl K. Brozek, “Solvation of Nanoscale Materials,” ChemRxiv, 2024. [65] R. G. dos Santos, Fundamentals of Surface Thermodynamics. Cham: Springer International Publishing, 2024. doi: 10.1007/978-3-031-52466-0. [66] B. W. Ninham, “On progress in forces since the DLVO theory,” Adv Colloid Interface Sci, vol. 83, no. 1–3, pp. 1–17, Dec. 1999, doi: 10.1016/S0001-8686(99)00008-1. [67] H. N. W. Lekkerkerker, R. Tuinier, and M. Vis, Colloids and the Depletion Interaction, vol. 1026. Cham: Springer International Publishing, 2024. doi: 10.1007/978-3-031-52131-7. [68] J. A. Carpenter, N. Passaleva, M. Häring, G. Mikl, and A. R. Studart, “3D Printing of Hierarchical Porous Steel and Iron‐Based Materials,” Adv Mater Technol, vol. 8, no. 3, Feb. 2023, doi: 10.1002/admt.202200971. [69] Z. Li, Y. Li, B. Shi, D. Tang, Y. Wang, and L. Hao, “Dual gradient direct ink writing of functional geopolymer-based carbonyl-iron/graphene composites for adjustable broadband microwave absorption,” Ceram Int, vol. 48, no. 7, pp. 9277–9285, Apr. 2022, doi: 73 10.1016/j.ceramint.2021.12.114. [70] Z. Lotfizarei, A. Mostafapour, A. Barari, A. Jalili, and A. E. Patterson, “Overview of debinding methods for parts manufactured using powder material extrusion,” Addit Manuf, vol. 61, p. 103335, Jan. 2023, doi: 10.1016/j.addma.2022.103335. [71] C. Xu, S. Yu, W. Wu, Q. Liu, and L. Ren, “Direct ink writing of Fe bone implants with independently adjustable structural porosity and mechanical properties,” Addit Manuf, vol. 51, p. 102589, Mar. 2022, doi: 10.1016/j.addma.2021.102589. [72] C. Xu et al., “Direct ink writing of porous Fe–Ti6Al4V and Fe-Inconel 718 bimetallic structures,” Journal of Materials Research and Technology, vol. 25, pp. 6613–6627, Jul. 2023, doi: 10.1016/j.jmrt.2023.07.103. [73] S. L. Taylor, A. E. Jakus, R. N. Shah, and D. C. Dunand, “Iron and Nickel Cellular Structures by Sintering of 3D‐Printed Oxide or Metallic Particle Inks ,” Adv Eng Mater, vol. 19, no. 11, Nov. 2017, doi: 10.1002/adem.201600365. [74] C. Xu, B. Quinn, L. L. Lebel, D. Therriault, and G. L’Espérance, “Multi-Material Direct Ink Writing (DIW) for Complex 3D Metallic Structures with Removable Supports,” ACS Appl Mater Interfaces, vol. 11, no. 8, pp. 8499–8506, Feb. 2019, doi: 10.1021/acsami.8b19986. [75] R. Bayaniahangar, S. Bayani Ahangar, Z. Zhang, B. P. Lee, and J. M. Pearce, “3-D printed soft magnetic helical coil actuators of iron oxide embedded polydimethylsiloxane,” Sens Actuators B Chem, vol. 326, p. 128781, Jan. 2021, doi: 10.1016/j.snb.2020.128781. [76] Y.-W. Moon, I.-J. Choi, Y.-H. Koh, and H.-E. Kim, “Macroporous alumina scaffolds consisting of highly microporous hollow filaments using three-dimensional ceramic/camphene-based coextrusion,” J Eur Ceram Soc, vol. 35, no. 16, pp. 4623–4627, Dec. 2015, doi: 10.1016/j.jeurceramsoc.2015.08.017. [77] Q. Meng, Z. Wang, X. Chai, Z. Weng, R. Ding, and L. Dong, “Fabrication of hematite (α-Fe 2 O 3 ) nanoparticles using electrochemical deposition,” Appl Surf Sci, vol. 368, pp. 303–308, Apr. 2016, doi: 10.1016/j.apsusc.2016.02.007. [78] S. Bagheri, C. K. G, and S. Bee Abd Hamid, “Generation of Hematite Nanoparticles via Sol-Gel Method,” 2013. [Online]. Available: www.isca.in [79] Wikipedia, “Iron Oxides,” https://en.wikipedia.org/wiki/Iron_oxide. [80] J. Lian, X. Duan, J. Ma, P. Peng, T. Kim, and W. Zheng, “Hematite (α-Fe2O3 ) with Various Morphologies: Ionic Liquid-Assisted Synthesis, Formation Mechanism, and Properties,” ACS Nano, vol. 3, no. 11, pp. 3749–3761, Nov. 2009, doi: 10.1021/nn900941e. [81] B. K. Ozcelik and C. Ergun, “Synthesis and characterization of iron oxide particles using spray pyrolysis technique,” Ceram Int, vol. 41, no. 2, pp. 1994–2005, Mar. 2015, doi: 10.1016/j.ceramint.2014.09.103. [82] J. Yu, X. Yu, B. Huang, X. Zhang, and Y. Dai, “Hydrothermal Synthesis and Visible-light Photocatalytic Activity of Novel Cage-like Ferric Oxide Hollow Spheres,” Cryst Growth Des, vol. 9, no. 3, pp. 1474–1480, Mar. 2009, doi: 10.1021/cg800941d. [83] H. Wan et al., “Advanced hematite nanomaterials for newly emerging applications,” Chem Sci, vol. 14, no. 11, pp. 2776–2798, 2023, doi: 10.1039/D3SC00180F. [84] P. P. Sarangi, B. Naik, and N. N. Ghosh, “Low temperature synthesis of single-phase α-Fe2O3 nano-powders by using simple but novel chemical methods,” Powder Technol, vol. 192, no. 3, pp. 245–249, Jun. 2009, doi: 10.1016/j.powtec.2009.01.002. [85] M. Khalil, N. Liu, and R. L. Lee, “Synthesis and characterization of hematite nanoparticles using ultrasonic sonochemistry method,” International Journal of Technology, vol. 4, pp. 582–590, 2017, doi: 10.14716/ijtech.v8i4.7285. [86] A. Ibrahim and B. A. Abubakar, “Some wet routes for synthesis of hematite nanostructures,” African Journal of Pure and Applied Chemistry, vol. 7, no. 3, pp. 114–121, Mar. 2013, doi: 74 10.5897/AJPAC12.002. [87] K. Wongwailikhit and S. Horwongsakul, “The preparation of iron (III) oxide nanoparticles using W/O microemulsion,” Mater Lett, vol. 65, no. 17–18, pp. 2820–2822, Sep. 2011, doi: 10.1016/j.matlet.2011.05.063. [88] C.-A. Chiu, K. D. Hristovski, R. Dockery, K. Doudrick, and P. Westerhoff, “Modeling temperature and reaction time impacts on hematite nanoparticle size during forced hydrolysis of ferric chloride,” Chemical Engineering Journal, vol. 210, pp. 357–362, Nov. 2012, doi: 10.1016/j.cej.2012.08.093. [89] W. Wang, J. Y. Howe, and B. Gu, “Structure and Morphology Evolution of Hematite (α-Fe2O3) Nanoparticles in Forced Hydrolysis of Ferric Chloride,” The Journal of Physical Chemistry C, vol. 112, no. 25, pp. 9203–9208, Jun. 2008, doi: 10.1021/jp800683j. [90] M. A. Vargas, J. E. Diosa, and E. Mosquera, “Data on study of hematite nanoparticles obtained from Iron(III) oxide by the Pechini method,” Data Brief, vol. 25, p. 104183, Aug. 2019, doi: 10.1016/j.dib.2019.104183. [91] J. Zhang, L. X. Rong, Y. Liu, and B. Z. Dong, “SAXS study on the microstructure of Fe2O3 nanocrystal,” 2003. [Online]. Available: www.elsevier.com/locate/msea [92] M. Farahmandjou and F. Soflaee, “Synthesis and characterization of α-Fe2O3 nanoparticles by simple co-precipitation method,” Physical Chemistry Research, vol. 3, no. 3, pp. 191–196, 2015, doi: 10.22036/pcr.2015.9193. [93] M. Tahir, M. Fakhar-e-Alam, M. Atif, G. Mustafa, and Z. Ali, “Investigation of optical, electrical and magnetic properties of hematite α-Fe2O3 nanoparticles via sol-gel and co-precipitation method,” J King Saud Univ Sci, vol. 35, no. 5, p. 102695, Jul. 2023, doi: 10.1016/j.jksus.2023.102695. [94] N. M. Hosny, N. Nowesser, A. S. Al Hussaini, and M. S. Zoromba, “Solid State Synthesis of Hematite Nanoparticles from Doped Poly o-aminophenol (POAP),” J Inorg Organomet Polym Mater, vol. 26, no. 1, pp. 41–47, Jan. 2016, doi: 10.1007/s10904-015-0284-4. [95] C. Diaz et al., “Solvent-less method for efficient photocatalytic α-Fe 2 O 3 nanoparticles using macromolecular polymeric precursors,” New Journal of Chemistry, vol. 40, no. 8, pp. 6768– 6776, 2016, doi: 10.1039/C6NJ00561F. [96] N. Kumar et al., “Molecular Insights of Oxidation Process of Iron Nanoparticles: Spectroscopic, Magnetic, and Microscopic Evidence,” Environ Sci Technol, vol. 48, no. 23, pp. 13888–13894, Dec. 2014, doi: 10.1021/es503154q. [97] E. Darezereshki, “One-step synthesis of hematite (α-Fe2O3) nano-particles by direct thermaldecomposition of maghemite,” Mater Lett, vol. 65, no. 4, pp. 642–645, Feb. 2011, doi: 10.1016/j.matlet.2010.11.030. [98] A. Hassanjani-Roshan, M. R. Vaezi, A. Shokuhfar, and Z. Rajabali, “Synthesis of iron oxide nanoparticles via sonochemical method and their characterization,” Particuology, vol. 9, no. 1, pp. 95–99, Feb. 2011, doi: 10.1016/j.partic.2010.05.013. [99] M. Tadić, N. Čitaković, M. Panjan, Z. Stojanović, D. Marković, and V. Spasojević, “Synthesis, morphology, microstructure and magnetic properties of hematite submicron particles,” J Alloys Compd, vol. 509, no. 28, pp. 7639–7644, Jul. 2011, doi: 10.1016/j.jallcom.2011.04.117. [100] Q. Meng, Z. Wang, X. Chai, Z. Weng, R. Ding, and L. Dong, “Fabrication of hematite (α-Fe2O3 ) nanoparticles using electrochemical deposition,” Appl Surf Sci, vol. 368, pp. 303–308, Apr. 2016, doi: 10.1016/j.apsusc.2016.02.007. [101] J. Toniolo, A. S. Takimi, M. J. Andrade, R. Bonadiman, and C. P. Bergmann, “Synthesis by the solution combustion process and magnetic properties of iron oxide (Fe3O4 and α-Fe2O3) particles,” J Mater Sci, vol. 42, no. 13, pp. 4785–4791, Jul. 2007, doi: 10.1007/s10853-006-07637. 75 [102] K. Suresh and K. C. Patil, “A combustion process for the instant synthesis of γ-iron oxide,” J Mater Sci Lett, vol. 12, no. 8, pp. 572–574, Jan. 1993, doi: 10.1007/BF00278328. [103] J. S. Nyarige, T. P. J. Krüger, and M. Diale, “Structural and optical properties of hematite and Larginine/hematite nanostructures prepared by thermal spray pyrolysis,” Surfaces and Interfaces, vol. 18, p. 100394, Mar. 2020, doi: 10.1016/j.surfin.2019.100394. [104] T. Jurkin, M. Gotić, G. Štefanić, and I. Pucić, “Gamma-irradiation synthesis of iron oxide nanoparticles in the presence of PEO, PVP or CTAB,” Radiation Physics and Chemistry, vol. 124, pp. 75–83, Jul. 2016, doi: 10.1016/j.radphyschem.2015.11.019. [105] H. Karami, “Synthesis and characterization of iron oxide nanoparticles by solid state chemical reaction method,” J Clust Sci, vol. 21, no. 1, pp. 11–20, Mar. 2010, doi: 10.1007/s10876-0090278-x. [106] J. Ding, T. Tsuzuki, and P. G. McCormick, “Hematite powders synthesized by mechanochemical processing,” Nanostructured Materials, vol. 8, no. 6, pp. 739–747, Sep. 1997, doi: 10.1016/S0965-9773(97)00221-3. [107] D. Bokov et al., “Nanomaterial by Sol-Gel Method: Synthesis and Application,” Advances in Materials Science and Engineering, vol. 2021, pp. 1–21, Dec. 2021, doi: 10.1155/2021/5102014. [108] J. J. Ebelmen, “Recherches sur les combinaisons des acides borique et silicique avec les éthers,” Annale de Chimie et de Physique, 1846. [109] T. Graham, “On the properties of silicic acid and other analogous colloidal substances,” J Chem Soc, vol. 17, pp. 318–327, 1864. [110] L. L. Hench and J. K. West, “The sol-gel process,” Chem Rev, vol. 90, no. 1, pp. 33–72, Jan. 1990, doi: 10.1021/cr00099a003. [111] A. E. Gash, T. M. Tillotson, J. H. Satcher, J. F. Poco, L. W. Hrubesh, and R. L. Simpson, “Use of Epoxides in the Sol−Gel Synthesis of Porous Iron(III) Oxide Monoliths from Fe(III) Salts,” Chemistry of Materials, vol. 13, no. 3, pp. 999–1007, Mar. 2001, doi: 10.1021/cm0007611. [112] R. RT, A.-A. SD, K. HH, and M. HS, “Preparation and Characterization of Hematite Iron Oxide (alpha-Fe2O3) by Sol-Gel Method,” Chem Sci J, vol. 09, no. 04, 2018, doi: 10.4172/21503494.1000197. [113] O. Opuchovic and A. Kareiva, “Historical hematite pigment: Synthesis by an aqueous sol–gel method, characterization and application for the colouration of ceramic glazes,” Ceram Int, vol. 41, no. 3, pp. 4504–4513, Apr. 2015, doi: 10.1016/j.ceramint.2014.11.145. [114] R. Ilmetov, “Photocatalytic activity of hematite nanoparticles prepared by sol-gel method,” Mater Today Proc, vol. 6, pp. 11–14, 2019, doi: 10.1016/j.matpr.2018.05.069. [115] M. Cepero Mejías, “Síntesis de polvos nanométricos de óxido de hierro para el uso del hidrógeno como energía limpia,” Trabajo Fin Máster, Máster en Ingeniería Industrial, Escuela Técnica Superior de Ingeniería, Universidad de Sevilla (US), Sevilla, 2021. [116] S. Nangare, Y. Vispute, R. Tade, S. Dugam, and P. Patil, “Pharmaceutical applications of citric acid,” Futur J Pharm Sci, vol. 7, no. 1, p. 54, Feb. 2021, doi: 10.1186/s43094-021-00203-9. [117] S. Khan et al., “Antimicrobial activity of citric acid functionalized iron oxide nanoparticles – Superparamagnetic effect,” Ceram Int, vol. 46, no. 8, pp. 10942–10951, Jun. 2020, doi: 10.1016/j.ceramint.2020.01.109. [118] Utari, H. Maulidina, R. Arilasita, H. Widiyandari, Suharno, and B. Purnama, “Citric acid concentration tune of structural and magnetic properties in hematite (α−Fe2O3) nanoparticles synthesized by sol−gel method,” Mater Res Express, vol. 10, no. 3, p. 036101, Mar. 2023, doi: 10.1088/2053-1591/acbf0c. [119] U. Khan, A. Akbar, H. Yousaf, S. Riaz, and S. Naseem, “Ferromagnetic Properties of Al-doped Fe2O3 Thin Films by Sol-gel,” Mater Today Proc, vol. 2, no. 10, pp. 5415–5420, 2015, doi: 10.1016/j.matpr.2015.11.061. 76 [120] I. Y. Y. Bu, “Sol–gel production of aluminium doped zinc oxide using aluminium nitrate,” Mater Sci Semicond Process, vol. 27, pp. 19–25, Nov. 2014, doi: 10.1016/j.mssp.2014.06.011. [121] M. I. F. Macêdo, C. C. Osawa, and C. A. Bertran, “Sol-Gel Synthesis of Transparent Alumina Gel and Pure Gamma Alumina by Urea Hydrolysis of Aluminum Nitrate,” J Solgel Sci Technol, vol. 30, no. 3, pp. 135–140, Jun. 2004, doi: 10.1023/B:JSST.0000039497.46154.8f. [122] S. G. Ullattil and P. Periyat, “Sol-Gel Synthesis of Titanium Dioxide,” 2017, pp. 271–283. doi: 10.1007/978-3-319-50144-4_9. [123] C. L. Luu, Q. T. Nguyen, and S. T. Ho, “Synthesis and characterization of Fe-doped TiO2 photocatalyst by the sol–gel method,” Advances in Natural Sciences: Nanoscience and Nanotechnology, vol. 1, no. 1, p. 015008, Mar. 2010, doi: 10.1088/2043-6254/1/1/015008. [124] D. V. Wellia, Q. C. Xu, M. A. Sk, K. H. Lim, T. M. Lim, and T. T. Y. Tan, “Experimental and theoretical studies of Fe-doped TiO2 films prepared by peroxo sol–gel method,” Appl Catal A Gen, vol. 401, no. 1–2, pp. 98–105, Jul. 2011, doi: 10.1016/j.apcata.2011.05.003. [125] J. Adánez, L. F. de Diego, F. García-Labiano, P. Gayán, A. Abad, and J. M. Palacios, “Selection of Oxygen Carriers for Chemical-Looping Combustion,” Energy & Fuels, vol. 18, no. 2, pp. 371–377, Mar. 2004, doi: 10.1021/ef0301452. [126] B. Wang, R. Yan, D. H. Lee, Y. Zheng, H. Zhao, and C. Zheng, “Characterization and evaluation of Fe2O3/Al2O3 oxygen carrier prepared by sol–gel combustion synthesis,” J Anal Appl Pyrolysis, vol. 91, no. 1, pp. 105–113, May 2011, doi: 10.1016/j.jaap.2011.01.010. [127] T. Mattisson, M. Johansson, and A. Lyngfelt, “Multicycle Reduction and Oxidation of Different Types of Iron Oxide ParticlesApplication to Chemical-Looping Combustion,” Energy & Fuels, vol. 18, no. 3, pp. 628–637, May 2004, doi: 10.1021/ef0301405. [128] L. ‐P. Lefebvre, J. Banhart, and D. C. Dunand, “Porous Metals and Metallic Foams: Current Status and Recent Developments,” Adv Eng Mater, vol. 10, no. 9, pp. 775–787, Sep. 2008, doi: 10.1002/adem.200800241. [129] S. Deville, “Freeze‐Casting of Porous Ceramics: A Review of Current Achievements and Issues,” Adv Eng Mater, vol. 10, no. 3, pp. 155–169, Mar. 2008, doi: 10.1002/adem.200700270. [130] M. Damizia, P. J. Lloreda-Jurado, P. De Filippis, B. de Caprariis, E. Chicardi, and R. Sepúlveda, “Green hydrogen production using doped Fe2O3 foams,” Int J Hydrogen Energy, vol. 51, pp. 834–845, Jan. 2024, doi: 10.1016/j.ijhydene.2023.09.008. [131] Francisco Castro Zafra, “Fabricación aditiva. Impresión de tinta con base de canfeno y partículas NiO,” US, Sevilla, 2022. [132] R. Velmurugan, B. Krishnakumar, R. Kumar, and M. Swaminathan, “Solar active nano-TiO2 for mineralization of Reactive Red 120 and Trypan Blue,” Arabian Journal of Chemistry, vol. 5, no. 4, pp. 447–452, Oct. 2012, doi: 10.1016/j.arabjc.2010.12.023. [133] B. Krishnakumar, A. Balakrishna, C. T. Arranja, C. M. F. Dias, and A. J. F. N. Sobral, “Chemically modified amino porphyrin/TiO2 for the degradation of Acid Black 1 under day light illumination,” Spectrochim Acta A Mol Biomol Spectrosc, vol. 176, pp. 134–141, Apr. 2017, doi: 10.1016/j.saa.2017.01.019. [134] Y. W. Phuan, M. N. Chong, T. Zhu, S.-T. Yong, and E. S. Chan, “Effects of annealing temperature on the physicochemical, optical and photoelectrochemical properties of nanostructured hematite thin films prepared via electrodeposition method,” Mater Res Bull, vol. 69, pp. 71–77, Sep. 2015, doi: 10.1016/j.materresbull.2014.12.059. 77 78 Annex List A. Production of Fe2O3 particles doped with Al2O3 and TiO2 by citric acid sol-gel hydrolysis. B. Phyton code for the generation of a geometry GCode for DIW with FullControl. C. GCode generated for DIW, square grid samples. 79 Annex A 86 Annex B import fullcontrol as fc from math import tau, sin from OLD.GRID import Circle_Grid from fullcontrol import Point design_name = 'Square_Grid' # Name of G-Code save_steps = False # Save steps of the design save_GCode = True # Save G_Code add_var_values = False # Expand name with main parameters (Go to line 180) ask_name = False # Use when you want to specify the name each time # ----------PRINTER PARAMETHERES class Print_parameters: def __init__(self): # PRINTING PARAMETERS self.nozzle_temp = 40 # Temperature of the nozzle (normal 70ºC) self.print_speed = 150 # Movement speed if printing (normal 400 mm/s) self.travel_speed = 400 # Movement speed if NOT printing (normal 550 mm/s) self.dia_feed = 12 # 1 layer feed diameter (normal 16 mm) ! Not a real dimension self.dia_feed_step = 0 # Consecutive layers feed diameter step up self.dia_feed_delta = 0 # Progressive layers feed diameter increase self.Extrusion_width = 1 # Extrusion width [e_units] ! Not a real dimension self.Extrusion_high = 0.3 # Extrusion height [e_units] ! Not a real dimension self.fan_percent = 0 # % Power for fun # LAYER PARAMETERS self.N_layers = 40 # Number of layers self.dz_offset = 0.15 # Extrusion offset [e_units] self.dz_2_layer = 0.3 # 2nd Layer high self.dz_x_layer = 0.3 # Layer high !Closer to reality self.rotation_angle = 0 # Rotation angle for each layer # AJUSTMENT PARAMETERS self.x_center = 89 # (114+64)/2 # X center position [e_units] self.y_center = 120 # (170+70)/2 # Y center position [e_units] self.z_center = 0 # Z center position [e_units] = Bed level self.bed_width = 50 # Bed width [e_units] self.bed_high = 100 # Bed high [e_units] printer_name = 'ultimaker2plus' # Select Initial and End G-Code PP = Print_parameters() # ----------INITAL MOVEMENT # Define basic points for FullControl center = fc.Point(x=PP.x_center, y=PP.y_center, z=PP.z_center) corner = fc.Point(x=PP.x_center - PP.bed_width / 2, y=PP.y_center - PP.bed_high / 2) bed_level = fc.Point(z=PP.z_center) steps = [] steps.append(fc.ManualGcode(text='\n;START DESIGN')) # Comment steps.append(fc.Point(z=PP.z_center + 10)) steps.append(corner) # Move to down-left corner point steps.append(bed_level) # Move down to bed level steps.append(fc.Extruder(on=True)) # Start printing # ----------DESIGN 87 Distance = 3 # [mm] layer1_steps = [] layer1_steps.append(fc.Point(x=-Distance * 2, y=-Distance * 2, z=0)) layer1_steps.append(fc.Point(x=-Distance * 2, y=Distance * 2, z=0)) layer1_steps.append(fc.Point(x=-Distance * 1, y=Distance * 2, z=0)) layer1_steps.append(fc.Point(x=-Distance * 1, y=-Distance * 2, z=0)) layer1_steps.append(fc.Point(x=0, y=-Distance * 2, z=0)) layer1_steps.append(fc.Point(x=0, y=Distance * 2, z=0)) layer1_steps.append(fc.Point(x=Distance * 1, y=Distance * 2, z=0)) layer1_steps.append(fc.Point(x=Distance * 1, y=-Distance * 2, z=0)) layer1_steps.append(fc.Point(x=Distance * 2, y=-Distance * 2, z=0)) layer1_steps.append(fc.Point(x=Distance * 2, y=Distance * 2, z=0)) layer2_steps = [] layer2_steps.append(fc.Point(x=Distance * 2, y=Distance * 2, z=0)) layer2_steps.append(fc.Point(x=-Distance * 2, y=Distance * 2, z=0)) layer2_steps.append(fc.Point(x=-Distance * 2, y=Distance * 1, z=0)) layer2_steps.append(fc.Point(x=Distance * 2, y=Distance * 1, z=0)) layer2_steps.append(fc.Point(x=Distance * 2, y=0, z=0)) layer2_steps.append(fc.Point(x=-Distance * 2, y=0, z=0)) layer2_steps.append(fc.Point(x=-Distance * 2, y=-Distance * 1, z=0)) layer2_steps.append(fc.Point(x=Distance * 2, y=-Distance * 1, z=0)) layer2_steps.append(fc.Point(x=Distance * 2, y=-Distance * 2, z=0)) layer2_steps.append(fc.Point(x=-Distance * 2, y=-Distance * 2, z=0)) # ----------INITIALS LAYERS layer1_steps = fc.move(layer1_steps, center) layer2_steps = fc.move(layer2_steps, center) steps += layer1_steps + fc.move(layer2_steps, fc.Vector(z=PP.dz_2_layer)) # ----------REPETITIVE LAYERS for i in range(1, PP.N_layers + 1): layerX = fc.move(layer1_steps, fc.Vector(z=PP.dz_x_layer * 2 * i)) layerX += fc.move(layer2_steps, fc.Vector(z=PP.dz_x_layer * (2 * i + 1))) steps += layerX # ----Design ends here steps.append(fc.Extruder(on=False)) # Stop printing # ----------OFFSET steps = fc.move(steps, fc.Vector(z=PP.dz_offset)) # ----------GENERATE GCODE SETTINGS if add_var_values: design_name += '_n_' + str(PP.N_layers + 2) \ + '_z_' + str(round(PP.z_center + PP.dz_offset, 3)) \ + '_dz_' + str(round(PP.dz_x_layer, 2)) \ + '_f_' + str(PP.dia_feed) \ + '_v_' + str(PP.print_speed) # + '_a_' + str(round(alpha / tau * 360, 2)) \ # + '_R_' + str(R_Ext) \ # + '_T_' + str(nozzle_temp) if ask_name: design_name = input('What sedign name would you like? ') gcode_controls = fc.GcodeControls( printer_name='custom', save_as=design_name, initialization_data={ 'bed_temp': 0, 'fan_percent': 0, 'print_speed': PP.print_speed, 'travel_speed': PP.travel_speed, 'nozzle_temp': PP.nozzle_temp, 'extrusion_width': PP.Extrusion_width, 'extrusion_height': PP.Extrusion_high, 'e_units': 'mm', 88 'relative_e': True, 'dia_feed': PP.dia_feed}) # ---------- CUSTOM INITIAL GCODE starting_steps = [] starting_steps.append(fc.ManualGcode(text=open('./InitialGCode/' + printer_name + '.txt', 'r').read())) # ---------- CUSTOM END GCODE ending_steps = [] ending_steps.append(fc.ManualGcode(text=open('./EndGCode/' + printer_name + '.txt', 'r').read())) # ---------- SAVE if save_steps: save('steps_' + design_name, steps) if save_GCode: gcode = fc.transform(starting_steps + steps + ending_steps, 'gcode', gcode_controls) 89 Annex C ; Time to print!!!!! ; GCode created with FullControl - tell us what you're printing! ; [email protected] or tag FullControlXYZ on Twitter/Instagram/LinkedIn/Reddit/TikTok M83 ; relative extrusion M140 S0 ; set bed temp and continue M104 S40 ; set hotend temp and continue M190 S0 ; set bed temp and wait M109 S40 ; set hotend temp and wait M106 S0 ; set fan speed ;----- ; STARTING PROCEDURE ;----- M82 ;absolute extrusion mode G28 ;Home ;Prime the extruder G92 E0 G1 F200 E0 G92 E0 M83 ;relative extrusion mode ;----- ; MAIN CODE ;----- ;START DESIGN G0 F40 Z10.15 G0 X64. Y70.0 G0 Z0.15 G1 F150 X83.0 Y114.0 E0.12713 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z0.45 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z0.75 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 90 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z1.05 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z1.35 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z1.65 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z1.95 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z2.25 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z2.55 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z2.85 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 91 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z3.15 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z3.45 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z3.75 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z4.05 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z4.35 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z4.65 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z4.95 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 92 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z5.25 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z5.55 E0.00079577 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z5.85 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z6.15 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z6.45 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z6.75 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z7.05 E0.000795775 93 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z7.35 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z7.65 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z7.95 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z8.25 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z8.55 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z8.85 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 94 G1 X83.0 E0.031831 G1 Z9.15 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z9.45 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z9.75 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z10.05 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z10.35 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z10.65 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z10.95 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 95 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z11.25 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z11.55 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z11.85 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z12.15 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z12.45 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 G1 X83.0 E0.031831 G1 Y117.0 E0.00795775 G1 X95.0 E0.031831 G1 Y114.0 E0.00795775 G1 X83.0 E0.031831 G1 Z12.75 E0.000795775 G1 Y126.0 E0.031831 G1 X86.0 E0.00795775 G1 Y114.0 E0.031831 G1 X89.0 E0.00795775 G1 Y126.0 E0.031831 G1 X92.0 E0.00795775 G1 Y114.0 E0.031831 G1 X95.0 E0.00795775 G1 Y126.0 E0.031831 G1 Z13.05 E0.000795775 G1 X83.0 E0.031831 G1 Y123.0 E0.00795775 G1 X95.0 E0.031831 G1 Y120.0 E0.00795775 Equation Chapter 1 Section 1 Resumen trabajo Fin de Máster Ingeniería Industrial Desarrollo y mejora del proceso de fabricación DIW de piezas de hematita porosa. Autor: Miguel Paúl Navarrón Tutor: Ranier Enrique Sepúlveda Ferrer Dpto. de Ciencia de Materiales Escuela Técnica Superior de Ingeniería Universidad de Sevilla Sevilla, 2024 iii Índice Índice ............................................................................................................................................................ iii Palabras Clave ............................................................................................................................................... iv 1 Introducción .......................................................................................................................................... 1 2 Sintetización .......................................................................................................................................... 3 3 Fabricación ............................................................................................................................................ 5 4 Impresión .............................................................................................................................................. 6 5 Post-procesado ...................................................................................................................................... 8 6 Resultados ............................................................................................................................................. 8 7 Conclusión ........................................................................................................................................... 10 Palabras Clave Hidrogeno, Hematita, Fabricación de Materiales. Direct Ink Writting v 1 INTRODUCCIÓN En un mundo con muchos problemas a solucionar desde la ingeniería, se aboga por la diversidad en tecnologías y no a las soluciones únicas. Sin embargo, siendo el transporte de energía el motor del desarrollo de la sociedad, este se encuentra limitado entre la electricidad y los combustibles fósiles. Bajo este contexto y con la vista puesta en los conflictos geopolíticos, surge el hidrógeno como un nuevo vector energético disponible en cualquier contexto geográfico donde el agua esté presente. El debate en torno al hidrógeno se centra en enfrentar a las antiguas tecnologías de producción como el reformado de metano, denominadas como negras-grises-azules según origen y nivel de contaminación de CO₂, frente a la electrólisis como vía verde de producción de hidrógeno, en cuanto la procedencia de la electricidad tenga esta misma etiqueta. La realidad del hidrógeno verde es su cuasi inexistencia, con cifras del 0.04% del total de hidrógeno producido en el año 2023, frente al 62% proveniente del reformado de metano. Esta realidad seguirá estando vigente mientras el precio del hidrógeno verde (4-7 $/kg) no compita frente al hidrógeno tradicional (1.5-3 $/kg). Dicho esto, el hidrógeno verde es una tecnología apoyada y respaldada con un conjunto de 1046 proyectos en torno a la producción de más de 1 MW, con un requisito en inversión de 320 B$ para el año 2030, según el Hydrogen Council 2023. Además, alejándonos de uso como fuente de energía, el hidrógeno verde está respaldado por su potencial en la descarbonización en sectores con alta demanda energética como la siderurgia. A su vez tecnologías en torno al uso del hidrógeno ya surgen en muchos contextos como su uso en carretillas transportadoras que permite la aplicación de motores con mayor potencia que los eléctricos, pero hábiles para su uso en almacenes de interior, al contrario que sus primos con motor de combustión. Así como el uso de pilas hidrógeno en autobuses, ferrocarriles o elementos de calefacción para el hogar. Todo esto se respaldada por el uso general del hidrógeno en procesos como la producción de armonía, haciendo de este un recurso conocido ya por la industria. Dentro de la batalla entre el hidrógeno azul y el verde surge un grupo de tecnologías que buscan permitir aligerar la transición entre ambas tecnologías, en especial enfocado en la producción de hidrógeno de alta pureza. El chemical looping water splitting (CLWS), es un proceso que permite la obtención de hidrógeno aprovechando el potencial redox de un metal, asignado como oxigen carrier (Transportador de oxígeno) (OC). En este proceso el metal es, en una primera etapa, oxidado por una corriente de vapor de agua que libera hidrógeno puro. Seguido, una corriente de un gas orgánico, como el etanol, en una segunda etapa, reduce el óxido metálico a su estado inicial, liberando a su vez una corriente de hidrógeno, esta vez con un porcentaje de dióxido de carbono, produciendo lo que se conoce en la industria como Syn-Gas. El potencial del CLWS surge de su capacidad de producir un hidrógeno de mayor pureza que el proceso de reducción de metano, eliminado los costes de purificación, pero con un consumo eléctrico menor y coste que los procesos de electrólisis. Sin embargo, los procesos CLWS, o su familia Chemical Looping Hydrogen (CLH), aplican camas fluidizadas del polvo metálico en cuestión, el OC, dentro de los reactores. Esta tecnología requiere de altas presiones en los gases para producir el movimiento del lecho, y por ello incurren en un gasto solo sostenible en aplicaciones de gran escala. Esto ata a estas tecnologías, problema común a la electrólisis y al reformado, a grandes núcleos industriales que requieren de una red de transporte e infraestructura inexistente para la viabilidad del uso del hidrógeno como vector energético. Una alternativa para los CLWS es la aplicación de modelos de un único reactor donde en vez de un lecho fluidizado se aplica un catalizador poroso como medio para ambas reacciones, esto imposibilita ajustar el diseño del reactor a la reacción, pero permite su reducción y aplicación distribuida, permitiendo así la producción local de hidrógeno, evitando la dependencia a las redes de transporte. Introducción 2 2 Como punto de partida para el diseño de estos catalizadores porosos está su desarrollo a partir de metales como el hierro. El hierro, aparte de su valor por su bajo coste y disponibilidad, es considerado uno de los mejores OC al contar con una capacidad de intercambio de oxígeno (Ro) muy elevada, principalmente debido a los múltiples grados de oxidación que este metal presenta (Fe → FeO → Fe2O3 → Fe3O4), con un Ro de 0.3 en el caso de la dupla con hematita (Fe → Fe2O3), solo superado por la dupla CaSO4/CaS con un Ro de 0.47. El principal inconveniente del hierro como OC es su desactivación a bajo número de ciclos, esta se produce por la sinterización de poros y partículas que va reduciendo la superficie libre, junto a la deposición de restos de carbón que no solo limita la reacción, sino que contribuye a la contaminación del hidrógeno en la etapa de oxidación con vapor. El desarrollo de catalizadores porosos ya contribuye a una mayor estabilidad frente a su contraparte porosa, pero para lograr una estabilidad real, el dopado de la estructura del hierro con otros metales con óxidos estables como el aluminio o el titanio ha demostrado ser la mejor opción. El ejemplo de referencia de este texto es el trabajo de Daminzia M. que demuestra la viabilidad del catalizador dopado con aluminio en proceso etanol-vapor, logrando la estabilidad de este tras 10 ciclos de trabajo. La obtención de catalizadores porosos metálicos o en la versión opuesta cerámica, partiendo del óxido metálico, parte de técnicas dentro del ámbito de la pulvimetalurgia. En este ámbito el uso de polvos mezclados en una matriz polimérica permite, en su forma tradicional, la obtención de piezas porosas mediante el moldeo del compuesto y la posterior eliminación del componente orgánico en un proceso térmico que permite a su vez la sinterización del polvo en una estructura porosa y rígida. Procesos más avanzados en este ámbito buscan obtener morfologías más complejas aplicando el material compuesto como medio para un proceso de deposición tridimensional, o impresión 3D. En este caso, el direct ink writting (DIW), es un proceso que permite la aplicación de un medio líquido, la tinta, como material a depositar, que a través de la cabeza extrusora controlada por un robot permite el trazado del camino que da lugar a la geometría tridimensional. En este trabajo se propone el desarrollo de catalizadores cerámicos de hematita a partir de DIW, planteando el proceso desde la obtención de nanopartículas de hematita dopada; su posterior procesado en un material para impresión mediante su combinación con un medio orgánico, conocido en el ámbito de la fabricación de piezas porosas, el canfeno; para su posterior impresión en un equipo de DIW, a estudiar y mejorar durante el propio proyecto; por último se aplicará una etapa de postprocesado para la obtención final de los catalizadores cerámicos porosos. Con el fin de resumir el trabajo desarrollado en la memoria en inglés de este mismo proyecto, se plantean a continuación de forma resumida las secciones de la producción de la hematita, el desarrollo de la tinta para DIW, el equipo aplicado, resultados y conclusiones. 3 2 SINTETIZACIÓN La sintetitación de las nanopartículas de hematita se desarrolla a partir de un proceso de coprecipitación, en este apartado se buscó no solo la obtención de las partículas dopadas con titanio y aluminio, sino también la formalización y estandarización del proceso de fabricación, facilitando así su uso por parte del equipo en trabajo a futuro. Con ello se estableció un proceso de cuatro días para la obtención del polvo en cuestión. La coprecipitación, de forma específica, el proceso Sol-Gel, seguido, busca obtener el óxido en cuestión a partir de precursores en forma de nitruro, en este caso nitruro de hierro III nona hidratado. Para ello, este precursor es disuelto en un medio a pH y molaridad controlada, permitiendo la liberación de los iones metálicos que posteriormente se oxidan para la formación del compuesto objetivo. En este proceso, la presencia de ácido cítrico permite la gelificación del medio, limitando la nucleación de las, permitiendo la obtención del polvo nanométrico esperado. El proceso de Sol-Gel permite un dopado simple mediante la adicción de nitruros del aditivo objetivo durante el propio proceso. Primera etapa La primera etapa del proceso consiste en la obtención de los pesos adecuados para cada precursor en función de la cantidad y la composición final deseada, relacionando los precursores y los compuestos finales por el número de iones metálicos. A su vez, se calculan los requisitos de ácido cítrico y agua en función de la molaridad deseada para el medio, considerando una relación de 1 a 1 entre iones metálicos y moles de ácido. Una vez obtenido los pesos de todos los recursos, estos se mezclan en el medio acuoso mediante agitación magnética, a 80 °C, durante tres horas, para permitir la correcta disociación de los nitruros, figura 2.1. El dopado con titanio requiere de una etapa previa, ya que como precursor de este se eligió el tetra isopropyl orto titanato, este compuesto que es altamente reactivo con la humedad, formando óxido de titanio, requiere su disolución previa en isopropilato y su incorporación lenta a la mezcla ya temperada y en agitación. 2.1 Figura: Mezcla precursora del proceso de coprecipitación. Sintetización 4 4 Segunda etapa La segunda etapa caracteriza el proceso dándole su apellido de Gel. En esta, la evaporación lenta del agua produce la formación de una macromolécula que da la apariencia gelatinosa de primera y luego en forma de estructura cristalina, figura 2.2. Esta gelificación limita la nucleación de los óxidos, dando lugar a las nanopartículas. La evaporación se realizó a 90 °C durante 6 horas. 2.2 Figura: Cristal resultante del proceso de gelificación-evaporación. Tercera etapa La tercera etapa del proceso consiste en la eliminación del material orgánico mediante la calcinación de la muestra seca resultante del proceso anterior. En esta etapa se sigue un ciclo donde inicialmente se alcanza una temperatura de 350 °C para permitir la calcinación de cualquier orgánico; seguido de una etapa a 620 °C que permite la obtención del óxido deseado en cuestión, la hematita. Resultado en la figura 2.3. 2.3 Figura: Polvo calcinado fruto de la tercera etapa. Cuarta etapa Una vez obtenido el polvo calcinado, este pasa por un postprocesamiento que busca separar las partículas obtenidas y funcional izar su superficie mediante el recubrimiento con un dispersarte, ácido esteárico. Este proceso se lleva a cabo mediante un molino de bolas aplicando como medio etanol. El molido se desarrolla durante 12 horas. Una vez concluida la molienda, las partículas son lavadas con etanol y separadas en un rotavapor para la extracción del alcohol a baja temperatura, evitando así la separación de las partículas al ácido esteárico, permitiendo la recuperación del medio para su uso a futuro. El polvo seco es molido a mano y almacenado para su uso en la tinta. Resultado en figura 2.4. 2.4 Figura: Polvo de hematita resultante del proceso de copreciptación