Het verbeteren van de resolutie bij Laser Poeder Bed Fusie van koper voor toekomstige warmte-overdrachtsapparaten

The ever-growing demand for more performant heat transfer devices has led to increasingly compact and complex designs. Additive manufacturing, and in particular Laser Powder Bed Fusion (LPBF), has been proposed as a viable route to produce such next-generation thermal components. While LPBF offers excellent resolution for most engineering materials, copper remains challenging due to its high thermal conductivity and low absorptivity at near-infrared (NIR) wavelengths, which hinder process stability and limit feature resolution. This dissertation addresses these challenges, ultimately enabling the fabrication of fine copper features via LPBF for advanced heat transfer applications. For clarity, the work is divided into two parts. Part I, titled "Exploring small-scale copper features produced by Laser Powder Bed Fusion", examines how laser-material interaction dictates processability and resolution in copper LPBF. Oxygen-free copper (OF-Cu) is benchmarked against copper powder surface-decorated with molybdenum nanoparticles, introduced to enhance laser absorption. Calorimetric measurements confirm the absence of a stable conduction-mode regime in OF-Cu, distinguishing it from conventional LPBF materials and explaining its reliance on high energy densities and keyhole-mode melting—conditions that inherently penalize resolution. Surface decoration significantly improves effective absorptivity at lower energy densities; however, stable processing still requires elevated energy input due to heat losses driven by thermal conduction, which limit the energy available for melting. To capture these effects, a unified scaling law is proposed, incorporating a conduction-loss correction factor and intrinsic powder absorptivity, enabling accurate prediction of effective absorptivity across materials and process conditions. Process optimization for bulk parts demonstrates that baseplate preheating and powder modification substantially expand the stable process window. Among all strategies, nano-molybdenum decoration proves most effective, doubling intrinsic powder absorptivity and reducing the energy input required for densification by up to 20%, while preserving high electrical conductivity (~93% IACS). Combined with single-bead scan strategies, these improvements enable fine-feature fabrication, achieving thin walls of 151 µm and pin fin diameters of 200 µm, compared to ~250 µm for pure copper. Microstructural analysis attributes these gains not only to enhanced absorptivity but also to increased melt viscosity, which stabilizes keyhole melt pools and mitigates surface roughness. While powder modification has become a common approach to improve absorptivity in copper LPBF, molybdenum nanoparticle decoration stands out due to its low solubility in the copper matrix and high thermal stability. The step-by-step analysis of the LPBF process for both modified and unmodified powders—from effective absorptivity to thermal and mechanical properties, and ultimately the resolution achievable with single-bead strategies—provides essential insights and a robust foundation for the broader adoption of copper LPBF in advanced thermal applications. Part II, titled "Thermal performance of LPBF-manufactured TPMS heat exchangers", evaluates the thermal properties of gyroid-based triply periodic minimal surfaces (TPMS) manufactured via LPBF in OF-Cu, A205 aluminum, and reaction-bonded silicon carbide (RBSC). Initially, gyroid lattices with varying unit cell sizes (UCS) ranging from 5 mm to 10 mm and relative lattice densities (RLD) from 10% to 50% were manufactured and characterized to evaluate their suitability for compact heat exchanger applications. Key thermal properties, including effective thermal conductivity and overall heat transfer coefficients, were quantified using four-point electrical conductivity measurements and forced convection experiments conducted in the laminar flow regime. The experimentally determined thermal conductivity exhibited a small relative error (<15%) compared to FEM-based results reported in the literature. For OF-Cu, the deviation between measured and simulated overall heat transfer coefficients remained below 5%, underscoring the accuracy and reliability of the experimental setup. A macro-scale Finite Volume Model (FVM) was developed to derive the homogenized local heat transfer coefficient (h̄g) from the overall heat transfer coefficient, and was validated against high-fidelity CFD simulations. The Nusselt-Reynolds (Nu-Re) relationship obtained from CFD was found to be independent of UCS and nearly independent of RLD. The Nusselt-Reynolds relationship derived from the experimental-FVM approach closely matched CFD results for A205 and OF-Cu. Larger deviations observed for RBSC point to the influence of manufacturing-induced imperfections such as porosity and surface roughness. Overall, gyroid structures consistently outperformed conventional straight-fin designs in terms of volumetric heat transfer in the laminar regime, primarily attributed to vortex-induced boundary layer disruption. The resolution advancements resulting from part I were put to the test by producing gyroid structures with the single bead scan strategy, which resulted in a 2 mm UCS and 1 mm UCS for OF-Cu and Cu1Mo, respectively. Experimental investigations at this scale remain limited, and to date, no comprehensive studies had addressed the combined effects of Relative Lattice Density (RLD), Unit Cell Size (UCS), and material selection on the local thermal performance of gyroid structures operating in the laminar flow regime. Additionally, this part presents the first experimentally derived local heat transfer coefficients from TPMS-based heat sink experiments, offering a novel contribution to the field. The results serve as a foundation for more computationally efficient macro-scale thermal simulations and optimization and support the use of gyroid structures in future thermal applications. In conclusion, this work establishes that physics-guided process optimization and powder modification can overcome copper's inherent LPBF challenges, enabling fine-feature fabrication without compromising thermal conductivity. Coupled with macro-scale Finite Volume Models and advanced scan strategies, these findings provide the necessary tools for designing compact, periodic heat transfer devices.