Controlling the Microstructure of Metal and Polymer Foams via Advanced Templating Approaches

This thesis investigates the use of microfluidic droplet generation for the synthesis of highly uniform copper and polystyrene foams, and the characterisation of these materials with an emphasis on high-resolution X-ray computed tomography. Specifically, this research addresses the application of macroporous foams in high energy density laser experiments, where they serve as targets for high-powered lasers to create plasmas that model extreme conditions. The stringent and evolving requirements of laser target materials present ongoing challenges for the synthesis and characterisation of low-density materials, so, in collaboration with AWE, droplet microfluidics-based templating was explored as a novel route to control microstructure and density in uniform and tuneable metallic and polymer foams.
Initially, droplet microfluidics was explored as a tool for fabricating monodisperse polymer microsphere template materials and assessing their potential for templating uniform copper foams. Flow focusing capillary microfluidic devices produced large 1,6 hexanediol diacrylate (HDDA) droplets of high uniformity, which were UV-cured to create polyHDDA microbeads. However, challenges with droplet size control and device fouling indicated that microfluidic photopolymer microspheres may not be suitable for templating metal foams required for plasma physics experiments.
Subsequently, uniform styrene-co-divinylbenzene (DVB) polymerised high internal phase emulsions (polyHIPEs) were synthesised using monodisperse microfluidic water-in-styrene emulsion droplets as templates. The resultant foams had small, highly uniform, tuneable pore sizes, and precise density control was achieved through novel centrifugation-based techniques. These foams were suitable for laser-plasma experiments and provided an ideal alternative to currently used disordered polyHIPEs.
To further enhance material characterisation, particularly to identify pores resulting from droplet coalescence, a deep learning model was developed for segmenting high-resolution X-ray tomograms of microfluidic polystyrene polyHIPEs. By optimising training data generation, and evaluating hyperparameters, a deep learning model was developed to segment tomogram volumes to identify large pores which did not arise due to direct templating by small microfluidic droplets. Building on these developments, the research quantitatively explored porosity in microfluidic templated polystyrene and poly(4-chlorostyrene) polyHIPEs. The inclusion of 4-chlorostyrene led to greater coalescence and larger pores and modification of curing temperature and internal phase volume fraction was shown to affect polyHIPE morphology. Quantitative volumetric analysis revealed the combined value of microfluidics and segmented X-ray CT for understanding polyHIPE properties.
Finally, the synthesis of copper foams was investigated using electroless plated polystyrene polyHIPEs as templates. The depth of copper deposition was correlated with the degree of openness of polyHIPEs. Increasing the reaction rate and reagent diffusion improved copper penetration, leading to the synthesis of uniform copper foams using microfluidic templated polyHIPEs.