Powder Spreading Characteristics of Monomodal and Bimodal Spherical Alumina for Ceramic Matrix Composite Binder Jet Additive Manufacturing
Restricted (Penn State Only)
- Author:
- Groeneveld-Meijer, Willem
- Graduate Program:
- Mechanical Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 22, 2023
- Committee Members:
- Guhaprasanna Manogharan, Thesis Advisor/Co-Advisor
Robert Carl Voigt, Committee Member
Robert Kunz, Professor in Charge/Director of Graduate Studies
Jeremy Schreiber, Special Signatory
Matt Lear, Special Signatory - Keywords:
- binder jet additive manufacturing
additive manufacturing
alumina powder
ceramic matrix composites
ceramic
composites - Abstract:
- Binder jet additive manufacturing (BJAM) allows for the manufacturing of complex geometries from traditionally difficult to manufacture materials [1]. These ceramic materials, such as alumina, Al2O3, and silica, SiO2, inherently lend themselves to BJAM for the purpose of manufacturing complex precursor geometries for ceramic matrix composites (CMCs) [2]. The toughness and ultimate tensile strength of these CMCs relies upon the liquid metal infiltration into a ceramic precursor pore network [3]. This paper presents empirical data in combination with discrete element model (DEM) simulations of the settling and spreading behavior of uniform and bimodal spherical Al2O3 powders in order to efficiently tailor bulk pore networks for CMCs. Specifically, the behavior of these powders during deposition and recoating were studied as well as the optimal large particle mole percentages required to achieve high packing factor in as printed preforms. The densities of these preforms depends upon the uniformity of the powder bed, any deviations in the powder bed would impact density of, and the ability to tailor the pore networks for CMCs. For BJAM using bimodal powders, it is important that the bimodal mixture remains uniform throughout the powder bed and does not separate into its component powder size distributions (PSDs). DEM simulations were utilized to find the highest packing factor for bimodal powders centered around measured volume mean particle sizes of 6.9μm and 21.7μm. This was determined to be 77.48% when the packing calculations were represented as body centered cubic (BCC) like [4], [5]. For the specific powders used, both monomodal and bimodal, it was shown that a higher percentage of small particles do not deposit more heavily at the beginning of the powder bed recoat cycle. Statistical data gathered with respect to p values and measured variables shows that the p values are statistically significant due to the plotted average means within each pairwise data group being statistically significantly different. Future investigation is required as these statistically significant surface roughness variables found via analysis of variance (ANOVA) suggest other possible causes of variations occurring within the powder bed. This knowledge can be directly applied to future work manufacturing BJAM of CMC preforms with precisely tailored part porosity, and therefore tuned material properties.