Date of Award


Degree Name

Master of Science


Mechanical Engineering


David Espalin


Ceramic additive manufacturing (AM) uses the same fundamentals as material extrusion (MEX) that utilizes polymers or resins for printing applications. Ceramic AM uses the same process of slicing an STL file to determine a toolpath for the printer that will facilitate material deposition to manufacture full parts; ceramic AM also provides a low-cost advantage when compared to processes that require high powered lasers for melting metal powders or heated nozzles used for dispensing thermoplastics. Despite having the similar design freedom and ability for rapid prototyping that other printing methods benefit from with the addition of low cost, ceramic AM has unique challenges and disadvantages that must be understood and characterized before benefitting from the roles that ceramic parts can fill.Ceramic pastes are vulnerable to drying, which can cause nozzle clogging and cracking after deposition. They are also heavily reliant on the initial chemical formulation and accompanying material characteristics since ceramic pastes need to be easily extruded while maintaining shape after deposition, mitigating the drying effects mentioned earlier, and must have the ability to produce a high-density ceramic part after sintering. These challenges compound when considering a final part since favorable changes that affect shape retention may negatively impact drying, and changes that positively impact drying may affect the density of the final part due to the inclusion of voids after extrusion. Ceramic AM depends on the shear thinning and shape retention characteristics of paste, thus, it is important to obtain rheological characteristics of the paste and the effects of printing conditions on that paste. This research focuses on the rheological characterization of barium titanate ceramic paste and dimensional characterization of printed single 50 mm long ceramic beads in response to changes in ceramic solid loading volume percent, PEG molecular weight, and print parameters such as extrusion factor, print velocity, and nozzle type. A full factorial design of experiments was made using 5 factors with 2 levels for a total of 32 runs. Beads were printed, then scanned 10 times in intervals of one minute. A separate container of paste was then taken and used to determine material properties using rheology. This was done to determine the significance of each variable on the dimensions and leveling of ceramic beads, which would provide insight into the requirements that would produce successful beads with favorable characteristics and measurements matching the intended dimensions. This research was impactful since characterizing single beads can be expanded to the outcome of a full part without the necessity of longer print times and more material waste that would come with printing full parts. Although full parts printed at the end of this experiment were not successful, bead characterization was used to determine the challenges associated with printing each paste using different factors. The printed beads exhibited errors that were also present in printed parts while also providing data to determine the most impactful factors without using the larger amount of time and material required for printing parts. Measurements of printed beads collected using a laser profilometer determined that lower solid loading vol% and PEG molecular weight caused greater deformation after extrusion and beads lost 22.2% of their height over 10 minutes. Beads printed using high solid loading vol% and PEG molecular weight had better shape retention, only losing 4.2% height over 10 minutes with no increase in width, however, these beads also experienced discontinuities due to nozzle clogging. Rheology of these pastes revealed that high solid loading vol% had significantly greater yield stress and flow point values, having a magnitude increase of 728% and 1,385.3%, respectively, when considering 39BTO-12-400 and 55BTO-12-400. A statistical analysis of the results determined that a combination of high extrusion factor and high PEG molecular weight was most impactful for producing viable beads using both solid loading vol%, while the best overall combination was high extrusion factor, high PEG molecular weight, and high solid loading. Based on the statistical analysis, print velocity and nozzle type had no impact, while solid loading was the most significant, followed by extrusion factor, and finally PEG molecular weight. Statistically significant parameters identified using ANOVA were used to print four parts to determine if dimensionally accurate parts could be printed using the best levels for each factor. Although printing was successful, the extrusion factor was insufficient for the designated standoff distance and layer height used for each part, resulting in discontinuities and clogging, which were the same issues present in single bead printing.




Received from ProQuest

File Size

99 p.

File Format


Rights Holder

Nicholas Durand

Available for download on Friday, May 23, 2025