Analysis of local manufacturing variability in polymer melt extrusion 3D printing

Polymer melt extrusion 3D printing produces spatially and temporally non-uniform local thermal histories as the printed domain grows through deposition, cooling, and reheating. In addition, inherent variability in material properties and process conditions can cause different thermal responses even at points nominally subjected to the same printing conditions. This variability affects repeatability and reproducibility of the printing process through its influence on local thermal histories and, consequently, on part quality and final material state. To address this, the thesis develops experimental, statistical, and numerical approaches to study local thermal-history variability and process-level material-state evolution in polymer melt extrusion additive manufacturing. Acrylonitrile butadiene styrene (ABS) is used as a case study for amorphous polymers. Temperature-dependent specific heat capacity and thermal conductivity are measured through repeated thermal tests, and their sample-to-sample variability is quantified using Bayesian models. The glass-transition zone is identified through complementary thermal and thermomechanical measurements, providing the basis for interpreting the temperature-dependent behavior of thermal conductivity over the processing range. These probabilistic material properties, together with measured fluctuations in deposition, chamber, and build-plate temperatures, are incorporated into a transient finite element model of fused filament fabrication. The model propagates uncertainty through the deposition and cooling sequence and predicts a range of local thermal histories, consistent with in-situ measurements, that can arise under nominally identical printing conditions. Polyether ether ketone (PEEK) is used as a case study for semi-crystalline polymers, where local thermal history governs both thermal and material-state evolution during processing. An adopted crystallization model is combined with a melting model developed in this work and implemented under additive-manufacturing-relevant thermal histories with repeated cooling and reheating. Temperature- and state-dependent thermophysical properties are formulated for process-level use, and a representative temperature is defined for consistent material-state updates. The resulting framework is used to examine how the coupled descriptions follow material-state evolution along complex thermal paths, including suboptimal printing conditions. Overall, this thesis advances process-level understanding of polymer melt extrusion additive manufacturing by establishing how thermal-response variability develops during printing and by developing the steps required to examine material-state evolution under process-relevant thermal histories.