Material extrusion-based polymer additive manufacturing (AM) processes such as fused deposition modeling (FDM) or fused filament fabrication (FFF) are increasingly being used for large-scale structural applications, necessitating an understanding of the underlying structure-process-property relationships in order to predict performance under load. The strength of material extrusion AM parts along the build direction, and perpendicular to the extruded strands, commonly referred to as “Z-strength” is of particular interest since it is lower than the strength along the extruded strand direction and is a limiting factor in design. Furthermore, the Z-strength is sensitive to the process parameters of the print and could change depending on the geometry of the part, or even at different locations within the same part. This presents a challenge for the qualification and characterization of the part’s structural performance and highlights the need for predictive models for Z-strength. The inter-dependence of process parameters further exacerbates the problem by making it difficult to independently control the parameters and characterize Z-strength using purely an experimental approach.Extensive studies have been performed in the literature to build analytical and computational predictive models for polymer AM. However, they all focus exclusively on polymer chain diffusion to predict interlayer adhesion strength, or coalescence to predict effective bead shape and contact area. While these are important contributors to Z-strength, there is a lack of a holistic predictive model that combines these factors together, along with other mechanical factors such as the effect of surface stress concentrations. Surface stress concentrations are especially critical with recent trends towards large format additive manufacturing which uses brittle, filled polymers, and also prints with tall layer heights – both of which increase the material sensitivity to surface notches. In this work, it was experimentally demonstrated that up to 38% of strength reduction could be attributed solely to the presence of surface beads. The theoretical stress concentration factors (kt) associated with the surface bead shapes were quantified using meso-scale finite element analysis (FEA) with detailed/realistic modeling of the individual beads. Critical parameters influencing kt were identified and parametric FEA studies were performed by varying the geometric parameters. A single empirical equation that can be used for determining kt for any bead shape was generated. In addition, a guideline was proposed for the effective wall thickness to be used when determining nominal stress (FEA or otherwise) such that kt is independent of the direction or type of load applied. Next, a novel framework for predicting Z-strength was proposed by breaking it down into a combination of interlayer adhesion, bead geometry, effective contact area, surface stress concentrations, and notch sensitivity. Interlayer adhesion was determined using a polymer chain diffusion-based model through indirect characterization of weld time, and validated using tensile test data at the coupon level. The evolution of the bead shape and its associated parameters was determined using a coalescence model. The resulting bead shape was used to obtain kt, which was then applied to the interlayer adhesion strength to predict the effective Z-strength. The results were validated with an extensive database of tensile test coupons cut from the walls of AM parts, with over 50 combinations of different process parameters including layer height, extrusion temperature, layer time, and print angle (layer offset). The methodology was successfully integrated into existing commercial software GENOA 3DP along with finite element software ABAQUS and validated to predict part-level structural performance with variable Z-strength. Overall, this work creates a holistic framework for predicting Z-strength in polymer AM as a function of process parameters.
Read
