Nodal geometry optimization in additively manufactured thin lattice structures: Application to the design of lightweight Ti–6Al–4V LPBF metamaterials

In recent years, the increasing use of additive manufacturing has enabled the development of metamaterials, such as lattice structures, which provide high strength-to-weight ratios and energy absorption capabilities. However, lattice structures have a flaw: the sharp intersections at the nodes act as stress concentrators, resulting in reduced mechanical properties. To avoid this, several approaches to modify the nodal geometry are known, such as nodal filleting or spherical joints. However, there is a lack of a systematic, generalizable understanding of the impact on nodal geometry and how to design it for optimal mechanical properties, backed by experimental work. This work investigates, at fixed relative densities, combining both experimental and computational approaches, the effect of node rounding on the mechanical performance of different unit cells. With this aim, the study initially focuses on analyzing the effect of node rounding on the mechanical properties of BCC cellular structures at two relative density levels (low at ≈ 3% and high at ≈ 24%). The constant relative density requirement is achieved through increasing fillet radii while decreasing strut diameters for each relative density. The results show increases of the elastic modulus ranging 50.4–103.5%, and 26.4–45.7% in the yield strength, showing a lower impact of the nodal geometry at lower relative densities. Finally, the study is generalized to other unit cells (simple cubic). The approach developed in this work reveals design indicators correlated with the mechanical performance of node-reinforced cellular metamaterials and enables exploration of a broader design space to optimize lattice structures for specific properties. • This study addresses the nodal topology optimization in thin lattice structures. • The work focus on Ti–6Al–4V body-centered cubic (BCC) thin lattice structures. • Increases of up to 103.5% in stiffness and 45.7% in the yield strength are reported. • Real-geometry-informed FE analysis shows excellent alignment with experiments. • Equivalent stress concentration coefficient is shown as accurate performance predictor.