Metal–polymer composites are widely used in aerospace, automotive, and marine applications due to their high specific strength and favorable wear and corrosion resistance. However, reliable interfacial joining remains challenging because of the diss...
Metal–polymer composites are widely used in aerospace, automotive, and marine applications due to their high specific strength and favorable wear and corrosion resistance. However, reliable interfacial joining remains challenging because of the dissimilar properties of metals and polymers, and the design and process freedom for complex joint geometries is often limited. In our previous work, mechanical interlocking joints were realized by filling and building polymers into machined negative (recessed) metallic features using vat photopolymerization (VPP). Nevertheless, the negative-feature approach inherently increases the metal volume fraction at the joint and constrains geometric design flexibility. To address these limitations, this study extends the VPP-based filling–building concept by incorporating both negative- and positive-feature metal architectures and proposes a manufacturing route for interlocking metal–polymer composites. In particular, positive (protruded) metal lattices fabricated by powder bed fusion (PBF) were infiltrated by exploiting over-curing in VPP, followed by continuous VPP building on the same platform. This approach aims to reduce the metal volume fraction at the joint while enabling diverse interlocking geometries. Eight unit-cell designs were investigated, including four pillar-type and four lattice-type architectures, designed to have comparable volumes for fair comparison. The maximum curing depth and depth-wise microhardness profiles were quantified as a function of over-curing time to establish process conditions that enable infiltration into the internal regions and to discuss the underlying indirect-curing mechanism. Under the selected over-curing and subsequent building conditions, all geometries showed complete infiltration without discernible voids in the observed regions, and the infiltrated polymer conformally occupied the internal unit-cell spaces. Tensile tests revealed geometry-dependent maximum load and fracture behavior: A1 (Pillar) and A4 (Pawn) among the pillar-type, and B2 (BCC) and B3 (FCC) among the lattice-type, exhibited relatively higher maximum loads. In contrast, B1 (SC) and B4 (Pyramid) showed lower loads, which is attributed to combined effects of geometry-dependent optical shielding and light transport that alter the effective curing region and hardness gradient, along with stress concentration induced by the interlocking geometry. Fractography indicated polymer-dominant failure in all specimens, while polymer residues locally remained on metal surfaces in some cases, suggesting a potential contribution of micro-interlocking associated with the as-built surface microstructure. This study demonstrates a process pathway that complements the limitations of negative-feature joints by leveraging positive-feature architectures and clarifies the relationships among geometry, infiltration, mechanical performance, and fracture behavior. The findings provide practical guidance for joint design and process-window selection for interlocking metal–polymer composites. Keywords : Vat photopolymerization, Powder bed fusion, Metal–polymer composites, Over-curing, Tensile test, Finite element analysis