TPMS structures are mathematical surfaces that divide space into two intertwined, non-intersecting regions while maintaining a mean curvature of zero at every point. Translated into physical form, they produce smooth, continuously curved cellular solids that resemble soap films frozen in place. Unlike traditional strut-based lattices built from beams and nodes, TPMS geometries distribute stress along smooth curved walls, avoiding the sharp junctions where cracks and stress concentrations typically initiate. This makes them attractive for applications ranging from bone implants and tissue scaffolds to heat exchangers, energy absorbers, and lightweight aerospace panels, where the ability to tune stiffness, strength, and porosity independently is a decisive advantage.

The researchers focused on four of the most widely used TPMS topologies: the Gyroid, the Schwarz Primitive (SP), the Schwarz Diamond (SD), and the Neovius surface. Each of these geometries has a distinctive internal architecture. The Gyroid, perhaps the most famous, twists through space in a helical, chiral pattern with no straight lines or flat faces. The Schwarz Primitive consists of a network of orthogonal channels, the Schwarz Diamond mimics the tetrahedral bonding arrangement of a diamond crystal lattice, and the Neovius surface is a more complex derivative of the Primitive family with additional saddle points and a higher degree of connectivity. Because each topology channels load differently, the choice between them can dramatically alter mechanical performance even when the amount of material used is identical.

To disentangle the effects of geometry from those of size and density, the team designed lattices with two unit cell sizes, 10 millimeters and 15 millimeters, and four relative densities: 20, 25, 30, and 35 percent. Relative density, the ratio of the lattice’s density to that of the fully dense parent material, is the single most important scaling parameter in cellular solids theory, while unit cell size governs how many cells fit within a structure and how boundary effects propagate through it. By holding all printing parameters constant across every specimen, the authors ensured that any differences in compressive behavior could be attributed to design rather than to process variability, a methodological discipline that strengthens the statistical conclusions drawn later in the study.

All specimens were fabricated from PLA filament using FFF, the most accessible and widely deployed 3D printing process, in which a thermoplastic filament is melted and extruded layer by layer. PLA was a deliberate choice: it is inexpensive, dimensionally stable, and prints reliably, making it a sensible model material for establishing design rules that can later be transferred to engineering thermoplastics or even metal additive manufacturing. The printed lattices were then subjected to uniaxial compression testing, the standard method for characterizing cellular materials, which reveals not only peak strength and stiffness but also the shape of the stress-strain curve, including how the structure collapses, plateaus, and eventually densifies as opposing walls crush together.

The results were striking in their spread. The Neovius lattice emerged as the clear champion, achieving a maximum compressive strength of 41.51 megapascals at 35 percent relative density with the smaller 10-millimeter unit cell. The Schwarz Diamond followed at 34.60 megapascals, confirming that highly connected, diamond-like architectures translate well into load-bearing polymer lattices. The Gyroid, by contrast, reached 23.81 megapascals but displayed a smoother, more gradual stress-strain response with progressive densification, a behavior that may prove more valuable than raw strength in applications demanding predictable, controlled energy absorption, such as protective packaging or cushioning. The Schwarz Primitive trailed the field at 20.26 megapascals, suggesting that its open orthogonal channel network concentrates load less effectively under compression in this material system.

To quantify which design variables actually mattered, the researchers performed an analysis of variance (ANOVA) on their experimental data. The verdict was unambiguous: lattice topology was the dominant factor, accounting for 56.03 percent of the variation in compressive performance. Relative density came second at 33.25 percent, while unit cell size contributed a comparatively modest 7.59 percent. For engineers, this hierarchy is genuinely useful knowledge. It means that selecting the right surface geometry matters more than fine-tuning wall thickness or cell dimensions, and that a designer’s first decision, the choice of TPMS family, carries more than half the weight in determining final mechanical performance.

The study then moved from characterization to prediction. The team fitted regression models to their experimental results, converting raw compression data into mathematical objective functions that relate compressive performance to unit cell size and relative density for each topology. These models became the fitness functions for the Lyrebird Optimization Algorithm (LOA), a bio-inspired metaheuristic introduced in 2023 that mimics the remarkable vocal mimicry behavior of the Australian superb lyrebird, a bird capable of imitating almost any sound it hears. In the algorithm, candidate solutions adjust and imitate the best-performing solutions in the population, balancing exploration of the design space with exploitation of promising regions, in a manner conceptually similar to genetic algorithms and particle swarm optimization but with distinct mimicry-driven update rules.

Running the LOA across the regression models produced a clear and consistent optimum: for every TPMS topology, the best combination of design parameters was approximately the 10-millimeter unit cell paired with 35 percent relative density, the densest and finest configuration tested. The Neovius lattice achieved the highest composite desirability score of 0.994, a near-perfect value indicating that the optimization criterion, which balances multiple performance objectives into a single figure of merit, was almost fully satisfied by this configuration. The convergence of the experimental data and the algorithmic search on the same design point provides mutual validation: the statistics did not merely fit the data, they pointed to a physically meaningful optimum.

The broader significance of this work lies in its demonstration of a complete, systematic pipeline for architected material design: fabricate a controlled matrix of specimens, test them, distill the results into regression models, and let a modern metaheuristic search the design space for optima. As additive manufacturing continues to expand into certified structural components, medical implants, and multifunctional devices, methods like this one transform lattice design from an intuitive art into a quantitative engineering discipline. The humble lyrebird, it turns out, has something to teach the world of materials science, and the Neovius surface, long a mathematical curiosity, now has the compressive test data to claim its place among the strongest printable lattice geometries available to designers working with desktop 3D printers.

Subject of Research: Mechanical characterization and optimization of 3D-printed TPMS lattice structures

Article Title: Mechanical characterization and lyrebird algorithm-based optimisation of fused filament fabricated triply periodic minimal surface lattice structures

Article References: Aryan, O. R., & Debnath, K. (2026). Mechanical characterization and lyrebird algorithm-based optimisation of fused filament fabricated triply periodic minimal surface lattice structures. Journal of Materials Science. https://doi.org/10.1007/s10853-026-13871-w

Image Credits: AI Generated

DOI: 10.1007/s10853-026-13871-w

Keywords: TPMS lattice structures, fused filament fabrication, PLA, Neovius surface, Gyroid, Schwarz Diamond, compressive strength, Lyrebird Optimization Algorithm, ANOVA, relative density, unit cell size, additive manufacturing

The study was led by Assistant Professor Farid Labib of the Research Institute for Science and Technology at Tokyo University of Science, together with Associate Professor Kazuhiro Nawa of the Institute of Multidisciplinary Research for Advanced Materials at Tohoku University and Professor Ryuji Tamura of Tokyo University of Science. The team focused on Tsai-type compounds, a class of multi-shell cluster-based structures in which moment-bearing rare-earth atoms occupy an icosahedral site. In these materials, nested atomic shells, including the rhombic triacontahedron, icosidodecahedron, icosahedron, dodecahedron, and an inner tetrahedron, build up a cluster architecture whose rare-earth icosahedral shell carries the magnetic moments responsible for the compounds’ magnetic response.

Historically, materials chemists have relied on chemically tunable parameters to organize the magnetic properties of complex intermetallics. The most influential of these is the valence-electron concentration, commonly expressed as the electron-per-atom, or e/a, ratio. In metallic systems such as Heusler alloys and approximant crystals, the e/a ratio has been widely used to classify magnetic ground states. In gold-based Tsai-type approximant crystals specifically, the e/a ratio was found to control whether the compounds developed long-range antiferromagnetic order, long-range ferromagnetic order, or a spin-glass state in which the magnetic moments freeze in random orientations. But the predictive power of e/a has a serious limitation: it does not transfer cleanly across different alloy families and constituent elements.

That limitation matters because quasicrystal-based intermetallics are considered promising platforms for exploring emergent magnetic phenomena and quantum states not found in ordinary periodic crystals. Without a reliable, experimentally accessible parameter to guide the search, researchers have lacked a systematic way to identify which compositions are worth synthesizing. As Dr. Labib explained, quasicrystals are among the most uniquely structured materials discovered to date and are expected to exhibit novel magnetic states and quantum phenomena not found in ordinary crystals, yet until now there has been no unified guideline for systematically exploring these phenomena in quasicrystals and their approximant crystals.

To close that gap, the team synthesized a family of gold-based 1/1 approximant crystals containing aluminum or gallium and the rare-earth elements terbium, dysprosium, and holmium, and then systematically measured their structural and magnetic properties. Approximant crystals are periodic structures whose local atomic environments closely mimic those of true quasicrystals, making them experimentally tractable stand-ins for the quasiperiodic originals. The measurements revealed a nearly monotonic inverse correlation between the e/a ratio and the lattice parameter, meaning that as the electron concentration per atom rises, the physical size of the cubic unit cell shrinks in a regular, predictable way.

The researchers then extended their experiments to non-Heisenberg Tsai-type compounds containing terbium, dysprosium, and holmium. These compounds are called non-Heisenberg because their magnetic moments do not behave as simple, freely rotating spins; instead, the crystal electric field generated by the surrounding lattice imposes strong uniaxial magnetic anisotropy that favors specific orientations of the moments. Under these conditions, the team observed characteristic whirling antiferromagnetic and ferromagnetic orders, in which the constrained moments arrange themselves in swirling patterns dictated by the interplay of anisotropy and the conduction-electron-mediated interactions between rare-earth moments on the icosahedral shell.

When the researchers plotted the magnetic ground states against the lattice parameter, the result was striking. The compounds organized with high accuracy according to the size of the unit cell, whereas the conventional e/a classification showed systematic shifts depending on which rare-earth element was present and what the alloy composition was. The phase boundaries were sharply defined. Compounds with lattice parameters above approximately 14.72 angstroms adopted the whirling antiferromagnetic state. Those with lattice parameters between 14.62 and 14.72 angstroms became whirling ferromagnets. And those with lattice parameters below approximately 14.62 angstroms froze into spin-glass states. These well-defined thresholds establish the lattice parameter as a unified and experimentally accessible structural descriptor for predicting magnetic ground states in the Tsai-type family.

The physical interpretation is that structural length scales, alongside electron concentration, must be explicitly considered when understanding the conduction-electron-mediated magnetic interactions in these complex alloys. In rare-earth intermetallics, the magnetic moments communicate with one another indirectly through the conduction electrons, and the strength and sign of that communication depend sensitively on the distances between the moment-bearing atoms. A change in the lattice parameter therefore directly reshapes the geometry of the magnetic interaction network on the icosahedral shell, which in turn determines whether the system settles into antiferromagnetic whirling, ferromagnetic whirling, or glassy freezing. The e/a ratio captures part of this physics but conflates it with element-specific electronic effects, which is why it fails to produce a single, element-independent classification.

The practical implications extend well beyond classification. Because the lattice parameter can be tuned through alloy composition and measured with routine structural characterization, it offers materials designers a direct control knob for targeting a desired magnetic ground state. Dr. Nawa noted that the unified magnetic phase diagram constructed in the study can serve as a practical roadmap for the systematic exploration of new magnetic quasicrystals and approximant crystals exhibiting novel magnetic orders and quantum phenomena, and that it can provide a guideline for designing new magnetic materials with targeted magnetic ground states, opening new opportunities for discovering unconventional magnetism in quasiperiodic and complex intermetallic systems.

The study was supported by the Japan Society for the Promotion of Science through multiple Grants-in-Aid for Scientific Research, by the Murata Science and Education Foundation, and by the Japan Science and Technology Agency through its CREST program. The authors declared no competing interests. For a field that has spent decades cataloguing the exotic magnetism of quasiperiodic materials one compound at a time, the message of this work is unusually clean: in the Tsai-type family, the size of the atomic cage appears to decide the fate of the spins inside it. If the lattice-parameter thresholds hold across further alloy families, the search for new magnetic quasicrystals may become far less of a treasure hunt and far more of an engineering exercise, with the next generation of unconventional magnets designed first on paper, in angstroms.

Subject of Research: Lattice-parameter control of magnetic ground states in Tsai-type quasicrystals and approximant crystals

Article Title: Lattice parameter as a unified parameter governing magnetic ground states in Tsai-type compounds

Article References: Lattice parameter as a unified parameter governing magnetic ground states in Tsai-type compounds. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: quasicrystals, Tsai-type clusters, lattice parameter, magnetic ground states, antiferromagnetism, ferromagnetism, spin glass, rare-earth intermetallics, electron-per-atom ratio, magnetic anisotropy, approximant crystals, Tokyo University of Science