Exploring relative size effects for strut-based and sheet-based scaffolds defined by repeating unit cell geometry fabricated via selective laser melting

Thumbnail Image




Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



With advancements in 3D printing, porous titanium implants have gained attention in the medical community as a suitable replacement for damaged bone. Additive manufacturing techniques like Selective Laser Melting (SLM) can create complex porous structures within the body of an implant that encourage osseointegration and result in implant stiffness that matches that of surrounding bone. This leads to better integration of the implant and decreases the risk complications due to stress shielding.

One concern with applying porous architectures to medical implants comes from the small size of the implants. Small porous devices can see boundary effects where a truncated pore is no longer contributing to the loading, which causes the porous material to be weaker than bulk properties would predict. As such, the ratio of the diameter of the loading cross-section to the unit cell size (D/u) becomes an important consideration when applying porous structures to load-bearing implants.

This study sought to find the saturation point of D/u, which is the point where the boundary effects are no longer significant and the properties of the porous material reflect bulk material properties. Three different porous architectures were tested in this study: gyroid-sheet, octet-truss, and stochastic-truss. Cubic unit cells ranging from 3x3x3mm to 12x12x12mm were applied to 10mm and 20mm diameter samples for each architecture, then samples were printed from Ti6Al4V powder using SLM. Specimens were then tested under compressive loading to determine compressive mechanical properties.

Testing revealed that the gyroid-sheet was the strongest and stiffest architecture, followed by the octet-truss and stochastic-truss architectures. Further analysis showed that the gyroid-sheet saturates at D/u≈3, while the octet-truss and stochastic-truss saturate at D/u≈5. The difference between the saturation points for the truss vs sheet-based architectures is likely due to the way the architectures are defined.

The gyroid-sheet is formed using a continuous sheet, so even when the pore is truncated it still contributes to loading. When a pore is truncated in the truss-based architectures, on the other hand, it no longer contributes to the loading. Because of this, the octet-truss and stochastic-truss architectures see much greater boundary effects, so more unit cells across the loading diameter are required to reach bulk material properties. This indicates that the gyroid-sheet is a suitable porous architecture to use in orthopedic implants.





Patterson, Jordan (2020). Exploring relative size effects for strut-based and sheet-based scaffolds defined by repeating unit cell geometry fabricated via selective laser melting. Master's thesis, Duke University. Retrieved from https://hdl.handle.net/10161/20831.


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.