Hi all,

I am trying to model an octet truss metamaterial using a brittle material model in Abaqus. The constituent material is purely brittle with no plasticity. It fails at an engineering stress of 7 GPa (approximately 10% engineering strain) and has an elastic modulus of 73 GPa.

My objective is to capture the layer-by-layer failure of the octet lattice using either beam or truss elements. However, the simulation results so far do not match the expected behavior. I am using deformable body for octet and top/bottom analytical rigid bodies for compression.

The expected stress-strain response consists of multiple distinct load drops, where each peak corresponds to the failure of half of an octet unit cell. Therefore, for a structure containing 7 unit cells, I would expect to see 14 peaks in the response. I attached Figure 1 for your reference.

To simplify the problem, I initially modeled the lattice using beam elements with a ductile damage initiation criterion and damage evolution. Unfortunately, the simulation still does not reproduce the sequential failure behavior that I expect.

Has anyone encountered a similar problem or have any suggestions on how to model this type of progressive, layer-by-layer failure in an octet truss lattice? In particular:

• Is there a more suitable damage or failure model for brittle lattice structures?
• Are beam or truss elements appropriate for capturing this behavior, or should I consider another element type or modeling approach?
• Are there any numerical settings (e.g., element deletion, stabilization, mesh density, or solution controls) that are important for capturing this sequential failure?

Any advice or suggestions would be greatly appreciated.

Thank you!

Best regards,
Mirhan

Expected result 

Model:

Result: