Machine Learning-Guided Design of Shell-Based Multistable Lattices for Superior Energy Absorption and Reusability.

Area of Science:

• Materials Science
• Mechanical Engineering
• Nanotechnology

Background:

• Hollow-tube micro/nanolattices are effective energy absorbers but suffer from permanent damage, limiting reusability.
• Multistable metamaterials offer reusability but often have constrained load-bearing and energy absorption due to beam instability.
• Existing designs struggle to balance recoverability, strength, and energy absorption for demanding applications.

Purpose of the Study:

• To design and develop novel, reusable energy-absorbing metamaterials with enhanced mechanical properties.
• To overcome the limitations of traditional hollow-tube lattices and constrained multistable designs.
• To explore the potential of shell-based architectures for tunable multistability and superior energy absorption.

Main Methods:

• Integration of finite element simulations and machine learning to design shell-based unit cells with continuously varying thickness.
• Fabrication of multistable microlattices using projection microstereolithography.
• Characterization of mechanical properties, including deformability, recoverability, reusability, strength, and energy absorption.

Main Results:

• Designed unit cells exhibit tunable multistable behaviors controlled by geometric parameters.
• Fabricated microlattices demonstrate large deformability, excellent recoverability, and reusability.
• Achieved high strength (≈91.7 kPa) and superior energy absorption (≈6.2 × 10^4 J m^-3), outperforming existing hollow-tube lattices.
• Demonstrated versatility through fabrication with various materials and 3D printing techniques, including 3D hierarchical structures with multi-directional multistability.

Conclusions:

• Shell-based unit cells provide a viable route for designing highly reusable and efficient energy-absorbing metamaterials.
• The developed microlattices offer a significant advancement over traditional designs, enabling exceptional energy absorption and tunable multistability.
• This mechanomaterial approach opens new possibilities for scalable and versatile metamaterial design in diverse applications requiring robust energy management.