Design and Advanced Manufacturing of Bio-inspired Optimal Microstructures using Machine Learning Method

Abstract

Bioinspired fibrillar structures have been promising for various disruptive adhesive applications. Especially micro/nanofibrillar structures on gecko toes can have strong and controllable adhesion and friction on a wide range of surfaces with residual-free, repeatable, self-cleaning, and other unique features. Also, in some environmental conditions (e.g., relative humidity, temperature), their adhesion performance increases according to literature. These findings can be integrated to design high-performance synthetic structural adhesives such as composite-based synthetic gecko-inspired adhesives. Additionally, there are some debates and theories about the reason for the increase of gecko adhesion in different environmental conditions. The related theories can be examined by studying them systematically. This investigation requires live geckos’ and gecko-inspired synthetic adhesives’ performance comparison in various environmental conditions. These findings can explore why adhesion increases and helps to design high-performance synthetic structural adhesives. Moreover, gecko-inspired synthetic adhesives’ adhesion performance highly depends on their fabrication method. Due to fabrication limitations, the desired complex fibril designs sometimes cannot be fabricated. Advanced fabrication techniques can be integrated to minimize fabrication limitations and fabricate the desired designs almost freely. As a result, a two-photon-lithography-based three-dimensional printing technique can be used with an elastomeric material to manufacture more advanced free-body design fibrils. After all these findings, we can try to explore the outperformance of optimal designs for gecko-inspired synthetic adhesives. Previously, synthetic dry fibrillar adhesives inspired by such biological fibrils have been optimized in different approaches to increase their performance. Previous fibril designs for shear optimization are limited by pre-defined standard shapes in a narrow range primarily based on human intuition, which restricts their maximum performance. In this aspect, we can combine the Bayesian optimization and finite-element-method-based shear mechanics simulations to find shear-optimized fibril designs automatically. In addition, fabrication limitations can be integrated into the simulations to have more experimentally relevant results. The computationally discovered shear-optimized structures are fabricated, experimentally validated, and compared with the simulations. Both experimental and simulation results show that the shear-optimized fibrils perform better than the pre-defined standard fibril designs. This design optimization method can be used in future real-world shear-based gripping or non-slip surface applications, such as robotic pick-and-place grippers, climbing robots, gloves, electronic devices, and medical and wearable devices.