MechE PhD Dissertation Defense: Xing Ye

TITLE: BIOINSPIRED MULTIFUNCTIONAL SOFT ENERGY SOURCES AND ACTUATOR NETWORKS

ABSTRACT: Sensory and motor functions are fundamental to how living organisms interact with their surroundings. The sensory receptors transmit electrical signals when a stimulus causes sudden changes in membrane voltage, a process powered by electro-chemical gradients within cells. This information is processed and integrated by the nervous system to generate motor commands, which are transmitted through neu-rons to initiate coordinated movements in muscles. The interplay between sensing, signal transmission, and actuation enables animals to perceive and respond to their environments. Efforts have been made to mimic these abilities in biologically inspired systems for applications ranging from wearable devices to soft robotics. Self-powered sensors that mimic the responsiveness of skin and soft actuators that emulate the signal-driven motion of muscle are essential in performing these sensing and mobility tasks in robotic and autonomous systems. However, simulating these biological func-tions in artificial devices remains challenging. In service of this goal, we present a bioinspired framework for designing soft sensory-motor structures that operate under mechanical deformation and environmental stimuli. The first part of the dissertation focuses on the development of liquid metal based soft sensors inspired by the mechanoreceptors in human skin. These self-powered sensors transmit electrical signals when mechanical stimuli are detected, similar to how the nerve endings embedded in skin respond to mechanical pressures. The power in the soft sensors comes from the electrochemical energy of rupturing liquid metal oxide. Liquid metals rapidly form a passivating oxide layer on surfaces that are ex-posed to oxygen-containing environments, which is generally considered a nuisance. We investigate a new approach to exploit the liquid metal oxide by mechanically rup-turing this protective skin and building an electrochemical energy source. We design an emulsion with dispersed liquid metal droplets in an ionic liquid that allows easy device fabrication and customization. Owing to the protection of liquid metal oxide skin, these devices do not self-discharge over time or fail in harsh environments, such as high temperature or aquatic conditions. Future applications are demonstrated by designing a strain-activated stretchable battery and a pressure-sensitive self-powered keypad. The second part of the dissertation explores a dendritic network of soft actua-tors inspired by the ubiquitous branched structures in biological systems. Branching offers an efficient and scalable solution for interacting with natural environments. Blood vessels use branched structures to transport nutrients, plants add branches to maximize light exposure, and fungi utilize branching to increase surface area for nutrient absorption. In particular, neurons form a branching network of dendrites for transmitting signals and processing complex information. Inspired by the ubiquitous utility of dendritic networks in nature, we propose a dendritic architecture consisting of soft bending actuators for shape morphing systems. Each actuator simply performs a basic task of out-of-plane bending under fluidic pressure. However, arranging these actuators in a non-overlapping dendritic network enables the deployment of volumet-ric 3D objects of unprecedented complexity from flat 2D structures. The dendritic architecture unlocks the full range of motion for bending actuators and enables vol-umetric shape transformations that are hard to achieve with existing structures in the literature, such as lattices and kirigami. Each bending dendrite is an integral part of performing the shape-shifting tasks, whose bending behavior is encoded by the geometric properties of the dendritic networks. We also propose an inverse de-sign strategy to construct customized dendritic architecture from prescribed target 3D volumetric objects. We demonstrate the versatility of dendritic actuator networks by showing their applications in autonomous robotic systems and programmable me-chanical metamaterials.

COMMITTEE: ADVISOR Professor J. William Boley, ME/MSE; CHAIR Professor Douglas Holmes, ME/MSE; Professor Joerg Werner, ME/MSE; Professor Andrew Sabelhaus, ME/SE; Professor Abigail Plummer, ME/MSE