TITLE: All-Printed Multi-Material Flexible Thermoelectric Devices

ABSTRACT: Wearable Thermoelectric (TE) devices have multiple applications that can benefit humanity. Through the Peltier effect, TE wearable devices create a thermal gradient that can regulate temperature at specific body locations. Conversely, they can turn body heat emissions directly into electricity to power other devices. Despite recent advances in creating TE devices for wearable applications, a viable materials and manufacturing approach remains challenging, owing to tradeoffs between device flexibility and TE performance, the generally low performance inherent in organic TE materials, and the complexity of current fabrication approaches. This project will solve these challenges by combining new multi-scale composite TE material designs with multi-material Direct Ink Writing (DIW) 3D Printing. The efficiency of a TE device is directly dependent on the TE figure of merit and power factor of the materials used as legs. The doping of organic materials has successfully produced flexible films with ultra-high Seebeck coefficients. Thus, a flexible, doped-conductive polymer composite is proposed as the base for the p-type TE material in the device. A systematic study is proposed to investigate the trade-offs between mechanical and thermoelectric properties of the doped-conductive polymer composite. We propose a Thermoplastic Polyurethane (TPU)/Carbon composite doped with an organic charge transfer salt as the n-type ink. Unlike past flexible TE organic materials, which rely on complex spin-casting manufacturing methods and are managed in ultra-thin film geometries, our TE material composites will initially be manufactured as printable inks, enabling automated manufacturing via DIW multi-material printing. A Silver-based ink is developed as the electrode, and a TPU-based ink is used to package the device. Although material design is imperative in creating an efficient TE device, improving power density through device architecture is essential to enhance performance. An initial device will be made using a lateral structure architecture, and a second device will be proposed using a radial architecture. In both cases, a parametric programming approach allows us to develop print paths for devices with varying TE leg geometries and print them sequentially. Since our individual material properties aren’t highly dependent on the manufacturing approach, DIW allows us to optimize device performance and carry out material performance independently.

COMMITTEE: ADVISOR/CHAIR Professor William J. Boley, ME/MSE; Professor Sean Lubner, ME/MSE; Professor Joerg Werner, ME/MSE