455
Thermoelectric Energy Harvesting Research at the Assist NSF Nano-Systems Engineering Research Center
The thermoelectrics effort focuses on combining nanostructured, high figure of merit materials with advances in material and device design to minimize parasitic electrical and thermal resistances. The three key challenges are: 1) Materials with high figures of merit, ZT; 2) Flexible yet efficient packaging that reduces the parasitic thermal resistances at the skin / device and device / ambient interfaces; 3) A device with a small form factor that is comfortable to wear.
The experimental work is supported by a modeling effort that focuses on achieving comprehensive understanding of fundamental electrical and thermal transport properties of lead and bismuth chalcogenide heterostructures. The theoretical treatment starts from a first principles approach based on density functional theory to extract the energy band details of electrons and phonons as well as the interaction between them in the bulk materials and in nanostructures. Then, the macroscopic electrical and thermal transport properties are evaluated by solving the Boltzmann transport equation in the diffusive regime, including those for the structures with dimensional constraints. A Monte Carlo approach is chosen to be used for both electrons and phonons that can accurately describe nonlinear effects. The key focus is placed on the nanoscaled superlattices as well as more complex quantum well quantum wires.
Different integration approaches are being considered including a) integration of epitaxial materials using conventional pick-and-place tooling and thermal compression bonding into a flexible polyamide package and b) electrochemical deposition of thermoelectric materials within the flexible package.
It is essential for the flexible package to conduct heat extremely well. Furthermore, the package has to be stable up to at least 300oC needed for solid diffusion bonding of the thermoelectric legs. In our work, we are exploring polyamide with embedded metal studs (Figure 1) to satisfy both the flexibility and temperature stability requirements.
The small form factor requires a small and flexible heatsink to effectively dissipate the heat via convection. Together with the thermal resistance of the skin, the thermal resistance of the heat sink limits the achievable temperature difference across the thermoelectric device. As shown in Figure 2, the performance of a generator capable of delivering 1 mW output power can be severely hampered by these thermal resistances. The calculations were carried out for heat sink and skin conductance ranges of 20-500 and 20-100 W/m2K respectively assuming a device thermal resistance of ~500 K/W. The total device area is 4 cm2. We are exploring millimeter tall carbon nanotubes to construct highly efficient, small-profile heatsinks (Figure 3). The advantages of CNTs include their large surface area, high thermal conductivity and their ability to grow as fins on patterned catalysts.
Another effort in the center is aiming at passive heat pipes using novel superhydrophilic hierarchical nanostructures. The goal is to enable a flexible, integrated two-phase heat transfer device to effectively transport heat to and from the thermoelectric generator (TEG). Heat pipes can offer thermal conductivity figures that are orders of magnitude higher than those offered by even the best metals. The approach is based on surface nanostructures with two-tier length scales to design both the fluid transport via improved wicking and thermal transfer properties independently. The devices are fabricated using laser nanolithography, atomic layer deposition, and nanowire synthesis. The key challenges are optimizing viscous fluid friction and thermal contact area in the nanostructures, as well as device packaging using thin metals sheets into a flexible platform.