Woodruff School of Mechanical Engineering
Faculty Candidate Seminar
Semiconductor Nanostructures for Efficient Thermo-electric Energy Conversion
Dr. Zlatan Aksamija
University of Wisconsin-Madison
Monday, February 4, 2013 at 11:00:00 AM
MRDC Building, Room 4211
Thermoelectric (TE), or solid-state, refrigeration using semiconductor-based nanostructures, such as nanowires, nanoribbons, and superlattices, is an attractive approach for targeted cooling of local hotspots inside integrated circuits due to inherently no moving parts, ease of miniaturization and on-chip integration, and the nanostructures°¶ enhanced TE conversion efficiency. In addition, thermoelectric power generation enables the reuse of waste heat in a variety of applications, from low-power and energy-efficient designs all the way up to internal combustion engines and solar cells. Thermoelectric efficiency, measured by the figure-of-merit ZT, is dictated by the ratio of electronic power factor S2É„ over the total thermal conductivity. Consequently, largest gains in TE conversion efficiency have come from the ability to reduce thermal conductivity. This is especially true in nanostructures, where small physical dimensions lead to reduced thermal transport due to the scattering of lattice waves, or phonons, with the boundaries and interfaces of the nanostructure. The design of efficient semiconductor thermocouples requires a thorough understanding of both charge and heat transport; therefore, thermoelectricity in semiconductor-based nanostructures requires that both electronic and thermal transport are treated on equal footing. Silicon-on-insulator (SOI) nano-membranes and membrane-based nanowires and ribbons show promise for application as efficient thermoelectrics, which requires both high electronic power factor and low thermal conductivity. In this talk, we present numerical simulation and modeling of both carrier and phonon transport in ultrathin silicon nanomembranes and gated nanoribbons. We show that the thermoelectric response of Si-membrane-based nanostructures can be improved by employing the anisotropy of the lattice thermal conductivity, revealed in ultrathin SOI nanostructures due to boundary scattering, or by using a gate to provide additional carrier confinement and enhance the thermoelectric power factor. Furthermore, we explore the consequences of nanostructuring on silicon/germanium and Si/SiGe alloy superlattices, and show that the drastic reduction of thermal conductivity in these structures comes from the increased interaction of lattice waves with rough interfaces and boundaries. Finally we demonstrate reduced thermal conductivity in both suspended and supported graphene nanoribbons (GNRs), which exhibit strong anisotropy due to interaction of lattice waves with line edge roughness (LER) and the competition between LER and substrate scattering. The talk will conclude with an outlook for future nanostructured thermoelectric based on nanocrystalline and nanocomposite semiconductors, and graphene.
Zlatan Aksamija received his B.S. in Computer Engineering in 2003, and his M.S. and Ph.D. in Electrical Engineering (with Computational Science and Engineering option) in 2005 and 2009, respectively, all from the University of Illinois at Urbana/Champaign. His dissertation work entitled Thermal effects in semiconductor materials and devices was supported by a DOE Computational Science Graduate Fellowship (2005-2009). Zlatan was ranked as an Outstanding TA by his students in the Fall of 2004. He was awarded an Outstanding Paper award at the EIT'07 conference and a Greg Stillman Memorial semiconductor graduate research award in 2008. From 2009 to 2011, Zlatan was a Computing Innovation Postdoctoral Fellow in the ECE department at the University of Wisconsin-Madison. His research, supported by the CIFellows program from the Computing Research Association, focused on semiconductor nanostructures for thermo-electric energy conversion applications, as well as numerical methods for the simulation of electronic and thermal transport in nanostructures. Zlatan is currently the NSF CI TraCS Fellow at the University of Wisconsin-Madison, where he continues to work on computational nanoscience for energy-efficient electronic and thermoelectric materials and devices.
Refreshments will be served.