Oxide Thermoelectrics
- PDF / 288,078 Bytes
- 7 Pages / 612 x 792 pts (letter) Page_size
- 26 Downloads / 225 Views
1044-U02-05
Oxide Thermoelectrics David Joseph Singh Materials Science and Technology Division, Oak Ridge National Laboratory, 1 Bethel Valley Road, Oak Ridge, TN, 37831-6114 ABSTRACT Thermoelectricity in oxides, especially NaxCoO2 and related materials, is discussed from the point of view of first principles calculations and Boltzmann transport theory. The electronic structure of this material is exceptional in that it has a combination of very narrow bands and strong hybridization between metal d states and ligand p states. As shown within the framework of conventional Boltzmann transport theory, this leads to high Seebeck coefficients even at metallic carrier densities. This suggests a strategy of searching for other narrow band oxides that can be doped metallic with mobile carriers. Some possible avenues for finding such materials are suggested. INTRODUCTION The use of waste heat from engines and other sources represents a significant potential energy savings. Thermoelectrics are an attractive technology for recovering this type of energy. However, widespread application will likely require the development of new thermoelectric materials that combine low cost, stability and high thermoelectric performance, as characterized by the dimensionless figure of merit, ZT = σS2T/κ = (S2/L)(κe/(κe+κl)),
(1)
where S is the thermopower, σ is the electrical conductivity, κe and κl are the electronic and lattice parts of the thermal conductivity, κ=κe+κl, and the last equality is derived from the Wiedemann-Franz relation (expected to hold for thermoelectrics at reasonable operating temperatures), with Lorentz number L. From this expression it is clear that high ZT requires both high S and sufficiently good conductivity to make the ratio κe/κ in the last term reasonable. Based on this, it was long held that thermoelectrics should be degenerately doped semiconductors in order to have a reasonable balance of high S with sufficiently high σ. It was also long thought that while oxides may offer advantages for applications, such as potentially low cost and convenient processing, they were unlikely thermoelectric materials because the carrier mobilities in oxides are usually lower than in good semiconductors. Thus the discovery by Terasaki and co-workers that the layered cobaltate, NaxCoO2 is a high ZT material [1] came as a surprise especially considering that the material is not only an oxide but that it is also a metal with a high carrier density of ~0.3 holes per Co.
The high ZT ≈1 of NaxCoO2 at temperatures relevant for waste heat recovery has since been confirmed [2] and other closely related layered misfit cobaltates with high ZT have been found [3,4]. However, oxide thermoelectrics with comparable ZT from outside this family have yet to be confirmed and applications are difficult because of materials issues, especially the fact that the properties of NaxCoO2 are highly anisotropic. Therefore it is useful to understand the origins of the thermoelectric performance of the cobaltates and if possible use this understanding
Data Loading...
