Coming: More efficient thermoelectric generators, thanks to nanocomposites

You can add nanocomposites to the list of materials that will turn heat into electricity. Researchers at the Massachusetts Institute of Technology will be describing their work in thermoelectric generators based on nanocomposites at the upcoming IEEE International Electron Devices Meeting (IEDM), Dec. 5-7 in Washington, D.C.

The idea behind thermoelectric power is as conceptually simple as its name implies — turning heat into electricity. However,real-world applicationof the idea is anythingbut easy. The conversionof sunlight into electricityis dominated by photovoltaic(PV) and solarthermal power generationusing optical concentratorsand mechanical heat engines.

Taking a place alongside these technologies is a promising flat-panel, solar thermal-to-electric power conversion technique based on the Seebeck effect. Developed by MIT's Dr. Gang Chen and his research team, the work has implications that could potentially be quite broad. The team's solar thermoelectric generators (STEGs) have hit a peak efficiency of 4.6% — a figure seven or eight times higher than the best flat-panel STEG values reported previously.

From one pair of pn junctions, each about 1 mm3 and developing between 0.06 and 0.08 V, researchers say they can get 60 to 80 mW. In real devices, these will go in series to boost output voltage, as typically done in thermoelectric devices. They also expect efficiency improvements will come from three directions: improving the materials' figure of merit; improving surfaces so they better absorb solar radiation and minimize thermal emission; and from combining the devices with some optical concentration.

“The new device employs high-performance nanostructured thermoelectric ma terials and spectrally-selective solar absorbers in a unique design that exploits high thermal concentration in a vacuum environment,” explains Chen. It works by generating a temperature difference of roughly 200°C between the device interior and ambient air. Rather than using movable mirrors that follow the sun and focus its rays on a tiny area, the new system uses stationary panels not unlike traditional solar panels. Doing so eliminates the need for expensive tracking systems. Similar to silicon PV cells, the team's creation is a solid-state device without moving parts.

Inside a glass vacuum chamber sits a thermoelectric generator covered with a black plate of copper, which absorbs sunlight but does not re-radiate it as heat. The generator's other side is exposed to ambient temperatures. When in the sun, the unit quickly heats up. In comparison with conventional PV panels, the economics of the new device look good. It requires less material and would be much cheaper to produce than typical PVs. Another benefit: It could be integrated into existing solar hot water systems, which would boost the bang for the capital expenditure buck. Although solar water heaters aren't widely deployed in the U.S., they're common in residences throughout China and Europe, where they provide affordable hot water.

“Integrating thermoelectric generators into an existing solar hot water system incurs very little additional expense. A fully installed system that uses selective sur faces and vacuum tubes costs around $300 per family in China,” says Chen.

Materials needed to build these STEGs are made by means of a nanostructured process developed a few years ago by Chen's lab at MIT and at researcher Zhifeng Ren's Boston College lab. The teams continue to improve the materials and use them in complete systems. Beyond the idea of piggybacking onto solar hot water heaters, Chen says that the DoE is supporting research to develop better thermoelectric materials for such applications as capturing waste heat from car and truck engines, which could play an important role in reducing carbon emissions.

Besides working on the new STEG application, Chen is also researching the fundamental aspects of electron and phonon transport in thermoelectric materials. (A phonon is basically a special type of vibrational motion within a crystal lattice.) Bismuth telluride-based alloys are among those having the most useful properties, but others hold promise as well.

Page 2 of 2

“Other materials include skutterudites, lead tellurides, and half heuslers. Although we knew about some of these before, engineering the transport of their phonons and electrons has significantly improved their ZT values,” says Chen. (Z is a figure of merit for thermoelectric devices. The dimensionless figure of merit ZT is Z multiplied by the average temperature the device sees.) Skutterudites are cobalt arsenide minerals containing nickel and iron. Heuslers are ferromagnetic metal alloys based on a Heusler phase, basically a particular composition with a face-centered cubic crystal structure.

The field of thermoelectric power generation continues to heat up (no pun intended), with more than 3,000 journal papers published since 2009. Chen and other research teams worldwide are studying new materials, strategies, fabrication techniques, and applications. While researchers are making, gains in materials and in understanding fundamentals, many new issues are being raised while old ones remain unanswered. For example, the understanding of phonon transport in bulk materials has notably advanced, yet basic questions around phonon and electron transport across interfaces and within thermoelectric materials remain, according to Chen. To make matters even more complex, nanocomposites add another layer of intricacy because they introduce many more interfaces.

“The structure of these interfaces, the phonon transmissivity at a single interface, and multiple scattering events associated with the interfaces are not well understood,” explains Chen.

So far, the main application area for thermoelectric devices involves cooling, for example, small mobile refrigerators, cooled car seats, temperature regulators for semiconductor lasers, and medical and scientific instruments. As far as power generation, thermoelectric generators have found use in space applications, while earthly applications have been limited to uses such as remote power along oil pipelines and bodytemperature-powered watches.

The main driver of further research is vehicle waste heat recovery, because automobile efficiency remains around 20%. Roughly one-third of the heat goes out the exhaust pipe, while another third is dissipated through the radiator. Between these two heat sources, the exhaust pipe sports the higher temperature and is better suited for thermoelectric technology. According to Chen, all major automakers have programs on thermoelectric waste heat recovery for conversion to electricity, driven by legislation on emission mandates for future vehicles. That said, vehicle applications are widely considered the most challenging because of varying driving conditions and size and weight limit requirements. Other researchers are looking into areas such as industrial waste heat recovery for thermoelectric power generation.