Thermoelectrics Gaining More Attention and Development Focus

22 July 2005

Exhaust heat = available energy—if you can convert it.

Thermoelectrics made it into the headlines recently from an unlikely source: two Utah teenagers won the first Ricoh Sustainable Development Award in May for using Peltier chips in a prototype automotive air conditioning unit they built for the Intel International Science and Engineering Fair.

The glowing story in the Salt Lake Tribune and subsequent online discussion (and some snark) on sites such as (Sustainability Zone and Slashdot unintentionally underscored an area of research that is gaining momentum, focus and support from the Department of Energy, GM and other automakers, and numerous international research institutes and universities: the use of thermoelectric materials to convert waste engine heat into useful electricity (Waste Heat Recovery). (Earlier posts.)

Thermoelectric materials exploit a phenomenon in which the application of heat to combinations of certain metals induces an electric current. Thomas Seebeck, a German physicist, first noticed the thermoelectric effect in 1821. (The Peltier effect is the reverse of the Seebeck effect—the creation of a heat difference from an electric voltage.)

Until fairly recently, thermoelectric devices have been hampered by low efficiency (as many of the comments on the Utah teens’ work noted) and high cost. Consequently, their applications have been limited.

In a speech to be given at next week’s International Conference on Energy, Environment and Disasters, John Fairbanks from the DOE will succinctly outline the opportunity:

Thermoelectrics in the last half century were based on bulk semiconductors and essentially limited to a Figure of Merit (ZT) of 1.0. In the mid 90’s Millie Dresselhauss at MIT analytically proposed a nano technology approach of quantum well superlattices that should lead to very efficient thermoelectric devices. Several organizations reported ZT’s of 2.4 and 4.0 with small laboratory specimens around the start of this Century.

Much of this work was supported by the Defense Advanced Research Program (DARPA)/ Office of Naval Research (ONR)’s Multiple University Research Initiative as well as DOE’s program with Hi-Z Technologies and our Pacific Northwest National Laboratory (PNNL).

The DOE High Efficiency Thermoelectrics Program brings the laboratory breakthrough to production scale and development of a direct heat to electricity device for energy recovery from a vehicle engines waste heat.

Waste heat recovery can be obtained from the radiator, lube oil cooler, exhaust gas, exhaust gas recirculation (EGR) loop, turbocharger compressor discharge. The latter two would do double duty as Oxides of Nitrogen (NO_x) is reduced by cooled EGR and cooling the turbocharged air increases the volume introduced to the combustion chamber which improves efficiency.

Development of this high thermoelectric technology mated with the beltless or “more electric” diesel engine wherein all the engine accessories are motor driven or the integrated starter/motor/alternator/damper could have diesel engine efficiency approach 60 percent efficiency! This could be achieved with a very wide range of liquid or gaseous fuels. This engine would reduce CO₂ emissions by 50 percent as a “bonus.”

Fairbanks is a big supporter of the thermoelectric effort within the DOE, and the topic will, once again, be prominent at this year’s upcoming DOE Diesel Engine Emissions Reduction conference (DEER 2005).

At the 24th International Conference on Thermoelectrics (ICT2005) held in June at Clemson University, Fairbanks, Jihui Yang of GM and Takenobu Kajikawa of Shonan Institute of Technology spoke separately about their plans for practical, industrial-scale thermoelectric waste heat recovery systems.

According to Cronin Vining’s ZTspam report from the conference:

From these and other groups, a consensus seems to have emerged that 15% conversion efficiency is achievable in the near term and also economically viable for thermoelectric technology. While there are many similarities in the actual technical work endorsed by these three men, there was a distinct difference in motivation. The number one concern for the Americans seems to be cost and/or fuel saving. The number one concern for the Japanese seems to be the environment.

Earlier this year,the DOE kicked off a more than $6-million project for the development of an efficient and practical thermoelectric system that will improve fuel economy by converting waste heat in automobile engine exhaust into electrical power.

The DOE selected BSST, a subsidiary of Amerigon, to lead the work. BSST will lead a development team that includes Visteon Corporation, Teledyne Energy Systems, BMW of North America, UC Santa Cruz, Purdue, the DOE’s National Renewable Energy Laboratory (NREL) and JPL/CalTech.

The basic insight the Utah teens had is essentially on the mark. In fact, thermoelectric devices are already used in heating and cooling in automobiles—but just in individual seats, rather than the entire cabin. BSST (the lead in the DOE TE project) currently supplies a million car seat heater/coolers per year to Ford, GM, Toyota and Nissan to augment the cabin cooling system.

Again, John Fairbanks:

Successful high-efficiency thermoelectrics could replace the current auto gas compressor air conditioning unit. They could be more efficient, therefore less CO₂ would be released to the atmosphere, and the solid state Peltier devices would not involve any refrigerant gases eliminating their contribution to Greenhouse Gases. As diesel and gasoline engines become more efficient, less heat is available for vehicle cabin heating. Peltier devices would be candidates for these vehicles as well as electric, hybrid and fuel cell vehicles.

Researchers Tammy Humphrey and Heiner Linke have published a design principle for nanostructured thermoelectric materials that could dramatically increase their efficiency, leading to widespread adoption in devices such as those described above.

The pair calculated that certain nanostructured materials could achieve values of the standard thermoelectric figure of merit Z of around 10. This is approximately twice the values theoretically achieved by another research team working with nanostructures in 2003. (Earlier post.)

(A hat-tip to Rod Edwards!)

Resources:

• International Thermoelectric Society

• ICT2005 proceeds available from http://shop.ieee.org and http://www.cpmt.org/proceedings/ in October. All orders must be placed before September 15th.