Thermoelectric (TE) heat recovery technologies can frequently reduce the operating costs for facilities by increasing energy productivity and efficiency. New systems using advanced ceramic materials are being developed, showing promising results at higher temperatures up to 1,000°C. Further, they are demonstrating that the material is not only functioning, but at potential device efficiency three times higher than current, state-of-the-art thermoelectric materials.                              

Waste heat recovery from diesel or gasoline engine exhaust can increase fuel efficiency while reducing weight, carbon deposit and manufacturing costs of both military and commercial vehicles. In an extreme case, jet-based aircraft have a huge available temperature difference between the jet exhaust and the passing air. The greater the temperature difference between the two sides of the thermoelectric materials, the more electricity can be harvested. The problem with current thermoelectric materials is they have not been able to operate efficiently above 300°C, or the materials are cost-prohibitive for industrial applications. Either the material or the system begins to break down under this heat. Thus, current technologies have been unable to capture energy where most of it is available, in the 700-1,000°C range.

The market demands a low cost of production, as well as flexibility in use. The technology solution must be adaptable to harvesting heat in different applications, ranging from power production plants to smaller vehicle power trains. Above 300°C, it can no longer function as a result of the materials or modules losing stability. Another example is in a coal power plant, where the furnace heats water to 370°C in order to make steam; however, to increase energy output, the steam is superheated above 540°C. Many industrial processes generate waste heat above 800°C. In order to integrate thermoelectric waste heat recovery in power plants, these systems and materials will have to be resistant to much higher temperatures.

Thermoelectric Markets

Although thermoelectric phenomena have been extensively used for heating and cooling applications, electricity generation has only seen limited market niche applications. It is only in recent years that interest has increased regarding new applications of energy generation through thermoelectric harvesting. This growth in interest is expected to continue, characterized by an overall market for thermoelectric energy harvesters that will reach $750 million by 2022. The four main markets for high-temperature thermoelectrics include vehicle waste heat recovery, consumer applications, military and aerospace applications, and industrial applications.

Waste Heat Recovery in Vehicles

The Environmental Protection Agency (EPA) reports that only 15% of fuel energy is used to move a vehicle down the road or run useful accessories, such as air conditioning. According to BMW, automobiles are an example of high energy usage with low efficiency. Roughly 75% of the energy produced during peak combustion is lost in the exhaust and engine coolant in the form of heat.

A large number of car companies, including Volkswagen, Volvo, Ford, and BMW, in collaboration with NASA and the U.S. Department of Energy (DOE), have been developing thermoelectric waste heat recovery systems in-house. Each company has achieved different types of performance, but all are expected to lead to improvements of 3-5% in fuel economy, while the power generated out of these devices could potentially reach up to 1,200 W.

Other automotive companies researching thermoelectrics include General Motors, Toyota/Denso, Bosch, Amerigon/BSST, Siemens, and Cummins. To date, the Ford Fusion, BMW X6, and Chevrolet Suburban concept cars have all been produced with thermoelectric generators in the exhaust stream.

One highly visible application might be to increase energy efficiency in combustion-engine-based transport, thereby reducing the U.S.’s carbon footprint and dependency on foreign oil imports. Converting waste heat produced by motors into electricity would make cars far more efficient. There is an untapped opportunity that exists to harvest energy from the engine and exhaust infrastructure.

Already, BMW states that fuel efficiency has been increased by 12% with early thermoelectric prototypes. With DOE funding, BMW and Ford are working on these prototypes for passenger cars. If successful, alternators would no longer be needed to recharge a car battery (in both hybrid and non-hybrid models). This would translate to reduced weight and air pollution, leading to improved fuel efficiency. Elimination of an alternator has another advantage with vehicles that operate under industrial and field environments.

Consumer Applications

In these applications, the type of solution that thermoelectric generators provide varies and could be related to saving energy while cooking or heating your home.

Military and Aerospace Applications

These applications have already become a market of several million dollars, having supported mature thermoelectric harvesting technologies for several decades now (e.g., radioisotope thermoelectric generators in space probes and satellites). These are purely performance-driven applications in a segment where cost considerations are not as important as the ability to efficiently and reliably provide power when needed most—i.e., in hostile, remote environments and applications.

Industrial Applications

In certain industrial applications, such as the melting of steel and glasses or to power nuclear or coal plants, heat is generated from the production facilities. In other words, waste heat is generated as part of the industrial manufacturing process and is then lost to the environment. Aluminum, glass, metal casting and steel: all have process furnaces discharging high-temperature waste heat combustion gases and melt pool gases (e.g., aluminum, ~ 775°C; and glass, ~ 1,425°C).

In some industries, this heat can be used to raise steam, preheat raw materials or combustion air, or be integrated with other processes at the manufacturing site. Other industries have limited opportunities to reuse this thermal energy, however. This is what makes the thermoelectric generator (TEG) so attractive.

The opportunity to recover waste heat should be large in metals industries, which use numerous heat-treatment furnaces (with relatively clean flue streams of combustion gases only), as well as in chemical industries, where process heaters are widely used (e.g., direct-fired reboilers, reactors, etc.). Additional waste heat opportunities exist in lime kilns, cement kilns, etc. Table 1 shows a breakdown of the potential waste heat by industry.

Proposed Solutions

With the increasing cost of fossil fuels, improved efficiencies can result in substantial cost savings. Current commercially available thermoelectric conversion systems and materials can only function at waste heat temperatures below 500°C. This limitation leaves out many industrial processes where waste heat is generated at high temperatures. The limitation is largely due to three factors: a lack of suitable low-cost, high-temperature thermoelectric materials; mechanical failure caused by thermal expansion mismatch between thermoelectric component and conductive or metalized materials; and design limitation in heat management. This has created an opportunity to develop new thermoelectric materials for harvesting the wasted energy.

Before discussing the current state of research in this area, it is worthwhile to define some basic terminology used in this field. The energy conversion efficiency of a TE material is gauged by a parameter called figure of merit (ZT). ZT is a dimensionless parameter and is defined as:

ZT = (Q² x s x T)/k

where Q (µV/K) is the Seebeck coefficient of the material, s (S/cm) is the electrical conductivity of the material, T (K) is the absolute temperature and k (watt/m-K) is the thermal conductivity of the material.                                                        

Early TE materials research in the 1950s and 1960s yielded bismuth telluride (Bi₂Te₃), lead telluride (PbTe) and silicon-germanium (SiGe) alloys as the materials with the best ZTs in three somewhat distinct temperature ranges. Bi₂Te₃ and its alloys have been used extensively in TE refrigeration applications and some niche low-power generation applications, and have a useful temperature range of 180-450 K. PbTe and SiGe materials have been used extensively in higher-temperature power generation applications, particularly spacecraft power generation, and have a useful temperature range of 500-900 K and 800-1,300 K, respectively.

The discovery of these materials as good candidates for TE devices led to the development of a fledgling TE industry. Many of its early participants are still active in the area today; they are focused on using BiTe for heating and cooling applications, PbTe for moderate temperature energy harvesting, and SiGe for high temperatures.

Proposed Technology

Research work will first focus on the development of high-temperature p-type and n-type ceramic oxide thermoelectric materials. The proposed technical approach is to increase the electrical conductivity of oxide ceramic thermoelectric compounds through doping and decrease the thermal conductivity through the introduction of structured porosity.

Some success has been achieved in developing such structures for use as a substrate in oxygen transport membrane applications for CO₂ sequestration (see Figure 1). The integrated system developed through this approach will be far more efficient and can harvest thermal waste energy at temperatures current materials cannot tolerate. It can also be produced in a low-cost manner and form-fitted to different industrial applications.

Multi-layer structures consisting of dense and porous layers will be built as shown in Figure 2. Figure 3 shows a sample configuration for TE measurement. Real devices based on the previously described concept can be designed and fabricated. Such a system is advantageous in that:

•  It pushes the operating temperature to a 700-1,000ºC range, at which no TE materials are currently available

•  It is based on conventional multi-layer ceramics technology currently practiced in the electronics industry

•  It leads to lower-cost TE material

The most important technical goal is to develop n-type and p-type oxide thermoelectric materials with ZT > 0.5. Particle size of the thermoelectric powders should be < 600 nm, and their sintering temperature should be lower than 1,500ºC. Other goals include the development of a process to manufacture porous tape and a process to make ceramics with a porosity designed to enhance their thermoelectric properties. With respect to cost, the aim is that TEG produced using this technology would meet a metric of < $4 (cost to own) per watt (output to gain).

Bismuth chalcogenides (Bi₂Te₃ and Bi₂Se₃) comprise some of the best-performing room temperature thermoelectrics, as they have the highest percentage of Carnot efficiency for thermoelectric generators. Bismuth tellurium is the best; however, overall telluride compounds tend to be limited in use due to toxicity, rarity and the high cost of tellurium. For higher-temperature applications, lead telluride and selenide have shown good ZT values (at Oak Ridge National Laboratory) between 1.4 and 1.8, with predictions of p-type ZT as high as 2 at 1,000 K; however, there are cost and toxicity issues with Pb.

Silicon-germanium alloys are currently the best-developed thermoelectric materials for applications up to 1,000ºC and are therefore used in advanced aerospace applications such as radioisotope thermoelectric generators (RTG). However, energy conversion has historically been low, and ZT values above 1 have not been commercialized, with ZT starting to degrade above 800ºC. Oak Ridge National Laboratory considers the practicality threshold ZT greater than 2.0. Silicon-germanium material is also very expensive compared to strontium-, niobium-, and titanium-containing oxides.*

To date, oxide thermoelectrics demonstrate exceptional thermal stability, especially at high temperatures, but their figure of merit is relatively low (0.34 at 1,000 K, although research suggests that some layered oxides can reach ZT values up to 2.7 at 900 K). The approach is to use low-cost, low-toxicity, and high-abundance materials and minerals, then process these raw materials using historically inexpensive processing techniques, such as solid-state reaction, ultimately realizing quick commercialization.

  *TAM’s n-type material

Projected Benefits

When this thermoelectric technology is fully developed and broadly adopted, it is expected that in the automobile industry alone, it could reduce the fuel consumption of U.S. drivers by over 10%, resulting in savings of as many as 2.1 million barrels of crude per day, according to the CIA World Factbook. Not only would drivers save at least 10% of the money they spend on fuel, but they would benefit due to the reduction in cost of buying and maintaining a car.

Initial expectations are that the system would cost only 50% of the cost of an alternator, and would have a much longer lifetime and fewer maintenance problems. When applied to other industrial processes such as a coal power plant, or a jet combustion engine, this technology could be equally as large and significant.

Such a new technology brings a significant opportunity to recapture wasted energy generated at high temperatures (1,000ºC). Capturing waste heat generated from the broad spectrum of industries that developed during the world’s industrial revolution has a profound impact. We may experience total efficiency from the many great inventions of engines for vehicles, power generation, and industrial processes. The future of new thermoelectric systems using advanced ceramic materials harvesting these high-temperature waste heat energies and converting them into electricity looks promising.  

For more information, visit www.tamceramics.com.

 Note: Funding for this project has been provided by the New York State Energy Research and Development Authority (NYSERDA) under Agreement Number 30370.  NYSERDA has not reviewed the information contained herein, and the opinions expressed in this report do not necessarily reflect those of NYSERDA or the State of New York.